+Model

JVAC 7854 1—11 ARTICLE IN PRESS

Vaccine (2008) xxx, xxx—xxx
1

available at [Link]

journal homepage: [Link]/locate/vaccine

2 Q1 The relative immunogenicity of DNA vaccines

delivered by the intramuscular needle injection,

OF
3

4 electroporation and gene gun methods

Shixia Wang c,d, Chunghua Zhang a,c, Lu Zhang a,c, Jun Li a,b,c,

RO
5

6 Zuhu Huang a,b,c,∗, Shan Lu a,c,d,∗∗

a
7 Jiangsu Province Key Laboratory in Infectious Diseases, The First Affiliated Hospital of Nanjing Medical University,
8 Nanjing 210029, China
DP
b
9 Department of Infectious Diseases, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China
c
10 China-US Vaccine Research Center, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China
d
11 Laboratory of Nucleic Acid Vaccines, Department of Medicine, University of Massachusetts Medical School, Worcester,
12 MA 01605, USA

13 Received 2 January 2008; received in revised form 8 February 2008; accepted 11 February 2008
TE

14
KEYWORDS Summary Immunogenicity of DNA vaccines varies significantly due to many factors including
the inherent immunogenicity of the protein antigen encoded in the DNA vaccine, the optimal
EC

15
DNA vaccine;
16 immune responses that can be achieved in different animal models and in humans with dif-
Antibody;
17 ferent genetic backgrounds and, to a great degree, the delivery methods used to administer
T cell;
18 the DNA vaccines. Based on published results, only the gene gun-mediated delivery approach
B cell;
19 has been able to elicit protective levels of immune responses in healthy, adult volunteers by
Influenza
20 DNA immunization alone without the use of another vaccine modality as a boost [Roy MJ, Wu
RR

21 MS, Barr LJ, et al. Induction of antigen-specific CD8+ T cells, T helper cells, and protective
22 levels of antibody in humans by particle-mediated administration of a hepatitis B virus DNA
23 vaccine. Vaccine 2000;19(November(7—8)):764—78; Drape RJ, Macklin MD, Barr LJ, Jones S,
24 Haynes JR, Dean HJ. Epidermal DNA vaccine for influenza is immunogenic in humans. Vaccine
25 2006 22;24(May(21)):4475—81]. Recent results from animal studies suggest that electropora-
26 tion is also effective in eliciting high level immune responses. However, there have been no
CO

27 reports to identify the similarities and differences between these two leading physical delivery
28 methods for DNA vaccines against infectious disease targets. In the current study, we compared
29 the relative immunogenicity of a DNA vaccine expressing a hemagglutinin (HA) antigen from
30 an H5N1 influenza virus in two animal models (rabbit and mouse) when delivered by either
UN

∗ Corresponding author at: Department of Infectious Diseases, The First Affiliated Hospital of Nanjing Medical University, 300 Guangzhou

Road, Nanjing 210029, China. Tel.: +86 25 83781217.

∗∗ Corresponding author at: Rm 304, Lazare Research Building, University of Massachusetts Medical School, 364 Plantation Street,

Worcester, MA 01605, USA. Tel.: +1 508 856 6791; fax: +1 508 856 6751.
E-mail addresses: zhhuang@[Link] (Z. Huang), [Link]@[Link] (S. Lu).

0264-410X/$ — see front matter © 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/[Link].2008.02.033

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
2 S. Wang et al.

31 intramuscular needle immunization (IM), gene gun (GG) or electroporation (EP). HA-specific anti-
32 body, T cell and B cell responses were analyzed. Our results indicate that, overall, both the GG
33 and EP methods are more immunogenic than the IM method. However, EP and IM stimulated a
34 Th-1 type antibody response and the antibody response to GG was Th-2 dominated. These findings
35 provide important information for the further selection and optimization of DNA vaccine delivery
36 methods for human applications.
37 © 2008 Elsevier Ltd. All rights reserved.

38 Introduction The goal of the current study was to compare the rel- 88

ative immunogenicity between these two leading physical 89

39 DNA immunization was introduced as a new method of delivery methods of DNA vaccines, gene gun or electropo- 90

40 immunization in the early 1990s. Several research groups ration, by using a prototypic DNA vaccine which expresses 91

OF
41 were able to demonstrate that immunization with plasmid a highly immunogenic antigen, the hemagglutinin (HA) anti- 92

42 DNA encoding for a specific protein antigen could elicit an gen from an H5N1 influenza virus. The more conventional 93

43 antigen-specific immune response [3—8]. While this vac- intramuscular needle immunization (IM) was included as a 94

44 cination technology showed promising results in eliciting control. HA-specific antibody, T cell and B cell responses 95

45 both humoral and cell-mediated immune responses and were analyzed in two animal models (rabbit and mouse) 96

46 generating protection against various pathogen challenges, in order to determine the relative strength and types of 97

RO
47 DNA immunization using conventional needle methodolo- immune responses produced by these various immunization 98

48 gies appeared less successful in non-human primates and methods. 99

49 humans. Although more recent human studies have showed

50 that DNA immunization via needle injection is effective in Results 100
51 priming human immune responses [9], high DNA doses in
the milligram (mg) range are usually required. On the other
DP
52
HA-specific antibody responses in NZW rabbits 101
53 hand, physical methods of delivering DNA plasmids, such as
54 a gene gun (GG) and electroporation (EP) have been shown
Groups of New Zealand White (NZW) rabbits were immunized 102
55 to produce an immune response in large animals and humans
with either H5 HA DNA vaccine or empty DNA vaccine vec- 103
56 using microgram (␮g) amounts of DNA [10—13].
tor as outlined in Table 1. The encoded HA antigen has the 104
57 Particle Mediated Epidermal Delivery (PMED), the com-
same amino acid sequence as that in an avian H5N1 virus iso- 105
58 mercial name for the current gene gun technology, is a
late A/VietNam/1203/04 but the gene sequence included in 106
TE

59 needle free device which delivers gold beads coated with

this H5 HA DNA vaccine was codon optimized as previously 107
60 DNA vaccine plasmids into the epidermal layer of skin. Pre-
reported [44]. Nine rabbits (3/group) received the H5 HA 108
61 clinical studies have shown that multiple plasmids can be
DNA vaccine by IM, EP or GG delivery. The dose range used 109
62 successfully combined into one formulation [14], and that
in the current study was based on the reported optimal doses 110
63 a single vaccination, in certain cases, can protect against
for each delivery method so the immune responses will not 111
EC

64 multiple potential pathogens. This vaccine delivery method

be biased due to inadequate doses of DNA vaccines. In addi- 112
65 has been used in small animal studies with strong immuno-
tion, 4 immunizations were given so the peak level antibody 113
66 genicity results [15], and it has also shown promise in larger
responses, as well as the kinetics of antibody responses can 114
67 animals, such as pigs [16] and non-human primates [17,18],
be evaluated. A total of 5 rabbits received the empty DNA 115
68 and in humans against hepatitis B virus [1,19—21], influenza
vector as the negative controls. 116
69 [2] and malaria [22].
RR

70 Electroporation was reported for in vitro gene transfer

71 (also named electropermeabilization) as early as in 1982 Table 1 Rabbit immunization groupsa
72 [23] and has been used for optimization of in vitro and ex
73 vivo gene transfer [24—28]. In the last decade, in vivo EP Groups Rabbit ID DNA vaccines Routeb Dose
74 has been proven to be an efficient approach for delivering (␮g)
genes into muscle tissue [29—32]. The local (in vivo) deliv-
CO

75
A IM-1, IM-2, IM-3 H5 HA DNA IM 200
76 ery of multiple long square-wave electric pulses (20—30 ms)
vaccine
77 of low voltage (50—200 V/cm) shortly after the administra-
B EP-1, EP-2, EP-3 H5 HA DNA EP 200
78 tion of naked DNA in various tissues improves transfection
vaccine
79 efficiency, resulting in a 100—1000-fold increase in pro-
C GG-1, GG-2, GG-3 H5 HA DNA GG 30
80 tein expression [33—37]. Indeed, in vivo electroporation
vaccine
markedly enhances the effectiveness of DNA vaccination
UN

81
D Vc-1, Vc-2, Vc-3 Empty vector GG 30
82 in eliciting both humoral and cellular immune responses in
E Vc-4 Empty vector EP 200
83 various animal models including mice [11,33,38,39], goats
F Vc-5 Empty vector IM 200
84 and cattle [40], and non-human primates [41]. Furthermore,
a All rabbits received DNA immunizations at weeks 0, 2, 4 and
85 DNA immunization by EP has been shown to induce prolonged
86 primary immune responses and maintained immune mem- 6.
b GG: gene gun immunization; IM: intramuscular injection; EP:
87 ory for up to 6 months following just one DNA immunization
electroporation.
[42,43].

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
The relative immunogenicity of DNA vaccines delivered by the intramuscular needle injection 3

OF
RO
DP
Figure 1 Temporal anti-H5 HA IgG responses in rabbits measured by ELISA at 1:500 serum dilution. (A) Rabbits received empty DNA
vector via intramuscular injection (IM, Vc-1), electroporation (EP, Vc-2) or gene gun (GG, Vc-3—Vc-5). Rabbits received H5 HA DNA
immunizations via different vaccination approaches: (B) IM, (C) EP, or (D) GG. The arrows indicate the time of DNA immunizations.
Each curve represents one rabbit.

Rabbits that received the HA DNA vaccine via IM inocula- ery approaches were based on the optimal doses reported
TE

117 145

118 tion were slower to produce HA-specific antibody responses in literature. The mice in the experimental and control GG 146

119 and the peak level antibodies were also lower when com- groups received 6 ␮g of DNA plasmid while those in the EP 147

120 pared to rabbits that received the HA DNA vaccine via EP and IM groups received 100 ␮g at each immunization. 148

121 or GG inoculations (Figs. 1 and 2). The temporal HA-specific

122 antibody responses measured by ELISA at 1:500 serum dilu-
EC

123 tion showed that HA-specific antibody responses started

124 to appear in rabbits from the IM group only after three
125 immunizations while rabbits in the EP and GG groups pro-
126 duced detectable antibody responses after one to two DNA
127 immunizations (Fig. 1). GG immunization was significantly
RR

128 more effective than EP method in eliciting higher antibody

129 responses (p < 0.01) (Fig. 2), particularly when the fact that
130 ∼5 times more HA DNA vaccine was used in the EP group
131 (200 ␮g) than the GG group (36 ␮g) at each immunization is
132 taken into account. As expected, rabbits that received the
133 empty DNA vector did not have any H5 HA-specific antibody
CO

134 responses after 4 immunizations (Fig. 1A).

135 HA-specific antibody responses in Balb/C mice

136 Four groups of Balb/C mice (5 mice/group) were immunized Figure 2 Peak level serum anti-H5 HA IgG antibody responses
with either H5 DNA vaccine or empty DNA vector as shown in
UN

137 in rabbits induced by different vaccination approaches: intra-

138 Table 2. Mice in Groups A, B, and C received the H5 HA DNA muscular injection (IM), electroporation (EP) or gene gun (GG).
139 vaccine by IM, EP or GG immunizations, respectively. For All rabbits received H5 HA DNA immunizations and sera were
140 the empty vector control group (Group D), we only included collected at 2 weeks after the 4th DNA immunization. Each bar
141 the gene gun method because it is the most immunogenic represents an individual rabbit serum. The broken horizontal
142 delivery approach based on the above rabbit study and thus, lines indicate the mean antibody titer of the relevant group.
143 may be the most likely approach to elicit any non-specific The statistical difference between each group was determined
144 immune responses. Again the doses used for different deliv- and the differences with p < 0.05 are indicated.

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
4 S. Wang et al.

specific antibody responses and produced lower responses 153

Table 2 Mouse immunization groupsa
when compared to mice that received the H5 DNA immu- 154

Groups No. of DNA vaccines Routeb Dose nization via EP or GG inoculation (Fig. 3). The temporal 155

mice (␮g) HA-specific antibody responses measured by ELISA at 1:500 156

serum dilution showed that HA-specific antibody responses 157

A 5 H5 HA DNA vaccine IM 100 started to appear in mice from the IM group after three (3) 158
B 5 H5 HA DNA vaccine EP 100 immunizations while the mice in the EP and GG groups pro- 159
C 5 H5 HA DNA vaccine GG 6 duced detectable antibody responses after only 2 DNA immu- 160
D 5 Empty vector GG 6 nizations (Fig. 3). Temporal antibody curves showed that 161
a All mice received DNA immunizations at weeks 0, 2, 4 and 6. antibody responses in the GG immunized group rose earlier 162
b GG: gene gun immunization; IM: intramuscular injection; EP:
and higher when compared to mice in the EP group (Fig. 3), 163
electroporation. although ∼15 times more HA DNA vaccine was used in the 164

EP group (100 ␮g) compared to the GG group (6 ␮g). How- 165

149 As with the rabbit immunizations, mice that received the ever, after four immunizations, the end titers at peak level 166

OF
150 empty DNA vector did not have any H5 HA-specific responses between these two approaches were very similar (Fig. 3). 167

151 after 4 inoculations (Fig. 3). The mice that received the HA Both EP and GG DNA immunization induced significantly 168

152 DNA vaccine via IM inoculation were slower to produce HA- higher HA-specific antibody titers when compared to IM DNA 169

immunization (p < 0.05, for both comparisons) (Fig. 3). 170

RO
HA-specific antibody secreting cells (ASC) in Balb/C 171

mice 172

In addition to analyzing serum HA-specific IgG responses 173

induced by various DNA vaccination approaches, we also 174

examined HA-specific antibody secreting cells from spleno- 175

DP
cytes of immunized mice 2 weeks after the 4th DNA 176

immunization. ELISPOT plates were coated with recombi- 177

nant H5 HA proteins. Mice that received HA DNA vaccines 178

displayed HA-specific ASC in splenocytes and the number of 179

HA positive ASC in each group (i.e., IM, GG, EP) were in gen- 180

eral consistent with their respective HA-specific antibody 181

responses (Fig. 4). Mice that received empty DNA vector did
TE

182

not display any HA-specific ASC in the splenocytes and that 183

background levels against a mock antigen coding (PBS) on 184

the ELISPOT plate for HA DNA vaccine groups were very low 185

(Fig. 4). Overall, mice that received an IM immunization of 186

the HA DNA vaccine generated significantly lower numbers

EC

187

of ASC in splenocytes following the 4th DNA immunization 188

when compared to mice in the EP and GG groups (p < 0.05, 189

both comparisons) (Fig. 4). 190

HA-specific T cell responses in Balb/C mice 191

RR

In addition to the antibody responses, HA-specific T cell 192

Figure 3 H5 HA-specific antibody responses in mice. (A) Tem- responses were also evaluated following HA DNA vaccina- 193

poral anti-H5 HA IgG responses in mice measured at 1:500 serum tion using the three different DNA delivery methods in mice. 194

dilution. Each curve represents the average antibody responses Mouse splenocytes were collected 2 weeks after the 4th 195

for each group of mice from three independent assays. ‘‘H5 DNA immunization and stimulated with the peptide IYST-
CO

196

HA—IM’’, ‘‘H5 HA-EP’’ and ‘‘H5 HA—GG’’ indicate the groups VASSL, a previously reported HA-specific T cell epitope. HA 197

receiving H5 HA DNA immunizations via intramuscular injec- peptide-specific IFN-␥ secreting cells were measured by IFN- 198

tion (IM), electroporation (EP) or gene gun (GG), respectively. ␥ ELISPOT. The mice that received the HA DNA vaccine 199

‘‘Vector—GG’’ indicates the mice receiving empty DNA vector generated HA peptide-specific IFN-␥ secreting cells follow- 200

by gene gun. Group pooled mouse sera from 5 mice in each group ing HA peptide stimulation (Fig. 5). Mice that received the 201

empty DNA vector via GG immunization did not have HA

UN

were used in the ELISA. The arrows indicate the time of DNA 202

immunizations. (B) Peak level serum anti-H5 HA IgG responses peptide-specific IFN-␥ spots in the splenocytes, and the 203

in mice immunized with empty DNA vector or with H5 HA DNA background level spots by stimulation with an unrelated 204

vaccine via IM, EP or GG immunization. Each bar represents the mock peptide were very low (Fig. 5). In general, mice in 205

average titer of each group of 5 mice with standard error. Mouse all three groups that received the HA DNA vaccine produced 206

sera were collected at 2 weeks after the 4th DNA immunization. a high level of IFN-␥ T cell responses (200—400 spot forming 207

The statistical difference between each group was determined cells/million splenocytes) after the 4th DNA immunization. 208

and the differences with p < 0.05 are indicated. Although mice in EP and GG groups had slightly higher IFN-␥ 209

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
The relative immunogenicity of DNA vaccines delivered by the intramuscular needle injection 5

OF
RO
DP

Figure 4 H5 HA-specific antibody secreting cells (ASC) in splenocytes of immune mice as measured by ELISPOT. Mice were immu-
TE

nized with empty DNA vector or with H5 HA DNA vaccine via IM, EP or GG immunization, as indicated. (A) Actual sample wells of
H5 HA-specific ASC spots with either mock antigen coated or H5 HA antigen coated plates. (B) Frequency of H5 HA-specific ASC
per million splenocytes in each group. Data represent the average of spot forming cells (SFCs) per million of splenocytes from
5 mice/group plus standard deviation. The splenocytes were collected 2 weeks after the 4th DNA immunization. The statistical
difference between each group was determined and groups with p < 0.05 are indicated.
EC

210 T cell responses than those in IM group, the difference was MacVector software 7.2 against codon preference of Homo 229

211 not significant (Fig. 5). sapiens. The less optimal codons in the HA genes were 230

212 Because the above results demonstrate that all three replaced by the preferred codons of mammalian systems 231

213 DNA vaccination approaches induced HA peptide-specific T in order to produce a higher expression of the HA pro- 232
RR

214 cell responses, we further examined if there is any dif- teins. Sequence optimization was also performed to make 233

215 ference in the types of T helper responses among these the mRNA more stable and the gene more favorable for tran- 234

216 approaches by using the IgG1/IgG2a ratio as a surrogate scriptional and translational processes. During the sequence 235

217 marker. Interestingly, the ratio of IgG1/IgG2a in mice receiv- optimization, the following cis-acting sequence motifs were 236

218 ing either IM or EP immunizations was much less than 1, avoided: internal TATA-boxes, chi-sites and ribosomal entry 237

indicating that these DNA immunization methods induced sites; AT-rich or GC-rich sequence stretches; ARE, INS,
CO

219 238

220 predominantly a Th1-type antibody response. However, the CRS sequence elements; cryptic splice donor and acceptor 239

221 ratio of IgG1/IgG2a in mice receiving GG immunizations was sites; and branch points. Despite these DNA level sequence 240

222 clearly greater than 1, indicating that this DNA immuniza- changes, the final codon optimized HA DNA sequences still 241

223 tion method induced predominantly a Th2-type antibody produce the same HA amino acid sequences as in the 242

224 response. original H5N1 virus A/VietNam/1203/04. The codon opti- 243

mized HA gene was chemically synthesized by Geneart

UN

244

(Regensburg, Germany). With a human tissue plasminogen 245

225 Materials and methods activator (tPA) leader sequence substituting the natural HA 246

leader sequence, the tPA-HA gene insert was cloned into 247

226 Codon optimized H5 HA DNA vaccine the pSW3891 vector. The DNA vaccine plasmid was pro- 248

duced from Escherichia coli (HB101 strain) with a Mega 249

227 The codon usage of the HA gene from influenza A purification kit (Qiagen, Valencia, CA) for both in vitro trans- 250

228 human viruses H5N1 A/VietNam/1203/04 was analyzed using fection and in vivo animal immunization studies. H5 HA 251

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
6 S. Wang et al.

OF
RO
DP
Figure 5 ELISPOT analysis for IFN-␥ secretion in mouse splenocytes immunized with empty DNA vector or with H5 HA DNA vaccine
via IM, EP or GG immunization, as indicated. (A) Actual sample wells with mock or HA peptide stimulations. (B) Frequency of HA
peptide-specific spots per million splenocytes in each group. Data represent the average of spot forming cells (SFCs) per million of
splenocytes from each group plus standard deviation. The splenocytes were collected 2 weeks after the 4th DNA immunization.
TE

252 expression was verified by in vitro transfection of 293T to gold beads (1 ␮m in size) at a ratio of 2 ␮g plasmid/mg Q2 278

253 cells and Western blot analysis, as previously described gold (Fig. 6). 279

254 [44]. All immunizations were given at weeks 0, 2, 4, 6. Serum 280

samples were taken prior to the first immunization and 2 281

EC

255 DNA immunization of New Zealand White (NZW)

256 rabbits and BALB/c mice

257 NZW rabbits (∼2 kg body weight) and BALB/c mice (6—8
258 weeks old) were purchased from Shanghai Animal Center,
RR

259 Chinese Academy of Science. Both rabbits and mice were

260 housed in the Animal Research Center at the Nanjing Medi-
261 cal University in accordance with approved protocol. Three
262 DNA vaccine immunization systems were used for rabbit and
263 mouse immunizations as follows: (1) Intramuscular injection
(IM): 200 ␮g (rabbits) or 100 ␮g, (mice) of HA DNA vaccine
CO

264

265 plasmid or vector control plasmid was delivered at 2 differ-

266 ent sites in the quadriceps muscle at each immunization.
267 (2) Electroporation: a Caliper Electrodes style electropora-
268 tor (SCIENTZ-2C) from Scientz Co., Ltd. (Ningbo, China) was
269 used. Following IM injection of DNA vaccines as described Figure 6 Subtype analysis of H5 HA-specific IgG in mouse sera
above, the injection sites were electroporated once with
UN

270 raised by the H5 HA DNA vaccine through different vaccination

271 the following parameters: 100 V, 60 ms and 60 Hz for rabbits approaches: IM, EP or GG. The H5 HA-specific IgG1 and IgG2a
272 and 50 V, 60 ms and 60 Hz for mice. (3) Gene gun immuniza- were measured by IgG isotype ELISA. The IgG1/IgG2a ratios were
273 tion: a Helios gene gun (Bio-Rad) was used to deliver the based on the concentration of IgG1 or IgG2a isotypes calculated
274 DNA vaccine at the shaved abdominal skin [45]. In gene gun using the standard curves established with known concentration
275 groups, a total of 36 ␮g (rabbits) or 6 ␮g (mice) of HA DNA of mouse IgG1 or IgG2a. The mouse sera were collected 2 weeks
276 vaccine plasmid or vector control plasmid was delivered at after the 3rd DNA immunization. The bars show the average
277 each immunization. Each shot delivers 1 ␮g of DNA coated IgG1/IgG2a ratios in each group plus standard deviation.

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
The relative immunogenicity of DNA vaccines delivered by the intramuscular needle injection 7

282 weeks after each immunization for the study of HA-specific A/VietNam1203/04. The non-HA peptide was from HIV Env 339

283 antibody responses. Mouse splenocytes were collected 1 V3 region (IGPGRAFYT) as a negative control. Peptides were 340

284 week after the last DNA immunization to measure the H5 purchased from AnaSpec Corp. (San Jose, CA). The peptides 341

285 HA-specific ASC and HA peptide-specific T cell responses. (final concentration 4 ␮g/ml) were added to the wells with 342

100 ␮l of freshly isolated splenocytes (500,000 cells/well 343

286 ELISA (enzyme-linked immunosorbent assay) in R10 medium) in duplicate. The plates were incubated 344

for 24 h overnight at 37 ◦ C in 5% CO2 . The plates were 345

then washed, incubated with 100 ␮l of biotinylated rat 346

287 An ELISA was conducted to measure the H5 HA-specific anti-
anti-mouse IFN-␥ Rat IgG1 (1 ␮g/ml in dilution buffer in 347
288 body (IgG) responses in immunized rabbits and mice, as
the kit), and incubated at 4 ◦ C overnight. After additional 348
289 previously described [44]. The 96-well flat-bottom plates
washes, 100 ␮l of HRP-conjugated streptavidin complex was 349
290 were coated with 100 ␮l/well of transiently expressed HA
added to each well in above dilution buffer for 2 h at room 350
291 antigen at 1 ␮g/ml from 293T cells transfected with the
temperature. The plates were washed, and spots repre- 351
292 H5 HA DNA vaccine plasmids. After being washed 5 times,
senting individual IFN-␥-producing cells were detected after 352
as described above, the plates were then blocked with

OF
293
a 25-min color reaction using AEC coloring system. IFN-␥ 353
294 200 ␮l/well of blocking buffer (5% non-fat dry milk, 4%
spot-forming cells (SFC) were counted. The results were 354
295 Whey, 0.5% Tween-20 in PBS at pH 7.2) for 1 h. After five
expressed as the number of SFC per 106 input cells. The 355
296 washes, 100 ␮l of serially diluted rabbit or mouse serum in
number of peptide-specific IFN-␥-secreting T cells was cal- 356
297 Whey buffer (4% Whey, 0.5% Tween-20 in PBS) was added
culated by subtracting the background (no-peptide) control 357
298 in duplicate wells and incubated for 1 h. To detect rab-
value from the established SFC count.

RO
bit antibody, the plates were incubated for 1 h at 37 ◦ C
358
299

300 with 100 ␮l of biotinylated anti-rabbit (Vector Laboratories,

301 Burlingame, CA) diluted at 1:2000 in Whey buffer. Then, Assays for antibody secreting cells 359
302 100 ␮l of horseradish peroxidase (HRP)-conjugated strepta-
303 vidin (Vector Laboratories) diluted at 1:5000 in Whey buffer IP plates were coated with H5 HA antigen produced from 360
304 was added to each well and incubated for 1 h. To detect the transient transfection of 293T cells at a concentra- 361
DP
305 the mouse antibody, a similar protocol was used except that tion of ∼2 ␮g/ml in PBS at 4 ◦ C overnight, then blocked as 362
306 the plates were incubated for 1 h at 37 ◦ C with 100 ␮l of HRP- described above. Freshly isolated splenocytes (100 ␮l/well, 363
307 conjugated goat-anti-mouse IgG diluted at 1:10,000 in Whey 500,000 cells/well) in R10 medium with 0.1% ␤-ME were 364
308 buffer. After the final wash, the plates were developed with incubated in duplicate wells for 4 h at 37 ◦ C. The plates were 365
309 3,3 ,5,5 tetramethylbenzidine (TMB) solution at 100 ␮l per then washed, incubated with 100 ␮l of biotinylated goat- 366
310 well (Sigma, St. Louis, MO) for 3.5 min. The reactions were anti-mouse IgG diluted at 1:1000 in dilution buffer from the 367
311 stopped by adding 25 ␮l of 2 M H2 SO4 , and the plates were ELISPOT kit above for 1 h. After additional washes, 100 ␮l
TE

368
312 read at an OD of 450 nm. The end titration titer was deter- of HRP-conjugated streptavidin complex diluted at 1:2000 369
313 mined as the highest serum dilution that has an OD reading in the dilution buffer was added to each well for 1 h at 370
314 above twice of that from the negative control serum. 37 ◦ C, then spots were developed using AEC coloring system 371
315 IgG1 or IgG2 isotype-specific ELISA was conducted, as pre- as described above. 372
316 viously described [46]. This assay is similar to the above total
EC

317 IgG ELISA, with the exception that horseradish peroxidase

318 (HRP)-conjugated goat-anti-mouse IgG1 or IgG2 (Southern Statistical analysis 373

319 Technology Associates, AL) at 1:2000 dilution was used. The

320 concentrations for HA-specific mouse IgG1 or IgG2a were cal- Student’s t test was used to analyze the differences in anti- 374

321 culated from the standard curve using a known amount of body responses between animal immunization groups, as 375

measured by antibody titers and spot forming cells in ASC

RR

322 purified mouse IgG. 376

and IFN-␥ ELISPOT assays. A p value of less than 0.05 was 377

considered significant. 378

323 Gamma interferon enzyme-linked immunospot
324 (IFN-␥ ELISPOT) assay
Discussion 379

IFN-␥ ELISPOT assays were performed on fresh mouse spleno-

CO

325

326 cytes, as previously described [47,48]. The IFN-␥ ELISPOT Since the discovery of DNA vaccination as a novel technology 380

327 assay for HA-specific T cells used the mouse IFN-␥ ELISPOT to induce antigen-specific immune responses, different DNA 381

328 kit from U-CyTech Biosciences, Netherland and Multiscreen plasmid delivery approaches can be grouped into two major 382

329 Immobilon P membrane White Sterile 96-well plates (IP types. The first type includes ‘‘chemical approaches’’, rely- 383

330 plates) from Millipore, MA. The assay was performed accord- ing on the chemical and biochemical interactions between 384

ing to manufacturers’ directions. Briefly, the plates were DNA molecules and the target cells. With this type of deliv-
UN

331 385

332 coated with 5 ␮g/ml of purified rat anti-mouse IFN-␥ IgG1 in ery, DNA plasmids are dissolved in various solutions, with 386

333 PBS at 4 ◦ C overnight. After the plates were washed three or without carrier polymers (in lipid-form or other chemi- 387

334 times with PBS, each plate was blocked by the addition of cal natures), and delivered by conventional intramuscular 388

335 200 ␮l of the blocking buffer in each well for 1 h at 37 ◦ C. or intradermal needle injections, transdermally or through 389

336 The known H2d mouse restricted HA epitope (IYSTVASSL) mucosal administration [4,50—56]. The second type of DNA 390

337 [49] was employed for measuring HA-specific T cells. This HA delivery includes ‘‘physical approaches’’. With this type of 391

338 epitope is well-conserved in the H5 HA sequence for H5N1 approach, the delivery of DNA plasmids is based on var- 392

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
8 S. Wang et al.

393 ious forms of physical forces, such as shock wave, high the extracellular space, as is true for conventional needle 455

394 pressure gas and electrical pulse. In general, the physi- delivery. 456

395 cal approaches require special devices that can produce Recently, additional methods based on physical princi- 457

396 external forces, such as a gene gun or an electroporation ples, such as electorporation, have demonstrated increased 458

397 device [3,12,13,57,58]. There have been limited studies to immunogenicity when compared to the chemical delivery 459

398 directly compare these two types of DNA vaccine delivery method via needle injection alone approach, as EP presum- 460

399 approaches, but available data indicates that the physi- ably creates transient pores in the cell membranes, and 461

400 cal method is more effective in eliciting higher immune increases movement of the DNA into the cells due to an 462

401 responses in animal studies and that the type of immune electrophoretic effect [68]. The effectiveness of EP has 463

402 responses elicited by these two types of delivery methods been shown in a number of studies [33,38—42,60,69—71]. 464

403 may also be different [15,59,60]. The application of electroporation, regardless of the site of 465

404 The difference between the chemical and physical injection, should favor the transfection of a greater vari- 466

405 approaches is further highlighted by the results from early ety of cells, including APCs. Furthermore, as an additional 467

406 phase DNA vaccine studies in humans. DNA vaccines deliv- mechanism, mild tissue damage, which may be induced by 468

OF
407 ered by chemical methods, including those with facilitating electroporation, could provoke an influx of APCs, induce 469

408 chemical agents, have not been shown to be particularly danger signals, and enhance the release of antigen from 470

409 immunogenic. In the first series of DNA vaccine studies injured cells, thereby increasing antigen presentation, and 471

410 in humans, a group of asymptomatic HIV-infected individ- also possibly provide adjuvant effect [72,73]. The chemi- 472

411 uals who received a DNA vaccine encoding an env gene cal approach of delivering DNA with conventional needle 473

and a rev gene from the HIV-1MN isolate via conventional delivery has a particular disadvantage in that the DNA is

RO
412 474

413 intramuscular needle injection. While this study confirmed deposited into the extracellular space and subsequently 475

414 the safety of DNA immunization, no consistent change in needs to be taken up by the cells—–a critical step, which so 476

415 CD4 or CD8 T lymphocyte counts or in plasma HIV con- far has not been able overcome by any particular chemical 477

416 centrations was observed [61,62]. In another study, it was carriers. 478

417 found that although a DNA vaccine encoding the P. falci- In the current study, we compared the relative immuno- 479

parum circumsporozoite protein for malaria, could elicit genicity of the two leading physical delivery methods: gene
DP
418 480

419 low level antigen-specific CTL responses through conven- gun and electroporation, in both rabbit and mouse mod- 481

420 tional intramuscular injection [63], measurable antibody els. Our data show that both of these methods are highly 482

421 responses were not detected in any of the 20 volunteers [64]. effective in inducing antigen-specific immune responses, 483

422 On the other hand, gene gun delivery of a plasmid express- and both are more immunogenic than the conventional nee- 484

423 ing hepatitis B surface antigen (HBsAg) in naive volunteers dle injection method. The difference in immunogenicity 485

424 showed promising immune responses in humans [1]. Three between these two physical approaches is somewhat lim- 486
TE

425 administrations of plasmid DNA by this approach resulted ited. While the gene gun approach is significantly more 487

426 in the generation of seroprotective antibody levels in all effective in eliciting antibody responses in rabbits and only 488

427 12 volunteers. Furthermore, all 12 volunteers showed cell- slightly more effective in mice, when compared to the EP 489

428 mediated immune responses to HBsAg post-administration, approach, this advantage appears to disappear when multi- 490

429 highlighting the ability of the physical delivery approach to ple immunizations are given. On the other hand, EP may be 491
EC

430 generate broad-based immune responses. More recently, a more effective in inducing higher Th1-type antibody immune 492

431 phase I human study using gene gun technology examined responses as shown in our study while GG predominantly 493

432 the safety and immunogenicity of a DNA vaccine, pPJV1671, induced Th2-type antibody immune responses as well doc- 494

433 which expresses the HA from the H3 Panama strain of the umented in literature [74,75]. This is interesting because 495

434 human influenza virus in healthy adults [2]. Volunteers in both are ‘‘physical delivery methods’’ and raise the ques- 496

435 the 4 ␮g dose group (highest dose) achieved the criteria on tion of whether IM injection, even after electroporation, is 497
RR

436 all three parameters required for licensure by the Commit- the key factor in controlling these differences. In the cur- 498

437 tee for Proprietary Medical Products (CPMP) in the European rent study, we did not directly measure secreted cytokines 499

438 Union (seroconversion, seroprotection and GMT) 21 days because it is not clear whether changes in the Th subtypes 500

439 after a single vaccination. These data highlight the poten- that would have been examined in mice would be valid 501

440 tial of DNA vaccines in humans, when delivered by a physical in non-human primates or humans, the ‘‘real’’ targets for 502

method. improved DNA vaccine delivery. With the increasing use of

CO

441 503

442 The above results suggest that the ability of DNA vaccines EP in large animal studies and the pending phase I clinical 504

443 to elicit an immune response in a host appear to greatly study in healthy human volunteers, a more complete under- 505

444 depend upon the route of administration (i.e., chemical vs. standing of antigen-specific T cell responses, including the 506

445 physical delivery). A high level immune response appears to unique profiles of cell-mediated immune responses related 507

446 rely on the ability of the DNA plasmids to efficiently enter to the use of different DNA delivery methods, will be gained 508

targeted cells and not so much on the amount of DNA that by using polyfunctional cytokine analyses.
UN

447 509

448 is administered. DNA immunization via gene gun adminis- EP delivery is clearly more effective than IM delivery in 510

449 tration allows the DNA plasmids to penetrate directly into our study but the magnitude of increase was less than pre- 511

450 the cytoplasm [3,65,66], presumably resulting in the DNA viously reported [33], reasons for which there may be a 512

451 being processed by antigen presenting cells (APCs) and sub- number of factors. First, the immunogenicity of DNA vac- 513

452 sequently presented to T cells [67]. The same holds true for cines is highly dependent on individual antigens. The HA 514

453 any method that increases the ability of the DNA to enter antigen is known to be highly immunogenic and since the HA 515

454 directly into the intracellular environment rather than into gene insert used in the current study is codon optimized, the 516

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
The relative immunogenicity of DNA vaccines delivered by the intramuscular needle injection 9

517 immunogenicity of the HA DNA vaccine is further improved References 575

518 [44]. Other factors which influence the immunogenicity of
519 DNA vaccines irrespective of whether or not GG or EP is used [1] Roy MJ, Wu MS, Barr LJ, et al. Induction of antigen-specific Q3 576
520 include the choice of DNA vaccine vector and the design CD8+ T cells, T helper cells, and protective levels of antibody 577
521 of the testing antigen as part of the antigen engineering in humans by particle-mediated administration of a hepatitis B Q4 578

522 process [76,77]. In combination, these factors may improve virus DNA vaccine. Vaccine 2000;19(November(7—8)):764—78. 579

523 the immunogenicity of the DNA vaccine included in the cur- [2] Drape RJ, Macklin MD, Barr LJ, Jones S, Haynes JR, Dean 580

524 rent study which may result in a smaller difference observed HJ. Epidermal DNA vaccine for influenza is immunogenic in 581

525 between IM and EP deliveries. humans. Vaccine 2006;24(May(21)):4475—81. 582

[3] Tang DC, Devit M, Johnston SA. Genetic immunization is 583
526 In the current study, a caliper electrode model elec-
a simple method for eliciting an immune response. Nature 584
527 troporator was used. Currently, there is no standardized 1992;356(6365):152—4. 585
528 technology in EP. There are at least 3—4 different major [4] Ulmer JB, Donnelly JJ, Parker SE, et al. Heterologous pro- 586
529 models in the field. The results learned from the current tection against influenza by injection of DNA encoding a viral 587
530 report may not completely represent the results from other protein. Science 1993;259(5102):1745—9. 588

OF
531 models of electroporators. Future studies may need to over- [5] Robinson HL, Hunt LA, Webster RG. Protection against a 589

532 come the commercial barrier and conduct parallel studies lethal influenza virus challenge by immunization with a 590

533 to examine the safety and efficacy of different models haemagglutinin-expressing plasmid DNA. Vaccine 1993;11(9): 591

534 of electroporators, especially in humans using the same 957—60. 592

535 antigens. [6] Lu S, Santoro JC, Fuller DH, Haynes JR, Robinson HL. 593

Use of DNAs expressing HIV-1 Env and noninfectious HIV- 594

While, in the current study, we utilized an ELISA to mea-

RO
536
1 particles to raise antibody responses in mice. Virology 595
537 sure antibody responses to vaccination, other assays such as 1995;209(1):147—54. 596
538 hemagglutinination inhibition (HI) would have been a more [7] Wang B, Ugen KE, Srikantan V, et al. Gene inoculation generates 597
539 useful measure of protective antibodies against influenza. immune responses against human immunodeficiency virus type 598
540 However, the main goal of the current study was to exam- 1. Proc Natl Acad Sci USA 1993;90(May(9)):4156—60. 599
541 ine the immunogenicity, rather than the protective qualities [8] Davis HL, Demeneix BA, Quantin B, Coulombe J, Whalen 600

of the different vaccination approaches and given that the RG. Plasmid DNA is superior to viral vectors for direct gene
DP
542 601

543 same antigen is used, the levels of protective antibodies transfer into adult mouse skeletal muscle. Hum Gene Ther 602

544 are usually proportional to the level of overall antibody 1993;4(December(6)):733—40. 603

545 response. [9] Wang S, Kennedy J, West K, et al. Balanced cellular and anti- 604

body responses induced by the polyvalent DNA prime-protein Q5 605

546 Previously, the relative immunogenicity of DNA vaccine
boost HIV-1 vaccine formulation DP6-001 in healthy human vol- 606
547 delivery methods by EP or GG was studied in an anti-tumor unteers. Vaccine 2008, doi:10.1016/jvaccine.2007.12.024. 607
548 model in transgenic mice [60] where the overall antigen- [10] Pertmer TM, Eisenbraun MD, McCabe D, Prayaga SK, Fuller DH, 608
TE

549 specific immune responses between these two methods were Haynes JR. Gene gun-based nucleic acid immunization: elicita- 609
550 very similar. The only difference was that GG is less effec- tion of humoral and cytotoxic T lymphocyte responses following 610
551 tive than EP in controlling the incidence and the growth of epidermal delivery of nanogram quantities of DNA. Vaccine 611

552 spontaneous tumors, however, there was no data to rule out 1995;13(15):1427—30. 612

553 a non-antigen-specific effect in this finding. Future studies [11] Dobano C, Widera G, Rabussay D, Doolan DL. Enhancement of 613
EC

554 should include the development of standardized protocols antibody and cellular immune responses to malaria DNA vac- 614

555 for EP and to compare the different EP models that are cines by in vivo electroporation. Vaccine 2007;25(36):6635—45. 615
[12] Fuller DH, Loudon P, Schmaljohn C. Preclinical and clinical 616
556 available in the field, especially for human clinical stud-
progress of particle-mediated DNA vaccines for infectious dis- 617
557 ies for vaccines against infectious diseases. Furthermore, eases. Methods 2006;40(September(1)):86—97. 618
558 since immune responses may differ dramatically with the [13] Dean HJ, Haynes J, Schmaljohn C. The role of particle- 619
559 use of this immunization method based on the innate proper- mediated DNA vaccines in biodefense preparedness. Adv Drug
RR

620
560 ties of the antigen used, future studies should also examine Deliv Rev 2005;57(June(9)):1315—42. 621
561 the factors that determine the immunogenicity of various [14] Talaat AM, Lyons R, Johnston SA. A combination vaccine 622

562 individual antigens when EP is used as the immunization confers full protection against co-infections with influenza, 623

563 method. The availability of more than one physical deliv- herpes simplex and respiratory syncytial viruses. Vaccine 624

564 ery approach in the field of DNA vaccine is a positive sign 2001;20(Nov(3—4)):538—44. 625

that DNA vaccine technology is maturing and proliferat- [15] Fynan EF, Webster RG, Fuller DH, Haynes JR, Santoro JC, Robin- 626
CO

565
son HL. DNA vaccines: protective immunizations by parenteral, 627
566 ing before being accepted as a vaccination method for
mucosal, and gene-gun inoculations. Proc Natl Acad Sci USA 628
567 general use. 1993;90(24):11478—82. 629

[16] Macklin MD, McCabe D, McGregor MW, et al. Immunization 630

of pigs with a particle-mediated DNA vaccine to influenza A 631
568 Acknowledgements virus protects against challenge with homologous virus. J Virol 632
UN

1998;72(February(2)):1491—6. 633

569 This work was supported in part by the grants U01 AI056536 [17] Lodmell DL, Ray NB, Parnell MJ, et al. DNA immunization 634

570 and R01 AI065250 from the National Institute of Allergy and protects nonhuman primates against rabies virus. Nat Med 635

571 Infectious Diseases, USA and Jiangsu Province Natural Sci- 1998;4(8):949—52. 636

572 ence Foundation grant BK2006728 and grant support for the [18] Fuller DH, Rajakumar PA, Wilson LA, et al. Induction of mucosal 637

protection against primary, heterologous simian immunode- 638

573 Jiangsu Province Key Lab in Infectious Diseases, China. The
ficiency virus by a DNA vaccine. J Virol 2002;76(August(7)): 639
574 authors thank Dr. Jill M. Grimes-Serrano for critical reading 3309—17. 640
of the manuscript.

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
10 S. Wang et al.

641 [19] Tacket CO, Roy MJ, Widera G, Swain WF, Broome S, Edelman tects mice against intranasal poxvirus challenge. Vaccine 709

642 R. Phase 1 safety and immune response studies of a DNA vac- 2007;25(10):1814—23. 710

643 cine encoding hepatitis B surface antigen delivered by a gene [40] Tollefsen S, Vordermeier M, Olsen I, et al. DNA injec- 711
644 delivery device. Vaccine 1999;17(July(22)):2826—9. tion in combination with electroporation: a novel method 712

645 [20] Rottinghaus ST, Poland GA, Jacobson RM, Barr LJ, Roy MJ. Hep- for vaccination of farmed ruminants. Scand J Immunol 713
646 atitis B DNA vaccine induces protective antibody responses in 2003;57(March(3)):229—38. 714

647 human non-responders to conventional vaccination. Vaccine [41] Luckay A, Sidhu MK, Kjeken R, et al. Effect of plasmid DNA 715

648 2003;21(November(31)):4604—8. vaccine design and in vivo electroporation on the resulting 716

649 [21] Roberts LK, Barr LJ, Fuller DH, McMahon CW, Leese PT, Jones S. vaccine-specific immune responses in rhesus macaques. J Virol 717

650 Clinical safety and efficacy of a powdered Hepatitis B nucleic 2007;81(May(10)):5257—69. 718

651 acid vaccine delivered to the epidermis by a commercial pro- [42] Tsang C, Babiuk S, van Drunen Littel-van den Hurk S, Babiuk 719

652 totype device. Vaccine 2005;23(September(40)):4867—78. LA, Griebel P. A single DNA immunization in combination 720

653 [22] McConkey SJ, Reece WH, Moorthy VS, et al. Enhanced T-cell with electroporation prolongs the primary immune response 721

654 immunogenicity of plasmid DNA vaccines boosted by recom- and maintains immune memory for six months. Vaccine 722
655 binant modified vaccinia virus Ankara in humans. Nat Med 2007;25(July(30)):5485—94. 723

OF
656 2003;9(June(6)):729—35. [43] Babiuk S, Tsang C, van Drunen Littel-van den Hurk S, Babiuk LA, 724
657 [23] Neumann E, Schaefer-Ridder M, Wang Y, Hofschneider PH. Gene Griebel PJ. A single HBsAg DNA vaccination in combination with 725

658 transfer into mouse lyoma cells by electroporation in high elec- electroporation elicits long-term antibody responses in sheep. 726

659 tric fields. EMBO J 1982;1(7):841—5. Bioelectrochemistry 2007;70(2):269—74. 727

660 [24] Potter H. Electroporation in biology: methods, applications, [44] Wang S, Taaffe J, Parker C, et al. Hemagglutinin (HA) pro- 728

661 and instrumentation. Anal Biochem 1988;174(November(2)): teins from H1 and H3 serotypes of influenza A viruses require 729

361—73. different antigen designs for the induction of optimal protec-

RO
662 730

663 [25] Tsong TY. Electroporation of cell membranes. Biophys J tive antibody responses as studied by codon-optimized HA DNA 731

664 1991;60(August(2)):297—306. vaccines. J Virol 2006;80(23):11628—37. 732

665 [26] Sukharev SI, Klenchin VA, Serov SM, Chernomordik LV, [45] Wang S, Joshi S, Lu S. Delivery of DNA to Skin by particle bom- 733

666 Chizmadzhev YuA. Electroporation and electrophoretic DNA bardment. In: Heiser WC, editor. Gene Delivery to Mammalian 734
667 transfer into cells. The effect of DNA interaction with elec- Cells. Humana Press; 2004. 735

668 tropores. Biophys J 1992;63(November(5)):1320—7. [46] Wang S, Heilman D, Liu F, et al. A DNA vaccine produc- 736
DP
669 [27] Wolf H, Rols MP, Boldt E, Neumann E, Teissie J. Control by ing LcrV antigen in oligomers is effective in protecting 737

670 pulse parameters of electric field-mediated gene transfer in mice from lethal mucosal challenge of plague. Vaccine 738
671 mammalian cells. Biophys J 1994;66(February(2 Pt 1)):524—31. 2004;22(September(25—26)):3348—57. 739

672 [28] Harrison RL, Byrne BJ, Tung L. Electroporation-mediated gene [47] Chen X, Rock MT, Hammonds J, et al. Pseudovirion particle 740

673 transfer in cardiac tissue. FEBS Lett 1998;435(September(1)): production by live poxvirus human immunodeficiency virus vac- 741

674 1—5. cine vector enhances humoral and cellular immune responses. 742

675 [29] Rols MP, Delteil C, Golzio M, Dumond P, Cros S, Teissie J. In J Virol 2005;79(May(9)):5537—47. 743
TE

676 vivo electrically mediated protein and gene transfer in murine [48] Rock MT, Crowe Jr JE. Identification of a novel human leuco- 744

677 melanoma. Nat Biotechnol 1998;16(February(2)):168—71. cyte antigen-A*01-restricted cytotoxic T-lymphocyte epitope 745

678 [30] Aihara H, Miyazaki J. Gene transfer into muscle by electropo- in the respiratory syncytial virus fusion protein. Immunology 746
679 ration in vivo. Nat Biotechnol 1998;16(September(9)):867—70. 2003;108(April(4)):474—80. 747

680 [31] Mir LM, Bureau MF, Gehl J, et al. High-efficiency gene trans- [49] Tamura M, Kuwano K, Kurane I, Ennis FA. Definition of amino 748

681 fer into skeletal muscle mediated by electric pulses. Proc Natl acid residues on the epitope responsible for recognition by 749
EC

682 Acad Sci USA 1999;96(April(8)):4262—7. influenza A virus H1-specific, H2-specific, and H1- and H2- 750

683 [32] Gehl J, Mir LM. Determination of optimal parameters for in cross-reactive murine cytotoxic T-lymphocyte clones. J Virol 751

684 vivo gene transfer by electroporation, using a rapid in vivo 1998;72(November(11)):9404—6. 752

685 test for cell permeabilization. Biochem Biophys Res Commun [50] Felgner PL, Tsai YJ, Sukhu L, et al. Improved cationic lipid 753

686 1999;261(August(2)):377—80. formulations for in vivo gene therapy. Ann N Y Acad Sci 754

687 [33] Widera G, Austin M, Rabussay D, et al. Increased DNA vac- 1995;772(November):126—39. 755
RR

688 cine delivery and immunogenicity by electroporation in vivo. J [51] Wolff JA, Malone RW, Williams P, et al. Direct gene transfer 756

689 Immunol 2000;164(May(9)):4635—40. into mouse muscle in vivo. Science 1990;247(March(4949 Pt 757
690 [34] Bloquel C, Fabre E, Bureau MF, Scherman D. Plasmid DNA elec- 1)):1465—8. 758

691 trotransfer for intracellular and secreted proteins expression: [52] Nabel GJ, Nabel EG, Yang ZY, et al. Direct gene transfer with 759

692 new methodological developments and applications. J Gene DNA-liposome complexes in melanoma: expression, biologic 760

693 Med 2004;6(February(Suppl. 1)):S11—23. activity, and lack of toxicity in humans. Proc Natl Acad Sci USA 761
CO

694 [35] Mathiesen I. Electropermeabilization of skeletal muscle 1993;90(December(23)):11307—11. 762

695 enhances gene transfer in vivo. Gene Ther 1999;6(April(4)): [53] Raz E, Carson DA, Parker SE, et al. Intradermal gene immu- 763

696 508—14. nization: the possible role of DNA uptake in the induction 764

697 [36] Rizzuto G, Cappelletti M, Maione D, et al. Efficient and of cellular immunity to viruses. Proc Natl Acad Sci USA 765
698 regulated erythropoietin production by naked DNA injec- 1994;91(September(20)):9519—23. 766
699 tion and muscle electroporation. Proc Natl Acad Sci USA [54] Herrmann JE, Chen SC, Jones DH, et al. Immune responses and 767

700 1999;96(May(11)):6417—22. protection obtained by oral immunization with rotavirus VP4 768
UN

701 [37] Khan AS, Pope MA, Draghia-Akli R. Highly efficient constant- and VP7 DNA vaccines encapsulated in microparticles. Virology 769

702 current electroporation increases in vivo plasmid expression. 1999;259(1):148—53. 770

703 DNA Cell Biol 2005;24(December(12)):810—8. [55] Chen SC, Jones DH, Fynan EF, et al. Protective immunity 771
704 [38] Dupuis M, Denis-Mize K, Woo C, et al. Distribution of DNA induced by oral immunization with a rotavirus DNA vaccine 772

705 vaccines determines their immunogenicity after intramuscular encapsulated in microparticles. J Virol 1998;72(7):5757—61. 773
706 injection in mice. J Immunol 2000;165(September(5)):2850—8. [56] Selinsky C, Luke C, Wloch M, et al. A DNA-based vaccine for 774

707 [39] Hooper J, Golden J, Ferro A, King A. Smallpox DNA the prevention of human cytomegalovirus-associated diseases. 775
708 vaccine delivered by novel skin electroporation device pro- Hum Vaccin 2005;1(January—February(1)):16—23. 776

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033
 +Model
JVAC 7854 1—11 ARTICLE IN PRESS
The relative immunogenicity of DNA vaccines delivered by the intramuscular needle injection 11

777 [57] Haynes JR. Particle-mediated DNA vaccine delivery to the skin. [67] Condon C, Watkins SC, Celluzzi CM, Thompson K, Falo Jr LD. 813

778 Expert Opin Biol Ther 2004;4(June(6)):889—900. DNA-based immunization by in vivo transfection of dendritic 814

779 [58] Haynes JR, McCabe DE, Swain WF, Widera G, Fuller JT. cells. Nat Med 1996;2(October(10)):1122—8. 815

780 Particle-mediated nucleic acid immunization. J Biotechnol [68] Zaharoff DA, Barr RC, Li CY, Yuan F. Electromobility of plas- 816

781 1996;44(January(1—3)):37—42. mid DNA in tumor tissues during electric field-mediated gene 817

782 [59] Pertmer TM, Roberts TR, Haynes JR. Influenza virus delivery. Gene Ther 2002;9(October(19)):1286—90. 818
783 nucleoprotein-specific immunoglobulin G subclass and cytokine [69] Babiuk LA, Pontarollo R, Babiuk S, Loehr B, van Drunen Littel- 819

784 responses elicited by DNA vaccination are dependent on the van den Hurk S. Induction of immune responses by DNA vaccines 820

785 route of vector DNA delivery. J Virol 1996;70(9):6119—25. in large animals. Vaccine 2003;21(January(7—8)):649—58. 821

786 [60] Smorlesi A, Papalini F, Amici A, et al. Evaluation of differ- [70] Wang Z, Troilo PJ, Wang X, et al. Detection of inte- 822

787 ent plasmid DNA delivery systems for immunization against gration of plasmid DNA into host genomic DNA following 823
788 HER2/neu in a transgenic murine model of mammary carci- intramuscular injection and electroporation. Gene Ther 824

789 noma. Vaccine 2006;24(March(11)):1766—75. 2004;11(April(8)):711—21. 825

790 [61] MacGregor RR, Boyer JD, Ugen KE, et al. First human trial of a [71] Heller LC, Ugen K, Heller R. Electroporation for targeted gene 826

791 DNA-based vaccine for treatment of human immunodeficiency transfer. Expert Opin Drug Deliv 2005;2(March(2)):255—68. 827

OF
792 virus type 1 infection: safety and host response. J Infect Dis [72] Prud’homme GJ. DNA vaccination against tumors. J Gene Med 828
793 1998;178(July(1)):92—100. 2005;7(January(1)):3—17. 829

794 [62] Boyer JD, Chattergoon MA, Ugen KE, et al. Enhancement [73] Prud’homme GJ, Glinka Y, Khan AS, Draghia-Akli R. 830

795 of cellular immune response in HIV-1 seropositive indi- Electroporation-enhanced nonviral gene transfer for the 831
796 viduals: A DNA-based trial. Clin Immunol (Orlando, Fla) prevention or treatment of immunological, endocrine and 832

797 1999;90(1):100—7. neoplastic diseases. Curr Gene Ther 2006;6(April(2)):243—73. 833

[63] Wang R, Doolan DL, Le TP, et al. Induction of antigen-specific [74] Feltquate DM, Heaney S, Webster RG, Robinson HL. Different T

RO
798 834

799 cytotoxic T lymphocytes in humans by a malaria DNA vaccine. helper cell types and antibody isotypes generated by saline and 835

800 Science 1998;282(5388):476—80. gene gun DNA immunization. J Immunol 1997;158(5):2278—84. 836

801 [64] Le TP, Coonan KM, Hedstrom RC, et al. Safety, tolerability and [75] Weiss R, Scheiblhofer S, Freund J, Ferreira F, Livey I, Thalhamer 837

802 humoral immune responses after intramuscular administration J. Gene gun bombardment with gold particles displays a par- 838

803 of a malaria DNA vaccine to healthy adult volunteers. Vaccine ticular Th2-promoting signal that over-rules the Th1-inducing 839

804 2000;18(March(18)):1893—901. effect of immunostimulatory CpG motifs in DNA vaccines. Vac- 840
DP
805 [65] Eisenbraun MD, Fuller DH, Haynes JR. Examination of parame- cine 2002;20(August(25—26)):3148—54. 841

806 ters affecting the elicitation of humoral immune responses by [76] Wang S, Farfan-Arribas DJ, Shen S, et al. Relative contributions 842
807 particle bombardment-mediated genetic immunization. DNA of codon usage, promoter efficiency and leader sequence to 843

808 Cell Biol 1993;12(November(9)):791—7. the antigen expression and immunogenicity of HIV-1 Env DNA 844

809 [66] Yang NS, Burkholder J, Roberts B, Martinell B, McCabe D. vaccine. Vaccine 2006;24(May(21)):4531—40. 845

810 In vivo and in vitro gene transfer to mammalian somatic [77] Lu S, Manning S, Arthos J. Antigen engineering in DNA immu- 846

811 cells by particle bombardment. Proc Natl Acad Sci USA nization. In: Lowrie DaW R, editor. Methods in Molecular 847
TE

812 1990;87(December(24)):9568—72. Medicine DNA Vaccines. Humana Press; 1999. p. 355—74. 848
EC
RR
CO
UN

Please cite this article in press as: Wang S, et al., The relative immunogenicity of DNA vaccines delivered JVAC
by the intra-
7854 1—11
muscular needle injection, electroporation and gene gun methods, Vaccine (2008), doi:10.1016/[Link].2008.02.033