Joined: May 2002
I got some replies to this quote I posted in response to Nelson Alonso:
Hmm, neither seems quite completely so true for the ATPase, because of the PPase. A simpler, partially sequence-similar system can perform the task. So even for a system older than the flagellum scientists are beginning to get hints indicating that ICness tain't all it's cracked up to be.
So lets review. A de-novo design hypothesis entails:
1. No evolutionary history
2. IC tied in with functional constraint (selection weeding out mutants because of ICness .
Well for starters, for the flagellum your reliance on Mike Gene has left you a bit out of date:
In the near future I want to bring Dembski into the mix. However, unless any relevant criticism of a specific system is brought up, I'm simply going to list them for now. And we can bring up another thread to discuss each system's history. For now, I'm just concerned with listing them.
The number of parts in a flagellum that don't have homologs with different, non-flagellar functions is getting to be rather low; mostly they are filament and shaft proteins, which all may be homologous with each other, and of course nonmotile filaments are known to have a wide degree of uses in bacteria...
Biochemistry 2001 Oct 30;40(43):13041-50
Conformational change in the stator of the bacterial flagellar motor.
Kojima S, Blair DF.
Department of Biology, University of Utah, Salt Lake City, Utah 84112, USA.
MotA and MotB are integral membrane proteins of Escherichia coli that form the stator of the proton-fueled flagellar rotary motor. The motor contains several MotA/MotB complexes, which function independently to conduct protons across the cytoplasmic membrane and couple proton flow to rotation. MotB contains a conserved aspartic acid residue, Asp32, that is critical for rotation. We have proposed that the protons energizing the motor interact with Asp32 of MotB to induce conformational changes in the stator that drive movement of the rotor. To test for conformational changes, we examined the protease susceptibility of MotA in membrane-bound complexes with either wild-type MotB or MotB mutated at residue 32. Small, uncharged replacements of Asp32 in MotB (D32N, D32A, D32G, D32S, or D32C) caused a significant change in the conformation of MotA, as evidenced by a change in the pattern of proteolytic fragments. The conformational change does not require any flagellar proteins besides MotA and MotB, as it was still seen in a strain that expresses no other flagellar genes. It affects a cytoplasmic domain of MotA that contains residues known to interact with the rotor, consistent with a role in the generation of torque. Influences of key residues of MotA on conformation were also examined. Pro173 of MotA, known to be important for rotation, is a significant determinant of conformation: Dominant Pro173 mutations, but not recessive ones, altered the proteolysis pattern of MotA and also prevented the conformational change induced by Asp32 replacements. Arg90 and Glu98, residues of MotA that engage in electrostatic interactions with the rotor, appear not to be strong determinants of conformation of the MotA/MotB complex in membranes. We note sequence similarity between MotA and ExbB, a cytoplasmic-membrane protein that energizes outer-membrane transport in Gram-negative bacteria. ExbB and associated proteins might also employ a mechanism involving proton-driven conformational change.
The occurrence of significant conformational change in the stator has implications not only for the present-day mechanism but also for the evolution of the flagellar motor. A membrane complex that undergoes proton-driven conformational changes could perform useful work in contexts other than (and simpler than) the flagellar motor, and ancestral forms of the MotA/MotB complex might have arisen independently of any part of the rotor. We queried the sequence database using the sequence of the best-conserved part of MotA (the segment containing membrane segments 3 and 4) from Aquifex aeolicus, a species whose lineage is deeply branched from other bacteria. In addition to the expected MotA homologues, the search returned a protein sequence from the archaeal species Methanobacterium thermoautotrophicum (protein MTH1022) that shows significant sequence similarity not only to MotA but also to the protein ExbB (Figure 9). ExbB is a cytoplasmic-membrane protein that functions in conjunction with ExbD, TonB, and outer-membrane receptors to drive active transport of certain essential nutrients across the outer membrane of Gram-negative bacteria. The energy for this transport comes from the proton gradient across the inner membrane. Thus, MotA and ExbB are both components of systems that tap the proton gradient to do work some distance away (at either the rotor-stator interface or the outer membrane; Figure 9).
Other features also point to a connection between the Mot and Exb systems. MotA functions in a complex with MotB, which as noted contains the critical residue Asp32 near the cytoplasmic end of its single membrane segment. ExbB functions in a complex with ExbD, which likewise has a single membrane segment with a critical Asp residue near its cytoplasmic end (Asp25 in ExbD of E. coli; ref 59). Although ExbB has only three membrane segments in contrast to the four in MotA, the membrane segments that show sequence similarity have the same topology. The protein TonB is also present in the complex with ExbB and ExbD (59, 60) and would provide an additional membrane segment to round out the topological correspondence (Figure 9). ExbB contains a well-conserved Pro residue (Pro141 in E. coli ExbB) that is the counterpart of Pro173 of MotA. Although MotB and ExbD do not share close sequence similarity apart from the critical Asp residue, in certain positions in the membrane segment the residues most common in MotB proteins are also common in ExbD proteins. Finally, like the MotA/MotB complex the ExbB/ExbD complex contains multiple copies of each protein (61). Together, these facts make a reasonable case for an evolutionary connection between the Mot proteins of the flagellar motor and the Exb proteins of outer-membrane transport (and by extension the TolQ/TolR proteins, which are related to ExbB/ExbD but whose functions are less understood).
So even for systems that are remote from us by 3 billion years there has been some recent progress.
...from this ARN thread:
...and then Nelson replied and I replied:
Hmm, Mike Gene didn't think that such a thing was irrelevant:
|Originally posted by Nelson Alonso:|
I also want to read your reference concerning bacterial flagellar motor, but to be honest, I already see it as irrelevant, since sequence similarity doesn't tell me much about how to make a bacterial flagellum via natural selection and random mutation.
And yet soon after he wrote this, exactly such homologous-but-non-flagellar ion channel was discovered. Looks to me like this family of enzymes couples proton flow to all kinds of mechanical processes, only one of which is rotating the base of the flagellum.
|MotA/MotB, on the other hand, could plausibly exist as some ion channel prior to the existence of the flagella, but there is no evidence of this.|
And he also wrote:
But a plausibly pre-existing function for stage 2 (now) *has* been provided. And, strangely enough (except from the perspective of the cooption hypothesis), the first reasonably strong homology evidence for a pre-existing functional subcomponent of the flagellum *just happens* to be the last two parts added in the cooption scenario, which just happen to already function together just like they do in the flagellum.
[snipped image for formatting sake]
Figure 1. The EFM Hypothesis. It begins with multi-component export machinery and invokes an initial cooption even to explain the origin of the filament. But because the "filament" of the flagellum is also a multi-component system, simultaneous, not gradual cooption is being invoked. Its non-flagellar function is not provided. The second cooption event, where an ion channel is added to create the flagellum, invokes the same thing.
(I don't see any good reason for MG's breaking what is otherwise convention and splitting off FlG's C-terminal end as a "separate" part, seems to me this is a special maneuvure not utilized elsewhere)
If you follow this analogy, the car when "sitting still" would be "jiggling" many times its body length every second due to brownian motion. Your driveway and house would be ruined. Intuition is a poor guide to life at the very small scale.
The bacterial motor allows movement at such a speed that, if the bacteria were resized to the weight of cars, they would supposedly break the sound barrier.
All depends on what you mean by "function". Kojima and Blair didn't think so. As for distance, the proposed cooption event did occur at something like the common ancestor of eubacterial, so there has been a bit o'time for divergence.
The Exb complex are homologs but quite distant from the Mot complex, and even the function is completely different.
....And when Nelson still wasn't impressed, I wrote:
Um, Nelson, you didn't even read the quote I provided:
See that? A homolog in archaebacteria. As far as our current understanding of microbial phylogeny allows us to say anything about what is basal to what, it appears that the MotA homolog is more basally distributed than the eubacterial flagellum. So if you're going to use distribution as a guide to what came first, you lose on this one.
We queried the sequence database using the sequence of the best-conserved part of MotA (the segment containing membrane segments 3 and 4) from Aquifex aeolicus, a species whose lineage is deeply branched from other bacteria. In addition to the expected MotA homologues, the search returned a protein sequence from the archaeal species Methanobacterium thermoautotrophicum (protein MTH1022) that shows significant sequence similarity not only to MotA but also to the protein ExbB (Figure 9). ExbB is a cytoplasmic-membrane protein that functions in conjunction with ExbD, TonB, and outer-membrane receptors to drive active transport of certain essential nutrients across the outer membrane of Gram-negative bacteria. The energy for this transport comes from the proton gradient across the inner membrane.
Of course, given the uncertainty surrounding phylogenetic events 3.5 billion years ago, not to mention lateral transfer events etc., the most reasonable thing to do would be to say that we have no particularly good information about what came before what. But then there goes any confidence in the assertion that bacteria have always had flagella or have no precursor homologous proteins, which has been a big chunk
of your argument.
Regarding the probability of this ion channel hooking up to the base of a primitive Type III pili, the exact mechanism of coupling proton flow to motion is still up in the air. However, you wouldn't have to get a fully functioning flagellum out of it, even undirected wiggling would enhance dispersal. This provides the starting point for natural selection to refine the procedure.
As you and Mike Gene point out, such a getting-the-process-started mutation is an unlikely "completely chance event" -- but just like any mutation, this is not a one-try event!!! Give a few gaztrillion bacteria a few million years! And we are clearly no longer dealing with the fortuitous de novo synthesis of a whole bunch of proteins, as Dembski suggests, or even the fortuitous cooption of dozens of individual proteins all at once, as Mike Gene sometimes mischaracterizes cooption. We are just hypothesizing a mutation crudely coupling two pre-existing systems.
As for similarity in function, Kojima and Blair note that this basic ion-channel system has been successfully coupled to diverse systems:
The point with the brownian motion analogy is that "faster than the speed of sound" is not quite so impressive on the molecular scale. Considering that some bacteria get by with relatively slow motility systems, and many get by with none at all, argues that we should keep the niftiness of the flagellum in perspective. Another random fact: the energetic efficiency of the flagellum is at best a few percent according to Berg's Random Walks in Biology.
|Thus, MotA and ExbB are both components of systems that tap the proton gradient to do work some distance away (at either the rotor-stator interface or the outer membrane; Figure 9).|
And then in reply to a few objections from Mike Gene:
An awful lot of entries in the NCBI protein database listing MotA, ExbB, and TolQ as being a "family", in opposition to what Mike Gene has been suggesting
People can read the quote I posted and decide for themselves exactly what the authors meant. Mentioning significant sequence similarity in the middle of a discussion of homologs seems like a strong indication to me that they meant homolog or at least "likely homolog". And see below for what some of those proteins are labeled as in the NCBI database.
|Originally posted by Mike Gene:|
Nic: Hey, I'm just following the peer-reviewed lit., man. Take it up in the pages of Biochemistry if you like...
The authors never claim MTH1022 is a homolog of motA. And even if they did, the claim would be unsupported.
Hmm, well a standard protein BLAST on E coli ExbB gives me this taxonomy report:
Actually, running a PSI-BLAST of ExbB gets me four hits on archaeabacteria, but whatever.
With e values of 0.005 and higher. That's not very impressive. Furthermore, even if there is a non-convergent and real relationship here, note that it is not widely distributed and would thus seem to have evolved long after archaea appeared.
A much stronger point is that the ExbB homologs are very widely distributed in eubacteria, in very basal lineages.
Think so? What groups did you have in mind?[/qb]
An E. coli "biopolymer transport exbB protein"
Here is the link
Here are some of the results *outside* the proteobacteria:
Note below that *even* some proteobacteria ExbB proteins have "unimpressive" similarity scores and yet are still called ExbB proteins (as are some **archaeabacteria** ExbB-family proteins):
|Code Sample |
| Chlamydophila pneumoniae CWL029 [chlamydias] taxid 115713|
gi|15618694|ref|NP_224980.1| Macromolecule transporter [Ch... 61 1e-08
Cytophaga hutchinsonii [CFB group bacteria] taxid 985
gi|23136461|gb|ZP_00118181.1| hypothetical protein [Cytoph... 60 1e-08
gi|23136843|gb|ZP_00118557.1| hypothetical protein [Cytoph... 36 0.35
Chlamydia trachomatis [chlamydias] taxid 813
gi|15605326|ref|NP_220112.1| polysaccharide transporter [C... 59 4e-08
Chlamydia muridarum (agent of mouse pneumonitis) [chlamydias] taxid 83560
gi|15835499|ref|NP_297258.1| MotA/TolQ/ExbB proton channel... 58 8e-08
Aquifex aeolicus [aquificales] taxid 63363
gi|15606982|ref|NP_214364.1| TolQ homolog [Aquifex aeolicu... 55 5e-07
gi|15606823|ref|NP_214203.1| biopolymer transport exbB [Aq... 42 0.005
Deinococcus radiodurans [eubacteria] taxid 1299
gi|15805483|ref|NP_294179.1| biopolymer transport protein,... 47 1e-04
Methanothermobacter thermautotrophicus str. Delta H [euryarchaeotes] taxid 187420
gi|15679040|ref|NP_276157.1| biopolymer transport protein ... 43 0.002
Synechocystis sp. PCC 6803 [cyanobacteria] taxid 1148
gi|16329196|ref|NP_439924.1| biopolymer transport ExbB pro... 41 0.013
gi|16329550|ref|NP_440278.1| biopolymer transport ExbB pro... 40 0.018
MotA and ExbB appear to have equally wide distributions
|Code Sample |
| Photobacterium damselae subsp. piscicida [g-proteobacteria] taxid 38294|
gi|19577341|emb|CAD27898.1| ExbB protein [Photobacterium d... 33 1.9
Plesiomonas shigelloides [enterobacteria] taxid 703
gi|13774064|gb|AAG23397.1| ExbB [Plesiomonas shigelloides] 33 2.0
Sulfolobus solfataricus [crenarchaeotes] taxid 2287
gi|15899059|ref|NP_343664.1| Amino acid transporter, putat... 33 2.6
Methanosarcina acetivorans C2A [euryarchaeotes] taxid 188937
gi|20089242|ref|NP_615317.1| MotA/TolQ/ExbB proton channel... 33 3.4
gi|20093212|ref|NP_619287.1| MotA/TolQ/ExbB proton channel... 33 3.8
gi|20093420|ref|NP_619495.1| MotA/TolQ/ExbB proton channel... 32 6.7
The distribution taxonomy report for ExbB is, in fact, rather like the taxonomy report for standard protein blast of E. coli's MotA protein: lots and lots of enterobacteria and proteobacteria, and a few hits out in spirochetes and other various deeply divergent bacteria.
Here is said taxonomy report
Are we to conclude that MotA is just as likely to be of late origin and derived as ExbB?
...methinks the database may be a wee bit biased towards certain intensively-studied gram-negative bacteria groups and that therefore seeing many hits in those groups and few outside means very little in terms of relative significance. As I showed, you have to put MotA and ExbB in the same distribution bucket regardless.
Regarding low e-values:
Based on your blanket skepticism of marginal e-values, you may want to argue that some MotA proteins are of independent origins and convergent on standard MotA's. Many of these scores are non too impressive, yet some are MotAs (or PomA, a related motor) despite this:
Regarding the e-values of archaeal ExbB homologs, PSI-BLAST on E. coli ExbB gives an archaeabacterial homolog with e-value of 2e-17, which ought to be good enough for Mike Gene:
|Code Sample |
| Treponema denticola [spirochetes] taxid 158|
gi|4426945|gb|AAD20619.1| flagellar motor protein MotA [Tr... 53 3e-06
Pseudomonas putida [g-proteobacteria] taxid 303
gi|2853602|gb|AAC08066.1| MotA [Pseudomonas putida] 53 5e-06
Leptospira interrogans serovar lai str. 56601 [spirochetes] taxid 189518
gi|24213362|ref|NP_710843.1| Chemotaxis motA protein [Lept... 52 6e-06
gi|24216276|ref|NP_713757.1| motility protein A [Leptospir... 39 0.056
Bacillus anthracis str. A2012 [eubacteria] taxid 191218
gi|21402555|ref|NP_658540.1| MotA_ExbB, MotA/TolQ/ExbB pro... 52 8e-06
gi|21399546|ref|NP_655531.1| MotA_ExbB, MotA/TolQ/ExbB pro... 51 1e-05
Shewanella oneidensis MR-1 [g-proteobacteria] taxid 211586
gi|24375769|ref|NP_719812.1| chemotaxis motA protein [Shew... 51 1e-05
gi|24373102|ref|NP_717145.1| chemotaxis motA protein [Shew... 38 0.13
Thermotoga maritima [thermotogales] taxid 2336
gi|15643440|ref|NP_228484.1| motility protein A [Thermotog... 50 3e-05
Campylobacter jejuni [e-proteobacteria] taxid 197
gi|15791705|ref|NP_281528.1| putative flagellar motor prot... 48 9e-05
Rhodopseudomonas palustris [a-proteobacteria] taxid 1076
gi|22963976|gb|ZP_00011582.1| hypothetical protein [Rhodop... 42 0.006
Magnetococcus sp. MC-1 [proteobacteria] taxid 156889
gi|22999422|gb|ZP_00043404.1| hypothetical protein [Magnet... 41 0.011
gi|23000123|gb|ZP_00044067.1| hypothetical protein [Magnet... 33 3.5
Vibrio alginolyticus [g-proteobacteria] taxid 663
gi|3024412|sp|O06873|POMA_VIBAL Chemotaxis pomA protein >g... 40 0.038
Bacillus halodurans [eubacteria] taxid 86665
gi|15615802|ref|NP_244106.1| flagellar motor apparatus [Ba... 39 0.058
Vibrio vulnificus CMCP6 [g-proteobacteria] taxid 216895
gi|27363787|ref|NP_759315.1| Flagellar motor component Mot... 37 0.18
Clostridium thermocellum ATCC 27405 [eubacteria] taxid 203119
gi|23021959|gb|ZP_00061601.1| hypothetical protein [Clostr... 35 0.87
Vibrio cholerae [g-proteobacteria] taxid 666
gi|15640908|ref|NP_230539.1| chemotaxis protein PomA [Vibr... 32 5.6
| Methanothermobacter thermautotrophicus str. Delta H [euryarchaeotes] taxid 187420|
gi|15679040|ref|NP_276157.1| biopolymer transport protein ... 90 2e-17
gi|15678698|ref|NP_275813.1| unknown [Methanothermobacter ... 43 0.003
gi|15678311|ref|NP_275426.1| protein kinase [Methanothermo... 33 3.4
These kinds of things ought to at least be mentioned and discussed before reckless statements are made about absolutely no evidence for precursors to flagellar proteins, that's my only point. For some reason you guys prefer to sweep it under the rug by unsupported arguments about ExbB's narrow distribution. Just acknowledge that this little bit of the biological world is a bit disharmonious with the flagellum-was-specially-created thesis. All I've been saying, and now documenting, is that ExbB homologs are at least as widely distributed as MotA, and possibly more widely distributed.
Edited by niiicholas on Feb. 17 2003,18:09