DroughtGard™ maize will be the first commercially available transgenic (GM) drought tolerant crop if it’s released in 2013 as planned. Hybrid seed sold under this trademark will combine a novel transgenic trait (based on the bacterial cspB gene) with the best of Monsanto’s conventional breeding program.
The Union of Concerned Scientists threw their usual wet blanket on the development. I think their gloomy assessment of transgenic drought tolerance is pretty biased but they do make a few good points that the public should understand. Primarily, “drought tolerance” does not mean that these plants can be grown with little to no water.
A typical maize crop burns through over half a million gallons of water on the way to harvest – and the physics of turning sunshine, minerals and water into over a hundred bushels per acre of grain doesn’t leave much room for improvement. To paraphrase a scientific presentation I once saw, modern maize varieties are like high performance race cars. You could re-engineer your race car to get 50 mpg, but it won’t be very fast anymore and you won’t make any money. All the same, getting more grain per input water (generally referred to as water use efficiency) remains of major interest.
Drought tolerance is different than water use efficiency and can operate in complex and unpredictable ways (as plant physiologists have been shouting for years). For example, it’s a glaringly common mistake in the literature to announce some new “drought resistance” gene that allows Arabidopsis to survive longer without water – only to reveal that the “resistant” plants grow slower and therefore simply take longer to drink up their little ration of soil water. This way of measuring drought is of questionable usefulness because helping a drought-ravaged crop to survive a few days longer is commercially worthless. As you’ve likely seen in the news this year, a maize crop that has been subjected to serious drought has absolutely no chance of producing enough grain to justify harvesting – either in the costs of harvest (e.g. diesel) or the value of crop insurance. Reducing growth rates generally also results in lower biomass and grain yield under good times, which farmers have zero tolerance for. Alternatively, reducing yield losses under minor to moderate drought (while delivering maximum yield under well-watered conditions) is a very valuable trait in the North American maize seed market. While this modest goal is nonetheless important (and very difficult to achieve!), the public should understand what ‘drought tolerance’ really means and hold their expectations accordingly.
DroughtGard maize contains the gene for “cold shock protein B” (cspB) from Bacillis subtilis. Cold shock proteins were discovered (and named) due to their rapid accumulation in cold shocked bacterial cells. Some CSPs such as CSPB act as RNA chaperones, which help to maintain normal physiological performance during stress events by binding and unfolding tangled RNA molecules so that they can function normally. Castiglioni et al., 2008 describes Monsanto’s initial research demonstrating cold resistance of Arabidopsis plants transformed with cspA and cspB, followed by cold, heat and drought resistance of similarly transformed rice plants. It might seem odd to spend time testing genes in Arabidopsis and rice prior to getting around to testing in the crop of interest but this is the appeal of model organisms – genes can be screened much more efficiently in small, quick and easy growing model organisms compared to often finicky and cumbersome crop species. This is critical as companies may need to consider hundreds of thousands of transgenic constructs for each one that becomes a successful product.
Following these initial experiments, maize was transformed with cspB and was assessed in a series of field experiments of increasing complexity and realism. First, 22 independent cspB transgenic events (i.e. independent transfers of the DNA construct into 22 maize plants) were crossed into commercial grade maize. These were tested in a preliminary field trial where they were exposed to drought stress during the two week period immediately preceding anthesis (flowering), when cereals are most susceptible to yield losses. Drought-induced biomass losses that occur early in the season and grain filling losses that occur after flowering can be partially compensated for if sufficient rains return. Drought stress that occurs adjacent to anthesis, however, leads to the abortion of individual grains, with an irreversible impact on yield. This first field experiment was designed so that non-transgenic plants subjected to drought suffered from a 50% reduction in growth rate relative to the well-watered control. Within this scheme, cspB transgenic maize lines exhibited up to a 24% increase in growth rate under drought conditions (accompanied by improved chlorophyll content and photosynthetic rates), while performing normally under well-watered conditions. Grain measurements demonstrated a 4% increase in the number of plants with kernel-bearing ears and a 11.7% increase in the number of kernels per plant for cspB transgenic lines under drought. Thus, the cspB gene appeared capable of minimizing kernel abortion, an irreversible (and therefore very important) component of yield loss under drought.
Larger grain yield trials were then performed on 10 cspA and 10 cspB lines at four locations. Again, the drought treatment was applied immediately before flowering, at a severity that produced a 50% yield loss for non-transgenic genotypes. Here, cspA lines showed an average yield increase of 4.6% while cspB lines had an average yield increase of 7.5%. The best performing lines of cspA and cspB showed increases of 30.8% and 20.4%, respectively. The best two cspB lines (CspB-Zm event 1 and 2) also showed significant gains in leaf growth, chlorophyll content and photosynthetic rates.
The most promising line, CspB-Zm event 1, was then assessed in several years of field trials. The transgenic locus was crossed into three different hybrid genotypes, which were tested under three controlled watering conditions (well-watered, drought immediately preceding flowering, drought during grain fill) at five replicated locations. Non-transgenic controls suffered 50% or 30-40% yield losses under the two drought stresses, respectively, while cspB lines produced 11-21% relative yield gains. The yield stability of CspB-Zm event 1 was also analyzed over four years of testing in three hybrid test-crosses, where cspB provided an average yield benefit of 10.5% (with a range of 6.7-13.4%). CspB lines were also tested under dryland (non-irrigated) conditions with a field design that approximated normal commercial planting densities (rather than a more open field plot design). Of the sites that happened to experience seasonal drought during the experimental timeframe, the maximum yield gain of the transgenic lines was 15%.
Monsanto conducted much more field testing than was described in this paper. Typically, genes are discovered in a model organism prior to testing in a pilot field trial under controlled conditions. Promising lines are ramped up into multi-location/year field trials of increasing size and agronomic realism until they’re eventually tested in preliminary farmer trials and prepared for commercial release. In this promotional presentation, Monsanto illustrates their network of 200 testing locations across different climatic regions of the United States. This entire process (for one transgenic variety) takes about 12 years.
In addition to field testing for agronomic performance, Monsanto had to generate huge amounts of data in order to petition for deregulation to allow their novel transgenic product to be grown in the U.S. and imported to foreign countries. If you’d like to get into the weeds on this, you can climb into this 500+ page Petition for the Determination of Non-Regulated Status for MON 87460). I looked through it to see some additional yield data and it appears consistent with what they stated in Castiglioni 2008.
According to the Plant Pest Risk Assessment for MON 87460 Corn, the final variety contains a single insertion of the following DNA sequences:
- Part of the TDNA border
- used to insert genes into plants with A. tumefaciens
- Promoter, leader and intron from rice actin gene, Ract1
- appears to drive constitutive (always on) expression of the cspB gene
- Full length coding sequence of cold shock protein B (cspB) from Bacillus subtilis
- functional cspB RNA chaperone protein
- 3′ untranslated terminator sequence of the transcript 7 gene from A. tumefaciens
- a short length of noncoding DNA to separate the two genes
- loxP sequence from Bacteriophage P1 for Cre-lox recombination
- potentially allows the whole nptII cassette to be specifically snipped out of the plant once its no longer needed, though they appear to have left it in
- P-35S promoter for the 35S RNA of Cauliflower mosaic virus
- most famous promoter in plant molecular biology, drives constitutive expression of nptII
- CS-nptII coding sequences from Tn5 from E. coli
- protein that confers neomycin and kanamycin resistance, which allows original plantlets that have been successfully transformed to be selected for by applying antibiotics that kill any non-transgenic plants. Once plants with functional gene insertions are identified, antibiotic resistance is no longer needed and can be removed.
- T-nos 3′ untranslated sequence from nopaline synthase (NOS) gene from A. tumefaciens
- terminator sequence for nptII gene
- loxP sequence from Bacteriophage P1 for Cre-lox recombination
- Part of the TDNA border
From what I’ve heard, the yield gain of this variety under drought occurs because it slows its growth specifically under drought stress such that existing soil moisture is saved for the critical period surrounding flowering, resulting in less kernel abortion, higher harvest index and greater yield. This is certainly ironic given the expectations of many field physiologists and breeders. It also occurs to me that this is consistent with the thesis of Denison’s Darwinian Agriculture – that evolution has already maxed out our crops’ ability to deal with most stresses and environments, and that the greatest potential for improvement exist in traits that only work in an ecosystem (such as a farm field), where plants aren’t in competition with their neighbors.
DroughtGard will begin its limited commercial roll-out in the Western Great Plains (from South Dakota down to Texas). This is the initial target environment for many drought tolerant maize products as it’s far enough west to experience regular drought, but not so far that irrigation is common. It’s convenient for Monsanto that their new variety will be ready for its first season following a major drought. Farmers are generally most willing to invest in premium traits like drought tolerance when memory of major yield losses are fresh on their minds.
Of course the real test will be in farmers’ fields over the next couple of years. It will be interesting to see how well this variety sells and performs.
Castiglioni, P., Warner, D., Bensen, R.J., Anstrom, D.C., Harrison, J., Stoecker, M., Abad, M., Kumar, G., Salvador, S., D’Ordine, R. & (2008). Bacterial RNA Chaperones Confer Abiotic Stress Tolerance in Plants and Improved Grain Yield in Maize under Water-Limited Conditions, Plant Physiology, 147 (2) 455. DOI: 10.1104/pp.108.118828