Quote:
|
the main difference here is that in selective breeding programs, nature has a multitude of safety levers to ensure potentially destructive genes are not passed from generation to generation.
|
Well this is a bit misleading. . . . If we're anthropomorphizing a plant (implying here that it doesn't want to pass on potentially destructive genes) I think we should at least limit that sentiment to the plant's interests. The plant doesn't "care" if it contains a gene that might harm somebody that's going to eat it. Or, to put it more officially, there is no selective advantage for a plant to evolve mating avoidance with another individual containing genes that are detrimental to potential herbivores. There certainly is selection to avoid mating with individuals containing genes that might reduce the fitness of that particular plant, yes. How effective that selection is in practice (ie "reinforcement") is one of the big disputes right now in plant evolutionary biology. But there is nothing in nature that "tries" to prevent dangerous genes from being passed on, until those genes actually exert an effect in an individual that reduces its personal fitness. If harming a herbivore doesn't reduce a plant's fitness, then there's nothing to prevent that plant's descendents from obtaining such a gene from a mate.
Quote:
|
You can mate carrots to varieties of carrots, but you can't mate tomatoes to carrots. And if you were successful, as some breeding programs have done with some animals, they aren't able to produce viable offspring.
|
Actually plants hybridize all over the place. About half of all plant species evolved through hybridization followed by polyploidy. In fact if you draw an evolutionary "tree" of most plant groups, it looks more like a net than a tree. Many plant genera are cross compatible.
But nevertheless, the fact that there are genetic barriers to hybridization is really not pertinent to the question of whether new anti-herbivore or human toxicity can evolve in a plant species "naturally"; such barriers never evolved to prevent evolution of anti-mammal toxicity. And further, there is tremendous variation within widespread species in secondary chemical defenses; any particular plant could gain many new toxicities by mating widely enough even staying within its taxonomic species.
The popular view seems to be: the more genetically distant two parents are, the more "dangerous" it is (to humans) to mix their genes. I really know of no biological evidence that could be used to support this view. I don't see any reason to believe that a natural cross of two individuals within a species is any less likely to produce a new kind of toxicity to humans than inserting a single gene from another species, no matter how distant. In fact, the natural cross is far more likely to produce an individual radically different from its parents (since 10s of thousands of genes are being mixed in random combinations) than the transfer of a single gene.
I think there is a lot of fear of the unknown packaged into this popular view.
I also say, full speed ahead on all these technologies.