- Email: [email protected]
Influence of the glucose concentration on the yield of biomass and lipopolysaccharide in Salmonella cultures
Zbl. Bakt. 281, 30-37 (1994)
© Gustav Fischer Verlag, Stuttgart· Jena . New York
Influence of the Glucose Concentration on the Yield of Biomass and Lipopolysaccharide in Salmonella Cultures S. SCHLECHT and C. GALANOS Max-Planck-Institut fur Immunbiologie, Freiburg With 3 Figures· Received August 10, 1993 . Revision received October 20,1993· Accepted November 12, 1993
Summary Two Salmonella S-forms and two R-mutants were cultivated in complex medium supplemented with different amounts (0.5-3%) glucose. Cultivation was performed batchwise in a fermentor under aerobic conditions. With all strains investigated, the yield of bacterial mass increased with increasing concentration of glucose. In the case of three strains, the % LPS content of bacteria also increased, thus achieving the aim of this investigation. The synthesis of bacterial mass and LPS did not proceed in parallel and differed from strain to strain. At optimal glucose concentration, the yield of LPS could be increased up to 250%. The chemical composition of the LPS was independent of the glucose concentration. The individual strains exhibited an identical composition with regard to lipid A and polysaccharide when cultured at different glucose concentrations. The uniformity of the molecular distribution of LPS could also be confirmed by SDS-polyacrylamide gel electrophoresis. In the S-form LPSs, also the proportion of the unsubstituted R-form LPS was not affected by the glucose concentration in the culture medium. The present results demonstrate that optimisation of the cultivation conditions with respect to the glucose concentration of the medium would be of advantage especially for Salmonella strains that are cultivated frequently. Introduction Microbiological nutrient media usually contain glucose as an easily convertible source of carbon and energy. While already small quantities of glucose act as a growthpromoting factor (3), larger amounts lead to a marked increase in the yield of bacterial mass and in the amount of the metabolic products formed (2, 6). For optimum growth conditions, however, the concentration of glucose that can be employed in a given medium is subject to certain limitations. Thus, on the one hand, because of the necessary oxygen requirements of aerobic organisms in a dense culture (14), and, on the other, because of the increased accumulation in the medium of metabolites with toxic activity. It is known from the culturing of Klebsiella aerogenes in a chemostat (12) that glucose can be completely converted to biomass and CO 2 only when it is used as a limiting factor; a surplus always leads to an accumulation of incomplete breakdown
Mass Cultivation of Salmonella
31
products. If the normal products of glucose catabolism are mixed with a growing culture of E. coli, growth is brought to a standstill (10). In contrast to this, the accumulation of toxic products in cultures of Shigella sonnei only takes place to a limited extent. Otherwise, they are of minor importance; after glucose has been utilized, a further addition to the culture in the stationary growth phase results in a renewed and powerful increase of growth (6). In some cases, repetitive addition of glucose was carried out during the culturing process to regulate changes in pH caused by metabolites produced during cultivation and other parameters (7). It has been known for a long time that the glucose (or other sugar) level of the culture medium influences the amount of the polysaccharide in the microorganisms (1). Nevertheless, there is little information available about the action on the synthesis of microbial surface antigens (4, 8, 13, 17). The present communication deals with the influence of different levels of glucose concentration in the medium during the culturing of Salmonella on the yield in biomass and lipopolysaccharide (LPS) and the composition of the LPS.
Material and Methods Bacteria. The investigation was carried out on the S forms of S. typhimurium 1135 and S. dublin, as well as on the R mutants S. typhimurium his 386 (chemotype Ra) and S. typhimurium SLl032 (chemotype Rdl). The two R mutants were descendants of the wild type strain S. typhimurium LT2. Culture conditions. 1 litre of the final medium contained the following ingredients: 15 g tryptone (Difco), 15 g tryptose (Difco), 10 g yeast extract (Difco), 3 g NaC!, 8 g Na2HP04 . 12H20, 2 g MgS04 . 7H 20 and 0.3 ml polyglycol nooo (Hedinger, Stuttgart). Glucose and MgS04 . 7H 2 0 were sterilized separately. For glucose, dextrose monohydrate cerelose (Maizena) was used in concentrations of 5, 10, 20 and 30 g per litre, which resulted in an osmotic pressure of 422,450,500 and 551 mOsm, respectively, as determined with a semimicro osmometer (Knauer, Berlin). Cultivation was carried out batch-wise in a laboratory fermentor (Eschweiler, Kiel) in 6litre quantities. For inoculation, 300ml of a shaking culture (16h, 37°C) containing 0.3% glucose were used. In the fermentor vessel, an aeration rate of 1 litre per litre medium per minute (LiL x min) was used, increasing the rate to 2 litres where necessary (see text). The fermentor was equipped with a p02 regulation device maintaining the oxygen partial pressure at 240 mm Hg p02. The pH was kept at 7.2 by adding NaOH or HCl (IN) and a constant temperature of 37 DC was maintained during the cultivation. Culture was terminated when the stationary phase of growth had been reached and the bacteria were killed with phenol (1 % end concentration). Yield of biomass and LPS. The dry weight of the bacteria was determined gravimetrically (15). The LPS was extracted from the S form bacteria by the phenol-water procedure at 65-68°C (23), and from the R mutants by the phenol!chloroformlpetroleum ether (PCP) method (5) using in each case 10 g of the respective ethanol dried bacteria. The yield was determined gravimetrically (1 S). Methods of chemical analysis. The gas chromatographic method for determining fatty acids and neutral sugars, and the colorimetric methods for determining glucosamine, abequose, tyvelose, 2-keto-.'l-deoxyoctonate and phosphorus have been described earlier (18). SDS polyacrylamide Rei-electrophoresis (SDS PAGE). The electrophoresis was carried out in flat gels as previously described in detail (19), using Laemmli's buffer system (9) and the LPS was stained with silver nitrate (22). For the densitometric analysis of the gels, a Quick Scan TLC Densitometer (DESAGA, Heidelberg) was used, with a 525 nm filter and a 0.1 X 1 mm slit.
32
S. Schlecht and C. Galanos Results Cultivation
A typical example of the course of cultivation is shown in Fig. 1. In order to maintain the required partial pressure of oxygen, the stirrer speed was automatically raised. In this way, it was increased to values of between 1000 and 1200 RpM, for the four strains cultured (Fig. 2). In the case where cultivation was performed in the presence of 3% glucose (or from 2% onwards in the case of R mutants), the aeration rate was increased additionally to between 1.5 and 2 LlL x min. In all experiments, it was therefore possible to maintain a pOz of 2: 40 mm Hg throughout the culturing period. With higher glucose concentration in the medium, the period required for a complete consumption of the glucose and thus the total duration of cultivation, was prolonged. With a glucose concentration of 0.5%, culturing came to an end after about 4-5% hours, and at 3% after about 6112-8314 hours. Yield of biomass and lipopolysaccharide
The biomass yield from all strains increased by about 50% to 70% when the glucose concentration was raised from 0.5% to 1% or 2%. Raising it to 3% led to a further increase in the yield in the case of 2 strains (Fig.2). The maximal increase in yield among the 4 strains thus amounted to about 60% to 100%.
~~ ~ .
CD
II
" Turbidity
\.
~
\
~
Temperature,
_--,--~'\~PHr2~--~\-:-~---:-;,:/-:t.:-l~-'""'~~ .,..../':....
/\
\.
!:
----\.., -...... , ' - Stirrer speed \,
6
5
4 H
0 U
3 r s
2
1
o
Fig. 1. Course of the batch culturing of S. dublin in a medium with 1% glucose at 37°C and pH 7.2. The regulation of pOl was set at 40 mm Hg and kept open upwards. Dropping of the oxygen partial pressure below the minimal value resulted in a stepwise increase of the stirrer speed. 1 Beginning of alteration of pH towards alkalinity as the glucose in the medium became exhausted 2 Entry of culture into the stationary growth phase.
Mass Cultivation of Salmonella Bacteria dry weighl gil
15
S Iml135 (S)
10
/ o
0-0-0
LPS conlenl
LPS yield
Cultivation
%
mg/I
15 10
1500
hrs 10
Max.
time
1000
.______0 0_0.._____0
500
33
_~"o
stirrer speed rpm 1300 1000 700
,o---~ ~ 0'-
300
0'
I
500~
5 du (5)
15
/e---
10
0-0
~o
.
...........
-.--.~.
----0
0- 0
10
100
0'
._____--e
,-~'
,,'-
-------
.
1300 1000 700
05 Slm hls386(Ra) /x_x
15 10
10
1000
500
x/x
200
15 10
5 1m. SL 1032 (Rdl)
"-,,.-----" ,/
15 10
x
/
10
~)(--)(
x-x
)C-)(~)(~-'-')( x'
1500 1000
1300 A--A
"-"~""""" ".,."", 3
x_ _ x _ x
x/
700
x"
10
500 300
1300 1000
%
3
1()(x)
700
/
A_A_A
%
Glucose concentration of medium
Fig. 2. Dry weight of bacteria, LPS content of bacteria, yield of cell-bound LPS, duration of the culturing ( - - stationary phase of growth, - - - - to exhaustion of glucose) and maximum stirrer speed for the batch culturing of 2 Salmonella S forms and 2 Salmonella R mutants.
The percentage content of LPS in the bacteria (Fig. 2) was reduced in the two S forms when the glucose content of the medium exceeded 0.5% or 1%. In contrast, it increased in the R mutants until the glucose concentration had reached 1 % and then remained at about this level. The maximal absolute yield of cell-bound LPS per litre of culture was achieved in the case of 3 strains at glucose concentrations ranging between 1 % and 3%. In this way; an increase in yield of 230% to 270% was obtained compared to the yield obtained with 0.5% glucose. The S. dublin strain proved to be an exception. At a glucose level of about 0.5% to 2%, the yield remained fairly constant and also significantly lower than with the other strains.
Chemistry of the lipopolysaccharides In their lipid and polysaccharide components in S forms or in the oligosaccharides in the R mutants, the lipopolysaccharides showed the typical composition that has been repeatedly described in earlier communications (16): The sugars of the O-repeating units of the polysaccharide chain - mannose, rhamnose and galactose - and the 3
Zbl. Bakt. 28111
34
S. Schlecht and C. Galanos
branched sugars, abequose (in S. typhimurium) and tyvelose (in S. dublin) were present in proportions of 1 : 1 : 1 : 1. 60% of abequose remained acetylated. The mean number of units in O-specific polysaccharide showed only slight variation in the two S forms. The proportion of glucosamine to heptose also remained constant in all strains. Migration of lipopolysaccharides in SDS PAGE
The electrophoretic separation of the LPS (Fig. 3) revealed no change in the pattern of migration depending on the glucose concentration in any of the strains. With both the S form LPSs, the distribution of LPS molecules of different molecular weights remained practically the same. The components of free unsubstituted R LPS also remained constant. The densitograms revealed an R LPS portion between 10.0% and 11. 7% from the lipopolysaccharides of S. typhimurium, and between 17.1 % and 20.3 % from those of S. dublin.
Fig. 3. Migration of lipopolysaccharides from typhimurium 1135 in SDS-PAGE silver staining. The bacteria were cultured in a medium with different glucose content (0.5%, 1%,2%, 3% from left to right).
s.
Discussion This investigation has demonstrated that after varying the glucose concentration of the culture medium, the biosynthesis of LPS undergoes greater quantitative alterations than does the biomass (see Table 1). This was shown to be true also for other culturing parameters (14, 15, 16). The synthesis of both LPS and biomass occassionally showed a divergent trend (Fig. 2), so that the maximal output of both could be maintained with different concentrations of glucose in the medium. This was particularly the case with the two S forms. The varying influence of the glucose concentration on growth of individual strains and an influence on the synthesis of antigens are also known for other gram-negative bacteria (13,21). In a biotechnological sense, the product in the cultures described here was cellbound LPS. The optimal glucose concentrations for this are listed in Table 1. They
Mass Cultivation of Salmonella
35
Table 1. Optimal glucose concentration for the biosynthesis of LPS and the duration of culturing for 4 strains of Salmonella Strain
S. typhimurium 1135 S. dublin S. typhimurium his386 S. typhimurium SLl032
Optimal glucose concentration for LPS synthesis %
Yield per litre of culture fluid
Increase factor corresponding to 0.5% glucose biomass LPS
Duration of the culturing
LPS mg
biomass g
1
1297
11,7
3,5
5h 14/
0,5
213
7,1
2*
750
12,7
3,7
1,7
6h 30/
3*
1105
13,1
3,3
2,0
7h 02/
1,5
4h 57'
* Identical with the optimum for the synthesis of biomass.
show that increases in yield of about 3.5 were achieved, but that the optimal glucose concentration was distributed over a relatively wide range of 0.5% to 3%. For routine culturing, a glucose concentration of 1 % is frequently used, which represents a compromise (compare Fig. 2). With this glucose concentration, an LPS decrease of about a third compared to the optimal yield was obtained in the case of S. typhimurium his 386 and S. typhimurium SLl032. An optimisation of the conditions in terms of the glucose concentration in the medium is therefore an obvious advantage with strains which are frequently cultured. Also, the effort involved in the subsequent extraction and purification of LPS, is more economical with a higher initial concentration of the product. With an optimized glucose concentration, the duration of the culturing was kept within reasonable limits (Table 1). The optimisation of the culture conditions may be furthermore improved if also optimal conditions of other culture parameters such as temperature (16) and pH (in preparation) are considered. Chemical analysis has shown that the composition of LPS both in the lipid and in the polysaccharide components is independent of the glucose concentration used in the medium. Its uniformity was also confirmed in SDS-PAGE. In the LPS of the S forms, the proportion of free R-LPS not substituted with 0 polysaccharide remained constant while this is usually altered significantly, for instance, by changes in the incubation temperature (20). We regard this also as the expression of a constant metabolism of LPS biosynthesis, proceeding independently of the glucose concentration under the experimental conditions described. An essential condition for the purposeful employment of higher glucose concentrations is the maintenance of optimal aerobic conditions throughout the entire course of the culture (14). With an efficient stirring equipment of the fermentor, this can be achieved without difficulty using a p02 regulating device. For stirring and mixing the aerated 6 litre cultures in our fermentor, fWO 6-leaf disc impellers with a relative diameter of 0.42 are employed. We should like to express our thanks to G. Backs and Mrs. Erika Metz for their careful work and assistance during the performance of the investigation.
36
S. Schlecht and C. Galanos References
1. Dagley, S. and E. A. Dawes: Factors influencing the polysaccharide content of Escherichia coli. Biochem. J. 45 (1949) 331-337 2. Ecker, R. E. and W. R. Lockhart: Relationships between initial nutrient concentration and total growth. J. Bact. 82 (1961) 80-84 3. Eyer, H. und H. W. v. Schrader-Beielstein: Uber ein das Bakterienwachstum fOrderndes Prinzip. Zbl. Bakt. I. Orig. 169 (1957) 108-113 4. Formal, S. B., L. S. Baron, and W. Spilman: The virulence and immunogenicity of Salmonella typhosa grown in continuous culture. J. Bact. 72 (1956) 168-175 5. Galanos, c., O. Luderitz, and O. Westphal: A new method for the extraction of Rlipopolysaccharides. Eur. J. Biochem. 9 (1969) 245-249 6. Johnson, R. B.: Some factors influencing the laboratory mass cultivation of Shigella sonnei. J. Bact. 66 (1953) 92-100 7. J06, 1., L. Zalay, I. Kiss,!. Fuvessy und S. Pusztai: Die Massenzuchtung pathogener Darmbakterien. I. Mitteilung: Der Zuchtungsvorgang. Zbl. Bakt. Orig. 176 (1959) 496-513 8. Lacey, B. W.: Non-genetic variation of surface antigens in Bordetella and other microorganisms, in: Microbial Reaction to Environment, pp. 343-390. Univ. Press, Cambridge (1961) 9. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685 10. Landwall, P.: and T. Holme: Removal of inhibitors of bacterial growth by dialysis culture. J. Gen. Microbiol. 103 (1977) 345-352 11. Monad,].: The growth of bacterial cultures. Ann. Rev. Microbiol. 3 (1949) 371-394 12. Neijssel, O. M. and D. W. Tempest: The regulation of carbohydrate metabolism in Klebsiella aerogenes NCTC 418 organisms, growing in chemostat culture. Arch. MicrobioI. 106 (1975) 251-258 13. Reissbrodt, R., G. Kemmer und G. Seltmann: Abhangigkeit der Ausbildung bestimmter Antigene von Proteus mirabilis Stamm 1095167 beim Wachstum auf Minimalnahrmedien. Z. Allgem. Mikrobiol. 15 (1975) 357-370 14. Schlecht, S.: Wachstum und Lipopolysaccharid-Gehalt von Salmonellen bei Submerszuchtungen nach dem "batch"-Verfahren. 1. Zuchtungen bei Anwendung unterschiedlicher Beluftungsbedingungen. Zbl. Bakt. Hyg., I. Abt. Orig. A 215 (1970) 51-72 15. Schlecht, S.: Wachstum und Lipopolysaccharidgehalt von Salmonellen bei Submerszuchtungen nach dem Batch-Verfahren. 2. Mitteilung: Einflug der Wachstumsphase auf die Lipopolysaccharidsynthese. Zbl. Bakt. Hyg., I. Abt. Orig. A 232 (1975) 61-72 16. Schlecht, S.: Einflug der Zuchtungstemperatur auf die Produktion von Biomasse und Lipopolysaccharid sowie auf die chemische Zusammensetzung und Antigenitat der Lipopolysaccharide von S-Formen und R-Mutanten des Genus Salmonella. Zbl. Bakt. Hyg. A 261 (1986) 147-160 17. Schlecht, S. und O. Westphal: Wachstum und Lipopolysaccharid-(O-Antigen)-Gehalt von Salmonellen bei Zuchtungen auf Agarnahrboden. Zbl. Bakt. I. Abt. Orig. 200 (1966) 241-259 18. Schlecht, S. und I. Fromme: Chemische und serologische Eigenschaften von SalmonellaLipopolysacchariden aus unterschiedlichen Wachstumsphasen. Zbl. Bakt. Hyg., I. Abt. Orig. A 233 (1975) 199-222 19. Schlecht, S., M. A. Freudenberg, and C. Galanos: Immunoblot analysis of the R-form lipopolysaccharide from Salmonella S-forms. Zbl. Bakt. 227 (1992) 288-295 20. Schlecht, S. and H. Mayer: The influence of low growth temperature on the expression of R-core determinants and on the length of O-chains in lipopolysaccharides of Salmonella enterica serovars. Zbl. Bakt., 280/4 (1994) 448-457 21. Shida, T., K. Komagata, and K. Mitsugi: Reduction of lag time in bacterial growth. 2. Effects of inoculum size, and glucose and sodium chloride supplemented in culture
Mass Cultivation of Salmonella
37
media, initial pH of media, and cultural temperature. J. Gen. Appl. Microbiol. 21 (1975) 293-303 22. Tsai, C. M. and C. E. Frasch: A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Ann. Biochem. 119 (1982) 115-119 23. Westphal, 0., O. Luderitz und F. Bister: Ober die Extraktion von Bakterien mit Phenol! Wasser. Z. Naturforsch. 7h (1952) 148-155 Dr. Siegfried Schlecht und Dr. Chris Galanos, Max-Planck-Institut fur Immunbiologie, Postfach 1169, D-79108 Freiburg, Germany
© Gustav Fischer Verlag, Stuttgart· Jena . New York
Influence of the Glucose Concentration on the Yield of Biomass and Lipopolysaccharide in Salmonella Cultures S. SCHLECHT and C. GALANOS Max-Planck-Institut fur Immunbiologie, Freiburg With 3 Figures· Received August 10, 1993 . Revision received October 20,1993· Accepted November 12, 1993
Summary Two Salmonella S-forms and two R-mutants were cultivated in complex medium supplemented with different amounts (0.5-3%) glucose. Cultivation was performed batchwise in a fermentor under aerobic conditions. With all strains investigated, the yield of bacterial mass increased with increasing concentration of glucose. In the case of three strains, the % LPS content of bacteria also increased, thus achieving the aim of this investigation. The synthesis of bacterial mass and LPS did not proceed in parallel and differed from strain to strain. At optimal glucose concentration, the yield of LPS could be increased up to 250%. The chemical composition of the LPS was independent of the glucose concentration. The individual strains exhibited an identical composition with regard to lipid A and polysaccharide when cultured at different glucose concentrations. The uniformity of the molecular distribution of LPS could also be confirmed by SDS-polyacrylamide gel electrophoresis. In the S-form LPSs, also the proportion of the unsubstituted R-form LPS was not affected by the glucose concentration in the culture medium. The present results demonstrate that optimisation of the cultivation conditions with respect to the glucose concentration of the medium would be of advantage especially for Salmonella strains that are cultivated frequently. Introduction Microbiological nutrient media usually contain glucose as an easily convertible source of carbon and energy. While already small quantities of glucose act as a growthpromoting factor (3), larger amounts lead to a marked increase in the yield of bacterial mass and in the amount of the metabolic products formed (2, 6). For optimum growth conditions, however, the concentration of glucose that can be employed in a given medium is subject to certain limitations. Thus, on the one hand, because of the necessary oxygen requirements of aerobic organisms in a dense culture (14), and, on the other, because of the increased accumulation in the medium of metabolites with toxic activity. It is known from the culturing of Klebsiella aerogenes in a chemostat (12) that glucose can be completely converted to biomass and CO 2 only when it is used as a limiting factor; a surplus always leads to an accumulation of incomplete breakdown
Mass Cultivation of Salmonella
31
products. If the normal products of glucose catabolism are mixed with a growing culture of E. coli, growth is brought to a standstill (10). In contrast to this, the accumulation of toxic products in cultures of Shigella sonnei only takes place to a limited extent. Otherwise, they are of minor importance; after glucose has been utilized, a further addition to the culture in the stationary growth phase results in a renewed and powerful increase of growth (6). In some cases, repetitive addition of glucose was carried out during the culturing process to regulate changes in pH caused by metabolites produced during cultivation and other parameters (7). It has been known for a long time that the glucose (or other sugar) level of the culture medium influences the amount of the polysaccharide in the microorganisms (1). Nevertheless, there is little information available about the action on the synthesis of microbial surface antigens (4, 8, 13, 17). The present communication deals with the influence of different levels of glucose concentration in the medium during the culturing of Salmonella on the yield in biomass and lipopolysaccharide (LPS) and the composition of the LPS.
Material and Methods Bacteria. The investigation was carried out on the S forms of S. typhimurium 1135 and S. dublin, as well as on the R mutants S. typhimurium his 386 (chemotype Ra) and S. typhimurium SLl032 (chemotype Rdl). The two R mutants were descendants of the wild type strain S. typhimurium LT2. Culture conditions. 1 litre of the final medium contained the following ingredients: 15 g tryptone (Difco), 15 g tryptose (Difco), 10 g yeast extract (Difco), 3 g NaC!, 8 g Na2HP04 . 12H20, 2 g MgS04 . 7H 20 and 0.3 ml polyglycol nooo (Hedinger, Stuttgart). Glucose and MgS04 . 7H 2 0 were sterilized separately. For glucose, dextrose monohydrate cerelose (Maizena) was used in concentrations of 5, 10, 20 and 30 g per litre, which resulted in an osmotic pressure of 422,450,500 and 551 mOsm, respectively, as determined with a semimicro osmometer (Knauer, Berlin). Cultivation was carried out batch-wise in a laboratory fermentor (Eschweiler, Kiel) in 6litre quantities. For inoculation, 300ml of a shaking culture (16h, 37°C) containing 0.3% glucose were used. In the fermentor vessel, an aeration rate of 1 litre per litre medium per minute (LiL x min) was used, increasing the rate to 2 litres where necessary (see text). The fermentor was equipped with a p02 regulation device maintaining the oxygen partial pressure at 240 mm Hg p02. The pH was kept at 7.2 by adding NaOH or HCl (IN) and a constant temperature of 37 DC was maintained during the cultivation. Culture was terminated when the stationary phase of growth had been reached and the bacteria were killed with phenol (1 % end concentration). Yield of biomass and LPS. The dry weight of the bacteria was determined gravimetrically (15). The LPS was extracted from the S form bacteria by the phenol-water procedure at 65-68°C (23), and from the R mutants by the phenol!chloroformlpetroleum ether (PCP) method (5) using in each case 10 g of the respective ethanol dried bacteria. The yield was determined gravimetrically (1 S). Methods of chemical analysis. The gas chromatographic method for determining fatty acids and neutral sugars, and the colorimetric methods for determining glucosamine, abequose, tyvelose, 2-keto-.'l-deoxyoctonate and phosphorus have been described earlier (18). SDS polyacrylamide Rei-electrophoresis (SDS PAGE). The electrophoresis was carried out in flat gels as previously described in detail (19), using Laemmli's buffer system (9) and the LPS was stained with silver nitrate (22). For the densitometric analysis of the gels, a Quick Scan TLC Densitometer (DESAGA, Heidelberg) was used, with a 525 nm filter and a 0.1 X 1 mm slit.
32
S. Schlecht and C. Galanos Results Cultivation
A typical example of the course of cultivation is shown in Fig. 1. In order to maintain the required partial pressure of oxygen, the stirrer speed was automatically raised. In this way, it was increased to values of between 1000 and 1200 RpM, for the four strains cultured (Fig. 2). In the case where cultivation was performed in the presence of 3% glucose (or from 2% onwards in the case of R mutants), the aeration rate was increased additionally to between 1.5 and 2 LlL x min. In all experiments, it was therefore possible to maintain a pOz of 2: 40 mm Hg throughout the culturing period. With higher glucose concentration in the medium, the period required for a complete consumption of the glucose and thus the total duration of cultivation, was prolonged. With a glucose concentration of 0.5%, culturing came to an end after about 4-5% hours, and at 3% after about 6112-8314 hours. Yield of biomass and lipopolysaccharide
The biomass yield from all strains increased by about 50% to 70% when the glucose concentration was raised from 0.5% to 1% or 2%. Raising it to 3% led to a further increase in the yield in the case of 2 strains (Fig.2). The maximal increase in yield among the 4 strains thus amounted to about 60% to 100%.
~~ ~ .
CD
II
" Turbidity
\.
~
\
~
Temperature,
_--,--~'\~PHr2~--~\-:-~---:-;,:/-:t.:-l~-'""'~~ .,..../':....
/\
\.
!:
----\.., -...... , ' - Stirrer speed \,
6
5
4 H
0 U
3 r s
2
1
o
Fig. 1. Course of the batch culturing of S. dublin in a medium with 1% glucose at 37°C and pH 7.2. The regulation of pOl was set at 40 mm Hg and kept open upwards. Dropping of the oxygen partial pressure below the minimal value resulted in a stepwise increase of the stirrer speed. 1 Beginning of alteration of pH towards alkalinity as the glucose in the medium became exhausted 2 Entry of culture into the stationary growth phase.
Mass Cultivation of Salmonella Bacteria dry weighl gil
15
S Iml135 (S)
10
/ o
0-0-0
LPS conlenl
LPS yield
Cultivation
%
mg/I
15 10
1500
hrs 10
Max.
time
1000
.______0 0_0.._____0
500
33
_~"o
stirrer speed rpm 1300 1000 700
,o---~ ~ 0'-
300
0'
I
500~
5 du (5)
15
/e---
10
0-0
~o
.
...........
-.--.~.
----0
0- 0
10
100
0'
._____--e
,-~'
,,'-
-------
.
1300 1000 700
05 Slm hls386(Ra) /x_x
15 10
10
1000
500
x/x
200
15 10
5 1m. SL 1032 (Rdl)
"-,,.-----" ,/
15 10
x
/
10
~)(--)(
x-x
)C-)(~)(~-'-')( x'
1500 1000
1300 A--A
"-"~""""" ".,."", 3
x_ _ x _ x
x/
700
x"
10
500 300
1300 1000
%
3
1()(x)
700
/
A_A_A
%
Glucose concentration of medium
Fig. 2. Dry weight of bacteria, LPS content of bacteria, yield of cell-bound LPS, duration of the culturing ( - - stationary phase of growth, - - - - to exhaustion of glucose) and maximum stirrer speed for the batch culturing of 2 Salmonella S forms and 2 Salmonella R mutants.
The percentage content of LPS in the bacteria (Fig. 2) was reduced in the two S forms when the glucose content of the medium exceeded 0.5% or 1%. In contrast, it increased in the R mutants until the glucose concentration had reached 1 % and then remained at about this level. The maximal absolute yield of cell-bound LPS per litre of culture was achieved in the case of 3 strains at glucose concentrations ranging between 1 % and 3%. In this way; an increase in yield of 230% to 270% was obtained compared to the yield obtained with 0.5% glucose. The S. dublin strain proved to be an exception. At a glucose level of about 0.5% to 2%, the yield remained fairly constant and also significantly lower than with the other strains.
Chemistry of the lipopolysaccharides In their lipid and polysaccharide components in S forms or in the oligosaccharides in the R mutants, the lipopolysaccharides showed the typical composition that has been repeatedly described in earlier communications (16): The sugars of the O-repeating units of the polysaccharide chain - mannose, rhamnose and galactose - and the 3
Zbl. Bakt. 28111
34
S. Schlecht and C. Galanos
branched sugars, abequose (in S. typhimurium) and tyvelose (in S. dublin) were present in proportions of 1 : 1 : 1 : 1. 60% of abequose remained acetylated. The mean number of units in O-specific polysaccharide showed only slight variation in the two S forms. The proportion of glucosamine to heptose also remained constant in all strains. Migration of lipopolysaccharides in SDS PAGE
The electrophoretic separation of the LPS (Fig. 3) revealed no change in the pattern of migration depending on the glucose concentration in any of the strains. With both the S form LPSs, the distribution of LPS molecules of different molecular weights remained practically the same. The components of free unsubstituted R LPS also remained constant. The densitograms revealed an R LPS portion between 10.0% and 11. 7% from the lipopolysaccharides of S. typhimurium, and between 17.1 % and 20.3 % from those of S. dublin.
Fig. 3. Migration of lipopolysaccharides from typhimurium 1135 in SDS-PAGE silver staining. The bacteria were cultured in a medium with different glucose content (0.5%, 1%,2%, 3% from left to right).
s.
Discussion This investigation has demonstrated that after varying the glucose concentration of the culture medium, the biosynthesis of LPS undergoes greater quantitative alterations than does the biomass (see Table 1). This was shown to be true also for other culturing parameters (14, 15, 16). The synthesis of both LPS and biomass occassionally showed a divergent trend (Fig. 2), so that the maximal output of both could be maintained with different concentrations of glucose in the medium. This was particularly the case with the two S forms. The varying influence of the glucose concentration on growth of individual strains and an influence on the synthesis of antigens are also known for other gram-negative bacteria (13,21). In a biotechnological sense, the product in the cultures described here was cellbound LPS. The optimal glucose concentrations for this are listed in Table 1. They
Mass Cultivation of Salmonella
35
Table 1. Optimal glucose concentration for the biosynthesis of LPS and the duration of culturing for 4 strains of Salmonella Strain
S. typhimurium 1135 S. dublin S. typhimurium his386 S. typhimurium SLl032
Optimal glucose concentration for LPS synthesis %
Yield per litre of culture fluid
Increase factor corresponding to 0.5% glucose biomass LPS
Duration of the culturing
LPS mg
biomass g
1
1297
11,7
3,5
5h 14/
0,5
213
7,1
2*
750
12,7
3,7
1,7
6h 30/
3*
1105
13,1
3,3
2,0
7h 02/
1,5
4h 57'
* Identical with the optimum for the synthesis of biomass.
show that increases in yield of about 3.5 were achieved, but that the optimal glucose concentration was distributed over a relatively wide range of 0.5% to 3%. For routine culturing, a glucose concentration of 1 % is frequently used, which represents a compromise (compare Fig. 2). With this glucose concentration, an LPS decrease of about a third compared to the optimal yield was obtained in the case of S. typhimurium his 386 and S. typhimurium SLl032. An optimisation of the conditions in terms of the glucose concentration in the medium is therefore an obvious advantage with strains which are frequently cultured. Also, the effort involved in the subsequent extraction and purification of LPS, is more economical with a higher initial concentration of the product. With an optimized glucose concentration, the duration of the culturing was kept within reasonable limits (Table 1). The optimisation of the culture conditions may be furthermore improved if also optimal conditions of other culture parameters such as temperature (16) and pH (in preparation) are considered. Chemical analysis has shown that the composition of LPS both in the lipid and in the polysaccharide components is independent of the glucose concentration used in the medium. Its uniformity was also confirmed in SDS-PAGE. In the LPS of the S forms, the proportion of free R-LPS not substituted with 0 polysaccharide remained constant while this is usually altered significantly, for instance, by changes in the incubation temperature (20). We regard this also as the expression of a constant metabolism of LPS biosynthesis, proceeding independently of the glucose concentration under the experimental conditions described. An essential condition for the purposeful employment of higher glucose concentrations is the maintenance of optimal aerobic conditions throughout the entire course of the culture (14). With an efficient stirring equipment of the fermentor, this can be achieved without difficulty using a p02 regulating device. For stirring and mixing the aerated 6 litre cultures in our fermentor, fWO 6-leaf disc impellers with a relative diameter of 0.42 are employed. We should like to express our thanks to G. Backs and Mrs. Erika Metz for their careful work and assistance during the performance of the investigation.
36
S. Schlecht and C. Galanos References
1. Dagley, S. and E. A. Dawes: Factors influencing the polysaccharide content of Escherichia coli. Biochem. J. 45 (1949) 331-337 2. Ecker, R. E. and W. R. Lockhart: Relationships between initial nutrient concentration and total growth. J. Bact. 82 (1961) 80-84 3. Eyer, H. und H. W. v. Schrader-Beielstein: Uber ein das Bakterienwachstum fOrderndes Prinzip. Zbl. Bakt. I. Orig. 169 (1957) 108-113 4. Formal, S. B., L. S. Baron, and W. Spilman: The virulence and immunogenicity of Salmonella typhosa grown in continuous culture. J. Bact. 72 (1956) 168-175 5. Galanos, c., O. Luderitz, and O. Westphal: A new method for the extraction of Rlipopolysaccharides. Eur. J. Biochem. 9 (1969) 245-249 6. Johnson, R. B.: Some factors influencing the laboratory mass cultivation of Shigella sonnei. J. Bact. 66 (1953) 92-100 7. J06, 1., L. Zalay, I. Kiss,!. Fuvessy und S. Pusztai: Die Massenzuchtung pathogener Darmbakterien. I. Mitteilung: Der Zuchtungsvorgang. Zbl. Bakt. Orig. 176 (1959) 496-513 8. Lacey, B. W.: Non-genetic variation of surface antigens in Bordetella and other microorganisms, in: Microbial Reaction to Environment, pp. 343-390. Univ. Press, Cambridge (1961) 9. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685 10. Landwall, P.: and T. Holme: Removal of inhibitors of bacterial growth by dialysis culture. J. Gen. Microbiol. 103 (1977) 345-352 11. Monad,].: The growth of bacterial cultures. Ann. Rev. Microbiol. 3 (1949) 371-394 12. Neijssel, O. M. and D. W. Tempest: The regulation of carbohydrate metabolism in Klebsiella aerogenes NCTC 418 organisms, growing in chemostat culture. Arch. MicrobioI. 106 (1975) 251-258 13. Reissbrodt, R., G. Kemmer und G. Seltmann: Abhangigkeit der Ausbildung bestimmter Antigene von Proteus mirabilis Stamm 1095167 beim Wachstum auf Minimalnahrmedien. Z. Allgem. Mikrobiol. 15 (1975) 357-370 14. Schlecht, S.: Wachstum und Lipopolysaccharid-Gehalt von Salmonellen bei Submerszuchtungen nach dem "batch"-Verfahren. 1. Zuchtungen bei Anwendung unterschiedlicher Beluftungsbedingungen. Zbl. Bakt. Hyg., I. Abt. Orig. A 215 (1970) 51-72 15. Schlecht, S.: Wachstum und Lipopolysaccharidgehalt von Salmonellen bei Submerszuchtungen nach dem Batch-Verfahren. 2. Mitteilung: Einflug der Wachstumsphase auf die Lipopolysaccharidsynthese. Zbl. Bakt. Hyg., I. Abt. Orig. A 232 (1975) 61-72 16. Schlecht, S.: Einflug der Zuchtungstemperatur auf die Produktion von Biomasse und Lipopolysaccharid sowie auf die chemische Zusammensetzung und Antigenitat der Lipopolysaccharide von S-Formen und R-Mutanten des Genus Salmonella. Zbl. Bakt. Hyg. A 261 (1986) 147-160 17. Schlecht, S. und O. Westphal: Wachstum und Lipopolysaccharid-(O-Antigen)-Gehalt von Salmonellen bei Zuchtungen auf Agarnahrboden. Zbl. Bakt. I. Abt. Orig. 200 (1966) 241-259 18. Schlecht, S. und I. Fromme: Chemische und serologische Eigenschaften von SalmonellaLipopolysacchariden aus unterschiedlichen Wachstumsphasen. Zbl. Bakt. Hyg., I. Abt. Orig. A 233 (1975) 199-222 19. Schlecht, S., M. A. Freudenberg, and C. Galanos: Immunoblot analysis of the R-form lipopolysaccharide from Salmonella S-forms. Zbl. Bakt. 227 (1992) 288-295 20. Schlecht, S. and H. Mayer: The influence of low growth temperature on the expression of R-core determinants and on the length of O-chains in lipopolysaccharides of Salmonella enterica serovars. Zbl. Bakt., 280/4 (1994) 448-457 21. Shida, T., K. Komagata, and K. Mitsugi: Reduction of lag time in bacterial growth. 2. Effects of inoculum size, and glucose and sodium chloride supplemented in culture
Mass Cultivation of Salmonella
37
media, initial pH of media, and cultural temperature. J. Gen. Appl. Microbiol. 21 (1975) 293-303 22. Tsai, C. M. and C. E. Frasch: A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Ann. Biochem. 119 (1982) 115-119 23. Westphal, 0., O. Luderitz und F. Bister: Ober die Extraktion von Bakterien mit Phenol! Wasser. Z. Naturforsch. 7h (1952) 148-155 Dr. Siegfried Schlecht und Dr. Chris Galanos, Max-Planck-Institut fur Immunbiologie, Postfach 1169, D-79108 Freiburg, Germany