Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Eur J Immunol. Author manuscript; available in PMC 2012 March 21.
Published in final edited form as:
PMCID: PMC3154578

Robust genital gag-specific CD8+ T-cell responses in mice upon intramuscular immunization with simian adenoviral vectors expressing HIV-1-gag


Most studies on E1-deleted adenovirus (Ad) vectors as vaccine carriers for antigens of HIV-1 have focused on induction of central immune responses although stimulation of mucosal immunity at the genital tract (GT), the primary port of entry of HIV-1, would also be highly desirable. In the present study, different immunization protocols using chimpanzee-derived adenoviral (AdC) vectors expressing gag of HIV-1 clade B given in heterologous prime-boost regimens were tested for induction of systemic and genital immune responses. Although i.n. immunization stimulated CD8+ T-cell responses that could be detected in the GT, this route only induced marginal cellular responses in systemic tissues and furthermore numbers of gag-specific CD8+ T-cells contracted sharply within a few weeks. In contrast, i.m. immunization induced higher and more sustained frequencies of vaccine-induced cells which could be detected in the GT as well as systemic compartments. Antigen-specific CD8+ T-cells could be detected 1 year after immunization in all compartments analyzed. Genital memory cells secreted IFN-γ, expressed high levels of CD103 and their phenotypes were consistent with a state of activation. Taken together, results presented here show that i.m. vaccination with AdC vectors is a suitable strategy to induce a long-lived genital CD8+ T-cell response.

Keywords: T Cell, Mucosal, HIV, Adenovirus


The efficacy of most vaccines is linked to their ability to induce neutralizing antibodies (nAb). This approach has thus far been elusive for vaccines to HIV-1 as the envelope proteins of this virus are heavily glycosylated [1], variable between isolates [2], and undergo structural changes during binding to its receptors and co-receptors [3]. Many HIV-1 vaccines currently in clinical or pre-clinical testing thus attempt to induce HIV-1-specific T-cells to more conserved viral antigens [4].

HIV-1 is most commonly transmitted through sexual contact [5]. Women are infected at a higher rate than men and can pass the virus to their offspring. One would assume that vaccine-induced CD8+ T-cells at the port of entry, i.e., the GT would be crucial for early clearance of infected cells before HIV-1 spreads to lymph nodes and then to the intestinal tract, which provides a rich source of HIV-1 target cells [6]. Although our knowledge on T-cells within the GALT is rapidly expanding, pertinent characteristics of T-cells that home to the female GT remain understudied.

The HIV-1 vaccine efforts of our group have focused on AdC vectors that induce potent transgene product-specific B and CD8+ T-cell responses in mice [7-9] and non-human primates (NHP) [10]. NAb to AdC are rare in humans and do not cross-react with human serotypes of Ad. AdC-induced responses are sustained as the vector persists at low levels [11] and can be increased by heterologous prime-boost regimens [10, 12]. As we reported recently, i.n. administration of AdC elicits high frequencies of CD8+ T-cells that home to the GT of female mice [13].

Here, we extended these studies and our results show that CD8+ T-cells that home to the GT can be induced at high frequencies by both mucosal and i.m. immunizations. I.m. immunization elicits stronger systemic and mucosal responses than i.n. or intravaginal (i.vag.) immunization. Genital CD8+ T-cells express phenotypic markers indicative for activation, and most importantly, are fully functional; they proliferate and secrete cytokines upon reencounter of their cognate antigen.


Frequencies of vaccine-induced CD8+ T-cells

Responses were analyzed upon a single immunization of female BALB/c mice with AdC6gag given i.m., i.n. or i.vag. Frequencies of gag-specific CD8+ T-cells were determined by tetramer staining using a gag peptide-(AMQMLKETI) and H-2Kd-specific tetramer of cells isolated from spleens, blood, iliac lymph nodes (ILN), GT and nasal-associated lymphoid tissue (NALT) at different times after administration. For most experiments, samples from the GT included cells from the vagina, cervix, uterus, uterine horns and ovaries. For some of the phenotypic analyses, cells from the vagina were isolated separately from the remaining GT, referred to as OUC (ovaries, uterus, uterine horns, cervix). Cells isolated from the same compartments of naïve mice were used as controls.

As reported previously [10] and shown in Fig. 1A, i.m. immunization induced a robust and sustained gag-specific CD8+ T-cell response in systemic compartments. Surprisingly, i.m. immunization induced high frequencies of gag-specific CD8+ T-cells within the GT that by week 2 after vaccination were close to 40% and by 10 weeks were still above 10% of all CD8+ T-cells. I.n. vaccination induced readily measurable responses within the GT and NALT, but only marginal responses in blood or spleens. I.vag. immunization was ineffective and only induced a low and transient response in all tissues analyzed. Importantly, i.n. or i.vag. immunization induced lower frequencies of gag-specific CD8+ T-cells within the GT than i.m. immunization. Differences in frequencies achieved by i.m. in comparison to i.n. or i.vag. immunization were statistically significant (p<0.05) in spleens, blood, ILN and GT at all post-vaccination time points tested.

Figure 1Figure 1
Frequencies of gag-specific CD8+ T-cells after administration of AdC vectors expressing HIV-1 gag. (A) Mice were immunized i.n. (open bars), i.m. (black bars) or i.vag (striped bars) with AdC6gag. Cells from different tissues were analyzed at different ...

In the next set of experiments, prime-boost regimens were tested to establish whether systemic and mucosal CD8+ T-cell responses could be enhanced by a second immunization with a heterologous AdC vector expressing the same transgene product. For these experiments, mice were primed either i.n., i.m. or i.vag. with AdC6gag. Six weeks later, they were boosted i.n., i.vag. or i.m. (i.m. for the i.m.-primed group only) with AdC68gag. Frequencies of gag-specific CD8+ T-cells were analyzed 2 weeks before and 2 and 4 weeks after the boost (Figure 1B). GT and NALT were assessed after immunization with regimens inducing the highest responses against HIV-gag in systemic compartments. I.m primed/i.m. boosted mice were also analyzed for frequencies of tet+CD8+ T-cells at one year after booster immunization to determine the longevity of the response.

Vaginal booster immunization failed to increase frequencies of gag-specific CD8+ T-cells in systemic compartments of i.m. primed mice. However, i.vag. boost of i.n.-primed mice elicited an increase of frequencies in spleen and blood, although less pronounced than the i.m./i.m. regimen (p<0.05). Frequencies were higher in spleen, blood, ILN and GT for the group receiving two doses through systemic routes in comparison to groups receiving at least one mucosal administration (p<0.05). Within the GT, frequencies of gag-specific CD8+ T-cells increased after i.n./i.vag. or i.m./i.m. regimens, being more pronounced in the group receiving the vectors systemically (p<0.01). At 2 and 4 weeks after the i.m/.im. prime-boost immunization, frequencies at the GT exceeded those from blood (p<0.01). At one year after the i.m./i.m. regimen, gag-specific CD8+ T-cells could still be detected in the GT, although frequencies were not statistically different from those in blood (p<0.05) and had decreased compared to those detected at 4 weeks after boost (p<0.05). At that time, frequencies in spleens and ILN remained stable and those in blood decreased, presumably reflecting a loss of the more activated effector/effector memory cells (p<0.05).

Cytokine secretion by gag-specific CD8+ T-cells

To gain insight into functional properties of gag-specific T-cells, we conducted ELISpot assays for IFN-γ and IL-2. Figure 2A shows IFN-γ secretion by splenocytes isolated from mice that received AdC6gag i.m. Concomitantly with the ELISpot assays, cells were tested by flow cytometry to determine frequencies of CD8+ T-cells and results were normalized to reflect spots per 106 CD8+ T-cells. In the ELISpot assay, cells were stimulated with either the AMQMLKETI peptide, which carries an immunodominat MHC class I epitope of gag for H-2kd mice or with a pool of peptides representing the entire Gag sequence. We observed no difference in the response and therefore conclude that, as observed previously, our AdC-based vaccine induces mainly CD8+ T-cells rather than CD4+ T-cells [14] to the transgene-product, and furthermore that Gag-specific CD8+ T-cells in BALB/c mice are directed mainly to the AMQMLKETI peptide. Consequently, the use of this one peptide for stimulation of specific cells would be expected to detect the majority of Gag-specific CD8+-T cells in this mouse strain.

Figure 2
IFN-γ production by vaccine-induced gag-specific CD8+ T-cells. IFN-γ ELISpot assay was performed with splenocytes from (A) mice immunized i.m. with AdC6gag 2 weeks earlier, and stimulated with the total pool of consensus gag clade B in ...

Independent of the route and number of immunizations, T-cells isolated from different tissues preferentially produced IFN-γ; significant numbers of IL-2 producing cells could not be detected (>55 spot-forming units [SFU]/106 lymphocytes). Examples for the results are shown in Figure 2B, which presents data from mice immunized 2 weeks earlier i.n. or i.m. with AdC6gag. Similar results were obtained at later time points or after prime-boost regimens (data not shown). Numbers of IFN-γ-secreting cells were higher in spleen, blood, ILN and the GT upon i.m. immunization (p<0.05). Although samples from the GT showed secretion of IFN-γ in response to antigen, we had expected higher frequencies from this compartment based on frequencies obtained by tetramer staining (higher in GT than in blood or spleen (p<0.05) for both i.n. and i.m administration). However, ELISpot assays showed significantly higher secretion in blood than in GT for the i.n. group (p<0.05) and comparable numbers for the i.m.-primed mice. It is feasible that cells from the GT or NALT secrete cytokines other than IFN-γ or IL-2 and therefore escaped detection by the ELISpot assays. Although this was not ruled out, we favor the explanation that vaccine-induced T cells from the GT and NALT are comparatively frail and thus more readily detected by staining procedures that do not require lengthy incubations. In order to further address this issue, mice were immunized with AdC6gag i.m. and tetramer frequencies were evaluated from cells isolated from the GT either directly without further culture, or after an overnight culture at 37°C with or without the specific peptide. Cells were stained with an antibody to CD8a, the specific tetramer, a live cell dye and analyzed by flow cytometry. We observed pronounced cell death after overnight incubation of cells especially upon stimulation with the specific peptide; accordingly numbers of tet+CD8+ T-cells declined ~ 25 or 150 fold upon overnight in vitro culture in medium or the Gag peptide respectively (data not shown).

Phenotypes of vaccine-induced CD8+ T cells

To elucidate potential differences between T-cells isolated from distinct compartments, expression levels of CD44, CD27 (two lymphocyte activation markers), CD62L, a lymph node homing marker differentially expressed by effector and central memory cells, and α4β7, an integrin that favors migration to the gut mucosa, were determined on tet+CD8+ T-cells induced by AdC6gag. Figure 3A shows data for naïve CD8+ lymphocytes compared to tet+CD8+ T-cells 4 and 10 weeks after a single i.n. immunization and 4 weeks after a vaginal boost of i.n.-primed mice. Figure 3B shows data for CD8+ T-cells tested 4 and 10 weeks after i.m. priming, at 4 weeks after a booster immunization of i.m.-primed mice given i.vag or i.m., and at 1 year after an i.m/i.m. prime-boost regimen. In all experiments, tet-CD8+ T-cells from immune mice were also analyzed and their phenotypes mirrored those of naïve mice (data not shown).

Figure 3Figure 3
Phenotype analysis of gag-specific CD8+ T-cells upon mucosal or systemic administration of AdC vectors. Lymphocytes from different tissues were harvested, stained with antibodies to CD8, CD44, CD62L, CD27 and α4β7, and analyzed by flow ...

Four weeks after i.n. immunization with AdC6gag, CD44 was up-regulated on gag-specific CD8+ T-cells from spleens, blood, ILN and NALT (Figure 3A). This increase was less pronounced on tet+CD8+ cells from the GT, presumably reflecting that most T-cells from the GT were already antigen-experienced. Most of the tet+CD8+ T-cells from the GT expressed comparable levels of CD62L, although a small population was CD62Lhi. It should be pointed out though that expression of CD62L was also low on most of the genital CD8+ T-cells from naïve mice. Expression of α4β7 was low on most cells except for a small population of tet+CD8+ T-cells present in spleen and blood. The booster immunization did not have a pronounced effect on expression of CD44, CD62L or CD27. α4β7 expression was again increased on some of the tet+CD8+ T cells from spleens and ILN.

At 4 weeks after i.m. immunization, CD44 expression was up-regulated on tet+CD8+ T-cells from spleens, ILN and GT (Figure 3B). We detected a down-regulation of CD62L expression on tet+CD8+ T-cells from spleens, blood and the GT but not on those from ILN. CD27 expression was decreased on a subpopulation of tet+CD8+ T-cells from blood, spleens and GT. At 4 weeks after i.n. or i.vag boost, expression levels of CD44, CD62L, CD27 and α4β7 mirrored those seen at 10 weeks after priming, and there were no striking differences among groups that received an AdC6gag i.m. prime followed by a heterologous boost through the i.m. or i.vag. routes. At 1 year after the i.m. prime-boost vaccine regimen, expression of CD44 on tet+CD8+ T-cells isolated from the different compartments (NALT was not tested in this experiment) overlapped with those seen on part of CD8+ T-cells of age-matched naïve mice. This may reflect an increase of CD44 expression on the control CD8+ T-cells due to immunosenescence [15]. Gag-specific CD8+ T-cells isolated from the ILN and GT showed an increase in CD62L expression, which was unexpected for the latter compartment. In blood and spleen, expression of CD62L and CD27 were similar or only slightly increased above those seen on unprimed CD8+ T-cells, suggesting that the gag-specific CD8+ T-cells had differentiated into resting memory cells.

Additional markers were analyzed on gag-specific CD8+ T-cells isolated from different compartments after an i.m./i.m. heterologous prime-boost regimen (Figure 4). For the two early time points, i.e., 4 weeks after priming or boosting, cells isolated from the vaginal mucosa were treated and analyzed separately from OUC. CD44, CD62L and CD27 were tested and found to mirror those shown in Figure 3. Figure 4A shows expression levels of CD69, an early activation marker enriched on mucosal cells derived from the intestinal tract, CD127, the IL-7 receptor that allows for homeostatic proliferation of memory CD8+ T-cells through IL-7, CD103, also known as integrin αEβ7, which has been linked to infiltration of cells into epithelial tissues [16, 17] and natural-killer group 2D (NKG2D), a transmembrane protein that acts as a primary activation receptor on NK cells and serves as a co-stimulator for CD8+ T-cells [18].

Figure 4Figure 4
Phenotype of gag-specific CD8+-T cells induced by i.m. administration of AdC vectors. Lymphocytes from different tissues were harvested and analyzed 4 weeks after i.m. prime with AdC6gag vector, and 4 weeks and 1 year after boost with the heterologous ...

By 4 weeks after i.m. prime or boost, CD69 was decreased on tet+CD8+ T-cells from spleens, blood and OUC, while its expression on the vagina was similar to that on unprimed CD8+ T-cells. By 1 year after the boost, CD69 expression on tet+CD8+ T-cells from all compartments was similar to that of naïve cells, suggesting that this molecule is unlikely to contribute for the sustained presence of vaccine-induced CD8+ T-cells within the GT (data not shown). Expression of CD127 was increased on tet+CD8+ T-cells from ILN and the vagina at 4 weeks after priming. A similar pattern was observed at 4 weeks after the boost but for a modest increase in OUC. By one year after the boost, CD127 expression was increased in tet+CD8+ T-cells from all compartments, being especially pronounced in cells from GT.

The most striking difference in the expression of CD103 was seen at 1 year after the boost, when this marker was markedly upregulated on tet+CD8+ T-cells from the GT, but otherwise comparable to naïve cells in the other compartments. No remarkable changes were seen in the profile of NKG2D on the compartments analyzed.

Figure 4B shows expression levels of granzyme B, a proteolytic enzyme that induces caspase-dependent apoptosis, and perforin, a pore-forming protein that facilitates granzyme access through the membrane into the cytosol of the target cell [19]. In addition, Figure 4B shows expression levels for cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4), a key molecule for down-regulation of T-cell responses, programmed death-1 (PD-1), which negatively regulates T-cell signaling and effector functions and is expressed at increased levels on so-called exhausted T-cells [20] and Ki-67, a protein associated with proliferation.

Expression of granzyme B mostly mirrored that of perforin, with a very pronounced increase in both enzymes in most tet+CD8+ T-cells isolated from the whole GT at 1 year after the boost. This was surprising as expression levels of other markers such as CD62L at the same time point suggest that T-cells isolated from the GT had differentiated into resting memory cells. Memory CD8+ T-cells typically do not carry granzyme or perforin, which are markers for fully activated effector CD8+ T-cells.

CTLA-4 expression was decreased in tet+CD8+ T-cells from spleens, ILN and vagina at 4 weeks after the prime, while there was an increase in its expression on those from OUC. By 1 year after the boost, most tet+CD8+ T-cells from the whole GT showed high levels of CTLA-4 expression, which was otherwise expressed at levels similar to those of unprimed CD8+-T cells in other compartments. Further analyses showed that in the GT, cells that were high in CTLA-4 concomitantly expressed high levels of lytic enzymes (not shown). By 1 year after the boost, Ki-67 levels were up-regulated on the GT. Expression of PD-1 was largely unremarkable.

In summary, the most striking differences in phenotypes between tet+CD8+ T-cells from blood and spleen in comparison to those from the GT and its draining lymph nodes were seen at 1 year after the i.m./i.m. prime-boost regimen. Subpopulations of tet+CD8+ T-cells from the GT showed marked increases in expression of CD103, CD127, CD62L, granzyme B, perforin, CTLA-4 and Ki-67 and thus clearly represented a stage of differentiation not seen in spleens or blood.

Origin of genital gag-specific CD8+ T-cells

To gain insight into the origin of CD8+ T-cells that homed to the GT, we conducted adoptive transfer experiments. BALB/c donor mice were primed with AdC6gag and boosted with AdC68gag given i.m. Fourteen days post-boost, splenocytes were isolated from the vaccinated mice and frequencies (Figure 5) of tet+CD8+ cells were determined. The remaining cells were injected i.v. at 5×107 cells/mouse into naïve Thy1.1 congenic recipient mice. The recipient mice were euthanized 7 days later. As AdC vectors persist at very low levels in activated CD8+ T-cells [11], we cannot rule out transfer of the vectors in splenocytes of donor origin. However, it is unlikely that the minute amount of vector present in T-cells of the donor would induce a detectable immune response in the host within the time frame of the experiment. Nevertheless, to ensure that results were not biased by activation of host-derived T-cells, we used a congenic mouse strain for the experiment, which allowed us to track cells of donor origin. As shown in Figure 5, gag-specific Thy1.1- CD8+ cells of donor origin could readily be detected in all compartments tested, including the GT. As seen after i.m. prime with AdC6gag (Figure 1), frequencies of tet+CD8+ T-cells were higher in the GT than in other compartments analyzed (p<0.01). The results clearly show that gag-specific CD8+ T-cells from spleens can migrate to and are enriched for in the GT. We tested tet+CD8+ cells from donor mice prior to transfer for expression of cell markers shown in Figures 3 and and4.4. CD69 and CD103, two molecules that have been implicated on the phenotype of mucosa-derived cells [21, 22], were expressed at the same levels on tet+CD8+ cells from donor mice prior to transfer and in control cells, and were thus unlikely to have contributed to the enrichment of gag-specific CD8+ T-cells within the GT. We also tested for expression levels of these markers in tet+CD8+ cells of donor origin that had homed to the GT of recipient mice. Levels of CD69 again were similar to those on tet-CD8+ T-cells, while CD103 was increased. This increase was only seen on tet+CD8+ cells from the GT but not on those from other compartments, suggesting that this molecule becomes up-regulated once CD8+ T-cells home to the GT (data not shown).

Figure 5
Migratory properties of adoptively transferred CD8+ T-cells. BALB/c mice were vaccinated with AdC6gag and boosted with AdC68gag. Fourteen days after the boost, splenocytes were isolated and transferred into naïve Thy1.1 recipient mice. Mice were ...


Women are most commonly infected by HIV-1 through heterosexual contact and immune mechanisms at or within the female GT would be expected to provide a crucial first barrier to transmission. As nAb that neutralize the countless HIV-1 variants remain elusive, many of the vaccines currently in clinical trials focus on the induction of HIV-1-specific CD8+ T-cells. Such response cannot prevent the initial infection, but if present at the port of entry, might rapidly eliminate infected cells and thus thwart or potentially prevent spread of the virus.

We showed in mice that a homologous prime-boost regimen using AdC vectors expressing gag induces transgene product-specific CD8+ T-cells that could be isolated from the GT [13]. This previous paper used intracellular cytokine staining assays, which may not be optimal for the study of the GT-derived lymphocytes. Here we extended these studies testing different routes of immunization, more efficacious heterologous prime-boost regimens, and assessed migratory patterns of such cells.

It is known that nasal immunization is able to induce immune responses not only in the respiratory tract but also at the GT [23]. Results reported here show that CD8+ T-cells, which home to the female GT, can be induced by i.n. immunization but this response is not sustained. In addition, vaginal booster immunization, as would be experienced in human vaccine recipients against HIV-1, causes only a slight local increase in i.n.-induced antigen-specific CD8+ T- cells and fails to increase responses systemically. Last but not least, i.n. immunization may be problematic for some vectors as this route allows access of the vaccine into the central nervous system. I.vag immunization, as reported by others [24], only induces very low levels of antigen-specific CD8+ T-cells, which combined with logistic problems in humans should discourage further pursuit of this route of immunization for Ad vectors.

Results are more promising after i.m. immunization, which not only elicits antigen-specific CD8+ T-cells in systemic tissues but also high and sustained responses within the GT, as also reported recently by another group [25]. A second immunization given i.m. causes a robust booster effect within the GT of i.m.-primed mice, and gag-specific CD8+ T- cells remain detectable for at least one year. I.m. immunization is thus overall superior at inducing genital CD8+ T- cell responses by AdC vectors compared to i.n. immunization, and offers the added benefit of also eliciting potent systemic CD8+ T-cell responses, which may serve as a second layer of defense in case the virus breaks through the mucosal barrier. These findings are in agreement with a study in mice showing that intraperitoneal infection with lymphocytic choriomeningitis virus is superior to i.n. infection for the induction of CD8+ T-cell responses in the vaginal mucosa [26]. Other reports reached similar conclusions upon vaccination with protein vaccines [27], DNA vaccines [28], or DNA vaccines combined with Modified Vaccinia Ankara and Semliki Forest vectors [29]. Taken together, these results indicate that induction of CD8+ T-cell responses at mucosal sites upon i.m. immunization is independent of a given vaccine platform.

Antigen-experienced CD8+ T-cells may traffic to the GT with the help of specific sensors that remain to be identified, or alternatively this process may be random. To gain further insight into the vaccine-induced CD8+ T-cells that homed to the GT, we conducted a detailed phenotypical analysis of gag-specific CD8+ T-cells induced by the different immunization protocols, comparing cells isolated from spleen, blood, ILN and the GT at different times after immunization. In some assays we also tested cells isolated from NALT; the latter were tested for comparison as a population of cells homing to a distinct mucosal site. Phenotypes of gag-specific CD8+ T-cells isolated from systemic sites and the GT were phenotypically distinct, and this was especially pronounced at 1 year after the i.m./i.m. prime-boost vaccine regimen. The phenotypes suggest that most tet+CD8+ T-cells present in the GT remain fully activated and would be expected to start target cell lysis immediately upon encounter of infected cells.

We evaluated markers that are known to be up-regulated on cells derived from the intestinal mucosa. Studies have demonstrated high levels of CD69 expression on intestinal CD8+ cells [22, 30], but expression of CD69 was not increased in the GT at any of the time points analyzed. Although α4β7 has been linked to the genital migration of subsets of CD4+ cells [31], and is a well-known marker for homing of T-cells to the intestinal mucosa, our results do not suggest that α4β7 affects homing of CD8+ T-cells to the GT. CD103 was slightly increased in tet+CD8+ T-cells from the GT at early time points, and by 1 year after immunization became strongly up-regulated. In the adoptive transfer experiment, CD103 was low on the gag-specific CD8+ T-cells isolated from the vaccinated donors and upon transfer remained low on cells isolated from all compartments but the GT, where an increase was observed. Again, these data argues against the notion that CD103 supports mucosal homing but rather suggest that CD103 may contribute to the retention of CD8+ T-cells within the GT. The adoptive transfer experiment also showed that gag-specific CD8+ T-cells from the spleen could readily migrate to the GT to a similar extend as observed in vaccinated mice. This argues against the need for a distinct differentiation pathway during activation to allow for migration of CD8+ T-cells to the mucosa, as had been described for T-cell homing to GALT [32] or for CD4+ T-cells of the female GT [33]. On the other hand, the observation that at 2 weeks upon i.m. immunization frequencies of Gag-specific CD8+ T-cells were ~10 fold higher in blood but only ~ 2 fold higher in the GT than upon i.n. immunization argues for a deliberate homing process that is dictated by the conditions under which the CD8+ T-cells are initially stimulated. We would therefore assume that migration of activated CD8+ T-cells to the GT is in part random and affected by their overall frequencies in blood, and in part driven by the expression of yet to be identified homing markers. In either case, we would assume that activated CD8+ T-cells receive signals from the microenvironment that favor their retention once they reach the GT leading to an enrichment of this cells at the mucosal surface, which is the port of entry for many pathogens.

The functionality of genital CD8+ T-cells remains to be investigated in more depth. Our data thus far show that T-cells from the GT produce IFN-γ but not IL-2 as has also been reported for genital T-cells in SIV-infected NHP [34]. In our study, gag-specific CD8+ T-cells from the GT expressed high levels of granzyme B, perforin and Ki-67, which suggests that they are highly activated cells able to immediately commence target cell lysis and proliferation. Other authors have demonstrated atypical T-cells within mucosal surfaces [22] and we speculate that the high levels of lytic enzymes seen in memory-type CD8+ T-cells from the GT could be a result of a specific microenvironment.

In summary, data presented here show that i.m. immunization with a replication defective AdC vector in mice induces a robust transgene product-specific CD8+ T-cell response within the GT that can be enhanced by a booster immunization given i.m. The response is sustained and can still be detected 1 year after immunization. Vaccine-induced genital CD8+ T-cells are functional, they carry lytic enzymes and release cytokines upon antigenic stimulation. Taken together, results shown should allow for guarded optimism that potent vaccines administered i.m. may induce a genital barrier to HIV-1 infection in women. In fact, systemic regimens would be preferable over mucosal ones in humans due to logistical factors and the lack of interference by flora or menstrual cycle, which may profoundly affect mucosal vaccine efficacy.

Materials and Methods

1. Mice

Female 6- to 8-week-old BALB/c mice were obtained from Ace Animals Inc. (Boyertown, PA). Female 6- to 8-week-old Thy1.1 mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Animals were housed at the Animal Facility of The Wistar Institute (Philadelphia, PA) and all experiments were performed according to institutionally approved protocols.

2. Viral vectors

Purified E1-deleted Ad vectors expressing gag of HIV-1 clade B, derived from simian serotypes C6 (AdC6) or C68 (AdC68) were produced and quality controlled as described [8, 35].

3. Immunization of mice

Groups of five to twenty BALB/c mice were immunized by i.m. or mucosal routes with AdC vectors diluted to 1010 viral particles in sterile saline to a total volume of 10ul (i.n. and i.vag.) or 100ul (i.m.). Mice were immunized i.m. by injection into the lower leg muscle, while mucosal immunization was given with an automatic pipette. For mucosal immunizations, i.e., i.n. or i.vag., mice were first anesthetized with ketamine and xylazine chloride given intraperitoneally. Five days before i.vag. immunization, mice were intraperitoneally injected with 3mg of medroxiprogesterone-acetate (Sigma-Aldrich). For prime-boost experiments, mice were boosted i.n., i.vag. or i.m. 6 weeks after the first immunization.

4. Isolation of lymphocytes

Lymphocytes were isolated as previously described [36]. Briefly, blood was collected in heparin and red blood cells were lysed using ACK Lysing Buffer (Invitrogen). Spleens, ILN and NALT were dissociated against metal screens and washed with Leibovitch’s L-15 modified medium (Mediatech Inc.). For isolation of lymphocytes from the GT, the vagina, cervix, uterine horns and ovaries were removed and cut into fragments. Tissue segments were submitted to constant shaking at 130rpm for 1h in RPMI 1640 (Mediatech Inc.) containing 5% FBS and 1% penicillin streptomycin (Sigma-Aldrich). Fragments were enzymatically digested with 1.4 mg/ml of collagenase type I (Gibco) for 15 min. Cells from the two cycles were pooled and lymphocytes purified by a discontinuous Percoll gradient (Sigma Aldrich) consisting of 40% fraction containing cells overlaid over a 70% fraction.

5. Adoptive transfer of lymphocytes

BALB/c mice were primed i.m. with AdC6gag and boosted 6 weeks later with AdC68gag given i.m. Splenocytes were isolated 14 days later, and 5×107 splenocytes were injected intravenously (i.v.) into naïve Thy1.1 recipient mice. Tissues were analyzed for the presence of tet+CD8+ donor cells 7 days after the transfer.

6. Tetramer staining, phenotypic analyses and flow cytometry

Staining was performed using a gag peptide- (AMQMLKETI) and H-2Kd-specific tetramer (NIAID Tetramer Facility). For phenotyping, cells were incubated with the tetramer, and antibodies to CD8a, α4β7, CD27, CD103 (BD Pharmingen), CD44, CD62L, PD-1 (Biolegend), CD69, CD127 and NKG2D (eBioscience). Cells were permeabilized with BD Cytofix/Cytoperm™ Fixation and Permeabilization Solution (BD Bioscience) and stained for granzyme B, Ki-67 (BD Pharmingen), CTLA-4 (RD Systems) and perforin (eBioscience). For transfer experiments, cells were also stained for CD90.1 (Thy 1.1) (BD Pharmingen). Prior to analysis, cells were fixed with BD Stabilizing Fixative (BD Bioscience). Flow cytometric analyses of cells were performed with a BD LSR II (Becton-Dickinson) flow cytometer. Data were analyzed using FlowJo V8.8 software (Tree Star Inc.). BD CompBeads Compensation Particles (Becton-Dickinson) were used to set distinct negative- and positive-stained populations for the fluorochrome-labeled antibodies used in the experiments. For assessment of background and non-specific activation, we immunized animals with AdC vectors expressing an unrelated transgene; phenotypes for those cells mirrored those from naïve groups, as did frequencies of tetramer+CD8+ T-cells (data not shown). Samples were gated on live lymphoid cells, and then on CD8+ cells vs. side scatter (SSC), followed by gating on CD8+ cells vs. forward scatter (FSC). The remaining cells were then plotted as CD44 cells vs. tetramer for further analysis.

7. ELISpot assay

IFN-γ and IL-2 ELISpot assays were performed with lymphocytes as previously described [10]. Briefly, 96-well Millipore polyvinylidene difluoride plates were coated with anti-mouse IFN-γ or IL-2 antibody (BD Pharmingen) diluted in PBS and incubated overnight at 4°C. Plates were then washed and blocked with 10% MLC media (DMEM supplemented with 10-6M of 2-mercaptoethanol and 10% FBS) for 2h at 37°C. Lymphocytes were added to plates at 2×105 cells per well in triplicates, and stimulated with the 9mer peptide AMQMLKETI or the total pool of 123 15-mer peptides derived from consensus Gag clade B, in presence of anti-mouse CD28 and CD49d (BD Pharmingen) for 18–20h at 37°C in 10% CO2. Cells were removed and plates incubated with biotin-labeled antibody (BD Pharmingen) for 2h at room temperature. Streptavidin alkaline phosphatase (Mabtech AB) was added for 1 hour, and the spots developed by adding BCIP/NBT (Pierce) for 5 min. Plates were washed in water and dried before counting using the C.T.L. Series 3A Analyzer and ImmunoSpot 3.2 (Cellular Technology Ltd). Data from unstimulated cells were used as background control, and values were subtracted from sample values before plotting. In parallel, cells were stained with an antibody to CD8a and analyzed by flow cytometry to determine frequencies of this cell subset. These results were used to normalize the data obtained by ELISpot assays and data show numbers of SFU per 106 CD8+ cells. Samples that resulted in less than 55 SFU/106 cells were considered negative.

8. Statistical analysis

Each experiment was conducted repeatedly with 5-20 mice and figures show means and standard deviations based on independent experiments. Statistical significance of differences between groups was calculated by unpaired two-sample Student t-test using GraphPad Prism (GraphPad Software, LaJolla, CA, USA). The p values of p<0.05 or p<0.01 were considered statistically significant.

Supplementary Material



We wish to thank Christina Cole for assistance with preparation of the manuscript. This work was supported by grant AI074078-01 from the National Institutes of Health, by Wistar Cancer Center Support Grant P30 CA 010815 from the National Cancer Institute, and by the Gates Foundation (GCGH). Partial support was also provided by CAPES and PNDST/Aids, Brazil.


Chimpanzee-derived (simian) adenovirus
CTLA-4 - cytotoxic T-lymphocyte
associated antigen 4
Genital tract
Iliac lymph node
Median fluorescence intensity
Nasal-associated lymphoid tissue
Non-human primate
Programmed-death 1
Spot-forming units


Conflict of Interest The authors disclose there are no conflicts of interest.


1. Pikora CA. Glycosylation of the ENV spike of primate immunodeficiency viruses and antibody neutralization. Curr HIV Res. 2004;2:243–254. [PubMed]
2. Simmonds P, Balfe P, Ludlam CA, Bishop JO, Brown AJ. Analysis of sequence diversity in hypervariable regions of the external glycoprotein of human immunodeficiency virus type 1. J Virol. 1990;64:5840–5850. [PMC free article] [PubMed]
3. Jones PL, Korte T, Blumenthal R. Conformational changes in cell surface HIV-1 envelope glycoproteins are triggered by cooperation between cell surface CD4 and co-receptors. J Biol Chem. 1998;273:404–409. [PubMed]
4. McMichael AJ. HIV vaccines. Annu Rev Immunol. 2006;24:227–255. [PubMed]
5. Chan DJ. Fatal attraction: sex, sexually transmitted infections and HIV-1. Int J STD AIDS. 2006;17:643–651. [PubMed]
6. Centlivre M, Sala M, Wain-Hobson S, Berkhout B. In HIV-1 pathogenesis the die is cast during primary infection. Aids. 2007;21:1–11. [PubMed]
7. Fitzgerald JC, Gao GP, Reyes-Sandoval A, Pavlakis GN, Xiang ZQ, Wlazlo AP, Giles-Davis W, Wilson JM, Ertl HC. A simian replication-defective adenoviral recombinant vaccine to HIV-1 gag. J Immunol. 2003;170:1416–1422. [PubMed]
8. Pinto AR, Fitzgerald JC, Giles-Davis W, Gao GP, Wilson JM, Ertl HC. Induction of CD8+ T cells to an HIV-1 antigen through a prime boost regimen with heterologous E1-deleted adenoviral vaccine carriers. J Immunol. 2003;171:6774–6779. [PubMed]
9. Xiang Z, Gao G, Reyes-Sandoval A, Cohen CJ, Li Y, Bergelson JM, Wilson JM, Ertl HC. Novel, chimpanzee serotype 68-based adenoviral vaccine carrier for induction of antibodies to a transgene product. J Virol. 2002;76:2667–2675. [PMC free article] [PubMed]
10. Reyes-Sandoval A, Fitzgerald JC, Grant R, Roy S, Xiang ZQ, Li Y, Gao GP, Wilson JM, Ertl HC. Human immunodeficiency virus type 1-specific immune responses in primates upon sequential immunization with adenoviral vaccine carriers of human and simian serotypes. J Virol. 2004;78:7392–7399. [PMC free article] [PubMed]
11. Tatsis N, Fitzgerald JC, Reyes-Sandoval A, Harris-McCoy KC, Hensley SE, Zhou D, Lin SW, Bian A, Xiang ZQ, Iparraguirre A, Lopez-Camacho C, Wherry EJ, Ertl HC. Adenoviral vectors persist in vivo and maintain activated CD8+ T cells: implications for their use as vaccines. Blood. 2007;110:1916–1923. [PubMed]
12. Tatsis N, Lin SW, Harris-McCoy K, Garber DA, Feinberg MB, Ertl HC. Multiple immunizations with adenovirus and MVA vectors improve CD8+ T cell functionality and mucosal homing. Virology. 2007;367:156–167. [PMC free article] [PubMed]
13. de Souza AP, Haut LH, Silva R, Ferreira SI, Zanetti CR, Ertl HC, Pinto AR. Genital CD8+ T cell response to HIV-1 gag in mice immunized by mucosal routes with a recombinant simian adenovirus. Vaccine. 2007;25:109–116. [PubMed]
14. Tatsis N, Ertl HC. Adenoviruses as vaccine vectors. Mol Ther. 2004;10:616–629. [PubMed]
15. Ernst DN, Weigle WO, Noonan DJ, McQuitty DN, Hobbs MV. The age-associated increase in IFN-gamma synthesis by mouse CD8+ T cells correlates with shifts in the frequencies of cell subsets defined by membrane CD44, CD45RB, 3G11, and MEL-14 expression. J Immunol. 1993;151:575–587. [PubMed]
16. Karecla PI, Bowden SJ, Green SJ, Kilshaw PJ. Recognition of E-cadherin on epithelial cells by the mucosal T cell integrin alpha M290 beta 7 (alpha E beta 7) Eur J Immunol. 1995;25:852–856. [PubMed]
17. Cepek KL, Shaw SK, Parker CM, Russell GJ, Morrow JS, Rimm DL, Brenner MB. Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the alpha E beta 7 integrin. Nature. 1994;372:190–193. [PubMed]
18. Raulet DH. Roles of the NKG2D immunoreceptor and its ligands. Nat Rev Immunol. 2003;3:781–790. [PubMed]
19. Cullen SP, Martin SJ. Mechanisms of granule-dependent killing. Cell Death Differ. 2008;15:251–262. [PubMed]
20. Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands in tolerance and immunity. Annu Rev Immunol. 2008;26:677–704. [PubMed]
21. Johansson-Lindbom B, Agace WW. Generation of gut-homing T cells and their localization to the small intestinal mucosa. Immunol Rev. 2007;215:226–242. [PubMed]
22. Klein JR. T-cell activation in the curious world of the intestinal intraepithelial lymphocyte. Immunol Res. 2004;30:327–337. [PubMed]
23. Holmgren J, Czerkinsky C. Mucosal immunity and vaccines. Nat Med. 2005;11:S45–53. [PubMed]
24. Gupta S, Janani R, Bin Q, Luciw P, Greer C, Perri S, Legg H, Donnelly J, Barnett S, O’Hagan D, Polo JM, Vajdy M. Characterization of human immunodeficiency virus Gag-specific gamma interferon-expressing cells following protective mucosal immunization with alphavirus replicon particles. J Virol. 2005;79:7135–7145. [PMC free article] [PubMed]
25. Kaufman DR, Liu J, Carville A, Mansfield KG, Havenga MJ, Goudsmit J, Barouch DH. Trafficking of antigen-specific CD8+ T lymphocytes to mucosal surfaces following intramuscular vaccination. J Immunol. 2008;181:4188–4198. [PMC free article] [PubMed]
26. Suvas PK, Dech HM, Sambira F, Zeng J, Onami TM. Systemic and mucosal infection program protective memory CD8 T cells in the vaginal mucosa. J Immunol. 2007;179:8122–8127. [PMC free article] [PubMed]
27. Barnett SW, Srivastava IK, Kan E, Zhou F, Goodsell A, Cristillo AD, Ferrai MG, Weiss DE, Letvin NL, Montefiori D, Pal R, Vajdy M. Protection of macaques against vaginal SHIV challenge by systemic or mucosal and systemic vaccinations with HIV-envelope. Aids. 2008;22:339–348. [PubMed]
28. Wu H, Dennis VA, Pillai SR, Singh SR. RSV fusion (F) protein DNA vaccine provides partial protection against viral infection. Virus Res. 2009;145:39–47. [PMC free article] [PubMed]
29. Martinon F, Brochard P, Ripaux M, Delache B, Auregan G, Vaslin B, Le Grand R. Improved protection against simian immunodeficiency virus mucosal challenge in macaques primed with a DNA vaccine and boosted with the recombinant modified vaccinia virus Ankara and recombinant Semliki Forest virus. Vaccine. 2008;26:532–545. [PubMed]
30. Kuo S, El Guindy A, Panwala CM, Hagan PM, Camerini V. Differential appearance of T cell subsets in the large and small intestine of neonatal mice. Pediatr Res. 2001;49:543–551. [PubMed]
31. Hawkins RA, Rank RG, Kelly KA. Expression of mucosal homing receptor alpha4beta7 is associated with enhanced migration to the Chlamydia-infected murine genital mucosa in vivo. Infect Immun. 2000;68:5587–5594. [PMC free article] [PubMed]
32. Svensson M, Johansson-Lindbom B, Zapata F, Jaensson E, Austenaa LM, Blomhoff R, Agace WW. Retinoic acid receptor signaling levels and antigen dose regulate gut homing receptor expression on CD8+ T cells. Mucosal Immunol. 2008;1:38–48. [PubMed]
33. Ibraghimov AR, Sacco RE, Sandor M, Iakoubov LZ, Lynch RG. Resident CD4+ alpha beta T cells of the murine female genital tract: a phenotypically distinct T cell lineage that rapidly proliferates in response to systemic T cell activation stimuli. Int Immunol. 1995;7:1763–1769. [PubMed]
34. Stevceva L, Kelsall B, Nacsa J, Moniuszko M, Hel Z, Tryniszewska E, Franchini G. Cervicovaginal lamina propria lymphocytes: phenotypic characterization and their importance in cytotoxic T-lymphocyte responses to simian immunodeficiency virus SIVmac251. J Virol. 2002;76:9–18. [PMC free article] [PubMed]
35. Roy S, Gao G, Lu Y, Zhou X, Lock M, Calcedo R, Wilson JM. Characterization of a family of chimpanzee adenoviruses and development of molecular clones for gene transfer vectors. Hum Gene Ther. 2004;15:519–530. [PubMed]
36. Xiang ZQ, Gao GP, Reyes-Sandoval A, Li Y, Wilson JM, Ertl HC. Oral vaccination of mice with adenoviral vectors is not impaired by preexisting immunity to the vaccine carrier. J Virol. 2003;77:10780–10789. [PMC free article] [PubMed]