Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Jun 26;11(6):e1005010.
doi: 10.1371/journal.ppat.1005010. eCollection 2015 Jun.

Phosphorylation of the Peptidoglycan Synthase PonA1 Governs the Rate of Polar Elongation in Mycobacteria

Free PMC article

Phosphorylation of the Peptidoglycan Synthase PonA1 Governs the Rate of Polar Elongation in Mycobacteria

Karen J Kieser et al. PLoS Pathog. .
Free PMC article


Cell growth and division are required for the progression of bacterial infections. Most rod-shaped bacteria grow by inserting new cell wall along their mid-section. However, mycobacteria, including the human pathogen Mycobacterium tuberculosis, produce new cell wall material at their poles. How mycobacteria control this different mode of growth is incompletely understood. Here we find that PonA1, a penicillin binding protein (PBP) capable of transglycosylation and transpeptidation of cell wall peptidoglycan (PG), is a major governor of polar growth in mycobacteria. PonA1 is required for growth of Mycobacterium smegmatis and is critical for M. tuberculosis during infection. In both cases, PonA1's catalytic activities are both required for normal cell length, though loss of transglycosylase activity has a more pronounced effect than transpeptidation. Mutations that alter the amount or the activity of PonA1 result in abnormal formation of cell poles and changes in cell length. Moreover, altered PonA1 activity results in dramatic differences in antibiotic susceptibility, suggesting that a balance between the two enzymatic activities of PonA1 is critical for survival. We also find that phosphorylation of a cytoplasmic region of PonA1 is required for normal activity. Mutations in a critical phosphorylated residue affect transglycosylase activity and result in abnormal rates of cell elongation. Together, our data indicate that PonA1 is a central determinant of polar growth in mycobacteria, and its governance of cell elongation is required for robust cell fitness during both host-induced and antibiotic stress.

Conflict of interest statement

The authors have declared that no competing interests exist.


Fig 1
Fig 1. PonA1 is essential in M. smegmatis and required for normal growth of M. tuberculosis.
(A) An allelic exchange system in M. smegmatis provides an efficient method to test the importance of PonA1 for bacterial survival. PonA1 is essential in M. smegmatis, as allelic exchange with a vector encoding ponA1 complements bacterial growth, while allelic exchange with a negative control vector fails to rescue growth. (B) C57Bl6 mice were aerosol infected with H37Rv wildtype, ΔponA1, ΔponA1::ponA1 wt, and ΔponA1::ponA1 TG- (an allele of PonA1 with a catalytic active site mutation in the transglycosylase (TG) domain), and CFU were enumerated were from lung and spleen homogenates at 15 and 42 days post infection (dpi). ΔponA1 and ΔponA1::ponA1 TG- cells are less fit than H37Rv wildtype or isogenic wildtype, respectively, at 15 and 42 dpi. Statistical significance was calculated by a one-tailed t-test (lungs at 15 dpi, H37Rv compared to ΔponA1 *** indicates p-value = 0.0005; lungs 42 dpi, H37Rv compared to ΔponA1 ** indicates p-value = 0.0089, and ΔponA1::ponA1 wt compared to ΔponA1::ponA1 TG- ** indicates p-value = 0.0042. Spleens at 42 dpi, H37Rv compared to ΔponA1 *** indicates p-value = 0.0001, and ΔponA1::ponA1 wt compared to ΔponA1::ponA1 TG- * indicates p-value = 0.0122). PonA1, while not required for growth of M. tuberculosis in culture, is required for normal bacterial multiplication and dissemination during infection in a mouse model of tuberculosis.
Fig 2
Fig 2. PonA1’s glycan synthesis is required in M. smegmatis and catalytic activity promotes normal cell elongation.
(A) Allelic exchange (see Fig 1A) with vectors encoding alleles of PonA1 with varying catalytic activities (TG-, transglycosylase mutant; TP-, transpeptidase mutant; TG-TP-, transglycosylase and transpeptidase double mutant) demonstrates that a TG- allele fails to rescue bacterial survival. In contrast, a TP- allele complements bacterial growth. It is likely that PonA1’s synthesis of the glycan backbone of peptidoglycan is required for cell survival while other enzymes coordinate with PonA1 to crosslink those glycan strands into the existing sacculus. (B) M. tuberculosis strains were grown in culture and total cell length was measured: H37Rv wildtype (234 cells), ΔponA1 (241 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test), ΔponA1 L5::ponA1 wt (212 cells), ΔponA1 L5::ponA1 TG- (220 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test), ΔponA1 L5::ponA1 TP- (227 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test), and ΔponA1 L5::ponA1 TG-TP- (209 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test). Loss of PonA1 or expression of a TG- allele of PonA1 reduces cell length compared to their isogenic wildtypes, supporting the role of PonA1’s TG activity in regulating cell growth in M. tuberculosis. Loss of TP activity had a more modest effect on cell length. (C) Msm PonA1 isoforms were produced in E. coli and assessed for their ability to bind bocillin-FL. Lane 1, negative control (wildtype E. coli); lane 2, PonA1wt; lane 3, PonA1TP-; lane 4, PonA1TG-; lane 5, PonA1TG-TP-. Mutants that lack TP activity cannot bind bocillin-FL. However, PonA1TG- binds bocillin-FL at a level similar to wildtype PonA1, showing that PonA1’s catalytic activities can be uncoupled. (D) Msm cells that express PonA1TP- do not exhibit gross cell shape changes, suggesting another enzyme coordinates peptidoglycan crosslinking with PonA1 to maintain cell integrity. Scale bar, 2 μm. (E) ΔponA1 L5::ponA1 TP- cells (229 cells) are shorter than wildtype M. smegmatis (wt Msm, 251 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test) or isogenic wildtype (ΔponA1 L5::ponA1 wt, 247 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test), indicating that PonA1’s crosslinking influences cell length.
Fig 3
Fig 3. PonA1’s periplasmic domains modulate cell shape in mycobacteria.
(A) Replacement of wildtype PonA1 with a PonA1 truncation containing the cytoplasmic tail and TG domain (PonA11-360) rescues bacterial survival. Cells exhibit shape changes, but survive, suggesting that PonA1’s TG domain is the sufficient periplasmic domain required for bacterial growth. Cell shape may change due to missing protein-protein interactions that occur along the TP domain. Scale bar, 2 μm. (B) PonA11-360 cells (266 cells) are shorter than isogenic wildtype (251 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test), which may indicate that the TP domain plays a role in complex formation that influences cell length. These measurements were taken on a different day than Fig 2C and 2D, hence the slight difference in isogenic wildtype cell length.
Fig 4
Fig 4. Excess PonA1 induces ectopic polar growth in M. smegmatis.
(A) Overexpression of PonA1 causes the formation of ectopic growth poles, regardless of PonA1’s catalytic activity, suggesting that PonA1’s complex formation is sufficient to induce ectopic polar growth. Scale bar, 2 μm. (B) Excess levels of PonA1, including wildtype PonA1, inhibit bacterial multiplication, likely because of ectopic polar growth. (C) Although PonA1’s catalytic activity is not required for ectopic pole formation, it does influence the frequency at which ectopic poles form. Cells that express a TG- allele form ectopic poles at more than 2 times the frequency of other PonA1 alleles, suggesting that local peptidoglycan architecture could influence ectopic pole formation (statistical significance was assessed by one-way analysis of variance with Bonferroni’s multiple comparison test, and multiplicity adjusted p-values are reported. PonA1wt compared to PonA1TG-, p-value = 0.0039; PonA1TG- compared to PonA1TP-, p-value = 0.0225; PonA1TG- compared to PonA1TG-TP-, p-value = 0.0042). (D) PonA1-RFP localizes to both growing tips of the ectopic pole, indicating that PonA1 is involved in nucleating elongation complexes at these growth tips.
Fig 5
Fig 5. Phosphorylation regulates the rate of cell elongation.
(A) Phospho-transfer profiling with the kinase domains of the major serine-threonine protein kinases of M. tuberculosis (Mtb Pkn) reveals that PknB efficiently phosphorylates Mtb MBP-PonA1cyto. (B) M. smegmatis cells that express a T50A allele of ponA1 (134 cells; approximate p-value = 0.0145 by the Kolmogorov-Smirnov test) are longer than isogenic wildtype cells (219 cells), while cells that express a T50D allele (139 cells; approximate p-value = 0.0082 by the Kolmogorov-Smirnov test) are shorter than isogenic wildtype, suggesting PonA1’s phosphorylation regulates cell elongation or division. (C) Timelapse microscopy revealed that cells that expressed a T50A (127 cells; approximate p-value = 0.0002 by the Kolmogorov-Smirnov test) allele elongated faster than isogenic wildtype cells (174 cells), which was phenocopied by a truncation of the cytoplasmic tail (Δcyto; 202 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test). These data suggest that PonA1’s phosphorylation negatively regulates cell elongation. (D) Similarly, phosphorylation status of PonA1 affects total cell length in M. tuberculosis. Cells that expressed a T34A allele (211 cells; approximate p-value = 0.0066 by the Kolmogorov-Smirnov test) exhibited an average cell length 5% longer than isogenic wildtype (202 cells), and cells that expressed a T34D allele (207 cells; approximate p-value < 0.0001 by the Kolmogorov-Smirnov test) were 11% shorter than isogenic wildtype. This suggests that PonA1’s unusual phosphorylation negatively regulates cell elongation in M. tuberculosis as in M. smegmatis. (E) Msm cells that encode a T50A,TP- allele of PonA1 are defective for normal cell separation. These cells form short chains of cells with multiple septa (white arrows). These data suggest that PonA1’s phosphorylation may regulate PonA1 TG activity, the remaining functional catalytic activity for this allele, and that alterations to PonA1’s TG activity impact the cell’s peptidoglycan and consequent cleavage of that peptidoglycan.
Fig 6
Fig 6. PonA1’s transpeptidase activity is required for normal teicoplanin tolerance in mycobacteria.
(A) Minimum inhibitory concentration (MIC) of teicoplanin for wildtype and different PonA1 mutant M. tuberculosis strains is tabulated. Fold change in MIC is calculated from wildtype Mtb (for ΔponA1) or isogenic wildtype (for the TG-, TP-, TG-TP-, T34A, and T34D strains). Cells that express a PonA1 TP mutant or the T34A allele are more sensitive to teicoplanin than isogenic wildtype. No change, no fold change in MIC from appropriate wildtype strain. (B) Teicoplanin MICs for M. smegmatis mutant strains are tabulated. Fold change is calculated from isogenic wildtype. Cells that express a PonA1 TP mutant are more sensitive to teicoplanin than isogenic wildtype.
Fig 7
Fig 7. A model for how PonA1 promotes cell elongation in mycobacteria.
(A) We propose a model wherein PonA1 localizes early to the growth tip and promotes pole elongation through PG synthesis. PonA1 may recruit other factors to form the elongation complex or they are recruited independently of PonA1 (factors not shown). Together with these factors, polar elongation proceeds (orange cells). PonA1 furthers cell elongation through its synthesis of PG. PonA1’s synthesis of glycan strands (colored subunits, panels 1–3) and its crosslinking of those glycan strands (black lines, panels 2–3) promote normal cell length. These catalytic activities are potentially modulated through PonA1’s phosphorylation (arrow), which acts to regulate the rate of cell elongation. (B) Excess PonA1 stimulates ectopic cell elongation (orange cells). The frequency of ectopic pole formation increases with imbalanced PG crosslinking and may result from changes in local peptidoglycan architecture (panels 1–2, factors removed from panel 2 for clarity). These architectural changes may act like a sink that ‘recruits’ additional protein-interactors of PonA1 or additional elongation complex components to spur cell elongation at ectopic sites (panel 3, dark gray factors).

Similar articles

See all similar articles

Cited by 24 articles

See all "Cited by" articles


    1. World Health Organization. Global tuberculosis report 2014. World Health Organization; 2014.
    1. Kieser KJ, Rubin EJ. How sisters grow apart: mycobacterial growth and division. Nat Rev Micro. 2014;12:550–62. - PMC - PubMed
    1. Typas A, Banzhaf M, Gross CA, Vollmer W. From the regulation of peptidoglycan synthesis to bacterial growth and morphology. Nat Rev Micro.; 2012;10:123–36. - PMC - PubMed
    1. Hett EC, Chao MC, Rubin EJ. Interaction and Modulation of Two Antagonistic Cell Wall Enzymes of Mycobacteria. PLoS Pathog. 2010;6:e1001020 10.1371/journal.ppat.1001020 - DOI - PMC - PubMed
    1. Chao MC, Kieser KJ, Minami S, Mavrici D, Aldridge BB, Fortune SM, Alber T, Rubin EJ. Protein Complexes and Proteolytic Activation of the Cell Wall Hydrolase RipA Regulate Septal Resolution in Mycobacteria. PLoS Pathog. 2013;9:e1003197 10.1371/journal.ppat.1003197 - DOI - PMC - PubMed

Publication types

MeSH terms