Front-loading with ribosomes

While outlining the logic of front-loading in The Design Matrix, I noted how the existence of multifunctional (moonlighting) proteins would serve the needs of front-loading. In essence, a protein with multiple functions can be viewed as a protein that is packed with preadaptations ready to be more fully exploited when the proper conditions arise.

This developing paradigm has allowed me to come up with a prediction. If evolution was front-loaded, and a significant aspect of this front-loading existed as multifunctional proteins, whereby secondary or tertiary functions could be unleashed as evolution proceeded into the future, an excellent candidate for storage of some of these secondary functions would be the ribosome, the protein-synthesizing factory of the cell. This is because a designer could count on the ribosome being retained, largely unchanged, throughout billions of years of evolution because it plays such an absolutely essential role in life. If it remains largely unchanged, secondary functions can be carried into the future. Thus, I would predict that ribosomal proteins, which normally function as chaperones to fold the ribosomal RNA and hold it together to form the functioning ribosome, would also exhibit secondary functions (moonlight).

And a survey of the literature does indeed support this prediction.

To illustrate the extent of moonlighting among ribosomal proteins, let’s begin with an assembly map of the ribosomal small subunit.

Fig. 1. The revised assembly map of the 30S subunit. The 16S ribosomal RNA is shown at the top, oriented from 5´ to 3´ direction. Each of the arrows indicates an observed dependency of binding for each ribosomal protein. The primary binding proteins depend solely on interactions with 16S rRNA (top row); the secondary and tertiary binding proteins depend on prior binding of other proteins. (Source)

Next, let’s put some blue boxes around the ribosomal proteins that are found in all three domains (archaea, bacteria, and eukarya), indicating these proteins were present in LUCA:

ribconserved

Next, let’s put red circles around the ribosomal proteins for which there is experimental evidence to support a moonlighting role (the moonlighting roles are listed below the figure)

moonlightrib

Moonlight Roles

S2 – plays a role in oogenesis in Drosophila

S3 – DNA repair

S4 – contains ETS DNA-binding motif

S6 – possible tumor suppressor

S7 – the fold of the core is similar to that of a DNA architectural factor; binds own mRNA

S9 – a B23/NPM-binding protein required for normal cell proliferation

S10 – Transcription elongation

S12 – enhance splicing in phage

S13 – controls splicing of own pre-mRNA

S14 – inhibits transcription of own mRNA

S19 – attracts monocytes/blood cell production

S20 – participates in antitermination of transcription

So these ribosomal proteins are involving in DNA repair, transcription control, RNA processing, the cell cycle, and development.

Now let’s look at both figures side-by-side

ribmooncons

This assembly maps involves 15 universal small subunit ribosomal proteins, and of these 15, ten have a moonlighting role (we shall explore several of these functions at a future date). What’s striking is that all proteins involved in binding to the 3’ domain of the rRNA are not only universal proteins, but also have moonlighting functions.

Those universal ribosomal proteins that don’t seem to have a moonlighting role are s17, s5, s8, s11, and s15.

Thus, the front-loading hypothesis further predicts these too will eventually be discovered to have moonlighting functions.

Advertisements

30 responses to “Front-loading with ribosomes

  1. Pingback: Front Loading with Ribosomes - Telic Thoughts

  2. Thanks

  3. Mike,

    Does that track-back thingy happen automatically, or did someone have to do something to make it show up here? If someone has to do something, then it wasn’t me. I don’t know how to do that track-back thingy.

  4. Mike, I have a question that is relevant to a debate I’m carrying on at TT. How did you discover the moonlighting roles of the ribosomal proteins? Is there a list of moonlighting proteins? Or is there a list of proteins and their functions, and you made the correlation, based on you FLE hypothesis?

  5. Bilbo,

    When I was working on the DM, I came up with this idea (I would hope it is clear the prediction fits in nicely with the discussion I provide in chapter 7 of TDM). So I spent some time slogging through abstracts at PudMed by searching with ‘ribosome s1’ , ‘ribosome s2’, etc (this stuff was going to be part of volume 2). I hit paydirt with S3, as there were several papers on S3’s role in DNA repair. Eventually, I ran across a review that did list many of moonlighting roles. I recently found another review paper that added to the list published a year or so ago. And a few days ago, I found a paper that shows a moonlighting role for s5 – one of the ones not on the list above. I’ll write that one up eventually, but the stuff about REST has really drawn my attention for now.

    I should point out these moonlighting roles are usually discovered through serendipity, as no one was looking for them on the basis of any evolutionary prediction.

    As for the track back thing, anytime someone with a WordPress blog links to this blog, a track back automatically shows up.

  6. Interesting idea. However, there is no evidence to show that evolution has a direction or that it predicts what it will do with itself in the future. It is certainly true to say that many evolutionary adaptations on the molecular level are through “recycling” of what is already there. But it is entirely another thing to conclude such recycling is possible because those molecules are some how predestined to be recycled; that they are “designed” with extra functions built in. I’m sorry to say but this borders on pseudoscience and the likes of astrology or tarot cards.

  7. Mike, over at TT hrun suggested the following:

    And hey, if you were to actually look at the review, maybe you’d find that the scientists who wrote them also expect to find more multifunctional ribosomal proteins.

    So I told him I would ask you if they indeed made that prediction.

  8. Quyen Tran writes:

    Interesting idea. However, there is no evidence to show that evolution has a direction or that it predicts what it will do with itself in the future. It is certainly true to say that many evolutionary adaptations on the molecular level are through “recycling” of what is already there. But it is entirely another thing to conclude such recycling is possible because those molecules are some how predestined to be recycled; that they are “designed” with extra functions built in. I’m sorry to say but this borders on pseudoscience and the likes of astrology or tarot cards.

    Quyen, you need to show that non-directed evolution has no problem explaining how two-thirds of the ribosomal proteins moonlight. Otherwise I’m afraid that it looks like directed evolution is the more likely candidate, regardless of how much you dislike it.

  9. Bilbo,

    The two reviews I used make no mention of such an expectation:

    Emerging functions of ribosomal proteins in gene-specific transcription
    and translation by Mikael S. Lindström (Biochemical and Biophysical Research Communications 379 (2009) 167–170).

    Extraribosomal functions of ribosomal proteins by Ira G. Wool. (TIBS 21 (1996) 164-165).

    Also, I can see from my above reply, that I made it sound like I found one protein while slogging through PubMed and used the reviews to fill in the rest. In reality, I found half through PubMed searching and I used the tables from these two reviews to fill in the other half. Here are the sources for the list I provide above (and I don’t include all the PubMed finds):

    S2: Wool paper

    S3: PubMed search:

    DNA Repair (Amst). 2007 Oct 1;6(10):1453-62.
    Translocation of human ribosomal protein S3 to sites of DNA damage is dependant on ERK-mediated phosphorylation following genotoxic stress.

    S4: Pub Med search

    EMBO J. 1998 Aug 17;17(16):4545-58.
    The crystal structure of ribosomal protein S4 reveals a two-domain molecule with an extensive RNA-binding surface: one domain shows structural homology to the ETS DNA-binding motif.

    S6: Wool paper

    S7: PubMed search

    Structure. 1997 Sep 15;5(9):1199-208.
    Ribosomal protein S7: a new RNA-binding motif with structural similarities to a DNA architectural factor.

    S9: PubMed search

    J Biol Chem. 2008 Jun 6;283(23):15568-76.
    Ribosomal protein S9 is a novel B23/NPM-binding protein required for normal cell proliferation.

    S10: PubMed search

    Mol Cell. 2008 Dec 26;32(6):747-8.
    Comment on:
    Mol Cell. 2008 Dec 26;32(6):791-802.
    Transcription by moonlight: structural basis of an extraribosomal activity of ribosomal protein S10.

    S12: Wool paper

    S13: Lindstrom paper

    S14:Lindstrom paper

    S19: PubMed search

    Pathol Int. 2007 Jan;57(1):1-11.
    Roles of the ribosomal protein S19 dimer and the C5a receptor in pathophysiological functions of phagocytic leukocytes.

    S20: Wool paper

    BTW, I was under the impression that hrun viewed “expectations” as something akin to a whimsical guess. Has he loosened up from his posturing about predictions to grant significance to mere expectations?

    Anyway, there are plenty more ribosomal blog entries in the pipeline here.

  10. Hi Quyen,

    However, there is no evidence to show that evolution has a direction or that it predicts what it will do with itself in the future.

    How can you be so sure? What would you count as evidence that showed evolution has a direction? What would make you suspect this?

    But it is entirely another thing to conclude such recycling is possible because those molecules are some how predestined to be recycled; that they are “designed” with extra functions built in.

    It wouldn’t be predestination. It’s more akin to loading the dice. Preadaptation, not predestination.

  11. Playing the messenger boy, again. Hrun responds:

    Odd. When I looked at the Lindström review I found this:

    This brief literature overview demonstrates an expanding list of novel r-protein functions outside of the ribosome itself. While by many regarded as contaminants in biochemical assays there are also reasons to not too quickly dismiss a ribosomal protein. The challenge is to identify those interactions which are physiologically relevant. This might be a difficult task given the often critical roles r-proteins play in cell growth, but conditional knock-out mouse models and RNA interference strategies could be very helpful tools. An interesting avenue for further research is the possibility that many mRNAs could be controlled by RNA binding r-proteins which then bind to 5′ and 3′ UTRs. This could result in complex interactions between micro-RNAs and r-proteins affecting the regulation of mRNA processing, association with ribosomes, and translation. With regards to cancer research, the task for the future will be to define those alterations in r-protein expression which are functionally relevant for tumor development, as well as those changes that may provide predictive information for a therapeutic purpose.

    So are you suggesting that Lindström tells people to look for more extraribosomal functions without the expectation that people will find such functions?

  12. And also from hrun:

    And as for the talking point by Ira Wool, I wonder if you looked at the book chapters Ira was referencing:

    the evidence for the putative extraribosomal functions is considered in greater detail in two essays 4,5 (from which the table is abstracted), which also have extensive lists of references.

    Since the talking point is quite short, one would suspect that speculations about extraribosomal functions would rather be found in the 10 page paper or book chapter he is referencing.

  13. Hi Bilbo,

    No, I did not go to the library and check out the book Ira Wool references. The table supplied in the talking point was clearly sufficient for the purposes of the above blog entry.

    As for the Lindström paragraph, I originally answered the following claim from hrun:

    “if you were to actually look at the review, maybe you’d find that the scientists who wrote them also expect to find more multifunctional ribosomal proteins.”

    Note that hrun has changed it from a) scientists to scientist and b) multifunctional ribosomal proteins to extraribosomal functions.

    I read the Lindström paragraph as a warning not to automatically assume ribosomal protein content in biochemical assays were contaminants given that science has discovered they can and do have extraribosomal functions. Lindström then speculates they could be binding to non-coding portions of mRNA and adding a further layer of regulation. I did not read it as some expectation that s5 or s17, for example, would turn out to have a extraribosomal function.

    So when hrun writes, “So are you suggesting that Lindström tells people to look for more extraribosomal functions without the expectation that people will find such functions?“ this doesn’t mean we should expect to find that s5 moonlights. It could just as well be an expectation about a more specific function, or yet another function, for, say s4 or s6.

    Anyway, I’m not sure why this quibbling is significant. Let’s re-focus on the main points.

    1. I laid out the logic of FLE, and its relation to moonlighting, in The Design Matrix.

    2. I built on this logic and laid out the testable hypothesis as outlined in this blog entry. This is a positive, stand-alone, example that shows how the logic of front-loading can guide us and provide a deeper understanding of biotic reality.

    3. Since many insist that FLE must somehow play against conventional evolutionary thinking, in the comments, I noted, “I should point out these moonlighting roles are usually discovered through serendipity, as no one was looking for them on the basis of any evolutionary prediction.”

  14. BTW, since this topic seems to be generating some interest, I think I’ll suspend the exploration of FAST and return to it.

  15. Bilbo, it appears that I have gotten myself banned form telicthoughts, so I will be unable to continue any discussion there.

  16. “Anyway, I’m not sure why this quibbling is significant. Let’s re-focus on the main points.

    1. I laid out the logic of FLE, and its relation to moonlighting, in The Design Matrix.

    2. I built on this logic and laid out the testable hypothesis as outlined in this blog entry. This is a positive, stand-alone, example that shows how the logic of front-loading can guide us and provide a deeper understanding of biotic reality.

    3. Since many insist that FLE must somehow play against conventional evolutionary thinking, in the comments, I noted, “I should point out these moonlighting roles are usually discovered through serendipity, as no one was looking for them on the basis of any evolutionary prediction.””

    The quibbling is indeed not particularly important.

    I simply to exception to the suggestion that this was a ‘prediction’ as it is usually understood in science. Likewise, I took exception that Bilbo credited you for ‘discovering’ that many ribosomal proteins have moonlighting functions.

    That is all there is to it.

  17. Hi hrun,

    I simply to exception to the suggestion that this was a ‘prediction’ as it is usually understood in science.

    Fair enough. I would just point out that a) I have always been up front that my book and this blog are not “science and b) checking with the thesaurus shows that the word ‘predict’ is synonymous with ‘predict’ in more relaxed settings. But I do understand your concern, so I will try to be more careful with my wording next time around.

    Likewise, I took exception that Bilbo credited you for ‘discovering’ that many ribosomal proteins have moonlighting functions.

    No, I did not experimentally detect that many ribosomal proteins moonlight so I cannot be credited with such discoveries. I simply pointed out the logic of FLE and how it led to an expectation that turned out to be correct.

  18. Hi hrun,

    Since I never figured out how to communicate with the other members of TT, I can’t verify that you are banned from TT. In case you’re not, I’ll post my reply here and at The Design Matrix.

    [b-quote] hrun: So, how does this support the distinction that Mike apparently is trying to draw between him (FLE adherent) vs. the scientists who write the reviews on Pubmed (not FLE adherent)?[/b-quote]

    Mike went looking for moonlighting roles for ribosomal proteins, based on his hypothesis, which predicts it.
    It’s not clear whether other scientists are expecting moonlighting roles for ribosomal proteins, and if they are, why they are expecting them.

    Let me add that when I said that Mike discovered the moonlighting ribosomal proteins, I was (and still am) under the impression that before this thread there was no exhaustive list of moonlighting ribosomal proteins. That means that no one thought it significant to make such a list, besides Mike. A crude analogy would be the blind men each listing the properties of the elephant that they had discovered, then someone coming along and compiling a list of all the properties of elephants that had been discovered. The compiled list is a discovery in and of itself.

    Let me also add that even though Mike doesn’t claim to be doing science, I believe he is. Stephen Meyer’s story about Watson and Crick is very enlightening in this regard. I’ll have to start a new thread about it, probably Monday.

  19. Hi Bilbo,

    Let me add that when I said that Mike discovered the moonlighting ribosomal proteins, I was (and still am) under the impression that before this thread there was no exhaustive list of moonlighting ribosomal proteins. That means that no one thought it significant to make such a list, besides Mike.

    Actually, the Wool paper has a list and the Lindström paper has a short list that adds to it.

    But, AFAIK, my list is the most exhaustive for SSU ribomsomal proteins.

    What I have done that is different is that I correlated three things with the figures above and the list – assembly map, universal SSU proteins, and moonlighting roles. I don’t know of anyone who attempted to make those connections.

  20. Fair enough. I would just point out that a) I have always been up front that my book and this blog are not “science and b) checking with the thesaurus shows that the word ‘predict’ is synonymous with ‘predict’ in more relaxed settings. But I do understand your concern, so I will try to be more careful with my wording next time around.

    I know that you have been quite straightforward and careful about it. Oftentimes, I have the feeling that the distinctions you make, are not necessarily shared by the posters I conversed with on TT.

    No, I did not experimentally detect that many ribosomal proteins moonlight so I cannot be credited with such discoveries. I simply pointed out the logic of FLE and how it led to an expectation that turned out to be correct.

    Yet again, that was simply the distinction I wanted to clarify. While ‘discovery’ can denote finding something that is new to oneself, in a science context, the term discovery is generally reserved for novel findings.

  21. Since I never figured out how to communicate with the other members of TT, I can’t verify that you are banned from TT. In case you’re not, I’ll post my reply here and at The Design Matrix.

    I am quite certain I am banned. My old username is blocked, my attempt to register with a new username to get clarification has been banned as well.

    If you like, you could simply announce this in one of the open threads so that the folks who I was in mid-conversation with, will know.

    Mike went looking for moonlighting roles for ribosomal proteins, based on his hypothesis, which predicts it.

    Bilbo, yet again, you are using two different terms here ‘hypothesis’ and ‘predict’. One (hypothesis) I think is entirely appropriate. The other (prediction) is not. You have admitted over on TT that according to FLE there could be many ways for a front-loader to implement FLE without using moonlighting proteins for the ribosome. So, in a scientific sense, this can not be a prediction of FLE. It simply is a hypothesis by Mike that he tested by looking through the published work available on Pubmed.

    Let me add that when I said that Mike discovered the moonlighting ribosomal proteins, I was (and still am) under the impression that before this thread there was no exhaustive list of moonlighting ribosomal proteins. That means that no one thought it significant to make such a list, besides Mike. A crude analogy would be the blind men each listing the properties of the elephant that they had discovered, then someone coming along and compiling a list of all the properties of elephants that had been discovered. The compiled list is a discovery in and of itself.

    In fact there have been such lists. And as more multifunctional ribosomal proteins are being discovered, more review papers are being listed and thus more proteins are added to the list. As a case in point, look at the following review (PMID: 19362532, How Common Are Extraribosomal Functions of Ribosomal Proteins? — to find it, just put the PMID number into the search box at pubmed.org) there is a table that is more exhaustive than all preceding tables.

    So if you follow the scientific literature, there have been such tables compiled for over a decade and they will continue to be compiled. I therefor don’t think this qualifies as a ‘discovery’.

    Let me also add that even though Mike doesn’t claim to be doing science, I believe he is. Stephen Meyer’s story about Watson and Crick is very enlightening in this regard. I’ll have to start a new thread about it, probably Monday.

    And that is exactly the problem of why we are arguing about terms like ‘prediction’ and ‘discovery’. If you think what Mike does is science, then I would suggest you use the terms like they are commonly used in science. It is not challenging (even for a non-scientists) to figure out how the term prediction is commonly used in a scientific context. I even pointed it out to you and quoted it.

    And this is exactly why I was not really critical of Mike’s blog-post in general, but simply about how it was portrayed on TT.

  22. hrunBilbo, yet again, you are using two different terms here ‘hypothesis’ and ‘predict’. One (hypothesis) I think is entirely appropriate. The other (prediction) is not. You have admitted over on TT that according to FLE there could be many ways for a front-loader to implement FLE without using moonlighting proteins for the ribosome. So, in a scientific sense, this can not be a prediction of FLE. It simply is a hypothesis by Mike that he tested by looking through the published work available on Pubmed.

    I admitted that there were many ways to implement FLE. But I also argued (cogently, I think), that moonlighting would be the most likely way, and long-lasting systems the most likely place to use moonlighting. Not finding moonlighting ribosomal proteins would have been a severe blow to Mike’s hypothesis. Finding them meant that his hypothesis passed the test. Until it’s established that some other hypothesis explains the data as well, FLE is the front-runner.

    hrun: “So if you follow the scientific literature, there have been such tables compiled for over a decade and they will continue to be compiled. I therefor don’t think this qualifies as a ‘discovery’.

    I like Mike’s answer: “But, AFAIK, my list is the most exhaustive for SSU ribomsomal proteins.

    What I have done that is different is that I correlated three things with the figures above and the list – assembly map, universal SSU proteins, and moonlighting roles. I don’t know of anyone who attempted to make those connections.

  23. hrun: “I am quite certain I am banned. My old username is blocked, my attempt to register with a new username to get clarification has been banned as well.

    If you like, you could simply announce this in one of the open threads so that the folks who I was in mid-conversation with, will know.

  24. Oops. I forgot to say, “Will do.”

  25. Todd Berkebile

    I was banned too, Bradford is getting really upset with people disagreeing with him and highlighting his mistakes. I’d say I deserved it a bit more than you, your posts were always informed and civil whereas I gave up on civil when Mike left and just provoked people. 😉

  26. I admitted that there were many ways to implement FLE. But I also argued (cogently, I think), that moonlighting would be the most likely way, and long-lasting systems the most likely place to use moonlighting. Not finding moonlighting ribosomal proteins would have been a severe blow to Mike’s hypothesis. Finding them meant that his hypothesis passed the test. Until it’s established that some other hypothesis explains the data as well, FLE is the front-runner.

    Bilbo, I don’t think I have the time or energy to go over this again. I agree that Mike’s hypothesis (that ribosomes have moonlighting function) would critically suffer if no such functions would be found. As far as FLE suffering, I think you are too confident about putting probabilities on how a designer would implement FLE. How in the world would you know what the ‘likeliest’ way is to implement FLE. You don’t even know all possible ways?

    Look at it this way: You might claim that when rolling dice a seven is the most likely number. To which I would reply that it depends on how many dice you roll and what numbers are on each side.

    The equivalent is that there may be a million ways to implement FLE and you (without knowing even a fraction of them) are making claims about which one is more likely than another.

    I like Mike’s answer: “But, AFAIK, my list is the most exhaustive for SSU ribomsomal proteins.

    It might have been the most exhaustive for SSU ribosomal proteins at that particular time. But that does not make it unique. In any case, would you call it a ‘discovery’ simply because it is more exhaustive than the previous ones.

    Again, I don’t think I will be able to convince about the proper usage of scientific terms like ‘prediction’, ‘hypothesis’ and ‘discovery’– so we might as well leave it at that.

    Oops. I forgot to say, “Will do.”

    Thanks.

  27. I was banned too, Bradford is getting really upset with people disagreeing with him and highlighting his mistakes. I’d say I deserved it a bit more than you, your posts were always informed and civil whereas I gave up on civil when Mike left and just provoked people. 😉

    I have to say I am still baffled by Bradford’s reaction. I still think that I was perfectly right and he had no way to support his notion that Denton was a ‘key founder’ or ID. Influential, maybe. An inspiration for some, sure. But ‘key founder’?

    Anyway, it used to be that if he didn’t like where a thread was going to memory hole the comments and ban you from the thread. I am quite shocked that he decided to delete my perfectly civil comment, his own cussing, to lock the thread and to simply ban me without warning or announcement.

    I guess I must have really hit a nerve there. Ah well, I guess I can spend less time now trying to teach basic biology to Bert.

  28. Todd B.: ” I was banned too, Bradford is getting really upset with people disagreeing with him and highlighting his mistakes. I’d say I deserved it a bit more than you, your posts were always informed and civil whereas I gave up on civil when Mike left and just provoked people. ”

    Sorry you were banned, too, Todd. I always figure the purpose of a controversial blog is to have controvery. That’s difficult to do when you ban the people who disagree with you.

    hrun: “Again, I don’t think I will be able to convince about the proper usage of scientific terms like ‘prediction’, ‘hypothesis’ and ‘discovery’– so we might as well leave it at that.”

    Sounds good. That will give me more time to read Mike’s stuff.

  29. Hi hrun,

    It might have been the most exhaustive for SSU ribosomal proteins at that particular time. But that does not make it unique.

    What makes it unique is that I provide the only list which attempts to determine how many universally conserved SSU proteins are likely to have a moonlighting function.

    Also, who else has noted, “What’s striking is that all proteins involved in binding to the 3’ domain of the rRNA are not only universal proteins, but also have moonlighting functions?”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s