- Email: [email protected]
Zebrafish retinal mutants
Vision Research 38 (1998) 1335 – 1339
Zebrafish retinal mutants Susan E. Brockerhoff a,*, John E. Dowling b, James B. Hurley c a
b
Department of Biochemistry, Uni6ersity of Washington, Box 357350, Seattle WA 98195, USA Department of Molecular and Cellular Biology, The Biological Laboratories, Har6ard Uni6ersity, 16 Di6inity A6e, Cambridge MA 02138, USA c Department of Biochemistry and Howard Hughes Medical Institute, Uni6ersity of Washington, Box 357370, Seattle WA 98195, USA Received 9 May 1997; received in revised form 11 July 1997
Abstract We have initiated a genetic analysis of the zebrafish visual system to identify novel molecules involved in vertebrate retinal function. Zebrafish are highly visual; they have four types of cones as well as rod photoreceptors, making it possible to study both rod and cone-mediated visual responses. To identify visual mutants, optokinetic responses of mutagenized larvae are measured in a three-generation screen for recessive mutations. By measuring visual behavior our genetic screen has been targeted towards identifying mutants that do not have gross morphological abnormalities. The electroretinogram (ERG) of optokinetic-defective mutants is recorded and their retinas are examined histologically to localize defects to the retina. In this report, we summarize our screening results and ERG and histological analyses of the five morphologically normal mutants we have analyzed to date. Additionally, the more detailed characterization of a red-blind mutant that we have isolated is summarized. Our results indicate that mutants with defects in various processes such as photoreceptor synaptic transmission, photoreceptor adaptation and cell-type specific survival and/or function can be identified using this approach. © 1998 Published by Elsevier Science Ltd. All rights reserved. Keywords: Retina; Zebrafish; Mutant; Electroretinogram
An experimental approach that has only recently been used to dissect vertebrate retinal function and development involves the use of genetic screens. The usefulness of a genetic approach for studying neurogenesis and neuronal function in the eye has been demonstrated by nearly 30 years of genetic studies on the visual system of an invertebrate, Drosophila [1,2]. Visual system mutants identified through genetic screens using this organism have helped define fundamental principles underlying the development and function of the Drosophila eye. However, vertebrate and invertebrate vision are fundamentally different so the genetic studies done with Drosophila are not directly applicable to the vertebrate visual system. For example, an increase in illumination elicits hyperpolarization of vertebrate photoreceptors but it depolarizes invertebrate photoreceptors [3]. In addition to differences in phototransduction, the types of cells involved in image processing and the overall morphology of the eyes differ substantially between vertebrates and invertebrates. * Corresponding author. Tel.: + 1 206 616 9464; fax: +1 206 685 2320; e-mail: [email protected].
Because of these significant differences, we are developing a genetic approach to understanding a model vertebrate visual system. We have chosen to use zebrafish (Danio rerio) for these studies because they are small and easy and inexpensive to maintain in large numbers. Furthermore, they mate with high efficiency, produce large broods and have a robust visual response. Another advantage of zebrafish is that the embryos are transparent and develop very rapidly outside of the mother. Some 4 days post-fertilization (dpf), zebrafish larvae begin to swim and exhibit vision-dependent behavior [4]. Furthermore, during the last few years, methods to induce a high frequency of mutations in zebrafish have been established and techniques for growing large numbers of fish have been optimized [5–8]. These procedures have enabled two groups to successfully conduct large-scale three-generation genetic screens for recessive mutations that cause visible morphological abnormalities early in the development of the zebrafish [9,10]. Finally, genetic linkage maps in zebrafish are now available so mutant genes can potentially be isolated by positional cloning [11–13].
0042-6989/98/$19.00 © 1998 Published by Elsevier Science Ltd. All rights reserved. PII: S0042-6989(97)00227-7
1336
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339
Another reason for choosing zebrafish to identify retinal mutants, is that zebrafish have a sophisticated visual system. In contrast to mice, which have a roddominated retina and poor color vision (mice only have two types of cones; [14]), zebrafish are tetrachromatic, having four types of cones with absorption maxima extending from 360– 570 nm ([15]). Zebrafish also possess rod photoreceptor cells so both scotopic and photopic vision can be analyzed in this organism [16]. In adults, the cones are arranged in a row mosaic [17] [15]. Red- and green-sensitive cones alternate in sequence and are separated by either a single blue cone (always adjacent to a red-sensitive cone) or a single UV cone (always adjacent to a green-sensitive cone). Furthermore, early eye morphogenesis and the development of the visual system is similar to that of other vertebrates [18]. Information obtained from a genetic analysis of the zebrafish visual system should be applicable to other vertebrates. Thus, in zebrafish, rod- and cone-mediated visual behavior, color vision and mosaic formation can be analyzed. We first tested two vision-dependent behaviors of zebrafish larvae, the optokinetic response (OKR) and phototactic behavior, to determine which behavior would be most useful in a genetic screen [19]. By measuring visual behavior we set out to identify mutants that had subtle defects in visual abilities but were otherwise normal. Two criteria for a useful screen were considered important. The assay must be rapid, so that reasonable numbers of fish can be analyzed, and also almost all wild-type zebrafish must exhibit a robust, easily identifiable positive response. Although zebrafish show a positive response in both assays, only optokinetic responses are reliably and rapidly detected in almost all fish examined. Including the time spent arranging fish, only 1 min is required to analyze optokinetic responses in each fish. Furthermore, \ 95% of wild-type fish show a positive response by 5 dpf. Thus, we have been using this assay to measure the visual abilities of larvae in a three-generation screen for ethylnitrosourea-induced recessive mutations in the visual system and we have isolated several mutants (see below). The initial screening was conducted in collaboration with Dr Wolfgang Driever’s group at the Massachusetts General Hospital in Boston, as part of their large-scale screen for morphological mutants [19]. We then continued the screen for visual system mutants at Harvard [20]. Currently, we are also screening for visual mutants at the University of Washington. Optokinetic responses (eye movements) are measured by placing 10–20 5 – 7 dpf mutagenized third-generation larvae in a Petri dish containing a methyl cellulose solution to partially immobilize the fish. Larvae kept in methyl cellulose for more than 1 h continue to develop normally when returned to fish water. Larvae are arranged for optimal viewing in the methyl cellulose using
a dissecting needle. The dish is placed in the center of a microscope stage to which a circular drum is mounted. The drum has black and white vertical stripes (20 deg/cycle) on the inside and is turned by a belt attached to an adjacent motor (6 rpm). For each larva, the drum is rotated in two directions and the eye movements are analyzed by watching the larva through the microscope. Normal larvae show a definitive optokinetic response consisting of smooth pursuit eye movements followed by rapid saccades in the opposite direction. The eye movements are instantly reversed when the direction of the rotating stripes is changed. A response is considered positive if a single smooth pursuit and saccade eye movement in the proper direction is observed after starting drum rotation in each direction. To date, we have screened approximately 500 genomes for recessive mutations affecting the visual system. We have identified nine potential mutants that are morphologically normal but have abnormal optokinetic responses. No defects in the brain, eye or other organs can be detected in these morphologically normal optokinetic-defective mutants by examination under a high-powered dissecting microscope. Although the mutants appear morphologically normal, curiously, they all display abnormal light-dependent regulation of melanophore size. For example, ‘no optokinetic response a’, noa, when raised in ambient light, has expanded melanophores compared with its optokinetic-normal siblings. In contrast, the mutant ‘partial optokinetic response b’, pob, appears lighter than its normal siblings when raised under ambient light. Whether these phenotypes are due to the retinal defects we detect in these mutants (see below), or whether the pineal is also defective in these mutants is unknown and we are currently investigating this in more detail. Since our optokinetic-defective mutant larvae may have a lesion at any number of loci, we record the electroretinogram (ERG) to localize defects to the outer retina (for details on the vertebrate ERG see [21]). Histological analyses of the eyes of mutants with abnormal optokinetic responses are also conducted to determine if visible structural defects in the eye are apparent. We have analyzed the ERG and retinal histology of 5/9 mutants and four of these appear to have defects in the outer retina. A summary of the retinal morphology using the light microscope and ERG phenotype is presented in Table 1. Although neither the ERG analysis nor the histological analysis defines the molecular defect in these mutants, these data suggest that we have identified a wide variety of mutants with defects in processes as diverse as photoreceptor adaptation (no optokinetic response b, nrb), photoreceptor synaptic transmission (noa; [19]) and cell-type specific survival and/or function (pob; [20]). In nrb mutants, for example, we have measured the ability of photoreceptors to adapt to changes in illumi-
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339
1337
Table 1 Summary of retinal histology and ERG phenotype of five mutants with defective optokinetic (OK) responses Mutant
OK phenotype
ERG
Histology
Defect
noa m631 ([19])
No OKR in white light
b-wave severely reduced and delayed b-wave reduced and delayed. Adaptation defect b-wave reduced and delayed
Retina appears normal at 5 dpf Retina appears normal at 5 dpf Tiering of photoreceptors absent at 6 dpf. OPL abnormal Retina appears normal at 5 dpf Red cones are missing at 5 dpf
Synaptic transmission
nrb a13 (this pa- No OKR in white light per) nrc a14 ([22]) No OKR in white light poa m724 ([19]) pob a1 ([20])
OKR small and quick (erratic) at all wavelengths No OKR in red light. OKR observed in white light
ERG waveform normal. bwave reduced in bright light No ERG above 580 nm. Larger-than-normal a-wave below 580 nm
Photoreceptor function ??? General neurological defect Defect in red cone maintenance or function
Although nrc, noa and nrb all do not show an OKR with white light, all three mutants show occasional spontaneous eye movements, indicating that they do not have eye movement defects. OPL, outer plexiform layer.
nation by recording the ERG in the presence of background light. In wild-type larvae, the ERG responses with and without dim background light (46 mW/cm2) are similar; the b-wave is reduced only slightly by such background illumination. In contrast, in the nrb mutant with the same background illumination, the b-wave is reduced substantially (Fig. 1). Qualitatively similar results were seen in eight wild-type and in nine mutant larvae at 5–6 dpf. Other abnormalities in the nrb ERG are also obvious. For example, the a-wave in the mutant ERG is larger than normal and the b-wave is delayed and slightly reduced in amplitude. We think it likely that the primary defect in nrb is in the photoreceptors and we are presently examining photoreceptor function in these mutants. Since our optokinetic behavioral assay is designed to select mutant larvae that are morphologically normal, these fish tend not to have obvious morphological abnormalities in retinal structure. Two of the four mutants that we know have outer retinal defects, noa and nrb, have normal appearing retinas when 1 mM retinal sections are viewed with a light microscope. The outer nuclear layer of a third mutant, (no optokinetic response c) nrc, appears slightly abnormal when examined under the same conditions (not shown). We are currently defining the structural defect in the retina of this mutant by electron microscopy (Allwardt et al., in preparation). Only one mutant thus far, pob, has a rather obvious structural defect in the retina. This mutant specifically lacks red cone photoreceptors and therefore has a thinner outer nuclear layer than wildtype larvae. To date, pob is the only color-blind mutant that we have identified. It has an optokinetic response in white light but no response in red light. In this mutant, red cone photoreceptors specifically degenerate between 3.5 and 5 dpf, resulting in a 20% reduction in cell number in the outer nuclear layer [20]. The red cone photoreceptors form and then degenerate by an apoptotic
mechanism. When we analyze the mutant and wild-type retinas at 5 dpf for DNA fragmentation, a characteristic of apoptotic nuclei, the mutant has significantly more nuclei with fragmented DNA than the control. Furthermore, as expected if only photoreceptors are degenerating, the increase in apoptotic nuclei is seen only in the photoreceptor layer (Table 2). Interestingly, the pob mutation is not in the one known gene specific to red cones, red opsin [20]. In addition to these morphologically normal visual mutants, we and others have identified mutants with gross morphological abnormalities in the eye [24,25]. These mutants appear at a higher frequency than the behavioral mutants ( 1 morphological mutant/15 F2 families vs 1 behavioral mutant/50 F2 families) and most of these ‘small eye’ mutants have non-specific retinal degeneration and degeneration in the brain. A subset of the morphological mutants appear to have a cell-type specific degeneration. One mutant, archie, selectively loses most ganglion and some amacrine cells [24]. Several additional mutants lose all photoreceptor cells, whereas the other retinal layers in these mutants do not appear to degenerate. To date, no retinal mutants that appear to have a defect in cell formation (i.e. the mutant lacks a cell-type because it does not form) have been identified, although Malicki and colleagues have isolated a class of mutations that lead to patterning defects within the retina and brain [25]. In conclusion, over the last few years we have initiated a genetic screen to identify zebrafish mutants primarily with subtle retinal defects. Although this work is still in its infancy, we have clearly demonstrated that it is possible to identify mutants with a variety of retinal defects. The long-term goal is to clone the genes that have been altered in these fish and to study the gene products using molecular and biochemical methods. Linkage maps of the zebrafish genome are available [11–13] and thus positional cloning should be feasible. Furthermore, a procedure for generating retro-
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339
1338
Fig. 1. Representative ERG responses from 5 dpf nrb and sibling responder larvae. The responder ERG is only reduced slightly by background illumination (middle recording, top), whereas the ERG response of nrb larvae is reduced substantially by the identical procedure (middle recording, bottom). Tha a-wave (a) and b-wave (b) are indicated in the left recording on the top. The procedure for recording ERGs was as described previously ([20]) except the fish were not dark-adapted prior to the experiment and the light flash was B 1 msec in duration (Strobe flash at 1/16th power: Canon Speedlite 300 TL). See text for more details.
viral-insertion mutants has recently been published [26,27], and genetic screens using this approach are in progress. Finally, we hope to learn information about retinal function from characterization of the various mutants. For example, it is unclear whether red cone
photoreceptors have unique molecules in addition to red opsin. By isolating RNA that is expressed in wildtype retinas but not in pob retinas we may identify molecules specific to red cones other than red opsin.
References Table 2 Number of apoptotic cells/retina in each retinal nuclear layer Wild-type
pob
5.5 ganglion cells 9.5 inner nuclear layer cells 1 photoreceptor cell
6 ganglion cells 15.5 inner nuclear layer cells 178.5 photoreceptor cells
Apoptotic cells were identified in wild-type (n = 2) and pob (n =2) retinas at 5 dpf using the in situ Apoptosis Detection Kit (Oncor Inc.) and the protocol as described ([23]). After staining, larvae were dehydrated with a series of ethanol solutions and embedded in epon-araldite ([18]). 5 mm sections were examined using a light microscope and the apoptotic cells/section counted.
[1] Benzer, S. The 1990 Helmerich Lecture: the fly and eye. In: Lam DM, Shatz CJ, editors. Development of the Visual System. Cambridge: MIT Press, 1991. [2] Pak WL. Drosophila in vision research. The Friedenwald Lecture. Invest Ophthalmol Vis Sci 1995;36:2340 – 57. [3] Yarfitz S, Hurley JB. Transduction mechanisms of vertebrate and invertebrate photoreceptors. J Bio Chem 1994;269:14329– 32. [4] Clark DT, Visual responses in the developing zebrafish (Brachydanio rerio). In: Biology. Eugene: University of Oregon, 1981. [5] Driever W, Stemple D, Schier A, Solnica-Krezel L. Zebrafish: genetic tools for studying vertebrate development. Trends Genet 1994;10:152 – 9.
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339 [6] Mullins MC, Hammerschmidt M, Haffter P, Nusslein-Volhard C. Large-scale mutagenesis in the zebrafish: in search of genes controlling development in a vertebrate. Curr Biol 1994;4:189 – 202. [7] Mullins MC, Nusslein-Volhard C. Mutational approaches to studying embryonic pattern formation in the zebrafish. Curr Opin Genet Dev 1993;3:648–54. [8] Solnica-Krezel L, Schier AF, Driever W. Efficient recovery of ENU-induced mutations from the zebrafish germline. Genetics 1994;136:1401– 20. [9] Driever W, Solnica-Krezel L, Schier AF, et al. A genetic screen for mutations affecting embryogenesis in zebrafish. Development 1996;123:37 – 46. [10] Haffter P, Granato M, Brand M, et al. The identification of genes with unique and essential functions in the development of the zebrafish Danio rerio. Development 1996;123:1–36. [11] Johnson SL, Gates MA, Johnson M, et al. Centromere-linkage analysis and consolidation of the zebrafish genetic map. Genetics 1996;142:1277– 88. [12] Knapik EW, Goodman A, Atkinson OS, et al. A reference cross DNA panel for zebrafish (Danio rerio) anchored with simple sequence length polymorphisms. Development 1996;123:451 – 60. [13] Postlethwait JH, Johnson SL, Midson CN, et al. A genetic linkage map for the zebrafish. Science 1994;264:699–703. [14] Sz’el A, Rohlich P, Caff’e AR, Juliusson B. Unique topographic separation of two spectral classes of cones in the mouse retina. J Comp Neurol 1992;325:327–42. [15] Robinson J, Schmitt EA, Harosi FI, Reece RJ, Dowling JE. Zebrafish ultraviolet visual pigment: absorption spectrum, sequence, and localization. Proc Nat Acad Sci USA 1993;90:6009 – 12. [16] Kljavin IJ. Early development of photoreceptors in the ventral retina of the zebrafish embryo. J Comp Neurol 1987;260:461 –
71. [17] Branchek T, Bremiller R. The development of photoreceptors in the zebrafish, Brachydanio rerio. I. Structure. J Comp Neurol 1984;224:107 – 15. [18] Schmitt EA, Dowling JE. Early eye morphogenesis in the zebrafish, Brachydanio rerio. J Comp Neurol 1994;344:532–42. [19] Brockerhoff SE, Hurley JB, Janssen-Bienhold U, Neuhauss CF, Driever W, Dowling JE. A behavioral screen for isolating zebrafish mutants with visual system defects. Proc Nat Acad Sci USA 1995;92:10545– 9. [20] Brockerhoff SE, Hurley JB, Niemi GA, Dowling JE. A new form of inherited red-blindness identified in zebrafish. J Neurosci 1997;17:4236 – 42. [21] Dowling JE, The retina: an Approachable Part of the Brain. Cambridge, MA: Belknap Press of Harvard University Press, 1987. [22] Lall AB, Brockerhoff SE, Dowling JE. Electroretinographic characteristics of nrc mutant larvae of zebrafish. Invest. Ophthal. Vis. Sci. Florida, May 11 – 15. ARVO c 2889, 1997. [23] Abdelilah S, Mountcastle-Shah E, Harvey M, et al. Mutations affecting neural survival in the zebrafish Danio rerio. Development 1996;123:217 – 27. [24] Fadool JM, Brockerhoff SE, Hyatt GA, Dowling JE. Mutations affecting eye morphology in the developing zebrafish (Danio rerio). Dev Genet 1997;20:1 – 8. [25] Malicki J, Neuhauss SC, Schier AF, et al. Mutations affecting development of the zebrafish retina. Development 1996;123:263 – 73. [26] Gaiano N, Allende M, Amsterdam A, Kawakami K, Hopkins N. Highly efficient germ-line transmission of proviral insertions in zebrafish. Proc Nat Acad Sci USA 1996;93:7777 – 82. [27] Gaiano N, Amsterdam A, Kawakami K, Allende M, Becker T, Hopkins N. Insertional mutagenesis and rapid cloning of essential genes in zebrafish. Nature 1996;383:829 – 32.
. .
1339
Zebrafish retinal mutants Susan E. Brockerhoff a,*, John E. Dowling b, James B. Hurley c a
b
Department of Biochemistry, Uni6ersity of Washington, Box 357350, Seattle WA 98195, USA Department of Molecular and Cellular Biology, The Biological Laboratories, Har6ard Uni6ersity, 16 Di6inity A6e, Cambridge MA 02138, USA c Department of Biochemistry and Howard Hughes Medical Institute, Uni6ersity of Washington, Box 357370, Seattle WA 98195, USA Received 9 May 1997; received in revised form 11 July 1997
Abstract We have initiated a genetic analysis of the zebrafish visual system to identify novel molecules involved in vertebrate retinal function. Zebrafish are highly visual; they have four types of cones as well as rod photoreceptors, making it possible to study both rod and cone-mediated visual responses. To identify visual mutants, optokinetic responses of mutagenized larvae are measured in a three-generation screen for recessive mutations. By measuring visual behavior our genetic screen has been targeted towards identifying mutants that do not have gross morphological abnormalities. The electroretinogram (ERG) of optokinetic-defective mutants is recorded and their retinas are examined histologically to localize defects to the retina. In this report, we summarize our screening results and ERG and histological analyses of the five morphologically normal mutants we have analyzed to date. Additionally, the more detailed characterization of a red-blind mutant that we have isolated is summarized. Our results indicate that mutants with defects in various processes such as photoreceptor synaptic transmission, photoreceptor adaptation and cell-type specific survival and/or function can be identified using this approach. © 1998 Published by Elsevier Science Ltd. All rights reserved. Keywords: Retina; Zebrafish; Mutant; Electroretinogram
An experimental approach that has only recently been used to dissect vertebrate retinal function and development involves the use of genetic screens. The usefulness of a genetic approach for studying neurogenesis and neuronal function in the eye has been demonstrated by nearly 30 years of genetic studies on the visual system of an invertebrate, Drosophila [1,2]. Visual system mutants identified through genetic screens using this organism have helped define fundamental principles underlying the development and function of the Drosophila eye. However, vertebrate and invertebrate vision are fundamentally different so the genetic studies done with Drosophila are not directly applicable to the vertebrate visual system. For example, an increase in illumination elicits hyperpolarization of vertebrate photoreceptors but it depolarizes invertebrate photoreceptors [3]. In addition to differences in phototransduction, the types of cells involved in image processing and the overall morphology of the eyes differ substantially between vertebrates and invertebrates. * Corresponding author. Tel.: + 1 206 616 9464; fax: +1 206 685 2320; e-mail: [email protected].
Because of these significant differences, we are developing a genetic approach to understanding a model vertebrate visual system. We have chosen to use zebrafish (Danio rerio) for these studies because they are small and easy and inexpensive to maintain in large numbers. Furthermore, they mate with high efficiency, produce large broods and have a robust visual response. Another advantage of zebrafish is that the embryos are transparent and develop very rapidly outside of the mother. Some 4 days post-fertilization (dpf), zebrafish larvae begin to swim and exhibit vision-dependent behavior [4]. Furthermore, during the last few years, methods to induce a high frequency of mutations in zebrafish have been established and techniques for growing large numbers of fish have been optimized [5–8]. These procedures have enabled two groups to successfully conduct large-scale three-generation genetic screens for recessive mutations that cause visible morphological abnormalities early in the development of the zebrafish [9,10]. Finally, genetic linkage maps in zebrafish are now available so mutant genes can potentially be isolated by positional cloning [11–13].
0042-6989/98/$19.00 © 1998 Published by Elsevier Science Ltd. All rights reserved. PII: S0042-6989(97)00227-7
1336
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339
Another reason for choosing zebrafish to identify retinal mutants, is that zebrafish have a sophisticated visual system. In contrast to mice, which have a roddominated retina and poor color vision (mice only have two types of cones; [14]), zebrafish are tetrachromatic, having four types of cones with absorption maxima extending from 360– 570 nm ([15]). Zebrafish also possess rod photoreceptor cells so both scotopic and photopic vision can be analyzed in this organism [16]. In adults, the cones are arranged in a row mosaic [17] [15]. Red- and green-sensitive cones alternate in sequence and are separated by either a single blue cone (always adjacent to a red-sensitive cone) or a single UV cone (always adjacent to a green-sensitive cone). Furthermore, early eye morphogenesis and the development of the visual system is similar to that of other vertebrates [18]. Information obtained from a genetic analysis of the zebrafish visual system should be applicable to other vertebrates. Thus, in zebrafish, rod- and cone-mediated visual behavior, color vision and mosaic formation can be analyzed. We first tested two vision-dependent behaviors of zebrafish larvae, the optokinetic response (OKR) and phototactic behavior, to determine which behavior would be most useful in a genetic screen [19]. By measuring visual behavior we set out to identify mutants that had subtle defects in visual abilities but were otherwise normal. Two criteria for a useful screen were considered important. The assay must be rapid, so that reasonable numbers of fish can be analyzed, and also almost all wild-type zebrafish must exhibit a robust, easily identifiable positive response. Although zebrafish show a positive response in both assays, only optokinetic responses are reliably and rapidly detected in almost all fish examined. Including the time spent arranging fish, only 1 min is required to analyze optokinetic responses in each fish. Furthermore, \ 95% of wild-type fish show a positive response by 5 dpf. Thus, we have been using this assay to measure the visual abilities of larvae in a three-generation screen for ethylnitrosourea-induced recessive mutations in the visual system and we have isolated several mutants (see below). The initial screening was conducted in collaboration with Dr Wolfgang Driever’s group at the Massachusetts General Hospital in Boston, as part of their large-scale screen for morphological mutants [19]. We then continued the screen for visual system mutants at Harvard [20]. Currently, we are also screening for visual mutants at the University of Washington. Optokinetic responses (eye movements) are measured by placing 10–20 5 – 7 dpf mutagenized third-generation larvae in a Petri dish containing a methyl cellulose solution to partially immobilize the fish. Larvae kept in methyl cellulose for more than 1 h continue to develop normally when returned to fish water. Larvae are arranged for optimal viewing in the methyl cellulose using
a dissecting needle. The dish is placed in the center of a microscope stage to which a circular drum is mounted. The drum has black and white vertical stripes (20 deg/cycle) on the inside and is turned by a belt attached to an adjacent motor (6 rpm). For each larva, the drum is rotated in two directions and the eye movements are analyzed by watching the larva through the microscope. Normal larvae show a definitive optokinetic response consisting of smooth pursuit eye movements followed by rapid saccades in the opposite direction. The eye movements are instantly reversed when the direction of the rotating stripes is changed. A response is considered positive if a single smooth pursuit and saccade eye movement in the proper direction is observed after starting drum rotation in each direction. To date, we have screened approximately 500 genomes for recessive mutations affecting the visual system. We have identified nine potential mutants that are morphologically normal but have abnormal optokinetic responses. No defects in the brain, eye or other organs can be detected in these morphologically normal optokinetic-defective mutants by examination under a high-powered dissecting microscope. Although the mutants appear morphologically normal, curiously, they all display abnormal light-dependent regulation of melanophore size. For example, ‘no optokinetic response a’, noa, when raised in ambient light, has expanded melanophores compared with its optokinetic-normal siblings. In contrast, the mutant ‘partial optokinetic response b’, pob, appears lighter than its normal siblings when raised under ambient light. Whether these phenotypes are due to the retinal defects we detect in these mutants (see below), or whether the pineal is also defective in these mutants is unknown and we are currently investigating this in more detail. Since our optokinetic-defective mutant larvae may have a lesion at any number of loci, we record the electroretinogram (ERG) to localize defects to the outer retina (for details on the vertebrate ERG see [21]). Histological analyses of the eyes of mutants with abnormal optokinetic responses are also conducted to determine if visible structural defects in the eye are apparent. We have analyzed the ERG and retinal histology of 5/9 mutants and four of these appear to have defects in the outer retina. A summary of the retinal morphology using the light microscope and ERG phenotype is presented in Table 1. Although neither the ERG analysis nor the histological analysis defines the molecular defect in these mutants, these data suggest that we have identified a wide variety of mutants with defects in processes as diverse as photoreceptor adaptation (no optokinetic response b, nrb), photoreceptor synaptic transmission (noa; [19]) and cell-type specific survival and/or function (pob; [20]). In nrb mutants, for example, we have measured the ability of photoreceptors to adapt to changes in illumi-
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339
1337
Table 1 Summary of retinal histology and ERG phenotype of five mutants with defective optokinetic (OK) responses Mutant
OK phenotype
ERG
Histology
Defect
noa m631 ([19])
No OKR in white light
b-wave severely reduced and delayed b-wave reduced and delayed. Adaptation defect b-wave reduced and delayed
Retina appears normal at 5 dpf Retina appears normal at 5 dpf Tiering of photoreceptors absent at 6 dpf. OPL abnormal Retina appears normal at 5 dpf Red cones are missing at 5 dpf
Synaptic transmission
nrb a13 (this pa- No OKR in white light per) nrc a14 ([22]) No OKR in white light poa m724 ([19]) pob a1 ([20])
OKR small and quick (erratic) at all wavelengths No OKR in red light. OKR observed in white light
ERG waveform normal. bwave reduced in bright light No ERG above 580 nm. Larger-than-normal a-wave below 580 nm
Photoreceptor function ??? General neurological defect Defect in red cone maintenance or function
Although nrc, noa and nrb all do not show an OKR with white light, all three mutants show occasional spontaneous eye movements, indicating that they do not have eye movement defects. OPL, outer plexiform layer.
nation by recording the ERG in the presence of background light. In wild-type larvae, the ERG responses with and without dim background light (46 mW/cm2) are similar; the b-wave is reduced only slightly by such background illumination. In contrast, in the nrb mutant with the same background illumination, the b-wave is reduced substantially (Fig. 1). Qualitatively similar results were seen in eight wild-type and in nine mutant larvae at 5–6 dpf. Other abnormalities in the nrb ERG are also obvious. For example, the a-wave in the mutant ERG is larger than normal and the b-wave is delayed and slightly reduced in amplitude. We think it likely that the primary defect in nrb is in the photoreceptors and we are presently examining photoreceptor function in these mutants. Since our optokinetic behavioral assay is designed to select mutant larvae that are morphologically normal, these fish tend not to have obvious morphological abnormalities in retinal structure. Two of the four mutants that we know have outer retinal defects, noa and nrb, have normal appearing retinas when 1 mM retinal sections are viewed with a light microscope. The outer nuclear layer of a third mutant, (no optokinetic response c) nrc, appears slightly abnormal when examined under the same conditions (not shown). We are currently defining the structural defect in the retina of this mutant by electron microscopy (Allwardt et al., in preparation). Only one mutant thus far, pob, has a rather obvious structural defect in the retina. This mutant specifically lacks red cone photoreceptors and therefore has a thinner outer nuclear layer than wildtype larvae. To date, pob is the only color-blind mutant that we have identified. It has an optokinetic response in white light but no response in red light. In this mutant, red cone photoreceptors specifically degenerate between 3.5 and 5 dpf, resulting in a 20% reduction in cell number in the outer nuclear layer [20]. The red cone photoreceptors form and then degenerate by an apoptotic
mechanism. When we analyze the mutant and wild-type retinas at 5 dpf for DNA fragmentation, a characteristic of apoptotic nuclei, the mutant has significantly more nuclei with fragmented DNA than the control. Furthermore, as expected if only photoreceptors are degenerating, the increase in apoptotic nuclei is seen only in the photoreceptor layer (Table 2). Interestingly, the pob mutation is not in the one known gene specific to red cones, red opsin [20]. In addition to these morphologically normal visual mutants, we and others have identified mutants with gross morphological abnormalities in the eye [24,25]. These mutants appear at a higher frequency than the behavioral mutants ( 1 morphological mutant/15 F2 families vs 1 behavioral mutant/50 F2 families) and most of these ‘small eye’ mutants have non-specific retinal degeneration and degeneration in the brain. A subset of the morphological mutants appear to have a cell-type specific degeneration. One mutant, archie, selectively loses most ganglion and some amacrine cells [24]. Several additional mutants lose all photoreceptor cells, whereas the other retinal layers in these mutants do not appear to degenerate. To date, no retinal mutants that appear to have a defect in cell formation (i.e. the mutant lacks a cell-type because it does not form) have been identified, although Malicki and colleagues have isolated a class of mutations that lead to patterning defects within the retina and brain [25]. In conclusion, over the last few years we have initiated a genetic screen to identify zebrafish mutants primarily with subtle retinal defects. Although this work is still in its infancy, we have clearly demonstrated that it is possible to identify mutants with a variety of retinal defects. The long-term goal is to clone the genes that have been altered in these fish and to study the gene products using molecular and biochemical methods. Linkage maps of the zebrafish genome are available [11–13] and thus positional cloning should be feasible. Furthermore, a procedure for generating retro-
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339
1338
Fig. 1. Representative ERG responses from 5 dpf nrb and sibling responder larvae. The responder ERG is only reduced slightly by background illumination (middle recording, top), whereas the ERG response of nrb larvae is reduced substantially by the identical procedure (middle recording, bottom). Tha a-wave (a) and b-wave (b) are indicated in the left recording on the top. The procedure for recording ERGs was as described previously ([20]) except the fish were not dark-adapted prior to the experiment and the light flash was B 1 msec in duration (Strobe flash at 1/16th power: Canon Speedlite 300 TL). See text for more details.
viral-insertion mutants has recently been published [26,27], and genetic screens using this approach are in progress. Finally, we hope to learn information about retinal function from characterization of the various mutants. For example, it is unclear whether red cone
photoreceptors have unique molecules in addition to red opsin. By isolating RNA that is expressed in wildtype retinas but not in pob retinas we may identify molecules specific to red cones other than red opsin.
References Table 2 Number of apoptotic cells/retina in each retinal nuclear layer Wild-type
pob
5.5 ganglion cells 9.5 inner nuclear layer cells 1 photoreceptor cell
6 ganglion cells 15.5 inner nuclear layer cells 178.5 photoreceptor cells
Apoptotic cells were identified in wild-type (n = 2) and pob (n =2) retinas at 5 dpf using the in situ Apoptosis Detection Kit (Oncor Inc.) and the protocol as described ([23]). After staining, larvae were dehydrated with a series of ethanol solutions and embedded in epon-araldite ([18]). 5 mm sections were examined using a light microscope and the apoptotic cells/section counted.
[1] Benzer, S. The 1990 Helmerich Lecture: the fly and eye. In: Lam DM, Shatz CJ, editors. Development of the Visual System. Cambridge: MIT Press, 1991. [2] Pak WL. Drosophila in vision research. The Friedenwald Lecture. Invest Ophthalmol Vis Sci 1995;36:2340 – 57. [3] Yarfitz S, Hurley JB. Transduction mechanisms of vertebrate and invertebrate photoreceptors. J Bio Chem 1994;269:14329– 32. [4] Clark DT, Visual responses in the developing zebrafish (Brachydanio rerio). In: Biology. Eugene: University of Oregon, 1981. [5] Driever W, Stemple D, Schier A, Solnica-Krezel L. Zebrafish: genetic tools for studying vertebrate development. Trends Genet 1994;10:152 – 9.
S.E. Brockerhoff et al. / Vision Research 38 (1998) 1335–1339 [6] Mullins MC, Hammerschmidt M, Haffter P, Nusslein-Volhard C. Large-scale mutagenesis in the zebrafish: in search of genes controlling development in a vertebrate. Curr Biol 1994;4:189 – 202. [7] Mullins MC, Nusslein-Volhard C. Mutational approaches to studying embryonic pattern formation in the zebrafish. Curr Opin Genet Dev 1993;3:648–54. [8] Solnica-Krezel L, Schier AF, Driever W. Efficient recovery of ENU-induced mutations from the zebrafish germline. Genetics 1994;136:1401– 20. [9] Driever W, Solnica-Krezel L, Schier AF, et al. A genetic screen for mutations affecting embryogenesis in zebrafish. Development 1996;123:37 – 46. [10] Haffter P, Granato M, Brand M, et al. The identification of genes with unique and essential functions in the development of the zebrafish Danio rerio. Development 1996;123:1–36. [11] Johnson SL, Gates MA, Johnson M, et al. Centromere-linkage analysis and consolidation of the zebrafish genetic map. Genetics 1996;142:1277– 88. [12] Knapik EW, Goodman A, Atkinson OS, et al. A reference cross DNA panel for zebrafish (Danio rerio) anchored with simple sequence length polymorphisms. Development 1996;123:451 – 60. [13] Postlethwait JH, Johnson SL, Midson CN, et al. A genetic linkage map for the zebrafish. Science 1994;264:699–703. [14] Sz’el A, Rohlich P, Caff’e AR, Juliusson B. Unique topographic separation of two spectral classes of cones in the mouse retina. J Comp Neurol 1992;325:327–42. [15] Robinson J, Schmitt EA, Harosi FI, Reece RJ, Dowling JE. Zebrafish ultraviolet visual pigment: absorption spectrum, sequence, and localization. Proc Nat Acad Sci USA 1993;90:6009 – 12. [16] Kljavin IJ. Early development of photoreceptors in the ventral retina of the zebrafish embryo. J Comp Neurol 1987;260:461 –
71. [17] Branchek T, Bremiller R. The development of photoreceptors in the zebrafish, Brachydanio rerio. I. Structure. J Comp Neurol 1984;224:107 – 15. [18] Schmitt EA, Dowling JE. Early eye morphogenesis in the zebrafish, Brachydanio rerio. J Comp Neurol 1994;344:532–42. [19] Brockerhoff SE, Hurley JB, Janssen-Bienhold U, Neuhauss CF, Driever W, Dowling JE. A behavioral screen for isolating zebrafish mutants with visual system defects. Proc Nat Acad Sci USA 1995;92:10545– 9. [20] Brockerhoff SE, Hurley JB, Niemi GA, Dowling JE. A new form of inherited red-blindness identified in zebrafish. J Neurosci 1997;17:4236 – 42. [21] Dowling JE, The retina: an Approachable Part of the Brain. Cambridge, MA: Belknap Press of Harvard University Press, 1987. [22] Lall AB, Brockerhoff SE, Dowling JE. Electroretinographic characteristics of nrc mutant larvae of zebrafish. Invest. Ophthal. Vis. Sci. Florida, May 11 – 15. ARVO c 2889, 1997. [23] Abdelilah S, Mountcastle-Shah E, Harvey M, et al. Mutations affecting neural survival in the zebrafish Danio rerio. Development 1996;123:217 – 27. [24] Fadool JM, Brockerhoff SE, Hyatt GA, Dowling JE. Mutations affecting eye morphology in the developing zebrafish (Danio rerio). Dev Genet 1997;20:1 – 8. [25] Malicki J, Neuhauss SC, Schier AF, et al. Mutations affecting development of the zebrafish retina. Development 1996;123:263 – 73. [26] Gaiano N, Allende M, Amsterdam A, Kawakami K, Hopkins N. Highly efficient germ-line transmission of proviral insertions in zebrafish. Proc Nat Acad Sci USA 1996;93:7777 – 82. [27] Gaiano N, Amsterdam A, Kawakami K, Allende M, Becker T, Hopkins N. Insertional mutagenesis and rapid cloning of essential genes in zebrafish. Nature 1996;383:829 – 32.
. .
1339