Info Service on Biodiversity and Traditional Knowledge (Oct17/06)
Session of the Governing Body
Information: A Pressing Concern for the Seed Treaty
Sequence information is a topic that governments cannot afford to ignore at the 7th meeting of the Governing Body (GB) of the International Treaty on Plant Genetic Resources for Food and Agriculture (ITPGRFA) in November 2017 in Kigali, Rwanda. New applications of sequence information are transforming how plant genetic resources are used, and have major long term implications for the Treaty.
The increasing availability of “free” sequence information of diverse varieties of crops is reducing the need for physical access to germplasm in commercial and other breeding programs. As the drivers of this phenomenon, including cheap sequencing, gene editing and other biotechnological approaches, continue to develop, the trend will accelerate.
These rapidly advancing technologies are upending traditional approaches to access and benefit sharing, not only for the Treaty’s Multi-lateral System (MLS) and plant genetic resources for food and agriculture (PGRFA) more generally, but for all biodiversity. A situation is quickly arising in which biopiracy is taking place because legal frameworks have not caught up to technical realities.
Sharing of sequence information is now central to many aspects of crop research, but so long as that information is generated and shared without applicant of benefit sharing obligations, the developing country governments, farmers, and indigenous peoples that created and nurtured that diversity, including the seeds held by national and international agricultural research center collections, will lose. Farmers’ varieties in-trust and in the field are and will continue to be privately “mined” for profitable sequences with little or no recompense.
This impending change – sometimes called “dematerialization” – promises to be so stark that even if the Treaty is able to successfully address the failures of its Benefit Sharing Fund (principally a lack of payments from the private sector), any agreement to fix the MLS that does not fully apply benefit sharing obligations to sequences will be a dead letter, undermined by “free” sequence information before a revised MLS ever even enters force.
This includes present concepts of a “subscription system”.
Governments should recognize that it is unrealistic to expect results from handing over the task of sequence information to the Treaty’s MLS Working Group, whose éminences grises are stuck in a decades-old quagmire over the terms of access to the seeds themselves. This state of affairs - having the Working Group stuck debating how to manage last century’s technology - is, as far as the seed industry is concerned, not a bad situation. Because it helps permit newer practices to become established in the absence of the application of benefit sharing rules.
That is, the MLS Working Group is busy unsuccessfully fixing the perpetually broken front door of the barn (seeds), while all the animals (sequences) are being led out the back.
For governments to stop the unfolding free-for-all on MLS and PGRFA sequence information, they must make the Treaty current by finding a way to apply benefit sharing rules to access and use of sequences.
Otherwise, as groups such as the International Rice Research Institute (IRRI) mine the in-trust collections, publishing sequences of Farmers’ Varieties without regard for benefit sharing, and even patenting farmers’ sequences that interest them, governments will find that the foundations on which the Treaty was built have been washed away by a sea of big data.
Rapid Advances in Sequencing and Synthesis
The easiest place to see the transformation currently underway is with the smallest organisms. In the health sector, cheap, deep and fast gene sequencing means that the full sequence of influenza viruses can be determined within hours of their isolation. And if that sequence is uploaded onto an Internet database or sent attached to an e-mail, gene synthesis technologies make it possible to recreate a living virus in under three days, at an appropriately equipped laboratory anywhere in the world.
Thus, in some cases, entire small organisms can now be moved across the planet more quickly when transmitted on the internet as sequence information than when physical samples are carried by a courier such as DHL or Federal Express.
In addition to speed increases, the complexity of gene constructs (measured by the size of their genome) that can be synthesized from a sequence is increasing. Poliovirus, the first virus to be wholly synthesized in the lab, in 2002, is about 7,500 nucleotides long. In November 2016, an American scientific team announced whole synthesis of adenovirus, with a genome of 34,000 nucleotides, four and a half times that of poliovirus.
The Synthetic Yeast Genome Project, an international collaboration of several labs, plans to have synthesized the 16 chromosomes of Saccharomyces cerevisiae by the end of 2017 – a total of 12 million base pairs of DNA comprising an entire eukaryotic genome.
Of course, it is presently not possible to synthesize from scratch more complex organisms such as crop seeds, but that is not necessary in order for sequence information to transform breeding and ABS. By combining sequence data with gene-editing technologies, such as CRISPR, genetic diversity from one place can be introduced in organisms in another without physical access taking place, and without a material transfer agreement – either a Standard MTA under the MLS or a different MTA for non-Annex 1 crops.
For example, corporate crop breeders interested in making tomatoes more tolerant of dry conditions might turn their attention to the gene sequences of tomato plants from the Ecuador, Peru, and Chile. With enough sequence and characterization data, they might identify mutations that make tomatoes more drought tolerant. Gene-editing techniques can then be used to introduce those mutations into commercial cultivars for sale in North America or Europe.
Such tomatoes are merely a semi-hypothetical example; the number of other traits in other crops that might similarly be accessed through data, and not physical transfer of materials, is practically infinite. Centres of diversity of crops and other species thus are unwittingly allowing access to their genetic resources when sequence information of their biodiversity is placed online without adequate controls.
And this of course includes when ex-situ collections, such as CGIAR Centres, share sequences outside of benefit sharing obligatons.
In Parallel: Consideration by the CBD and WHO
As is apparent, managing sequence information is an issue of relevance beyond ITPGRFA. It reaches across biology to include medicinal plants, industrial microbes, pathogens, and other species under the Convention on Biological Diversity (CBD) and, in some cases, the World Health Organization (WHO) and FAO Commission on Genetic Resources for Food and Agriculture (CGRFA).
While the implications of sequence information have been anticipated for several years, including by ITPGRFA, their consideration by the CBD is of particular import because of the Convention’s broad scope and membership, and the importance of the Treaty’s approach to sequence information maintaining consistency with that of the Convention.
The origin of the present CBD discussion is found in the 2015 report of the Ad Hoc Technical Expert Group (AHTEG) on Synthetic Biology,which identified an important potential impact of sequence information on the Convention as “access without benefit-sharing”. Parties to the CBD Conference of the Parties in Mexico in December 2016 then adopted a decision on sequence information that sets in motion a plan intended to lead to an important decision at their next meeting in Egypt in 2018.
Many developing countries at the Mexico meeting took the position that the CBD should adopt a decision clarifying that sequence information should be treated equivalently to physical samples for the purposes of benefit-sharing. In the Treaty context, this would be similar to Governing Body stating in a decision the Treaty’s benefit sharing obligations are applicable to access and use of sequences generated from MLS germplasm.
The World Health Organization, which is considering sequence information in the context of disease research (access to and use of pathogen sequences, particularly influenza virus), has engaged in some practical – but as yet unfinished – consideration of methods to manage access and use of sequences inside a multilateral access and benefit sharing arrangement, specifically, the WHO Pandemic Influenza Preparedness Framework (PIP Framework).
WHO has consulted with databases that host influenza sequence data and is considering “front end” approaches involving database user agreements, data hosting approaches including sequence tagging, and more difficult and complex “back end” monitoring of the appearance of multilateral system (for influenza) sequences in intellectual property claims.
Terms related to sequence information
The several international processes that are considering sequence information are using a variety of overlapping terms. Recognizing a need for consistency, the CBD has decided to review terminology before its next meeting, and it may be hoped that the terms it identifies for use will also be employed elsewhere to aid understanding and consistency. For now, various terms are used:
“Dematerialization” has been used for several years by ITPGRFA. It refers to genetic resources and to the fact that sequence information can, for some Treaty purposes, particularly access, supplant the need for physical specimens. Thus you have genetic resources being moved and used without movement of the origin material (germplasm).
“Genetic sequence data”, or GSD, is a term used by the WHO PIP Framework. Discussion of GSD under the Framework includes the RNA sequence of the influenza virus and that sequence in alternative forms, mainly modifications pertinent to research, diagnosis, and therapeutics, such as complementary DNA (cDNAs) and insertions / deletions with effects on virulence, growth in cell culture, etc.
“Digital Sequence Information”, or DSI, is a term originating in the CBD AHTEG, whose scope has not been defined but which many consider to include DNA, RNA, and amino acid / protein sequences in their various forms. It has been pointed out that the word “digital” may be eliminated from the term since it could be limiting for reasons including that future information (computer) systems may not be “digital”, and that sequence information that is not presently stored “digitally” should also be part of the discussion. The term is now also in use by FAO, although use by both the CBD and FAO may change based on the outcome of the ruminations of CBD experts tasked to review terms.
Key concepts to enforce in rules and policies for agricultural sequence information
Sequence information must be considered the equivalent of physical samples (e.g. seeds).
Accessing sequence information increasingly satisfies many of the same purposes previously served by accessing material, including use in the creation of new commercial products (varieties) which may be placed under patent and plant breeder’s rights. Since sequences are used this way, and will increasingly be used this way henceforth, the access and benefit sharing rules that apply to seeds should apply to sequences.
ABS agreements, including the MLS SMTA, must be updated to cover sequence data
Most ABS agreements and laws are grounded in materials physically changing hands and may not be applicable to sequence information as presently drafted. ABS agreements may also permit physical access to materials without addressing benefit sharing obligations of the recipient if she generates sequence information from them. If sequences are not treated as the equivalent of material in ABS agreements, biopiracy will be facilitated, and ways by which users can escape from benefit sharing obligations will continue proliferating.
Hosts of sequence information must require users to agree to benefit sharing
Hosts of sequence information of MLS seeds, such as CGIAR Centers and databases like Genbank and the European Nucleotide Archive (ENA) must be required to have their users to agree to ITPGRFA benefit sharing as a condition of access to MLS sequence information. The Treaty should develop rules for such user agreements (e.g. “click-wrap” terms and conditions), and the hosts should be required to implement them.
Sequence information includes DNA, RNA, and amino acid sequences
Genomic nucleotides are only part of the relevant sequences. The hereditary material of organisms is not just DNA but in some cases is RNA. And because of the complementarity between the molecules, and their important functions, the sequences of both must be covered. The sequences of amino acids that nucleotides encode are similarly valuable and can be used to replicate and modify natural compounds and in design of biological systems.
ITPGRFA’s action on sequences cannot be wholly tied to the MLS Working Group
While any revised MLS SMTA needs to incorporate provisions on sequence information, given the great uncertainty and delays in the current Working Group’s outcome, ITPGRFA must also act on sequences separately, as soon as possible, and independently of the MLS Working Group’s uncertain outcome.
Such a course of action would have similarities to what unfolded at the CBD COP in Cancun, where consideration of sequence information was removed from consideration as as item within the CBD’s synthetic biology agenda item, in order to take a COP Decision on sequences and to further consider sequences, as a COP agenda item at its next meeting.
Large scale agricultural sequencing projects already underway or planned to start soon will generate hundreds of thousands of genomes of crop plants, many of them Farmers’ Varieties. If these are distributed online without users being required to observe benefit sharing obligations, benefit sharing from use of this diversity may be lost, and more patents on farmers’ genes will result.
It is, for most Treaty intents and purposes, effectively impossible to recall sequence information once it has been released into the electronic realm without benefit sharing strings attached. But stopping sequence information publication per se, of course, should not be the objective of Treaty governments. Rather, the Governing Body needs to quickly move to clarify that MLS sequences come with the same benefit sharing obligations as apply to seeds, and that position of the Treaty may then be worked into the output(s) of the MLS Working Group. In that way, the generation and use of MLS sequence data will not be impaired, rather, it will occur among actors committed to observance of the Treaty.
It being established by the Treaty that benefit sharing fully applies to MLS sequences, the stage will then be set for the Treaty’s next great sequence information challenge: how to ensure that the capabilities of big data are out to use in a way that genuinely serves the interest of small farmers and indigenous peoples and local communities, and not those companies that would merely use the technology to advance the presently inequitable situation.
IRRI Seeks Patents on Yield-Boosting Gene Taken from Indonesian Farmers’ Rice. TWN Information Service on Intellectual Property Issues, 10 March 2017. URL: https://www.twn.my/title2/intellectual_property/info.service/2017/ip170303.htm
See UNEP/CBD/SYNBIO/AHTEG/2015/1/3. URL: http://www.cbd.int/doc/meetings/synbio/synbioahteg-2015-01/official/synbioahteg-2015-01-03-en.pdf
CBD COP Decision XIII/16. Digital sequence information on genetic resources. URL: https://www.cbd.int/doc/decisions/cop-13/cop-13-dec-16-en.pdf
See the WHO PIP Framework website, URL: http://www.who.int/influenza/pip/en/