Gut Mucosal FOXP3⁺ Regulatory CD4⁺ T Cells and Nonregulatory CD4⁺ T Cells Are Differentially Affected by Simian Immunodeficiency Virus Infection in Rhesus Macaques^▿

Animals and experimental SIV infection.

Rhesus macaques (Macaca mulatta) of Chinese origin were housed at the German Primate Center, Goettingen, Germany. Animal care and handling were performed under the German Animal Protection Law and in accordance with the guidelines of the German Primate Center and were reviewed and approved by the appropriate authorities. Test animals were seronegative for SIV, simian T-lymphotropic virus type 1, and simian retrovirus. Animals were infected with 300 50% tissue culture infective doses (TCID₅₀) of SIVmac251 intravenously. Three animals (group 1) were infected for at least 9 months before euthanasia, which was performed in accordance with federal guidelines and institutional policies. Colon tissue was obtained at necropsy. For longitudinal analysis, colonic biopsy specimens from five additional animals (group 2) were collected under combined ketamine, xylazine, and atropine anesthesia 1 week before and 1, 2, 3, 4, 7, 10, and 12 weeks after infection as described previously (2). Plasma viral load peaked at 2 weeks postinfection and remained detectable throughout the course of SIV infection in all animals. None of the animals developed AIDS-like disease during the observation period. The levels of viral load and CD4⁺ T cells in peripheral blood and colonic mucosa are given in Table 1.

Isolation of mucosal cell subsets.

Mucosal lamina propria lymphocytes were isolated from colon sections by EDTA treatment and enzymatic digestion (49). CD4⁺ CD25⁺ and CD4⁺ CD25⁻ lymphocytes were isolated by using the nonhuman primate CD4⁺ CD25⁺ regulatory T-cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer's protocol, using two passages on magnetic columns. Immediately thereafter, the purity of isolated cell subsets was verified by flow cytometric analysis, and remaining cells were quickly snap-frozen and stored at −80°C for subsequent DNA/RNA extraction.

Flow cytometric analysis.

Flow cytometric analysis of T_reg was performed using antibodies against CD4 (clone L200; BD Biosciences, Heidelberg, Germany), CD8 (DK25; Dako, Hamburg, Germany), CD25 (ACT-1; Dako), and FOXP3 (PCH101; eBioscience, San Diego, CA) according to the manufacturer's FOXP3 staining protocol. Antibodies were conjugated to fluorescein, phycoerythrin, peridinin chlorophyll protein, or allophycocyanin. Data were acquired on the FACSCalibur flow cytometer (BD) and analyzed with CellQuest software (BD). T_reg were phenotypically identified by their coexpression of FOXP3 and CD25 within the CD4⁺ CD8⁻ lymphocyte gate.

Quantification of viral RNA and DNA in mucosal tissue and mucosal cells.

Total DNA and RNA were simultaneously purified from biopsy specimens or mucosal cell subsets with the use of the Allprep DNA/RNA minikit (Qiagen, Hilden, Germany) according to the manufacturer's protocol, except that both DNA and RNA were eluted in 70 μl. On-column digestion of DNA was performed during RNA purification using the RNase-free DNase set (Qiagen) in order to remove DNA carryover. Viral RNA was reverse transcribed by using the SuperScriptTM III reverse transcriptase kit (Invitrogen, Karlsruhe, Germany) with 0.5 μM of the oligonucleotide SIVgag1521rc (5′-TCAATTTTACCCAGGCATTTAATGT-3′). Viral DNA or viral RNA was quantified by a TaqMan-based real-time PCR (ABI-Prism 7500 real-time PCR system; Applied Biosystems, Darmsatdt, Germany) using 50 to 500 ng and 0.5 to 1 μg template DNA extracted from biopsy specimens and cell subsets, respectively, or 10 μl cDNA and the following oligonucleotides specific for the gag region of SIVmac251: SIVgag1444fw (5′-ACCCAGTACAACAAATAGGTGG-3′), SIVgag1521rc, and the 6-carboxyfluorescein (FAM)-labeled probe SIVgag1472 (5′-FAM-TGTCCACCTGCCATTAAGCCCGAG-BBQ-3′). In parallel with each DNA sample, simian CD4 gene copies were quantified in order to normalize the SIV DNA input using the following oligonucleotides: CD4mm163776fw (5′-TGTAGTGTGCCAGTTTAGTGC-3′), CD4mm163918rc (5′-GCACTATGGCAGTTAACATCATC-3′), and the YAK-labeled probe CD4mm163798 (5′-YAK-TGGACCCTGGTTGAAATCCTGGTTCTGC-BBQ-3′). Samples were analyzed in a total volume of 50 μl of a mixture containing PlatinumQ PCR SuperMix-UDG (Invitrogen), 400 nM each oligonucleotide, and 100 nM probe. Thermal cycling conditions consisted of 2 min at 50°C and 2 min at 95°C, followed by 50 cycles of 15 s at 95°C and 45 s at 60°C. Amplification, data acquisition, and analysis were performed using the ABI Prism 7500 sequence detection system (Applied Biosystems). The limit of detection of the SIVgag assay and the CD4 assay was 20 copies/reaction. Based on two CD4 gene copies per cell, the input number of cells was calculated in each assay, and SIV DNA copy numbers were related to cell number. Results are given as SIV DNA copies/10⁶ cells. SIV RNA copies/10⁶ cells were calculated in each assay from SIV RNA copy numbers/10 μl based on the volume of input DNA used for the corresponding quantification of SIV DNA copies/10⁶ cells.

Gene expression analysis.

RNA was extracted from isolated mucosal CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cell subsets using the NucleoSpin RNA II kit (Macherey-Nagel, Düren, Germany) according to the manufacturer's instructions. Quality control for the integrity and purity of the RNA preparation was performed on a capillary electrophoresis system (Agilent 2100 Bioanalyzer; Agilent Technologies, Böblingen, Germany). RNA was amplified and labeled with the fluorescent dye carbocyanine 3 (Cy3) using the Agilent low-RNA input linear amplification kit (Agilent Technologies) following the manufacturer's protocol. The amount of cRNA and the Cy3 incorporation rate were measured using the ND-100 spectrophotometer (NanoDrop Technologies, Wilmington, DE). cRNA microarray analysis was carried out using the 44K Agilent whole rhesus-genome oligo microarray (Agilent Technologies). Hybridization was performed according to the Agilent 60-mer oligo microarray processing protocol using the Agilent gene expression hybridization kit (Agilent Technologies), and fluorescence signals were detected using Agilent's scanner system (Agilent Technologies). Raw microarray image files were read out and processed using Agilent feature extraction (Agilent Technologies) and analyzed with the Rosetta Resolver gene expression data analysis system (Rosetta Biosoftware, Seattle, WA). All procedures were performed by customer service of Miltenyi Biotec, Germany. Normalized signal intensities for two sets of selected genes were analyzed for transcriptional characterization of mucosal T_reg and non-T_reg.

Immunohistochemistry and immunofluorescence staining.

For immunostaining, 2- or 3-μm sections were cut, deparaffinized, and subjected to a heat-induced epitope retrieval step before incubation with antibodies. Sections were immersed in sodium citrate buffer solution at pH 6.0 and heated in a high-pressure cooker. The slides were rinsed in cool running water, washed in Tris-buffered saline (pH 7.4), and incubated with primary antibodies. The primary antibodies included monoclonal antibodies against CD3 (clone UCHT1, dilution 1:25; Dako, Glostrup, Denmark), CD4 (1F6, 1:5; Novocastra Laboratories, Newcastle upon Tyne, United Kingdom), FOXP3 (PCH101, 1:100; eBioscience), Ki67 (MIB-1, 1:2,000; Dako), and Perforin (Pf-344). For detection, the streptavidin-AP kit (K5005; Dako) and the Envision PO kit (K5007; Dako) were used. Only the lamina propria was analyzed, and lymphoid aggregates and intraepithelial lymphocytes were excluded. For quantification of perforin⁺ cells, lamina propria and intraepithelial cells were analyzed. The frequency of T_reg per mucosal CD4⁺ T cells was determined by dividing the number of CD3⁺ FOXP3⁺ cells by the number of mucosal CD4⁺ cells. The frequency of proliferating cells within mucosal CD4⁺ cells was determined by dividing the number of CD4⁺ Ki67⁺ cells by the number of mucosal CD4⁺ cells. For immunofluorescence labeling, sections were incubated with rat anti-FOXP3 antibody (PCH101, 1:20; eBioscience) followed by Alexa Fluor 555-conjugated anti-rat antibody (1:100; Invitrogen, Carlsbad, CA), washed three times in PBS, and incubated with mouse anti-Ki67 (MIB-1, 1:1,000; Dako,) followed by Alexa Fluor 488-conjugated anti-mouse antibody (1:100; Invitrogen). Nuclei were counterstained with DAPI (4′,6-diamidino-2-phenylindole) (1:1,500; Roche, Mannheim, Germany), and slides mounted in Fluoromount-G (Southern Biotech, Birmingham, AL). Images were acquired using a fluorescence microscope (AxioImager Z1) equipped with a charge-coupled-device (CCD) camera (AxioCam MRm) and processed with Axiovision software (Carl Zeiss MicroImaging, Inc.). Negative controls were performed by omitting the primary antibodies. Positive cells were quantified in intestinal tissues per high-power field (hpf) (1 hpf = 0.237 mm²), and 10 hpf were averaged in each case.

Statistical analysis.

Data are represented as medians ± the standard deviation (SD) and were analyzed by using the two-tailed Student t test or the Wilcoxon test. A P value of less than 0.05 was considered to be statistically significant. Correlations were determined using linear regression (Pearson correlation; GraphPad Prism; GraphPad Software, San Diego, CA).

Isolation of mucosal T_reg and non-T_reg subsets.

Mucosal CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cells were isolated from the colon sections of three chronically SIV-infected animals (Table 1, group 1) and a healthy SIV-uninfected animal. SIV-infected animals were examined immunohistochemically for absolute numbers of mucosal T_reg and their frequency within the absolute number of mucosal CD4⁺ T cells (relative frequency of mucosal T_reg) 1 week before the infection and at necropsy. As expected and consistent with our previous findings for chronic HIV infection, absolute and relative frequencies of T_reg in colonic mucosa were markedly increased in the chronically SIV-infected animals compared to preinfection levels: absolute and relative frequency increased from 9.1 ± 6.6/10 hpf and 5.6 ± 2.1% to 25.0 ± 13.4/10 hpf and 53.8 ± 12.8%, respectively.

After isolation, mucosal CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cells were analyzed for CD4, CD25, and FOXP3 expression by flow cytometry. The purity of the CD4⁺ CD25⁺ cell subset and the CD4⁺ CD25⁻ cell subset ranged from 87.8 to 97.4% and from 89.7 to 98.2%, respectively. The CD4⁺ CD25⁺ cell subset contained 77.0 to 93.3% FOXP3-expressing cells. Flow cytometric analysis of mucosal cell subsets isolated from colonic mucosa of one SIV-infected animal is shown in Fig. 1. In this animal, FOXP3 transcript expression analyzed by microarray was detectable with isolated CD4⁺ CD25⁺ cells but not with isolated CD4⁺ CD25⁻ cells. Since the majority of the isolated CD4⁺ CD25⁺ cells expressed FOXP3, representing the key factor for the function of regulatory CD4⁺ T cells, we refer to these cells as the mucosal T_reg subset. Isolated CD4⁺ CD25⁻ cells are designated the mucosal non-T_reg subset.

Low efficiency of SIV RNA production in infected mucosal T_reg.

For the gut mucosal tissue of chronically HIV-infected individuals, an increase in the percentage of T_reg within CD4⁺ T cells has been described, although absolute numbers of mucosal CD4⁺ T cells were decreased (15). The reason for this dichotomy remained unclear. One possible explanation might be that mucosal T_reg are selectively spared from infection-mediated depletion. In order to assess the in vivo level of SIV infection in mucosal T_reg and non-T_reg, we quantified cell-associated SIV DNA and SIV RNA copy numbers in both mucosal CD4⁺ T-cell subsets, which were isolated in parallel from each of three chronically SIV-infected animals (Fig. 1). In one animal, viral sequences could not be detected in the mucosal T_reg subset due to poor cell counts and the subsequently low recovery rate of viral DNA and RNA. Therefore, the mucosal T_reg subset of this animal was not included in the analysis. Mucosal T_reg contained SIV DNA copies that were 11.7-fold higher than those of non-T_reg (Fig. 2). These cell-associated SIV DNA copies may reflect integrated proviruses and/or unintegrated reverse transcripts, which both are a record of infection events. Interestingly, despite being infected at a higher level, the mucosal T_reg subset contained 4.5-fold fewer SIV RNA copies than the non-T_reg subset (Fig. 2). These findings indicate that SIV RNA production is less efficient in mucosal T_reg than in their nonregulatory counterpart, either by insufficient integration of SIV DNA or transcription of proviral DNA.

Mucosal T_reg are less affected by SIV-related cell death than mucosal non-T_reg.

The observed differences in the efficiency levels of SIV RNA production suggest that virus replication is attenuated in T_reg compared to non-T_reg. As a consequence, mucosal T_reg and non-T_reg could be differentially affected by SIV-mediated cell death, either via virus-induced cytopathic effects or immune response-mediated cytolysis. To determine the level of SIV-related cell death-inducing pathways, mucosal T_reg and non-T_reg from a chronically SIV-infected animal and a healthy SIV-uninfected animal were compared for cell death-related gene expression by microarray experiments. Flow cytometric analysis of the mucosal cell subsets isolated from the SIV-infected animal is shown in Fig. 1. The purity was comparable to that of T_reg and non-T_reg subsets isolated from the SIV-uninfected animal. Normalized signal intensities for single genes were analyzed to identify mRNA expression levels of different genes in mucosal T_reg and non-T_reg. Genes whose normalized signal intensity values were less than 5 in both subsets of one animal were excluded from the analysis. In the SIV-uninfected animal, transcript expression levels of proapoptotic/cell death-related genes did not differ between mucosal T_reg and non-T_reg (Fig. 3A). This was different with the chronically SIV-infected animal, for which the transcript expression levels of genes with cell death-related function were significantly higher in the mucosal non-T_reg subset than in the T_reg subset (P = 0.0134) (Fig. 3B). These included genes previously demonstrated to be directly associated with HIV production, such as CDK4, GADD45G, and PUMA (3, 21, 33, 38, 42). Transcription of anti-apoptotic genes was found rarely with both animals (Fig. 3A and B). In both the SIV-infected animal and the SIV-uninfected animal, transcript levels of genes related to control and regulation of proliferation were comparable in mucosal T_reg and non-T_reg (data not shown). This provides additional support that the high level of cell death-related gene expression found in the non-T_reg subset of the SIV-infected animal was mediated by direct viral infection and/or immune defense mechanisms rather than increased homeostatic or activation-induced proliferation of mucosal non-T_reg over T_reg. Taken together, these results suggest that SIV-related cell death-inducing mechanisms occurred preferentially in the mucosal non-T_reg subset.

Longitudinal study of mucosal CD4⁺ T cells.

Longitudinal analysis of alterations in the mucosal CD4⁺ T-cell pool was performed with five rhesus macaques during the course of SIV infection (Table 1, group 2). Acute HIV/SIV infection is characterized by high viral replication and massive loss of CD4⁺ T cells, which is most pronounced in the gastrointestinal mucosa and persists during chronic infection (11, 35, 46). In the colonic mucosa of all animals, high viral replication occurred within the first 4 weeks of infection, as determined by comparing the cell-associated copy numbers of SIV RNA and SIV DNA (Fig. 4). Consequently, and consistent with previous observations (12), we refer to the first 4 weeks of infection as the acute phase and to the following weeks as the chronic phase of SIV infection.

SIV infection is associated with increased proliferation of mucosal CD4⁺ T cells, including T_reg.

Given that mucosal T_reg are spared from SIV-mediated cell death, proliferation of the mucosal CD4⁺ T-cell pool may, in the course of SIV infection, favor the accumulation of T_reg. In order to assess this possibility, mucosal CD4⁺ T cells in colon biopsy specimens taken longitudinally during SIV infection were phenotypically analyzed in situ for coexpression of the proliferation marker Ki67. Immunohistochemical double staining revealed less than 10% of mucosal CD4⁺ T cells expressing Ki67 before infection (Fig. 5A and B). Ki67 expression of CD4⁺ T cells increased to 40.0% ± 11.9% by week 2 and to 74.5 ± 5.4% by week 3 after infection (Fig. 5B). This high number of CD4⁺ cells expressing Ki67 in mucosal CD4⁺ T cells was maintained during the follow-up, indicating a remarkably increased proliferation of mucosal CD4⁺ T cells in chronic SIV infection (Fig. 5A and B). To examine whether T_reg proliferate within the mucosal CD4⁺ T-cell pool, we analyzed coexpression of FOXP3 and Ki67 in mucosal cells by immunofluorescence staining. We found coexpression of FOXP3 and Ki67 in the chronic phase of SIV infection, indicating proliferation of T_reg within the mucosal CD4⁺ T-cell pool (Fig. 5C). Thus, in addition to preferential SIV-mediated cell death in non-T_reg, local expansion may account for increased mucosal T_reg in chronic SIV infection.

Increase of mucosal T_reg is most pronounced after SIV-induced depletion of mucosal CD4⁺ T cells.

To determine the dynamics of mucosal T_reg during the course of SIV infection, we immunohistochemically examined the changes in the absolute frequency of mucosal T_reg in sequential biopsy specimens. In addition, we analyzed the changes of T_reg relative to the number of CD4⁺ cells (relative frequency of mucosal T_reg). As expected, the depletion of mucosal CD4⁺ T cells was almost complete within 2 weeks after infection, and mucosal CD4⁺ T cells remained low throughout the observation period (Fig. 6A). Mucosal T_reg numbers remained unchanged at week 1 and declined below preinfection levels by week 2 postinfection (Fig. 6A). Relative frequencies of mucosal T_reg increased compared to preinfection levels (P = 0.0002) (Fig. 6B), suggesting a preferential loss of non-T_reg during the first 2 weeks of infection, which occurs probably due to SIV-mediated cell death. Thereafter, the relative frequencies of mucosal T_reg increased remarkably to 44.9 ± 14.4% at week 7 after infection compared with only 4.0 ± 2.2% at preinfection (P = 0.0136) and remained high during the follow-up (Fig. 6B). During the course of infection, loss of mucosal CD4⁺ T cells correlated with a relative increase of mucosal T_reg, indicating that mucosal T_reg were spared from SIV-mediated depletion (P = 0.0018) (Fig. 7A). Consistent with this, we found a highly significant correlation between the number of CD4⁺ T cells and the number of mucosal non-T_reg (P < 0.0001) but not of mucosal T_reg (7B and C). These observations support our aforementioned findings that T_reg are less susceptible to SIV-related cell death than non-T_reg.

After week 2 postinfection, mucosal T_reg increased also in absolute numbers (Fig. 6A). Therefore, the high relative frequency of mucosal T_reg in the chronic phase not only was attributed to a preferential loss of non-T_reg but may also have resulted from an additional local expansion of T_reg within the colonic mucosa. Consistent with this, an increase of T_reg numbers after week 2 postinfection coincided with high proliferation of the mucosal CD4⁺ T-cell pool, which included T_reg (Fig. 5).

Increase of mucosal T_reg correlates with a reduction of perforin-expressing cells in colonic mucosa.

The accumulation of T_reg in colonic tissue might impair mucosal immune responses. In order to assess the potential contribution of mucosal T_reg to the counterregulation of the cytotoxic immune response, we examined in situ the expression of the cytotoxic molecule perforin after SIV infection and compared the numbers of perforin⁺ cells to the frequencies of mucosal T_reg. Consistent with previous findings (43), perforin expression in the colonic tissue of all five animals increased during acute SIV infection and returned to preinfection levels afterward (data not shown). There was a significant inverse correlation between the relative frequency of T_reg and the number of perforin-expressing cells (Fig. 8), which is consistent with the suggestion that mucosal T_reg inhibit cytotoxic immune responses in chronic HIV/SIV infection.