There’s more to vision than meets the eye

If you live in Oxford and have an interest in how the brain works, then it is worth checking out Cortex Club. Every few weeks the society organises for an academic working in an area of neuroscience to come in and give a talk about their work. The talks are relatively informal and so provide a great opportunity for people to discuss recent developments with leading researchers. Any Oxford student or staff member can attend Cortex Club, and one of the aims of the society is to foster an interdisciplinary community of neuroscientists. The presentations for the rest of this term focus largely on movement and sensation; a timetable can be found here.

Recently at Cortex Club, Professor Russell Foster presented an engrossing talk about recent work – much of it coming out of his own lab – involving the discovery and characterisation of a novel kind of photoreceptor in the eye. I was so fascinated by the discovery of this new kind of photoreceptor and the implications of its existence that I wanted to share the story here.

Photoreceptors are the cells in the eye which convert photons of light into electrical impulses, the “language” of the brain. Traditionally, it has been thought that we have two types of photoreceptors, rods and cones, with slightly different functions (e.g. night-time vs day-time vision; achromatic vs colour vision), and which, together, seem to fulfil all the requirements of vision. After converting photons of light to electrical impulses, these cells communicate with other retinal cells (horizontal cells, amacrine cells, bipolar cells, and retinal ganglion cells) which combine information from multiple rods and cones before transmitting that information to the brain.

In a paper published in 1991, however, Professor Foster and his colleagues challenged the notion that rods and cones constitute all the photoreceptors in the eye, by showing that light can affect certain behaviours even in the absence of these cells. Light cues help to control animals’ circadian rhythms, influencing hormonal fluctuations, body temperature, sleep/wake behaviour and so on across the course of the day. The researchers specifically examined whether mice with a variant of a gene associated with degeneration of the classical photoreceptors – and which therefore lacked vision – were still able to synchronise their behaviour to environmental light cues.

Foster and colleagues found that despite lacking rods and cones, these mice started moving around at the beginning of the dark period of a light-dark cycle, just like mice with intact rods and cones (mice are, of course, nocturnal). That is, they seemed to synchronise their behaviour with the environmental light, despite conventional wisdom suggesting that they possessed no cells able to respond to this light.

The mice were subsequently kept in darkness, and showed the typical sleep-wake cycles of behaviour that occur in the absence of light cues (these are called “freerunning” cycles, and in mice actually last a little bit less than 24 hours). But when the animals were exposed to a 15 min pulse of light a few hours after the start of their awake, moving period, they showed what is known as a “phase shift”: the following day the time at which the animals woke up and started moving around was delayed. Again, the animals were showing a behaviour that was a direct consequence of being exposed to light.

Other studies subsequently found similar results, and used immunostaining methods to confirm that there were no remaining rod or cone photoreceptors that could explain the behavioural responses to light. These results therefore all pointed to the existence of a third kind of cell acting as a photoreceptor.

The question remained as to where these novel photoreceptors could be. Because the cells apparently played a vital role in regulating circadian behaviour, it seemed likely that they would project to the suprachiasmatic nucleus (SCN) of the hypothalamus in the brain. The SCN is considered the brain’s “pacemaker”, involved in the orchestration of circadian behaviour, and importantly, its activity is modulated by light. Another group of researchers therefore decided to identify the new kind of photoreceptor by tracing fibres projecting to the SCN back to their origin. The investigators injected a fluorescent retrograde tracer into the SCN of mice. When the retinas were removed and examined, a small proportion of retinal ganglion cells (RGCs) were found to be labelled. Moreover, these cells showed an electrical depolarisation in response to light, even when detached from the rest of the retina. Thus it seemed that a subtype of RGCs – now termed photosensitive RGCs (pRGCs) – were able to respond to light, and were apparently important in controlling circadian behaviour.

In his talk Professor Foster went into much more detail than I have here, and the situation is not as clear-cut as I have made out. For example, there are now known to be different subtypes of pRGCs, which may serve different functions and have distinct anatomical connections. The relative contribution of rods and cones and pRGCs to circadian regulation is also uncertain. Professor Foster’s group is continuing to conduct important research to further our understanding of these cells and their exact role.

However, as someone who is very much interested in the application of neuroscientific research to humans, I was drawn towards the clinical implications of the discovery. In particular, the discovery of this third kind of photoreceptor requires some fundamental changes in the way in which we help people with visual disorders.

Perhaps the biggest implication is that we can no longer necessarily think of a blind person’s eyes as functionally useless. As Professor Foster pointed out, under the advice of an opthamologist, a blind person will sometimes have his or her eyes removed if, for example, they are prone to infection. If they can’t see anyway, then removing the eyes should do no harm, right? But in fact, if that person’s blindness has spared their retinal ganglion cells, then in removing their eye, an important regulator of circadian rhythms is being taken away. The person could end up suffering from sleep disruptions for the rest of their life. For the same reason, constantly wearing sunglasses could also prevent light from getting to the pRGCs of blind people and cause disruption to sleep.

Something Professor Foster didn’t really talk about is whether there could be a sort of opposite form of “blindness”, in which vision is spared, but in which there are problems with circadian regulation due to damage to the pRGCs. The possibility of such a form of blindness has been demonstrated in the animal literature: Göz and colleagues selectively destroyed cells that expressed melanopsin, a protein involved in phototransduction exclusively in the pRGCs, in mice. These animals displayed no deficits in measures of visual perception; however, they had difficulty synchronising their behaviour with a light-dark cycle.

Could such a form of blindness exist in humans? So far, the evidence seems pretty scarce. One possibility put forward is that seasonal affective disorder (SAD), a disorder closely associated with sleep disturbances and often treated with light therapy, could involve dysfunction of the pRGCs. In a recent study Roecklein and colleagues found that 5% of SAD patients, but no controls, had a mutation within the gene coding for melanopsin – that protein so vital to the pRGC response.  Five percent is a pretty small number, and cause-and-effect cannot really be established here. But at the very least, the study demonstrates that abnormalities in the visual system could underlie circadian disorders, even in the absence of any deficits in vision.

One of the things I enjoyed most about Professor Foster’s talk is that it provided a great example of how basic research can yield important clinical knowledge. It is clear that many patients will benefit from the work that he and others have performed over the last couple of decades.