Two pieces turned up in the British journal Nature this week that shed a little more light on the potential value of genetic engineering in addressing the dangers cimate change is going to impose on agriculture. (I've been pretty vigorous on this site proclaiming that potential value, despite the clearly pernicious effects of Roundup-ready crops, the likely pernicious effects of Bt maize, and Monsanto's ugly business models.)
One is a news item that notes methods and advances in creating drought-resistance maize for Africa. Genetic engineering has helped a little, but so far conventional breeding has made a much more substantial contribution. The other - an online paper highlighted in one of this week's editorials - is more bullish, indicating a path forward for engineering into food crops the much more efficient photosynthesis carried out by cyanobacteria and by many weeds.
Below the fold are the links - behind a firewall unless you or your university subscribe to Nature - and my attempts at a summary.
African maize. Maize is particularly susceptible to drought, because under dry conditions, the female organs, the silks, develop late. If that stretches too far, the wind-borne pollen has already been carried away, and no kernels grow. So in addition to general drought-hardy characteristics, researchers in this field have been looking for strains with shorter periods between the development of male and female organs.
Work has relied on seed banks provided by Mexico's International Maize and Wheat Improvement Center (CIMMYT). The Drought Tolerant Maize for Africa Project, running from 2006 through 2016, has concentrated on breeding 153 drought tolerant lines from that bank, crossing them in the final generation with varieties that have grown particularly well in Africa. Many of these are already in the field, performing well, and are expected to reduce poverty on the continent by up to 9% by the end of 2014. Meanwhile, under the aegis of the African Agricultural Technology Foundation in Nairobi (alas, with Monsanto among its half dozen collaborators), transgenic drought-tolerant maize is expected to make its debut no earlier than the end of 2016.
Why is conventional artificial selection outperforming genetic engineering? Because drought tolerance is a complex characteristic, involving an unknown number of mostly unidentified genes. GE technology is best suited to manipulation of a small number of known genes. Another powerful desideratum for African maize cultivation, plants that thrive on lower levels of nitrogen and spare farmers some of the heavy expense of fertilizer, is similarly influenced by a plethora of distinct genes, and so will give natural breeding the same kind of edge. Given the breadth of the challenge, I submit we can't afford not to explore all of the tools in our toolbox. But it's clear the tried and true natural methods will be the stars for the next decade or two.
Better photosynthesis. By way of contrast, a single enzyme, Rubisco, drives photosynthesis. So this is the kind of process that genetic engineering should be able to fiddle with successfully. And there's a powerful motivation for fiddling.
There are two different photosynthetic systems in plants. Each culminates with its own variation of Rubisco, but "C4" plants start with an extra step that concentrates the CO2 before it's used. "C3" plants save energy by skipping the extra step - but at a price. Some of the C4 Rubisco fixes oxygen rather than CO2, reducing the efficiency of the process. Under hot, dry conditions, that inefficiency skyrockets - and the globe is now prepped to see a lot more of hot, dry conditions. Most food crops (corn and sugarcane being notable exceptions) are C3 plants, and it will be mightily handy in decades to come if they were available in C4 versions.
In their online paper, Lin et al. describe the successful replacement in tobacco of several subunits of Rubisco with the C4 version taken from cyanobacteria. This is a very preliminary result. As expected, the modified enzyme processed CO2 much more efficiently. But C3 plants produce a lot more Rubisco than C4 plants do, and the engineered tobacco produced only 12% as much of the stuff; so its growth rates were actually lower. It will take more work to replace the remaining wild-type subunits of the enzyme. Also, C4 plants carry out their preliminary CO2-concentrating step in structures called carboxysomes; it wasn't clear to me reading the paper whether those will have to be constructed somehow to gain the full benefit from the imported Rubisco enzyme. Still, I found it an intriguing harbinger of things to come.