UNIVERSIDADE DE LISBOA

FACULDADE DE FARMÁCIA

“HIV infection in Angola: Molecular epidemiology, diagnosis

and antibody neutralization”

Francisco Neves Dos Santos Martin

Orientadores: Prof. Doutor Nuno Eduardo Moura dos Santos da Costa Taveira
Prof. Doutor João Manuel Braz Gonçalves

Tese especialmente elaborada para a obtenção do grau de Doutor em Farmácia,

especialidade Microbiologia.

2022
 UNIVERSIDADE DE LISBOA

FACULDADE DE FARMÁCIA

“HIV infection in Angola: Molecular epidemiology, diagnosis

and antibody neutralization”

Francisco Neves Dos Santos Martin

Orientadores: Prof. Doutor Nuno Eduardo Moura dos Santos da Costa Taveira
Prof. Doutor João Manuel Braz Gonçalves

Tese especialmente elaborada para obtenção do grau de Doutor em Farmácia (Microbiologia)

Júri:
Presidente: Doutora Maria Alexandra de Oliveira Silva Braga Pedreira de Brito, Professora
Associada com Agregação e membro do Conselho Científico da Faculdade de Farmácia da
Universidade de Lisboa
Vogais: Doutora Ana Barroso Abecasis, Professora Auxiliar, Instituto de Higiene e Medicina
Tropical da Universidade Nova de Lisboa
Doutor Nuno Eduardo Moura dos Santos da Costa Taveira, Professor Catedrático, Instituto
Universitário Egas Moniz, orientador
Doutora Perpétua da Conceição Rodrigues Gomes Cavaco Silva, Professora Associada, Instituto
Universitário Egas Moniz
Doutor Fernando Manuel Tavares Maltez, Professor Auxiliar Convidado, Faculdade de Medicina
da Universidade de Lisboa
Doutor José Miguel Azevedo Pereira, Professor Auxiliar com Agregação, Faculdade de Farmácia
da Universidade de Lisboa

Francisco Martin teve o apoio financeiro da Fundação para a Ciência e Tecnologia através de
uma bolsa de doutoramento (SFRH/BD/87488/2012).
2022
Todas as afirmações efetuadas no presente documento são da exclusiva responsabilidade do seu
autor, não cabendo qualquer responsabilidade à Faculdade de Farmácia de Lisboa pelos conteúdos
nele apresentados.
Francisco Martin teve o apoio financeiro da Fundação para a Ciência e Tecnologia através de uma
bolsa de doutoramento (SFRH/BD/87488/2012).
“Faz-te um Homem”

-Amaro Jesus Neves

Acknowledgements

Foi uma longa, feliz e fascinante etapa que irei recordar sempre com bastante carinho e

que me permitiu abrir horizontes e ser o ser humano que sou hoje.

Em primeiro lugar, gostaria de agradecer ao Professor Doutor Nuno Taveira, na qualidade

de orientador desta tese, por me ter envolvido e dado a conhecer desde cedo a área da

investigação. Posso afirmar que a investigação científica supera tudo o que fiz até hoje,

em termos de conhecimento, superação, persistência, cooperação, envolvência e

principalmente resiliência. É fascinante! “Ficou o bichinho”.

Obrigado pela oportunidade de participar em diferentes projetos em que estivemos juntos

e por toda a partilha que levou ao crescimento do meu conhecimento científico. Fui muito

feliz nesta equipa.

Obrigado ainda por toda a compreensão, supervisão, orientação e apoio nesta reta final.

Em segundo lugar gostaria de agradecer ao Professor Doutor José Moniz Pereira, como

ex-diretor do Departamento de Microbiologia da Faculdade de Farmácia, por ter

permitido desenvolver o meu trabalho nesta unidade e também por ter iniciado a minha

coorientação.

O meu muito obrigado ao Professor Doutor João Gonçalv es por ter abraçado a

coorientação do meu projeto de doutoramento já na fase final da redação da dissertação.

Obrigado também a todos os docentes que trabalham no departamento, em particular à

Prof. Quirina, ao Prof. José Miguel, à Prof. Isabel Portugal e ao Prof. Jorge Vítor.

Um especial agradecimento à Dr.ª Sofia Clemente e restantes colaboradores do Hospital

da Divina Providência em Luanda, Angola. Sem a vossa colaboração este trabalho não

teria sido possível. Muito obrigado!

ii
Aos doentes Angolanos infetados por VIH que participaram nos estudos desenvolvidos

no decurso deste projeto, o meu muito obrigado!

À Fundação para a Ciência e Tecnologia o meu obrigado pelo apoio financeiro que

permitiu desenvolver este trabalho.

Obrigado a todos os meus colegas de laboratório, Joana Duarte, Andreia Martins, Cláudia

Paladino, Cheila Rocha, Inês Bártolo, Pedro Borrego, Carla Silva, João Perdigão, Marta

Calado, Marta Gíria, Joana Vital, Ana Rita, Inês Figueiredo, Inês Moranguinho, Rita

Calado, Rita Mateus, José Marcelino, Rute Marcelino e Jaciara pela amizade, conselhos,

esclarecimentos, apoio e momentos tão bem passados que deixam saudades.

À minha família, um muito obrigado por todo o apoio, compreensão e por sempre me

motivarem a fazer o melhor, e ser melhor.

Em especial aos meus pais, ao meus sogros, tios, primos, irmão, avó, à minha esposa e

aos meus filhos por nunca desacreditarem em mim e sentirem-se sempre orgulhosos do

meu esforço e dedicação.

A estabilidade, educação e formação que me têm dado e continuam a dar torna tudo mais

fácil. Vocês conhecem-me como ninguém e sabem o que penso em relação a este assunto

de querer estar sempre a melhorar e de ser um eterno insatisfeito. Contudo, deixo aqui

por escrito, eu não vejo o meu sucesso como algo extraordinário que deva ser celebrado,

mas sim como uma obrigação por tudo o que fizeram e continuam a fazer por mim.

iii
iv
PREFACE
The research described in this work was performed under the supervision of Prof. Doutor

Nuno Taveira and the cosupervision of Prof. Doutor João Gonçalves. All experimental

work was performed at HIV Evolution, Epidemiology and Prevention laboratory,

Research institute for Medicines ([Link]), Faculty of Pharmacy, Universidade de

Lisboa.

The results obtained in this thesis were described in the following publications:

• Bártolo I, Zakovic S, Martin F, Palladino C, Carvalho P, Camacho R, Thamm

S, Clemente S, Taveira N. HIV-1 diversity, transmission dynamics and primary

drug resistance in Angola. PLoS ONE. 2014; 9(12): e113626

• Martin F, Palladino C, Mateus R, Bolzan A, Clemente S, Gomes P, Taveira N.

Early infant diagnosis of HIV-1 infection in Luanda, Angola, using a new DNA

PCR assay and dried blood spots. PLoS ONE. 2017;12(7): e0181352.

• Martin F, Marcelino J, Palladino C, Bártolo I, Tracana S, Moranguinho I,

Gonçalves P, Mateus R, Calado R, Borrego P, Leitner T, Clemente S, Taveira

N. Long-term and low-level envelope C2V3 stimulation from highly diverse

virus isolates leads to frequent development of broad and elite antibody

neutralization in HIV-1 infected individuals. Available at medRxiv

[Link] (Manuscript submitted to

eBioMedicine)

v
Peer-reviewed abstracts published in international journals:

Martin F, Palladino C, Mateus R, Clemente S, Gomes P, Taveira N. Evaluation of an in-

house molecular HIV-1 Test to assess Mother-to-Child HIV-1 transmission in angola (the

APEHC cohort) JAIDS Journal of Acquired Immune Deficiency Syndromes. 2016; (71):

Abstract_P83

Martin F, Palladino C, Mateus R, Diniz AR, Calado R, Figueiredo I, Bártolo I, Borrego

P, Clemente S, Taveira. Broad and potent neutralizing antibody responses in HIV-1

infected Angolan patients: implications for vaccine design and efficacy. HIV Medicine,

20 (Suppl. 9), 3–316. Abstract_PE38/9

Oral communications:

Martin F, Palladino C, Mateus R, Diniz A.R, Calado R, Figueiredo I, Bártolo I, Borrego

P, Clemente S, Taveira N. Broad And Potent Neutralizing Antibody Responses In HIV-

1 Infected Angolan Patients: Implications For Vaccine Design And Efficacy. 1 st Annual

Meeting in Pharmaceutical Sciences, Coimbra, Portugal, November, 20 19. (Invited

communication)

Martin F, Palladino C, Mateus R, Diniz A.R, Calado R, Figueiredo I, Bártolo I, Borrego

P, Clemente S, Taveira N. Broad And Potent Neutralizing Antibody Responses In HIV-

1 Infected Angolan Patients: Implications For Vaccine Design And Efficacy. 11 th iMed

UL. Postgraduate Students Meeting, Lisbon, Portugal, 15th of July, 2019. (awarded best

oral communication)

vi
Martin F, Palladino C, Mateus R, Diniz AR, Calado R, Clemente S, Taveira N.

Characterization of the neutralizing antibody responses in HIV-1 infected patients from

Angola and their impact on disease progression. 4o Congresso Nacional de Medicina

Tropical. Instituto de Higiene e Medicina Tropical. 1o encontro lusófono de SIDA,

Tuberculose e Doenças oportunistas. Instituto de Higiene e Medicina Tropical. 19-21

April 2017.

Poster Communications:

Martin F, Palladino C, Mateus R, Diniz AR, Calado R, Clemente S, Taveira N.

Neutralizing antibody response in HIV-1 infected patients from Angola. Keystone

Symposia, Molecular and Celular Biology, March, 2018.

Martin F, Palladino C., Mateus R, Diniz AR, Calado R, Clemente S, Taveira N.

Neutralizing antibody response in HIV-1 infected patients from Angola. 9th

[Link] Postgraduate Students Meeting and 2nd i3DU Meeting. Poster number 92.

13-14 July 2017. Faculty of pharmacy. Lisbon.

Martin F, Palladino C, Mateus R, Diniz AR, Calado R, Clemente S, Taveira N.

Neutralizing antibody response in HIV-1 infected patients from Angola. Ciência 2017. 3-

5 July 2017. Centro de Congressos de Lisboa.

Martin F, Palladino C, Mateus R, Clemente S, Gomes P, Taveira N. Evaluation of an In-

house Molecular HIV-1 Test to Assess Mother-to-Child HIV-1 Transmission in Angola

(the APEHC Cohort). 17th Annual International Meeting of the Institute of Human

Virology, Baltimore, USA, 26th September 2015.

vii
Martin F, Mateus R, Calado R, Diniz AR, Palladino C, Clemente S, Taveira N.

Characterization of the neutralizing antibody responses in HIV-1 infected patients in

Angola. 7th [Link] Postgraduate Students Meeting, Lisbon, Portugal, 16th of July,

2015.

Martin F, Palladino C, Taveira N. Development of a new in-house real-time DNA PCR

assay for the early diagnosis of perinatal HIV-1 infection in dried blood spots. 6th

[Link] Postgraduate Students Meeting, Lisbon, Portugal, 2nd July, 2014.

Martin F, Bártolo I, Carvalho P, Camacho R, Clemente S, Taveira N. HIV-1 genetic

diversity and transmitted drug resistance for the recent introduction of CRF01_AE. 18th

International Bioinformatics workshop on virus evolution and molecular epidemiology,

Gainesville, Florida, USA, 25th August 2013.

Martin F, Palladino C, Taveira N. Development of a new in-house real-time DNA PCR

assay for the early diagnosis of perinatal HIV-1 infection in dried blood spots. 5th

[Link] Postgraduate Students Meeting, Lisbon, Portugal, 2nd June, 2013.

Other publications

The research described below was produced during the PhD, however is not reflected in

the present thesis. The studies were presented in the form of research papers, scientific

book chapters, oral and poster communications in international and national meetings.

The following publications are listed by order of publication date, the latest being the

first listed.

viii
Research papers:

Marcelino R, Gramacho F, Martin F, Brogueira P, Janeiro N, Afonso C, Badura R,

Valadas E, Mansinho K, Caldeira L, Taveira N, Marcelino J. Antibody response against

selected epitopes in the HIV‐1 envelope gp41 ectodomain contributes to reduce viral

burden in HIV‐1 infected patients. Scientific Reports, 2021, 11:8993.

Calado R, Duarte J, Borrego P, Marcelino JM, Bártolo I, Martin F, Figueiredo I,

Almeida S, Graça L, Vítor J, Aires da Silva F, Dias I, Carrapiço B, Taveira N. A Prime-

Boost Immunization Strategy with Vaccinia Virus Expressing Novel gp120 Envelope

Glycoprotein from a CRF02_AG Isolate Elicits Cross-Clade Tier 2 HIV-1 Neutralizing

Antibodies. Vaccines (Basel). 2020 Apr 7;8(2):171.

Cunha‐Santos C, Ricardo P, Perdigao L, Martin F, Gomes Oliveira J, Cardoso M,

Manuel A, Taveira N, Goncalves J. Inhibition of HIV replication through siRNA carried

by CXCR4‐targeted chimeric nanobody. Cellular and Molecular Life Sciences. 2019.

Ezeonwumelu I, Bártolo, Martin F, Abecasis A, Campos T, Romero-Severson E O,

Leitner T, Taveira N. Accidental Father-to-Son HIV-1 Transmission During the

Seroconversion Period. Aids Research and Human Retroviruses, 2018, 34 (10): 857 -

862.

Book Chapters:

ix
Taveira N, Moranguinho I, Martin F. O agente. Manual sobre SIDA – 6ª Edição. Editor

Francisco Antunes e Fernando Maltez. Permanyer Portugal, 2021.

Taveira N and Martin F. O agente. Manual sobre SIDA – 5ª Edição. Editor Francisco

Antunes, Fernando Maltez. Permanyer Portugal, 2018.

Peer-reviewed abstracts published in international journals:

Martin F, Martins A, Maia F, Rocha C, Borrego P, Antunes F, Caldeira L, Valadas E,

Taveira N. Resistance mutations to protease inhibitors in proviral DNA of HIV-2 infected

patients predict response to treatment. Abstract Supplement HIV Glasgow 2018. Journal

of the International AIDS Society 2018, 21(S8):e25187. Abstract_P305.

Martins A, Maia F, Martin F, Rocha C, Valadas E, Antunes F, Borrego P, Taveira N.

Protease diversity and resistance to protease inhibitors of viruses archived in peripheral

blood mononuclear cells of HIV-2 infected patients. Reviews in Antiviral Theraphy &

Infectious Diseases. 2014;(2): Abstract_P35

Gonçalves P, Martin F, Borges P, Espirito Santo M, Taveira N, Marcelino J. Neutralising

and non-neutralising antibodies response in HIV-1-infected individuals from

Mozambique. BMJ global health. 2019 EDC.161: Abstract_A61.

Oral communications:

x
Taveira N, Diniz AR, Borrego P, Martin F, Gomes P, Gonçalves F, Caixas U, Vaz Pinto

I, Barahona, I, Bártolo I. Dolutegravir has a highly potent activity against most primary

HIV-2 isolates includ- ing those that have acquired resistance to Raltegravir. – June 1-5,

2017 – ASM Microbe – New Orleans, USA.

Diniz AR, Borrego P, Martin F, Gomes P, Gonçalves F, Caixas U, Vaz Pinto I,

Barahona, I, Bártolo I, Taveira N. Dolutegravir has a highly potent activity against most

primary HIV-2 isolates includ- ing those that are resistant to raltegravir. – 3-9 May, 2017

– rede SAÚDE 4th Annual Conference, ULisboa Innovation Week – Lisbon, Portugal.

Calado R, Duarte J, Borrego P, Marcelino JM, Wilton J, Bártolo I, Martin F, Barroso H,

Almeida SCP, Graça L, Taveira N. Novel HIV-1 gp120 glycoproteins derived from non-

B subtypes induces tier 2 cross-clade neutralizing antibodies in mice. 4º Congresso

Nacional de Medicina Tropical. 1o encontro lusófono de SIDA, Tuberculose e Doenças

oportunistas. Instituto de Higiene e Medicina Tropical. 19-21 April 2017.

Poster Communications:

Calado R, Duarte J, Borrego P, Marcelino JM, Wilton J, Bártolo I, Martin F, Almeida

SCP4, Barroso H, Graça L, Taveira N. A prime-boost immunization strategy with

Vaccinia virus expressing novel HIV-1 gp120 glycoproteins induces tier 2 neutralizing

antibodies in mice. 9th [Link] Postgraduate Students Meeting and 2nd i3DU

Meeting.13-14 July 2017. Faculty of pharmacy. Lisbon.

xi
Calado R, Duarte J, Borrego P, Marcelino JM, Wilton J, Bártolo I, Martin F, Almeida

SCP4, Barroso H, Graça L, Taveira N. Induction of tier 2 cross-clade neutralizing

antibodies in mice by a prime-boost immunization strategy with Vaccinia virus

expressing novel HIV-1 gp120 glycoproteins from non-B subtypes. Ciência 2017. 3-5

July 2017. Centro de Congressos de Lisboa.

Figueiredo IB, Calado R, Martin F, Borrego P, Cardoso F, Taveira N, Barroso H.

Antibody response to an HIV-1/HIV-2 chimeric envelope glycoprotein in mice. 3a

Semana da inovação. 3-9 May 2017. Reitoria da Universidade de Lisboa (ULisboa).

Poster presentation.

Figueiredo IB, Calado R, Martin F, Borrego P, Cardoso F, Taveira N, Barroso H.

Antibody response to an HIV-1/HIV-2 chimeric envelope glycoprotein in mice. Keystone

Symposia, HIV Vaccines, Keystone, March, 2017.

Calado R, Duarte J, Borrego P, Marcelino JM, Wilton J, Bártolo I, Martin F, Clemente

S, Almeida SCP, Barroso H, Graça L, Taveira N. A novel prime-boost immunization

strategy with Vaccinia virus expressing HIV-1 gp120 derived from non-B subtypes

induces autologous and heterologous Tier 2 HIV-1 cross-clade neutralizing antibodies

in mice. Keystone Symposia, HIV Vaccines, Keystone, March, 2017.

Martins A, Maia F, Martin F, Rocha C, Valadas E, Antunes F, Borrego P, Taveira N.

Protease diversity and resistance to protease inhibitors of viruses archived in peripheral

blood mononuclear cells of HIV-2 infected patients. 12th European Workshop on HIV &

xii
Hepatitis treatment strategies & antiviral drug resistance 26 – 28 March 2014, Barcelona,

Spain.

xiii
Resumo
As metas definidas pela UNAIDS para o controlo da infeção por HIV até 2030, definem

objectivos de reduzir 75% dos novos casos e mortes por HIV até 2020 e 90% até 2030,

em comparação com 2010 [1]. Contudo, estima-se que nenhum país da África Subsariana

atingiu a meta dos 75% de redução dos novos casos de infeção [2]. Inclusive, em Angola

entre 2010 e 2018 a incidência da infeção por HIV-1 aumentou 61,2%. Contudo, as mais

recentes estimativas sugerem que a prevalência parece estar agora estabilizada com

340.000 pessoas a viverem com HIV [2, 3]. As mulheres e crianças estão entre as

populações mais afetadas [2, 3]. A terapêutica antirretrovírica está disponível em Angola

desde 2000 para os doentes infetados por HIV-1, contudo só a partir de 2004 passou a ser

dispensada de forma gratuita. No final de 2020 o número de pessoas a viver com HIV em

tratamento eram 111.168, aproximadamente 33% das pessoas a necessitar de tratamento

[3].

Os objetivos desta tese foram: 1) obter um melhor conhecimento da diversidade da

infeção por HIV-1, da dinâmica de transmissão e resistência aos antirretrovíricos em

Angola; 2) melhorar o diagnóstico da infeção por transmissão vertical em Angola; e 3)

caracterizar a resposta neutralizante contra o vírus, e determinar os factores víricos e do

hospedeiro a ela associada. A epidemia da infeção por HIV é peculiar em Angola, uma

vez que circulam na população formas altamente divergentes do vírus. Dados já

publicados sobre a epidemiologia molecular da infeção por HIV-1 em Angola

demonstram que circulam na população praticamente todos os subtipos puros bem como

vários subsubtipos, várias formas recombinantes circulantes, bem como formas

recombinantes únicas [4, 5]. Esta elevada diversidade genética dos vírus circulantes

espelha uma epidemia antiga, que data da primeira metade do século 20 [6].

xiv
Neste contexto, o primeiro objetivo específico desta tese foi caracterizar a diversidade

genética do HIV-1 em Angola, e a dinâmica de transmissão dos subtipos e formas

recombinantes do HIV-1 presentes no país, bem como conhecer a prevalência de

transmissão de mutações de resistência aos antirretrovíricos, 5 anos após a massificação

da distribuição da terapêutica antirretrovírica (capítulo 2). Para tal, sequenciou-se e

analisou-se filogeneticamente sequências víricas, de 139 amostras de plasma colhidas em

2009, de doentes Angolanos HIV-1 positivos sem experiência prévia de tratamento,

maioritariamente residentes em Luanda. Por forma a determinar as tendências evolutivas

comparam-se estas sequências de 2009 do gene pol com sequências de 2001

anteriormente analisadas[7].

Verificou-se uma diminuição significativa da prevalência do subtipo A de 2001 para 2009

(40,0% para 10,8%, p = 0,0019), enquanto a prevalência de formas recombinantes únicas

aumentou 2 vezes (40,0% para 83,1%, p <0,0001). Em 2009, 47,1% dos vírus eram

subtipos puros (foram identificados todos os subtipos exceto o B), 47,1% eram vírus

recombinantes e 5,8% eram não tipáveis. Os recombinantes únicos mais frequentemente

identificados (U/H) formaram um grupo monofilético altamente suportado, sugerindo

uma origem local e comum destas formas recombinantes. No que diz respeito à

transmissão de mutações de resistência aos antirretrovíricos, neste estudo em 2009,

verificámos uma prevalência muito baixa (0,7%), somente num doente foi identificada a

K103N o que compara com 1,6% de prevalência observada em 2001 [7]. O elevado

número de recentes e pequenos grupos de transmissão que identificámos neste estudo,

bem como a identificação de novas formas recombinantes únicas, são consistentes com

uma epidemia de HIV-1 em crescimento, impulsionada principalmente pela transmissão

heterossexual.

xv
Uma epidemia de HIV em crescimento com elevada diversidade genética coloca desafios

a diferentes níveis, principalmente no diagnóstico, tratamento e prevenção da infeção.

Este facto aliado à elevada taxa de transmissão vertical registada em Angola, 25 ,0% em

2014 [8], estiveram na base do estudo apresentado no capítulo 3, que teve como principal

objetivo o desenvolvimento e validação de um teste molecular qualitativo, sensível e

barato de deteção de DNA pró-viral de HIV-1, em amostras de sangue total periférico

colhidas em papel de filtro, por forma a permitir o diagnóstico da infeção perinatal em

recém-nascidos expostos ao HIV-1. O gene da integrase foi usado como alvo dado ser

um dos genes mais conservados do genoma do vírus. O limite de deteção determinado

por regressão Probit foi avaliado usando diluições limite de plasmídeos recombinantes

contendo o gene da integrase de todos os subtipos de HIV-1 relevantes, a forma

recombinante circulante CRF02_AG e células ACH-2, que têm a particularidade de

conter uma única cópia de HIV-1 integrada no genoma. O teste teve a capacidade de

detetar todos os subtipos de HIV-1 com um limite de deteção de 14 cópias. A

sensibilidade e especificidade clínicas foram avaliadas em 100 amostras de adultos

infetados por HIV-1, 5 amostras de crianças também infetadas, 50 amostras de

voluntários saudáveis e 139 amostras de recém-nascidos de mães Angolanas infetadas

por HIV-1, com dados serológicos aos 18 meses de vida, que foram usados como

referência de infeção. O teste teve também a capacidade de detetar a infeção por HIV-1,

4 semanas após o nascimento. A sensibilidade e especificidade clínicas do método foram

de 100,0% avaliadas nas 139 amostras dos recém-nascidos expostos ao HIV-1. A taxa de

transmissão vertical registada do nosso coorte foi de 2,2% entre Janeiro de 2012 e

Outubro de 2014, que contrasta com a taxa registada a nível nacional em período

semelhante de 25,0%, e espelha os excelentes cuidados materno-infantis prestados no

Hospital da Divina Providência em Luanda, Angola. O baixo custo e a simplicidade do

xvi
teste tornam-no adequado para a implementação em Angola e outros países com recursos

limitados.

O quarto capítulo desta tese teve como principal objetivo a caracterização da resposta

neutralizante contra o HIV-1 em 322 doentes Angolanos infetados e perceber quais os

principais determinantes virológicos e do hospedeiro dessa resposta imunológica. A

indução de anticorpos neutralizantes de largo espectro contra o HIV é considerada de

extrema importância para o controlo da infeção, contudo, até à data nenhuma vacina

testada demonstrou eficácia em induzir este tipo de anticorpos [9-20]. Para caracterizar a

resposta neutralizante usámos um painel de 12 pseudovirus de Env de difícil neutralização

(tier 2) representativos das estirpes que circulam a nível mundial e usámos amostras de

plasma de doentes Angolanos infetados por HIV-1 colhidas em 2009 (n=178) e 2014

(n=58). Amplificámos e determinámos por análise filogenética o subtipo e tropismo do

vírus no gene env, especificamente na região C2V3C3 nas amostras colhidas em 2009

(n= 110 sequências) e comparámos com resultados anteriores determinados em amostras

colhidas em 2001 (n= 96 sequências). Nas amostras de 2014 não foi possível a

amplificação dado que a larga maioria dos doentes estava sob terapêutica antirretrovírica.

Para além da caracterização da resposta neutralizante e tipagem das amostras,

determinámos os possíveis epitopos neutralizantes por comparação com anticorpos

neutralizantes monoclonais de largo espectro anteriormente caracterizados. O título de

ligação contra polipéptidos recombinantes, compreendendo as regiões C2, V3, C3 de

diferentes subtipos de HIV-1, numa subpopulação de doentes com resposta neutralizante

conhecida (n= 48 em 2009 e n= 16 em 2014) foi também avaliado. De seguida explorámos

possíveis associações e correlações entre a resposta neutralizante, o subtipo do vírus, o

título de ligação e as características demográficas e da infeção, da população Angolana

infetada por HIV-1. A análise filogenética permitiu identificar diversos subtipos de HIV-

xvii
1 incluindo o A1, A2, B, C, D, F1, G, H, J, estirpes não tipáveis e formas recombinantes

circulantes. O subtipo A, foi o subtipo puro predominante em ambos os anos de colheita,

2001 e 2009, contudo, o subtipo C aumentou significativamente (2,2 vezes, p= 0,0095)

em 2009. Dos 176 isolados em que estavam disponíveis ambas as sequências da protease

e da região C2V3C3, 74 (42,0%) eram não-recombinantes e 102 (58,0%) eram

recombinantes, sendo que os recombinantes prevaleceram relativamente aos subtipos

puros em ambos os anos. Notavelmente, considerando as 236 amostras em que

caracterizámos o perfil de neutralização, aproximadamente 56,0% dos doentes Angolanos

exibiram respostas neutralizantes de largo espectro. A frequência de doentes com

capacidade de neutralização elite foi elevadada em 2014, quando os doentes estavam sob

terapêutica antirretrovírica e com virémias de baixo nível comparativamente com 2009,

onde a larga maioria dos doentes não tinha experiência prévia de terapêutica

antirretrovírica. Esta resposta neutralizante de largo espectro foi associada ao subtipo C,

idade mais avançada e contagens de linfócitos T CD4+ mais baixas. Verificámos também,

uma forte associação entre a resposta neutralizante e o título de ligação contra os

polipéptidos recombinantes C2V3C3 dos diferentes subtipos. O título de ligação de

anticorpos contra a região C2V3C3 no presente estudo foi um bom indicador do espectro

e potência da neutralização. A resposta neutralizante teve como alvo na maioria dos casos

o super epitopo baseado em N-glicanos na V3, mas anticorpos específicos para o vértice

da V2, o local de ligação do CD4, a região proximal da membrana na gp41 e contra

epitopos desconhecidos, também foram identificados nalguns doentes. As regiões V3 e

C3 foram significativamente menos variáveis e menos sujeitas a seleção positiva nos

doentes com respostas neutralizantes elite em comparação com os doentes com respostas

neutralizantes mais fracas ou ausentes, o que sugere um papel ativo dos anticorpos

neutralizantes de largo espectro, que têm como alvo estas regiões, no controlo da

xviii
replicação e diversificação do vírus. Concluindo, o desenvolvimento de anticorpos de

largo espectro de neutralização contra o HIV-1 requer estimulação de longo prazo e de

baixo nível da região V3C3 do invólucro por parte de isolados de subtipo C altamente

diversos. Estes resultados têm implicações diretas para o desenho de uma nova geração

de vacinas contra o HIV-1.

Palavras-chave: diversidade genética do HIV, epidemia de HIV em Angola, resposta

neutralizante, diagnóstico da infeção perinatal por HIV.

xix
Abstract
The UNAIDS fast-track goals established the need to reduce 75% of new HIV cases and

deaths by 2020 and 90% by 2030, compared to 2010 [1]. However, it is estimated that no

country in sub-Saharan Africa has reached the target of 75% reduction in new cases of

HIV infection [2]. HIV-1 incidence in Angola increased 61.2% from 2010 to 2018 and

the prevalence is now stable with 340,000 people living with HIV [2, 3]. Women and

children are the most affected populations [3]. The aims of this thesis were: 1) to get a

better understanding of the HIV-1 diversity, transmission dynamics and transmitted drug

resistance (TDR) in Angola, 2) improve the early infant diagnosis (EID) in Angola and

3) characterize the neutralizing antibody responses against HIV-1 and assess the possible

viral and host factors associated with it.

The HIV epidemics in Angola is peculiar, since highly divergent forms of the virus

circulates in the population. In this context our first objective was to assess HIV -1

diversity, transmission dynamics and prevalence of transmitted drug resistance (TDR) in

Angola in 2009, in 139 drug naïve HIV-1 infected individuals and compare the results

before ART scale-up (chapter 2). We saw an increase in genetic diversity between 2001

and 2009, with the prevalence of subtype A decreasing significantly while the prevalence

of unique recombinant forms (URFs) increased 2-fold. Also, local U/H recombinants

were newly identified. TDR mutation K103N was found in one (0.7%) patient. Overall,

transmission of drug resistant strains was still negligible in Luanda in 2009 and the

emergence of new URFs are consistent with a rising HIV-1 epidemics. Our second

objective was to develop and validate a sensitive, simple and cheap qualitative proviral

DNA PCR-based assay for early infant diagnosis (EID) in HIV-1-exposed infants

(n=139) using dried blood spots (DBS) (chapter 3). We were able to successfully validate

the assay, the limit of detection (LOD) using several integrase recombinant plasmids was

xx
14 copies and clinical sensitivity and specificity were high. The percentage of HIV-1

mother-to-child-transmission (MTCT) between January 2012 and October 2014 was only

2.2%. In chapter 4 we performed the first detailed characterization of the neutralizing

antibody response in 322 Angolan HIV-1 infected patients and identified its determinants.

Remarkably, 56% of the individuals had broad cross-neutralizing activity. Also, cross-

clade neutralization was positively associated with subtype C infection and negatively

associated with CD4 counts and antibody binding titers against envelope C2V3C3 region

was a good indicator of neutralization breadth and potency. In chapter 4 we concluded

that development of broad and elite antibody neutralization against HIV-1 requires long-

term and low-level envelope V3C3 stimulation from highly diverse subtype C isolates.

Keywords: HIV genetic diversity, HIV epidemic in Angola, neutralizing antibody

response, early infant diagnosis (EID) of HIV

xxi
xxii
Abbreviations
Ab Antibody

Ad Adenovirus

ADCC Antibody-dependent cell-mediated cytotoxicity

ADCVI Antibody-dependent cell-mediated virus inhibition

aLRT Approximate likelihood-ratio test

AIDS Acquired Immunodeficiency Syndrome

APC Antigen presenting cells

ART Antiretroviral therapy

ARV Antiretroviral

BSA Bovine serum albumin

BSL-2 Biosafety level 2

bNAb Broadly neutralizing antibody

CA Conic shaped viral capsid

CD4bs CD4 binding site

CCR5 C-C chemokine receptor type 5

CDC Center for disease control and prevention

CMV Cytomegalovirus

CO2 Carbon dioxide

COVID-19 Coronavirus disease 2019

CRF Circulating recombinant form

CTL T cytotoxic lymphocytes

DC Dendritic cell

DMEM Dulbecco´s minimal essential medium

xxiii
DNA Deoxyribonucleic acid

dsDNA Double stranded DNA

DRC Democratic Republic of Congo

DRM Drug resistance mutations

EID Early infant diagnosis

ELISA Enzyme-Linked Immunosorbent Assay

Fab Antigen-binding fragment

FBS Fetal bovine serum

Fc Fragment crystallizable region

FcRs Fc receptors

FP Fusion peptide

GALT Gut associated lymphoid tissue

GC Germinal Centers

GPCR G-protein-coupled receptor

HIV-1 Human immunodeficiency virus type 1

HIV-2 Human immunodeficiency virus type 2

HLA Human leukocyte antigen

HR1 Heptad Region 1

HR2 Heptad region 2

HTLV Human T-cell leukaemia viruses

Hu-BLT Humanized mice

HuMAbs Human Monoclonal antibodies

ICOS Inducible T cell co-stimulator

IDUs Injecting Drug users

IFN Type I interferon

xxiv
IgG Immunoglobulin

IN Integrase

INIs Integrase inhibitors

IQR Interquartile range

KIR Killer immunoglobulin receptor

LTR Long terminal repeat

MA Matrix protein

ML Maximum likelihood

MPER Membrane proximal external region

mRNA Messenger RNA

MRCA Most recent common ancestor

MSM Men who have sex with men

MTCT Mother-to-child Transmission

NAb Neutralizing antibody

NC Nucleocapsid protein

NHPs Non-human primates

NNRTIs non-nucleoside reverse transcriptase inhibitor

NK Natural killer cells

NVP Nevirapine

OD Optical density

PAMPs Pathogen-associated molecular patterns

PBMC Peripheral blood mononuclear cell

PBS Phosphate buffered saline

PCR Polymerase chain reaction

PEP Post-exposure prophylaxis

xxv
PI Protease Inhibitors

PIC Pre-integration complex

PMTCT prevention of Mother-to-child transmission

PNGS potential-N-linked glycosylation sites

PrEP Pre-exposure prophylaxis

PRRs Pathogen-recognition receptors

PR Protease

RC Republic of Congo

RIP Recombinant Identification Program

RLU Relative light units

RNA Ribonucleic acid

RRE Rev Response element

RT Reverse transcriptase

RTC Reverse transcriptase complexes

SD Standard deviation

SHM Somatic Hypermutation

SHIV Simian-Human Immunodeficiency Virus

SIV Simian immunodeficiency virus

SPF Specific pathogen free

SU Surface glycoprotein

TAR Trans-acting response

TB Tuberculosis

TD Transmembrane domain

TDR transmitted drug resistance

Tfh T Follicular helper cells

xxvi
Tfr T Follicular regulatory cells

TLR Toll-like receptor

TM Transmembrane glycoprotein

TNF- α Tumor necrosis factor α

URF Unique recombinant form

UNAIDS Joint United Nations Program on HIV/AIDS

VLPs Virus-like particles

VSV Vesicular stomatitis virus

VV Vaccinia virus

WHO World Health Organization

Units

ºC Celsius degrees

kb Kilobase

kDa kilodalton

ml milliliter

nm nanometers

μg micrograms

μl microliter

xxvii
xxviii
Table of contents
Acknowledgements ..........................................................................................................ii
Preface ..............................................................................................................................v
Resumo ..........................................................................................................................xiv
Abstract ..........................................................................................................................xx
Abbreviations ..............................................................................................................xxiii
Chapter 1
1. General introduction ...................................................................................................2
1.1 HIV Discovery ........................................................................................................2
1.2 HIV in Sub-Saharan Africa and Angola ..................................................................2
1.3 HIV-1 structure, genome organisation and replication cycle ..................................5
1.3.1 Viral structure and genome organization .........................................................5
1.3.2 Replication cycle .............................................................................................7
1.4 HIV envelope glycoproteins .................................................................................12
1.4.1 The gp120 molecule ......................................................................................12
1.4.2 The gp41 molecule ........................................................................................14
1.5 HIV co-receptor usage and tropism ......................................................................14
2. HIV genetic diversity .................................................................................................15
2.1 HIV classification .................................................................................................15
2.2 Global distribution of HIV subtypes .....................................................................17
2.3 Consequences of HIV-1 genetic diversity .............................................................19
2.3.1 Impacts of HIV-1 genetic diversity on transmission and disease progression
................................................................................................................................19
2.3.2 Impact of HIV-1 genetic diversity on diagnosis ...........................................22
2.3.3 Impact of HIV-1 genetic diversity on antiretroviral therapy..........................23
3. Neutralizing antibodies against HIV infection …....................................................24
3.1 Antibody responses to HIV ...................................................................................24
3.1.1 Non-neutralizing antibodies (ADCC and ADCVI) .......................................25
3.1.2 Neutralizing antibodies (nAbs) .....................................................................26
3.1.3 Broad neutralizing antibodies (bNAbs).........................................................27
3.2 Vaccine development ...........................................................................................30
Aims and work plan ......................................................................................................32
Chapter 2

xxix
HIV-1 Diversity, Transmission Dynamics and Primary Drug Resistance in Angola
.........................................................................................................................................37
Chapter 3
Early Infant Diagnosis of HIV-1 Infection in Luanda, Angola, Using a New DNA PCR
Assay and Dried Blood Spots ..........................................................................................74
Chapter 4
Long-term and low-level envelope C2V3 stimulation from highly diverse virus isolates
leads to frequent development of broad and elite antibody neutraliz ation in HIV-1
infected individuas...........................................................................................................98
Chapter 5
General Discussion and Conclusions ............................................................................141
References ....................................................................................................................153

xxx
xxxi
 Chapter 1

Introduction
 Chapter 1 – General Introduction

1. General introduction

1.1 HIV Discovery

The first cases of patients with the acquired immunodeficiency syndrome (AIDS) were

reported in 1981 in the United States following the observation of opportunistic infections

in young homosexual men [21]. Soon after the first reports in 1983, HIV the causative

agent of AIDS was identified by Luc Montagnier and Françoise Barré-Sinoussi at the

Pasteur Institute (France) [22, 23].

In 1986 HIV-2 was isolated in patients from Guinea-Bissau and Cape Verde Islands

(West Africa) interned at Hospital Egas Moniz in Lisbon (Portugal) [24, 25]. Isolation of

this new virus was possible due to the insightful work of Maria Odette Santos Ferreira at

the Faculty of Pharmacy Universidade de Lisboa [25].

The identification and characterization of HIV was a major achievement, duely

acknowledged by The Nobel Foundation in 2008, with the award of the Nobel Prize for

Medicine to Luc Montagnier and Françoise Barré-Sinoussi. The knowledge built over the

years trying to reduce the burden of HIV-1 infection globally also played an important

role in minimizing the impact of other infectious diseases such as COVID-19 in terms of

diagnosis, transmission, prevention and treatment [26, 27].

1.2 HIV in Sub-Saharan Africa and Angola

Despite recent progress in reducing the number of new infections, 1.5 million [1.0 million

- 2.0 million] people became newly infected with HIV in 2020 globally and HIV/AIDS

is still among the leading causes of disease burden and mortality in sub-Saharan Africa

[28]. The UNAIDS fast-track goals for the control of HIV infection by 2030, define clear

2
 Chapter 1 – General Introduction

and measurable goals for the implementation of public health policies and establish the

need to reduce 75% of new HIV cases and deaths by 2020 and 90% by 2030, compared

to 2010 [1]. However, none of the 44 Sub-Saharan African countries has reached the

target of 75% reduction in new cases of HIV infection in 2020 [2]. Sub-Saharan Africa

remains severely affected by the epidemic accounting for 67% of the people living with

HIV in the world and for 60% of the new infections [28].

Angola is a South-western African country bordered by Republic of Congo, Democratic

Republic of Congo, Zambia and Namibia. According to the UNAIDS report on the global

AIDS epidemic 2013 [29] the estimated HIV prevalence and new infections in adults

have decreased between 2001 and 2012 in all the bordering countries of Angola. For

example, in the Republic of Congo HIV prevalence decreased from 4.7% to 2.8% and the

number of new infections decreased from 6,600 to 3,400. In contrast, the estimated

number of adults living with HIV in Angola has increased in the same period from

110,000 to 220,000 (1.8% vs 2.3% prevalence) and the estimated number of new

infections rose from 16,000 to 23,000 [29]. This increasing tendency continued from 2010

onwards [2]. HIV-1 incidence in Angola increased 61.2% from 2010 to 2018 and the

prevalence is now stable with 340,000 people living with HIV [2, 3]. Women and children

are the most vulnerable populations [3]. Despite the recent advances in prevention of

mother to child transmission (MTCT), Angola reported one of the highest rates of MTCT

(25%) among the 22 priority countries included in the UNAIDS global plan in 2015 [8,

30]. Since then, the incidence of HIV-1 infection in children has been decreasing,

however final vertical transmission rate including breastfeeding was 18.6% in 2020 [8]

and is estimated that 5,200 children (aged 0-14 years) acquired HIV in 2020 [3]. In 2020,

11,000 women aged 15 and over became newly infected by HIV in Angola, more than 2

times the number of man in the same age group [3].

3
 Chapter 1 – General Introduction

The development and availability of antiretroviral drugs helped to significantly improve

the conditions of life of the HIV infected individuals, and especially to increase their life

expectancy [31]. However, due to poverty related barriers, access to antiretroviral therapy

remains limited in Africa and an effective vaccine is hoped to be a practical and cost-

effective intervention to control the spread of the HIV/AIDS pandemic. ART has been

available in Angola since 2000 for those infected with HIV who could afford buying ARV

drugs. Since 2004, a national plan has been implemented to provide free ARV drugs to

HIV-1 infected individuals using the WHO public health approach to ARV delivery [32].

At the end of 2012 the number of people on ART was 39,704 [29], 48% of the adults in

need of treatment based on WHO 2010 guidelines [33]. More recent data shows that the

access to antiretroviral therapy (ART) in Angola is still reason for concern, since at the

end of 2020 the number of people living with HIV undergoing treatment was 111,168,

approximately 33% of those in need of treatment [3]. The rising trend in the number of

people living with HIV in Angola associated with the decrease in mortality due to ART

[3], will inevitably increase the demand for ART in the near future. Increasing ART

coverage of adults and children will inevitably lead to increasing drug resistance and raise

the need for newer antiretroviral (ARV) drugs, such as 2 nd generation Integrase Inhibitors

(INIs) and Protease Inhibitors (PIs) that are not commonly available in Angola [34].

The frequency of transmitted drug resistance (TDR) in Angolan patients has risen from

1.6% in 2001 [7] to 16.3% in 2008–2010 [35, 36] suggesting that TDR may be an

important public health problem in Angola. A more recent observational study registered

a prevalence of 18% of drug resistant mutations (DRM) in 42 HIV-1 infected pregnant

women naive to antiretroviral therapy [37]. Further work is required to characterize TDR

level in Luanda as only a few patients living in this province have been included in

previous surveys.

4
 Chapter 1 – General Introduction

1.3 HIV-1 structure, genome organisation and replication cycle

1.3.1 Viral structure and genome organisation

HIV-1 is a lentivirus belonging to the retroviridae family. A common feature of all

retroviruses is that they synthesize DNA from their RNA genome via the reverse

transcriptase enzyme. The mature HIV virion, is spherical in form with a diameter of

around 120 nanometers (nm). The virion is composed of two identical copies of positive

single-strand RNA together with the reverse transcriptase (RT), protease (PR), integrase

(IN) and RNase H enzymes encapsulated within the viral core (CA; p24), together with

the accessory proteins Nef, Vif, Vpr and Vpu, which is enclosed within the matrix (MA;

p17) [13, 38]. A lipid bilayer that is derived from the membrane of the host cell, as a

result of the process of budding, envelopes the HIV virion, into which the gp41 is

embedded, which in turn anchors the gp120 in a trimer of heterodimers (Figure 1).

Figure 1- Schematic representation of a mature HIV-1 virion illustrating major viral

components. Figure from [Link]

(Consulted in July 2021).

5
 Chapter 1 – General Introduction

HIV-1 virus consists of three structural genes: gag, pol, and env with flanking long

terminal repeat (LTR) sequences at each end of the genome. In addition, HIV-1 possesses

regulatory genes (tat, rev), and accessory genes (vif, vpr, vpu, and nef) (Figure 2).

The gag codes for the internal structural proteins of the virus: matrix (MA, p17), capsid

(CA, p24), and nucleocapsid (NC, p7). The pol codes for the viral enzymes: reverse

transcriptase (RT), which contains both DNA polymerase and associated ribonuclease H

(RNase H) activity, integrase (IN) and protease (PR) [38, 39]. The env codes for viral

envelope glycoproteins as a precursor (gp160), which is then processed to a surface

glycoprotein, gp120 and a trans-membrane glycoprotein, gp41. The mature gp120.gp41

proteins are bound by non-covalent interactions and are associated as a trimer on the

surface of virions. The envelope (Env) protein is responsible for recognition of cellular

receptors and viral entry into cells.

The tat and rev regulatory genes modulate transcriptional and post-transcriptional steps

of virus gene expression and are essential for virus replication and propagation. Tat acts

by binding to the trans-activation response (TAR) RNA element and activates

transcription initiation and elongation from the LTR promoter [40]. Rev acts by binding

to Rev response element (RRE) and promotes the nuclear export, stabilization, and

utilization of the viral mRNAs containing RRE [41, 42]. The genes vif, vpr, vpu, and nef

are termed acessory genes however they play an important role modulating diferent steps

of viral replication. Vif promotes the infectivity but not the production of viral particles.

In the absence of Vif, the produced viral particles are defective, but the cell to cell

transmission of virus is not affected significantly [43]. In addition, Vif prevents the action

of the cellular APOBEC.3G (Apo-lipoprotein B mRNA-editing enzyme, catalytic

polypeptide-like 3G) a potent antiretroviral cytidine deaminase [44]. APOBEC.3G exerts

its antiviral effect during reverse transcription to trigger G-to-A hyper-mutation in the

6
 Chapter 1 – General Introduction

nascent retroviral DNA [45]. Vpr that is incorporated into the virion facilitates the nuclear

localization of the preintegration complexes, and modulate cell division in order to

accelerate the production of HIV proteins by arresting infected cells at the G2 phase of

the cell cycle [46]. Vpx is found in HIV-2, but not in HIV-1. This accessory gene is a

homolog of HIV-1 vpr, however the function in relation to Vpr is not fully elucidated;

both are incorporated into virions at levels comparable to Gag proteins through

interactions with Gag p6 [47]. Vpu is unique to HIV-1 and some SIV (e.g. SIVcpz). It

has two essential biological functions; it degrades CD4 in the endoplasmic reticulum [48]

and promotes extracellular release of viral particles [49]. Nef is important in pathogenesis,

it down-regulates cell surface CD4 molecules, and increases the infectivity of viral

particles [50].

Figure 2- Organization and landmarks of the HIV-1 DNA genome. Open reading frames
are shown as rectangles. The gene start is indicated by number in the upper left corner of
each rectangle (ATG start codon). The number in the lower right indicates the position of
the stop codon. The tat and rev spliced exons are shown in dark grey and grey,
respectively. The numbering positions in the bottom of the figure are relative to HXB2
strain. Figure from Los Alamos National Library,
[Link] (consulted in July
2021)

1.3.2 Replication cycle

The replication cycle of HIV takes place in different phases (Figure 3). Entry is the first

step in the process of HIV infection and requires binding of HIV gp120 to CD4 in host

7
 Chapter 1 – General Introduction

cells [51]. This results in conformational changes in gp120, which then exposes the co-

receptor binding sites. The co-receptors required for entry of HIV-1 are mainly CCR5

and CXCR4, and define viral tropism. Immediately following gp120 and co-receptor

binding, further conformational changes take place in gp41, which allows it to expose the

fusion peptide. The fusion peptide is then inserted into the target cell membrane enabling

further conformation changes in gp41 that drive virus to cell fusion and subsequent entry

into the cells [51, 52]. Cell to cell transmission of HIV is another effective mechanism of

virus proliferation that do not necesseraly involve binding of HIV Env to the host cell receptor

[53].

Following membrane fusion the virus core enters and uncoats into the cytoplasm of the

target cell. The virus RNA genomes are converted into double-stranded DNA. Reverse

transcriptase (RT) has two distinct enzymatic activities: it is a DNA polymerase capable

of copying either RNA or a DNA template into a complementary DNA sequence; and it

is an RNase H, capable of degrading the RNA strand of an [Link] duplex into small

pieces once it has been used as a template for the first DNA strand [54]. The reverse

transcription is initiated, and viral RNA is converted into a DNA/RNA hybrid. The

template RNA is degraded by the RNase H activity and RT polymerase synthetises the

complementary DNA strand to form the double helix DNA molecule. The viral DNA was

thought to be translocated from the cytoplasm to the nucleus as part of the pre-integration

complex (PIC) (viral DNA associated with MA, RT, and IN). However, very recently

new insights about HIV-1 replication cycle have been published showing that HIV-1 CA

might travel intact to the nucleous of the host cell (Figure 4), apparently the diameter of

the nucleous pore is sufficient to allow the passage of undisrupted CA. This disruptive

finding is particularly important since a new drug class of ARVs, capsid inhibitors, will

soon be available [55-57]. The viral capside containing the PIC is transported by the

8
 Chapter 1 – General Introduction

microtubules to the vinicity of the nucleus, in a process that involves the binding to

different nucleoporins (NUP62, NUP153, NUP358) and cleavage polyadenylation

specificity factor 6 (CPSF6). In the nucleus reverse transcription is completed and the CA

and PIC desagregates probably by the mechanic pressure of the newly synthesized viral

DNA. The integration of the viral DNA in the genome of the host cell is mediated by the

viral IN and the cellular cofactors NUP153 and LEDGF/p75. The integration process is

completed when cellular repair enzymes fill in the gaps between the integrated viral DNA

and the host target DNA, and is then designated proviral DNA [58].

Uncoating might also

occur in the nucleus

Figure 3 - Schematic representation of the life cycle of HIV. (Figure adapted from [58])

After integration, transcription is mediated by the host cell RNA polymerase II using

integrated provirus as the template and generates full length viral RNAs that serve as

mRNA. The HIV-1 LTR serves as the site of transcriptional initiation and harbors cis-

acting elements required for RNA synthesis. The transcription initiates at the U3/R

junction that contains binding sites for numerous proteins that participate in the regulation

of the expression of the viral genome (e.g. Sp.1 and two NF.κB binding sites). Successful

transcription leads to the generation of HIV viral transcripts that are derived from a single

full-length transcript by alternative splicing, generating messenger RNA (mRNA) with

9
 Chapter 1 – General Introduction

common 5´and 3´ends [59]. HIV-1 transcripts can be grouped into three different classes:

double spliced mRNA or early transcripts encode early regulatory proteins such as Tat,

Nef and Rev; single spliced mRNA or late transcripts encoding Env, Vif, Vpr and Vpu;

and unspliced and complete mRNA that encode for the polyprotein precursors Gag and

Gag-Pol and are incorporated in the viral particles as genomic RNA [59, 60]. To complete

the expression of the late transcript proteins from the early transcripts, Tat and Rev are

necessary. Tat binds to a short secondary structure located in the R region of the 5’LTR

sequence known as the transactivation response region (TAR). Tat plays a critical role in

up-regulating transcription from the LTR by more than 100-fold. There are two possible

ways that Tat can increase HIV-1 RNA synthesis, one is to augment transcription

initiation and the other one is to improve the activity of RNA polymerase [61, 62]. The

transport of unspliced and single spliced mRNA outside the nucleus to the cytoplasm to

be translated is performed by Rev. This process is mediated by the binding of Rev to the

Rev response element (RRE), a 240 base region of complex RNA secondary structure,

present in the midle of the env gene. Via the nuclear export factor (NES), Rev binds to

the chromosome region maintenance 1 (Crm1) (nuclear export factor) in the presence of

RAs-related Nuclear protein (Ran GTPase) connected to guanisine triphosphate (GTP).

Posteriorly, DEAD-box polypeptide (DDX1) and DEAD-box polypeptide 3 (DDX3) bind

to the N-terminal domain of Rev. The translocation process through the nucleus pore

seems to be mediated by the DDX1 and DDX3 that bind to the nucleoporins. In the

cytoplasm the hydrolyses of GTP induces the release of Rev from the viral mRNA. The

translation of the viral mRNA is determined by the internal ribosome entry segment

(IRES) present in the beginning of the mRNA [62, 63].

10
 Chapter 1 – General Introduction

The HIV Env glycoprotein is synthesized in the rough endoplasmic reticulum (RER) to

generate the Env precursor protein, gp160 and then transported to the Golgi apparatus,

where it is cleaved by a host protease (furin) into gp120 and gp41. The Env protein is

incorporated in the plasma membrane of the cell, while Gag and Gag-Pol are assembled

into viral capsids in the cytoplasm [60, 64]. Two identical copies of the viral genomic

RNA bound by p7 Gag product form a nucleoprotein complex. The newly formed

nucleoprotein particle migrates to the plasma membrane at the site of insertion of Env,

and Gag and Gag-Pol precursors [60]. Immature virus particles are assembled at the

plasma membrane and are released by budding through the plasma membrane, acquiring

a portion of the plasma membrane that contains gp41 and gp120 [62]. Pol, the protease,

and the Gag proteins are generated by proteolytic cleavage mediated by the protease

domain of the Gag-Pol precursor polypeptides upon release of the particles from the cell,

thereby producing mature virus particles [65].

Figure 4- New mechanism of CA nuclear import. Different potential core states can reach
the nucleus: nearly intact (1a), remodeled (1b), or partially uncoated (1c). Figure adapted
from [56]

11
 Chapter 1 – General Introduction

1.4 HIV envelope glycoproteins

HIV-1 Env glycoproteins are assembled as Env spikes. They are responsible for

interacting with cellular receptors and initiating the fusion of the viral and cell

membranes. They are the only viral target of neutralizing antibodies. The functional

envelope spike consists of a trimer of heterodimers formed by two glycoproteins, gp120

(the exterior envelope glycoprotein) and gp41 (the transmembrane glycoprotein)[13].

Three gp120 molecules interact non-covalently with three gp41 units forming an

oligomer, where the trimeric structure is maintained by the interactions between the gp41

domains. The Env trimer oscillates between open and closed conformations, most of the

time the trimer is closed except when it interacts with the host cell receptors[13].

1.4.1 The gp120 molecule

The gp120 molecule consists of five relatively conserved regions (C1-C5), interspersed

between five variable regions (V1-V5), which are delimited by cysteine residues forming

disulfide bonds [66]. Gp120 is a highly glycosylated protein, N-linked glycans account

for half of its mass, with a small proportion being O-linked sugars [13, 67]. The gp120

core is composed of three general areas: the inner domain, the outer domain, and the

bridging sheet (Figure 5). The inner domain is formed mainly by the C1 and C5 regions

which interact with the gp41 trans-membrane unit [68]. The inner domain is devoid of

glycosylation [69]. The outer domain is heavily glycosylated to shield its antigenic

surface, protecting it from antibody recognition. The proximal end of the outer domain

includes V4 and V5 variable loops, whereas the distal end includes the base of the V3

loop, which interacts through hydrogen-bonds with the V1/V2 stem emanating from the

inner domain. In between the outer and inner domains is the bridging sheet region, formed

by four antiparallel β-sheets: β2 and β3, which constitute the stem of the V1/V2 loop; and

12
 Chapter 1 – General Introduction

the β20 and β21 of the C4 region [70]. Besides these structural sites, gp120 has two main

functional sites: the CD4 binding site (CD4bs) and co-receptor binding site. The CD4bs

is a conformational region that is only apparent in the context of the liganded structure of

gp120. The CD4 binding loop projects away from the centre of the outer domain and the

β20-β21 segment of bridging sheet [62]. The α-helices of the inner domain, the CD4

binding loop and the β20-β21 segment of bridging sheet create a long, narrow cavity,

lined principally with hydrophobic side chains, in which many of the residues that are

presumed to contact CD4 are located near or within this long cavity [71]. The binding of

CD4 induces large conformational changes in the inner domain, which leads to the

formation of the bridging sheet and co-receptor binding site [72]. Neither the receptor

(CD4) nor the co-receptor (CXCR4 or CCR5) site is properly formed in the unliganded

conformation of the gp120 core. In the unliganded conformation, the bridging sheet can

close up to create the co-receptor binding surface, which is flanked by the V1–V2 and V3

loops [73]. The V3 loop has been shown as the major determinant of HIV tropism, which

demonstrates its involvement in the co-receptor-binding site [74].

13
 Chapter 1 – General Introduction

Figure 5 - gp120 molecular structure representing the outer domain, the inner domain
and the V1/V2 small domain. CD4 binding site and CCR5 in the base of the V3 are
coloured red and green respectively. Adapted from [62]

1.4.2 The gp41 molecule

The gp41 molecule is a transmembrane glycoprotein that is less variable and less

glycosylated than gp120. It interacts non-covalently with gp120 and is responsible for

maintaining the trimeric structure of the envelope glycoprotein, although its structure in

the native conformation is unknown. The HIV-1 envelope glycoprotein gp41 is a

homotrimeric structure formed by three gp41 monomers, with each monomer non-

covalently associated with gp120. Each gp41 molecule consists of three domains: an

extracellular domain (ectodomain), a transmembrane domain (TMD), and a C terminal

cytoplasmic tail [13, 72].

1.5 HIV co-receptor usage and tropism

Not long after the discovery of the CD4 molecule as the major receptor for HIV [75], new

evidence started to accumulate indicating that CD4 alone was not sufficient for HIV to

enter the target cells. In 1986, Maddon et al. showed that CD4 expressed on mouse cells

allowed the virus to bind but did not confer virus entry [76]. These results supported the

conclusion that this restriction was due to the requirement for a cofactor of unknown

identity that was specific to human cells [77, 78]. Subsequently, several in vitro studies

using CD4+ T cell lines and macrophages showed dif ferent levels of HIV-1 infectivity

depending on the cell lines used, which led to the identification of two main cell co -

receptors that HIV-1 requires to enter the host cell CCR5 and CXCR4. The first HIV-1

co-receptor was identified in 1996 and was named fusin, because it mediated HIV -1

fusion [79] and was thereafter renamed CXCR4 [80]. The second HIV-1 co-receptor was

14
 Chapter 1 – General Introduction

identified based on the finding that the CC-chemokines (i.e. RANTES, MIP.1α, and

MIP.1β) that are the natural ligands of CCR5 could block the infection of certain HIV-1

isolates in primary macrophages [77, 78, 80, 81]. HIV-1 strains able to use both co-

receptors were termed dual-tropic (D-tropic). Both CCR5 and CXCR4 belong to the

superfamily of seven transmembrane (7TM) G protein-coupled receptors. More than

fourteen other 7TM receptors or structural-related molecules have been suggested to act

as co-receptors for entry of HIV-1 in vitro. Currently, there is little evidence to suggest

that co-receptors other than CCR5 and CXCR4 are used significantly in vivo, although

there is a body of literature that demonstrates that HIV-1 employs a number of methods

to get into target cells [82-84].

HIV tropism is now commonly defined based on the co-receptor usage which is defined

as the ability of a particular HIV-1 virus to infect a target cell using a specific co-receptor,

either CCR5, or CXCR4 or both. The major genotypic determinant for HIV-1 co-receptor

usage is the V3 variable loops of the gp120 envelope glycoprotein [85]. Different

bioinformatic tools have been developed to predict HIV-1 co-receptor usage from the

amino acid sequence of V3, taking into account the key amino acids at positions 11 and

25, plus other sites in V3 that differ between CCR5 and CXCR4. The most commonly

known and used is geno2pheno from the Max Planck Institute Informatik and was used

to determine coreceptor usage in chapter 4 of this thesis [86].

2. HIV genetic diversity

2.1 HIV classification

HIV is a member of the lentivirus genus of the Retroviridae family. The name lentivirus

refers to slowly replicating viruses, because these viruses take a long time before it

15
 Chapter 1 – General Introduction

induces the full-blown disease. Two types of HIV have been characterized: HIV-1 and

HIV-2 and both share 40% to 50% genetic homology, with th e greatest sequence

divergence localized in the envelope gene [87]. HIV-1 is thought to be the result of cross-

species transmission of simian immunodeficiency viruses (SIVs) isolated from

chimpanzees (SIVcpz) or gorilas (SIVgor)[88], while HIV-2 is most closely related to a

virus found in sooty mangabeys (SIVsm) [89].

HIV-1 is divided into group M (main), the group O (outlier), the group N (non -M/ non-

O), and the more recently identified group P. Group M is responsible for the worldwide

HIV-1 epidemic, the other groups represent a minority of HIV-1 strains and don’t have

global expression. HIV group M is classified into 10 subtypes: A, B, C, D, F, G, H, J, K

and more recently identified subtype L[90] and 8 subsubtyes (A1-A6 e F1-F2). [91, 92].

Studies have shown that intra-subtype genetic distance can differ by up to 20% in the env

gene and inter-subtype genetic distances can reach up to 35% [92] [13]. Sequencing full-

length genomes have led to the identification of inter-subtype recombinants know as

circulating recombinant forms (CRFs) [93]. These are presumed to be the result of

recombination between different subtypes within an individual patient concurrently

infected with HIV-1 of two or more subtypes. The inter-subtype recombinant genomes

become designated as CRFs if; i) the identical recombinant viruses are identified in at

least three epidemiologically unlinked people, ii) are characterized by full-length genome

sequencing that share the same recombinant structure, and iii) form a monophyletic

cluster in all regions of the genome; and as URFs, if only one or two sequences are

available [94]. Recombinants are currently estimated to be responsible for at

approximately 22% of HIV-1 infections worldwide [93]. The CRFs are named with a

number sequential in the order in which they are reported in the literature and followed

by the letters of the subtype involved, starting with CRF01_AE. If the recombinants

16
 Chapter 1 – General Introduction

consist of more than two subtypes involved, they are therefore replaced by designation

“cpx”, meaning complex, e.g. CRF04_cpx (A, G, H, K, and U). Taxonomically, the CRFs

are at the same level as the subtype [93]. Currently, more than 100 CRFs for HIV-1

(CRF01 to CRF102) and only one for HIV- 2 (HIV2.CRF01_AB) are found in the HIV

database at Los Alamos National Laboratory

([Link] most of them

having been described in Africa. Some CRFs are major strains circulating in certain

regions and responsible of for more recent epidemics. For example CRF0l_AE and

CRF02_AG are dominant in Thailand, Asia and West Africa respectively [93].

2.2 Global distribution of HIV subtypes

HIV-1 genetic subtypes are unevenly distributed in different geographical locations.

According to recent studies, the most prevalent HIV-1 subtypes are subtypes A, B and C

[93]. Subtype C accounts for nearly half (48%) of all HIV-1 infections in 2010-2015,

while subtypes B, A, G and D account for 12.1%, 10.3%, 4.6% and 2.7%, respectively

[92]. The subtypes F, H, J, and K all together account for approximately 1% of infections.

The circulating recombinant forms CRF02_AG and CRF01_AE are responsible for 7.7%

and 5.3% of cases respectively. Other CRFs account for the remaining 3.7% of infections.

All recombinant forms (all CRFs and URFs) are responsible for over 20% of inf ections

worldwide [93]. The global distribution and prevalence of HIV-1 worldwide are shown

in Figure 6. This distribution reflects the present situation and might be susceptible to

modifications in the next years. Subtype A viruses are predominant in Central and Eastern

Africa (Kenya, Rwanda, Uganda, and Tanzania) and in Eastern European countries

formerly constituting the Soviet Union. Subtype B, which is the most widely

disseminated subtype, is predominant in North and Latin America, the Caribbean,

17
 Chapter 1 – General Introduction

Europe, and Australia. It is also common in several countries of Southeast Asia, North

Africa, Middle East (Israel), and among South-African and Russian homosexual men.

Subtype C is predominant in southern Africa, Ethiopia and India. Subtype D viruses are

found principally in East Africa and to a lesser extent in West Africa. CRF01_AE and

subtype B co-circulate in South-East Asia, whereas CRF02_AG, along with other

recombinants, dominates in West and West Central Africa. In South America, the

epidemic is a mixture of subtype B and BF recombinants, with a small proportion of

subtype C infections. In East Asia subtypes B, C and BC recombinant strains dominate.

Central Africa harbors a complex mixture of rare subtypes (F, G, H, J and K) and

recombinants, without any predominant strain. [91-93].

Figure 6 – Global distribution of HIV-1 pure subtypes and CRFs. The pie charts in the
map represent the distribution of HIV-1 subtypes and CRFs from 2014 to 2017 for each
region. The dimension of the pie charts are proportional to the number of people living
with HIV in that region. The colours representing the dif ferent subtypes and CRFs are
indicated in the legent on the left. In the bottom right is represented a phylogenetic tree
showing the more prevalent subtypes and CRFs in the world based in the HIV-1 pol gene
sequences obtained from
([Link] adapted from [95].

18
 Chapter 1 – General Introduction

Angola is one of the countries with the most diverse HIV epidemic in the world [91]. In

Angola the HIV-1 epidemic is highly complex with all HIV-1 subtypes, several circulant

recombinant forms (CRFs), unique recombinant forms (URFs) and untypable (U) strains

reported [4, 5, 7, 34-37, 96-99]. This genetic complexity poses significant challenges to

laboratory diagnosis, antiretroviral treatment (ART) and effectiveness of prevention

strategies [15, 34, 37, 100].

2.3 Consequences of HIV-1 genetic diversity

2.3.1 Impacts of HIV-1 genetic diversity on transmission and disease

progression

Earlier cohort studies found an association between CRF01_AE and heterosexual

transmission as well as between subtype B and intravenous drug use [101, 102]. However

later on, contrary results were published from a longitudinal study performed in Thailand

that found an increased probability of CRF01_AE transmission among IDUs compared

with subtype B [103]. A more recent study performed in HIV-discordant couples in

Uganda found that subtype A was associated with a significant higher rate of heterosexual

transmission than subtype D [104]. The rate of transmission may be related to differences

in subtype-specific co-receptor tropism. The R5 tropic HIV strains are more frequently

transmitted than strains that use CXCR4, however X4 tropic viruses emerge later in

infected patients and are associated with more rapid disease progression [105]. It was

found that HIV-1 subtype D was more prone to use CXCR4 in early infection which may

in part explain their reduced heterosexual transmissibility when compared to other genetic

forms [104, 106, 107] whereas subtype A mostly used CCR5 even in late infection. This

may explain why HIV-1 subtype D infected patients had more rapid progression that

those infected with subtype A in Uganda, Kenya, and Tanzania. The percentage of

19
 Chapter 1 – General Introduction

CXCR4 viruses appears lower in subtype C than in subtype B, even when the viruses are

obtained from patients with advanced AIDS [108]. A previous study in Tanzania

suggested that subtypes A and C are more likely to be perinatally transmitted than subtype

D [109], and that pregnant women infected with subtype C were more frequently

susceptible to transmit HIV to their children than those infected with subtype B [110].

Also a recent publication from the Swiss HIV cohort study showed a substancial increase

in transmission of non B-subtypes between 1990 and 2019 among men who have sex with

men (MSM) [111].

A number of studies have suggested differences in the rate of disease progression

associated with infection by different HIV-1 subtypes [112]. The best studied of these are

subtypes A, D and C, which co-circulate in several African countries [15, 113-115]. Large

cohort studies addressed the issue of the impact of HIV-1 subtype on disease progression

[15, 115, 116]. Most of these studies point to higher virulence of subtype C strains in

comparison to other HIV-1 subtypes [15, 115-117]. One of those studies an European

seroconverter cohort showed that HIV-1 subtype significantly influenced CD4 count at

seroconversion and rate of decline, subtype C-infected participants had the lowest CD4

levels at seroconversion [116]. A multinational clinical trial PEARLS (ACTG A5175)

has shown that HIV-1 subtype C infection was associated with higher virological risk

failure compared to subtype B and other non-subtype B infections [117]. A large sub-

Saharan African cohort also found that subtype C and subtype D-infected participants had

faster disease progression in comparison to subtype A [15]. Moreover, a retrospective

cohort study (1996-2007) reported that African patients infected with HIV-1 non-B

subtypes had slower rates of disease progression compared to Haitians and Canadians

infected with subtype B viruses [118]. Another study showed that Senegalese woman

infected with non-A subtypes were 8 times more likely to develop AIDS than those

20
 Chapter 1 – General Introduction

infected with subtype A [119]. Also, a study of a Kenyan cohort showed that patients

infected with subtype D had a higher mortality rate and a faster decline in CD4+ count

than those infected with subtype A or C [120]. The propensity of subtype D to exhibit a

greater degree of using dual co-receptor than other subtypes [106] may help to explain

the observation that subtype D appears to be associated with a more rapid rate of disease

progression than other subtypes.

A very recent study, conduct in sub-saharan Africa comparing the viral replication

capacity of viruses isolated from HIV-1 infected patients from East and West Africa,

found that the viruses isolated from West African patients had higher replication capacity

in a subtype dependent manner and that CRF02_AG was associated with lower

replication capacity due to the presence of a specific polymorfism in Gag [121].

An important unanswered question is the biological basis for these differences. A possible

clue comes from data suggesting that emergence of X4 variants is less common in HIV-

1 subtype A infection, compared with HIV-1subtype D infection, but the differences were

not statistically significant, and the cross-sectional design precluded analysing the effect

of coreceptor usage on the risk of disease progression [113]. It is important to explore the

relationship between emergence of X4 virus and subsequent disease progression in other

cohorts with different viral subtypes in circulation. A recent study analysing the

determinants of HIV-1 subtype C pathogenesis found that R5 tropic subtype C virus

strains in the absence of co-receptor switch have the ability to use alternative coreceptors

in vitro during the course of infection, in particular FPRL1 [122], which may account for

the apparent higher virulence of subtype C strains. However, finding a cause effect in all

of these cohort studies is very difficult to measure since there are several other factors

that can impact transmission and disease progression besides viral diversity, e.g.

behavioral, epidemiological and immunological factors [123].

21
 Chapter 1 – General Introduction

2.3.2 Impacts of HIV-1 genetic diversity on diagnosis

HIV-1 fourth-generation immunoassays are able to detect all known HIV-1 group M

subtypes, group O and HIV-2 positive samples with very high sensitivity and specificity.

In contrast, the results obtained from antigen-antibody rapid tests are far from satisf actory

with more diverse HIV strains [124]. These fourth-generation immunoassays and the ultra

sensitive 5 th generation assays detecting Ag p24 provide an advantage for detection of

infection during the window period prior to seroconversion since the diagnostic window

may be reduced by an average of 5 days relative to an IgM-sensitive EIA [124, 125].

However, such advanced assays are often not available in resource limited countries

where most new infections occur. In field situations with a high diversity of circulating

HIV strains, such as in Angola, the performance of rapid diagnostic tests is much less

satisfactory with sensitivities ranging from 94.1 to 100% and specificities ranging from

88.0% to 98.8% [126]. Also, early antiretroviral therapy diminishes the production of

antibodies which compromises the performance of serological tests [127, 128]. PCR-

based assays for viral load measurements, also have difficulty detecting and reliably

quantifying HIV-1 RNA when testing diverse genetic variants of HIV-1 from Africa, and

different assays frequently yield discordant viral load results [129-131]. Furthermore,

relative to mother-to-child-transmission (MTCT), serological assays do not allow the

early diagnosis of HIV-1 infection because of the persistence of maternal HIV-1

antibodies in infants during the first 12-18 months of life. The WHO recommends the use

of molecular-based virological testing to determine the infection status for HIV-1-

exposed infants during the first 4-6 weeks of life or at the earliest opportunity thereafter

[132]. Despite the high accuracy of fourth-generation and HIV RNA tests, their sensitivity

could potentially be affected in settings of expanded ART for prevention of MTCT

(option B and B+), which reduce circulating HIV-1 RNA and viral particles [133].

22
 Chapter 1 – General Introduction

Qualitative DNA PCR test which detect proviral DNA in peripheral blood mononuclear

cells (PBMCs) is therefore recommended for early infant diagnosis (EID) of HIV-1 and

is the most widely implemented test in resource-limited settings [134, 135].

2.3.3 Impact of HIV-1 genetic diversity on antiretroviral therapy

The development of resistance to antiretroviral drugs continues to be an important

problem in the treatment of HIV-infected individuals. Studies around the world have

demonstrated that different HIV-1 subtypes have similar susceptibilities to currently used

antiretroviral drugs, which were originally developed based on subtype B viruses [136,

137]. However, viral subtype can be associated with treatment virological failure [117].

Different HIV genetic forms carry in their genomes genetic signatures and

polymorphisms that could alter the structure of viral proteins which are targeted by drugs,

thus impairing ART efficacy. Several mutations are generally required for the virus to

become resistant to protease inhibitors (PI) and integrase inhibitors (INIs), whereas a

single amino acid substitution can induce resistance to the non -nucleoside reverse

transcriptase inhibitor (NNRTIs) [138]. The NNRTI resistance mutation V106M occurs

in subtype C and CRF01_AE, but not in subtype B, and the protease inhibitor (PI)

mutation L89I/V occurs in subtypes C, F and G, but not in B [139]. Among non-B

subtypes differences are also notable, as nevirapine resistance mutations developed more

frequently in subtype D than A in a mother-to-child transmission prevention study using

single-dose nevirapine [140]. The probability of selecting resistance varies between HIV-

1 subtypes. There are specific drug resistance mutations pathways characteristic of certain

HIV subtypes which might pose a major challenge in epidemics driven by highly

divergent HIV strains, such as in Angola. In fact, a small surveillance study conducted in

Angola recently showed a high prevalence (18%) of drug resistance mutations in HIV-1

infected pregnant women [34, 37].

23
 Chapter 1 – General Introduction

3. Neutralizing antibodies against HIV infection

3.1 Antibody response to HIV

Following acute HIV infection an abundance of antibodies are elicited. Antibodies have

the ability to inhibit HIV-1 infection through multiple pathways: they can bind cell-free

virus and prevent the infection or they can complex with Fcδ receptor to block HIV-1

through effector cell mechanisms [13]. The progression of these antibodies include: i)

binding antibodies that first develop within 8 days after plasma virus detection and

initially exist as antigen–antibody complexes, followed a few days later by circulating

anti-gp41 antibodies, and further few weeks later with anti-gp120 antibodies targeting

mostly the V3 loop [141]; ii) non-neutralizing antibodies that act together with innate

immune cells to kill virus infected cells known as antibody dependent cell-mediated

cytotoxicity (ADCC) and antibody dependent cell-mediated viral inhibition (ADCVI)

[142]; iii) neutralizing antibodies (Nabs). Neutralizing antibodies aim to block viral entry

and subsequent infection by binding to exposed regions on the envelope. It is generally

recognised that a successful HIV-1 vaccine should elicit potent and broadly neutralizing

antibodies similar to those found in some HIV-1 controllers [13, 143, 144]. Such

antibodies may protect patients from disease progression and can neutralize a wide range

of genetically diverse HIV-1 subtypes [13].

3.1.1 Non-neutralizing antibodies (ADCC and ADCVI)

24
 Chapter 1 – General Introduction

The HIV-1 Env glycoprotein is highly immunogenic but the antibodies elicited by it

during infection are generally either non-neutralizing (nnAbs) or lack neutralizing

breadth or potency against primary HIV-1 strains thus failing to inhibit viral replication

in infected individuals [145, 146]. Nonetheless, non-neutralizing antibodies can clear the

virus by binding to the infected cells and initiate the recruitment of activated effectors

cells, which in turn induces cytolysis or apoptosis of infected cells. ADCC is the result of

the formation of a complex between the IgG Fab portion of the antibody with the viral

protein on the cell surface and binding of the Fc portion to the Fc receptors of the antibody

[142]. Fc receptors are expressed on natural killer cells, monocytes, macrophages,

dendritic cells and neutrophils [142]. The binding to the Fc receptors can lead to release

of antiviral cytokines [147] resulting in the killing of the infected target cell [148].

ADCVI also involves the interaction between a target cell, ADCVI antibody and an

effector cell. However, rather than causing cell death, ADCVI antibodies aim to reduce

the viral output from infected target cells [149].

Vaccine studies in both humans and non-human primate model systems have brought

some evidence that non-neutralizing antibodies may provide protection from infection.

In fact, in human vaccine studies the few correlates of protection identified until now, in

the only HIV-1 vaccine clinical trial (RV144) that demonstrated some level of protection

against HIV-1 infection, were related to non-neutralizing antibodies to V1/V2, high

levels of antibody-dependent cellular cytotoxicity (ADCC) after controlling for IgA, and

HIV-1-specific IgG3 responses [9, 10, 12, 13, 117, 145, 150-154].

3.1.2 Neutralizing antibodies (nAbs)

In HIV-1 infection, nAbs can block the virus-cell interaction by inhibiting the binding of

the virion to CD4 and co-receptors on the cell surface, therefore preventing

25
 Chapter 1 – General Introduction

conformational changes of the virus envelope that are required for subsequent steps in the

virus life cycle. The earliest neutralizing antibodies can be detected within months of

infection in most HIV-1 infected individuals [155, 156]. Most HIV-1 infected individuals

produce antibodies capable of inhibiting their own virus (autologous virus) and are known

as autologous neutralizing antibodies [155-158]. These antibodies target immunogenic

exposed regions of the HIV-1 virion; however their neutralization capacity is transient

and have little impact in the control of the infection already established due to the

continuous capacity of HIV-1 to diversify and escape recognition of these antibodies [13,

159-163]. The appearance of neutralization escape variants soon after the autologous

response supports the notion that these antibodies exert immunological pressure on the

virus [155, 164]. Autologous antibody response primarily target variable regions rather

than the conserved regions of the HIV-1 envelope, explaining the strain specificity of

these antibodies [165, 166]. The V1V2 region was therefore shown to be a frequent target

of autologous nAbs in HIV-1 and Simian–Human Immunodeficiency Virus (SHIV) [167]

[168]. A study reported on the detection of autologous nAbs as early as 52 days after

detection of HIV-specific antibodies in acutely infected patients [155]. Gray et al, have

evaluated autologous and heterologous neutralizing antibody responses in 14 HIV -1

subtype C acutely infected individuals [156]. Their results revealed that potent autologous

neutralizing antibodies are produced within 3 to 12 months post-infection with an increase

in autologous antibody production observed within the first 6 months [156]. Another

study identified the C3V4 region, specifically the C3 α.2-helix as a major target of

autologous neutralization antibodies in subtype C infections [168]. V4 does not appear to

be a significant autologous nAb target, although changes in this region may mediate

neutralization escape [169] while the role of V5 is less clear. This continuous

interdependent cycle of neutralizing antibodies driving the selection of escape mutants,

26
 Chapter 1 – General Introduction

that in turn drive the maturation of dif ferent B-cell percursors selecting new antibodies,

might increase the probability of eliciting broad neutralizing antibodies (bNAbs).

3.1.3 Broad neutralizing antibodies (bNAbs)

In contrast to the early autologous Nab responses, antibodies capable of neutralizing

heterologous viruses (heterologous neutralization) develop later in infection and can be

remarkably potent and cross reactive [13, 155, 156]. Only a small percentage of

chronically infected patients can develop broadly cross-reactive nAbs against multiple

HIV-1 viruses [146, 170-172]. The reasons why some individuals develop broadly cross-

reactive nAbs is still not fully understood, however some determinants of development

of bNAbs were already identified such as viral diversity, duration of infection and viral

load levels [146, 173]. Broadly cross-reactive nAbs response was initially thought to be

quite uncommon [174], but recent studies have described the presence of broadly cross-

reactive nAbs in different cohorts [173, 175, 176]. These studies indicate that such

antibodies are not rare [173, 175, 176], providing evidence that the natural B cell response

can generate broadly cross-reactive nAbs against HIV-1. This type of antibody response

could be the key to stop or at least contain HIV-1 transmission if they could be induce by

vaccines prior to virus exposure [159-161], since some bNAbs have showed the ability to

prevent and suppress HIV-1 infection in Humans [177-182]. However, the effectiveness

of antibodies elicited by candidate immunogens and vaccines to neutralize heterologous

primary HIV-1 strains in vitro and in vivo tested so far has been very limited [9-14].

Broadly neutralizing monoclonal antibodies against HIV-1 have been identified and

studied comprehensively over the last years [10, 15, 136, 150, 151, 159-161, 172, 183-

212]. To date more than one hundred bNAbs against HIV-1 have been produced

([Link] bNAbs target six

27
 Chapter 1 – General Introduction

indentified epitopes in the HIV-1 envelope: the CD4 binding site (CD4bs); the V1/V2;

V3 glycan; gp41 MPER, gp41 and gp120 interface residues and the fusion peptide [13,

213, 214] (Figure 7).

Figure 7- Molecular structure of the HIV-1 envelope gp160 showing the functional
domains and neutralizing epitopes. Model based on the glycosylated BG505SOSIP.664
trimer. PDB ID 4ZMJ. gp120 is coloured in olive green and gp41 in gray. The six bNAbs
target epitope regions are highlighted in different colours. V1/V2 in pink (V1 dark pink,
V2 light pink); V3/V4 in blue (V3 dark blue, V4 light blue); CD4bs in red; fusion peptide
in light green; gp120 and gp41 interface residues in magenta; MPER region in dark green.
Example of bNAbs targeting the specific regions are given. Adapted from [62]

These new bNAbs were derived from donors infected with different HIV-1 subtypes and

the success of this effort was based on the combination of three strategies: a) the selection

of chronically infected individuals with potent and cross-subtype reactive serum

antibodies; b) the use of novel selection of screening approaches; and c) the development

of efficient methods to isolate human monoclonal antibodies [13].

28
 Chapter 1 – General Introduction

However, the mechanisms underlying the elicitation of such antibodies by B cell

populations has just started to be unveiled [2, 14, 26, 55, 56, 111, 121, 215-228]. Guiding

the immune system to elicit such bNAbs remains a major challenge due to extremely

complex antibody maturation pathways and high levels of somatic hype rmutations

(SHM) required by bNAbs to acquire neutralization breadth [215].

HIV-1 has developed multiple escape mechanisms to avoid neutralization. Such features

include the inaccessibility of relevant epitopes due to the trimeric structure of envelope,

shielded envelope glycoproteins by glycans which makes them recognised as host-

derived; conformational masking of receptor-binding sites; and hypervariable loops that

are exposed but capable to change Nab epitopes, nucleotide substitutions, insertions and

deletions, modification of envelope [9, 10, 155, 203, 229], temporary epitope exposure

and non-functional envelope spikes which are not expressed by mature functional spikes

that may deviate the immune response from functional targets [230]. HIV envelope is

heavily glycosylated, with almost 50% of the total mass consisting of poorly or non-

immunogenic glycans [66, 231] shielding antibody access to the epitope. Also, the

unfavourable stoichiometry with very limited number of gp160 glycoproteins per virion

(between 21-42 SU molecules or 7-14 trimers per particle) likely reduce the ability of

antibodies to bind simultaneously to two Env molecules (bivalent antibody binding) [232,

233].

Passive immunization of primates challenged with chimeric SHIV strains has shown that

human bNAbs can protect against infection and are effective against intravenous [234,

235], oral [236] or intravaginal challenges [235, 237]. Passive administration of bNAbs

has been shown to protect humanized mice and macaques against high-dose challenge

with HIV or SHIVs viruses [234, 238, 239]. In addition, bNAbs have been shown to have

an active role in therapy contributing to the decline of plasma viremia to undetectable

29
 Chapter 1 – General Introduction

levels in HIV-1 infected humanized mice and SHIV-infected macaques, especially when

combined with other bNAbs and/or ART [240, 241]. Recently, Nussenzweig et al. have

demonstrated that the passive administration of HuMAb 3BNC117 was safe and effective

in reducing viral load in HIV-1 infected individuals [177, 242]. Also, the potential use of

bNAbs as long acting agents for the prevention and treatment of HIV-1 infection when

used together with other ARVs is being explored [224]. Thus, bNAbs can be used to

prevent HIV-1 infection but also disease progression.

3.2 Vaccine development

Eliminating HIV from the human population will require a successful vaccine. However,

after 40 years of HIV-1 pandemic, no vaccine exists for clinical use to prevent HIV-1

infection. Candidate vaccines evaluated to date have either failed or have shown very

modest efficacy. The reasons are multiple and include the remarkable high HIV-1

diversity; and the host’s inability to mount antibodies to targets within conserved

envelope regions that confer broad neutralization [13]. Recent insights of ways that

vaccines can potentially stimulate protective T- and B-cell immunity, the identification

of new targets for bNAbs, and the discovery of new mechanisms of host control of HIV-

bNAbs induction offer renewed hope for the development of a well-tolerated and

effective preventive HIV-1 vaccine [13, 144].

A new generation of optimized immunogens that mimic the native Env trimer such as

BG505 SOSIP.664 gp140 [9, 10, 15-20], capable of expressing multiple epitopes for

bNAbs including quaternary epitopes, may lead to the development of an effective HIV-

1 vaccine. However to date such trimers failed to consistently elicit bNAbs capable of

heterologous neutralization of tier 2 viruses (the tier of the majority of the primary HIV-

1 strains) [9, 10, 12, 15-20].

30
 Chapter 1 – General Introduction

Traditional vaccine strategies have focused on live attenuated viruses, whole killed

viruses and protein subunits [243, 244]. These approaches have proven successful against

other viruses such as the influenza virus but raise great safety concerns with regard to

HIV-1 [245, 246]. More recent vaccine strategies have made use of gene delivery

technologies such as plasmid DNA vaccines, recombinant viral vectors (attenuated or

replication-incompetent viruses) such as adenoviruses, poxviruses and Vaccinia virus

[14, 212, 247-249].

To date more than 100 clinical trials have been performed in last 30 years to evaluate HIV

vaccine candidates. So far, only seven HIV-1 vaccine candidates have completed efficacy

trials and none has succeeded in inducing bNAbs, as noted in the Thai Phase III clinical

trial (RV144) conducted in Thailand in 2009 [13, 144]. The STEP study conducted in

America, Caribbean and Australia, which comprised of a recombinant adenovirus vector

that expressed HIV-1 subtype B gag, pol and nef genes unexpectedly was brought to an

early end due to safety concerns [247] and more recently the clinical trial (HVTN 702)

was stoped due to lack of efficacy [14].

In summary there are several obstacles for the development of an effective HIV vaccine,

i) the high genetic diversity of the virus due to the lack of proofreading ativity of the RT,

ii) the early and rapid destruction of the cellular immune system, iii) the early

establishment of latent viral reservoirs in different cells and compartments, iv) the unclear

immune correlates of protection, v) safety concerns regarding the use of attenuated

viruses in humans, vi) the conformational changes of the Env trimer (open and closed),

vii) the occlusion of neutralizing epitopes in the outer surface of the virus and also the

difficulty in exposing these epitopes in vaccine candidates [13, 144].

Nevertheless, the efforts for finding an effective HIV vaccine continue, the major

strategies that are being pursued include different approaches, passive immunization with
31
 Chapter 1 – General Introduction

bNAbs and induction of broad nAbs with vaccine candidates, trying to augment the

quality and the quantity of non-neutralizing V1V2 antibodies as seen in the RV144

immune correlates study, the development of vaccine vectors that better represent critical

T-cell epitopes and the viral diversity of circulating strains [13] and mRNA vaccines that

proved efficacious against certain variants of SARS-CoV-2 [250].

Aims and work plan

For more than 30 years our group and other investigators from the Faculty of Pharmacy,

Universiade de Lisboa, have been studying the HIV epidemics in Angola, in terms of

seroprevalence, genetic diversity, antiretroviral drug resistance surveillance and

characterization of new recombinant forms and full genome sequences [4, 5, 7, 99, 251,

252]. However, in an epidemic constantly being fueled by new infections, where highly

divergent forms of the virus circulate, it is important to monitor the evolution of the

epidemics and continue to further characterize the impact of this rising HIV epidemics

on transmission, TDR, diagnosis and disease progression.

In the 2 nd chapter of this thesis the main goal was to further characterize the HIV-1

epidemic in Angola by assessing the HIV-1 diversity and prevalence of transmitted drug

resistance (TDR) in Angola in 2009, five years after ART scale-up and determine the

prevalent transmission dynamics.

First we extracted, amplified, sequenced and subtyped by phylogenetic analysis HIV-1

pol sequences from plasma samples collected in 2009, from 139 HIV-1 infected Angolan

patients naïve to antiretroviral therapy residing in Luanda. To check for differences in

viral subtype and diversity and in order to analyse the evolutionary trends we compared

our reults with a similar survey that our group performed in 2001[7]. Resistance mutation

32
 Chapter 1 – General Introduction

analysis was performed using the Stanford genotypic resistance interpretation algorithm.

Mutations specifically associated with transmitted HIV-1 drug resistance were analyzed

with the Calibrated Population Resistance Tool (CPR). For performing the transmission

network analysis we extended the study population to all other Angolan patients for which

pol sequences were available in the Los Alamos HIV Sequence Database (n=364), in

order to avoid overestimation of relatedness between sequences due to the use of scarce

data.

Women and children are the most affected populations by HIV/AIDS in Angola [3]. HIV-

1 MTCT is the main route of infection among the pediatric population. In 2014 alone

more than 4,500 children acquired HIV-1 through mother-to-child transmission (MTCT)

[8] and is estimated that 5,200 children acquired HIV in 2020 [3]. Early infant diagnosis

(EID) of HIV-1 infection and early antiretroviral treatment (ART) initiation reduces HIV-

1-related mortality and long-term morbidity as well as the size of the HIV-1 reservoirs

[253, 254]. Molecular-based virological testing enables determination of the infection

status of HIV-1-exposed infants in the first days or weeks of life. However, the high cost

of the commercially available tests and the relatively poor performance with dried blood

spots (DBS) and highly diverse isolates have hampered its implementation in Angola.

In the 3 rd Chapter of this thesis the main goal was to develop and validate a sensitive,

simple and low-cost qualitative DBS-based HIV-1 DNA PCR for early infant diagnosis

of HIV-1 infection in Angola and potentially other less resourced countries. We started

by selecting a specific pair of primers targeting a highly conserved region in the integrase

of HIV-1 with the capability of paring with all HIV-1 genotypes prevailing in Angola.

We then constructed control plamids containing the integrase gene of all prevailing HIV-

1 subtypes in Angola in order to determine the analytical sensitivity of our method. We

performed serious dilutions of these control plasmids and ACH-2 cells, which contain a

33
 Chapter 1 – General Introduction

single copy of HIV-1B provirus per cell, in seronegative blood and spiked filter papers.

Clinical sensitivity was determined on dried blood spots (DBS) samples from 100 HIV-

1-infected adult patients, 5 local samples of HIV-1 infected infants, 50 healthy volunteers

and 139 HIV-1-exposed infants of the Angolan Pediatric HIV Cohort (APEHC) with

serology at 18 months of life. For extracting the viral DNA from all DBS we used a simple

and inexpensive method based on the cationic resin Chelex. For the detection of the viral

DNA we used a nested PCR to amplify a small (194 bp) fragment in the integrase region

and the human gene C-C chemokine receptor 5 (CCR5) was used as an internal control

to confirm the presence and quality of genomic DNA. All samples were screened in

triplicate, using HIV-1 serology results at 12 months as diagnostic reference.

Considering that vaccine effectiveness will depend on the extent to which induced

antibodies will be capable to neutralize the global diversity of circulating HIV-1 variants,

studying HIV-1 epidemics like the Angolan makes most sense. Angola HIV-1 epidemics

is peculiar, as it is a very old epidemic driven by highly divergent forms of the virus

including all M subtypes, circulating recombinant forms (CRFs), unique recombinant

forms (URFs) and many untypable (U) sequences [5, 7, 96-99]. However, host immunity

to these complex HIV-1 strains in Angola has never been studied.

In the 4 th Chapter of this thesis the aim was to make the first detailed characterization of

the neutralizing antibody response and determine viral and host factors associated with

this response in Angolan patients infected with HIV-1.

We performed a cross–sectional population study with a total of 375 HIV-1 infected

Angolan plasma samples, collected in 2001, 2009 and 2014. To characterize the

neutralizing antibody responses we used a panel of 12 tier 2 Env-pseudotyped viruses

representative of the strains circulating worldwide and used plasma samples from HIV-1

infected Angolan patients collected in 2009 (n=178) and 2014 (n=58). Briefly, Env-

34
 Chapter 1 – General Introduction

pseudotyped viruses were produced by co-transfection of Env-expressing plasmids in

293T cells with PSG3.1Δenv plasmid as backbone and titrated in TZM-bl cells. A plasma

dilution of 1:40 was used on an initial neutralizing screening assay and a neutralization

score was determined in order to be reflective of neutralization potency and breadth.

Neutralization of Env-pseudotyped viruses was measured using Tat-regulated firefly

luciferase (luc) reporter gene expression to quantify reductions in virus infection in TZM-

bl cells. VSV-G-pseudotyped particles were used to test neutralization specificity. A

subset of plasma samples showing broad cross-neutralizing activity in the initial

neutralizing screening assay were selected for determinining the 50% inhibitory dilution

(ID50) (n=28 in 2009 and n=10 in 2014). We also amplified and determined by

phylogenetic analysis the viral subtype and tropism in the env gene, specifically in the

C2V3C3 region in samples collected in 2009 (n=110 sequences) and compared with

previous results determined in samples collected in 2001 (n=96 sequences). In the 2014

samples, amplification was not possible as the vast majority of patients were under

antiretroviral therapy. In addition to characterizing the neutralizing response and

subtyping the samples, we determined the possible neutralizing epitopes of the polyclonal

plasma samples by comparison with neutralizing epitopes from previously characterized

bNAbs. Binding titer against constructed recombinant polypeptides, comprising the C2,

V3, C3 regions of different HIV-1 subtypes, was determined in a subpopulation of

patients with known neutralizing response (n=48 in 2009 and n=16 in 2014). We then

explored possible associations and correlations between neutralizing response, vir al

subtype, binding titer to the C2V3C3 recombinant polypeptides, and demographic and

infection characteristics of the HIV-1 infected Angolan population.

35
Chapter 1 – General Introduction

36
 Chapter 2

HIV-1 Diversity, Transmission Dynamics and Primary Drug

Resistance in Angola

Inês Bártolo 1,2 , Suzana Zakovic 2 , Francisco Martin 1,2 , Claudia Palladino 1 , Patrícia Carvalho 3,
Ricardo Camacho 3,4 , Sven Thamm 5 , Sofia Clemente 6 , Nuno Taveira 1,2*

1Unidade dos Retrovírus e Infecções Associadas, Centro de Patogénese Molecular e Instituto de

Investigacão do Medicamento ([Link]), Faculdade de Farmácia, Universidade de Lisboa,
Lisboa, Portugal. 2 Centro de Investigação Interdisciplinar Egas Moniz, Instituto Superior de
Ciências da Saúde Egas Moniz, Monte de Caparica, Portugal. 3 Laboratório de Biologia
Molecular, Centro Hospitalar Lisboa Ocidental, Hospital Egas Moniz, Lisboa, Portugal. 4 Rega
Institute for Medical Research, KU Leuven, Leuven, Belgium. 5 Abbott GmbH & Co. KG,
Wiesbaden, Germany. 6 Hospital da Divina Providência, Serviço de Doenças Infecciosas, Luanda,
Angola

PLoS ONE December 2014 9(12): e113626.

[Link]

Keywords: HIV-1 epidemics; Drug resistance; Transmission dynamics; Angola

38
 Chapter 2 – HIV-1 diversity in Angola

Abstract

Objectives: To assess HIV-1 diversity, transmission dynamics and prevalence of

transmitted drug resistance (TDR) in Angola, five years after ART scale-up.

Methods: Population sequencing of the pol gene was performed on 139 plasma samples

collected in 2009 from drug-naive HIV-1 infected individuals living in Luanda. HIV-1

subtypes were determined using phylogenetic analysis. Drug resistance mutations were

identified using the Calibrated Population Resistance Tool (CPR). Transmission

networks were determined using phylogenetic analysis of all Angolan sequences present

in the databases. Evolutionary trends were determined by comparison with a similar

survey performed in 2001.

Results: 47.1% of the viruses were pure subtypes (all except B), 47.1% were

recombinants and 5.8% were untypable. The prevalence of subtype A decreased

significantly from 2001 to 2009 (40.0% to 10.8%, P=0.0019) while the prevalence of

unique recombinant forms (URFs) increased 2-fold (40.0% to 83.1%, P<0.0001). The

most frequent URFs comprised untypable sequences with subtypes H (U/H, n=7, 10.8%),

A (U/A, n=6, 9.2%) and G (G/U, n=4, 6.2%). Newly identified U/H recombinants formed

a highly supported monophyletic cluster suggesting a local and common origin. TDR

mutation K103N was found in one (0.7%) patient (1.6% in 2001). Out of the 364

sequences sampled for transmission network analysis, 130 (35.7%) were part of a

transmission network. Forty eight transmission clusters were identified; the majority

(56.3%) comprised sequences sampled in 2008–2010 in Luanda which is consistent with

a locally fueled epidemic. Very low genetic distance was found in 27 transmission pairs

sampled in the same year, suggesting recent transmission events.

39
 Chapter 2 – HIV-1 diversity in Angola

Conclusions: Transmission of drug resistant strains was still negligible in Luanda in

2009, five years after the scale-up of ART. The dominance of small and recent

transmission clusters and the emergence of new URFs are consistent with a rising HIV-1

epidemics mainly driven by heterosexual transmission.

40
 Chapter 2 – HIV-1 diversity in Angola

Introduction

Despite the recent decline in the number of people newly infected with HIV, around 35.3

million people were still living with HIV at the end of 2012 [29]. Sub-Saharan Africa

remains severely affected by the epidemic accounting for 71% of the people living with

HIV in the world and for 69.5% of the new infections [29]. Angola is a South-western

African country bordered by Republic of Congo, Democratic Republic of Congo, Zambia

and Namibia. According to the UNAIDS report on the global AIDS epidemic 2013 [29]

the estimated HIV prevalence and new infections in adults have decreased between 2001

and 2012 in all the bordering countries of Angola. For example, in the Republic of Congo

HIV prevalence decreased from 4.7% to 2.8% and the number of new infections

decreased from 6,600 to 3,400. In contrast, the estimated number of adults living with

HIV in Angola has increased in the same period from 110,000 to 220,000 (1.8% vs 2.3%

prevalence) and the estimated number of new infections rose from 16,000 to 23,000 [29].

However a recent HIV seroprevalence survey performed on pregnant women in 36

sentinel sites in 18 provinces of Angola has found that on aggregate HIV prevalence did

not vary significantly from 2004 up to 2011 (median 2.8%, range 2.7%–3.2%) although

there was considerable variation across provinces [255]. Additional studies are clearly

needed to better characterize the dynamics of the HIV epidemic in Angola.

HIV-1 epidemic in Angola is highly complex with all HIV-1 group M subtypes (except

B), several circulant recombinant forms (CRFs), unique recombinant forms (URFs) and

untypable (U) strains reported [4, 5, 7, 35, 36, 96-99]. This genetic complexity may pose

a significant challenge to laboratory diagnosis and antiretroviral treatment (ART)

effectiveness [15, 100], underscoring the importance of implementing regular surveys of

HIV-1 diversity and its impact in this country.

41
 Chapter 2 – HIV-1 diversity in Angola

Transmitted drug resistance (TDR) is a major public health problem, especially in

resource-limited settings as it can determine rapid loss of effectiveness of firstline

antiretroviral (ARV) regimens [206, 210]. Drug-naive individuals that acquire a virus

with drug resistance mutations (DRMs) begin ART with a higher risk of virologic failure

and of developing resistance [206, 256]. The absence of proper patient monitoring may

lead to increased emergence and transmission of resistant strains [257]. ART has been

available in Angola since 2000 for those infected with HIV who could buy ARV drugs.

Since 2004, a national plan has been implemented to provide free ARV drugs to HIV-1

infected individuals using the WHO public health approach to ARV delivery [32]. At the

end of 2012 the number of people on ART was 39,704 [29], 48% of the adults in need of

treatment based on WHO 2010 guidelines [33]. The frequency of TDR in Angolan

patients has risen from 1.6% in 2001 [7] to 16.3% in 2008–2010 [35, 36] suggesting that

TDR may be an important public health problem in Angola. However, further work is

required to characterize TDR level in Luanda as only a few patients living in this province

have been included in previous surveys. In this study we aimed to better characterize the

genetic diversity of HIV-1 and determine the prevalence of TDR in drug-naive patients

in Luanda five years after ART scale-up in 2009. Additionally, to better understand the

dynamics of the HIV-1 epidemics we performed the first investigation of HIV-1

transmission networks in Angola.

Materials and Methods

Study population

One hundred and thirty nine plasma samples were collected during 2009 from drug-naive

HIV-1 positive individuals attending the Hospital da Divina Providência (HDP) in

Luanda, Angola. This hospital is located in the Kilamba-Quiaxe district serving an

42
 Chapter 2 – HIV-1 diversity in Angola

estimated population of 990,892 inhabitants, 13.4% of Luanda’s population (7,395,977

habitants) [195]. Besides the patients attended at the main building, the hospital works

with patients attending four health centers located in different regions of Luanda. The

main criteria for patient inclusion in the study were those recommended by the WHO for

this type of study [257]: confirmed diagnosis of HIV-1 infection, no pregnancy or first

pregnancy (to exclude previous use of ARV for the prevention of mother-to-child

transmission during delivery), no clinical diagnosis of AIDS (stage 1 and 2 WHO

classification system for HIV infection) and no ART exposure. Epidemiological, clinical,

and virological characterization of the patients is given in Table 1. Serological diagnosis

of HIV-1 infection was done using the rapid tests Determine HIV-1/2 (Abbott) and Uni-

Gold Recombigen (Trinity Biotech). The number of CD4+ T cells was determined using

the ABACUS 5 Junior Hematology analyzer. Plasma viral load was determined in a

subset of patients using the Abbott Real Time HIV-1 assay (Abbott Laboratories). The

study was conducted according to the Declaration of Helsinki and was reviewed and

approved by the Board of Directors of Hospital da Divina Providência (Luanda, Angola)

and the National Ethics Committee of Angola. Written informed consent was obtained

from all participants. The study was verbally explained to the patients before they signed

the written consent. For the transmission network study, to avoid overestimation of

relatedness between sequences due to the use of scarce data [258] we extended the study

population to all other Angolan patients for which pol sequences were available in the

Los Alamos HIV Sequence Database [259]. Hence, in addition to our present sequences

we used 226 Angolan pol sequences collected from the Los Alamos HIV Sequence

Database, counting in total 364 sequences. These sequences were derived from samples

collected in 1993 and 2001 (n=86) [5, 7, 98, 99], 2009 (n=39) [259] and 2008-2010

(n=101) [35, 36]. Most sequences (n=64, 28.3%) were obtained from patients attending

43
 Chapter 2 – HIV-1 diversity in Angola

different medical facilities in and near Luanda (including Hospital Sanatório de Luanda,

Laboratório da Força Aérea Nacional Angolana, Hospital Militar Principal, Clínica

Sagrada Esperança, Centro Nacional de Sangue and São Lucas Medical Center in the

village Kifangondo). Remaining sequences were obtained from patients attending

Hospital services in Cabinda (n==20, 8.8%), Namibe (n=4, 1.8%), Benguela (n=4, 1.8%),

Zaire (n=3, 1.3%), Cuanza Norte, Bengo and Huila (n=1, 0.4%, each), and from patients

living in Central (n=7, 3.1%), North (n=3, 1.3%) and South (n=2, 0.9%) of Angola. Origin

of 116 (51.3%) patients was not available. Fourteen patients were on ART. Because HIV

transmission networks are mainly confined to a country [260] no sequences outside

Angola were included in the present study.

Table 1. Epidemiological, clinical, and virological characteristics of HIV-1 Angolan

patients analyzed in this study.

Variables Samples
Patients [n (%)] 139 (100)
Age [mean (SD), years] 36 (14) (n=139)
Gender [n (%)]
Male 50 (36.0)
Female 87 (62.6)
Unknown 2 (1.4)
Transmission route [n (%)]
Heterosexual 120 (86.3)
Vertical 14 (10.1)
Blood transfusion 2 (1.4)
Unknown 3 (2.2)
240.5 (1-1914)
CD4 [mean (range), cells/ml] (n=106)
HIV RNA [mean, (SD) log10,
copies/ml] 5.1 (1.0) (n=86)
Pure subtype [n (%)] 65 (47.1) (n=138)
Untypable 8 (5.8) (n=138)
Recombinants [n (%)] 65 (47.1) (n=138)
doi:10.1371/[Link].0113626.t001

Viral RNA extraction, PCR amplification and sequencing

44
 Chapter 2 – HIV-1 diversity in Angola

Viral RNA was extracted from 140 ml plasma using QIAmp Viral RNA Mini Kit

(Qiagen). RT-PCR was performed with Titan One Tube RT-PCR System (Roche). Nested

PCR was done using an in-house method described elsewhere [5, 7, 98, 99]. Thermal

cycling conditions for PCR and primers sequence and position were previously described

[5, 7, 98, 99]. DNA sequences were obtained with Big Dye Terminator Cycle Sequencing

Kit (Applied Biosystems) and an automated sequencer (3100-Avant Genetic Analyzer,

Applied Biosystems).

Phylogenetic and recombination analysis

Sequences were aligned with reference strains collected from the Los Alamos HIV

Sequence Database [259] using ClustalX [261]. Maximum-likelihood (ML) phylogenetic

analyses [262] were performed using the best-fit model of molecular evolution estimated

by Modeltest v3.7 under the Akaike information criterion [263]. ML trees were inferred,

with program PhyML using Seaview software [264]. To find the ML tree, an iterative

heuristic method combining two different tree rearrangement methods was used: nearest

neighbor interchange (NNI) and subtree pruning and regrafting (SPR). The reliability of

the obtained topology was estimated with the approximate likelihood-ratio test (aLRT)

[264]. Recombination analysis was performed by bootscanning using SimPlot [265]. For

transmission network analysis, protease (PR) and reverse transcriptase (RT) sequences

were concatenated in SeaView [264]. Sequences were aligned with ClustalX [261] and

manually edited in MEGA [211]. Codons associated to drug resistance were stripped from

the alignment to exclude convergent evolution [206]. Phylogenetic analysis was

performed using a single alignment with all subtypes and CRFs included as previously

described [260, 266]. Best-fit model was chosen with Modeltest v3.7 under the Akaike

information criterion [263]. ML tree was constructed in PhyML incorporated in Bioportal

server [267]. Reliability of the tree was assessed using bootstrap replication (1000

45
 Chapter 2 – HIV-1 diversity in Angola

replicates). Genetic distance for clusters with bootstrap support ≥90% was measured in

MEGA [211]. Sequences with genetic distance, 0.05 (range 0.000–0.049) substitutions

per site were considered genetically related and patients were assumed to belong to the

same transmission cluster [268]. Automatic cluster detection was used to confirm the

transmission clusters that were initially detected. This was performed with PhyloPart

program based on Approximate ML tree obtained with FastTree program [269]. Clusters

were detected through the depth-first search of reliable nodes with patristic distance under

1st percentile threshold of the whole tree distance [199]. According to this threshold,

transmission clusters were recognized with patristic distance, 0.07 substitutions per site

and node reliability ≥90%.

Resistance mutation analysis

Resistance mutation analysis was performed using the Stanford genotypic resistance

interpretation algorithm [270]. Mutations specifically associated with transmitted HIV-1

drug resistance were analyzed with the Calibrated Population Resistance Tool (CPR)

([Link] [271, 272].

Statistical analysis

Statistical analysis was performed with GraphPad Prism version 5.00 for Windows,

(GraphPad Software). The Spearman rank test and linear regression analysis were used

to quantify the magnitude and direction of the correlation between viral load and CD4+

T cells. The Mann-Whitney U test was used to compare independent groups. The

frequencies of drug resistance mutations of Angolan viruses were compared with those

available at the Stanford HIV Drug Resistance Database [270] for the same subtypes

using Fisher’s exact test. P-values <0.05 were considered significant.

GenBank accession numbers

46
 Chapter 2 – HIV-1 diversity in Angola

Sequences have been assigned the following GenBank accession numbers: KF853612–

KF853892.

Results

HIV-1 genetic diversity

Plasma samples were obtained from 139 HIV-1 infected individuals. The mean age of the

patients was 36 years (SD, 14) and most (62.6%) were women (Table 1). The main route

of transmission was heterosexual contact (86.3%). As expected, plasma viral load was

high in most patients (mean 5.1 log10 copies/ml) and the number of CD4+ T cells was

low (mean, 240.5 cells/ml). Viral load and CD4+ T cells were negatively correlated

(n=73, Spearman r=-0.3319, P=0.0041). Sequencing and phylogenetic analysis of the PR

region was completed successfully for 139 (100%) patients; RT sequences were also

obtained for all but one of these patients (n=138, 99.3%). Phylogenetic analysis showed

that all viruses belonged to HIV-1 group M (Figure 1 A and B). Out of the 138 isolates

for which there was PR and RT sequences, 65 (47.1%) sequences were non-recombinant

and 65 (47.1%) were recombinant, of which 11 (16.9%) were CRF02_AG, and 8 (5.8%)

were untypable (U). The following pure subtypes and sub-subtypes were identified: A

(n=3, 4.6%), A1 (n=1, 1.5%), A2 (n=3, 4.6%), C (n=24, 36.9%), D (n=9, 13.8%), F1

(n=13, 20.0%), G (n=7, 10.8%), H (n=3, 4.6%) and J (n=2, 3.2%). Thirty different

patterns of recombination were found. Subtypes A1, A2, A3, C, D, F1, G, H, J and K,

and CRF02_AG and U sequences were involved in recombination events. Most of the

recombinants (n=54, 83.1%) were URFs; in almost half of the recombinants (n=31,

47.7%) one of the regions was untypable. The most frequent URFs comprised untypable

sequences with subtypes H (U/H, n=7, 10.8%), A (U/A, n=6, 9.2%) and G (G/U, n=4,

6.2%). The U/H recombinants had a mean genetic distance of 0.062 substitutions per site

47
 Chapter 2 – HIV-1 diversity in Angola

and formed a highly supported monophyletic cluster in both genomic regions indicating

that they share the same origin (Figure 1A and B). A similar U/H cluster has been

described recently in Angola but the Province of origin of the patients in this cluster has

not been disclosed [35, 36]. Phylogenetic analyses revealed a close evolutionary

relatedness of all U/H sequences suggesting that the origin of this emerging URF is

Luanda (Figure 2). The analyses of the evolution of HIV-1 genetic diversity in Luanda

from 2001 to 2009, showed that there was a significant decrease in the prevalence of

subtype A (3.7 fold difference, P=0.0019) which was replaced by subtype C as the

dominant subtype (Table 2). Moreover, the percentage of URFs increased more than

twice during the same period (P<0.0001).

A)

48
 Chapter 2 – HIV-1 diversity in Angola

B)

Figure 1. Genetic subtypes and evolutionary relationships of the viruses sequenced in

this study. Maximum likelihood phylogenetic trees of PR (A) and RT (B) regions were
constructed with reference sequences from all HIV-1 subtypes and sub-subtypes (empty
circles) and with the Angolan sequences (filled circles). In each tree, the aLRT values
supporting the internal branches defining a subtype or a sub-subtype are shown. The scale
represents number of base substitutions per site.

Drug resistance mutations and other polymorphisms

There were no major mutations associated with resistance to protease inhibitors (PIs).

The minor resistance mutations L10I and L10V, associated with resistance to most of the

PIs when present with other mutations [273, 274], were found in 15.7% and 17.6% of the

isolates, respectively (Table S1). This is higher than the frequencies previously described

for untreated patients (6.8% and 8.2%) [270]. V11I, associated with resistance to

darunavir [274, 275], was detected in 7.7% of subtype F isolates and 13.3% of

49
 Chapter 2 – HIV-1 diversity in Angola

CRF02_AG isolates. This frequency is significantly higher than that found in sequences

of the same subtypes available in the Stanford Database [270] (Table S1). K20I was found

in almost all G and CRF02_AG isolates and is a natural polymorphism of both genetic

forms [276]. K20V was found in one patient harboring a CRF02_AG virus. K20I/V

codons are non-polymorphic in most subtypes [136, 270]. They appear to be selected

most commonly by nelfinavir and to reduce its susceptibility [277, 278]. A71T was found

in one patient infected with a subtype C virus and A71V was found in two patients infect

with subtype D. The latter mutation has never been described for subtype D [270].

A71T/V are polymorphisms that occur in 2–3% of untreated individuals but the frequency

increases in patients receiving PIs [279-281]. In subtype D isolates the frequency of

polymorphism I13V was significantly lower than that found in sequences of the same

subtype available in the Stanford database [270] (Table S1). Similar findings were

obtained for K14R, E35D and R57K in subtype A and for L89M in subtype G. For all the

other polymorphisms the frequencies found in the Angolan isolates were significantly

higher when compared with the worldwide sequences available from untreated patients

[270]. Subtype F isolates were the most polymorphic followed by subtype A and

CRF02_AG.

In the RT, we detected the K103N mutation in one patient (1/138, 0.7%) that was infected

with a subtype G virus (Table S2). This mutation confers high-level resistance to

nevirapine, delavirdine and efavirenz [270]. In subtype F the frequency of polymorphisms

A272P and I326V was significantly lower than that found in sequences of the same

subtype available in the Stanford Database [270]. Similar findings were obtained for

K11T, D123AS, K173S, Q174K, V179I, Q207E, R211S, T286A, E312D and G335DE

in subtype A, I293V, I329L and G335D in subtype C and T200A, V292I and G335D in

CRF02_AG. For all the other polymorphisms the frequencies found in the Angolan

50
 Chapter 2 – HIV-1 diversity in Angola

isolates were significantly higher when compared with the sequences available from

untreated patients worldwide. Compared with the 2001 survey there was a 2.3-fold

decrease in the prevalence of TDR (1.6% vs 0.7%).

51
 Chapter 2 – HIV-1 diversity in Angola

Figure 2. Evolutionary relationships of the U/H recombinants. Sequences of U/H

recombinants (named with 09AOHDP) were aligned with those of a previous study
(named gb, GenBank) [35, 36]. The aLRT values supporting the internal branches
defining a subtype or a sub-subtype are shown. The scale represents number of base
substitutions per site.

Table 2. Evolution of HIV-1 genetic diversity in Luanda from 2001 to 2009

HIV-1 genetic forms 2001 2009 P value a

n (%) n (%)
Pure subtypes 46/86 (53.5) 65/138 (47.1) 0.6771
Recombinants 40/86 (46.5) 65/138 (47.1)
First generation CRFs 5/40 (16) 11/65 (16.9) 0.5902
URFs 16/40 (40) 54/65 (83.1) <0.0001
Prevalence of subtype A 17/46 (40) 7/65 (10.8) 0.0019
Prevalence of subtype C 15/46 (33) 24/65 (36.9) 0.6899
Prevalence of subtype F 6/46 (13) 13/65 (20) 0.4453
a Fisher’s exact test.

Transmission network analysis

To better characterize the dynamics of the current HIV-1 epidemics in Angola and assist

in the implementation of more focused prevention strategies we performed a transmission

network analysis. The majority of the 364 sequences included in this sub-study were from

patients residing in Luanda (n=202, 55.5%); the remaining sequences were derived from

patients from seven other provinces of Angola (n=46, 12.6%) or their origin was unknown

(n=116, 31.9%). Forty eight transmission clusters were identified comprising 130 patient

sequences (35.7% of the sampled patients) (Figure 3); more than half of these (52.3%)

reported being heterosexual (Table S3). Consistent with this, small clusters comprising

two closely related strains were dominant (n=33, 68%). Only three large transmission

chains, each comprising seven individuals, were found. As expected, most sequences

(n=98, 75.4%) in the transmission clusters were sampled in 2008, 2009 and 2010 and

most clusters (N=27, 56.3%) comprised only sequences sampled in these dates (Table

52
 Chapter 2 – HIV-1 diversity in Angola

S3). All but one sequence (from Namibe) with available information on the origin were

from Luanda indicating that the current epidemic is mostly sustained by local

transmission. Notably, 14 (29.2%) transmission clusters comprised 2001 and 2008–2010

sequences suggesting a long standing presence of some of the more current viruses

circulating in Luanda. In 3 (21.4%) of these clusters 2001 sequences originated from

Cabinda, Benguela and Lunda Norte Provinces. Six clusters (12.5%) comprised only

sequences sampled in 2001 and one cluster (2.1%) comprised only sequences sampled in

Cabinda in 1993 (Table S3). These clusters were considered uninformative for the current

study.

Figure 3. Transmission cluster analysis. Maximum likelihood tree with the 48

transmission clusters colored in green. Maximum likelihood tree was constructed in
PhyML. Reliability of the tree was assessed using bootstrap resampling (1000 replicates).
Bootstrap values and cluster number are indicated in each cluster. A bootstrap value of
0.7 (70%) or greater indicate significant support for the clusters. The scale represents
number of base substitutions per site.

53
 Chapter 2 – HIV-1 diversity in Angola

Finally, in 27 transmission pairs sampled in the same year (including pairwise clusters

within larger transmission chains) very low genetic distance was found (median 0.005

nucleotide substitutions per site; range 0.000–0.007) suggesting recent transmission

events [282]. Potential sample mix-up for pairwise clusters with 0.000 genetic distances

was excluded based on visual inspection of pairwise alignments and origin of the samples.

The main features of the individuals included in the transmission networks were no

different compared to individuals outside the transmission networks (Table 3).

Table 3. Demographic, immunologic and virologic characteristics of HIV-1 patients

included in the transmission networks and patients outside of the networks.

Transmission Remaining
Variables Total networks patients P value
Patients [n (%)] 364 (100) 130 (35.71) 234 (64.28) -
Gender [n (%)]
Male 100 (27.47) 44 (33.85) 56 (23.93)
Female 142 (39) 47 (36.15) 95 (40.6) 0.1250 a
Unknown 122 (33.52) 39 (30) 83 (35.47)
Transmission route [n (%)]
Heterosexual 194 (53.3) 68 (52.3) 126 (53.85)
Vertical 15 (4.12) 7 (5.38) 8 (3.42) 0.8208 a
Others 10 (2.74) 4 (3.08) 6 (2.56)
Unknown 145 (39.83) 51 (39.23) 94 (40.17)
Origin [n (%)]
Luanda 200 (54.94) 78 (60) 122 (52.14)
Cabinda 20 (5.49) 4 (3.08) 16 (6.84) 0.2850 a
Others 28 (7.29) 8 (6.15) 20 (8.55)
Unknown 116 (31.87) 40 (30.77) 76 (32.48)
CD4 [mean (range), 335.39 (1-1914) 300.12 (12-790)
cells/µL (n=41) (n=78) 0.7309 b
HIV RNA [mean, (range) 5.06 (1.6-6.68)
log10, copies/mL] 5.15 (1.94-7) (n=33) (n=52) 0.6391 b
a – Chi-square test; b – Mann-Whitney test;

Discussion

54
 Chapter 2 – HIV-1 diversity in Angola

We assessed HIV-1 diversity, transmission dynamics and prevalence of TDR in Luanda

in 2009, five years after the scale-up of ART and compared these data with our previous

survey performed in 2001 [5, 7, 98, 99]. Individuals included in this study had a low CD4

count which was directly related with high viral load. These features are consistent with

the reported absence of ART [283, 284]. Like in 2001, no major PIs resistance mutations

were found in the study population which is consistent with the fact that first-line

regimens used in Angola do not include PIs [285]. However, some minor mutations in

the PR and many unusual polymorphisms were detected suggesting that some Angolan

isolates might have a low genetic barrier for resistance to some PIs [270]. The K103N

mutation, which confers high-level resistance to nevirapine, delavirdine and efavirenz

[270], was found in one patient accounting for a 0.7% prevalence rate of TDR which is

2.3 fold lower compared to the 2001 survey [7]. This residual TDR prevalence is similar

to that of several African countries that also use the public health approach to ART [200,

201, 286-288] and suggests that the most common first-line ARV regimens will be

effective in this population. Similar to previous studies, the HIV-1 epidemic in Luanda in

2009 was highly

complex being characterized by the presence of almost all subtypes (A, C, D, F, G, H and

J; 47.1%), complex recombinants viruses (47.1%) and untypable (5.8%) strains [4, 5, 7,

35, 36, 96-99, 289]. A high number of our sequences fall at basal positions on the

phylogenetic trees (pre-subtype branches) which is consistent with the long standing

presence of HIV-1 in Angola [5, 7, 98, 99]. In addition, some strains from Angola have

little organized substructure and form weaker clusters within phylogenetic trees than the

global reference sequences, not allowing a clear distinction between subtypes. As a

consequence, the current global subtype classification may not reflect the extent of

diversity in this region [290]. The prevailing subtype in 2009 in Luanda was subtype C

55
 Chapter 2 – HIV-1 diversity in Angola

(36.9%) followed by sub-subtype F1 (20.0%) whereas in 2001 it was subtype A followed

by subtype C [5, 7, 98, 99]. The significant decrease in the prevalence of subtype A and

increase in subtype C observed in 2009 could be explained by the increasing

predominance of subtype C in the bordering countries, namely in the south region of

Democratic Republic of Congo [202] and in Zambia [35, 36, 200, 207]. As in 2001,

almost half of our sequences were recombinant comprising all group M subtypes as well

as CRF02_AG and U sequences. The frequency of CRFs did not change between 2001

and 2009, but the frequency of URFs more than doubled in the same period.

This is on contrast to the global and regional distribution of HIV-1 genetic forms between

2000 and 2007, where there was a notable increase in the proportion of CRFs and a

decrease in URFs [203]. Importantly, the results indicate that the Angolan HIV-1

epidemic is still increasing in genetic complexity and suggest high rates of co-infection

and/or superinfection [208] which is consistent with an increasing HIV-1 incidence and

prevalence [29, 255, 291]. The most common URF was U/H found in seven strains

(10.8% of the recombinants and 5.1% of the total population). This new recombinant

strain was found in unrelated patients and its sequences clustered in a highly supported

monophyletic group suggesting that it was originally produced in Luanda. The close

relationship with U/H sequences recently reported elsewhere in Angola [35, 36], indicates

that this new recombinant is already established in Angola. Sequencing the full-length

genome of this recombinant strain will be needed to determine if this is a new CRF.

A large number of transmission clusters were identified in this study which included

35.7% of the analyzed samples. This is not uncommon in HIV epidemics as within a

smaller population or even globally HIV infected individuals are often part of wide

transmission networks [195, 260, 292]. Small clusters mostly comprising two sequences

were dominant over large clusters which is consistent with heterosexual contact being the

56
 Chapter 2 – HIV-1 diversity in Angola

main route of transmission reported in most patients [293]. While most transmission

clusters comprised only sequences from Luanda and were therefore consistent with a

locally propelled epidemic, some clusters contained sequences from Luanda and from

other locations in Angola consistent with a more complex origin and transmission

dynamics going well beyond the borders of the capital city. Finally, based on high

sequence homology between patients in transmission clusters, a large number of potential

recent infections were inferred. Overall, the results are consistent with a rising HIV-1

epidemic in Luanda [29, 255, 291]. Further surveys are required to obtain a clearer picture

of the dynamics of the current HIV-1 epidemics at the national level. In conclusion,

transmission of drug resistant strains was still negligible in Luanda in 2009, five years

after the scale-up of ART. The dominance of small and recent transmission clusters and

the emergence of new URFs are consistent with a rising HIV-1 epidemics mainly driven

by heterosexual transmission.

57
 Chapter 2 – HIV-1 diversity in Angola

Supporting Information

Table S1. Minor mutations and natural polymorphisms detected in the PR of drug-naive patients from Luanda.

Patients (n)
P value (Angola vs World)#
Angola World *

Codons A C D F1 G AG A C D F G AG
A C D F G AG
(n =
(n =21) (n = 13) (n = 13) (n = 15) (n = 15) (n = 3736) (n = 7638) (n = 1373) (n = 817) (n = 1208) (n = 3183)
25)
L10I 6 1 3 3 2 1 411 122 47 77 212 166 0.0132 0.875 0.0098 0.0992 1 0.7381
L10V 4 - - 9 2 2 250 - - 250 35 267 0.0009 - - < 0.0001 0.073 0.8242
V11I - - - 1 2 - - - 0 - 73 - - - 0.0133 - 0.0496
T12A - 1 - - 2 - - 107 - - 37 - - 0.8019 - - 0.0801 -
T12I - - - 1 - - - - 33 - - - - 0.4218 -
T12K - - - 1 2 - - - - 25 36 - - - - 0.3409 0.0806 -
T12M - - - 1 - - - - 8 - - - - - 0.1331 0.0001 -
T12N - - - - 2 - - - - - 0 - - - - - - -
T12P - - 1 - - 1 - - 21 - - 0 - - 0.1885 - - < 0.0001
T12S - 18 2 - - 1 - 5041 32 - - 0 - 0.6738 0.0385 - - < 0.0001
I13A - - - - 1 1 - - - - 0 225 - - - - 0.0123 0.6568
I13V 17 4 2 11 14 14 3250 298 919 123 1184 2928 0.6209 0.0096 0.0002 < 0.0001 0.2677 0.7753
K14E - 2 - - - - - 0 - - - - - < 0.0001 - - - -
K14R 2 2 1 1 12 9 1569 840 84 98 773 1942 0.0053 0.8743 0.5624 1 0.2805 0.9809
I15L - - - - - 1 - - - - - 64 - - - - - 0.7204
I15V 7 19 - 8 4 3 1137 6492 - 621 181 226 0.96 0.3289 - 0.3232 0.2642 0.1524
G16A 1 - - - - - 60 - - - - - 0.7830 - - - - -
G16E 9 4 - 2 - 6 523 458 - 131 - 700 0.0005 0.0935 - 1 - 0.1721
L19I 1 11 3 1 1 - 112 4736 84 47 64 0.8661 0.1 0.0432 0.5136 0.5613 -
L19P - - - 1 - 3 - - - 0 - 477 - - - 0.0145 - 0.857
L19T - 6 - - - - - 1069 - - - - - - - - - -
L19V 2 8 2 - - 0 840 0 - - - < 0.0001 0.0025 < 0.0001 - - -
K20I 1 - - - 15 14 164 - - - 1184 3024 0.6928 - - - 1 0.7662

58
 Chapter 2 – HIV-1 diversity in Angola

K20M 1 - - 1 - - 0 - - 15 - - < 0.0001 - - 0.225 - -

K20R 6 5 2 5 - - 523 1375 136 245 - - 0.1096 0.9991 0.6698 0.5463 - -
K20V - - - - - 1 - - - - 35 - - - - - 0.4166
L33F - - 1 - - - - 33 - - - - 0.0094 - - -
E35D 13 4 5 12 6 3 3437 1991 247 735 628 732 < 0.0001 0.3592 0.0697 1 0.4393 0.9742
E35G - - - - 1 - - - - - 0 - - - - - 0.0123 -
E35N 1 - - - - 1 60 - - - - 80 0.8806 - - - - 0.8432
M36I 21 21 10 13 15 14 3661 6492 851 784 1196 3119 0.8994 0.8877 0.3914 1 1 0.7204
M36L - 3 - - - - - 244 - - - - - 0.0547 - - - -
N37A - 1 - - - - - 0 - - - - - < 0.0001 - - - -
N37D 2 - 4 1 - - 560 - 107 60 - - 0.694 - 0.0158 1 - -
N37E 2 - - - - - 45 - - - - - 0.0148 - - - - -
N37S - 2 - 1 3 6 - 649 - 0 29 207 - 0.787 - 0.0157 0.006 < 0.0001
P39A 1 - - - - - 0 - - - - - < 0.0001 - - - - -
P39Q - 1 - - - - - 0 - - - - - - < 0.0001 - - -
P39S - 1 1 - - 2 - - 44 - - 70 - 0.7191 0.3501 - - 0.0426
P39T - - 1 - - - - - 15 - - - - 0.1407 - - -
R41K 15 21 13 13 15 15 3587 6416 1318 744 1160 2960 < 0.0001 0.7848 1 0.6186 1 0.5790
R57K 3 - - 11 - - 1718 - - 686 - - 0.0071 - - 1 - -
Q61H - - - 2 1 - - - - 12 0 - - - - 0.0185 0.0123 -
Q61N - - 1 - - - - - 29 - - - - - 0.2441 - -
I62V 2 - 5 3 - - 168 - 206 98 - - 0.7208 - 0.0484 0.2032 - -
L63N - - - 2 - - - - - 0 - - - - - 0.0001 -
L63P 2 - 3 - 2 3 411 - 426 - 230 509 0.8749 0.6529 0.7648 - 0.7495 0.9446
L63Q - - - - - 1 - - - - - 0 - - - - - < 0.0001
L63S - 2 5 - - - - 359 80 - - - - 0.7606 0.0007 - - -
L63T 4 7 2 - - - 163 2521 58 - - - 0.0041 0.2658 0.1058 - - -
L63V 4 5 - 12 2 1 135 556 - 147 0 0 0.0016 0.04 - < 0.0001 0.0001 < 0.0001
I64L - - - - 1 2 - - - - 41 194 - - - - 0.4098 0.531
I64M - 1 1 - - - - 84 0 - - - - < 0.0001 0.0094 - - -
E65D - - 1 11 - 1 - - 165 196 - 0 - - 1 < 0.0001 < 0.0001
C67E - - - - 4 - - - - - 411 - - - - - 0.7846 -
C67G - - - - 2 - - - - - 18 - - - - - 0.0235 -
C67N - - 1 - - - - 0 - - - - - 0.0094 - - -
C67S - - - - 2 - - - - - 145 - - - - - 0.6994 -
C67Y - 3 - 1 - - - 0 - 0 - - - < 0.0001 - 0.0157 - -
H69K 19 - 1 - 1148 15 3624 - 47 - 15 3088 0.271 - 0.3689 - 1 0.9339

59
 Chapter 2 – HIV-1 diversity in Angola

H69Q 1 - 2 1 - - 71 - 77 74 - - 0.8763 - 0.1668 1 - -

H69R 1 - - - - 0 - - - - - < 0.0001 - - - - -
H69Y - - 1 3 - - - - 220 35 - - - - 0.7052 0.0184 - -
K70Q - - - 1 - - - - 0 - - - - - 0.0157 - -
K70R 4 - 1 2 1 3 273 - 84 27 110 350 0.1022 - 0.5624 0.0725 1 0.5743
A71P - 1 - - - - 0 - - - - - < 0.0001 - - - -
A71T - 1 - - - - 497 - - - - - 0.9193 - - - -
A71V - - 2 - - - - - 0 - - - - - < 0.0001 - - -
I72M - - - - 1 - - - - - 73 - - - - - 0.7924
I72T - - 1 4 - - - - 62 188 - - - - 0.4553 0.5114 - -
I72V - - - - 1 - - - - - 19 - - - - - 0.2202 -
V77L - - - - - 1 - - - - - 0 - - - - - 0.0123
V77I - - - 1 - - - - - 61 - - - - - 1 - -
V82I - - - - 14 2 - - - 6798 102 - - - - 0.9024 0.1397
L89F - - - 1 - - - - - 0 - - - - - 0.0157 - -
L89I 2 - - - 2 1 0 - - - 22 53 < 0.0001 - - - 0.0332 0.4941
L89M 19 23 - 12 13 14 3661 6340 - 400 1184 1148 0.0985 0.3527 - 0.0016 0.0385 0.5379
T91V 2 - - - - - 64 - - - - - 0.0596 - - - - -
I93L 2 24 9 - - - 897 7333 47 - - - 0.1953 0.6102 < 0.0001 - - -
*
According to the Stanford HIV Drug Resistance Database; #Fisher exact test; Common accessory mutations are in bold letters; Common highly polymorphic compensatory mutations are in bo ld underline letters.
doi:10.1371/[Link].0113626.s001

60
 Chapter 2 – HIV-1 diversity in Angola

Table S2. Drug resistance mutations and natural polymorphisms detected in the RT of drug-naive patients from Luanda

Patients (%)

Angola World P value (Angola vs World)#+

Codons A C D F1 G AG A C D F G AG

A C D F G AG
(n = 28) (n = 26) (n = 9) (n = 14) (n = 7) (n = 11) (n = 3108) (n = 6794) (n = 1095) (n = 413) (n = 1197) (n = 2269)

K103N - - - - 1 - - - - - 0 - - - - - <0.0001 -
P4S - 2 - - - - - 272 - - - - - 0.6486 - - -
P4T - 1 - - 1 - - 0 - - 0 - - <0.0001 - 0.0058 - -
P4QK - - - 1 - - - - - 0 - - - - - 0.0328 - -
I5V 2 - - 1 - - 59 - - 0 - - 0.1891 - - 0.0328 - -
E6D - 1 1 - - - - 19 31 - - - - 0.1366 1 - -
E6K 1 1 - - - 1 106 150 - - - 30 0.634 0.8065 - - - 0.3604
E6N 1 - - - - - 53 - - - - - 0.9792 - - - -
K11D - 1 - - - - - 0 - - - - - <0.0001 - - - -
K11Q - - - - 1 - - - - - 12 - - - - - 0.0734
K11R - - - - - 1 - - - - - 0 - - - - - <0.0001
K11S 4 - - - - - 0 - - - - - <0.0001 - - - - -
K11T 5 1 - - - - 1274 0 - - - - 0.0222 <0.0001 - - - -
K20R 4 1 3 5 1 1 591 462 76 50 168 132 0.694 0.836 0.0152 0.0238 1 0.855
V21A 1 - - - - - 0 - - - - - <0.0001 - - - - -
V21I - - - 1 4 - - - - 5 65 - - - 0.1823 0.0003 -
T27S 1 2 - - 1 - 31 197 - 0 - 0.6857 0.4144 - - 0.0058 -
V35K - 4 - 2 - - - 211 - 10 - - - 0.0026 - 0.0546 - -
V35I - - 1 1 - - - 59 1 - - - - 0.3617 0.0674 - -
V35T 28 22 7 11 7 11 2890 6318 986 347 1125 2042 0.2803 0.1995 0.5697 0.4815 - -
E36A 1 22 - - - - 93 4824 - - - - 0.7056 0.1899 - - 1 0.548
E36V 1 - - - - - 0 - - - - - <0.0001 - - - - -

61
 Chapter 2 – HIV-1 diversity in Angola

T39A 1 - 2 1 2 5 162 - 80 83 60 60 0.9695 - 0.1141 0.3208 0.0464 <0.0001

T39D 1 3 - - - - 0 1495 - - - - <0.0001 0.0004 - - - -
T39E - 22 - - - - - 4756 - - - - - 0.1588 - - - -
T39K 9 - - 2 1 - 870 - - 16 83 - 0.7829 - - 0.1131 0.398 -
T39L 1 - 3 3 1 - 47 - 22 28 49 - 0.9121 - 0.0005 0.073 0.2574 -
T39M 3 - - - - - 112 - - - - - 0.1368 - - - - -
T39N 4 - - - - - 177 - - - - - 0.1521 - - - - -
T39P - - - - 1 - - - - 0 - - - - - 0.0058 - -
T39R - - - 1 - - - - - 7 - - - - - 0.2358 - -
T39S 1 - - 1 - - 34 - - 5 - - 0.7348 - - 0.1823 - -
E40A 1 - - - - - 0 - - - - - <0.0001 - - - - -
E40D 7 4 5 1 4 5 373 482 285 83 239 227 0.0707 0.084 0.0332 0.3208 0.034 <0.0001
K43E 3 - - 1 - - 183 - - 0 - - 0.5 - - 0.035 - -
K43Q - - - - 1 - - - - - 0 - - - - - 0.0058 -
K43R - 1 - - 2 - - 50 - - 204 - - 0.4858 - - 0.3427 -
S48T 1 24 - - - - 78 5911 - - - - 0.8035 0.6094 - - - -
K49R 4 - 7 2 1 - 342 - 722 79 132 - 0.8035 - 0.278 1 0.5603 -
K64R 5 - - - - - 0 - - - - - <0.0001 - - - - -
V69I 11 5 8 1 6 3 1492 2718 1051 74 922 771 0.4657 0.1313 1 0.4803 1 0.8812
S68G - - - - 1 2 - - - - 24 79 - - - - 0.1369 0.0701
S68R 1 - - - - - 0 - - - - - <0.0001 - - - - -
A98S - - - - 4 - - - - - 407 - - - - - 0.2385 -
K102Q 3 - - - - - 53 - - - - - 0.0041 - - - - -
K103R - 1 - - - - - 0 - - - - - <0.0001 - - - -
K104R 1 1 - 2 - 2 0 197 - 10 - 129 <0.0001 0.7655 - 0.0546 - 0.3387
K104T - 1 - - - - - 0 - - - - <0.0001 - - - -
V118I 32 1 - 1 - - 56 170 - 13 - - <0.0001 0.0024 - 0.3776 - -
V118M - - 1 - - - - - 0 - - - - - 0.0073 - - -
D121C - - 2 - - - - - 48 - - - - 0.8175 0.0473 - - -
D121H 4 1 1 - 1 - 373 285 107 - 65 - 0.9377 0.688 0.5627 - 0.3268 -
D121Y - 3 3 - - - - 815 372 - - - - 0.8175 1 - - -

62
 Chapter 2 – HIV-1 diversity in Angola

K122A 1 - - - - - 0 - - - - - <0.0001 - - - - -
K122E 25 20 7 3 5 3 2797 6115 898 66 778 817 0.8478 0.059 1 0.4815 1 0.7739
K122V - 1 - - - - - 0 - - - - - <0.0001 - - -
D123G - 3 - - 1 1 - 1427 - - 40 75 - 0.2703 - - 0.2158 0.8224
D123N 11 3 - 1 1 - 777 951 - 0 132 - 0.1295 0.9382 - 0.0328 0.5603 -
D123S 7 5 - - 3 1 1461 1563 - - 443 123 0.0329 0.6737 - - 0.7144 0.8958
I135K - 1 - - - - - 0 - - - - - <0.0001 - - - -
I135L - - - 1 - - - - - 95 - - - - - 0.208 - -
I135R - - 1 - - - - - 26 - - - - - 0.1804 - - -
I135T 8 - 1 2 1 1492 - 252 40 323 - 0.063 - 0.6903 0.6378 0.6818 -
I135V - 5 1 9 3 5 - 435 3 120 335 1498 - 0.024 0.0287 0.0137 0.4083 0.2641
I135Y - 1 - - - - - 0 - - - - - <0.0001 - - - -
E138A - 2 - - - - - 387 - - - - - 0.9371 - - - -
E138D 1 - - - - - 0 - - - - - <0.0001 - - - -
E138V - 1 - - - - - 0 - - - - - <0.0001 - - - -
I142T - 1 - - - - - 37 - - - - - 0.2453 - - - -
I142V 5 5 1 - - - 404 285 105 - - - 0.6326 0.0019 0.5556 - - -
K154R 1 1 - - - - 0 0 - - - - <0.0001 <0.0001 - - - -
P157L - 1 - - - - - 0 - - - - - <0.0001 - - - -
P157R - - - 2 - - - - - 0 - - - - - 0.001 - -
A158S 5 4 1 - - - 496 747 12 - - - 0.9889 0.6893 0.0908 - - -
S162A 1 - - - 1 9 137 - - - 395 2133 0.8042 - - - 0.4368 0.2904
S162C 1 7 1 11 2 - 177 747 0 219 49 - 0.9416 0.0397 0.0073 0.0988 0.0322 -
S162E - - - - - 1 - - - - - 0 - - - - - <0.0001
S162N 1 - - - - - 0 - - - - - <0.0001 - - - - -
S162T - - - - - 1 - - - - - 0 - - - - - <0.0001
S162W - - - 1 - - - - - 5 - - - - - 0.1823 - -
T165I 1 3 2 - 1 - 146 414 81 - 47.0 - 0.8663 0.4554 0.1165 - 0.2484 -
T165P - 1 - - - - - 0 - - - - - <0.0001 - - - -
K166R 2 3 1 - - - 81 951 22 - - 0.3694 0.9382 0.1556 - - -
E169D 1 - 2 2 - 1 109 - 120 153 - 27 0.6188 - 0.2186 0.7159 - 0.3166

63
 Chapter 2 – HIV-1 diversity in Angola

E169G 1 - - - - - 0 - - - - <0.0001 - - - - -
K173A 15 14 - 2 2 - 435 4212 - 215 168 - <0.0001 0.2671 - 0.0058 0.2587 -
K173D - - - 1 - - - - - 0 - - - - - 0.0328 - -
K173I - - - 7 - 3 - - - 35 - 193 - - - 0.0001 - 0.0937
K173L 3 - - 1 - - 839 - - 0 - - 0.0852 - - 0.0328 - -
K173M - - - 1 - - - - - 0 - - - - - 0.0328 - -
K173N - - - - - 1 - - - - - 0 - - - - - <0.0001
K173R - 1 - 1 1 - - 0 - 7 65 - <0.0001 - 0.2358 0.3268 -
K173S 4 - - - - 1 1554 - - - - 43 0.0004 - - - - 0.5273
K173T 1 10 - - 3 3 96 1699 - - 515 1520 0.6881 0.1759 - - 1 0.3303
Q174K 14 11 - 14 5 - 2300 2174 - 372 910 - 0.0078 0.5341 - 0.3797 0.6760 -
Q174N - 1 1 - - - - 11 0 - - - - 0.0332 0.0073 - - -
Q174R - 1 1 - 1 7 - 401 11 - 63 1566 - 0.9783 0.0840 - 0.3184 0.2381
D177E 21 20 7 4 5 7 2642 5299 996 149 1101 2133 0.2272 0.9160 0.5337 0.7782 0.1052 0.0004
D177G 2 - - - - - 87 - - - - - 1 - - - - -
D177N - - - 1 - - - - - 10 - - - - - 0.3101 - -
I178L 3 2 3 3 - - 106 482 131 50 - - 0.1136 0.7917 0.0620 0.3973 - -
I178M 3 - 2 1 - 3 174 - 471 19 - 635 0.4494 - 0.4779 0.4945 - 0.7764
I178V - 3 - - - - - 115 - - - - - 0.0020 - - - -
V179D - 2 1 - - - 0 - 1 - - - <0.0001 - 0.1823 - -
V179G - - - 1 - - - - - 5 - - - - - 0.1823 - -
V179I 1 - - 1 - - 1678 - - 20 - - <0.0001 - - 0.1585 - -
G196E 8 2 1 - 2 - 193 299 39 - 71 - <0.0001 0.7360 0.2566 - 0.0625 -
G196K - 1 - - - - - 122 - - - - - 0.9634 - - - -
G196S 1 - - - - - 27 - - - - - <0.0001 - - - - -
T200A 19 26 - 14 7 7 1150 6250 - 182 1101 2065 0.0015 0.2537 - - 1 0.0088
T200E - - 5 - - - - - 164 - - - - - 0.0030 - - -
T200I 2 - - - - - 118 - - - - - 0.6715 - - - - -
T200M - - 1 - - - - - 34 - - - - - 0.2280 - - -
T200R 1 - - - - - 0 - - - - - <0.0001 - - - - -
T200V - - - - - 4 - - - - - 0 - - - - - <0.0001

64
 Chapter 2 – HIV-1 diversity in Angola

I202V 2 2 - 2 - - 622 510 - 26 - - 0.1442 0.7361 - 0.2320 - -

E204D 1 1 1 - - - 0 0 0 - - - <0.0001 <0.0001 0.0073 - - -
E204K - - 1 - - 1 - - 81 - - 52 - - 0.4621 - - 0.6241
E204L - - 1 - - - - - 0 - - - - - 0.0073 - - -
E204Q 1 - 2 - - - 0 - 0 - - - <0.0001 - <0.0001 - - -
Q207A 18 5 - - 1 3 2704 584 - - 55 172 0.0011 0.1148 - - 0.2841 0.0601
Q207D 4 2 - 1 - - 87 428 - 30 - - 0.0024 0.9103 - 1 - -
Q207E 1 17 7 11 1 7 180 4552 756 223 515 1679 0.9247 0.9728 0.4470 0.0998 0.2495 -
Q207G - - - - - 1 - - - - - 95 - - - - - 0.9558
Q207K - - - - 3 - - - - - 359 - - - - - 0.4354 -
Q207N 2 1 1 44 211 - 16 - - 0.0854 0.7271 - 0.4389 - -
Q207S - 1 - - - - - 0 - - - - - <0.0001 - - - -
R211A - - - 1 - - - - - 12 - - - - - 0.3558 - -
R211K 15 14 6 7 1 9 466 4484 879 314 443 1203 <0.0001 0.2722 0.6609 0.0516 0.4338 0.1081
R211N - - - - 1 1 - - - - 37 23 - - - - 0.2015 0.2552
R211S 10 - - 1 5 - 2238 - - 5 383 - <0.0001 - - 0.1823 0.0388 -
F214L 6 5 3 - 1 - 404 883 38 - 144 - 0.3003 0.5152 0.0129 - 0.5937 -
P243A 2 2 1 1 - - 0 0 14 66 - - <0.0001 <0.0001 0.1041 0.7067 - -
P243S - 1 1 1 - - - 0 35 0 - - - <0.0001 0.2338 0.0328 - -
P243T 1 3 1 - - 0 163 - 10 - - <0.0001 0.0173 - 0.3101 - -
I244V - - 5 - - - - 44 - - - - <0.0001 - - -
V245E 7 1 1 - - - 528 211 164 - - - 0.3845 0.7271 1 - - -
V245H - - - 1 - - - - - 0 - - - - - <0.0001 - -
V245K 1 2 4 1 - 1 292 482 482 5 - 168 0.4666 0.7971 0.7368 0.1823 - 0.7160
V245L 1 - - - - - 0 - - - - - <0.0001 - - - - -
V245M 3 - - - - - 466 - - - - - 0.7144 - - - - -
V245Q 13 21 1 11 17 10 1243 5571 175 380 1065 1072 0.6184 0.9260 1 0.1050 0.3887 0.0096
V245S - - - 1 - - - - - 0 - - - - - <0.0001 - -
V245T 1 - 1 - - - 402 - 153 - - - 0.0015 - 1 - - -
E248D 10 8 - 7 4 4 995 679 - 244 814 386 0.8303 0.0014 - 0.5844 0.6867 0.1939
E248N 2 - - 4 - - 205 - - 58 - - 0.7901 - - 0.1306 - -

65
 Chapter 2 – HIV-1 diversity in Angola

D250T 1 - - - - - 0 - - - - - <0.0001 - - - - -
S251A 1 - - - - - 0 - - - - - <0.0001 - - - - -
S251D - 1 - - 1 4 - 14 - - 36 0 - 0.0763 - - 0.1967 <0.0001
S251H 3 - - - - - 40 - - - - - 0.0006 - - - -
S251N 1 1 - 1 - - 71 88 - 7 - - 0.8365 0.8221 - 0.2358 - -
S251T 1 - - - - - 0 - - - - - <0.0001 - - - -
K275Q 3 1 - 1 2 121 279 - 4 77 - 0.1748 0.6684 - 0.1542 0.1263 -
K275R 1 2 - 1 2 1 53 394 - 78 0 116 0.9792 0.9935 - 0.4825 <0.0001 0.9296
A272K - 2 - - - - - 82 - - - - - 0.0356 - - - -
A272P 11 18 7 11 6 - 1108 5435 788 392 1053 - 0.8402 0.2613 0.4550 0.0378 0.5937 -
A272S 6 1 2 1 106 214 - 10 57 - <0.0001 0.7193 - 0.0546 0.2928 -
A272T - - - - - 1 - - - - - 0 - - - - - <0.0001
V276I - 1 1 - 1 2 - 88 49 - 62 170 - 0.7808 0.3109 - 0.3142 0.4431
V276T - - - 1 - - - - - 0 - - - - - 0.0328 - -
K277Q - - - 1 - - - - - 0 - - - - - 0.0328 - -
K277R 8 12 6 3 1 - 528 4076 624 120 227 - 0.1711 0.2161 0.4779 0.7656 1 -
K277S - - - - 1 1 - - - - 0 408 - - - - 0.0058 0.7093
Q278H - 2 3 1 3 - - 299 64 66 91 - - 0.7360 0.0096 0.7067 0.0128 -
Q278N - 1 - 1 - - - 183 6 - - - 0.8070 - 0.2095 - -
K281R 6 4 1 4 2 1 653 598 105 27 109 522 0.8580 0.4039 0.5556 0.0137 0.1302 0.4620
L283I 3 2 - - - - 127 197 - - - - 0.2022 0.3868 - - - -
R284K - 1 - - - 4 - 95 - - - 34 - 0.8231 - - - <0.0001
T286A 18 20 6 13 7 11 2673 4416 712 355 1053 2156 0.0026 0.2861 0.7206 0.7033 1 0.9498
T286V - 1 - - - - - 183 - - - - - 0.8070 - - - -
E291D 23 24 - 11 6 11 2766 6386 - 384 1089 2019 0.3963 0.9584 - 0.079 0.4862 0.4946
E291I - - - 1 - - - - - 0 - - - - - 0.0328 - -
V292I 20 23 1 13 7 6 1958 6318 110 326 1137 1883 0.4694 - 0.5732 0.3182 1 0.0361
I293V 27 20 6 13 7 10 3046 6386 843 405.0 1149 2224 0.9326 0.0013 1 0.2613 1 0.5500
P194A 1 - - - - - 127 - - - - - 0.7319 - - - - -
P294S - - - 1 - - - - - 7 - - - - - 0.1434 - -
P294T 16 2 1 - 2 9 2020 347 41 - 68 1566 0.5043 0.8799 0.2677 - 0.0579 0.5556

66
 Chapter 2 – HIV-1 diversity in Angola

L295M 2 - - - - - 106 - - - - - 0.5771 - - - - -

L295W 1 - - - - - 0 - - - - - <0.0001 - - - - -
E297A 1 - - 1 5 - 0 - - 0 802 - <0.0001 - - 0.0328 - -
E297D 2 - - - - - 0 - - - - - <0.0001 - - - - -
E297K 4 2 6 - - - 199 102 142 - - - 0.1929 0.0768 0.0001 - - -
E297R 2 - - 8 - - 53 - - 244 - - 0.1445 - - 1 - -
E297T - - - 3 1 - - - - 95 103 - - - - 1 0.4695 -
L310I - 1 - 1 - 4 - 82 - 5 - 41 - 0.7422 - 0.1823 - <0.0001
K311R 3 1 - - - - 118 401 - - - - 0.1617 0.9783 - - - -
E312D 8 - - - - - 1865 - - - - - 0.0015 - - - - -
E312N 1 - - - - - 342 - - - - - 0.3419 - - - - -
E312P 1 - - - - - 0 - - - - - <0.0001 - - - - -
E312Q 1 - - - - - 0 - - - - - <0.0001 - - - - -
E312T 3 - - - - - 37 - - - - - 0.0003 - - - - -
V317A 3 5 - 1 - - 246 747 - 58 - - 0.8459 0.3056 0.7033 - -
V317E - 1 - - - - - 0 - - - - - <0.0001 - - - -
S322A 1 - - - - 7 106 - - - - 163 0.6340 - - - - <0.0001
S322E 1 - - - - - 0 - - - - - <0.0001 - - - - -
S322T - - - - - 1 - - - - - 102 - - - - - 0.9964
K323E 1 - - - - - 0 - - - - - <0.0001 - - - - -
K323I 1 - - - - - 0 - - - - - <0.0001 - - - - -
K323R 1 - - - - 1 0 - - - - 0 <0.0001 - - - - <0.0001
D324E 7 6 1 3 4 - 224 747 45 50 850 - 0.0013 0.0993 0.2896 0.3973 0.4214 -
D324F - - - - - 1 - - - - - 0 - - - - - <0.0001
L325I - - - - 2 - - - - - 22 - - - - - 0.0075 -
I326K - - - - - - - - - - - - - - - - - -
I326L - - - 1 - - - - - 0 - - - - - 0.0328 - -
I326V 11 - - 1 1 8 497 - - 245 40 1339 0.0021 - - <0.0001 0.2158 0.5382
E328D 1 1 2 - - - 0 82 0 - - - <0.0001 0.7422 <0.0001 - - -
E328K - - - - - 1 - - - - - 0 - - - - - <0.0001
I329L - 1 1 1 - - - 353 57 27 - - - 0.0438 0.1546 1 - -

67
 Chapter 2 – HIV-1 diversity in Angola

I329V 1 1 - 1 6 - 106 129 - 50 1041 - 0.6340 0.0004 - 1 1 -

G333D - - 1 - - - - - 0 - - - - - 0.0073 - - -
G333E - - 1 1 - - - - 0 0 - - - - 0.0073 0.0328 - -
G333L - - - - 1 - - - - - 0 - - - - - 0.0058 -
Q334A - - - 1 - - - - - 0 - - - - - 0.0328 - -
Q334C - 1 - - - - - 102 - - - - - 0.8627 - - - -
Q334D - 1 - - - - - 747 - - - - - 0.3954 - - - -
Q334E - - - - 1 - - - - - 14 - - - - - 0.0842 -
Q334H 2 - 1 - - 1 56 - 41 - - 0 0.1664 - 0.2677 - - <0.0001
Q334K - 1 - - - - - 0 - - - - - <0.0001 - - - -
Q334N 2 4 - 1 - - 0 747 - 14 - - <0.0001 0.6893 - 0.3987 - -
Q334R - - - 1 - - - - - 0 - - - - - 0.0328 - -
Q334Y - - 1 1 - - - - 11 0 - - - - 0.0840 0.0328 - -
G335A 1 - - - - - 0 - - - - - <0.0001 - - - - -
G335D 4 1 - - - 2 2704 5435 - - - 1951 <0.0001 <0.0001 - - <0.0001
G335L - 1 - - - - - 0 - - - - - <0.0001 - - - -
G335E 1 - - - - - 2705 - - - - - <0.0001 - - - - -
G335R 1 - - - - - 0 - - - - - <0.0001 - - - - -
G335S - - 1 - - - - - 50 - - - - - 0.3161 - - -

*According to the Stanford HIV Drug Resistance Database; #Fisher exact test; Drug resistance mutations are in bold letters. Additional NRTI-
selected mutations are in underline letters. Additional NNRTI polymorphic accessory mutations are in italic letters.
doi:10.1371/[Link].0113626.s002

68
 Chapter 2 – HIV-1 diversity in Angola

Table S3. Epidemiological characteristics of the patients included in the transmission

clusters.

Year Reported
Cluster Sampling
Sequence of Gender transmission Region
number date
birth route
09AGHDP119 2009 1980 F Heterosexual Luanda
09AGHDP226 2009 n.a. F n.a. Luanda
09AGHDP208 2009 1970 F Heterosexual Luanda
1 JN937038 2008 1975 F n.a. n.a.
01AOHM176 2001 1963 M Heterosexual Luanda
JQ616884 2009 n.a. n.a. n.a. n.a.
JN937034 2008 n.a. n.a. n.a. n.a.
JQ616880 2009 n.a. n.a. n.a. n.a.
2
JN937047 2009 n.a. n.a. n.a. n.a.
09AGHDP237 2009 1957 M Heterosexual Luanda
3 01AOCSE126 2001 1976 F n.a. Lunda
Norte
01AOLFA13 2001 1971 M Heterosexual Luanda
4
01AOHAB86 2001 1963 F n.a. Luanda
JN937098 2010 n.a. n.a. n.a. n.a.
5
JN937104 2010 n.a. n.a. n.a. n.a.
09AGHDP186 2009 1982 F Heterosexual Luanda
6
JQ616882 2009 n.a. n.a. n.a. n.a.
09AGHDP62 2009 1973 F Heterosexual Luanda
JN937097 2010 n.a. n.a. n.a. n.a.
7 01AOSNS09 2001 1981 F Homosexual Luanda
09AGHDP42 2009 1967 M Heterosexual Luanda
09AGHDP289 2009 1985 F Heterosexual Luanda
93AOHDC247 1993 n.a. F n.a. Cabinda
8
93AOHDC253 1993 n.a. F n.a. Cabinda
JN937116 2010 n.a. n.a. n.a. n.a.
9
JN937112 2010 1976 M Heterosexual Central
JN937050 2009 n.a. n.a. n.a. n.a.
JQ616883 2009 n.a. n.a. n.a. n.a.
10
09AGHDP279 2009 1960 M Heterosexual Luanda
01AOHJM06 2001 1968 F Heterosexual Luanda
01AOSNS01 2001 1959 M Heterosexual Luanda
11
01AOHDP73 2001 1958 M Heterosexual Luanda
JN937046 2009 n.a. n.a. n.a. n.a.
12
JN937037 2008 n.a. n.a. n.a. n.a.

69
 Chapter 2 – HIV-1 diversity in Angola

JN937061 2009 n.a. n.a. n.a. n.a.

13
JN937101 2010 n.a. n.a. n.a. n.a.
09AGHDP233 2009 1979 M Heterosexual Luanda
09AGHDP280 2009 2004 M Vertical Luanda
14
JN937054 2009 1964 F Heterosexual Central
01AOSNS56 2001 1965 M n.a. Luanda
09AGHDP231 2009 1987 F Heterosexual Luanda
15 09AGHDP164 2009 1992 F Heterosexual Luanda
01AOHDC229 2001 1964 M Heterosexual Cabinda
09AGHDP44 2009 2007 M Vertical Luanda
16
09AGHDP68 2009 1985 F Heterosexual Luanda
09AGHDP50 2009 1963 M Heterosexual Luanda
17
09AGHDP118 2009 1978 F Heterosexual Luanda
09AGHDP242 2009 1970 M Heterosexual Luanda
18
09AGHDP111 2009 1965 F Heterosexual Luanda
JN937017 2008 1987 F Heterosexual Namibe
19
JN937040 2008 1967 M MSM Luanda
JN937095 2010 1965 M Heterosexual Benguela
20
01AOHDP71 2001 1949 F Heterosexual Luanda
01AOHJM64 2001 1975 F n.a. Luanda
21 01AOSNS55 2001 1966 F n.a. Luanda
09AGHDP240 2009 1982 F Heterosexual Luanda
09AGHDP74 2009 1939 M Heterosexual Luanda
22
JQ616899 2009 n.a. n.a. n.a. n.a.
09AGHDP94 2009 1973 M Heterosexual Luanda
23
09AGHDP30 2009 2007 M Vertical Luanda
JN937059 2009 n.a. n.a. n.a. n.a.
JN937060 2009 n.a. n.a. n.a. n.a.
24
JN937115 2010 1981 F n.a. Central
09AGHDP82 2009 2006 M Vertical Luanda
09AGHDP212 2009 1981 F Heterosexual Luanda
01AOSNS04 2001 1969 M Bisexual Luanda
25
JQ616913 2009 n.a. n.a. n.a. n.a.
09AGHDP263 2009 1978 F Heterosexual Luanda
09AGHDP267 2009 1971 M Heterosexual Luanda
26
01AOSNS24 2001 1973 F Heterosexual Luanda
01AOLFA90 2001 1977 M n.a. Luanda
27
01AOSNS49 2001 1957 M Bisexual Cabinda
JN937081 2010 n.a. n.a. n.a. n.a.
JN937070 2010 n.a. n.a. n.a. n.a.
28
JN937073 2010 n.a. n.a. n.a. n.a.
JN937079 2010 n.a. n.a. n.a. n.a.
29 09AGHDP169 2009 1986 M Heterosexual Luanda

70
 Chapter 2 – HIV-1 diversity in Angola

09AGHDP167 2009 1980 M Heterosexual Luanda

09AGHDP130 2009 1984 F Heterosexual Luanda
30
09AGHDP145 2009 1976 M Heterosexual Luanda
01AOLFA19 2001 1963 M Heterosexual Luanda
31 01AOLFA17 2001 n.a. F Heterosexual Luanda
JQ616890 2009 n.a. n.a. n.a. n.a.
09AGHDP49 2009 1973 M Heterosexual Luanda
32
01AOCSE136 2001 1972 M Heterosexual Luanda
JQ616894 2009 n.a. n.a. n.a. n.a.
33
JQ616891 2009 n.a. n.a. n.a. n.a.
09AGHDP64 2009 1963 F Heterosexual Luanda
09AGHDP200 2009 1969 M Heterosexual Luanda
JN937085 2010 n.a. n.a. n.a. n.a.
34 09AGHDP296 2009 1976 F Heterosexual n.a.
01AOLFA94 2001 1969 M Heterosexual Luanda
01AOSNS36 2001 n.a. F Heterosexual Luanda
JN937075 2010 n.a. n.a. n.a. n.a.
09AGHDP281 2009 1978 F Heterosexual Luanda
35
JQ616886 2009 n.a. n.a. n.a. n.a.
01AOSNS03 2001 1964 M Heterosexual Luanda
36 01AOLFA14 2001 1956 M Heterosexual Luanda
01AOLFA18 2001 1958 F Heterosexual Luanda
09AGHDP274 2009 1976 M Heterosexual Luanda
37 09AGHDP204 2009 2006 F Vertical Luanda
(a2)
09AGHDP106 2009 2006 M Vertical Luanda
38
09AGHDP57 2009 1966 F Heterosexual Luanda
09AGHDP290 2009 1980 M Heterosexual Luanda
39
JN937058 2009 n.a. n.a. n.a. n.a.
09AGHDP100 2009 1969 F Heterosexual Luanda
40
JN937033 2008 n.a. n.a. n.a. n.a.
JN937035 2008 n.a. n.a. n.a. n.a.
41 JN937072 2010 n.a. n.a. n.a. n.a.
09AGHDP37 2009 1972 F Heterosexual Luanda
09AGHDP157 2009 1958 M Heterosexual Luanda
42
JQ616905 2009 n.a. n.a. n.a. n.a.
JQ616885 2009 n.a. n.a. n.a. n.a.
43 09AGHDP20 2009 1982 F Heterosexual Luanda
09AGHDP245 2009 1989 F Heterosexual Luanda
01AOCSE125 2001 1969 M Heterosexual Luanda
44
01AOHDP75 2001 1973 F Heterosexual Luanda
JN937031 2008 n.a. n.a. n.a. n.a.
45
JN937106 2010 1991 F Heterosexual Central

71
 Chapter 2 – HIV-1 diversity in Angola

09AGHDP258 2009 2006 M Heterosexual Luanda

09AGHDP266 2009 1986 F Heterosexual Luanda
01AOSNS40 2001 1976 F Heterosexual Luanda
46 09AGHDP86 2009 2006 M Vertical Luanda
JN937028 2008 n.a. n.a. n.a. Central
JQ616898 2009 n.a. n.a. n.a. n.a.
09AGHDP105 2009 1976 M Heterosexual Luanda
09AGHDP201 2009 1951 M Heterosexual Luanda
47
JQ616897 2009 n.a. n.a. n.a. n.a.
01AOLFA97 2001 1958 F Heterosexual Luanda
48
01AOSNS37 2001 1973 F n.a. Luanda
n.a.- not available. doi:10.1371/[Link].0113626.s003

Author Contributions

Conceived and designed the experiments: NT IB SC RC SZ. Performed the experiments:

IB SC PC FM SZ. Analyzed the data: IB SZ CP ST RC NT. Contributed

reagents/materials/analysis tools: IB SZ FM ST SC. Wrote the paper: IB SZ NT FM.

72
73
 Chapter 3

Early Infant Diagnosis of HIV-1 Infection in Luanda, Angola,

Using a New DNA PCR Assay and Dried Blood Spots

Short title: Early Infant Diagnosis of HIV-1 in Luanda

Francisco Martin 1, ¶, Claudia Palladino 1, ¶, Rita Mateus1, Anna Bolzan 2, Perpétua

Gomes3,4, José Brito 4, Ana Patrícia Carvalho3, Yolanda Cardoso5, Cristovão Domingos6,

Vanda Sofia Lôa Clemente 2, Nuno Taveira1,4

1Research Institute for Medicines ([Link]), Faculdade de Farmácia, Universidade

de Lisboa, Lisbon, Portugal. 2 Hospital da Divina Providência, Luanda, Angola.
3Laboratório de Microbiologia Clínica e Biologia Molecular, Serviço de Patologia

Clínica, Centro Hospitalar de Lisboa Ocidental, E.P.E., Lisbon, Portugal. 4 Centro de

Investigação Interdisciplinar Egas Moniz, Instituto Superior de Ciências da Saúde Egas
Moniz, Caparica, Portugal. 5 Ministério da Saúde de Angola - Instituto Nacional de Saúde
Pública (INSP), Luanda, Angola. 6 Ministério da Saúde de Angola - Instituto Nacional de
Luta contra a Sida (INLS), Luanda, Angola.

¶These authors contributed equally to this work.

PLoS ONE 2017 July; 12(7): e0181352. [Link]

Keywords: HIV-1 early infant diagnosis; proviral DNA; dried blood spots; Luanda,

Angola

74
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Abstract

Background: Early diagnosis and treatment reduces HIV-1-related mortality, morbidity

and size of viral reservoirs in infants infected perinatally. Commercial molecular tests

enable the early diagnosis of infection in infants but the high cost and low sensitivity with

dried blood spots (DBS) limit their use in sub-Saharan Africa.

Objectives: To develop and validate a sensitive and cheap qualitative proviral DNA

PCR-based assay for early infant diagnosis (EID) in HIV-1-exposed infants using DBS

samples.

Study design: Chelex-based method was used to extract DNA from DBS samples

followed by a nested PCR assay using primers for the HIV-1 integrase gene. Limit of

detection (LoD) was determined by Probit regression using limiting dilutions of newly

produced recombinant plasmids with the integrase gene of all HIV-1 subtypes and ACH-

2 cells. Clinical sensitivity and specificity were evaluated on 100 HIV-1 infected adults;

5 infected infants; 50 healthy volunteers; 139 HIV-1-exposed infants of the Angolan

Pediatric HIV Cohort (APEHC) with serology at 18 months of life.

Results: All subtypes and CRF02_AG were amplified with a LoD of 14 copies. HIV-1

infection in infants was detected at month 1 of life. Sensitivity rate in adults varied with

viral load, while diagnostic specificity was 100%. The percentage of HIV-1 MTCT cases

between January 2012 and October 2014 was 2.2%. The cost per test was 8-10 USD

which is 2- to 4-fold lower in comparison to commercial assays.

Conclusions: The new PCR assay enables early and accurate EID. The simplicity and

low-cost of the assay make it suitable for generalized implementation in Angola and other

resource-constrained countries.

76
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Introduction

HIV-1 mother-to-child-transmission (MTCT) is the main mode of infection among the

pediatric population and is disproportionately affecting children in impoverished

countries. Despite the decline in MTCT rate in recent years in most of the sub-Saharan

Africa, it is estimated that 150,000 children became newly infected with HIV in 2015

[30]. Children infected perinatally are at high risk of rapid disease progression and death

during the first year of life without antiretroviral therapy (ART) [197]. Given the reported

benefits of early ART initiation in reducing HIV-1-related mortality and long-term

morbidity [253] and reducing the size of the HIV-1 reservoirs [254], early HIV-1

diagnosis in newborns represents the critical gateway to timely initiation of life-saving

ART. Serological assays do not permit the early diagnosis of HIV-1 infection because of

the persistence of maternal HIV-1 antibodies in infants during the first 12-18 months of

life. The WHO recommends the use of molecular-based virological testing to determine

the infection status for HIV-1-exposed infants during the first 4-6 weeks of life or at the

earliest opportunity thereafter [132]. Despite the high accuracy of tests which detect HIV

RNA or p24, their sensitivity could potentially be affected in settings of expanded ART

for prevention of MTCT (option B and B+), which reduce circulating HIV-1 RNA and

viral particles [133]. Qualitative DNA PCR test which detect proviral DNA in peripheral

blood mononuclear cells (PBMCs) is recommended for early infant diagnosis (EID) of

HIV-1 and is the most widely implemented test in resource-limited settings [134, 135].

The considerable uptake of HIV-1 DNA molecular tests is driven by the lower costs

compared with quantitative assays along with their good sensitivity when performed on

blood microsamples or dried blood spots (DBS) [134]. The use of DBS presents several

advantages such as reduced costs for collection, storage and shipping. Thus, DBS samples

are convenient for increasing access to testing in settings with poor healthcare provision

77
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

and referral laboratories [294]. Currently, two HIV-1 DNA assays are commercially

available for EID using DBS: the COBAS® AmpliPrep/COBAS® TaqMan® HIV-1

Qualitative Test (Roche Molecular System Inc., Branchburg, NJ), which recently

replaced the Roche Amplicor® DNA test v1.5, and the RealTime HIV-1 Qualitative Test

(Abbott Molecular, Des Plaines, IL) [135]. These assays are highly sensitive but require

sophisticated instrument platforms whose cost with related equipment (e.g. centrifuge;

biosafety cabinet; freezer) can range from about US$ 100,000 to more than US$ 200,000.

Recently, innovative technologies designed for use at or near the point-of-care (such as

the Cepheid Xpert® HIV-1 Qual assay and the Alere™ q HIV-1/2 Detect) have been

developed and the Xpert® HIV-1 Qual assay has been validated for both whole blood

and DBS specimens [135].

Despite the recent advances in prevention of MTCT, Angola reported one of the highest

rates of MTCT (25%) among the 22 priority countries included in the UNAIDS global

plan [30]. The EID national program implemented in 2007 based on the Roche Amplicor®

DNA test v1.5 was interrupted in 2012; consequently the coverage of virological testing

for infants is currently very low and only 14% of HIV-1-exposed newborns received

virological HIV-1 testing within the first 2 months of life in 2014 [30]. Several challenges

may prevent the implementation of molecular diagnostic tests such as the high cost of

available commercial PCR assays and their limited sensitivity with highly divergent HIV-

1 subtypes which co-circulate in the country [5-7, 98, 99, 212, 252].

In this article, we developed and validated a new qualitative HIV-1 DNA PCR assay for

the early infant diagnosis of HIV-1 infection in Angola and other countries with similar

complex epidemics. Using this highly sensitive and specific assay we identified the new

cases of MTCT of HIV-1 in the APEHC pediatric cohort recently established in a major

hospital in Luanda.

78
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Participants and Methods

Study Design

This assay validation study describes the use of DBS to diagnose HIV-1 in infants born

to HIV-infected mothers in Angola between January 2012 and October 2014. The

STROBE checklist was used to help design and conduct the study [295].

Ethics Statement

This study was conducted according to the Declaration of Helsinki with the approval of

the National Ethical Committee of Angola and the Ethic Committee of the Centro

Hospitalar de Lisboa Ocidental, E.P.E in Portugal. Written informed consent was

obtained from all participants or from parents/guardians for their infants and from HIV-

1 seronegative healthy volunteers.

The Angolan Pediatric HIV Cohort and clinical samples

The Angolan Pediatric HIV Cohort (APEHC) has prospectively collected data on HIV-

1-infected pregnant women and their infants attending the municipal Hospital da Divina

Providência (HDP) since March 2012. HDP is located in the Luanda district, which is the

most populated district of Luanda city, Angola. Epidemiological, clinical and laboratory

data were collected at study entry and every 6 months thereafter for women. Infants were

followed according to the perinatal care service offered at the HDP, which includes

clinical and biological examination at months 1, 3, 6, and 18 of life. No specific

recommendations for HIV treatment was made for women enrolled in the cohort and

physicians followed the WHO guidelines for prevention of MTCT [134]. Infants received

NVP once daily from birth through age 4-6 weeks in accordance with option B [134] .

HIV-1 testing for both mothers and newborns was free of charge and was performed using

two rapid tests for detection of antibodies against HIV-1/2 (Determine HIV ½ and Uni-
79
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Gold HIV) as recommended by WHO [296]. The diagnosis required consistent results of

the two different tests. An infant was considered as infected if anti-HIV-1 antibodies were

detected on two separate samples collected at least three months apart and persisted after

18 months of age; it was considered non-infected if serological testing was negative on

two separate samples before or after 18 months. Undetermined cases were those with

discordant results between the two rapid tests and were further tested using an ELISA

assay (Vironostika HIV Uni-form II Ag/Ab ELISA test; bioMérieux, France). DBS

specimens were obtained from all infants born between January 2012 and October 2014

from HIV-1 infected women enrolled in the APEHC. Additionally, five DBS samples

from HIV-1 infected infants aged 2- to 12-days old were obtained from the Instituto

Nacional de Saúde Pública (INSP). DBS samples were also collected from six women

attending the HDP with indeterminate HIV serologic testing results. DBS samples were

prepared by spotting 125 µL of whole blood, collected by heel prick for infants and by

finger-prick for adults, onto filter paper cards (Whatman® Human ID Blood Stain Cards

BFC 180). DBS were dried over night at room temperature, individually enveloped in

Glassine paper, inserted in a zip-lock polyethylene bag (Deltalab S.L., Spain) with silica

desiccant and stored at -20ºC. DBS were subsequently shipped at room temperature to

our laboratory at the University of Lisbon for testing.

Two sets of samples were used as clinical controls. One-hundred DBS specimens were

collected during 2014 from adults attending a central hospital in Lisbon (Centro

Hospitalar de Lisboa Ocidental, E.P.E.), who had a confirmed HIV-1 diagnosis done by

a 4th generation assay followed by an immunoblot assay on two separate samples. Plasma

HIV-1 RNA level was determined in these patients by COBAS® AmpliPrep/COBAS®

TaqMan® HIV-1 Test, v2.0 (Roche Diagnostic Systems) with a lower limit of detection

80
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

of 20 copies/mL. CD4 + T-lymphocytes were quantified by flow cytometry. Fifty DBS

collected from HIV-1 seronegative healthy volunteers were used as negative controls.

To further test the diagnostic specificity of the assay, samples obtained from patients

infected with HBV, HCV and CMV were also analyzed. These samples were collected

from HIV-1 seronegative adults attending the Centro Hospitalar de Lisboa Ocidental,

E.P.E. who had a confirmed diagnosis of hepatitis B (HBV) done by detection of hepatitis

B surface antigen, HBsAg, (N=10), or hepatitis C (HCV) done by detection of

antibodies to HCV and viral RNA (N=10), or cytomegalovirus (CMV) done by detection

of antibodies to CMV (N=10).

Production of DBS samples with control plasmids

A 1553 bp pol gene product comprising the highly conserved IN gene region was

amplified from the reference subtype B HIV-1 SG3.1 [297] and from nine primary viral

isolates belonging to the most common HIV-1 clades circulating in Angola S1 Table by

RT-PCR as described previously [4, 5, 7, 98, 99]. Primers for this PCR were designed

using PerlPrimer® v1.1.21 software and reference sequences present in the Los Alamos

HIV sequence database ([Link] Their sequence and location in the

HIV-1 genome are shown in S2 Table. PCR products were cloned into pcDNA 3.1D/V5-

His-TOPO plasmid using the protocol indicated by the manufacturer (Invitrogene Corp.,

Carlsbad, CA) and sequenced by the Sanger method. Phylogenetic analysis was used to

confirm the subtype of the isolates (data not shown). Purified control plasmids were

quantified by spectroscopy at 260 nm using a calibration curve and then serially diluted

in HIV-1 seronegative blood and spotted (125 µL) onto Human ID blood stain cards to

obtain a concentration of 50 to 5,000 copies/DBS.

Production of DBS samples with ACH-2 cells

81
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

ACH-2 cells were used as analytical control for DNA PCR assays as they contain a single,

integrated HIV-1 subtype B proviral DNA per cell [298, 299]. ACH-2 cells were diluted

in 125 µL of HIV seronegative blood (5 log serial dilutions) and spotted on Human ID

blood stain cards to obtain a DBS control panel containing 50-5,000 cells.

Chelex DNA extraction method

DNA was extracted from DBS samples using the polyvalent cationic resin Chelex 100®

(Bio-Rad Laboratories, Hercules, CA, USA). Briefly, six circles of 5 mm diameter were

punched from each DBS spot into a 1.5 mL tube. After 30 min washing with sterile water

and 3 min centrifugation (15,400 g), 250 µL of 5% Chelex solution was added to the pellet

and samples were incubated at 56ºC for 15 minutes. The samples were vortexed and

centrifuged (15,400 g) for 3 min, prior to a final incubation at 100ºC for 8 minutes.

Finally, samples were centrifuged and stored at -20ºC until the PCR reaction was

performed.

PCR amplification of proviral HIV-1 DNA

A nested PCR was used to amplify a 194 bp fragment of the IN gene. First- and second-

round amplifications were performed using the same reaction and cycling conditions. The

25 µL reaction volume contained 1X NH4 buffer, 3 mmol/L MgCl2, 0.5 µmol/L of each

dNTP, 0.3 µmol/L forward and reverse primers S2 Table, 1U of Taq DNA polymerase

(Bioline® Reagents Ltd, London, UK) plus 2.5 µL of DNA solution from the clinical or

control samples (first-round PCR). For the second-round PCR reaction we used 2.5ul of

the first-round PCR reaction. Amplification cycling conditions were as follows:

denaturation step of 94°C/3 min followed by 35 cycles of denaturation at 94°C/ 45 sec,

annealing at 56°C/35 sec, extension at 72°C/ 1 min followed by a single final extension

step at 72ºC/ 15 min. Amplified products were visualized with green safe staining after

electrophoresis in 2% agarose gel. In all cases the human gene C-C chemokine receptor

82
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

5 (CCR5) was used as an internal control to confirm the presence and quality of genomic

DNA. CCR5 was amplified using primers CCR5c and CCR5d [107] and the cycling

conditions used for the amplification of the IN gene region.

Limit of detection (LoD)

DNA extracted from DBS samples spotted with serial dilutions of control plasmids or

ACH-2 cells was subjected to the nested PCR protocol described above. Each template was

amplified ≥10 times and the LoD for each subtype and its 95% fiducial confidence

interval were estimated by probit regression analysis.

Results

Analytical sensitivity

The LoD of the assay was determined by probit regression analysis with DBS spiked

with serial dilutions of control plasmids containing the IN gene regions from HIV-1

subtypes A, C, D, F, G, H, J and CRF02_AG. For HIV-1 subtype B, DBS spiked with

increasing number of ACH-2 cells, which contain one integrated proviral DNA copy per

cell, were used. When the subtype was included in the probit model as an independent

factor, the parallelism test chi-square was significant (χ2 = 64.7; df = 8; p < 0.001) which

rejects the assumption of equal slopes across subtypes. Therefore, the probit analysis was

implemented separately for each subtype. Under these conditions the LoD of the assay

varied between 4.3 and 14.4 copies depending on subtype Table 1.

Table 1. Limit of detection of the assay for different HIV-1 subtypes as determined using

probit regression analysis.

83
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Probit results*
Significance of
Subtype
Pearson χ2 goodness of 95% Confidence
LoD (copies/PCR)
fit test Interval
A1 0.330 5.0 4.1 – 8.4
H 0.300 10.1 7.9 – 19.7
B 0.885 9.3 5.0 – 33.1
F 0.914 10.4 7.7 – 27.2
G 0.997 11.8 9.6 – 22.6
J 0.267 4.3 3.3 – 6.3
CRF02_AG 0.252 4.4 3.4 – 6.4
C 0.077 5.5 4.7 – 31.0
D 0.624 14.4 10.4 – 25.2
* Determined in DBS spiked with serial dilutions of control plasmids except for subtype
B that was determined in DBS spiked with ACH-2 cells (S1 Fig. and S3 Table).

One hundred and twenty five µl of infant’s blood, which is the amount present in each

DBS, has 0.4-0.6x106 PBMCs [300]. HIV-1 infected infants harbor an estimate of 13,000

to 75,400 HIV proviral copies per 10 6 PBMCs [301]. In each PCR reaction, we used 1/100

of the extracted DNA solution (2.5 out of 250 µl); considering only the lower limit of

PBMCs that the infants may have in this amount of blood this corresponds to 0.4x104

PBMCs per reaction. Considering the lower number of proviral copies that the HIV-1

infected infants may have in this number of cells this corresponds to a minimum of 52

HIV-1 proviral copies. This is more than 11-fold higher than the lower LoD of our PCR

assay assuring that it has high enough sensitivity to detect HIV-1 infection in all infected

infants.

Diagnostic sensitivity and specificity

Diagnostic specificity of the EID assay was 100% since all the HIV-1 seronegative

samples tested (n=186, 50 adults from Portugal and 136 infant DBS samples from the

APEHC cohort) were negative for the presence of HIV-1 proviral DNA. This was further

confirmed using 30 samples from patients infected with HBV, HCV or CMV as all gave

negative results using the new PCR assay S2 Fig.

84
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

The clinical sensitivity of the assay was determined with DBS samples collected from

100 HIV-1 infected adult patients from Portugal (patients with chronic infection), from 5

confirmed HIV-1 positive infants obtained from the Instituto Nacional de Luta contra a

Sida (INLS), Luanda, Angola (infection in these patients was confirmed by serology at

month 18 of life and by detection of HIV-1 DNA using the Nuclisens EasyMag/EasyQ,

Biomérieux) and from the infants of the APEHC cohort. Regarding the Portuguese adult

patients, the median CD4 count was 608 cells/mm3 (min-max:83-2,075). Nine subjects

were severely immunosuppressed (<200 cells/mm3), 35 had CD4 count of 200-499

cells/mm3, and 52 had ≥500 cells/mm3. HIV-1 proviral DNA was detected in 14.3%,

56.3% and 85.7% of the patients with plasma viral load of <20 copies/mL, 20-1,000

copies/mL and >1,000 copies/mL, respectively Table 2. All five HIV-1 infected infants

from the INLS were HIV-1 DNA positive by our assay.

Table 2. Performance of the new PCR assay in HIV-1-infected adults from Portugal.

Adult HIV-1-infected patients

(N=100)
EID assay Undetectable Viral load of Viral load of Total
viral load 20-1,000 >1,000
(<20 copies/mL) copies/mL copies/mL
Positive 11 9 6 26

Negative 66 7 1 74

Total 77 16 7 100
Percentage of
14.3 56.3 85.7 --
detection

A total of 154 HIV-1-exposed infants were enrolled in the APEHC cohort and one DBS

card containing 4 blood spots per infant was available for the lab tests. The median age

was 1 month: 83% (129/154) of infants were 1 month of age, 7% (11/154) were 2-5

85
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

months of age, and 9% (14/154) were 6-12 months of age; 50% were girls (n=77). For

the specificity and sensitivity analyses, 15 patients were excluded as follows: 11 infants

dropped-out from routine clinical care and 4 infants died before the serology results were

confirmed. Those patients were all negative according to our assay. Three out of the 139

samples that were analyzed by our assay were HIV-1 DNA positive; infection with HIV-

1 in these infants was confirmed by serology at month 18 (Table 3 and S3 Fig.).

Table 3. Sensitivity and specificity of the new assay for early infant diagnosis of HIV-1

infection in Angola.

Infants with HIV-1 serology at

month 18 and/or HIV-1 DNA test
(N=144)*

Our HIV-1 DNA

Positive Negative
PCR assay

Positive 8 0
Negative 0 136
Total 8 136
Sensitivity 100.0 %
Specificity 100.0 %

* Infants from the APEHC cohort (N=139) plus infants (N=5) with HIV-1 infection
confirmed at the INLS in Luanda, Angola.

All 136 infants with negative results with our assay were HIV-1 seronegative at month

18. Therefore, 3 out of 139 (2.2%) infants from the APEHC cohort were infected with

HIV-1 between January 2012 and October 2014 acquiring the virus through MTCT.

Among the HIV-1-infected infants, one was an 8-month-old girl born in healthcare

facilities at the end of 2012 who received oral zidovudine and formula feeding; her mother

initiated ART (lamivudine/zidovudine/nevirapine) during the second trimester of

86
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

pregnancy and received intrapartum zidovudine. The second infant was a one month old

boy born at home in July 2014 who received formula feeding and his mother initiated

ART (tenofovir/lamivudine/efavirenz) during the third trimester of pregnancy; no

prophylaxis with zidovudine was administered. The third HIV-1-infected newborn was a

disabled 7–month-old girl who was transferred to another hospital soon after delivery. No

information on prophylaxis was available.

Finally, we calculated the cost per test using our assay, including cost of filter papers,

reagents, equipment maintenance, and human resources to be 8-10 USD which is about

1/2 to 1/4 of the cost of commercialized tests in Angola. Hands-on time required to

perform the assay was comparable based on information taken from company websites

and references [302-304] (Table 4).

87
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

1 Table 4. Comparison of cost and operational features of our in-house assay with commercial assays.

Features In-house EID assay AMPLICORTM HIV-1 DNA Abbott RealTime HIV-1
Test v1.5 Qualitative

Type of assay PCR/qualitative PCR/qualitative Real Time PCR/Qualitative

Specimen volume 100-125 µl 200-500 µl 100-200 µl
Target of amplification Pol (IN) Gag Pol (IN)

Genotypes detected All subtypes, CRF02 Subtypes A-H Subtypes A-H, CRF01, CRF02,
Groups O and N

Analytical sensitivity (for 112 copies/ml Not available 839 copies/ml

DBS specimens)
Time for result 6-7 hours 7-8 hours 8 hours

Cost/test (USD)* 8-10 15-30 37

Number of samples/run 1-30 samples 9-21 samples 21-96 samples
Equipment required Thermocycler, microcentrifuge, heat Thermocycler, ELISA, reader/washer, M2000sp, M2000rt equipment
block, gel electrophoresis microcentrifuge

Equipment cost 14,000$ 25,000$ 150,000$

2 *Pricesof tests performed with commercial assays vary considerably depending on quantities, infrastructure, support required and country of

3 implementation.

88
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Discussion

Children are among the most vulnerable to be at risk for HIV infection but in spite of this

the AIDS response in sub-Saharan Africa has largely left them behind [30]. In this regard,

Angola has only registered a 25% reduction of new infections among infants since 2009

[30] which led the Assistant Secretary-General of the United Nations to declare in 2015

that the epidemic in Angola might worsen if an effective AIDS response is not

reinvigorated. EID of HIV-1 infection in infants at risk enables early treatment and care

of the infected infants. Significant progress has been made in many sub-Saharan countries

in implementing EID services following the introduction of HIV DNA testing on DBS

[198]. However, the high cost of commercially available HIV-1 DNA tests and the

perceived sensitivity problems related to the very diverse and complex viral strains

circulating in Angola have prevented their implementation in this country. At the HDP

where our cohort is based and in most other hospitals in Angola, pediatric diagnosis of

HIV-1 infection is still done by serology at month 12 of life which significantly delays

the initiation of treatment [305]. To support EID service expansion in Luanda we

developed and validated a new HIV-1 DNA assay to be used on DBS samples. To account

for the very diverse HIV-1 strains present in Luanda, we had to produce a new set of

control plasmids containing the IN gene, our target for amplification, from local HIV-1

subtypes. Phylogenetic analysis showed that most of the new IN sequences fall at basal

positions on the phylogenetic trees (pre-subtype branches) which is consistent with the

complexity of the HIV-1 strains circulating in Luanda and with Angola being one of the

epicenters of the HIV-1 epidemic [4-7, 98, 99, 212, 252].

To lower the costs, we used the Chelex method of DNA extraction which is also quick

and easy to perform. This method had been previously applied to diagnostic screening in

89
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

HIV-1-exposed infants in Rwanda showing reliable results when used in combination

with either an in-house nested PCR or the Roche Amplicor HIV-1 DNA assay version

1.5 [306]. For the purpose of the EID, the Chelex DNA extraction method represents a

low-cost alternative to commercial kits such as the QIAamp™ DNA Investigator Kit

which costs 6.8 USD per reaction in Portugal and at least twice that in Angola. The Chelex

method that we have used costs less than 3 USD per reaction, is very quick to perform,

and does not use hazardous solvents. Moreover, a recent study that compared the yield of

DNA extracted from blood samples applied to Whatman™ FTA™ cards using different

methods showed that the amount of DNA recovered with the Chelex procedure (2.5 ng

from a blood spot of 6.0 mm) is similar to or larger than the amount of DNA recovered

with the other methods [307].

Our nested PCR assay showed a very low LoD for all the complex HIV-1 genotypes that

we used as controls, suggesting that it was appropriate for early diagnosis of HIV-1

infection in infants in Angola. Indeed, using this assay we could detect all HIV-1 infected

infants at month 1 of life. The good performance of the assay was also demonstrated in

HIV-1 infected adults where we could detect HIV-1 DNA in 14.3% of patients with

undetectable viral load (plasma viral load of <20 copies/mL).

The low percentage of HIV-1 infected infants (2.2%) in the APEHC cohort between 2012

and 2014 contrasts with the rate reported in 2014 at a national level of 25% [30] and

confirms the effectiveness of the WHO-based prevention program implemented since

2007 at the HDP [305]. The detection of HIV-1 infection in infants as early as 1 month

after birth makes this new assay suitable to health care centers following option B+ of

WHO guidelines that recommend EID at 4-6 weeks of life. Moreover, our test might be

useful to determine HIV-1 infection status when serology results are indeterminate after

12-18 months of age.

90
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

One possible shortcoming of our study is that we could not make a head-to-head

comparison of our assay with a commercial test because currently there is no adequate

platform for EID testing from DBS in Angola. In fact, the EID national program

implemented at the Instituto Nacional de Saúde Pública (INSP) with the support of the

Clinton Foundation was discontinued in 2012. NucliSens® HIV-1 QT test (bioMérieux,

Inc., Durham, NC) is still used at the INLS but its performance is severely affected by the

genetic heterogeneity of HIV-1. As reported by several studies, the test has low accuracy

in detecting or quantifying specific group M subtypes (A, C, F, and G), recombinants

(CRF01_AE, CRF02_AG), and group O [308-311]. Another possible shortcoming is the

limited size of the prospectively enrolled cohort and consequently the clinical evaluation

of the assay. Thus, further study in the clinical setting is likely warranted.

Conclusions

The high analytical and clinical sensitivity of our EID assay have enabled accurate, early

and low cost diagnosis of HIV-1 infection in exposed infants in Angola. The low

percentage of HIV-1 MTCT case observed within the APEHC cohort is consistent with

the current high standard of pediatric care provided at HDP. The simplicity and low-cost

of the assay make it suitable for generalized implementation in Angola and other

resource-constrained countries.

Competing interests

The authors declare that they have no conflicts of interest.

Acknowledgements

We greatly acknowledge the contribution and efforts of the staff involved at the Hospital

da Divina Providência in Luanda for helping us to conduct the study. ACH-2 cells were

91
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

obtained through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH. We

appreciate the participation of all the patients without whom this study would not have

been possible. Financial support was provided by the Fundação para a Ciência e a

Tecnologia (FCT) Portugal (PTDC/SAU-EPI/122400/2010), part of the EDCTP2

programme supported by the European Union. Francisco Martin is supported by the

Portuguese Fundação para a Ciência e Tecnologia (FCT) (grant number

SFRH/BD/87488/2012). Claudia Palladino is supported by the Portuguese Fundação

para a Ciência e Tecnologia (FCT) (grant number SFRH/BPD/77448/2011).

Supporting Information

92
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

S1 Fig. Representative example of the results of the limit of detection (LoD) of the new

PCR assay for subtype J control plasmid. 500, 250 and 100 copies of control plasmid

were added to 125 ul of seronegative HIV blood and spotted in Human ID bloodstain

cards. Extracted DNA was amplified by nested-PCR and amplified products were run on

a 2% agarose gel with green safe staining. Each samples was amplified >10 times. (M)

Molecular weight marker (NZY Leader VI); (IN) HIV-1 integrase fragment (194 bp);

(R5) CCR5 gene fragment (189 bp); (SN) HIV-1 seronegative control; (-): ddH2O. The

LoD for the subtype J was 4.3 copies/PCR (95% confidence interval: 3.3-6.3).

1A

1B

1C

93
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

S2 Fig. Results of the diagnostic specificity experiments using the new PCR assay with

samples collected from adult patients infected with HBV, HCV or CMV. Amplification

of samples from patients infected with HBV, HCV or CMV using the new PCR assay

conditions. A) Samples of HBV-infected patients (n=10); B) Samples of HCV-infected

patients (n=10); C) Samples of CMV-infected patients (n=10). Samples were tested in

triplicate. PCR products were run on 2% agarose gel and stained with green safe. (M)

Molecular weight marker (NZY Leader VI); (-) HIV-1 seronegative sample; (+) HIV-1

seropositive sample (194 bp); (R5) CCR5 gene (189 bp).

S3 Fig. Representative example of the results obtained using the new PCR assay on

samples collected from infants enrolled in the APHEC cohort. Each infant was assigned

an anonymized code (C126-C132). C126-C131 are uninfected infants whereas C132 is

an HIV-1 infected infant. Samples were tested in triplicate. Amplified products were run

on a 2% agarose gel with green safe staining. (M) Molecular weight marker (NZY leader

VI); (IN) HIV-1 integrase fragment (194 bp); (R5) CCR5 gene fragment (189 bp).

94
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Table S1. Origin and genotype of the virus isolates used to produce the control plasmids.

Country
Accession Sampling Genotype
Isolate of Origin Patient
number date (IN gene)
(Province)

93AOHDC249 KU296949 1993 Angola adult A1

(Cabinda)
93AOHDC251 KU296950 1993 Angola adult H
(Cabinda)
93AOHDC253 KU296951 1993 Angola adult J
(Cabinda)

09AOHDP34 KU296952 2009 Angola adult C

(Luanda)

09AOHDP110 KU296953 2009 Angola adult D

(Luanda)

09AOHDP157 KU296954 2009 Angola adult G

(Luanda)

09AOHDP237 KU296955 2009 Angola adult F1

(Luanda)

01PTHDECJN KU296956 1998 Portugal* infant CRF02_AG

(Lisbon)

00PTHDEEBB KU296957 2000 Portugal* infant G

(Lisbon)
*Patients infected in Luanda

Table S2. Sequence and location of PCR primers used in this study and size of amplified.

This table reports the sequence and location of primers used for the amplification of the

IN gene in clinical specimens, reference plasmids and ACH-2 cells.

Position in the Band

PCR type Primers Sequence (5’-3’)
HIV-1 HXB2 size (bp)

F_IN_out AACATAGTAACAGAYTCACARTATGC 4,029-4,055

First-round
1,553
PCR
R_IN_out TGGTCTTCTGGGGCTTGTTCCAT 5,582-5,559

APEHC_IN_F AATTGGAGAGCAATGGCTAGTGA 4,281-4,303 194

95
 Chapter 3 Early infant diagnosis of HIV-1 in Angola

Second-
APEHC_IN_R CACTGGCTACATGGACTGCTAC 4,473-4,452
round PCR

Table S3. Limit of detection (LoD) of HIV-1 subtype B DNA in ACH-2 cells using probit

regression analysis. This table relates to the determination of the LoD of the in-house EID

molecular test in ACH-2 cells using probit regression analysis. The same principle was

applied to the control plasmids in order to determine the LoD for the different subtypes

tested.

Cells and
Cells and No. Detected
provirus per Probit value
provirus per DBS (%)
PCR

5,000 50 10 (100) NA *

1,000 10 10 (100) NA *

500 5 8 (80) 5.84

250 2.5 6 (60) 5.25

100 1 3 (30) 4.48

50 0.5 2 (20) 4.16

* NA, not applicable.

96
97
 Chapter 4

Long-term and low-level envelope C2V3 stimulation from

highly diverse virus isolates leads to frequent development of

broad and elite antibody neutralization in HIV-1 infected

individuals

Francisco Martin 1, José Marcelino 1,2, Claudia Palladino 1, Inês Bártolo 1, Susana Tracana1,

Inês Moranguinho 1, Rita Mateus1, Rita Calado 1, Pedro Borrego 1, Thomas Leitner3, Sofia

Clemente4, Nuno Taveira 1,2#

1Research Institute for Medicines ([Link]), Faculty of Pharmacy, Universidade

de Lisboa, Lisbon, Portugal; 2Centro de Investigação Interdisciplinar Egas Moniz CIIEM,
Instituto Superior de Ciências da Saúde Egas Moniz, Caparica, Portugal; 3Theoretical
Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM
87545, USA; 4Hospital da Divina Providência, Luanda, Angola;
Manuscript submitted to eBioMedicine (Available at medRxiv

[Link]

Keywords: HIV-1 evolution; genetic diversity; antibody neutralization; Angola

98
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Abstract
Elicitation of potent neutralizing antibodies against genetically diverse HIV-1 isolates is

important for an effective HIV-1 vaccine. Some HIV-1 infected patients produce such

broadly neutralizing antibodies (bNAbs). Identification of host and viral correlates of

bNAb production may help develop the next generation of HIV-1 vaccines. We carried

out the first detailed characterization of the neutralizing antibody response and identify

viral and host factors associated with the development of bNAbs in HIV -1 infected

patients from Angola, one of the oldest, more dynamic, and diverse HIV-1 epidemics in

the world. Plasma samples from 322 HIV-1 infected patients were collected in 2001, 2009

and 2014. Phylogenetic analysis of C2V3C3 envelope sequences identified a diverse

array of subtypes including A1, A2, B, C, D, F1, G, H, J, untypable strains, and

recombinant forms which prevailed over pure subtypes. Notably, 56% of the patients

developed cross, broad, or elite neutralizing responses against a reference panel of tier 2

Env-pseudoviruses far exceeding results obtained elsewhere in the world. The frequency

of elite neutralizers was higher in 2014, when patients were on ART and had low viremia,

than in 2009 when patients were drug naive. In drug naïve patients, broad neutralization

was associated with subtype C infection, lower CD4+ T cell counts, higher age, or higher

titer of C2V3C3-specific antibodies relative to patients that did not develop bNAbs.

Neutralizing antibodies targeted the V3-glycan supersite in most patients but antibodies

specific for the V2 apex, the CD4 binding site, the gp41 membrane -proximal external

region (MPER) and unknown epitopes were also found in some patients. V3 and C3

regions were significantly less variable and less subject to positive selection in elite

neutralizers compared to weak or no neutralizers suggesting an active role of bNabs

directed against these regions in controlling HIV-1 replication and diversification. Hence,

development of broad and elite antibody neutralization against HIV-1 requires long-term

100
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

and low-level envelope V3C3 stimulation from highly diverse subtype C isolates. These

results have direct implications for the design of a new generation of HIV-1 vaccines.

101
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Introduction

The HIV-1 Env glycoprotein is highly immunogenic but, in general, the antibodies

elicited by it during infection lack neutralizing breadth or potency against primary HIV-

1 strains thus failing to inhibit viral replication in infected individuals [145, 146]. In

natural infection only 5 to 30% of adult individuals develop broadly neutralizing

antibodies (bNAbs) after several years of infection [92, 145, 146, 159, 161-163, 173, 181,

182, 189, 194, 196, 212, 312-319] and these bNAbs have little impact in the control of

the infection due to the continuous capacity of HIV-1 to diversify and escape these

antibodies [13, 159-163]. However, some recombinant human bNAbs supress viral

replication in HIV-1 infected individuals [177-182], prevent human infection by some

HIV-1 strains [14], and passive immunization in animal models can protect from infection

and/or disease progression (reviewed in [320]). Therefore, bNAbs are promising tools to

restrict HIV-1 transmission and control disease progression if they could be induced by

vaccination. Unfortunately, so far, antibodies elicited by candidate immunogens and

vaccines have shown a very limited ability to neutralize heterologous primary HIV-1

strains [9-20].

bNAbs target five highly conserved epitopes in the HIV-1 envelope: the CD4 binding site

(CD4bs); the V2 apex; V3 glycan supersite; gp41 MPER, and gp41/gp120 interface [213,

214, 321]. However, the mechanisms underlying the elicitation of such antibodies by B

cell populations are still largely unknown [215, 216, 223, 322]. Guiding the immune

system to elicit such bNAbs remains a major challenge due to extremely complex

antibody maturation pathways and high levels of somatic hypermutation (SHM) required

by bNabs to acquire neutralization breadth (reviewed in [322]). An exception to this rule

is the V3-glycan supersite bNAb lineage that does not require extensive antibody-affinity

102
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

maturation [314, 323] allowing their development in early stages of infection [181, 314,

315], and explaining their high prevalence in recently infected individuals [173].

Furthermore, V3-specific IgG binding and neutralizing responses in pregnant woman

living with HIV-1 predict low risk of mother-to-child-transmission of HIV-1 [316]. In

HIV-2 infected individuals the V3 loop is a dominant target of bNAbs such that V3

undergoes extensive sequence, conformational and functional alterations to escape

antibody neutralization [212, 217, 324-326]. Such findings, together with the proved

therapeutic value of V3-glycan supersite bNAbs [181], highlight this epitope as a key

target for HIV vaccine design.

Understanding the mechanisms underlying the production of bNabs against HIV-1 in

some individuals during natural infection is of crucial importance for the development of

improved immunogens and immunization strategies. Gray et al. [165] showed that

patients infected with HIV-1 clade C rarely produce antibodies binding to the 2F5

neutralizing epitope in gp41 suggesting a correlation between HIV-1 subtype and

neutralizing response. However, other studies found limited to no impact of HIV -1

subtype in plasma neutralization, suggesting that HIV-1 group M subtypes and

neutralization response evolved independently [157, 162, 163, 327-329]. More recently,

in a large longitudinal Sub-Saharan HIV primary infection cohort, cross-clade plasma

neutralization was strongly correlated with subtype C infection [173]. Also, Rusert et al.

[146] found a strong association between plasma neutralization specificity and HIV-1

subtype, with subtype B viruses being more vulnerable to CD4 -binding-site specific

antibodies and non-B subtype viruses being more vulnerable to V2-glycan specific

neutralizing antibodies. In this study, V3-glycan and MPER-specific neutralizing

responses were independent of viral subtype. The differences observed between studies

might be related with the different assay conditions used to assess neutralizing activity,

103
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

in particular with the selected indicator virus panel that should represent the global HIV-

1 diversity and be standardized to allow inter-study comparison [162].

Considering that vaccine effectiveness will depend on the extent to which induced

antibodies will neutralize the global diversity of circulating HIV-1 variants, it is important

to characterize HIV-1 antibody responses in different epidemics and geographies. For

example, we have shown recently that the frequency and level of binding antibody

response to selected epitopes in the envelope transmembrane gp41 glycoprotein differ

between HIV-1 infected patients from Germany, France, and Portugal that have different

subtype distribution [330]. In Switzerland, data analysis of a large Swiss HIV Cohort

showed that ethnicity was associated with bnAb induction being black participants more

prone to develop bNAb responses [146].

The neutralizing antibody response of HIV-1 infected patients from Angolan has never

been evaluated. Angolan HIV-1 epidemic is peculiar, as it is driven by all subtypes and

multiple CRFs and URFs [5-7, 96-99, 212, 252]. In addition, because it is a very old

epidemic, highly divergent and ancestral forms of the different subtypes are often present

[5-7, 96-99, 331]. It has been suggested that the genetic complexity of the virus

quaisespecies present in HIV-1 individuals is directly related to the development of

neutralization breadth regardless of infection duration (reviewed in [171]). This should

be particularly evident in old epidemics such as the one of Angola. Hence, characterizing

the antibody responses and HIV-1 evolution in this population may provide new insights

into the development and evolution of the neutralizing antibody response against HIV-1

and into vaccine design. Here, we carried out the first detailed characterization of the

neutralizing antibody response against HIV-1 in Angola and identified viral and host

factors associated with the neutralizing response.

104
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Materials and Methods

Study population and ethics statement

This cross-sectional observational retrospective study included 322 HIV-1 infected

adults. Plasma samples were collected in 2001, 2009 and 2014 at the Hospital da Divina

Providência (HDP), a referral hospital in Luanda, the capital city of Angola. Eligible

participants had ≥19 years of age, were not pregnant and had a serological diagnosis of

HIV-1 [Determine HIV-1/2 (Abbott) and Uni-Gold Recombigen (Trinity Biotech) rapid

tests]. Plasma viral load and number of CD4+ T cells were determined in a subset of

patients using the Abbott Real Time HIV-1 assay (Abbott Laboratories) and the

ABACUS 5 Junior Hematology analyser, respectively. The study was conducted

according to the Declaration of Helsinki and was reviewed and approved by the National

Ethics Committee of Angola. The study was verbally explained to all the patients before

obtaining their written consent.

Cell lines

TZM-bl and HEK-293T cell lines were obtained from the NIH AIDS Reagent Program

([Link] TZM-bl cells were

engineered from HeLa cells that constitutively express CXCR4 to express large amounts

of CD4, CCR5 and a firefly luciferase reporter gene under the control of the HIV-1 LTR

[332]. Cells were cultured at 37oC, 5% CO2 using Dulbecco minimal essential medium

(DMEM) supplemented with 10 % heat-inactivated fetal bovine serum and with 100

units/ml of penicillin and 100 µg/ml of streptomycin.

Viral RNA extraction, PCR amplification, sequencing, and phylogenetic

analysis

105
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Before viral RNA extraction, 1 ml of plasma was centrifuged at 35.000rpm (61,793g) for

1h at 4ºC to concentrate viral particles. Supernatant was stored at -80oC for other

applications and pelleted material was resuspended with 560μl of Buffer AVL+RNA

carrier from the QIAmp® Viral RNA Mini Kit (Qiagen) and the manufacturer’s protocol

was followed. Reverse transcription was performed with NZY First-Strand cDNA

Synthesis Kit (NZYtech, Portugal) and a 534bp fragment comprising the C2V3C3 env

region was amplified by PCR using an in-house method described elsewhere [5, 7, 98,

99]. Sequencing of the C2V3C3 amplicons was performed with BigDye Terminator

Cycle Sequencing Kit (Applied Biosystems). Sequences were aligned using Muscle in

MEGA version 6 software [211, 333] with reference strains collected from the Los

Alamos HIV Sequence Database ([Link] Maximum-likelihood (ML)

phylogenetic analyses were performed using the best-fit model of nucleotide substitution

as estimated by Modeltest v3.7 under the Akaike information criterion [263]. Maximum

Likelihood (ML) trees were inferred with PhyML 3.0 [264, 334]. Tree searching was

done with nearest neighbor interchange (NNI) and subtree pruning and regrafting (SPR).

The reliability of the obtained topology was estimated with bootstrap with 500 replicates

[264, 334]. Determination of coreceptor usage was made based on the V3 loop sequence

using geno2pheno [coreceptor] webtool ([Link] [86]. False

positive rates (FPR) used were 10% as recommended [191]. Selective pressure was

examined with the DATAMONKEY web-server [335], after removing all positions

containing gaps and missing data from the dataset. All estimations were performed using

the MG94 codon substitution model crossed with the nucleotide substitution model GTR

previously selected with Modeltest. Four different approaches were used to identify

codons under selection: single-likelihood ancestor counting (SLAC), fixed-effects

likelihood (FEL), internal fixed effects likelihood (IFEL) and relaxed -effects likelihood

106
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

(REL) methods [336]. While SLAC, FEL and REL detect sites under selection at the

external branches of the phylogenetic three, IFEL identifies such sites only along the

internal branches. To classify a site as positively or negatively selected the cut-off P-value

was 10% for SLAC, FEL and IFEL. For REL, codons under selection were detected with

a cut-off value for the Bayes factor of 50.

Entropy and N-linked glycosylation analysis

Potential N-linked glycosylation sites were identified using the N-Glycosite software

[337], and the entropy at each amino acid position was measured with Shannon’s entropy-

one and Shannon’s entropy-two online tools, all available at the Los Alamos National

Laboratory HIV sequence database ([Link]

Production of C2V3C3 polypeptides and analysis of antibody reactivity

Six 178 amino acids polypeptides comprising the part of C2, V3 and part of C3 envelope

regions (position 212–390 in gp120 in HIV-1 HXB2) of HIV-1 isolates circulating in

Angola (subtypes C, G, H, J, and CRF02_AG) and Portugal (subtype B) were expressed

in Escherichia coli and purified as described previously [212]. Briefly, a DNA fragment

of 534 nucleotides comprising the C2, V3 and C3 coding regions (position 6858-7392 in

HIV-1 HXB2) was amplified from plasmids containing the full-length envelope gene

using the primers described elsewhere and cloned into the bacterial expression vector

pTrcHis (Invitrogen) [212]. Expression of C2V3C3 polypeptides in Escherichia coli

strain TOP10 was induced with isopropyl-β-D-thiogalactopyranoside (IPTG) according

to the manufacturer´s instructions, and protein purification was performed using

Dynabeads® His-tag Isolation & Pulldown (Life Technologies). Bradford assay (Bio-

Rad) was performed to determine protein concentration. Purified recombinant

polypeptides were analysed by SDS-12% PAGE.

107
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Antibody reactivity against these polypeptides was determined using an ELISA assay as

described [212]. In brief, 96- wells ELISA plates were coated overnight at 4 0C with 100

μl of 0.05M bicarbonate-coating buffer (pH=9.4) containing 100 ng of the different

C2V2C3 polypeptides. Plates were blocked with 2% gelatine in Tris-buffered saline

(TBS) 1X for 1h at room temperature. Serial dilutions of plasma (1:100 to 1:32 00) in

primary antibody buffer (TBS 1X + 0.05 % Tween 20 + 5 % blocking solution) were

added to the wells and the plates were incubated for 2h at room temperature. Goat anti-

human IgG conjugated to alkaline phosphatase, diluted 1:2000 in the primary antibody

buffer and 5% goat serum was added to the wells. Plates were washed between steps with

TBSt (TBS 1X + 0.05 % Tween 20). Plates were developed by adding Sigma -fast p-

nitrophenol phosphate diluted in deionized water. The plates were incubated for 20

minutes at room temperature away from light. Optical density was read at 405nm on a

microplate reader. The clinical cut-off value of the assay was calculated as the mean OD

value of HIV-seronegative samples plus 2 times the standard deviation (SD). Binding

antibody titers were calculated as the highest plasma dilutions giving a positive reaction

(OD / cut-off> 1).

Production of Env-pseudotyped viruses

A reference panel of 12 tier 2 HIV-1 Env-pseudotyped viruses of subtypes C (n=3), A

(n=1), CRF07_BC (n=2), CRF01_AE (n=2), B (n=2), G (n=1) and AC recombinant (n=1)

were produced using the Global panel of HIV-1 Env clones [162], obtained through the

NIH AIDS reagent program. Env-pseudotyped viruses were produced by transfection of

Env-expressing plasmids in 293T cells using PSG3.1Δenv as backbone, in a 1:3 ratio

using JetPRIME® DNA transfection reagent. Viral stocks were filtered through 0.45 µm

pore size filters after 48 hours and stored at -80 oC until use.

108
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Plasma neutralization assay

Neutralization of the Env-pseudotyped viruses was assessed in TZM-bl cells using Tat-

induced luciferase (Luc) reporter gene expression to quantify the reduction in virus

infection as described previously [338]. Briefly, TZM-BL cells (10,000 cells/well) were

seeded the day before the neutralization assay to allow adherence of the cells to the bottom

of the wells. Heat inactivated plasma samples (56 0C for 30 min) were incubated at 1:40

dilution in triplicate with the respective Env-pseudotyped virus for 1 hr at 37 0C before

transfer to TZM-bl cells. Following 48 hours, percent neutralization was determined by

calculating the difference in average RLU between test wells containing plasma samples

and the wells containing the Env-pseudotyped virus from the indicator panel after the

normalization of the results using the average RLU of cell control wells. Results were

considered valid if the average RLU of virus wells was >10 times the average RLU of

cell control wells. A virus pseudotyped with the envelope glycoprotein of vesicular

stomatitis virus (VSV-G) was used as neutralization specificity control.

Neutralizing antibody titers were determined for a subset of plasma samples showing

broad cross-neutralizing activity (n=38). In this case, 100 µL of 2-fold serial dilutions

beginning at 1:40 were mixed with 100 µL of each virus (200 TCID50/well) and

incubated for 1 h before adding to the cells. After 48 h, culture medium was removed

from each well, and plates were analyzed for luciferase activity as described above. Wells

with medium were used as background control, and virus-cell wells were included as

infection control. Neutralizing titer (ID50) was defined as the highest dilution for which

50% neutralization was achieved.

Neutralization score and plasma categorization

109
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

To categorize the neutralizing activity of the Angolan samples in terms of potency and

breadth we used a previously described scoring system [146, 167, 330]. A score of 0 was

attributed when neutralizing activity against a given virus of the panel was less than 20%,

a score of 1 when neutralization ranged between 20-<50%, a score of 2 for 50-<80%

neutralization and a score of 3 for ≥80% neutralization. The overall neutralization score

(NS) for a given plasma was obtained by adding the scores against the 12 Env-

pseudoviruses of the panel and reflects neutralization potency and breadth. As a validated

and worldwide accepted classification system to define neutralizing activity is lacking,

for the purpose of the present study we classified plasmas with scores 25 -36 as elite

neutralizers, 18-24 as broad neutralizers, 6-17 as cross neutralizers and <6 as weak or no

neutralizers. According to this classification an elite neutralizer sample must neutralize

≥9 viruses of the panel with a neutralization potency ≥80%.

Prediction of bnAb epitope specificities by clustering analysis

The neutralizing antibody specificities were determined for a subset of patients exhibiting

broad and elite neutralization capacity using cluster analysis with human bNAbs targeting

the main neutralizing epitopes on the viral envelope and capable to neutralize at least half

of the 12 Env-pseudotyped virus panel as described previously [162]. Neutralization

heatmaps and clusters were computed via the online tool ClustVis using a predefined

correlation clustering distance method (Pearson correlation subtracted from 1) based on

the average distance of all possible pairs. ClustVis is a web tool for visualizing clustering

of multivariate data (available at [Link] [339].

Statistical analysis

The statistical analysis was performed with GraphPad Prism version 5.01 or 9.0

(GraphPad Software Incorporated, San Diego, California, USA). The Mann -Whitney,

110
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Kruskal-Wallis or Fisher’s exact tests were used to compare differences between groups.

The Spearman rank test was used to quantify the magnitude and direction of the

correlation between antibody neutralization activity and plasma binding titers against

C2V3C3 polypeptides, CD4+ T cell counts, viral subtype and age of patients. Hypothesis

tests were two-tailed and P values <0.05 were considered significant.

To test the potential correlation between neutralization score and genetic distance of the

clinical samples to the neutralization panel viruses, we used amino acid sequences and

Hamming distances that included gaps as characters because: 1) neutralization occurs on

the amino acid level, 2) does not depend on the evolutionary path to a state-combination,

and 3) indels may have significant effects on antibody binding. Genetic distances were

calculated using DECIPHER [340], regression analysis was performed using R version

R-4.0.3 [341], and visualization using ggplot2 [342].

Results

Characterization of the study population and infecting HIV-1 isolates

Overall, 375 plasma samples from 322 adult HIV-1 infected patients from three sampling

years, 2001 (n=106), 2009 (n=210) and 2014 (n=59) were included in the analysis.

Epidemiological, clinical, demographic, and virological characterization of the patients

is given in Table S1. The median age of the patients was 34 years and most (n=242,

64.5%) were women. The main route of transmission was heterosexual contact (n=304,

81.1%). There were no significant differences in age and gender between sampling years.

The median plasma viral load (VL) at the time of sampling was significantly higher in

2001 relative to 2009 (4.2-fold higher) and 2014 (33.5-fold higher). The median number

of CD4+ T cells in 2014 was 1.8-fold higher when compared to 2009 (p=0.0015). The
111
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

significantly lower VL and higher CD4+ T cell number in 2014 is consistent with most

patients being on cART which was not the case in 2001 and 2009.

Sequencing and phylogenetic analysis of the C2V3C3 Env region was completed

successfully for 206 patients from 2001 (n=96/106, 90.6%) and 2009 (n=110/210,

52.4%). The following subtypes were identified: A1 (2001, n=33, 34.4%; 2009, n=32,

29.1%), A2 (2001, n=6, 6.3%; 2009, n=3, 2.7%), B (2001, n=2, 2.1%; 2009, n=2, 1.8%),

C (2001, n=12, 12.5%; 2009, n=30, 27.3%), D (2001, n=2, 2.1%; 2009, n=8, 7.3%), F1

(2001, n=5, 5.2%; 2009, n=6, 5.5%), G (2001, n=8, 8.3%; 2009, n=11, 10%), H (2001,

n=19, 19.8%; 2009, n=15, 13.6%), and J (2001, n=3, 3.1%; 2009, n=0, 0.0%). Untypable

U strains were 4.2% (n=6) in 2001 and 2.7% (n=3) in 2009 (Figure S1). Subtype A

prevailed in 2001 and 2009, but subtype C increased significantly (2.2 -fold, P= 0.0095)

in 2009. Out of the 176 isolates for which there were protease (PR) and C2V3C3

sequences available, 74 (42.0%) were non-recombinant and 102 (58.0%) were

recombinant. Recombinant strains prevailed over pure subtypes in 2001 and 2009 (Table

S2).

The genotypic analysis of tropism showed that most viruses were R5 in 2001 (82.3%,

N=79) and in 2009 (85.5%, N=94), without significant differen ces between sampling

years (Figure S2). Unfortunately, we could not sequence the C2V3C3 region from most

of the 2014 samples due to their low or undetectable viral load (Table S1). Moreover, the

lack of plasma prevented further analysis in samples collected in 2001.

Characterization of the antibody response

In total, 236 Angolan plasma samples were screened for neutralization breadth and

potency against the 12 ENV-pseudotyped indicator panel, 178 samples from 2009 and 58

from 2014, amounting to 2832 plasma/virus combinations (Figure S3). In 2009, 80.9%

112
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

(144/178) of Angolan patients had the capacity to neutralize at least one virus from the

indicator panel; this increased to 93.1% (54/58) in 2014 (Figure 1). Likewise, the mean

percent neutralization (27.43%, 95%CI: 25.94, 28.92 in 2009 vs 60.52%, 95%CI: 57.37,

63.66 in 2014, p<0.0001) and the mean neutralization breadth (4.39, 95%CI: 3.83, 4.95

in 2009 vs 8.40, 95%CI: 7.32, 9.48 in 2014, p<0.0001) were higher in 2014 relative to

2009.

Figure 1– Neutralization potency and breadth per sampling year. A) Potency of

neutralization (% neutralization at 1:40 plasma dilution) of samples collected in 2009 and
2014 as assessed against the 12-Env-pseudotyped virus indicator panel. Mean and 95%
confidence intervals are shown. B) Neutralization breadth (number of Env-pseudotyped
virus neutralized with a neutralization value >20%) in samples collected in 2009 and
2014. Median and interquartile range are shown. P values were obtained using the Mann
Whitney U test.

Percent neutralization for each plasma-virus combination was recorded as a breadth-

potency matrix: ≥80% neutralization received a score of 3, 50% to <80% a score of 2,

20% to <50% a score of 1, and <20% received a score of 0. Plasma samples were then

ranked by the sum of scores in order to reflect their potency and breadth [146, 167].

Breadth, potency and neutralization score were directly correlated as expected (Figure

S4).
113
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Mean NS was 11.71, 95% CI [10.22, 13.19] ranging from 0 to 36 and median was 7, IQR

[2.0, 21.0]. Remarkably, approximately 30% (n= 68/236) of the patients developed

antibody responses with the capacity to potently neutralize at least half the viruses from

the panel (Figure 2A). Overall, considering both sampling years, 18.6% (n= 44/236) of

study participants were elite neutralizers (NS ≥ 25), 10.2% (n= 24/236) were broad

neutralizers (18 ≤ NS < 25), 27.1% (n= 64/236) were cross neutralizers (6 ≤ NS < 17),

and 44.1% (n= 104/236) were weak neutralizers or did not neutralize any virus of the

panel (Figure 2B).

Correlates of the neutralizing response

The neutralizing antibody response has been previously associated with viral load, CD4+

T cell count, viral diversity and infection time [146, 173]. We first analyzed the impact

of sample collection time on the neutralizing antibody responses of the HIV infected

Angolan patients. Strikingly, median NS was 6.2-fold higher in 2014 relative to 2009

(31.00, IQR [10.50, 33.00] vs 5.00, IQR [1.00, 13.25], p<0.0001) (Figure 2A). Consistent

with this, the frequency of elite neutralizers was 9.5-fold higher in 2014 than in 2009

[57% (n=33/58) vs 6% (n= 11/178), p<0.0001], and weak or no neutralizers were 2.7-

fold more frequent in 2009 than in 2014 [52% (n= 93/178) vs 19.0% (n= 11/58),

p<0.0001] (Figure 2B). Broad neutralizers were 2.4-fold more frequent in 2009 than in

2014 [12% (n= 21/178) vs 5% (n= 3/58), p=0.2107] and a similar trend was observed for

cross neutralizers [30% (n= 53/179) vs 19.0% (n= 11/58), p=0.1270]. We also analysed

matching plasma pairs from 2009 and 2014 to determine the evolution of neutralizing

antibody response as a function of infection time. In line with the previous results,

neutralizing score increased in 2014 relative to 2009 in 31 out of the 38 matched plasma

pairs analysed (81.6%). (Figure 2C). The NS was unrelated with the sex of the patients

(Figure 2D).

114
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Figure 2- Neutralization score (NS) in HIV-1 infected patients from Angola as a function
of year of sampling and sex. A) NS in 2009 is represented in blue and in 2014 in red; NS
in all patients is in green. B) Angolan patients were categorized into 4 groups according
to the NS as follows: no or weak neutralizers, <6 (grey); Cross neutralizers, 6 -17 (yellow);
Broad neutralizers, 18-24 (orange); Elite neutralizers, 25-36 (red). C) Neutralization score
in matched samples collected in 2009 and 2014, showing the NS categories. D) NS in
males and females. Median and interquartile range are shown. P values were obtained
using the Mann Whitney U test.

The 50% neutralization titers (ID50) against the 12 Env-pseudotyped virus indicator

panel were determined in a subset of plasma samples from 2009 (n=28) and 2014 (n=10)

showing broad and elite neutralizing activity (Figure 3A). When comparing unmatched

samples, neutralization titers were significantly higher in 2014 than in 2009 [median

log10 ID50 in 2009= 1.903, IQ: 1.602-2.505 (n=336 plasma-virus pairs) vs median log10

ID50 in 2014= 2.204, IQ: 1.903- 2.806 (n= 120 plasma-virus pairs), p=0.0013] (Figure3

A/B)

115
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Figure 3- Antibody neutralization titers in a subset of unmatched plasma samples from

elite and broad neutralizers from 2009 (n=28) and 2014 (n=10). A) Heatmap of the
neutralization titers (ID50) and neutralization score against the 12 Env -pseudotyped virus
indicator panel. ID50 values are color-coded, with darker colors implying higher ID50
values. *HIV subtype determined in the pol gene. + Number of CD4+ T cells determined
in 2009; ND- not done due to lack of sample. B) Comparison of antibody neutralization

116
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

titers in 2009 and 2014. Log10 ID50 values obtained by each patient sample against the
12 Env-pseudotyped virus indicator panel are plotted. Lines indicate the median with
interquartile range. P value was obtained using the Mann Whitney U test.

Overall, these results suggest that duration of infection is an important correlate of the

potency and breadth of the neutralizing antibody response in Angolan patients infected

with HIV-1.

To analyse the impact of HIV-1 subtype on neutralization by Angolan samples we

compared the neutralization score (NS) in patients infected with subtypes C (n=27) and

A1 (n=26), the two prevailing subtypes in Angola, and in patients infected with the other

subtypes and recombinant forms (n=56). Only samples collected in 2009 were included

in this analysis due to the limited number of samples genotyped in 2014. NS varied

significantly with infecting virus subtype (p=0.014), with subtype C leading to

significantly higher NS than subtype A1 [median NS= 17.00, IQR (6.00, 25.00) vs 6.00

IQR (3.50, 15.00), p=0.0103] or other subtypes [median NS= 17.00, IQR (6.00, 25.00) vs

8.00, IQR (1.00, 17.75), p=0.0087] (Figure 4A). These results indicate that virus subtype

is a major determinant of the neutralizing antibody response in our patients.

The indicator virus panel used in the neutralization experiments contains three subtype C

strains (25710, CE1176, and CE0217) that could be more closely related to the subtype

C isolates from the Angolan patients and explain the higher NS observed in patients

infected with subtype C viruses. To examine this issue, we compared the susceptibility of

the reference panel isolates to neutralization and found a significant variation related to

virus subtype (Figure 4B). The easiest viruses to neutralize were isolate 25710, a subtype

C from India, and 398F1, a subtype A from Tanzania. On the other hand, viruses most

resistant to neutralization were 2278, a subtype B from Spain and CNE8, a CRF01_AE

117
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

from China. Interestingly, subtype C isolates CE1176 and CE0217 from Malawi were

significantly more resistant to neutralization than 25710 suggesting a closer relationship

of subtype C isolates from Angola to this Indian subtype C isolate and showing that

subtype per is not the main determinant of susceptibility to antibody neutralization. To

investigate the impact of the evolutionary distance between the HIV-1 Angolan isolates

and the indicator virus panel on the neutralizing antibody responses, we aligned the

C2V3C3 amino acid sequences f rom the Angolan isolates (year 2009) with those from

the indicator virus panel. As expected, subtype C viruses from the patients were more

closely related with subtype C viruses of the indicator panel relative to other subtypes

(Figure S5A). There was a significant negative correlation of amino acid distance of the

indicator panel to NS (Spearman r= -0.2319, p = 0.019) (Figure S5B). Hence, the closer

the isolate from the indicator panel was to the patient’s C2V3C3 amino acid sequence,

the easier it was neutralized. On average, clade C reference strain 25710 from the

indicator panel was the closest indicator virus panel member to the Angolan isolates and,

not surprisingly, it was the easiest virus to neutralize. At the other end, clade B reference

strain 2278 was the furthest away from the C2V3C3 Angolan sequences and was the most

difficult virus to neutralize along with the CRF01_AE virus (CNE8). Nevertheless, many

patients infected with all subtypes developed potent bNAb responses despite the high

genetic distance to the viruses of the indicator panel, indicating that other factors besides

the relatedness with the indicator panel contribute to the potency of the neutralizing

response.

118
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Figure 4– Impact of HIV-1 subtype on antibody neutralization. A) Neutralization score

in patients infected with the two most common subtypes in Angola (year 2009), C (n=27)
and A1 (n=26), and in patients infected with other subtypes and recombinant forms
(n=56). The Kruskal-Wallis nonparametric test was used to analyse the difference in
median NS for all subtypes (p=0.014). Dunns multiple comparison test was used to
analyse differences in NS between subtypes. Median and interquartile range are shown.
B) Percent neutralization of each of the 12 Env-pseudotyped virus indicator panel by the
plasma samples (at 1:40 dilution) from the Angolan patients (N=236). Mean percent
neutralization and 95% confidence interval bars against a given virus from the indicator
panel are shown. Statistically significant differences are represente d by the P values
obtained with Dunns multiple comparison test. *** p<0.0001, ** p<0.001, *p<0.05.

Drug naïve patients (year 2009) with ≤200 CD4+ T cells/μl at study entry had

significantly higher NS values than patients with >200 CD4+ T cells/μl [median NS in

patients ≤200 CD4+ T cell counts was 7.00 (IQR, 3.50, 21.00) vs 4.00 (1.00, 12.00) in

patients with > 200 CD4+ T cell counts, p = 0.0193] (Figure 5A). Moreover, NS values

119
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

were inversely associated with CD4+ T cell counts (Spearman r=-0.3043, p=0.0005)

(Figure 5B) and directly associated with age (Spearman r=0.1644, p =0.0302) in these

patients (Figure 5C). These results suggest that elicitation of high levels of broadly

neutralizing antibodies in these patients is directly correlated with prolonged a ntigenic

stimulation [343].

Figure 5- Correlation between neutralization score, CD4+ T cell counts and patient’s age.
A) Neutralization score differences between 2009 patient’s with ≤ 200 CD4+ T cell
counts at study entry and patients with >200 CD4 T cell counts. Median and interquartile
range are shown. P values were obtained using the Mann Whitney U test; B) Correlation
of neutralization score with CD4+ T cell counts in 2009 and 2014; C) Correlation of
neutralization score with patient’s age in 2009 and 2014. Samples collected in 2009 are
shown in blue and samples collected in 2014 in red. Linear trend is shown with mean and
95% CI bands; Spearman r and P values are indicated.

Epitope specificities of the plasma neutralizing antibodies

120
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

In the same subset of 38 plasma samples (n=28 from 2009 and n=10 from 2014) from

broad and elite neutralizers, the epitope specificities were mapped using a computational

clustering tool based on the epitope specificities of a panel of human bNAbs [339]. Six

(15.8%) samples did not cluster with any of the bNAbs. Thirty -two (84.2%) samples

clustered with one of the bNAbs. Of these most samples (68.8%, 22/32) clustered with

PGT128 and 2G12, two bnAbs that target the V3 glycan supersite with important contact

residues in V3 and V4 (Figure 6) [188, 204, 328]. Five (15.6%) samples clustered with

bnAb 4E10 that targets the gp41 membrane-proximal external region (MPER) [344]. Four

(12.5%) samples clustered with VRC01 and VRC-CH31 bnAbs that target the CD4

binding site. Finally, one (3.1%) sample clustered with PG16 and PG9 that target the

V1V2 glycans. These results indicate that the V3 glycan supersite is the dominant broadly

neutralizing epitope in Angolan patients.

Figure 6- Cluster analysis and heatmap of the predicted epitope specificity in the top
neutralizing patients from Angola. In the top of the columns, known bnAb epitopes are
coloured according to the respective epitope specificities as shown by the legend. The

121
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

identification of the plasma samples and bnAbs is shown in the bottom of the columns.
Cluster analysis for both rows and columns were computed according to the Pearson
correlation [339]. Blue colours in the heatmap represent lower neutralization activity and
red colours higher neutralization activity. Each column represents the neutralization
values of a given plasma sample or a bnAb of known specificities against the 12 Env -
pseudotyped virus panel whose names are indicated to the left.

Neutralization score is directly related with titer of C2V3C3-binding

antibodies in all subtypes

The antibody binding reactivity against a panel of recombinant polypeptides comprising

the C2, V3, and C3 envelope regions of subtypes B, C, G, H, J, and CRF02_AG was

characterized in a subset of samples from 2009 (n=48) and 2014 (n=16) with known

antibody neutralization profile. All but the B polypeptide were derived from Angolan

isolates. All but six samples from five patients reacted with all C2V3C3 polypeptides

demonstrating the high antigenicity of this envelope region (Figure S6). In 2009, patients

had significantly higher median antibody binding titers against subtype C than against

subtypes G (p=0.0007), H (p=0.0282), J (p=0.0052), and CRF02_AG (p=0.0149). Of

note, median antibody binding titers were always higher in 2014 relative to 2009

regardless of the C2V3C3 polypeptide subtype but this was not significant except for

CRF02_AG.

The higher antibody reactivity against subtype C antigen could be related with the higher

neutralizing responses observed in subtype C infected patients. We therefore investigated

possible associations between neutralization score, C2V3C3 antibody binding titer and

subtype. Remarkably, C2V3C3 antibody binding titer was positively associated with NS

values, i.e., patients with higher antibody binding titers to C2V3C3 polypep tides had

higher neutralizing antibody responses (Figure 7). This was significant for all subtypes

122
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

of the C2V3C3 recombinant polypeptides tested confirming this epitope as an important

neutralizing domain in these patients independent of subtype.

B
C
CRF02_AG
4.0
4 4
r=0.5204 O

Ab binding titer
r=0.4323 3.5 O O r=0.2753

Ab binding titer
Ab binding titer

O O O O O p<0.0001 O O O
3 p=0.0030 O
3 O O O O O O O OO
O O O O p=0.0416
O

(log10)
O O O O O

(log10)
(log10)

OO
O O 3.0
O OO O OO O OO O O O O
2O 2 O O O
2.5
O O
O
1 1
2.0 O

0 O 0 1.5
0 10 20 30 40 0 10 20 30 40 0 10 20 30 40
Neutralization Score Neutralization Score Neutralization Score

G H J
4 5 4
r=0.6045 r=0.4992
Ab binding titer

Ab binding titer
Ab binding titer

O O 4 O O O
3 O
p<0.0001 r=0.4058 3 O O O O p=0.0005
(log10)

O O O O O O p=0.0058 O OO
O
(log10)

(log10)
O OO 3 O O O O O O O O
OO O O
2O O O O O
O OOO 2 OOO O
2 O

1 1
1

0 OO O 0 0 O
0 10 20 30 40 0 10 20 30 40 0 10 20 30 40
Neutralization Score Neutralization Score Neutralization Score

Figure 7– Association between antibody binding titer to C2V3C3 recombinant

polypeptides of different subtypes and neutralization score (year 2009). Filled symbols
are the antibody binding titers of the broad/elite neutralizers to a given C2V3C3 subtype.
Unfilled symbols are the antibody binding titers of the no/weak and cross neutralizers to
a given C2V3C3 subtype. Associations were assessed by Spearman analyses. P-values
and Spearman r values are indicated. Linear trend is shown with mean and 95% CI bands.

Impact of the neutralizing antibodies in the diversity and evolution of

C2V3C3

Neutralizing antibodies targeting the C2, V3 and C3 envelope regions are common in

HIV-1 infected individuals [173] and escape from these antibodies leads to higher

diversity in these regions as well as to higher positive selection and convergent evolution

123
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

[345-347]. To investigate the impact of the neutralizing antibodies in the diversity and

evolution of the envelope glycoproteins of the viruses infecting our patients, we analysed

amino acid entropy and the sites under selective pressure in the C2, V3 and C3 regions in

the different neutralization categories for samples collected in 2009. Considering the three

regions together, mean overall entropy values were similar in all neutralization categories:

weak/no neutralizers= 0.5727 [95% confidence interval (CI): 0.4753, 0. 6241]; cross

neutralizers= 0.5931 (0.5012, 0.6849); broad neutralizers= 0.5381 (0.4472, 0.6291); and

elite neutralizers= 0.4106 (0.3313, 0.4899). Regardless of neutralization category, the

region with higher mean entropy was C3 [0.8528 (0.7556, 0.9500)] fo llowed by V3

[0.4659 (0.3903, 0.5414)] and C2 [0.3635 (0.3092, 0.4178)] (p<0.0001). We then plotted

Shannon’s entropy differences in C2V3C3 between no/weak neutralizers and cross, broad

and elite neutralizers. This analysis revealed that viruses from elite neutralizers were far

less variable than viruses from weak/no neutralizers as seen by the number of amino acids

with positive entropy differences relative to amino acids with negative entropy

differences (37 vs 16 sites, p=0.0023) (Figure 8). On the other hand, viruses from broad

and cross neutralizers did not vary significantly from viruses from no/weak neutralizers.

Relative to no/weak neutralizers, the most variable amino acid residues in the broad and

elite neutralizers were found in V3 and/or C3 (broad neutralizers: 1 site in C2 vs 7 sites

in C3, p<0.0001; elite neutralizers: 3 sites in C2 vs 13 sites in V3C3, p<0.0001), two

regions that contain broadly neutralizing epitopes (Figure 8).

Diversifying selection in C2V3C3 varied according to the different neutralization

categories (p<0.001) (Table S3). Considering only sites that were selected by at least two

methods, weak/no neutralizers had a total of 9 positively selected sites, cross neutralizers

6, broad neutralizers 3 and elite neutralizers 1. Regardless of neutralization category most

sites under selective pressure were present in the C3 region.

124
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Figure 8 – Amino acid entropy difference in the C2V3C3 region between neutralization
categories. A) Shannon’s entropy difference between No/Weak and Cross neutralizers.
B) Shannon’s entropy difference between No/Weak and Broad neutralizers. C) Shannon’s
entropy difference between No/Weak and Elite neutralizers. Sites with significant entropy
difference (p ≤0.05) are shown in red. Gray boxes delimitate the V3 region. Numbers in
the x axes indicate the amino acid position in HIV-1 HXB2.

The mean number of N-glycosylation sites in C2V3C3 was similar in all neutralization

categories [Non-neutralizers: 9.2 (range: 8-12); cross-neutralizers: 9.2 (range: 7-11),

broad-neutralizers: 9.4 (range: 7-11); elite-neutralizers: 9.8 (range: 8-11)] (Table S4). C3

had more potential N-glycosylation sites than C2 or V3 but sites in V3 and C2 were more

conserved. For example, sites 241, 262, 276 and 289 in C3 were present in ≥70% of

strains and site 301 in the V3 crown was present in all but two strains (97% ). In C3, site

332, which together with site 301 in V3 and other elements in V1, V3 and V4 is part of

125
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

the V3-glycan supersite [170, 348], was highly conserved (70%) in all neutralization

categories.

Discussion

HIV-1 was introduced in Angola from Kinshasa, the capital city of the Democratic

Republic of Congo (DRC), likely in 1910-1940 making the Angolan epidemic the second

oldest in the world [6]. Like in the DRC, the Angolan epidemic has been driven by all

subtypes but B, untypable and highly divergent strains, and multiple CRFs and URFs [5,

96, 99, 252]. In this study we confirmed the extremely high diversity and evolving

complexity of HIV-1 strains present in Angola. Subtypes A and C dominated over other

subtypes but all other Env subtypes were present along with untypable basal strains and

recombinant strains that prevailed over pure subtypes. The remarkable diversity and

evolution of HIV-1 in Angola is driven by the increasing number of new infections [2],

the limited access to antiretroviral therapy, and high levels of drug resistance [99, 349].

The high diversity and rapid evolution of HIV-1 in this country can pose a serious

challenge to vaccination and other preventive efforts. At the individual level, the long-

term B cell stimulation by this highly diversified ensemble of viruses may have promoted

the development of exceptional neutralization breadth [146, 166, 170, 172, 173, 184, 350,

351]. We found that the majority (56%) of the patients in our cohort developed cross,

broad, or elite neutralizing responses. These results far exceed those from previous cohort

studies in sub-Saharan Africa [167, 172, 173, 176, 352]. For example, Beirnaert et al.

found 10.6% broad neutralizers in Cameroon[352] and Landais et al. found about 15%

broad neutralizers in a cohort of HIV-1 infected patients from Eastern and South

Africa[173]. When compared to cohort studies from other geographies where subtype B

dominates, the frequency of patients with bNAb responses reported in our study was also

126
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

much higher[146, 163, 353]. For example, Rusert et al. [146] in Switzerland found that

most patients (79.1%) showed weak or no neutralization breadth, which compares to 44%

in our cohort, and that only 1.3% were elite neutralizers which compares to 19% in our

cohort. This divergence may be related with many factors besides the diversity of

infecting viruses, such as the HLA genotype and ethnicity of the patients, viral load,

CD4+ T cell counts, and duration of infection [146, 166, 172, 173, 354].

In agreement with other studies, neutralization score was inversely correlated with CD4+

T cell counts in 2009 when the patients were naïve to ART and had high viral load s [146,

166, 170, 173]. This is generally associated with high envelope stimulation of B cells and

inevitably leads to B-cell exhaustion in chronic viraemic HIV-1 infection [355].

Remarkably, however, the frequency of elite neutralizers and the mean neutralization

score in matched and unmatched patients increased significantly in 2014, when patients

were already undergoing ART, relative to 2009. The boost in the quality of the

neutralizing response in these patients suggest good restoration of the B cell compartment

with ART which is uncommon in chronic HIV-1 infection [355-357]. Moreover, the

moderate level of plasma viremia (median 11,660 HIV-1 RNA copies /ml, IQR, 380-

30,060) found in these patients may have provided the low-level antigenic stimulation

needed for the full maturation of memory B cells and bNAb production [355, 358, 359].

This has precedent in HIV-2 infection where most patients are infected for long periods

and produce potent and broadly neutralizing responses in a setting of low plasma viremia

[324, 360, 361]. In this model, B cell exposure to low-level envelope antigens, likely in

lymphoid tissues, during prolonged infection periods leads to the generation of highly

specific envelope C2V3C3- specific antibodies as well as broad and potent neutralizing

antibodies [317].

127
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Viral type and subtype as well as the nature of the epitope target on the viral envelope

impact the antibody maturation process as seen by the frequency of elicitation [212, 361]

and epitope specificity of bnAbs [146, 171, 348]. Differences in envelope structure and

epitope exposure, length of variable loops, type of V3 motifs, N-glycosylation patterns,

and conservation of key sites have helped to explain why certain HIV-1 subtypes like

subtype C are better at promoting the elicitation of neutralizing antibodies [146, 171, 173,

176, 212, 348, 362]. In line with these studies, we found that infection with subtype C

viruses was associated with enhanced neutralization breadth and potency. In general,

subtype C infected individuals have shown a bias to V2 -glycan directed antibody

responses, and subtype C envelope from transmitted viruses have been less prone to

neutralization by V3-directed antibodies due to the absence of the N332- glycan in the

C3 region [146, 173, 176, 348, 363-365]. This was not the case in our study as most top

neutralizers had V3-directed antibodies that were able to neutralize the subtype C isolates

from the virus panel, and the N301 and N332 glycans defining the V3 -glycan supersite

were highly conserved in the patient’s isolates. Supporting the major role of the V3 and

C3 envelope regions in the development of bNAbs in our cohort, we found a strong direct

correlation between the titer of antibodies binding to C2V3C3 envelope polypeptides

from all subtypes and neutralization score. Nonetheless, antibodies specific for the V2

apex, the CD4 binding site, the gp41 MPER and/or unknown epitopes were also found in

some patients revealing the complexity of the neutralizing antibody responses in th ese

patients.

We also looked at the variability of patient’s sequences in the envelope C2V3C3 region

to assess the impact of escape from neutralizing antibodies on viral evolution and

diversity. V3 and C3 were the most variable regions which is consistent with the dominant

role of neutralizing antibodies targeting these regions in these patients [173]. V3 bNAb

128
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

recognition sites and sites associated with resistance to neutralization such as N295 [166,

170, 173] were under positive selection in broad and elite neutralizers. However, elite

neutralizers exhibited far less variability and lower number of sites under selective

pressure in V3 and C3 relative to weak or low neutralizers. The convergence of the viral

swarm to a few resistant strains provides convincing evidence for the crucial role of V3-

and C3-directed bNAbs in controlling HIV-1 replication and diversification in these

patients [176, 212, 316, 366].

In conclusion, an exceptionally high number of Angolan patients infected with HIV-1

elicits broad and elite neutralizing antibodies mostly targeting the V3 -glycan supersite.

This is associated with long-term and low-level V3- and C3- antigenic stimulation by the

highly diverse isolates circulating in this country especially subtype C. These results have

direct implications for the design of a new generation of HIV-1 vaccines.

Acknowledgments

We greatly acknowledge the contribution and efforts of the staff an d patients from the

Hospital da Divina Providência in Luanda for this study. TZM-bl cells were obtained

through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH. This work was

supported by Fundação para a Ciência e a Tecnologia (FCT), Portugal, u nder project

grants UIDB/04138/2020 and UIDP/04138/2020. This study was in part supported by the

NIH/NIAID under grant R01AI087520. Francisco Martin was supported by FCT under

PhD grant number SFRH/BD/87488/2012. CP is funded by FCT under a contract-

program as defined by DL No. 57/2016 and Law No. 57/2017.

Supporting Information

Table S1- Characteristics of the HIV-1-infected patients.

129
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Characteristics 2001 2009 2014

Total number of patients, n (%) 106 (28.3) 210 (56.0) 59* (15.7)
Age (years) median (IQR) 32 (26-40) 32 (28-39) 39 (36-46)
Sex, n (%)
Female 53 (50.0) 145 (69.0) 44 (74.6)
Male 43 (40.6) 65 (31.0) 15 (25.4)
Unknown 10 (9.4) -- --
Geographic origin, n (%):
Angola 94 (88.7) 207 (98.6) 57 (96.6)
RDC -- 3 (1.4) 1 (1.7)
Unknown 12 (11.3) -- 1 (1.7)
HIV-1 mode of transmission, n (%):
Heterosexual 38 (35.8) 210 (100.0) 56 (94.9)
Bisexual 4 (3.8) -- --
IDU 1 (0.9) -- --
Transfusion 1 (0.9) -- --
Unknown 62 (58.5) -- 3 (5.1)
CD4+ T cell count -- N=162 N=21
CD4+ T cell count/mm 3, median N/A 265 (133-448) 475 (343-569)
(IQR)
Plasma viral load N=16 N=71 N=13
VL (copies/ml), median (IQR) 390,877 (209,172- 93,391 (28,222- 11,660 (380-
704,286) 510,579) 30,060)
Undetectable, n (%) -- -- N=9 (69.2)
Unknown, n (%) 90 (84.9) 139 (66.2) 46 (78.0)
Co-morbidities, n (%):
Tuberculosis -- 33 (15.7) --
HBV -- 17 (8.1) --
TB+HBV co-infections -- 3 (1.4) --
Other -- 48 (22.9) --
WHO Clinical stage, n (%):
Asymptomatic 13 (12.3) -- --

130
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Symptomatic intermediate 20 (18.9) -- 1 (1.7)

AIDS 11 (10.4) -- 1 (1.7)
Unknown 62 (58.5) 210 (100.0) 57 (96.6)
cART, n (%):
cART-naïve 102 (96.2) 202 (96.2) 1 (1.7)
cART-exposed 4 (3.8) 1 (0.5) 21 (35.6)
Unknown -- 7 (3.3) 37 (62.7)
Legend: N/A, not available; RDC, Republic Democratic of Congo; IDU, intravenous drug user;
VL, viral load; cART, combined antiretroviral therapy; IQR, interquartile range; HBV, Hepatitis B
virus. *53/59 HIV-1 infected patients were followed longitudinally from 2009.

Table S2 - Main C2V3C3 subtypes in Angola in 2001 and 2009

Genetic forms 2001 2009 P value a

N (%) N (%)

Pure subtypes 35/88 (39.8) 39/88 (44.3) 0.6470

Recombinant forms 53/88 (60.2) 49/88 (55.7)

Subtype A 33/96 (34.4) 32/110 (29.1) 0.4542

Subtype C 12/96 (12.5) 30/110 (27.3) 0.0095

Subtype H 19/96 (19.8) 15/110 (13.6) 0.2625

aFisher’s exact test

131
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

Table S3- Positively selected sites in the C2, V3 and C3 regions in the four neutralization

categories selected at least by two methods

Neutralization Codon* SLAC p- REL PP FEL p- IFEL p-

category value value value

no/Weak 293 3.673 0.101 1.191 1.000 1.028 0.254 4.219 0.059

335 3.816 0.048 1.612 0.993 0.809 0.036 1.376 0.016

336 2.793 0.139 1.307 1.000 3.023 0.105 8.145 0.020

343 3.820 0.092 1.237 1.000 0.418 0.650 0.316 0.747

344 1.933 0.277 1.457 0.997 0.599 0.053 0.067 0.853

346 2.924 0.134 1.386 1.000 1.587 0.046 0.781 0.089

347 4.361 0.039 1.414 1.000 0.980 0.100 0.432 0.440

361 3.785 0.097 1.267 1.000 0.540 0.715 0.420 0.754

362 4.384 0.034 1.307 1.000 1.015 0.140 0.344 0.562

Cross 318 2.646 0.085 0.496 0.647 0.317 0.053 0.000 1.000

336 2.582 0.096 0.719 0.889 0.735 0.045 0.522 0.152

337 3.791 0.041 0.721 0.887 0.520 0.060 0.132 0.469

344 2.185 0.188 0.852 0.989 0.599 0.012 1.184 0.095

346 4.272 0.009 0.862 0.991 0.587 0.002 0.366 0.066

365 2.034 0.053 -0.040 0.068 0.210 0.038 0.000 1.000

Broad 335 2.928 0.055 0.935 0.835 0.999 0.047 0.000 1.000

347 3.383 0.034 0.927 0.830 1.282 0.031 0.603 0.409

363 2.908 0.068 0.583 0.609 0.978 0.274 4.188 0.027

Elite 295 2.067 0.138 4.302 1.000 10.915 0.020 8.826 0.724

*Codons selected with 10% level of significance (SLAC, FEL and IFEL) or above a
Bayes Factor of 50 (REL) selected by at least 2 methods and numbered according to
codon position of HIV-1 HXB2. PP, posterior probabilities. Codons selected

132
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

simultaneously by SLAC, FEL and REL are bold and underlined. Bold dN-dS differences
correspond to significant P-values or posterior probabilities.

Table S4- Frequency and distribution of potential N-glycosylation sites in the C2, V3 and

C3 regions across neutralization categories.

*Relevant N-glycosylation sites are highlighted and coloured according the position in
C2V3C3. Higher frequency glycosylation sites are boxed in red. Sites were numbered
according to the reference strain HIV-1 HXB2.

133
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

2 Figure S1- Phylogenetic relationship between the Angolan HIV-1 C2V3C3 sequences.

3 Maximum likelihood phylogenetic tree of C2V3C3 region was constructed with reference

4 sequences from all HIV-1 subtypes (yellow dots) with the 2001 (blue dots) and 2009

5 (green dots) Angolan sequences and the virus sequences from the indicator panel (red

6 dots).

134
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

8 A) B)

10 Figure S2- HIV tropism and amino acid diversity in 2001 and 2009. A) HIV V3 -based

11 tropism as determine in geno2pheno considering a false positive rate cut-off of 10%. The

12 red dots are predicted X4 tropic virus. B) C2V3C3 amino acid diversity as assessed by

13 Shannon’s entropy. Mean entropy in the C2V3C3 region for each patient is shown, 2001

14 samples are represented in red filled dots and 2009 in blue unfilled dots. Variability at

15 the amino acid level was calculated using Shannon’s entropy -one online tool

16 ([Link] Mean and

17 95% confidence intervals are represented. P values were obtained using the Mann

18 Whitney U test.

19

20

21

22

23

24

25

26

135
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

136
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

53

54

55

56

57

58

59

60

61

62 Figure S3- Heatmap showing the neutralizing activity of plasma samples from 2009 and

63 2014 against the 12-Env pseudotyped virus indicator panel. Percent neutralization was

64 determined in TZM-bl cells with plasma samples diluted 1:40. White cells indicates non

65 determined values; Yellow cells indicate <20% neutralization; orange highlighting

66 indicates 20 to <50% neutralization; light brown highlighting indicates 50% to <80%

67 neutralization; red highlighting indicates ≥80% neutralization. Virus subtype is indicated

68 below the isolate common name of the Env-pseudotyped virus.

69

70 Figure S4- Neutralization breadth and potency predict neutralization score (NS). A)

71 Correlation between neutralization breadth and potency in the 236 samples. B)

137
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

72 Correlation between NS and breadth. C) Correlation between NS and potency. Breadth

73 was considered the number of pseudoviruses that were neutralized >20% and potency

74 was the geometric mean of %neutralization against a given virus of the 12-virus indicator

75 panel. Linear trend is shown with mean and 95% CI bands. The linear regression line is

76 represented showing mean and 95% confidence interval error bands, goodness of fit r2

77 and P values are indicated.

138
 Chapter 4- Neutralizing antibody responses against HIV-1 in Angola

78 Figure S5- Impact of HIV-1 clade on antibody neutralization of the 12-virus panel. A)

79 C2V3C3 amino acid distance of the viruses from the indicator panel to the viruses

80 infecting the patients. B) Correlation between neutralization score and C2V3C3 amino

81 acid distance of the viruses from the indicator panel to the viruses infecting the patients.

82 Subtype of the patient’s virus is indicated by color, linear trend is shown with mean and

83 95% CI bands.

84

85

86

87

88

89

90

91

92

93

94

95

96

97 Figure S6- Antibody binding titers against the C2V3C3 recombinant polypeptides of

98 different subtypes in patients from 2009 and 2014. Blue circles correspond to patients

99 from 2009 and red circles to patients from 2014. Median and interquartile range are

100 shown. P values were obtained using the Mann Whitney U test. P values <0.05 are shown

101 in bold.

139
140
 Chapter 5

General Discussion and Conclusions

141
 Chapter 5- General discussion and conclusion

In Chapter 2 of this thesis we characterized the HIV-1 diversity, transmission dynamics

and prevalence of TDR in Luanda in 2009, five years after the scale-up of ART and

compared these data with our previous survey performed in 2001 [5, 7, 98, 99]. Like in

2001, no major PIs resistance mutations were found in the study population which is

consistent with the fact that first-line regimens used in Angola didn’t include PIs in 2009

[285]. The K103N mutation, which confers high-level resistance to nevirapine, and

efavirenz [270], was found in only one patient accounting for a 0.7% prevalence rate of

TDR which was lower than the 2001 survey [7]. This residual TDR prevalence was

similar to that of several African countries that also use the public health approach to

ART [200, 201, 286-288]. However, more recent studies conducted in Luanda, by

Sebastião et al showed an increase in HIV drug resistance, where more than 17% of drug

naïve HIV-1 infected pregnant women presented DRMs mainly to the NNRTIs which is

particular concerning due to the limited number of ART options available in Angola for

the treatment of HIV-1 infection, that is still mainly based on NNRTIs with a backbone

of two NRTIs [37, 349]. The unregulated and unmonitored use of ART obtained in the

black market or abroad, the lack of alternative ART regimens as well as the displacement

of HIV infected people to countries where ART is accessible for a long time are the most

likely explanations for the increase of the HIV drug resistance seen in more recent years

in Angola. Similar to previous studies and in line with more recent publications [34, 37,

349], the HIV-1 epidemic in Luanda in 2009 was highly complex being characterized by

the presence of almost 50% of complex recombinant virus and all subtypes with the

exception of subtypes B and K [4, 5, 35, 96, 289]. We saw that some strains from Angola

had little organized substructure and formed weaker clusters within phylogenetic trees

than the global reference sequences, not allowing a clear distinction between subtypes.

As a consequence, the current global subtype classification may not reflect the extent of

143
 Chapter 5- General discussion and conclusion

diversity in this region [290] which might reflect the 5.8% of untypable sequences

observed in our study. Moving forward, full genome sequences should be sequenced more

frequently in order to get more detail about the the possible recombination events, as

recently Bartolo et al identified a new H/U/CRF02_AG recombinant genome isolated

from an HIV-1 infected Angolan patient [252]. The prevailing subtype in 2009 in Luanda

was subtype C (36.9%) followed by subsubtype F1 (20.0%) whereas in 2001 it was

subtype A followed by subtype C [7]. In 2018, in 34 HIV-1 infecteted pregnant women

from Luanda, also analyzing the pol gene, Sebastião et al. identified the same prevailing

subtypes as we did in 2009, 38% of subtype C the most frequent, followed by the subtype

F1 18% [37]. The significant decrease in the prevalence of subtype A and increase in

subtype C observed in 2009 and in more recent studies [34, 37] could be explained by the

increasing predominance of subtype C in the bordering countries, namely in the south

region of Democratic Republic of Congo [202] and in Zambia [35, 200, 207].

Importantly, the results indicate that the Angolan HIV-1 epidemic is still increasing in

genetic complexity and suggest high rates of co-infection and/or superinfection [208]

which is consistent with an increasing HIV-1 incidence and prevalence [29, 255, 291].

In chapter 2 we also attempted to perform the first assessement of HIV-1 transmission

dynamics in Angola. We sucessfully identified a large number of transmission clusters in

this study which included 35.7% of the analyzed samples. Overall, based on high

sequence homology between patients in transmission clusters, a large number of potential

recent infections were inferred which is consistent with a rising HIV-1 epidemic in

Luanda driven by heterosexual transmission [29, 255, 291].

In chapter 3 we performed a much needed validation of a molecular test for the EID of

HIV-1 infection with pratical and direct benefits for the children of the Angolan Perinatal

HIV Cohort (APEHC). The high seroprevalence of HIV-1 infection in pregnant women
144
 Chapter 5- General discussion and conclusion

in Angola make children among the most at risk for HIV infection [367]. In spite of this

the AIDS response in sub-Saharan Africa has largely left them behind [30]. In this regard,

Angola has only registered a 25% reduction of new infections among infants since 2009

[30] which led to the implementation of several programs for prevention of mother to

child transmission (PMTCT), due to the implementation of such programs 4,100 new

HIV infections were averted in 2020 [3]. However, final vertical transmission rate

including breastfeeding in 2020 in Angola was 18.6% [3]. Significant progress has been

made in many sub-Saharan countries in implementing EID services following the

introduction of HIV DNA testing on DBS [198]. Several recent studies shown the

feasibility and advantages of using point of care (POC) platforms for the EID of HIV-1

infection in different African countries [226, 227, 368]. However, at the HDP where our

cohort was based and in most other hospitals in Angola, pediatric diagnosis of HIV-1

infection was still done by serology at month 12 of life which significantly delays the

initiation of treatment [305]. To support EID service expansion in Luanda we developed

and validated a new HIV-1 DNA assay to be used on DBS samples.

To lower the costs, we used the Chelex method of DNA extraction which is also quick

and easy to perform. This method had been previously applied to diagnostic screening in

HIV-1-exposed infants in Rwanda showing reliable results when used in combination

with either an in-house nested PCR or the Roche Amplicor HIV-1 DNA assay version

1.5 [306]. Also, the same extraction method was used before successfully to detect CMV

co-infection in infants infected with HIV [369]. The Chelex method that we have used

costs less than 3 USD per reaction, is very quick to perform, and does not use hazardous

solvents. Our nested PCR assay showed a very low LoD for all the complex HIV-1

genotypes that we used as controls, suggesting that it was appropriate for early diagnosis

of HIV-1 infection in infants in Angola. Indeed, using this assay we could detect all HIV-

145
 Chapter 5- General discussion and conclusion

1 infected infants at month 1 of life which makes this new assay suitable to health care

centers following option B+ of WHO guidelines that recommend EID at 4-6 weeks of

life. The low percentage of HIV-1 infected infants (2.2%) in the APEHC cohort between

2012 and 2014 confirms the effectiveness of the WHO-based prevention program

implemented since 2007 at the HDP [305] and the high standard of care provided at HDP

to the pregnant women infected with HIV. More recently, using a similar methodology,

filter papers and nested PCR for the diagnosis of perinatal HIV-1[370] and other

infectious disease [371] in rural and low income areas, has been sucessfuly used by others

[370, 371]. Which shows the feasibility and adaptability of our method to be used not

only to diagnose HIV infection but to detected other pathogens in diferent geographic

areas [371]. In fact using an optimized version of the protocol that we developed and

validated for the perinatal HIV-1 diagnosis, we as well as others [372] managed to detect

and diagnose SARS-CoV-2 infection, which was especially important since the COVID-

19 pandemic caused supply shortages of diagnostic tests. In comparison with current

conventional methods for detecting SARS-CoV-2 (COVID-19) [373] the Chelex method

presents important advantages in terms of safety, costs, and sensitivity [372].

In chaper 4 we performed the first detailed characterization of the antibody response and

its viral and host determinants on a large cohort of HIV-1-infected Angolan patients.

Overall, 29% (elite and broad neutralization categories) of the patients developed broad

and potent neutralizing responses, roughly corresponding to more than 50% breadth on a

diverse and representative 12 ENV-pseudotyped virus panel [162]. This result is in line

with other studies suggesting that some degree of neutralization breadth develops in the

majority of HIV infected individuals. However it is higher than that observed in previous

cohort studies in sub-Saharan Africa [167, 172, 173, 176] and in Europe [146, 163]. This

divergence is likely related with differences between HIV populations such as the HLA

146
 Chapter 5- General discussion and conclusion

genotype and ethnicity of the patients, diversity of infecting viruses, viral load, CD4

counts, and duration of infection [146, 166, 172, 173, 354]. Time of infection and viral

diversity has been shown to influence the developing of neutralization breadth [146, 172,

173, 374]. In this sense, the age of the HIV-1 epidemic, the unusual set of HIV-1 clades

and the overall high viral diversity found in Angola may have contributed to the high rate

of broad neutralizers found in this study [2, 5, 6, 34, 37, 99, 252]. Indeed, in this thesis

(Chapter 2 and 4) we observed an increasing viral variability that might impact

neutralization capacity, both processes are mutually dependent, Env escape variants being

selected in response to neutralizing responses and a greater diversity of escape variants

selecting a greater variety of Abs creating a repetitive cycle resulting in the incre ased

probability of eliciting bNAbs [166, 170].

Also, we saw a significant boost in the neutralizing response from 2009 to 2014.

Although, we can argue that in 2009 some patients could still be in the process of

mounting an effective neutralizing response as more than two thirds of the samples

collected in 2014 were follow-ups of 2009 [146, 172, 173, 374]. Nevertheless, this result

is no less remarkable since in 2014 the vast majority of the patients were under ART and

it was shown before that restoration of the B cell function with ART is uncommon in

chronic HIV-1 infection [356, 357]. However, we can hypothesize that similar to what

happens in HIV-2, persistent stimulation of germinal centers in the lymph nodes in a

seting of reduced viremia is linked to memory B-cell exhaustion resulting in increased

HIV antibody affinity maturation [317]. In fact, moving forward we are going to further

characterize and test the neutralizing activity of the Angolan HIV-1 infected patients

against HIV-2 clinical isolates, preliminary results shows that cross-type neutralization is

much more common than expected [375, 376], which raises the hypothesis that these

147
 Chapter 5- General discussion and conclusion

ancestral founder viruses that circulate in Angola have envelope features structurally

more similar to HIV-2 that has important exposed epitopes, such as the V3 region [326].

Previous publications found that the development of neutralization breadth was strongly

associated with viral load (VL) [173, 377]. Unfortunately VL determination was not

standard of care at the time of our study in Angola, so CD4 count and clinic were the

main markers used to measure disease progression. The finding that late presenters (<200

CD4+ T cell counts) had higher neutralization scores was in line with other studies and

likely reflects higher viral replication in these patients. An intrinsic difference in CD4

levels between patients with or without neutralizing capacity cannot be totally excluded

as CD4 levels prior to infection were not available. Even so, if there is indeed a direct

impact from CD4+ T cell loss on bnAb development, this could potentially be due to the

increase of survival time of auto-reactive B cells or by restraining the polyactivation of B

cells in that way decreasing unspecific activation of precursor B cell populations [172].

In our study infection by subtype C viruses was associated with enhanced neutralization

breadth and potency. Specific phenotypic and genotypic features of subtype C

transmitted/founder viruses have been described that possibly support the development

of neutralization breadth, such as less variable V3 region, shorter V1-V2 region and

unique mutational patterns in the α2-helix in the C3 region [146, 168, 171, 176, 212, 348,

362, 364]. Highlighting the importance of the infecting virus in guiding the selection of

bNAbs, a comprehensive analysis by Koyous et al., where they prospectively followed

selected transmission pairs identifying specific characteristics of the virus that make them

more prone to develop bNAbs, emphasizing the importance that select HIV-1 viral

variants have on the capability to initiate bnAb responses [353]. Furthermore, it has been

shown that viral subtype impacts the elicitation [212] and epitope specificity of bNAbs

[146, 171] which shows the importance of the subtype of the infecting virus in guiding

148
 Chapter 5- General discussion and conclusion

bnAb induction and maturation. Yet, in our study we must consider that subtype C viruses

from the indicator panel more closely related to the infecting virus of the Angolan patients

were more easily neutralized than subtypes with higher genetic differences such as

CRF01_AE, evidencing a within clade neutralization tendency.

Despite not having assessed the neutralizing activity against the autologous virus, we

looked at the variability of the Angolan HIV sequences in the C2V3C3 region to assess

the impact of antibodies targeting this region on viral diversity. The V3 and C3 were the

most variable regions across neutralizing categories which is in line with other studies

that show that bNAbs targeting the V3 glycan are the most abundant [173] and first

bNAbs [316] to be selected during the course of the natural infection, mainly because of

the exposed nature of the V3 loop and the fact that contrary to other bNAbs, V3 targeting

bNAbs do not require extensive somatic hypermutations [323]. Interestingly, patients

exhibiting elite neutralizing capacity against the virus panel had far less variability in the

C2V3C3 in comparison to patients with low neutralizing responses, indicating that the

neutralizing response constrained the diversification having selected quasispecies more

resistant to neutralization. Moreover, site-by-site analysis revealed that diversifying

selection follow the same tendency as viral diversity and neutralizing responses, since the

Angolan patients exhibiting low neutralizing responses were the ones showing more sites

under selective pressure. Taken together these findings point to an increasing diversifying

selection in Env evolving together with development of breadth to a certain extent and

then a convergence of the resistant viral quasispecies with the acquisition of breadth.

Of note, important V3 bnAb recognition sites and sites associated with resistance to

neutralization such as N295 were under positive selection only in patients with broad and

elite neutralizing capacities, and N-glycans at positions 339 and 355 in C3 were only

present simultaneously in elite neutralizers. These findings suggest that most Nab

149
 Chapter 5- General discussion and conclusion

epitopes in these patients are located in these V3 and C3 regions. In fact, despite N295

glycan not being a direct contact of bnAb PGT 135 it was shown by Kong et. al. [19] that

N295 is an important recognition site for some HIV-1 strains. Also Seabright et al. shown

that deletion of N295A resulted in a considerable decrease in 2G12 antibody binding

[378]. Furthermore, as hypothesized by Gray et al. that clade C resistance to 2G12 was

related to the lack of N295 glycan [379], in our study site N295 was under positive

selective pressure in the elite neutralizers (60% of which harbouring clade C virus) despite

lacking glycosylation at position 295 which may have prompted the high number of 2G12

and PGT 128 like bnAb responses seen in our study. Increasing the number of N-glycans

in the envelope gp120 surface glycoprotein, or varying the position of glycosylation sites,

has been associated with escape from IgG neutralizing antibody response in simian

immunodeficiency virus (SIV) and HIV-1 infection [155, 380-382]. A recent study that

followed HIV-1 subtype C infected individuals longitudinally from acute infection

reported associations between the development of bNAbs and the presence of specific

glycans in the C2V3C3 region on the gp120 [176], specifically on sites 301 and 332. In

our study both these N-glycosylation sites were very conserved, being present in almost

all sequences tested, irrespective of neutralization category and viral subtype, which also

might have played a role in the high number of PGT128 and 2G12 bnAb like responses

that we have seen. In contrast, N-glycosylation sites N339 and N355 were mainly present

in the patients exhibiting elite neutralization capacity. These sites and particularly N339

were recently identified as an important epitope for CP506 lineage mAbs [366].

We found a strong association between antibody binding titers targeting the C2V3C3

region and neutralization capacity irrespective of viral subtype and patients’

characteristics. These results confirm the important role of the C2V3C3 HIV-1 region as

a target for antibody neutralization [176, 212, 316, 366]

150
 Chapter 5- General discussion and conclusion

Consistent with this we found that most of the neutralizing antibodies in Angolan patients

had neutralization fingerprints similar to bNAbs targeting the V3 glycan, concordantly

Landais et al. in a large cohort of Eastern and South African HIV infected patients found

that the most prevalent target epitope was V3 glycan N332 supersite followed by the V2

apex [173]. The C2V3C3 region of HIV-1, as previously found, has an extended and

highly accessible V3 loop [383]. Such conformation is entirely consistent with its

immunodominant and neutralizing nature and with its crucial role in HIV-1 co-receptor

binding and tropism. Concordantly, Calado et al. tested recently a vaccine HIV candidate

in mice and showed that cross clade neutralization of tier 2 HIV-1 isolates was being

driven essentially by antibodies targeting the V3 crown which provides further support

for the V3 crown as a crucial vaccine target [212, 250]. However, we have to take into

consideration that the envelope conformations of the global Env-pseudotyped virus panel

might naturally favour the selection and identification of antibodies targeting V3, since it

was shown by Han et al. [383] that Envs of distinct standardised virus panels are in an

open conformation with the V3 loop exposed. Still, the high prevalence of Angolan

patients with cross-clade neutralizing activity is consistent with long-term antigenic

stimulation with highly divergent and diverse HIV-1 clades. Also, the antibody binding

titer against envelope C2V3C3 region was a good indicator of neutralization breadth and

potency and may also be a good indicator of vaccine efficacy.

Overall, the studies developed in these thesis show an increasing diversity of the HIV-1

epidemics in Angola which poses significant challenges to the diagnosis, treatment and

management of people living with HIV. Lastly, the results obtained by characterizing the

neutralizing antibody responses in this challenging setting confirms the importance of

further characterize epidemics such as the Angolan with ancient HIV strains.

151
References

References

153
References

154
 References

References

1. UN General Assembly Political declaration on HIV and AIDS: on the fast track

to accelerating the fight against HIV and to ending the AIDS epidemic by 2030.

2016. [Link]

declaration-HIV-AIDS_en.pdf.

2. Subnational mapping of HIV incidence and mortality among individuals aged 15-

49 years in sub-Saharan Africa, 2000-18: a modelling study. Lancet HIV, 2021.

8(6): p. e363-e375.

3. AIDSinfo, Global data on HIV epidemiology and response. Country factsheets:

Angola 2020

[Link]

4. Bartolo, I., et al., High genetic diversity of human immunodeficiency virus type 1

in Angola. AIDS Res Hum Retroviruses, 2005. 21(4): p. 306-10.

5. Bartolo, I., et al., Highly divergent subtypes and new recombinant forms prevail

in the HIV/AIDS epidemic in Angola: new insights into the origins of the AIDS

pandemic. Infect Genet Evol, 2009. 9(4): p. 672-82.

6. Pineda-Pena, A.C., et al., On the contribution of Angola to the initial spread of

HIV-1. Infect Genet Evol, 2016. 46: p. 219-222.

7. Bartolo, I., et al., Antiretroviral drug resistance surveillance among treatment-

naive human immunodeficiency virus type 1-infected individuals in Angola:

evidence for low level of transmitted drug resistance. Antimicrobial agents and

chemotherapy, 2009. 53(7): p. 3156-8.

8. UNAIDS, 2015 PROGRESS REPORT ON THE GLOBAL PLAN: towards the

elimination of new HIV infections among children and keeping their mothers

alive.

155
 References

[Link]

obalPlan.

9. Pugach, P., et al., A native-like SOSIP.664 trimer based on an HIV-1 subtype B

env gene. J Virol, 2015. 89(6): p. 3380-95.

10. Sanders, R.W., et al., HIV-1 VACCINES. HIV-1 neutralizing antibodies induced

by native-like envelope trimers. Science, 2015. 349(6244): p. aac4223.

11. Burton, D.R. and J.R. Mascola, Antibody responses to envelope glycoproteins in

HIV-1 infection. Nat Immunol, 2015. 16(6): p. 571-6.

12. Lee, J.H., et al., Antibodies to a conformational epitope on gp41 neutralize HIV-

1 by destabilizing the Env spike. Nat Commun, 2015. 6: p. 8167.

13. Stephenson, K.E., et al., Vaccines and Broadly Neutralizing Antibodies for HIV-

1 Prevention. Annu Rev Immunol, 2020. 38: p. 673-703.

14. Gray, G.E., et al., Vaccine Efficacy of ALVAC-HIV and Bivalent Subtype C gp120-

MF59 in Adults. N Engl J Med, 2021. 384(12): p. 1089-1100.

15. Amornkul, P.N., et al., Disease progression by infecting HIV-1 subtype in a

seroconverter cohort in sub-Saharan Africa. AIDS, 2013. 27(17): p. 2775-86.

16. van der Sluis, R.M., et al., Dendritic cell-induced activation of latent HIV-1

provirus in actively proliferating primary T lymphocytes. PLoS Pathog, 2013.

9(3): p. e1003259.

17. Sanders, R.W., et al., A next-generation cleaved, soluble HIV-1 Env trimer,

BG505 SOSIP.664 gp140, expresses multiple epitopes for broadly neutralizing

but not non-neutralizing antibodies. PLoS Pathog, 2013. 9(9): p. e1003618.

18. Kijak, G.H., et al., Molecular evolution of the HIV-1 Thai epidemic between the

time of RV144 immunogen selection to the execution of the vaccine efficacy trial.

J Virol, 2013. 87(13): p. 7265-81.

156
 References

19. Kong, L., et al., Supersite of immune vulnerability on the glycosylated face of HIV-

1 envelope glycoprotein gp120. Nat Struct Mol Biol, 2013. 20(7): p. 796-803.

20. Julien, J.P., et al., Design and structure of two HIV-1 clade C SOSIP.664 trimers

that increase the arsenal of native-like Env immunogens. Proc Natl Acad Sci U S

A, 2015. 112(38): p. 11947-52.

21. Hymes, K.B., et al., Kaposi's sarcoma in homosexual men-a report of eight cases.

Lancet, 1981. 2(8247): p. 598-600.

22. Barre-Sinoussi, F., et al., Isolation of a T-lymphotropic retrovirus from a patient

at risk for acquired immune deficiency syndrome (AIDS). Science, 1983.

220(4599): p. 868-71.

23. Chermann, J.C., et al., Isolation of a new retrovirus in a patient at risk for

acquired immunodeficiency syndrome. Antibiot Chemother (1971), 1983. 32: p.

48-53.

24. Clavel, F., et al., Isolation of a new human retrovirus from West African patients

with AIDS. Science, 1986. 233(4761): p. 343-6.

25. Clavel, F., et al., Human immunodeficiency virus type 2 infection associated with

AIDS in West Africa. N Engl J Med, 1987. 316(19): p. 1180-5.

26. Rhodes, S.D. and F.S. Sy, Effectively Confronting the COVID-19 Pandemic:

Critical Lessons From HIV Prevention, Care, and Treatment and Innovative

Strategies to Conduct Community-Based and Community-Engaged Research

Safely. AIDS Educ Prev, 2020. 32(6): p. 455-471.

27. Hans, L., et al., Preparing for the next pandemic: Lessons from rapid scale-up of

SARS-CoV-2 testing in a South African high-throughput automated HIV

molecular laboratory. Int J Infect Dis, 2021. 110: p. 1-3.

28. UNAIDS. Global HIV & AIDS statistics - 2020 fact sheet

157
 References

[Link] 2020.

29. UNAIDS, UNAIDS report on the global AIDS epidemic 2013. Geneve. Avalaible

[Link]

2013/gr2013/UNAIDS_Global_Report_2013_en.pdf. Accessed 2014 Jan 30.

2013.

30. UNAIDS, AIDS by the numbers 2016 2016, Joint United Nations Programme on

HIV/AIDS (UNAIDS).

31. Reeves, J.D. and A.J. Piefer, Emerging drug targets for antiretroviral therapy.

Drugs, 2005. 65(13): p. 1747-66.

32. Gilks, C.F., et al., The WHO public-health approach to antiretroviral treatment

against HIV in resource-limited settings. Lancet, 2006. 368(9534): p. 505-10.

33. WHO (2010) Antiretroviral therapy for HIV infection in adults and adolescents.

Recommendations for a public health approach. 2010 revision Avalaible:

[Link] Accessed

15 January 2014.

34. Sebastiao, C.S., J. Morais, and M. Brito, Clinical and Public Health Implications

of HIV- Genetic Diversity and Drug Resistance Mutations in Angola: A

Systematic Review. AIDS Rev, 2020. 23(1): p. 48-56.

35. Afonso, J.M., et al., HIV-1 genetic diversity and transmitted drug resistance

mutations among patients from the North, Central and South regions of Angola.

PLoS One, 2012. 7(8): p. e42996.

36. Afonso, J.M., M.G. Morgado, and G. Bello, Evidence of multiple introductions of

HIV-1 subtype C in Angola. Infect Genet Evol, 2012. 12(7): p. 1458-65.

158
 References

37. Sebastiao, C.S., et al., Genetic diversity and drug resistance of HIV-1 among

infected pregnant women newly diagnosed in Luanda, Angola. PLoS One, 2019.

14(11): p. e0225251.

38. Coffin, J.M., The virology of AIDS: 1990. AIDS, 1990. 4 Suppl 1: p. S1-8.

39. Bedwell, G.J. and A.N. Engelman, Factors that mold the nuclear landscape of

HIV-1 integration. Nucleic Acids Res, 2021. 49(2): p. 621-635.

40. Roy, S., et al., A bulge structure in HIV-1 TAR RNA is required for Tat binding

and Tat-mediated trans-activation. Genes Dev, 1990. 4(8): p. 1365-73.

41. Sherpa, C., et al., Evolution of the HIV-1 Rev Response Element during Natural

Infection Reveals Nucleotide Changes That Correlate with Altered Structure and

Increased Activity over Time. J Virol, 2019. 93(11).

42. Cullen, B.R. and M.H. Malim, The HIV-1 Rev protein: prototype of a novel class

of eukaryotic post-transcriptional regulators. Trends Biochem Sci, 1991. 16(9):

p. 346-50.

43. Dettenhofer, M., et al., Association of human immunodeficiency virus type 1 Vif

with RNA and its role in reverse transcription. J Virol, 2000. 74(19): p. 8938-45.

44. Sheehy, A.M., N.C. Gaddis, and M.H. Malim, The antiretroviral enzyme

APOBEC3G is degraded by the proteasome in response to HIV-1 Vif. Nat Med,

2003. 9(11): p. 1404-7.

45. Stavrou, S. and S.R. Ross, APOBEC3 Proteins in Viral Immunity. J Immunol,

2015. 195(10): p. 4565-70.

46. Le Rouzic, E. and S. Benichou, The Vpr protein from HIV-1: distinct roles along

the viral life cycle. Retrovirology, 2005. 2: p. 11.

47. Laguette, N., et al., SAMHD1 is the dendritic- and myeloid-cell-specific HIV-1

restriction factor counteracted by Vpx. Nature, 2011. 474(7353): p. 654-7.

159
 References

48. Willey, R.L., et al., Human immunodeficiency virus type 1 Vpu protein induces

rapid degradation of CD4. J Virol, 1992. 66(12): p. 7193-200.

49. Klimkait, T., et al., The human immunodeficiency virus type 1-specific protein vpu

is required for efficient virus maturation and release. J Virol, 1990. 64(2): p. 621-

9.

50. Roeth, J.F. and K.L. Collins, Human immunodeficiency virus type 1 Nef: adapting

to intracellular trafficking pathways. Microbiol Mol Biol Rev, 2006. 70(2): p.

548-63.

51. Gomez, C. and T.J. Hope, The ins and outs of HIV replication. Cell Microbiol,

2005. 7(5): p. 621-6.

52. Hladik, F. and M.J. McElrath, Setting the stage: host invasion by HIV. Nat Rev

Immunol, 2008. 8(6): p. 447-57.

53. Agosto, L.M., et al., Highly active antiretroviral therapies are effective against

HIV-1 cell-to-cell transmission. PLoS Pathog, 2014. 10(2): p. e1003982.

54. Arnold, E. and S.G. Sarafianos, Molecular biology: an HIV secret uncovered.

Nature, 2008. 453(7192): p. 169-70.

55. Zila, V., et al., Cone-shaped HIV-1 capsids are transported through intact nuclear

pores. Cell, 2021. 184(4): p. 1032-1046 e18.

56. Scoca, V. and F. Di Nunzio, The HIV-1 Capsid: From Structural Component to

Key Factor for Host Nuclear Invasion. Viruses, 2021. 13(2).

57. AlBurtamani, N., A. Paul, and A. Fassati, The Role of Capsid in the Early Steps

of HIV-1 Infection: New Insights into the Core of the Matter. Viruses, 2021. 13(6).

58. Engelman, A. and P. Cherepanov, The structural biology of HIV-1: mechanistic

and therapeutic insights. Nat Rev Microbiol, 2012. 10(4): p. 279-90.

160
 References

59. Wu, Y., HIV-1 gene expression: lessons from provirus and non-integrated DNA.

Retrovirology, 2004. 1: p. 13.

60. Murphy, R.E. and J.S. Saad, The Interplay between HIV-1 Gag Binding to the

Plasma Membrane and Env Incorporation. Viruses, 2020. 12(5).

61. Moranguinho, I. and S.T. Valente, Block-And-Lock: New Horizons for a Cure for

HIV-1. Viruses, 2020. 12(12).

62. Taveira, N. and F. Martin, O agente. Manual sobre SIDA – 5ª Edição. Editor

Francisco Antunes, Fernando Maltez. 2018. Permanyer Portugal., 2018.

63. Jackson, P.E.H., et al., Sequence and Functional Variation in the HIV-1 Rev

Regulatory Axis. Curr HIV Res, 2020. 18(2): p. 85-98.

64. Klingler, J., et al., How HIV-1 Gag Manipulates Its Host Cell Proteins: A Focus

on Interactors of the Nucleocapsid Domain. Viruses, 2020. 12(8).

65. Bird, S.W., et al., The ins and outs of viral infection: keystone meeting review.

Viruses, 2014. 6(9): p. 3652-62.

66. Leonard, C.K., et al., Assignment of intrachain disulfide bonds and

characterization of potential glycosylation sites of the type 1 recombinant human

immunodeficiency virus envelope glycoprotein (gp120) expressed in Chinese

hamster ovary cells. J Biol Chem, 1990. 265(18): p. 10373-82.

67. Bernstein, H.B., et al., Human immunodeficiency virus type 1 envelope

glycoprotein is modified by O-linked oligosaccharides. J Virol, 1994. 68(1): p.

463-8.

68. Helseth, E., et al., Human immunodeficiency virus type 1 gp120 envelope

glycoprotein regions important for association with the gp41 transmembrane

glycoprotein. J Virol, 1991. 65(4): p. 2119-23.

161
 References

69. Wyatt, R., et al., The antigenic structure of the HIV gp120 envelope glycoprotein.

Nature, 1998. 393(6686): p. 705-11.

70. Wyatt, R., et al., Involvement of the V1/V2 variable loop structure in the exposure

of human immunodeficiency virus type 1 gp120 epitopes induced by receptor

binding. J Virol, 1995. 69(9): p. 5723-33.

71. Kwong, P.D., et al., Structure of an HIV gp120 envelope glycoprotein in complex

with the CD4 receptor and a neutralizing human antibod y. Nature, 1998.

393(6686): p. 648-59.

72. Pancera, M., et al., Structure of HIV-1 gp120 with gp41-interactive region reveals

layered envelope architecture and basis of conformational mobility. Proc Natl

Acad Sci U S A, 2010. 107(3): p. 1166-71.

73. Huang, C.C., et al., Structure of a V3-containing HIV-1 gp120 core. Science,

2005. 310(5750): p. 1025-8.

74. Chen, B., et al., Structure of an unliganded simian immunodeficiency virus gp120

core. Nature, 2005. 433(7028): p. 834-41.

75. Dalgleish, A.G., et al., The CD4 (T4) antigen is an essential component of the

receptor for the AIDS retrovirus. Nature, 1984. 312(5996): p. 763-7.

76. Maddon, P.J., et al., The T4 gene encodes the AIDS virus receptor and is expressed

in the immune system and the brain. Cell, 1986. 47(3): p. 333-48.

77. Deng, H., et al., Identification of a major co-receptor for primary isolates of HIV-

1. Nature, 1996. 381(6584): p. 661-6.

78. Dragic, T., et al., HIV-1 entry into CD4+ cells is mediated by the chemokine

receptor CC-CKR-5. Nature, 1996. 381(6584): p. 667-73.

162
 References

79. Alkhatib, G., et al., CC CKR5: a RANTES, MIP-1alpha, MIP-1beta receptor as a

fusion cofactor for macrophage-tropic HIV-1. Science, 1996. 272(5270): p. 1955-

8.

80. Moriuchi, M., et al., Cloning and analysis of the promoter region of CXCR4, a

coreceptor for HIV-1 entry. J Immunol, 1997. 159(9): p. 4322-9.

81. Trkola, A., et al., CD4-dependent, antibody-sensitive interactions between HIV-1

and its co-receptor CCR-5. Nature, 1996. 384(6605): p. 184-7.

82. Allen, A.G., et al., Gene Editing of HIV-1 Co-receptors to Prevent and/or Cure

Virus Infection. Front Microbiol, 2018. 9: p. 2940.

83. Gombos, R.B., et al., Inhibitory Effect of Individual or Combinations of Broadly

Neutralizing Antibodies and Antiviral Reagents against Cell-Free and Cell-to-

Cell HIV-1 Transmission. J Virol, 2015. 89(15): p. 7813-28.

84. Dufloo, J., T. Bruel, and O. Schwartz, HIV-1 cell-to-cell transmission and broadly

neutralizing antibodies. Retrovirology, 2018. 15(1): p. 51.

85. Hwang, S.S., et al., Identification of the envelope V3 loop as the primary

determinant of cell tropism in HIV-1. Science, 1991. 253(5015): p. 71-4.

86. Lengauer, T., et al., Bioinformatics prediction of HIV coreceptor usage. Nat

Biotechnol, 2007. 25(12): p. 1407-10.

87. De Cock, K.M., F. Brun-Vezinet, and B. Soro, HIV-1 and HIV-2 infections and

AIDS in West Africa. AIDS, 1991. 5 Suppl 1: p. S21-8.

88. Peeters, M., et al., Isolation and partial characterization of an HIV-related virus

occurring naturally in chimpanzees in Gabon. AIDS, 1989. 3(10): p. 625-30.

89. Chen, Z., et al., Human immunodeficiency virus type 2 (HIV-2) seroprevalence

and characterization of a distinct HIV-2 genetic subtype from the natural range

163
 References

of simian immunodeficiency virus-infected sooty mangabeys. J Virol, 1997. 71(5):

p. 3953-60.

90. Yamaguchi, J., et al., Brief Report: Complete Genome Sequence of CG-0018a-01

Establishes HIV-1 Subtype L. J Acquir Immune Defic Syndr, 2020. 83(3): p. 319-

322.

91. Hemelaar, J., et al., Country Level Diversity of the HIV-1 Pandemic between 1990

and 2015. J Virol, 2020. 95(2).

92. Hemelaar, J., et al., Global and regional molecular epidemiology of HIV-1, 1990-

2015: a systematic review, global survey, and trend analysis. Lancet Infect Dis,

2019. 19(2): p. 143-155.

93. Hemelaar, J., et al., Global and regional epidemiology of HIV-1 recombinants in

1990-2015: a systematic review and global survey. Lancet HIV, 2020. 7(11): p.

e772-e781.

94. Robertson, D.L., et al., HIV-1 nomenclature proposal. Science, 2000. 288(5463):

p. 55-6.

95. Hemelaar, J., The origin and diversity of the HIV-1 pandemic. Trends Mol Med,

2012. 18(3): p. 182-92.

96. Abecasis, A., et al., HIV-1 genetic variants circulation in the North of Angola.

Infect Genet Evol, 2005. 5(3): p. 231-7.

97. Abecasis, A.B., et al., HIV-1 subtype distribution and its demographic

determinants in newly diagnosed patients in Europe suggest highly

compartmentalized epidemics. Retrovirology, 2013. 10: p. 7.

98. Bartolo, I., et al., HIV-1 genetic diversity and transmitted drug resistance in health

care settings in Maputo, Mozambique. J Acquir Immune Defic Syndr, 2009.

51(3): p. 323-31.

164
 References

99. Bartolo, I., et al., HIV-1 diversity, transmission dynamics and primary drug

resistance in Angola. PLoS One, 2014. 9(12): p. e113626.

100. Bartolo I, T.N., HIV-1 Diversity and Its Implications in Diagnosis, Transmission,

Disease Progression, and Antiretroviral Therapy. In: Caliskan M, editor. Genetic

Diversity in Microorganisms

Rijeka, Croatia: InTech - Open Access Publisher. pp. 171–214., 2012.

101. Gao, F., et al., The heterosexual human immunodeficiency virus type 1 epidemic

in Thailand is caused by an intersubtype (A/E) recombinant of African origin. J

Virol, 1996. 70(10): p. 7013-29.

102. Soto-Ramirez, L.E., et al., HIV-1 Langerhans' cell tropism associated with

heterosexual transmission of HIV. Science, 1996. 271(5253): p. 1291-3.

103. Hudgens, M.G., et al., Subtype-specific transmission probabilities for human

immunodeficiency virus type 1 among injecting drug users in Bangkok, Thailand.

Am J Epidemiol, 2002. 155(2): p. 159-68.

104. Kiwanuka, N., et al., HIV-1 subtypes and differences in heterosexual HIV

transmission among HIV-discordant couples in Rakai, Uganda. AIDS, 2009.

23(18): p. 2479-84.

105. Berger, E.A., et al., A new classification for HIV-1. Nature, 1998. 391(6664): p.

240.

106. Huang, W., et al., Coreceptor tropism in human immunodeficiency virus type 1

subtype D: high prevalence of CXCR4 tropism and heterogeneous composition of

viral populations. J Virol, 2007. 81(15): p. 7885-93.

107. Huang, Y., et al., The role of a mutant CCR5 allele in HIV-1 transmission and

disease progression. Nature medicine, 1996. 2(11): p. 1240-3.

165
 References

108. Cilliers, T., et al., The CCR5 and CXCR4 coreceptors are both used by human

immunodeficiency virus type 1 primary isolates from subtype C. J Virol, 2003.

77(7): p. 4449-56.

109. Renjifo, B., et al., Differences in perinatal transmission among human

immunodeficiency virus type 1 genotypes. J Hum Virol, 2001. 4(1): p. 16-25.

110. Renjifo, B., et al., In-utero transmission of quasispecies among human

immunodeficiency virus type 1 genotypes. Virology, 2003. 307(2): p. 278-82.

111. Duran Ramirez, J.J., et al., Increasing Frequency and Transmission of HIV-1 Non-

B Subtypes among Men Who Have Sex with Men in the Swiss HIV Cohort Study.

J Infect Dis, 2021.

112. Kuritzkes, D.R., HIV-1 subtype as a determinant of disease progression. The

Journal of infectious diseases, 2008. 197(5): p. 638-9.

113. Kaleebu, P., et al., Effect of human immunodeficiency virus (HIV) type 1 envelope

subtypes A and D on disease progression in a large cohort of HIV-1-positive

persons in Uganda. The Journal of infectious diseases, 2002. 185(9): p. 1244-50.

114. Kiwanuka, N., et al., Effect of human immunodeficiency virus Type 1 (HIV-1)

subtype on disease progression in persons from Rakai, Uganda, with incident

HIV-1 infection. The Journal of infectious diseases, 2008. 197(5): p. 707-13.

115. Mlisana, K., et al., Challenges of diagnosing acute HIV-1 subtype C infection in

African women: performance of a clinical algorithm and the need for point-of-

care nucleic-acid based testing. PLoS One, 2013. 8(4): p. e62928.

116. Touloumi, G., et al., Impact of HIV-1 subtype on CD4 count at HIV

seroconversion, rate of decline, and viral load set point in European

seroconverter cohorts. Clin Infect Dis, 2013. 56(6): p. 888-97.

166
 References

117. Kantor, R., et al., Pretreatment HIV Drug Resistance and HIV-1 Subtype C Are

Independently Associated With Virologic Failure: Results From the Multinational

PEARLS (ACTG A5175) Clinical Trial. Clin Infect Dis, 2015. 60(10): p. 1541-9.

118. Keller, M., et al., Impact of HIV-1 viral subtype on CD4+ T-cell decline and

clinical outcomes in antiretroviral naive patients receiving universal healthcare.

AIDS, 2009. 23(6): p. 731-7.

119. Kanki, P.J., et al., Human immunodeficiency virus type 1 subtypes differ in disease

progression. J Infect Dis, 1999. 179(1): p. 68-73.

120. Baeten, J.M., et al., HIV-1 subtype D infection is associated with faster disease

progression than subtype A in spite of similar plasma HIV-1 loads. J Infect Dis,

2007. 195(8): p. 1177-80.

121. Farinre, O., et al., Subtype-specific differences in Gag-protease replication

capacity of HIV-1 isolates from East and West Africa. Retrovirology, 2021. 18(1):

p. 11.

122. Cashin, K., et al., Linkages between HIV-1 specificity for CCR5 or CXCR4 and in

vitro usage of alternative coreceptors during progressive HIV-1 subtype C

infection. Retrovirology, 2013. 10: p. 98.

123. Attia, S., et al., Sexual transmission of HIV according to viral load and

antiretroviral therapy: systematic review and meta-analysis. AIDS, 2009. 23(11):

p. 1397-404.

124. Stone, M., et al., Comparison of Detection Limits of Fourth- and Fifth-Generation

Combination HIV Antigen-Antibody, p24 Antigen, and Viral Load Assays on

Diverse HIV Isolates. J Clin Microbiol, 2018. 56(8).

167
 References

125. Fiebig, E.W., et al., Dynamics of HIV viremia and antibody seroconversion in

plasma donors: implications for diagnosis and staging of primary HIV infection.

AIDS, 2003. 17(13): p. 1871-9.

126. Aghokeng, A.F., et al., Inaccurate diagnosis of HIV-1 group M and O is a key

challenge for ongoing universal access to antiretroviral treatment and HIV

prevention in Cameroon. PLoS One, 2009. 4(11): p. e7702.

127. Hage-Sleiman, M., et al., False-negative Results of Human Immunodeficiency

Virus (HIV) Rapid Testing in HIV Controllers. Clin Infect Dis, 2020. 70(8): p.

1754-1757.

128. Fogel, J.M., et al., Brief Report: Impact of Early Antiretroviral Therapy on the

Performance of HIV Rapid Tests and HIV Incidence Assays. J Acquir Immune

Defic Syndr, 2017. 75(4): p. 426-430.

129. Delaugerre, C., et al., Clinical and resistance consequences of misquantification

of plasma and cerebrospinal fluid human immunodeficiency virus type 1 (HIV-1)

RNA in samples from an HIV-1 subtype G-infected patient. J Clin Microbiol,

2009. 47(11): p. 3763-4.

130. Korn, K., et al., Single-point mutations causing more than 100-fold

underestimation of human immunodeficiency virus type 1 (HIV-1) load with the

Cobas TaqMan HIV-1 real-time PCR assay. J Clin Microbiol, 2009. 47(4): p.

1238-40.

131. Rouet, F., et al., Comparison of the Generic HIV Viral Load assay with the

Amplicor HIV-1 monitor v1.5 and Nuclisens HIV-1 EasyQ v1.2 techniques for

plasma HIV-1 RNA quantitation of non-B subtypes: the Kesho Bora preparatory

study. J Virol Methods, 2010. 163(2): p. 253-7.

168
 References

132. WHO, Consolidated guidelines on HIV prevention, diagnosis, treatment and care

for key populations. 2016 update. . 2016, World Health Organization: Geneva.

133. Penazzato, M., et al., Early infant diagnosis of HIV infection in low-income and

middle-income countries: does one size fit all? Lancet Infect Dis, 2014. 14(7): p.

650-5.

134. WHO, March 2014 Supplement to the 2013 Consolidated Guidelines on the Use

of Antiretroviral Drugs for Treating and Preventing HIV Infection.

Recommendations for a Public Health Approach. . 2014, World Health

Organization Geneva.

135. UNITAID, HIV/AIDS diagnostics technology landscape - semi-annual update.

2015, World Health Organization: Geneva.

136. Kantor, R., et al., Impact of HIV-1 subtype and antiretroviral therapy on protease

and reverse transcriptase genotype: results of a global collaboration. PLoS Med,

2005. 2(4): p. e112.

137. Kantor, R., Impact of HIV-1 pol diversity on drug resistance and its clinical

implications. Curr Opin Infect Dis, 2006. 19(6): p. 594-606.

138. Saag, M.S., et al., Antiretroviral Drugs for Treatment and Prevention of HIV

Infection in Adults: 2020 Recommendations of the International Antiviral Society-

USA Panel. JAMA, 2020. 324(16): p. 1651-1669.

139. Martinez-Cajas, J.L., et al., Differences in resistance mutations among HIV-1 non-

subtype B infections: a systematic review of evidence (1996-2008). J Int AIDS

Soc, 2009. 12: p. 11.

140. Eshleman, S.H., et al., Comparison of mother-to-child transmission rates in

Ugandan women with subtype A versus D HIV-1 who received single-dose

169
 References

nevirapine prophylaxis: HIV Network For Prevention Trials 012. J Acquir

Immune Defic Syndr, 2005. 39(5): p. 593-7.

141. Tomaras, G.D., et al., Initial B-cell responses to transmitted human

immunodeficiency virus type 1: virion-binding immunoglobulin M (IgM) and IgG

antibodies followed by plasma anti-gp41 antibodies with ineffective control of

initial viremia. J Virol, 2008. 82(24): p. 12449-63.

142. Sarmay, G., et al., Mapping and comparison of the interaction sites on the Fc

region of IgG responsible for triggering antibody dependent cellular cytotoxicity

(ADCC) through different types of human Fc gamma receptor. Mol Immunol,

1992. 29(5): p. 633-9.

143. Burton, D.R., Advancing an HIV vaccine; advancing vaccinology. Nat Rev

Immunol, 2019. 19(2): p. 77-78.

144. Dieffenbach, C.W. and A.S. Fauci, The search for an HIV vaccine, the journey

continues. J Int AIDS Soc, 2020. 23(5): p. e25506.

145. Horwitz, J.A., et al., Non-neutralizing Antibodies Alter the Course of HIV-1

Infection In Vivo. Cell, 2017. 170(4): p. 637-648 e10.

146. Rusert, P., et al., Determinants of HIV-1 broadly neutralizing antibody induction.

Nat Med, 2016. 22(11): p. 1260-1267.

147. Russell, J.H. and T.J. Ley, Lymphocyte-mediated cytotoxicity. Annu Rev

Immunol, 2002. 20: p. 323-70.

148. Baum, L.L., et al., HIV-1 gp120-specific antibody-dependent cell-mediated

cytotoxicity correlates with rate of disease progression. J Immunol, 1996. 157(5):

p. 2168-73.

149. Overbaugh, J. and L. Morris, The Antibody Response against HIV-1. Cold Spring

Harb Perspect Med, 2012. 2(1): p. a007039.

170
 References

150. Huang, J., et al., Broad and potent neutralization of HIV-1 by a gp41-specific

human antibody. Nature, 2012. 491(7424): p. 406-12.

151. Haynes, B.F., et al., Immune-correlates analysis of an HIV-1 vaccine efficacy

trial. N Engl J Med, 2012. 366(14): p. 1275-86.

152. Sogaard, O.S., et al., The Depsipeptide Romidepsin Reverses HIV-1 Latency In

Vivo. PLoS Pathog, 2015. 11(9): p. e1005142.

153. Baldwin, K.M., et al., HLA class II diversity in HIV-1 uninfected individuals from

the placebo arm of the RV144 Thai vaccine efficacy trial. Tissue Antigens, 2015.

85(2): p. 117-26.

154. Kim, J.H., J.L. Excler, and N.L. Michael, Lessons from the RV144 Thai phase III

HIV-1 vaccine trial and the search for correlates of protection. Annu Rev Med,

2015. 66: p. 423-37.

155. Wei, X., et al., Antibody neutralization and escape by HIV-1. Nature, 2003.

422(6929): p. 307-12.

156. Gray, E.S., et al., Neutralizing antibody responses in acute human

immunodeficiency virus type 1 subtype C infection. J Virol, 2007. 81(12): p. 6187-

96.

157. Moore, J.P., et al., Inter- and intraclade neutralization of human

immunodeficiency virus type 1: genetic clades do not correspond to neutralization

serotypes but partially correspond to gp120 antigenic serotypes. J Virol, 1996.

70(1): p. 427-44.

158. Arendrup, M., et al., Autologous HIV-1 neutralizing antibodies: emergence of

neutralization-resistant escape virus and subsequent development of escape virus

neutralizing antibodies. J Acquir Immune Defic Syndr, 1992. 5(3): p. 303-7.

171
 References

159. Walker, L.M., et al., Broad and potent neutralizing antibodies from an African

donor reveal a new HIV-1 vaccine target. Science, 2009. 326(5950): p. 285-9.

160. Scheid, J.F., et al., Broad diversity of neutralizing antibodies isolated from

memory B cells in HIV-infected individuals. Nature, 2009. 458(7238): p. 636-40.

161. Simek, M.D., et al., Human immunodeficiency virus type 1 elite neutralizers:

individuals with broad and potent neutralizing activity identified by using a high-

throughput neutralization assay together with an analytical selection algorithm.

J Virol, 2009. 83(14): p. 7337-48.

162. deCamp, A., et al., Global panel of HIV-1 Env reference strains for standardized

assessments of vaccine-elicited neutralizing antibodies. J Virol, 2014. 88(5): p.

2489-507.

163. Hraber, P., et al., Impact of clade, geography, and age of the epidemic on HIV-1

neutralization by antibodies. J Virol, 2014. 88(21): p. 12623-43.

164. Deeks, S.G., et al., Neutralizing antibody responses against autologous and

heterologous viruses in acute versus chronic human immunodeficiency virus

(HIV) infection: evidence for a constraint on the ability of HIV to completely

evade neutralizing antibody responses. J Virol, 2006. 80(12): p. 6155-64.

165. Moore, P.L., et al., Limited neutralizing antibody specificities drive neutralization

escape in early HIV-1 subtype C infection. PLoS Pathog, 2009. 5(9): p. e1000598.

166. Moore, P.L., The Neutralizing Antibody Response to the HIV-1 Env Protein. Curr

HIV Res, 2018. 16(1): p. 21-28.

167. Mishra, N., et al., Broadly neutralizing plasma antibodies effective against

autologous circulating viruses in infants with multivariant HIV-1 infection. Nat

Commun, 2020. 11(1): p. 4409.

172
 References

168. Moore, P.L., et al., The c3-v4 region is a major target of autologous neutralizing

antibodies in human immunodeficiency virus type 1 subtype C infection. J Virol,

2008. 82(4): p. 1860-9.

169. Sato, S., et al., Potent antibody-mediated neutralization and evolution of antigenic

escape variants of simian immunodeficiency virus strain SIVmac239 in vivo. J

Virol, 2008. 82(19): p. 9739-52.

170. Landais, E. and P.L. Moore, Development of broadly neutralizing antibodies in

HIV-1 infected elite neutralizers. Retrovirology, 2018. 15(1): p. 61.

171. Stefic, K., et al., Impact of HIV-1 Diversity on Its Sensitivity to Neutralization.

Vaccines (Basel), 2019. 7(3).

172. Gray, E.S., et al., The neutralization breadth of HIV-1 develops incrementally

over four years and is associated with CD4+ T cell decline and high viral load

during acute infection. J Virol, 2011. 85(10): p. 4828-40.

173. Landais, E., et al., Broadly Neutralizing Antibody Responses in a Large

Longitudinal Sub-Saharan HIV Primary Infection Cohort. PLoS Pathog, 2016.

12(1): p. e1005369.

174. Mascola, J.R. and D.C. Montefiori, The role of antibodies in HIV vaccines. Annu

Rev Immunol, 2010. 28: p. 413-44.

175. Stamatatos, L., et al., Neutralizing antibodies generated during natural HIV-1

infection: good news for an HIV-1 vaccine? Nat Med, 2009. 15(8): p. 866-70.

176. Ndlovu, B., et al., Envelope characteristics in individuals who developed

neutralizing antibodies targeting different epitopes in HIV-1 subtype C infection.

Virology, 2020. 546: p. 1-12.

177. Caskey, M., et al., Viraemia suppressed in HIV-1-infected humans by broadly

neutralizing antibody 3BNC117. Nature, 2015. 522(7557): p. 487-91.

173
 References

178. Caskey, M., et al., Corrigendum: Viraemia suppressed in HIV-1-infected humans

by broadly neutralizing antibody 3BNC117. Nature, 2016.

179. Schoofs, T., et al., HIV-1 therapy with monoclonal antibody 3BNC117 elicits host

immune responses against HIV-1. Science, 2016. 352(6288): p. 997-1001.

180. Scheid, J.F., et al., HIV-1 antibody 3BNC117 suppresses viral rebound in humans

during treatment interruption. Nature, 2016. 535(7613): p. 556-60.

181. Caskey, M., et al., Antibody 10-1074 suppresses viremia in HIV-1-infected

individuals. Nat Med, 2017. 23(2): p. 185-191.

182. Nishimura, Y. and M.A. Martin, Of Mice, Macaques, and Men: Broadly

Neutralizing Antibody Immunotherapy for HIV-1. Cell Host Microbe, 2017.

22(2): p. 207-216.

183. Ng, O.T., et al., HIV type 1 polymerase gene polymorphisms are associated with

phenotypic differences in replication capacity and disease progression. The

Journal of infectious diseases, 2014. 209(1): p. 66-73.

184. Huang, J., et al., Broad and potent HIV-1 neutralization by a human antibody that

binds the gp41-gp120 interface. Nature, 2014. 515(7525): p. 138-42.

185. Falkowska, E., et al., Broadly neutralizing HIV antibodies define a glycan-

dependent epitope on the prefusion conformation of gp41 on cleaved envelope

trimers. Immunity, 2014. 40(5): p. 657-68.

186. Bosque, A., et al., Homeostatic proliferation fails to efficiently reactivate HIV-1

latently infected central memory CD4+ T cells. PLoS Pathog, 2011. 7(10): p.

e1002288.

187. Mehta, S.R., et al., Using phylogeography to characterize the origins of the HIV-

1 subtype F epidemic in Romania. Infect Genet Evol, 2011. 11(5): p. 975-9.

174
 References

188. Li, Y., et al., Mechanism of neutralization by the broadly neutralizing HIV-1

monoclonal antibody VRC01. J Virol, 2011. 85(17): p. 8954-67.

189. Tomaras, G.D., et al., Polyclonal B cell responses to conserved neutralization

epitopes in a subset of HIV-1-infected individuals. J Virol, 2011. 85(21): p. 11502-

19.

190. Raymond, S., et al., Genotypic prediction of HIV-1 subtype D tropism.

Retrovirology, 2011. 8: p. 56.

191. Vandekerckhove, L.P., et al., European guidelines on the clinical management of

HIV-1 tropism testing. Lancet Infect Dis, 2011. 11(5): p. 394-407.

192. Lewis, D.J., et al., Phase I randomised clinical trial of an HIV-1(CN54), clade C,

trimeric envelope vaccine candidate delivered vaginally. PLoS One, 2011. 6(9):

p. e25165.

193. Garcia, F., et al., Safety and immunogenicity of a modified pox vector-based

HIV/AIDS vaccine candidate expressing Env, Gag, Pol and Nef proteins of HIV-

1 subtype B (MVA-B) in healthy HIV-1-uninfected volunteers: A phase I clinical

trial (RISVAC02). Vaccine, 2011. 29(46): p. 8309-16.

194. Churchyard, G.J., et al., A phase IIA randomized clinical trial of a multiclade HIV-

1 DNA prime followed by a multiclade rAd5 HIV-1 vaccine boost in healthy adults

(HVTN204). PLoS One, 2011. 6(8): p. e21225.

195. Leigh Brown, A.J., et al., Transmission network parameters estimated from HIV

sequences for a nationwide epidemic. The Journal of infectious diseases, 2011.

204(9): p. 1463-9.

196. Spearman, P., et al., A trimeric, V2-deleted HIV-1 envelope glycoprotein vaccine

elicits potent neutralizing antibodies but limited breadth of neutralization in

human volunteers. J Infect Dis, 2011. 203(8): p. 1165-73.

175
 References

197. Marston, M., et al., Net survival of perinatally and postnatally HIV-infected

children: a pooled analysis of individual data from sub -Saharan Africa. Int J

Epidemiol, 2011. 40(2): p. 385-396.

198. Chatterjee, A., et al., Implementing services for Early Infant Diagnosis (EID) of

HIV: a comparative descriptive analysis of national programs in four countries.

BMC Public Health, 2011. 11: p. 553.

199. Prosperi, M.C., et al., A novel methodology for large-scale phylogeny partition.

Nat Commun, 2011. 2: p. 321.

200. Price, M.A., et al., Transmitted HIV type 1 drug resistance among individuals with

recent HIV infection in East and Southern Africa. AIDS Res Hum Retroviruses,

2011. 27(1): p. 5-12.

201. Hamers, R.L., et al., HIV-1 drug resistance in antiretroviral-naive individuals in

sub-Saharan Africa after rollout of antiretroviral therapy: a multicentre

observational study. Lancet Infect Dis, 2011. 11(10): p. 750-9.

202. Djoko, C.F., et al., High HIV type 1 group M pol diversity and low rate of

antiretroviral resistance mutations among the uniformed services in Kinshasa,

Democratic Republic of the Congo. AIDS Res Hum Retroviruses, 2011. 27(3): p.

323-9.

203. Hemelaar, J., et al., Global trends in molecular epidemiology of HIV-1 during

2000-2007. AIDS, 2011. 25(5): p. 679-89.

204. Walker, L.M., et al., Broad neutralization coverage of HIV by multiple highly

potent antibodies. Nature, 2011. 477(7365): p. 466-70.

205. Santos, A., et al., Evaluation of the diagnostic performance of the rapid test VIKIA

HIV1/2 in a highly complex HIV-1 epidemic. Diagn Microbiol Infect Dis, 2011.

71(1): p. 90-2.

176
 References

206. Gupta, R.K., et al., Global trends in antiretroviral resistance in treatment-naive

individuals with HIV after rollout of antiretroviral treatment in resource -limited

settings: a global collaborative study and meta-regression analysis. Lancet, 2012.

380(9849): p. 1250-8.

207. Sullivan, P.S., et al., Prevalence of seroconversion symptoms and relationship to

set-point viral load: findings from a subtype C epidemic, 1995 -2009. AIDS, 2012.

26(2): p. 175-84.

208. Yebra, G., et al., Most HIV type 1 non-B infections in the Spanish cohort of

antiretroviral treatment-naive HIV-infected patients (CoRIS) are due to

recombinant viruses. J Clin Microbiol, 2012. 50(2): p. 407-13.

209. Delatorre, E. and G. Bello, Phylodynamics of the HIV-1 epidemic in Cuba. PLoS

One, 2013. 8(9): p. e72448.

210. Stadeli, K.M. and D.D. Richman, Rates of emergence of HIV drug resistance in

resource-limited settings: a systematic review. Antivir Ther, 2013. 18(1): p. 115-

23.

211. Hall, B.G., Building phylogenetic trees from molecular data with MEGA. Mol

Biol Evol, 2013. 30(5): p. 1229-35.

212. Calado, R., et al., A Prime-Boost Immunization Strategy with Vaccinia Virus

Expressing Novel gp120 Envelope Glycoprotein from a CRF02_AG Isolate Elicits

Cross-Clade Tier 2 HIV-1 Neutralizing Antibodies. Vaccines (Basel), 2020. 8(2).

213. Julg, B. and D.H. Barouch, Neutralizing antibodies for HIV-1 prevention. Curr

Opin HIV AIDS, 2019. 14(4): p. 318-324.

214. Cohen, Y.Z. and M. Caskey, Broadly neutralizing antibodies for treatment and

prevention of HIV-1 infection. Curr Opin HIV AIDS, 2018. 13(4): p. 366-373.

177
 References

215. McGuire, A.T., et al., Diverse recombinant HIV-1 Envs fail to activate B cells

expressing the germline B cell receptors of the broadly neutralizing anti-HIV-1

antibodies PG9 and 447-52D. J Virol, 2014. 88(5): p. 2645-57.

216. McGuire, A.T., et al., Specifically modified Env immunogens activate B-cell

precursors of broadly neutralizing HIV-1 antibodies in transgenic mice. Nat

Commun, 2016. 7: p. 10618.

217. Nityanandam, R. and R. Serra-Moreno, BCA2/Rabring7 targets HIV-1 Gag for

lysosomal degradation in a tetherin-independent manner. PLoS Pathog, 2014.

10(5): p. e1004151.

218. Troyano-Hernaez, P., et al., Update in natural antiretroviral-associated mutations

among HIV-2 variants and discrepancies across HIV-2 resistance interpretation

tools. AIDS Res Hum Retroviruses, 2021.

219. Laville, P., M. Petitjean, and L. Regad, Structural Impacts of Drug-Resistance

Mutations Appearing in HIV-2 Protease. Molecules, 2021. 26(3).

220. Zhang, Z., et al., HIV-2 Vif and foamy virus Bet antagonize APOBEC3B by

different mechanisms. Virology, 2021. 554: p. 17-27.

221. Umunnakwe, C.N., et al., Specific Guanosines in the HIV-2 Leader RNA are

Essential for Efficient Viral Genome Packaging. J Mol Biol, 2021. 433(2): p.

166718.

222. Berzow, D., et al., Human Immunodeficiency Virus-2 (HIV-2): A Summary of the

Present Standard of Care and Treatment Options for Individuals Living with HIV-

2 in Western Europe. Clin Infect Dis, 2021. 72(3): p. 503-509.

223. Hu, Y., et al., Virus Evolution and Neutralization Sensitivity in an HIV-1 Subtype

B' Infected Plasma Donor with Broadly Neutralizing Activity. Vaccines (Basel),

2021. 9(4).

178
 References

224. McFarland, E.J., et al., Safety, Tolerability, and Pharmacokinetics of a Long-

Acting Broadly Neutralizing HIV-1 Monoclonal Antibody VRC01LS in HIV-1-

Exposed Newborn Infants. J Infect Dis, 2021.

225. Sutcliffe, C.G., et al., The NSEBA Demonstration Project: implementation of a

point-of-care platform for early infant diagnosis of HIV in rural Zambia. Trop

Med Int Health, 2021.

226. De Broucker, G., et al., The cost-effectiveness of scaling-up rapid point-of-care

testing for early infant diagnosis of HIV in southern Zambia. PLoS One, 2021.

16(3): p. e0248217.

227. Sutcliffe, C.G., et al., Point-of-care p24 antigen detection for early infant

diagnosis of HIV infection: cross-sectional and longitudinal studies in Zambia.

BMC Infect Dis, 2021. 21(1): p. 118.

228. Etoori, D., et al., 'If the results are negative, they motivate us'. Experiences of

early infant diagnosis of HIV and engagement in Option B. Glob Public Health,

2021. 16(2): p. 186-200.

229. Krumm, S.A., et al., Mechanisms of escape from the PGT128 family of anti-HIV

broadly neutralizing antibodies. Retrovirology, 2016. 13(1): p. 8.

230. Mouquet, H., Antibody B cell responses in HIV-1 infection. Trends Immunol,

2014. 35(11): p. 549-61.

231. Chen, B., et al., Determining the structure of an unliganded and fully glycosylated

SIV gp120 envelope glycoprotein. Structure, 2005. 13(2): p. 197-211.

232. Zhu, P., et al., Distribution and three-dimensional structure of AIDS virus

envelope spikes. Nature, 2006. 441(7095): p. 847-52.

233. Brandenberg, O.F., et al., The HIV-1 Entry Process: A Stoichiometric View.

Trends Microbiol, 2015. 23(12): p. 763-774.

179
 References

234. Baba, T.W., et al., Human neutralizing monoclonal antibodies of the IgG1 subtype

protect against mucosal simian-human immunodeficiency virus infection. Nat

Med, 2000. 6(2): p. 200-6.

235. Mascola, J.R., S.S. Frankel, and K. Broliden, HIV-1 entry at the mucosal surface:

role of antibodies in protection. AIDS, 2000. 14 Suppl 3: p. S167-74.

236. Hofmann-Lehmann, R., et al., Postnatal passive immunization of neonatal

macaques with a triple combination of human monoclonal antibodies against oral

simian-human immunodeficiency virus challenge. J Virol, 2001. 75(16): p. 7470-

80.

237. Veazey, R.S., et al., Protection of macaques from vaginal SHIV challenge by

vaginally delivered inhibitors of virus-cell fusion. Nature, 2005. 438(7064): p. 99-

102.

238. Sun, M., et al., VRC01 antibody protects against vaginal and rectal transmission

of human immunodeficiency virus 1 in hu-BLT mice. Arch Virol, 2016. 161(9): p.

2449-55.

239. Shibata, R., et al., Neutralizing antibody directed against the HIV-1 envelope

glycoprotein can completely block HIV-1/SIV chimeric virus infections of

macaque monkeys. Nat Med, 1999. 5(2): p. 204-10.

240. Caskey, M., F. Klein, and M.C. Nussenzweig, Broadly Neutralizing Antibodies

for HIV-1 Prevention or Immunotherapy. N Engl J Med, 2016. 375(21): p. 2019-

2021.

241. Horwitz, J.A., et al., HIV-1 suppression and durable control by combining single

broadly neutralizing antibodies and antiretroviral drugs in humanized mice. Proc

Natl Acad Sci U S A, 2013. 110(41): p. 16538-43.

180
 References

242. Caskey, M., et al., Corrigendum: Viraemia suppressed in HIV-1-infected humans

by broadly neutralizing antibody 3BNC117. Nature, 2016. 535(7613): p. 580.

243. Learmont, J.C., et al., Immunologic and virologic status after 14 to 18 years of

infection with an attenuated strain of HIV-1. A report from the Sydney Blood Bank

Cohort. N Engl J Med, 1999. 340(22): p. 1715-22.

244. Baba, T.W., et al., Live attenuated, multiply deleted simian immunodeficiency

virus causes AIDS in infant and adult macaques. Nat Med, 1999. 5(2): p. 194-

203.

245. Giri, M., K.E. Ugen, and D.B. Weiner, DNA vaccines against human

immunodeficiency virus type 1 in the past decade. Clin Microbiol Rev, 2004.

17(2): p. 370-89.

246. Estcourt, M.J., A.J. McMichael, and T. Hanke, DNA vaccines against human

immunodeficiency virus type 1. Immunol Rev, 2004. 199: p. 144-55.

247. Priddy, F.H., et al., Safety and immunogenicity of a replication-incompetent

adenovirus type 5 HIV-1 clade B gag/pol/nef vaccine in healthy adults. Clin Infect

Dis, 2008. 46(11): p. 1769-81.

248. Zhang, M., et al., DNA prime-protein boost using subtype consensus Env was

effective in eliciting neutralizing antibody responses against subtype BC HIV-1

viruses circulating in China. Hum Vaccin Immunother, 2012. 8(11): p. 1630-7.

249. Chapman, R., et al., Heterologous prime-boost vaccination with DNA and MVA

vaccines, expressing HIV-1 subtype C mosaic Gag virus-like particles, is highly

immunogenic in mice. PLoS One, 2017. 12(3): p. e0173352.

250. Morris, L., mRNA vaccines offer hope for HIV. Nat Med, 2021. 27(12): p. 2082-

2084.

181
 References

251. Santos-Ferreira, M.O., et al., A study of seroprevalence of HIV-1 and HIV-2 in six

provinces of People's Republic of Angola: clues to the spread of HIV infection. J

Acquir Immune Defic Syndr, 1990. 3(8): p. 780-6.

252. Bartolo, I., et al., Rare HIV-1 subtype J genomes and a new H/U/CRF02_AG

recombinant genome suggests an ancient origin of HIV-1 in Angola. AIDS Res

Hum Retroviruses, 2016.

253. Violari, A., et al., Early antiretroviral therapy and mortality among HIV-infected

infants. N Engl J Med, 2008. 359(21): p. 2233-44.

254. Bitnun, A., et al., Early initiation of combination antiretroviral therapy in HIV-1-

infected newborns can achieve sustained virologic suppression with low

frequency of CD4+ T cells carrying HIV in peripheral blood. Clinical infectious

diseases : an official publication of the Infectious Diseases Society of America,

2014. 59(7): p. 1012-9.

255. Catumbela E, F.A., Serrano D, Furtado ML, Gomes M, Shiraishi RW et al., The

Angola HIV epidemic, 2004-11: a case of change or stability? . Lancet, 2013.

382(suppl 2): p. s17.

256. Little, S.J., et al., Antiretroviral-drug resistance among patients recently infected

with HIV. N Engl J Med, 2002. 347(6): p. 385-94.

257. Bennett, D.E., et al., Recommendations for surveillance of transmitted HIV drug

resistance in countries scaling up antiretroviral treatment. Antivir Ther, 2008. 13

Suppl 2: p. 25-36.

258. Albert, T.L.a.J., Reconstruction of HIV-1 Transmission Chains for Forensic

Purposes. AIDS Reviews, 2000(2): p. 241-251.

259. Los Alamos Sequence Database (2014) Los Alamos Sequence Database. Los

Alamos National Laboratory, Los Alamos, New Mexico. Available:

182
 References

[Link] Accessed 7

August 2014.

260. Frentz, D., et al., Limited cross-border infections in patients newly diagnosed with

HIV in Europe. Retrovirology, 2013. 10: p. 36.

261. Larkin, M.A., et al., Clustal W and Clustal X version 2.0. Bioinformatics, 2007.

23(21): p. 2947-8.

262. Felsenstein, J., Evolutionary trees from DNA sequences: a maximum likelihood

approach. J Mol Evol, 1981. 17(6): p. 368-76.

263. Posada, D. and K.A. Crandall, MODELTEST: testing the model of DNA

substitution. Bioinformatics, 1998. 14(9): p. 817-8.

264. Gouy, M., S. Guindon, and O. Gascuel, SeaView version 4: A multiplatform

graphical user interface for sequence alignment and phylogenetic tree building.

Mol Biol Evol, 2010. 27(2): p. 221-4.

265. Lole, K.S., et al., Full-length human immunodeficiency virus type 1 genomes from

subtype C-infected seroconverters in India, with evidence of intersubtype

recombination. J Virol, 1999. 73(1): p. 152-60.

266. Chalmet, K., et al., Epidemiological study of phylogenetic transmission clusters

in a local HIV-1 epidemic reveals distinct differences between subtype B and non-

B infections. BMC Infect Dis, 2010. 10: p. 262.

267. Kumar, S., et al., AIR: A batch-oriented web program package for construction

of supermatrices ready for phylogenomic analyses. BMC Bioinformatics, 2009.

10: p. 357.

268. Lewis, F., et al., Episodic sexual transmission of HIV revealed by molecular

phylodynamics. PLoS Med, 2008. 5(3): p. e50.

183
 References

269. Price, M.N., P.S. Dehal, and A.P. Arkin, FastTree 2--approximately maximum-

likelihood trees for large alignments. PLoS One, 2010. 5(3): p. e9490.

270. Stanford HIV Drug Resistance Database (2014) Stanford HIV Drug Resistance

Database. Stanford University. Available: [Link] Accessed

2014 Jan 22.

271. Bennett, D.E., et al., Drug resistance mutations for surveillance of transmitted

HIV-1 drug-resistance: 2009 update. PLoS One, 2009. 4(3): p. e4724.

272. Johnson, V.A., et al., Update of the drug resistance mutations in HIV-1: March

2013. Top Antivir Med, 2013. 21(1): p. 6-14.

273. Baxter, J.D., et al., Genotypic changes in human immunodeficiency virus type 1

protease associated with reduced susceptibility and virologic response to the

protease inhibitor tipranavir. J Virol, 2006. 80(21): p. 10794-801.

274. Poveda, E., et al., Prevalence of darunavir resistance mutations in HIV-1-infected

patients failing other protease inhibitors. J Antimicrob Chemother, 2007. 60(4):

p. 885-8.

275. de Meyer, S., et al., Resistance profile of darunavir: combined 24-week results

from the POWER trials. AIDS Res Hum Retroviruses, 2008. 24(3): p. 379-88.

276. Holguin, A., et al., Efficacy of antiretroviral therapy in individuals infected with

HIV-1 non-B subtypes. AIDS Rev, 2006. 8(2): p. 98-107.

277. van Westen, G.J., et al., Significantly improved HIV inhibitor efficacy prediction

employing proteochemometric models generated from antivirogram data. PLoS

Comput Biol, 2013. 9(2): p. e1002899.

278. Vermeiren, H., et al., Prediction of HIV-1 drug susceptibility phenotype from the

viral genotype using linear regression modeling. J Virol Methods, 2007. 145(1):

p. 47-55.

184
 References

279. Kempf, D.J., et al., Analysis of the virological response with respect to baseline

viral phenotype and genotype in protease inhibitor-experienced HIV-1-infected

patients receiving lopinavir/ritonavir therapy. Antivir Ther, 2002. 7(3): p. 165-

74.

280. Pellegrin, I., et al., Interpretation of genotype and pharmacokinetics for resistance

to fosamprenavir-ritonavir-based regimens in antiretroviral-experienced

patients. Antimicrob Agents Chemother, 2007. 51(4): p. 1473-80.

281. Pellegrin, I., et al., Virological responses to atazanavir-ritonavir-based regimens:

resistance-substitutions score and pharmacokinetic parameters (Reyaphar study).

Antivir Ther, 2006. 11(4): p. 421-9.

282. Brenner, B.G., et al., High rates of forward transmission events after acute/early

HIV-1 infection. J Infect Dis, 2007. 195(7): p. 951-9.

283. Langford, S.E., J. Ananworanich, and D.A. Cooper, Predictors of disease

progression in HIV infection: a review. AIDS Res Ther, 2007. 4: p. 11.

284. Mahnke, Y.D., et al., Early immunologic and virologic predictors of clinical HIV-

1 disease progression. AIDS, 2013. 27(5): p. 697-706.

285. Instituto Nacional de Luta Contra a SIDA - Angola (2010) UNGASS 2010 -

Relato´ rio sobre o Progresso do Paı´s para dar Seguimento aos Compromissos

da Sessa˜o Especial sobre VIH e SIDA da Assembleia Geral das Nac¸o˜es Unidas,

2008–2009. Avalaible:

[Link] gressrepor

ts/2010countries/angola_2010_country_progress_report_es.pdf. Accessed 2014

Feb 15.

185
 References

286. Nwobegahay, J., et al., Low prevalence of transmitted genetic drug resistance in

a cohort of HIV infected naive patients entering antiretroviral treatment programs

at two sites in northern South Africa. J Med Virol, 2012. 84(12): p. 1839-43.

287. Hunt, G.M., et al., Surveillance of transmitted HIV-1 drug resistance in Gauteng

and KwaZulu-Natal Provinces, South Africa, 2005-2009. Clin Infect Dis, 2012.

54 Suppl 4: p. S334-8.

288. Aghokeng, A.F., et al., Evaluation of transmitted HIV drug resistance among

recently-infected antenatal clinic attendees in four Central African countries.

Antivir Ther, 2009. 14(3): p. 401-11.

289. Castelbranco, E.P., et al., Frequency of primary resistance to antiretroviral drugs

and genetic variability of HIV-1 among infected pregnant women recently

diagnosed in Luanda-Angola. AIDS Res Hum Retroviruses, 2010. 26(12): p.

1313-6.

290. Rambaut, A., et al., The causes and consequences of HIV evolution. Nat Rev

Genet, 2004. 5(1): p. 52-61.

291. Ministério da Saúde de Angola (2009) Boletim Epidemiológico VIH e SIDA.

Instituto Nacional de Luta Contra a SIDA. Ano 0 N˚ 02 - Julho a Dezembro de

2009. Angola. Avalaible: [Link] Accessed 20 July 2011.

292. Hue, S., et al., Genetic analysis reveals the complex structure of HIV-1

transmission within defined risk groups. Proc Natl Acad Sci U S A, 2005. 102(12):

p. 4425-9.

293. Brenner, B., M.A. Wainberg, and M. Roger, Phylogenetic inferences on HIV-1

transmission: implications for the design of prevention and treatment

interventions. AIDS, 2013. 27(7): p. 1045-57.

186
 References

294. Smit, P.W., et al., Systematic review of the use of dried blood spots for monitoring

HIV viral load and for early infant diagnosis. PLoS One, 2014. 9(3): p. e86461.

295. Field, N., et al., Strengthening the Reporting of Molecular Epidemiology for

Infectious Diseases (STROME-ID): an extension of the STROBE statement.

Lancet Infect Dis, 2014. 14(4): p. 341-52.

296. WHO, HIV Assays: Laboratory performance and other operational

characteristics rapid diagnostic tests Combined detection of HIV-1/2 antibodies

and discriminatory detection of HIV-1 and HIV-2 antibodies. . 2015, World

Health Organization: Geneva.

297. Ghosh, S.K., et al., A molecular clone of HIV-1 tropic and cytopathic for human

and chimpanzee lymphocytes. Virology, 1993. 194(2): p. 858-64.

298. Folks, T.M., et al., Tumor necrosis factor alpha induces expression of human

immunodeficiency virus in a chronically infected T-cell clone. Proceedings of the

National Academy of Sciences of the United States of America, 1989. 86(7): p.

2365-8.

299. Clouse, K.A., et al., Monokine regulation of human immunodeficiency virus-1

expression in a chronically infected human T cell clone. Journal of immunology,

1989. 142(2): p. 431-8.

300. Denny, T., et al., Lymphocyte subsets in healthy children during the first 5 years

of life. JAMA, 1992. 267(11): p. 1484-8.

301. Tetali, S., et al., Virus load as a marker of disease progression in HIV-infected

children. AIDS Res Hum Retroviruses, 1996. 12(8): p. 669-75.

302. Chang, J., et al., Field evaluation of Abbott Real Time HIV-1 Qualitative test for

early infant diagnosis using dried blood spots samples in comparison to Roche

187
 References

COBAS Ampliprep/COBAS TaqMan HIV-1 Qual test in Kenya. Journal of

virological methods, 2014. 204: p. 25-30.

303. Abbott and Molecular, RealTime HIV-1 Qualitative. Product Description, Abbott

Molecular.

304. WHO, Sources and prices of selected medicines and diagnostics for people living

with HIV/AIDS/ a joint UNICEF, UNAIDS, WHO, MSF project- 6th edition. 2005,

World Health Organization: Geneva.

305. Lussiana, C., et al., Effectiveness of a prevention of mother-to-child HIV

transmission programme in an urban hospital in Angola. PLoS One, 2012. 7(4):

p. e36381.

306. Fischer, A., et al., Simple DNA extraction method for dried blood spots and

comparison of two PCR assays for diagnosis of vertical human immunodeficiency

virus type 1 transmission in Rwanda. J Clin Microbiol, 2004. 42(1): p. 16-20.

307. Sciences, G.H.L., Reliable extraction of DNA from Whatman™ FTA™ cards GE

Healthcare Application note 28-9822-22 AA. 2010, GE Healthcare Life Sciences.

308. Swanson, P., et al., Impact of human immunodeficiency virus type 1 (HIV-1)

genetic diversity on performance of four commercial viral load assays: LCx HIV

RNA Quantitative, AMPLICOR HIV-1 MONITOR v1.5, VERSANT HIV-1 RNA

3.0, and NucliSens HIV-1 QT. J Clin Microbiol, 2005. 43(8): p. 3860-8.

309. Amendola, A., et al., Underevaluation of HIV-1 plasma viral load by a

commercially available assay in a cluster of patients infected with HIV-1 A/G

circulating recombinant form (CRF02). Journal of acquired immune deficiency

syndromes, 2002. 31(5): p. 488-94.

310. Antunes, R., et al., Evaluation of the clinical sensitivities of three viral load assays

with plasma samples from a pediatric population predominantly infected with

188
 References

human immunodeficiency virus type 1 subtype G and BG recombinant forms. J

Clin Microbiol, 2003. 41(7): p. 3361-7.

311. Nkengasong, J.N., et al., Quantification of RNA in HIV type 1 subtypes D and G

by NucliSens and Amplicor assays in Abidjan, Ivory Coast. AIDS Res Hum

Retroviruses, 1999. 15(6): p. 495-8.

312. Hu, X., et al., Profiling the neutralizing antibody response in chronically HIV-1

CRF07_BC-infected intravenous drug users naive to antiretroviral therapy. Sci

Rep, 2017. 7: p. 46308.

313. Brandenberg, O.F., et al., Predicting HIV-1 transmission and antibody

neutralization efficacy in vivo from stoichiometric parameters. PLoS Pathog,

2017. 13(5): p. e1006313.

314. Bonsignori, M., et al., Staged induction of HIV-1 glycan-dependent broadly

neutralizing antibodies. Sci Transl Med, 2017. 9(381).

315. Anthony, C., et al., Cooperation between Strain-Specific and Broadly

Neutralizing Responses Limited Viral Escape and Prolonged the Exposure of the

Broadly Neutralizing Epitope. J Virol, 2017. 91(18).

316. Martinez, D.R., et al., Maternal Binding and Neutralizing IgG Responses

Targeting the C-Terminal Region of the V3 Loop Are Predictive of Reduced

Peripartum HIV-1 Transmission Risk. J Virol, 2017. 91(9).

317. Rocha, C., et al., Potency of HIV-2-specific antibodies increase in direct

association with loss of memory B cells. AIDS, 2017. 31(17): p. 2431-2433.

318. Doria-Rose, N.A., et al., Mapping Polyclonal HIV-1 Antibody Responses via

Next-Generation Neutralization Fingerprinting. PLoS Pathog, 2017. 13(1): p.

e1006148.

189
 References

319. Martin, F., et al., Early infant diagnosis of HIV-1 infection in Luanda, Angola,

using a new DNA PCR assay and dried blood spots. PLoS One, 2017. 12(7): p.

e0181352.

320. Spencer, D.A., et al., Polyfunctional Tier 2-Neutralizing Antibodies Cloned

following HIV-1 Env Macaque Immunization Mirror Native Antibodies in a

Human Donor. J Immunol, 2021. 206(5): p. 999-1012.

321. Spencer, D.A., et al., Advancing HIV Broadly Neutralizing Antibodies: From

Discovery to the Clinic. Front Public Health, 2021. 9: p. 690017.

322. Kwong, P.D. and J.R. Mascola, HIV-1 Vaccines Based on Antibody Identification,

B Cell Ontogeny, and Epitope Structure. Immunity, 2018. 48(5): p. 855-871.

323. Simonich, C.A., et al., HIV-1 Neutralizing Antibodies with Limited

Hypermutation from an Infant. Cell, 2016. 166(1): p. 77-87.

324. Marcelino, J.M., et al., Resistance to antibody neutralization in HIV-2 infection

occurs in late stage disease and is associated with X4 tropism. AIDS, 2012.

26(18): p. 2275-84.

325. Rocha, C., et al., Evolution of the human immunodeficiency virus type 2 envelope

in the first years of infection is associated with the dynamics of the neutralizing

antibody response. Retrovirology, 2013. 10: p. 110.

326. Serra, P.A., N. Taveira, and R.C. Guedes, Computational Modulation of the V3

Region of Glycoprotein gp125 of HIV-2. Int J Mol Sci, 2021. 22(4).

327. Kostrikis, L.G., et al., Quantitative analysis of serum neutralization of human

immunodeficiency virus type 1 from subtypes A, B, C, D, E, F, and I: lack of direct

correlation between neutralization serotypes and genetic subtypes and evidence

for prevalent serum-dependent infectivity enhancement. J Virol, 1996. 70(1): p.

445-58.

190
 References

328. Trkola, A., et al., Human monoclonal antibody 2G12 defines a distinctive

neutralization epitope on the gp120 glycoprotein of human immunodefic iency

virus type 1. J Virol, 1996. 70(2): p. 1100-8.

329. Ditse, Z., et al., Effect of HIV Envelope Vaccination on the Subsequent Antibody

Response to HIV Infection. mSphere, 2020. 5(1).

330. Marcelino, R., et al., Antibody response against selected epitopes in the HIV-1

envelope gp41 ectodomain contributes to reduce viral burden in HIV-1 infected

patients. Sci Rep, 2021. 11(1): p. 8993.

331. Faria, N.R., et al., HIV epidemiology. The early spread and epidemic ignition of

HIV-1 in human populations. Science, 2014. 346(6205): p. 56-61.

332. Platt, E.J., et al., Effects of CCR5 and CD4 cell surface concentrations on

infections by macrophagetropic isolates of human immunodeficiency virus type 1.

J Virol, 1998. 72(4): p. 2855-64.

333. Tamura, K., et al., MEGA6: Molecular Evolutionary Genetics Analysis version

6.0. Mol Biol Evol, 2013. 30(12): p. 2725-9.

334. Guindon, S., et al., New algorithms and methods to estimate maximum-likelihood

phylogenies: assessing the performance of PhyML 3.0. Syst Biol, 2010. 59(3): p.

307-21.

335. Weaver, S., et al., Datamonkey 2.0: A Modern Web Application for

Characterizing Selective and Other Evolutionary Processes. Mol Biol Evol, 2018.

35(3): p. 773-777.

336. Kosakovsky Pond, S.L. and S.D. Frost, Not so different after all: a comparison of

methods for detecting amino acid sites under selection. Mol Biol Evol, 2005.

22(5): p. 1208-22.

191
 References

337. Zhang, M., et al., Tracking global patterns of N-linked glycosylation site variation

in highly variable viral glycoproteins: HIV, SIV, and HCV envelopes and

influenza hemagglutinin. Glycobiology, 2004. 14(12): p. 1229-46.

338. Mascola, J.R., et al., Recommendations for the design and use of standard virus

panels to assess neutralizing antibody responses elicited by candidate human

immunodeficiency virus type 1 vaccines. J Virol, 2005. 79(16): p. 10103-7.

339. Metsalu, T. and J. Vilo, ClustVis: a web tool for visualizing clustering of

multivariate data using Principal Component Analysis and heatmap. Nucleic

Acids Res, 2015. 43(W1): p. W566-70.

340. Firth, H.V., et al., DECIPHER: Database of Chromosomal Imbalance and

Phenotype in Humans Using Ensembl Resources. Am J Hum Genet, 2009. 84(4):

p. 524-33.

341. R Core Team, R: A language and environment for statistical computing, 2017.

Disponível em [Link]

342. Wickham, H., ggplot2: Elegant Graphics for Data Analysis. Springer -Verlag,

2016.

343. Mellors, J.W., et al., Plasma viral load and CD4+ lymphocytes as prognostic

markers of HIV-1 infection. Ann Intern Med, 1997. 126(12): p. 946-54.

344. Brunel, F.M., et al., Structure-function analysis of the epitope for 4E10, a broadly

neutralizing human immunodeficiency virus type 1 antibody. J Virol, 2006. 80(4):

p. 1680-7.

345. Barroso, H., et al., Evolutionary and structural features of the C2, V3 and C3

envelope regions underlying the differences in HIV-1 and HIV-2 biology and

infection. PLoS One, 2011. 6(1): p. e14548.

192
 References

346. Mabvakure, B.M., et al., Positive Selection at Key Residues in the HIV Envelope

Distinguishes Broad and Strain-Specific Plasma Neutralizing Antibodies. J Virol,

2019. 93(6).

347. Dingens, A.S., et al., An Antigenic Atlas of HIV-1 Escape from Broadly

Neutralizing Antibodies Distinguishes Functional and Structural Epitopes.

Immunity, 2019. 50(2): p. 520-532 e3.

348. Bricault, C.A., et al., HIV-1 Neutralizing Antibody Signatures and Application to

Epitope-Targeted Vaccine Design. Cell Host Microbe, 2019. 26(2): p. 296.

349. Sebastiao, C.S., J. Morais, and M. Brito, Factors Influencing HIV Drug

Resistance among Pregnant Women in Luanda, Angola: Findings from a Cross-

Sectional Study. Trop Med Infect Dis, 2021. 6(1).

350. Bhiman, J.N., et al., Viral variants that initiate and drive maturation of V1V2-

directed HIV-1 broadly neutralizing antibodies. Nat Med, 2015. 21(11): p. 1332-

6.

351. Liao, H.X., et al., Co-evolution of a broadly neutralizing HIV-1 antibody and

founder virus. Nature, 2013. 496(7446): p. 469-76.

352. Beirnaert, E., et al., Identification and characterization of sera from HIV-infected

individuals with broad cross-neutralizing activity against group M (env clade A-

H) and group O primary HIV-1 isolates. J Med Virol, 2000. 62(1): p. 14-24.

353. Kouyos, R.D., et al., Tracing HIV-1 strains that imprint broadly neutralizing

antibody responses. Nature, 2018. 561(7723): p. 406-410.

354. Sather, D.N., et al., Factors associated with the development of cross-reactive

neutralizing antibodies during human immunodeficiency virus type 1 infection. J

Virol, 2009. 83(2): p. 757-69.

193
 References

355. Moir, S. and A.S. Fauci, B-cell responses to HIV infection. Immunol Rev, 2017.

275(1): p. 33-48.

356. Badura, R., et al., Early ART in Acute HIV-1 Infection: Impact on the B-Cell

Compartment. Front Cell Infect Microbiol, 2020. 10: p. 347.

357. Moir, S., et al., B cells in early and chronic HIV infection: evidence for

preservation of immune function associated with early initiation of antiretroviral

therapy. Blood, 2010. 116(25): p. 5571-9.

358. Gach, J.S., et al., HIV-1 specific antibody titers and neutralization among

chronically infected patients on long-term suppressive antiretroviral therapy

(ART): a cross-sectional study. PLoS One, 2014. 9(1): p. e85371.

359. Cizmeci, D., et al., Distinct clonal evolution of B-cells in HIV controllers with

neutralizing antibody breadth. Elife, 2021. 10.

360. Kong, R., et al., Broad and potent neutralizing antibody responses elicited in

natural HIV-2 infection. J Virol, 2012. 86(2): p. 947-60.

361. Rodriguez, S.K., et al., Comparison of heterologous neutralizing antibody

responses of human immunodeficiency virus type 1 (HIV-1)- and HIV-2-infected

Senegalese patients: distinct patterns of breadth and magnitude distinguish HIV-

1 and HIV-2 infections. J Virol, 2007. 81(10): p. 5331-8.

362. Bai, H., et al., The breadth of HIV-1 neutralizing antibodies depends on the

conservation of key sites in their epitopes. PLoS Comput Biol, 2019. 15(6): p.

e1007056.

363. Rademeyer, C., et al., Genetic characteristics of HIV-1 subtype C envelopes

inducing cross-neutralizing antibodies. Virology, 2007. 368(1): p. 172-81.

194
 References

364. Kumar, S., et al., An HIV-1 Broadly Neutralizing Antibody from a Clade C-

Infected Pediatric Elite Neutralizer Potently Neutralizes the Contemporaneous

and Autologous Evolving Viruses. J Virol, 2019. 93(4).

365. Cheedarla, N., et al., Evolution of Neutralization Response in HIV-1 Subtype C-

Infected Individuals Exhibiting Broad Cross-Clade Neutralization of HIV-1

Strains. Front Immunol, 2018. 9: p. 618.

366. Lei, L., et al., The HIV-1 Envelope Glycoprotein C3/V4 Region Defines a

Prevalent Neutralization Epitope following Immunization. Cell Rep, 2019. 27(2):

p. 586-598 e6.

367. Vueba, A.N., et al., Prevalence of HIV and hepatitis B virus among pregnant

women in Luanda (Angola): geospatial distribution and its association with socio-

demographic and clinical-obstetric determinants. Virol J, 2021. 18(1): p. 239.

368. Matsinhe, M., et al., Inpatient Point-of-Care HIV Early Infant Diagnosis in

Mozambique to Improve Case Identification and Linkage to Antiretroviral

Therapy. Glob Health Sci Pract, 2021. 9(1): p. 31-39.

369. Leruez-Ville, M., et al., Detection of cytomegalovirus DNA on dried blood spots

collected from infants infected with HIV: an in-house method adaptable in

resource-limited settings. J Virol Methods, 2013. 193(2): p. 503-7.

370. Moragas, M., M.D. Golemba, and A. Mangano, A new highly sensitive single-tube

nested real-time PCR assay: Clinical utility in perinatal HIV-1 diagnosis. J Virol

Methods, 2021. 297: p. 114273.

371. De Vitis, E., et al., Real-time polymerase chain reaction on filter paper spotted

samples: a gateway to molecular diagnosis of invasive bacterial diseases for rural

areas in low-income countries. Trans R Soc Trop Med Hyg, 2022. 116(3): p. 233-

241.

195
 References

372. Guan, B., et al., Sensitive extraction-free SARS-CoV-2 RNA virus detection using

a chelating resin. iScience, 2021. 24(9): p. 102960.

373. WHO, Diagnostic testing for SARS-CoV-2: interim guidance, 2020. Available:

[Link]

Accessed 10th December 2021

374. Anthony, C., et al., Correction for Anthony et al., "Cooperation between Strain-

Specific and Broadly Neutralizing Responses Limited Viral Escape and

Prolonged the Exposure of the Broadly Neutralizing Epitope". J Virol, 2018.

92(9).

375. Bottiger, B., et al., Envelope cross-reactivity between human immunodeficiency

virus types 1 and 2 detected by different serological methods: correlation between

cross-neutralization and reactivity against the main neutralizing site. J Virol,

1990. 64(7): p. 3492-9.

376. Vidyavijayan, K.K., et al., Cross Type Neutralizing Antibodies Detected in a

Unique HIV-2 Infected Individual From India. Front Immunol, 2018. 9: p. 2841.

377. Doria-Rose, N.A., et al., Breadth of human immunodeficiency virus-specific

neutralizing activity in sera: clustering analysis and association with clinical

variables. J Virol, 2010. 84(3): p. 1631-6.

378. Seabright, G.E., et al., Networks of HIV-1 Envelope Glycans Maintain Antibody

Epitopes in the Face of Glycan Additions and Deletions. Structure, 2020. 28(8):

p. 897-909 e6.

379. Gray, E.S., et al., N-linked glycan modifications in gp120 of human

immunodeficiency virus type 1 subtype C render partial sensitivity to 2G12

antibody neutralization. J Virol, 2007. 81(19): p. 10769-76.

196
 References

380. Polzer, S., et al., The N-linked glycan g15 within the V3 loop of the HIV-1 external

glycoprotein gp120 affects coreceptor usage, cellular tropism, and neutralization.

Virology, 2002. 304(1): p. 70-80.

381. Chackerian, B., L.M. Rudensey, and J. Overbaugh, Specific N-linked and O-linked

glycosylation modifications in the envelope V1 domain of simian

immunodeficiency virus variants that evolve in the host alter recognition by

neutralizing antibodies. J Virol, 1997. 71(10): p. 7719-27.

382. Sagar, M., et al., Human immunodeficiency virus type 1 V1-V2 envelope loop

sequences expand and add glycosylation sites over the course of infection, and

these modifications affect antibody neutralization sensitivity. J Virol, 2006.

80(19): p. 9586-98.

383. Han, Q., et al., Difficult-to-neutralize global HIV-1 isolates are neutralized by

antibodies targeting open envelope conformations. Nat Commun, 2019. 10(1): p.

2898.

197