More Recent Comments

Monday, May 26, 2008

Centromere DNA

During mitosis in eukaryotic cells the chromosomes are duplicated and the two sister chromosomes separate and move to opposite ends of the dividing cell. This segregation is controlled by spindle microtubules that attach to specific regions of the chromsomes called centromeres.

Centromeres are easily seen in the light microscope following chromosome condensation. They appear as a constricted region where the daughter chromosomes remain attached to each other. In non-dividing cells the centromere region is heterochromatic, which means that it remains relatively condensed compared to the rest of the chromatin that contains active genes (euchromatin).

Yeast centromeres are very simple but mammalian centromere DNA has not been extensively characterized because it consists largely of multiple repeats of simple sequence DNA. Because of the repetitive nature of centromeric DNA these region are difficult to clone. They are missing from the human genome database.


Genomes & Junk DNA

Total Junk so far

Nevertheless, we have a pretty good idea of the organization of centromere DNA from the few centromeres that have been sequenced. In humans the dominant repeat is α satellite DNA, a 171 bp sequence that is repeated about 18,000 times at an average centromere. Kinetochore proteins bind to the central region of the centrosome and the spindle microtubules attach to the kinetochore (Cheeseman and Desai, 2008).

Fluorescent hybridization studies with α satellite DNA light up all centromeres on human chromosome indicating an abundance of α satellite DNA at all centromeres. We don't know how much of this DNA is essential for chromosome segregation. There are rare examples of neocentromeres (newly formed centromeres) that have very little α satellite DNA suggesting that much of it is non-essential. Artificial human chromosomes segregate at mitosis with only a few copies of α satellite DNA at their centromeres.

Not all α satellite DNA is associated with functional centromeres since the presence of inactive, nonfunctional centromere sequences in the human genome is well known. (Such as one of the ancestral centromeres associated with the formation of human chromosome 2 from a fusion of two separate primate chromosomes. See Stanyon et al. (2008) for a review of the evolution of primate chromosomes with an emphasis on the formation of new centromeres and the loss of ancient ones.)

There are also at least 68,214 monomeric α satellite sequences in the human genome (Alkan et al. 2007).

Human centromeres range from 0.3Mb to 5Mb in size (Cleveland et al. 2003). If the average centromeric region is 3Mb (3,000 kb) in size then 23 centromeres represents 2% of the entire genome sequence. Not all of this DNA is essential because, among other reasons, there is considerable variation between individuals in the length of a given centromere. Nevertheless, lets assume for the sake of our junk DNA calculation that all of it is essential.

Monomeric α satellite sequences make up about 0.3% of the genome (Alkan et al. 2007). These bits of DNA are almost certainly non-essential "escapees" from centromeric regions or fossil centromeres. The total amount of α satellite DNA in the human genome is between 2% and 5%. The vast majority of these sequences are not in the databases. If we add in the fossil centromeres we can estimate that the total amount of junk α satellite DNA comes to about 1% of the genome.

[Image Credits: The drawing of a centromere is from Alberts et al. (2002) Figure 4-50. The photograph of chromosomes is from Hunt Willard (Schueler et al. (2001)]

Alberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K. and Walter, P. (2002) The Molecular Biology of the Cell 4th ed., Garland Science, New York (USA)

Alkan, C., Ventura, M., Archidiacono, N., Rocchi, M., Sahinalp, S.C., et al. (2007) Organization and Evolution of Primate Centromeric DNA from Whole-Genome Shotgun Sequence Data. PLoS Comput Biol 3: e181. [doi:10.1371/journal.pcbi.0030181]

Cheeseman, I.M. and Desai, A. (2008) Molecular architecture of the kinetochore–microtubule interface. Nature Reviews Molecular Cell Biology 9:33-46. [doi:10.1038/nrm2310]

Cleveland, D.W., Mao, Y., and Sullivan, K.S. (2003) Centromeres and Kinetochores From Epigenetics to Mitotic Checkpoint Signaling. Cell 112:407-421. [doi:10.1016/S0092-8674(03)00115-6 ]

Schueler, M.G., Higgins, A.W., Rudd, M.K., Gustashaw, K. & Willard, H.F. (2001) Genomic and genetic definition of a functional human centromere. Science 294:109-115.

Stanyon, R., Rocchi, M., Capozzi, O., Roberto, R., Misceo, D., Ventura, M., Cardone, M.F., Bigoni, F., and Archidiacono, N. (2008) Primate chromosome evolution: Ancestral karyotypes, marker order and neocentromeres. Chromosome Research 16:17-39. [doi: 10:1007/s10577-007-1209-z]


RPM said...

The accumulation of satellite DNA at cetromeres (and telomeres) may be, in part, because selection is not as efficacious in regions of low recombination. Therefore, slightly deleterious insertions of repetitive sequence are not purged. That's not to say that some of it is not useful for proper segregation.

Also, some of the most interesting work on mammalian centromeres comes from marsupials, where there are a lot of robertsonian fusions and centromeric shifts.

Anonymous said...

rpm: what makes you think it's not recombining efficiently? since it's so repetitive, how could you tell?

RPM said...

Empirical measurements of recombination rate show that it is decreased near centromeres. It's not recombination that's not efficient, but selection is less efficient in regions of low recombination.

Larry Moran said...

RPM says,

It's not recombination that's not efficient, but selection is less efficient in regions of low recombination.

Isn't selection for a single locus independent of recombination?

I think what you mean to say is that deletions and insertions of centromeric repeats do not occur as often as they would if the repeats were located in the chromosome arms.

This is supported by the fact that fossil centromeres deteriorate rapidly once they stop acting as centromeres.

RPM said...

Selection at a single locus is independent of recombination, but genomes aren't composed of independent loci.

Selection is less efficient in regions of low recombination because of interference between loci. A selective sweep at a locus drives all linked alleles to fixation, regardless of whether they are deleterious of advantageous. Also, the removal of deleterious mutations removes all linked variation, even if they are beneficial alleles (Muller's ratchet).

Therefore, fixing advantageous mutations and purging deleterious ones will be less efficient in regions of low recombination. So, we expect the accumulation of deleterious variants (like repetitive insertions).