PLoS Pathogens (Feb 2011)

BosR (BB0647) controls the RpoN-RpoS regulatory pathway and virulence expression in Borrelia burgdorferi by a novel DNA-binding mechanism.

  • Zhiming Ouyang,
  • Ranjit K Deka,
  • Michael V Norgard

DOI
https://doi.org/10.1371/journal.ppat.1001272
Journal volume & issue
Vol. 7, no. 2
p. e1001272

Abstract

Read online

In Borrelia burgdorferi (Bb), the Lyme disease spirochete, the alternative σ factor σ⁵⁴ (RpoN) directly activates transcription of another alternative σ factor, σ(S) (RpoS) which, in turn, controls the expression of virulence-associated membrane lipoproteins. As is customary in σ⁵⁴-dependent gene control, a putative NtrC-like enhancer-binding protein, Rrp2, is required to activate the RpoN-RpoS pathway. However, recently it was found that rpoS transcription in Bb also requires another regulator, BosR, which was previously designated as a Fur or PerR homolog. Given this unexpected requirement for a second activator to promote σ⁵⁴-dependent gene transcription, and the fact that regulatory mechanisms among similar species of pathogenic bacteria can be strain-specific, we sought to confirm the regulatory role of BosR in a second virulent strain (strain 297) of Bb. Indeed, BosR displayed the same influence over lipoprotein expression and mammalian infectivity for strain Bb 297 that were previously noted for Bb strain B31. We subsequently found that recombinant BosR (rBosR) bound to the rpoS gene at three distinct sites, and that binding occurred despite the absence of consensus Fur or Per boxes. This led to the identification of a novel direct repeat sequence (TAAATTAAAT) critical for rBosR binding in vitro. Mutations in the repeat sequence markedly inhibited or abolished rBosR binding. Taken together, our studies provide new mechanistic insights into how BosR likely acts directly on rpoS as a positive transcriptional activator. Additional novelty is engendered by the facts that, although BosR is a Fur or PerR homolog and it contains zinc (like Fur and PerR), it has other unique features that clearly set it apart from these other regulators. Our findings also have broader implications regarding a previously unappreciated layer of control that can be involved in σ⁵⁴-dependent gene regulation in bacteria.