A Blog Around The Clock

Blogging on Peer-Reviewed Research

i-710d005c8660d36282911838843a792d-ClockWeb logo2.JPGIf you really read this blog ‘for the articles’, you know some of my recurrent themes, e.g., that almost every biological function exhibits cycles and that almost every cell in every organism contains a more-or-less functioning clock. Here is a new paper that combines both of those themes very nicely, but I’ll start with a little bit of background first.

Daily Rhythms in Sensory Sensitivity

If almost every biochemical, physiological and behavioral function exhibits daily cycles, it is no surprise that such rhythms have been discovered in sensory sensitivity of many sensory modalities – vision, hearing, smell, etc. Even the last lone holdout – electeroreception – has recently been shown to exhibit daily rhythms of sensitivity.

But the phrase “sensory sensitivity” is vague. What it means is that there is a daily switch in the power of detection versus power of resolution. In most senses, the two are incompatible. If the sensory organ is tuned to detect the lowest possible amount of energy, it is incapable of maximal resolution and vice versa. What has been seen so far in all the sense studied to date is that the power of detection is maximal during the time the animal is inactive (e.g., quiescent or asleep), while the power of fine resolution is maximal whne the animal is active.

This makes sense from an adaptive perspective. When you are out and about looking for food or mates and on a lookout for enemies, it is a good idea to be able to finely discrimate the details of the environment, be it visual, auditory, olfactory, geomnagnetic, etc. At the same time, lowering the treshold for detection is useful when you are asleep and need to respond by sudden awakening to possible cues of danger in the environment – there is no immediate need for fine discrimination, but there is a need to overpower the sleep mechanism and wake the animal up quickly.

The difference between a pacemaker and a clock

You may already be aware that I am quite obsessed with the concept of circadian organization, i.e., the idea that there is not just one brain pacemaker that times everything, buit an entire circadian system – an organ system on par with all the others you learn about in school (circulatory, endocrine, reproductive, immune, etc.) – in which every cell in the body is a part of it, with different cells and tissues playing different roles. Some tissues are pacemakers (The Master Clocks), while others are peripheral clocks (The Slave Clocks). The criteria for something to be considered a pacemakers are: a) rhytmicity in vivo and in vitro, b) loss of overt rhythmicity of the organism if the pacemaker is lesioned, and c) transplanation of the tissue carries period and phase of the transplant and forces it onto the host.

For a long time, it was thought that the circadian system in mammals was much simpler than that in other vertebrates – the SCN (suprachiasmatic nucleus) being the only pacemaker. The work on non-mammalian vertebrates led to discoveries of complex systems composed of multiple, mutually-synchronized neural pacemakers, as well as myriads of peripheral clocks in all the other cells in the body. Those findings prompted the mammalian researchers to start looking for complexity in the mammalian systems as well – finding semi-autonomous pacemaking tissues both within the SCN and outside of it.

One of the most saught-after pacemakers outside of the SCN is a food-entrainable clock which is normally in synchrony with the SCN but can be forced to desynchronize by unusual feeding schedules and is capable of functioning as a pacemaker in the complete absence of the SCN. Some other candidate organs, while shown to contain clocks, have not been shown to be pacemakers so far.

So, over the past several years, our understanding of circadian systems has changed dramatically and this may be a good time for you to go and re-read this post and particularly take a look at the last figure at the very end – that is how we think about the problem today: multiple pacemakers interacting with each other; many peripheral clocks interacting with each other, being entrained by the pacemakers and feeding back onto pacemakers; and multiple sensory inputs from the environment synchronizing the pacamakers, as well as themselves being regulated by pacemakers (the circadian regulation of sensory sensitivity, as discussed above).

In early studies of the peripheral clocks, cells from the tissues like liver, muscle, lung, heart or skin usually showed only a couple of oscillations when placed in a dish. At the same time, pacemaking tissues like mammalian SCN, chicken pineal, or retina of various vertebrates, tended to cycle in vitro indefinitely – as long as one could keep them alive in a dish. So, the “official” distinction between a pacemaker and a clock was the ability to sustain oscillaitons in a dish.

However, improvements in culturing methods have since shown that all the cells – from fibroblasts to muscle cells to liver cells – are capable of indefinite cycling in vitro. The distinction between pacemakers and peripheral clocks is getting fuzzy. Thus, a new “official” definition is needed. For a clock to be a pacemaker, it needs to be able to force the phase and period onto another tissue, and it needs to be entrainable by some type of input from the outside environment. Thus, a pacemaker is directly entrainable by light (or food-schedules, etc.) and it “drives” the phase and period of other clocks. Peripheral clocks are entrained only by internal clues – neural impulses or timed release of hormones – and only drive events within the same tissue. Perhaps a new nomenclature is needed. How about: global vs. local pacemakers?

Another problem with the past studies is the almost ubiqutous choice of only very few outputs as measures of the function of the underlying pacemaker. The most often used outputs are locomotor activity (e.g., wheel-running in rodents or perch-hopping in songbirds), sleep-wake cycle, body temperature, and blood levels of melatonin. Apparently, all those outputs are strongly driven by the SCN, which left other possible pacemakers undetectable by these methods. The other pacemakers presumably drive other functions: an intestinal pacemaker, if it exists, would regulate the rhythms of gut peristalsis and release of digestive enzymes and be entrained by feeding schedules; an ovarian pacemaker, if it exists, would, likewise control the daily rhythms of hormone release and ovulation, entrained, perhaps indirectly, by the behavior of conspecifics.

Olfactory Bulb. Olfactory Bulb?!

Let us now take a look at the new paper by Erik Herzog. If the name seems familiar, it is because you have encountered it here before. Here is an excellent press release well worth your time to read (also here).

Erik is currently working on the circadian organization in mammals, both within and outside the SCN, but he did not always do that. His PhD work was, under Bob Barlow, on the circadian modulation of visual sensitivity in the horseshoe crab Limulus. While there, Erik did one of the coolest experiments ever. He caught some crabs and inserted multiple electrodes into the cells of Limulus eyes. The electrodes measured the electrical activity of the eye cells and – via radiotelemetry – sent the data to a computer. At the same time, Erik attached miniature cameras on the backs of the crabs. Then, he released the animals back into the murky waters of their natural environment and continuosly monitored both what could be seen (by cameras) and what the crabs could see (by analysis of electrophysiological data). What he found out is exactly what I mentioned in the beginning – a daily switch between power of detection and power of resolution.

So yes, Erik is a cool guy (he did some research at the ocean floor as well, going down on the Alvin submersible – I am so jealous!). But let’s get back to the newest paper. It does something similar to the Limulus work, but in mice instead. Erik, his postdoc Daniel Granados-Fuentes and an undegraduate student Alan Tseng, tested the hypothesis that the olfactory bulb is a circadian pacemaker, separate from the SCN.

Olfactory bulb is a big part of the brain in most vertebrates (our large cortex makes everything else look small in comparison, but even our olfactory bulbs are still pretty big parts of the brain). Odorant molecules picked up by the odorant receptors in the olfactory epithelium lining the insides of the nose trigger electrical activity of olfactory neurons. These neurons send their processes to the olfactory bulb for primary processing of the olfactory informaton. Olfactory bulbs then project to a bunch of other brain areas involved in many functions, including memory, emotions, stress response and feeding behavior, as well as to the Piriform Cortex for second-level processing of the olfactory information – the place which tells your cortex that rose by any other name would still smell like a rose.

It’s been known for a while now that there is a circadian modulation in olfactory acuity in a number of mammals, as well as Drosophila. In another rodent, Octogon degus, it has been shown that odors affect circadian rhythms of locomotor activity. So, is the olfactory bulb itself a circadian pacemaker driving the rhythm of sensory sensitivity and, under the influence of odors, affecting the other pacemaker in the SCN?

First, Erik and Co. measured c-Fos induction in the olfactory bulb and the piriform cortex upon exposure to an odor at different times of day. This gene, c-Fos, is a so-called “early-immediate” gene and is often used as an indicator of increased activity of a cell when it is not known exactly which genes may be induced by the stimulus – the c-Fos is activating many genes, so it really does not matter which particular one is important for the response. It is a rather crude measure of an On-Off state of the cell. Both the olfactory bulb and the piriform cortex showed a circadian rhythm in c-Fos induction in response to the odorant, in agreement with the notion that in mice, nocturnal animals, the power of detection should be high during the day and the power of resolution during the night.

In a previous study, they also showed that canonical clock genes cycle in the olfactory bulb placed in a dish, both spontaneously and under the influence of a stimulus – temperature cycles. These two experiments together show that there is a clock in the olfactory bulb but not a pacemaker.

In the next experiment, they show that the rhythm in the piriform cortex is driven by the olfactory bulb. Moreover, the connection is strictly lateralized – when the left olfactory bulb is removed there is no rhythm in the left piriform cortex, while the rhythm in the right piriform cortex remains unchanged. Thus, piriform cortex is NOT a circadian pacemaker.

Next, they removed the SCN. The rhythms, both spontaneous and smell-induced, remained normal. The period of the rhythm, just like with the olfactory bulb in a dish, is a little shorter than in the presence of the SCN, suggesting that there is an influence of the SCN on the clock in the olfactory bulb. But, its independence from the SCN, as seen in this experiment, strongly suggests that the olfactory bulb is not just a clock, but also a circadian pacemaker driving the rhythm of olfactory sensitivity in an independent fashion.

Finally, they removed the olfactory bulbs and monitored the locomotor activity driven by the SCN. Lo and behold, the period of the rhythm changed, as well as the rate of re-entrainment to shifted light-cycles. This tells us that the two sites – the olfactory bulb and the SCN both affect each other’s rhythm-generating properties. This means that olfactory bulb definitely IS a circadian pacemaker. It fulfills all the criteria I listed above: it is entrained by odors from the environment [Edit: not yet - this is under current investigation], it feeds back on the SCN, and it drives rhythms in other tissues (piriform cortex).

This is quite a tour-de-force. In a single paper, Erik’s lab identified a non-SCN circadian pacemaker (which is about 100 times larger than the SCN) and showed that it is indeed a pacemaker. Conrast that to the decades-long search for the food-entrainable pacemaker which is still not as certain in its findings. No need to mention that this finding will spur future search for other pacemakers and the study of circadian organization – how do all those multiple pacemakers and clocks interact with each other. After all, when they alll get out of sync, the state is called “jet-lag”, something that NIH may be interested in funding the research on.


  1. #1 Jonathan Vos Post
    July 18, 2007


    Is this the kind of breakthrough that gets one shortlisted for a Nobel Prize in Medicine and Physiology?

  2. #2 coturnix
    July 18, 2007

    I hope so! But this whole field is chockful of people who deserve it for various revolutionary discoveries. Whichever three people in the field eventually get it, it will not be fair to dozens of others.

  3. #3 Stagyar zil Doggo
    August 5, 2007

    Thank you for a well written and informative article. However, as an Engineer who last took biology waay back as a Sophmore in High School, I would appreciate some clarifications. You write –

    But the phrase “sensory sensitivity” is vague. What it means is that there is a daily switch in the power of detection versus power of resolution. In most senses, the two are incompatible. If the sensory organ is tuned to detect the lowest possible amount of energy, it is incapable of maximal resolution and vice versa.

    This sounds like you are talking about adjusting gain(s) – either of the sensor or any of the amplifiers downstream in the signal path. Setting any gain too high results in saturation of the corresponding sensor or amplifier, even for low values of the input signal. I’m guessing this is the situation you describe as having “high power of detection, but low power of resolution” – like when stepping into the midday sun after being in a dark-room. The contrary situation, would then be to dial all the gains back down just enough to avoid saturation at the anticipated strength of the input. You still keep them as high as possible to keep the signal to noise ratio high and thus get “maximum resolution, but low power of detection”.

    Is this indeed what you are referring to or am I taking things out of context? I mean, I think you are saying that some or all gains in all biological sensory circuits have a circadian periodicity, but I feel unsure because the terms used – detection v resolution – are somewhat unusual to me.

  4. #4 coturnix
    September 9, 2007

    Yes, this is it. And the control of the gain appears to be at the level of the sensor. Sorry – the terminology you were using was unusaul to me: different traditions use different words to say the same things, I guess.