Fungal Symbionts Produce Prostaglandin E2 to Promote Their Intestinal Colonization

Abstract

Candida albicans is a ubiquitous fungal symbiont that resides on diverse human barrier surfaces. Both mammalian and fungal cells can convert arachidonic acid into the lipid mediator, prostaglandin E2 (PGE₂), but the physiological significance of fungus-derived PGE₂ remains elusive. Here we report that a C. albicans mutant deficient in PGE₂ production suffered a loss of competitive fitness in the murine gastrointestinal (GI) tract and that PGE₂ supplementation mitigated this fitness defect. Impaired fungal PGE₂ production affected neither the in vitro fitness of C. albicans nor hyphal morphogenesis and virulence in either systemic or mucosal infection models. Instead, fungal production of PGE₂ was associated with enhanced fungal survival within phagocytes. Consequently, ablation of colonic phagocytes abrogated the intra-GI fitness boost conferred by fungal PGE₂. These observations suggest that C. albicans has evolved the capacity to produce PGE₂ from arachidonic acid, a host-derived precursor, to promote its own colonization of the host gut. Analogous mechanisms might undergird host-microbe interactions of other symbiont fungi.

Keywords: Candida albicans; arachidonic acid; host-microbe interactions; phagocyte; prostaglandin E2; symbiont; virulence.

Figure 1

In vitro PGE₂ production by various C. albicans strains. Levels of PGE₂ metabolite, a stable derivative of PGE₂, in the supernatants of in vitro cultures of various fungal species (A) or C. albicans strains (B), were assayed by ELISA. Arachidonic acid (AA) was added (+) or not (–) to the cultures. Data pooled from 3 to 7 independent experiments. Mean ± s.d. *p < 0.05; One-way ANOVA with Dunnett's multiple comparisons test.

Figure 2

Fungal PGE₂ is not required for the in vitro competitive fitness of C. albicans. WT C. albicans or the ole2/ole2 mutant was grown in rich (YPD) or minimal (YNB) media in competition with a fluorescently-tagged WT strain at a 1:1 ratio in serial batch cultures for 80 h. (A) Relative frequencies of the WT or ole2/ole2 strain throughout the serial passages. (B) Fitness coefficients of the WT or ole2/ole2 strain, determined as outlined in Materials and Methods. Data pooled from 3 independent experiments. Mean ± s.d. ns, not significant. Student's t-test.

Figure 3

Fungal PGE₂ is not required for virulence. (A,B) Hyphae formation of the WT and ole2/ole2 C. albicans strains in response to fetal calf serum (FCS) (A) or when plated on Spider agar. (B) AA was added where indicated. (C) In vitro cytotoxicity of the indicated fungal strains to J774A.1 murine macrophages, determined by quantifying the release of lactate dehydrogenase (LDH) from the macrophages. Data presented as a percentage of the maximal amount of LDH released by completely lysed cells. (D) Survival of mice infected intravenously with 2 × 10⁵ cells of the indicated C. albicans strains. (E,F) Kinetics of the fungal CFUs in the tongue (E) and vaginal lavage fluid (F) of mice infected sublingually (E) or intravaginally (F) with the indicated fungal strains. Data pooled from 2 independent experiments. Mean ± s.d. (C,E,F). n = 8–10 mice/group. ns, not significant. Student's t-test (C,E,F); Log-rank test (D).

Figure 4

Fungal PGE₂ promotes intestinal colonization by C. albicans. Antibiotics-treated mice were gavaged with the indicated C. albicans strains and stools were plated on YPD agar at regular intervals over 9 days. (A,B) Relative frequencies (A) and fitness coefficients, normalized to the mean of the WT coefficients (B), of the indicated fungal strains in the intestines of gavaged mice. (C,D) Relative frequencies (C) and fitness coefficients (D) of the indicated fungal strains in mice supplemented with vehicle or 16,16-dimethyl-PGE₂ (dmPGE₂). Coefficients for each fungal strain were normalized to the mean of the respective vehicle coefficients. Fitness coefficients were determined as described in Materials and Methods. Data pooled from 2 independent experiments. Mean ± s.d. n = 9−10 mice/group. ****p < 0.0001; ***p < 0.001. Student's t-test (A,B); Two-way ANOVA with Sidak's multiple comparisons test (C,D).

Figure 5

Ablation of colonic phagocytes abrogates the fitness defect of the ole2/ole2 mutant. In vivo intestinal colonization experiments were performed as in Figure 4. (A,B) Mice were treated with control or clodronate-containing liposomes. (C,D) WT or Ccr2-KO mice were used for intestinal colonization. (A,C) Kinetics of the relative frequencies of the indicated fungal strains. P-values indicate comparisons of relative frequencies of the two fungal strains within each treatment (A) or genetic (C) group. (B,D) Fitness coefficients of the indicated fungal strains determined as described in Materials and Methods. Coefficients for each treatment group were normalized to the mean of the respective WT coefficients. (E) J774A.1 murine macrophages were infected with the indicated C. albicans strains for 24 h and the percentage survival of each strain was determined as outlined in Materials and Methods. (F) Infection of macrophages as in (A), but with addition of exogenous PGE₂ or vehicle to the cultures. Data pooled from 2 (A–D) or 4–6 (E,F) independent experiments. Mean ± s.d. (A–D). n = 9–10 mice/group (A–D). ****p < 0.0001; ***p < 0.001; *p < 0.05; ns, not significant. Two-way ANOVA with Sidak's multiple comparisons test (A–D); paired student's t-test (E,F).