Skip navigation

Every molecule in your body has a story to tell. This tale is about hemoglobin, the class of oxygen-binding molecules that make oxygen transport in complex creatures such as us possible. Hemoglobins are proteins that contain the heme moiety, a planar molecule binding iron (Fe) and maintaining it in the +2 oxidataion state (denoted Fe(II)).

Hemoglogins in the Animal Kingdom
Hemoglobins in almost all vertebrates is a heterotetramer (that is, four (‘tetra’) individual of different (‘hetero’) proteins) composed of two α-globin and two β-globin polypeptides (a fancy word for proteins). The heme part is bound in a pocket formed by structure of each globin unit. The movement and interactions between the subunits increases hemoglobin’s ability to bind oxygen.

Hemoglobin structure

Hemoglobin structure

The α- and β-globin subunits themselves are quite different in their amino acid sequence, with only 50% homology no matter what vertebrate you took the blood from. Thus, it is necessary to look to animals that have a common ancester quite a way back in time from us and most other vertebrates. If animals having a homotetrameric structure (that is, hemoglobin composed of four identical globin protein units) were found ‘higher up’ on the tree of life (say, for instance, a rabbit), then evolution itself would be falsified and such an instance would require a new explanation.

This is not the case, however. We have to go back to modern representatives of some of the earliest vertebrates – the jawless fishes, which include the lampreys and hagfish. Technically, the aren’t really vertebrates, but are representive of intermediate forms on the way towards evolving with vertebrae, though they do have a spinal cord. A very old body form indeed.

The hemoglobin of jawless fishes (Agnatha) are the latest common ancestors to have a single hemoglobin gene. The α- and β-globin genes that continue on to present day occurred through gene duplication and mutation along with natural selection and/or genetic drift some 650 Mya. As is predicted by evolutionary theory, no animals with common ancestry prior to the appearance of the genus Agnatha have a β-globin gene.

A very few, or even a single amino acid substitution in hemoglobin can produce startling changes in oxygen affinity. This allows for adaptation to extreme environments such as high altitudes as experienced by migrating birds and llamas in the Andes, alters the thermodynamics of oxygen exchange in tuna and other endothermic fish to ensure that oxygen doesn’t simply bubble away, enables reindeer and yaks to survive extreme cold, and human fetal hemoglobin (which is different from the adult form) enables the fetus to dissipate energy produced metabolically and aids in maintaining a constant body temperature.

Hemoglobins in Plants
Plants use oxygen in the respiratory chain of mitochondria, the energy-producing cellular organelles. Hemoglobins are used to transport oxygen to mitochondria. The first hemoglobins were found in the root nodules of legumes. While these leghemoglobins differ considerably in their amino acid sequences (a whopping 80%), their three-dimensional structures are very similar to those found in vertebrates. This argues strongly for common ancestry with a very ancient concestor, to use Richard Dawkins word for ‘common ancestor’. I rather like that terminology.

The existence of leghemoglobins in a very specific group of plants at first was a source of consternation, but in the years since it has been found to be quite common within the plant world. If this had not happened, yet again an explanation would have been necessary. Perhaps horizontal gene transfer (something which is common in bacteria) would have sufficed so long as there was ample experimental evidence, but it would definitely have been a problem. Or perhaps leghemoglobin is a product of a divergence from an ancient plant hemoglobin gene.

Evolution of Hemoglobins

Evolution of Hemoglobins

But it turns out that the latter possibility is the likely explanation for the existence of leghemoglobin, particularly in light of the fact that distinctly different hemoglobins have been found in nonleguminous plant root nodules. In fact, hemoglobins abound in both monocots and dicots, and two classes of hemoglobins have been characterized: nonsymbiotic and symbiotic types. The nonsymbiotic type is the more widely distributed hemoglobin type and the symbiotic type is found in the root nodules of only a relatively few plant families (two so far). One very interesting finding is that in the legume soybean a hemoglobin has been found closely resembling nonsymbiotic hemoglobin. The transcribed messenger RNA (mRNA) is quite scarce, much less abundant than the mRNA from the symbiotic form. This is very strong evidence for the source of the leghemoglobin gene resulting from a gene duplication event, which would have occurred prior to the separation of monocots and dicots (roughly 150 Mya).

Hemoglobins in Protists, Fungi and Bacteria
More recently, it has been shown that the hemoglobin gene precedes even the prokaryote/eukaryote split. A variety of intron/exon arrangements exist in protists: some have no introns while others have several. The homology that exists in the hemoglobin gene, however, points to several possibilities: many intron insertions in the evolutionary tree with differential loss of introns, or a good deal of intron sliding, or finally repeated insertions of introns.

Yeasts contain a flavohemoglobin, a fusion of a heme-binding domain and a flavin adenine dinucleotide (FAD) binding domain. FAD serves as an oxidizing agent in a number of metabolic pathways, including the citric acid cycle. This yeast version of hemoglobin has no introns.

Bacteria also have flavohemoproteins and while their heme-binding domain is somewhat different from plant or animal hemoglobins, the similarity of their three-dimensional structure to that for animals and plants argues in favor of a homology between bacterial and eukaryotic hemoglobins.

Hemoglobins Connect Almost All Life on the Planet
We normally think of hemoglobin in the context of our own blood, comprising about 85% of the dry weight of red blood cells and a whopping 35% wet weight. For a protein, this is extremely concentrated. What surprises most people is the ubiquitousness and varied functions of hemoglobins in not just the animal world. Plants use hemoglobins, as do fungi, protists and bacteria. The gene encoding hemoglobin is thus very old, going back to the ancestor common to essentially all life on this planet. In fact, the only group for which hemoglobins have not been found are within the kingdom Archaea, considered to be the living relatives of the most ancient form of life we know about on this planet.

How do we know that this is the case, that there is a common link between all species of life through the hemoglobin gene? In plants the hemoglobin gene (and leghemoglobin gene, evidence of a common source for this class of molecules as well) is separated into four exons (parts of the gene which are transcribed into RNA) by three introns (parts which are ‘edited out’ by RNA transcriptase, the protein which does the transcribing). The first and third introns are in positions homologous to those in the α- and β-globin and myoglobin genes. The second intron interrupts the section of DNA encoding the E helix structure of hemaglobin. This second intron is also part of the gene encoding leghemoglobin, significant evidence pointing to both stemming from a common ancestral hemoprotein gene.

The available evidence points to the ancestor to plants and animals having a hemoglobin gene with three introns. Nematodes have this intron arrangement in their hemoglobin gene, but the location of this intron is not the same as for plants. At this point it is unclear as to whether this represents a significant divergence from the gene that the common ancestor had, is a result of ‘intron sliding’, even independent intron insertion events is unclear and remains an open question. Even so, the evidence for a hemoglobin gene from a common ancestor living about 1500 million years ago (Mya) is compelling.

Regardless, this central intron was lost prior to the rise of Annelids (earthworms) and Arthropods (insects, spiders, scorpions, etc.) about 670 Mya. As a result, the genes for globin vertebrates do not contain this central intron. If they did, common descent would again be hard-pressed to explain the anomaly. The homology of the first and third introns makes for a testable hypothesis. It is possible to perform a statistical analysis using the variation within the intron codons and plot this against a priori knowledge of the evolutionary pathway to show whether the hypothesis of a common ancestry for hemoglobin in plants and animals extends as far back as 1500 Mya.

A direct comparison of the primary amino acid sequence is probably not all that useful. Even if a gene sequence is conserved piece-wise in time, the process of mutation and natural selection and/or genetic drift would probably make hemoglobins from common ancestors separated by over 1500 Mya unrecognizable. And it isn’t all that well conserved. To date, about 1000 mutations in the globin genes have been described in humans alone. More useful is the comparison of three-dimensional structures by x-ray crystallography or NMR spectroscopy as a test for homology.

A great deal of speculation has gone into how hemoglobins arose in the first place. Perhaps it initially evolved in order to help protect the cell from the continually rising amounts of the toxin known as oxygen from photosynthetic organisms. Once cells began to use oxygen in respiratory processes, hemoproteins could act as electron-transfer agents or to scavenge oxygen for respiration. Gene duplications with mutation would allow other redox agents (like the cytochrome b molecules) to evolve from the ancestral hemoprotein gene. Once multicellular organisms arose, hemoglobins could then evolve into their current function of oxygen transporers. At an early stage of evolution, it has been proposed that worm-like animals had large polymeric aggreggations of hemoglobin subunits circulating through primitive circulatory systems. Next would be monomeric hemoglobin in blood cells functioning as oxygen storage.

An Animal Without Hemoglobin?
About 80 years ago there were rumors of a fish that had colorless blood. Biologists treated it like El Chupacabra or Big Foot and ignored Norwegians who fished off the Antarctic. That was until someone brought in a specimen. Their blood is almost totally plasma, with < 1% leukocyte content and almost no erythrocytes (which themselves contain absolutely no hemoglobin). Today, there are 15 known species of icefish of the family Channichthytidae and all are endangered by global warming for reasons we shall see momentarily.

How does the icefish accomplish this feat? In two ways: first, the metabolism of these fish is very slow, even by fish standards. Second, oxygen dissolves in water better at lower temperatures than at higher. The oxygen-saturated waters in which icefish inhabit are indeed very cold – below 2°C, and often below 0°C. This is the reason that global warming, if not stopped in its track, will make the icefish extinct.

As Theodosius Dobzhansky once stated, “Nothing in Biology makes sense except in the light of Evolution”. Is there an evolutionary explanation for the lack of hemoglobin? The original hypothesis of JT Ruud was that the inactivation of genes encoding hemoglobin (lethal for fishes in warmer waters) was selectively neutral or deleterious but non-lethal. If so, there should then be evidence of the presence of one-time working globin genes.

It turns out that the gene encoding the β-globin unit is completely missing from the Chanichthyid genome, having been removed through a gene deletion mutation event. The gene encoding the α-globin unit has been inactivated and truncated, and some of intron 2 and all of intron 3 remain in the sequence. Even the embryonic and juvenile forms of globin have been deleted or inactivated. Ruud was right on track in his hypothesis.

Chalk up another one to Evolution.

Hemoglobins are an ancient class of molecule uniting almost all forms of life on this planet. Only in the most ancient organisms, those species of the kindom Archaea, is hemoglobin unknown. It’s presence in a number of roles demonstrates a common theme in evolution, namely new tricks for old dogs. These various functions also give us some ideas on how hemoproteins arose and evolved.


  1. di Prisco G, Cocca E, Parker SK and Dietrich III HW. Tracking the evolutionary loss of hemoglobin expression by the white-blooded Antarctic icefishes. Gene 295:185-191 (2002)
  2. Hardison RC. A brief history of hemoglobins: Plant, animal, protist, and bacteria. Proc Natl Acad Sci USA 93:5675-5679 (1996)
  3. Hardison R. Hemoglobins from bacteria to man: Evolution of different patterns of gene expression. J Exp Biol 201:1099-1117 (1998)
  4. Wells RMG. Evolution of haemoglobin function: Molecular adaptations to environment. Clin Exp Pharm Physiol 26:591-595 (1999)
  5. Clementi ME, Condo SG, Castagnola M and Giardina B. Hemoglobin function under extreme life conditions. Eur J Biochem 223:309-317 (1994)


  1. Yet another thread connecting us to our ancestral life. Thanks for this, but I can hear the argument from creationists: common design because it works. I had written a post a long time ago on the existence of genes for fingers, and not hands, in sharks. A creationist patiently explained to me that the Creator pre-loaded the genes because of his plan for tetrapods.

    • shamelesslyatheist
    • Posted November 13, 2008 at 9:54 am
    • Permalink
    • Reply

    Such arguments are easily countered. My standard answer to that kind of nonsense is that, as Carl Sagan noted, such a claim is extraordinary and requires extraordinary evidence. Claiming that a creator “pre-loaded” the genes is not only extraordinary, but unverifiable. Unverifiable claims are baseless rationalistic nonsense. It was exactly these kinds of unverifiable claims which led Christians who were in the early years of what we now call Science to discard these dogmatic answers and search for real ones.

    “Common design because it works” fails when one considers the traceability of pseudogenes, genes which do absolutely nothing and are thus not essential or even noticed and so does not explain their presence in the genome. A prime example of this is the NANOGP8 gene we share with chimps, right down to the identical inclusions of Alu retroelements, yet it is a gene which transcribes no RNA and is thus just occupying space in the chromosome. This is not explainable by the template hypothesis. Why include useless genes?

    In the end, however, these answers are good enough for them because they fit into their world view and confirm their bias. They will ignore real answers no matter how much reality bites.

    Thanks for the comment!

2 Trackbacks/Pingbacks

  1. […] Original post: Hemoglobin: A Molecule That Connects All Species […]

  2. By Tangled Bank #118: Yes We Did Edition on 12 Nov 2008 at 4:16 pm

    […] – is based. New research continues to demonstrate the vitality of evolutionary biology, from the ubiquity of hemoglobin (the presence of which even in plants demonstrates common ancestry, and the ability of “old […]

Leave a Reply

Please log in using one of these methods to post your comment: 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

%d bloggers like this: