1995- Mike Romanos- Advances in the use of Pichia pastoris for high-level gene expression.pdf

8/10/2019 1995- Mike Romanos- Advances in the use of Pichia pastoris for high-level gene expression.pdf

1/7

Advances in the use o f i c h i a p a s t o r i sfor h igh leve l gene express ion

i k e R o m a n o s

G l a x o We l l c o m e B e c k e n h a m U K

The P i c h i a p a s t o r i ssys tem has the potential for very high level p rodu ctionof foreign proteins. This together with the recent availabil ity of this systemin kit form has changedP i c h i a f rom being an express ion sys tem wi th onlyspecialized biotech nolog ical applications to one that is wid ely u sed for rapidexpression in the laboratory. The developme nt of G41 8 selection pro tocols hassimplified the rapid isolation of mu lticopy transformants efficiently expressingintracellular proteins. In addit ion several strategies are now also availablefor optim izing secretion such as the generation of clones with progressivelyincreasing vector copy num ber expression screening in microtitre plates and

minim iz ing proteolysis b y a num ber of techniques.

Current Op inion in Biotechnology 1995 6 :527 -533

I n t r o d u c t i o n

In the pas t few years , thePichia pastorisexpress ion sys temhas rapid ly achieved mu ch w ider use thanks to increas ingawarene ss of the early successes and to i ts sale in k it formby Invi t rogen (San Diego, U SA). F rom bein g a special is tsy s t em, u sed p r imar i l y by yeast g roups i n b io t echn o logycompanies , i t i s now a mains t ream express ion hos tused a longs ideEscherichia coli, Saccharomyces cerevisiae,andbaculovi rus . This was c lear f rom the d iverse in teres ts ofde legates a t a recentPichia meet ing (San Diego, USA,1994) organized by Invi t rogen.

Pichia i s an indus t r ia l methylo t rophic yeas t in i t ia l lychosen fo r p roduc t ion o f s i ngl e -ce l l p ro t e in because o f i tsabi l i ty to grow to very h ig h ce ll dens i ty in s imple d ef inedmedia . From th is bas is , a h ighly eff ic ient express ionsys t em was des igned us ing t he m e thano l - i nduc ib l ea lcohol oxidase 1 A O X 1 ) promote r and vec to r s t ha tin tegra te in to thePichia gen om e [1] . Several examplesshow tha t Pichiacan rout ine ly achieve percentage y ie lds(5-40 of ce l l pro te in) mu ch h igher than baker 's yeas t,and o f t en equ iva l en t t oE. co l ior baculovirus [1,2].Addi t ional ly, sca le-up ofPich i a cul ture to h igh ce l ldens i ty i s s imple and has resul ted in eno rmo us y ie lds on avolu me tr ic basis (e .g . >12 g l - I for te tanus tox in f ragm entC [3] and >3 g l-1 secreted hum an seru m a lbu min [HSA]

[ 4 ] .A l t h o u g h m o s tPichiaexpression ve ctors are v ery similar,d i fferent st rategies can be emp loyed because a se lec t iono f hos t pheno types and ch romosom a l s it es o f i n t eg ra t i onis available. This gives greater scope for optimizingexpress ion , but has caused some confus ion , par t icu lar ly

among users unfami l ia r wi th yeas t genet ics . In th isreview, I at tem pt to clarify the choic es that are availableand g ive sugges t ions as to when they should be used . Ia lso d iscuss some o f the m ost recent develo pmen ts andresul ts wi th thePichiasys t em, and p rov ide an upda t e ons o m e o f t he m a n yPichia-der ivedproduc t s i n commerc i a ldeve lopmen t o r abou t t o r each t he marke t .

P i c h i a a d v a n t a g e s a n d d i s a d v a n t a g e s

Pichiahas the fo l lowin g ma in advantages : f ir s t, ex t reme lyhigh y ie lds of in t race llu lar pro te ins ; second, very h ighlevels of secre t ion in to an a lmost pro te in- f ree m ediu m;th i rd , ease of fe rment a t ion to h igh ce l l dens i ty ; andfour th , genet ic s tab i l i ty and sca le-up wi thout loss ofy ie ld . I t i s proving va luable in producing la rge amountsof pro te in for analy t ica l studies; on e in teres t ing recentapplication is in efficientin vivo i so topic label l ing ofprote ins for N M R [5] . Th e sys tem is a lmost cer ta in lythe s implest of any to sca le up , a fea ture tha t makes i t veryat t rac t ive for rapid produ ct ion of b io logica l products forclinical trials.

Like any o ther express ion sys tem, however,Pichia isno panacea , and ex amples o f low yie lds or fa i lureof express ion are a lso accumula t ing , thou gh m anyrema in unpub l i shed . P robab ly t he commones t p rob l emenco unte red is pro teolys is of secre ted polypept ides[1] , t houg h a num ber o f ways o f ove rcoming th ishave become ava i l ab l e . Ano the r common p rob l em i sineff ic ient secre tion of com plex fore ign prote ins (e.g.

AbbreviationsA O X l a l c o h o l o x id a s e 1 ;EGF-~epidermal growth factor; HSA--hum an serum album in; IGF -l--insulin-like growth factor 1

MutS--methanol utilization slow;PH O 1 ac id phospha t a se 1 .

Current Biology Ltd ISSN 0958 1669 5


2/7

5 8 Expression ystems

(a)

Sad

N o t lAvrl I

EcoRliSnaBlNcol I

B a m H l [

A O X I t

A O X I

A p pPIC37 7 6 8 b p

glU3 A O X I

HIS4

b )

Sac[B e / It ~

gl~l

N o dAvrl I

EcoRI iSnaBI iNco l I

BamH I [

A O X I t

A O X I

A p pPIC3K HIS49 0 0 9 b p

kan~'

3 ' A O X I

(r) 1995 CurremOp inion l~ Biotechnology

F ig . 1 . Two typ i ca lPichia exp re s s ionvec to r s , pP IC3 and pP IC3K. ( a ) VecpPIC3 compr i s e s a SA O X 1 r eg ion i n -c lud ing p romo te r, c l on ing s i t e po ly l i nA O X I t e rmina to r (AOX1 t) ,H I S 4 marker,3 A O X 1 r eg ion and amp ic i l l i n - r e st ance marke r (Ap) . ( b ) Vec to r pP ICin add i t i on t o t he above , con t a in s kanam yc in - r e s i s t ance (kan r ) ma rke r G418 s e l ec t i on o f mu l t i copy t r an s fman t s .

HIV -1 gp l2 0 [6] ). Desp i t e many example s o f y i e ld sin the grammes per l i t re range , he tero logous secre t ionis more demanding than in t race l lu lar express ion and notguaran teed to wo rk . F ina l ly, some genes do n ot g ive anydetec table pro te in , of ten because yeas t t ranscr ip t ionalt e rmina to r s r e su l t i n t r unca t ed mRNA. The p rob l emwas f i rs t descr ibed for the h ighly AT-r ic h te tanus toxing e n e i n S. cerevisiaeand so lved by gene synthes is toinc rea se t he G C con ten t [2 ] . The s ame p rob l em hasbeen s een i n Pichia and so lved in the same way (e .g .w i t h t h e Bacillus sphaericusBsp2 insecticidal toxin [P1]and wi th HIV-1 Env [6] ) . I t i s notewor thy tha t theEn vDN A, wh ich i s no t AT - r i ch , is e~c i e n t ly t r ansc r ibed i n

S. cerevisiae.

i c h i a vectors and s t ra ins

As Pichia has no stable episomal vectors, integratingvectors a re employe d. Al l useH I S 4 as a selectable marke rand have the same genera l organiza t ion , as exempl i f iedby pP IC3 (Fig. 1). Some of the m ost usefu l vec torsand the i r proper t ies a re l i s ted in Table 1 . The mostcommonly u sed hos t s t r a in i s GS l15his4), bu t t hemore recently available protease-deficient strains (e.g.S M D l 1 6 8 ; his4, pep4) are f inding increas ing use inreduc ing proteoly t ic c leavage of secre ted pro te ins . KM 71his4, aoxl)and other strains that have an inactiveA O X I

gene , a re fundam enta l ly d ifferent in tha t the y grow verys lowly on methanol as a carbon source , for example ,du r ing i nduc t ion ( i . e . t hey have a Mut s pheno type ,r a the r t han t he M ut + pheno ty pe o f GS l15 ) .Express ion vec tors a re d i rec ted to in tegra te in to thePichia g e n o m e i n o n e o f tw o w a y s , d e p e n d i n g o nwhere t he DNA i s cu t be fo re t r ans fo rma t ion ( s eeTab le 2 ) . D iges t i on t o g ive a DNA f r agmen t w i thh o m o l o g y t oA O X 1 at both ends leads to replacement

o f g e n o m i cA O X 1 by th e f rag men t ( i.e. t ransplacementand genera tes a Muts recombinant s t ra in . Linear iza t iof the vec tor, by cut t in g e i ther 5 ' to theA O X 1 p r o m o t e ro r w i t h i n t h e H I S 4 marker, d i rec ts in tegra t ion o f thep l a smid a t t he hom ologou s s it es i n t he genome . T he Mphen o type o f t he l a t t e r t ype o f t r ans fo rman t is d i c ta tby the hos t s t ra in used .Some unexpec t ed f ea tu re s o fPichia t ransformat ionsinc rea se t he num ber o f t ypes o f t r ansfo rman t and t hthe po t en t i a l f o r con fus ion . Wi t h i n t eg ra ti on a tA O X 1o r H I S 4 , mult ico py t ransformants (up to 10) can arif rom repea t ed r ecombina t ion even t s . On the o thhand , t r ansp l acemen t wou ld be expec t ed t o y i e ld on

s ing l e - copy Muts transformants; however, a detaileanalys is has revealed three sources of he te rogeneamong the t ransformants [3] . F i rs t , a h igh propor t iof ' t ransforman ts ' conta in no vec tor, express no fore ipro te in and probably representhis4 gene convers ions .Second , only 5-30 are t rue t ransplacements (Mutthe remainder have in tegra t ions a tA O X 1 o r H I S 4and are M ut + . F ina lly, 1-10 of M uts t ransformantshave up to 30 in teg ra ted copies of the t ransplac if ragment as tandem head to ta i l repeats tha t probabar ise by a mechanism involv ingin v ivo l igation [3].Th e M ut + popu la t i on a lso con t a in s m u l t i copy c lonAl though t ransplacement usual ly requi res the labor io

sphaeroplas t t ransform at ion techniq ue because o f lof requencies , i t y ie lds a h ighly d ivergent popula t iont ransformants , which i s usefu l for de ta i led opt imiza t is tudies . Also , it seems to be the only w ay o f obta in ive ry h igh copy num ber t r ans fo rman t s .

Ge ne dosage is c r i t i ca l fo r m axim al express ion

In s om e o f the earl iest studies, expression o f galactosidase and hepati t is B surface antigen was n


3/7

H i g h le v e l e x p r e s s i o n i ni c h i a pas to r i R o m a n o s 5 2 9

Ta b l e1. Pichia expression vectors.

Vector Un iqu e clon ing Selection Comm ents Referencesnam e sites markers

I n t r a c e l l u l a r p r o d u c t i o npPIC3 BamHI, NcoI*

SnaBI, EcoRI,Avrll

andNott

:~PIC3K BamHI, Ncol*5naBI, EcoRI

Avrll and Nott

pHIL-DI EcoRI

EcoRHIL-D2

HIS4 Po lylinke r vector [I 2 *]

HIS4and kan Polyl inker vector wi th G418- [12 ]selection for multicopy clones

HIS4 On e of the first basic vectors [9]

HIS4 Contains fl ori andNotl [a]sites for transplacement

pHIL-D3 Asull and EcoRI HIS4

pHIL-D4 EcoRI HI54and karl

pHIL-D5 EcoRI

pHIL-D7 Asull and EcoRI

) A O 8 1 5 EcoRI

Xhol, EcoRI, 5maland BamHI

S e c r e t i o n~HIL-SI

~)PIC9 Xhot, 5naBI,EcoRI, Avrlland Noll

pPIC9K Xhol*, 5naBI,EcoRI, Avdland Notl

5tul, EcoRI, BglllNotl, Xhol, 5peland

BamHI

pYAM7SP6

Contains fl ori and nativeAsull site forunaltered AOX1 5' untranslated region

[a]

G41 8-selection of [a]mul t icopy c lones

HIS4and kar~ Contains fl ori ,Nofl sites for [a]transplacement and G418-selection

for mul t icopy c lones

HIS and kanr Contains fi ori, a nativeAsull [a]cloning site,Nolt sites for

transplacement, and uses G418-selection for mu lticopy clones

HIS4 Contains fl ori and aBamHIsite for [1,16]generation of multicopy expression units

in vitro

HIS4 Conta insPichia PH01 signal and fl ori [a]

HIS4 Contains a 5.cerevisiae [12 ]a-factor leader

HIS4and kanr Contains a S.cerevisiae et-factor leader, [1 2 ']wi th G418-se lect ion for mul t icopy c lones

HIS4 Conta ins a hybr idPichia Phol s ignal wi th a [5 ]Kex2 endopeptidase cleavage site

*The c loning s i te i s not unique, thus three-way l iga t ions are used.deta i l s contact K Sreekrishna, Marion Me rre l l Do w Inc , 2110 East

[a] K Sreekrishna, K Kropps, personal com mu nica tion. (For furtherGalbra i th Road, PO Box 156300, Cincinn at i , Ohio 45215 , USA.)

improved by i nc rea s ing vec to r copy number [7 ,8 ] . I nnumerous subsequent examples , however, the i so la t ionof mu l t icop y in tegrants has resul ted in dramat ica l lyhig her y ie lds [3 ,9-11] . In a de ta i led s tudy o f te tanustoxin f ragment C, our group [3] showed tha t express ionwas co r re l a t ed w i th copy num ber 1 -14 cop i e s) , whe reass it e o f i n t eg ra t i on and M ut p heno type had , a t mo s t ,

only a minor effec t on y ie ld . This i s perhaps surpr is ingbecause t he Mu t phe no typ e a f fec ts bo th t he g row th r a tedu r ing i nd uc t ion and t he a ccumula t i on o f l a rge amoun t sof end oge nou s a lcoho l oxidase. M ut s s trains ma y y ie lda h ighe r p ropo r t i on o f co r r ec t l y fo lded p roduc t i ns i tua t ions where fo ld ing is ra te- l imi t ing e .g . in the caseo f hepati t is B surface antig en [8]) .


4/7

53 Expression ystems

Table2. Types ofPichia transformant.

Host Vector digest Resulting His+ Commentsstrain to direct transformants

integration*

GSl15 Sad(Mut+) AOX1

integration)

Sa~ and StulHIS4 integration)

BgnltAOX1

transplacement)

KM71 5acl o r 5all/Stul(Muts)

Vector is integrated 5' to the genom icAOXI gene, leaving heAOXI geneundisrupted (i.e. M ut*phenotype)

Vector i s integra ted wi th in g enom ichis4 locus. A O X Igeneundisrupted (i .e. Mut+phenotype)

Bglllfragment replaces genomicAOX1gene, generating Mutsphenotype. (Only 5-25 oftransformants are of this type, theremainder are mainlyA O X t orHIS4 integrants l i .e. Mut+phenotype])

Same transformants, asGSI 15, except all are M ut sbecause host A O X Igeneis already disrupted.

High -freque ncy transformation usin geither sphaeroplasting or electroporation.Ideal for rout ine use . M 00 oftransformants express protein. M ulti -co pyintegrants (up to 10 copies)arise at low frequencies

High-frequency ransformation usingeither sphaeroplasting or electroporation.-100 of transformantsexpress protein.Note p otential to generate His+ 'pop-outs'lacking foreign geneL M ulti-copy ntegrants(up to 10 copies) arise at low freq uency

Low-frequency transformation, preferablyusing the sphaeroplast method. This is bestmethod for mul t i -copy clones , y ie ld in g a l - I 0frequency and up to 30 copies. Generates aheterogeneous pool of Mu ts and Mut +transformants, including somenon-expressers

Higher transformation frequency thanGSI I 5, especially with electrop oration

*These sites are com mo n to all vectors and can be us ed generally, unless present in the foreign gene . *The vector p HIL-D 2 hasNot1sites in place ofBgllland can be u sed whe n the fore ign gene conta insBglllsites. ~;This problem does no t often o ccu r in sm all-scalecultures.

The ex t r eme ly h igh l evel o f a l coho l ox idase (5 -30 )expressed f rom A O X I i n i nduced w i ld - typePichiahad sugges t ed t ha t one copy o f anA O X I express ionvec to r wou ld p roduce max ima l rnRNA l eve l s . I n arecent paper, m R N A levels were analyzed f rom a ser ieso f t r ans fo rman t s con t a in ing i nc rea s ing copy number s(1 -12 ) o f an HIV -1Env express ion vec tor [12] . TheEnv m R N A leve l inc r eased p rog res s ive ly w i th cop ynumber up t o t he max imum number t e s t ed ; a t a s i ng l ecopy o f the v ec tor, i t was two - to threefo ld less tha nA O X 1 mR.NA, bu t w i th g r ea t e r t han t h r ee cop i e s i texceededA O X 1 m R N A . A progressive increase was alsoseen wi th f ragm ent C , sugges t ing tha t t ranscr ip t ion i s, ingenera l , l imi t ing inPichia con ta in ing on ly a s i ng l e copyof the vec to r and tha t i t is sensible to rout ine ly max imizecopy nu mb er. Despi te th is , o ther fac tors , such as pro te instabili ty, mu st hav e a maj or effe ct because final yieldsof d i fferent pro te ins vary grea t ly, even wi th mu l t icopyclones . For example , 12 copies ofEnv vec to r y i e lded2.5 of to ta l ce ll pro te in , whereas 14 copies of f ragrnentC vector y ie lde d 27 .

Aside f rom gene dosage , a c r i t ica l fac tor tha t haslong been r ecogn ized t o a f f ec t i nduc t ion e f f i c i encyis aera t ion of Pichia cultures. This results from the

tend ency of cu l tures , especial ly M ut + s t rains , to b eco moxy gen- l imi te d in shake-flask induct ions , an d i t probabexplains the consisten tly large increase (e.g. 5-10 -fold ) y i e ld t ha t is obse rved wh en swi t ch ing t o f e rmen te r s.

Secre t ion in i c h i a

Many users have been a t t rac ted toPichia by the ve ryhigh re por ted levels of pro te in secre t ion in h igh -denscul tures, such tha t the prod uct can com prise >80 the pro te in in the medium. Yet , secre t ion i s compleand i s dependen t no t on ly on f ac to r s such a s gendosage and Mut pheno type , bu t a l so on o the r f ac totha t a ffec t the y ie ld and qu al i ty of prod uct (e.g. s ignsequence , processing , pro teolysis and g lycosyla t ion) .

Perhaps the most content ious i ssue , as demonst ra ted the l ive ly debate a t the recent meet ing onPichia in SanDiego, i s Mut phenotype . Because secre tory ves ic lin S cerevisiae loca l ize to the bud, i t had been widelbe l ieved tha t secre t ion could only occur in d iv id icel ls . Secre t ion of mouse a- amy lase f romS cerevisiae has,howeve r, been show n to be a s e f f i ci en t i n non -d iv id i


5/7

H i g h l e v e l e x p r e s s io n ni c h i a p a s t o r i sR o m a n o s 5 3

a s d iv id ing ce l l s [ 13 ] . The g roup a t S I BI A In c ( LaJo l la , Ca l i f o rn i a ) had f avou red M ut + s tr a in s an d haveu s e d t h e s e i n m a n y e x a m p l e s o f h i g h - le v e l s e c r e t io n(e . g . ep ide rma l g rowth f ac to r [EGF] , i n su l i n - l i k e g r owt hfac tor-1 [ IG F-1 ] , apro t in in , e tc . [14 ,P2 ,P3 ] ) . Ev en so ,r e su l ts w i t h H S A [4 ] a n d m u r i n e E G F [1 5] d e m o n s t r a t etha t Mu tS s t r a i n s c an a l so y i e ld h igh l eve l s . Man y mo rer ecen t succe s s fu l examp le s u t i li z e e i t he r M u t + o r M u t sstrains.

I n the ca se o f ge n e dosage , ag r ee me n t i s ab s o l u t e th a t ano p t i m a l , r a t h e r ' t h a n m a x i m a l , c o p y n u m b e r i s u s u a ll yr equ i r ed . I n m any cases , s ec r e t ion e f f i c i en c y ha s bee nimp roved w i th s eve ral vec to r cop i e s [P2 ,P3 ]. E xam p l e sa l so ex i st w h e r e t h e m a x i m u m c o p y n u m b e r t e s te d w a so p t i m a l ( e .g . w i t h m u r i n e E G F [ 15 ]) . W i t h b o v i n el y s o z y m e , h o w e v e r , i n c re a s in g t h e c o p y n u m b e r f r o mo n e t o t h r e e r e d u c e d t h e l e v e l o f se c r e te d p r o d u c t[ 1 6] , a n d f o r H I V g p 1 2 0 , a c o p y n u m b e r o f g r e a t e rt h a n o n e r e d u c e d s e c r e t i o n a n d i n c r e a s e d a c c u m u l a t i o no f i n tr ace l lu l a r p roduc t s [ 6] . I t wo u ld ap pe a r t h a t l e sse f f i c i en t l y s ec r e t ed p ro t e in s a r e l i ke ly t o b l ock t hesec r e to ry pa thway a t h ighe r exp re s s ion l e ve l s .

Seve ra l f o r e ign p ro t e in s have been e f f i c i en t l y s e c r e t e dus ing t he i r na t i ve s igna l pep t i de ( e. g. H S A, w h e r e as t ra i n con t a in in g t h r ee cop i e s o f t he gene , ex p r e s s e d f ro ma m o d i f i e dA O X 2 p romote r, gave a s t agge r in g y i e ld o f10 g1 -1 [4 ,P4 ] ). W i th bac t e r i a l a - amy la se , h o w e v e r, t hey i e ld u s ing a yea s t s i gna l was two - t o t h r e e fo ld h igh e r[17] , and in genera l , yeas t s igna ls a re more l ike ly tob e s u c c e s sf u l [2 ]. S e c r e t i o n o f E G F a n d r o u t i n e E G Fus ing t he S. cerevisiaea - f a c t o r l e a d e r w a s s h o w n t o b eh igh ly e f f i c i en t , and ana ly s is o f t he p ro d u c t sh o w eda u t h e n t i c p r o c e s s i n g a t t h e K e x 2 e n d o p e p t i d a s e c l e a v a g es it e [ 1 5 ,P 2 ] . T h e v e c t o r s p P I C 9 a n d p P I C 9 K , w h i c hc o n t a i n t h e c t - fa c t o r l e a d e r s e q u e n c e w i t h c o n v e n i e n tc loning s i tes [12] , have recent ly become avai lab lef r o m I n v i t r o g e n a n d a r e n o w w i d e l y u s e d . R e c e n tsucce s s fu l exam p le s u s ing t he o r- f ac to r l e ad e r i n c lu d e t h efo l l owing : s i ng l e - cha in Fv an t i body f r agm e n t s [ 1 8 ] ; a9 k D a t h r o m b o m o d u l i n f r a g m e n t [ 19 ]; b l o o d fa c t o r X I I[19 ]; a f r agm en t o f am y lo id ~ -p ro t e in [20 ]; on co s t a t i n M( S J M c A n d r e wet al., abs t r ac t , Cu r r en t Top i c sn G e n eExpre s s ion Sys t ems : P i ch i a pa s to r i s , San D ieg o , U S A,O c t o b e r 1 9 9 4 ) c o f f e e - b e a n c t -g a la c to s id a s e (A Z h u , L FKimb a l l , Cu r r e n t To p i c s i n Ge ne Exp re s s i o n S y s t e ms :P i c h i a p a s t o r i s , S a n D i e g o , U S A , O c t o b e r 1 9 9 4 ) , a n dca theps in B ( JS M or tet al., abs t r ac t , Cu r r en t To p i c s i nGene Exp re s s ion Sys t ems : P i ch i a pa s to r i s , San D ieg o ,O c to be r 1 994 ) . An a l te rna t ive s i gnal s e q uen c e , t ha t o ft h e Ih chiaac id phospha t a se 1( P H 0 1 ) gene , i s used in thevec to r pHIL-S1 , wh ich i s a l so ava i l ab l e f r om I nv i t r o ge n .

Studies us ing S .cerevisiaehave shown tha t t h e p a r t i cu l a ryea s t s i gnal pep t i d e u sed can a f f ec t e f f i c i en cy o f s ec r e t i o n[2 ] . Recen t l y, a syn the t i c hyb r id s i gna l b a s e d on Ph o l ,w i th an add i t i ona l 19 r e s idues , i nc lud ing a K e x2 c l e a vages it e , ha s be en fou nd t o im prove s ec r e t i o n o f t i cka n t i c o a g u l a n t p r o t e i n a n d s o m e o t h e r p r o t e i n s t w o - t oth r ee fo ld i n Pichia [5] . Even so, i t is not clear that

g e ne ra l r u l e s c on c e rn ing t he s i gn a l pep t i de can beap p l i ed t o an y p ro t e i n .

T h e p r o b l e m o f p r o t e o l y ti c i n st a b il i ty i n t h e m e d i u mh a s b e e n e n c o u n t e r e d w i t h s e ve r al p r o t e i n s s e c r e t ed f r o mPichia. I t has bee n see n in shake f lasks , bu t usua l ly appearst o b e f a r w o r s e i n f e r m e n t e r s , b e c a u s e o f th e h i g h e rc o n c e n t r a t i o n o f p r o t e a s e s o r th e d i f f e r en t m e d i u m u s e d .

T h r e e d i f f e r e n t ap p r oac hes have bee n u s e d succe s s fu l l y t oo v e rco me p r o t e o l y s i s [ 1 ] : ad d i n g a mi n o a c id o r pep t i des u p p l e m e n t s t o t h e g r o w t h m e d i u m , b u f f e r i n g t h e p Ho f t h e m e d i u m t o a v a l u e w h e r e d e g r a d a t i o n i s r e d u c e d(e .g . pH 3) , o r us ing pro tea se-d ef ic ie n t hos t s t ra ins . Int h e c a se o f I G F - 1 , n o p r o t e i n w a s p r o d u c e d u n l es s t h em e d i u m w a s b u f fe r e d to p H 3 , b u t a 5 0 % i n c r ea s e w a st h e n a ch i ev e d u s in g ap e p 4 s tr a in g r o w n i n a m e d i u mbuffe red to p H 3 [P3].

S. cerevisiaeha s bee n a vo id e d a s a ho s t f o r t he de -v e l o p m e n t o f h u m a n t h e r a p e u ti c g l y c o p r o t e i n s b e c a u sey eas t - de r i ved g ly c op ro t e i n s a r e a n t i g en i c and f r equen t l yh y pe rg l yc osy l a t e d ( i . e . t h ey co n t a in ex t ens ive ou t e r

c ha i n man no s e u n i t s [ 5 0 - 1 50 r e s i du es ] t ha t c an maskf unc t i o n ) . Pichia-derived i nv e r t a s e i s , h oweve r, no th y p e rg l y c o s y l a t e d a n d h a s a n o u t e r - c h a i n l e n g t h o f8 - -1 4 un i t s , com pa r e d w i th >5 0 i nS. cerevisiae[21].Bulk P ich iag ly c op ro t e i n was f ou n d t o b e l e s s f r equen t l yhy pe rg ly c osy l a t e d , t o h a ve a sh o r t e r ou t e r- cha in l eng th(


6/7

53 Expression ystems

e l ec t ropo ra t i on , wh ich had been t hough t t o y i e ldonly s ingle-copy t ransformants . This grea t ly s impl i f iesthe r ap id i so l a ti on o f mu l t i copy c lones and ma y becons ide red t he m e th od o f cho i ce fo r gene ra l l abo ra to ryuse . In prac t ice , th is met ho d wo rks bes t us ing the KM 71s t ra in and Sad -d iges t ed vec to r because o f t he h ighe rt ransformat ion f requencies tha t a re a t ta inable . Wherevery h igh copy numbers (10-30) are requi red , however,i t then appears tha t t ransplacement u s ing the sphaeroplas tt r ans fo rma t ion me tho d mus t be u sed . Transfo rman t s cansubsequen tly be s c r eened fo r h igh copy num ber ' j ackpo tc lones ', e i t he r M ut s o r M ut + , u s ing a r ap id D NAdot -b lo t m e tho d [11 ]. P rec is e copy number s shou ld bede t e rm ined us ing quan t i ta t ive D N A do t b lo t s [3] .For secre t ion , in i t ia l s tudies could be car r ied out wi ths ingle-copy t ransformants , but i t would be preferableto s ta r t wi th a ser ies w i th d i fferent copy numb ers .Such a series could be created in several ways. A rangeo f m uk icop y t r ans fo rman t s , i so l a ted by G418- se l ec t ion[12 ] or D N A dot b lo t [11] , could be analyzed forcopy number [3 ] . An a l t e rna t i ve me thod deve loped by

SIBIA is to u t i lize a p lasmid tha t can be used to genera temul t i -casse t te vec tors wi th up to e ight copiesin vitro[1 ,16] . This method has the d isadvantage tha t severa lD N A clonin g s teps are requi red , but the advantage tha tt he copy number i s de t e rmined be fo re t r ans fo rma t ion .An y o f t hese app roaches can be ca r r ied o u t i n e i t he rM ut + or Muts s t ra ins for com par ison .An a l te rna t ive empir ica l approach to opt imiza t ionhas been used for t ick ant icoag ulant pro te in [5 ] .Transplacement was used to genera te a he terogeneouspoo l o f t r ans fo rman t s and t he l eve l o f s ec r e ted p rodu c tf rom 91 Mut s and M ut + c lones was t es t ed i n i nduc t ionsin microt i t re p la tes . Eleven t ransformants were fur ther

evalua ted in shake-f lask induct ions and one chosen forf e rmen te r op t imiza t i on , g iv ing a a f i na l p roduc t l eve l o f1 .7 g l - I o f p roduc t . Th i s m e tho d co u ld be u sed i n o the rcases where a s imple method exis t s for quant i ta t ing thep roduc t .Pichia t ransformants a re genera l ly tes ted in fermentersas soon as possible because shake-flask inductions aresub-opt ima l . M ut + and M ut s st ra ins each have the i radvantages : M ut + s t ra ins are less l ike ly to beco mepoiso ned by meth ano l , whereas M ut s s tra ins are lessl i kely t o becom e oxygen - l imi t ed . A l though a ma t t e ro f deba t e , r epo rt s ex i st o f i nduc t ion u s ing M ut s t ha tis equa lly rapid (48 h) as that u sing M ut + strains

[3 ,10 ,11 ,15] . Because the increase in y ie ld in goingfrom shake-f lask to fermenter induct ions i s not a lwayspredictable [11], i t may be prudent to select severalt r ansfo rman t s be fo re p roceed ing t o de t a i l ed f e rmen te ropt imiza t ion .

o n c l u s i o n s

The num ber o f d i f f e r en t p ro t e in s be ing exp res sed i nPichia i s expanding rapid ly, and resul t s f rom ongoing

s tudies wi l l increase our know ledge of the capabi li t iand l imi ta t ions of th is expression sys tem. The morw i d e a d o p t i o n o fPichia,however, w i l l requi re addi t ion alref inements (e .g . more auxot rophic and mutant s t ra ina l te rna t ive se lec t ion markers , a drug-se lec t ion sys tetha t shows g reat e r dose -dependence t han G 418 , vec tofo r s imukaneou s exp re s sion o f two p ro t ein s and a l t e rnt i ve p romote r s t oA O X 1 ) . These r e f i nemen t s , many o fwh ich should be avai lable shor t ly, wi l l enable theI~ chiasys t em to ga in some o f t he f l ex ib il it y o fS. cerevisiae.

Pichia i s a l ready widely accepted as an impor tanb io t echno log i ca l hos t o rgan i sm, and we a r e now a t t hexc i t i ng st age o f obse rv ing p roduc t s mov ing t h roug h c l in i ca l t ri al s and beyo nd . IG F-1 and H SA shou ld soobe m arke t ed p roduc t s fo r the t r ea tmen t o f amy lo t rophlateral sclerosis and as a serum replacement, respectiveNumerous cy tok ines , vacc ines and o the r b io log i cp roduc t s a r e unde r deve lopmen t , and a Cuban g rouhas developed a hepat i t i s B vaccine tha t i s cur rentbe ing so ld i n Sou th Amer i ca .

Acknowledgements

I wo uld l ike to than k m y col leagues in the f ie ld especia l ly JC la r e Ko t i S r eek ri shna J im Cregg R ich Buckho l z Dav id Hg ins Mick Hu n te r and Ben ne t Cohen fo r con t i nua l l y sha ri n f o r m a t i o n w i th m e .

References and recom mend ed reading

Papers o f par t i cu la r in te res t p ub l i shed wi th in the annua l pe r iodr e v i e w h a v e b e e n h i g h l ig h t e d a s : of special interest of outstanding interest

I . Cregg JM Vedvick TS Raschke WC:R e c e n t a d v a n c e s i nthee x p r e s s i o nof fo re igng e n e s i nPichia pastoris Biotechnolog1993 11 :905-910 .

2. Romanos MA Scorer CA Clare IJ:F o r e ig n g e n e e x p r e s s io n y e a s t : a r e v i e wYeast1992 8 :423-488 .

3. Clare Jl Rayment FB Ballantine SP Sreekrishna K Rom anMA: High-lev el expression oft e t a n u s t o x i n f r a g m e n t C Pichia pastoriss t ra in s c o n t a i n i n g m u l t ip l e t a n d e m i n t e go f t h e g e n e giotechnology1991 9 :455-460 .

4. Barr KA Hopkins SA Sreekrishna K:P r o t o c o l f o r e f f i c i es e c r e t i o no f H S A d e v e l o p e d f r o mPichia pastoris Ph arm En1992 12 :48-51 .

5. Laroche Y Storme V De Meutter J Messens l Lauwereys - High-levelsecretion and very efficient isotopic lab ell ing of ti

anticoagulant peptide TAP) expressed in the methylotrophy e a s t Pichia pastoris Biotechnology1994 12:1119-1124.

A rapid m icroti tre plate expression screen is employ ed to obta in sevefficient TAP-secret ing transformants from a po ol for further optim izaThe authors achieved very high level secret ion using a hybrid Pho l sipep tide containing a Ke x2 cleavage si te . Also reports an eff icient metfor ~SN and 13C isotopic labe ll ing of a protein that can then be usedete rmine the solution structure by N MR.

6. Scorer CA Buckholz RG Clare JJ Romanos MA :T h ei n t r a c e l l u l a r p r o d u c t i o n a n d s e c r e t i o nof HIV-1 e n v e l o p ep r o t e i n i n th e m e t h y l o t r o p h i c y e a s tPichia pastoris Gene1993136:111-119.


7/7

7. Cregg JM, Ma dden KN: Dev elopm ent of transformation systemsa n d cons t ruc t ion of methanol -u t i li sa t ion-defec t ive mutants ofPichia pastori by gene disruption. InBiological Resea rch onIndustrial Yeastvol 2. E dited by S tewart GG, Russell I , KleinRD, H iebsch RR. Boca Raton: CRC Press; 1987:1-18.

8. Cregg JM, Tscho pp JF, Stillman C, Sie gel R, Akong M, CraigWS, Buckholz RG, Madden KR, Kellaris PA, Davies GRet al.:High-level expression and efficient assembly of hepati t is Bsurface an t igen in the methylo t rophic yeas tPichia pastori ,Biotechnology1987, 5:479-485.

9. Sr ee kris hn a K, Nelles L, Potenz R, Cruze J, Mazz aferro P,Fish W, Mo tohir o F, Holde n K, Ph elps D, W ood P, ParkerK: High-level expression, purification, and characterisationo f r e c o m b i n a n th u m a n t u m o u r necrosis factor synthesisedin the methylo t rophic yeas tPichia pastorb. Biochemistry,28:411 7-412 5.

10 . Sreekr ishnaK, Potenz RB, Cruze JA, McCombie WR, ParkerKA, Nelles L, Mazzaferro PK, Holden KA, Harrison RG, WoodPJ et al.: High level expression of heterologous proteinsinmethylo t rophlc yeas tPichia pastori~ ] Basic Mic robio l1988,28:265-278.

11 . R o m a n o sMA , Clare JJ, Bee sley KM, Ray me nt FB, 8allantineSP, Makoff AJ, Dougan G, Faitweather NF, Charles IG:Recombinant Bordetella pertussiper tac t in (P69) f rom theyeast Pichia pastoris:,high-level production and immunologicalproperties. Vaccine 1991, 9:901-906.

1 2 . S c o r e rCA, C lare JJ, Mc Com bie WR, Romanos MA, Sreekrishna K: Rapid selection using G418 of high cofly numbe r

transformants ofPichia pastori for h igh- leve l fore igng e n eexpression. Biotechnology1994, 12:181 I 84.

Polylinker vectors for intracellular expression (pPIC3 and pPIC3K) andn-factor secretion vectors (pPIC 9 and pPIC 9K) are constructed w ithand without the kanamycin-resistance marker for G418-selection. Theauthors describe a rapid method for selecting multicopy transformantsusing G418-selection an d electroporation. They analyze H IVEnvm R N Alevels from strains with 1-12 copies of the expression vector and showa progressive increase with copy number.

13. Ho vlan d P, Flick J, Johnston M, Sc lafani RA: Galactose as agratuitous induc er ofGALgene expression in yeasts grow ingon glucose. Gene 1989, 83:57-64.

14. Vedvick T, 8uc kholz RG, Engel M, Urcan M, Kinney J, ProvowS, Siegel RS, Thill GP: H igh-level secretion of biolog icallyac t ive apro t in ln f rom the yeas tPichia pastori . J Ind Micro biol1991, 7 :197-201.

15. Clare JJ, Romanos MA, R aym ent F8, Rowedder JE, Smith MA ,Payne MM , Sreekrishna K, Henw ood CA: Production of mouseepide rmal grow th fa ctor in yeast: high-level secretionusingPichia pastori strains containing multiple gene copies.Gene1991, 105:205-212.

16. Thil l GP, Davis GR, Sti llman C, Holtz G, Brierley R, Engel M,Buckho ltz R, Kinn ey J, Provow S, Vedvick T, Seigel RS: Positivea n d negative effects o f m ulticop y integrated expression vectorson protein expression inPichia pastori .In Proceedings of the6th International Symposium on Genetics of Microorganisms,vo l 2. Edited by Heslot H, Davies J, Florent J, Bobichon L,Durand G, Penasse L. Paris: Societe Francaise de M icrobiolog ie;1990:477-490.

17. Paifer E, Margolles E, Crem ata J, Mon tesino R, Herrera L, Delgad o J-M: Efficient expression and secretion of recom binant

a lpha amylase in Pichla pastorisusing two different signalsequences. Yeast1994 10:1415-1459.

Improved secretion efficienc y and product yield (from 0.9 g1-1 to 2.5 gl-])is achieved b y replacing the bacterial signal peptide sequence with thatfrom S.cerevisiae SU C2.

18. Ridder R, Schmitz R, Legay F, Gram H: Genera tion of rabbitmon oc lon al ntibody fragments from a combinatorial phage

H i g h le v e l e x p r e s s i o n ini c h i a p a s t o r i sR o m a n o s 5 3

disp lay l ibrary and the i r product ion in the yeas tPichiapastori . Biotechnology1995, 13:255-260.

The ct-factor leader vector, pPICg, is used to direct expression, al low ingsecretion of fu nctional antibody single-chain Fv fragment at >100mgl-1in P. pastoris.

19. Wh ite CE, Kempi NM , Komives EA: Exp ression of highlydisulfide bonded proteins inPichia pastoris. Structure1994,2 : 1 0 0 3 - 1 0 0 5 .

20. Van Nostrand WE, Schm aier AH , Neiditch BR, Sieg el RS,

Raschke WC, Sisodia SS, Wagner SL: Expression, purificationa n d characterisation of the Kunit-type proteinase inhib itordomain of the amyloid-pro te in precursor- l ikepro te in -2 .Biochim Biophys Acta1994, 1209:165-170.

21. Ts ch op p F, Sverlow G, Kosson R, Craig W, Grinn a L: High-leve lsecretion of glycosylated invertase in the methylotrophic yeastPichia pastoris. Biotechnology1987, 5 :1305-1308.

22. Grinna LS, Tschopp iF: Size distr ibution and general structuralfeatures of N-llnked oligosaccharides from the methylotrophicyeast, Pichia pastoris. Yeast1989, 5:107-115.

23. Trimble RB, Atkinson PH , Tscho pp JF, Tow nsend RR, Ma ley F:Structure of oligosaccharides on.Saccharomyce SUC 2invertasesecreted by the methylotrophic yeastPichia pastoris. ] BiolChem 1991, 266:22807-22817.

24. Rodriguez M, R ubiera R, Penichet M, Montesinos R, Cremata

J , De La Fuen te J: High level expression of the B.microplus B m 8 6 a n t i g e n i nthe yeast Pichia pastorisforminghighly immunogenic particles for catt le.] Biotechno/1994,3 3 : 1 3 5 - 1 4 6 .

The 8m 86 m embrane glycoprotein from gut e pithelial cells of the catt let ick is expressed nPichiausing the invertase signal peptide. The produc tis purified as 17 -45 nm particles that are able to partial ly protect catt lein immunization experiments.

P a t e n t s

of special interest * of outstanding interest

P1. Sreek rishna K, Prevatt WD , T hill GP, Davis GR, Koutz P,Barr KA, Hopkins SA: Production ofBacillusentomotoxins inmethy lo t rophic yeas t . 1993 EP0586892A1.

P 2 . S i e g e lRS, Buckholz RG, Thill GP, Wondrack LM: Productionof ep iderm al growth fac tor in methylo t rophic yeas t ce ll s . 1990W O 9 0 / 1 0 6 9 7 .

P3. Brierley RA, Davis GR, Holtz GC: Produ ction ofinsulin-l ike grow th factor-1 in me thylo lroph ic yeast cells . 1994US5,324,639.

These authors use in vitrol igation to generate a multicopy o.-factorexpression vector for IGF-1 secretion. Secretion levels increase pro-gressively with copy number, up to six copies. Because of proteo)ysis,no product is detected, unless the induction me dium is buffered to pH 3.A further 50 increase is observed when using a pep4 protease-deficientstrain.

P4. Miura M, Ishida, Y. Oi H, Murakami K, Nakagawa Y, Kawabe

H: Mutant A O X 2 promoter, microorganism carrying same,method of prepara t ion thereof and produ ct ion o f he tero logousprote in using such microorganism. 1992 EP92105201.5.

M Romanos Glaxo Wellcome Langley Cour t Beckenham KentBR3 3BS UK.

1995- Mike Romanos- Advances in the use of Pichia pastoris for high-level gene expression.pdf

Documents