Basal Bodies of Bacterial In

This paper investigates further the question of whether the flagella of Proteus mirabilis emerge from basal bodies. The bacteria were grown to the stage of swarmer differentiation, treated lightly with penicillin, and then shocked osmotically. As a result of this treatment, much of the cytoplasmic content and also part of the plasma membrane were removed from the cells. When such fragmented organisms were stained negatively with potassium phosphotungstate, the flagella were found to be anchored--often by means of a hook--in rounded structures approximately 50 m/z wide, thus confirming Part I of our study. In these rounded structures a more brilliant dot was occasionally observed, which we interpret as being part of the basal granule. A prerequisite for the demonstration of the basal granules within the cells was, however, the removal of both the cytoplasm and the plasma membrane from their vicinity. In some experiments, the chondrioids were "stained" positively by the incorporation into them of the reduced product of potassium tellurite. The chondrioids were here observed to be more or less circular areas from which rodlike structures extended. The chondrioids adhered so firmly to the plasma membrane that they were carried away with it during its displacement by osmotic shocking, while the basal bodies were left behind. This observation disproves our previous suggestion that the flagella might terminate in the chondrioids. The basal bodies often occur in pairs, which suggest that they could be selfreproducing particles. I N T R O D U C T I O N All electron microscope observations of the basal bodies of bacterial flagella have so far been made with whole mount preparations of cells, which were either shadowed or stained negatively. No records have, as yet, been published of basal bodies on bacterial flagella observed in thin sections. However, when living bacteria in preparation for ultramicrotomy are first embedded in agar and then fixed, the flagella are well preserved and may subsequently be seen in the section to pierce the cell wall and enter the cytoplasm. In a study in which Proteus mirabilis was prepared according to this adaptation of technique (20), we sometimes observed the flagella to originate from round strut603 on A uust 8, 2017 jcb.rress.org D ow nladed fom tures, approximate ly 25 to 45 m # wide, a t the per iphery of the cytoplasm. Only in organisms which were very poor in thei r cytoplasmic content could the boundar ies of the bodies be dist inguished with sufficient clarity; therefore, i t is desirable tha t fur ther proof of the genuine existence of these bodies be provided by other techniques of investigation. To this end, we applied negat ive staining with potassium phosphotungs ta te to organisms in which m u c h of the cellular content had been released by osmotic shock after the cell wall had been weakened by penicillin. A n earlier investigation of Proteus vulgaris (17, 18) on shadowed prepara t ions of whole cells suggested tha t the flagella might emerge from the chondrioids (cf. footnote, reference 20). The chondrioids themselves can readily be made visible by incuba t ing the bacter ia with ei ther potassium tellurite (17) or te t rani t ro-blue te t razol ium (23, 24). Reduced products of these compounds are deposited in ra ther restricted cytoplasmic areas contiguous wi th the p lasma membrane . Such areas in Proteus do not have the m e m b r a n o u s configurat ions characteris t ic of chondrioids in m a n y Grampositive bac ter ia (for references, see 18). I n order to reinvestigate the relat ionship ot the basal bodies of flagella to the chondrioids in Proteus mirabilis and to analyze fur ther the s t ructure of the chondrioids themselves, some cells whose walls had a l ready been loosened by penicillin were incuba ted wi th potassium tellurite. They were then shocked osmotically and stained negatively wi th potassium phosphotungstate . The resuits of these experiments do not confirm the previous suggestion t ha t the chondrioids function as the basal bodies of Proteus flagella. Finally, some observations are included on the structure of free flagella which had been released from autolyzed organisms. M A T E R I A L S A N D M E T H O D S In this work, two strains of Proteus mirabilis were used : one was obtained from Dr. E. Klieneberger-Nobel (cf. reference 32) ; the other from the Central Health Laboratories, Toronto, Canada (10-12). The bacteria were grown out overnight, usually by passing them through a motility tube of semisolid agar (heart infusion broth + 0.3% agar), then suspended in physiological saline to a concentration of about 10 ~ bacteria/ml. One ml of this solution was spread uniformly over the surface of heart infusion agar (Difco) in a 1 5 ~ m diamctcr Petri dish; the plates were incubated for 31/~ to 5 hr at 35°C, i.e. to the differentiation of swarmers as checked with the light microscope (11). The bacteria were removed from the plates in heart infusion broth to which was subsequently added 2,000 I U / m l penicillin, 0.25 M sucrose, 0.01 M MgCI~, and 5% horse serum. Such suspensions were incubated for 45 to 60 rain, i.e. until many of the bacteria had begun forming spheroplasts when checked under phase-contrast. The suspension was then divided into two parts: the first was centrifuged directly at 4,000 RPM and the resulting pellet shocked osmotically by diluting the organisms in distilled water; the second part was incubated under semianaerobic conditions for an additional hour with 0.05% potassium tellurite in stationary tubes filled to the br im and stoppered, then centrifuged, and shocked. In a few experiments, the bacteria were removed from the plates of heart infusion agar in distilled water and allowed to autolyze by being kept overnight either in the refrigerator, for partial breakdown of the organisms, or in the 35°C incubator. The latter treatment was applied in the hope of completely freeing the flagella and their basal structures. In all cases, the organisms--either whole or fragm e n t e d w e r e stained by mixing the suspension with an equal volume of 2% (w/v) potassium phosphotungstate (PTA) (41). The stained preparations were placed on electron microscope grids, coated with either carbon alone or Formvar reinforced with carbon, by breaking a thin film in a plat inum loop onto the surface of the supporting membrane. The grid itself rested on a pad of filter paper so that excess fluid drained away as the film of bacteria dried down rapidly. Electron micrographs were taken with a Philips EM 200 operating at 60 or 80 kv with the double condensor lens system, a 50-~ objective aperture, and the specimen cooling device. O B S E R V A T I O N S Flagella and Basal Bodies W h e n cells of Proteus mirabilis are shocked osmotical ly in distilled water after t r ea tmen t with penicillin, m u c h of thei r content is released; yet m a n y of the bacter ia re ta in thei r shape and flatten on the support ing film as the potassium phosphotungstate drics down. A n interest ing example of such a cell is provided in Fig. 1 : ha rd ly any cytoplasm is left wi th in the cell wall, apar t from remnan t s of the plasma m e m b r a n e (Pro) and a n u m b e r of e lec t ront ransparent bodies, ca. 50 m/~ wide, at the bases of the flagella. Occasionally, two such bodies arc in close contact (see arrows), each bear ing a flagellum (cf. also reference 2). Of interest are the several hoodlike folds (H) tha t 604 THE JOURNAL OF CELL BIOLOGY • VOLUME 81, 1966 on A uust 8, 2017 jcb.rress.org D ow nladed fom