Effects of Agitation, Aeration and Temperature on Production of a Novel Glycoprotein GP-1 by

14 Pages·2017·1.33 MB·English

molecules
Article
Effects of Agitation, Aeration and Temperature on
Production of a Novel Glycoprotein GP-1 by
Streptomyces kanasenisi ZX01 and Scale-Up Based on
Volumetric Oxygen Transfer Coefficient
YongZhou1,†,Li-RongHan1,2,†,Hong-WeiHe1,BuSang3,Dai-LinYu3,Jun-TaoFeng1,2and
XingZhang1,2,*
1 ResearchandDevelopmentCenterofBiorationalPesticides,NorthwestAgriculture&ForestryUniversity,
Yangling712100,Shaanxi,China;[email protected](Y.Z.);[email protected](L.-R.H.);
[email protected](H.-W.H.);[email protected](J.-T.F.)
2 ShaanxiResearchCenterofBiopesticideEngineering&Technology,Yangling712100,Shaanxi,China
3 AgricultureResearchInstitute,TibetAcademyofAgriculturalandAnimalHusbandryScience,
Lhasa850032,Tibet,China;[email protected](B.S.);[email protected](D.-L.Y.)
* Correspondence:[email protected],Tel.:[phone]
† Theseauthorscontributedequallytothiswork.
Received:8December2017;Accepted:5January2018;Published:11January2018
Abstract: The effects of temperature, agitation and aeration on glycoprotein GP-1 production
by Streptomyces kanasenisi ZX01 in bench-scale fermentors were systematically investigated. The
maximum final GP-1 production was achieved at an agitation speed of 200 rpm, aeration rate
of 2.0 vvm and temperature of 30 ◦C. By using a dynamic gassing out method, the effects of
agitation and aeration on volumetric oxygen transfer coefficient (k a) were also studied. The
L
valuesofvolumetricoxygentransfercoefficientinthelogarithmicphaseincreasedwithincreaseof
agitationspeed(from14.53to32.82h−1)andaerationrate(from13.21to22.43h−1). Inaddition,
asuccessfulscale-upfrombench-scaletopilot-scalewasperformedbasedonvolumetricoxygen
transfercoefficient,resultinginfinalGP-1productionof3.92,4.03,3.82and4.20mg/Lin5L,15L,
70Land500Lfermentors,respectively. Theseresultsindicatedthatconstantvolumetricoxygen
transfercoefficientwasappropriateforthescale-upofbatchfermentationofglycoproteinGP-1by
StreptomyceskanasenisiZX01,andthisscale-upstrategysuccessfullyachieved100-foldscale-upfrom
bench-scaletopilot-scalefermentor.
Keywords: scale-up;glycoprotein;Streptomyces;anti-TMV;volumetricoxygentransfercoefficient
1. Introduction
As an important part of natural products, microbial metabolites exhibit high diversity in
producing species, functions, bioactivities and chemical structures, which have gradually become
ascientificresearchhotspotinrecentyears[1]. Streptomycesisacommonandwellknownspecies
of microorganism, and metabolites isolated from Streptomyces account for nearly half of the total
numberofmicrobialcompounds[1–3]. Antibioticsandsimilarsmallmoleculecompoundsconstitute
an important group of metabolites produced by Streptomyces and usually show various biological
activities,representingthemaintopicofstudiesandliteratureallthetime[4,5].Withtherapidprogress
ofmethodsandtechnologiesforscreeningandisolation,thenumberofnovelmicrobialmetabolites
continuestogrow,butanumberofbiologicalmacromolecules,suchaspolysaccharides,polypeptides
andglycoproteins,werealsosimultaneouslydiscovered[6–10].
Molecules2018,23,125;doi:10.3390/molecules23010125 www.mdpi.com/journal/molecules
Molecules2018,23,125 2of14
It is well known that microbial growth and their metabolite production in bioreactors are
greatly influenced by the medium components and physical factors, such as agitation, aeration,
temperature, fermentation time and dissolved oxygen (DO) [11]. Both agitation and aeration are
important parameters for all aerobic processes, and have a significant effect on the production of
mostbiopolymers,includingxanthan[12],gellan[13],pullulan[14],curdlan[15],andsoon[16–18].
Agitationplaysanimportantmixingandshearingroleinfermentationprocesses. Itnotonlyimproves
massandoxygentransferbetweenthedifferentphases,butalsomaintainshomogeneouschemicaland
physicalconditionsinthemediumbycontinuousmixing. Ontheotherside,agitationcancauseshear
forces,whichinfluencemicroorganismsinseveralways,suchaschangesinmorphology,variationin
growthandmetaboliteformationandevencausingdamagetocellstructures. Aerationdetermines
theoxygenationofthefermentationprocess,andalsocontributestomixingofthefermentationbroth,
especiallywheremechanicalagitationspeedsarelow[18,19].
Process optimization is generally performed in bench-scale bioreactors, and then scaled up
to larger scale for commercial production [20]. The objective of scale-up is to design and build
a larger scale system on the basis of the results obtained from small scale devices [21]. The
scale-upstrategyneedstobebasedonsomecriteriasuchasvolumetricoxygentransfercoefficient
(k a), volumetricpowerconsumption(P/V),impellertipspeed(Vs)andmixingtime(t )[22–24].
L m
Foraerobicfermentation,oxygensupplyisoneofthekeylimitingfactorsformicrobialgrowthand
productformation,thusvolumetricoxygentransfercoefficient(k a)isgenerallyusedasascale-up
L
criterion[25,26]. k aplaysanimportantroleinthedesign,scale-upandeconomyofthefermentation
L
process. This coefficient is considerably influenced by a lot of factors, such as geometrical and
operationalcharacteristicsofvessels,mediumcomponentsandthemicroorganisms’morphology[27].
Anumberofmethodshavebeendevelopedfork ameasurement,suchaschemicalmethods,sodium
L
sulfite oxidation method, absorption of CO , physical methods, and dynamic methods [28–30].
2
The dynamic gassing out method is the most common method and includes the determination of
oxygentransferrate(OTR)andoxygenuptakerate(OUR),respectively[31].
StreptomyceskanasenisiZX01(CGMCC4893)wasoriginallyisolatedfromsoilaroundKanasLake
(XinjiangProvince,China)[32]. OurpreviousstudyindicatedthatstrainZX01canproduceanovel
glycoproteinGP-1withactivityagainstplantviruses,especiallytobaccomosaicvirus(TMV)[33,34].
However, batch fermentation of strain ZX01 was found to be extremely unstable, affording low
GP-1productionandrequiringlongfermentationperiods,whichnotonlyreducetremendouslythe
efficiencyoffutureresearch,butalsoseriouslyhinderpotentialindustrialproductionofthestrainand
commercialdevelopmentofitsrelatedfermentationproducts[35].
In the present work, the effects of agitation, aeration and temperature on the production of
glycoproteinGP-1byStreptomyceskanasenisiZX01ina5Lfermentorwereinvestigatedsystematically.
Inaddition,theprocesswasscaledupfrombench-scale(5L)topilot-scale(15L,70Land500L)on
thebasisofmaintainingthevalueofk atostudythefeasibilityforfutureindustrialproductionof
L
glycoproteinGP-1asanovelanti-plantvirusagent.
2. Results
2.1. EffectofTemperatureontheBenchScaleFermentationofStrainZX01
Figure1showsfermentationresultsunder25,30,35and40◦C,withaerationrateandagitation
speedcontrolledat1.0vvmand150rpm,respectively. AsshowninFigure1A,timecourseofcell
growthwasclearlydividedintothreephases(logarithmicphase,stationaryphaseanddeclinephase).
Themaximumvaluesofdrycellweight(DCW)were3.15,2.94,2.85and2.78g/Lat25,30,35and
40◦C,respectively(Table1). Inotherwords,alowertemperaturewasbeneficialtogrowthofstrain
ZX01althoughcellgrowthrateincreasedwithincreaseoftemperatureinthebeginning(24h).
Molecules2018,23,125 3of14
Molecules 2018, 23, 125 3 of 13

Figure 1. The effect of temperature on DCW (A); GP-1 production (B) and DO (C) during batch
Figure 1. The effect of temperature on DCW (A); GP-1 production (B) and DO (C) during batch
fermentation of Streptomyces kanasenisi ZX01 in a 5 L fermentor.
fermentationofStreptomyceskanasenisiZX01ina5Lfermentor.
Table 1. The final GP-1 production, the maximum DCW and kLa under different temperatures, agitation
Table1.spTeheedsfi annadl aGePra-t1iopnr roadteusc itni oan 5, Lth feermmeanxtiomr.u mDCWandkLaunderdifferenttemperatures,agitation
speedsandaerationratesina5Lfermentor.
Parameters The Final GP-1 Production (mg/L) The Maximum DCW (g/L) kLa (h−1)
Temperature (°C) 25 1.32 3.15 -
30 Th2e.47F inalGP-1 TheM2a.9x4i mum -
Paramete rs 35 Prod2.u20c tion(mg/L) DCW2.85( g/L) kL-a (h−1)
40 1.95 2.78 -
TempAergaittautioren ((r◦pCm)) 150 25 2.59 1.32 33..1025 14.53-
200 30 3.05 2.47 23..9124 18.23-
250 2.16 2.67 27.21
35 2.20 2.85 -
300 1.87 2.61 32.82
40 1.95 2.78 -
Aeration (vvm) 0.5 2.50 3.00 13.21
1.0 2.96 3.10 16.70
Agitation(rpm) 150 2.59 3.02 14.53
1.5 3.41 3.25 18.91
200 3.05 3.12 18.23
2.0 3.86 3.39 22.43
250 2.16 2.67 27.21
The time course of GP-1 p3r0o0duction are illust1r.a8t7ed in Figure 1B. In log2a.r6i1thmic phase, GP-312 b.8eg2an to
be sAynetrhaetisoizned(v avnmd) was excre0t.e5d quickly with c2e.l5l 0growth. After a low 3p.0ro0duction rate in 1st3a.t2i1onary
phase, GP-1 was largely accu1m.0ulated in the ferm2.9en6tation broth due to 3g.r1a0dual cell death d1u6r.7in0g the
decline phase. GP-1 producti1o.n5 initially increas3e.4d1 with the increase of3 t.2em5perature. After1 87.29 1h, the
higher temperature of 35 and2 4.00 °C accelerated 3e.n8z6yme inactivation and3. 3ce9ll senescence, re2s2u.4lt3ing in
a reduction of the GP-1 production rate. The final GP-1 production from 25 to 40 °C were 1.66, 2.47,
2.25 and 1.95 mg/L, respectively (Table 1).
ThetimecourseofGP-1productionareillustratedinFigure1B.Inlogarithmicphase,GP-1began
The DO concentration profiles under different temperatures are presented in Figure 1C. All DO
tobesycnotnhceesnitzreatdioanns ddewcraesaseexdc dreratemdatqicuailclyk tloy aw mitihnicmeullmg lreovwelt wh.itAhifnt e2r4 ah,l aonwd pthreond ruocseti solnowralyt eduinrinstga tionary
phase,G P-1waslargelyaccumulatedinthefermentationbrothduetogradualcelldeathduringthe
declinephase. GP-1productioninitiallyincreasedwiththeincreaseoftemperature. After72h,the
highertemperatureof35and40◦Cacceleratedenzymeinactivationandcellsenescence,resultingina
reductionoftheGP-1productionrate. ThefinalGP-1productionfrom25to40◦Cwere1.66,2.47,2.25
and1.95mg/L,respectively(Table1).
Molecules2018,23,125 4of14
TheDOconcentrationprofilesunderdifferenttemperaturesarepresentedinFigure1C.AllDO
concentrationsdecreaseddramaticallytoaminimumlevelwithin24h,andthenroseslowlyduring
therestMofolefceurlems 2e01n8t, a23t,i o12n5 time. DOconcentrationsweresolowat30,35and40◦C,evendecr4e aofs 1i3n gto0%
saturationduringlogarithmicphaseandstationaryphase. Bycontrast,DOconcentrationmaintained
above2t0h%e rseastt uofr afetrimonendtautrioinn gtimthee. DwOh ocolencfeenrtmraetinontast wioenrep sroo lcoews sata 3t0l,o 3w5 atnedm 4p0e °rCa,t euvreeno dfe2cr5ea◦sCin.gT thoa tisthe
0% saturation during logarithmic phase and stationary phase. By contrast, DO concentration
reasonthatGP-1productionat25◦Cismuchlowerthanothers,becausestrainZX01couldnottake
maintained above 20% saturation during the whole fermentation process at low temperature of 25 °C.
advantaTgheato ifs othxey rgeeasnonfu tlhlayt tGoPs-1ypnrtohdeuscitzioena ant d25p °rCo ids umcuecgh llyocwoepr rtohateni nothGePrs-,1 b.eTcahueses tsutrdaiyn iZnXd0i1c acoteudld thatthe
optimumnott teamkep aedrvaatnutargeef oofr oSxytgreepnt foumllyy ctoe ssyknatnhaessieznei asnidZ pXr0od1utcoe pglryocdopurcoeteignl yGcPo-1p. rTohtee sitnudGyP in-d1icwataesd 3th0at◦ C.
the optimum temperature for Streptomyces kanasenisi ZX01 to produce glycoprotein GP-1 was 30 °C.
2.2. EffectofAgitationonFermentationofStrainZX01onBenchScale
2.2. Effect of Agitation on Fermentation of Strain ZX01 on Bench Scale
Theexperimentswereconductedattheconstantaerationrateof1.0vvmandtemperatureof
The experiments were conducted at the constant aeration rate of 1.0 vvm and temperature of
30◦Cwithdifferentagitationspeedsof150,200,250and300rpm,respectively(Figure2).
30 °C with different agitation speeds of 150, 200, 250 and 300 rpm, respectively (Figure 2).

Figure 2. The effect of agitation speed on DCW (A); GP-1 production (B) and DO (C) during batch
Figure2. TheeffectofagitationspeedonDCW(A);GP-1production(B)andDO(C)duringbatch
fermentation of Streptomyces kanasenisi ZX01 in 5 L fermentor.
fermentationofStreptomyceskanasenisiZX01in5Lfermentor.
At first the production of GP-1 increased approximately at the same level under the four
Atafigristatttiohne sppreoeddsu, catniodn thoefnG bPec-a1mine csrigenaisfeicdanatplyp droifxfeirmenatt eafltyera 2t4t hhe. Tshaem mealxeivmeulmun fidnearl pthroedfuocutirona gitation
of GP-1 was achieved at 200 rpm (3.05 mg/L), and the final GP-1 production at 150, 250 and 300 rpm
speeds,andthenbecamesignificantlydifferentafter24h. ThemaximumfinalproductionofGP-1was
were 2.69, 2.36 and 1.87 mg/L, respectively. The highest values of DCW from 150 to 300 rpm were
achievedat200rpm(3.05mg/L),andthefinalGP-1productionat150,250and300rpmwere2.69,
2.92, 3.10, 2.77 and 2.61 g/L, respectively (Table 1). The higher agitation speeds of 250 and 300 rpm
2.36and1.87mg/L,respectively. ThehighestvaluesofDCWfrom150to300rpmwere2.92,3.10,
could cut off mycelium and damage the cell structure owing to unbearable shear force.
2.77and2.6T1hge/ DLO,r ecsopnceecntitvraetliyon( Tparbofliele1s) .wTehree ghriegahtleyr daigffietraetniot nunsdpeere dfosuor fle2v5e0lsa onfd a3g0it0atriopnm spcoeeudlsd cutoff
myceliu(mFigaunred 2dCa).m DaOg ceonthceenctrealtliosntr wucatsu driestionwctilyn glowtoeru antb 1e5a0r rapbmle, asnhde maraifnotracinee.d below 5% saturation
ThfeorD mOostc oofn tcheen titmraet. iIonn copnrtroafistl,e DsOw ceonrecengtrreaatitolnys dwiefrfee rmeonsttlyu nredmearinfeodu arbolevve e2l0s%o, 3f0%ag aintadt i4o0n% speeds
saturation at 200, 250 and 300 rpm, respectively. The results showed that 200 rpm was the optimal
(Figure2C).DOconcentrationwasdistinctlylowerat150rpm,andmaintainedbelow5%saturation
agitation speed for Streptomyces kanasenisi ZX01 to produce glycoprotein GP-1.
formostofthetime. Incontrast,DOconcentrationsweremostlyremainedabove20%,30%and40%
saturati onat200,250and300rpm,respectively. Theresultsshowedthat200rpmwastheoptimal
agitationspeedforStreptomyceskanasenisiZX01toproduceglycoproteinGP-1.
Molecules2018,23,125 5of14
2.3. EffeMctoloecfuAlese 2r0a1t8i, o2n3, 1o2n5 FermentationofStrainZX01onBenchScale 5 of 13
Th2e.3f.e Ermffecetn otf aAteiroantisonw one rFeercmaernrtiaetdiono ouf tStartaidn iZffXe0r1e nont Baeenrcaht iSocnaler atesof0.5,1.0,1.5and2.0vvm,with
sameagitationspeedof200rpmandtemperatureof30◦C.Figure3showstheeffectofaerationrate
The fermentations were carried out at different aeration rates of 0.5, 1.0, 1.5 and 2.0 vvm, with
onGP-1sapmreo adguitcattiioonn ,spDeeCdW of a20n0d rpDmO ancdo ntecmenpetrraattuioren odf 3u0r °iCn.g Ftighueref e3r smhoewnsta tthieo enffpecrto ocfe asesr.atIionnt rhaitse partof
experimoenn Gt,Pt-1h eprfionduacltiporno, dDuCcWti oanndo DfOG Pco-1nciennctrraetaiosne dduarsinage trhaet ifoernmreantteatiionnc rperaosceesds. (I2n. 5th0i,s 2p.9ar6t, o3f .41 and
3.85mge/xLpearitm0e.n5t,, 1th.0e, f1in.5al apnroddu2c.0tiovnv omf ,GrPe-s1p ineccrteivaseeldy )as( Taaerbalteio1n) .ratTeh iencereffaesecdt o(2f.5a0e, r2a.9t6io, n3.4r1a taendo n GP-1
3.85 mg/L at 0.5, 1.0, 1.5 and 2.0 vvm, respectively) (Table 1). The effect of aeration rate on GP-1
productionwasmoresignificantthanbothagitationspeedandtemperature. Ontheotherhand,the
production was more significant than both agitation speed and temperature. On the other hand, the
maximumDCWalsoincreasedwhenaerationrateincreased(3.00,3.10,3.25and3.39g/Lat0.5,1.0,
maximum DCW also increased when aeration rate increased (3.00, 3.10, 3.25 and 3.39 g/L at 0.5, 1.0,
1.5and12..50 avnvdm 2.,0 rvevsmp,e rcetsipveecltyiv)e(lTya) b(Tleab1le). 1).

Figure 3. The effect of aeration rate on DCW (A); GP-1 production (B) and DO (C) during batch
Figure 3. The effect of aeration rate on DCW (A); GP-1 production (B) and DO (C) during batch
fermentation of Streptomyces kanasenisi ZX01 in a 5 L fermentor.
fermentationofStreptomyceskanasenisiZX01ina5Lfermentor.
All DO concentration profiles decreased greatly before 24 h, and then remained above 10%, 15%,
All2D5%O ancdo n30c%en startautriaotinonp arto 0fi.5l,e 1s.0d, 1e.c5r aenads e2d.0 vgvrmea, trleyspbeecftiovreely2 (4Fihgu,raen 3dC)t.h Tehne DreOm leavienles din astbaotivonea1ry0 %,15%,
phase increased with increase of aeration rates. The DO concentrations were never less than 10% in all
25%and30%saturationat0.5,1.0,1.5and2.0vvm,respectively(Figure3C).TheDOlevelsinstationary
conditions, which indicated that oxygen supply was enough for cell depletion. The result indicated
phaseincreasedwithincreaseofaerationrates. TheDOconcentrationswereneverlessthan10%inall
that 2.0 vvm was the optimal aeration rate to obtain a higher production of glycoprotein GP-1.
conditions,whichindicatedthatoxygensupplywasenoughforcelldepletion. Theresultindicated
that2.02v.4v.m Effwectass oft hAegiotaptitoinm aandl aAeerraattiioonn onr aktLae otno Boebntcah iSncaaleh igherproductionofglycoproteinGP-1.
The volumetric oxygen transfer coefficient (kLa) represents the capacity of oxygen supply and
2.4. EffectsofAgitationandAerationonk aonBenchScale
transfer in the fermentor, which is relatLed with agitation speed, aeration rate, geometrical characteristic of
Thfeervmoelnutmore atnrdic rohxeoylgogeincatlr cahnasrafectrerc ooef fmfiecdieiunmt ([k16]a. )Erfefepcrtse soef nagtsitatthioenc sappeaedci tayndo afeoraxtyiogne nratseu opnp lyand
L
transferkiLna itnh ea 5fe Lr mfeermnteonrt,owr ahriec shhoiswrne liant Tedabwle i1t.h Tahge iktLaat ivoanluseps einecdr,eaaseerda taiso angritaatteio,ng espoemede tarnicda alecrhataioranc teristic
rate increased on bench scale. At agitation speeds of 150 to 300 rpm, the kLa values ranged from
offermentorandrheologicalcharacterofmedium[16]. Effectsofagitationspeedandaerationrate

on k a in a 5 L fermentor are shown in Table 1. The k a values increased as agitation speed and
L L
aerationrateincreasedonbenchscale. Atagitationspeedsof150to300rpm,thek avaluesranged
L
from14.53to32.82h−1. Ataerationratesof0.5to2.0vvm, thevaluesofk awereincreasedfrom
L
13.21to22.43h−1. Thek avalueshadawiderrangeofchangeundertheeffectofagitationspeed.
L
Molecules2018,23,125 6of14
Thek avalueswere18.23and22.43h−1 attheoptimalagitationspeed(200rpm)andaerationrate
L
(2.0vvm),respectively.
In this study, the final GP-1 production and the maximum DCW decreased when k a values
L
increasedfrom27.21to32.82h−1undertheimpactofagitationspeed(Table1). Inotherwords,too
highk ahasnegativeeffectonthefinalGP-1productionandthemaximumDCW.Thereasonshould
L
beattributedtotremendousshearforcecausedbyhighagitationspeedthatcoulddestroythestructure
ofcellandmycelium,andaffectthebiosynthesisofglycoproteinGP-1.
2.5. BenchScaleVerificationExperiments
By using one-factor-at-a-time method, temperature, agitation speed and aeration rate on
bench scale were sequentially optimized to obtain higher glycoprotein GP-1 production by
StreptomyceskanasenisiZX01. FourbatchesfermentationofStreptomyceskanasenisiZX01werecarried
outwith200rpm,2.0vvmand30◦Ctoverifytheresultsofoptimizationexperiment. ThefinalGP-1
production, the maximum DCW and k a values of verification test are presented in Table 2. The
L
resultsrevealthatundertheoptimalconditions,thefinalGP-1productionandthemaximumDCW
couldreach3.92mg/Land3.31g/Lonaverage. ThefinalGP-1productionof3.92mg/Lonabench
scalewasincreasedby54.33%comparedwith2.54mg/Lbeforeoptimization[35]. Ontheotherside,
averagevalueofk awas21.62h−1,whichwasanimportantparameterforscale-upfermentationfrom
L
benchscaletopilotscale.
Table2.ThefinalGP-1production,themaximumDCWandk ain5Lfermentor.
L
Fermentation TheFinalGP-1 TheMaximum
Fermentor k a(h−1)
Condition Production(mg/L) DCW(g/L) L
3.89 3.55 22.83
4.05 3.42 21.31
5L 200rpm,2.0vvm
3.90 3.41 21.57
3.83 3.45 20.78
Average 3.92 3.46 21.62
2.6. Scale-UpFermentationonPilotScale
Thevolumetricoxygentransfercoefficient(k a)hasbeenapreferredcriterionforscale-upof
L
aerobicfermentations,owingtounderlyingprincipleoftheoxygentransferrateinordertoachievea
similaroxygendemandformicrobialpopulationfromasmallerbioreactortoalargerfermentor[25].
Becausebench-scaleandpilot-scalefermentorshavedifferentgeometriccharacteristicsandagitation
systems,thek avalueswereslightlydifferentdespitethefactthesameoperationalconditionswere
L
used [36]. A previous experiment with the optimal conditions obtained from the bench scale was
testedinpilot-scalefermentors,andbothfinalGP-1productionandk avaluesin15L,70Land500L
L
fermentorswerefoundtobelowerthanthoseonbenchscale(Table3).
Therefore,aseriesofscale-upfermentationexperimentswithagitationspeedsof225,250,275and
300rpmwereperformedin15L,70Land500Lfermentors,whentemperatureandaerationratewere
thesameasinthebench-scaleexperiments(Table3). TheresultsindicatedthatthehighestfinalGP-1
productionfoundat225rpmwas4.03mg/LwiththemaximumDCWof3.54g/Lin15Lfermentor.
Asagitationspeedsincreasedfrom250to300rpm,thefinalGP-1productionandthemaximumDCW
decreasedduetothenegativeeffectofshearforce. Similarresultswerealsoobtainedin70Land500L.
Molecules2018,23,125 7of14
Table3.Measuredk avaluesatdifferentfermentationconditionsinpilot-scalefermentors.
L
TheFinalGP-1 TheMaximum
Fementor Agitation(rpm) k a(h−1)
Production(mg/L) DCW(g/L) L
15L 200 3.75 3.55 19.15
225 4.03 3.54 22.52
250 3.92 3.40 25.74
275 3.57 3.20 29.51
300 3.20 3.14 31.43
70L 200 3.55 3.60 17.53
225 3.75 3.60 20.32
250 3.89 3.76 23.11
275 3.50 3.58 26.18
300 3.16 3.32 29.78
500L 200 3.81 3.77 15.04
225 3.96 3.81 18.38
250 4.13 4.01 22.31
275 4.00 3.85 27.04
300 3.80 3.66 32.47
In order to verify the stability of fermentation scale-up process, four batches of repeated
experimentswereperformedinpilot-scalefermentors. TheresultslistedinTables2and4showedthat
5L,15L,70Land500LfermentorsachievedfinalGP-1productionof3.92,4.03,3.82and4.20mg/L,
themaximumDCWof3.46,3.53,3.69and3.95g/L,withsimilark aof21.62,22.80,23.20and22.24h−1.
L
Consequently,thebatchfermentationofglycoproteinGP-1bystrainZX01wassuccessfullyscaled
upfrombench-scaletopilot-scalethroughcontrollingagitationspeedtomaintaink aconstant. For
L
scale-upofaerobicfermentation,k aisfrequentlyusedasthecriteriontoensureequaloxygentransfer
L
conditionsondifferentscales[25]. Thepresentstudyalsodemonstratedthatk awasapplicabletothe
L
scale-upofbatchfermentationofglycoproteinGP-1productionbyStreptomyceskanasenisiZX01from
bench-scaletopilot-scalefermentor.
Table4.ThefinalGP-1production,themaximumDCWandk ainpilot-scalefermentors.
L
Fermentation TheFinalGP-1 TheMaximum
Fermentors k a(h−1)
Condition Production(mg/L) DCW(g/L) L
3.98 3.52 23.63
4.10 3.55 23.71
15L 225rpm,2.0vvm
4.10 3.61 22.72
3.94 3.45 21.14
Average 4.03 3.53 22.80
3.81 3.71 23.42
3.93 3.76 24.27
70L 250rpm,2.0vvm
3.78 3.65 23.51
3.77 3.63 21.59
Average 3.82 3.69 23.20
4.23 3.93 22.20
4.28 3.90 23.28
500L 250rpm,2.0vvm
4.09 4.06 21.81
4.20 3.91 21.68
Average 4.20 3.95 22.24
Molecules2018,23,125 8of14
3. Discussion
Theresearchoneffectsoffermentationconditionssuchastemperature,agitationandaerationand
otherparametersduringtheprocess,suchasdissolvedoxygen,pHandreducingsugaronbiopolymers
productioncanplayanimportantroleinunderstandingthesynthesisofbiopolymers,fermentation
kineticsandprocessscale-up[25].
Temperatureisoneofimportantfactorsaffectingmicrobialfermentation,duetoitscorrelation
with all biochemical enzyme catalysis processes. Growth and metabolism of microorganisms
shouldproceedundertheappropriatetemperature. Temperaturehasvariousimpactsonmicrobial
fermentation,includingcellgrowth,productsynthesis,biosynthesisdirectionandphysicalproperties
ofbroth. Accordingtothekineticsofenzymecatalysis,highertemperaturesresultinfasterreaction
rates, leading in turn to sped-up growth and product synthesis. However, when the temperature
exceeds the optimum growth temperature of a microorganism, inactivation and denaturation of
enzymeswilloccur,resultinginmicroorganismdeathandfinallyinreductionofthefermentationcycle.
Furthermore,theoptimumtemperaturesforcellgrowthandmetaboliteaccumulationarefrequently
different. Theoptimumtemperatureforgrowthofglutamicacid-producingstrainis30–34◦C,for
example,whiletheproductionrateofglutamicacidreachesthemaximuminatemperaturerangeof
35–37◦C[37–39]. OurresultsshowthattheoptimumtemperatureforstrainZX01growthandGP-1
accumulationwere25◦Cand30◦C,respectively.
Agitationcouldmainlycausemixingandshearinthefermentationprocess, whichcanmake
oxygen, heat and nutrients mix fully and be transferred efficiently in the fermentation broth, and
dispersetheairintosmallbubblestoimprovethegas-liquidcontactarea,andpreventmyceliafrom
clustering to favor of oxygen absorption [19]. Too high an agitation speed not only increases the
powerconsumption,butalsocreatesheterogeneousmixingandshearforcesthatcandamagefragile
microorganismsandaffectproductformation[13]. Ontheotherhand,whentheagitationspeedis
toolow, theviscosityoffermentationbrothwillincrease, leadingto anreductionin masstransfer
efficiency[36]. Theresearchrevealedthat200rpmwastheoptimalagitationspeedforstrainZX01to
growandproduceglycoproteinGP-1.
Aerationnotonlysuppliesthenecessaryoxygenforcellgrowth,butalsoeliminatesexhaustgas
generatedduringthefermentationprocess[19]. However,higheraerationrateresultsinareduction
in the volume of fermentation broth. Oxygen supply is necessary for microorganisms’ growth in
aerobicfermentation,butsomemicroorganismsmaybeaffectedbyoxygentoxicityatexcessiveoxygen
concentration[36]. Thisoxygentoxicitysituationdidnotoccurinourstudy,sincethemaximumDCW
andfinalGP-1productionwereachievedatthehighestaerationrateof2.0vvm.
The scale-up of the fermentation process based on constant k a for improvement of GP-1
L
production from bench scale to pilot scale was another important feature in this article, leading
toafinalGP-1productionof3.92,4.03,3.82and4.20mg/Lin5L,15L,70Land500Lfermenters,
respectively. Theseresultsindicatedthatk awasquitesuitableforthescale-upofbatchfermentation
L
processofglycoproteinGP-1byStreptomyceskanasenisiZX01. Inordertoensureequaloxygentransfer
conditions on different scales, k a has been preferred as the main criterion for scale-up of aerobic
L
fermentation[25].Kshirsagaretal.reportedthatPHAproductionbyHalomonascamposalisMCMB-1027
wassuccessfullyscaledupfrom14Lto120Lbyfixingk a[40].Quetal.achievedamaximum150-fold
L
scale-upofDHAproductionfroma10Ltoa1500Lfermentorbasedonmatchedk avalues[20].
L
TherewasslightreductionofthefinalGP-1productioninthe70Lfermentor. Incontrast,the
result at 500 L was much higher. The hydrodynamic environment in pilot and plant fermenters,
includingmixingeffects,shearforcesandoxygentransferefficiency,isnormallymorecomplexthan
thatinbench-scalebioreactors[25]. Theincreaseofbrothviscositycausedbyhighmyceliabiomass
actsasabarriermakingoxygentransferfromtheliquidphasetocellsdifficult[25]. Thoughagitation
speedhasadirecteffectonmixingthatcouldreduceviscosityandincreaseoxygensolubilityand
transferinbroth,shearforcescausedbyhigh-speedimpellerwillhaveanegativeeffectoncellgrowth
andmetaboliteaccumulation[41]. MoreandmoreresearchershaveusedtheComputationalFluid
Molecules2018,23,125 9of14
Dynamics(CFD)techniquetoacquireinformationabouttheflowfieldinfermentorsbysimulating
the fermentor structure. Many of them have reported that the impeller types and combinations
haveanobviouslycorrelationwiththevolumetricoxygentransfercoefficient. Homogeneousmixing,
moderateshearenvironmentandhigherproductproductioncouldbeachievedthroughoptimizing
impellernumber,typesandcombinations[42–46]. Hence,inordertoimprovetheflowfieldandair
distributioninsidethefermentor,andfurtherincreaseglycoproteinGP-1production,differentimpeller
combinationsinlarge-scalefermentorwillbeoptimizedbyusingCFDtechniqueasfuturework.
In conclusion, our work showed that the maximum glycoprotein GP-1 production by
StreptomyceskanasenisiZX01wasobtainedatanagitationspeedof200rpm,aerationrateof2.0vvm
andtemperatureof30◦Cina5Lfermentor. Onthebasisofvolumetricoxygentransfercoefficient,
glycoproteinGP-1wassuccessfullyscaledupfrom5Lto15L,70Land500L,resultinginproductions
of3.92,4.03,3.82and4.20mg/L,respectively.
4. MaterialsandMethods
4.1. Microorganism
StreptomyceskanasenisiZX01,obtainedfromtheResearchandDevelopmentCenterofBiorational
Pesticides(Yangling,China),wasisolatedfromsoilatKanasLake(XinjiangProvince,China). Strain
ZX01isregisteredattheChinaGeneralMicrobiologicalCultureCollectionCenter(CGMCC)under
strain number CGMCC 4893. The strain was maintained on Gauze’s No. 1 agar medium and
subculturedatmonthlyintervals,orstoredin20%glycerolat−70◦C.
4.2. InoculumPreparationandMedia
StrainZX01waspre-culturedinGauze’sNo. 1agarplatefor72h. Afterpre-culture,aloopof
inoculum from the plate was inoculated into the seed medium of 100 mL for preparation of seed
inoculum. Forthe5,15and70Lfermentations,seedinoculumwasculturedin500mLshakeflasks
with200mLworkingvolumeat30◦Cand150rpmfor72h. For500Lfermentation,seedinoculum
wasculturedina50Lseedfermentorcontaining17.5Lseedmediumat30◦Cand150rpmfor72h.
The seed medium composition was (in g/L): soluble starch, 20; KNO , 1; K HPO , 0.5;
3 2 4
MgSO ·7H O, 0.5; NaCl, 0.5; FeSO ·7H O, 0.01. The fermentation medium was (in g/L): millet
4 2 4 2
steepliquor,10;solublestarch,10;yeastextract,3;NaCl,2.5;CaCO ,0.2.
3
4.3. Fermentors
Four sizesof fermentors were usedin this study, and theirspecifications are listed inTable 5.
Bench-scalefermentationwascarriedoutin5Lquadrupleglassbioreactors(GBCN-5C,ZhenjiangEast
BiotechEquipmentandTechnologyCo.,Ltd.,Zhenjiang,Zhenjiang,China)with3.5Lworkingvolume.
Thebioreactorwasequippedwithathermometer,pHsensor,dissolvedoxygensensor,tachometer,
air-flowmeterandinternalpressuresensorandfoam-sensingprobe. Theagitationsystemconsisted
of two impellers with four-flat-blade and the magnetic base. The agitation rate was controlled by
electromagneticimpulse. Theaerationsystemwasanairinletthrougharingspargerwithanair-flow
meterandfilter. Temperaturewasmaintainedconstantbyaheatingsysteminthebottomandcooling
water. Thebioreactorandallitspartscontaining3.5Lmediumwassterilizedbyhighpressuresteam
sterilizationpotat121◦Cfor30min. Aftersterilization,thefermentationmediumwasinoculatedwith
5%(v/v)seedinoculum. Thesignalofthefoam-sensingprobewasconnectedtoanelectromagnetic
valvethrougharelaytoaddantifoam.
Stainless steel fermentors (GUJS-15, GUJS-70 and GJ-500 with nominal 15 L, 70 L and 500 L
capacity,andworkingvolumesof10L,50Land350L,respectively,ZhenjiangEastBiotechEquipment
andTechnologyCo.,Ltd.,Zhenjiang,Zhenjiang,China)wereusedinpilot-scalefermentation. The
equipmentalsoincludedathermometer,pHsensor,dissolvedoxygensensor,tachometer,air-flow
meter and internal pressure sensor and foam-sensing probe. The agitation was provided by two
Molecules2018,23,125 10of14
six-flat-bladeimpellersanddrivenbymechanicalstirredfromthetop. Airflowwassuppliedand
controlledbyspidersparger,airfilterandair-flowmeter. Temperaturewasmaintainedconstantby
thermostaticwatertankandcoolingwater. Fermentorscouldbesterilizedin-situwithexogenous
steam. AlldatasuchasDO,pHandtemperaturewereonlinemonitoredoverthefermentationprocess
andinputintoacomputerat1hintervals.
Table5.Thespecificationsoffermentors.
Fermentor Bench-Scale Pilot-Scale
Totalvolume(L) 5 15 70 500
Workingvolume(L) 3.5 10 50 350
Diameteroffermentor(m) 0.15 0.25 0.38 0.75
Diameterofimpeller(m) 0.07 0.10 0.15 0.30
Heightoffermentor(m) 0.30 0.50 0.75 1.50
Baffle 3 4 4 6
Impeller Twoimpellerswithfour-flat-blade Twoimpellerswithsix-flat-blade
Typeofdrive Magneticstirred Mechanicalstirred
Sterilization Ex-situ In-situ
4.4. ExtractionandDeterminationofGlycoproteinGP-1Production
The fermentation broth (100 mL) was centrifuged at 10,000 rpm for 20 min to separate the
precipitate and supernatant. The supernatant was concentrated to a volume of 10 mL by rotary
evaporatorandthenprecipitatedbyadding4-foldvolumesofethanolat4◦C.Theprecipitatewas
redissolvedindistilledwater(10mL)andcentrifuged(10,000rpm,10min)againtoremovethose
water-insoluble materials. The supernatant was applied to a DEAE-52 Cellulose anion-exchange
column(2cm×60cm)elutedwithdeionizedwaterfirst,andthenwith0.1MNaClataflowrate
of 5mL/min. The 0.1 M NaCl fraction was collected and centrifuged (10,000 rpm, 15 min) with
centrifugalfilterdevices(3K,0.5mL)toremoveNaCl. ThefractionwassubjectedtoHiTrapTMConA
4Belutedwithbindingbuffer(20mMTris-HCl,0.5MNaCl,1mMMnCl ,1mMCaCl ,pH7.4)and
2 2
elutionbuffer(0.1Mmethyl-α-D-glucoside,20mMTris-HCl,0.5MNaCl,pH7.4)sequentiallyata
flowrateof1mL/min. ThefractionelutedwithelutionbufferthatcontainedGP-1wasconcentrated
to100µLandanalyzedbyhighperformanceliquidchromatography(HPLC).
The concentration of GP-1 was analyzed by a HPLC (Waters, Milford, MA, USA) with a gel
filtration column (TSK-gel G2000SWXL, 7.8 × 300 mm, 5 µm, TOSOH, Tokyo, Japan) and a 996
photodiodearraydetectorat280nm. HPLCwasperformedon10µLsamplewith20%acetonitrileata
flowrateof0.5mL/minand28◦C.GP-1purifiedpreviously(purity>99%)wasdilutedto10,5,2.5,
1.25and0.625mg/mLasstandards.
4.5. DeterminationofDryCellWeight
Theprecipitateobtainedbycentrifugingbrothsamplewaswashedwithdistilledwatertwice,and
thenfreeze-driedtoaconstantweightbyfreezedryer(SIMInternationalGroupCo.,Ltd.,Norwood,
MA,USA),expressedasdrycellweight(DCW,g/L).
4.6. DeterminationofVolumetricOxygenTransferCoefficient
OUR,OTRandk aweredeterminedbydynamicgassingoutmethodusingDOprobe[32]. Based
L
onthedynamicmassbalanceinbatchfermentation,thefollowingequationforchangesinDOcould
beestablished:
dC (cid:16) (cid:17)
O2 =OTR−OUR= k a C∗ −C −Q ·X (1)
dt L O2 O2 O2
∗
whereC isthedissolvedoxygenconcentrationofthefermentationbroth(g/L),C isdissolved
O2 O2
oxygen concentration in equilibrium with oxygen partial pressure (g/L), Q is specific oxygen
O2