Appl Microbiol Biotechnol (1984) 20:256-261

'erobiology Biotechnology © Springer-Veriag 1984

Initiation of solvent production, clostridial stage and endospore formation in Clostridium acetobutylicum

Susan Long, David T. Jones, and David R. Woods

Department of Microbiology, University of Cape Town, Rondebosch 7700, South Africa

P262

Summary. A minimal medium was used to investigate the triggers regulating the initiation of solvent production and differentiation in Clostridium aceto- butylicum P262. The accumulation of acid end-prod- ucts caused the inhibition of cell division and the initiation of solvent production and cell differentia- tion. Initiation only occurred within a narrow pH range. Glucose or ammonium limited cultures failed to achieve the necessary threshold of acid end-prod- ucts and solvent production and differentiation were not initiated. The addition of acid end-products or ammonium to cultures containing suboptimal levels of glucose or nitrogen respectively, enhanced solvent production. Resuspension of cells in media contain- ing the threshold level of acid end-products and residual glucose induced endospore formation. Glu- cose or ammonium limitation did not induce sporu- lation and there was a requirement for glucose and ammonium during solventogenesis and endospore formation. Initiation of solvent production and clostridial stage formation were essential for sporu- lation. The induction of endospore formation in C. acetobutylicum P262 differs from that in the aerobic endospore forming bacteria where sporula- tion is initiated by nutrient starvation.

Introduction

An understanding of the triggers regulating the transition from the acidogenic phase to the solven- togenic phase in Clostridium acetobutylicum is impor- tant for future developments aimed at improving the industrial production of acetone, butanol, and etha- nol (ABE) by fermentation. The solventogenic phase

Offprint requests to: D. R. Woods

is complicated by the induction of endospore for- mation and the differentiation of the bacterium. The C. acetobutylicum P262 strain differs from the type culture strain, ATCC 824, in that it has a clearly defined clostridial stage (swollen, granulose-contain- ing, phase-bright, cigar-shaped cells surrounded by an extracellular polysaccharide capsule) which pre- cedes endospore formation. The differentiation of C. acetobutylicum P262 appears to be closely linked to solvent production and we reported the isolation of clostridial stage (cls) mutants which were unable to produce solvents and failed to sporulate (Jones et al. 1982). Gottschal and Morris (1981b) showed that a loss of spore-forming ability during continuous cul- ture was associated with a loss of the ability to produce solvents.

Although there have been a number of recent reports on the regulation of solvent production, the relationship to differentiation in C. acetobutylicum has not been investigated. Little or no solvents were formed by C. acetobutylicum grown in continuous culture under glucose or ammonium limitation (Gott- schal and Morris 1981b; Andersch et al. 1982; Bahl et al. 1982) and Monot et al. (1982) reported that minimal levels of glucose and potassium were nec- essary for solvent production in continuous culture. The importance of acetate and butyrate in triggering solvent production and in affecting the final solvent ratio has been reported (Davies and Stephenson 1941; Nakhmanovich and Shcheblykina 1960; Hfigg- str6m and Molin 1980; Gottschal and Morris 1981b; Bahl et al. 1982; Martin et al. 1983). A defined minimal medium (CAMM) (Long et al. 1983) which supported sporulation and solvent production was utilized to investigate the relationships between vegetative growth, clostridial stage formation, endo- spore formation, solvent production and the way in which glucose, ammonium and acid end-product concentration are involved in the regulation of these events.

S. Long et al.: Initiation of solvent production and sporulation in C. acetobutylicum 257

Materials and methods

Bacterial strain and media. The C. acetobutylicum P262 strain utilized in this study has been described previously (Jones et al. 1982; Robson and Jones 1982; Long et al. 1983). The bacterium was maintained as a spore suspension in sterile distilled water stored at 4 ° C. Spore suspensions were prepared from cultures grown anaerobically at 34°C on the buffered Clostridium basal medium (CBM), described by O'Brien and Morris (1971), solidified with 1.5% agar. After 3 days of incubation, the spores were scraped into distilled water, incubated with l m g of lysozyme/ml at 37 ° C for 30 min, and then washed three times in sterile distilled water. All manipulations were carried out under stringent anaerobic conditions in an anaerobic glove cabinet (Forma Scientific, Marietta, Ohio).

Initiation of solvent production and sporulation. The induction and inhibition of solvent production and differentiation were studied in the sporulation minimal medium (CAMM) described by Long et al. (1983). Spores were activated by heat shock at 75 ° C for 2 rain and then cooled on ice before inoculation of 5 ~tl of spore suspension into 10ml of CBM. CAMM was inoculated with exponential phase cells (optical density at 600 nm, 0.45) harvested from CBM and washed once with sterile distilled water. Cultures were incubated at 34 ° C and agitated every 6 -8 h by gentle mixing. Total bacterial counts, clostridial stage counts, and spore counts were determined with a Thoma counting chamber (Weber Scientific International, Lancing, England) and Zeiss photomi- croscope fitted with phase- and interference-contrast optics.

The presence of capsules was determined by negative staining with India ink, the presence of granulose was determined by staining with iodine, and the presence of forespores was deter- mined by the methods of Smith and Ellner (1957) and Hoeniger and Headley (1968).

6 0

. , ~ 4 5

(5

ua 3 0 q3 0 ° z 15 .,J

O

10 9

1 0 8

Z

1 0 7

1 0 6

'15

... s

ua 0 1 0 0

8o

g e oo

= ~ .o

~- I,- 2 0

i i ~ i i i i a

o / o - - o o - o • - -

o o _ - - - - o ~ / ~

i A//'// : b i i I i i i i

. . • . \ : \ D \ ° / , > / c

i F " J o

/ 7 ' ~ . ~ - . - - . - - i

i i J i i c o •

10 2 0 3 0 4 0 5 0 6 0 7 0 8 0

T I M E ( h )

8

7

67=

5

4

Fig. 1. Growth (a), physiological (b) and morphological (c) changes in C. acetobutylicum P262 grown in CAMM (60 g/l glucose), a (@) total cell count; (O) colony forming units; and (at) mature endospores, b ([]) residual glucose; ( I ) pH; (~) total acids; and (<5) total solvents, c (©) motility; (@) granulose; (at) clostridial forms; and (k 0 forespore septa

End-products and glucose determinations. Acetate, butyrate, acetone, butanol, and ethanol were measured using a Hewlett Packard 5880 A gas chromatograph with a flame ionization detector. A glass column (1.86 m x 4 ram) packed with 10% diethylene glycol adipate on 80-100 mesh chromosorb-W-HP (Supelco) was used with propanol as an internal standard. The conditions used for the determinations were as follows: column temperature 60-180°C progressive at 20 ° C/rain, detector tem- perature 300 ° C, carrier gas (N2) flow rate 30 ml/min. Glucose was determined by glucose oxidation of culture filtrates in a Beckman Glucose Analyser 2.

Results

Physiological and developmental changes associated with growth, solvent production, and sporutation in minimal medium

The production of solvents was determined in the defined CAMM of Long et al. (1983) which gave high levels of sporulation. In batch culture CAMM containing 50-60 g/1 glucose produced 3 - 5 x 108 endospores/ml (70-80%) and 10-16 g/1 of total solvents after 60-80 h (Fig. 1). ABE were produced in the ratio of 7 : 2 : <1. The cessation of solvent production coincided with the exhaustion of glucose although in some experiments residual glucose remained (up to 5 g/l). The pH breakpoint (pH 5.5)

demarcated the transition from the acidogenic phase to the solventogenic phase. During the acidogenic phase 25-40% of the glucose was consumed, resulting in the accumulation of 4 - 6 g/1 of total acids.

The shift to the solventogenic phase was charac- terised by the cessation of cell division, the onset of solvent production, granulose accumulation and the conversion of vegetative rods into clostridial forms (Fig. 1). The ability of the cells to form colonies decreased sharply at this stage but there was a gradual recovery in the number of colony forming units (c.f.u.) as sporulation proceeded. Although there was a decrease in c.f.u, the total biomass of the culture increased as the cells continued to accumulate granulose indicating that little or no loss in cell viability occurred. During the solventogenic phase the pH increased and solvent production was asso- ciated with enhanced glucose consumption while the concentration of acid end-products remained rela- tively constant.

Effect of nitrogen source

The effect of varying the type of ni trogen source on the production of solvents and differentiation was

258 S. Long et al.: Initiation of solvent production and sporulation in C. acetobutylicum

wo / • o ~

O~10 / / e / . /

LU 0 ~

0

b / . ~ J ;

10 20 3 0 4 0 5 0 6 0 0 10 2 0 3 0 4 0 5 0

CONCENTRATLON CONCENTRATION OF GLUCOSE G//L OF GLUCOSE G~L

.~60 . . . . . . . . . ~ ~ 600 = ~ ~ ~ - - - . . . .

/ Z ~ , 5 0 0 0

o~- v- soo ~ ~o=4 o 400 z

• ~ ~ a o 300 # 3oo ~ % =

1 0 0 r- 100 ~ ~Z3 a- 10 ~ "

w 0 I I IF 0

O 1 2 3 4 6 0 1 2 3 4 CONCENTRATION CONCENTRATION

nAP G/t_ nap G/~

0 60

Fig. 2. Effect of glucose on end-product formation (a), and growth, clostridial stage and endospore formation (b) in C. acetobutylicurn P262. Glucose (5-60 g/l) was added to CAMM cultures and samples were assayed at the end of the acidogenic phase (20-30 h) and at the end of the solventogenic phase (60-80 h). a (0) glucose consumed after 20-30 h; (O) glucose consumed after 60-80 h; (A) acid end-products after 20-30 h; (11) acid end-products after 60-80 h; (A) total solvents after 60-80 h. b ([3) total cell count; (II,) clostridial stage count; (O) endospore count

Fig. 3. Effect of ammonium on glucose utilization and end-product formation (a), and growth, clostridial stage and endospore formation (b) in C. acetobutylicum P262. Cells were grown in CAMM (60 g/1 glucose) containing different concentrations of DAP (0.06-6.00 g/l). Excess phosphate was present as K2H POg. Samples were assayed at the end of the acidogenic phase (20-30 h) and at the end of the solventogenic phase (60-80 h). a (©) glucose consumed after 20-30 h; (O) glucose consumed after 60-80h; (A) acid end-products after 20-30 h; (11) acid end-products after 60-80 h; (A) total solvents after 60-80 h. b ([Z) total cell count; (I~) clostridial form count; (O) endospore count

investigated in CAMM. Cultures containing diam- monium hydrogen phosphate (DAP), ammonium nitrate, ammonium acetate, ammonium sulphate (which were added at equivalent ammonium concen- trations) and yeast extract all produced constant and relatively high concentrations of solvents (13-15 g/l) and a high proportion of the cells accumulated granulose and formed clostridial stages (60-90%). However the nitrogen source affected endospore formation and 70-80% sporulation occurred in cultures containing DAP, ammonium nitrate and yeast extract. Endospore formation was variable with ammonium acetate (5-60%) and low levels of sporulation occurred in cultures containing ammoni- um sulphate (10-25%).

Effect of glucose and ammonium on the solventogenic phase and endospore formation

The effects of varying the glucose (Fig. 2) and ammonium (diammonium hydrogen phosphate, DAP) (Fig. 3) concentrations in CAMM on cell growth, the development of the solventogenic phase and endospore formation were investigated. The initiation of the solventogenic phase only occurred when glucose and DAP were present at concentra- tions above 10 and 0.3 g/1 respectively. When the glucose concentration was below this threshold level all the glucose was consumed during the acidogenic phase and cell yields were reduced (Fig. 2). At low levels of ammonium (< 0.3 g/1 DAP) the cell yields

were also reduced but only < 25% of the glucose was utilized (Fig. 3). The overall effects of glucose and ammonium limitation were similar in that the amount of acid end-products produced at the low glucose or ammonium concentrations was approximately 50% of that obtained with higher concentrations and only a small decrease in the pH of the cultures occurred. Under these conditions the cells remained as vege- tative rods and did not form clostridial stages, endospores or produce solvents.

The conversion of vegetative rods to the clostrid- ial stage and the production of solvents only occurred in cultures in which the concentration of glucose was above the threshold level and the amount of acid end-products was > 4 g/1 (Fig. 2). The proportion of cells which produced clostridial stages increased from approximately 1% at a glucose concentration of 10 g/1 to a maximum of 80-90% at a concentration of 40 g/1. Increased clostridial stage formation was accompanied by an increase in solvent production from < 1 g/l to > 12 g/1. Endospore formation was only observed in CAMM containing > 30 g/1 glucose and reached a maximum of 70-80% endospores at a concentration of 60 g/1 glucose.

Four levels of ammonium, 0.6, 2.0, 3.0, and 4.0 g/1 DAP, were identified which were required for maximum cell yield, clostridial stage formation, solvent production, and endospore formation respec- tively (Fig. 3). At low concentrations of ammonium (< 0.6 g/1 DAP), growth was limited, low concen- trations of acid end-products were produced, solvent production and differentiation were not initated.

S. Long et al.: Initiation of solvent production and sporulation in C. acetobutylicum 259

Table 1. Effect of addition of ammonia, DAP, potassium 40 phosphate, and acetate/butyrate on the production of solvents, clostridial forms, and endospores in C A M M (60 g/1 glucose) "#,30 containing D A P (1 g/l). The number of cells which formed clostridial forms and endospores is expressed as a percentage of the total count

Addition Con- Time Total Clos- Endo- cen- of ad- sol- tridial spores tration dition vents forms (g/l) (h) (g/l) (%) (%)

None - 0 7.57 80-90 < 1 NH 3 0.5 0 8.62 80-90 30-50 NH3 2.5 0 14.74 80-90 60-70 NH 3 2.5 24 11.72 80-90 40-50 D A P 5.0 24 11.25 70 -80 50-60

K2H PO4 7.8 0 7.34 70-80 < 1 KH2 PO 4 Acetate & 1 + 1 0 7.46 70 -80 < 5 butyrate

., 20 u~ 0

z lO O

0 0

i i i i i i

a

• [ 1 / I I

i i i l I i

b

2<\,_._. 0 10 2 0 3 0 4 0 5 0 6 0 0 10 20 30 40 50 60

T I M E (h)

Fig, 4a, b. Effect of acetate and butyrate on end-product formation in C. acetobutylicum P262 grown in CAMM (30 g/1 glucose), a No addition, b Addition of acetate (1 g/l) and butyrate (1 g/l) at 26 h ( ,~ ). ( I ) Residual glucose; (O) acetate; ( e ) butyrate + acetate; and ([~) total solvents

e - m 4 ~ z

.-t

if-

Table 2. Effect of glucose, acetate and butyrate concentrations on endospore formation in C. acetobutylicum P262. Exponential phase cells grown in CAMM (30 g/1 glucose) were harvested, washed, and resuspended in CAMM containing different concen- trations of glucose, acetate, and butyrate. The number of cells which formed endospores is expressed as a percentage of control cultures resuspended in C A M M (60 g/l glucose). Control cultures contained 1 . 5 - 2 x 10 s endospores/ml

Resuspension medium Endospores (%)

Glucose Acetate Butyrate g/1 g/l g/1

5 0 0 0 5 1 1 0 5 2 2 0

10 0 0 < 1 10 1 1 < 1 10 2 2 < 1

20 0 0 3 20 1 1 11 20 2 2 30

30 0 0 13 30 1 1 53 30 2 2 100 30 3 3 100 30 4 4 40

60 0 0 100

Although maximum cell yield ocurred at 0.6 g/1 DAP, solvent production and clostridial stage development was reduced and endospore formation did not occur. Maximum clostridial stage formation and solvent production were obtained at 2 and 3g/1 DAP respectively. Optimum endospore formation oc- curred at 4 g/1 DAP and there was a direct relation-

ship between endospore formation and ammonium concentration between 1 and 4 g/1 DAP.

The effect of adding extra ammonium in the form of ammonia or DAP was monitored in 60 g/1 glucose CAMM containing 1 g/1 DAP (Table 1). The addition of increasing concentrations of ammonium resulted in enhanced solvent production and an increase in endospore formation. Adding ammonium at the start was more effective than adding it at the end of the acidogenic phase. Solvent production and endospore formation were not affected by the addition of phosphate.

Effect of acid end-products on the solventogenic phase and endospore formation

The role of acetate and butyrate in the development of the solventogenic phase and clostridial stage and endospore production was investigated by adding the acids to cultures during the acidogenic phase (Fig. 4). In control cultures containing 30 g/1 glucose between 33-50% of the initial glucose still remained at the end of the acidogenic phase, 50-70% of the cells were converted to clostridial forms, 5 - 6 g/1 solvents were produced and < 10% endospore formation occurred. After the addition of i g/1 acetate and 1 g/1 butyrate the elevated level of acetate was maintained but the level of butyrate decreased. The total cell counts in the control and acid treated cultures remained the same but in the latter cultures there was an increase in both the number of clostridial forms (65-85%) and the concentration of solvents (7.5 g/l) after 60 h. Endospore formation was also increased

260 S. Long et al.: Initiation of solvent production and sporulation in C. acetobutylicum

from approximately 1% in the control to approxi- mately 25% in the acid treated cultures.

Cells grown to late exponential phase were also harvested and resuspended in CAMM containing different concentrations of glucose, acetate and butyrate to determine the effect on clostridial stage and endospore production (Table 2). Cells resus- pended in 5 g/1 glucose with or without the addition of acetate and butyrate failed to form clostridial stages and endospores. Cultures resuspended in increasing concentrations of glucose and acetate and butyrate showed a corresponding increase in the number of clostridial forms and endospores. The highest num- ber of clostridial forms and endospores occurred in CAMM containing 30 g/1 glucose to which either 2 or 3 g/1 acetate and 2 or 3 g/1 butyrate had been added. When higher concentrations of each of these acids (4 g/l) were added to 30 g/1 glucose the development of clostridial forms and endospores was decreased. The addition of acetate and butyrate to cultures containing 1 g/1 DAP had no effect on the enhance- ment of either solvent production or endospore formation (Table 1).

The resuspension technique was utilized to deter- mine whether there was an actual requirement for acetate and butyrate for the induction of the clostridial stage and endospore production. Late exponential phase cells grown in CAMM containing 30 g/1 glucose were resuspended in CAMM (30 g/1 glucose) containing either acetate (2 g/l) and butyrate (2 g/l) or the non-metabolisable acid HC103 (1.4 g/l) which produced similar pH values. Resuspension in acetate and butyrate resulted in a six-fold increase in clostridial stage and endospore formation but HC103 caused a six-fold decrease in clostridial stage and endospore formation. The addition of HC103 at the same concentration as butyrate and acetate caused a marked decrease in pH and cell viability.

Discussion

The use of a chemically defined medium which supported endospore formation and produced levels of solvents only slightly lower than those observed in the industrial fermentation and complex media (Spivey 1978; Jones et al. 1982; Robson and Jones 1982) facilitated a study of factors involved in the initiation of solvent production and differentiation in C. acetobutylicum P262. The shift to the solventog- enic phase was linked both to the cessation of cell division and the triggering of various physiological and morphological changes. Anaerobic bacteria are strongly inhibited when the end-products resulting from metabolism are present at relatively low con- centrations (Herrero 1983) and in C. acetobutylicum

P262 cell division was inhibited when 4 -5 g/1 of acid end-products were present. At this stage none of the nutrients were limiting and the enhanced glucose consumption which was observed following the cessation of growth was indicative of inhibition by acid end-products. The inhibition of cell division by acetate and butyrate was associated with the initiation of solvent production and differentiation. The involvement of acetate and butyrate in triggering solvent production and differentiation is indicated by the results of our studies on the resuspension of vegetative cells in the acids and the addition of the acids to non-differentiated cultures. At low glucose or ammonium concentrations growth was inhibited by nutrient limitation rather than by the accumulation of acid end-products and the conversion to the solven- togenic phase and cell differentiation did not occur. The involvement of acetate and butyrate in triggering solvent production, but not differentiation, in C. ace- tobutylicum has been suggested by Gottschal and Morris (1981a); Bahl et al. (1982); Martin et al. (1983) and Yu and Saddler (1983).

The absence of solvent production in ammonium limited batch cultures of C. acetobutylicum P262 is in agreement with the results of Gottschal and Morris (1981b) and Andersch et al. (1982) who did not obtain solvents in ammonium limited chemostat cultures. However Monot and Engasser (1983) reported the production of significant levels of acetone and butanol (8.5 g/l) by C. acetobutylicum ATCC 824 under ammonium limitation in batch and continuous cultures. Although the results using a different strain are not necessarily directly compar- able, our results suggest that the ammonium con- centration used by Monot and Engasser (1983) (0.18 g/1 ammonium) was sufficient for the pro- duction of the threshold concentration of acid end-products and the initiation of solvent production. It is interesting that in our experiments with C. ace- tobutylicum P262 a solvent concentration of 8.5 g/1 was also obtained at 0.21 g/1 ammonium.

In addition to the concentration of acetate and butyrate the pH of the culture also plays an important role in the successful triggering of the second phase of the fermentation and in differentiation. The results reported by Gottschal and Morris (1981a) and our results using a non-metabolisable acid to replace acetate and butyrate indicate that the attainment of a low pH alone is not sufficient to initiate the shift to the second phase of the fermentation. Although HC103 reduced the pH of the culture we do not know what other effects it may have on C. acetobutylicum. However the pH of the culture must be maintained within a relatively narrow pH range in order for the shift to the solventogenic phase to occur and for induction of clostridial forms and endospores. The

S. Long et al.: Initiation of solvent production and sporulation in C. acetobutylicum 261

actual pH range appears to differ between C. aceto- butylicum strains and the threshold of acid end-prod- ucts required to inhibit growth and trigger the shift to the solventogenic phase and differentiation may vary depending on the pH of the culture. In strain P262 cell viability decreases when very low pH values occur in poorly buffered media.

Although a threshold concentration of acid end-products and the correct pH levels were essen- tial, they were not sufficient on their own to ensure the development of the solventogenic phase and induce differentiation. These developments also required the presence of residual glucose and ammonium at the end of the acidogenic phase.

The factors responsible for the intitiation of sporulation in C. acetobutylicum P262 differ marked- ly from those in Bacillus subtilis where sporulation is initiated by nutrient starvation (Young and Mandel- stare 1979). In C. acetobutylicum glucose or ammo- nium limitation did not induce sporulation and there was an actual requirement for glucose and ammo- nium during endospore formation. At lower concen- trations of glucose or ammonium the yield of endospores was proportional to the concentration of glucose or ammonium respectively. The intracellular processes responsible for the induction of sporulation in C. acetobutylicum are not known but acid end-product inhibition of growth, pH and the initia- tion of solvent production and the clostridial stage appear to be essential for the initiation of endospore formation in C. acetobutylicum P262.

References

Andersch W, Bahl H, Gottschalk G (1982) Acetone-butanol production by Clostridium acetobutylicum in an ammo- nium-limited chemostat at low pH values. Biotechnol Lett 4 : 29- 32

Bahl H, Andersch W, Braun K, Gottschalk G (1982) Effect of pH and butyrate concentration on the production of acetone and butanol by Clostridium acetobutylicum grown in continuous culture. Eur J Appl Microbiol Biotechnol 14:17-20

Davis R, Stephenson M (1941) Studies on the acetone-butyl alcohol fermentation. I. Nutritional and other factors involved in the preparation of active suspensions of Clostridium acetobutylicum (Weizmann). Biochem J 35:1320-1331

Gottschal JC, Morris JG (1981a) The induction of acetone and butanol production in cultures of Clostridium acetobutylicum by elevated concentration of acetate and butyrate. FEMS Microbiol Lett 12:385-389

Gottschal JC, Morris JG (1981b) Non-production of acetone and butanol by Clostridium acetobutylicum during glucose- and ammonium-limitation in continuous culture. Biotechnol Lett 3 : 525-530

Hfiggstr6m L, Molin N (1980) Calcium alginate immobilized cells of Clostridium acetobutylicum for solvent production. Biotech- nol Lett 2:241-246

Herrero AA (1983) End-product inhibition in anaerobic fermen- tations. Trends Biotechnol 1 :49-53

Hoeniger JFM, Headley CL (1968) Cytology of spore germination in Clostridium pectinovorum. J Bacteriol 96:1835-1847

Jones DT, van der Westhuizen A, Long S, Allcock ER, Reid SJ, Woods DR (1982) Solvent production and morphological changes in Clostridium acetobutylicum. Appl Environ Micro- biol 43:1434-1439

Long S, Jones DT, Woods DR (1983) Sporulation of Clostridium acetobutylicum P262 in a defined medium. Appl Environ Microbiol 45:1389-1393

Martin JR, Petitdemange H, Bavongue J, Gay R (1983) Effect of acetic and butyric acids on solvent production by Clostridium acetobutylicum. Biotechnol Lett 5 :89-94

Monot F, Engasser JM (1983) Production of acetone and butanol by batch and continuous culture of Clostridium acetobutylicum under nitrogen limitaton. Biotechnol Lett 5:213-218

Monot F, Martin J-R, Petitdemange H, Gay R (1982) Acetone and butanol production by Clostridiurn acetobutylicum in a syn- thetic medium. Appl Environ Microbiol 44:1318-1324

Nakhmanovich BM, Shcheblykina NA (1960) Fermentation of calcium salts of acetic and butyric acids by Clostridium acetobutylicum. Mikrobiologiya 29:67-72

O'Brien RW, Morris JG (1971) Oxygen and the growth and metabolism of Clostridium acetobutylicum. J Gen Microbiol 68 : 307-318

Robson PM, Jones DT (1982) Production of acetone-butanol by industrial fermentation. In: Chaude D, Durand G (eds) Production D'Intermediares Industriels Par Culture Anaero- bie. Societe Francaise De Microbiologie, pp 169-213

Smith AG, Ellner PD (1957) Cytological observations on the sporulation process of Clostridium perfringens. J Bacteriol 7 3 : 1 - 7

Spivey MJ (1978) The acetone/butanol/ethanol fermentation. Proc Biochem 13 : 2 -5

Young M, Mandelstam J (1979) Early events during bacterial endospore formation. Adv Microb Physiol 20:102-162

Yu EKC, Saddler JN (1983) Enhanced acetone-butanol fermen- tation by Clostridium acetobutylicum grown on D-xylose in the presence of acetic or butyric acid. FEMS Microbiol Lett 18 : 103-107

Received April 9, 1984