Research Article: R pyocin tail fiber structure reveals a receptor-binding domain with a lectin fold

Date Published: February 5, 2019

Publisher: Public Library of Science

Author(s): Adam J. Salazar, Mukul Sherekar, Jennifer Tsai, James C. Sacchettini, Olga Mayans.

http://doi.org/10.1371/journal.pone.0211432

Abstract

R pyocins are ɸCTX-like myophage tailocins of Pseudomonas sp. Adsorption of R pyocins to target strains occurs by the interaction of tail fiber proteins with core lipopolysaccharide (LPS). Here, we demonstrate that N-terminally truncated R pyocin tail fibers corresponding to a region of variation between R-subtypes are sufficient to bind target strains according to R-subtype. We also report the crystal structures of these tail fiber proteins and show that they form an elongated helical trimer composed of three domains arranged linearly from N- to C-terminus: a baseplate proximal head, medial shaft, and distal foot. The head and shaft domains contain novel structural motifs. The foot domain, however, is composed of a conserved jellyroll fold and shares high structural similarity to the tail fiber of myophage AP22, podophage tailspike C-terminal domains (LKA-1 and ɸ297), and several eukaryotic adhesins (discoidin I/II, agglutinin, and octocoral lectin). Many of these proteins bind polysaccharides by means of their distal loop network, a series of highly variable loops at one end of the conserved jellyroll fold backbone. Our structures reveal that the majority of R-subtype specific polymorphisms cluster in patches covering a cleft formed at the oligomeric interface of the head domain and in a large patch covering much of the foot domain, including the distal loop network. Based on the structural variation in distal loops within the foot region, we propose that the foot is the primary sugar-binding domain of R pyocins and R-subtype specific structural differences in the foot domain distal loop network are responsible for binding target strains in an R-subtype dependent manner.

Partial Text

Discovered in 1954 by François Jacob, pyocins are highly diverse peptide inhibitors of pseudomonad growth [1–3]. They include S, F, L, and R-types, which are further classified into subtypes by strain sensitivity [1, 4–7]. Of these, F and R pyocins resemble phage tail assemblies in both structure and function, and are more accurately categorized as “tailocins”. Tailocins are prophages that lack head structural genes and genome packaging mechanisms, but retain conserved elements of phage tails, including the inner tube, outer sheath, baseplate, and tail fibers [1, 4]. F pyocins are closely related to lambda-like siphophages by morphology and genomic organization, and contain flexible, non-contractile tails [1, 5]. R pyocins, on the other hand, are related to the P2-like myophage ɸCTX, and contain rigid, contractile tails [1]. They are also auspiciously resistant to protease treatment, temperatures up to 60°C, and inactivation by sterilizing UV exposure [8].

R2-type pyocins from P. aeruginosa strain PAO1 were purified and tested for bacteriocidal activity against known R1 producing strain, LESB58, by agar overlay spotting assay. As anticipated, we found that strain PAO1 was resistant to R2-type pyocin treatment and strain LESB58, on the other hand, was highly sensitive to R2-type pyocins (Fig 2A). To determine if the R2-NTF is sufficient to bind target cells according to the observed R-subtype pyocin sensitivity pattern, R2-NTF was incubated with strain PAO1 or LESB58. Following several rounds of centrifugation and washing to remove unbound R2-NTF, cells were lysed and protein samples normalized for total bacterial protein by SDS-PAGE. We observed (n = 3) in an anti-his western blot that the R2-NTF was specifically bound to LESB58, but not PAO1, confirming that the region of dense polymorphism encapsulated by the R2-NTF defines R-subtype specificity (Fig 2B).

In our investigation, we successfully elucidated the structures of R1 and R2-subtype pyocin tail fibers, allowing us to describe the R-subtype variable region, its domain composition, and cell binding function. These structures are in agreement with the recently reported structures of Buth et al (2018) that contain different fragments of R1 and R2-subtype tail fibers than described in this report. The R-subtype variable region investigated here contains 99% of the polymorphic variation between R-subtype tail fibers. Because there is no known specific conserved function of tail fibers beyond binding host or target cells, we hypothesized that sequence differences between R-subtype tail fibers would ostensibly correspond to residues involved in R-subtype selective binding and most of the residues involved in these interactions would belong to the variable region. We, therefore, propose that the head or foot domains found within this variable region contain putative LPS binding motifs. Our structures suggest that the discoidin-like “foot” is responsible for strain specificity and core LPS binding. In our model, R pyocin specificity by means of core LPS sugar binding is accomplished by host sugar interactions with the distal loop network of the foot domain. Supporting our assertion is evidence that structurally similar distal loop regions of discoidin and related lectins interact with polysaccharides (see below). In both our R1 and R2-subtype tail fiber structures, the distal loop network contains charged metal binding sites atypical in characterized discoidin-like lectins. However, unrelated lectin structures contain charged metal binding sites in which Ca2+ interactions stabilize residues of a sugar binding groove indirectly through orienting residues involved in sugar binding (Concavalin A, PDB:1I3H) and, in which Ca2+ is directly involved in coordinating hydroxyl interactions of the target carbohydrate (Rat mannose binding C-type lectin, PDB:2SMB) [25]. Therefore, we have reason to believe that metal interactions are involved in mediating LPS binding of R pyocin tail fibers.

 

Source:

http://doi.org/10.1371/journal.pone.0211432

 

Leave a Reply

Your email address will not be published.