Key features of σS required for specific recognition by Crl, a transcription factor promoting assembly of RNA polymerase holoenzyme

Abstract

Bacteria use multiple sigma factors to coordinate gene expression in response to environmental perturbations. In Escherichia coli and other γ-proteobacteria, the transcription factor Crl stimulates σ(S)-dependent transcription during times of cellular stress by promoting the association of σ(S) with core RNA polymerase. The molecular basis for specific recognition of σ(S) by Crl, rather than the homologous and more abundant primary sigma factor σ(70), is unknown. Here we use bacterial two-hybrid analysis in vivo and p-benzoyl-phenylalanine cross-linking in vitro to define the features in σ(S) responsible for specific recognition by Crl. We identify residues in σ(S) conserved domain 2 (σ(S)2) that are necessary and sufficient to allow recognition of σ(70) conserved domain 2 by Crl, one near the promoter-melting region and the other at the position where a large nonconserved region interrupts the sequence of σ(70). We then use luminescence resonance energy transfer to demonstrate directly that Crl promotes holoenzyme assembly using these specificity determinants on σ(S). Our results explain how Crl distinguishes between sigma factors that are largely homologous and activates discrete sets of promoters even though it does not bind to promoter DNA.

Keywords: RNAP formation; RpoS; bacterial stress response; curli fiber; transcription initiation.

Fig. 1.

Crl interacts with the DNA-melting domain of σ^S (σ^S₂). BTH analysis of the interaction of σ^S fragments with Crl (black) or the unfused control (light gray) by measurement of β-galactosidase activity. Error bars show SD between two separate experiments of two independent cultures each.

Fig. 2.

Identification of σ^S₂ residues needed for interaction with Crl (BTH analysis). (A) σ^S₂ variants with reduced interaction with Crl (<0.40 relative to WT; black bar, WT and teal bars, variants) that retain an interaction with the β′CH (>0.51 relative to WT; gray bars) and whose identities are different in σ^S and σ⁷⁰. Error bars: SD between ratio of variant/WT β-galactosidase activity in two separate experiments of two independent cultures each. Complete results of BTH screen are shown in Table S1 and Fig. S2A. (B) Amino acid sequence alignment of σ^S₂ and σ⁷⁰₂. Identity (|), similarity (:); residues identified in A are colored teal. Colored bar indicates σ⁷⁰ regions defined by homology [region 1.2 (blue), nonconserved region (green), 2.1 (yellow), 2.2 (orange), 2.3 (red), and 2.4 (purple)]. (C and D) Homology model of σ^S₂ based on σ⁷⁰ structure (1SIG) with colors as in B. (E and F) Surface representation of σ^S₂ from C and D; residues with Crl interaction <0.4 relative to WT are in teal, as in A.

Fig. 3.

Crl activation of osmY promoter is dependent on the DPE patch in vivo and in vitro. (A) osmY promoter–lacZ expression in early stationary phase (OD_A600 ∼2.4). Error bars represent SD between two experiments of two independent cultures each. (B) Multiple round in vitro transcription from plasmid-borne osmY promoter (pRLG8941) by Eσ^S(WT) or Eσ^S(E137Q) in the presence of a range of concentrations of Crl. Representative experiment is shown. (C) Effect of Crl on Eσ^S(WT) (filled triangles) or Eσ^S(E137Q) (open circles) normalized to the same reaction without Crl. The plot shows the average and SD from three separate experiments (six reactions).

Fig. 4.

BPA-mediated cross-linking of σ^S to Crl. (A) SDS/PAGE gel showing results of UV exposure on ³²P–Crl incubated with WT and four different σ^S BPA variants. Cross-linked ³²P–Crl–σ^S complexes are indicated. (B) σ^S₂ homology model showing positions of σ^S BPA variants that cross-link to Crl (orange) and σ^S substitutions that disrupt Crl–σ^S interaction identified in the BTH analysis (cyan).

Fig. 5.

σ⁷⁰ variant with a σ^S DPE motif and an NCR deletion interacts with Crl. (A) Conservation between σ^S₂ and σ⁷⁰₂. Surface representation of σ^S₂ (homology model) and σ⁷⁰₂ crystal structure (1SIG). Identities, light gray; similarities, dark gray; not conserved, purple; position of NCR, green. DPE motif (σ^S D135/P136/E137) and σ⁷⁰ NCR (I123–R374) are indicated. (B) Amino acid sequence of the σ⁷⁰₂–σ^S₂ chimera showing positions of the σ^S₂ substitution for the NCR (green) and flanking residues (yellow) and the σ^S DPE motif (blue). The positions of these substitutions are illustrated on two views on the σ^S₂ homology model. (C) BTH analysis of interaction of σ₂ variants with Crl (red) and β′CH (gray) compared with background level (white). Error bars show SD between two separate experiments of two independent cultures each.

Fig. 6.

Crl promotes assembly of Eσ^S(WT) but not Eσ^S(E137Q) in vitro. (A) Equilibrium binding of σ^S or (B) σ^S(E137Q) variant to core RNAP (10 nM) in the presence or absence of Crl (16 µM). Eσ^S holoenzyme assembly represents the equilibrium LRET signal (acceptor/donor ratio) 1 h after mixing terbium-labeled core RNAP with fluorescein-labeled σ^S. (C) Histogram showing core RNAP interaction with 37 nM σ^S ± Crl. Error bars indicate average of four samples. (D) Crl promotes assembly of Eσ^S but not Eσ⁷⁰ or Eσ^H in vitro. Fold effect of Crl (0, 3, 9, and 27 µM) on RNAP holoenzyme assembly [interaction between fluorescein-labeled σ^S, σ⁷⁰, or σ^H (10 nM) and terbium-labeled core RNAP (8.3 nM)] using an LRET assay.