Gut CD4 T cells are massively infected during primary HIV infection1 and then extensively depleted in lamina propria and gut-associated lymphoid tissue.2 Most studies of CD4 T-cell HIV/SIV infection under highly active antiretroviral therapy (HAART) have assessed these infection levels in blood, reporting the persistence of a long-lived cellular HIV reservoir after long-term treatment.3,4 Particularly, central and transitional memory CD4 T cells (TCM and TTM) constitute the largest HIV reservoir persisting over time, mostly due to the homeostatic proliferation of TTM and the long half-life of TCM.3 Viral persistence under HAART might also be potentiated by ongoing viral replication, as evidenced by elevated levels of HIV DNA in short-lived activated CD4 T cells in this context.4,5
Whether these mechanisms of viral persistence in blood under HAART are relevant in the gut remains unknown. Overall, studies using gut biopsies to determine the size of the HIV reservoir under treatment report that infection levels were highly correlated to those in peripheral blood mononuclear cell (PBMCs).3,6 Two studies reported that gut CD4 T cells are infected at a rate 5 times that of their circulating counterparts.5,7 Nonetheless, none of these studies directly assessed gut CD4 T-cell-associated HIV DNA, whereas other gut-resident cells might also host the virus.
To clarify the contribution of infected gut CD4 T cells to the total cell-associated HIV (cell HIV DNA) viral load in a major site of residence of HIV reservoir cells,5,7 we have directly investigated the infection rates of purified cell-sorted rectal (R-) and peripheral blood (PB-) memory CD4 T cells in patients under long-term HAART. We clearly show for the first time that infection levels of R-memory CD4 T cells and of their blood counterparts did not differ. Moreover, lower frequencies of these CD4 T cells in rectum might underline contribution of other gut resident cells to cellular HIV reservoir in this tissue. These results contribute to a better understanding of the HIV-1 reservoir dynamic in gut, which is still a major hurdle to eradication.
MATERIALS AND METHODS
Patients and Tissue Collection
This transversal prospective single-center study included 11 patients treated with HAART over a median of 11 years and with <1 HIV-RNA copy/mL when studied (Table 1). All patients consented in writing (form approved by the Agence Française de Sécurité Sanitaire des Produits de Santé, AFSSAPS protocol reference #2008-A01405-50) to blood collection and rectal biopsies. Rectal biopsies were obtained all around the upper part of the rectum by anoscopy; a median of 15 tissue pieces were collected for (1) anatomopathological analyses (see Pathology Assessment), (2) virological and pharmacological investigations, and (3) immunologic and virological investigations (see Rectal Tissue Dissociation).
Tissue samples were fixed in 4% formalin and embedded in paraffin. Tissue sections 4 μm thick were stained with hematoxylin-eosin-saffron. The immunostaining procedure used a commercial kit (Thermo Fisher Scientific, Philadelphia, PA) according to the manufacturer's instructions after the initial steps: antigen retrieval, blocking of nonspecific background staining, and then incubating deparaffinized, formalin-fixed tissue sections 4 μm thick for 1 hour at room temperature with anti-CD4 antibody (clone 4B12, Thermo Fisher Scientific) diluted by 1:20. A digital image of whole-tissue sections from each rectal mucosal biopsy specimen (3 per patient) was captured (magnification: ×20) with a high-resolution digital camera (Axiocam; Zeiss, Le Pecq, France) coupled with a light microscope. The CD4+ lymphocytes on the section were counted manually with Image J 1.34s software (National Institutes of Health, Bethesda, MD). Degree of fibrosis was evaluated semiquantitatively on a 4-point scale (0 = none; 1 = mild; 2 = moderate; and 3 = severe).
Rectal Tissue Dissociation
The rectal biopsy specimens intended for flow cytometry and cell sorting were collected in complete RPMI 1640 medium supplemented with a cocktail of antibiotic and antimycotic agents (Gibco, Life Technologies, Grand Island, NY), containing 10% fetal calf serum (Gibco) and then finely ground before digestion for 30 minutes at 30°C in 400 U/mL of collagenase type IV (Sigma-Aldritch, St. Louis, MO), a process punctuated by a rapid vortex every 10 minutes. The digested specimens were then further dissociated with the plunger of a 1 mL syringe on a 70-μm cell strainer (Becton Dickinson, Franklin Lakes, NJ). Rectal mononuclear cells (RMCs) were stained with antibodies for flow cytometry experiments.
Antibodies and Flow Cytometry
Rectal mononuclear cells and PBMCs were stained with the following antibody combinations: Pacific-Blue-CD3 (UCHT1), Alexa700-CD4 (RPA-T4), Cy7PE-CCR7 (3D12), APC-CD27 (L128), PE-CCR5 (2D7), FITC-CD69 (L78), PE-β7 integrin (FIB504), and HLA-DR (L243) (Becton-Dickinson Pharmingen, San Jose, CA), energy-coupled-dye-CD45RA (2H4), and FITC-CD25 (B1.49.9) (Beckman-Coulter, Brea, CA). Total R-CD4 T cells and total PB-memory CD4 T cells (CD3+CD4+CD45RA−) were sorted on a 4-laser FACSAria flow cytometer using FACSDiva software (Becton-Dickinson Biosciences, San Jose, CA). The CD4 T-cell subsets of the R- and PB-memory CD4 T cells were defined as follows: naive (TN: CD45RA+CCR7+CD27+) and memory cells, including central (TCM: CD45RA−CCR7+CD27+), transitional (TTM: CD45RA−CCR7−CD27+), and effector memory (TEM: CD45RA−CCR7−CD27−) cells, and terminal effector CD4 TEMRA cells (CD45RA+CCR7−CD27−).
Cell-associated HIV-1 DNA and Plasma Viral Load Quantification
Residual plasma viremia was measured as previously described8 with a limit of quantification of 1 copy/mL. Cell HIV DNA was quantified in each patient's undissociated tissue sample (hereafter, biopsy specimen), sorted R-CD4 T cells, PBMCs, and sorted PB-memory CD4 T cells, as previously described.9
Antiretroviral Drug Quantification
Antiretroviral drugs in plasma and rectal tissue were determined with an UPLC-MS/MS method (Acquity UPLC-Acquity TQD) as previously described.10
The median CD4 count at treatment initiation was 179 (2–501) cells/mm3, and 4 patients had AIDS-defining clinical conditions (Table 1). A median of 11 (5–13) years afterward, all had undetectable plasma viral loads (<1 copy/mL) and normal CD4 counts: 790 (499–1202) CD4 cells/mm3.
Alteration of Rectal Tissue after Long-Term treatment
We scored rectal tissue fibrosis by immunohistochemistry of 3 biopsy sections per patient: results for 4 patients showed only mild fibrosis in the lamina propria (data not shown), even though one had AIDS-defining clinical conditions. The median frequency of CD4 T cells was estimated at 1.3 × 10−4 (2.7 × 10−5 to 1.9 × 10–4) CD4 T cells per square micrometer of rectal tissue (data not shown).
Rectal and Peripheral Blood Memory CD4 T Cell Displayed Similar HIV-1 Infection Levels
We first quantified cell HIV-DNA content in total PBMCs and rectal biopsy specimens. Sensitivity of the real-time quantitative polymerase chain reaction9 allowed us to detect cell HIV DNA in 10 of 11 PBMC samples and in 8 of 11 biopsy specimens. Their level of infection were in median: 305 (67–4305) HIV-DNA copies per million cells in PMBCs and 265 (37–2067) HIV-DNA copies per million cells in biopsy specimens (P > 0.05) (Fig. 1A).
However, the proportion of CD4 T cells analyzed by flow cytometry was significantly lower among RMCs (R-CD4 T cells) than PBMCs (PB-CD4 T cells) [respective medians: 1.4% (0.1–5.2) and 19.3% (13.3–49.9) of mononuclear cells, P = 0.0078; Fig. 1B, left panel] and the vast majority of R-CD4 T cells was composed of memory cells [median: 91.1% (85.9–96.9)] (Fig. 1B, right panel), mostly CD45RA−CCR7−CD27+ TTM [median: 35.1% (7–69.5); Fig. 1C]. Moreover, comparison of the percentages of memory CD4 T-cell subsets among RMCs and PBMCs expressing the CCR5 HIV coreceptor and at least one of the CD25, CD69, or HLA-DR activation markers showed that the proportions of both were significantly higher in rectal tissue than in blood (P < 0.0156), except for CD45RA−CCR7−CD27− TEM, which also contained high proportions of CCR5+ cells in blood (Fig. 1D). These results overall indicate that with similar infection levels of PBMCs and RMCs reported herein, isolated CD4 T cells from RMCs may contain much more cell HIV DNA than do PB-CD4 T cells.
To test this hypothesis, we therefore evaluated infection levels of CD4 T cell with similar differentiation profiles in RMCs and PBMCs. We sorted R-CD4 T cells by flow cytometry and compared their infection levels with those of matched PB-memory CD4 T cells. Strikingly, infection levels did not differ between R-CD4 T cells [median: 4000 (600–22,000) cell HIV-DNA copies per million cells] and their matched PB-memory CD4 T cells [5641 (2078–18,042), P > 0.05] (Fig. 1E, right panel). These results were obtained on 5 of 11 patients, whereas for the 6 remaining patients, numbers of R-CD4 T cells collected were too small to allow retrieval of sufficient amount of genomic DNA for cell HIV-DNA quantification.
Antiretroviral Drug Concentrations in the Rectum and Blood
As an efficient antiretroviral delivery in rectal compartment might contribute to stabilize the viral reservoir turnover, we measured antiretroviral drug biodisponibility in blood and rectal compartments of the 5 aforementioned patients. Our results indicate that concentrations of non–nucleosidic reverse transcriptase inhibitor predominated in both compartments compared with protease inhibitor and nucleosidic/nucleotidic reverse transcriptase inhibitor. Therefore, pharmacokinetic exposure in the rectum seemed to be more efficient for non–nucleosidic reverse transcriptase inhibitor than for nucleosidic/nucleotidic reverse transcriptase inhibitor or protease inhibitors and, most importantly, resemble with the one observed in blood (Fig. 1F).
This first direct assessment of the infection level of sorted R- and PB-memory CD4 T cells in long-term treated patients revealed that both CD4 T cell subsets host similar amounts of HIV DNA. These data imply that the extensive contribution of the gut to HIV reservoirs5,7 under HAART is not exclusively supported by extremely high levels of R-CD4 T-cell infection.
The predominance of highly activated and CCR5-expressing rectal memory CD4 T cells suggested that CD4 T cells in the gut might be highly susceptible to HIV infection as proposed by other groups.5,7 Assuming that the vast majority of cell HIV DNA was in CD4 T cells, gut CD4 T cell infection frequencies have been previously calculated,7 combining the infection level of gut mononuclear cells and the proportion of gut CD4 T cells.7 In our study, such an extrapolation would have overestimated infection by log10 compared with the actual measurements.
Although infected memory CD4 T cells are major contributors to the HIV reservoir in blood compared with naïve CD4 T cells (data not shown and Refs. 3,11), our results suggest that in gut compartment, the involvement of infected non-CD4 T cells to the overall HIV-DNA load should be considered. Among others, cells from the myeloid lineage such as macrophages play a debated role in the HIV persistence.12 Our study reinforced the necessity to clarify the contribution of macrophages to viral reservoir, particularly, in HAART-treated patients.
Several mechanisms may account for this unexpectedly modest level of R-CD4 memory T-cell infection. First, one can speculate that despite treatment, a high cell death rate in this compartment would help clear infected R-CD4 T cells from gut lymphoid and non-lymphoid tissues and lower their apparent infection levels. Indeed, extensive T-cell depletion2 by viral-induced cytopathic effects,13 tissular fibrosis that deprives cells of homeostatic factors,14 and cytotoxic antiviral immunity,15,16 all persist in the gut despite treatment. This CD4 T-cell clearance occurs at even higher frequencies during primary HIV infection, when, however, gut CD4 T cells seem to be infected at a higher rate than total PB-CD4 T cells.1 This suggests that additional antiretroviral treatment-related mechanisms may explain the equivalent infection levels we found in these 2 CD4 T-cell subsets after long-term HAART. Therefore, a second hypothesis, still controversial, is that long-term treatment that is started early is associated with effective reconstitution of the sigmoid colon CD4 T-cell compartment16 and would promote reseeding of the ileum by mostly uninfected ß7hi PB-TCM cells.3,16,17 Their migration would lower the apparent infection level of R-CD4 T cells. Finally, a third possibility is that the similarity of the drug bioavailability profiles found in blood and rectal tissue means treatment would efficiently affect viral replication and cell HIV DNA in both compartments. In accordance, Yukl et al7 reported similar replication rates, as evidenced by cellular usRNA quantification, in both compartments. More extensive studies would be necessary to clarify the mechanisms of low infection levels of these R-memory CD4 T cells and the infection levels of other gut-resident cells.
Our study thus shows that strong contribution of R-CD4 T cells to the HIV reservoir after long-term HAART cannot be related to elevated infection levels of resident CD4 T cells, at least in the rectal compartment. Elevated activation levels and turnover of CD4 T cells in this tissue must be considered as parameters strongly influencing HIV reservoir homeostasis. Therefore, tissue-specific investigations of viral persistence must be conducted to provide accurate keys for efficient therapeutic strategies.
The authors thank Catherine Blanc and the Inter IFR Flow Cytometry Plateform for their support in flow cytometry cell sorting. They thank patients for their participation to this study, Mélanie Lavenu for her technical assistance, and Jo Ann Cahn for help with manuscript preparation.
1. Mehandru S, Poles MA, Tenner-Racz K, et al.. Mechanisms of gastrointestinal CD4+ T-cell depletion during acute and early human immunodeficiency virus type 1 infection. J Virol. 2007;81:599–612.
2. Brenchley JM, Schacker TW, Ruff LE, et al.. CD4+ T cell depletion during all stages of HIV disease occurs predominantly in the gastrointestinal tract. J Exp Med. 2004;200:749–759.
3. Chomont N, El-Far M, Ancuta P, et al.. HIV reservoir size and persistence are driven by T cell survival and homeostatic proliferation. Nat Med. 2009;15:893–900.
4. Chun TW, Nickle DC, Justement JS, et al.. HIV-infected individuals receiving effective antiviral therapy for extended periods of time continually replenish their viral reservoir. J Clin Invest. 2005;115:3250–3255.
5. Chun TW, Nickle DC, Justement JS, et al.. Persistence of HIV in gut-associated lymphoid tissue despite long-term antiretroviral therapy. J Infect Dis. 2008;197:714–720.
6. Avettand-Fenoel V, Prazuck T, Hocqueloux L, et al.. HIV-DNA in rectal cells is well correlated with HIV-DNA in blood in different groups of patients, including long-term non-progressors. AIDS. 2008;22:1880–1882.
7. Yukl SA, Gianella S, Sinclair E, et al.. Differences in HIV burden and immune activation within the gut of HIV-positive patients receiving suppressive antiretroviral therapy. J Infect Dis. 2010;202:1553–1561.
8. Palmer S, Wiegand AP, Maldarelli F, et al.. New real-time reverse transcriptase-initiated PCR assay with single-copy sensitivity for human immunodeficiency virus type 1 RNA in plasma. J Clin Microbiol 2003;41:4531–4536.
9. Avettand-Fenoel V, Chaix ML, Blanche S, et al.. LTR real-time PCR for HIV-1 DNA quantitation in blood cells for early diagnosis in infants born to seropositive mothers treated in HAART area (ANRS CO 01). J Med Virol. 2009;81:217–223.
10. Jung BH, Rezk NL, Bridges AS, et al.. Simultaneous determination of 17 antiretroviral drugs in human plasma for quantitative analysis with liquid chromatography-tandem mass spectrometry. Biomed Chromatogr. 2007;21:1095–1104.
11. Brenchley JM, Hill BJ, Ambrozak DR, et al.. T-cell subsets that harbor human immunodeficiency virus (HIV) in vivo: implications for HIV pathogenesis. J Virol. 2004;78:1160–1168.
12. Eisele E, Siliciano RF. Redefining the viral reservoirs that prevent HIV-1 eradication. Immunity. 2012;37:377–388.
13. Haase AT. Population biology of HIV-1 infection: viral and CD4+ T cell demographics and dynamics in lymphatic tissues. Annu Rev Immunol. 1999;17:625–656.
14. Zeng M, Smith AJ, Wietgrefe SW, et al.. Cumulative mechanisms of lymphoid tissue fibrosis and T cell depletion in HIV-1 and SIV infections. J Clin Invest. 2011;121:998–1008.
15. Ferre AL, Hunt PW, Critchfield JW, et al.. Mucosal immune responses to HIV-1 in elite controllers: a potential correlate of immune control. Blood. 2009;113:3978–3989.
16. Sheth PM, Chege D, Shin LY, et al.. Immune reconstitution in the sigmoid colon after long-term HIV therapy. Mucosal Immunol. 2008;1:382–388.
17. Verhoeven D, Sankaran S, Silvey M, et al.. Antiviral therapy during primary simian immunodeficiency virus infection fails to prevent acute loss of CD4+ T cells in gut mucosa but enhances their rapid restoration through central memory T cells. J Virol. 2008;82:4016–4027.
DECAMUNE Study Group
Roland Tubiana, Christine Katlama, Michèle Pauchard, Brigitte Autran, Laura Papagno, Cathia Soulié, Vincent Calvez, and François Lecardonnel.
ORVACS Study Group
Clinical: R. Tubiana, A.H. Mohand, and M. Pauchard (Paris, France); B. Berzins (Chicago, IL); L. Ruiz (Badalona, Spain); J. Gatell and J. Joseph (Barcelona, Spain); S. Staszewski (Frankfurt, Germany).
Immunology: O. Pelle' and I. Théodorou (Paris, France); Andrew McMichael and Lucy Dorrell (Oxford, United Kingdom); B. Walker (Boston, MA); M. Plana (Barcelona, Spain); L. Ruiz (Badalona, Spain); and F. Barin (Tours, France).
Virology: V. Calvez and C. Soulié (Paris, France).