Multiple light damage protocols have been described to damage photoreceptors and consequently induce a retinal regeneration response in adult zebrafish. This protocol describes an improved method that can be used in pigmented animals and that damages the vast majority of rod and cone photoreceptors across the entire retina.
Light-induced retinal degeneration (LIRD) is commonly used in both rodents and zebrafish to damage rod and cone photoreceptors. In adult zebrafish, photoreceptor degeneration triggers Müller glial cells to re-enter the cell cycle and produce transient-amplifying progenitors. These progenitors continue to proliferate as they migrate to the damaged area, where they ultimately give rise to new photoreceptors. Currently, there are two widely-used LIRD paradigms, each of which results in varying degrees of photoreceptor loss and corresponding differences in the regeneration response. As more genetic and pharmacological tools are available to test the role of individual genes of interest during regeneration, there is a need to develop a robust LIRD paradigm. Here we describe a LIRD protocol that results in widespread and consistent loss of both rod and cone photoreceptors in which we have combined the use of two previously established LIRD techniques. Furthermore, this protocol can be extended for use in pigmented animals, which eliminates the need to maintain transgenic lines of interest on the albino background for LIRD studies.
Light-induced retinal degeneration (LIRD) is commonly used in both rodents and zebrafish to damage rod and cone photoreceptors. In adult zebrafish, photoreceptor degeneration triggers Müller glial cells to re-enter the cell cycle and produce transient-amplifying progenitors. These progenitors continue to proliferate as they migrate to the damaged area, where they ultimately give rise to new photoreceptors. Currently, there are two widely-used LIRD paradigms, each of which results in varying degrees of photoreceptor loss and corresponding differences in the regeneration response. As more genetic and pharmacological tools are available to test the role of individual genes of interest during regeneration, there is a need to develop a robust LIRD paradigm. Here we describe a LIRD protocol that results in widespread and consistent loss of both rod and cone photoreceptors in which we have combined the use of two previously established LIRD techniques. Furthermore, this protocol can be extended for use in pigmented animals, which eliminates the need to maintain transgenic lines of interest on the albino background for LIRD studies.
All procedures described in this protocol were approved by the animal use committee at Wayne State University School of Medicine.
1. Dark Adaptation
2. UV Light Exposure
3. Halogen Lamp Light Exposure
4. Tissue Collection
The heretofore described light treatment protocol was compared to each individual method of LIRD. In dark-treated adult albino animals (Figures 3-5), the individual light treatments resulted in significant loss of rod (Figure 3) and cone (Figure 4) photoreceptors. However, both individual treatments primarily damaged photoreceptors in the dorsal half of the retina, leaving the ventral retina relatively protected from the light treatments (Figures 3 and 4). In addition, compared with the halogen light treatment, the UV light treatment damaged more cone photoreceptors in the dorsal half of the retina (compare Figures 4B with C, respectively). Combining the UV and halogen light treatments resulted in significantly greater loss to both rod and cone photoreceptors throughout the retina, largely eliminating the neuroprotection of the ventral half of the retina (Figures 3 and 4). Compared with the individual LIRD methods, a corresponding increase in proliferation was also observed in the ventral retina when the combined light treatment protocol was used (Figure 5).
Pigmented animals are more resistant to LIRD due to retinal pigmented epithelium, which absorbs light and protects the photoreceptors from phototoxicity. As expected, dark-adapted pigmented animals were nearly completely resistant to the halogen light treatment alone (Figures 6B, J, and N). We found that 48 hr after light onset, rod photoreceptors were reduced in the dorsal retina, but not ventral retina (Figures 6E, F, and Q). Cone photoreceptor nuclei were present in normal numbers (Figure 6R). Compared with untreated retinas, halogen light treatment alone also did not increase Müller glial cell proliferation 48 hr after light onset (Figures 7B and F). In albino animals, UV light treatment alone resulted in slightly greater damage to both rod and cone photoreceptors in the dorsal retina (Figures 3I and 4I). However, in pigmented animals, UV light treatment alone did not significantly reduce rod or cone cell number (Figures 6C, G, K, O, Q, and R). In addition, UV treatment alone only elicited a weak regenerative response in the dorsal half of the retina in pigmented animals (Figure 7C). In contrast to the individual LIRD results, combining the UV and halogen light treatments resulted in significantly greater photoreceptor damage and regenerative response in pigmented animals. Importantly, significant loss of both rods and cones was observed in both the dorsal and ventral retinas (Figures 6D, H, L, P, Q, and R). Compared with untreated and individual LIRD methods, a corresponding increase in proliferation was also observed in both dorsal and ventral halves of the retina when the combined light treatment protocol was used (Figure 5).
Figure 1. Photograph depicting the halogen light treatment setup. Two sets of four 250 W halogen lamps were spaced 29 cm on either side of two clear 1.8 L acrylic tanks. The aerator and tubing were placed so as not to obstruct the light. The lids of the tanks were cracked 2 cm to allow airflow from the fan. A thermometer was placed in one of the tanks for the temperature to be monitored. Click here to view larger image.
Figure 2. Photograph depicting the UV light treatment setup. The UV light source from a fluorescent stereomicroscope was securely fixed to a glass Petri dish ~2 cm off the bench top. A 250 ml glass beaker, partially wrapped in foil, was filled will 100 ml of system water. The fish were placed inside the 250 ml beaker. The 250 ml beaker was centered inside a 4 L beaker, which was filled with system water to the level of the water in the 250 ml beaker. The 4 L beaker was placed adjacent to the light filament centered at the 250 ml beaker. A large opaque shield was placed around the entire setup. Click here to view larger image.
Figure 3. Rod photoreceptor loss in albino zebrafish following various light damage paradigms. Retinal sections from adult Tg(nrd:egfp)/alb immunolabeled with GFP to show rod photoreceptor density (green) in the dorsal (A-D) and ventral (E-H) halves of the retina. A, E) Prior to light onset (0 hr), EGFP-positive rod photoreceptor nuclei (ONL) and outer segments (ROS) can be visualized. B, F) At 48 hr following halogen light onset (hLt), truncated ROS and significantly fewer rod nuclei (I) were present in the dorsal, but not ventral retina. C-G) At 48 hr after 30 min of UV exposure (hpUV), significantly fewer rod nuclei were present in the dorsal and ventral retinas compared to 0 hr (C, G), but not 48 hLt treatments (G, I). D-H) 48 hr following onset of combined light treatment (30 min UV exposure followed by halogen treatment; UV+48 hLt), almost no ROS were present in dorsal or ventral retinas, in addition to a significant loss of rod photoreceptor nuclei (I). I) Quantification of rod photoreceptor nuclei across 300 µm of the central dorsal or ventral retina. Significant differences (p ≤ 0.05) between groups are depicted as: "a" for different from 0 hr, "b" for different from 48 hLt and "c" for different from 48 hpUV. Scale bar in panel A represents 50 µm and applies to panels A-H. Click here to view larger image.
Figure 4. Double cone loss in albino zebrafish following various light damage paradigms. Retinal sections from adult albino zebrafish immunolabeled with Zpr-1 to show double cones (gold) in the dorsal (A-D) and ventral (E-H) halves of the retina following each of the light damage paradigms (48 hLt, 48 hpUV, and UV+48 hLt). A, E) At 0 hr, double cone nuclei were present in dorsal and ventral retinas. B, F) At 48 hLt, significantly fewer double cones were present in the ventral retina only (F, I). C-G) At 48 hpUV, significantly fewer double cones (I) and a large amount of debris was present in the dorsal retina only (C), while no changes were observed in the ventral retina (G). D-H) At UV+48 hLt, significantly fewer double cones (I) and very little debris was present in dorsal and ventral retinas. I) Quantification of double cones across 300 µm of the central dorsal or ventral retina. Significant differences (p ≤ 0.05) between groups are depicted as, "a" for different from 0 hr, "b" for different from 48 hLt and "c" for different from 48 hpUV. Scale bar in panel A represents 100 µm and applies to panels A-H. Click here to view larger image.
Figure 5. Changes is proliferation response following various light damage paradigms. Retinal sections from adult albino zebrafish immunolabeled with PCNA to show cell proliferation (red) in the dorsal (A-D) and ventral (E-H) halves of the retina following each of the light damage paradigms (48 hLt, 48 hpUV, and UV+48 hLt). A-E) At 0 hr, proliferation was absent from the inner nuclear layer (INL). B, F) At 48 hLt, single progenitor cells were present in the INL across the dorsal retina and in a small portion of the ventral retina. C-G) At 48 hpUV, columns of progenitor cells were present in the dorsal retina, while only single progenitor cells were present in a small portion of the ventral retina. D, H) At UV+48 hLt, large columns of progenitor cells were present across the dorsal and ventral retina. Scale bar in panel A represents 100 µm and applies to panels A-H. Click here to view larger image.
Figure 6. Rod and cone photoreceptor loss in pigmented zebrafish following various light damage paradigms. Retinal sections from adult wild-type (AB) zebrafish stained with TO-PRO-3 to show all nuclei (A-P; blue) and immunolabeled with Zpr-3 to show rod outer segments (A-H; magenta) or Zpr-1 to double cones (I-P; gold) in the dorsal (A-D, I-J) and ventral (E-H, M-P) halves of the retina following each of the light damage paradigms (48 hLt, 48 hpUV, and UV+48 hLt). A-H) Significant decreases in rod photoreceptor nuclei were present in the dorsal retina at 48 hLt (B, Q) and in the dorsal and ventral retina at UV+48 hLt (D, H, Q). I-P) Although cone nuclei were hypertrophied in the dorsal retina at 48 hpUV (K), no significant decreases in double cone nuclei were present at 48 hLt or 48 hpUV (J-K, N-O, R). Significant decreases in cone nuclei were present in the dorsal and ventral retina at UV+48 hLt (L, P, R). Q-R) Quantification of rod photoreceptor and double cone nuclei across 300 µm of the central dorsal or ventral retina. Significant differences (p ≤ 0.05) between groups are depicted as, "a" for different from 0 hr, "b" for different from 48 hLt and "c" for different from 48 hpUV. Scale bar in panel A represents 200 µm and applies to panels A-H. Click here to view larger image.
Figure 7. Proliferation response in pigmented zebrafish following various light damage paradigms. Retinal sections from adult wild-type (AB) zebrafish immunolabeled with PCNA to show cell proliferation (red) in the dorsal (A-D) and ventral (E-H) halves of the retina following each of the light damage paradigms (48 hLt, 48 hpUV, and UV+48 hLt). A-D) In the dorsal retina, only a few single progenitor cells were present in the INL at 48 hLt and 48 hpUV (B, C), while columns of progenitor cells were present at UV+48 hLt (D). E-H) In the ventral retina, no changes in proliferation were observed except at UV+48 hLt (H), at which time columns of progenitor cells were present. Scale bar in panel A represents 100 µm and applies to panels A-H. Click here to view larger image.
Here we show that combining a short UV exposure with a continual bright light exposure results in widespread photoreceptor loss and a robust regeneration response. Compared with the individual LIRD methods, this combined method is also the most effective protocol to damage both rods and cones in both halves of the retina. Importantly, this treatment is effective in pigmented animals as well as albino animals.
Although we provide evidence that the combined protocol results in more widespread and consistent damage compared with the individual LIRD methods, scientific considerations should be discussed before selecting a LIRD paradigm. Our data suggest that the combined method should be utilized when the entire retinal tissue is collected for analysis (i.e. proteomics, real-time PCR). We found that the combined method had the greatest lesion size, followed by the halogen exposure, and finally the UV exposure. We suggest that the combined method also be used when investigating a gene or pathway that could differentially effect rod and cone regeneration. If this is not possible for practical reasons, our data suggest that the UV exposure is the next best method for these studies, if the analysis is limited to the central dorsal retina.
Practical considerations are also warranted when selecting the appropriate LIRD method. The first concern is the cost of the necessary equipment. All the protocols require a prolonged period of dark adaptation (data not shown). A self-contained dark enclosure that is built into the zebrafish housing module is extremely convenient, but not necessary. Even a cardboard box could be used, if care is taken to tape or seal the seams, and the water parameters are monitored. The halogen-based LIRD method is extremely economical and convenient. All items can be purchased for under $100 US at a local hardware store and set up takes just a few minutes. The only practical disadvantage of this method is the inconvenience of working in the same room as the light treatment. The lights generate a good amount of heat when running continuously for 4 days and the light intensity is both distracting and potentially damaging to human vision. The UV and combined methods require a UV source, which is not available to every laboratory. In addition, great care should be taken to ensure the safety of the lab members when performing a UV light treatment. Thus, while it is more convenient to have the halogen light apparatus in an isolated space, the safety concerns of the UV treatment necessitate such a space. This becomes an additional inconvenience if the UV source used for the light treatment is also typically used for a fluorescent microscope housed in a "microscope room" or imaging core. Performing multiple rounds of UV treatments for control and experimental groups of animals would necessitate shutting down the microscope room for an extensive length of time.
Perhaps the greatest advantage to using the combined method is the ability to perform retinal regeneration studies on pigmented animals. We show that neither individual method results in significant loss of rods and cones in the dorsal and ventral retina. In contrast, the combined LIRD paradigm results in similar levels of widespread rod and cone loss in both pigmented and albino animals. The ability to apply this LIRD method to pigmented animals saves the researcher significant space, time, and per diem animal care costs (if any), as it eliminates the need to generate and maintain transgenic lines of interest on the albino background for retinal regeneration studies. Together, we feel that the scientific and practical advantages outweigh the potential inconveniences of this improved LIRD method.
The authors have nothing to disclose.
The authors would like to thank Xixia Luo for excellent fish husbandry and technical support. This work was funded by the National Institutes of Health grants R21EY019401 (RT) and P30EY04068 (RT), and start-up funds to RT, including an unrestricted grant from Research to Prevent Blindness to the Wayne State University, Department of Ophthalmology. JT was supported by a Thomas C. Rumble Fellowship provided by the Wayne State University Graduate School.
UV light source | Leica | EL600 |
Glass Petri dish (150 x 20 mm) | Sigma-Aldrich/Pyrex | CLS3160152BO |
250 ml glass beaker | Sigma-Aldrich/Pyrex | CLS1000250 |
4 L glass beaker | Sigma-Aldrich/Pyrex | CLS10004L |
Aluminum foil | Fisher | 01-213-105 |
250 W halogen lamps | Workforce | 265-669 |
1.8 L clear acrylic tanks | Aquaneering | ZT180T |
1.8 L clear acrylic tank lids | Aquaneering | ZT180LCL |
Fan | Honeywell | HT-900 |
Aerator | Tetra | 77853-900 |
Thermometer | Cole-Parmer | YO-08008-58 |
Bent forceps (5/45) | World Precision Instruments | 504155 |