Effect of defined lipopolysaccharide core defects on resistance of ...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1981, p. 843-849 0099-2240/81/110843-07$02.00/0

Vol. 42, No. 5

Effect of Defined Lipopolysaccharide Core Defects on Resistance of Salmonella typhimurium to Freezing and Thawing and Other Stresses GEANNIE M. BENNETT, ALAN SEAVER, AND PETER H. CALCOTTt* Department of Biological Sciences, Wright State University, Dayton, Ohio 45435 Received 21 April 1981/Accepted 15 July 1981

A family of mutants of Salmonella typhimurium with altered lipopolysaccharide (LPS) core chain lengths were assessed for sensitivity to freeze-thaw and other stresses. Deep rough strains with decreased chain length in the LPS core were more susceptible to novobiocin, polymyxin B, bacitracin, and sodium lauryl sulfate during growth, to ethylenediaminetetraacetic acid and sodium lauryl sulfate in resting suspension, and to slow and rapid freeze-thaw in water and saline, and these strains exhibited more outer membrane damage than the wild type or less rough strains. Variations in the LPS chain length did not dramatically affect the sensitivity of the strains to tetracycline, neomycin, or NaCl in growth conditions or the degree of freeze-thaw-induced cytoplasmic membrane damage. The deeper rough isogenic strains incorporated larger quantities of less-stable LPS and less protein into the outer membrane than did the wild type or less rough mutants, indicating that the mutations affected outer membrane synthesis or organization or both. Nikaido's model of the role of LPS and protein in determining the resistance of gram-negative bacteria to low-molecular-weight hydrophobic antibiotics is discussed in relation to the stress of freeze-thaw.

The stress of freezing and thawing causes considerable damage to microbes, including loss of viability, membrane damage, deoxyribonucleic acid damage, ultrastructural changes, and loss of respiratory, active uptake, nutrient retention, and protein synthesis activities (3, 4, 6, 10, 11, 14, 15, 17, 23, 28, 31, 36, 44). In addition, wall damage (outer membrane) has been detected in gram-negative bacteria as evidenced by release of periplasmic proteins (14, 15), loss or alteration of lipopolysaccharide (LPS) (44,45), loss of outer membrane proteins (23), and increases in the permeability of the wall to certain enzymes (36, 44; A. Kohn and W. Szybalski, Bacteriol. Proc., p. 126-127, 1959), dyes (36, 44), detergents (10, 18, 36, 44), and large-molecular-weight material (15). This type of damage appears to play an important role in enumerating gram-negative bacteria in environmental situations, particularly stressful ones (5, 7-9, 19, 22, 27, 33, 38, 44, 50, 52). Recently, the outer membrane of a variety of gram-negative bacteria has been studied at the biochemical, physiological, and genetic levels (20, 21, 26, 29, 40, 42). In a number of systems defined mutants are available that are defective in one or more of the major components of the t Present address: CR-Bioproducts, Dow Chemical Co., Midland, MI 48640.

cell structure (21, 46). Examination of a family of Salmonella typhimurium mutants with defined mutations that result in altered LPS core structure has enabled us to determine the role played by the LPS in conferring resistance to the bacterium against freeze-thaw and other stresses. (Parts of this study were presented at the 81st Annual Meeting of the American Society for Microbiology, Dallas, Tex., March 1981.) MATERIALS AND METHODS Organisms and cultural conditions. S. typhi-

murium PC1 and a family of mutants were obtained from R. Roantree, Department of Microbiology, Stanford University, Calif., and are described in Table 1. They were grown at 37°C aerobically to late log-early stationary phase in nutrient broth (18). Organisms were harvested from the culture medium and washed by centrifugation and resuspension in 0.85% saline. Before treatment, samples were resuspended in water or saline. Freezing and thawing regimens. Populations in water or saline at approximately 1 mg dry weight per ml were subjected to slow (1 to 2°C/min) or rapid (100 to 200'C/min) cooling to -196°C. After 5 min at this temperature, the samples were thawed slowly (5 to 10'C/min) (12). After thawing, samples were serially diluted into 2 mM MgSO4 and surface plated onto nutrient agar with or without the supplements 0.5 M NaCl or sodium lauryl sulfate (SLS). The minimum

843

844

BENNETT, SEAVER, AND CALCOTT

TABLE 1. Genetic properties of a family of S. typhimurium mutants Strain SL3770 SL3749

DesigGenotype nation PC1 rfa+ PC2

rfaL446

SL3750

PC3

rfaJ417

SL3748

PC4

rfa-432

SL3769

PC5

rfaG471

SL3789

PC6

rfaF511

SL1102

PC7

LT2 metA22 trpB2 Hib H2e flaA66 strA120 metE551 rfaE543

Origin Roantree et al.

(46)

Derived from PCi by transduction Derived from PCi by transduction Derived from PCi by transduction Derived from PCi by transduction Derived from PC1 by transduction SL1027 rfaE543

APPL. ENVIRON. MICROBIOL. Tris-2 mM MgSO4 (pH 7.4) buffer and incubated at 370C for 1 h (48). The washed particles were recovered by ultracentrifugation, resuspended in the same buffer, and incubated at 370C for a further 1 h before recovery by ultracentrifugation and resuspension in the TrisMgSO4 buffer. These depleted particles (analogous to the crude outer membranes of Schnaitman [48] and Smit et al. [49]) were designated Triton-insoluble material (TIM). Biochemical analyses. Protein in the CFE and TIM was determined by the Lowry method with bovine serum albumin, fraction V, as the standard (35). LPS was assayed as keto-deoxyoctulosonic acid (KDO) by the thiobarbituric acid method (43). Chemicals. All chemicals were purchased from Sigma Chemical Co., St. Louis, Mo., and Fisher Scientific Co., Cincinnati, Ohio.

RESULTS A series of strains of S. typhimurium carrying defined mutations which result in alterations in the LPS core (46) were assessed for their MIC concentration of SLS which allowed >90% plating for five antibiotics. The antibiotics chosen inefficiency was determined for each strain. Plates were cluded three, bacitracin, novobiocin, and polyincubated at 37°C for 24 to 48 h before counting. myxin B, of which the relative MIC is dependent Viability was determined with reference to the colony on the LPS polysaccharide chain length, and counts obtained from untreated samples (17). two, neomycin and tetracycline, in which chain MICs of antibiotics. The minimum inhibitory con- length plays a minor role in determining the centration (MIC) of each antibiotic was determined by a standard plating procedure. Routinely, antibiotics MIC (46). As demonstrated by Roantree et al. at a variety of concentrations were incorporated into (46), the relative MICs for bacitracin, novobionutrient agar, and a diluted population of the orga- cin, and polymyxin B were greatly reduced, to nisms (containing 150 to 250 viable organisms per 0.1 less than 20% in the rough mutant PC5 and the ml) was surface plated onto the medium. After growth deeper rough mutants (PC6 and PC7), compared for 24 to 48 h at 37°C, colony counts were performed. with the wild type and the less deep rough The MIC was defined as that concentration of the strains (PC2 and PC3) (data not shown). Howantibiotic that reduced the colony count to 90%

plating efficiency

Strain

SLS (%) 0.5 ± 0.03 0.5 ± 0.05 0.35 ± 0.03 0.1 ± 0.02 0.05 ± 0.02 0.01 ± 0.005 0.2 ± 0.02

NaCl (M)

PC1

0.5 0.5 0.5 0.5 0.5 0.5 0.5

PC2 PC3 PC4 PC5 PC6 PC7 a Late-log-early-stationary-phase organisms were surface plated onto nutrient agar supplemented with NaCl or SLS to the indicated levels. The maximum concentration of agent that allowed >90% plating efficiency was determined.

100

-

10

-

A

--

0

B

M-cr

-

.

--

A

0.1 J 0.01 c

100

Uf) 10

-

I.

-

0.1

-

0.01

12 3 4 5 6 7

1 2 3 4 5 6 7

STRAIN PC FIG. 1. Effect offreezing and thawing on a family of S. typhimurium mutants. Washed late-log-earlystationary-phase organisms were subjected to slow (A, C) or rapid (B, D) freezing in water (A, B) or 0.85% saline (C, D) and slow thawing. The populations were suitably diluted and surface plated onto nutrient agar (U), supplemented with 0.5 M NaCI (A) or with SLS (0) to levels described in Table 2. The plates were incubated at 37°C to constant count. The percentage of survivors was calculated for each condition with reference to the count of unstressed orga-

nisms. This represents the average of three experiments performed. The variation noted was ±8% of the values given.

846

APPL. ENVIRON. MICROBIOL.


gent even in the absence of EDTA. This sensitivity was enhanced in the presence of EDTA. In general, the wild type and the less deep teria in the presence and absence of EDTA. The wild type and some of the rough mutant strains rough strains were less susceptible to antibiotics, to SLS both in growing and resting cells, to were relatively resistant to low concentrations of the detergent in the absence of EDTA (PC1, EDTA, and to both rapid and slow freezing and PC2, PC3, PC4, and PC5), whereas the deep thawing in water or saline, and also exhibited rough mutants (PC6 and PC7) were extremely more outer membrane damage after freeze-thaw susceptible (Fig. 2). EDTA was able to increase than the deepest rough strains. Thus these muthe rate of SLS-induced lysis of all rough strains, tations caused alterations in the stability and particularly the deepest (PC5, PC6, and PC7). sensitivity of the outer membrane to stress. When challenged with high concentrations of While it is certain that these mutations caused detergent, the deepest rough strains (PC5, PC6, a decrease in the chain length of the carbohyand PC7) showed extreme sensitivity to deter- drate residue of the LPS, it is possible that they might have interfered in the synthesis and incorporation of another outer membrane com100 ponent, protein. Ames et al. (1) have presented evidence that the concentrations of the outer membrane proteins are reduced in the deepest rough mutants of S. typhimurium. We have attempted to confirm these observations in this family of mutants by comparing the amounts of LPS (KDO-positive material) and protein that 50 were not solubilized from disrupted organisms by washing in 2% Triton-Tris-MgSO4 buffer. This procedure yields particles or TIM which are analogous to the crude outer membranes of Schnaitman (48). No attempt was made to purify the outer membrane by other procedures such as equilibrium sucrose density gradient ultracenoT trifugation (40). Table 3 illustrates one typical 2 3 4 6 7 experiment of three performed. As can be seen, STRAIN PC the wild-type PC1 and the rough mutants PC2 FIG. 2. Effects of SLS and EDTA on cellular in-. and PC3 retained the majority of their LPS tegrity of a family of S. typhimurium mutants. (KDOTIM/KDOCFE) on their TIM (65 to 87%), Washed late-log-early-stationary-phase organisms whereas the deeper rough mutant strains (PC4, were suspended in Tris-hydrochloride (50 mM; pH PC5, and PC6) retained very little (25 to 33%). 7.4) with (-, A) or without (0, A) 1 mM EDTA, and The PC7 strain, which had shown aberrant rethen SLS was added to a final concentration of 0.1% (0, 0) or 1% (A, A). Lysis of the organisms was sults in the other experiments described, refollowed on a recording spectrophotometer at 540 nm. tained intermediate levels of LPS on the TIM. The rate of lysis was determined for the linear portion In addition, whereas the wild type (PC1) and the rough mutants PC2 and PC3 contained simof the slope of the curve.

these mutations on the ability of SLS to permeate the outer membrane and lyse resting bac-

U)

5

TABLE 3. Preparation of crude outer membranes from a family of S. typhimurium mutantsa CFE

Stram,

Protein

LPS (ug of KDO)

TIM

Protein

LPS (,Lg of KDO)

LPSTIM/

LPSCFE

ProteinTIM/

proteinCFE

LPSCFE/

proteinTIM

(mg) (mg) PC1 90:5 320 7.4 277 0.87 0.082 43.2 63.1 345 7.8 234 0.68 PC2 0.124 44.0 PC3 76.2 266 6.0 172 0.65 0.078 44.4 PC4 119.0 630 5.2 204 0.33 0.044 120.0 111 61.2 365 PC5 3.5 0.30 0.057 103.0 PC6 94.8 315 3.8 80 0.25 0.040 83.6 PC7 53.6 276 4.7 135 0.49 0.088 59.0 aIn this experiment washed organisms were disrupted in a French pressure cell to produce CFE. Protein (Lowry method) and LPS (thiobarbituric acid-positive material, or KDO) were determined. The CFEs were separated into particulate and soluble material by ultracentrifugation. The particles were washed twice in 2% Triton-50 mM Tris-hydrochloride (pH 7.4)-2 mM MgCl2 buffer for 1 h at 370C to solubilize cell membrane and were recovered by ultracentrifugation. The crude outer membrane, or TIM, was assayed for protein and LPS.

VOL. 42, 1981

LPS CORE DEFECTS IN S. TYPHIMURIUM

ilar amounts of LPS per unit of cell protein (KDOcFE/proteincFE), the deeper rough mutants (PC4, PC5, and PC6) contained substantially more LPS. Again, strain PC7 was unusual and contained quantities of LPS similar to those in the wild type and distinct from that contained in the other deeper rough mutants. As reported by Ames et al. (1), the deepest rough mutants (in our system, PC4, PC5, and PC6) contained less protein in TIM per unit of cell protein (proteinTnm/proteinCFE) than the wild type or strains PC2 and PC3. Again the deepest rough mutant, PC7, was aberrant, containing more protein in TIM than the other deep rough mutants. Analysis of these crude outer membranes (TIM) by SLS-polyacrylamide gel electrophoresis indicated that the protein profiles were similar both qualitatively and quantatively in all seven strains as reported by Ames et al. (1) (P. H. Calcott, unpublished data).

DISCUSSION Defined mutations which result in alteration in the carbohydrate moiety of the LPS are pleiotropic inasmuch as they alter the quantity and relative proportions of two major components of the outer membrane, protein and LPS, in S. typhimurium. Decreasing the carbohydrate chain length by mutation resulted in a decrease in the incorporation of outer membrane proteins into the structure (1, 26; this paper) and an increase in the amount of LPS incorporated (1; this study). Also associated with this is a decrease in stability of the LPS in the membrane. This cannot be explained simply by an altered cell volume or surface area (34, 49). Consequently, these mutations alter the molecular arrangement or architecture or both in the outer membrane, perhaps by altering the spatial arrangement of the components (20, 21). Those mutations that result in a substantial decrease in the LPS chain length do not affect the susceptibility of the cytoplasmic membrane to NaCl or increase the incidence of membrane damage on freeze-thaw. However, they alter the susceptibility of the cells when the stress is directed, at least in part, to the outer membrane. Whether the sensitivities of the deepest rough strains with shorter LPS carbohydrate chains compared with the wild type are a direct result of the shorter LPS chains or an indirect effect via a decreased protein content in the outer membrane cannot be conclusively ascertained with the data presented. However, the carbohydrate chain length must play a role since strain PC7 resembles the wild type in protein content of the outer membrane, quantity of cellular LPS, and stability of this macromolecule in the outer membrane, yet it is not as resistant

847

to the stresses as the wild type and is almost as sensitive as the deep rough strain PC6. At least some of the sensitivity conferred by the mutations can be attributed to indirect effects, since PC7 was significantly more resistant to SLS during growth and exhibited less outer membrane damage on freeze-thaw than the other deeper rough strain (PC6). The involvement of components other than LPS, such as protein, in the outer membrane in determining sensitivity to stress is strengthened by two studies in our laboratory. Mutants of Escherichia coli lacking the porn outer membrane protein (ompB) are extremely resistant whereas mutants lacking heat-modifiable protein (ompA) or minor portein TSX (tsx) are sensitive to freeze-thaw stress (P. H. Calcott, R. S. Petty, and L. Manzo, manuscript in preparation). In another system, a plasmid-bearing strain of Pseudomonas aeruginosa was unusually sensitive to SLS after freezethaw and to EDTA (18), indicating a plasmidmediated alteration in the outer membrane structure. Although the plasmid did not alter the lipid composition or the LPS composition of the bacterium, it increased the quantity of a 55,000-molecular-weight protein in the outer membrane (P. H. Calcott, G. Campbell, N. H. Truong, C. Ellis, and M. Thomas, manuscript in preparation). Nikaido (41) has presented a model to explain the role of LPS and protein in conferring resistance to gram-negative bacteria to certain lowmolecular-weight hydrophobic antibiotics. He envisaged that while hydrophilic antibiotics enter the periplasmic space through protein pores (porins), hydrophobic antibiotics pass the barrier by dissolving in the phospholipids of the outer membrane. The long chain of the LPS acts to prevent entry of these latter antibiotics by preventing them from approaching and dissolving in the phospholipids. Since freeze-thaw sensitivity, particularly in the deeper rough mutants, is significantly increased compared with the wild type, we can conclude that if Nikaido's model is correct, the stress of freeze-thaw acts on the phospholipid moiety of the outer membrane. In addition, since deep rough strains become extremely susceptible to SLS after freezethaw, we can conclude that the stress allows more complete or rapid entry (or both) of SLS to the cytoplasmic membrane, most probably by exposing more phospholipid or breaking up phospholipid-protein-LPS interactions; LPS and protein are removed from the outer membrane on freeze-thaw (23, 45). This study indicates that LPS structure and most probably protein makeup and organization of outer membrane play important roles in determining sensitivity of S. typhimurium to stress.

848


Variations in membrane composition can be brought about by changes in the genetic makeup (1, 18, 21, 39, 46) or by nutritional factors (13, 16, 21, 24, 25, 47), which can result in dramatic differences in sensitivity of these organisms to a variety of stresses (13, 18, 24, 46). This might help to explain the variability in sensitivity and resistance of gram-negative bacteria in stressful environments such as water bodies, chlorinetreated water, frozen food, and other environmental samples where organisms with injured walls can be detected (7, 9, 19, 44, 50).

APPL. ENVIRON. MICROBIOL.

of Escherichia coli from freeze-thaw damage: the relative importance of wall and membrane damage. Can. J. Microbiol. 21:1960-1968. 16. Calcott, P. H., and R. S. Petty. 1980. Phenotypic viability of lipids of Escherichia coli grown in chemostat culture. FEMS Lett. 7:23-27. 17. Calcott, P. H., S. M. Steenbergen, and R. S. Petty. 1979. The effects of freezing and thawing on the uptake and retention of amino acids by Escherichia coli. FEMS Lett. 6:267-272. 18. Calcott, P. H., C. Zaborowski, W. L. Levine, and N. H. Truong. 1979. Drug resistance plasmid (pPL1) mediated changes in the susceptibility of Pseudomonas aeruginosa to stress. FEMS Lett. 7:75-80. 19. Camper, A. K., and G. A. McFeters. 1978. Chlorine injury and the enumeration of waterborne coliform ACKNOWLEDGMENTS bacteria. Appl. Microbiol. 37:633-641. We acknowledge the generous support of the U.S. Army 20. Costerton, J. W., J. M. Ingram, and K. J. Cheng. Department of Research (grant no. DRXRO-CB-15525-L). 1974. Structure and function of the cell membrane of We thank W. S. Brewer for critically reviewing this manugram-negative bacteria. Bacteriol. Rev. 38:87-110. script. 21. DiRienzo, J. M., K. Nakamura, and M. Inouye. 1978. The outer membrane proteins of Gram-negative bacLITERATURE CITED teria: biosynthesis, assembly and functions. Annu. Rev. 1. Ames, G. F.-L., E. N. Spudich, and H. Nikaido. 1974. Biochem. 47:481-532. Protein composition of the outer membrane of Salmo- 22. Dutka, B. J. 1973. Coliform are an inadequate index of nella typhimurium: effect of lipopolysaccharide mutawater quality. J. Environ. Health 36:39-46. tions. J. Bacteriol. 117:406-416. 23. Ghani, B. A., and P. H. Calcott. 1980. Protein synthesis 2. Asbell, M. A., and R. G. Eagan. 1966. Role of multivain Escherichia coli recovering from freezing and thawlent cations in the organization, structure, and assembly ing damage. Ohio J. Sci. 80:69. of the cell wall of Pseudomonas aeruginosa. J. Bacte- 24. Gilbert, P., and M. R. W. Brown. 1978. Influence of riol. 92:380-387. growth rate and nutrient limitation on the gross cellular 3. Bank, H. 1973. Visualization of freezing damage. II. Struccomposition of Pseudomonas aeruginosa and its resisttural alterations during warming. Cryobiology 10:157ance to 3- and 4-chlorophenol. J. Bacteriol. 133:1066170. 1072. 4. Bank, H., and P. Mazur. 1973. Visualization of freezing 25. Gill, C. O., and J. R. Suisted. 1978. The effects of and damage. J. Cell. Biol. 57:729-742. temperature and growth rate on the proportion of un5. Berg, G., D. R. Dahling, G. A. Brown, and D. Berman. saturated fatty acids in bacteriol lipids. J. Gen. Micro1978. Validity of fecal coliforms, total coliforms, and biol. 104:31-36. fecal streptococci as indicators of viruses in chlorinated 26. Gmeiner, J., and S. Schlecht. 1979. Molecular organiprimary sewage effluents. Appl. Environ. Microbiol. 36: sation of the outer membrane of Salmonella typhimu880-884. rium. Eur. J. Biochem. 93:609-620. 6. Beuchat, L. R. 1978. Injury and repair of Gram-negative 27. Grabow, W. 0. K., and M. du Preez. 1979. Comparison bacteria, with special consideration of the involvements of m-Endo LES, MacConkey, and Teepol media for of the cytoplasmic membrane. Adv. Appl. Microbiol. membrane filtration counting of total coliform bacteria 23:219-243. in water. Appl Environ. Microbiol. 38:351-358. 7. Bissonnette, G. K., J. J. Jezeski, G. A. McFeters, and 28. Ingram, M., and B. M. Mackay. 1976. Inactivation by D. G. Stuart. 1975. Influence of envirommental stress cold, p. 111-151. In C. Board and B. M. Mackay (ed.), on enumeration of indicator bacteria from natural waInhibition and inactivation of vegetative microbes. Fifth ters. Appl. Microbiol. 29:186-194. Symposium of the Society of Applied Bacteriology. 8. Brown, A. D. 1976. Microbial water stress. Bacteriol. Oxford University Press, Oxford. Rev. 40:803-846. 29. Kamio, Y., and H. Nikaido. 1977. Outer membrane of 9. Cabelli, V. P. 1979. What do water quality indicators Salmonella typhimurium: identification of proteins exindicate?, p. 305-336. In R. R. Colwell and J. Foster posed on cell surface. Biochim. Biophys. Acta 464:589(ed.), Aquatic microbial ecology: proceedings of the 601. ASM conference. University of Maryland, College Park. 30. Knowles, C. J., P. H. Calcott, and R. A. MacLeod. 10. Calcott, P. H. 1978. Freezing and thawing microbes. 1974. Periplasmic CO-binding c-type cytochrome in a Patterns of Progress Microbiology Series PPM 14. Mermarine bacterium. FEBS Lett. 49:78-83. row Publishing, Shilden, England. 31. Lee, S. K., P. H. Calcott, and R. A. MacLeod. 1977. 11. Calcott, P. H., and A. M. Gargett. 1981. Mutagenicity Relationship of cytochrome content to the sensitivity of freezing and thawing. FEMS Lett. 10:151-155. of bacteria to NaCl on freezing and thawing. Can. J. 12. Calcott, P. H., and R. A. MacLeod. 1974. Survival of Microbiol. 23:413-419. Escherichia coli from freeze-thaw damage: a theoretical 32. Leive, L. 1965. Release of lipopolysaccharide by EDTA and practical study. Can. J. Microbiol. 20:671-681. treatment of E. coli. Biochem. Biophys. Res. Commun. 13. Calcott, P. H., and R. A. MacLeod. 1974. Survival of 21:290-296. Escherichia coli from freeze-thaw damage: influence of 33. Litsky, W. 1979. Gut critters are stressed in the environnutritional status and growth rate. Can. J. Microbiol. ment, more stressed by isolation procedures, p. 345-347. 20:683-689. In R. R. Colwell and J. Foster (ed.), Aquatic microbiol 14. Calcott, P. H., and R. A. MacLeod. 1975. The survival ecology: proceedings of the ASM conference. University of Escherichia coli from freeze-thaw damage: permeaof Maryland, College Park. bility barrier damage and viability. Can. J. Microbiol. 34. Lounatmaa, K. 1977. Effect of acrolein-glutaraldehyde 21:1724-1732. on the outer membrane of Salmonella typhimurium 15. Calcott, P. H., and R. A. MacLeod. 1975. The survival smooth and rough strains. FEMS Lett. 2:167-171.

V 1PS CORE DEFECTS IN S. TYPHIMURIUM VOL. 42, 1981 35. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 36. MacLeod, R. A., and P. H. Calcott. 1976. Cold shock and freezing damage to microbes, p. 81-109. In J. R. Postgate and T. G. R. Gray (ed.), Survival of vegetative bacteria. 26th Symposium of the Society for General Microbiology. Cambridge University Press, Cambridge, England. 37. Mazur, P. 1970. Cryobiology: the freezing of biological systems. Science 168:939-949. 38. McFeters, G. A., G. K. Bissonette, J. J. Jezeski, C. A. Thompson, and D. G. Stuart. 1974. Comparative survival of indicator bacteria and enteric pathogens in well water. Appl. Microbiol. 27:823-829. 39. Meadow, P. M., P. L. Wells, M. Salkinoja-Salonen, and E. L. Nurmiaho. 1978. The effect of lipopolysaccharide composition on the ultrastructure of Pseudomonas aeruginosa. J. Gen. Microbiol. 105:23-28. 40. Mizuno, T., and M. Kageyama. 1978. Separation and characterization of the inner and outer membrane of Pseudomonas aeruginosa. J. Biochem. 84:179-191. 41. Nikhido, H. 1976. Outer membrane of Salmonella typhimurium: transmembrane diffusion of some hydrophobic substances. Biochim. Biophys. Acta 433:118-132. 42. Ohkawa, I., S. Shiga, and M. Kageyama. 1979. An esterase on the outer membrane of Pseudomonas aeruginosa for the hydrolysis of long chain acyl esters. J. Biochem. 86:643-656. 43. Osborn, M. J. 1963. Studies on the Gram-negative cell wall. I. Evidence for the role of 2 keto-3 deoxyoctonate in the lipopolysaccharide of Salmonella typhimurium.

849

Proc. Natl. Acad. Sci. U.S.A. 50:499-506. 44. Ray, B., and M. L. Speck. 1973. Freezing-injury in bacteria. Crit. Rev. Clin. Lab. Sci. 4:161-213. 45. Ray, B., M. L. Speck, and W. J. Dobrogosz. 1976. Cell wall lipopolysaccharide damage in Escherichia coli due to freezing. Cryobiology 13:153-160. 46. Roantree, R. J., T. T. Kuo, and D. G. MacPhee. 1977. The effect of defined lipopolysaccharide core defects on the antibiotic resistance of Salmonella typhimurium. J. Gen. Microbiol. 103:223-234. 47. Robinson, A., and D. W. Tempest. 1973. Phenotypic variability of the envelope proteins of Klebsiella aerogenes. J. Gen. Microbiol. 78:361-370. 48. Schnaitman, C. A. 1972. Effect of ethylenediaminetetraacetic acid, Triton X-100, and lysozyme on the morphology and chemical composition of isolated cell walls of Escherichia coli. J. Bacteriol. 108:553-563. 49. Smit, J., Y. Kamino, and H. Nikaido. 1975. Outer membrane of Salmonella typhimurium: chemical analysis and freeze-fracture studies with lipopolysaccharide mutants. J. Bacteriol. 124:942-958. 50. Stuart, D. G., G. A. McFeters, and J. E. Schillinger. 1977. Membrane filter technique for the quantification of stressed fecal coliforms in the aquatic environment. Appl. Environ. Microbiol. 34:42-46. 51. Unemoto, T., and R. A. MacLeod. 1975. Capacity of the outer membrane of a gram-negative marine bacterium in the presence of cations to prevent lysis by Triton X100. J. Bacteriol. 121:800-806. 52. Verstraete, W., and J. P. Voets. 1976. Comparative study of E. coli survival in two aquatic ecosystems. Water Res. 10:129-136.