Microbial Translocation and Immune Activation in HIV-1 Infected Pregnant Women

Abstract

Background:

Immune Activation (IA) has been previously documented in both pregnant (PG) and non-PG HIV-1 infected (HIV+) women as well as in HIV- uninfected PG women; the latter as a result of the fetal allograft. To determine whether the combined effects of HIV and pregnancy result in increased IA and whether IA is associated with Microbial Translocation (MT), we performed a prospective, longitudinal, controlled study during pregnancy and the postpartum (PP) period.

Methods:

HIV+ PG women had biomarkers of IA and MT tested at 12–20 weeks (T1), and 24–36 weeks (T2) of pregnancy and at 6–8 weeks Postpartum (T3). HIV+, non-PG women were tested at comparable time points. HIV- PG women were tested at T1 only. HIV+ women were not started on antiretroviral therapy (ART) until T1. Biomarkers of IA assessed included: CD4DR+, CD4CD38+, CD4DR+CD38+, CD8DR+, CD8CD38+, and CD8DR+CD38+. Biomarkers of MT included LPS, sCD14, and 16SrDNA.

Results:

30 HIV+PG women, 18 HIV+ non-PG and 10 HIV-PG were enrolled. In the HIV+ women, there were no differences in median age, viral load, % or absolute CD4 at entry. Significant differences between T1 and T2 and between T1 and T3 were noted in CD8DR+CD38+ in HIV+PG women after ART. CD4DR+, CD4DR+CD38+, and CD8DR+ decreased post ART in HIV+PG women but a decline in IA was less evident in HIV+ non-PG. LPS decreased post ART by T3 in both HIV+PG and HIV+ non-PG groups; 16SrDNA was elevated at all time points in both groups when compared to control values, and declined post ART in the HIV+PG group. A subgroup of HIV-PG at T1 had IA and MT as evidenced by several IA markers and increased LPS.

Conclusion:

The degree of IA and MT was similar among HIV+PG and HIV+ non-PG women followed longitudinally. There was no incremental increase due to the combined effects of HIV and pregnancy. Several markers of IA and MT (LPS, 16SrDNA) decreased post ART. IA and MT occurred in a subgroup of HIV-PG women during the 1^st trimester. Further study must be done to confirm whether MT consistently occurs in some healthy women during PG.

1. INTRODUCTION

While pregnancy is known to be associated with a state of maternal immune tolerance and an increased risk of morbidity and mortality resulting from certain intercurrent infections (i.e. influenza has an increased risk of hospitalizations and death among both HIV+ and HIV− women [1]), it has also been shown to be associated with IA. The latter has been thought to occur as a result of exposure to the fetal allograft [2]. More recently, immune activation (IA) has been documented in both PG and non-PG HIV+ women [3–6]. The latter observations are consistent with what is now known about the role of microbial translocation and the resultant immune activation in promoting HIV pathogenesis [7–14]. In order to study the relationship that microbial translocation may have with immune activation in PG HIV+ women, we performed a prospective, longitudinal observational study to determine whether the immune activation in PG, HIV+ women was associated with microbial translocation and whether the intensity of IA and MT in PG HIV+ women was greater than these same parameters among non-PG HIV+ women followed for a comparable period of time.

2. METHODS

2.1. Study Design

We conducted a prospective, longitudinal, cohort controlled study which enrolled HIV-1 infected pregnant (Experimental Group), and HIV+ non-PG women (Control Group 1) and followed both groups over 3 time points. The Pregnant, HIV+ women were enrolled between the 12–20^th weeks of gestation (Baseline or visit#1, T1), and seen for follow-up visits at the 24–36^th weeks of gestation (study visit#2, T2) and at 6–8 weeks postpartum (study visit#3, T3). The Control Group 1 subjects were seen at Baseline (T1), and again at 12+/−2 weeks later (T2), and 12+/− 2 weeks after that for the 3^rd and final visit (T3). All Control Group 1 subjects were documented to not be pregnant at the time of entry. Ten HIV uninfected PG women between 10–20 weeks of gestation were also enrolled (Control Group 2). These women only had a single visit at baseline to provide values for both Immune Activation and Microbial Translocation from this control population.

2.2. Study Participants

All Experimental Group Subjects were recruited from the population of HIV+ PG women receiving care at the Special Immunology Prenatal Clinic (PRIM) at the University of Miami Miller School of Medicine/Jackson Memorial Hospital (UMMSOM/JMH). Control Group#1 subjects were recruited from the Adult Special Immunology Clinics at the UMMSOM/JMH while Control Group#2 subjects were recruited from the Routine Prenatal Care Clinics. All HIV infected study subjects were not on Highly Active Antiretroviral Therapy (HAART) at the time of the first study visit being either naïve for HAART or not having received antiretroviral agent for the prior 6 months. All Experimental Group subjects were started on a HAART regimen at the time of their baseline visit but only after study bloods were drawn. All Experimental Subjects remained on HAART throughout their pregnancy. Similarly, all Control Group#1 subjects were not started on HAART until after their study bloods were drawn during the baseline visit and remained on HAART throughout all 3 study visits. All Control subjects were monitored during the study to determine if they had become pregnant, with the proviso that they would be replaced in the study by another Control subject should this happen. The study was approved by the Human Subjects Committee at the UMMSOM/JMH. Informed consent was obtained from each study subject prior to enrollment.

2.3. Clinical Data

The following demographic and clinical data was collected from the study subject’s medical records: date of birth, sex, ethnic and racial group, HAART naïve versus off HAART for more than 6 months, gestational age at enrollment and at delivery, live birth, mode of delivery, birth weight, complete blood counts, and absolute CD4 and percent CD4 count. Women previously known to be co-infected with either Hepatitis C or B were excluded from the study because of their potential association with microbial translocation [15–17].

2.4. Laboratory Measurements

Screening for T cell activation:

Flow cytometry for activated T were performed in whole blood and measures of LPS and 16sRibosomal DNA were determined in plasma isolated from blood collected in EDTA. All blood specimens were drawn without regard to whether the subjects were fasting or not. Activation markers on T cells: Activation markers on T cells were assessed by measuring the expression of CD38 and HLA-DR on gated CD3+CD8+ and CD3+CD8-(henceforth referred to as CD3+CD4+) cells. In brief, 100 μl of EDTA-anticoagulated whole blood was dispensed into a 5-mL tube followed by 20 μl of each fluorochrome-labeled monoclonal antibody. Antibodies used include anti-CD38-phycoerythrin (PE), anti-CD8-peridinin chlorophyll protein (APC), anti-CD3- peridinin chlorophyll protein (PerCP) anti-HLA-DR-fluorescein isothiocyanate (FITC) and appropriate isotype control mAbs (BD PharMingen, San Jose, CA). After 15 min incubation at room temperature followed by lysis of the red blood cells by use of commercial lysis reagents (FACS Lysing Solution [Becton-Dickinson, BD]), the cells were washed with PBS and the samples fixed in 1% EM-grade paraformaldehyde, followed by analysis by flow cytometry on BD FACSCalibur (BD Biosciences). Flow Cytometry data analysis was performed using FlowJo v8.6 software (Tree Star, Inc, OR). Values in healthy volunteers for CD38+ HLADR+ cells were <5%.

Lipopolysaccharide (LPS):

Plasma, diluted to 20% with endotoxin-free water was heated to 80°C for 10 min to inactivate plasma proteins and eliminate false positives caused by serine proteases. Plasma LPS was quantified using a commercially available LAL (Limulus Amebocyte Assay, QCL-1000, from Lonza, Walkersville, MD) according to the manufacturer’s protocol. Samples were run in duplicate and background was subtracted. Endotoxin in the test sample was quantified between the ranges of the endotoxin concentrations used to construct a standard curve. Precautions were taken to avoid false positive and false negative reactions. The results were reported as EU/ml, and are converted to pg/mL. Using this method, values in healthy volunteers were <75 pg/mL.

16s Ribosomal DNA:

The assay was performed as per Jiang et al. [18.] by real time PCR. The sequences of the primers and probe were as follows: 8F Forward Primer AGAGTTTGATCCTGGCTCAG, 355R Reverse Primer CTGCTGCCTCCCGTAGGAGT, 63P TaqMan Probe FAM-GCAGGCCTAACACATGCAAGTC-BHQ1, and the size of the product insert is 347bp in length. DNA was extracted from 200μl of plasma using the QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA). 50 μl of AE buffer was used to elute the DNA. 5 μl of DNA was used for each reaction for real time PCR. Each PCR contained platinum buffer, 2.5μl; 50mM Mg, 1.75μl; 10mM dNTP, 0.5μl; 12.5uM 5’ primer, 1μl; 12.5uM 3’ primer, 1μl; 5uM probe, 1μl; ROX 100x, 0.25μl; Platinum Taq, 0.125μl; and 5μl of template/standard in a total volume of 25μl made up with water. The PCR conditions were as follows: 95°C for 5min to activate the Platinum Taq, 95°C 30s and 60°C 1min for 40 cycles. Nontemplate controls were also included. The real time fluorescence detection was with the ABI PRISM 7700 Sequence Detector (Perkin-Elmer Applied Biosystems) to quantify the bacterial 16S rRNA gene (rDNA) level in serum and/or plasma. The SDS sequence detection software version 1.9.1 (ABI) of the instrument determined the threshold cycle value (CT) which represents the PCR cycle at which an increase in reporter fluorescence above a baseline signal can be detected. The software generated a calibration curve of CT values against a standard serial dilution of calibrator DNA and determined the unknown value in the samples by interpolation. Real time PCR was performed in triplicate for each standard dilution of the samples and the mean CT value of the triplicate PCRs was determined and used for calculation of 16s Ribosomal DNA levels. Values in healthy donors were <2 copies/ml.

2.5. Statistical Analysis

Comparisons between groups were evaluated using standard t tests. The tests were considered statistically significant with a p-value ≤ 0.05. All the statistical analyses were performed using GraphPad Prism version 7.04 for Windows, GraphPad Software, La Jolla California USA (www.graphpad.com). Comparison between 2 groups was performed using Unpaired T test or Mann-Whitney Test respectively for normally or not-normally distributed data evaluated by D’Agostino-Pearson Normality Test. For the comparison between proportions, the Chi-squared test was applied. The number of * noted above the panels in the Figures denotes the following: * p</= 0.05, ** p</= 0.01, *** p</= 0.001, and **** p</= 0.0001.

3. RESULTS

3.1. Study Participants and Demographics

Between June 23, 2011, and April 7, 2015, we enrolled and completed follow-up on 30 HIV-infected pregnant women in the Experimental Group, and 18 HIV infected, non-pregnant women in Control Group 1. We initially targeted HIV infected women (both PG and non-PG) who were HAART naïve but because our efforts were hampered by the small numbers of such patients encountered, we subsequently expanded our entry criteria to allow enrollment of HIV-1 infected women who had been off HAART for 6 or more months as well.

Nine of the 30 subjects in the Experimental Group were HAART naïve while 21 had been off HAART for six or more months. Similarly, only four of the Control Group1 subjects (HIV infected, non-pregnant) were HAART naïve while 14 had been off HAART for six or more months. Ten HIV uninfected, pregnant women were enrolled into Control Group 2 to provide baseline values for the Experimental Group and were tested only one time; at entry. We were only able to enroll 18 of the targeted 30 for Control Group 1 within the period funded for the study. All HIV-infected subjects were started one of 3 general categories of ART regimens: 1. 2 nucleoside reverse transcriptase inhibitor (NRTI) and a non-nucleoside reverse transcriptase inhibitor (NNRTI), 2. 2 NRTI plus a protease inhibitor (PI), or 3. Triple NRTI). Compliance with ART assessed by verbal recall was reported to vary between 70–100% in the HIV+PG group and 0–100% among the HIV+ non-PG group. Table 1 describes the Demographics of the study population including both the HIV viral load (VL) and AbsCD4 and %CD4 data at entry and again at the time of the last visit. We initially attempted to match the Experimental and the Control Groups by age and ethnic group but these efforts were hampered but the difficulties encountered in enrolling into Control Group#1. While there were significant differences in the 3 groups in terms of their ethnicity, there was no significant difference between the 3 groups in maternal age. There was also no significant difference at baseline between the HIV+PG+ group and the HIV+ non-PG group in the following parameters: 1. number of women who were ART naive relative to the number who had been off ART for >/= 6 months. 2. HIV viral load, 3. absolute and 4. percent CD4 counts. The mean viral load at the third visit among the HIV+ non-PG was significantly higher than the viral load in the HIV+PG+ group likely due to poorer compliance in the former group. There was a significant difference in the proportions of infants delivered via C/Section between the HIV+PG+ and the HIV-PG+ group but this was likely secondary to the small size of the latter group. None of the subjects experienced a major acute infectious complication during the study although one profoundly immunosuppressed woman in the Experimental Group died post- delivery with Clinical AIDS and Progressive Multifocal Leukoencephalopathy (PML). None of the HIV infected non-pregnant group became pregnant during the course of the study and none of the perinatally exposed infants were infected themselves.

Table 1.

Demographics and laboratory parameters.

	Pregnant HIV+ Women (30) Experimental Group	Non-pregnant HIV+ Women (18) a. Control Group 1	HIV uninfected, pregnant Women (10 b. Control Group 2
Median Maternal Age (range) *	26.5 (19–36)	30.3(18–58)	26.6 (19–36)
Ethnicity African American	21(70%)	12(66%)	4(36%)
White Hispanic	9(30%)	3(17%)	6(55%)
White nonHispanic#		3(17%)	1(9%)
Naive	9	6	NAc.
Off ART > 6mos. *	21	10	NA
Mode of Delivery(Vagnal/CS)^	10/20	NA	4/6
Mean Gest.Age at Entry(weeks) *	15(9–38)	NA	16(10–23)
Live Births	30	NA	10
Mean Gest.Age at birth *	37(34–40)	NA	37(28–41)
Mean Birth Weight(grams; range) *	2831(1215–3690)	NA	2873(1070–4270)
Mean VL at Entry(range) *	12615(20–126627)	44329(95–229627)	NA
Mean VL at postpartum /3rd visit+	264(20–6549)	33885(20–290290)	NA
Mean AbsCD4(%) at Entry *	453(24–960)	516(19–1093)	NA
Mean AbsCD4(%) at the Postpartum/3rd visit *	533(77–1080)	631(45–1359)	NA

^a.Two women had incomplete demographic data.

^b.One visit only at entry. One mother had incomplete demographic data.

^c.Not Applicable.

^•Non-significant

^#p= 0.0001 Chi square

^{^}p=0.0003

⁺p=0.0253

Despite being an exclusion criteria, 6 women were subsequently found to be co-infected with either Hepatitis B Virus (one in the HIV+PG+ group) or Hepatitis C Virus (5 in the HIV+ non-PG group). The data from these 6 women was not included in the following analysis of the IA and MT.

3.2. Immune Activation in HIV infected Pregnant, and Nonpregnant Women.

There was a significant drop in the degree of IA as reflected by the CD4+DR+ cell surface markers between the baseline visit prior to starting HAART (T1), and T2, and between T1 and T3 in the pregnant, HIV-infected group (Fig. 1B). There was no significant decrease in the IA marker CD4+CD38+ between T1 and T3 although there was a significant decrease between T2 and T3 (Fig. 1A). When analyzed using dual IA marker staining, CD4+DR+CD38+, there was a significant decrease in IA between T1 and T2 and between T1 and T3 (Fig. 1C), suggesting that HAART decreased but did not suppress IA in these pregnant HIV-infected women.

Fig. (1). CD4 Immune Activation.

Comparison of CD4 T cell activation markers between HIV+ pregnant women, HIV+ non-pregnant women and HIV− pregnant women. HIV+ pregnant time points correspond to weeks of gestation. HIV+ non-pregnant time points are at similar times after enrollment, not conception. The red line in each set represents the median value. The blue line represents the 90% value for healthy controls. The t test was performed on paired data (i.e. longitudinal values for a given subject).

Among the HIV-infected non-pregnant women, while there was a significant decrease between T1 and T3 in IA, as measured by the CD4+DR+ marker (Fig. 1B), there was no significant decrease in IA between the 3 time points with CD4+CD38+ or dual staining using CD4+DR+CD38+(Fig. 1C).

A similar pattern was seen in HIV-infected pregnant group when IA was assessed using CD8+DR+, and CD8+ CD38+, as well as with dual staining with CD8+DR+CD38+ (Fig. 2). There was a consistent decrease between T1 and T3 in all three parameters measured, with the most significant difference between T1 and T2, and T1 and T3 noted with CD8+DR+ (Fig. 2B), and CD8+DR+CD38+ (Fig. 2C).

Fig. (2). CD8 Immune Activation.

Comparison of CD8 T cell activation markers between HIV+ pregnant women, HIV+ non-pregnant women and HIV− pregnant women. HIV+ pregnant time points correspond to weeks of gestation. HIV+ non-pregnant time points are at similar times after enrollment, not conception. The red line in each set represents the median value. The blue line represents the 90% value for healthy controls. The t test was performed on paired data (i.e. longitudinal values for a given subject).

There was no significant decrease noted between any of the time points with CD8+DR+, CD8+CD38+, or CD8+ DR+CD38+ among HIV infected women who were not pregnant (Figs. 2C). There was a significant decrease in the CD8+CD38+ cells between T1 and T3 in the HIV+ non-PG group.

Curiously, among the 10 pregnant women who were not infected with HIV, about half of them had T cell immune activation at entry (at the one time point they were studied) as evidenced by CD4+DR+CD38+, CD8+DR+, CD8+ CD38+, and CD8+DR+CD38+ staining confirming an earlier report (Figs. 1 and 2) [6.].

3.3. Microbial Translocation and Macrophage Activation in HIV infected Pregnant, and Nonpregnant Women.

Among pregnant, HIV-1 infected women, there was also significant changes in the two biomarkers of microbial translocation (LPS, and 16SrDNA) and sCD14; a marker of macrophage activation over the three study visits (Fig. 3). Lipopolysaccharide (LPS) decreased significantly between T1 and T3 (Fig. 3A); while 16SrDNA decreased significantly between T1 and T2 (Fig. 3C). In contrast, sCD14 increased significantly between T1 and T3 and between T2 and T3 (Fig. 3B).

Fig. (3). Microbial Translocation.

Comparison of the plasma levels of microbial translocation markers (A) LPS, (C) 16SrDNA and immune activation marker (B) sCD14 between HIV+ pregnant women, HIV+ non-pregnant women and HIV− pregnant women. HIV+ pregnant time points correspond to weeks of gestation. HIV+ non-pregnant time points are at similar times after enrollment, not conception. The red line in each set represents the median value. The blue line represents the 90% value for healthy controls. The t test was performed on paired data (i.e. longitudinal values for a given subject).

Among non-pregnant HIV infected women, LPS decreased between T1 and T2 and between T1 and T3 but there were no significant changes in either 16SrDNA or sCD14 over the course of the study (Fig. 3).

Among the 10 HIV negative pregnant women tested at baseline, six of them had LPS levels which were above the upper limit of the range of normal values while all of the values for 16SrDNa and sCD14 in this group of women were essentially within range of normal. The noted increase in LPS in these 6 were not matched with an increase in the levels of sCD14 (Fig. 3).

A repeat analysis of both the IA and MT data from the entire group of enrolled subjects including the data from the 6 excluded women did not reveal significant differences from the results described above except for a significant decrease in the CD8+CD38+ cells between T1 and T3 in the HIV+ non-PG group.

4. DISCUSSION

Pregnancy is known to be associated with an increased risk of morbidity and mortality resulting from certain infections (e.g. Influenza) and have an increased risk of hospitalizations and death among both HIV infected and uninfected pregnant women [1.], but the nature of the immune processes that occur during pregnancy that may contribute to this defect have not been well characterized. In general, pregnancy has traditionally been considered an example of maternal immune tolerance of the fetus as a “semi-allograft” [1]. A growing body of evidence, however, now suggests that the immunological relationship between the mother and the fetus is more complex, and is characterized by variable degrees of maternal immune activation [19], which may serve to promote implantation of the embryo and subsequently growth and development of the conceptus. Loewendorf et al. [20] documented that the proportions of HLA-DR+/CD38+ effector CD45RA+ CD8 cells are significantly increased in both the peripheral blood and at the utero-placental interface (UPI) in HIV negative pregnant women suggesting an inflammatory milieu. They also demonstrated the presence of activated, HLA DR+ T regulatory cells (Tregs) in these same patients. The Treg cells were present in higher concentrations at the UPI than in the blood; their function presumably to prevent the rejection of the fetus by the maternal immune system. In a 2017 paper in Nature Reviews Immunology, Mor et al. [19] described the occurrence of 3 successive immunological stages during gestation. The first stage occurring during the first trimester is pro-inflammatory and facilitates blastocyst implantation and placentation; the second stage between weeks 13–27 of gestation is anti-inflammatory and promotes tolerance of the fetus while facilitating its growth. The third stage between weeks 28 and 40 is again pro-inflammatory and is responsible for the initiation of parturition. Mor further proposed in this same paper that inflammation induced by a concomitant viral infection (ie. CMV, Zika, HIV) could disrupt this ordered sequence and increase the risk of fetal damage and premature birth.

Mikyas et al.in a much earlier study [3] found that HLA-DR+,CD38+CD8+ cells, serum neopterin, and B-2 microglobulin all increased during late pregnancy in HIV negative pregnant women compared to non-pregnant women. Truong in a more recent study from 2010 [5], documented a rise in the mean number of absolute CD4 cells and CD8 cells among HIV infected treated and untreated pregnant women during pregnancy. They also found that beta-2 microglobulin and neopterin were increased at all time points among HIV infected versus uninfected pregnant women. They proposed that the heightened immune activation observed in pregnant HIV infected women resulted from the combined effects of both pregnancy and HIV-mediated inflammation, which in turn promoted higher HIV viral loads.

Microbial Translocation has been documented to occur in every group of HIV-infected individuals studied [10, 12, 13] but this is the first paper documenting the occurrence of microbial translocation among HIV pregnant women during gestation and the post-partum period. We found that Immune activation and microbial translocation were both demonstrated during pregnancy and the postpartum period. While both decreased after starting HAART, both entities continued to be present. The degree of immune activation was similar among HIV+, pregnant women and HIV+, nonpregnant women. Hence there was no incremental increase due to the combined effects of both HIV and pregnancy at these time points. There was significant decrease among the HIV+, pregnant women in immune activation as measured by CD4+ and CD8+cells dually positive for DR+CD38+ between entry and the second gestational visit and entry and the postpartum visits after starting HAART. There was also a similar pattern of decline in CD4DR+ and CD8DR+ cells at these same time points. Among the HIV+, non-pregnant women starting HAART, there was a significant decrease in immune activation only in CD4+DR+ cells noted between the entry and the post-partum visits.

Following the initiation of HAART, significant drops in LPS were noted among both HIV + pregnant and nonpregnant women; in contrast, a significant decrease in 16sDNA was observed only in HIV+ pregnant women. Surprisingly, the median level of sCD14 increased following the initiation of HAART among HIV+ pregnant women whereas the median levels of sCD14 among HIV+ non-pregnant women remained the same.

Among the 10 HIV negative pregnant women tested one time at Entry, approximately half of them had IA as evidenced by CD4+- and CD8+-DR+CD38+ staining. A similar number also had elevated LPS although there was no increases in either the 16sDNA or sCD14.

One limitation of this study was that we did not assess medication compliance at each visit. The respective median HIV viral loads at Entry and at the time of the last visit, however, suggested that compliance with HAART was generally good in the Experimental Group and more variable in Control Group#1. Compliance with ART in the HIV+ PG group was reported to be 70–100% with the greater majority claiming they were 100% compliant. Compliance among the HIV+ non-PG group was more variable with reported rates of compliance varying between 0–100%. In the authors’ experience based upon over 30 years of experience of working with HIV-affected families in Miami (including HIV infected women and their infected children), the greater majority of HIV+ pregnant women are almost 100% compliant with HAART during gestation in an effort to protect their unborn child. As we did not determine how many of the women were still taking HAART at the last visit, it is possible that our results may have been influenced by diminished compliance. Diminished compliance among the HIV+ non-PG women during the later study visits may be one reason why there was less dramatic changes in IA this latter group.

While none of the participants were diagnosed with an overt infectious complication during the study, we did not assess the study participants for any subclinical infectious process, ie. CMV, [21] which might have contributed to IA.

Our target accrual for both HIV-infected study groups was 30. While we were able to reach the targeted accrual in the Experimental group, we were only able to enroll 18 HIV-infected non-pregnant control subjects because of the difficulty in finding women who were either HAART naïve or had been off HAART for more than 6 months. The increasing availability of HAART coupled with the fact that many HIV infected women were diagnosed and started on HAART at an outside facility before being seen at our institution was the basis for this difficulty. Despite this limitation, we believe that the data derived from these patients has provided us with valid control data.

CONCLUSION

Immune activation as well as microbial translocation were both repeatedly documented during pregnancy and the postpartum period among HIV infected pregnant women before and after starting HAART. The degree of immune activation was similar among HIV+ pregnant women and non-pregnant women; hence there was no incremental increase in immune activation due to the combined effects of both HIV and pregnancy. Interestingly, immune activation and microbial translocation were both observed in a subgroup of HIV uninfected pregnant women; the significance of which warrants further study.

LIST OF ABBREVIATIONS

16SrDNA

16 S ribosomal Deoxynucleic acid

ART

Antiretroviral Therapy

DHHS

Department of Health and Human Services

HAART

Highly Active Antiretroviral Therapy

HIV

Human Immunodeficiency Virus

IA

Immune Activation

LPS

Lipopolysaccharide

MT

Microbial Translocation

PG

Pregnant

UMMSOM/JMH

University of Miami Miller School of Medicine/Jackson Memorial Hospital

VL

Viral Load