Striking diversity of vmp1, a variable gene encoding a putative membrane protein of the stolbur phytoplasma

Abstract

Studies of phytoplasma-insect vector interactions and epidemiological surveys of plant yellows associated with the stolbur phytoplasma (StolP) require the identification of relevant candidate genes and typing markers. A recent StolP genome survey identified a partial coding sequence, SR01H10, having no homologue in the "Candidatus Phytoplasma asteris" genome but sharing low similarity with a variable surface protein of animal mycoplasmas. The complete coding sequence and its genetic environment have been fully characterized by chromosome walking. The vmp1 gene encodes a protein of 557 amino acids predicted to possess a putative signal peptide and a potential C-terminal transmembrane domain. The mature 57.8-kDa VMP1 protein is likely to be anchored in the phytoplasma membrane with a large N-terminal hydrophilic part exposed to the phytoplasma cell surface. Southern blotting experiments detected multiple sequences homologous to vmp1 in the genomes of nine StolP isolates. vmp1 is variable in size, and eight different vmp1 RsaI restriction fragment length polymorphism types could be distinguished among 12 StolP isolates. Comparison of vmp1 sequences revealed that insertions in largest forms of the gene encode an additional copy of a repeated domain of 81 amino acids, while variations in 11-bp repeats led to gene disruption in two StolP isolates. vmp1 appeared to be much more variable than three housekeeping genes involved in protein translation, maturation, and secretion and may therefore be involved in phytoplasma-host interactions.

FIG. 1.

(A) Genome walking and assembly of the final chromosomal fragment of 4,939 bp. SSH fragments (7) are shown as light gray boxes, and the GenomeWalker PCR fragments are represented by white boxes. The restriction sites indicated in white boxes indicate the type of library that allowed producing the PCR genome walking fragment. (B) Localization of the three coding sequences corresponding to the complete VMP1 protein, the N-terminal part of NAD-dependent DNA ligase, and the beginning of the excinuclease ATPase subunit UVRA. Gray arrows in vmp1 indicate 11-bp direct repeats where duplication (isolate T2_92) or deletion (isolate Moliere) occurred. Black arrows indicate primers 1H10F and 1H10R. (C) Structural domains of the VMP1 protein. SP, N-terminal signal peptide (27 amino acids); TM, transmembrane alpha helix (22 amino acids [aa]); B (84 amino acids) and B′ (80 amino acids), repeated domains.

FIG. 2.

Southern blots for vmp1 detection in the StolP genome. (A) Hybridization at high stringency with an SR1H10 digoxigenin-labeled PCR-amplified probe. Lanes: 3 and 5, healthy periwinkle total DNA; 2 and 4, StolP PO-infected periwinkle DNA; 2 and 3, undigested; 4 and 5, digested with HindIII; 1, the native pGEMT-Easy plasmid containing the SR1H10 SSH product. (B) Hybridization at low stringency with an SR1H10 digoxigenin-labeled PCR-amplified probe hybridized to HindIII-digested DNA from periwinkle infected by different StolP isolates (indicated above the lanes).

FIG. 3.

(A) Size polymorphism of PCR products obtained with primers H10F1 and H10R1 (vmp1) from total DNA of periwinkles infected with different StolP isolates (indicated above the lanes) (Table 2). No amplification was detected for stolbur isolate C, healthy periwinkle DNA, and H₂O. Electrophoresis was performed on a 0.8% agarose gel. (B) RsaI RFLP analysis of H10F1/R1 products (vmp1) on an 8% polyacrylamide gel. (C) RFLP analysis of tuf amplicons with HpaII restriction enzyme. Electrophoresis was performed on a 1.5% agarose gel. M, 1-kb ladder from Invitrogen.

FIG. 4.

Evolutionary relationships of secY-map and vmp1 genetic loci for 12 StolP isolates. The evolutionary history was inferred using the maximum-parsimony method. (A) One tree out of the 170 most parsimonious trees for the map-secY genetic locus; (B) 1 tree out of the 4 most parsimonious trees for vmp1. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 replicates) is shown next to the branches. The trees are drawn to scale, with branch lengths calculated using the average pathway method, and represent the number of nucleotide changes over the whole sequence. All positions containing gaps and missing data were eliminated from the data set. There were a total of 1,888 positions in the final data set for the map-secY genetic locus and 1,522 positions in the final data set for vmp1. Phylogenetic analyses were all conducted with MEGA4.