Distinct gut microbiome features characterize Fasciola hepatica infection and predict triclabendazole treatment outcomes in Peruvian patients

Abstract

Background: Fasciola hepatica, a globally distributed helminth, causes fasciolosis, a disease with significant health and economic impacts. Variability in triclabendazole (TCBZ) efficacy and emerging resistance are remaining challenges. Evidence suggests that the gut microbiome influences host-helminth interactions and is associated with anthelmintic effects, but its association with human F. hepatica infection and TCBZ efficacy is not well understood.

Methods: In this study, we investigated the relationship between Fasciola hepatica infection and the gut microbiome through metagenomic shotgun sequencing of 30 infected and 60 age- and sex-matched uninfected individuals from Peru. Additionally, we performed a longitudinal analysis to evaluate microbiome dynamics in relation to TCBZ treatment response.

Results and discussion: Infection was associated with specific microbial taxonomic and functional features, including higher abundance of Negativibacillus sp900547015, Blautia A sp000285855, and Prevotella sp002299635 species, and enrichment of microbial pathways linked to survival under stress and depletion of pathways for microbial growth. Unexpectedly, we identified that responders to TCBZ treatment (who cleared infection) harbored many microbiome features significantly different relative to non-responders, both before and after treatment. Specifically, the microbiomes of responders had a higher abundance Firmicutes A and Bacteroides species as well as phospholipid synthesis and glucuronidation pathways, while non-responders had higher abundance of Actinobacteria species including several from the Parolsenella and Bifidobacterium genera, and Bifidobacterium shunt and amino acid biosynthesis pathways.

Conclusions: Our findings underscore the impact of helminth infection on gut microbiome and suggest a potential role of gut microbiota in modulating TCBZ efficacy, offering novel insights into F. hepatica-microbiome interactions and paving the way for microbiome-informed treatment approaches.

Keywords: Fasciola hepatica; intestinal microbiome; liver fluke; longitudinal study; metagenomic shotgun sequencing; treatment response; triclabendazole.

Figure 1

Schematic overview of study design. (A) This study includes 90 Peruvian subjects: 30 infected with Fasciola hepatica and 60 uninfected, age- and sex-matched controls. The 30 infected subjects were treated with triclabendazole (TCBZ), resulting in 13 responders, 11 non-responders, and 6 lost to follow-up. (B) Samples were collected from six different locations from the Cusco highlands in Peru. The number next to each region indicates the number of samples collected at that location, and the size of each circle is proportional to the sample count. Map data collected from Google Maps using the ‘ggmap’ R package (v4.0.0, in 2024).

Figure 2

Comparisons of beta-diversity by infection status, and treatment response. Microbiome beta diversity comparisons grouped by (A) pre-treatment infection status, (B) infected pre-treatment responders and non-responders, and (C) post-treatment responders and non-responders. Microbiome composition was compared using non-metric multidimensional scaling (NMDS) analysis based on Bray-Curtis distances. Statistical significance between groups was calculated using the permutational multivariate analysis of variance (PERMANOVA) test.

Figure 3

Within-sample (alpha) diversity comparisons by infection status and treatment response. Differences in alpha diversity were calculated between samples grouped by (A) pre-treatment infection status, and (B) responders and non-responders before and after TCBZ treatment. Alpha diversity index was calculated based on Faith’s phylogenic diversity index. The box plot displays the median, 25th to 75th percentiles, with whiskers extending from min to max. Statistical significance between groups was calculated using Welch’s t-tests.

Figure 4

Comparison of between-sample (beta) diversity using Bray-Curtis dissimilarities within and between groups. Beta diversity within and between groups was compared using the Bray-Curtis dissimilarity matrix, grouped by (A) infection status and treatment response, (B) before treatment, and (C) after treatment. The violin plots display the median (dashed line), 25th and 75th percentiles (dotted lines), with whiskers extending from the minimum to maximum values. Data distribution is visualized with wider sections (indicating higher data density) or narrower sections (indicating lower density). Statistical significance between groups was determined using Welch’s t-test.

Figure 5

Differential microbial features by F hepatica infection status. (A) Microbial species or (B) metabolic pathways significantly differentially abundant between F hepatica-infected patients and uninfected controls at both timepoints, based on LEfSe analysis. Features with LDA score 2 and P-value < 0.05 are ranked by -Log of P-value, and normalized expression values (copies per million) were transformed to z-scores to show the abundance of each feature in each sample (1 sample per column). Additionally, the -Log of P-value from LEfSe and the mean decrease of accuracy (MDA) value from random forest analysis are presented as bar graphs. SPW, super pathway.

Figure 6

Differential microbial features by treatment response before and after treatment. (A) Counts of microbial species and pathways significantly associated with responder samples or non-responder samples, both before and after treatment. (B) Microbial species or (C) metabolic pathways significantly differentially abundant between responders and non-responders at both timepoints, based on LEfSe analysis. Features with LDA score 2 and P-value < 0.05 are ranked by -Log of P-value, and normalized expression values (copies per million) were transformed to z-scores to show the abundance of each feature in each sample (1 sample per column). Additionally, the -Log of P-value from LEfSe and the mean decrease of accuracy (MDA) value from random forest analysis are presented as bar graphs. SPW: super pathway.