Wall turnover was studied in Bacillus subtilis. The loss of radioactively labeled wall polymers was followed during exponential growth in batch and chemostat cultures. Turnover kinetics were identical under all growth conditions; pulse-labeled wall material was lost with first-order kinetics, but only after exponential growth for 1 generation time after its incorporation. Similarly, continuously labeled cells showed an accelerating decrease in wall-bound radioactivity starting immediately after removal of the labeled precursor and also reached first-order kinetics after 1 generation time. A mathematical description was derived for these turnover kinetics, which embraced the concept of "spreading" of old wall chains (H. M. Pooley, J. Bacteriol. 125:1127-1138, 1976). Using this description, we were able to calculate from our experimental data the rate of loss of wall polymers from cells and the fraction of the wall which was sensitive to turnover. We found that about 20% of the wall was lost per generation time and that this loss was affected by turnover activity located in the outer 20 to 45% of the wall; rather large variations were found with both quantities and also between duplicate cultures. These parameters were quite independent of the growth rate (the specific growth rate varied from 1.3 h-1 in broth cultures to 0.2 to 0.3 h-1 in chemostat cultures) and of the nature of the anionic polymer in the wall (which was teichoic acid in cultures with an excess of phosphate and teichuronic acid in phosphate-limited chemostat cultures). Some implications of the observed wall turnover kinetics for models of wall growth in B. subtilis are discussed.

The walls of Bacillus subtilis var. niger WM, grown in a Mg2+-limited chemostat culture (carbon source glucose, dilution rate = 0.2 h-', 37 "C, pH 7) contained 45 (w/w) teichoic acid, a polymer composed of glycerol, phosphate and glucose in the molar ratio 1.00 : 1.00 : 0.88, respectively.Alkaline hydrolysis of this teichoic acid yielded I-U-b-glucosylglycerol phosphate (together with small amounts of glycerol phosphate) and 13C nuclear magnetic resonance spectra of this hydrolysis product, and its derivative after alkaline phosphatase treatment, confirmed that the monomeric unit was l-O-~-glucosylglycerol-3-phosphate. Assignment of the resonances in the spectrum of undegraded teichoic acid revealed that the polymer was a poly [(2,3)glycerol phosphate], glucosidically substituted on C-1 of glycerol with /?-glucose.The walls of Bacillus subtilis var. niger contain teichoic acid under conditions of chemostat culture where phosphorus is not the growth-limiting nutrient [l, 21. However, the precise structure of this polymer is not known with certainty and therefore, in order to study its biosynthesis, it is first necessary to determine its exact composition. Previous investigations have shown that B. subtilis var. niger produces a teichoic acid built up from glycerol, phosphate and glucose [l], but no conclusive evidence has been presented about the sequence of the monomcrs in this polymer [3]; because of the complexity of the chemical techniques generally used, we looked for a more direct and unequivocal method for the sequence determination.13C nuclear magnetic resonance (I3C NMR) has scarcely been used for structural analysis of bacterial wall polymers, although Bundle et al. [4] have shown the value of this technique in elucidating the structure of 2-acetamino-2-deoxy-~-glucose phosphate polymers from Neisseriu meningitidis and Staphylococcus lactis. However, no reports have been published, to the best of our knowledge, on the application of 13C NMR techniques to teichoic acid analysis.In this paper we describe the analysis of the structure of the monomeric units of teichoic acid from Ahhrr~iarion. NMR, nuclear magnetic resonance.B. suhtilis var. niger WM by an alkaline hydrolysis procedure [5,6], and the use of 13C NMR as an analytical tool for the determination of the sequence of the monomers in the undegraded teichoic acid. MATERIALS AND METHODS Cultivation of BacteriaBacillus subtilis var. niger WM (a spontaneously occurring mutant from B. suhtilis var. niger which forms white instead of reddish colonies on peptone agar + 1 %glucose) was cultured in a 1-1 (LH-Engineering) chemostat with the dilution rate set to a value of 0.2 h-', the pH value controlled at 7.0 and the temperature regulated at 37 "C. The MgZ+-limited medium was made up essentially as described by Evans et al. [7], except that the overall concentration of each nutrient was three-quarters of that specified. Glucose was provided as the carbon source and was added to a final concentration of 22.5 g/l. The outflow tube of the chemostat was connec...

S U M M A R YProtoplasts of a strain of Pythiuii? (~~~2 1 4 2 ) could be obtained by incubating the mycelium with a combination of helicase and cellulase in the presence of an osmotic stabilizer. This process was inhibited when organic compounds were used as osmotic stabilizers but stimulated by pretreatment of the mycelium with a detergent. The most effective detergent was Triton X 100. Regeneration of protoplasts could be demonstrated in liquid or in solid media, but less than 30:/; of the protoplasts were able to grow into new mycelium. In contrast to protoplast formation, regeneration was inhibited by the inorganic salts used as osmotic stabilizer. The results suggest that the hyphal wall of Pythium is covered with a protective layer of lipid material. I N T R O D U C T I O NThe regeneration of fungal protoplasts is of current interest because of the importance of cell-wall formation in morphogenesis. To study this process in Pythiurn P R L~I Q a convenient method is needed to make a large number of protoplasts.It is possible to make protoplasts of Pj)thium P R L~I~~ with an enzyme system produced by a Streptomyces species, containing cellulase, exo-and endolaminaranase activity (Sietsma, Eveleigh & Haskins, 1969). The disadvantages of this method are that it takes a long incubation time (24 h) to get a sizeable number of protoplasts and that the preparation and purification of the enzyme complex is very tedious and laborious. A method has been found which makes large quantities of protopIasts of Pythium within a period of 4 h by using a combination of commercially available helicase and cellulase. The regeneration of these has been studied. M E T H O D SOrganism and growth conditions. Pythiuin PRL 2142 described by Haskins (1963) was grown in shake culture at 28 "C in a medium containing (g/l): KH,P04, 0.5; K2HP04, 0.5 ; K,S04, 0.5 ; MgC1,. 6H,O, 0.6 ; glucose, I o ; asparagine, I -2 ; and thiamine, 0-00 I .Before sterilization the pH of the medium was adjusted to 6.2, andfcholesterol (0.1 g) added as a 10% (w/v) solution in ethanol. After the culture had grown for a certain time, specified under results, the mycelium was collected by filtration, washed twice with distilled water, and then with an incubation medium containing an osmotic stabilizer dissolved in 0.005 M-acetate buffer, pH 6.2. The following substances were used as osmotic stabilizers: KCI, 0.4 M; NaCl, 0.4 M; NH,Cl, 0.4 M; MgS04, 0.65 M; sorbitol, 0.65 M; mannitol, 0.65 M ; sucrose, 0.65 M. The mycelium was finally suspended in an incubation medium at a concentration of 30 mg organismfml, helicase ( I -2 mglml) and cellulase (2 mgfml) added, and incubated at 30 "C. In some experiments the mycelium was either pretreated or incubated with SH-compounds, chelating agents or detergents as described below.

A study was made to determine whether factors other than the availability of phosphorus were involved in the regulation ofsynthesis ofteichoic and teichuronic acids in Bacillus subtilis subsp. niger WM. First, the nature of the carbon source was varied while the dilution rate was maintained at about 0.3 h-'. Irrespective of whether the carbon source was glucose, glycerol, galactose, or malate, teichoic acid was the main anionic wall polymer whenever phosphorus was present in excess of the growth requirement, and teichuronic acid predominated in the walls of phosphate-limited cells. The effect of growth rate was studied by varying the dilution rate. However, only under phosphate limitation did the wall composition change with the growth rate: walls prepared from cells grown at dilution rates above 0.5 h-1 contained teichoic as well as teichuronic acid, despite the culture still being phosphate limited. The wall content of the cells did not vary with the nature of the growth limitation, but a correlation was observed between the growth rate and wall content. No indications were obtained that the composition of the peptidoglycan of B. subtilis subsp. niger WM was phenotypically variable.

Baci1lu.s stearothermophilus B65 and Bacillus subtilis var. niger. WM both contain teichoic acids in their walls composed of glycerol, phosphate and glucose. The 13C nuclear magnetic resonance spectrum of B. steavotherinophilus teichoic acid showed 13C-31P coupling on the signals from the C-5 and C-6 carbon atoms of the glucose molecule and an a-glucosidic linkage between glucose and the C-1 atom of the glycerol moiety. These data are consistent with a poly[glucosylglycerol phosphate] as the cell-wall teichoic acid in this organism.B. subtilis var. niger WM teichoic acid was oxidized by periodate and incubated in glycine buffer at pH 10.5. This treatment did not significantly increase the phosphoinonoester content (by /,'-elimination of the phosphate groups) of the teichoic acid molecule (7.1 to 9.5 yo), which is in accordance with earlier data derived from 13C nuclear magnetic resonance spectroscopy [De Boer et al. (1976) Eur. J . Biochem. 62, 1-61, that in this organism the glucose is not an integral part of the polymer chain. Similar treatment of B. stearotlzermophilus B65 teichoic acid increased the phosphomonoester content of the preparation from 0.15 to 68.1 "/;;.The chemical structures of teichoic acids from a number of bacterial species are well documented (for review see [I]). However, conflicting evidence has been published for the structures of teichoic acids produced by two strains of Bacillus. Thus, Wicken [2] analysed the structure of teichoic acid from Bacillus stt.arotliermt~I~hilus B65 and found a 2,3-linked poly[glycerol phosphate] with a-glucosyl residues attached to position 1 of the glycerol molecule (Fig.1, I). This observation was reinforced later by Kennedy [3], who showed the synthesis of this polymer by a membrane preparation of the same organism.Although expected to form its structure (presence of a-glucosyl residues with C-3, C-4 and C-6 unsubstituted [4]), the teichoic acid from B. stearothermoplzilus B65 did not react with the lectin, concanavalin A [5]. Therefore the location of the phosphodiester groups in the polymer chain was reinvestigated. From these experiments [5] it was shown that the teichoic acid isolated from the same organism was composed of poly[glucosylglycerol phosphate] as shown in structure I1 (Fig. I); this compound would not be expected to react with concanavalin A.Abbreviation. NMR, nuclear magnetic resonance.The reverse situation has arisen with the teichoic acid from B. .subtilis var. niger. While earlier results, based upon analysis of the products of alkaline hydrolysis, pointed towards a teichoic acid 01' structure TI ( Fig. 1) with P-glucosidic linkages [6], De Boer et al.[7] concluded from the I3C nuclear magnetic resonance (I3C NMR) spectra of the teichoic acid and its alkaline hydrolysis products, that B. suhtilis var. nigrr produccs a teichoic acid with 2,3-glycerolphosphate linkages as in structure I (Fig. 1). In the present communication, we have elaborated further our earlier observations [5,7], by the interpretation of the I3C NMR spectrum of B. s...