Effects of vlsE complementation on the infectivity of Borrelia burgdorferi lacking the linear plasmid lp28-1

Abstract

The loss of linear plasmid lp28-1, which contains the vls antigenic variation locus, is associated with reduced infectivity of Borrelia burgdorferi in immunocompetent mice. The recombinant shuttle vector pBBE22, which includes the virulence determinant BBE22 from lp25 and restores infectivity to readily transformable B. burgdorferi lacking lp25 and lp56, was used to determine the effect of trans expression of vlsE on virulence. Spirochetes lacking lp28-1 were complemented with the plasmid pBBE22:vlsE, containing both BBE22 and vlsE. VlsE protein produced by this construct was expressed and surface accessible in in vitro-cultured B. burgdorferi, as determined by surface proteolysis and immunoblot analysis. Clones lacking lp25 but containing lp28-1 and either pBBE22 or pBBE22:vlsE were reisolated consistently from immunocompetent mice 8 weeks after infection. In contrast, a clone lacking both lp25 and lp28-1 and complemented with pBBE22:vlsE was isolated from only a single tissue of one of six C3H/HeN mice 8 weeks postinfection. These results indicate that either an intact vls antigenic variation locus or another determinant on lp28-1 is required to restore complete infectivity. In addition, an isogenic clone that retained lp28-1 was complemented with the vlsE shuttle plasmid and was examined for vlsE sequence variation and infectivity. Sequence variation was not observed for the shuttle plasmid, indicating that the cis arrangement of vlsE and the vls silent cassettes in lp28-1 facilitate vlsE gene conversion. Lack of vlsE sequence variation on the shuttle plasmid thus did not result in clearance of the trans-complemented strain in immunocompetent mice under the conditions tested.

FIG. 1.

pBBE22:vlsE plasmid map. KpnI and XbaI restriction sites used during cloning of segments containing BBE22 and vlsE into the pBSV2 shuttle vector are shown (20, 41, 46).

FIG. 2.

Western blot analysis demonstrating VlsE expression in vlsE-complemented B. burgdorferi clones. Lane 1, 5A4 (wild type, containing all of the native B. burgdorferi plasmids); lane 2, 5A8 (lacking the VlsE-encoding plasmid lp28-1); lane 3, 5A13/pBBE22 (lp25⁻, complemented with BBE22); lane 4, 5A10/pBBE22:vlsE-2 (lp25⁻ and lp28-1⁺, complemented with BBE22 and vlsE); lane 5, 5A10/pBBE22:vlsE-5 (lp25⁻ and lp28-1⁻, complemented with BBE22 and vlsE). Numbers at the left are molecular weight standards (in thousands). A quantity of bacterial lysate equivalent to 10⁷ organisms was loaded in each lane.

FIG. 3.

Effect of surface proteolysis on VlsE in vlsE-complemented B. burgdorferi clones. Intact B. burgdorferi were treated with proteinase K for 40 min and then subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (10⁷/lane) and Western blotting. Two identical blots were reacted with either mouse polyclonal anti-VlsE (left panel) or a mouse monoclonal antibody against the flagellar protein FlaB (right panel). Lanes 1, clone 5A4 (containing all native B. burgdorferi plasmids); lanes 2 and 3, clone 5A10/pBBE22:vlsE-5 (lacking lp25, lp56, and lp28-1 but containing shuttle vector pBBE22:vlsE); lanes 4 and 5, clone 5A10/pBBE22:vlsE-2 (lacking lp25 and lp56 but containing both lp28-1 and pBBE22:vlsE). Proteinase K treatment of intact B. burgdorferi (lanes 3 and 5) dramatically reduced VlsE immunostaining but did not affect levels of the periplasmic protein FlaB. Numbers on the left are molecular weight markers (in thousands).

FIG. 4.

Alignment of predicted amino acid sequences of cassette regions of VlsE from B. burgdorferi 5A13/pBBE22 reisolated from joints (jt) and hearts (ht) of mice 2 and 8 weeks postinoculation. The cassette region of VlsE from B. burgdorferi 5A13/pBBE22 was used to infect mice. Identical amino acid sequences are indicated by periods, and gaps are indicated by hyphens. The variable regions (52) are indicated by shaded boxes.

FIG. 5.

IgG response elicited by infected mice to B. burgdorferi and VlsE recombinant protein. Graphs represent mean absorbencies obtained by ELISAs using B. burgdorferi sonicates and recombinant VlsE as antigens (see Materials and Methods).

FIG. 6.

Quantitation of B. burgdorferi in tibiotarsal joints of C3H/HeN mice. All tissues were assessed in a blinded fashion, and each symbol represents the average value for a single animal. Triangles, samples taken 2 weeks postinfection; diamonds, samples taken 4 weeks postinfection; squares, samples taken 8 weeks postinfection; *, absence of samples for this strain and time point.