PPrreettrreeaattmmeenntt hhyyddrroollyyssaattee ((ssppeenntt lliiqquuoorr)) SSiizzee e e PPrreettrreeaattmmeenntt tt aa rreedduuccttiioonn rr ss tt ll ss aa bb cc uu mimi ss dd ee hh lili oo CC SS SStteeaamm Lignosulfonate and elevated pH can enhance enzymatic saccharification of lignocelluloses Wang et al. Wangetal.BiotechnologyforBiofuels2013,6:9 http://www.biotechnologyforbiofuels.com/content/6/1/9 Wangetal.BiotechnologyforBiofuels2013,6:9 http://www.biotechnologyforbiofuels.com/content/6/1/9 RESEARCH Open Access Lignosulfonate and elevated pH can enhance enzymatic saccharification of lignocelluloses ZJ Wang1,2, TQ Lan2,3 and JY Zhu2* Abstract Background: Nonspecific (nonproductive) binding (adsorption) of cellulase by lignin has been identified as a key barrier to reduce cellulase loading for economical sugar and biofuel production from lignocellulosic biomass.Sulfite Pretreatment to Overcome Recalcitrance of Lignocelluloses (SPORL)is a relatively new process, but demonstrated robust performance for sugar and biofuel production from woody biomass especially softwoods in terms of yields and energy efficiencies.This study demonstrated therole of lignin sulfonation inenhancing enzymatic saccharification of lignocelluloses –lignosulfonate from SPORL can improve enzymatic hydrolysis of lignocelluloses, contrary to the conventional belief that lignin inhibits enzymatic hydrolysis due to nonspecific binding ofcellulase. Results: The study found thatlignosulfonate from SPORL pretreatment and from a commercial source inhibits enzymatic hydrolysis of pure cellulosic substrates at low concentrations due to nonspecific binding ofcellulase. Surprisingly, the reduction in enzymatic saccharification efficiency of a lignocellulosic substrate was fully recovered as the concentrations of thesetwo lignosulfonates increased.We hypothesize that lignosulfonate serves as a surfactant to enhanceenzymatichydrolysis athigher concentrations and that this enhancement offsets itsinhibitive effect from nonspecific binding of cellulase, when lignosulfonateis applied to lignocellulosic solid substrates. Lignosulfonate can blocknonspecific bindingof cellulase by bound lignin on thesolid substrates, in thesame manner as a nonionic surfactant, to significantly enhance enzymatic saccharification. This enhancement is linearly proportional to the amount of lignosulfonate applied which is very important to practical applications. For a SPORL-pretreatedlodgepole pinesolid, 90% cellulose saccharification was achieved at cellulase loading of 13 FPU/g glucan with theapplication of itscorresponding pretreatment hydrolysate coupled with increasing hydrolysis pHto above 5.5comparedwith only51% for thecontrol run without lignosulfonate atpH 5.0. The pH-inducedlignin surface modification atpH 5.5furtherreduced nonspecific binding of cellulase by lignosulfonate. Conclusions: The results reported inthis study suggest significant advantages for SPORL-pretreatmentin terms of reducing water usage and enzyme dosage, and simplifying process integration, i.e., it should eliminate washing of SPORL solid fraction for direct simultaneous enzymatic saccharification and combinedfermentationof enzymatic and pretreatment hydrolysates (SSCombF). Elevated pH 5.5or higher, rather thanthe commonly believedoptimal and widely practiced pH 4.8-5.0, should be used inconductingenzymatic saccharification of lignocelluloses. Keywords: Enzymatic hydrolysis/saccharification, Lignin/lignosulfonate, Nonspecific/nonproductivebinding/ adsorption, Cellulase enzymes, Pretreatment *Correspondence:[email protected] 2USForestService,ForestProductsLaboratory,Madison,WI,USA Fulllistofauthorinformationisavailableattheendofthearticle ©2013Wangetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited. Wangetal.BiotechnologyforBiofuels2013,6:9 Page2of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Background reduce nonspecific binding of cellulase to the bound The interactions between lignin and cellulase play a lignin fraction in lignocelluloses in the same manner as major role in enzymatic hydrolysis of lignocelluloses for other surfactants to result in a gain in enzymatic hy- sugar and biofuel production from biomass [1]. These drolysis efficiency. Furthermore, it is possible that the interactions can be described as (1) lignin physical application of certain lignosulfonate, e.g., spent liquors blockage to limit cellulose accessibility to cellulase [2], from SPORL pretreatments, with low nonspecific bind- and (2) nonspecific adsorption or binding of cellulase ings of cellulase – hypothesized in the previous para- enzymes to lignin [3-7]. Reported studies indicate that graph – may result in a net enhancement of enzymatic these two mechanisms produced negative effects on en- saccharificationoflignocelluloses. The validationorcon- zymatic saccharification of lignocelluloses. Pretreatment firmation of this argument has significant scientific and of lignocelluloses such as Organosolv [8,9] and SPORL practical implications for sulfite pretreatment technolo- [10] is able to partially remove lignin physical blockage gies, such as SPORL [10] that has been demonstrated bysolubilizingafractionoflignininto thehemicellulosic robust performance for sugar and biofuel production sugar stream (pretreatment spent liquor). However, fur- with very high yields from woody biomass including very therprocessinglignocellulosestoremove ligninblockage recalcitrantsoftwoodspecies[24-26].Specifically,thesep- such as by delignification is not only expensive but also aration of the SPORL pretreatment hydrolysate (spent li- maynotbenecessary intermsofimprovingcellulose ac- quor) from the SPORL pretreated solid ligno cellulosic cessibility. It is probably more effective to address the fraction and washing of the SPORL solid fraction would issue of nonspecific binding of cellulase to lignin to fur- notbe required in order to enhance enzymatic saccharifi- ther enhance enzymatic saccharification. One passive cation. This can significantly simplify biorefinery process approach was to wash the solid fraction of pretreated integrationandsaveasignificantlyamountofwater. lignocelluloses to eliminate nonspecific binding of cellu- Previously, we demonstrated that the application of lase to free lignin (referring to lignin separated from divalent metal compound, such as Ca(II) applied for solid lignocellulosic substrate), as commonly described neutralizing pretreatment hydrolysate, can eliminate in the literature [11]. Another passive approach is to use nonspecific cellulase binding to lignosulfonate in the un- surfactant, protein and metal compound to block bound washed SPORL pretreated aspen (hardwood) solids lignin (referring to lignin retained in solid lignocellulosic [18,19]. The aspen lignosuflonate in the SPORL spent li- substrate) and free lignin, reducing their nonspecific quor had a low degree of lignin sulfonation because of a binding to cellulase [3,12-19]. However, washing is a sig- low sulfite loading of 3% on oven dry (od) wood used in nificant environmental concern because the amount of pretreatment. This study is a step further in demonstrat- water required is on the order of 10 m3 water/ton ligno- ing the role of lignin sulfonation in enhancing enzymatic cellulose, based on pulp mill experience [19]. The appli- saccharification. SPORL spent liquor that contain ligno- cations of surfactant and protein are both expensive at sulfonatewithhighdegreeofsulfonation,producedfrom therequiredlevels. lodgepole pine (softwood) by SPORL at 8% sulfite load- Hydrophobic interaction hasbeenidentifiedasthepri- ing on od wood, was directly mixed with SPORL pre- mary driving force for protein adsorption [20]. Increas- treated lodgepole pine and dilute acid pretreated aspen. ingthehydrophobicity ofa substrate resultsinenhanced The objectives of the present study are: (1) to verify the adsorption of protein or cellulase [1,21]. This suggests two hypotheses proposed above by directly comparing that nonspecific binding (adsorption) of cellulase (made enzymatic cellulose saccharification efficiencies of sev- of protein) to lignin will be different for lignins of differ- eral SPORL-pretreated lignocellulosic substrates with ent hydrophobicities. Sulfonated lignin, has good hydro- and without the applications of a SPORL-pretreatment philic properties. We can hypothesize that sulfonated hydrolysate from lodgepole pine at different loadings; lignin such as lignosulfonate in the sulfite pretreatment (2) to verify our previous finding [27,28]: significant im- hydrolysates (spent liquor) [9,10] may produce less non- provement in cellulose saccharification when enzymatic specific binding (adsorption) to cellulase enzymes. This hydrolysis of lignocelluloses is conducted at an elevated hypothesis is indirectly corroborated by the excellent en- pH 5.5 – 6.2 as oppose to pH 4.8 – 5.0 as exclusively zymatic digestibility of lignocellulosic substrates after used in the literature. The elevated pH study was con- sulfite pretreatment such as SPORL[10], sulfite pulping ducted with the application of SPORL pretreatment hy- [22],andligninsulfonation[23]. drolysate for further enhancement of saccharification Lignosulfonate functions as a surfactant due to its efficiency.Thisworklaysthefoundationfordirectsimul- stronghydrophilicity.Thispropertyhasbeenusedtode- taneous enzymatic saccharification and combined fer- velop commercial surface modification products such as mentation (SSFCombF) of the whole lignocellulosic dispersants and plasticizers. Therefore, we can further slurry at high solids loadings without a separation and hypothesize that the application of lignosulfonate can washingstepafterpretreatment[28]. Wangetal.BiotechnologyforBiofuels2013,6:9 Page3of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Results and discussions recovery achieved at lignosulfonate concentration ap- EnzymatichydrolysisofWhatmanpaperwiththe proximately 0.3 g/L as Klason lignin (Figure 2a). Al- additionoflignosulfonates though a few data points have large standard deviations Enzymatic hydrolysis of a pure cellulose substrate of for the duplicate data sets of 24 and 48 h, the results Whatman paper can be inhibited by a commercial ligno- showedthesimilartrend.Thesephenomena suggestthat sulfonate as revealed in our previous study [18]. This in- lignosulfonate can be inhibitive to enzymatic hydrolysis hibitive phenomenon was verified in this study at high of cellulose due to nonspecific binding of cellulase as concentrations upto 10 g/L using the same commercial previously known. The recovery of SED at high lignosul- lignosulfonate (Figure 1). The pH of the buffer solution fonate concentrations can be explained first by the ionic was4.8andthemeasuredpHinthehydrolysissuspension strength effect [29]. The neutralization of the pretreat- was 4.53. The reduction in substrate enzymatic digestibil- ment hydrolysate using NaOH solution resulted in a Na+ ity(SED),definedasthepercentageofglucaninthecellu- concentration in the mixed enzymatic and pretreatment losic solid substrate converted to glucose enzymatically, hydrolysate of 23.5 mmol/L. Our previous study found wasrapid at lowlignosulfonate concentrationofless than that enzymatic hydrolysis of a pretreated lignocellulosic 0.5 g/L (observable after 6 h hydrolysis) and then slowed substrate can be improved by 4% at K+ concentration of significantly, in agreement with our previous study [18]. 2 mmol/L [19]. Lignosulfonate as a surfactant may also This reduction in enzymatic hydrolysis is due to the non- playaroletoenhance enzymatichydrolysisof pure cellu- specificlignosulfonatebindingofcellulase. lose [3]. The extent of this positive effect of surfactant is A very interesting phenomena was observed when a small forpurecellulosesamples,however, and varieswith SPORL-pretreatment lodgepople hydrolysate (L-BD4- the amount of surfactant applied. This explains the grad- T85-3) was added into the enzymatic hydrolysis suspen- ual recovery trend of SED with the amount of lignosulfo- sion (buffered at pH=4.8) of Whatman paper. The nate added (Figure 2a and b). The fact that recovery of SPORL hydrolysate, L-BD4-T85-3, was first neutralized SED did not take place in Figure 1 using a commercial from pH approximately 1.5 to 4.8 using NaOH before lignosulfonate alone without the application of ions adding to the Whatman paper suspension. The reduc- suggests that ionic strength mechanism may play an im- tion in SED was also apparent (observable after 24 h portant role for the observed recovery of glucose concen- hydrolysis) and substantial at low lignosulfonate concen- trationinFigure2a. tration of less than 0.2 g/L as Klason lignin (Figure 2a). Our previous study indicated that a divalent metal can However, SED started to recover when lignosulfoante form complex with lignosulfoante to reduce its affinity concentration was further increased (the glucose in the to cellulase [18]. We conducted a separate study by neu- pretreated hydrolysate was subtracted from the mea- tralizingthesameSPORL-pretreatmenthydrolysateusedin sured glucose concentration in the mixed hydrolysate in theprevioussetofexperiments(Figure2a)usingCa(OH) . 2 calculating SED). The recovery was also rapid with 100% Thesametrendof SEDreductionandrecoverywiththe addition of lignosulfonate as discussed in the previous paragraph was observed. However, slightly higher SEDs than that for the control run without lignosulfonate were obtained after 48 hours hydrolysis when lignosul- fonate concentration was greater than 0.3 g/L as Klason lignin (Figure 2b). The difference in SED between runs using NaOH (Figure 2a) and Ca(OH) neutralization 2 (Figure2b)werewithintheerrormargin,suggestingthe lodgepole pine lignosulfonate produced using a high sulfite dosage of 8% on wood may already has very low affinity to cellulase due to its high degree of sulfonation or hydrolphlicity. The formation of lignosulfonate-Ca(II) complex on lignin nonspecific binding to cellulase is not significant. EnzymatichydrolysisofaSPORL-pretreatedlodgepolepine withtheapplicationofSPORL-pretreatmenthydrolysates The application of non-ionic surfactant can significantly Figure1Effectoftheapplicationofacommerciallignosulfonate reduce nonspecific binding of cellulase by lignin to en- onsubstrateenzymaticdigestibility(SED)ofWhatmanpaper hance enzymatic hydrolysis of lignocellulsosic substrates (purecellulose)atcellulaseloadingof15FPU/gglucan. [3,5,17]. Similarly, the application of lignosulfonate, as a Wangetal.BiotechnologyforBiofuels2013,6:9 Page4of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Figure2EffectoftheapplicationofaSPORL-pretreatedlodgepolepinehydrolysate(L-BD4-T85-3)onsubstrateenzymaticdigestibility (SED)ofWhatmanpaper(purecellulose)atcellulaseloadingof15FPU/gglucan.(a)L-BD4-T85-3neutralizedusingNaOHtopH4.8;(b)L- BD4-T85-3neutralizedusingCa(OH) topH4.8. 2 surfactant,can enhance enzymatic hydrolysisoflignocel- non-ionic surfactant at equivalent cellulase loadings [3]. lulosic substrates. A SPORL – pretreatment hydrolysate The amount of glucose in L-BD4-T85-3 (Table 1) was from lodgepole pine (L-BD4-T85-3) was first neutralized subtracted from the measured glucose concentration in using NaOH to pH 4.8 and then applied to enzymatic the mixed hydrolysate for calculating SED. Lignosulfo- hydrolysis of a SPORL pretreated lodgepole pine solid nate concentration in the mixed hydrolysate at 1.0 g/L substrate (SP-BD4) at 2% solids (i.e., dilution with as Klason lignin is equivalent to lignin concentration in water). The results indicate cellulose conversion of SP- the whole slurry of the pretreated material when hydro- BD4 represented by SED increased linearly with the in- lyzed at2%waterinsolublesolids. crease in lignosulfonate concentration (> 0.2 g/L) in the Thegains inSED per unit mass oflignosulfonate load- mixed hydrolysate (Figure 3a), in agreement with that ing, ΔSED/m , for the results presented in Figure 3a lig reported in the literature [3,19]. Furthermore, the in- were calculated under different lignosulfonate concen- crease in SED, i.e., from approximately 49% to 72% at trations and various enzymatic hydrolysis durations. The 24 h, or from 52% to 78% (by 50%) at 72 h, by applying results indicate negative SED gains at low lignosulfonate lignosulfonate at 1.0 g/L (as Klason lignin) was also loadings(<0.2g/L asKlasonlignin)andshorthydrolysis qualitatively in agreement with the gain in SED of a SO duration of 6 h (Figure 3b). This again confirmed that 2 catalyzed steam exploded spruce substrate by applying lignosulfonate does have inhibitive effect on enzymatic Figure3EffectoftheapplicationofaSPORL-pretreatedlodgepolepinehydrolysate(L-BD4-T85-3)onenzymaticcellulose saccharificationofaSPORL-pretreatedlodgepolepine(SP-BD4)atcellulaseloadingof13FPU/gglucan.(a)Substrateenzymatic digestibility(SED);(b)GainsinSEDpermgoflignosulfonateapplied(%L/g). Wangetal.BiotechnologyforBiofuels2013,6:9 Page5of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Table1Listofsubstratesproducedalongwiththe with the increase in lignin concentration, has significant productionconditions importance for simultaneous enzymatic saccharification Sample Method Chemicalcharges T(°C) Duration@ and combined fermentation of the enzymatic and pretreat- Label1 onwood(%) T(min) ment hydrolysates (SScombF) using the whole slurry at a LodgepolePine–Solidcellulosicsubstrates highsolidsloadings. SP-BD4 SPORL HSO:2.2 180 20 Three SPORL-pretreatment hydrolysates, i.e., L-BD4- 2 4 T85-1, L-BD4-T85-2, L-BD4-T85-3, obtained from three NaHSO:8.0 3 pretreatmentswithdifferentpretreatmenttimes(Table1) LodgepolePine–Liquidsubstrates:Pretreatmenthydrolysates (spentliquor) were used to further verify the positive effects of the application of lignosulfonate on enzymatic hydrolysis of L-BD4-T85-1 SPORL HSO:2.2 185 5 2 4 lignocelluloses. Both NaOH and Ca(OH) were used to L-BD4-T85-3 NaHSO:8.0 185 25 2 3 neutralize the hydrolysates to pH 4.8 before application. L-BD4-T85-5 185 45 The results indicated that the enzymatic cellulose con- Aspen–Solidcellulosicsubstrates version efficiency, or SED, of a SPORL-pretreated lodge- SP-AS SPORL HSO:1.1 170 25 pole pine, SP-BD4, was increased by over 50% when 2 4 NaHSO:3.0 either one of the pretreatment hydrolysate was applied 3 (Figure 4a and b). The hydrolysate with shortest pre- SPH-AS HSO:0 170 25 2 4 treatment duration of5 min(Table 1), i.e.,L-BD4-T85-1, NaHSO:3.0 3 with the highest lignosulfonate concentration of 12.6 g/L DA-AS Diluteacid(DA) HSO:1.1 170 25 2 4 as Klason lignin (Table 1) produced highest gain in en- 1SPstandsforSPORL;SPHstandsforSPORLwithsodiumbisulfiteonly;DA zymatic saccharification, i.e., SED at 72 h was increased standsfordiluteacid.ASstandsforaspen. from approximately 52% to 87%, or by 67% (Figure 4a). hydrolysis. However, there are two competing processes, While the hydrolysate with the longest pretreatment lignosulfonate adsorbed by cellulase to reduce cellulase duration of 45 min (Table 1), L-BD4-T85-5, with lowest activityand lignosulfonateservesasa surfactant toblock lignosulfonate concentration of 9.7 g/L as Klason lignin lignin on the solid substrate to prevent cellulase adsorp- (Table 1), produced smallest gain in SED at 72 h tion on the solid substrate lignin. The second process increased by 48% (Figure 4a). This difference is mainly prevails at high lignosulfonate concentrations and longer attributed to the differences in applied lignosulfonate hydrolysis duration. The results also indicate that the SED dosage and in the composition of cellulase inhibitive gain at 48 h per unit mass of lignosulfonate, ΔSED/m , compounds other than lignin in the pretreatment hydro- lig initiallydecreasedrapidlyandthendecreasedveryslowlyat lysates [30]. It is expected that a longer pretreatment lignosulfonate concentration exceeding 0.2 g/L as Klason duration resulted in more inhibitive compounds (i.e., lignin.Atalignosulfonatedosageof1.0g/LasKlasonligin, furans) in the pretreatment hydrolysate, L-BD4-T85-5, ΔSED/m increased from approximately 2% L/g at hy- and therefore a less enhancement of SED, than a shorter lig drolysis time of 1 h to 25% L/g at 48 h. The slow- pretreatment. The differences in the properties of the diminishing effect, i.e., ΔSED/m decreased very slowing lignosulfonates due to the varied degree of sulfonation lig Figure4EffectoftheapplicationofdifferentSPORL-pretreatmentlodgepolepinehydrolysates(spentliquors)producedunder differentpretreatmentdurationsonSEDofSP-BD4.(a)LiquorsneutralizedusingNaOHto4.8;(b)LiquorsneutralizedusingCa(OH) to4.8. 2 Wangetal.BiotechnologyforBiofuels2013,6:9 Page6of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 under different pretreatment durations may also affect 13 FPU/g glucan after 72 h of hydrolysis for SP-BD4. The SEDenhancement. SED of the corresponding control run(pH 4.8without lig- Hydrolysate neutralization using Ca(OH) showed nosulfonate) was only 51%. When a similar substrate pro- 2 consistently but only slightly increased enhancement of duced from the same batch of lodgpole pine (BD4) wood enzymatic saccharification at 72 h than those observed chips under the identical SPORL-pertreatment conditions, using NaOH by comparing the results in Figure 4a to a cellulase loading of 24 FPU/g glucan (15 FPU/g sub- those in Figure 4b. The fact that the differences in SED strate) was required to achieve 90% cellulose conversion enhancements are pretty much within the measurement [31]. This suggests a savings in cellulase application of errors suggests the lignosulfonate produced at a high 46% can be achieved by using lignosulfonate at elevated sulfite dosage of 8% on wood may already has avery low pH. Furthermore, the increased pH is also favorable for affinity to cellulase. The role of divalent metal to form yeastfermentationsthroughSSFCombF. complex with lignosulfonate to reduce lignosulfonate A similar study was also conducted using a dilute binding of cellulase [18,19] is not very important. This is acid- (DA-AS) and SPORL- (SP-AS) pretreated aspen inagreementwith thatobservedfrom Figure2aandb. substrates with the addition of SPORL-pretreatment hy- drolysate at lignosulfonate concentration of 1.0 g/L as Improvedenzymatichydrolysisofalodgepoplepineby Klason lignin, L-BD4-T85-3, at elevated pH of 5.5. The combiningwithitspretreatmenthydrolysateandusing results indicate that using a high pH alone can enhance anelevatedpHof5.5 enzymatic saccharification as represented by SED gains Previouslywedemonstratedthatenzymaticsaccharification (Figure 5b). With the application of L-BD4-T85-3, SED efficiency of lignocelluloses can be increased significantly decreased slightly during the first 10 hours for both sub- whenhydrolysiswasconductedatanelevatedpHapproxi- strates (by comparing the solid symbols on solid line mately 5.0 or higher for SPORL-pretreated substrates with the corresponding open symbols on dash line), sug- [27,28].WemixedthesameSPORL-pretreatedhydrolysate gesting lignoslfonate can be inhibitive to enzymes. But (L-BD4-T85-3) with the SPORL-pretreated solid substrate the lignosulfonate can enhance enzymatic hydrolysis (SP-BD4) to conduct enzymatic hydrolysis at an elevated after24hours.TheSEDof bothsubstratesafter24hours buffer solution pH of 5.5. The results show that SED at were further enhanced at elevated pH 5.5. When com- 72hincreaseslinearlywithlignosulfonate concentrationin pared with the control run using buffer solution pH of the mixed hydrolysate (Figure 5a). Furthermore, an add- 4.8 without the application of pretreatment hydrolysate itional gain in SED at 72 h of 10 percentage points was (L-BD4-T85-3), the SED at 72 h were increased by 25 achieved when the pH of the buffer solution was elevated and 62% for the DA-AS and SP-AS, respectively. The to5.5.Thisdemonstratesthattheapplicationoflignosulfo- pH effect on improving enzymatic saccharification is nateatelevatedpHcanfurtherenhanceenzymatichydroly- more pronounced for the SPORL-pretreated sample sisoflignocelluloses.Cellulosesaccharificationefficiencyor compared with the dilute acid pretreated sample, and SED of 90% can be achieved at a cellulase loading of only agreeswith ourprevious study [28]. Figure5ComparisonsofSEDofdifferentpretreatedsolidsubstratesattwoenzymaticsaccharificationpHwiththeapplicationsofa SPORL-lodgepolepinehydrolysate(L-BD4-T85-3).(a)SEDofSPORLpretreatedlodgepolepine(SP-BD4)at72hundervariousL-BD4-T85-3 loadingsexpressedasconcentrationoflignosulfonateinthemixedhydrolysate,Celluclaseloading:13FPU/gglucan;(b)Time-dependentSEDsof SPORL-anddiluteacid-pretreatedaspensubstrates(DA-AS,SP-AS)atafixedL-BD4-T85-3loadingof1.0g/LoflignosulfonateasKlasonlignin, Celluclaseloading:7.5FPU/gglucan. Wangetal.BiotechnologyforBiofuels2013,6:9 Page7of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Commerciallignosulfonatetoenhanceenzymatic increasedbyapproximately75%comparedwithsaccharifi- saccharificationoflignocellulose cationofthewashedsolidsubstratealoneatpH4.8.These The commercial lignosulfonate (D748, LignoTech USA, resultsdemonstratethefeasibilityofeliminatingsolidsub- Rothschild, WI) was applied to enzymatic hydrolysis of stratewashingfordirectsimultaneousenzymaticsacchari- dilute acid-pretreated (DA-AS) and SPORL-pretreated fication and combined fermentation of enzymatic and aspen (SPH-AS), and lodgepole pine (SP-BD4) solid sub- pretreatment hydrolysates(SSCombF)usingSPORLtech- strates at lignosulfonate concentration of 10 g/L at an nologywithreducedwaterandenzymeuse. elevated buffer solution pH of 5.5. The SED at 72 h was increased by approximately 20 and 10% (Figure 6), re- Materials and methods spectively, for SPH-AS and SP-BD4, indicating the com- All experiments were carried out according to the ex- mercial lignosulfonate is also effective to increase perimentalprocessflowsimilartopreviousdescribedex- enzymatic hydrolysis of lignocelluloses. However, no sig- cept the pretreatment spent liquor was pH adjusted nificant effect was observed to the dilute acid pretreated before added to the solid fraction to conduct enzymatic aspen (DA-AS) though slight inhibitive effect to reduce saccharification[32]. SED before 24 h and the graduate recovery of SED after 24 was observed, suggesting the interaction of lignosul- Materials fonate with the bound lignin on the lignocellulose solids A lodgepole (Pinus contorta) tree killed by mountain alsodependsonthestructure ofthesolidlignin. pine beetle (Dendroctonus ponderosae) (estimated infest- ation age of 4 years, abbreviated BD4) was harvested Conclusions from the Canyon Lakes Ranger District of the Arapaho– Thisstudydemonstrateslignosulfonatehasa low affinity Roosevelt National Forest, Colorado. The details of the to cellulase. As a surfactant, certain lignosulfonate, such tree were described in our previous studies [31,33]. as lignosulfonate from SPORLpretreatment of lodgepole Fresh aspen (Populus tremuloides) wood logs were pine, can be applied to significantly enhance enzymatic obtained from northern Wisconsin, USA. All wood logs hydrolysis of lignocelluloses. Elevated pH can also en- were shipped to the U.S. Forest Products Laboratory, hance enzymatic saccharification of lignocelluloses. By Madison, Wisconsin, and chipped using a laboratory directly mixing SPORL pretreatment hydrolysate of a chipper. The wood chips were then screened to remove lodgepole pine with the corresponding SPORL pretreated all particles greater than 38 mm and less than 6 mm in lodgepolepinesolidsubstrate,90%ofenzymaticsacchari- length. The thickness of the accepted chips ranged from fication of pretreated lodgepole pine can be achieved at 1 to 5 mm. The chips were kept frozen at a temperature pH 5.5 and cellulase loading of 13 FPU/g glucan, or ofabout−16°Cuntilused. Celluclast 1.5 L and Novozyme 188 (β-glucosidase) were generously provided by Novozymes North America (Franklinton, NC). Sodium acetate buffer, sulfuric acid, and sodium bisulfite were ACS reagent grade and used asreceived from Sigma-Aldrich(St.Louis,MO). Substrateproduction Several lignocellulosic solid substrates were produced fromboththelodgepolepineandaspenwoodchipsusing pretreatment methods with different process chemistries. Table 1 lists the various substrates produced along with production process conditions. Because dilute acid is a widely studied pretreatment and effective on many feed- stocks except softwoods such as lodgepole pine used in this study. We also produced solid substrates from aspen (hardwood) using dilute acid and SPORL (Table 1) to evaluate the effectiveness of the application of lignosulfo- nate to enhance saccharfication of these aspen substrates. Figure6Effectoftheapplicationofacommercial A laboratory wood pulping digester of capacity of 23 L lignosulfonateontimedependentSEDsofpretreatedaspen wasusedtoconductpretreatmentasdescribedinourpre- (SPH-AS,DA-AS)andapretreatedlodgepolepine(SP-BD4). vious study [10]. Thedigesterwasheatedbya steamjacket Celluclaseloading7.5FPU/gglucanandhydrolysisbuffered androtatedat2rpmformixing.Theovendry(od)weightof atpH5.5. woodchipsineachpretreatmentwas2kg.Thepretreatment Wangetal.BiotechnologyforBiofuels2013,6:9 Page8of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Pretreatment hydrolysate (spent liquor) e at r st b u Pretreatment Size d s reduction oli s S al c mi Press e h C Water Steam Figure7Aschematicexperimentalflowdiagram. liquidtowoodratio(L/W)waskeptat3(v/w).The chemical hydrolysates were analyzed according to the methods charges, reaction temperature, and duration of different describedlaterinthetext.TheresultsarelistedinTables2 pretreatments are listed in Table 1. and3.Itshouldbepointedoutthattheexperimentalflow The pretreated wood chips remained intact and were showninFigure7istodemonstratetheconceptproposed separated from the pretreatment hydrolysate (hemicellu- inthisstudy.Thefreshwaterusedfordiskmillingshould losic sugar stream) by a screen. The pretreated wood be replaced by the pretreatment hydrolysate, i.e., the sep- chips were disk milled using disk plates of pattern D2B- aration of pretreatment hydrolysate from the solids is not 505 with a plate gap of 0.25 mm and adjusted to a re- necessary, as practiced in our laboratory currently after finer discharge consistency of 10% with dilution water theproposedconceptbeingproventhroughthisstudy. (Figure 7). The energy consumption for disk milling was The pH of the pretreatment hydrolysates were adjusted recordedasdescribedelsewhere[34,35].Thesize-reduced to 4.8 using NaOH before applied to the suspension of solids were directly dewatered to a solids content of ap- solidsubstratetoconduct enzymatichydrolysis. Forcom- proximately 30% by vacuum pressing in a canvas bag (as parisonpurposes,thepHofthepretreatmenthydrolysates prewashing). The yields of solid substrate after washing werealsoadjustedto4.8usingCa(OH) insomehydroly- 2 werethendeterminedfromtheweightandmoisture con- sis experiments as indicated. The glucose was measured tentofthecollectedsubstrate.Thechemicalcompositions using the glucose analyzer described in the “Analytical of both the solid substrates along with the pretreatment methods”section. A pure cellulosic substrate, a commercial Whatman filter paper (Grade 3, Cat No 1003 150, Whatman Inter- Table2Chemicalcompositionsandyieldsoftheuntreated national,England)wasalsoused.The manufacturer spe- andpretreatedlignocellulosicsubstrateslistedinTable1 cified ash content is 0.06%. High purity lignosulfonate Samplelabel Klignin Glucan Xylan Mannan Solidsyield (%) (%) (%) (%) (wt%) D748 from softwood sulfite pulping was donated by LignoTech USA (Rothschild, WI). Untreatedwood BD4 28.6 41.9 5.5 11.7 100 FL 29.2 39.1 6.0 10.0 100 Table3Glucoseandlignosulfonateconcentrationsinthe pretreatmenthydrolysates(hemicellulosicsugarstream) Aspen 20.2 45.6 16.4 1.4 100 listedinTable1 Pretreatedwood Liquidsubstrate Glucose(g/L) Lignosulfonateas SP-BD4 34.7 57.4 1.5 0.6 58.6 Klasonlignin(g/L) SP-AS 28.1 66.2 1.9 0.3 58.2 L-BD4-T85-1 2.95 12.6 SPH-AS 22.2 66.6 5.3 0.8 60.9 L-BD4-T85-3 5.34 11.1 DA-AS 30.0 61.6 3.3 0.4 63.5 L-BD4-T85-5 7.18 9.7 Wangetal.BiotechnologyforBiofuels2013,6:9 Page9of10 http://www.biotechnologyforbiofuels.com/content/6/1/9 Enzymatichydrolysis Competinginterests Enzymatic hydrolysis was conducted using commercial Theauthorsdeclarethattheyhavenocompetinginterests. enzymes at 2% substrate solids (w/v) in 50-mL of buffer Authors’contributions solutionsonashaker/incubator(ThermoFisherScientific, ZJWconductedmostoftheexperiments.TQLinitiatedandassisted Model4450,Waltham,MA)at50°Cand200rpm.Unless conductingthepHeffectexperiments.JYZdirectedtheresearchand indicated,the pH of the buffersolutionof sodium acetate analyzedthedataandwrotetheentirepaper.Allauthorsreadandapproved thefinalmanuscript. was4.8.Celluclast1.5Lloadingsvariedbetween7.5to13 FPU/gglucan.TheratioofNovozyme188(β-glucosidase) Authors’information loading (in CBU) to Celluclast 1.5 L loading (FPU) was WangandLanwerevisitingPh.DstudentsattheUSDAForestService maintained at 1.5 for all experiments. Selected hydrolysis (USDA-FS),ForestProductsLab(FPL),fromSouthChinaUniversityof experiments were carried out in duplicates to ensure ex- Technology,GuangZhou,China.WangiscurrentlywithKeyLabofPaper Science&Technology,ShandongPolytechnicUniversity,Jinan,China.Zhuis perimentalrepeatability.Hydrolysatewassampledperiod- aScientificTeamLeaderattheUSDA-FS-FPL.Heisaco-inventortheSPORL ically for glucose concentration analysis. Each data point pretreatmentprocessandpublishesextensivelyinwoodybiomass istheaverageoftworeplicates.Theaveragerelativestand- bioconversionforbiofuel,fiber,andnanocellulosicmaterials.Zhuisan electedFellowoftheInternationalAcademyofWoodScience(IAWS)andan arddeviationwasapproximately2%. officeroftheForestProductsDivisionoftheAmericanInstituteofChemical Engineering(AIChE)andtheTechnicalAssociationofthePulpandPaper Industry(TAPPI).Heservesontheeditorialboardsofseveraltechnical journals.ThisworkwasconductedonofficialgovernmenttimeofZhu. Analyticalmethods The chemical compositions of the original and pre- Acknowledgements treated biomass were analyzed by the Analytical and Mi- ThisworkwaspartiallysupportedbyaUSDASmallBusinessInnovative croscopy Laboratory of the Forest Products Laboratory Research(SBIR)PhaseIIproject(ContractNumber:2010-33610-21589)to BiopulpingInternational,Inc.Thisprojectprovidedpartialfinancialsupport as described previously [31]. All lignocelulosic samples toWangandfullsupporttoLanfortheirvisitingappointmentsattheUS were Wiley milled (model #2, Arthur Thomas Co, Phile- ForestService(USFS),ForestProductsLaboratory(FPL).TheChinese delphia, PA). The milled sample of 20 mesh (~1 mm) in ScholarshipCouncilprovidedtheotherpartialsupportforWang.We acknowledgeFredMatt(FPL)forcarryingoutcarbohydrateanalyses. size was hydrolyzed in two stages using sulfuric acid of 72% (v/v) at 30°C for 1 h and 3.6% (v/v) at 120°C for Authordetails 1 h. The hydrolysate supernatant and remaining solids 1KeyLabofPaperScience&Technology,ShandongPolytechnicUniversity, Jinan,China.2USForestService,ForestProductsLaboratory,Madison,WI, are then filtered through a Gooch Crucible lined with a USA.3CollegeofLightIndustryandFoodSciences,SouthChinaUniversityof 21mmWhatmanfilterintoavolumetricflask.Thesuper- Technology,Guangzhou,China. natant was used for carbohydrate analysis using high- Received:31July2012Accepted:13September2012 performanceanionexchangechromatographywithpulsed Published:28January2013 amperometricdetection(HPAEC-PAD)[36].Klasonlignin (acidinsoluble)retainedonthefilterpaperwasquantified References gravimetricallyafterdrying. 1. NakagameS,ChandraRP,SaddlerJN:Theinfluenceofligninonthe enzymatichydrolysisofpretreatedbiomasssubstrates.InSustainable The saccharides in the pretreatment hydrolysates ProductionofFuels,Chemicals,andFibersfromForestBiomass.EditedbyZhuJY, (spentliquors)wereanalyzedusingaDionexHPLCsystem ZhangX,PanXJ.Washington,DC:AmericanChemicalSociety;2011:145–167. (ICS-3000) equipped with integrated amperometric de- 2. MooneyCA,MansfieldSD,TouhyMG,SaddlerJN:Theeffectofinitialpore ™ volumeandlignincontentontheenzymatichydrolysisofsoftwoods. tector and Carbopac PA1 guard and analytical columns BioresourTechnol1998,64(2):113–119. at 20°C. Eluent was provided at a rate of 0.7 mL/min, 3. ErikssonT,BorjessonJ,TjerneldF:Mechanismofsurfactanteffect according to the following gradient: 0→25 min, 100% inenzymatic hydrolysis of lignocellulose.EnzymMicrobTechnol2002, water; 25.1→35 min, 30% water and 70% 0.1 M NaOH; 31(3):353–364. 4. MansfieldSD,MooneyC,SaddlerJN:Substrateandenzymecharacteristics 35.1→40 min, 100% water. To provide a stable baseline thatlimitcellulosehydrolysis.BiotechnolProg1999,15:804–816. and detector sensitivity, 0.5 M NaOH at a rate of 5. SewaltVJH,GlasserWG,BeaucheminKA:Ligninimpactonfiberdegradation.3. Reversalofinhibitionofenzymatichydrolysisbychemicalmodificationof 0.3 mL/min was used as post-column eluent. For fast ligninandbyadditives.JAgricFoodChem1997,45(5):1823–1828. analysis, glucose in the enzymatic hydrolysate was mea- 6. BerlinA,BalakshinM,GilkesN,KadlaJ,MaximenkoV,KuboS,SaddlerJ: sured in duplicate using a commercial glucose analyzer Inhibitionofcellulase,xylanaseandbeta-glucosidaseactivitiesby softwoodligninpreparations.JBiotechnol2006,125(2):198–209. (YSI 2700S, YSI Inc., Yellow Springs, OH). 7. PalonenH,TjerneldF,ZacchiG,TenkanenM:AdsorptionofTrichoderma The lignosulfonate in the SPORL hydrolysate was cal- reeseiCBHIandEGIIandtheircatalyticdomainsonsteampretreated culatedasKlasonligninbasedonKlasonligninyieldloss softwoodandisolatedlignin.JBiotechnol2004,107(1):65–72. 8. PanXJ,AratoC,GilkesN,GreggD,MabeeW,PyeK,XiaoZ,ZhangX, through SPORL-pretreatment with the assumption that SaddlerJ:Biorefiningofsoftwoodsusingethanolorganosolvpulping: all Klason lignin losses were solublized as lignosulfonate Preliminaryevaluationofprocessstreamsformanufactureoffuel-grade into the SPORL hydrolysate. The amount of lignosulfo- ethanolandco-products.BiotechnolBioeng2005,90(4):473–481. 9. IakovlevM,vanHeiningenA:EfficientfractionationofsprucebySO2- nate produced from the acid soluble lignin was not ethanol-watertreatment:Closedmassbalancesforcarbohydratesand accountedfor. sulfur.ChemSusChem2012,5(8):1625–1637.