Stability of Lactic Acid Bacteria to Freezing as Related to Their Fatty Acid Composition

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLoGY, Mar. 1977, p Copyright 1977 American Society for Microbiology Vol. 33, No. 3 Printed in U.S.A. Stability of Lactic Acid Bacteria to Freezing as Related to Their Fatty Acid Composition I. GOLDBERG* AND L. ESCHAR Laboratory ofapplied Microbiology, Institute ofmicrobiology, Hebrew University-Hadassah Medical School, Jerusalem, Israel Received for publication 9 September 1976 The viability of Streptococcus lactis and Lactobacillus sp. A-12 after freezing at -17 C for 48 h was better preserved when the cells were grown in medium supplemented with oleic acid or Tween 80 (polyoxyethylene sorbitan monooleate). A pronounced change in the cellular fatty acid composition was noted when the bacteria were grown in the presence of Tween 80. In S. lactis the ratio of unsaturated to saturated fatty acids increased from 1.18 to 2.55 and in Lactobacillus sp. A-12 it increased from 0.85 to 1.67 when Tween 80 was added to the growth medium. The antibiotic cerulenin markedly inhibited the growth of lactic acid bacteria in tomato juice (TJ) medium but had almost no effect on the growth of the bacteria in TJ medium containing Tween 80 (or oleic acid). The antibiotic inhibited markedly the incorporation of [1-'4C]acetate but had no inhibitory effect on the incorporation of exogenous [1-'4C]oleate (or [1- '4C]palmitate) into the lipid fractions of lactic acid bacteria. Thus, the fatty acid composition of lactic acid bacteria, inhibited by the antibiotic cerulenin, can be modulated by exogenously added oleic acid (or Tween 80) without the concurrent endogenous fatty acid synthesis from acetate. The data obtained suggest that cerulenin inhibits neither cyclopropane fatty acid synthesis nor elongation of fatty acid acyl intermediates. The radioactivity of cells grown in the presence of [1-'4C]oleate and cerulenin was associated mainly with cyclopropane A19:0, 20:0 + 20:1, and 21:0 acids. As a consequence, cerulenin caused a decrease in the ratio of unsaturated to saturated fatty acids in lactic acid bacteria as compared with cells grown in TJ medium plus Tween 80 but without cerulenin. Cerulenin caused a decrease in the viability of S. lactis and Lactobacillus sp. A-12 after freezing at 17 C for 48 h only when Tween 80 - was present in the growth medium. We conclude that the sensitivity of lactic acid bacteria to damage from freezing can be correlated with specific alterations in the cellular fatty acids. Concentrated frozen cultures of lactobacilli acid contained higher levels of octadecanoic and lactic streptococci are used routinely in the (18:1) and cyclopropane nonadecanoic (19:0) manufacture of various fermented products. fatty acids and lower levels of saturated fatty However, certain lactobacilli (22, 23) and individual strains of lactic streptococci (1, 5, 7, 13, acid. Statistical analysis of different strains in- acids than cells grown in the absence of oleic 16, 24) are not sufficiently stable to freezing. dicated that the amount of cyclopropane A19:0 The instability of frozen cultures of these bacteria presents a difficulty with regard to their use of lactobacilli after the freezing process. fatty acid was most closely related to viability for fermentation processes, especially for the Gilliland and Speck (7) reported that concentrated cultures of different strains of lactic direct inoculation of the food with these cultures. streptococci varied with respect to survival at It was found (22, 23) that concentrated cultures of different strains of Lactobacillus bul- (18:1) acid in the cellular lipids of different cul C for 7 days. As the percentage of oleic garicus were affected by freezing in liquid nitrogen for 24 h. The degree of viability after the after freezing decreased (7). tures increased, the percentage of survivors freezing process varied among the different In this study we investigated what kind of strains studied. Viability of all strains examined was improved by supplementing the maintain high viability of Streptococcus lactis cellular fatty acid composition was required to growth medium with oleic acid or with Tween and Lactobacillus sp. A-12 after freezing at 80. All strains grown in the presence of oleic -17 C. Also, we studied whether or not specific 489

2 490 GOLDBERG AND ESCHAR alterations in the endogenous fatty acid composition, caused by the antibiotic cerulenin (2, 19, 20) in the presence of exogenously added oleic acid (or Tween 80), would have an effect on the viability of these bacteria after the freezing process. MATERIALS AND METHODS Bacteria, media, and growth conditions. S. lactis is used commercially in the manufacture of dairy products in Israel. Lactobacillus sp. A-12 was isolated from Israeli red wines and is used for inducing malo-lactic fermentation in these wines. Lactic acid bacteria were grown in tomato juice (TJ) medium containing, per liter: tryptone (Difco), 20 g; peptone (Difco), 5 g; yeast extract (Difco), 5 g; glucose, 3 g; lactose, 2 g; and tomato juice, 200 ml. The ph of the medium was adjusted to 5.5 with 6 N HCl. Tomato juice was prepared by adding 560 ml of canned juice (Tal, Israel) to 1,800 ml of water. The solution was kept at 4 C for 20 h and centrifuged at 120 x g for 10 min, and the supernatant solution was filtered through a Whatman no. 541 filter paper. The clear solution obtained was kept at -20 C until use. In some experiments TJ medium was supplemented with 0.1% Tween 80 (Sigma Chemical Co., St. Louis, Mo.). Cultures were grown in test tubes containing 2 to 4 ml of the appropriate medium. Cerulenin (Makor Chemicals Ltd., Jerusalem, Israel) was added to the cell suspensions to a final concentration of 200,ug/ ml. Fatty acid supplements were in the form of K+ salt solution of oleate and palmitate (Sigma, 99% pure) in slightly alkaline aqueous ethanol (1:3, vol/ vol). Growth of cells was followed by measuring the culture density at 650 nm. Preparation of concentrated cell suspensions. S. lactis and Lactobacillus sp. A-12 were grown statically in tubes containing 5 ml of TJ medium with or without Tween 80. Bacteria were transferred three times in the same medium and then inoculated into 250-ml Erlenmeyer flasks containing 100 ml of the appropriate medium. After the cultures reached the stationary phase of growth (18 h at 30 C), the cells were harvested by centrifugation and suspended in 2 ml of TJ medium. Cell suspensions were stored at -17 C for 48 h. CFU counts. Colony-forming units (CFU) were determined before and after freezing by plating samples of cell suspensions on agar plates prepared with TJ medium with or without Tween 80. The frozen cell suspensions were thawed by immersion in water at 17 C for 5 min. Duplicate plates were used for each dilution. Incubations were carried out under anaerobic conditions (GasPak anaerobic jars; BBL, Cockeysville, Md.) at 30 C for 48 h. Mean values for CFU were calculated by using the equation m = dl(y(cl + W2) n + n2d2 where mh is the mean colony count, d, and d2 are two consecutive dilutions, C, and C2 are the total colony APPL. ENVIRON. MICROBIOL. counts on all the plates from dilutions d, and d2, and ni and n2 are the number of plates used for each dilution. Isotope experiments. [1-'4C]acetate (62 Ci/mol), [1-'4C]palmitate (50 Ci/mol), and [1-'4C]oleate (57 Ci/mol) were obtained from the Radiochemical Centre, Amersham, England. The labeled fatty acids were added undiluted to the growing cultures. Samples were counted in vials containing 10 ml of 2,5-diphenyloxazole-toluene (4 g/liter) in a Packard Tri-Carb liquid scintillation spectrometer model [1-'4C]acetate (2.5 uci), [1-_4C]palmitate (2 uci), or [1-'4C]oleate (0.5,uCi) was added to 2-ml portions of cell suspensions. After 3 h of incubation, the cell suspensions were cooled to 4 C and centrifuged at 12,000 x g for 2 min. Cell pellets were washed twice with 3 ml of cold water. Five milliliters of carrier cells (absorbancy at 650 nm = 1.8) of S. lactis or Lactobacillus sp. A-12 grown in TJ medium or TJ medium plus Tween 80 was added in order to minimize loss of labeled lipids, and the cells were washed twice with 3 ml of cold water. Lipids were extracted from the cells with chloroform-methanol (2:1, vol/ vol). Samples (1 ml) of the chloroform-methanol extracts were evaporated to dryness at 60 C, dissolved in 10 ml of 2,5-diphenyloxazole-toluene, and measured for radioactivity. Fatty acid analysis. The fatty acid methyl esters were prepared from the chloroform-methanol extracts of the cells according to method B described by Moss et al. (17). A Varian model 1200 or an F & M Scientific model 5750 (Packard) gas chromatograph equipped with flame ionization detectors was used for analysis of samples of methyl esters of fatty acids. Samples were analyzed on stainless-steel columns (5 feet by 1/8 inch [about 1.5 m by 0.3 cm]) packed either with 20% diethyleneglycol succinate on Chromosorb W (60/80 mesh) or with 10% Apiezon L on Chromosorb W/AW/DMCS (70/80 mesh) (Applied Science Co., State College, Pa.) The column temperature was 190 C. The injector temperature was 240 C, and the detector temperature was 190 C. The carrier gas was nitrogen at a flow rate of 9 to 20 ml/min. The methyl ester peaks were tentatively identified by comparison of their retention times with highly purified methyl ester standards (ClO to C20) (Applied Science Co.). Methyl esters were collected from the column by passing through capillary tubes that were cooled with liquid nitrogen. Peak areas were determined by weighing, and the percentage of each fatty acid was calculated from the ratio of the area of its peak to the total area of all peaks. RESULTS Effect of growth medium composition on stability during freezing. Cultures of S. lactis grown in TJ medium exhibited decreased viability after freezing at -17 C for 48 h (Table 1). When Tween 80 was added to the growth medium, a profound increase in the stability of these bacteria was noted. The addition of Tween 80 to the suspending medium had no protective effect for S. lactis grown in TJ me-

3 VOL. 33, 1977 TABLE 1. Effect ofgrowth medium composition on the stability of lactic acid bacteria after freezing at -1 7 C and storage for 48 h Bacterium Growth medium Death (%)a S. lactis TJ S. lactis TJ + Tween Lactobacillus sp. A-12 TJ Lactobacillus sp. A-12 TJ + Tween ± 7.3 a The numbers given represent average values ± interval estimated (with confidence interval of 95%). dium without added Tween 80 (results not shown). In the absence of Tween 80, cultures of Lactobacillus sp. A-12 were less affected by the freezing process than were cultures ofs. lactis (Table 1). With Lactobacillus sp. A-12, an improvement in the stability was noted when Tween 80 was added to the growth medium. However, this improvement was smaller than that observed with S. lactis. The addition of Tween 80 to the freezing menstruum had almost no effect on the viability oflactobacillus sp. A-12 cells frozen in TJ medium (results not shown). Similar results were obtained when the bacteria were frozen at -17 or - 70 C for 7 days (results not shown). Effect of Tween 80 on lipid fatty acids. Analyses of methyl esters of fatty acids prepared from the lipid fractions of S. lactis and Lactobacillus sp. A-12 grown with or without Tween 80 are given in Table 2. The main fatty acids found in both bacteria grown in TJ medium were 16:0, 18:1, and cyclopropane A19:0. In Lactobacillus sp. A-12, 16:1 fatty acid was present in relatively high quantity. The ratio of unsaturated to saturated fatty acids in S. lactis was 1.18, whereas this ratio in Lactobacillus sp. A-12 was 0.85 (Table 2). The improvement in the viability of both bacteria after the freezing process caused by the addition of Tween 80 to the growth medium (Table 1) was accompanied by profound changes in the composition of the fatty acids in the lipid fractions of the bacteria (Table 2). In S. lactis the proportions of 16:0, 16:1, and 18:0 fatty acids decreased, whereas the content of 18:1 acid increased from 10.1 to 28.8%. Almost no change in the proportion of cyclopropane A19:0 acid was noted in S. lactis when Tween 80 was added to the growth medium. In Lactobacilus sp. A-12, the addition of Tween 80 to the growth medium caused a marked decrease in the proportions of 12:0, 14:0, 16:0, 16:1, and 18:1 acids and an increase in the proportions of 18:1 and cyclopropane A19:0 acids (Table 2). In both bacteria the addition of Tween 80 to the growth medium caused an increase in the ra- FREEZING OF LACTIC ACID BACTERIA tio of unsaturated to saturated fatty acids (Table 2). Effect of cerulenin on the growth of lactic acid bacteria. When cerulenin was added at zero time to a suspension of S. lactis cells grown in TJ medium, growth was inhibited to a degree that depended on the concentration of the antibiotic. At a final cerulenin concentration of 200,ug/ml, the rate of growth was initially decreased, and after about 3 h growth stopped completely (Fig. 1). In a cerulenincontaining culture supplemented with oleate (or Tween 80), the growth rate was almost the same as in the control to which cerulenin was not added. The addition of cerulenin to S. lactis cells grown in TJ medium plus Tween 80 caused an initial decrease in the rate of growth, but after about 3 h growth resumed at its original rate (Fig. 1). When cerulenin was added at zero time to a suspension of Lactobacillus sp. A-12 cells grown in TJ medium, the rate of growth was initially decreased and after 2 to 3 h had TABLE 2. Fatty acid composition of the lipid fractions from bacteria grown with or without Tween 80a Percentage of fatty acid in: S. lactis Lactobacillus sp. Fatty acidb A-12 TJ + Ti TJ TJ + TJ Tween 80 Tween 80 10:0 0.2 < : : : : Xc :0 2.6 < : A19: : :Od Unsatu ratede/saturated a Samples of methyl ester fatty acids were separated on diethylene-glycol succinate column as described in Materials and Methods. b Number preceding the colon indicates number of carbons, and number after the colon designates degree of unsaturation. A19:0, Cyclopropane nonadecanoic acid. " Unidentified fatty acid; retention time between 16:1 and 18:0. d Tentative identification. e Cyclopropane A19:0 acid was considered an unsaturated acid.

4 492 GOLDBERG AND ESCHAR E c= AJ TIME (hours) FIG. 1. Effect of cerulenin on the growth of Streptococcus lactis. Cells grown at 30 C for 16 h in TJ medium with or without Tween 80 were inoculated into fresh medium. Cerulenin (200 Mglml) was added (1). Cells grown in TJ medium plus Tween 80 with (U) or without (J) cerulenin. Cells grown in TJmedium with (@) or without (0) cerulenin. Cells were grown in TJ medium with cerulenin, and palmitic acid (50.gIml) (x) or oleic acid (50 pg/ml) (A) was added (2). APPL. ENVIRON. MICROBIOL. stopped almost completely (Fig. 2). However, in Lactobacillus sp. A-12 cells grown in TJ medium plus Tween 80, the addition of ceru-. lenin had almost no effect on the rate of growth for several generations (Fig. 2). Effect of cerulenin on synthesis of cellular 0.1 constituents. The effect of cerulenin on lipid synthesis was studied in order to ascertain whether the antibiotic interfered with the de TIME (hours) novo synthesis of fatty acids in gram-positive FIG. 2. Effect of cerulenin on the growth of Lactobacillus sp. A-12. Cells grown at 30 C for 16 h in TJ- bacteria. Exponentially growing cultures ofs. lactis and Lactobacillus sp. A-12 were transferred to fresh TJ medium with or without medium with or without Tween 80 were inoculated into fresh medium. Cerulenin (200 M.g/ml) was added ( a). Culture Tween 80. grown in TJ medium with (O) or Cerulenin (final concentration, 200 without (O) cerulenin. Culture grown in TJ medium,ug/ml,) was added, and 1 h later labeled precursors were added for an additional 2 h. Ceru- At this time the cultures were diluted in the appropri- plus Tween 80 with (O) or without (0) cerulenin. ( t ) lenin inhibited the incorporation of [1- ate medium in the presence or absence ofcerulenin. "4Clacetate into the total lipid fractions of both bacteria by 83 to 94% as compared with the control (Table 3). Cerulenin inhibited the incorporation of [1-14C]palmitate and [1- "4C]oleate into the total lipid fractions of both bacteria grown in TJ medium by 21 to 50% as compared with the control. However, no inhibition by cerulenin of the incorporation of either [1-14C]palmitate or [1-14C]oleate was observed when these bacteria were grown on TJ medium plus Tween 80. During a 3-h incubation with cerulenin, the incorporation of [1-14C]acetate into the total protein fractions (hot trichloroacetic acid precipitate of the cells) of both bacteria was not affected by the antibiotic (results not shown). Effect of cerulenin on fatty acid composition in the lipid fractions of lactic acid bactedomae ct = Ln LU I- -LJ

5 VOL. 33, 1977 TABLE 3. FREEZING OF LACTIC ACID BACTERIA 493 Effect ofcerulenin on the incorporation of[i4c]acetate, ['4C]palmitate, and [14C]oleate into the lipid fractions of lactic acid bacteriaa Bacterium Growth medium Supplement Radioactivity in lipids (cpm/fraction) ['4C]acetate ['4C]palmi- ['4C]oleate tate S. lactis TJ None 6, ,400 96,700 TJ Cerulenin ,900 57,900 TJ + Tween 80 None 4, ,900 60,000 TJ + Tween 80 Cerulenin ,000 76,700 Lactobacillus sp. A-12 TJ None 12, ,500 78,400 TJ Cerulenin 2, ,000 63,200 TJ + Tween 80 None 25, , ,800 TJ + Tween 80 Cerulenin 1, , ,500 a Cells were grown at 30 C for 16 h in TJ medium with or without Tween 80. The 16-h cultures were diluted with the same medium in which they had been grown, in a total volume of 2 ml, and incubated at 30 C until exponential growth resumed. Cerulenin was added (final concentration of 200,ug/ml), and after 1 h [14C]acetate (2.5 MCi), [14C]palmitate (2 0Ci), or [14C]oleate (0.5 MCi) was added and the incubation was continued for an additional 2 h. Radioactivity in the lipid fraction of the cells was measured as described in Materials and Methods. ria. In lactic acid bacteria treated with cerulenin, the exogenously added oleic acid (or Tween 80) is incorporated into cellular lipids (Table 3), thus altering the fatty acid composition (Table 4). The proportions of 12:0, 14:0, 16:0, 16:1, and 18:1 acids decreased and the proportions of cyclopropane A19:0, 20:0 + 20:1, and 21:0 acids increased as compared with the control (Table 2). These results suggest that both the exogenously added acid and the cyclopropane A19:0 acid derived from it (9, 10, 14, 26) were elongated in the presence of cerulenin. Direct confirmation of this contention was obtained when Lactobacillus sp. A-12 cells were grown in the presence of [1-14C]oleic acid and cerulenin and the radioactive fatty acids in the lipid fraction of the cells were analyzed (Table 4). Most of the radioactivity was found in cyclopropane A19:0, 20:0 + 20:1, and 21:0 acids. Cerulenin caused an increase in the ratio of unsaturated to saturated fatty acids in both bacteria (Table 4) as compared with cells grown in TJ medium with Tween 80 but without cerulenin (Table 2). Effect of cerulenin on the viability of lactic acid bacteria after freezing at -17 C. The viability of S. lactis and Lactobacillus sp. A-12 after freezing at -17 C for 48 h changed very little when the bacteria were incubated in TJ medium with cerulenin (for 3.5 h) before the freezing process (Table 5). Cells grown in TJ medium plus Tween 80 were less affected by the freezing process than cells grown in TJ medium (Tables 1 and 5). The addition of cerulenin to cells incubated in TJ medium plus Tween 80 (before freezing at -17 C) caused a decrease in the stability of these bacteria after freezing, and this affect was more pronounced with S. lactis than with Lactobacillus sp. A-12 (Table 5). When cerulenin was removed from the cell suspensions, after 3.5 h of incubation in TJ medium plus Tween 80, viability was improved after freezing at -17 C (Table 5). DISCUSSION In this work we have shown that the viability oflactobacillus sp. A-12 and S. lactis after freezing at - 17 C for 48 h was better preserved when the cells were grown in medium supplemented with oleic acid or Tween 80. The increase in the viability of these bacteria was concurrent with changes in the fatty acid composition. In Lactobacillus sp. A-12, the major changes were the increase in the proportions of 18:1 and cyclopropane A19:0 acids and the increase in the ratio of unsaturated to saturated fatty acids (see Table 2), similar to the results obtained with other species of bacteria belonging to the genus Lactobacillus (14, 18, 23). S. lactis, like other lactic streptococci species (7), contained much higher levels of cyclopropane A19:0 acid than did lactobacilli (see Table 2). The addition of Tween 80 to cell suspensions of S. lactis caused changes in fatty acid composition similar to Lactobacillus sp. A-12 except for the content of cyclopropane A19:0 acid, which remained almost unchanged. Our results are in contrast with those of Gilliland and Speck (7), who have shown that the higher the relative amount of 18:1 acid in different strains of lactic streptococci, the lower the viability after freezing at -17 C. In addition, several reports have sug-

6 494 GOLDBERG AND ESCHAR TABLE 4. Effect of cerulenin on fatty acid composition in the lipid fractions of lactic acid bacteria grown in TJ medium supplemented with oleic acida Lactobacillus sp. A-12 S. lactis Radioac- Fatty acid (% of fatty % of tivity in acid) fatty acid fatty acid (cpm) 10:0 < NDb 12: ND 14: : : X( : : : ,200 20:0 + 20: ,300 21:0d ,600 >21:0 ND Unsaturated'/ saturated a Cells were grown statically with TJ medium plus Tween 80 at 30 C for 16 h. They were transferred three times in the same medium and then inoculated to flasks containing 10 to 20 ml of TJ medium supplemented with oleic acid (300 Ag/ml). After the cultures had reached the exponential phase of growth (about 1.5 h), cerulenin (200,ug/ml, final concentration) was added, and the cultures were then divided into two portions, one of which was incubated for 4 h. Methyl ester fatty acids were prepared from the lipid fraction of the cells. The other portion was incubated for 1 h, and then ['4C]oleic acid (5,uCi) was added and incubation was continued for an additional 2 h. Lipids were extracted, and the methyl ester fatty acids were separated on a Apiezon column. Radioactivity in the different methyl ester fatty acids was determined as described in Materials and Methods. ND, Radioactivity could not be detected. Unidentified fatty acid with retention time between 16:0 and 18:1. d Tentative identification. e Cyclopropane A19:0 acid was considered an unsaturated acid. gested that cyclopropane A19:0 fatty acid is important in maintaining the proper fluidity of the cell membrane and thus preventing death of certain cells, including lactobacilli, during freezing. However, we have shown that although the content of cyclopropane A19:0 fatty acid in S. lactis (TJ medium) was about 3.7 times higher than inlactobacillus sp. A-12 (TJ medium), S. lactis was more affected by the freezing process than was Lactobacillus sp. A-12 (see Tables 1 and 2 and references 7 and 3). APPL. ENVIRON. MICROBIOL. Both in our experiments (Table 2) and in the work of Speck and his co-workers (7, 22, 23), the exogenously added Tween 80 (or oleic acid) did not stop de novo synthesis of endogenous fatty acid from acetate. As a consequence, the changes observed in the proportions of either 18:1 or cyclopropane A19:0 fatty acid (in TJ medium plus Tween 80) were accompanied by changes in the proportions of other fatty acids as well (see Table 2), so that the overall composition of fatty acids in these bacteria was altered. Therefore, by using these experimental conditions, it is not possible to draw meaningful conclusions regarding the role of a single fatty acid in maintaining viability of cells after the freezing process. Two ways are known to modify fatty acid composition of microorganisms by exogenous fatty acids without the concurrent de novo synthesis of endogenous fatty acids from acetate: (i) by using lipid-deficient mutants that are either temperature sensitive (20) or auxotrophic for fatty acid(s) (21), and (ii) by using the antibiotic cerulenin [(2S)(3R)-2,3-epoxy-4- oxo-7,10-dodecadienoylamide] (2, 17, 19, 20) in the presence of exogenous fatty acids (3, 8). It was pointed out previously (8) that although the same information can, in principle, be obtained by the two methods (compare reference 3 with 21 and reference 8 with 11), the cerulenin technique has the advantage as compared to the use of mutants in that it is not subject to temperature limitation and is applicable to diverse microorganisms or cells in which fatty acid synthesis occurs. Therefore we used cerulenin in this study. Previous reports (8, 25) have shown that this antibiotic blocks an essential step in the synthesis of both saturated and unsaturated fatty acids. The antibiotic specifically inhibits the,f-ketoacyl thioester synthetase of Mycobacterium phlei (25) and f3-ketoacyl acyl carrier protein synthetase ofescherichia coli (6). Cerulenin inhibited almost completely the growth of gram-positive bacteria (lactic acid bacteria) grown in TJ medium (Fig. 1 and 2), similar to data obtained (3, 8, 15) on the inhibitory effects of the antibiotic on the growth of a variety of yeasts, fungi, and gram-negative bacteria. The inhibitory effects of cerulenin on the growth of lactic acid bacteria were prevented by supplementing the cultures with oleic acid (or Tween 80), but not with palmitic acid (Fig. 1 and 2). These findings are similar to the results reported for the yeast Saccharomyces cerevisiae (3), where the addition of certain single fatty acids (oleic acid gave the best result) restored the growth of cerulenin-inhibited cells. In contrast to these results, the

7 VOL. 33, 1977 FREEZING OF LACTIC ACID BACTERIA 495 TABLE 5. Effect of cerulenin on the stability of lactic acid bacteria during freezing at -17C for 48 ha Bacterium Growth medium before freezing Death (%) 0 hw 1.5 h 2.5 h 3.5 h (1.5)c S. lactis TJ S. lactis TJ + Tween Lactobacillus sp. A-12 TJ Lactobacillus sp. A-12 TJ + Tween a Cells were grown as described in Table 4. Immediately after the addition of cerulenin (200 gg/ml, final concentration) and 1.5, 2.5, and 3.5 h after the addition of the antibiotic, samples of cells were taken. Samples were then divided into two portions; one was diluted and CFU were determined, and the other was frozen at - 17 C for 48 h and CFU were determined after thawing. Percentage of death was calculated in each sample from the ratio of the colony counts made before and after freezing. b Incubation time in the presence of cerulenin. c Cerulenin was removed after 3.5 h from the cell suspensions by centrifuging the cells at 10,000 x g for 3 min. The cells were incubated in fresh medium for an additional 1.5 h. inhibitory effect of cerulenin on the growth of E. coli was reversed only by supplementing the culture medium with a mixture of oleic acid (unsaturated acid) and palmitic acid (saturated acid) (8). Cerulenin inhibited the incorporation of [1- "4C]acetate into the total lipid fractions of lactic acid bacteria grown in the presence or absence of Tween 80 by 83 to 94% (Table 3). However, under these experimental conditions, cerulenin did not inhibit the incorporation of [1-14C]oleic acid (or [1-_4C]palmitic acid) into the lipid fractions of these bacteria (Table 3). These results indicate that when de novo synthesis of fatty acids is prevented, grampositive bacteria are able to utilize exogenous oleic acid, thus achieving normal growth. Cerulenin, when added to lactic acid bacteria incubated in TJ medium, did not significantly influence the viability of these cells after the freezing process (Table 5). During this incubation period in the presence of cerulenin (3.5 h), the growth of lactic acid bacteria in TJ medium was inhibited but continued to some extent (Fig. 1 and 2) without the concurrent de novo synthesis of fatty acids (Table 3). Thus incubation with cerulenin in TJ medium resulted in a reduction in the total content of fatty acids per cell without affecting the proportions of the different fatty acids in the lipid fractions ofthe cells. These effects appear to be more critical for the growth of lactic acid bacteria than for maintaining viability after freezing at -17 C. Different results were obtained when lactic acid bacteria were grown in TJ medium plus Tween 80 in the presence of cerulenin. In this experimental system, the growth of the bacteria was affected very little (Fig. 1 and 2), but their viability after freezing at - 17 C for 48 h was reduced profoundly (Table 5). As was mentioned earlier in the discussion, under these experimental conditions the de novo synthesis of fatty acids from acetate was inhibited almost completely (see Table 3), but the incorporation of exogenous Tween 80 (or oleic acid) into the lipid fraction of the cells was not affected (Tables 3 and 4). The exogenous oleic acid in the presence of cerulenin was in part elongated or was converted to cyclopropane A19:0 fatty acid and then elongated (Table 4; see also references 9, 10, 14, 26). Thus the only change that occurred in the fatty acid composition of these cells, as compared with control cells grown without cerulenin, was the increase in the relative amounts of cyclopropane A19:0 and 20:0 + 20:1 acids and mainly the increase in the content of 21:0 fatty acid, resulting in a decrease in the ratio of unsaturated to saturated fatty acids (Table 4). In this experimental system, the increase in the content of long-chain fatty acids (>18 carbon atoms) altered the integrity of the membrane and caused greater damage to lactic acid bacteria during freezing. In contrast to the conclusions of Speck and his co-workers (7, 23), the increase in the content of cyclopropane A19:0 and the decrease in the content of 18:1 acid in both lactic acid bacteria, as compared with the control cells grown without cerulenin (Tables 2 and 4), resulted in a decrease in the viability of the bacteria after freezing (Table 5). It can be concluded that the resistance of lactic acid bacteria (both lactobacilli and lactic streptococci) to freezing is dependent upon a favorable balance between de novo synthesis and other reactions (such as methylation and elongation) that could modify the different fatty acids. This balance results in a specific ratio between unsaturated and saturated fatty acids in each bacterium, which is important for maintaining viability after the freezing process.

8 496 GOLDBERG AND ESCHAR LITERATURE CITED 1. Accolas, J. P., and J. Auclair Storage of highly concentrated suspensions of lactic acid bacteria in the frozen state. I. Mesophilic lactic acid bacteria. Lait 47: Arison, B. H., and S. Omura Revised structure of cerulenin. J. Antibiot. 27: Awaya, J., T. Ohno, and S. Omura Substitution of cellular fatty acids in yeast cells by the antibiotic cerulenin and exogenous fatty acids. Biochim. Biophys. Acta 409: Christie, W. W Cyclopropane and cyclopropene fatty acids, p In F. D. Gunstone (ed.), Topics in lipid chemistry, vol. 1. John Wiley and Sons, Inc., New York. 5. Cowman, R. A., and M. L. Speck Ultra-low temperature storage of lactic streptococci. J. Dairy Sci. 48: D'Agnolo, G., J. S. Rosenfeld, J. Awaya, S. Omura, and P. R. Vagelos Inhibition of fatty acid synthesis by the antibiotic cerulenin. Specific inactivation of,3- ketoacyl-acyl carrier protein synthetase. Biochim. Biophys. Acta 326: Gilliland, S. E., and M. L. Speck Relationship of cellular components to the stability of concentrated lactic streptococcus cultures at -17 C. Appl. Microbiol. 27: Goldberg, I., J. R. Walker, and K. Bloch Inhibition of lipid synthesis in Escherichia coli cells by the antibiotic cerulenin. Antimicrob. Ag. Chemother. 3: Goldfine, H Use of a filter-paper disk assay in the measurement of lipid biosynthesis. J. Lipid Res. 7: Halper, L. A., and S. J. Norton Regulation of cyclopropane fatty acid biosynthesis by variations in enzyme activities. Biochem. Biophys. Res. Commun. 62: Harder, M. E., J. R. Beacham, J. E. Cronan, Jr., K. Beacham, J. L. Honegger, and D. F. Silbert Temperature-sensitive mutants of Escherichia coli requiring saturated and unsaturated fatty acids for growth: isolation and properties. Proc. Natl. Acad. Sci. U.S.A. 69: Kodicek, E Aspects of the constitution of bacterial membranes, p In N. E. Gibbons (ed.), APPL. ENVIRON. MICROBIOL. Recent progress in microbiology, vol. 8. University of Toronto Press, Toronto, Ontario, Canada. 13. Lamprech, E. D., and E. M. Foster The survival of starter organisms in concentrated suspensions. J. Appl. Bacteriol. 26: Liu, T. Y., and K. Hoffman Cyclopropane ring biosynthesis. Biochemistry 1: Matsumae, A., S. Nomura, and T. Hata Studies on cerulenin. IV. Biological characteristics of cerulenin. J. Antibiot. Ser. A 17: Moss, C. W., and M. L. Speck Injury and death of Streptococcus lactis due to freezing and frozen storage. Appl. Microbiol. 11: Moss, C. W., M. A. Lambert, and W. H. Merwin Comparison of rapid methods for analysis of bacterial fatty acids. Appl. Microbiol. 28: O'Leary, W. M Studies of the utilization of 14Clabeled octadecanoic acid by Lactobacillus arabinosus. J. Bacteriol. 77: Omura, S., M. Katagiri, A. Nakagawa, Y. Sano, S. Nomura, and T. Hata Studies on cerulenin. V. Structure of cerulenin. J. Antibiot. Ser. A 20: Omura, S., A. Nakagawa, K. Dekikawa, M. Otani, and T. Hata Studies on cerulenin. VI. Some spectroscopic features of cerulenin. Chem. Pharm. Bull. 17: Orme, T. W., J. McIntyre, F. Lynen, L. Kuhn, and E. Schwizer Fatty-acid elongation in a mutant of Saccharomyces cerevisiae deficient in fatty-acid synthetase. Eur. J. Biochem. 24: Smittle, R. B., S. E. Gilliland, and M. L. Speck Death of Lactobacillus bulgaricus resulting from liquid nitrogen freezing. Appl. Microbiol. 24: Smittle, R. B., S. E. Gilliland, M. L. Speck, and W. M. Walter, Jr Relationship of cellular fatty acid composition to survival of Lactobacillus bulgaricus in liquid nitrogen. Appl. Microbiol. 27: Stadhouders, J., L. A. Jansen, and G. Hup Preservation of starters and mass production of starter bacteria. Neth. Milk Dairy J. 23: Vance, D., I. Goldberg, K. Bloch, S. Omura, and S. Nomura Inhibition of fatty acid synthetases by the antibiotic cerulenin. Biochem. Biophys. Res. Commun. 48: Zatkin, H., J. H. Law, and H. Goldfine Enzymatic synthesis of cyclopropane fatty acids catalyzed by bacterial extracts. J. Biol. Chem. 238: