Applications of recombinant DNA technology 315
An alternative to the use of antifungal proteins or
metabolites is to manipulate the hypersensitive
response, which is a physiological defence mechan-
ism used by plants to repel attacking pathogens.
Resistance occurs only in plants carrying a resist-
ance gene (R) that corresponds to an avirulence
(avr) gene in the pathogen. Elicitors (signalling
molecules) released by pathogens are detected by
the plant and activate a range of defence responses,
including cell death, PR-gene expression, phyto-
allexin synthesis and the deposition of cellulose at the
site of invasion, forming a physical barrier. Import-
antly, the hypersensitive response is systemic, so
that neighbouring cells can pre-empt pathogen
invasion. A recently developed strategy is to transfer
avirulence genes from the pathogen to the plant,
under the control of a pathogen-inducible promoter.
This has been demonstrated in tomato plants
transformed with the avr9 gene from Chladosporium
fulvum, resulting in resistance to a range of fungal,
bacterial and viral diseases (Keller et al. 1999,
reviewed by Melchers & Stuiver 2000).
Resistance to insects and other pests
Insect pests represent one of the most serious biotic
constraints to crop production. For example, more
than a quarter of all the rice grown in the world is
lost to insect pests, at an estimated cost of nearly
US$50 billion. This is despite an annual expenditure
of approximately US$1.5 billion on insecticides for
this crop alone. Insect-resistant plants are therefore
desirable not only because of the potential increased
yields, but also because the need for insecticides
is eliminated and, following on from this, the
undesirable accumulation of such chemicals in the
environment is avoided. Typical insecticides are
non-selective, so they kill harmless and beneficial
insects as well as pests. For these reasons, transgenic
plants have been generated expressing toxins that
are selective for particular insect species.
Research is being carried out on a wide range of
insecticidal proteins from diverse sources. However,
all commercially produced insect-resistant transgenic
crops express toxin proteins from the Gram-positive
bacterium Bacillus thuringiensis (Bt). Unlike other
Bacillus species, Bt produces crystals during sporula-
tion, comprising one or a small number of ~130 kDa
protoxins called crystal proteins. These proteins are
potent and highly specific insecticides. The specificity
reflects interactions between the crystal proteins
and receptors in the insect midgut. In susceptible
species, ingested crystals dissolve in the alkaline
conditions of the gut and the protoxins are activated
by gut proteases. The active toxins bind to receptors
on midgut epithelial cells, become inserted into the
plasma membrane and form pores that lead to cell
death (and eventual insect death) through osmotic
lysis. Approximately 150 distinct Bt toxins have
been identified and each shows a unique spectrum of
activity.
Bt toxins have been used as topical insecticides
since the 1930s, but never gained widespread use,
because they are rapidly broken down on exposure
to daylight and thus have to be applied several times
during a growing season. Additionally, only insects
infesting the exposed surface of sprayed plants are
killed. These problems have been addressed by the
expression of crystal proteins in transgenic plants.
Bt genes were initially introduced into tomato
(Fischhoff et al. 1987) and tobacco (Barton et al.
1987, Vaeck et al. 1987) and later into cotton
(Perlak et al. 1990), resulting in the production of
insecticidal proteins that protected the plants from
insect infestation. However, field tests of these plants
revealed that higher levels of the toxin in the plant
tissue would be required to obtain commercially
useful plants (Delannay et al. 1989). Attempts to
increase the expression of the toxin gene in plants by
use of different promoters, fusion proteins and leader
sequences were not successful. However, examina-
tion of the bacterial cry1Ab and cry1Ac genes
indicated that they differed significantly from plant
genes in a number of ways (Perlak et al. 1991). For
example, localized AT-rich regions resembling plant
introns, potential plant polyadenylation signal
sequences, ATTTA sequences that can destabilize
mRNA and rare plant codons were all found. The
elimination of undesirable sequences and modifica-
tions to bring codon usage into line with the host
species resulted in greatly enhanced expression of
the insecticidal toxin and strong insect resistance
of the transgenic plants in field tests (Koziel et al.
1993). By carrying out such enhancements, Perlak
and colleagues expressed a modified cry3A gene in
potato to provide resistance against Colorado beetle
POGC14 9/11/2001 11:12 AM Page 315