By Mike Gene (8/7/03)
The Factory Known as the Cell
I have long predicted that the bacterial cytoplasm will turn out to be a highly organized (high-information) state. The hypothesis works as follows: Bacteria probably best reflect the original life forms that first appeared on this planet. Being the products of design at the hands of a highly advanced human-like intelligence , we expect them to more closely share attributes of advanced forms of human design than those structures produced by non-teleological forces. This prediction follows from the Paleyian design inference, which attributes machine-like complexity to design. A non-teleological viewpoint simply cannot lead to this type of expectation. In fact, it hasn't. For decades, non-teleologists (scientists) have expected that the cell's cytoplasm would be dominated by random forces and mass action. This expectation, however, fits nicely with the conception of the first life forms aggregating/self-assembling in a random, prebiotic soup. A couple of recent review papers add support to my hypothesis.
1. Bridges, B. 2000. DNA polymerases and SOS mutagenesis: can one reconcile the biochemical and genetic data? BioEssays 22:933-937.
In this article, Bridges explores the relationship between Pol III, the main DNA replication machine in bacteria, and Pol V, a newly discovered DNA polymerase that repairs DNA lesions. Bridges considers how the biochemical and genetic evidence appear to contradict each other and proposes a speculative model to resolve these contradictions. This model moves away from the notion that the bacterial cytoplasm, and DNA replication, is dominated by random interactions and mass action and instead proposes a "replisome held together by protein-protein interactions and located as a structural entity within the cell." Evidence already exists to indicate that the DNA polymerase is not mobile during DNA replication. That is, it appears DNA is replicated much like proteins are synthesized and music is heard on a tape player: by running a linear thread over something that reads the thread. In other words, the replisome is a machine that is anchored and DNA is replicated by threading it through the machine. Bridges' speculation is intriguing. He proposes that the "head" of the replisome is composed of a cassette of polymerases which can shift so that the appropriate polymerase is used. In other words, the Pol III head normally replicates the DNA, but upon being confronted with lesions, the Pol V unit shifts into position and engages the DNA. If this model emerges as a valid representation of the replication process, the teleological implications are significant. For not only would we have yet another very powerful example of nanotechnology to further dissect, but the door is opened to exploring one way in which evolutionary mechanisms are designed.
2. Kisters-Woike, et al. 2000. On the conservation of protein sequences in evolution. TIBS 25: 419-421.
As the evidence that likens a cell to a factory/city accumulates, one could argue that this highly organized state evolved, through the classical blind watchmaker mechanism, from an initially random state. After all, the examples of muscles cells (more recently evolved states) might represent just this. However, at this point, we could make another prediction - the highly organized states we see coincided with the origin of life on this planet. Thus, if we turned to bacteria, the simplest life forms best reflecting the ancestral state, we would predict to find just this.
Kisters-Woike (K-W) et al. make a rather interesting case that much of the amino acid sequence conservation that is seen in very distantly related proteins is probably a function of those proteins being components in a multi-protein complex. That is, a protein as a monomer should be able to tolerate significant changes to its surface residues unless those residues are part of an interface that is involved in complementary interactions with another protein.
K-W et al. look at the sequence conservation of a bacterial and human glycolytic enzyme (glycolysis is an ancient and central metabolic pathway for breaking down glucose) in three-dimensions. There is a 65% identity over 332 amino acids and when the position of these conserved residues is mapped to the three-dimensional crystal structures of the proteins, it becomes quite clear that there are many extensive surface patches shared by both. K-W et al. interpret this conservation as regions important for protein-protein interactions:
An interface where several enzymes are involved would tolerate hardly any amino acid substitutions. When several different proteins interact, any mutant in one protein will influence the binding of several other enzymes in such a manner that this cannot be counteracted by a single suppressor mutation.
The teleological implications are two-fold:
a. K-W thus interpret this data to mean that this glycolytic enzyme is part of a large complex. There is already evidence that this is true in eukaryotic cells, including single-celled protozoa. But this sequence data also indicates such a state exists in bacteria . In fact,
In contrast to multi-complex enzymes, proteins like monomeric periplasmic transporter proteins or repressors are free to change drastically in evolution as long as their 3D structure is maintained in general. The enzymes of the glycolytic pathway remain the way they have been all along - involved in a large complex. The surfaces of the enzymes that take part in such a large complex have to be conserved. (emphasis added)
Thus, we do have evidence that the high-information states that minimize diffusion existing as part of the initial conditions associated with the origin of life. Or, at the very least, such states did not recently evolve along side vertebrate cells.
b. When K-W argue that that an interface where several enzymes are involved would tolerate hardly any amino acid substitutions because when several different proteins interact, any mutant in one protein will influence the binding of several other enzymes in such a manner that this cannot be counteracted by a single suppressor mutation , one can hear the echoes of Behe. In fact, we might even be able to develop a thesis whereby large-scale IC systems possess a threshold level of complexity that is indeed a true barrier to all four possible routes of Darwinian evolution. This would support the design inference based on machine-like complexity (the first thread of the teleological approach). More yet, such a designed state would constitute front-loading, as such a state would ensure your design would propagate throughout evolution and resist the noise of mutations. In other words, a designer who designs the glycolytic pathway as a large protein complex fulfills two design objectives - efficiency and propagation of the design. Both attributes thus allow the designer to indirectly design vertebrate glycolysis, for example, by designing bacterial or protozoan glycolysis.
Front-Loading Evolution
We've already seen an example whereby an initial design event expresses itself throughout its subsequent history, but there is another more tantalizing example on the horizon. In this case, we may have an example whereby something is designed only so that it's design potential is not realized until much later in evolution. The paper I have in mind is:
Ausio, J. 2000. Are linker histones (histone H1) dispensible for survival? BioEssays 22: 873-877.
In this paper, Ausio covers a lot of evidence whereby histone H1, which functions to link nucleosomes and thus more efficiently package DNA in eukaryotes, is not essential for survival and reproduction in filamentous fungi. If we eliminate H1 function in Ascolobus and Aspergillus, the cells are perfectly viable with no deleterious consequence on the sexual reproduction cycle. The same results were previously seen in the protozoan Tetrahymena. However, in the fungi mentioned, elimination of H1 does result in the cessation of growth within a week or two. In other words, elimination of H1 does not affect viability or reproduction, but only the life-span of the individual organism (however, with Aspergillus, elimination of H1 does not even effect the life span of the organism and has no apparent effect).
Three more points. First, thus far H1 is ubiquitous in eukaryotes. Secondly, H1 may not be crucial in single-celled organisms; in addition to the Tetrahymena data, Ausio observes, "These results suggest that while linker histones may be dispensable for the relatively short life span of an individual cell, they are most likely indispensable for survival of higher eukaryote organisms." Thirdly, Ausio argues that this is probably not true for multi-cellular organisms, where compaction of the genome is an important ingredient in the regulatory schemes used in generating and maintaining a multicellular body plan. Why is all this significant?
If H1 was indeed designed, given its minimal role in protozoa, it might constitute a very good example of front-loading evolution such that the initial eukaryotic state was prepared to evolve a multicellular state. In other words, the existence of H1 in protozoa may best be explained by the existence of H1 in metazoans. And that is one hypothesis that simply cannot be entertained, for the briefest of all moments, from a non-teleological perspective.
An important caveat is in order, however. Tetrahymena are fairly specialized protozoa and may not be representative of most protists. However, given that H1 is not essential in simple metazoans, such as filamentous fungi and also in specialized protozoa, we have good reason to suspect it might likewise be nonessential for less specialized protozoa. Here is yet another example where a teleological approach can generate experimental research. Instead of assuming Tetrahymena is unusual, and thus irrelevant, with regard to its lack of need for H1, we need to go into the lab and knock out H1 genes from other protozoans. And in keeping with the general argument of my speculation, we have yet another example of using a teleological approach to generate a prediction - if there is something to my hypothesis, then we will find other protists where H1 is not essential. In fact, we might even find some protists without H1.
Finally, keep in mind that "nonessential" does not mean H1 will have no role. Useless H1 is not a way to front-load (as useless things decay into nonexistence). Front-loading may entail giving a higher eukaryotic protein some role in protozoa to ensure it persists until something like higher eukaryotes evolve. But it is not until it is coopted into its primary designed role that it becomes essential.
Another COG Problem
A COG is a "chicken-or-egg" problem in biology, namely, which came first? COGs are indeed real problems, as evidence by the widespread acceptance of the hypothetical RNA world in science. One of the main reasons for this widespread acceptance is that the RNA world allows the non-teleologist to escape the COG posed by DNA replication - proteins are needed to replicate DNA, but the proteins needed are encoded by DNA.
Well, biology is full of COGs which have not been resolved. One such COG entails the chaperonin GroEL/ES.
Houry, WA et al. 1999. Identification of in vivo substrates of the
chaperonin GroEL. Nature 402: 147-154.
GroEL is an elegant protein-folding machine that is universally present in bacteria. In the research presented by Houry et al., many of the proteins which are folded by GroEL were identified. Included in the list are three ribosomal proteins, the subunits of the bacterial RNA polymerase, and the translation elongation factor Tu (all are universally present in bacteria). So why does this amount to a COG? GroEL exists because its genes are transcribed by RNA polymerase and translated by the ribosome. But for transcription and translation to take place, the machinery components must be folded by GroEL. So which came first?
Of course, one could argue that the transciption and translation machinery were originally folded by some unknown ancestral chaperone that has be replaced by GroEL. But that would be a fairly ad hoc exaplanation. If, on the other hand, the original cells that appeared on this planet were designed as functioning wholes, there is no COG - both appeared simultaneously.