In dieser Arbeit konnte gezeigt werden, dass die Transkription, Replikation und Verpackung (Passage) viraler Fremdsegmente über die Länge der größten Virus-eigenen Segmente (2341 nt) hinaus bis auf 4000 nt ausgedehnt werden kann. Transkriptions- und Replikations-Effizienz sinken jedoch mit zunehmender Länge ab. Mit Hilfe des RNA-Polymerase I-Transkriptionssystems konnte der Einfluss von gezielt gesetzten Mutationen in der cRNA-Promotorstruktur detektiert werden. Für die direkte Untersuchung des cRNA-Promotors wurde die DNA-Sequenz, die in dem RNA-Polymerase I-Transkriptionskonstrukt für das RNA-Segment kodiert, so verändert, dass nach in vivo Transfektion eine cRNA entsteht, die ihrerseits die Transkription einer Reportergen-mRNA initiieren kann. Nach Infektion bzw. Expression der viralen Polymerase und des NP-Proteins liess sich die Aktivität des Reportergens nachweisen. Somit wurde die Fähigkeit des cRNA-Promotors zum Transkriptionsstart erstmalig nachgewiesen. Zu diesem Zweck wurde das kodierende CAT-Gen in inverser Orientierung (als Minusstrang-Sequenz) eingesetzt und am 5´-cRNA-Promotorende zusätzlich eine Folge von 6 Uridin-Basen eingefügt, um die Polyadenylierung der mRNA zu ermöglichen. Basierend auf der Mutationsanalyse des vRNA-Promotors ist die komplementäre cRNA-Promotorsequenz in gleicher Weise durch Substitution und Doppelsubstitution untersucht worden. Die Mutationsanalyse des cRNA-Promotors zeigt, dass die aktive cRNA-Promotorstruktur, wie die des vRNA-Promotors, eine 'corkscrew'-Sekundärstruktur aufweist. Allerdings unterscheidet sich die Erkennungs- und Bindungs-Konformation der viralen RNA-Polymerase in der Interaktion mit einer optimierten cRNA-'corkscrew'-Struktur in einigen Einzelheiten von der Wechselwirkung mit der vRNA-'corkscrew'-Struktur. Zur Bestimmung aller basenspezifischen Erkennungselemente der viralen Polymerase bedarf es weiterer cRNA-Promotoranalysen. Die Analyse der 5´-Endsequenzen der aus der cRNA synthetisierten mRNA (c-mRNA) nach der RACE-Methode ergab, dass die 5´-Promotorsequenz durch Fremdsequenzen verlängert vorliegt, wie es für die aus der vRNA gebildete mRNA (v-mRNA) typisch ist. Auch der c-mRNA-Transkriptionsstart ist daher ein 'cap-snatching'-Prozess. Das unterschiedliche Verhalten der CAT-Expression der cRNA und vRNA bei den NP-Abhängigkeitsversuchen und die Ergebnisse der RACE-Analyse bestätigen, dass die in unserem System nach einer Infektion nachgewiesene CAT-Expression die Transkriptionsaktivität der cRNA- bzw. der vRNA-Moleküle widerspiegelt und nicht die Replikation. Nachdem gezeigt werden konnte, dass die cRNA außer der bekannten Replikations- eine Transkriptionsaktivität aufweist, war es möglich, cRNA- und vRNA-Transkription in einem System zu kombinieren. Dies enthält in einem ambisense bicistronischen RNA-Segment zwei Transkriptionseinheiten mit zwei einander entgegengesetzt orientierten Genen. Darin kann die Transkription des einen Gens von dem ambisense cRNA-Promotor, die des anderen Gens von dem ambisense vRNA-Promotor aus gestartet werden. Für die Polyadenylierung ist beiderseits eine U6-Sequenz benachbart zur 5´-Promotorsequenz eingeführt worden. In den primär transfizierten Zellen lässt sich die Expression der ambisense Gene CAT und GFP nachweisen. In den passagierten Zellen lässt sich bei diesen Konstrukten die Expression desjenigen Gens mit deutlicher Leistung nachweisen, dessen Transkription von dem vRNA-Promotor des ambisense Segmentes aus gestartet wird. Um die Expression desjenigen Gens weiter zu stärken, dessen Transkription von dem cRNA-Promotor des ambisense Segmentes aus gestartet wird, bedarf es weiterer Experimente. Jedoch zeigen die Ergebnisse, dass ein solches ambisense Segment von der viralen RNA-Polymerase transkribiert und repliziert wird. Außerdem wird das virale ambisense Segment in die Tochter-Virionen verpackt. Mit diesem System kann man zwei verschiedene Gene, z.B. ein Influenza-eigenes und ein beliebiges Fremdgen, zu einem ambisense Segment kovalent zusammenfügen, mit der Gewähr für einen stabilen Aufbau eines solchen rekombinanten Influenza-Virus. Influenza viruses belong to the Orthomyxoviridae family which are negative stranded RNA viruses. Their RNA is segmented (7 segments in influenza C and 8 segments in influenza A and B viruses). The segments of negative single-stranded viral genome RNA (vRNA) are contained in the form of ribonucleoprotein (RNP). Small amounts of the polymerase complex are associated with the RNPs, consisting of the subunits PB1, PB2 and PA (see influenza virus structure, page 12, Abb. 11). The in vivo analysis in this work was based on the polymerase I transcription system (Zobel et al., 1993; Neumann et al., 1994, see page 32, Abb. 1.7,). In this reverse genetic system cDNA coding for one of the influenza RNA segments is inserted between the cellular RNA polymerase I promoter (in this work human promotor) and terminator (murine) sequence. The coding region in the cDNA is replaced by CAT reporter gene. After transfection of such constructs in 293T cells, the human RNA polymerase delivers the precise viral RNA (vRNA, negative strand). Later the transfection is followed by infection with the FPV influenza A virus. During the infection the different influenza proteins are delivered beside the nucleoprotein (NP) and the polymerase subunits (PB1, PB2, PA), so that functional RNPs are built. The viral RNA polymerase and NP transcribe the vRNA (negative strand) to mRNA (positive strand). The replication is also catalyzed by the viral RNA polymerase and NP. For the replication a complementary RNA (cRNA, positive strand) is synthetized, which functions as a template for synthetizing further vRNAs. All of these processes (see page 26, Abb. 1.6) occur in the cell nucleus. CAT activity is then detectable in these cells. The foreign vRNA CAT segment is selectivly packaged in the progeny viruses by the end of an infection cycle additional to the eight influenza A segments. The package of this CAT segment is detectable by passaging the supernatant of the primer infection to MDCK cells. Detecting CAT activity in those infected MDCK cells means that the CAT segment was succesfully packaged. To raise the transcription and replication levels and therewith the possibility of packaging foreign segment tested in this system, we use the 1104 mutations (G`3 -> A`3, U`5 -> C`5, C`8 -> U`8) in the noncoding promoter region of the vRNA (Neumann and Hobom, 1995). The analogous mutations in the 5´end of the cRNA promoter (C 3 -> U 3, A 5 -> G 5, G8 -> A 8) were concerned in the cRNA constructs. While the longest influenza segment is only 2341 nt (nucleotides) in length, we were able to show that foreign segments of about 3000 nt and 4000 nt are transcribed, replicated and packaged in influenza A progeny viruses. However the efficiency of transcriptrion, replication and packaging of influenza A viral foreign segments decrease the longer these segments are (see page 84, Abb. 3.5). The RNA polymerase I transcription system was used to detect the influence of site- directed mutations in the cRNA promoter. For the direct analysis of the cRNA promoter and in contrary to the nature of the positive stranded cRNA of the influenza virus, the CAT reporter gene was inserted in the antisense orientation. While the ends remained to be cRNA ends, beside the fact that the polyadenylation signal consisting of 6 uridine nucleotides was added to the distal region of the 5´ end of the cRNA promoter. The 6 Adenin nucleotides in the 3´ end of the cRNA promoter were deleted to avoid base pairing with the polyadenylating signal in the 5´end (compare page 109, Abb. 3.26 with page 89, Abb. 3.11). After transfection in 293T cells and followed infection of such a cRNA transcription construct, a positive CAT signal was detected. This means, that the cRNA was transcribed by the viral RNA polymerase to the positive stranded mRNA, which was translated to the CAT protein. The replication of this cRNA produces a positive stranded vRNA, which replicates to negative stranded cRNA. The transcription of the positive stranded vRNA would produce a negative stranded mRNA, which can´t lead to the CAT expression. This way we guarantee that the detected CAT activity can only arise from the mRNA, which is transcribed from the cRNA by the action of the cRNA promoter, not the vRNA. The mutational analysis of the cRNA promoter showed that the active cRNA promoter also builds a corkscrew secondary structure (see page 97, Abb. 3.17) like the vRNA promoter (see page 96, Abb. 3.16). Nevertheless, the interaction of the viral RNA polymerase with the corksrew of the cRNA promoter differs from that with the corkscrew of the vRNA Promoter. These differences are obvious in the nucleotide specifity during the recognition of the promoter by the viral RNA polymerase. The 5´ ends of the mRNA and vRNA resulted from the cRNA transcription and replication were analysed using the RACE method (see page 111, Abb. 3.27). This analysis showed, that the 5´ end of the mRNAs detected were always elongated by foreign sequence, which is typical for mRNA transcribed from the vRNA. This means that the mRNA transcribed from the cRNA is also primer-dependent which requires cap-snatching process, known for mRNA transcribed from vRNA. In this process the polymerase subunit PB2 recognizes the 5´ cap ends of newly synthetized cellular mRNAs, the PB1 subunit cleaves 12-15 nucleotides at those ends which are used as primer for the mRNA synthesis from vRNA and cRNA as we introduce here. The levels of CAT activities detected in this work after transfection of cRNA constructs differs from those after transfection of vRNA constructs. This proves that detected CAT activities in this system are originated from the transcription process of the cRNA and the vRNA and not from the replication process. The transcription ability of the cRNA to mRNA led us to a new design of the polymerase I transcription system. We combined the cRNA and vRNA molecules to an ambisense molecule and inserted this in a polymerase I transcription unit (see page 122, Abb. 3.37 and page 123, Abb. 3.38). We used CAT and GFP as reporter genes, which were inserted in opposite orientations. The simultaneous expression of the two genes was detected in those ambisense constructs, where the GFP transcription is resulted from the vRNA by the action of the vRNA promoter and the CAT transcription is resulted from the cRNA by the action of the cRNA promoter, both proteins were detected in 293T cells after the primer infection. After passaging in MDCK cells only the protein, whose transcription originated from the vRNA strand was detected, which is GFP in this case. In the other ambisense case, the GFP expression was originated from cRNA and the CAT expression from the vRNA, only CAT expression was detected after the primer infection, and the same result was detected after the passage in MDCK cells. According to this results, the construction of the cRNA strand in the ambisense constructs must be optimized to support the transcription of the cRNA. Nevertheless, these results show, that such an ambisense segment can be transcribed, replicated and packaged in the progeny influenza viruses. Influenza genes or even other foreign genes can also be combined in such an ambisense system which might give a chance for producing stable recombinant influenza virus or other viral vectors.