REVIEW doi: 10.1111/j.1365-3083.2007.01974.

x
..................................................................................................................................................................

Rheumatic Fever and Rheumatic Heart Disease:

Genetics and Pathogenesis
L. Guilherme*, , R. Ramasawmy*, & J. Kalil*, ,à

Abstract
*Heart Institute (InCor), School of Medicine, Molecular mimicry between streptococcal and human proteins is considered as
University of São Paulo; Institute for the triggering factor leading to autoimmunity in rheumatic fever (RF) and
Immunology Investigation, Millenium Institute;
rheumatic heart disease (RHD). Here, we present a review of the genetic sus-
and àClinical Immunology and Allergy,
Department of Clinical Medicine, University of ceptibility markers involved in the development of RF ⁄ RHD and the major
São Paulo, School of Medicine, São Paulo, immunopathological events underlying the pathogenesis of RF and RHD. Sev-
Brazil eral human leucocyte antigen (HLA) class II alleles are associated with the dis-
ease. Among these alleles, HLA-DR7 is predominantly observed in different
ethnicities and is associated with the development of valvular lesions in RHD
Received 13 April 2007; Accepted in revised
patients. Cardiac myosin is one of the major autoantigens involved in rheuma-
form 19 May 2007
tic heart lesions and several peptides from the LMM (light meromyosin) region
Correspondence to: L. Guilherme, Laboratório were recognized by peripheral and intralesional T-cell clones from RF and
de Imunologia, Instituto do Coração (InCor), RHD patients. The production of TNF-a and IFN-c from heart-infiltrating
HC-FMUSP. Av. Dr Eneas de Carvalho Aguiar, mononuclear cells suggests that Th-1 type cytokines are the mediators of
44 - 9 andar. 05403-000 São Paulo, SP, Brazil.
RHD heart lesions while the presence of few interleukin-4 producing cells in
E-mail: luizagui@[Link]
the valve tissue contributes to the maintenance and progression of the valvular
lesions.

giation are the most common events caused by valvulitis

Introduction
leading to chronic RHD. RF ⁄ RHD is still a major public
Rheumatic fever (RF) is a delayed sequel to throat infec- health burden in developing countries, leading to
tion by Streptococcus pyogenes and affects susceptible 233,000 deaths annually [2].
untreated children. Based on the major criteria established The incidence of RHD in the world is at least 15.6
by Jones and revised by the American Heart Association, million cases and the highest documented prevalence of
[1] the disease manifests as polyarthritis, carditis, chorea, the disease among children from developing countries is
erythema marginatum and ⁄ or subcutaneous nodules. 5.7 per 1000 in sub-Saharan Africa [2]. In Brazil, the
Nearly 75% of affected children display arthritis and incidence of ARF was 20,000 new cases in 1992; how-
30–45% develop carditis, which causes heart damage with ever, in the last 10 years it decreased by 75% but was
pericardial, myocardial and endocardial involvement fol- still high, reaching 5000 new cases in 2002 (data from
lowed by progressive and permanent valvular lesions lead- the Brazilian Health Ministry).
ing to rheumatic heart disease (RHD). Sydenham’s chorea The pathogenesis of RF ⁄ RHD is complex and both
is characterized by involuntary movements, especially, of environmental and genetic factors contribute to its aetiol-
the face and limbs, muscular weakness, and disturbances of ogy. The manifestation of the disease in only a small sub-
speech, gait and voluntary movements. Children usually set of children untreated for strep throat and also the fact
exhibit concomitant psychological dysfunction, especially, that only one-third of the affected children progress to
obsessive-compulsive disorder, increased emotional labil- the development of RHD suggests the involvement of
ity, hyperactivity, irritability and age-regressed behaviour. host genetic factors. Furthermore, familial clustering
It is usually a delayed manifestation, and is often the sole and high concordance of RF ⁄ RHD among monozygous
manifestation of acute rheumatic fever (ARF) [1]. twins provide conclusive evidence for the presence of
Life-threatening complications from RHD include genetic determinants of susceptibility to RF ⁄ RHD.
valvulitis. Valvular lesions and mitral and aortic regur- Nonetheless, the presence of different clinical manifestations

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207 199
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
200 Rheumatic Fever and Rheumatic Heart Disease L. Guilherme et al.
..................................................................................................................................................................

of RF ⁄ RHD may also reflect extensive genetic hetero- presented a B-cell alloantigen designated 883 [20]. The
geneity. alloantigen 883 was suggested to be related to the HLA
In this review, we will highlight the genetic influen- class II molecules located in the HLA-DR region of the
ces involved in the pathogenesis of RF ⁄ RHD as well the MHC. A monoclonal antibody was produced against B
immunological mechanisms involved in the development cells from patients harbouring the 883 alloantigen; how-
of rheumatic heart lesions. ever, this antibody recognizes another antigen named
D8 ⁄ 17, which is expressed on the surface of 10–20% of
B cells of RF patients. The D8 ⁄ 17 antigen was identified
Family studies in RF ⁄ RHD
in 90–95% of RF patients from different regions [21].
The observation that RF clusters in families by Cheadle From this followed an investigation of HLA class II
in 1889 led to the search for the mode of inheritance in molecules in different populations. The presence of an
the early 20th century and later, for the host determi- association between HLA class II alleles and RF ⁄ RHD
nants of susceptibility to the disease. Several studies sug- was first described by Ayoub et al. [22], who found an
gested a simple recessive model of distribution of cases association between HLA-DR2 in African-American
among children in rheumatic families [3, 4] while other patients and HLA-DR4 in Caucasian-American patients
studies revealed that there is no clear mode of inheritance [22]. The association with DR4 in Caucasian-American
[5, 6]. Despite the controversy in the mode of inheritance, patients was confirmed by Anastasiou-Nana [23].
these studies provide evidence that RF ⁄ RHD occurs in HLA-DR4 was also found in Saudi Arab patients [24]
genetically predisposed individuals. A twin study revealed (Table 1). The most consistent HLA class II allele
a higher concordance rate (18.7%) in monozygotic than associated with RF ⁄ RHD is HLA-DR7. In Brazilian-
in dizygotic (2.5%) twins, suggesting very low Mullatos patients, HLA-DR7 and HLA-DRw53 were
penetrance of RF [7]. Also, the similarity of clinical mani- revealed to be markers for susceptibility to RF and RHD
festations in siblings of patients with RF was observed to [25, 26]. In Brazilian-Caucasians, HLA-DR7 was further
be higher than would be expected by chance [7]. confirmed [27] (Table 1). It is important to note that
HLA-DRw53 is of broad specificity and is associated
with HLA-DR4, -DR7, and, -DR9. HLA-DR7 was also
Genetic influence of the major histocompatibilty
found in other populations, such as Turkish [28, 29],
complex
Latvian (in which the association of HLA-DRB1*07 and
In an effort to identify gene(s) responsible for the devel- DQB1*0302 or DQB1*0401-2 was also associated with
opment of RF, many studies were based on the presence the development of valvular lesions [30]), and also in
or absence of selected blood groups (ABO) and the secre- RHD patients from Northern Egypt with severe mitral
tor status (ABH) of patients with RF [8–10]. Several valve disease [31] (Table 1). Other HLA class II alleles
studies have sought an association of RF ⁄ RHD with (HLA-DR and HLA-DQ) were also found in association
either human leucocyte antigen (HLA)-A or -B loci of with RF ⁄ RHD in different populations [32–39] as shown
the major histocompatibility complex (MHC) [11–17]. in Table 1. The strength of correlation of HLA class II
Controversial results have been obtained with some find- alleles ranges from 2.3 times to 13.6 (RR = 2.3–13.6)
ing a borderline association. The major role of HLA class (Table 1) indicating that these alleles play a key role in
I antigens is to present intracellular antigens (self and the development of the disease. The fact that different
intracellular pathogen antigens) to the T-cell receptor class II alleles were found associated with the disease
(TCR). RF ⁄ RHD, as mentioned before, is a consequence could be due to the following aspects: (1) ethnic differ-
of untreated infection caused by S. pyogenes. Extracellular ences of the populations studied; (2) selection of RF
antigens mainly use MHC class II antigens to activate patients with defined clinical manifestation of the disease;
the adaptive immune response. This may be one explan- (3) methodology of the study (serological or molecular
ation for the absence of HLA class I association with biology approach); (4) size of RF ⁄ RHD patient and
RF ⁄ RHD. It is interesting to note that in 1975, HLA control samples; or (5) specific streptococcal strains in
class II was not known, but an association with HLA-B5 different regions.
and an increased response in vitro to streptococcal anti-
gens [18] were found. A significantly more pronounced
Genetic influence of non-MHC
immune response by RF ⁄ RHD patients as measured by
circulating immune complexes [19] was also found. In Considering that both innate and adaptive immune
both cases, the reactivity was probably due to an associ- responses are involved with the development of RF ⁄ RHD,
ation with HLA class II. non-MHC genes probably increase the susceptibility
Strong support for this idea came from studies by to RF in combination with the HLA molecules. Identi-
Patarroyo et al. [20]. They observed that 72% of patients fying these genes will be informative in understanding
with RF from Bogotá, Colombia or New York, USA the pathogenesis of the disease.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
L. Guilherme et al. Rheumatic Fever and Rheumatic Heart Disease 201
..................................................................................................................................................................

Table 1 HLA class II and RF ⁄ RHD.

Country Population HLA risk Clinical picture of disease RR Ref.

South África African DR1, DR6 RF ⁄ RHD 5.2 ⁄ 2.6 33

Martinique Admixed DR1 RHD – 32
Brazil Mulatto DR1 Sydenham’s chorea – 34
USA African-American DR2 RF ⁄ RHD 3.86 22
Índia Indian DR3, DQW2a RF ⁄ RHD 3.76a 35, 36
USA Caucasian-American DR4 RF ⁄ RHD 3.55 22
USA Caucasian-American DR4 RF ⁄ RHD 2.3 23
Saudi Arábia Arabians DR4 RF ⁄ RHD 13.6 24
Turkey Turkish DR11 RF ⁄ RHD – 37
Turkey Turkish DR3, DR7 RHD – 28
Brazil Mulatto DR7, DR53 RF ⁄ RHD 3.8 ⁄ 4.2 25
Brazil Mulatto Allogenotope TaqI DRb 13.81 kbb RF ⁄ RHD – 26
Brazil Caucasian DR7 RF ⁄ RHD 2.4 27
Egypt Egyptian DRB1*0701 DQ A1*0201 RHD–MVR 3.0 31
DRB1*13 DQA1*0501, DQ A1*0301 RHD –
Latvia Latvian DRB1*0701; DQB1*0302 RF ⁄ RHD–MVL 4.18 ⁄ 3.13 30
DRB1*0701; DQB1*0401 RF ⁄ MVR, Syndeham’s chorea 4.18 ⁄ 4.33
Turkey Turkish DRB1*07 RHD 2.78 29
Japan Japanese DQA1*0104, DQB1*05031 RHD – Mitral stenosis 2.8 ⁄ 3.23 39
Mexico Mestizo DRB1*1602, DQA1*0501,DQB1*0301 RHD 5.3 38

HLA, human leucocyte antigen; RF, rheumatic fever; RHD, rheumatic heart disease; MVL, multivalvular lesions; MVR, mitral valve regurgitation;
RR, relative risk.
Studies from Egypt, Japan, Latvia and Mexico employed molecular methods (PCR).
a
DQw2 RR value.
b
Defined by RFLP study, 13.81 kb fragment corresponds to the HLA-DR53 antigen.
(–), not done.

Mannose binding lectin (MBL) is an acute phase observe any association and the frequency of the
inflammatory protein and functions as a soluble pathogen mutated allele is very low in the population (R. Rama-
recognition receptor. MBL binds to a wide variety of sawmy, K. C. Faé, A. C. Tanaka, G. Spina, A. C. Goldberg,
sugars on the surface of pathogens and plays a major role J. Kalil, L. Guilherme, unpublished observations).
in innate immunity due to its ability to opsonize patho- There are many genes with inflammatory function and
gens, enhancing their phagocytosis and activating the one of these is the interleukin (IL)-1 gene cluster that is
complement cascade via the lectin pathway [40]. MBL is located on chromosome 2 and includes the genes expres-
encoded by the gene MBL2 located on the chromosome sing the proinflammatory cytokines IL-1a and IL-1b and
10q11.1-q21 region. Mutations in exon 1 of the MBL2 their inhibitor IL-1 receptor antagonist (IL-1RA). One
gene correlate with deficient MBL in the plasma and have study from Taiwan reported that variations in IL-1b and
been shown to be associated with recurrent infections in IL-1RA are not associated with RHD [45]. Further studies
children and several infectious diseases [40]. N-acetylg- are warranted in other populations with larger sample size
lucosamine, present in the cell wall of streptococcus, is a before excluding IL-1 as a susceptible or protective factor.
strong ligand for MBL. Recently, one study reported that Tumour necrosis factor A(TNFA) is another gene with
genotypes correlated with high MBL were associated with inflammatory function which is situated in the MHC class
RHD [41]. III regions. Interestingly, three independent studies have
Toll-like receptors are also pathogen recognition shown an association of TNFA polymorphism with
receptors that sense invading pathogens to initiate innate RF ⁄ RHD. In the Brazilian population, the presence of
immune responses [42]. The search for polymorphisms in either one of the TNFA alleles ()308A and )238A) was
TLR related to bacterial infections is of interest in associated with the development of RF and RHD and this
RF ⁄ RHD. Berdeli et al. [43] reported a very strong associ- association was stronger in patients with aortic valve
ation of a non-synonymous polymorphism Arg753Gln lesions [46]. In Turkish and Mexican populations, the
in the gene encoding TLR2 with RF among Turkish TNFA )308A allele was associated with RF ⁄ RHD [47,
children. However, a second study in this population 48]. Of note, one study did not confirm the association in
could not confirm this association [44]. The discrepancy the Turkish population [49]. Thus, variants of TNFA may
between the two studies may be due to an error in be one of the predisposing risk factors for RF ⁄ RHD, which
genotyping or it is a reflection of distinct populations. is likely to act in synergy with other factors, both genetic
Among Brazilian patients with RF ⁄ RHD, we did not and environmental, in the development of the disease.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
202 Rheumatic Fever and Rheumatic Heart Disease L. Guilherme et al.
..................................................................................................................................................................

The molecular mechanism underlying the association the cytokines IL-1, IL-6 and TNF-a should be produced
of TNFA )308A ⁄ G with susceptibility to RF ⁄ RHD is to eliminate the bacteria. All these proteins are genetic-
not known. TNFA )308A was reported to be associated ally controlled and in RF ⁄ RHD patients, alterations or
with high TNF-a production [50]. TNF increases the mutations can lead to differential expression and ⁄ or
synthesis of proinflammatory cytokines and also leads to secretion, consequently causing damage. MBL binds to
the activation of NFjB which subsequently activates the the N-acetyl b D-glucosamine in the streptococcal cell
transcription of proinflammatory genes. Several lines of wall, activating the complement-lectin pathway that
evidence indicate a key role for TNF-a in the immuno- induces the clearance of the bacteria. Some mutations in
pathogenesis of RF ⁄ RHD as presented below in the path- the gene that codes for MBL production present in RF
ogenesis mechanisms section. and RHD patients probably interfere in the clearance of
Several other candidate genes have been investigated S. pyogenes (R. Ramasawmy, G. Spina, K. C. Faé, A. C.
with negative or positive findings, including angiotensin- Pereira, R. Nisihara, I. Messias Reason, M. Grinberg,
converting enzyme [51–53], Fcgamma RIIA and F. Tarasoutchi, J. Kalil, L. Guilherme, unpublished
Fcgamma RIIIB [54] and TGF-b [55]. Further follow-up observations). In addition, RF ⁄ RHD patients presented
studies will be needed with much larger sample sizes to increased levels of TNF-a in the plasma [58–60] and it
replicate the findings. was demonstrated that stimulation of peripheral blood
Overall, the identification of the HLA-DR molecules mononuclear cells from children with ARF with strepto-
associated with RF ⁄ RHD led to the elucidation, in part, coccal antigen produced higher levels of TNF-a when
of the mechanism underlying the pathogenesis of the dis- compared with controls [61]. IL-1 is also reported to
ease, which will be discussed in the pathogenesis mech- increase during active rheumatic carditis [62].
anism section. Currently, only a few genes for RF ⁄ RHD
have been identified through candidate gene-association
Molecular mimicry and autoimmune reactions
studies. Future studies will need to focus on immune
response gene pathways to unravel the breaking of toler- Molecular mimicry is a mechanism by which host and
ance in these patients to tip them towards the develop- pathogen antigens that exhibit some degree of homology
ment of autoimmunity. are recognized through cross-reactivity by both T and B
lymphocytes. The antibodies secreted by B cells recognize
the conformations of proteins and other antigen mole-
Pathogenic mechanisms
cules. T cells recognize peptide fragments combined with
Rheumatic fever or rheumatic heart disease is the most MHC molecules.
convincing example of molecular mimicry in human The pathogenesis of RF ⁄ RHD seems to result from an
pathological autoimmunity, given the cross-reactions overt immune response involving either humoral or cellu-
between streptococcal antigens and human tissue pro- lar reaction or both, triggered by group-A streptococci
teins, mainly heart tissue proteins, which follow throat infection. The concept of an involvement of autoimmune
infection by S. pyogenes in susceptible individuals. reactions in the pathogenesis of RF was introduced only
The S. pyogenes, or group A streptococci (GAS), cell in the 1960s by Kaplan who demonstrated that antibod-
wall consists of carbohydrates such as N-acetyl b D-gluco- ies against GAS reacted with human heart preparations
samine linked to a polymeric rhamnose backbone. Group [63, 64]. Subsequent studies conducted by Zabriskie
A streptococci contain M, T and R surface proteins and et al. gave support to the hypothesis that RF has an auto-
lipoteichoic acid (LTA), all of which are involved in immune origin by describing the presence of antibodies
bacterial adherence to throat epithelial cells. The M that were cross-reactive with streptococcal membrane
protein, which extends from the cell wall, is composed of antigens in acute RF sera [65, 66]. Goldstein and col-
two polypeptide chains with approximately 450 amino leagues showed that antibodies to the N-acetylglucosa-
acid residues, in an a-helical coiled-coil configuration. mine carbohydrate cross-reacted with glycoproteins
The amino-terminal (N-terminal) portion presents anti- present in the heart valves that contain N-acetylglucosa-
genic variations but is highly homologous, with the mine [67]. Cunningham’s group has shown that human
exception of the first 11 amino acid residues that define monoclonal antibodies also react with N-acetylglucosa-
the different serotypes [56], of which 200 have been iden- mine, cardiac myosin and laminin. They have also dem-
tified to date. The M protein is the most important anti- onstrated the in vitro cytotoxic activity of human and
genic structure of the bacteria and shares structural murine monoclonal antibodies [57]. Anti-M protein anti-
homology with a-helical coiled-coil human proteins like bodies were shown to cross-react with vimentin and car-
cardiac myosin, tropomyosin, keratin, laminin, vimentin diac myosin, suggesting that these proteins were the
and several valvular proteins [reviewed 57]. target autoantigens recognized in the heart [68–72].
During the acute phase of the throat infection, Using anti-myosin antibodies purified by affinity from
inflammatory acute phase proteins, such as MBL, and ARF patient sera, the same group identified cross-reactive

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
L. Guilherme et al. Rheumatic Fever and Rheumatic Heart Disease 203
..................................................................................................................................................................

epitopes on myosin and the M5 ⁄ M6 proteins [73]. In Table 2 M protein peptides recognized by human and murine T cells.
addition, they demonstrated the potential role of these
Amino acid residues
cross-reactive antibodies in the development of RHD, by
showing that they are able to bind to the endothelial sur- M5(1–35) AVTRGTINDPQRAKEALDKYELENHDLKTKNEGLKa
face, which may lead to inflammation, cellular infiltration M5(1–25) TVTRGTISDPQRAKEALDKYELENHb
and valve scarring [74]. The upregulation of the adhesion NT5(59–76) KKEHEAENDKLKQQRDTLa
NT6(72–89) QRDTLSTQKETLEREVQNa
molecule VCAM-1 after binding of cross-reactive anti- M5(81–96) DKLKQQRDTLSTQKETb
bodies to the valvular endothelium facilitates cellular M5(83–103) LKQQRDTLSTQKETLEREVQNb
infiltration [75] (Fig. 2). M5(91–103) STQKETLEREVQNb
The first evidence of CD4+ T-cell involvement in
NT5 ⁄ NT6 myosin cross-reactive peptides described by Cunningham
RHD lesions was described by Raizada et al. [76], while et al. [81], and human valvular tissue proteins cross-reactive
their role in the development of heart tissue lesions in M5(1–25), M(81–96), M5(81–103), M5(91–103) peptides described
RHD was described later by us [77]. We succeeded in by Guilherme et al. [77]; M5(1–35) peptide described by Robinson
growing in vitro these heart tissue infiltrating T cells and et al. [82].
through a molecular analysis we demonstrated the Bold typed and underlined regions correspond to the identical residues
among the different peptides.
molecular mimicry between b-haemolytic streptococci a
Murine lymph node T cells.
and heart tissue proteins. By generating T-cell clones b
Human heart infiltrating T-cell clones.
from heart lesions of four severe RHD patients, we dem-
onstrated for the first time the ability of 7.5% of these B1B2, B2, B2B3A, and B3A, were predominantly recog-
cells to simultaneously recognize M protein peptides and nized [81]. NT5 ⁄ 6 and B1B2 ⁄ B2 aligned with the M5
heart tissue-derived proteins. Three M5 regions (residues regions previously identified by our group, namely
1–25, 81–103 and 163–177) were cross-reactive with M5(81–103) and M5(163–177) respectively. Robinson
several heart protein fractions, mainly those derived et al. [82], obtained lymph node T-cell clones from mice
from valvular tissue with molecular masses of 95–150, immunized with recombinant M5 protein, and the
43–65 and 30–43 kDa [77](Fig. 2). The frequencies of M5(1–35) peptide aligned with the M5(1–25) region
autoreactive T cells were higher in RHD patients in recognized by human heart-infiltrating T-cell clones
acute phase of the disease (67%) when compared with (Table 2).
RHD patients in chronic phase (around 25%) [78]. Pep- Cardiac myosin is one of the major autoantigens
tides included in the three immunodominant M5 regions involved in rheumatic heart lesions [57, 71]. Recently, we
described above and heart tissue protein fractions were showed a very high frequency (63.2%) of heart tissue
also recognized by peripheral T cells from RF ⁄ RHD infiltrating T-cell clones recognizing the myosin ß chain,
patients [79]. The M5(81–96) epitope was preferentially the LMM fragment. Thirty-four per cent of T-cell clones
recognized by patients with severe RHD that expressed exhibited cross-reactivity with different patterns such as,
HLA-DR7+DR53+ [79]. These data suggest a role for the (1) myosin and valve-derived proteins; (2) myosin and
HLA class II molecules DR7 and DR53 in presenting streptococcal M5 peptides; and (3) myosin, valve-derived
the streptococcal immunodominant peptide to the TCR. proteins and M5 peptides [83]. The reactivity against
In addition, show the significance of the propensity of LMM peptides of heart tissue infiltrating T cells may be
DR7 and ⁄ or DR53 to cause severe RHD in Brazilian attributable to the stimulation of these cells initially by
patients and probably other populations in which this the a-helical coiled-coil streptococcal M protein. The
class II allele is associated with RHD and with the devel- myosin cross-reactivity with several other valvular pro-
opment of valvular lesions. teins may occur first through mimicry [77, 84] and even-
Yoshinaga et al. [80], isolated T cells from heart valves tually by an epitope spreading mechanism [85] in which
and compared the reactivity of PHA-stimulated T-cell the evolution of an autoimmune pathology may easily
lines derived from heart-valve specimens and peripheral obliterate any evidence of the initial target antigen, such
blood lymphocytes of RF patients and showed that, even as streptococcal M protein and cross-reactive self-antigens.
though these cells recognized cell wall and membrane Cross-reactive peripheral T-cell clones from a RHD
streptococcal antigens, they failed to react with the M patient were responsive to group A recombinant M6 pro-
protein, myosin, or other mammalian cytoskeletal pro- tein, cardiac myosin, tropomyosin, and laminin, a valve
teins. The absence of reactivity against M protein and protein [84]. The authors showed that the cross-reactive
self-antigens probably was a result of low frequencies of response was MHC class I- and II-restricted for T-cell
specific T cells. clones that were CD4+ and CD8+ respectively. These cells
Myosin ⁄ M5 protein cross-reactive T-cell epitopes were recognized streptococcal M6 epitopes on the B region, and
also investigated in mice immunized with intact cardiac S2 and LMM regions of human cardiac myosin [84].
myosin. Lymph node T cells cross-reacted with overlap- Both peripheral and rheumatic heart lesion-derived
ping M5 peptides and seven regions, termed NT4 ⁄ 5 ⁄ 6, T-cell clone studies demonstrate mimicry at the T-cell

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
204 Rheumatic Fever and Rheumatic Heart Disease L. Guilherme et al.
..................................................................................................................................................................

Patient # 1

Vβ13-Cβ Vβ13-Cβ

HEART- LESIONS
Infiltrating T- cell line
10 10
Oligoclonal profile of

PBMC T cell expansion

bearing
Polyclonal
profile TCR-Vß13 Jß2.7

Jß2.7

Figure 1 CDR3 size patterns of a heart tissue expanded T-cell population. Top left, patient no. 1, Vß13 family found in peripheral blood mononu-
clear cells with a polyclonal pattern. Top right, Vß13 family found in the mitral valve-derived T-cell line as an oligoclonal expansion that combines
with some Jß segments. Vß13Jß2S7 – an oligoclonal expansion that corresponds to 61.8% of the expanded T cells bearing the TCRVß13 (Table 3).

level between streptococcal M protein and human cardiac The heart tissue infiltrating T-cell repertoire is probably
myosin epitopes and confirm our previous work reported composed by migrating peripheral primed T cells. We
for T-cell mimicry in the valve. assessed the clonality of T cells by analysing the Vß and Jß
We observed that IFN-c, TNF-a (inflammatory cytok- genes of the TCR in both peripheral blood and intralesion-
ines) and IL-10 (regulatory cytokine) positive cells were al T-cell lines obtained from heart tissue from several
consistently predominant in both myocardium and valvu- RHD patients. The relative frequency of the 22 Vß famil-
lar tissue whereas IL-4 (regulatory cytokine) was scarce in ies in both peripheral and intralesional T-cell lines showed
the valves. The predominance of inflammatory cytokines no particular expansion of any Vß family. The analysis of
in the heart lesions confirms that RHD is mediated by TCR Vß and Jß rearrangement showed the clonal presence
inflammatory immune responses. In addition, the signifi- of expanded families in the heart tissue and a polyclonal
cantly lower IL-4 expression in the valvular tissue may distribution in the peripheral blood [88]. Here, we present
contribute to the progression of RHD leading to perma- some examples of these patterns in Fig. 1 and Table 3. The
nent damage in the valves [86] (Fig. 2). hypothesis that streptococcal primed T cells migrate from

Table 3 Relative frequencies and CDR3 patterns of T cells from the peripheral blood and mitral valve-infiltrating T-cell lines.

Patient no. Origin of T cells ⁄ subsets Vb family CDR3 pattern Jb family CDR3 pattern

1 PBMC Vb 13–4.5% Polyclonal ND

Mi V Vb 13–8.3% Oligoclonal Jb 1.5–11.1% All oligoclonal
87.2% CD4+ Jb 2.7–61.8%
7.4% CD8+ Jb 2.2–24.8%
Jb 2.3–18.8%
2 PBMC Vb 1–2.7% Polyclonal ND
Mi V Vb 1–7.6% Oligoclonal Jb 1.2–17.5% All oligoclonal
91.0% CD4+ Jb 1.6–40.3%
6.0% CD8+ Jb 2.5–11.1%
3 PBMC Vb 5–1.7% Polyclonal ND
Mi V Vb 5–5.3% Oligoclonal Jb 2.3–24.0% All oligoclonal
49.4% CD4+ Jb 2.7–53.0%
38.5% CD8+

PBMC, peripheral blood mononuclear cell.

CDR3 patterns were analysed by the Immunoscope approach [87].
Bold typed, means major expansions.
ND, not done.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
L. Guilherme et al. Rheumatic Fever and Rheumatic Heart Disease 205
..................................................................................................................................................................

Valvular tissue

CD4+ T cells Cross reactive Mononuclear

antibodies cells

•Streptococcal/self antigens cross reactive

antibodies facilitate heart- tissue T cells
infiltration
•CD4+ T cells recognize streptococcal
antigens and heart-tissue proteins by
molecular mimicry
•Inflammatory cytokines (TNF-α IFN-γ) are
produced by mononuclear cells
•Few mononuclear cells produce IL-4
(regulatory cytokine) – leading to the
Surgical specimen of acute RF showing small vegetations progression and maintenance of valvular
on the line of closure of the mitral valve (arrows). lesions.
Macroscopic photo done by Aiello, VD
Heart Institute (InCor).

Figure 2 Major events triggering rheumatic valvular lesions in RHD. Heart tissue cross-reactive antibodies bind to the surface of the valve endot-
helial and facilitate mainly CD4+ T-cell infiltration [75–77]. Streptococcal primed T cells trigger an autoimmune reaction and produce inflammatory
cytokines. Low numbers of regulatory IL4+ cells are found in the valvular tissue [88]. The permanent ongoing inflammation leads to the valvular
lesions such as the small vegetations.

the periphery to the heart was confirmed by the identifica- leading to RHD valve lesions, such as the small vegeta-
tion of some M protein ⁄ heart tissue cross-reactive intrale- tions shown on the left side of the picture. Streptococcal
sional CD4+ T-cell clones. As an example, an intralesional and host antigen cross-reactive antibodies facilitate heart
T-cell clone that recognized two cardiac myosin peptides tissue infiltration by T lymphocytes. CD4+ T cells are
(LMM39 and LMM41, amino acid residues 1790–1807 the major effectors of the heart tissue, leading to RHD
and 1816–1833 respectively) and a mitral valve-derived lesions. Inflammatory cytokines such as, TNF-a and IFN-
protein with a molecular mass of 56–53 kDa and isoelec- c are the mediators of valve lesions and the presence of
tric point of 6.76 displayed the TCR-Vß13Jß2S7 [80]. few mononuclear cells producing IL-4, a regulatory cyto-
Taken together, these results demonstrate that T-cell kine, leads to the progression and maintenance of valve
populations are expanded in heart lesions, driven by self- lesions in RHD patients.
antigen recognition, and probably contribute to the main-
tenance and progression of tissue damage.
Acknowledgment
We acknowledge Dr Vera Demarchi Aiello from the
Concluding remarks
Heart Institute (InCor) for providing us with the macro-
All the findings described in the last 50 years about the scopic photo of ARF valvulitis.
immune response leading to RF and RHD have led to a
significant understanding of the pathogenesis of RF and
References
RHD. The association of HLA class II alleles, considered as
genetic markers for RF ⁄ RHD, in fact pointed out HLA 1 Dajani AS, Ayoub E, Bierman FZ et al. Guidelines for the diagnosis
molecules that preferentially present streptococcal and self- of Rheumatic Fever: Jones criteria, 1992 uptade. JAMA
1992;268:2069–73.
antigens to the TCR. The gene TNFA, located on the same 2 Carapetis JR, Steer AC, Mulholland EK, Weber M. The global burden
chromosome as the HLA genes, carrying a mutation in the of group A streptococcal diseases. Lancet Infect Dis 2005;5:685–94.
promoter region leading to the increased production of 3 Wilson MG, Schweitzer MD. Pattern of hereditary susceptibility in
TNF-a, an inflammatory cytokine, is associated with the rheumatic fever. Circulation 1954;10:699–704.
disease. Other genes related to immune regulation are now 4 Gray FG, Quinn RW, Quinn JP. A long term survey of rheumatic
and non-rheumatic families; with particular reference to environ-
being investigated and will certainly contribute to the defi- ment and heredity. Am J Med 1952;13:400–12.
nition of new genetic markers and an understanding of 5 Uchida IA. Possible genetic factors in the etiology of rheumatic
how these genes act in the complex network of autoim- fever. Am J Hum Genet 1953;5:61–9.
mune reactions that occur in RF ⁄ RHD. 6 Stevenson AC, Cheeseman EA. Heredity and rheumatic fever; a
Both B and T cells play important role in the devel- study of 462 families ascertained by an affected child and 51 famil-
ies ascertained by an affected mother. Ann Eugen 1953;17:177–210.
opment of RF. In Fig. 2, we summarize the major events

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
206 Rheumatic Fever and Rheumatic Heart Disease L. Guilherme et al.
..................................................................................................................................................................

7 Spagnuolo M, Taranta A. Rheumatic fever in siblings. Similarity of 28 Ozkan M, Carin M, Sonmez G, Senocak M, Ozdemir M, Yakut C.
its clinical manifestations. N Engl J Med 1968;278:183–8. HLA antigens in Turkish race with rheumatic heart disease. Circula-
8 Clarke CA, McConnell RB, Sheppard PM. ABO blood groups and tion 1993;87:1974–8.
secretor character in rheumatic carditis. Br Med J 1960;1:21–3. 29 Haydardedeoglu FE, Tutkak H, Kose K, Duzgun N. Genetic sus-
9 Glynn LE, Holborow EJ. Relation between blood groups, secretor ceptibility to rheumatic heart disease and streptococcal pharyngitis:
status and susceptibility to rheumatic fever. Arthritis Rheum association with HLA-DR alleles. Tissue Antigens 2006;68:293–6.
1961;4:203–7. 30 Guedez Y, Kotby A, El-Demellawy M et al. HLA class II associations
10 Buckwalter JA, Naifeh GS, Auer JE. Rheumatic fever and the blood with rheumatic heart disease are more evident and consistent among
groups. Br Med J 1962;2:1023–7. clinically homogeneous patients. Circulation 1999;99:2784–90.
11 Falk JA, Fleischman JL, Zabriskie JB, Falk RE. A study of HL-A 31 Stanevicha V, Eglite J, Sochnevs A, Gardovska D, Zavadska D,
antigen phenotype in rheumatic fever and rheumatic heart disease Shantere R. HLA class II associations with rheumatic heart disease
patients. Tissue Antigens 1973;3:173–8. among clinically homogeneous patients in children in Latvia. Arth-
12 Caughey DE, Douglas R, Wilson W, Hassall IB. HL-A antigens in ritis Res Ther 2003;5:R340–6.
Europeans and Maoris with rheumatic fever and rheumatic heart 32 Monplaisir N, Valette I, Bach JF. HLA antigens in 88 cases of
disease. J Rheumatol 1975;2:319–22. rheumatic fever observed in Martinique. Tissue Antigens
13 Joysey VC, Roger JH, Ashworth F et al. Parallel studies of HLA 1986;28:209–13.
antigens in patients with rheumatic heart disease and scleritis: com- 33 Maharaj B, Hammond MG, Appadoo B, Leary WP, Pudifin DJ.
parisons with three control populations. J Rheumatol Suppl HLA-A, B, DR, and DQ antigens in black patients with severe
1977;3:84–8. chronic rheumatic heart disease. Circulation 1987;76:259–61.
14 Leirisalo M, Kovuranta P, Laitinen O. Rheumatic fever and its 34 Donadi EA, Smith AG, Louzada-Junior P, Voltarelli JC, Nepom
sequels in children. A follow-up study with HLA analysis. J Rheu- GT. HLA class I and class II profiles of patients presenting RF with
matol 1980;7:506–14. Sydenham’s chorea. J Neurol 2000;247:122–8.
15 Read SE, Reid H, Poon-King T, Fischetti VA, Zabriskie JB, 35 Jhinghan B, Mehra NK, Reddy KS, Taneja V, Vaidya MC, Bhatia
Rapaport FT. HLA and predisposition to the nonsuppurative ML. HLA, blood groups and secretor status in patients with estab-
sequelae of group A streptococcal infections. Transplant Proc lished rheumatic fever and rheumatic heart disease. Tissue Antigens
1977;9:543–6. 1986;27:172–8.
16 Murray GC, Montiel MM, Persellin RH. Distribution of HLA anti- 36 Taneja V, Mehra NK, Reddy KS et al. HLA-DR ⁄ DQ antigens and
gens in a Mexican–American population and a comparison with two reactivity to B cell alloantigen D8 ⁄ 17 in Indian patients with rheu-
similar populations from different geographical locales. Tissue Anti- matic heart disease. Circulation 1989;80:335–40.
gens 1978;12:337–40. 37 Olmez U, Turgay M, Ozenirler S et al. Association of HLA class I
17 Matsumori A, Hirose K, Wakabayashi A et al. HLA in hypertrophic and class II antigens with rheumatic fever in a Turkish population.
cardiomyopathy and rheumatic heart disease. Jpn Circ J Scand J Rheumatol 1993;22:49–52.
1979;43:445–9. 38 Hernandez-Pacheco G, Aguilar-Garcia J, Flores-Dominguez C et al.
18 Greenberg LJ, Gray ED, Yunis EJ. Association of HL-A 5 and MHC class II alleles in Mexican patients with rheumatic heart dis-
immune responsiveness in vitro to streptococcal antigens. J Exp Med ease. Int J Cardiol 2003;92:49–54.
1975;141:935–43. 39 Koyanagi T, Koga Y, Nishi H et al. DNA typing of HLA class II
19 Yoshinoya S, Pope RM. Detection of immune complexes in acute genes in Japanese patients with rheumatic heart disease. J Mol Cell
rheumatic fever and their relationship to HLA-B5. J Clin Invest Cardiol 1996;28:1349–53.
1980;65:136–45. 40 Jack DL, Klein NJ, Turner MW. Mannose-binding lectin targeting
20 Patarroyo ME, Winchester RJ, Vejerano A et al. Association of a the microbial world for complement attack and opsonophagocytosis.
B-cell alloantigen with susceptibility to rheumatic fever. Nature Immunol Rev 2001;180:86–99.
1979;278:173–4. 41 Messias Reason IJ, Schafranski MD, Jensenius JC, Steffensen R. The
21 Zabriskie JB, Lavenchy D, Willians RC Jr et al. Rheumatic fever- association between mannose-binding lectin gene polymorphism
associated B cell alloantigen as identified by monoclonal antibodies. and rheumatic heart disease. Hum Immunol 2006;67:991–8.
Arthritis Rheum 1985;28:1047–51. 42 Medzhitov R. Toll-like receptors and innate immunity. Nat Rev
22 Ayoub EM, Barrett DJ, Maclaren NK, Krischer JP. Association of Immunol 2001;1:135–45.
class II human histocompatibility leukocyte antigens with rheuma- 43 Berdeli A, Celik HA, Ozyurek R, Dogrusoz B, Aydin HH. TLR-2
tic fever. J Clin Invest 1986;77:2019–26. gene Arg753Gln polymorphism is strongly associated with acute
23 Anastasiou-Nana MI, Anderson JL, Carlquist JF, Nanas JN. HLA- rheumatic fever in children. J Mol Med 2005;83:535–41.
DR typing and lymphocyte subset evaluation in rheumatic heart 44 Duzgun N, Duman T, Haydardedeoglu FE, Tutkak H. The lack of
disease: a search for immune response factors. Am Heart J genetic association of the Toll-like receptor 2 (TLR2) Arg753Gln
1986;112:992–7. and Arg677Trp polymorphisms with rheumatic heart disease. Clin
24 Rajapakse CN, Halim K, Al-Orainey I, Al-Nozha M, Al-Aska AK. Rheumatol 2007;26:915–9.
A genetic marker for rheumatic heart disease. Br Heart J 45 Chou HT, Tsai CH, Chen WC, Tsai FJ. Lack of association of gen-
1987;58:659–62. etic polymorphisms in the interleukin-1beta, interleukin-1 receptor
25 Guilherme L, Weidebach W, Kiss MH, Snitcowsky R, Kalil J. antagonist, interleukin-4, and interleukin-10 genes with risk of
Association of human leukocyte class II antigens with rheumatic rheumatic heart disease in Taiwan Chinese. Int Heart J 2005;
fever or rheumatic heart disease in a Brazilian population. Circula- 46:397–406.
tion 1991;83:1995–8. 46 Ramasawmy R, Fae KC, Spina G et al. Association of polymor-
26 Weidebach W, Goldberg AC, Chiarella J et al. HLA class II anti- phisms within the promoter region of the tumor necrosis factor-
gens in Rheumatic Fever: analysis of the DR locus by RFLP and alpha with clinical outcomes of rheumatic fever. Mol Immunol
Oligotyping. Hum Immunol 1994;40:253–8. 2007;44:1873–8.
27 Visentainer JE, Pereira FC, Dalalio MM, Tsuneto LT, Donadio PR, 47 Sallakci N, Akcurin G, Koksoy S et al. TNF-alpha G-308A poly-
Moliterno RA. Association of HLA-DR7 with rheumatic fever in morphism is associated with rheumatic fever and correlates with
the Brazilian population. J Rheumatol 2000;27(6):1518–20. increased TNF-alpha production. J Autoimmun 2005;25:50–4.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207
 13653083, 2007, 2-3, Downloaded from [Link] by Cochrane Philippines, Wiley Online Library on [04/03/2024]. See the Terms and Conditions ([Link] on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
L. Guilherme et al. Rheumatic Fever and Rheumatic Heart Disease 207
..................................................................................................................................................................

48 Hernandez-Pacheco G, Flores-Dominguez C, Rodriguez-Perez JM 68 Baird RW, Bronze MS, Kraus W, Hill HR, Veasey LG, Dale JB.
et al. Tumor necrosis factor-alpha promoter polymorphisms in Epitopes of group A streptococcal M protein shared with antigens
Mexican patients with rheumatic heart disease. J Autoimmun of articular cartilage and synovium. J Immunol 1991;146:3132–7.
2003;21:59–63. 69 Cunningham MW, Hall NK, Krisher KK, Spanier AM. A study of
49 Berdeli A, Tabel Y, Celik HA, Ozyurek R, Dogrusoz B, Aydin anti-group A streptococcal monoclonal antibodies cross-reactive with
HH. Lack of association between TNFalpha gene polymorphism at myosin. J Immunol 1986;136:293–8.
position -308 and risk of acute rheumatic fever in Turkish patients. 70 Cunningham MW, Swerlick RA. Polyspecificity of antistreptococcal
Scand J Rheumatol 2006;35:44–7. murine monoclonal antibodies and their implications in autoimmu-
50 Kroeger KM, Carville KS, Abraham LJ. The )308 tumor necrosis nity. J Exp Med 1986;164:998–1012.
factor-alpha promoter polymorphism effects transcription. Mol 71 Krisher K, Cunningham MW. Myosin: a link between streptococci
Immunol 1997;34:391–9. and heart. Science 1985;227:413–5.
51 Davutoglu V, Nacak M. Influence of angiotensin-converting enzyme 72 Dale JB, Beachey EH. Epitopes of streptococcal M proteins shared
gene insertion ⁄ deletion polymorphism on rheumatic valve involve- with cardiac myosin. J Exp Med 1985;62:583–91.
ment, valve severity and subsequent valve calcification. J Heart Valve 73 Cunningham MW, McCormack JM, Fenderson PG, Ho MK, Beac-
Dis 2005;14:277–81. hey EH, Dale JB. Human and murine antibodies cross-reactive with
52 Chou HT, Tsai CH, Tsai FJ. Association between angiotensin streptococcal M protein and myosin recognize the sequence GLN-
I-converting enzyme gene insertion ⁄ deletion polymorphism and risk LYS-SER-LYS-GLN in M protein. J Immunol 1989;143:2677–83.
of rheumatic heart disease. Jpn Heart J 2004;45:949–57. 74 Galvin JE, Hemric ME, Ward K, Cunningham M. Cytotoxic mAb
53 Ozisik K, Emir M, Ulus AT et al. The renin-angiotensin system from rheumatic carditis recognizes heart valves and laminin. J Clin
genetic polymorphisms and rheumatic mitral valve disease. J Heart Invest 2000;106:17–24.
Valve Dis 2004;13:33–7. 75 Roberts S, Kosanke S, Dunn TS, Jankelow D, Duran CMG, Cun-
54 Berdeli A, Celik HA, Ozyurek R, Aydin HH. Involvement of ningham MW. Pathogenic mechanism in rheumatic carditis: focus
immunoglobulin FcgammaRIIA and FcgammaRIIIB gene poly- on valvular endothelium. J Infect Dis 2001;183:507–11.
morphisms in susceptibility to rheumatic fever. Clin Biochem 76 Raizada V, Williams RC Jr, Chopra P et al. Tissue distribution of
2004;37:925–9. lymphocytes in rheumatic heart valves as defined by monoclonal
55 Chou HT, Chen CH, Tsai CH, Tsai FJ. Association between anti-T cells antibodies. Am J Med 1983;74:90–6.
transforming growth factor-beta1 gene C-509T and T869C poly- 77 Guilherme L, Cunha-Neto E, Coelho V. Human heart-infiltrating
morphisms and rheumatic heart disease. Am Heart J 2004;48:181– T-cell clones from rheumatic heart disease patients recognized both
6. streptococcal and cardiac proteins. Circulation 1995;92:415–20.
56 Fischetti VA. Streptococcal M protein. Sci Am 1991;264:58–65. 78 Guilherme L, Cunha-Neto E, Tanaka AC, Dulphy N, Toubert A,
57 Cunningham MW. Pathogenesis of group A streptococcal infections. Kalil J. Heart-directed autoimmunity: the case of rheumatic fever.
Clin Microbiol Rev 2000;13:470–511. J Autoimmun 2001;16:363–7.
58 Narin N, Kutukculer N, Ozyurek R, Bakiler AR, Parlar A, Arcasoy 79 Guilherme L, Oshiro SE, Fae KC et al. T cell reactivty against
M. Lymphocyte subsets and plasma IL-1 alpha, IL-2, and TNF- streptococcal antigens in the periphery mirrors reactivity of heart-
alpha concentrations in acute rheumatic fever and chronic rheumatic infiltrating T lymphocytes in rheumatic heart disease patients. Infect
heart disease. Clin Immunol Immunopathol 1995;77:172–6. Immun 2001;69:5345–51.
59 Samsonov MY, Tilz GP, Pisklakov VP. Serum-soluble receptors for 80 Yoshinaga M, Figueiroa F, Wahid MR, Marcus RH, Suh E, Zabriskie
tumor necrosis factor-a and interleukin-2 and neopterin in acute JB. Antigenic specificity of lymphocytes isolated from valvular speci-
rheumatic fever. Clin Immunol Immunopathol 1995;74:31–4. mens of rheumatic fever patients. J Autoimmun 1995;8:601–13.
60 Yegin O, Coskun M, Ertug H. Cytokines in acute rheumatic fever. 81 Cunningham MW, Antone SM, Smart M, Liu R, Kosanke S.
Eur J Pediatr 1997;156:25–9. Molecular analysis of human cardiac myosin-cross-reactive B and
61 Miller LC, Gray ED, Mansour M et al. Cytokines and immunoglob- T-cell epitopes of the group A streptococcal M5 protein. Infect
ulin in rheumatic heart disease: production by blood and tonsillar Immun 1997;65:3913–23.
mononuclear cells. J Rheumatol 1989;16:1436–42. 82 Robinson JH, Atherton MC, Goodacre JÁ, Pinkney M, Weightman
62 Morris K, Mohan C, Wahi PL, Anand IS, Ganguly NK. Enhance- H, Kehoe MA. Mapping T-cell epitopes in group A streptococcal
ment of IL-1, IL-2 production and IL-2 receptor generation in type 5 M protein. Infect Immun 1991;59:4324–31.
patients with acute rheumatic fever and active rheumatic heart dis- 83 Fae KC, Silva DD, Oshiro SE. Mimicry in recognition of cardiac
ease; a prospective study. Clin Exp Immunol 1993;91:429–36. myosin peptides by heart-intralesional T cell clones from rheumatic
63 Kaplan MH, Suchy ML. Immunological relation of streptococcal heart disease. J Immunol 2006;176:5662–70.
and tissue antigens II. Cross-reactions antisera to mammalian heart 84 Ellis NM, Li Y, Hildebrand W, Fischetti VA, Cunningham MW.
tissue with cell wall constituent of certain strains of group A strep- T cell mimicry and epitope specificity of cross-reactive T cell clones
tococci. J Exp Med 1964;119:643–50. from rheumatic heart disease. J Immunol 2005;175:5448–56.
64 Kaplan MH, Svec KH. Immunologic relation of streptococcal anti- 85 Lehmann PV, Forsthuber T, Miller A, Sercarz EE. Spreading of
body cross-reactive with heart tissue: association with streptococcal T-cell autoimmunity to cryptic determinants of an autoantigen.
infection, rheumatic fever and glomerulonephritis. J Exp Med Nature 1992;358:155–7.
1964;19:51–66. 86 Guilherme L, Cury P, Demarchi LM. Rheumatic heart disease:
65 Zabriskie JB. Mimetic relationships between group A streptococci proinflammatory cytokines play a role in the progression and
and mammalian tissues. Adv Immunol 1967;7:147–88. maintenance of valvular lesions. Am J Pathol 2004;165:1583–91.
66 Zabriskie JB, Hsu KC, Seegal BC. Heart-reactive antibody associ- 87 Pannetier C, Even J, Kourilsky P. T-cell repertoire diversity and
ated with rheumatic fever:characterization and diagnostic signifi- clonal expansions in normal and clinical samples. Immunol Today
cance. Clin Exp Immunol 1970;7:147–59. 1995;16:176–81.
67 Goldstein I, Rebeyrotte P, Parlebas J, Halpern B. Isolation from 88 Guilherme L, Dulphy N, Douay C et al. Molecular evidence for
heart valves of glycopeptides which share immunological properties antigen-driven immune responses in cardiac lesions of rheumatic
with Streptococcus haemolyticus group A polysaccharides. Nature heart disease patients. Int Immunol 2000;12:1063–74.
1968;219:866–8.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 66, 199–207