Department of Agricultural Sciences, Imperial College London, Wye campus, Ashford, Kent TN25 5AH, UK
Correspondence
Nicholas J. Russell
nicholas.russell{at}imperial.ac.uk
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Present address: Grain Quality and Nutrition Division, Bangladesh Rice Research Institute, Joydebpur, Gazipur-1701, Bangladesh.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We are interested in the mechanisms of membrane adaptation that are utilized by vegetative cells of B. cereus to enable them to grow once the spores have germinated. In particular, the vegetative cells must adapt to changing temperature and aw conditions. It is well known that in other bacteria the adaptations to temperature are mainly in the fatty acyl moieties of the membrane phospho(glyco)lipids (Russell, 1984, 1990
), whereas the response to low aw is a change in the lipid head group (Russell, 1989
, 1993
). Although such adaptive changes have been described for a wide range of bacteria, including bacilli, there is little published information available for B. cereus.
We have isolated a number of new strains of B. cereus from Bangladeshi rice, and have characterized their lipid composition and shown that all the strains are mesophilic and that their growth is stimulated by rice starch (Haque, 2002). Therefore, it was of interest to investigate the effects of stress due to lowered temperature and aw, as well as the influence of rice starch, on membrane lipid adaptation to understand better how this organism can grow in heated rice as it cools and cause food poisoning. In this paper, we show that not only do the different strains of B. cereus vary in their membrane lipid compositions, but that they do not respond in a uniform manner to the stresses of lowered temperature or aw. Moreover, the response to lowered aw is solute-specific. The expected changes in phospholipid fatty acyl branching and chain length during growth at reduced temperature were observed, but there were unexpected long-term changes in unsaturation. Growth in rice starch stimulated growth and modified phospholipid composition in a way that could not be explained by changes in growth rate or altered aw.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Preparation of soluble rice-starch medium.
Soluble rice starch was prepared by boiling 100 g rice in 1 l sterile water for approximately 17 min (according to cooking time), and the rice was removed by filtration through 12-fold cheese cloth. The filtrate was termed 100 % soluble rice starch and appropriate volumes were added to subtilis minimal medium, prepared according to Clowes & Hayes (1968), to give 70 % (v/v) soluble rice-starch medium; when required, this medium was solidified by the addition of agar (Difco; 1·5 %, w/v, final concn).
Extraction of lipid and quantification of phospholipid.
Cultures of B. cereus were grown in nutrient broth (Oxoid) at 37 or 15 °C in the presence or absence of NaCl or sucrose to lower aw. Bacteria were harvested in the late-exponential phase by centrifuging at 16 266 gav for 15 min (rav=10 cm) and the cell pellet was washed twice with 0·1 M phosphate buffer (pH 7·0). To test the effect of rice starch on lipid composition, bacteria were grown in subtilis minimal medium containing 70 % (v/v) soluble rice starch. Total lipids were extracted using the method of Bligh & Dyer (1959) as described by Kates (1986)
. Phospholipids were separated by two-dimensional thin-layer chromatography and quantified on the basis of their phosphorus content as described by Ntougias & Russell (2001)
.
Fatty acid analysis by capillary GC.
Fatty acid methyl esters (FAMEs) were produced from total lipid via transmethylation with 2·5 % (v/v) concentrated sulphuric acid in dry methanol. Hydrogenation of FAMEs was carried out to identify those fatty acids containing unsaturated bonds (Kates, 1986); double bond positions were not determined. The FAMEs were analysed by capillary GC using a Pye Unicam PU4500 gas chromatograph equipped with a Supelco SP2380 fused silica capillary column (length 30 m, i.d. 0·25 mm and film thickness 0·20 µm) and a Spectraphysics SP420 integrator. The carrier gas was nitrogen with a flow rate of 40 ml min1. Samples were analysed isothermally at 160 °C. The injector and detector temperatures were 250 °C. The peaks were identified by comparison of their retention times with those of a standard bacterial FAME mix (Supelco 4-7080) and confirmed by GC-MS when necessary.
Statistical analysis.
Student's t-test was used to determine the statistical significance of paired values, with 95 % being taken as the significant limit.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
When the B. cereus strains were grown in the presence of 40 % (w/v) sucrose to lower aw, the proportions of both PG and PE decreased, with a concomitant approximate doubling of the proportion of DPG, the only exception being strain BRS-533 in which PE did not increase (cf. Table 1 with Table 5
and supplementary data Table 1
). There were no major changes in the proportions of the minor components. The ratio of anionic to zwitterionic phospholipids was increased substantially at lower aw in all of the strains, except BR5-533, in which there was no significant change in the ratio even though the relative proportions of PG and DPG changed.
|
When NaCl (7 %, w/v) was added to the growth medium to lower aw, there was a decrease in the proportions of PG and PE in all strains except BR5-529, in which the proportion of PG did not change significantly (cf. Table 1 with Table 6
and supplementary data Table 3
). The proportion of DPG increased in all strains apart from BR5-529. As a result of these changes the ratio of anionic to zwitterionic lipids increased significantly in all strains, except BR28-535 in which it decreased.
|
Effect of rice starch on lipid composition
When the strains were grown in 70 % (v/v) rice-starch medium, the proportion of DPG approximately doubled, whereas the proportion of PG approximately halved, in all strains (cf. Table 1 with Table 5
and supplementary data Table 5
). The proportion of PE did not change significantly except in strains BR5-530 and BR5-532, in which it increased by 17 and 9 % respectively compared to controls grown in nutrient broth.
The presence of 70 % (v/v) rice starch in the growth medium resulted in approximately twofold increases in i15 : 0 and i17 : 0, whereas there were decreases in i16 : 0 and n16 : 0, but little or no change in a15 : 0 and a17 : 0 (cf. Table 3 with Table 6
and supplementary data Table 6
). There were some individual exceptions to this generalization for specific acids, in particular for strain BR5-533 in which the major fatty acid, i15 : 0, as well as a15 : 0, were decreased, whereas the proportion of i17 : 0 increased approximately fourfold. The proportion of branched-chain fatty acids was 1060 % higher in cultures grown in soluble rice-starch medium compared with control cultures or those grown in sucrose or NaCl medium (Table 6
). The major unsaturated fatty acid components in rice starch are n18 : 1 and n18 : 2, but the amount of n18 : 1 (usually present when cultures are grown in nutrient broth) did not increase and n18 : 2 could not be detected in the acyl lipids of any B. cereus strains grown in rice-starch-supplemented medium.
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Lowering of growth temperature generally does not have very much effect on phospholipid head-group composition, and, to our knowledge, there is no published information specifically in relation to B. cereus. The data in Table 2 show that in contrast to most other bacteria, there is a significant effect of lowered growth temperature on phospholipid composition in all except one of the B. cereus strains, with an approximate two- to threefold decrease in the ratio of anionic/zwitterionic phospholipids (cf. Tables 1 and 2
). Such a change will increase the propensity for the formation of non-bilayer phases, since the PE contains largely branched-chain fatty acyl chains that give the molecule a cone-shape (Macdonald et al., 1985
). However, a reduction in temperature will lower acyl chain fluidity and therefore the diameter of the lower dimension (i.e. distal to the head group) of the cone so that it was more cylindrical and therefore more likely to form a lamellar (bilayer) phase. Thus, one can speculate that the phospholipid head-group compositional change would counteract the effect of lowered temperature in terms of membrane fluidity and phase behaviour, but not of the change in surface charge, since the membrane lipid bilayer would have a shift towards a preponderance of zwitterionic rather than anionic phospholipids.
The fatty acid composition of the B. cereus strains in the present study was very similar to that reported for other strains of this bacterium (Kaneda, 1967, 1968
, 1969
). Although there were differences in the fatty acid compositions between individual strains, they all adapted to a lowering of growth temperature from 37 to 15 °C in the same qualitative manner. All of the changes, namely decreases in the proportion of branched-chain fatty acids and their equivalent chain-length, and increases in the anteiso-/iso-branched ratio and proportion of unsaturated fatty acids, would contribute to the maintenance of membrane fluidity at the lower growth temperature. The increase in unsaturation was particularly marked in being approximately ninefold (mean of the seven strains). Kaneda (1972)
observed a 20 % increase in unsaturated fatty acids in B. cereus grown at 21 °C compared with 37 °C, which is in good agreement with the values of 2037 % increases for the strains grown at a lower temperature of 15 °C. The cultures for lipid analysis were grown for many generations at the appropriate temperature, so the change in fatty acyl unsaturation is not a reflection of the short-term (transient) adaptation that is seen in some other Bacillus spp. (Fulco, 1984
). Nonetheless, the B. cereus strains display the full variety of fatty acid adaptive changes that are available to organisms containing straight- and branched-chain, saturated and unsaturated fatty acids.
The common bacterial adaptive strategy to low aw stress, whether caused by high cation concentrations or the addition of uncharged osmotically active solutes, is an increase in anionic membrane lipids to maintain the stability and integrity of the membrane bilayer (Russell, 1989; Sutton et al., 1990
, 1991
). In all of the B. cereus strains, grown at their optimum temperature at low aw, there is already a dominance of anionic phospholipids that would provide a high negative charge to the membrane lipid lamellar phase. The response of all B. cereus isolates grown in low aw due to sucrose or NaCl was to further increase the proportion of anionic lipids (cf. Tables 1 and 5
), although the magnitude of the changes was not particularly great. The effects of 7 % NaCl and 40 % sucrose on aw are roughly equivalent, but there was a greater or lesser effect of NaCl compared with sucrose for different strains. Strain BR28-535 was anomalous in that the proportion of anionic phospholipids decreased slightly on lowering aw with NaCl, despite the fact that its control phospholipid composition is not markedly different from other strains and it displays a normal response to sucrose. Similarly, strain BR5-533 did not significantly alter the proportion of anionic phospholipids in response to lowering of aw with sucrose, although there was a shift in the relative proportions of PG and DPG. Thus, the strains not only show distinct solute effects, but also different responses to the low aw stress. Such anomalous responses have been observed previously. For example, Jones et al. (2000)
found that a Bacillus sp. responded by increasing zwitterionic PE instead of an anionic phospholipid such as PG. In contrast, for another food-associated Bacillus sp., Cummings & Russell (1996)
observed that the adaptive response was similar to that of other bacteria, including halotolerant ones.
In the present study, most of the increase in anionic phospholipid was in DPG rather than PG, which actually decreased in most strains. The shift from PG to DPG could be the result of lowered growth rate, but it was not due to growth phase in batch culture because all cultures were harvested in late-exponential phase. In their study of Bacillus subtilis, López et al. (1998) found that growth in medium with aw lowered using NaCl also resulted in an increase of DPG at the expense of PG, and a rise in the amount of anionic lipid (due to increased glycolipid). The mechanism of the shift in anionic/zwitterionic lipid ratio in bacilli is not known, although it has been demonstrated in the halotolerant Salinivibrio costicola that salt stimulates PG biosynthesis rather than inhibits PE biosynthesis in order to raise the anionic/zwitterionic ratio (Russell et al., 1985
). It is generally believed that in bacteria PG/DPG (anionic) and PE (zwitterionic) are synthesized by separate routes (Harwood & Russell, 1984
; Heath et al., 2002
). However, information about bacterial lipid biosynthesis is based very largely on experiments with Escherichia coli, and Fulco (1984)
found that PG might be converted directly to PE in some Bacillus species, although it is not known if that mechanism occurs in the B. cereus strains studied here.
In common with other bacteria subjected to stress by low aw, the strains of B. cereus also modified the fatty acyl composition of the membrane lipids. The responses to low aw with NaCl or sucrose were not the same, and the various strains reacted differently. A comparison with some other Gram-positive bacteria reveals a similar variety of responses. For example, different Planococcus species respond in apparently opposite ways as far as branched-chain fatty acids are concerned (Miller, 1985; Monteoliva-Sanchez & Ramos-Cormenzana, 1987
), whilst in Staphylococcus aureus the proportion of branched-chain fatty acids increased (Kanemasa et al., 1972
). The variety of responses and whether they would fluidize the membrane at low aw are discussed by Russell (1993)
. In the B. cereus strains grown in NaCl medium there was no significant change in the proportion of branched-chain fatty acids, although there was generally an increase in the anteiso-/iso-branched ratio. Although no direct measurements have been made, one would predict that these changes in fatty acyl composition of the membrane phospholipids would give an increase in fluidity of the B. cereus strains. This response is similar to that of halotolerant bacteria, which also increase membrane fluidity by altering membrane fatty acid composition in response to low aw (Russell, 1993
). In contrast, in sucrose medium there was an approximately 31 % decrease in the proportion of branched fatty acids, but the anteiso-/iso-branched ratio did not alter except for two strains in which it increased i.e. the changes would generally be predicted to give a decrease in membrane fluidity. Thus there are clearly not only solute-specific effects, but also quite opposite membrane responses. However, caution must be exercised in making such predictions. In the related B. subtilis, growth in NaCl leads to changes in fatty acid composition that one would predict as being fluidizing (López et al., 1998
), but direct measurements of membrane fluidity using fluorescent probes showed that the opposite was the case (López et al., 2000
).
When the strains of B. cereus were grown in minimal medium containing soluble rice starch, there was very little change in the proportion of anionic lipids, although an even higher amount of DPG was present in the membranes compared with control cultures or those grown under conditions of low aw due to NaCl or sucrose. The observed changes are not due to being grown in minimal medium rather than in nutrient broth, as control experiments demonstrated (Haque, 2002). In addition, care was taken to harvest all cultures in the late-exponential phase of batch culture to avoid changes due to this factor. Neither can this shift in lipid composition be an adaptation to the stress of low aw, because the large molecular size of the starch molecules means that they exert little osmotic pressure in solution. This was tested directly and the aw value of 70 % rice-starch medium was measured as 1·00, i.e. the same as control (nutrient broth) medium. Nor can the increase in DPG relative to PG be due to a slower growth rate, because on the basis of plate counts of viable cells the strains grew faster as the amount of rice starch was increased to 70 % and only slowed at higher concentrations (Haque, 2002
). Presumably the lipid change is due to some other factors in the rice-starch medium. It is not due to uptake of DPG from the medium because rice starch does not contain this phospholipid. This was checked directly by extracting rice-starch medium. The major phospholipid component of rice grains was phosphatidylcholine (
81·0 %), together with smaller amounts of phosphatidylinositol (
5 %), PG (
5 %) and PE (
7 %), plus one unknown minor component (
1 %). As a further check, it was shown that B. cereus cultures grown in rice starch did not contain even traces of n18 : 2 fatty acid, which is a major fatty acyl component of lipids in rice-starch medium. It is not known what factor in the rice-starch medium triggered the additional conversion of PG to DPG in the B. cereus strains. In model systems in vitro, DPG forms non-bilayer structures under specific conditions, e.g. in the presence of Ca2+ (Vasilenko et al., 1982
). However, the rice-starch medium used in the present study did not contain additional amounts of Ca2+. Cleveland et al. (1976)
proposed that DPG protects cells from autolysis, and it also acts as a barrier to Na+ permeability (Kanemasa et al., 1972
). Perhaps, there are other ionic components in rice starch that acted as the trigger for DPG formation. There were no particularly consistent changes in the fatty acid composition in the different strains of B. cereus in response to growth in rice-starch medium, apart from a 35 % decrease in the anteiso-/iso-branched ratio, which would lower membrane fluidity. In view of the association of the emetic form of food poisoning with B. cereus growth in rice, it would be interesting to determine just what factor(s) is responsible for the stimulation of growth as this could inform food-handling practice and reduce the problem in the catering industry.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bulla, L. A., Bennett, J. R. G. A. & Shotwell, O. L. (1970). Physiology of the spore-forming bacteria associated with insects. II. Lipids of vegetative cells. J Bacteriol 104, 12461253.
Cleveland, R. F., Wicken, A. J., Daneo-Moore, L. & Shockman, G. D. (1976). Inhibition of wall autolysis in Streptococcus faecalis by lipoteichoic acid and lipids. J Bacteriol 126, 192197.[Medline]
Clowes, R. C. & Hayes, W. (1968). Appendix A, media. In Experiments in Microbial Genetics. Edited by R. C. Clowes & W. Hayes. Oxford: Blackwell.
Cummings, S. P. & Russell, N. J. (1996). Osmoregulatory responses in bacteria isolated from fresh or composted olive-mill waste waters. World J Microbiol Biotechnol 12, 6167.
Fulco, A. J. (1984). Regulation and pathways of membrane lipid biosynthesis in bacilli. In Biomembranes, vol. 12, pp. 303327. Edited by M. Kates & L. Manson. New York & London: Plenum.
Granum, P. E. (1994). Bacillus cereus and its toxins. J Appl Bacteriol 76 (suppl.), 615665.
Haque, M. A. (2002). Characterisation of Bacillus cereus strains in Bangladeshi rice. PhD thesis. University of London, UK.
Harwood, J. L. & Russell, N. J. (1984). Lipids in Plants and Microbes. London: George Allen and Unwin.
Heath, R. J., Jackowski, S. & Rock, C. O. (2002). Fatty acid and phospholipid metabolism in prokaryotes. In Biochemistry of Lipids, Lipoproteins and Membranes, pp. 5592. Edited by D. E. Vance & J. E. Vance. Amsterdam: Elsevier.
Houtsmuller, U. M. T. & van Deenen, L. L. M. (1963). Identification of a bacterial phospholipid as an O-ornithine ester of phosphatidylglycerol. Biochim Biophys Acta 70, 211213.[CrossRef][Medline]
Jones, C. E., Murphy, P. J. & Russell, N. J. (2000). Diversity and osmoregulatory responses of bacteria isolated from two-phase olive oil extraction waste products. World J Microbiol Biotechnol 16, 555561.[CrossRef]
Kaneda, T. (1967). Fatty acids in the genus Bacillus. I. Iso- and anteiso-fatty acids are characteristic constituents of lipids in 10 species. J Bacteriol 93, 894903.[Medline]
Kaneda, T. (1968). Fatty acids in the genus Bacillus. II. Similarity in the fatty acid compositions of Bacillus thuringiensis, Bacillus anthracis and Bacillus cereus. J Bacteriol 95, 22102216.[Medline]
Kaneda, T. (1969). Fatty acids in Bacillus larvae, Bacillus lentimorbus and Bacillus popilliae. J Bacteriol 98, 143146.[Medline]
Kaneda, T. (1972). Positional preferences of fatty acids in phospholipids of Bacillus cereus and its relation to growth temperature. Biochim Biophys Acta 280, 297305.[Medline]
Kanemasa, Y., Yoshioka, T. & Hayashi, H. (1972). Alteration of the phospholipid composition of Staphylococcus aureus cultured in medium containing NaCl. Biochim Biophys Acta 280, 444450.[Medline]
Kates, M. (1986). Techniques of Lipidology. Amsterdam: Elsevier.
Lang, D. R. & Lundgren, D. G. (1970). Lipid composition of Bacillus cereus during growth and sporulation. J Bacteriol 101, 483489.[Medline]
López, C. S., Heras, H., Ruzal, S. M., Sánchez-Rivas, C. & Rivas, E. A. (1998). Variations of the envelope composition of Bacillus subtilis during growth in hyperosmotic medium. Curr Microbiol 36, 5561.[CrossRef][Medline]
López, C. S., Heras, H., Garda, H., Ruzal, S. M., Sánchez-Rivas, C. & Rivas, E. A. (2000). Biochemical and biophysical studies of Bacillus subtilis envelopes under hyperosmotic stress. Int J Food Microbiol 36, 137142.
Macdonald, B. M., Sykes, B. D. & McElhaney, R. N. (1985). Fluorine-19 nuclear magnetic resonance studies of lipid fatty acyl chain order and dynamics in Acholeplasma laidlawii B membranes. Gel-state disorder in the presence of methyl iso and anteiso branched-chain substituents. Biochemistry 24, 24122419.[Medline]
Miller, K. J. (1985). Effects of temperature and sodium chloride concentration on the phospholipid and fatty acid compositions of a halotolerant Planococcus sp. J Bacteriol 162, 263270.[Medline]
Monteoliva-Sanchez, M. & Ramos-Cormenzana, A. (1987). Cellular fatty acid composition of Planococcus halophilus NRCC 14033 as affected by growth temperature and salt concentration. Curr Microbiol 15, 133136.
Nichols, G. L., Little, C. L., Mithani, V. & de Louvis, J. (1999). The microbiological quality of cooked rice from restaurants and take-away premises in the United Kingdom. J Food Prot 62, 877882.[Medline]
Ntougias, S. & Russell, N. J. (2001). Alkalibacterium olivoapovliticus gen. nov., sp. nov., a new obligately alkaliphilic bacterium isolated from edible-olive wash-waters. Int J Syst Evol Microbiol 51, 11611170.[Abstract]
Russell, N. J. (1984). Mechanisms of thermal adaptation in bacteria: blueprints for survival. Trends Biochem Sci 9, 108112.[CrossRef]
Russell, N. J. (1989). Adaptive modifications in membranes of halotolerant and halophilic microorganisms. J Bioenerg Biomembr 21, 93113.[Medline]
Russell, N. J. (1990). Cold adaptation of microorganisms. Philos Trans Roy Soc London B326, 595611.
Russell, N. J. (1993). Lipids of halophilic and halotolerant microorganisms. In the Biology of Halophilic Bacteria, pp. 163210. Edited by R. H. Vreeland & L. Hochstein. Boca Raton, FL: CRC Press.
Russell, N. J., Kates, M. & Kogut, M. (1985). Phospholipid biosynthesis in a moderately halophilic bacterium, Vibrio costicola, during adaptation to changing salt concentrations. J Gen Microbiol 131, 781789.
Sutton, G. C., Russell, N. J. & Quinn, P. J. (1990). The effect of salinity on the phase behaviour of purified phosphatidylethanolamine and phosphatidylglycerol isolated from a moderately halophilic eubacterium. Chem Phys Lipids 56, 135147.[CrossRef]
Sutton, G. C., Russell, N. J. & Quinn, P. J. (1991). The effect of salinity on the phase behaviour of total lipid extracts and bilayer mixtures of the major phospholipids isolated from a moderately halophilic eubacterium. Biochim Biophys Acta 1061, 235246.[Medline]
Vasilenko, I., de Kruijff, B. & Verkleij, A. J. (1982). Polymorphic phase behaviour of cardiolipin from bovine heart and from Bacillus subtilis as detected by 31P NMR and freeze-fracture techniques. Biochim Biophys Acta 689, 282286.
Received 15 September 2003;
revised 18 January 2004;
accepted 27 January 2004.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |