An NlpC/P60 protein catalyzes a key step in peptidoglycan recycling at the intersection of energy recovery, cell division and immune evasion in the intracellular pathogen Chlamydia trachomatis

The obligate intracellular Chlamydiaceae do not need to resist osmotic challenges and thus lost their cell wall in the course of evolution. Nevertheless, these pathogens maintain a rudimentary peptidoglycan machinery for cell division. They build a transient peptidoglycan ring, which is remodeled during the process of cell division and degraded afterwards. Uncontrolled degradation of peptidoglycan poses risks to the chlamydial cell, as essential building blocks might get lost or trigger host immune response upon release into the host cell. Here, we provide evidence that a primordial enzyme class prevents energy intensive de novo synthesis and uncontrolled release of immunogenic peptidoglycan subunits in Chlamydia trachomatis. Our data indicate that the homolog of a Bacillus NlpC/P60 protein is widely conserved among Chlamydiales. We show that the enzyme is tailored to hydrolyze peptidoglycan-derived peptides, does not interfere with peptidoglycan precursor biosynthesis, and is targeted by cysteine protease inhibitors in vitro and in cell culture. The peptidase plays a key role in the underexplored process of chlamydial peptidoglycan recycling. Our study suggests that chlamydiae orchestrate a closed-loop system of peptidoglycan ring biosynthesis, remodeling, and recycling to support cell division and maintain long-term residence inside the host. Operating at the intersection of energy recovery, cell division and immune evasion, the peptidoglycan recycling NlpC/P60 peptidase could be a promising target for the development of drugs that combine features of classical antibiotics and anti-virulence drugs.


Supplementary
Text. Primary sequence alignment of YkfCBce, YkfCCtr, YkfCCpn and NlpC/P60Ela. The secondary structure of YkfCBce characterized by Xu et al., 2010 [1] is indicated above its sequence with the Sh3b1 domain in blue, the Sh3b2 domain in green and the NlpC/P60 domain in orange. The in silico predicted secondary structures of the chlamydial proteins are shown in grey above the respective sequences. The catalytic triad comprising Cys, His and a polar residue is shown in red, residues contributing to the S1 and S2 sites essential for substrate recognition in YkfCBce are highlighted in yellow and blue, respectively. Residues contributing to both sites are further highlighted in green. The signal peptide sequence of YkfCBce is displayed in grey.  5 h and (f) 6 h was compared by one-way ANOVA analysis. P-values are shown as asterisks: n.s.: P ≥ 0.05, *: P = 0.01 to 0.05, **: P = 0.001 to 0.01, ***: P < 0.001. Error bars indicate ± s.d. (n=3).

Fig C in S1
Text. Purification and in vitro activity of recombinant YkfCCtr. Recombinant YkfCCtr as well as its active site mutant YkfCCtrC172A were analyzed by SDS-PAGE (a). In vitro peptidase activity over time was analyzed by TLC using L-Ala-γ-D-Glu-mDAP tripeptide as substrate (b). YkfCCtr activity was time dependent reaching a plateau at t=30 min and was not decreased in presence of the reaction product mDAP (c) or the chelator EDTA (d).    Inhibitors were tested on confluent HEp-2 monolayer for 6 h, 24 h or 30 h. Cytotoxicity was visualized using alamarBlue cell viability reagent. Error bars indicate ± s.d.

Fig H in S1
Text. Wild-type YkfCCtr expression levels are required for optimal chlamydial growth. Constructs encoding dCas12-crRNA CRISPRi system targeting ykfCCtr or a non-targeting control were transformed into C. trachomatis (-pL2). McCoy cells were infected with these transformants, and dCas12 expression was induced or not with 10 nM aTc at 4 hpi. At 24 hpi, RNA and genomic DNA were isolated and used for (RT)-qPCR to measure transcript and genomic DNA levels. (a) Transcript levels of euo, omcB, and ykfC in dCas12-induced conditions at 24 hpi in the non-targeting or ykfCCtr-targeting knockdown strains. (b) The levels of genomic DNA of nontargeting and ykfCCtr-targeting knockdown strains at 24 hpi in uninduced (UI) or induced (I) conditions. (c) Inclusion forming units of non-targeting and ykfCCtr-targeting knockdown strains at 24 hpi in uninduced and induced conditions. (a-c) Error bars represent ± standard deviation (n = 3). *: p<0.05; **: p<0.001 (d, e) IFA controls of non-targeting and ykfCCtr-targeting knockdown strains at 24 hpi. MOMP (green) and dCas12 (red) were stained. The images were acquired on a Zeiss AxioImager.Z2 equipped with an Apotome2 using a 100X objective. Scale bar: 2 μm.     The in silico predicted secondary structure of NlpC/P60Sne is shown in grey above the respective sequences. The catalytic triad comprising Cys, His and a polar residue is shown in red, residues contributing to the S1 and S2 sites essential for substrate recognition in YkfCBce are highlighted in yellow and blue, respectively. Residues contributing to both sites are further highlighted in green. The signal peptide sequence of YkfCBce is displayed in grey. . Antibodies raised against NlpC/P60 of E. lausannensis (b) and S. negevensis (c) specifically label bacteria. Cells infected with S. negevensis or E. lausannensis were labelled with mice serum taken before (pre-immune) or after immunization with the indicated protein (green), with species-specific antibodies (red) and DAPI (blue). Bar = 2 μm.

Supplementary Material and Methods
Cloning of C. trachomatis. For ctl0328 (homolog of ct127)-knockdown in C. trachomatis serovar L2 (strain 434/Bu), a gBlock (IDTDNA, Coralville, IA) encoding a crRNA cassette was designed to target the 5' intergenic region and inserted into pBOMBL12CRia linearized by BamHI using HiFi (NEB) as described [3]. As a control, a scrambled non-targeting crRNA gBlock cassette was also inserted into the BamHI-digested pBOMBL12CRia empty vector. The HiFi products were transformed into NEB10β competent cells (NEB) and plated on LB agar containing 100 μg/mL ampicillin.
Immunofluorescence RT-qPCR analysis. McCoy cells were infected with the C. trachomatis L2 transformants carrying the aTc-inducible constructs encoding dCas12 with or without constitutively expressed crRNA targeting ykfCCtr or the non-targeting control. At 4hpi, the constructs were induced or not with 10 nM aTc. RNA and genomic DNA were collected using Trizol (Invitrogen) and DNeasy Tissue (Qiagen) kit as described previously [4]. The isolated RNA was treated with DNase to remove DNA contamination, and cDNA was synthesized with the DNased RNA using Superscript III reverse transcriptase (Invitrogen). Equal volumes of cDNA were used in qPCR reactions [4]. Similarly, equal mass of DNA was used from each sample in qPCR reactions using the primers listed in Supplementary Table 1 to quantify chlamydial genomes, which were used to normalize the amount of cDNA.