Effects of Co-Trimoxazole on Microbial Translocation in HIV-1-Infected Patients Initiating Antiretroviral Therapy

Abstract

Microbial translocation (MT) contributes to immune activation during HIV-1 infection, and persists after initiation of antiretroviral therapy (ART). We investigated whether levels of MT markers are influenced by the use of co-trimoxazole (TMP-SMX) in HIV-1 patients. Plasma samples were obtained from HIV-1-infected patients initiating ART with (n=13) or without (n=13) TMP-SMX prophylaxis. Markers of MT [lipopolysaccharide-binding protein (LBP), lipopolysaccharide (LPS), soluble CD14 (sCD14), and intestinal fatty acid binding protein (I-FABP)] were assessed at baseline (BL), at 1 month, and at 1 year by the Limulus Amebocyte Lysate Assay or ELISA. BL levels of LBP were elevated in both categories of patients; they were highest in patients starting ART and TMP-SMX (median, μg/ml: 36.7 vs. 4.3, respectively, p=0.001) and correlated inversely with CD4⁺ T cell counts (ρ=−0.65; p=0.005). Patients receiving ART and TMP-SMX had a significant reduction in LBP between BL and 1 year (median, μg/ml: 36.7 vs. 11.1; p=0.003). In contrast, levels of LPS at BL were lower in patients starting ART and TMP-SMX compared to those without TMP-SMX (median, pg/ml: 221 vs. 303 respectively; p=0.002) and did not change at 1 year. The increased BL levels of sCD14 had declined in both groups at 1 year. No difference in I-FABP levels was found between BL and 1 year. Concomitant use of ART and TMP-SMX reduces microbial translocation markers LBP and sCD14, probably due to its impact on the gut microbiota. Effective ART for 1 year does not restore gut–blood barrier dysfunction.

Introduction

The lymphoid tissue in the gut is infected by human immunodeficiency virus type 1 (HIV-1) very early after infection,^1,2 resulting in damage to the gut–blood barrier. This includes a permanent loss of Th17⁺ CD4⁺ T cells, impaired antimicrobial mucosal immunity, changes in enterocyte homeostasis,³ and translocation of microbial products to the blood. Although antiretroviral therapy (ART) leads to a pronounced inhibition of the viral replication and recovery of CD4⁺ T cells, this translocation persists.⁴ The level of microbial translocation (MT) is similar to the levels in diseases such as inflammatory bowel diseases, coeliac disease, and sepsis, and its role in chronic immune activation and inflammation in HIV-1 infection has been detailed.^5
–7

MT can be studied using several direct markers in blood such as lipopolysaccharide (LPS) and bacterial 16S rDNA.^8

–11 In addition, indirect markers are used such as soluble CD14 (sCD14), a protein secreted from monocytes activated by bacterial or viral stimuli, lipopolysaccharide-binding protein (LBP), an acute-phase protein synthetized mainly in the liver upon LPS stimuli, interleukin (IL)-6, D-dimer, and antiflagellin antibodies.^12

–16 Additionally, intestinal fatty acid binding protein (I-FABP), a protein released from enterocytes undergoing cell death, is used as a marker of gut permeability.¹⁷

Antibiotics influence the composition of the gut microbiota. Hence, Helicobacter pylori eradication with metronidazole and clarithromycin for 7 days in patients with dyspeptic disorders had both short-term and long-term effects on the intestinal microbiota.¹⁸ In addition, treatment with oral vancomycin for 1 week reduced fecal microbial diversity, causing a population shift from gram-positive to gram-negative bacteria.¹⁹ Thus, use of antibiotics might change the commensal gut microbiota and reduce the total bacterial load, which may impact the magnitude of MT.

The link between antibiotic use and MT has emerged in several studies. For example, rifaximin, a nonabsorbable broad-spectrum oral antibiotic, was shown to reduce MT in patients with decompensated cirrhosis.²⁰ In a model with simian immunodeficiency virus-infected monkeys, 2 weeks of treatment with neomycin, metronidazole, and cefotaxime led to a significant decrease in MT.⁴ However, although several studies have reported an increased MT in HIV-1-infected patients,^4,8,21 only a few have reported the impact of antibiotics.^10,16 In our previous work, levels of sCD14 were affected by antibiotic use in HIV-1 patients starting ART.¹⁰ In addition, we have shown that the use of tuberculostatics decreases MT.¹⁶

Based upon these observations, we hypothesized that the levels of MT markers may be influenced by the use of co-trimoxazole (TMP-SMX) in patients who receive it as prophylaxis against Pneumocystis jiroveci. Such knowledge can be valuable for the interpretation of studies of MT in HIV-1-infected patients with advanced disease. We therefore analyzed several biomarkers for MT in plasma obtained at baseline, at 1 month follow-up, and at 1 year in treatment-naive patients with immunodeficiency who received TMP-SMX as prophylaxis together with first line ART.

Materials and Methods

Subjects

Patients with HIV-1 infection (n=26) followed at the Department of Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden, who initiated first line ART, were selected from a larger cohort based on sample availability (Table 1). They were divided into two groups depending on whether they started ART and concomitantly TMP-SMX prophylaxis against Pneumocystis jiroveci (160 mg/800 mg three times a week) (n=13) or not (n=13). TMP-SMX was started at a median of 7 days (IQR 1.5–21) and ART 12 days (7.5–15) after the first visit (baseline; BL) at the clinic. In the non-TMP-SMX group ART was initiated 21 days (8.5–52) after BL. ART consisted of two nucleoside reverse transcriptase inhibitors (NRTIs): tenofovir (n=15) and abacavir (n=6) or zidovudine (n=4) in combination with lamivudine/emtricitabine, and the nonnucleoside reverse transcriptase inhibitor (NNRTI) efavirenz (n=7) or one of the ritonavir-boosted protease inhibitors (PI/r) lopinavir (n=8), darunavir (n=4), or atazanavir (n=6), with the exception of one patient in the group of ART who was treated only with the integrase inhibitor raltegravir+darunavir/r.

Table 1.

Characteristics of the HIV-1-Infected Patient-Cohorts at Start of Antiretroviral Therapy Together With or Without Concomitant Co-trimoxazole (TMP-SMX) Treatment

	All	TMP-SMX group	Non-TMP-SMX group
Subjects, number	26	13	13
Age, years: median (range)	39 (25–70)	42 (25–70)	37 (25–70)
Sex (male/female)	16/10	8/5	8/5
Hepatitis, number B/C^a	1/3	1/2	0/1
CD4⁺ T cells/μl: median (range)	180 (0–350)	100 (0–190)^b	260 (110–350)
HIV-1 RNA copies/ml: median (range)	153,000 (3,690–10,000,000)	276,000 (26,500–10,000,000)	99,700 (3,690–48,300)
Mode of transmission
Heterosexual	14	7	7
MSM	6	3	3
PWID	3	2	1
Blood transfusion	1	0	1
Unknown	2	1	1
Use of tenofovir, number	15	10	5
Days to follow-up 1: median (IQR)	47 (36–62)	38 (31–56)	58 (40–70)
Weeks to follow-up 2: median (IQR)	50 (44–56)	51 (44–52)	46 (41–57)

One (in TMP-SMX group, with hepatitis C) of four patients with hepatitis had cirrhosis.

The baseline CD4⁺ T cell count was significantly lower in the co-trimoxazole (TMP-SMX) group (p<0.0001).

MSM, men who have sex with men; PWID, people with intravenous drug use.

Plasma samples had been drawn as a part of clinical care and kept at −70°C until analysis. Three samples of each patient were analyzed at BL, 1 month (follow-up 1; FU1), and 1 year (FU2). The time between BL and FU1 was a median of 38 days (IQR 31–56) in the TMP-SMX group and 58 days (40–70) in the non-TMP-SMX group. TMP-SMX was given for a median of 27 days (25–50) before FU1. The time between BL and FU2 was a median of 51 weeks (IQR 44–52) and 46 weeks (41–57), respectively. Five of the 13 patients continued the prophylaxis until FU2. The other eight were on TMP-SMX for a median of 176 days (IQR 58–208). Three patients had AIDS events (extrapulmonary tuberculosis, candida esophagitis, Kaposi's sarcoma) at the time of HIV diagnosis in the TMP-SMX group, but none in the other group. The study was approved by the Regional Ethics Committee at the Karolinska Institutet (Dnr 2009/1485-31).

CD4⁺ T cells and plasma HIV-1 RNA

Analysis of CD4⁺ T cell counts and plasma HIV RNA load was performed as part of the clinical routine with flow cytometry and COBAS Taqman HIV-1 RNA test v.2.0 (Roche Molecular Systems Inc., Branchburg, NJ), respectively.

Quantification of microbial translocation markers

LPS levels were determined by the Limulus Amebocyte Lysate Assay (LAL, Lonza, MD) according to the manufacturer's instructions, with the following modifications: samples were diluted 5-fold to avoid interference with background color and preheated to 70°C for 12 min prior to analyses to dissolve immune complexes, as described elsewhere.⁸ Linear regression analysis was performed to obtain the corresponding endotoxin concentration of the samples from their absorbance values (background was subtracted after each run). Quantifications of LBP, sCD14, and I-FABP levels were performed by enzyme-linked immunosorbent assays (Hycult biotech, the Netherlands; R&D Systems, USA; DS Pharma Biomedical Co., Japan, respectively) according to the manufacturer's instructions.

Statistical analysis

Comparisons between independent groups were performed by the Mann–Whitney U test. Longitudinal analyses were performed by the Wilcoxon matched pairs test. The relationship between two variables was determined with Spearman's correlation. Changes in LPS, LBP, sCD14, and I-FABP from BL to FU1 and FU2 were analyzed with a generalized linear mixed-effects model adjusted by gender, age, viral load, and CD4⁺ T cells at BL. The model included a random intercept and random slope that allowed individual variations in the change of the MT markers over time. Data were analyzed by GraphPad Prism 5.04 and STATA 12 SE/12 and presented as median and as IQR if not stated otherwise. A p-value<0.05 was considered significant.

Results

CD4⁺ T cells and HIV-1 RNA

At BL, the median CD4⁺ T cell count was 100 cells/μl (IQR 50–145) and 260 cells/μl (205–325), respectively, in the patients who thereafter received prophylaxis or not (p<0.0001) (Table 1). The T cell recovery from BL to FU2 did not differ significantly between the groups. No significant difference in HIV-1 load was found between the groups at BL (median, copies/ml: 276,000 vs. 99,700) (p=0.087). At FU1, the viral load was higher in the TMP-SMX group (1,740 vs. 800 copies/ml; p=0.027). At FU2, no difference was found in the proportion of patients with >50 copies/ml [in the TMP-SMX group 4/12 (of whom there were two blips) and in the non-TMP-SMX group 2/13 (both blips)]. Blips were defined as intermittent periods of detectable low-level viremia that return spontaneously to an undetectable level without any change in ART.

Lack of change in LPS levels during 1 year of ART

CD4⁺ T cells and LPS correlated positively at BL (ρ=0.46; p=0.02). The LPS levels were lower in the TMP-SMX group compared to the non-TMP-SMX group (pg/ml: 221; 159–266 vs. 302; 289–346; p=0.002), despite higher CD4⁺ T cells in the latter group. The difference persisted at FU1 and FU2 (p=0.003 vs. 0.004, respectively) (Fig. 1A). LPS levels increased significantly between BL and FU1 in the non-TMP-SMX group only (pg/ml: 302; 289–346 vs. 329; 313–354; p=0.0007). There was no significant change in LPS levels between BL and FU2 within the two groups, and the same result was revealed by the generalized linear mixed-effects model. When stratifying the patients into two groups based upon LPS increase or decrease between BL and FU2, individuals with declining LPS (n=7) tended to have a better CD4⁺ T cell recovery (cells/μl: 340; 200–400 vs. 150; 110–215; p=0.057).

FIG. 1.

Baseline (BL) levels of lipopolysaccharide (LPS) were lower in the co-trimoxazole group (TMP-SMX) compared to the non-co-trimoxazole group (non-TMP-SMX) (p=0.002) and did not change significantly in any of the groups between BL and follow-up 2 (FU2) (A). Lipopolysaccharide-binding protein (LBP) levels at BL were elevated in the TMP-SMX group compared to the non-TMP-SMX group (p=0.001) and were significantly reduced at FU2 in TMP-SMX only (p=0.003) (B). CD4⁺ T cells and LBP were significantly negatively correlated at BL (C). BL levels of sCD14 tended to be higher in TMP-SMX (p=0.07), and decreased longitudinally until FU2 in both groups (TMP-SMX; p=0.001, non-TMP-SMX; p=0.02) (D). Horizontal bars indicate median values. Comparisons between independent groups were performed using the Mann–Whitney U test, and the Wilcoxon matched pairs test was used for longitudinal analyses. Spearman's rank test was applied for correlations. *p<0.05, **p<0.01, ***p<0.001.

Decrease of LBP levels after 1 year in TMP-SMX-treated patients

At BL, LBP levels were elevated in the TMP-SMX group compared to the non-TMP-SMX group (μg/ml: 36.7; 14.1–49.9 vs. 4.3; 2.2–14.5; p=0.001) (Fig. 1B). The levels remained elevated in the TMP-SMX group at FU1. No differences between the groups were found at FU2. In the TMP-SMX group, a reduction in the LBP level was seen between BL and FU2 (11.1; 4.5–18.7) (p=0.003), but not in the other group (8.9; 5.9–14.5) (p=0.47) (Fig. 1B).

The different kinetics of LBP between the groups persisted in the generalized linear mixed-effects model, where patients who did not receive TMP-SMX prophylaxis had a nonsignificant increase in LBP at FU2, as opposed to patients with prophylaxis who had a decrease (Table 2A).

Table 2.

Adjusted Analysis of Mean Change of LBP (A) and sCD14 (B) During Antiretroviral Therapy, With or Without Concomitant Co-Trimoxazole (TMP-SMX) Treatment

A. LBP	Coefficient (95% CI) ^a	p-value
Non TMP-SMX
FU	−2.9 (−11.8–6.1)	0.53
FU2	1.2 (−7.3–9.7)	0.79
TMP-SMX
FU	−0.6 (−13.3–12.1)	0.93
FU2	−24.4 (−36.5– −12.4)	<0.001

B. sCD14	Coefficient (95% CI) ^a	p-value
Non TMP-SMX
FU	−147067.7 (−807317.9–513182.5)	0.662
FU2	−1271497 (−2039733–−503260.7)	0.001
TMP-SMX
FU	−56090.7 (−989825.5–877644.1)	0.906
FU2	−217981 (−1304168–868205.6)	0.694

Generalized linear mixed-effects model with a random intercept and random slope adjusted by gender, age, viral load, and CD4 at baseline. FU, first time point of follow-up; FU2, second time point of follow-up. Non TMP-SMX refers to patients who did not receive prophylaxis and TMP-SMX to those who received prophylaxis.

LBP, lipopoly saccharide-binding protein; sCD14, soluble CD14.

At BL, a negative correlation between CD4⁺ T cells and LBP was seen (ρ=−0.65; p=0.005) (Fig. 1C) and a positive correlation was seen between LBP and sCD14 (ρ=0.42; p=0.03), but not at FU1 and FU2. LPS and LBP were negatively correlated at BL (ρ=−0.45; p=0.03) and FU2 (ρ=−0.65; p=0.002).

Levels of sCD14 were reduced by ART

At BL, VL and sCD14 were positively correlated (ρ=0.55; p=0.005) and the levels of sCD14 tended to be higher in the TMP-SMX group (μg/ml: 3.7×10⁶; 3.1–4.3×10⁶ vs. 2.9×10⁶; 1.8–3.9×10⁶; p=0.07). A significant reduction was seen between BL and FU2 in both groups (p=0.001; p=0.02, respectively) (Fig. 1D). After adjustment for covariables in the generalized linear mixed-effects model, sCD14 levels decreased significantly between BL and FU2 in the non-TMP-SMX group only (Table 2B).

Elevated levels of I-FABP associated with use of tenofovir

At BL, the groups had similar levels of I-FABP (Fig. 2A). In the TMP-SMX group, the levels increased from BL to FU1 (1.68; 0.99–2.7 vs. 4.67; 1.49–8.19) (p=0.001) and a decrease was seen at FU2, to levels similar to those at BL. In the non-TMP-SMX group, I-FABP levels at FU1 and FU2 were similar to BL levels. No significant differences in the kinetics of I-FABP were revealed in the generalized linear mixed-effects model.

FIG. 2.

Similar levels of intestinal fatty acid binding protein (I-FABP) were observed in the co-trimoxazole (TMP-SMX) group and the non-co-trimoxazole (non-TMP-SMX) group at baseline (BL), and increased between BL and follow-up (FU) in the TMP-SMX group (p=0.001), but did not change significantly between BL and follow-up 2 (FU2) in any group (A). Tenofovir (TDF)-treated patients had a significant elevation of I-FABP at FU compared to BL (p=0.0003) (B). Horizontal bars indicate median values. Comparisons between independent groups were performed using the Mann–Whitney U test, and the Wilcoxon matched pairs test was used for longitudinal analyses. **p<0.01, ***p<0.001.

After stratifying the patients into groups based upon I-FABP increase or decrease between BL and FU1, the CD4⁺ T cell recovery was found to be 103 cells/μl in the group with increasing I-FABP levels and 210 cells/μl in the group with decreasing I-FABP levels, respectively (p=0.02). Patients were also categorized according to chosen NRTI-based ART. In the group (n=15) with tenofovir-based ART, I-FABP levels increased from BL (ng/ml 1.92; 1.26–3.15) to FU1 (4.13; 2.62–6.44; p=0.0003) (Fig. 2B). At FU2, the levels had declined to levels similar to the BL (data not shown). However, patients with non-tenofovir-based ART (n=11) did not have significantly altered I-FABP levels.

Discussion

It is well established that MT is increased in immunodeficient HIV-1-infected patients, although there are diverging results regarding the extent of the MT and how it is influenced by ART.^4,9,21 Also, metagenomics data show that the gut microbiota is changed in HIV-1-infected individuals.^22,23 One possible confounding factor is the use of antibiotics/antifungals since the effects of these drugs on the microbiome are significant and can persist for a long time.^18,20,24 We hypothesized that the influence of such a drug may also be reflected in the markers of systemic inflammation and MT. Therefore, the kinetics of MT markers was assessed in treatment-naive patients starting ART with or without TMP-SMX.

Our findings at baseline clearly confirmed that MT is present in untreated patients with immunodeficiency. Thus, levels of well-established MT markers (LPS, LBP, sCD14, and I-FABP) were increased compared to historical healthy controls and the CD4⁺ T cells correlated negatively with LBP.²⁵ However, we report a positive correlation between LPS levels and the CD4⁺ T cells, in contrast to our findings in a previous work.¹⁰ A similar finding was reported by Marchetti et al. with low LPS levels in severely immune compromised patients.²⁶ A possible explanation is a compositional shift of patients microbiome in the severely immune compromised subjects²³ studied here and/or the impact of a more frequent use of antibiotics in such patients.

Thus, in our study five out of 13 patients in the TMP-SMX group received antibiotics and/or antifungals within 3 months, and an additional three patients 6 months before the baseline visit, compared to only one patient in the ART group. These antibiotic-treated patients tended to have lower LPS levels at BL, but the difference did not reach statistical significance (data not shown). Thus, a consequent change in the composition of the gut microbiota with a relative decrease in gram-negative species could have resulted in the lower LPS levels. Other alternative explanations could be an increased LPS clearance, mirrored by a negative correlation between LPS and LBP, or intestinal bacterial overgrowth with a gram-positive microbiota shift.⁴

It is clearly unlikely that 1 month of ART will improve the damage to the gut–blood barrier and any changes in MT biomarkers at that time point represent other biological events. In our study, the I-FABP levels increased after 1 month among the patients who received ART and TMP-SMX. Tenofovir use was more prevalent in this group, and tenofovir-treated individuals had significantly higher levels of I-FABP at 1 month. We speculate that the transient enterocyte damage (reflected by higher I-FABP levels) might be linked to temporary enterocyte toxicity, as higher tenofovir concentrations have been observed in the gut compartment compared to the blood.²⁷ Still, an immune reconstitution syndrome (IRIS) in the gut cannot be dismissed. The latter hypothesis is supported by the finding of increased LPS levels at month 1 in the ART-treated group since an IRIS in the gut is likely to increase the permeability for bacterial products. However, in a recent large cohort study on HIV-TB-coinfected patients receiving TB treatment and starting ART, levels of I-FABP after 1 month of ART were lower, but LBP levels were higher in patients when compared to patients without IRIS.²⁸ Intestinal biopsies are needed to determine if IRIS in the gut compartment influences I-FABP levels.

Conflicting results of LPS kinetics during ART have been reported with both decreasing and unaltered levels.^4,8,9,29 We have previously reported that MT decreases, but still persists, after 72 weeks of successful ART.¹⁰ The present study included a shorter follow-up (median 51 weeks), but the patients had a more advanced immunodeficiency at baseline than in our earlier study. In line with our earlier results, a reduction in the sCD14 levels was observed at 1 year in both treatment groups.^10,16,29 In addition, the LPS levels did not decline or normalize during this short follow-up.

Overall, LPS levels have declined between BL and FU2 in only seven patients (four in the TMP-SMX group), who tended to have a more favorable CD4 T cell recovery. It is difficult to reach a solid conclusion concerning this relationship because of the small cohort size, but the higher CD4 T cell reconstitution in blood may reflect decreased bacterial translocation across the gut–blood barrier due to local reconstitution in gut-associated lymphoid tissue. In contrast, the LBP levels declined in the patients who received TMP-SMX. LPB can be induced and bind other bacterial products (such as peptidoglycans) from, e.g., gram-positive bacteria. Thus, these broad microbiological triggers of LBP will be reduced by TMP-SMX use, as the total bacterial load in the gut decreases, contributing to the reduction in LBP in the group of antibiotic-treated patients only. The discrepancy in LBP kinetics between the groups persisted even when adjusting for CD4⁺ T cell count in the generalized linear mixed-effects model. This suggests that the reduction in LBP in the TMP-SMX group was related to the antibiotic prophylaxis, and not to different levels of baseline immunodeficiency between the groups.

The strength of our study is the longitudinal design; to date there have been relatively few studies assessing the long-term kinetics of MT markers during HIV-1 infection.^{10,16,29
–31} Additionally, none has analyzed the impact of the use of antibiotics on this process except for our earlier studies in which data indicate that antibiotics may influence the levels of MT biomarkers.^10,16This issue was highlighted in a recent review of co-trimoxazole prophylaxis during HIV infection.³² Our study has obvious limitations. We retrospectively analyzed a small cohort of patients, which limits the opportunities to detect more subtle differences between groups. The groups had differences in CD4⁺ T cell counts at the start, but those did not influence the results when the model adjusting for significant covariables was applied. As it was a retrospective study, we could not avoid the differences in timing of specimen collection, which might have influenced the differences between groups at FU1. Furthermore, characterization of the gut microbiome was not performed, restricting the possibility of correlating our findings with antibiotic-related alternations of the gut flora.

In summary, our results indicate that use of TMP-SMX is likely to affect LBP, which is an important biomarker of MT. This should be considered when studies on the effects of ART on MT in HIV-1-infected patients are analyzed. Additionally, an I-FABP elevation was detected after 1 month of ART, with a possible link to tenofovir. Our pilot study warrants large-scaled studies to further clarify whether HIV-1 drugs affect the level of enterocyte cell death and the influence of antibiotics on MT in HIV-1-infected patients.

Footnotes

Acknowledgments

We thank Professor Anders Sönnerborg for valuable comments. The study was supported by Stockholm's County Council (SLL-KI), Swedish Physician Against AIDS Research Fund, County Council of Gävleborg, and Swedish Society for Medical Research (SSMF for Piotr Nowak).

Parts of the data were presented at a poster session during the HIV Nordic Conference, October 2–3, 2014, Stockholm, Sweden.

Author Disclosure Statement

No competing financial interests exist.

References

Guadalupe

, Reay

, Sankaran

, et al.: Severe CD4+ T-cell depletion in gut lymphoid tissue during primary human immunodeficiency virus type 1 infection and substantial delay in restoration following highly active antiretroviral therapy. J Virol, 2003; 77(21):11708–11717.

Kotler

, Reka

, Borcich

, and Cronin

: Detection, localization, and quantitation of HIV-associated antigens in intestinal biopsies from patients with HIV. Am J Pathol, 1991; 139(4):823–830.

Brenchley

, Paiardini

, Knox

, et al.: Differential Th17 CD4 T-cell depletion in pathogenic and nonpathogenic lentiviral infections. Blood, 2008; 112(7):2826–2835.

Brenchley

, Price

, Schacker

, et al.: Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med, 2006; 12(12):1365–1371.

Lakatos

, Kiss

, Palatka

, et al.: Serum lipopolysaccharide-binding protein and soluble CD14 are markers of disease activity in patients with Crohn's disease. Inflamm Bowel Dis, 2011; 17(3):767–777.

Assimakopoulos

, Papageorgiou

, and Charonis

: Enterocytes' tight junctions: From molecules to diseases. World J Gastrointest Pathophysiol, 2011; 2(6):123–137.

Opal

, Scannon

, Vincent

, et al.: Relationship between plasma levels of lipopolysaccharide (LPS) and LPS-binding protein in patients with severe sepsis and septic shock. J Infect Dis, 1999; 180(5):1584–1589.

Troseid

, Nowak

, Nystrom

, et al.: Elevated plasma levels of lipopolysaccharide and high mobility group box-1 protein are associated with high viral load in HIV-1 infection: Reduction by 2-year antiretroviral therapy. AIDS, 2010; 24(11):1733–1737.

Merlini

, Bai

, Bellistri

, et al.: Evidence for polymicrobic flora translocating in peripheral blood of HIV-infected patients with poor immune response to antiretroviral therapy. PLoS One, 2011; 6(4):e18580.

10.

Vesterbacka

, Nowak

, Barqasho

, et al.: Kinetics of microbial translocation markers in patients on efavirenz or lopinavir/r based antiretroviral therapy. PLoS One, 2013; 8(1):e55038.

11.

Svard

, Sonnerborg

, Vondracek

, et al.: On the usefulness of circulating bacterial 16S rDNA as a marker of microbial translocation in HIV-1-infected patients. J Acquir Immune Defic Syndr, 2014; 66(4):e87–89.

12.

Nowak

, Abdurahman

, Lindkvist

, et al.: Impact of HMGB1/TLR ligand complexes on HIV-1 replication: Possible role for flagellin during HIV-1 infection. Int J Microbiol, 2012; 2012:263836.

13.

Sandler

and Douek

: Microbial translocation in HIV infection: Causes, consequences and treatment opportunities. Nat Rev Microbiol, 2012; 10(9):655–666.

14.

Lichtfuss

, Hoy

, Rajasuriar

, et al.: Biomarkers of immune dysfunction following combination antiretroviral therapy for HIV infection. Biomark Med, 2011; 5(2):171–186.

15.

Grube

, Cochane

, Ye

, et al.: Lipopolysaccharide binding protein expression in primary human hepatocytes and HepG2 hepatoma cells. J Biol Chem, 1994; 269(11):8477–8482.

16.

Abdurahman

, Barqasho

, Nowak

, et al.: Pattern of microbial translocation in patients living with HIV-1 from Vietnam, Ethiopia and Sweden. J Int AIDS Soc, 2014; 17:18841.

17.

Pelsers

, Hermens

, and Glatz

: Fatty acid-binding proteins as plasma markers of tissue injury. Clin Chim Acta, 2005; 352(1–2):15–35.

18.

Jakobsson

, Jernberg

, Andersson

, et al.: Short-term antibiotic treatment has differing long-term impacts on the human throat and gut microbiome. PLoS One, 2010; 5(3):e9836.

19.

Vrieze

, Out

, Fuentes

, et al.: Impact of oral vancomycin on gut microbiota, bile acid metabolism, and insulin sensitivity. J Hepatol, 2014; 60:824–831.

20.

Kalambokis

and Tsianos

: Rifaximin reduces endotoxemia and improves liver function and disease severity in patients with decompensated cirrhosis. Hepatology, 2012; 55(2):655–656.

21.

Cassol

, Malfeld

, Mahasha

, et al.: Persistent microbial translocation and immune activation in HIV-1-infected South Africans receiving combination antiretroviral therapy. J Infect Dis, 2010; 202(5):723–733.

22.

Vujkovic-Cvijin

, Dunham

, Iwai

, et al.: Dysbiosis of the gut microbiota is associated with HIV disease progression and tryptophan catabolism. Sci Transl Med, 2013; 5(193):193ra191.

23.

Nowak P

, Avershina

, Noyan

, et al.: Decreased Diversity of Gut Microbiota Is Associated with Immune Status during HIV-1 Infection. Abstract Conference on Retroviruses and Opportunistic Infections (CROI) 2014. 2014.

24.

Dethlefsen

and Relman

: Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation. Proc Natl Acad Sci USA, 2011; 108(Suppl 1):4554–4561.

25.

Ipp

, Zemlin

, Glashoff

, et al.: Serum adenosine deaminase and total immunoglobulin G correlate with markers of immune activation and inversely with CD4 counts in asymptomatic, treatment-naive HIV infection. J Clin Immunol, 2013; 33(3):605–612.

26.

Marchetti

, Merlini

, Sinigaglia

, et al.: Immune reconstitution in HIV+ subjects on lopinavir/ritonavir-based HAART according to the severity of pre-therapy CD4+. Curr HIV Res, 2012; 10(7):597–605.

27.

Di Mascio

, Srinivasula

, Bhattacharjee

, et al.: Antiretroviral tissue kinetics: In vivo imaging using positron emission tomography. Antimicrob Agents Chemother, 2009; 53(10):4086–4095.

28.

Goovaerts

, Jennes

, Massinga-Loembe

, et al.: LPS-binding protein and IL-6 mark paradoxical tuberculosis immune reconstitution inflammatory syndrome in HIV patients. PLoS One, 2013; 8(11):e81856.

29.

Wallet

, Rodriguez

, Yin

, et al.: Microbial translocation induces persistent macrophage activation unrelated to HIV-1 levels or T-cell activation following therapy. AIDS, 2010; 24(9):1281–1290.

30.

Chevalier

, Petitjean

, Dunyach-Remy

, et al.: The Th17/Treg ratio, IL-1RA and sCD14 levels in primary HIV infection predict the T-cell activation set point in the absence of systemic microbial translocation. PLoS Pathog, 2013; 9(6):e1003453.

31.

Pallikkuth

, Fischl

, and Pahwa

: Combination antiretroviral therapy with raltegravir leads to rapid immunologic reconstitution in treatment-naive patients with chronic HIV infection. J Infect Dis, 2013; 208(10):1613–1623.

32.

Church

, Fitzgerald

, Walker

, et al.: The expanding role of co-trimoxazole in developing countries. Lancet Infect Dis, 2015; 15(3):327–339.

Effects of Co-Trimoxazole on Microbial Translocation in HIV-1-Infected Patients Initiating Antiretroviral Therapy

Abstract

Introduction

Materials and Methods

Subjects

CD4+ T cells and plasma HIV-1 RNA

Quantification of microbial translocation markers

Statistical analysis

Results

CD4+ T cells and HIV-1 RNA

Lack of change in LPS levels during 1 year of ART

Decrease of LBP levels after 1 year in TMP-SMX-treated patients

Levels of sCD14 were reduced by ART

Elevated levels of I-FABP associated with use of tenofovir

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

CD4⁺ T cells and plasma HIV-1 RNA

CD4⁺ T cells and HIV-1 RNA