The capsular network of Klebsiella pneumoniae
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
Itlstitirte qf Microbiology, U t ~ i ~ ~ e r ofRot71e, sity R o m e , Ittrly Accepted January 28, 1977
CASSONE, A., and E. GARACI.1977. The c a p s ~ ~ l anetwork r of Klehsielln pt~errtnonicre. Can. J. Microbiol. 23: a4-689. Attempts at improving chemical fixation for electron-microscopic observation of the capsule of Klehsiellnp1rero71ot1irrewere made. The capsule was preserved by using alcian blue - lanthanum and tris-(I-aziridinyl) phosphine oxide (TAPO) - aldehyde - osmium procedures. Despite the different retention of the overall capsular material and minor variations in morphological details, in both cases the interpretation of ~~ltrastructural patternssuggested that thecapsule be composed of a meshed netwol-k of thin polysaccharide fibrils radiating from the cell wall. This organization is in keeping with all recognized chemical properties of bacterial polysaccharide capsules or, at least, does not contradict them. Moreover, an effective preservation of bacterial structures other than capsule has been obtained. mostly in specimens fixed by the TAPO-aldehyde-osmium method, a fact which gives further reliability to the technical approach used for capsule visualization. CASSONE, A,. et E. GARACI.1977. The capsular network of Klel~siellrrptrelrmonicre. Can. J . Microbiol. 23: 684-689. Des experiences ont ete conduites en vue d'amkliorer la fixation chimique pour I'examen en microscopie electronique de la capsule de Klrhsielln ptlerrnlot~itre. La capsule est fixke par les methodes utilisant soit I'alcian blue - lanthanum, soit le tris-(I-aziridinyl) phosphine oxide (TAPO) -aldehyde - osmium. Malgre que, par ces methodes, la rttention des divers materiaux capsulaires est differente et que des variations mineures dans les details morphologiques se presentent. I'interpretation des patterns ultrastructuraux suggere que la capsule est constituee d'un reseau en reticules de fines fibrilles polysaccharidiques irradiant de la paroi cellulaire. Cette organisation est en accord avec les pro~rieteschimiques reconnues des capsules polysaccharidiques des bacteries ou, du moins, ne les contredit pas. De plus, une fixation efficace des structures bacteriennes autres que les capsules est obtenue, principalement pour les specimens fixes par la methode TAPO-aldChyde-osmium, ce qui conftre plus de fiabilite B I'approche technique utilisie pour I'examen des capsules. [Traduit par le journal]
Because of their chemical properties, bacterial capsules are poorly preserved in electron microscopy. In fact, they are built up of highly hydrated, delicate gels, generally polysaccharide in nature (14), which can be coarsely extracted or collapse during any one of the standard preparative stages for electron microscopy, mostly during dehydration (1 1). Techniques are now available for embedding without dehydration in water-miscible resins (6) or for electron-microscopic observation after freezing procedures (7). The exploitation of these techniques in Klebsiella and in other capsulated bacteria, while furnishing important new information, did not prove, however, to be fully satisfactory since it gave images of difficult and non-uniform interpretation (I 1, 12). In our opinion, it would be advantageous to use conventional procedures for electron-microscopic visualization of bacterial capsule provided
that chemical fixation was improved. In principle, the ideal fixative should cross-link a n d stabilize the capsular components strongly so as to avoid or minimize any damage due to water removal. In this note we describe attempts for a n improvement in chemical fixation of the capsule in Klebsiella pneu~woniae.
Materials and Methods Several strains of Klebsiellnpt~rutnotriae(IMM-Rome 1 to 5) of unidentified serological type were used throughout this study. They were isolated from pathological specimens and maintained in culture on brain heart infusion medium (BBL, Maryland, U.S.A.). At intervals during growth, the microrganisms were prepared for electron microscopy as follows. Slants were directly flooded with the prefixation mixture for the required time, then cells were removed from the slant, washed, and subsequently treated in suspension as described below. Chemical fixation has always been carried out at 4-6'C. T o this purpose, two main procedures have been
C A S S O N E A N D GARACl
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
TABLE1. Fixation techniques
Prefixation components
Glutaraldehyde (3%) TAPO (0.5%)
Primary stain Vehicle for prefixation pH Time (min) Postfixative Post stain Vehicle for postfixation pH Time (h)
Alcian blue (0.5%) Cacodylate buffer 0.05 M
Glutaraldehyde (0.75%) T A P 0 (1%) Acrolein (1%) None Phosphate buffer 0.05 M
6.0 30 OsOa (1%) Lanthanum nitrate (1%) S-collidine buffer 0.05 M
7.2 30 o s o , (4%) None Hz0
used. The first one, conventionally named A, was characterized by the use of a polycationic dye, alcian blue, and a lanthanum salt in a slight modification of the technique described by Shea (10). Accordingly, the slant was prefixed for 30 min with a mixture of glutaraldehyde, TAPO (Tris-(1-aziridinyl) phosphine oxide; Polysciences, Pennsylvania, U.S.A.), and alcian blue (8GX, Allied Chem., NJ, U.S.A.), 3% (v/v), 0.5% (v/v), and 0.5% (w/v) respectively, in cacodylate buffer, 0.05 M, pH 6.0, and subsequently postfixed for 6 h with a mixture of OsO4 and lanthanum nitrate, both 1% (w/v) in S-collidine buffer, 0.05 M, pH 7.35. The second type of fixation, named B, characteristically involved the use of TAPO and aq~leousosmium, coupled with aldehyde fixative (1, 4). The slant was prefixed for 30 min with a mixture of glutaraldehyde, TAPO, and acrolein, 0.75%, 1%, and 1% (v/v) respectively, in phosphate buffer, 0.05 M, pH 7.2, and postfixed in unbuffered Os04, 4% (w/v) in HzO, pH 5.8, for 12 h. The two fixation techniques are conlparatively sunimarized in Table 1. All fixed specimens were carefully washed in S-collidine buffer and stained overnight at 4'C with uranyl acetate, 0.5% (w/v), in Verona1 acetate buffer, 0.05 M, pH 5.1. Subsequently they were dehydrated and infiltrated (in ethanol-graded series and propylene oxide, at 4"C), then embedded in Spurr low-viscosity resin (13), according to conventional procedures. Thin sections w h ~ c hhad been cut at the Porter-Blum MT-2 ultramicrotome were stained with lead citrate (8) and alcoholic uranyl acetate and observed with a Siemens 1A electron microscope operating at 60-80 kV.
Results After fixation of Kleb.~iellaptze~rmoniaewith method A, the capsule was easily preserved. It was seen as a medium-density, compact structure directly laid over the wall with a regular and continuous profile all around the cell (Fig. I). It didn't show any coarse distortion and its general ultrastructural aspect was that of a fibrous or
fibrogranular net measuring about 120 nm in thickness (Figs. 1, 2). The cytoplasmic matrix and the cell wall appeared sufficiently resolved as compared to other conventional fixation methods, while the nucleoid was slightly swollen and coagulated (Fig. 2). A more distinct resolution of capsular components has been achieved using the fixation method B whereby the capsule was composed of very thin (25-30 A the best resolved ones), lowdensity, and intermingling fibrils radiating from the cell wall (Figs. 3, 4). The capsular average thickness now measures about 55 nm, a value approximately half that measured in A-fixed specimens, suggesting a possible extraction of a marked amount of capsular material. Accordingly, a few individual fibers appear t o be longer than the average fiber (Fig. 3, arrows). As shown by method A as well, the amount of capsular material varied according t o the cell age, being thickest as the culture approaches the stationary phase of growth and thinnest in young and dividing cells especially at the constriction site (Fig. 4). (The distinct phases of growth, however, were not recorded, so this observation, though closely paralleling some biochemical findings (14), needs further experiments). The preservation of the other cell structures following the B method of fixation could be considered satisfactory. The nuclear component was clearly seen as well-resolved and contrasted fibrils, like in thestandard Kellenberger procedure (9), while the cell wall showed distinctly both the outer wall membrane and the inner peptido-
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
CAN. J. MICROBIOL. VOL. 23, 1977
NOTE:The length of marker bars correspond t o 200 nm (Fig. I), 100 nm (Figs. 2, 3, 4), and 50 nm (Fig. 5). cn: capsule; CW, cell wall; CM, cytoplasmic membrane; NU, nucleoid; WM, wall membrane; R, peptidoglycan layer. FIGS1, 2. Klebsielln pneumoniae, strain IMM-Rome 1 and 3, respectively, fixed by method A (for details, see text).
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
CASSONE AND GARACI
FIGS3,4, 5. Klebsiellapnerrmo~riae,strains IMM-Rome 1 (Fig. 3) and IMM-Rome 4 (Figs 4, 5) fixed by method B (see text). Simple arrows in Fig. 3 point to the capsule while head arrows point to longer fibrils. Simple arrows in Fig. 4 point to the constriction area of cell separation where scanty capsular material is seen.
687
688
CAN. J. MICRO BIOL. VOL. 23. 1977
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
glycan layer (Fig. 5). N o artifactual discontinuities were seen in the capsular material nor was the capsule separated from the underlying cell wall as frequently happens in poorly fixed specimens (5).
Discussion In a critical evaluation of these results, it is pertinent to consider that the two procedures of chemical fixation, significantly different from each other, used during this investigation, gave a 'comparable' visualization of the bulk of the capsule which essentially consisted of a network of interconnected fibers evenly distributed around the cell. It is also clear that some degree of 'intermingling' among capsular fibers, as visualized here, could be artifactual resulting in some coarse floccularity of the capsule itself. Indeed, a coalescence of capsular components in imperfectly fixed specimens is likely to be caused by dehydration, as outlined elsewhere (5) and by Springer and Roth (1 1, 12) (method A, see mainly Fig. 2). On the other hand, excessive or improper reactivity of the fixatives could lead to some material extraction, as it seems to have occurred in B-fixed cells. Despite these possible limitations we think that an intermeshed network structure is the most realistic visualization of the capsule at the electron microscope. With its extensive surface, this network accounts for all chemical properties of bacterial capsules which can bind very high amounts of water, ions, and eventually organic molecules (14, 15). Although all fixation components used here are required for the visualization of the Klebsiella capsule, the use of TAPO should be especially stressed in view of its activity of polysaccharide crosslinker and stabilizer, two properties of utmost importance in chemical fixation for electron microscopy. This compound, indeed, had already proved useful for the visualization of several polysaccharide and glycoprotein-rich structures (I, 2, 3, 4). Thus, by using original fixation for electronmicroscopic observation of Klebsiella pnertmoniae, we extend previous reports by Springer and Roth (1 I), mainly as far as the resolution of capsular components is concerned. These authors did firstly and consistently show the ultrastructure of the capsule in both K. pneutnoniae
and Diplococcus pneumoniae. But, at variance with their ruthenium red method (I]), the fixation and staining procedures used here (in particular, the TAPO-aldehyde-osmium technique) gave also a satisfactory preservation of the other bacterial structures (but not of ribosomes), a fact which is of relevance in judging the reliability of a particular procedure for electron microscopy and, eventually, in adopting it. I n this view, it should be stressed that neither the Ryter and Kellenberger method (9) nor t h e widely used glutaraldehyde-osmium fixations are suitable for a consistent and reproducible electron-microscopic observation of the bacterial capsule. Finally, we think it isof particular interest that the distinct layers of the cell wall are well preserved in B-fixed cells since this could facilitate studies on both the structural relationship between capsule and wall, and the behaviour of the capsular components during cell division, a fact which is almost completely unknown. I. CASSONE, A. 1973. Improved visualization ofcell wall Experientia, structure in Sacchcrronlyces c.er.ri~i.sicrc~. 29: 1303-1306. A , , N. S I M O N E T Ti I~,n dV. STRIPPOLI. 1973. 2. CASSONE, Ultrastructu~.alchanges in the wall during germ-tube formation from blastospores of Ctrrrdid~rol11ic~rn.s.J . Gen. Microbiol.77: 417-428. 3. CASSONE,A , , N. S I M O N E T T Iand , V. STRIPPOLI. 1974. Wall structure and bud fol-mation in Cryptococerrs ~~ec?fi)l-nzcr~ls. Arch. Microbial. 95: 205-212. 4. DJACZENKO, W.. and A. CASSONE.1972. Visualization of new ultrastructural components in the cell wall of Ccrndid(r trlbiccr~lswith fixatives containing TAPO. J. Cell Biol. 52: 18G190. 5. G A R A C IE, . , and A. CASSONE.1974. Capsular structures in Klehsiell(rp~~c.rrnzo~~itre. Boll. Soc. Biol. Sper. 50: 1491-1495. 6. HAYAT,M. A . 1970. Principles and techniques ofelectron microscopy. Vol. 1. Van Nostrand Reinhold Company, New York. 7. MOOR,H. 1969. Freeze-etching. Int. Rev. Cytol. 25: 391-412. 8. REYNOLDS, J . E. 1963. The use of lead citrate as an electron opaque stain for electron microscopy. J. Cell Biol. 17: 208-213. 9. RYTER,A., and E . KELLENBERGER. 1958. Etude au microscope Clectronique des plasmas contenant d e I'acide desoxyribonuclCique. 1. Le nucleide d e batteries en croissance active. Z. Naturforsch. 13b: 597-605. 10. S H E A S. , 1971. Lanthanum staining ofthe surface coat of cells. Its enhancement by t h e use of fixatives containing alcian blue orcetyl pyridinum chloride. J. Cell Biol. 51: 61 1-620. E. L., and I. L. ROTH. 1973. The ultra11. SPRINGER, structure of the capsules of Diplococcr~spnerlrnonicl~
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
CASSONE A N D GARACI
stained with red ruand Klebsielln pn~~irrnot~irrc. thenium. J. Gen. Microbial. 74: 21-31, 12. S P R I N G E RE., L., A N D I . L. ROTH. 1973. Ultrastrucand slime tureof the capsuleof Klehsiellrrpne~rt~~or~itre of Etrrerohocler nerogetles a s revealed by freezeetching. Arch. Mikrobiol. 93: 277-286. 13. S P U R R A. , R. 1969. A low-viscosity, epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26: 31-38.
689
14. SUTHERLAND,I. W. 1972. Bacterial exopolysaccharides. 111Advances in microbial physiology. vol. 8. Edired by A. H . Rose and D. W. Tempest. Academic Press, London and N e w York. pp. 143-213. 15. W I L K I N S O NJ., F. 1958. T h e extracellular polysaccharides of bacteria. Bacterial. Rev. 22: 46-73.
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
Itlstitirte qf Microbiology, U t ~ i ~ ~ e r ofRot71e, sity R o m e , Ittrly Accepted January 28, 1977
CASSONE, A., and E. GARACI.1977. The c a p s ~ ~ l anetwork r of Klehsielln pt~errtnonicre. Can. J. Microbiol. 23: a4-689. Attempts at improving chemical fixation for electron-microscopic observation of the capsule of Klehsiellnp1rero71ot1irrewere made. The capsule was preserved by using alcian blue - lanthanum and tris-(I-aziridinyl) phosphine oxide (TAPO) - aldehyde - osmium procedures. Despite the different retention of the overall capsular material and minor variations in morphological details, in both cases the interpretation of ~~ltrastructural patternssuggested that thecapsule be composed of a meshed netwol-k of thin polysaccharide fibrils radiating from the cell wall. This organization is in keeping with all recognized chemical properties of bacterial polysaccharide capsules or, at least, does not contradict them. Moreover, an effective preservation of bacterial structures other than capsule has been obtained. mostly in specimens fixed by the TAPO-aldehyde-osmium method, a fact which gives further reliability to the technical approach used for capsule visualization. CASSONE, A,. et E. GARACI.1977. The capsular network of Klel~siellrrptrelrmonicre. Can. J . Microbiol. 23: 684-689. Des experiences ont ete conduites en vue d'amkliorer la fixation chimique pour I'examen en microscopie electronique de la capsule de Klrhsielln ptlerrnlot~itre. La capsule est fixke par les methodes utilisant soit I'alcian blue - lanthanum, soit le tris-(I-aziridinyl) phosphine oxide (TAPO) -aldehyde - osmium. Malgre que, par ces methodes, la rttention des divers materiaux capsulaires est differente et que des variations mineures dans les details morphologiques se presentent. I'interpretation des patterns ultrastructuraux suggere que la capsule est constituee d'un reseau en reticules de fines fibrilles polysaccharidiques irradiant de la paroi cellulaire. Cette organisation est en accord avec les pro~rieteschimiques reconnues des capsules polysaccharidiques des bacteries ou, du moins, ne les contredit pas. De plus, une fixation efficace des structures bacteriennes autres que les capsules est obtenue, principalement pour les specimens fixes par la methode TAPO-aldChyde-osmium, ce qui conftre plus de fiabilite B I'approche technique utilisie pour I'examen des capsules. [Traduit par le journal]
Because of their chemical properties, bacterial capsules are poorly preserved in electron microscopy. In fact, they are built up of highly hydrated, delicate gels, generally polysaccharide in nature (14), which can be coarsely extracted or collapse during any one of the standard preparative stages for electron microscopy, mostly during dehydration (1 1). Techniques are now available for embedding without dehydration in water-miscible resins (6) or for electron-microscopic observation after freezing procedures (7). The exploitation of these techniques in Klebsiella and in other capsulated bacteria, while furnishing important new information, did not prove, however, to be fully satisfactory since it gave images of difficult and non-uniform interpretation (I 1, 12). In our opinion, it would be advantageous to use conventional procedures for electron-microscopic visualization of bacterial capsule provided
that chemical fixation was improved. In principle, the ideal fixative should cross-link a n d stabilize the capsular components strongly so as to avoid or minimize any damage due to water removal. In this note we describe attempts for a n improvement in chemical fixation of the capsule in Klebsiella pneu~woniae.
Materials and Methods Several strains of Klebsiellnpt~rutnotriae(IMM-Rome 1 to 5) of unidentified serological type were used throughout this study. They were isolated from pathological specimens and maintained in culture on brain heart infusion medium (BBL, Maryland, U.S.A.). At intervals during growth, the microrganisms were prepared for electron microscopy as follows. Slants were directly flooded with the prefixation mixture for the required time, then cells were removed from the slant, washed, and subsequently treated in suspension as described below. Chemical fixation has always been carried out at 4-6'C. T o this purpose, two main procedures have been
C A S S O N E A N D GARACl
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
TABLE1. Fixation techniques
Prefixation components
Glutaraldehyde (3%) TAPO (0.5%)
Primary stain Vehicle for prefixation pH Time (min) Postfixative Post stain Vehicle for postfixation pH Time (h)
Alcian blue (0.5%) Cacodylate buffer 0.05 M
Glutaraldehyde (0.75%) T A P 0 (1%) Acrolein (1%) None Phosphate buffer 0.05 M
6.0 30 OsOa (1%) Lanthanum nitrate (1%) S-collidine buffer 0.05 M
7.2 30 o s o , (4%) None Hz0
used. The first one, conventionally named A, was characterized by the use of a polycationic dye, alcian blue, and a lanthanum salt in a slight modification of the technique described by Shea (10). Accordingly, the slant was prefixed for 30 min with a mixture of glutaraldehyde, TAPO (Tris-(1-aziridinyl) phosphine oxide; Polysciences, Pennsylvania, U.S.A.), and alcian blue (8GX, Allied Chem., NJ, U.S.A.), 3% (v/v), 0.5% (v/v), and 0.5% (w/v) respectively, in cacodylate buffer, 0.05 M, pH 6.0, and subsequently postfixed for 6 h with a mixture of OsO4 and lanthanum nitrate, both 1% (w/v) in S-collidine buffer, 0.05 M, pH 7.35. The second type of fixation, named B, characteristically involved the use of TAPO and aq~leousosmium, coupled with aldehyde fixative (1, 4). The slant was prefixed for 30 min with a mixture of glutaraldehyde, TAPO, and acrolein, 0.75%, 1%, and 1% (v/v) respectively, in phosphate buffer, 0.05 M, pH 7.2, and postfixed in unbuffered Os04, 4% (w/v) in HzO, pH 5.8, for 12 h. The two fixation techniques are conlparatively sunimarized in Table 1. All fixed specimens were carefully washed in S-collidine buffer and stained overnight at 4'C with uranyl acetate, 0.5% (w/v), in Verona1 acetate buffer, 0.05 M, pH 5.1. Subsequently they were dehydrated and infiltrated (in ethanol-graded series and propylene oxide, at 4"C), then embedded in Spurr low-viscosity resin (13), according to conventional procedures. Thin sections w h ~ c hhad been cut at the Porter-Blum MT-2 ultramicrotome were stained with lead citrate (8) and alcoholic uranyl acetate and observed with a Siemens 1A electron microscope operating at 60-80 kV.
Results After fixation of Kleb.~iellaptze~rmoniaewith method A, the capsule was easily preserved. It was seen as a medium-density, compact structure directly laid over the wall with a regular and continuous profile all around the cell (Fig. I). It didn't show any coarse distortion and its general ultrastructural aspect was that of a fibrous or
fibrogranular net measuring about 120 nm in thickness (Figs. 1, 2). The cytoplasmic matrix and the cell wall appeared sufficiently resolved as compared to other conventional fixation methods, while the nucleoid was slightly swollen and coagulated (Fig. 2). A more distinct resolution of capsular components has been achieved using the fixation method B whereby the capsule was composed of very thin (25-30 A the best resolved ones), lowdensity, and intermingling fibrils radiating from the cell wall (Figs. 3, 4). The capsular average thickness now measures about 55 nm, a value approximately half that measured in A-fixed specimens, suggesting a possible extraction of a marked amount of capsular material. Accordingly, a few individual fibers appear t o be longer than the average fiber (Fig. 3, arrows). As shown by method A as well, the amount of capsular material varied according t o the cell age, being thickest as the culture approaches the stationary phase of growth and thinnest in young and dividing cells especially at the constriction site (Fig. 4). (The distinct phases of growth, however, were not recorded, so this observation, though closely paralleling some biochemical findings (14), needs further experiments). The preservation of the other cell structures following the B method of fixation could be considered satisfactory. The nuclear component was clearly seen as well-resolved and contrasted fibrils, like in thestandard Kellenberger procedure (9), while the cell wall showed distinctly both the outer wall membrane and the inner peptido-
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
CAN. J. MICROBIOL. VOL. 23, 1977
NOTE:The length of marker bars correspond t o 200 nm (Fig. I), 100 nm (Figs. 2, 3, 4), and 50 nm (Fig. 5). cn: capsule; CW, cell wall; CM, cytoplasmic membrane; NU, nucleoid; WM, wall membrane; R, peptidoglycan layer. FIGS1, 2. Klebsielln pneumoniae, strain IMM-Rome 1 and 3, respectively, fixed by method A (for details, see text).
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
CASSONE AND GARACI
FIGS3,4, 5. Klebsiellapnerrmo~riae,strains IMM-Rome 1 (Fig. 3) and IMM-Rome 4 (Figs 4, 5) fixed by method B (see text). Simple arrows in Fig. 3 point to the capsule while head arrows point to longer fibrils. Simple arrows in Fig. 4 point to the constriction area of cell separation where scanty capsular material is seen.
687
688
CAN. J. MICRO BIOL. VOL. 23. 1977
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
glycan layer (Fig. 5). N o artifactual discontinuities were seen in the capsular material nor was the capsule separated from the underlying cell wall as frequently happens in poorly fixed specimens (5).
Discussion In a critical evaluation of these results, it is pertinent to consider that the two procedures of chemical fixation, significantly different from each other, used during this investigation, gave a 'comparable' visualization of the bulk of the capsule which essentially consisted of a network of interconnected fibers evenly distributed around the cell. It is also clear that some degree of 'intermingling' among capsular fibers, as visualized here, could be artifactual resulting in some coarse floccularity of the capsule itself. Indeed, a coalescence of capsular components in imperfectly fixed specimens is likely to be caused by dehydration, as outlined elsewhere (5) and by Springer and Roth (1 1, 12) (method A, see mainly Fig. 2). On the other hand, excessive or improper reactivity of the fixatives could lead to some material extraction, as it seems to have occurred in B-fixed cells. Despite these possible limitations we think that an intermeshed network structure is the most realistic visualization of the capsule at the electron microscope. With its extensive surface, this network accounts for all chemical properties of bacterial capsules which can bind very high amounts of water, ions, and eventually organic molecules (14, 15). Although all fixation components used here are required for the visualization of the Klebsiella capsule, the use of TAPO should be especially stressed in view of its activity of polysaccharide crosslinker and stabilizer, two properties of utmost importance in chemical fixation for electron microscopy. This compound, indeed, had already proved useful for the visualization of several polysaccharide and glycoprotein-rich structures (I, 2, 3, 4). Thus, by using original fixation for electronmicroscopic observation of Klebsiella pnertmoniae, we extend previous reports by Springer and Roth (1 I), mainly as far as the resolution of capsular components is concerned. These authors did firstly and consistently show the ultrastructure of the capsule in both K. pneutnoniae
and Diplococcus pneumoniae. But, at variance with their ruthenium red method (I]), the fixation and staining procedures used here (in particular, the TAPO-aldehyde-osmium technique) gave also a satisfactory preservation of the other bacterial structures (but not of ribosomes), a fact which is of relevance in judging the reliability of a particular procedure for electron microscopy and, eventually, in adopting it. I n this view, it should be stressed that neither the Ryter and Kellenberger method (9) nor t h e widely used glutaraldehyde-osmium fixations are suitable for a consistent and reproducible electron-microscopic observation of the bacterial capsule. Finally, we think it isof particular interest that the distinct layers of the cell wall are well preserved in B-fixed cells since this could facilitate studies on both the structural relationship between capsule and wall, and the behaviour of the capsular components during cell division, a fact which is almost completely unknown. I. CASSONE, A. 1973. Improved visualization ofcell wall Experientia, structure in Sacchcrronlyces c.er.ri~i.sicrc~. 29: 1303-1306. A , , N. S I M O N E T Ti I~,n dV. STRIPPOLI. 1973. 2. CASSONE, Ultrastructu~.alchanges in the wall during germ-tube formation from blastospores of Ctrrrdid~rol11ic~rn.s.J . Gen. Microbiol.77: 417-428. 3. CASSONE,A , , N. S I M O N E T T Iand , V. STRIPPOLI. 1974. Wall structure and bud fol-mation in Cryptococerrs ~~ec?fi)l-nzcr~ls. Arch. Microbial. 95: 205-212. 4. DJACZENKO, W.. and A. CASSONE.1972. Visualization of new ultrastructural components in the cell wall of Ccrndid(r trlbiccr~lswith fixatives containing TAPO. J. Cell Biol. 52: 18G190. 5. G A R A C IE, . , and A. CASSONE.1974. Capsular structures in Klehsiell(rp~~c.rrnzo~~itre. Boll. Soc. Biol. Sper. 50: 1491-1495. 6. HAYAT,M. A . 1970. Principles and techniques ofelectron microscopy. Vol. 1. Van Nostrand Reinhold Company, New York. 7. MOOR,H. 1969. Freeze-etching. Int. Rev. Cytol. 25: 391-412. 8. REYNOLDS, J . E. 1963. The use of lead citrate as an electron opaque stain for electron microscopy. J. Cell Biol. 17: 208-213. 9. RYTER,A., and E . KELLENBERGER. 1958. Etude au microscope Clectronique des plasmas contenant d e I'acide desoxyribonuclCique. 1. Le nucleide d e batteries en croissance active. Z. Naturforsch. 13b: 597-605. 10. S H E A S. , 1971. Lanthanum staining ofthe surface coat of cells. Its enhancement by t h e use of fixatives containing alcian blue orcetyl pyridinum chloride. J. Cell Biol. 51: 61 1-620. E. L., and I. L. ROTH. 1973. The ultra11. SPRINGER, structure of the capsules of Diplococcr~spnerlrnonicl~
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by University of Auckland on 12/07/14 For personal use only.
CASSONE A N D GARACI
stained with red ruand Klebsielln pn~~irrnot~irrc. thenium. J. Gen. Microbial. 74: 21-31, 12. S P R I N G E RE., L., A N D I . L. ROTH. 1973. Ultrastrucand slime tureof the capsuleof Klehsiellrrpne~rt~~or~itre of Etrrerohocler nerogetles a s revealed by freezeetching. Arch. Mikrobiol. 93: 277-286. 13. S P U R R A. , R. 1969. A low-viscosity, epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26: 31-38.
689
14. SUTHERLAND,I. W. 1972. Bacterial exopolysaccharides. 111Advances in microbial physiology. vol. 8. Edired by A. H . Rose and D. W. Tempest. Academic Press, London and N e w York. pp. 143-213. 15. W I L K I N S O NJ., F. 1958. T h e extracellular polysaccharides of bacteria. Bacterial. Rev. 22: 46-73.
