Synthetic Genes: Exploring Man-Made Genes

[Therian]

Hi I was talking to my friend and she said that it would be possible (maybe) to make synthetic genes and that those genes would be unique genes- she said they would be like plastic (because plastic doesn't occur in nature yet it's a combination of things that exist in nature) (Because the synthetic genes wouldn't exist in nature yet they would be a combination of what does exist naturally) So I guess my question is could somebody give me some info on synthetic- eg man- made- genes. I did find this website http://www.genscript.com/cgi-bin/site2/faq.cgi?category=custom+gene+synthesis

thanks

 

[Moonbear]

That site you linked to refers to a research tool where instead of cloning genes, you can have a DNA sequence synthesized. This is usually used when you want just a short sequence from a single gene that you want to use for something such as transfecting into cells in vitro. You can synthesize any sequence you want with that technology (there may be limits to the length), and sometimes people do use it as a way to introduce an intentional mutation, or they may generate a random sequence as a control (I don't use those sorts of tools in my own research, so one of the molecular biologists could probably comment in more detail about their applications). However, just generating random sequences doesn't mean you'll wind up with anything that does anything in a cell.

I don't think the plastic analogy is a very good one. With DNA synthesis in vitro, you're not creating molecules different from those found in nature, the same 4 nucleotides are still used, it's just the sequences that you can vary.

 

[pattylou]

tangential comment:

One genetic manipulation of cotton, has been to clone in genes from bacteria that *do* make plastics (such as poly beta hydroxybutarate).

There are cotton fields, at present, that grow poly-cotton blends. Cool, hunh?

 

[quasi426]

I likewise must disagree with the plastic analogy because a plastic is simply a material that will remain in its deformed state even after the load is taken away. In other words if I pull on a piece of material and cause it to deform, it will stay deformed when I stop pulling.

I think you're friend might be trying to say that we can make a protein behave the way we want through directed mutagenisis. Another way of doing this would be to know the structure of the protein and make decisions about changing certain amino acids.

But in terms of an organism we have a long way to go.

DNA sequence leads to biomolecule (RNA,protein, and lipids indirectly)
Biomolecules have functions directly related to the DNA sequence and their enviroment.
Biomolecules also interact with other biomolecules which lead to overall function.

 

[Therian]

So basically before I was thinking, oh neat, if you want to use a gene but it doesn't exist in nature, you can just use it by using it as a synthetic man-made gene (Since you couldn't use it as a non-synthetic gene Since it wouldn't exist in any other form) But anyways so I thought you could actually do ANYTHING with genetic engineering since you could make synthetic genes (genes that don't exist in nature) but now because I read this

With DNA synthesis in vitro, you're not creating molecules different from those found in nature, the same 4 nucleotides are still used, it's just the sequences that you can vary.

I'm confused about the limitations of synthetic genes; could you try explaining that in layman's terms? I mean I understand what the person who wrote that is saying, but what would that MEAN in terms of the limitations we would have with synthetic genes?

Thanks

 

[selfAdjoint]

So basically before I was thinking, oh neat, if you want to use a gene but it doesn't exist in nature, you can just use it by using it as a synthetic man-made gene (Since you couldn't use it as a non-synthetic gene Since it wouldn't exist in any other form) But anyways so I thought you could actually do ANYTHING with genetic engineering since you could make synthetic genes (genes that don't exist in nature) but now because I read this


With DNA synthesis in vitro, you're not creating molecules different from those found in nature, the same 4 nucleotides are still used, it's just the sequences that you can vary.

I'm confused about the limitations of synthetic genes; could you try explaining that in layman's terms? I mean I understand what the person who wrote that is saying, but what would that MEAN in terms of the limitations we would have with synthetic genes?

Thanks

Craig Venter has a project to create a synthetic bacterium. His idea is to take a natural bacterium, remove its DNA, insert instead some synthetic DNA that has been built in the lab, and try to get the bacterium to live and do what the DNA codes for. One idea is to have bacteria that can split gases like methane to supply unlimited quantities of hydrogen for an hydrogen economy. I don't believe he's very far along that trail yet, though.

 

[Therian]

Do you think that we could use synthetic genes in living organisms within the next century?
Could we hypothetically do anything that we wanted to with synthetic genes?

 

[Moonbear]

Okay, to try to clarify a bit (I hope...keep asking if I'm still too technical)...

If you're talking about making a new DNA sequence like none found in nature already, you might wind up with something that once inserted into an organism encoded some sort of protein (via the usual transcription-translation process), but it would be as hit or miss as evolution in terms of chances of making a sequence out of all the possible combinations that actually does something (maybe somewhat faster only because we could directly tinker with sequences rather than waiting for chance mutations) but either way, it would be quite a tedious way of doing things with no clear goal in mind that no respectable scientist would bother.

On the other hand, if you're talking about taking known gene sequences from organisms that we know code for certain proteins, and inserting it into some other organism to make that protein in an organism that normally doesn't make it, this is already being done as both a research tool and as a means of using organisms as "bioreactors" to make things like protein pharmaceuticals (sometimes referred to as farmaceuticals). This is also essentially what we're talking about when we say we've made a recombinant protein, that it's not made in the native species, but the gene is put into, for example, a bacteria and then the protein (or RNA, depending on the application) harvested from those bacteria when they are grown in huge quantities.

 

[Therian]

Okay I just wanted to clarify:

Somebody said theoretically anything is possible with genetic engineering

Then somebody said, no theoretically anything is NOT possible with genetic engineering because the things we could do with genetic engineering- even with synthetic genes- are NOT infinite, just really really numerous

So anyways, I just wanted to know which one of the above things was the case. If the second thing is the case, do we know what we could and couldn't do with genetic engineering? Do we know what we could and couldn't do with genetic engineering within the next century?

 

[Moonbear]

Okay I just wanted to clarify:

Somebody said theoretically anything is possible with genetic engineering

Then somebody said, no theoretically anything is NOT possible with genetic engineering because the things we could do with genetic engineering- even with synthetic genes- are NOT infinite, just really really numerous

So anyways, I just wanted to know which one of the above things was the case. If the second thing is the case, do we know what we could and couldn't do with genetic engineering? Do we know what we could and couldn't do with genetic engineering within the next century?

The second is correct. While a lot may be possible, there are limits. More examples of limits are that there are limits to the size of a sequence you can insert in cells (this depends a lot on the cell and existing DNA structure), keeping existing genes functional (if that's the goal), plus some genes end up lethal to a cell. We can do a lot more in bacteria than we can in other organisms, and a lot more in plants than animals.

Right now, predicting what will happen in the next century is really difficult to do. There's a lot of growth in the field right now, but there's no way to know whether that will continue or hit a wall beyond which we get stuck and can't progress much further, or what direction new findings and progress will lead us in. There is also a limitation of what ethics allow us to do and how far to go in certain directions.

 

[selfAdjoint]

Okay, to try to clarify a bit (I hope...keep asking if I'm still too technical)...

If you're talking about making a new DNA sequence like none found in nature already, you might wind up with something that once inserted into an organism encoded some sort of protein (via the usual transcription-translation process), but it would be as hit or miss as evolution in terms of chances of making a sequence out of all the possible combinations that actually does something (maybe somewhat faster only because we could directly tinker with sequences rather than waiting for chance mutations) but either way, it would be quite a tedious way of doing things with no clear goal in mind that no respectable scientist would bother.

Well Venter thinks he can do better than that. See his talk at EDGE, http://www.edge.org/3rd_culture/biocomp05/biocomp05_index.html. Scroll down to his description of his organisation's current research on artificial chromosomes.

 

[Therian]

So if somebody said "Can I do this with genetic engineering" Could you tell them whether or not it would be possible to do that (Depending on what it was) based on the fact that there are limits to what we could do? Or do we just have way too little an idea of what limits there would be to be able to tell anyone for sure what we could and couldn't do with genetic engineering?

 

[pattylou]

A big problem with understanding gene expression, currently, is that we don't always understand how proteins fold. We may know that ATG codes for a methionine, but we may not know where that methionine will end up, in the final protein.

Our protein-folding computer simulations may predict the best folding for a protein (based on... free? Biochemistry help?...energy) but in the cell oftentimes what is predicted doesn't happen. Proteins fold in ways we don't expect, they splice out regions we didn't think they would... about the only things we have figured out with any certainty is the primary sequence of proteins (immediately following translation) and (to some extent) whether they will prefer to be cytoplasmic (soluble in aqueous solutions) or membrane - bound (containing exposed hydrophobic regions.)

So there are a couple of constraints. (1) (From Moonbear's above, IIRC) If you wanted someone to "make" (engineer) a protein with a property that proteins don't normally have, such as --- I don't know ---- ability to withstand superheated pH14 solutions,then you will have little luck (the amino acids themselves would have trouble withstanding this!). (I know, I know, prions are very resistant to most physical treatments - )

And (2) we can't predict well enough, how the sequence we engineer, will be "seen" by the cell. It may fold unusually. It may have some short sequence that you unwittingly engineered in, that tells the cell to cut it in half, or to add a phosphate group, or something. It may have a secondary function that you didn't expect - like something that would kill the cell that is making it.

Most engineering at present, is not making new protein sequences from scratch, but rather finding where such proteins exist in nature and isolating the genes for your personal use. You want something that can withstand great heat and pressure? Isolate organisms from deep sea vents. Hope to clone a gene you can use, from one of them. And so on.

(Venter's work sounds neat. Thanks for the info, Selfadjoint.)

 

[Moonbear]

Well Venter thinks he can do better than that. See his talk at EDGE, http://www.edge.org/3rd_culture/biocomp05/biocomp05_index.html. Scroll down to his description of his organisation's current research on artificial chromosomes.

I only skimmed because that article was rather lengthy, but I don't see where he's saying anything to contradict my statements. He seems to be also talking about basically taking the mix-and-match method of using what we know about one organism to engineer something for another organism.

Patty, very good points about protein folding. Thanks.

 

[iansmith]

Most engineering at present, is not making new protein sequences from scratch, but rather finding where such proteins exist in nature and isolating the genes for your personal use. You want something that can withstand great heat and pressure? Isolate organisms from deep sea vents. Hope to clone a gene you can use, from one of them. And so on.

Just to add to this statement. People are also creating fusion/hybrid proteins. This is basicly using desired functional region of at least two different protein and fusing the region. This is closely related to the synthetic gene idea. Some people in my lab are doing this; however success (functional proteins) has been limited. It may be due to misfolding (based on in silico prediction) or due to codon bias. This brings me to my other point. The problem with expressing protein from a different organism in another organism is that the set of codon (tRNA) use in one maybe differrent and therefore it limits protein production. Some E. coli strains have been modified in order to eliminate this codon bias. Rare tRNA have been added in the gemone. However, it does not always work.

The next step of engineering would be to eliminate codon bias in a sequence.

We might eventually come to a point where we build gene with specific function. This could be "synthetics"This would probably be a fusion/hybrid type. We known that, at least for some, conserved region usually have a specific function. Our knowledge on this is, however, very small. So the idea would be to determine the function of the protein and select the conserved regions that you need. The second step would be to create the gene using the proper codon bias, test its folding with (improved) bioinformatics tools and revised. The third step would be synthesised oligonucleotides, fused and amplify the whole gene. This will not be achieve before we improve our knowledge about protein and conserved regions, and improve the tools we used.