Expressionsvektoren zur Anreicherung eines rekombinant hergestellten Proteins in unterschiedlichen Zellkompartimenten
Die vorliegende Erfindung betrifft einen Expressionsvektor (A), der dadurch gekennzeichnet ist, daß er mindestens zwei Kopien eines für ein gewünschtes Protein kodierenden Gens jeweils mit einem Promotor funktioneil verknüpft enthält, oder eine Zusammensetzung (B), die mindestens zwei Expressionsvektoren umfaßt, die jeweils mindestens eine Kopie eines für ein gewünschtes Protein kodierenden Gens mit einem Promotor funktionell verknüpft enthalten, wobei die einzelnen Genkopien dadurch gekennzeichnet sind, daß sie (a) für das gewünschte Protein (I) als Fusionsprotein mit einem gewünschten Lokalisierungssignal (II) kodieren; wobei (b) bei den einzelnen Genkopien der für (I) kodierende Anteil im wesentlichen identisch, der für (II) kodierende Anteil jedoch unterschiedlich ist, wobei nach Einschleusen des Expressionsvektors (A) oder der Zusammensetzung (B) in einen Wirt das gewünschte Protein in unterschiedliche Kompartimente des Wirts transportiert wird. Die vorliegende Erfindung betrifft auch eine Zusammensetzung (C), die mindestens einen in Piastiden lokalisierten und einen in Mitochondrien lokalisierten Expressionsvektor umfaßt, die jeweils mindestens eine Kopie eines für ein gewünschtes Protein kodierenden Gens mit einem Promotor funktionell verknüpft enthalten, wobei die einzelnen Expressionsvektoren dadurch gekennzeichnet sind, daß die einzelnen Genkopien im wesentlichen identisch sind und das gewünschte Protein in Piastiden und Mitochondrien des Wirts exprimiert wird. Die vorliegende Erfindung betrifft auch die Kombination der vorstehenden Ausführungsformen. Die erfindungsgemäßen Expressionsvektoren/ Zusammensetzungen finden Verwendung in einem Verfahren, das die Steigerung der Produktionsrate für rekombinante Proteine erlaubt.
Inzwischen werden die meisten Proteine von kommerziellem Interesse in pro- oder eukaryotischen Wirtsorganismen rekombinant hergestellt. Während die Expressionsraten in prokaryotischen Organismen, insbesondere in Bakterien, oft sehr hoch sind (bis zu 50% des Gesamtproteins), werden in eukaryotischen Zellen oft nur geringe Expressionsraten erzielt (deutlich unter 1% bis höchstens zu wenigen Prozenten im Vergleich zu dem löslichen Gesamtprotein). Somit ist die Expression eines gewünschten Proteins in Bakterien in vielen Fällen bevorzugt. Allerdings tritt hier
oft das Problem der Bildung von "inclusion bodies" (Einschlußkörpern) auf, in denen das überproduzierte Protein in denaturierter Form vorliegt. Die natürlichen Biosynthese-Apparate von Pflanzenzellen sind in der Lage, auch solche, z.T. komplexe, Proteine zu synthetisieren, die der Expression in prokaryotischen Zellen nicht zugänglich sind. Allerdings sind die Expressionsraten in Pflanzenzellen oft sehr gering und in verschiedenen Kompartimenten dieser Zellen auch unterschiedlich. Dies wird z.B. bei rekombinanten scFv-Antikörpem in transgenen Pflanzen hinsichtlich der Kompartimente Cytoplasma, Apoplast und endoplasmatisches Retikulum (ER) untersucht (Conrad und Fiedler, Plant Molecular Biology 38 (1998), 101-109). Es zeigte sich, daß eine cytoplasmatische Expression von scFv-Antikörpem kaum realisierbar war, nur in Einzelfällen konnten überhaupt Expressionsraten nachgewiesen werden, wobei diese allerdings extrem gering waren. Die subzelluläre Lokalisation im endoplasmatischen Retikulum (ER) führte dagegen sowohl in Samen von Tabak oder Arabidopsis (Conrad und Fiedler, supra) als auch in Kartoffelknollen (Artsaenko et al., Molecular Breeding 4 (1998), 313-319) zu einer effizienteren Produktion des Fremdproteins. Apoplastische Expression konnte ebenfalls nachgewiesen werden, wenn auch nur mit etwa einem Zehntel der Menge im Vergleich zur Lokalisation im ER.
Andererseits ist bekannt, daß in der Natur bestimmte Proteine in mehrere subzelluläre Kompartimente einer Zelle gleichzeitig sortiert werden. So wird z.B. in Erbsen die Glutathion-Reduktase aufgrund des Vorhandenseins einer einzigen Prosequenz sowohl in Chloroplasten als auch in Mitochondrien transportiert (Creissen et al., Plant Journal 8 (1995), 167-175). Bezüglich der rekombinanten Produktion von Proteinen, beispielsweise in Pflanzenzellen, ist allerdings zu betonen, daß ein bestimmtes Kompartiment einer Pflanzenzelle aus biologischen und physikalischen Gründen nur eine begrenzte Speicherfähigkeit für ein zu produzierendes Protein aufweist, da der notwendige physikalische Speicherplatz mit anderen Proteinen geteilt werden muß. Außerdem sind teilweise bei Proteinfaltung und -transport Chaperone und Rezeptoren beteiligt. Bei hohen Expressionsraten kann demzufolge eine Überlastung des Transportapparates oder der Speicherkapazität der Pflanzenzelie eintreten. Dies hat zur Folge, daß deren maximal mögliche Fremdproteinbiosynthese-Kapazität aufgrund der eingeschränkten Leistungsfähigkeit des Transkriptions- und Translationsapparates
nicht vollständig ausgenutzt wird. Schließlich gibt es in Pflanzenzellen neben dem Kemgenom und dessen Transkriptions- und Translationsapparat zwei weitere Genome mit (zumindest teilweise) eigenständigen Transkriptions- und Translationsapparaten, nämlich die der Piastiden und Mitochondrien. Erfolgt eine Integration des Fremdgens allerdings ausschließlich in eines dieser Genome, so wird nur ein Teil der maximal möglichen Fremdproteinbiosynthese-Kapazität des zellulären Stoffwechsels genutzt. Außerdem wird durch ein Expressionskonstrukt mit nur einem definierten Targetting-Signal nur das Speicherpotential eines subzellulären Kompartimentes genutzt. Die vorstehend beschriebenen Probleme hinsichtlich der Effizienz der rekombinanten Herstellung von Proteinen in Eukaryoten, insbesondere Pflanzen, wirdn bisher nicht bzw. nur wenig zufriedenstellend gelöst. Auch beschränkte sich die Fachliteratur bisher nur auf die Schilderung von Experimenten hinsichtlich der vergleichenden Bestimmung der Expressionsraten in einzelnen subzellulären Kompartimenten.
Somit liegt der Erfindung im wesentlichen das technische Problem zugrunde, Mittel zur Steigerung der Ausbeute von in Eukaryoten rekombinant hergestellten Fremdproteinen zur Verfügung zu stellen.
Die Lösung dieses technischen Problems wird durch die Bereitstellung der in den Patentansprüchen gekennzeichneten Ausführungsformen erreicht.
Die vorliegende Erfindung betrifft ein Verfahren, das die bisher im Stand der Technik vorhandenen Beschränkungen in der Effizienz der Fremdproteinbiosynthese in Eukaryoten, insbesondere in Pflanzenzellen, minimiert. So bietet sich für das "Molecular Farming" in transgenen Pflanzen durch die vorliegende Erfindung die Möglichkeit, wesentliche Erhöhungen der Expressionsraten fremder Proteine zu erreichen. Im Gegensatz zu den bisherigen Vorgehensweisen wird dabei das zu produzierende Protein nicht in ein einziges subzelluläres Kompartiment geleitet und nur das Speicherpotential dieses einen Kompartimentes genutzt, vielmehr werden in einer Ausführungsform zwei oder mehrere homologe Gene für die Expression eines definierten Proteins in die Wirtszelle eingebracht, wobei sich die dadurch kodierten Proteine darin unterscheiden, daß sie Targetting-Signale (Lokalisierungssignale) für
unterschiedliche subzelluläre Kompartimente enthalten. Beispielsweise kann ein diagnostischer scFv-Antikörper durch Fusion mit einem Signalpeptid in den Apoplasten, gleichzeitig aber auch durch Fusion mit einem anderen Signalpeptid und mit dem KDEL-Targetting-Signal im ER lokalisiert werden. Somit werden beispielsweise die Speicherpotentiale des ER und des Apoplasten gleichzeitig genutzt. Weiterhin können zusätzlich Gene eingebracht werden, die für vakuolär oder im Golgi- Apparat lokalisierte scFv-Antikörper kodieren. Somit können in diesem Ausführungsbeispiel der vorliegenden Erfindung die Speicherpotentiale von vier subzellulären Kompartimenten miteinander kombiniert werden. In einer alternativen Ausführungsform werden die Fremdproteine nicht durch nukleare Expression und Transport in Organelien, d.h. Piastiden oder Mitochondrien, produziert, sondern die entsprechenden Gene werden durch direkte Transformation von Piastiden oder Mitochondrien mittels geeigneter Vektoren in diese Organellen eingeschleust und das für das betreffende Protein kodierende Gen direkt in diesen Kompartimenten exprimiert. Beide Vorgehensweisen können kombiniert werden, d.h. in einer weiteren Ausführungsform der vorliegenden Erfindung wird das Speicherpotential von bestimmten Kompartimenten, z.B. von Piastiden oder Mitochondrien, dadurch effizienter ausgenutzt, daß sowohl eine Organellen- als auch eine Kerntransformation mit den entsprechenden Genen durchgeführt wird, also die direkte Proteinsynthese in Organellen mit der nuklearen Proteinsynthese und nachfolgendem Transport des Proteins in die Organellen, d.h. Piastiden und Mitochondrien, mittels geeigneter Signalsequenzen kombiniert wird. Mit der vorliegenden Erfindung können somit die einzelnen Speicherkapazitäten typischer Speicherkompartimente, wie z.B. der Piastiden, des endoplasmatischen Retikulums oder von Speichervakuolen, kombiniert und synergistisch ausgenutzt werden.
Somit betrifft eine Ausführungsform der vorliegenden Erfindung einen Expressionsvektor (A), der dadurch gekennzeichnet ist, daß er mindestens zwei Kopien eines für ein gewünschtes Protein kodierenden Gens jeweils mit einem Promotor funktionell verknüpft enthält, oder eine Zusammensetzung (B), die mindestens zwei Expressionsvektoren umfaßt, die jeweils mindestens eine Kopie eines für ein gewünschtes Protein kodierenden Gens mit einem Promotor funktionell verknüpft enthalten, wobei die einzelnen Genkopien dadurch gekennzeichnet sind,
daß sie (a) für das gewünschte Protein (I) als Fusionsprotein mit einem gewünschten Lokalisierungssignal (II) kodieren, wobei (b) bei den einzelnen Genkopien der für (I) kodierende Anteil im wesentlichen identisch, der (II) kodierende Anteil jedoch unterschiedlich ist, wobei nach Einschleusen des Expressionsvektors (A) oder der Zusammensetzung (B) in einen Wirt das gewünschte Protein in unterschiedliche Kompartimente des Wirts transportiert wird.
Eine alternative Ausführungsform der vorliegenden Erfindung betrifft eine Zusammensetzung (C), die mindestens einen in Piastiden lokalisierten und einen in Mitochondrien lokalisierten Expressionsvektor umfaßt, die jeweils mindestens eine Kopie eines für ein gewünschtes Protein kodierenden Gens mit einem Promotor funktionell verknüpft enthalten, wobei die einzelnen Expressionsvektoren weiter dadurch gekennzeichnet sind, daß die einzelnen Genkopien im wesentlichen identisch sind und das gewünschte Protein in den Mitochondrien und Piastiden des Wirts exprimiert wird.
In einer bevorzugten Ausführungsform der vorliegenden Erfindung sind die beiden vorstehenden Ausführungsformen kombiniert, d.h., es wird sowohl eine Organellen- als auch eine Kerntransformation mit den für das gewünschte Protein kodierenden Genen durchgeführt, also die direkte Proteinsynthese in Organellen, d.h. in Piastiden und Mitochondrien, mit der nuklearen Proteinsynthese und nachfolgendem Transport des Proteins in die Organellen durch die Verwendung von Expressionsvektoren, bei denen das gewünschte Protein mit geeigneten Signalsequenzen verknüpft ist, kombiniert. Vektoren bzw. Vektorsequenzen, die die Expression des gewünschten Proteins in Organellen, d.h. in Piastiden oder Mitochondrien, erlauben, sind dem Fachmann bekannt. Beispielsweise wird auf Svab et al., Proc. Natl. Acad. Sei. USA 87 (1990) 8526 - 8530; Khan and Maliga, Nature Biotechnology 1 (1999), 910 - 915; Sidorov et al., Plant Journal 19 ( 1999), 209 - 216 verwiesen.
Der hier gebrauchte Ausdruck "im wesentlich identisch" bezüglich der einzelnen Genkopien bedeutet, daß die davon kodierten Proteine zumindest über die gleiche biologische Aktivität verfügen. Die einzelnen Genkopien können z.B. Modifikationen enthalten, die für die Expression des Gens, beispielsweise in Abhängigkeit von der
Expression in bestimmten Organellen, von Vorteil sind; vgl. dazu auch den nachstehenden Abschnitt zu "gene silencing".
Verfahren zur Konstruktion der Expressionsvektoren, die für das gewünschte Gen in exprimierbarer Form enthalten, sind dem Fachmann bekannt und beispielsweise auch in gängigen Standardwerken beschrieben (vgl. z.B. Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, 2. Aufl. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY). Die erfindungsgemäßen Expressionsvektoren können auf einem Plasmid, Cosmid, Virus, Bacteriophagen oder einem anderen in der Gentechnik üblichen Vektor basieren. Diese Vektoren können weitere Funktionseinheiten besitzen, die eine Stabilisierung des Vektors im Wirtsorganismus bewirken. Ferner können bei Verwendung einer Pflanze als Wirtsorganismus "left border"- und "right border"- Sequenzen agrobakterieller T-DNA enthalten sein, wodurch eine stabile Integration in das Erbgut von Pflanzen ermöglicht wird. Ferner kann eine Terminationssequenz vorhanden sein, die der korrekten Beendigung der Transkription sowie der Addition einer Poly-A-Sequenz an das Transkript dient. Derartige Elemente sind in der Literatur beschrieben (vgl. Gielen et al., EMBO J. 8 (1989), 23-29) und beliebig austauschbar.
Die mit dem gewünschten Protein zu fusionierende Signalsequenz richtet sich nach dem Zellkompartiment, in das der Transport des Proteins erfolgen soll. Geeignete Signalsequenzen (Signalpeptide) und für diese kodierenden DNA-Sequenzen sowie Verfahren zur Verknüpfung mit dem gewünschten Protein, z.B. einem scFv-Antikörper, auf eine solche Weise, daß sowohl das Protein noch aktiv ist als auch der Transport in das gewünschte Zellkompartiment erfolgt, sind dem Fachmann bekannt. Beispielsweise wird auf das Signalpeptid der α-Amylase aus Gerste hinsichtlich des Apoplasten (Düring et al., Plant Journal 3 (1993), 587 - 598), auf die Kombination aus Maus-Signalpeptid und KDEL-ER-Retentionssignal hinsichtlich des ER (Artsaenko et al., supra), auf das Targetting einer Säuger-α-2,6-Sialyltransferase hinsichtlich des Golgi-Apparats (Wee et al., Plant Cell IV (1998), 1759 - 1768, auf das Vakuolen- Lokalisierungssignal einer vakuolären Chitinase aus Gurke hinsichtlich der Vakuolen (Neuhaus et al., Proc. Natl. Acad. Sei. USA 88 (1991), 10362 - 10366), auf das Ferredoxin-Transitpeptid hinsichtlich der Chloroplasten und Piastiden, und auf das Transitpeptid von Tryptophanyl-t RNA-Snthetase aus Hefe hinsichtlich der
O 01/73085 7
Mitochondrien (Schmitz and Lonsdale, Plant Cell 1 (1989), 783-791) verwiesen.
Für die Expression des für das gewünschte Protein kodierenden Gens geeignete Promotoren sind dem Fachmann bekannt und dazu zählen beispielsweise bei der Verwendung einer Pflanze als Wirtsorganismus der Cauliflower Mosaic Virus 35S Promotor (Odell et al., Nature 313 (1995), 810 - 812) der Agrobacterium tumefaciens Nopalin Synthase Promotor und der Mannopin Synthase Promotor (Harpster et al., Molecular and General Genetics 212 (1988), 182 - 190). Das für das gewünschte Protein kodierende Gen kann auch mit einem induzierbaren Promotor verknüpft sein, was z.B. die Steuerung der Synthese des gewünschten Proteins z.B. in einer transgenen Pflanze, zu einem gewünschten Zeitpunkt erlaubt, wie bei der nachstehend beschriebenen Nach-Ernte-Produktionstechnologie. Geeignete Promotoren sind dem Fachmann bekannt und dazu zählen beispielsweise der anaerob induzierbare Gap C4 Promotor aus Mais (Bülow et al., Molecular Plant- Microbe Interactions 12 (1999), 182 - 188), PR-Promotoren, wie L-Phenylalanin Ammonium Lyase-, Chalcon Synthase- und "Hydroxyproline rieh glycoprotein"- Promotoren induzierbar durch Ethylen (Ecker and Davies, Proc. Natl. Acad. Sei. USA 84 (1987), 5202 - 5210) und durch Dexamethason induzierbares chimäres Transkriptions-Induktionssystem (Kunkel et al., Nature Biotechnology 17 (1990), 916 - 918), IncW-Promotor aus Mais, induzierbar durch Sucrose oder D-Glucose (Chen et al., Proc. Natl. Acad. Sei. USA 96 (1999), 10512 - 10517. Ergänzend wird auf Datla et al., Biotechnology Annual Review 3 (1997), 269 - 290 und Gatz and Denk, Trends in Plant Science 3 (1998), 352 - 358 verwiesen.
Zur Herstellung der Expressionsvektoren zur Einführung in Wirtsorganismen, z.B. Pflanzen, stehen eine große Anzahl von Klonierungsvektoren zur Verfügung, die ein Replikationssignal für E.coli und ein Markergen zur Selektion transformierter Bakterienzellen enthalten. Beispiele für derartige Vektoren sind pBR322, pUC-Serien, M13mp-Serien, pACYC184, etc.. Die gewünschte Sequenz kann an einer passenden Restriktionsschnittstelle in den Vektor eingeführt werden. Der erhaltene Vektor wird für die Transformation von E.coli-Zellen verwendet. Transformierte E.coli-Zellen werden in einem geeigneten Medium gezüchtet, anschließend geerntet und lysiert. Der Vektor wird dann wiedergewonnen. Als Analysemethode zur Charakterisierung der
gewonnenen Vektor-DNA werden im allgemeinen Restriktionsanalysen, Gelelektrophoresen und weitere biochemisch-molekularbiologische Methoden eingesetzt. Nach jeder Manipulation kann die Vektor-DNA gespalten und gewonnene DNA- Fragmente können mit anderen DNA-Sequenzen verknüpft werden. Jede Vektor-DNA- Sequenz kann in den gleichen oder anderen Vektoren kloniert werden.
Für die Einführung der vorstehenden Expressionsvektoren in eine Pflanzzelle stehen eine Vielzahl von Techniken zur Verfügung. Diese Techniken umfassen die Transformation von Pflanzenzeilen mit T-DNA unter Verwendung von Agrobacterium tumefaciens oder Agrobacterium rhizogenes als Transformationsmittel, die Fusion von Protoplasten, die Injektion, die Elektroporation von DNA, die Einbringung von DNA mittels der biolistischen Methode sowie weitere Möglichkeiten.
Bei der Injektion und Elektroporation von DNA in Pflanzenzellen werden an sich keine speziellen Anforderungen an die verwendeten Vektoren gestellt. Es können einfache Plasmide, wie z.B. pUC-Derivate, verwendet werden. Sollen aber aus derartig transformierten Zellen ganze Pflanzen regeneriert werden, sollte ein selektierbarer Marker vorhanden sein. Geeignete selektierbare Marker sind dem Fachmann bekannt und dazu zählen beispielsweise das Neomycinphosphotransferase Il-Gen aus E. coli (Beck et al., Gene 19 (1982), 327 - 336), das Sulfonamid-Resistenzgen (EP-369637) und das Hygromycin-Resistenzgen (EP-186425). Je nach Einführungsmethode für die gewünschten Gene in die Pflanzenzelle können weitere DNA-Sequenzen erforderlich sein. Wird z.B. für die Transformation der Pflanzenzelle das Ti- oder Ri-Plasmid verwendet, so muß mindestens die rechte Begrenzung, häufig jedoch müssen die rechte und linke Begrenzung der Ti- und Ri-Plasmid T-DNA als Flankenbereich mit den einzuführenden Genen verbunden werden.
Werden für die Transformation Agrobakterien verwendet, muß die einzuführende DNA in spezielle Vektoren kloniert werden, und zwar entweder in einen intermediären Vektor oder in einen binären Vektor (vgl. dazu auch die nachstehenden Beispiele 1 bis 3). Die intermediären Vektoren können aufgrund von Sequenzen, die homolog zu Sequenzen in der T-DNA sind, durch homologe Rekombination in das Ti- oder Ri- Plasmid der Agrobakterien integriert werden. Dieses enthält außerdem die für den
Transfer der T-DNA notwendige vir-Region. Intermediäre Vektoren können nicht in Agrobakterien replizieren. Mittels eines Helferplasmids kann der intermediäre Vektor auf Agrobacterium tumefaciens übertragen werden. Binäre Vektoren können sowohl in E.coli als auch in Agrobakterien replizieren. Sie enthalten ein Selektionsmarker-Gen und einen Linker oder Polylinker, welche von der rechten und linken T-DNA Grenzregion eingerahmt werden. Sie können direkt in die Agrobakterien transformiert werden. Das als Wirtszelle dienende Agrobakterium sollte ein Plasmid, das eine vir- Region trägt, enthalten. Die vir-Region ist für den Transfer der T-DNA in die Pflanzenzelle notwendig. Zusätzliche T-DNA kann vorhanden sein. Das derartig transformierte Agrobakterium wird zur Transformation von Pflanzenzellen verwendet.
Für den Transfer der DNA in die Pflanzenzelle können Pflanzen-Expiantate zweckmäßigerweise mit Agrobacterium tumefaciens oder Agrobacterium rhizogenes cokultiviert werden. Aus dem infizierten Pflanzenmaterial (z.B. Blattstücke, Stengelsegmente, Wurzeln, aber auch Protoplasten oder Suspensions-kultivierte Pflanzenzellen) können dann in einem geeigneten Medium, welches Antibiotika oder Biozide zur Selektion transformierter Zellen enthalten kann, wieder ganze Pflanzen regeneriert werden. Die so erhaltenen Pflanzen können dann auf Anwesenheit der eingeführten DNA untersucht werden. Alternative Systeme zur Transformation von monokotylen Pflanzen sind die Transformation mittels des biolistischen Ansatzes, die elektrisch oder chemisch induzierte DNA-Aufnahme in Protoplasten, die Elektroporation von partiell permeabilisierten Zellen, die Makroinjektion von DNA in Blütenstände, die Mikroinjektion von DNA in Mikrosporen und Pro-Embryonen, die DNA-Aufnahme durch keimende Pollen und die DNA-Aufnahme in Embryonen durch Quellung (zur Übersicht: Potrykus, Physiol. Plant (1990), 269 - 273). Während die Transformation dikotyler Pflanzen über Ti-Plasmid-Vektorsysteme mit Hilfe von Agrobacterium tumefaciens wohl etabliert ist, weisen neuere Arbeiten darauf hin, daß auch monokotyle Pflanzen der Transformation mittels auf Agrobacterium basierender Vektoren sehr wohl zugänglich sind.
B e i d e r A n w e n d u n g d e r e r f i n d u n g s g e m ä ß e n Expressionsvektoren/Zusammensetzungen (Kombinationen) ist zu beachten, daß bei der Produktion therapeutischer Proteine die Homogenität des produzierten Proteins
nicht beeinträchtigt wird und nur solche subzellulären Kompartimente miteinander gekoppelt werden, die keine Beeinträchtigung der Qualität des produzierten Proteins herbeiführen. Andererseits ist bei diagnostischen Proteinen oder Proteinwerkstoffen hierhin a priori keine Beeinträchtigung gegeben, so daß sich das erfindungsgemäße Verfahren bzw. die erfindungsgemäßen Expressionsvektoren insbesondere für derartige Proteine mit niedrigeren wirtschaftlichen Gewinnspannen eignen, weil hierbei die Expressionsraten einen wesentlich höheren Einfluß auf die wirtschaftliche Realisierbarkeit haben als bei therapeutischen Proteinen. Dem Problem eines potentiellen "gene silencing" bei Vorhandensein mehrerer gleicher Gene, z.B. in einer transgenen Pflanzenzelle/einer transgenen Pflanze, kann der Fachmann dadurch entgegengewirken, daß z.B. bei mehreren Kern-kodierten Fremdgenen für dasselbe Protein der Einsatz der möglichen unterschiedlichen Codons für eine Aminosäure unter Berücksichtigung der in einem bestimmten Organismus bevorzugten Codons eine unterschiedliche Nucleinsäuresequenz für die verschiedenen Gene z.B. in Form synthetisch hergestellter Gene benutzt wird.
In einer bevorzugten Ausführungsform enthalten die erfindungsgemäßen Expressionsvektoren bzw. die diese umfassenden Zusammensetzungen (Kombinationen) Lokalisierungssignale für die Lokalisierung im endoplasmatischen Retikulum (ER), im Apoplasten, im Golgi-Apparat, in Piastiden, in Peroxisomen, in Mitochondrien und/oder in Vakuolen. Es wird auf vorstehende Ausführungen hinsichtlich Signaisequenzen (Signalpeptide) verwiesen. Besonders bevorzugt sind als Lokalisierungssignale das KDEL-ER-Targeting-Peptid, das Golgi-Lokalisierungssignal der ß-1 ,2-N-Acetylglucosamintransferase (GnTI), die kleine Untereinheit der Ribulose- Biphosphat-Carboxylase und/oder das vakuoläre Targeting-Signal SKNPIN.
Bei den in dem erfindungsgemäßen Verfahren nützlichen Eukaryoten handelt es sich vorzugsweise um Pflanzen. Dabei kann es sich prinzipiell um Pflanzen jeder beliebigen Pflanzenspezies handeln, d.h. sowohl monokotyle als auch dikotyle Pflanzen. Der hier verwendete Begriff "Pflanze" umfaßt auch Pilze, Moose und Algen. Bevorzugt handelt es sich um Nutzpflanzen, z.B. um Pflanzen wie Weizen, Gerste, Reis, Mais, Zuckerrübe, Zuckerrohr, Brassicaceen, Leguminosen, Tabak oder Kartoffel. Besonders bevorzugt als transgene Pflanzen sind Brassicaceen,
Leguminosen und Kartoffel. Die für die Expression des gewünschten Proteins gewünschten Pflanzenteile betreffen im Prinzip jedes beliebige Pflanzenteil, jedenfalls Vermehrungsmaterial und Ernteprodukte dieser Pflanzen, beispielsweise Früchte, Samen, Knollen, Wurzelstöcke, Sämlinge, Stecklinge etc.
Ferner betrifft die vorliegende Erfindung auch Pflanzensamen, eine Pflanzenzelle oder eine Pflanze, die einen erfindungsgemäßen Expressionsvektor oder eine erfindungsgemäße Zusammensetzung enthält.
Schließlich betrifft die vorliegende Erfindung ein Verfahren zur Steigerung der Produktionsrate eines gewünschten Proteins in einem Wirt, vorzugsweise einer Pflanze, das dadurch gekennzeichnet ist, daß zur simultanen Expression des für das gewünschte Protein kodierenden Gens in verschiedenen Kompartimenten des Wirts und/oder zum simultanen Transport des gewünschten Proteins in verschiedene Kompartimente des Wirts der Wirt mit dem erfindungsgemäßen Expressionsvektor und/oder der erfindungsgemäßen Zusammensetzung transformiert und das gewünschte Protein aus dem Wirt gewonnen wird. Geeignete Verfahren zur Gewinnung des Proteins und der am besten geeigneten Pflanzenteile in Abhängigkeit von der Lokalisierung des Proteins sind dem Fachmann bekannt und z.B. auch in den nachstehenden Beispielen 1 bis 3 beschrieben.
Die vorliegende Erfindung zeichnet sich dadurch aus, daß in einem Wirt, insbesondere einer Pflanze, eine hohe Expressionsrate eines gewünschten Proteins erreicht werden kann, ohne daß es zu physiologischen Störungen im Zellhaushalt und damit im Wachstum und der Lebensfähigkeit des Wirts kommt. Dies wird durch die Verwendung der vorstehenden Expressionsvektoren bzw. Zusammensetzungen (Kombinationen) erreicht, vor allem aber auch durch die Kombination der vorliegenden Erfindung hinsichtlich der multiplen subzellulären Lokalisation des gewünschten Proteins beispielsweise mit der Nach-Emte-Produktionstechnologie (vgl. dazu das nachstehende Beispiel 3), bei der die Expression des für das gewünschte Protein kodierenden Gens erst nach Ernte der Pflanze bzw. bestimmter Pflanzenteile erfolgt. Bei dieser durch den Anmelder entwickelten Technologie erfolgt die Induktion der Expression, beispielsweise die Induktion des zu aktivierenden Promotors, über
chemische Stimuli (z.B. Veränderung der Zusammensetzung der umgebenden Gasphase, vorzugsweise anaerobe Induktion, Vernebelung von Lösungen fester induzierender (Bio-)chemikaiien oder Verwendung flüchtiger Substanzen und deren gleichmäßige Verteilung im Reaktionsraum) oder Aufhebung der funktionellen Inhibition der Transkription des betreffenden Fremdgens im transgenen Wirtsorganismus (z.B. durch Insertion von transkriptionsinhibierenden Sequenzen zwischen Promotor und zu exprimierendem Gen, welche extern steuerbar entfernt oder inaktiviert werden können). Diese Vorgehensweise ist vor allem im Hinblick auf transgene Pflanzen von Interesse, die hierfür ein hervorragend geeignetes Bioreaktor-System darstellen, vor allem da der großflächige Anbau von Kulturpflanzen und die großtechnische Verarbeitung großer Pflanzenmengen, z.B. für die Öl - und Stärkegewinnung, etabliert sind. Solches, unter Einbezug der Nach-Emte-Produktions- Technologie beispielsweise für die Produktion hochwertiger Proteine oder anderer (bio-)chemischer Substanzen in großem Maßstab eröffnet neue (bio)medizinische und technologische Anwendungsgebiete. Außerdem kann bei dieser Vorgehensweise das Pflanzengewebe, beispielsweise beim Einsatz eines gasförmigen Induktionsstimulus, intakt und unzerstört bleiben, da es nicht zerkleinert werden muß. Ferner kommt es zu einer sehr gleichmäßigen und schnellen Verteilung des Induktionsstimulus um das Gewebe herum, wie auch zwischen den Zellen im Zellverband.
Somit ist eine bevorzugte Ausführungsform des erfindungsgemäßen Verfahrens dadurch gekennzeichnet, daß die Expression des für das gewünschte Protein kodierenden Gens erst nach der Ernte des Wirts erfolgt, wobei das für das gewünschte Protein kodierende Gen so auf den Expressionsvektoren inseriert ist, daß dessen Expression erst in Anwesenheit einer induzierenden Verbindung (Induktor) erfolgt und das Inkontaktbringen mit dem Induktor nach der Ernte des Wirts erfolgt. Geeignete Vorgehensweisen sind dem Fachmann bekannt. Der Induktor kann z.B. über die Veränderung der den Wirtsorganismus umgebenden Gasphase, eine Vernebelung einer Lösung eines Induktors oder ein aktives Überströmen mit einem flüchtigen Induktor erfolgen. Bei der Veränderung der Gasphase kann es sich z.B. um einen Sauerstoffentzug handeln, wobei dann für die Expression ein unter anaeroben Bedingungen aktiver Promotor verwendet wird, z.B. der GapC4-Promotor aus Mais (Bülow et al., supra). Die Induktion der Expression des für das gewünschte Protein
kodierenden Gens kann auch durch Aufhebung der funktionellen Hemmung der Transkription und/oder Translation erfolgen, beispielsweise dadurch, daß bei den erfindungsgemäßen Expressionsvektoren zwischen dem Promotor und dem Gen eine Nucleinsäure inseriert ist, die dadurch gekennzeichnet ist, daß sie die Transkription und/oder Translation des Gens verhindert und nach Induktion exzisiert werden kann, was zur Expression des Gens führt. Diese Nucleinsäure kann z.B. eine durch eine induzierbare Rekombinase exzisierbare Nucleinsäure sein. Die exzisierbare Nucleinsäure und die Rekombinase können Bestandteile des Rekombinase-LBD- Systems (WO 95/00555) sein.
Die folgenden Beispiele erläutern die Erfindung.
Beispiel 1: Produktion von scFv(ox)-Antikörpern im ER, im Apoplasten, im Golgi-Apparat und in Piastiden von transgenen Kartoffelknollen
Die in Artsaenko et al. (supra) beschriebene cDNA, die für einen im endoplasmatischen Retikulum (ER) lokalisierten scFv(ox)-Antikörper mit KDEL-ER-Targetting-Sequenz kodiert, wird mittels einer Linker-Ligation (am 5'-Ende mit CATGCCATGGCATG [5'-phosphoryliertes Oligonucleotid]), am 3'-Ende mit GCTCTAGAGC [5'-phosphoryliertes Oligonucleotid] derart modifiziert, daß sie am 5'- Ende eine Ncol-Restriktionsschnittstelle und am 3'-Ende eine Xbal-Restriktionsschnittstelle aufweist. Nach Verdau des Plasmids pRT 100 (Töpfer et el. (1987) Nucleic Acids Research 15, 5890) mit Ncol und Xbal wird das scFv-kodierende DNA-Fragment in diesen Expressionsvektor inseriert. Es wird das Plasmid pRT 100/scFv(ox)ER erhalten. Nach Spaltung mit Hindill wird die Expressionskassette 35S-scfv(ox)ER Isoliert.
In einer weiteren Klonierung wird die cDNA für den scFv(ox)-Antikörper ohne die kodierende Sequenz für das KDEL-ER-Targetting-Signal (Artsaenko et al., (supra)) mittels einer Linker-Ligation (am 5'-Ende mit CATGCCATGGCATG [5'- phosphoryliertes Oligonucleotid]; am 3'-Ende mit GCTCTAGAGC [5'-phosphoryliertes Oligonucleotid]) derart modifiziert, daß sie am 5'-Ende eine Ncol-Restriktionsschnittstelle und am 3'-Ende eine Xbal-Restriktionsschnittstelle
aufweist. Nach Verdau des Plasmids pRT 100 mit Ncol und Xbal wird das scFv-kodierende DNA-Fragment in diesen Expressionsvektor inseriert. Es wird das Plasmid pRT 100/scFv(ox)AP erhalten. Nach Spaltung mit Hindlll wird die Expressionskassette 35S-scFv(ox)AP isoliert.
Für die Lokalisierung des scFv(ox)-Antikörpers im Golgi-Apparat wird das Signalpeptid in Plasmid pRT 100/scFv(ox)AP durch das Golgi-Lokalisierunggssignal der ß-1 ,2-N- Acetylglucosaminyltransferase (GnTI) aus Kaninchen (Burke et el., J.Biol.Chem. 267 (1992), 24433-24440) ersetzt. Dazu wird mittels einer PCR-Reaktion die dafür kodierende DNA-Sequenz so modifiziert, daß sie am 5'-Ende eine Ncol-Schnittstelle und am 3'-Ende die GTCGAC-Sequenz aufweist, für die eine Sall-Schnittstelle codiert. Für die PCR-Reaktion wird das folgende Primerpaar verwendet:
5'-GnTI-Primer: CCATGGATGCTGAAGAAGCAGTCTGCTGG 3'-GnTI-Primer: GTCGACACGTGTCCAGAAGAAGAGGAGGAG
Die cDNA für den reifen scFv(ox)-Antikörper wird mittels einer Linker-Ligation (am 5'-Ende mit CCGTCGACAT [5'-phosphoryliertes Oligonucleotid] und am 3'-Ende mit GCTCTAGAGC [5'-phosphoryliertes Oligonucleotid]) derart modifiziert, daß sie am 5'-Ende eine Sall-Schnittstelle und am 3'-Ende eine Xbal-Restriktionsschnittstellle aufweist. Beide cDNA-Fragmente werden mit Ncol + Sall bzw. Sall + Xbal verdaut, so daß überhängende Enden entstehen. Die erhaltenen cDNA-Fragmente werden gleichzeitig in den mit Ncol und Xbal geöffneten Vektor pRT 100 inseriert. Es wird das Plasmid pRT 100/scFv(ox)GO erhalten. Nach Spaltung mit Hindlll wird die Expressionskassette 35S-scFv(ox)GO isoliert.
Für den Transport des cytopiasmatisch synthetisierten scFv(ox)-Antikörpers in Chloroplasten wird eine für ein Transitpeptid-scFv(ox) kodierendes cDNA-Fragment erzeugt. Dazu wird die cDNA für das Transitpeptid aus dem für die kleine Untereinheit der Ribulose-Bisphosphat-Carboxylase kodierenden Gens (Anderson et al., Biochemical Journal 240 (1986), 709-715) mittels einer PCR-Peaktion so modifiziert, daß am 5'-Ende eine Ncol-Schnittstelle und am 3'-Ende die GTCGAC-Sequenz, die für eine Sall-Schnittstelle kodiert, eingefügt werden. Dafür wird das folgende
Primerpaar verwendet:
5'-rbcT Primer: CCATGGCTTCTATGATATCCTCTTCAG 3'-rbcT Primer: GTCGACGCACTTTACTCTTCCACCATTGC
Die cDNA für den reifen scFv(ox)-Antikörper wird mittels einer Linker-Ligation am 5'-Ende mit CCGTCGACAT [5'-phosphoryliertes Oligonucleotid] und am 3'-Ende mit GCTCTAGAGC [5'-phosphoryliertes Oligonucleotid]) derart modifiziert, daß sie am 5' Ende eine Sall-Schnittstelle und am 3'-Ende eine Xbal-Restriktionsschnittstelle aufweist. Beide cDNA-Fragmente werden mit Ncol + Sall bzw. Sall + Xbal verdaut, so daß überhängende Enden entstehen. Die erhaltenen cDNA-Fragmente werden gleichzeitig in den mit Ncol und Xbal geöffneten Vektor pRT 100 inseriert. Es wird das Plasmid pRT 100/scFv(ox)CH erhalten. Nach Spaltung mit Hindlll wird die Expressionskasette 35S-scFv(ox)CH isoliert.
Der binäre Vektor pSR 8-30 (Düring et al., Plant Journal 3 (1993), 587-589; Porsch et al., Plant Molecular Biology 37 (1998), 581-585) wird mit Hindlll geöffnet. Alle vier vorstehend beschriebenen Expressionskassetten (35S-scFv(ox)ER, 35S-scFv(ox)AP, 35S-scFv(ox)GO" und 35S-scFv(ox)CH) werden gleichzeitig In die Hindlll-Schnittstelle des binären Vektors ligiert Die eindeutige Identifizierung positiver Klone, die alle vier Fragmente enthalten, erfolgt durch DNA-Sequenzierung. Es wird der binäre Vektor pSR 8-30/scFv(ox)EAGC erhalten.
Der Expressionsvektor pSR 8-30/scFv(ox)EAGC wird zur Transformation von E.coli SM10 verwendet. Transformanten werden mit Agrobacterium GV 3101 gemischt und über Nacht bei 28°C inkubiert (Koncz und Scheli, Molecular and General Genetics 204 (1986), 383-396; Koncz et al., Proc.Natl.Acad.Sci.USA 84 (1987), 131 - 135). Es wird auf Carbenicillin selektioniert, wobei das hierfür notwendige bla-Gen in den vorstehenden Expresssionsvektoren vorliegt. Selektionsklone von Agrobacterium tumefaciens werden auf abgeschnittenen und mehrfach an der Mittelrippe eingeritzten Blättern der Kartoffelpflanze cv. Desiree aufgebracht und die Blätter werden 2 Tage bei 20°C im Dunkeln inkubiert. Danach werden die Agrobakterien abgewaschen und den Kartoffelblättern Pflanzenwuchsstoffe zugesetzt, so daß bevorzugt Sprosse
lo
regenerieren. Ferner werden durch die Zugabe von Kanamycin 100 mg/l in das Pflanzenmedium nicht-transformierte Zellen in den Kartoffelblättern abgetötet. Heranwachsende Sprosse werden abgeschnitten und auf das Medium ohne Pflanzenwachstumsstoffe, aber mit Kanamycin, bewurzelt. Die weitere Kultivierung der Kartoffelpflanzen erfolgt in üblicher Weise.
Der Nachweis des exprimierten scFv-Antikörpers in transgenem Blatt- und Knollenmaterial wird durch Antikörper, die an scFv oder Protein L binden, im Western Blot bzw. ELISA erbracht. Hierzu wird das Gesamtprotein des Kartoffelmaterials isoliert und in die entsprechenden Nachweisverfahren eingesetzt. Im Vergleich zu den in Artsaenko et al. (supra) publizierten Ergebnissen zur Knollen-Expression von scFv(ox) ausschließlich im ER (Expressionsraten von bis zu 2% des löslichen Gesamtproteins) werden durch die simultane Expression in insgesamt vier Kompartimenten bis zu 5% scFv(ox)-Antikörper-Protein vom löslichen Gesamtprotein erreicht.
Beispiel 2: Produktion von scFv(ox)-Antikörpern durch nukleare
Transformation für die Expression im ER und in Vakuolen und plastidäre
Transformation für die Expression in Piastiden
Für die Herstellung eines Plastiden-Transformationsvektors wird Chloroplasten-DNA (cp-DNA) aus 8 Wochen alten Tabakpflanzen, angezogen im Gewächshaus, nach dem Verfahren von Koladner und Tewarl (Biochim. Biophys. Acta 404 (1975), 372-390) isoliert. Die isolierte Chloroplasten-DNA wird mit Bglll verdaut und die entstandenen Bglll- Fragmente werden "shot gun" in den mit BarnHI geöffneten Vektor pBluescriptKS (Stratagene) kloniert. Über Kolonie-Hybridisierung wird ein Klon identifiziert, der das Bglll-Fragment mit den ndhF-, rpl32 und trnL-Genen enthält (Shinozaki et al., EMBO J. 5 (1986), 2043-2049; 4.656bp-Fragment, Nucieotide Nr. 111515-116171). Es wird das Plasmid pBKSnrt erhalten.
In die Hincll-Schnittstelle des 4.656bp-Bglll-Fragments im Plasmid pBKSnrt wird das Gen für den scFv(ox)-Antikörper aus Artsaenka et al. (supra) unter die Kontrolle des
psbA-Promotors und des psbA-Terminationssignals (Shinozaki et al., supra) aus Tabak kloniert. Dafür wird die cDNA für den reifen scFv(ox)-Antikörper mittels einer Linker-Ligation am 5'-Ende mit CGCGAATTC [5'-phosphoryliertes Oligonucleotid] und am 3'-Ende mit GACTAAGCTT [5'-phosphoryliertes Oligonucleotid] derart modifiziert, daß sie am 5'-Ende eine EcoRI-Schnittstelle und am 3'-Ende eine Hindill- Restriktionsschnittsteile aufweist. Dieses cDNA-Fragment wird nach Verdau mit EcoRI und Hindlll in das Plasmid pBluescriptSK (Stratagene) kloniert. Es wird das Plasmid pBSK/scFv(ox) erhalten. Der 5'-Bereich des psbA-Gens (Plastiden-spezifischer Promotor) wird aus Chloroplasten-DNA von Tabak durch eine PCR-Reaktion amplifiziert. Dafür wird folgendes phosphoryliertes Primerpaar verwendet:
psbA1 Primer: GTCATGTTATACTGTTG
(Nucleotide 1700-1684, Shinozaki et al., supra) psbA2 Primer: GGTAAAATCTTGGTTTATTTAATCATC
(Nucleotide 1596-1622, Shinozaki et al., supra)
Das PCR-Produkt wird "blunt end" in die Smal-Schnittstetle des Vektors pBSK/scFv(ox) kloniert. Es wird das Plasmid pBSK/psbA-scFv(ox) erhalten. Der 3'-Bereich des psbA-Gens (Terminationssignal) wird ebenfalls über eine PCR-Reaktion amplifiziert. Dazu wird folgendes phosphoryliertes Primerpaar verwendet:
psbA3 Primer: AGTCTATGTAAGTAAAATAC
(Nucleotide 401-420, Shinozaki et al., supra) psbA4 Primer: CCTGGCCTAGTCTATAGGAG
(Nucleotide 530-511 , Shinozaki et al., supra)
Das PCR-Produkt wird "blunt end" in die aufgeschnittene und aufgefüllte Sall-Schnittstelle des Vektors pBSK/psbA-scFv(ox) inseriert. Es wird das Plasmid pBSK/psbA-scFv(ox)-PgbA erhalten. Die Expressionskassette wird aus diesem Plasmid durch Verdau mit BamHI und Kpnl herausgeschnitten, aufgefüllt und "blunt end' in die Hincll-Schnittstelle des Vektors pBSKnrt kloniert. Es wird das Plasmid pBSKnnVscFv(ox) erhalten. In die SnaBI-Schnittstelle des 4.656bp-Bglll-Fragments des
1 o
Plasmids pBSKnnVscFv(ox) wird weiterhin als Selektionsmarker das nptll-Gen unter die Kontrolle des psbA-Promotors und des psbA-Terminationssignals aus Tabak kloniert. Dafür wird das nptll-Gen mit einer PCR-Amplifizierung aus dem Vektor pCR2.1 (Invitrogen) isoliert. Dazu wird folgendes Primerpaar verwendet:
Kan1 -Primer: CCGAATTCCAAGAGACAGGATGAGGATC
Kan2-Primer: CGCGAAGCTTCAATTCAGAAGAACTCAAG
Das PCR-Produkt wird über EcoRI- und Hindlll-Verdau in das Plasmid pBlueschptSK (Stratagene) kloniert. Es wird das Plasmid pBSK/NPT erhalten. Der 5'-Bereich des psbA-Gens (wie vorstehend beschrieben isoliert) wird "blunt end' in die Smal-Schnittstelle des Vektors pBSK/NPT kloniert. Es wird das Plasmid pF3SK/psbA-NPT erhalten. Der 3'-Bereich des pshA-Gens (Terminationssignal) wird (wie vorstehend beschrieben) ebenfalls "blunt end' in die aufgeschnittene und aufgefüllte Sall-Schnittstelle des Vektors pBSK/psbA-NBT inseriert, wobei das Plasmid pBSK/psbA-NPT-psb erhalten wird. Die Expressionskassette wird aus diesem Plasmid mit BamHI und Kpnl herausgeschnitten, aufgefüllt und "blunt end" in die SnaBI- Schnittstelle des Vektors pBKSnrt/scFv(ox) kloniert, wodurch das Plasmid pBKSnrt/scFv(ox)-NPT erhalten wird.
Die Transformation von Kartoffelplastiden und die Selektion von transgenen Kartoffelpflanzen, die das vorstehend beschriebene Konstrukt in das Piastidengenom eingebaut haben, erfolgt nach den von Svab et al., supra, Carrer et al. (Mol.Gen.Genet. 241 (1993), 49-56) und Sidorov et al. (Plant J. (1999), supra beschriebenen Protokollen (biolistischer DNA-Transfer) mit dem Helium-betriebenen "Particle-Gun" Gerät PDS-1000/He (Du Pont Willmington, USA). Der Nachweis der Piastidentransformation erfolgt durch Isolierung plastidärer DNA und Hybridisierung mit dem scFv(ox)-Gen. Der Nachweis des produzierten scFv(ox)-Antikörpers erfolgt analog zu vorstehendem Beispiel 1.
Die vorstehend beschriebene transgene Kartoffellinie, die das Gen für die plastidäre Expression des scFv(ox)-Antikörpers in den Piastiden enthält, wird weiterhin mit den beiden Expressionskassetten 35S-scFv(ox)ER aus Beispiel 1 und einer weiteren
Expressionskassette 35S-scFv(ox)VA für die vakuoläre Lokaiisation des produzierten scFv(ox)-Antikörpers transformiert.
Für die Klonierung der Expressionskassette 35S-scFv(ox)VA wird das Plasmid pSR9-12 ein Derivat des Plasmids pRT 100, welches stromabwärts vom CaMV 35S- Promotor die kodierende Sequenz für ein Maus IgG-Signalpeptid enthält, mit Sall und Xbal geöffnet. Die cDNA für den reifen scFv(ox)-Antikörper wird mittels einer Linker-Ligation (am 5'-Ende mit CCGTCGACTCTAAGAACCCAATTAAC [5'- phosphoryliertes Oligonucleotid) und am 3'-Ende mit GCTCTAGAGC [5'- phosphoryliertes Oligonucleotid]) derart modifiziert, daß sie am 5'-Ende eine Sall-Schnittstelle und am 3'-Ende Ende eine Xbal-Restriktionsschnittstelle aufweist. Gleichzeitig wird damit die für das vakuoläre Targetting-Signal SKNPIN aus dem 20 kDa-Protein aus der Kartoffelknolle kodierende DNA-Sequenz (PT20; Kolde, Plant Cell Physiol. 40 (1999), 1152-1159) integriert. Dieses cDNA-Fragment wird mit Sall und Xbal verdaut, so daß überhängende Enden entstehen, und in den geöffneten Vektor inseriert. Es wird das Plasmid pSR9-12/scFv(ox)VA erhalten. Nach Spaltung mit Hindlll wird die Expressionskassette 35S-scFv(ox)VA isoliert.
Die beiden Expressionskassetten 35S-scFv(ox)ER und 35S-scFv(ox)VA werden gleichzeitig als Hindlll-Fragmente in den mit Hindlll geöffneten binären Vektor pSR 8 -30 hyg , e i n D e rivat d es Vekto rs p S R 8 -30 , d e r an stel le d es Neomycinphosphotransferare Il-Gens das Hygromycinphosphotransferare-Gen enthält, ligiert. Die eindeutige Identifizierung positiver Klone, die beide Fragmente enthalten, erfolgt durch DNA-Sequenzierung. Es wird der binäre Vektor pSR 8-30hyg/scFv(ox)EV erhalten. Die Kartoffeltransformation erfolgt wie im vorstehenden Beispiel 1 beschrieben, wobei die Selektion mit 5mg/l Hygromycin erfolgt. Der Nachweis des produzierten scFv(ox)Antikörpers erfolgt wie im vorstehenden Beispiel 1 beschrieben. Die Integration des vorstehend beschriebenen Konstrukts in das Kerngenom wird durch Isolierung kemgenomischer DNA und Hybridisierung mit dem scFv(ox)-Gen gemäß Standardverfahren nachgewiesen. Gegenüber Beispiel 1 kann die Menge des exprimierten scFv(ox) nochmals gesteigert werden, nämlich auf 8% scFv(ox)- Antikörper-Protein vom löslichen Gesamtprotein.
Beispiel 3: Nach-Ernte-Produktion von scFv(ox)-Antiköpern durch anaerobe Expression im ER und im Golgi-Apparat
Der anaerob induzierbare GapC4-Promotor aus Mais (DE 195 47 272) wird mittels einer PCR-Reaktion derart modifiziert, daß er am 5'-Ende eine Hincll- Restriktionsschnittstelle und am 3'-Ende eine Ncol-Restriktionschnittstelle erhält. Für die PCR-Reaktion wird folgendes Primerpaar verwendet:
Hincll-pGapC4 Primer: CATGTCAACACATAAGGAAGAAGAGGTAGAAAG pGapC4-Ncol Primer:
CATGCCATGGATCGATGACGGGGTTGGCGAGTGTG
Die in Artsaenko et al. (supra) beschriebene cDNA, die für den im endoplasmatischen Retikulum (ER) lokalisierten scFv(ox)-Antikörper kodiert, wird mittels einer Linker- Ligation am 5'-Ende mit CATGCCATGGCATG [5'-phosphoryliertes Oligonucleotid] und am 3'-Ende mit GCTCTAGAGC [5'-phosphoryliertes Oligonucleotid] derart modifiziert, daß sie am 5'-Ende eine Ncol-Restriktionsschnittstelle und am 3'-Ende eine Xbal-Restriktionsschnittstelle aufweist. Aus dem Plasmid pRT 100 wird der CaMV 35S-Promotor mittels Restriktionsverdau mit Hincll und Xbal entfernt. Stattdessen werden die beiden oben beschriebenen Nucleinsäurefragmente, die für den GapC4- Promotor und den scFv(ox)-Antikörper kodieren, inseriert. Es wird das Plasmid pRT 100GAP/scFv(ox)ER erhalten. Nach partieller Spaltung mit Hindlll wird die Expressionskassette GAP-scFv(ox)ER isoliert.
Für die anaerobe Expression des im Golgi-Apparat lokalisierten scFv(ox)-Antikörpers wird das Plasmid pRT 100/scFv(ox)GO aus Beispiel 1 mit Hincll und Ncol verdaut. Dadurch wird der CaMV-35S Promotor entfernt. Statt dessen wird das vorstehend beschriebene PCR-modifizierte pGapC4-Promotorfragment inseriert. Es wird das Plasmid pRT 100GAP/scFv(ox)GO erhalten. Nach Spaltung mit Hindlll wird die Expressionskassette GAP-scFv(ox)GO isoliert.
Der binäre Vektor pSR 8-30 wird mit Hindlll geöffnet. Beide vorstehend beschriebenen Expressionskassetten (GAP-scFv(ox)ER und GAP-scFv(ox)GO)
werden gleichzeitig in die Hindlll-Schnittstelle des binären Vektors ligiert. Die eindeutige Identifizierung positiver Klone, die beide Fragmente enthalten, erfolgt durch DNA-Sequenzierung. Es wird der binäre Vektor PSR 8-30GAP/scFv(ox)EG erhalten.
Der Expressionsvektor pSR 8-30GAP/scFv(ox)EG wird wie im vorstehenden Beispiel 1 beschrieben zur Transformation von Kartoffeln verwendet. Zum Nachweis der Expression des scFv-Antikörpers wird abgeschnittenes Blattmaterial bzw. intaktes (oder geschnittenes) Knollenmaterial mittels des "Anaeroculf-Systems (Fa. Merck, Darmstadt, Deutschland) wie bei Bülow et al. Molecular Plant-Microbe Interactions 12 (1999), 192-198 beschrieben induziert. Nach 40 Stunden wird das Blattmaterial entnommen und gemörsert. Der Nachweis des exprimierten scFv-Antikörpers wird wieder durch Antikörper, die an den scFv-Antikörper oder Protein L binden im Western Blot bzw. ELISA erbracht. Hierzu wird das Gesamtprotein des Kartoffelmaterials isoliert und in die entsprechenden Nachweisverfahren eingesetzt.
Als expressions-positiv ermittelte transgene Kartoffelpflanzen werden im Gewächshaus (im Topf oder im Erdbeet) oder im Freiland unter üblichen gartenbaulichen bzw. landwirtschaftlichen Bedingungen angebaut. Die Knollen werden nach üblicher Handhabung geerntet und gelagert. Für die Nach-Ernte-Produktion des scFv-Antikörpers werden die Knollen in einen Reaktionsbehälter aus Stahl (oder Kunststoff) verbracht, der unten ein Gaszufuhrventil und oben ein Gasabführventil aufwies. Die Raumluft im Behälter wird schnell durch Zufuhr von beispielsweise technischem Stickstoff (oder Kohlendioxid) verdrängt. Unter langsamem Luftstrom (1 m3 Gaszufuhr pro Stunde pro m2 Grundfläche) wird eine konstante Zusammensetzung der Gasphase im Reaktionsbehälter eingestellt. Nach 40 Stunden werden die Knollen aus dem Reaktionsbehälter entnommen, homogenisiert, der Festanteil wird abzentrifugiert und der wäßrige Überstand der chromatographischen Aufreinigung des scFv-Antikörpers zugeführt. Durch Vergleich von Proben vor und nach Sauerstoffentzug dieser transgenen Kartoffellinie mit den transgenen Kartoffellinien, die den scFv(ox)-Antikörper nur im ER exprimieren, zeigt sich, daß durch die kombinierte Expression in zwei Zellkompartimenten (hier: ER und Golgi-Apparat) deutlich höhere Expressionsraten erzielt werden.