The interview we have just published with Tim Hunt was recorded as a gentle reminder that there is more to biological science than can be captured by reporting checklists, now rapidly proliferating in response to increasing demands for more rigor in research.
And in case anyone was wondering, the interview was recorded in May, and has nothing at all to do with the much too well publicized events of June and their sequellae. We were at that time planning our own reporting checklist – more on this can be found in our editorial and our accompanying blog – and were reflecting on the need to take into account the wayward character of biology.
There is an impeccable case for imposing methodological and statistical rigor in preclinical research that has explicit practical implications. But more latitude may be allowable – even essential – where research is exploratory, the bench is at a very safe distance from the bedside, and systems or organisms may be intransigent to adequate sampling or current established methodology.
Major discoveries are often cited as object lessons in the dangers of being too hidebound, fastidious or doctrinaire.
Major discoveries are often cited as object lessons in the dangers of being too hidebound, fastidious or doctrinaire. Tim Hunt’s discovery of cyclin isn’t a bad example of how fundamental insights can emerge, eventually, at the end of a semi-random excursion along a murky path.
It certainly wasn’t obvious to the world at the time that it was a major breakthrough, and he is reasonably confident that the paper reporting it would not now be published in the high-profile journal where it appeared in 1983.
Phenomenology and wild guesses
At that time, nothing was known about the regulation of the cell cycle (and Hunt wasn’t working on it). The only clue was an activity (sic) called MPF, assumed to be a protein, that induced the completion of the meiotic divisions of frog oocytes – a rather specialized cell cycle transition.
All Hunt and his colleagues had was an unknown protein that appeared and disappeared cyclically in sea urchin eggs when they were fertilized. They were not looking for a cell cycle regulator at the time, and they had not even used up-to-date 2D methodology for running their gels (luckily, as it turned out, since for unclear reasons it is difficult to see cyclin on 2D gels).
The speculation in the discussion to the paper, which was published after revision despite the scathing comments of one of the referees, is clearly off the mark: the authors were plainly convinced they had something important, but equally plainly had no idea what was going on. They thought it must have something to do with MPF (which it did), though they didn’t quite say they thought it might actually be MPF (which it wasn’t).
In fact they couldn’t have known what was going on. The phenomenon of a protein that came and went with each division cycle didn’t fit with anything that was then known about how proteins behaved. Nor did the gradual accumulation of cyclin during the cell cycle fit at all with the abrupt transitions that actually occur when cells enter mitosis.
With the benefit of hindsight…
We now know that cyclin is a regulatory protein that is continuously synthesized and periodically degraded, and that it binds and activates the central cell-cycle regulator now known as Cdk – cyclin-dependent kinase – then only known as the activity MPF. Activated Cdk phosphorylates downstream targets that execute the steps in the cell cycle, including those that launch cyclin on the path to destruction, switching the kinase off.
The abrupt re-entry into the cycle as cyclin reaccumulates is due to regulatory phosphorylations that emerged only well after the discovery of cyclin, when MPF/Cdk had been characterized, and with the help of yeast genetics.
That’s the outline of the story. The entire system of controls converging on the central Cdk-cyclin pair is so complex that it is still not completely understood, and an interesting account of the progressive understanding of its details can be found in a recent article from John Tyson and Bela Novak explaining the attempt to capture its logic with formal modelling.
There is no way on earth that anyone at the time that cyclin was discovered could have projected the regulatory system by which cell division is fine-tuned and constrained.
It is sobering to note that according to Hunt it took five years for the significance of his paper to be recognized, and during those years it was hardly cited at all.
Meanwhile it is sobering to note that according to Hunt it took five years for the significance of his paper to be recognized, and during those years it was hardly cited at all. So it wouldn’t have done much for the impact factor of the journal it was published in.
Reporting standards, rigor, and the boundaries of knowledge
What has this to do with reporting standards and rigor? Not much. The actual data in the original cyclin paper are not in dispute – although had the authors been asked to repeat or – more likely – extend them to establish their significance or mechanism, as they probably would have in today’s climate, they would have been stuck. It would have been another year before they could harvest sea urchin eggs again.
And they would presumably have been in trouble if they had been asked to use current 2D gel technology to confirm their findings.
It does, arguably, show how papers pushing the boundaries of the known may fall foul of the criteria now generally applied by high-profile journals.
Ron Vale has recently speculated, in this vein, on how today’s editors might have responded to the 1953 paper by James Watson and Francis Crick on the structure of DNA.
- The scientific Odyssey: Pre-registering the voyage - 3rd May 2017
- A year of almost anything you can think of in life science – an eclectic pick from BMC Biology - 13th January 2017
- Peer review: opting out - 23rd September 2016