Exposing the secrets of two well-known Lactobacillus casei phages, J-1 and PL-1, by genomic and structural analysis

Abstract

Bacteriophage J-1 was isolated in 1965 from an abnormal fermentation of Yakult using Lactobacillus casei strain Shirota, and a related phage, PL-1, was subsequently recovered from a strain resistant to J-1. Complete genome sequencing shows that J-1 and PL-1 are almost identical, but PL-1 has a deletion of 1.9 kbp relative to J-1, resulting in the loss of four predicted gene products involved in immunity regulation. The structural proteins were identified by mass spectrometry analysis. Similarly to phage A2, two capsid proteins are generated by a translational frameshift and undergo proteolytic processing. The structure of gene product 16 (gp16), a putative tail protein, was modeled based on the crystal structure of baseplate distal tail proteins (Dit) that form the baseplate hub in other Siphoviridae. However, two regions of the C terminus of gp16 could not be modeled using this template. The first region accounts for the differences between J-1 and PL-1 gp16 and showed sequence similarity to carbohydrate-binding modules (CBMs). J-1 and PL-1 GFP-gp16 fusions bind specifically to Lactobacillus casei/paracasei cells, and the addition of l-rhamnose inhibits binding. J-1 gp16 exhibited a higher affinity than PL-1 gp16 for cell walls of L. casei ATCC 27139 in phage adsorption inhibition assays, in agreement with differential adsorption kinetics observed for both phages in this strain. The data presented here provide insights into how Lactobacillus phages interact with their hosts at the first steps of infection.

FIG 1

Dot plot comparison of Lactobacillus phages J-1 and PL-1. A sequence file containing J-1 was compared against a file containing PL-1 using Gepard (91).

FIG 2

(A) Annotated genome maps of bacteriophages J-1 and PL-1. The viral genomes of J-1 and PL-1 are represented in four tiers, with markers spaced at 1-kbp and 100-bp intervals. The predicted genes are shown as boxes either above or below the genome, depending on whether they are rightward or leftward transcribed, respectively. Gene numbers are shown within each box. Putative genes can be divided in the following six modules: packaging (light blue), virion structure (yellow), lysis (purple), integration and immunity (red), and replication (orange). The putative proteins found in the extreme right region are colored in green, while the ORFs lacking function are white. (B) Global comparison of Lactobacillus phages. Two pairwise nucleotide alignment of phages J-1, PL-1, and related Lactobacillus phages (Lrm1, A2, phiAT3, and Lc-Nu) using Phamerator (43). The genomes are represented by horizontal lines, with putative genes shown as boxes above (transcribed rightward) or below (transcribed leftward) each genome; the number of each gene is shown within each box.

FIG 3

Analysis and functional assignment of J-1 and PL-1 structural proteins. Illustrated are electron micrographs of J-1 (left) and PL-1 (right) phage particles and SDS gel electrophoresis of virion proteins showing the predicted gene products. Molecular mass markers (Mk) are, from top to bottom, 170, 130, 100, 70, 55, 40, 35, and 25 kDa.

FIG 4

Translational frameshift of capsid and chaperone mRNAs. A translational −1 frameshifting near the end of transcribed genes 6 and 13 results in synthesis of two different-length products, gp6-gp7 (A) and gp13-gp14 (B), respectively. The proposed slippery sequence is shaded in gray. Amino acid sequences depicted in bold letters correspond to peptides detected by MALDI-MS for the short and long forms of the protein. The underlined sequence in gp6-gp7 (A) corresponds to the N-terminal peptide detected by MALDI-MS and confirms the predicted proteolytical processing.

FIG 5

gp16 alignment and structure prediction. (A) Amino acid sequence alignment of gp16 with similar proteins found in Lactobacillus phages, Orf19.1 of Spp1 and Orf46 of TP901-1. Uppercase letters correspond to aligned Pfam domain (Sypho_tail). (B) Predicted structure of gp16 of J-1 and PL-1 based on Spp1 Orf19.1 crystal structure (PDB code 2X8K, chain A). Dom1 and Dom2 correspond to the regions that could not be modeled with this PDB code. The template is shown in yellow, and the colors in the modeled structure correspond to the domains shown in panel A. (C) Predicted structure of Dom1 J-1 (orange) and Dom1 PL-1 (brown) of gp16 based on the carbohydrate-binding module (CBM) crystal structure of the endo-β-1,4-galactanase from Thermotoga maritima (PDB code 2XON, chain L) (gray).

FIG 6

gp17 alignment and structure prediction. (A) Amino acid sequence alignment of gp17 with similar proteins found in Lactobacillus phages and gp44 of phage Mu. Uppercase letters correspond to the aligned Pfam domain (Prophage_tail). (B) Predicted structure of the first 400 amino acids of gp17 of J-1 and PL-1 (green) based on Mu gp44 crystal structure (PDB code 1WRU) (yellow).

FIG 7

Kinetics of adsorption of J-1 and PL-1. L. paracasei subsp. paracasei ATCC 27092 (A) or L. casei subsp. casei ATCC 27139 (B) cells were incubated with J-1 (circles) or PL-1 (squares). At the indicated time points, PFU/ml in the supernatant was measured and the percentage of adsorption was calculated.

FIG 8

Binding of GFP-gp16 to Lactobacillus cells. Recombinant proteins J-1 GFP-gp16 and PL-1 GFP-gp16 were incubated with Lactobacillus casei subsp. casei ATCC 27139 (a-b), L. paracasei subsp. paracasei ATCC 27092 (c-d), Lactobacillus acidophilus (e-f), and L. casei subsp. casei ATCC 27139 in the presence of l-rhamnose (g-h) or glucose (i-j). Cells were visualized by phase-contrast microscopy (left image) and fluorescence microscopy (right image). Magnification, 1,000×.

FIG 9

Adsorption of J-1 and PL-1 in the presence of sugars. J-1 (gray bars) and PL-1 (black bars) were preincubated with 0.25 M the indicated sugars and further incubated with cell walls of L. casei subsp. casei ATCC 27139. PFU/ml in the supernatant was measured, and the percentage of adsorption compared to the control was calculated. The error bars represent the standard deviations from experiments done in triplicate.

FIG 10

Adsorption inhibition assays. Adsorption inhibition was determined when L. casei subsp. casei ATCC 27139 cell walls were incubated with increasing amounts of J-1 gp16 (gray bars) or PL-1 gp16 (black bars), followed by adsorption assays using phage J-1 (A) or PL-1 (B). The error bars represent the standard deviations from experiments done in triplicate.