In two previous posts (Part 1 and Part 2) we looked into the Alcohol and Aldehyde Dehydrongenase (ADH and ALDH respectively) enzimes that metabolize alcohol in humans.
We noticed that they appear in all human populations but, in different genetic variations, which seem to be distributed globally following certain dispersal routes, acquiring higher frequencies among some peoples while in others they are absent or found at lower frequencies.
These variations in genes in turn affect how these people process their alcoholic intake. Some may have very ancient origins while others seem to be quite recent. Today we will present the final data and complete this third and last post of this series.
ADH and orthodoxy
A paper that is now 11 years old (Mulligan et al., 2003) [1] types Alcohol Dehydrogenase (ADH) and looks into the differente alleles for various populations. As expected, these are found in different frequencies for ADH1C / ADH1B among Amerindians, Asians and Africans (other populations were not included in this paper).
I was surprised at how the authors "edited" their data: (Bold mine) [1]
"For each of the five studied populations, all inferred haplotypes were removed if they did not significantly improve the model as assessed by a log likelihood ratio. Based on this criterion, the following numbers of unconfirmed haplotypes were removed from each population: American Indian, 3; Siberian, 2; Mongolian, none; Chinese, none; Nigerian, none.
The American Indian population had more unconfirmed haplotypes than the others, most likely because it had the largest sample size and contained relatives. However, all of the unconfirmed haplotypes were rare and increased the likelihood of particular individuals only trivially."
They used a program (MLOCUS) which conducts a probabilistic analysis and assigns log likelihood values, so as we can see above, they "removed... haplotypes" because they "did not significantly improve the model" or "increased the likelyhood... only trivially" [1].
Rare Amerindian haplotypes were therefore discarded (and in a high proportion compared to other populations), this surely biased the outcome, moreso since their impact on the "model" was minimum.
Then we must add the assumption that African and Eurasian genes are much older than Native American ones:
The assumption that Amerindians are "young"
Mulligan et al., conclude "No evidence of recombination was detected at the ADH or ALDH loci in the American Indian population. One Chinese and three Nigerian ADH haplotypes appear to have arisen by recombination and/or gene conversion, a result that is consistent with a deeper evolutionary history in these populations compared with American Indians." [1]
The fact is that the Chinese recombination is the one that originates Ht 6 affecting 8% of the Chinese population. It is not "deep" (i.e. old, archaic, ancient), actually this
Ht6 "possessed the ADH1B*47His allele" which, as we have seen (Part 2 of this post) is very recent among Chinese:
Hui Li, et al., 2011 [2] gives the following dates for the ADH1B*47His alleles (this paper's nomenclature for them is H6 and H7): "The estimated ages of H6 and H7 both indicate relatively recent coalescents or expansion times of the haplogroups.The age of H7 is estimated at only around 2.8 thousand years... This young age is unexpected..." [2]
Clearly recent, so why do Mulligan et al. asume that the allele is "ancient"? [1]
I believe that they are trying to fit the paper to adjust to the current orthodoxy of a late peopling of America, therefore Asian lineages must be ancient and Amerindian ones young.
To do so, they also "manually" fit the Ht 7 found in Africans (Nigerians), which they believe to be the "ancestral" haplotype, between two Amerindian alleles, suggesting it is the most parsimonious setting for it; then they derive both Amerindian alleles, Ht 2 and Ht 3, from it; the write (my comments bold between brackets) [1]
Alternatively, Osier et al. (2002) inferred a haplotype in two American Indian populations that would reverse the order of ADH1C HaeIII and ADH1C Ile349Val.
[exactly the opposite to what Mulligan et al. suggest!] Osier et al. (2002) also inferred five additional ADH haplotypes present at low frequencies in four American Indian populations.
Two of Osier and co-worker’s (2002) haplotypes, including the one that would reverse the order of ADH1C HaeIII and ADH1C Ile349Val, were removed from our dataset
[they deliberately removed the data that contradicted their assumption] based on insignificant improvement of the model, suggesting that a more minimal set of haplotypes may
exist for the populations investigated in Osier et al. (2002). [1]
What did Ossier et al. find?
Since Mulligan et al., mentioned Ossier et al., 2002, [3] we will take another look at Ossier & team's work (basically Table 4 and pp. 95):
Ossier et al., defined an Ancestral haplotype (212111) [corresponding to Ht 2 in Mulligan et al.], found in all populations aorund the world, with some minor exceptions, at relatively low frequencies (those with the highest value are shown in brackets): Africans: 0 - 10.6%, Europeans: 0.7 - 30.7% [Finns], East Asia: 0 - 9.8%, Pacific-PNG: 14.6 - 38%, Siberia: 13.1%, N. America: 11.7 - 48.1% [Mexican Pima], S. America: 3.2 - 10.8%.
I wonder why do Finns and Mexican Pima have such high frequencies
The Ancestral haplotype then mutated into two alleles:
- 112111 [Ht 1 in Mulligan et al.] This is the ADH1C HaeIII site-absent allele, and it differs from the ancestral haplotype only at the ADH1C EcoRI site. (it is what Mulligan et al. named Ht1), which is found in most populations around the world and very common in Europe: Africa: 2.8 - 17.2%, Europe: 5.0 - 30.8% [Basque], East Asia: 0 - 14%, nil in Pacific - PNG, Siberia: 14%, N. America: 2.3 - 21%, S. America: 6.9 - 22.7%
- 211111 [Ht 7]. Originated by "an independent mutation of the ADH1C Ile349Val site on the ancestral haplotype... This haplotype is rarely seen today but is present in the !Kung San (8.6%), Biaka Pygmies (10.8%), and African Americans (1.2%)". Some outliers: Yakut 1.3%, Micronesians 3%, Danes 1%, Irish 0.8%, San Francisco Chinese 1%. Zero elsewhere.
- 221111 [Ht 3], is the mutation of 211111 and is "most common around the world" :
Africa: 35.7 - 87.5%, Europe: 1.9 - 39%, East Asia: 9,4 - 20.5%, Pacific - PNG: 23.4 - 42.3%, Siberia: 42.4%, N. America: 28.6 - 68.3%, S. America: 64.9 - 82.3%. Note the very high frequencies in South America compared to much lower values in Asia - Siberia.
Who is correct? Mulligan or Ossier? You can decide based upon the evidence, please check both papers.
Unsurprisingly Mulligan et al. hint that the "difference in haplotype distribution may reflect the fact that different Asian and African populations were analyzed by Osier et al. (2002)." They also recognize that founder effects and possible population bottlenecks may have influenced Amerindian haplotype frequencies.[1]
Amerindian oddities
Nevertheless, the striking facts are that American Indians have, for both Mulligan and Ossier some peculiarities:
- The highest frequency of Ht 2 in the whole world (this is the Ancestral allele)
- The highest frequency of Ht 1 in the world (and this is only one mutation away from the ancestral allele, so it is evidently "old" too)
- The highest frequency of Ht 3 in the world if we consider only South American Natives (78.4%) or second highest 62.3% (for both North and South Amerindians) vs. 65.1% for Sub Saharan Africans. This is the most common allele worldwide, yet its highest frequency is found in Africa -cradle of Mankind- and... of all places, America!
An explanation is that a bottleneck effect in America eliminated some alleles leaving others, which later expanded to fill the void, hence a larger frequency of some alleles in comparison to the rest of the world.
I find this difficult to believe because there are some extremely rare alleles are present in America which are only found in Africa and in the Pyrenean foothills in Spain. A bottleneck would have eliminated these too, how did they survive? Were they present in even larger numbers before the bottleneck? Do we see the imprint of a once larger population? (I do believe that European contact in the 1500s wiped out a large number of alleles unique to Amerindians. So the "lack of diversity" does not mean a "recent" origin of Amerindians, it is merely the outcome of attrition due to disease and war).
Below are some Rare alleles and their frequencies [3]:
- 111111: 1.1% Karitiana, 9.2% Maya, 1.1% Arizona Pima. While in the rest of the world it is only found among: Basque 1.6%, North Moroccans 2.2%, !Kung San 2.3% and S.E. Bantu speakers: 1.4%. Zero in Asia, Siberia and the rest of the world.
- 121111: 1.1% Karitiana, 1.5% R. Suri, and (again) 1.6% Basque, 0.7% Catalans, 0.6 - 8.9% North Africans, !Kung San 2.6% and S.E. Bantu speakers: 4.2%. Zero in the rest of the world.
We could argue that Spanish genes admixed with those of Amerindian after the discovery of America: this would account for Catalan and Basque alleles. Additionally Moroccan genes surely got into the Spanish genome during the Moorish occupation of Spain (711 - 1491 C.E.), but, what about the Sub Saharan Africans? Did slave trade introduce !Kung San genes into America? I find it unlikely. Actually you would expect other African genes but not !Kung San genes.
Now, since Ht 1, Ht 2 (ancestral allele) and Ht 3 are all found in Africa you might expect them to have originated there. But the !Kung San do not carry Ht 1 and Sub Saharan Africans frequency for it is very low (6.1%), but not so in North Africa (13.6%) or Middle East (12%) so perhaps it originated out of Africa and back-migrated later, skipping East Asia and Siberia which also have low frequencies (3.7%), it is absent in PNG and Micronesia, but remained strong in America (11.8%) and in Europe (27.3%) where it probably originated.
Could they be a Denisovan allele? Not likely since it is not found among Papuans and they have the highest Denisovan admixture.
Perhaps it is Neanderthal. If so, its low rate among East Asians (which should have a very high Neanderthal admixture) is easily explained due to the exponential growth of the recent Ht 6 and Ht 7 haplotypes in that area, which were positively selected at the expense of the other alleles.
Closing Comments
As a summary of this "three part" post, I will highlight two things:
First; there is a tendency to "fit" the data to corroborate the orthodox theory of a late peopling of America by a small group of people with limited genetic diversity (founder effect and bottleneck). These then expanded in the New World filling it with people with a very different mix of genes than those found in the Old World.
This is akin to Astronomy before the "Big Bang" theory (the age of the Universe and the cosmological constant were "adjusted" to fit the prevailing theories) or Geology before Plate Tectonics.
Second, the different alleles found among American natives (and this is even more noticeable among South American natives, since North American ones have the imprint of a recent Asian migratory wave) are not similar to those found in East Asia. Some appear to be more similar to those of Caucasians or Africans than to those of Asians, which is unusual since one would expect Amerindians to resemble their supposed Asian ancestors.
Post discovery admixture, bottlenecks and founder effects are used to explain away these differences, but it is highly probable that a stronger archaic admixture is found among Amerindians than elsewhere. Probably due to an Admixture that took place in America, during a peopling event that predated the appearance of Modern East Asians or Siberians.
Part 1 Part 2
Sources
[1] Connie J. Mulligan, et. al., (2003). Allelic variation at alcohol metabolism genes (ADH1B, ADH1C, ALDH2) and alcohol dependence in an American Indian population. Hum Genet (2003) 113 : 325–336 doi: 10.1007/s00439-003-0971-z
[2] Hui Li, et al., (2007). Geographically Separate Increases in the Frequency of the Derived ADH1B*47His Allele in Eastern and Western Asia. Am. J. Hum. Genet. 2007;81:842–846. doi: 10.1086/521201
[3] Michael V. Osier, Andrew J. Pakstis, David Goldman, Howard J. Edenberg, Judith R. Kidd, and Kenneth K. Kidd., (2002). A Proline-Threonine Substitution in Codon 351 of ADH1C Is Common in Native Americans.
doi: 10.1097/01.ALC.0000042013.13899.75, Alcohol Clin Exp Res, Vol 26, No 12, 2002: pp 1759–1763
Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2014 by Austin Whittall ©
No comments:
Post a Comment