Plant Molecular Biology 48: 183–200, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
Recombinase-directed plant transformation for the post-genomic era
David W. Ow
Plant Gene Expression Center, Agricultural Research Service, United States Department of Agriculture, Albany,
CA 94710, USA (
author for correspondence; e-mail firstname.lastname@example.org) and Plant & Microbial Biology,
University of California, Berkeley, CA 94720, USA
Key words: Cre- lox, DNA integration, FLP-FRT, phiC31, site-speciﬁc recombination
Plant genomics promises to accelerate genetic discoveries for plant improvements. Machine-driven technologies
are ushering in gene structural and expressional data at an unprecedented rate. Potential bottlenecks in this crop
improvement process are steps involving plant transformation. With few exceptions, genetic transformation is
an obligatory ﬁnal step by which useful traits are engineered into plants. In addition, transgenesis is most often
needed to conﬁrm gene function, after deductions made through comparative genomics, expression proﬁles, and
mutation analysis. This article reviews the use of recombinase systems to deliver DNA more efﬁciently into the
Since the pioneering transformation advances of the
early 1980s, much of the research efforts have been
directed, and rightly so, to a horizontal spread of the
technology. As a result of this emphasis, it is now
possible to transform a wide variety of plant species.
The trade-off, however, has been less attention de-
voted to advancing the efﬁciency of the transformation
process itself. Compared to many microbial systems,
plant transformation appears somewhat antiquated.
Whereas millions of independent transformants are
routinely obtained with many microbial systems, in
plants the numbers are generally in the single- to
double-digit range. Hence a shotgun transformation
approach to gene discovery is an option that has not
been seriously entertained.
Gene transfer in many microbial systems also pro-
duces highly consistent phenotypes. Relatively few
representative clones are needed for every construct
analyzed. In plants, independent transformants show
highly variable levels and patterns of expression. So,
for a typical DNA construct, 20–50 independent pri-
mary transformants are needed. For the commercial
development of a new trait, hundreds of independent
transformants are screened for the few with suitable
transgene structure and expression.
The underlying reasons for the high variability
in transgene expression are not completely under-
stood, but at least four factors are involved in this
1. Tissue culture. Somaclonal variation has long been
associated with tissue culture-regenerated plants.
Changes in chromosome structure and ploidy,
DNA sequence, DNA modiﬁcation, and transpo-
son activity have all been reported in somaclonal
variants (Peschke and Phillips, 1992; Kaeppler
et al., 2000).
2. Integration site. Chromosomal structures such as
telomeres or heterochromatin are known to affect
the expression of nearby genes (Stavenhagen and
Zakian, 1994; Howe et al., 1995; Wallrath and
Elgin, 1995). As a transgene integrates at ran-
dom locations, chromosomal inﬂuences on trans-
gene expression can be expected to differ among
independent transformants (Meyer, 2000).
3. Transgene redundancy. Transformed plants often
contain variable numbers of transgenes. Rarely is
there a positive correlation between gene expres-
sion and copy number. On the contrary, many cases
have linked extra full or partial transgene copies