The Basis for ABO Blood Typing Undermines Common Design

When Darwin first proposed the idea of common descent there seemed to be no evolutionary connection between humans and apes. Then the bones and fossils of extinct Neanderthals and other human/ape-like creatures began to come forth--some more like humans, some more like apes--providing confirmation that Darwin was on the right track. Meanwhile scientists began to discover that the proteins of humans and other apes were similar. When the structure and coding of DNA was worked out, scientists began determining the sequence of genes, culminating in the sequencing of whole genomes in the last 15 years or so, including that of Neanderthals. What scientists found in genetic sequence solidified the inferences already arrived at by other means: that among living species, humans are most closely related to chimpanzees and gorillas, with which they share a common ancestor. The evidence for common descent contained in our DNA is rich and multi-layered.

Creationists, however, counter that similarity of DNA is not evidence for common descent. Rather, it is evidence of common design. The basic idea is that since humans and chimpanzees (or any other grouping of species) have similar anatomy and physiology, it makes sense that the Designer would use similar genetic sequences. This assertion certainly has a commonsensical appeal since it casts biology as an extension of human experience in the modern era of mass production, computer programming, and bio-engineering.

The central paradox of this assertion is that in attributing commonalities to common design, one must also attribute to design many ordinary and mundane genetic characteristics that are otherwise explainable on principles of microevolution. Those who make the common design assertion rarely address the patterns of both similarities and differences.

The ABO blood system makes for an interesting case study. The system is a classic example of genetic co-dominance that is often taught in high school biology class, and is also a classic example of the need for matching in blood transfusions or organ transplants. The basis for the system is the pattern of certain molecules expressed on human blood cells and other organs and tissues. Everyone expresses the basic molecule called the H antigen, and if your blood type is O, that's the end of the story. People with A, B, or AB blood-type also have genes for glycosytransferases, which are enzymes that add an extra sugar molecule onto the H antigen--kind of like topping off a Christmas tree with a star, or adding a cherry to an ice cream sundae. The difference between A and B is the type of sugar molecule added to the H antigen. The A enzyme adds one kind, the B enzyme adds a different one. People with type O don't technically lack the transferase genes, they just have versions that are broken due to mutation, though there are no known diseases associated with this. The immune system of people with O or A type will react against type B antigen as a foreign invader, and the same applies to people with O or B in relation to A antigen. People with both genes (A and B) will make both antigens (A and B), and therefore their immune system will not react against A antigen, B antigen, or H (O) antigen.

The A and B glycosyltransferases are actually extremely similar. In fact, there are only four amino acids different between them, and only two of those determine which sugar the enzyme adds to H antigen. I have aligned the protein sequences [1] of the A and B transferases in the following figure. Letters are standard abbreviations for different amino acids (building blocks of proteins), and periods represent identical amino acids. I have highlighted the four differences, and the blue arrows point to the two key amino acids that determine which sugar is added:

If we were to look at the DNA sequence we would see seven differences, instead of just four. However, three of those differences do not change the amino acid sequence. The reason is that many amino acids have several three-letter DNA codes (codons), so changing one of those DNA letters may simply change the three-letter codon to an equivalent codon. Such mutations are called silent because they have no effect on protein function. In contrast, in the four highlighted examples the change in the underlying DNA resulted in a codon for a different amino acid.

These two sequences are so similar that it shouldn't surprise you to learn that scientists believe that one of them originated from the other. At some point one of the genes was duplicated (a common occurrence), and mutations gave one of the two copies a different specificity. And the changes didn't stop there. What I have shown are the two most common sequences for A and B, but some people have variants (alleles) that have other mutations. Most of those mutations are silent, but some of them do change amino acids in various parts of the protein.

The ABO blood groups have been looked at in other animals, and primates have the same basic system. The following (modified) figure shows the evolutionary relationship of various primate species and the blood types that have been found [2]. (Click for bigger image.)

The ABO types are found in varying combinations among hominoids, old world monkeys, and new world monkeys. Previously it was thought that the B alleles were re-created from the A allele in several different lineages. However, a more recent study [2] found that the A and B alleles of the various species are more closely related to one another (A to A, B to B) than any of the A alleles are to the B alleles (with the apparent exception of orangutans). This implies that both A and B were present in the ancestor of primates, and that in some lineages one or the other has been independently lost. In contrast, the O alleles are not all closely related to each other (since there are many ways to break a gene).

Remember, the part of the protein that determines whether the transferase acts as type A or B is dependent on those two amino acids. Interestingly, some primates have a silent mutation in their A gene at one of the important amino acids. I have illustrated this difference with the following figure (compare to the blue arrows in the first figure above):

For each transferase gene, A and B, the DNA sequence is given with the amino acid translation below. The A gene has two versions of the DNA sequence, with the silent mutation (T) in red. Although you might expect CTG and TTG to code for different amino acids, a quick consultation of a codon table shows that both code for leucine (L).

Now here's where things get interesting. Let's look again at the figure showing primate evolutionary relationships, but this time I'll show it without modification [2].

It turns out that the silent mutation in the A gene is only found in old world monkeys; hominoids and new world monkeys have the other version (CTG). Overall, hominoids are more closely related to old world monkeys than new world monkeys, as shown in the figure. But for this piece of the A gene, old world monkeys are different. The simple evolutionary interpretation is that the silent mutation, which is probably selectively neutral, occurred in the lineage of old world monkeys just after they split away from the hominoid lineage. An alternative explanation is that there were multiple independent mutations in the old world monkey lineage. Clearly, several different species independently having the exact same mutation arise and become dominant in the population is of much lesser probability.

So What?

With all of that as background, let's turn our attention to the assertion that similar DNA represents similar design rather than an evolutionary relationship. If the Designer was using similar genetic sequence to make similar organisms, why make a silent DNA change only in old world monkeys?

There are three basic creationist responses that I can think of. First, it could be that each old world monkey species with type A independently had that mutation arise and become fixed in the population. This is essentially the same alternative that was rejected above.

Second, a creationist might argue that the pattern of A alleles represents a sense of whimsy and artistry by the designer. After all, sometimes we do things just because we feel like it. That such alleged artistry happens to also fit an evolutionary scenario might give us pause. At any rate, if the defense of someone's hypothesis when contradicted by the data is simply to assert that the Designer just felt like it, they clearly are not engaged in a scientific discussion and are instead simply seeking to rationalize a conclusion they have already made.

A third creationist argument is more subtle. He/she might note that silent mutations can have an effect on the efficiency with which a protein is made. So perhaps that silent mutation in the A gene actually plays an important role in the context of old world monkey gene expression and physiology. This argument contains at least a grain of legitimate science. That this one silent mutation affects the efficiency of gene expression is highly doubtful, but is testable in principle. However, the notion that it is important for the long-term survival of old world monkeys becomes absurd when you consider that:

1. Like humans, not every member of a particular old world monkey species will have the A gene. Whatever effect the mutation has is irrelevant for those individuals.

2. At least two old world monkey species have apparently lost the A gene altogether! Clearly the mutation has no relevance for those species.

3. As mentioned above, there are other alleles of the A gene found in humans that contain other mutations, and the same holds for old world monkeys. Natural mutations clearly do occur in the A gene. How can naturally occurring mutations be differentiated from designed ones?

Conclusion

Here we summarize and circle back to the paradox of the common design argument. Considering the variability of the presence of A and B genes within individuals, populations, and species, and considering the fact that various polymorphisms exist within each species (some of which are silent, some of which are not), can the difference of a single silent nucleotide really be justified on principles of common design?


Notes:
1. Source of sequences: Human A - http://www.ncbi.nlm.nih.gov/nuccore/58331215; Human B - http://www.ncbi.nlm.nih.gov/nuccore/AB844269.1

2. Ségurel L, Thompson EE, Flutre T, Lovstad J, Venkat A, Margulis SW, Moyse J, Ross S, Gamble K, Sella G, Ober C, Przeworski M. The ABO blood group is a trans-species polymorphism in primates. Proc Natl Acad Sci U S A. 2012 Nov 6;109(45):18493-8. Figure 1 from the paper contained a mistake; the corrected version is here.

---

Continue reading...

The Difference Between Ancestry and Race

The current issue of the journal Science has a policy article discussing the pitfalls of using race in medical and genetic studies (Taking race out of human genetics). It is very easy to conflate race and ancestry, so I was struck by this paragraph that concisely explains the difference:

It is important to distinguish ancestry from a taxonomic notion such as race. Ancestry is a process-based concept, a statement about an individual's relationship to other individuals in their genealogical history; thus, it is a very personal understanding of one's genomic heritage. Race, on the other hand, is a pattern-based concept that has led scientists and laypersons alike to draw conclusions about hierarchical organization of humans, which connect an individual to a larger preconceived geographically circumscribed or socially constructed group.

Maybe that doesn't seem straightforward on a first reading, but it isn't that difficult to unpack. Ancestry is about who your ancestors were and where they came from. Basically it's the process of how you came to be. Race is a categorization that is made based on how you match up to a pattern. Barack Obama, for example, is equally of European and African ancestry. Yet in the United States his divided ancestry is usually collapsed into the racial category of black, based primarily on outward features that match the pattern associated with that race. If he had grown up in northern Minnesota, associated only with white people, and married a woman exclusively of Scandinavian ancestry, he still would never be considered white in the United States because he just doesn't look white (which is to say that his features don't match the pattern associated with being white).

I suppose this is all fairly mundane, but it is very easy to forget the distinction, and perhaps difficult to articulate. If I ever have reason to explain the difference, I'll remember that it boils down to process vs. pattern.

---

Continue reading...