If you've seen BladeRunner, you know the short soliloquy at the end by one of the android replicants, Roy, as he's about to expire from a genetically programmed early death.

"I've seen things you people wouldn't believe. Attack ships on fire off the shoulder of Orion. I watched c-beams…glitter in the dark near Tanhauser Gate. All those…moments will be lost…in time, like tears…in rain. Time…to die."

There's an interesting idea here, that death can be an intrinsic property of our existence, a kind of internal mortality clock that is always ticking away, and eventually our time will run out and clunk, we'll drop dead. There is a germ of truth to it; there are genetic factors that may predispose one to greater longevity, and in the nematode worm C. elegans there are known mutants that can greatly extend the lifetime of the animal under laboratory conditions.

However, in humans only about 25% of the variation in life span can be ascribed to genetic factors to any degree, and even in lab animals where variables can be greatly reduced, only 10-40% of the life span variation has a genetic component. There is a huge amount of chance involved; after all, there aren't likely to be any genes that give you resistance to being run over by a bus. Life is like a long dice game, and while starting with a good endowment might let you keep playing for a longer time, eventually everyone craps out, and a run of bad luck can wipe out even the richest starting position rapidly.

In between these extremes of genetic predetermination and pure luck, though, a recent paper in Nature Genetics finds another possibility: factors in the organism that are not heritable, yet from an early age can be reasonably good predictors of mortality.

Here's the experiment. Start with an isogenic line of Caenorhabditis elegans; these are all nearly perfectly identical genetically, eliminating most of the possibilities of a genetic component. When raising a colony of isogenic animals, of course, they don't all abruptly kick the bucket on the same day at the end of their maximum lifespan (under two months for these worms), but instead a few die every day until they are all gone. The question is, is there anything that will allow one to predict whether a newly hatched worm will die in one week, or in 8 weeks?

Into this worm strain, a marker construct is introduced. The construct consists of a copy of some regulatory elements from the worm genome, attached to the sequence for Green Fluorescent Protein (GFP), a gene that glows bright green. Whenever the normal gene controlled by the regulatory elements is turned on, the GFP gene is also turned on. This is a fairly common technique in developmental biology, to take advantage of the regulatory circuitry in the cell to turn on a gene we can see. It's very clever—it means every time the gene we're interested in is active, the whole cell glows like a Christmas tree light, allowing us to watch genes flick on and off in living cells.

What gene? In this case, the GFP construct lets us see whenever a gene called HSP-16.2 is turned. HSP stands for Heat Shock Protein; it's one of a class of genes that are switched on when cells are stressed by harsh environmental conditions, such as when the temperature is artificially elevated. You can also induce them with cold or oxygen deprivation or disease. Many HSPs are chaperones, or proteins that assist in the folding of other proteins—they prevent proteins from being denatured. Others assist in transporting proteins from one compartment of the cell to another, or in flagging damaged proteins for destruction.

So Rea et al. have a line of worms where they can see an HSP gene being turned on. When these newborn nematodes are examined with a microscope, nothing is glowing; the investigators then crank up the temperature in the incubator from a comfortable 20°C to an unpleasant 35°C for an hour or two, and here's what they see in the scope:

On the left is the view with normal optics, and on the right the view with fluorescence optics, and you can see the worms are glowing greenly—as expected, they've turned on the heat shock proteins in response to a heat shock. Look carefully, though, and you'll see that there is variation. Some worms turn on HSP strongly (the one labeled "H"), others turn it on to a moderate level ("M"), and still others activate the gene to only low levels ("L", which are nearly invisible in the fluorescence image).

So far, so good, and not too surprising. The gene is turned on when expected, and there is variability, which is also quite common in biological systems. The degree of variability is a little bit surprising, since these are isogenic animals; the variability is not at all genetic, since they can be bred and there is no pattern of inheritance of the brightness.

So here is the very interesting correlation. The worms could be sorted out on the basis of their brightness after one hour of heat exposure, and then the longevity of the high, medium, and low HSP expressers was measured. The worms with a strong HSP response lived significantly longer than the ones with the low response.

(a) Data from a typical longevity analysis shown (mean s.e.m.; high = 24.4±1.1 d, median = 18.7±1.1 d, low = 15.35±1.0 d; N = 40; P < 0.025). (d) Survival trajectories of worms used to generate the data in a. "Presort" is the measure of survivorship in a population that was not sorted into high, median, and low HSP expression, and "Preheat" is a population that was not stressed with a heat shock.

Keep in mind that HSP expression is only a marker of some general and incompletely understood processes by which the organism copes with stress; artificially increasing the quantity of HSP in a nematode does improve its life span by a small amount, but not as much as was seen here. They're just seeing one note in the symphony, which is enough to tell it is being played, but not enough to know all the players.

The fascinating thing to me is that they are finding so much significant (I think a 50% increase in average life span is certainly significant!) variation in animals that is not a consequence of heredity, but is clearly an outcome of early, random processes in development and physiology. The authors offer some explanations:

Stochastic variation arises from fundamental thermodynamic and statistical mechanical considerations. A large fraction of individual variation in lifespan must stem from the fact that life results from an integrated series of metabolic reactions that themselves are under physical constraints of the specificity and rigidity with which they, too, can be regulated. At the molecular level, two points are germane to this study. First, when the number of molecules regulating a biological process becomes countably small, chance distributions come into play such that some regulatory molecules can vary severalfold between individual cells. Second, the Maxwell-Boltzmann (M-W) equation specifies the distribution of kinetic energies among molecules and requires kinetic energy to be a distributed function. This equation was used to develop a general theory explaining mortality kinetics. Several sources of variation at the molecular level could conceivably alter GFP (HSP-16.2) expression level and simultaneously affect more global processes. These include intracellular differences and fluctuations in the rates of molecular processes such as transcription, ribosome loading and translation (as previously postulated). Chance variation in the number of HSF effecter molecules present in each cell at the time of heat shock also could have marked phenotypic consequences. Variation in the frequency of mitochondrial genomic rearrangements, as previously observed in isogenic populations of C. elegans could have an effect. There is a growing body of research describing variation among isogenic individuals at the molecular level, typically in microbial or yeast cultures where such effects can be visualized. Substantial variation among genetically identical individuals is a fact of nature, and inherent molecular variability implies that biochemical and molecular genetic processes must have inherent variability.

By the way, since I began with Bladerunner, I've got to complain about one scene in the movie. The replicant Roy confronts his maker Tyrell, and wants to know how to stop his death clock. Tyrell makes excuses.

TYRELL: The facts of life. I'll be blunt. To make an alteration in the evolvement of an organic life system, at least by men, makers or not, it fatal. A coding sequence can't be revised once it's established.

ROY: Why?

TYRELL: Because by the second day of incubation any cells that have undergone reversion mutation give rise to revertant colonies—like rats leaving a sinking ship. The ship sinks.

ROY: What about E.M.S. recombination?

TYRELL: We've already tried it—ethyl methane sulfonate is an alkylating agent and a potent mutagen—it creates a virus so lethal the subject was destroyed before we left the table.

ROY: nods grimly.

ROY: Then a repressor protein that blocks the operating cells.

TYRELL: Wouldn't obstruct replication, but it does give rise to an error in replication, so that the newly formed DNA strand carries a mutation and you're got a virus again…but all this is academic—you are made as good as we could make you.

All of that is total gobbledygook. For instance, EMS is commonly used as a mutagen, but it doesn't create viruses; the contrived rule about coding sequences being incapable of revision doesn't make sense; the stuff about repressors causing replication errors is just babble. That part of the movie always makes me cringe, because it is so clear the writer just plucked random biology words out of some textbook and strung them together in ways that make no sense.

He should have just said his lifespan was in part an ontogenetic consequence, and it couldn't be corrected short of rewinding his entire life history back to fertilization, and replaying his development. That's one of the lessons of this paper, at least.

Rea SL, Wu D, Cypser JR, Vaupel JW, Johnson TE (2005) A stress-sensitive reporter predicts longevity in isogenic populations of Caenorhabditis elegans. Nature Genetics, published online 24 July 2005.