Positive Association between HIV RNA and IL-6 in the Genital Tract of Rwandan Women

Abstract

Infections and inflammation in the genital tract can influence HIV expression or HIV susceptibility. The goal of this study was to determine if significant relationships exist between cytokines and HIV in genital tract secretions from 57 HIV-seropositive Rwandan women. Genital tract secretions were obtained by cervicovaginal lavage (CVL). Ten different cytokines in CVL were measured by multiplex cytometric bead arrays. HIV RNA in CVL and plasma were measured by quantitative PCR. In univariate analysis, genital tract HIV RNA was significantly associated with plasma HIV RNA and several of the cytokines, while in multivariate analysis, genital tract HIV RNA was significantly associated only with plasma HIV RNA and IL-6. This association of IL-6 with HIV RNA levels suggests that IL-6 is an indicator for conditions that induce HIV expression and that IL-6 may contribute to induction of HIV expression in the genital tract.

H IV is transmitted predominantly through heterosexual contact and higher levels of HIV in genital secretions are believed to increase the risk of transmission.¹ Strong correlations between levels of HIV in genital secretions and plasma HIV levels have been shown and heterosexual transmission is increased in the setting of higher plasma HIV levels.^2
–4 Inflammation or infections are known to increase HIV levels locally in the genital tract and promote HIV transmission. Thus, genital shedding of HIV has been observed in women who have no detectable plasma virus¹ and women with altered vaginal flora, sexually transmitted infections (STI), or genital tract inflammation have higher levels of genital HIV shedding than women without these conditions.^5
–7 Treatment of STI reduces HIV-1 shedding.^8,9 While these studies suggest that local inflammation caused by genital infections and/or immune responses to infections increases HIV shedding, the specific immune/inflammatory mediators that are associated with genital HIV shedding have not been well characterized. Identification of such mediators could lead to their use as markers of risk or as targets for preventing HIV transmission. To identify such mediators, we analyzed the relationship between levels of HIV RNA and a panel of 10 cytokines and chemokines in genital secretions from 57 Rwandan women.

The Rwanda Women's Inter-association Study and Assessment (RWISA) is an observational prospective cohort study investigating the effectiveness and toxicity of antiretroviral therapy (ART) and comorbidities in HIV-infected Rwandan women. Between May 15 and November 15, 2005, women were enrolled from community-based contacts. Informed consent was obtained, in Kinyarwanda, in accordance with protocols approved by the Rwandan Ethics Committee and the Institutional Review Board of Montefiore Medical Center, Bronx NY. At each study visit historical information, blood, and genital samples were collected.

Vaginal swabs were collected from the women for wet mount and gram stain for Nugent evaluation.¹⁰ Subsequently, cervicovaginal lavage (CVL) was performed by irrigation of the cervix with 10 ml of nonbacteriostatic sterile saline, followed by aspiration from the posterior fornix. CVL was held on ice until processing, (within 3 h of collection). CVL was gently vortexed to evenly distribute cells before aliquoting into 1-ml portions and freezing at −70°C. Upon thawing, 0.4 ml of CVL was used for HIV RNA measurement while the remaining sample was centrifuged for 30 min at 13,000 × g. The pellet was used for DNA isolation for quantification of bacteria,⁵ while the supernatant material was immunoassayed for cytokines and Trichomonas vaginalis.

HIV RNA in plasma was measured by COBAS TaqMan HIV-1 Test or Amplicor HIV-1 Monitor test v1.5 (both from Roche Diagnostic Corporation, Indianapolis, IN). HIV RNA in all CVL samples was quantified using the Amplicor test with a lower limit of detection of 400 copies/ml.

CVL samples clarified by centrifugation were added to Cytometric Bead Array kits (BD Biosciences, San Jose, CA). The Human Inflammation kit [interleukin (IL)-8, IL-1β, IL-6, IL-10, IL-12, and tumor necrosis factor (TNF)-a] and the Human Chemokine kit (IL-8, RANTES, MIG, MCP1, and IP10) were assayed by FACSCaliber (Becton Dickinson, San Jose, CA). Since IL-8 was measured in two different kits, IL-8 levels from only the inflammation kit were used for analysis. T. vaginalis p65 antigen was measured in CVL according to a previously published method¹¹ using a monoclonal capture antibody to p65 (Biodesign International, Saco, ME), a polyclonal rabbit anti-p65 detection antibody (HyTest Ltd., Turku, Finland), and recombinant p65 (HyTest) as a standard. The sensitivity and specificity of this test are reported as 89% and 97%, respectively.¹¹

For statistical analysis, a censored linear regression model was used in the analyses. The cytokine measures, the genital HIV RNA, and the plasma virus load were all log₁₀ transformed in the univariate and multivariate analysis as well as in the correlation analysis.

The median age of the subjects was 35 years (range 25–59) (Table 1). At the time of sample collection, none of the subjects was pregnant, none was menstruating, and four were postmenopausal. The median CD4 cell count was 264/μl (range 52–778), the median plasma HIV RNA was 4.16 log₁₀ copies/ml (range 1.60–6.29), and the median genital tract HIV RNA was 1.98 (range <400–4.54). Five participants were found to be using or had been exposed to antiretroviral drugs at the time that CVL samples were obtained.

Table 1.

Patient Characteristics and Quantitative Genital Organism and Cytokine Levels

	Mean ± SD	Median	Range
Age, years	35.8 ± 6.5	35	25–59
Plasma HIV RNA, log₁₀ copies/ml plasma	3.77 ± 1.48	4.16	1.60–6.29
CVL HIV RNA, log₁₀ copies/ml CVL	1.69 ± 1.76	1.98	0–4.54
CD4 cells/mm³ blood	271 ± 149	264	52–778
IL-12p70, pg/ml CVL	13 ± 6	14	0–24
TNF-α, pg/ml CVL	6 ± 5	5	0–35
IL-10, pg/ml CVL	8 ± 3	8	0–16
IL-6, pg/ml CVL	71 ± 144	14	0–831
IL-1β, pg/ml CVL	352 ± 837	69	0–4713
IL-8, pg/ml CVL	2695 ± 2947	1525	53–10,000^a
IP-10, pg/ml CVL	787 ± 1298	217	0–5000^a
MCP-1, pg/ml CVL	60 ± 112	18	0–547
MIG, pg/ml CVL	503 ± 938	196	0–5000^a
RANTES, pg/ml CVL	108 ± 342	11	0–1900
Nugent score (0–10 scale)	4.9 ± 2.3	5	0—10

Three of the samples had the maximal detectable IL-8 (10,000 pg/ml), three samples had maximal IP-10 (5000 pg/ml), and one sample had maximal MIG (5000 pg/ml).

Wet mount examination for T. vaginalis showed five positive, 51 negative, and one undetermined. ELISA of CVL supernatant for T. vaginalis p65 confirmed three of the positives by wet mount and found an additional five. Wet mount examination for clue cells showed two positive and 55 negative. Nugent scoring of gram stains showed 12 subjects with BV-like flora (score 7–10), 13 with normal flora (score 0–3), 30 with intermediate flora (score 4–6), and two with missing slides.

In analysis of cytokines in CVL samples, IL-8 was present at relatively higher levels than other cytokines (median 1525 pg/ml, Table 1) and was detected in all samples. MIG (median 196) and IP-10 (median 217) were found at intermediate levels and detected in all but three and four samples, respectively. IL-6, IL-1β, MCP-1, IL-12p70, and RANTES were found at somewhat lower levels (medians of 14, 69, 18, 11, and 14, respectively), but were detected in the majority of samples. TNF-a and IL-10 were not detected or detected at very low levels (<12 pg/ml) in all but one sample each.

The association of genital tract HIV RNA levels with each of the variables was assessed. Plasma HIV RNA levels were significantly associated with genital tract HIV RNA levels (p < 0.0001; coefficient 1.26; 95% confidence interval 0.79, 1.74). In univariate analysis, controlling only for plasma virus levels, genital tract HIV RNA levels were significantly negatively associated with CD4 and IL-10 and positively associated with IL-6, IL-1β, IL-8, ICP10, MCP-1, MIG, and RANTES (Table 2). Other variables were not associated with genital tract HIV levels (other cytokines, Nugent score, and T. vaginalis positivity). In a multivariate model, only IL-6 and plasma HIV RNA were significantly associated with genital tract HIV RNA levels.

Table 2.

Significant Univariate and Multivariate Regression Associations with Cervicovaginal Lavage HIV RNA Levels Controlling for Plasma HIV Level ^a

	Univariate model		Multivariate model
	Regression coefficeint (95% CI)	p	Regression coefficient (95% CI)	p
CD4 cells/mm³ blood	−3.7 (−6.6, −0.85)	0.01
IL-10, pg/ml CVL	−2.7 (−5.10, −0.32)	0.03
IL-6, pg/ml CVL	1.04 (0.58, 1.49)	<0.0001	1.04 (0.58, 1.49)	<0.0001
IL-1β, pg/ml CVL	0.71 (0.32, 1.10)	0.0004
IL-8, pg/ml CVL	0.98 (0.50, 1.45)	<0.0001
IP-10, pg/ml CVL	1.64 (0.51, 2.76)	0.004
MCP-1, pg/ml CVL	1.87 (0.50, 3.23)	0.007
MIG, pg/ml CVL	2.00 (0.82, 3.18)	0.0009
RANTES, pg/ml CVL	2.02 (0.95, 3.09)	0.0002

All data were log transformed for analysis. Univariate and multivariate analyses were performed controlling for plasma HIV RNA levels.

Since five of the subjects had potential exposure to ART at the time samples were obtained, which could affect genital tract or plasma HIV levels, the data were reanalyzed after their exclusion. Reanalysis showed again that IL-6 (p < 0.0004; coefficient 0.85; 95% confidence interval 0.38, 1.32) and plasma HIV RNA (p < 0.0001; coefficient 1.71; 95% confidence interval 1.07, 2.45) were significantly associated with genital tract HIV RNA levels.

To determine if IL-6 levels were associated with infections or inflammation, IL-6 levels in Trichomonas-positive and -negative subjects were compared as well as in subjects with Nugent 0-3 versus Nugent 7-10. There was no significant difference in IL-6 between the groups. However, the subjects with absent, rare, or few neutrophils observed on Nugent slides had a median IL-6 of 12 pg/ml while subjects with moderate or many neutrophils had a median IL-6 level of 23 pg/ml (p = 0.059, Mann–Whitney).

In this study we found significant univariate associations between genital tract HIV RNA levels and a panel of cytokines, but a multivariate model indicated that only IL-6 was independently associated with genital tract HIV RNA. The cross-sectional nature of the study does not allow inference of a cause and effect relationship between IL-6 and HIV RNA shedding. However, previous in vitro studies show that proinflammatory cytokines including IL-1, TNF-α, and IL-6 can induce HIV expression in cells.¹³ IL-6 has been shown to induce HIV replication by both transcriptional and post-transcriptional mechanisms and can also induce expression of other proinflammatory mediators.¹⁴ A study of women treated for cervical intraepithelial lesions showed that after treatment, there were increased levels of genital HIV, TNF-α, IL-6, and other activation markers in CVL, while plasma HIV levels were not altered.⁷ A previous study that also assessed the relationship between IL-6 and HIV in vaginal secretions of women found no significant association.¹⁵ However, this relationship could have been obscured by the use of antiretroviral drugs and in fact an association between antiretrovirals and IL-6 was observed. In the current study, only five women used antiretroviral drugs and exclusion of those women from analysis did not change the relationship between IL-6 and HIV in CVL.

A recent study showed that intravaginal inoculation of imiquimod, a TLR7 agonist, increased susceptibility to SIV transmission in macaques. Imiquimod also increased peak SIV plasma loads in infected animals, and increased proinflammatory cytokine expression, including IL-6, in the vaginal mucosa,¹⁶ but genital SIV levels were not measured in that study. Several groups have proposed measuring cytokines in genital samples to predict the inflammatory effect of microbicides. In vitro exposure of epithelial cell monolayers to certain microbicides induces production of IL-6.¹⁷

A limitation of this study was that we did not have data regarding Chlamydia or gonorrhea infection. Such information might have permitted the identification of the infections that contributed to increased levels of cytokines. During menses in HIV-postive women, several cytokines including IL-1β, IL-6, IL-8, and RANTES are reported elevated in vaginal washings.¹⁸ In the current study, none of the CVL samples was collected during menses.

Measuring cytokines in the genital tract has been used as a marker of inflammation and also as a window that may provide insight into pathways that affect HIV expression in the genital tract. The current study suggests that measuring IL-6 is useful in these situations.

Footnotes

Acknowledgments

Financial support was provided by a Rush University Translations Science Consortium Pilot award. RWISA was funded through a supplement to the Bronx/Manhattan Women's Interagency HIV Study funded by the National Institute of Allergy and Infectious Diseases (U01 A1035004-14), and also supported in part by the Albert Einstein College of Medicine Center for AIDS Research funded by the National Institutes of Health (NIH AI-51519) and by the National Institute of Diabetes and Digestive and Kidney Disease (DK54615). Carolyne Gichinga assayed the cytokines by cytometric bead arrays.

References

Kovacs

, Wasserman

, Burns

et al. Determinants of HIV-1 shedding in the genital tract of women. Lancet, 2001; 358:1593–1601.

Wawer

, Gray

, Sewankambo

et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis, 2005; 191:1403–1409.

Quinn

, Wawer

, Sewankambo

et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med, 2000; 342:921–929.

Operskalski

, Stram

, Busch

et al. Role of viral load in heterosexual transmission of human immunodeficiency virus type 1 by blood transfusion recipients. Transfusion Safety Study Group. Am J Epidemiol, 1997; 146:655–661.

Sha

, Zariffard

, Wang

et al. Female genital-tract HIV load correlates inversely with Lactobacillus species but positively with bacterial vaginosis and Mycoplasma hominis. J Infect Dis, 2005; 191:25–32.

Ghys

, Fransen

, Diallo

et al. The associations between cervicovaginal HIV shedding, sexually transmitted diseases and immunosuppression in female sex workers in Abidjan, Cote d'Ivoire. AIDS, 1997; 11:F85–93.

Lawn

, Subbarao

, Wright

et al. Correlation between human immunodeficiency virus type 1 RNA levels in the female genital tract and immune activation associated with ulceration of the cervix. J Infect Dis, 2000; 181:1950–1956.

McClelland

, Wang

, Mandaliya

et al. Treatment of cervicitis is associated with decreased cervical shedding of HIV-1. AIDS, 2001; 15:105–110.

Wang

, McClelland

, Reilly

et al. The effect of treatment of vaginal infections on shedding of human immunodeficiency virus type 1. J Infect Dis, 2001; 183:1017–1022.

10.

Nugent

, Krohn

, Hillier

. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation. J Clin Microbiol, 1991; 29:297–301.

11.

Lisi

, Dondero

, Kwiatkoski

, Spence

, Rein

, Alderete

. Monoclonal-antibody-based enzyme-linked immunosorbent assay for Trichomonas vaginalis. J Clin Microbiol, 1988; 26:1684–1686.

12.

Sha

, Chen

, Wang

, Zariffard

, Cohen

, Spear

. Utility of Amsel criteria, Nugent score, and quantitative PCR for Gardnerella vaginalis, Mycoplasma hominis, and Lactobacillus spp. for diagnosis of bacterial vaginosis in human immunodeficiency virus-infected women. J Clin Microbiol, 2005; 43:4607–4612.

13.

Osborn

, Kunkel

, Nabel

. Tumor necrosis factor alpha and interleukin 1 stimulate the human immunodeficiency virus enhancer by activation of the nuclear factor kappa B. Proc Natl Acad Sci USA, 1989; 86:2336–2340.

14.

Poli

, Bressler

, Kinter

et al. Interleukin 6 induces human immunodeficiency virus expression in infected monocytic cells alone and in synergy with tumor necrosis factor alpha by transcriptional and post-transcriptional mechanisms. J Exp Med, 1990; 172:151–158.

15.

Zara

, Nappi

, Brerra

, Migliavacca

, Maserati

, Spinillo

. Markers of local immunity in cervico-vaginal secretions of HIV infected women: Implications for HIV shedding. Sex Transm Infect, 2004; 80:108–112.

16.

Wang

, Abel

, Lantz

, Krieg

, McChesney

, Miller

. The Toll-like receptor 7 (TLR7) agonist, imiquimod, and the TLR9 agonist, CpG ODN, induce antiviral cytokines and chemokines but do not prevent vaginal transmission of simian immunodeficiency virus when applied intravaginally to rhesus macaques. J Virol, 2005; 79:14355–14370.

17.

Fichorova

, Bajpai

, Chandra

et al. Interleukin (IL)-1, IL-6, and IL-8 predict mucosal toxicity of vaginal micro-bicidal contraceptives. Biol Reprod, 2004; 71:761–769.

18.

Al-Harthi

, Kovacs

, Coombs

et al. A menstrual cycle pattern for cytokine levels exists in HIV-positive women: Implication for HIV vaginal and plasma shedding. AIDS, 2001; 15:1535–1543.