- Email: [email protected]
Host factors that influence the behaviour of bacterial pathogens in vivo
Int. J, Med, Microbial. 290, 207-213 (2000) © Urban & Fischer Verlag http://www.urbanfischer.de/journals/ijmm
Host factors that influence the behaviour of bacterial pathogens in vivo Harry Smith The Medical School, University of Birmingham, Birmingham B15 2TT, UK Received April 13, 20.0.0. ' Accepted May 16,200.0.
Abstract Interest is increasing in how bacteria behave and produce virulence determinants within the infected host. There are three aspects of this process; observations on the bacteria themselves, recognition of host factors that affect them and investigation of metabolic interactions between the two. The first aspect is relatively easy to investigate and attracts much interest. The second and third are difficult to work on and hence understudied. The review aims to stimulate interest in them by indicating methods of investigation and describing some successful studies. After discussing host factors that determine growth in vivo consideration is given to factors that influence the production of the determinants of mucosal colonization, penetration, interference with host defence and damage to the host. The final section deals with the influence of host-derived cytidine 5' -monophospho-N-acetylneuraminic acid and lactate on the pathogenicity of gonococci, meningococci and Haemophilus il1f!uel1zae. Key words: host factors - bacterial behaviour in vivo - CMP-NANA, lactate
Introduction Bacterial pathogens have unique biological properties which enable them to colonize mucous surfaces, penetrate them, grow in the environment of the host, inhibit or avoid host defences and damage the host. The bacterial products responsible for these five biological requirements are the determinants of pathogenicity (virulence). Current knowledge comes mainly from studies in vitro but interest is increasing in how bacteria behave and produce virulence determinants within the infected host. There are three aspects: observations on the bacteria and identification of virulence determinants formed during infection; recognition of host factors that affect this bacterial behaviour; and investigation of metabolic interactions between them.
The first aspect is easy to investigate and attracts much interest. New methods have been designed for looking at bacterial activity in vivo, e. g. in vivo expression technology (IVET) for detecting genes expressed in vivo and signature-tagged mutagenesis (STM) for identification of virulence genes in vivo. Much knowledge is accumulating from their use (Dorman et aI., 2000). The second and third aspects are difficult to work on and hence are understudied. Most references to the influence of host factors are made in relation to the environment in vivo as a whole, rather than to specific factors and their changes during infection (Dorman et ai., 2000). This review aims to stimulate interest in identifying such factors and studying their interactions with bacterial pathogens.
Corresponding author: Professor Harry Smith, The Medical School, University of Birmingham, Birmingham B15 2TT, UK, Phone: +44 (0.)121-414-6920., Fax: +44 (0.)121-414-5925 1438-4221100/290/3-20.7 $ 12.00/0
208
H. Smith
Host factors that determine bacterial growth in vivo The multifactorial nature of pathogenicity means that the determinants of all five requirements are essential for its manifestation. However, growth holds the primary position because without it, other determinants would not be formed. The new methods for recognizing genes expressed in vivo (Dorman et al., 2000) are underlining its importance; many of these genes are involved in nutrient acquisition and metabolism. There are several methods for recognizing nutrients that determine growth in vivo. The first depends on the fact that most key nutrients e. g. iron, are also necessary for growth in vitro. Pathogens are grown in defined media and key nutrients identified by lack of growth when they are deleted. Their presence in vivo is often well documented in the literature (Lentner, 1981, 1984) or can be ascertained by analysis of tissue samples. Another method is to see if tissue extracts stimulate growth of the pathogen in minimal media and then to purify the stimulant. Erythritol, which facilitates rapid growth of Brucella abortus in foetal tissues and fluids of pregnant cattle thereby causing abortion (Smith et al., 1962; Williams et al., 1964; Anderson and Smith, 1965; Keppie et al., 1965) was discovered in this way. Nutrient availability in vivo can be indicated by auxotrophic mutants. Auxotrophs of Salmonella typhi, Salmonella typhimurium and Shigella {lexneri unable to synthesize p-aminobenzoic acid, purines, pyrimidines and histidine have low virulence for mice, rabbits, monkeys or man (Ahmed et al., 1990; Curtiss et al., 1988; Fields et al., 1986; Kamell et al., 1993; Leuing and Finlay, 1991; Levine et al., 1987; O'Callaghan et al., 1988). Their impaired growth indicates absence or scarcity of the nutrients in these animals. Finally, the new methods for identifying genes expressed in vivo that are involved in acquiring and metabolizing nutrients have provided a powerful tool. The enzymes coded by the genes indicate nutrients used in vivo. In STM studies of staphylococci and Vibrio cholerae in mice, virulence genes were identified which indicated that proline and biotin (respectively) are important for growth in vivo (Schwan et al., 1998; Chiang and Mekalanos, 1998). Two previously unrecognized genes of Escherichia coli, guaA and argC, were induced in urine indicating that synthesis of guanine and arginine is important in uropathogenesis (Russo et al., 1996). After identification, the concentrations of these key nutrients in uninfected tissues and the changes that occur during infection need to be ascertained. The first can be obtained easily, -either from the literature (Lentner 1981; 1984) or by analysis of tissue samples. The
second is extremely difficult. However, some parameters have been measured within infected cells. Quantitative fluorescence microscopy measured intraphagosomal pH in macrophages (Aranda et al., 1992) and lacZ reporter genes indicated the approximate levels of Ca 2 +, Fe 2 + and Mg2+ in tissue culture cells (Pollack et al., 1986; Garcia del Portillo et al., 1992). It should be kept in mind that as infection proceeds nutrients could be generated by the action of bacterial enzymes e. g. proteases on host high Mr components. The relevance of identified nutrients to infection in vivo should be tested. The virulence of the pathogen may be enhanced by including additional nutrient in the inoculum, e. g., biotin with V. cholerae in mouse intestine (Chiang and Mekalanos 1998). Also, mutants unable to use the nutrient may have reduced virulence, e. g., the above auxotrophs. The existence in various tissues of high osmolarity, low pH and anaerobic conditions, which could inhibit growth in vivo, can be ascertained from the literature or by analysis of tissue samples. Also, these adverse influences, and how they are overcome, may be indicated by the functions of genes that are expressed in vivo. IVET identified cadA as a gene expressed by V. cholerae in rabbit ileal loops and orally infected mice. This gene codes for lysine decarboxylase, an important element in the acid tolerance response of bacteria. Thus, its expression fits with resistance to low pH in the stomach and the fatty acid content of the ileum (Merrell and Camilli, 2000). The final questions for consideration are; how are nutrients metabolized in vivo and by what bacterial determinants; and how do bacteria compensate for nutrient deficiencies and hostile biochemical conditions? It is well-nigh impossible to investigate the metabolism of pathogens growing in infected tissues. The aim has to be to simulate in vitro an aspect of behaviour relevant to growth in vivo, e. g. a rapid doubling time, by addition of nutrients that appear relevant in vivo. Then, meaningful studies can be made by established methods of bacterial physiology. Knowledge of genes expressed in vivo and metabolic potential revealed by genomics (Huynen et al., 1999) may be helpful. Genomics aided identification of the transport protein for bacterial acquisition of zinc for the metalloproteins in DNA and RNA polymerases (Lewis et al., 1999). This section ends with a summary of the well known and erudite studies on iron acquisition in vivo to illustrate that all the steps outlined above are achievable. In vitro studies showed iron is an essential nutrient for all bacJeria. Its availability in vivo is restricted by chelation to host transferrin and lactoferrin, and injection of iron salts enhances virulence (Bullen, 1981). Using iron-limiting conditions in vitro, molecular studies revealed strategies for overcoming iron restriction (Wein-
Host factors on bacterial behaviour in vivo
berg, 1995). Siderophores excreted by some pathogens return to bacteria via cell-surface receptors to give up their iron under the action of reductases (Halle and Meyer, 1992; Weinberg, 1995). In other cases, transferrin-bearing iron interacts with cell wall receptors and delivers iron into the bacteria (Cornelissen et aI., 1992). The cell wall receptors were demonstrated on bacteria in patients or infected animals (Brown and Williams, 1985; Cornelissen et aI., 1992; Smith 1990; 1996). Mutants deficient in determinants of iron acquisition were attenuated in virulence tests (Cornelissen et aI., 1997).
Host factors that affect virulence determinant production Two types of host factors influence production of the determinants of mucosal colonization, penetration, interference with host defence and damage to the host. The first are the substrates for production of the determinants. The second are the factors that provide the environment for operation of the requisite regulatory processes. In both cases, we are referring to factors additional to those required for growth. Relevant substrates can be recognized when comparisons of in vivo and in vitro grown bacteria have revealed a biological property appertaining to virulence, e. g. resistance to serum killing, possessed by the former but not by the latter. Bacteria are then grown in media supplemented by plasma or host cell extracts in the hope that the biological property is reproduced in vitro and can be used as an assay for purification of the responsible host factor. Examples of this approach are described in the final section. There are two methods for studying factors in the environment in vivo which affect regulation of virulence determinant production. The first is similar to that for recognizing nutrients involved in growth. The presence and concentrations of potential host factors (ions, metabolites, osmolarity, pH, and Eh ) in the tissues that become infected are obtained from the literature or by analysis. Then, attempts are made to mimic in vitro the conditions in vivo and observe effects on virulence determinant production and regulation. Up to the present, most studies have concerned intestinal conditions. Six invasion genes of S. typhimurium were maximally expressed in vitro at an oxygen tension, osmolarity and pH likely to exist in the ileum (Bajaj et aI., 1996). Two type III secretion genes of S. typhi, invG and prgH, were induced by high osmolarity, anaerobic conditions and pH 6,$ (Leclerc et aI., 1998). The Yst regulated toxin of Y. enterocolitica is only produced in vitro at temperatures below 30°C, but, 'at an osmolar-
209
ity and pH similar to those in the ileum, it is formed at 3]oC (Mikulskis et aI., 1994). EspB, the protein of E. coli which triggers processes resulting in effacement of microvilli and cytoskeletal changes, is maximally secreted at the pH, temperature, and osmolarity of the intestine (Kenny et a!., 1997). Aeromonas hydrophila produces rough lipopolysaccharide (LPS) in cultures at 37°C but smooth LPS and increased virulence at the osmolarity of the ileum (Aguilar et aI., 1997). The second approach is to investigate whether the environmental conditions that affect regulons in vitro exist and operate in vivo. Unfortunately, it has been successful only in a few cases. Mg2+ concentrations appear to control the PhoPlPhoQ system of S. typhimurium in vivo as in vitro. Within phagosomes, where the bacteria reside and resist intracellular killing, the Mg2+ concentration (estimated 20-100 ~M) is permissive for phoPiphoQ expression, unlike the high concentrations (0.5-1.0 mM) found in cytosols and body fluids (Garcia Vescovi et a!., 1996). Also, 37°C is permissive for expression of virB and virF and hence the invasion proteins of Shigella spp. (Maurelli and Sansonetti, 1988). However, for the BvgAS system which regulates production of many virulence determinants of Bordetella spp., it is controlled in vitro by MgS0 4 , nicotinic acid derivatives and temperature but in vivo only the permissive temperature (3]OQ is known to be relevant (Melton and Weiss, 1993; Martinez de Tejada et a!., 1996). Studies of the effect of different concentrations of MgS04 and various nicotinic acid derivatives in vitro suggested that the latter may have an affect in vivo but this was not proven (Melton and Weiss, 1993). The position regarding environmental regulation of the ToxRf[oxS system in vivo is even more obscure (Skorupski and Taylor, 1997; DiRita et a!., 2000). In vitro, V. cholerae forms toxin maximally at low temperature, under aerobic conditions and at low pH; and bile salts are repressive. In the human intestine, toxin production occurs at 37°C, under anaerobic conditions, at high pH and in the presence of bile salts . .
The influence of host-derived cytidine S'-monophospho-N-acetyl neuraminic acid (CMP-NANA) and lactate on the pathogenicity of gonococci, meningococci and Haemophilus influenzae CMP-NANA is the donor for sialylation of gonococcal LPS which confers resistance to killing by serum (Smith et a!., 1995; Vogel and Frosch, 1999). Lactate stimulates gonococcal growth, metabolism and virulence determinant production in a medium containing glucose, i. e. the conditions occurring in vivo (Parsons et a!.,
210
H. Smith
1996a, b; Gao et al., 1998). These factors were discovered by investigating why gonococci in urethral exudates are resistant to killing by fresh human serum and lose their resistance on subculture. Incubation with blood cell extracts restored resistance to serum killing, and fractionation showed that the active compound was CMP-NANA (Smith et al., 1995). Sodium dodecyl sulphate polyacrylamide electrophoresis on gonococci grown with CMP_ 14 CNANA showed some LPS components were sialylated. When sialyl groups were removed by neuraminidase reversion to serum sensitivity resulted. Sialylation of LPS interfered with the following host defence mechanisms: adsorption of complement component C3; ingestion and killing by neutrophils; killing by antibody against gonococcal porin 1B; and stimulation of the immune response, but sialylation prevented invasion of epithelial cells. A sialyltransferase was demonstrated in gonococcal extracts, and a sialyltransferase-deficient mutant, in contrast to its parent, did not become serum resistant when incubated with CMP-NANA (Bramley et al., 1995) nor resistant to the other host defence mechanisms mentioned above (Gill et al., 1996). Blood cell extracts enhanced the ability of CMPNANA to sialylate LPS and to induce serum resistance. The factor concerned was purified and shown to be lactate (Parsons et al., 1996a); minute quantities were active in a defined medium containing high concentrations of glucose. Lactate alone did not induce serum resistance and its action is separate from that of CMPNANA, because enhanced sialylation occurred when gonococci were pretreated with lactate (Parsons et al., 1996b). Lactate markedly increased LPS, protein and ribose synthesis (Gao et al., 1998). Increased LPS synthesis could explain enhancement of sialylation because more receptors become available. The changes also occurred in gonococci when glucose and lactate concentrations were adjusted to levels akin to those in vivo (Gao et al., 1998). This evidence of general stimulation of gonococcal metabolism is backed by other observations. Lactate increased oxygen consumption by gonococci in a salt solution containing glucose (Britigan et al., 1988); and in the defined medium containing glucose, gonococcal growth was faster with lactate which was metabolized concomitantly with glucose and more rapidly (Regan et al., 1999). The increased growth and metabolism might have been due to general stimulation of nutrient uptake but no evidence for this was obtained from uptake of [14C]adenine, [14C]glucose and [14C]proline (Gao et al., 2000). When gonococci were grown with 14C-Iabelled lactate aerobically in the defined medium containing glucose and lactate, label was concentrated in a low-Mr component X and LPS. Some label was detected in a few proteins. Thin layer chromatography showed that
component X was membrane lipid. N-terminal sequencing of the three proteins most heavily labelled by [14C]lactate showed one was GroEL and another porin 1B (Yates et al., 1999). Production of these proteins probably increases along with the general stimulation of protein synthesis that occurs with lactate (Gao et al., 1998) but, unlike for LPS, specific measurements have not been made. Gonococcal lipids are complex mixtures of ph osphatidyl ethanolamines and glycerols containing 4 principal fatty acids (Sud and Feingold, 1975; Senff et al., 1976). Nuclear magnetic resonance spectroscopy on lipid extracted from gonococci grown with [13C]lactate and unlabelled glucose, showed that lactate carbon was in fatty acid components, not ethanolamine/glycerol moieties, in contrast to when [13C]glucose was used alone or with unlabelled lactate (Yates et al., 1999). The pattern of glucose carbon incorporation changes when lactate is present. More carbon is incorporated into lipid, and different fatty acids may be formed. Lactate seems to have two overlapping functions in gonococcal pathogenicity; stimulation of growth and metabolism, and increased production of LPS and possibly other virulence determinants. GroEL, a chaperone, ensures correct folding during increased protein synthesis; porin 1B plays a metabolic role in membrane transport; and stimulation of lipid synthesis aids membrane formation. As regards virulence determinants, LPS synthesis is increased and GroEL is another stimulator of inflammatory cytokines (Coates and Henderson, 1998). Porin 1B inserts into membranes of neutrophils and impairs their action (Bjerknes et al., 1995). The LPS of some meningococci is sialylated, endogenously for sero groups B, C, Wand Y and exogenously, by CMP-NANA, for groups A and 29E (Smith et al., 1995). Also, the degree of sialylation of the former can be increased by incubation with exogenous CMPNANA (Estabrook et al., 1997). LPS sialylation affects several aspects of pathogenicity that are similar to those described above for gonococci but not as markedly, because meningococci have another virulence determinant, capsular polysaccharide (Smith et al., 1995). Sometimes it is difficult to distinguish between their respective roles. For example, the position over serum resistance is equivocal; some papers indicate sialylation of LPS is important (Estabrook et al., 1997; Kahler et al., 1998), others not (Vogel and Frosch, 1999). Unfortunately, the effect of lactate on meningococci has not yet been studied. The LPS of some strains of H. influenzae is also sialylated and it increases serum resistance (Mandrell et al., 1992; Hood et al., 1999). Sialylation can be increased by incubation with exogenous CMP-NANA or
Host factors on bacterial behaviour in vivo
sialic acid for some strains (Hood et al., 1999) but not for others (Kuratana et al., 1989). Like meningococci, H. influenzae has a capsule which contributes to virulence. Organisms in blood samples and nasal washes are·more resistant to killing by fresh serum than organisms grown in broth (Shaw et al., 1976; Rubin and Moxon 1985). Incubation of the latter organisms for 30-40 minutes with a low-Mr filtrate of serum or nasal washings made them serum resistant; this was lost on broth subculture (Shaw et al., 1976). These organisms were also resistant to other defence mechanisms and more virulent for rats (Kuratana et al., 1991). There are two mechanisms of serum resistance (Kuratana et al., 1990). In one, both capsular deficient mutants and capsulated wild types were converted to resistance and LPS content increased. In the second, only capsular wild types were converted to resistance and capsular polysaccharide increased. Substitution of the serum filtrate by a buffer containing lactate, glucose, urea and bicarbonate produced the first mechanism of resistance induction; and a mixture of lactate and Ca 2 + the second (Kuratana et al., 1991). In the first case, lactate may play a similar role to that for gonococci, namely, stimulation of LPS synthesis thereby producing more receptor for sialylation and hence increased serum resistance. For H. influenzae, the sialylation occurs endogenously 'for the strains concerned, so that lactate alone could induce serum resistance. For gonococci, exogenous CMP-NANA is needed for lactate to have an effect. In the second mechanism lactate may be a carbon source for increased production of capsular polysaccharide. It is fascinating that LPS sialylation plays a similar J role in the pathogenicity of gonococci, meningococci and H. influenzae and that lactate also has a strong influence, at least for two of them. Such mechanisms may operate for other pathogens. Acknowledgements. I am indebted to]. A. Cole, M.J. Gill, N. ]. Parsons, C. W. Penn, N. Woodward and E. A. Yates for reading the manuscript.
References Aguilar, A., Merino, S., Rubines, X., Tomas,]. M.: Influence of osmolarity on lipopolysaccharides and virulence of Aeromonas hydrophila serotype 0: 34 strains grown at 3rc. Infect. Immun. 65, 1245-1250 (1997). Ahmed, Z. u., Serker, M. R., Sack, D. A.: Protection of adult rabbits and monkeys from lethal shigellosis by oral immunization with thymine-requiring and temperaturesensitive mutants of Shigella {lexneri. Vaccine 8, 153-158 (1990) . Anderson, ]. D., Smith: H.: The metabolism of erythritol by Brucella abortus.]. Gen. Microbiol. 38, 100-124 (1965).
211
Aranda, C. M. A., Swanson, J. A., Loomis, W. P., Miller, 5.1.: Salmonella typhimurium activates virulence gene transcription within acidified macrophage phagolysomes. Proc. Nat!. Acad. Sci. USA 89, 10079-10083 (1992). Bajaj, V., Lucas, R. L., Hwang, c., Lee, C. A.: Coordinate regulation of Salmonella typhimurium invasion genes by environmental and regulatory factors is mediated by control of hilA expression. Mol. Microbiol. 22, 703-714 (1996). Bjerknes, R. , Guttormsen, H. , Solberg, C. 0., Wetzler, L. M.: Neisserial porins inhibit human neutrophil actin polymerization, degranulatation, opsonin receptor expression and phagocytosis but prime the neutrophils to increase their oxidative burst. Infect. Immun. 63, 160-167 (1995). Bramley, J., Demarco de Hormaeche, R., Constantinidou, c., Nassif, X., Parsons N.]., Jones, P., Cole,]. A., Smith, H.: A serum-sensitive, sialyltransferase-deficient mutant of Neisseria gonorrhoeae defective in conversion to serum resistance by CMP-NANA or blood cell extracts. Microb. Pathog. 18, 187-195 (1995). Britigan, B. E., Klapper, D., Svendsen, T., Cohen, M.S.: Phagocyte-derived lactate stimulates oxygen consumption by Neisseria gonorrhoeae. ]. Clin. Invest. 81, 318-324 (1988). Brown, M. R. W., Williams, P.: The influence of environment on envelope properties affecting survival of bacteria in infections. Anml. Rev. Microbiol. 39,527-556 (1985). Bullen, J. J.: The significance of iron in infection. Rev. Infect. Dis. 3,1127-1138 (1981). Chiang, S. L., Mekalanos,].].: Use of signature-tagged transpo son mutagenesis to identify Vibl'io cholerae genes critical for colonization. Mol. Microbiol.27, 797-805 (1998). Coates, A. R. M., Henderson, B.: Chaperonins in health and disease. Ann. N. Y. Acad. Sci. 851,48-53 (1998). Cornelissen, C. N., Biswas, G. D., Tsai, J., Parachuri, D. K., Thompson, S. A. Sparling, P. E: Gonococcal transferrinbinding protein 1 is required for transferrin utilization and is homologous to Ton-B-dependent outer membrane receptors. ]. Bacteriol. 174,5788-5797 (1992). Cornelissen, C. N., Kelly, M., Hobbs, M. M., Anderson,]. E, Cannon, ]. G., Cohen, M. S., Sparling P. E: The transferrin receptor expressed by gonococcal strain FA1090 is required for experimental infection of human male volunteers. Mol. Microbiol. 27, 611-616 (1997). Curtiss, R. III, Kelly, S. M., Gulig, P. A., Gentry-Weeks, C. R., Galan, j . E.: Avirulent salmonellae expressing virulence antigens from other pathogens for use as orally administered vaccines. In: Virulence mechanisms of bacterial pathogens (J. A. Roth, ed.) pp. 311-328. American Society for Microbiology Press, Washington DC 1988. DiRita, V. ]., Engleberg, N. c., Heath, A., Miller, A., Crawford, ]. A., Yu, R.: Virulence gene regulation inside and outside. Phil. Trans. R. Soc. Lond. B 355, 657-665 (2000). Dorman, c.]., Dougan, G., Holden, D. W., Williams, P., Smith, H.: The activities of bacterial pathogens in vivo. Phil. Trans. R. Soc. Lond. B 355, 547-712 (2000). Estabrook, M. M ., Griffiss, J. M., Jarvis, G. A.: Sialylation of Neisseria meningitidis lipooligosaccharide inhibits serum bactericidal activity by masking lacto-N-neotetraose. Infect. Immun. 65, 4436-4444 (1997) . <
212
H. Smith
Fields, P. 1., Swanson, R. V., Haidaris, C G., Heffron, F.: Mutants of Salmonella typhimurium that cannot survive within macrophages are avirulent. Proc. Nat!. Acad. Sci. USA 83, 5189-5193 (1986). Gao, L., Parsons, N.]., Curry, A., Cole,]. A., Smith, H.: Lactate causes changes in gonococci including increased lipopolysaccharide synthesis during short-term incubation in media containing glucose. FEMS Microbiol. Lett. 169, 309-316 (1998). Gao, L., Linden, L., Parsons, N.]., Cole,]. A., Smith, H.: Uptake of metabolites by gonococci grown with lactate in a medium containing glucose: evidence for a surface location of the sialyltransferase. Microb. Pathog. 28, 257266 (2000). Garcia-del Portillo, F., Foster, ]. W., Maguire, M. E., Finlay, B. B.: Characterization of the microenvironment of Salmonella typhimurium containing vacuoles within MDCK epithelial cells. Mol. Microbiol. 6, 3289-3297 (1992). Garcia Vescovi, E., Soncini, F. C, Groisman, E. A.: Mgz+ as an extracellular signal: environmental regulation of Salmonellaevirulence. Cell 84, 165-174 (1996). Gill, M. J., McQuillen, D. P., van Putten, ]. P. M., Wetzler, L. M., Bramley, J., Crooke, H., Parsons, N. J., Cole, ]. A., Smith, H.: Functional characterization of a sialyltransferase-deficient mutant of Neisseria gonorrhoeae. Infect. Immun. 64,3374-3378 (1996). Halle, F., Meyer,]. M.: Iron release from ferrisiderophores. A multi-step mechanism involving a NADHlFMN oxidoreductase and chemical Areduction by FMNH z. Eur.]. Biochern. 209, 621-627 (1992). Hood, D. W., Makepeace, K., Deadman, M. E., Rest, R. F., Thibauld, P., Martin, A., Richards, ]. C, Moxon, E. R.: Sialic acid in the lipopolysaccharide of Haemophilus in{luenzae: strain distribution, influence on serum resistance and structural characterization. Mol. Microbiol. 33, 679692 (1999). Huynen, M. A., Dandekar, T., Bork, P.: Variation and evolution of the citric-acid cycle: a genomic perspective. Trends Microbiol. 7,281-291 (1999). Kahler, C M., Martin, L. E., Shih, G. C, Rahman, M. M., Carlson, R. W., Stephens, D. S.: The (u2 ~ 8)linked polysialic acid capsule and lipooligosaccharide structure both contribute to ability of serogroup N Neisseria meningitidis to resist the bactericidal activity of normal human serum. Infect. Immun. 66, 5939-5947 (1998). Karnell, A., Cam, P. D., Verma, N., Lindberg, A. A.: AroD deletion attenuates Shigella {lexneri strain 2457T and makes it a safe and efficacious oral vaccine in monkeys. Vaccine 8, 830-836 (1993). Kenny, B., Akio, A., Stein, M., Finlay, B. B.: Enteropathogenic Escherichia coli protein secretion is induced in response to conditions similar to those in the gastrointestinal tract. Infect. Immun. 65, 2606-2612 (1997). Keppie, ]., Williams, A. E., Witt, K., Smith, H.: The role of erythritol in the tissue localization of the brucellae. Brit. ]. Exp. Pathol. 46,104-108 (1965). Kuratana, M., Anderson, P.: Host metabolites that phenotypically increase the resistance of Haemophilus in{luenzae type b to clearance mechanisms.]. Infect. Dis. 163, 10731079 (1991).
Kuratana, M., Hansen, E.]., Anderson, P.: Multiple mechanisms in serum factor-induced resistance of Haemophilus in{luenzae type b antibody. Infect. Immun. 58, 914-917 (1990). Kuratana, M., Inzana, T.]., Anderson, P.: Source of lowmolecular-weight host factors that phenotypically increase the resistance of Haemophilus in{luenzae Type b to bacteriolysis: nonidentity of a factor active in gonococci. Microb. Pathog. 7, 73-77 (1989). Leclerc, G.]., Tartera, C, Metcalf, E. S.: Environmental regulation of Salmonella typhi invasion-defective mutants. Infect. Immun. 66,682-691 (1998). Lentner, C: Units of Measurement, Body Fluids, Composition of the Body, Nutrition. In Geigy Scientific Tables, vol. 1., Ciba Geigy, Basle 1981. Lentner, C: Physical Chemistry, Composition of the Blood, Haematology, Sonatometric Data. In Geigy Scientific Tables, vol. 3., Ciba Geigy, Basle 1984. Leuing, K. Y., Finlay, B. B.: Intracellular replication is essential for virulence of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 88, 11470-11474 (1991). Levine, M. M., Herrington, D., Murphy, J. R., Morris, ]. G., Losonsky, G., Tall, B., Lindberg, A. A., Svenson, S., Bagar, S., Edwards, M. F., Stocker, B.: Safety, infectivity, immunogenicity and in vivo stability of two attenuated auxotrophic mutant strains of Salmonella typhi 541 Ty and 543Ty used as oral vaccines for man.]. Clin. Invest. 79, 888-902 (1987). Lewis, D. A., Klesney-Tait, ]., Lumbley, S. R., Ward, C K., Latimer, ]. L., Ison, C A., Hansen, E.].: Identification of the znuA-encoded periplasmic zinc transported protein of Haemophilus ducreyi. Infect. Immun. 67, 5060-5068 (1999). Mandrell, R. E., McLaughlin, R., Kwaik, Y. A., Lesse, A., Yamasaki, R., Gibson, B., Spinola, S. M., Apicella, M. A.: Lipooligosaccharides (LOS) of some Haemophilus species mimic human glycosphingolipids, and some LOS are sialylated. Infect. Immun. 60, 1322-1328 (1992). Martinez de Tejada, G., Miller,]. F., Cotter, P. A.: Comparative analysis of virulence control systems of Bordetella pertussis and Bordetella bronchisepticus. Mol. Microbiol. 22, 895-908 (1996). Maurelli, A. T., Sansonetti, P.].: Identification of chromosome genes controlling temperature-regulated expression of Shigella virulence. Proc. Nat!. Acad. Sci. USA 85, 2820-2824 (1988). Melton, A. R., Weiss, A. A.: Characterization of environmental regulators of Bordetella pertussis. Infect. Immun. 61, 807-815 (1993). Merrell, D. S., Camilli, A.: Detection and analysis of gene expression during infection by the in vivo expression technology. Phil. Trans. R. Soc. Lond. B 355, 587-599 (2000). Mikulskis, A. V., Delor, I., Thi, V. H., Cornel is, G. R.: Regulation of the Yersinia enterocolitica enterotoxin Yst gene. Influence of growth phase, temperature, osmolarity, pH and bacterial host factors. Mol. Microbiol. 14,905-915 (1994). O'Callaghan, D., Maskell, D., Liew, F. Y., Easmon, C S. F., Dougan, D.: Characterization of aromatic- and purinedependent Salmonella typhimurium; attenuated persis-
Host factors on bacterial behaviour in vivo tence, and ability to induce protective immunity In BALB/c mice. Infect. Immun. 56,419-423 (1988). Parsons, N.]., Boons, G.J., Ashton, P. R., Redfern, P. D., Quirk, P., Gao, Y., Constantinidou, c., Patel, J., Bramley, J., Cole, ]. A., Smith, H.: Lactic acid is the factor in blood cell extracts which enhances the ability of CMP-NANA to sialylate gonococcal lipopolysaccharide and induce serum resistance. Microb. Pathog. 20, 87-100 (1996a). Parsons, N. J., Emond, ]. P., Goldner, M., Bramley, J., Crooke, H., Cole, J. A., Smith, H.: Lactate enhancement of sialylation of gonococcal lipopolysaccharide and induction of serum resistance by CMP-NANA is not due to direct activation of the sialyltransferase: metabolic events are involved. Microb . Pathog. 21, 193-204 (1996b). Pollack, c., Straley, S. c., Klempner, M. S.: Probing the phagolysosome environment of human phagocytes with a Ca 2 +-responsive operon fusion in Yersinia pestis. Nature (Lond) 332, 834-836 (1986). Regan, T., Watts, A., Smith, H., Cole, J. A.: Regulation of the lipopolysaccharide-specific sialyltransferase activity of gonococci by growth state of the bacteria, but not by carbon source, catabolite repression or oxygen supply. Antonie Van Leeuwenhoek 75, 369-379 (1999). Rubin, L. G., Moxon, E. R.:The effect of serum-factor induced resistance to somatic antibodies on the virulence of Haemophilus inf/uenzae type b. J. Gen. Microbiol. 131, 515-520 (1985). Russo, T. A., Jadush, S. T., Brown, J.J., Johnson, J. R.: Identification of two previously unrecognised genes (guaA and argC) important for uropathogenesis. Mol. Microbiol. 22,217-228 (1996). Schwan, W. R., Coulter, S. N., Ng, E. Y. W., Langhorne, M . H., Ritchie, H. D., Brody, L. L. , West brock-Wadman, S., Bayer, A. S., Folger, K. R., Stover, C. K.: Identification and characterization of the PutP proline permease that contributes to in vivo survival of Staphylococcus au reus in animal models. Infect. Immun. 66, 567-572 (1998). Senff, L. M., Wegener, W. S., Brooks, G. E, Finnerty, W. R., Makula, R. A.: Phospholipid composition and phospho-
213
lipase A activity of Neisseria gonorrhoeae. ]. Bacteriol. 127,874-880 (1976). Shaw, S., Smith, A. L., Anderson, P., Smith, D. H.: The paradox of Haemophilus inf/uenzae type b bacteremia in the presence of serum bactericidal activity. J. Clin. Invest. 58, 1019-1029 (1976). Skorupski, K., Taylor, R. K.: Control of the ToxR virulence regulon in Vibrio cholerae by environmental stimuli. Mol. Microbiol. 25, 1003-1009 (1997). Smith, H .: Pathogenicity and the microbe in vivo. J. Gen. Microbial. 136,377-393 (1990). Smith, H .: What happens in vivo to bacterial pathogens? Ann. N. Y. Acad. Sci. 797, 77-92 (1996 ). Smith, H., Parsons, N.J., Cole, J. A.: Sialylation of neisserial lipopolysaccharide: a major influence on pathogenicity. Microb. Pathog. 19,365-377 (1995). Smith, H., Williams, A. E., Pearce, J. H., Keppie, J., HarrisSmith, P. W., Fitzgeorge, R. B., Witt, K.: Foetal erythritol: a cause of the tissue localization of Brucella abortus in bovine contagious abortion. Nature Lond. 193, 47-49 (1962). Sud, I.J., Feingold, D. S.: Phospholipids and fatty acids of Neisseria gonorrhoeae. J. Bacteriol.124, 713-717 (1975). Vogel, U., Frosch, M.: Mechanisms of neisserial serum resistance. Mol. Microbiol. 32, 1133-1139 (1999). Weinberg, E. D.: Acquisition of iron and other nutrients in vivo. In: Virulence mechanisms of bacterial pathogens, 2nd ed. (J. A. Roth, C. A. Bolin, R. A. Brogden, E C. Minion, M.J. Wannemuehler, eds.), pp. 79-93. American Society for Microbiology Press, Washington DC 1995. Williams, A. E., Keppie, ]., Smith, H.: The relation of erythritol usage to virulence in Brucellas.J. Gen. Microbiol. 37, 265-292 (1964). Yates, E. A., Gao, L., Parsons, N.J., Cole, ]. A. , Smith, H. : When lactate is used by gonococci growing in a medium containing glucose, as occurs in vivo, its carbon is incorporated into lipid, LPS, GroEL and porin lB. Abstract 37. A Royal Society Discussion Meeting on the 'Activities of bacterial pathogens in vivo'. Society for General Microbiology, Reading, UK 1999.
Host factors that influence the behaviour of bacterial pathogens in vivo Harry Smith The Medical School, University of Birmingham, Birmingham B15 2TT, UK Received April 13, 20.0.0. ' Accepted May 16,200.0.
Abstract Interest is increasing in how bacteria behave and produce virulence determinants within the infected host. There are three aspects of this process; observations on the bacteria themselves, recognition of host factors that affect them and investigation of metabolic interactions between the two. The first aspect is relatively easy to investigate and attracts much interest. The second and third are difficult to work on and hence understudied. The review aims to stimulate interest in them by indicating methods of investigation and describing some successful studies. After discussing host factors that determine growth in vivo consideration is given to factors that influence the production of the determinants of mucosal colonization, penetration, interference with host defence and damage to the host. The final section deals with the influence of host-derived cytidine 5' -monophospho-N-acetylneuraminic acid and lactate on the pathogenicity of gonococci, meningococci and Haemophilus il1f!uel1zae. Key words: host factors - bacterial behaviour in vivo - CMP-NANA, lactate
Introduction Bacterial pathogens have unique biological properties which enable them to colonize mucous surfaces, penetrate them, grow in the environment of the host, inhibit or avoid host defences and damage the host. The bacterial products responsible for these five biological requirements are the determinants of pathogenicity (virulence). Current knowledge comes mainly from studies in vitro but interest is increasing in how bacteria behave and produce virulence determinants within the infected host. There are three aspects: observations on the bacteria and identification of virulence determinants formed during infection; recognition of host factors that affect this bacterial behaviour; and investigation of metabolic interactions between them.
The first aspect is easy to investigate and attracts much interest. New methods have been designed for looking at bacterial activity in vivo, e. g. in vivo expression technology (IVET) for detecting genes expressed in vivo and signature-tagged mutagenesis (STM) for identification of virulence genes in vivo. Much knowledge is accumulating from their use (Dorman et aI., 2000). The second and third aspects are difficult to work on and hence are understudied. Most references to the influence of host factors are made in relation to the environment in vivo as a whole, rather than to specific factors and their changes during infection (Dorman et ai., 2000). This review aims to stimulate interest in identifying such factors and studying their interactions with bacterial pathogens.
Corresponding author: Professor Harry Smith, The Medical School, University of Birmingham, Birmingham B15 2TT, UK, Phone: +44 (0.)121-414-6920., Fax: +44 (0.)121-414-5925 1438-4221100/290/3-20.7 $ 12.00/0
208
H. Smith
Host factors that determine bacterial growth in vivo The multifactorial nature of pathogenicity means that the determinants of all five requirements are essential for its manifestation. However, growth holds the primary position because without it, other determinants would not be formed. The new methods for recognizing genes expressed in vivo (Dorman et al., 2000) are underlining its importance; many of these genes are involved in nutrient acquisition and metabolism. There are several methods for recognizing nutrients that determine growth in vivo. The first depends on the fact that most key nutrients e. g. iron, are also necessary for growth in vitro. Pathogens are grown in defined media and key nutrients identified by lack of growth when they are deleted. Their presence in vivo is often well documented in the literature (Lentner, 1981, 1984) or can be ascertained by analysis of tissue samples. Another method is to see if tissue extracts stimulate growth of the pathogen in minimal media and then to purify the stimulant. Erythritol, which facilitates rapid growth of Brucella abortus in foetal tissues and fluids of pregnant cattle thereby causing abortion (Smith et al., 1962; Williams et al., 1964; Anderson and Smith, 1965; Keppie et al., 1965) was discovered in this way. Nutrient availability in vivo can be indicated by auxotrophic mutants. Auxotrophs of Salmonella typhi, Salmonella typhimurium and Shigella {lexneri unable to synthesize p-aminobenzoic acid, purines, pyrimidines and histidine have low virulence for mice, rabbits, monkeys or man (Ahmed et al., 1990; Curtiss et al., 1988; Fields et al., 1986; Kamell et al., 1993; Leuing and Finlay, 1991; Levine et al., 1987; O'Callaghan et al., 1988). Their impaired growth indicates absence or scarcity of the nutrients in these animals. Finally, the new methods for identifying genes expressed in vivo that are involved in acquiring and metabolizing nutrients have provided a powerful tool. The enzymes coded by the genes indicate nutrients used in vivo. In STM studies of staphylococci and Vibrio cholerae in mice, virulence genes were identified which indicated that proline and biotin (respectively) are important for growth in vivo (Schwan et al., 1998; Chiang and Mekalanos, 1998). Two previously unrecognized genes of Escherichia coli, guaA and argC, were induced in urine indicating that synthesis of guanine and arginine is important in uropathogenesis (Russo et al., 1996). After identification, the concentrations of these key nutrients in uninfected tissues and the changes that occur during infection need to be ascertained. The first can be obtained easily, -either from the literature (Lentner 1981; 1984) or by analysis of tissue samples. The
second is extremely difficult. However, some parameters have been measured within infected cells. Quantitative fluorescence microscopy measured intraphagosomal pH in macrophages (Aranda et al., 1992) and lacZ reporter genes indicated the approximate levels of Ca 2 +, Fe 2 + and Mg2+ in tissue culture cells (Pollack et al., 1986; Garcia del Portillo et al., 1992). It should be kept in mind that as infection proceeds nutrients could be generated by the action of bacterial enzymes e. g. proteases on host high Mr components. The relevance of identified nutrients to infection in vivo should be tested. The virulence of the pathogen may be enhanced by including additional nutrient in the inoculum, e. g., biotin with V. cholerae in mouse intestine (Chiang and Mekalanos 1998). Also, mutants unable to use the nutrient may have reduced virulence, e. g., the above auxotrophs. The existence in various tissues of high osmolarity, low pH and anaerobic conditions, which could inhibit growth in vivo, can be ascertained from the literature or by analysis of tissue samples. Also, these adverse influences, and how they are overcome, may be indicated by the functions of genes that are expressed in vivo. IVET identified cadA as a gene expressed by V. cholerae in rabbit ileal loops and orally infected mice. This gene codes for lysine decarboxylase, an important element in the acid tolerance response of bacteria. Thus, its expression fits with resistance to low pH in the stomach and the fatty acid content of the ileum (Merrell and Camilli, 2000). The final questions for consideration are; how are nutrients metabolized in vivo and by what bacterial determinants; and how do bacteria compensate for nutrient deficiencies and hostile biochemical conditions? It is well-nigh impossible to investigate the metabolism of pathogens growing in infected tissues. The aim has to be to simulate in vitro an aspect of behaviour relevant to growth in vivo, e. g. a rapid doubling time, by addition of nutrients that appear relevant in vivo. Then, meaningful studies can be made by established methods of bacterial physiology. Knowledge of genes expressed in vivo and metabolic potential revealed by genomics (Huynen et al., 1999) may be helpful. Genomics aided identification of the transport protein for bacterial acquisition of zinc for the metalloproteins in DNA and RNA polymerases (Lewis et al., 1999). This section ends with a summary of the well known and erudite studies on iron acquisition in vivo to illustrate that all the steps outlined above are achievable. In vitro studies showed iron is an essential nutrient for all bacJeria. Its availability in vivo is restricted by chelation to host transferrin and lactoferrin, and injection of iron salts enhances virulence (Bullen, 1981). Using iron-limiting conditions in vitro, molecular studies revealed strategies for overcoming iron restriction (Wein-
Host factors on bacterial behaviour in vivo
berg, 1995). Siderophores excreted by some pathogens return to bacteria via cell-surface receptors to give up their iron under the action of reductases (Halle and Meyer, 1992; Weinberg, 1995). In other cases, transferrin-bearing iron interacts with cell wall receptors and delivers iron into the bacteria (Cornelissen et aI., 1992). The cell wall receptors were demonstrated on bacteria in patients or infected animals (Brown and Williams, 1985; Cornelissen et aI., 1992; Smith 1990; 1996). Mutants deficient in determinants of iron acquisition were attenuated in virulence tests (Cornelissen et aI., 1997).
Host factors that affect virulence determinant production Two types of host factors influence production of the determinants of mucosal colonization, penetration, interference with host defence and damage to the host. The first are the substrates for production of the determinants. The second are the factors that provide the environment for operation of the requisite regulatory processes. In both cases, we are referring to factors additional to those required for growth. Relevant substrates can be recognized when comparisons of in vivo and in vitro grown bacteria have revealed a biological property appertaining to virulence, e. g. resistance to serum killing, possessed by the former but not by the latter. Bacteria are then grown in media supplemented by plasma or host cell extracts in the hope that the biological property is reproduced in vitro and can be used as an assay for purification of the responsible host factor. Examples of this approach are described in the final section. There are two methods for studying factors in the environment in vivo which affect regulation of virulence determinant production. The first is similar to that for recognizing nutrients involved in growth. The presence and concentrations of potential host factors (ions, metabolites, osmolarity, pH, and Eh ) in the tissues that become infected are obtained from the literature or by analysis. Then, attempts are made to mimic in vitro the conditions in vivo and observe effects on virulence determinant production and regulation. Up to the present, most studies have concerned intestinal conditions. Six invasion genes of S. typhimurium were maximally expressed in vitro at an oxygen tension, osmolarity and pH likely to exist in the ileum (Bajaj et aI., 1996). Two type III secretion genes of S. typhi, invG and prgH, were induced by high osmolarity, anaerobic conditions and pH 6,$ (Leclerc et aI., 1998). The Yst regulated toxin of Y. enterocolitica is only produced in vitro at temperatures below 30°C, but, 'at an osmolar-
209
ity and pH similar to those in the ileum, it is formed at 3]oC (Mikulskis et aI., 1994). EspB, the protein of E. coli which triggers processes resulting in effacement of microvilli and cytoskeletal changes, is maximally secreted at the pH, temperature, and osmolarity of the intestine (Kenny et a!., 1997). Aeromonas hydrophila produces rough lipopolysaccharide (LPS) in cultures at 37°C but smooth LPS and increased virulence at the osmolarity of the ileum (Aguilar et aI., 1997). The second approach is to investigate whether the environmental conditions that affect regulons in vitro exist and operate in vivo. Unfortunately, it has been successful only in a few cases. Mg2+ concentrations appear to control the PhoPlPhoQ system of S. typhimurium in vivo as in vitro. Within phagosomes, where the bacteria reside and resist intracellular killing, the Mg2+ concentration (estimated 20-100 ~M) is permissive for phoPiphoQ expression, unlike the high concentrations (0.5-1.0 mM) found in cytosols and body fluids (Garcia Vescovi et a!., 1996). Also, 37°C is permissive for expression of virB and virF and hence the invasion proteins of Shigella spp. (Maurelli and Sansonetti, 1988). However, for the BvgAS system which regulates production of many virulence determinants of Bordetella spp., it is controlled in vitro by MgS0 4 , nicotinic acid derivatives and temperature but in vivo only the permissive temperature (3]OQ is known to be relevant (Melton and Weiss, 1993; Martinez de Tejada et a!., 1996). Studies of the effect of different concentrations of MgS04 and various nicotinic acid derivatives in vitro suggested that the latter may have an affect in vivo but this was not proven (Melton and Weiss, 1993). The position regarding environmental regulation of the ToxRf[oxS system in vivo is even more obscure (Skorupski and Taylor, 1997; DiRita et a!., 2000). In vitro, V. cholerae forms toxin maximally at low temperature, under aerobic conditions and at low pH; and bile salts are repressive. In the human intestine, toxin production occurs at 37°C, under anaerobic conditions, at high pH and in the presence of bile salts . .
The influence of host-derived cytidine S'-monophospho-N-acetyl neuraminic acid (CMP-NANA) and lactate on the pathogenicity of gonococci, meningococci and Haemophilus influenzae CMP-NANA is the donor for sialylation of gonococcal LPS which confers resistance to killing by serum (Smith et a!., 1995; Vogel and Frosch, 1999). Lactate stimulates gonococcal growth, metabolism and virulence determinant production in a medium containing glucose, i. e. the conditions occurring in vivo (Parsons et a!.,
210
H. Smith
1996a, b; Gao et al., 1998). These factors were discovered by investigating why gonococci in urethral exudates are resistant to killing by fresh human serum and lose their resistance on subculture. Incubation with blood cell extracts restored resistance to serum killing, and fractionation showed that the active compound was CMP-NANA (Smith et al., 1995). Sodium dodecyl sulphate polyacrylamide electrophoresis on gonococci grown with CMP_ 14 CNANA showed some LPS components were sialylated. When sialyl groups were removed by neuraminidase reversion to serum sensitivity resulted. Sialylation of LPS interfered with the following host defence mechanisms: adsorption of complement component C3; ingestion and killing by neutrophils; killing by antibody against gonococcal porin 1B; and stimulation of the immune response, but sialylation prevented invasion of epithelial cells. A sialyltransferase was demonstrated in gonococcal extracts, and a sialyltransferase-deficient mutant, in contrast to its parent, did not become serum resistant when incubated with CMP-NANA (Bramley et al., 1995) nor resistant to the other host defence mechanisms mentioned above (Gill et al., 1996). Blood cell extracts enhanced the ability of CMPNANA to sialylate LPS and to induce serum resistance. The factor concerned was purified and shown to be lactate (Parsons et al., 1996a); minute quantities were active in a defined medium containing high concentrations of glucose. Lactate alone did not induce serum resistance and its action is separate from that of CMPNANA, because enhanced sialylation occurred when gonococci were pretreated with lactate (Parsons et al., 1996b). Lactate markedly increased LPS, protein and ribose synthesis (Gao et al., 1998). Increased LPS synthesis could explain enhancement of sialylation because more receptors become available. The changes also occurred in gonococci when glucose and lactate concentrations were adjusted to levels akin to those in vivo (Gao et al., 1998). This evidence of general stimulation of gonococcal metabolism is backed by other observations. Lactate increased oxygen consumption by gonococci in a salt solution containing glucose (Britigan et al., 1988); and in the defined medium containing glucose, gonococcal growth was faster with lactate which was metabolized concomitantly with glucose and more rapidly (Regan et al., 1999). The increased growth and metabolism might have been due to general stimulation of nutrient uptake but no evidence for this was obtained from uptake of [14C]adenine, [14C]glucose and [14C]proline (Gao et al., 2000). When gonococci were grown with 14C-Iabelled lactate aerobically in the defined medium containing glucose and lactate, label was concentrated in a low-Mr component X and LPS. Some label was detected in a few proteins. Thin layer chromatography showed that
component X was membrane lipid. N-terminal sequencing of the three proteins most heavily labelled by [14C]lactate showed one was GroEL and another porin 1B (Yates et al., 1999). Production of these proteins probably increases along with the general stimulation of protein synthesis that occurs with lactate (Gao et al., 1998) but, unlike for LPS, specific measurements have not been made. Gonococcal lipids are complex mixtures of ph osphatidyl ethanolamines and glycerols containing 4 principal fatty acids (Sud and Feingold, 1975; Senff et al., 1976). Nuclear magnetic resonance spectroscopy on lipid extracted from gonococci grown with [13C]lactate and unlabelled glucose, showed that lactate carbon was in fatty acid components, not ethanolamine/glycerol moieties, in contrast to when [13C]glucose was used alone or with unlabelled lactate (Yates et al., 1999). The pattern of glucose carbon incorporation changes when lactate is present. More carbon is incorporated into lipid, and different fatty acids may be formed. Lactate seems to have two overlapping functions in gonococcal pathogenicity; stimulation of growth and metabolism, and increased production of LPS and possibly other virulence determinants. GroEL, a chaperone, ensures correct folding during increased protein synthesis; porin 1B plays a metabolic role in membrane transport; and stimulation of lipid synthesis aids membrane formation. As regards virulence determinants, LPS synthesis is increased and GroEL is another stimulator of inflammatory cytokines (Coates and Henderson, 1998). Porin 1B inserts into membranes of neutrophils and impairs their action (Bjerknes et al., 1995). The LPS of some meningococci is sialylated, endogenously for sero groups B, C, Wand Y and exogenously, by CMP-NANA, for groups A and 29E (Smith et al., 1995). Also, the degree of sialylation of the former can be increased by incubation with exogenous CMPNANA (Estabrook et al., 1997). LPS sialylation affects several aspects of pathogenicity that are similar to those described above for gonococci but not as markedly, because meningococci have another virulence determinant, capsular polysaccharide (Smith et al., 1995). Sometimes it is difficult to distinguish between their respective roles. For example, the position over serum resistance is equivocal; some papers indicate sialylation of LPS is important (Estabrook et al., 1997; Kahler et al., 1998), others not (Vogel and Frosch, 1999). Unfortunately, the effect of lactate on meningococci has not yet been studied. The LPS of some strains of H. influenzae is also sialylated and it increases serum resistance (Mandrell et al., 1992; Hood et al., 1999). Sialylation can be increased by incubation with exogenous CMP-NANA or
Host factors on bacterial behaviour in vivo
sialic acid for some strains (Hood et al., 1999) but not for others (Kuratana et al., 1989). Like meningococci, H. influenzae has a capsule which contributes to virulence. Organisms in blood samples and nasal washes are·more resistant to killing by fresh serum than organisms grown in broth (Shaw et al., 1976; Rubin and Moxon 1985). Incubation of the latter organisms for 30-40 minutes with a low-Mr filtrate of serum or nasal washings made them serum resistant; this was lost on broth subculture (Shaw et al., 1976). These organisms were also resistant to other defence mechanisms and more virulent for rats (Kuratana et al., 1991). There are two mechanisms of serum resistance (Kuratana et al., 1990). In one, both capsular deficient mutants and capsulated wild types were converted to resistance and LPS content increased. In the second, only capsular wild types were converted to resistance and capsular polysaccharide increased. Substitution of the serum filtrate by a buffer containing lactate, glucose, urea and bicarbonate produced the first mechanism of resistance induction; and a mixture of lactate and Ca 2 + the second (Kuratana et al., 1991). In the first case, lactate may play a similar role to that for gonococci, namely, stimulation of LPS synthesis thereby producing more receptor for sialylation and hence increased serum resistance. For H. influenzae, the sialylation occurs endogenously 'for the strains concerned, so that lactate alone could induce serum resistance. For gonococci, exogenous CMP-NANA is needed for lactate to have an effect. In the second mechanism lactate may be a carbon source for increased production of capsular polysaccharide. It is fascinating that LPS sialylation plays a similar J role in the pathogenicity of gonococci, meningococci and H. influenzae and that lactate also has a strong influence, at least for two of them. Such mechanisms may operate for other pathogens. Acknowledgements. I am indebted to]. A. Cole, M.J. Gill, N. ]. Parsons, C. W. Penn, N. Woodward and E. A. Yates for reading the manuscript.
References Aguilar, A., Merino, S., Rubines, X., Tomas,]. M.: Influence of osmolarity on lipopolysaccharides and virulence of Aeromonas hydrophila serotype 0: 34 strains grown at 3rc. Infect. Immun. 65, 1245-1250 (1997). Ahmed, Z. u., Serker, M. R., Sack, D. A.: Protection of adult rabbits and monkeys from lethal shigellosis by oral immunization with thymine-requiring and temperaturesensitive mutants of Shigella {lexneri. Vaccine 8, 153-158 (1990) . Anderson, ]. D., Smith: H.: The metabolism of erythritol by Brucella abortus.]. Gen. Microbiol. 38, 100-124 (1965).
211
Aranda, C. M. A., Swanson, J. A., Loomis, W. P., Miller, 5.1.: Salmonella typhimurium activates virulence gene transcription within acidified macrophage phagolysomes. Proc. Nat!. Acad. Sci. USA 89, 10079-10083 (1992). Bajaj, V., Lucas, R. L., Hwang, c., Lee, C. A.: Coordinate regulation of Salmonella typhimurium invasion genes by environmental and regulatory factors is mediated by control of hilA expression. Mol. Microbiol. 22, 703-714 (1996). Bjerknes, R. , Guttormsen, H. , Solberg, C. 0., Wetzler, L. M.: Neisserial porins inhibit human neutrophil actin polymerization, degranulatation, opsonin receptor expression and phagocytosis but prime the neutrophils to increase their oxidative burst. Infect. Immun. 63, 160-167 (1995). Bramley, J., Demarco de Hormaeche, R., Constantinidou, c., Nassif, X., Parsons N.]., Jones, P., Cole,]. A., Smith, H.: A serum-sensitive, sialyltransferase-deficient mutant of Neisseria gonorrhoeae defective in conversion to serum resistance by CMP-NANA or blood cell extracts. Microb. Pathog. 18, 187-195 (1995). Britigan, B. E., Klapper, D., Svendsen, T., Cohen, M.S.: Phagocyte-derived lactate stimulates oxygen consumption by Neisseria gonorrhoeae. ]. Clin. Invest. 81, 318-324 (1988). Brown, M. R. W., Williams, P.: The influence of environment on envelope properties affecting survival of bacteria in infections. Anml. Rev. Microbiol. 39,527-556 (1985). Bullen, J. J.: The significance of iron in infection. Rev. Infect. Dis. 3,1127-1138 (1981). Chiang, S. L., Mekalanos,].].: Use of signature-tagged transpo son mutagenesis to identify Vibl'io cholerae genes critical for colonization. Mol. Microbiol.27, 797-805 (1998). Coates, A. R. M., Henderson, B.: Chaperonins in health and disease. Ann. N. Y. Acad. Sci. 851,48-53 (1998). Cornelissen, C. N., Biswas, G. D., Tsai, J., Parachuri, D. K., Thompson, S. A. Sparling, P. E: Gonococcal transferrinbinding protein 1 is required for transferrin utilization and is homologous to Ton-B-dependent outer membrane receptors. ]. Bacteriol. 174,5788-5797 (1992). Cornelissen, C. N., Kelly, M., Hobbs, M. M., Anderson,]. E, Cannon, ]. G., Cohen, M. S., Sparling P. E: The transferrin receptor expressed by gonococcal strain FA1090 is required for experimental infection of human male volunteers. Mol. Microbiol. 27, 611-616 (1997). Curtiss, R. III, Kelly, S. M., Gulig, P. A., Gentry-Weeks, C. R., Galan, j . E.: Avirulent salmonellae expressing virulence antigens from other pathogens for use as orally administered vaccines. In: Virulence mechanisms of bacterial pathogens (J. A. Roth, ed.) pp. 311-328. American Society for Microbiology Press, Washington DC 1988. DiRita, V. ]., Engleberg, N. c., Heath, A., Miller, A., Crawford, ]. A., Yu, R.: Virulence gene regulation inside and outside. Phil. Trans. R. Soc. Lond. B 355, 657-665 (2000). Dorman, c.]., Dougan, G., Holden, D. W., Williams, P., Smith, H.: The activities of bacterial pathogens in vivo. Phil. Trans. R. Soc. Lond. B 355, 547-712 (2000). Estabrook, M. M ., Griffiss, J. M., Jarvis, G. A.: Sialylation of Neisseria meningitidis lipooligosaccharide inhibits serum bactericidal activity by masking lacto-N-neotetraose. Infect. Immun. 65, 4436-4444 (1997) . <
212
H. Smith
Fields, P. 1., Swanson, R. V., Haidaris, C G., Heffron, F.: Mutants of Salmonella typhimurium that cannot survive within macrophages are avirulent. Proc. Nat!. Acad. Sci. USA 83, 5189-5193 (1986). Gao, L., Parsons, N.]., Curry, A., Cole,]. A., Smith, H.: Lactate causes changes in gonococci including increased lipopolysaccharide synthesis during short-term incubation in media containing glucose. FEMS Microbiol. Lett. 169, 309-316 (1998). Gao, L., Linden, L., Parsons, N.]., Cole,]. A., Smith, H.: Uptake of metabolites by gonococci grown with lactate in a medium containing glucose: evidence for a surface location of the sialyltransferase. Microb. Pathog. 28, 257266 (2000). Garcia-del Portillo, F., Foster, ]. W., Maguire, M. E., Finlay, B. B.: Characterization of the microenvironment of Salmonella typhimurium containing vacuoles within MDCK epithelial cells. Mol. Microbiol. 6, 3289-3297 (1992). Garcia Vescovi, E., Soncini, F. C, Groisman, E. A.: Mgz+ as an extracellular signal: environmental regulation of Salmonellaevirulence. Cell 84, 165-174 (1996). Gill, M. J., McQuillen, D. P., van Putten, ]. P. M., Wetzler, L. M., Bramley, J., Crooke, H., Parsons, N. J., Cole, ]. A., Smith, H.: Functional characterization of a sialyltransferase-deficient mutant of Neisseria gonorrhoeae. Infect. Immun. 64,3374-3378 (1996). Halle, F., Meyer,]. M.: Iron release from ferrisiderophores. A multi-step mechanism involving a NADHlFMN oxidoreductase and chemical Areduction by FMNH z. Eur.]. Biochern. 209, 621-627 (1992). Hood, D. W., Makepeace, K., Deadman, M. E., Rest, R. F., Thibauld, P., Martin, A., Richards, ]. C, Moxon, E. R.: Sialic acid in the lipopolysaccharide of Haemophilus in{luenzae: strain distribution, influence on serum resistance and structural characterization. Mol. Microbiol. 33, 679692 (1999). Huynen, M. A., Dandekar, T., Bork, P.: Variation and evolution of the citric-acid cycle: a genomic perspective. Trends Microbiol. 7,281-291 (1999). Kahler, C M., Martin, L. E., Shih, G. C, Rahman, M. M., Carlson, R. W., Stephens, D. S.: The (u2 ~ 8)linked polysialic acid capsule and lipooligosaccharide structure both contribute to ability of serogroup N Neisseria meningitidis to resist the bactericidal activity of normal human serum. Infect. Immun. 66, 5939-5947 (1998). Karnell, A., Cam, P. D., Verma, N., Lindberg, A. A.: AroD deletion attenuates Shigella {lexneri strain 2457T and makes it a safe and efficacious oral vaccine in monkeys. Vaccine 8, 830-836 (1993). Kenny, B., Akio, A., Stein, M., Finlay, B. B.: Enteropathogenic Escherichia coli protein secretion is induced in response to conditions similar to those in the gastrointestinal tract. Infect. Immun. 65, 2606-2612 (1997). Keppie, ]., Williams, A. E., Witt, K., Smith, H.: The role of erythritol in the tissue localization of the brucellae. Brit. ]. Exp. Pathol. 46,104-108 (1965). Kuratana, M., Anderson, P.: Host metabolites that phenotypically increase the resistance of Haemophilus in{luenzae type b to clearance mechanisms.]. Infect. Dis. 163, 10731079 (1991).
Kuratana, M., Hansen, E.]., Anderson, P.: Multiple mechanisms in serum factor-induced resistance of Haemophilus in{luenzae type b antibody. Infect. Immun. 58, 914-917 (1990). Kuratana, M., Inzana, T.]., Anderson, P.: Source of lowmolecular-weight host factors that phenotypically increase the resistance of Haemophilus in{luenzae Type b to bacteriolysis: nonidentity of a factor active in gonococci. Microb. Pathog. 7, 73-77 (1989). Leclerc, G.]., Tartera, C, Metcalf, E. S.: Environmental regulation of Salmonella typhi invasion-defective mutants. Infect. Immun. 66,682-691 (1998). Lentner, C: Units of Measurement, Body Fluids, Composition of the Body, Nutrition. In Geigy Scientific Tables, vol. 1., Ciba Geigy, Basle 1981. Lentner, C: Physical Chemistry, Composition of the Blood, Haematology, Sonatometric Data. In Geigy Scientific Tables, vol. 3., Ciba Geigy, Basle 1984. Leuing, K. Y., Finlay, B. B.: Intracellular replication is essential for virulence of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 88, 11470-11474 (1991). Levine, M. M., Herrington, D., Murphy, J. R., Morris, ]. G., Losonsky, G., Tall, B., Lindberg, A. A., Svenson, S., Bagar, S., Edwards, M. F., Stocker, B.: Safety, infectivity, immunogenicity and in vivo stability of two attenuated auxotrophic mutant strains of Salmonella typhi 541 Ty and 543Ty used as oral vaccines for man.]. Clin. Invest. 79, 888-902 (1987). Lewis, D. A., Klesney-Tait, ]., Lumbley, S. R., Ward, C K., Latimer, ]. L., Ison, C A., Hansen, E.].: Identification of the znuA-encoded periplasmic zinc transported protein of Haemophilus ducreyi. Infect. Immun. 67, 5060-5068 (1999). Mandrell, R. E., McLaughlin, R., Kwaik, Y. A., Lesse, A., Yamasaki, R., Gibson, B., Spinola, S. M., Apicella, M. A.: Lipooligosaccharides (LOS) of some Haemophilus species mimic human glycosphingolipids, and some LOS are sialylated. Infect. Immun. 60, 1322-1328 (1992). Martinez de Tejada, G., Miller,]. F., Cotter, P. A.: Comparative analysis of virulence control systems of Bordetella pertussis and Bordetella bronchisepticus. Mol. Microbiol. 22, 895-908 (1996). Maurelli, A. T., Sansonetti, P.].: Identification of chromosome genes controlling temperature-regulated expression of Shigella virulence. Proc. Nat!. Acad. Sci. USA 85, 2820-2824 (1988). Melton, A. R., Weiss, A. A.: Characterization of environmental regulators of Bordetella pertussis. Infect. Immun. 61, 807-815 (1993). Merrell, D. S., Camilli, A.: Detection and analysis of gene expression during infection by the in vivo expression technology. Phil. Trans. R. Soc. Lond. B 355, 587-599 (2000). Mikulskis, A. V., Delor, I., Thi, V. H., Cornel is, G. R.: Regulation of the Yersinia enterocolitica enterotoxin Yst gene. Influence of growth phase, temperature, osmolarity, pH and bacterial host factors. Mol. Microbiol. 14,905-915 (1994). O'Callaghan, D., Maskell, D., Liew, F. Y., Easmon, C S. F., Dougan, D.: Characterization of aromatic- and purinedependent Salmonella typhimurium; attenuated persis-
Host factors on bacterial behaviour in vivo tence, and ability to induce protective immunity In BALB/c mice. Infect. Immun. 56,419-423 (1988). Parsons, N.]., Boons, G.J., Ashton, P. R., Redfern, P. D., Quirk, P., Gao, Y., Constantinidou, c., Patel, J., Bramley, J., Cole, ]. A., Smith, H.: Lactic acid is the factor in blood cell extracts which enhances the ability of CMP-NANA to sialylate gonococcal lipopolysaccharide and induce serum resistance. Microb. Pathog. 20, 87-100 (1996a). Parsons, N. J., Emond, ]. P., Goldner, M., Bramley, J., Crooke, H., Cole, J. A., Smith, H.: Lactate enhancement of sialylation of gonococcal lipopolysaccharide and induction of serum resistance by CMP-NANA is not due to direct activation of the sialyltransferase: metabolic events are involved. Microb . Pathog. 21, 193-204 (1996b). Pollack, c., Straley, S. c., Klempner, M. S.: Probing the phagolysosome environment of human phagocytes with a Ca 2 +-responsive operon fusion in Yersinia pestis. Nature (Lond) 332, 834-836 (1986). Regan, T., Watts, A., Smith, H., Cole, J. A.: Regulation of the lipopolysaccharide-specific sialyltransferase activity of gonococci by growth state of the bacteria, but not by carbon source, catabolite repression or oxygen supply. Antonie Van Leeuwenhoek 75, 369-379 (1999). Rubin, L. G., Moxon, E. R.:The effect of serum-factor induced resistance to somatic antibodies on the virulence of Haemophilus inf/uenzae type b. J. Gen. Microbiol. 131, 515-520 (1985). Russo, T. A., Jadush, S. T., Brown, J.J., Johnson, J. R.: Identification of two previously unrecognised genes (guaA and argC) important for uropathogenesis. Mol. Microbiol. 22,217-228 (1996). Schwan, W. R., Coulter, S. N., Ng, E. Y. W., Langhorne, M . H., Ritchie, H. D., Brody, L. L. , West brock-Wadman, S., Bayer, A. S., Folger, K. R., Stover, C. K.: Identification and characterization of the PutP proline permease that contributes to in vivo survival of Staphylococcus au reus in animal models. Infect. Immun. 66, 567-572 (1998). Senff, L. M., Wegener, W. S., Brooks, G. E, Finnerty, W. R., Makula, R. A.: Phospholipid composition and phospho-
213
lipase A activity of Neisseria gonorrhoeae. ]. Bacteriol. 127,874-880 (1976). Shaw, S., Smith, A. L., Anderson, P., Smith, D. H.: The paradox of Haemophilus inf/uenzae type b bacteremia in the presence of serum bactericidal activity. J. Clin. Invest. 58, 1019-1029 (1976). Skorupski, K., Taylor, R. K.: Control of the ToxR virulence regulon in Vibrio cholerae by environmental stimuli. Mol. Microbiol. 25, 1003-1009 (1997). Smith, H .: Pathogenicity and the microbe in vivo. J. Gen. Microbial. 136,377-393 (1990). Smith, H .: What happens in vivo to bacterial pathogens? Ann. N. Y. Acad. Sci. 797, 77-92 (1996 ). Smith, H., Parsons, N.J., Cole, J. A.: Sialylation of neisserial lipopolysaccharide: a major influence on pathogenicity. Microb. Pathog. 19,365-377 (1995). Smith, H., Williams, A. E., Pearce, J. H., Keppie, J., HarrisSmith, P. W., Fitzgeorge, R. B., Witt, K.: Foetal erythritol: a cause of the tissue localization of Brucella abortus in bovine contagious abortion. Nature Lond. 193, 47-49 (1962). Sud, I.J., Feingold, D. S.: Phospholipids and fatty acids of Neisseria gonorrhoeae. J. Bacteriol.124, 713-717 (1975). Vogel, U., Frosch, M.: Mechanisms of neisserial serum resistance. Mol. Microbiol. 32, 1133-1139 (1999). Weinberg, E. D.: Acquisition of iron and other nutrients in vivo. In: Virulence mechanisms of bacterial pathogens, 2nd ed. (J. A. Roth, C. A. Bolin, R. A. Brogden, E C. Minion, M.J. Wannemuehler, eds.), pp. 79-93. American Society for Microbiology Press, Washington DC 1995. Williams, A. E., Keppie, ]., Smith, H.: The relation of erythritol usage to virulence in Brucellas.J. Gen. Microbiol. 37, 265-292 (1964). Yates, E. A., Gao, L., Parsons, N.J., Cole, ]. A. , Smith, H. : When lactate is used by gonococci growing in a medium containing glucose, as occurs in vivo, its carbon is incorporated into lipid, LPS, GroEL and porin lB. Abstract 37. A Royal Society Discussion Meeting on the 'Activities of bacterial pathogens in vivo'. Society for General Microbiology, Reading, UK 1999.