Comparison of the carbohydrate composition of the cell walls of strains of Streptomyces griseus.

In our previous experiments (Arch. Microbiol. 40:261, 1951), no correlation was found between any of the morphological or physiological characteristics of various Streptomyces griseus strains and their streptomycin-producing capacity. There was but one common feature of those strains which did not produce streptomycin: they all had a short life cycle, which was also characterized by the early appearance of short fragments of hyphae after a brief vegetative mycelial growth. This behavior was reminiscent of the development of the Nocardia. Avery and Blank (Can. J. Microbiol. 1:140, 1954), Cummins and Harris (J. Gen. Microbiol. 15:9, 1956), and Romano and Sohler (J. Bacteriol. 72:865, 1956), studying the composition of isolated cell walls of Streptomyces and Nocardia strains, found differences between the genera and declared that the chemical composition of the cell wall was of taxonomic importance. Streptomyces strains contained hexose, whereas the cell walls of Nocardia had pentose in addition. However, Sohler, Romano, and Nickerson (J. Bacteriol. 75:283, 1958) later found pentose in the cell wall of some Streptomyces strains. Those strains which were lysed by lysozyme contained only hexoses, whereas those Streptomyces which were not lysed contained both pentose and hexose. We have studied the cell-wall composition of an S. griseus strain used industrially for the production of streptomycin (strain Ls-1; no. 52-1, according to the code number of our laboratory) and a stable mutant strain (no. 45) which did not produce streptomycin. This was done in the hope that the difference in the time of fragmentation of the hyphae between the two strains would be reflected in the carbohydrate composition of their cell walls. The characteristics of the strains were described earlier (Arch. Microbiol. 40:261, 1961). The strains were cultivated in shake culture in soybean medium for various periods of time. After harvesting, the cells were washed twice with distilled water and disintegrated with an ultrasonic disintegrator (Measuring & Scientific Equipment Ltd., London.). After centrifuging, the cell walls were washed with water, with 0.067 M phosphate buffer (pH 7.6), and then again with water several times. Cell-wall preparations were hydrolyzed with 1 N H2SO4 for 4 hr in boiling water. The method of cell-wall isolation is a slight modification of that used by Salton and Horne (Biochim. et Biophys. Acta 7:177, 1951). Detailed results will be published elsewhere. Hexose and pentose were present in the cell walls in both of our strains. Strain 52-1 contained 17% reducing substance, based on the hexose content determined by the anthrone reaction, and 2.6% pentose, as determined by the orcinol test (Meybaum, modified by Dische, In S. P. Colowick and N. 0. Kaplan [ed.], Methods in Enzymology, vol. 3, p. 88, Academic Press, Inc., New York, 1955). The cell wall of strain 45 had 12.5% hexose and 3.5% pentose. It is clear that the cell walls of both of our S. griseus strains contain pentose. This is in contrast to the finding of Romano and Sohler and in some respects to the data of Sohler, Romano, and Nickerson. Namely, both of our strains were susceptible to lysozyme action (Lysosim, Fluka AG) and, in spite of this. their cell walls contained pentose. We are not able to explain the pentose content by ribonucleic acid impurities because: (i) the pentose content did not decrease after ribonuclease treatment, and (ii) the pentose was proven by paper chromatography to be arabinose. Repetition of Romano and Sohler's experiments by cultivating our S. griseus strains in their media did not change our results concerning the pentose content. We think that the difference between their and our results is to be explained by strain specificity. Microscopically, we found more and less lysozyme-susceptible parts along the hyphae. 1342 NOTES