Heterologous expression of MreBMx in E. coli causes MreBEc to aggregate at cell poles.

a) Heterologous expression of MreBMx in E. coli strains DH5α, MG1655, and BW25113 causes E. coli cells to form dark aggregates at cell poles. Insets show the morphology of cells carrying the empty pMAT3 vector. b) CryoEM images of an E. coli MG1655 cell that expresses MreBMx. White arrow points to the polar aggregate. In the cells with polar aggregates, PG thickens at cell poles (inset i, red arrow), whereas PG is rarely visible in nonpolar regions (insert ii). c) MreBMx expresses at very low levels, even when induced by high concentration of CuSO4. The native expression of MreBMx from the same number of M. xanthus (Mx) cells is shown in the first lane as a reference. Black arrow points to the bands detected by an anti-MreBMx antibody. d) MreBEc-msfCFPSW forms dispersed foci in the NO59 strain. e) In BN04 (NO59 pMAT3-mreBMx-PAmCherrySW) cells, instead of forming filaments, MreBMx and MreBEc colocalize in polar aggregates. f) While BN04 strain retains rod shape, expressing MreBMx increases both the length and width of cells. BF, bright field.

Heterologous expression of MreBMx causes E. coli to elongate PG at cell poles.

a) NO59 cells elongate PG at nonpolar regions in a dispersed manner, consistent with MreBEc localization pattern. PG elongation was labeled by BADA. White arrows point to cell poles. b) Quantitative analysis of PG elongation across 156 NO59 cells. Cell lengths were normalized to 1 and BADA fluorescence intensity profiles were standardized using Z-score normalization (0 indicates the mean value for each cell). Insets here and in panel e illustrate patterns of PG elongation, with newly synthesized PG shown in green. Imaged cell regions are highlighted by dotted-line boxes. c) PG elongation mode in NO59 cells visualized by pause-chase sequential labeling with BADA and TADA, where both dyes were incorporated in a dispersed, nonpolar pattern. d) BN04 cells that express MreBMx-PAmCherrySW elongate PG at poles, consistent with the localization patterns of both MreBEc and MreBMx. PG elongation was visualized by BADA. e) Quantitative analysis of PG elongation across 135 BN04 cells. For each cell, its BADA fluorescence intensity profile was measured starting from the only or brighter cell pole. f) Elongation mode in BN05 cells visualized by pause-chase sequential labeling with BADA and TADA. While the new growth (TADA) occurred at cell poles, the old growth (BADA) receded to nonpolar regions. BF, bright field. Scale bars, 5 μm.

The relocalized Rod system carries out the reprogrammed, polar PG elongation.

a) In the absence of heterologous MreB, E. coli cells elongate PG using both the Rod system and aPBPs. Neither moenomycin (MOE, 10 μg/ml) that inhibits all aPBPs, mecillinam (MEC, 100 μg/ml) that inhibits PBP2 in the Rod system, or A22 (10 μg/ml) that inhibits the polymerization of MreB filaments and thus disrupts the assembly of Rod complexes, abolishes PG elongation completely. PG elongation was visualized using BADA. b) E. coli cells expressing MreBMx solely rely on the Rod system for polar PG elongation, as MEC alone is sufficient to abolish PG elongation. Such polar PG elongation is resistant to A22, suggesting that polar localized Rod enzymes elongate PG independent of MreB filaments. PG elongation was visualized using BADA. c) Heterologous expression of MreBMx recruits E. coli PBP2 to cell poles. d) In the absence of RodZ, MreBEc still forms aggregates at pole-like locations (white arrows) but fails to induce focused PG growth. PG elongation was visualized using BADA. e) The reprogrammed, polar PG elongation does not depend on the divisome as FtsZ does not colocalize with polar MreB aggregates.

Targeting PBP2 to cell poles is sufficient to drive polar PG elongation.

a) When PBP2 (encoded by mrdA) is ectopically expressed with a polar-targeting tag (PopZ-H3H4) by an arabinose-inducible promoter under 0.4% arabinose (Ara), MreBEc aggregates at cell poles and cells switch to polar PG elongation and this reprogrammed growth mode is resistant to A22 (10 μg/ml). The white arrow points to a cell that still elongated PG at nonpolar sites. b) Quantitative analysis of PG elongation. Cell lengths were normalized to 1 and BADA fluorescence intensity profiles were standardized using Z-score normalization (0 indicates the mean value for each cell). Insets illustrate patterns of PG elongation, with newly synthesized PG shown in green. The numbers of cells that adopt each elongation pattern were presented above the insets. Imaged cell regions are highlighted by dotted-line boxes. c) Expressing pole-targeting PBP2 in the absence of MreBEc neither induces focused PG elongation (labeled by BADA) nor restores rod-shape, despite that Rod enzymes still incorporate BADA. aPBPs were inhibited by moenomycin (MOE, 4 μg/ml) for 1 h before adding BADA for 15 min. Scale bars, 5 μm.

Bacterial strains, plasmids, and primers used in this study.

BN04 cells grow slower than its parental strain NO59.

Averages and standard deviations were calculated from three technical repeats.