Research ArticleMICROBIOLOGY

Mapping of host-parasite-microbiome interactions reveals metabolic determinants of tropism and tolerance in Chagas disease

See allHide authors and affiliations

Science Advances  22 Jul 2020:
Vol. 6, no. 30, eaaz2015
DOI: 10.1126/sciadv.aaz2015
  • Fig. 1 Spatial impact of T. cruzi infection is reflected by spatial modulation of the tissue small-molecule profile.

    C3H/HeJ male mice (n = 5 per group and replicate) were mock-infected or infected with 1000 luminescent T. cruzi strain CL Brener trypomastigotes in two biological replicates. GI samples were collected 12 and 89 days after infection. (A) Parasite burden at each sampling site. To correct for variations in sample size and background signal, luminescence counts were normalized to signal from matched uninfected samples and to sample weight, for each sampling position. (B and C) Median normalized luminescent signal, at each sampling site, 12 days after infection (B) and 89 days after infection (C). Common logarithmic scale for (B) and (C), scaled from lowest luminescent signal (dark blue) to highest signal (dark red). (D) PCoA analysis showing separation between sampling sites in terms of overall chemical composition, even within a given organ (positive mode, all time points combined, Bray-Curtis-Faith distance metric). (E to I) PCoA analysis showing chemical composition differences (positive mode analysis) between infected and uninfected samples in the esophagus (E), stomach (F), small intestine (G), cecum (H), and large intestine (I). (J) R2 at each sampling site [common logarithmic scale, scaled from lowest R2 (dark blue) to highest R2 (dark red)]. PCo, principal coordinate (e.g., PCo1, principal coordinate 1).

  • Fig. 2 Common and tissue-specific metabolic changes identified by random forest demonstrate persistence of these alterations at sites of CD.

    (A to C) Infection-induced elevation of C20:4 acylcarnitine in the esophagus and small intestine in the acute stage, persisting in the esophagus in the later infection stage. (D to F) Infection-induced elevation of PC(20:4) in the esophagus and large intestine in the acute stage, persisting in the esophagus 89 days after infection. PC(20:4) was decreased in the acute stage in the infected small intestine. (G and H) Infection-induced elevation of kynurenine in stomach and large intestine in the acute stage, persisting in the large intestine 89 days after infection. (I and J) Infection-induced decrease in tryptophan at sites of increased kynurenine in the large intestine (12 and 89 days after infection). Infection also increased tryptophan in the esophagus 89 days after infection. (K to M) Infection-induced increase in the small intestine, cecum, and large intestine cholic acid (all detected adducts combined), acute stage only. E, esophagus; SC, stomach; SI, small intestine; C, cecum; LI, large intestine. Black lines in (A and B), (D and E), and (G to L) indicate false discovery rate (FDR)–corrected Mann-Whitney P < 0.05. Spatial distribution plots in (C), (F), and (M) are scaled for each specified metabolite, from lowest median normalized metabolite abundance (dark blue) to highest abundance (dark red). Linear scale in (C) and (F) and logarithmic scale in (M). Asterisks (*) above GI tract positions in (C), (F), and (M) indicate sites with FDR-corrected Mann-Whitney P < 0.05 compared to matched uninfected samples.

  • Fig. 3 T. cruzi infection has a persistent, spatially heterogeneous impact on the microbiota.

    Representation of between-sample differences in microbial community composition through principal coordinate transformation of unweighted UniFrac distances. (A to D) Comparison of acute-stage infected and uninfected samples from stomach (A), small intestine (B), cecum (C), and large intestine (D). (E and F) Comparison of cecum (E) and large intestine (F) samples from uninfected and persistently infected mice. Spatial heterogeneity was also observed within an organ (G), with the highest disturbances in the microbiota in the proximal large intestine (sampling position 11). (G) R2 at each sampling site in the acute stage [logarithmic scale, scaled from lowest R2 (dark blue) to highest R2 (dark red)]. *PERMANOVA P < 0.05.

  • Fig. 4 Chemical cartography reveals a causal role for carnitine metabolism in acute CD tolerance.

    (A) Acylcarnitine molecular network (all tissue sites and time points combined). (B) Infection-induced increases in acute-stage long-chain acylcarnitines in the distal small intestine (black lines, FDR-corrected Mann-Whitney P < 0.05). (C) Spatial distribution of acute-stage long-chain acylcarnitines [medians; common logarithmic scale from lowest (dark blue) to highest abundance (dark red)]. *Above sampling sites, P < 0.05 FDR-corrected Mann-Whitney. (D) Carnitine treatment prevents acute-stage mortality (n = 5 per group, 50,000 trypomastigote infection). (E and F) Comparable parasite burden between carnitine-treated and vehicle groups. (E) Overall whole-body luminescence. Mean and SEM are displayed. (F) Representative bioluminescent imaging, week 3 after infection (common scale). (G and H) Carnitine treatment mitigates plasma infection-induced metabolic disturbances (n = 24 benznidazole and carnitine groups, n = 22 vehicle group, and n = 9 uninfected group). (G) PCoA analysis (Bray-Curtis-Faith distance metric) of plasma samples. (H) Heat map showing metabolite features distinguishing vehicle-treated individuals from carnitine-treated, benznidazole-treated, and uninfected individuals (Kruskal-Wallis, FDR-corrected P < 0.05). (I) Carnitine treatment mitigates cardiac infection-induced metabolic disturbances. PCoA analysis (Bray-Curtis-Faith distance metric) of heart samples [n = 5 for uninfected, vehicle, and carnitine groups; n = 4 for benznidazole group (one biological replicate; see fig. S6D for data from second replicate)]. (J) Carnitine treatment reduces cardiac Bnp gene expression (n = 10 for benznidazole group and carnitine group, n = 9 for vehicle group, and n = 4 for uninfected group; Student’s t test P = 0.01).

Supplementary Materials

  • Supplementary Materials

    Mapping of host-parasite-microbiome interactions reveals metabolic determinants of tropism and tolerance in Chagas disease

    E. Hossain, S. Khanam, D. A. Dean, C. Wu, S. Lostracco-Johnson, D. Thomas, S. S. Kane, A. R. Parab, K. Flores, M. Katemauswa, C. Gosmanov, S. E. Hayes, Y. Zhang, D. Li, C. Woelfel-Monsivais, K. Sankaranarayanan, L.-I. McCall

    Download Supplement

    The PDF file includes:

    • Figs. S1 to S7
    • Tables S1 and S2
    • Legend for data file S1

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article