Reconstituting Spore Cortex Peptidoglycan Biosynthesis Reveals a Deacetylase That Catalyzes Transamidation

Some bacteria survive in nutrient-poor environments and resist killing by antimicrobials by forming spores. The cortex layer of the peptidoglycan cell wall that surrounds mature spores contains a unique modification, muramic-δ-lactam, that is essential for spore germination and outgrowth. Two proteins, the amidase CwlD and the deacetylase PdaA, are required for muramic-δ-lactam synthesis in cells, but their combined ability to generate muramic-δ-lactam has not been directly demonstrated. Here we report an in vitro reconstitution of cortex peptidoglycan biosynthesis, and we show that CwlD and PdaA together are sufficient for muramic-δ-lactam formation. Our method enables characterization of the individual reaction steps, and we show for the first time that PdaA has transamidase activity, catalyzing both the deacetylation of N-acetylmuramic acid and cyclization of the product to form muramic-δ-lactam. This activity is unique among peptidoglycan deacetylases and is notable because it may involve the direct ligation of a carboxylic acid with a primary amine. Our reconstitution products are nearly identical to the cortex peptidoglycan found in spores, and we expect that they will be useful substrates for future studies of enzymes that act on the spore cortex.


Expression and purification of CwlD and PdaA -general protocol used for all proteins. E. coli C43(DE3)
containing the appropriate plasmid was grown in 500 mL LB broth supplemented with kanamycin at 37 ºC with shaking until the OD600 was 0.5-0.6. The culture was cooled to 16 ºC before inducing protein expression with 500 µM IPTG and shaking for 16 h. Cells were harvested by centrifugation (4,200 ´ g, 15 min, 4 ºC) and resuspended in 20 mL lysis buffer (20 mM Tris pH 7.5, 400 mM NaCl). The cell suspension was supplemented with DNase (0.5 mg/mL) and phenylmethanesulfonyl fluoride (1 mM) and the cells lysed by passage through a cell disruptor (EmulsiFlex-C3, Avestin) at ³ 10,000 psi. Cell debris was pelleted by centrifugation (20,000 ´ g, 30 min, 4 ºC). The resulting supernatant was supplemented with 40 mM imidazole and then rocked with 0.5 mL Ni-NTA resin (Qiagen) for 45 min at 4 ºC. The resin was collected in a column by gravity flow and then washed twice with 5 mL wash buffer (20 mM Tris pH 7.5, 400 mM NaCl, 40 mM imidazole). The protein was eluted in 10 mL elution buffer (20 mM Tris pH 7.5, 400 mM NaCl, 200 mM imidazole) and concentrated by centrifugal filtration. The protein was then further purified by fast protein liquid chromatography (FPLC, AKTA Pure, Cytiva) on a Superdex 200 Increase 10/300 column (Cytiva) in a running buffer consisting of 20 mM Tris pH 7.5, 400 mM NaCl. Pooled elution fractions were concentrated by centrifugal filtration and the protein absorbance measured at 280 nm. The predicted extinction coefficient of the protein (via ProtParam 3 ) was used to estimate concentration. Protein was diluted to 200 µM in running buffer with 10% glycerol (v/v), aliquoted, and stored at -80 ºC.
CwlD and PdaA reactions -general conditions. B. subtilis Lipid II was polymerized with SgtB, a monofunctional peptidoglycan glycosyltransferase from Staphylococcus aureus. Pooled polymerization reactions of up to 1 mL total volume were assembled under the following conditions: 50 mM HEPES, pH 7.5, 2 mM CaCl2, 20 µM Lipid II, 0.2 µM SgtB, 10% DMSO (v/v). Reactions were incubated at room temperature for 30 min. CwlD and/or PdaA were added at 2 µM and the reactions incubated static at room temperature. Reactions were quenched by addition of EDTA at a final concentration of 10 mM and the mixture immediately frozen at -20 ºC. We note that trial experiments revealed that BsCwlD and BsPdaA exhibited measurable activity between pH 6 and 8.5 with the optimal pH for both enzymes being 7-7.5.
Digestion and LC-MS analysis of reaction products. Peptidoglycan products of CwlD/PdaA reactions were digested with mutanolysin to enable LC-MS analysis. To 50 µL aliquots of peptidoglycan reaction products, 8 U of mutanolysin was added and the reaction incubated at 37 ºC for 2 h. Aqueous sodium borohydride (10 mg/mL, 50 µL) was added and the reaction incubated for 30 min at room temperature. The solution pH was adjusted to ~4 by addition of 20% phosphoric acid (approximately 5 µL) and the reactions lyophilized to dryness. The residue was dissolved in 25 µL of water and analyzed by LC-MS.
LC-MS was conducted using an Agilent Technologies 1200 series HPLC in line with an Agilent 6520 Q-TOF mass spectrometer using electrospray ionization and operating in positive ion mode. Products were separated on a

Supporting Information
Tobin et al.

S-4
Waters Symmetry Shield RP18 column (5 µm, 3.9 ´ 150 mm) with matching column guard using the following prepared under the general reaction conditions described above. A 50 µL aliquot was removed and frozen at -20 ºC (the "peak A" sample). BsCwlD (2 µM) was added and the reaction incubated at room temperature for 1 h. A 100 µL aliquot was removed, EDTA added at 10 mM, and the sample frozen at -20 ºC (the "peak B" sample). CdPdaA1 (2 µM) was added to the pooled mixture and the reaction incubated at room temperature for 5 min. A 100 µL aliquot was removed, EDTA added at 10 mM, and the sample frozen at -20 ºC (the "peak C" sample). The remaining reaction mixture was incubated at room temperature for 20 h then split into two 100 µL aliquots. Mutanolysin (8 U) was added to all samples and the reactions incubated at 37 ºC for 2 h. To generate the "peak D" sample, 100 µL of 2 mg/mL aqueous sodium borohydride was added to an aliquot of 20 h reaction products and the reaction incubated at room temperature for 30 min. All other samples were reduced similarly with 10 mg/mL sodium borohydride with the second aliquot of 20 h reaction products representing the "peak E" sample. The pH of all reactions was adjusted to ~4 by addition of 20% phosphoric acid and the samples lyophilized to dryness. The residue was dissolved in 25 µL of water.
The muropeptide reaction products were analyzed using the same LC-MS method and instrumentation as above. Mutanolysin digestion and LC-MS analysis were conducted as described above. Relative product amounts were calculated by integrating the product peaks and dividing by the total peak area.

Lactam Cyclization Assay.
To generate linear peptidoglycan substrates enriched in muramic acid residues (product C), pooled reactions were assembled as described in the generic reaction conditions, above. Linear B. subtilis peptidoglycan was treated with BsCwlD (2 µM) for 1 h at room temperature. BsPdaA or CdPdaA1 (2 µM) was then added for 10 min and the mixture immediately passed over a 100 µL plug of settled Ni-NTA resin (Qiagen) by gravity flow. The eluate was passed over the resin a second time to maximize protein binding. The eluate was then split into 50 µL aliquots and treated as indicated in the main text. CwlD or PdaA variants were re-added at 2 µM and the resulting mixture incubated at room temperature for 20 h. All reactions were quenched by addition of 10 mM EDTA and stored at -20 ºC until needed. Mutanolysin digestion and LC-MS analysis were conducted as described above.
Western Blotting. Proteins separated by SDS-PAGE were transferred to a PVDF membrane. The membrane was blocked in tris-buffered saline with 5% dry milk (w/v) and 0.05% tween-20 (v/v) for 1 h at room temperature. His6tagged proteins were detected by incubation with THE™ His Tag Antibody [HRP] (Genscript, 1:5000 in blocking buffer) for 1 h at room temperature. The membrane was washed with tris-buffered saline, 0.05% tween-20. His-tagged proteins were detected by incubating the membrane in ECL reagent (ThermoFisher) and measuring chemiluminescence using an Azure 500 imager (Azure Biosystems).   Figure S1. Coomassie-stained SDS-PAGE of purified proteins. Each lane contains 2 µg of protein.

Supporting Information
Tobin et al.   We note that while simultaneous addition of BsCwlD and BsPdaA gives lactam as the major product, PdaA gave optimal conversion to lactam when added after preincubation with BsCwlD ( Figure 2). At 48 h, the C. difficile enzymes gave yields of up to 31% product E. This corresponds to 24% of all MurNAc monomers having been converted to muramic-d-lactam.

S-8
We note that CdCwlD exhibits low activity due to the absence of the activating protein GerS, which is also required for muramic-d-lactam synthesis in C. difficile. 6,7 CdPdaA1 exhibited robust activity when provided substrates enriched in products B or C, as demonstrated in Figure 3, S4, and S8.