Oral administration of live- or heat-killed Candida albicans worsened cecal ligation and puncture sepsis in a murine model possibly due to an increased serum (1→3)-β-D-glucan

Abstract

Candida albicans is the most common fungus in the human intestinal microbiota but not in mice. To make a murine sepsis model more closely resemble human sepsis and to explore the role of intestinal C. albicans, in the absence of candidemia, in bacterial sepsis, live- or heat-killed C. albicans was orally administered to mice at 3h prior to cecal ligation and puncture (CLP). A higher mortality rate of CLP was demonstrated with Candida-administration (live- or heat-killed) prior to CLP. Fecal Candida presented only in experiments with live-Candida administration. Despite the absence of candidemia, serum (1→3)-β-D-glucan (BG) was higher in CLP with Candida-administration than CLP-controls (normal saline administration) at 6h and/or 18h post-CLP. Interestingly, fluconazole attenuated the fecal Candida burden and improved survival in mice with live-Candida administration, but not CLP-control. Microbiota analysis revealed increased Bacteroides spp. and reduced Lactobacillus spp. in feces after Candida administration. Additionally, synergy in the elicitation of cytokine production from bone marrow-derived macrophages, in vitro, was demonstrated by co-exposure to heat-killed E. coli and BG. In conclusion, intestinal abundance of fungi and/or fungal-molecules was associated with increased bacterial sepsis-severity, perhaps through enhanced cytokine elicitation induced by synergistic responses to molecules from gut-derived bacteria and fungi. Conversely, reducing intestinal fungal burdens decreased serum BG and attenuated sepsis in our model.

Fig 1

Survival analysis of mice with cecal ligation and puncture (CLP) with normal saline (NSS), NSS with fluconazole and heat-killed Candida oral administration (A) and with live-Candida with and without fluconazole (B) at the time of operation is shown. Organ injury, as demonstrated by serum creatinine (Scr) for kidney injury (C), alanine transaminase (ALT) for liver injury (D), serum cytokines (TNF-α, IL-6 and IL-10) (E-G), fecal Candida burdens (H) and serum (1→3)-β-D-glucan (BG) (I) was analyzed.(n = 6-7/group); #, p < 0.05 vs. NSS; *, p < 0.05

Fig 2

Blood bacterial count (A), peritoneal bacterial count (B) and bacterial count from internal organs (C) are shown. (n = 7/ group). The time-course of fecal Candida burdens (D) and serum (1→3)-β-D-glucan (BG) in different treatment groups (E, F) is described. [n (at 0h) = 10-12/group and n (at 6h and 18h) = 5-6/group]. #, p < 0.05 vs. 0h in each group; *, p < 0.05 vs. NSS

Fig 3. Serum (1→3)-β-D-glucan (BG) (A) and serum cytokines (TNF-α, IL-6 and IL-10) (B-D) in the mice at 96h post-sham surgery (n = 5) and mice survived at 96h post-CLP administered with normal saline (NSS; n = 5), fluconazole (n = 4) and live-Candida with fluconazole (n = 2) are shown (survival rate of these groups were demonstrated in Fig 1A and 1B).

*, p < 0.05 vs. sham; No data, non survivor in this group.

Fig 4. Gut microbiota analysis from individual mouse feces.

Control (N1-3), heat-killed Candida administration (HC1-3) and live-Candida (C1-3) by relative abundance of bacterial diversity at genus level (A) are shown. Abbreviation “p”, “c” and “o” means unclassified family sequences in phylum, class and order, respectively, and Genera that contain ≤ 0.01% relative abundance were removed. A representative of gut microbiota structures demonstrated by non-metric multidimensional scaling (NMDS) are shown (B).

Fig 5

Time-course of cytokines in supernatant of macrophages after incubation with heat-killed E. coli preparation (see Method) with or without purified (1→3)-β-D-glucan (Pachyman) (BG) (at 1 or 10 μg/ml) or BG 10 μg/ml alone was shown (independent experiments were done in triplicate) (A-C). Area under the curve of cytokine responses of graph A-C are demonstrated (D-F). **, p < 0.05 vs. PBS+heat-killed E. coli.