Purification and properties of a protein factor stimulating peptidoglycan synthesis in toluene- and Licl-treated Bacillus megaterium cells.

A protein factor, called PG-I, can be solubilized from toluene-treated Bacillus megaterium cells by LiCl extraction. After LiCl extraction, peptidoglycan synthesis by the toluene-treated cells is decreased. Protein PG-I can be added back to the extracted cells to stimulate peptidoglycan synthesis. This factor has now been purified 124-fold. It has a molecular weight of 42,000 as estimated by Sephadex gel filtration in the presence of 0.4 M KCl and 52,000 as determined by sodium dodecyl sulfate disc gel electrophoresis. Periodate-Schiff staining of the polyacrylamide gel indicates that factor PG-I is a glycoprotein. The reconstitution of LiCl-extracted cells requires Mg2+ with an apparent Km of 1.9 X 10(-3) M. The Mg2+ ions can be replaced by Ca2+ and by Mn2+ ions to some extent; Zn2+ and Cu2+ ions had no effect. The available data suggest that factor PG-I is essential for peptidoglycan synthesis and requires at least one thiol group for stimulatory activity.

1% Triton X-100 and 1 rnM dithiothreitol, for 12 hours with two changes of buffer. The dialyzed protein, after removal of insoluble materials by centrifugation at 48,000 x g was applied to a hydroxylapatite column (3.7 x 6 cm) which had been equilibrated with the same buffer. The column was washed with the same buffer without the Triton X-100 until the A,,, had decreased to less than 0.15 (320 ml). The column was further washed with 320 ml of 0.2 M sodium phosphate buffer, pH 6.8, containing 1 mM dithiothreitol. Then, a linear gradient from 0 to 0.25 M ammonium sulfate in 0.2 M sodium phosphate buffer, pH 6.8, containing 1 'mM dithiothreitol was used to elute the column. The total gradient was 1000 ml, and fractions of 13.7 ml were collected at a flow rate of 30 ml/hour.
The factor PG-I activity in 5 ~1 of each fraction was determined.  gave a periodate-Schiff positive band on sodium dodecyl apparent K, value for the Mg2+ needed for the reconstitution sulfate-polyacrylamide gel electrophoresis. This band was coin-  Elution was carried out using the same buffer, and fractions of 2.5 ml were collected at a flow rate of 16 ml/hour. B, 10% polyacrylamide gels were run for 4 hours at 8 mA per tube in 0.1% sodium dodecyl sulfate as described under "Experimental Procedure." The relative mobility for each protein including factor PG-I (0) is the average value from 5 separate gels. The marker proteins (0) employed were bovine serum albumin (68,000), ovalbumin (43,660), bovine pancreas carboxypeptidase A (34,000), and egg white lysoxyme (14,300). cident with the major band of Coomassie blue-stained protein (Fig. 5). These data indicated that protein PG-I is probably a glycoprotein and could explain the discrepancy in the molecular weights as determined by Sephadex gel filtration and sodium dodecyl sulfate gel analysis. If factor PG-I contained carbohydrates which could bind sodium dodecyl sulfate ions  LiCI-extracted cells (9 x 10') were kept at -20" for 3 months and then were reconstituted with 0.5 pg of purified factor PG-I. The cells were assayed both for peptidoglycan synthesis and for the incorporation of free diaminopimelate into acid-insoluble material (2). The assay conditions were -as described under "Experimental Procedure" except that [aH]diaminopimelate (0.25 mM, 185,000 cpm per assay) was included in the reaction mixtures of Experiment B. of diaminopimelate into previously formed wall. In addition, the ability of factor PG-I to stimulate diaminopimelate incorporation in the presence of peptidoglycan synthetic precursors can be explained by the LiCl-extracted cells retaining enzyme(s) capable of incorporating diaminopimelate into newly formed peptidoglycan. As the synthesis of peptidoglycan is increased by addition of factor PG-I, more wall substrate is made available for the diaminopimelate-incorporating enzyme(s) responsible for activity I. These enzyme(s) might well be transpeptidases.

Peptidoglycan
synthesis can be stimulated in toluene-and LiCl-treated B. megaterium cells by the addition of the glycoprotein factor PG-I with a molecular weight of 42,000 as measured by Sephadex gel filtration.
We have purified this factor 124fold to one major protein band on sodium dodecyl sulfate-polyacrylamide disc gel electrophoresis. At no stage during the purification procedure did the factor behave like more than one protein. Thus we conclude that a LiCl-extract of B. megaterium cells contains only one protein responsible for the stimulation of peptidoglycan synthesis in LiCl-treated cells. Furthermore, the LiCl-treated cells lost essentially all peptidoglycan synthetic activity upon prolonged storage in the freezer and this activity could be regained after reconstitution with purified factor. Hence, the PG.1 protein seems to play an essential role in peptidoglycan synthesis. The other enzymes involved in the synthesis must have remained relatively active during storage. The precise role of factor PG-I in peptidoglycan synthesis is as yet unknown, but is now under investigation.
Bacterial peptidoglycan synthesis by cytoplasmic membranes reauires the presence of magnesium ions. The optimal concentration of magnesium ions required for synthesis is about 10 to 20 mM in both membrane preparations and toluene-treated cells of B. megaterium (1,12). This coincides with the optimal concentration for reconstitution of LiClextracted cells with factor PG-I. When LiCl-extracted cells were preincubated with magnesium ions alone, no change in peptidoglycan synthetic activity was observed (Fig. 6), indicating that magnesium is essential for the reattachment of factor PG-I to the cells as well as for peptidoglycan synthesis. These observations are like those with the reattachment of the membrane ATPases of Escherichiu coli (13) and Staphylococcus fuecalis (14), which is also known to be facilitated by magnesium ions.
It had been found previously that factor PG-I could not bind to purified walls of B. meguterium, but could attach to membranes prepared from the same cells by sonic disruption (2). In our present experiments, LiCl-treatment seems to be able to remove factor PG-I from membranes isolated from cells opened by grinding with alumina, because peptidoglycan synthesis is reduced after the extraction. The synthesis was recovered after reconstitution with purified factor in the presence of magnesium ions. These observations suggest that the stimulatory factor is normally associated with the cell membrane and can be released and reattached to the membrane experimentally.
Because factor PG-I participates in a complex series of reactions involved in wall biosynthesis, it would be reasonable for the factor to be attached close to other enzymes involved in peptidoglycan synthesis. The requirement of active sulfhydryl group(s) in factor PG-I might mean that these group(s) might be essential for the activity of the protein.
Factor PG-I can stimulate the incorporation of diaminopimelate, which is dependent on simultaneous peptidoglycan synthesis. However, as discussed above, the phenomenon can be explained solely on the basis of factor PG-I stimulating peptidoglycan synthesis. Therefore, we have no evidence for this factor participating in any reactions other than those involved in peptidoglycan synthesis.
The involvement of factor PG-I in peptidoglycan synthesis in bacteria other than B. meguterium is explored in experiments where membranes of other bacteria were reconstituted with purified factor. A very slight but reproducible stimulation was observed, but the level was too low to be as yet considered significant. Additional experiments will be required to determine whether this factor is actually involved.
In our experiments characterizing factor PG-I, we used material which still showed minor contaminating bands on polyacrylamide gel electrophoresis. Removal of these materials by Sephadex G-75 fractionation permitted us to assign the activity to the major band and to determine the molecular weight by gel electrophoresis. The contamination probably did not seriously affect our other characterization experiments because there we were examining functional and not physical characteristics.
It seems unlikely that the minor band could contain the PG-I activity, because the activity remains after the material in the small band is removed by Sephadex G-75 chromatography.