About AE   About NHM   Contact Us   Terms of Use   Copyright Info   Privacy Policy   Advertising Policies   Site Map
Custom Search of AE Site
spacer spacer


Ivar Kljavin, Ph.D. and Andrew Cuthbertson, M.D.

A. Introduction

When you look at something, your "perception" of whatever you're looking at occurs in two stages. The first stage involves light entering the eye where it is converted into electrical information and the second stage involves further processing of the visual information in the brain. In very simple terms you can think of the eye as a ball with a small hole cut out on one side of it. In the hole is a lens (and cornea) which helps project an image of whatever you're looking at onto the back most inside portion of the ball. On this inside surface, a specialized light capturing organ called the neural retina processes the image and then transmits the acquired information through the optic nerve to higher centers in the brain.

The retina is a multilayered "sheet" containing an array of light sensitive neuronal cells called photoreceptor cells (there are two types, called rods and cones) which are connected to a complex array of four other neuronal cell types. The photoreceptor cells, which for the most part have a shape that looks like a pencil, can be considered biological transducers that convert absorbed light into electrochemical signals. They do this through a complex process involving light absorbing molecules localized within a specialized region of the photoreceptor cell called the outer segment, which by analogy to the pencil, would be the eraser. The outer segments are an elaborate folded membrane structure which look somewhat similar to a stack of discs neatly stuffed into a tube (or "Pogs" stuffed into a Pog holder- you'll understand that description if you have children 4-7 years of age). The outer segment discs are constantly being renewed; in rods the discs are formed at the base of the outer segment, about three each hour and migrate outward to the tip where they are shed away. Again by analogy, the point of the pencil would be the portion of the photoreceptor which is connected to the other neuronal cells in the retina. These other neuronal cells participate in the initial processing and transmission of the visual information from the photoreceptor cells to the brain.

If you look at the retina in cross section, you would see a series of cell layers, with the photoreceptor cells positioned at the back most portion of the retina. Photoreceptor outer segments point towards the back of the eye and are positioned the furthest away from the lens. In fact, light actually passes through all the other cell layers even before reaching the photoreceptor cell outer segments. The photoreceptor side of the retina is closely in contact to another cell layer called the retinal pigmented epithelium (abbreviated from this point on as the RPE). These cells control the exchange of materials between photoreceptors and their blood supply and participate in the replenishment of the light absorbing molecules that photoreceptors need throughout their life. This replenishment processes involves, at least in part, the RPE "eating" the shed outer segment discs. Additionally, the RPE absorbs excess light energy that is not trapped by the photoreceptor cells, thereby reducing scattered light and improving the clarity of images. This latter function is due to black melanin granules present within the RPE cytoplasm, thus the name "pigmented" epithelium.

B. Photoreceptor cells are easily injured

Diseases or injury to the retina that lead to blindness by directly affecting the retinal cells do so by injuring or ultimately killing them. Problems that cause blindness mostly disrupt either the input cells of the retina, the photoreceptor cells, or the output cells called ganglion cells which project axons all the way from the retina to the brain. For the rest of this discussion, we will focus only on the specialized light capturing organ, the retina and more specifically on the photoreceptor cells. A great deal of research in the retina is now focused on preventing blindness by developing therapies that have potential for reversing, or even at best slowing down the loss of photoreceptor cells as a result of disease. The retina has a long history of being studied in terms of it's anatomy, physiology and development owing to the fact that it is easy to isolate and examine as an intact piece of nerve tissue. Studying brain, or specific brain regions is a bit more challenging since many brain regions of interest are rather inaccessible. The most urgent questions asked in all the research going on in the retina is, " why do photoreceptors die?" and "what keeps photoreceptor cells alive in the first place?". Like elsewhere in the central nervous system, if photoreceptor cells or other retinal neurons become damaged they are incapable of being replaced by cellular division, which would be a nice mechanism of self replacement similar to new skin cells making more of themselves to grow over a wound. In any event, an answer to either of the above research questions has not been found. This most likely is due to the fact that injury to photoreceptor cells can come about in a number of different ways and their survival may depend on factors derived from the circulation or from factors derived from retinal cells themselves.

The photoreceptor cells, both rods and cones, seem particularly susceptible to injury as they are often the first cells to degenerate or suffer damage. Hereditary defects in specific photoreceptor genes, retinal detachment, circulatory disorders, overexposure to light, toxic effects to drugs and nutritional deficiencies are among the wide array of causes that can result in death of photoreceptor cells. For example, diseases affecting the retinal circulation such as what can occur in diabetes are among the major causes of vision impairment or blindness. In addition, hereditary diseases of the retina account for around 20 percent of all legal blindness in the United States. An example is a set of progressive hereditary defects in humans known as retinitis pigmentosa (RP) which is characterized by a progressive loss of peripheral vision and night blindness, due in large part, to the loss of rod photoreceptor cells. Total blindness is the usual outcome in more progressive stages of this disease. Macular degeneration , a complex group of disorders that affects the central predominately cone portion of the retina involved in our acute vision, is another major cause of blindness. Additionally, retinal detachment which involves the separation of the neural retina from the RPE inevitably leads to death of photoreceptor cells.

C. Death of photoreceptor cells may be influenced by something inside the cell or outside the cell.

The loss of photoreceptor cells can be due to a direct action on the cell itself. For example, in humans and several animals, mutations in several photoreceptor-cell specific genes have been shown to be associated with or cause retinal diseases like retinitis pigmintosa described above. These include mutations in the gene encoding the light absorbing molecule found within the outer segment called rhodopsin, the gene encoding a molecule called peripherin which helps maintain the normal structure of the outer segment and a gene encoding a molecule important for converting absorbed light into an electrochemical signal. Mice with spontaneous mutations in at least two of these genes have served as an important tool for confirming the ultimate fate of cells with these genetic defects. In addition, molecular biology has been used to produce transgenic mice with specific mutations in the rhodopsin gene, which also mimics the human defect seen in retinitis pigmintosa.

The major question concerning these genetic diseases is how do these mutations lead to the death of photoreceptor cells? With mutations in the rhodopsin gene, it has been suggested that the mutant rhodopsin molecules might be responsible for somehow activating the production and accumulation of toxic substances. Alternatively, other researchers have suggested that the mutant molecules may not be appropriately placed into the outer segment, thus accumulating in abnormal amounts in the photoreceptor cell body, which may also lead to cell death. Mutations in the gene encoding the molecule important for converting absorbed light into an electrochemical signal may likewise lead to toxic effects and finally, mutations in the peripherin gene may lead to structural abnormalities within the outer segment. At the present time we don't really know how the primary genetic defect in the photoreceptor cell leads to cell death. However, recent results from several laboratories have demonstrated that defects in the above genes may induce activation of other genes which actively kill the cell. This active process is called apoptosis, which is a form of programmed cell death where so-called suicide genes encode proteins which shrink and fragment the cell nucleus and DNA, effectively killing the cell. Although nothing is known about how apoptosis is triggered in retinal neurons, these data hold important implications for directing research efforts that attempt to interfere with the expression of the suicide genes.

The health of photoreceptor cells may also be influenced by factors from neighboring cells. The primary example of this relates to the close association between the RPE and photoreceptor cells. As already noted above, separation of the retina from the RPE leads to photoreceptor cell death and the degree of cell death is dependent on the duration of the separation. Detachments usually occur because of breaks in the neural retina which may appear in a variety of forms including tears and holes of various sizes. Retinal detachment may result from a number of conditions like blood vessel disease, diabetes or trauma. Additionally, problems with the RPE itself can lead to photoreceptor cell death as with the animal model called the RCS (Royal College of Surgeons) rat. In this animal, the RPE are unable to properly chew up the shed outer segment discs which results in the accumulation of cellular debris between the photoreceptor cells and the RPE. Photoreceptor cells usually die by two month of life in these animals.

Another cell in the retina not yet described in this discussion, called the Mčller cell, may also be important for maintaining the health of photoreceptor cells. Mčller cells are an example of a class of non-neuronal cell called glia, which traverse the entire retina in a radial direction from the ganglion cells to the photoreceptor cell bodies. In addition to providing structural support within the retina, Mčller cells regulate the control of ionic concentrations, degradation of neurotransmitters, removal of certain metabolites and may be important for the normal development of photoreceptor cell connections with other retinal neurons. Although a search for defects in Mčller cells has not specifically been done, any disease or injury affecting their normal function most likely would have a dramatic influence on the health of photoreceptor cells. Finally, the death of rod photoreceptor cells may influence the health of cones. One common feature recently observed in the mouse with mutations in rod-specific genes (i.e., rhodopsin) is that cones also eventually die. This finding is surprising since rhodopsin is found only in rods and the cones appear, at least initially, not to have any noticeable abnormalities. The reason for the loss of cones has not been determined, although it has been suggested that dying rods may release a toxin within the retina.

D. Reversing the loss of vision due to photoreceptor cell death has been attempted by two major approaches: transplantation strategies, or "Rescue" of dying photoreceptor cells.


Recall from the discussion that the RCS rat has problems with its RPE in that they can't chew up outer segment discs very efficiently. Researchers have used this animal as a model to demonstrate that transplanting healthy RPE to a degenerating RCS retina is possible and can rescue photoreceptor cells from dying. Just prior to photoreceptor cell death in this rat, healthy RPE were grafted into the back of the rats eye. After the transplant surgery, the grafted healthy RPE were able to chew up outer segment debris normally, in fact even outer segment debris that had accumulated during the course of the RCS rats early life were chewed up. The noticeable result even 6 months after the surgery was that photoreceptor cells in the area of the healthy RPE were alive, whereas the photoreceptors in areas of the unhealthy RPE were dying. Although no human disease has been found that is comparable to the RSC rat, there are other functional problems with human PE that may be relevant to 'macular' degeneration of photoreceptor cells. The macula is a tiny central part of the retina rich in cone photoreceptor cells, important for our central vision and, therefore, our ability to read and see fine detail. Therefore, RPE transplants in humans have recently been attempted, but thus far this strategy has shown limited success most likely because the photoreceptors in the patients had already been severely damaged.

The neural retina itself has also been transplanted to several regions in the brain and retina of rats and mice with photoreceptor degenerations. Why put a retina in the brain? Early transplant studies were done in this way in order to first demonstrate that retinal tissues can survive a transplant procedure and develop rather normally and independently as retinal cells even in a foreign environment. Retinal tissues have usually been placed in two regions of the retina, the vitreal chamber (the vitreous is a clear gelatinous substance that fills the eyeball between the retina and the lens) and between the photoreceptor cells and RPE, an area called the subretinal space. Some of the donor tissues have included isolated photoreceptor cell layers (as "tissue sheets") from rat and human or dissociated retinal cell suspensions derived from postnatal or embryonic tissues. The use of embryonic retina as donor tissue is of interest because it contains cells made up mostly of what we call undifferentiated progenitor cells. Early in retinal development, the retina is not multilayer as described in the introduction, instead it made up of actively dividing cells that look all alike, long and slender like match sticks all nicely aligned in a box. What is so special about these cells is that they are the precursors or predecessors to all the retinal cell types seen in later stages of development, including photoreceptor cells. Thus, the basic idea of using these cells as donor tissue is that the progenitor cells might turn into photoreceptors after being transplanted, thus replacing the cells lost as a result of disease. A lot of research is now underway focusing on how progenitor cells "decide" to make photoreceptors versus other retinal cell types. At the present time, we don't know how to control progenitor cells to a point where we could add a "factor" which instructs them to make only rods or cones. Is transplantation of healthy retinal tissue to a diseased eye a useful therapy for vision loss? It's too soon to tell, but all of the above studies have been important at very least in establishing the techniques and feasibility of grafting retinal tissues to a normal or damaged host retina. However, while transplanted tissues may survive in a new environment, whether or not successful functional integration of the transplanted cells with the host retina has taken place remains an open question. We can discuss why this might be a problem in the up coming discussion questions.

"Rescue" of dying photoreceptor cells

In addition to transplantation following loss of photoreceptor cells, there have been other strategies that focus more on "rescuing" or slowing the loss of visual cells. One such strategy might be to insert a normal gene into the photoreceptor cells carrying a mutation, this strategy is called gene therapy. Some recent studies in the retina have demonstrated that by injecting viruses carrying a gene encoding what is known as a reporter (a gene that will express a product you can see under the microscope) into the subretinal space, some retinal cells can be induced to express the reporter product. At the present time these studies are mostly designed to find the best way to deliver genes into mature photoreceptor cells. However, gene therapy may not immediately be useful for retinal disease since photoreceptor cells are not easily amenable to sticking a new gene into them, mostly because rods and cones are rather inaccessible and there are limitations to transferring genes into non-dividing cells. Furthermore, gene replacement may not help a disease where the mutant gene product itself is toxic to the photoreceptor cells. Perhaps the best time to use gene therapy may be in cases where defects mostly involve loss-of-function genes. Nevertheless, researchers have recently introduced artificial genes into 1-day old mouse embryos affected with mutations in the peripherin gene and the gene encoding the molecule important for converting absorbed light into an electrochemical signal (recall that both of these mouse mutants have dying photoreceptor cells). The embryos that now have the artificial genes which encode normal products of the mutant genes were found to have healthy photoreceptor cells.

Another potential way to rescue dying photoreceptors is to provide them with survival/growth factors. Growth factors have been found to participate in diverse roles such as controlling cell division of retinal progenitor cells and regulating the growth of neural connections between retinal neurons. However, only recently has their role been studied during retinal disease. Perhaps the first indication that secreted growth factors can rescue photoreceptor cells from dying was obtained from studies on rats constructed to contain both normal and RCS RPE, as described above. In the retina of these "mixed-RPE" animals, photoreceptor cells adjacent to RCS RPE were dying as expected, and those that were lying next to normal healthy RPE were alive and well. However, photoreceptor cells that were lying "just" beyond immediate contact with the normal healthy RPE also appeared alive and well. This basic observation suggested to researchers that direct photoreceptor-RPE contact was not needed and that normal healthy RPE might be secreting a survival promoting factor.

Perhaps one of the best characterized survival/growth factors in the retina is fibroblast growth factor (FGF). Injections of FGF into the vitreous of an eye from the RCS rat or rats with light damaged retina prevents photoreceptor cell death for several months, even as outer segment debris accumulates. Similar results have been seen when FGF is injected into the subretinal space. However, strangely, even sham injections where no growth factor is injected can delay photoreceptor cell death, although the effect is small and very localized to the point of injection. Why did the researchers find a survival response with the sham injections? One possible explanation is that various growth factors were locally released from damaged retinal tissues or even macrophages which are often seen migrating into the damaged area. Macrophages (phagocytic cells found in the wall of blood vessels and in loose connective tissues) are known to produce many different growth factors, some of which could have the ability to promote photoreceptor cell survival.

Other experiments have shown that vitreal injections of human subretinal fluid as well as other growth factors can rescue dying photoreceptor cells. For example, one recent study tested eight different factors injected into the retina of rats with dying photoreceptor cells, all showing the ability to delay the degeneration of photoreceptor cells. The data from these experiments clearly show that the mechanisms with which potential survival agents act can be extremely complex. This is because many of the growth factors don't necessarily interact specifically with photoreceptor cells and they are known to activate different receptor systems on the cell surface. In addition, as described above, photoreceptor cells are in intimate contact with RPE, Mčller cells, other photoreceptors and retinal neurons. Any one or a combination of these cells could contain or perhaps be induced to express factors of their own with photoreceptor cell survival promoting activity.

So, how do you fix an eye? Or rather, how can we fix problems associated with photoreceptor cells? Obviously, we don't really know yet. Research of the retina is still focused on the basics, like how do the cells die and what keeps them alive in the normal retina? For now, a more realistic short term view is that we may be able to, at very least, slow down the inevitable death of photoreceptor cells, which is significant if someone with a retinal disease or injury might have an extra month or year of useful vision.

Now, you may want to review a Resource List on Vision.

Science Seminars Archive

Let's Collaborate Index

Custom Search on the AE Site