HIV RNA is detected in the genital tract of 40%–80% of HIV-infected women with detectable plasma viral loads.1–4 Several investigators have also shown that shedding of HIV in the genital tract can occur in 20%–30% of women with low or undetectable plasma HIV RNA.5,6 Our previous studies and those of others have demonstrated that HIV genital tract shedding is most closely related to higher plasma HIV RNA levels1 but is also associated with the presence of sexually transmitted infections (STIs), cervical inflammation,7–9 and the vaginal microbiome.10,11 In the era of highly active antiretroviral therapy (HAART), the likelihood of genital shedding is significantly lower in treated women.1,12,13 Although genital shedding decreased rapidly after antiretroviral (ARV) therapy initiation in one prospective observational study, suppression of genital HIV shedding was not complete.12 We have shown that although potent ARV therapy was associated with a significant decrease in genital tract shedding, 59% of women receiving potent ARV therapy still exhibited genital tract shedding and 16% of these women had shedding even when plasma RNA was <500 (50–500) copies per milliliter.1
Generalized immune activation is the hallmark of HIV infection and is associated with clinical outcome and affects mucosal immunity. Our current study aimed to evaluate the impact of T-cell activation in blood, as measured by the expression of CD38 and DR on CD4+ and CD8+ T cells, on genital tract HIV shedding. We measured HIV viral load in paired blood and cervicovaginal lavage (CVL) samples and blood immune activation markers by flow cytometry in a cohort of HIV-infected women. We hypothesized that systemic immune activation would be associated with an increase in HIV genital shedding.
MATERIALS AND METHODS
Subjects were a subset of the 2059 HIV-infected women enrolled in the Women's Interagency HIV Study (WIHS). Women were enrolled during 1994–1995 or 2001–2002 at 6 sites in the United States. WIHS methods and baseline cohort characteristics have been described previously.14 This subgroup consists of 226 women enrolled in WIHS during 1994–1995 who had at least one quantitation of HIV in paired plasma and genital tract specimens available for evaluation from November 1994 through September 2001.
Women enrolled in WIHS were monitored at 6-month intervals. At baseline (entry into the cohort) and subsequent semiannual visits, standardized interviews collected demographic, clinical, and historical data, physical and pelvic examinations were performed, and clinical and research specimens were obtained, including blood, Papanicolaou (Pap) smear, and genital tract specimens. The gynecological examination consisted of visual inspection and speculum examination for lesions and discharge. CVL was collected after a lavage was performed using 10 mL of sterile nonbacterostatic saline into the vagina and against the cervical os and then aspirating from the posterior vaginal fornix.
Baseline serology for herpes simplex virus 1 (HSV) 1 and HSV-2 was performed with Western blot in one central laboratory (University of Washington). Syphilis testing was done at each local clinical laboratory using the rapid plasma reagin test.
Viral Load Determination
Plasma HIV viral load (plasma viral load) quantitation was performed in laboratories participating in the Division of Acquired Immunodeficiency Syndrome Viral Quality Assurance Program and who were CLIA certified as required by Division of Acquired Immunodeficiency Syndrome. Plasma viral load was measured real time using isothermal nucleic acid sequence–based amplification. Initial tests used NucliSens (Organon Teknika Corp., Durham, NC, with a lower limit of detection [LLD] of 80 copies per milliliter). After 2007, it was measured using COBAS AmpliPrep/COBAS TaqMan HIV-1 Test (Roche Molecular Systems, Branchburg, NJ, with LLD of 48 copies per milliliter). HIV RNA testing of CVL for all visits was performed using the NucliSens assay (LLD of 80 copies per milliliter). HIV genital shedding was defined as HIV RNA >80 copies per milliliter.
Lymphocyte subset analyses were performed real time using standard flow cytometric clones and techniques in laboratories that were certified by the National Institute of Allergy and Infectious Diseases Immunology Quality Assurance Program as previously reported.15 Women were evaluated for expression of activation of CD4+ and CD8+ T cells in either fresh whole blood collected in EDTA tubes using 3-color flow cytometry or frozen peripheral blood mononuclear cells using 3- or 4-color flow cytometry as reported elsewhere.16–18 The following fluorochrome-conjugated antibodies were used: anti-CD3, -CD4, -CD8, -HLA-DR, and -CD38 (Becton Dickinson Immunophenotyping Kit II; Becton Dickinson, San Jose, CA). Flow cytometric analysis was performed using a FACSCalibur flow cytometer using CELLQuest software (Becton Dickinson).
Genital Tract Evaluations
CVL specimens were collected for saline and potassium hydroxide preparation for microscopy, pH, candida culture, Gram stain for bacterial vaginosis (BV), swabs for gonorrhea/chlamydia nucleic acid detection test (baseline only), trichomonas by wet mount or culture, and Pap smear. Swabs for HSV culture and syphilis screening were obtained if ulcer/vesicle/fissure was present. Inflammation (leukocytes on Pap smear) and inflammation-associated cellular changes (cellular changes found with inflammation), including basophilic cytoplasm, enlarged unevenly sized nuclei, enlarged, irregular, or multiple nucleoli, were ascertained from Pap smear. Pap smears were read in a central laboratory (Kyto Diagnostics, New York, NY). Candidiasis was diagnosed by the presence of pseudohyphae or positive culture. Gram stains were interpreted for BV in a central laboratory (University of Washington) using the Nugent score criteria, with categorization as normal (0–3), intermediate (4–6), or consistent (7–10).19 Trichomoniasis was diagnosed by visualization of motile trichomonads on wet mount or with positive culture. Human papillomavirus (HPV) was identified by polymerase chain reaction in 2 central laboratories.20
Women in the study were classified as “shedders” if they had at least one HIV shedding visit (CVL viral load >80 copies per milliliter) and as “nonshedders” if they never shed HIV. Demographic and clinical characteristics at study enrollment (baseline) were compared between HIV shedders and nonshedders by χ2 tests. Characteristics evaluated at baseline and subsequent visits included age at visit, race/ethnicity (White, African American, Hispanic, or other), ARV therapy used since the last visit (no therapy, monotherapy, combination, or HAART), plasma viral load level (≤80, 81–1000, 1001–10,000, 10,001–100,000, >100,000 copies per milliliter), CD4 cell count (≤200, 201–350, 351–500, >500 cells per cubic millimeter), number of lifetime male sex partners, BV Gram stain score (normal: 0–3, intermediate: 4–6, or consistent: 7–10), candidiasis, chlamydia, gonorrhea, HPV, HSV, syphilis, trichomonas, cervical lesions, vaginal pH (<4.5, 4.5–5.4, ≥5.5), inflammation, and inflammation-associated cellular change.
To assess the relationship between systemic T-cell activation and genital infections, generalized estimating equation models with identity link function and an exchangeable correlation matrix were used to account for multiple visits per subject. Percentages of CD4+ and CD8+ T cells in blood that were DR+CD38+ or DR−CD38−were treated as the continuous dependent variables. Adjusting for plasma viral load, the analyses were performed separately for each of the variables used to indicate genital infection in a previous study.21 Plasma viral load was treated as a continuous independent variable and was log10 transformed to reduce positive skew. Chlamydia, gonorrhea, syphilis, HSV-1, and HSV-2 were measured at baseline and were modeled as fixed variables. HPV, candidiasis, Trichomonas vaginalis, BV Gram stain score, inflammation, inflammation-associated cellular changes, cervical lesions, and vaginal pH, which were obtained at each visit, were modeled as time-dependent variables.
The relationship between systemic T-cell activation and HIV shedding in the genital tract was assessed using generalized estimating equations with logit link function and an exchangeable correlation matrix. The dependent variable was whether HIV shedding was present in the genital tract at a given visit. The immune markers of interest, CD38+DR+ and CD38−DR− on both CD4 and CD8 T cells, were categorized by tertiles. The longitudinal associations were first evaluated univariately (not adjusting for other covariates) and then were adjusted for HIV plasma viral load (log10 transformed). The associations were also assessed by adjusting for HIV viral load and markers of genital infections that were found significantly associated with immune activation (P < 0.05). Vaginal pH was not included in these adjustments as it may result in overadjustment as a marker for genital infections. Associations were presented as odds ratios (OR) with 95% confidence intervals (CIs). STATA software (version 11.2) was used for all statistical analyses.
The 226 HIV-positive women in this study had 569 genital evaluations with a median of 2.5 visits per women (range: 1–6 visits); women exhibited HIV genital shedding on 159 (28%) of these visits. There were few differences in demographic, virologic, behavioral, and clinical characteristics at WIHS enrollment between the nonshedders (50.9%, n = 115) and shedders (49.1%, n = 111) (Table 1). The HIV shedders tended to be younger, with a median age of 34 vs. 37 years for nonshedders (P = 0.03). Among the HIV shedders, African Americans accounted for 58.6%, Hispanics 20.7%, and Whites 19.8%. At baseline, only 5% of all women had plasma viral load <80 copies per milliliter. The shedders tended to have higher viral loads at baseline (median: 17,500 copies per milliliter, interquartile range: 4400–76,000) vs. nonshedders (median: 15,051 copies per milliliter, interquartile range: 1400–53,000), although this was not statistically significant (P = 0.07). The median CD4 cell count was similar in both groups, with approximately one-third in each group having CD4 cell counts >500 cells per cubic millimeter. Approximately half of the women were not receiving any ARV therapy at baseline (50% vs. 40%, nonshedders vs. shedders), and only one woman in the shedder cohort was on HAART at enrollment (0.9%). The median vaginal pH was 5 for both nonshedders and shedders. Very few women had evidence of active HSV infection by culture (0.9% in both groups, data not shown), chlamydia (0% vs. 0.9%, nonshedders vs. shedders), and gonorrhea (0% in both groups) at enrollment. Shedders were more likely to have cervical inflammation (leukocytes on Pap smear) than nonshedders (20.6% vs. 8.9%, respectively, P = 0.02). Other common genital infections, including BV, HPV (all types), and trichomonas, did not differ by group. A large percentage of women in both cohorts had evidence of HSV seropositivity: HSV-1 (79% vs. 80%, nonshedders vs. shedders) and HSV-2 (76% vs. 80%, nonshedders vs. shedders). Few women in either cohort had cervical lesions, candidiasis, syphilis, or inflammation-associated cellular changes.
Immune Activation Markers and Genital Infections
Each of the immune markers was analyzed for their association with cervical inflammation and STIs (Table 2). Adjusting for plasma viral load, CD4+CD38+DR+ was associated with HSV-1 seropositivity (β = 3.18, standard error [SE] = 1.52, P = 0.04) and HPV infection (β = 2.2, SE = 1.04, P = 0.04). The presence of T. vaginalis was associated with lower levels of resting CD8+ (CD38−DR−) T cells (β = −2.19, SE = 0.72, P = 0.002), and candidiasis was associated with lower levels of resting CD4+ (CD38−DR−) T cells (β = −2.8, SE = 0.97, P = 0.004). Of the other markers of cervical inflammation, only vaginal pH of 5.5+ (β = 2.58, SE = 1.04, P = 0.01) was associated with higher levels of resting CD4+ (CD38−DR−) T cells.
Univariate and multivariate models analyzing the association between the immune markers and genital shedding are shown in Table 3. In the univariate model, high levels of T-cell activation (CD38+DR+) were significantly associated with genital shedding for both CD8+ T cells (OR = 3.52, 95% CI: 2.05 to 6.03, P < 0.001 for highest vs. lowest tertile) and CD4+ T cells (OR = 4.31, 95% CI: 2.35 to 7.91, P < 0.001 for highest vs. lowest tertile). Conversely, more resting cells (CD38−DR−) were associated with less shedding for both CD4+ and CD8+ T cells (P < 0.001 for both), and these associations remained significant after adjusting for plasma viral load. However, after adjusting for plasma viral load and those significant genital cofactors measured at every visit, including candidiasis, HPV, and trichimoniasis, only higher levels of resting CD8+ (CD38−DR−) T cells (OR = 0.44, 95% CI: 0.21 to 0.9, P = 0.02 for highest vs. lowest tertile) were significantly inversely associated with the presence of HIV shedding in the genital tract.
This is one of the largest prospective studies to evaluate the association between systemic T-cell activation and genital HIV shedding. This study has several notable findings. First, in our unadjusted model, systemic T-cell activation was significantly associated with genital tract shedding as supported by other studies.22 However, when controlling for HIV RNA alone and with STIs/genital tract infections, the association of systemic immune activation and genital shedding was lost. Second, although we did not see an association with specific genital infections and HIV shedding at baseline, cervical inflammation was more common in the HIV shedder cohort. Third, when analyzing the association of STI, genital tract infections, cervical inflammation, and immune activation, we found that only HSV-1 seropositivity and HPV infection were associated with CD4 activation, but not CD8 activation. Our findings are consistent with other studies, which have shown that subclinical or asymptomatic herpes infection and coinfection with HPV are associated with increased plasma and genital HIV viral loads in addition to systemic and/or local immune activation.20 Finally, and perhaps most importantly, when adjusted for plasma viral load, we found that women with higher percentages of nonactivated resting CD4 and CD8 T cells (CD38−DR−) were significantly less likely to have genital tract shedding. This held true for CD8 resting cells in our fully adjusted model, which included significant genital tract infections.
Activated T cells are necessary for HIV replication and are more likely to have HIV reservoirs and replication-competent virus than resting T cells. Activation promotes virus entry, provirus integration into host cell DNA, and viral RNA transcription23–25 and predicts adverse prognosis for HIV-infected patients.26–28 Our findings support results of previous studies suggesting that systemic immune activation levels may reflect local activation levels, which may promote genital tract HIV replication.22,25,29 In one small study, investigators showed that the presence of immune activation in the cervix was independently associated with genital HIV shedding and that cervical immune activation can be predicted by immune activation in the blood.22 Although we did not specifically look at activation of T cells in the genital mucosa, we did find that plasma activation levels in the unadjusted models and levels of resting T cells in the fully adjusted models were associated with shedding.
These data suggest that although levels of systemic activation and HIV replication are reflected in the genital tract, local factors, such as STIs, cervical inflammation, and infection, may contribute to local activation of T cells and replication of HIV in the genital tract. However, when there are more resting CD8 T cells circulating, local genital tract HIV replication may be diminished, which underscores the importance of achieving maximal viral suppression with ARV therapy with the aim of achieving higher levels of resting T cells.
Our study has several limitations. First, we may have underestimated the percentage of activated T cells as we defined activation as coexpression of both CD38 and HLA-DR and did not include T cells that express CD38 or HLA-DR alone. Second, our study was not designed to evaluate the impact of local genital immune activation on systemic activation. In addition, we did neither sample the genital tract at the same time during the menstrual cycle nor evaluate the levels of local inflammatory markers and cytokines as done in other studies.22,30 However, it is possible that similar to gut-associated T-cell depletion and immune activation,31–33 HIV replication in the genital tract results in depletion of mucosal T cells and translocation of bacterial products resulting in systemic immune activation. In fact, Bull et al30 demonstrated that an increase in local inflammatory cytokines/chemokines and genital tract HIV shedding was associated with a depletion in CD4 T cells (FOXP3+) in the genital tract although it is unclear what effect this had on systemic immune activation. Alternatively, it is possible that cells that are activated before circulating through the female genital tract may increase local activation through secretion of cytokines and thereby HIV replication locally. In addition, protective factors within the vaginal microbiome, such as H2O2-producing lactobacillus, may inhibit viral replication and local immune activation34 and thus systemic immune activation.
In conclusion, our study suggests that preventing or controlling systemic immune activation by suppressing HIV viral replication and/or genital inflammation and treating coinfections may decrease the risk of genital immune activation and local HIV replication and therefore decrease the risk of HIV transmission to partners and offspring. Having higher percentages of resting T cells, which is seen with maximal viral suppression with treatment, may also decrease the likelihood of mucosal replication and, in turn, transmission. Further studies to better characterize the associations and causality are warranted.
Data in this article were collected by the Women's Interagency HIV Study (WIHS). The contents of this publication are solely the responsibility of the authors and do not represent the official views of the National Institutes of Health. WIHS (Principal Investigators): UAB‐MS WIHS (Michael Saag, Mirjam‐Colette Kempf, and Deborah Konkle‐Parker), U01‐AI‐103401; Atlanta WIHS (Ighovwerha Ofotokun and Gina Wingood), U01‐AI‐103408; Bronx WIHS (K. Anastos), U01‐AI‐035004; Brooklyn WIHS (H. Minkoff and Deborah Gustafson), U01‐AI‐031834; Chicago WIHS (M. Cohen), U01‐AI‐034993; Metropolitan Washington WIHS (M. Young), U01‐AI‐034994; Miami WIHS (Margaret Fischl and Lisa Metsch), U01‐AI‐103397; UNC WIHS (Adaora Adimora), U01‐AI‐103390; Connie Wofsy Women's HIV Study, Northern California (Ruth Greenblatt, Bradley Aouizerat, and Phyllis Tien), U01‐AI‐034989; WIHS Data Management and Analysis Center (Stephen Gange and Elizabeth Golub), U01‐AI‐042590; Southern California WIHS (Alexandra Levine and Marek Nowicki), U01‐HD‐032632 (WIHS I–WIHS IV).
1. Kovacs A, Wasserman SS, Burns D, et al.. Determinants of HIV-1 shedding in the genital tract of women. Lancet. 2001;358:1593–1601.
2. Khopkar P, Sane S, Bichare S, et al.. Association of plasma viremia with HIV-1 RNA levels in cervicovaginal lavage secretions in HIV-1 seropositive ART naïve women from Pune, India. J Clin Virol. 2013;58:730–732.
3. Natividad-Villanueva GU, Santiago GUE, Manalastas RM Jr, et al.. Human immunodeficiency virus in plasma and cervicovaginal secretions in Filipino women. Int J STD AIDS. 2003;14:826–829.
4. Uvin SC, Caliendo AM. Cervicovaginal human immunodeficiency virus secretion and plasma viral load in human immunodeficiency virus-seropositive women. Obstet Gynecol. 1997;90:739–743.
5. Spinillo A, et al.. Factors associated with nucleic acids related to human immunodeficiency virus type 1 in cervico-vaginal secretions. BJOG. 2001;108:634–641.
6. Debiaggi M, Zara F, Spinillo A, et al.. Viral excretion in cervicovaginal secretions of HIV-1-infected women receiving antiretroviral therapy. Eur J Clin Microbiol Infect Dis. 2001;20:91–96.
7. Ghys PD, Fransen K, Diallo MO, et al.. The associations between cervicovaginal HIV shedding, sexually transmitted diseases and immunosuppression in female sex workers in Abidjan, Côte d'Ivoire. AIDS. 1997;11:F85–F93.
8. Kreiss J, Willerford DM, Hensel M, et al.. Association between cervical inflammation and cervical shedding of human immunodeficiency virus DNA. J Infect Dis. 1994;170:1597–1601.
9. Kovacs A, Chan LS, Chen ZC, et al.. HIV-1 RNA in plasma and genital tract secretions in women infected with HIV-1. J Acquir Immune Defic Syndr. 1999;22:124–131.
10. Borgdorff H, Tsivtsivadze E, Verhelst R, et al.. Lactobacillus-dominated cervicovaginal microbiota associated with reduced HIV/STI prevalence and genital HIV viral load in African women. ISME J. 2014;8:1781–1793.
11. Cone RA. Vaginal microbiota and sexually transmitted infections that may influence transmission of cell-associated HIV. J Infect Dis. 2014;210(suppl 3):S616–S621.
12. Graham SM, Holte SE, Peshu NM, et al.. Initiation of antiretroviral therapy leads to a rapid decline in cervical and vaginal HIV-1 shedding. AIDS. 2007;21:501–507.
13. Fiore JR, Suligoi B, Saracino A, et al.. Correlates of HIV-1 shedding in cervicovaginal secretions and effects of antiretroviral therapies. AIDS. 2003;17:2169–2176.
14. Barkan SE, Suligoi B, Saracino A, et al.. The Women's interagency HIV study. WIHS collaborative study group. Epidemiology. 1998;9:117–125.
15. Nicholson J, Kidd P, Mandy F, et al.. Three-color supplement to the NIAID DAIDS guideline for flow cytometric immunophenotyping. Cytometry. 1996;26:227–230.
16. Landay A, Benning L, Bremer J, et al.. Correlates of immune activation marker changes in human immunodeficiency virus (HIV)-seropositive and high-risk HIV-seronegative women who use illicit drugs. J Infect Dis. 2003;188:209–218.
17. Al-Harthi L, Voris J, Du W, et al.. Evaluating the impact of hepatitis C virus (HCV) on highly active antiretroviral therapy-mediated immune responses in HCV/HIV-coinfected women: role of HCV on expression of primed/memory T cells. J Infect Dis. 2006;193:1202–1210.
18. Kovacs A, Al-Harthi L, Christensen S, et al.. CD8(+) T cell activation in women coinfected with human immunodeficiency virus type 1 and hepatitis C virus. J Infect Dis. 2008;197:1402–1407.
19. Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation. J Clin Microbiol. 1991;29:297–301.
20. Palefsky JM, Minkoff H, Kalish LA, et al.. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst. 1999;91:226–236.
21. Homans J, Christensen S, Stiller T, et al.. Permissive and protective factors associated with presence, level, and longitudinal pattern of cervicovaginal HIV shedding. J Acquir Immune Defic Syndr. 2012;60:99–110.
22. Jaspan HB, Liebenberg L, Hanekom W, et al.. Immune activation in the female genital tract during HIV infection predicts mucosal CD4 depletion and HIV shedding. J Infect Dis. 2011;204:1550–1556.
23. Wahl SM, Orenstein JM. Immune stimulation and HIV-1 viral replication. J Leukoc Biol. 1997;62:67–71.
24. Copeland KF, Heeney JL. T helper cell activation and human retroviral pathogenesis. Microbiol Rev. 1996;60:722–742.
25. Appay V, Sauce D. Immune activation and inflammation in HIV-1 infection: causes and consequences. J Pathol. 2008;214:231–241.
26. Giorgi JV, Liu Z, Hultin LE, et al.. Elevated levels of CD38+ CD8+ T cells in HIV infection add to the prognostic value of low CD4+ T cell levels: results of 6 years of follow-up. The Los Angeles Center, Multicenter AIDS Cohort Study. J Acquir Immune Defic Syndr. 1993;6:904–912.
27. Giorgi JV, Hultin LE, McKeating JA, et al.. Shorter survival in advanced human immunodeficiency virus type 1 infection is more closely associated with T lymphocyte activation than with plasma virus burden or virus chemokine coreceptor usage. J Infect Dis. 1999;179:859–870.
28. Benito JM, Zabay JM, Gil J, et al.. Quantitative alterations of the functionally distinct subsets of CD4 and CD8 T lymphocytes in asymptomatic HIV infection: changes in the expression of CD45RO, CD45RA, CD11b, CD38, HLA-DR, and CD25 antigens. J Acquir Immune Defic Syndr Hum Retrovirol. 1997;14:128–135.
29. Zara F, Nappi RE, Brerra R, et al.. Markers of local immunity in cervico-vaginal secretions of HIV infected women: implications for HIV shedding. Sex Transm Infect. 2004;80:108–112.
30. Bull ME, Legard J, Tapia K, et al.. HIV-1 shedding from the female genital tract is associated with increased Th1 cytokines/chemokines that maintain tissue homeostasis and proportions of CD8+FOXP3+ T cells. J Acquir Immune Defic Syndr. 2014;67:357–364.
31. Brenchley JM, Price DA, Schacker TW, et al.. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med. 2006;12:1365–1371.
32. Brenchley JM, Douek DC. HIV infection and the gastrointestinal immune system. Mucosal Immunol. 2008;1:23–30.
33. Dandekar S. Pathogenesis of HIV in the gastrointestinal tract. Curr HIV/AIDS Rep. 2007;4:10–15.
34. Klebanoff SJ, Coombs RW. Viricidal effect of Lactobacillus acidophilus on human immunodeficiency virus type 1: possible role in heterosexual transmission. J Exp Med. 1991;174:289–292.