Creating Artificial Proteins

Spy der Mann writes "By examining how proteins have evolved, UT Southwestern Medical Center researchers have been able to design genes to create artificial proteins. The researchers have discovered a set of simple "rules" that nature appears to use to design proteins. By feeding these rules into a computer program, they were able to obtain a sequence of artificial genes. These genes were then inserted into laboratory bacteria, producing the artificial proteins as expected."

Comments

Creating artificial drugs

[Khyber]

Well, we know we've been able to modify DNA to produce insulin from bacteria.

We've got bacteria that crap out metal wires (Can't remember if we discovered them or made them)

Now where's the bacteria that will make substances like xanax or other drugs, so it can make the entire market cheaper and more affordable to those who need it but don't have insurance, and "naturally" at that? (Naturally as in not needing a buttload of power from a processing plant for the drug and wasting energy uselessly)

Re:Creating artificial drugs

[k98sven]

Now where's the bacteria that will make substances like xanax or other drugs, so it can make the entire market cheaper and more affordable to those who need it but don't have insurance, and "naturally" at that? (Naturally as in not needing a buttload of power from a processing plant for the drug and wasting energy uselessly)

Um.. news flash: Drugs have been made that way for years.

But first: This works for proteins such as insulin. Most drugs are not proteins, however.

And for those who are, there is nothing about it which necessarily makes it cheaper or less power-consuming. Bacteria need food. Bacteria need to be kept warm. And most importantly, you've got to seperate and purify your drug from the bacteria and growth substrate and whatnot.

Of course, for proteins you've got no choice. It's practically impossible to synthesize proteins using conventional chemistry. And it's very very difficult (and likely uneconomical) to use bacteria to produce other organic compounds. So these things are complimentary to eachother, really.

Re:I hate to turn this into a flamewar so soon, bu

[salemnic]

I think you might be using backwards logic here. TFA states that by examinig 100 proteins they were able to notice some standard common things about the proteins they were looking at. When they made rules around those common things they could make new proteins.

It's like having 100 pieces of example code to look at before trying to create your own, not generating the code from nothing.

s

Re:I hate to turn this into a flamewar so soon, bu

[Anonymous Coward]

Their "rules" were derived from observing nature, not computed or by any ab-initio means.

The original summary of the article is quite off base, as many of these biochemistry-related revelations are.

I would better summarize the Nature paper as saying that the researchers have found a somewhat reliable method of duplicating a three-dimensional structure by using existing sequences as a simple template. The concept of truly "designing" a protein from scratch remains the Grail of this field.

Link to Nature article

[JuliusSu]

Full text of article, institutional/personal subscription required.

Abstract: Classical studies show that for many proteins, the information required for specifying the tertiary structure is contained in the amino acid sequence. Here, we attempt to define the sequence rules for specifying a protein fold by computationally creating artificial protein sequences using only statistical information encoded in a multiple sequence alignment and no tertiary structure information. Experimental testing of libraries of artificial WW domain sequences shows that a simple statistical energy function capturing coevolution between amino acid residues is necessary and sufficient to specify sequences that fold into native structures. The artificial proteins show thermodynamic stabilities similar to natural WW domains, and structure determination of one artificial protein shows excellent agreement with the WW fold at atomic resolution. The relative simplicity of the information used for creating sequences suggests a marked reduction to the potential complexity of the protein-folding problem.

From this page : a WW domain is the smallest, monomeric, triple-stranded, anti-parallel beta-sheet protein domain that is stable in the absence of disulfide bonds, cofactors or ligands.

Re:Name that film...

[ScrewMaster]

Nope. From the original movie "The Fly" with David Hedison.

Rephrasing the Story Description

[Fractal Dice]

The rules for how DNA encodes protiens have been known since before I was born. The evolutionary mapping of how the genes coding different protiens duplicated and evolved into new structures is fairly easy to map out (give or take a brute force algorithm that runs in double-factorial-time to search through all evolutionary trees looking for the one that minimizes the number of mutations required along the way).

So this group has calculated the most likely common ancestor of the gene that now codes for a whole family of protiens, encoded the solution in real DNA, stuck it into bacteria and shown that it actually does produce a protien that they have been able to isolate the actual protien so that they can explore what it does/did.

(the term "articial protien" seems very odd to read - before I think it through, it sounds as though its hinting there is something mystical to "natural" protiens untouched by humans)

Re:Information Theory Usages?

[Nutty_Irishman]

It would also, of course, be interesting if you could use this to work backwards through the genome to a set point


There actually is research that looks at predicting the last common ancestor between two species. For example, given man and ape, you can make a prediction on what the man/ape gnome was before they diverged into two species (not to go into details, but a lot of species divergence is the result of some kind of large scale chromosome rearrangement that makes it impossible to sexually reproduce). Remember, we didn't evolve from an ape, we diverged from an ape. The man and ape have had the same amount of time to evolve their genomes to become the species that we are today. Most people assume that at one time in our past we looked exactly like present day apes, but then evolved into humans. Where in fact we(again, both man and ape) probably looked something like a cross between an ape and a human-- whatever that might be.

Ignorant Comment

[Anonymous Coward]

Before there is a large debate in the ethics community, they (you) ought to get the facts straight.

E. coli is not a virus. Depending on the strain, the genome size is anywhere from 4.6 to 5.2 millions of base pairs. Putting one of your very own E. coli genomes together would be difficult and expensive. Better yet, why not just grow some? They'll spit up their genome after a few biochemical steps at the lab bench. If you were talking about a phage, their genomes are variable, averaging 35-50 kbp, or thousands of base pairs. That's still difficult and expensive. Even when you got your phage genome put together, now what? You need infectous phage particles to go about the infection/replication cycle. Phage particles aren't infectous to mammalian cells, so that's not what the terrorists are after.

Working with and engineering virus particles isn't reliant upon ordering oligos and assembling the entire genome. The hallmark of being a genejockey is allowing the DNA-containing object to do the replication work for you. You let the bacterial culture divide. You let the mammalian cells divide. You let the phages infect the bacteria and replicate. You let the viruses infect the mammalian cells and replicate. You then harvest the DNA from the millions of [whatever], make some CTL+C and CTL+V with some enzymes, and package the DNA back up in the infectous particles. It's a non-trivial process, and the ability to order oligos on the web doesn't magically give someone the ability to hack out a super virus.

Drexler first proposed de novo protein design

[Savantissimo]

K. Eric Drexler, (primary originator of molecular nanotechnology) first came up with the idea that the protein folding problem did not have to be solved in order to design and produce proteins of specified shapes and functions:

From

http://www.imm.org/PNAS.html:
Proc. Natl. Acad. Sci. USA
Vol. 78, No. 9, pp. 5275-5278, September 1981
Chemistry section
Molecular engineering:
An approach to the development of general capabilities for molecular manipulation

K. Eric Drexler
Space Systems Laboratory, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139

Communicated by Arthur Kantrowitz, June 4, 1981 .... What can be built with these tools? Gene synthesis (3) and recombinant DNA technology can direct the ribosomal machinery of bacteria to produce novel proteins, which can serve as components of larger molecular structures....

natural scientists seek a more general understanding than design engineers require. Science seeks the ability to predict the conformations of all natural polypeptides. In attempting this, protein chemists can search for a minimum-energy chain conformation (in hope that the protein assumes not a local but a global minimum-energy conformation) (6) or can attempt to follow the chain-folding mechanism to find the final conformation (7). Prediction will be easier if the natural conformation has outstanding stability or if its folding mechanism proceeds in a sequence of strongly preferred steps....

Engineers (in contrast to scientists) need not seek to understand all proteins but only enough to produce useful systems in a reasonable number of attempts.... Through use of strategically placed charged groups, polar groups, disulfide bonds, hydrogen bonds, and hydrophobic groups, the engineer should be able to design proteins that not only fold predictably to a stable structure (sometimes) but that serve a planned function as well.

***

PROTEIN ENGINEERING:
A 1988 view of some 1981 predictions
K. Eric Drexler
Visiting Scholar, Stanford University. Box 60775, Palo Alto, CA 94306

A 1981 paper [1] discussed de novo protein design as part of a long-term strategy for developing complex molecular devices and systems. It presented arguments against the view that the fold-design problem is an extension of the classical (and still unsolved) fold-prediction problem (i.e., predicting folds from sequences without homologous models), a view which has discouraged efforts at design.

Fold prediction is a scientific problem: it must deal with naturally evolved sequences, but natural selection's 'design goals' enforce only the physical reliability of folding -- not its human predictability. This results in folds of only minimal stability. Fold design, in contrast, is an engineering problem. Protein engineers, exploiting their freedom of design, can work with sequences artificially selected for superior predictability and stability of folding. These observations indicated that "the difficulties encountered in predicting the conformations of natural proteins do not seem insurmountable obstacles to protein engineering" [1].

In accord with the implications of this argument, we have seen the successful, de novo design of a globular protein (alpha-4) [2,3] while the classical fold prediction problem remains unsolved [4]. Likewise confirmed has been the suggestion that design can increase protein stability beyond that enforced by natural selection. In recent years, deliberate single-residue modifications have raised protein stabilities through a variety of mechanisms [5,6]. Owing to design choices consistently biased toward stability, the protein alpha-4 has a stability of 22 kcal/mole, substantially greater than the 4-9 kcal/mole of typical natural proteins of similar size [3].

Successful protein engineering marks a milestone in a research agenda leading toward capabilities of broad technological significance [1,7].

Re:I hate to turn this into a flamewar so soon, bu

[Byzboy]

>the concept of "species" is a completely man-made concept and actually rather arbitrary. Nature does not make any such distinction.

Not true species, is the only testable (ie scientific, ie non-arbitrary) classification. All the rest are arbitrary. A species is defined as a population of organisms which have reproductive isolating mechanisms which prevent the production of a viable (including sexual) offspring. ie two organisms are members of a different species when they are unable to reproduce and/or produce a viable offspring.

For the benefit of the slashdot crowd

[soren.harward]

Since protein engineering is my field of study, for the benefit of the /. crowd (and my karma) I'll fill in the gaping holes left in the New Scientist article, and give you a little more background on the Nature paper. Because the writeup on /. is a perfect example of "scientific telephone": a semi-interesting result gets written up into a paper, which once it's been through several layers of editors suddenly seems like a major breakthrough.

The Nature paper isn't a breakthrough. It's not even really a major advance. Scientists in my field have been creating artificial proteins for five to ten years now. And yes, even some of them designed completely from scratch (though they're really simple; nothing as complex as, say, ATP synthase) instead of just taking a known fold pattern, known as a "motif." The "WW domain" (domain, in protein parlance, is a small, independent structure within a much larger protein---think of it like a module within the kernel or Apache) is a common fold in hundreds of different proteins. Basically, they analyzed the sequences of all of these WW domains, and figured out which positions were meaningful. It's kinda like reading through some code in a programming language you don't know, and figuring out which lines are comments and which lines are actual compilable code. This group found that the number of interesting positions is small, that they could identify them just from the amino acide sequence instead of having to mess with the whole complicated 3D structure of the domain, and that if they put together a protein with the meaningful amino acids intact and the non-meaningful positions randomized, then in many cases they could still get a pretty decent protein (in terms of structural similarity to the "natural" protein) out of it. Most of the paper is devoted to showing via various methods that they did get a pretty decent protein.

So what does this mean for me, assuming that this paper is absolutely correct (which I admit is a little hard for me to determine with one quick reading, given that I'm just a first-year grad student)? It means that the number of meaningful amino acids in a protein (at least in terms of overall structure) is pretty low, and that they can be identified without knowing what the full 3D structure is. This is good, because for a lot of proteins, the 3D structure is difficult to get. However, they picked an easy target: a small domain where there are over 100 unique sequences known. We'll see how well this method holds up with longer domains and fewer unique sequences. The S/N ratio won't be nearly as good.