This chapter is an updated synthesis of Appendixes 2.1–2.4 of this Working Paper.
The overall objective of the FAO work programme on forest biotechnology is to contribute to a global assessment of the discipline’s status and trends. The purpose of this chapter is to summarize the findings and conclusions of four individual studies seeking to define the current state of forest biotechnology research and application. This ‘statistical survey’ of the types of biotechnologies being used globally will ultimately be coupled with analyses of economic and other forces driving biotechnology development in forestry to provide a comprehensive view of the current status of and trends in the discipline. This Working Paper is not intended to provide descriptive summaries of biotechnology tools, their precise classifications, areas of applications, or perceived contributions to forestry development. These topics have been extensively treated in other publications (see citations in references and annexes in this Working Paper).
Between 2002 and 2004, the FAO Forestry Department commissioned four studies to investigate the extent and patterns of biotechnology research and application on forest trees, worldwide. The studies are presented in this Working Paper (the Appendix numbers are indicated in brackets).
• Chaix, G and Monteuuis, O. CIRAD Forestry Department. Biotechnology in the forestry sector. 2004. (Appendix 2.1)
• Walter, C. and Killerby, S. New Zealand Forest Research. A global study on the state of forest tree genetic modification. 2003. (Appendix 2.2)
• Wang, H. Chinese Academy of Forestry. The state of genetically modified forest trees in China. 2003. (Appendix 2.3)
• El-Kassaby, Y. A. University of British Columbia. Anticipated contribution to and scale of impact of biotechnology in forestry. 2004. (Appendix 2.4)
The authors of these reports have compiled comprehensive information sources, including databases from questionnaires, systematic searches in CAB Abstracts and associated global scientific databases, Internet searches of public and private websites, and personal communications with scientists active in the field.
All attempts to be thorough and accurate have been made but capturing a sense of the status and trends of forest biotechnology research and application is a daunting task given the global spread of practitioners and the extremely rapid rate of progress in some technological areas such as genomics. Despite the authors’ best efforts, the FAO-sponsored studies reported here are sure to suffer from some largely unavoidable deficiencies such as:
• Inability to access, literally or in a timely manner, the vast literature and databases in Russian, Chinese and several other languages.
• Inability to access information on progress of and future plans for research and application in the private sector. This is particularly true for, but by no means exclusive to, R&D on genetic modification.
• Inability to stay abreast of rapid progress, the results of which may have immediate and large impacts on the science.
• Inability to report progress on work currently underway or completed, but not reported (often a lag of a few years).
• Inability to equate the number of citations of particular technologies with the actual resources invested in those technologies.
As an example of the last point, the number of citations and patents noted for somatic embryogenesis does not accurately reflect the several millions of dollars invested in the technology in North America alone over the last 15 years. In short, this Working Paper is likely to be a good indicator and possibly predictor of trends in forest biotechnology activity, but is equally likely to be lagging in defining the most current research activities and applied results. Future updates of the global status of forest biotechnologies will address many of the drawbacks pointed out here.
The current working definition of biotechnology used by the FAO is “any technological application that uses biological systems, living organisms, or derivatives thereof, to make or modify products or processes for specific use” (Convention on Biological Diversity4). In the broadest sense, this definition would include historical or traditional tree improvement technologies. For this report, the interpretation is narrowed to consider only modern DNA, molecular biology and plant reproductive technologies. Genetic modification or genetic engineering is defined here as the use of recombinant DNA and asexual gene transfer methods to alter the structure or expression of specific genes and traits.
Over 2 700 references representing the main forest biotechnology activities (publications of major research findings, field plantings, propagation programmes, etc.) reported within the last decade were compiled and synthesized in project surveys. The majority of these references (60 percent) were reported within the last 5 years, suggesting the scope and scale of biotechnology research and application is accelerating. Approximately 81 percent of the cited activities refer to biotechnologies excluding genetic modification. Chaix and Monteuuis
(see Appendix 2.1) differentiate three categories of these activities (Figure 1.1):
• marker development and estimates of diversity (referred to below as ‘diversity’);
• mapping, marker-assisted selection (MAS) and genomics (collectively referred to as MMG);
Figure 1.1 Proportion of biotechnology activities, by major categories, indicated in the public domain (from Chaix and Monteuuis, Appendix 2.1)
Forest biotechnology research and application is truly global in scope with activities identified in 76 countries. A significant majority of cited activities occurs in developed countries
(68 percent), with the United States (14 percent), France (9 percent) and Canada (8 percent) the most active participants represented in the data set (percentages are of total citations of main biotechnology activities). India (9 percent) and China (6 percent)5 were far and away the most active of the developing countries and countries in transition. Regionally, forest biotechnology activities were most numerous in Europe (39 percent), Asia (24 percent) and North America (23 percent), and least numerous in Oceania (6 percent), South America
(5 percent), Africa (3 percent) and the Near East (less than one percent).
While forest biotechnology research and application has spread to at least 140 genera, the great majority of activity (62 percent) has been focused on only six genera (Pinus, Eucalyptus, Picea, Populus, Quercus and Acacia, in descending order of activity).
Total activities in biotechnology excluding genetic modification were relatively evenly distributed between the categories identified by Chaix and Monteuuis (see above and Apppendix 2.1), diversity, MMG and micropropagation (32 percent, 26 percent and
42 percent, respectively), but became skewed when the number of genera studied was considered (diversity, 75 genera; MMG, 36 genera; micropropagation, 78 genera). Each category will be discussed at length in the following sections.
Biochemical and molecular markers play a significant role in many forest biotechnology activities, and marker development for trees has closely followed that for humans and agricultural crops. Isozymes, randomly amplified polymorphic DNAs (RAPDs) and restriction fragment length polymorphisms (RFLPs) have been widely used for genetic diversity and mapping studies, though the current trend favours microsatellites (nuclear and cytoplasmic) and AFLPs (amplified fragment length polymorphisms). Currently, ESTs (expressed sequence tags) and SNPs (single nucleotide polymorphisms) represent the most active area of marker development, though the total number of citations for these markers remains small suggesting the literature does not yet reflect the current status of this area of research. Work with these markers is being driven by large genomic and association genetics projects.
The vast majority of biotechnology activities in forestry appears to be restricted to the laboratory (99 percent, 97 percent and 94 percent for diversity, MMG and micropropagation, respectively) with some supporting field trials (one percent, 2.5 percent and 5 percent, respectively). Reported commercial application of biotechnologies appears to be restricted to micropropagation (but forms less than one percent of reported activities for that category), though it is known from non-public sources that commercial applications exist to a modest degree for all categories.
In general, reported activities in this category relate to the use of molecular markers to characterize amounts and patterns of genetic diversity in tree populations (studies on population structure), and to estimate the influence of human activities on forest genetic resources. More specifically, these activities include descriptive studies of (i) measures of diversity (57 percent of total activities in this category) and tree conservation with logging assessment (8 percent); (ii) tree mating systems, estimation of effective population size and gene flow within and among populations (15 percent); (iii) paternity analysis and fingerprinting applications (14 percent); and (iv) taxonomic and phylogenetic applications
(6 percent). Most studies targeting genetic diversity appear to be addressing and validating population genetic models or providing baseline data for the elaboration of conservation strategies. However, the application of these data for conservation purposes at the operational level is generally not obvious in the literature. Marker applications for fingerprinting and paternity analysis have found considerable utility in both basic science and commercial applications, though the commercial applications are not widely publicized (e.g. Lambeth et al. 2001).
Ninety-nine forest tree genera are represented in this category, though efforts are directed largely at the six noted in Section 1.5.1, Overview. In addition, Fagus, Abies, Betula, Castanea, Pseudotsuga and Larix have received considerable attention. While 66 percent of the studies on genetic diversity characterization occur in Europe or North America, studies of species originating from (native to) these regions represent only 37 percent of the total activities. The remaining distribution, by region, for the origin of species studied is evenly divided among Asia (19 percent), Oceania (New Zealand and Australia, 11 percent), South America (9 percent), Africa (4 percent) and the Near East (less than one percent), with
19 percent of species of non-ascertained origin, or hybrids.
Studies on genetic diversity characterization have been largely based on RAPD and AFLP markers in the past, with a gradually increasing use of nuclear and cytoplasmic microsatellites. Virtually all of these markers are considered neutral (not related to selection pressure) and may not be ideal for understanding genetic patterns influenced by evolutionary forces.
There is a small but growing trend to use SNP markers located in coding regions of candidate genes controlling phenotypic expression in adaptive traits to study the influence of evolutionary forces on allele frequencies.
During the decade of the 1990s significant biotechnology activity centred on the development of molecular markers, test populations, genetic linkage maps, and statistical means of identifying quantitative trait loci (QTLs). QTLs represent statistical associations between markers and genes that control some proportion of the genetic and phenotypic variation of a quantitative trait (generally less than 10 percent per QTL). QTLs have several potential applications including (i) genetic dissection of complex quantitative traits, (ii) providing the basis for MAS, and (iii) providing guidance for selection and prioritization of candidate genes (discussed later). Linkage and QTL maps have been created for over two dozen tree species6 and though more maps are likely to appear, most current efforts appear to focus on increasing the density and type of markers located on these maps. The current trend in MAS is towards the selection of superior alleles in candidate genes directly controlling phenotypic variation in traits of interest. This approach, termed association genetics, differs in application from traditional QTL studies primarily in the form of the test population being studied. Traditional methods use pedigreed populations for within-family selection while association studies rely on populations of unrelated individuals. Though MAS using QTLs has found utility for specialized populations of commercial species in a few developed and developing countries, association genetics holds promise for application across many populations, species and countries following appropriate development.
Over the last half-dozen years tremendous resources have been invested in genomics sciences, though this may not yet be reflected in the activities compiled here. Genomics encompasses a wide range of activities including gene discovery (ESTs), gene space and genome sequencing, gene function determination (database blast searches, expression profiling using arrays and slides, etc.), comparative studies among species, genera and families, physical mapping and the burgeoning field of bioinformatics. The ultimate goal of genomics is to identify every gene and its related function in an organism.
The recent completion of a whole-genome sequence for Populus (a project led by the United States Department of Energy) has laid the foundation for reaching this goal for a model species, and efforts follow to replicate this deed in Eucalyptus, though progress is slower. Public and private EST libraries for conifers now probably exceed one million entries.
The immediate applications of genomics include identification of candidate genes for association studies and targets for genetic modification studies. Also, comparative studies of ESTs from different trees have revealed the tremendous similarity among taxa throughout the conifers, and raise hope that what is learned from one species will benefit many others.
Finally, this category includes activities in the areas of proteomics and metabolomics, disciplines that currently enjoy very modest efforts in Europe and North America, but which are likely to expand in level of effort and geographic area in the near future.
Data set analyses, though confounded by the many technologies noted here, remain useful for tracking trends. MMG activities are conducted globally, but are concentrated in Europe
(43 percent) and North America (34 percent). About 40 genera have been studied, though four genera dominate the MMG landscape (Pinus, Populus, Eucalyptus, Picea).
Methodological approaches to MMG studies heavily favour gene discovery, functional genomics and candidate gene identification (65 percent of total studies), while linkage and QTL mapping account for 31 percent of activities. Proteomics activities are modest
(2 percent). Where identified, MMG studies have targeted primarily genes controlling wood property traits (57 percent) and resistance to abiotic (20 percent) and biotic (4 percent) stresses. Growth rate (6 percent), genetic diversity (8 percent) and flowering (5 percent) make up the remaining major categories.
Micropropagation is a term used here to describe methods of in vitro vegetative multiplication including rooted cuttings, organogenesis and somatic embryogenesis. Micropropagation is used to create large numbers of individual clones or genotypes. Because vegetative propagation bypasses the genetic mixing associated with sexual reproduction, it represents an ideal way to deliver genetic gain: select individuals are replicated precisely. The majority of biotechnology activities excluding genetic modification compiled by Chaix and Monteuuis (Appendix 2.1) (42 percent) relate to micropropagation.
Micropropagation by rooted cutting is commonly used in more than 20 species of commercial importance, the majority of which are angiosperms. Many of the activities noted suggest the technology is advanced and commercially viable. Conifers are less easily rooted than angiosperms, though modest programmes for several genera exist. Somatic embryogenesis is defined by an array of steps that result in the creation of embryos from somatic tissues
(as opposed to zygotic embryos from germinal cell lines). Though technically difficult, the technology has the potential to produce literally millions of genetically identical individual plants. It has received considerable R&D attention for highly valued conifer species, primarily in developed countries, for many years. Though large-scale commercial plantings of somatic embryos do not yet exist, progress in the technology appears promising and small-scale field testing is increasing (for example, on Pinus taeda in the United States). The delivery of somatic embryos to the field remains a significant hurdle to reducing plantlet costs and, therefore, large-scale use. Excellent progress in the creation of manufactured seed7 appears to provide a solution to this problem, though further research is likely to be needed.
Organogenesis, or the creation of plantlets from tissues such as cotyledons, has largely fallen out of favour in forestry operations and is used infrequently. In the future, the use of vegetatively propagated trees for intensively managed, high-yielding plantations is anticipated to increase significantly in all regions of the world. It is the author’s view that rooted cuttings are likely to dominate in angiosperm propagation in developing countries while somatic embryos will dominate in conifer propagation in developed countries.
Micropropagation activities occur in at least 64 countries in all regions of the world, though efforts are concentrated in Asia (38 percent of total), Europe (33 percent) and North America (16 percent). These activities include more than 80 genera of forest trees, the most commonly used being Pinus, Picea, Eucalyptus, Acacia, Quercus, Tectona, Populus and Larix.
A breakdown of activities within micropropagation suggests most reported work is within somatic embryogenesis (65 percent), along with cell and tissue culture (17 percent) and microcuttings (13 percent).
Few if any technological advancements in crop improvement methods have attracted as much attention from the scientific and lay communities as genetic modification. This is as true for forestry as it is for agriculture. In fact, genetic modification is so embedded in the public conscientiousness, that it is often considered synonymous with the term biotechnology. As detailed earlier, biotechnology encompasses much more than genetic modification, and publicly cited work on biotechnology activities excluding genetic modification significantly outnumbers genetic modification activities in the forestry sector (see Figure 1.1). However, in view of the intense scrutiny of and interest in genetic modification today, and its potential for significant influence on the ecological and economic landscapes, it is given disproportionate attention in this review.
The FAO-commissioned studies reviewed in this Working Paper provide detailed accounts of and insights into the state of genetic modification activities in world forestry as of early 2004. Walter and Killerby (Appendix 2.2) have summarized results of a detailed questionnaire sent to well over 500 institutions and scientists potentially engaged in or concerned with work on genetic modification in forestry. In addition, like Chaix and Monteuuis (Appendix 2.1), Walter and Killerby have searched public Internet databases, such as the APHIS (US Department of Agriculture – Animal and Plant Health Inspection Service) website for field trial permit applications in the United States, for any reference to work on genetic modification not typically reported in the literature. Collectively, these reports, taken together with information collated by Wang (Appendix 2.3), provide a reasonably consistent picture of the status of and trends in forestry research on genetic modification.
Forestry genetic modification activities are taking place in at least 35 countries, 16 of which host some form of experimental field trials (Figure 1.2). These field trials are generally very small (12 to 2 850 plants in reported studies) and typically of short duration. In many countries such trials must be destroyed before seed bearing occurs. In the remaining
15 countries, experimentation is restricted to laboratories or greenhouses. To date, only China (see Wang, Appendix 2.3) has reported the establishment of approved, commercial plantations of GM trees (discussed in subsequent sections).
Nearly two-thirds of the 520 reported genetic modification activities in forestry (see Chaix and Monteuuis, Appendix 2.1) occur in North America (48 percent) and Europe (32 percent). Asia follows with 14 percent of reported activities, Oceania with 5 percent, South America with one percent and Africa with less than one percent. The year 2004 has seen a sharp increase in field test applications for GM forest trees in the United States (Figure 1.3).
Populus was the first tree to be genetically modified (1986) and is by far the most commonly studied tree genus for genetic modification purposes today (47 percent of activities). This is no doubt a function of the ease with which some genotypes of the genus can be transformed and vegetatively propagated for experimental purposes. Pinus (19 percent), Eucalyptus
(7 percent), Liquidambar (5 percent) and Picea (5 percent) make up the majority of the remaining experimental studies. Field trials of GM trees are restricted largely to four top genera (Populus, 51 percent; Pinus, 25 percent; Liquidambar, 11 percent; and Eucalyptus,
Figure 1.2. Forest genetic modification activities worldwide
Figure 1.3. Field test applications for GM forest trees in the United States
Source: US Animal and Plant Health Inspection Service (APHIS) GM field test release permits database, according to APHIS notification categories.
Approximately half of all reported genetic modification activities with trees are related to methods development (e.g. gene stability, gene expression) or basic biological questions
(e.g. functional genomics, tissue culture). Of the remaining activities, ostensibly guided by commercial deployment objectives, herbicide tolerance (13 percent), biotic resistance
(12 percent), wood chemistry (9 percent) and fertility issues (6 percent) dominate the most studied groups of traits. This area will be discussed in greater detail later.
The questionnaire was designed not only to assess the current status of genetic modification in trees worldwide, but also to capture the views of practitioners on future developments and means to keep the public informed of their activities. A copy of the questionnaire is available from the FAO.
Of the 418 questionnaires sent out by Walter and Killerby8, 49 were returned (11.7 percent) and of these, 23 respondents were conducting research on genetic modification in forest trees9. For simplicity, these 23 respondents will be referred to as the ‘core’ respondents henceforth. Respondents did not include some private companies known to be conducting genetic modification activities in trees, most notably in North America where the majority of activity is reported. Regardless, the comments and trends noted, while not comprehensive, are likely to reflect those of the global research community. Of the core respondents, ten represented research institutions, nine universities, two private industries, and two were categorized as ‘other’. Two of the total respondents represented non-profit organizations concerned with appropriate use of biotechnology. Fifty-seven percent of the core respondents currently publish information about their work on the Internet.
The first three questions sought to determine the scope of genetic modification activities in contained laboratory experiments (68 reported), field-based experiments (26 reported) or commercial plantings (none reported), primarily with respect to traits and species evaluated. Of the 68 laboratory projects, all but two were identified as research targeted with the remaining two related to commercial applications. Four projects serviced both functions. Seventeen different gene/trait groups were identified in the laboratory, the majority of which were related to methods development or stability (markers, reporters, antibiotic resistance;
26 projects), reproductive development (19) or herbicide resistance (11). Other notable traits included wood properties (eight), insect resistance (seven) and wood chemistry (eight).
The species studied were largely as noted in Section 1.6.1, Overview, though the percentages of each varied slightly from the list generated by the complete data set. Most projects were reported to be working with more than one species and on more than one trait or gene. Eleven respondents reported a total of 26 projects with field trials, 20 of which have been established (1–3 per year) over 13 of the last 15 years. Of these 26, 20 (77 percent) involved Populus, two Pinus, and one each Betula, Eucalyptus and Picea. Markers (nine projects), herbicide resistance (six), lignin biosynthesis (five), insect resistance (three), and reproductive development (two) dominated the traits being investigated in field trials. Approximately half of the respondents with field trials identified the issues addressed by these trials to be related to plant growth/performance (five projects), gene expression stability (four), environmental risk assessment (three), horizontal gene transfer (two), herbicide applications (two) and four others (one each).
To judge future trends better, respondents were asked whether they had pending plans for future trials of GM trees. A total of 12 laboratory and field trials were specified for four genera in five countries, with establishment anticipated over the next 3 years.
One respondent from China reported the commercial release of two transgenic poplar genotypes transformed for leaf-eating insect resistance. The largest release was with Populus nigra transformed with the Bt gene cry1Ac. Following a successful 1-ha pilot planting in 1994, 80 ha of field trials were established on eight sites in seven provinces in 1999. Subsequent authorization from state regulating agencies permitted the establishment of ca 1 million trees in 2002, on some 300 ha. A smaller release with a hybrid poplar clone transformed with both cry1Ac and API followed in 2003. No other respondents to the questionnaire reported plans for commercialization.
The majority of the 68 projects involving GM trees had no or very weak connections to traditional breeding programmes (60 percent) or vegetative propagation programmes
(66 percent), though ongoing development of some links was noted. The remainder noted strong links to such efforts. Of the core respondents, 65 percent noted the use of genetic modification for underpinning other research goals such as testing gene function, evaluating environmental impact, QTL detection, MAS evaluation, microarray and EST mining for genes, etc.
Respondents were asked to identify whether, and from whom, regulatory oversight existed for their genetic modification projects. Core respondents representing 15 countries identified regulatory agencies (see Appendix 2.1) and categorized their involvement in (i) pre-risk assessment (13 countries), (ii) surveillance, monitoring and quarantine (12 countries), or
(iii) management of failure, redress and control (12 countries). At least nine countries regulate all three levels of involvement, and no country is totally unregulated. When asked whether the regulatory framework in their country was adequate to assess the benefits and risks related to their experiments, 15 of the core respondents (65 percent) indicated it was, five thought it inadequate, and three did not answer. Comments on how the process could be improved were many, and seemed to focus on how the process could be streamlined and made less expensive. Others suggested more risk assessment research was required. None of the core respondents indicated they were testing GM trees abroad.
Of the core respondents, eight (35 percent) indicated they owned intellectual property (IP) in relation to their production of GM trees, ten did not, and five respondents did not answer. Nine of the respondents confirmed they had IP agreements with other institutions. None of the respondents indicated they had IP arrangements with institutions in developing countries and countries in transition, though some indicated willingness to have these where appropriate and required. Eleven areas were noted as being the target for IP ownership including, but not restricted to, genes, gene expression, transgene development and testing, modulation of lignin biosynthesis, flowering control and embryogenesis.
Many of the 49 respondents contributed suggestions as to how GM trees might be used for environmental or health benefits in the future, but only a few respondents predicted further field releases, suggesting that high costs and regulatory burdens of conducting such trials are too great. Some suggestions for future plans included development of GM trees for tolerance to drought and extreme temperatures, candidate gene testing, production of secondary compounds (pharmaceuticals), site remediation, fibre quality, etc. The generally dour view of the future of genetic modification in forestry can be attributed to a number of factors including, but not restricted to: consumer rejection of or unease with GM products; public relations risks for companies engaged in research on genetic modification; unpredictable and costly government regulations, not only for conducting the research but for international trade; inadequate research support; intellectual property costs; and so forth. These factors are in contrast to the largely unregulated and less costly biotechnologies excluding genetic modification or traditional means of tree improvement.
Respondents were asked what commercial, environmental or human health benefits might be attributed to the use of genetic modification technology in forestry. Respondents often indicated more than one benefit per category. With respect to commercial benefits, increased wood production (15 responses) or improved wood quality (12) were most cited, followed by resistance to insects (nine) and disease (seven), reduced production and processing costs of wood or chips (five), and reduced chemical costs for pulping (four). Several other responses were given once each (a total of 60 possible benefits mentioned). Two notable environmental benefits were reduced pressure on natural forests (12) and reduced use of chemicals in forests and in processing (12). The use of GM trees for phytoremediation and carbon sequestration10 were also identified as environmental benefits (seven responses each), while increased productivity per managed hectare, adaptation to stresses, reduced erosion, and renewable energy largely completed the list. Human health benefits from GM forests were mentioned least often, though reduction in pollen and allergy problems was identified seven times. Reduced environmental pollution (five responses) and environmental protection and restoration (five) were also noted. Three respondents felt there were no likely health benefits. Benefits of other kinds were also noted. It was noted that genetic modification in forestry could lead to potential economic benefits for developing countries and countries in transition, and could significantly accelerate conventional breeding programmes. There were indications that genetic modification could also provide basic biological knowledge and employment, and contribute to reduce global warming.
Respondents were next asked to identify anticipated commercial, environmental or human health risks associated with the use of genetic modification in forestry. Commercial risks ranged from public resistance (nine) and large financial investment risks (five), to biological risks such as transgene instability (three), plantation failure (two), wood quality issues (two) and use of monocultures (two). Ecoterrorism and development of tolerance by insects to GM resistance were each mentioned once in a total of 31 suggested risks. Of 16 responses to a question about human health risks, 11 suggested these were negligible or non-existent. Two votes each were given for potentially new allergens or toxic metabolites being created.
Potential environmental risks noted included gene (eight) or plant (seven) escape into natural ecosystems, impact of resistance genes on non-target species (two), and clonal failure in plantations (two). Four other responses were given once each and eight respondents indicated there were no or limited environmental risks. A broad array of general responses was given about other potential risks, for example: (i) GM forests may render natural forests valueless; (ii) means exist to ameliorate environmental risks such as gene escape; (iii) regulatory/legal/commercial constraints will always slow commercialization of best technologies; and (iv) genetic modification will encourage clonal plantation forestry, which is undesirable.
When asked how risks involved with GM trees can be best addressed respondents noted most frequently the need for increased risk assessments in field and laboratory studies (13 total responses), introduction of sterility genes in GM crops (five), and research and education (two). Several other comments were given once each.
Respondents were specifically asked how their own R&D programmes addressed risk. Responses were many, varied and constructive. They included anticipated use of flowering control in GM products, inclusion and promotion of biodiversity conservation and ecological restoration of native/natural forests in association with GM plantations, studies of pollen dispersal and ecological interactions between GM trees and herbivores, and use of harmless selectable markers (to detect the presence of transgenic cells) rather than antibiotics.
Finally, respondents were asked about the risks of not using genetic modification technology. Seven respondents felt that this would compromise the ability to reduce pressure on the world’s natural forests, while four felt there was little or no risk at all. Several other answers were recorded, including (i) non-users could be at a competitive disadvantage in intensive plantation production; (ii) loss of one approach for saving endangered species (e.g. the American chestnut, Castanea dentata); (iii) loss of the best, long-term opportunity for increasing forest productivity; and (iv) disproportionate loss of the technology could lead to a decrease of the timber industry in some countries relative to others.
The questions of what obstacles to the application of genetic modification technologies in trees exist, and what means exist to address those obstacles, were asked. The overwhelming response to the first question was that public perception, regulatory issues and adversity toward science constituted the major set of obstacles (22 of 49 responses). The remaining nine obstacles, which drew 2–4 responses each, included such items as technical hurdles, lack of interest and support from the timber industry and growers, lack of funding, environmental risks, lack of integration with traditional breeding programmes and scientists, forest certification schemes and IP issues. Sixty-five percent of respondents (13 of 20) listed better education of and communication with the public as ways to overcome obstacles. Several other excellent suggestions dealt with both technical and strategic use issues, including:
(i) target traits with obvious benefit to consumers; (ii) provide technology to ensure gene flow does not occur between GM and natural trees; (iii) seek means to certify products from GM plantations; (iv) increase R&D funds, particularly from public sources; and (v) eliminate the need for GUS (β-glucuronidase) and antibiotic resistant genes in the genetic modification process. There was also support for increased field testing to validate benefits and risks, possibly even to the exclusion of seeking commercial release until data are solidly supportive.
The final series of questions in the questionnaire addressed issues of public perception of specific programmes and how organizations using genetic modification communicated their findings to the public. Twenty-five respondents (22 core, three additional) contributed an array of answers summarized by Walter and Killerby (see Appendix 2.2). These covered the gamut from very positive to very negative (vandalism of GM trials), and virtually all recognized the importance of public perception to their ultimate success. Some respondents perceived resistance from local governments while sensing support at the national level, while in some countries resistance was experienced from all aspects of society and government. Of core respondents, 61 percent reported they have some form of direct public communication, either as part of a corporate strategy or as individual communication strategies. Three respondents had no form of communication and two relied on collaboration with outside agencies for their communication about GM trees.
Over 70 field trials of GM fruit and ornamental trees, conducted in 13 countries, were noted by respondents and from database searches. A significant majority (43 of 73) of the trials are located in the United States. Taxa most studied include Carica (papaya, 18), Malus (33), and Prunus (13). These trials target fundamentally different suites of genes from the forestry sector, concentrating on viral (15), fungal (17) and insect (11) resistance, rooting (five) and fruit ripening (11).
While the information provided by respondents to the questionnaire was immensely useful in gauging the status of and trends in genetic modification in forestry, it was clear from surveys of public databases and regulatory sites that all activities were not accounted for in the formal survey. This was particularly true for field trials of GM trees in the United States11.
In general, getting an accurate estimate of the total number of field trials with GM trees is difficult. Best estimates place the number at greater than 225 trials, with as many as 150 or more in the United States. While the general world trend appears to be a gradual decline in the number of field trials being established, the current trend in the United States appears to be that field trial establishment is stable or increasing. This is largely a function of one or more private organizations focusing on the delivery of GM products to the forestry community.
Biotechnology in forestry, as in agriculture in general, encompasses a wide range of research tools used to understand and manipulate the genetic make-up of select organisms. Forest biotechnology is much broader than genetic modification, and includes micropropagation, molecular marker applications, and the rapidly expanding area of genomics. Indeed, a significant majority (81 percent) of all the biotechnology activities in forestry compiled from the last 10 years was not related to genetic modification.
Though most biotechnology tools were originally developed for use with human beings and agricultural crops, the forestry research community has been fast to adapt them.
In many respects, forest trees, traditionally considered difficult organisms to study genetically, may constitute model organisms for technologies seeking to understand and use natural sources of variation. This is due in part to the unparalleled genetic and phenotypic diversity common to forest trees, relative to most crop species.
The global forestry biotechnology community, though small in comparison to agriculture, is fairly robust and widely divergent in research, development and application targets. Cited activities in forest biotechnology have increased nearly three-fold in the most recent 5-year period, compared to the previous 5-year period. Thus, there is an upward trend in the development of biotechnologies focused on characterization and utilization of naturally occurring genetic variation within species, while publicly funded and reported programmes on genetic modification technologies in trees, particularly those targeting field testing and deployment, seem flat or diminishing. Private sector genetic modification activities, though narrowly based, appear dynamic and the world’s first commercial GM forest plantations have been established on a small scale in China.
Overall understanding of tree molecular biology has advanced dramatically in recent years. The completion of genome sequences for model plants such as Arabidopsis and especially Populus, along with development of high throughput genomics tools for gene discovery and functional assignment have set the stage for continued rapid progress. This includes our understanding of the genetic characterization of complex traits such as growth, cold and drought tolerance, insect and disease resistance, and so forth. Knowledge of, and access to, genes, will, in the author’s view, cease to be a significant hindrance to progress in genetic modification or other biotechnology applications, over the next 10 years. To capture fully the value of this expanding resource, significant investment in bioinformatics infrastructure is needed. Finally, though small steps are appearing in some countries, globally the coordination and complementarity of biotechnology tools and conventional applied tree breeding programmes seem poorly developed.
Perceived trends in the development and application of biotechnology tools are highly varied.
• Molecular markers: The creation of new types of markers has recently declined (in the last 5 years), but the development and optimization of specific markers such as microsatellites and SNPs for tree species is currently very active. Markers will probably continue to enjoy increased application in forest genetic studies (diversity and conservation, phytogeography, mating systems) and tree improvement (fingerprinting, paternity analysis, breeding and testing, QTL mapping, MAS, association genetics), though most effort is likely to be concentrated on a few highly valued species. These applications are becoming increasingly commercial in scope.
• Genomics/proteomics: This is the most rapidly expanding area of publicly funded research in forest biotechnology, and centres on gene discovery and function elucidation. Whole genome or expressed gene sequences are expected to be largely known for at least one conifer (Pinus) and possibly two angiosperms (Populus and Eucalyptus) within
5 years. Thereafter, functional genomics tools (microarrays, model species comparisons, genetic modification) will probably receive most support. Within 10–15 years, virtually complete physical and genetic maps are likely to identify and locate most of the genes in model conifer and angiosperm tree species and large unigene sets may be available for expression studies. Comparative genomics studies suggest this information will be applicable across species, genera, and even family boundaries. Applications for this information will gradually increase, including technologies complementing traditional tree improvement such as association genetics (identification of superior alleles at known genes) and identification of genes for studies on genetic modification.
• Micropropagation: Vegetative propagation will probably expand concomitantly with intensively managed, clonal plantations throughout the world. The development of more intensive plantations seems a driving force in micropropagation biotechnology and genetic modification. Application will certainly increase if and when genetic modification finds acceptance due to the desire to deliver gain through clones in most cases. Research in rooted cutting technology is mature and declining, but use of rooted cuttings will continue to grow. Somatic embryogenesis (SE) research still enjoys significant support, particularly for application in conifers, but some technical hurdles remain to commercializing the technology on a large scale. The use of manufactured seed to deliver embryos may make SE more widely affordable and open the door to more extensive clonal forests for select species, markets and applications. Advances in automation will also be necessary in all steps of the SE process.
• Genetic modification: Globally, publicly funded research in genetic modification targeting forest deployment seems to be flat or diminishing, and some scientists have expressed mixed feelings about future applications of the technology. A notable exception is China, which appears to sponsor a robust R&D programme in several institutes and has sanctioned small commercial releases of GM poplar. A relatively few but apparently quite active privately funded organizations in the United States, New Zealand, and possibly elsewhere, are pursuing R&D with commercial deployment of GM trees in their business plans before 2010. Many private forestry products companies are reluctant to engage in research on genetic modification, probably as a function of several factors: consumer unease, public relations risks for companies engaged in the research, unpredictable and costly government regulations, government bans against genetic modification, limitations due to fragmented patent estates and stringent intellectual property rights, etc. Despite these uncertainties, some authors anticipate an increasing use of GM forest trees in the near future (10 years). This will probably occur largely in short-rotation clonal forests, where investment can be rapidly recouped, for specialty applications such as saving species endangered by pests or, more commonly, as a genomics tool for studying gene function. Many of the challenges to securing the benefits of GM crops for the poor, noted for agricultural crops (FAO 2004), will be equally or more difficult to overcome in forestry.
With the exception of some micropropagation tools, research and application of forest biotechnology tools will continue to be primarily sponsored by and used in developed countries. With the shifting of industrial, high-yield forestry to semi-tropical climates, application of biotechnologies in developing countries is likely to follow relatively quickly. The biggest challenge to the forest biotechnology community may be to find ways to enhance growth and yield of non-industrial forests for use in developing countries and countries with economies in transition.
FAO. 2004. The state of food and agriculture (SOFA) 2003–2004. Agricultural biotechnology: meeting the needs of the poor? FAO Agriculture Series No. 35. Rome.
Lambeth, C., Lee, B.-C., O’Malley, D.M. & Wheeler, N.C. 2001. Polymix breeding combined with paternal analysis (PMX/WPA) of progeny: An alternative to full-sib breeding and testing systems. Theor. Appl. Genet., 103: 930–943.
4 See: www.biodiv.org/.
5 Figures should be taken with caution: for example, a significant amount of literature in Chinese is not covered by international databases.
6 See: http://dendrome.ucdavis.edu/.
7 Source: Weyerhaeuser Company, Federal Way, WA, USA.
8 Other questionnaires were sent out by FAO or posted on mailing lists and biotechnology discussion fora.
9 Several respondents commented on genetic modification activities in ornamental and fruit trees. Data on these will be noted where appropriate.
10 Questionnaires were filled in before the ninth session of the Conference of the Parties to the UN Framework Convention on Climate Change in December 2003.
11 See: www.isb.vt.edu/cfdocs/fieldtests1.cfm or www.aphis.usda.gov/.