Figure 1. Gene transfer by murine leukaemia virus (MLV)-derived vectors. Generation of the two components of an MLV-based gene transfer system. Top shows a diagram of the integrated DNA genome of MLV (about 9000 nucleotides). Among the cis-acting control sequences are the packaging signal () for encapsidation of RNA into viral particles and the long terminal repeat (LTR) regions that harbour control elements for transcription, reverse transcription and integration. The three open reading frames (gag, pol and env) encode all trans-acting protein factors. The left column shows packaging constructs engineered to direct the production of viral proteins. Simple packaging constructs harbour all three open reading frames in one construct. In complex packaging constructs, the gag-pol and env reading frames are engineered into two separate constructs. , deleted packaging signal; p(A), heterologous polyadenylation signal; Prom, heterologous promoter. The right column shows various designs of MLV-based transfer vectors. (a) Transfer vector derived by simple replacement of the coding regions of the viral proteins by the coding region of a foreign gene. (b) Transfer vector harbouring an internal heterologous promoter in addition to the viral LTR promoter. Due to the scheme of duplication of LTR sequences during retroviral replication a two-promoter transfer vector may be designed such that transcriptional control sequences of the LTR are nonoperative in the target cell (self-inactivating transfer vectors). (c) Transfer vector in which expression of two genes is directed by distinct promoters, the LTR promoter and an internal heterologous promoter, respectively. (d) Transfer vector with two genes in which gene product 1 is produced from unspliced RNA and gene product 2 from spliced messenger RNA directed by the splice donor (SD) and splice acceptor (SA) signals for viral env messenger RNA. (e) Transfer vector producing a single messenger RNA directing the translation of two open reading frames, translation of gene 2 being controlled by an internal ribosome entry site (IRES).

Figure 2. Gene transfer by a two-component retroviral vector system (see Figure 1). A packaging cell line has been generated by stable insertion of packaging construct DNA by a nonviral method such as transfection. Transfer vector DNA is introduced into the packaging cells. The transfer vector RNA is packaging competent while the packaging construct RNAs are defective for incorporation into viral particles. Viral particles harbouring transfer vector RNA are used for transduction of target cells. Subsequently target cells producing the desired transfer vector-encoded RNA and protein may be identified and expanded.

Figure 3. Activation of an oncogene by provirus insertion. In the native conformation the proto-oncogene is regulated by its native promotor and enhancer elements illustrated in the upper part of the figure. Insertion of retroviral vector enhancer elements from the integrated provirus may influence the transcriptional level of the oncogene and increase its level of transcription. Depicted is a case of enhancer activation of an oncogene where the provirus has been inserted in the opposite transcriptional direction. Curved twin lines represent chromatin; thin straight lines represent RNA transcripts; thick arrow indicates enhancer activation of host promoter.

Figure 4. Generation of replication-competent viruses by recombination between transfer vector and packaging constructs. Recombination takes place at the level of reverse transcription of two RNAs encapsidated into the same particle. All three examples cover a simple transfer vector as shown in Figure 1. In the case of a simple packaging construct with viral control sequences at both ends, only one crossover event is needed to generate a fully functional virus. In the simple packaging construct with a heterologous poly(A) signal, two crossover events are needed. In the complex packaging construct, three crossover events are needed. Not shown is the possible contribution of endogenous retroviral RNAs.

Figure 5. HIV-1 based gene transfer systems. Top: Map of the human immunodeficiency virus type 1 (HIV-1) genome in the DNA form with indication of the long terminal repeat (LTR), and Rev-responsive element (RRE) regions and open reading frames encoding viral proteins. Left: Basic packaging construct impaired in the Vpu and Env. Heterologous envelopes, the amphotropic MLV (MLV (A) env) or the vesicular stomatitis virus G protein (VSV G) provided from separate expression constructs. All constructs harbour heterologous promoters (Prom) and polyadenylation signals (p(A)). Advanced packaging constructs express gag-pro-pol, lack all auxiliary HIV-1 genes and retain the Rev-responsive element (RRE). Separate expression constructs provide Rev and envelope function (in casu VSV G). Right: The simple transfer vector harbours the extended packaging signal () partly overlapping with gag, Gag production being impaired by mutation. The foreign gene is inserted after a heterologous promoter. In the self-inactivating transfer vector the heterologous promoter (Prom) is independent of Tat in the packaging cell and the LTR translocated to the upstream position in the target cell impaired by mutation. The lines below the self-inactivating transfer vector maps indicate predicted RNA species. SD, splice donor site; SA, splice acceptor site.

Figure 6. Ex vivo and in vivo gene transfer. Transfer vector-bearing virus particles are liberated into the supernatant of cultured packaging cells (upper left). The right part of the figure (blue arrows) illustrates the ex vivo scheme, the left part (red arrows) the in vivo scheme. Candidate target tissues and prototypic diseases are listed. These are not restricted to those having been approached in retroviral vector-based clinical trials.