Mammalian Genome 13, 543±547 (2002). Ó Springer-Verlag New York Inc. 2002
An RCAS-TVA-based approach to designer mouse models
Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021, USA
Received: 25 June 2002 / Accepted: 22 July 2002
The twentieth-century reductionist approach to molecular bio-
logy dissected complex biological phenomena into distinct
biochemical pathways. Today, we are challenged by these
distinct biochemical pathways in an attempt to recapitulate
tumor formation in complex biological settings, such as the
mouse. This review describes a recently developed mouse
modeling system that utilizes avian RCAS retroviruses as ve-
hicles to deliver genes of interest into transgenic mice that
express the gene encoding the avian receptor TVA in speci®c
cell types. This system allows for precise spatio-temporal in-
troduction of multiple de®ned genetic elements into selected
cell types, making it an ideal system for studying the cooper-
ating events in tumor formation as well as the roles of gene
expression in dierent cell lineages during development.
Creation of mouse models for developmental and tumorigenic
processes by using avian retroviral gene delivery
Retroviruses have the ability to introduce new genetic infor-
mation into the chromosomes of target cells, thus serving as
vehicles for the stable transfer of genes. A retroviral vector that
has been extensively used in avian systems is the avian leukosis
virus-A (ALV-A)-derived vector called RCAS (Replication-
Competent ASLV long terminal repeat with Splice acceptor).
RCAS was derived from RSV-A (Rous sarcoma virus-A) by
replacement of the src gene with a multicloning site that stably
accommodates inserts up to 2.5 kb (Boerkoel et al. 1993;
Greenhouse et al. 1988; Hughes et al. 1987; Petropoulos and
Hughes 1991). Since the src-derived splice acceptor site is re-
tained in RCAS vectors, the expression of an inserted gene is
driven by the viral long terminal repeat (LTR). There is a par-
allel family of RCAN vectors (Replication-Competent ASLV
long terminal repeat with No splice acceptor) that lack the src
splice acceptor and can express an inserted gene from an ap-
propriate internal promoter (Petropoulos et al. 1992). The most
commonly used RCAS vector, RCASBP, was created by sub-
stituting the pol region from the Bryan high-titer strain of RSV.
These vectors became very popular for use in gene transfer
because they are replication competent in avian cells, and high-
titer viral stocks can be produced without a helper component.
Although widely used in avian systems, RCAS vectors
could not be used in mammalian hosts because of the lack of
suitable cell surface receptors for virus entry and infection.
This has changed with the cloning of the tv-a gene, which
encodes for the receptor for ALV-A, called TVA (Bates et al.
1993; Young et al. 1993). It was shown that ectopic expression
of tv-a in mammalian cells allows for viral entry and chro-
mosomal integration by ALV-A (Bates et al. 1993; Young et al.
1993). This discovery enabled the generation of mammalian
cell lines and transgenic mouse lines that express the TVA cell
surface receptor, thus rendering the cells susceptible to infec-
tion with RCAS viruses (Federspiel et al. 1994, 1996; Young
et al. 1993). The practical aspects of generating RCAS-TVA
mouse tumor models have been reviewed (Fisher et al. 1999).
TVA transgenic mice have an important advantage over the
avian systemsÐthe TVA receptor can be expressed under the
control of a cell- or tissue-speci®c promoter in order to develop
a system for cell- or tissue-speci®c gene targeting (Dunn et al.
2000, 2001; Federspiel et al. 1994; Holland et al. 1998; Holland
and Varmus 1998; Murphy and Leavitt 1999). Only mouse cells
engineered to express the TVA receptor can be infected, while
mouse cells that do not harbor a tv-a gene homolog are resis-
tant to RCAS infection. As a result, the retroviral RCAS in-
fection enables the introduction of a gene of interest into a
speci®c cell type or tissue. After entry into TVA-positive
mammalian cells, a newly synthesized DNA copy of the viral
genome integrates into the host DNA, and viral LTR promotes
a high level of transcription of the integrated provirus.
In contrast to replication-defective vectors, RCAS vectors
do not require a packaging cell line. High-titer viral stocks can
be prepared by transferring a plasmid encoding the vector into
a chicken ®broblast cell line (Himly et al. 1998; Schaefer-Klein
et al. 1998). RCAS infection does not, however, result in in-
fectious progeny in mammalian cells (Federspiel et al. 1994).
The lack of viral protein production prevents cell-to-cell
spreading of infection and decreases the probability of an
immune response by the host (Pinto et al. 2000). Furthermore,
because the RCAS env gene is poorly expressed in mammalian
cells, the TVA receptor is not occupied by viral envelope
protein, and the infected TVA-positive mammalian cells re-
main susceptible to repeated rounds of infection with multiple
RCAS vectors carrying dierent genes. As a result, multiple
genes can be introduced sequentially into speci®c cells of a
single transgenic mouse strain (Federspiel et al. 1994; Holland
et al. 1998; Murphy and Leavitt 1999).
There are, however, several potential limitations to using the
RCAS-TVA system for generating mouse models. The ®rst in-
volves the 2.5 kb insert size limit of RCAS. Although pseudo-
typing allows for the generation of viruses with large inserts,
viral titers are generally lower. Second, the experimentalist has
no control over the integration site of the viral DNA. The virus
embeds itself as a transgene at random sites in the host genome.
Although the provirus has its own promoter, the activity of the
viral promoter is in¯uenced by the adjacent host sequences.
This can be corrected by modifying the RCAS vectors to express
an inserted gene from an appropriate internal promoter. The
more signi®cant concern, however, arises from the knowledge
that the insertion of the provirus can also in¯uence the ex-
Correspondence to: S. Orsulic; E-mail: email@example.com