There's a special distinction between computer hardware/software and biology: whilst computer and other man-made technology is subject to the force of the 2nd law of thermodymamics (law of increased entropy) and results in inevitable failure, it appears that biological systems have developed a way to circumvent this. Consider that the cell lines that all lifeforms are composed of are in fact derived from the very beginnings of life on this planet and rather than succumbing to entropy, biological systems appear able to increase in complexity (RNA viruses > prokaryotes > multicellular organisms). Therefore, its not very useful to be reliant on metaphors of human technology when seeking to understand aging.
Actually, the inappropriate analogy is the one you're making between Life and individual biological systems (organisms). Mark Hamalainen states it much more eloquently than I could:
Thermodynamics and Information in Aging
In order for life to stabilize information in a viable form it needs a method of entropy export. This is provided by replication coupled with natural selection. Random mutations to a germ-line's genetic information are constantly occurring and, of those which affect function, most are deleterious. However, if the information is continually copied at sufficiently high levels, it is always possible to maintain a viable subset. This process works most efficiently when single copies of genetic information are passed between generations. Each extra copy would increase the probability of the new organism carrying a deleterious mutation, decreasing the fractional size of the viable subset. Also, single copies can be scrutinized completely by natural selection, whereas in the case
of multiple copies, some mutations would be silent unless every copy carried them and could accumulate rapidly.
The somatic body is a continually replicating organism that passes every copy of its genetic information from one instant to another. From this perspective, it is obvious why natural selection or some analogous internal selection would be difficult. Each copy of genetic information within us continually undergoes changes independent of other copies. Our bodies can try to destroy or silence deleterious mutations by internal selection, but there are two obstacles to this method: selection criteria and positive selection. In natural selection, only a single copy of information is passed from an organism to the new generation and the selection criteria is life or death. Whether the organism is viable and able to reproduce, or not, determines the survival of that copy. In this way, natural selection is able to scrutinize the entire content of a copy of information that seeded an organism. Internal selection cannot use these criteria. Whether a cell lives or dies while in the somatic body does not depend on the functionality of its entire genetic content. Therefore, internal selection must try to approximate natural selection's abilities by other means. There are only two ways to accomplish this: either a copy can perform selection on itself (intracellular selection), or it can be scrutinized through comparison to another copy (intercellular selection). Intracellular selection is
performed by built-in mechanisms that detect deleterious changes to a cell's biochemistry and trigger death (apoptosis) or growth arrest (cellular senescence). This system eventually fails; since these mechanisms are encoded by the DNA they are also vulnerable to incapacitated by mutations. Intercellular selection is performed by surface recognition (adaptive immunity for example), and upon detection of an abnormal biochemical state, results in either attacking (by phagocytosis for example) or signaling to the intracellular machinery of the abnormal cell to destroy itself. These systems are also subject to incapacitation by mutation, and cannot work indefinitely. In addition, intra/intercellular selection is limited by the ingenuity of the selection criteria. It is not feasible to select for every possible deleterious mutation (indeed this problem emerges in the context of stem cell culturing), and this highlights the second obstacle to internal selection: it has the potential
to become deleterious. Mutant copies of information that increase their replication rate and difficulty of being selected against will be positively selected for. This problem is most obviously
manifested in cancer, where treatments are designed to try to kill the cancer based on differences (such as growth rate) detected between the cancer and healthy tissue. Such treatments are ultimately flawed for the same reasons as intra/intercellular selection. Hence, natural selection acts only on generations. It does not stabilize informational deterioration within the somatic body, and no internal selection process can fully supplement it. From an evolutionary perspective, maintenance of the germ-line is most important. The somatic body is just an elaborate shell designed to protect it and be periodically discarded.
Actually this does not need to be necessarily the case. It is conceivable that if there exists a sufficient level of inter-connectivity between these intracellular selection mechanisms, such that the loss of function of a few, regardless of which, results in activation of the remaining ones, then the possibility of failure of said system, will be for all practical purposes negligible, if engineering comes into the equation. As there would need to be X number of simultaneous failures-mutations above a given design threshold for the system to fail.
To a certain degree nature has developed something partially like this, we human's should
SOMEDAY be able to engineer tissues with practically indefinite functional life. You could put additional genes here and there throughout the various chromosomes, and interconnect their products such that they interact with each other in such a way, that failure only occurs if X(the number engineers eventually found acceptable) fail at the same time. As for safeguarding the functional capacity of the population of cells of a specific tissue, you similarly tie them or their products to the aformentioned safety net-web of selection. Similar to a raid array. That is individual cells behave like a house of cards, engineered to be exponentially more susceptible to total failure than to deviation from function. Fast to die if compromised in function, but being easily replaced by the remaining cells.
I believe thanks to the presence of many copies, and the capacity to easily make more, that is thanks to the self-replication ability of cells, that it should be possible to design immortal biological systems. Self-replication + intelligent modification = systems that have additional error-correction, damage protection, redundancy coupled with a vast web of interacting mechanisms that ensure any deviation from designed function results in self-destruction. I believe the same goes for any alternate self-replicating molecular machinery developed, stuff made of it if properly designed should be able to function pretty much indefinitely.
Edited by Cameron, 17 April 2009 - 11:20 AM.
LongeCity
