By David N. Leff

Mammals aren¿t the only forms of life on earth that contract cancer. Plants, too, can acquire botanical tumors, called crown galls.

¿Crown gall,¿ observed plant molecular biologist Abhaya Dandekar, ¿is an important disease that affects a lot of woody plants ¿ ornamentals, fruit trees, nut trees, that kind of thing. A big round burl at the base of trees, crown gall wood is prized to make veneer or smokers¿ pipes. When a crown gall grows on these plants, they go into a decline, and are no longer productive. This causes a tremendous amount of economic damage to perennial crops worldwide.¿

Dandekar, a professor of pomology (fruit culture) at the University of California at Davis, continued: ¿This crown gall blight is one of the reasons people started studying Agrobacterium tumefaciens, the soil bacterium that causes the disease, back in the 30s and 40s. They were trying to find a solution to this plant tumor infection. However, A. tumefaciens turned out to be much more interesting as a genetic engineer, because it transmits its DNA ¿ and foreign genes ¿ into plants. This discovery jump-started the entire agricultural biotechnology enterprise.¿

The Proceedings of the National Academy of Sciences (PNAS) dated Nov. 6, 2001, but electronically released Oct. 30, carries an article titled: ¿RNAi-mediated oncogene silencing confers resistance to crown gall tumorigenesis.¿ Dandekar is its senior author.

¿It had dawned on us that we might target the bacterial genes involved in the tumor formation,¿ he told BioWorld Today. ¿There are only three genes in this bacterium that generate crown galls. Two of those genes make auxin, a hormone involved in root forming. The third gene expresses cytokinin, the growth-promoting hormone that forms the plant¿s shoots. And together,¿ he pointed out, ¿when you have a suite of oncogenes expressing a high level of auxin and cytokinin, you get uncontrolled growth that results in this tumor.

¿We found,¿ Dandekar recounted, ¿that these genes were over 90 percent conserved among a wide variety of Agrobacteria. So it occurred to us that we use gene silencing as a strategy to shut them off. But we never anticipated that it would work so well.¿

A. tumefaciens Seeks Cut Or Abrasion

Besides the plant pathogen itself, it takes one other factor ¿ a wound ¿ to gain the bacterial DNA entry into the plant. ¿Typically, what happens,¿ Dandekar said, ¿is that in the field people are hoeing or removing weeds, and accidentally cut the plant. Also in the nursery, where there¿s a lot of cutting and chopping, workers have to be careful how they clean their tools to prevent crown gall.

¿T-DNA is a piece of DNA that is transferred from the bacteria into the plant genome,¿ he continued. ¿That¿s the unique thing about Agrobacterium infections. Unlike any of the usual plant pathogens that will invade the plant, this one sends its DNA inside to be expressed in the plant cell, which results in the tumor formation. In the paper, we show that we can transform the plants by inserting a different piece of bacterial DNA, and this particular sequence containing the same hormone genes. So we are not preventing Agrobacterium from infecting, just preventing it from forming the tumor.¿

Dandekar explained: ¿In other words, if you would target these bacteria-directed hormone genes, you would not target the endogenous plant genes, because auxin and cytokinin are very important in plants. However, the genes we engineered are different. We designed constructs that express a self-complementary RNA, as we reported. And we found that these double-stranded RNAs really worked very efficiently in silencing genes.

¿This is like the FlavrSavr tomato that came out many years ago,¿ he recalled. ¿That was an antisense transformation that delayed the fruit¿s softening by keeping ethylene from premature synthesis. Now it¿s becoming clear that all these mechanisms are basically the same. Antisense works the way sense works. Many times if you express the sense message it will shut down the endogenous expression. And the reason they do so is that they are triggering gene silencing. However, they are not as efficient as if you use self-complementary methods. That means we are expressing both the sense and the antisense as part of the same message. So it sort of slips over on itself and makes double-stranded RNA.

¿This double-stranded RNA is the key intermediate that triggers gene silencing. What we¿ve done is make these self-complementary messages that encode a significant portion of these two messenger RNAs that encode those plant hormones. So the i¿ in the RNAi of our paper¿s title stands for interference¿ ¿ silencing.¿

In one in vivo experiment, Dandekar related, ¿We made transgenic tomato plants ¿ Lycopersicon esculentum ¿ that expressed genes introduced into these plants by Agrobacterium tumefaciens. They were expressing the self-complementary genes and making double-stranded RNA to those genes. This we did by plant transformation, using the bacterial vector. When the hormone genes had been removed, we put in a T-DNA encoding the various genes that we wanted to put in. Then we extracted tomato plant cells and generated transgenic tomato plants that now expressed these double-stranded RNAs in the whole plant. In those tomatoes, less than 24 percent grew tumors, in comparison to 100 percent of the controls.¿

Next: From Fruit To Nuts

¿What we¿re doing now,¿ he observed, ¿is trying to have this succeed in a fruit or nut tree. We are working on walnuts, in which crown gall is a severe problem. We hope to engineer a rootstock to show that this can be done in walnuts. Initially, we will do a lot of experiments in the lab, but then take them into the field and do outdoor experiments. In California alone,¿ Dandekar pointed out, ¿walnuts represent a $250 million industry, or thereabouts. So you can see the effect of crown gall. If at some point you lose productivity, you lose the whole orchard, because you have to maintain productivity to a certain level to be economically viable.

¿I think gene silencing is a very important area,¿ he summed up. ¿In human biology, it¿s already being used a lot to study gene function and gene discovery, because they can create a functional knockout.¿ Dandekar concluded: ¿We are perhaps the pioneer gene silencers in agriculture.¿