In the E. coli linear divisome assembly pathway (Du and Lutkenhaus, 2017), FtsK is the first protein downstream of FtsZ encoded by Ct (Figure 1A). In other organisms, FtsK is uniformly distributed at the septum of dividing cells (Yu et al., 1998; Wang et al., 2006; Veiga and Pinho, 2017). To investigate the localization of FtsK during cell division in Ct, HeLa cells were infected with Ct. To overcome the challenges associated with assessing cell morphologies in densely packed inclusions in infected cells, we analyzed FtsK localization in Ct derived from lysates of infected HeLa cells at 21 hr post-infection (hpi) as described previously (Ouellette et al., 2022). Ct were stained with an antibody against the chlamydial major outer membrane protein (MOMP) and an antibody that recognizes endogenous FtsK. Blotting analysis revealed that this FtsK antibody recognizes a single protein with the predicted molecular mass of FtsK (Figure 1—figure supplement 1A). Our results showed that, unlike FtsK in other organisms, chlamydial FtsK accumulates in discrete foci in the membrane of coccoid cells (Figure 1B). In cell division intermediates, FtsK localized in foci at the septum, foci at the septum and at the base of the progenitor mother cell, or foci at the base of the progenitor mother cell only (Figure 1C). The chlamydial FtsK foci observed during cell division were not uniformly distributed at the septum, rather septal foci of FtsK were restricted to one side of the MOMP-stained septum. In addition, the FtsK foci were often above or below (marked with arrowheads in Figure 1C) the MOMP-stained septum. Similar analyses were performed using Ct transformed with the pBOMB4-Tet (-GFP) plasmid encoding FtsK with a C-terminal mCherry tag. The expression of this mCherry fusion is under the control of an anhydrotetracycline (aTc)-inducible promoter. HeLa cells were infected with the transformant, and the expression of the fusion was induced by the addition of 10 nM aTc to the media of infected cells at 19 hpi. The induced cells were harvested at 21 hpi and a lysate was prepared from Ct present in the infected cells. Blotting analysis of the lysate with mCherry antibodies revealed that a single protein with the predicted molecular mass of FtsK-mCherry was present in the lysate (Figure 1—figure supplement 1B). Imaging analyses of Ct in the lysate revealed that like endogenous FtsK, FtsK-mCherry accumulated in foci in coccoid cells (Figure 1D), and in division intermediates, it localized in foci at the septum, foci at the septum and at the base of the progenitor mother cell, or foci at the base of the progenitor mother cell only (Figure 1E). The foci of FtsK-mCherry, like endogenous FtsK, were often offset relative to the plane defined by MOMP staining at the septum (arrowhead in Figure 1E). Inclusion forming unit (IFU) assays demonstrated that overexpression of the FtsK-mCherry fusion had no effect on chlamydial developmental cycle progression and the production of infectious EBs (Figure 1—figure supplement 1C). While it is possible that the population of FtsK at the base of the mother cell is a remnant of FtsK from a previous division, ~20% of dividing cells have a secondary bud (Figure 1F), and FtsK and FtsK-mCherry accumulate in foci at the base of secondary buds (arrowheads in Figure 1G), suggesting that the population of FtsK at the base of the mother cell corresponds to a nascent divisome complex that forms at the site where the daughter cell will arise in the next round of division.

To investigate the distribution of other putative chlamydial divisome components during budding, we transformed Ct with plasmids encoding PBP2, PBP3, or MreC with an N-terminal mCherry tag. The PBP2, PBP3, and MreC fusions were induced by the addition of 10 nM aTc to infected cells at 19 hpi, and the induced cells were harvested at 21 hpi and stained with MOMP antibodies. Imaging analyses revealed that the PBP2, PBP3, and MreC fusions accumulated in foci in coccoid cells (Figure 2A), and in cell division intermediates, the fusions accumulated in foci at the septum, in foci at the septum and at the base of the progenitor mother cell, or in foci only at the base of the progenitor mother cell (Figure 2B). Similar analyses with an MreB_6xHis fusion (Lee et al., 2020) revealed that MreB exhibited a similar localization profile (Figure 2A and B). Each of these fusions, like FtsK, were restricted to one side and were often slightly above or slightly below (marked with arrowheads in Figure 2B) the MOMP-stained septum in dividing cells. Blotting analyses revealed that mCherry antibodies primarily detected single species with the predicted molecular mass of the PBP2, PBP3, and MreC fusions in lysates prepared from induced cells (Figure 2—figure supplement 1A), and IFU assays demonstrated that the aTc induced overexpression of the PBP2, PBP3, and MreC fusions had no effect on the developmental cycle progression of Ct (Figure 2—figure supplement 1B). Like FtsK, the foci of the PBP2, PBP3, MreC and MreB fusions were also detected at the base of secondary buds (Figure 2—figure supplement 1C). Quantification of the localization profiles of endogenous FtsK and the various fusion proteins revealed that the distribution profile of FtsK-mCherry accurately reflected the distribution of endogenous FtsK (Figure 2C). Furthermore, a greater percentage of FtsK was associated with the base of dividing cells (including cells with septum and base, and cells with base alone) suggesting that FtsK associates with nascent divisomes at the base of dividing cells prior to the other putative divisome proteins assayed. Finally, this analysis suggested that MreB associated with nascent divisomes at the base of dividing cells after mCherry-PBP2 and mCherry-PBP3 (marked with # in Figure 2C). The localization profiles of the chlamydial divisome proteins (Figures 1 and 2) likely reflect the assembly of divisome complexes at the septum and at the base of the progenitor mother cell, and the disassembly of the septal divisome when divisome proteins are only present at the base of the mother cell.

Since it was possible that the localization profiles of the mCherry-PBP2 and mCherry-PBP3 fusions were at least in part due to their induced over-expression, we performed similar studies with rabbit antibodies generated against peptides derived from PBP2 or PBP3 (Ouellette et al., 2012). Blotting analyses (Figure 2—figure supplement 2A) with these antibodies revealed that they recognized mCherry-PBP2 and mCherry-PBP3 in Ct lysates, and immunofluorescent staining with the PBP2 and PBP3-specific antibodies (Figure 2—figure supplement 2B) completely overlapped the mCherry fluorescence in cells when the mCherry PBP2 and PBP3 fusions were inducibly expressed in Ct. Imaging analyses with the antibodies that recognize endogenous PBP2 and endogenous PBP3 indicated that these antisera detected foci in coccoid cells, and in cell division intermediates, the PBP2 and PBP3 antibodies detected foci at the septum, foci at the septum and at the base of the mother cell, or foci at the base alone (Figure 2—figure supplement 3A). Quantification revealed that the localization profiles of endogenous PBP2 and PBP3 in division intermediates (Figure 2—figure supplement 3B) were not statistically different than the localization profiles observed for the mCherry fusions of PBP2 and PBP3 (Figure 2C).

The quantification in Figure 2C suggested that FtsK is recruited to nascent divisomes that form at the base of dividing cells prior to the other divisome components assayed. This hypothesis was tested by staining cells expressing the PBP2, PBP3, MreC, or MreB fusions with antibodies that recognize endogenous FtsK. Imaging analyses revealed that in a subset of cells, FtsK was detected in foci at the septum and at the base of dividing cells, while each of the fusions was only detected at the septum where they overlapped the distribution of septal FtsK (Figure 2D), indicating that FtsK is recruited to nascent divisomes at the base of the cell prior to the other divisome components assayed. Additional analyses revealed that FtsK also overlapped the distribution of the PBP2, PBP3, MreC, and MreB fusions following their appearance at the base of dividing cells (Figure 2—figure supplement 4A). Quantification revealed the percentage of cells in which FtsK overlapped the distribution of each of the fusions at the septum, at the septum and the base, and at the base alone in division intermediates (Figure 2—figure supplement 4B).

MreB was one of the last components that associated with nascent divisomes forming at the base of the progenitor mother cell (Figure 2C). To investigate whether MreB filament formation was required for the formation of foci by the other chlamydial divisome components, HeLa cells were infected with Ct transformed with the FtsK, PBP2, PBP3, MreC, or MreB fusions, and the fusions were induced by adding 10 nM aTc to the media of the infected cells at 20 hpi for 1 hr. During the induction period, cells were incubated in the absence (Figure 3B and D) or presence (Figure 3C and E) of the MreB inhibitor, A22, which inhibits MreB filament formation (Bean et al., 2009). RBs were harvested at 21 hpi and stained with the appropriate antibodies to assess the effect of A22 on the cellular distribution of the fusions. As previously shown (Ouellette et al., 2012; Cox et al., 2020), A22 inhibits chlamydial budding and most cells in the population were coccoid following A22 treatment (Figure 3A). Furthermore, approximately 50% of the untreated control cells were coccoid, which is consistent with prior estimates of the number of non-dividing RBs at this stage of the developmental cycle (Lee et al., 2018), indicating that our lysis procedure does not lead to a bias in the number of non-dividing coccoid cells in the population. MreB in coccoid cells adopted a diffuse pattern of localization following A22 treatment (Figure 3). A22 also had a statistically significant effect on the percent of coccoid cells containing MreC foci, but it did not affect the ability of FtsK, PBP2, or PBP3 to form foci in coccoid cells (Figure 3D and E). These data indicate that MreB filaments do not function as a scaffold that is necessary for the assembly of all divisome components in Ct.

To further investigate the mechanisms that regulate divisome assembly in Ct, we inducibly repressed the expression of ftsK or pbp2 using CRISPRi technology, which has been used to inducibly repress the expression of genes in Ct (Ouellette et al., 2021). CRISPRi employs a constitutively expressed crRNA that targets an inducible dCas enzyme (dCas12) to specific genes where it binds but fails to cut, thus inhibiting transcription. We transformed Ct with the pBOMBL12CRia plasmid that constitutively expresses an ftsK or pbp2-specific crRNA, which target sequences in the ftsK or pbp2 promoter regions. To determine whether ftsK and pbp2 transcript levels were altered using this CRISPRi approach, dCas12 expression was induced by the addition of 5 nM aTc to the media of infected cells at 8 hpi. Control cells were not induced. Nucleic acids were isolated from induced cells and uninduced control cells at various times, and RT-qPCR was used to measure ftsK or pbp2 transcript levels. This analysis revealed that the induction of dCas12 resulted in ~10 fold reduction in ftsK transcript levels by 15 hpi in cells expressing the ftsK-targeting crRNA (Figure 4A), and ~ eightfold reduction in pbp2 transcript levels in cells expressing the pbp2-targeting crRNA (Figure 4B), while these crRNAs had minimal or no effect on chlamydial euo and omcB transcript levels, suggesting that the ftsK and pbp2 crRNAs specifically inhibit the transcription of ftsK and pbp2 (Figure 4A and B). To investigate the effect of ftsK or pbp2 down-regulation on developmental cycle progression, dCas12 was induced by the addition of aTc to the media of infected cells at 4 hpi. Control cells were not induced. The cells were then fixed at 24 hpi and stained with MOMP and Cas12 antibodies. Imaging analysis revealed that Ct morphology was normal and dCas12 was undetectable in the inclusions of uninduced control cells, while foci of dCas12 were observed in the induced cells, and Ct in the inclusion exhibited an enlarged aberrant morphology (Figure 4C and D), suggesting that the inducible knockdown of ftsK or pbp2 blocks chlamydial cell division. In additional studies, we induced dCas12 at 17 hpi in cells expressing the ftsK or pbp2-targeting crRNAs. Lysates were prepared and the cells were fixed at 21 hpi, and localization studies revealed that foci of endogenous FtsK and PBP2 were almost undetectable when ftsK or pbp2 were transiently knocked down using this CRISPRi approach (Figure 4E).

To assess whether the knockdown of ftsK or pbp2 arrests Ct division at a specific stage of polarized budding, HeLa cells were infected with Ct transformed with the pBOMBL12CRia plasmids encoding the ftsK or pbp2-targeting crRNAs. At 17 hpi, dCas12 was induced by the addition of 20 nM aTc to the media. Control cells were not induced. RBs were harvested from induced and uninduced control cells at 22 hpi and stained with MOMP antibodies, and imaging analyses quantified the division intermediates present in the population. These analyses revealed that >60% of the Ct in the uninduced controls were at various stages of polarized budding (Figure 5A and B), while ~90% of the cells were coccoid when ftsK was knocked down (Figure 5A), and ~85% of the cells were coccoid following pbp2 knockdown (Figure 5B), suggesting that the initiation of polarized budding of Ct is inhibited when ftsK or pbp2 are knocked down.

We next examined whether the knockdown of ftsK affected foci formation by PBP2 or PBP3 in coccoid cells. HeLa cells were infected with Ct transformed with the pBOMBL12CRia plasmid encoding the ftsK-targeting crRNA. At 17 hpi, dCas12 was induced by the addition of 10 nM aTc to the media. Control cells were not induced. RBs were harvested from induced and uninduced control cells at 22 hpi, and the localization of endogenous PBP2 and PBP3 in coccoid cells was assessed (Figure 5C and D). This analysis revealed that the number of coccoid cells with polarized foci of PBP2 and PBP3 were reduced by approximately 80% following ftsK knockdown (Figure 5E). In similar analyses, we assessed the effect of pbp2 knockdown on the ability of FtsK and PBP3 to form foci in coccoid cells. While FtsK retained its ability to form foci in coccoid cells following pbp2 knockdown, foci of PBP3 were almost entirely absent in pbp2 knockdown cells (Figure 5C and D). Quantification of these assays revealed that FtsK is necessary for foci formation by both the PBP2 and PBP3 transpeptidases, while PBP2 is necessary for foci formation by PBP3 (Figure 5E and F). Our data place FtsK upstream of, and necessary for, the addition of PBP2 and PBP3 to the Ct divisome, and PBP2 upstream of and necessary for the addition of PBP3 to the Ct divisome. These results are consistent with inhibitor studies that indicated PBP2 acts upstream of PBP3 in the polarized budding process of Ct (Cox et al., 2020).

To investigate whether the catalytic activity of PBP2 is necessary to maintain its association with the Ct divisome, HeLa cells were infected with Ct, and mecillinam, an inhibitor of the transpeptidase activity of PBP2 (Kocaoglu et al., 2015; Cox et al., 2020), was added to the media of infected cells at 20 hpi. Cells incubated in the absence of mecillinam were included as a control. Mecillinam-treated and control cells were harvested at 22 hpi and the effect of inhibiting the catalytic activity of PBP2 on the localization of FtsK, PBP2, and PBP3 was determined. As shown previously, mecillinam blocks chlamydial division (Ouellette et al., 2012; Cox et al., 2020), and most cells in the population assumed a coccoid morphology (Figure 6A). We then determined the localization of endogenous FtsK, PBP2, and PBP3 in drug-treated and control coccoid cells. Mecillinam treatment resulted in a ~50% reduction in the number of cells with polarized foci of PBP2 (Figure 6B-E). There was a similar reduction in polarized foci of PBP3 following mecillinam treatment (Figure 6B-E). These data indicate that the catalytic activity of PBP2 is necessary for PBP2 to efficiently associate with or maintain its association with polarized divisome complexes. Furthermore, consistent with pbp2 knockdown studies, PBP3 association with the divisome complex is dependent on the prior addition of PBP2 to the complex, but foci formation by FtsK is unaffected when PBP2 foci are reduced in number (Figure 6).

To assess the morphology of PG at the septum and the base of dividing cells, we used an EDA-DA labeling strategy (Liechti et al., 2014; Cox et al., 2020). This approach enabled the detection of PG foci, bars, and rings in dividing Ct (Liechti, 2021). Our imaging analysis revealed that in some instances PG foci were detected at both the septum and at the base of the progenitor mother cell (Figure 7A), and in other instances, PG organization at the septum and the base differed. In the example shown (Figure 7B), a PG foci was detected at the septum while a PG ring was detected at the base of the mother cell. In additional analyses, we compared the localization of mCherry-PBP3 to the localization of PG in cells where the expression of this mCherry fusion had been induced by the addition of aTc to the media. This analysis, which was restricted to PG formation at the septum of dividing cells, revealed that multiple foci of PBP3 were associated with a septal PG ring (Figure 7C). Furthermore, the PG ring was at an angle relative to the MOMP-stained septum. Similar analyses revealed that two foci of endogenous FtsK were associated with PG that was again at an angle relative to the MOMP-stained septum. This was true even though the PG had not fully reorganized into a ring structure (Figure 7C).

We then determined the effect of A22 and mecillinam on PG synthesis/organization. Since both of these drugs induce Ct to assume a coccoid morphology, we initially characterized PG organization in untreated coccoid cells. We detected foci, bars, or rings in ~80% of untreated coccoid cells (Figure 7D and E), which make up ~50% of the cells in the inclusion at this stage of the developmental cycle (Lee et al., 2018). Furthermore, each of these PG intermediates exhibited a polarized distribution in untreated coccoid cells (Figure 7D). Although we cannot rule out that continued PG synthesis and reorganization occurs in polarized division intermediates, PG rings can arise prior to any of the morphological changes that occur during the polarized division of Ct.

Prior studies have shown that inhibitors of MreB filament formation prevent the appearance of PG-containing structures in Ct (Liechti et al., 2014; Ouellette et al., 2022). To assess whether MreB filament formation is required for PG synthesis and/or PG reorganization, we infected HeLa cells with Ct, and EDA-DA and A22 were added to the media of infected cells at 18 hpi. The cells were harvested at 22 hpi, lysates were prepared, and PG localization was determined. These analyses revealed that PG was diffuse/undetectable in the majority of A22-treated cells, and, in those cells where PG was still detected, it could not convert into ring structures (Figure 7E and F). In similar experiments, we assessed the effect of mecillinam on the appearance of PG intermediates in coccoid cells. These analyses revealed that PG formed discrete foci or bars in 60% of mecillinam-treated cells (Figure 7E and F). However, these PG intermediates could not convert into PG rings when the transpeptidase activity of PBP2 was inhibited. Finally, we assessed PG organization in cells where ftsK was knocked down by inducing dCas12 in the ftsK knockdown strain by the addition of aTc to the media of infected cells at 17 hpi. Cells were fixed at 21 hpi, and localization studies revealed that ftsK knockdown had the most dramatic effect on PG localization, which was diffuse/undetectable in ~90% of the cells assayed. Inhibiting divisome assembly by knocking down ftsK almost entirely prevented the accumulation of all PG-containing intermediates in Ct (Figure 7E and F).

HeLa cells (clone S3, authenticated by ATCC, Manassas, VA) were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Invitrogen, Waltham, MA) containing 10% fetal bovine serum (FBS, Hyclone, Logan, UT) at 37 °C in a humidified chamber with 5% CO2. HeLa were only passaged five times before fresh aliquots of cells were thawed for use. The Hela cell stock was routinely tested for mycobacterial contamination by PCR. HeLa cells grown in DMEM containing 10% FBS were infected with Ct serovar L2 434/Bu. Infections of HeLa cells with chlamydial transformants were performed in DMEM containing 10% FBS and 0.36 U/mL penicillin G (Sigma-Aldrich).

The plasmids and primers used for generating mCherry fusions of FtsK, PBP2, PBP3, and MreC are listed in Supp. File 1 A. The chlamydial ftsK, pbp2, pbp3, and mreC genes were amplified by PCR with Phusion DNA polymerase (NEB, Ipswich, MA) using 10 ng C. trachomatis serovar L2 genomic DNA as a template. The PCR products were purified using a PCR purification kit (Qiagen) and inserted into the pBOMB4-Tet (-GFP) plasmid, which confers resistance to β-lactam antibiotics. The plasmid was cut at the NotI (FtsK-mCherry) or the KpnI (mCherry-PBP2, mCherry-PBP3, mCherry-MreC) site, and the chlamydial genes were inserted into the cut plasmid using the HiFi DNA Assembly kit (NEB) according to the manufacturer’s instructions. The products of the HiFi reaction were transformed into NEB-5αI^q competent cells (NEB) and transformants were selected by growth on plates containing ampicillin. DNA from individual colonies was isolated using a mini-prep DNA isolation kit (Qiagen), and plasmids were initially characterized by restriction digestion to verify the inserts were the correct size. Clones containing inserts of the correct size were DNA sequenced prior to use.

Total nucleic acids were extracted from HeLa cells infected with Ct plated in 6-well dishes as described previously (Ouellette et al., 2015; Ouellette et al., 2021). For RNA isolation, cells were rinsed one time with PBS, then lysed with 1 mL Trizol (Invitrogen) per well. Total RNA was extracted from the aqueous layer after mixing with 200 μL per sample of chloroform following the manufacturer’s instructions. Total RNA was precipitated with isopropanol and treated with DNase (Ambion) according to the manufacturer’s guidelines prior to cDNA synthesis using SuperScript III (Invitrogen). For DNA, infected cells were rinsed one time with PBS, trypsinized and pelleted before resuspending each pellet in 500 μL of PBS. Each sample was split in half, and genomic DNA was isolated from each duplicate sample using the DNeasy extraction kit (Qiagen) according to the manufacturer’s guidelines. Quantitative PCR was used to measure C. trachomatis genomic DNA (gDNA) levels using an euo primer set. 150 ng of each sample was used in 25 μL reactions using standard amplification cycles on a QuantStudio3 thermal cycler (Applied Biosystems) followed by a melting curve analysis. ftsK, pbp2, euo, and omcB transcript levels were determined by RT-qPCR using SYBR Green as described previously (Ouellette et al., 2021) (see Supp. File 1B for primers used for measuring gDNA levels and RT-qPCR). Transcript levels were normalized to genomes and expressed as ng cDNA/gDNA.

Ct was transformed as described previously (Wang et al., 2011). Briefly, HeLa cells were plated in a 10 cm plate at a density of 5×10⁶ cells the day before beginning the transformation procedure. Ct lacking its endogenous plasmid (-pL2) was incubated with 10 μg of plasmid DNA in Tris-CaCl₂ buffer (10 mM Tris-Cl pH 7.5, 50 mM CaCl₂) for 30 min at room temperature. HeLa cells were trypsinized, washed with 8 mL of 1 x DPBS (Gibco), and pelleted. The pellet was resuspended in 300 μL of the Tris-CaCl₂ buffer. Ct was mixed with the HeLa cells and incubated at room temperature for an additional 20 min. The mixture was added to 10 mL of DMEM containing 10% FBS and 10 μg/mL gentamicin and transferred to a 10 cm plate. At 48 hpi, the HeLa cells were harvested and Ct in the population were used to infect a new HeLa cell monolayer in media containing 0.36 U/ml of penicillin G to select for transformants. The plate was incubated at 37 °C for 48 hr. These harvest and re-infection steps were repeated every 48 hr until inclusions were observed.

HeLa cells were seeded in 10 cm plates at a density of 5×10⁶ cells per well the day before the infection. Ct L2 or chlamydial strains transformed with plasmids encoding FtsK-mCherry, mCherry-PBP2, mCherry-PBP3, or mCherry-MreC or with plasmids that direct the constitutive expression of the crRNAs targeting the pbp2 or ftsK promoters were used to infect HeLa cells from ATCC?? in DMEM. For experiments with the transformants, aTc was added to the media of infected cells at the indicated concentration and time. At 21 hpi, cells were detached from the 10 cm plate by scraping and pelleted by centrifugation for 30 s. The pellet was resuspended in 1 mL of 0.1 x PBS (Gibco) and transferred to a 2 mL tube containing 0.5 mm glass beads (Thermo Fisher Scientific). Cells were vortexed for 3 min. then centrifuged at 800 rpm for 2 min in a microfuge. 20 μLs of the supernatant was mixed with 20 μLs of 2 x fixing solution (6.4% formaldehyde and 0.044% glutaraldehyde) and incubated on a glass slide for 10 min at room temperature. Cells were washed with three times with PBS, and the cells were permeabilized by incubation with PBS containing 0.1% Triton X-100 for 1 min. Cells were washed with PBS two times. For experiments with Ct L2, the cells were incubated with a goat primary antibody against the major outer-membrane protein (MOMP; Meridian, Memphis, TN), and the mouse primary antibody that recognizes endogenous FtsK raised against recombinant CT739 protein (https://doi.org/10.1099/mic.0.047746-0), or with rabbit antibodies raised against peptides derived from PBP 2 or PBP3 (Ouellette et al., 2012). Briefly, chlamydial antigens or peptides emulsified with Freund’s incomplete adjuvant were used to immunize animals via intramuscular injections three times with an interval of 2 wk. Antisera were collected from the immunized animals 2–4 wk after the final immunization as the primary antibodies. After the primary antibody labeling, the cells were then rinsed with PBS and incubated with donkey anti-goat IgG (Alexa 488) and donkey anti-mouse IgG (Alexa 594) or donkey-anti-rabbit IgG (Alexa 594) secondary antibodies (Invitrogen). Experiments in which we visualized the distribution of the various mCherry fusions, the localization of the mCherry fluorescence was compared to the distribution of MOMP. In some experiments, we determined the distribution of the MreB_6 x His fusion by staining cells expressing the fusion with a rabbit anti-6x His antibody (Abcam, Cambridge, MA) and the goat anti-MOMP antibody followed by the appropriate secondary antibodies. Cells were imaged using Zeiss AxioImager2 microscope equipped with a 100 x oil immersion PlanApochromat objective and a CCD camera. During image acquisition, 0.3 μm xy-slices were collected that extended above and below the cell. The images were collected such that the brightest spot in the image was saturated. The images were deconvolved using the nearest neighbor algorithm in the Zeiss Axiovision 4.7 software. Deconvolved images were viewed and assembled using Zeiss Zen-Blue software. For each experiment, three independent replicates were performed, and the values shown for localization are the average of the three experiments. In some instances, 3D projections of the acquired xy slices were generated using the Zeiss Zen-Blue software.

PG was labeled by incubating cells with 4 mM ethylene-D-alanine-D-alanine (E-DA-DA) as described (Cox et al., 2020). The incorporated E-DA-DA was fluorescently labeled using the Click & Go labeling kit (Vector Laboratories). The distribution of fluorescently labeled PG was compared to the distribution of MOMP and endogenous FtsK or the distribution of mCherry-PBP3. Three independent replicates were performed, and the values shown are the average of the three experiments.

HeLa cells were infected with Ct (-pL2) transformed with the pBOMB4 Tet (-GFP) plasmid encoding the indicated aTc-inducible gene. At 8 hpi, aTc was added to the culture media at the indicated concentration. Control cells were not induced. At 48 hpi, the HeLa cells were dislodged from the culture dishes by scraping and collected by centrifugation. The pellet was resuspended in 1 mL of 0.1 x PBS (Gibco) and transferred to a 2 mL tube containing 0.5 mm glass bead tubes (Thermo Fisher Scientific). Cells were vortexed for 3 min followed by centrifugation at 800 rpm for 2 min. The supernatants were mixed with an equal volume of a 2 x sucrose-phosphate (2SP) solution (ref) and frozen at –80 °C. At the time of the secondary infection, the chlamydiae were thawed on ice and vortexed. Cell debris was pelleted by centrifugation for 5 min at 1 k x g at 4 °C. The EBs in the resulting supernatant were serially diluted and used to infect a monolayer of HeLa cells in a 24-well plate. The secondary infections were allowed to grow at 37 °C for 24 hr before they were fixed and labeled for immunofluorescence microscopy by incubating with a goat anti-MOMP antibody followed by a secondary donkey anti-goat antibody (Alexa Fluor 594). The cells were rinsed in PBS and inclusions were imaged using an EVOS imaging system (Invitrogen). The number of inclusions were counted in five fields of view and averaged. Three independent replicates were performed, and the values from the replicates were averaged to determine the number of inclusion-forming units. Chi-squared analysis was used to compare IFUs in induced and uninduced samples.

HeLa cells were infected with Ct transformed with the pBOMB4-Tet (-GFP) plasmid encoding FtsK-mCherry, mCherry-PBP2, mCherry-PBP3, mCherry-MreC, or MreB-6xHis. The fusions were induced at 20 hpi with 10 nM aTc for 1 hr in the absence or presence of 75 μM A22. At 22 hpi, cells were harvested and prepared for staining as described above. Three independent replicates were performed, and the values shown for localization are the average of the three experiments.

HeLa cells were infected with Ct L2 and 20 μM mecillinam (Sigma) was added to the media of infected cells at 17 hpi. Control cells were untreated. At 22 hpi, infected cells were harvested and RBs were prepared and stained with MOMP, FtsK, PBP2, or PBP3 antibodies as described above. Alternatively, cells were incubated with 4 mM EDA-DA at 17hpi in the presence or absence of 20 μM mecillinam. The cells were harvested at 22 hpi, and RBs were prepared and PG was click-labeled and its distribution was visualized in MOMP-stained cells as described above. Three independent replicates were performed, and the values shown for localization are the average of the three experiments.

HeLa cells infected with Ct L2 were harvested by scraping the infected cells from the plate at 24 hpi. Uninfected HeLa cells were included as a control. The HeLa cells were pelleted by centrifugation, resuspended in SDS sample buffer, and electrophoresed on a 10% SDS polyacrylamide gel. The gel was electrophoretically transferred to nitrocellulose (Schleicher and Schuell), and the filter was incubated with mouse polyclonal antibodies raised against chlamydial FtsK. The filter was rinsed and incubated with 800 donkey anti-mouse IgG secondary antibodies (LICOR, Lincoln, NE) and imaged using a LICOR Odyssey imaging system.

HeLa cells were infected with Ct transformed with plasmids that inducibly express FtsK-mCherry, mCherry-PBP2, mCherry-PBP3, or mCherry-MreC. The fusions were induced by the addition of 10 nM aTc to the media of infected cells at 17 hpi. The cells were harvested and pelleted at 21 hpi. The cell pellet was resuspended in 1 mL of 0.1 x PBS (Gibco) and transferred to a 2 mL tube containing 0.5 mm glass beads (Thermo Fisher Scientific). Cells were vortexed for 3 min followed by centrifugation at 800 rpm for 2 min. The supernatant was collected and centrifuged for 3 min at 13,000 rpm and the pellet containing Ct was resuspended in TBS containing 1% TX-100, 1 X protease inhibitor cocktail (Sigma), and 5 μM lactacystin. The suspension was sonicated three times on ice and centrifuged at 13,000 rpm for 3 min. The supernatant was collected and mixed with SDS sample buffer. The samples were boiled and electrophoresed on a 10% SDS polyacrylamide gel, and the gel was electrophoretically transferred to nitrocellulose. The blots from these analyses were probed with a rabbit anti-mCherry primary antibody (Invitrogen) and a 800 donkey anti-rabbit IgG secondary antibodies (LICOR, Lincoln, NE). The filters were imaged using a LICOR Odyssey imaging system.

HeLa cells were infected with Ct transformed with the pBOMB4-Tet (-GFP) plasmid encoding mCherry-PBP2 or mCherry-PBP3. The fusions were induced with 10 nM aTc at 17 hpi. Uninduced cells were included as a control. The cells were harvested at 21 hpi, and samples were processed for immunoblotting as described above. The blots were probed with rabbit polyclonal antibodies raised against peptides derived from chlamydial PBP2 or PBP3 (Ouellette et al., 2012). The blots were then rinsed and incubated with 800 donkey anti-rabbit IgG secondary antibodies (LICOR, Lincoln, NE). The filters were imaged using a LICOR Odyssey imaging system.