A Blog Around The Clock

Blogging on Peer-Reviewed Research

There is nothing easier than taking a bad paper – or a worse press release – and fisking it with gusto on a blog. If you happen also to know the author and keep him in contempt, the pleasure of destroying the article is even greater.

It is much, much harder to write (and to excite readers with) a blog post about an excellent paper published by your dear friends. But I’ll try to do this now anyway (after the cut).

Paul Shaw is a friend, and Indrani Ganguli is a good, good, good friend. Faculty and graduate students in biology are usually a pretty smart lot. A subset of those, as self-selected friends discussing the Big Picture, tends to be even smarter. Then, within such a group, you have a person who stands out with brilliance. That was Indrani. She did an excellent MS thesis and an excellent PhD Dissertation (on olfaction in Drosophila), but I suspected that she would do some really great stuff only once she gets to be independent and do whatever she wants. The Neurosciences Institute is just such a place, fostering indpendence and deep thinking.

About a month ago they published a paper (pdf) in Science. You know how Science papers tend to be short one-pagers with lots of information concentrated in a very small space (and some more info provided online). Each word, each sentence and each figure is there for a reason. One needs to look for what is missing in order to poke holes in the argument. Well, Indrani’s paper is almost 7 pages long – that is War And Peace! Enormous amount of information highly concentrated in one spot. It took me a while to read it through and think about each sentence and why it is in there.

The questions of evolution and adaptive function of sleep are often marred by a historically anthropocentric view of sleep. While the sleep in mammals may have taken on many different functions over geological time, it is not neccessary that all those functions were associated with sleep in its incipient evolutionary stages. They could have been added later.

Enormous amount of research has been done on the role of sleep in learning and memory, as this putative function of sleep is one of the hypothesized adaptive functions of sleep, i.e., the original reasons why sleep appeared in animals at all, in other words: what sleep is for. While the evidence for importance of sleep is mounting, the mechanisms are still completely elusive and the findings are hotly contested by some. Most of this research has been performed in human subjects and in rodent models. Indrani’s study is, to my knowledge, the first one to demonstrate the connection in an invertebrate.

Indrani also went to the historical sleep and clocks conference in which the human-centered sleep research community was first blown away by research on sleep in fruitflies. Perhaps this is where she got her ideas what she wanted to do after grad school, and here it is:

Waking Experience Affects Sleep Need in Drosophila (pdf)

I read through the paper slowly and carefully. Every time I read a portion I would come up with a question: but, but, what about X? In the next paragraph, there would be an answer – here is an experiment that tests for X. Then Y. Then Z. All bases covered, no holes left unplugged. I ended up not having any problems with the paper at all.

What she did first was to show that social environment affects the amount of sleep. Fruitflies kept in isolation sleep less than fruitflies kept in groups. As these insects already sleep through the night, the only time where additional sleep can be fitted in is during daytime – taking naps. Flies in group settings did not have more naps, but their naps lasted much longer. The volume of the environment (the size of the vial) had no effect on this. On the other hand, the number of flies kept together did matter: more flies in the vial, longer the naps.

The increased sleep was not due to physical tiredness provoked by social interactions – the levels of locomotor activity were the same between isolated and group-housed flies. A bout of sleep deprivation (both high and low levels of agitation) did not change the pattern – group-housed flies rebounded to their pattern of sleep, while isolated flies rebounded to their pattern of sleep. There was no effect of sex, reproductive state, sexual experience or sexual composition of the group.

Deletion of circadian clock genes and genes related to hearing had no effect either. However, deletion of genes related to vision and olfaction eliminated or greatly reduced the effect, suggesting further that input of information in a group setting causes the increase in sleep.

The response was plastic. Flies reared in isolation developed the short-naps pattern, but once moved to a group setting switched to the long-nap pattern. It also worked in reverse and it worked repeatedly.

Levels of dopamine, neurotransmitter impicated in learning and memory, were three times higher in mushroom bodies of group-setting flies than in the isolated flies. Mutations in genes involved in dopamine regulation and processing greatly affected the response of sleep duration to social setting. Mutations in a number of genes (but not all) impicated in learning and memory in flies (though I do not know how contentious this role is for any of these genes) also disrupted the experience-related changes in sleep. Most, but not all of these genes are expressed in mushroom bodies.

Training flies (a courtship learning test) increased sleep. In a series of behavioral tests, flies were taught and then sleep deprived. Flies deprived during the first post-training night neither remembered the task nor increased their nap-times. Flies not sleep-deprived and flies sleep-deprived during the following day both remembered the task and increased their nap-duration.

Put together, the paper provides evidence at several different levels of organization (genetics, neurochemistry, development, behavior) that sleep in fruitflies is related to learning and memory of social interactions. Unlike press releases and media reports, the paper itself is very cautious about its claims. The paper pretty much only states that Drosophila appears to be a good model for future studies of the connection between sleep and memory in flies – which sounds quaint but is actually a big claim as nobody has shown such a possible link in anything outside of mammals. I would say that the data show much more, but can’t wait to see what future studies Indrani publishes next.

Check the media reports on these science websites:

Scientific American
Cosmos Magazine

In the mainstream media:

Fox News

On science blogs:

3 quarks daily


  1. #1 melior
    October 26, 2006

    I wonder if flies in groups are able to get away with longer naps partly as a result of sharing the responsibility for watching out for threats?

  2. #2 gg
    October 26, 2006

    I enjoyed your post (as I do for most of the blog in fact).
    It’s always nice to see someone using their blogs to actually try to provide some original contents in form of genuine opinions rather then the usual copy and paste.

    I believe Indrani’s paper is a very interesting one indeed but I wouldn’t be as celebrative as you seems.
    I think that most of the paper is really based on the data shown in Fig 3; those are the ones that actually keep the whole story standing and interesting.
    The paper has its flaws and some of them are quite puzzling I must say (as example: compare Fig 1A and 2D. They are virtually the same experiment if you think about it but they give opposite results) but it does offer some new hints to the field.

  3. #3 Mustafa Mond, FCD
    October 26, 2006
  4. #4 Sir Oolius
    October 26, 2006

    gg: You mean 1D vs 2D?

    melior: Weren’t they all tested individually but raised either I or E?

  5. #5 gg
    October 26, 2006

    Sir Oolius, I meant 1D,E vs 2D (sorry for the typo).
    In 1E flies are kept in Enriched environment for 4 days, than measured in isolation for 6 days. In 2D are kept in EE for 4 days, then isolated for 4 days, then analysed. First 2 days of the analisys in fig 2D should be same as fig 1E but they are very different. Anyway, it doesn’t really matter. It is a quite nice paper.

  6. #6 Indrani Ganguly-Fitzgerald
    October 26, 2006

    gg– puzzlement is well-founded, for the simple reason that despite the Editorial team’s generosity in giving me more than my deserved share of space in the journal, i was not able to be as explicit as i would have liked. the reason for the difference between figures 1E and 2D is because the trikinetic tube condition (the glass tubes in which all flies are individually tested) is qualitatively different from the plastic tubes/fly vials in which individual isolated flies were reared. the trikinetics tubes mandate back and forth walking due to space constraints, very different from a fly vial where there is ample space, ample food, and the ability to move freely. the only treatment in fly vial is missing con-specifics. isolated individuals reared in trikinetics tubes sleep the same as socialized flies. why? possibly due to a confounding effect of stress (and possibly rises in DA)on sleep? do not have an answer as to why. however, given that all flies were tested in trikinetics tubes — the confounding (if you want to call it that) effect of trikinetics tubes on sleep was cancelled out in comparisons…hope this helps.

  7. #7 Indrani Ganguly-Fitzgerald
    October 26, 2006

    and, btw– noone could doubt our friendship, coturnix. esp not following that rather *restrained* vote of confidence you have in my abilities..haha..so glad you liked the work. you were always one of my strongest critics (the other one being my mom). remember our smoke-laden debates on the back porch of our labs?

  8. #8 coturnix
    October 26, 2006

    Thank you for that clarification, Indrani and welcome to my blog. Eh, the good old times on the back porch… and of course, mutual criticism made us think even sharper.

  9. #9 gg
    October 27, 2006

    Hi Indira, did you try doing the experiment described in to 2D forcing the 4 days isolation in the DAM tubes rather than in the regular fly vials?

  10. #10 gg
    October 27, 2006

    Hi Indrani, did you try doing the experiment described in to 2D forcing the 4 days isolation in the DAM tubes rather than in the regular fly vials?
    (sorry I mispelled your name in the previous post, I apologise).

  11. #11 Indrani Ganguly-Fitzgerald
    October 27, 2006

    Hi Indrani, did you try doing the experiment described in to 2D forcing the 4 days isolation in the DAM tubes rather than in the regular fly vials?
    (sorry I mispelled your name in the previous post, I apologise).

    Posted by: gg | October 27, 2006 12:26 AM

    gg- i did. when reared in social groups (E) and then placed in DAM tubes for 4 days — they slept significantly more than (I) flies tested in DAM tubes (this is the protocol used for data presented in Figure 1B-E). In figure 2D, flies were reared in social groups (E) and then placed in single fly vials (not DAM tubes) for 4 days (creating treatment: E=>I) and then tested in DAM tubes. So, in other words, placing flies in DAM tubes did not override the effect of socialization or isolation, merely sustained it (Figure 1E).