Journal
of General Virology (1998), 79, 1935–1944. Printed in Great Britain
...................................................................................................................................................................................................................................................................................
In vivo resistance to simian immunodeficiency virus
superinfection depends on attenuated virus dose
Martin P. Cranage,1 Sally A. Sharpe,1 Adrian M. Whatmore,1† Natasha Polyanskaya,1
Stephen Norley,2 Nicola Cook,1 Sharon Leech,1 Michael J. Dennis1 and Graham A. Hall1
1
2
Centre for Applied Microbiology and Research, Porton Down, Salisbury SP4 0JG, UK
Paul-Ehrlich-Institute, Paul Ehrlich Strasse 51–59, Langen, Germany
Infection of macaques with attenuated simian
immunodeficiency virus (SIV) induces potent superinfection resistance that may be applicable to the
development of an AIDS vaccine but little information exists concerning the conditions necessary
for the induction of this vaccine effect. We report
that only a high dose of attenuated SIVmac protected macaques against intravenous challenge with
more virulent virus 15 weeks after primary infection.
Three of four animals given 2000–20 000 TCID50 of
SIVmacC8, a molecular clone of SIVmac251(32H)
with a 12 bp deletion in the nef gene, essentially
resisted superinfection with uncloned SIVmac. In
two animals challenge virus was never detected by
PCR and in one animal challenge virus was detected
on one occasion only. Although animals given
Introduction
The development of an effective AIDS vaccine requires
investigation of multiple strategies. This task is complicated by
the pathobiology of immunodeficiency virus infection and the
high degree of genetic variability exhibited by these viruses.
Simian immunodeficiency virus (SIV) shares many of the
biological properties of HIV and infection of macaques with
SIV has proved invaluable for modelling AIDS vaccine
strategies. One of the most promising observations to emerge
from the use of this model is the finding that macaques infected
with live attenuated SIV are resistant to superinfection (Daniel
et al., 1992). Importantly, such protection works not only
against intravenous challenge with cell-free virus, but also
Author for correspondence : Martin Cranage.
Fax ­44 1980 611310. e-mail martin.cranage!camr.org.uk
† Present Address : School of Biological Sciences, University of
Warwick, Coventry CV4 7AL, UK.
2–200 TCID50 of attenuated virus were superinfected they were spared from the loss of CD4 cells
seen in infected naive controls. Protection from
superinfection did not correlate with immune responses, including the levels of virus-specific antibodies or virus-neutralizing activity measured on
the day of challenge ; although, after superinfection
challenge, Nef-specific CTL responses were detected only in animals infected with high doses of
attenuated SIV. Unexpectedly, cell-associated virus
loads 2 weeks after inoculation were significantly
lower in animals infected with a high dose of
attenuated SIV compared to those in animals infected with a low dose. Our results suggest that the
early dynamics of infection with attenuated virus
influence superinfection resistance.
against challenge with virus-infected cells (Almond et al., 1995)
and against challenge via mucosal surfaces (Putkonen et al.,
1996 ; Cranage et al., 1997 ; Miller et al., 1997).
The use of a live attenuated immunodeficiency virus as a
vaccine in man is controversial, principally because of safety
concerns. The observation that infant macaques given SIV that
is attenuated for growth in adult macaques can develop disease
(Baba et al., 1995) underscores the importance of host factors as
well as viral determinants in the definition of virulence. Despite
these important concerns, the potency of protection elicited by
infection with attenuated virus cannot be ignored. The
parameters and mechanisms of protection with live attenuated
virus are still to be fully defined : an essential step if such a
strategy is to be used safely in man.
Our previously reported studies and those of our collaborators, as part of a European Community Concerted Action
Programme, have utilized the SIVmacC8 molecular clone, a
naturally occurring attenuated variant of SIVmac251-32H
derived and characterized by Rud et al. (1994). The principle
0001-5508 # 1998 SGM
BJDF
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
M. P. Cranage and others
difference between this virus and the more virulent SIVmacJ5
molecular clone is a 12 bp deletion in the nef}3« LTR. Infection
of macaques with SIVmacC8 results in an early, high, cellassociated viraemia followed by a rapid clearance of virus from
the peripheral circulation, so that by 12 weeks after infection
virus can only be recovered intermittently from peripheral
blood mononuclear cells (PBMC) (Whatmore et al., 1995).
Infection of macaques with this virus conferred resistance to
superinfection with closely related and heterologous virus
following intravenous challenge (Whatmore et al., 1995) and
rectal challenge (Cranage et al., 1997). Furthermore, a high
degree of protection against intravenous challenge with cellfree virus has been seen as early as 10 weeks after infection
(Norley et al., 1996).
To investigate the relationship between dose of attenuated
virus, the dynamics of infection and the outcome of superinfection challenge early after vaccination we inoculated ten
macaques in pairs with graded doses of SIVmacC8. Fifteen
weeks after primary inoculation, animals were challenged
intravenously with SIVmac220, a cell-free virus stock derived
ex vivo from a macaque infected with the SIVmacJ5 molecular
clone (Polyanskaya et al., 1997). We show here that only
animals primed with high doses of the attenuated virus resisted
superinfection. Furthermore, cell-associated virus loads in these
animals, 2 weeks after inoculation, were lower than in those
animals given low doses of attenuated virus. Protection was
correlated neither with humoral nor cellular immune responses,
although Nef-specific CTL were detectable after challenge only
in animals given high doses of attenuated SIV.
Methods
+ Virus stocks and animal inoculations. The derivation of
SIVmacC8 from the uncloned stock of SIVmac251(32H) and its
characterization have been described previously (Rud et al., 1994). This
molecular clone has an attenuated phenotype in vivo (Rud et al., 1994 ;
Whatmore et al., 1995 ; Almond et al., 1995), differing from the more
virulent SIVmacJ5 molecular clone by a 12 bp deletion in the nef}3«LTR
(nucleotides 9501–9512) and six nucleotide substitutions. Two of the
three substitutions in the overlapping nef}3« LTR result in conservative
amino acid changes and one is a silent change. The inoculum stock,
designated 9}90, was produced from the cell-free supernatant of infected
C8166 cells as described previously (Rud et al., 1994) and has a titre of 10&
TCID }ml on C8166 cells.
&!
The derivation of the cell-free SIVmac220 challenge stock, designated
6}94, from the spleen of a rhesus macaque infected with SIVmacJ5 was
described by Polyanskaya et al. (1997). The stock has a titre of 10%
TCID }ml on C8166 cells and the MID in rhesus macaques is equal to
&!
&!
or less than 0±3 TCID .
&!
Juvenile rhesus macaques used in this study were bred within the
United Kingdom and housed according to the Home Office Code of
Practice. Ketamine hydrochloride sedation was used for all procedures
requiring the removal of animals from their cages. Animals were
challenged with virus by inoculation into the femoral vein.
+ Virus isolation and cell-associated virus loads. PBMC were
isolated from heparinized blood by centrifugation onto Ficoll–Paque
(Pharmacia). Virus isolation was assayed by co-cultivation of PBMC with
BJDG
C8166 cells either at a single concentration of 10' PBMC or additionally
in a 5 fold dilution range from 4¬10& cells to 130 cells. PBMC}C8166
cell cultures were put up in duplicate in 25 cm# flasks and replenished
with fresh medium and C8166 cells every 3–4 days maintaining the total
volume at approximately 15 ml. Cultures were examined regularly for
cytopathic effect (CPE) and where none was evident cultures were
maintained for 28 days. Cultures not resulting in CPE were tested for the
presence of virus antigen by immunofluorescent staining of acetone–
methanol-fixed cells using an antiserum from an SIV-infected macaque.
Fifty percent end-points were calculated using the Ka$ rber formula and
cell-associated virus loads expressed as the number of virus-infected cells
per 10' PBMC.
+ PCR. DNA was extracted from PBMC essentially as described by
Kitchin et al. (1990). SIV nef sequences were amplified by nested PCR
using primer sets SE9044N}SN9866C and SN9272}SN9763C. PCR
products having the full-length nef sequence (SIVmac220) were distinguished from those with the truncated sequence encoded by SIVmacC8
by restriction enzyme digestion with RsaI and the specificity of PCR
products was confirmed by Southern blotting as described by Rose et al.
(1995).
+ Assay of anti-SIV antibodies and virus infectivity neutralization. Binding antibody titres against SIV were determined by ELISA
using standard methods. Plates were coated with recombinant SIVmac251 gp140 (Repligen ; MRC AIDS Reagent Project ARP625.1) or
recombinant SIVmac251 p27–GST fusion protein (MRC AIDS Reagent
Project ARP643) at predetermined optimal concentrations. Immunoblots
were prepared by standard methods using lysates of crude virus pellets
prepared from the culture supernatants of SIVmac251(32H)-infected
C8166 cells that were concentrated approximately 1000-fold by highspeed centrifugation. Binding of SIV-specific antibodies to SIV-coated
membranes was detected with rabbit anti-monkey IgG HRP and
visualized by enhanced chemiluminescence (ECL, Amersham).
Virus infectivity serum neutralizing activity was determined using a
yield reduction assay based on the comparison of virus titre in the
presence and absence of constant serum dilution. Briefly, 25 µl of a 1}50
dilution of each serum was aliquoted into each well of a microtitre plate ;
25 µl aliquots of SIVmac32H, titrated in a 2±5-fold dilution series, were
then added to each plate with eight replicates per dilution. After mixing,
plates were incubated at 37 °C for 1 h after which 150 µl of a suspension
of C8166 cells (2000 cells per well) was added. Plates were incubated for
7 days at 37 °C before detection of virus by antigen capture ELISA. Cells
were lysed by addition of 20 µl of 2 % (v}v) Tween 20 per well before
transferring 50 µl of the lysates to the corresponding wells of ELISA
plates coated with an anti-SIVmac Gag monoclonal antibody. The
presence of trapped viral antigen was determined by addition of diluted
plasma from an SIVmac-infected rhesus macaque followed by anti-human
IgG–peroxidase conjugate and finally substrate}chromogen buffer. The
number of positive wells was used to calculate the TCID . The yield
&!
reduction was calculated as the virus titre in the absence of serum divided
by the titre in the presence of serum.
+ PBMC phenotype enumeration. A whole blood lysis method
was used to analyse PBMC for CD4+ (OKT4, Ortho Diagnostics) and
CD8+ (DK25, Dako). Fifty µl volumes of heparinized whole blood were
combined with 15 µl of monoclonal or isotype-matched control antibody,
mixed and incubated at room temperature for 1 h. The cells were washed
once with 3 ml PBS, resuspended in 1 ml of 4 % (v}v) Immunolyse in PBS
(Coulter) and mixed vigorously. After 30 s, 3 ml PBS containing 1 %
(w}v) formaldehyde was added and the cells pelleted (8 min at 400 g).
Cells were resuspended in 1 ml of this medium and analysed using an
EPICS Profile I flow cytometer (Coulter). The lymphocyte population
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
SIV vaccine dose effect
was delineated by forward scatter versus side scatter and a bitmap placed
around this population was used to gate the fluorescence signal. Total
white cell counts and blood smear differential counts were determined
manually for each animal, allowing the calculation of the absolute
numbers of each lymphocyte subset.
Results
+ Assay of cytotoxic T cell activity. Ficoll–Hypaque isolated PBMC
were stimulated in vitro with autologous SIV-infected PHA blasts
(Gallimore et al., 1995). One-fifth of the isolated PBMC were prepared for
use as stimulator cells while the remaining cells were maintained
overnight at 37 °C. Stimulator cells were treated overnight with 1 %
(v}v) PHA (Gibco) and infected with SIVmacC8. The resulting blasts
were washed twice with serum-free RPMI, resuspended and inactivated
with mitomycin C (Sigma) at 25 µg}ml for 1 h at 37 °C. The inactivated,
SIV-infected PHA blasts were washed three times prior to addition to the
culture of unstimulated, autologous PBMC. On day 3, 10 Units of
Lymphocult T (Biostat) was added to the culture. Cultures were fed
subsequently every 3 days and assayed for CTL activity after 10–14
days. Later in the study the in vitro stimulation method was modified to
increase the efficiency of pCTL expansion. Autologous, SIV-infected,
PHA-induced blasts were prepared from PBMC isolated 7 days before
the cells to be stimulated. Blasts were prepared and infected as previously
and then cultured at 1¬10' cell}ml at 37 °C for 3–4 days. Cells were
resuspended at 10( cells}ml in medium containing one-tenth (v}v) 0±5 M
Tris pH 8±0 ; β-propiolactone (Sigma) was added to a final concentration
of 0±2 % and cells were incubated in the dark at 37 °C for 2 h. The reaction
was stopped by the addition of sodium azide to a final concentration of
0±056 M and cells were washed prior to addition to effector cell cultures.
The inactivated, SIV-infected PHA-induced blasts were combined
with freshly isolated PBMC at ratios between one to two and one to five.
On day two, Lymphocult T was added at 10 Units}ml. Cultures were fed
on day four and tested for CTL activity on day seven.
Target cells were prepared from herpesvirus papio-transformed
autologous B-lymphoblastoid cell lines (B-LCL) by overnight infection
with 5 p.f.u. per cell of recombinant vaccinia viruses expressing either SIV
Nef (VV 9011 MRC AIDS Reagent Programme, ARP274.1) or influenza
A virus nucleocapsid protein PB2 (a kind gift from F. Gotch, Chelsea &
Westminster Hospital, London, UK). The following day the cells were
labelled with 100 µCi of Na &"CrO (Amersham) for 1 h at 37 °C.
#
%
Effector cells were combined with labelled target cells at various effector
to target ratios ranging from 200 : 1 to 5 : 1 in 96-well U-bottom plates.
Following incubation at 37 °C for 4 h, 100 µl of supernatant was
harvested from each well and the radioactivity determined by γ-counting.
Total &"Cr incorporation was determined by incubation of targets with
2±5 % (v}v) Triton X-100 and spontaneous release determined after
incubation of targets with medium alone. The tests were performed in
triplicate and the geometric mean count was determined. CTL activity
was expressed as : percent specific release ¯ 100¬(experimental c.p.m.spontaneous c.p.m.)}(total c.p.m.-spontaneous c.p.m.).
Five pairs of macaques were inoculated intravenously with
10-fold dilutions of a cell-free virus stock of the attenuated
molecular clone SIVmacC8 (Table 1). Virus was recovered 2
and 4 weeks later from the PBMC of all animals receiving
20 000 ®20 TCID . At weeks 13 and 15 six of these animals
&!
had become virus isolation negative and in the remaining two
virus was isolated on a single occasion only. Infection of all of
these animals was confirmed by PCR detection of SIVmacC8
proviral DNA. PCR also revealed that one of the animals (15S)
that had received 2 TCID was infected, despite being virus
&!
isolation negative.
Cell-associated virus loads were measured by limiting
dilution of PBMC and cultivation with the C8166 cell line, a
sensitive indicator of SIV infection. As shown in Fig. 1, a peak
of cell-associated viraemia was detected 2 weeks after infection
and virus loads had decreased by week 6 in all animals. The
mean virus load at 2 weeks in animals inoculated with 20 000
(4S, 14S) and 2000 (33S, 21S) TCID was significantly lower
&!
than in animals inoculated with 200 (23S, 33R) or 20 (22S, 7S)
TCID (131 virus infected cells per 10' PBMC compared to
&!
4627 virus infected cells per 10' PBMC ; P ¯ 0±04, Student’s ttest). At week 6, no significant difference in virus load was seen
between these two groups (6±7 infected cells per 10' PBMC
compared to 7±5 infected cells per 10' PBMC).
Numbers of CD4 and CD8 cells in the peripheral circulation
increased following inoculation of animals with SIVmacC8 and
then declined so that mean levels at the time of superinfection
challenge were similar to pre-infection levels (Fig. 2). This
pattern was seen in all animals inoculated with the attenuated
virus, including animal 15S from which virus could not be
recovered but which was positive by PCR, and animal 13S,
which showed no evidence of infection. When data from
individual animals were analysed there was no obvious
relationship between CD4 or CD8 cell numbers and the dose
of attenuated virus with which each animal was infected. On
the day of superinfection challenge, the mean CD4 and CD8
cell numbers were similar in vaccinated and control groups
(Fig. 2). The mean counts in naive controls were low when
compared to baseline mean values determined from three
separate analyses performed prior to challenge (1±46³0±55
¬10' CD4 cells}ml ; 1±2³0±52¬10' CD8 cells}ml). Thus,
overall, all animals, regardless of infection status, had depressed
cell counts on the day of challenge.
+ Lymphoproliferation assay. U-bottom 96-well microtitre plates
were prepared with 100 µl aliquots of antigen, mitogen or medium alone
in each well. Antigens were tested in triplicate over a range of
concentrations from 4 to 0±02 µg}ml. PBMC were prepared at a
concentration of 5¬10' cell}ml, 100 µl aliquots dispensed into each well,
and then incubated at 37 °C for 6 days in an atmosphere of 5 % CO ­air.
#
Cells were pulsed with 1 µCi [$H-methyl]thymidine 6 h prior to harvesting
with a Skatron Combi cell harvester (Helis Bio) and the incorporation of
label was measured by liquid scintillation counting. Tests were performed
in triplicate and the geometric mean c.p.m. were determined for each test.
Stimulation index (SI) was calculated as the ratio of experimental c.p.m. to
control c.p.m. and indices of greater than 2 were considered to be
positive.
Cell-associated virus loads measured early after
infection with attenuated virus are dependent upon
inoculum dose
Outcome of superinfection challenge
Fifteen weeks after inoculation with the attenuated SIVmacC8, all ten macaques and four naive control animals were
inoculated intravenously with 30 MID of a cell-free stock of
&!
virus designated SIVmac220. This virus stock was derived
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
BJDH
BJDI
­
­
­
­
­
­
­
­
­
®
®
®
®
®
4S
14S
33S
21S
23S
33R
22S
7S
15S
13S
9S
26S
189
190
C8
C8
C8
C8
C8
C8
C8
C8
C8
®
®
®
®
®
­
­
­
­
­
­
­
­
®
®
®
®
®
®
No.
20 000
20 000
2 000
2 000
200
200
20
20
2
2
0
0
0
0
Presuperinfection
SIVmacC8
challenge
dose
(TCID50) VI PCR S
PCR
®
C8
®
C8
­
C8
®
C8
­
C8
®
C8
­ C8}220
®
C8
­
C8
­
220
­
220
­
220
­
220
­
220
VI
2
®
­
­
®
­
­
­
­
®
­
­
­
­
­
VI
C8
C8
C8
C8}220
C8}220
C8}220
C8}220
C8}220
C8
220
220
220
220
220
PCR
4
®
­
®
®
­
­
­
­
­
­
­
­
­
­
VI
C8
C8}220
C8
C8
C8}220
C8
C8}220
C8}220
C8
220
220
220
220
220
PCR
6
­
­
®
­
­
­
­
®
­
­
­
­
­
­
VI
C8
C8}220
C8
C8
C8}220
C8}220
220
C8
C8}220
220
220
220
220
220
PCR
8
®
­
­
®
­
­
­
­
®
­
­
­
­
­
VI
C8
C8}220
®
®
C8}220
C8}220
C8}220
®
C8
220
220
220
220
220
PCR
12
Weeks after superinfection challenge :
VI, virus isolation on C8166 cells ; PCR, polymerase chain reaction for PBMC-associated proviral DNA ; S, seroconversion.
PCR
®
C8
­
220
®
C8
­
C8
­ C8}220
­
220
­ C8}220
®
C8
­
220
­
220
­
220
­
220
­
220
­
220
VI
16
PCR
®
C8
­
®
®
C8
®
®
­ C8}220
­
220
® C8}220
® C8}220
­
220
­
220
­
220
­
220
­
220
­
220
VI
21
Table 1. Effect of dose on infection of macaques with attenuated SIVmacC8 and resistance to superinfection challenge with SIVmac220
®
­
®
®
­
­
­
­
­
­
­
­
­
­
VI
C8
C8}220
®
C8
C8}220
C8}220
220
C8}220
220
220
220
220
220
220
PCR
24
M. P. Cranage and others
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
SIV vaccine dose effect
Fig. 1. Cell-associated virus loads. The number of cells per 106 PBMC required for the recovery of SIV is shown relative to the
number of weeks after intravenous inoculation of rhesus macaques with different doses of attenuated SIVmac. Details of dose
are given in Table 1.
Fig. 2. CD4 and CD8 PBMC subsets. The mean CD4 and CD8 counts in
macaques following primary infection with attenuated SIVmacC8 (E) or
virulent SIVmac220 (D) is shown. Macaques infected with attenuated
virus were superinfection challenged at week 15. Bars indicate standard
deviations and were determined for pre-infection time-points of
attenuated SIVmac-infected animals on two occasions. Values for naive
control animals are shown for the day of challenge only (see text).
from the spleen of a rhesus macaque 167 days after infection
with the J5 molecular clone of SIVmac251-32H.
Virus was isolated consistently from week 2 through to
week 24 from the four control animals and from the animal that
had received 2 TCID of SIVmacC8 but was not infected
&!
(13S). Differential PCR confirmed that all these animals were
infected with the challenge virus (SIVmac220). In contrast,
virus was recovered with relatively high frequency in only six
of nine animals that had been primed with attenuated virus and
in three of these animals virus isolation was delayed relative to
the controls. Differential PCR revealed that three of four
animals primed with the two highest doses of attenuated virus
essentially resisted superinfection. Challenge proviral DNA
was never detected in two of these animals (4S and 33S) and
was detected on a single occasion only, 4 weeks after challenge,
in the third (21S). Furthermore, the profile of virus isolation
from these three animals was consistent with the presence of
attenuated virus only. Both challenge and attenuated provirus
signals were detected in the remaining animals, although in
only one animal (15S) did the challenge virus become
consistently the predominant proviral species. With the
exception of animal 7S, the frequency of virus recovery from
PBMC was high (75–100 %) in dually infected animals,
consistent with the presence of the more virulent challenge
virus.
Following superinfection challenge, mean CD4 and CD8
cell numbers in the peripheral circulation increased to values
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
BJDJ
M. P. Cranage and others
similar to those determined in the naive controls. Although
individual values in each animal fluctuated over time (data not
shown) there was no sustained loss or increase of either CD4+
or CD8+ PBMC in animals infected with SIVmacC8 and
challenged with SIVmac220. In all animals except one (15S),
CD4 cell numbers had increased by 2 weeks after superinfection challenge, regardless of protection status. In animal
15S, the CD4 cell count remained unaltered after challenge but
had risen by week 8. CD4 cell counts in animal 13S, which
remained uninfected following primary inoculation with
SIVmacC8, were below pre-challenge levels on two occasions
following infection with SIVmac220.
The majority of SIVmacC8-infected animals also showed an
increase in CD8+ cell numbers 2 weeks after superinfection
challenge, with the exception of animal 23S where the cell
count fell but nevertheless rebounded by 8 weeks after
challenge. CD8 cell numbers in animal 13S changed in a similar
manner to those of 23S.
In contrast to animals already infected with SIVmacC8,
numbers of circulating CD4 and CD8 cells in controls fell
following infection with SIVmac220. CD4 cell numbers fell in
three animals to values below 0±5¬10' cells}ml. In a fourth
animal (26S) the CD4 cell count was low on the day of
challenge (0±39¬10' cells}ml) and fell to 0±26¬10' cells}ml
at 16 weeks after challenge. The mean CD4 cell count in the
control group was significantly different from the SIVmacC8infected group at 23 weeks and 31 weeks after challenge (P ¯
0±005, P ¯ 0±006 respectively ; Student’s t-test). CD8 cell
numbers in controls did not show a sustained reduction and
were not significantly different from those of SIVmacC8infected animals.
Immune responses following infection with live
attenuated SIVmac and superinfection challenge
Immunoblotting revealed all but one animal (13S) to have
seroconverted to the SIV major structural proteins following
inoculation with attenuated virus (data not shown), including
animal 15S which failed to yield virus upon culture of PBMC.
Only in animals given high doses (2000 and 20 000 TCID ) of
&!
attenuated virus (three of four) were binding antibodies to SIV
envelope detectable 2 weeks after inoculation (Fig. 3). Initially,
the anti-envelope response in 15S was lower than in other
animals of the group. Following superinfection challenge, the
four control animals and animal 13S seroconverted to SIV
antigens. In general, it was not possible to distinguish animals
that were superinfection resistant from those that became
dually infected on the basis of anamnestic antibody responses
to SIV envelope.
Neither titre of binding antibody nor virus neutralizing
activity, measured on the day of superinfection challenge,
correlated with protection (Table 2).
In order to determine how quickly CTL responses were
generated after infection with live attenuated virus, PBMC
BJEA
Fig. 3. Antibody responses. Anti-SIV gp140 antibody titres measured by
ELISA, in macaques inoculated with attenuated SIVmacC8 (week 0) and
challenged at week 15 with virulent SIVmac220.
Table 2. Serum antibody levels and SIV infectivity
neutralizing activity on the day of superinfection
challenge
Antibody titre
SIVmacC8
Animal* dose (TCID50) Anti-p27 Anti-gp140
4S
14S
33S
21S
23S
33R
22S
7S
15S
13S
20 000
20 000
2 000
2 000
200
200
20
20
2
2
1300
1700
500
600
1900
300
500
1000
1800
! 100
52 000
40 000
40 000
6 000
68 000
7 000
7 000
30 000
13 000
! 100
Neutralization
(yield
reduction)
154
154
386
123
218
69
154
123
485
0±9
* Animals shown in bold were resistant to superinfection with
SIVmac220 (Table 1).
from SIVmacC8-infected animals were restimulated in vitro
using SIV-infected autologous PHA blasts. CTL activity was
measured against autologous B-LCL targets expressing SIV
Nef from recombinant vaccinia virus. This protein was chosen
since it is known to contain CTL epitopes and a strong anti-Nef
CTL response has been associated with a vaccine effect
(Gallimore et al., 1995).
A transient Nef-specific CTL response was detected in
three of nine animals tested following infection with SIVmacC8. Later, at the time of superinfection challenge, Nefspecific CTL responses could no longer be detected in any of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
SIV vaccine dose effect
Table 3. SIV Nef specific CTL activity in PBMC following infection of macaques with
SIVmacC8 and superinfection challenge at week 15
CTL activity was determined after restimulation of PBMC using autologous short-term infected PHA blasts.
Activity of less than 10 % after subtraction of activity against control targets expressing influenza A virus PB2
was considered as not significant. , Not done.
Percent specific release (Nef-specific CTL)
(E : T ¯ 50 : 1)
Animal*
4S
14S
33S
21S
23S
33R
22S
7S
15S
13S
SIVmacC8
dose
(TCID50)
2
8
20 000
20 000
2 000
2 000
200
200
20
20
2
2
! 10
! 10
! 10
! 10
! 10
14
22


! 10
14



! 10
! 10
10


! 10
Weeks after SIVmacC8 inoculation :
13
15
17
27
! 10
! 10

! 10

! 10
! 10

! 10
! 10

! 10
! 10
! 10
! 10

! 10

! 10
! 10
! 10
! 10

! 10
! 10
! 10
! 10

! 10
! 10
! 10




! 10




29
! 10




! 10
! 10


! 10
* Animals shown in bold were protected from superinfection challenge with SIVmac220.
Fig. 4. CTL activity against SIV Nef. The percent specific release of 51Cr from autologous B cell lines infected with recombinant
vaccinia virus expressing SIV Nef (D) or recombinant vaccinia virus expressing influenza A virus PB2 (E) is shown for a range
of effector to target ratios (E : T) in five attenuated SIVmac-infected macaques tested between 17 and 25 weeks after challenge
with virulent virus.
the animals tested (Table 3). Furthermore, Nef-specific CTLs
could not be detected following superinfection challenge when
tested up to 14 weeks after challenge. This result suggested
that the in vitro restimulation method may have been inefficient
in expanding CTL precursors. Between 17 and 25 weeks after
superinfection challenge, a modified restimulation method was
adopted, using a higher stimulator to effector ratio and PHA
blasts that had been inactivated after a longer exposure to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
BJEB
M. P. Cranage and others
virus. This method revealed the presence of Nef-specific CTL
in three of three animals tested that had been primed with the
higher doses of attenuated virus. Nef-specific CTL were absent
in two of two (33R and 22S) animals tested given lower doses
of attenuated virus, despite their earlier detection after primary
infection (Fig. 4). The presence of Nef-specific CTL did not,
however, correlate directly with protection since animal 14S
became infected upon superinfection challenge.
When sufficient PBMC were available, lymphoproliferative
responses against whole, heat-inactivated SIV and recombinant
envelope were determined. Although the magnitude of
responses in individual animals varied on each occasion tested
PBMC from each of the protected animals (4S, 21S and 33S)
had whole SIV antigen-driven responses on two of three
occasions tested (SI range 2±7–22). A response to gp140 was
seen on one occasion with PBMC from 33S (SI 57) and on two
occasions with PBMC from 21S (SI 2±9 and 9), corresponding
to times when responses to whole virus were detected. PBMC
from four animals that were infected with both attenuated
virus SIVmacC8 and SIVmac220 (14S, 22S, 7S and 15S)
responded to whole SIV on at least one occasion (SI range
2±7–46). PBMC from 14S, 7S and 15S also responded to gp140
(SI range 2±1–7±5).
Discussion
In general, live attenuated vaccines against virus infections
are extremely successful. The efficacy of this type of vaccination is believed to be due to the stimulation of immune
responses that mimic those seen during natural infection and
thus the immune system is primed for encounter with the
natural pathogen. It is therefore important to determine the
parameters of infection that influence superinfection resistance.
One study of macaques infected with attenuated SIV has
suggested that superinfection resistance developed fully only
after a considerable period of time (Wyand et al., 1996). Other
studies, however, indicated a high degree of protection at 10
and 20 weeks (Norley et al., 1996) or 22 weeks (Stahl-Hennig
et al., 1996) after primary infection. The results of the present
study show that this apparent discrepancy may be related to
the dose of attenuated SIV administered. Furthermore, different
doses of attenuated SIV influenced the virus recovery profile
prior to superinfection challenge, suggesting that the early
dynamics of infection relate to the ‘ vaccine ’ efficacy.
Dose-related vaccine efficacy has been reported with other
live attenuated viruses including rabies virus (Flamand et al.,
1984), rotavirus (McNeal et al., 1992 ; Feng et al., 1994) and
herpes simplex virus (Mercadal et al., 1993). The results of the
present study also show a dose-related effect but are unusual in
that animals receiving high doses of attenuated SIV had lower
cell-associated virus loads 2 weeks after infection than did
animals receiving lower doses, suggesting that less virus
replication had occurred in these former animals. The possibility that the peak of cell-associated viraemia had shifted in
BJEC
these animals cannot be excluded. The results from Norley et
al. (1996), using a high dose of SIVmacC8, who found a similar
level of early viraemia that had declined by 4 weeks after
inoculation, indicate that any shift in replication dynamics is,
therefore, likely to be towards an earlier peak in viraemia. The
earlier rise in anti-SIV envelope antibodies in animals given a
high dose of attenuated virus may indeed support this notion.
Taken together, it is clear that further experiments with a
higher frequency of sampling and including plasma RNA
determination are now needed to define the relationship
between virus dose and early replication dynamics.
The level of protection from superinfection with virulent
virus was similar to that reported in two previous studies
(Norley et al., 1996 ; Stahl-Hennig et al., 1996), both of which
used high doses of SIVmacC8. An effect on pathogenesis was
also found in the present study in the majority of animals
infected with lower doses of attenuated virus. Although these
animals became superinfected, as evidenced by PCR detection
of challenge virus-derived proviral DNA, there was a sparing
from CD4 cell loss, delayed detection of challenge virus in all
but one animal and strong SIV-driven lymphoproliferative
responses. Normally, SIV-driven lymphoproliferative responses are weak in animals infected with virulent SIV
(McGraw et al., 1990 ; Ahmed-Ansari et al., 1990 ; Dittmer et al.,
1994 ; Sharpe, 1997). Protection from the consequences of
superinfection with virulent virus in the absence of ‘ sterilizing ’
protection has been reported in macaques infected with SIVmac
1A11 or HIV-2 (Marthas et al., 1990 ; Putkonen et al., 1990),
both of which replicate relatively poorly in macaques.
Interestingly, in the animal infected following inoculation
with the lowest dose of attenuated virus the outcome was
different with no recovery of virus but evidence of infection by
PCR and seroconversion. Wyand et al. (1996) reported
transient infection in animals given an apparently low dose of
SIVmac239 ∆nef intravenously and others have reported
transient infections following low-dose mucosal exposure
(Pauza et al., 1993 ; Clerici et al., 1994 ; Miller et al., 1994 ;
Trivedi et al., 1996). Exposure to low doses of virus has been
associated with protection from subsequent virus challenge
(Clerici et al., 1994 ; Trivedi et al., 1996) ; however, in the
present study no such association was found. Neither the
animal with modified SIVmacC8 infection (15S) nor the animal
that remained uninfected (13S) resisted superinfection.
Although detection of challenge virus was delayed in 15S, the
challenge virus proviral DNA subsequently became the
predominant species and CD4 cells declined relative to prechallenge levels. Similarly, Wyand et al. (1996) found no
protection in animals transiently infected with attenuated virus
after intravenous exposure, suggesting that the low-dose
protection effect may be unique to mucosal exposure.
The mechanism of superinfection resistance remains unknown. No direct correlate of immune protection was evident
in the study described here. The importance of serum
neutralizing antibodies in superinfection resistance has been
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
SIV vaccine dose effect
proposed (Clements et al., 1995 ; Wyand et al., 1996) but no
association between such activity and protection was found in
this study in agreement with a previous study using SIVmacC8
(Norley et al., 1996).
The detection of CTL activity, early after infection with
attenuated virus is consistent with a role for CTL in the early
clearance of virus infected cells and the inability to detect
activity at later time points may have been due to inefficient in
vitro restimulation of precursor CTL (Sharpe, 1997 ; Cranage et
al., 1997). A longer period of stimulation in vitro may have
revealed low frequency CTL precursors. The detection of CTL
activity only in animals primed with high doses of attenuated
virus, regardless of protection status, together with the
tendency for an earlier antibody response to envelope in these
animals, suggests that these animals may have been primed for
qualitatively or quantitatively different responses to those in
animals that received lower doses of attenuated virus. Infection
of macaques with a high dose of SIVmacC8 has been associated
with an overexpression of IL-2, IL-4 and IFN-γ mRNAs in
PBMC during the first few weeks of infection (Benveniste et al.,
1996) and it remains to be determined if this effect is virusdose-related.
The lack of definitive correlates of protective immunity in
the study described here may simply reflect the limitations of
the assays employed with respect to breadth of antigen
recognition, specific function and sampling sites. Alternatively,
superinfection resistance induced with attenuated SIV may be
a non-immunological phenomenon. It is possible that a high
dose of attenuated virus saturates a critical population of target
cells creating a bottleneck to the establishment of subsequent
infection. This may involve virus interference as has been
described for murine retroviruses (Mitchell & Risser, 1992) or
may be associated with the induction of intrinsic antiviral
factors such as the CD8 suppressor factors (Mackowicz &
Levy, 1992 ; Cocchi et al., 1995 ; Paxton et al., 1996). PBMC
from macaques infected with SIVmacC8 are relatively less
sensitive to infection with virulent virus in vitro than are cells
from uninfected macaques and this difference is influenced by
the presence of CD8 cells (P. Silvera, personal communication).
Finally, defective particles in the primary inoculum may
influence superinfection resistance. Such material may modulate the early dynamics of infection by the induction of cell
factors as discussed above and may account for the threshold
effect of protection seen in the current experiment. Such a
model would not exclude classical immunity in the superinfection resistance effect, since the primary inoculum may
determine the optimal environment for the stimulation of
subsequent protective immunity. Experiments using defined
particle to infectivity ratios would test this hypothesis.
The dose of attenuated virus needed to obtain protection
from challenge with virulent virus is a particularly important
parameter in the light of the findings that, at least in neonates
infected orally, high doses of virus may induce disease (Baba et
al., 1995 ; Wyand et al., 1997). It is clear that the relationship
between virus dose and time to superinfection resistance
requires further investigation to better understand the parameters and mechanisms of protection elicited with live
attenuated vaccines against immunodeficiency viruses.
We are grateful to Dr E. J. Stott, Dr N. Almond and Dr E. W. Rud for
useful discussions and to Dr Harvey Holmes and the MRC AIDS Reagent
Programme for the provision of many of the reagents used. The work
was supported by a grant from the UK Medical Research Council and by
the UK Department of Health.
References
Ahmed-Ansari, A., Powell, J. D., Jenson, P. E., Yehuda-Cohen, T.,
McClure, H. M., Andersen, D., Fultz, P. N. & Sell, K. W. (1990).
Requirements for simian immunodeficiency virus antigen-specific in vitro
proliferation of T cells from infected rhesus and sooty mangabeys. AIDS
4, 399–407.
Almond, N., Kent, K., Cranage, M., Rud, E., Clarke, B. & Stott, E. J.
(1995). Protection by attenuated simian immunodeficiency virus in
macaques against challenge with virus-infected cells. Lancet 345,
1342–1344.
Baba, T. W., Jeong, Y. S., Penninck, D., Bronson, R., Greene, M. F. &
Ruprecht, R. M. (1995). Pathogenicity of live attenuated SIV after
mucosal infection of neonatal macaques. Science 267, 1820–1825.
Benveniste, O., Vaslin, B., LeGrand, R., Cheret, A., Matheux, F.,
Theodoro, F., Cranage, M. P. & Dormont, D. (1996). Comparative
interleukin (IL)-2}interferon (IFN)-γ and IL-4}IL-10 responses during
acute infection of macaques inoculated with attenuated nef-truncated or
pathogenic SIVmac251 virus. Proceedings of the National Academy of
Sciences, USA 93, 3658–3663.
Clements, J. E., Montelaro, R. C., Zink, M. C., Amedee, A. M., Miller, S.,
Trichel, A. M., Jagerski, B., Hauer, D., Martin, L. N., Bohm, R. P. &
Murphey-Corb, M. (1995). Cross-protective immune responses induced
in rhesus macaques by immunization with attenuated macrophage-tropic
simian immunodeficiency virus. Journal of Virology 69, 2737–2744.
Clerici, M., Clark, E. A., Polacino, P., Axberg, I., Kuller, L. R., Casey,
N. I., Morton, W. R., Shearer, G. M. & Benveniste, R. E. (1994). T-cell
proliferation to subinfectious SIV correlates with lack of infection after
challenge of macaques. AIDS 8, 1391–1395.
Cocchi, F., DeVico, A. L., Garzino-Demo, A., Arya, S. K., Gallo, R. C. &
Lusso, P. (1995). Identification of RANTES, MIP-1 alpha, and MIP-1
beta as the major HIV-suppressive factors produced by CD8+ T cells.
Science 270, 1811–1815.
Cranage, M. P., Whatmore, A. M., Sharpe, S. A., Cook, N., Polyanskaya,
N., Leech, S., Smith, J. D., Rud, E. W., Dennis, M. J. & Hall, G. A.
(1997). Macaques infected with live attenuated SIVmac are protected
against superinfection via the rectal mucosa. Virology 229, 143–154.
Daniel, M. D., Kirchhoff, F., Czajak, S. C., Sehgal, P. K. & Desrosiers,
R. C. (1992). Protective effects of a live attenuated SIV vaccine with a
deletion in the nef gene. Science 258, 1938–1941.
Dittmer, U., Luke, W., Stahl-Hennig, C., Coulibaly, C., Petry, H.,
Bodemer, H., Hunsmann, G. & Voss, G. (1994). Early helper T-cell
dysfunction in simian immunodeficiency virus but not in human
immunodeficiency virus type-2-infected macaques. Journal of Medical
Primatology 23, 298–303.
Feng, N., Burns, J. W., Bracy, L. & Greenberg, H. B. (1994).
Comparison of mucosal and systemic humoral immune responses and
subsequent protection in mice orally inoculated with a homologous or a
heterologous rotavirus. Journal of Virology 68, 7766–7773.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04
BJED
M. P. Cranage and others
Flamand, A., Coulon, P., Pepin, M., Blancou, J., Rollin, P. & Portnoi, D.
(1984). Immunogenic and protective power of avirulent mutants of
rabies virus selected with neutralizing monoclonal antibodies. In Modern
Approaches to Vaccines, pp. 289–294. Edited by R. M. Chanock & R. A.
Lerner. Cold Spring Harbor, NY : Cold Spring Harbor Laboratory.
Gallimore, A., Cranage, M., Cook, N., Almond, N., Bootman, J., Rud, E.,
Silvera, P., Dennis, M., Corcoran, T., Stott, J., McMichael, A. & Gotch,
F. (1995). Early suppression of SIV replication by CD8+ nef-specific
cytotoxic T cells in vaccinated macaques. Nature Medicine 1, 1167–1173.
Kitchin, P. A., Almond, N., Szotyori, Z., Fromholc, C. E., McAlpine, L.,
Silvera, P., Stott, E. J., Cranage, M., Baskerville, A. & Schild, G. (1990).
Paxton, W. A., Martin, S. R., Tse, D., O’Brien, T. R., Skurnick, J.,
VanDevanter, N. L., Padian, N., Braun, J. F., Kotler, D. P., Wolinsky, S.
M. & Koup, R. A. (1996). Relative resistance to HIV-1 infection of CD4
lymphocytes from persons who remain uninfected despite multiple highrisk sexual exposures. Nature Medicine 2, 412–417.
Polyanskaya, P., Sharpe, S., Cook, N., Leech, S., Banks, J., Dennis, M.,
Hall, G., Stott, J. & Cranage, M. (1997). Anti-major histocompatibility
complex antibody responses to simian B cells do not protect macaques
against SIVmac infection. AIDS Research and Human Retroviruses 13,
923–931.
Putkonen, P., Thorstensson, R., Albert, J., Hild, K., Norrby, E.,
Biberfeld, P. & Biberfeld, G. (1990). Infection of cynomolgus monkeys
The use of the polymerase chain reaction for the detection of simian
immunodeficiency virus in experimentally infected macaques. Journal of
Virological Methods 28, 85–100.
with HIV-2 protects against pathogenic consequences of a subsequent
simian immunodeficiency virus infection. AIDS 4, 783–789.
McGraw, T. P., Vowels, B. R., Gardner, M. B., Ahmed-Ansari, A. &
Gershwin, M. E. (1990). Simian immunodeficiency virus-specific T cell-
Putkonen, P., Nilsson, C., Ma$ kitalo, B., Biberfeld, G., Thorstensson, R.,
Cranage, M. & Rud, E. (1996). Immunization with live attenuated
mediated proliferative response of infected macaques. AIDS 4, 191–198.
Mackowicz, C & Levy, J. A. (1992). CD8+ cell anti-HIV activity : nonlytic
suppression of virus replication. AIDS Research and Human Retroviruses 8,
1039–1050.
McNeal, M. M., Sheridan, J. F. & Ward, R. L. (1992). Active protection
against rotavirus infection of mice following intraperitoneal immunization. Virology 191, 150–157.
SIVmac can protect macaques against mucosal infection with SIVsm. In
Vaccines 96, pp. 235–241. Edited by F. Brown, E. Norrby, D. Burton & J.
Mekalanos. Cold Spring Harbor, NY : Cold Spring Harbor Laboratory.
Rose, J., Silvera, P., Flanagan, B., Kitchin, P. & Almond, N. (1995). The
development of PCR based assays for the detection and differentiation of
simian immunodeficiency virus in vivo. Journal of Virological Methods 51,
229–239.
Marthas, M. L., Sujipto, S., Higgins, J., Lohman, B., Torten, J., Luciw,
P. A., Marx, P. & Pederson, N. C. (1990). Immunization with a live
attenuated simian immunodeficiency virus (SIV) prevents early disease
but not infection in rhesus macaques challenged with pathogenic SIV.
Journal of Virology 64, 3694–3700.
Mercadal, C. M., Slaoui, M., Brown, S. M. & Rouse, B. T. (1993).
Efficacy of the herpes simplex virus types 1 and 2 mutant viruses to
confer protection against zosteriform spread in mice. Viral Immunology 6,
35–42.
Miller, C. J., Marthas, M., Torten, J., Alexander, N. J., Moore, J. P.,
Doncel, G. F. & Hendrickx, A. G. (1994). Intravaginal inoculation of
rhesus macaques with cell-free simian immunodeficiency virus results in
persistent or transient viremia. Journal of Virology 68, 6391–6400.
Miller, C. J., McChesney, M. B., Lu, X., Dailey, P. J., Chutkowski, C., Lu,
D., Brosio, P., Roberts, B. & Lu, Y. (1997). Rhesus macaques previously
infected with simian}human immunodeficiency virus are protected from
vaginal challenge with pathogenic SIVmac239. Journal of Virology 71,
1911–1921.
Mitchell, T. & Risser, R. (1992). Interference established in mice by
infection with Friend murine leukemia virus. Journal of Virology 66,
5696–5702.
Norley, S., Beer, B., Binninger-Schinzel, D., Cosma, C. & Kurth, R.
(1996). Protection from pathogenic SIVmac challenge following short-
term infection with a nef-deficient attenuated virus. Virology 219,
195–205.
Pauza, C. D., Emau, P., Salvato, M. S., Trivedi, P., MacKenzie, D.,
Malkovsky, M., Uno, H. & Schultz, K. T. (1993). Pathogenesis of
SIVmac after atraumatic inoculation of the rectal mucosa in rhesus
#&"
monkeys. Journal of Medical Primatology 22, 154–161.
BJEE
Rud, E. W., Cranage, M., Yon, J., Quirk, J., Ogilvie, L., Cook, N.,
Webster, S., Dennis, M. & Clarke, B. E. (1994). Molecular and biological
characterization of simian immunodeficiency virus macaque strain 32H
proviral clones containing nef size variants. Journal of General Virology
75, 529–543.
Sharpe, S. (1997). T cell responses to simian immunodeficiency virus. PhD
thesis, Open University, UK.
Stahl-Hennig, C., Dittmer, U., Nißlein, T., Petry, H., Jurkiewicz, E.,
Fuchs, D., Wachter, H., Ma$ tz-Rensing, K., Kuhn, E.-M., Kaup, F.-J.,
Rud, E. W. & Hunsmann, G. (1996). Rapid development of vaccine
protection in macaques by live-attenuated simian immunodeficiency
virus. Journal of General Virology 77, 2969–2981.
Trivedi, P., Horejsh, D., Hinds, S. B., Hinds, P. W., II, Wu, M. S.,
Salvato, M. S. & Pauza, C. D. (1996). Intrarectal transmission of simian
immunodeficiency virus in rhesus macaques : selective amplification and
host responses to transient and persistent viraemia. Journal of Virology 70,
6876–6883.
Whatmore, A. M., Cook, N., Hall, G. A., Sharpe, S., Rud, E. W. &
Cranage, M. P. (1995). Repair and evolution of nef in vivo modulates
simian immunodeficiency virus virulence. Journal of Virology 69,
5117–5123.
Wyand, M. S., Manson, K. H., Garcia-Moll, M., Montefiori, D. &
Desrosiers, R. C. (1996). Vaccine protection by a triple deletion mutant
of simian immunodeficiency virus. Journal of Virology 70, 3724–3733.
Wyand, M. S., Manson, K. H., Lackner, A. A. & Desrosiers, R. C.
(1997). Resistance of neonatal monkeys to live attenuated vaccine
strains of simian immunodeficiency virus. Nature Medicine 3, 32–36.
Received 12 February 1998 ; Accepted 23 April 1998
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 23:23:04