There are three general types of Genetic Modification as I define them, and will be discussing them as follows:
Artificial Selection (can be facilitated by radiation or mutagens.)
Direct Genetic modification (Via gene insertions, deletions, and controlled mutations)
Now, I’m not going it get down into the nitty gritty of how to do all these things. I’m just going to talk generally about how they work and what they mean for the genome of the resulting organism.
One important idea we need to understand about genetic modification (GM) is that GM is a tool. Genetically modified organisms are the final results. It’s very important we do not confuse the tool with the product: just like a hammer and saw is not a cabinet, GM is not a GMO. And like any tool it has it’s positives and negatives, and can be misused, but the process of modern Direct GM is difficult and time consuming and require a lot of resources, expertise, and knowledge before you can commence with creating a GMO that will do anything at all, let alone something useful. Hybridization and artificial selection (selective breeding) are both in turn much simpler and much more time consuming. Modern wheat, before any genetic tinkering in the lab, has had thousands of years of GM done to it already via selective breeding. The modern wheat in our bread does not closely resemble the wild wheat cultivated by early farmers many thousands of years ago. Though wheat isn’t necessarily the best example…
…Since wheat is a hybrid of two different grains that, from what I can tell, don’t exist anymore, but we’ll get back to hybrids soon enough.
Artificial selection. This is maybe how we did it, and when I say that I mean artificial selection could be thought as one of the keys to human success. This is ultimately how we made agriculture work for us for the last 10’000 years. Artificial selection alone is responsible for historic agriculture and is still relevant today. So what is artificial selection? Well it harnessing evolution by natural selection and bending that selection to the whims of humanity.
For those of you who are not familiar with the scientific theory of evolution via natural selection (there are other selection pressures, but you don’t need to know them to understand this), well first that’s a minor tragedy, but no worries, there is no shortage of good information out there for you, here’s a bare bones basic run down. (Disclaimer this is by no means complete, but it enough for this post)
- You have population of living things.
- There is an amount of generic variation in the population.
- This variation lead to different morphological, and or behavioral traits.
- Members of this population can reproduce with one another (they are the same species)
- Some traits give an advantage or disadvantage to the individuals that possess them. (if no traits give those advantages or disadvantages then natural selection does not occur)
- Over time individuals with advantageous traits will produce more offspring while individuals with less advantageous traits produce less offspring (offspring is best thought of in terms of how many grand children). Over many hundreds or thousands of generations this will gradually change the morphology/behaviors of the entire population. Some times to an extent that it no longer resembles it’s ancient ancestors.
So how does this relate back to artificial selection? Well instead of an unaided gradual change as some individuals do slightly better than others, in artificial selection the breed can exert much greater control. For example if you have individual with undesirable traits, like wheat with 6 foot tall stalks (as a historical example) then you only breed those wheat plants which have shorter stalks, and do not save the grain (sell it or what not) of the longer stemmed wheat plants. This ability to quickly remove undesirable traits speeds the process of changes from thousand or millions of years to decades or centuries.
Now that said, you only have so much control with artificial selection, prior to knowledge of genes let alone gene sequencing, you could only select individuals based on detectable traits. Like hair length, colour, and thickness. And if an undesirable trait is a repressive gene you can’t actually remove it 100% from the breeding population with out blind luck, and heterozygous individuals (individuals with an allele for both the recessive and dominate trait) will often look/behave just like homozygous dominate (having both alleles for the dominate trait) individuals.
Also when you using artificial selection you can’t simplely choose one trait. I apologize it I didn’t make this clear before. When your breeding an organism your getting the whole genome, so you can’t just choose one trait and cross only that trait. you have to cross all the traits in the organism and hope you get a good combination. If you do this over and over eventually you’ll eventually get many, though not necessarily all, the traits you where after.
Now allow me to give another example. Say you have two breeds of apple tree one is a common baking apple and another is a half wild crab apple which produces a compound which you find gives the apple a delicious after taste, though the whole apple is not very tasty and is very small. So you try breeding the two lines together, and low and behold the resulting apples have none of the traits you where after, they are all small and sour with out the nice after taste you where after. However you expected this to happen so you cross that second generation with itself, and this time you get a more interesting mix, some trees have big apples and a sour taste, and some are small and sweet, and some have that after taste. You select those trees with traits closest you what you want and cross them, and do this six or so more times. Then you die. Because remember apple trees take a long time to grow and you got old, but don’t fret! Your grand child takes up your apple tree hobby and does a few more crosses and finally the Holy Grail of apples is born! Big, Tasty, and with your magic after taste! To bad you didn’t just isolate the trait and insert it into baking apples genome. Then you might of had the chance to enjoy it yourself… Now on to hybrids.
There are two terms you’ll need to understand if you look up more about hybrids: interspecific and intraspecific hybridization. Interspecific hybridization is the kind I’m talking about here and is between different species (generally closely related, although evolutionarily diverged to a significant degree, often to the point that they can’t normally reproduce in the wild). Intraspecific hybridization is hybridization within the same species and falls under general selective breeding above or just the natural crossing within the same species.
Hybridization is, in my opinion, the most uncontrollable, and therefore most problematic tool in the genetic modification tool box. Then again it is pretty safe as well, but the Africanized honey bees, killer bees, are an intraspecific hybrid species of the African honey bee and one of a number of the other subspecies of the western (European) honey bee. Hybrid plants are quite safe overall only posing environmental risks if they out compete native species, and I’m not currently aware of hybrid plant running amuck, it’s more that hybrid animals pose direct risks. Though, fun fact: there are a number of gentle Africanized honey bees which are now useable by bee keepers, and there is a move to try to make all Africanized honey bees gentled, though that’s a long ways off yet.
Why do I find hybridization more problematic? Well, because you have whole genomes, and your mixing them together and seeing what you get before selectively breeding in the conventional sense. This leaves the very real opportunity for unexpected events to occur since you are less able to control which genes will be crossed with which. While generally this effect is muted by the fact that only very similar organisms are able to be crossed, generally sub species or sister taxa are the limit of successful and fertile hybrid crosses. Though so long as two organisms are in the general area of the tree of life, fill similar niches, and have the same or close to the same number of chromosomes, you might be able to get a hybrid, but there is not clear cut rules for what will work, so some very odd combination might be possible. Plants are much easier to work with compared to animals, since you can keep a line going even if it’s infertile through numerous asexual mechanisms, such as runners cutting, or the artificial method of taking plant cells (normally from apical meristem or an embryo from a seed) and growing them in vitro to produce a large number of genetically identical plants from a single source.
Finally, we have genetic manipulation on the gene level. This is the most complex method in its totality, but actually producing the plants is not that difficult once you’ve isolated the gene(s) in question. The first step in adding or taking away a gene or genes is to identify what your after. Identifying a gene is not a simple task and requires much work, to such an extent that to adequately inform you all would be a series in itself. So rather than give a good explanation I’ll give a basic run down of one common type of method used in plants:
To identify a gene, researcher will often attempt to “shut it off” or disable the gene by mutating plant seeds and looking at the morphology of the mature plant and seeing if the desired traits have been lost or altered. If they have, the genes of those plant will be sequenced and comparing to a linage which has not been mutated.
Once the comparisons are made, the mutated genes are located and can be produced and inserted into bacteria, yeast or viruses for further testing. You need to test more since you only know which genes were mutated at this point, not which one(s) effect the trait your interested in. So you see what each of the genes do, and which proteins they are responsible for. This too is no simple task, but for the sake of brevity I won’t go into it further.
So now we have the gene(s) in question, and now it is time to put them into the plant you want. There are two main ways of to do this: Agrobacterium and Gene guns.
Gene guns? Yep you can shoot genes into a cell! Generally how this is done is a large number of plant cells are cultivated in the same manner I mentioned in the part of this post about hybrids. And then microscopically small gold pellets coated in the gene in question are fired in to the plant cells using a high powered air gun/microscope thingy called a gene gun. Some of those genes will enter some of the cell nuclei. From there you grow those cell into mature plants and pick out those plants which have the traits you want (most of them will be unsuccessful and will be unchanged). Then you sequence those plants and select those which have taken up the gene best, and then breed them into an existing line for several generations to produce a seed stock.
Agrobacterium works similarly to a gene gun, but instead uses the mechanism of the Agrobacterium to insert the genes in question. Otherwise it is basically identical.
All this work and you get one highly control change in a plant, with many inbuilt checks and controls, plus all of what happens above can be contained 100% in lab, and much of it has to happen in the lab, so the escape of seeds is exceedingly unlikely.
So that’s a bare bones run down of the three basic forms of genetic modification, and why Genetic insertion and deletion are not scary, but a very precise tool, which gives us a scalpel to broad strokes of Hybridization and artificial selection.