Genetically engineering plants for crop improvement.
ABSTRACT Dramatic progress has been made in the development of gene transfer systems for higher plants. The ability to introduce foreign genes into plant cells and tissues and to regenerate viable, fertile plants has allowed for explosive expansion of our understanding of plant biology and has provided an unparalleled opportunity to modify and improve crop plants. Genetic engineering of plants offers significant potential for seed, agrichemical, food processing, specialty chemical, and pharmaceutical industries to develop new products and manufacturing processes. The extent to which genetically engineered plants will have an impact on key industries will be determined both by continued technical progress and by issues such as regulatory approval, proprietary protection, and public perception.
Genetically Engineering Plants for Crop Improvement
Author(s): Charles S. Gasser and Robert T. Fraley
Source: Science, New Series, Vol. 244, No. 4910 (Jun. 16, 1989), pp. 1293-1299
Published by: American Association for the Advancement of Science
Stable URL: http://www.jstor.org/stable/1704384 .
Accessed: 03/09/2011 13:46
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact email@example.com.
American Association for the Advancement of Science is collaborating with JSTOR to digitize, preserve and
extend access to Science.
CHARLES S. GASSER AND ROBERT T. FRALEY
Dramatic progress has been made in the development of
gene transfer systems for higher plants. The ability to
introduce foreign genes into plant cells and tissues and to
regenerate viable, fertile plants has allowed for explosive
expansion of our understanding of plant biology and has
provided an unparalleled opportunity to modify and
improve crop plants. Genetic engineering of plants offers
significant potential for seed, agrichemical, food process-
ing, specialty chemical, and pharmaceutical industries to
develop new products and manufacturing processes. The
extent to which genetically engineered plants will have an
impact on key industries will be determined both by
continued technical progress and by issues such as regula-
tory approval, proprietary protection, and public percep-
HE STABLE INTRODUCTION
plants represents one of the most significant developments in
a continuum of advances in agricultural technology that
includes modern plant breeding, hybrid seed production, farm
mechanization, and the use of agrichemicals to provide nutrients
and control pests. The first-generation applications of genetic
engineering to crop agriculture are targeted at issues that are
currently being addressed by traditional breeding and agrichemical
discovery efforts: (i) improved production efficiency, (ii) increased
market focus, and (iii) enhanced environmental conservation. Ge-
netic engineering methods complement plant breeding efforts by
increasing the diversity of genes and germplasm available for
incorporation into crops and by shortening the time required for the
production of new varieties and hybrids. Genetic engineering of
plants also offers exciting opportunities for the agrichemical, food
processing, specialty chemical, and pharmaceutical industries to
develop new products and manufacturing processes.
The first transgenic plants expressing engineered foreign genes
were tobacco plants produced by the use of Agrobacterium
vectors (1). Transformation was confirmed by the presence of
foreign DNA sequences in both primary transformants and their
progeny and by an antibiotic resistance phenotype conferred by a
chimeric neomycin phosphotransferase gene. These early transfor-
mation experiments often utilized plant protoplasts as the recipient
cells; the subsequent development of transformation methods based
on regenerable explants (2) such as leaves, stems, and roots contrib-
OF FOREIGN GENES
The authors are at Monsanto Company, 700 Chesterfield Village Parkway, St. Louis,
i6 JUNE i989
uted significantly to the facile and routine transformation methods
that are used today for many dicotyledonous plant species. A variety
of free DNA delivery methods, including microinjection, electropo-
ration, and particle gun technology are being developed for the
transformation of monocotyledonous plants such as corn, wheat,
and rice. In view of the rapid progress that is being made, it is likely
that all major dicotyledonous and monocotyledonous crop species
will be amenable to improvement by genetic engineering within the
next few years.
In this article, we describe transformation methods that have been
developed for plants and discuss some of the applications of
genetically engineered plants in agriculture. We also address some of
the critical issues that will influence the commercialization of
genetically engineered crops.
Methods for Introducing Genes into Plants
Agrobacterium tumefaciens-mediated gene transfer. Derivatives of
the plant pathogen Agrobacterium tumefaciens have proved to be
efficient, highly versatile vehicles for the introduction of genes into
plants and plant cells. Most transgenic plants produced to date were
created through the use of the Agrobacterium system. Agrobacterium
is the etiological agent of crown gall disease and produces
tumorous crown galls on infected species. The utility of this
bacterium as a gene transfer system was first recognized when it was
demonstrated that the crown galls were actually produced as a result
of the transfer and integration of genes from the bacterium into the
genome of the plant cells (3). Virulent strains of Agrobacterium
contain large Ti (for tumor inducing) plasmids, which are responsi-
ble for the DNA transfer and subsequent disease symptoms. Genetic
and molecular analyses showed that Ti plasmids contain two sets of
sequences necessary for gene transfer to plants; one or more T-DNA
(transferred DNA) regions that are transferred to the plant, and the
Vir (virulence) genes which are not, themselves, transferred during
infection. The T-DNA regions are flanked by border sequences that
were shown to be responsible for the definition of the region that is
to be transferred to the infected plant cell. The T-DNA contains 8 to
13 genes (4), including a set for production of phytohormones,
which are responsible for formation of the characteristic tumors
when transferred to infected plants. Several excellent reviews on the
biology of this and other pathogenic species of Agrobacterium have
been published for those who desire more detailed information (4).
Early experiments demonstrated that heterologous DNA inserted
into the T-DNA could be transferred to plants along with the
existing T-DNA genes (5). Efficient plant transformation systems
were constructed by removing the phytohormone biosynthetic
genes from the T-DNA region, thereby eliminating the ability of the
bacteria to induce aberrant cell proliferation (6). Modern plant
transformation vectors are capable of replication in Escherichia coli as
well as Agrobacterium, allowing for convenient manipulations (7).
The general features of these vectors and the process of transfer to
plant cells are outlined in Fig. 1. Recent technological advances in
vectors for Agrobacterium-mediated
provements in the arrangements of genes and restriction sites in the
plasmids that facilitate construction of new expression vectors.
Vectors in current use have convenient multilinker regions, which
may be flanked by a promoter and a polyadenylate addition site for
direct expression of inserted coding sequences (8).
constitutes an excellent system for introducing genes
into plant cells, since (i) DNA can be introduced into whole plant
tissues, which bypasses the need for protoplasts, and (ii) the
integration of T-DNA is a relatively precise process. The region of
DNA to be transferred is defined by the border sequences; occasion-
al rearrangements do occur, but in most cases an intact T-DNA
region is inserted into the plant genome (9). This contrasts with free
DNA delivery systems in which the plasmids routinely undergo
rearrangment and concatenation reactions before insertion and can
lead to chromosomal rearrangements during insertion in both
animal (10) and plant (11) systems. Sequencing of insertion sites
shows that only small duplications or other changes occur in
flanking sequences during T-DNA integration (12). The stability of
expression of most genes that are introduced by Agrobacterium
appears to be excellent. Published studies have shown that integrat-
ed T-DNAs give consistent genetic maps and appropriate segrega-
gene transfer have involved im-
Fig. 1. Agrobacterium-medi-
ated plant transformation.
(A) Generalized plant trans-
formation vector (PTV).
The plasmid contains an ori-
gin of replication that allows
it to replicate in Agrobacter-
ium (Ori-Agro), and a high
copy number origin of repli-
cation functional in E. coli
(Ori-E. coli). This allows for
easy production and testing
of engineered plasmids in E.
coli prior to transfer to Agro-
for subsequent in-
troduction into plants. Two
resistance genes are usually
carried on the plasmid, one
for selection in bacteria, in
this case for spectinomycin
resistance (SpJ), and the
other that will express in
plants; in this example en-
coding kanamycin resistance
(Kanr). Also present are sites
for the addition of one or
more inserted genes (IG)
and directional T-DNA bor-
der sequences which, when
recognized by the transfer
finctions of Agrobacterium,
delimit the region that will be transferred to the plant. (B) Diagram of the
process. The PIV constructed in E. coli is transferred to
an engineered Agrobacterium
by a "triparental"
engineered Agrobacterium contains a "disarmed" Ti plasmid (D-Ti) from
which the genes necessary for pathogenesis have been removed (6). Viru-
lence functions on the D-Ti interact in trans with the border sequences on
the PTsV mobilizing the region between them into a plant cell and inserting it
into one of the plant's chromosomes within the nucleus. The kanamycin-
resistant phenotype conferred by the Kan^r gene allows the selection of
transformed plant cells during plant regeneration.
B E. co/i
mating procedure (6). The
tion ratios (1, 13). Introduced traits have been found to be stable
over at least five generations during cross-breeding and seed increase
on genetically engineered tomato and oilseed rape plants (14). This
stability is critical to the commercialization of transgenic plants. The
list of plant species that can be transformed by Agrobacterium has
been greatly expanded and now includes several of the most
important broadleaf crops (Table 1).
Advances in other transformation
technologies. In those systems where
transformation is efficient, it is the method
of choice because of the facile and defined nature of the gene
transfer. Few monocotyledonous plants appear to be natural hosts
although transgenic plants have been produced in
asparagus with Agrobacterium
vectors (15) and transformed tumors
have been observed in yam (16). Cereal grains such as rice, corn, and
wheat have not been successfully transformed by Agrobacterium,
despite encouraging evidence for T-DNA transfer in corn (17).
Extensive efforts have consequently been directed toward the devel-
opment of systems for the delivery of free DNA into these species.
The first of these systems to give demonstrable transformation of
plant cells relied on physical means similar to those used in the
transformation of cultured animal cells. Transformation has been
achieved in plant protoplasts through facilitation of DNA uptake by
calcium phosphate precipitation, polyethylene glycol treatment,
electroporation, or combinations of these treatments (18). These
methods have allowed the production of transgenic cells for the
study of gene expression in systems that cannot be transformed by
other means (19).
The applicability of these systems to the production of transgenic
plants is limited by the difficulties involved in regenerating plants
from protoplasts. There have been significant advances in the
regeneration of cereals (traditionally one of the most recalcitrant
groups) from protoplasts. Several laboratories have succeeded in
regenerating fertile rice plants from protoplasts (20). This advance
was rapidly followed by the production of transgenic rice plants
through the delivery of free DNA to protoplasts followed by
regeneration (21). Progress in regeneration of corn has been more
limited; one group demonstrated regeneration of mature plants
from protoplasts and succeeded in producing transgenic plants (22,
23). However, all plants were sterile, apparently as a result of the
necessary period in culture or the regeneration procedure. While this
progress is encouraging, limitations remain in the application of this
technology to cereal crop improvement. In corn and rice, the ability
to form regenerable protoplasts appears to be primarily confined to
a small number of varieties. Even if the fertility problems are
overcome, introduction of the transferred genes into the broad
range of commercial varieties in use today would require a lengthy
period of backcrossing.
In parallel with the work on protoplast transformation, efforts to
find novel ways to introduce DNA into intact cells or tissues have
been emphasized. Regeneration of cereals from immature embryos
or from explants is relatively routine (24). One of the most
significant developments in this area has been the introduction of
"particle gun" or high-velocity microprojectile technology. In this
system, DNA is carried through the cell wall and into the cytoplasm
on the surface of small (0.5 to 5 [im) metal particles that have been
accelerated to speeds of one to several hundred meters per second
(25-27). The particles are capable of penetrating through several
layers of cells and allow the transformation of cells within tissue
explants. Production of transformed corn cells (28) and fertile, stably
transformed tobacco (26) and soybean (27) plants with particle guns
has already been demonstrated. By eliminating the need for passage
through a protoplast stage, the particle gun method has the
potential to allow direct transformation of commerical genotypes of
cereal plants. Intensive efforts to produce transgenic cereals by the
SCIENCE, VOL. 244
use of particle guns are currently under way in many laboratories
around the world.
Other methods that have the potential to influence the production
of transgenic cereals include gene transfer into pollen (29), direct
injection into reproductive organs (30), microinjection into cells of
immature embryos (31), and rehydration of desiccated embryos
(32). There has been some demonstration of transient or stable gene
expression through the use of each of these methods in some species,
but the range of their applicability remains to be demonstrated.
Application of Genetic Engineering to
The availability of efficient transformation systems for crop
species is of intense interest to biotechnology, agrichemical, and
seed companies for the application of this technology to crop
improvement. Initial research has been focused on the engineering
of traits that relate directly to the traditional roles of industry in
farming, such as the control of insects, weeds, and plant diseases.
Progress has been rapid, and genes conferring these traits have
already been successfully introduced into several important crop
species. Genetically engineered soybean, cotton, rice, corn, oilseed
rape, sugarbeet, tomato, and alfalfa crops are expected to enter the
marketplace between 1993 and 2000.
Weed control. Engineering herbicide tolerance into crops represents
a new alternative for conferring selectivity and enhancing crop safety
of herbicides. Research has largely concentrated on those herbicides
with properties such as high unit activity, low toxicity, low soil
mobility, and rapid biodegradation and with broad spectrum activi-
ty against various weeds. The development of crop plants that are
tolerant to such herbicides would provide more effective, less costly,
and more environmentally attractive weed control. The commercial
strategy in engineering herbicide tolerance is to gain market share
through a shift in herbicide use (33)-not to increase the overall use
of herbicides, as is popularly held. Herbicide-resistant plants will
have the positive impact of reducing overall herbicide use through
substitution of more effective and environmentally acceptable prod-
Two general approaches have been taken in engineering herbicide
tolerance: (i) altering the level and sensitivity of the target enzyme
for the herbicide and (ii) incorporating a gene that will detoxify the
herbicide. As an example of the first approach, glyphosate, the active
ingredient of Roundup herbicide, acts by specifically inhibiting the
enzyme 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS)
(34). Glyphosate is active against annual and perennial broadleaf and
grassy weeds, has very low animal toxicity, and is rapidly inactivated
and degraded in all soils (35). Tolerance to glyphosate has been
engineered into various crops by introducing genetic constructions
for the overproduction of EPSPS (36) or of glyphosate-tolerant
variant EPSPS enzymes (37, 38). Similarly, resistance to sulfonyl-
urea compounds, the active ingredients in Glean and Oust herbi-
cides, has been produced by the introduction of mutant acetolactate
synthase (ALS) genes (39). Glean and Oust are broad-spectrum
herbicides and are effective at low application rates. Since both
EPSPS and ALS activities are present in wild-type plants, the
possibility of deleterious effects on crop performance or product
quality due to their reintroduction is unlikely. The use of these
herbicides in new crop applications may require reexamination of
residues of the herbicides; however, since the residue safety levels for
these two compounds in food crops have already been established,
this is not an issue unique to genetically engineered plants.
Resistance to gluphosinate (40) and bromoxynil (41) has been
achieved by the alternative approach of introducing bacterial genes
16 JUNE 1989
encoding enzymes that inactivate the herbicides by acetylation or
nitryl hydrolysis, respectively. In field tests the gluphosinate-tolerant
plants have shown excellent tolerance to the herbicide (42). Evalua-
tion of the biological activity of the specific herbicide conjugates and
metabolites that may be present in the transgenic plants will be
carried out according to existing chemical residue regulations.
Current crop targets for engineered herbicide tolerance include
soybean, cotton, corn, oilseed rape, and sugarbeet. Factors such as
herbicide performance, crop and chemical registration costs, poten-
tial for out-crossing to weed species, proprietary rights issues, and
competing herbicide technologies must all be considered before final
decisions on commercialization of specific herbicide-tolerant crops
can be made.
Insect resistance. The production of insect-resistant plants is another
application of genetic engineering with important implications for
crop improvement and for both the seed and agrichemical indus-
tries. Progress in engineering insect resistance in transgenic plants
has been achieved through the use of the insect control protein
genes of Bacillus thuringiensis (B.t.). Bacillus thuringiensis is an entomo-
cidal bacterium that produces an insect control protein which is
lethal to selected insect pests (43). Most strains of B.t. are toxic to
lepidopteran (moth and butterfly) larvae, although some strains
with toxicity to coleopteran (beetle) (44) or dipteran (fly) (45) larvae
have been described. The insect toxicity of B.t. resides in a large
protein; this protein has no toxicity to beneficial insects, other
animals, or humans (46). The mode of action of the B.t. insect
control protein is thought to be exerted at the level of disruption of
ion transport across brush border membranes of susceptible insects
Table 1. Species for which the production of transgenic plants have been
reported. Abbreviations: At, Agrobacterium tumefaciens; Ar, Agrobacterium
rhizogenes; FP, free DNA introduction into protoplasts; PG, particle gun;
MI, microinjection; IR, injection of reproductive organs.
At (1)), FP (85), PG (26)
At (89), MI (31)
At (38), PG (27)
Transgenic tomato, tobacco, and cotton plants containing the B.t.
gene exhibited tolerance to caterpillar pests in laboratory tests (48).
The level of insect control observed in the field tests with tobacco
and tomato plants has been excellent; in one such test tomato plants
containing the B.t. gene suffered no agronomic damage under
conditions that led to total defoliation of control plants (49).
The excellent insect control observed under field conditions
indicates that this technology may have commercial application in
the near future. Early market opportunities for caterpillar resistance
are leafy vegetable crops, cotton, and corn. Crop targets for beetle
resistance are potato and cotton. Other types of insecticidal mole-
cules are necessary to extend biotechnology approaches for control-
ling additional insect pests in these and other target crops. Plants
genetically engineered to express a proteinase inhibitor gene are
partially resistant to tobacco budworm in laboratory experiments
(50); field tests will be necessary to determine the agronomic utility
of this approach.
Disease resistance. Significant resistance to tobacco mosaic virus
(TMV) infection, termed "coat protein-mediated protection," has
been achieved by expressing only the coat protein gene of TMV in
transgenic plants (51). This approach produced similar results in
transgenic tomato, tobacco, and potato plants against a broad
spectrum of plant viruses, including alfalfa mosaic virus, cucumber
mosaic virus, potato virus X, and potato virus Y (52). One mecha-
nism of coat protein-mediated cross protection appears to involve
interference with the uncoating of virus particles in cells before
translation and replication (53).
Transgenic tomatoes carrying the TMV coat protein gene have
been evaluated in greenhouse and field tests and shown to be highly
resistant to viral infection (Fig. 2) (54). The transgenic plants
showed no yield loss after virus inoculation, whereas the yield was
reduced 23% to 69% in control plants. The level of capsid protein in
the engineered plants [typically 0.01% to 0.5% of the total protein
(52)] is well below the levels found in plants infected with this
endemic virus. This fact should facilitate registration and commer-
cialization of virus-resistant plants. Virus resistance could provide
significant yield protection in important crops such as vegetables,
corn, wheat, rice, and soybean.
While limited success in engineering resistance to fingal diseases
has been reported (55), genetically engineered resistance to fungal
pathogens and to bacteria remains in the early research stages.
Key Advances in Expression and Gene
Dramatic progress has been made in our understanding of and
ability to alter the regulation of gene expression in plants and in
techniques for the identification and isolation of genes of interest. In
many cases, this progress has been facilitated by the availability of
efficient gene transfer systems. The engineered plants discussed in
the previous section generally depend on the use of continuously
expressed promoters driving dominant single gene traits. Future
plant genetic engineering will probably include alteration of traits
that require subtle temporal and spatial regulation of gene expres-
sion and introduction or alteration of entire biosynthetic pathways.
Regulated gene expression. Genes that show precise temporal and
spatial regulation in leaves (56), floral organs (57), seeds (58), and
other plant organs have now been identified and isolated from a
number of species of higher plants (59). Within the next few years,
genetic engineers will have in hand a large battery of regulatory
sequences that will allow for accurate targeting of gene expression to
specific tissues within transgenic plants. In addition, a number of
genes that respond to external influences, such as heat shock,
anaerobiosis, wounding, nutrients, and applied phytohormones,
have been isolated and characterized (60). The control regions of
these genes may also find utility in genetic engineering strategies.
The ability to decrease the expression of a gene in a transgenic
plant also has potential utility in the study of plant gene expression
and function as well as in crop improvement. Significant successes
have already been achieved with genes that produce antisense RNAs
to the messengers for polygalacturonase in tomato fruits (61) and
chalcone synthase in petunia and tobacco plants (62). In all of these
studies, substantial reductions (up to 90%) in the levels of the
mRNA and protein products of the target genes were observed.
Striking phenotypic alterations were observed in some of these
transgenic plants (62). This method of constructing mutant pheno-
types will significantly enhance biochemical and physiological stud-
ies on protein and enzyme function. In an alternative approach to
reducing expression of a gene, the enzymatic regions derived from
self-splicing RNA molecules are used to design RNA enzymes
capable of specific RNA cleavage (63). In vitro studies have
demonstrated the potential of this method, but it has yet to be
applied in plants (63). Preliminary work on insertion of donor DNA
into plant chromosomes by homologous recombination (64) indi-
cates that it may also be possible to use this approach for the
selective inactivation of a gene.
Gene tagging. Advances in methods for the identification and
isolation of new gene coding sequences are of great importance to
the engineering of improved plants. The cloning of transposon
sequences has allowed the isolation of genes from several species by
transposon-mediated gene tagging (65). The demonstration that
mobile elements isolated from maize are able to transpose when
introduced into dicot species (66) indicates that this powerful
technique is applicable to any plant species for which transformation
is possible. It has also been shown that under appropriate transfor-
mation conditions, the T-DNA of a plant transformation vector can
itself serve as an insertional mutagen (67).
Gene mapping. Major efforts have been mounted to obtain high-
resolution restriction fragment length polymorphism (RFLP) ge-
netic maps in a number of plant species (68). The availability of such
a map in tomato has already led to the resolution of several loci
affecting quantitative quality traits (69). The RFLP mapping tech-
nique will be especially powerful in Arabidopsis, where the small
genome size and lack of significant repetitive sequences (70) will
simplify the process of genome "walking" from an RFLP marker to
a closely linked gene. The availability of Arabidopsis genomic libraries
in cosmids, which can also act as plant transformation plasmids (71),
will allow direct testing of the isolated DNA for its ability to
complement the mutation of interest at each step of the walking
process. In addition, such libraries may be used in large-scale
transformation experiments to directly rescue genes by complement-
ing mutants with a selectable phenotype (71).
Key Issues Affecting Introduction of
Genetically Engineered Plants
The advances in crop improvement by genetic engineering have
occurred so rapidly that the initial introduction of these crops in the
marketplace will be primarily influenced by nontechnical issues.
These issues include regulatory approval, proprietary protection,
and public perception.
Regulatory approval. In the United States, genetically engineered
plants potentially come under the statutory jurisdiction of three
federal agencies: the United States Departmlent of Agriculture
(USDA), Food and Drug Administration (FDA), and Environmen-
tal Protection Agency (EPA). The field testing of genetically
SCIENCE, VO L. 24'1
engineered crops has been less controversial than the introduction of
other recombinant organisms into the environment. In the last 3
years there have been over a dozen tests of engineered crops in
diverse locations across the United States (72)-by year end there
will be over 30 such tests. All of these tests have been reviewed in
detail by the USDA, with input from the other government
agencies. The key consideration in approval of these tests has been a
scientific evaluation of the risk and environmental impact of a
particular field test experiment. Several studies and discussions of
the issues and perceptions that surround the release of genetically
engineered crops have produced a consensus that such engineered
crops present virtually no direct risk to human or animal health (73).
The specific knowledge of the introduced DNA sequences, the
detailed understanding of the known functions of the gene prod-
ucts, and the high level of biological or physical containment were
cited as key reasons for the inherent low risk to human and animal
The "success" of such small field tests, while important, has
overshadowed other needs in the regulatory process. For example,
many unanswered questions remain regarding the cost and regula-
tory requirements for large-scale multisite field tests. It is important
that an approval process be developed to accommodate the rapid
transition that will occur as testing of engineered crops goes from
small, isolated field plots to large-scale, multisite testing; the devel-
opment of genetically engineered crop varieties and hybrids will
ultimately occur in the fields around the world-not in the research
laboratory. The mechanism for FDA or EPA approval or endorse-
ment of genetically engineered plants and food products remains
undefined. Issues such as regulatory requirements, registration
costs, and commercialization
timelines are already becoming signifi-
cant issues for companies attempting to develop improved genetical-
ly engineered crops for te mid-199Os. Several groups (74), such as
the International Food Biotechnology Council (IFBC) and the
Federation of American Scientists for Experimental Biology (FA-
SEB) expert panel on criteria for determnicng the regulatory status
of food and food ingredients produced by new technologies,
consisting of academic scientists and representatives of major food,
chemical, biotechnology, and seed companies, are working with
government agencies to develop appropriate registration guidelines.
The regulation of transgenic plants must be based on scientific
principles that (i) meet the general publics need for a safe and
reasonably priced food supply and (ii) recognize the inherent low
risk of gene transfer technology and the benefits afforded by
genetically engineered crops to growers, food processors, and
Proprietary protection. Patent protection for genetically engineered
plants is considered essential to offset the cost of developing crops
with significant new traits. The Supreme Court decision in Dia-
mond v. Chakrabarty (75) ruled that microorganisms were not
unpatentable simply because they were living cells, and in 1985, the
U.S. Board of Patent Appeals and Interferences ruled specifically
that whole plants were patentable (76). Numerous companies have
since filed patent applications that cover the genes, the processes of
isolating genes, and making the genetically modified plants and
seeds themselves. Patent protection provides a broader proprietary
right than is provided under either the International Union for the
Protection of New Varieties of Plants (UPOV) or the U.S. Plant
Variety Protection Act (PVPA). T he scope of the proprietary right
of a patent on a plant is broadened by the absence of the "experi-
mental use" exceptions found in protection afforded by plant varietal
certification status. Although no one disputes that companies that
have invested heavily in R&ED to isolate, test, and commercialize
genes are entitled to protection for their inventions, there is
considerable debate within the seed industry concerning how much
16 JUNE 1989
protection is deserved and what impact patents will have on the
cooperative nature of the seed industry itself (77). The concern has
been voiced that patents on plants will favor large seed companies
and reduce the overall number of companies. In contrast, while
there were three private soybean seed companies before PVPA, now
there are more than 40; patenting plants will likely create finrther
incentive to invest in the seed industry in order to position it to meet
the technological challenges and supply needs of the future. Much of
this debate results from confusion surrounding the restrictions
imposed by patent rights versus the incentive they provide for the
competitive research and product development that stimulates inno-
vation. Many of the conciliatory proposals, including patenting of
genes (but not plants) and compulsory licensing in the event that
plant patenting .is permitted, if implemented, could significantly
reduce the incentive for private industry funding in this field.
Lack of proprietary protection for genetically engineered plants
outside the United States remains a serious limitation; plant and
animal varieties are largely excluded from patent protection by
European countries that signed the 1973 European Patent Conven-
tion. At this time only specific processes can be patented. The
European Patent Office (EPO) is currently readdressing the patent-
ing of plants and animals, but this seems certain to be appealed and
it may be several years before the situation is clear, and only then will
.. . . ~~~
~ ~ ~ ~ ~ . . .
Fig. 2. Virus-resistant
cotiigthe. TMV coat protein gen~e. Tomato cotyledons were trans-
fomd(83) with an Agrobactenum
straincnann a TMV coat protein
cDNA chimeric gene (51). Transgemic
protein production by inmunoblot analysis. The R, progeny of a repre-
snaieplant that expressed high levels of coat protein were anlayzed for
after iouaonwith TMV. The control plant on the left is a
segregant that lacks the TMV coat protein gene; the plant on the right has
inherited the gene. (B) Field test (1988) of tomato plants containing TMV
cotat protein gene. Controilandl transgrenic
~ ~ ~ ~ ~ ~ ~ ~ ~~ATCE
plats. (A) Greenhouse
evaluation of tomato plants
were screened for coat
seelings were growrn
Jre Couty Ilios
ek ae otefedts
lnswr h oto rgt
len to tha it
begin the wave of oppositions, appeals, and infringement actions
that have marked the early pharmaceutical patents in the biotechnol-
ogy area (78). Enforceability of plant patents in other countries,
including Japan, China, and Eastern Bloc countries, is questionable.
While there are numerous initiatives to harmonize both registration
and proprietary protection throughout the key trading countries in
the world, the outcome is not imminent and will be unlikely to have
an impact on first-generation products.
Public perception. Genetically engineered crops are being developed
at a time when a lack of understanding regarding the importance of
agricultural research exists. Current issues, including concerns about
(i) periodic, temporary production surpluses, (ii) changing farm
infrastructure, (iii) inconsistency in farm policies, and (iv) a general
distrust for new technologies, have at times overshadowed the long-
term need for the provision of economical, high-quality food
products for a growing world population. Currently, at the begin-
ning of the 1989 cropping season, world reserves of grain are at
their lowest level since the years immediately following World War
II; another drought in 1989 could create a world food emergency
Despite this background, recent polls conducted by the Office of
Technology Assessment indicate that most people believe that the
benefits of agricultural biotechnology research outweigh remote
risks (72). In view of the initial public debate that has occurred over
the last several years on field testing and environmental release of
genetically engineered organisms, it would seem that agricultural
biotechnology has indeed passed its first major public perception
The next test of the public acceptance of this technology will come
in several years when food products derived from genetically
engineered crops enter the general food supply. The current focus
on issues of risk and environmental release has heightened the need
for increased science education and open discussion of issues. It is
essential that the safety and benefits of agricultural biotechnology
research and the critical role that it will play in providing for world
food demand (80) be communicated and understood, so that
informed decisions by the public are possible.
A Future Perspective on Genetically
During the last 5 years, the availability of gene transfer systems
has catalyzed a major refocusing on plants as a biological system; the
use of genetically engineered plants as an analytical tool to explore
unique aspects of gene regulation and development and the poten-
tial to produce novel commercial crop varieties has created a high
level of scientific excitement and has driven research into many new
areas. The breadth of information to be gained from the study of
transgenic plants is serving as an important focus for unifying basic
plant science research in plant breeding, pathology, biochemistry,
and physiology with molecular biology. Regulation of gene expres-
sion is the fundamental basis for manipulating cellular metabolism,
and this new research tool offers the possibility of extending
physiological and genetic observations to a mechanistic level. In the
next few years we can expect to see major advances in our
understanding of basic plant processes.
These advances, in turn, will accelerate the application of geneti-
cally engineered plants in the seed production and agrichemical
industries. The major crops that can currently be improved with
genetic techniques are soybean, cotton, rice, and alfalfa (Table 1),
and commercial introductions of genetically engineered varieties are
likely in the mid- 1990s. Rapid progress is being made in the genetic
engineering of corn, and it is likely that genetically engineered corn
hybrids carrying traits for resistance to herbicides, insects, and viral
diseases will reach the marketplace by the year 2000. The timing of
commercialization of genetically engineered crops is ultimately
determined by the need to address each of the following issues: (i)
evaluation of field performance, (ii) breeding and seed increase for
commercial-scale release, (iii) establishment of optimal agronomic
practices, and (iv) regulatory approval and crop certification.
The worldwide agrichemical industry has been and will continue
to be a leading sponsor of agricultural biotechnology research. All
major agrichemical companies have R&D efforts in the area of
biotechnology for crop improvement. These companies see oppor-
tunities to develop new products and extend the use of existing
products, as well as to be positioned at the leading edge of new
technologies that may have a significant impact on existing agri-
Genetic engineering of plants also offers exciting opportunities for
the food processing industry to develop new products and more
cost-effective processes. While many of the early successful examples
of genetically engineered plants have focused on agronomic genes, it
is possible that the food processing and specialty chemical industries
may represent the greatest commercial opportunity for biotechnolo-
gy. Examples of such applications include production of (i) larger
quantities of starch or specialized starches with various degrees of
branching and chain length to improve texture and storage proper-
ties, (ii) higher quantities of specific oils or the elimination of
particular fatty acids in seed crops, and (iii) proteins with nutrition-
ally balanced amino acid composition. The ability to reduce process-
ing costs by the elimination of anti-nutritive or off-flavor compo-
nents in foods is quite feasible with antisense nucleic acid technolo-
gy. The enzymes and genes involved in biosynthesis of coloring
materials and flavors are important to the food industry and to the
consumer. Studies on the biosynthesis of some of these compounds
have been hampered by the low quantities of enzymes present in the
producing cells, but new techniques based on gene tagging may
overcome these difficulties.
Enormous opportunity lies in the successful use of crops for both
commodity and specialty chemical products. Plants have traditional-
ly been a source of a wide range of polymeric materials. These range
from starch and celluloses, which are carbohydrate-based, to polyhy-
drocarbons such as rubber and waxes. Many of these polymers have
been replaced in the last two to three decades by synthetic materials
derived from petroleum-based products. However, the cost, supply,
and waste-stream problems often associated with petroleum-based
products are issues that are focusing renewed attention on the use of
biological polymers. Genetic engineering will significantly enlarge
the spectrum and composition of available plant polymers.
Plants also offer the potential for production of foreign proteins
with various applications to health care. Proteins such as neuropep-
tides, blood factors, and growth hormones could be produced in
plant seeds, and this may ultimately prove to be an economical
means of production. Several mammalian proteins have been pro-
duced in genetically engineered plants (81), and expression of
pharmaceutical peptides in oilseed rape plants has been reported
REFERENCES AND NOTES
1. R. B. Horsch et al., Science 223, 496 (1984); M. De Block, L. Herrera-Estrella,
M. Van Montagu, J. Schell, P. Zambryski, EMBOJ. 31, 681 (1984).
2. R. B. Horsch et al., Science 227, 1229 (1985).
3. M.-D. Chilton et al., Cell 11, 263 (1977).
4. E. W. Nester, M. P. Gordon, R. M. Amasino, M. F. Yanofsky, Annu. Rev. Plant
Physiol. 35, 387 (1984); G. Gheysen, P. Dhaese, M. Van Montagu, J. Schell, in
Genetic Flux in Plants, B. Hohn and E. S. Dennis, Eds. (Springer-Verlag, New
York, 1985), pp. 11-47; P. Zambryski, J. Tempe, J. Schell, Cell 56, 193 (1989).
5. K. A. Barton et al., Cell 32, 1033 (1983).
SCIENCE, VOL. 244
6. R. T. Fraley et al., Bio/Technology 3, 629 (1985).
7. H. Klee and S. G. Rogers, in Plant DNA Infectious Agents, T. Hohn and J. Schell,
Eds. (Springer-Verlag, New York, 1985), pp. 179-203.
8. S. G. Rogers et al., Methods Enzymol. 153, 253 (1987).
9. A. Spielmann and R. B. Simpson, Mol. Gen. Genet. 205, 34 (1986); R.
Jorgensen, C. Snyder, J. D. G. Jones, ibid. 207, 471 (1987).
10. W. Mark, K. Signorelli, E. Lacy, Cold Spring Harbor Symp. Quant. Biol. 50, 453
(1985); L. Covarrubias, Y. Nishida, B. Mintz, Proc. Natl. Acad. Sci. U.S.A. 83,
11. I. Potrykus et al., in Plant DNA Infectious Agents, T. Hohn and J. Schell, Eds.
(Springer-Verlag, New York, 1985), pp. 229-247.
12. G. Gheysen et al., Proc. Natl. Acad. Sci. U.S.A. 84, 61 (1987).
13. M. Wallroth et al., Mol. Gen. Genet. 202, 6 (1986); S. C. Deroles and R. C.
Gardner, Plant Mol. Biol. 11, 355 (1988).
14. X. Delannay, private communication.
15. B. Bytebier et al., Proc. Natl. Acad. Sci. U.S.A. 84, 5345 (1987).
16. W. Schaifer, A. G6rz, G. Kahl, Nature 327, 529 (1987).
17. N. Grimsley, T. Hohn, J. W. Davies, B. Hohn, ibid. 325, 177 (1987).
18. I. Potrykus et al., Mol. Gen. Genet. 199, 183 (1985); H. Lorz, B. Baker, J. Schell,
ibid., p. 178; M. Fromm, L. P. Taylor, V. Walbot, Nature 319, 791 (1986); H.
Uchimiya et al., Mol. Gen. Genet. 204, 204 (1986).
19. J. Callis, M. Fromm, V. Walbot, Genes Dev. 1, 1183 (1987); W. R. Marcotte, Jr.,
C. C. Bayley, R. S. Quatrano, Nature 335, 454 (1988).
20. T. Fujimura, M. Sakurai, H. Akagi, T. Negishi, A. Hirose, Plant Tissue Culture
Lett. 2, 74 (1985); K. Toriyama, K. Hinata, T. Sasaki, Theor. Appl. Genet. 73, 16
(1986); Y. Yamada, Z. Q. Yang, D. T. Tang, Plant Cell Rep. 4, 85 (1986); R.
Abdullah, E. C. Cocking, J. A Thompson, BiolTechnology 4, 1087 (1986).
21. K. Toriyama et al., BiolTechnology 6, 1072 (1988).
22. C. Rhodes, K. Lowe, K. Ruby, ibid., p. 56.
23. C. A. Rhodes et al., Science 240, 204 (1988).
24. I. K. Vasil, BiolTechnology 6, 397 (1988).
25. T. M. Klein, E. D. Wolf, R. Wu, J. C. Sanford, Nature 327, 70 (1987).
26. T. M. Klein et al., Proc. Natl. Acad. Sci. U.S.A. 85, 8502 (1988).
27. D. E. McCabe et al., Bio/Technology 6, 923 (1988).
28. T. M. Klein, B. A. Roth, M. E. Fromm, in Genetic Engineering, vol. 11, J. K.
Setlow, Ed. (Academic Press, New York, in press).
29. G. Zhou et al., Methods Enzymol. 101, 433 (1983); D. Hess, Intern Rev. Cytol.
107, 367 (1987); Z.-x. Luo and R. Wu, Plant Mol. Biol. Reporter 6, 165 (1988).
30. A. de la Pefia, H. L6rz, J. Schell, Nature 325, 274 (1987).
31. G. Neuhaus, G. Spangenberg, 0. Mittelsten Scheid, H.-G. Schweiger, Theor.
Appl. Genet. 75, 30 (1987).
32. R. T6pfer, B. Gronenborn, J. Schell, H.-H. Steinbiss, Plant Cell 1, 133 (1989).
33. C. Benbrook and P. Moses, in Proceedings: BioExpo 1986 (Butterworth, Stoneham,
MA, 1986), pp. 27-54.
34. H. Steinricken and N. Amrhein, Biochem. Biophys. Res. Commun. 94, 1207
(1980); D. M. Mousdale and J. R. Coggins, Planta 160, 78 (1984).
35. D. Atkinson, in Glyphosate, E. Grossbard and D. Atkinson, Eds. (Butterworth,
London, 1985), pp. 127-133; L. Torstensson, ibid., pp. 137-150.
36. D. M. Shah et al., Science 233, 478 (1986).
37. L. Comai et al., Nature 317, 741 (1985).
38. M. A. Hinchee et al., BiolTechnology 6, 915 (1988).
39. G. W. Haughn et al., Mol. Gen. Genet. 211, 266 (1988).
40. M. De Block et al., EMBOJ. 6, 2513 (1987).
41. D. M. Stalker, K. E. McBride, L. D. Malyj, Science 242, 419 (1988).
42. W. De Greef et al., BiolTechnology 7, 61 (1989).
43. H. T. Dulmage, in Microbial Control of Pests, H. D. Burgess, Ed. (Academic Press,
New York, 1981), pp. 193-222; A. I. Aronson, W. Beckman, P. Dunn, Microbiol.
Rev. 50, 1 (1986).
44. S. A. McPherson et al., Bio/Technology 6, 61 (1988).
45. T. Yamamoto and R. E. McLaughlin, Biochem. Biophys. Res. Commun. 103, 414
(1981); L. J. Goldberg and J. Margalit, Mosquito News 37, 355 (1977).
46. A. Llausner, BiolTechnology 2, 408 (1984); D. R. Wilcox et al., in Protein
Engineering, M. Inouye and R. Sarma, Eds. (Academic Press, New York, 1986), p.
47. V. F. Sacchi et al., FEBS Lett. 204, 213 (1986).
48. D. A. Fischhoffet al., BiolTechnology 5, 807 (1987); M. Vaeck et al., Nature 238,
33 (1987); D. A. Fischhoff, private communication.
49. X. Delannay et al., BiolTechnology, in press.
50. V. A. Hilder et al., Nature 330, 160 (1988).
51. P. Powell-Abel et al., Science 232, 738 (1986).
52. N. E. Turner et al., EMBO J. 6, 1181 (1987); C. Hemenway, R.-X. Fang, W. K.
Kaniewski, N.-H. Chua, N. E. Turner, ibid. 7, 1273 (1988); M. Cuozzo et al.,
Bio/Technology 6, 549 (1988); A. Hoekema, M. J. Huisman, L. Molendijk, P. J.
M. van den Elzen, B. J. C. Cornelissen, ibid. 7, 273 (1989); C. M. P. Van Dun, J.
F. Bol, L. Van Vloten-Doting, Virology 159, 299 (1987); E. C. Lawson and P. R.
Sanders, private communication.
53. J. C. Register and R. N. Beachy, Virology 166, 524 (1988).
54. R. S. Nelson et al., Bio/Technology 6, 403 (1988).
55. J. D. G. Jones et al., Mol. Gen. Genet. 212, 536 (1988).
56. F. Nagy, G. Morelli, R. T. Fraley, S. G. Rogers, N.-H. Chua, EMBO J. 4, 3063
(1985); R. Fluhr and N.-H. Chua, Proc. Natl. Acad. Sci. U.S. A. 83,2358 (1986).
57. M. A. Anderson et al., Nature 321, 38 (1986); C. S. Gasser, K. A. Budelier, A. G.
Smith, D. M. Shah, R. T. Fraley, Plant Cell 1, 15 (1989).
58. C. Sengupta-Gopalan et al., Proc. Natl. Acad. Sci. U.S.A. 82, 3320 (1985); R. N.
Beachy et al., EMBOJ. 4, 3047 (1985).
59. D. P. 5. Verma and R. B. Goldberg, Eds., Temporal and Spatial Regulation of Plant
Genes (Springer-Verlag, New York); R. B. Goldberg, Science 240, 1460 (1988);
P. N. Benfey and N.-H. Chua, ibid. 244, 174 (1989).
i6 JUNE i989
60. R. Broglie et al., ibid. 224, 838 (1984); D. E. Rochester, J. A. Winter, D. M.
Shah, EMBOJ. 5, 451 (1986); A. Theologis, Annu. Rev. Plant Physiol. 37, 407
(1986); M. M. Sachs and T.-H. D. Ho, ibid., p. 363 (1986); R. W. Thornberg,
G. An, T. E. Cleveland, R. Johnson, C. A. Ryan, Proc. Natl. Acad. Sci. U.S. A. 84,
744 (1987); J. G6mez et al., Nature 334, 262 (1988); N. M. Crawford, M. Smith,
D. Bellissimo, R. W. Davis, Proc. Natl. Acad. Sci. U.S.A. 85, 5006 (1988); C. J.
Lamb, M. A. Lawton, M. Dron, R. A. Dixon, Cell 56, 215 (1989).
61. C. J. S. Smith et al., Nature 334, 724 (1988); R. E. Sheehy, M. K. Kramer, W. R.
Hiatt, Proc. Natl. Acad. Sci. U.S.A. 85, 8805 (1988).
62. A. R. van der Krol et al., Nature 333, 866 (1988).
63. J. Haseloff and W. L. Gerlach, ibid. 334, 585 (1988).
64. J. Paszkowski, M. Baur, A. Bogucki, I. Potrykus, EMBO J. 7, 4021 (1988).
65. N. V. Federoff, D. B. Furtek, 0. E. Nelson, Proc. Natl. Acad. Sci. U.S.A. 81,
3825 (1984), K. C. Cone, F. A. Burr, B. Burr, ibid. 83, 9631 (1986); E. M.
Rinchik, L. B. Russell, N. G. Copeland, N. A. Jenkins, Genetics 112, 321 (1986);
R. J. Schmidt, F. A. Burr, B. Burr, Science 238, 960 (1987).
66. B. Baker, G. Coupland, N. Federoff, P. Starlinger, J. Schell, EMBOJ. 6, 1547
(1987); M. A. Van Sluys, J. Tempe, N. Federoff, ibid., p. 3881.
67. K. A. Feldmann et al., Science 243, 1351 (1989).
68. R. Bernatzky and S. D. Tanksley, Genetics 112, 887 (1986); T. Helentjaris, Trends
Genet. 3, 217 (1987); B. S. Landry, R. Kesseli, H. Leung, R. W. Michelmore
Theor. Appl. Genet. 74, 646 (1987); B. Burr, F. A. Burr, K. H. Thompson, M. C.
Albertson, C. W. Stuber, Genetics 118, 519 (1988); C. Chang et al., Proc. Natl.
Acad. Sci. U.S.A. 85, 6856 (1988).
69. A. H. Paterson et al., Nature 335, 721 (1988); S. D. Tanksley and J. Hewit, Theor.
Appl. Genet. 75, 811 (1988).
70. R. E. Pruitt and E. M. Meyerowitz, J. Mol. Biol. 187, 169 (1986).
71. H. J. Klee, M. B. Hayford, S. G. Rogers, Mol. Gen. Genet. 210, 282 (1987).
72. U. S. Congress, Office of Technology Assessment, New Developments in Biotechnol-
ogy-Field Testing Engineered Organisms: Genetic and Ecological Issues, OTA-BA-350
(U.S. Government Printing Office, Washington, DC, May 1988).
73. Council of the NatioAal Academy of Sciences, Introduction of Recombinant DNA-
Engineered Organisms into the Environment: Key Issues (National Academy Press,
Washington, DC, 1987); Boyce Thompson Institute for Plant Research, Regula-
tory Considerations: Genetically Engineered Plants (Center for Science Information,
San Francisco, CA, 1987)..
74. International Food Biotechnology Council, Food and Chemical News, 6 February
1989, pp. 10-12; Expert Panel on Criteria for Determining the Regulatory
Status of Food and Food Ingredients Produced by New Technologies, Federal
Register, vol. 53, no. 168, 30 August 1988.
75. Diamond v. Chakrabarty, 447 U.S. 303 (1980).
76. Ex parte Hibberd, United States Patents Quarterly 227, 303 (1985).
77. J. Johnson, Seedsmen's Digest, 6 to 8 October 1987; Executive Committee of the
American Seed Traders Association, Diversity 14, 27 (1988).
78. B. J. Fowlston, BiolTechnology 6, 911 (1988).
79. L. R. Brown, Worldwatch Paper 85 (Worldwatch Institute, Washington, DC,
80. Council for Agricultural Science and Technology, Report No. 114 (University of
Iowa, Ames, IA, 1988).
81. D. A. Eichholtz et al., Somat. Cell. Mol. Genet. 13, 67 (1987); D. D. Lefebvre, B.
L. Miki, J.-F. Lalibertd, Bio/Technology 5, 1053 (1987); A. Barte et al., Plant Mol.
Biol. 6, 347 (1986).
82. J. Vandeleerckhove et al., BiolTechnology, in press.
83. S. McCormick et al., Plant Cell Rep. 5, 81 (1986).
84. G. Ooms, M. M. Burrell, A. Karp, M. Bevan, J. Hille, Theor. Appl. Genet. 73,
744 (1987); S. Sheerman and M. W. Bevan, Plant Cell Rep. 7, 47 (1988).
85. J. Paszkowski et al., EMBO J. 3, 2717 (1984).
86. A. M. Lloyd et al., Science 234, 464 (1986).
87. R. Michelmore, E. Marsh, S. Seely, B. Landry, Plant Cell Rep. 6, 439 (1987).
88. N. P. Evrette et al., Bio/Technology 5, 1201 (1987).
89. J. E. Fry, A. Barnason, R. B. Horsch, Plant Cell Rep. 6, 321 (1987). E.-C. Pua, A.
Mehra-Palta, F. Nagy, N.-H. Chua, Bio/Technology 5, 815 (1987).
90. N. Basiran, P. Armitage, R. J. Scott, J. Draper, Plant Cell Rep. 6, 396 (1987); M.
C. Jordan and A. McHughen, ibid. 7, 281 (1988).
91. P. Umbeck, G. Johnson, K. Barton, W. Swain, BiolTechnology 5, 263 (1987); E.
Firoozabady et al., Plant Mol. Biol. 10, 105 (1987).
92. A. R. Barnason and J. E. Fry, private communication.
93. D. Catlin et al., Plant Cell Rep. 7, 100 (1988).
94. E. A. Shahin et al., Crop Sci. 26, 1235 (1986).
95. M. Deak, G. B. Kiss, C. Koncz, D. Dudits, Plant Cell Rep. 5, 97 (1986).
96. H. P. Hernalsteens et al., EMBOJ. 3, 3039 (1984).
97. F. K6hler, C. Golz, S. Eapen, H. Kohn, 0. Schieder, ibid. 6, 313 (1987).
98. A. J. Trulson, R. B. Simpson, E. A. Shahin, Theor. Appl. Genet. 73, 11 (1986).
99. C. David and M.-D. Chilton, J. Temp6, BiolTechnology 2, 73 (1984).
100. C. David and J. Temp6, Plant Cell Rep. 7, 88 (1988).
101. T. Noda et al., ibid. 6, 283 (1987).
102. D. Tepfer, Cell 37, 959 (1984).
103. J. J. Fillatti et al., Mol. Gen. Genet. 206, 192 (1987); F. Pythoud, V. P. Sinkar, E.
W. Nester, M. P. Gordon, BiolTechnology 5, 1323 (1987).
104. G. H. McGranahan et al., BiolTechnology 6, 800 (1988).
105. D. J. James et al., Plant Cell Rep. 7, 658 (1989).
106. M. E. Horn, R. D. Shillito, B. V. Conger, C. T. Harms, ibid., p. 469.
107. We thank X. Delannay, D. Fischhoff, P. Sanders, C. Lawson, A. Barnason, J. Fry,
and J. Leemans for giving us perrnision to cite their work prior to publication. We
thank G. Kishore, S. Rogers, H. Klee, J. Callis, D. Gunning, N. Turner and S.
Brown for help with the content, readability, and references in the manuscript.
Special thanks to all other members of the Plant Molecular Biology Group at