July 16, 2008

28,000 year-old Cro-Magnon mtDNA from Italy

The researchers could verify that the sequence of the Cro-Magnon (which was the Cambridge Reference Sequence, common among modern Caucasoids) was genuine, since it was different from that of all possible contaminating individuals who handled the find since 2003.

This raises an interesting methodological problem. Should researchers with common mtDNA sequences be handling ancient remains? It seems like blind good luck that no one out of individuals had the quite common CRS. In the recent mtDNA paper on the Mycenaeans for example, they were able to identify contaminant sequences by the fact that the author and experimenter had a particular mutation; if she was plain CRS, they wouldn't have been able to disprove possible contamination for the CRS individual from Mycenae.

Ascertaining authenticity is challenging. In an ideal situation the sample is handled by only a single individual, and one who is unlikely to possess the same mtDNA type as the sample. An obvious solution to this would be to recruit a person of remote geographic origin to do the lab work, e.g. a Japanese person to work on European samples and vice versa.

PLoS ONE 3(7): e2700. doi:10.1371/journal.pone.0002700

A 28,000 Years Old Cro-Magnon mtDNA Sequence Differs from All Potentially Contaminating Modern Sequences

David Caramelli et al.

Abstract

Background

DNA sequences from ancient speciments may in fact result from undetected contamination of the ancient specimens by modern DNA, and the problem is particularly challenging in studies of human fossils. Doubts on the authenticity of the available sequences have so far hampered genetic comparisons between anatomically archaic (Neandertal) and early modern (Cro-Magnoid) Europeans.

Methodology/Principal Findings

We typed the mitochondrial DNA (mtDNA) hypervariable region I in a 28,000 years old Cro-Magnoid individual from the Paglicci cave, in Italy (Paglicci 23) and in all the people who had contact with the sample since its discovery in 2003. The Paglicci 23 sequence, determined through the analysis of 152 clones, is the Cambridge reference sequence, and cannot possibly reflect contamination because it differs from all potentially contaminating modern sequences.

Conclusions/Significance:

The Paglicci 23 individual carried a mtDNA sequence that is still common in Europe, and which radically differs from those of the almost contemporary Neandertals, demonstrating a genealogical continuity across 28,000 years, from Cro-Magnoid to modern Europeans. Because all potential sources of modern DNA contamination are known, the Paglicci 23 sample will offer a unique opportunity to get insight for the first time into the nuclear genes of early modern Europeans.

Link

20 comments:

Maju said...

Fascinating. Finally clear direct evidence that mtDNA H has been in Europe since so long ago.

Any idea if the individual belongs to a Gravettian or still Aurignacian culture?

Crimson Guard said...

Its a small matter, but it seems there's a slight discrepancy(?) in the timeframe between the two articles, one says 24,000 the other 28,000 years.

http://dienekes.50webs.com/blog/archives/000200.html

How much does this blow out C. Loring Brace comment,"If this analysis shows nothing else, it demonstrates that the oft-repeated European feeling that the Cro-Magnons are ‘‘us’’ is more a product of anthropological folklore than the result of the metric data available from the skeletal remains."?

His own data suggested otherwise too, showing them more close to Southern Europeans like Greeks.

Maju said...

I've just checked Julien Riel-Salvatore's blog entry on this cave (from 2006, http://averyremoteperiodindeed.blogspot.com/2006/10/new-paper-on-pagliccis-eup-levels.html) and he mentions 28,000 BP uncalibrated for the Gravettian layer. He also mentions that there are other Gravettian findings in Veneto dated to 30-31,000 BP (again uncalibrated).

So guess my question can be replied with a "most likely Gravettian".

miz RAND BLOWTON said...

Wow! I didn't know lab workers could possibly contaminate sample DNA with their own DNA.I mean,I guess it could happen ,but I would think it a rare occurance.Also wouldn't there be a difference in the age of the DNA samples that are ancient and recent?

Maju said...

Wow! I didn't know lab workers could possibly contaminate sample DNA with their own DNA.

It has been a major concern with all aDNA research so far, specially when the results happened to be close to modern DNA, as in this case. In specimens discovered long ago such checks may not be made at all (or would be extremely difficult in the best case) but Paglicci is a relatively recent find and that helps.

In another case (in Denmark, if I don't recall badly) strict asepsia procedures were taken since the very discovery of the remains at the site in order to ensure zero contamination.

Also wouldn't there be a difference in the age of the DNA samples that are ancient and recent?

I think you are mixing pears and oranges here: the age of a haplogroup (an estimate) and the age of a specific individual sequence (that is always that of the individual). They have not much to do with each other.

terryt said...

Maju: " "most likely Gravettian". Certainly. Presumably this is the oldest surviving mtDNA line in Europe. The evidence suggests any earlier 'modern' human mtDNA lines (along with Neanderthal ones) have become extinct.

Maju said...

Presumably this is the oldest surviving mtDNA line in Europe. The evidence suggests any earlier 'modern' human mtDNA lines (along with Neanderthal ones) have become extinct.

Not at all. Haplogroup U is generally thought to be older, of Aurignacian origin in the case of the European subclades. Some of these subclades make up now the second most common clade (among all them) in Europe. U5 and the rare U8a clade are generally believed to that old. Other cases could be too IMO.

Crimson Guard said...

I think we have to go with what is provable rather than hypothetical. Haplogroup information is always changing the more we find out out on them through various methods and means. Its the same thing with how R1b is much younger than previously stated and can not be associated with Cromagnon's anymore.

Ebizur said...

Perhaps mtDNA haplogroup HV (including its subclades, H and V) dispersed together with Y-DNA haplogroup IJ (and its subclades, I and J).

There is no logical reason for maju to take this finding of haplogroup H mtDNA in Paleolithic remains from Italy as evidence of the presence of Y-DNA haplogroup R1b in Paleolithic Western Europe.

pconroy said...

WOW, this is fascinating.

So "Paglicci 23" has CRS at HVS-1. My wife, whose maternal side is Sicilian, also has CRS at HVS-1...

So I welcome "Paglicci 23" as family!

Maju said...

Its the same thing with how R1b is much younger than previously stated and can not be associated with Cromagnon's anymore.

I think you are wrong in accepting Y-DNA estimates as something real. Can you explain me how O subclades are like 30,000 y.o. and R subclades are much younger? There are too many inconsistencies in MCH estimates. Archaeology is more solid (specially in Europe) and a total replacement, even if only by the male side, looks extremely unlikely.

Only future research will tell for sure anyhow.

There is no logical reason for maju to take this finding of haplogroup H mtDNA in Paleolithic remains from Italy as evidence of the presence of Y-DNA haplogroup R1b in Paleolithic Western Europe.

I have been re-reading carefully all my comments in this thread and I did not say that. I haven't mentioned Y-DNA at all until this very post.

I do think that R1b is of about that age and that this clade looks like very much parallel to mtDNA H in distribution and structure. But I have not made that claim before you raised the issue (in his topic at least).

I wonder what you two were thiniking about when replying to something I didn't say. Not here at least.

...

So "Paglicci 23" has CRS at HVS-1. My wife, whose maternal side is Sicilian, also has CRS at HVS-1...

So I welcome "Paglicci 23" as family!


Congratulations! :)

Ebizur said...

maju said,

"I wonder what you two were thiniking about when replying to something I didn't say. Not here at least."

http://leherensuge.blogspot.com/

Ebizur said...

In his comment on his own blog entry at http://leherensuge.blogspot.com/2008/07/gravettians-belonged-to-haplogroup-h.html, maju said,

"Anyhow I am persuaded that R1b is parallel to H in its arrival and spread. Its very similar distribution, its almost identical starlike structure... all points to the same process. We just need to wait for some aDNA testing in this sense to reach the same level of certainty."

Maju said...

Hehe! I knew I had written that (or something of the like) I just wondered why that discussion had been transfered here. It's not like I eat commenters for breakfast, I think.

But, well, guess it doesn't really matter. It is a worthy discussion but I don't think we will reach to any agreement until Y-DNA is tested in ancient remains too.

That's why I still think this finding is most important: at least for female ancestry there not anymore big doubts: most European mtDNA is quite clearly of Paleolithic origin.

This brings me to Ebizur's suggestion: Perhaps mtDNA haplogroup HV (including its subclades, H and V) dispersed together with Y-DNA haplogroup IJ (and its subclades, I and J).

J is massively thought to be of Neolithic (West Asian) origin. But I is not: most authors seem to think of it as European because of its distribution.

But think about this: mtDNA H clearly decreases in density towards SE Europe, precisely the area where haplogroup I is most common (excepting Sweden maybe?).

I must say I don't have a too clear opinion on Y-DNA I but considering it's rather rare in the western European "refuges" (Basque Country, Ireland, Britain and Iberia), where the original West European gene pool may have been preserved better, I tend to think of it as fundamentally Central/SE European with expansion along the Neolithic and post-Neolithic E>W migration trends.

Also, unlike with mtDNA H, you don't see any subclade of Y-DNA I that has greater concentration or diversity in the SW and that can be related to the Magdalenian expansion from the Franco-Cantabrian province.

I have pondered about this hypothesis and I really cannot make much sense of it, sorry.

So I am left with R1b (or rather R1b1b2) as only possible actor for that apparently well documented post-LGM Magdalenian expansion. It is that or nothing. And I just cannot believe it's nothing left in the Y-DNA side when so much is in the mtDNA.

The most probable scenario, no matter how much I consider other posibilities, is that R1 or R1b arrived at Europe with Gravettian and that R1b1b2 (former R1b1c) expanded from the SW with Magdalenian.

Sorry if that doesn't fit some age guesstimates. It's the last factor I put in the equation, as it's a very unreliable kind of "data".

terryt said...

Maju asked, "Can you explain me how O subclades are like 30,000 y.o. and R subclades are much younger?" What do you see as the problem?

Dienekes said...

That's why I still think this finding is most important: at least for female ancestry there not anymore big doubts: most European mtDNA is quite clearly of Paleolithic origin.

The CRS doesn't occur in most Europeans, so you can't conclude this, at least on the basis of this data.

Maju said...

@ Terry:

Maju asked, "Can you explain me how O subclades are like 30,000 y.o. and R subclades are much younger?" What do you see as the problem?

That R and O, or P and NO are exactly parallel in the Y-DNA tree. They shoul have evolved in similar timelines, maybe not necesarily exact but close to each other.

Also earlier other authors claimed O to be "very young". Things change too much in this issue of age guesses depending who you read. There's no consistent system, to many unknown variables, and often it just doesn't make any sense with each other nor with what we know of the archaeological record.

The CRS doesn't occur in most Europeans, so you can't conclude this, at least on the basis of this data.

Haplogroup H. CRS is just a technical reference sequence within it.

What matters anyhow is that haplogroup H was part of the European genetic pool some 28,000 years ago already. That it's not something that arrived who-knows-how in the middle of the late UP. Now it has at least one known archaeological member - there were probably many more though.

It means that any estimate that places the arrival of H to Europe after that date (for instance Richards-2000 suggested a date of c. 16,000 BP) are wrong.

This finding also ratifies integrative multidisciplinary common sense over fanaticism of MCH age estimates.

cacio said...

Strictly speaking, CRS seems to have been the ancestral value for all R subhaplogroups, so we cannot really equate CRS with H. In a previous paper, the author found another CRS at the cave and determined that it was HV (the other skeleton was N* or N1*). So I guess what we can conclude is perhaps that HV types have been around for a while in Europe, though whether they are H or not is not yet clear. Also, it is not clear whether Paglicci, in southern Italy, is representative of European cro-magnons.

cacio

Anonymous said...

I have this to say: Cro Magnons only exist in France, found under a rock shelf at Cro Magnon. Those Italian remains are Cro Magnoid that is similar to Cro Magnon remains found in France.

Also mtDNA H originated in the Middle East. The date of origin is 25,000 years ago. The haplogroup is younger than the purported age of those Italian skeletal remains. In addition the CRS sequence of mtDNA H, mtDNA H2b, is much younger than 28,000 years old.

The dates are way out. It tells me that those remains have been handled badly since they were found in that cave. The result probably belongs to some archaeologist, or osteologist or some museum worker i.e a modern human born in the 20th century C.E.

Maju said...

Also mtDNA H originated in the Middle East. The date of origin is 25,000 years ago. The haplogroup is younger than the purported age of those Italian skeletal remains. In addition the CRS sequence of mtDNA H, mtDNA H2b, is much younger than 28,000 years old.

The dates are way out. It tells me that those remains have been handled badly since they were found in that cave. The result probably belongs to some archaeologist, or osteologist or some museum worker i.e a modern human born in the 20th century C.E.


Whoa! That's the kind of attitude I'd expect from a true believer in the molecular clock hypothesis. Thanks for such a great example of MCH fanaticism: it is very illustrative.

Sadly, faith and science do not go well along. What I think instead is that it is a good evidence that the current MCH paradigm is largely wrong, whatever the reasons. Even if the molecular clock is ticking, it seems to be doing at a different pace than expected and/or that the assumptions for a late OOA (or also a late Pan-Homo split) are just plainly wrong.

Haplogroup H has been detected(aboundantly, dominant) in other European Paleolithic populations (albeit of later date), just for the record.