Capsular Polysaccharide Cross-Regulation Modulates Bacteroides thetaiotaomicron Biofilm Formation

Abstract

Bacteroides thetaiotaomicron is one of the most abundant gut symbiont species, whose contribution to host health through its ability to degrade dietary polysaccharides and mature the immune system is under intense scrutiny. In contrast, adhesion and biofilm formation, which are potentially involved in gut colonization and microbiota structure and stability, have hardly been investigated in this intestinal bacterium. To uncover B. thetaiotaomicron biofilm-related functions, we performed a transposon mutagenesis in the poorly biofilm-forming reference strain VPI-5482 and showed that capsule 4, one of the eight B. thetaiotaomicron capsules, hinders biofilm formation. We then showed that the production of capsules 1, 2, 3, 5, and 6 also inhibits biofilm formation and that decreased capsulation of the population correlated with increased biofilm formation, suggesting that capsules could be masking adhesive surface structures. In contrast, we showed that capsule 8 displayed intrinsic adhesive properties. Finally, we demonstrated that BT2934, the wzx homolog of the B. thetaiotaomicron glycosylation locus, competes with capsule production and impacts its adhesion capacity. This study therefore establishes B. thetaiotaomicron capsule regulation as a major determinant of B. thetaiotaomicron biofilm formation, providing new insights into how modulation of different B. thetaiotaomicron surface structures affects in vitro biofilm formation.IMPORTANCE The human gut harbors a complex bacterial community that plays important roles in host health and disease, including nutrient acquisition, maturation of the immune system, and resistance to infections. The capacity to adhere to surfaces and form communities called biofilms is believed to be important for niche colonization and maintenance of gut bacteria. However, little is known about the adhesion capacity of most gut bacteria. In this study, we investigated biofilm formation in Bacteroides thetaiotaomicron, one of the most abundant bacteria of the normal mammalian intestine. We identified that B. thetaiotaomicron capsules, a group of eight surface-exposed polysaccharidic layers mediating important interactions with the gut environment, are also major determinants of biofilm formation that mask or unmask adhesion factors. Studying how B. thetaiotaomicron regulates its adhesion properties will allow us to better understand the physiology and specific properties of this important gut symbiont within anaerobic biofilms.

Keywords: Bacteroides thetaiotaomicron; biofilm; capsule.

FIG 1

Capsule 4 inhibits biofilm formation in B. thetaiotaomicron VPI-5482. (A) Organization of B. thetaiotaomicron capsular operon 4 (CPS4). The first two genes (BT1358 and BT1357) code regulators of capsular biosynthesis. BT1356-1338 code the enzymes involved in Cps4 capsular polysaccharide biosynthesis. The arrows indicate 5 individual transposon insertions within the CPS4 operon. (B) The 96-well plate biofilm assay after 48 h of growth in BHIS. The Mean of the WT is adjusted to 100%. In the min-max boxplot of 6 biological replicates for each strain, each replicate is the mean of two technical replicates. ***, P < 0.0005; Mann-Whitney test, comparing the indicated mutant to the WT. (C) The 96-well plate biofilm assay after 48 h of growth in BHIS. The mean of the WT is adjusted to 100%. In the min-max boxplot of 6 to 9 biological replicates for each strain, each replicate is the mean of two technical replicates. ***, P < 0.0005; Mann-Whitney test, comparing the indicated mutant to the WT. The images shown under each boxplot correspond to representative CV-stained microtiter wells after resuspension of the biofilm.

FIG 2

Capsule cross-regulation modulates biofilm formation in B. thetaiotaomicron. (A) The 96-well plate crystal violet biofilm assay after 48 h of growth in BHIS. (B) Organization of B. thetaiotaomicron capsular operon 4 (CPS4) with identified transposon insertion points in the first two genes of the operon (BT1358 and BT1357), coding regulators of capsular biosynthesis. (C) The 96-well plate crystal violet biofilm assay after 48 h of growth in BHIS. The Mean of the WT is adjusted to 100%. (D) The 96-well plate crystal violet biofilm assay after 48 h of growth in BHIS. In panels A, C, and D, the mean of the WT is adjusted to 100%. In the min-max boxplot of 6 to 9 biological replicates for each strain, each replicate is the mean of two technical replicates. **, P < 0.005; Mann-Whitney test. upxY, upxY^BT1358; upxZ, upxZ^BT1357.

FIG 3

Capsule expression in B. thetaiotaomicron is heterogenous and has consequences for biofilm formation. (A and D) The 96-well plate biofilm assay after 48 h of growth in BHIS. The mean of the WT is adjusted to 100%. In the min-max boxplot of 6 biological replicates for each strain, each replicate is the mean of two technical replicates. **, P < 0.005; Mann-Whitney test, comparing the indicated mutant to the WT. The pictures shown under the boxplot in panel A correspond to representative CV-stained microtiter wells after resuspension of the biofilm. (B) Transmission electron microscopy (TEM) images of overnight cultures fixed with ferritin. Arrows indicate some examples of acapsulated cells. (C) Percentage of acapsulated cells of indicated strains counted on TEM pictures. For each strain, at least 100 cells were counted. ***, P < 0.0005; prop.test (R). upxY, upxY^BT1358; upxZ, upxZ^BT1357.

FIG 4

BT2934 is a novel capsule inhibitor. (A) Organization of the B. thetaiotaomicron protein glycosylation BT2934 locus with identified transposon insertion point. (B and C) The 96-well plate crystal violet biofilm assay after 48 h of growth in BHIS. The mean of the WT is adjusted to 100%. In the min-max boxplot of 6 to 12 biological replicates for each strain, each replicate is the mean of two technical replicates. *, P < 0.05; **, P < 0.005; ***, P < 0.0005; Mann-Whitney test. (D) TEM images of ΔCPS4ΔBT2934-2938, ΔCPS4ΔBT2935-2938, and ΔCPS4ΔBT2934-2938+pBT2934 overnight cultures fixed with ferritin. The arrows indicate some acapsulated cells as examples. (E) Percentage of acapsulated cells in overnight cultures counted on TEM pictures. For each strain, at least 100 cells were counted. ***, P < 0.0005; prop.test (R). (F) Overnight cultures of indicated strains in BHIS. Only ΔCPS1-8 showed aggregation.

FIG 5

BT2934 and CPS4 contribute to in vivo colonization in axenic mice. Min-max boxplot of CFU/ml/dry weight of feces, numbered from feces from 5 to 10 axenic mice after cocolonization with indicated strains. *, P < 0.05; **, P < 0.005; ***, P < 0.0005; Mann-Whitney test. (A) WT (ermR) versus WT (tetR). (B) WT (ermR) versus ΔCPS4 (tetR). (C) (ermR) versus ΔBT2934-2938 (tetR). (D) ΔCPS4 (ermR) versus ΔCPS4ΔBT2934-2938 (tetR).