updated on Dec. 04, 2004

Question 041118b: What is a biological clock (sometimes called protein clock, or genetic tree)?

karthaus@photon.chitose.ac.jp

日本語

Answer 041118b: A protein clock is used to describe the relation between species (distant or close), and from this data, the time when the two species separated in evolutionary timeline is calculated. This approach looks nice, since the older the branching is, the more distant is the relation and the more different the protein structure should be. Frogs are more different from humans than cows. Thus the frog proteins should be more different from human ones than cow ones.
But actually, this is not the case.

The central dogma of evolution is common ancestry. Every mammal evolved from a common mammal in the distant past. Every vertebrae evolved from a common animal in the distant past, and so on.
Bacteria evolved into plants which evolved into animals, which evolved into fish, which evolved into amphibians, which evolved into reptiles, which evolved into mammals, which evolved into humans.
During this evolution, the proteins evolved, too.
Thus a protein of a close relative (evolutionary speaking) should be less different than the same protein of a more distant relative.

But this is not the case. Here is one example, cytochrome c.
Cytochrome c is an enzyme that plays a central role in the energy conversion of a cell. It is a protein that all animals, and many bacteria and plants, have. In animals it consists of 104 amino acids. Out of those 104 amino acids, at least 34 are the same for all species. The role of the cytochrome c is so basic and essential for the cell, that any mutation that leads to a change of one of those 34 amino acids is lethal. Furthermore, the heme-group in this enzyme is always between amino acid number 14 and 17.
The other 70 amino acids can be exchanged against other amino acids. Mutation occurs over time, and thus the number of different amino acids between two species should be a measure for the evolutionary distance. But this clock does not work as accurate as it seems.

For example, human and tuna(very distant relatives in evolutionary terms) have 20 differences, while close relatives like humans and horses already have 12 differences.

Here are a few more funny relationships:

"Rattlesnake cytochrome c is closer to human (14 differences) than to rhesus monkey (15), dog (21), penguin (30), or even a fellow reptile, the snapping turtle (22)! Human beings are closer to the pekin duck (11) than to a fellow mammal, the horse (12). Tomatoes and sunflowers appear closer to human than to other mammal, reptile, amphibian, fish, insect, or yeast.

Mammals are said to have evolved from reptiles; these came from amphibians, and they arose from fish. By cytochrome c comparison, however, the dog as one extreme example is closer to carp (a fish) than to rattlesnake (reptile) or bullfrog (amphibian).

The data show that in closely related organisms, the corresponding place in the chain may be occupied by amino acids which differ greatly in size, acidity, and electrical properties. Some of the differences, moreover, would have required multiple mutations in the same codon to evolve one from the other - in the face of the laws of probability." (Citation: Coppedge (1973))

The sunflower and human have 42 differences, while penicillus and humans have 47 differences. But sunflower and penicillus, both plants, have 55 differences. The bullfrog and human have 18 differences, while the bullfrog has 23 differences with a reptile rattlesnake (which is evolutionary closer). The bullfrog is closer to a snapping turtle (10 differences), even though a turtle is not a amphibian, but a reptile. (Reference: Loennig)

Fitch wrote (1984, Cladistic and other methods, problems, pitfalls and other potentials, subtitle "So many Trees and so Little Difference") that the investigated cytochrome c sequences of eight bacteria species allow already 10 995 different evolutionary trees. They all differ only slightly.
(Citation: The worst tree is only 41 replacements worse than the best one. If you know that you are distributing more than 10,000 trees into only 41 categories, it becomes a little more difficult to say that a tree close to the low end is all that much better.)

So, the relation between species is not clear at all.
Even worse, by comparing sequences of other proteins, different trees have to be constructed.

Other authors came to the following conclusions about molecular clocks:

Citation: (Andrews, 1985): "Two problems affect the accuracy of the protein clock. First, the fossil calibration date may be wrong because it is based an the earliest appearance in the fossil record of a species belonging to a particular lineage. It gives only a minimum date for the origin of the lineage, and the 'earliest appearance' is something that is always subject to change from new evidence. Second, the rates of change of the proteins being analysed may change along or between lineages; Goodman and others have shown that, particularly for the Hominoidea, the rate of change in a number of proteins is not constant. The Protein clock must therefore be said to be uncertainly calibrated and to keep irregular time."

Citation (Patton and Avise, 1986): "Since it now appears quite possible that homologous proteins can evolve at different rates in different phylads, molecular-based conclusions about absolute divergence time for species with a poor fossil record should remain appropriately reserved."

Citation: (Howgate, M.E. (1986): THE GERMAN 'OSTRICH' AND THE MOLECULAR CLOCK. Nature 324, 516):
One of the major problems with molecular clocks is that they need to be located in an absolute time frame determined by the fossil record. Undoubtedly the best such clock available at the moment is the DNA-DNA hybridization clock of Sibley and Ahlquist, but Peter Houde...argues that the DNA-DNA clock needs to be reset. ...If the picture envisaged by Houde is correct...then the datum on which DNA-DNA hybridization data is tied in to the fossil record is out by tens of millions of years.

Citation : (Houde, P. (1986): OSTRICH ANCESTORS FOUND IN THE NORTHERN HEMISPHERE SUGGEST NEW HYPOTHESIS OF RATITE ORIGINS. Nature 324, 563 - 565.):
The far reaching implications of this study lie with the temporal calibration of the DNA hybridization molecular clock. Sibley and Ahlquist postulated that the divergence of ostrich DNA from rhea DNA was initiated by, and therefore could be dated by, the spreading of the Atlantic seafloor. They have since applied this calibration to a variety of other avian and mammalian taxa. With the simplicity of the vicariance biogeography hypothesis of ratite origins now challenged, one cannot accept this part of the calibration of the DNA molecular clock without at least some degree of scepticism.

Citation: (Li, W.-H., M. Tanimura and P.M. Sharp (1987): AN EVALUATION OF THE MOLECULAR CLOCK HYPOTHESIS USING MAMMALIAN DNA SEQUENCES. J. Mol. Evol. 25, 330 - 342.):
The data summarized in this study clearly indicate that no global clocks apply to all mammals. In the past, rate standards have been obtained under the assumption of equal rates among mammals (Miyata et al. 1980; Li et al. 1985 a,b). It was estimated that the average synonymous rate for mammals is about 5 x 10-9 per site per year. The present study suggests that the average synonymous rate is about 6.5 x 10-9 in the rodent lineage and may be as high as 10 x 10-9 in mouse and rat, but is only 3 x 10-9 in the primate and artiodactyl lineages and may be as slow as 1 x 10-9 in humans and apes. ...nonsynonymous rate varies greatly among genes...

Citation: (Fitch, W.M. (1987): COMMENTARY ON THE LI AND WU, EASTEAL LETTERS. Mol. Biol. Evol. 4, 81.):
In the face of the evidence, it appears rather forced to postulate an ancient gene that is no longer with us in order to preserve the uniform-rate hypothesis. One might better conclude that the uniform-rate hypothesis is probably not reliable as a universal assumption.

But neodarwinists need the "Mutation and Selection" mechanism for evolution, and they came up with interesting ideas like this one:

Citation: (Gillespie, J.H. (1986): NATURAL SELECTION AND THE MOLECULAR CLOCK. Mol. Biol. Evol. 3, 138 - 155.):
A basic conclusion of the present paper is that molecular evolution is compatible with an episodic model. To arrive at this conclusion we began with a model with varying rates of evolution. However, other starting points could lead to models without an episodic structure that are nonetheless compatible with the observed values of R (t).

It seems that evolutionists change their model according to the result they want to obtain. More on that topic in Question 041118c: Is the "Theory of evolution" a theory in the scientific sense?

But even neodarwinists sometimes ask critical questions like this one by Gillespie:
"Why should the rates of mutation be altered only at the time of origin of the orders of mammals and then remain unaltered for the next 60 Myr even though other lineages are branching off to form families, genera, and species? A more realistic model would have the neutral mutation rates varying through time in a single lineage to the same extent that they vary between lineages."

Citation (S. Ferguson-Miller, D.L. Brautigan and E. Margoliash, CORRELATION OF THE KINETICS OF ELECTRON TRANSFER ACTIVITY OF VARIOUS EUKARYOTIC CYTOCHROMES C WITH BINDING TO MITOCHONDRIAL CYTOCHROME C OXIDASES (1976; J. Biol. Chemistry 251, 1104 - 1115) :
...the apparently constant rate at which amino acid substitutions occur in cytochrome c (13-16) were consistent with the neutral mutation hypothesis for protein evolutionary change (17-20). However, when an examination of the much more extensive data now available demonstrated that the rate of residue variation in cytochrome c was not constant, either in a single line of descent, at various evolutionary intervals or in different lines of descent during the same time, the neutral mutation hypothesis was no longer tenable for this protein (21-24).

Here is a very serious comment about using the biological clock for human evolution

Citation: (Lewin, R. (1987): AFRICA: CRADLE OF MODERN HUMANS. Science 237, 1292 - 1295.):
For many researchers...there is a head-on conflict between the molecules and the fossils. "If the molecular evidence is correct, then the fossils become inexplicable", states Milford Wolpoff of the University of Michigan. "But I believe the fossil evidence shows that the molecular biology is being wrongly interpreted."
Berkeley biochemists...conclusion...all modern humans derive from a population that lived about 200,000 years ago in Africa from which populations migrated to the rest of the Old World about 100,000 years later. Little or no interbreeding with existing Archaic sapiens populations occurred, suggest Wilson and his collegues. "They have calibrated the mutation rate incorrectly", claims Wolpoff. "With a much slower rate they would get a time of origin of 850,000 years ago, which I believe is correct." Smith also hesitates to accept the molecular biology evidence at face value. ..."

and yet another citation about human ancestery and settlement in Australia

Citation: (2003, Meeting: Modern Human Origins. Tracing the Road Down Under, Science, 302, 555):
She [Harding] argues that the steadily ticking "molecular clock" traditionally used by geneticists is unrealistic and produces ages that are too young. Instead Harding uses random mutation rates, which she claims more accurately reflect the natural world. Assuming two different rates, Harding concludes that the ancient haplogroups arose either 60,000 or 90,000 years ago."

Harding replaces one uncertain theory with another. In two sentences there are four words that indicate ambiguity (unrealistic, claims, assuming, either - or).
That shows how scientists themselves are uncertain about their models.

In conclusion, I want to cite the following scientist

Citation: (Patterson, 1987):
Since it was first proposed in the mid-1960s, the clock hypothesis has been one of the most controversial topics in evolutionary theory.


Why is it then that evolutionary closely related species (human and ape) have only one difference in cytochrome c structure, and far relatives (humans and sunflowers) have 42? Isn't this a proof for evolution, anyway?

No it is not. Humans and apes live in similar environment, have similar body structure, eat similar food. Thus the cytochrome c, which also interacts in a very complicated way with other enzymes and proteins, needs to have a similar structure. The human cytochrome c and the ape cytochrome c were designed with a similar structure.

The sunflower and humans live a fairly different lifestyle, so their needs for energy transformation in the cell is quite different. So sunflower cytochrome c and human cytochrome c need a different design.

A racing car has a different purpose than a taxi or a truck. So their design differs. But a Honda racing car and a BMW racing car run on the same course, so they need a similar design.

back to Main page