Identifying the function of a locus of adaptation in high-altitude deer mice

Posted by: Bryn Gaertner

Evolutionary and functional insights into the mechanism underlying high-altitude adaptation of deer mouse hemoglobin.

Storz JF, Runck AM, Sabatino SJ, Kelly JK, Ferrand N, Moriyama H, Weber RE, Fago A.  Proc Natl Acad Sci U S A. 2009 Aug 25;106(34):14450-5. Epub 2009 Aug 10. PubMed PMID: 19667207; PubMed Central PMCID: PMC2732835

This paper performed both evolutionary and functional analysis of duplicated genes adapted to altitudinal zones.

They found parallel functional differentiation at multiple unlinked duplicates, the two alpha-globin genes on chromosome 8 and the two beta-globin genes on chromosome 1.  In this paper, they found that there is a change in beta-chain binding affinity: The difference is in 2,3-DPG affinity, which is an allosteric cofactor that stabilizes the de-oxy conformation of the heme molecule.  This difference encourages O2 binding in high altitude environments, which would enhance pulmonary loading under hypoxia.  However, our discussion lead us to wonder whether this is actually a *low* altitude adaptation


The Big question this paper addresses is how do we get adaptation in different environments, particularly in proteins with a complex quaternary structure?  The hemoglobin molecule is actually composed of two alpha- and two beta-chains.  Hemoglobin forms a heterotetramer by combining two alpha and two beta chains.  This creates two semi-rigid dimers.  When oxygen binds, both dimers rotate.  When oxygen is not bound, then allosteric cofactors come in and bind to positively-charged amino acids to stabilize the unbound molecule.

Modifications to this interaction have already been implicated in hypoxia survival, particularly in high-altitude mammals.  But how do we make this fragile interaction among many dimers work?

The answer may be by changing affinity not necessarily for oxygen (for which you’re more likely to get deleterious mutations), but for these allosteric cofactors. In particular, the 2,3 diphosphoglycerate (DPG) and Cl- ions.  Usually, these two cofactors bind to positively charged sites in deoxy hemoglobin.  This stabilizes the molecule, making it harder for oxygen to rebind.  But, this would require coordinated changes in alpha and beta subunits, and these genes are unlinked in the genome.

There is a history of studying hemoglobin binding affinity on an altitudinal cline using deer mice.  Deer mice have a wide geographic range spreading two or three different mountain ranges, plus all the land in between.  It had previously been shown that adaptive variation is associated with allelic variation in the two tandem alpha chains.  The two-locus genotype gives high blood-O2 affinity, superior aerobic performance under high altitude hypoxia or cold temperatures.  What is also interesting is that there is intermediate performance in heterozygotes, which is consistent with latitudinal clines.

The reason this allele enhances performance is because high affinity Hb facilitates pulmonary O2 loading in hypoxic conditions.  However, the cost is clear in low altitude environments: it’s hard to get rid of the oxygen once it’s bound, so you’re starving your muscles of oxygen even though there’s plenty in your bloodstream.

There is strong directional selection for this pattern of adaptation when looking at alpha-globin allelism, where the high-affinity allele is very frequent in high altitude, and low affinity is fixed or nearly fixed in low populations.  The patterns of linkage disequilibrium and nucleotide divergence are consistent with these conclusions.  Beta-globin is a more confusing story.  Electromorphs of the protein indicate that there are amino acid substitutions between the high- and low- altitude mice, but performance testing didn’t find any difference in performance or blood-O2.

Experiment 1: What are the patterns of b-globin polymorphisms?

Thin layer isoelectric focusing identifies differences in proteins that would change their electrical properties.  They found that there are two distinct b-globin variants that appeared to be segregating Mendelianly.  However, there are definitely two copies of the beta-globin gene.  Though there is plenty of recombination between the two genes, within one chromosome you only find one morph of the gene.  The authors interpret this as a history of interparalog conversion, just like what is in the alpha-globin gene.  This demonstrates very marked linkage disequilibrium.

They then cloned and sequenced the two alleles (now two haplotypes) and found that there are four AA substitutions between the two.  Both genes show lots of nucleotide divergence.  The amino acid substitutions exist on the E- and H- helicies, which are right around the binding pocket of the heme group.

Experiment 2: Is there selection do to altitude between these populations?

To perform this experiment, they first had to simulate what the population structure would look like under a null model.  They used eight “neutral” genes to simulate what the level of heterozygosity and LD should look like if there is no population structure, then tested the hypothesis of population structure at the beta-globin loci.  They found that there is more nucleotide divergence when compared to the neutral genes. They also found that the nucleotide divergence is partitioned differently in the alpha- vs beta-globin genes: There was more divergence in HBB and more LD.  Thus, there is local adaptation in HBB and the alternative alleles are maintained as balanced polymorphism by spatially varying selection.  So, perhaps HBB is doing something after all.

Experiment 3: what’s the function of this spatially varying selection?

To assay the function, the group compared binding affinity of the blood from mice in both high- and low-altitude.  They controlled for the alpha-hemoglobin polymorphisms by using two groups of mice, ones with one type of substitution and one with another.  In this assay, you test how much oxygen you have to put in the chamber to get 50% of the hemoglobin to bind.  With no cofactors, all four groups (high and low beta-globin, high- and low- alpha-globin) required very little oxygen for saturation.  After adding DPG, the low-beta-globin groups had a pronounced right-shift in the curve.  THis means it takes more oxygen to get saturation.  However, this shift was not seen in the high-beta-globin group: here, the curve stayed the same.  THis suggests that even with DPG, the affinity of this allele remains high.  Or, it is not sensitive to DPG.  However, this effect disappears with the addition of KCl solution, which would increase the amount of Cl ions hanging around.  Now the curves look the same.  Adding DPG to the KCl treatment doesn’t have an additional effect.

The authors conclude that this allele increases O2 affinity by decreasing DPG sensitivity.  This mirrors the patterns of altitude differences in alpha-hemoglobin and demonstrates that both the alpha- and beta-globin subunits of hemoglobin are important for adaptation to high-altitude environments.

Our Discussion

In discussion we determined that there was a moderate disconnect between the authors’ experimental results and the conclusions they drew.  Though we were all impressed at the level of functional assays performed, we couldn’t quite make sense of the results.  Certainly part of this is due to the fact that none of us are physiologists.  However, the result with the Cl- ions are odd.  Is Cl- a limiting factor in red blood cells?

One member of our group had an interesting take on the situation.  A textbook example of acclimation is as follows: When a low-altitude individual goes into high-altitude, many immediate physiological adaptations occur.  We breathe more, meaning our blood becomes more acidic.  This could limit Cl- circulating in the bloodstream, and it shifts the loading curve of oxygen farther to the left.  DPG production is upregulated and reshifts the curve back to the right.  By itself this is interesting, but particularly important when considering the role of myoglobin, which picks up the oxygen bound in hemoglobin for use in the muscles: Myoglobin is far left-shifted relative to hemoglobin on a loading curve; if the curves are too close together then you don’t get oxygen to your muscles.  High-altitude mice are insensitive to this DPG-induced shift, and these mice in particular were sampled from Mt. Evans, which is one of the highest points in the continental US.  So perhaps this is more of a retained adaptation for low-altitude mice than a derived adaptation for high-altitude mice.  In keeping with this though process, it seems like the mutations in beta-globin could be more important for releasing oxygen, where the mutations in alpha-globin are more imporant for binding oxygen.  Their coevolution is consistent with this hypothesis, and may help explain why they weren’t able to find a performance effect due to beta-globin polymorphisms.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s