Drink before sleep

Among the many humorous sayings, puns, and jokes that one inevitably encounters on any social medium account, one that was popular this year was about the similarity between putting a 2 year old to bed and putting your drunk friend to bed, which went like this: they both sing to themselves, request water, mumble and blabber incoherently, do some weird yoga posses, cry, hiccup, and then they pass out. The joke manages to steal a smile only if someone has been through both situations, otherwise it looses its appeal.

Being exposed to both situations, I thought that while the water request from the drunk friend is a response to the dehydrating effects of alcohol, the water request from the toddler is probably nothing more than a delaying tactic to postpone bedtime. Whether there may or may not be some truth to my assumption in the case of the toddler, here is a paper to show that there is definitely more to the water request than meets the eye.

Generally, thirst is generated by the hypothalamus when its neurons and neurons from organum vasculosum lamina terminalis (OVLT) in the brainstem sense that the blood is either too viscous (hypovolaemia) or too salty (hyperosmolality), both phenomena indicating a need for water. Ingesting water would bring these indices to homeostatic values.

More than a decade ago, researchers observed that rodents get a good gulp of water just before going to sleep. This surge was not motivated by thirst because the mice were not feverish, were not hungry and they did not have a too viscous or a too salty blood. So why do it then? If the rodents are restricted from drinking the water they get dehydrated, so obviously the behavior has function. But is not motivated by thirst, at least not the way we know it. Huh… The authors call this “anticipatory thirst”, because it keeps the animal from becoming dehydrated later on.

Since the behavior occurs with regularity, maybe the neurons that control circadian rhythms have something to do with it. So Gizowski et al. (2016) took a closer look at  the activity of clock neurons from the suprachiasmatic nucleus (SCN), a well known hypothalamic nucleus heavily involved in circadian rhythms. The authors did a lot of work on SCN and OVLT neurons: fluorescent labeling, c-fos expression, anatomical tracing, optogenetics, genetic knockouts, pharmacological manipulations, electrophysiological  recordings, and behavioral experiments. All these to come to this conclusion:

SCN neurons release vasopressin and that excites the OVLT neurons via V1a receptors. This is necessary and sufficient to make the animal drink the water, even if it’s not thirsty.

That’s a lot of techniques used in a lot of experiments for only three authors. Ten years ago, you needed only one, maybe two techniques to prove the same point. Either there have been a lot of students and technicians who did not get credit (there isn’t even an Acknowledgements section. EDIT: yes, there is, see the comments below or, if they’re missing, the P.S.) or these three authors are experts in all these techniques. In this day and age, I wouldn’t be surprised by either option. No wonder small universities have difficulty publishing in Big Name journals; they don’t have the resources to compete. And without publishing, no tenure… And without tenure, less research… And thus shall the gap widen.

Musings about workload aside, this is a great paper, shedding light on yet another mysterious behavior and elucidating the mechanism behind it. There’s still work to be done though, like answering how accurate is the SCN in predicting bedtime to activate the drinking behavior. Does it take its cues from light only? Does ambient temperature play a role and so on. This line of work can help people that work in shifts to prevent certain health problems. Their SCN is out of rhythm and that can influence deleteriously the activity of a whole slew of organs.

Summary of the doi: 10.1038/nature19756 findings. 1) The light is a cue for suprachiasmatic nulceus (SCN) that bedtime is near. 2) The SCN vasopressin neurons that project to organum vasculosum lamina terminalis (OVLT) are activated. 3) The OVLT generates the anticipatory thirst. 4) The animal drinks fluids.

Reference: Gizowski C, Zaelzer C, Bourque CW. (28 Sep 2016). Clock-driven vasopressin neurotransmission mediates anticipatory thirst prior to sleep. Nature, 537(7622): 685-688. PMID: 27680940. DOI: 10.1038/nature19756. ARTICLE

By Neuronicus, 5 October 2016

EDIT (12 Oct 2016): P.S. The blog comments are automatically deleted after a period of time. In case of this post that would be a pity because I have been fortunate to receive comments from at least one of the authors of the paper, the PI, Dr. Charles Bourque and, presumably under pseudonym, but I don’t know that for sure, also the first author, Claire Gizowski. So I will include here, in a post scriptum, the main idea of their comments. Here is an excerpt from Dr. Bourque’s comment:

“Let me state for the record that Claire accomplished pretty much ALL of the work in this paper (there is a description of who did what at the end of the paper). More importantly, there were no “unthanked” undergraduates, volunteers or other parties that contributed to this work.”

My hat, Ms. Gizowski. It is tipped. To you. Congratulations! With such an impressive work I am sure I will hear about you again and that pretty soon I will blog about Dr. Gizowski.


5 thoughts on “Drink before sleep

  1. There is an acknowledgment section, so perhaps before discrediting the authors, you should do your research, and give credit where it’s due.


    1. First: There is indeed an Acknowledgments section, which, inexplicably, I missed, probably because it is AFTER the References section, and I’m accustomed to see it before. Nevertheless, thanks for pointing it out, my apologies! Here is it: ‘Acknowledgements This work was supported by a Foundation Grant from the Canadian Institutes of Health Research (FDN 143337), a James McGill Chair, and a Studentship awarded by the Research Institute of the McGill University Health Centre (RIMUHC). The RIMUHC receives generous funding from the Fonds de Recherche Québec Santé. We thank M. Bouvier for human V1aR; H. Gainer for anti-VP neurophysin primary antibodies; and N. Cermakian, K-F. Storch, and M. Prager-Khoutorsky for comments on an early draft of the manuscript‘. Although to be fair (and little nasty, I admit) the authors thank here people for donations and comments and not work, am I correct? Which would mean that if there were people who contributed to the work, they were not acknowledged. Please, before you get all your feathers up, relax, understand the meaning of IF.

      Second: I did not discredit anyone. As you encourage me to do my research by, presumably, re-reading the paper and I did so and made the necessary corrections, so I encourage you to read my words again. I merely pointed out that either there were more people doing work or the 3 authors really know a lot of different techniques and how neither option would surprise me. I would be terribly naive (and misinformed) if I thought that everybody received their deserved credit everywhere, everytime, wouldn’t I? It is an opinion, not a fact, and I present it as such. But, furthermore, if you continue reading, you will also observe how I emphasized the workload required today to make a point that would have been accepted with less work a mere decade ago. Thus acknowledging the possibility that the authors being prolific in all these techniques is a LIKELY one. Maybe I was not clear enough with all those ‘significant pauses’ I inserted. Still, please, un-ruffle your feathers, there was no discrediting. Which brings me to my final point:

      Third: If I indeed wanted to discredit the researchers (which I tell you I have no reason to do so), I would have been far more straightforward in my endeavor, without any innuendos. In the same vein, if I was biased in any way, I doubt I would have allowed your comment on my blog, let alone answer it. But, since I am trying my darnest to present the facts as they are WHILE also voicing my own opinions, here you go :). Hope this clarifies matters.


  2. The techniques used are pretty common for well-rounded researchers. Knockout animals can be ordered from commercial suppliers. Immunofluoresence is a mainstay of neuroscience, and is a straight forward protocol (same for c-fos). Performing optogenetic viral injection requires the same skills needed for tracing, electrophysiology is a specialized but common lab technique, and behaviour isn’t particularly difficult to assay, often just needing special equipment and software. I would expect that a student / postdoc, albeit a talented one, should be capable of performing all of these techniques over the course of a year or two.


    1. You are absolutely right about the rounded researchers today, Jim. But that’s my point, a well rounded researcher TODAY. A decade ago, a well rounded neuroscientist was one who could asses behavior and do some immuno. I’m pretty sure the Journal of Neuorscience’s Behavioral a.k.a. Pharmacology section from the early naughts would prove my point. The researcher trained in E-Phys more often than not was not also trained in behavior (could think s/he could do it, but installing a software a behaviorist does not make). Of course, there are exceptions. Many. But how many fMRI specialists in the early 2000 had any idea about how to adjust the ph of a solution? Also, a decade ago, most neuroscientists were trained by Psychology departments and I know for a fact that many PhDs granted from these departments at the time did not require a Molecular Biology and Genetics course for graduation. Sure, anybody can now order a kit and do some epigenetics, for example, but I’m sure you agree that just following the steps on the back of the box like a beginner cook trying a new cake recipe will probably not lead you to a Nature paper. A little bit of theory and understanding offered by graduate courses would not come amiss, don’t you agree? Courses which are more and more offered in the so-called humanist departments these days, to their credit. As for knockouts, that is easy now, but not in the 2000s when a lot of behavior work was done in rats because they are bigger and better behaved than mice (so I’ve heard) and you couldn’t order any knockouts because the genome was not done. So the rat PIs used oligonucleotides to alter gene expression, while trying to decide whether or not is worth writing a new grant and buying new equipment for the lab to transition to mice to take advantage of the the genetic mods. I grant you that immuno is easy and even an undergrad can get reliable results, without understanding what’s happening. Same with c-fos, although I also remember that in 2005 there wasn’t a reliable c-fos automatic count software (our lab tried to write a patch for I forgot what, ImageJ perhaps?, but then we said the hell with it and counted by hand). Now is easy-peasy. You are again right about opto requiring the same skills any intra-cranial surgery, but surely not the same equipment? How much do those lasers run for these days, by the way? So small universities and labs would be at a financial disadvantage because even if they have the talent, I doubt all that small labs can afford all this equipment, space, reagents etc. That was part of my point, lack of resources may be more deleterious to science production in small universities than lack of talent. As for a year or two about the talented postdocs… what can I say? Of course is possible, almost everything is possible, but only if they are trained by someone that knows all these techniques by heart, all the troubleshouting has been done, the lab has awesome SOPs on the main computer for all the techniques to be printed at will, so the poor postdoc does not have to introduce de novo a technique because than I can guarantee that troubleshooting can take up to a year. Finally, (wow, I am loquacious AND babbling today), I will say it again: I really did not have any intent of disparaging the authors, nor any reason. I don’t know them and their paper is well done. I am surprised that my musing on lack of resources of small universities and othe coutnries along with decade old reminiscences have ruffled feathers. Oh well, I hope that my desperately long and unchecked for errors answers to the comments on this post have shed some light where my laconic writing did not. Anyway, thanks for stopping by, Jim.


  3. Hi, I am the PI of that study (Charles Bourque). The PhD student who did the work (Claire Gizowski) was a little ruffled by the implication at first. But I did tell her the writer had fairly put both possibilities in the text and that, other than this side opinion, this was a wonderful piece and she should be very happy about it. I certainly am, THANK YOU! Let me state for the record that Claire accomplished pretty much ALL of the work in this paper (there is a description of who did what at the end of the paper). More importantly, there were no “unthanked” undergraduates, volunteers or other parties that contributed to this work. I agree that three authors on a study of this type is impressive. But this is to Claire Gizowski’s merit.

    Liked by 1 person

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