posted 11. July 2002 12:11                   

quote:

---

What I would like to focus on here, instead, is the question of archetypes as an explanation for analogies, as opposed to "selection pressure" as the main mode of explanation of these striking phenomena. You only allude to archetypes' being in the mind of God, but what if they are somehow immanent in the dynamics of the cell itself? More specifically, I would be very interested to know what you think of recent (and not so recent) "structuralist" trends in biological thought?

---

quote:

---

I am thinking, for example, of D'Arcy Thompson's "On Growth and Form" (orig. publ. 1917, numerous editions), of Gerry Webster and Brian Goodwin's "Form and Transformation" (Cambridge UP, 1996), of the numerous Santa Fe Institue and A-Life studies on the abstract patterns underlying biological order, and the fewer but very interesting efforts to quantify an abstract "morphospace" for a particular type of organic form quantitatively (e.g., R.D.K. Thomas and W.E. Reif, "The Skeleton Space: A Finite Set of Organic Designs," Evolution, 1993, 47: 341--360).

I know that most Darwinian thinkers are fairly comfortable with this work, feeling that it will simply be incorporated into an expanded Darwinian "paradigm." I personally feel it is far more subversive than that. But, at any rate, I was just wondering what you thought about the naturalistic, structuralist alternative to understanding analogies as archetypes.

---

posted 11. July 2002 13:09                   

quote:

---

Once again, between the respective ring types, there is a very clear pattern. The alignments look good, and clearly show distinct types. In other words, there are two 1 billion year+ patterns that are nicely kept.

---

quote:

---

But, if you just assume that they *must* be related (since they are obviously too similar to not be related, right?), you can doubtless find a few areas of 'homology'. I found, for instance, a three amino acid stretch that happened to line up between the two when I forced them to align with each other.

You can go fishing for sequence similarity between almost anything if you really want to, but you need to look closer at the overall historical pattern before suggesting that they really fit a homology pattern. I do not believe that was done here.

---

posted 13. July 2002 02:21                   

quote:

---

Hello Nic!

I was so tempted to 'accidentally' slip and call you that last reply, just for fun. But I remember you sharing with us how you didn't like your identity getting out when you didn't want it to, so I didn't want to ruin it for you. Anyway...

---

quote:

---

I'm not trying to bring this up again to beat a dead horse, but I wanted to just quickly expound on one of those papers I cited to put this sequence homology myth to rest:

__________________________
The ring-type polymerase sliding clamp family. Bruck I, O'Donnell M.
Genome Biology 2001, 2:reviews3001.1-3001.3 (9 January 2001)

---

quote:

---

"The ring-type polymerases are found in all organisms, both prokaryote and eukaryote. The existing body of genome sequence information indicates that the B sliding clamp proteins are highly conserved in prokaryotes, and PCNA is highly conserved among eukaryotes. Interestingly, B and PCNA show no sequence homology, even though they have very similar three-dimensional structure [6]."

They do, in fact, mention below that idea of the 6-1 hypothesis. But.. I'll have more to say on that one later, as I do not think it to be a robust explanation.

---

quote:

---

"The structure of eukaryotic PCNA is practically superimposable on that of the B clamp [4,9]. The monomeric unit is only about two-thirds the size of B, however; it consists of two globular domains instead of three and trimerizes to form a six-domain ring the size of the B dimer. Although the PCNA domain structure is essentially the same as that of the domain structure in B, no sequence homology is detected between the two families. Perhaps the multidomain structure evolved from a common ancestral gene encoding one domain that later underwent duplications and fusion events to form the three-domain monomer."

---

quote:

---

[snip]
'Amazing similarity' indeed! But note the insistence on the part of these guys (as well as others), who actually did alignments and looked into this stuff - even after PSI BLAST came out, IMO - that there really is no significant sequence similarity.

So... we have two types that have been around for at least a billion years each, and yet are pretty darn conserved within their types, but completely different between themselves. I stress this over and over because it will frame the way that we approach any thoughts of sequence divergence and structural conservation.

---

quote:

---

Nucleic Acids Res 2000 Sep 15;28(18):3570-80

PSI-BLAST searches using hidden markov models of structural repeats: prediction of an unusual sliding DNA clamp and of beta-propellers in UV-damaged DNA-binding protein.

da link

[the clamp-relevant portion of the abstract]

We have designed hidden Markov models (HMMs) of structurally conserved repeats that, based on pairwise comparisons, are unconserved at the sequence level. To model secondary structure features these HMMs assign higher probabilities of transition to insert or delete states within sequence regions predicted to form loops. HMMs were optimized using a sampling procedure based on the degree of statistical uncertainty associated with parameter estimates. A PSI-BLAST search initialized using a checkpoint-recovered profile derived from simulated sequences emitted by such a HMM can reveal distant structural relationships with, in certain instances, substantially greater sensitivity than a normal PSI-BLAST search. This is illustrated using two examples involving DNA- and RNA-associated proteins with structurally conserved repeats. In the first example a putative sliding DNA clamp protein was detected in the thermophilic bacterium Thermotoga maritima. This protein appears to have arisen by way of a duplicated beta-clamp gene that then acquired features of a PCNA-like clamp, perhaps to perform a PCNA-related function in association with one or more of the many archaeal-like proteins present in this organism.

---

quote:

---

It should also be noted that researchers in this field really aren't sure what to make of these two types. They are not really sure whether to call them homologous or not, and as we see from O'Donnell's 6-1 suggestion, the idea that they were related is recieved as tentative at best.

---

quote:

---

Sliding DNA clamps are ring-shaped proteins that allow DNA polymerase to achieve high processivity during chromosome replication by tethering the polymerase catalytic subunit to DNA. From the crystal structures of these proteins it appears that they can encircle duplex DNA without steric hindrance, a property that presumably allows them to non-specifically attach to and move rapidly along DNA without dissociating (for reviews see 9-12). The structures of three distinct families of sliding clamps are available and include the Escherichia coli ß-clamp (13), the human and yeast proliferating cell nuclear antigen (PCNA) (14,15) and the bacteriophage RB69 and T4 sliding clamp proteins (16,17). All of these structures share a 12-fold symmetry around the ring consisting of a simple ß-[alpha]-ß-ß-ß structural repeat (13), though there is structural divergence in some of the repeats. Bacterial ß-clamps contain six ß-[alpha]-ß-ß-ß repeats per subunit with two subunits per ring while the eukaryotic and bacteriophage clamps contain four repeats per subunit with three subunits per ring. Pairs of these repeats form a domain, which has been termed the ‘processivity fold’ (18); thus the ring of the sliding clamp contains six domains and therefore is often described as having 6-fold symmetry (19). A structural representative of a fourth family of processivity fold proteins, namely the herpes simplex virus UL42 protein, is also available (18). UL42 does not form a ring-shaped clamp, however, but rather functions as a monomer and interacts with DNA quite differently than do sliding clamps; it has been suggested that UL42 resembles a primitive ancestor of sliding clamps (see 18 and references therein). Despite their structural similarity, proteins in each of these four families lack significant pairwise sequence similarity to proteins in the other families, suggesting that additional, unrelated processivity fold proteins remain to be found. Sensitive sequence analysis methods offer an opportunity to identify additional sliding clamps. For example, the fission yeast DNA repair proteins Rad1p, Rad9p and Hus1p were recently predicted to be distant relatives of PCNA using PSI-BLAST and similar procedures (20-22).

---

quote:

---

So, the question then becomes whether they somehow held onto their 3D structures, but underwent incredibly rapid evolutionary divergence.

---

I think this reveals several fundamental misconceptions you are operating under:

1) You think sequence divergence while maintaining the same structure is impossible, or very unlikely. On the contrary, it is a well-established pattern that structure decays very little even as sequence similarity dwindles to randomness. E.g., this kind of pattern is common: consider a pair of proteins, e.g. one from humans, and one chosen from increasingly distant relatives. Cytochrome C would be an archetypal example:

Sequence similarity | Structural similarity
100% | Superimposable
90% | Superimposable
80% | Superimposable
70% | Superimposable
60% | Superimposable
50% | Superimposable
40% | Superimposable
30% | Superimposable

I believe cytochrome C sequence differences bottom out at around 30%. The same pattern could be cited for numerous other proteins. See the pattern? Since you apparently accept common descent for the eukaryotes (your arguments based on the peculiarities of prok/euk relations force you to, as these pecularities do not really exist within eukaryotes)

Notably, this massive sequence redundancy for structure (zillions of potential amino acid sequences get you cytochrome C, Hubert Yockey estimated there were 10^93 of them IIRC) means that practically any sequence similarity (except perhaps small active sites of a few amino acids) is functionally unnecessary similarity (homology on my/Owen's defintion) that needs a function-independent explanation.

2) "(I)ncredibly rapid evolutionary divergence" is required to explain the twin facts of (a) high sequence conservation within e.g. the eukaryotes and eubacteria, respectively (How high, BTW?; you haven't given us numbers...50% similarity among all eukaryotes, or what?) and (b) low sequence conservation between the euks/eubacteria. You are exagerating to an extreme degree here (I am not exagerating ).

Recall how many billions of years separate eubacteria from archebacteria/euks (it is very unconstrained, but estimates range from 3 billion -- ~1 billion AFAIK). On evolutionary accounts, this should be the most extreme divide in known extant life. You are arguing that the degree of divergence is too large for this hypothesis, because within e.g. eukaryotes the clamp proteins have diverged only (say) 50% within the last 750-1500 million years (say, for a yeast-human or plant-human comparison).

But what do we know about rates of molecular divergence:

It will be slowest if all of these things are constant:

- function
- interaction with other proteins
- environment, particularly temperature

If these stay constant, then the rate of change will be slow (and approximately constant, the protein being subject only to stochastic substitutions). But if any of these things change, then all bets are off, primarily because natural selection can drive substitution to adapt to new function, new/addition interactions, or changed environments (and yes, we have numerous examples of rather rapid, although geologically moderate, sequence change driven by selection, as evidenced by e.g. rates of synonymous and non-synonymous substitution between closely related organisms, e.g. various Drosophila species).

In the case of DNA sliding clamps, the basic function has remained the same across all the groups, but eukaryotes attach a number of extra proteins to the clamps in comparison to prokaryotes, and archaebacteria are well-known extremophiles. The common ancestor of archaebacteria and eukaryotes was probably hyperthermophilic (adapted to ultrahigh temperatures that few-to-no eubacteria have ever reached). Adaptation to extreme temperature change is well-known to necessitate both changes in DNA/RNA G+C ratios (to maintain stability) and amino acid changes (to avoid denaturing the protein -- think taq polymerase, used in high-temperature PCR and discovered in an organism in a hot spring). Moving from high temperature back to low temperature would necessitate similar changes (although perhaps not as extreme).

These kinds of considerations make it perfectly plausible to propose accelerated substitution rates for (a) a eubacterial-->archaebacterial transition, and (b) an archaebacterial --> eukaryote transition. Even an orders-of-magnitude acceleration in substitution rate for a normally conserved protein would still be a substition rate much lower than that known for other proteins.

In the case of the FtsZ/tubulin relationship, we have a functional change in addition to the above considerations applying.

quote:

---

Also, I see that you remember my hypothesis that all three domains are independent designs. And you are right that these rings would support (or at least not contradict) a 2 domain ancestry. I can accept that. However, I base my 3 domain hypothesis on a much wider array of genes, and have found that some genes, for instance, cluster eubacteria and archs, some archs and euks (like the rings) and some even eubacteria and euks. If we had this discussion over another protein, you might be accusing me of claiming a 2 domain hypothesis with eubacteria and eukaryotes as one group!

---

quote:

---

In truth, I believe that during common design, some objects were directly shared (meaning down to the sequence level) between some of the domains (and in some cases, all of the domains), while in other cases, only the ideas were shared, while in others, an idea only pops up in one of them... but that is another discussion. Suffice to say that whenever I find any sort of break like this between any of the domains (whatever combination), I see it as supporting the three commonly designed cell types.

---

quote:

---

Anyway, I have to get back to work. I'll try to get back to the structural evolution later. But are you on record, then, as feeling that the most likely explanation for these two ring types is large divergence with structural conservation? What data leads you to think this?

---

quote:

---

Naturalists try to arrange the species, genera, and families in each class, on what is called the Natural System. But what is meant by this system? ... But many naturalists think that something more is meant by the Natural System; they believe that it reveals the plan of the Creator; but unless it be specified whether order in time or space, or what else is meant by the plan of the Creator, it seems to me that nothing is thus added to our knowledge.

[...]

Nothing can be more hopeless than to attempt to explain this similarity of pattern in members of the same class, by utility or by the doctrine of final causes. The hopelessness of the attempt has been expressly admitted by Owen in his most interesting work on the `Nature of Limbs.' On the ordinary view of the independent creation of each being, we can only say that so it is; that it has so pleased the Creator to construct each animal and plant.

[...]

How inexplicable are these facts on the ordinary view of creation! ... Why should similar bones have been created in the formation of the wing and leg of a bat, used as they are for such totally different purposes? Why should one crustacean, which has an extremely complex mouth formed of many parts, consequently always have fewer legs; or conversely, those with many legs have simpler mouths? Why should the sepals, petals, stamens, and pistils in any individual flower, though fitted for such widely different purposes, be all constructed on the same pattern ?

On the theory of natural selection, we can satisfactorily answer these questions. In the vertebrata, we see a series of internal vertebrae bearing certain processes and appendages; in the articulata, we see the body divided into a series of segments, bearing external appendages; and in flowering plants, we see a series of successive spiral whorls of leaves. An indefinite repetition of the same part or organ is the common characteristic (as Owen has observed) of all low or little-modified forms; therefore we may readily believe that the unknown progenitor of the vertebrata possessed many vertebrae; the unknown progenitor of the articulata, many segments; and the unknown progenitor of flowering plants, many spiral whorls of leaves. We have formerly seen that parts many times repeated are eminently liable to vary in number and structure; consequently it is quite probable that natural selection, during a long-continued course of modification, should have seized on a certain number of the primordially similar elements, many times repeated, and have adapted them to the most diverse purposes. And as the whole amount of modification will have been effected by slight successive steps, we need not wonder at discovering in such parts or organs, a certain degree of fundamental resemblance, retained by the strong principle of inheritance.

---

The similarity of Darwin's argument from the common "duplicate structure and modify" pattern in biology, and my argument from a similar pattern in proteins (go back to the post documenting same-fold, different function, and same function, different fold), is, I am sure, accidental.



PS:

Here is another model of a DNA sliding clamp, with a cross-section of the DNA double helix inside.



Consider how many other potential ways there probably are to form a ring of these approximate dimensions, with a loose affliation for DNA on the inside (recall that it is a *sliding* clamp). Is the 6-domain pattern functionally necessary? Couldn't 5 or 7 or 8 domains work if designing from scratch? Couldn't you have the ring specified by just one big domain, with a gap on one side to clamp open or shut (which is how the current sliding clamps work IIRC)?

Superimposable structure is not functionally necessary; you can attempt to explain it by designer archetype, but then are you really explaining much at all, asks Darwin. Common descent, OTOH...

I am going in circles now. Too many rings in my head I guess. I've said my bit for now...

nicsinia

[ 13 July 2002, 02:29: Message edited by: yersinia ]

posted 13. July 2002 15:40                   

quote:

---

2) Therefore superimposable structural similarity is similarity that is not explained by common function (and indeed, for a pair like eubacterial-cytokinesis FtsZ and ciliary tubulin, there is not common function available to do any explaining). It is "similarity in excess of that required for the similarity in function", i.e. Owen's homology.

---

quote:

---

If we entertain teleological explanations, there is nothing that compels us to move beyond a form/function similarity. But an ID theorist can take it further than this and ask why such similar structures would be employed differently? There should be good design reasons for employing essentially the same structure differently. First, what is the functional similarity? Apparently, to bind and hydrolyzes GTP such that assembly into protofilaments occur that tap into the design principle of dynamic instability (where GTP binding and hydrolysis functions as a switching mechanism). Let's call this the GTPase-dependent Protofilament Design (GPD) and we can think of it as a framework. But this framework/function, by itself, is biologically useless. In fact, unless it is tied to a biological function, it works merely as an wasteful energy sink (a futile GTPase cycle). Of course, for the biological function we need a biological context. This context then works to specify how one decorates the framework to tap into GPD's functional potential. In this case, we have the distinct bacterial and eukaryal contexts that differ in many significant ways. In a eukaryote, GPD is built upon to become tubulin, whereby it will play an important scaffold function in accord with the tensigrity model. What's more, it will serve as tracks for eukayal motor proteins, kinesin and dynein. It will also be called upon to form motility structures (flagella) and to play an essential role in segregating newly formed genetic material. In bacteria, GPD is built upon differently to serve the bacterial context. Here, it will play an important role in cytokinesis and thus interact with a different set of players. It is probably also associated with a bacterial cytoskeleton which may play more of a organizational role than a true skeletal role (the latter role is largely fulfilled by the cell wall). [1] The point is that a similar framework may have simply been exploited by a designer in different contexts (as a crude analogy, I look at my fan blowing air on me this hot night and am reminded that in another context, a very similar structure is used to move boats about). This ID hypothesis predicts several things...

---