First life with 'alien' DNA | Hacker News Reader

First life with 'alien' DNA

nature.com

60 pointsjamesbritt46 comments

  “If you read a book that was written with four letters, you’re not
  going to be able to tell many interesting stories,” Romesberg
  says. “If you’re given more letters, you can invent new words, you
  can find new ways to use those words and you can probably tell more
  interesting stories.”

He needs a better way to motivate his work to a lay audience without losing those whose immediate response will be dismissive because binary can tell interesting stories in two letters.

That quote sounds rather silly, I agree, but if you think about it, these results don't have too many immediate applications or benefits that would resonate with a layman. Therefore it's challenging to try and explain the importance of this achievements to a wider audience. In this light, I kind of understand the book example.

>> you can probably tell more interesting stories.”

books aren't good example. For today's layman it should have been translated to "probably build Super- and X-men"

Honestly, it actually is a pretty good example, but you need to be creative.

If I tell you that your book can only have 26 characters, you'll say, 'great, who cares, that's the whole alphabet'.

But imagine the books you cannot write: No punctuation, no foreign languages, no mathematics, no accents, etc. Give yourself an escape character, and all of a sudden, you have access to a LOT more creativity.

I think the implication was being restricted to English words containing only 4 given letters.

But you still have no real sense how big the effect is. Saying that now 216 instead of only 64 triplets are possible give you some vague sense (but may make you overestimate it because of redundancy, forbidden combinations and what not may be relevant).

Quite, and even without binary, it's those four letters that got him to write, made his laptop, etc.

Written with "only the ten hundred words people use the most often".

http://xkcd.com/1133/

That's such a cool idea

Any idea how I can get a printed/framed version of that?

http://store.xkcd.com/products/up-goer-five-poster

Not a biologist, but isn't he omitting the key fact that those 4 letters can't construct completely new meanings? As in the 4 base letters can only make 20 "words" (the fundamental amino acids), which each have some pre-defined "semantics" to them. Sure we still have tons and tons of protein "sentences", but the analogy fails because we can't (yet) build our own arbitrary syntax + lexicon using those words; they have specific chemical interactions that limit their usage.

As in the 4 base letters can only make 20 "words"...the analogy fails because we can't (yet) build our own arbitrary syntax + lexicon using those words

Isn't that just the circumstance of the particular hardware we have? There's nothing fundamental preventing the construction of a different set of molecular machinery with its own "opcodes." It may be beyond us right now, but there's nothing fundamental preventing it.

Since the actual paper is paywalled, can someone answer this question?

This paper uses d5SICS-dNaM base pair. How does it compare to dZ-dP pair in the following paper? I checked that two pairs are different.

Enzymatic incorporation of a third nucleobase pair. (2007) http://www.ncbi.nlm.nih.gov/pubmed/17576683

The major accomplishment of the paper referenced in the original article is that they demonstrated the feasibility of unnatural bases in vivo. The work in the Yang, et. al. paper occurred in vitro, whereas the Malyshev paper was in an E. coli system. Malyshev, et al. noticed that the unnatural bases were being degraded in the space between inner and outer cell membranes (periplasm), so they added a membrane transport protein to import the unnatural bases into the cell. From there, the E. coli was able to incorporate the unnatural bases into replicates of an introduced DNA segment (plasmid) using its endogenous replication machinery. They also demonstrated that the unnatural bases were not a hinderance to growth, and they they were not excised by the DNA repair enzymes.

There have been other papers demonstrating the use of unnatural bases in various cases in vitro (some cited in the paper), but this is notable because it is a thorough example of use in vivo. Peter Schultz has done similar work, as well as exciting work on unnatural bases in tRNA.

The bases in the Yang paper and those in the Malyshev paper both exhibit Watson-Crick bonding geometry, and pair by hydrogen bonding. One distinction of those mentioned in the Malyshev paper is the presence of a sulfur substituent, though there is no mention of its participation in the bond.

  "Malyshev sees the ability to control the uptake of
   foreign DNA bases as a safety measure that would prevent 
   the survival of alien cells outside the lab, should they 
   escape."

This reminds me of that line from Jurassic Park where Ian Malcom says: "I'm simply saying that life, uh... finds a way."

Anyway, this is interesting in the sense that perhaps the role of humanity in the big picture of life evolution is to do exactly this kind of thing...

if we can imagine each civilization to have its own "black box" what would be typical last words recorded by it?

"ability to control the uptake of foreign DNA bases as a safety measure that would prevent the survival of alien cells outside the lab, should they escape. But other researchers, including Benner, are trying to engineer cells that can make foreign bases from scratch, obviating the need for a feedstock"

sounds like a darn good candidate :)

Is there some behaviour that can't be coded with natural base pairs?

This is the 'spot' where you put the ink in your 3D atom printer.

Life had used its entire 'ANSI-code' for its own purposes, (naturally (literally)). But if we want to insert our own stuff, we need a place to put it. So these guys refactored the entire code of e. Coli to avoid one of the lesser used characters, which can then be used as an escape character for any new code you want to use. Basically, refactoring an old ASCII program (e. Coli (20 amino acids)) to allow compatibility with UTF-8 (modern chemistry (~unlimited amino acids)) by shifting all uses of one of the letters to a compatible, but not identical case. And now you can use that letter for anything you like (when designing proteins).

You get to use new amino acids in building your proteins. The current amino acids have been evolutionarily defined by being metabolically creatable, useful, etc. and are restricted in number. Some get modified after the fact, but in general, there are only a limited number of amino acids used to build proteins. (Protein = biological machine)

If you have an 'escape character' in your DNA, you can insert any amino acid you want into that protein by registering it in that empty spot. This can include introducing amino acids you created chemically in a lab. So now your new organism can start to use amino acids in its proteins that are impossible to build in nature. Creativity gets opened wide. Right-handed amino acids are the obvious choice to start with - biologically 'inert' (immune system doesn't recognize them), but functionally identical. Theoretically you could start to use non-organic atoms. New biologically orthogonal reactive groups. Or entirely new structural features.

You can now incorporate non-natural chemical moieties with atomic resolution.

In addition to the applications mentioned by toufka, one would be data storage.

There has been some work[1] done on using DNA as a medium of high-density self-replicating storage for digital data, and it would be exciting to have another couple bases to work with.

1. http://www.nature.com/nature/journal/vaop/ncurrent/nature118...

What would be exciting is if the bases had properties that made them amenable to easy read operations using non biological means. Currently it takes 1 - 3 days fort a DNA sequencer to read what is effectively a few gigabytes of data in a human DNA sequence. That is painfully slow for any practical use. However if we had DNA bases that had atomic properties that could interact with some sort of electronic device, we might be able to engineer a sequencer that is quite easy and cheap. In fact, we might even be able to design an enzyme that translates regular DNA into a synthetic form for easy sequencing (similar to how RNA polymerase turns DNA to RNA)

Is there a term for Singularity due to runaway evolution of an artificially created species?

It's similar to http://en.wikipedia.org/wiki/Grey_goo.

Keep in mind, life has been trying to 'runaway' since it was made. We're really good at biological warfare. A species from much earlier in our evolutionary history wouldn't stand a chance against a modern organism. We're really well optimized. It would be very very difficult for a novel organism to get a leg up on the few billion years of adaptation we've accumulated, for this particular environment in which we live.

"A species from much earlier in our evolutionary history wouldn't stand a chance against a modern organism [in the modern environment]". I doubt most species would stand a chance going back in time to an Earth with no oxygen!

I am worried about we becoming obsolete wrt to a more modern organism :)

They solved that in a nifty way: the new bacteria can only reproduce in the lab.

The "Alien" DNA contains two new base pairs. If it wants to reproduce, it must construct a new cell which also contains the new base pairs. So it must get these new base pairs from somewhere.

Natural base pairs can either be synthesized by the cell or eaten. These Alien pairs cannot be synthesized by the bacteria, and do not exist in the Earth's environment.

The only way for these bacteria to reproduce is in a lab where they are "fed" the unnatural base pairs.

dr_ian_malcolm.jpg

(The parent comment references the "Life finds a way" quote from Jurassic Park, possibly in jest)

Can't happen: Evolution is limited to small steps caused by mutation. First, there is nothing in a cell which creates anything remotely like the missing base-pairs so there is nothing available to mutate into a base-pair synthesizer. Second, as the cells are grown in a vat of artificial base-pairs, there is no evolutionary advantage to developing the ability to create base-pairs.

It's like scurvy: humans have no evolutionary pressure (outside of 19th Century sailors) to suddenly start producing Vitamin C themselves because they eat fresh food.

"Mid-21st Century History?"

Four isn't the interesting number, 20 is.

Ugh, that headline.

It's a good thing they put quotes around 'alien' otherwise I wouldn't have been tempted to click on it.

Somewhere in La Jolla, Peter Schultz is asking why he's not getting credit for having done this... years ago.

EDIT: I had also completely misremembered this, thanks to cjensen for the correction. On a similar note, their next efforts are to use these new bases in the genetic code. While it's very interesting and could have some great uses for restricting the spread of modified organisms, an expanded amino acid repertoire is a much more useful achievement.

The weird thing is that it's another lab in La Jolla that published this, so they must have known about Schultz's work. What's new about this work vs. Schultz's work, and if he's in the same institution, why didn't they work with him on this? I saw Schultz talk several years ago, and was really impressed; what is the improvement that justifies the claims of a huge milestone?

Science reporting is completely fucked up.

Assuming Wikipedia describes Schultz's work correctly, this is a different achievement.

DNA is essentially an instruction set. Each instruction is three base-four digits (a sequence of three base pairs). A single instruction is either STOP, START, and for every amino acid there is one or more ADD AMINO ACID TO SEQUENCE instruction.

Schultz altered the interpretation of some of the ADD AMINO instructions so that a different, non-natural amino acid is added.

This new creation alters the mechanism by changing from a base-four instruction set to a base-eight instruction set. As a side benefit, bacteria based on this new mechanism cannot reproduce in the wild.

The interesting part of schultz's work is the use of modified tRNA adapters. Personally, I think this is a far more tractable approach.

I think the main breakthrough here is that these nucleotides are not only unnatural but that the host can actually duplicate them; the E coli in this experiment passed the new nucleotides to 24 child cells.

And also, "When the supply of foreign nucleotides ran out, the bacteria replaced the foreign bases with natural ones."

This could have interesting implications in the theory of the origin of life; with DNA/carbon-based life shown to be adaptable to slightly different forms of the chemicals it needs in the deepest levels of genetic encoding, the set of initial conditions that could give rise to the life we see today grows.

Schultz's work (as I've understood) modified tRNA to alter the genetic code[0] which converts nucleobases into amino acids. This allows for a different code compared to the one found in nature.

This work, on the other hand, adds 2 new nucleobases to the existing 4 bases.

That's not the same. IMHO it's a very different and novel work: in principle the genetic code can now accept an alphabet of up to 216 amino acids, compared to 64 previously (with nature essentially using 20 of those; this assumes three-nucleotide codons).

It's true that Schultz's work, at a higher level, is necessary for this to be of any practical use. But this work has significant additional benefit on its own: for example, I believe you could immediately construct a genetic code that used the additional redundancy to be far more stable against mutation.

[0]https://en.wikipedia.org/wiki/Genetic_code

Somewhere in La Jolla, Peter Schultz is probably not all that worried about it.

Currently I would say it seams kind of useless, he could have spent his time on something better...

This is basic science. It doesn't need to be directly useful. The possibility of creating lifeforms that use different base pairs in their DNA is fascinating. It has implications about how similar or dissimilar life that evolved separately elsewhere in the universe could be. I hope they keep up the good work!

Right why doesn't he move to SV and work on Serious Real Problems like how to get people to click on ads and the latest Laundry As A Service app?

I was thinking more like doing something with the DNA.

To me this seams like adding the letter ö and ä to the English alphabet there is no need for them. If we put them in the alphabet we still won't use it cause they have no purpose.

Have you ever gone to the movies, or watched any TV shows? If so, couldn't you have spent that time doing something better?

Reply on news.ycombinator.com