One of the assumptions in the study of circadian organization is that, at the level of molecules and cells, all vertebrate (and perhaps all animal) clocks work in roughly the same way. The diversity of circadian properties is understood to be a higher-level property of interacting multicellular and multi-organ circadian systems: how the clocks receive environmental information, how the multiple pacemakers communicate and synchronize with each other, how they convey the temporal information to the peripheral clocks in all the other cells in the body, and how peripheral clocks generate observable rhythms in biochemistry, physiology and behavior.
Most of the studies have always been performed in mammals, but the uniformity and simplicity of mammalian circadian systems makes a number of alternative hypotheses impossible to differentiate from each other and picking to study yet another species of rodent is unlikely to answer these questions. In mammals, there is only one way the environmental light enters the system - via eyes. In mammals, there is only one pacemaker - the suprachiasmatic nuclues of the hypothalamus (SCN) and all the other organs are peripheral clocks, including the pineal organ whih secretes melatonin with a daily (really nightly) rhythm. All the studies of the complexity of the mammalian system have to focus on subpopulations of pacemaker neurons within the tiny SCN and neccessitates difficult and expensive cellular and molecular techniques to resolve.
Studies in some non-mammalian vertebrates provided the answers to some questions unanswerable in mammals. Most of the early non-mammalian models, e.g., the house sparrow and some lizards, just happened to be species in which the major pacemaker is the pineal organ. Without the pineal, no rhythms are observed. Again, a very simple system, but sufficiently different from mammals to be able to resolve some old questions of circadian organization. Furthermore, and that is where focus of non-mammalian research lied in the 1980s, all vertebrates except mammals can perceive environmental light via three routes: the eyes, the pineal organ and the deep-brain photoreceptors.
A series of papers by my advisor and the students and postdocs preceding me in the lab established that Quail Is Different: the pineal appeared to play no greater role than as a peripheral clock and the major pacemakers are in the retina of the eye. Thus, this was a promising new model for the study of circadian organization, especially the question of coupling between paired clocks, something that we addressed several years later.
So, it was not surprising that, at the time I joined the lab, the mood was quite eye-centric. Pineal was a nobody and Eye was the God. It was expected that eyes would have a major role in every aspect of circadian rhythmicity in this species, including photoperiodic time measurement and the control of circadian timing of ovulation and oviposition in the female quail. The study I was given to do right off the bat was designed to test exactly those notions.
Female Japanese quail is reproductively stimulated by photoperiod lengths greater than about 12.5 hours. When sexually mature, the quail lays approximately one egg per day. The first egg is laid in the early afternoon (it was ovulated about 25-27 hours before). The next day, the egg is laid a little bit later in the day, the third day even later, and so on, until the timing of egg-lay encroaches into the night at which point the laying stops, takes a break for a day or two and starts all over again with a new clutch beginning in the early afternoon.
This pattern of egg-laying is consistent with mathematical models of two coupled physical oscillators: in particular phase-relationships ovulation can occur (the "permissive phase") while in other phase-relationships it cannot. In short photoperiods, the permissive phase does not exist, thus no ovulation occurs. In the somewhat longer photoperiods, the permissive phase is narrow and the oscillator driving ovulation is forced into "relative coordination" or even entrainment by the oscillator entrained to the light-dark cycle. As the photoperiod gets even longer in summer, the permissive phase gets broader and the ovulatory rhythm is capable of freerunning through the permissive phase with its own inherent period of around 27 hours.
The eye-centered thinking at the time was that one of the oscillators was the clock in the eye driving the SCN, while the other, the one driving ovulation, was located either in the ovary or in the SCN. If that was true, manipulation of the ocular pacemakers would in some way modify or disrupt both the photoperiodic response and the circadian timing of ovulation and oviposition. So, I manipulated the ocular pacemakers by either removing them (EX) or by preventing them from perceiving the light by placing black, opaque patches over the eyes (EP).
What happened? Nothing. Both in LD14:10 and in LD18:6 the patterns of egg-laying were identical between sighted and EX birds (black is body temperature higher than the daily mean, circles mark the tiem of egg-laying):
There was no change in egg-laying patterns in birds before and after eye-patching, either:
EX birds even had an unperturbed freerun of the ovulatory rhythm in constant light and constant darkness:
The only really novel thing I could get from the data was that light perceived by the eyes, but not light perceived by the extra-retinal photoreceptors, exerts a masking (i.e. direct) stimulatory effect on body temperature on top of its phase-entraining effect - see the difference between square-wave shapes in sighted and sinusoid shapes in blind (EX) and eye-patched (EP) birds:
Being young and new at the time, I was quite distraught to get negative results. But my advisor was quite excited about the data because of the way they changed the way we think about this system. I did not understand it at the time because, as soon as I started looking at the data I started thinking in the new way not realizing it was a new way. I was new and not yet familiar with all the literature. Only later I realized that my advisor had to change his mind because of these data (and one of the people who reviewed my paper, person I knew only from numerous papers I read, later got me a beer at a meeting for "such a good first paper").
So, what did this change? It dethroned the eye to some extent and made us focus on the hypothalamus. Up till then, we treated the SCN almost as if it was a peripheral clock itself. But it is a pacemaker in its own right. It can remain coherent and drive circadian rhythms for as long as 1-2 weeks in the absence of the eyes or light-dark input from the outside. This study showed that the SCN can remain coherent when exposed to light via extraretinal photoreceptors even when the reproductive system is trying to drive it with a longer period of 27 hours. Thus, it is more robust than we thought.
Also, it is not just the eyes entraining the SCN - the influence is bi-directional. When the eyes were covered with patches, the SCN was capable of entraining the eyes to the 24-hour cycle.
And, the two-oscillator interaction that determines the location and breadth of the "permissive phase" for ovulation is not between the eye and the SCN or the eye and the ovary, but between two clocks both residing in the hypothalamus. In the Discussion section we published a new model of the circadian system in quail:
The way we drew the system was somewhat conservative. We were still not entirely sure about the feedback from the SCN to the eyes so those arrows were omitted. Also, we depicted the two hypothalamic components as two cell-types residing within the same nucleus - there is nothing to say that these cannot be located in two distinct pairs of nuclei.
In short, instead of a unitary circadian system in which eyes drive the SCN which drives all the other rhythms and the ovary may feed back onto the eyes, we here have a two-component system which, for all practical purposes can be treated as two functionally (and perhaps anatomically) distinct circadian systems in quail.
One system, let's call it retinohypothalamic system (RHS), consists of pacemakers in the eyes and pacemakers in the SCN engaged in a feedback loop. They drive the melatonin rhythm in the pineal which feeds back onto the system and makes it more robust. This component/system is entrainable by light perceived both by the eyes and by the extraretinal photoreceptors in the deep brain and the pineal itself.
The other component/system, let's call it gonadohypothalamic system (GHS), consists of another set of cells in the hypothalamus which may or may not reside within the SCN in a feedback loop with the ovary. It is still not clear if the hypothalamus is the pacemaker and the ovary a peripheral clock, or if the ovary is the pacmaker and the hypothalamus a driven oscillator, or if both the hypothalamus and the ovary are pacemakers synchronized with each other, or if both sites are peripheral oscillators which, when working in concert, together act as a pacemaker. What is clear is that this system, the GHS, is not directly entrainable by light, but gets its temporal information from the RHS. If the RHS is disrupted (e.g., by exposure to constant light), the GHS freeruns with its own period of about 27 hours. If the RHS is entrained to the 24-hours cycle, the GHS freeruns through the permissive phase only. If the RHS is freerunning (with a period of around 22 hours), the
GHS also freeruns (with a period of around 27 hours) and as the two rhythms cross each other, a day of ovulation may get skipped (I would desribe this as a "hit-and-run permissive phase").
This work was about one half of my Master thesis and was published back in 2000.