By Emily Crotteau
What if your eyes could regenerate over time, saving you from the declining vision many people face as they age? Scientists are hoping to find answers to this question by studying fish, whose eyes possess regenerative capacity throughout their lives. Fish and humans are distantly related, but if fish evolved their eye-regenerating abilities before the human lineage diverged, it is possible that similar mechanisms can be reactivated in the mammalian eye. We, too, may have a latent capacity to regenerate our eyes, if only the right chemical signals are sent. The quest for such chemical signals in humans can only be undertaken if we have a working understanding of eye regeneration in fish; otherwise, we would not know where to begin to look. In 2008, Morris et al took a novel approach to this question by studying two zebrafish models that experience different forms of eye degeneration. They found that these two types of fish respond to eye degeneration in different ways, indicating that there are multiple modes of regeneration in the eye. If even one of these processes has an analog in humans, it could open whole new doors in the treatment of degenerative eye diseases.
When you want to understand how a system works, it is useful to control as many variables as possible, so you can study the effect of each one individually. This lets you ask questions about correlations between the variables: When I change this, how does it affect that? What if I change this other thing instead? What is its effect on that? Cell regeneration is difficult to study because it occurs in response to cell damage or deterioration, and when scientists intentionally cause such damage, it can introduce additional variables with unknown effects. For instance, if a tissue is damaged using lasers, a common technique, how is the regenerative response impacted by the fact that lasers, an unlikely culprit in most medical cases, caused the damage? Perhaps the lasers introduced other external or collateral effects that are not relevant to someone with age-related eye degeneration. Since inducing damage introduces added complexity and uncertainty to studies of regenerative processes, Morris et al looked for another method of studying deterioration in fish eyes. This was difficult because in normal circumstances, fish eyes do not deteriorate over time the way human eyes do. Fortunately, modern genetic techniques provided a solution: create mutant fish!
Morris et al characterized two strains of mutant zebrafish, one which experiences selective degeneration of rods, which are responsible for low-light vision, and one which experiences selective degeneration of cones, which provide color vision. Rods and cones are photoreceptor cells, the only cells in the eye that directly react to light; all the rest are part of the process that transforms the information about light gathered by these cells into an electrochemical signal that can be interpreted by the brain, a complex information relay that produces the images we see. Because the two types of fish experience “natural” degeneration, Morris et al did not need to depend on inducing damage to observe the process of rod and cone regeneration independently.
Equipped with their two strains of mutant fish, Morris et al got out their microscopes and labeling dyes and set out to learn what they can about eye regeneration. To understand their results, it is useful to understand two distinct regions within the eye called the inner nuclear layer (INL) and outer nuclear layer (ONL). Rod and cone cells live in the outer nuclear layer and form connections with cells in the inner nuclear layer which are responsible for sending signals to the brain. One special type of cell called the Müller glial cell spans both layers, and is thought to play a structural role in the eye and also seems to have a role in retinal stem cell production. Scientists often measure cell regeneration by observing how many cells are proliferating a given point in time, and where this proliferation is occurring. Intuitively, one would expect the mutant fish to experience proliferation in the outer nuclear layer of their eyes, because this is where cells are degenerating. Evidence of proliferation elsewhere in the eye would provide insight into other cells involved in the regeneration process. This is exactly what Morris et al were looking for, and exactly what they found, but only in one of their mutants!
As expected, Morris et al observed proliferation in the outer nuclear layer in both of their mutant fish strains. More surprising, and unexpected, was that proliferation also occurred in the inner nuclear layer and Müller glial cells of the fish that underwent cone degeneration, but not in those that underwent rod degeneration. This result suggests that two different methods of regeneration exist for the two major types of photoreceptor cells. Damage-induction experiments had also observed proliferation in the inner nuclear layer, but until now, it was uncertain whether this was a collateral effect of the damage or a genuine part of the regenerative response. Now, a more nuanced answer can be given: the inner nuclear layer and Müller glia are directly involved in photoreceptor regeneration, but only for cone cells. Having identified where the regenerative response is occurring, the next question is, what, specifically, is going on?
This question can begin to be tackled with another where question” where do the new cells come from? This is part of the broader study of cell differentiation, the investigation how and when cell identity is determined. In the zebrafish eye, populations of retinal stem cells in the inner nuclear layer can differentiate into all types of eye cells throughout the life of the fish. These cells give rise to a variety of what are called progenitor cells, a kind of less-differentiated or “parental” cell type that can produce one or many types of offspring that differentiate further, usually into a fully-differentiated form. The observation that cone degeneration in zebrafish provokes a proliferative response in the inner nuclear layer suggests that retinal stem cells are involved in the regenerative process. In contrast, the observation that rod degeneration only produced proliferation in the outer nuclear layer suggests that a population of rod progenitor cells, distinct from cone progenitors, produced the replacement cells.
This piqued the interest of Morris et al, and they conducted a further experiment that detected rod-specific cellular markers in rod progenitors, suggesting that “rod identity” begins earlier in the differentiation process than had been previously thought, with distinct cone and rod lineages arising from retinal stem cells, rather than a generic photoreceptor progenitor. This discovery is akin to finding out that two boys you had thought were fraternal twins are actually half-siblings! Rod and cone cells are still related, but not as closely as we had thought.
In 2009, Alvarez-Delfin et al, a research group that included members of the Morris et al team, revisited the question of rod and cone development in a third zebrafish mutant, the lots-of- rods mutant, which exhibits an overabundance of rod photoreceptor cells. The extra rods, surprisingly, emerge from a type of cone precursor cell, rather than from the normal rod lineage. Their findings suggest that normal cone cells must repress the tendency to become rods, and that the lots-of-rods mutant lacks the ability to repress “rod-ness” in certain cone precursors. Thus, it seems that cone cells get their identity by “avoiding” chemical triggers that would push them into another identity. This interpretation makes sense in light of the results from Morris et al, because it explains why rod progenitors in the outer nuclear layer can provide replacements in response to rod degeneration, but not cone degeneration. Because rod progenitors have embraced their “rod-ness”, rather than suppressing it as cone precursors must, they cannot produce replacement cone cells. Thus, new cone cells must be recruited from differentiating retinal stem cells in the inner nuclear layer.
These findings bring us back to the question that continues to motivate this research: how can our understanding of zebrafish eye regeneration aid in the development of treatment and therapies for humans with degenerative eye diseases who may be facing a lifetime of blindness. In the course of this research, Morris et al and Alvarez-Delfin et al identified many of the genes involved in the proliferative response to cone and rod degeneration. If analogs to these genes can be found in humans, they will open up whole new avenues of study. Such genes, if they are present, are unlikely to play an identical role to those in fish, because human eyes lack the regenerative power of zebrafish eyes. However, it is possible that these genes could be stimulated or otherwise manipulated into performing such a role. The fruits of this research are unlikely to make their way into human medicine for a long time, but for now, speculation like this fuels interest in further funding and research into the mysteries of the zebrafish eye. The next step is to move from fish into mouse models, a fellow mammal which is also unable to regenerate photoreceptors. By exploring mouse mutants using techniques similar to those of Morris et al, we will gain a better understanding of the similarities and differences between fish and mammalian eyes and will be able to make more realistic predictions about the future of treatment for human eye disease. Even as research moves into mouse models, there is still much work to be done in zebrafish. The story of the zebrafish eye is far from complete, and surely many further insights are in store!
If you’re interested, you can migrate to the primary articles on which this piece is based.
Morris, Ann C., Tamera L. Scholz, Susan E. Brockerhoff, and James M. Fadool. (2008). Genetic Dissection Reveals Two Separate Pathways for Rod and Cone Regeneration in the Teleost Retina. Dev Neurobiol. 68(5): 605–619.