NOTE: Below's a very interesting comment by the molecular biologist Dr Michael Antoniou on the widely reported paper published last week in Nature (VOL.460 NO.7258) on so called "snorkel rice", involving genes that allow rice to survive in deepwater flooding by shooting up in height.
http://www.nature.com/nature/journal/v460/n7258/edsumm/e090820-14.html

The paper was reported in both the science and popular media with headlines that were often dramatic, eg "Snorkel rice could feed millions" (BBC). Some of these reports mistakenly claimed that this was a GM rice.

One widely circulated popular science piece, for example, was headlined, "Genetically Engineered Rice Plants Grow 'Snorkels' To Survive Floods."
http://www.popsci.com/scitech/article/2009-08/snorkel-genes-allow-rice-plants-survive-floods
http://esciencenews.com/sources/popsci/2009/08/19/genetically.engineered.rice.plants.grow.snorkels.to.survive.floods

Jon Hamilton, a correspondent for National Public Radio's science desk, had his report on "snorkel rice" billed as: "Scientists are racing to genetically engineer strains of rice that can prevent the deadly famines that come with drought and floods in Asia. One new strain causes part of the plant to elongate, acting as a snorkel."
http://www.npr.org/templates/rss/podlayer.php?id=2100615
http://www.wbur.org/news/npr/people/2100615

Such reports may have been behind comments claiming that "snorkel rice" provides, "A perfect example of the benefits of genetic modification of staple crops and why we need to fund research on them."
http://www.reddit.com/r/science/comments/9cgv4/genetically_engineered_rice_grows_snorkel_during/

And blog comments such as:

"But it's genetically modified food. We can't have that! Who cares if it feeds the world, it's still not organic and we all know organic foods keep us from getting sick. Or do we.."
http://steve-lewis.blogspot.com/2009/08/snorkel-rice-could-feed-millions.html

As Dr Antoniou's comments make clear, "snorkel rice" is very far from being a "perfect example of the benefits of genetic modification". In fact, the paper leads to the conclusion that non-GM biotechnological approaches, like marker assisted selection (MAS), can successfully deliver new crops with complex traits.

Of course, "snorkel rice" is not the first flood-resistant rice to be passed off as a GM rice when it's not. The UK's former Chief Scientist, Prof Sir David King, even pulled this trick in a BBC radio programme. There is a submergence tolerant (flood resistant) rice on the market but it is non-GM. The researchers, in fact, tried to use a GM approach to develop it and failed.
http://www.guardian.co.uk/commentisfree/2008/dec/09/david-king-gm-crops
–-
–-
COMMENT by Dr Michael Antoniou, Head of the Nuclear Biology Group at King's College London

The article by Hattori Y et al. in Nature (volume 460, pages 1026-1031) describes the mapping of the "Snorkel" genes (designated  SK1 and SK2) that stimulate a high rate and length of stem growth in rice under flood conditions. A DNA marker statistical analytical (quantitative trait loci – QTL) approach was used initially to roughly identify the location within the plant total DNA (genome) where the flood response genes were located. A total of 3 QTL regions were identified on three different chromosomes (numbers 1, 3 and 12) with one of them (on chromosome 12) being particularly important (QTL12). QTL12 was followed up by a direct gene mapping (positional cloning) approach to specifically identify genes that may be involved in deepwater response residing in this region of the genome. This identified two candidate genes (subsequently named SK1 and SK2). The SK1 and SK2 genes were then confirmed as having a functional role in the deepwater flood response by transgenic (GM) techniques in a "gain-of-function" analysis; i.e., add the genes to a rice variety which lacks them (T65) and see if it now acquires a flood-induced growth response; it worked!

I mention all this because it is important when it comes to understanding the transfer of the genes to a high yielding variety.

On page 1028 of the article in Nature the authors state:

"Deepwater rice is mainly cultivated in lowland areas, and it is the only crop that can be grown in flooded areas in the rainy season. One problem with the cultivation of deepwater rice is its low yield (2). Deepwater rice bred for these areas must possess not only internode elongation ability under deepwater conditions, but must also produce higher yields. To date, we have identified three QTLs on chromosomes 1, 3 and 12 that regulate the deepwater response (9,10). To evaluate the deepwater responses of these QTLs, NILs and QTL pyramiding lines (21,22) were produced. Four QTL pyramiding lines were produced, possessing two QTLs on chromosomes 1 and 3 (NIL-113), 3 and 12 (NIL-3112), 1 and 12 (NIL-1112), and three QTLs on chromosomes 1, 3 and 12 (NIL-113112) in a T65 genetic background (Fig. 4c, d). In particular, the QTL pyramiding line NIL-113112 showed almost the same deepwater response as C9285. These results indicate that introducing these three QTLs into non-deepwater rice enables it to behave like deepwater rice. Therefore, the QTL pyramiding strategy is a powerful and efficient tool for breeding rice varieties for flood-prone areas. Recently, genes that regulate yield have been identified (21,23-26). The combination of deepwater and yield QTLs has the potential to help breeding in these areas."

What does this all mean you may ask? Basically the researchers CROSSED the deepwater, flood responding but low yielding rice (C9285) with the non-deepwater responding T65 rice variety. Using the fact that they had identified the DNA regions that regulate deepwater response as QTL markers, they were able to use an MAS strategy to identify progeny from this cross which contained various combinations of the three QTL markers within an otherwise T65 variety genetic background. They found that the T65 cross that contained all 3 QTL markers showed the same flood response to that of the parental C9285 type.

So, the message is that although various non-GM and GM biotechnological methods were used to identify the genomic regions and two of the genes responsible for the flood response, it is MAS [non-gM Marker Assisted Selection] that was used to exemplify the transfer of the flood response gene loci to an otherwise non-responsive line of rice. It is interesting that the authors in their statement above seem to imply that their cross-breeding (QTL pyrimiding) strategy is the way forward to exploit their findings and do NOT mention a transgenic approach, probably because what is involved is so complex that GM could not deliver what is required. We have to remember that THREE QTL regions were identified as being linked with the flood response. So the QTL12-linked SK1 and SK2 are only part of the story albeit a very important one. So I presume the authors chose a QTL pyramiding cross-breeding plus MAS strategy to make sure they got all three QTL regions in place in the new variety; something totally beyond what GM can even dream about doing!

This is another example of how an MAS and NOT a GM transgenic approach can successfully deliver new crops with complex traits.