Synopsis: Last time I discussed the existence of several very long palindromes on the Y chromosome. Why would they be there in the first place? Rozen et al. (2003) argue that the Y chromosome basically recombines with itself through gene conversion! But what’s the evidence for such a possibly game-changing assertion?
Skaletsky et al. (2003) observed that roughly 30% of the MSY is >99.9% identical to other MSY sequences (Figure 1). Once the similarity falls below that 99.9%, however, the proportion of sequences showing intrachromosomal similarity falls precipitously. These may indicate distinct regions (which we have previously discussed). Because the similarity is so high, Skaletsky et al. infer this 30% undergoes gene conversion, or non-reciprocal recombination.
Whereas crossing over is reciprocal (both chromosomes exchange genetic material), gene conversion imposes one genetic sequence upon another. In the special case of the Y chromosome, one arm of a palindrome can impose a genetic sequence on the other arm. Theoretically, Y-Y gene conversion could counteract the forces of gene decay: any mutations in the ampliconic genes are copied over by the “correct” version. This may make further sense as the Y chromosome’s testes-specific genes (aside from SRY) are all located within these palindromes. Figure 2 shows how this may be possible.
Is there any direct evidence of Y-Y gene conversion?
Rozen et al. (2003) (of the Page lab) found direct evidence of Y-Y gene conversion in the two CDY1 genes at symmetrical positions in the first palindrome. They looked at a specific mutation in CDY1 (position 381) in 171 different Y chromosomes representing “42 distinct branches of a robust tree of human Y chromosome genealogy.” All of the chromosomes had C/C (C on one arm, C on the other) at this site except for a small cluster of 5 branches which have C/T and T/T (Figure 3). (These mutations were of no functional consequence.) Rozen et al. reasoned that the ancestral genotype was C/T as only one mutation would be required.
Three of the five branches contain both C/T and T/T genotypes which may mean gene conversion sometimes replaced the C with a T. In contrast, one of the branches also has a C/C genotype, indicating gene conversion in the opposite direction. Rozen et al. also found two more examples in CDY1.
Could these variants have arisen by mutation?
Perhaps, but here are several reasons Rozen et al. do not believe so. While nucleotides in the example cited above were switching from C/C to C/T, T/T, and C/C again, As and Gs did not appear. Instead, one base was being substituted by the base on the opposite arm. This does not necessarily refute the case of random mutations, but it does lend more credibility to the gene conversion hypothesis.
Additionally, Rozen et al. note (in the supplementary material):
“Because the C sequence variant corresponds to a C in a CpG dinucleotide, with an expected high rate of C-to-T mutations, we considered the alternate possibility that observations of the C/T chromosomes could be due to recurrent C-to-T substitutions. However, this possibility is inconsistent with the absence of other instances of this mutation in 133 chromosomes (37 haplotypes) outside of the highlighted cluster.”
So even taking into account a mutation bias, Rozen et al. reason that gene conversion is still a viable hypothesis. Rozen et al. also claim they found two more examples of Y-Y gene conversion in CDY1.
Rozen et al. (2003) further estimated the rate of gene conversion. Using time since divergence and assumptions of mutation rates, they calculated the estimated rate of gene conversion to be 2.2 x 10-4 conversions per duplicated nucleotide per generation, or in other words, 600 duplicated nucleotides experience gene conversion for every son born in recent human evolution.
These results have two important implications.
The first is that the Y chromosome actually recombines! With itself!
The second is that gene conversion may “save” the Y chromosome! If an arm suffers a mutation, gene conversion from the other arm can fix the mistake.
However, there is a problem: In the example above, some of the C/T chromosomes were converted to T/T, not the ancestral C/C. In other words, gene conversion is adirectional. It could very well further doom the Y, and not save it. My next post will investigate this problem.
Rozen, S., Skaletsky, H., Marszalek, J., Minx, P., Cordum, H., Waterston, R., Wilson, R., & Page, D. (2003). Abundant gene conversion between arms of palindromes in human and ape Y chromosomes Nature, 423 (6942), 873-876 DOI: 10.1038/nature01723
Skaletsky, H., Kuroda-Kawaguchi, T., Minx, P., Cordum, H., Hillier, L., Brown, L., Repping, S., Pyntikova, T., Ali, J., Bieri, T., Chinwalla, A., Delehaunty, A., Delehaunty, K., Du, H., Fewell, G., Fulton, L., Fulton, R., Graves, T., Hou, S., Latrielle, P., Leonard, S., Mardis, E., Maupin, R., McPherson, J., Miner, T., Nash, W., Nguyen, C., Ozersky, P., Pepin, K., Rock, S., Rohlfing, T., Scott, K., Schultz, B., Strong, C., Tin-Wollam, A., Yang, S., Waterston, R., Wilson, R., Rozen, S., & Page, D. (2003). The male-specific region of the human Y chromosome is a mosaic of discrete sequence classes Nature, 423 (6942), 825-837 DOI: 10.1038/nature01722