Statistical analysis of the Bacterial Carbohydrate Structure Data Base (BCSDB): characteristics and diversity of bacterial carbohydrates in comparison with mammalian glycans

Abstract

Background: There are considerable differences between bacterial and mammalian glycans. In contrast to most eukaryotic carbohydrates, bacterial glycans are often composed of repeating units with diverse functions ranging from structural reinforcement to adhesion, colonization and camouflage. Since bacterial glycans are typically displayed at the cell surface, they can interact with the environment and, therefore, have significant biomedical importance.

Results: The sequence characteristics of glycans (monosaccharide composition, modifications, and linkage patterns) for the higher bacterial taxonomic classes have been examined and compared with the data for mammals, with both similarities and unique features becoming evident. Compared to mammalian glycans, the bacterial glycans deposited in the current databases have a more than ten-fold greater diversity at the monosaccharide level, and the disaccharide pattern space is approximately nine times larger. Specific bacterial subclasses exhibit characteristic glycans which can be distinguished on the basis of distinctive structural features or sequence properties.

Conclusion: For the first time a systematic database analysis of the bacterial glycome has been performed. This study summarizes the current knowledge of bacterial glycan architecture and diversity and reveals putative targets for the rational design and development of therapeutic intervention strategies by comparing bacterial and mammalian glycans.

Figure 1

Distribution of carbohydrate sequences for various taxonomic classes. For the combined BCSDB and GLYCOSCIENCES.de databases the pie chart sector areas correspond to the distribution (in percent) of the 13775 assigned sequences within the taxonomic classes shown, while the labels give the absolute numbers of sequences. The white pie sector contains all classes which each have < 1% of the total assigned records. This category is expanded in the bar chart at the right, where the bottom block "all other" contains all classes which each have < 0.1% of the assigned records. Class names ending with ...opsida or ...mycetes correspond to plants or fungi, respectively; Actinopterygii contains fish while Chondrichthyes contains sharks.

Figure 2

Size distribution of carbohydrate sequence units. The normalized frequency distribution for total carbohydrate residue count per sequence unit is shown in percent of total sequences for each taxonomic group. A. For taxonomy Set 1 the solid blue curve represents the cumulative values for bacterial oligomers (blue-shaded region) and the repeating units of polymers (cyan region) in comparison with mammals (black curve). B. The frequency distribution for carbohydrate residue count per sequence unit is shown for each of the bacteria groups defined in taxonomy Set 2, using the color coding defined in the legend. For comparison the dotted curve shows the distribution for mammals. The curves are smoothed for visual clarity.

Figure 3

Branching index distribution. The normalized frequency distributions for the number of branching points per residue are shown for the carbohydrate sequence units of taxonomy Set 1 (A) and Set 2 (B), analogous to the graphs in Fig. 2.

Figure 4

Mean charge density distribution. Normalized frequency distributions for the maximum possible mean charge density per residue are plotted for the carbohydrate sequence units of taxonomy Set 1 (A) and Set 2 (B), analogous to the graphs in Fig. 3.

Figure 5

Diversity of monosaccharides in bacteria and mammals. For taxonomy Set 1 the 35 most abundant monosaccharide residues are listed from left to right in decreasing order of their total abundance across all taxonomic groups. The three rows of circles correspond to the three groups: mammals (gray), bacteria (polymers, cyan), bacteria (oligomers, blue), as defined in Table 1. Circle areas reflect the relative abundance of a monosaccharide residue within each group (monosaccharide count/total residue count per taxonomic group). Residues that are the result of analytical artifacts or incomplete structure elucidation (hexosamine alditols, glycerol, D-Glc, etc.) are highlighted with gray bars. In this and subsequent Figures the anomeric designators α and β are written as a and b, the ring type designators p and f are shown as p and f.

Figure 6

The most abundant monosaccharides in bacteria. Circle areas reflect relative abundances of the 30 most common monosaccharide residues (A) or basic entities (B) for the carbohydrates found in the BCSDB for each of the bacterial taxonomy groups of Set 2. Within each group the abundances are normalized to the total number of residues per group. The color code (see legend at bottom) is the same as in Figs. 2–4; for comparison, the open circles represent data for mammals. The residues are sorted from top to bottom in order of decreasing total abundance in bacteria (order differs from Fig. 5). The residue Kdo (without anomeric configuration) results from analytical artifacts and is highlighted in gray. For the basic entities defined in B, no distinction is made between anomeric configurations and ring types.

Figure 7

The most abundant unique monosaccharides. Circle areas reflect relative abundances within a taxonomic group for those unique monosaccharide residues which appear exclusively, or nearly so, in a single taxonomic group of Set 1 (A) or Set 2 (B). Uniqueness is defined here as: frequency in the selected group > 0.1%, frequency in other groups < 0.1%. Residues that result from analytical artifacts and those that are ambiguous due to incomplete structure elucidation are highlighted in gray. The symbol X represents any substituent.

Figure 8

Monosaccharide ring type distributions. The distribution of residue ring type (pyranose, furanose, linear, unknown) is shown schematically for various taxonomic groups. A. For taxonomy Set 1 the areas of the colored bars are proportional to the absolute occurrences (numbers shown) of a given ring type in each taxonomic group. The vertical scale is expanded in the inset. B. For taxonomy Set 2 two different ways of viewing the data are presented. (1) For a given ring type the area or height of each colored bar in a stack represents the relative abundance (%) of that ring type for each of the taxonomic groups, normalized to the total occurrence of that ring type across all groups (stack height = 100% for each ring type). Thus, the bar heights within a stack represent the distribution of a single ring type across all taxonomic groups. (2) The number labeled in each bar represents the frequency (in %) of residues with the corresponding ring type within the bar's taxonomic group, normalized to the total number of residues for that group. The numbers sum horizontally to 100% for each taxonomic group (color) and, therefore, represent the distribution of the different ring types within an individual group.

Figure 9

Distribution of monosaccharide modifications. Frequency distributions (in %) for secondary modifications of monosaccharide residues are shown for taxonomy Set 1 (A) and Set 2 (B), normalized to the total number of carbohydrate residues within each taxonomic group. In A the bars for bacterial oligomers (blue) and polymers (cyan) are stacked to give the cumulative values for all bacteria studied.

Figure 10

Distribution of monosaccharide pairs in various bacterial groups. For taxonomy Set 2 the matrix presents relative abundance data for monosaccharide residue pairs of all linkage types, involving the 20 most common residues serving as donor (children) or acceptor (parent). Each circle area reflects the relative abundance of a given donor-acceptor pair (matrix coordinates) within the corresponding taxonomic group, normalized to the total number of pairs within that group.

Figure 11

Glycosidic linkages in mammalian carbohydrates. Frequency distribution of specific disaccharide linkages in mammalian carbohydrates. Plotted circle areas represent the relative frequencies for disaccharides formed from the 9 most common donors (children) and 9 most common acceptors (parents) in a defined glycosidic linkage (color code in legend). The areas of the circles are proportional to the relative abundances of disaccharide pairs, normalized to the total number of specific disaccharide pairs. The linkage codes α1-n and β1-n correspond to a linkage to any exocyclic carbon at the acceptor, e.g. C6 in hexopyranoses. For donor residues in keto form the linkage is at the anomeric carbon C2 instead of C1. For better visualization some of the circles for a given linkage are offset somewhat from the matrix coordinate corresponding to a given linkage type.

Figure 12

Glycosidic linkages in bacterial carbohydrates. Frequency distribution of specific disaccharide linkages in bacterial carbohydrates. Plotted circle areas represent the relative frequencies for disaccharides formed from the 15 most common donors (children) and 15 most common acceptors (parents) in a defined glycosidic linkage (color code in legend) for bacterial oligomers (A) or polymers (B). The areas of the circles are proportional to the relative abundances of specific disaccharide pairs, normalized to the total number of disaccharide pairs. The linkage codes and plotting offsets are used as in Fig. 12. The lower left corner of each diagram is plotted at the right with rescaling for better visualization.