Cloning Mammoths

Anonym Feigling writes "For your consideration... An article over at the New Zealand Herald discusses some of the challenges a japanes team faces as it attemps to develop a system to create a clone from 20,000 year-old mammoth tissue samples discovered in Siberia. It seems to me that shortly after death, any animal's/plant's "cellular repair mechanisms" (for the lack of a better...) will fail, and thus the probability of finding a single cell with perfectly intact DNA from which to create a clone is pretty well zero. Interesting stuff, but it seems that practical considerations (think code rot) would make it difficult."

Comments

Code rot probably not the best analogy

[Quixotic Raindrop]

I get what you're saying, but code doesn't really "rot" in the same way that living cells do. I believe that if there are any intact cells they'd be lucky, but you don't need an intact cell to extract a DNA sample.

Re:Code rot probably not the best analogy

[bcrowell]

They had a pretty hard time with the neanderthal femur they found a while back.
AFAIK, they only analyzed the mitochondrial DNA (mtDNA) from the neanderthal sample, not the nuclear DNA that codes for proteins. They also did Oetzi, the 5000-year-old man they found in an alpine glacier.

mtDNA is only inherited from your mother, and is useful as a clock because it's not strongly selected for. There's also a lot more mtDNA in a cell than nuclear DNA, which makes it easier to recover mtDNA from an old sample than it is to recover nuclear DNA. Even though mtDNA is easier than nuclear DNA, they didn't even try to recover the complete mtDNA genome on these samples -- they just used them statistically, as clocks.

The neanderthal DNA showed that our last common ancestor with the neanderthals was 500,000 years ago, which implies that we're separate species, i.e., it supports the total replacement model (we lived alongside neanderthals and Homo erectuses for a long time, and then they went extinct) rather than the multiregional model (where H sapiens arose through worldwide interbreeding with other archaic hominids).

Oetzi's mtDNA was virtually indistinguishable from the mtDNA of the people who currently inhabit the region.

Re:Code rot probably not the best analogy

[Sgt York]

Getting intact full DNA from a living cell, prepared freshly is a challenge. And it does "rot", just like living cells. It's a molecule, and anything that will decay a cell will take the DNA in that cell along with it.

There is also molecular decay, independent of decomposition from microorganisms. DNA is subject to autocatalytic acidic hydrolysis; that is, in any solution with a pH lower that about 6 or 7, it will break itself up into little chunks. Most tissues become quite acidic after death. The DNA is still there, kinda, but it's broken up into little bits.

You don't need intact cells to get DNA, that is true. But to clone, you need a full genome, intact. Each chromosome needs to be a full sequence. For other applications, busted up DNA is fine. You can sequence, look for similarities, etc. But to clone, you need the whole thing, all in it's correct pieces.

Re:Huh?

[TCQuad]

It's significantly more difficult than that. You know those nice, clean chromosome pictures that you see in textbooks? Those are actually from actively dividing cells, the only time that the chromosomes are actually large enough to see in a microscope. Most of the time they're not compacted to that extent and therefore aren't very visible. So picking pieces by any sort of visual inspection at that level is out.

You could run these out on low percent agarose gels, and you'd be able to separate on the basis of size, but you wouldn't be able to tell the difference between a fragment of the X chromosome and the Y chromosome, so it'd be almost impossible to isolate based on size.

And the determination of good v. bad is much more difficult. How long should a mammoth chromosome be? Was this small one cut or that large one ligated? And when you don't know the genome, you're flying blind.

You also can't tell about base-pair damage, etc. without knowing what the sequence was originally. And to get the accurate original sequence, we need living tissue. Therefore, by definition, this is going to be our best guess as to what a wooly mammoth should be. There may be gene differences, etc. compounding the dietary differences that the mammoth will encounter.

The best scenario is to isolate one intact nucleus. Why? Well, if over the course of time the nucleus managed to remain intact, then the cell was probably fairly protected. If the cell was protected to a significant extent, then you're good with most major chromosomal damage and any ones with significant minor base pair damage will be nixed (i.e., the pregancy won't take).

Re:Don't always need an intact DNA

[TCQuad]

Three considerations:

1) In order to PCR amplify something, you need primers which bind to the target areas and begin the replication. The primers need to have a known sequence, and we don't really know the mammoth genome, so we don't know what we're looking to amplify.

2) Mammoths have multiple chromosomes, so this isn't a one-step process. You'd need to repeatedly amplify section after section on each chromosome. Not impossible (per se), but not really feasible with todays technology due to:

3) Good day, high wind, Herculase (a PCR enzyme for long targets) can get 48,000 base pairs in one cycle with reasonable accuracy. The E. coli genome is 5.4 million base pairs. To PCR the entire E. coli genome you'd need to repeat the process 113 times to get the entire genome; if you're lucky enough to get the max every time, it'd take a lot of complex stitching to get it done. Of course, a mammoth is a lot bigger and more complex than a bacterium. The Fugu (pufferfish) genome is ~100 times bigger than E. coli (300 million), humans ~1000 times (3-4 billion). You can see the difficulty in using PCR for this type of application.

You are right, in principle, that you should be able to do all of these (eventually) but you also have to remember that each of these processes (not to mention troubleshooting!) takes materials (original DNA) in significant quantities. If we don't have a herd of mammoths, we probably don't have enough for what you suggest.

Making a hybrid is not really cloning...

[geoswan]

The initial plan was to find mammoth sperm cells, which could be used to inseminate an elephant to create a mammoth-elephant hybrid. But no sperm cells have been found, and other samples retrieved have been rendered unusable by time and climate changes.

Finding mammoth sperm, and impregnating an elephant is not cloning, it is just artificial insemination.

Worth noting is that if it turns out that the mammoth is closely enough related to a modern elephant for a pup to be born that doesn't mean the beginning of mammoth-elephant ranching. Lots of hybrids aren't fertile, like mules.

You ever hear of anyone crossing Indian and African elephants?

Re:Making a hybrid is not really cloning...

[willtsmith]

True.

However, this method (if successful) would be the quickest, easiest way to get a living mammoth (hybrid). Once you had your hybrids you could bank up lots of samples and clone it using the Dolly technique.

Gene therapy could be used to "mammothize" the hybrids. Subsequent clones from "gene-therapized" samples would be even more "mammoth" then previous generations.

Ultimately, even if a fertile mammoth hybrid could be produced, it would take a VERY long time to produce a near-pure mammoth via selective inbreeding. It takes to darn long for the females to reach fertility (which probably contributed to their extinction).

Ultimately, we will have the technology to reproduce a live mammoth given the VERY pristine condition of the frozen mammoth sample. It is a certainty that one could re-constitute a complete DNA through multiple samples.

I think it will teach us a LOT about evolutionary biology (and genetics as a whole) to effectively DEVOLVE elephants (to a common anscestor) and re-evolve them into Mammoths.

Don't forget that modern DNA is littered with discarded/unused sequences. For example, there have been experiments done where researchers generated chickens with TEETH. The information to create teeth and grow them is there (likely from their dinasour anscestors). It just had to be turned back on.

Re:Would you want to?

[alleycat0]

The disappearance of Pleistocene megafauna coincided with the extinction of numerous other species, many of which likely would have been considered as sources of neither food nor other resources, thus casting doubt on the theory that humans hunted mammoths into extinction.

ObCredentials: IANAP (P=paleontologist) but i spent most of my life as a professional archaeologist.