7000191a 322.328

Journal of Industrial Microbiology & Biotechnology (2001) 27, 322–328D 2001 Nature Publishing Group 1367-5435/01 $17.00 Fermentation characterization and flux analysis of recombinantstrains of Clostridium acetobutylicum with an inactivated solRgene LM Harris1, L Blank1, RP Desai1, NE Welker2 and ET Papoutsakis1 1Department of Chemical Engineering, Northwestern University, Evanston, IL 60208, USA; 2Department of Biochemistry,Molecular Biology, and Cell Biology, Northwestern University, Evanston, IL 60208, USA The effect of solR inactivation on the metabolism of Clostridium acetobutylicum was examined using fermentationcharacterization and metabolic flux analysis. The solR - inactivated strain ( SolRH ) of this study had a higher rate ofglucose utilization and produced higher solvent concentrations ( by 25%, 14%, and 81%, respectively, for butanol,acetone, and ethanol ) compared to the wild type. Strain SolRH( pTAAD ), carrying a plasmid - encoded copy of thebifunctional alcohol / aldehyde dehydrogenase gene ( aad ) used in butanol production, produced even higherconcentrations of solvents ( by 21%, 45%, and 62%, respectively, for butanol, acetone, and ethanol ) than strain SolRH.
Clarithromycin used for strain SolRH maintenance during SolRH( pTAAD ) fermentations did not alter productformation; however, tetracycline used for pTAAD maintenance resulted in 90% lower solvent production. Journal ofIndustrial Microbiology & Biotechnology ( 2001 ) 27, 322 – 328.
Keywords: Clostridium acetobutylicum; metabolic engineering; flux analysis; solvents; solR; tetracycline; clarithromycin strain carries an additional plasmid pTAAD - encoded copy of theaad gene, which encodes a bifunctional butanol formation Clostridium acetobutylicum, a Gram - positive, spore - forming, enzyme [ 15 ]. SolRH was selected as the host strain in this study obligate anaerobe, has long been of industrial interest for the since it has a defined genetic modification; one non - replicative production of the solvents acetone and butanol from renewable plasmid is integrated into the chromosome of SolRH in contrast sources [ 8 ]. Due to its ability to catabolize a wide range of to strain SolRB, which has multiple non - replicative plasmids substrates into solvents, C. acetobutylicum was in widespread use integrated into its chromosome [ 16 ]. Additionally, the tetracy- internationally for fermentative production of acetone and butanol cline minimum inhibitory concentrations ( MICs ) in SolRH was between 1910 and 1960. However, unprecedented growth of the less than 5 g / ml compared to 10 g / ml in SolRB ( data not petroleum industry during the middle of the 20th century led to a shown ). Thus, use of the tetracycline resistance plasmid pTAAD drastic decline in industrial C. acetobutylicum fermentations since it ( as reported here ) was more practical in strain SolRH. Strain was more economical to produce solvents from petroleum - based SolRH( pTAAD ) was constructed in order to determine if an sources [ 8 ]. Metabolic engineering of this organism to construct increased aad gene dosage would further enhance the solvent - superior solvent - producing strains may be the key to a resurgence producing capabilities of strain SolRH. The use of the antibiotics of industrial - scale clostridial fermentations.
clarithromycin and tetracycline for selection was also examined While the majority of genes directly responsible for formation to examine their effects on product formation.
of acids and solvents in this strain have been cloned andstudied, less is known about mechanisms for controlling geneexpression in solventogenic clostridia. Recently, a putativerepressor of solvent formation genes, SolR, was identified and a molecular characterization of solR - modified strains was reported [ 16 ]. Inactivation of the solR gene ( located on the Bacterial strains and plasmids are listed in Table 1.
pSOL1 megaplasmid ) [ 6 ], using genomic integration of a non -replicative plasmid, resulted in the generation of strains whichproduce higher levels of solvents. Preliminary characterization of one of these strains ( renamed SolRB, formerly Mutant B ) Escherichia coli was grown aerobically at 378C in Luria – Bertani included a single batch fermentation [ 16 ]. The present study ( LB ) medium, and C. acetobutylicum was grown anaerobically at focuses on detailed fermentation characterization and metabolic 378C in Clostridium Growth Medium ( CGM [ 18,22 ] ). Colonies of flux analysis [ 2 ] of a second strain ( SolRH; formerly Mutant E. coli and C. acetobutylicum were obtained on agar - solidified LB H ) and a derivative of that strain [ SolRH( pTAAD ) ]. The latter or Reinforced Clostridial Medium ( RCM; Difco, Sparks, MD ),respectively. For recombinant strains, liquid media were appropri-ately supplemented with erythromycin ( Em; 100 g / ml ), tetra-cycline ( Tc; 10 g / ml ), and chloramphenicol ( Cm; 35 g / ml ); Correspondence: ET Papoutsakis, Department of Chemical Engineering, North- 40 g / ml of erythromycin and 10 g / ml of tetracycline were western University, Evanston, IL 60208, USAReceived 12 September 2000; accepted 21 July 2001 used in solid media as needed. For bioreactor fermentations, Analysis of C. acetobutylicum solR inactivation strainsLM Harris et al ( single copy of plasmid integrated into genome ),MIC 5 g / ml ( multiple copies of plasmid integrated into genome ),MIC 10 g / ml aAbbreviations: solR putative repressor of sol operon genes; MLSr, macrolide lincosamide streptogramin B - resistant; recA, homologous recombinationabolished; lacZ,  - galactosidase; mcrBC, methlycytosine - specific restriction system; Cmr, chloramphenicol - resistant; È3TI, È3T methylase; Tcr,tetracycline - resistant; Apr, ampicillin - resistant; AAD, aad, alcohol / aldehyde dehydrogenase.
bAmerican Type Culture Collection, Manassas, VA.
cNew England Biolabs, Beverly, MA.
erythromycin was substituted by its pH - stable derivative clari- which point it was controlled through addition of 6 M NH4OH.
thromycin ( 100 g / ml ). E. coli strains were stored long - term at À 858C in LB medium with 10% glycerol. For long - term storage of acetone, ethanol, butanol, and acetoin, 3%; acetate and butyrate, C. acetobutylicum, strains were maintained as spores on RCM agar at a pH of 6.8, or frozen at À 858C in CGM with 15% glycerol.
Metabolic flux analysis involves the calculation of specific ( per The isolation of plasmids from E. coli via the alkaline lysis unit cell mass ) in vivo intracellular reaction rates for various method and further manipulations of E. coli plasmid DNA were enzymatic reactions ( which will be referred to as ‘‘fluxes’’ ). These performed using standard protocols [ 9 ]. A modified alkaline fluxes [ in units of mmol ( g cells ) À 1 h À 1 ] were calculated from lysis method was used for isolation of plasmid DNA from substrate utilization and product formation data using a system of C. acetobutylicum [ 7 ]. A previously published method [ 14 ] was linear equations developed from the metabolic reaction stoichiom- etry, a widely used and validated technique [ 1 – 3,7,17 ]. A mass methylated plasmid DNA [ 13 ] using a Bio - Rad Gene Pulser extinction coefficient of 51 ( g cells ) À 1 cm À 1 was used to convert optical density measurements ( A 600 ) into cell dry weight concen-trations ( g cells l À 1 ). The flux analysis was performed using software developed specifically for analysis of C. acetobutylicumfermentation data [ 2,17 ], and the pathways fluxes considered are Plasmid pTAAD was generated by cloning a 3.06 - kb DNA depicted in Figure 1. For these analyses, the fermentation times fragment containing aad gene with its two promoters into the were re - scaled [ 1,3,7 ]: the normalized scale, T tetracycline - resistant E. coli – C. acetobutylicum shuttle vector pTLH1 [ 7 ]. These tetracycline - resistant vectors were suitable for h at A 600 = 1. On this time scale, the transition from exponential phase to stationary phase typically occurs at T use in the MLSr ( macrolide lincosamide streptogramin B - resistant ) Therefore, the stages of the fermentations are classified as ‘‘Early’’ for T N < 10 h, and ‘‘Late’’ for T N > 10 h. The calculation of bothkinetic and integral fluxes and the errors of such calculations have been discussed previously [ 1,3,7 ].
concentration analysis were performed as previously reported[ 3,7 ]. Batch fermentations were performed in either a BiostatM ( Braun Biotech, Allentown, PA ) or a BioFloII bioreactor ( NewBrunswick Scientific, Edison, NJ ) with working volumes of 1.5 Strain SolRH [ 16 ] and newly developed strain SolRH( pTAAD ) and 4.01, respectively. After inoculating the growth medium were characterized using fermentation experiments and metabolic ( CGM pH 6.2 ), the culture pH was allowed to fall to 5.0, at flux analysis to determine the impact of two genetic alterations, the Analysis of C. acetobutylicum solR inactivation strains Figure 1 C. acetobutylicum primary metabolic pathways and corresponding in vivo fluxes. The conversion between major carbon containing speciesis depicted without cofactors. Selected enzymes are shown in bold and abbreviated: PTA, phosphotransacetylase;AK, acetate kinase;CoAT, CoAtransferase;AADC, acetoacetate decarboxylase;PTB, phosphotransbutyrylase;BK, butyrate kinase, AAD, alcohol / aldehyde dehydrogenase( DH );BDHA and BDHB, butanol DHG isozymes A and B. Reactions involved with the intracellular fluxes ( e.g., rBIO and rHYD ) examinedhere are indicated by dashed boxes. Further details on these metabolic fluxes can be found in Refs. [ 2,3,7,17 ].
solR inactivation alone and in combination with increased aad gene formation pathways ), rACUP, rBYUP ( acid re - utilization ), and rBUOH ( butanol formation ). Strain SolRH exhibited a consistentelevation in rGLY1 from T N = 0 to 5 h ( Figure 2A ). Thisincreased flux of glucose utilization was also reflected by Fermentation and flux analysis of solRH versus wild elevated acid formation fluxes ( rPTAAK and rPTBBK ) from T N = 0 to 5 h compared to the WT strain ( Figure 2B and C ). In Fermentation characterizations showed that inactivation of solR in addition, strain SolRH exhibited an earlier reversal ( indicated by strain solRH resulted in altered product concentration profiles negative flux values ) of flux through the butyrate formation compared to the WT organism, C. acetobutylicum ATCC 824 pathway ( rPTBBK ). The most dramatic effect of the solR ( Table 2 ). Although peak butyrate levels did not differ significantly inactivation was on solvent formation fluxes. Acetone formation between strains SolRH and WT, final SolRH butyrate concen- flux is the sum of the acetate re - utilization flux ( rACUP ) and the trations were 68% lower than WT values ( 13 vs. 41 mM ). Both butyrate re - utilization flux ( rBYUP ) and showed significantly maximum and final acetate levels were slightly higher in SolRH higher peak levels compared to the WT strain. The acetate re - compared to WT. solR inactivation had a profound effect on solvent utilization flux, rACUP, in strain SolRH reached a peak value formation, resulting in final butanol, acetone, and ethanol twice that of the WT strain ( Figure 2D ). The butyrate re - concentrations that were 25%, 14%, and 81% higher, respectively, utilization flux, rBYUP, in strain SolRH reached a peak value up to 60% higher than in the WT ( Figure 2E ). In addition, the peak Metabolic flux analysis of the fermentation data was used to in rBYUP was 5 h earlier in strain SolRH than in the WT strain.
further characterize strain SolRH. The key pathway fluxes ( i.e., However, the elevated acetone formation fluxes in strain SolRH specific intracellular reaction rates ) discussed ( Figure 1 ) are were sustained for a shorter period of time than in the WT strain.
rGLY1 ( glucose utilization ), rPTAAK and rPTBBK ( acid The butanol formation flux reached a peak up to 30% higher ( but Analysis of C. acetobutylicum solR inactivation strainsLM Harris et al Table 2 Product formation in fermentation experiments at pH 5.0 Data are shown as mean ± SEM values from two experiments. Glucose was fed in strain SolRH( pTAAD ) fermentations to prevent glucose depletion andpremature termination of the fermentation.
was sustained for a shorter period of time ) in strain SolRH than pathways fluxes ( rGLY1, rGLY2, rTHL, and rBYCA ) were twice as high, and acid formation fluxes, rPTAAK and rPTBBK, were In addition to the time profiles, an integral flux analysis ( see 120% and 840% higher, respectively, in strain SolRH than in the Materials and Methods; data not shown ) further reinforced the WT strain. Solvent formation fluxes, rACUP, rBUOH, and rETOH, observations discussed above, but also showed significant differ- were ca. 130%, 150%, and 400%, respectively, higher in strain ences that are not apparent from the kinetic analysis: the central SolRH. While acetone formation via butyrate re - utilization Figure 2 Time course profiles of metabolic fluxes in cultures of strain SolRH ( solid symbols ) and of the WT ( open symbols ). Different symbolsrepresent data from the two replicate experiments of Table 2.
Analysis of C. acetobutylicum solR inactivation strains ( rBYUP ) was elevated by ca. 60%, this pathway is responsible for When tetracycline and clarithromycin were used in combina- only a small amount of the carbon reutilization in strain SolRH. The tion, acetate and butyrate levels were both significantly elevated.
calculated metabolic pathway fluxes were used to estimate changes Final butanol, acetone, and ethanol concentrations were all reduced in the split ratio at the butyryl – CoA branchpoint. During the by ca. 80%, and cell densities significantly decreased ( Table 3 ).
stationary phase of the WT strain, the central pathway ( rBYCA ) Patterns of product formation when clarithromycin alone was provided 82% of the carbon used in butanol formation, while used during the SolR( pTAAD ) fermentation were similar to the rBYUP and rPTBBK provided 10% and 8%, respectively. In control fermentation with no selective pressure ( Table 3 ).
contrast, in strain SolRH, rBYCA provided 65% of the carbon, These results indicate that the use of clarithromycin to select while rPTBBK provided 28% of the carbon used for butanol for SolRH( pTAAD ) does not alter product formation patterns, formation. In fact, this reversal of the butyrate formation pathway is growth rates, or cell densities in this organism. In contrast, the largely responsible for the 150% increase in rBUOH, while glucose data show that the use of tetracycline in SolRH( pTAAD ) fermentations, alone and in combination with clarithromycin,inhibits solvent production and enhances acidogenesis. Althoughtetracycline is useful for selection of plasmid - carrying strains, use Effects of tetracycline and clarithromycin on product of this antibiotic drastically reduces solvent production and is therefore not appropriate for use in fermentations for solvent Initially, two antibiotics were used for maintenance of C. aceto- production. Loss of this family of plasmids during batch butylicum strains SolRH( pTLH1 ): the control strain carrying a fermentations of recombinant C. acetobutylicum strains in the plasmid without the cloned aad gene and SolRH( pTAAD ).
absence of antibiotics as a selection pressure is known to be Clarithromycin ( 100 g / ml ), a stable derivative of erythromycin, was used to select for the host strain SolRH, and tetracycline use Another example of inhibition of product formation in the ( 10 g / ml ) assured presence of the replicative plasmids pTLH1 presence of tetracycline involves the use of strain 34, a derivative and PTAAD. Clarithromycin resistance was conferred by the of C. acetobutylicum ATCC 824 with a single chromosomal presence of the non - replicative plasmid integrated into the genomic insertion of the tetracycline resistance conjugative transposon solR [ 16 ], while tetracycline resistance was conferred by the tetM Tn916. When tetracycline was used to select for strain 34, no gene on the multicopy vectors pTLH1 and pTAAD. During butyraldehyde dehydrogenase activity was detected and no fermentations, the use of two antibiotics coincided with a drastic butanol was produced ( C. Cass, unpublished results ). Tn916 decrease in growth and solvent production. Thus, experiments were has been used extensively for generating clostridial strain designed to determine the effects of tetracycline and clarithromycin mutants, including solvent formation mutants ( e.g., Ref. [ 12 ] ).
use separately and together ( Table 3 ).
Our data suggest that the use of tetracycline - resistant Tn916 in When tetracycline was used during SolRH( pTAAD ) fermen- generating solvent formation mutants may be inappropriate. Use tations, cell densities were decreased, peak butyrate concen- of tetracycline and its derivatives in various microbial systems trations were increased by over 35%, while final butyrate values inhibits several types of enzymes, including some dehydrogenase exceeded the control values by over 1000% ( Table 3 ). Peak and systems [ 4 ]. While the mechanism by which tetracycline affects final acetate concentrations were not significantly affected.
product formation is not proven, it is possible that tetracycline Tetracycline inhibition of solvent production was severe: butanol, interferes with the dehydrogenase activity required for solvent acetone, and ethanol concentrations were decreased by 92%, production in recombinant strains of C. acetobutylicum such as 94%, and 90%, respectively, compared to the no - antibiotic Fermentation and flux analysis of strains solRH Table 3 Tetracycline ( Tc ) and clarithromycin ( Clt ) effects on strain ( pTAAD ) versus solRH( pTLH1 ): effect of plasmid - encoded aldehyde alcohol dehydrogenaseStrain SolRH( pTAAD ) was characterized in order to determine the effect of pTAAD - encoded aldehyde alcohol dehydrogenase [ 5 ] onproduct formation in strain SolRH ( Table 2 ). No antibiotics were used during SolRH( pTLH1 ) fermentations. However, selectivepressure was applied for maintenance of strains SolRH( pTAAD ) and SolRH( pTLH1 ) until the time that reactors were inoculated at the start of each fermentation. Comparative plating studies ( data not shown ) on selective and non - selective media indicated that the plasmid - carrying strains were stable throughout the course of fermentations without selective pressure.
Fermentations of strain SolRH( pTAAD ) were performed at pH 5.0 based on other studies showing that pH = 5.0 resulted in higher solvent production than either pH 4.7 or 5.5 ( data not shown ).
These SolRH( pTAAD ) cultures were supplemented with glucose during mid to late exponential growth to avoid prematuretermination of solvent production due to glucose exhaustion.
Fermentation experiments were carried out at pH 5.0. There was no glucose Concentrated glucose ( 3.8 M ) was added when the residual glucose fed into these fermentations. For the duplicate experiment, values arereported as mean ± SEM.
concentration dropped below 150 mM in order to keep glucose Analysis of C. acetobutylicum solR inactivation strainsLM Harris et al concentration above 100 mM until solvent production began to [ SolRH( pTLH1 ) ] fermentations. Final butanol, acetone, and plateau. Equivalent glucose supplementation of either the WT or ethanol concentrations were elevated by 22%, 42%, and 161%, SolRH( pTLH1 ) strains did not affect solvent production ( data not respectively. While the SolRH( pTAAD ) fermentation doubling shown ). The results of the SolRH( pTAAD ) fermentations were time was 16% lower than the doubling time of the control, maximal compared to the control strain SolRH( pTLH1 ), which was optical densities did not vary significantly.
developed to account for host – plasmid effects [ 21 ] using a Kinetic metabolic flux analysis ( Figure 3 ) indicated noticeable plasmid without a cloned clostridial gene. Inactivation of solR, differences between strains SolRH( pTAAD ) and SolRH( pTLH1 ).
combined with increased aad gene dosage, resulted in a strain Glucose utilization in SolRH( pTAAD ) was transiently elevated at [ SolRH( pTAAD ) ] with improved solvent - producing ability. At the shift to the stationary phase ( Figure 3A ). A similar trend was final concentrations of 248 mM ( 17.6 g / l ) butanol, 141 mM observed in the other central pathway fluxes ( data not shown ). A ( 8.2 g / l ) acetone, and 47 mM ( 2.2 g / l ) ethanol, total solvent stationary phase elevation was also observed in rPTAAK fluxes, production ( 28 g / l ) by SolRH( pTAAD ) far exceeded the 19 g / l but not in rPTBBK fluxes ( Figure 3B and C ). Instead, the butyrate solvents produced by the WT. It should be noted that aad formation pathway supported increased butyrate re - utilization. As overexpression in the WT strain ( ATCC 824 ) did not result in with strain SolRH, solvent formation fluxes showed the most increased solvent formation [ 14 ]. This suggests that the inactiva- dramatic differences. The peak acetone formation fluxes were tion of solR is necessary for the positive effect of aad over- evaluated in strain SolRH( pTAAD ) by up to 200% and 35% for Maximum and final acetate concentrations produced by SolRH( pTLH1 ) ( Figure 3D and E ). The peak butanol formation SolRH( pTAAD ) were 30% higher than control values. Peak flux, rBUOH, was elevated by up to 200% during the stationary butyrate levels were 23% higher in SolRH( pTAAD ) than in the phase ( Figure 3F ), and the ethanol formation flux was control; however, final butyrate concentrations were 40% lower.
significantly elevated ( data not shown ). It should also be noted This implies better re - utilization of butyrate by strain Sol- that while strain SolRH( pTAAD ) exhibited higher peak fluxes, RH( pTAAD ). Solvent production was significantly increased in strain SolRH( pTLH1 ) sustained the fluxes for a longer period of Figure 3 Time course profiles of metabolic fluxes in strains SolRH( pTAAD ) ( solid symbols ) and SolRH( pTLH1 ) ( open symbols ). Differentsymbols represent data from the two replicate experiments of Table 2. For strain SolRH( pTAAD ), data from two additional replicates ( not shownon Table 2 ) are also plotted.
Analysis of C. acetobutylicum solR inactivation strains As before, integral flux analysis ( data not shown ) was used to 6 Green EM, ZL Boynton, LM Harris, FB Rudolph, ET Papoutsakis and detect differences between SolRH( pTAAD ) and SolRH( pTLH1 ) GN Bennett. 1996. Genetic manipulation of acid formation pathwaysby gene inactivation in Clostridium acetobutylicum ATCC 824.
that are less noticeable from the kinetic flux analysis. After the shift to stationary phase, The differences between strains Sol- 7 Harris LM, RP Desai, NE Welker and ET Papoutsakis. 2000.
RH( pTAAD ) and SolRH( pTLH1 ) were most pronounced: glucose Characterization of recombinant strains of the Clostridium acetobuty- utilization ( rGLY1 ) and the cental pathways ( rGLY2, rTHL, and licum butyrate kinase inactivation mutant: need for new physiologicalmodels for solventogenesis and butanol inhibition? Biotechnol Bioeng rBYCA ) were ca. 100 – 130% higher, and the acid formation fluxes, rPTAAK and rPTBBK, were elevated by 180% and 225%, 8 Jones DT and DR Woods. 1986. Acetone – butanol fermentation respectively, in strain SolRH( pTAAD ). The acetone formation revisited. Microbiol Rev 50: 484 – 524.
fluxes, rACUP and rBYUP, were elevated by 160% and 70%, 9 Lee SY and S Rasheed. 1990. A simple procedure for maximum yield respectively, although rBYUP played a lesser role than rPTBBK in of high quality plasmid DNA. BioTechniques 9: 676 – 679.
10 Lee SY, GN Bennett and ET Papoutsakis. 1992. Construction of butyrate re - utilization. Finally, the alcohol formation fluxes, Escherichia coli – Clostridium acetobutylicum shuttle vectors and rBUOH and rETOH, were elevated by 125% and 370%, transformation of Clostridium acetobutylicum stains. Biotechnol Lett The results of the fermentation characterizations and the 11 Lee SY, LD Mermelstein and ET Papoutsakis. 1993. Determination of plasmid copy number and stability in Clostridium acetobutylicum metabolic flux analyses clearly illustrate the potential for rational ATCC 824. FEMS Microbiol Lett 108: 319 – 323.
design of superior solvent - producing strains through manipulation 12 Mattsson DM and P Rogers. 1994. Analysis of Tn916 - induced mutants of key regulatory and product formation genes. Although the of Clostridium acetobutylicum altered in solventogenesis and sporula- putative role of solR as a transcriptional repressor is now being tion. J Ind Microbiol 13: 258 – 268.
challenged [ 20 ], our results show that solR plays a significant role 13 Mermelstein LD and ET Papoutsakis. 1993. In vivo methylation in Escherichia coli by the Bacillus subtilis phage È3T I methyltransferase to protect plasmids from restriction upon transformation of Clostridiumacetobutylicum ATCC 824. Appl Environ Microbiol 59: 1077 – 1081.
14 Mermelstein LD, NE Welker, GN Bennett and ET Papoutsakis. 1992.
Expression of cloned homologous fermentative genes in Clostridiumacetobutylicum ATCC 824. Bio / Technology 10: 190 – 195.
Supported by National Science Foundation grant BES - 9905669, 15 Nair RV, GN Bennett and ET Papoutsakis. 1994. Molecular and National Institutes of Health Predoctoral Biotechnology characterization of an aldehyde / alcohol dehydrogenase gene fromClostridium acetobutylicum ATCC 824. J Bacteriol 176: 871 – 885.
Training grant GM - 08449. We thank Abbott Laboratories ( Abbott 16 Nair RV, EM Green, GN Bennett and ET Papoutsakis. 1999. Regulation Park, IL ) for donation of clarithromycin.
of the sol locus genes for butanol and acetone formation in Clostridiumacetobutylicum ATCC 824 by a putative transcriptional repressor.
J Bacteriol 181: 319 – 330.
17 Papoutsakis ET. 1984. Equations and calculations for fermentations of butyric acid bacteria. Biotechnol Bioeng 26: 174 – 187.
1 Desai RP and ET Papoutsakis. 1999. Antisense RNA strategies for the 18 Roos JW, JK McLaughlin and ET Papoutsakis. 1985. The effect of pH metabolic engineering of Clostridium acetobutylicum. Appl Environ on nitrogen supply, cell lysis, and solvent production in fermentations of Clostridium acetobutylicum. Biotechnol Bioeng 27: 681 – 694.
2 Desai RP, LK Nielsen and ET Papoutsakis. 1999. Stoichiometric 19 Su YA, P He and DB Clewell. 1992. Characterization of the tet( M ) modeling of Clostridium acetobutylicum fermentations with non - linear determinant of Tn916: evidence for regulation by transcription constraints. J Biotechnol 71: 191 – 205.
attenuation. Antimicrob Agents Chemother 36: 796 – 778.
3 Desai RP, LM Harris, NE Welker and ET Papoutsakis. 1999. Metabolic 20 Thormann K and P Duerre. 2001. Orf5 / solR: a transcriptional repressor flux analysis elucidates the importance of the acid formation pathway in of the sol operon of Clostridium acetobutylicum? J Ind Microbiol regulating solvent production by Clostridium acetobutylicum. Metab 21 Walter KA, LD Mermelstein and ET Papoutsakis. 1994. Host – plasmid 4 Gottlieb D and PD Shaw. 1967. Antibiotics: I. Mechanisms of Action.
interactions in recombinant strains of Clostridium acetobutylicum Springer - Verlag, New York, p. 785.
ATCC 824. FEMS Microbiol Lett 335 – 342.
5 Green EM and GN Bennett. 1996. Inactivation of an aldehyde / alcohol 22 Wiesenborn DP, ET Papoutsakis and FB Rudolph. 1988. Thiolase from dehydrogenase gene from Clostridium acetobutylicum ATCC 824. Appl Clostridium acetobutylicum ATCC 824 and its role in the synthesis of acids and solvents. Appl Environ Microbiol 54: 2717 – 2722.

Source: http://www.papoutsakis.org/sites/www.papoutsakis.org/uploaded/Analysis%20of%20Recombinant%20strains%20of%20Clostridium%20acetobutylicum%20with%20an%20inactivated%20solR%20gene.pdf