16 <strong>Bt</strong> <strong>Brinjal</strong>: <strong>The</strong> <strong>GEAC</strong> <strong>environmental</strong> <strong>risk</strong> <strong>assessment</strong>Inserted TransgeneFinding 8. <strong>The</strong> description <strong>of</strong> <strong>the</strong> transgene in <strong>Bt</strong> brinjal EE-1 is inadequate to support <strong>risk</strong> <strong>assessment</strong>. It isnot clearly known that <strong>the</strong>re is only one transgene incorporated into <strong>the</strong> brinjal genome, that <strong>the</strong> inserted transgeneexpresses <strong>the</strong> intended gene product or that <strong>the</strong> transgene does not interrupt a functional plant gene.(A) Number <strong>of</strong> inserts. <strong>The</strong> Dossier indicates that one transgene insert remains in <strong>Bt</strong> brinjal EE-1. Based on <strong>the</strong> evidenceprovided, <strong>the</strong> conclusion is not fully justified. <strong>The</strong>re are several sources <strong>of</strong> uncertainty that weaken <strong>the</strong> conclusion, <strong>and</strong> three <strong>of</strong><strong>the</strong>se sources <strong>of</strong> uncertainty may affect <strong>risk</strong>. Two may increase <strong>risk</strong> <strong>and</strong> one may increase or decrease <strong>risk</strong>.• Uncertainty arises because <strong>the</strong> size <strong>of</strong> <strong>the</strong> digoxigenin (DIG)-labelled <strong>Bt</strong> probe is not specified. If it is too large, it may notdetect independently inserted fragments;if it is too small, it may not detectBox 1independently inserted fragments notBackcrossing <strong>and</strong> Transgene Characterizationassociated with <strong>the</strong> probe.Backcrossing is one way to create crop varieties that are similar to previous elite• Uncertainty also arises because probes varieties with a few new traits added. It is also a useful way to clean up “dirty”transformation events that generate transgenic plants.for o<strong>the</strong>r parts <strong>of</strong> <strong>the</strong> plasmid, such asAt <strong>the</strong> time <strong>Bt</strong> brinjal event EE-1 was created, all know methods for creating<strong>the</strong> nptII gene <strong>and</strong> several o<strong>the</strong>rs, are not transgenic events created “dirty” transgenic events, <strong>and</strong> EE-1 is no exception. Aused. If fragments are inserted in such “dirty” event is one in which <strong>the</strong>re are multiple transgene insertions, <strong>of</strong>ten withinversions <strong>of</strong> <strong>the</strong> transgene, interspersion <strong>of</strong> plant DNA, artificially createda way that a new open reading frame isintrons, <strong>and</strong> transgene fragments. All <strong>of</strong> <strong>the</strong> “dirty” elements <strong>of</strong> <strong>the</strong> event couldcreated that is associated with a plant create serious <strong>risk</strong>s to <strong>the</strong> environment.promoter, novel gene products may be<strong>The</strong> multiple insertions typically end up on different plant chromosomes or somedistance apart on <strong>the</strong> same chromosome. Recurrent backcrossing can help toexpressed at high rates.clean up <strong>the</strong>se dirty events.• <strong>The</strong> third source <strong>of</strong> uncertainty isOnce a transgenic plant cell is created, it must be regrown into a whole plant.Backcrossing involves crossing <strong>the</strong> newly created transgenic plant with an eliteassociated with <strong>the</strong> breeding program.variety or line that <strong>the</strong> breeder wants <strong>the</strong> transgenic plant to resemble. RecurrentAlthough <strong>the</strong> file makes it clear that <strong>the</strong> backcrossing involves multiple generations <strong>of</strong> crossing <strong>of</strong>fspring back to <strong>the</strong> sametransformed plant is backcrossed to elite elite variety or line as diagrammed in <strong>the</strong> figure below.brinjal germplasm, it is not clear howRecurrent Backcrossing to Elite Line(B 1is <strong>the</strong> first backcross generation, etc.)many backcross generations were done.B 1B 2B 3B 4Each backcross generation reduces <strong>the</strong>EliteLineprobability that an additional integrationevent is retained into <strong>the</strong> final brinjalX X X Xproduct. In o<strong>the</strong>r words, <strong>the</strong> moreTransformedPlantbackcross generations, <strong>the</strong> less likely anTransgeneundetected integration event is retained in<strong>the</strong> final brinjal product (Box 1). Because Proportion <strong>of</strong><strong>the</strong> number <strong>of</strong> backcross generations is Transformed Plant (½) 1 (½) 2 (½) 3Remaining in Offspringnot specified for each <strong>of</strong> <strong>the</strong> <strong>Bt</strong> brinjalproducts, it is uncertain how well <strong>the</strong> For brinjal’s 12 linkage groups (Doganlar et al. 2002), about 10 backcrossinserted transgene has been characterised. generations are necessary to eliminate all but <strong>the</strong> chromosome with <strong>the</strong> transgenewith 99% probability. Ten BC generations would also eliminate all but 0.1% (on(B) Sequence <strong>of</strong> <strong>the</strong> inserted transgene. average) <strong>of</strong> <strong>the</strong> genetic material associated with <strong>the</strong> transgene. If only 5 backcrossUnder <strong>the</strong> assumption that <strong>the</strong>re is only generations were used, <strong>the</strong> probability that all o<strong>the</strong>r chromosomes have beeneliminated is only 62%, <strong>and</strong> about 3% <strong>of</strong> <strong>the</strong> chromosome associated with <strong>the</strong>one transgene inserted into <strong>Bt</strong> brinjal EE-1transgene remains. If only 5 backcross generations were used, a significantas asserted by <strong>the</strong> file, <strong>the</strong> sequence <strong>of</strong>amount <strong>of</strong> germplasm that potentially contains unintended transgenes remains inthis one transgene was not reported. <strong>The</strong> <strong>the</strong> final variety. This creates unnecessary <strong>risk</strong>s to <strong>the</strong> environment that couldhave been easily reduced with a bit more breeding effort. Hence <strong>the</strong> number <strong>of</strong>sequence <strong>of</strong> <strong>the</strong> inserted transgene is typicallybackcross generations is an essential element to characterizing a transgene.required by regulatory authorities around <strong>the</strong>
Characterisation <strong>of</strong> Transgene 17world. Although <strong>the</strong> sequence <strong>of</strong> <strong>the</strong> <strong>Bt</strong> gene, <strong>and</strong> all <strong>of</strong> <strong>the</strong> marker genes was known (but not provided), this is <strong>the</strong> sequence<strong>of</strong> <strong>the</strong> plasmid, not <strong>the</strong> sequence <strong>of</strong> <strong>the</strong> inserted gene. <strong>The</strong>re are typically differences between <strong>the</strong> two. <strong>The</strong> sequence <strong>of</strong> <strong>the</strong>inserted transgene is necessary to substantiate <strong>the</strong> logic behind <strong>the</strong> reasoning that <strong>the</strong> marker genes pose no <strong>risk</strong>, <strong>and</strong> that <strong>the</strong>Cry1Ac protein is virtually identical to o<strong>the</strong>r Cry1Ac proteins that have been studied previously. <strong>The</strong> sequence will also enableidentification <strong>of</strong> unintended open reading frames (ORFs), which could result in unintended protein expression.(C) Flanking regions. Information on flanking regions is not provided. Although <strong>the</strong>re are limitations on <strong>the</strong> extent <strong>of</strong>genetic (DNA) analysis that can be done on brinjal because <strong>the</strong> entire brinjal genome has not yet been sequenced, <strong>the</strong>re remainseveral significant <strong>and</strong> important elements that were not done. Typically an inserted transgene will be flanked by a mixture<strong>of</strong> plasmid <strong>and</strong> plant DNA, which is fur<strong>the</strong>r flanked by only plant DNA associated with <strong>the</strong> original plant chromosome.Sequencing through <strong>the</strong> mixed flanking regions to <strong>the</strong> plant flanking regions is now routinely required by many regulatoryauthorities around <strong>the</strong> world. <strong>The</strong> flanking sequences are needed to determine if <strong>the</strong> transformation process has inadvertentlycreated novel open reading frames <strong>and</strong> to determine if a functioning plant gene was interrupted by <strong>the</strong> insertion event. Lacking<strong>the</strong> full brinjal sequence, it is not possible to determine unambiguously if a plant gene was or was not interrupted, but because<strong>of</strong> <strong>the</strong> high correspondence <strong>of</strong> <strong>the</strong> brinjal <strong>and</strong> tomato genomes, <strong>the</strong> flanking region sequences can be blasted against <strong>the</strong> tomatogenome until such time that a full brinjal sequence is available.Needed Experiments <strong>and</strong> Observations1. Probe sensitivity. Provide <strong>the</strong> size <strong>and</strong> sequence <strong>of</strong> <strong>the</strong> DIG-labelled <strong>Bt</strong> probe <strong>and</strong> identify <strong>the</strong> part <strong>of</strong> <strong>the</strong> transgene to whichit is expected to bind.2. Transgene number. Use additional probes that provide complete coverage <strong>of</strong> <strong>the</strong> original plasmid to explore for incorporatedtransgene fragments associated with o<strong>the</strong>r parts <strong>of</strong> <strong>the</strong> transgene. Report <strong>the</strong> size <strong>of</strong> <strong>the</strong>se probes <strong>and</strong> <strong>the</strong> detection limits <strong>of</strong><strong>the</strong> Sou<strong>the</strong>rn blots.3. Transgene sequence. Sequence <strong>the</strong> inserted transgene <strong>and</strong> any o<strong>the</strong>r inserted fragments identified in 1) <strong>and</strong> 2).4. Flanking regions. Sequence <strong>the</strong> regions flanking <strong>the</strong> inserted transgene (<strong>and</strong> transgene fragments) until continuous brinjalDNA is encountered.5. Gene disruption. Identify <strong>the</strong> brinjal DNA in <strong>the</strong> flanking regions <strong>and</strong> determine if it is associated with a potentiallyfunctioning brinjal gene. As long as <strong>the</strong> brinjal genome has not been sequenced, <strong>the</strong> tomato <strong>and</strong> potato genomes can beused for this purpose. If it is associated with a potentially functioning brinjal gene, determine <strong>the</strong> function <strong>and</strong> its potentialeffects on <strong>the</strong> plant.