The Structure and Occurrence of Pili (Fimbriae) on Pseudomonas aerug inosa

Transcription

1 Journal of General Microbiology (1971), 67, Printed in Great Britain The Structure and Occurrence of Pili (Fimbriae) on Pseudomonas aerug inosa By R. L. WEISS Microbiology Department, Indiana University, Bloomington, Indiana 4740 I, US. A. (Accepted for publication 12 May 1971) S U M M AR Y The structure and occurrence of pili on Pseudomonas aeruginosa were investigated with the electron microscope. Piliation depended on the growth of the organism. The number and length of pili reached a maximum in the logarithmic phase and then declined during the stationary phase. A change in colonial morphology from a dry to a moist colony was correlated with the decline from maximum piliation. The distribution of pili as mono- or bipolar was not a function of the growth phase of the culture. Moreover, the number of pili per cell was not uniform within the culture during the growth cycle. Pili, observed by negative staining, had mean diameters of nm. The wide variation of diameter on individual filament was thought to be due to the flexible nature of the pili. This interpretation was supported by the observation that filaments were capable of extreme coiling and bending. When negatively stained the pili did not appear hollow and seemed to differ from the rigid tube-like structure of type I pili (observed in Escherichia coli) by being flexible and rod-like. INTRODUCTION Filamentous appendages known as pili or fimbriae were first reported on Pseudomonas aeruginosa by Houwink & van Iterson (1950). Pseudomonas pili possess a number of properties which permit them to be easily distinguished from flagella with the electron microscope. Pili do not present the characteristic sinusoidal longitudinal waviness of flagella (Heumann & Marx, 1964), are usually of thinner diameter and often are more numerous (Marx & Heumann, 1962) than flagella. Pili of the genus Pseudomonas may have the same position as flagella or may cover the entire cell surface, depending on the species (Fuerst & Hayward, 1969). Tweedy & Park (I 968) described similarities between filaments observed on Pseudomonas multivorans and common pili (type I) of Escherichia coli (Brinton, 1965). Because most studies of pili have focused attention on organisms in the Enterobacteriaceae, it is often assumed that the occurrence of such pili is common among Gram-negative bacteria. With few exceptions (Bradley, 1965, 1966) since their discovery, pili of P. aeruginosa have largely been ignored. Little was known of Pseudonomas pili until Fuerst & Hayward (1969) examined the occurrence of pili in this genus. In view of the recent interest in the pili possessed by the polar flagellates and the general need for more information about pili, we have pursued studies on their occurrence and morphology in P. aeruginosa. METHODS Bacteria and cultural conditions. A Gram-negative rod, obtained from Veterans Administration Hospital, Long Beach, California, U.S.A., was identified as Pseudomonas aeruginosa on the basis of morphological and physiological characteristics. Bacteria were grown logarith-

2 136 R. L. WEISS mically at 37" in I % (w/v) tryptone (Difco) and 0.5 % (w/v) NaCl to a concentration of 2 to 4 x I O bacteria/ml. ~ Stationary phase bacteria were obtained by growing cultures 40 to 48 h. at 37" without aeration. The media were inoculated with isolated colonies grown overnight at 37" on nutrient agar (Difco) plates. Colony morphology. For studies on colony morphology brain heart infusion agar (Difco) plates were inoculated from an overnight culture and incubated at 37" for 24 to 48 h. Negative staining. Carbon-parlodion coated grids were placed on a drop of culture to allow bacteria to adhere to the carbon membrane. After 5 min. the grids were washed three times with glass distilled water and stained with I % potassium phosphotungstic acid (ph 7.0) or I % uranyl acetate (ph 4.6). Ultrathin sectioning. Exponentially growing bacteria were fixed in veronal-buffered I % osmium tetroxide (ph 6.1), essentially by Ryter & Kellenberger's (I 958) procedure, washed three times in buffer and embedded in Epon 812 by the method of Luft (1961). Sections 60 to 80 nm. thick were cut with a glass knife on a Porter-Blum MT-2 ultramicrotome, picked up on 400 mesh copper grids, poststained with uranyl acetate and Reynolds' (1963) lead citrate, and examined with an Hitachi HU-IIA or HU-IIC electron microscope operated at an accelerating voltage of 50 kv. Measurement of pilus diameters. Cultures used for measurement of pilus diameters were grown 20 to 22 h. in tryptone medium at 37". Diameters were measured in negatively stained preparations made with both phosphotungstic acid (PTA) and uranyl acetate. Measurements were made directly in the centre of the electron-image plates in an area of about 25 cm2. by means of a Nikon Shadowgraph projector, and checked with a carbon-grating replica (2160 lines/mm.). RESULTS Colony morphology. Brinton (1965) found that piliated cells of Escherichia coli formed smaller and smoother colonies than non-piliated cells. We found that while Pseudomonas aeruginosa was always piliated, the development of a colony could be divided into two stages differing in degree of piliation and separated by the production of a fluorescent pigment. These stages corresponded to logarithmic and stationary phase growth. In the first stage the bacteria formed a dry, granular colony which was compact and quite adhesive. The bacteria in the colony remained firmly connected; they could not be dispersed by washing or subsequent treatment. The onset of pigment production marked a gradual loss of the dry character and adhesive nature of the colony which gradually became glossy or moist with a smooth surface. The firm attachment between the bacteria was lacking in the moist colony, and the cells were easily separated by suspending them in water. Numbers of pili per cell. Negatively stained preparations of the exponential and stationary phases of growth were examined to determine the numbers of pili per bacterium in broth cultures. Cultures on solid media were not examined because of the difficulty encountered in separating organisms during logarithmic growth. In the logarithmic phase 64 % of the cells were piliated, while in the stationary phase 54 % were. The distribution of the numbers of pili per cell in the exponential and stationary phases is shown in Fig. I. Most of the bacteria growing exponentially had fewer than 10 pili; about 35 % had more than 10 pili; less than 10 % had more than 20 pili. Appearance of green pigment marked the onset of the stationary phase in which about 52 % of the population was piliated, and the number of pili per cell was sharply reduced. Most cells had from I to 5 pili and less than 15 % had more than 10. Only I % of the cells observed in the stationary phase had over 20 pili. The reduction in the pro-

3 The strircture and occiirrence of pili (fimbriae) on Pseudomonas aeruginosa 137 portion of piliated cells and the number of pili per cell was a characteristic feature of stationary phase growth. Shape, position, and appearance of pili. Both polar and bipolar pili were observed on individual cells of Pseudomonas aeruginosa; pili were bipolar in about 60 % of the cells in rm n I Number of pili per celi Fig. I. Number of pili per cell on IOO piliated bacteria. Monopolar and bipolar cells were counted. (a) Logarithmic phase cells with 64 % piiiation. (6) Stationary phase ceils with 54 % piliation. logarithmic phase growth. Polar pili were observed with a frequency of 25 % on the flagellated pole and I 5 % on the non-flagellated pole. In the stationary phase pili on the flagellated pole, the non-flagellated pole or on both poles were observed with about equal frequency. Pili were usually seen to extend away from the cell for a distance of I to 2 pm. However, numerous filaments several microns long were observed in some preparations. The longest fibres measured about 10 pm. Because numerous rigidly straight, long filaments located near the flagellum might interfere with flagellar motion, these filaments were probably flexible rather than rigidly straight. In negatively stained preparations of cells in the logarithmic phase the pili could be seen to extend away from the cell as rather irregular filaments which sometimes underwent extreme bending or curving without breakage (Fig. 2a). Because of their flexibility pili could also spread out from the pole in an irregular pattern as shown in Fig. 2 b. Thus the flexible nature of these pili permits both straight and unusually curved fibres to be observed. All the fibres observed were irregular in appearance and seemed to lack the rigidity commonly associated with type I pili. Heumann & Marx (1964) described cells of P. echimides which cohere in a rosette or star-like cluster by means of monopolar pili. In P. aeruginosa small groups of pili (Fig. 2) could attach to other cells, producing networks of cells rather than a rosette because of the bipolar distribution of pili. One such network was observed to contain bundles of pili (arrows, Fig. 3a) as well as unattached irregular fibres. Size of pili. Previous measurements of Pseudomonas pili by negative staining have relied on PTA as a stain. Pili stained by this method seem to yield similar measurements to those stained with uranyl acetate. Measurements are shown in Fig. 4. Individual pili

4 R. L. WEISS Fig. 2. (a) Electron micrograph of a cell showing the flexible nature of a cluster of pili formed into a coil. Negatively stained with uranyl acetate. (6) Electron micrograph showing several pili spread out in an irregular pattern from the polar surface of the cell. Negatively stained with uranyl acetate.

5 The structure and occurrence of pili (jimbriae) on Pseudomonas aeruginosa I 39 Fig. 3. (a) Electron micrograph of late logarithmic cells showing a dense network of pili (arrows) formed between cells. Negatively stained with uranyl acetate. (b) Electron micrograph of a cell showing the rod-like nature of the filaments. Note that the pili present an irregular structure. Negatively stained with phosphotungstate. I0 MIC 67

6 140 R. L. WEISS varied in diameter along their length. Some varied by 7.0 nm., others only by 1.0 nm. A representative distribution of ranges (Fig. 2) clearly shows that some pili were narrower than others. However, statistical analysis by the t-test shows that this difference was not significant. The mean diameter measurements for IOO pili (six measurements per pilus) showed a normal distribution with a range from about 4-0 to 8.0 nm. This organism appeared to have a single type of pilus on the basis of size with a mean measured diameter of 6.0 & 2.8 nm. - I I I I I I I I I I I I Diameter (nm.) Fig. 4. Range of diameters of pili formed by Pseudomonas aeruginosa. Each thick line represents the diameter range of an individual pilus. The thin line represents the mean diameter range for IOO pili. Structure of pili. Negative contrast preparations of pili stained with PTA are shown in Fig. 3 b. The structure appeared rod-like. The negative stain seemednot to have penetrated the pilus in the same way as it does with peritrichously distributed Pseudomonas pili (Tweedy & Park 1968) or type I pili (Brinton, 1965). The pili showed a definite irregularity along the length of individual fibres. When negatively stained with uranyl acetate (Fig. 5a) the pili appeared filamentous with uneven width (Fig. 2 b; arrows, Fig. 5 a). This irregularity was interpreted to mean that the pili were not uniformly round along their length and might have been oval with a rod-like construction. This was judged from the observation that locations along the pilus that appeared wider adjoined narrow areas. This might have resulted from a twist of an oval filament as it was adhering to the grid. Staining deposits along the filament and the nearby flagellum appeared to be uniform and there was an apparent flagellar substructure which consists of dense granules arranged in parallel rows. Thin sections of pili. Thin sections of cells in the logarithmic phase are shown in Fig. 5 by c. Several thin filaments may be observed. Fig. 5 b shows a number which are sharply curved. In Fig. 5 c the thin filaments extend from the pole of a tangentially sectioned cell. The filaments probably represent the pili observed by negative staining of whole cells. Diameters of the thin-sectioned filaments ranged from 7.5 to 10.5 nm. and were only slightly above the diameter of negatively stained pili. The diameters of the sectioned filaments were well below the values for negatively stained flagella. The sectioned filaments were numerous, as one might have predicted- from observations of negatively stained cells.

7 The structure and occurrence of pili (Jimbriae) on Pseudomonas aeruginosa I 4 I Fig. 5. (a) Electron micrograph of logarithmic cells showing locations along the pilus axis (arrows), where the diameter is not uniform. In each case a narrow area is adjacent to an area with a wider diameter. Negatively stained with uranyl acetate. (6) Electron micrograph showing thin-sectioned filaments which are sharply curved. The attachment of filaments to the cell is not shown. (c) Electron micrograph showing thin-sectioned filaments which appear to extend from the pole of the cell. 10-2

8 142 R. L. WEISS DISCUSSION Most pseudomonads grown from stock cultures show a variety of colony types. In the strain studied here, a change in colony morphology from a dry to a moist surface was correlated with a decrease in the proportion and number of pili per cell during the transition from the logarithmic to the stationary phase of growth. Possibly the dry colony associated with rich piliation of the logarithmic phase reflected restricted motility, while the moist colony correlated with reduced piliation in the stationary phase denoted unrestricted movement of bacteria over the surface of the colony. Some preliminary results suggest that this correlation may be fairly common among clinical isolates. The outgrowth of pili in the culture is related to the stage of growth of the organism. Both the proportion of piliated cells and the number and length of pili were greatest during the logarithmic phase. More than 30 pili were observed on a single cell with individual filaments in excess of 10 pm. Fuerst & Hayward (1969) stated that they observed a maximum of 10 pili per cell on Pseudomonas aeruginosa. They report the maximum length of filaments to be 1-7pm. However, their growth medium, temperature and time of incubation were different from those used in the present study. The distribution of pili in the polar position was similar to that observed by previous workers (Fuerst & Hayward, 1969; Houwink & van Iterson, 1950). In stationary phase cultures the proportion of cells with bipolar pili decreased to about half the value observed in the logarithmic phase. A partial constriction indicative of septum formation was observed on many cells with bipolar pili. This was interpreted to mean that the distribution of pili on one or the other pole of the daughter cells is related to cellular division. If the new daughter cells are to have polar pili the cell about to divide must have bipolar pili. Measurements of pilus diameters have usually relied on negative staining. Bradley (I 966) reported that pili of Pseudomonas aeruginosa observed by negative staining of cells whose pili were marked by infection with ribonucleic acid bacteriophage had a diameter of 4-5 nm. Fuerst & Hayward (1969) state that the diameter of pili measured on negative contrast preparations of two strains of P. ueruginosa was 5-4 t I -4 nm. and 5.2 k 1-4 nm. In the present study the range of measured diameters was greater than that expected according to the standard deviation presented in measurements of Fuerst & Hayward (1969) for pili of P. aeruginosa. However, the mean diameter of 6.0 k 2.8 nm. was similar to these previous measurements. Polar pili of Pseudomonas aeruginosa appear to be unique in shape and morphology. The suggestion that polar pili of Pseudonomas species are flexible rather than straight (Fuerst & Hayward, 1969) has been confirmed, but flexible fibres have not yet been observed in other Pseudomonas species. Flexibility in polar pili would be of great benefit to the pseudomonads in general, since large numbers of such filaments located on the flagellated pole of the bacterium would not be likely to interfere with bacterial flagellar motion. The relation between motility and the presence of pili needs to be studied further. Bradley (1966) observed that pili on P. aeruginosa which acted as RNA phage receptors appeared to be hollow in places. These pili were distinct from type I pili because no helical structure was observed, Fuerst & Hayward (1969) found a helical structure and central hollow core in pili of P. multivorans which were similar to type I pili, but did not find these features in pili observed on p. aeruginosa. We observed that in negatively stained preparations the stain did not appear to penetrate the pili of P. aeruginosa. Thin sections of P. multivorans with peritrichous pili did not show detailed substructure or the hollow core observed by negative staining (Fuerst & Hayward, 1969). Thin filaments observed in sectioned cells of P. ueruginosa with polar pili

9 The structure and occurrence of pili (jimbriae) on Pseudomonas aeruginosa 143 appeared rod-like, but did not differ substantially from filaments observed in sections of P. multivorans. Like Fuerst & Hayward, we were unable to observe the penetration of filaments into the cell. Thin sections of the organism show filaments with a smooth outer surface and a rod-like structure. Negatively stained pili also appear rod-like. I have not seen the hollow core typical of type I pili. I envisage the pili of P. aeruginosa as consisting of a flexible rod-like structure with an elliptical or oval cross-section. This investigation was supported in part by USPHS grant no. TI-GM-503, National Institutes of Health. I wish to thank Dr Charles C. Brinton, Jun. for helpful suggestions during part of this study. REFERENCES BRADLEY, D. E. (1965). The morphology and physiology of bacteriophages as revealed by the electron microscope. Journal of the Ropl Microscopical Society 84, BRADLEY, D. E. (1966). The structure and infective process of a Pseudomonas aeruginosa bacteriophage containing ribonucleic acid. Journal o f General Microbiology 45, BRINTON, C. C., JUN. (1965). The structure, function, synthesis and genetic control of bacterial pili and a molecular model for DNA and RNA transport in Gram-negative bacteria. Transactions of the New York Academy of Sciences 21, FUERST, J. A. & HAYWARD, A. C. (1969). Surface appendages similar to fimbriae (pili) on Pseudomonas species. Journal of General Microbiology 58, HEUMANN, W.& MARX, R. (1964). Feinstruktur und Funktion der Fimbrien bei dem sternbildenden Bakterium Pseudomonas echinoides. Archiv fur Mikrobiologie 47, HOUWINK, A. L. & VAN ITERPON, W. (1950). Electron microscopal observations on bacterial cytology. 11. A study on flagellation, Biochimica et biophysica acta 5, LUFT, J. H. (1961). Improvements in epoxy resin embedding methods. JourHal of Biophysical and Biochemical CYtO~ogY 9, MARX, R. & HEUMANN, W. (1962). Uber Geisselfeinstrukturen und Fimbrien bei zwei Pseudomonas - Stamen. Archiv fir Mikrobiologie 43, REYNOLDS, E. S. (1963). The use of lead citrate at high ph as an electron-opaque stain in electron microscopy. Journal of Cell Biology 77, RYTER, A. & KELLENBERGER, E. (1958). Gtude au microscope electronique de plasmas contenant de l acide deoxyribonucleique. Zeitschrgt fur Naturforschung 133, TWEEDY, J. M. & PARK, R. W. A. (1968). Evidence for the presence of fimbriae (pili) on Vibrio species. Journal of General Microbiology 51,