Impact of intestinal parasitic infections on gut epithelial barrier and inflammation among foreign-born persons living with HIV

Introduction

Intestinal parasitic infections are often asymptomatic yet can cause chronic activation of immune responses in the human host^1–3. Some parasites, such as Strongyloides stercoralis, can continuously autoinfect the host for the entire lifetime if left untreated^4,5. Although treatable, intestinal parasitic infections are underdiagnosed due to lack of testing and low sensitivity of stool microscopy^6,7. It is known that both HIV and enteric parasitic infection cause gut damage and increased microbial translocation^8–10. Still, there are gaps in knowledge regarding the effects of simultaneous HIV and enteric parasitic infections on enterocyte turnover, microbial translocation, and systemic inflammation.

Parasitic infections increase CCR5⁺ CD4⁺ T-cell frequency, creating more HIV target cells in persons living with HIV (PWH)^10,11. Parasitic infections also increase CD8⁺ T-cell activation¹. Malnutrition, weight loss, chronic diarrhea, and changed immune responses caused by intestinal parasitic infections may further accelerate HIV/AIDS progression^12–16. Reciprocally, HIV infection may impair the immune system’s ability to defend against parasitic infections^17,18. Soluble CD14 (sCD14) is a marker of microbial translocation released upon LPS stimulation¹⁹. Both sCD14 and soluble CD163 (sCD163) are markers of macrophage and monocyte activation^19–21. Intestinal fatty acid binding protein (I-FABP) is a marker of intestinal permeability and enterocyte turnover²⁰. Increased levels of each of these biomarkers have been shown to predict morbidity and mortality in treated HIV infection, including cardiovascular disease and neurocognitive decline^19–25.

We conducted a single-center, prospective cohort study to evaluate the impact of intestinal parasitic infections on microbial translocation (sCD14), systemic inflammation (sCD163), and enterocyte damage (I-FABP), and consequently, the potential for impacting morbidity and mortality of PWH.

Materials and Methods

Participants were enrolled during a routine clinic visit at Thomas Street Health Center (TSHC), Harris Health System, the outpatient HIV center affiliated with Baylor College of Medicine and University of Texas Health Sciences Center-Houston. Inclusion criteria were as follows: age >18 years; foreign-born; documented HIV-1 infection; on antiretroviral therapy (ART) for >1 year; plasma HIV RNA < 20 copies/ml for ≥6 months (1 blip of < 200 copies/ml was permitted); and willingness to be treated with antiparasitic medication if indicated. Exclusion criteria were: inflammatory bowel disease; malignancy of the gastrointestinal tract; major intestinal surgery in the last 2 years; antiparasitic therapy in the last 1 year; pregnancy, plans to become pregnant, or breastfeeding; use of antibiotics in the last 60 days; use of probiotics and prebiotics (supplements and products) for more than three consecutive days within the 60 days prior to screening (yogurt with live cultures allowed). Ethical approval was obtained from Baylor College of Medicine and the University of Texas Health Sciences Center Institutional Review Boards. Administrative approval was granted by the Harris Health System Office of Research (H-37138). Written informed consent was obtained before any study procedures began.

Approximately 30g of stool and 20 mL of blood were collected from participants. A recombinant NIE-antigen ELISA was utilized to detect Strongyloides stercoralis infection, including positive and negative controls and a standard curve to quantify anti-Strongyloides-specific IgG⁷.

Multi-parallel real-time quantitative PCR (qPCR) was performed on stool to diagnose infection caused by Ancylostoma duodenale, Ascaris lumbricoides, Blastocystis species, Cryptosporidium hominis/parvum, Cystoisospora belli, Entamoeba histolytica, Giardia intestinalis, Necator americanus, Trichuris trichiura, Schistosoma spp, and Strongyloides stercoralis. Multi-parallel qPCR was performed as previously described²⁶ with the same species-specific primers and FAM-labeled minor groove binder probes (Applied Biosystems, Foster City, CA). Samples were run on an ABI ViiA7 PCR in Houston, Texas, with default parameters and 40 cycles. DNA concentrations were calculated using a standard curve.

The biomarkers sCD14 and sCD163 were measured by Quantikine ELISA kit (R&D systems, MN), and I-FABP was measured by DuoSet ELISA kit (R&D systems, MN) according to the manufacturer’s instructions. The plasma samples were diluted as needed by the methods protocol.

We utilized stool quantitative PCR (qPCR) testing and Strongyloides stercoralis recombinant IgG ELISA (Strongy IgG) for sensitive diagnosis of parasitic infection to compare levels of sCD14, sCD163, and I-FABP to those without parasitic coinfection.

Results of these studies (qPCR and Strongy IgG) were reported to the participant’s clinician, who considered treatment according to the clinical condition of the patient and the local standard of care, as well as discussed risks and benefits with the participant. A medical chart review confirmed that each participant with positive Strongy IgG was treated with ivermectin. A follow-up evaluation of 6 months post-treatment was scheduled for each Strongy IgG positive participant, which consisted of a clinical evaluation and repeat blood samples for sCD14, sCD163, and I-FABP level testing.

Statistical analyses were performed with R²⁷ and Prism 10.3.1 (GraphPad, Boston, MA). Correlations were estimated by calculating Spearman rank correlation. Mann-Whitney u-tests and Wilcoxon signed-rank tests were used to compare biomarker concentrations in parasite-infected and uninfected groups, pre-and post-treatment, respectively.

Results

Fifty-two participants living with HIV born in 14 different countries were enrolled in February and March of 2019. Thirty-four of the 52 (65%) were male, and the median age of participants was 50 years (IQR 43-57). Participants had a median of 19 years living in the US (Table 1). Median CD4⁺ T-cell count was 464 cells/mm³ (IQR 325-562). Plasma samples were obtained from all participants; stool samples were provided by 45 of the 52 participants. Fourteen (27%) participants tested positive for infection with Strongyloides stercoralis by Strongy IgG. Seven (15.5%) of the 45 participants who provided stool tested positive for pathogenic parasitic infection by stool qPCR: four (8.9%) with Trichuris trichiura (whipworm), one (2.2%) with Necator americanus (hookworm), one participant (2.2%) with Strongyloides stercoralis, and one (2.2%) with the protozoan infection Entamoeba histolytica (Supplemental Figure 1). Three participants (7%) were coinfected with more than one parasitic infection (either Strongy IgG positive and stool PCR positive or stool PCR positive for two parasites). Of note, 37 (82.2%) tested positive for Blastocystis species, which for this study was considered non-pathogenic. All participants were negative for Ancylostoma duodenale, Ascaris lumbricoides, Cryptosporidium hominis/parvum, Giardia intestinalis, and Schistosoma spp. Seventeen of 52 (32.7%%) participants were positive for either Strongy IgG or stool qPCR (not including Blastocystis infection). One participant was positive for Strongy IgG and 3 parasites by stool qPCR-Necator americanus (hookworm), Strongyloides stercoralis, and Trichuris trichiura (whipworm). CD4⁺ T-cell counts were not significantly different between participants who tested positive for intestinal parasitic infection and those negative for parasitic infection (431 cells/mm³ [IQR 305-525] versus 492 cells/mm³ [IQR 320-655], respectively; p = 0.50). The median sCD14 level in (+) Strongy IgG participants was significantly higher than in (-) Strongy IgG participants (1.69 μg/ml [IQR 1.51-1.88)] versus 1.48 μg/ml [IQR 1.31-1.67], p = 0.03). The median sCD14 level was not significantly different between participants with or without parasitic infection (by Strongy IgG or stool PCR), not including Blastocystis: 1.69 μg/ml (IQR 1.23-2.51) compared to 1.38 μg/ml (0.78-4.22; p = 0.07) (Figure 1). The median sCD163 level did not differ between (+) Strongy IgG vs (-) Strongy IgG participants (674 ng/ml [IQR 551-743] versus 583 ng/ml [IQR 480-760], p = 0.43). Similarly, the median sCD163 level did not differ between participants with parasitic infection (not including Blastocystis) versus parasite-negative participants (663 ng/ml [IQR 573-778] versus 569 ng/ml [471-721]; p = 0.20). I-FABP levels were not significantly different between (+) Strongy IgG and (-) Strongy IgG participants (832 pg/ml [IQR 380-1513] versus 921 pg/ml [IQR 362-2283], p = 0.98) or parasite-positive and parasite-negative participants (825 pg/ml [IQR 374-1898] versus 898 pg/ml [IQR 342-2214], p = 0.76). In a sensitivity analysis, participants positive and negative for parasites, including Blastocystis species, did not have statistically significant differences in any of the three biomarker levels.

• Download figure
• Open in new tab

Figure 1. Soluble CD14 levels in those with and without parasites.

Participants with strongyloidiasis had higher sCD14 levels (µg/ml) compared to participants without strongyloidiasis (1.69 vs. 1.48 µg/ml, respectively) (A). There was a higher sCD14 level (µg/ml) in individuals with either positive Strongyloides serology or positive qPCR for parasites not including Blastocystis (1.69 vs. 1.51 µg/ml, respectively) although not statistically significant (B).

*Strongyloides positive indicates positive by Strongyloides IgG serology.

**Parasite-positive indicates positive by either Strongyloides IgG serology or positive by stool PCR for pathogenic parasites

Ten of the 14 participants (71%) positive by Strongy IgG on the initial visit were evaluated 6 months after parasitic infection treatment with post-treatment biomarker levels. Four of the 14 (29%) were lost to follow-up or unable to return because of restrictions due to the COVID-19 pandemic. Pre-treatment and post-treatment levels of sCD14 (1.69 vs. 1.735 µg/ml, p = 0.829), sCD163 (673 vs. 688 ng/ml, p = 0.752), and I-FABP (832.1 vs. 1418 pg/ml, p = 0.37) were not significantly different.

Logistical regression was used to detect predictors of Strongy IgG positivity. The number of years living in the United States, country of origin, years since visit to home country, and baseline absolute eosinophil count were not associated with test positivity.

Discussion

Both HIV and parasitic infections are known to cause intestinal damage and affect inflammatory biomarkers, but the effects of HIV and parasite co-infection remain understudied. Our study provides a meaningful contribution to the literature because we evaluated levels of sCD14, sCD163, and I-FABP in PWH on suppressive ART in the setting of comprehensive testing for parasitic infections. We found that the median sCD14 level in PWH with (+) Strongy IgG was significantly higher than in PWH with (-) Strongy IgG. Given that sCD14 is an independent predictor of mortality in PWH¹⁹, the increased levels of sCD14 seen in participants of our study with parasitic infection raise concern about the long-term effects of enteric parasitic infection. The median sCD163 and I-FABP levels did not differ significantly between PWH with and without parasitic infections.

Our study detected a prevalence of 38% for pathogenic parasitic infection, demonstrating that intestinal parasitic coinfection is common in foreign-born PWH in the US. Most of these (27%) were due to infection with Strongyloides stercoralis detected by recombinant Strongyloides IgG ELISA. Strongyloides stercoralis infection is an auto-infectious helminth^4,5. Thus, participants in our study could have been infected years before participation in our study. The Strongyloides adult worm releases only a few ova per day, making diagnosis difficult by stool microscopy and qPCR. A positive Strongy IgG can be considered a test for active infection in an untreated host⁷. Prior studies have shown that parasitic coinfection can negatively impact CD4 counts^28–31 and plasma HIV RNA levels^32,33, but most studies have evaluated PWH, not on suppressive ART. PWH in Southern Ethiopia not taking ART who were treated for Ascaris lumbricoides had a significant increase in CD4⁺ T-cells from 469 to 551 cells/mm³ at 6 months, compared to a non-significant decrease in helminth-uninfected participants³⁴. Similar results were observed in a randomized, double-blind, placebo-controlled trial in ART-naive Kenyans treated for Ascaris lumbricoides³⁵. PWH coinfected with hookworm diagnosed by stool qPCR in Uganda had significantly lower CD4⁺ T-cell counts, suggesting that intestinal parasitic infections could contribute to the low CD4⁺ T-cell count and delayed recovery in resource-limited settings²⁸. Higher plasma HIV RNA levels were observed in pregnant women with hookworm or Trichuris trichiura infections³⁶. Deworming of T. trichiura or hookworm in an observational study increased CD4⁺ T-cell counts and decreased the percentage of pregnant women on ART with detectable HIV RNA levels³⁷.

In contrast to the studies above, a meta-analysis of randomized controlled trials of antiparasitic medications or placebo in PWH with unknown parasitic infection status showed no significant effect of one-time or repeated deworming on plasma HIV RNA levels or CD4⁺ T-cell count in people not taking ART³⁸. A retrospective, observational study reported empiric deworming did not increase CD4⁺ T-cell counts among Ugandans on ART except when restricted to women in the first year of ART³⁹. A study conducted in outpatient clinics in Lilongwe, Malawi, found no improvement in plasma HIV RNA levels after treatment of parasitic infection⁴⁰.

It is important to note that most of the studies above used stool microscopy to identify those with parasitic infections. Given the low sensitivity of stool microscopy, the cohort of participants identified as parasite-negative may have included several participants with unidentified parasitic infections. Additionally, studies of parasitic infections in PWH on suppressive ART are limited. With the studies above reporting the variable impact of parasitic infection on markers of HIV infection, our study measured three biomarkers as an indirect evaluation of parasitic infection on HIV infection in coinfected participants. Soluble CD14 is a marker of microbial translocation and LPS-bioactivity¹⁹; both sCD14 and sCD163 are markers of monocyte and macrophage activation^19–21. A case-control analysis of 74 PWH who died and 148 matched controls showed that participants with the highest quartile of soluble CD14 (sCD14) levels had a 6-fold increased risk of death compared to participants with the lowest quartile, even after adjusting for CD4⁺ T-cell count and other inflammatory markers¹⁹. The mechanism of this increase in mortality is not clear, but additional studies have shown an increase in cardiovascular disease and neurocognitive decline in patients with elevated biomarkers^20–23, supporting the hypothesis that prolonged inflammation is harmful to the host, even if asymptomatic.

A key limitation of our study is the small sample size. Additionally, participants in our study left their home country at variable numbers of years before enrollment in the study, had various parasitic infections, and had blood, but no stool samples were available from some participants, which could have contributed to sampling error. Levels of sCD163 were higher in (+) Strongy IgG participants compared to (-) Strongy IgG participants and higher in participants positive by stool qPCR compared to participants negative by stool qPCR, but neither reached statistical significance. A study with a larger sample size would probably be needed further to study sCD163 levels in PWH with intestinal parasitic infections. We did not see a decrease in any of the three biomarkers measured following treatment of intestinal parasitic infection. This finding may have been related to our small sample size at follow-up visits, short-term follow-up, or true lack of improvement of LPS-bioactivity or change in monocyte and macrophage activation following treatment. Further studies are needed to determine the impact of treatment on these biomarkers. Another limitation is that we only had post-treatment biomarker results from the positive participants for Strongy IgG.

It is important to note that our study benefited from highly sensitive diagnostic tests that are not readily available outside the research setting. In most resource-limited clinical settings, the diagnosis of intestinal parasitic infection relies on poorly sensitive stool microscopy or empiric treatment is given based on clinical signs and symptoms. If additional studies show similar results with elevated levels of sCD14 and sCD163, or if a study with a larger post-treatment cohort shows improvement of biomarkers following treatment, then the clinical question of routine empiric deworming in PWH in parasite-endemic regions should be reconsidered. Given that sCD14 is an independent predictor of mortality in PWH, our study’s increased levels of sCD14 seen in PWH with parasitic infection raise concern about the long-term effects of enteric parasitic infection. Further studies are needed to determine the clinical and virological consequences of intestinal parasitic infections in PWH.

Financial Support

RM received funding from the Texas Developmental Center for AIDS Research (NIH 5P30AI161943-04).

Disclosures regarding real or perceived conflicts of interest

None

Co-Author Contact Information

1. Melissa Reimer-McAtee

   Division of Infectious Diseases, Department of Internal Medicine, University of Texas Health Sciences Center, Houston, TX, USA

   melissa_reimer-mcatee{at}med.unc.edu

2. Jose Serpa

   Division of Infectious Diseases, Department of Internal Medicine, University of Texas at Tyler, Tyler, TX, USA

   Jose.SerpaAlvarez{at}uttyler.edu

3. Casey McAtee

   Department of Pediatrics, Baylor College of Medicine, Houston, TX, USA

   Casey.McAtee{at}bcm.edu

4. Emma Ortega

   Department of Tropical Medicine and Infectious Disease, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA, USA

   emmaortega27{at}gmail.com

5. Anoma Somasunderam

   Division of Infectious Diseases, Department of Internal Medicine, University of Texas Health Sciences Center, Houston, TX, USA

   Anoma.Somasunderam{at}uth.tmc.edu

6. Roberto Arduino

   Division of Infectious Diseases, Department of Internal Medicine, University of Texas Health Sciences Center, Houston, TX, USA

   Roberto.C.Arduino{at}uth.tmc.edu

7. Rojelio Mejia

   Department of Pediatrics: Tropical Medicine, Baylor College of Medicine, Houston, TX, USA

8. Netanya S. Utay

   Division of Infectious Diseases and Geographic Medicine, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, TX, USA