I like to think about chromosomes as Torah scrolls in the Arc of the cell nucleus. As the Torah contains the germ of all Jewish law, the chromosomes are repositories of all the information that the cell uses to do its business. Like the Torah, the chromosome is a copy of a copy of a copy of an ancient document. Like the Torah, the chromosome is scrolled up most of the time for safekeeping, wrapped around a molecular spool made of protein (histone). And like the Torah, the chromosome is useless unless it is read; and yet every time it is read, there is a risk of damage. The Torah can be torn, the chromosome becomes broken.
The DNA consists of two complementary strands. When one breaks, then repairing the other one is easy— the complementary strand is used as a guide. But when double-stranded breaks occur, proper repair is a complex problem.
The cell has several redundant methods and a great deal of machinery for repairing double-stranded breaks. The machinery is very similar, not just in plants and animals, but also in bacteria and archaea. Evidently, DNA repair evolved very early in the history of life, and has been conserved for well over a billion years.
The person who repairs the Torah is called a Sofer, and he has been apprenticed for years, memorizing more than a thousand rules about calligraphy, materials and fabrication. DNA is repaired by smart protein molecules, which have no apprenticeship.
Making a mistake in DNA repair can be fatal for the cell, so there’s a lot of motivation to do it right. If one letter is miscopied, the result is occasionally disastrous, but usually is not so important. If a letter is missing or an extra one added, however, this can spell big trouble. DNA is read in three-letter words, and if there is one two-letter word thrown in, the reader just grabs a letter from the next word, which then starts in the wrong place, and so on all the way down the line. This is called a “frame shift error”. It’s the difference between ABC ABC ABC ABC… and ABA BCA BCA BCA… Since ABC has a completely different meaning from BCA, the result is not a small difference in one word, but a monstrosity all along its length.
The best and most reliable mode of repair is called homologous recombination, and is closely related to the process that takes place when sperm and egg combine, and some of their genes are permitted to cross over. There is a molecule called MRN that clips frayed ends to create a clean break. There is a molecule called RPA that binds to one strand on each side of the break, and a molecule called Rad51 that makes a local copy of the nucleotide sequence near each broken end, and then goes wandering through the nucleus, looking for a match. When all goes well, the match is found, because there are two almost-identical copies of each chromosome in the nucleus (one inherited from the mother and one from the father). So the unbroken, homologous chromosome is found and unzipped at the point matching the Rad51 template.
From this point there are two different sub-pathways. I don’t think anyone knows how one is chosen over the other. But the net effect of each is to “invade” the good copy of the chromosome, cut out one strand only, then use base pairing to repair the single-strand gap. The excised piece is carried back to the broken spot (how to find it?) and inserted at one end. Meanwhile, another Rad51 molecule has done the same thing for the other broken end.
Now the tricky part: The two broken ends each have one strand extended, and these single strands come together, joining up via complementary pairing. Another smart molecule called DNA ligase performs ligation=snipping off the excess, and re-joining in just the right place.
This story leaves me breathless, contemplating two mysteries: How did random variation and natural selection manage to discover such smart protein molecules? And how did resourceful teams of dedicated humans (who can’t, after all, sit inside the nucleus and watch this happen) ever figure all this stuff out?
There is no Sofer, no little man inside the cell nucleus directing the process and applying good judgment when things don’t go as expected. There are only a handful of molecules that perform the repair. Smart molecules. Really smart molecules.
The B Team
I’m hoping you find such stories as jaw-dropping as I do, but you may be wondering: What’s it doing in my aging blog?
I was hoping you’d ask.
Last month, the team of Vera Gorbunova and Andrei Seluanov (University of Rochester) published in PLoS Genetics (summarized and explained here in ScienceBlog) an article about DNA repair in old and young mice. One very efficient mode, slightly different from that described above, is NHEJ, non-homologous end-joining. It is the preferred method in young mice. But when mice get older, they switch to MMEJ=microhomology-mediated end joining, and make more mistakes. This contributes to misformed proteins and inefficient cell metabolism at best, genome instability and cancer at worst.
“DNA repair by NHEJ declines with age in mice, which could provide a mechanism for age-related genomic instability and increased cancer incidence with age.”
If DNA breaks are accidents that happen with fixed frequency, there’s no reason to assume they would be more common in older than younger mice. Gorbunova and Seluanov modified the genomes their mice so that every time a DNA break was being repaired, the cell would glow green. There was a lot more green in the older mice. They showed that this is because the MMEJ process, used in older mice, had more failures.
Why does the body switch from a better to a worse mode of DNA repair? Is there something that goes wrong with the highly-efficient NHEJ process over time, so that in old age (when the mice need it most) it’s not available for them? Or is it programmed death, one more example of a way in which the body gets old and dies on a schedule fixed by evolution?