Chlamydial GroEL autoregulates its own expression through direct interactions with the HrcA repressor protein

Abstract

In the pathogenic bacterium Chlamydia trachomatis, a transcriptional repressor, HrcA, regulates the major heat shock operons, dnaK and groE. Cellular stress causes a transient increase in transcription of these heat shock operons through relief of HrcA-mediated repression, but the pathway leading to derepression is unclear. Elevated temperature alone is not sufficient, and it is hypothesized that additional chlamydial factors play a role. We used DNA affinity chromatography to purify proteins that interact with HrcA in vivo and identified a higher-order complex consisting of HrcA, GroEL, and GroES. This endogenous HrcA complex migrated differently than recombinant HrcA, but the complex could be disrupted, releasing native HrcA that resembled recombinant HrcA. In in vitro assays, GroEL increased the ability of HrcA to bind to the CIRCE operator and to repress transcription. Other chlamydial heat shock proteins, including the two additional GroEL paralogs present in all chlamydial species, did not modulate HrcA activity.

FIG. 1.

Western blot analysis of DNA affinity-purified proteins. Equal amounts of lysates purified by CIRCE DNA beads (lane 1) and non-CIRCE DNA beads (lane 2) were resolved by SDS-PAGE. A diluted, crude 28-h-postinfection C. trachomatis serovar L2 lysate was loaded as an antibody control (lane 3). The SDS-PAGE gel was transferred to nitrocellulose and probed with the indicated antibody. Following detection, the same blot was stripped and reprobed with additional antibodies as indicated. Antibody information is provided in Materials and Methods.

FIG. 2.

Native HrcA EMSA. ³²P-labeled DNA probe containing the CIRCE element was incubated with several protein preparations, and the resulting complexes were resolved by native polyacrylamide gel electrophoresis. Lane 1, no protein; lane 2, recombinant HrcA; lane 3, CIRCE DNA bead-purified lysate; lane 4, CIRCE DNA bead-purified lysate with α-HrcA antibody; lane 5, α-HrcA antibody alone.

FIG. 3.

EMSA protein complex analysis. A. Western blot analysis of proteins extracted from EMSA gel. Vertical lanes were cut from an EMSA gel and sectioned horizontally into 10 samples. Proteins were extracted from each gel slice sample and subjected to SDS-PAGE followed by Western blotting. Each vertical lane in the blot represents a section of the original EMSA gel lane. The blot was probed with a series of antibodies as indicated for each image. B. Quantification of ³²P-labeled DNA in gel samples. Gel slices generated in parallel with extracted samples described above were analyzed by liquid scintillation counting. Normalized values for ³²P radioactivity signal of samples corresponding to lanes in the Western blot are represented in a bar graph.

FIG. 4.

Effect of GroEL on recombinant HrcA DNA binding activity. A. ³²P-labeled DNA probe containing the CIRCE element was incubated with several protein preparations, and the resulting complexes were resolved by native polyacrylamide gel electrophoresis. Lane 1, no protein; lane 2, 40 nM GroEL; lane 3, 75 nM HrcA; lane 4, 75 nM HrcA and 40 nM GroEL. B. Graph showing the effects of several concentrations of GroEL on HrcA DNA binding activity. HrcA concentrations are indicated on the x axis. GroEL concentrations are as follows: no GroEL (⧫), 40 nM GroEL (▪), 75 nM GroEL (▴), 120 nM GroEL (×). Data were combined from three independent reactions, and error bars are standard deviations from the means.

FIG. 5.

Effect of chlamydial heat shock proteins on recombinant HrcA. ³²P-labeled DNA probe containing the CIRCE element was incubated with several protein preparations, and the resulting complexes were resolved by native polyacrylamide gel electrophoresis. Lane 1, no protein; lane 2, 70 nM HrcA; lane 3, 70 nM HrcA and 50 nM GroEL; lane 4, 70 nM HrcA and 50 nM GroEL, 50 nM GroES, and 100 μM ATP; lane 5, 70 nM HrcA and 100 nM GroEL2; lane 6, 70 nM HrcA and 100 nM GroEL3; lane 7, 70 nM HrcA and 100 nM DnaK.

FIG. 6.

DNA binding activity of native and denatured HrcA. ³²P-labeled DNA probe containing the CIRCE element was incubated with several protein preparations and the resulting complexes were resolved by native polyacrylamide gel electrophoresis. Lane 1, no protein; lane 2, recombinant HrcA; lane 3, CIRCE DNA bead-purified lysate; lane 4, CIRCE DNA bead-purified lysate denatured in 6 M urea followed by dialysis against storage buffer; lane 5, CIRCE DNA bead-purified lysate denatured in 6 M urea followed by dialysis against storage buffer with the addition of 50 nM GroEL and 50 nM GroES.

FIG. 7.

Effect of GroEL on recombinant HrcA-mediated transcriptional repression. The graph represents the results of in vitro transcription of CIRCE-regulated dnaK promoter in the presence of HrcA and GroEL. Transcription was performed using partially purified chlamydial RNA polymerase under DNA substrate-limiting conditions in the presence of ³²P-labeled nucleotides. RNA products are resolved by urea-PAGE. Data from three independent reactions were combined for each time point, and error bars are standard deviations from the means. Recombinant proteins added to reactions are as follows: no recombinant protein (⧫), 400 nM GroEL alone (□), 800 nM HrcA alone (○), and 800 nM HrcA and 400 nM GroEL (×).