Intracellular growth of Chlamydia trachomatis leads to global histone hypermethylation by impairing demethylation
- Institut Pasteur, Université Paris Cité, UMR3691 CNRS, Cellular biology of microbial infection, France
- Sorbonne Université, Collège Doctoral, France
- Department of Integrative Immunobiology, Duke University School of Medicine, United States
- Université Paris Cité, CNRS, Epigenetics and Cell Fate, UMR7216, France
- Institut de Recerca Biomèdica (IRB Barcelona), Spain
- Barcelona Institute of Science and Technology (BIST), Spain
- Institut Pasteur, Université Paris Cité, CNRS UAR 2024, Proteomics Platform, Mass Spectrometry for Biology, France
- Institut Pasteur, Université Paris Cité, Bioinformatics and Biostatistics HUB, France
- Institut Pasteur, Université Paris Cité, UMR 3523 CNRS, Epigenetic Chemical Biology, France
- Institut Pasteur, Université Paris Cité, UMR3691 CNRS, BioImage Analysis, France
- Institució Catalana de Recerca i Estudis Avançats (ICREA), Spain
- Department of Pharmacology, UT Southwestern Medical Center, United States
Figures
Figure 1 with 4 supplements
C. trachomatis infection induces protein hypermethylation.
(A) Cells were infected at MOI = 1 with mCherry expressing C. trachomatis serovar L2 (blue) for 48 hr before fixation, permeabilization, and staining with the indicated antibodies. DAPI staining (grey) of DNA allows for the detection of host nuclei and bacteria. Bar, 50 μm. Scatter plots display the correlation between mean intensities for the indicated methylated residues in the nuclei of infected cells. a.u., arbitrary unit. R=Pearson’s coefficient of correlation. (B) Mean intensities for the indicated methylated residues were quantified in the nuclei of non-infected and infected cells of the same coverslip. Histograms display the percentage of positive nuclei for the indicated histone methylation mark, defined as nuclei with fluorescence intensity greater than twofold the mean intensity of nuclei of non-infected samples. The p-value of a Welch’s t-test is shown ****p<0.0001, **p<0.01, *p<0.05. (C) Mouse-derived organoids infected for 40 hr with GFP-expressing C. trachomatis serovar L2 (top) or for 30 h with GFP-expressing C. muridarum (bottom) were fixed and stained for H3K9me3 (purple). Arrow heads point to nuclei with increased H3K9me3 intensity. Quantification of the intensity of the signal in nuclei of infected cells and of neighboring non-infected cells is shown. The p-value of a Welch’s t-test is shown. Bar, 20 μm. (D) Volcano plot of the relative enrichment of methylated host proteins in cells infected with C. trachomatis. The site of methylation (on R or K) is indicated for the proteins that are significantly hypomethylated (green quadrant) or hypermethylated (orange quadrant) in infected cells. Proteins or methylated peptides that were recovered in only one condition are absent from this representation (Supplementary file 1). (E) Analysis of the biological pathways associated with proteins differentially methylated late in infection using g:Profiler (Kolberg et al., 2023).
Figure 1—figure supplement 1
C. trachomatis infection induces histone hypermethylation.
HeLa cells were infected or not with mCherry-expressing C. trachomatis L2 for 48 hr before fixation, permeabilization, and staining with antibodies specific for the indicated histone methylation marks, followed by Alexa-488 conjugated secondary antibodies (green). DNA was labeled with DAPI (white), mCherry is displayed in blue. Bar, 50 μm.
Figure 1—figure supplement 2
Histone hypermethylation occurs in HeLa cells and MEFs infected with C. muridarum.
HeLa cells were infected with C. muridarum for 48 hr before fixation, permeabilization, and staining for H3K4me3 and H4K20me3. Mouse embryonic fibroblasts (MEFs) were infected with C. trachomatis L2 (top) or C. muridarum (bottom) for 48 hr. DNA was stained with DAPI; inclusions are highlighted with an asterisk.
Figure 1—figure supplement 3
Histone hypermethylation in infected primary cells.
Primary human epithelial cells isolated from the ectocervix were infected for 48 hr with C. trachomatis serovar L2 before proceeding as in B. Right panel shows quantification of relative distribution of histone marks.
Figure 1—figure supplement 4
Quantification of DNA methylation.
Nuclei were isolated from cells infected for 48 hr with C. trachomatis L2 or from control cells prior to measurement of DNA methylation by mass spectrometry. Results of four independent experiments are shown, with mean ± SD, and the p-value of the Student’s paired t-test is indicated (**p<0.01).
Figure 2 with 1 supplement
Histone hypermethylation correlates with bacterial proliferation.
(A) HeLa cells were infected with mCherry-expressing C. trachomatis serovar L2 (MOI = 1) for the indicated time before fixation, permeabilization, and staining for H3K9me3 and H4K20me3. The mCherry signal is displayed in blue, DNA in white. Bar, 50 μm. Dot plots display the mean intensities for the indicated methylated residues in one representative experiment. a.u., arbitrary unit. Histograms display the percentage of positive nuclei for the indicated histone methylation mark, defined as nuclei with fluorescence intensity greater than twofold the mean intensity of nuclei of non-infected cells. Data from three independent experiments and the p-values of Welch’s t-tests are shown ****p<0.0001, **p<0.01, *p<0.05. (B) HeLa cells were infected with C. trachomatis L2 expressing mCherry for 48 h before fixation, permeabilization, and staining for H3K9me3. The mean fluorescence intensity of mCherry of each individual inclusion is shown for infected cells with a low level of H3K9me3 signal (signal < twofold mean H3K9me3 signal in nuclei of non-infected cells) and with a high level of H3K9me3 signal (signal > twofold mean H3K9me3 signal in nuclei of non-infected cells). The p-value of a Welch’s t-test is shown. (C). Cells were infected with mCherry-expressing C. trachomatis serovar L2 and 15 μg/ml ciprofloxacin or 10 μg/ml chloramphenicol was added to the culture medium 24 hpi. Cells were fixed 48 hpi, permeabilized and stained for H3K4me3 and H3K27me3. The dot plots display the mean mCherry intensities in inclusions of control and antibiotic-treated cultures (left) and the relative mean intensities of H3K4me3 (top) and H3K27me3 (bottom) in the nuclei from infected cells and non-infected cells in one representative experiment. Histograms display the percentage of positive nuclei, each dot represents an independent experiment. The p-values of Welch’s t-tests are shown.
Figure 2—figure supplement 1
Histone hypermethylation is independent of Nue.
Cells were infected for 48 hr with C. trachomatis L2 WT or with a mutant that does not express Nue (Nue*=CTL2M340rs59, see methods), fixed, permeabilized, and stained with the indicated antibodies. The dot plots display the intensity for the indicated histone methylation marks in the nuclei of infected cells. Absence of Nue expression in cell lysates 24 hpi was verified by western blot. Antibodies against the chlamydial protein Hsp60 were used to control for infection levels.
-
Figure 2—figure supplement 1—source data 1
pdf file containing original western blots for Figure 2—figure supplement 1.
- https://cdn.elifesciences.org/articles/110111/elife-110111-fig2-figsupp1-data1-v3.pdf
-
Figure 2—figure supplement 1—source data 2
Original files of Figure 2—figure supplement 1.
- https://cdn.elifesciences.org/articles/110111/elife-110111-fig2-figsupp1-data2-v3.zip
Figure 3
S-adenosyl methionine (SAM) levels display cell-to-cell heterogeneity and do not increase in infected cells.
(A) Cells were infected with C. trachomatis serovar L2 for 48 hr before isolation of the nuclei. SAM concentrations were normalized to protein content. Each dot represents an independent experiment; the mean ± SD is shown. The p-value of a ratio paired t-test is indicated. (B) Cells transfected with the SAM sensor 5–1 and infected or not for 40 hr with mCherry-expressing C. trachomatis serovar L2 (MOI = 1) were imaged live after addition of BI dihydrochloride and DFHO to the culture medium. SAM binding activates the orange fluorescence by promoting Squash folding and subsequent binding with DFHO. The constitutive green fluorescence of Broccoli–BI is used for normalization. Bar, 30 μm. (C) The orange to green fluorescence ratio was measured in the cytoplasm (left) and in the nucleus (right) of infected and non-infected cells, with or without cycloleucine. The experiment is representative of three; black bars indicate the mean. The p-value of Welch’s t-test is shown. (D) Dot plot of the orange to green fluorescence ratio against the mCherry signal, used as a read-out of bacterial load, in the cytoplasm (left) or nucleus (right) of infected cells. R=Pearson’s coefficient of correlation, showing an absence of correlation between SAM concentration and bacterial load in both compartments.
Figure 4 with 3 supplements
The activities of histone demethylases are low in late C. trachomatis infected cells.
(A) Cells were infected with mCherry-expressing C. trachomatis serovar L2. After 24 hr, the culture medium was refreshed with or without the indicated concentration of JIB-04. Cells were fixed 48 hpi, permeabilized, and stained for H3K9me3 (green). DNA was labeled with DAPI (white), mCherry is displayed in blue. Bar, 50 μm. The dot plot displays H3K9me3 mean fluorescence intensity in nuclei of non-infected or infected cells. a.u., arbitrary unit. Black bars indicate the mean. The p-value of Welch’s t-test are shown. The histogram displays the percentage of positive nuclei in five independent experiments. The p-value of Welch’s t-tests are shown ****p<0.0001; ***p<0.001; **p<0.01; *p<0.05; n.s., not significant. Nuclei were scored positive when H3K9me3 intensity was greater than two times its mean intensity in nuclei of non-infected cells from the same culture condition. (B) Cells were transfected with the indicated HA-tagged constructs (dKDM: catalytically dead mutant) and infected 3 h later with mCherry-expressing C. trachomatis L2. Cells were fixed 48 hr later, permeabilized, and stained for HA and for the indicated methylated residues. DNA was labeled with DAPI (white), mCherry is displayed in blue. Bar, 50 μm. The dot plots display mean fluorescence intensities of the indicated marks in one representative experiment, each dot representing one infected cell. The histograms display the percentages of positive nuclei in the transfected infected cells in four independent experiments. Cells were considered transfected when the HA signal in the nucleus was greater than the twofold value of the mean intensity in nuclei in non-transfected cells of duplicate wells. Nuclei were scored positive as in B. (C) Demethylase activity against H3K4me3 was measured in nuclear extracts from cells infected for 40 h, normalized to protein concentration and expressed relative to the activity measured in nuclear extracts from non-infected cells. The p-value of a Welch’s t-test is indicated.
Figure 4—figure supplement 1
Infected cells are sensitized to JIB-04.
Same as Figure 4 panel B, showing H4K20me3 staining.
Figure 4—figure supplement 2
JIB-04 treatment impairs bacterial development.
HeLa cells were infected with C. trachomatis L2 expressing GFP at MOI = 0.2 and treated with the indicated concentration of JIB-04 24 hpi. Infectious forming units (IFUs) collected 48 hpi were quantified by reinfecting fresh HeLa cells as described in the methods. One-way ANOVA with Dunnett’s multiple comparison test was used to determine significance.
Figure 4—figure supplement 3
Infected cells are sensitized to KDM4 inhibitors.
The KDM4 inhibitors SD70 (2.5 μM), TACH101 (10 nM), or vehicle (DMSO 1:1000) were added in the culture medium 24 hpi. Samples were processed 48 hpi and analyzed as Figure 4A. The mCherry intensity in the inclusions is used as an indicator of bacterial load; the effect of JIB-04 (2 μM) is shown for comparison.
Figure 5 with 2 supplements
Supply of demethylase cofactors reduces histone hypermethylation.
(A) Succinate concentrations were measured in nuclear extracts from cells infected for 40 hr or not infected and normalized to protein concentrations. The p-value of a ratio paired t-test is indicated. (B) Cells were infected with mCherry-expressing C. trachomatis serovar L2 (MOI = 2). After 24 hr, the culture medium was left untouched (control) or refreshed with the addition of 5 mM DMKG, solvent alone (DMSO), or 100 μM ferric ammonium citrate (iron). Cells were fixed 48 hpi, permeabilized, and stained for H3K4me3 (green) and H4K20me3 (red). DNA was stained with DAPI (displayed in white). Bar, 50 μm. (C) The dot plots show the intensities for the indicated methylated residues in nuclei of infected and non-infected cells for one representative experiment. The second condition (-) corresponds to refreshment of the medium without addition of DMSO. The p-value of Welch’s t-tests are shown. The histograms display the percentages of positive nuclei in five independent experiments. The p-value of Welch’s t-tests are shown.
Figure 5—figure supplement 1
Effect of medium supplementation on progeny.
The culture medium was replaced with fresh medium complemented with 5 mM DMKG (left), 100 μM ferric ammonium citrate, or a combination of the two (right). The result of three to five independent experiments and the p-values of ratio-paired t-tests are shown.
Figure 5—figure supplement 2
Competition between iron and JIB-04.
Where indicated, 2 μM JIB-04 or vehicle (DMSO 1:2500) were added to the culture medium 24 hpi, immediately followed or not with the addition of 100 μM ferric ammonium citrate. Samples were processed 48 hpi and analyzed as in Figure 4A. The dot plots display H3K9me3 mean fluorescence intensity in nuclei of infected cells and mCherry intensity in the inclusions, respectively, for one representative experiment. a.u., arbitrary unit. Black bars indicate the mean. The histogram displays the percentage of positive nuclei in two independent experiments with technical duplicates. The p-values of Welch’s t-tests are shown.
Figure 6 with 3 supplements
H3K4me3 and H3K9me3 chromatin landscape.
(A) Distribution of genomic sites enriched in H3K4me3 and H3K9me3 marks. The functional annotation was based on the chromatin states established from ChromHMM (ChromHMM track from ENCODE/Broad Institute). The relative abundance in HeLa cells of each chromatin state is shown on the right. (B) Profile of the H3K4me3 ChIP-seq signal (read density) from merged biological replicates, i.e., pileups of the H3K4me3 enrichment relative to chromatin input (subtraction mode, input from IP of BAM files) in non-infected (NI) and infected HeLa cells on genomic loci defined as 3 kb upstream and downstream of annotated Transcriptional Start Site (TSS) of genes (NCBI RefSeq). (C) Boxplots showing read coverage (log2(CPM + 1); CPM, counts per million; bin = 500 bp) from H3K4me3, H3K9me3 ChIP-seq experiments and input in non-infected (NI), and infected HeLa cells. The median and the p-value (Wilcoxon test) is indicated within each boxplot. Outliers are not shown for clarity.
Figure 6—figure supplement 1
Quality controls for the chromatin immunoprecipitation followed by DNA sequencing (ChIP) experiments.
Electrophoresis profile of input DNA isolated from infected and uninfected cells and corresponding ChIP with antibodies to the indicated histone marks. The DNA concentrations measured in each lane are reported in the table.
-
Figure 6—figure supplement 1—source data 1
pdf file containing original gels for Figure 6—figure supplement 1.
- https://cdn.elifesciences.org/articles/110111/elife-110111-fig6-figsupp1-data1-v3.pdf
-
Figure 6—figure supplement 1—source data 2
Original files of Figure 6—figure supplement 1.
- https://cdn.elifesciences.org/articles/110111/elife-110111-fig6-figsupp1-data2-v3.zip
Figure 6—figure supplement 2
Number of reads and peaks analyzed for each experiment.
Figure 6—figure supplement 3
Hierarchical clustering of correlation matrix of H3K4me3 and H3K9me3 ChIP-seq replicates .
Hierarchical clustering of correlation matrix of H3K4me3 and H3K9me3 ChIP-seq replicates (1 and 2) performed in non-infected (Nl) and infected (Inf) HeLa cells. Spearman correlations were calculated in deepTools the multiBamSummary (Galaxy Version 3.5.4+galaxy0) was followed by plotCorrelation (Galaxy Version 3.5.4+galaxy0) using the read counts split into 500 bp bins across the human genome.
Figure 7
Dimethyl-ketoglutarate (DMKG) supplementation modifies the host transcriptional response to infection.
(A) Transcripts from infected (INF) and non-infected (NI) samples, supplemented or not with 5 mM DMKG 24 hpi, and extracted 48 hpi were analyzed by microarray. Two-by-two comparisons of differentially expressed genes are shown (cut-off: twofold difference). (B) Venn diagram showing about one third (364/1055) of the differentially regulated genes upon infection were no longer differentially regulated with DMKG supplementation. (C) For each of the 10 genes tested, the histogram on the left displays the microarray result (p-values were extracted from Supplementary file 5) and the histogram on the right shows the result of the reverse transcription quantitative PCR (RT-qPCR) validation. For RT-qPCR validation, samples were prepared in the same conditions as the microarray samples. Specific sets of primers for the indicated genes were used to measure transcript levels by RT-qPCR normalized to actin transcripts following the ΔΔCt method. Each experiment was performed at least four times (dots). The data are presented as relative mRNA levels compared to uninfected cells and shown as the mean± SD. Statistical comparisons were obtained using contrast analysis after linear modeling, 2-by-2 multiple testing and p-value adjustment (See Methods for details). ****p<0.0001; ***p<0.001; **p<0.01. (D) KEGG pathways associated to the 55 infection responsive genes differentially regulated by DMKG supplementation (Supplementary file 5, analysis with g:Profiler). (E) Cells were infected with C. trachomatis L2 for 48 hpi. Levels of tRNAGly-GCC and tRNAArg-UCG measured by northern blot are expressed relative to 5 S RNA and U6 snRNA. See the Method section for statistical analysis of the data.
-
Figure 7—source data 1
pdf file containing original northern blots for Figure 7E.
- https://cdn.elifesciences.org/articles/110111/elife-110111-fig7-data1-v3.pdf
-
Figure 7—source data 2
Original files of Figure 7E.
- https://cdn.elifesciences.org/articles/110111/elife-110111-fig7-data2-v3.zip
Figure 8
JIB-04 administration shows limited benefit in vivo.
(A) Schematic view of animal infections. JIB-04 treatment was administered intraperitoneally (i.p.) and infections with C. muridarum were intravaginal. (B) Bacterial load was measured on DNA extracted from vaginal lavages at D4 and normalized to host DNA. Each dot represents one mouse; the p-value of Welch’s t-test are shown. (C) The genital tracts were excised at D35 and the presence and severity of hydrosalpinx was scored. (D) Example of the segmentation and detection of the skeletons of uterine horns. The radius along the length of the uterine horn was computed starting from the point of convergence of the right and left skeletons (yellow and red lines). Each color trace corresponds to one animal; measures of the left and right horns are displayed in left and right panels, respectively.
Additional files
-
Supplementary file 1
Relative abundance of host methylated proteins 48 hpi with C. trachomatis.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp1-v3.xlsx
-
Supplementary file 2
Relative abundance of host proteins 48 hpi with C. trachomatis.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp2-v3.xlsx
-
Supplementary file 3
ChIP-seq analyses.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp3-v3.xlsx
-
Supplementary file 4
Differentially expressed host genes 48 hpi with C. trachomatis.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp4-v3.xlsx
-
Supplementary file 5
DMKG sensitive genes in C. trachomatis infection.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp5-v3.xlsx
-
Supplementary file 6
Antibodies used.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp6-v3.docx
-
Supplementary file 7
Oligonucleotides used.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp7-v3.xlsx
-
Supplementary file 8
CTL2M340rs59 variants, identified by genotyping.
- https://cdn.elifesciences.org/articles/110111/elife-110111-supp8-v3.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/110111/elife-110111-mdarchecklist1-v3.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Intracellular growth of Chlamydia trachomatis leads to global histone hypermethylation by impairing demethylation
eLife 15:e110111.
https://doi.org/10.7554/eLife.110111
