<<

J Clin Pathol: first published as 10.1136/jcp.11.6.483 on 1 November 1958. Downloaded from

J. clin. Path. (1958), 11, 483.

THE BACTERIAL BY ROBERT CRUICKSHANK From the Bacteriology Department, University of Edinburgh Our approach to the bacterial cell depends on to these later. The nuclear or chromatin material our particular interests. The geneticist finds it a has the chemical and reactions of desoxy- most useful plastic tool for studying and ribonucleic acid (D.N.A.), but cannot be regarded variations, transductions and transformations; the as a nucleus in the sense that we use the term for chemist is interested in it as a source of cells, since there is no nuclear , of bewildering variety ; he also helps in identifying no , and no unequivocal evidence of divi- the various components of the cell by fractiona- sion by . tion and detailed chemical analysis of bacterial cell masses; the cytologist is mainly concerned with the of the cell, using modern staining and Although a rigid lining membrane for the bac- microscopic methods, including phase-contrast and terial cell, similar to that for the mycelium, electron microscopy; the bacteriologist is in- was postulated by Cohn in 1875, it is only since terested in the cell's interaction with its immediate the introduction of electron microscopy that cell environment, whether that be living or dead walls have been clearly demonstrable. From bac- terial suspensions which have been mechanically and decaying matter. copyright. For clinical pathologists gathered together to do or sonically disrupted, the cell walls can be separ- honour to Virchow, it would seem most important ated from the electronically more opaque cyto- to discuss, if only in outline, our modern concepts plasm by high-speed at 8,000 to of the anatomy and of the bacterial 10,000 r.p.m., after first removing any intact cells cell. at 2,000 to 3,000 r.p.m. It is less easy to get pure cell walls with Gram-negative than with Gram- Definition positive cells, and mechanical disruption gives belong to the kingdom of Protista, better results than sonic disintegration. The cell which also includes and animal forms and wall constitutes about 20% of the dry weight of http://jcp.bmj.com/ has many resemblances to the blue-green . Gram-positive cocci and as much as 45 % for According to majority opinion at the present time, C. diphtheriae. The thinner walls of Gram- the bacterium may be unicellular or multicellular, negative bacteria probably account for consider- with haploid nuclear or chromatin material occur- ably less than 20 % of the total weight. The thick- ring as a single body and dividing by transverse ness of the staphylococcal cell wall has been esti- . depends on constrictive mated at 15 to 20 mr,, while that of Gram-negative ingrowth of the cell wall and cytoplasmic mem- bacilli, like Bact. coli and salmonellae, is around on October 2, 2021 by guest. Protected brane. 10 to 15 m,u and of Mycobacterium tuberculosis Let us start with a simple diagrammatic repre- 23 mju. The selective rigidity of the outer casing sentation (Fig. 1) of the bacterial cell. It is en- is demonstrated by the finding that the cell walls closed in a cell wall from which there may be of rod-shaped when specially prepared extruded flagella and fimbriae and may be sur- for electron microscopy retain their cylindrical rounded by a capsule. The main of the form, whereas the protoplast assumes a spherical cell wall is to give form and rigidity and some form when the cell wall is dissolved by . protection to the functional cell or protoplast. The Cell wall suspensions have a milky white opales- has a lining or membrane which acts as cent appearance, and in the case of chromogenic a selectively permeable osmotic barrier and con- bacteria there is no contained pigment which is tains within it many granules varying in size from associated with the small particles in the cyto- 10 to 20 mI,u composed mainly of ribonucleic acid plasm. (R.N.A.). Various bodies, such as There is considerable variation in the chemical volutin granules, granules, etc., may be con- composition of the cell walls in different bacterial tained within the cytoplasm, and we shall refer species, the main constituents being -poly- J Clin Pathol: first published as 10.1136/jcp.11.6.483 on 1 November 1958. Downloaded from

484 ROBERT CRUICKSHA NK

MEMBRANE copyright. %1NCLUSION GRANULES

'CELL WALL: EXTRA- PROTOPLASTIC PROTOPLASt FIG. 1.-Diagrammatic representation of a bacterial cell (extracellular structures omitted at bottom of figure). saccharide complexes. Certain major differences a cell wall precursor, since it contained muramic http://jcp.bmj.com/ occur between the Gram-positive and Gram- acid and the amino-acids in the same ratio (Park negative organisms, the former having a limited and Strominger, 1957). These findings may be linked range of amino-acids, whereas the Gram-negative to the early suggestion by Duguid (1946), follow- bacteria have the same full range of amino-acids ing observed morphological changes in - as have most . The lipid content of the inhibited bacteria, that penicillin interferes speci- cell wall in Gram-negative bacteria is usually much fically with the formation of the cell wall while greater than that of Gram-positive organisms allowing growth to proceed until the on October 2, 2021 by guest. Protected (around 20% compared with 2-4%), and this high bursts its defective envelope and so undergoes content of lipid may be related to the lipo-poly- , and to the recent observations by Cooper saccharide of the 0 or endotoxin. Poly- (1954, 1955) that penicillin is specifically bound to saccharides and hexosamine are present in about a bacterial lipid fraction which could be the cyto- equal amounts in both Gram-positive and Gram- plasmic membrane. At last we seem to be learning negative bacteria. something about the mode and site of action of In the past few years more detailed analyses penicillin and perhaps of other antibiotics. have been made of the chemical content of the cell Another application of our knowledge of the wall and its possible precursors (see Park, 1958). chemistry of the cell wall concerns the action of Thus, analysis of staphylococcal cell walls has lysozyme, which acts on a mucopolysaccharide shown that muramic acid, D-glutamic acid, lysine with the release of N-acetyl hexosamine. When and DL-alanine are present in a ratio of about resistant variants of the susceptible M. lysodeik- I: 1: 1: 3, while examination of penicillin-inhibited ticus are obtained, chemical analysis shows no staphylococci revealed an accumulation of a change in their or amino-acid content but uridine-5'- complex which could be a 100-fold increase in the 0-acetyl group. These J Clin Pathol: first published as 10.1136/jcp.11.6.483 on 1 November 1958. Downloaded from

THE BACTERIAL CELL 485 findings suggest that lysozyme resistance may be correct term than capsular swelling reaction) to due to an ability of the organism to acetylate study the structural elements in the capsule and certain cell wall hydroxyl groups associated with also as an indicator of the localization of surface the , which in the sensitive strains are free . When specific antiserum is added to and receptive (Brumfitt, Wardlaw, and Park, 1958). capsulated bacteria, a precipitation reaction quickly Other lytic enzymes, such as that derived from occurs which renders the capsule visible without Streptomyces albus, act on cell walls and can be changing its size or shape. This reaction has been used for preparing bacterial solutions for precipita- used in the identification of such bacterial species tion reaction, such as Lancefield grouping. as pneumococcus, , Klebsiella, Haemophilus, Neisseria, Pasteurella, and Bacillus, Capsules and has helped to show that the capsular substance Much controversy and confusion have arisen of the anthrax bacillus is not a but over the definition of bacterial capsules, and vari- a glutamic acid polypeptide, while that of ous alternative or additional names have been used, Pasteurella pestis is also a or a protein- such as envelope, slimy layer, , and outer polysaccharide complex. coat. We may define a capsule as a microscopi- cally demonstrable slimy or gelatinous layer cover- Flagella ing the cell wall and having a definite external sur- face. The term "envelope" is used by some Flagella are unusual in that they lie outside the bacteriologists if the external boundary is not cell wall and may be removed without killing the sharp or the slime has a tendency to dissolution; cell, yet have a characteristic movement which in- others have described a " capsule membrane," but dicates that they are part of the living cell. Despite there seems no sound support for either of these the ingenious artistry and arguments of Pijper, we may accept the flagella as locomotor

two terms. A good deal of the confusion has copyright. arisen from difficulties in the demonstration of the with their origin in a basal inside the cell capsule. The two most reliable methods are the wall. Structurally, they are non-tapering filaments wet-film India ink technique, used as a negative in the shape of a cylindrical helix, which may have stain, and the specific antigen- reaction either one of two different wavelengths. The fila- erroneously called the Neufeld Quellung or cap- ments vary in length up to 12 ,u and are approxi- sular swelling reaction. Duguid (1951), who gives a mately 12 mpt thick. Chemically they are com- detailed description of the India ink method, found posed predominantly of protein made up of elon- that of dry-film staining techniques the eosin serum gated (molecular weight 41,000) with a method of Howie and Kirkpatrick (1934) was the characteristic amino-acid composition from which http://jcp.bmj.com/ best. Recently, Novelli (1953) and others have , tryptophane, proline, and cysteine are used Alcian blue as a capsule stain, since it is the lacking. The great variety of antigenic entities only dye so far recorded which will stain the bac- among flagella with very similar protein content terial capsule without previous mordanting. With suggests that these antigenic differences depend on one or other of these reliable methods, there is the arrangement rather than the composition of unanimity about the capsules of organisms like the the amino-acids; this plasticity may also account

pneumococcus, the anthrax bacillus, or Fried- for the variation from specific to group phase on October 2, 2021 by guest. Protected lInder's bacillus. With Pasteurella pestis, on the which characterizes many salmonella flagella. other hand, there has been much divided opinion In stained preparations of flagellated bacteria about capsule and envelope. Amies (1951) believed the actual number of flagella is much greater than that the " envelope," dissolving in saline at 600 C., are seen in vivo under dark-ground or phase- is no more than a particularly well developed cap- contrast microscopy. This difference is probably sule which can be shown to persist despite solvents. due to a tendency for flagella to aggregate with In addition to these demonstrable capsules, certain any ; aggregation into quite thick filaments bacteria may carry surface antigens or "micro- is particularly well seen in the elongated forms of capsules" (Wilkinson, 1958) separable and anti- B. proteus grown on agar in the presence of peni- genically distinct from the cell wall, for example, cillin (see Fleming, Voureka, Kramer, and Hughes, the M antigen of Streptococcus pyogenes, which 1950). When flagella have been broken off from can be dissolved off by trypsin, without affecting a young culture they regenerate within half an the cell wall or viability. hour; on the other hand, there is no evidence that Tomcsik and his colleagues (see Tomcsik, 1956) they wear themselves out or even become frayed have used the specific capsular reaction (a more by active movement. J Clin Pathol: first published as 10.1136/jcp.11.6.483 on 1 November 1958. Downloaded from

486 ROBERT CRUICKSHANK Space does not allow any discussion of the in- (1956), who has carried out many interesting teresting studies of Stocker and others on the studies with protoplasts: genetic transductions of flagella and . " Intact bacteria and the protoplasts derived from them have closely parallel capabilities. Both forms Finbriae respire; both synthesize proteins and nucleic acids Before we pass inside the cell wall mention must and form adaptive enzymes; both can support the be made of another type of appendage, the short, multiplication of virulent and temperate bacterio- phages; both can support the development of endo- slender and very numerous fibres, called fimbriae spores; both grow in appropriate media; and both by Duguid, Smith, Dempster, and Edmunds (1955) can divide. Moreover, many of these activities are and Duguid and Gillies (1957) to denote a fringe carried out at approximately similar rates and to which surrounds many species of Gram-negative similar extents by the two forms, the whole and the part. And yet the part is not the whole ; there are bacteria (Bact. coli, Bact. cloacae, Salmonella, differences in behaviour. To what extent some of Shigella flexneri, but not Sh. sonnei or shigae. these differences are due to the use of unsuitable They are about half the thickness of flagella, and conditions is not yet known." fimbriation is enhanced by frequent passage in Protoplasts cannot build a cell wall, possibly broth culture but has no relationship to motility. because a starter of cell wall material is needed These fimbriae have strong adhesive properties before more can be laid down. Similarly, the and are responsible for agglutination of the red failure to sporulate unless the process has already cells of various animal species. The fimbriae been initiated probably means that cell wall is from different serological types of Shigella flexneri needed to complete spore formation. Again, al- are antigenically similar and may be responsible though phage will multiply inside the protoplast, for some of the non-specific agglutinins present in this body cannot be infected with virulent phage, the in many individuals. It is still uncertain which seems to need the cell wall for initial pene- what function, if any, they serve. tration. copyright. The Protoplast The Nuclear Material Separation of the cytoplasm from the cell wall Fierce arguments have been raging for some was observed many years ago by Fischer (1900), years on the nature and interpretation of the who described as " plasmoptysis " the ejection of changes that are demonstrable in the chromatin the cytoplasm from organisms like B. anthracis bodies or nuclear material of the bacterial cell and Vibrio proteus. The empty casing either ad- during the process of growth and division. The hered to or was separated from the cytoplasm, argument largely turns on whether the chromatin which then assumed a spherical form. This pheno- bodies which grow in number with the growth of menon was recently rediscovered by Stahelin the cell are composed of which http://jcp.bmj.com/ (1953, 1954) in weak-walled anthrax bacilli. The divide like the chromosomes of animal and plant term protoplast was, however, first used by Wei- cells, i.e., by mitosis, or whether they divide bull, who dissolved the cell wall of B. megaterium directly or amitotically, as happens with many of by lysozyme and maintained the spherical naked the higher bacteria and Protista. The opposing cell in a stabilizing solution of sucrose or poly- views are set out by Robinow and DeLamater re- ethylene glycol for at least 24 hours. Attached spectively in their chapters in the Symposium on flagella can be demonstrated in the intact cell al- Bacterial Anatomy (1956). Robinow, using Feul- on October 2, 2021 by guest. Protected though motility does not occur, probably for phy- gen and acid Giemsa stains, has studied parti- sical reasons. The protoplasts disintegrate rapidly cularly the development of the single chromatin in non-stabilizing solutions or if the suspension is body in germinating spores, and his interpretations aerated and the lysate is seen to contain " ghosts " of the observed phenomena carry convincing evi- or empty vesicles and small granules. The ghosts, dence for direct amitotic division. DeLamater which probably represent the cytoplasmic lining, has used the actively growing cell for his studies, are composed of delicate and contain and, having perhaps too readily assumed that bac- most of the pigmented material of the cell. The teria divide by mitosis, interprets his findings to fit of protoplasts and whole bacterial this hypothesis. He claims that both B. mega- cells is remarkably similar, so that, besides the terium and Bact. coli have three chromosomes cytochrome system, most of the cell's enzymic which can be shown pictorially to divide simul- activity is contained within the protoplast. Phage taneously by a complex mitotic mechanism inside production and spore formation also occur, pro- the nuclear membrane. He and his colleagues also vided these phenomena have been initiated before describe mitochondria and a centriole, which dissolution of the cell wall. To quote McQuillen Robinow, on the other hand, regards as an acces- J Clin Pathol: first published as 10.1136/jcp.11.6.483 on 1 November 1958. Downloaded from

THE BACTERIAL CELL 487 sory chromatin granule. It is impossible for the small size of the granule and the multiplicity of layman to adjudicate on the merits of these diver- enzymes in many bacteria suggest that there will gent stories by highly specialized cytologists, but be numerous enzymically different types of support can be given to DeLamater's viewpoint granules. They are probably derived from the that the truth will best be elicited by the simul- nucleus, growing by the formation or accumula- taneous application of other techniques, such as tion of proteins. In discussing the rela- those of and bacterial physiology tionship of bacterial cytoplasm to that of other along with the cytochemical methods of the cyto- lowly creatures, Bradfield states that: logist. " the most striking structural difference between If we turn for a moment to the evidence of the bacterial cytoplasm and the cytoplasm of microbial geneticist, it is obvious that genetic ex- and is the absence from bacteria of all kinds nuclear material takes place as a result of intra-cytoplasmic membranes; the oxidative change of granules are not wrapped in membranes to form of sexual pairing in certain species; in addition, mitochondria; the RNA-rich granules are not transfer of occurs to cause the phenomena arrayed on membranes to form endoplasmic reticu- of transformation in pneumococcus and haemo- lum; and finally the cytoplasm as a whole is not philus types and of transduction of toxigenicity in separated from the nucleus by a nuclear mem- C. diphtheriae or motility in salmonella. The cyto- brane." logists have so far been unable to contribute to a Among , the most commonly better understanding of these strange happenings, seen are the Babes-Ernst metachromatic bodies, although there has been some correlation between often called volutin granules. These meta- cytological changes and genetic exchange in higher chromatic granules, seen best in C. diphtheriae organisms like and maize. There is and ranging in size up to 0.6 /u, are very electron little doubt that D.N.A. is the active principle in opaque and chemically contain polymetaphosphate mediated some- and probably other compounds. the genetic mutations of bacteria, copyright. times through bacteriophage. It has also been Smith, Wilkinson, and Duguid (1954) found that, shown that genetic differentiation of the D.N.A. although volutin granules might account for 20% material is linear, i.e., the "bacterial chromo- of the bacterial volume, only 1 % of the dry weight some" may be regarded as a long, thread-like was polymetaphosphate. D.N.A. , carrying the genes in an orderly Wilkinson and his colleagues (Williamson and sequence like the inscription on a tape. The Wilkinson, 1958; Macrae and Wilkinson, 1958) D.N.A. molecule, pictured by Watson and Crick have also studied the lipid inclusions found parti- (1953) as a cylindrical double helix with a diameter cularly in the larger varieties of the Bacillus of 2.5 m,u, would correspond to a thread about species. These lipid granules, which are composed http://jcp.bmj.com/ 1,000 times the length of the parent bacterial cell. mainly of a polymer of f8 hydroxybutyric acid, are Attempts at cytological study of the D.N.A. mole- most numerous in cells about to sporulate or when cule are now being made by electron microscopic grown in semi-synthetic culture media containing examination of ultra-thin sections of bacteria. It an excess of , pyruvate, or ,B hydroxy- is suggested that the thread is coiled up to form a butyrate as the main carbon sources for metabol- tube, and in the photomicrographs of Maal0e and ism. They are regarded by the Edinburgh workers

Birch-Andersen (1956) tube-like structures cut in as stores of carbon and and are probably on October 2, 2021 by guest. Protected all different directions can be demonstrated. It used up in the process of sporulation. Whether may be possible in the future to learn more about inclusion bodies like volutin and the lipid the nucleus and its mode of division by detailed globules serve a specific function in cellular studies of this kind. metabolism, or whether they represent an ab- normal or pathological state of the bacterial cell, Cytoplasm and Inclusion Bodies is a debatable point. As already mentioned, the In electron photomicrographs of sectioned bac- pigments of chromogenic bacteria are found pre- teria (staphylococci, streptococci, paracolon bacilli, dominantly within the cytoplasm, and Schachman, aerobacter, mycobacteria) the cytoplasm seems to Pardee, and Stanier (1952) have observed many consist of fine granules of 10 to 20 m,u with some- well-defined chromatophores of approximately times larger granules, the total number in a single 40 m,u diameter in the cytoplasm of Rhodo- cell being of the order of 104 to 105. Chemically spirillum rubrum. They were absent from - the cytoplasm is composed mainly of ribonucleic isms grown in the dark. Probably many other acid and protein and contains most of the enzymic photosynthetic bacteria possess similar chromato- activity of the cell, partly in its membrane. The phores. J Clin Pathol: first published as 10.1136/jcp.11.6.483 on 1 November 1958. Downloaded from

488 ROBERT CRUICKSHANK

Osmotic Activity of the Bacterial Cell cell wall must also be permeable, and Mitchell Having discussed some aspects of bacterial and Moyle (1956) suggested that it consists of anatomy, we may conclude with a reference to the a network of chains or fibres, with an effective bacterial cell's mode of living. As with other cells, pore diameter of about 1 m,u. Indeed, although the active and growth of bacteria depend on the plasma membrane is the important osmotic an interchange of nutrient and waste products barrier to small molecules, the cell wall probably between the interior of the cell and the external acts as the barrier to solutes of 10,000 molecular environment. This involves a selective intake of weight or over. However, the cytoplasmic mem- nutrient molecules and the expulsion of waste brane is, according to the experiments by Mitchell molecules, while the vital enzymic and nuclear and Moyle and others, more than a semi-permeable substances of the cytoplasm are retained. There barrier, for it contains the bulk of the cytochrome is good evidence that the cytoplasmic lining mem- system and other enzymes, and it may act as a brane functions as a selective osmotic barrier, carrier of substances into and out of the cell, a while the outer wall maintains the structural process called exchange diffusion. This may ex- rigidity of the cell. The bacterial cell is remark- plain the selectivity of the osmotic barrier, which ably resistant to violent changes in its environ- would on this hypothesis be dependent on the ment and will remain intact in media ranging from presence of carrier enzymes which act as ferry distilled water to strong salt solutions, although boats for the transport of selected passengers into this tolerance is considerably reduced in the young, and out of the cell. actively growing phase. Bacteria react by swelling REFERENCES or shrinking in media of high or low osmotic pres- Amies, C. R. (1951). Brit. J. exp. Path., 32, 259. Brumfitt, W., Wardlaw, A. C., and Park, J. T. (1958). Nature (Lond.), sures, but the reaction is selective in that, as 181, 1783. Fischer showed more than 50 years ago, plasmo- Cohn, F. (1875). Beitr. Biol. Pfl., 1, Heft 2, p. 127. Cooper, P. D. (1954). J. gen. Microbiol., 10, 236. lysis of Gram-negative bacteria occurs in high con- -(1955). Ibid., 12, 100. copyright. DeLamater, E. D. (1956). Bacterial Anatomy (6th Symposium of the centrations of salt or sucrose but not in similar Society for General ), p. 215. University Press, of glycerol or urea. During Cambridge. concentrations Duguid, J. P. (1946). Edinb. med. J., 53, 401. the cell wall does not shrink with the -(1951). J. Path. Bact., 63, 673. - and Gillies, R. R. (1957). Ibid., 74, 397. protoplast, the contents of which stream out in - Smith, I. W., Dempster, G., and Edmunds, P. N. (1955). Ibid., a series of fine droplets. Gram-positive bacteria 70, 335. Fischer, A. (1900). Z. Hyg. Infect.-Kr., 35, 1. are much more resistant to plasmolysis, but they, A., Voureka, A., Kramer, I. R. H., and Hughes, W. H. Fleming,(1950). J. gen. Microbiol., 4, 257. too, respond by contraction selectively to hyper- Howie, J. W., and Kirkpatrick, J. (1934). J. Path. Bact., 39, 165. tonic solutes of sucrose but not of glycerol. This Maaloe, O., and Birch-Andersen, A. (1956). Bacterial Anatomy (6th Symposium of the Society for General Microbiology), p. 261. http://jcp.bmj.com/ indicates that they resemble Gram-negative bac- University Press, Cambridge. McQuillen, K. (1956). Ibid., p. 127. teria except that the plasma membrane does not Macrae, R. M., and Wilkinson, J. F. (1958). J. gen. Microbiol., 19, the cell wall, although there 210. easily separate from Mitchell, P., and Moyle, J. (1956). Bacterial Anatomy (6th Sym- is good evidence that it is a distinct . posium of the Society for General Microbiology), p. 150. University Press, Cambridge. The inside the cell may be as Novelli, A. (1953). Experientia (Basel), 9, 34. of Park, J. T. (1958). The Strategy of (8th Symposium of high as 20 atmospheres, and the tensile strength the Society for General Microbiology), p. 49. University Press, the cell wall, greater in spherical than in cylindrical Cambridge.

-- and Strominger, J. L. (1957). Science, 125, 99. on October 2, 2021 by guest. Protected organisms, is responsible for withstanding the high Robinow, C. F. (1956). Bacterial Anatomy (6th Symposium of the if tensile is Society for General Microbiology), p. 181. University Press, pressures. Obviously this strength Cambridge. weakened, as by the action of penicillin, the pheno- Schachman, H. K., Pardee, A. B., and Stanier, R. Y. (1952). Arch. Biochem., 38, 245. menon of plasmolysis or plasmoptosis will readily Smith, I. W., Wilkinson, J. F., and Duguid, J. P. (1954). J. Bact., 68, 450. occur. Stahelin, H. (1953). Schweiz. Z. allg. Path., 16, 892. active plasma membrane is - (1954). Ibid., 17, 296. The functionally Tomcsik, J. (1956). Bacterial Anatomy (6th Symposium of the composed of lipid and protein and is estimated Society for General Microbiology), p. 41. University Press, Cambridge. to be 15 m,u thick if hydrated, or 5 mrnt if un- Watson, J. D., and Crick, F. H. C. (1953). Nature (Lond.), 171, 737. This corresponds to the classical semi- Wilkinson, J. F. (1958). Bact. Rev., 22, 46. hydrated. Williamson, D. H., and Wilkinson, J. F. (1958). J. gen. Microbiol., permeable membrane 2-4 molecules thick. The 19, 198.