An exopolysaccharide pathway from a freshwater Sphingomonas isolate

Abstract

Bacteria embellish their cell envelopes with a variety of specialized polysaccharides. Biosynthesis pathways for these glycans are complex, and final products vary greatly in their chemical structures, physical properties, and biological activities. This tremendous diversity comes from the ability to arrange complex pools of monosaccharide building blocks into polymers with many possible linkage configurations. Due to the complex chemistry of bacterial glycans, very few biosynthetic pathways have been defined in detail. As part of an initiative to characterize novel polysaccharide biosynthesis enzymes, we isolated a bacterium from Lake Michigan called Sphingomonas sp. LM7 that is proficient in exopolysaccharide (EPS) production. We identified genes that contribute to EPS biosynthesis in LM7 by screening a transposon mutant library for colonies displaying altered colony morphology. A gene cluster was identified that appears to encode a complete wzy/wzx-dependent polysaccharide assembly pathway. Deleting individual genes in this cluster caused a non-mucoid phenotype and a corresponding loss of EPS secretion, confirming the role of this gene cluster in polysaccharide production. We extracted EPS from LM7 cultures and determined that it contains a linear chain of 3- and 4-linked glucose, galactose, and glucuronic acid residues. Finally, we show that the EPS pathway in Sphingomonas sp. LM7 diverges from that of sphingan-family EPSs and adhesive polysaccharides such as the holdfast that are present in other Alphaproteobacteria. Our approach of characterizing complete biosynthetic pathways holds promise for engineering polysaccharides with valuable properties.

Importance: Bacteria produce complex polysaccharides that serve a range of biological functions. These polymers often have properties that make them attractive for industrial applications, but they remain woefully underutilized. In this work, we studied a novel polysaccharide called promonan that is produced by Sphingomonas sp. LM7, a bacterium we isolated from Lake Michigan. We extracted promonan from LM7 cultures and identified which sugars are present in the polymer. We also identified the genes responsible for polysaccharide production. Comparing the promonan genes to those of other bacteria showed that promonan is distinct from previously characterized polysaccharides. We conclude by discussing how the promonan pathway could be used to produce new polysaccharides through genetic engineering.

Keywords: Sphingomonas; biopolymer; polysaccharide; wzy.

Fig 1

Isolation of Sphingomonas sp. LM7. (A) Sphingomonas sp. LM7 growing on agar plates. The isolate grows as yellow, mucoid colonies. (B) Phylogenetic tree showing select isolates from the Sphingomonadales order of Alphaproteobacteria. The neighbor-joining tree was constructed from 16S rDNA sequences aligned with ClustalX. Bootstrapping values are indicated at the relevant nodes. * indicates canonical sphingan-producing strains. # highlights the non-canonical sphingan called sanxan produced by Sphingomonas sanxanigenes NX02. Note that the sanxan biosynthesis genes appear distinct from the sphingan or promonan genes (22).

Fig 2

Identification of two polysaccharide biosynthesis clusters in LM7. (A) Screening for transposon insertion mutants with altered colony morphology. Mutagenized cells were plated on agar plates. Under these conditions, the majority of colonies showed mucoid morphology. The blue carat points to a non-mucoid mutant. (B) Map of the LM7 genome showing the locations of the two gene clusters that influence colony morphology. (C) Phenotypes of ∆wzyA and ∆wzyB mutants. The top image shows the growth of LM7 strains on a solid medium containing Congo Red dye. Wild type and the ∆wzyB mutant appear mucoid and react with the Congo Red dye, while the ∆wzyA mutant does not. The middle image shows an EPS extraction. Wild-type and ∆wzyB cultures produce secreted matrix. ∆wzyA cells do not secrete this matrix. The bottom image shows the density gradient centrifugation of LM7 strains. The ∆wzyB mutant displays increased cell density relative to wild-type and ∆wzyA cells.

Fig 3

The promonan biosynthesis cluster. (A) Map of the prm cluster. Open-reading frames are colored by predicted function. Purple: monosaccharide incorporation; green: assembly/secretion; and gray: polysaccharide modification. (B) Morphologies of individual prm mutants on solid medium supplemented with Congo Red.

Fig 4

Genetic complementation of mutations affecting promonan production. All images show agar plates with wild-type LM7 on the left, the indicated deletion mutant containing a cumate-inducible form of the complementing gene on the right, and the respective deletion mutant containing an empty vector control in the center. (A) Low levels of prmJ induction give an intermediate complementation phenotype in the ∆prmJ mutant background, but higher levels of induction are needed for a full restoration of mucoidy. Complementation of the ∆prmK mutant is most effective at low inducer concentrations, while high levels of induction fail to restore mucoidy. (B) Complementation of promonan-associated phenotypes in individual prm mutants. The deletion mutant being complemented and the optimal inducer concentration are indicated below each image.

Fig 5

Conservation of polysaccharide biosynthesis genes in Alphaproteobacteria. (A) Schematic of Alphaproteobacterial phylogeny showing the relationship among isolates with putative polysaccharide adhesin pathways. (B) Phenotypes associated with holdfast production. Fluorescently labeled wheat germ agglutinin staining shows the loss of holdfast production in the Caulobacter crescentus ∆hfsE ∆pssY ∆pssZ and ∆hfsJ mutants. Crystal violet (CV) staining of cultures grown in microtiter plates shows the loss of surface adhesion in the ∆hfsE ∆pssY ∆pssZ and ∆hfsJ mutants. (C) Inferred activities for initiation and subsequent glycosyltransferase enzymes in four polysaccharide pathways. (D) CV staining assay testing the effect of introducing PHPT genes from various bacteria into the C. crescentus ∆hfsE ∆pssY ∆pssZ mutant. All tested genes appear capable of converting UndPP to UndPP-Glc. (E) CV staining assay testing the effect of introducing wecG-family GT genes from various bacteria into the C. crescentus ∆hfsJ mutant. wecGs from Brucella ovis, Rhizobium leguminosarum, and Phaeobacter inhibens can carry on the HfsJ reaction, while wecG from Escherichia coli (negative control) and prmH from Sphingomonas sp. LM7 cannot.