NK cell responses to HIV/SIV infection have been well studied in acute and chronic infected patients/monkeys, but little is known about NK cells during viral transmission, particularly in mucosal tissues. In this article, we report a systematic study of NK cell responses to high-dose vaginal exposure to SIVmac251 in the rhesus macaque female reproductive tract (FRT). Small numbers of NK cells were recruited into the FRT mucosa following vaginal inoculation. The influx of mucosal NK cells preceded local virus replication and peaked at 1 wk and, thus, was in an appropriate time frame to control an expanding population of infected cells at the portal of entry. However, NK cells were greatly outnumbered by recruited target cells that fuel local virus expansion and were spatially dissociated from SIV RNA+ cells at the major site of expansion of infected founder populations in the transition zone and adjoining endocervix. The number of NK cells in the FRT mucosa decreased rapidly in the second week, while the number of SIV RNA+ cells in the FRT reached its peak. Mucosal NK cells produced IFN-γ and MIP-1α/CCL3 but lacked several markers of activation and cytotoxicity, and this was correlated with inoculum-induced upregulation of the inhibitory ligand HLA-E and downregulation of the activating receptor CD122/IL-2Rβ. Examination of SIVΔnef-vaccinated monkeys suggested that recruitment of NK cells to the genital mucosa was not involved in vaccine-induced protection from vaginal challenge. In summary, our results suggest that NK cells play, at most, a limited role in defenses in the FRT against vaginal challenge.

Although antiretroviral therapies have converted HIV-1 infection to a chronic and largely manageable disease for many individuals, it is clear that ending the pandemic depends on more effective measures for prevention, particularly for the principal route of sexual mucosal transmission and in the population of young women in the pandemic’s epicenter in sub-Saharan Africa, who are the most at risk for heterosexual acquisition of HIV-1 (1). To that end, we have been seeking design principles for development of an effective vaccine to prevent HIV-1 transmission to women in the SIV-rhesus macaque nonhuman primate model. We report the results of an investigation into the role of NK cells in preventing transmission by analyses of their responses to high-dose vaginal challenge with pathogenic SIVmac251 in naive animals, as well as animals vaccinated with the live attenuated vaccine, SIV∆nef.

We focused these studies on the immunological and virological events of transmission in the early stages of infection at the portal of entry for two reasons. First, it was shown that the SIV-infected founder populations were initially quite small in the monkey female reproductive tract (FRT) mucosal tissues, despite vaginal exposure to high doses of SIVmac251 (2, 3). Consequently, there should be an opportunity for pre-existing or rapidly induced innate or adaptive immune responses to operate at high E:T ratios to prevent or restrict the establishment of the initial virus-replicating population. Second, because local expansion of the infected founder population usually precedes the dissemination of infection into the circulation and the establishment of a robust self-propagating systemic infection in mainly lymphoid tissues (2, 3), pre-existing and rapid immune responses could efficiently prevent or attenuate systemic infection by constraining this local expansion.

The innate immune system, particularly NK cells, could potentially play important roles in this early window of opportunity as a rapid responder population that could kill the infected cells through either Ab-dependent cellular cytotoxicity or Ab-independent (NK cytotoxicity) mechanisms (4). NK cells also express β-chemokines, including MIP-1α/CCL3, MIP-1β/CCL4, and RANTES/CCL5, which could inhibit continued propagation of infection by blocking viral entry (4, 5). Furthermore, preservation of NK cell functions has been associated with improved disease outcome in patients and monkeys chronically infected with HIV-1 and SIV, respectively (68). Increased NK cell activity has been correlated with protection in HIV-1 highly exposed seronegative subjects (9). Studies on the activating and inhibitory receptors on NK cells revealed a protective correlation between certain KIR and HLA polymorphisms (1013), as well as a role for NK cells in controlling viral infection in vivo based on HIV-1 adaptation to KIR receptors (14). However, little is known about NK cell responses in mucosal tissues, particularly at the early stage of SIV/HIV mucosal transmission. We address, in part, that gap in our knowledge of the role of NK cells in protecting the FRT in early infection by analyzing and comparing NK cell responses in the lower FRT mucosa to high-dose vaginal challenge with pathogenic SIVmac251 in naive animals, as well as in animals vaccinated with the live attenuated vaccine, SIV∆nef.

Archived genital tissues from previous studies of SIV high-dose vaginal infection (2) and SIV Δnef vaccines (15, 16) were used in this study. Briefly, fresh tissues obtained at necropsy were fixed in 4% paraformaldehyde, Streck tissue fixative, or SAFEFIX II (Fisher Scientific, Kalamazoo MI) Tissue Fixative and embedded in paraffin, as previously reported (2).

Single and double immunohistochemical staining, fluorescent immunohistochemical staining, and quantitative image analysis on confocal microscopy were performed as described elsewhere (17, 18). The primary Abs used in this study are summarized in Table I.

Table I.
Primary Abs used in this study
AbsClone (Catalog No.)Tissue FixationAg Retrieval
NKG2A Epitomics (T3308) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
Granzyme H Sigma (HPA029200) PFA 1 mM EDTA buffer (pH 8) 98°C 20 min 
Granzyme B Spring Bioscience (E2580) PFA 1 mM EDTA buffer (pH 8) 98°C 20 min 
IFN-γ Abcam (ab25101) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
IL-2Rβ (CD122) Novus Biologicals (NBP1-19140) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
HLA-E MEM-E/06 (ab3984) PFA 1 mM EDTA buffer (pH 8) 98°C 20 min 
CD3 AbD Serotec (MCA 1477) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CCL3/MIP-1α Neomarkers (RB-10489-P) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CCL4/MIP-1β R&D (AF-271-NA) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CCL5/RANTES R&D (AF-278-NA) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
Ki67 Neomarkers (RM9106) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
HLA-DR/DQ/DP DAKO (M0073) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CD69 Novacastra (NCL-CD69 CH11) SAFEFIX II 1 mM EDTA buffer (pH 8) 98°C 20 min 
CD38 Vector (VP-C348 SPC32)  SAFEFIX II 1 mM EDTA buffer (pH 8) 98°C 20 min 
IP10 R&D Systems (AF-266-NA) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CD107a LifeSpan (LS-B580) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CD68 DAKO (KP1 M0814) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
Vimentin Neomarkers (RB-9063) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
AbsClone (Catalog No.)Tissue FixationAg Retrieval
NKG2A Epitomics (T3308) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
Granzyme H Sigma (HPA029200) PFA 1 mM EDTA buffer (pH 8) 98°C 20 min 
Granzyme B Spring Bioscience (E2580) PFA 1 mM EDTA buffer (pH 8) 98°C 20 min 
IFN-γ Abcam (ab25101) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
IL-2Rβ (CD122) Novus Biologicals (NBP1-19140) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
HLA-E MEM-E/06 (ab3984) PFA 1 mM EDTA buffer (pH 8) 98°C 20 min 
CD3 AbD Serotec (MCA 1477) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CCL3/MIP-1α Neomarkers (RB-10489-P) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CCL4/MIP-1β R&D (AF-271-NA) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CCL5/RANTES R&D (AF-278-NA) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
Ki67 Neomarkers (RM9106) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
HLA-DR/DQ/DP DAKO (M0073) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CD69 Novacastra (NCL-CD69 CH11) SAFEFIX II 1 mM EDTA buffer (pH 8) 98°C 20 min 
CD38 Vector (VP-C348 SPC32)  SAFEFIX II 1 mM EDTA buffer (pH 8) 98°C 20 min 
IP10 R&D Systems (AF-266-NA) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CD107a LifeSpan (LS-B580) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
CD68 DAKO (KP1 M0814) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 
Vimentin Neomarkers (RB-9063) PFA 10 mM citrate buffer (pH 6) 98°C 20 min 

PFA, paraformaldehyde.

SIV RNA was detected in paraformaldehyde-fixed and paraffin-embedded tissues by in situ hybridization (ISH), as previously described (2). Briefly, 5-μm sections were deparaffinized in xylene, rehydrated in PBS, and permeabilized sequentially in HCl, digitonin, and proteinase K. The sections were then acetylated and hybridized to 35S-labeled SIV-specific riboprobes. After wash and digestion with ribonucleases, the sections were coated with nuclear-track emulsion before exposure and development. For fluorescent ISH, digoxigenin-labeled SIV-specific riboprobes were used, followed by sequential staining with Goat Anti-Digoxigenin Abs (Roche, Indianapolis, IN) and Donkey Anti-Goat Abs conjugated with Alexa Fluor 555 (Invitrogen, Eugene, OR) (17, 18).

The Wilcoxon rank-sum test was used to measure the variations in NK cell, macrophage, and SIV RNA+ cell densities over the course of infection, vaccination, and challenge. The paired t test was used to compare the number of NK cells between vaginal and cervical tissues from the same animal. The Spearman rank correlation test was used to analyze the relationship between mucosal NK cells and local SIV RNA+ cells. Statistical analyses were carried out using Prism 4 software.

We systematically examined early NK cell responses in the vaginal and cervical mucosa of naive unvaccinated rhesus macaque FRTs within 4 wk of high-dose vaginal inoculation of SIVmac251 (1–2 × 105 TCID50). NK cells, defined as NKG2A+CD3 cells (Fig. 1A), as previously reported (1923), were detected in fixed and paraffin-embedded necropsy samples using immunohistochemistry (see Table I for Ab information). In general, NK cells primarily were found in vaginal and cervical submucosa (Fig. 1B–D). In contrast to gut mucosal tissues (24), we did not find intraepithelial NK cells in vagina or cervix, and we found very few NK cells at the site where infected founder populations have been consistently noted (2, 3) in the palmate folds of endocervical tissues in any of the 32 cervical tissues examined (Fig. 1B).

FIGURE 1.

Spatial distribution of NK cells in the lower FRT mucosa of rhesus macaques. (A) NK cells in rhesus macaques were defined as NKG2A+CD3 using immunohistochemistry (original magnification ×200). NK cells primarily were located in the submucosal area in endocervix (B), ectocervix (C), and vagina (D). (B) There were remarkably few NK cells in the palmate folds of the endocervix. All images were taken from stained sections of tissues collected 7 d postinfection. NKG2A (green), CD3 (red). In (B–D), NK cells were labeled as green dots for better visibility at the magnification shown (original magnification ×200).

FIGURE 1.

Spatial distribution of NK cells in the lower FRT mucosa of rhesus macaques. (A) NK cells in rhesus macaques were defined as NKG2A+CD3 using immunohistochemistry (original magnification ×200). NK cells primarily were located in the submucosal area in endocervix (B), ectocervix (C), and vagina (D). (B) There were remarkably few NK cells in the palmate folds of the endocervix. All images were taken from stained sections of tissues collected 7 d postinfection. NKG2A (green), CD3 (red). In (B–D), NK cells were labeled as green dots for better visibility at the magnification shown (original magnification ×200).

Close modal

We quantified the number of NK cells in vaginal (48 animals) and cervical (32 animals) tissues in naive animals prior to and following SIV vaginal challenge. As shown in Fig. 2A and 2B, NK cells were rapidly recruited into the cervical and vaginal mucosa after vaginal inoculation, with significantly higher numbers in cervix compared with vagina (cervical median = 1017, vaginal median = 319) at the end of the first week (Fig. 2C). However, even at the peak of expansion, NK cells were greatly outnumbered by expanded populations of the CD4+ T cells that fuel local virus expansion (2, 3), as well as by macrophages (Fig. 2D).

FIGURE 2.

NK cells are recruited into the lower FRT mucosa after vaginal challenge. The density of NK cells in the genital mucosa was quantified in cervical (A) and vaginal (B) tissues. (C) In the same individual animals, there were more mucosal NK cells in the cervix than vagina. (D) However, even at the peak of recruitment, NK cells were significantly outnumbered by recruited CD4 T cells and macrophages. Data in (D) are from tissues collected 7 d postinfection. Each point or line represents an individual animal.

FIGURE 2.

NK cells are recruited into the lower FRT mucosa after vaginal challenge. The density of NK cells in the genital mucosa was quantified in cervical (A) and vaginal (B) tissues. (C) In the same individual animals, there were more mucosal NK cells in the cervix than vagina. (D) However, even at the peak of recruitment, NK cells were significantly outnumbered by recruited CD4 T cells and macrophages. Data in (D) are from tissues collected 7 d postinfection. Each point or line represents an individual animal.

Close modal

Although small foci of SIV RNA+ cells have been detected in the cervix as early as 3–4 d (2, 25) postvaginal inoculation, it seemed unlikely that viral replication per se would be the driving force for the early NK cell influx into the genital mucosa, given the magnitude of local infection. The recruitment of NK cells to cervical tissues of animals vaginally inoculated with infectious wild type SIV (WT-SIV) or AT-2–inactivated SIV (AT-2–SIV) is consistent with this conclusion. As shown in Fig. 3A, the densities of NK cells were comparable between WT-SIV and AT-2–SIV groups through 4 d after vaginal inoculation. Moreover, the decrease in the numbers of mucosal NK cells in both vagina and cervix during the second week (the peak of infection in the FRT), when only 13.6% (cervix) and 24.8% (vagina) (percentage of median) of the peak values of NK cells remained (Fig. 2A, 2B), also argues against viral replication–driven NK cell recruitment.

FIGURE 3.

(A) AT-2–SIV was as potent as WT-SIV in recruiting NK cells into the FRT. Each point represents an individual animal. (B) Macrophages (CD68+) and fibroblasts (vimentin+) were the major CXCL10/IP-10–expressing cell populations in the FRT mucosa (original magnification ×200). (C) These CXCL10+ cells were in close proximity to NKG2A+ NK cells in the submucosa. All images were from tissues collected 7 d postinfection. Original magnification ×200.

FIGURE 3.

(A) AT-2–SIV was as potent as WT-SIV in recruiting NK cells into the FRT. Each point represents an individual animal. (B) Macrophages (CD68+) and fibroblasts (vimentin+) were the major CXCL10/IP-10–expressing cell populations in the FRT mucosa (original magnification ×200). (C) These CXCL10+ cells were in close proximity to NKG2A+ NK cells in the submucosa. All images were from tissues collected 7 d postinfection. Original magnification ×200.

Close modal

NK cells are most likely recruited by chemokine expression in the FRT. Because CXCL10/IP-10 is well known as a potent NK cell chemoattractant, we examined its expression profile in the FRT mucosa of infected animals. Macrophages (CD68+) and fibroblasts (vimentin+) were the major CXCL10-producing cell populations in the genital mucosa (Fig. 3B). These CXCL10+ cells resided close to the basal layer of epithelium and often were found in close proximity to the majority of NKG2A+ NK cells in the submucosa (Fig. 3C). We favor local recruitment of NK cells by these CXCL10+ cells rather than recruitment by CXCL10 in the inoculum (26), which we would expect to elicit a general and immediate recruitment of NK cells to the mucosal border, rather than the observed focal and delayed recruitment 3 d after exposure.

We next investigated the potential role of NK cells recruited in the first week of infection in containing local viral replication by examining the density and spatial relationships of the mucosal NK cells and SIV RNA+ cells. We enumerated SIV RNA+ cells detected by ISH and showed that SIV RNA+ cells were barely detectable in the first week and then increased to peak in the second week (Fig. 4A, 4B). Because the mucosal NK cells peaked in the first week, when the local expansion of infected founder foci of infected cells had just begun to expand, there was an expected negative correlation between the densities of SIV RNA+ cells and NK cells, which was significant in cervix but not vagina (Fig. 4C, 4D). However, in montage images of the transformation zone (TZ), where SIV RNA+ cells are consistently concentrated in early infection (2, 3), there was complete spatial separation of NKG2A+CD3 NK and SIV RNA+ cell populations (Fig. 5). Indeed, in all animals examined, the SIV RNA+ cells were always located in the endocervix close to the TZ where there were few, if any, NK cells (Fig. 1B). Although these images are snapshots of interactions of cells in FRT tissues, the spatial dissociation between NK cells and SIV RNA+ cells (Fig. 5) does not support the hypothesis that recruited NK cells contain infection by contact-dependent mechanisms in the endocervix and TZ where expanding founder populations of infected cells have been consistently documented (2, 3). However, this spatial dissociation does not exclude a possible role for NK cells in eliminating infected cells at sites close to NK cells before the SIV RNA reaches a detectable level.

FIGURE 4.

Increase in the density of SIV RNA+ cells in cervix and vagina and negative correlation with the density of mucosal NK cells at the end of the first week following vaginal inoculation. (A and B) SIV RNA+ cells detected by ISH were enumerated in cervix and vagina. Each point represents an individual animal. (C and D) In cervical and vaginal tissues, the densities of NK cells correlated negatively with the densities of SIV RNA+ cells at the end of the first week.

FIGURE 4.

Increase in the density of SIV RNA+ cells in cervix and vagina and negative correlation with the density of mucosal NK cells at the end of the first week following vaginal inoculation. (A and B) SIV RNA+ cells detected by ISH were enumerated in cervix and vagina. Each point represents an individual animal. (C and D) In cervical and vaginal tissues, the densities of NK cells correlated negatively with the densities of SIV RNA+ cells at the end of the first week.

Close modal
FIGURE 5.

Spatial dissociation between cervical NK cells and SIV RNA+ cells. Montage images of the density and spatial distribution of mucosal NK cells and SIV RNA+ cells in entire cervical tissue sections. (A and B) NK cells and SIV RNA+ cells are too small to be visible in montage images of entire tissue section at the resolution shown. Therefore, the NK cells and SIV RNA+ cells were labeled as green and red dots, respectively. All images were from tissues collected 7 d postinfection. Original magnification ×200.

FIGURE 5.

Spatial dissociation between cervical NK cells and SIV RNA+ cells. Montage images of the density and spatial distribution of mucosal NK cells and SIV RNA+ cells in entire cervical tissue sections. (A and B) NK cells and SIV RNA+ cells are too small to be visible in montage images of entire tissue section at the resolution shown. Therefore, the NK cells and SIV RNA+ cells were labeled as green and red dots, respectively. All images were from tissues collected 7 d postinfection. Original magnification ×200.

Close modal

We examined the chemokine and cytokine expression profiles of cervical NK cells at 7 d postvaginal inoculation for evidence that might indirectly support the hypothesis that NK cells play a protective role in containing local infection following vaginal exposure to SIV. The majority of cervical NK cells produced IFN-γ (74.3–98.1%, median 80.3%, n = 7) (Fig. 6A), as well as the HIV-1 inhibitory β-chemokine MIP-1α (CCL3) (Fig. 6B), but not MIP-1β (Fig. 6E) or RANTES (Fig. 6F) (data not shown). Only approximately one in four cervical NK cells expressed granzyme B (10.3–29.2%, median 25.4%, n = 7) (Fig. 6C), and less than half of the NK cells were positive for the more abundant NK cytotoxic effector (27), granzyme H (14.6–42.6%, median 38.5%, n = 7) (Fig. 6D). Moreover, the NK cells were negative for the degranulation marker, CD107a (Fig. 6G), and also negative for the phenotypic markers of activation and replication Ki67, CD69, CD38, and HLA-DR (Fig. 6H–K).

FIGURE 6.

Cervical NK cells express IFN-γ (A), MIP-1α (CCL3) (B), granzyme B (C), and granzyme H (D). However, the majority are CCL4 (E), CCL5 (F), CD107a (G), Ki67 (H), CD69 (I), CD38 (J), and HLA-DR (K). Original magnification ×200.

FIGURE 6.

Cervical NK cells express IFN-γ (A), MIP-1α (CCL3) (B), granzyme B (C), and granzyme H (D). However, the majority are CCL4 (E), CCL5 (F), CD107a (G), Ki67 (H), CD69 (I), CD38 (J), and HLA-DR (K). Original magnification ×200.

Close modal

In seeking an explanation for the latter phenotype, we had noticed in unpublished microarray analyses, in which we compared transcriptional profiles of cervical tissues from uninfected and vaginally infected animals, that expression of the HLA-E (Mamu-E) inhibitory ligand of NKG2A receptors was upregulated in cervical tissues within 4 d after virus infection (Fig. 7A) (A.J. Smith, S.W. Wietgrefe, L. Shang, C.S. Reilly, P.J. Southern, K.E. Perkey, L. Duan, H. Kohler, S. Muller, J. Robinson, J.V. Carlis, Q. Li, R.P. Johnson, and A.T. Haase, unpublished observations). By immunohistochemical staining, HLA-E+ cells were detectable as soon as 24 h postchallenge (Fig. 7B). These HLA-E+ cells included Langerhans cells in the genital epithelium and cells in the submucosa close to epithelium, where NK cells usually reside (Fig. 7B). Consistent with the inhibitory function of HLA-E, the resident NK cell population was CD122 (IL-2Rβ) negative (Fig. 7C) and, therefore, would be unable to respond to IL-2 and IL-15. Thus, collectively, the cervical NK cells recruited by SIV vaginal inoculation lacked many markers of activation and cytotoxic function that might be expected for an effector population.

FIGURE 7.

(A) Microarray analysis comparison of gene expression in uninfected versus infected cervical tissue reveals upregulation of HLA-E expression in cervical tissues after vaginal challenge. (B) The HLA-E–expressing cells in cervical–vaginal submucosa as early as 24 h postinfection include Langerhans cells in epithelium (right panel) (DAB staining, original magnificaiton ×100). (C) NKG2A+CD3 NK cells in cervical mucosa were CD122 (IL-2Rβ) negative (original magnification ×200).

FIGURE 7.

(A) Microarray analysis comparison of gene expression in uninfected versus infected cervical tissue reveals upregulation of HLA-E expression in cervical tissues after vaginal challenge. (B) The HLA-E–expressing cells in cervical–vaginal submucosa as early as 24 h postinfection include Langerhans cells in epithelium (right panel) (DAB staining, original magnificaiton ×100). (C) NKG2A+CD3 NK cells in cervical mucosa were CD122 (IL-2Rβ) negative (original magnification ×200).

Close modal

SIVmac239Δnef vaccination was shown to protect against high-dose vaginal challenge (15, 16, 28) (e.g., vaccinated animals either had sterilizing immunity or lower peak and set point viral loads compared with unvaccinated animals at 20 wk postvaccination; R.K. Reeves, manuscript in preparation). We evaluated potential NK cell contributions to this protection by asking whether vaccinated animals had denser vaginal and cervical NK cell populations compared with naive animals prior to and after vagina challenge. To the contrary, we found that the cervical and vaginal NK cell populations prior to challenge were not statistically different between naive and vaccinated animals (Fig. 8A, 8B), and, surprisingly, high-dose vaginal challenge of vaccinated animals with pathogenic SIVmac251 did not induce the influx of NK cells or macrophages (Fig. 8C) into the genital mucosa that we had documented (Fig. 1) in unvaccinated animals.

FIGURE 8.

NK cell population densities in the FRT mucosa do not correlate with SIVmac239Δnef vaccine protection before or after high-dose vaginal challenge. NK cells in the cervical (A) and vaginal (B) mucosa were quantified by immunohistochemistry in tissues from naive, SIVmac251 vaginally infected, SIVmac239Δnef-vaccinated, and vaccinated plus vaginally challenged rhesus macaques. There was no correlation between NK cell density prior to and after vaginal challenge with protection, and NK cell densities did not increase following challenge in the vaccinated animals. (C) Macrophage recruitment also was inhibited in the vaccinated animals. *Not significant.

FIGURE 8.

NK cell population densities in the FRT mucosa do not correlate with SIVmac239Δnef vaccine protection before or after high-dose vaginal challenge. NK cells in the cervical (A) and vaginal (B) mucosa were quantified by immunohistochemistry in tissues from naive, SIVmac251 vaginally infected, SIVmac239Δnef-vaccinated, and vaccinated plus vaginally challenged rhesus macaques. There was no correlation between NK cell density prior to and after vaginal challenge with protection, and NK cell densities did not increase following challenge in the vaccinated animals. (C) Macrophage recruitment also was inhibited in the vaccinated animals. *Not significant.

Close modal

In this article, we report characterization of the NK cell response to SIV infection in the female genital mucosa of rhesus macaques after a high-dose vaginal exposure. In this animal model, the first 3–4 d postinfection define a time frame of vulnerability for the virus in establishing and expanding infected founder populations and, thus, a window of opportunity for the host immune system to inhibit viral replication and prevent mucosal transmission. We found that NK cells were recruited into the genital mucosa in this window of opportunity by mechanisms that can be attributed, at least in part, to cells in the FRT expressing CXCL10/IP-10, the most potent NK cell chemoattractant (29), in close proximity to NK cells. The peak values of recruited cervical NK cells at 7 d postvaginal exposure correlated negatively with the number of SIV RNA+ cells in cervix. Moreover, most of the NK cells in the FRT mucosa were positive for IFN-γ and MIP-1α/CCL3, a β-chemokine that was shown to block viral entry in culture (4, 5). Collectively, these results initially provided evidence to support the hypothesis that the NK cells recruited into the genital mucosa could contribute to containing SIV infection in the early window of opportunity at the portal of entry.

However, we document aspects of the quantity, location, and quality of observed mucosal NK cell responses that are inconsistent with NK cells as effective gatekeepers to prevent viral transmission. First, only a very small number of NK cells was recruited into the genital mucosa relative to the orders of magnitude greater recruitment of CD4 T cells that fuel the local expansion of viruses and infected cells prior to systemic dissemination and infection (2, 25). The numbers of NK cells also declined at the peak of virus replication in the lower FRT tissues, similar to the decreased numbers of NK cells in the gut mucosa of SIV-infected animals 2–4 wk following SIV vaginal transmission (23, 30). Thus, the effective in vivo ratio of NK effector cells/infected targets was even more unfavorable for a protective role.

Second, the spatial dissociation between recruited NK cells and expanding populations of SIV RNA+ cells in the endocervix does not support a model in which mucosal NK cells kill their targets by cell–cell contact mechanisms, such as Ab-dependent cellular cytotoxicity and death receptor–mediated cytotoxicity (31). Third, the recruited NK cells lacked many phenotypic markers associated with a functioning effector population, such as evidence of degranulation (CD107a), activation (CD38, CD69, and HLA-DR), and other cytokines/chemokines previously associated with an activated effector population of NK cells, although CD69 and Ki67 expression on NK cells in blood and gut has been reported (23, 32). In the FRT, we show increased expression of the inhibitory HLA-E ligand and downregulation of activating receptor CD122 following vaginal exposure to SIV, highlighting the rapidly altered balance between activating and inhibitory signals during mucosal transmission, as was shown in other acute and chronic infections (3335).

SIVmac239Δnef vaccination provides robust protection against subsequent WT-SIV challenge by parenteral and mucosal routes (16, 28). Although safety issues have precluded advancing this and other live attenuated viruses as candidates for an effective HIV-1 vaccine, understanding the correlates of protection could provide critical insights to guide HIV-1 vaccine development. We have been focusing particularly on identifying correlates associated with the Δnef vaccine–mediated maturation of protection, defined by little protection if animals are challenged vaginally at 5 wk postvaccination but substantial protection at 15–20 wk or even 40 wk postvaccination. In this study, we assessed the role of NK cells in the FRT mucosa in mediating the vaccine-induced protection in animals that have been vaccinated for 20 wk (versus naive animals). Surprisingly, NK cell numbers in the FRT mucosa of animals vaccinated 20 wk prior to challenge were comparable to those in naive animals. Even more remarkably, we did not observe any influx of NK cells or macrophages into the FRT mucosa in the vaccinated animals following high-dose vaginal challenge. We show elsewhere that the formation of immune complexes following vaginal challenge, as well as the interaction of these immune complexes with FcR for IgG, FcγRIIb, induces an inhibitory program that blocks CD4 T cell recruitment (A.J. Smith, S.W. Wietgrefe, L. Shang, C.S. Reilly, P.J. Southern, K.E. Perkey, L. Duan, H. Kohler, S. Muller, J. Robinson, J.V. Carlis, Q. Li, R.P. Johnson, and A.T. Haase, manuscript in preparation). This inhibitory program may similarly be involved in blocking NK cell, as well as macrophage, recruitment. Thus, our studies collectively provide only limited evidence of correlations between NK cells in the FRT mucosa of naive animals and containment of infection at the portal of entry and no evidence for a role for NK cells in protection against vaginal challenge in SIVmac239Δnef-vaccinated animals.

We thank Dr. Ronald Desrosiers for providing SIVmac239∆nef; Jacqueline Gillis, Fay Eng Wong, and Elizabeth Curran for assistance with tissue processing; and Dr. Angela Carville and other members of the New England Primate Research Center Division of Veterinary Resources for expert animal care.

This work was supported by the International AIDS Vaccine Initiative and National Institutes of Health Grants AI071306, AI090735, AI095985, and RR00168 (currently OD011103).

Abbreviations used in this article:

AT-2–SIV

AT-2–inactivated SIV

FRT

female reproductive tract

ISH

in situ hybridization

TZ

transformation zone

WT-SIV

wild type SIV.

1
United Nations Joint Programme on HIV/AIDS (UNAIDS). 2012. Global Report: UNAIDS Report on the Global AIDS Epidemic. http://www.unaids.org/en/resources/publications/2012/name,76121,en.asp. Accessed: May 23, 2014.
2
Miller
C. J.
,
Li
Q.
,
Abel
K.
,
Kim
E. Y.
,
Ma
Z. M.
,
Wietgrefe
S.
,
La Franco-Scheuch
L.
,
Compton
L.
,
Duan
L.
,
Shore
M. D.
, et al
.
2005
.
Propagation and dissemination of infection after vaginal transmission of simian immunodeficiency virus.
[Published erratum appears in 2005 J. Virol. 79: 11552.]
J. Virol.
79
:
9217
9227
.
3
Li
Q.
,
Estes
J. D.
,
Schlievert
P. M.
,
Duan
L.
,
Brosnahan
A. J.
,
Southern
P. J.
,
Reilly
C. S.
,
Peterson
M. L.
,
Schultz-Darken
N.
,
Brunner
K. G.
, et al
.
2009
.
Glycerol monolaurate prevents mucosal SIV transmission.
Nature
458
:
1034
1038
.
4
Fauci
A. S.
,
Mavilio
D.
,
Kottilil
S.
.
2005
.
NK cells in HIV infection: paradigm for protection or targets for ambush.
Nat. Rev. Immunol.
5
:
835
843
.
5
Oliva
A.
,
Kinter
A. L.
,
Vaccarezza
M.
,
Rubbert
A.
,
Catanzaro
A.
,
Moir
S.
,
Monaco
J.
,
Ehler
L.
,
Mizell
S.
,
Jackson
R.
, et al
.
1998
.
Natural killer cells from human immunodeficiency virus (HIV)-infected individuals are an important source of CC-chemokines and suppress HIV-1 entry and replication in vitro.
J. Clin. Invest.
102
:
223
231
.
6
Forthal
D. N.
,
Landucci
G.
,
Haubrich
R.
,
Keenan
B.
,
Kuppermann
B. D.
,
Tilles
J. G.
,
Kaplan
J.
.
1999
.
Antibody-dependent cellular cytotoxicity independently predicts survival in severely immunocompromised human immunodeficiency virus-infected patients.
J. Infect. Dis.
180
:
1338
1341
.
7
Ahmad
R.
,
Sindhu
S. T.
,
Toma
E.
,
Morisset
R.
,
Vincelette
J.
,
Menezes
J.
,
Ahmad
A.
.
2001
.
Evidence for a correlation between antibody-dependent cellular cytotoxicity-mediating anti-HIV-1 antibodies and prognostic predictors of HIV infection.
J. Clin. Immunol.
21
:
227
233
.
8
Banks
N. D.
,
Kinsey
N.
,
Clements
J.
,
Hildreth
J. E.
.
2002
.
Sustained antibody-dependent cell-mediated cytotoxicity (ADCC) in SIV-infected macaques correlates with delayed progression to AIDS.
AIDS Res. Hum. Retroviruses
18
:
1197
1205
.
9
Tomescu
C.
,
Abdulhaqq
S.
,
Montaner
L. J.
.
2011
.
Evidence for the innate immune response as a correlate of protection in human immunodeficiency virus (HIV)-1 highly exposed seronegative subjects (HESN).
Clin. Exp. Immunol.
164
:
158
169
.
10
Alter
G.
,
Altfeld
M.
.
2009
.
NK cells in HIV-1 infection: evidence for their role in the control of HIV-1 infection.
J. Intern. Med.
265
:
29
42
.
11
Carrington
M.
,
Martin
M. P.
,
van Bergen
J.
.
2008
.
KIR-HLA intercourse in HIV disease.
Trends Microbiol.
16
:
620
627
.
12
Bashirova
A. A.
,
Thomas
R.
,
Carrington
M.
.
2011
.
HLA/KIR restraint of HIV: surviving the fittest.
Annu. Rev. Immunol.
29
:
295
317
.
13
McMichael
A. J.
,
Jones
E. Y.
.
2010
.
Genetics. First-class control of HIV-1.
Science
330
:
1488
1490
.
14
Alter
G.
,
Heckerman
D.
,
Schneidewind
A.
,
Fadda
L.
,
Kadie
C. M.
,
Carlson
J. M.
,
Oniangue-Ndza
C.
,
Martin
M.
,
Li
B.
,
Khakoo
S. I.
, et al
.
2011
.
HIV-1 adaptation to NK-cell-mediated immune pressure.
Nature
476
:
96
100
.
15
Wyand
M. S.
,
Manson
K.
,
Montefiori
D. C.
,
Lifson
J. D.
,
Johnson
R. P.
,
Desrosiers
R. C.
.
1999
.
Protection by live, attenuated simian immunodeficiency virus against heterologous challenge.
J. Virol.
73
:
8356
8363
.
16
Johnson
R. P.
,
Lifson
J. D.
,
Czajak
S. C.
,
Cole
K. S.
,
Manson
K. H.
,
Glickman
R.
,
Yang
J.
,
Montefiori
D. C.
,
Montelaro
R.
,
Wyand
M. S.
,
Desrosiers
R. C.
.
1999
.
Highly attenuated vaccine strains of simian immunodeficiency virus protect against vaginal challenge: inverse relationship of degree of protection with level of attenuation.
J. Virol.
73
:
4952
4961
.
17
Smith
A. J.
,
Li
Q.
,
Wietgrefe
S. W.
,
Schacker
T. W.
,
Reilly
C. S.
,
Haase
A. T.
.
2010
.
Host genes associated with HIV-1 replication in lymphatic tissue.
J. Immunol.
185
:
5417
5424
.
18
Smith
A. J.
,
Schacker
T. W.
,
Reilly
C. S.
,
Haase
A. T.
.
2010
.
A role for syndecan-1 and claudin-2 in microbial translocation during HIV-1 infection.
J. Acquir. Immune Defic. Syndr.
55
:
306
315
.
19
Carter
D. L.
,
Shieh
T. M.
,
Blosser
R. L.
,
Chadwick
K. R.
,
Margolick
J. B.
,
Hildreth
J. E.
,
Clements
J. E.
,
Zink
M. C.
.
1999
.
CD56 identifies monocytes and not natural killer cells in rhesus macaques.
Cytometry
37
:
41
50
.
20
Webster
R. L.
,
Johnson
R. P.
.
2005
.
Delineation of multiple subpopulations of natural killer cells in rhesus macaques.
Immunology
115
:
206
214
.
21
Mavilio
D.
,
Benjamin
J.
,
Kim
D.
,
Lombardo
G.
,
Daucher
M.
,
Kinter
A.
,
Nies-Kraske
E.
,
Marcenaro
E.
,
Moretta
A.
,
Fauci
A. S.
.
2005
.
Identification of NKG2A and NKp80 as specific natural killer cell markers in rhesus and pigtailed monkeys.
Blood
106
:
1718
1725
.
22
Bostik
P.
,
Takahashi
Y.
,
Mayne
A. E.
,
Ansari
A. A.
.
2010
.
Innate immune natural killer cells and their role in HIV and SIV infection.
HIV Ther.
4
:
483
504
.
23
Reeves
R. K.
,
Gillis
J.
,
Wong
F. E.
,
Yu
Y.
,
Connole
M.
,
Johnson
R. P.
.
2010
.
CD16- natural killer cells: enrichment in mucosal and secondary lymphoid tissues and altered function during chronic SIV infection.
Blood
115
:
4439
4446
.
24
Sips
M.
,
Sciaranghella
G.
,
Diefenbach
T.
,
Dugast
A. S.
,
Berger
C. T.
,
Liu
Q.
,
Kwon
D.
,
Ghebremichael
M.
,
Estes
J. D.
,
Carrington
M.
, et al
.
2012
.
Altered distribution of mucosal NK cells during HIV infection.
Mucosal Immunol.
5
:
30
40
.
25
Zhang
Z.
,
Schuler
T.
,
Zupancic
M.
,
Wietgrefe
S.
,
Staskus
K. A.
,
Reimann
K. A.
,
Reinhart
T. A.
,
Rogan
M.
,
Cavert
W.
,
Miller
C. J.
, et al
.
1999
.
Sexual transmission and propagation of SIV and HIV in resting and activated CD4+ T cells.
Science
286
:
1353
1357
.
26
Del Prete
G. Q.
,
Scarlotta
M.
,
Newman
L.
,
Reid
C.
,
Parodi
L. M.
,
Roser
J. D.
,
Oswald
K.
,
Marx
P. A.
,
Miller
C. J.
,
Desrosiers
R. C.
, et al
.
2013
.
Comparative characterization of transfection- and infection-derived simian immunodeficiency virus challenge stocks for in vivo nonhuman primate studies.
J. Virol.
87
:
4584
4595
.
27
Sedelies
K. A.
,
Sayers
T. J.
,
Edwards
K. M.
,
Chen
W.
,
Pellicci
D. G.
,
Godfrey
D. I.
,
Trapani
J. A.
.
2004
.
Discordant regulation of granzyme H and granzyme B expression in human lymphocytes.
J. Biol. Chem.
279
:
26581
26587
.
28
Johnson
R. P.
1999
.
Live attenuated AIDS vaccines: hazards and hopes.
Nat. Med.
5
:
154
155
.
29
Robertson
M. J.
2002
.
Role of chemokines in the biology of natural killer cells.
J. Leukoc. Biol.
71
:
173
183
.
30
Reeves
R. K.
,
Rajakumar
P. A.
,
Evans
T. I.
,
Connole
M.
,
Gillis
J.
,
Wong
F. E.
,
Kuzmichev
Y. V.
,
Carville
A.
,
Johnson
R. P.
.
2011
.
Gut inflammation and indoleamine deoxygenase inhibit IL-17 production and promote cytotoxic potential in NKp44+ mucosal NK cells during SIV infection.
Blood
118
:
3321
3330
.
31
Russell
J. H.
,
Ley
T. J.
.
2002
.
Lymphocyte-mediated cytotoxicity.
Annu. Rev. Immunol.
20
:
323
370
.
32
Reeves
R. K.
,
Evans
T. I.
,
Gillis
J.
,
Johnson
R. P.
.
2010
.
Simian immunodeficiency virus infection induces expansion of alpha4beta7+ and cytotoxic CD56+ NK cells.
J. Virol.
84
:
8959
8963
.
33
Chang
J. J.
,
Altfeld
M.
.
2010
.
Innate immune activation in primary HIV-1 infection.
J. Infect. Dis.
202
(
Suppl. 2
):
S297
S301
.
34
Sandler
N. G.
,
Douek
D. C.
.
2012
.
Microbial translocation in HIV infection: causes, consequences and treatment opportunities.
Nat. Rev. Microbiol.
10
:
655
666
.
35
Klatt
N. R.
,
Funderburg
N. T.
,
Brenchley
J. M.
.
2013
.
Microbial translocation, immune activation, and HIV disease.
Trends Microbiol.
21
:
6
13
.

The authors have no financial conflicts of interest.