Life is like an opera, which involves a number of soloists, an often large chorus and many musicians playing a variety of instruments. To produce something worth listening to they all need to produce the correct notes, at an appropriate volume, in the right order, at the appropriate time. If everyone did what they wanted, whenever they felt like it, the result would be an uncoordinated cacophony. A conductor is needed to bring order to this potential chaos.
The many complex physiological and biochemical process responsible for an organism’s life must be similarly co-ordinated. You wouldn’t want to eat, sleep and urinate at the same time after all. In all organisms the conductor of life’s complex orchestra is a biological clock that ensures all the body’s processes happen in the right order and at an appropriate time across the 24-hour day.
The force that has shaped much of Earth’s biology is the daily transition between light and darkness as the Earth spins on its axis and changes its position relative to the sun. Most aspects of an organism’s physiology and behaviour are tuned to the daily light/dark cycle. This is as true for humans as it is for birds, insects, plants and bacteria. Thus, at dusk, oceanic creatures ascend from the deep waters, where they hide during the day to escape visual predators, to feed in the productive surface waters under cover of the night. Many flowers close their leaves and petals at night (perhaps to minimise predation and water loss) and open them at dawn to exploit the sun’s rays for photosynthesis.
The human body clock
Humans are similarly slaves to the rising and setting of the sun, most clearly exemplified by our sleep/wake cycle. Other aspects of our metabolism change rhythmically in tune with the light/dark cycle too; body temperature, blood pressure, alertness, cognitive function and the levels of many hormones (such as cortisol) increase prior to dawn in anticipation of daytime activity and then decline in the evening in preparation for sleep. Hormones, such as pineal melatonin and growth hormone (from the pituitary) are normally only released at night. Even our libido waxes and wanes across the 24-hour day, with a notable increase in males associated with the morning rise in testosterone. Many aspects of visual physiology and function also change regularly during a 24-hour period: photoreceptor disk shedding, ocular melatonin release, dopamine synthesis, electroretinogram amplitude, extracellular pH, visual sensitivity and, well known to readers, intraocular pressure all vary with time of day.
Such 24-hour cycles in physiology and behaviour are called ‘circadian rhythms’, from the Latin ‘circa’ meaning about, and ‘diem’ meaning day. Under constant conditions they persist for multiple cycles, but their period can be either a little longer or shorter than 24 hours. So in constant darkness the rhythm set by the clock has a period of ‘about a day’, but not exactly 24 hours. The endogenous nature of circadian rhythms was first shown in plants in 1792 when the French astronomer Jean Jaques d’Ortous de Marian put a mimosa plant in a cupboard and saw that it opened its leaves during the day, even though there was no light in the cupboard. Of course, we have circadian rhythms too. Jurgen Aschoff showed this when he isolated student volunteers in an underground ‘bunker’ out of contact with each other and the outside world and allowed them to sleep and eat when they wanted. They adopted a daily rhythm that had a period of longer than 24 hours and all were soon living a life completely out of phase with those in the real world. Therefore, our clock, like all circadian clocks, needs daily resynchronising (known as entraining) with the Earth’s light/dark cycle, otherwise bodily functions drift out of phase with the environment.
Another reason that the body clock needs adjusting is because, apart from at the equator, day length is not constant throughout the year. Because the axis of the Earth’s rotation is tilted, the further you are away from the equator the greater the difference in day length between the winter and summer. Therefore, a clock with a constant period, which for example signalled a constant 12 hours light and 12 hours dark, would be of little use as it would only be correct for a few days every year. There has to be a system for adjusting the internal clock to keep step with the ever changing light/dark cycle.
The nature of the clock
The mismatch between circadian rhythms and the external day and its resetting by light exposure with which we are most familiar is the jet lag we experience after flying across multiple time zones. Initially our internal circadian system causes us to sleep and wake, feel hungry, release hormones, visit the lavatory at inappropriate times – a biological ‘cacophony’. But after a few days our biological clock aligns to the new light/dark cycle. How the clocks are synchronised with ambient illumination will be discussed later.
In mammals, the ‘master body clock’ is located within a small anterior hypothalamic structure on top of the optic chiasma – the suprachiasmatic nucleus (SCN). Although it is the main body clock, the SCN is not the only one. Most, if not all, cells of the body have the capability to generate autonomous 24-hour rhythms of activity. Thus the mammalian retina has at least one, and probably more, circadian clocks controlling ocular function. The daily rhythms of such ‘peripheral clocks’ are co-ordinated by the SCN. As a result, the ‘day within’ arises from a complex network of billions of coupled circadian oscillators.
The advantage of such endogenous timers is obvious. As every schoolchild in the 17th century knew, ‘the early bird catcheth the worm’. Another way of looking at this would be ‘the late worm gets eaten’. Either way, the adage betrays an understanding of the value of circadian clocks even 400 years ago. It clearly benefits a hungry bird to be ready to hunt before it gets light and for the worm to withdraw underground before dawn breaks. It pays to prepare for predictable events.
How such clocks work is beyond the scope of this article. Essentially, they are transcription/translation feedback loops, and the basic building blocks have been conserved across all animal life. In simple terms, about 12 key ‘clock genes’ are activated within the cell’s nucleus and their messenger RNA (mRNA) strands transcribed. These mRNAs then get translated into proteins in the cytoplasm of the cell where they assemble. The resulting protein complex moves into the nucleus and inhibits the transcriptional drive of the clock genes. Thus no more mRNA is produced from the genes and the clock proteins slowly degrade. With the loss of clock proteins, the genes are released from inhibition, and once more make mRNA which is translated into protein. The rate of transcription, translation, protein assembly, protein degradation, and so on, produces an approximately 24-hour cycle of cellular activity, and it is this molecular oscillation that needs entraining to the light/dark cycle to ensure internal and external time coincide.
Ocular role
So, to return to the question of how these clocks are synchronised to the daily light/dark cycle. In all non-mammalian vertebrates (fish, birds, reptiles and amphibians), photoreceptors within the pineal complex, and even deeper within the brain, detect light for the regulation of the circadian system. In mammals, however, it is the retina which sends projections to the SCN and allows the internal clock to be synchronised to the astronomical day. It is therefore clear that the eye is not only important for what we normally think of as vision, which involves forming a detailed image of the world around us. It is also involved in other, non-image forming processes such as the synchronising of the body’s rhythms, for which detailed information about the image is not required. What is important is a measurement of the absolute light level.
It has long been known that rods and cones underlie the image-forming (IF) functions of the eye. Until recently, however, it was unclear which photoreceptor(s) were responsible for the eye’s non-image forming (NIF) responses and provided input to the SCN. Was it rods or cones or perhaps both? The answer surprised everyone – it was neither. Circadian rhythms in mice, that had been genetically altered to have no functional rods and cones, were unaffected, with activity and hormone cycles still perfectly aligned to the periods of light and darkness. The only possible conclusion was that there was another non-rod, non-cone photoreceptor supplying information about ambient light levels to the SCN. We now know that this novel photoreceptor in mammals is a subset of directly photosensitive retinal ganglion cells (pRCGs) containing the photopigment melanopsin. These cells, which make up about 1 per cent of all ganglion cells in the human retina, respond to light without any inputs from the rods or cones.
A third retinal photoreceptor
Human rods contain a visual pigment absorbing maximally at 498nm and we have three cone types maximally sensitive at 420, 534 and 563nm. However, if you look around the animal kingdom, you will find a large variety of other rod and cone pigments ranging in maximal sensitivity from 360 to 630nm. Each species has a set of visual pigments most suited to its specific visual needs. The sensitivity of melanopsin, however, is remarkably consistent, being maximally sensitive in the ‘blue’ part of the spectrum at around 480nm in all species. There has been much speculation why this should be so, and the favoured suggestion is that these receptors are tuned to the blue-enriched wavelengths of light across the dome of the sky at dawn and dusk. At twilight, the immediate horizon is red/orange; but look up. The sky is blue, peaking at around 480nm.
The announcement in 1999 of the existence of a third photoreceptor type in the mammalian retina was met with widespread scepticism. After a talk by one of the authors in what shall remain an unnamed optometry department, the head of the institution said ‘I do not believe you!’ When asked why, he said ‘I do not see the point!’ So, what is the point of this third, novel, class of retinal photoreceptor? Why not use the traditional photoreceptors to inform the SCN about light levels?
The simple answer is that the properties of the photoreceptors required for the two functions of the eye, IF and NIF vision, are largely incompatible. To detect detailed information for IF vision, a photoreceptor must respond quickly and briefly. It must therefore, for example, adapt rapidly, while one that measures the amount of light in the environment for NIF vision needs to keep responding for much longer. Thus, while the responses of cones, and to a lesser extent, rods diminish after a few seconds of light exposure, pRGCs cells continue to respond for many minutes; they do not adapt. Thus, the existence of separate photoreceptor classes for IF and NIF vision is, in fact, not surprising.
There are at least five morphological types of pRCGs that project to different parts of the brain. The precise details of where they project, how these projections overlap, and how their responses to light differ is still being unravelled. However, the expectation is that those pRGCs that project to the SCN will encode NIF information in a slightly different manner compared to those pRGCs that project to the olivary pretectal nuclei and control the pupil responses, or those that project to the ventro-lateral preoptic nuclei and regulate sleep and alertness. It is also likely that pRGCs have a role to play in conscious brightness perception through projections to the dorsal lateral geniculate nuclei. But again, details are lacking.
Understanding that many functions are controlled by pRGCs has profound clinical implications. For example, historically following degeneration of the traditional photoreceptors, the eyes of functionally blind individuals were sometimes removed as they were considered of no value. As a result, patient’s circadian rhythms were severely disrupted and their quality of life impaired. However, as long as inner retinal function is maintained, even if the rods and cones have degenerated, it is sensible to retain the eye and allow the NIF processes controlled by the pRGCs to function normally.
This is not to suggest that rods and cones have no role to play in NIF behaviours. All three photoreceptor types (rods, cones and pRGCs) are involved in different aspects of most of these responses. Take the pupil response in mice as an example. Ablating any one of the three photoreceptor types leaves the pupil response largely intact. Removing all three results in a static pupil. As rods and cones are both more sensitive and respond more rapidly than pRGCs they are responsible for the rapid initial phase of pupil constriction. The pRGCs, although they are slower and less sensitive than the traditional retinal photoreceptors, have the advantage that they do not adapt. They therefore take over from the rods and cones at higher light levels and mediate sustained pupillary constriction.
Health implications
As circadian clocks influence most bodily processes, not surprisingly they have also been implicated in many disease processes (Table 1).
For example, the degeneration of the SCN that occurs in Alzheimer’s disease (AD) causes sleep disturbances in patients which also disrupt the sleeping patterns of their carers. Such disrupted sleep in carers is a key reason why AD sufferers enter residential care earlier than they would otherwise need to, costing the health service millions of pounds every year. A disrupted body clock and sleep/wake cycle is also a characteristic feature of mental illness, and the partial stabilisation of sleep in patients with schizophrenia has been shown to improve symptoms. The retinal circadian clock has also been implicated in several ocular pathological processes. These include the susceptibility of photoreceptors to light damage, retinal degeneration and the development of myopia.
The introduction of cheap electricity and artificial light in the late 19th century and the subsequent restructuring of working hours has progressively distanced us from the solar 24-hour cycle of light and dark. The consequence has been a disruption of the circadian and sleep systems, resulting in performance deficits including increased errors, reduced vigilance, memory loss, reduced mental and physical reaction times and reduced motivation. Sleep deprivation and disruption is also associated with a range of metabolic abnormalities, including the glucose/insulin axis. For example, sleep disrupted individuals take longer to regulate blood glucose levels and insulin can fall to levels usually seen in the early stages of diabetes – abnormalities that can be reversed by the resumption of normal sleep patterns. Such results have suggested that long-term sleep and circadian rhythm disruption might contribute to chronic conditions such as diabetes, obesity, and hypertension. Furthermore, obesity is strongly correlated with sleep apnoea and this often contributes to further sleep disturbance. Under these circumstances a dangerous positive feedback loop of obesity and sleep disturbance can result.
Sleep pattern regulation
Sleep loss and circadian rhythm disruption are obvious in nightshift workers. More than 20 per cent of the population of employment age work at least some of the time outside the 7am to 7pm day. Josephine Arendt at the University of Surrey makes the following point: ‘Because of their rapidly changing and conflicting light-dark exposure and activity-rest behaviour, shift workers can have symptoms similar to those of jet lag. Although travellers normally adapt to the new time zone, shift-workers usually live out of phase with local time cues.’ Even after 20 years of night-shift work individuals will not normally shift their circadian rhythms in response to the demands of working at night. Despite the great variety and complexity of ‘shift systems’, none have been able to alleviate fully the circadian problems associated with shift work. Metabolism, along with alertness and performance, are still high during the day when the nightshift worker is trying to sleep and low at night when the individual is trying to work. A misaligned physiology (along with disrupted sleep patterns) in nightshift workers has been associated with increased cardiovascular mortality, an eightfold higher incidence of peptic ulcers, and a higher risk of some forms of cancer. Other problems include a greater risk of accidents, chronic fatigue, excessive sleepiness, difficulty sleeping and higher rates of substance abuse and depression. Nightshift workers are also much more likely to view their jobs as extremely stressful.
So why don’t shift-workers shift their clocks? After all, and as we discussed above, if we travel across multiple time zones, we do eventually recover from jet lag and adjust our clocks to the local time zone. The answer seems to be that the pRGCs that entrain the circadian system are fairly insensitive to light. The clock responds to bright natural sunlight in preference to the dim artificial lights commonly found in the workplace. Shortly after dawn, natural light is some 50 times brighter than normal office lighting (300-500 lux), and at noon natural light can be 500-1,000 times brighter, even in northern Europe. Thus, exposure to strong natural light on the journey to and from work, combined with low levels of light in the workplace, locks the nightshift worker onto local time. In this way biological and social time are persistently misaligned.
Influencing the clock
In view of the blue light sensitivity of melanopsin, the prediction was that light of around 480nm would be most the most effective for shifting circadian rhythms and alerting the arousal systems. In all studies undertaken to date, this has been shown to be the case. Blue light exposure is indeed the most effective at adjusting the circadian clock, reducing sleepiness, improving reaction times and reducing the symptoms of seasonal affective disorder (SAD). For these reasons there has been some concern that ‘yellow’ blue-light blocking intraocular lenses used in cataract surgery may filter out sufficient blue light to impair pRGC-mediated biological responses. A recent study addressed this issue and demonstrated that most commercially available blue blocking intraocular lenses filter out no more blue light (at around 480nm) than the normal crystalline lens of a 50-year-old, and an examination of sleep quality in hundreds of patients showed that there were no differences in the timing or quality of sleep in individuals implanted with either blue-blocking (yellow) or UV-blocking (clear) intraocular lenses.
It is worth reflecting that we are a species that has evolved under bright light conditions. Even on an overcast day in Europe, natural light is around 10,000 lux, and may be as high as 100,000 lux on bright sunny days. Yet we live inside homes, offices, factories, schools and hospitals that are often isolated from natural light, and where artificial light is often around 200 lux and seldom exceeds 400 to 500 lux. While this level of light allows IF responses to light, the pRGCs are much less sensitive to the light than the traditional photoreceptors, so that short light exposure that is easily detected by the rods and cones is not recognised by the pRGCs. Robust setting of the clock requires 1,000 to 2,000 lux for 30-60 minutes, and the morning is the best time to get this light. However, dim light can have an effect if it is delivered over long periods of time, and particularly if the light contains wavelengths of around 480nm. Thus, the relatively dim indoor room light from bedside lamps, and blue-enriched computer screens or tablets could potentially have measurable effects on the clock and alertness systems, and there is an emerging consensus that there is a small but significant effect of such light sources. For example, in one recent study, the continuous use of a light-emitting eReader for four hours prior to bedtime delayed the onset of sleep by a small but significant amount (around 10 minutes) compared to four hours continuous reading of a paper book, with room light reflected from the page. Clearly this is not a big effect and could easily be swamped by multiple other environmental variables, not least the effect of caffeine in coffee or tea prior to bed.
Conclusion
The hullaballoo generated over blue-blocking contact lenses or the impact of light from computer screens should not detract from the bigger picture. Until the late 1990s, it seemed inconceivable to most vision scientists and ophthalmologists that there could be an unrecognised class of photoreceptor within the vertebrate eye. After all, the eye was the best understood part of the central nervous system. One hundred and fifty years of research had explained how we see. Photons are detected by the rods and cones and their graded potentials are assembled into an image by inner retinal neurones, followed by advanced visual processing in the brain. This representation of the eye left no room for an additional class of ocular photoreceptor. However, we now know that the rods and cones are not the only photoreceptors. The discovery that the eye contains another class of photoreceptor based on a small number of photosensitive retinal ganglion cells (pRGCs) that detect environmental brightness and regulate a wide range of physiology and behaviour (including the regulation of 24-hour body clocks, sleep, alertness, mood and even pupil size) has transformed our understanding of how the eye detects light. But this discovery is not just a new and exciting area of sensory biology. There are important clinical implications, not least on the classification of blindness. Ophthalmologists now appreciate that complete eye loss plunges individuals into a world devoid of both vision and a proper sense of time, and the assessment, treatment and care of individuals with eye disease must take these dual functions of the eye into account.
Model answers
Correct answer is in bold italics
1 Which of the following has not been reported as a possible association with disrupted circadian rhythms?
A Obesity
B Hyperactivity
C Type 2 diabetes
D Reduced immunity
2 Which of the following is the most likely circadian periodicity for a human in the absence of external stimuli?
A 12 hours
B Less than 24 hours
C 24 hours
D Greater than 24 hours
3 Where is the central coordinating site for circadian mechanisms located in a human?
A Visual cortex
B Ganglion cell layer of the retina
C Just superior to the optic chiasma
D Pituitary gland
4 What percentage of ganglion cells are photosensitive?
A 1 per cent
B 5 per cent
C 25 per cent
D All have this property
5 What is the peak sensitivity for melanopsin?
A 420nm
B 480nm
C 534nm
D 563nm
6 Which of the following best represents light levels in bright sunshine?
A 100 lux
B 1000lux
C 10000lux
D 100000lux
Further reading
- Alexander I, Cuthbertson FM, Ratnarajan G, et al. Impact of cataract surgery on sleep in patients receiving either UV-blocking or blue-filtering intraocular lens implants. Invest Ophthal & Vis Sci, 2014; 55, 4999-5004.
- Chang A M, Aeschbach D, Duffy JF, Czeisler CA. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS, 2015; 112, 1232-1237.
- Foster RG, Kreitzman L. Rhythms of Life. London: Profile Books, 2004.
- Foster RG, Kreitzman L. The Rhythms of Life – What your body clock means to you. Experimental Physiology, 2014; 99, 599–606.
- Hankins MW, Peirson SN, Foster RG. Melanopsin: an exciting photopigment. Trends in Neurosciences, 2008; 31, 27-36.
- Ruan G X, Zhang D Q, Zhou T, et al Circadian organization of the mammalian retina. PNAS, 2006; 103, 9703-9708.
Ron Douglas is Professor of Visual Science at the Department of Optometry & Visual Science, City University London. Russell Foster is Professor of Circadian Neuroscience and Head of the Department of Ophthalmology at the University of Oxford