Content uploaded by Angela M Amedee
Author content
All content in this area was uploaded by Angela M Amedee on Sep 25, 2014
Content may be subject to copyright.
Characterization of SIV in the Oral Cavity and in Vitro
Inhibition of SIV by Rhesus Macaque Saliva
Jessica S. Thomas,
1
Nedra Lacour,
1
Pamela A. Kozlowski,
1,2
Steve Nelson,
3
Gregory J. Bagby,
4
and Angela M. Amedee
1
Abstract
Human immunodeficiency virus (HIV) infections are rarely acquired via an oral route in adults. Previous studies
have shown that human whole saliva inhibits HIV infection in vitro, and multiple factors present in human saliva
have been shown to contribute to this antiviral activity. Despite the widespread use of simian immunodeficiency
virus (SIV)-infected rhesus macaques as models for HIV pathogenesis and transmission, few studies have
monitored SIV in the oral cavity of infected rhesus macaques and evaluated the viral inhibitory capacity of
macaque saliva. Utilizing a cohort of rhesus macaques infected with SIV
Mac251
, we monitored virus levels and
genotypic diversity in the saliva throughout the course of the disease; findings were similar to previous ob-
servations in HIV-infected humans. An in vitro infectivity assay was utilized to measure inhibition of HIV/SIV
infection by normal human and rhesus macaque whole saliva. Both human and macaque saliva were capable of
inhibiting HIV and SIV infection. The inhibitory capacity of saliva samples collected from a cohort of animals
postinfection with SIV increased over the course of disease, coincident with the development of SIV-specific
antibodies in the saliva. These findings suggest that both innate and adaptive factors contribute to inhibition of
SIV by whole macaque saliva. This work also demonstrates that SIV-infected rhesus macaques provide a
relevant model to examine the innate and adaptive immune responses that inhibit HIV/SIV in the oral cavity.
Introduction
Human immunodeficiency virus (HIV) infection and
AIDS remain significant public health challenges
worldwide. The majority of new HIV-1 infections are ac-
quired across mucosal surfaces through high risk sexual
contact with infected individuals.
1
Epidemiologic and clinical
reports agree that transmission of virus rarely occurs via an
oral route in adults, despite the presence of detectable levels of
virus in the saliva, salivary glands, and oropharyngeal tissues
of HIV-infected patients.
2–4
The level and potential infectivity of virus present in the
oral cavity of HIV-positive individuals have been evaluated
using various methodologies.
2,5
Studies using culture-based
methods have reported low frequencies of HIV-1 detection in
saliva, while more quantitative methods, such as polymerase
chain reaction (PCR), have led to higher frequencies of
detection.
5
One study detected HIV-1 RNA by PCR in 96% of
cell-free whole saliva samples, with a median viral load of 162
copies/ml saliva (range 0–72,080 copies/ml).
4
Another study
examined virus levels in saliva by the Nuclisens (nucleic acid
sequence-based amplification) assay and detected HIV-1
RNA in 42% of salivary secretions, at a mean level of
794 copies/ml (range 79–794,328 copies/ml).
6
HIV-1 RNA
levels in saliva are typically lower than those of matched
plasma samples; however, discordant viral loads in saliva and
plasma have been observed, leading investigators to suggest
that the oral cavity may be a viral reservoir.
4,7
Genotypic
analyses of HIV-1 envelope diversity in saliva and plasma
have demonstrated homogeneous viral populations in these
compartments.
8,9
Homogeneity of the HIV-1 populations in
saliva and plasma suggests that viral variants in saliva and
plasma originate from a common source and raises the pos-
sibility that the oral cavity may be used as a noninvasive
site to monitor viral evolution and disease progression in
HIV-infected individuals.
8,9
Attempts to culture infectious virus from the saliva of
HIV-infected individuals have been largely unsuccessful, with
recovery rates ranging from 0% to 5%, despite detectable and,
in some cases, high levels of viral RNA in saliva.
10
In a number
of studies, human saliva collected from HIV seronegative
and seropositive individuals has been shown to inhibit HIV
1
Department of Microbiology, Immunology, and Parasitology,
2
Gene Therapy Program,
3
Department of Medicine, and
4
Department of
Physiology, Louisiana State University Health Sciences Center, New Orleans, Louisiana.
AIDS RESEARCH AND HUMAN RETROVIRUSES
Volume 26, Number 8, 2010
ªMary Ann Liebert, Inc.
DOI: 10.1089/aid.2009.0235
901
infection in vitro.
11–15
Multiple endogenous inhibitory factors
have been identified in saliva, which contribute to this anti-
viral effect, including mucins, cystatins, defensins, secretory
leukocyte protease inhibitor, lactoferrin, anti-HIV antibodies,
and other factors.
2,10
The innate inhibitory capacity of saliva
is, at least in part, responsible for limiting viral expression and
transmission of HIV in the oral cavity.
2,3,16
Simian immunodeficiency virus (SIV)-infected rhesus ma-
caques are a widely accepted model for the study of HIV
transmission and pathogenesis. Experimental infection of
macaques with SIV leads to immune dysfunction and pro-
gression through advanced disease similar to HIV infection in
humans.
17
Although SIV in the oral cavity has not been fully
investigated, one recent report quantified SIV RNA levels in
the saliva and peripheral blood of five SIV
Mac251
-infected
rhesus monkeys over the first 90 days postinfection (dpi).
18
Levels of SIV RNA in saliva were variable between the five
animals. Saliva and peripheral blood viral RNA levels peaked
at 14 dpi, and median levels were 10
5
and 10
7
viral RNA
copies/ml, respectively. Following 35 dpi, only occasional
saliva samples had detectable levels of SIV RNA, which were
observed mainly in animals with high set point plasma viral
RNA levels.
Rhesus macaques may also provide a good model to
evaluate the innate protective factors present in the oral cav-
ity, including the potential SIV-inhibitory capacity of ma-
caque saliva. Initial reports demonstrated that both human
and chimpanzee saliva were able to inhibit HIV-1 infection of
peripheral blood mononuclear cells in vitro.
11
However, hu-
man submandibular saliva had little to no inhibitory capacity
against HIV-2
ROD
or SIV
Mac239
infection in vitro.
13
To evaluate
the utility of the SIV-infected macaque model for the study of
oral HIV pathogenesis, we examined oral SIV loads and ge-
notypes in saliva collected from a cohort of SIV
Mac251
-infected
rhesus macaques throughout the course of disease and eval-
uated the inhibitory capacity of rhesus macaque whole saliva
against SIV in vitro.
Materials and Methods
Cohort and sample collection
All samples were collected using protocols approved by
LSUHSC IRB and LSUHSC and Tulane IACUC committees.
Whole saliva samples were collected from eight normal hu-
man volunteers. Volunteers were asked to abstain from food
and drink only water for 1 h prior to sample collection. Saliva
(5 ml) was collected from each volunteer by expectoration into
15-ml polypropylene tubes on ice. Whole saliva samples were
clarified by centrifugation (12,000gfor 1 min ) and aliquots
of supernatants were stored at 808C until use.
Whole saliva samples were collected from eight normal,
untreated juvenile male rhesus macaques (Macaca mulatta)
housed at the LSU Health Sciences Center animal facility.
Whole saliva samples were also collected from 16 juvenile
male rhesus macaques infected intravenously with SIV
Mac251
and housed at the Tulane National Primate Research Center in
Covington, LA, at several time points post-SIV infection (1, 2,
4, 6, 8, 12, 16, 30, or 32, and 48 weeks postinoculation). All
animals in this report were enrolled in larger studies at
LSUHSC with similar treatment protocols. Whole saliva
samples were collected from anesthetized macaques by
placement of two Weck-Cel surgical sponges (Solon-
Medtronics, Solon, OH) in the buccal cavity for 5–10 min.
19
Sponges soaked in saliva were placed in 15-ml polypropylene
tubes on ice immediately following collection. Whole saliva
was recovered from the sponges by centrifugation (12,000g
for 1 min) through a sterile 0.5-ml tube containing small holes
at the bottom into a 1.5-ml collection tube. Whole fluid was
then clarified by centrifugation, and supernatants were stored
at 808C until analyzed.
As varying volumes of saliva were recovered, the quality of
macaque whole saliva samples was assessed for total protein
levels by BCA assay (Thermo Pierce), according to manufac-
turer’s instructions. Typical total protein levels in macaque
whole saliva ranged from 2000 to 4000 mg/ml.
Peripheral blood samples were collected in tubes contain-
ing EDTA from 16 rhesus macaques at various time points
following SIV
Mac251
infection, coincident with the time points
of saliva collection. Plasma aliquots were stored at 808C
until analyzed.
Quantitation of SIV RNA
Levels of SIV RNA in rhesus macaque plasma and saliva
samples were quantified by real-time TaqMan, reverse-
transcriptase PCR (RT-PCR) using SIV gag primers and
probe, as previously described.
20
Briefly, viral RNA was iso-
lated from 1 ml plasma using the Trizol reagent (Life Tech-
nologies, Rockville, MD) and reverse transcribed to cDNA.
Products were PCR amplified in duplicate reactions using
1/10 of the total volume of cDNA.
Whole saliva SIV RNA levels were quantified by the same
method, with some modifications for accurate measurement
of viral RNA in mucosal secretions. Briefly, RNA was isolated
from 50 ml of whole saliva supernatant using the Trizol-LS
reagent (Life Technologies, Rockville, MD) and reverse tran-
scribed into cDNA using Multiscribe Reverse Transcriptase
(Applied Biosystems, Foster City, CA). PCR amplification
was performed in duplicate reactions utilizing the total
amount of cDNA. Due to the limited volumes of saliva
available for assessment, the assay had a quantitation limit of
100 copies/ml of saliva.
Characterization of SIV genotypes
Viral genotypes expressed in plasma and saliva over the
course of disease were evaluated in SIV-infected animals by
heteroduplex tracking assays (HTA) utilizing SIV envelope
V1 sequences amplified by RT-PCR from time-matched
plasma and saliva samples, as previously described.
21,22
Briefly, viral RNA was isolated using Trizol reagent (Life
Technologies, Rockville, MD) and reverse transcribed into
cDNA using Multiscribe Reverse Transcriptase (Applied
Biosystems, Foster City, CA). Using conserved primer pairs, a
480-bp fragment of SIV envelope V1/V2 sequences was am-
plified from cDNA by nested PCR. Products from indepen-
dent V1 PCR amplifications, at least three per plasma sample
and two per saliva sample, were pooled for more accurate
analysis of viral diversity by HTA.
A single-stranded
32
P-labeled probe was constructed by
end point dilution of SIV
Mac251
stock and PCR amplification of
the V1 region from the predominant envelope genotype
present in the stock. Pooled V1 PCR products generated from
plasma and saliva samples were mixed with 1 ml
32
P-labeled
probe and 2 ml annealing buffer in a total volume of 20 ml. This
902 THOMAS ET AL.
mixture was heated to 908C for 3 min, chilled on ice for 3 min,
and then separated by electrophoresis on a 12% polyacryl-
amide gel for 1600 V-h. DNA migration patterns were visu-
alized and evaluated by phosphorimaging and ImageQuant
software (GE Healthcare).
Culture and coculture of SIV from whole saliva
CEMx174-R5 and MT4-R5 cell lines and rhesus macaque
peripheral blood mononuclear cells (RhPBMCs) were used for
SIV culture/coculture experiments and expansion of virus
stocks. CEMx174-R5 and MT4-R5 T cell lines were a gift from
James Robinson, Tulane University School of Medicine.
23
Both cell lines express CCR5 cell surface receptors under pu-
romycin selection. RhPBMCs were isolated from normal (SIV-
seronegative) rhesus macaque blood using Ficoll-Hypaque
gradient separation and stimulated with phytohemagglutinin
(PHA) in interleukin (IL)-2-containing media for 72 h prior to
experiments. PBMC and T cell line cultures were evaluated for
cytotoxic effects of whole saliva by cell viability counts using
Trypan blue staining. At a 1:20 saliva dilution, cells did not
show any evidence of decreased viability over 14 days in
culture.
Acute (2 wpi) and chronic (18 months postinoculation)
whole saliva samples from SIV
Mac251
-infected rhesus ma-
caques were clarified by centrifugation. The supernatant
fraction was used for SIV culture studies, and the cell fraction
was used for SIV coculturing studies. Cultures and cocultures
were maintained at 378C in six-well plates and contained
110
6
RhPBMC, CEMx174-R5, or MT4-R5 cells and 1:20 di-
lution of cell-free whole saliva supernatant or whole saliva cell
fraction in 2 ml appropriate culture media. Culture/coculture
aliquots were assessed at 4, 7, 10, and 14 days for SIV RNA
and DNA levels by real-time RT-PCR, as described above.
Viral isolates
HIV isolate, HIV-1
Bal
, and HIV-1
JRFL
, were obtained from
the NIH Reference and Reagent Program. The SIV
Mac251
virus
was generously provided by Preston Marx, Tulane Regional
Primate Center, Covington, LA. Viral isolates were expanded
on CEMx174-R5 or MT4-R5 cell lines in RPMI media con-
taining 10% fetal calf serum for 3 weeks at 378C. Culture su-
pernatants were clarified by centrifugation (250 rcf for 10 min)
and vacuum filtered using the Steriflip apparatus (Millipore).
An SIV
Mac239
molecular infectious clone, containing the full
length proviral DNA genome in a modified pBR322 plasmid
vector, was a gift from Toshi Kodama, University of Pitts-
burgh School of Medicine, Pittsburgh, PA (GenBank acces-
sion: M33262).
24,25
To examine the inhibitory capacity of
saliva against a neutralization-sensitive SIV envelope, we re-
placed the gp160 region of the SIV
Mac239
molecular clone with
the predominant viral envelope of the SIV
DeltaB670
quasis-
pecies, an isolate shown to be sensitive to plasma neutrali-
zation in vitro.
26
SIV
Mac239-Cl3env
is a molecular infectious
clone that contains the SIV
DeltaB670
-clone 3 envelope (GenBank
accession: FJ842859) in an SIV
Mac239
background.
To obtain virus stocks, plasmids containing the proviral
genomes of SIV
Mac239
and SIV
Mac239-Cl3env
(3 mg DNA) were
transfected, using the Fugene 6 transfection reagent (Roche),
into 110
6
HEK293T cells seeded overnight in T25 cell
culture flasks at 378C. Culture supernatants were harvested
after 72 h at 378C and were used to infect 510
5
MT4-R5 cells
in RPMI containing 10% fetal calf serum for expansion of
SIV
Mac239
and SIV
Mac239-Cl3env
virus stocks. After 3 weeks,
virus culture supernatants were prepared as described
above. All virus stocks were aliquoted and stored at –808C.
The50%tissuecultureinfectiousdose(TCID
50
)wasdeter-
mined by serial dilution of virus stocks and infection of
TZM-bl cells as described.
27
Inhibition assay
Inhibition of HIV and SIV isolates by human and rhesus
macaque whole saliva was evaluated using an in vitro neu-
tralization assay previously described for evaluation of HIV
plasma neutralizing antibody responses.
27–29
This assay uti-
lizes TZM-bl cells, a HeLa cell line that stably expresses CD4,
CXCR4, and CCR5 receptors and contains a tat-responsive
firefly luciferase (Luc) reporter gene. This cell line requires
only a single round of infection to accurately measure in vitro
neutralization of virus. Briefly, human or macaque whole
saliva samples were diluted in DMEM with 10% fetal calf
serum (FCS) and incubated with 100 TCID
50
s of cell-free virus
at 378C in 96-well plates for 1 h prior to addition of cells. TZM-
bl cells were counted and diluted in DMEM containing 10%
FCS and 75 mg/ml DEAE-Dextran (Sigma); 8000 cells were
added to each reaction in a final volume of 250 ml per well.
Plates were then incubated at 378C for 48 h. Luminescence for
each reaction was assessed using the Bright-Glo Luciferase
Assay System (Promega) and quantified in 96-well black
plates using a Hidex Oy CHAMELEON V plate reader. In-
hibition of virus by saliva samples was expressed as the re-
duction in relative light units (RLU) as compared to virus
controls and reported as the percent reduction in viral infec-
tion in vitro. Using 100 TCID
50
doses, virus only control wells
produced RLU that were on average 10 times those of back-
ground. All virus stocks used in these studies produced
similar RLU in control wells. Saliva samples were assayed in
duplicate; cell only (negative) and virus only (positive) con-
trols were included in replicates of eight within each plate.
Only samples that displayed a 50% or more reduction in RLU
as compared to the average of control wells were considered
to have quantifiable inhibitory effects.
To evaluate potential cytotoxic effects of whole saliva on
TZM-bl cells, viability and metabolic activity of cells treated
with 1:10 and 1:20 dilutions of whole saliva were assessed and
compared to untreated controls. Cell viability counts using
Trypan blue staining were similar in treated and control cul-
tures, as were measurements of background RLU following
analysis in the Bright-Glo Luciferase Assay. Similarly, no
differences in metabolic activity were observed following
culture of TZM-bl cells in the presence of saliva as compared
to mock-treated control wells when assessed by a cell prolif-
eration MTS assay (Promega CellTiter 96 AQ
ueous
One Solu-
tion Cell Proliferation Assay), according to manufacturer
protocols.
30
Measurement of total and gp130 (SIV-envelope-specific)
antibody levels by ELISA
Total IgG and IgA levels in macaque saliva were measured
by ELISA as previously described.
31
Briefly, Fisherbrand mi-
crotiter plates were coated overnight at 48C with 100 ml PBS,
pH 7.4 containing 0.4 mg/ml goat F(ab0)
2
to monkey IgG
(1.2 mg/ml total protein; MP Biochemicals, Solon, OH) or
SIV INHIBITION BY WHOLE SALIVA 903
containing 0.5 mg/ml goat antimonkey IgA (Alpha Diag-
nostics, San Antonio, TX). Blocking was done at room tem-
perature for 30 min in PBS (1) containing 0.05% Tween-20
and 2% goat serum (Equitech-Bio Inc., Kerrville, TX). Whole
saliva was diluted 1:1000 in blocking buffer, added to the first
row of wells, and then further diluted in a 2-fold series down
the plate. The standard was pooled normal monkey serum
containing known amounts of IgA and IgG.
32
After overnight
storage at 48C, plates were developed by consecutive treat-
ment with biotin-conjugated goat antihuman IgG (South-
ernBiotech, Birmingham, AL) or antimonkey IgA (Alpha
Diagnostics) for 1 h at 378C, then 0.5 mg/ml streptavidin-
peroxidase (Sigma) for 30 min at room temperature. Tetra-
methylbenzidine Supersensitive substrate (Sigma) was added
for 30 min, followed by 1M H
2
SO
4
stop solution. Absorbance
was read at 450 nm in a SpectraMax 5 plate reader (Molecular
Devices, Sunnyvale, CA). The SoftMax Pro computer pro-
gram (Molecular Devices) was then used to construct four-
parameter standard curves and calculate the total IgG and
IgA in saliva samples.
SIV envelope-specific IgG concentrations were similarly
measured by ELISA using recombinant SIV
mac251
gp130
(ImmunoDiagnostics, Woburn, MA) as coating reagent, 1:50
starting dilutions of saliva, and a standardized pool of plasma
from SIV-infected macaques. Plates were developed with
secondary antibody to IgG as described above.
Statistical analyses
All statistical comparisons in this study were performed by
the Mann–Whitney Utest. pvalues <0.05 were considered
significant.
Results
SIV levels in the oral cavity of rhesus macaques
Whole saliva and peripheral blood samples were obtained
from 16 SIV
Mac251
-infected rhesus macaques over a 48-week
period following inoculation. SIV RNA levels were quantified
in plasma and saliva by real-time TaqMan RT-PCR and
compared (Fig. 1). As previously shown in HIV-infected in-
dividuals, detectable levels of viral RNA were present in the
saliva of SIV-infected rhesus macaques.
4,6
The level of viral
RNA present in saliva of each animal was approximately 2–3
logs lower than time-matched plasma viral RNA levels. Only
20 of the 111 total saliva samples evaluated (18%) contained
undetectable levels of SIV RNA (less than 100 viral RNA
copies/ml of saliva). These 20 samples were collected from 10
of the 16 rhesus macaques between 4 and 32 weeks following
infection, representing 1–2 time points per animal. Therefore,
during the majority of the disease course, each animal in the
cohort had detectable levels of virus in their saliva, with
several time points having >1000 copies/ml. The highest
levels of SIV RNA in macaque saliva were observed at 2
weeks postinoculation (wpi), ranging between 150 and
340,000 copies/ml of saliva (Fig. 1).
Despite the detectable levels of SIV RNA measured in
macaque saliva, we were unable to culture virus from the
saliva of SIV-infected macaques collected during the acute
(2 wpi) or chronic (18 months pi) stages of SIV disease using
rhesus macaque PBMCs or CEMx174-R5 and MT4-R5 T cell
lines. We were also unable to detect productive SIV infection
through coculture of the cellular fraction of the same acute
and chronic stage whole saliva samples.
To examine the genotypic diversity of SIV present in sa-
liva, we amplified the V1/V2 region of the SIV envelope by
RT-PCR in matched macaque plasma and saliva samples
collected over 48 weeks following infection and examined
the resulting amplicons by HTA. Samples were evaluated
from each of the 16 SIV-infected macaques over the course of
disease; four animals were selected as representative exam-
ples of the cohort, and these data are shown in Fig. 2, along
with the longitudinal SIV RNA levels in plasma and saliva.
We were unable to amplify envelope V1 PCR products from
all saliva samples, particularly samples with unquantifiable
viral RNA levels. HTA was performed on all PCR-positive
saliva samples, as shown in Fig. 2, with amplicons from time-
matched plasma samples (boxes). Plasma samples collected
from SIV
Mac251
-infected macaques at 1 or 2 wpi contained
four predominant envelope genotypes, which were also
present in saliva. These four genotypes are representative of
the SIV
Mac251
stock used for inoculation of the animals (data
not shown). Some of the inoculum envelope genotypes
persisted in plasma and saliva of each animal throughout the
disease course, but new genotypes appeared in both com-
partments at various time points (typically appearing from
8 to 12 wpi). Late in the disease course (30–48 wpi), saliva
samples from animals A, B, and C contained only one or two
of the predominant V1 genotypes found in more diverse
plasma samples. The loss of genotypic heterogeneity
observed in saliva over the disease course may reflect limi-
tations in assay sensitivity for genotypes present at very low
levels, lower viral levels, and/or the presence of antiviral
salivary factors. Animal D had high levels of saliva SIV RNA
and succumbed early to disease at 4 months following
infection. At end-stage disease in animal D (16 wpi), four
envelope genotypes were identified in both plasma and
saliva.
Among the cohort of macaques, the SIV envelope geno-
types present in saliva reflected the genotypes present in
plasma. In a few instances, we did observe unique envelope
V1 genotypes present in saliva, which were not found in
matched plasma samples. However, the majority of saliva SIV
genotypes were observed in plasma samples at concurrent or
earlier time points in the disease course.
In vitro measurement of SIV inhibition by human
and rhesus macaque whole saliva
To determine whether whole human or rhesus macaque
saliva was capable of inhibiting SIV infection in vitro,we
collected whole saliva samples from eight normal human
volunteers and eight normal rhesus macaques. The inhibi-
tory capacity of each normal human and macaque saliva
(diluted 1:20) against two isolates of SIV, SIV
Mac239
and
SIV
Mac239-Cl3env
, was evaluated by their ability to inhibit
infection of MT4-R5 T cells. Both normal human and rhesus
macaque whole saliva samples from this collection were
capable of near complete inhibition of SIV
Mac239-Cl3env
and
variable levels of SIV
Mac239
inhibition, demonstrating that
whole saliva can inhibit SIV infection in vitro. However,
infection levels in the MT4-R5 cell line by both SIV isolates
were variable across replicate wells, making standardization
of this assay difficult.
904 THOMAS ET AL.
Weeks Post-Infection
1E+1
1E+2
1E+3
1E+4
1E+5
1E+6
1E+7
1E+8
1E+9
SIV RNA level (copies/ml)
Plasma
Saliva
12 4 6 8 12 16 30-32 48
FIG. 1. Levels of SIV RNA in time-matched plasma (&) and saliva (~) samples collected from SIV
Mac251
-infected rhesus
macaques (n¼16) over 48 weeks following inoculation. The real-time RT-PCR assay quantitation limit of 100 copies/ml is
indicated on the chart by the dotted line. Samples shown that are below the dotted line contain less than 100 copies of SIV
RNA/ml fluid.
FIG. 2. SIV envelope V1 genotyping by heteroduplex tracking assay (HTA) on time-matched plasma and saliva samples
collected from four cohort-representative SIV
Mac251
-infected macaques over 48 weeks following inoculation. (A, B, C, and D)
32
P-labeled V1 DNA banding patterns in the upper portion and time-coincident SIV RNA levels in the lower portion for
plasma and saliva samples over the SIV disease time course collected from each of the four representative animals. Time-
matched plasma and saliva samples are outlined in boxes. The dotted lines indicate the lower assay limit for measurement of
SIV RNA levels.
SIV INHIBITION BY WHOLE SALIVA 905
To develop a more sensitive, standardized assay to evalu-
ate the virus-inhibitory capacity of saliva in vitro, we adapted
an assay commonly used to measure antibody neutralization
of HIV. Utilizing the TZM-bl cell line, we evaluated the in-
hibitory capacity of a set of normal human and rhesus ma-
caque whole saliva against five viral isolates. Comparisons of
human and monkey saliva against a specific virus were con-
ducted in a single experiment. The results are shown in Fig. 3.
Percent inhibition of HIV
BaL
, an HIV-1 isolate commonly
used to examine human salivary inhibition, varied widely
among the eight normal human saliva samples tested, rang-
ing from <10% to 75% inhibition. Human whole saliva was
also capable of inhibiting HIV
JRFL
infection to a similar level.
Rhesus macaque whole saliva was less inhibitory to the HIV
isolates tested as compared to human saliva, with only one
rhesus sample capable of inhibiting either HIV isolate by more
than 50% in this assay. All macaque saliva samples (eight of
eight) and six of eight human saliva samples tested were ca-
pable of inhibiting SIV
Mac251
infection. All normal human and
macaque saliva samples were capable of inhibiting SIV
Mac239
and SIV
Mac239-Cl3env
infection by greater than 50%. SIV
Mac239-
Cl3env
was the isolate most sensitive to inhibition by human
and macaque saliva, with all saliva samples in this collection
inhibiting SIV
Mac239-Cl3env
infection by greater than 75% in this
assay. These data indicate that both normal human and rhe-
sus macaque whole saliva can inhibit in vitro infection of SIV,
and that SIV
Mac239-Cl3env
is highly sensitive to inhibition
in vitro, providing a tool to evaluate the inhibitory capacity of
macaque whole saliva. Samples which exhibited <50% inhi-
bition showed varying levels of inhibition ranging between 0
and 50% in replicate experiments, and quantification of this
amount was not reproducible with values below 50%.
Therefore, these saliva samples were categorized as negative
for virus inhibition.
Volumes of whole saliva obtained from anesthetized
SIV
Mac251
-infected macaques in this study were limited, and
we were unable to perform repeated SIV inhibition mea-
surements on individual saliva samples collected from many
animals in the cohort. To assess the variability in SIV inhibi-
tory capacity of multiple saliva samples collected from a sin-
gle animal, saliva samples (n¼7) were collected over 3 months
from two animals and SIV
Mac239-Cl3env
inhibition was mea-
sured using the TZM-bl assay. We observed consistent,
reproducible levels of SIV inhibition by replicate saliva sam-
ples collected from both animals; average percent inhibition
by saliva was 89 4.5 (standard deviation) and 82 6.4.
Because the volume of saliva collected varied among ani-
mals, we assessed the quality of samples collected by mea-
suring protein levels. Total protein levels were measured by a
BCA assay in all macaque whole saliva samples collected.
Levels of total salivary proteins among individual macaques
ranged between 2000 and 4000mg/ml saliva and were not
associated with same-sample measures of SIV inhibitory
capacity in vitro. These data suggest that the quality of saliva
samples collected from macaques in this study was compa-
rable.
The small volumes of saliva available from monkeys also
limited our ability to analyze the SIV-inhibitory capacity of
FIG. 3. Inhibition of HIV and SIV infection by normal human (n¼8) and rhesus macaque (n¼8) whole saliva was
evaluated using the TZM-bl assay. Saliva samples were diluted 1:20 and measured for inhibitory capacity against five viral
isolates: HIV-1
BaL
, HIV-1
JRFL
, SIV
Mac251
, SIV
Mac239
, and SIV
Mac239-Cl3env
. Percent inhibition of each virus is shown for normal
human (&) and normal macaque (~) saliva samples. Gray area indicates saliva samples that were negative for virus
inhibition in this assay (less than 50% inhibition of viral isolate).
906 THOMAS ET AL.
saliva at different dilutions. To determine the range of saliva
dilutions capable of inhibiting SIV in vitro, a collection of
samples with larger volumes available was diluted 1:20, 1:80,
and 1:200 and evaluated for levels ofSIV
Mac239-Cl3env
inhibition
using the TZM-bl assay. Based on assessment of replicate ex-
periments, wedetermined that saliva samples diluted 1:20 that
demonstrated >75% inhibition were also capable of inhibiting
infection by >50% at a dilution of 1:80, whereas saliva samples
with inhibitory levels between 50% and 75% of controls were
not capable of inhibiting SIV infection at a higher dilution of
saliva. Therefore to classify the SIV inhibitory capacity of
volume-limited macaque saliva following analysis at a single
dilution of saliva (1:20), we used the designations of less than
50% inhibition of SIV
Mac239-Cl3env
infection as having little to no
SIV-inhibitory capacity, greater than 50% inhibition as having
SIV-inhibitory capacity, and greater than 75% as having a
higher titer of inhibitory factors (1:80).
Inhibitory capacity of rhesus macaque saliva
over the SIV disease course
The SIV-inhibitory capacity of whole saliva collected from
the cohort of 16 SIV
Mac251
-infected macaques was evaluated
over 48 weeks following inoculation, as shown in Fig. 4.
During the first 8 weeks following SIV infection, a wide range
of inhibitory capacity among individual animals was ob-
served; however, the majority of macaque saliva samples
tested demonstrated >50% inhibition of SIV infection. Saliva
collected from eight normal, untreated macaques all demon-
strated levels of inhibition of >75% against SIV
Mac239-Cl3env
(Fig. 3). The wider range of SIV inhibition observed by saliva
from the SIV-infected macaque cohort may be due to SIV in-
fection itself or may represent the variability of saliva inhibi-
tory capacity among a larger group of macaques.
At 2 wpi, 14 of 16 of animals had positive saliva SIV-
inhibitory capacities (>50% inhibition), 8 of which were >75%
inhibition, and 2 of 16 animals had negative saliva
SIV-inhibitory capacities (<50% inhibition). Averages among
the cohort are shown in Fig. 5A. At 8 wpi, all animals had
positive saliva SIV-inhibitory capacities, with half of the ani-
mals demonstrating >75% inhibition of SIV infection. By
16 wpi, all saliva samples exhibited positive SIV-inhibitory
capacities of >75% inhibition. Although we did not observe
any significant changes in inhibitory capacity during the acute
stage of SIV disease (2–8 wpi), there was a significant im-
provement in the SIV-inhibitory capacity of saliva later in the
disease course (16–48 wpi). The improvement in levels of SIV
inhibition by infected macaque saliva was statistically
significant between 2 and 16 wpi ( p<0.001) and between 8
and 16 wpi ( p<0.01). The SIV-inhibitory capacity of infected
macaque saliva remained statistically higher ( p<0.05) than
the 8 wpi time point throughout the disease course. There
was no correlation between salivary SIV
Mac239-Cl3env
inhibi-
tory capacity and salivary SIV RNA levels over the disease
course; however, at 16 wpi, virus levels in the saliva of all
animals were less than 1000 copies/ml and remained low to
undetectable (less than 100 copies/ml saliva) through
48 weeks. These data suggest that the development of
effective adaptive host immune responses in the oral cavity
between 8 and 16 weeks following infection may contribute
FIG. 4. Saliva samples (~) were collected from 16 SIV
Mac251
-infected rhesus macaques at the indicated time points fol-
lowing inoculation. Saliva was diluted 1:20 and inhibitory capacity against SIV
Mac239-Cl3env
was measured by TZM-bl assay
for each animal over the course of disease. Changes in salivary inhibitory capacity were evaluated for statistical significance
by the Mann–Whitney Utest: (*) p<0.001; ({)p<0.01; ({)p<0.05.
SIV INHIBITION BY WHOLE SALIVA 907
to limiting expression of virus and to the SIV-inhibitory
capacity of saliva.
To evaluate humoral immune responses in saliva of SIV-
infected macaques, we examined total IgG and IgA antibody
levels in macaque saliva as well as SIVgp130-specific IgG
antibody levels. Figure 5 shows the comparison between
salivary inhibition of SIV
Mac239-Cl3env
and antibody levels
over the course of disease. There were no significant changes
in the levels of total IgG and IgA antibodies in macaque
saliva corresponding to the increase in saliva inhibition of
SIV infection in vitro over the course of disease. However, we
did observe a statistically significant increase in SIVgp130-
specific IgG levels in SIV
Mac251
-infected macaque saliva be-
tween 6–8 wpi and 28–30 wpi ( p<0.01). The increases in
SIV-specific humoral immune responses are coincident with
the improvement in the SIV-inhibitory capacity of saliva
observed later in the disease course of this cohort and sug-
gest that SIV-specific antibody responses in the oral cavity of
macaques contribute to the inhibition of SIV infection in vitro
by saliva.
Discussion
In this study, we have demonstrated that SIV levels in the
oral cavity of rhesus macaques are highly similar to previ-
ously reported viral loads in HIV-infected humans. SIV
Mac251
-
infected macaques (n¼16) had detectable levels of SIV RNA
present in saliva over the course of disease. Among individual
animals, saliva SIV RNA levels were 2–3 logs lower than the
corresponding plasma levels and reflected fluctuations in the
plasma SIV RNA levels over the disease course. However,
among the animal cohort, we did not observe a direct corre-
lation between viral RNA levels in plasma and saliva at any
time point, in agreement with a recent report by Whitney
et al.
18
In that report, viral RNA was detected over a pro-
longed period of time in the blood and saliva of five SIV
Mac251
-
infected macaques, similar to our observations. Analysis of
the genotypic diversity of virus populations in matched
macaque plasma and saliva samples indicated the presence of
identical SIV envelope variants in both fluids, similar to the
observations in HIV-infected humans.
8,9
Attempts to culture
FIG. 5. Comparison of total and SIV envelope-specific antibody levels in saliva collected from 16 SIV
Mac251
-infected rhesus
macaque saliva at various time points over the course of disease to SIV-inhibitory capacity in vitro.(A) The average percent
inhibition of SIV
Mac239-Cl3env
infection by all saliva samples over time. Salivary antibody levels in each saliva sample were
measured by ELISA, and total IgG and IgA levels in saliva (B) and SIVgp130 IgG (binding antibody) levels in saliva (C) are
shown for select time points over the disease course. Changes in antibody levels were assessed for statistical significance by
the Mann–Whitney Utest; ({)p<0.01.
908 THOMAS ET AL.
or coculture SIV from infected macaque saliva were unsuc-
cessful, indicating that like human saliva, macaque saliva may
contain inhibitory factors that reduce the viability of virus in
the oral cavity.
5
Multiple endogenous human salivary factors with specific
and nonspecific mechanisms of anti-HIV activity have
been described since Fultz first demonstrated inhibition of
HIV infection in vitro by whole human and chimpanzee
saliva.
10,33,34
Despite the widespread use of SIV-infected
rhesus macaques as an animal model for HIV transmission
and pathogenesis, few studies have utilized this model to
evaluate the inhibitory components of macaque saliva. One
study, by Nagashunmugam et al., was unable to demonstrate
any inhibitory activity of human submandibular saliva
against HIV-2
ROD
or SIV
Mac239
infection of HUT78 cells
in vitro, as measured by reverse transcriptase activity at 7 days
postinfection.
13
In this study, we first examined the SIV-
inhibitory capacity of normal human and macaque saliva
against SIV
Mac239
and SIV
Mac239-Cl3env
using an MT4-R5 T cell
inhibition assay. We were able to confirm saliva inhibition of
SIV infection in vitro using the MT4-R5 cell line. However,
variable levels of infection in positive controls were observed
by both SIV isolates, making standardized and sensitive as-
sessment of SIV inhibition by saliva difficult. To address
similar difficulties with assessment of plasma neutralization
of HIV isolates in vitro, rapid and sensitive assays for mea-
surement of HIV-1 neutralization have recently been stan-
dardized using the TZM-bl reporter cell line.
28,29
We utilized
this assay to examine inhibition of viral infection by normal
human and rhesus macaque saliva, and both were capable of
inhibiting in vitro infection of HIV and SIV isolates to varying
degrees using the TZM-bl assay. Therefore, lack of assay
sensitivity may have played a role in the previous inability to
measure saliva inhibition of SIV infection.
13
The virus isolate used in previous studies may have also
contributed to the inability of saliva to inhibit SIV infection
in vitro.SIV
Mac239
is a lab-adapted isolate of SIV and has
been characterized as resistant to in vitro neutralization by
plasma antibodies.
35
It has been reported that characteristics
of SIV
Mac239
determined in vitro, such as cell tropism, were
not predictive of in vivo tropism or pathogenesis of the virus,
indicating that SIV
Mac239
may not be an ideal isolate to ex-
amine saliva inhibition against SIV in vitro.
36
For assessment
of in vitro neutralization of primary or biological HIV-1
isolates, investigators have utilized pseudotyping and mo-
lecular infectious cloning methods to limit T cell line pas-
sage and allow for more accurate measurement of antibody-
mediated neutralization of heterologous HIV-1 envelopes.
37
We hypothesized that the use of an SIV molecular infectious
clone containing a plasma neutralization-sensitive envelope
would allow for more accurate measurement of inhibition of
SIV infection in vitro by macaque saliva. SIV
Mac239-Cl3env
,
used in this study, is a replication-competent virus created
from an SIV
Mac239
molecular infectious clone and contains
a neutralization-sensitiveenvelopederivedfromthe
SIV
DeltaB670
quasispecies.
24,38
In this study, SIV
Mac239-Cl3env
was the viral isolate most sensitive to inhibition by saliva
using the TZM-bl assay.
Utilizing a cohort of 16 SIV-infected macaques, at 2 wpi, we
observed varying levels of SIV-inhibitory capacity by saliva,
ranging from <50 to >90% inhibition, with average levels of
72%. In contrast, the collection of normal macaque saliva
(shown in Fig. 3) exhibited more uniform levels of inhibition
against SIV
Mac239-Cl3env
, where each of the eight samples had
>75 % inhibition. Samples were not available from time
points prior to SIV infection of the 16 animal cohort to de-
termine if SIV infection had an impact on the inhibitory ca-
pacity of saliva. It is also possible that these results represent
the variation found among different animals, which was not
reflected in the small cohort of the eight normal macaques
evaluated.
Among the SIV-infected cohort, we observed a significant
increase in the SIV-inhibitory capacity of saliva collected
16 wpi, corresponding to a statistically significant increase in
levels of SIV-specific IgG in saliva and reduced salivary SIV
RNA levels (less than 1000 copies/ml saliva). The develop-
ment of SIV-specific humoral immune responses may play a
key role in the control of viral replication in the oral cavity of
infected macaques and together with endogenous antiviral
factors in saliva, these responses may serve to limit virus in
this compartment.
Many endogenous factors have been identified that con-
tribute to the antiviral activity of human saliva in vivo and
in vitro.
33
The current study demonstrates that both normal
human and macaque whole saliva are capable of inhibiting
SIVandHIVinfectionin vitro using a sensitive, standard-
ized assay. These data suggest that the same innate antiviral
components identified in human saliva may also be present
in macaque saliva, and that the inhibitory activities of these
salivary components are not limited to HIV-1. Utilization of
the rhesus macaque model to study viral pathogenesis in
the oral cavity may be of particular importance in HIV
vaccine and treatment studies. Recent reports indicate that
frequent or high-risk contact between HIV serodiscordant
individuals elicits protective, HIVgp160-specific neutraliz-
ing antibody responses at multiple mucosal surfaces,
including IgA responses in the genital tract secretions of
HIV-negative female Kenyan sex workers and in the
parotid saliva of HIV-negative babies exposed to breast milk
from HIV-infected mothers.
39–41
Further examination of
these findings is necessary to determine if these mucosal
HIVgp160-specific antibodies may truly be protective
against HIV infection in vivo or have potential as a pro-
phylactic therapy. Although more work is needed to iden-
tify the specific protective factors present in macaque saliva
and to determine similarities of these factors to human
salivary molecules with anti-HIV activity, our findings
demonstrate the relevance of the SIV-infected rhesus
macaque for studying HIV pathogenesis in the oral cavity,
as well as for identifying innate and acquired responses
protective against HIV at mucosal surfaces.
Acknowledgments
We would like to thank the LSU Comprehensive Alcohol
Research Center, Core Laboratories, Jason Dufour, DVM
(Tulane Regional Primate Center), Ashok Aiyar, Ph.D.
(LSUHSC), and Patricia Molina, M.D., Ph.D. (LSUHSC), for
support, technical help, and macaque specimens. We thank
Jane Schexnayder for technical assistance and James Ro-
binson, M.D. (Tulane HSC) for technical advice and cell lines.
This work was supported, in part, by the Louisiana Vaccine
Center and the South Louisiana Institute for Infectious Dis-
ease Research sponsored by the Louisiana Board of Regents
SIV INHIBITION BY WHOLE SALIVA 909
and by NIH Grants AA009803 (S. Nelson), AA007577 (G.J.
Bagby), and AI058896 (P.A. Kozlowski).
Author Disclosure Statement
No competing financial interests exist.
References
1. Hall HI, Song R, Rhodes P, et al.: Estimation of HIV inci-
dence in the United States. JAMA 2008;300(5):520–529.
2. Campo J, Perea MA, del Romero J, Cano J, Hernando V, and
Bascones A: Oral transmission of HIV, reality or fiction? An
update. Oral Dis 2006;12(3):219–228.
3. Moutsopoulos NM, Greenwell-Wild T, and Wahl SM: Dif-
ferential mucosal susceptibility in HIV-1 transmission and
infection. Adv Dent Res 2006;19(1):52–56.
4. Liuzzi G, Clementi M, Bagnarelli P, Valenza A, Cataldo PT,
and Piazza M: Analysis of HIV-1 load in blood, semen and
saliva: Evidence for different viral compartments in a cross-
sectional and longitudinal study. AIDS 1996;10(14):F51–F56.
5. Rothenberg RB, Scarlett M, del Rio C, Reznik D, and
O’Daniels C: Oral transmission of HIV. AIDS 1998;12(16):
2095–2105.
6. Shugars DC, Slade GD, Patton LL, and Fiscus SA: Oral and
systemic factors associated with increased levels of human
immunodeficiency virus type 1 RNA in saliva. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 2000;89(4):
432–440.
7. Shugars DC and Wahl SM: The role of the oral environment
in HIV-1 transmission. J Am Dent Assoc 1998;129(7):
851–858.
8. Freel SA, Williams JM, Nelson JAE, et al.: Characterization of
human immunodeficiency virus type 1 in saliva and blood
plasma by V3-specific heteroduplex tracking assay and ge-
notype analyses. J Virol 2001;75(10):4936–4940.
9. Freel SAFS, Pilcher CD, Menezes P, Giner J, Patrick E, Len-
nox JL, Hicks CB, Eron JJ Jr, and Shugars DC: Envelope
diversity, coreceptor usage and syncytium-inducing pheno-
type of HIV-1 variants in saliva and blood during primary
infection. AIDS 2003;17(14):2025–2033.
10. Shugars DC: Endogenous mucosal antiviral factors of the
oral cavity. J Infect Dis 1999;179(Suppl 3):S431–S435.
11. Fultz PN: Components of saliva inactivate human immu-
nodeficiency virus. Lancet 1986;2(8517):1215.
12. Shine N, Konopka K, and Duzgunes N: The anti-HIV-1 ac-
tivity associated with saliva. J Dent Res 1997;76(2):634–640.
13. Nagashunmugam T, Davis C, Kennedy S, Goldstein LT, and
Malamud D: Human submandibular saliva specifically in-
hibits HIV type 1. AIDS Res Hum Retroviruses 1997;13(5):
371–376.
14. Nagashunmugam T MD, Davis C, Abrams WR, and Fried-
man HM: Human submandibular saliva inhibits human
immunodeficiency virus type 1 infection by displacing en-
velope glycoprotein gp120 from the virus. J Infect Dis
1998;178:1635–1641.
15. Malamud D, Berthold P, Roth E, and Friedman HM: Human
submandibular saliva aggregates HIV. AIDS Research and
Human Retroviruses. 1993;9:633–637.
16. Challacombe SJ and Sweet SP: Oral mucosal immunity and
HIV infection: Current status. Oral Dis. 2002;8(2):55–62.
17. Lackner AA and Veazey RS: Current concepts in AIDS
pathogenesis: Insights from the SIV/macaque model. Annu
Rev Med. 2007;58:461–476.
18. Whitney JB, Luedemann C, Bao S, et al.: Monitoring HIV
vaccine trial participants for primary infection: Studies in the
SIV/macaque model. AIDS 2009;23(12):1453–1460.
19. Kozlowski PA, Lynch RM, Patterson RR, Cu-Uvin S, Flani-
gan TP, and Neutra MR: Modified wick method using
Weck-Cel sponges for collection of human rectal secretions
and analysis of mucosal HIV antibody. J Acquir Immune
Defic Syndr 2000;24(4):297–309.
20. Bagby GJ, Zhang P, Purcell JE, Didier PJ, and Nelson S:
Chronic binge ethanol consumption accelerates progression
of simian immunodeficiency virus disease. Clin Exp Res
2006;30(10):1781–1790.
21. Amedee A, Rychert J, Lacour N, Fresh L, and Ratterree M:
Viral and immunological factors associated with breast milk
transmission of SIV in rhesus macaques. Retrovirology
2004;1(1):17.
22. Rychert J, Lacour N, and Amedee AM: Genetic analysis of
simian immunodeficiency virus expressed in milk and se-
lectively transmitted through breastfeeding. J Virol 2006;
80(8):3721–3731.
23. Krowicka H, Robinson JE, Clark R, Hager S, Broyles S, and
Pincus SH: Use of tissue culture cell lines to evaluate HIV
antiviral resistance. AIDS Res Hum Retroviruses 2008;24(7):
957–967.
24. Kestler HW 3rd, Naidu YN, Kodama T, et al.: Use of in-
fectious molecular clones of simian immunodeficiency vi-
rus for pathogenesis studies. J Med Primatol 1989;18(3–4):
305–309.
25. Kodama T, Burns DP, Kestler HW 3rd, Daniel MD, and
Desrosiers RC: Molecular changes associated with replica-
tion of simian immunodeficiency virus in human cells. J Med
Primatol 1990;19(3–4):431–437.
26. Cole KS, Rowles JL, Jagerski BA, et al.: Evolution of enve-
lope-specific antibody responses in monkeys experimentally
infected or immunized with simian immunodeficiency virus
and its association with the development of protective im-
munity. J Virol 1997;71(7):5069–5079.
27. Montefiori DC: Evaluating neutralizing antibodies against
HIV, SIV, and SHIV in luciferase reporter gene assays. Curr
Protocall Immunol 2005;Chapter 12:Unit 12 11.
28. Mascola JR, D’Souza P, Gilbert P, et al.: Recommendations
for the design and use of standard virus panels to assess
neutralizing antibody responses elicited by candidate hu-
man immunodeficiency virus type 1 vaccines. J Virol 2005;
79(16):10103–10107.
29. Polonis VR, Brown BK, Rosa Borges A, et al.: Recent ad-
vances in the characterization of HIV-1 neutralization assays
for standardized evaluation of the antibody response to in-
fection and vaccination. Virology 2008;375(2):315–320.
30. Cory AH, Owen TC, Barltrop JA, and Cory JG: Use of an
aqueous soluble tetrazolium/formazan assay for cell growth
assays in culture. Cancer Commun 1991;3(7):207–212.
31. Kozlowski PA, Chen D, Eldridge JH, and Jackson S: Con-
trasting IgA and IgG neutralization capacities and responses
to HIV type 1 gp120 V3 loop in HIV-infected individuals.
AIDS Res Hum Retroviruses 1994;10(7):813–822.
32. Wang SW, Kozlowski PA, Schmelz G, et al.: Effective in-
duction of simian immunodeficiency virus-specific systemic
and mucosal immune responses in primates by vaccination
with proviral DNA producing intact but noninfectious viri-
ons. J Virol 2000;74(22):10514–10522.
33. Shugars DC, Alexander AL, Fu K, and Freel SA: En-
dogenous salivary inhibitors of human immunodeficiency
virus. Arch Oral Biol 1999;44(6):445–453.
910 THOMAS ET AL.
34. Cohen MS, Shugars DC, and Fiscus SA: Limits on oral
transmission of HIV-1. Lancet 2000;356(9226):272.
35. Choi WS, Collignon C, Thiriart C, et al.: Effects of natural
sequence variation on recognition by monoclonal antibodies
neutralize simian immunodeficiency virus infectivity. J Virol
1994;68(9):5395–5402.
36. Borda JT, Alvarez X, Kondova I, et al.: Cell tropism of
simian immunodeficiency virus in culture is not predictive
of in vivo tropism or pathogenesis. Am J Pathol 2004;
165(6):2111–2122.
37. Follis KE, Trahey M, LaCasse RA, and Nunberg JH: Con-
tinued utilization of CCR5 coreceptor by a newly derived T-
cell line-adapted isolate of human immunodeficiency virus
type 1. J Virol 1998;72(9):7603–7608.
38. Cole KS, Alvarez M, Elliott DH, et al.: Characterization of
neutralization epitopes of simian immunodeficiency virus
(SIV) recognized by rhesus monoclonal antibodies derived
from monkeys infected with an attenuated SIV strain. Vir-
ology 2001;290(1):59–73.
39. Hirbod T, Reichard C, Hasselrot K, et al.: HIV-1 neutralizing
activity is correlated with increased levels of chemokines in
saliva of HIV-1-exposed uninfected individuals. Curr HIV
Res 2008;6(1):28–33.
40. Hasselrot K, Saberg P, Hirbod T, et al.: Oral HIV-exposure
elicits mucosal HIV-neutralizing antibodies in uninfected
men who have sex with men. AIDS 2009;23(3):329–333.
41. Farquhar C, VanCott T, Bosire R, et al.: Salivary human
immunodeficiency virus (HIV)-1-specific immunoglobulin A
in HIV-1-exposed infants in Kenya. Clin Exp Immunol
2008;153(1):37–43.
Address correspondence to:
Angela M. Amedee
LSU Health Sciences Center
1901 Perdido Street, P6-1
New Orleans, Louisiana 70112
E-mail: aamede@lsuhsc.edu
SIV INHIBITION BY WHOLE SALIVA 911
