The Gene ail for the Attachment-Invasion Locus Protein of Yersinia enterocolitica Biotype 1A Strains Is Located on the Genome of Novel Prophages

1. Introduction

Yersiniosis is an infectious disease of the gastrointestinal tract caused by Yersinia (Y.) enterocolitica and, to a lesser extent, by Y. pseudotuberculosis [1,2]. It is the third most common bacterial enteritis in Europe. Infections are mainly caused by the consumption of raw or undercooked pork [3,4,5,6,7,8,9]. Y. enterocolitica comprises six biotypes (BT) and up to 70 serotypes. While the BTs 1B, 2, 3, 4 and 5 possess the 70 kb virulence plasmid pYV encoding a type III secretion system and effector proteins (Yersinia outer proteins = YOPs) with in part toxic activity for eucaryotic cells, BT 1A strains are generally devoid of pYV [10]. Moreover, some important chromosomally encoded virulence factors of pathogenic biotypes are only rarely or not existing in BT 1A [11,12]. Two of those are the enterotoxin YstA and the attachment-invasion locus protein (Ail), which is involved in both the invasion of eucaryotic cells as well as in serum resistance [13,14,15]. The ail gene has yet only been detected in some BT 1A strains [16,17]. Because of the lack of these virulence factors, BT 1A strains have been considered to be non-pathogenic for a long time. However, during the last years, there is an increasing number of reports describing BT 1A isolates from clinical cases [18,19,20,21,22,23,24,25,26]. This biotype obviously comprises at least two phylogenetic lineages, each with different virulence factors, some of which are toxins [17,27]. The question arises, whether horizontal gene transfer may be involved in the heterogeneity of this biotype. In this study, we sequenced 370 BT 1A isolates from different sources, of which 41 contain the virulence gene ail. A closer analysis of this gene revealed that unlike in the strictly pathogenic biotypes, ail of most BT 1A isolates is located on the genome of prophages, which were characterized by in silico analyses.

2. Results

2.1. Forty-One Out of 370 BT 1A Genomes Contain a Prophage-Associated Ail Gene.

Sequence analyses of 370 Y. enterocolitica BT 1A isolates revealed that they harbor the gene ail. They were isolated between 2013 and 2025 from different sources (mainly minced meat and wild boars) and belong to 17 different MLST allele profiles (Figure 1A). All isolates possess the enterotoxin gene ystB, 30, 10 and 3 of them additionally the virulence-associated genes hreP, myfA and tccC, respectively (Table S1). For almost all isolates, plasmid-borne sequences of approximately 3 to 106 kb were predicted.

A closer look at the ail gene of these isolates showed that it is up to 99% identical to ail of pathogenic biotypes, e.g. the bio/serotype 1B/O:8 strain 8081. In addition, the upstream sequence containing the promoter and ribosome binding site of ail are similarly related to their counterpart in other biotypes suggesting that ail may be active in BT 1A.

However, the regions encompassing ail diverge significantly in the various biotypes. In most BT 1A isolates like in 24-YE00064 studied here, the gene is surrounded by partitioning genes and genes for cell lysis (lysin and holin) and DNA packaging (small and large terminase) typically associated with phage (Table 1). Moreover, genes for phage assembly (capsid and tail), the genetic switch as well as an integrase and excisionase were also identified in most chromosomes suggesting that ail is part of a prophage (Table S2).

2.2. Comparison of the Prophages Indicates Relationships Between Them.

The 41 prophage sequences identified in the investigated isolates by comparison with the vB_Yen-24-YE00064 prophage form five major clusters (Figure 1A), of which the prophages in the clusters C2 to C5 harbor the ail gene, whereas prophages in cluster C1 are devoid of ail, since here, the gene is located at a different position on the chromosome (see below). An alignment disclosed that most regions of the prophages, particularly those encoding structural proteins, are closely related, whereas e.g. the integrase genes show major differences (Figure 1B). Short read sequencing allowed the prediction of four whole prophage genomes (vB_Yen_16-YE00051, vB_Yen_20-YE00187, vB_Yen_24-YE00064 and vB_Yen_25-YE00027) belonging to four clusters (Figure 1A). They have genome sizes of 34,918 to 50,744 bp and are composed of 55 to 78 Open Reading Frames (ORFs, Table S1). The overall genome organization of the prophages is similar (Figure 2A). As with other temperate phages, ORFs for repressor proteins, cell lysis, DNA packaging, capsid and tail assembly are clustered, even though some ORFs, particularly those encoding structural proteins, are missing in vB_Yen_16-YE00051. The prophages showed no identities to other phages and only some relatedness to two Y. enterocolitica BT 1A chromosomes (Y201, CP124238.1 and Y115, CP124259.1).

Three prophages (vB_Yen_16-YE00051, vB_Yen_20-YE00187 and vB_Yen_25-YE00027) have an identical attachment site att of 40 bp (Figure 2B). Their integrases are 100% identical (Figure 2C). By contrast, the prophage vB_Yen_24-YE00064 has an att site of only 27 bp (Figure 2B). The integrase of this prophage is only approximately 35% identical to those of the other ones (Figure 2C). Thus, it is not surprising that the two groups have different integration sites on the bacterial chromosome. While the prophages vB_Yen_16-YE00051, vB_Yen_20-YE00187 and vB_Yen_25-YE00027 are integrated between two genes for hypothetical proteins, vB_Yen_24-YE00064 is integrated between a gene for an integrase and a YebY family protein.

2.3. Relocation of ail in Cluster C1 and Analysis of 1B/O:8 Strains.

Sequence analysis of cluster C1 revealed similar prophage sequences as in isolate 24-YE00064. However, at the position of ail in vB_Yen_24-YE00064, there is a gap in the C1 prophage genomes (Figure 3A). In this cluster, ail is located approximately 600 kb apart from the vB_Yen_24-YE00064-related prophage sequences. Interestingly enough, the gene and its adjacent sequences are surrounded by DNA segments, which are similarly present in the corresponding prophages, as shown for the prophage vB_Yen_23-YE00044.2 (Figure 3B). The homologous upstream and downstream sequences of ail have a length of 144 bp and 925 bp, respectively, but are not related to each other. Therefore, it remains open how ail was relocated in isolates belonging to cluster C1. Nevertheless, it is noteworthy that even in the highly pathogenic Y. enterocolitica 1B/O:8 strains 8081 (AM286415.1), WA (CP009367.1) and Billups-1803-68 (CP173224.1), ail is associated with phage genes. Indeed, a stretch of approximately 20 kb of strain 8081 containing ail is similar to vB_Yen_24-YE00064 (Figure 3C). This stretch essentially corresponds to ØYE200 identified in 8081 (Thomson et. al, 2006), which, however, has been determined as a smaller prophage (15.5 kb) without ail. Besides ail, the 20 kb prophage of strain 8081 comprises genes for e.g. an integrase, lysis proteins (holin and lysin) and the terminase large subunit. Moreover, the fact that this DNA segment also contains the 27 bp attachment site of vB_Yen_24-YE00064 upstream of the integrase gene and that is linked to tRNA genes suggests that ail was once associated with a similar prophage. It is conspicuous that the DNA segment in strain 8081 harbors several transposase genes which are lacking in vB_Yen_24-YE00064 and which might have been involved in genetic reassortments.

3. Discussion

The attachment-invasion locus protein Ail of Y. enterocolitica is an important virulence factor, which is produced by all pathogenic biotypes of this species, as well as by Y. pseudotuberculosis and Y. pestis [28]. It has yet only rarely been described in BT 1A strains of Y. enterocolitica and in Y. enterocolitica-like species, e.g. Y. kristensenii, some of which have been reported to contain an additional ail-related gene which may be associated with plasmids or phages [11,29,30]. For that reason, ail is routinely used as target for the detection of pathogenic Y. enterocolitica and Y. pseudotuberculosis by RT-PCR (ISO TS 18867:2015). Though, sequencing of 370 BT 1A isolates showed that ail is more commonly present in this biotype as expected. We identified the gene in 41 (11.1%) out of 370 isolates recovered from food and wild boars in the last 12 years. The ail-positive isolates represent a broad range of MLST alleles profiles, even though some types (ST304, ST428, ST832) were prevailing. The question arises, how BT 1A strains may acquire ail. This study suggests that it may occur by lysogenic conversion via temperate phages. Indeed, the analysis of ail-positive BT 1A isolates showed that in most of them the gene is located on a prophage. The prophages are related to each other, but form five different clusters. Up to now, four prophage genomes containing ail could be analysed in detail. The analysis suggests that three of them (vB_Yen_20-YE00187, vB_Yen_24-YE00064 and vB_Yen_25-YE00027) may be active, since all elements required for the formation of a phage particle are obviously present, whereas the prophage vB_Yen_16-YE00051 is presumably defective, because some essential ORFs for capsid and tail proteins are lacking. The fact that genes for structural proteins of the complete prophages are very similar indicates that the corresponding phage particles may have the same morphology. A striking difference, however, pertains to the integration site of them on the Y. enterocolitica chromosome. Unlike some other parts of their genomes, genes for the integrase and the attachment sites are in part highly diverse. As a consequence, the prophages are not integrated at the same position in BT 1A strains. Whether these sites also exist in other Y. enterocolitica biotypes or even other Yersinia species and whether those strains may also acquire ail by phage-mediated transfer has still to be studied. It has to be taken into account that for lysogenic conversion, the host range of a temperate phage is of major importance. Yersinia enterocolitica BT 1A strains belong to various serotypes, which may determine the host specificity of a phage. It appears, however, that ail-prophages are subjected to genetic recombinations. This can be clearly seen in cluster C1 comprising related prophages, in which ail including adjacent sequences has been relocated. Similarly, highly pathogenic 1B strains like 8081 contain remnants of ail-prophages suggesting recombination events or even a horizontal gene transfer in the past. Regrettably, information on temperate Y. enterocolitica phages and their potential to exchange genes is scarce [31,32]. We will therefore now determine the inducibility of the identified ail-prophages and phenotypic properties of possibly produced phage particles.

4. Materials and Methods

4.1. Typing of Y. enterocolitica Strains.

Isolates were initially cultivated on Columbia agar supplemented with 5% sheep blood (bioMérieux Deutschland GmbH, Nürtingen, Germany) at 28°C for 16-20 hours for whole-cell matrix-assisted laser-desorption/ionization time-of-flight mass spectrometrical identification (MALDI-TOF MS) using the direct transfer method with HCCA matrix on a Biotyper (Bruker Daltonics GmbH & Co. KG, Bremen, Germany). In addition, physiological and biochemical tests using classical tube and plate procedures for species confirmation and biochemical differentiation were conducted as previously described [33]. Unless otherwise indicated, YP cultivation was conducted at aerobic conditions at 28°C for 18-24 hours using lysogeny broth (LB)-based media. For solid media preparation, LB medium was supplemented with 1.8% bacto agar no. 1 (Oxoid Deutschland GmbH, Wesel, Germany) [34,35].

4.2. Genome Sequencing and Bioinformatics Analysis

Whole-genome sequencing (WGS) of Y. enterocolitica BT 1A isolates was performed by short-read, paired-end sequencing (2 x150 cycles) on a NextSeq500 benchtop device (Illumina Inc., San Diego, CA, USA). Bacterial genomic DNA was extracted from liquid cultures grown at 37°C for 20-24 hours using the PureLink Genomic DNA Mini Kit (Invitrogen, Ebersberg, Germany) according to the recommendation of the manufacturers. DNA sequencing libraries were prepared using the Nextera XT DNA Sample Preparation Kit (Illumina Inc.) [31,36,37,38]. Raw sequencing data were subjected to the Aquamis pipeline (Deneke et al., 2021) for quality evaluation, demultiplexing and trimming, while general in silico typing purposes were conducted using BakCharak (https://gitlab.com/bfr_bioinformatics/bakcharak; access: Jul-2025) [39]. Prophage detection was conducted using the Phastest tool for initial screening, while manual curation using Accelrys DS Gene (v2.5; Accelrys Inc., San Diego, CA, USA) was performed to determine the complete prophage sequence from attachment sites of the bacteria (attB) and prophages (attP). Genome annotation of the prophage genomes was conducted using the Bacterial and Viral Bioinformatics Resource Center (BV-BRC). Initial functional prediction of open reading frames (ORFs) was manually curated according to predicted functions of closely related protein sequences derived from blastp searches at NCBI (National Center for Biotechnology Information) [31,36,37,38]. Phylogenetic relationship of vB_Yen_24-YE00064 (reference) to Y. enterocolitica genome datasets encoding Ail were conducted using CSI phylogeny (v1.4; default settings; https://cge.food.dtu.dk/services/CSIPhylogeny/; access: Sep-2025)

4.3. Genome Accession Numbers

Deposition of prophage genomes at NCBI Genbank was conducted using BankIt for the prophages vB_Yen_24-YE00064 (PV779719), vB_Yen_16-YE00051 (SUBMITTED), vB_Yen_20-YE00187 (PX109664) and vB_Yen_25-YE00027 (PX109665).