Can GM Crops be Introduced Into Crop Centres of Origin and Diversity?Prof. C. Kameswara Rao and Dr. S. Shantharam
Critics of the use of GM technology for crop improvement argue that the introduction of transgenic (GM) varieties into the Centres of Origin (CO) and/or Centres of Diversity (CD) of the concerned crop plants would eliminate the existing diversity and impoverish natural genetic resources. This is scare mongering that has now become an emotional and sentimental issue of serious proportions, but is bereft of any rationality, with science being paid a Nelson's eye.

There is certainly a possibility of transgenics and their wild or cultivated relatives inter-crossing in nature. Since this is a very broad generalization, only a crop-wise and region-wise scientific evaluation of possible events and their consequences should guide our decisions and not rhetoric.
Centres of Origin of cultivated plants are identified on the basis of the number and diversity of wild species as well as the number of endemic species of the concerned genus in a given region, while the Centres of Diversity are recognized on the basis of the number and diversity of different varieties, wild and cultivated, of the species. The Centres of Origin and Centres of Diversity of crop plants as known to us are largely based on circumstantial evidence. In the cases of crops that are extensively cultivated over wide geographical ranges, a large number of new varieties were continuously developed, involving a large number of parents, making the issues virtually intangible. For example, IR-64 rice appears to have had more than 100 parents, with consequent extensive genomic rearrangements, some natural and the others induced.

It is often forgotten that species are placed in the same genus based on various taxonomic criteria and this does not necessarily mean that all the species in a genus are genetically related to each other. Genetic relationship of species and varieties should be determined on the basis of the degree of crossability. For species to interbreed producing fertile offspring with ease, their genomes should be compatible. Such considerations did not enter the early taxonomic treatments or the current criticism.

For various natural causes, the number/diversity of species/varieties in the CO may dwindle in time while the same species/varieties may be very successful in their new homes, which are not the CO. Hence CO and CD need not necessarily be the same. It is not a settled issue that India is a CO of rice while it is certainly a CD. So far as cotton is concerned it is neither a CO nor a natural CD.
Desired genes from the gene pools of crop plants need not be sourced from living plants from the CO or CD. They may come from anywhere, such as collections in research institutes, gene banks, pollen banks or such ex situ sources. The seed in various collections may not be viable after a time, but with the deployment of techniques of molecular biology, any gene(s) from any source can now be isolated and utilized.

Distinctness of species and varieties is maintained in nature by the operation of reproductive barriers in various forms. If all species, even of the same genus, can intercross freely in nature, we would not have had so many wild species considered as related to the cultivated species. Varieties of a crop plant species may cross with each other in a greater frequency than species, but even this is not of such a common occurrence. Farmers have maintained the distinctness of varieties crops, of even those that interbreed freely in nature. Cabbage, cauliflower, Brussels' sprouts, broccoli and knoll-kohl are taxonomic varieties of the same species Brassica oleracea. Grown in the proximity of each other and unattended, these crops freely interbreed and lose their identity in a few generations, because they are not reproductively isolated from each other. For over a couple of centuries, farmers have maintained distinct, not just these five crops, but several cultivars under each of them, which is not an easy task. Most other crop plant species do not pose similar difficulties.

If natural hybridization is a rare event, it is proportionately difficult under experimental conditions. It was estimated that it needed over 100,000 artificial crosses performed by several research groups over several decades, before a successful hybrid between Raphanus and Brassica was produced. It involved a dogged pursuit of over 100 years to produce a fertile Triticale by a very large number of plant breeders.

It is possible that there may be a few instances of natural hybridization in otherwise non-crossing species. But a few chance hybridizations do not mean anything unless the hybrids produce fertile offspring from the first generation onwards, and can back- cross. Even after such introgression, unless the new characters have an adaptive value, the new gene combinations do not survive in nature. Alternatively, plant breeders should select them for their agronomic or economic potential. Actually this was what has happened before artificial means of crop improvement were developed.
If transgenic varieties hybridize with their wild or cultivated relatives in nature, the frequency of such events cannot be more than what has been happening in nature, before the introduction of transgenics into the environment. It is an absurd contention that transgenes enhance the promiscuity of crop plants.

Since rice and cotton crops are often at the forefront of arguments against the introduction of transgenics of these crops into the environment through commercial cultivation, more particularly into areas supposed to be rich in diversity, issues related to these two crops are discussed here.

RICE: Of about 25 species in Oryza, the following species have genomes designated AA, similar to that of Oryza sativa, the cultivated species (the additional areas of distribution are given in parenthesis):

Tropical Australia: Oryza meridionalis (AA) Endemic; Oryza rufipogon (AA) (Tropical Asia) Tropical Asia: Oryza nivara (AA) (India); Oryza rufipogon (AA) (Australia) India: Oryza nivara (AA) (Tropical Asia); Oryza rufipogon (AA) (Australia) Tropical Africa: Oryza barthii (AA) Endemic; Oryza longistaminata (AA) Endemic West Africa: Oryza glaberrima (AA) Endemic? Central and South America: Oryza glumaepatula (AA) Endemic

While considering the introduction of rice transgenics into a particular area, reliable data on the accurate distribution of these species are essential. Several of these species are endemics and so are of concern only in the respective regions.

The following is a set of results of experimental inter-specific hybridization in Oryza, with the percentage of seed set given in parenthesis. The species listed are pollinators while the cultivated Oryza sativa was the female parent.

1. Oryza sativa complex (genome AA; diploids); Oryza nivara (9.1 to 56.7); Oryza rufipogon (18.5 to 73.0)*; Oryza glaberrima (29.5 to 56.7)
2. Oryza mayeriana complex (genome not designated; diploids); Oryza mayeriana ssp. granulata (zero)*
3. Oryza ridleyi complex (genome not designated; tetraploids) ; Oryza ridleyi (0.0 to 7.7)*
4. Oryza officinalis complex (diploids); Oryza officinalis (CC; 5.9 to 17.3)*; Oryza australiensis (EE;0.5 to 3.8)*; Oryza latifolia (CCDD; 0 to 25)*; Oryza grandiglumis (CCDD; 6.7)*
5. Not in any complex Oryza brachyantha (FF; 0 to 1.1)*

* Embryo rescue techniques were needed for the recovery of seed of these crosses. No data are available on the viability of this first generation seed and on the subsequent generations.
Only Oryza nivara and Oryza rufipogon have produced hybrids with Oryza sativa, with seed set of any significance. Interspecific hybridization between the cultivated and wild species does not seem to occur in nature. Hybrids of Oryza sativa with species even in the same genome group are highly sterile and may require embryo rescue. Such hybrids cannot survive in nature.

For the following reasons the impact of the introduction of new rice varieties in the CDs of rice is manageable, with a few precautions in place:

1. Commercial varieties of Oryza sativa are neither normally sympatric nor naturally panmictic with the wild species of Oryza.
2. Rice is a predominantly self-pollinated crop. The viability of rice pollen is only for about five minutes and the receptivity of the stigma is about 20 minutes, though the florets may remain open for over an hour. The distance of horizontal dispersion pollen is short, about six to seven meters. Even if it is 100 meters, a mere dispersion of pollen, if they are not viable, is of no consequence.
3. Data on the distribution of the wild species are generally vague. Generalized descriptions as 'South India' or 'Tropical Asia' or 'Central America', are misleading. The wild species have a pronounced disjunct distribution and appear and disappear, in the same area, time and again. Even within a particular country, precise data on the distribution of the wild species are needed. For example, in India, the distribution of Oryza meyeriana is given as 'Southern, Eastern and North Eastern India'. but the species occurs only in small isolated pockets and not throughout these regions. The introduction of new varieties needs to be done cautiously only in the pockets of the occurrence wild relatives. Even then, a separation distance of 20 meters eliminates the chances of contamination and even 10 meters makes it unlikely.
4. Planting a refuge of some species which are taller the than the rice plants, around the field of the transgenic variety will provide a pollen screen. If these species are of some of economic importance (such as green manure), the farmer derives an additional benefit. Since rice grows in waterlogged conditions for most part, a non-rice refuge within the rice field is not a workable proposition.
5. The spread of the introduced genes requires positive selection potential in the hybrid background. Granting this, genes for nutritional enhancement such as _-carotene, iron and others, do not cause any adverse impact on the environment. Genes for herbicide tolerance operate only when triggered by the herbicide. Since rice plants cannot hybridize with any other grass species, introduction of transgenes for herbicide tolerance, abiotic stress such as salt, drought, shade and flood tolerance, are not of any appreciable consequence. Rice is a very delicate crop requiring great care. Even if any of these genes get incorporated into other varieties of rice, the chances of survival of these hybrid swarms and the consequences of that event cannot be alarming. Transgenics with such genes are any way meant for introduction in small areas.
6. It is actually necessary to bring transgenic (GM) varieties of rice into large-scale cultivation in order to understand the probable problems and to device mechanisms to contain them. The problems, if any, are easily taken care of, if the impact of the new varieties is evaluated case-by-case, region-wise.
7. To a naturally possible extent, commercial varieties of rice have been exchanging genes among themselves and with the wild species of Oryza all along and there is no evidence of the hybrids surviving to any considerable extent or of having any significant impact on the environment.
8. In conclusion, while caution should certainly be the watch word, there are no alarming possibilities that warrant a blanket ban on the introduction of transgenic (GM) rice for commercial cultivation.

COTTON: Since 1890, when the first hybrids of cotton with superior qualities were produced, cotton breeders have put in enormous efforts and time to continuously develop inter-specific and inter-varietal hybrids, with the result the current cultivated cottons cannot be ascribed to any taxonomic species. They never existed in nature as they are synthetic and have not evolved through natural means.
Species of Gossypium and cultivated cotton are only rarely sympatric. In India, the only wild species is Gossypium stocksii, a migrant from the Middle East, which occurs in North Western Gujarat, where cotton is not cultivated.

Cotton pollen are among the heaviest in the angiosperms, with about 70 per cent of hydration. They are spinescent, sticky and tend to clump. This and the structure of the cotton flower make it extremely difficult for the cotton pollen to be wind borne. Cotton being a self- pollinated crop, the chances for cross-pollination are quite low. The refuge takes care of any pollen drift.

The most prominent and controversial transgenes in cotton are the Bt genes for pest resistance. Natural hybrids between species of Gossypium and cultivated cotton varieties are not a common event. There can be some small degree of hybridization between the transgenic and the cultivated cotton varieties, but this cannot be more than what has been happening in nature, among the cotton varieties, all the time. Even in such a rare event, even if the Bt gene is transferred to another variety of cultivated cotton, how can the consequences be catastrophic? In fact, such an event amounts to a free transfer of expensive and time-consuming technology without the burden of the regulatory process. If we are innovative we can make the best use of the situation.

The scientists should provide information to answer the arguments against introduction of transgenics into CO and CD. Otherwise, the public would get misled and avoidable opposition, based in ignorance and/or misinformation, builds up.

C. Kameswara Rao is at Foundation for Biotechnology Awareness and Education, Bangalore, India; krao@vsnl.com; and S. Shantharam is at Biologistics International, Ellicott City, MD, USA; sshantharam@biologistics.us