Accessing intellectual property to develop transgenic horticultural crops

120 CALIFORNIA AGRICULTURE, VOLUME 58, NUMBER 2

Access to intellectual property is a major obstacleto developing transgenic horticultural crops

RESEARCH ARTICLE

▲▲

Gregory D. GraffBrian D. WrightAlan B. BennettDavid Zilberman

▼

Inefficiencies in accessing intellectualproperty (IP) appear to be hinderingotherwise valuable research and de-velopment (R&D) in horticultural cropvarieties. While leading private-sector agricultural biotechnologyfirms with strong IP positions andcommercial freedom to operate (FTO)see insufficient incentives in thesmall, fractured markets of horticul-tural products, researchers with pub-lic-sector support for horticulturalprojects but weak IP positions mayfind that the best way of gainingFTO and moving forward is to bandtogether and provide mutual accessto one another’s technologies. ThePublic Intellectual Property Resourcefor Agriculture (PIPRA), headquar-tered at UC Davis, is a new coalitionof U.S. universities and foundationscommitted to this strategy.

Stories and rumors have circulatedfor years about biotechnology

projects in horticulture being shelvedbecause of intellectual property (IP)conflicts. In a typical situation, a plantscientist at a university agricultural-ex-periment station or a smaller seed firmhas developed a remarkable new vari-ety using the cutting-edge scientifictools of plant biotechnology. Then, asthey or the nursery or the growers’ as-sociation with whom they work takethe next steps to develop and releasethe new variety to commercial growers,their efforts are quickly and quietlyshut down by a letter from an attorney.

The letter alleges that the new varietycontains a piece of technology that in-fringes upon a client’s IP claims. Fur-thermore, the patent owner appears noteven to be interested in negotiating a li-cense. And to this day, the legendaryvariety sits in storage somewhere in agreenhouse or a freezer, unused andsadly neglected.

Of course, it is difficult to establishthe definitive reasons why a projectdoes not come to fruition, especiallywhen there are numerous factors si-multaneously affecting the outcome.Prior patents may be just a convenientexcuse — and the patent owners ascapegoat — for tough decisionsmade to terminate unpromising oreconomically unattractive projects. Still,while patents do provide convincingincentives for private firms to invest inagricultural research and development(R&D), taking the necessary steps to re-spect the rights of patent ownershipdoes add an additional layer of costsfor developing new crop varieties.

Economists call these additional costs“transaction costs”; they include legalfees for searching and filing patentsand expenses for negotiating and draft-ing licenses. Royalties paid for usinganother’s technology are not IP transac-tion costs. Rather, they are “rent” paidto use the technology and to compen-sate for the R&D expenditures spent tocreate it.

Commercial developers of agricul-tural biotechnologies often take mea-sures to avoid incurring these IPtransaction costs. They may shift theirR&D strategies or even acquire othercompanies to avoid dependence onoutside technologies, thereby limitingexpenses and preventing the compli-cations and uncertainties inherent in“renting” them (Graff, Rausser et al.2003). These measures, however, canbe costly too. Either way, costs facedunder an IP system can, in theory, can-cel out the private incentives created byIP to pursue innovation. More trou-bling, IP can even prevent publicly

Gathering the legal rights to all the intellectual property necessary for marketing agenetically engineered product can be daunting, especially for smaller companies andpublic institutions. And the transaction costs to secure “freedom to operate” can beconsiderable.

Article 1, Section 8 of the U.S. Constitutionstates, “Congress shall have power . . . to promotethe progress of science and useful arts by securing forlimited times to authors and inventors the exclusiveright to their respective writings and discoveries.”

http://CaliforniaAgriculture.ucop.edu • APRIL- JUNE 2004 121

funded innovation from having its in-tended social impact. Horticultural ge-netics may be one such area of stalledinnovation. Yet are there any good in-dicators of this stalling beyond just sto-ries and rumors? And if so, can weestablish links with IP?

Biotech R&D trends

Recent U.S. Department of Agricul-ture (USDA) registrations for field trialsof transgenic crops show that R&D inhorticultural crops is lagging whencompared with the major row crops.Even leading transgenic horticulturalcrops such as melon, lettuce, straw-berry, grape, apple and sunflower arehardly represented in field trials (fig.1A; see page 106). Horticultural cropsare completely dwarfed by corn, thesingle most commonly testedtransgenic crop, which by itself is thesubject of almost half of all transgenicfield trials.

Of course, U.S. production of anysingle horticultural crop is far less valu-able than U.S. production of corn. Lessfield-testing is to be expected for lessvaluable crops. But, even when apply-ing a rough calculation to account forthe differences in size and value of in-dividual crops — dividing by the an-nual value of each crop’s U.S.production (fig. 1B) — horticulturalcrops tend to show a greater farm-gatevalue per field trial. In other words,horticultural crops are subject to fewergenetic field trials, and presumably re-ceive less biotech R&D, for every dollarof crop production.

Furthermore, the proportion oftransgenic field trials conducted bypublic-sector research organizations,such as state universities or the USDA,versus the proportion conducted bycommercial firms, varies widely bycrop type (fig. 1C). Public-sector in-volvement in the field-testing of the10 leading transgenic crops — mostlymajor row crops — averages just 15%.Yet, in the next 20 mostly horticulturaltransgenic crops, public-sector in-volvement averages much higher,around 40%.

These numbers should be inter-preted cautiously, as the samples repre-senting many of the horticultural cropsare small and the ratios are taken over

Fig. 1. (A) Top 30 transgenic crops, ranked by total number of field trials registered atUSDA-APHIS from June 1997 to May 2002; (B) value of U.S. crop production in 1997divided by the number of U.S. field trials of the top 30 transgenic crops; (C) percentage ofU.S. field trials of top 30 transgenic crops that were conducted by public-sectoragricultural R&D organizations. In figures 1B and 1C, the dotted lines draw simple linearcomparisons across crops, ranked according to total transgenic field trials, in order toillustrate questions about broad categorical differences in R&D investment for differentcrops; as such, they do not represent statistically tested trends or relationships.

*Indicates row crops; corn is primarily field and some sweet corn.


just a few field trials. For example, 16field trials have been done on trans-genic papaya (all by public research or-ganizations) and only 11 on transgenicwalnut (10 by UC and one by a USDAlab in California).

Despite this variability, there ap-pears to be less investment in biotechfor horticultural crops than for majorrow crops, both in absolute terms andrelative to overall crop values, while agreater proportion of that smaller R&Dinvestment in horticultural crops comesfrom the public sector. Involvement bycommercial firms in horticultural cropsseems to be missing. While this data istoo sketchy to conclude outright thatcommercial firms are underinvestingin horticultural biotechnology, it al-lows us to ask whether they might be,and if so, why.

After a few early excursions intohorticultural crops — most notably byMonsanto, Asgrow and Calgene (bothnow Monsanto subsidiaries) as well as bySyngenta’s predecessors at Zeneca —major agricultural biotechnology firmshave virtually shut down their productdevelopment in horticultural crops.Long-shelf-life tomatoes, virus-resistantsquash and insect-resistant potatoeshave not taken off as did Bt corn andherbicide-tolerant soybeans. Some ofthe specialized vegetable seed firms,such as PetoSeed (which becameSeminis), and some of the smaller agri-cultural biotechnology firms that spe-cialized in vegetable crops, such asDNA Plant Technologies (later bought

out by Seminis), continued their bio-technology efforts a bit longer. Yetthose efforts appear to have all butdried up in recent years.

Instead, fruit and vegetable seedcompanies with active research andproduction activities, such as Seminis,Danson, Golden Valley, Harris Moranand others, continue to pursue theirproduct development goals throughconventional breeding techniques. Oneexception is the Scotts Company, whichis currently seeking regulatory ap-proval for a biotech product for golfcourses, a glyphosate-resistantbentgrass. Indeed, most of the biotechwork in horticultural varieties is con-ducted in university laboratories doingbasic plant science. Occasionally, thoseprojects spin out a commercially inter-esting trait or technology, but univer-sity technology-transfer offices have ahard time finding commercial partnersamong the seed firms, nurseries orgrowers’ associations.

Factors discouraging investment

As with any investment, there is adegree of risk involved in putting re-sources into the development of a newtransgenic horticultural variety. Futurereturns are uncertain, and expected re-turns are weighed against costs in-curred to enter the marketplace. Suchconsiderations also apply, more gener-ally, to public-sector investments in re-search. Although the measures ofsuccess may be more in terms of scien-tific advancements than earned profits,

the practical importance of a new dis-covery is still important. (Consider, forexample, the scientific as well as com-mercial impact of virus-resistant pa-paya [see page 92]).

Market demand. The size andstrength of demand for a newtransgenic variety will determine thesize of returns on the investment. Mar-ket uncertainties for agricultural prod-ucts are nothing new, due to suchfactors as disruptive competition insupply, cyclical price fluctuations andchanges in consumer demand. How-ever, some food consumers, such as inEurope, are skeptical of foods producedusing biotechnology. While a majority ofU.S. consumers seem relatively unfazedby the genetic contents of processedbulk commodities such as soybeans andfeed corn, consumers could react morestrongly to obvious modifications ofproducts in the produce aisle. Yet specificmarket uncertainties surrounding the useof transgenics could be addressed by theselection of technologies and traits thatdeliver real tangible benefits to con-sumers in ways that are perceived asunambiguously safe.

Regulatory approvals. The processof regulatory approvals for GM crops isessential to assure the safety of thetechnology. The R&D costs associatedwith gaining approval are consideredup-front or “sunk” investments, andthey must be spent to gain access to themarket. These costs can be greater if thetransgenic crop contains novel proteinsor pest-control components, as addi-

With grant support from the California Strawberry Commission (CSC), UC scientists genetically engineered the strawberry variety‘Selva’ with a pear fruit gene for resistance to fungal pathogens. However, the engineered lines have not been tested because ofthe CSC’s subsequent reluctance to support that effort. Left, young genetically engineered strawberries in the greenhouse. Right,conventional strawberries.

Cour

t esy

of U

SDA

- ARS

Dar

io C

antu


tional assessments are required.In major row crops, investments to

obtain regulatory approval can be re-couped from the small technology feescharged on each bag of transgenic seed,which are multiplied out over millionsof acres planted; however, with horti-cultural crops the distribution of regu-latory costs is often concentrated ontomuch smaller markets. In many horti-cultural crops, several different variet-ies are commercially important. Ifintrogression of the new trait via back-crossing is not an option, such as maybe the case for clonally propagated vari-eties that do not breed true, each varietymust be separately transformed in thelab, and each must be separately testedand approved. Regulatory costs wouldadd up, but they could not be spread outover nearly as large a market as theycould for row crops. Still, returns per acrefrom horticultural varieties tend to bemuch higher, and the costs of specializedpesticides replaced by transgenic traitsmay also be higher. In addition, regula-tory costs can be expected to decline asmore risk assessments are completed,government agencies become more adeptat judging the merits of different biotech-nologies, and the policies and proce-dures become streamlined and finelytuned. In addition, the extension of anapproach similar to the IR-4 program,which provides regulatory assistancefor pesticides targeted to the needs ofspecialty crops (see page 110), could re-duce the regulatory burden ontransgenic specialty crops.

Access to intellectual property.Transaction costs for gaining freedomto operate (FTO) in the relevant IP-protected technologies can be consider-able. As with regulatory costs, the totalIP transaction costs are independent ofmarket size, and a larger number oftransgenic varieties means more costlynegotiations and more deals to cut. Oneindustry estimate put the costs of nego-tiating a single crop genetics deal ashigh as $100,000. When multiple pat-ented genetic technologies are stackedin a cultivar, as is increasingly the case,the problem is compounded.

Uncertainty over the total amount ofIP transaction costs scares off invest-ment in R&D projects, unless the ex-pected returns are particularlyattractive. This will continue as long asthere is uncertainty in the IP landscapefor plant biotechnologies and geneticmaterials. With the number of patentsin this area growing at an exponentialrate, IP access could be a deterrent tobiotech R&D in horticultural varietiesfor years to come.

IP hurdles for horticultural crops

IP access is a general problem for allof crop biotechnology. The reasons liein the cumulative nature of the geneticsand biotechnologies embodied intransgenic varieties. Plants are com-plex systems, and a healthy, produc-tive crop plant has numerous geneticand metabolic pathways functioningtogether. Those genetics are inheritedfrom breeding stock or can be added

using biotechnology. A geneticallyengineered seed or plant cultivar maycontain three different kinds of tech-nological components that can be pro-tected as IP, including (1) thegermplasm of the plant variety, (2)the specific genes that confer a newtrait and (3) the fundamental tools ofbiotechnology such as genetic mark-ers, promoters and transformationmethods. The IP situation is compli-cated by a number of additional factorsthat add to the transaction costs.

Complex intellectual-property law.Different technological components of atransgenic crop variety are covered inthe United States under different formsof IP law. If a variety is clonally propa-gated, the germplasm — the plant vari-ety itself — can be claimed as IP at theU.S. Patent and Trademark Office(USPTO) under a Plant Patent, estab-lished in 1930 by the Plant Patent Act toprotect against cuttings being taken,repropagated and directly resold underanother name. Seed-propagated variet-ies can be claimed as a form of IP underthe USDA system of Plant Variety Pro-tection (PVP) certification, establishedby the Plant Variety Protection Act in1970. And, since 1980 — following alandmark decision by the SupremeCourt in Diamond v. Chakrabarty overthe patenting of a genetically engi-neered microorganism — all kinds of“invented organisms,” including novelplant germplasm, have come to beclaimed as IP under standard U.S. util-ity patents. (In practice, plant varieties

Several different fruits and vegetables have been geneticallyengineered to resist viral diseases, for which there are often fewsources of natural protection. A single gene from the virus itself isinserted into the plant genome, thereby preventing the virus from

phot

os:T

r evo

r Su

slow

making copies of itself and causing disease symptoms, fruit damageand crop losses. Left, yellow zucchini affected by viral diseases.Right, in a field test, genetically engineered zucchini (right) wasmuch hardier than the conventional crop (left).


being claimed by inventors are almostexclusively corn and soybeans, not hor-ticultural varieties.)

Subsequent technological and legaldevelopments following Diamond v.Chakrabarty now allow utility patents toprotect invented genes, proteins andother gene products, as well as biotech-nology tools such as transformation ofgenetic contents, selection using geneticmarkers, and regulation of expressionusing genetic promoters. Finally, a sig-nificant part of the value of an agricul-tural variety often lies not in itstechnological or biological characteris-tics per se but rather in its recognitionand reputation among consumers inthe marketplace. That “brand” namecan be protected as IP by registering itas a trademark with the USPTO.

The challenges posed by multiplelayers of IP law are, if anything, greaterfor horticultural varieties than for rowcrops: plant patents, PVPs or utilitypatents may cover the germplasm; util-ity patents typically cover the gene andbiotechnology tools used; and trade-marks are more often used to protectvariety names. In leading row cropssuch as corn and soybeans, germplasmas well as the genes and biotechnolo-gies are protected more consistentlyunder only utility patents. While trade-marks like Roundup Ready or LibertyLink refer to input traits and may be ofsome value in marketing to farmers, theidentities of such agronomic traits com-mand little notice or value from finalfood consumers.

Exporting to global markets. Formany important horticultural crops, ex-ports constitute a large share of output,so FTO under IP must include freedomin foreign markets. Since the various IPrights important for plants are adminis-tered nationally, an exporter mustcheck FTO separately in each foreign

market. In general, the tools of biotech-nology are more likely to be patented injust the major markets — such as theUnited States, Europe and Japan — andless likely to be patented in countrieswith smaller markets. Uses of biotech-nologies specifically for minor cropsare less likely to be widely patented inmultiple countries than are uses in im-portant field crops. However, as a re-sult of the International Union for theProtection of New Varieties of Plants(UPOV) agreement first established in1961, PVP systems are widely availableoverseas for the protection of clonallypropagated varieties, and such varietiesdo tend to be widely registered in mul-tiple countries. Still, not all types of bio-technologies, genes or plant germplasmcan be protected in all countries. Forexample, utility patenting of plants isallowed in only a few countries (includ-ing the United States).

Beyond these trends, however, thereare no hard-and-fast rules as to whichtechnology will be protected in whichcountry, as each inventor decideswhere to seek protection (Binenbaum etal. 2003). As a result, those seeking FTOare confronted by an often bewilderinginternational patchwork of IP rights,where the negotiations needed for aparticular transgenic variety can differsignificantly each time it crosses a na-tional border.

Intellectual-property holders. Un-less a new transgenic variety is devel-oped by an integrated effort at a largecompany backed by a broad IP portfo-lio, a number of different owners — in-cluding companies, individuals,universities and even governments —will have valid IP claims over the tech-nologies and genetic contents that endup being included in it. That meansthere are numerous owners to trackdown, negotiations to conduct, billable

legal time to hire, and multiple royaltypayments to administer. The costs andheadaches involved in working out“who owns what” and “who oweswhat to whom” can balloon into whateconomists call the “tragedy of the anti-commons” and render the developmentprocess unfeasible. The “tragedy” is ar-guably worse in horticultural cropsthan in row crops. Given the smallermarkets involved, there is less incentivein industry to consolidate IP portfoliosaround horticultural crops. Also, notone of the public-sector organizationsor their typically smaller commercialpartners in horticultural crop develop-ment has a complete IP portfolio inplant biotechnology.

Uncertain ownership of rights.When technologies are patented, it isoften not clear who currently owns par-ticular aspects of each technology. Thisuncertainty is cleared up in the courtsthrough patent interference cases,where attorneys and scientists under-take intensive “surveying” of the“property lines” between the patentsand technologies in question. Some-times these cases drag on for years,keeping key technologies in legal limboand the R&D community guessing as towho is the rightful owner. Yet, for mostregistered patents there is no such scru-tiny. As a result, the boundaries for aconsiderable expanse of technologicalterritory are not clearly demarcated,creating considerable uncertainty as towhen a new application could be con-sidered to be infringing or “trespass-ing.” In horticultural crops, the lack ofclarity about the scope and validity ofpatent claims is especially important.Because the markets are smaller, fewerproducts have been developed andfewer contests have been fought to es-tablish legal precedents. Furthermore,just the anticipation of possible legal

The U.S. Patent and Trademark Office manages intellectual propertyunder the Plant Patent Act of 1930, the Plant Variety Acts of 1970and 1994, and general utility patents that can cover the products of

Ruth

Nyb

old/

USP

TO

Skid

mor

e, O

win

gs &

Mer

r ill L

LP

and processes used to develop genetically engineered seeds andcrops. Left to right, U.S. patent office locations past (Washington,D.C.), present (Arlington, Va.) and future (Alexandria, Va.).


costs can shut a project down before itever gets off the ground.

Transfer of rights. IP covering acrop variety may be sold, licensed ortransferred to another organization atany time. The transfer of rights can oc-cur either in part (nonexclusively) or inwhole (exclusively). The transfer canhappen in just one territory where it isprotected (such as the United States) orin multiple territories. The transfer ofrights for a biotechnology tool or genecould be specified for use in just onecrop (such as corn), in several crops(such as all cereal grains), or in any andall crops. Finally, to make mattersworse, the fact that the IP rights havebeen transferred may be consideredcommercially sensitive information andnot be made public.

Other issues. Any organizationmanaging the release of a new crop va-riety faces uncertainty about which IPrights actually cover what technolo-gies, who holds those rights in whichcountries, and to what degree a spe-cific new transgenic variety infringeson those rights. Resolution of suchuncertainty is not less costly for cropswith smaller market values. Even af-ter reliable information is obtained,uncertainty remains about negotiatingthe permissions. IP owners are not re-quired to negotiate licenses, and theymay feel there is not enough potentialrevenue in minor crops to make theirlicensing efforts worthwhile. Theymay also be concerned about technol-ogy stewardship, given the nervous-ness among consumers about foodbiotechnology and its status as a hotmedia topic. They may worry that themishandling of their technology by asmall and relatively inexperiencedhorticultural player could lead tostronger regulations, potentially erod-ing that technology’s value in its major

crops, or jeopardize public perceptionsabout biotechnology overall.

Public-sector IP management

In response to IP congestion andcontinuing uncertainties, several lead-ing U.S. public-sector agricultural re-search organizations have cometogether to create the Public IntellectualProperty Resource for Agriculture(PIPRA), an organization providing col-laborative IP management solutions topublic-sector and smaller private-sectorplayers in horticulture (Atkinson et al.2003; see sidebar, page 127). While indi-vidual universities and even the USDAhave small and uncoordinated IP port-folios in plant genetics, together theyhold a fairly comprehensive set of tech-nologies that could be useful for devel-oping transgenic varieties (Graff,Cullen et al. 2003). PIPRA seeks to coor-dinate the disparate portfolios of itsmember organizations to support spe-cialty crop applications. With the of-fices of technology transfer of itsmember organizations, PIPRA is pursu-ing several cooperative strategies.

Licensing terms. First, PIPRA seeksto develop and adopt more preciselyfocused terms of licensing, with specificdistinctions for the “fields of use” towhich a technology is licensed. A com-pany that licenses a technology in-vented at a university can still get the

full benefit of using the technology inthose major row crops in their line ofbusiness, even if the license clearly de-fines and grants exclusive use of thetechnology in just those crops. Such alicense effectively “reserves” the rightsto use the technology in any othercrops. Horticultural firms could thenmake separate agreements with theuniversity to use the technology in onlytheir defined specialty crops. An ad-vantage of this strategy is that it canalso apply to other minor uses, includ-ing “alternative” crops (such as cassavaor millet) and humanitarian applica-tions in staple crops for developingcountries (such as vitamin A–fortifiedGolden Rice). By discriminating be-tween big markets and multiplesmaller markets — including thosewith limited commercial value but im-portant social benefits — public-sectorscientists could see their inventionsearn royalties in the big markets of ma-jor row crops while still helping to im-prove smaller crops or increase foodsecurity in world’s poorest regions.

PIPRA database. A database will, forthe first time, list in one place currentinformation about all of the patents ofPIPRA’s members and their availabilityfor licensing alongside informationabout technologies published in the sci-entific literature (and thus publiclyavailable), in sufficient detail to identify

New biotech crops must meet theintellectual-property and regulatoryrequirements of importing countries, andthere are no firm rules as to whichtechnologies will be protected or regulatedin which countries. This situation can createserious difficulties for exporters. Right, foodmarket in Benin; far right, Ethiopia.

IFPR

I/Ele

ni G

abr e

- Mad

hin

1999

IFPR

I/Phi

l ippe

Ber

ry


which technologies can be accessed forwhich uses. The database will offer aclear, complete and certain “universallisting” of technologies available fromPIPRA’s member organizations andthe public domain.

Commercial patent databases andprofessional legal staff are available toresearchers in large private companiesfor searching through the “prior art”(records of what is already patented) tomake FTO analyses of a new product’sIP position. Such resources are seldomavailable to academic and governmentresearchers. The PIPRA database willdecrease uncertainty about what cannotbe used by showing what can be used.

Patent-pooling. PIPRA is investigat-ing the creation of patent-poolingmechanisms, which would collect IPsubmitted from its member organiza-tions, package the technologies to-gether and offer unified licenses for the“bundled” IP in a field of use, such as aspecific crop, or in a particular state orcountry. This process mimics, in a vir-tual way, how large commercial firmshave assembled their IP portfolios toprovide FTO in major field crops. Itsfeasibility will depend — at least at theoutset — on the extent to which public-sector organizations are able and will-ing to provide access to patentscovering key enabling biotechnologytools already licensed to the corporatesector.

Even if used to access technologieson just a patent-by-patent basis, coor-dinated information and streamlinedaccess to academic and government-owned IP could help decrease trans-action costs and improve efficiency intechnology-transfer markets. There isample room for improvement here, assome have complained that negotiatinglicenses from universities and govern-ment agencies is often less efficientthan negotiating licenses from firms.PIPRA can improve public-sector tech-nology transfer for agriculture by pro-viding information, tools andprecedents for efficient licensing.

Greater opportunities lie in the

The costs and headaches involved in working out “who owns what” and“who owes what to whom” can balloon into what economists call the“tragedy of the anti-commons” and render the development process unfeasible.

steps being taken to coordinate ac-cess, package IP bundles and targetuses in lower-value markets such ashorticultural crops and traits impor-tant for food security in developingcountries. These are, generally speak-ing, areas that commercial firms arenot interested or capable of serving.Such collaboration is not surprising,given the history and ethos of coopera-tion among agricultural experiment sta-tions within the land-grant system.Public-sector institutions also havegreater legal flexibility to enter into col-lective IP management arrangements,given historical antitrust concernsabout abuses of patent-coordinationefforts in industry.

Even more important will be the es-tablishment of ongoing precedents andmechanisms for the treatment of futureIP. Academic and government re-searchers will go on making importantdiscoveries and inventing new tech-nologies for agriculture. Those futureinventions will, from their inception, behandled in ways — such as being listedin the universal database, licensed fortargeted “fields of use” and included inIP pools — that will make them acces-sible in a carefully proscribed manner,not just to top commercial bidders, butto anyone else in the broader agricul-

tural community who can make gooduse of the technology, including horti-cultural researchers and growers.

G.D. Graff is Researcher, B.D. Wright isProfessor, and D. Zilberman is Professor,Department of Agricultural and ResourceEconomics, UC Berkeley; and A.B. Bennettis Professor, Department of Vegetable CropScience, UC Davis, and Executive Direc-tor, Office of Technology Transfer, UC Of-fice of the President. Wright and Zilbermanare members, Giannini Foundation.

References

Atkinson RC, Beachy RN, Conway G, et al.2003. Public sector collaboration for agricul-tural IP management. Science 301:174–5.

Binenbaum E, Nottenburg C, Pardey PG,et al. 2003. South-North trade, intellectualproperty jurisdictions and freedom to oper-ate in agricultural research on staple crops.Ec Dev Cultural Change 51:309–35.

Graff GD, Cullen SE, Bradford KJ, et al.2003. The public-private structure of intel-lectual property ownership in agriculturalbiotechnology. Nature Biotech 21:989–95.

Graff GD, Rausser GC, Small AA. 2003.Agricultural biotechnology’s complementaryintellectual assets. Review Econ Statistics85:349–63.

Right, the orange canola seeds have beengenetically engineered to produce highlevels of beta-carotene. Monsanto haslicensed the technology and is workingwith the Tata Energy Research Institute inIndia and Michigan State University’sAgricultural Biotechnology SupportProgram to develop high beta-carotenemustard for possible use in India. Above, aconventional canola plant.

Sacr

amen

t o B

ee/C

hris

Cr e

wel

l

Cano

la C

ounc

i l of

Can

ada


Deborah Delmer

Public resource for ag

Universities and other nonprofitinstitutions have generated many keypatents related to agricultural bio-technology and they will most likelyremain an important source of inno-vation. However, no single institutionhas the complete package of technolo-gies required for commercialization ofa biotech variety. Although some insti-tutions are developing ways to dealwith these problems, there are stillmany examples of public-sector inven-tions that have been licensed exclu-sively to private-sector partners. In late2002, representatives of more than adozen U.S. public-sector agricultural re-search institutions (including UC)joined with the U.S. Department of Ag-riculture and the Rockefeller andMcKnight Foundations to discuss ac-cess to patented agricultural technolo-gies for the development anddistribution of improved specialtycrops and those targetedfor the developing world.

The participating institu-tions are pursuing an initia-tive called the PublicIntellectual Property Re-source for Agriculture (PIPRA), whichwill explore the following collaborativeIP management strategies:

• Developing licensing strategies thatretain specific rights for humanitarianand specialty crops while allowingand encouraging licensing of suchtechnologies to the private sector.

• Developing and maintaining a data-base of public-sector IP assets thatincludes information on IP licensingstatus and technologies that are inthe public domain.

• Creating technology “packages”that would provide both FTO andthe material resources (such as vec-tors, promoters and trait genes) re-quired for specialty or humanitarianapplications.

At present, PIPRA focuses only onIP generated by them, whereas another

new entity called the African Agricul-tural Technology Foundation is dealingwith access to private-sector technologiesfor targeted crop applications in sub-Saharan Africa (www.aftechfound.org).Other issues that limit the commercializa-tion of genetically engineered crop variet-ies — such as public acceptance and highcosts for regulatory approval — must beaddressed by other initiatives.

Next steps

PIPRA is moving forward with aprocess that will forge collective actionin technology management among asignificant number of nonprofit institu-tions active in agricultural biotechnol-ogy research. This initiative is intendedto be widely inclusive; at present about25 major institutions are involved andmore are being sought. UC Davis pro-fessor Alan Bennett was recently se-lected as PIPRA’s executive director,and its offices will be based on campus.PIPRA intends to move forward in a

cooperative manner and craft solutionsto problems that arise. To this end, ac-tivities are under way that will answerimportant questions about the types ofcollaboration needed, including devel-oping case studies, assembling a data-base, involving additional public-sectorinstitutions and engaging private-sectorand other key stakeholders. The groupis stimulating broad discussion to helpuncover the implications of new IPmanagement strategies, and to identifycritical issues that must be addressed tomake this initiative successful.

D. Delmer is Associate Director for FoodSecurity, Rockefeller Foundation, NewYork. For more information go to:www.pipra.org.

Nonprofit institutions form intellectual-propertyresource for agriculture

No single institution has the completepackage of technologies required forcommercialization of a biotech variety.


Current practices in patenting andintellectual property (IP) protec-

tion have created barriers to the use ofbiotechnology and advanced agricul-tural technologies for the creation andcommercialization of new crop variet-ies. The complex and cumulative na-ture of biological innovation requiresaccess to multiple technologies that areoften exclusively owned or licensed.For example, commercializing a singlevariety of transgenic tomato could in-volve obtaining the rights to use a vari-ety of technologies and genes from nu-merous life-sciences companies,government agencies and universities.

Obtaining “freedom to operate”(FTO: the ability to clear all IP barriersand bring a product to market) fortransgenic crop varieties is difficult.There is considerable uncertainty as towho holds what rights to particulartechnologies, and negotiating access tothose rights is time-consuming andcostly. This is a problem for the majorinternational agricultural companiesthat focus primarily on high-volumecrops such as corn, soybeans and cot-ton; for research institutions that workon specialty crops grown on muchsmaller acreages, such as tomatoes,strawberries, apples and cabbage; andfor public institutions that work onstaple crops for humanitarian use in de-veloping countries.

The international agricultural com-panies have taken steps to solve theirFTO problems through mergers andcross-licensing agreements that bringtogether essential IP components withinone company. However, public-sector in-stitutions — such as universities, govern-ment agencies, international agriculturalresearch centers and others working onspecialty and staple crops — are stillstruggling to find ways to gain FTO. Inaddition, donor agencies such as theRockefeller and McKnight Foundations,which have a long history of investing inagricultural research that benefits subsis-tence farmers in developing countries,have also found that IP constraints are re-ducing the flow of technology.

Documents

Accessing intellectual property to develop transgenic horticultural crops