One of the things I've been doing recently is marinating myself in the literature on avian genomes. My interest in it was sparked by an offhand conversation with my boss about birds and their small genomes, and so this post is going to be the first in a series that will be part historical, part summary, and part speculative, so as to clarify my understanding about the issue. I hope I don't foul it up!
The fact that avian genomes are, as a general characteristic of their clade, much smaller than the average vertebrate genome has been known for decades.[1] Many people have tried to explain this size discrepancy with both adaptive and non-adaptive arguments, with some ideas perhaps more convincing than others in my limited understanding. But whatever the explanation, the small nature of the avian genome likely had a hand in derived adaptations in avian immunity, specifically B cell biology. B cells were discovered in the 1950's when researchers (among them Max Cooper who now plays with lampreys) removed an organ called the "bursa of Fabricius" from baby chickens. These bursa-less chickens demonstrated a loss of antibodies, which had two big implications in that: 1) cells in the bursa probably had a role in making antibodies, and 2) there were fewer (but not zero) cells called lymphocytes in the blood. The latter implication set up for the discovery of T cells and also illustrated that lymphocytes were not a homogenous population of small uninteresting cells and deserved further study. Anyway, cells that make antibodies were named B cells, B for bursa. The funny thing is that researchers then tried to look for a bursa in mammals but failed because mammals don't have a bursa. Mammals and most other vertebrates that aren't fish or birds make B cells from the bone marrow, thus confusing many an undergrad who naively assume B stands for bone.[2] My point in this quick historical tangent is that avian B cell biology substantially differs in some aspects from the normal (mammalian) state of affairs.
So, how different? For one, the molecular mechanism by which avian B cells generate their repertoire of B cell receptors, i.e. antibodies. I'm going to assume the reader (you!) already knows of the mechanism by which mammals largely generate their repertoire, but Wikipedia has an entry on it. Anyway, avian B cells still undergo V(D)J recombination but the resulting repertoire is of very limited diversity due to a much smaller pool of V, D, and J segments to paste together. However, B cells then undergo a process called gene conversion in a manner analogous to what I blogged previously about in lampreys. If you didn't read that particular post, then essentially what happens is that an enzyme called AID causes DNA damage within the rearranged receptor. This then recruits a DNA repair process called gene conversion that fixes DNA damage by copying from a similar but not necessarily the exact same sequence and substituting it for the chunk of DNA containing the damage. I say not necessarily the exact same sequence because this DNA repair process normally borrows from other cognate chromosome, as diploid cells are wont to do. But what actually happens is that the repair process borrows chucks of DNA from a family of pseudo-V(D) and V genes upstream respectively of the real heavy and light chain genes. Iterative processes of this modified gene conversion introduce diversity into the initially limited V(D)J rearrangements and therefore generate antibody diversity in birds. This method of generating antibody diversity depends, in part, on the physical distance between the donor pseudogenes and the actual receptor gene. That is, the closer a donor is to the acceptor, the more likely that donor is used in gene conversion. The end result is that the distances between all the component parts of the heavy and light chain loci have to be shorter than is the case in mammals. By way of comparison, the chicken light chain locus is about 25 kb; the human lambda chain locus is about 885 kb. So, small genome, small Ig locus. As we'll see, the miniaturisation of avian genomes seems to have miniaturised everything, from intergenic spaces to introns and also the Ig loci. So provisionally, it seems likely that the evolution of a small genome was a necessary pre-adaptation for the unique aspects of avian B cell biology.
Edited Oct. 11, 2010 for clarity
The fact that avian genomes are, as a general characteristic of their clade, much smaller than the average vertebrate genome has been known for decades.[1] Many people have tried to explain this size discrepancy with both adaptive and non-adaptive arguments, with some ideas perhaps more convincing than others in my limited understanding. But whatever the explanation, the small nature of the avian genome likely had a hand in derived adaptations in avian immunity, specifically B cell biology. B cells were discovered in the 1950's when researchers (among them Max Cooper who now plays with lampreys) removed an organ called the "bursa of Fabricius" from baby chickens. These bursa-less chickens demonstrated a loss of antibodies, which had two big implications in that: 1) cells in the bursa probably had a role in making antibodies, and 2) there were fewer (but not zero) cells called lymphocytes in the blood. The latter implication set up for the discovery of T cells and also illustrated that lymphocytes were not a homogenous population of small uninteresting cells and deserved further study. Anyway, cells that make antibodies were named B cells, B for bursa. The funny thing is that researchers then tried to look for a bursa in mammals but failed because mammals don't have a bursa. Mammals and most other vertebrates that aren't fish or birds make B cells from the bone marrow, thus confusing many an undergrad who naively assume B stands for bone.[2] My point in this quick historical tangent is that avian B cell biology substantially differs in some aspects from the normal (mammalian) state of affairs.
So, how different? For one, the molecular mechanism by which avian B cells generate their repertoire of B cell receptors, i.e. antibodies. I'm going to assume the reader (you!) already knows of the mechanism by which mammals largely generate their repertoire, but Wikipedia has an entry on it. Anyway, avian B cells still undergo V(D)J recombination but the resulting repertoire is of very limited diversity due to a much smaller pool of V, D, and J segments to paste together. However, B cells then undergo a process called gene conversion in a manner analogous to what I blogged previously about in lampreys. If you didn't read that particular post, then essentially what happens is that an enzyme called AID causes DNA damage within the rearranged receptor. This then recruits a DNA repair process called gene conversion that fixes DNA damage by copying from a similar but not necessarily the exact same sequence and substituting it for the chunk of DNA containing the damage. I say not necessarily the exact same sequence because this DNA repair process normally borrows from other cognate chromosome, as diploid cells are wont to do. But what actually happens is that the repair process borrows chucks of DNA from a family of pseudo-V(D) and V genes upstream respectively of the real heavy and light chain genes. Iterative processes of this modified gene conversion introduce diversity into the initially limited V(D)J rearrangements and therefore generate antibody diversity in birds. This method of generating antibody diversity depends, in part, on the physical distance between the donor pseudogenes and the actual receptor gene. That is, the closer a donor is to the acceptor, the more likely that donor is used in gene conversion. The end result is that the distances between all the component parts of the heavy and light chain loci have to be shorter than is the case in mammals. By way of comparison, the chicken light chain locus is about 25 kb; the human lambda chain locus is about 885 kb. So, small genome, small Ig locus. As we'll see, the miniaturisation of avian genomes seems to have miniaturised everything, from intergenic spaces to introns and also the Ig loci. So provisionally, it seems likely that the evolution of a small genome was a necessary pre-adaptation for the unique aspects of avian B cell biology.
Edited Oct. 11, 2010 for clarity
1. Gregory TR. 2002. A bird's-eye view of the C-value enigma: genome size, cell size, and metabolic rate in the class Aves. Evolution. 56(1):121-130.
2. As I'm writing this, I keep thinking of those interactive toys that teach kids about farm animals. I'll tell you, higher education would be a breeze if all I did everyday is to play with toys.
