Long-chain N-acyl amino acid synthases are linked to the putative PEP-CTERM/exosortase protein-sorting system in Gram-negative bacteria

Abstract

Clones that encode the biosynthesis of long-chain N-acyl amino acids are frequently recovered from activity-based screens of soil metagenomic libraries. Members of a diverse set of enzymes referred to as N-acyl amino acid synthases are responsible for the production of all metagenome-derived N-acyl amino acids characterized to date. Based on the frequency at which N-acyl amino acid synthase genes have been identified from metagenomic samples, related genes are expected to be common throughout the global bacterial metagenome. Homologs of metagenome-derived N-acyl amino acid synthase genes are scarce, however, within the sequenced genomes of cultured bacterial species. Toward the goal of understanding the role(s) played by N-acyl amino acids in environmental bacteria, we looked for conserved genetic features that are positionally linked to metagenome-derived N-acyl amino acid synthase genes. This analysis revealed that N-acyl amino acid synthase genes are frequently found adjacent to genes predicted to encode PEP-CTERM motif-containing proteins and, in some cases, other conserved elements of the PEP-CTERM/exosortase system. Although relatively little is known about the PEP-CTERM/exosortase system, its core components are believed to represent the putative Gram-negative equivalent of the LPXTG/sortase protein-sorting system of Gram-positive bacteria. During the course of this investigation, we were able to provide evidence that an uncharacterized family of hypothetical acyltransferases, which had previously been linked to the PEP-CTERM/exosortase system by bioinformatics, is a new family of N-acyl amino acid synthases that is widely distributed among the PEP-CTERM/exosortase system-containing Proteobacteria.

Fig. 1.

Long-chain N-acyl amino acids isolated from antibacterially active eDNA clones: N-tetradecanoyl tyrosine (1), N-tetradecanoyl phenylalanine (2), N-tetradecanoyl tryptophan (3), and N-tetradecanoyl arginine (4). An N-acylhomoserine lactone (5) is produced by LasI from P. aeruginosa PAO1 [N-(3-oxododecanoyl)homoserine lactone]. A lyso-ornithine lipid (6) is produced by OlsB from S. meliloti strain 1021 [N-(3-hydroxyhexadecanoyl)ornithine].

Fig. 2.

(A) Open reading frame maps of the eDNA inserts containing previously identified N-acyl amino acid synthase genes (either NasY or NasR genes) and genes encoding PEP-CTERM motif proteins. These maps are divided into two sections to separate those that contain whole or partial genes for PrsT, PrsK, and PrsR (middle section) from those that do not (top section). For comparison, the bottom section shows similar regions from the genomes of Nitrosococcus oceani ATCC 19707 and Thiobacillus denitrificans ATCC 25259. (B) In this overview of the proposed PrsK/PrsR-mediated regulation of PEP-CTERM motif protein expression, an unknown signal initiates phosphorelay between PrsK and PrsR. Phosphorylated PrsR then binds to a proposed response regulator binding motif and activates the alternate σ⁵⁴-RNA polymerase holoenzyme (Eσ⁵⁴) in an ATP-dependent process leading to transcription of downstream gene(s) encoding PEP-CTERM motif protein(s). PEP-CTERM motif proteins, anchored to the plasma membrane by their PEP-CTERM motifs, are subsequently cleaved by the transpeptidase EpsH, presumably for targeting across the outer membrane. The contributions of ExoAT and NAS-like enzymes to this process are currently unknown.

Fig. 3.

(A) UV-Vis chromatograms from the HPLC-MS analysis of organic extracts of E. coli heterologous expression constructs containing the NasY2 or ExoAT132 gene or empty vector. Prominent signals from N-acyltyrosines occur between minutes 12 and 17. (B) Atmospheric pressure ionization-positive (API+) mode ionization data corresponding to minutes 11 to 18 of the HPLC-MS analysis of the NasY2 (top panel) and ExoAT132 (bottom panel) genes. The peak with m/z (M+H)⁺ of 392, corresponding to 1 (N-tetradecanoyl tyrosine), is highlighted in both panels. The relatively larger peaks with m/z (M+H)⁺ of 418 and 420 that are marked by a single asterisk in the ExoAT132 gene panel correspond to the saturated and monounsaturated 16-carbon N-acyl derivatives of tyrosine. (C) Relative ion counts (electrospray ionization-positive [ESI+] mode) over the course of the HPLC-MS analysis of organic extracts of E. coli heterologous expression constructs containing the NasR or ExoAT11 gene or empty vector. (D) API-positive mode ionization data corresponding to minutes 10 to 18 of the HPLC-MS analysis of the NasR (top panel) and ExoAT11 (bottom panel) genes. The peak with m/z (M+H)⁺ of 385, corresponding to 4 (N-tetradecanoyl arginine), is highlighted in both panels. The largest peak in the ExoAT11 gene panel with m/z (M+H)⁺ of 439, marked by a single asterisk, corresponds to the monounsaturated 18-carbon N-acyl derivative of arginine.

Fig. 4.

16S rRNA phylogram of the 69 species of Proteobacteria whose genomes contain PrsK and PrsR genes (out of 71 species total). The phylogram was constructed using 16S sequences from the Silva comprehensive rRNA database SSU Ref 104, trimmed to the region corresponding to positions 133 to 1178 of the E. coli K-12 MG1655 16S rRNA gene rrsG, and aligned using Clustal W. The trimmed 16S rRNA sequence from Verrucomicrobiae bacterium DG1235 (a PrsK/PrsR-containing member of the phylum Verrucomicrobia) was used as an outgroup. For each ExoAT gene homolog (InterPro IPR022484) found within a genome, an orange dot is appended next to the species name. A purple dot is appended next to the name of each species whose genome contains a putative homolog of the group 1 NasY genes.