Executive Summary

One of the most promising applications of agricultural biotechnology is the large-scale production of recombinant proteins in transgenic crop plants for medical and industrial uses, or "molecular farming". Two issues that limit the use of maize in molecular farming are the relatively low protein concentration of the grain and concerns about "transgene contamination" of commodity corn. This project addresses each of these issues through the development of maize genotypes that are optimized for recombinant protein production. Existing transgenic maize lines expressing the benign GUS protein under the control of a variety of seed-specific promoters have been crossed to Illinois High Protein (IHP), a genotype with three times the grain protein concentration of conventional maize. These populations have been subjected to breeding and selection for maximal GUS yields and concentration in maize seeds. The use of IHP in combination with the classic maize zein reduction mutant opaque2 (o2) shows promise as a genotype to increase yields of recombinant proteins. In addition, two biological approaches to transgene containment, male sterility of transgenic lines and visually obvious seed anthocyanin pigmentation, are being incorporated into IHP to minimize risks associated with transgene contamination of commodity corn. The genotypes generated will serve as a model test system for initial "proof-of-concept" evaluations that can then be extended to the production of other high-value recombinant proteins.

Research Activities and Progress

This project has four main objectives. Activities and progress related to each of these objectives are detailed in the following sections.

1. Introgression of promoter-GUS transgenes into IHP1 and selection for maximal GUS expression. Two events for each of the Z22pro-GUS, Z19pro-GUS, Z27pro-GUS and Ubi1-GUS transgenes have been backcrossed four times (the BC3 generation) to either the IHP1 or B73 inbred lines. We initially proposed to use Western blot analysis to quantify GUS protein accumulation, but found that it was possible to make selections based on visual observation of the blue histochemical staining for GUS activity. This method is more rapid and also has the advantage of selection being based on the amount of active protein accumulated, rather than total protein that may not be in an active form.

Figure 1 shows the results from GUS staining representative kernels of the different transgene constructs in the B73 and IHP1 backgrounds. On the left side are the Z27pro-GUS and Ubi-GUS constructs that show strong expression in both B73 and IHP1. The Z27 promoter is also endosperm-specific, whereas Ubi1 is active throughout the kernel. These expression patterns are consistent with expectations from other characterizations of Z27 and Ubi1 expression. The Z22pro-GUS and Z19pro-GUS transgenes exhibit stronger expression in IHP1 compared to B73, which is similar to the expression behavior of the endogenous 22-kDa and 19-kDa a-zein genes. The activities of the Z22 and Z19 promoters show significant overlap and are expressed more strongly in the periphery of the endosperm.

The stained kernels shown in Figure 1 are from the ears that show the highest GUS expression levels. Generally, the spatial distribution of GUS activity is similar among different ears harboring the same transgene, but the intensity of expression varies from weak to strong. Expression levels among individuals with the same transgene are more uniform in the BC3 kernels compared to earlier generations, suggesting that expression variation is most likely due to genetic segregation for key factors that regulate promoter activities.

2. Test whether GUS yields are enhanced by the o2 mutation. The initial crosses of the Ubi1-GUS and Z27pro-GUS transgenes to IHP1: o2 were made in the summer 2005 nursery, using the BC2 plants in the IHP1 background as donor parents. A second backcross to IHP1:o2 will be made in the greenhouse this winter to produce ears segregating for o2 and the transgenes in a predominantly IHP1 background. Opaque2 mutant and wild-type seeds from these ears will be grown in the 2006 summer nursery, ears self-pollinated, and GUS levels measured in the seeds obtained.

A second activity related to this objective was the field evaluation of B73: o2 X IHP1: o2 hybrids, to serve as a benchmark for what might be expected for GUS-expressing hybrids. The B73 x IHP1 o2 hybrid was evaluated along with the B73 x IHP1, B73 x Mo17, and B73 x Mo17 o2 hybrids for comparison. Each hybrid was grown in two replicate plots at two rates of supplemental nitrogen: 0 (N deficient) and 180 pounds per acre (N sufficient). Ears were harvested from the plots, shelled grain was weighed to estimate yield, and a sample of grain was analyzed by near-infrared reflectance to obtain concentrations of starch, oil, and protein. Protein yields on a land area basis were also calculated. Table 1 shows the results.

Hybrid	Grain Yield (bu./acre)	Protein Yield (bu/acre)	Starch %	Protein %	Oil %
B73 x Mo17 - 0 N	148	9.3	74.5	6.3	3.5
B73 x Mo17 - 180 N	202	15.8	72.9	7.8	3.5
B73: o2 x Mo17: o2 - 0 N	98	5.6	75.6	5.7	4.3
B73: o2 x Mo17: o2 - 180 N	135	12.3	72.0	9.1	3.8
B73 x IHP1 - 0 N	50	4.2	70.4	8.3	4.8
B73 x IHP1 - 180 N	131	16.5	65.5	12.6	5.2
B73: o2 x IHP1: o2 - 0 N	41	3.8	70.2	9.3	4.5
B73: o2 x IHP1: o2 -180 N	110	14.2	66.7	12.9	4.4

Table 1: Yield and grain composition estimates for evaluation of enhanced transgene protein production. Values represent the mean from ten ears, five from each replicate plot.

It is clearly evident in all genotypes that protein concentration and protein yields increase dramatically (2-4 fold) when provided a fertilizer rate typically used in current corn production. Thus, nitrogen fertilization will likely remain important for maximizing yields of recombinant proteins. Each of these hybrids shared B73 as a female parent, but using IHP1 as a male parent increased protein concentration from 2-5% compared to hybrids with Mo17 as male parent. Grain yields were significantly less for the B73 x IHP1 compared to B73 x Mo17 hybrids at both N rates; however, the increased protein concentration of IHP1 hybrids led to greater protein yields at the high N rate. The o2 mutation reduced protein yields of each of the hybrids, having a greater negative effect on grain yield compared to protein concentration. Interestingly, the B73:o2 x IHP1:o2 hybrid maintained similar protein concentrations as its B73 x IHP1 counterpart. Because the proportion of zein protein is expected to decrease in o2 mutant hybrids, it appears that other proteins compensate for the reduction in zeins. This result suggests that even though overall protein yields may be slightly lower in the B73: o2 x IHP1: o2 hybrid compared to B73 x IHP1 or even B73 x Mo17, the potential yield of a recombinant non-zein protein may indeed still be the greatest in the B73: o2 x IHP1: o2 background. The higher concentration of protein in B73: o2 x IHP1: o2 hybrid grain, coupled with its softer endosperm texture, may facilitate recovery of extracted recombinant proteins. Furthermore, the higher concentration of oil in the IHP1 hybrids may add value to processing by-products.

3. Develop a male-sterile and r-navajo version of IHP1 as a transgene containment system. The r-nj cms-S stock has been backcrossed three times to IHP1 (summer 2004, winter 2005, summer 2005). The unexpected presence of the Rf3 fertility restorer gene in IHP1 has complicated the recovery of backcross plants with the desired cms-S male sterile phenotypes, but three different ears were successfully obtained from male sterile plants. The r-nj phenotype has proven to expressed in a dominant manner with an interesting pattern in IHP1, where the anthocyanin pigmentation is localized to a vary small area surrounding the silk attachment region (Figure 2). This subtle, yet robust phenotype indicates r-nj will indeed be a useful marker gene that should gain rapid acceptance once implemented.Figure 2: IHP1: r-nj kernels.

4. Make all data publicly accessible through publications, presentations, and a project website. The results from this project will be posted at the Moose laboratory website (http://cropsci.uiuc.edu/faculty/moose/index.cfm) after the completion of the 2006 field studies.

Work Planned

The backcrossing of transgenes will continue to the BC4 generation in the greenhouse this winter. BC3 lines will also be crossed to B73 and B73:o2 to produce hybrid seed for field evaluation during the 2006 summer nursery. The onset, rate, and duration of promoter activities in the different transgene and background genotype combinations will be assessed by staining for GUS in developing seeds. Genetic crosses followed by measurement of GUS activity will be performed to assess whether the maternal control of expression observed for the endongenous zein genes also occurs with the zein promoter-GUS transgenes.

Transgenes introgressed into IHP1 will be crossed to the male-sterile, r-nj IHP line in the winter greenhouse to produce progeny segregating for the male-sterile phenotype, r-nj, and the transgenes. Large families will be grown in the 2006 summer nursery and plants selected on the basis of male sterile phenotype, GUS content in mature seeds, and r-nj phenotype.

The field evaluation of the hybrids listed in Table 1 will be repeated in 2006.

Publications

The results of this work have not yet been published. I anticipate the submission of at least two peer-reviewed journal articles after the completion of the 2006 field experiments. One publication will describe the spatial-temporal patterns of promoter activities in the different genetic backgrounds. A second will report the results from the field evaluation of IHP1: o2 hybrids. Should we be able to show that GUS recovery is significantly increased in the IHP1 or IHP1:o2 hybrids compared to normal hybrids, I would consider filing a patent application for this technology.

Other Significant Impacts

This project has been one component of the thesis research for Elizabeth Wrage, who intends to obtain her M.S. degree in December, 2005. The project has provided Elizabeth with practical training in the breeding of maize transgenic lines and the handling of regulated transgenic field experiments. These skills make Elizabeth an attractive employee to the private agricultural biotechnology sector.