I recently wrote a review to celebrate the twentieth anniversary of the first Mediator complex EM reconstruction. This blog post is an update with clickable links (links to the papers and to the PDB/EMDB codes).
Recently I started a Twitter thread highlighting a couple of seminal discoveries in molecular biology going back in the 40’s 50’s #classicpaper. I received many positive feedback and the thread was quite popular. I am of course very pleased but this was also quite unexpected. Unexpected because it’s classic textbook knowledge really, unexpected (but again a pleasant surprise) because I wrongly thought apparently that nobody anymore read old scientific papers.
I certainly did not read all these papers when I was a student but as I am getting older, I have a growing interest in the history of science. The birth of molecular biology is a fascinating golden period from the revelation of the double helix of DNA to the cracking of the genetic code and first glimpses of gene regulation. This thread was inspired by reading ‘She has her mother’s laugh’ by Carl Zimmer, ‘Life’s greatest secret’ by Matthew Cobb, ‘The eighth day of creation’ by Horace Judson, ‘Brave genius’ by Sean B Carroll and ‘Histoire de la Biologie Moléculaire’ by Michel Morange (that is going to be translated in English by Matthew Cobb).
I strongly encourage students to read these books and these seminal papers (unfortunately many of these seminal papers are still under paywall but you know what to do*). Again, this golden age of creative thought and hypothesis-driven research is such an inspiration. The strategies and methodologies created from scratch for almost each experiments, the concepts developed that are still true and central to modern biology, pure joy. As I sometimes said, if time travel was possible I would love to have lived in the 50’s-60’s to witness the birth of Molecular Biology and meet all these fantastic scientists.
1⃣ 1944 Oswald Avery, Colin MacLeod & Maclyn McCarty – DNA, not protein as was commonly believed, is the hereditary material for bacteria, and the cause of bacterial transformation.
2⃣ 1947 André Boivin & Roger Vendrely – a near forgotten 2 pages in French that suggest almost explicitly that DNA –> RNA –> protein.
3⃣ 1952 & 1953 Alexander Dounce – like Boivin & Vendrely, Dounce is one of the first to propose that DNA might serve as a template for the synthesis of RNA, which in turn serves as a template for the synthesis of proteins.
4⃣ 1952 Alfred Hershey & Martha Chase – They confirmed that DNA was the molecule of heredity a.k.a as the blender experiment. (However as Matthew Cobb told me, After the experiment, after the double helix, Hershey still thought protein played a role. See this recount for example).
6⃣ 1956 & 1958 Francis Crick – The Central Dogma: once ‘information’ is passed into protein it cannot get out again.
7⃣ To read a clear explanation of the 2 unrelated hypotheses ‘The Central Dogma’ and ‘DNA -> RNA -> Proteins’, have a look at Dan Graur blog post.
8⃣ 1958 Francis Crick – The adaptor hypothesis (in On protein Synthesis): to explain how information encoded in DNA is used to specify the amino acid sequence of proteins.
9⃣ 1958 Mahlon Hoagland – Discovery of the adaptors = soluble RNAs a.k.a. tRNA.
1⃣0⃣ 1957 Vernon Ingram – The first demonstration that the abnormal haemoglobin in sickle cell anaemia patients is caused by an alteration in one amino acid.
1⃣1⃣ 1958 Matthew Meselson and Franklin W. Stahl – experimental proof of Semi-Conservative DNA replication.
1⃣2⃣ 1959 Pardee, Jacob & Monod – The PaJaMo experiment that supported the hypothesis that a molecule mediated the production of proteins from DNA (cytoplasmic messenger).
1⃣3⃣ 1961 Jacob & Monod – The fundamental basis of gene regulation, one of the most influential paper in the history of modern biology (& I am not saying that because Jacob & Monod were French). And yes RNA was already proposed by Jacob and Monod in 1961 to control the operon.
1⃣5⃣ For an historical point of view of the discovery of mRNA, see also this great recount by Matthew Cobb.
1⃣6⃣ 1961 Marshall W. Nirenberg & J. Heinrich Matthaei – A poly-U RNA was translated into polyphenylalanine in a cell-free system. This experiment provided the initial clue to breaking the genetic code. See also the didicated NIH web site.
1⃣7⃣ 1961 Crick, Barnett, Brenner & Watts-Tobin – The existing knowledge in 1961 & the experimental procedures were certainly not sufficient to allow anyone to deduce the general nature of the genetic code but they nearly solved the riddle.
1⃣8⃣ 1965 Margarita Salas – The first experimental results indicating that the direction of reading of the genetic message is from the 5’ to the 3’ end (see also My scientific life by Margarita Salas in 2016).
2⃣0⃣ 1966 Francis Crick – The Wobble hypothesis. A visionary Crick again explains why multiple codons can code for a single amino acid.
2⃣1⃣ 1967 Brenner – The last of the 64-Triplet Genetic Code is cracked.
2⃣1⃣ bis 1964 Allfrey Faulkner Mirsky – Acetylation & methylation of histones & their possible role in the regulation of RNA synthesis.
2⃣2⃣ 1969 Britten & Davidson – Like the Monod & Jacob paper in 1961, a very influential paper on gene regulation. Their theory stated the hypothesis that repetitive non-coding sequences are at the core of genetic regulation.
2⃣3⃣ 1968 Karin Ippen-Ihler – Studies using the lac operon identified the promoter as a cis controlling element for gene transcription.
2⃣4⃣ 1969 Bob Roeder & William J. Rutter – the discovery of 3 chromatographically separable forms of eukaryotic RNA polymerase from sea urchin embryos (I, II and III).
2⃣5⃣ 1970 Kedinger, Gniazdowski, Mandel, Gissinger & Chambon – Pierre Chambon also isolated 2 activities from calf thymus, Pol A (Pol I) & Pol B (Pol II), of which only Pol B was inhibited by the Amanita toxin α-amanitin.
2⃣6⃣ 1974 Roger Kornberg – The organizing principle of the nucleosome, a histone octamer, and its mode of interaction with DNA. Here and here.
2⃣7⃣ 1975 P. Oudet, M. Gross-Bellard, & P. Chambon – The first electron microscopy of reconstituted histone–DNA complexes a.k.a. beads on a string.
I am often wondering how others are dealing with their unpublished data. Do you have a drawer full of unfinished projects which will never see the light of day? There are a number of drivers that may cause projects to be stopped before completion: Unfunded, priorities changes, data are not worth the additional energy to convert them into a publication, wrong hypothesis,……
Here is the story of a project I have abandoned.
When I came back in France after my post-doc (a long time ago), one of my project seeks to define the role of sumoylation (an ubiquitin-related post-translational modification) in controlling the functional properties of proteins focusing on the possible structural and dynamic consequences of this modification. Did sumoylation act as structurally independent docking module rather than through the induction of a conformational change in the modified protein?
One obvious target, at least for me, was the transcriptional co-repressor CtBP. It was my post-doc favorite protein and CtBP was known to be sumoylated (ref 1) within its unstructured C-terminal domain (ref 2).
To promote the recombinant expression of SUMOylated proteins in bacteria, we decided to transfer the complete set of enzymes essential for this post translational modification (E1 and E2 enzymes + SUMO-1) to E. coli based on previous work (ref 3). Separating the sumoylated protein from the undesired unmodified fraction (you seldom obtain 100% modification) is often technically challenging so we decided to introduce a 6xHis tag SUMO-1 (ref 4). This strategy did work for IĸBα (ref 4) but I have never been able to optimize the protocol for CtBP. We did obtain roughly 50% of modified and 50% of unmodified CtBP (see the figure below) but because CtBP dimerizes/multimerizes, the separation of the modified from the unmodified substrate in this case was unsuccessful (a single peak in gel filtration).
As the saying goes, every cloud has a silver lining. At the time, I was very positive thinking that the structural study was still possible and even more exciting if I could obtain the structure of a CtBP dimer, one unmodified monomer and one modified monomer. Ah, the foolishness of youth.
Frustration, Frustration
So it was time to set up crystallization screenings and we did obtain some exciting crystal ‘hits’. In crystallography, salt crystal is frustration, crystal optimization is frustration, and no diffraction is frustration. CtBP may have a high degree of structural instability in its C-terminus, one modified and one unmodified monomer may lead to sample heterogeneity, whatever but unfortunately NO diffraction. So after a couple of tries (additives, drop ratios, other conditions), I have finally abandoned this project.
Maybe I did not try hard enough, maybe one day someone will solve the structure of sumoylated CtBP or maybe one day, I will try again.
Eventually they will understand that there are no such things as applied science, only applications of science (Louis Pasteur).
Eventually they will understand that it takes 15-20 years for basic research evidence to reach clinical practice.
Eventually they will understand that it is impossible to predict what questions will actually find practical applications in the future.
Eventually they will understand that spending on Basic research now provides the raw material for the next generation of technological advances that fuel our economic growth.
Eventually they will understand that Basic research is the pacemaker of technological progress.
Eventually they will understand that Science is a public good.
Eventually they will understand that the power of CRISPR Cas9 –based genome editing is something nobody could have predicted at the outset.
Eventually they will understand that without more than 30 years of research about gene expression, Transcription factor-based cellular reprogramming would have never opened the way to converting somatic cells to a pluripotent state.
Eventually they will understand that accidental discoveries continue to flourish in basic research with direct consequences for everyday life.
Eventually they will understand that human curiosity, creativity and inquisitiveness are the driving forces behind basic research.
Eventually they will understand that excellence should be the goal of funding, not a barrier to it (Mike Galsworthy).
Eventually they will understand that if you can accurately predict outcome, what you are doing is not research, it is development (Jim Woodgett).
Eventually they will understand that with funding rate for NSF/NIH/NHMRC/CIHR/ANR/…. less than 15%, very talented junior scientists will be leaving research.
Eventually they will understand that with funding rate for NSF/NIH/NHMRC/CIHR/ANR/…. less than 15%, basic research will suffer from the loss of knowledge and expertise.
Eventually they will understand but it is already too late.
Some of my colleagues often asked me what I am doing on Twitter. Below are some of my answers.
Twitter, the micro-blogging platform may be viewed as fascinating for some people but also frightening and boring for others. It is certainly a controversial subject. But Twitter is a diamond in the rough for the scientific community: keeping up with current research in real time, follow conferences, improve your professional network, bibliography search,…
This post does not aim to be a scientist’s guide to social media in general and to Twitter in particular. The objective is simply to share my experiences as a scientist in social media. As Zen Faulkes (@DoctorZen) quite rightly stated here : ‘Everything that happens on social media has been happening at conference for as long as there have been conference (informal conversations). Social media is just the biggest research conference in the world’.
(click to enlarge the images)
1- A bibliography search tool
1a- Scientific journals twitter accounts. Forget Pubmed, RSS feed or eTOCs. Just follow your favorite journal on Twitter. So far, I have a list of 291 journals.
1b- Keeping up using Twitterbots. An increasing number of people are exploring the use of twitterbots for more productive academic purposes (for more info see here a great explanation by @caseybergman). For example, I have a list of domain-specific literature bots here.
1c- The keyword search. Twitter has of course a search engine. Simply use keyword search as you would search Google or Pubmed. Below is an example with ‘CRISPR’.
1d- Sharing information. Twitter is particularly well suited to sharing information from your own scientific readings. A classical tool in your Twitter belt is the ability to share a link (and a photo) to the article.
1e- #icanhazpdf. The famous hashtag used to coordinate the exchange of scholarly papers. Suppose that you need a certain journal article but do not have a subscription to the journal. Anyone who notices checks to see if they have access, such as via their university’s institutional subscriptions, and if they do, they download the article and send it to you.
2- To keep you updated and engaged
2a- Outreach from scientific conferences. Looking for a forthcoming conference? Many scientific societies that organized academic conferences are on twitter: CSHL meetings, Keystone Symposia, Cell Symposia. 2014 saw the increased interaction with many conference twitter accounts with delegates actively tweeting about the meetings. For example with the Annual scientific meeting of the Australian diabetes society @ADS ADEA.
2a-bis Conference live-tweeting. Twitter lends itself particularly well to sharing information from a conference and live-tweeting is a growing trend. Live-tweeting is simply when twitter users tweet key points from presentations that they attend at conferences. Just follow the hashtag for the conference.
2b- Funding opportunities and notices. Many funding opportunities can be found on Twitter.
2c- Pharma and biotech companies. Find company information such as new products, promotional items.
2d- Career opportunities. Many principal investigators do advertises their PhD and postdoctoral positions on Twitter.
3- Help, share, discuss
3a- Lauching a new topic. Do you have a particular question ? Is there any specific topic you would like to discuss ? Twitter is the place to be for the scientific community.
3b- Direct contact. Do you have a particular question for a biotech company ? Ask directly your question via Twitter. Communication is faster than ever.
3c- Promoting your research. Publicise yourself, promote and present your work, your papers, your blog. Twitter is as a global science communication tool.
3d- Networking. Having real-time scientific discussions from your bedroom with people across the world, conversations that you would not necessarily have otherwise, expending your network.
4- Further reading
Online collaboration : scientists and the social network (link) by Richard Van Noorden (@Richvn)
Burning platforms: friending social media’s role in #scicomm (link) by Jim Woodgett (@jwoodgett)
10 simple rules of live tweeting at scientific conferences (link) by Ethan O. Perlstein (@eperlste)
Science and social media: some academics still don’t ‘get it’ (link) by Kirk Englehardt (@kirkenglehardt)
[1] A recent paper attributed a large proportion of variation in incidence of cancers across different tissues to the number of stem cell divisions in them, and
stochastic errors in cell division.
[2] The paper grouped tumour types with known external causes as “deterministic” and those without as “stochastic”
[3] I have seen people being hostile to the notion of stochasticity in cancer who’ve postulated other deterministic factors, with the implicit assumption that what is stochastic is really deterministic processes with as-of-now undiscovered causes.
[4] Here I explain why processes with known causes are still stochastic, leading to my gripe with both the misunderstanding that has permeated discussion of the paper as well as the iffy notion of grouping tumours into stochastic and deterministic ones in the paper. My assertion is that even those cancers strongly driven by external carcinogens involve randomness/stochasticity.
Academic publishing in general and the peer-review process in particular, if not broken, are seriously under strain. We all remember Arsenic life or the more recent STAP cells fiasco. Pre-publication peer-review is unfortunately not always getting the job done as a filter.
Many publishers have already embarked on experiments/alternatives/developments with respect to improving transparency and efficiency. Unfortunately, each journal has its own version of peer review. This blog post deals with these currently available alternatives, hoping (dreaming) that one day those ‘new’ policies may becoming the norm for all the publishers societies.
Referee cross-commenting
As a reviewer I would love to be able to see the final decision and other Reviewer’s comments. But it not always the case. I have probably reviewed about 20-25 papers since the beginning of my career. Only a couple of time I was informed of the final decision and in only one recent reviewing, the editor send me a message to let me know of other reviewer’s comments.
So for me, to have access to the comments provided by the other reviewers should be compulsory. EMBO press employs this review format and I think it’s great and very useful. It is partly designed to minimize contradictory statements.
The full Monty- Uncropped scans of Western blots included in supplemental figures.
If you have nothing to hide, you have nothing to fear. We all do want to show pretty and clean data but you don’t have to make results look better than reality. With regards to ‘representative data’, a lot of journals such as Nature Cell Biology now require to send all the unedited, uncropped scans of Western blots with your manuscript. Peer review should be a gatekeeper for possible doctored images and doing the ‘full Monty’ appears to me to be going in the right direction. Systematic image screening similar to those made at EMBO press should also become a standard for all publishers.
Transparent review
Transparency is one of the fundamental guiding principles in science. Would the publication of referee reports and editorial decision benefits the debate? EMBO press certainly thinks so (an example here), so do I. One direct benefit is to know to what extent a paper has been improved during the peer review process.
Pre-Print servers
An invite approach that has worked well for the physics community is the use of the pre-print server arXiv. Seeing the emergence of several preprint servers to biology (fighare, BioRxiv, peerj and F1000Research) is certainly a good sign. It is an effective way to share and get a collegial feedback not restricted to 2-3 reviewers. Other advantages include rapid dissemination and immediate visibility. We are constantly answering questions about our work at meetings, seminars, conferences and with our publications. But you usually answer to a couple of people. By sharing your work openly you can answer 100, 1000! The more feedback you receive, the better your work will be. Unfortunately not all academic journals submission policies allow pre-prints and its very regrettable.
That’s it for today. Please feel free to comment and give your thoughts/feedback.
Rule 1: Avoid conflict of interest. I will disqualify myself from review if I feel unable for any reason to provide an unbiased assessment.
Rule 2: Ask yourself honestly whether the paper falls within the scope of your expertise. If I don’t fell qualified for the paper, I will decline to do the review.
Rule 3: Punctuality is a virtue of kings. I will return my review within the specified deadline. There are many sources of unnecessary time loss in the publication process. Everyone loses out if some do not play by the rules.
Rule 4: Review unto others as you would have them review unto you. I will not propose a bunch of new experiments, especially the ones that I do not perform for my own work.
Rule 5: Leave it to the future to judge a manuscript’s impact. I will only evaluate the evidence for the claims. Impact is unpredictable. Peer review is only a process of ‘pre-filtering’. Readers are the ‘post-filter’, in other words = peer validation.
Rule 6: It is their papers, not yours. I will not try to turn author’s paper into a paper I would have written.
Rule 7: Review the work, not the authors. Whether the author is a Nobel laureate or a graduate student, I will judge the paper the same.
Rule 8: Don’t hide behind a cloak of anonymity, sign your review. I will have the courage to stand by my reviews, including negative reviews.
Wang et al. (Cell research 2014) and Tsai et al. (Cell 2014) recently describe a substantially improved cryo-EM of the Mediator Complex which permitted an unambiguous and topsy-turvy assignment of the Head, Middle and Tail modules.
A major point of agreement that emerges from these extensive biochemical studies is that Mediator subunits are organized into three core modules (Head, Middle, Tail) and a dissociable CDK8 kinase module. However, information about subunit localization and boundaries of the three core modules has remained rather elusive, even contradictory. For example a recent cryo-EM of yeast Mediator at 28 Å resolution identified a previously additional independent module referred to as the arm domain (Cai et al., 2009).
In two recent reports, the Cai (Wang et al., 2014) and Asturias (Tsai et al., 2014) labs describe a cryo-EM analysis that completely redefine the modular organization of the core Mediator. In particular the head module was previously assigned to one end of the Mediator structure with the middle and tail modules folded back on one another to from the upper portion of Mediator (Chadick and Asturias, 2005).
Using either tagged or deleted individual subunits combined with unequivocal docking of the X-ray structure of the Head module, the authors arrive at an impressive improved cryo-EM reconstruction of Mediator at a resolution of ~ 18 Å. And in a dramatic topsy-turvy twist, the Head and Middle modules form now the upper portion while the dense domain at the base corresponds to the Tail. As a consequence of the enhanced resolution, previously unassigned metazoan-specific subunits are now clearly localized. For example, MED27, MED28, MED29 and MED30 make extensive contacts with the Head module while MED26 associates with the Middle module.
Does this completely new pattern of modules rearrangement provide a more concrete view of the conformational changes that these modules undergo upon interaction with the RNA polymerase II ? Previously, the most prominent change resulted from the relative rotation and translation of the Middle and Tail modules that leads to a complete repositioning of the Middle module (Cai et al., 2009). These module movements triggered by formation of the holoenzyme are still carried on but from now on that is the Head and the Middle modules that undergo a coordinated rotation.
These observations directly challenge the previous holoenzyme model in which the reported RNA pol II binding site was located near the Head module. Astonishingly, the entire interaction surface of the Head module is now highly exposed and extensive Mediator/RNA pol II contacts are mediated through the Middle and Tail modules.
In the future, with the cryo-EM resolution revolution (Kühlbrandt 2014), the near-atomic resolution of full Mediator and even of the full transcription pre-initiation complex (PIC) may soon become a reality.
(
An infinity of hypotheses 
