N-acetylgalactosamine utilization pathway and regulon in proteobacteria: genomic reconstruction and experimental characterization in Shewanella

Abstract

We used a comparative genomics approach to reconstruct the N-acetyl-d-galactosamine (GalNAc) and galactosamine (GalN) utilization pathways and transcriptional regulons in Proteobacteria. The reconstructed GalNAc/GalN utilization pathways include multiple novel genes with specific functional roles. Most of the pathway variations were attributed to the amino sugar transport, phosphorylation, and deacetylation steps, whereas the downstream catabolic enzymes in the pathway were largely conserved. The predicted GalNAc kinase AgaK, the novel variant of GalNAc-6-phosphate deacetylase AgaA(II) and the GalN-6-phosphate deaminase AgaS from Shewanella sp. ANA-3 were validated in vitro using individual enzymatic assays and reconstitution of the three-step pathway. By using genetic techniques, we confirmed that AgaS but not AgaI functions as the main GalN-6-P deaminase in the GalNAc/GalN utilization pathway in Escherichia coli. Regulons controlled by AgaR repressors were reconstructed by bioinformatics in most proteobacterial genomes encoding GalNAc pathways. Candidate AgaR-binding motifs share a common sequence with consensus CTTTC that was found in multiple copies and arrangements in regulatory regions of aga genes. This study provides comprehensive insights into the common and distinctive features of the GalNAc/GalN catabolism and its regulation in diverse Proteobacteria.

FIGURE 1.

Reconstruction of GalNAc utilization pathways in Proteobacteria. Different pathway variants in E. coli and Shewanella are highlighted by yellow and green background arrows, respectively. Multiple nonorthologous variants of proteins for several functional roles are listed in the same box and marked by uppercase Roman numerals.

FIGURE 2.

Genomic organization of the GalNAc utilization pathway genes. Genes and candidate AgaR-binding sites are shown by arrows and circles, respectively. Genes are colored according to their functional roles in Fig. 1 and named by the last letter of the corresponding protein. Genes from the same genetic locus that are not located immediately next to each other are separated by a slash. Genes from different genetic loci are separated by a double slash or are shown within two lines united by braces. Candidate AgaR sites are colored according to the corresponding AgaR-binding motifs shown at the bottom. Sequence logos for AgaR-binding motifs were generated by the WebLogo package. Genomic locus tags for all displayed genes are available in supplemental Table S2.

FIGURE 3.

Quantitative RT-PCR analysis of aga gene expression. Total RNA was isolated from Shewanella sp. ANA-3 cells grown on 10 mm GalNAc or GlcNAc as a sole carbon source. The expression levels of each gene were normalized to the gene expression in the GlcNAc-grown cells. Average ± S.D. (error bars) for six independent experiments are shown.

FIGURE 4.

In vitro reconstitution of the Shewanella GalNAc pathway. A, biochemical transformations in the GalNAc pathway and the assay used to assess the pathway reconstitution. The pathway product, ammonium, was detected at 340 nm by coupling to NADH-NAD⁺ conversion via glutamate dehydrogenase. B, pathway reconstitution by using the purified Shewanella sp. ANA-3 enzymes. All samples contained 2 mm GalNAc and 5–10 μg of each enzyme.

FIGURE 5.

Effect of agaI or agaS gene deletion on GalNAc-dependent cell growth of E. coli. A, growth of ΔagaI and ΔagaS mutant strains compared with the wild type (WT) strain of E. coli ATCC 8739. B, complementation of the ΔagaS mutant of E. coli ATCC 8739 by introducing plasmid constructs expressing both agaS and agaY genes or only the agaY gene from E. coli ATCC 8739. Cells were grown in M9 minimal medium containing 10 mm GalNAc as a sole carbon source. The cell growth was monitored by measuring the A_{600 nm}.