Genetic Organization of the aprX-lipA2 Operon Affects the Proteolytic Potential of Pseudomonas Species in Milk

Abstract

Psychrotolerant Pseudomonas species are a main cause of proteolytic spoilage of ultra-high temperature (UHT) milk products due to the secretion of the heat-resistant metallopeptidase AprX, which is encoded by the first gene of the aprX-lipA2 operon. While the proteolytic property has been characterized for many different Pseudomonas isolates, the underlying aprX-lipA2 gene organization was only described for a few strains so far. In this study, the phylogenomic analysis of 185 Pseudomonas type strains revealed that the presence of aprX is strongly associated to a monophylum composed of 81 species, of which 83% carried the aprX locus. Furthermore, almost all type strains of known milk-relevant species were shown to be members of the three monophyletic groups P. fluorescens, P. gessardii, and P. fragi. In total, 22 different types of aprX-lipA2 genetic organizations were identified in the genus, whereby 31% of the species tested carried the type 1 operon structure consisting of eight genes (aprXIDEF prtAB lipA2). Other genetic structures differed from type 1 mainly in the presence and location of genes coding for two lipases (lipA1 and lipA2) and putative autotransporters (prtA and prtB). The peptidase activity of 129 strains, as determined on skim milk agar and in UHT-milk, correlated largely with different aprX-lipA2 gene compositions. Particularly, isolates harboring the type 1 operon were highly proteolytic, while strains with other operon types, especially ones lacking prtA and prtB, exhibited significantly lower peptidase activities. In conclusion, the phylogenomic position and the aprX-lipA2 gene organization specify the proteolytic potential of Pseudomonas isolates. In addition, however, an interplay of several environmental factors and intrinsic traits influences production and activity of AprX, leading to strain-specific proteolytic phenotypes.

Keywords: AprX peptidase; aprX-lipA2 operon; genus Pseudomonas; milk spoilage; proteolytic potential.

Figure 1

Phylogenomic tree of the genus Pseudomonas. Shown is the rooted maximum likelihood phylogenomy of the genus Pseudomonas comprising 185 type strains. The tree is based on a multiple sequence alignment composed of 92 universal bacterial core genes (69,704 alignment positions). Evolutionary distances were estimated using the GTR+G+I model. Branches with high bootstrap support (≥70%) are marked with blue circles. In total, 200 bootstrap replicates were calculated. Cellvibrio japonicus Ueda107^T (RefSeq ID NC_010995.1) is used as outgroup. For reasons of clarity the length of the outgroup branch was downscaled to 0.05 substitutions per site (dashed branch). Black triangles highlight type strains containing the aprX gene. Black stars refer to type strains of species classified as milk-relevant in the past. The genus is separated into 22 distinct monophyletic groups and 8 singletons. Group names are shown at the outer-most border.

Figure 2

(Continued)FIGURE 2Proteolytic activity and aprX-lipA2 genetic organizations of selected Pseudomonas strains. The rooted phylogenomy shows 178 Pseudomonas strains for which an agar diffusion assay was performed and/or the aprX gene was found. The maximum-likelihood method and the GTR+G+I model were applied to reconstruct the tree based on 92 concatenated universal bacterial core genes (75,078 alignment positions). Branches containing blue circles are of high bootstrap support (≥70% of 200 replicates). The monophylum represented by strains P. psychrotolerans DSM 15758^T, P. oryzihabitans DSM 6835^T, and P. oryzihabitans WS 5017 was used to root the tree (outgroup). Strain-specific proteolytic activity after 3, 4, and 7 days is visualized as heatmap illustrating non (dark blue), weak (light blue), moderate (yellow), strong (light red) and very strong (dark red) activities. AprX-lipA2 genetic organizations are shown at single-nucleotide resolution next to the heatmap and in a strain-wise manner. Coding sequences not belonging to one of the genes aprX (green), aprI (black), aprD (brown), aprE (orange), aprF (yellow), prtA (magenta), prtB (cyan), lipA1 (pink), and lipA2 (antique pink) are colored gray. Individual genes or gene clusters of the aprX-lipA2 operon being present at distinct genomic locations are separated by a double slash (//). In such cases, the different portions of genetic information do not necessarily have to lie on the same strand (be encoded in the same direction). Strains containing partial gene sequences are marked with an asterisk (*). In this context, left-open gene sequences indicate that the 5'-portion of the coding sequence is missing, whereas right-open gene sequences highlight cases lacking the 3'-part. Each aprX-lipA2 genetic organization is followed by its type number. To provide a more structured overview the phylogenomy is divided into 18 monophyletic groups and two singletons (P. massiliensis CB1^T and P. yamanorum LMG 27247^T). Group names are listed on the right-hand of the figure.

Figure 3

Most abundant aprX-lipA2 genetic organizations. Schematic representation of the four most abundant aprX-lipA2 genetic organizations (type 1–4) in Pseudomonas, as determined by number of species per type. The double slash (//) indicates the presence of particular gene clusters at distinct genomic locations.

Figure 4

Proteolytic activities of most abundant aprX-lipA2 genetic organizations. Boxplot representation of proteolytic activities at day 4 with regard to the most abundant aprX-lipA2 genetic organizations, namely aprXIDEF prtAB lipA2 (type 1), aprXIDEF lipA2 (type 2), aprXIDEF | prtAB (type 3) and aprXIDEF prtAB lipA1A2 (type 4). Each dot refers to the average proteolytic activity of a particular Pseudomonas species. One-way ANOVA revealed a significant difference in mean proteolytic activity between aprX-lipA2 genetic organizations (p =1.7 x 10^-5). Post hoc Tukey’s Test indicates that significant differences exist between type 1 and all other genetic organizations.

Figure 5

Proteolytic activity of 28 Pseudomonas strains, grouped according to their aprX-lipA2 genetic organization into highly proteolytic (type 1, 9, 12) and middle to low proteolytic isolates (type 2, 3, 8). Extracellular peptidase activity of strains, grown in UHT milk at 12°C, was determined after day 3 and 4 via azocasein assay. Each strain was cultivated in triplicates and EPA measurement was carried out in duplicates for each sample; means were calculated and standard deviations are shown as error bars. Dashed lines point out the classification into low, moderate and highly proteolytic isolates according to proteolytic activity values determined on day 4.