by Jennifer Phillips, Ph.D.
The time has come to delve into the retinal component of Usher syndrome. In Part II, I briefly described the results of protein localization studies, in which most members of the Usher cohort were found at the connecting cilium of the photoreceptor and at the photoreceptor synapse. The following diagram summarizes these findings:
Recall the structure of the photoreceptor cell described in Part I. The inner segment, just above the nucleus, contains all the standard-issue cell operating equipment: specialized molecules required for producing protein, degrading cellular waste products and performing various other metabolic functions. The outer segment contains the intricately folded membrane discs with which light sensitive molecules are associated. Between these two cellular compartments lies the connecting cilium, which grows out of the inner segment, extends up into the outer segment, and is surrounded by a structure known as the periciliary ridge, which encircles the connecting cilium like a little cuff. The cilium serves as a functional connection between the inner and outer segments, as well as a structural one. Proteins and other cellular materials synthesized in the inner segment need to get to the outer segment in order to perform their particular jobs up there, and materials that are no longer needed in the outer segment need to be carried away and dealt with in the inner segment. The connecting cilium acts as a transport system to which motor proteins can anchor and pull their molecular cargo up or down as needed.
The localization studies of the Usher proteins reveal that many of them are in the vicinity of the connecting cilium, but a closer look at this region of the cell shows that they are specifically either in the periciliary ridge (the ‘cuff’) or the space between the periciliary ridge and the connecting cilium:
The congenital deafness in human patients and mouse models of the disease, and the defects in stereocilia formation seen in the Usher mice are nicely explained by the model of protein interaction and function in the developing hair cells, discussed in Part II. The retinal cells, however, appear to develop normally and apparently function normally until they begin to degenerate. I say ‘apparently’ because the ‘pre-death’ state of the photoreceptors has been difficult to observe thus far. Historically, the first sign of a problem in human Usher patients occurs when the hearing-impaired child or teenager begins to experience night blindness due to a loss of rod photoreceptors in the periphery of the eye. By the time of this first clinical exam in such cases, the degeneration is already well underway. Fortunately (from an investigative point of view, at least) this is changing with genetic screening and early identification of Usher patients, but even with earlier eye exams it’s still not at all clear what is going wrong in the retina at the molecular level.
At this point you might well ask what clues the Usher mutant mice, which proved so valuable in adding to our understanding of the disease progression in the ear, can tell us about the events leading up to retinal degeneration. To our great consternation, most of the originally identified Usher mice do not undergo retinal degeneration at all! A number of these mutant mice have been examined expectantly until the end of their natural lives (around 2 years) and most do not exhibit any abnormality in their retinas. The exceptions to this are older mice with mutations in the Cadherin 23 (ush1d) gene, which show a slight reduction in visual function older ages, and Myo7a (ush1b) mutant mice, which exhibit a fairly distinct defect in moving proteins around in the retinal pigmented epithelium. Neither type of mouse shows any retinal degeneration.
Several theories have been put forth to explain this discrepancy between the mouse and human forms of the retinal disease. One possibility is that mice, being nocturnal animals and usually raised in low-light laboratory conditions, may not endure the bright light exposure that human retinas must withstand. Another explanation may lie in the slow, progressive nature of the human disease and the relatively short life cycle of the mouse—perhaps two years just isn’t long enough for the retinal defects to manifest in the mouse retina. A third theory centers on the fact that all of the known Usher proteins actually exist in multiple variations—the genetic code that specifies each of these proteins can be cut and spliced in a few different ways, giving rise to similar, but not identical protein products. The exact roles of the different isoforms of every gene aren’t yet clear, but some of them do appear to be more important in the ear than in the eye. It’s possible that the mutations in mouse Usher genes that give rise to such a strong ear phenotypes don’t affect the part of the protein that’s important for retinal cell function, and thus the mouse is spared the vision loss that characterizes the human disease. In further support of this latter theory is that fact that many of the Usher syndrome genes are also linked to non-syndromic deafness in humans—hearing loss without associated blindness.
None of the above theories are mutually exclusive, and it may turn out to be a combination of genetics, environment and life span that has limited the retinal phenotype of these Usher mutant mice. Encouragingly, excellent progress has been made through the use of genetically engineered mice, in which an Usher protein is removed completely (see knockout mice for more on this technique) or, alternatively, a targeted mutation is introduced into a particular Usher gene that renders the encoded protein non-functional. Thus far, these genetically modified mice show late-onset retinal degeneration (I’ve blogged about two such strains previously, here and here) and are providing important new avenues for therapeutic research. In these mice, therefore, we have a more complete model of Usher syndrome, although the retinal degeneration still appears to initiate later in mouse development than in the corresponding human lifespan. Moreover, as useful as these new strains have been, the Knockout technique isn’t foolproof. Knockout mice for at least two Usher genes (Ush1c and Clrn1) have not displayed significant retinal degeneration in any studies published to date.
In short, there are still a great many unanswered questions surrounding the pathophysiology of Usher syndrome, particularly with respect to the retinal phenotype. To complement the data being collected from the mouse models, myself and other scientists have investigated various Usher proteins in the zebrafish. There are some differences in the retinal anatomy of zebrafish and humans, but basic cell structure and function is conserved between the two species. Additionally, there are some similarities that make zebrafish an especially appealing organism for this type of study, including the fact that fish are diurnal animals with rich color vision--even better than humans, in fact, as they can see light in the ultraviolet range of the spectrum. Other advantages to using zebrafish are related to their development. Zebrafish embryos undergo fertilization and development outside the mother’s body, and usually several hundred embryos are produced from a single mating. They develop rapidly and are able to swim, see and hear just a few days after fertilization. Thus, we can conduct vision, hearing and balance tests within the first week of development and obtain results quite rapidly to learn more about the consequences of losing Usher gene function.
Understanding the cellular events that precede the death of these cells will be crucial in identifying ways to improve diagnosis and treatment of Usher syndrome. In the conclusion of the Usher story, I’ll discuss current clinical practices for managing Usher syndrome, and the direction of the research efforts designed to enhance these treatments.