Yangetal.AMBExpress2013,3:3 http://www.amb-express.com/content/3/1/3 ORIGINAL ARTICLE Open Access A comparison of three pH control methods for revealing effects of undissociated butyric acid on specific butanol production rate in batch Clostridium acetobutylicum fermentation of Xuepeng Yang1, Maobing Tu1,2*, Rui Xie1, Sushil Adhikari2 and Zhaohui Tong3 Abstract pH control has been essential for butanol production with Clostridium acetobutylicum.However, it is not very clear atwhat pH level theacid crash will occur, atwhat pH level butanol production will be dominant, and atwhat pH levelbutyricacid production will be prevailing.Furthermore, contradictory results have been reported about required acidic conditions for initiation of solventogenesis. Inthis study, with the aim of further understanding the roleof undissociatedbutyric acid in butanol production,weinvestigated thecorrelation between undissociated butyric acid concentration and specific butanol production rate inbatchfermentation of Clostridium acetobutylicum bycomparing threepH control approaches:NaOH neutralization (at 12, 24 or 36 h), CaCO supplementation (2, 5, 3 or 8g/l) and NaOAc buffering (pH 4.6, 5.0or 5.6). By neutralizing thefermentationpH to ~5.0 at different time, we observed that neutralization should take place atthe beginning of exponential phase (12 h), and otherwise resulting in lower concentrations ofundissociated butyric acid, cell biomass and final butanol. CaCO 3 supplementation extended cell growthto 36 h and resulted in higher butyrate yield under 8g/L of CaCO .In the 3 NaOAc buffering, the highest specific butanol rate (0.58 h−1) was associated with the highest undissociatedbutyric acid (1.92 g/L). The linear correlation of theundissociated butyric acid with thespecific butanol production rates suggested theundissociated butyric acid could be themajor driving force for butanol production. Keywords: Butanol, Butyric acid, Fermentation, Undissociated butyricacid, Clostridium acetobutylicum Introduction fermentation or ABE fermentation” (Jones and Woods Butanol is one of the promising advanced biofuels and 1986). The typical products ratio from this process with an important intermediate in chemical synthesis. It is ClostridiumacetobutylicumisA:B:E=3:6:1.C.acetobu- being pursued by industry and the U.S. government (the tylicum is capable of fermenting hexose and pentose DOE, USDA and NSF) for the next generation of alter- sugars to butanol, but solvent yields and fermentation native fuels (Hess 2006; Li et al. 2011a; Li et al. 2011b; rates have varied depending on strains and fermentation Milne et al. 2011; Qureshi and Ezeji 2008). Butanol, as conditions(Qureshietal.2008).Thefermentationprocess one of the alternative biofuels, has several advantages by C. acetobutylicum is divided into two distinct phases: over ethanol (the main transportation biofuel candidate) rapid cell growth and production of butyric acids take such as low vapor pressure and tolerance to water con- place in a first acidogenic phase accompanied by pH de- tamination (Hess 2006). Butanol production from sugars crease, which is then replaced by a second solventogenic by fermentation is known as “acetone-butanol-ethanol phaseinwhichbutanolandothersolvents(mainlyacetone andethanol)areproduced,leadingtopHincrease(Bryant *Correspondence:[email protected] and Blaschek 1988; Holt et al. 1984; Huang et al. 1986; 1ForestProductsLabandCenterforBioenergyandBioproducts,Auburn HusemannandPapoutsakis1988;Monotetal.1984). University,520DevallDrive,Auburn,AL36849,USA 2DepartmentofBiosystemsEngineering,AuburnUniversity,Auburn,AL Previously, extensive research has been focused on the 36849,USA mechanismsof“acidcrash”andinitiationofsolventogenesis Fulllistofauthorinformationisavailableattheendofthearticle ©2013Yangetal.;licenseeSpringer.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproduction inanymedium,providedtheoriginalworkisproperlycited. Yangetal.AMBExpress2013,3:3 Page2of8 http://www.amb-express.com/content/3/1/3 (Martin et al. 1983; Wang et al. 2011; Zverlov et al. 2006). Clostridia Medium (RCM) with the addition of agar in a Acid crash occurs in pH uncontrolled fermentation and RT Anaero-Indicator. Isolated colony was cultured an- results in the cessation of glucose uptake, acids production aerobically in 50 mL of 38 g/L of RCM at 35°C for 24 h andbutanolproduction(Maddoxetal.2000).Theoverpro- and left aerobically in ambient temperature overnight ductionofundissociatedacids(>60mM)withlowpHcould for the production of spores. The spores were washed be the main cause for acid crash (Maddox et al. 2000). Re- and re-suspended in sterile water as the pre-inoculum cently, formic acid has also been suggested to trigger the (OD =2). Before inoculation, the pre-inoculum was 600 acid crash of butanol production by C. acetobutylicum treated with heat shock at 80°C for 10 min. The heat (Wang et al. 2011). Initiation of solventogenesis has been shocked spores were cultured anaerobically into 50 mL correlated well with the concentration of undissociated of RCM liquid medium at 35°C for 12 h as seed inocu- butyricacid(Huangetal.1986;HusemannandPapoutsakis lum. 5 mL of the seed inoculum was added into the 1988;Monotetal.1984).Ithasbeensuggestedthatsolvent 50 mL fermentation broth. All the culture experiments productioninitiatedattheminimumof1.5g/Lofundissoci- were conducted in 125 mL serum bottles. All the media ated butyric acid (Monot et al. 1984). A linear correlation were autoclaved at 121°C for 20 min before the inocula- has been found between butanol and undissociated butyric tion.CelldensitywasmeasuredbyaUV–visspectrometer acid concentration at the onset of solvent formation at 600 nm. The dry cell weight (DCW) was calculated (HusemannandPapoutsakis1988).Theadditionofacetate, from the optical density based on the following equation butyrate and propionate could improve the final solvent DCW(g/L)=0.3A . 600 yields, but not much on the initiation of solventogenesis (Fond et al. 1985; Husemann and Papoutsakis 1990; Yu FermentationandpHcontrolmethods and Saddler 1983). However, the solvent formation could The fermentation was carried out in 125 mL serum bot- also be initiated at neutral pH (6.8–7.0) with the supple- tles with 50 mL of P2 medium containing (g/L): glucose, mentation of high concentration of acetate plus butyrate 50; yeast extract, 1.0; ammonium acetate, 2.2; KH PO , (GeorgeandChen1983;Holtetal.1984). 2 4 0.5; K HPO , 0.5; MgSO ·7H O, 0.2; MnSO ·7H O, It is believed that the accumulation of a threshold 2 4 4 2 4 2 0.01; FeSO ·7H O, 0.01; NaCl, 0.01. All the media were amount of butyric acid initiates the solvent production. 4 2 bubbled through nitrogen for 10 min to remove oxygen Particularly, the undissociated butyric acid excreted in andautoclavedat121°Cfor20minbeforefermentationin- the fermentation media re-enters the cells and serves oculation. During the fermentation, samples were taken at as the precursor for butanol production (Bryant and regular intervals for analysis. Three different pH control- Blaschek 1988; Husemann and Papoutsakis 1988). lingmethodswereexaminedfortheireffectsonfermenta- However, the excess of acids can be produced in the tion by C. acetobutylicum. In the NaOH neutralization, batch fermentation of C. acetobutylicum without pH 5 mL of 0.5 M NaOH was added into the fermentation control, and subsequently causes the cessation of glu- broths at 12 h, 24 h, or 36 h respectively during the fer- cose utilization (Maddox et al. 2000). Therefore, pH mentationto adjust thepHto~5.0. IntheCaCO supple- control has been essential for butanol production with 3 mentation, 2 g/L, 5 g/L or 8 g/L of CaCO powder was C. acetobutylicum (Bryant and Blaschek 1988; Huang 3 addedintothemediatocontrolthepHatthebeginningof et al. 1986; Monotetal.1984;Roosetal.1985).However, fermentation. In the sodium acetate buffering, the initial it is not very clear at what pH level the acid crash will fermentation pH was controlled at 4.6, 5.0, or 5.6 by vary- occur,atwhatpHlevelbutanolproductionwillbedomin- ing the ratio of HOAc/NaOAc. All the treatments were ant, and at what pH level butyric acid production will be conducted in duplicate. Fermentation broth without pH prevailing. Also, contradictory results have been reported controlwasusedasacontrol. thatacidicconditionsarenotrequiredforinitiationofsol- ventogenesis (George and Chen 1983). In this study, with theaimofimprovingbutanolproductionyield,weinvesti- Analyticalmethods gatedtheeffectsofdifferentpHcontrolmethodsonbuta- Glucose concentration was determined by HPLC or nol and butyric acid production in batch fermentation by using the DNS method. Butanol and butyric acid were C. acetobutylicum. In addition, we explored the potential analyzed by a Varian 3800 Gas Chromatography mechanisms for the persistent butanol production under equipped with aVarian CP8400 autosampler, splitless in- effectivepHcontrol. jector system, and flame ionization detector. The separ- ation was conducted using a Stabilwax-DA column Material and methods fitted with a 5 m deactivated guard column. The separ- Microbialstrainandmedium ation conditions were used by a previously described Clostridium acetobutylicum (ATCC 824) was used in method (Robinson et al. 2002). Butanol and butyric acid this study. The strain was maintained on Reinforced were quantified bychromatographicgradestandards.pH Yangetal.AMBExpress2013,3:3 Page3of8 http://www.amb-express.com/content/3/1/3 W was determined using pH test strips (BDH , pH range The pH profile showed the fermentation pH was main- 3.6–6.1,pH graduation0.3/0.5). tained between 4.7 and 5.0 and the cell biomass reached 0.68g/Lat24h,andincreasedby70%ascomparedtothe Results control. It appeared that acid flush was overcome under EffectofNaOHneutralizationoncellgrowth,butanoland thisscenario.However,whenweneutralizedthefermenta- butyricacidproductionbyC.acetobutylicum tion pH at 24 h and 36 h respectively, the glucose con- To determine at what time and pH level the acid crash sumption was still ceased at 24 h and the butanol yields will occur, we added 5 mL of 0.5 MNaOH at 12 h, 24 h, were very low at 72 h, both cases were similar to the fer- and 36 h individually during the fermentation to control mentationcontrol.Itindicatedtheacidflushprobablyoc- the pH between 4.7 and 5.3. These three time points curred early in the exponential phase (fermentation corresponded to the beginning of the exponential phase, 12–24 h) in the pH range of 3.6–4.1, which likely is the end of the exponential phase, and the stationary the reason why the later neutralization could not help phase of cell growth. The results of butanol and butyric improve glucose consumption and butanol production. acid yields were given in Table 1 and further details of As for butyric acid production, with the NaOH pH profiles, cell growth, glucose consumption and buta- neutralization at 12 h, the concentration of butyrate nol production were shown in Figure 1. Without pH increased significantly from 1.4 g/L (without pH control) control (Figure 1A), the fermentation pH dropped to 4.1 to4.3g/L at24h(Figures1AandB),andthendecreased at 12 h, 3.6 at 24 h,and remained at 3.6 until 36 h.After significantlyforbutanolproduction.Itappearedthatmore that, the fermentation pH started to increase and butyrate was converted into butanol when the fermenta- reached 4.4 at 72 h. The cells grew slowly in the first tion pH is controlled around 4.7. Our results agreed well 12 h of lag phase, and then increased dramatically from withthepreviousreportthatathresholdamountofundis- 12 to 24 h (the exponential phase). When the pH sociated butyric acid was required to start the solvento- dropped to approximately 3.6, the cell growth ended at genesis (Husemann and Papoutsakis 1988; Monot et al. 24 h and the cell biomass reached 0.40 g/L. Similar 1984). Failing to control the pH at 12 h was presumed to results have been reported previously when pH uncon- cause deactivation of key enzymes for acid and butanol trolled batch fermentation of C. acetobutylicum was per- productionandevenkillthecells(Welchetal.1989).Asa formed in a fermentor (Husemann and Papoutsakis result, neutralizing pH at 24 and 36 h after acid flush 1990). The butanol production started from 12 h and couldnotleadtoanyimprovementofbutanolproduction. slowly increased to 5.7 g/L at 72 h. More than 50% of This assumption was in agreement with the glucose glucose was not consumed due to the potential acid utilization, which was stopped after 24 h in these two crash. The acid crash seemed to take place before 24 h casesaswellasinthefermentationcontrol(Figure1). and the glucose consumption was ceased afterwards. This phenomenon can be described more precisely as EffectofCaCO supplementationoncellgrowth,butanol 3 “acid flush” because the excessive acids or low pH cause andbutyricacidproductionbyC.acetobutylicum a weak fermentation in the solventogenic phase, rather CaCO has been used to neutralize organic acids during 3 “crash” solvent production and fermentation completely fermentation and to maintain pH at a certain range (Figure 1A). (Huangetal.2005;Vandaketal.1997).Toevaluateitsef- WhenweneutralizedthepHto5.3withNaOHat12h, fect on fermentation by C. acetobutylicum, we examined the glucose was completely consumed and the final buta- the effects of CaCO supplement (2, 5, 8 g/L) on butanol 3 nol yield increased dramatically to 11.9 g/L (Figure 1B). and butyric acid production by C. acetobutylicum. The Table1EffectofdifferentpHregulationmethodsoncellgrowthandbutanolproductionbyC.acetobutylicum pHcontrol NaOHneutralization CaCO supplementation NaOAcbuffering(pH) 3 12h 24h 36h 2g/L 5g/L 8g/L 4.6 5.0 5.6 pHat24ha 4.7±0.0 3.6±0.0 3.6±0.0 4.4±0.0 4.7±0.0 5.3±0.0 4.1±0.0 4.4±0.0 5.0±0.0 Cb 11.9±0.7 3.8±0.0 4.9±0.1 8.8±0.4 10.8±0.1 7.9±0.4 6.5±0.3 12.3±0.9 8.3±0.4 butanol Cc 4.3±0.3 1.8±0.2 1.7±0.1 4.0±0.3 5.4±0.5 6.2±0.2 2±0.3 2.7±0.5 4.0±0.2 butyricacid Biomass 0.69±0.03 0.45±0.06 0.39±0.02 0.45±0.05 0.41±0.02 0.53±0.01 0.51±0.03 0.63±0.02 0.70±0.04 Cd 2.18±0.14 1.0±0.08 1.6±0.09 2.68±0.18 3.01±0.26 1.50±0.05 1.70±0.28 1.92±0.32 1.32±0.07 UBA q 0.41±0.02 0.22±0.09 0.42±0.02 0.49±0.06 0.52±0.01 0.25±0.02 0.44±0.01 0.58±0.01 0.36±0.02 p apHat24h:pHvaluesat24hoffermentation.bC (g/L):Finalbutanolconcentrationat72h.Biomass(g/L):Cellbiomassat24h.cC :Butyricacid butanol butyricacid concentrationat24h(peakconcentration).C (g/L):Concentrationofundissociatedbutyricacidat24h. UBA q:Specificbutanolproductionrates(between24and36h,h−1)werecalculatedbasedonthefollowingequation:q =(C −C )/(12× p p butanol,36h butanol,24h Xcellbiomass,24h). Yangetal.AMBExpress2013,3:3 Page4of8 http://www.amb-express.com/content/3/1/3 A B 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 7 7 50 Glucose 12 0.8 50 Glucose Butanol 12 0.8 6 40 pH 10 6 40 pH 10 0.6 0.6 8 8 pH45 Glucose (g/L) 2300 Butanol 46 Products (g/L) 00..24Biomass (g/L) pH 45 Glucose (g/L)2300 Butyric acid Acetone 46 Products (g/L) 00..24 Biomass g/L () 10 Biomass Acetic acid Acetone 2 10 Biomass Acetic acid 2 3 0 Butyric acid Ethanol 0 0.0 3 0 Ethanol 0 0.0 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Fermentation time (h) Fermentation time (h) C D 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 7 7 50 Glucose 12 0.8 50 Glucose 12 0.8 6 40 pH 10 6 40 pH 10 0.6 0.6 8 8 pH45 Glucose (g/L)2300 Butanol 46 Products (g/L) 0.4Biomass (g/L) pH 45 Glucose (g/L)2300 Butanol 46Products (g/L) 0.4Biomass (g/L) 0.2 0.2 10 Biomass BAucteytriicc aacciidd 2 10 Biomass Butyric acid Acetic acid 2 3 0 Acetone Ethanol 0 0.0 3 0 AEctehtaonnoel 0 0.0 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Fermentation time (h) Fermentation time (h) Figure1EffectofNaOHneutralizationonbutanolproductionbyC.acetobutylicum(A:fermentationcontrolwithoutpHcontrol;B neutralizationat12h;C:neutralizationat24h;D:neutralizationat36h). initial pH was 6.5 for all the CaCO supplement treat- fermentation,thepHwas4.4,4.7,and5.3withtheCaCO 3 3 mentsandthecontrol.TheresultsshowedCaCO supple- supplementation at 2, 5 and 8 g/L respectively. And the 3 mentation improved glucose uptake significantly resulting corresponding undissociated butyric acid was 2.7, 3.0 and innearlycompleteconsumptionofglucose(Figure2A-C). 1.5g/Lat24h.Thefinalbutanolconcentration(10.8g/L) Interestingly, we found that the CaCO supplementation was higher when the fermentation pH (24 h) was kept at 3 at the concentration of 2, 5 or 8 g/L extended the cell 4.7thanthat(8.8g/L)atpH5.3.Itsuggestedthatbutanol growth to 36 h, and correspondingly increased the cell production was not only related to the cell biomass, but biomassfrom0.43g/L(control)to0.51,0.57and0.67g/L alsocorrelated tothe fermentationpHand the undissoci- at 36 h respectively. Consequently, we observed that the ated butyric acid. With the CaCO supplementation, the 3 butanol production rates (between 36–48 h) increased fermentation pH controlled between 4.4 and 5.3 at 24 h dramatically from 0.05 g/L h (the control) to 0.10, 0.36 could overcome the acid flush, but higher undissociated and0.24g/L·hwhentheCaCO supplementationofat2, butyric acid probably was a prerequisite for fast butanol 3 5,and8g/Lrespectively.Thisindicatedthestrongcorrel- production in the solventogenic phase. The results also ation between cell biomass and butanol production, but indicated that keeping higher pH (5.3) in fermentation highcell biomass could also result in higherbutyrate, not favored higher butyric acid production. Although the butanol ifthe pHwasnot controlled within the appropri- butyric acid could be converted into butanol in the later ate range. For example, the higher CaCO supplementa- solventogenic phase; the final butyric acid concentration 3 tion (8 g/L) resulted in the highest cell biomass and was still high. The similar results of pH effect on butyric butyrateconcentration(4.0g/L)at72h,whilethebutanol acid productionwerereportedinthepreviousresearchin concentration was only 7.8 g/L. (Figure 2B and 2C). The which Monot et al. found that the medium with higher highest butanol concentration reached 10.8 g/L at 72 h pH level produced more acids than the medium with with the supplementation of 5 g/L CaCO . At 24 h of lowerpHlevel(Monotetal.1984). 3 Yangetal.AMBExpress2013,3:3 Page5of8 http://www.amb-express.com/content/3/1/3 A B 0 10 20 30 40 50 60 70 7 50 Glucose 12 0.8 7 0 10 20 30 40 50 60 70 6 40 pH 10 50 Glucose Butanol 12 0.8 Butanol 0.6 6 40 pH 10 8 0.6 pH45 Glucose (g/L)2300 Butyric acid Acetone 46Products (g/L) 00..24Biomass (g/L) pH45 Glucose (g/L)2300 Butyric acid 468Products (g/L) 0.4Biomass (g/L) 10 Biomass Acetic acid 2 10 Biomass Acetic acid Acetone 2 0.2 3 0 Ethanol 0 0.0 3 0 Ethanol 0 0.0 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Fermentation time (h) Fermentation time (h) C 0 10 20 30 40 50 60 70 7 12 50 0.8 6 40 Glucose pH 10 0.6 pH45 Glucose (g/L)2300 Butyric acid Butanol 468 Products (g/L) 0.4Biomass (g/L) Acetone 0.2 10 Biomass Acetic acid 2 3 0 Ethanol 0 0.0 0 10 20 30 40 50 60 70 Fermentation time (h) Figure2EffectofCaCO supplementationonbutanolproductionbyC.acetobutylicum(A:CaCO supplementationat2g/L;BCaCO 3 3 3 supplementationat5g/L;C:CaCO supplementationat8g/L). 3 Effectofsodiumacetatebufferingonbutanolandbutyric 4.4 resulted in the termination of glucose consump- acidproductionbyC.acetobutylicum tion. We also observed the lag phase was reduced sig- Bufferingisanothercommonmethodusedinvariousfer- nificantly at the high initial buffering pH (5.0 and 5.6) mentation processes to regulate pH (Bryant and Blaschek compared with that at a low buffering pH (4.6). Buffer- 1988). Therefore, the effects of different initial NaOAc ing the initial pH at 5.0 produced the highest butanol buffering pH on butanol and butyric acid production concentration 12.3 g/L at 72 h. Although buffering the by C. acetobutylicum were investigated to evaluate its initial pH at 5.6 had the highest butyric acid concen- effectiveness in comparison with those of CaCO sup- tration 4.0 g/L at 24 h, further conversion of butyric 3 plementation and NaOH neutralization. Sodium acet- acid to butanol was slow and the butanol yield ate was chosen to buffer pH at 4.6, 5.0, and 5.6 initially remained at a low level at the end of fermentation. by changing the ratio of NaOAc and HOAc. The results Comparing three pH control methods revealed high showed that NaOAc buffering at different initial pH similarities of the fermentation patterns between the could affect the fermentation pH profiles, cell growth, sodium acetate buffering and the CaCO supplementa- 3 glucose consumption and butanol production signifi- tion. In both methods, maintaining the pH above 5.0 cantly (Figure 3A-C). Buffering at initial pH 5.0 and at 24 h favored the butyric acid production, maintain- 5.6 appeared to overcome the acid flush and enable ing the pH between 4.4 and 4.7 at 24 h promoted the complete glucose consumption, while buffering at ini- butanol production. More interestingly, this optimal tial pH 4.6 resulted in incomplete consumption of glu- pH favoring the butanol production was also reflected cose and ~20 g/L glucose was not utilized at 72 h. At in the NaOH neutralization, in which the pH at 24 h the end of exponential phase (24 h), the cell biomass was 4.7 by adjusting pH with NaOH at 12 h. The reached 0.51, 0.63, 0.70 g/L and the fermentation pH observed cell growth took place mainly between 12 dropped to 4.1, 4.4, and 5.0 respectively (for initial buf- and 24 h in fermentation, the cell biomass (DCW) fering pH at 4.6, 5.0 and 5.6). This indicated that pH increased from 0.1 g/L (at 12 h) to 0.7 g/L (at 24 h), 4.4 probably was a threshold for acid flush, pH below and then remained almost constant. It suggested the Yangetal.AMBExpress2013,3:3 Page6of8 http://www.amb-express.com/content/3/1/3 A B 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 7 10 50 Glucose 12 0.8 50 Glucose Butanol 12 0.8 8 6 40 10 40 10 0.6 0.6 pH45 Glucose (g/L)2300 pH Butanol 468Products (g/L) 00..24Biomass (g/L) pH46 Glucose (g/L)2300 BipoHmass Butyric acid Acetone 468Products (g/L) 00..24Biomass (g/L) 10 Biomass Acetic acid Acetone 2 2 10 Acetic acid 2 Butyric acid 3 0 Ethanol 0 0.0 0 Ethanol 0 0.0 0 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Fermentation time (h) Fermentation time (h) C 0 10 20 30 40 50 60 70 7 50 12 Glucose 0.8 6 40 10 0.6 pH Butanol 8 pH5 Glucose (g/L)2300 Butyric acid 46Products (g/L) 0.4Biomass (g/L) 4 0.2 10 Biomass Acetone Acetic acid 2 3 0 Ethanol 0 0.0 0 10 20 30 40 50 60 70 Fermentation time (h) Figure3EffectofNaOAcbufferingonbutanolproductionbyC.acetobutylicum(A:NaOAcbufferingatinitialpH4.6;BNaOAc bufferingatinitialpH5.0;C:NaOAcbufferingatinitialpH5.6). first 12–24 h probably was the exponential phase of cell growth and butyric acid was produced associated withcellgrowthmainlyinthisphase.Attheendofexpo- nential phase, bacteria started to generate the butanol dehydrogenase, which was responsible for producing butanol. Therefore, controlling the fermentation pH between 4.4–4.7 at 12–24 h is essential for the maximum 0.80 CaCO supplementation 3 butanolproduction. Sodium acetate buffering 0.75 buffering at pH 5.6 h) 6 Discussion s (3 0.70 s a EffectsofpHcontrolonthemaximumbiomass m 0.65 To seek the potential mechanisms involved in persistent Bio buffering at pH 5.0 8 g/L CaCO3 m 0.60 butanol production, we observed both the cell biomass u m buffering at pH 4.6 and the undissociated butyric acids correlated well with Maxi 0.55 5 g/L CaCO3 butanol production rates. The cell biomass could be related to the total enzyme activity in the later solvent 0.50 2 g/L CaCO3 productionphase.UnderthepHcontrolwithCaCO sup- 3 0.45 plementation and NaOAc buffering, the cell biomass 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 reachedthemaximumat36h.Weobservedalinearrela- Fermentation pH at 24 h tionship between the maximum cell biomass and the fer- Figure4CorrelationbetweenfermentationpH(24h)and mentation pH at 24 h (Figure 4). This indicated that the maximumbiomass(36h)inthebatchfermentationofC. acetobutylicum. higher pH at the end of exponential phases (in the pH Yangetal.AMBExpress2013,3:3 Page7of8 http://www.amb-express.com/content/3/1/3 range of 4.1–5.3), the higher the maximum biomass was achieved. This finding was consistent with the previ- 0.60 ous report with a pH-controlled batch fermentation of -1h) 0.55 C. acetobutylicum using a fermentor (Monot et al. 1984), s ( e inwhichtheincreaseofthefermentationpHbetweenpH at 0.50 n r 4.5and5.5enhancedthemaximumbiomass,butlowered o cti 0.45 the maximum biomass at a pH level higher than pH 6.0. du o In our work, we did not carry out the experiment at pr 0.40 pH≥6.0 because the low butanol yield would be pro- anol 0.35 duced at a higher pH. Figure 4 also showed thatbuffer- but CaCO3 supplementation ing the initial pH at 4.6 resulted in the fermentation cific 0.30 Sodium acetate buffering e pH 4.1 at 24 h and the lowest maximum biomass at sp 0.25 36 h. It indicated the low pH (~4.1) at the end of expo- q, p 0.20 nential phase inhibited the cell growth and the glucose 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 utilization. The decease of pH from 4.6 to 4.1 was C , Undissociated butyric acid (g/L) UBA likely caused by the production of butyric acid and acetic Figure5Correlationbetweenundissociatedbutyricacid acid in the acidogenic phase, where both acids reached concentration(24h)andspecificbutanolproductionrate their maximum concentrations at 24 h (Figures 1, 2, 3). (between24and36h)inthebatchfermentationofC. Butyric acid and acetic acid have been reported as two acetobutylicum. majortoxicproductsinhibitingcell growthofC.acetobu- tylicum based on their low inhibition constants (0.07 M required to initiate the solventogenesis (Husemann and and 0.19 M respectively)(Costa and Moreira 1983). As a Papoutsakis 1988; Monot et al. 1984). In our experi- result, microorganism probably develops a detoxification ments, the onset of solvent production took place at process involved a series of enzymes to convert butyric much lower level of the undissociated butyric acid acid to butanol, and acetic acid to acetone in a solven- (0.20–0.30 g/L), which may attributed to the use of dif- togenic phase (Monot et al. 1984). Therefore, relatively ferent media and pH controlled methods. The persistent low pH (<5.0) at the end of exponential phase is essen- butanol production could be more important than initial tial to maintain the persistent solvent production. butanol production, since the persistent solvent produc- More precisely, the concentration of undissociated tion would be essential for a higher final butanol con- butyric acid probably controls the butanol production centration at the end of fermentation. The specific rates in solventogenesis. butanol production rates at the end of exponential phase could represent the capability of persistent butanol pro- Effectsoftheundissociatedbutyricacidonspecific duction. For example, in the both cases of CaCO sup- 3 butanolproductionrates plementation (5 g/L) and NaOAc buffering (pH 5.0), the The correlation of the undissociated butyric acid (at higher the specific butanol production rates (0.52 h−1 24 h) with the specific butanol production rates between and 0.58 h−1), the higher the final butanol concentration 24 and 36 h was established in the fermentations with (10.8 g/L and 12.3 g/L) (Table 1). The cell biomass is es- CaCO supplementation and NaOAc buffering. The bu- sential for the evaluation of the persistent butanol prod- 3 tanol production rates increased dramatically between uctivity and therefore has been taken into accounts for 24 and 36 h. The undissociated butyric acid reached the the specific butanol production rates, because the maximum concentration at 24 h (the end of exponential enzymes (butanol dehydrogenase and CoA transferase) phase) except for the cases by adding 8 g/L of CaCO or induced or produced for solventogenesis are related to 3 buffering the initial pH at 5.6. In these two special cases, the cell biomass. The undissociated acetic acid was the maximum concentration of undissociated butyric suggested not required for solventogenesis (Husemann acid wasobtained at 36h;consequently the butanol pro- and Papoutsakis 1988). But the addition of acetate and duction rates were maximized between 36 and 48 h propionate could enhance final solvent concentrations probably due to the extended cell growth. Figure 5 (Husemann and Papoutsakis 1990; Mattaelammouri showed the specific butanol production rates (between et al. 1987). And the addition of butyrate demonstrated 24 and 36) were a function of the undissociated butyric the potential to initiate the solvent production at pH 5 acid concentration (at 24 h). The linear relationship be- (Holt et al. 1984). The addition of butyrate probable tween them indicated the undissociated butyric acid was increases the concentration of undissociated butyric the major driving force for the production of butanol. acid, which can induce the corresponding enzymes Previously, others have reported a minimum amount synthesis and drive the butanol production (Husemann (0.5 g/L–1.5 g/L) of undissociated butyric acid was and Papoutsakis 1989). Yangetal.AMBExpress2013,3:3 Page8of8 http://www.amb-express.com/content/3/1/3 In this study, the pH control was essential for the im- HusemannMHW,PapoutsakisET(1990)Effectsofpropionateandacetate provement of butanol production by C. acetobutylicum, additionsonsolventproductioninbatchculturesofclostridium- acetobutylicum.ApplEnvironMicrob56(5):1497–1500 because the appropriate pH in the exponential phase JonesDT,WoodsDR(1986)Acetone-butanolfermentationrevisited.Microbiol could enhance the specific butanol production rates by Rev50(4):484–524 producing higher undissociated butyric acid. NaOH LiSY,SrivastavaR,ParnasRS(2011a)Studyofinsitu1-butanolpervaporation froma-B-EfermentationusingaPDMScompositemembrane:validityof neutralization was a simple method to control the pH, solution-diffusionmodelforpervaporativea-B-Efermentation.Biotechnol and the key time to adjust the fermentation pH is the Progr27(1):111–120 beginning of exponential phase. CaCO supplementation LiSY,SrivastavaR,SuibSL,LiY,ParnasRS(2011b)Performanceofbatch,fed- 3 batch,andcontinuousa-B-EfermentationwithpH-control.Bioresource resulted in relatively low content of cell biomass and Technol102(5):4241–4250 high yield of butyrate. NaOAc buffering at pH 5.0 MaddoxIS,SteinerE,HirschS,WessnerS,GutierrezNA,GapesJR,SchusterKC enabled a highest butanol production rate in the station- (2000)Thecauseof“acidcrash”and“acidogenicfermentations”dupingthe batchacetone-butanol-ethanol(ABE-)fermentationprocess.JMolMicrob ary phase. The linear correlation of the undissociated Biotech2(1):95–100 butyric acid with the specific butanol production rates MartinJR,PetitdemangeH,BallongueJ,GayR(1983)Effectsofaceticandbutyric suggests the undissociated butyric acid could be the acidsonsolventsproductionbyclostridium-acetobutylicum.BiotechnolLett 5(2):89–94 major driving force for butanol production. The key to MattaelammouriG,JanatiidrissiR,JunellesAM,PetitdemangeH,GayR(1987) achieve high butanol concentration is to maintain a pH Effectsofbutyricandacetic-acidsonacetonebutanolformationby level around 4.7 and a higher level of undissociated clostridium-acetobutylicum.Biochimie69(2):109–115 MilneCB,EddyJA,RajuR,ArdekaniS,KimPJ,SengerRS,JinYS,BlaschekHP, butyric acidduringtheexponential phaseofcellgrowth. PriceND(2011)Metabolicnetworkreconstructionandgenome-scalemodel ofbutanol-producingstrainclostridiumbeijerinckiiNCIMB8052.BMCSyst Biol5(130).doi:10.1186/1752-0509-5-130 Competinginterests MonotF,EngasserJM,PetitdemangeH(1984)InfluenceofpHandundissociated Theauthorsdeclarethattheyhavenocompetinginterests. butyric-acidontheproductionofacetoneandbutanolinbatchculturesof clostridium-acetobutylicum.ApplMicrobiolBiot19(6):422–426 Acknowledgements QureshiN,EzejiTC(2008)Butanol,‘asuperiorbiofuel’productionfrom ThisstudywassupportedinpartbygrantsfromAlabamaAgricultural agriculturalresidues(renewablebiomass):recentprogressintechnology. ExperimentalStation. BiofuelBioprodBior2(4):319–330 QureshiN,SahaBC,CottaMA(2008)Butanolproductionfromwheatstrawby Authordetails simultaneoussaccharificationandfermentationusingclostridiumbeijerinckii: 1ForestProductsLabandCenterforBioenergyandBioproducts,Auburn partII-Fed-batchfermentation.BiomassBioenerg32(2):176–183 University,520DevallDrive,Auburn,AL36849,USA.2Departmentof RobinsonJR,KeatingJK,BoussaidAB,MansfieldSM,SaddlerJS(2002)The BiosystemsEngineering,AuburnUniversity,Auburn,AL36849,USA. influenceofbarkonthefermentationofdouglas-firwhitewoodpre- 3DepartmentofAgricultureandBiologicalEngineering,UniversityofFlorida, hydrolysates.ApplMicrobiolBiotechnol59(4):443–448.doi:10.1007/s00253- Gainesville,FL32611,USA. 002-1055-z RoosJW,MclaughlinJK,PapoutsakisET(1985)TheeffectofpHonnitrogen Received:6December2012Accepted:3January2013 supply,cell-lysis,andsolventproductioninfermentationsofclostridium- Published:7January2013 acetobutylicum.BiotechnolBioeng27(5):681–694 VandakD,ZigovaJ,SturdikE,SchlosserS(1997)EvaluationofsolventandpHfor extractivefermentationofbutyricacid.ProcessBiochem32(3):245–251 References WangSH,ZhangYP,DongHJ,MaoSM,ZhuY,WangRJ,LuanGD,LiY(2011) BryantDL,BlaschekHP(1988)Bufferingasameansforincreasinggrowthand Formicacidtriggersthe“acidcrash”ofacetone-butanol-ethanol butanolproductionbyclostridium-acetobutylicum.JIndMicrobiol3(1):49–55 fermentationbyclostridiumacetobutylicum.ApplEnvironMicrob CostaJM,MoreiraAR(1983)Growth-inhibitionkineticsfortheacetone-butanol 77(5):1674–1680 fermentation.AcsSymSer207:501–512 WelchRW,RudolphFB,PapoutsakisET(1989)Purificationandcharacterizationof FondO,MattaammouriG,PetitdemangeH,EngasserJM(1985)Theroleofacids thenadh-dependentbutanoldehydrogenasefromclostridium- ontheproductionofacetoneandbutanolbyclostridium-acetobutylicum. acetobutylicum(atcc824).ArchBiochemBiophys273(2):309–318 ApplMicrobiolBiot22(3):195–200 YuEKC,SaddlerJN(1983)Enhancedacetone-butanolfermentationby GeorgeHA,ChenJS(1983)AcidicconditionsAreNotobligatoryforonsetof clostridium-acetobutylicumgrownonD-xyloseinthepresenceofaceticor butanolformationbyclostridium-beijerinckii(synonym,clostridium- butyric-acid.FEMSMicrobiolLett18(1–2):103–107 butylicum).ApplEnvironMicrob46(2):321–327 ZverlovVV,BerezinaO,VelikodvorskayaGA,SchwarzWH(2006)Bacterialacetone HessG(2006)BPanddupontplan‘biobutanol’.ChemEngNews84(26):9 andbutanolproductionbyindustrialfermentationinthesovietunion:use HoltRA,StephensGM,MorrisJG(1984)Productionofsolventsbyclostridium- ofhydrolyzedagriculturalwasteforbiorefinery.ApplMicrobiolBiot acetobutylicumculturesmaintainedatneutralpH.ApplEnvironMicrob 71(5):587–597 48(6):1166–1170 HuangL,ForsbergCW,GibbinsLN(1986)InfluenceofexternalpHand doi:10.1186/2191-0855-3-3 fermentationproductsonclostridium-acetobutylicumintracellularpHand Citethisarticleas:Yangetal.:AcomparisonofthreepHcontrol cellular-distributionoffermentationproducts.ApplEnvironMicrob methodsforrevealingeffectsofundissociatedbutyricacidonspecific 51(6):1230–1234 butanolproductionrateinbatchfermentationofClostridium HuangLP,JinB,LantP,ZhouJT(2005)Simultaneoussaccharificationand acetobutylicum.AMBExpress20133:3. fermentationofpotatostarchwastewatertolacticacidbyrhizopusoryzae andrhizopusarrhizus.BiochemEngJ23(3):265–276 HusemannMHW,PapoutsakisET(1988)Solventogenesisinclostridium- acetobutylicumfermentationsrelatedtocarboxylic-acidandproton concentrations.BiotechnolBioeng32(7):843–852 HusemannMHW,PapoutsakisET(1989)Enzymeslimitingbutanolandacetone formationincontinuousandbatchculturesofclostridium-acetobutylicum. ApplMicrobiolBiot31(5–6):435–444