Integrin α₄β₇ Blockade Preferentially Impacts CCR6⁺ Lymphocyte Subsets in Blood and Mucosal Tissues of Naive Rhesus Macaques

Integrin α₄β₇ primarily mediates leukocyte migration to the gut and the GALTs by adhering to the mucosal vascular addressin cell adhesion molecule-1 (MAdCAM-1) on high endothelial venules of Peyer’s patches (PPs) and mesenteric lymph nodes (LNs), and on postcapillary venules of gut lamina propria (1, 2).

A humanized mAb specific for the α₄β₇ heterodimer, vedolizumab (Entyvio; Takeda Pharmaceuticals), was approved in 2014 for the treatment of moderately to severely active Crohn’s disease (CD) (3) and ulcerative colitis (UC) (4), two forms of inflammatory bowel disease (IBD). Within a relatively short period of time, vedolizumab has become a front-line therapy for UC and CD, achieving clinical remission at rates similar to those seen with TNF-α antagonists (5), the gold standard for therapeutic efficacy in UC and CD. The prevalent hypothesis is that vedolizumab reduces inflammation in the gut by blocking trafficking of α₄β₇⁺ lymphocytes to the gut. However, its mechanism of action is likely more complex because only a mild lymphoid depletion was observed in the PPs of macaques treated with up to 100 mg/kg of vedolizumab (6). Additionally, a significant fraction of IBD patients fail to respond to vedolizumab therapy for reasons that remain unclear (7).

Of note, the gut and GALT are critical sites for HIV replication, particularly during the acute phase of infection (8–10), and cells that express high levels of α₄β₇, α₄β₇^high CD4⁺ T cells are highly susceptible to HIV and SIV infection and preferentially depleted during the early stages of SIV infection (11–14). Moreover, increased frequencies of α₄β₇^high CD4⁺ T cells at the time of viral challenge were shown to correlate with increased susceptibility to rectal SIV infection and increased early plasma viral loads (15). Finally, it has been shown that factors associated with higher risk of HIV type 1 acquisition, like HSV type 2 infection and high progesterone levels (16, 17), are characterized by higher frequencies of α₄β₇^high CD4⁺ T cells within the mucosal tissues (18–20).

The i.v. administration of a simianized anti-α₄β₇ mAb (Rh-α₄β₇) with the same Ag-binding variable regions as the humanized vedolizumab, just prior to and during i.v. and rectal SIV infection, resulted in lower gut and lymphoid tissue viral loads, higher blood, and gastrointestinal CD4⁺ T cell counts and the absence of disease progression (21, 22). Of note, infusion of Rh-α₄β₇ prior to and during repeated low-dose vaginal challenges with SIVmac251 prevented SIV acquisition in half of the macaques challenged and, when infection did occur, Rh-α₄β₇ protected the GALT (23). More recently, short-term administration of Rh-α₄β₇ in combination with cART to treat acutely infected rhesus macaques led to prolonged (over 2 y) viral suppression after withdrawal of all therapeutic interventions (24).

A number of mechanisms have been proposed to explain the beneficial activity of Rh-α₄β₇ against SIV, including preventing the homing of CD4⁺ T cells to the GALT and interfering with HIV infection of α₄β₇^high CD4⁺ T cells (23). However, more recent data suggest that Rh-α₄β₇ does not block trafficking of total CD4⁺ T cells to the gut in uninfected RMs. Instead, it restores CD4 counts in the gut of SIV-infected RMs. This restoration is associated with the selective reduction of tissue viral loads in the gut and specific LNs (P.J. Santangelo, submitted for publication). Importantly, the studies with Rh-α₄β₇ were carried out at a dose of 50 mg/kg and it is not known if lower doses would be equally efficacious. In contrast, the current approved treatment of UC and CD with vedolizumab consists of 300 mg infusions (∼5 mg/kg) at weeks 0, 2, and 6, and every 8 wk thereafter, a dose that was shown to be saturating in blood in humans and macaques (6, 25).

To understand in greater detail the effects of Rh-α₄β₇ in vivo and to lay the foundation for studies aimed at defining the mechanisms by which Rh-α₄β₇ mediates its protective effect, we studied the phenotypic changes of subsets of lymphoid cells including activation markers in blood and mucosal tissues of naive rhesus macaques prior to and following infusion of Rh-α₄β₇. The studies were performed in two phases, with a focus on characterizing changes in the blood during the first study (blood-oriented study) and on changes in tissue lymphoid cells in the second (tissue-oriented study). Overall, treatment with Rh-α₄β₇ resulted in a trend toward leukocytosis and increased absolute numbers of both CD4⁺ and CD8⁺ T cells, but not of B cells. Moreover, profound changes in the frequencies and absolute numbers of different immune cell subsets, markers of cell activation, and cell trafficking were noted. The results of these studies constitute the basis of this report.

For the blood-oriented study, four juvenile-adult male rhesus macaques were housed and maintained at the Yerkes National Primate Research Center in accordance with the rules and regulations of the Committee on the Care and Use of Laboratory Animal Resources. They were treated and sampled as described in the results, and all the procedures were reviewed and approved by the Emory University Institutional Animal Care and Use Committee, fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International since 1985.

For the tissue-oriented study, 16 juvenile-adult female rhesus macaques were socially housed indoors, in climate-controlled conditions at the Tulane National Primate Research Center in compliance with the regulations under the Animal Welfare Act, the Guide for the Care and Use of Laboratory Animals (26). Animals were treated and sampled as described in the results and all procedures were approved by the Animal Care and Use Committee of the Tulane National Primate Research Center and in compliance with animal care procedures. None of the animals were treated with depot-medroxyprogesterone and the menstrual cycle was not monitored.

Levels of rhesus Rh-α₄β₇ Ab [clone Act-1; National Institutes of Health (NIH) Nonhuman Primate Reagents Resource center, MassBiologics, Boston, MA] in macaque plasma were measured using the α₄β₇-expressing human T cell line HuT 78 (NIH AIDS Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, NIH: HUT 78 from Dr. R. Gallo) in a flow cytometry–based assay as described previously (23). Briefly, HuT78 cells were first incubated in vitro for 3–4 d in RPMI 1640 media containing 100 nM retinoic acid to increase the surface expression of α₄β_7. Aliquots of these HuT78 cells (10⁵ cells) were incubated for 30 min at 4°C with plasma to be tested (undiluted, 1:5 or 1:10 diluted in PBS/2% FBS) obtained from each macaque before (day 0) and after (days 4 and 20) Rh-α₄β₇ mAb administration. Cells were then washed and incubated for 30 min at 4°C with biotinylated anti-rhesus IgG1 (Ab 7H11; NIH Nonhuman Primate Reagents Resource center), washed again, and resuspended in neutravidin-PE (Thermo Fisher Scientific) for 20 min at 4°C. Stained cells were analyzed on a flow cytometer for PE fluorescence. The plasma concentration of the Rh-α₄β₇ Ab was quantified using the standard curve method comparing the mean channel fluorescence intensity of cells treated with macaque plasma to the mean channel fluorescence intensity of cells treated with serially diluted Rh-α₄β₇ mAb (250–0 μg/ml; lower limit of detection 2 μg/ml; interassay coefficient of variability calculated on standard curves of seven different assays is 6.7%).

Soluble factors in plasma from treated macaques were measured using the macaque Novex multiplex Luminex assay (Cytokine Monkey Magnetic 29-Plex Panel; Invitrogen) on a Luminex 200 instrument (Luminex, Austin, TX). The factors measured included: IL1RA, CXCL11, MIF, FGF-Basic, CCL2, G-CSF, IFN-γ, CCL22, IL-15, CXCL8, EGF, HGF, VEGF, CXCL9, CCL5 (Rantes), CCL11, CCL4, CXCL10 (MCP-1), GM-CSF, TNF-α, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12, CCL3, and IL-17. IL-21 was measured by ELISA assay (Human ELISA Max Deluxe set; BioLegend) following the manufacturer’s instructions. Absorbance was read at 450 nm using an ELISA plate reader (Emax; Molecular Devices). Data were calculated using a linear regression curve-fitting algorithm with standard samples and interpolation of unknowns from standard curve (GraphPad Prism, version 5.0).

PBMCs were isolated using Ficoll-Hypaque density gradient centrifugation. Cells from mucosal tissues were isolated by enzymatic digestion with 1 mg/ml collagenase IV (Worthington Biochemical) and 1 mg/ml of human serum albumin (Sigma-Aldrich). The resulting cell suspension was passed through a 40 μm cell strainer. LNs were cut in small pieces and passed directly through a 40 μm cell strainer. Aliquots of the cell suspensions were stained with the viability dye LIVE/DEAD Aqua dye (Molecular Probes) before being incubated with a mixture of different panels of monoclonal Abs as listed in Supplemental Fig 1. IgA Abs were conjugated using labeling kits (Innova Biosciences). To quantify α₄β₇ receptor coverage following the administration of Rh-α₄β₇, cells isolated from blood and tissues before and at the indicated time points after treatment were incubated with mouse PE-Act-1 mAb (NIH Nonhuman Primate Reagent Resource center, MassBiologics). PE⁺ cells are indicative of free receptors. In addition, in efforts to distinguish the blocking effects of the Rh-α₄β₇ administration from depletion of cells expressing α₄β₇, we used a noncompetitive anti-α₄ mAb (we found no evidence for cell depletion). More than 200,000 events were acquired in the lymphocyte live cells gate using a BD LSRII flow cytometer. Data were analyzed using FlowJo 9.9.4 software (TreeStar, Ashland, OR).

Data from experiments involving a single treatment (blood-oriented study; Figs. 1–3, Supplemental Figs. 2, 3) were analyzed by a repeated measure one-way ANOVA with Bonferroni multiple comparison test; data from experiments including two treatments (tissue-oriented study; Figs. 4–9, Supplemental Fig. 2) were analyzed by repeated measure two-way ANOVA with Bonferroni multiple comparison test when applicable. In the instance of missing data points, statistical analysis was performed using a Wilcoxon matched-pairs t test (Wilcoxon signed rank) or a Mann–Whitney U test whenever pairing was not possible. The tests used for analysis of the data are specified in the corresponding figure legend. All analyses were performed using GraphPad Prism software. Significant p values < 0.05 (*), < 0.01 (**) and < 0.001 (***) are indicated.

To determine the effect of systemic Rh-α₄β₇ administration on blood lymphocytes in naive macaques, four rhesus macaques were treated with a single i.v. dose (50 mg/kg) of Rh-α₄β₇. This dose was used in all previous studies in SIV-infected macaques (21, 23, 24). Blood was collected at baseline and 2, 4, 6, and 8 wk after administration to monitor phenotypic changes in lymphoid cell populations. Receptor coverage was rapidly achieved and maintained for 6 wk (Fig. 1A). Blood cell counts revealed a trend toward leukocytosis at week 2, which reverted to baseline at subsequent time points (Supplemental Fig. 2A). Although Rh-α₄β₇ induced an increase in the absolute numbers of total CD4⁺, CD8⁺ T, and NK cells (but not B cells; Fig. 1B), the relative frequencies of naive, central memory (CM) and effector memory (EM) CD4⁺ and CD8⁺ T cell subsets did not significantly change with treatment (Fig. 1C). A small decrease in the frequency of CM CD4⁺ T cells was detected after Rh-α₄β₇ clearance at the end of the 8 wk follow-up period compared with earlier time points (Fig. 1C). In contrast, no changes in the frequencies of naive (CD21⁺/CD27⁻), resting (CD21⁺/CD27⁺), and memory (CD21⁻/CD27⁺) mature B cells (CD10⁻/CD20⁺/CD19⁺) were noted (data not shown). The frequencies of CD127⁻ CD25^high regulatory T cells (Treg-like) increased nonsignificantly at week 2 postinfusion followed by a significant decrease at week 8 compared with week 2 (Supplemental Fig. 2B). The examination of activation markers revealed a significant decrease in the frequencies of CD69⁺ CD4⁺ T cells and B cells at 2 wk posttreatment and a nonsignificant decrease in CD69⁺ CD8⁺ T cells at 2 wk, which gradually increased (Fig. 2; absolute numbers in Supplemental Fig. 3). In contrast, the frequency of CD25⁺ cells increased within the CD4⁺ and CD8⁺ lymphocytes (Fig. 2A, 2B), but decreased within B cells (Fig 2C; absolute numbers in Supplemental Fig. 3). Finally, the expression of HLA-DR on B cells and the frequency of HLA-DR⁺ NK cells decreased 2 wk postinfusion followed by a gradual return to baseline after mAb clearance (Fig. 2D). The analysis of cell adhesion and homing markers revealed that the Rh-α₄β₇ profoundly decreased the frequencies of CD103⁺ cells within all three subsets (CD4⁺ T, CD8⁺ T, and B cells; Fig. 2) and they also decreased in absolute numbers (Supplemental Fig. 3). In contrast, CCR6⁺ CD4⁺ T cells and B cells substantially increased by week 4 posttreatment both in terms of relative frequency (Fig. 2) and absolute numbers (Supplemental Fig. 3). However, the relative frequency of CCR6⁺ CD8⁺ T cells remained stable (Fig. 2). Interestingly, gut-homing CCR9⁺ cells increased significantly only within the CD4⁺ T cells (Fig. 2A), whereas the frequency of CCR5⁺ cells increased in both CD4⁺ and CD8⁺ T cells. Specifically, increases in CCR5⁺ and CCR9⁺ cells were detected within naive and CM subsets, but not within the EM subsets (Fig. 3). Finally, the frequency of CD62L⁺ cells increased significantly in naive and CM CD8⁺ T cells, but not in CD4⁺ T cells (Fig. 3B, data not shown). Of note, although we detected an increase in the frequency of CCR6⁺ CD4⁺ T cells, no significant changes compared with baseline were present in Th17-precursors CCR6 CD161 double-positive CD4⁺ T cells (27) (Supplemental Fig. 2C). In summary, following Rh-α₄β₇ administration, specific changes were noted in blood lymphocytes, particularly in tissue-homing CD4⁺ T cell and B cell subsets (Table I).

Following the in-depth analysis of blood lymphocyte subsets, we sought to determine how Rh-α₄β₇ impacts lymphocyte subsets within mucosal tissues. To this end, we performed a separate study in naive macaques focused on tissue-specific changes (tissue-oriented study). We also decided to investigate potential differences between the 50 mg/kg dose used in the macaque/SIV studies (21, 23, 24, 28) and a 5 mg/kg dose that is analogous to that employed in vedolizumab treatment of IBD patients (29). Groups of naive female rhesus macaques were treated with a single i.v. dose of Rh-α₄β₇ (50 mg/kg, n = 6 or 5 mg/kg, n = 4) or with a control Rh-IgG1 (50 mg/kg, n = 3 or 5 mg/kg, n = 3). Blood and tissue biopsies (vaginal, rectal, duodenum, and LNs) samples were obtained at baseline, 4, 6, and 20 d after treatment. A schematic of the sampling schedule is shown in Fig. 4A. Peak circulating levels of ∼150 μg/ml were reached at day 4, and sustained levels (≥100 μg/ml) persisted at day 20 at both the high and low dosages. Both doses resulted in comparable blood Rh-α₄β₇ levels (Fig. 4B). Although no significant changes in either CD4⁺ or CD8⁺ T cell counts were yet observable after administration of 50 mg/kg of Rh-α₄β₇ 4 d after injection compared with baseline, lymphocytosis was already detectable in the animals treated with the lower dose (Fig. 4C; day 20 not measured). Complete α₄β₇ receptor coverage on CD4⁺ T cells was achieved by both doses at 4 d postinfusion in blood, but not in rectal and vaginal tissues, where complete coverage was achieved by day 20 (Fig. 4D). Of note, although significant, receptor coverage was incomplete in the duodenal biopsies at either dose (small intestine; Fig. 4D) at days 4–6 (the only time point measured).

To determine the impact of Rh-α₄β₇ on trafficking of α4β7 CD4⁺ T cell subsets in tissues, we monitored the frequencies of α4⁺ CD4⁺ T cells with an anti-α4 mAb, whose binding is not affected by Rh-α4β7 binding to α4β7. We monitored the frequencies of total α4⁺ CD4⁺ T cells as well as of α4⁺ CD95⁺ CD4⁺ T cells, because α4β7^high CD4⁺ T cells, which are preferentially infected or depleted by HIV and SIV (11, 12, 18, 30) are all CD95⁺ memory CD4⁺ T cells. Surprisingly, both total α4⁺ and the α4⁺ CD95⁺ CD4+ T cell subset in blood and tissues were mostly unaffected by the Rh-α4β7 treatment (data not shown, Fig. 4D) with only a small decline in the frequency of α4⁺ CD95⁺ CD4⁺ T cells in the rectal tissue that was significant at day 4 compared with day 20 of sampling.

Interestingly, we found that among the soluble factors measured by the 29-Plex Luminex assay 4 d postinfusion, the plasma levels of IL-1β, CCL5 (Rantes), and CCL2 (MCP-1) were significantly increased in Rh-α₄β₇–treated animals relative to baseline, but not in the controls (Fig. 5; day 20 not measured). Moreover, because IL-21 was one of the factors significantly increased in Rh-α₄β₇-antiretroviral–treated macaques that controlled viral replication after analytical treatment interruption (24), we measured the levels of IL-21 in our naive animals. However, no impact of Rh-α₄β₇ on IL-21 was detectable at least on the samples collected on day 4 after infusion (Fig. 5), which implies the effect on IL-21 maybe a consequence of SIV infection.

Because Rh-α₄β₇ did not significantly decrease the frequencies of α₄⁺ CD4⁺ T cells in tissues, we investigated whether it affected other specific cell subsets that are known to play a role in HIV/SIV infection. The data from the different doses were pooled, because similar circulating levels and receptor coverage were achieved with the two doses of Rh-α₄β₇ and no detectable difference in cell phenotype was observed between doses (with one exception; see below and Fig. 6 legend). Interestingly, although there was a nonsignificant increase in the frequency of CD69⁺ CD4⁺ T cells in the rectal tissue at day 4 posttreatment, which then decreased by day 20 posttreatment, a decrease in the frequency of CD69⁺ CD4⁺ T cells was already present at day 4 in the vaginal tissue (Fig. 6). A profound decrease in the frequency of CCR6⁺ CD4⁺ T cells was already present at day 4 posttreatment in the rectal tissue, whereas a nonsignificant decrease was present in the vaginal tissue after 20 d (Fig. 6). Interestingly, the frequencies of CCR6^high CD95⁺ and CXCR3⁺ CCR6⁺ CD4⁺ T cells were decreased in the rectal tissue, but not the frequency of CXCR3^- CCR6⁺ true Th17-like CD4⁺ T cells (27, 31) (Supplemental Fig. 2). A significant decrease in the frequency of CXCR3⁺ CD4⁺ T cells was also detectable in the rectal tissue by day 20 posttreatment and by day 4 in the vaginal tissue (Fig. 6). Moreover, a significant decrease in the frequency of CCR5⁺ CD4⁺ T cells was present by day 4 in the rectal tissue only in the animals treated with the 50 mg dose and by day 20 in both rectal and vaginal tissues. Of note, the frequency of CD4⁺ T cells expressing the active conformation of integrin LFA-1 (detected with clone MEM148) decreased significantly by day 20 posttreatment in the vaginal tissue (Fig. 6), whereas it increased in the small intestine by day 6 (Fig. 6 and day 20 not examined), and it did not change in the colorectal tissues (data not shown).

To study the effects of Rh-α₄β₇ on lymphoid cells and, in particular, CD4⁺ T cells in LNs, inguinal LNs from both treatment groups were sampled at day 6 postinfusion and the data of the different doses pooled. Significant, near-complete α₄β₇ receptor coverage was achieved on CD4⁺ T cells (Fig. 7A) and, in contrast to the other tissues examined, in the LNs the treatment resulted in a significant reduction of α₄⁺ CD4⁺ T cells (both total, data not shown, and CD95⁺ Fig. 7B). Moreover, the frequencies of CD69⁺ and CD62L⁺ CD4 T cells were significantly lower in the LNs of Rh-α₄β₇–treated animals compared with IgG1-treated controls (Fig. 7C, 7D). Interestingly, the frequency of Foxp3⁺ CD4⁺ Tregs also decreased significantly (Fig. 7E; even more when gated within total CD3⁺ T cells, Fig. 7F). Of note, we found that virtually all the CXCR5⁺ PD1^high T follicular helper cells express α4β7 (Fig. 7H, Supplemental Fig. 4). However, perhaps because of the poor receptor coverage (∼30%, Fig. 7H), which may indicate low penetration of the Rh-α4β7 in the follicular area, there was no significant difference in the frequencies of these cells between the treatment groups (Fig. 7G).

Although no detectable changes were noted on the absolute numbers of circulating B cells following the administration of Rh-α₄β₇ over time (Fig. 1B, data not shown), we observed a transient significant increase of naive (CD27^-) and a corresponding decrease of memory (CD27⁺) B cells 4 d after treatment in lymphoid cells isolated from rectal tissues. However, by day 20 the frequencies reverted to baseline values (Fig. 8). Similarly, we detected a significant increase of naive and a decrease of memory B cells in the small intestine 6 d after treatment (Fig. 8). Moreover, although the frequency of IgA⁺ B cells was significantly decreased in the small intestine at day 6 posttreatment, it was not affected in the LNs (data not shown).

The frequencies of plasmacytoid dendritic cells (pDC) (defined as CD123⁺, CD11c⁻ HLA-DR⁺ LIN⁻) and myeloid DC (mDC) (defined as CD123⁻, CD11c⁺ HLA-DR⁺ LIN⁻) were characterized by flow cytometry in both blood and tissues. Although complete receptor coverage was not achieved on either population of DCs in blood or in tissues (Fig. 9B, 9D), we observed a significant decrease in the frequency of pDC within total live cells in the rectal tissue by day 4 posttreatment (Fig. 9A). Similarly, a nonsignificant trend toward decreased levels of pDCs was observed in the vaginal tissue 4 d after treatment (Fig. 9A; day 20 not measured). In contrast, a significant increase in the frequency of mDC was detected 6 d posttreatment in the small intestine and a nonsignificant increase in the colorectal tissue of Rh-α₄β₇–treated animals compared with IgG1 controls (Fig. 9C).

Integrin α₄β₇ has an important role in trafficking immune cells to the gut lamina propria and inductive sites (2, 32, 33); however, α₄β₇ also acts as a costimulatory molecule and its conformational activation contributes to cell proliferation (34, 35). We reason that different immune cell subsets may have different requirements for trafficking to different mucosal tissues. However, little is known about subset-specific immune cell trafficking. Integrin β₇^−/− deficient mice have been used to demonstrate the requirement for integrin β₇ in the formation of morphologically mature PPs and lymphoid aggregates of the large intestine (36, 37). Moreover, the overall amount of CD4⁺ and CD8⁺ T cells seem to be similar in the gut lamina propria of β₇^−/− and wild-type mice, whereas B cells and Treg are reduced both in the lamina propria and gut lymphoid aggregates (36). It is important to note that integrin β₇ also pairs with α_E, forming α_Eβ₇. By binding to E-cadherin, α_Eβ₇ contributes to mucosal-specific retention of leukocytes within epithelia (38). β₇^−/− deficient mice reflect the loss of both α₄β₇ and α_Eβ₇, whereas the impact of specific blocking of α₄β₇ with the mAb Rh-α₄β₇ on trafficking of specific immune cell subsets is still largely unexplored, even though this anti-α₄β₇ constitutes a first-line treatment for IBD (39) and has shown to be effective in protecting against SIV infection (21, 23, 24). In this study, we sought a better understanding of the impact of the simianized anti-α₄β₇ Rh-α₄β₇ on lymphocyte subsets in blood and tissues. Interestingly, although it was previously reported that CD4⁺ T cells in the gut do not seem to show any major changes following anti-α₄β₇ treatment in the uninflamed intestine [Ref. (36) and P.J. Santangelo, submitted for publication], we detected a significant increase in the absolute counts of CD4⁺ and CD8⁺ T cells and NK cells in Rh-α₄β₇–treated animals. This is in contrast with reports that vedolizumab does not induce lymphocytosis (40). Of note, Rh-α₄β₇ differs from vedolizumab in several aspects, including being a simianized mAb, retaining the ability to bind Fc receptors, and in pharmacokinetics (6, 25). It is possible that this explains some of the differences between the in vivo effects of vedolizumab and Rh-α₄β₇. Interestingly, Rh-α₄β₇ did not impact the absolute number of circulating B cells and, despite the significant increase in CD4⁺ and CD8⁺ T cells, there was no change compared with the baseline in the relative frequencies of naive, CM, and EM subsets in the blood of Rh-α₄β₇–treated animals. Notably, using an α₄ mAb that does not compete with the Rh-α₄β₇ mAb, we could not detect a significant decrease of α₄⁺ lymphocytes in any of the tissues, with the exception of LNs, in contrast with several reports of a significant decrease in the frequency of β₇⁺ lymphocytes in the gut of Rh-α₄β₇/vedolizumab-treated naive animals (6, 41). This may be due to those studies using anti-β₇ monoclonal or polyclonal Abs. Because the parental anti-α₄β₇ clone, Act-1, binds specifically to the β₇ subunit to create an epitope that forms when β₇ is in complex with α₄ (42), it is likely that the Rh-α₄β₇ interferes with most, if not all, available anti-β₇ mAbs (we have observed interference with the anti-β₇ mAb clone FIB21, E. Martinelli, unpublished observations). We submit that this may explain the profound decrease in β₇⁺ cells in the mucosa of Rh-α₄β₇/vedolizumab-treated animals seen by others (6, 41). A lack of decrease in α₄⁺ could not be attributed to the presence of α₄β₁⁺ β₇⁻ cells because in mucosal tissue the vast majority of α₄⁺ CD4⁺ T cells coexpress both β₁ and β₇. However, a compensatory influx of α₄β₁⁺ β₇⁻ cells in Rh-α₄β7⁺–treated animals could not be excluded. In contrast, we found that circulating T and B cells expressing the tissue-specific markers CD103 and CD69 were relatively decreased compared with baseline in Rh-α₄β₇–treated animals. This is despite the fact that circulating CD103⁺ and CD69⁺ CD4⁺ T cells coexpress α₄β_7. In terms of absolute numbers, circulating CD69⁺ CD4⁺ T cells were mostly unchanged (Supplemental Fig. 2) and thus, probably simply unaffected by Rh-α₄β₇ treatment. In contrast, CD103⁺ cells decreased both in terms of frequency and absolute number (nonsignificant for CD4⁺ T cells; Supplemental Fig. 3). This could be a result of the migration of this subset to the tissues. However, CD103 was not monitored in the tissue-oriented study. Instead, the changes in frequencies of CD69⁺ CD4⁺ T cells in the tissues may be explained by a loss of CD69 expression due to an impact of Rh-α₄β₇ treatment on the local microenvironment (43, 44).

The increased frequencies of T cell subsets expressing CD25 and CCR5 in blood following Rh-α₄β₇ treatment suggests that the administration of Rh-α₄β₇ may result in blocking the trafficking of these subsets to the tissues. Indeed, in the tissue-oriented study, we found a decreased frequency of CCR5⁺ CD4 T cells in both vaginal and rectal tissue compared with baseline. Despite a profound effect of Rh-α₄β₇ on these cell subsets, only about half of circulating CCR5⁺ or CD25⁺ cells express α₄β₇. Notably, the CCR6⁺ CD4⁺ T cell subset was the most profoundly impacted by Rh-α₄β₇, with a preferential increase in blood and a corresponding decrease in the colorectal tissue. This may be related to the recognized role of CCR6 in mucosal homing (45, 46). However, only about half of CCR6⁺ cells express α₄β₇ and about half of circulating α₄β₇⁺ CD4⁺ T cells are CCR6⁺. Thus, direct interference with trafficking does not entirely explain the profound preferential impact on this marker. Nonetheless, the preferential decrease of CCR6⁺ cells in the mucosa may contribute to the explanation of the protective effect of Rh-α₄β₇ against vaginal SIV acquisition, because of the high susceptibility of CCR6⁺ CD4⁺ T cells to HIV/SIV infection (47–49).

Interestingly, cells expressing the mucosal homing receptor CCR9 were increased within blood CD4⁺ T cells, but not within CD8⁺ T cells. The opposite is true for CD62L⁺ cells, whose frequency in blood was increased within the naive and CM CD8⁺ T cells, but relatively unchanged within the CD4⁺ T cell subsets. This suggests that Rh-α₄β₇ has a different effect on diverse subsets of CD4⁺ and CD8⁺ T cells.

Notably, the number of B cells and relative frequencies of blood B cell subsets was stable. Although it was not possible to obtain absolute cell counts in the tissues, the frequency of B cells within total live cells in tissues did not decrease following Rh-α₄β₇ treatment. However, interestingly, the frequency of blood CD69⁺ and CD25⁺ B cells profoundly decreased in Rh-α₄β₇–treated animals and the decrease was significant even when considering the absolute number (Supplemental Fig. 4). This suggests that Rh-α₄β₇ has selectively induced the loss of CD69 and CD25 expression on B cells as opposed to their redistribution to the tissues. These findings, together with reduced HLA-DR expression, suggest that Rh-α₄β₇ may be decreasing the Ag-presentation capability of circulating B cells (50). Rh-α₄β₇ also profoundly increased the frequency of circulating CCR6⁺ B cells, confirming the critical role of CCR6 in gut homing of B cells and the overlapping phenotype of CCR6^−/− deficient mice and mice treated with anti-α₄β₇ mAb, especially in the formation of cryptopatches and isolated lymphoid follicles (36, 46). Moreover, Rh-α₄β₇ induced a significant change in the relative frequency of memory and naive B cells in both the small intestine and colorectal tissue, which, considering the absence of parallel changes in blood, cannot be simply attributed to interference with trafficking of memory B cells. In line with this result, we found a decreased frequency of IgA-expressing B cells in the small intestine. A similar decrease in the frequency of IgA⁺ B cells is reported in the gastrointestinal tract of β₇^−/− deficient mice (2). One possible explanation is that an interaction between α₄β₇-MAdCAM-1 is needed for appropriate B cell maturation within the gut.

Finally, although coverage of α₄β₇ by Rh-α₄β₇ on blood DCs was not complete, we found a reduced frequency of pDCs in the rectal tissue of Rh-α₄β₇–treated animals compared with baseline. Reduced pDC trafficking to the colorectal area following SIV challenge has already been reported in animals that were treated with the Rh-α₄β₇ prior to challenge (22). However, it was not known that this phenomenon is independent of SIV infection. In contrast, mDCs were impacted only in the small intestine. This effect of Rh-α₄β₇ on DCs subsets again suggests that Rh-α₄β₇ does not generically block trafficking of all α₄β₇⁺ cells to the gut, but it selectively impacts specific DC subsets that perhaps depend more on α₄β₇ for trafficking or maturation. We also cannot exclude that some of the effects on DCs and CD8⁺ T cells were the result of indirect effects of changes in other lymphocyte subsets.

In conclusion, in this study we show that the administration of Rh-α₄β₇ profoundly and preferentially impacts trafficking of CCR6⁺ CD4⁺ and B cells to mucosal tissues and reduces the frequencies of memory B cells in the gut even in the absence of detectable inflammation. This suggests that interaction of α₄β₇ with MAdCAM-1 may be more critical for trafficking and perhaps maturation of these select subsets. Future work will need to address the impact of HIV/SIV infection on the Rh-α₄β₇–driven changes and how Rh-α₄β₇ affects other immune cells especially those of the innate immune system.

We are very grateful to the veterinary staff and study coordinators at both the Yerkes Primate Research Center of Emory University, Atlanta, GA, and the Tulane Primate Research Center of Tulane University, Covington, LA, for superb assistance and technical help during the course of these studies.

The authors have no financial conflicts of interest.