Alterations in redox status of key proteins regulate and drive chlamydial differentiation. (A) Key characteristics of chlamydial developmental forms. (B) Hypothetical model for triggering secondary differentiation through oxidative stress (black angled line). Increasing oxidation of critical protein(s) may lead to earlier differentiation whereas maintaining a reducing environment may delay differentiation. (C) Schematic representation of experimental model for triggering secondary differentiation through altered activity of AhpC. ahpC knockdown may lead to earlier differentiation while overexpression of ahpC may delay differentiation.

Overexpression of ahpC affects chlamydial growth and differentiation. (A) Transcriptional analysis of ahpC in ahpC overexpression (ahpC) and empty vector (EV) control using RT-qPCR following induction at 10 hpi with 1 nM aTc. RNA and genomic DNA (gDNA) were harvested at 14 and 24 hpi and processed as mentioned in materials and methods. Data are presented as a ratio of cDNA to gDNA plotted on a log scale. ***p< 0.0001 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates. (B) Immunofluorescence assay (IFA) of ahpC and EV at 14 and 24 hpi. Construct expression was induced or not at 10 hpi with 1 nM aTc, and samples were fixed with methanol at 24 hpi then stained for major outer membrane protein (MOMP - red) and DAPI (blue) to label DNA. Scale bars = 2 µm. Images were captured using a Zeiss Axio Imager Z.2 with Apotome2 at 100x magnification. Representative images of three biological replicates are shown. (C) Impact of ahpC overexpression on inclusion area. Inclusion area of ahpC overexpression and EV strains was measured using ImageJ. Experimental conditions were the same as mentioned in section (B). The area of 50 inclusions was measured per condition for each sample. ***p< 0.001 vs uninduced sample by using ordinary one-way ANOVA. Data were collected from three biological replicates. (D) IFU assay of ahpC overexpression and empty vector control. Expression of the construct was induced or not at 10 hpi, and samples were harvested at 24 or 48 hpi for reinfection and enumeration. IFUs were calculated as the percentage of uninduced samples. **p< 0.001 vs uninduced sample by using multiple paired t test. Data represent three biological replicates. (E) Quantification of genomic DNA (gDNA) determined by qPCR in ahpC overexpression and empty vector control. Construct expression was induced or not at 10 hpi with 1 nM aTc, gDNA was harvested at 14 and 24 hpi, and ng gDNA were plotted on a log scale. ***p< 0.0001 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates. (F) Ratio of log10 IFUs and log10 gDNA. IFU/ml from (E) was normalized with gDNA from (D). **p< 0.001 vs uninduced sample by using multiple unpaired t test. Data represent three biological replicates.

Higher expression of ahpC provides resistance to peroxides in Chlamydia. IFA of ahpC (A) or EV (C) exposed to oxidizing agents. Construct expression was induced or not at 10 hpi with 1 nM aTc, and samples were treated with three different concentrations of hydrogen peroxide (H2O2) at 16 hpi for 30 min then fixed with methanol at 24 hpi and stained and imaged as described in the legend of Fig. 2B. Representative images from three biological replicates are shown. IFU analysis of ahpC (B) or EV (D) following treatment with oxidizing agents, CHP-Cumene hydroperoxide, H2O2-Hydrogen peroxide, TBHP-Tert-butyl hydroperoxide, and PN-Peroxynitrite. Samples were processed as described for (A) and (C), and IFUs harvested at 24 hpi. IFUs of treated samples were compared with respective untreated controls. ***p< 0.0001 vs untreated sample by using two-way ANOVA. Data represent three biological replicates.

Reduced levels of AhpC negatively impact chlamydial growth. (A) Transcriptional analysis of ahpC in knockdown (ahpC KD) and non-target (NT) control using RT-qPCR following induction at 10 hpi with 1 nM aTc. RNA and gDNA were harvested at 14 and 24 hpi. Quantified cDNA was normalized to gDNA, and values were plotted on a log scale. ***p< 0.0001 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates. (B) IFA was performed to assess inclusion size and morphology using the same induction conditions as in section (A). At 24 hpi, cells were fixed with methanol and stained using primary antibodies to major outer membrane protein (MOMP), Cpf1 (dCas12), and DAPI. All images were acquired on Zeiss Axio Imager Z.2 with Apotome2 at 100x magnification. Bars, 2 µm. Representative images of three biological replicates are shown. (C) Quantification of genomic DNA (gDNA) determined by qPCR in ahpC KD and NT strains. dCas12 expression was induced or not at 10 hpi, and gDNA was harvested at 14 and 24 hpi and plotted on a log scale. ***p< 0.0001 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates. (D) IFU titers following induction at 10 hpi with 1 nM aTc. IFUs were counted from 24 and 48 hpi samples and calculated as percentage of uninduced samples. ***p< 0.0001, **p< 0.001 vs uninduced sample by using multiple paired t test. Data represent three biological replicates. (E) Ratio of log10 IFUs by log10 gDNA. IFU/ml from (D) was normalized with gDNA from (C). **p< 0.001 vs uninduced sample by using multiple unpaired t test. Data represent three biological replicates.

Intracellular ROS levels were measured to investigate the function of AhpC in reducing ROS. (A) HeLa cells were infected or not with ahpC knockdown, and construct was induced or not at 10 hpi with 1 nM aTc. At 24 or 48 hpi, samples were washed with DPBS and incubated with CellROX Deep red dye for 30 min in dark. ROS levels were measured at wavelengths of 640 nm (excitation) and 665 nm (emission). **p< 0.001, *p< 0.01 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates. (B) Microscopy images were acquired using live cells on Zeiss Axio Imager Z.2 with Apotome2 at 100x magnification. Scale Bars= 10 µm. Representative images of three biological replicates are shown.

Chlamydia is hypersensitive to oxidizing agents in ahpC knockdown condition. IFA of ahpC KD (A) or NT (C) treated with 62.5 µM H2O2. dCas12 expression was induced or not at 10 hpi with 1 nM aTc, treated or not with H2O2 at 16 hpi for 30 min, and allowed to grow until 24 hpi. Coverslips were fixed with methanol at 24 hpi and stained major outer membrane protein (MOMP), Cpf1 (dCas12), and DAPI. Scale bars = 2 µm. Images were captured using a Zeiss Axio Imager Z.2 with Apotome2 at 100x magnification. Representative images from three biological replicates are shown. IFU analysis of ahpC KD (B) or NT (D) following treatment with oxidizing agents, CHP-Cumene hydroperoxide, H2O2-Hydrogen peroxide, TBHP-Tert-butyl hydroperoxide, or PN-Peroxynitrite. dCas12 expression was induced or not, and samples were treated or not as mentioned in the legend of Fig. 3B. IFUs of treated samples were calculated as percentage of respective untreated samples. ***p< 0.0001 vs untreated sample by using two-way ANOVA. Data represent three biological replicates.

Complementation of the phenotypes observed in the ahpC knockdown. (A) Confirmation of complementation (comp) of ahpC knockdown by RT-qPCR. Samples were processed and quantified as mentioned previously in the legend of Fig. 4A. Values were plotted on a log scale. *p< 0.01 vs uninduced sample using ordinary one-way ANOVA. Data represent three biological replicates. (B) IFA of comp strain was performed at 24 hpi, and staining and imaging were performed as mentioned in the legend of Fig. 4B. Scale bars, 2 µm. Representative images from three biological replicates are shown. (C) Genomic DNA quantitation was performed by qPCR. Construct expression was induced or not at 10 hpi, and gDNA was harvested at 14 and 24 hpi and plotted on a log scale. Statistical analysis was calculated using ordinary one-way ANOVA. Data represent three biological replicates. (D) IFU analysis of comp strain. Statistical analysis was calculated using multiple paired t test. Data represent three biological replicates. (E) IFA of comp strain following treatment with 62.5 µM H2O2. Samples were treated, stained, and images acquired as mentioned in the legend of Fig. 6A. Scale bars = 2 µm. Representative images from three biological replicates are shown. (F) IFU analysis of comp strain following treatment with oxidizing agents. Experiments were performed as mentioned in the legend of Fig. 6B. IFUs were calculated as percentage of respective untreated samples. ***p< 0.0001 vs untreated sample by using two-way ANOVA. Data represent three biological replicates. (G) ahpC knockdown growth defect rescued by ROS scavengers. IFU analysis of ahpC knockdown treated with or without scavengers, α-Tocopherol (100 µM) and DMTU (10 mM), as mentioned in materials and methods. IFUs were calculated as a percentage of the untreated, uninduced sample. ***p< 0.0001 vs untreated, uninduced or induced sample by using two-way ANOVA. Data represent three biological replicates. (H) IFA of ahpC knockdown treated with or without scavengers. Experimental conditions were similar as in section (G). Staining and imaging were performed as mentioned in Fig. 6A. Representative images from three biological replicates are shown.

Effects of ahpC knockdown/overexpression on chlamydial developmental cycle progression. (A) RT-qPCR analysis of late cycle genes (hctA, hctB, omcB, tsp, and glgA) in ahpC knockdown. Experimental conditions were the same as mentioned in the legend of Fig. 4A. Quantified cDNA was normalized to gDNA, and values were plotted on a log scale. ***p< 0.0001, **p< 0.001, *p< 0.01 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates. (B) Volcano plot of RNA-sequencing of ahpC knockdown. Experimental conditions were the same as mentioned in the legend of Fig. 4A. The volcano plot was prepared using GraphPad Prism software. The vertical dashed lines indicate a fold change of 1.7 as compared to the respective uninduced control. The horizontal dashed line indicates an p value of 0.05. The black dots represent genes not significantly different and grey dots represent significantly altered but fold change lesser than 1.5. Blue and red spots represent statistically significant altered genes with more than 1.5-fold change in transcription levels between uninduced and induced samples. Red dots indicate ahpC and canonical late genes also shown in Table 1. (C) One-step growth curve of ahpC knockdown. Samples were induced or not with 1 nM aTc at 10 hpi and harvested at 16, 18, 20, 22, and 24 hpi. IFUs recovered are displayed as log10 values. ***p< 0.0001 vs uninduced sample by using two-way ANOVA. Data represent three biological replicates.

List of canonical late genes significantly increased during ahpC knockdown at 14 hpi.

Altering the activity of AhpC in Ctr L2 impacts its developmental cycle progression. (Top) In Chlamydia, secondary differentiation is asynchronous and RBs divide through an asymmetric budding mechanism. In such conditions, either the mother or daughter cell may inherit more oxidized proteins (represented by a darker shade), which can then impact whether a given RB will divide again or undergo secondary differentiation. (Bottom) The black dots represent EBs, the orange circles show RBs. The developmental cycle of wild-type C. trachomatis (Ctr L2) is shown. In ahpC KD, highly oxidized conditions lead some RBs to cross the oxidative threshold sooner, allowing activation of late genes and secondary differentiation earlier than other RBs. In ahpC overexpression, a reducing environment results in a delay in achieving the oxidative threshold, thus allowing RBs to continue to divide before committing to secondary differentiation.