0780-B4
A.D. Yanchuk and B. Jaquish 1
Although advances in forest biotechnology have produced important products and discoveries, there are several associated problems surfacing from these technologies for forest genetic resource managers. The new technologies attract substantial attention from private investors and government funding agencies, which can detract from funding traditional genetic management programmes. In many cases, the proprietary nature of such investments restricts the flow of information back to the scientific community. Lastly, the products of genes are not simple entities. Most gene products have small effects and complex interactions that greatly affect the phenotype. For traits managed at the field level, population levels of management will still be required regardless of whether the products arise through biotechnology or traditional breeding.
The forest genetic resource management system developed in British Columbia attempts to consider the dynamic nature of genetic variation and the mechanisms that both maintain and change it. Research in tissue culture, transgenics and molecular genetics will undoubtedly continue to provide tree breeders and forest genetic resource managers with new research tools. In a few situations, they may even have direct economic applications. However, the details of how traits are affected by genetic variation and how this variation must be managed in space and time will only come from the long-term evaluation of experimental trees in the field. Local, national and international funding agencies and all levels of government need to acknowledge that biotechnology should only be viewed in a supporting role to traditional plant improvement approaches, not as any kind of a replacement.
Over the past two decades, a rapid expansion of research in biotechnology has provided many new techniques and opportunities. This trend is likely to continue. However, various aspects of forest biotechnology (e.g., tissue culture, molecular genetics and transgenics) create potential problems for forest genetic resource managers. First, there is the basic problem of fitting the evolutionary development of populations together with single-gene technologies. Molecular genetic techniques tend to force a reduction rather than an expansion in the use of available genetic variation (Namkoong et al. 2002). Second, new genetic modification techniques raise several issues around socio-economic consequences (e.g., farmers' rights) and in equitable development of genetic resources (e.g., government support for traditional breeding versus higher technology). Moreover, the advantages of new technologies will usually belong only to those who have invested in them. Finally, biotechnology has created a large machinery around variety property rights or patents. This tends to prevent widespread dissemination of the results and materials (Santos and Lewontin 1997), not to mention a diversion of scarce funds to legal costs.
A large number of plant manipulation techniques can be referenced under the label of forest biotechnology; however, three main areas will likely be important in forestry: 1) tissue culture, 2) transgenics or genetically modified trees (GMT's), and 3) molecular genetic markers.
Vegetative propagation comprises a broad range of techniques involving manipulation of plant tissue that will ultimately allow for complete vegetative `re-propagation' of the whole plant, i.e., the production of a clonal `variety' or line. In forestry, most commercial scale cloning methods have relied on rooted cuttings, which has been largely uncontroversial. However, advanced micropropagation, such as somatic embryogenesis (SE), presents many new opportunities for clonal forestry with forest trees. It is also the basic plant type required for the development of GMT's. Cost aside, SE has all the necessary attributes of becoming a valuable research and breeding tool. Long-term storage of lines is possible and large-scale production from field-tested clonal lines is likely in a few species (e.g., loblolly and radiata pine). However, the development costs of these advanced tissue culture technologies are high compared to simpler techniques. Moreover, it is not yet clear if clonal genetic gains will justify the investments.
To be of economic value, GMT's must offer unique features that cannot be economically delivered through conventional breeding programmes. They must also be capable of offsetting the costs in developing the technology.
To date, the traits that have been considered for genetic modification (e.g., herbicide resistance, reduced flowering or sterility, insect resistance and wood chemistry) all have potential commercial value. However, they will likely be important for a very limited number of species in very few situations around the world. It is important to note that conventional tree breeding programmes have also made significant improvement in insect resistance and wood quality traits.
Currently, in the development of insect resistant plants, research is underway to reduce the reliance on the relatively narrow group of natural Bt toxins by examining other compounds (ffrench-Constant and Bowen, 1999). Nevertheless, the complex ecological ramifications and public concerns regarding the release of these proteins into food chains will dictate that high levels of scientifically sound laboratory and field testing programs are in place. In wood science, genes important in the pathway of lignin development have been modified to produce unique wood in very young trees (e.g. Akim et al. 2001). In this case, two important questions will linger regarding the development of lignin-modified varieties or clones: 1) "How much extra value will there be in plantations using such trees?" and 2) "Does altered wood show susceptibilities to environmental stresses?". Once again, substantial periods of field testing will be required to answer such questions.
The primary ecological concern with the deployment of GM plants appears to be from potential problems arising from gene exchange with wild populations. However, in many situations these GM trees will be deployed only as exotic plantation species, so gene exchange may not be a major concern. If GM trees are deployed locally, reduced flowering or sterility is likely to be a basic requirement, and deployment restrictions, over and above the concerns arising from clonal forestry per se, will be necessary.
Genetic markers are essentially DNA sequences that are indicative of common ancestry. With correct interpretation, genetic markers are invaluable for examining patterns of genetic variation among and within populations, assessing levels of outcrossing and inbreeding, and genetic identification or `fingerprinting'. A large amount of genetic marker research has been directed toward indirect early selection of better genotypes (i.e., marker-assisted selection or MAS). However, MAS has had limited commercial successful in animal, crop or tree breeding for two main reasons. First, the additional genetic gains are generally not large enough to offset the costs of developing and applying the technology, and second, most traits of commercial interest are affected by a large number of genes each with small effect. Therefore, it is likely that MAS will only be useful for a handful of species and situations, and in very few tree breeding applications. Instead, the real value of molecular genetics in areas such as detecting quantitative trait loci (QTL's) is likely to be in locating candidate regions for investigation of the genes sets that affect measurable traits (Namkoong et al. 2002).
Within the scientific community, it is reasonably well accepted that it will be the new, or novel, gene products that pose a risk, and not the technology used to obtain them. However, to fully evaluate these risks, in-depth knowledge of the particular genetic transformation is necessary. What are the protein products being produced? What are their possible interactions with other gene products? How do they interact and express themselves in the developing organism? To further complicate matters, in most cases we cannot predict accurately the phenotypes of GM or non-GM genotypes without adequate testing in different environments. We will then find ourselves comparing the expression of one, two or three-gene GM genotypes, probably with non-transformed genotypes, and both requiring long-term field testing in several relevant environments. The problem of genotype-environment interaction becomes increasingly important as higher technologies are used and capital investment increases (Lewontin 1977).
To date, the major finding of molecular genetics is that genes, although simple structures on their own, vary and change due to many genomic mechanisms. They produce products that are anything but simple and do not have simple effects on phenotypes. Therefore, the management of gene effects that influence traits such as yield and resistance must be managed at stand and landscape levels, and still require population levels of management.
There are several levels of concern in the conservation of forest genetic resources and tree breeding. However, in terms of population sampling, these generally reduce to two categories, in situ and ex situ. The impetus for attempting to build integrated strategies for the conservation of rare and low-frequency genes, and the development of genetic resources through breeding populations, comes from the fact that most organizations will be challenged by financial limitations to manage many large ex situ populations. This is certainly true in British Columbia, and one can only imagine how true it is in other jurisdictions where resources and expertise are scarce, and social and ecological settings are more complex. In British Columbia, we currently manage 40 programs with ex situ collections in 12 species, as well as insuring that in situ reserves are in place to support these and other species not in development programs (Yanchuk 2001).
We view the problem of conserving genes in breeding programs as substantially different from conserving genes in non-select populations. This is because we are actively changing the frequency of genes at "targeted" loci. Therefore, we need to manage the depletion of genetic variation under strong selection, rather than erosion through ignorant neglect. More importantly, by changing gene frequencies across many gene loci and across multiple populations, we expect that variation will be maintained and will likely even increase in early generations. This provides us with a means to diversify populations and leave options open for unknown future climate, environmental and economic needs.
Our confidence in breeding and developing several small populations for each species is based on two factors. First, it is well known that even in small populations the loss in heterozygosity (H), or additive genetic variance, is slow. The loss of H is predicted by, HN = Ht (1 - 1/2Ne)N (where N is number of generations, t is the initial time, and Ne is the effective population size). For example, if we maintain an effective population size as small as 10, over 10 generations, we expect to lose only ~40% of the original genetic variation. Second, in short-term selection experiments (i.e., 5-10 generations), population sizes of a few dozen have been shown to be adequate to achieve expected gains (Namkoong et al. 1988, pg. 63). New genetic variations arising from mutation (Barker 1995) and other mechanisms that generate or maintain genetic variation (Rasmusson and Phillips 1997) appear to maintain genetic variation even after strong selection in early generations of selection and breeding.
Therefore, a sensible strategy of the preservation of genetic resources must take into account this selectable variance and the evolutionary dynamics that maintain it across many gene loci (Namkoong et al. 2002). For species in which we will not develop breeding or pre-breeding populations, we must rely on an array of in situ gene conservation strategies. The principles for managing these populations differ only by the extent of different selection pressures being applied (artificial versus natural), and the need for large populations to maintain lower frequency genes.
One challenge in forest gene conservation will be the desire to utilise genes from in situ conservation populations once breeding populations have moved substantially past the phenotypes of wild relatives. Nevertheless, without the option to consider these important genes in the future, we fail to meet our basic gene stewardship responsibilities. In these situations, transgenics may have a role to play, but this would be highly dependent upon the local situation.
Molecular genetics, or genomics research, will likely help us understand the complexity of gene expression, epistasis, and give more clarity to quantitative genetic models. However, as stated earlier, the development of an increased understanding of gene action and the number of genes affecting traits will require the co-operation of developmental biologists, physiologists, and tree breeders with long-term field testing under a variety of environments. Most certainly this should not come at the expense or exclusion of material developed by the traditional breeding methods.
Private investors have taken the lead for many investments in modern biotechnology, and in so doing have accepted the associated economic risks. In many instances these investment risks are patent protected and new knowledge is withheld from science for corporate advantage. For others to use or share the technology or material can be prohibitively expensive. To help offset these concerns, the role of governments in research and development may have to expand so as to provide a flow of material and information that can be used and shared by both private and public institutions (Santos and Lewontin, 1997). Regardless, the allocation of funds, whether through private or public agencies, needs to achieve a balance between building scientific capabilities and knowledge, and supporting more applied, well proven forestry technologies (Burdon, 1994). In this regard, the investment and use of any biotechnology needs to be assessed carefully on a case-by-case basis.
Public and government acceptance of GM plants, clonal forestry, or even products from traditional breeding programs is now as dependent on biological risk assessment and risk issue management as it is on technical or economic issues (Leiss, 1999). The idea that GM trees might be functionally analogous to some invasive exotic species does not seem likely (Strauss and Bradshaw, 2001); however, in all types of improved forest trees, potential risks must be balanced with benefits. All deployment strategies must be designed to minimize the risk of economic losses (e.g., stand volume, Roberds and Bishir, 1997),) as well as future biological losses (e.g., inappropriate use of resistance genes and the subsequent development of virulence in pest and disease populations). Again, this is irrespective of the technology used to develop the plant products. Other forest management decisions with potentially more serious ecological consequences than deployment of transgenic trees could include the introduction of exotic species, the inappropriate use of untested provenances, or the failure to establish seed transfer guidelines for local populations using sound genecological research.
To help address these issues around deployment, several countries have developed regulations and restrictions specifying the requirements of confined field testing needed before commercial release of GM plants (OECD, 2000). These requirements will undoubtedly continue to evolve, as will national laws and regulations, and other broader international agreements on biosafety, e.g. the Cartagena Protocol on Biosafety (CBD, 2000).
The advancements in genetic transformation, combined with delivery systems that impose the use of limited sets of clones, force forest genetic resource managers further into the clonal forestry problem. Social concerns about GMT's throughout the world, will likely elevate all aspects of clonal forestry, even if GMT's are not in the local discussions. Ironically, despite all the attention the issue is receiving, no commercial scale plantations of GM trees will be available in B.C., or Canada, for many years.
For species, populations and traits managed at the field level, population levels of management will be required for products of both traditional plant breeding and biotechnology. In British Columbia, the economic realities of relatively slow growth rates and long generation intervals will continue to be a major challenge to investors in biotechnology. However, it is clear that biotechnology may have a role in a limited number of places and for a limited number of species. It is important that investments in biotechnology not detract from the greater needs of allocating resources to many of the neglected species that may have unique features or benefits much greater than any single gene technology is likely to provide. Local, national and international funding agencies need to examine seriously the benefits associated with each investment. Moreover, they require appropriate levels of political support in evaluating proposals and awarding scarce resources.
Akim, L.G., Argyropoulos, D.S., Jouanin, L., Leple, J.C., Pilate, G., Pollet, B. & Lapierre, C. 2001. Quantitative P-31 NMR spectroscopy of lignins from transgenic poplars. Holzforschung. 55:386-390.
Barker, J.S.F. 1995. Quantitative genetic models: past, present and future. In: Proc. 25th Canadian Tree Improvement Assoc., Victoria, B.C. pp. 3-17.
Burdon, R.D. 1994. The role of biotechnology in forest tree breeding. Forest Genetic Resources, 22: 2-5. FAO, Rome.
CBD 2000. Cartagena Protocol on Biosafety. Convention on Biological Diversity. Available on the Internet at http://www.biodiv.org/biosafe/Protocol/Protocol.html
ffrench-Constant, R. & Bowen, D. 1999. Photorhabdus toxins: novel biological insecticides. Current Opinion in Microbiology, 2: 284-288.
Leiss, W. 1999. The trouble with science: public controversy over genetically-modified foods. Presented at the Eastern Regional Meeting of the Canadian Society of Plant Physiologists, Kingston, Ontario, Canada, 12 December. Available on the Internet at http://www.ucalgary.ca/~wleiss/news/trouble_with_science.htm.
Lewontin, R.C. 1977. The relevance of molecular genetics to plant and animal breeding. In: Pollack, E., Kempthorne, O. and T.B. Bailey Jr. (eds.). Proc. of the Int. Conf. on Quantitative Genetics, Iowa State Univ. Press, Ames. P. 55-62.
Namkoong, G., Kang, H.C. and J.S. Brouard. 1988. Tree breeding: principles and strategies. Monographs on Theor. and Applied Genetics 11. Springer-Verlag, New York.
Namkoong, G., Lewontin, R.C. and A.D. Yanchuk. 2002. Plant genetic resource management: the next investments in quantitative and qualitative genetics. (submitted, Genetic Resources and Crop Evolution).
OECD. 2000. Biotechnology regulatory developments in OECD member countries. Organisation for Economic Co-operation and Development (Internet document (http://www.oecd.org/ehs/country.htm).
Rasmusson, D.C. and R.L. Phillips. 1997. Plant breeding progress and genetic diversity from de novo variation and elevated epistasis. Crop Science. 37:303-310.
Roberds, J.H. & Bishir, J. 1997. Risk analysis in clonal forestry. Canadian Journal of Forest Research, 27: 425-432.
Santos, M. de Miranda & Lewontin, R.C. 1997. Genetics, plant breeding and patents: conceptual contradictions and practical problems in protecting biological innovations. IPGRI Plant Genetic Resources Newsletter, 112: 1-8.
Strauss, S.H. and H.D. Bradshaw. 2001. Tree biotechnology in the new millennium: International symposium on ecological and societal aspect of transgenic plantations. Oregon State University. (http://www.fsl.orst.edu/tgerc/iufro2001/eprocd.htm).
Yanchuk, A.D. 2001. A quantitative framework for breeding and conservation of forest tree genetic resources in British Columbia. Canadian Journal of Forest Research 31: 566-576.
1 Research Branch, British Columbia Forest Service, 712 Yates St. 3rd floor, Victoria, Canada. V8W 9C2. [email protected]