Outer membrane lipoprotein NlpI scaffolds peptidoglycan hydrolases within multi‐enzyme complexes in Escherichia coli

Abstract The peptidoglycan (PG) sacculus provides bacteria with the mechanical strength to maintain cell shape and resist osmotic stress. Enlargement of the mesh‐like sacculus requires the combined activity of peptidoglycan synthases and hydrolases. In Escherichia coli, the activity of two PG synthases is driven by lipoproteins anchored in the outer membrane (OM). However, the regulation of PG hydrolases is less well understood, with only regulators for PG amidases having been described. Here, we identify the OM lipoprotein NlpI as a general adaptor protein for PG hydrolases. NlpI binds to different classes of hydrolases and can specifically form complexes with various PG endopeptidases. In addition, NlpI seems to contribute both to PG elongation and division biosynthetic complexes based on its localization and genetic interactions. Consistent with such a role, we reconstitute PG multi‐enzyme complexes containing NlpI, the PG synthesis regulator LpoA, its cognate bifunctional synthase, PBP1A, and different endopeptidases. Our results indicate that peptidoglycan regulators and adaptors are part of PG biosynthetic multi‐enzyme complexes, regulating and potentially coordinating the spatiotemporal action of PG synthases and hydrolases.

: Appendix Table S2: Appendix Table S3: In vivo and in vitro proteomic interaction assays link NlpI to several classes of PG hydrolases. NlpI dimerizes and interacts with several endopeptidases in vitro. Assaying for interactions between endopeptidases and with NlpI by Ni 2+ -NTA pulldown assay. In vitro PG digestion assays. NlpI localizes along the cell envelope. Reconstitution of PG multi-enzyme complexes with PBP1A and LpoA. Initial fluorescence scans for MST experiments between PBP1A/LpoA with NlpI/EPases. PBP4 interacts simultaneously with NlpI and the PBP1A-LpoA synthase complex. PBP1A-LpoA activity is not significantly affected by the presence of MepS and/or NlpI. MepS*-NlpI does not affect the TPase activity of PBP1A-LpoA. BW25113 vs. BW25113∆nlpI MP analysis. Expression of NlpI in a nlpI strain partially restores wild-type morphology. MepS levels impact cell width.

Fig S3. Assaying for interactions between EPases and with NlpI by Ni 2+ -NTA pulldown assay.
A. SDS-PAGE (12-15%) analysis of applied (A) and eluted (E) samples from Ni 2+ -NTA pull down assay of NlpI-EPase combinations. Equimolar concentrations of proteins for each respective assay (1 or 2 μM) were incubated with Ni 2+ -NTA beads in combination or alone. Retention of tagless protein in the presence of His-tagged partner indicates an interaction. Tagless PBP5 is not retained in the presence of His-tagged NlpI nor is tagless NlpI retained by His-Slt. B. SDS-PAGE (12-15%) analysis of applied (A) and eluted (E) samples from Ni 2+ -NTA pull down assay of His-MepS with PBP7 and His-PBP7 with PBP4. Equimolar concentrations of proteins for each respective assay (1-2 μM) were incubated with Ni 2+ -NTA beads in combination or alone.

Fig. S4. in vitro PG digestion assays.
NlpI moderately affects the activity of MepM, but less so the activity of MepS, PBP4 and PBP7. Representative HPLC chromatograms of assays containing the respective EPase with or without NlpI. Control samples were incubated with no enzyme. Muropeptides were separated and the PG profiles determined as previously described (Glauner, 1988). Figure shows representative chromatograms of 3-6 independent experiments.

Fig. S5. NlpI localizes along the cell envelope.
A. MC4100 cells grown to steady state in minimal glucose medium at 28°C were immunolabeled with antibodies specific for NlpI. The map of diameters (magenta) and map of fluorescence (green) of NlpI localization where cells are sorted according to their cell length are shown. The concentration of NlpI in molecules per µm -2 wall area as function of the normalized cell division cycle age is constant. The grey dots are the values measured for the individual cells and the black markers are the 5% age bins with the 95 confidence error bars. The extra fluorescence at midcell compared to the rest of the cell (FCPlus) plotted as function of the normalized cell division cycle. NlpI is not specifically present at midcell (number of analyzed cells is 6087). B. A high resolution fluorescence SIM image of BW23115 cells that were grown in LB at 37°C immunolabeled with anti-NlpI on the right and the corresponding Phase contrast image on the left is shown in the bottom panel. Scale bar equals 2 µm. C. Wildtype BW25113 and the isogenic ∆nlpI strain were grown in TY at 37°C and fixed. The serum against NlpI was absorbed (pre) to the ∆nlpI cells and the remaining antibodies were used to label the wildtype strain (upper images) and another sample of the ∆nlpI cells (lower images). From left to right are shown: a phase contrast and the corresponding fluorescence image of the labelled cells and the map of diameters (magenta) and map of fluorescence (green) of the cells sorted according to ascending length. Scale bar equals 2 µm. Graph of the concentration NlpI (AU) in the envelope of the cells plotted against the cell division cycle time (%) are also shown. The dots are the data on the individual cells, whereas the markers are the 5% age bins with the 95 confidence error bars. D. Western blot using anti-NlpI for constructs carrying different chromosomally tagged NlpI versions, and the wild-type and ΔnlpI controls. Cells were grown exponentially for 2 hours, harvested and adjusted to 2 μg total protein per lane. Samples were separated by SDS PAGE and visualised by Western Blot using primary antibodies against NlpI or RpoD, and anti-rabbit HRP. Samples are as followed: WT: BW25113; Δ: BW25113ΔnlpI::kan ; SF: BW25113nlpI::strep::flag::kan; HA: BW25113nlpI::HA::frt. E. Wildtype BW25113 and the isogenic ∆nlpI strain and the same strain chromosomally expressing NlpI-HA were grown in TY at 37°C and fixed and labelled with anti HA. From left to right are shown: a phase contrast and the corresponding fluorescence image of the labelled cells and the map of diameters (magenta) and map of fluorescence (green) of the cells sorted with ascending length. Scale bar equals 1 µm. Also shown in panel D is a graph of the concentration HA-NlpI (AU) in the cells plotted against the cell division cycle time (%). The dots are the data on the individual cells, whereas the markers are the 5% age bins with the 95% confidence error bars. F Cell morphology complementation assay of various NlpI versions. The area of the cells was determined from phase contrast images assuming the shape of the cells to consist of a cylinder with two half spheres as poles. BW is BW25113 the parental strain of the nlpI deletion strain ΔnlpI. Next sample is the ΔnlpI transformed with a plasmid that expresses NlpI under control of an arabinose inducible promotor (either without arabinose of after induction for 2 hours with arabinose (no difference between 01% or 0.2%). The last two samples are chromosomal fusions of NlpI with either a C-terminal HA-tag or a Strep-Flag tag.  Capillary scans for samples showing ligand concentration dependent changes in fluorescence were repeated after boiling samples in reducing agent and SDS to abolish ligand binding. The fluorescence reads for samples containing highest and lowest concentration of ligand were within the margin of error, suggesting that differences in initial fluorescence were due to ligand binding and not different Fl-protein concentration, validating the use of the raw fluorescence data to plot binding curves.  Fig. S6B and is displayed here again for completeness. MST curves plotted are the mean data +/-SD of three independent experiments. Fl, fluorescently labelled; FNorm, normalized fluorescence. B. Fixed concentration MST assays with LpoA-PBP4-NlpI showed differences in initial fluorescence. These differences were eliminated upon boiling of samples in SDS (which abolishes ligand binding), which suggests that the initial differences were due to ligand binding and not due to inaccurate pipetting, validating the use of the raw fluorescence data to plot binding curves.

D.
The initial gradient of slope (in panel C) against time is plotted as a graph to indicate lipid II incorporation. n.s., not significant. (p-value < 0.05 was calculated using Students T-test) E. Transpeptidase (TPase) activity of PBP1A ± LpoA in the presence or absence of MepS* and NlpI. Experiments were done in duplicate and error bars indicate range. Representative chromatograms shown in Fig. S10. and muropeptide quantifications are displayed in Table EV 6.    Variations of MepS levels have no impact on growth rate. Strains were grown in LB at 30°C. For the strain ΔmepS, pBAD30-mepS, the medium was supplemented with carbenicillin (100 μg/mL) and 0,2% glucose to repress mepS or 0,2% L-arabinose for induction. OD600 was measured over time. B. Alternative representation of Fig. 3f and S13c. Points indicate single-cell width (left plot) and length (right plot). Squares indicate the population average, and error bars indicate the standard deviation between single cells. The calculated p-value (p) between two populations is indicated and the value of the mean width and length is indicated in red ( m). The number of cells (n) measured for the WT strain is 187,