Friday, January 3, 2014

GMO methods

Here's where we get into the real nitty-gritty science of genetically modified organisms (GMOs): the technology used to make them.  I defined (mostly) what a GMO is in the last post; now we see just what the methods are which set them apart.
A. tumifaciens Ti plasmid, courtesy of BacMap

Gene splicing of this sort really took off some thirty years ago, when the first trials of herbicide-resistant tobacco were done in the U.S. and France.1  Many lab experiments had been done over the previous decade, but when the tobacco field trials were done, along with the test of a spliced bacterial strain that successfully helped protect a field from frost damage,2 the era of commercially viable GMOs had arrived.

Since then, the specific technologies have changed a bit, but the process remains the same. I'll lay it out, step by step.

Step 1: Identify the gene you want.

This could be a gene from a mutant of the same plant variety, a different variety, a different but related plant, a wildly different plant, or something from another kingdom entirely, such as bacteria or animals.  The more closely related it is, the easier it is to get the target plant to express that gene (read it and make a functional protein).  But there are a few tricks to get even foreign genes to work.

Gene sequencing has made selection a lot easier.  We can compare the genetics of the plant we have with those of a plant that does what we want, and pinpoint genes which might be responsible for the difference.  We have maps for what all the genes in some simple organisms do, such as some bacteria, and we can use those to find genes that look similar in other things.  And there are methods to "knock out" the genes we're interested in, to see whether the feature goes away.

So once the researchers have picked out what looks like a good candidate, the next thing is...

Step 2: Isolate the gene.

There are a bunch of little steps here, all with the same goal: get a specific tidbit of DNA into the open.  First all the DNA is pulled out of the cells in a sample, and washed clean of the remnants of cell wall, cellular machinery, and so on.  Then an enzyme or two is added to the DNA, to break it into bite-size pieces; the enzymes used have particular sequences they like to "snip" at, and one is picked which will cut as close as possible to the target gene, but leave it intact.

Since there still isn't a lot of material to work with, the bits are run through a process called polymerase chain reaction (PCR), which multiplies the target gene over and over.  Ideally, with good primers (the "hooks" used to fasten on to the piece so it can be multiplied) only that gene is amplified, and it comes out with thousands or tens of thousands of copies.  Then it's separated out from anything else in the solution, and you have your little vial of the copied gene.

Step 3: Add some extras.

Most of the time, a new gene won't get used in the plant it's put into.  Or, if it is, it won't make as much as we want of the protein it codes for.  How often a gene is used is controlled, largely, by promoters: little sequences at the beginning of the gene which tells the cell when to use it.  A plant virus known as Cauliflower mosaic virus (CaMV) came up with a couple of promoters of its own, which it added to the genes it inserted into its host during infection.  The main CaMV promoter, known as 35S, is "always on"; if the promoter is there, the gene it's attached to gets expressed all the time, and cranks out lots of orders (RNA transcripts) for its protein.3  There are other promoters used in making GMOs, including another from CaMV, but the 35S promoter is still very widely used.4

Once you've got your new plant, how can you tell whether your gene actually got into it?  Many insertion methods are pretty chancy.  Therefore, having some easy way to tell your gene is in there is helpful.  As a result, there are a bunch of available marker genes,5 which give the new plant some new, harmless ability which it's easy to test for.  They include selectable markers, such as antibiotic resistance (kanamycin is common), and reporter genes, which make some substance which can be seen or tested for chemically.  The gene for green fluorescent protein (GFP) makes a protein that glows under blacklight, and GUS creates a stain when an enzyme is used on a test cell.

These markers are typically tacked onto the desired gene before packing the whole thing up, and they are present in the mature modified plant.  That said, they're pretty useless in the field, and unlikely to have any effect on the plant's survival or the animals who eat it.  There's been more of a question about the effect of the promoters, but I'll go into that in a later segment.

Finally, we're ready to...

Step 4: Get the gene into the plant.

There are a bunch of ways to do this.6  The earliest method for introducing genes into plants used a bacteria called Agrobacterium tumefaciens, better known as the cause of crown gall (if you grow roses, you might have heard of it).  A.t. has a loop of genetic material which can cut into the DNA of just about any non-grassy plant, and insert its cargo of a handful of genes.  Scientists took that carrier loop, known as a plasmid, and deleted the native A.t. genes; then they could splice in anything they wanted.  All they had to do then was to breed up a bunch of these bacteria, with their new genetic cargo, and let them infect some cells from the target plant.  It worked pretty well, as methods go, and it was the means used to create the first generation of genetically modified plants.

Beyond Agrobacterium, there are some viruses that can also carry genes when they infect a plant, and viruses or other bacteria which can modify bacterial GMOs.

When a plant cell is stripped of its cell wall, so that all it has is a cell membrane similar to those of our own cells, a small electric shock can open up holes in the membrane long enough to slip some DNA in.  If it's done right, that DNA is taken into the cell and treated as if it belonged.  Specific chemicals can have the same effect of temporarily punching little holes in the cell membrane to allow new genes in.  These methods, electroporation and chemical poration, are much more successful than Agrobacterium at modifying grassy plants.  Vacuum infiltration, using vacuum to open up cell walls enough to infuse genetic material, has a similar end result.

Microinjection uses a small needle to inject DNA into plant cells.

Finally, the most recent and most widely applied category of modification is bioballistics.  The DNA cargo is loaded onto tiny "bullets" made of tungsten or gold, which are then fired at the plant like a shotgun.  Some of those bullets penetrate cells and drop their cargo, which is then taken up.  It sounds very haphazard and inefficient, but it works surprisingly well, and has been used to create many of the newer GMOs.  It helps that this "gene gun" works on grassy plants, such as rice and corn, which can be hard to modify by other means.

Step 5: Make a new plant.

Now that you've got the gene into the plant... but wait, it's only in a handful of cells at this point.  The next thing is to make whole plants, where every cell is modified.

In some cases, like bioballistics, the embryo of a seed can be the part modified, so it's pretty simple to grow it up into a whole plant.  With Agrobacterium modification, chemical poration, or electroporation, it takes some work, because the cells that have been modified are random tissue cells.  Fortunately, plants have the ability to grow whole new individuals from any single cell, if the conditions are right.  The procedure to do that is called micropropagation, or tissue culture, and it's used for other purposes such as creating disease-free plants from virus-infested parents.7  With time and care, the individual modified cells are coaxed into making roots and shoots, and finally they're indistinguishable from those grown from seed.  It's more difficult with grassy plants, though many of the problems have been solved over the years.

At this point, the plants can be test-grown in the greenhouse to make sure they grow properly and have all the desired traits.  In many cases with crop plants, the last step is...

Step 6: Breed a new variety.

Plants which are easy to modify aren't always the ones that offer the best performance as a crop.  So, once it's clear that the target gene (the transgene) has successfully made it into new plants, those modified plants are bred with a high-yielding commercial variety, using traditional methods.  The offspring are examined to make sure they still have the transgene, and several generations down the road, the new variety is stable and ready for general use.


So there you have it.  The process of creating modified plants has brought up a bunch of questions (separate from those raised by the genes themselves), including some about the safety of the promoters, how genetically stable the new plants are, and whether randomly-inserted genes might disrupt the plant in such a way that it would become harmful to eat.  I'll address those issues all together in a later segment, under "Concerns".

Next up: which plants, and which genes, have been thrown into the GMO pool so far.

[This article is part of the series on GMOs. Jump to the first post for an overview and index.]