A research paper by Villanea FA, Huerta-Sanchez E, Fox K., (2021) investigated the ABO Genetic Variation in Neanderthals and Denisovans (Mol Biol Evol. 2021 Jul 29;38(8):3373-3382. doi: 10.1093/molbev/msab109. Erratum in: Mol Biol Evol. 2021 Dec 9;38(12):5835. doi: 10.1093/molbev/msab261. PMID: 33892510; PMCID: PMC8321519). This is very interesting, because contrary to what one would expect, these ancient relatives belong to the same ABO blood groups that modern humans have.
Blood Groups: A, O, B
This study looked into the ABO allele (gene) that is located on chromosome 9 and that codes certain enzymes known as glycosyltransferase that in turn modify H antigens on red cells, this defines an individual's blood group: group O individuals have the H antigen, a precursor for A and B antigens, group A has only A antigen, group B carries B antigen, and group AB has both A and B antigens, and as mentioned, group O have neither A or B antigens.
Original and Later A blood group
The original and ancestral A allele appeared ~3.5 million years ago. However, according to Kitano et al., 2012 it vanished around 2 million years (My) ago and remained absent for 1.75 My. The modern A blood group found in humans originates from an ABO*A allele that was created recently in the first H. sapiens some 260,000 years ago, from the fusion of ABO*B with ABO*O (B and O blood group alleles).
O Blood Group
O blood group allele appeared after the Δ261 deletion in the A gene, around 2 million years ago as the O02 variant.
It is therefore older than Neanderthals, Denisovans, and Modern humans. It was first carried by the common ancestor of those hominins. Lalueza-Fox C., Gigli E., et al., (2008) detected genes for blood group O (finding a variant known as O01) in the remains two Neanderthal men from El Sidrón, Spain, dated to ~50.000 years ago (50 kya). But O01 is a "newer" variant, O02 is the basal one.
B Blood Group
The B blood group, according to Gueuning et al., 2023 emrged "from the recombination between ancestral alleles of ABO∗A2 and ABO∗O.O2."
O haplotypes
There are two main branches within the O haplogroup: O01 and O02, and they include sub-branches.
Haplotype O02 is also the most ancestral of the O blood group variants. Kitano et al., 2012 place it as the oldest branch of this group ~2 million yers old, followed by the second variant, O01, 1.5 million years ago.
Recent research conducted by Gueuning et al., 2023 studied ABO blood groups using samples gathered in Zurich, Switzerland (so we can imagine a farily uniform Central European population). I will focus on their findings regarding the O blood group.
They found the following frequencies: ABO∗O.01.01: 40.95%, and ABO∗O.01.02: 22.65%, followed by a very low frequency of the ancestral variant, ABO∗O.02: 2.11%.
Regarding the phylogenetic tree it is shown below adapted from Fig. 3 in Guening et al., 2023:
The tree shows O02 separated from the O01 variant, and within the O01 variant, two separate branches, one O.01.02 more divergent and basal, followed by O.01.01 (which contains to sub-branches).
The authors noted these splits: "we observed in the phylogenetic tree deep splits separating the ABO∗O clades. The split between ABO∗O.02 and ABO∗O.01 alleles was even more pronounced in the phylogeny... The 2 major O-phenotype groups (ie, ABO∗O.01 and ABO∗O.02) were paraphyletic to each other with deep (ie, ancient) splits, showing that these groups are not closely related in evolutionary terms." Furthermore, the younger O01.01 "ABO∗O.01.01 appeared evolutionary closer to the cluster of ABO∗A1, A2, and B than ABO∗O.01.02."
O01.02 and Native Americans
This is relevant and interesting, because the O.01.02 variant represents half of the Amerindian O blood group variants. And O itself is found at very high frequencies in Native Americans (up to 100% in some groups). More on this below.
Native American Anomaly: extremely high O frequency, and their haplotypes
Blood group O is the most frequent blood group in the world. Its approximate global the frequencies are the following (Source): African: 50.2%, East Asian: 40.4%, South Asian: 39.4%, Middle East: 38.3%, European: 42.3%, and the highest prevalence is found among people of Latin American descent, with 57.4%. But Latinos are an blend of Native Americans (the ones that survived the Great Dying caused by the Discovery of America), African slaves and Europeans, so what is the prevalence among genuine or pure Native Americans?
Native Americans
The Native people of South American "are unique in that they have completely lost both the A and B alleles" the O blood group is found at frequencies close to 100% (Source), especiall among South American Native Americans.
The Nomenclature! an equivalence table
As with most things related to genetic names, the nomenclature changed over the years, papers published 10 or 20 years ago employ one notation, and more recent ones use another. This is frustrating for a layman like myself. However, to simplify things see the following equivalence table:
in italics the dbRBC or old name and in brackets, the (alt.) name (Source).
- ABO*O.01.01 = O01 (O1)
- ABO*O.01.02 = O02 (O1v)
- ABO* O.01.75 = O75 (O1v(G542A))
Below is the table, modified to show the new naming, it shows the mutations in the bases that distinguish the basal O(1) from A1, all are the same except for the deletion at 261. Then there are the mutations from O(1) to O(1v) where "v" stands for Variant. This table comes from Llop et al, 2008
I will be using the "old" notation from now on.
Almost all Amerindians belong to one of the three following haplotypes: O(1), O(1v) —O1 variant, and a specific mutation of O(1v) known as O(1v(G542A)), also written as O1v542.
O(1v), including its mutation G542A is the most common among Native Americans, accounting for 70-80% of all Amerindian O blood type individuals (Source) except for the peculiar Ecuadorian natives, the Cayapas, where O1 is more prevalent. They also have a rare mtDNA which may hint at a different source for their original, founding population.
Haplotype O(1) is the next most frequent, followed by some other very rare alleles as you can see in the following frequency table (1 = 100%), which includes some Amerindian groups.
The second table shwon above compares different Amerindian Populations with Asian ones. Translation of the Spanish words: Población is population, Aymara from Bolivia (Aymara b) and Chile (Aymara Chi), Japanese (Japoneses), Chinese (Chinos), and Koreans (Coreanos) and Mixed Latin-Native - Mestizo (Mest Chi).
The Mutation G542A
The G542A mutation is consistently found across Native American populations in frequencies of 12% to 45% of the total O blood group haplotypes.
It is rare in the rest of the world and found at very low frequencies (Estrada-Mena et al., 2010 and Llop et al, 2008 and Gueuning et al., 2023.)
Below is another table showing the mutations that distinguish O variants from the basal A1 blood group. The table is adapted from Yip, 2000. I added the O(1v) G542A line, and highlighted the two mutations, the deletion at position 261, and the marker for the 6542A mutation, where a guanine base flips to adenine G→A at position 542.
As you can see, O1 differs from the A1 allele by the deletion Δ261 (marked with "—) at position 261. The O1v allele has another 9 substitutions besides de Δ261. O1v is found in Eurasians and Table 1 from Villanea et al., 2021 show it at a 28% frequency.
Rare outside of America
Estrada-Mena, 2010 indicates that the O1v(G542A) haplotype has been found in some non-American populations: 6 Danes, 2 Kuwaitis, and 1 person from England. (He cites Olsson et al. 1998; Yip 2000; and Yip et al. 2006).
The study conducted in Zurich, Switzerland identified Gueuning et al., 2023 two, 2, (out of 154 samples) carrying the O1v(G542A) mutation (in the modern nomenclature it is called ABO∗O.01.75, and described the mutation as the regular O1v (modern name ABO*O.01.02) as follows: "ABO∗O.01.02 background with additional c.542G>A." The ABO∗O.01.02 is what they call O1(v), present in 31 out of 154 samples, or 22.65% frequency.
According to Estrada-Mena, this extreme rarity in the Old World of the O(1v(G542A)) or ABO* O.01.75 allele could be due to three options:(1) Gene flow of Native Americans or Latinamericans into the Old World, after European discovery. (2) Recurring mutation, it turns up by chance, infrequently, the G542A mutation crops up in different populations, but this is unlikely, as it needs to appear within the O(1v) haplotype to cause this mutation which is almost exclusively Amerindian. (3) It is an ancestral polymorphism, that appeared before human beings, in one of our ancestors. But since it is found at such a low frequency, and absent in all other Eurasian and African populations, this is also unlikely. Option (1) seems the most logical. I agree.
The Society of Blood Transfusion (ISBT) nomenclature shows that ABO*O.06 (former O53 or O6) and ABO*O.01.11 (former O11) both carry the c.542G>A mutation, but neither are within O1v, and they also carry other mutations.
Why is it so prevalent in America?
While the North American native people have a high prevalence of O, the Aleut, Eskimo/Inuit and Athabaskans of northern North America have high frequencies of A and B haplogroups. This difference is due to a later migration into America and a different genetic origin compared to the first wave to reach America ~20 kya (orthodox dating).
The high frequency of blood type O among current Native Americans could be due to a founder effect (it was already present at a very high rate among the people who first reached America), genetic drift (random mutations led other blood groups to fade away and vanish while O increased its frequency), natural selection due to the new environment and pathogens found in America that favored O over the other blood groups. The diseases introduced by Europeans, like smallpox could have also played a role in enhancing O group among native populations.
A study by Halverson and Bolnick, 2008 analyzed the blood types of ancient remains from eastern North America and found that "The precontact ABO frequencies were not significantly different from those observed in extant Native Americans from the same region, but they did differ significantly from the ABO frequencies in extant Siberian populations... [these] findings are most consistent with the hypothesis of a founder effect during the initial settlement of this continent.
The similar frequencies in ancient and contemporary populations seems to show that selection hasn't played a role. The difference with Siberians, on the other hand is interesting, orthodox intrepretation is of course, the founder effect, but it could also imply that current Siberians are not related to the people that were the source of the migrants that moved into America./p>
Interestingly, one of the co-authors of the Neanderthal, Denisovan, and human O-type blood paper that is the subject of this post, Fernando Villanea, in his 2010 thesis, Evolution of the ABO Blood Group Locus in Pre-Columbian Native Americans, evaluated and analyzed "four hypotheses... to explain the low overall Native American diversity and the high within Native American variance of ABO allele frequencies: a) Founder’s effect in the original population, b) Post-European contact reduction of diversity, c) PostEuropean contact natural selection and d) High population structuring following subsequent migrations into the continent."
In his conclusions, Villanea did not attribute the blood-type distribution among Amerindians to an initial founder effect, but to one due to the dispersal of small populations across the vast continent. He found that the "results are consistent with ABO diversity being lost as a result of isolation as populations migrated deeper into the continent and not as a result of a bottleneck in the original founder population." (Hypothesis d). Small groups or bands of natives moving across an empty territory, and expanding quickly in them plus genetic drift associated with these small populations shaped the blood groups of the New World.
Regarding smallpox, it is true that ABO bood types play a role in immunity, and it is possible that the O allele provided better fitness against that disease, but Villanea also found (like Halverson and Bolnick, 2008) that ancient pre-contact blood types that he sampled, this time, from West Coast remains from the U.S. show a similar diversity to the current one.
These comments about the relevance of O-type blood among Native Americans are significant, yet the paper by Villanea FA, Huerta-Sanchez E, Fox K., (2021) did not include any Amerindian or American data!! Below I will discuss this paper.
Neanderthal and Denisovan Blood Group O in modern humans
Going back to the paper by Villanea, Huerta-Sanchez, and Fox, 2021, which is the subject of today's post, the relevant part is that they detected the allele that is responsible for O blood group in two Siberian Neanderthals, the Altai Cave specimen, and the Chagyrskaya person, as well as in Denisovans.
These two Neanderthals lived in different periods, with the Altai Cave person living around 120,000 years ago (120 kya) and the other individual from nearby Chagyrskaya, living 80 kya.
We all carry two copies of an allele for blood group, one inherited from our mothers, the other from our fathers. The Altai Cave Neanderthal carried two O2 variants (he was homozygous, as both were identical, meaning they were both Os), while the Chagyrskaya Neanderthal had one O2 and one O1 (also homozygous). They also analyzed the genes of the Croatian Vindija Neanderthal from 50 kya, who carried one O1 allele and a rare variang called-cis-AB that does not code for O blood group (he was heterozygous); the Chagyrskaya Neanderthal individual was homozygous carying O1 and O2 alleles. Finally they analyzed a Denisovan from ~50 kya and found that it carried two different O1 alleles which have not yet been found in humans.
The authors note that "perhaps more surprisingly, the O2 allele variant found in Siberian Neanderthals can be found at low frequencies in modern Europeans and Southeast Asians, and the O1 allele variant found in Siberian and European Neanderthal is also found at very low frequency in modern East Asians." They then suggest that both alleles O1 and O2 were passed on to humans due to admixing with Neanderthals and that they have been positively selected for as they are more divergent than other chuks of Neanderthal genome in humans, meaning that they are still with us because they serve an adaptive purpose.
The image below summarizes Villanea et al.'s paper's findings:
Below is Table 1 from this paper, captioned "Genotypes for Eight ABO Variants and their Frequencies in the 1,000 Genomes Panel, Including Chromosome 9 Position in the Human Genome (Hg19, build37), as well as Genotypes for Each Chromosome in the Four Archaic Individuals."
The Nomenclature Changes!!
As you can see in the table above, Villanea et al. don't use the terminology "exon 7, 542 G→A", instead they employ numbers (top row) like 136131056 but this terminology which replaced the exon and SNP one, has also been superseded! The NIH gene viewer, for the ABO locus now uses a different number system (see here) so the numbers that ranged from 136125788 to 13615061709 for the ABO gene now are 133250401 to 133275201. See the gene online here.
The Amerindians Were Ignored in This Study
Surprisingly, this paper only included data from Old World modern humans, the authors discarded American populations, which is surprising, considering that Native Americans have the highest frequencies of O blood group in the world, almost 99% of Amerindians in South America carry O blood group. Why ignore it?
Villanea with Bolnick and others, 2013 studied the uniqueness of Amerindians finding that "Our results support a Beringian origin of O1v(G542A) , which is distributed today among all Native American groups that have been genotyped in appreciable numbers at this locus." So the preceding, basal root for this 542 mutation, the O1v had to be either Beringian or Siberian. Not an introgression from Europe after 1492. But see what the current paper says:
The authors justify this exclusion of Amerindians arguing that the O group samples that they found were of European origin (even among the African Americans they sampled). I quote them, and have introduced my comments in bold font:
"Identifying the origin of archaic introgression in admixed populations is complex. For example, American populations can trace portions of their genomes to European and African ancestry, as a consequence of European colonization of Native Americans, and the African slave trade. Because of these historical events, archaic introgression present in modern American individuals can be inherited from any of these sources in brackets.
[But what about isolated and almost pure natives like those of the Amazon? Why not include ancient samples of pre-contact people, like the ones used by Villanea, co-author of this paper, in his thesis?]
For the American populations in the 1,000 Genomes Project panel, including two African populations sampled in the Southwest United States and Barbados, we used ancestry calls from Martin et al. (2017) to distinguish if purported archaic ABO alleles sit in European-ancestry genome tracks. This allowed us to properly track archaic introgression to Europe, rather than being retained ancestrally in African populations, or introgression with the Asian ancestral populations from which Native Americans descend. For our genetic distance results, we thus excluded six populations sampled in America (MXL, PUR, PEL, CLM), including two African populations (ASW, ACB), after determining that all introgresssed ABO haplotypes are exclusively located in European ancestry tracks, thus providing a confusing look at archaic introgression."
[Why use Latinos? They are admixed by definition, MXL (Mexicans from Loa Angeles), PUR (Puerto Ricans), PEL (Peruvians from Lima), and CLM (Colombians from Medellín) all carry European and also African genes from the past 500 years! "Pure" contemporary and Amerindian samples are available, and would have come in handy to expand the scope of this study.]
Furthermore, study after study has shown that Amerindians and Latinamericans carry particular variants of blood group O, that are prevalent in America, how can these belong to "European ancestry tracks"? It seems odd to me.
Diego and Neanderthals
Estrada-Mena, 2010 suggests that the uniquely Amerindian mutation O1v(G542A) emerged in Beringia during the "standstill" there as a founder effect. This would explain its absence in Siberia and its exclusive presence in the Americas. A sound explanation.
But I am not bound by orthodoxy so, as I posted back in 2014 "the ample distribution of O1v(G542A) could also be explained by admixture with a Neanderthal population carrying the allele, which arose promoted by the selective advantage of conferring resistance against the infectious diseases of the New World. On the other hand, the alleged "cradle" of Amerindians, Eastern Asia, has the highest global frequency for type B blood allele. America... has the lowest, mostly among in Western Alaska due to recent East Asian admixture" Yes, we could also attribute the lack of B blood group to a founder effect or genetic drift, but there is additional proof.
There is a very rare blood group, called Diego (see my 2015 post about it), found in America and Central Asia. It is an Amerindian trait and dates back to the peopling of the New World. According to a paper published last year in Nature by Stéphane Mazières, Silvana Condemi, Wassim El Nemer & Jacques Chiaroni, 2025, it is ancient, the AR33K woman, from Amur in China, dated to 33 kya, carried the Diego D1*01 allele, however, her blood group was B. The paper states: "... before 30ky the Diego blood group might have already taken shape, before settling the Americas and comigrating with the nomads of the Mongolian steppes. Unfortunately, we could not corroborate any other DI*01 in Siberian individuals dating from the initial colonization of the Americas. " The next earliest ancient individual comes from a later population from Yana River, NEO239, Who lived 7,500 years ago. The paper also reports that "two Band3 Memphis alleles (Diego blood group system) have been observed in ... Denisova11 and Yana1." Denisova 11 or "Denny" is 90 ky old, and a hybrid of Denisovan and Neanderthal, while Yana1 id 30 ky old. The authors add that the D1*01 "may have appeared in South Siberia before migrating to the Americas and Central Asia."
O Blood Group and Amerindians
There is a clear link between Neanderthals, Denisovans, and blood group O. Apart from those mentioned futher up, the O2 type which is currently rare (1.1% frequency) was reported ( Mazières et al., 2025) in other ancient human beings from Russia 34 ky old, Sunghir2 and Sunghir3, as well as two Neandertals, Chagyrskaya D (50-60 kya) and GoyetQ56-1 (43 kya), so it was probably more prevalent in the distant past. Then there is Denisova 3 ("Pinky") from 52-72 kya, who carries the O blood group allele (Condemi et al., 2021).
Kitano et al., 2012 speculated that "Homo erectus had only B and O alleles until its later stage... [and] that the frequency of O allele was quite high among Neandertals."
So, blood group B, virtually absent among the first wave of people to reach America is also ancient, found in the heterozygous Neanderthal from Vindija. and in the AR33K woman. It was also present among ancients, but at lower frequencies.
There is an association between O blood group, Diego group, and Neanderthals, Denisovans, and modern Amerindians. The admixing could have taken place in Asia, with a swift migration, as mentioned in my previous three part posts on MUC 19 from South Siberia, into America. It could also have ocurred with Neanderthals in America.
Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2026 by Austin Whittall ©
