h tt p : / / w w w . b j m i c r o b i o l . c o m . b r /
Industrial
Microbiology
Simultaneous
production
of
amylases
and
proteases
by
Bacillus
subtilis
in
brewery
wastes
Alina
Sánchez
Blanco
a,
Osmar
Palacios
Durive
a,
Sulema
Batista
Pérez
a,
Zoraida
Díaz
Montes
a,
Nelson
Pérez
Guerra
b,∗aDepartamentodeFundamentosQuímicosyBiológicos,FacultaddeIngenieríaQuímica,sedeMella,UniversidaddeOriente,Santiagode
Cuba,Cuba
bDepartamentodeQuímicaAnalíticayAlimentaria,FacultaddeCiencias,CampusdeOurense,UniversidaddeVigo,Spain
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:
Received1August2015 Accepted8January2016 Availableonline20April2016 AssociateEditor:AdalbertoPessoa Junior Keywords: Amylases Bacillussubtilis Brewerywastes Proteases
Responsesurfacemethodology
a
b
s
t
r
a
c
t
Thesimultaneousproductionofamylase(AA)andprotease(PA)activitybyBacillussubtilis UO-01inbrewerywasteswasstudiedbycombiningtheresponsesurfacemethodologywith thekineticstudyoftheprocess.Theoptimumconditions(T=36.0◦CandpH=6.8)forhigh biomassproduction(0.92g/L)weresimilartotheconditions(T=36.8◦CandpH=6.6)for highAAsynthesis(9.26EU/mL).However,themaximumPAlevel(9.77EU/mL)wasobtained atpH7.1and37.8◦C.Undertheseconditions,aconsiderablyhighreduction(between69.9
and77.8%)oftheinitialchemicaloxygendemandofthewastewasachieved.Inverification experimentsundertheoptimizedconditionsforproductionofeachenzyme,theAAand PAobtainedafter15hofincubationwere,respectively,9.35and9.87EU/mL.Byusingthe LuedekingandPiretmodel,bothenzymeswereclassifiedasgrowth-associatedmetabolites. Proteaseproductiondelayseemedtoberelated totheconsumptionofnon-proteinand proteinnitrogen.Theseresultsindicatethatthebrewerywastecouldbesuccessfullyused forahighscaleproductionofamylasesandproteasesatalowcost.
©2016SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.Thisis anopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/ licenses/by-nc-nd/4.0/).
Introduction
TheamylasesandproteasesproducedbytheBacillusspecies are of world-wide interest for their important industrial applications.1Therefore,productionoftheseenzymesshould
becarriedoutatalowproductioncostbyusingeconomically availableculturemedia(suchasfoodwastesoragro-industrial residues)andoptimizedfermentationconditions.2,3
∗ Correspondingauthor.
E-mail:nelsonpg@uvigo.es(N.PérezGuerra).
Differentstudieshaveshownthatproductionofamylases and proteases is affected by a variety of physicochemical factors, including the type and composition of the sub-strate, incubationtime andtemperature, pH, agitation and the concentration and type of the carbon and nitrogen sources.1,3However,insomecases,theeffectsofthese
vari-ables on enzyme production have been studied by using the “one variable ata time” method1,3,4 rather than using
responsesurfacemethodology(RSM).Thefirstmethodistime
http://dx.doi.org/10.1016/j.bjm.2016.04.019
1517-8382/©2016SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).
consuming and couldlead toanincomplete interpretation ofthebehaviorofthesystem,resultinginalackof predic-tiveability,mainlywhenthereareinteractionsbetweenthe independentfactors.5Incontrast,theuseofRSMallowsthe
reductioninthenumberofexperimentsandobtains empiri-calmathematicalmodelsdescribingtheeffect(bothlinearand quadratic)ofeachindependentfactorandtheirinteractions ontheresponsevariables.5
Consideringthesubstantialavailabilityofbrewerywastes (BWs)withahighchemicaloxygendemand(COD)atverylow pricesfromabreweryplantinSantiagodeCuba(Cuba),the useofthiseffluentasafermentationsubstratecouldofferan attractivealternativeforthelowcostproductionofamylases andproteasesbyBacillussubtilisUO-01.However,tothebest ofourknowledge,thereisnoinformationavailableontheuse ofbrewerywasteforproductionofamylasesandproteasesby
Bacillusstrains.
In the present study, we investigated the suitability of thiswastetosupportboththegrowthandenzyme
produc-tionbyB.subtilisUO-01atdifferentinitialvaluesofpHand
incubationtemperatures.Forthispurpose,thetimecourses ofthe synthesisofamylases, proteases,biomassand total sugars consumption were followed for 30h. Subsequently, theconcentrationsofenzymes andbiomassobtainedwere modeledbyusingthecorrespondinglogisticmodeltosmooth theexperimentaldataobtainedandreducetheexperimental error.Thedatapredictedbythelogisticmodelineachcaseat theappropriatefermentationtimewerethenusedfortheRSM analysistoobtaintheoptimumpHandtemperaturevaluesfor highproductionofamylases,proteasesandbiomass.In addi-tion,theenzymeproductionsystemwasstudiedunderthe optimumconditionstoverifytheeffectivenessandthe accu-racyoftheempiricalenzymemodelobtained.Aftertypifying boththeamylaseandproteaseproductionwiththeLuedeking andPiretmodel,6therelationshipbetweenprotease
produc-tion and consumption of total nitrogen and proteins was studiedatdifferentculturepH valuesandatthe optimum temperatureforgrowthofB.subtilisUO-01.
Materials
and
methods
Bacterialcultures
B.subtilisUO-01,theamylase-andprotease-producingstrain,
wasacquiredfromtheBiotechnologyCenteroftheUniversity ofOriente(SantiagodeCuba,Cuba).Stockcultureswere main-tainedat4◦Connutrientagarslants.Theworkingcultures werepreparedmonthlyfromfrozenstockculturesand main-tained at4◦C on nutrient agar (CultimedPanreac Química S.A.).
Culturemediumpreparation,inoculumandfermentation conditions
Brewerywastes(BWs),whichwereusedtopreparethe cul-turemedia,wereobtainedfromalocalbreweryinSantiagode Cuba,Cuba.Thewasteswerecentrifugedat12,000×g/15min toremovethesolidsinsuspension.Thesupernatantobtained fromthe BWs contained(g/L):COD, 3.40;totalsugars,1.98;
Table1–Experimentaldomainandcodificationofthe
independentvariables(T:temperatureandpH)usedin
theexperimentaldesign.
Codedvalues Actualvalues
T(◦C) pH –1.267 28.0 4.0 –1 30.0 4.5 0 37.8 6.2 1 45.5 8.0 1.267 47.6 8.5
reducingsugars,1.46;totalnitrogen,0.095;totalphosphorus,
0.034.
Inoculumwaspreparedbytransferring(withasterile
inoc-ulationloop)somecoloniesofa24-hold slantcultureinto
250mLErlenmeyerflaskscontaining50mLofsterilemedium
composed of(g/L):glucose,20; bacteriologicalpeptone,2.5;
KH2PO4·3H2O,1.5;Na2SO4,1.5;MgSO4·7H2O,0.15,FeSO4·7H2O,
0.03;MnCl2·4H2O,0.1;CaCl2·2H2O,0.45.Afterinoculation,the
mediumwasadjustedtopH6.8andsterilizedat121◦C/15min.
The inoculated culture was then incubated at 36◦C/12h
(200rpm).
The production BWs medium was supplemented with
the same ingredients as the mediumused to prepare the
inoculum,but inthiscase,soluble potatostarch(at a
con-centrationof10g/L)wasusedinsteadofglucose.Themedia
werebufferedatdifferentinitialpHvalueswiththe
appropri-atebuffer(0.1Mpotassiumhydrogenphthalate–HClbufferfor
pHvaluesof4.0,4.5and6.2or0.1Msodiumphosphatebuffer
forpHvaluesof8.0and8.5)accordingtotheexperimental
designdefinedinTable1andthensterilized(121◦C/15min).
Batchcultureswereperformedintriplicatein250mL Erlen-meyerflaskscontaining50mLofthecorrespondingbuffered medium.Eachflaskwasinoculatedwitha2%(v/v)inoculum level(withanabsorbanceof0.5at600nm)ofa12-hinoculum cultureandincubatedinanorbitalshaker(200rpm)for30hat thecorrespondingtemperatureaccordingtotheexperimental matrixdefinedinTable1.
To study the relationshipbetween nitrogenand protein consumptionandproteaseproductionbystrainUO-01at dif-ferent initialpHvalues,theBWsmediawerebufferedwith the same buffers used inthe former experimentto obtain initialpHvaluesof4,5,6,7,and 8.Themediawere incu-bated(200rpm)attheoptimumtemperatureforgrowthofthe enzyme-producingbacteriumfor21h.
Analyticalmethods
In each fermentation, threeflasks were collected each 3h, and triplicatesamples(runs)weretakenfromeach flaskto performanalyticaldeterminations.TheCODvalueswere mea-suredattheendofthecultureperiod(30h).
Growthwasmonitoredbyabsorbanceat600nmand con-vertedtocelldryweight(CDW)fromastandardcurve.Cells were harvested by centrifugation (12,000×g for 15min at 4◦C)ofculturesamplesandwashedtwicewithsaline(0.8% NaCl).Theculturesupernatantswereusedtomeasuretotal sugars (TS), total nitrogen (TN), proteins, enzyme produc-tion and COD. The methods for determining total sugars
(phenol–sulfuricacidmethod),nitrogen(micro-Kjeldahl, sub-stitutingdistillationfor aspectrophotometric method)and proteins (method of Lowry) were described in a previous work.7
Amylaseactivity(AA)ofthecell-freemediumwas deter-mined at 40◦C by appropriately mixing the diluted crude enzymeextract(80L)with400Lof0.15Mcitrate–phosphate buffer (pH 5.0) and 800L of 4% soluble starch previously maintained at 40◦C/15min. The reaction was stopped by immediatelyadding480Lof3,5-dinitrosalicylicacid(DNS). Thereducingsugars(glucoseequivalents)releasedfromthe enzymaticreactionweredetermineafter10minbyusingthe DNSmethod.Oneunitofamylaseactivity (enzymaticunits (EU)/mL) was defined as the amount of enzyme releasing 1mg/mLofreducingsugars/minundertheassayconditions.8
Proteaseactivity (PA)ofthecell-freemediumwas deter-mined as described by Tekin et al.9 The reaction mixture
with1mLof1%(w/v)casein in0.02MNaOH,2mLof0.4M phosphatebufferpH6,and1mLofcell-freemedium (suit-ablydiluted)wasincubatedat30◦C/10min.Thereactionwas stopped with 3mL of 10% trichloroacetic acid, mixed and after5min,centrifugedat12,000×gfor5min,then0.5mLof thesupernatantswereincubatedwith2.5mLof0.1MNaOH in2%(w/v)Na2CO3 for10min.Thereafter,0.25mLofFolin phenolicreagent(commercialsolutiondiluted1:1indistilled water)wasadded,mixedandheldfor30minatroom temper-ature.Theabsorbancemeasuredat750nmwasconvertedto mgoftyrosine/Lusingacalibrationcurve(mgtyrosine/Lvs. absorbance).Thetyrosinesolutions dissolvedin0.01MHCl weretreatedinthe samemannerasthecell-freemedium. Oneunitofprotease(EU/mL)wasdefinedastheamountof enzymethatproducedanabsorbanceat750nmequivalentto 1moloftyrosine/minundertheassayconditions.
CODanalyseswerecarriedoutintriplicateintheculture supernatants atthe end ofeach fermentation(30h)using the Standard Methods for the Examination of Water and Wastewater.10
Statisticalanalyses
Afterobtainingtheexperimentalconcentrationsofbiomass (CDW), AA and PA foreach sampling time, the data were smoothedbyusing,incaseofgrowthandPAproduction,the correspondingformofthelogisticmodel6andamodifiedform
ofthelogisticmodel,inthecaseofAA:
CDW(g/L)= CDWmax 1+e(c−b·t), beingc=ln
CDW max CDW0 − 1 (1) AA (EU/mL) = A0·e (−A1·t) 1+e(A2+A3·t+A4·t2)− A0 1+e(A2) (2) PA (EU/mL) = A0 1+e(A2+A3·t+A4·t2)− A0 1+e(A2) (3)whereCDWmaxandCDW0are,respectively,themaximumand theinitialcelldryweightconcentration(g/L),bisaconstant ofproportionality(h−1),tisthetime(h)andA0,A1,A2,A3and A4areconstantsinmodels(2)and(3).Thelattermodelswere alsoadjustedtodescribeAAproduction.However,thebestfits
wereobtainedwiththeuseofmodel(2)thatincludesaterm (e(−A1·t))thatconsiderstheAAdecayobservedfromthe18h
ofincubationuntilthe endoftheculture.Aftersmoothing thecellgrowthandenzyme(AAandPA)productioncurves, thebestsamplingtimetotakethedataforthecorresponding experimentaldesignwasselectedforstudyingthecombined effectofpHandtemperatureonthegrowthandenzyme syn-thesisbyB.subtilisUO-01intheBWsmedium.
Thecentralcomposite(orthogonal)designwasbasedon twolevelsandtwovariablesandconsistedof13experiments withfour(22)factorialpoints,fouraxialpointstoforma cen-tral composite designwith ˛=1.267 and five center points forreplication (Table1).The corresponding smootheddata obtainedat15hoffermentationwereusedineachcase.The results were analyzed by the Experimental Design Module of the Statistica software package (Statistica 12.0 for Win-dowscomputer programmanual; StatSoftInc., 2013,Tulsa, OK,USA).Thecoefficientsofthemodelswithpvalueslower than0.05wereconsideredstatisticallysignificant.TheFisher’s
F-test foranalysis of variance (ANOVA) was performed on
experimentaldatatoevaluatethe statisticalsignificanceof themodel.
ModelingtheenzymesproductionwiththeLuedekingand Piretmodel
BothAAandPAproductionweremodeledbyusingthe Luedek-ing and Piret model.6 Themodel parameters in each case
wereobtainedbyusingthenon-linearcurve-fittingsoftware ofSigmaPlotforWindowsversion12.0(SystatSoftware,Inc., 2011).Thecoefficientsofthemodelswithpvalueslowerthan 0.05wereconsideredstatisticallysignificant.
Results
and
discussion
Kineticsofgrowth,enzymeproductionandtotalsugars consumptionbyB.subtilisUO-01inBWsmedia
Thekineticsofgrowth(CDW),productionofamylase(AA)and protease(PA)activityandtotalsugars(TS)consumptionbyB.
subtilisUO-01inBWsmediawasfollowedinbatchculturesat
differentinitialpHvaluesandincubationtemperatures(Fig.1) accordingtotheexperimentaldesigndefinedinTable1.
Accordingtothepredictionsofmodel(1),biomass produc-tionincreaseduntil15hofincubationandremainedconstant afterwards inall the culturesalthough,in some fermenta-tions, the total sugars were not totally consumed (Fig. 1). However,theevolutionofenzyme(AAandPA)synthesisand biomassproductionshowedsomedifferences.Thus,AAlevels increaseduntilreachingthemaximumconcentrationat15h ofculture,thentheAAconcentrationwasmaintainedalmost constantuntil18handdecreasedslightlyafterwards(Fig.1), althoughnotalwayssignificantly,becausethevaluesobtained forthecoefficientA1inmodel(2)werenotalwayssignificant. Becausethepresenceofstarchinducestheproductionof ␣-amylases,2duringthefirst15hoffermentation,AAwas
syn-thesizedbyB.subtilisUO-01tohydrolyzethestarchpresent inthemediumandproducemoreeasilyassimilablesugars (glucoseandmaltose)forthegrowingstrain.1
12 1 5 9 10 6 7 11 8 12 13 2 3 4 0.6 0.4 0.2 0 0.4 0.3 0.2 0.1 0 0.8 0.6 0.2 0.4 0 0.8 0.6 0.2 0.4 0 0.6 0.4 0.2 0 0 10 20
Time (h) Time (h) Time (h)
Time (h) Time (h) Time (h)
Time (h) Time (h) Time (h)
Time (h) Time (h)
CDW: Cell dry weight (g/L),
TS: Total sugars (g/L) AA: Amylase activity (EU/mL), PA: Protease activity (EU/mL),
Time (h) Time (h) TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W TS , AA, P A CD W CD W TS , AA, P A 30 T=45.5 ºC, pH=8.0 T=47.6 ºC, pH=6.2 T=28.0 ºC, pH=6.2 T=37.8 ºC, pH=8.5 T=37.8 ºC, pH=4.0 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=45.5 ºC, pH=4.5 T=30.0 ºC, pH=8.0 T=30.0 ºC, pH=4.5 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 0 0.3 4 0.6 8 0.9 12 1.2 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 0 0.2 0.4 0.6 0.8 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 16 0 4 8 12 16 0 4 8 12 16 0 4 8 12 16 0 4 8 12 16 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 9 6 3 0 12 0.12 9 0.09 6 0.06 0.03 0 3 0 12 9 6 3 0 12 9 6 3 0 12 9 6 3 0 0.3 0.2 0.1 0 12 9 6 3 0 12 9 6 3 0 12 9 6 3 0
Fig.1–Kineticofproductionofcelldryweight,amylaseandproteaseactivity,andconsumptionoftotalsugarsbyB.subtilis
UO-01attheinitialpHandTvalues,asdefinedinTable1.Eachdatapointrepresentsthemeanofthreeindependent cultures(thestandarderrorswerelessthan5%ofthemeans).
Table2–Concentrationsofbiomass(CDWing/L),amylase(AAinEU/mL)andprotease(PAinEU/mL)activityobtainedat
15hofincubationandchemicaloxygendemand(COD)reductionpercentages(%)attheendofthecultures(30h).No.:
numberoftheexperiment.
No. Codedvalues Actualvalues Responsevariablevalues
T(◦C) pH T(◦C) pH CDW AA PA CODreduction 1 1 1 45.5 8.0 0.49 3.97 7.16 40.4 2 1 –1 45.5 4.5 0.10 1.73 2.72 8.1 3 –1 1 30.0 8.0 0.68 5.30 7.50 56.3 4 –1 –1 30.0 4.5 0.37 3.20 3.06 30.1 5 1.267 0 47.6 6.2 0.35 3.08 5.42 28.8 6 –1.267 0 28.0 6.2 0.63 4.94 5.91 51.3 7 0 1.267 37.8 8.5 0.68 5.20 8.26 56.7 8 0 –1.267 37.8 4.0 0.25 2.25 2.64 19.9 9 0 0 37.8 6.2 0.91 9.40 8.84 74.4 10 0 0 37.8 6.2 0.89 9.27 8.81 73.1 11 0 0 37.8 6.2 0.93 9.40 9.48 77.8 12 0 0 37.8 6.2 0.81 8.75 9.63 69.9 13 0 0 37.8 6.2 0.87 9.10 9.48 72.1
TheobserveddeclineinAAsynthesisafter18hof
incuba-tioncouldberelatedtoapossibleproteolyticdegradationof
theamylasesbytheactionoftheextracellularproteases
pro-ducedbytheB.subtilisstrain.1,2 However,otherresearchers
attributedthisAAdecreasetoapossibledenaturationand/or decompositionoftheenzymeduetointeractionswithother compoundsinthefermentedmedium.3
Proteaseproductionthat reachedthe maximumPA lev-elsat15hofincubation (Fig.1)wasonlydetectedafter9h offermentationinall cultures, whenthe cells enteredthe post-exponentialphase,asitwasobservedbeforeforother
Bacillus species.11 Thisbehavior hasbeen attributed tothe
needforobtainingnutrientsforpost-exponentialsurvival11
ortoanincreasedneedforturnoverofcell proteinsatthe slowergrowthrate.1
Becauseallthemediawerebufferedatthecorresponding initialpHvalues,thecessationofgrowthfrom15hof fermen-tationcouldberelatedtothefollowingreasons:
1. IntheculturesatinitialpH≥6.2,thecarbonsourcereached levels considerablylower(<3g/L),exceptforthe cultures atinitialpH8.0and T=45.5◦C and atinitialpH6.2and T=47.6◦C(Fig.1).TheselowTSlevelscouldleadtoalow availabilityofthecarbonsource12thatlimitsthegrowthof
B.subtilisUO-01.
2. Theexhaustioninall theculturesofsomenutrient(the sources of nitrogen and/or phosphorus) or micronutri-ent(vitamins,mineralsaminoacidsorcations)essential forthegrowthoftheenzyme-producingstrain,asitwas observedbeforeforotherbacteria.12,13
Theremoval percentagesofinitial CODload in the fer-mented media are shown in Table 2. The higher COD reductions(69.9–77.8%)wereobtainedinthecultures charac-terizedbythehighestbiomassproductioninthecenterpoints oftheexperimentaldesign,suggestingthatB.subtilisUO-01 wasabletoutilizetheorganicmatterpresentinthebrewery wastewaterasasourceofnutrients.Thus,bacterialgrowth andenzymeproductionbyB.subtilisUO-01couldbeusedas analternative forremovingtheinitialCODload present in brewerywastes.
KineticclassificationoftheAAandPAproduction
Totypifytheproductionofthetwoenzymes(E),theclassical LuedekingandPiret(LP)model6wasused:
dE dX=˛·
dX
dt +ˇ·X,ordividingbythebiomass (4)
qE=(˛·+ˇ) (5)
whereXisthebiomassconcentration(gofCDW/L),tisthe time(h),qEisthespecificenzymeproductionrate(EU/mg/h), isthe specificgrowth rate(h−1), ˛isagrowth-associated constant forenzymeproduction(EU/mg),and ˇisthe non-growth-associatedconstant(EU/mg/h).
TheCDW,AAandPAvalues(symbolsinFig.2)predicted bymodels(1),(2)and(3)(solidlinesinFig.1)wereusedto calculatethevaluesoftheparametersoftheLPmodel(5).
The trajectories for the productiondynamic of the two enzymes(solidlinesinFig.2)werecalculatedbynumerical integrationoftheqEvaluesobtainedwiththeLPmodel(5)for AAandPAsynthesisinthecorrespondingbatchculturesofB.
subtilisUO-01.
Forbothenzymes,significantvalues(p<0.05)for˛and non-significantvalues(p>0.05)forˇwerealwaysobtained(Fig.2). Thus,AAandPAcanbecharacterizedasgrowth-associated metabolites(˛ /= 0,ˇ=0).However,AAproduction(R2
AA
val-uesbetween0.985and0.999)wasmoresatisfactorilydescribed bytheLPmodelthanPAsynthesis(R2
PAvaluesbetween0.876
and0.969),mainlybecauseproductionofthelatterenzyme startedafter9hofincubation,whilethegrowthandAA pro-ductionstartedfromthebeginningoftheculture(Fig.2).
EffectofinitialpHandincubationtemperatureonthe growthandproductionofamylasesandproteases
TheresultsontheeffectofinitialpHandtemperatureon pro-ductionofbiomass(CDW),AAandPAbyB.subtilisUO-01in BWsmediumafter15hofincubationareshowninTable2.
T=45.5 ºC, pH=8.0
AA, P
A
Time (h) Time (h) Time (h) Time (h)
Time (h) Time (h) Time (h)
Time (h) Time (h)
Time (h) AA: Amylase activity (EU/mL), PA: Protease activity (EU/mL)
Time (h) Time (h) Time (h) AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A AA, P A T=47.6 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=28.0 ºC, pH=6.2 T=45.5 ºC, pH=4.5 T=30.0 ºC, pH=8.0 T=37.8 ºC, pH=8.5 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=6.2 T=37.8 ºC, pH=4.0 T=30.0 ºC, pH=4.5 9 1 2 6 5 9 10 11 12 13 3 7 4 8 αPA=13.5±0.3 αPA=22.8±0.7 αAA=9.1±0.2 αAA=22.7±1.8 R2 PA=0.946 R2PA=0.876 αPA=11.3±0.8 R2 PA=0.969 αPA=6.8±0.1 R2 PA=0.932 R2 AA=0.999 αPA=14.6±1.0 αPA=8.4±0.5 αAA=9.5±0.8 R2 PA=0.945 R2 PA=0.950 R2 AA=0.997 αAA=8.6±0.7 R2 AA=0.998 αPA=10.5±0.9 R2 PA=0.936 αPA=8.7±0.8 R2 PA=0.943 αAA=8.6±0.7 R2 AA=0.998 αAA=10.9±0.3 R2 AA=0.999 αPA=9.9±0.8 R2 PA=0.932 αAA=11.8±0.9 R2 AA=0.992 αPA=8.6±1.2 R2 PA=0.939 αAA=11.4±1.4 R2 AA=0.992 αPA=9.6±1.2 R2 PA=0.944 αAA=11.9±1.3 R2 AA=0.987 αPA=8.7±1.1 R2 PA=0.939 αAA=11.5±1.2 R2 AA=0.989 αPA=8.3±1.0 R2 PA=0.939 αAA=11.5±0.8 R2 AA=0.993 R2 AA=0.993 αAA=9.4±0.7 R2 AA=0.985 αAA=9.3±0.7 R2 AA=0.992 6 3 0 9 6 3 0 3 2 1 0 6 4 2 0 12 9 6 3 0 12 9 6 3 0 12 9 6 3 0 12 9 6 3 0 12 9 6 3 0 12 9 6 3 0 3.5 2.8 2.1 1.4 0.7 0 10 8 6 4 2 0 5 4 3 2 1 0 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40
Fig.2–ModelingtheproductionofamylaseandproteaseactivitiesbyB.subtilisUO-01inBWsmediaatdifferent temperaturesandpHvalues(experiments1–13,Table1).Thesymbolscorrespondtotheamylaseandproteaselevels obtained,respectively,withmodels(2)and(3).ThesolidlinesdrawnthroughthesymbolswereobtainedaccordingtotheLP model(5).Theparameters˛AAand˛PAarethegrowthassociatedconstants(EU/mg)forAAandPAproductionsintheLP
model,andR2
AAandR2
PAarethecorrespondingcorrelationcoefficients.
The empirical models obtained with their significant coefficientsare:
CDW=0.88−0.11·T+0.17·pH−0.23·T2−0.25·pH2 (6) AA=9.11−0.72·T+1.12·pH−2.88·T2−3.06·pH2 (7) PA=9.21+2.22·pH−2.07·T2−2.21·pH2 (8) Becausethepvalueswerehigherthan˛(˛=0.050)andthe Fvalueswerelowerthan6.390forthethreemodels,thereis astatisticallysignificant relationshipbetweenthe response andthe independentvariablesatthe 95%confidencelevel. Thehighervaluesoftheadjusteddeterminationcoefficient (adjR2=0.989,0.983and0.989)obtainedforthesemodelsalso
indicatedtheirhighsignificanceandexcellentadequacyfor theexperimentaldata.
Theresponsesurfacesgeneratedwiththeempirical mod-els(6),(7)and(8)areshowninFig.3A–C,respectively.
Thehighestbiomassconcentration(0.92g/L)predictedby model (6) was obtained at T=36.0◦C and pH=6.8. These optimum T and pH values were almost similar to those (T=36.8◦C andpH=6.6) obtainedforhigh amylase produc-tion (9.26EU/mL). The same behavior was observed before
foraB.subtilisstrainisolatedfromsoilsamples3 andthree
mesophilicBacillus(Bacillussp.,B.subtilisandB.
amyloliquefa-ciens)isolates.14
However, comparisons of the optimum amylase level producedbyB.subtilisUO-01withtheoptimumenzyme pro-ductions for other Bacillus strains is very difficult due to the differentmethodologiesusedtodeterminetheenzyme
CD W (g/L)
A
B
C
AA (EU/mL) pH pH Temp (ºC) Temp (ºC) Temp (ºC) pH P A (EU/mL) 1.0 47.6 37.8 28.0 6.0 6.9 7.7 47.6 37.8 28.0 6.0 6.9 7.7 12 9 6 3 0 0.8 0.6 0.4 0.2 0 1.0 47.6 37.8 28.0 6.0 6.9 7.7 8 6 4 2 0CDW: Cell dry weight AA: Amylase activity, PA: Protease activity
Fig.3–ResponsesurfacesshowingtheeffectoftemperatureandpHontheproductionofcelldryweight,amylaseand proteaseactivitiesbyB.subtilisUO-01inBWsmedia.
activity,thedifferentformstoexpresstheenzyme concentra-tionandthedifferentformstodefinetheenzymeactivityunit. Forexample,insomecases,theenzymeactivitywas deter-minedbyquantifyingtheamountsofreducingsugarsreleased fromthehydrolysisofstarchbyincubatingthereaction mix-turesattemperaturesof40◦C,150◦C14or60◦C15for10min.
Inthesecases,theenzymeactivityunitwasdefinedasthe amountofenzymereleasing1molofglucose14,15ormaltose1
equivalentsfromthesubstrateperminundertheassay condi-tions,andthereactionmixtureshaddifferentcompositions.
Withregardtoprotease production,the maximumlevel (9.77EU/mL)predictedbymodel(8)wasobtainedatpH=7.1 and T=37.8◦C. Similar results were obtained forB. subtilis strainRand,whichproducedthemaximumrelativeprotease activity(%)atpH=7andT=37◦C.16
Asindicatedabove,comparisonoftheoptimumprotease levelproduced byB. subtilisstrainUO-01 withtheprotease levelsfoundbyotherresearchersisverydifficultasa con-sequence ofthe different forms ofexpressing the enzyme concentrationsordefiningthe unitofproteaseactivity.For example, protease levels have been expressed as relative productioninU/log10CFU17orasrelativeactivity(%).16In addi-tion,theunitofproteaseactivitywasdefinedastheamount ofenzymerequiredto:(i)produceanincreaseof0.1in opti-caldensityat700nm,1(ii)release1goftyrosinepermLper
minute,18or(iii)toobtainanincreaseof0.001absorbanceunit
at450nmperminute.16
The increase in temperature over its optimum value inhibitedproduct(CDW,AAandPA) formationbyB.subtilis strainUO-01,probablybysuppressionofcellviability19 and
enzymaticinactivation.20Incontrast,lowtemperaturevalues
mayslowdownthemetabolismofthemicroorganism20and
consequently,enzymesynthesis,becausethelatterproducts weresynthesizedasgrowth-associated(primary)metabolites, asitwasdemonstratedwiththeLPmodel(Fig.2).
The decrease in growth and enzyme (AA and PA) pro-duction observed at pH levels lower and higher than the optimumvaluecouldberelatedtoareductioninthemetabolic activityofB.subtilisUO-01,3probablycausedbyalimitation
in micronutrients or nutrient transport.13,21 Inactivation or
instabilityoftheenzymescouldbeotherreasonstoexplain thedecreaseobservedinenzymeproduction.22
Verificationofthepredictedresultsintheoptimal conditionsforenzymeproduction
Triplicateincubationexperimentswerecarriedoutunderthe optimizedconditionsforAAandPAproductionstoverifythe resultspredictedbymodels(7)and(8).Theexperimentaldata forgrowthandproductionofAAandPAinbothcaseswere smoothedbyusingthemodels(1),(2)and(3)(Fig.4).Underthe optimalconditionsforAAproduction(T=36.8◦CandpH=6.6), thepredictedvaluesofbiomass,AAandPAafter15hof incu-bationwere,respectively,0.94g/L,9.35EU/mLand9.34EU/mL
Optimum conditions for AA T=36.8 ºC, pH=6.6
CDW: Cell dry weight (g/L), PA: Protease activity (EU/mL),
AA: Amylase activity (EU/mL), TS: Total sugars (g/L) Time (h) 12 9 6 3 0 1.2 0.9 0.6 0.3 0 0 10 20 30 40 0 10 20 30 40 12 9 6 3 0 1.2 0.9 0.6 0.3 0 Time (h) TS , AA, P A TS , AA, P A CD W CD W
Optimum conditions for PA T=37.8 ºC, pH=7.1
Fig.4–Kineticsoftheproductionofcelldryweight,amylaseandproteaseactivityandconsumptionoftotalsugarsbyB. subtilisUO-01undertheoptimumconditionsforamylaseandproteaseproductions,respectively.Eachdatapointrepresents themeanofthreeindependentassays(thestandarderrorswerelessthan5%ofthemeans).
(leftpartofFig.4).Inaddition,undertheoptimumAA con-ditions,theconcentrations ofbiomassand PApredicted by theempiricalmodels(6)and(8)were0.91g/Land9.51EU/mL, respectively.
Inthesameway,undertheoptimumconditionsforPA pro-duction(T=37.8◦CandpH=7.1),themeanconcentrationsof biomass,AAandPAafter15hofincubationpredictedbythe models(1),(2)and(3)were,respectively,0.91g/L,8.97EU/mL and9.87EU/mL(rightpartofFig.4).UndertheoptimumPA
conditions,theconcentrationsofbiomassandAApredictedby theempiricalmodels(6)and(7)were0.90g/Land8.90EU/mL, respectively.Therefore,themodels(6),(7)and(8)canbe con-sideredtobeaccurateandreliableforpredictingthegrowth andproductionofamylasesandproteasesbyB.subtilisUO-01 asafunctionofinitialpHandincubationtemperatureinBWs media.
Asobservedinthepreviouscultures(Fig.1),protease pro-ductionstartedafter9hofincubationintheculturesatthe
CDW: Cell dry weight (g/L), AA: Amylase activity (EU/mL), PA: Protease activity (EU/mL), TS: Total sugars g/L)
1.2 12 9 6 3 0 12 9 6 3 0 2 0.4 1 0.3 0.2 0 0.9 0.6 0 9 6 3 0 0 5 10 15 Time (h) Time (h) Prot (g/L) AA (EU/mL) CD W (g/L) P A (EU/mL) TN (g/L) TS (g/L) Time (h) 20 25 0 5 10 15 20 25 0 5 10 15 20 25 0.3 pH=4 ( ), pH=5 ( ), pH=6 ( ), pH=7 ( ), pH=8 ( )
Fig.5–Timecourseofcelldryweight,amylaseandproteaseactivityandconsumptionoftotalsugars,proteinsandtotal nitrogenatdifferentinitialpHvaluesandatafixedtemperatureof36◦C.
optimalconditionsforAAandTAproduction(Fig.4). Inan attempttoexplaintheobserveddelayinproteaseproduction, thekineticsofgrowthandenzymeproductionwerefollowed inparallelwiththeevolutionofthecorrespondinginductor nutrient(totalsugarsandproteins)foreachenzyme.
Becauseboththenutrientconsumptionandproduct for-mation by different bacteria (Lactococcus lactis, L. cremoris,
LactobacillusrhamnosusandLact.casei)dependedontheculture
pH,13,21aseriesofcultureswithstrainUO-01inBWsmedia
adjustedtoinitialpHvaluesof4,5,6,7and8wasstudied. Thecultureswereincubatedat36◦C(theoptimum temper-atureforbiomassproduction)for21h.Consideringthatthe presenceofproteinandnon-proteinnitrogenintheculture mediacanaffectproteinconsumption,7theevolutionoftotal
nitrogenwasalsofollowedinthefermentations.
As observed in Fig. 5, production of biomass and AA increasedinparallelwiththedecreasesinTSandTN.In addi-tion,adirectrelationshipbetweenthetimecoursesofproteins andPAproductionwasobserved.Inthelattercase,proteins decreasedat9hoffermentationwhenproteaseproduction wasdetectedintheculturemedium.However,totalnitrogen wasslightlyconsumedduringthefirst6hofincubation,but afterthistime,theconcentrationoftotalnitrogendecreased rapidly,coincidingwiththestartoftheintenseprotein con-sumption.
IfthetimecoursesofproteinsandTNarejointlyanalyzed, adirectrelationshipbetweenbothvariablesonlyexistsafter 9hofincubation.Accordingtothemanufacturers(Pronadisa, LaboratoriosConda,SA),thebacteriologicalpeptoneusedas thenitrogensourceintheBWsmediahasatotalnitrogenand anaminonitrogencontent of15.5%and 3%(w/w), respec-tively. Thus, B. subtilis UO-01 seemed to consumefirst the non-proteinnitrogenfraction (presumablyeasier to assim-ilate) and then thenitrogen present inthe proteinsofthe culturemedia (Fig. 5), as it was observed before for other bacteria.7,23,24
These observations suggest that the delay in protease productioncould be closely relatedto the consumption of non-proteinandproteinnitrogenbyB.subtilisUO-01(Fig.5). However,theconsumptionofbothtotalnitrogenandproteins dependedon theculturepH (Fig. 5)asobservedbefore for nutrient(sugars,aminoacidsandphosphorus)transportin differentstrainsoflacticacidbacteria.13,21
Conclusions
Theresultsobtainedinthisworkshowedthatbrewerywastes couldbeusedforenzyme(amylasesandproteases)
produc-tionbyB.subtilisUO-01,withtheadditionaladvantageofan
effectiveandcost-effectivereductionoftheinitialCOD.Then, themethodologicalproceduredescribedinthisworkandthe resultsobtained could beused as areference for the pro-ductionofotherimportantmicrobialmetabolitesinbrewery wastesandinothereffluentsfromthefoodindustry.
Conflicts
of
interest
Theauthorsdeclarenoconflictsofinterest.
r
e
f
e
r
e
n
c
e
s
1.MukhtarH,HaqIU.Concomitantproductionoftwo proteasesandalpha-amylasebyanovelstrainofBacillus subtilisinamicroprocessorcontrolledbioreactor.BrazJ Microbiol.2012;43:1072–1079.
2.JamrathT,LindnerC,Popovi ´cMK,BajpaiR.Productionof amylasesandproteasesbyBacilluscaldolyticusfromfood industrywastes.FoodTechnolBiotechnol.2012;50:355–361. 3.UnakalC,KallurRI,KaliwalBB.Productionof␣-amylase usingbananawastebyBacillussubtilisundersolidstate fermentation.EurJExpBiol.2012;2:1044–1052.
4.RaulD,BiswasT,MukhopadhyayS,DasSK,GuptaS. Productionandpartialpurificationofalphaamylasefrom Bacillussubtilis(MTCC121)usingsolidstatefermentation. BiochemResInt.2014:1–5.
5.AgregánR,AlonsoE,TorradoA,GuerraNP.Areviewonsome importantfactorsaffectingbacteriocinproductionby lactococci,lactobacilliandpediococci.CurrBiochemEng. 2014;1:9–24.
6.GuerraNP.Modelingthebatchbacteriocinproduction systembylacticacidbacteriabyusingmodified
three-dimensionalLotka-Volterraequations.BiochemEngJ. 2014;88:115–130.
7.GuerraNP,TorradoA,LópezC,FajardoP,PastranaL. Dynamicmathematicalmodelstodescribethegrowthand nisinproductionbyLactococcuslactissubsp.lactisCECT539in bothbatchandre-alkalizedfed-batchcultures.JFoodEng. 2007;82:103–113.
8.TorradoA,VázquezJA,PrietoMA,etal.Amylaseproduction byAspergillusoryzaeinasolid-statebioreactorwith
fed-batchoperationusingmusselprocessingwastewatersas feedingmedium.JChemTechnolBiotechnol.2013;88:226–236. 9.TekinN,CihanAC¸,Takac¸ZS,TüzünCY,Tunc¸K,C¸ökmüs¸C. AlkalineproteaseproductionofBacilluscohniiAPT5.TurkJ Biol.2012;36:430–440.
10.APHA.StandardMethodsfortheExaminationofWaterand Wastewater.21steditionWashington,DC:AmericanPublic HealthAssociation;2005.
11.AronsonAI,BellC,FulrothB.Plasmid-encodedregulatorof extracellularproteasesinBacillusanthracis.JBacteriol. 2005;187:3133–3138.
12.VanNielEWJ,Hahn-HägerdalB.Nutrientrequirementsof lactococciindefinedgrowthmedia.ApplMicrobiolBiotechnol. 1999;52:617–627.
13.FajardoP,RodríguezI,PastranaL,GuerraNP.Productionofa potentiallyprobioticcultureofLactobacilluscaseisubsp.casei CECT4043inwhey.IntDairyJ.2008;18:1057–1065.
14.NusratA,RahmanSR.Comparativestudiesonthe productionofextracellular␣-amylasebythreemesophilic Bacillusisolates.BanglJMicrobiol.2007;24:129–132.
15.AmuthaK,PriyaKJ.EffectofpH,temperatureandmetalions onamylaseactivityfromBacillussubtilisKCX006.IntJPharma BioSci.2011;2.B-407-B-413.
16.AbushamRA,RahmanRNZRA,SallehAB,BasriM.
Optimizationofphysicalfactorsaffectingtheproductionof thermo-stableorganicsolvent-tolerantproteasefroma newlyisolatedhalotolerantBacillussubtilisstrainRand. MicrobCellFact.2009;8:20.
17.ShivanandP,JayaramanG.Productionofextracellular proteasefromhalotolerantbacterium,Bacillusaquimaris strainVITP4isolatedfromKumtacoast.ProcBiochem. 2009;44:1088–1094.
18.OlajuyigbeFM.Optimizedproductionandpropertiesof thermostablealkalineproteasefromBacillussubtilisSHS-04 grownongroundnut(Arachishypogaea)meal.AdvEnzRes. 2013;1:112–120.
19.FrancisF,SabuA,NampoothiriKM,SzakacsG,PandeyA. Synthesisofa-amylasebyAspergillusoryzaeinsolid-state fermentation.JBasicMicrobiol.2002;5:320–326.
20.MazuttiM,CeniG,DiLuccioM,TreichelH.Productionof inulinasebysolid-statefermentation:effectofprocess parametersonproductionandpreliminarycharacterization ofenzymepreparations.BioprocessBiosystEng.
2007;30:297–304.
21.PanesarPS,KennedyJF,KnillCJ,KossevaM.Productionof L(+)lacticacidusingLactobacilluscaseifromwhey.BrazArch BiolTechn.2010;53:219–226.
22.AqelH.EffectsofpH-values,temperatures,sodiumchloride, metalions,sugarsandTweensontheacidphosphatase activitybythermophilicBacillusstrains.EurJSciRes. 2012;75:262–268.
23.LetortC,NardiM,GaraultP,MonnetV,JuillardV.Casein utilizationbyStreptococcusthermophilusresultsinadiauxic growthinmilk.ApplEnvironMicrobiol.2002;68:3162–3165. 24.GuerraNP,FajardoP,PastranaL.Modellingthestress
inducingbiphasicgrowthandpediocinproductionby PediococcusacidilacticiNRRLB-5627inre-alkalizedfed-batch cultures.BiochemEngJ.2008;40:465–472.