Mitochondrial DNA haplogroup H structure in North Africa
Background
The Strait of Gibraltar separating the Iberian Peninsula from North Africa is thought to be a stronger barrier to gene flow for male than for female lineages. However, the recent subdivision of the haplogroup H at mitochondrial DNA (mtDNA) level has revealed greater genetic differentiation among geographic regions than previously detected. The dissection of the mtDNA haplogroup H in North Africa, and its comparison with the Iberian Peninsula and Near-East profiles would help clarify the relative affinities among these regions.Results
Like the Iberian Peninsula, the dominant mtDNA haplogroup H subgroups in North Africa are H1 (42%) and H3 (13%). The similarity between these regions is stronger in the North-West edge affecting mainly Moroccan Arabs, West Saharans and Mauritanians, and decreases eastwards probably due to gene flow from Near East as attested for the higher frequencies of H4, H5, H7, H8 and H11 subgroups. Moroccan Berbers show stronger affinities with Tunisian and Tunisian Berbers than with Moroccan Arabs. Coalescence ages for H1 (11 ± 2 ky) and H3 (11 ± 4 ky) in North Africa point to the possibility of a late Palaeolithic settlement for these lineages similar to those found for other mtDNA haplogroups. Total and partial mtDNA genomic sequencing unveiled stronger mtDNA differentiation among regions than previously found using HVSI mtDNA based analysis.Conclusion
The subdivision of the mtDNA haplogroup H in North Africa has confirmed that the genetic differentiation found among Western and Eastern populations is mainly due to geographicalrather than cultural barriers. It also shows that the historicalArabian role on the region had more a cultural than a demic effect. Whole mtDNAsequencing of identical H haplotypes based on HVSI and RFLP information has unveiled additional mtDNA differences between North African and Iberian Peninsula lineages, pointing to an older mtDNA genetic flow between regions than previously thought. Based on this new information, it seems that the Strait of Gibraltar barrier affected both male and female gene flow in a similar fashion.
Pay dirt! I’ve been looking for some kind of study into H in North Africa ever since I saw the Taforalt A-DNA study and noticed the H in it. I was curious how the H appeared in North African Ibero- Maurasian bones before the Capsian era (which seems to be near Eastern in origin) if H doesn’t show the same time depth as X1, U and M1 in North Africa. This allows for the Taforalt remains to have some relatively recent input from the near east, which explains the lesser time depth for H but it’s presence in the 12,000 year old Moroccan bones. Thank you Biomed.
Selected quotes for my own reference:
A principal component analysis (PCA) points to subhaplogroups H1 and H3 as being primarily responsible for the Iberian-Moroccan-Saharan connection, whereas H4, H5, H7, H8 and H11 testify the Near East influence
Thus, our HVSI based coalescence ages for H1 (14.2 ± 3.0 ky) and H3 (10.3 ± 2.6 ky), in the Iberian Peninsula, are very close to those published by Pereira et al. [40] in the same area for H1 (14.0 ± 3.0 ky) and for all of Europe for H3 (11.0 ± 3.0 ky)
From our data, it can be also deduced that the presence of the H1 and H3 subgroups in North Africa could have similar expansion times as in Europe and, therefore, a late Palaeolithic settlement in the region.
As a consequence, it has been proposed that the North African gene pool has had Palaeolithic and Neolithic influences from the East, but that the impact of the historicalinvasions, such as the Arabic role, had more a cultural than a demic effect. The lack of exclusive haplotypic matches between North Africa and the Arabian Peninsula found here is in accordance with that hypothesis.
The H here seems to have the same time depth as the Y chromosome J-m267 in North Africa, so I’m going to assume they made the trip from the near East together. this seems to match the expansion of the ‘escargotiere’ people who ate massive amounts of edible snails, who seem to have an origin somewhere between Southern Turkey and the Levant.
Mathilda's Anthropology Blog. 
But notice this other paper that states that H1 and H3 diversity in Tunisia is a subset of Iberian (as well as younger) and that therefore Iberia has to be considered the radiating center for these North African lineages.
Nevertheless it’s interesting (by the first paper cited in this post) that H7, common in North Africa, cannot be traced to Iberia but has to be of either West Asian origin or a local North African developement. Much of the same can be said of H4 and maybe other rare H subclades with some presence in North Africa (excepted H2a* that also seems better connected to Iberia than to anywhere else).
I was curious how the H appeared in North African Ibero- Maurasian bones before the Capsian era (which seems to be near Eastern in origin) if H doesn’t show the same time depth as X1, U and M1 in North Africa. This allows for the Taforalt remains to have some relatively recent input from the near east, which explains the lesser time depth for H but it’s presence in the 12,000 year old Moroccan bones. Thank you Biomed.
But remember that these age estimates are in fact inconsistent with factual data on haplogroup H in fossil samples, always de facto older than the usual TRMCA hunches. You mention the finding of haplogroup H in Taforalt but you also have the find of CRS-H2 in Paglicci (early Gravettian), when in theory it should not even exist at all (the usual age hunches for H2 are of some 12 kya, 16 milennia after it actually was there). So you can start multiplying all mtDNA age hunches by at least 2.5 in order to get some realistic results.
Instead with my SNP-based approach I gather that R0-HV was expanding (branching out) soon after the main Eurasian pulses ended and even quite parallely to the maybe more complex expansion of U (often considered significatively older, though I can’t see it). This aproach seems to fit better with the factual archaeogenetical data.
you also have the find of CRS-H2 in Paglicci (early Gravettian),
I’m not the only persion that spotted that then. It has to be over 45k old, pre expansion into Europe.
H pre-expansion into Europe? That’s something I do not really see either because most of H diversity is confined to Europe and secondarily North Africa. Only H8 has a really Asian (and Central Asian especifically) distribution.
HV* and R0a are West Asian (so R0 as a whole was West Asian originally) but H (and V) is basically European with a Central Asian branch (one among many) and a strong undifferentiated influence in North Africa (mostly Western clades).
I am not good at judging these items but from the latest R0 paper it would seem that the highest diversity is either in North Caucasus or Central Europe. North Caucasus unclear sample tagged as Karachay-Balkarian (two different ethinicities: one Turkic and the other NW Caucasian) gives peaks for H nucleotide (but not haplotype) diversity but the other Caucasian samples give totally average diversity levels, so guess it could well be an artifact of mixing different ethnicities in the same sample.
Next in diversity is Germany, which is top of the scale both in haplotype and nucleotide diversity, after apportioning to sample (raw numbers are much higher for Austria but the Austrian sample is huge).
H* also appears concentrated in mainland Europe, from France to the Aegean. So guess that a Central European urheimat related to Gravettian expansion makes sense. Aurignacian also spread from Central Europe, closer to the age you mention, but, unlike Gravettian, it did not expand to the East.
A usual problem anyhow is that nobody really knows the origin of Gravettian, which appears first of all in Central Europe but could well have been influenced by the rather continuous Perigordian-like tendencies of West Asia (or vice versa).
Another problem is that, while we can trace Aurignacian quite neatly, we know little of the so-called transitional cultures, often attributed to Neanderthals but with very limited evidence (just one skull finding, that’s all).
But, skipping this, your age hunch of 45 kya for H could well fit with Bacho-Kiro, that many consider an Aurignacian precursor and in any case placed at the necesary passage from West Asia, from where proto-Aurignacian must have arrived (either from West or Central Asia or a mixture of both). So if you’re right, H was already formed in Bachokirian times.
Just noticed that there is a sample tagged “Balcans” that has also a very high level of haplotype diversity albeit not as high in nucleotide diversity. So yeah, maybe you’re right and it was Bachokirian/Aurignacian.
The question may be then: and what abut U? I understand (very provisionally, on my own private research) that U3-4-8-9, U5 and U6 expanded at about the same time that HV did. So guess that they could be Aurignacian too. Like H, Eruopean U also has relatives to the East (U2 and U7), more like towards India than Central Asia but not fundamentally different in their spread (U2 is NW South Asia and southern Central Asia, U7 is like West Asia, Mediterranean and NW South Asia) .
Of course, this understanding requires throwing the usual age hunches to the trash bin and thinking in different, less dogmatic, parameters.
Luis, what a hypothesis! Multiply TRMCA by 2.5 to obtain a true age. Luis you are not a scientist, that is evident from your hypothesis. And your proof is that the age estimate of some Italian bone human bone scraps is 28 ky, and the age of mtDNA H CRS is half that age, therefore chuck all the scientific and statistical work done on estimating age of haplogroups out the window as it does not comply with your ideas of the age of certain haplogroups in Europe.
I doubt the CRS is 45 ky anywhere in the world. The CRS is more common in eastern Europe and further east than it is in Italy or Western Europe indicating an eastern origin, out of the Middle East.
Sorry, but both of you have lost any credibility to me. I accept science’s conclusions until it is disproved scientifically by better science.
A more rational opinion to finding the CRS in the old Italian bone is contamination by archaeologists, osteologists and museum workers. Doesn’t take much to cause dna contamination.
mtDNA H is supposed to have originated in the Middle East about 30 kya.
I am glad that North Africans have a Paleolithic origin and that the Arabians had minimal genetic effects but there is no need to outdo Lysenko.
Before you have a go on about trusting science- bear in mind the the calculated age for one downstream variation of hg H is a lot older than it’s parent and is etimated at 40k or more- so no they are quite often very obviously innaccurate. One estimated date for on M hg in the Far east is way older than the estimate age for M as well.
The Paglicci remains were very carefully tested, and show H at 26k old in Southern Europe. I needs to predate the population movement inot Europe.
“Coalescence ages for H1 (11 ± 2 ky) and H3 (11 ± 4 ky) in North Africa”.
That’s not particularly ancient. A little before undisputed evidence for Mediterranean boating but not too much before. And certainly long after undisputed evidence for open water travel in the Far East.
@Ponto:
… therefore chuck all the scientific and statistical work done on estimating age of haplogroups out the window as it does not comply with your ideas of the age of certain haplogroups in Europe.
Statistical maybe, scientifical not so much, IMO.
It is not that it does not comply with “my ideas” as much as it does not comply even with their own ideas. As Mathilda says, TRMCA age estimates are often contradictory with each other (derived haplogroups that appear much older than their underived ancestors for example) and fail almost systematically to provide age estimates that make any sense with what we know of Prehistory (that in the European case at least it’s a lot).
So no, they are not reliable much less scientifical. For, in order to be scientifical, a theory must make predictions that can be confirmed and this system fails to do that all the time. TRMCA hypothesis is at most scholastic, academic maybe, but not scientifical, as it does not fulfill the requirements of the scientfic method.
The CRS is more common in eastern Europe and further east than it is in Italy or Western Europe indicating an eastern origin, out of the Middle East.
As per the paper mentioned before, H2a is somewhat common indeed in Eastern Europe (but not in West Asia) and is also somewhat common in some Central and Western European areas, notably Galicia. The CRS (H2a2a) especifically was not detected anywhere in fact but its direct ancestor H2a2 was detected only in Cantabria (1 H2a2/52 H/135 total) and the huge Austrian sample (21 H2a2/859 H/2214 total). That means 1.9% and 2.4% respectively (in relation to H, not total sample).
The high frequency of H2a among Eastern Europeans is in fact caused by H2a1 (E. Slavs) and H2a3 (Daghestan), which are slightly more distant from the CRS than (mostly) Central-Western H2a2.
Anyhow, a Eastern European dominance does not mean necesarily (nor likely in fact) a West Asian origin. Eastern Europe was first populated (after some early erratic epysodes like Kostenki) by Gravettians from Central Europe, not from anywhere in West Asia. While we can’t discard some West Asian Neolithic influences in later times, we should not ignore either that the Epipaleolithic of the Zagros (Zarzian) is most likely rooted in Eastern European Epigravettian as well.
Sorry, but both of you have lost any credibility to me. I accept science’s conclusions until it is disproved scientifically by better science.
Obviously I don’t care too much on others’ opinions on me, otherwise I’d be always frightful of thinking on my own. But, regardless, your mouth-filling with the word “science” while rejecting solid, methodic, scientifical research of factual data, backed only by mere algorithmic speculation makes no sense to me.
I’d admit that finding the CRS in such an old person is somewhat surprising (I’d be more confortable with some less derived H* in fact) but facts are facts. If it would have been S or L5, we would have to accept it as well.
mtDNA H is supposed to have originated in the Middle East about 30 kya.
That is not serious. Even ignoring the fancy TRMCA, most H sublineages are found mostly in Europe (the main, albeit only partial, exception is H8, also minor subclades H14 and H21 AFAIK). H* is also found mostly in Europe. R0 and HV are, it seems, West Asian but H (and V) not anymore.
The fact that there has been back-migration from Europe into West Asia confuses everything (for some). Geneticists are for the most part not really good at Prehistory and there is this idea that migrations have always followed an East>West pattern. While there is some truth to that, the opposite is also true (i.e. Europeans migrating eastward and probably southward too), this caused the diversity in West Asia to increase and confuses matters for those that are not careful enough.
Important caveat to our discussion: the “CRS” sequence found in Paglici 23 is not the full CRS but the CRS haplotype within the HVR-I.
This means that most likely Paglici 23 was not H2a2 after all but some sort of H or even maybe HV. AFAIK the CRS haplotype in HVR-I is most commonly found in haplogroup H1 but guess it’s surely common in other H clades and might be even found in ancestor lineages like HV.
…my hypothesis on the forming of the gravettian is that it comes out of our collaboration at close quarters with
neander in the chatelperronian, 35-29k bc,
cf., tzopilotl wordpress. big catskulls adorn
the hide tents of the gravettian, pointing to cave
origin, totem animal of firedrill goddess,
tlatla tzol teotl, closely related to the basque
cavers and shows up in their subsequent
word for witch=tzol-tzintli(N)=sorgin(B),
cf., tletl blogspot for basque hodge-podge.
and that the gravettian gave rise to the north
atlantic solutrean crossing that fathered clovis and the amerindian languages from its
iberian/dorgogne source/tzol/(r)ce(N)=
one hole.
kostenkij/koct(russ)=bone/cotoni(N)=pinch(cut) has needles(therefore, firedrill) 45k bc.
the gravettian gives neander and his cromag allies an exit from a europe filling up with
ooaf(rika)s.
haplo needs linguistic archeology to help it along, otherwise non-productive snarlys
arise, and the subject lies there like a retired
bladder with not enough air to kick it around.
…i am going to my blog, tzopilotl wordpress,
mati, thanks for your hospitality.
“AFAIK
the CRS haplotype in HVR-I ”
Thank you, I was looking all over the web for this info! -Mark
Yes, the report is misleading. The CRS in HVR1 is not particularly significant. The CRS in HVR1 is found in many haplogroups. A search using mitosearch shows the CRS in HVR1 present through a wide representation of existing haplogroups including L1, L3, X, HV, H, V, C, A…..The finding of the CRS in HVR1 basically means absolutely nothing. So it begs the question why was it reported as being significant in an ancient human in Europe living 28 kya?
Well there were also other SNPs that delineated H, It was not just the CVS in HVR1.
And for the record some subclades of H are estimated at over 40,000 BP (eg H6). Haplogroup H can’t be younger than its subclades. The overall age calculations for H are likely distorted by a massive lost of European diversity in the ice age.
Haplogroup H can’t be younger than its subclades
A point I’ve made myself when looking at other age calculations for H. It kind of had to predate the expansion into Europe…