How long does a protein last? That depends very much on the protein. Down there at the molecular level in our cells, there’s a lot of turnover – ribosomes are constantly pumping out new protein strands, while other proteins are constantly heading for the various shredders and recycling facilities (various types of proteasome, the lysosomes, and so on). When you pick out individual proteins and follow their formation and destruction, though, you find a very wide range. There are a lot that fall into the hours/days/week range, broadly speaking, but the outliers at both ends of that scale can be pretty interesting.
At the short end you have (for example) powerful transcription factors like c-Myc. In normal cells, Myc’s half-life appears to be around twenty or thirty minutes most of the time. As a cell heads into the mitotic cycle, though, Myc hangs around longer and builds up from its usual trace levels in preparation for the big events of cell division, but then it drops right back down to baseline again. Interestingly, a great many types of tumor cell seem both to have more Myc than usual and more long-lived Myc than usual (with the half-life extended up to leisurely stretches like a couple of hours), and if you have some ideas about fixing that there’s plenty of room for you to come down and try them out. Don’t be too shocked if they don’t work, though, because people have been taking swings at that target for decades now.
Another fast-moving class of proteins are involved in circadian rhythm and cellular timekeeping in general. It became clear many years ago that there were indeed cells that could apparently keep track of time, but it was not obvious how they managed to do that. The answer (subject of a 2017 Nobel) comes down to cycles of protein synthesis and degradation, spooling and unspooling with the various speeds of the transcription complexes, the ribosomes and the proteasomes as the ultimate timekeepers. It’s as if we have little medieval water clocks constantly running inside us.
But then there are the long-lived proteins. Some of those are to be found in long-lived cells, of course, since cellular lifetimes vary a great deal as well (athough even long-lasting cells are still probably turning their c-Myc over within the hour). We have the ever-present example of long-term memory, and while we’re still discovering what the exact cellular processes are in neurons as these are laid down, there are certainly some proteins involved that mark these changes for a very long time. There’s one tissue, though, that’s famous for having proteins that are basically as old as you are: the lens of the eye. It forms from the inside out, in a sort of clean-as-you-go process, with distinctive lens proteins being produced in large quantities by specialized cells that then die off, and whose debris is cleared by the next layer. There are no cells left in the nucleus of the lens, and no enzyme activity either. Now, there is water around – the lens itself is overall sort of a very concentrated protein solution or gel with some structural motifs in it – but those proteins themselves are just sitting there for decades.
With no enzymes around, you then have to wonder: do these proteins ever just fall apart on their own? Do they have half-lives measured not by the enzymes that are waiting to turn them over, but by their intrinsic chemical stability? We now know that they do, and in fact cataract formation in the lens with age seems to be driven by these processes. They fall into several classes. A big one is nonenzymatic cleavage: some amino acid sequences that are known to break more readily than others just on standing, with peptide bonds next to an aspartic acid or an asparagine residue as some of the leading examples. In those cases, the side chains of those amino acids come around and attack the carbonyl group of an adjacent peptide bond – a slow process, to be sure, but one that can’t happen at all next to something like an isoleucine, whose side chain is an inert lump of hydrocarbon. A new paper has demonstrated that the same thing can happen next to glutamate and glutamine residues; the same process but just with an extra atom in the cyclic intermediate that forms.
Then there’s protein crosslinking. One of the big mechanisms for that is when phosphorylated serines or threonines gradually do an elimination reaction to form reactive alkenes, which then react with glutathione or nearby Cys residues. Another slow problem is that the chiral centers where the amino acid side chains branch off can slowly racemize, turning the peptide chain into a contaminated mix of D and L amino acids and affecting the overall protein structure. Aspartic acid and asparagine residues are especially prone to this (cyclic intermediates again, this time in self-attack mode), but serine shows elevated levels as well. A fourth process is deamidation, because that same asparagine cyclic degradation can fall apart in that mode as well, with rates varying depending on what other amino acids happen to be nearby.
One does get the impression that perhaps the adoption of asparagine as a canonical amino acid might have been a mistake, at least from our current perspective. You’ll notice that a lot of the data on these problems is indeed from lens tissue, and that makes sense. It’s a well-defined tissue and known to be full of proteins that don’t turn over. But the same processes are surely going on with all the other proteins with long lifespans in other organs. I would guess that there’s often not much evolutionary pressure to do anything about this stuff, either. Most of these are gradual enough not to have much impact on reproductive fitness, which is the main thing that evolution cares about. If some your proteins turn yellow and cloudy decades after you’ve passed on your genes, natural selection is unlikely to notice.
But we don’t have to play that hand. That’s the story of civilization and of science, the human race’s gradual refusal to abide by the house rules. Perhaps we can find ways to do a better job protecting these long-lived proteins, as we better come to understand what goes wrong with them!
