Involvement of lipoprotein NlpI in the virulence of adherent invasive Escherichia coli strain LF82 isolated from a patient with Crohn's disease

Abstract

Escherichia coli strain LF82 recovered from a chronic lesion of a patient with Crohn's disease (CD) is able to adhere to and invade cultured intestinal epithelial cells and to replicate within macrophages. One mutant selected for its impaired ability to invade epithelial cells had an insertion of a Tn phoA transposon within the nlpI gene encoding the lipoprotein NlpI. A NlpI-negative isogenic mutant showed a 35-fold decrease in its ability to adhere to and a 45-fold decrease in its ability to invade Intestine-407 cells, but its ability to survive and to replicate within macrophages was similar to that of wild-type strain LF82. In addition, this mutant did not express flagella and synthesized very small amounts of type 1 pili. Downregulation of type 1 pili in the NlpI-negative mutant resulted from a preferential switch toward the OFF position of the invertible DNA element located upstream of the fim operon. The FimB and FimE recombinases act in concert to control the switch, and a large decrease in fimB and fimE mRNA levels was observed. The absence of flagellar structures correlated with a drastic 19-fold decrease in the fliC mRNA level, regardless of the FlhD(2)C(2) transcriptional regulator and of the sigma(28) transcription factor. The key role of NlpI in virulence is independent of type 1 pili and motility, since induced type 1 pilus expression and/or forced contact between bacteria and intestinal epithelial cells did not restore the ability of the NlpI mutant to adhere to and to invade intestinal epithelial cells.

FIG. 1.

Allelic replacement of the nlpI gene by the kanamycin resistance cassette in the LF82-ΔnlpI isogenic mutant. (A) Schematic representation of the locations of different primers on the DNA of strain LF82 and the LF82-ΔnlpI isogenic mutant. (B) PCR amplification product analysis. Amplification products were generated by using specific primers for kanamycin resistance cassette sequence (A2- GBL-3 and B2-GBLnp-5) (lanes 1), for the intragenic region of the nlpI gene (nlpI-3 and nlpI-4) (lanes 2), for location of the kanamycin resistance cassette (nlpI-5 and B2-GBLnp-5) (lanes 3) and (nlpI-6 and A2-GBL-3) (lanes 4), and to amplify the extragenic region of nlpI gene (nlpI-5 and nlpI-6) (lanes 5). DNA to be amplified was released by boiling from LF82-ΔnlpI bacteria (lanes a) and LF82 bacteria (lanes b).

FIG. 2.

Transcomplementation of the adhesion (A) and invasion (B) defects of the LF82-ΔnlpI isogenic mutant with plasmid pPBI03 harboring the entire LF82 nlpI gene. Cell-associated bacteria were quantified after a 3-h infection. Invasion was determined after gentamicin treatment for an additional 1 h. The results are expressed as cell-associated (adherent plus intracellular) or intracellular bacteria relative to those obtained for wild-type strain LF82, taken as 100%. Each value is the mean of at least five separate experiments. Error bars indicate standard errors of the means for five separate experiments. *, P < 0.001.

FIG. 3.

Phenotype of the LF82-ΔnlpI isogenic mutant within J774-A1 macrophages cells. (A) Uptake of LF82, the LF82-ΔnlpI isogenic mutant, the LF82-ΔnlpI isogenic mutant transformed with the cloned nlpI gene (pPBI03), and the LF82-ΔnlpI isogenic mutant harboring the vector alone (pHSG575). Results were expressed in CFU per well after 1 h of gentamicin treatment. (B) Bacterial survival and replication after 24 h of gentamicin treatment. Results are expressed as the number of intracellular bacteria at 24 h relative to that obtained at 1 h after gentamicin exposure, taken as 100%. Data are means ± standard errors of the means for five separate experiments. *, P < 0.05.

FIG. 4.

Type 1 pilus and flagellum expression and/or regulation. Experiments were performed with strain LF82, the LF82-ΔnlpI isogenic mutant, the LF82-ΔnlpI isogenic mutant transcomplemented with the cloned nlpI gene (pPBI03), and the LF82-ΔnlpI isogenic mutant transformed with the vector alone (pHSG575). (A) Colony immunoblotting using polyclonal antibodies raised against purified type I pili. (B) Determination of the invertible element orientation of the fim operon. The orientation was determined by PCR analysis, as described in Materials and Methods. A 450-bp product revealed the ON-orientation and a 750-bp product revealed the OFF-orientation of the invertible element. (C) Motility on a 0.3% agar plate after 16 h at 37°C. Motility was visualized as a halo of radial diffusion of bacteria around the primary inoculum. (D) Western immunoblot analysis of FliC synthesis, using polyclonal antibodies raised against purified flagellin H1.

FIG. 5.

Adhesion (A) and invasion (B) abilities of the isogenic mutant LF82-ΔnlpI with induced type 1 pilus expression and forced contact between bacteria and Intestine-407 epithelial cells. Induced type 1 pilus expression was determined by transformation with pPBI01 harboring the entire fim operon. Experiments were performed without (black) or with (grey) centifugation. See the legend to Fig. 2. Data are means ± standard errors of the means for five separate experiments. *, P < 0.05.