Histone methylation by NUE, a novel nuclear effector of the intracellular pathogen Chlamydia trachomatis

Abstract

Sequence analysis of the genome of the strict intracellular pathogen Chlamydia trachomatis revealed the presence of a SET domain containing protein, proteins that primarily function as histone methyltransferases. In these studies, we demonstrated secretion of this protein via a type III secretion mechanism. During infection, the protein is translocated to the host cell nucleus and associates with chromatin. We therefore named the protein nuclear effector (NUE). Expression of NUE in mammalian cells by transfection reconstituted nuclear targeting and chromatin association. In vitro methylation assays confirmed NUE is a histone methyltransferase that targets histones H2B, H3 and H4 and itself (automethylation). Mutants deficient in automethylation demonstrated diminished activity towards histones suggesting automethylation functions to enhance enzymatic activity. Thus, NUE is secreted by Chlamydia, translocates to the host cell nucleus and has enzymatic activity towards eukaryotic substrates. This work is the first description of a bacterial effector that directly targets mammalian histones.

Figure 1. Sequence analysis of chlamydial SET domain protein.

(A) The SET domain of chlamydial protein NUE (CT737) sequence was compared to eukaryotic SET domain proteins G9a, MLL4, Set8 and Suv39H1. Residues important for SAM binding, catalysis and the structurally important pseudoknot are noted. (B) Homologous SET domain proteins were compared for each of the 6 Chlamydia genomes sequenced, the SET domain is underlined. “.” indicates a semi-conserved residue, “:” a conserved residue and “*” identical residues in all sequences.

Figure 2. NUE is secreted from bacteria and detected during late infection.

(A) Chlamydia trachomatis NUE and its C. pneumoniae and C. caviae homologs were fused to the Cya reporter protein and expressed in Shigella flexneri ipaD (constitutive TTS) or mixD (defective TSS) strains. Bacteria were pelleted and both pellet and supernatant were loaded on SDS-PAGE gels, transferred to a PVDF membrane and probed with anti- Cya (to detect chlamydial fusion proteins), anti-IpaD (Shigella secreted protein) or anti-CRP (Shigella non-secreted protein) antibodies. (B) HeLa cells were infected with C. trachomatis for the times indicated, lysed and analyzed by Western blot using antibodies for the chlamydial proteins NUE and Hsp60. Actin was used to ensure equal loading of samples. All results shown are representative of at least 2 separate experiments.

Figure 3. NUE localizes to the host cell nucleus.

(A) HeLa or 3T3 cells were transfected with FLAG-NUE, fixed with paraformaldehyde, permeabilized and analyzed by immunofluorescence using anti-FLAG antibody and Hoescht dye to stain nuclei. (B) HeLa cells were transfected with NUE or the C-terminal truncation mutants NUEΔ12 and NUEΔ122 and treated as above for immunofluorescence analysis. (C) The number of transfected cells demonstrating predominant nuclear localization was divided by the number of total transfected cells and multiplied by 100 to determine % nuclear localization. Each experiment was performed 3 times with >500 transfection events counted for each protein. Error bars reflect the difference in calculated percentage among the 3 separate experiments. (D) Western blot analysis of transfected cells. HeLa cells were transfected with FLAG-tagged NUE, NUEΔ12 or NUEΔ122 for 24 h and cells were then fractionated into cytosolic and nuclear fractions. Protein concentration was measured by Bradford and for each fraction 20 µg of protein was loaded on a gel. The remaining pellet after protein extraction from nuclei was also analyzed as the nuclear insoluble fraction. NUE distribution in the different fractions was analyzed by Western blot using anti-FLAG antibody. Anti-GDIβ (cytosolic), anti-PARP1 (nuclear) and anti-H3 (nuclear insoluble) were used to demonstrate purity of the fractions.

Figure 4. NUE is found in the nucleus of infected cells and associates with chromatin.

(A) HeLa cells were infected with C. trachomatis for 20, 25, 30 or 40 h. Nuclei were isolated as described in the Method section and both cytosolic and nuclear fractions were analyzed by Western blot using anti-NUE antibody. The membrane was stripped and re-probed for GDIβ as a cytosolic marker, PARP1 as a nuclear marker and the chlamydial protein EF-Tu to ensure the absence of bacteria in the nuclear fraction. Blots are representative of 2 separate experiments. (B) A schematic of the extraction protocol used to detect chromatin-associated proteins. (C) HeLa cells were transfected with GFP-NUE or GFP only as a negative control. Nuclei were separated from the cytosolic fraction of transfected HeLa cells. The nuclei were resuspended in 3 mM EDTA to extract soluble nuclear proteins versus chromatin-associated proteins. Samples were analyzed by Western blot using anti-GFP. HeLa cells transfected with Suv39H1 were used as a positive control and anti-Suv39H1 was used to detect the protein. Histone H3 was used to ensure chromatin enrichment in our chromatin fraction. Blots are representative of 3 separate experiments. (D) HeLa cells were infected with C. trachomatis for 40 h and analyzed for chromatin-association of NUE as in (C). Histone H3 was used to demonstrate purity of our chromatin-associated fractions and EF-Tu was used to control for bacterial contamination. Blots are representative of 2 separate experiments.

Figure 5. NUE methylates mammalian histones in vitro.

(A–C) Recombinant GST-NUE was purified from E. coli and incubated with either core histones (A), individual recombinant histones (B) or recombinant His-tagged chlamydial protein Hc1 (C) in the presence of ¹⁴C-SAM for 1 h at 30°C. Samples were then separated on a 15% SDS-PAGE gel and stained with Coomassie blue prior to gel dehydration and 24 h exposure to capture radioactive events. PRMT1, a histone H4 methyltransferase, was used as a positive control in panel A. Various concentrations of histone H2B were used as a positive control in panel C.

Figure 6. NUE automethylation regulates its activity towards histones.

(A) NUE automethylation is SAM-dependent. Recombinant GST-NUE was incubated in the presence of ¹⁴C-SAM and increasing amounts of non-radioactive SAM for 1 h at 30°C. Samples were then separated by SDS-PAGE electrophoresis and stained with Coomassie blue prior to gel dehydration and 24 h exposure to capture radioactive events. (B) NUE automethylates outside the GST moiety. GST-NUE was incubated with chlamydial protein GST-CT529 (1∶4 or 1∶2 molar ratio) or GST alone (1∶5, 1∶3 or 1∶1 molar ratio). Samples were then treated as in panel A. (C) NUE mutants have diminished activity towards histones. NUE, NUEΔ12, NUEGG (KK→GG) or NUEAA (KK→AA) were incubated in the presence of core histones for 1 h at 30°C. Samples were then treated as in panel A. All gels are representative of at least 2 independent experiments.