Total Pageviews

Thursday, June 21, 2012

Lactose Tolerance

The origin of domestic livestock (Goat, sheep, pig, cattle) started in the Middle East at least 10,000 years ago and spread from there all over the world. In the scientific publication “Domestication and early agriculture in the Mediterranean Basin: Origins, diffusion, and impact,” by Zeder et al., 2008, a comprehensive map is shown highlighting the historical importance of this region for mankind and its current diet.



The change of the diet from hunter/gatherers to farmers impacted not only the genetics of crop and livestock but also the genetics of humans. The most known and best described change is the ability of digesting and thus tolerating of milk as a food product, not only as infant but also as adult. Lactose, the sugar of milk, is the critical ingredient in milk that defines if someone can digest and thus tolerate milk or not. The responsible enzyme for digestion of lactose is called lactase, an enzyme that is highly expressed at early age in the small intestine and down-regulated after weaning, that's why adult hunter/gatherers are intolerant towards milk (lactose intolerant), while farmer populations can tolerate milk (lactose tolerant).

Dairy products (milk, yogurt, cheese) play an important role in the diet of the Iranian and Kurdish people, probably for thousands of years. It should be noted that Kurdish dairy products mostly rely on sheep and goat milk, while in Arabia camels were the main source of milk. In Europe bovine milk is dominating. Due to the size of camels and cattle it is very likely that Bedouins and Europeans consumed more dairy products than e.g. Iranians.

However, for quiet a while I am surprised to see how many Kurds are genetically lactose intolerant but physically lactose tolerant, even at high age. The best known SNP that is responsible for lactose tolerance is rs4988235.
1. People with "CC" at rs4988235 (also called C/T-13910) are genetically lactose intolerant; at 23andme it is "GG" (because MCM6 gene is in minus orientation but 23andme does not address it).
2. People with "CT" or "TT" at rs4988235 (also called C/T-13910) are genetically lactose tolerant; at 23andme it is "AG" or "AA" (because MCM6 gene is in minus orientation but 23andme does not address it)


Note: The actual gene for the enzyme lactase is called LCT and it is close to the gene MCM6. The MCM6 gene region seems to be important for the gene regulation of the LCT gene.

Let's take a look at the worldwide frequencies of rs4988235 (also called C/T-13910) in the Alfred database. The yellow bar shows the frequency of people that are genetically lactose intolerant (based on this one SNP), the green bar shows the frequency of people that are lactose tolerant:










Only 2-6% of Arabs, Bedouins, Druze, and Palestinians are genetically lactose tolerant.
Unfortunately, Northern Middle Eastern populations are not included in the ALFRED database but just looking at available data frequency of several Middle Eastern populations for this SNP rs4988235 it becomes obvious that rs4988235 cannot be only SNP responsible for lactose tolerance, there must be other independent mutations in the Middle Eastern genepool resulting in lactose tolerance.

Yuval Itan presented a nice map in his dissertation regarding the worldwide frequency of  C-13910:


So, where can we find more data about other SNPs that are related to lactose tolerance in the Middle East?

Enattah et al., 2008, showed that other Middle Eastern populations are lactose tolerant without having the mutation at SNP rs4988235 (also called C/T-13910). Instead they found mutations very close to rs4988235 (all in intron13) and one on exon17 of the MCM6 gene. The latter one on exon17 (T/C-3712) seems to be Arab specific.





The allele frequency of these SNPs in the Middle East (from Enattah et al., 2008):

Based on these data it seems that the major SNPs for Saudi-Arabian lactose tolerance are found:
1. It is T/G-13915 + T/C-3712, both SNPs always occur in pairs, which shows the strong positive selection for lactose tolerance and the common origin of both SNPs.
2. Even though 23andme is just determining T/G-13915 but not T/C-3712, it's okay because the two SNPs always come as pairs.

So, if you are an Arab (especially Saudi-Arabian) and lactose tolerant you most likely have at least one C allele at T/C-3712 and one G at T/G-13915.
The latter one is covered by 23andme, look up rs41380347 (Note: Again, because MCM6 gene is in the minus orientation and 23andme does not address it, all nucleotides are switched in the following manner: "G=>C", "C=>G", "T=>A" and "A=>T"):
rs41380347: If you have "AA" then you don't have the "Arabian lactose tolerance".
rs41380347: If you have "AC" or even "CC", then you do have the "Arabian lactose tolerance".

Finally, let's take a look at the worldwide frequencies of rs41380347 (also called T/G-13915) in the Alfred database. The yellow bar shows the frequency of people that are genetically lactose intolerant (based on this "Arabian" SNP), the green bar shows the frequency of people that are lactose tolerant: 






Yuval Itan presented a nice map in his dissertation regarding the worldwide frequency of T/G-13915 and two more African SNPs known to cause lactose persistence:





Here is another map summarizing all known lactose persistence SNPs and their worldwide frequencies:






Contrary, here is a map showing lactose persistence phenotype frequencies:






Summary:
Not all lactose tolerance SNPs are known, several SNPs resulting in lactose tolerance still need to be discovered. 


One way to discover the relevant regions in the human genome for such SNPs in the Middle Eastern genepool is by family studies:
If some of your family members are lactose tolerant while other family members are lactose intolerant and if you have enough 23andme results of both groups within your family, you might be able to pinpoint the genomic region by exclusion of not relevant genomic regions.

4 comments:

  1. Palisto, in case you don't know, there is a new paper on early North African dairying:

    http://www.nature.com/nature/journal/v486/n7403/full/nature11186.html

    And here is a summary of its findings:

    http://www.nature.com/news/pottery-shards-put-a-date-on-africa-s-dairying-1.10863

    ReplyDelete
  2. Thanks, Onur. I just started to take a look at it. The most important sentences from the paper are:

    "Direct evidence for the practice of dairying, beginning in the seventh millennium BC in northwestern Anatolia, appearing in the sixth millennium BC in eastern Europe and reaching Britain in the fourth millennium BC, has been established through the compound- specific stable carbon isotope analysis of animal fat residues preserved in archaeological pottery."
    and
    "Here we report the first unequivocal chemical evidence, based on the d13C and D13C values of the major alkanoic acids of milk fat, for the adoption of dairying practices by prehistoric Saharan African people in the fifth millennium BC."

    Dairy products were consumed 7000 years ago in Northern Africa, most likely from cattle. My guess is that these first North African Pastorals had the Y-haplogroup E1b1b1b1 (E-M81).

    ReplyDelete
    Replies
    1. Hey, are you sure you have got the 23andMe data right? It says they do the test for rs4988235 (https://www.23andme.com/health/Lactose-Intolerance/), which is C/T -13910(http://snpedia.com/index.php/Rs4988235). You have stated that "Even though 23andme is just determining T/G-13915 but not T/C-3712, ...".

      Delete
    2. 23andme is testing C/T -13910 alias rs4988235; 23andme is not testing T/C-3712.

      Delete