(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Trichinella spiralis -induced immunomodulation signatures on gut microbiota and metabolic pathways in mice [1]
['Xi-Meng Sun', 'Department Of Medical Microbiology', 'Parasitology', 'School Of Basic Medical Sciences', 'Capital Medical University', 'Beijing', 'Chun-Yue Hao', 'An-Qi Wu', 'Ze-Ni Luo', 'Saeed El-Ashram']
Date: 2024-01
The hygiene hypothesis proposes that decreased exposure to infectious agents in developed countries may contribute to the development of allergic and autoimmune diseases. Trichinella spiralis, a parasitic roundworm, causes trichinellosis, also known as trichinosis, in humans. T. spiralis had many hosts, and almost any mammal could become infected. Adult worms lived in the small intestine, while the larvae lived in muscle cells of the same mammal. T. spiralis was a significant public health threat because it could cause severe illness and even death in humans who eat undercooked or raw meat containing the parasite. The complex interactions between gastrointestinal helminths, gut microbiota, and the host immune system present a challenge for researchers. Two groups of mice were infected with T. spiralis vs uninfected control, and the experiment was conducted over 60 days. The 16S rRNA gene sequences and untargeted LC/MS-based metabolomics of fecal and serum samples, respectively, from different stages of development of the Trichinella spiralis-mouse model, were examined in this study. Gut microbiota alterations and metabolic activity accompanied by parasite-induced immunomodulation were detected. The inflammation parameters of the duodenum (villus/crypt ratio, goblet cell number and size, and histological score) were involved in active inflammation and oxidative metabolite profiles. These profiles included increased biosynthesis of phenylalanine, tyrosine, and tryptophan while decreasing cholesterol metabolism and primary and secondary bile acid biosynthesis. These disrupted metabolisms adapted to infection stress during the enteral and parenteral phases and then return to homeostasis during the encapsulated phase. There was a shift from an abundance of Bacteroides in the parenteral phase to an abundance of probiotic Lactobacillus and Treg-associated-Clostridia in the encapsulated phase. Th2 immune response (IL-4/IL-5/IL-13), lamina propria Treg, and immune hyporesponsiveness metabolic pathways (decreased tropane, piperidine and pyridine alkaloid biosynthesis and biosynthesis of alkaloids derived from ornithine, lysine, and nicotinic acid) were all altered. These findings enhanced our understanding of gut microbiota and metabolic profiles of Trichinella -infected mice, which could be a driving force in parasite-shaping immune system maintenance.
The complex interactions between gastrointestinal helminths, gut microbiota, and the host immune system are unknown. Here, we studied stool and serum samples from Trichinella-infected mice at different stages of development by 16S rRNA gene sequencing and non-targeted LC/MS metabolomics. We found that there is abundant Bacteroides in the gut microbiota during the parenteral phase of Trichinella infection, and the encapsulated phase shifts to a large number of probiotics Lactobacillus and Treg-associated-Clostridia. Metabolic activities also change in the enteral and parenteral phases, such as abnormal cholesterol metabolism. It is accompanied by changes in the Th2 immune response, lamina propria Treg, and immune hypo-responsiveness metabolic pathways. These findings contribute to a better understanding of the complex interactions between gastrointestinal helminths, gut microbiota, and host immune system. And the system-level knowledge of parasite metabolism and gene regulation through multi-omics joint analysis can help identify new targets for anthelmintic drugs and develop metabolite-based diagnostic tools.
As a multicellular parasitic helminth, T. spiralis is an interesting infection model for studying the induction of type 2 responses and Treg cells in response to infection. During T. spiralis infection, the entire life cycle is completed in the same host, sequentially containing adult, newborn larval (NBL), and muscle larval (ML) stages. The release of NBLs from sexually mature adult worms in the intestine is followed by their migration and the eventual establishment of a chronic infection in which MLs are present in the host’s skeletal muscles [ 38 ]. Each of these life stages may have a distinct impact on the host’s immune response. These characteristics make the T. spiralis-mouse system a good model for studying host-parasite relationships. This study used the T. spiralis-mouse model to investigate how gut microbiota and metabolic pathways support immunomodulation signatures induced by helminths. An integrated experimental design investigated T. spiralis infection-induced type 2 immunity and lamina propria resident Treg cells. Using 16S rRNA gene sequence and gas chromatography/mass spectrometry (LC/MS)-based metabolomic analyses, we investigated the effect of T. spiralis infection on intestinal microbiota and serum metabolism of infected animals. A time-course experiment was undertaken to monitor mice’s changes in intestinal microbiota and the effects on metabolomic profiles. This is one of the first studies to integrate helminth immunomodulation, gut microbiome, and serum metabolic profiles in mice with a bioinformatic approach to identify inter-relationships between host, gut microbiota, and metabolism in mice.
While still in its infancy, the application of metabolomics to the study of helminth infections is rapidly emerging as a fertile approach to better understand host-parasite interaction, especially on the aspect of helminth-induced immunomodulation. Changes in the metabolomic profiles of host fluids and tissues in response to parasite infection are critical for understanding parasite pathogenesis and parasite-induced immune responses and show great promise for diagnostics [ 26 ]. Metabolomic analysis of host samples (plasma, urine, or stool) has been widely used to identify biomarkers of parasitic helminth infection [ 27 – 29 ], which can distinguish infected from uninfected animals, often early in the disease’s course, by using either abundance changes in the metabolomic profile or unique metabolites to infected animals. There is a shared finding among these studies of significant changes in metabolites produced by the intestinal flora following infection with intestinal and extra-intestinal helminths. This shows that changes in the intestinal flora’s composition can significantly affect the host’s metabolite profile [ 29 , 30 ]. It has been reported that protozoan infections can alter host metabolism. These studies provided novel insights into disease pathogenesis and infection responses [ 31 , 32 ]. The urine and plasma samples of Plasmodium berghei-infected mice showed an upregulation of glycolysis, increased energy demand, and disruption of the gut microbiota [ 33 – 35 ]. Many of the intermediates involved in amino acid and energy metabolism were altered in mouse sera and brain samples after infection with Toxoplasma gondii [ 36 , 37 ].
Accumulating evidence supports the hypothesis that gastrointestinal (GI) helminth infections are associated with quantitative and qualitative modifications in the gut commensal microbiota with different downstream effects on parasite-mediated immune modulation, which can influence immune regulation and control of harmful inflammation [ 14 – 20 ]. The immune-molecular mechanisms underlying host-parasite–microbiota interactions remain elusive. Infections with GI helminths have been shown to affect the composition of the host gut microbiota and the relative abundance of individual microbial species, with likely downstream effects on host immunity and metabolic potential. Infections with gastrointestinal helminths have been shown to alter the gut microbiota composition and the relative abundance of individual microbial species [ 16 ]. These changes may impact host immunity and metabolism [ 17 ]. Fecal microbiota transplants from Heligmosomoides polygyrus-infected to naïve mice were enough to induce activation of regulatory T cells (Tregs) in the naïve mice and modulate host responses to concomitant diseases, ultimately affecting their clinical outcome [ 19 , 21 ]. Trichuris muris infections directly affected the composition of the caecal gut flora, comprising expanded populations of Firmicutes, Actinobacteria, Deferribacteres, and Proteobacteria, and a concomitant reduction in Bacteroidales [ 22 ]. In addition, selected populations of gut microbes may indirectly assist the establishment, or the elimination, of selected helminth parasites via immune-mediated events. Administration of Lactobacillus species to mice before experimental infection with H. polygyrus resulted in substantially increased worm burdens, which were mediated by bacterial-induced expansions of populations of Treg cells [ 23 ]. Recent studies had shown that parasite-elicited Th2 immune responses played a key role in helminth-associated qualitative and quantitative gut microbiota changes since such changes were not observed in Th2-deficient knock-out mice [ 24 , 25 ]. The host type 2 responses to parasite infection led to a depletion of the intestinal segmented filamentous bacteria (SFB), which reduced the expression of the pro-inflammatory cytokine IL-17 in the intestine [ 25 ].
The hygiene hypothesis assumes that the decreased incidence of infections with parasites and bacteria in developed countries may cause immune system deregulation and the consequent development of allergic and autoimmune diseases [ 1 ]. By neutralizing immunological pathways that would normally expel them and reset the limits of immunological response, helminths may effectively establish themselves inside the mammalian body. They also reduced responses to unrelated allergens and autoantigens, which may benefit the host [ 2 , 3 ]. Helminth-induced immunomodulation signatures on autoimmunity suppression fall into two categories: those that limit Th1 and Th17 responses through regulatory T cell (Treg) induction and those that activate helminth-specific type 2 responses (Th2). There is now a substantial body of experimental evidence supporting the role of gastrointestinal (GI) helminth parasites as immune regulators because of their ability to skew host immune responses toward Th2/Treg phenotypes [ 4 – 11 ]. Infection with helminths induces polarized type 2 immunity characterized by elevated expression of cytokines, including IL-4, IL-5, IL-10, and IL-13 [ 12 , 13 ].
A heat map of the Spearman correlation coefficient score and statistical significance were calculated by Package Vegan from the R program. It showed the associations between serum metabolites and host immune characteristics during T. spiralis infection. (A) The serum metabolites were analyzed in positive ionic mode. (B) The serum metabolites were analyzed in negative ionic mode. V.C represented villus height/crypt depth, num. GC represented the number of goblet cells per villus, GC. size represented the size of goblet cells, inf. Score represented the inflammatory score. The color of each cell represented the direction of the slope (red was positive, blue was negative). *P <0.05, **P <0.01, ***P <0.001.
We investigated the associations between serum metabolites and host immune characteristics using Spearman correlation test analysis during infection to elucidate the correlation between serum metabolites and immune characteristics during infection and further explore whether metabolism supports the outline of the host immune landscape ( Fig 12A and 12B ). With a P value of 0.05, we found a significant correlation between immunomodulation signatures (IL-4, IL-5, IL-13, IL-10, MLN Treg, and LP Treg) and duodenum pathological indicators (villus height/crypt depth, the number of goblet cells per villus, goblet cell size, and inflammatory score). The Th2/Treg immune response and duodenum tissue damage indicators for the potential biomarkers inosine, hypoxanthine, noradrenaline, salidroside, and indole all had significant positive slopes on days 6 and 15. While glycocholic, taurochenodeoxycholic acid, and 10,16-Dihydroxyhexadecanoic acid had significant negative slopes for Th2/Treg immune response and duodenum tissue damage indicators on days 6 and 15. In contrast, inosine, hypoxanthine, 1-piperidine, and xanthosine exhibited opposite patterns of association with the Th2/Treg immune response on day 30.
Receiver operator characteristic (ROC) was drawn with the true positive rate (sensitivity) as the ordinate and the false positive rate (1-specificity) as the abscissa. The obtained differential metabolites were evaluated by the ROC curve for potential biomarkers. The Area below the ROC Curve was known as the Area Under Curve (AUC), and the AUC was used to assess the sensitivity and specificity of the biomarker for predicting the occurrence of an event. The closer the AUC value is to 1, the higher the accuracy of the prediction. ROC analysis revealed up-regulated inosine, hypoxanthine, noradrenaline, salidroside, and indole and down-regulated glycocholic, taurochenodeoxycholic acid, and 10,16-dihydroxyhexadecanoic acid as having the highest sensitivity and specificity to detect T. spiralis infection on days 6 and 15 ( Fig 11A and 11B ). In contrast, on day 30, the down-regulated inosine, hypoxanthine, 1-piperidine, and xanthosine had the highest sensitivity and specificity for detecting T. spiralis infection ( Fig 11C ).
Kyoto Encyclopedia of Genes and Genomes (KEGG) was a database for systematic analysis of gene function and genomic information, which helped researchers to communicate genes and their expression information as a whole network. KEGG was a powerful tool for metabolism analysis and metabolic network research in vivo. DAVID v6.7 bioinformatic resources were used to predict the functions of T. spiralis infection-altered metabolic profiles and to identify potential therapeutic targets. Five and three pathways were significantly affected on days 6, 15, and 30, respectively (p<0.05 and enrichment ≥1). On day 60, there were no noticeable differences in KEGG pathways ( Fig 10 ). On day 6, KEGG pathway analysis revealed that primary and secondary bile acid biosynthesis (Map ID: 00120 and 00121) were the most significantly down-regulated metabolic pathways, while purine metabolism (Map ID: 00230) and tyrosine metabolism (Map ID: 00350) were the most significantly up-regulated metabolic pathways. By day 15, cholesterol metabolism (Map ID: 04979) was still down-regulated. On day 15, there was also an increase in protein digestion and absorption (Map ID: 04974) and phenylalanine, tyrosine, and tryptophan biosynthesis (Map ID: 00400). Meanwhile, there was a corresponding decrease in hyperimmune metabolic pathways, including purine metabolism (Map ID: 00230), tropane, piperidine, and pyridine alkaloid biosynthesis (Map ID: 00960), and the biosynthesis of alkaloids derived from ornithine, lysine, and nicotinic acid (Map ID: 01064) on day 30, which returned to homeostasis in the encapsulated phase.
Each point represented a metabolite. The horizontal coordinate represented the multiple change of the relative substances in the group (taking the logarithm of base 2). The vertical coordinate represented the P value of T test (taking the logarithm of base 10). The size of the scatter point represented the VIP value of the PLS-DA model. The larger the scatter point, the larger the VIP value. Significantly up-regulated metabolites were shown in red, significantly down-regulated metabolites were shown in blue, and non-significantly differentiated metabolites were shown in gray. (A) I6 vs. C6 group. (B) I15 vs. C15 group. (C) I30 vs. C30 group. (D) I60 vs. C60 group.
From the PLS-DA models, we used the Variable Importance in the Projection (VIP), combined with the P-value of T-test to find metabolites of differential expression. The threshold was set as VIP>1, Fold Change (FC) > 2.0 or FC < 0.5 and P value<0.05. We identified differentially produced compounds that included 184 metabolites in positive ionic mode and 176 metabolites in negative ionic mode on day 6, 257 metabolites in positive ionic mode and 272 metabolites in negative ionic mode on day 15, 88 metabolites in positive ionic mode and 131 metabolites in negative ionic mode on day 30, and 27 metabolites in positive ionic mode and 48 metabolites in negative ionic mode on day 60 ( Table 1 ). We visualized the results for differential metabolites in the form of a volcano plot. On day 6 following infection, 124 metabolites were up-regulated under the positive ionic mode, 60 were down-regulated, 88 were up-regulated, and 88 were down-regulated under the negative ionic mode as shown in the volcano plot. The number of different metabolites peaked on day 15, 136 metabolites were significantly up-regulated, 121 metabolites were significantly down-regulated under the positive ionic mode, 160 metabolites were significantly up-regulated, and 112 metabolites were significantly down-regulated under the negative ionic mode after infection. On day 30, 26 metabolites were significantly up, 62 were significantly down-regulated under positive ionic mode, 62 were significantly up-regulated, and 69 were significantly down-regulated under negative ionic mode. On day 60, only 11 metabolites were significantly up-regulated, and 16 were significantly down-regulated, under positive ionic mode. In contrast, 9 metabolites were significantly up-regulated, and 39 were significantly down-regulated. On day 60, only 11 metabolites were significantly up-regulated, and 16 were significantly down-regulated under positive ionic mode. In contrast, 9 metabolites were significantly up-regulated, 39 were significantly down-regulated (Figs 9 and S2 ). There was a significant difference between the infected group and the corresponding uninfected group at each time point, according to heat maps generated using the all-affected metabolites ( S3 Fig ).
Metabolomics was the simultaneous qualitative and quantitative analysis of all small molecule metabolites of an organism or cell during a specific physiological period. We identified the serum metabolites of the infected and uninfected mice at each time point and then generated untargeted metabolite profiles, which allowed them to identify and measure all of the metabolites in the samples, by Liquid Chromatograph Mass Spectrometer (LC-MS). We optimized the detection for both positive and negative ionic modes, which referred to the charge of the metabolites, considering the variable physicochemical properties of metabolites, such as molecular weight, polarity, solubility, and charge. In total, 3638 metabolites with positive and 2277 with negative ionic modes had been found. Because metabolome data had the characteristics of multi-dimension and high correlation among some variables, the analysis of those data required the use of multivariate statistical methods. Partial least squares discriminant analysis (PLS-DA), a commonly used method in metabolome analysis, could generally reflect the overall metabolic difference between the samples of each group and the variation degree within the group. Based on the PLS-DA models of metabolite profiling data, we found that the serum metabolites significantly differed between infected and control mice (Figs 8 and S1 ). The significant difference in metabolites suggested that the parasite infection impacted mouse metabolism.
A heat map of the Spearman correlation coefficient score and statistical significance was calculated by Package Vegan from the R program. It showed the associations between variations in the gut microbiota and host immune characteristics during T. spiralis infection. (A) The associations were analyzed on day 6 after infection with T. spiralis. (B) The associations were analyzed on day 15 after infection with T. spiralis. (C) The associations were analyzed on day 30 after infection with T. spiralis. (D) The associations were analyzed on day 60 after infection with T. spiralis. V.C represented villus height/crypt depth, num. GC represented the number of goblet cells per villus, GC. size represented the size of goblet cells, inf. Score represented the inflammatory score. Each cell’s color represents the slope’s direction (red is positive, blue is negative). *P <0.05, **P <0.01.
We explored the associations between variations in the gut microbiota and host immune characteristics using Spearman correlation test analysis during infection. Here, we showed that the microbiota communities significantly correlated with the immunomodulation signatures (IL-4, IL-5, IL-13, IL-10, MLN Treg, and LP Treg) and duodenum pathological indicators (villus height/crypt depth, the number of goblet cells per villus, the size of goblet cells, and the inflammatory score). Enterococcus and Staphylococcus had significantly negative slopes for Th2/Treg immune response and duodenum tissue damage indicators, excluding villus height/crypt depth, showing that their abundance was lower in infected mice on day 6( Fig 7A ). The highly prevalent genera Lactobacillus, Bacillus, Dubosiella, Hepatoplasma, Turicibacter, Arthromitus, and Staphylococcus also had significant negative slopes for Th2/Treg immune responses and duodenum tissue damage indicators as the newborn larvae circulation elicited systemic inflammation on day 15. For the genera Butyricimonas, Parabacteroides, Bacteroides, Mucispirillum, Rombousia, and Muribaculum, however, the slope of the Th2/Treg immune response and duodenum tissue damage indicators were strongly positive ( Fig 7B ). With the healing of tissue damage, bacteria such as Ruminococcaceae, Blautia, Lachnospiraceae, Ruminiclostridium, Clostridiales, Turicibacter, Romboutsia, Butyricicoccus, Anaerotruncus, Faecalibacterium, Roseburia, Desulfovibrio, Bifidobacterium, Alistipes, Lactobacillus, Harryflintia, Helicobacter, and Acetatifactor exhibited similar patterns of association with the Th2/Treg immune response, which indicated that these genera may be attempted to promote muscle larval colonization ( Fig 7C ). On day 60, we found that the genera Alistipes, Clostridiales, and Turicibacter had significant positive slopes for Treg ( Fig 7D ).
The quantitative microbiome analyses showed that the different amounts of Bacteroidales to Firmicutes in the infected and uninfected groups were statistically significant. Clearly, the major change was observed because of T. spiralis infection, with a linear discriminant analysis (LDA) effect size (log 10) of approximately 4. In the infected group, only Erysipelotrichia was reduced on day 6, Bacteroidales, Bacteroides vulgtus, Alloprevotella and prevotellaceae were more abundant on day 15, while Lactobacillus, Bacilli, Erysipelotrichia, Duboisella, Jeotgalicoccus and lactobacillus animals were reduced on day 15. In contrast, there was an increase in Clostridia, lactobacillus, Bacilli, Alistipes, Ruminococcaceae, Lachnospiraceae, and a decrease in Rikenellaceae, Muribaculaceae and Bacteroidales in infected mice on day 30 ( Fig 5B–5E ). Intriguingly, Clostridia and Ruminococcaceae continued to increase after day 60. Organizing these results into cladograms at the genus level provided an easily appreciated view of the microbiome shifts because of colonization with T. spiralis ( Fig 6 ).
The data showed the abundance of different bacterial groups in the infection and control groups at each time point. For the quantitative microbiome analyses, LEfSe method was used to identify significant bacterial groups between the infected and uninfected groups, which was performed by LEfSe software with a default LDA Score threshold of 4. (A) Community bar plot analysis was shown at the phylum level. N, represented the result of the samples collected before T. spiralis infection on day 0. (B) C6 group. (C) I60 group. (D) I15 vs. C15 group. (E) I30 vs. C30 group.
Community bar plot analysis ( Fig 5 ) showed that the gut microbiota community composition of the infected group differed significantly from the uninfected group (P = 0.235 on day 6; P = 0.001 on days 15 and 30; P = 0.014 on day 60, P value from Adonis analysis). The identified gut microbiota comprised 10 phyla. The most abundant bacterial group was the Bacteroidetes, followed by the Firmicutes, Proteobacteria, Actinobacteria, Deferribacteres, Chloroflexi, Verrucomicrobia, Acidobacteria, and Fusobacteria. The most notable differences were the relatively greater amounts of Bacteroidales and the fewer amounts of Firmicutes on day 6 and day 15, and they reversed these differences on day 30 and day 60. In particular, Bacteroidetes accounted for over 66% of the sequences in the infection group but only about 43% of the sequences in the control group on day 15. In contrast, T. spiralis colonization was associated with a 2.4-fold increase in the relative number of Firmicutes sequences on day 30 ( Fig 5A ). Additionally, T. spiralis colonization had similar effects on the Proteobacteria, shifting from a 1.7-fold increase on day 15 to a 2.7-fold decrease on day 60.
The experimental design of murine infection, as well as the taxonomic and α and β diversity of gut microbiomes on days 0, 6, 15, 30, and 60 after infection with T. spiralis, ten infected mice at each time point were sacrificed, and the feces were collected. The α diversity refers to the diversity within a sample, while β diversity refers to the diversity between samples. (A) The murine infection experimental design. (B) Comparison of Chao 1 index (an α diversity indicator used to estimate the total number of microbial species) between infection and control groups. (C) Comparison of observed species between two groups. (D) Comparison of the Shannon index (a measure of diversity considering the number and abundance of species in a sample) between two groups. (E) Comparison of Simpson index (a measure of diversity considering the number of species and their relative abundance) between two groups. (F) Unweighted pairwise UniFrac distances were averaged within each group to calculate an average diversity value (a proxy for inter-individual variation). (G) PCoA plot based on the unweighted UniFrac distance (to compare whether there were significant differences in microbial communities in specific evolutionary lineages of samples) of gut microbiota by positioning each sample from infected vs. uninfected group (I6 vs. C6 group, P = 0.3631; I15 vs. C15 group, P = 0.7912; I30 vs C30 group, P = 0; I60 vs. C60 group, P = 0). N, represented the result of the samples collected before T. spiralis infection on day 0. Median, dispersion degree, maximum, minimum, outliers and were shown as boxes. *P <0.05, **P <0.01, ***P <0.001. P values are from Wilcox test.
The study investigated the alterations in the gut microbiome of infected mice by sequencing the fecal 16S rRNA V4 region. The overall experimental design of the study was summarized in Fig 4A , which provided an overview of the methods used and the study’s timeline. We used various diversity measures to compare infected mice’s gut microbiota with healthy control mice at different time points (day 0, 6, 15, 30, and 60). The α diversity of the gut microbiota, which referred to the diversity within a sample, was not significantly different between the infected and uninfected groups on day 6 and day 60 ( Fig 4A–4G ). The Chao1 index, an α diversity indicator used to estimate the total number of microbial species, showed no significant difference at each time point (day 0, 6, 15, 30, and 60) ( Fig 4B ). However, on day 15, the infected group exhibited a higher Simpson index ( Fig 4E ), a measure of diversity considering the number of species and their relative abundance. On day 30, the infected group had higher values of observed species ( Fig 4C ) and a higher Shannon index ( Fig 4D ), which are also diversity measures considering the number and abundance of species in a sample. But the subtle difference from the Shannon index was that the Simpson index focused more on the relative abundance between different species, while the Shannon index focused more on species richness. To investigate the overall structure of the gut microbiota, we constructed a Principal Coordinate Analysis (PCoA) score plot based on weighted UniFrac distances. Weighted UniFrac distances was used to compare whether there were significant differences in microbial communities in specific evolutionary lineages of samples. Thus, PCoA measures the similarity between bacterial communities. The results showed that on days 30 and 60, the structure and composition of the microbiota differed significantly between the infected group and the uninfected group, indicating that the infection had a lasting effect on the gut microbiome. However, colonization with T. spiralis on days 6 and 15 did not significantly affect β diversity ( Fig 4F–4G ), which referred to the diversity between samples.
To monitor the Treg cells during T. spiralis infection, 48 Foxp3 EGFP mice were randomly assigned to infected (N = 24) and uninfected (N = 24) groups. Twenty-four Foxp3 EGFP mice were infected with 400 muscle larvae of T. spiralis via the gastrointestinal tract. At each time point, 6 infected or uninfected mice were sacrificed and the lamina propria of the colon were isolated. Foxp3 + regulatory T cells in the lamina propria of the colon were identified and analyzed by EGFP/CD4 immunohistochemistry. We stained CD4 with CD4-Cy3 antibody (red light) and nuclei with DAPI (blue light). Foxp3 in mice had its own EGFP green fluorescence, and an automatic slide scanner was used for fluorescence mode scanning. We found that infected colon lamina propria showed a higher accumulation of CD4 + Foxp3 + Treg cells, which were sustained by the encapsulated phase ( Fig 3 ).
On Days 6, 15, 30, and 60 after infection with T. spiralis, five infected mice at each time point were sacrificed, and MLNs were isolated. (A) The cells were sorted with CD3, CD4, CD25, and Foxp3 by FACS, and the percentage of CD25 + Foxp3 + cells in the CD4 + T cells population was shown. (B) The culture supernatants were collected after being incubated with α-CD3/CD28 for 48 h, and the concentrations of IL-4, IL-5, IL-10, IL-13, IL-2, IFN-γ, IL-17A, and IL-6 were measured by ELISA. Data are from three independent experiments expressed as mean SEM, with * P < 0.05, ** P < 0.01, and *** P< 0.001 compared to the uninfected group.
This study aimed to examine the T. spiralis infection-induced immune response by measuring the Th2 cell cytokine concentrations of IL-4, IL-5, IL-13, and IL-10 in response to restimulation with anti-CD3 and anti-CD28 mAbs, and the frequency of Treg cells by mesenteric lymph node (MLN) at each time point. The study found elevated IL-4, IL-5, IL-13, and IL-10 in infected mice on day 6, which remained until day 30 and returned to normal levels by day 60 ( Fig 2B ). The percentage of Treg cells in the mesenteric lymph nodes was also higher in infected mice, peaking on day 15 and returning to normal levels by day 60 ( Fig 2A ). In contrast, after restimulation with anti-CD3 and anti-CD28 mAbs, infection-induced mouse MLN cells secreted low levels of Th1 cell cytokines IL-2 and IFN-γ. This response only increased on day 15. We also found that the level of IL-17A decreased while IL-6 increased on day 6. T. spiralis infection results in a Type-2 immune response.
We collected duodenal tissue on days 6, 15, 30, and 60 after gavage. The pathological section results suggested that the duodenum of infected mice had mucosal lesions compared with uninfected mice ( Fig 1D ). Histological morphometry suggested a decreased ratio of villus height to crypt depth ( Fig 1E ), an increased number of goblet cells per villus ( Fig 1F ), goblet cell size ( Fig 1G ), and histological score ( Fig 1H ) in infected mice compared with uninfected mice. The duodenum was scored (0–3) for both inflammation and edema. These gut damage parameters of the duodenum indicated that the most severe pathological changes were observed in the enteral phase and were most often confined to the lamina propria and epithelium, resolved by day 30.
Wild type mice were randomly divided into infected and uninfected (control) groups. We administered 400 muscle larvae (ML) to mice via gavage (day 0) to quantify the effects of T. spiralis infection in mice. We then monitored the infection’s progressive effects on worm burden, weight loss, and duodenal pathology on day 6, 15, 30, and 60 after gavage. Based on the life cycle ( Fig 1A ), the larvae invaded the small bowel mucosa on day 6 after gavage, which developed into adult worms (AD). After incubation, approximately 70 adult worms entered the small intestine on day 6 of the enteral phase (day 6). In the parenteral phase (day 15), females released newborn larvae that reached the bloodstream and lymph drainage system. Approximately 15,000 larvae migrated to the striated muscles where they encyst (day 30), and the larval burden reached its peak in the encapsulated phase (day 60). ( Fig 1B ). These results were consistent with the life cycle of T. spiralis in mice. Slowly rising of body weight was a common symptom of a severe T. spiralis infection. Infected mice increased slowly in body weight compared to uninfected controls following worm burden ( Fig 1C ). These differences were consistent in all life stages, showing that T. spiralis infection impacts body weight increasing.
Discussion
Helminths have immunomodulatory mechanisms that can help trigger Th2 and Treg responses, and helminth infection can influence the Th1 and Th17 responses [39]. Intestinal helminths have been shown to cause major changes in the structure and function of the host gut microbiota [20,40–42]. Furthermore, the gut microbiome and metabolites are intricately connected, and alterations in the gut microbiota profoundly influence downstream metabolic pathways [43]. The evidence for a relationship between the gut microbiota and the metabolome is robust and promising [44–49]. The host immune system is also modulated by changes to the gut microbiota and metabolites, which affect the host immune regulatory network and, ultimately, helminth infection and parasitism [50,51]. The gut microbiota is important in determining host permissiveness for parasites [22,52]. The parasite manipulates host metabolism via insulin signaling and the gut microbiome [53,54]. Determining the effects of helminth infection-induced immune regulation characteristics on the intestinal microbiota and metabolic pathways in mice, as well as the connections between helminth-induced immune regulation characteristics and changes in mouse intestinal microbiota and metabolic pathways, are therefore crucial. It will contribute to a better understanding of the host-parasite interaction, define "worm therapy," clarify its mechanism and principle, and give a theoretical framework for the use and development of worm therapy for intestinal inflammatory diseases.
The interaction between helminth-induced immune regulation, gut microbiota, and metabolic pathways has yet to be investigated. To evaluate changes in immunological features, 10 T. spiralis-infected mice were collected at different time points (day 0, day 6, day 15, day 30, and day 60). To investigate the impact of Trichinella colonization on gut microbiota and serum metabolites and to study changes in gut microbiota or serum metabolites and immunological signatures, gut microbiota sequencing will be achieved in stool samples, and metabolomic characterization of serum samples will be performed. Understanding how helminth infection affects the gut microbiota or metabolic pathways by modifying immunity may help helminths cure intestinal diseases.
We investigated the changes in Trichinella’s physiological and immune characteristics at different developmental stages of mouse infection. Each C57BL/6 mouse was gavage with 400 Trichinella muscle larvae. The worm burden fluctuated according to developmental stage, and the body weight was always lower than normal mice. We also conducted duodenal tissue sections and active inflammatory parameters of the host at each developmental stage of T. spiralis. We found that the host’s intestinal lesions were more severe in the early intestinal stage of Trichinella infection (days 6 and 15). Trichinella causes damage to the host’s gastrointestinal tract, and changes in damage to the intestinal tract follow Trichinella’s developmental cycle. On day 6 of infection, the study found elevated IL-4, IL-5, IL-13, and IL-10 levels in infected mice. These levels remained elevated until day 30 and returned to normal by day 60. These findings aligned with the findings of previous studies which found the early immune response in the intestine during T. spiralis infection was mediated by cytokines and intestinal mucosal immune cells, including dendritic cells (DCs), eosinophils, goblet cells, and mast cells. Th2 cells produced cytokines that led to changes in these immune cells during T. spiralis infection. DCs were important antigen-presenting cells in the intestine and played a key role in regulating the mucosal immune response. T. spiralis infection could induce a strong Th2 response, which contributed to worm expulsion and goblet cell hyperplasia. Mastocytosis in the intestine during T. spiralis infection depended on the interplay between stem cell factor (SCF) and Th2-derived cytokines, such as IL-3, IL-4, and IL-9. These immune responses contribute to the host’s defense against T. spiralis infection [55–59]. Th1 cell cytokines IL-2 and IFN-γ showed a delayed increase, with levels only rising on day 15, in contrast to the Th2 cell cytokines. On day 6, we also noticed that IL-17A levels declined while IL-6 levels surged. T. spiralis infection had been shown to increase the recruitment of Treg cells, which could suppress IL-17A production and reduce inflammation [60].
Helminth infection can induce Th2 and Treg immune responses, which means that Th2 cytokines and Treg are significantly increased in mesenteric lymph nodes early in Trichinella infection and gradually return to normal levels over time. This is also consistent with Trichinella’s developmental cycle. T-helper type-2 (Th2) cells are activated during the complicated immune-mediated process of T. spiralis worm expulsion from the small intestine. T. spiralis larvae invade the intestine, and Th2-related cytokines increase immediately. IL-4 and IL-13 levels peak before the formation of full-muscle larvae [61]. This may be because of the early stage of T. spiralis infection, which damages the intestinal epithelium and the underlying tissue layer, which cause the body’s immune system to respond to regulate the immune system. Additionally, T. spiralis infection increased the number of small intestine mucosal immune cells and provided further evidence that the intestinal mucosal immune system of infected mice was induced towards a mixed Th1/Th2 phenotype, with a predominance of Th2 responses in the early stage of infection [62]. However, these finding were inconsistent with the findings of other studies, which suggested a different pattern of immune response during T. spiralis infection. According to these studies, the initial response during the intestinal stage was dominated by Th1 cytokines such as IFN-γ and IL-12, which were involved in activating immune cells to fight the parasite. However, during the muscle stage of infection, there was a shift towards a Th2 response, characterized by the production of cytokines such as IL-4, IL-10, and IL-13, which were involved in regulating inflammation and tissue repair [63,64]. Similarly, the first response during the intestinal phase was a moderate Th1 response, which was subsequently shifted to a powerful Th2 response. Th2 immunity was also a powerful muscle defense mechanism, and the regulatory arm of immunity was engaged to protect both the parasite and the host [65]. Notably, levels of Th1-type cytokines IL-2 and IFN-γ were elevated during the migratory stage of newborn larvae (day 15), which could be attributed to macrophage stimulation and induction the DCs maturation during larval migration. TNF-α and IFN-γ production during T. spiralis infection may result from macrophage activation [66]. Many studies have shown that Th1 cytokines, such as IL-12, INF-γ, IL-1β, and TNF-α are much more abundant in the early stages of intestinal infection by T. spiralis [63,67–69]. Inflammatory mediators, such as prostaglandin E2 (PGE2) and nitric oxide (NO), as well as the production of Th1 cytokines, such as, IL-12, INF-γ, IL-1β and TNF-α, were found to be produced in greater amounts during the intestinal phase of the T. spiralis infection, which was linked to an increased number of eosinophils and the development of intestinal pathology [64,70]. This suggested that T. spiralis infection induced an increase of small intestine mucosal immune cells and a mixed Th1/Th2 phenotype with a predominance of Th2 response at the early stage of infection.
Infection with T. spiralis induced a Type-2 immune response, leading to helminth expulsion. However, Treg cell levels in the mesenteric lymph nodes were elevated in infected mice, with a peak on day 15 and a return to normal levels by day 60. The increase in the frequency of Tregs helped the helminth persist. The number of Treg in the colonic lamina propria remained high in late-stage T. spiralis infection, showing that other factors in the gut regulate the immune system in response to T. spiralis damage to the intestinal lamina propria and epithelium. Tregs were vital in maintaining intestinal immunological homeostasis, particularly in the colon, raising the possibility that bacterially driven modulation of host immune responses might contribute to the altered worm burdens [71]. Helminth infections had been shown to increase the amount of colonic Tregs and Foxp3 expression, both of which were critical in maintaining the immunological balance in the colon [14,72]. On the same vein, T. spiralis infection can induce a polarized Th2 immune response, characterized by producing Th2 cytokines (IL-4, IL-10) and developing Treg cells [73,74].
Through 16S rRNA gene-based sequencing, this study examined the influence of Trichinella infection on gut microbiota throughout chronic infection. We have determined that helminth infection in female C57BL/6 considerably impacted the host microbiota. At days 6 and 60, there were no significant differences in the α-diversity of the gut microbiota between the infected and uninfected groups. Compared to the uninfected group, the infected group showed a higher α-diversity using the Simpson index on day 15, higher observed species values, and a higher α-diversity using the Shannon index on day 30. Our findings were inconsistent with a previous study that found a significant decrease in α-diversity on day 28 because of infection, which became more apparent by day 41 [75]. The β diversity of the microbiota differed significantly between the infection and uninfected groups at days 30 and 60. The difference in the β-diversity increased by day 41, showed by both DGGE and sequencing, confirmed the impact seen in our study assessing changes in cecal microbiota [75]. Several studies had found changes in the microbiota in the colons of pigs on day 21, which corresponded to the larval stages [76,77]. Helminth infection had resulted in a slight increase in diversity, with people carrying a range of helminth infections, including Necator americanus and Ascaris spp., making it difficult to determine cause and effect [78]. However, no significant alterations in microbiota were reported in a human investigation when N. americanus was the only soil-transmitted helminth [79]. There might be a variety of explanations for the observed variations between these experiments, and it might be partly because various parasites were studied, including mixed infections. Similarly, changes in sample preparation and DNA amplification for sequencing, subsequent sequence analysis, and reference database selection might account for some of these variances.
Bacteroidetes were the most abundant bacterial phylum, followed by Firmicutes, Proteobacteria, Actinobacteria, Deferribacteres, Chloroflexi, Verrucomicrobia, Acidobacteria, and Fusobacteria. The most obvious changes were the relatively higher numbers of Bacteroidales and lower numbers of Firmicutes on days 6 and 15, which were reversed on days 30 and 60. These findings coincided with the result that infection with T. spiralis decreased the ratio of Firmicutes to Bacteroidetes in the gut, which helps restore the previously increased ratio observed in high-fat diet-fed mice [80]. Bacteroidetes, in particular, accounted for over 66% of sequences in the infection group on day 15, compared to only about 43% in the uninfected group. At day 30, colonization by T. spiralis was associated with a 2.4-fold increase in the relative number of Firmicutes sequences. Additionally, colonization by T. spiralis had similar effects on Proteobacteria, which decreased from a 1.7-fold increase on day 15 to a 2.7-fold decrease on day 60. Our findings on how infection affected microbiota were consistent with previous research that found helminth infection to significantly impact the host microbiota in C57BL/6 mice [75]. During intestinal helminth infection, a complex relationship emerged between the worm, the host immune system, and the intestinal microbiome. The intestinal microbiome interacted with the host immune system to control helminth infection by shaping microbial communities and activating host immunoregulatory pathways [81]. The presence of a complex bacterial microbiota provides the host with resistance against intestinal helminths by regulating intestinal motility [82].
Changes in the gut microbiome and vice versa might affect the immune response to T. spiralis infection. Infection with T. spiralis induced a switch from elevated Bacteroidales in the parenteral phase to Treg-associated Clostridiales, Ruminococcaceae, Lachnospiraceae and Bacteroides in the encapsulated phase. T. spiralis induced the expansion of colon lamina propria Tregs, which might occur through indirect mechanisms involving the intestinal microbiome. In this respect, Bacteroidetes were gastrointestinal bacteria that broke down carbohydrates [83]. The anaerobic breakdown of carbohydrates released short-chain fatty acids (SCFAs), which were easily absorbed by the host and enable the host to obtain energy from sources that would otherwise be inaccessible by Bacteroidetes fermentation. Variations in the SCFA levels in the intestine could alter the lumen’s pH and impact the composition of the microbiota present [84]. The mice used in the study were given a diet rich in plant-based fiber that includes feed manufactured from wheat, barley, soy, and maize. The intestinal immune system had recently been shown to be significantly influenced by bacterially produced SCFA, which might also directly trigger CD4+ Foxp3+ Treg cells in the colon [85]. Helminth-regulation of inflammatory disease, demonstrating a crucial function for bacteria-derived SCFAs acting through ffar2 in H. diminuta-amelioration of colitis, the critical necessity of IL-10 that could up-regulate the expression of SCFA transporters/receptors, and butyrate regulation of IL-10 receptor expression [43]. However, the decrease in IL-17 production is correlated with an increase in CD4+ Foxp3+ Treg cells in the germ-free colon. Animals were reconstituted with a complex and diversified microbiota that lacked the major phyla. Bacteroidetes can’t restore the proper immunological balance, demonstrating how various organisms could affect the pro- and anti-inflammatory responses in the gut [86]. Similarly, T. spiralis induced the expansion of colon lamina propria Tregs through the release of IL-33, which led to the expansion of GATA3+ Tregs. Loss of DRA from colonocytes triggers the release of IL-33, which drove a type 2 immune response and specifically expanded a subset of GATA3+ Tregs [87]. The underlying mechanisms of action involved an IL-17/TNF-alpha synergistic reaction, suppression of Th1 and Th2 responses, and an upregulation of the regulatory cytokines IL-10 and TGF-β1 [88]. The microflora was required to accumulate cytokine-producing CD4+ T cells in the colon lamina propria, including Th17 cells that produce IL-17 [89]. The exact mechanisms by which T. spiralis achieved inflammation modulation and induced the expansion of colon lamina propria Tregs have not been fully elucidated yet [90].
Immunomodulation signatures (IL-4, IL-5, IL-13, IL-10, MLN Treg, and LP Treg) and duodenal pathological indicators (villus height/crypt depth, the number of goblet cells per villus, the size of goblet cells, and the inflammatory score) were substantially connected with bacterial communities. Except for villus height/crypt depth, Enterococcus and Staphylococcus displayed substantially negative slopes for Th2/Treg immune response and duodenal tissue damage markers, showing that their abundance was decreased in infected animals at day 6. As the newborn larvae circulation elicited systemic inflammation at day 15, the highly prevalent genera Lactobacillus, Bacillus, Dubosiella, Hepatoplasma, Turicibacter, Arthromitus, and Staphylococcus also had significant negative slopes for Th2/Treg immune responses and duodenum tissue damage indicators, showing that this microbial community was decreased by inflammation. However, the Th2/Treg immune response slope and duodenum tissue damage indicators were strongly positive for the genera Butyricimonas, Parabacteroides, Bacteroides, Mucispirillum, Rombousia, and Muribaculum, showing that this microbial community is likely to be strongly stimulated by inflammation. Turicibacter, Romboutsia, Butyricicoccus, Anaerotruncus, Faecalibacterium, Roseburia, Desulfovibrio, Bifidobacterium, Alistipes, Lactobacillus, Harryflintia, Helicobacter, and Acetatifactor showed similar patterns of association with the Th2/Treg immune response with the healing of tissue damage, suggesting that these genera may be attempts to promote muscle larval colonization. We observed substantial positive slopes for Treg in the taxa Alistipes, Clostridiales, and Turicibacter at day 60. These results showed that immunological characteristics in the GALT were correlated with infection-related changes in the abundance of certain microbial species. During primary infection, Th2 and duodenal mucosal inflammation were mounted against intestinal helminths to eliminate luminally dwelling worms, and subsequent Treg expansion was required for effective parasite establishment. This was followed by a shift from elevated Parabacteroides and Butyricimonas to elevated Lactobacillus, Turicibacter, Romboutsia, Faecalibacterium, Bifidobacterium, Alistipes, Harryflintia, Helicobacter, and Acetatifactor, raising the possibility of bacterially driven modulation of host immune responses during infection. In line with these findings, mice treated with Bifidobacteria infantis in a model of pathogen-induced inflammation experienced a reduction in intestinal inflammation, increased CD4+CD25+Treg cells, and decreased Th1 and Th17 responses [91]. Treg cells are significantly reduced in number and functionality in germ-free animals [92,93]. There was limited information on the relationship between immunomodulation signatures, duodenal pathological indicators, and bacterial communities. To better understand their relationship, further research was needed to explore the mechanisms and interactions between immunomodulation signatures, duodenal pathological indicators, and bacterial communities.
A non-targeted LC-MS/MS-based serum metabolomic study was conducted in mice to detect T. spiralis-specific alterations in metabolites and metabolic pathways, as well as possible biomarkers of Trichinella infection. Metabolism was one of the key components of the host-parasite relationship, which allowed parasites to successfully invade and infect their hosts. This host-parasite metabolic relationship was dynamically regulated, reflecting distinct immune-metabolic demanded at each stage of parasite differentiation. Primary and secondary bile acid biosynthesis was the most significantly down-regulated metabolic pathways on day 6 of the study, and purine metabolism and tyrosine metabolism were the most up-regulated metabolic pathways. Cholesterol metabolism was still down-regulated on day 15. Similarly, T. spiralis infection in mice resulted in alterations in several metabolites. The levels of total cholesterol were significantly reduced in mice infected with T. spiralis [80]. On day 15, there was also an increase in protein digestion and absorption and phenylalanine, tyrosine, and tryptophan production. There were no discernible changes in the KEGG pathways on day 60. Studies had shown that cholesterol accumulation had contributed to inflammation recently, while bile acids and purines control inflammation [94,95]. The activation of the NLRP3 inflammasome might be influenced by various lipids and their derivatives. Bile acids derived from cholesterol degradation could induce the ubiquitination of NLRP3 via the upregulation of protein kinase A (PKA) and inhibit its activation [96,97]. Most acute-phase proteins were rich in aromatic amino acids (phenylalanine, tyrosine, and tryptophan) during inflammation [98]. When glycolysis was inhibited, amino acids such as alanine or serine could also be converted to pyruvate, which was thought to affect NLRP3 activation [97]. Therefore, during the early stages of T. spiralis infection, these changes in metabolic pathways might develop a protective immune system for the host, followed by a balanced immune response to ensure muscle larvae colonization. On day 30, there was a corresponding decrease in hyperimmune metabolic pathways such as purine metabolism, tropane, piperidine, and pyridine alkaloid biosynthesis, and alkaloid biosynthesis from ornithine, lysine, and nicotinic acid. During the encapsulated phase, this decreased activity returned to homeostasis. These findings suggested that T. spiralis switched from an immune activation to an immune suppression metabolic pathway to colonize the host. On days 6 and 15, ROC analysis revealed that inosine, hypoxanthine, noradrenaline, salidroside, and indole had the highest sensitivity and specificity to detect T. spiralis infection, while glycocholic, taurochenodeoxycholic acid, and 10,16-dihydroxyhexadecanoic acid had the lowest. These findings are comparable to other findings [99].
However, the down-regulated metabolites inosine, hypoxanthine, 1-piperidine, and xanthosine had the highest sensitivity and specificity for identifying T. spiralis infection on day 30. This suggested that these metabolites may serve as diagnostic biomarkers of T. spiralis infection in mice during distinct periods. The putative biomarkers inosine, hypoxanthine, noradrenaline, salidroside, and indole all showed significant positive slopes on days 6 and 15, showing that these metabolites increased as inflammation increased. However, on days 6 and 15, glycocholic, taurochenodeoxycholic acid, and 10,16-Dihydroxyhexadecanoic acid had significant negative slopes for Th2/Treg immune response and duodenum tissue damage indicators, indicating that these metabolites were decreased by inflammation. Inosine, hypoxanthine, 1-piperidine, and xanthosine; however, exhibited opposite patterns of association with the Th2/Treg immune response at day 30, implying that lowering these metabolites may be proposed to promote muscle larvae colonization. These findings raise the possibility that metabolites influence host immune responses during infection. Th2 and duodenum mucosal inflammation during primary infection are mounted against intestinal helminths to expel luminally dwelling worms with up-regulated inosine, hypoxanthine, noradrenaline, salidroside, indole and down-regulated glycocholic, taurochenodeoxycholic acid, and 10,16-Dihydroxyhexadecanoic acid. Subsequent Treg increases are essential for successful parasitism establishment with the healing of tissue damage, accompanied by down-regulated inosine, hypoxanthine, 1-piperidine, and xanthosine. Taurochenodeoxycholic acid (TCDCA) has anti-inflammatory and immune-regulating properties as a signaling molecule [100–102]. Xanthine and hypoxanthine are purine metabolites that induce inflammation by producing reactive oxygen species (ROS). Hesperidin supplementation increased ATP production and reduced oxidative stress in muscle in both in vitro and mouse models, while helminthic invasion may reduce the amount of muscle-protective compounds [99].
The experimental design had limitations, such as the difficulty of accounting for the effects of sampling on the microbiome (Fig 4A), immune responses, and metabolomics profiles (Fig 12). However, we addressed these limitations by standardizing sampling methods, using statistical methods to account for individual variation, conducting replicate experiments, using non-invasive sampling methods, and carefully monitoring the health of the mice. We used a cross-sectional design to assess histopathological and immunohistochemical changes at different stages of infection.
In summary, there are changes in the gut microbiota and metabolic activity, as well as parasite-induced immunomodulation. During the enteral and parenteral phases, the disrupted metabolisms adapt to infection stress and then return to homeostasis during the encapsulated phase. In the encapsulated phase, there was a change from an abundance of Bacteroides to an abundance of probiotic Lactobacillus and Treg-associated-Clostridia. Metabolite-based diagnostics based on potential biomarkers of Trichinella infection showed enormous promise for the future, allowing for rapid evaluation of animals and humans in the field. The present work focused on metabolites from the host; however, using metabolomics to investigate the microbiota, various phases of Trichinella, and integrate global metabolite levels with genomic, transcriptomic, and proteomic data to get systems-level knowledge of parasite metabolism and gene regulation might be of interest for the identification of novel targets for anthelmintic drugs and the development of metabolite-based diagnosis tools.
[END]
---
[1] Url:
https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1011893
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
