Supplemental Material
Coordinating bacterial cell division with nutrient availability: a role for glycolysis
Leigh G. Monahana, Isabella V. Hajduka, Sinead P. Blaberb, Ian G. Charlesa and Elizabeth J. Harrya,*
a The ithree institute, University of Technology, Sydney, NSW, Australia
b Department of Chemistry and Biomolecular Sciences, Macquarie University, North Ryde, NSW, Australia
Supplemental Results
Suppression of the ts1 division mutant in the absence of pyruvate kinase is not caused by growth effects or induction of stress responses
As described in the main text, deletion of the gene encoding pyruvate kinase (pyk) restores viability to the temperature sensitive ts1 strain of B. subtilis at 48°C. In addition, loss of pyruvate kinase also has a moderate effect on the growth rate of both ts1 and wild-type B. subtilis cells. For example, at 48°C ts1 cells containing the pyk deletion exhibited a 1.4-fold increase in mass doubling time relative to the parent strain in rich media (~35 min versus ~25 min). Importantly, we found that this reduction in growth rate alone is not responsible for the suppression of ts1. First, we observed that out of a collection of 12 metabolic mutations, 11 failed to restore viability in the ts1 strain at the non-permissive temperature (described in the main text). Many of these mutations did however affect cell growth. For example, pdhC and citC mutants were found to grow at a similar rate to the pyk mutant at 48°C (doubling time ~35 min), while a pdhD mutant grew even more slowly (~60 min) without suppressing ts1 thermosensitivity. In addition, we have shown previously that a deletion of the ClpX protease, which has major effects on cell growth, does not suppress ts1 (1). Finally, we grew ts1 cells on minimal media (SMM containing either 1% glucose or 1% succinate + 0.2% glutamate as the carbon source; see Materials and Methods) to slow down cell growth, and found that this did not restore viability at 48°C. Together, these results suggest that a mere decrease in growth rate is not sufficient to rescue the ts1 strain.
As well as affecting cell growth, mutations in central carbon metabolism genes can also cause changes in the concentrations of metabolites and energetic compounds. Because unbalanced metabolism might be sensed by cells as stress, we examined the effect of inducing stress on the temperature sensitivity of ts1. The ts1 strain was grown on plates supplemented (separately or in combination) with chemicals previously shown to induce energetic (125-250 mM sodium azide) or nutritional (0.5-2 mg/ml DL-norvaline) stresses in B. subtilis (2, 3). Suppression was not observed for any condition. Importantly, control experiments showed that for wild-type cells, growth was slowed but viability was not affected under these conditions. This indicates that the added compounds were active but at sub-lethal concentrations, and is consistent with the induction of stress responses. These results suggest that stress responses are not required for the suppression of ts1.
Pyruvate kinase does not modulate FtsZ assembly directly
The simplest model to account for the observed link between pyk and cell division is that the pyruvate kinase protein itself plays a direct role in modulating Z ring formation (see main text). However, we tested this idea in a number of ways and the resulting data argue strongly against the model. First, we were unable to detect a direct interaction between FtsZ and pyruvate kinase in the yeast two-hybrid system, using pyruvate kinase as either the bait or prey (data not shown). FtsZ interactions have been shown to be amenable to yeast two-hybrid analysis, and we could readily detect the well-known FtsZ-FtsA interaction by this method, indicating that the system itself was functional in our hands. It is important to note that false negatives are a genuine possibility in yeast two-hybrid studies, and it is also possible that pyruvate kinase interacts with another component of the division machinery to effect Z ring formation.
For this reason, we also examined the cellular localization of pyruvate kinase. All cell division and FtsZ regulatory proteins are known to exhibit a defined spatial pattern in vivo, most co-localizing with the Z ring (4, 5). However, a functional Pyk-GFP fusion protein was observed to localize homogenously throughout the cell, showing no evidence of any form of specific spatial localization when expressed as the sole copy of pyruvate kinase under the native promoter (Fig. S1).
Finally, we showed that normal Z ring assembly can be restored in a pyk mutant even in the complete absence of pyruvate kinase by the addition of exogenous pyruvate (see main text and below). Taken together, these data confirm that the observed link between pyk and Z ring formation is not mediated by the pyruvate kinase protein itself.
Exogenous pyruvate restores normal Z ring assembly in pyk mutant cells growing in defined media
Results presented in the main text (Fig. 5) show that the cell division defects of a pyk mutant can be completely rescued by the addition of 1% pyruvate to the growth medium. This strongly suggests that the defects are caused by a reduction in intracellular pyruvate levels stemming from the loss of pyruvate kinase activity. However, these experiments were performed in a complex growth medium (L broth) that is likely to contain a wide range of metabolites, including pyruvate, at unknown concentrations. It is therefore unclear to what extent pyruvate levels are actually reduced in pyk mutant cells under these conditions.
Importantly, previous studies have confirmed that intracellular pyruvate levels are severely depleted in B. subtilis pyk mutants when grown in defined media with glucose as the sole carbon source (6, 7). Taking advantage of these findings, we grew strain SU664 (Dpyk) in a minimal medium containing 1% glucose as the carbon source (SMM). We then examined whether cells were defective in Z ring formation, and if so, whether the addition of 1% pyruvate could alleviate the defect. In these experiments we used immunofluorescence microscopy to visualize Z rings rather than the xylose-inducible FtsZ-YFP fusion protein that was used routinely throughout this study. This approach was chosen as it avoids the need to add xylose to the growth medium, which could potentially serve as an alternative carbon source.
As shown in Fig. S2, SU664 (Dpyk) cells exhibited clear defects in Z ring assembly when grown in SMM. Similar to L broth, Z rings were frequently found at polar locations away from the normal midcell site and cells with multiple Z rings could often be seen (Fig. S2C). Importantly, the addition of exogenous pyruvate completely restored normal Z ring formation under these conditions (Fig. S2D). These results provide further evidence that pyruvate plays an important role in linking glycolysis with bacterial cell division.
Supplemental Materials and Methods
Strain construction
All B. subtilis strains used in this study are listed in Table S1. Strain SU664 (Dpyk) was constructed by allelic replacement of the pyk gene with a chloramphenicol resistance marker using the PCR joining approach of Fabret et al. (8). First, the 1.0 kb resistance marker was amplified from pJH101 (9) using primers 5¢-TTTTCGCTACGCTCAAATCC-3¢ and 5¢-AACCTTCTTCAACTAACGGG-3¢. DNA fragments flanking the pyk gene were then amplified from the B. subtilis chromosome (strain SU5) using primers 5¢-GGTGAATGGAGAATGAAACG-3¢ and 5¢-GGATTTGAGCGTAGCGAAAATGAAATCTTCAGCCTTCAGC-3¢ for the upstream fragment (1.0 kb), and 5¢-CCCGTTAGTTGAAGAAGGTTGCCAGGATATTACAGTTGAC-3¢ and 5¢-CTAACAGCAAAGCAATCAGC-3¢ for the downstream fragment (1.1 kb). Equal amounts of each PCR product (1 mg) were mixed and subjected to a final PCR reaction in the absence of added primers. Due to engineered complimentarity between the genomic fragments and the resistance marker, this reaction produced a joined fragment in which the marker was flanked by DNA from upstream and downstream of the pyk gene (see reference 8). The PCR mixture was transformed directly into wild-type strain SU5, selecting for chloramphenicol resistance to generate SU664. Correct integration was verified by PCR and sequencing.
Strain SU592 was constructed by single crossover insertion of a recombinant vector into the pyk gene, placing downstream genes under control of the Pspac promoter. First, a 406 bp fragment within pyk was amplified using primers 5¢-GCCAAGCTTTGTAGATGCCGCTAAACG-3¢ and 5¢-GGGGATCCCGCGTTGCATTTTTTGATC-3¢. The PCR product was then cloned into pMutinT3 (10) at the HindIII and BamHI sites, and correct clones were identified by restriction digestion and sequencing. The recombinant vector was transformed into B. subtilis strain SU111 to produce SU592. Correct single crossover integration was confirmed by PCR.
A pyk complementation strain (SU610), in which the sole copy of pyk is expressed under xylose-inducible control at the ectopic amyE locus, was constructed as follows. First, the pyk gene was amplified using primers 5¢-GGGGTACCATGAGAAAAACTAAAATTGTTTGT-3¢ and 5¢-CGGGATCCAATTAAAGAACGCTCGCACG-3¢, then cloned into pSG1191 (11) at the KpnI and BamHI sites (excising yfp from the plasmid backbone). Correct clones were identified by restriction digestion and sequencing, and the recombinant vector was transformed into strain SU592 to produce SU610. Double crossover integration of the Pxyl-pyk construct at amyE was confirmed by PCR.
Strain SU699 harbours a pyk-gfp fusion in place of the native pyk gene and was constructed as follows. First, pyk was amplified with primers 5¢-GGGGTACCATGAGAAAAACTAAAATTGTTTGT-3¢ and 5¢-TGCCCTCGAGAAGAACGCTCGCACGGCCTTG-3¢. Using the KpnI and XhoI sites, the PCR product was cloned into pSG1151 (13). Correct clones were identified by restriction digestion and sequencing, and the recombinant vector was transformed into strain SU5 to produce SU699. Single crossover insertion of pyk-gfp was confirmed by PCR. Further characterization of SU699 cells revealed that the Pyk-GFP fusion protein is functional. These cells displayed none of the phenotypes associated with the loss of pyruvate kinase, including no detectable defects in growth, Z ring assembly or positioning (data not shown). Moreover, cells expressing pyk-gfp did not form minicells and exhibited a normal level of tolerance to FtsZ overproduction (data not shown).
Strain SU739 contains a YFP fusion to the E1a subunit of pyruvate dehydrogenase, enocoded by a xylose-inducible pdhA-yfp construct. Importantly, the pdhA-yfp gene was inserted onto the chromosome at the ectopic amyE locus in this strain, leaving the native pdhABCD operon intact. This operon encodes the four component subunits of the pyruvate dehydrogenase complex, which are transcribed from a promoter upstream of pdhA as well as at least one internal promoter (12). We reasoned that the correct localization of PDH E1a may depend on proper regulation of pdhABCD expression, and thus chose to insert pdhA-yfp at an ectopic locus. To construct SU739, the pdhA gene was first amplified using primers 5¢-GCGGTACCATGGCTGCAAAAACGAAAAAAGC-3¢ and 5¢-TGCCCTCGAGCTTCGACTCCTTCTGTGTA-3¢, then cloned into pSG1193 (11) at the KpnI and XhoI sites. Correct clones were identified by restriction digestion and sequencing, and the recombinant vector was transformed into strain SU5 to produce SU739. Double crossover integration of the Pxyl-pdhA-yfp construct at amyE was confirmed by PCR. Interestingly, we found that PDH E1a-YFP is able to respond to changes in the level of its substrate pyruvate (see main text), which suggests that the fusion is at least partially functional.
A similar approach was employed to create a PDH E1a overproduction strain (SU741) harbouring a second copy of pdhA under xylose-inducible control at the amyE locus. Strain SU741 was constructed as outlined above for SU739 (amyE::Pxyl-pdhA-yfp) except that a different reverse primer was used to amplify the pdhA gene, including its stop codon (5¢-TGCCCTCGAGTTACTTCGACTCCTTCTGTG-3¢).
Finally, a PDH E1a-depletion strain (SU791) was constructed by deleting the native pdhA gene from the chromosome of SU741 (amyE::Pxyl-pdhA), leaving a single copy of pdhA under the inducible Pxyl promoter. To avoid polar effects on downstream genes in the pdhABCD operon, we generated an in-frame pdhA deletion by replacing the first first 649 base pairs of the pdhA open reading frame with the open reading frame of the cat gene (651 base pairs including stop codon) conferring chloramphenicol resistance. DNA flanking the cat insertion was left unchanged, placing the expression of cat (along with the pdhBCD genes) under control of the natural operon promoters. The in-frame pdhA deletion was constructed using a modified version of the PCR joining approach described for strain SU664 (Dpyk; see above). First, the cat open reading frame was amplified from pJH101 (9) using primers 5¢-ATGAACTTTAATAAAATTGATTTAG-3¢ and 5¢-TTATAAAAGCCAGTCATTAGG-3¢. DNA fragments flanking pdhA were then amplified from the B. subtilis chromosome (strain SU5) using primers 5¢-ATTAACAGGATGCATGGGAC-3¢ and 5¢-TCAATTTTATTAAAGTTCATACTAAGTCACCTCTTCCTTTC-3¢ for the upstream fragment (1.0 kb), and 5¢-CTAATGACTGGCTTTTATAACTGCAGCTGAAACAATTGC-3¢ and 5¢-CGGTAAAGCTTCATATGCTC-3¢ for the downstream fragment (1.0 kb). Equal amounts of each PCR product (1 mg) were mixed and subjected to a final joining PCR reaction in the absence of added primers. The PCR mixture was transformed directly into strain SU741, selecting for chloramphenicol resistance to generate SU791. Correct integration of the cat gene by homologous recombination was verified by PCR and sequencing.
References
1. Monahan LG, Robinson A, Harry EJ. 2009. Lateral FtsZ association and the assembly of the cytokinetic Z ring in bacteria. Mol. Microbiol. 74:1004-1017.
2. Zhang S, Haldenwang WG. 2005. Contributions of ATP, GTP, and redox state to nutritional stress activation of the Bacillus subtilis sigma B transcription factor. J. Bacteriol. 187:7554-7560.
3. Eymann C, Homuth G, Scharf C, Hecker M. 2002. Bacillus subtilis functional genomics: global characterization of the stringent response by proteome and transcriptome analysis. J. Bacteriol. 184:2500-2520.
4. Errington J, Daniel RA, Scheffers DJ. 2003. Cytokinesis in bacteria. Microbiol. Mol. Biol. Rev. 67:52-65.
5. Adams DW, Errington J. 2009. Bacterial cell division: assembly, maintenance and disassembly of the Z ring. Nat. Rev. Microbiol. 7:642-653.
6. Fry B, Zhu T, Domach MM, Koepsel RR, Phalakornkule C, Ataai MM. 2000. Characterization of growth and acid formation in a Bacillus subtilis pyruvate kinase mutant. Appl. Environ. Microbiol. 66:4045-4049.
7. Pan Z, Zhu T, Domagalski N, Khan S, Koepsel RR, Domach MM, Ataai MM. 2006. Regulating expression of pyruvate kinase in Bacillus subtilis for control of growth rate and formation of acidic byproducts. Biotechnol. Prog. 22:1451-1455.
8. Fabret C, Ehrlich SD, Noirot P. 2002. A new mutation delivery system for genome-scale approaches in Bacillus subtilis. Mol. Microbiol. 46:25-36.
9. Ferrari FA, Nguyen A, Lang D, Hoch JA. 1983. Construction and properties of an integrable plasmid for Bacillus subtilis. J. Bacteriol. 154:1513-1515.