Lmb, a protein with similarities to the LraI adhesin family, mediates attachment of Streptococcus agalactiae to human laminin

Abstract

Streptococcus agalactiae is a leading cause of neonatal sepsis and meningitis. Adherence to extracellular matrix proteins is considered an important factor in the pathogenesis of infection, but the genetic determinants of this process remain largely unknown. We identified and sequenced a gene which codes for a putative lipoprotein that exhibits significant homology to the streptococcal LraI protein family. Mutants of this locus were demonstrated to have substantially reduced adherence to immobilized human laminin. The nucleotide sequence of the gene was subsequently designated lmb (laminin binding) and shown to be present in all of the common serotypes of S. agalactiae. To determine the role of Lmb in the adhesion of S. agalactiae wild-type strains to laminin, a recombinant Lmb protein harboring six consecutive histidine residues at the C terminus was cloned, expressed, and purified from Escherichia coli. Preincubation of immobilized laminin with recombinant Lmb significantly reduced adherence of the wild-type strain O90R to laminin. These results indicate that Lmb mediates the attachment of S. agalactiae to human laminin, which may be essential for the bacterial colonization of damaged epithelium and translocation of bacteria into the bloodstream.

FIG. 1

Amino acid sequence alignment of LraI proteins (S. pneumoniae PsaA, S. parasanguis FimA, S. sanguis SsaB, S. gordonii ScaA, S. crista ScbA, and E. faecalis EfaA) and S. agalactiae Lmb. Alignment and determination of consensus sequence were performed with the MultAlin program (http://www.toulouse.inra.fr). Amino acid residues of Lmb matching the consensus sequence are shown in boxes. Highly conserved residues (consensus level of 90%) are represented as capital letters in the consensus sequence; small letters denote a consensus level of ≥50%. !, I or V; $, L or M; %, F or Y; #, N, D, Q, E, or B. Parameters: gap weight, 12; gap length weight, 2.

FIG. 2

(A) Comparison of Lmb with the LraI domains. The Lmb protein (A) is designated by the open box showing a putative cleavage site for signal peptidase II. At amino acids (aa) 165 and 229, the plasmid insertion sites of the pG⁺host5 vector in mutants Lmb-k1 and Lmb-k2 are indicated. A region corresponding to the α region of LraI proteins with similarity to the B2 chain of human laminin is represented by a filled rectangle. (B) Structural features of the LraI family as proposed by Jenkinson (14). Numbers refer to amino acid residues, demarcating four regions: leader peptide, cleaved off by signal peptidase II; B1; B2; and the α region, which is presumed to be the solute binding domain. (C) Amino acid sequence alignment of the α region of Lmb with the laminin B2 chain. The alignment was performed with the BLASTp program at the National Center for Biotechnology Information web site.

FIG. 3

(A) Analysis of the lmb gene in various GBS serotypes by Southern hybridization. Genomic DNA was digested with EcoRI and transferred to a nylon membrane. Hybridization was performed with a Dig-dUTP-labeled fragment of lmb generated by PCR. Lanes: M, molecular size markers (in nucleotides); 1, GBS strain O90R; 2 to 11, specific serotypes as indicated above the lanes. (B) Southern analysis of GBS strain O90R and the isogenic mutant Lmb-k1. Genomic DNA of the parent (lanes 1 and 2) and (lanes 3 and 4) mutant strains was digested with EcoRI (lanes 1 and 3) or XbaI (lanes 2 and 4). Hybridization was performed with a probe directed to the internal fragment of lmb that was used for insertion duplication mutagenesis.

FIG. 4

Transcription analysis of the lmb locus by RT-PCR. RNA was extracted from S. agalactiae R268 and subjected to RT-PCR. Lanes: 1, DNA size marker; 2, PCR with chromosomal DNA as the template; 3, PCR with 1 μg of RNA as the template; 4, control for DNA contamination in which the RNA preparation was subjected to PCR without prior RT-PCR.

FIG. 5

Subcellular localization of Lmb. Bacterial cells were disrupted by a high-pressure cell homogenizer, separated into membrane and cytoplasmic fractions by centrifugation at 100,000 × g, and subjected to denaturing gel electrophoresis. Western immunoblotting was performed as described in Materials and Methods with a polyclonal anti-Lmb antibody (A) or preimmune serum (B) at a dilution of 1:1,000. Lane 1, cytoplasmic fraction; 2, membrane fraction; 3, recombinant Lmb (3 ng).

FIG. 6

Analysis of the surface exposure of Lmb by immunofluorescence staining. S. agalactiae cells were labeled with a polyclonal anti-Lmb antibody and viewed by fluorescence microscopy.

FIG. 7

Adherence of S. agalactiae wild-type (O90R) and lmb mutant (Lmb-k1 and Lmb-k2) strains to immobilized laminin. Adherence was tested in Terasaki wells coated with 100 μg of ECM protein per ml. Assays were carried out as described in Materials and Methods. Bars represent the mean ± standard deviation for six wells. Results are representative for at least three independent experiments performed for each strain. (A) Adherence to laminin with increasing incubation time; (B) adherence to laminin for different bacterial inocula; (C) adherence to laminin after preincubation of immobilized laminin with recombinant Lmb protein; (D) adherence to laminin upon Mn²⁺ substitution.

FIG. 8

SDS-PAGE analysis of recombinant Lmb. Bacterial lysates and purified recombinant Lmb were separated on an 8 to 25% SDS-polyacrylamide gel along with molecular mass markers and silver stained. Lanes 1, molecular mass marker; 2, crude bacterial lysate of E. coli BL21(DE3)(pET21a::lmb), uninduced; 3, crude bacterial lysate of E. coli BL21(DE3)(pET21a::lmb) after induction of protein expression with IPTG; 4, recombinant Lmb after purification over a Ni²⁺-NTA affinity matrix.