A living cell that is neither growing nor shrinking can be considered quiescent, a good example being a fibroblast deep in uninjured connective tissue in the body, a highly differentiated state. Many other cell types of the adult body are highly differentiated, but are by no means quiescent. Liver cells, for example, are extremely active, as are goblet cells of the intestinal mucosa. But these cells, despite their high rate of metabolism, also maintain constant size.
Proteins are continually being synthesized and degraded in all cells, the result being zero net change in non-growing cells, synthesis balancing degradation. Protein synthesis and its complex regulatory systems have been studied since the mid 50′s, and the process of “translation” from mRNAs is reasonably well understood. Protein degradation is a less mature field. Although the actual degradative process itself is broadly understood today, much less is known about the mechanisms that control the kinetics of protein degradation mediated by proteasomes and “free” proteases. Before the discovery of proteasomes, proteases responsible for degradation were thought to be loose in the cytosol (Schimke and Doyle, 1970). The two systems were thought to be involved, the lysosomal system and free proteolytic enzymes in the cytoplasm. By blocking the lysosomal activity with basic compounds that accumulate in their interna, it was clear that the lysosomes played little part in the bulk of protein turnover in the non-autophagic differentiated non-growing cell. Although proteasomal activity is now assumed to account for the bulk of degradation, a little further information would be welcome on the contribution that free proteases (and lysosomes, which might still be involved even where autophagy is not evident) might make to general and specific protein turnover.
While protein synthesis and degradation have received considerable attention as independent processes, few investigators have thought about the “dialogue” that exists to achieve protein balance, a question that has be one of basic cell biology. Protein stabilization has to equal destabilization. We remain generally ignorant of the regulatory mechanism(s) maintaining protein balance, which is only to be expected since we have been almost completely bereft to date of hypotheses that might explain it. By re-posing the problem more clearly here, new ideas might now be forthcoming. If an extensive and complex intracellular regulatory mechanism exists, one would have expected someone to stumble upon some evidence of it in the last 60 years. It is an issue that needs to be more assiduously investigated experimentally, not just theoretically; more data is needed on many basic facets to help formulate new hypotheses, and with new technology this should be relative easy (see below). We might also reflect on the goblet cell again. Although it synthesizes excessive amounts of mucus proteins, it nevertheless complies with this principle of balance, not only by intracellular synthesis and proteolysis, but by secreting the excess of mucus proteins that it synthesizes. Secretion must also contribute to protein loss in many cell types, but probably on a miniscule scale compared with goblet cells; nevertheless it has to be taken into the overall balance.
Data emerging from protein degradation studies prior to the discovery of proteasomes strongly indicated that there were two distinct classes of protein, short- and long-lived (Schimke and Doyle, 1970, Goldberg and Dice, 1974). Generally, short-lived proteins were seen as highly active enzymes that were used up and rapidly turned over. Long-lived proteins were the less active enzymes along with the many structural and housekeeping proteins of the cell. This notion has been upheld by nearly all investigations in the interim; there are indeed two distinct turnover rates, one fast and the other slow, both showing first-order kinetics indicative of the apparent randomness of degradation (an important observation with regard to any possible explanation), with the former usually being about an order of magnitude faster than the latter.
However, very different results were obtained from separate laboratories on the relative pool sizes of these fast-turnover and the slow-turnover groups of proteins, largely because different laboratories dod not used standardized techniques for studying turnover kinetics—usually that of following the fate of proteins pulse-labeled with a radioisotopic amino acid during a “cold” chase period in an excess of the unlabelled radioactive amino acid, to prevent any re-incorporation. Briefly, the longer the “pulse”, the fewer fast-turnover proteins were found at the end of the pulse (which was often 30 min or more). Clearly these were already being degraded as the labeling period proceeded, such that a 1 h labelling would have a bulk of slow-turnover proteins remained tagged, but literally a vanishingly small amount of fast-turnover would still be present. To surmount the problem, the ideal strategy is to “flash” label proteins and chase these “nascent” products immediately for many hours to days, which was achieved in the 1970-80s. The very short labeling period clearly showed that a very considerable proportion of newly labelled (nascent) proteins were present, regularly 20-40% of the total proteins that had been labelled (Wheatley et al., 1980). While this exposed the technical problem of clearly identifying two “classes” of fast and slow turnover proteins, few investigators of protein turnover seriously considered that such a large proportion of proteins could have such short half-lives, and most were largely dismissive of the findings, which then languished in the literature for at least 20 years. This high turnover rate of nascent proteins has since been independently established that such a large proportion of proteins could have such short half-lives, and most were largely dismissive of the findings, which then languished in the literature for at least 20 years (Yewdell, 2005; Yewdell and Nicchitta, 2006).
A second important observation emerging from the flash-labelling experiments was and still remains unfortunately ignored. Although gel electrophoresis was relatively underdeveloped at the time with regard to resolution of individual proteins (reliable 2-dimensional systems were still being developed), it was nevertheless clear that the banding seen in labelled proteins from cells taken just after pulse-chasing and those taken an hour or so later (i.e. when the short-lived proteins had largely been degraded) showed barely any significant change. What became immediately apparent was that the two “class” notion was not a major distinction between different species of the thousand or two resolvable proteins in a cell, but that, there was a fast turnover component and a slower turnover component for each and probably every species of protein (Wheatley et al., 1980). In brief, after flash labelling, usually 20-40% of probably all nascent proteins are degraded within an hour, leaving a slower turnover component that had half-lives of many hours and often days.
While these findings necessitated a complete rethink of how (and why) cells might so peremptorily destroy such an abundance of proteins, they ought to have heralded the beginning of an appraisal of how protein balance is maintained in the differentiated cell at that time. Although in 1980 the proteasome had yet to be discovered, knowledge of the actual mechanism(s) degrading nascent proteins does not change matters, the kinetics would be the same. A cell, in order to maintain itself, has to have a constant supply of new proteins. To ensure that this “satisfies its needs”, they would need to be made in excess. On this point, a difference between the views of Schubert et al. (2000) referring to “defective ribosomal products (DRiPS)” and my own is that probably the (greater) part of the rapidly turning over complement of proteins should not necessarily be considered defective, but surplus to needs and therefore disposable because they remain at risk if they cannot be integrated in the cytomatrix (become functional in their appropriate sites). It is acknowledged, however, that any nascent protein that cannot complete its proper folding (or perhaps find a suitable chaperone) remains at greater risk of degradation. The term introduced for these various states in nascent proteins was “metastable” [Figure 1; for a fuller treatment see Wheatley et al. (1982) and Wheatley (1984)].
While we may have some insight today of the way proteins are synthesized and then degraded, the following two questions seem to be central to any further discussion on the subject: (i) how do proteins that are synthesized get from the ribosomes to their appropriate sites in the cell matrix in which they can fully express their functions, i.e. become integrated as mentioned above (see also Medicherla and Goldberg, 2008; Qian et al., 2006), and is referred to in Figure 1 as “stabilization”; and (ii) what is the fate of proteins that lose their function - and presumably, but not necessarily - their integration for one reason or another (e.g. damage during improper reactions, the action of ROS, etc.) referred to in Figure 1 as “destabilization”? In brief, we need to know a lot more about the processes and consequences of protein stabilization and destabilization, which simply do not equate with those of synthesis and degradation per se. This question will not be answered until we have better techniques for following literally the ins-and-outs of proteins within the fabric of the living cell (which has to include their kinetics, selection, relative amounts, loss by secretion vs. degradation, etc.). We also need more basic data on which to formulate more and better hypotheses to guide the experimental work.
Modern techniques are both more sophisticated and many orders of magnitude more powerful than those used to formulate the above working hypothesis, and might more readily resolve some of the problems raised by the earlier work. However, at present there are two downsides to this optimism. No matter how powerful some new technique might be, it only becomes truly useful when it is applied to answering some crucial question that has been properly formulated. A critical appraisal of methods for measuring the stabilities of individual proteins by Yewdell et al. (2011; in this issue) points out their limitations, and draws particular attention to difficulties, not in their application to the study of protein stabilization and balance, but to how they are interpreted under these circumstances.
One way or another - and perhaps with techniques yet to be devised - we might reach a better understanding of this difficult but all too obvious problem of cell biology that has been largely bypassed or ignored for years.