Why Are Some Types of People More At Risk of Schizophrenia Than Others ?

Posted on December 15, 2013 | 1 comment

Schizophrenia is defined by the World Health Organisation as:

“a severe mental disorder, characterized by profound disruptions in thinking, affecting language, perception, and the sense of self. It often includes psychotic experiences, such as hearing voices or delusions. It can impair functioning through the loss of an acquired capability to earn a livelihood, or the disruption of studies.”

Schizophrenia is typically diagnosed between the ages of 15 and 35.  It is one of the most common serious mental health conditions. There is now strong evidence that males have an increased risk of developing schizophrenia.  Heritability of schizophrenia is estimated at 83% according to a population-based twin study.  

The incidence of schizophrenia in Europe can range from 7 per 100,000 [Aarhus, Denmark], up to 14 per 100,000 [Nottingham, UK]. [World Health Organisation 10 country study, Jablensky et al 1992. See here].

Although it is treatable, schizophrenia imposes a huge cost on society.  It is estimated that in the US, schizophrenia consumes a total of $63 billion a year for direct treatment, family and societal costs.  According to National Institutes of Mental Health, almost one third of that figure is for direct treatment and the remainder comprises other factors such as lost time from work for patients and caregivers, social services and criminal justice resources.  Schizophrenia accounts for a quarter of all mental health costs.  As most schizophrenia patients are never able to work, they also must be supported long term by public assistance and Medicaid.  In the UK, 80 million working days are lost each year and according to a report by the London School of Economics and Political Science, schizophrenia costs UK society around £11.8 billion [$18.8 billion] a year [http://www.lse.ac.uk/LSEHealthAndSocialCare/pdf/LSE-economic-report-FINAL-12-Nov.pdf].  In addition, people with schizophrenia have around a 50 times greater risk of committing suicide than the general population.

So what actually causes schizophrenia, and who most is at risk?  The exact cause remains unknown, however many experts support what is known as the stress vulnerability model theory of schizophrenia.  This theory states that everyone has a particular vulnerability to schizophrenia which is determined by a combination of biological, psychological and environmental factors.  In this model, a traumatic or stressful event can trigger schizophrenia, with relatively minor events triggering it in people with higher vulnerability, and more significant events triggering it in people with lower vulnerability.

Several risk factors and/or causes are believed to responsible for risk of susceptibility to schizophrenia. [Note that we can’t always be sure if the association is correlational or causational].   

These could include: genetic heritability; infections such as toxoplasmosis and herpes; sensitivity to the neurotransmitter dopamine; mutations in the gene for serotonin receptor 2a;  parasitic infection interfering with the development of the foetal brain during maternal pregnancy; gut inflammation or dysbiosis [which has numerous causes]; being born in winter; reactions to gluten and casein (food proteins found in wheat and dairy); hormonal changes (eg during puberty); pregnancy or birth complications; children born of consanguineous marriages; climate; latitude; Vitamin D status; fish consumption; advanced paternal age (and/or possibly maternal age); regular use of narcotics, eg cannabis, and especially strong herbal cannabis [‘skunk’], particularly at a younger age; being raised in an urban environment; being an immigrant; having experienced traumatic or stressful life events; being of African/Caribbean [or African-American] heritage; and being male.

Regarding genetics, in terms of the relationship between the Major Histocompatibility Complex (MHC) and schizophrenia, Human Leukocyte Antigen (HLA) loci were among the earliest to be studied.  Initial results had proved promising.  However, a meta-analysis a few years later with larger volumes of data proved to be somewhat less promising.  More studies were carried out since then however.  Recently the subject has been looked at once more, with a number of genome wide association studies (GWAS) and a combined analysis by the Psychiatric Genomics Consortium which demonstrated a link between schizophrenia and single-nucleotide polymorphisms (SNPs) in the MHC region.

Previously HLA alleles have been associated with various immunological disorders.  However there has been no conclusive proof that schizophrenia had an auto-immune related cause.  The only non-immune related disorder with a clear HLA association was hemochromatosis.  Also, a problem with HLA is that associations with certain conditions are hard to pin down as alleles can be tightly linked with each other in blocks as part of haplotypes (‘linkage disequilibrium’ or LD) and don’t recombine quickly.  So that an association between an immune disorder and HLA can be easily identified, but pinpointing the actual causal variants is not quite so easy. Also, the situation is compounded by other issues, including the fact that humans tend to mate with MHC-dissimilar partners which leads to greater HLA heterogeneity in their offspring.  Also, due to the MHC’s role in protecting against pathogens the region is under intense selective pressure which increases diversity [the HLA loci being the most polymorphic].  And finally, in many western countries with ethnically diverse populations and many highly admixed individuals, false-positive disease associations may also arise, so it is important to carefully match case and control subjects ethnically.

Despite these issues, however, it now definitely looks as though HLA’s association with schizophrenia is genuine.  As the numbers of studies carried out and sample sizes have greatly increased, results have increasingly been replicated and p-values [for explanation of p-values, see here] have dramatically reduced. Moreover, research increasingly suggests that the MHC region plays a significant role in neuronal function, brain development (prenatal and postnatal), plasticity, etc.

So which HLA alleles have been found to confer susceptibility to or protection from schizophrenia?  And what are the frequencies of these alleles in worldwide populations?

The Irish Schizophrenia Genomics Consortium and the Wellcome Trust Case Control Consortium 2 genome-wide association study (GWAS) [see here] identified HLA Class I allele C*01:02 as conferring susceptibility to schizophrenia:-

File:Cw-0102.gif

1  C*01:02  China Southwest Dai  0.2940  
 2  C*01:02  China Guizhou Province Shui  0.2840  
 3  C*01:02  Australia Yuendumu Aborigine  0.2470  
 4  C*01:02  China Yunnan Province Han  0.2180  
 5  C*01:02  Taiwan pop 2  0.2150  
 6  C*01:02  Taiwan Han Chinese  0.2120  
 7  C*01:02  China Guizhou Province Bouyei  0.2100  
 8  C*01:02  USA Hawaii Okinawa  0.2050  
 9  C*01:02  China South Han pop 2  0.2017  
 10  C*01:02  Taiwan Minnan pop 1  0.2010  

So we can see from this study that peak world frequencies for this allele are to be found in certain regions of China, plus Taiwanese aboriginals and Australian aboriginals.

But what about alleles which have been found to confer protection from schizophrenia?  The same study replicated a previous study, and also found that HLA Class I alleles B*08:01 and DRB1*03:01  were both protective:-

File:B-0801.gif

1  B*08:01  Ireland South  0.1820  
 2  B*08:01  Australia New South Wales Caucasian  0.1730  
 3  B*08:01  Ireland Northern  0.1620  
 4  B*08:01  England North West  0.1540  
 5  B*08:01  Germany DKMS – United Kingdom minority  0.1342  
 6  B*08:01  Germany DKMS – Netherlands minority  0.1125  
 7  B*08:01  USA Caucasian Bethesda  0.1120  
 8  B*08:01  Oman  0.1100  
 9  B*08:01  Austria  0.1100  
 10  B*08:01  USA Caucasian pop 2  0.1090  

DRB1-0301.gif

1  DRB1*03:01  Italy Sardinia pop2  0.5570  
 2  DRB1*03:01  United Arab Emirates pop 2  0.1810  
 3  DRB1*03:01  Ireland South  0.1700  
 4  DRB1*03:01  India Northeast Shia  0.1530  
 5  DRB1*03:01  France Rennes  0.1520  
 6  DRB1*03:01  India Andhra Pradesh Golla  0.1510  
 7  DRB1*03:01  Tunisia pop 3  0.1510  
 8  DRB1*03:01  France Rennes pop 3  0.1500  
 9  DRB1*03:01  Tunisia pop 2  0.1500  
 10  DRB1*03:01  Norway  0.1400  

You will recall from earlier where I mentioned the results of the World Health Organisation ten country study, where the incidence of schizophrenia in European cities ranged from a low of 7 per 100,000 in Aarhus, Denmark, up to a high of 14 per 100,000 in Nottingham, England.

Now both of these cities are broadly similar in size, population, and wealth.  Aarhus is at a more northerly latitude than Nottingham.  The population of Aarhus is more than 85% of white Danish ethnic background, while the population of Nottingham is approximately 65% of white British ethnic background.  Numerous studies have indicated that British and Danish people are genetically very close to one another.  So unless there is a major lifestyle difference between the populations of the two cities which is somehow causing the incidence of schizophrenia in Nottingham to be double that of Aarhus, clearly there is another factor at play.  In addition, it is probably unlikely that British health professionals would be far more zealous in terms of diagnosing schizophrenia than their Danish counterparts are.

Could there be a significant difference in rates of cannabis use? According to the United Nations Office on Drugs and Crime (UNODC) World Drug Report 2011, the annual prevalence of cannabis use as a percentage of the population aged 15-64 was 5.5% in Denmark (2008); and 6.6% in England and Wales (2010).  I haven’t looked at historical trends or regional trends in either country, but this is probably a reasonable proxy for estimating cannabis use rates in Nottingham and Aarhus.  If cannabis use was the primary or major cause of schizophrenia, we would likely see far higher rates of cannabis use in England and Wales in order to account for such a disparity in schizophrenia rates between the two cities.

In any case, it is not at all clear that cannabis use really is a cause anyway.  The first study on schizophrenia and cannabis took place in Jamaica in the 1960s.  Doctors had observed that Rastafarians who regularly smoked cannabis were at greater risk of psychosis.  Further studies in the UK and Sweden in the 1980s showed that schizophrenia patients were several times more likely to be cannabis smokers.  However these studies merely demonstrated correlation rather than causation.  In fact some schizophrenia patients reported that smoking cannabis made them feel better, so it was theorised that people who are more predisposed to schizophrenia may choose to “self-medicate” with cannabis.  Some more recent studies, however, suggest that cannabis use may be a partial cause of schizophrenia, when used by teenagers or young adults whose brains are still developing.  Nevertheless further research is required.

Alcohol use is unlikely to be a major factor.  According to the World Health Organisation, total pure alcohol consumption among adults aged 15+ in litres per capita per year is the same figure for both UK and Denmark [13.37 litres]. [See here].    

One major difference between the two cities that we haven’t looked at yet is the ethnic composition.  

According to Wikipedia, in 2008 12% of Aarhus’ population were immigrants.  If we assume that, including those born in Denmark to non-Danish parents, the total population of non-Danish origin was 15%, that means Aarhus’ population is 85% of White Danish origin.  According to Wikipedia, the largest immigrant groups (2008 data) are Palestinians (1.4%),Turks (1.4%), Somalis (1.2%), Iraqis (1.0%), Vietnamese (0.8%), and Iranians (0.7%).

We have recent ethnic data available for Nottingham, UK, following the publication of the 2011 Census by the Office for National Statistics (ONS) – [see here].

The ethnicities of Nottingham’s minority populations are as follows: South Asian 9.0%; Black 7.3%; White Other [mainly East European, especially Polish] 6.1%; Black/White Mixed 4.7%; Asian Other 2.1%; Chinese 2.0%; Other Ethnic Group 1.5%; Asian/White Mixed 1.1%; Other Mixed 0.9%.

So from comparing this data we can conclude that the current population of Nottingham in the UK is far more ethnically diverse than that of Arhus in Denmark.

But how different was the ethnic profile of the two cities back in 1992 when the WHO published their 10 country study?

In the case of Nottingham we can refer back to extracts from the 1991 census.

In 1991, Nottingham’s population was lower, being 235000 [as compared with 289301 in 2011].

Nottingham’s population at that time was considerably less diverse than today.  Although there was a small Irish community, in 1991 it was still many years before large-scale migration of East Europeans to the UK following the expansion of the European Union in 2004.  So the majority of white people in Nottingham were White British.  Also the UK government had yet to begin its programme of dispersing refugees and asylum seekers out to provincial cities (the majority were settled in London).  It was also many years before the expansion of numbers of international students enrolled in British universities, and the National Health Service (NHS) had yet to begin a new phase of large-scale recruitment of doctors and nurses abroad.  

Therefore most ethnic minorities residing in Nottingham in 1991 were first and second generation migrants of South Asian origin (approximately 4.4%) and African-Caribbean origin (approximately 3.2%), derived from migration in the 1950s and ’60s.

I haven’t been able to locate data on Aarhus’ population in the early 1990s, but it is probably fair to assume that the proportion of immigrants was likely considerably lower in 1991/2 than it was in 2008.

It seems fair to conclude that the fact that the rate of schizophrenia diagnosed in Nottingham in the UK was double the figure for Aarhus in Denmark [according to the World Health Organisation’s 1992 10 country study] is likely due to Nottingham’s very different ethnic composition compared with Aarhus.

In their paper “Migration and schizophrenia: an examination of five hypotheses” [Social Psychiatry 1987, Volume 22, Issue 4] Raymond Cochrane and Sukhwant Singh Bal looked at rates of admission for schizophrenia for the native born and the four largest foreign born populations living in England in 1981 (Irish, Indian, Pakistani, and Caribbean born).  These groups were also the main ones represented in Nottingham in the 1991 census.

Interestingly, schizophrenia rates for the Irish living in England were comparable with Irish people living in Ireland, but this figure was still actually higher than that of the English themselves.  The Irish do not greatly differ from the English genetically, so perhaps other factors may account for the difference [recent history of consanguineous marriages within Irish families? poorer nutrition?] Despite popular stereotypes, high alcohol consumption is unlikely to be the issue as consumption is only marginally higher among the Irish as compared with the British, according to World Health Organisation data.

Very high rates of schizophrenia were found among both Caribbean-born men and women; high rates were found among Indian-born men and women; and high rates among Pakistani-born men only [this is believed to be due to cultural reasons, with mentally ill Pakistani women opting out of the formal mental health system &/or returning to their country of origin].

My conclusion is that economic migrants to cities in western countries in northern latitudes who originate from non-western, tropical countries are genetically predisposed to a risk of schizophrenia, but also placed at greater risk of certain trigger factors for schizophrenia risk in terms of their new lifestyle.  

In the case of South Asian and African/Caribbean migrants to the UK, it may be the case that there could be several such trigger factors. These could include, for example, increased psychological stress [for many different reasons]; lack of exposure to bright sunlight/reduced vitamin D levels; poorer nutrition; reduced immune system function brought on by several of the above factors plus colder air temperature in UK resulting in parasitic infection [impacting either the adult migrant themselves or their unborn foetus], etc.

From the point of view of genetics, in particular looking at the Major Histocompatibility Complex (MHC) again, what can be determined from the available HLA data with regard to South Asian and African/Caribbean populations and schizophrenia risk, relative to British?

I decided to use HLA data from the Allele Net Frequency Database (ANFD) to find out [see here].

I decided to look at the following populations: Representing ‘White British’, a sample of Brits [n=1043] living in Germany; representing ‘South Asian’, two populations living in Delhi, India [n=112 & n=90, respectively]; representing ‘Black’ [including West African and West African-derived populations] are several groups including African Americans, African Brazilians, African Caribbeans, various West African groups, and Congolese  This data ought to be representative of the main three ethnic groups residing in Nottingham at the time of the World Health Organisation’s ten country study on schizophrenia in the early 1990s.

I will look at frequencies of the three main HLA alleles which I previously mentioned in regards to schizophrenia: C*01:02 [confers susceptibility]; B*08:01 [protective]; and DRB1*03:01 [protective].

C*01:02 [Confers susceptibility to schizophrenia]:

 C*01:02  Germany DKMS – United Kingdom minority  0.0319    
 C*01:02  India Delhi pop 2  0.0310  
 C*01:02  Senegal Niokholo Mandenka  0.0310  
 C*01:02  USA African American Bethesda  0.0160  
 C*01:02  USA African American pop 4  0.0085  
 C*01:02  Mali Bandiagara  0.0080  

 

 

B*08:01 [Confers protection from schizophrenia]:

 B*08:01  Germany DKMS – United Kingdom minority  0.1342  
 B*08:01  India Delhi pop 2  0.0160  

 
 B*08  Senegal Dakar  0.1110  
 B*08:01  Guinea Bissau  0.0770  
 B*08  Brazil Parana Afro Brazilian  0.0650  
 B*08  USA African American Bethesda  0.0640  
   B*08:01  Cameroon Yaounde  0.0540  
 B*08:01  Senegal Niokholo Mandenka  0.0480  
 B*08:01  USA African American  0.0460  
 B*08  USA OPTN African American  0.0430  
 B*08:01  USA African American pop 3  0.0400  
 B*08:01  USA African American pop 4  0.0384  
 B*08:01  USA African American pop 8  0.0380  
 B*08  Martinique  0.0300  
 B*08  Burkina Faso Mossi  0.0190  
 B*08:01  Cameroon Beti  0.0110  
 B*08:01  Mali Bandiagara  0.0070  
 B*08  Trinidad African  0.0000  

 

DRB1*03:01 [Confers protection from schizophrenia]:

 DRB1*03:01  Germany DKMS – United Kingdom minority  0.1395  
 DRB1*03:01  India Delhi  0.1220  
 DRB1*03:01  USA Colorado Univ Cord Blood Bank African American  0.0780  
 DRB1*03:01  USA African American pop 4  0.0707  
 DRB1*03:01  USA African American Bethesda  12.6 0.0700  
 DRB1*03:01  Congo Kinshasa Bantu  0.0670  
 DRB1*03:01  USA African American pop 7  0.0670  
 DRB1*03:01  Martinique  0.0600  
 DRB1*03:01  Congo Kinshasa expatriates living in Belgium  0.0460  
 DRB1*03:01  Jamaica  0.0230  
 DRB1*03:01  Senegal Dakar  4.4 0.0220  

Discussion

In looking at this HLA data, one can observe that the White British and South Asian populations both have similar frequencies of HLA C*01:02 which confers susceptibility to schizophrenia, while in the absence of suitable data on Jamaicans one can guess using the data from other populations of similar origin that their frequency of the allele would be either equal to or lower than the White British and South Asian samples [definitely not likely to be much higher].

Regarding the B*08:01 allele which confers protection from schizophrenia, this allele is at a very high frequency in the White British sample, very low frequency in the South Asian sample, and found at verious frequencies in the West African-derived populations [but still usually far lower than the British].

Regarding our final allele, DRB*03:01, which is also protective from schizophrenia. This allele is often found within the same high frequency haplotype as B*08:01 in northern European-derived populations [A*01:01-B*08:01-C*07:01-DRB1*03:01-DQB1*02:01], so it is no surprise that it would also be found at high frequency in the British population.  This allele is also found at high frequency in our South Asian group [12.2%].  However, the frequency is much lower in general for the African-derived populations, ranging from as low as 2.0% to as high as 7.0%. We are fortunate for this allele to actually have data from Jamaicans, which falls at the low end of the scale (just 2.3%).

So to recap: it appears that the White British, South Asian, and Black [West African-derived] populations all possess similar, rather low frequencies of the schizophrenia susceptibility-conferring C*01:02 allele.

The protective allele B*08:01, found at very high frequencies in the British, is at a low frequency in the South Asian sample, and with the exception of certain outliers is found at much lower frequencies in Black [West African-derived] populations as compared with the British.

Finally, the protective allele DRB1*03:01 is again found at a very high frequency in the British sample, but is also found at a high frequency in the South Asian sample this time.  Whereas the Black [West African-derived] populations again have much lower frequencies, and in the case of the Jamaican sample in particular, the figure is particularly low.

Conclusion

We could benefit from many more populations being typed for HLA, and more studies or datasets being published, and in particular more studies into HLA disease associations.

However using the HLA data to compare three different ethnic groups [White British, South Asian, and Black] which resided in Nottingham at the time of the World Health Organisation schizophrenia study in 1992, it is possible to conclude that Nottingham’s schizophrenia rate was double that of Aarhus in Denmark not so much for environmental or lifestyle factors, but more likely due to its ethnic profile.

The British have similar levels of the allele that confers susceptibility to schizophrenia as the South Asian and Black groups do, but very high levels of the two alleles which confer protection from schizophrenia.

Whereas the South Asian sample only shows a high frequency of just one of the two protective alleles, the Black [West African-derived] groups generally have high frequencies of neither, and in a number of cases [eg Jamaicans and DRB1*03:01] actually have very low frequencies.

Clearly there are both environmental and genetic reasons behind the higher rates of schizophrenia among Black and South Asian migrants to the UK relative to the native British population.

As more studies are carried out in the coming years the picture will become much clearer.

To take a free, online schizophrenia test, click here.  

  

  

1 Comment

Posted in Uncategorized

Tagged ,

An Introduction to HLA (Human Leukocyte Antigens) – Re-Post! [First Posted On HBD Chick’s Blog As Guest Post 15/05/13]

Posted on October 29, 2013 | Leave a comment

*EDIT – PLEASE SEE:-   http://hbdchick.wordpress.com/category/guest-post/

[AND:  http://hbdchick.wordpress.com/]

I would like to give a shout out to HBD Chick, who I am very grateful to for giving me a chance to write a guest post on her excellent blog a few months ago.  The post was an explanation about what ‘Human Leukoctye Antigens’ (HLA) are, and how they relate to those of us with an interest in Human BioDiversity (HBD), Evolutionary Genetics, Molecular Anthropology, or whatever else floats your boat in those areas.  If you haven’t been there already, I very strongly urge you to visit HBD Chick’s blog: ( hbdchick.wordpress.com ). She is an hbd blogger with a difference.  She has an amazingly specialised knowledge of mating patterns and family types in different human populations around the world, and the surprising effects these patterns have had on societies historically, and right up to the present day socially, economically, and politically.  If you think that those type of things don’t matter then you are very wrong and you need to read her blog.

Anyway, for those of you who have just stumbled across my obscure blog, and who don’t have a clue what the hell ‘HLA’ is (or why they should even care), I implore you to take a couple of minutes of your time to read my post below, and allow me to give you an explanation:-

An Introduction to HLA (Human Leukocyte Antigens)

I would like to thank HBD Chick, who recently asked me if I wanted to write a guest post here about Human Leukocyte Antigens (HLA). I have been a reader of this blog for several months now, and I really enjoy it. This blog is among the most innovative of the HBD-related blogs out there and takes a completely different approach to the subject. So it is a pleasure to write a post here, as I don’t write a blog myself.

In the comments section on some posts here recently I have talked about ‘HLA haplotypes’. I have had an interest in HLA haplotypes for a few years, following discussions with a molecular biologist who writes (or rather wrote) in newsgroups including Usenet. But what are ‘HLA haplotypes’, and more importantly what is their usefulness or relevance to those of us with an interest in ‘HBD’?

‘HLA’ means Human Leukocyte Antigen. The HLA system is simply the name given to the human version of the Major Histocompatibility Complex (MHC). MHC genes are found in most vertebrates, and this group of genes can be found on chromosome 6.

HLA genes are important in immune function and disease defence. There are three classes: Class I (A,B,C); Class II (DP, DM, DOA, DOB, DQ, DR); and Class III. All play different roles.

In addition to conferring resistance or susceptibility to various diseases and conditions (eg autoimmune diseases, such as type I diabetes), they are also vitally important in organ and bone marrow transplants. For example, if a donor is not a close enough match in terms of HLA, an organ could be rejected. Because diversity of HLAs in human populations is a part of disease resistance, it is very uncommon to find two unrelated individuals with identical HLA molecules at all loci.

In addition, it has been claimed that HLA may play a role in human mate selection through people’s perception of the odor of other people.

HLA haplotypes are like strings of HLA genes by chromosome, with one being inherited maternally and the other paternally. A multigene haplotype is a set of inherited alleles covering several genes, or gene-alleles. Common multigene haplotypes are generally the result of identity by descent from a common ancestor. As distance from the ancestor increases, chromosomal recombination causes multigene haplotypes to fragment.

An example of an HLA haplotype is shown below:

A*01:01 ; C*07:01 ; B*08:01 ; DRB1*03:01 ; DQA1*05:01 ; DQB1*02:01
[Which by serotyping, is more simply: A1-Cw7-B8-DR3-DQ2].

This particular haplotype btw is found at high frequency among people of North-West European descent, including Irish, British, Dutch, Germans, Scandinavians, and of course a high percentage of Americans and Australians. It is associated with autoimmune disorders like coeliac disease, and allergic diseases like hayfever and asthma.

In addition to medical use, HLA haplotypes can also be used as a means of tracing migrations in the human population as they are like a fingerprint of an event that has occurred in evolution.

So alongside Y Chromosome DNA and Mitochondrial DNA (mtDNA) markers, HLA haplotypes are useful tools for molecular anthropologists in determining evolutionary links between ancient and modern human populations. They also help to determine closeness of relationships between or within populations, and commonality of geographical origin between groups. Therefore patterns of migration and settlement can be traced, giving insight into how contemporary populations have formed and progressed over time.

Because the HLA system is under selection, diversity in HLA should be looked at in conjunction with other markers like Y-DNA and mtDNA in building a bigger picture when tracing migrations of modern humans.

However, HLA has several advantages over Y-DNA and mtDNA. mtDNA is often biased towards the founders in a core population and can show a punctuated distribution, while Y-DNA is biased more towards relatively recent migrations, and with both of these markers the effects of genetic drift are more rapid. With HLA there is intense heterozygous selection which works to preserve diversity.

Also with HLA, one tends not to see a punctuated distribution of haplotypes across geographical regions between groups, but rather one sees a fluidity of migrations tracing back to their original source population. Examples of this occur when one looks at diversity of HLA in northern Africa, which appears to be the source for a number of haplotypes that migrated into Europe in the Holocene. This picture is less clear from looking at mtDNA or Y-DNA data only.

Finally, the other advantage of HLA is the sheer volume of data available. [See: www.allelefrequencies.net]. The numbers of people worldwide who have been typed for HLA are huge. There are nearly 22 million donors and blood cord units in the BMDW (Bone Marrow Donors Worldwide) database, encompassing many different ethnicities and nationalities with hundreds of thousands of HLA alleles and extended haplotypes identified. Y-DNA and mtDNA cannot offer this volume of data. Whether used for tracing ancient human migrations, or researching diseases in different populations, this data can be extremely helpful.

The association of HLA haplotypes in populations with certain diseases in Western countries is interesting to consider from an HBD perspective. For example, if resistance or susceptibility to obesity, heart disease, diabetes, breast cancer, etc. are all affected by HLA type, then there could be wildly varying outcomes between different population groups, in addition to the effects of environmental or lifestyle factors.

This in turn could have serious implications. For example, certain disease-related charities’ fund-raising campaigns are aimed at convincing the general public to donate cash on the basis that everyone is almost equally at risk from that particular disease, when different ethnic groups living in the same country with the same lifestyle may often vary in risk considerably from very high risk down to very low risk, with HLA type being an important determining factor. But the charity may not publicise this information in order not to jeopardise their fund-raising. This, I should stress, is speculation on my part however.

If want to find out more about Human Leukocyte Antigens, please refer to some of the sources listed below:

References
– The Allele Frequency Net Database (AFND). A database and online repository for immune gene frequencies in worldwide populations: [www.allelefrequencies.net]
– “Tracking Human Migrations by the Analysis of the Distribution of HLA Alleles, Lineages and Haplotypes in Closed and Open Populations.”[Fernandez Vina, M. et al, 2012].
– “HLA 1991: Proceedings of the Eleventh International Histocompatibility Workshop and Conference” (Volumes I & II) [Kimoshi Tsuji; Miki Aizawa; Takehiko Sasazuki] Oxford Science Publications, 1991.
– “HLA-Net: A European Network of the HLA Diversity for Histocompatibility, Clinical Transplantation, Epidemiological and Population Genetics.” [www.hla-net.eu].
– “AHPD: Analysis of HLA Population Data to Reconstruct the History of Modern Humans and Infer the Role of Natural Selection.” [geneva.unige.ch/ahpd].
– “Challenging Views on the Peopling of East Asia: the Story According to HLA Markers.” [Di D, Sanchez-Mazas A.]. Am J Phys Anthropol. 2011 May; 145(1):81-96.doi:10.1002/ajpa.21470.Epub2011Jan4. PMID:21484761 [Pubmed – indexed for MEDLINE].
– Wikipedia: “Human Leukocyte Antigen.”
– Wikipedia: “Major Histocompatibility Complex.”

Leave a comment

Posted in Uncategorized

Origins of Americans (Looking At HLA Haplotypes)

Posted on October 26, 2013 | 2 comments

Thought I would start my blog with a semi-regular look at the origins of various human populations, using genetic data from the population under investigation, and then comparing it with data from other human populations around the world to ascertain where they derive from [or which populations derive from them].

There are numerous genetic markers which are used for populations when investigating [prehistoric or historic] ancestry.  Some of these include Y-DNA  [Y Chromosome DNA], mtDNA [Mitochondrial DNA], atDNA [Autosomal DNA], and HLA [Human Leukocyte Antigens].  All of them can be useful, but also have certain limitations.  I choose to predominantly use HLA alleles and haplotypes as ancestry markers when researching populations.

I’m going to start with looking at the origins of Americans, using HLA haplotypes.  I have access to a very large database of HLA haplotypes sampled from many different populations around the world.

On http://www.allelefrequencies.net , below are the top five most frequent HLA haplotypes in USA Caucasians [N = 2,361,208]:

1.  A*01 – B*08 – DRB1*03:01

2. A*03 – B*07 – DRB1*15:01

3. A*02 – B*44 – DRB1*04:01

4. A*02 – B*07 – DRB1*15:01

5. A*29 – B*44 – DRB1*07:01

Now for each haplotype, I will look at the top five populations [besides USA Caucasians] with the highest frequencies of these haplotypes, in order to ascertain where they originated from.  In future posts I may look further back in time, by breaking each haplotype down into its component alleles in order to find out approximately where [and when] the haplotype may have formed [in prehistory].  I will also look at potential risk of susceptibility to medical conditions and diseases associated with certain HLA alleles.

A*01 – B*08 – DRB1*03:01

1. Ireland South [N=250] 11.50%

2. England North West [N=298] 9.50%

3. Ireland Northern [N=1000] 9.00%

4. Cornwall [N=101] 8.40%

5. Netherlands [N=1374] 6.70%

[Sources: http://www.allelefrequencies.net ; & “HLA 1991: Proceedings of the Eleventh International Histocompatibility Workshop and Conference” – Tsuji et al, Oxford Science Publications 1991.]

A*03 – B*07 – DRB1*15:01

1. Spain Pas Valley [N=88] 6.50%

2. Ireland Northern [N=1000] 4.90%

3. Russia Chuvash [N=82] 4.90%

4. Ireland South [N=250] 4.30%

5. England North West [N=298] 4.20%

[Sources: see above]

A*02 – B*44 – DRB1*04:01

1. Cornwall [N=101] 7.90%

2. Ireland Northern [N=1000] 4.10%

3. Ireland South [N=250] 4.00%

4. England North West [N=298] 3.90%

5. Generic British [N=1043] 3.19%

[Sources: see above]

A*02 – B*07 – DRB1*15:01

1. England North West [N=298] 3.80%

2. Sweden Southern Sami [N=130] 3.80%

3. Spain Pas Valley [N=88] 3.80%

4. Ireland South [N=250] 3.10%

5. Ireland Northern [N=1000] 3.00%

[Sources: see above]

A*29 – B*44 – DRB1*07:01

1. Spain Arratia Valley Basque [N=83] 5.30%

2. Spain Murcia [N=173] 5.10%

3. Tunisia [N=100] 4.00%

4. Spain Pas Valley [N=88] 2.70%

5. England North West [N=298] 2.20%

[Sources: see above]

Note: Only populations with sample sizes >50 were included.

Conclusions

As expected, populations in the British Isles feature very prominently as source populations for USA Caucasians.  Need to obtain better data for Scottish.  Note that A29-B44-DR7 is found at highest frequency in eastern Spain; second highest European frequency is in western British Isles and it is from there that the #5 haplotype in USA Caucasians most likely derives, rather than, for example, USA Hispanics identifying themselves as ‘Caucasian’ as opposed to Hispanic.

As mentioned, in future posts I will weigh up the evidence from the data and look at the likely original prehistoric sources for the above haplotypes within Europe itself.

I will also look at disease associations for some of the HLA alleles, and address the origins of Native Americans, USA Hispanics, and African Americans using HLA data.

Commenters are very much welcome here, but please keep it relevant and polite.

I appreciate this blog covers a niche topic, so I don’t anticipate a large volume of traffic, however anybody is welcome to read and comment here.

Next post due..when I get round to it. ;-)

 

2 Comments

Posted in Uncategorized