In 20051, the Liu lab described a new central element in the Neurospora circadian clock. They found that an RNA helicase with similarity to the yeast exosome cofactor Dob1p/Mtr4p, associated with FRQ, a core clock component essential for circadian clock function in this fungus (it was termed FRH, for "FRQ-interacting RNA helicase"). Their evidence suggested that FRH played an important role in the circadian negative-feedback loop. The same lab showed last year that FFC (the FRQ + FRH complex) and the exosome were "part of a posttranscriptional negative feedback loop that regulates frq transcript levels and the circadian output pathway"2.
OK, so FRH is an important component of the central oscillator. As I have mentioned before, circadian systems are composed of a central oscillator and two signaling pathways: input pathways convey external signals to the oscillator, so that it can be synchronized with the environment, and output pathways allow the oscillator to temporally regulate diverse cellular processes.
Even though discussing both of these papers at length would be a great blogging topic, I just want to direct your attention to a little detail. Let's take a look at figure 5 from the 2005 paper:
You'll notice that frq mRNA levels, as it had been known for many years now, oscillates in constant conditions (DD= constant darkness). Note that frq peaks at around DD16 and then at DD 36. As Cheng et al. show, this oscillation is absent in a strain in which frh has been downregulated (dsfrh), and frq mRNA levels remain high throughout the subjective day. Further, they show that the oscillation of an output gene3, ccg-2 (for clock-controlled gene 2), is also altered in the dsfrh strain as compared to the WT (lower panel).
This is all very nice and fascinating, but wait...what happened at DD24? You'll notice that there is a blank space in the Northern membrane for ccg-2.
This is what the authors say about it in the figure legend:
(My emphasis)Figure 5. (A) Northern blot analyses showing the expression of frq and ccg-2 in DD. Cultures were harvested at the indicated hours in DD. QA was present in the medium for both strains.. The RNA samples at DD24 for ccg-2 were mishandled. (...)
"Mishandled"!? Really?
I realize that the conclusions of the paper do not depend on that particular time point and that the figure's general idea is perfectly clear without it [the downregulation of frh a) alters the rhythmic expression of a central clock component (frq) and b) disrupts circadian rhythmicity, as measured by the daily levels of ccg-2], but couldn't have they done the experiment again for the paper figure? I mean, they must have done it a couple of times to be sure that the result they were getting was reproducible, so why are they showing the Northern blot with the missing time point?
If I lose a sample, my PI ( and, basically common sense!) will tell me to do the experiment again, as I'm sure most PIs would. So, why they decided to go with this figure and justify the missing time point, instead of sending the figure with a different biological replicate, is a mystery to me.
Thoughts?
____
1Cheng P, He Q, He Q, Wang L, & Liu Y (2005). Regulation of the Neurospora circadian clock by an RNA helicase. Genes & development, 19 (2), 234-41 PMID: 15625191
2Guo J, Cheng P, Yuan H, Liu Y. (2009) The exosome regulates circadian gene expression in a posttranscriptional negative feedback loop.Cell 138(6):1236-46. PMID:19747717
3 An output gene is basically (and traditionally) a gene whose time-of-day specific expression is dependent on the circadian oscillator and that it doesn't play any role on the functioning of the central clock. Its disruption then, has no impact on the functioning of the clock, even though it may lead to other phenotypes.
5 Comments:
I have to agree with you fully. I cannot see why they would use the "mishandled" data. As you mentioned they must have more complete results that could have been used. I would like to hear their reason(s) for using that particular one.
PNAS direct submission? I cant imagine a scientist that´s performing, analyzing, writing the paper or reviewing it that would accept that.
Seems quite suspicious to me... Makes you wonder if they ever did biological replicates... Anyway, I doubt my PI would have accepted such a figure for a paper :(
Perhaps it was difficult to do the RNA extractions + blotting in the time-sensitive way? From what I know, time course experiments are a bitch to run, and require experimenters to sacrifice their own health and sanity for days on end. Perhaps this was the better of their gels, and if it's just some subsidiary figure of secondary importance, it just wasn't worth the time and money to run again. RNA work is notoriously finicky, so I wouldn't necessarily jump to conclusions so quickly...
Also, given the competition in that field, I wouldn't be surprised if the choice was between polishing up the experiment risking scoopage, or actually getting the paper published. In an ideal world, no one would have to sacrifice the quality of their science to get published, but in the real world it would be outright irresponsible for the supervisor to risk losing a key publication for a team of young scientists just because of a glitch in a diagram.
Unfortunately, we simply can't afford to do perfect science. Or, nowadays, even good science sometimes...
@Psi
We do those sort of experiments fairly frequently in our lab, and while they are not the type of experiments you'd hope for, they can certainly be performed in a week or so.
So, I'd have to disagree with the "difficulty" or the "possibility of scoopage" arguments, in this context.
I think the picture you portrait regarding losing the paper by doing this sort of experiment again, is bit exaggerated, considering the timespan involved.
I agree that without that timepoint, you can still get the idea across, but justifying not including it by saying that it was "mishandled", seems a little surreal.
Post a Comment