International Immunology International Immunology Advance Access published December 4, 2014
Sendai virus recombinant vaccine expressing a secreted, unconstrained respiratory syncytial virus fusion protein protects against RSV in cotton rats
Manuscript ID: Manuscript Type:
Complete List of Authors:
INTIMM-14-0098.R1
Original Research
12-Oct-2014
Pe
Date Submitted by the Author:
International Immunology
er
Zhan, Xiaoyan; St. Jude Children's Research Hospital, Department of Infectious Diseases Slobod, Karen; St. Jude Children's Research Hospital, Infectious Diseases Jones, Bart; St. Jude Children's Research Hospital, Department of Infectious Diseases Sealy, Robert; St. Jude Children's Research Hospital, Department of Infectious Diseases Takimoto, Toru; St. Jude Children's Research Hospital, Department of Infectious Diseases Boyd, Kelli; St. Jude Children's Research Hospital, Animal Resource Center Surman, Sherri; St. Jude Children's Research Hospital, Department of Infectious Diseases Russell, Charles; St. Jude Children's Research Hospital, Department of Infectious Diseases Portner, Allen; St. Jude Children's Research Hospital, Department of Infectious Diseases Hurwitz, Julia; St. Jude Children's Research Hospital, Infectious Diseases
ew
vi
Re
Keywords:
Sendai virus , Respiratory syncytial virus, Unconstrained F, secreted fusion protein, protection, vaccine
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
r Fo Journal:
Page 1 of 31
International Immunology
1
REVISED INTIMM-14-0098
2
Sendai virus recombinant vaccine expressing a secreted, unconstrained respiratory
3
syncytial virus fusion protein protects against RSV in cotton rats
4
Xiaoyan Zhan1, Karen S. Slobod1, Bart G. Jones1, Robert E. Sealy1,Toru Takimoto1, Kelli Boyd2,
5
Sherri Surman1, Charles J. Russell1, Allen Portner1, and Julia L. Hurwitz1,3
6 Department of Infectious Diseases and 2Animal Resource Center, St. Jude Children’s
8
Research Hospital; 3Department of Microbiology, Immunology and Biochemistry, University of
9
Tennessee Health Science Center, Memphis, Tennessee
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
1
r Fo
7
Running title: SeV-expressed RSV F secreted protein confers protection against RSV
11
Keywords: protection, respiratory syncytial virus, secreted fusion protein, Sendai virus,
12
unconstrained F, vaccine
13
Current Addresses:
er
Pe
10
Xiaoyan Zhan: Department of Pharmacology, Vanderbilt University, Nashville, TN; Karen
15
Slobod: Infectious Diseases and Pediatrics, Department of Technical Development, Novartis
16
Vaccines, 350 Mass Ave, Cambridge, MA; Toru Takimoto: Department of Microbiology and
17
Immunology, University of Rochester, Rochester, New York; Kelli Boyd: Vanderbilt University
18
School of Medicine, Division of Animal Care, AA-6206 MCN, Nashville, TN;
19 20
*
21
Julia L. Hurwitz
22
Department of Infectious Diseases
23
St. Jude Children’s Research Hospital
24
262 Danny Thomas Place
25
Memphis, TN 38105
26
Phone: 901-595-2464
To whom correspondence should be sent:
ew
vi
Re
14
1
27
Fax: 901-595-3099
28
e-mail:
[email protected] Page 2 of 31 International Immunology
er Pe
ew vi Re
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
r Fo
2
Page 3 of 31
International Immunology
30 31
The respiratory syncytial virus (RSV) is responsible for as many as 199,000 annual deaths
32
worldwide. Currently there is no standard treatment for RSV disease, and no vaccine. Sendai
33
virus is an attractive pediatric vaccine candidate because it elicits robust and long-lasting virus-
34
specific B cell and T cell activities in systemic and mucosal tissues. The virus serves as a gene
35
delivery system as well as a Jennerian vaccine against its close cousin, human parainfluenza
36
virus type 1. Here is described the testing of a recombinant SeV (SeVRSV-Fs) that expresses
37
an unconstrained, secreted RSV-F protein as a vaccine against RSV in cotton rats. After a
38
single intranasal immunization of cotton rats with SeVRSV-Fs, RSV-specific binding and
39
neutralizing antibodies were generated. These antibodies exhibited cross-reactivity with both
40
RSV A and B isolates. RSV-F-specific interferon-γ-producing T cells were also activated. The
41
SeVRSV-Fs vaccine conferred complete protection against RSV challenge without enhanced
42
immunopathology. In total, results showed that an SeV recombinant that expresses RSV F in
43
an unconstrained, soluble form can induce humoral and cellular immunity that protects against
44
infection with RSV.
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
ABSTRACT
r Fo
29
er
Pe
ew
vi
Re
3
International Immunology
Page 4 of 31
46
Respiratory syncytial virus (RSV) can cause close to 200,000 deaths in a single year worldwide
47
(1-6). The most vulnerable individuals are premature infants, children with congenital heart/lung
48
disease, and immunocompromised patients. Currently, there is only one form of successful RSV
49
prophylaxis, which involves the injection of preformed anti-RSV antibodies into vulnerable
50
infants (7). However, because of the high cost and logistical difficulty associated with
51
treatments, pre-formed antibody prophylaxis is rarely available to the children who need it most
52
(8;9). RSV vaccines have been studied for decades, but there remains no licensed product.
53
Clearly, the development of a successful RSV vaccine is a critical, unmet need in the pediatric
54
healthcare arena (10;11).
Pe
55
Here we describe the testing of a Sendai virus-based vaccine that expresses an unconstrained,
57
soluble, RSV F protein. Sendai virus was chosen as a vector, because it induces rapid and
58
durable B cell and T cell responses in blood and in mucosal tissues of the respiratory tract. The
59
virus has already been tested in adults and 3-6 year old children, and shown to be immunogenic
60
and well tolerated.
er
56
vi
Re
61
ew
62
This study addresses a question pertinent to recent literature: can an unconstrained, soluble
63
RSV F protein, truncated to remove transmembrane and intracellular protein regions, suffice as
64
a recombinant vaccine target? The question is based on an understanding that in the absence
65
of its transmembrane region, the pre-fusion form of RSV F protein is metastable and often
66
converts to a post-fusion form, a low-energy six helix bundle that lacks certain antibody
67
neutralizing determinants (12;13). Our data show that even though the RSV F protein expressed
68
by SeVRSV-Fs is unconstrained, the vaccine elicits RSV-specific binding and neutralizing
69
antibodies and T cell responses in cotton rats. The vaccine fully protects against RSV challenge
70
without enhanced immunopathology.
4
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
INTRODUCTION
r Fo
45
Page 5 of 31
71
International Immunology
MATERIALS AND METHODS
72
74
To make SeVRSV-Fs cDNA, we inserted a His tag sequence and a stop codon after the codon
75
for residue 524 in the RSV F gene, using QuickChange Site-directed Mutagenesis Kit
76
(Stratagene). The gene encodes a a protein truncated by 50 amino acids at its C-terminus and
77
therefore lacking transmembrane and intracellular regions The mutated gene was inserted into
78
the NotI site of pSV(E), Figure 1A (14;15).
79
Reverse genetics rescue was performed as described previously(16). The 293T cell line was
81
infected with a UV-inactivated, T7 RNA polymerase-expressing recombinant vaccinia virus
82
(vTF7.3) for 1 hr at 37°C. Cells were then co-transfected with plasmids containing recombinant
83
SeV cDNA plasmid and with supporting T7-driven plasmids respectively expressing the NP, P,
84
and L genes of SeV (pTF1SVNP, pTF1SVP, and pTF1SVL) in the presence of Lipofectamine
85
(Life Technologies, Grand Island, NY). After 40 hrs, cell lysates were prepared and amplified by
86
inoculation into embryonated hens eggs. Allantoic fluids were harvested after three days and
87
virus was cloned by plaque purification on LLC-MK2 cells. Cloned recombinant virus (termed
88
SeVRSV-Fs) was amplified in hen’s eggs to prepare stocks for further testing.
er
Pe
80
ew
vi
Re
89 90
Purification and testing of RSV F protein
91
Allantoic fluids were harvested three days after the inoculation of eggs with SeVRSV-Fs . Fluids
92
were centrifuged (approximately 300 x g, 10 min) to remove debris and filtered (0.45 µM filter).
93
Samples were passed over a PBS-equilibrated Sepharose column bound by the anti-F
94
monoclonal antibody Palivizumab (Synagis; MedImmune Inc., Gaithersburg, MD). The column
95
was washed with PBS and eluted with 0.2 M glycine (pH 2.8), after which pooled protein
96
fractions were dialyzed overnight against PBS. As controls for RSV F protein analysis, RSV-
5
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
Construct design
r Fo
73
International Immunology
Page 6 of 31
97
infected cell lysates were used. These were from HEp-2 cells that were infected with RSV-A2 in
98
PBS at room temperature for 1 hr. After washing, cells were cultured in Dulbecco’s modified
99
eagle’s medium (Cambrex Bio Science Walkersville, Inc, Walkersville, MD) supplemented with
101
lysed with TNE buffer (10 mM Tris [pH 7.4], 150 mM NaCl, 0.5% NP-40, and 1 mM EDTA).
102
Supernatants were clarified by centrifugation (15,000 x g, 10 min).Western blots were
103
performed with test and control samples by separating proteins on SDS polyacrylamide gels
104
under non-reducing conditions and transferring proteins to an Immobilon membrane (Millipore,
105
Danvers, Mass.). Membranes were treated with 5% milk/TBST (Tris buffered saline plus 0.5%
106
Tween), incubated with an RSV F-specific monoclonal antibody (antibody 1269, kindly provided
107
by Dr. Coelingh, NIAID(17)). Development was with a goat anti-mouse IgG (H+L) horse radish
108
peroxidase (HRP) conjugate (BioRad, Hercules, CA) followed by a SuperSignal West Pico
109
Chemiluminescent (HRP) Substrate (Pierce, Rockford, Il).
er
Pe
110
Re
Immunizations and RSV challenges
112
Immunizations and RSV challenges were conducted as described previously (16;18-20). Cotton
113
rats (Sigmodon hispidus; Harlan Sprague Dawley, Indianapolis, IN) were grouped (up to 5
114
animals per group) to receive SeVRSV-Fs or unmanipulated SeV intranasally (i.n., 2 x106
115
PFU/animal). Additional controls were with PBS alone. After five weeks, animals were
116
challenged i.n. with RSV-A2 at a dose of 1.5 x 106 PFU/cotton rat. Unless described as
117
preliminary, experiments were repeated to ensure reproducibility.
ew
vi
111
118 119
Enzyme-linked immunosorbent assay (ELISA)
120
ELISAs were described previously (16). Briefly, ELISA microtiter plates were coated with 1
121
ug/ml purified F protein for overnight incubation. Plates were blocked with PBS/3% bovine
122
serum albumin (BSA, Sigma, St Louis, MO), after which serially diluted serum samples from the
6
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
glutamine, gentamicin and 10% fetal calf serum. Three days later, cells were harvested and
r Fo
100
Page 7 of 31
International Immunology
123
test and control cotton rats were added and incubated for at least 1 hr at 37 °C. Plates were
124
washed and incubated with rabbit anti–cotton rat antibody (Virion Systems, Rockville, MD) for
125
30 min at room temperature. Plates were developed by incubation with an anti-rabbit IgG-
126
horseradish peroxidase conjugate (diluted in PBS/1% BSA, Bio-Rad, Hercules, CA, Cat# 170-
127
6515) for 30 min, followed by 2,2′-azino-bis-(3-ethylbenzthiazolinesulfonic acid) (ABTS,
128
Southern Biotechnology Associates, Inc, Birmingham, AL). Plates were read at O.D. 405 nm.
129
131
Neutralizing activity was measured using an RSV plaque assay as described previously (16).
132
Serially diluted test and control sera were mixed with RSV (100-150 PFU/well) in EMEM
133
(Cambrex Bio Science Walkersville, Inc, Walkersville, MD) for 1 hr at 37°C. The virus-serum
134
mixtures were inoculated onto HEp-2 cell monolayers in 12-well plates, except for virus RSV B1
135
that was inoculated onto Vero cells. Plates were incubated for 1 hr (37°C, 5% CO2) and then
136
overlaid with medium supplemented with glutamine, antibiotics, 10% fetal calf serum and 0.75%
137
methylcellulose (Fisher Scientific, Fair Lawn, NJ). After culture for 5-6 days (37°C, 5% CO2), the
138
methylcellulose was removed. Cells were treated with formalin phosphate and stained with
139
hematoxylin and eosin. Plaques were counted visually. Inhibition was defined as the percentage
140
reduction of plaques comparing post-vaccine sera with pre-vaccine sera. Viruses included RSV
141
A2 (Laboratory isolate, Type A RSV, American type culture collection, ATCC, Rockville, MD),
142
RSV B1 (Laboratory isolate, Type B RSV, ATCC), VR 1580 (Type B RSV, ATCC), K1014, and
143
K1016 (primary isolates, Type A RSV, kindly provided by Dr. J. DeVincenzo, Le Bonheur
144
Children’s Hospital, Memphis, TN).
er
Pe
ew
vi
Re
145 146
Peptide synthesis and IFN-gamma ELISPOT assays
147
The T cell assays were conducted as described previously (16). Briefly, overlapping F peptides
148
(derived from the RSV A2 F sequence) were prepared by the Hartwell Center for Bioinformatics
7
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
Neutralization assays
r Fo
130
International Immunology
Page 8 of 31
149
and Biotechnology at St. Jude Children’s Research Hospital. Peptides were generally 15 amino
150
acids in length. Peptide sequences were initiated at 5 amino acid intervals to cover the entire
151
length of the RSV F protein. Twenty two peptide pools were made for T cell testing.
152
154
goat anti-cotton rat IFN-gamma antibody (R & D Systems, Minneapolis, MN) and incubated
155
overnight at 4°C. Plates were washed, and incubated for at least 1 hr at 37°C with complete
156
tumor medium (CTM (21;22), a modified Eagle’s medium [Invitrogen, Grand Island, NY]
157
supplemented with 10% fetal calf serum, dextrose [500 µg/m], glutamine [2 mM], 2-
158
mercaptoethanol [3 x 10-5 M], essential and non-essential amino acids, sodium pyruvate,
159
sodium bicarbonate, and antibiotics). Mediastinal lymph node cells were harvested from cotton
160
rats 10 days after vaccination. Cells were suspended in CTM and added to wells (106 cells/well)
161
containing the peptide pools. The final concentration of each peptide was approximately 10 µM.
162
As positive controls, cells were stimulated with 4 µg/ml Con A (Sigma-Aldrich, St. Louis, MO).
163
The plates were incubated for 48 hr at 37°C. Plates were then washed with PBS followed by
164
PBS wash buffer (PBS with 0.05% Tween 20). Plates were next incubated with biotinylated goat
165
anti-cotton rat IFN-gamma antibody (100 µl of 0.5 µg/ml; R & D Systems, Minneapolis, MN) in
166
PBS containing 0.05% Tween 20 and 1% FCS (incubated at 37°C for at least 2 hr) followed by
167
streptavidin-conjugated alkaline phosphatase (Cat# D0396, DAKO, Copenhagen, Denmark)
168
diluted 1:500 in PBS wash buffer. After 1 hr, plates were rinsed with wash buffer and water and
169
spots were developed with 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tetrazolium alkaline
170
phosphatase substrate (Sigma-Aldrich). Spots were counted with an Axioplan 2 microscope
171
and software (Carl Zeiss, Munich-Hallbergmoos, Germany).
er
Pe
ew
vi
Re
172 173
Histopathology studies
8
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
Multiscreen-hemagglutinin filtration plates (Millipore, Bedford, MA) were coated with 3.3 µg/ml
r Fo
153
Page 9 of 31
International Immunology
175
described (16;18;20;23). Briefly, lungs were inflated via the trachea with 10% neutral buffered
176
formalin. They were then submerged in formalin for overnight fixation. Tissue was embedded in
177
paraffin, sectioned at 4 µm, and stained with hematoxylin and eosin. Lung pathology was
178
scored on the basis of three parameters: peribronchiolitis (inflammatory cells around small
179
airways), alveolitis (inflammatory cells within alveolar spaces), and interstitial pneumonitis
180
(inflammatory cell infiltrates and thickening of alveolar walls). Scoring was performed by a
181
pathologist using a scale of 0 (no inflammation) to 4 (maximum inflammation within the
182
experiment), defined independently for each parameter. The pathologist was blinded to the
183
origin of tissues from test or control groups while scoring.
Pe
184
RSV titration in cotton rat lungs after challenge
186
RSV measurement in cotton rat lungs has been described previously (16). Three days after i.n.
187
RSV challenge, cotton rats were sacrificed. Lungs were harvested and homogenized on ice with
188
a mechanical Dounce homogenizer (PowerGen125 PCR Tissue Homogenizing kit; Fisher
189
Scientific). Homogenates were centrifuged (approximately 1500 x g, 10 min) and supernatants
190
were collected. They were stored frozen prior to testing. For virus titration, serially diluted
191
supernatants were inoculated onto HEp-2 cell monolayers (except for virus B1 that was
192
measured on Vero cells) in 12-well plates in EMEM. After 1 hr at 37 °C, 5% CO2, the wells were
193
overlaid with medium supplemented with glutamine, antibiotics, 10% fetal calf serum and 0.75%
194
methylcellulose. Cells were incubated for 5-6 days (37°C, 5% CO2), after which the
195
methylcellulose was removed. Cells were fixed with formalin phosphate and the plates were
196
stained with hematoxylin and eosin for plaque counting. The total pulmonary virus burden was
197
expressed as the plaque count (PFU) per cotton rat.
er
185
ew
vi
Re
198 199
SeVRSV measurement in cotton rat lungs after vaccination
9
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
Five days after RSV challenge, lungs were prepared for histological analyses as previously
r Fo
174
International Immunology
Page 10 of 31
Replication-competent SeVRSV was monitored in cotton rat lungs by plaque count as described
201
previously (24) after i.n. vaccination with 2 x 106 PFU virus. Briefly, lungs were homogenized in
202
5 ml PBS, centrifuged briefly to remove debris, and plated using serial 10-fold dilutions on
203
adherent LLC-MK2 cell monolayers in six-well plates. After incubation at 33°C for 1 hour, the
204
inoculum was removed and replaced with an overlay containing acetylated trypsin (5 µg/ml).
205
Plates were inverted and cultured for 4-6 days at 33°C. A second overlay was added containing
206
neutral red. Plaques were visualized and counted after an additional incubation of cultures for 1-
207
3 days.
209
211
er
212
Pe
210
r Fo
208
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
200
ew
vi
Re 10
Page 11 of 31
213
International Immunology
RESULTS
214
216
Recombinant SeV particles were designed to express a truncated RSV F protein as described
217
in the Materials and Methods section and as illustrated in Figure 1A. To confirm F protein
218
secretion by infected cells, gel electrophoresis and Western blotting was performed with
219
allantoic fluids from infected eggs. The secreted F protein from SeVRSV-Fs was predictably
220
smaller than the full-length RSV F from RSV particles (data not shown). In a preliminary
221
experiment to ensure vaccine growth, three cotton rats received SeVRSV-Fs by the i.n. route
222
and were examined on days 1, 3, 5, and 7 for vaccine amplification. As shown in Figure 1B,
223
virus peaked on day 3 after inoculation and was cleared by day 7 in all animals.
er
Pe
224
Vaccination with SeVRSV-Fs elicits RSV-specific binding and neutralizing antibodies in
226
cotton rats.
227
To determine whether the SeVRSV-Fs vaccine induced an antibody response, we inoculated
228
groups of five cotton rats by the i.n. route (2 x 106 PFU/cotton rat). Animals injected with PBS or
229
unmodified SeV were used as controls to define baseline activities. Blood was collected before
230
and four weeks after vaccination for testing of RSV- specific antibodies. Post-vaccine antibodies
231
bound antigen in an ELISA (Figure 2A). Pooled serially diluted serum samples were also tested
232
for neutralization by incubating with RSV-A2 prior to infection of cultured cells. Sera were found
233
to be neutralizing against this homologous virus (Figure 2B). Neutralizing antibodies were also
234
detected against non-homologous viruses, including clinical isolates of the RSV A subtype
235
(Figures 2C and 2D), a laboratory isolate of the RSV B subtype (B1, Figure 2E), and a clinical
236
isolate of the RSV B subtype (Figure 2F, VR1580).
ew
vi
Re
225
237 238 11
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
RSV F protein secreted by SeVRSV-Fs-infected cells.
r Fo
215
International Immunology
Page 12 of 31
240
To examine T cell responses, we performed IFNγ-ELISPOT assays. Mediastinal lymph nodes
241
(MLN) were harvested from cotton rats ten days after vaccination with SeVRSV-Fs or
242
unmanipulated, non-recombinant virus (Sendai). Cell suspensions prepared from MLN were
243
combined for each experimental group (3 or more animals per group). To examine IFNγ
244
expression, we incubated cell suspensions with pools of overlapping peptides representing the
245
entire RSV F protein (listed in Table 1). As shown in Figure 3, SeVRSV-Fs induced IFNγ-
246
producing T cells against RSV peptide pools.
247
SeVRSV-Fs confers complete protection against RSV challenge in the cotton rat model.
249
SeVRSV-Fs was next tested for protection of animals from RSV challenge. Five weeks after
250
vaccination, animals were challenged i.n. with RSV-A2. Three days after challenge, animals
251
were sacrificed and lungs were collected for measurement of RSV burden. As shown in Figure
252
4, animals vaccinated with a single dose of SeVRSV-Fs were completely protected from RSV
253
challenge. In contrast, all animals inoculated with control SeV and subsequently challenged with
254
RSV were infected.
er
Pe
248
ew
vi
Re
255 256 257
No enhanced immunopathology upon RSV challenge following vaccination with SeVRSV-
258
Fs.
259
Cotton rats inoculated with SeVRSV-Fs or PBS and were sacrificed five days after challenge
260
with RSV to test for immunopathology. The lungs were harvested, perfused with formalin,
261
sectioned, and stained with hematoxylin and eosin for histologic examination. Peribronchiolitis,
262
interstitial pneumonitis, and alveolitis were detected in control cotton rats five days after i.n. RSV
263
challenge, and were not enhanced in vaccinated animals. Scores for test and control animals
264
are shown in Table 2.
12
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
SeVRSV-Fs vaccination elicits RSV-specific T cell responses
r Fo
239
Page 13 of 31
International Immunology
266
This report describes the immunization of cotton rats with SeVRSV-Fs, an SeV-based vaccine
267
produced by reverse genetics to deliver a truncated RSV F gene to mammalian cells. Cells
268
infected with SeVRSV-Fs expressed a secreted, unconstrained RSV F protein. Vaccination
269
generated binding and neutralizing antibody responses as well as T cell responses toward RSV.
270
Following a single i.n. inoculation with SeVRSV-F, cotton rats were fully protected from RSV
271
infection after challenge. Immunopathology was not enhanced after RSV challenge in animals
272
that were vaccinated with SeVRSV-Fs compared to controls.
273
Our previous studies with SeV vaccines have demonstrated the induction of rapid and durable B
275
and T cell responses systemically and at upper respiratory tract (URT) and lower respiratory
276
tract (LRT) mucosal surfaces (25;26). An unmanipulated SeV was previously demonstrated to
277
confer complete protection against challenge with its cousin, human parainfluenza virus-type 1
278
(hPIV-1), in African green monkeys (24). Unmanipulated SeV is currently in clinical trials. It has
279
been well tolerated in adults (27) and 3-6 year old children (data not shown), paving the way for
280
similar studies with recombinant SeV vaccine products.
er
ew
vi
Re
281
Pe
274
282
A recombinant SeV that expresses the full length RSV F has already been tested in cotton rats
283
and has been shown to protect against challenges with both RSV A and B isolates (16).
284
Protection is demonstrated even in the presence of passively transferred RSV-specific and SeV-
285
specific neutralizing antibodies in a cotton rat maternal antibody model (28). The vaccine safely
286
protects against RSV challenge in African green monkeys (29). Sendai viruses are attractive
287
vaccine candidates, in part because of their appealing immunogenicity and safety profiles, and
288
in part because they maintain genetic stability when successively passaged and amplified to
289
high titers in hen’s eggs or mammalian (e.g. Vero) cell cultures (data not shown). The study
13
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
DISCUSSION
r Fo
265
International Immunology
Page 14 of 31
290
described here demonstrates that an SeV recombinant expressing a truncated, secreted RSV F
291
protein can serve as a successful vaccine against RSV.
292
294
protein in a vaccine. Some researchers suggest that RSV F should be constrained in its pre-
295
fusion form, based on arguments that: (i) there are important antibody determinants (e.g. D25,
296
AM22, 5C4) that exist only within the pre-fusion form of F, and (ii) the majority of neutralizing
297
antibodies in RSV immune human sera bind to the pre-fusion F molecule (12;13). Opposing
298
arguments are that: (i) several important antigenic determinants including 101F exist on post-
299
fusion F proteins (30;31), (ii) the Palivizumab neutralizing monoclonal antibody binds post-fusion
300
F and provides excellent prophylaxis against RSV A and B isolates in clinical practice (7;9), (iii)
301
antibodies clear viruses by a variety of mechanisms that function both before and after the
302
fusion of virus to its target cell (32-34), and (iv) the artifactual manipulations required to
303
constrain soluble F protein in its pre-fusion form may inadvertently alter important native
304
antigenic epitopes, a situation that was associated with significant morbidity and mortality when
305
a formalin-treated RSV vaccine was tested in children (11;35). Our results support the argument
306
that RSV F constraint is not required, consistent with data from Swanson et. al. who tested a
307
subunit F protein as an RSV vaccine in mice (36).
er
Pe
ew
vi
Re
308 309 310
It is important to consider that antibodies protect against virus infections using sophisticated
311
mechanisms that function before and after virus fusion. In the context of HIV-1 infection, for
312
example, antibodies can inhibit infection well after virus has penetrated the cell. This is in part
313
due to the shuttling of virus particles to proteasomes when antibodies bind viral antigens and
314
TRIM-21 (32). Antibodies are also well known for their support of antibody-dependent cell-
315
mediated cytotoxicity (ADCC), a mechanism by which cytotoxic cells (NK or CD8+ T cells) are
14
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
Our results address a recent question in the RSV field that concerns the configuration of RSV F
r Fo
293
Page 15 of 31
International Immunology
316
linked to their virus-infected targets (37;38). Moreover, antibody-antigen complexes and free
317
antibodies modulate inflammation by up- and down-regulation of innate and adaptive immune
318
cells (39-41). Together, there are a plethora of antibody-mediated mechanisms that counter
319
virus infection and disease, and that are clearly not dependent on the pre-fusion structure of
320
RSV F. To date, the most clinically-relevant propohylactic regimen against RSV infections is
321
with an antibody that binds the post-fusion form of RSV F (8).
322 Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
r Fo
323
In conclusion, we have demonstrated that SeVRSV-Fs confers protection against RSV
325
challenge in a cotton rat model and that constraint of the secreted F protein in its pre-fusion
326
form is not required. Clearly, a number of vaccine products that present soluble F in an
327
unconstrained form remain viable vaccine candidates for the protection of humans from RSV
328
infection.
er
Pe
324
ew
vi
Re 15
International Immunology
Page 16 of 31
329
ACKNOWLEDGEMENTS
330
We thank Virion Systems for providing cotton rat antibody reagents. We thank Dr. J.
331
DeVincenzo, Le Bonheur Children’s Hospital, Memphis, TN for kindly providing primary RSV
332
isolates for neutralization assays. This work was supported by NIH NIAID P01 AI054955, NIH
333
NIAID R01 AI088729, NIH NCI P30-CA21765, and the American-Lebanese Syrian Associated
334
Charities (ALSAC).
335 Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
337
Conflict of interest statement: The authors declared no conflict of interests.
r Fo
336
er
Pe ew
vi
Re 16
Page 17 of 31
338
International Immunology
Table 1. RSV F peptide sequences for T cell testing
er
Pe
339
Re
Overlapping peptides were synthesized to cover the full sequence of secreted F. The
341
synthesized peptides were generally 15 amino acids in length and were initiated at 5-amino acid
342
intervals to cover the entire length of the RSV F protein. Amino acids represented in each
343
peptide pool are shown. Pools 21 and 22 included amino acids in the transmembrane and
344
cytoplasmic regions of F that were not included in the vaccine (as indicated by dashed lines).
345
Each pool comprised approximately 10 peptides. Seven peptides were not synthesized due to
346
technical difficulties: ITTILTAVTFCFASG (pool 1); KIKSALLSTNKAVVS (pool 7);
347
NFYDPLVFPSDEFDA (pool 20); and STTNIMITTIIIVII, MITTIIIVIIVILLS, IIVIIVILLSLIAVG, and
348
VILLSLIAVGLLLYC (pool 21).
ew
vi
340
349 350
17
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
RSV F Sequences MELLILKANAITTILTAVTFCFASGQNITEEFYQS QNITEEFYQSTCSAVSKGYLSALRTGWYTSVITIE GWYTSVITIELSNIKENKCNGTDAKVKLIKQELDK VKLIKQELDKYKNAVTELQLLMQSTPPTNNRARRE PPTNNRARRELPRFMNYTLNNAKKTNVTLSKKRKR NVTLSKKRKRRFLGFLLGVGSAIASGVAVSKVLHL GVAVSKVLHLEGEVNKIKSALLSTNKAVVSLSNGV KAVVSLSNGVSVLTSKVLDLKNYIDKQLLPIVNKQ KQLLPIVNKQSCSISNIETVIEFQQKNNRLLEITR KNNRLLEITREFSVNAGVTTPVSTYMLTNSELLSL MLTNSELLSLINDMPITNDQKKLMSNNVQIVRQQS NNVQIVRQQSYSIMSIIKEEVLAYVVQLPLYGVID VQLPLYGVIDTPCWKLHTSPLCTTNTKEGSNICLT TKEGSNICLTRTDRGWYCDNAGSVSFFPQAETCKV FFPQAETCKVQSNRVFCDTMNSLTLPSEINLCNVD PSEINLCNVDIFNPKYDCKIMTSKTDVSSSVITSL DVSSSVITSLGAIVSCYGKTKCTASNKNRGIIKTF NKNRGIIKTFSNGCDYVSNKGMDTVSVGNTLYYVN SVGNTLYYVNKQEGKSLYVKGEPIINFYDPLVFPS NFYDPLVFPSDEFDASISQVNEKINQSLAFIRKSD QSLAFIRKSDELLHNVNAGKSTTN---------------------------------------------
r Fo
Peptide Pool 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
International Immunology
351
Page 18 of 31
Table 2. Histopathology following vaccination and subsequent RSV challenge
352
SeVRSV-Fs SeVRSV-Fs SeVRSV-Fs SeVRSV-Fs SeVRSV-Fs
2 2 2 0 1
2 1 1 0 1
6 7 8 9 10
PBS control PBS control PBS control PBS control PBS control
4 4 2 4 3
2 3 3 4 3
353
Bronchiolits
Alveolitis
Interstitial Pnemonitis 1 1 1 0 1 2 2 3 4 3
Cotton rats were vaccinated with SeVRSV-Fs or PBS and then challenged with RSV.
355
Five days later, cotton rats were scored for immunopathology. The most severe disease
356
in each category within the experiment was scored as ‘4’ and the absence of disease
357
was scored as ‘0’.
er
ew
vi
Re
358
Pe
354
18
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
Vaccine Group
r Fo
Animal Number 1 2 3 4 5
Page 19 of 31
International Immunology
359
FIGURE LEGENDS
360
Figure 1. Design and characterization of recombinant SeV expressing RSV F as a
361
secreted protein.
362 A. The diagram illustrates the design of SeVRSV-Fs. A unique NotI restriction enzyme site was
364
created in the non-coding region of the HN gene of the full genome SeV cDNA for insertion of
365
the RSV Fs gene. The RSV gene was isolated from RSV-A2 with RT-PCR, digested with NotI
366
and cloned into the NotI site of pSV(E). T7=T7 promoter; ribo=hepatitis delta virus ribozyme
367
sequence.
368
B. In a preliminary experiment, cotton rats were vaccinated i.n. with 2 x 106 PFU SeVRSV-Fs
369
and examined throughout a one week period for vaccine virus amplification in the lower
370
respiratory tract, by testing homogenized lungs in plaque assays on LLC-MK2 cells. Each
371
symbol represents virus measurements from a different animal. ND=Not detected.
r Fo
Downloaded from http://intimm.oxfordjournals.org/ at Laurentian University on December 9, 2014
363
er
Pe
372
Re
373
Figure 2. Cotton rats primed with SeVRSV-Fs generate RSV-specific neutralizing
375
antibodies.
376
Groups of five cotton rats were inoculated with 2 x 106 SeVRSV-Fs or unmodified SeV. Sera
377
were collected after four weeks and pooled for each group of animals. (A) Sera were diluted
378
1:250 and 1:1000 for testing in an ELISA for RSV-F-specific binding antibody. (B-F) Pooled
379
serum samples from cotton rats inoculated with SeVRSV-Fs or unmodified SeV were tested for
380
inhibition of RSV growth in a plaque assay. Results are the mean percentage plaque reduction
381
for post-vaccine test sera compared to pre-vaccine control sera from the same animal set.
382
Standard error bars are shown. Neutralization experiments were conducted with different
383
viruses including RSV A2 (laboratory type A isolate, panel B), K1014 (primary type A isolate,
384
panel C), K1016 (primary type A isolate panel D), B1 (laboratory type B isolate, panel E), and
ew
vi
374
19
International Immunology
Page 20 of 31
385
VR1580 (primary type B isolate, panel F). Test and control samples were plated in triplicate in
386
ELISAs and in neutralization assays. All ELISAs and neutralization assays revealed statistically
387
significant differences between vaccinated and control samples using unpaired T tests (p3 animals per group) were tested against each of 22
395
peptide pools or against no peptide. Spot-forming cells (SFC) were counted per 106 plated
396
cells. Values are means of duplicate wells with standard errors. Asterisks indicate peptide pools
397
for which mean test results were ≥4x control values. Unpaired T tests were conducted using
398
GraphPad Prism software and showed that results for pools 2, 3, and 10 were all statistically
399
significant (p