JAIDS Journal of Acquired Immune Deficiency Syndromes:
Basic and Translational Science
Systemic and Mucosal Differences in HIV Burden, Immune, and Therapeutic Responses
Wahl, Sharon M PhD*; Redford, Maryann DDS*; Christensen, Shawna MS†; Mack, Wendy PhD‡; Cohn, Jon MD§; Janoff, Edward N MD‖; Mestecky, Jiri MD, PhD¶; Jenson, Hal B MD#; Navazesh, Mahvash DDS**; Cohen, Mardge MD††; Reichelderfer, Patricia PhD‡‡; Kovacs, Andrea MD†; for the DATRI 009 Study Group
From the *Oral Infection and Immunity Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD; †Maternal, Child and Adolescent Center for Infectious Diseases and Virology, University of Southern California Keck School of Medicine, Los Angeles, CA; ‡Department of Preventive Medicine, University of Southern California Keck School of Medicine, Los Angeles, CA; §Division of Infectious Diseases, Wayne State University, Detroit, MI; ‖Mucosal and Vaccine Research Program Colorado (MAVRC), VA Medical Center/University of Colorado Denver, Denver, CO; ¶Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL; #Department of Pediatrics, University of Texas Health Science Center, San Antonio, TX; **Division of Periodontology, Diagnostic Sciences, and Dental Hygiene, Herman Ostrow School of Dentistry of USC, Los Angeles, CA; ††Department of Medicine, Stroger (Cook County) Hospital, Chicago, IL; and ‡‡National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD.
Received for publication September 17, 2010; accepted December 22, 2010.
This study was supported by the Division of AIDS Treatment Research Initiative (DATRI) program, NIAID, NIH, Bethesda MD (Contract NO1-AI-15123); the Program Support Center, DHHS (Contract 282-98-0015, Task Order Number 21); and the Women's Interagency HIV Study, funded by NIAID, NIDCR, NCI, NICHD, NIDAA, the Agency for Health Care Policy and Research, the Centers for Disease Control and Prevention (Contracts U01-AI-35004, U01-AI-31834, U01-AI-34994, AI-34989, U01-HD-32632, U01-AI-34993, U01-AI-42590, NO1-DE-72621, and P01-AI-28147), and by the Intramural Research Program of the NIH, NIDCR.
The Woman's Interagency HIV Study Protocol was reviewed and approved by the institutional review boards at each participating center, and written and informed consent was obtained from all patients. Human experimentation guidelines of the U.S. Department of Health and Human Services were followed in the conduct of this research.
The authors have no conflicts of interest to disclose.
The members of DATRI 009 Study Group are listed in Appendix I.
Correspondence to: Dr. Sharon M. Wahl, PhD, Oral Infection and Immunity Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, 30 Convent Drive, MSC4352; Bldg 30, Room 320, Bethesda, MD 20892-4352 (e-mail: email@example.com).
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.jaids.com).
Background: Mucosal tissues represent major targets for HIV transmission but differ in susceptibility and reservoir function by unknown mechanisms.
Methods: In a cross-sectional study, HIV RNA and infectious virus were compared between oral and genital compartments and blood in HIV-infected women, in association with clinical parameters, copathogens, and putative innate and adaptive HIV inhibitors.
Results: HIV RNA was detectable in 24.5% of women from all 3 compartments, whereas 45% had RNA in only 1 or 2 sites. By comparison, infectious HIV, present in blood of the majority, was rare in mucosal sites. Innate mediators, secretory leukocyte protease inhibitor and thrombospondin, were highest in mucosae. Highly active antiretroviral therapy was associated with an 80% decreased probability of shedding. Multivariate logistic regression models revealed that mucosal HIV RNA was associated with higher plasma RNA, infectious virus, and total mucosal IgA, but not IgG. There was a 37-fold increased probability of detecting RNA in both genital and oral specimens (P = 0.008; P = 0.02, respectively) among women in highest versus lowest IgA tertiles.
Conclusions: Mucosal sites exhibit distinct characteristics of infectious HIV, viral shedding, and responses to therapy, dependent upon both systemic and local factors. Of the putative innate and adaptive mucosal defense factors examined, only IgA was associated with HIV RNA shedding. However, rather than being protective, there was a striking increase in probability of detectable HIV RNA shedding in women with highest total IgA.
The primary site of transmission and acquisition of HIV-1 in both women and men is through the genital mucosa.1 Enhanced levels of HIV in genital mucosal secretions have been associated with multiple factors including ulcers, local infections, inflammation, and plasma HIV RNA.2,3 By comparison, epidemiologic studies indicate that oral mucosa is an uncommon conduit for either transmission or acquisition of HIV,4-6 unlike herpesviruses (HSV1,2) and cytomegalovirus (CMV) that are transmitted by both oral and genital routes, and Epstein-Barr virus (EBV) and human herpesvirus-6 (HHV6) transmitted orally.7 However, bleeding in the oral cavity was suspected in HIV transmission in premasticated food from caregiver-to-infant,8 confounding the evidence. The relative infrequency of oral transmission may reflect a paucity of HIV replication locally, unique mucosal architecture, and/or presence of antiviral molecules,4,9-12 but differences in parameters underlying rates of genital and oral mucosal transmission are not well understood.
Multiple local and systemic virologic and immunologic differences must be considered in deciphering susceptibility and resistance profiles of mucosal sites relative to HIV transmission, including viral load, CD4+ T-cell counts, coinfections, antiviral therapy, innate, and adaptive immune responses. Among the endogenous innate mucosal factors that may differentially impact on HIV transmission and acquisition are mucins, defensins, secretory leukocyte protease inhibitor (SLPI), and thrombospondin (TSP-1), which have been shown to exhibit anti-HIV activity13 and adaptive immune response mediators and HIV-specific antibodies. Differences in these molecules among mucosal compartments or in peripheral blood and their relationship to HIV remain understudied, as does the relative impact of antiviral therapy. As the prevalence of HIV infection in women continues to increase, it remains critical to understand gender-specific susceptibilities, including routes of infection, localization of viral replication, role of innate factors, co-infections, potential compartmentalization of antiretroviral therapies (ARTs) and vaccine targeting. This cross-sectional study of HIV-infected women was designed to (1) compare HIV RNA (viral shedding) and infectious virus in 2 mucosal compartments, oral cavity, and genital tract, with these quantifiable measures of viral burden in peripheral blood, and (2) to identify selected influential intrinsic and extrinsic factors that may differ between compartments to influence HIV levels.
This was a cross-sectional substudy within Division of AIDS Treatment Research Initiative (DATRI 009).3 After recruitment of the first 225 women from Women's Interagency HIV Study (WIHS), a multicenter prospective study of HIV infection,14 the study was amended to enroll an additional 115 women (DATRI 009b) to include oral mucosa and expanded immunologic panel. Nonpregnant 19-year-old to 45-year-old infected women with intact uterus and cervix were invited to participate under separate informed consent, and excluded if presenting with AIDS-defining illness, had begun or changed ART within 1 month or reported coitus, douching, spermicide, or antimicrobial therapy in prior 48 hours. Women were evaluated at a single visit (6-month WIHS visit) that included collection of data regarding demographics, medications, illicit drug use, limited physical, oral, and vaginal examinations, and collection of blood, oral, and vaginal specimens.
Blood was collected in sodium citrate cell preparation tubes (Vacutainer, Becton-Dickinson, Franklin Lakes, NJ) and mononuclear cells and plasma separated by centrifugation (400g, 10 minutes). Plasma was centrifuged (1000g, 10 minutes) before storage (−70°C). Analyses of lymphocytes for CD4+ and CD8+ were performed using standardized protocols.3
Specimens were sequentially obtained as follows: (1) 4 swabs (buccal mucosa, tongue, floor of mouth, palate) and swabs of any active lesions, (2) buccal cytobrush for HIV culture, (3) unstimulated saliva collected in tubes on ice over 10 minutes, and (4) expectorated gargle obtained with 7.2 mL sterile distilled H2O. A dental examiner assessed mucosal and periodontal health in a subset. Swabs were cultured for Candida albicans. For culture of HIV, HSV, CMV, and respiratory viruses (influenza, parainfluenza, respiratory syncytial virus, adenovirus), specimens were placed in vials containing 2 mL RPMI 1640 with L-glutamine, 20% fetal bovine serum, 100 U/mL penicillin, 100 μL streptomycin, and 160 U/mL nystatin. For HIV polymerase chain reaction (PCR), swabs in medium with 4M guanidine isothiocyanate and mercaptoethanol were frozen until tested.3
Saliva was unspun or centrifuged (1500g, 20 minutes, 4°C). Unspun saliva was used for HIV RNA and culture, Human Papillomavirus (HPV) PCR and blood (Chemstrip-OB, Boehringer-Mannheim Diagnostics, Indianapolis, IN). Supernatants were diluted in Ca++Mg++-free phosphate-buffered saline (PBS) (1:3) and frozen until determination of immunoglobulins (Ig), SLPI, and TSP-1. Gargle mixed with 10X RPMI 1640 and 20% fetal bovine serum with antibiotics was aliquoted for EBV, HHV6, and HHV8 assays.15,16
After cervicovaginal lavage (CVL) with 10 mL sterile PBS, 4 endocervical swabs (for Candida albicans, pH, potassium hydroxide, amine odor test, microscopy, HIV RNA) and 1 cytobrush (HIV culture) were collected.3 Mixed swabs were tested for HSV PCR17 and cultured for HSV1, HSV2, CMV, adenovirus, enterovirus, and Varicella zoster.3 If present, genital lesions were swabbed for pathogen identification. Unspun CVL was tested for blood and semen (SemaTest, Humagen Fertility Diagnostics, Charlottesville, VA) and frozen for HIV RNA and HPV PCR18 or centrifuged (400g, 10 minutes) for Ig, SLPI, and TSP-1. Trichomonas vaginalis was assessed by wet mount and bacterial vaginosis by Amsel.19
Semiquantitative HIV cultures of blood and qualitative HIV cultures of genital (unspun CVL, swab) and oral (unspun saliva, swab) specimens were performed within 6 hours by protocol.20 Infectious virus was expressed as infectious units/106 cells (IUPM). For qualitative analysis, blood IUPM values >0.3219 were considered positive and ≤0.3219 negative.3
HIV RNA was assayed in oral swab, plasma, CVL, and unspun saliva with Nuclisens (Organon Teknika Corporation, Raleigh, NC), whereas cervical swabs were assayed with Roche Amplicor HIV-1 Monitor test (Roche Molecular Systems, Branchburg, NJ). Lower limit of detection (LLD) for plasma and cervical swabs was 400 copies per milliliter, and CVL, oral swabs, and saliva had LLD of 80 copies per milliliter.21
Centrifuged saliva (1:3 in PBS), CVL, and plasma were tested for SLPI and TSP-1 by enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN) or as described.12 Total IgG and IgA and HIV-specific antibodies were determined by capture enzyme-linked immunosorbent assay.22 IgG and IgA anti-gp120 were measured against standard curves of polyclonal secretory IgA or serum(Moni-Trol ES, Baxter, Stone Mountain, GA) with high anti-HIV-1 Ig as positive control and mucosal secretions and sera of uninfected individuals as negative controls. The cutoff (nonspecific) was set at 100 ng/mL.
For our analyses, HIV shedding was defined as presence of HIV RNA in oral swab or saliva and in genital swab or CVL using LLD above. We used DHHS/Kaiser Panel (DHHS/Kaiser 2004) guidelines to define HAART usage: (A) ≥2 NRTIs in combination with ≥1 PI or 1 NNRTI (88% of observations classified as HAART); (B) 1 NRTI in combination with ≥1 PI and ≥1 NNRTI (5%); (C) regimen containing ritonavir and saquinavir in combination with 1 NRTI and no NNRTIs (1%); and (D) abacavir or tenofovir containing regimen of ≥3 NRTIs in absence of PI and NNRTI (6%), except for 3 NRTI regimens consisting of: abacavir + tenofovir + lamivudine or didanosine + tenofovir + lamivudine.
Frequency distributions were calculated for categorical variables and median (range) for continuous variables. Associations among detection of HIV RNA in 3 compartments (oral, genital, blood) were assessed using a log-linear model; models were analyzed in total sample and also stratified by ART. HIV RNA was considered detectable in mucosal compartments if either swab or fluid was above LLD; women with both tests not completed were treated as unknown. To compare median levels of innate and adaptive molecules between compartments, Wilcoxon signed rank tests were used. Using logistic regression to assess relationships between HIV RNA in oral and genital compartments with clinical and laboratory assessments, odds ratios (ORs), and 95% confidence limits were calculated. The relationship between each variable and presence of HIV RNA in oral cavity and genital tract was analyzed in multivariate logistic regression models adjusting for ART (none, monotherapy, combination therapy, HAART) and plasma HIV RNA (<400, 400-10,000, ≥10,000 copies/ml) or ART only where appropriate. Additional multivariate models adjusted for CD4+ T cells (<200, 200 to <500, 500+ cells/mm3). In logistic regression models, SLPI, TSP, IgA, and IgG were categorized by tertile. Relationships between total IgA and HIV RNA within each compartment were estimated by Spearman rank correlation coefficients. Two-sided hypotheses were used throughout testing at 5% significance level.
HIV RNA and Infectious HIV in Blood, Oral Cavity and Genital Tract
Demographic and clinical characteristics of the WIHS cohort, including this subset (n = 115) 3,14 are summarized in Supplemental Table 1 (see Supplemental Digital Content 1, http://links.lww.com/QAI/A134). The majority were African American (57%) and Hispanic (20%) women who reported heterosexual contact as primary risk factor for transmission; only 24% reported injection drug use. Most had detectable plasma viral burden with approximately one-third having HIV RNA levels <400, one-third from 400-10,000, and the remainder >10,000 copies per milliliter (see Table, Supplemental Digital Content 1, http://links.lww.com/QAI/A134).
We evaluated presence of HIV RNA across compartments among the 106 women with available data in oral, blood and genital tract compartments in relationship to antiretroviral therapy (Table 1). Overall, HIV RNA was detectable in all 3 compartments in 24.5% of women and in none of the compartments for 29.3% (Table 1, Total). Women not receiving ART were most likely to have detectable HIV RNA in all 3 (35.1%) or any compartment (86.5%). Among women with undetectable plasma RNA, 8% had detectable RNA in genital secretions, and 3% in oral cavity (data not shown). Typically, HIV RNA levels were lower in genital and oral compartments than plasma [genital RNA/plasma RNA, median (range) =15.9% (0.3%-937.4%); oral RNA/plasma RNA = 8.6% (0.1%-300%)]. Although HIV RNA was detectable in the oral compartment of 30%, none had infectious virus and in genital secretions, 40% had detectable HIV RNA, but only 2 had detectable infectious virus (see Table, Supplemental Digital Content 1, http://links.lww.com/QAI/A134), irrespective of therapy, demonstrating differential dynamics of infectious HIV in systemic and mucosal compartments.
Mucosal and Systemic Innate and Adaptive Immune Factors and HIV
To explore additional factors that may contribute to disparate HIV detection in mucosal sites compared with blood, we monitored 2 innate factors reported to inhibit HIV, SLPI, and TSP-1.10-13,23 Although TSP-1 was highest in blood [Fig. 1A; median 193 ng/mL (range 9-1201)], these values were below those with demonstrable HIV inhibitory activity in vitro.12 Median TSP-1 in saliva was 22 ng/mL (0-128) and undetectable in CVL. On the other hand, median saliva SLPI levels (1225 ng/mL; 64-15,800) exceeded those in CVL (139 ng/mL; 2-4930) by nearly 10-fold (P < 0.0001) and in blood (40 ng/mL; 24-100) by 30-fold (P < 0.0001) (Fig. 1A), suggesting minimal, if any, contribution of these mucosal factors in control of HIV systemically, but potentially limiting infectious virus in mucosal compartments. Highest SLPI levels were found in oral cavity where no infectious virus (IUPM <0.32) was detectable, in contrast to blood, where SLPI was low and infectious virus was frequently detectable.
Mucosal fluids are also distinct from the systemic compartment by their levels of adaptive immune factors and antibodies that influence viral infectivity. We found that highest levels of both total and HIV-specific IgG and IgA were present in blood (Fig. 1B, C), where HIV RNA and culturable virus were also maximal (see Table, Supplemental Digital Content 1, http://links.lww.com/QAI/A134), demonstrating their inability to adequately control blood HIV RNA expression or culturable virus. At the mucosal level, differences were also evident in total and specific IgA and IgG. Total IgA was about 16-fold higher in saliva than CVL (Fig. 1B, P < 0.0001) and HIV gp120-specific IgA was also higher in saliva compared with CVL (median ∼40 μg/mL vs. undetectable in CVL) (Fig. 1C). Potentially, IgA directed to HIV antigens other than those specific for gp120, measured here, could also be represented in total IgA. By comparison, total IgG was about 3-fold higher in CVL than saliva (P < 0.0001) (Fig. 1B) but considerably lower than in blood. Similar to levels of total IgG, HIV-specific IgG was represented 2-fold to 3-fold more in CVL than saliva (P < 0.0001) (Fig. 1C). Despite predominance of total IgA in saliva, levels of HIV-specific IgG were comparable with those of specific IgA, as evident in CVL, suggesting that IgA responses to HIV are quite limited at mucosal sites.
Association of HIV RNA Shedding With Clinical, Virologic, and Immunologic Factors
To identify potential factors that could differentially impact HIV RNA shedding in oral (Table 2) and genital compartments (Table 3), we examined relationships between local and systemic virologic (HIV RNA, infectious virus) and immunologic factors (CD4 count, TSP, SLPI), as well as co-infections in each compartment. Multiple co-infections that were commonly detected in oral cavity included HHV6 (54%), Candida (49%) and EBV (21%), whereas herpes viruses (CMV, HSV1, HSV2, HHV8) were uncommon (0%-3%). In unadjusted models that examined association of these infections and SLPI and TSP with oral HIV shedding, we found that antiretroviral therapy (ART), plasma HIV RNA, CD4+ counts, quantitative PBMC cultures (IUPM) and EBV were all significantly associated with oral HIV shedding (Table 2), whereas the innate molecules SLPI and TSP were not associated with HIV RNA/shedding. When we adjusted for ART and plasma RNA, presence of blood infectious virus (IUPM) retained statistically significant associations with oral shedding, whereas EBV and CD4 count were no longer associated with shedding of HIV. For each log increase in plasma RNA, odds of oral HIV shedding increased 2.2-fold [95% confidence interval (CI): 1.6 to 2.9; P < 0.001]. Furthermore, women receiving HAART had 80% decreased odds (OR: 0.2; 95% CI: 0.1 to 0.8; P = 0.02) of oral HIV RNA compared with untreated women, independent of plasma RNA and CD4. There was no association observed between oral HIV shedding and presence of gingival bleeding, as a measure of inflammation (n = 28); however, a statistically significant association was observed for oral shedding with periodontal disease [maximum loss of attachment and pocket depth (P = 0.01 and P = 0.03, respectively)] and marginally with presence of oral lesions (n = 30; P = 0.08; data not shown).
Similar to oral cavity, there were multiple co-infections in genital tract, including Candida (26%), HPV (39%, including pathogenic strains 16,18,31,33,35 in 40%), and bacterial vaginosis (18%). CMV and HSV were infrequently detected (1.9% and 8.4%, respectively). As in oral cavity, plasma HIV RNA, CD4+ T cells and IUPM were all significantly associated with genital HIV shedding in unadjusted analyses (Table 3). Additional significant correlates were HPV (P = 0.01) and other viral infections (P = 0.01). TSP and SLPI were not associated with genital HIV RNA. However, in adjusted analyses, only plasma HIV RNA and infectious virus remained predictors for genital HIV shedding. Multivariate models showed that for each log increase in plasma RNA, odds of genital HIV shedding increased 2.8-fold (95% CI: 2.0 to 4.0; P < 0.001). ART was not associated with genital HIV shedding in this model and 41.2% of women on HAART still had detectable genital HIV RNA.
We next examined the influence of levels of IgG and IgA on HIV RNA shedding in multivariate models (Table 4). Unexpectedly we found that total oral IgA showed a strong positive association with oral HIV RNA levels. Women with the highest tertile of total IgA (≥64.968 μg/mL) were 37-times (95% CI: 1.8 to 744.6; P = 0.02) more likely to have detectable HIV RNA than women in the lowest tertile (<42.728 μg/mL), adjusting for ART and plasma HIV RNA (Table 4; P value for trend = 0.03). Adjusting for CD4+ counts did not substantially alter these results. Total IgA correlated directly with HIV RNA in unspun saliva (Spearman r = 0.54, P = 0.003) (Fig. 2A). Level of HIV-specific IgA was also significantly associated with oral HIV shedding (≥0.041 vs. <0.027 μg/mL: OR = 9.9; 95% CI: 1.0 to 97.6; P = 0.05) in adjusted analyses, whereas total IgG and HIV-1-specific IgG were not associated with HIV RNA/shedding. Because local mucosal co-infections could influence altered immunoglobulin levels, in addition to distorting immune function, disrupting the epithelial barrier and recruiting HIV target cells, we also assessed the potential relationship between oral pathogens and immunoglobulin levels. We found a marginal negative association between specific IgG and HHV6 (P = 0.08), with a statistically significant negative association between total IgG and HHV6 (P = 0.02).
Analogous to oral cavity, women in the highest tertile of genital IgA (≥3.8 μg/mL) were >37 times (95% CI: 2.6 to 531.1; P = 0.008) more likely to have genital HIV shedding as women in the lowest tertile (<2.0915 μg/mL) after adjusting for ART and plasma HIV RNA (Table 4; P value for trend = 0.007). Total IgA, but not HIV-specific IgA in CVL correlated directly with HIV RNA in both cervical swab (Spearman r = 0.45, P = 0.0008) and CVL (r = 0.44, P = 0.001) (Fig. 2B). Genital total IgG (≥15.532 vs. <7.758 μg/mL: OR = 4.7; 95% CI: 0.8 to 28.1; P = 0.09) showed a similar relationship with viral shedding after adjusting for ART and plasma HIV RNA or after adjusting for ART, plasma RNA, and CD4 (≥15.532 vs.<7.758 μg/mL: OR = 9.1; 95% CI: 1.1 to 77.7; P = 0.05), but HIV-specific IgG did not. Thus, despite being a relatively heterogeneous group of HIV+ women, it seems that levels of mucosal immunoglobulin, particularly IgA, are linked to local HIV RNA, independent of ART status or plasma virus. Further analyses revealed that in the genital compartment, a marginal positive association between total IgA and SLPI (r = 0.27; P = 0.08) was evident (data not shown).
This study evaluating potential determinants of infectious HIV and HIV RNA/shedding simultaneously in 3 compartments resulted in a number of notable findings. First, despite HIV RNA being detectable in mucosal fluids, it was rarely culturable, whereas virus was frequently cultured from blood. Second, of several putative innate and adaptive mucosal defense factors examined, only IgA was associated with shedding HIV RNA in the mucosae. However, rather than finding a protective effect, there was a striking increase in the probability of having detectable HIV RNA in the oral or genital tracts in women with highest total IgA levels. Finally, women who were on HAART, albeit less potent than HAART regimens in current use, often had detectable oral or genital HIV RNA, in addition to blood HIV. Nevertheless, among women with undetectable plasma HIV, 8% still had genital shedding and 3% had oral HIV RNA.
Because plasma HIV RNA and infectious virus are prominent determinants of HIV RNA/shedding in both oral and genital compartments, it was surprising that systemic HIV levels were not associated with infectious/culturable virus in the mucosae. Although the typically lower levels of mucosal HIV RNA may reflect this disparity in infectious titers, even when plasma HIV RNA was undetectable or at low levels, virus could still be cultured from some blood samples. These data support the notion that there are mucosal-specific factors that compromise the ability of virions to transmit infection, consistent with epidemiologic evidence that rates of mucosal HIV transmission are relatively low per high-risk exposure.24,25 A number of local factors were evaluated as potential determinants of infectious HIV and shedding HIV RNA in mucosal fluids. Although neither SLPI nor TSP-1, 2 innate mediators, correlated with presence or level of HIV RNA, highest SLPI levels and lack of infectious HIV were found in mucosae. The ability of SLPI to neutralize elastase, associated with HIV shedding,2 blunt inflammation,2,26 inhibit bacterial and fungal co-infections,27 and/or target host cells to inhibit HIV,10,11,13,28,29 may contribute to a resistant, albeit not impenetrable mucosal phenotype. Lower SLPI levels in CVL than saliva are perhaps related to inability to differentiate between active and inactivated SLPI.30,31 In mucosa, SLPI levels are reportedly altered by Candida albicans, HSV and/or HIV itself.32-34 Increased risk of HIV in HSV2-seropositive individuals35,36 in the context of decreased SLPI,34 the association between CVL SLPI titers and perinatal HIV transmission,37 and influence of breast milk SLPI on maternal-infant transmission38 underscore the potential significance of this mucosal antimicrobial peptide. Comparative analyses of additional anti-HIV factors, including defensins and chemokines, may reveal molecules that independently, or in the aggregate, may antagonize infection, particularly if antibodies are inadequately neutralizing.
In this regard, the most significant and consistent association in both genital and oral fluids was between the dominant IgA mucosal antibody isotype and HIV RNA. However, high total IgA levels in the 2 mucosal compartments (and oral HIV-specific IgA) were associated with a higher likelihood of HIV RNA/shedding. Enhanced mucosal total IgA may reflect polyclonal stimulation of IgA-secreting cells by co-infections, chronic activation, or HIV stimulation of mucosal IgA differentiation and secretion.39 Total IgA levels may also be influenced by high mucosal levels of SLPI, which regulates IgA through activation induced deaminase.40 In chronic HIV infection, polyclonal B-cell activation, exhaustion and hypergammaglobulinemia occur in the periphery, and antiviral therapy is associated with a rapid decline in plasma HIV antibodies,41,42 seemingly inconsistent with the mucosal data in these chronically infected women, the majority of whom are on antiviral therapy. Although studies have identified HIV-specific IgA in blood and mucosal secretions that neutralizes HIV in in vitro assays,39,43-46 higher IgA in our studies was not related to decreased viral RNA systemically or mucosally. Whether this mirrors the reported HIV-enhancing ability of serum IgA from HIV+ individuals,47 failure of neutralizing antibodies to interfere with HIV transcytosis through epithelial cells48 or other mechanism(s) is unclear, but the connection between IgA and enhanced HIV shedding may be relevant in vaccine development49 and needs further study. In one recent study, HIV-specific IgA not only did not protect but was also detected more frequently in breast milk of transmitters.50 Although our analyses were limited by the cross-sectional design, such observations may relate to the recently described ability of IgA antibodies to bind intact HIV virions and coordinate their basolateral to apical transport across epithelium.51 If this were the case, the increase in mucosal HIV RNA that we observed in the context of elevated total and/or HIV-specific IgA may be related to IgA-mediated HIV excretion into the mucosal lumen, which could also be consistent with reduced infectious virus despite high HIV RNA. Although such possibilities await further study, these intriguing findings highlight the need to better understand mucosal innate and adaptive factors that may modify HIV transmission between sexual partners and from mother-to-child and also confound vaccine development.
Further assessment of suspect local factors affecting mucosal immunoglobulins and HIV demonstrated that inflammatory lesions and the presence of copathogens had less influence than anticipated, albeit these analyses may have been hampered by relatively small numbers of exposed women with certain infections. In prior cross-sectional and prospective studies, mucosal co-infections have been associated with enhanced HIV replication,2,3 likely related to breach of mucosal integrity, recruitment of HIV-susceptible targets and/or HIV-infected cells, activation of NFκB, and other pathways.29 In this regard, CMV, HPV, and herpes viruses may directly or indirectly transactivate HIV.52,53 EBV+ women reportedly have increased risk of perinatal HIV transmission,54 and in our study, were more likely to have genital or oral HIV RNA in unadjusted analyses. However, when adjusted for plasma RNA levels, this association was lost, suggesting shared mechanisms may influence local viral replication.
No single factor controls HIV shedding and/or replication, but the collective components of the local milieu, including mucosal architecture, co-infections, innate and adaptive immune factors, and susceptible target cells with their unique viral tropisms, together with circulating viral burden and systemic immunity, likely determine mucosal expression and infectivity of HIV. Because these studies were performed on samples obtained before the use of more potent HAART regimens currently available, we were able to take advantage of the ability to more readily monitor infectious HIV and HIV RNA in parallel. Though clearly not independent of systemic determinants of viral replication and infectivity, mucosae also provide an independent source of viral determinants. Characterization of local factors is essential to not only understand HIV transmission and pathogenesis, but also support vaccine development. Because preventing infection is a primary goal to limit the AIDS pandemic, full characterization of endogenous antiviral mechanisms may provide new insights into development of prophylactic and therapeutic approaches against HIV, particularly those involving mucosal tissues, the most common sites of primary HIV transmission.
We would like to thank the following for their contributions: S. Lewis, V. Goveia, A. Soloviov, S. Beckner, V. Dodgin, S. Wasserman, and D. Wright at Westat, Rockville, MD.
1. Smith P, Wahl S. Immunobiology of mucosal HIV-1 infection. In: Mestecky J, Bienenstoch J, Lamm ME, et al, eds. Mucosal Immunology. 3rd ed. Amsterdam, The Netherlands: Elsevier/Academic Press; 1199-1203.
2. Anderson DJ, Politch JA, Tucker LD, et al. Quantitation of mediators of inflammation and immunity in genital tract secretions and their relevance to HIV type 1 transmission. AIDS Res Hum Retroviruses. 1998;14(Suppl 1):S43-S49.
3. Kovacs A, Wasserman SS, Burns D, et al. Determinants of HIV-1 shedding in the genital tract of women. Lancet. 2001;358:1593-1601.
4. Shugars DC, Wahl SM. The role of the oral environment in HIV-1 transmission. J Am Dent Assoc. 1998;129:851-858.
5. Schacker T, Collier AC, Hughes J, et al. Clinical and epidemiologic features of primary HIV infection. Ann Intern Med. 1996;125:257-264.
6. Page-Shafer K, Shiboski CH, Osmond DH, et al. Risk of HIV infection attributable to oral sex among men who have sex with men and in the population of men who have sex with men. AIDS. 2002;16:2350-2352.
7. Bruce AJ, Rogers RS III. Oral manifestations of sexually transmitted diseases. Clin Dermatol. 2004;22:520-527.
8. Gaur A, Dominguez K, Kalish M, et al. Practice of offering a child pre-masticated food: an unrecognized possible risk factor for HIV Triansmission. Paper presented at: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, MA.
9. Hussain LA, Lehner T. Comparative investigation of Langerhans' cells and potential receptors for HIV in oral, genitourinary and rectal epithelia. Immunology. 1995;85:475-484.
10. McNeely TB, Shugars DC, Rosendahl M, et al. Inhibition of human immunodeficiency virus type 1 infectivity by secretory leukocyte protease inhibitor occurs prior to viral reverse transcription. Blood. 1997;90:1141-1149.
11. McNeely TB, Dealy M, Dripps DJ, et al. Secretory leukocyte protease inhibitor: a human saliva protein exhibiting anti-human immunodeficiency virus 1 activity in vitro. J Clin Invest. 1995;96:456-464.
12. Crombie R, Silverstein RL, MacLow C, et al. Identification of a CD36-related thrombospondin 1-binding domain in HIV-1 envelope glycoprotein gp120: relationship to HIV-1-specific inhibitory factors in human saliva. J Exp Med. 1998;187:25-35.
13. Malamud D, Wahl SM. The mouth: a gateway or a trap for HIV? AIDS. 2010;24:5-16.
14. Barkan SE, Melnick SL, Preston-Martin S, et al. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology. 1998;9:117-125.
15. Pellett PE, Spira TJ, Bagasra O, et al. Multicenter comparison of PCR assays for detection of human herpesvirus 8 DNA in semen. J Clin Microbiol. 1999;37:1298-1301.
16. Jenson HB, Ench Y, Sumaya C, et al.Manual of Clinical Laboratory Immunology. 5th ed. Washington, DC: American Society for Microbiology; 1997.
17. Schacker T, Ryncarz AJ, Goddard J, et al. Frequent recovery of HIV-1 from genital herpes simplex virus lesions in HIV-1-infected men. JAMA. 1998;280:61-66.
18. 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.
19. Amsel R, Totten PA, Spiegel CA, et al. Nonspecific vaginitis. Diagnostic criteria and microbial and epidemiologic associations. Am J Med. 1983;74:14-22.
20. Division of AIDS N. DAIDS Virology Manual for HIV Laboratories. Washington, DC: US Department of Health and Human Services; 1997.
21. Nowicki MJ, Benning L, Bremer JW, et al. Longitudinal variability of human immunodeficiency virus type 1 RNA viral load measurements by nucleic acid sequence-based amplification and NucliSens assays in a large multicenter study. J Clin Microbiol. 2001;39:3760-3763.
22. Mestecky J, Jackson S, Moldoveanu Z, et al. Paucity of antigen-specific IgA responses in sera and external secretions of HIV-type 1-infected individuals. AIDS Res Hum Retroviruses. 2004;20:972-988.
23. Mestecky J, Moldoveanu Z, Russell MW. Immunologic uniqueness of the genital tract: challenge for vaccine development. Am J Reprod Immunol. 2005;53:208-214.
24. Powers KA, Poole C, Pettifor AE, et al. Rethinking the heterosexual infectivity of HIV-1: a systematic review and meta-analysis. Lancet Infect Dis. 2008;8:553-563.
25. Wawer MJ, Gray RH, Sewankambo NK, 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.
26. Orenstein JM, Wahl SM. The macrophage origin of the HIV-expressing multinucleated giant cells in hyperplastic tonsils and adenoids. Ultrastruct Pathol. 1999;23:79-91.
27. Hiemstra PS, Fernie-King BA, McMichael J, et al. Antimicrobial peptides: mediators of innate immunity as templates for the development of novel anti-infective and immune therapeutics. Curr Pharm Des. 2004;10:2891-2905.
28. Ma G, Greenwell-Wild T, Lei K, et al. Secretory leukocyte protease inhibitor binds to annexin II, a cofactor for macrophage HIV-1 infection. J Exp Med. 2004;200:1337-1346.
29. Wahl SM, Greenwell-Wild T, Vazquez N. HIV accomplices and adversaries in macrophage infection. J Leukoc Biol. 2006;80:973-983.
30. Vogelmeier C, Hubbard RC, Fells GA, et al. Anti-neutrophil elastase defense of the normal human respiratory epithelial surface provided by the secretory leukoprotease inhibitor. J Clin Invest. 1991;87:482-488.
31. Draper D, Donohoe W, Mortimer L, et al. Cysteine proteases of Trichomonas vaginalis degrade secretory leukocyte protease inhibitor. J Infect Dis. 1998;178:815-819.
32. Jana NK, Gray LR, Shugars DC. Human immunodeficiency virus type 1 stimulates the expression and production of secretory leukocyte protease inhibitor (SLPI) in oral epithelial cells: a role for SLPI in innate mucosal immunity. J Virol. 2005;79:6432-6440.
33. Chattopadhyay A, Gray LR, Patton LL, et al. Salivary secretory leukocyte protease inhibitor and oral candidiasis in human immunodeficiency virus type 1-infected persons. Infect Immun. 2004;72:1956-1963.
34. Fakioglu E, Wilson SS, Mesquita PM, et al. Herpes simplex virus downregulates secretory leukocyte protease inhibitor: a novel immune evasion mechanism. J Virol. 2008;82:9337-9344.
35. Smith D. HIV and herpes co-infection, an unfortunate partnership. J HIV Ther. 2004;9:1-3.
36. Celum C, Levine R, Weaver M, et al. Genital herpes and human immunodeficiency virus: double trouble. Bull World Health Organ. 2004;82:447-453.
37. Pillay K, Coutsoudis A, Agadzi-Naqvi AK, et al. Secretory leukocyte protease inhibitor in vaginal fluids and perinatal human immunodeficiency virus type 1 transmission. J Infect Dis. 2001;183:653-656.
38. Farquhar C, VanCott TC, Mbori-Ngacha DA, et al. Salivary secretory leukocyte protease inhibitor is associated with reduced transmission of human immunodeficiency virus type 1 through breast milk. J Infect Dis. 2002;186:1173-1176.
39. Karray S, Juompan L, Maroun RC, et al. Structural basis of the gp120 superantigen-binding site on human immunoglobulins. J Immunol. 1998;161:6681-6688.
40. Xu W, He B, Chiu A, et al. Epithelial cells trigger frontline immunoglobulin class switching through a pathway regulated by the inhibitor SLPI. Nat Immunol. 2007;8:294-303.
41. Morris L, Binley JM, Clas BA, et al. HIV-1 antigen-specific and -nonspecific B cell responses are sensitive to combination antiretroviral therapy. J Exp Med. 1998;188:233-245.
42. Moir S, Fauci AS. Pathogenic mechanisms of B-lymphocyte dysfunction in HIV disease. J Allergy Clin Immunol. 2008;122:12-19; quiz 20-11.
43. Hirbod T, Kaul R, Reichard C, et al. HIV-neutralizing immunoglobulin A and HIV-specific proliferation are independently associated with reduced HIV acquisition in Kenyan sex workers. AIDS. 2008;22:727-735.
44. Kaul R, Plummer F, Clerici M, et al. Mucosal IgA in exposed, uninfected subjects: evidence for a role in protection against HIV infection. AIDS. 2001;15:431-432.
45. Moja P, Tranchat C, Tchou I, et al. Neutralization of human immunodeficiency virus type 1 (HIV-1) mediated by parotid IgA of HIV-1-infected patients. J Infect Dis. 2000;181:1607-1613.
46. Devito C, Hinkula J, Kaul R, et al. Cross-clade HIV-1-specific neutralizing IgA in mucosal and systemic compartments of HIV-1-exposed, persistently seronegative subjects. J Acquir Immune Defic Syndr. 2002;30:413-420.
47. Janoff EN, Wahl SM, Thomas K, et al. Modulation of human immunodeficiency virus type 1 infection of human monocytes by IgA. J Infect Dis. 1995;172:855-858.
48. Chomont N, Hocini H, Gody JC, et al. Neutralizing monoclonal antibodies to human immunodeficiency virus type 1 do not inhibit viral transcytosis through mucosal epithelial cells. Virology. 2008;370:246-254.
49. Watkins DI, Burton DR, Kallas EG, et al. Nonhuman primate models and the failure of the Merck HIV-1 vaccine in humans. Nat Med. 2008;14:617-621.
50. Kuhn L, Trabattoni D, Kankasa C, et al. Hiv-specific secretory IgA in breast milk of HIV-positive mothers is not associated with protection against HIV transmission among breast-fed infants. J Pediatr. 2006;149:611-616.
51. Wright A, Lamm ME, Huang YT. Excretion of HIV-1 through polarized epithelium by immunoglobulin A. J Virol. 2008;82:11526-11535.
52. Gage JR, Sandhu AK, Nihira M, et al. Effects of human papillomavirus-associated cells on human immunodeficiency virus gene expression. Obstet Gynecol. 2000;96:879-885.
53. Yurochko AD, Huong SM, Huang ES. Identification of human cytomegalovirus target sequences in the human immunodeficiency virus long terminal repeat. Potential role of IE2-86 binding to sequences between -120 and -20 in promoter transactivation. J Hum Virol. 1999;2:81-90.
54. Pitt J, Schluchter M, Jenson H, et al. Maternal and perinatal factors related to maternal-infant transmission of HIV-1 in the P2C2 HIV study: the role of EBV shedding. Pediatric Pulmonary and Cardiovascular Complications of Vertically Transmitted HIV-1 Infection (P2C2 HIV) Study Group. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;19:462-470.
APPENDIX I: THE DATRI 009 STUDY GROUP
The DATRI 009 study group also includes the following: A. Levine, USC Keck School of Medicine; H. Minkoff, Maimonides Med. Center; M. Young, Georgetown Univ.; David Burns, NICHD; Paolo Miotti, NIAID; Penny Baron, MSKCC; Larry Corey, Univ. of Wash.; Kathleen Weber, Cook County Hospital; Alan Landay, Rush Presbyterian; Barbara Weiser, N.Y. State Dept. of Health; Patricia Garcia, Northwestern Univ.; Beverly Sha, Rush Presbyterian; Ronald Hershow, Univ. of Illinois; Bill Meyers, Quest Diagnostics; Bob Grant, UCSF; Yvonne DeSouze, UCSF; Maria Wamerdam, UCSF; Joel Palefsky, UCSF; Marek Nowicki, USC Keck School of Medicine; Cheryl Jennings, Cook County Hospital; James Bremer, Rush Presbyterian; Eric Peterson, Univ. of Wash.; Alex Ryncarz, Univ. of Wash.; Anne Cent, Univ. of Wash.; William Hardy, MSKCC; Jan Englund, Baylor College of Medicine.
adaptive immunity; HIV-1; mucosa; innate immunity; IgA; SLPI
Supplemental Digital Content
© 2011 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.