Fig 1.
S. marcescens UMH9 genetic loci with predicted function in surface polysaccharide production.
Genes with predicted roles in surface polysaccharide production were identified and manually annotated from the S. marcescens UMH9 genome sequence. Individual genes (arrows) are colored according to their predicted function. The coordinates and G+C nucleotide content of each locus are indicated.
Fig 2.
Phylogenetic analysis of the S. marcescens KL.
KL were extracted from S. marcescens genome sequences and MAFFT alignments were used to generate a neighbor-joining tree. The isolation source was determined from NCBI genome entry metadata and is indicated by colored circles in the outer rings. Selected clades representing different capsule genetic types are shaded and arbitrarily designated KL1-5.
Fig 3.
Annotation of KL from selected S. marcescens strains.
KL were identified from the indicated genomes and annotated manually. Individual genes are represented by arrows and were colored according to predicted function. The conserved and variable (CPSv) regions of each KL are indicated. Abbreviations: AT, acetyltransferase; GT, glycosyltransferase; GH, glycosylhydrolase; hyp. hypothetical protein of unknown function; PT, pyruvyltransferase.
Fig 4.
The wzi gene is required for cellular association of CPS.
Total uronic acids were quantified from S. marcescens strains cultured in M9 medium supplemented with glucose using a colorimetric assay. Uronic acid levels were normalized to culture density and calculated based on a glucuronic acid standard curve. Cell-associated CPS levels (A) were determined from S. marcescens strains collected by centrifugation and free CPS levels (B) were determined from conditioned medium cleared of cells by centrifugation and subsequent filtration. An unpaired t-test was used to assess significant differences between strains (**, P<0.01; ***, P<0.001; ****, P<0.0001).
Fig 5.
SDS-PAGE of CPS and LPS from S. marcescens mutant strains.
Total cell-associated polysaccharides were isolated from S. marcescens strains following overnight culture. (A) Purified polysaccharides were resolved on 7.5% SDS-PAGE gels and visualized by Alcian Blue staining. The molecular weights (kDa) of pre-stained protein standards (MW standard) are indicated. (B) Total cell-associated polysaccharides were electrophoresed on 4–20% gradient SDS-PAGE gels. LPS was visualized using the Pro-Q Emerald 300 Liposaccharide Gel Stain Kit with purified LPS from E. coli serving as a positive control. The region containing the characteristic O-antigen laddering pattern is indicated along with the high molecular weight CPS that co-stains with this procedure.
Fig 6.
Acidic CPS production by S. marcescens strains isolated from BSI.
Total cell-associated uronic acids from S. marcescens strains cultured in M9 medium supplemented with glucose were quantified by colorimetric assay. Uronic acid levels were normalized to culture density and calculated based on a glucuronic acid standard curve. Significant differences relative to the acapsular strain ATCC 13880 were assessed by one-way ANOVA with Dunnett’s multiple comparisons test (****, P<0.0001).
Fig 7.
SDS-PAGE of S. marcescens CPS and LPS from five KL clades.
Total cell-associated polysaccharides were isolated from S. marcescens strains following overnight culture. (A) Purified polysaccharides were resolved on 7.5% SDS-PAGE gels and visualized by Alcian Blue staining. The molecular weights (kDa) of pre-stained protein standards (MW standard) are indicated. (B) Total cell-associated polysaccharides were also electrophoresed on 4–20% gradient SDS-PAGE gels. LPS was visualized using the Pro-Q Emerald 300 Liposaccharide Gel Stain Kit with purified LPS from E. coli serving as a positive control. The region containing the characteristic O-antigen laddering pattern is indicated.
Fig 8.
Detection of sialic acids in S. marcescens strains UMH9 and gn773.
Sialic acid production by S. marcescens strains was measured by thiobarbituric acid assay. Sialic acid levels were normalized to the number of bacteria assayed from each culture and a 100 μM solution of N-acetylneuraminic acid (Neu5Ac) was included as a positive control. (A) Differences between strains were assessed by one-way ANOVA with Dunnett’s multiple comparisons test relative to the acapsular strain ATCC 13880 (****, P<0.0001). (B) Statistical significance for the loss of sialic acid production by the ΔneuB::nptII and ΔneuAB mutant strains compared to wild-type and complemented mutants was determined by unpaired t-test (*, P<0.05; **, P<0.01; ***, P<0.001).
Fig 9.
The neuB gene is required for CPS synthesis.
CPS production by the ΔneuB::nptII (A) and ΔneuAB (B) mutants was evaluated using an uronic acid assay. Cell-associated uronic acids were normalized to the density of the original culture and calculated based on a glucuronic acid standard curve. Statistical significance for differences between strains was assessed by unpaired t-test (**, P<0.01; ***, P<0.001; ****, P<0.0001).
Table 1.
S. marcescens strains and recombinant plasmids used in this study.
Fig 10.
GC-MS of CPS purified from S. marcescens strains.
(A-F) Trimethylsilyl derivatives of CPS monosaccharides were detected by GC-MS. Inositol (1 μg) was included in each run as an internal standard (IS). The ribose and rhamnose peak that appears in some samples is not resolved under the tested conditions. Abbreviations: Gal, galactose; GalNAc, N-acetylgalactosamine; Glc, glucose; GlcNAc, N-acetylglucosamine; Hep, heptose; KDN, ketodeoxynonulonic acid; Man, mannose; Rha, rhamnose; Rib, ribose; U, unknown.
Fig 11.
Identification of uronic acids in S. marcescens polysaccharide preparations.
(A-F) Uronic acids and other monosaccharides were detected in 10 μg preparations of S. marcescens CPS by HPAEC-PAD. The amount (nmole) of monosaccharides detected are shown in parentheses. G. Standards consisting of 1 nmole of selected monosaccharides were assessed simultaneously. Abbreviations: Fuc, fucose; Gal, galactose; GalA, galacturonic acid; GalN, galactosamine; Glc, glucose; GlcA, glucuronic acid; GlcN, glucosamine; Hep, heptose; IdoA, iduronic acid; Man, mannose; Rha, rhamnose; Rib, ribose; Xyl, xylose.
Fig 12.
RP-UPLC analysis of sialic acids.
(A-F) Polysaccharides were acid hydrolyzed and liberated sialic acids were derivatized with DMB for fluorescence detection following separation by RP-UPLC. (G) RP-UPLC of sialic acid monosaccharide standards (5 pmole). (H) The abundance of selected monosaccharides in each sample was determined based on the standards. Abbreviations: KDN; ketodeoxynonulonic acid; KDO, 3-deoxy-D-manno-octulosonic acid; Neu5Ac, N-acetylneuraminic acid; Neu5Gc, N-glycolylneuraminic acid.
Fig 13.
Serum susceptibility and phagocyte internalization of S. marcescens strains.
(A) Bacteria were exposed to 40% pooled human serum for 90 minutes. Viability was calculated relative to the number of bacteria present at the start of the experiment. Differences in serum sensitivity were assessed by one-way ANOVA with Dunnett’s multiple comparisons test relative to strain ATCC 13880. (B) Serum sensitivity of UMH9 capsule mutant strains was determined as described for panel A and an unpaired t test was used to determine statistical significance. (C) Adherent U937 cells were infected with S. marcescens strains at an MOI of 20. The proportion of intracellular bacteria was determined relative to the total bacterial inoculum after a 1 h treatment with gentamicin to kill extracellular bacteria. A one-way ANOVA with Dunnett’s multiple comparisons test relative to ATCC 13880 was used to assess statistical significance. (D) Internalization of the UMH9 capsule mutant strain by U937 cells was tested as described for panel C. A unpaired t-test was used to determine statistical significance. Symbols: **, P<0.01; ***, P<0.001; ****, P<0.0001.