Among genetic traits that we all share, the most familiar single gene trait in humans is probably our ABO blood type. Other familiar shared heritable human traits, such eye color, are not controlled by a single gene. Many traits that are inherited as a single gene are associated with genetic diseases, such as cystic fibrosis and sickle cell anemia, so they are familiar because their effects have profound and often tragic effects in some families, but they do not affect every person. Still other traits that are sometimes taught as being controlled by a single gene, such as tongue rolling and widow’s peak, have a highly questionable genetic basis since identical twins can differ in these phenotypes. So, when it comes to familiar single gene traits that every person has, the ABO blood type probably stands supreme; it almost certainly is used more often in murder mysteries and crime dramas than any other genetic trait.
This familiarity does not mean that we know all there is to know about the ABO blood types. While we traditionally (and appropriately) consider three alleles for the gene based on blood clotting assays or agglutination tests in the clinic, there are many more alleles in natural populations, particularly when the DNA sequences are analyzed. Even based on a simple blood clotting test, subtypes for these three alleles can sometimes be detected, with two common and 18 less common subtypes of the IA allele , for example. Nonetheless, the general blood clotting phenotype is convenient to assay and well-known, and is reliable enough for most purposes.
The biological basis of ABO blood types
The underlying biology for ABO blood types is relatively simple, as discussed in the chapter, and depicted in Figure A. The protein encoded by the ABO locus is a glycosyl transferase, an enzyme which attaches certain sugars to a polysaccharide stem molecule (the H antigen) found on the surface of red blood cells. The IA allele attaches the sugar N-acetyl galactosamine, while the IB allele attaches the sugar galactose. The i (O) allele encodes an inactive form of the enzyme, often because of a deletion in the gene or a premature stop codon, and attaches no additional sugars to the stem. The two sugars that are attached act as antigens in an immune response; we have immunological tolerance for the antigen of our own blood type, and we generate isoantibodies to the antigen with the other sugar during our first year of life. Thus, an individual with the IA allele produces antibodies against the antigen made by the IB allele, while a person with the IB allele generates antibodies to the antigen made by the IA allele.
Having neither antigen, a person with type O blood produces antibodies against both alleles. As a result, type O people can donate blood to any other blood type since they are not introducing any antigens that will cause agglutination, so they are sometimes referred to as universal donors. On the other hand, they can’t accept blood of any other blood type since they have antibodies against those antigens. By contrast, a person with type AB blood can receive blood from anyone since both antigens are already present and no antibodies are being produced; ABO individuals are sometimes referred to as universal recipients.
Given the existence of this system, we might ask how and why these blood types exist. Blood types were discovered around 1900 when blood transfusions were first being done, which resulted in large amounts of blood from one individual coming into contact with the blood of an unrelated individual. Austrian physician Karl Landsteiner is credited with identifying type A, type B, and type O blood in 1900, and was awarded the Nobel Prize in Physiology or Medicine in 1930; the less common type AB was discovered a few years after the others. While Landsteiner is credited with this discovery, the Czech doctor Jan Jansky had found all four types at about the same time as Landsteiner but his work was not widely known until years later. Landsteiner also discovered the Rh blood type and was among the team that discovered the polio virus, so the elucidation of the ABO blood types was not his only contribution to human genetics.
What is the evolutionary basis of the ABO blood types?
The time at which blood types were discovered is significant when thinking about the evolutionary biology of ABO. Different blood types were identified only when blood transfers and blood transfusions were being done regularly, which was millions of years after the blood types themselves arose in the human population. In other words, blood types existed long before we had any routine mechanism to detect these differences.
Based on other primates, the evolutionarily ancestral blood type was probably type A. The IB allele is thought to have arisen about 3.5 million years ago while the i allele is thought to have arisen more than 1.5 million years ago. Since i is a non-functional version of the gene, it probably arose many times independently, and other higher primates have a type O that is molecularly distinct from ours. Both the IA and an i allele are present in the genomes of Neanderthals, indicating that they had at least these two types. In other words, the variation in the phenotype that we know best for the ABO locus—blood clotting during transfusion—was almost certainly not the phenotype under which these different alleles arose. The variation at the ABO locus must be attributed to some evolutionary force other than that arising from blood transfusions.
What then was the evolutionary basis for variation at the ABO locus? Why don’t all humans have the same blood type? Have there been specific selective pressures that have led to different ABO alleles? (This is an even more intriguing question for the Rh blood locus since there has been a selective disadvantage known as Rh incompatibility arising when Rh positive children are born to an Rh negative mother.)
The answers are not so clear. It has been recognized for nearly a century that human populations differ in the relative frequency of the three alleles. In particular, the IB allele is absent from some populations such as Native Americans, which has been interpreted by some as evidence for different selective pressures on different blood types. In fact, the evolutionary forces such as selection may not have been directed against the blood types at all; approximately 80% of people secrete their ABO antigens into other body fluids such as saliva and plasma, and the ABO antigens are present on the surfaces of other cells in addition to red blood cells. Secretion of these antigens is encoded as part of the Lewis blood type on chromosome 19. Thus, it may be that the important biological effects of ABO during human evolution occurred on cells other than the red blood cells.
With such a long history of genetic differences for a familiar trait, many hypotheses have been posed about the effect of our ABO blood type on other phenotypes. Most of the more scientific hypotheses are based on correlations between blood type frequencies and various disease or pathogen frequencies. Some of the well-established correlations are summarized in Table A, but more diseases or pathogen correlations have been proposed; all of these correlations are associated with small changes in risk, typically less than 10% over other blood types.
To stress, these are merely correlations, and very little data exist on a mechanism by which a particular blood type may increase or decrease a cancer risk, for example. In the popular literature, ABO blood type has also been suggested to correlate with beauty, dietary behaviors, personality, and recovery from the effects of alcohol among other things, but there is no scientific support for any of these claims (even if a subjective phenotype such as beauty could be assayed accurately).
One hypothesis based on a possible molecular mechanism arises from proposed structural similarities between the antigens and some infectious agents. It has been proposed that the antigen produced by the IA allele, with N-acetyl galactosamine attached to the stem, is structurally similar to a glycoprotein making up the coat of some subtypes of the influenza virus. Thus, antibodies against the type A antigen (as would be found in people with type O or type B blood) may confer a slight resistance to influenza. Similarly, the antigen produced by the IB allele with galactose attached to the stem is proposed to be similar to a glycoprotein component of the cell wall found in Gram negative bacteria such as E. coli, Salmonella sp., and Vibrio cholera, and may confer some resistance to infections by those bacteria.
Yet another idea, and perhaps the most likely, is that the selective pressures on variation at the ABO locus were small; certainly all of the effects we observe now other than agglutination are relatively small in scale compared to other examples of alleles with selective benefits or risks. With such small effects, mutations in the locus arose and may have persisted or been lost based on evolutionary forces other than selection. In Chapter 16, we discuss some of these evolutionary forces as they relate to other traits, and it is possible that these also were important for blood types.