Genomic and Proteomic Study of Andreprevotia ripae Isolated from an Anthill Reveals an Extensive Repertoire of Chitinolytic Enzymes

Abstract

Chitin is an abundant natural polysaccharide that is hard to degrade because of its crystalline nature and because it is embedded in robust co-polymeric materials containing other polysaccharides, proteins, and minerals. Thus, it is of interest to study the enzymatic machineries of specialized microbes found in chitin-rich environments. We describe a genomic and proteomic analysis of Andreprevotia ripae, a chitinolytic Gram-negative bacterium isolated from an anthill. The genome of A. ripae encodes four secreted family GH19 chitinases of which two were detected and upregulated during growth on chitin. In addition, the genome encodes as many as 25 secreted GH18 chitinases, of which 17 were detected and 12 were upregulated during growth on chitin. Finally, the single lytic polysaccharide monooxygenase (LPMO) was strongly upregulated during growth on chitin. Whereas 66% of the 29 secreted chitinases contained two carbohydrate-binding modules (CBMs), this fraction was 93% (13 out of 14) for the upregulated chitinases, suggesting an important role for these CBMs. Next to an unprecedented multiplicity of upregulated chitinases, this study reveals several chitin-induced proteins that contain chitin-binding CBMs but lack a known catalytic function. These proteins are interesting targets for discovery of enzymes used by nature to convert chitin-rich biomass. The MS proteomic data have been deposited in the PRIDE database with accession number PXD025087.

Keywords: CBM; Chitinase; GH18; GH19; LPMO; carbohydrate-binding module; chitin; chitinolytic machineries; genome analysis; proteomics.

Figure 1

Putative chitin-active enzymes in the proteome of A. ripae. The figure shows all predicted A. ripae proteins containing domains annotated as glycosyl hydrolases in families GH18, GH19, or GH20 or as LPMO, with their domain architecture. The embedded bar charts show the average abundance during growth on α-chitin, glucose or N-acetylglucosamine for the five analyzed time points (1, 3, 5, 7, 13 days); the y-axis indicates protein abundance from 17 to 32 log2(LFQ); more detailed quantitative data is shown in Figure 2. The roman numbers in the column labeled “Cluster” refer to the clusters depicted in Figure 2. GH18 domains (green) that are predicted to lack catalytic activity are marked by “–”, whereas GH18 domains for which the prediction is uncertain are marked by “+/–”; see text for more details. Domains were annotated using InterProScan. HEX_bac_N: N-terminal domain of beta-hexosaminidases; GbpA_2: N-acetylglucosamine binding domain; CHB_HEX_N: N-terminal domain of chitobiases and beta-hexosaminidases, similar to CBM2/3, possibly involved in substrate binding; CHB_HEX_C: C-terminal domain of chitobiases and beta-hexosaminidases, no proposed catalytic or binding function; Gal_BD: Galactose binding domain; CE2_N: N-terminal domain of CE2 acetyl esterases; SGNH_hydro: SGNH hydrolase-type esterase domain with a similar fold to flavoproteins, often found in esterases and lipases; Chi_C: C-terminal domain found in some GH18s; LytTR: DNA-binding domain found in response regulators.

Figure 2

Heat map of secreted CAZymes. The figure shows a heat map of the 39 detected CAZymes that are predicted to be secreted for growth on α-chitin, glucose or N-acetylglucosamine, at five different time points (1–13 days). The color indicates the protein abundance, log2(LFQ), and represents the average of three biological replicates; gray color means not detected. The columns show protein ID’s, CAZy annotation for the catalytic and binding domains (auxiliary activity (AA), carbohydrate esterase (CE), glycosyl hydrolase (GH), polysaccharide lyase (PL)), the presence of carbohydrate-binding modules (CBMs), and the secretion pathway as predicted by LipoP. Superscripts at “GH18” indicate that this GH18 domain lacks (−) or possibly lacks (+/−) catalytic activity; see text for more details. The proteins were hierarchically clustered based on protein abundance patterns and manually divided into four groups as indicated: (I) Low expression on all three substrates; (II) Medium expression on α-chitin but not on glucose or N-acetylglucosamine; (III) High expression on all three substrates but clearly higher on α-chitin compared to glucose and N-acetylglucosamine; and (IV) Medium expression on all three substrates. GbpA_2: N-acetylglucosamine binding domain; Gal_BD: Galactose binding domain.

Figure 3

Domain structure of IGB_0282 and CbpD from P. aeruginosa. Domain boundaries are based on sequence analysis using InterPro. The gray box with a ? indicates a putative CBM with no current CAZy annotation, which is likely related to CBM5/12; see text for details. SP, signal peptide; LC-linker, low complexity region containing mainly Pro, Thr, Val, and Ala.

Figure 4

Detected secreted proteins with a putative chitin-binding domain but no known chitin-active catalytic domain. All these proteins, except IGB42_00583, were upregulated during growth on chitin (Figure 2). InterPro accession numbers for the non-CAZy domains are Peptidase M60: IPR031161; Beta/gamma Crystallin: IPR001064; Fibronectin type III (Fn3): IPR003961; Uncharacterized: IPR007541; Serralysin-like metalloprotease: IPR011049. See main text for more details. SP: signal peptide.

Figure 5

Heat map of secreted non-CAZymes. (A) The figure shows a heat map of the 177 detected non-CAZymes that are predicted to be secreted for three different substrates at five different time points (1–13 days, as shown in the zoomed regions). The color indicates protein abundance, log2(LFQ), and is based on the average of three biological replicates; gray color means not detected. The proteins were hierarchically clustered based on protein abundance patterns. NAG: N-acetylglucosamine. (B) Domain architecture of the 10 hypothetical proteins found in the lower of the two enlarged clusters, all upregulated on NAG and α-chitin. Signal peptides are shown in yellow. DUF: domain of unknown function; TPR: Tetratricopeptide (IPR011990); ADH: Alcohol dehydrogenase (IPR011047).