In nature, the phytopathogenic bacterium Agrobacterium tumefaciens of the family Rhizobiaceae infects susceptible plants and causes crown gall tumors. The disease results from the transfer of effector virulence (Vir) proteins and the transfer DNA (T-DNA) derived from a large bacterial tumor-inducing (Ti) plasmid. T-DNA transfer from A. tumefaciens into a plant cell requires the expression of several virulence (vir) genes that reside on the Ti plasmid [1,2,3,4]. The uncontrolled growth of crown gall tumors results from the transfer and expression of oncogenes encoded by the wild-type T-DNA, which directs overproduction of the plant growth hormones cytokinin and auxin . Another set of genes in wild-type T-DNA causes the production of bacterial nutrients, called opines, which are then utilized by A. tumefaciens as a carbon and sometimes nitrogen source.
A. tumefaciens uses a VirA/VirG two-component regulatory system to sense various environmental signals, including acidity, monosaccharides, and phenolic compounds, and induce vir gene expression [6,7]. With the help of VirD1 and VirD2 proteins, the single-stranded T-DNA is processed and then transported into plants via a type IV secretion system (T4SS). The T4SS is used by many pathogens to deliver protein and/or DNA into the cell cytosol and modulate eukaryotic cell functions [8,9,10,11]. The process involves the recognition of cognate substrates and delivery of the substrates across membrane barrier(s).
The T4SS consists of two functional components, a transmembrane transporter comprising VirD4 and VirB1-11 proteins, and a filamentous pilus (T-pilus) [12,13,14]. The T-pilus is a long, semi-rigid, flexuous filament 10 nm in diameter that may play an important role in virulence. The T-pilus contains at least two VirB proteins. The major component, VirB2, is translated as a 12.3-kD pro-pilin protein but is processed to a 7.2-kD pilin protein by removal of a N-terminal signal peptide (1–47 amino acid residues) [15,16,17]. T-pilin, 74 amino acid residues long, is coupled between the amino terminal residue Gln-48 to Gly-121 at the carboxy terminus in a head-to-tail peptide bond, forming an unusual cyclic peptide . VirB5 co-fractionates as a minor component in T-pilus preparations and contributes to T-pilus assembly . VirB5 is localized at the tips of the cell-bound T-pili and might mediate host cells and bacteria contact via interactions with the host protein during A. tumefaciens infection . After T-DNA enters plant cells, T-DNA, along with the attached VirD2 protein, will be transported into the plant nucleus and integrated into the plant chromosome with the assistance of VirE2, VirF, other Vir proteins, and plant proteins. During T-DNA nuclear import, VirE2 may interact with the plant VirE2-interacing plant protein (VIP1) in the cytoplasm to assist in nuclear targeting of T-DNA and to block endogenous VIP1 from activating plant defense responses [4,21,22]. Successful A. tumefaciens-mediated plant transformation involves a continuous battle of plant cells activating a defense response to repel bacterial infection and bacteria using Vir proteins and manipulating plant proteins to elude the plant’s immunity systems.
A previous study  identified plant-encoded proteins that may mediate the initial contact of A. tumefaciens T-pilus with the host cell. Yeast two-hybrid and in vitro assays revealed two classes of Arabidopsis proteins that interact with VirB2. The first class consists of three related proteins: reticulon-like protein B1 (RTNLB1), 2, and 4. The second class is a RAB8B GTPase. Yeast two-hybrid assay and in vitro interaction studies demonstrated that the three RTNLB proteins interact with themselves, each other, and RAB8B, so these proteins may form a multimeric complex . Pre-incubation of induced A. tumefaciens with GST-RTNLB1 protein reduced the A. tumefaciens transformation efficiency of Arabidopsis suspension cells. The level of RTNLB1 protein transiently increased immediately after A. tumefaciens infection. Arabidopsis rtnlb1 mutant plants were recalcitrant to Agrobacterium-mediated transformation, whereas Arabidopsis RTNLB1-overexpressing transgenic plants were hypersusceptible to A. tumefaciens infection . The three RTNLB proteins all have a carboxyl-terminal 150–201 amino acid reticulon (RTN) homology domain composed of two large hydrophobic regions and a ~66 amino acid loop in between. The RTN1 protein, a membrane-anchored component of the endoplasmic reticulum (ER), is the first identified member of this family and is expressed in the central nervous system and in neuroendocrine cells [24,25,26]. RTN proteins may interact with themselves or recruit other proteins to form a complex and perform specific functions. In mammalian, yeast, and plant cells, RTN proteins are involved in various endomembrane-related processes, which includes intracellular transport, vesicle formation, and membrane curvature [27,28,29,30,31,32,33].
More than 250 reticulon-like (RTNL) genes have been identified in divergent eukaryotes, fungi, plants, and animals. RTNL genes appear to have evolved from an intron-rich ancestor [27,34]. There are 21 RTNLB proteins in Arabidopsis thaliana sharing amino acid sequence similarity to the reticulon domain at the C terminus [30,35]. Consistent with the peripheral location of RTNLB1-GFP , RTNLB1 and RTNLB6 were found by proteomic analyses of plasma membrane-enriched preparations . Fluorescent-labeled RTNLB2, 4,  and RTNLB13  are localized in ER tubules. RTNLB1-4 and 13 can co-localize and constrict tubular ER membranes, so RTNLB proteins may bend the membrane and form multimeric, arc-like structures to shape the ER tubules . In addition, the C-terminal RHD domain is required for RTNLB1-4 to reside in ER membranes and efficiently constrict ER tubules but is not necessary for their homo- and heterotypic interactions .
The RTNLB3 and 6 proteins may participate the formation of the desmotubule, membrane structures derived from the cortical ER that transverse through plasmodesmata (PD) . Many viral movement proteins can help viruses spread via interactions with the PD [37,38,39]. RTNLB3 and 6 co-localize with the viral movement protein of Tobacco mosaic virus at the primary PD . Potato virus X movement protein is also detected in the desmotubules of Nicotiana benthamiana PD . A protein microarray screen identified RTNLB1 and 2 proteins that interact with the Arabidopsis FLAGELIN-SENSITIVE2 (FLS2) protein, one of the pattern recognition receptors (PRRs) for the bacterial flagellin . The rtnlb1,2 double mutant and RTNLB1 overexpression plants show increased susceptibility to Pseudomonas syringae pv. tomato DC3000 (Pst) infection and decreased FLS2-mediated immunity responses . FLS2 levels at the plasma membrane are lower in the rtnlb1,2 double mutant and RTNLB1 overexpression plants, so RTNLB1 and 2 may control the trafficking of the FLS2 protein to the plasma membrane . However, relatively little is known about the function of RTNLB proteins in plant–microbe interactions.
In this study, we further identified two additional RTNLB proteins, RTNLB3 and 8, that interact with the A. tumefaciens VirB2 protein. A. tumefaciens-mediated transient transformation efficiency was lower in rtnlb3 and rtnlb8 mutant than wild-type plants. Furthermore, overexpression of RTNLB3 or 8 in transgenic Arabidopsis plants enhanced both stable and transient A. tumefaciens transformation efficiency. Also, RTNLB3 or 8 overexpression plants were hypersusceptible to Pst DC3000 infection. This study further reveals the involvement of RTNLB3 and 8 in plant–microbe interactions.
2.1. Interactions Among RTNLB3 and 8 and Vir Proteins in Yeast and In Vitro
A previous study demonstrated that RTNLB1, 2, and 4 interacted with the C-terminal-processed portion of VirB2 protein in yeast two-hybrid and in vitro assays . From the phylogenetic tree results of the Arabidopsis RTNLB family, RTNLB1-8 proteins belong to the Group I proteins containing an N-terminal domain with 43–93 amino acid residues and a short C-terminal domain [27,30]. Therefore, we cloned RTNLB3 and RTNLB5-8 from Arabidopsis cDNA and examined whether RTNLB3 and RTNLB5-8 could interact with A. tumefaciens VirB2 bait protein in yeast two-hybrid assays. The RTNLB8 prey protein but not the RTNLB3 and RTNLB5-7 proteins interacted with the VirB2 bait protein in yeast (Figure 1). RTNLB1, 2, and 4 proteins interacted with the VirB2 protein as well, which was consistent with previous results , and were used as positive controls in the yeast two-hybrid assays. As expected, the RTNLB1-8 prey proteins did not interact with the unrelated Lamin C bait protein in yeast and was used as the negative control (Figure 1). We also examined whether RTNLB3 and RTNLB5-8 proteins could interact with other Vir proteins, including VirB5 (the minor component of T-pili), VirB1, ViB1*, VirD2, VirE1, VirE2, and VirF. RTNLB3 and RTNLB5-8 did not interact with other tested Vir proteins, which was similar to the results for RTNLB1, 2, and 4 proteins .
Previous studies have demonstrated that RTNLB1, 2, and 4 can interact with each other and with themselves [23,33]. We next tested whether the RTNLB3 and RTNLB5–8 proteins interacted with themselves and/or other RTNLB1-8 proteins in yeast two-hybrid assays. RTNLB2 used as a bait fusion protein interacted with the RTNLB3 or 8 but not RTNLB5, 6, or 7 (Figure 1 and Table S1). RTNLB3 used as the bait fusion protein interacted with RTNLB2, 4 or 8 but not RTNLB3 or RTNLB5-7. Figure 1 results demonstrated that the RTNLB4 protein interacted with RTNLB8 but not RTNLB3 or RTNLB5-7. As well, RTNLB5, 6, 7, or 8 used as bait fusion proteins did not interact with RTNLB1-8 proteins in yeast (Figure 1). Similarly, RTNLB1 did not interact with the RTNLB3 or RTNLB5-8 proteins in yeast two-hybrid assays (Figure 1). Some of the positive yeast two-hybrid interactions were not observed when the tested bait protein was swapped with the prey proteins (Figure 1 and Table S1). For example, the RTNLB2 bait protein interacted with the RTNLB8 prey protein in yeast; whereas the RTNLB8 bait protein did not interact with the RTNLB2 prey protein. These inconsistent findings may result from different conformations of the bait and prey fusion proteins in yeast, as previously reported for interactions between RTNLB and RAB8 .
We next performed β-galactosidase activity assays to quantify the interaction strengths in these yeast strains. The white colony yeast strains on the SD media with X-gal substrates showed zero β-galactosidase activity, so the liquid-based β-galactosidase activity assays showed similar results as the plate-based yeast two-hybrid assays (Table S1). Yeast strains expressing the RTNLB3 bait with RTNLB2, 4, or 8 prey proteins showed relatively lower β-galactosidase activities as compared with yeast strains expressing the RTNLB2 or 4 bait proteins with the same tested prey proteins, which suggests that the interaction strengths might be lower among RTNLB3 interacting with other RTNLB proteins (Table S1).
In vitro glutathione-S-transferase (GST) pull-down assays were used to determine direct protein–protein interactions of RTNLB3 and 8 with VirB2 and other RTNLB proteins. T7-tagged-RTNLB1, 2, 3, 4, and 8 proteins interacted with the GST-VirB2 fusion protein but not the GST protein in vitro (Figure 2A,B). In addition, GST-RTNLB1, 2, 3, and 4 fusion proteins but not the GST-RTNLB8 fusion protein interacted with the T7-tagged-VirB2 protein. The in vitro protein interactions between RTNLB1-4 and RTNLB8 were examined with GST pull-down assays by using the GST fusions and T7-tagged versions of RTNLB1-4 and 8. GST-RTNLB1 fusion proteins interacted with T7-tagged-RTNLB1, 2, 3, 4, but not T7-tagged-RTNLB8 proteins in vitro, whereas GST-RTNLB2 fusion protein interacted with T7-tagged-RTNLB1, 2, 4, 8, but not T7-tagged-RTNLB3 protein in vitro (Figure 2C,D). Furthermore, GST-RTNLB3 and GST-RTNLB4 fusion proteins interacted with the five tested RTNLB proteins (Figure 2E,F). However, only T7-tagged-RTNLB2 and 3 directly interacted with the GST-RTNLB8 fusion protein (Figure 2G). These interaction results were consistent with previous observations showing that RTNLB1-4 proteins may have homo- and heterotypic interactions . The yeast two-hybrid assay and GST pull-down assay results summarized in Table S1
Review: intellectual property aspects of plant transformation
Jim M. DunwellCorresponding author
- School of Plant Sciences, The University of Reading, Reading RG6 6AS, UK
Correspondence (fax 0118 378 8160; e-mail firstname.lastname@example.org)
One of the recurring themes of the debates concerning the application of genetic transformation technology has been the role of Intellectual Property Rights (IPR). This term covers both the content of patents and the confidential expertise usually related to methodology and referred to as ‘Trade Secrets’. This review explains the concepts behind patent protection, and discusses the wide-ranging scope of existing patents that cover all aspects of transgenic technology, from selectable markers and novel promoters to methods of gene introduction. Although few of the patents in this area have any real commercial value, there are a small number of key patents that restrict the ‘freedom to operate’ of new companies seeking to exploit the methods. Over the last 20 years, these restrictions have forced extensive cross-licensing between ag-biotech companies and have been one of the driving forces behind the consolidation of these companies. Although such issues are often considered of little interest to the academic scientist working in the public sector, they are of great importance in any discussion of the role of ‘public-good breeding’ and of the relationship between the public and private sectors.
The present and future status of genetically modified (GM) (transgenic) crops has been the subject of several recent reviews (Dunwell, 2000, 2002, 2004). Although these reviews have included some information extracted from patent databases, this analysis has been necessarily limited in scope. The present review expands on the information published previously (Sechley and Schroeder, 2002) and extends to a discussion of intellectual property from the perspective of the research scientist (Shear and Kelley, 2003) and of those interested in international developments (Blakeney, 2000; Binenbaum et al., 2003), globalization (DaSilva, 1998; Parayil, 2003) and the more general ethical aspects of the public- and private-sector relationships (Korn and Heinig, 2003; Hails, 2004).
What are patents?
Discussions concerning the merits, or otherwise, of patenting plants are not a recent occurrence. For example, during the Third International Conference on Genetics, organized by the Royal Horticultural Society, held in London in 1906, and most famous for the coining of the term ‘genetics’ by William Bateson, there was a session entitled: ‘“Copyright” for Raisers of Novelties’ (Anonymous, 1907). It is reported that Mr George Paul, whilst remarking on the absence of several well-known plant breeders, stated: ‘The fact is, these gentlemen do not like to tell us, or to show, what they have done in their experiments, because once their knowledge becomes public, they have not the slightest chance of receiving any pecuniary reward for their labours. If they were properly protected from being deprived of the due reward of their labours, they would no doubt be much more willing to come forward and help us and place their experience at our disposal’. During discussion, Professor Hansen responded: ‘I believe, in law, a seedling is regarded as the gift of God, and it would be hard to patent that; but could we not hope to have some law fashioned that would give a bonus to the man who does such skilled and valuable work as that which has come before us over and over again during the sessions of this conference’.
The chairman of the session, whilst sympathizing with Mr Paul, concluded that it would be unwise to pass a resolution on the subject as the discussions had demonstrated: ‘What very great difficulty there would be in enforcing such a law, because we have gentlemen from all parts of the world maintaining that a thing is new, and others, equally capable, maintaining that it is old’. Much of the debate today, almost 100 years later, follows the same themes.
The history of patent law dates back several centuries but, in summary: ‘A patent gives an inventor a period of exclusive exploitation (up to 20 years in the UK) in return for a disclosure of the invention’ (Huskisson, 1996). Central to the patent system are the criteria that have to be met for the granting of a patent. These criteria are for the invention to be novel, non-obvious, industrially applicable and not in several excluded categories. Considering these requirements in more detail:
- 1novel: ‘new’ means more than ‘state of the art’, i.e. everything made available to the public prior to the ‘priority date’ of the patent;
- 2non-obvious: an inventive step, not obvious to the ‘skilled man’ (who knows everything and has no imagination, may be a team);
- 3industrially applicable: something which can be made or used in any kind of industry, including agriculture;
- 4exclusions: (i) anything immoral; (ii) plant and animal varieties per se (except in USA); (iii) essentially biological processes for the production of plants and animals; (iv) ideas, theories, computer programs, etc.
One of the most important features of this process is the need for any invention to be kept confidential and not to be disclosed prior to the filing date of the patent application. In European patent law, an invention counts as published if it forms ‘part of the state of the art’, with the ‘state of the art’ being defined as: ‘everything made available to [even one member of] the public [anywhere in the world] by means of a written or oral description, by use, or in any other way’. Examples of publication include: learned papers, journals and magazines; abstracts; theses; the Internet; poster displays; exhibitions and open days; oral and casual disclosure; and confidential disclosure to many people (http://www.btgplc.com/info/links_lit.php).
More generally, the role of the patent system is to encourage industry and innovation by rewarding invention and protecting investment in product development. The concept behind most patent legislation is that disclosure and publication allow others to test the invention and attempt improvements, whilst expressly not allowing the routine use of a patented process or processes. This disclosure takes the form of a publication from the relevant patent office. In the case of most authorities, the patent procedure involves submission (filing) of a preliminary patent application followed, after 12 months, by a final application, which is published 6 months later and is then available for inspection. Exceptionally, until 15th March 2001, the USA maintained secrecy until the time the patent was granted, a period that can range from an average of 2–3 years upwards to more than 20 years. As an example of the lengths of time sometimes required for the resolution of complicated cases, the main US case, which covered elements of Agrobacterium-based transformation of dicotyledonous crops, was only finally resolved in October 2004, some 20 years after the date of filing. It was reported subsequently (February 2005) (http://www.bayercropscience.com/bayer/cropscience/cscms.nsf/id/Patent_dispute_resolved) that, under a related agreement, Max Planck Society, Bayer CropScience, Garching Innovation and Monsanto will cross-license their respective Agrobacterium-mediated transformation technologies worldwide. Bayer CropScience, Max Planck's exclusive licensee, and Monsanto will provide each other, in selected areas of the world, non-exclusive licences related to the development, use and sale of transgenic crops. Monsanto will also provide Max Planck Society with a licence in the USA for research purposes. In a similarly delayed resolution, the US patent claiming the direct transformation of Brassica protoplasts with a plasmid was granted in the USA (6 603 065) on 5th August 2003, some 18 years after the first application on this subject filed on 3rd May 1985.
Another important difference between the US and other systems is that the 17 years’ duration of a US patent filed prior to 2001 only starts from the time at which it was granted, whereas, in Europe (and now in the USA), the 20-year period of exclusivity starts from the time of filing the application. Some of the consequences of this change are discussed in more detail below.
At a practical economic level, the submission costs of patent applications are small (> £100) initially, but may rise rapidly if professional agents are employed, if the patent coverage is geographically extensive and if the granted patent is maintained over several years. Individual patent offices should be consulted for precise costings.
Sources of patent information
During the preparation of this review, extensive use has been made of the freely available patent databases in the USA (http://www.uspto.gov/patft/index.html), Europe (http://ep.espacenet.com/), World International Patent Organization (http://pctgazette.wipo.int/) and other international sites (e.g. http://www.surfip.gov.sg/sip/site/sip_home.htm; http://www.cambiaip.org/cgi-bin/cipr/TT3/simple.cgi). A very useful site with a summary of granted US ag-biotech patents from 1976 to 2000 is provided by the Economic Research Service (ERS) of the US Department of Agriculture (USDA) (http://www.ers.usda.gov/Data/AgBiotechIP/). Some sites are dedicated to specific species. For example, details of Arabidopsis patents are available from Lehle Seeds (http://www.arabidopsis.com/home_RR03.html), and information about trees can also be obtained (Sedjo, 2004). It should be noted that the most detailed forms of patent analysis require commercial subscription from companies such as Derwent (http://www.derwent.com), MicroPatent (http://www.micropat.com) or patentmaps.com. The last two companies have advanced graphical methods of displaying the relationships between patents within given sectors of technology.
In addition, this review includes data from various sites providing information on the field tests of GM crops in the USA (http://www.nbiap.vt.edu/cfdocs/fieldtests1.cfm) and elsewhere (http://www.nbiap.vt.edu/cfdocs/globalfieldtests.cfm) (http://gmoinfo.jrc.it), as these often provide useful perspectives into future commercial products and international trends. Another source of data on the release of GM crops is available at http://www.agbios.com/dbase.php.
Patents and agricultural biotechnology
There are a range of methods that can be used to protect novel types of plants produced by one company from being exploited by commercial competitors, with these methods varying from one country to another (Bennett, 1994; Erbisch and Velazquez, 1998; Blakeney et al., 1999; Cahoon, 2000). An introduction to the various approaches, namely plant breeders rights and patents (known collectively as Intellectual Property Rights – IPR), is available from several authors (Scalize and Nugent, 1995; Mauria, 2000; Henson-Apollonio, 2002; Brown, 2003; Fagerlin, 2003), and from the Intellectual Property Division of CAMBIA (Center for the Application of Molecular Biology to International Agriculture) based in Australia (http://www.cambiaip.org/Tutorials/Plant_Protection/plant_protection.htm). Specific information relating to individual countries is available at the respective patent offices. For example, the latest note on patenting of plants in the UK, ‘Examination Guidelines for Patent Applications Relating to Biotechnological Inventions in the UK’, was published by the Patent Office in November 2003 (http://www.patent.gov.uk/patent/reference/biotechguide/plant.htm). The relevant paragraphs of this document are as follows:
Par. 65 As confirmed by the EC Directive plant and animal varieties are not patentable. Plant varieties are currently protected under the Plant Varieties Act 1997. Both the 1997 Act and a separate European Community regime (Council Regulation (EC) No. 2100/94) are based on the 1991 UPOV Convention (http://www.upov.int/en/about/upov_system.htm). In the UK the system for granting plant variety rights is administered by the Plant Variety Rights Office (PVRO) at Cambridge. This system differs substantially from the patent system and to gain protection a variety must be tested for distinctness from other varieties, uniformity and stability.
Par. 66 Plant variety rights are confined to individual varieties. Patents may claim plant genera or species but they cannot claim individual varieties.
Par. 67 In the early days of granting plant patents neither the EPO nor the UK Patent Office had a problem with granting claims to plants in general even though it could be argued that such claims could be regarded as covering, in reality, a number of plant varieties. The EPO's Enlarged Board of Appeal was eventually called on to consider this issue. The Enlarged Board found:
A claim wherein specific plant varieties are not individually claimed is not excluded from patentability under Article 53(b) EPC even though it may embrace plant varieties;
When a claim to a process for the production of a plant variety is examined, Article 64(2) EPC is not to be taken into consideration;
The exception to patentability applies to plant varieties irrespective of the way in which they were produced. Therefore, plant varieties containing genes introduced into an ancestral plant by recombinant gene technology are excluded from patentability.
Thus, claims to transgenic plants are perfectly acceptable, unless expressed in plant variety terms or the invention is confined to modifying a particular plant variety. It may be therefore that if all the examples in an application are directed towards modifying a single variety, there could be a presumption that the invention is specifically for a plant variety.
Similar information is available concerning the patentability of plants in the USA (Merrill et al., 2004), Europe (Perdue, 1999; Fleck and Baldock, 2003), New Zealand (Ministry of Economic Development, 2002) and China (http://www.cnpvp.net/old-www/rules_and_regulations.htm#REGULATONS%20OF%20 THE%20PEOPLE’%20S%20REPUBLIC%20OF%20CHINA%20ON%20THE%20PROTECTION%20OF%20%20NEW%20VARIETIES%20OF%20PLANTS). In addition, the results of a detailed survey of actual practice of patent examiners in the three key patent offices, USA, Europe and Japan, has been published recently (Howlett and Christie, 2003). As opposed to the speculation that is widespread in this area, this survey gives information about what is actually happening in relation to gene patenting and, it is suggested, therefore, that it provides a basis for informed decisions and policy development. In a complementary study restricted to the present and future position in the USA (Merrill et al., 2004), the authors conclude that the continuing high rates of innovation suggest that the patent system there is working well and does not require fundamental changes, although they note that both economic and legal changes are putting new strains on the system. Specifically, patents are being more actively sought and vigorously enforced. The consequence of this activity is that the sheer volume of applications to the US Patent and Trademark Office (USPTO) – more than 300 000 a year – threatens to overwhelm the examination staff by reducing the quality of their work and creating a huge backlog of pending technology; cases of defending against patent infringement violations in court are also rising rapidly. This report also states that, in some cases, patenting appears to have departed from its traditional role, as firms build large portfolios to gain access to others’ technologies and reduce their vulnerability to litigation (Evenson, 2000). This issue and the effect on company interaction are considered below.
There have been several extensive reviews of the consequences and implications of applying patent (and other IPR) protection to plants (Plant Intellectual Property, 2001; Farnley et al., 2004), and the reader is referred to these publications, most of which are freely available on the Web. In one of the most comprehensive of these reviews (Binenbaum et al., 2003), the important conclusion is reached that, as patenting becomes even more prevalent in biotechnology, the diversity of innovations utilized in developing modern cultivars will mean that the number of separate rights needed to produce a new innovation will proliferate. Where ownership of relevant rights is sufficiently diffuse, the multilateral bargaining problem can become very difficult, although not impossible (see section on ‘Golden rice’ below), to resolve. This problem is further compounded by uncertainty. For example, those who develop new technology by building on existing technologies often know neither the extent to which the latter has been claimed as Intellectual Property nor the strength of any claims. Both the conduct of research and development and subsequent commercialization therefore entail navigating through a potential minefield of patent applications that have been filed but remain invisible pending publication by the patent office. For example, public breeders in the USA received an unwelcome surprise when a patent issued to Monsanto for the cauliflower mosaic virus (CaMV) 35S promoter (see below) surfaced after they had used it in the breeding of crop cultivars on the brink of commercialization. Fortunately, the uncertainty arising from such ‘submarine’ patents is becoming less important as the USA has harmonized with the rest of the world, first by awarding a patent term of 20 years from the date of filing (previously 17 years from the date the patent was awarded), and secondly by publishing (from November 2000) patent applications within 18 months of filing. In addition, the existence of ‘submarine patents’ can sometimes be inferred from the publication of foreign filings (e.g. EU/World Intellectual Property Organization), although detailed claims and coverage may differ.
Despite the complexity of biotechnological IPR, it should be pointed out that similar problems exist in the electronics industry where products are assembled from numerous internationally sourced components covered by a multiplicity of patents.
Patents and plant transformation
During the period since the production of the first transgenic plants, a wide diversity of patents have been sought on all aspects of the process, ranging from the underlying tissue culture methods through to the means of introducing the heterologous DNA, and to the composition of the DNA construct so introduced (Kesan, 2000). It would be impossible to summarize all this information in the space available here; the amount of patent information available in the area of plant transformation (Tables 1 and 2) can be judged by the fact that a search of the US application database for ‘transgenic plant’ and ‘method’ returned 2160 records on 26th August 2004, with at least six relevant patent applications being submitted on the single day in question.
|Selectable marker techniques||486|
|Culture growth, cell differentiation, etc.||1632|
|February to August 2004|
|Use of auxin precursor (Matsunaga et al.)||Nippon Paper||20040163143|
|Whisker-mediated method (Petolino et al.)||Dow Agrosciences||20040128715|
|Monocot transformation (Elliot et al.)||CSIRO||20040123342|
|Transformation of Brassica (Chen et al.)||Cargill||20040045056|
|Micro-vibration and ovary injection (Liou)||Unknown||20040045048|
|Electrical shock and ovary injection (Liou)||Unknown||20040045047|
|Transformation of soybean (Khan)||Syngenta||20040034889|
|Transformation of Camelina sativa||Unknown||20040031076|
|26th August 2004|
|Enhanced tissue culture response (Lowe et al.)||Pioneer||20040168217|
|Seed targeting (Jiang and Sun)||Unknown||20040168215|
|Phloem promoter (Kwart et al.)||Max Planck||20040168214|
|Nematode resistance (Verbsky et al.)||Monsanto||20040168213|
|Antifungal enzyme (Duvick et al.)||Pioneer||20040168212|
|Altered nicotine levels (Conkling et al.)||Unknown||20040168211|
For a detailed analysis of several of the key areas under discussion, the reader is referred to the extensive summaries published elsewhere, for example in the series of comprehensive CAMBIA White Papers (Roa-Rodrigues, 2003; Roa-Rodrigues and Nottenburg, 2003a, b; Mayer et al., 2004), aspects of which are considered below.
The first point to be emphasized is that patents only operate in the countries in which they are granted, subject to the caveat that products exported to countries in which the patent operates may also infringe. Frequently, however, the main point of interest has been the coverage of the patent(s) in question. There are some well-known examples of patents with very broad coverage and this is often a topic of debate and the cause of concerted opposition. For example, European Patent 301749, granted to Agracetus (then a subsidiary of WR Grace & Co.) on 2nd March 1994, is an exceptionally broad ‘species patent’ which grants this company rights to all forms of transgenic soybean varieties and seeds – irrespective of the genes used or the transformation technique employed. Agracetus was purchased by Monsanto in April 1996, after which it withdrew its previous opposition to this patent. However, opposition continued from other companies and organizations and a hearing was finally agreed by the European Patent Office (EPO) in May 2003, at which the patent was upheld, with the exception of Claim 25 covering plants other than soybean (http://www.european-patent-office.org/news/pressrel/2003-05-06_e.htm; http://www.european-patent-office.org/news/pressrel/pdf/bginfo_soya_e.pdf). The patent is due to expire in July 2008.
There are several techniques for the introduction of recombinant vectors containing heterologous genes of interest into plant cells, and the subsequent regeneration of plants from such cells. The two main methods are the use of Agrobacterium or the direct introduction of DNA on microparticles of metal, a technique known as biolistics. Some of the patents covering these methods are summarized in Table 3, which also includes details of other, related patents on techniques such as direct uptake into protoplasts, electroporation or vortexing with needle-shaped crystals (whiskers) of silicon carbide (Dunwell, 1999). Most of these methods involve a tissue culture step (Hall et al., 1996), and many of these enabling protocols are also the subject of patent claims. Recent related applications in this area are given in Table 4.
|Particle bombardment||Cornell||US 4945050|
|Dekalb||US 5538877, 5538880|
|Agracetus||US 5015580, 5120657|
|Agrobacterium||University Toledo||US 5177010, WO 02/102979|
|Texas A & M University||US 5104310, WO 03/048369|
|Leiden University||EP 120516, 159418, 176112|
|US 5149645, 5469976, 6464763|
|US 4940838, 4693976|
|Max Planck||EP 116718, 290799, 320500|
|Japan Tobacco||US 5591616|
|EP 604662, 627752|
|Ciba-Geigy||EP 267159, 292435|
|Washington University||US 6051757|
|Calgene||US 5463174, 4762785|
|Agracetus||US 5004863, 5159135|
|Whiskers||Zeneca||US 5302523, 5464765|
|Electroporation||Boyce-Thompson Instit.||WO 87/06614|
|Dekalb||US 5472869, 5384253|
|PGS||US 5679558, 5641664|
|WO 92/09696, 93/21335|
|Phosphinothricin, Basta||Aventis/AgrEvo||EP 531716 et al. |
US 5767371 et al.
|Kanamycin||Monsanto||EP 131623 |
US 6174724 et al.
|Hygromycin||Novartis||EP 186425 et al. |
US 5668298 et al.
|Cyanamide||Syngenta/Mogen||EP 97201140 |
|Aldehyde||Calgene||EP 0800583 |
|Mannose/xylose||Novartis||US 5767378 et al.|
|2,4-Dichlorophenoxyacetic acid||Unknown||EP 0738326 |
The most extensive publication in this area is the 360-page CAMBIA White Paper (Roa-Rodrigues and Nottenburg, 2003a) on Agrobacterium-mediated transformation. This document focuses on the patents directed to the methods and materials used for transformation, mainly of plants, but also of other organisms such as fungi. It should be stressed that, although much of the early development of this technique was performed in universities, most of the patents are consolidated in the hands of a few companies. In the case of one of the most important of the US applications, that covering Agrobacterium-mediated transformation of dicot plants, there was a 12-year dispute between applicants before the case was decided in favour of Monsanto in late 2004.
One recent advance that might circumvent the IPR limitations to Agrobacterium technology is the development of gene transfer techniques using other bacteria, such as Sinorhizobium (Broothaerts et al., 2005). It is claimed that these methods may offer an ‘open source’ alternative to the established transformation technologies.
Patents and DNA sequences
Almost all the significant components of the constructs used in plant transformation have been the subject of patent coverage. These include the ‘effect gene’ as well as its associated regulatory sequences (Kay et al., 1987), the selectable or screenable marker and additional sequences that might be required for the subsequent excision of the transgene. This review does not cover details of the gene of interest (Martin, 1998) and the reader is referred to other recent reviews that include summaries of the range of present and future transgenic crops (Dunwell, 2002, 2004).
Much of the debate in this area concerns the ability to apply for patents on DNA sequences of unproven function. There have been several attempts to do so, and the decisions on such applications have not been finalized. However, the fact remains that there is much useful sequence information available in patent databases and much of it is ignored by academic research scientists. Specifically, it is estimated that some 30−40% of all DNA sequences are only available in patent databases, as there is of course no obligation for commercial (or other) applicants to submit their sequences to public databases. Possibly, the best way to access this information is via the GENESEQ system, a commercial (Derwent) service.
Selection and identification of transformants
The production of transgenic organisms, including plants, involves the delivery of a gene of interest and the use of a selectable marker that enables the selection and recovery of transformed cells. This is necessary because only a minor fraction of the treated cells become transgenic, while the majority remain untransformed. It has been estimated recently (Miki and McHugh, 2004) that approximately 50 marker genes used for transgenic and transplastomic plant research or crop development have been assessed for efficiency, biosafety, scientific applications and commercialization.
Selectable marker genes (see Table 4 for selected patents) can be divided into several categories depending on whether they confer positive or negative selection and whether selection is conditional or non-conditional on the presence of external substrates.
The most common strategy currently used for selection is negative selection, the elimination of non-transformed cells in conditions in which the transformed cells are allowed to thrive. Elimination is often effected by treatment of cells with chemicals (e.g. antibiotics or herbicides) in conjunction with a transgene that confers resistance or tolerance to the chemical through detoxification or modification of the chemical. Much of the original work was conducted using antibiotic resistance marker (ARMs) genes, which confer resistance to antibiotics such as neomycin, kanamycin and hygromycin. A summary of the most important scientific aspects of such resistance genes has been published recently, together with an analysis of selected patents that relate to the most widely used ARMs (Roa-Rodrigues and Nottenburg, 2003b). Positive selectable marker genes are defined as those, such as phosphomannose isomerase, that promote the growth of transformed tissue. Many of these marker genes are covered by patents or patent applications (Table 4), with the most thorough Intellectual Property analysis available probably being that published on antibiotic markers and Basta resistance by CAMBIA (Mayer et al., 2004). A notable example of patent coverage is that of Monsanto which holds patent rights on the use of any antibiotic resistance gene as a selectable marker for plant transformation. Importantly, these proprietary rights apply only in the USA and are covered by three granted patents (US 5034322, US 6174724 and US 6255560).
As an alternative, or addition, to the use of selectable markers, transformants are often identified through the use of reporter or visualization molecules. The term ‘reporter’ relates to genes and their products used to identify transformed cells. In agricultural biotechnology, the most common reporters are β-glucuronidase (GUS) and green fluorescent protein (GFP).
Promoters and other regulatory elements
Regulatory elements are crucial to gene expression in all organisms. The patent landscape of transcriptional regulators that are constitutively active, spatially active (e.g. tissue-specific) and temporally active (e.g. induced or active in response to a certain chemical or physical stimulus) has been well summarized recently (Roa-Rodrigues, 2003). In this review, an assessment is presented of the possibilities for and limitations on further development of regulation of gene expression. The analyses include general patent information, such as patent numbers, total number of patents on a particular promoter, applicant names, dates of filing and grant. In addition, the analysis includes claims of relevant patents, including aspects of the prosecution history of the patents where appropriate.
Although the inventions protected by individual patents cannot be exactly the same, in certain cases there are patents that, due to the breadth of their scope, may encompass other protected inventions, or there may be patents that share common features. Where this is the case, this review points out the juxtaposition of the different inventions and the possible room left to manoeuvre around the different entities in the field. It also needs to be taken into account that there are patents that, although not totally directed to promoters, may have an effect on gene expression control. This is the case for the restrictive reproductive technologies, for example those termed ‘terminator technologies’, which may have a great impact on the use and development of methods to regulate the expression of genes related to plant reproduction and seed generation.
Although many people regard promoters as being confined to the research community, media press releases and information provided by ag-biotech companies illustrate how patent rights are often used as commercial assets in this industry. Patents on plant regulatory regions play a role in the subsequent development and innovation in promoters and in areas that rely heavily on mechanisms for controlling gene expression, such as chemically switchable systems (Sweetman et al., 2002).
Novel products and the freedom to operate
One of the issues of overriding importance to all companies is whether or not they are free to commercialize any particular product (Shear, 1999; Lence et al., 2002). Such ‘freedom to operate’ is determined by the status of any IPR that might cover the product in question, and the analysis of such IPR requires continuous (and therefore expensive) surveillance (Rausser and Small, 1996).
A well-known example that can be used to demonstrate the complexity of this issue is ‘golden rice’, a transgenic line that is enhanced for β-carotene (provitamin A) (Ye et al., 2000). It provides hope for alleviating the severe vitamin A deficiency that causes blindness in half a million children every year (Brooks and Barfoot, 2003). It has been suggested that extensive patenting has hampered the delivery of this rice to those in need as some 40 organizations hold 72 patents on the technology underlying its production (Kryder et al., 2000). The range of patents covering various components of the pBIN19hpc plasmid, used in the production of this rice, are shown in Table 5. These include patents on the phytoene trait genes (Fraser et al., 1992; Misawa et al., 1993; Burkhardt et al., 1997; Schledz et al., 1996), the promoter sequences (Okita et al., 1989), the selectable marker (Waldron et al., 1985; Wünn et al., 1996) and the transit peptide (Schreier et al., 1985). Such perceived problems with access to golden rice and essential medicines have stimulated debate within the USA on the obligations of American universities to facilitate the provision of goods for the public benefit (Kowalski and Kryder, 2002), an issue also considered below in the section on ‘Public- and private-sector issues’. A recent symposium (http://www.lifesci.consortium.umn.edu/conferences/ip.php) at the University of Minnesota addressed this specific question (Phillips et al., 2004).
Patents and commercial consolidation
One classic example of the effect of patent coverage and the difficulty of acquiring freedom to operate (see above) concerns the history of the bar gene that is used both as a selectable marker and as a gene providing resistance to the herbicide Basta (Botterman and Leemans, 1989). An excellent summary of this case has been published by CAMBIA (Mayer et al., 2004).
The acquisition in 1996 of the Belgian company Plant Genetic Systems (PGS) by AgrEvo (part of the German chemical company Hoechst) was an important strategic move to gain access to a broad portfolio of traits and enabling technologies required to participate in the highly competitive market of transgenic crops. At the time, AgrEvo had fallen behind Monsanto and the ag-biotech section of Novartis (now Syngenta) in securing a competitive market position in the area of GM insect- and herbicide-resistant crops. With the acquisition of PGS, AgrEvo made a major effort to enter the US and the Canadian markets, two of the largest markets in the world. By this acquisition, AgrEvo also gained access to the PGS Intellectual Property portfolio that included such areas as gene promoters, marker genes, techniques to insert specific genes into plant cells and gene expression technology to optimize the efficacy of expression of foreign genes in plants. In addition, PGS had engaged in research and development of novel technologies, particularly in the area of functional genomics, but also in engineering disease-tolerant plants and modifying certain quality traits. PGS's products included corn, oilseed rape (canola) and selected vegetables engineered for insect protection (based on the expression of Bacillus thuringiensis toxin), herbicide tolerance and pollination control.
PGS's herbicide tolerance technology was developed in collaboration with AgrEvo, based on tolerance to AgrEvo's herbicide Liberty™ (glufosinate) by virtue of the bar gene. PGS's SeedLink™ pollination control technology is also based on tolerance to Liberty™. After the subsequent merger of AgrEvo and Rhône-Poulenc, which gave rise to Aventis, the agricultural section of this merger was called Aventis Crop Science. The Hoechst conglomerate, holder of Aventis and other companies, finally decided to shed its agrichemicals section by selling to Bayer AG, which recently gave rise to Bayer Crop Science, explaining thereby the migration and inheritance of the bar gene portfolio over time.
Similar analyses could be conducted on the other remaining large companies, namely Syngenta, Monsanto and DuPont, and their respective Intellectual Property portfolios.
Public- and private-sector issues
The most detailed review of this aspect of ag-biotech patents is probably that conducted by Graff et al. (2003), who have summarized the ownership of critical patents and compared the relative significance of the private and public sectors in each area of research relevant to the commercialization of transgenic plants. The main findings of this review, and others (Huete-Perez, 2003), are given below.
Six companies hold 75% of all agricultural patents and it has been suggested that such concentration exacerbates the challenge of delivering agricultural inventions to the most needy segments of the world's population (Solleiro, 1995). One solution could be the compulsory licensing of patented inventions that have failed to reach the most needy markets (see below for further details). An alternative would be based on the fact that, while the public sector holds less than 3% of all patents, it does have 24% of agricultural biotechnology patents, many covering genes of great potential interest (Table 6). By exploiting these resources, universities and other public organizations therefore have opportunities to deliver affordable biotechnological innovations.
|Plant enzymes||291 (92)||25 (8)|
|Bacillus thuringiensis toxin||497 (90)||57 (10)|
|Industrial enzymes||229 (89)||29 (11)|
|Metabolic pathways||198 (86)||32 (14)|
|Rice disease resistance||206 (77)||61 (23)|
|Male sterility systems||133 (75)||44 (25)|
|Viral proteins||378 (71)||153 (29)|
|Herbicide resistance||194 (69)||88 (31)|
|Product quality||291 (65)||157 (35)|
|Flowering control||48 (58)||35 (42)|
|Pathogen resistance||53 (44)||67 (56)|
Concern has also been expressed about the potential dangers (financial or otherwise) associated with the use of patented technologies by academic establishments. This is a complicated issue involving ‘experimental use exception’ (Janis, 2003; Hoffman, 2004), the concept that allows others to examine and test the patented discovery, but not to put it to routine use. In the past, industry brought few lawsuits against research universities, but the implications can be significant. One of the most notable is the 11-year battle over the widely used DNA-replicating enzyme, Thermus aquaticus DNA polymerase, known as Taq. Swiss drug giant, Roche, argued that 200 scientists who had published research using Taq obtained from Promega (Madison, WI, USA) were patent infringers. The outcome of the case is likely to determine whether researchers have to pay more or less for Taq polymerase chain-reaction products (Agres, 2004).
Patents, ethics and international development
Some people consider that the commercialization of biotechnology, especially research and development, by transnational pharmaceutical and ag-biotech companies is already excessive and is increasingly dangerous to distributive justice, human rights and access of marginal populations to basic human goods (Cahill, 2001). Focusing on gene patenting in particular, this author argues that such patenting ought to be more highly regulated, that this regulation should be conducted with international participation and that account should be taken of solidarity and the common good.
The various trends associated with the socio-economic aspects of ag-biotech development have also been reviewed recently (Parayil, 2003). In the words of this author: ‘The dynamics of technology development along the technological trajectories of the Green Revolution and the Gene Revolution could be explicated by the social morphologies of modernization and globalization. The Green Revolution was shaped by the exigencies of modernization, while the Gene Revolution is being shaped by the imperatives of neo-liberal economic globalization’. In other words, the processes of innovation, development and diffusion of technologies followed different paths during these two periods because of the different innovation systems. The success of the Green Revolution was based on international collaboration that included the free exchange of genetic diversity and information. Most of the ‘added value’ present in modern crops has been accumulated over the centuries by farmers themselves as they selected their best plants as the source of seed for the next planting. These ‘land races’ have traditionally been provided free of charge by developing countries to the world community. Amongst the agencies involved, the various Consultative Group on International Agricultural Research (CGIAR) centres add value through selective breeding, and the superior varieties they generate are widely distributed without charge, thereby benefiting both developing and developed countries (Bragdon, 2000; Anon, 2001).
During the Gene Revolution, the situation changed, and much has been written over the last few years on the potentially deleterious effects of plant IPR on the freedom and commercial opportunities of farmers in developing countries (DaSilva, 1998; Conway and Toenniessen, 1999; Lesser et al., 1999; Blakeney, 2000; Wright, 2000; Nuffield Council on Bioethics, 2003; Toenniessen et al., 2003). One of the major reasons that IPR have become an important factor in plant breeding is through the greater use of utility patents (Summers, 2003). Such patents have stimulated greater investment in crop improvement research in industrialized countries, but they are also creating major problems and potentially significant additional expense for the already financially constrained public-sector breeding programmes that produce seeds for poor farmers. For example, it has been calculated (Phillips et al., 2004) that developed countries spend about $5 in research and development for every $100 in agricultural output, whereas developing countries spend only 66 cents.
Patents on biotechnology methods and materials, and even on plant varieties, are thus complicating and undermining the collaborative relationships between international institutions. Public-sector research institutions in industrialized countries no longer fully share new information and technology. Rather, they are inclined to patent and license (Erbisch and Fischer, 1998), and have special offices charged with maximizing their financial return from licensing (Brazell, 2000). Commercial production of any GM crop variety requires dozens of patents and licences (see above, Table 5). It is only the big companies that can afford to put together the IPR portfolios necessary to give them the freedom to operate (Barton, 1997). In addition, now, under the Trade-Related Aspects of Intellectual Property Rights (TRIPS) agreement of the World Trade Organization, most developing countries are required to put in place their own IPR systems, including IPR for plants (Giannakas, 2001). Furthermore, all of this ‘ownership’ of plant genetic resources is causing developing countries to rethink their policies concerning access to the national biodiversity they control (Lesser, 2000), and new restrictions are likely. The trade-related aspects of patents covering the specific example of vitamin A rice (see above) are summarized in Table 7, which shows the number of relevant patents in each of the major rice-producing and rice-importing countries. The complexity of trade between any two countries increases with the sum of the patents in the countries involved.