April 27, 2012

Ancient mtDNA from hunter-gatherers and farmers of Northern Spain (Hervella et al. 2012)

Having just finished writing my mega-post on the new exciting ancient DNA results from Sweden, a new paper has caught my attention which presents data from Paleolithic-to-Bronze Age inhabitants of northern Spain. Since this is an open-access paper, I'll let you read it, and post my own comments later as an update to this post.

UPDATE: Just three observations on the study:

  1. 50% of the hunter-gatherers belonged to haplogroup U subclades, hence confirming for this region as well that this lineage was over-represented in pre-farming populations; the remainder were assigned by the authors to H subclades on the basis of HVR polymorphisms.
  2. Haplogroup V that had been proposed as having originated in the Franco-Cantabrian region is again conspicuously absent in the ancient data
  3. It now appears that the N1a dominance in LBK samples (and in a French Megalithic) was not shared by all Neolithic groups, with no N1a turning up in the Spanish samples

From the paper:
The Neolithic sample of France (NEO_FR) is closer to present-day populations in the Near East, because it shows similar frequencies for haplogroups TX, W, J, H and U (the ones with the highest correlation for the first axis in the PCA, data not shown). However, the Neolithic populations from the Iberian Peninsula (NEO_CAT and NEO_NAVARRE) are located between the variability of present-day European populations and those in the Near East. Likewise, the Chalcolithic populations in the Basque Country (Longar, SJaPL and Pico Ramos) occupied a similar position to the Neolithic groups of the Iberian Peninsula (Figure 2).
The results from this paper are hence in agreement with lots of other lines of evidence pointing to an exogenous population element in the formation of the Neolithic in Southwest Europe as in Scandinavia.
The authors attribute this to maritime colonization:
Maritime colonization, transporting small and different Neolithic groups from the Near East pool could contribute to explain the difference.
Hopefully, we will be able to obtain autosomal DNA from some of the Neolithic samples as we did in Scandinavia, and I would be much surprised if these Neolithic inhabitants of Iberia were not Oetzi-like and hence intermediate between most modern populations of Southern Europe and the Near East.

PLoS ONE 7(4): e34417. doi:10.1371/journal.pone.0034417

Ancient DNA from Hunter-Gatherer and Farmer Groups from Northern Spain Supports a Random Dispersion Model for the Neolithic Expansion into Europe

Montserrat Hervella et al.

Abstract


Background/Principal Findings

The phenomenon of Neolithisation refers to the transition of prehistoric populations from a hunter-gatherer to an agro-pastoralist lifestyle. Traditionally, the spread of an agro-pastoralist economy into Europe has been framed within a dichotomy based either on an acculturation phenomenon or on a demic diffusion. However, the nature and speed of this transition is a matter of continuing scientific debate in archaeology, anthropology, and human population genetics. In the present study, we have analyzed the mitochondrial DNA diversity in hunter-gatherers and first farmers from Northern Spain, in relation to the debate surrounding the phenomenon of Neolithisation in Europe.

Methodology/Significance

Analysis of mitochondrial DNA was carried out on 54 individuals from Upper Paleolithic and Early Neolithic, which were recovered from nine archaeological sites from Northern Spain (Basque Country, Navarre and Cantabria). In addition, to take all necessary precautions to avoid contamination, different authentication criteria were applied in this study, including: DNA quantification, cloning, duplication (51% of the samples) and replication of the results (43% of the samples) by two independent laboratories. Statistical and multivariate analyses of the mitochondrial variability suggest that the genetic influence of Neolithisation did not spread uniformly throughout Europe, producing heterogeneous genetic consequences in different geographical regions, rejecting the traditional models that explain the Neolithisation in Europe.

Conclusion

The differences detected in the mitochondrial DNA lineages of Neolithic groups studied so far (including these ones of this study) suggest different genetic impact of Neolithic in Central Europe, Mediterranean Europe and the Cantabrian fringe. The genetic data obtained in this study provide support for a random dispersion model for Neolithic farmers. This random dispersion had a different impact on the various geographic regions, and thus contradicts the more simplistic total acculturation and replacement models proposed so far to explain Neolithisation.

Link

23 comments:

truth said...

So we have two Paleolithic mtDNA H, one belongs to H6 (sample CH) the other to H (sample PS1) ) (this last one can only be seen in the Word document in the Supp.Info).
And the Iberian Hunter-Gatherers are mot similar to those from Scandinavia.
The Neolithics are not at all similar to those of Central Europe (mainly due to lack of N1), and are most similar to those of France and Catalonia.

Maju said...

Wanted a full fledged peer-reviewed paper that reports mtDNA H among pre-Neolithic Western Europeans? Here you have one.

It is interesting that while the Basque Magdalenian/Epipaleolithic sites are in the line of Central Europe: no apparent H, loads of U (U5 essentially), the Cantabrian ones are 100% H, which is consistent with what was as reported for Epipaleolithic (and Neolithic) Portugal (Chandler 2005) and separately for Taforalt in North Morocco.

So far the only more or less clearly reported Paleolithic mtDNA H is in West Iberia and North Morocco. There are some sequences around Europe anyhow that may well be that (notably Sunghir in Russia) but there's always the doubt.

Interestingly not a single mtDNA H has been reported with any certainty yet outside Europe (or in the Balcans) in Pre-Neolithic contexts.

I'm more and more flirting with the notion of a demic expansion associated with Dolemenic Megalithism, expansion originated in SW Iberia, what is now Portugal "grosso modo". This expansion could explain at least in part the adjustment of mtDNA pools after the early Neolithic in favor of haplogroup H (very specially H1).

Dienekes said...

Wanted a full fledged peer-reviewed paper that reports mtDNA H among pre-Neolithic Western Europeans? Here you have one.

The HVR is not sufficient to assign samples to haplogroups. Still, 50% in haplogroup U for the pre-Neolithic samples is consistent with what we know about pre-Neolithic Europeans.

Dienekes said...

And the Iberian Hunter-Gatherers are mot similar to those from Scandinavia.

What are you talking about? There are no significant differences between the Iberian HG and the Central European/Scandinavian ones, whereas the latter have significant differences with all populations.

jeanlohizun said...

This is what the authors said:

"Sequencing of mtDNA HVR-I, nucleotide positions (nps) 15,998–16,400, and mtDNA HVR-II, nps 16504-429 as per [61], was undertaken in six overlapping fragments, each with a length of approximately 100 bp (base pair). HVR-II sequencing was carried out in samples with no polymorphic positions in HVR-I (Table S6). Similarly, the fragment between primers 8F and 8R (Table S6) was amplified in all samples to determine position 73 of HVR-II."

As far as I know, when a haplotype that has the rCRS in the HVR-I is tested in the position 73 of the HVR-II it is enough to determine if this is an H or a U. Well Table-S1 clearly shows that the sample from La Pasiega, Cantabria is an H.

As for the frequency, we have 3 Magdalenian sites(i.e. 12,000 ybp old), and 1 Mesolithic site(6900 ybp old). The Magdalenian sites are 2/3 H, 1/3 U. The Mesolithic site is 100%U.

truth said...

@Dienekes

It's not me, It's what the study says, I copy it directly :

"The three groups of hunter-gatherers considered in this analysis (from Scandinavia, Central Europe and the Cantabrian fringe on the Iberian Peninsula) did not show statistically significant differences between one another. However, the hunter-gatherers from Scandinavia (HG_SCA) and those from Central Europe (HG_CE) showed statistically significant differences with other Neolithic, Chalcolithic and present-day populations. (HG_CANT)"

Dienekes said...

I am not going to repeat the whole discussion on phylogenetics that we've already had in a different thread.

The fact is that a Magdalenian sample is U, a Mesolithic sample is U, and Magdalenian samples from Germany are also U (Bramanti et al. 2009).

So, while it's not impossible that these two Paleolithic samples are in fact H, the evidence is not really strong enough to make that claim with certainty.

jeanlohizun said...

So basically checking the HVR-II region 73 is enough to determine whether if a haplotype that is rCRS is U, but not enough to determine if is H.

I think the evidence is very strong to support that both Magdalenian samples from Cantabria are H. Why? Because if the evidence isn't strong, then it isn't strong to support that the Magdalenian samples from Guipuzcoa or Germany are U. It doesn't work like that Dienekes.

PS: They did test for the SNPs in the coding regions.

"Sequencing of mtDNA HVR-I, nucleotide positions (nps) 15,998–16,400, and mtDNA HVR-II, nps 16504-429 as per [61], was undertaken in six overlapping fragments, each with a length of approximately 100 bp (base pair). HVR-II sequencing was carried out in samples with no polymorphic positions in HVR-I (Table S6). Similarly, the fragment between primers 8F and 8R (Table S6) was amplified in all samples to determine position 73 of HVR-II.
...

In order to classify the mitochondrial variability of the individuals analyzed in this study, we proceeded to amplify 11 markers, which are required for defining the 10 Caucasian haplogroups described [63]. The protocol and primers are described in [17], [20], [42].
"

Maju said...

"... it's not impossible that these two Paleolithic samples are in fact H"...

I want to see in this sentence an advance in the way you perceive the factual data. Hope so.

There is not a single H sample from West Asia nor the Balcans before Neolithic. It's not impossible either... of course.

"... the evidence is not really strong enough to make that claim with certainty".

I believe that you are familiar with the English idom: "grasping for straws", not sure how it may be in Greek.

"The HVR is not sufficient to assign samples to haplogroups".

HVS-II certainly excludes U. I have yet to look in detail at the haplotypes and I'm sure I will disagree with some of the assignation but in this case as in Chandler 2005, the HVS-II certainly makes U-CRS an impossible result.

"Still, 50% in haplogroup U for the pre-Neolithic samples is consistent with what we know about pre-Neolithic Europeans".

True. Actually 100% of Basque samples but 0% of the Cantabrian ones. They are only 2 and 2 but I find the appearance of a clear cut line very interesting. More so when the Cantabrian facies of Magdalenian is more related to Asturias and this one may have got a genetic legacy from Iberian (not Basque but Iberian proper) Solutrean. The consistency of lots of H with Portuguese and Moroccan samples (as reported) is very interesting, suggesting to my mind the possibility that Western Europe was partly recolonized in the Neolithic/Chalcolithic from Iberia (Dolmenic Megalithic culture most likely).

Just a hypothesis right now but a most intriguing one to my eyes (and one that would fit well with ALL the reported ancient DNA).

eurologist said...

I think the interpretation in this paper is rather weak - they basically start out with the idea that anything "exotic" must be from the Near East.

The lack of N1a can again be explained if we assume that it was local Balkan, and as such picked up with LBK (as, documented, many other "exotic" mtDNA haplogroups) - but not by Cardium. As to the U/H (and perhaps V) conundrum in Franco-Cantabria, it could be that "U" were the last ones to seek refuge during LGM, and (like a spring) the first ones to leave after LGM.

Unknown said...

As I see there is no dating for La Chora and La Pasiega Magdalenian sites. If so, the right thing to say is: "mtDNA H samples were found in hunter-gatherer contexts", "Upper Paleolitic" is misleading. It's a pity, again Iberian aDNA paper with undated "paleolithic" mtDNA H samples.

Maju said...

I have asked the authors of the paper and both Concepción de la Rúa (corresponding author) and Montserrat Hervella (lead author) answered that they had indeed tested for RFLPs and that the reported mtDNA H is certainly H per -7025a Alu I.

There is no reasonable doubt anymore: mtDNA H (and H6) existed in Magdalenian Cantabria some 15 Ka ago.

As for what "Unknown" objects, we have to realize that the authors are just deferring to other papers for a lot of the background data (for example the RFLP sequencing was dispatched in a single short paragraph with three footnotes) just because they seem to be used to write in that less-explicit somewhat cryptic manner, so I'd suggest to track the references or just search online for the exact archaeological data on La Pasiega and the other cave.

But if they say "Magdalenian" it is from c. 17 to c. 11 Ka ago. No kidding.

Dienekes said...

@Unknown,

Haplogroup H originated around ~13,000 years ago, so it is quite unlikely that it would have been sampled in the context in question. The only other research on Magdalenian mtDNA from Hohler Fels turned up haplogroup U, as did 2 of the 4 hunter-gatherer samples from Iberia and the great majority of samples from Central Europe and Scandinavia.

In short, I am quite skeptical of this result, and in all probability it will turn out to be either contamination or misplaced archaeological context.

The only references provided in the paper for the two offending samples are dated (1952) and (1963) which certainly does not exactly increase my confidence.

The Hohler Fels sample, on the other hand has been assigned to the Magdalenian on both cultural context and carbon dating of bone, and underwent two extractions.

So, for the time being, the result is noted, but not believed.

Unknown said...

Reference #44 for La Pasiega cave:
González-Echegaray PJ, Ripoll E (1952) Hallazgos en la cueva de La Pasiega. In Ampurias XV-XVI Barcelona.43–65.

Reference #45 for La Chora cave:
González-Echegaray PJ, García-Guinea MA, Gegines A (1963) Cueva de La Chora. Excavaciones Arqueológicas de España, 26: 43–65.

Caves are known for stratification problem, artifacts from different periods could be mixed up.

So, no dating for the artifacts, no dating for the remains from 50-60 years old excavations.

If you want to prove something, do it right way, like the Swedish team:
"The Neolithic farmer sample (Gök4) was excavated from a megalithic burial structure in Gökhem parish, Sweden, and has been directly 14C-dated to 4921 ± 50 calibrated yr B.P. (cal yr B.P.), similar to the age (5100 to 4900 cal yr B.P.) of the majority of other finds in the area (15). There were no indications from the burial context suggesting that Gök4 was different from other TRB individuals (15, 16), and strontium isotope analyses indicate that Gök4 was born less than 100 km from the megalithic structure, similar to all other analyzed TRB individuals from the area (17). The three Neolithic hunter-gatherer samples were excavated from burial grounds with single inhumation graves on the island of Gotland, Sweden, for which associated remains have been dated to 5300 to 4400 cal yr B.P. (16)."

GailT said...

I checked the reference on the 9 RLFPs, they cite this paper http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1377722/pdf/9915963.pdf

and I see 8 SNPs mentioned, 4216
4577
4914
12308
13704
14465
14766
15254

Since they are already going to so much work, why not test the full genome on these samples? That would remove all doubt and also anchor the age estimates.

jeanlohizun said...

Dienekes said:

"In short, I am quite skeptical of this result, and in all probability it will turn out to be either contamination or misplaced archaeological context.
"

The authors went through great lengths to avoid contamination. It is by far more likely that Behar et al(2012) estimated the age of Haplogroup H erroneously, than this samples are either contaminated or not Magdalenian. You have gotten to the point where you are arguing against empirical evidence based on pre-conceive notions.

Dienekes said...

You have gotten to the point where you are arguing against empirical evidence based on pre-conceive notions.

Incorrect, I am sceptical of anomalous and incomplete evidence.

It is by far more likely that Behar et al(2012) estimated the age of Haplogroup H erroneously

Your claim that it is "far more likely" is your opinion. Both Behar et al. (2012) and Fu et al. (2012) are consistent with a late origin of mtDNA haplogroup H. This haplogroup was also not detected in many other hunter-gatherer samples from all over Europe and by many different teams of researchers.

It is indeed a valuable new piece of evidence, but should be evaluated in the context of the totality of the evidence. If mtDNA haplogroup H was indeed in Europe since Upper Paleolithic times, we will surely find it in other places and by other teams. Until we do, "one swallow doesn't make a summer".

Maju said...

But it was indeed detected by some other researchers (which you insist in disdaining) precisely in that same region of Europe/North Africa

It has also never been detected in your alleged areas of origin (West Asia and such) before Neolithic ever, at least not with any certainty.

While I do not think it is the case, it's even possible that H was semi-restricted to parts of SW Europe and NW Africa before Neolithic. Although we still need a source for the H found in Neolithic Central Europe (Austria?)

But in any case this is much more solid than "molecular-clock-o-logy". You're grasping for straws.

G Horvat said...

I would have liked to have seen more mutations; even HVR II. Three sequences, classified as 'H' have C285T which is found in U1 but never (?) in H. These three also happen to have 73G.

Annie Mouse said...

Fascinating to see H6 in a Magdalenian sample. The theory was that arrived in Western Europe recently (~5kya).

The Behar dating of H at 13k is clearly wrong. Sub groups like H6 and H8 have much older calculated dates (~40 kya). H cannot be younger than its daughter haplogroups.

Fanty said...

"The Behar dating of H at 13k is clearly wrong. Sub groups like H6 and H8 have much older calculated dates (~40 kya). H cannot be younger than its daughter haplogroups."

I am pretty sure, whatever calculation led to an age estaminate of 13K for the mother HG, gives even younger dates for the doughters aswell.

When I first heard about "H" ever, it had an age estaminate of 60K years and was guessed to originate in Greece.

But then again, R1b was estaminated to be 50K and to be the dominating haplogroup of the paleolithic western europeans.

Now people claim R1b as possibly as young as 15K years and not even on European soil 5K years ago.

Theories change. At least something is unexpected, exiting and interesting. ;-)

Ponto said...

Mitochondrial haplogroup V is of Neolithic origin, like its close relative H. The dating of haplogroups is the problem.

The problem with R1b being of Pleistocene origin in Europe is the lack of hard evidence. Even haplogroup I, common in Scandinavia and Finland, has questions that are never answered. One question: Why did haplogroup I and haplogroup J, both derivatives of haplogroup IJ get so geographically separated when they have the same parent?

eurologist said...

Why did haplogroup I and haplogroup J, both derivatives of haplogroup IJ get so geographically separated when they have the same parent?

Ponto, you can ask the same question about ED, or many other pairs. The answer is simple: one is representative of the people who stayed behind, the other is representative of people who moved away and (of all that did) were successful.