letter
© 1999 Nature America Inc. • http://genetics.nature.com
Retinopathy and attenuated circadian entrainment in Crx-deficient mice
© 1999 Nature America Inc. • http://genetics.nature.com
Takahisa Furukawa1,4, Eric M. Morrow1, Tiansen Li2, Fred C. Davis3 & Constance L. Cepko1
Crx, an Otx-like homeobox gene, is expressed specifically in the photoreceptors of the retina and the pinealocytes of the pineal gland1,2. Crx has been proposed to have a role in the regulation of photoreceptor-specific genes in the eye and of pineal-specific genes in the pineal gland. Mutations in human CRX are associated with the retinal diseases, cone-rod dystrophy-2 (adCRD2; refs 3–5), retinitis pigmentosa5 (RP) and Leber congenital amaurosis5,6 (LCA), which all lead to loss of vision. We generated mice carrying a targeted disruption of Crx. Crx–/– mice do not elaborate photoreceptor outer segments and lacked rod and cone activity as assayed by electroretinogram (ERG). Expression of several photoreceptor- and pineal-specific genes was reduced in Crx mutants. Circadian entrainment was also affected in Crx–/– mice.
To create a targeted disruption in Crx, we deleted a restriction fragment encoding the homeodomain (Fig. 1a). We used Southern blots to confirm the deletion (Fig. 1b,c). Both Crx+/– and Crx–/– mice showed no gross morphological abnormalities, and are viable and fertile. Primer pairs derived from the sequence of Crx exon 3 amplified a 249-bp product from cDNA in wild-type animals. In contrast, we detected no amplification in Crx–/– mice (Fig. 1d). We analysed retinal sections from two-week and three-week wild-type, Crx+/– and Crx–/– mice (Fig. 2). There were no differ-
ences among Crx genotypes at post-natal day (P) 10 (data not shown), before outer segments begin to form. Outer segments were present in wild-type mice at P14 and P21 (Fig. 2a,d), but Crx–/– mice at these ages showed no outer segment structures (Fig. 2c,f ). At P21, Crx–/– mice had a thin outer nuclear layer (ONL) and dislocated photoreceptor cells just beneath the pigment epithelium (PE; Fig. 2f). Crx+/– mice developed outer segments, but their length was shorter than those in wild type at P14 (Fig. 2b). At P21 there were no differences between Crx+/– and wild-type mice (Fig. 2d,e). We recorded ERGs for wild-type, Crx+/– and Crx–/– mice (Fig. 3a–c). The rod ERGs showed that one-month-old Crx–/– mice have less than 1.0% of normal rod activity (Fig. 3a), with both the a-wave and the b-wave completely absent. We did not detect the cone ERGs (Fig. 3a), indicating that cone function was reduced by at least 90%. In Crx+/– mice, rod ERG amplitudes were reduced compared with those of wild type (Fig. 3a). The cone ERG was not detected in heterozygous animals at our routine light intensity (Fig. 3e), although an ERG was detected at higher intensity (data not shown). We next conducted ERG analyses of older Crx mutant mice (up to six months; Fig. 3b,c). The rod ERG of two-month-old Crx+/– mice showed reduced activity compared with that of
a
b
c
d
Fig. 1 Targeted mutation at the Crx locus. a, Strategy for the targeted deletion of Crx. The filled box indicates the homeodomain coding region. The open box indicates the exons. Exon 1 is located ∼10 kb upstream of exon 2 (not shown). Exon 3 and part of exon 4 are replaced with the PGK-neo cassette in the targeting vector, and the EcoRI sites and a BamHI site are removed. The probes (a and b) used for Southern-blot analysis are shown below the wildtype genomic map. E, EcoRI; H, HindIII; B, BamHI; A, AvrII; RV, EcoRV. b,c, Southern-blot analysis of genomic DNA isolated from mouse tails. EcoRI-digested tail DNA was hybridized with probe a, and BamHIdigested tail DNA was hybridized with probe b. Probe a detects 10-kb wild-type and 21-kb mutant bands. Probe b detects 2.0-kb wild-type and 5.5-kb mutant bands. d, RT-PCR analysis of total RNA extracted from retina derived from wild-type and Crx–/– mice, using primers specific for Crx and Gapd. The positions of primers in exons 3 and 4 of Crx are shown in (a).
1Department of Genetics and Howard Hughes Medical Institute, Harvard Medical School, Boston, Massachusetts, USA. 2The Berman-Gund Laboratory for the Study of Retinal Degenerations, Harvard Medical School, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts, USA. 3Department of Biology, Northeastern University, Boston, Massachusetts, USA. 4Present address: Center for Developmental Biology, The University of Texas, Southwestern Medical
Center at Dallas, Dallas, Texas, USA. Correspondence should be addressed to C.L.C. (
[email protected]) or T.F. (
[email protected]).
466
nature genetics • volume 23 • december 1999
letter
© 1999 Nature America Inc. • http://genetics.nature.com
© 1999 Nature America Inc. • http://genetics.nature.com
a
b
Fig. 2 Retinal histology of Crx mutant mice. a,b,c, Plastic sections (0.5 µm) stained with toluidine blue from P14 Crx wild-type, heterozygous and homozygous mice, respectively. Crx–/– mice do not develop outer segments, and the outer segments of Crx+/– mice are shorter than those of wild type. PE, pigment epithelium; OS, outer segment; IS, inner segment; ONL, outer nuclear layer. d,e,f, P21 wild-type, heterozygous and homozygous Crx mice, respectively. Crx–/– mice do not develop outer segments, and degeneration of photoreceptor cells is evident by this stage. The outer segments of Crx+/– mice are similar to those of wild type at this stage. Scale bars, 10 µm.
c
We examined the expression of photoreceptor-specific genes in retina of P10 Crx–/– mice (Fig. 4). At P10 expression of genes encoding rhodopsin, cone opsins, rod transducin α-subunit, cone arrestin and recoverin were reduced in Crx–/– mice. The level of RNAs encoding cone transducin, rod PDE-β, cone PDEα´, rod cGMP-gated channel (RNG), rod arrestin and peripherin were slightly reduced. In contrast, genes encoding cone cGMPgated channel (CNG) and Neurod were upregulated in Crx–/– e d mutants. The gene encoding blue opsin was slightly reduced in Crx+/– mice (Fig. 4). The gene encoding green/red opsin was f upregulated, although it was absent in Crx–/– mice (Fig. 4). The pineal gland from Crx–/– mice did not show any gross abnormalities at one month (Fig. 5a–f). Arylalkylamine N-acetyltransferase (encoded by Aanat) and hydroxyindole-O-methyltransferase (encoded by Hiomt), key enzymes for melatonin synthesis, contain Crx-binding sites in their regulatory regions7. We examined expression of Aanat transcripts in the pineal gland by RT-PCR analysis. Primer pairs for Nat amplified a 488-bp product in wild-type animals, but only weak amplification was detected in Crx–/– mice (Fig. 5g). We also examined the expression of phototransduction component genes encoding rhodopsin, blue cone opsin, cone arrestin and Cng, which are expressed at a wild-type mice (Fig. 3b). The cone ERG of Crx+/– mice had relatively high level in the rodent pineal gland8. We amplified PCR developed by two months (Fig. 3b). Neither rod nor cone ERGs fragments of 518 and 401 bp with primer pairs for genes encoding of six-month Crx+/– mice were different a from those of wild-type mice (Fig. 3c). We also examined retinal sections of these mice (Fig. 3d,e), and found that there were no differences between retina of six-month Crx+/– and wild-type mice. At two months, Crx–/– mice had 8–10 rows of nuclei in the ONL (Fig. 3f) compared with 14–16 rows of nuclei in wild type (data not shown). By six months, the thickness of the ONL in Crx–/– mice was reduced to 1–3 rows of nuclei (Fig. 3g). Fig. 3 Phenotypes of Crx-mutant mice at 1–6 months. a, Dark-adapted (rod) and light-adapted (cone) ERG responses from four-week wild-type, Crx+/– and Crx–/– mice. ERG potentials were recorded from wild-type, heterozygous and homozygous mice. Response amplitudes (µV) of b-waves are indicated below waves with s.e.m. No significant amplitude was detected for ERGs from homozygous mice. a, a-wave; b, b-wave. b,c, ERGs from two-month (b) and sixmonth (c) mice. d,e, Photoreceptor morphology of six-month wild-type and Crx+/– mice. PE, pigment epithelium; OS, outer segment; IS, inner segment; ONL, outer nuclear layer. f,g, Photoreceptor degeneration of Crx–/– mice. Haematoxylin-and-eosin-stained retinal sections of two-month (f) and six-month Crx–/– (g) mice are shown. Six-month Crx–/– retina contains a photoreceptor layer of only 2–4 rows of cells. The inner plexiform layer (ipl) is also thinner than that of wild type. At this time, degeneration was more severe in the central region of the retina than in the periphery, with the central retina showing no ONL (data not shown). opl, outer plexiform layer; ipl, inner plexiform layer, gcl, ganglion cell layer. Scale bars, 10 µm.
nature genetics • volume 23 • december 1999
b
d
e
c
f
g
467
© 1999 Nature America Inc. • http://genetics.nature.com
letter
© 1999 Nature America Inc. • http://genetics.nature.com
cone arrestin
Rbp3
Pdebc
Rhok
Abc10
Gcp
Cncg
Rcvrn
Neurod
Gnat1
Cnga3
Rom1
Gapd
Gnat2
Sag
Prph2
Rho
Pdeb
Bcp
Fig. 4 Levels of expression of photoreceptor-specific genes in Crx mutant retina. Opsin (rhodopsin (Rho) and cone opsins (Bcp, Gcp)), transducin (rod (Gnat1) and cone (Gnat2)), rod phosphodiesterase (Pde) β-subunit (Pdeb), cone phosphodiesterase α-subunit (Pdebc), cGMP-gated channel (rod (Cncg) and cone (Cnga3)), arrestin (rod (Sag) and cone), rhodopsin kinase (Rhok) and recoverin (Rcvrn) are involved in the phototransduction pathway. Rom1 (Rom1) and peripherin (Prph2) are structural components of photoreceptor discs in the outer segment25. Abcr (Abc10) is an ATP-dependent transporter expressed in rods26. Neurod is expressed in a subset of differentiated photoreceptors in the retina27. Northern-blot analysis of total RNA (5 µg) extracted from P10 retinae of wild-type, Crx+/– and Crx–/– mice is shown. Probes used for hybridization were specific for Rho (8,301–8,800 bp), Bcp (blue), Gcp (green/red), Gnat1 (7–581 bp), Gnat2 (408–941 bp), Pdeb (2,154–2,663 bp), Pdebc (1,861–2,352 bp), Cncg (681–1,407 bp), Cnga3 (944–1,646 bp), Sag, cone arrestin (1–279 bp), Rhok (121–1,044 bp), Rcvrn (13–884 bp), Rom1 (695–1,847 bp), Prph2, Rbp3 (interphotoreceptor retinal-binding protein, Irbp), Abc10 (6,601–7,200 bp), Neurod and Gapd. The relative normalized intensity of each band measured by densitometer is indicated.
mouse rhodopsin and blue cone opsin, respectively, in wild-type animals. We did not detect these PCR products in Crx–/– mice (Fig. 5g). Primer pairs from genes encoding CNG or cone arrestin amplified PCR products with similar intensities in both wild-type and Crx–/– mice (data not shown). The pineal gland functions as a photoreceptive organ in some lower vertebrates. Our data suggest that Crx may have a role in the development of the pineal phototransduction pathway in lower vertebrates.
g
Aanat
a
b Rho
To examine photoentrainment activity in Crx–/– mice, we recorded the activity of wild-type (n=6) and Crx–/– (n=6) mice running on an exercise wheel (Fig. 6). All mice showed robust 24hour rhythms in activity during entrainment (Fig. 6a–d) and in constant darkness (data not shown). All mice showed the most activity during the dark interval of a 24-hour period (Fig. 6a–d). The percentage of total activity that occurred at night, however, was significantly different between wild-type and Crx–/– mice (95.9±0.84% s.e.m. for wild-type versus 86.3±2.80% for Crx–/–; P=0.015, Mann Whitney test). All mice reentrained to the light-dark cycle following an advance of four hours in the cycle (Fig. 6a–d). The number of days required to complete the shift, however, differed between wild-type and Crx–/– mice (Fig. 6e). Crx–/– mice lack outer segments, most likely caused by a reduction in photoreceptor-specific molecules including rhodopsin and cone opsins, or a mild reduction in a key molecule, perhaps peripherin. This is supported by the absence of outer segments in rhodopsin-deficient
Bcp
c
d
e
f
468
Gapd
Fig. 5 Pineal gland of Crx-mutant mice. a,b, Pineal glands viewed from the ventral side. Pineal glands of one-month wild-type (a) and Crx–/– (b) mice are indicated by arrows. Haematoxylin-and-eosinstained pineal gland of wild type (c) and Crx–/– (d) mice are shown. e,f, Higher magnification of (c) and (d), respectively. g, RT-PCR analysis of total RNA extracted from pineal glands derived from wild-type and Crx–/– mice using primers specific for N-acetyltransferase (NAT) gene, rhodopsin, cone opsin (blue) and Gapd. Examination of the production of melatonin in Crx mutant mice is not possible because most of the laboratory strains have mutations in Aanat or Hiomt.
nature genetics • volume 23 • december 1999
letter
© 1999 Nature America Inc. • http://genetics.nature.com
e
© 1999 Nature America Inc. • http://genetics.nature.com
Fig. 6 Effect of Crx genotype on photoentrainment in mice. a–d, Effect on phase shifts in response to change of light-dark cycle. Light-dark cycle was advanced by 4 h on the thirteenth day of measurement, indicated by the arrow. Representative double-plotted actograms of Crx+/+ (a,b) and Crx–/– (c,d) mice are shown. e, The average number of days (±s.e.m.) for the mice to be re-entrained to the advanced light/dark cycle (wild type, 3.33d±0.21; Crx–/–, 7.00d±1.24, P=0.04, Mann Whitney test). In an additional series of experiments in which mice were released into constant darkness, all of the mice freeran from phases near the new time of lights-off. The average phases determined from the subsequent freerunning rhythms were not significantly different between the groups (wild type, 15.14 ±0.29 h and Crx–/–, 14.20 ±1.02 h). The average freerunning periods over the first 10 d in constant darkness were not significantly different (wild type, 23.56±0.12 h and Crx–/–, 23.72±0.13 h). The deletion of Crx did not affect the persistence or the average period of circadian rhythms in constant darkness. This is consistent with the observation that Crx mRNA was not detected in the SCN of wild-type mice at night or during the day (data not shown).
a
c
b
d
mice9 and Prph2–/– mice10, in which peripherin is mutated. Crx is also essential for the phototransduction pathway in both rods and cones, which can be independent of outer segment formation to some extent. In Prph2–/– mice, which lack all the structures of rod and cone outer segments10, there are reduced ERG potentials11. Crx is not essential for photoreceptor cell fate determination, as there are cells that express some photoreceptor-specific genes in Crx–/– animals. Our results show that Crx is required for high expression of many photoreceptor-specific genes at P10. These genes, however, were still detectable in Crx mutants, except that encoding green/red cone opsin. The requirement for Crx in transcription of the cone opsin loci in mice supports the notion that the human red/green locus requires CRX activity at the CRX-binding element (CBE) in the locus control region1. The proximal regulatory region of Irbp, which contains a CBE, is critical for the photoreceptor-specific expression of Irbp (refs 12,13). Thus, it was an unexpected finding that the expression of Irbp was not affected in Crx–/– mice. Perhaps Crx is redundant with another factor such as Otx2 (ref. 14). Some genes specific to the photoreceptor, such as Cnga3 and Neurod, were upregulated in Crx–/– mice. Crx thus can work directly or indirectly as a negative regulator. The gene encoding blue cone opsin was slightly downregulated in Crx+/– animals, showing haploinsufficiency of Crx in the presumed transactivation of the blue cone opsin. It is notable that the green cone opsin gene was upregulated in Crx+/– mice, considering its expression was lost in Crx–/– animals. Mutations of human CRX are associated with three human retinal diseases, adCRD2, RP and LCA. Crx–/– mice are a model for LCA in that they showed flat rod and cone ERGs (ref. 15). It is not clear whether the CRX alleles in adCRD2 patients are loss-offunction or dominant-negative alleles. Further analysis of older Crx+/– mice may help to clarify this. Attenuation of photoentrainment of Crx mutant mice is subtle, but might indicate a role for rod or cone photoreceptors in this type of non-visual photoreception. Because rod photoreceptors are especially affected in Pdeb–/– mice, which do not show attenuated entrainment, the phenotype of Crx mutant mice may be due to the absence of cone function. The role of cones in circadian entrainment is, however, still controversial16. Photoentrainment was not eliminated in Crx–/– mice. This indicates that photoreception can occur in the absence of detectable ERGs for nature genetics • volume 23 • december 1999
both rods and cones. There have been reports of the expression of opsin-like photopigments in horizontal cells17, pigment epithelium18 and subsets of horizontal and amacrine cells19. Crx might be expressed at a low level in other cells, and Crx mutation may affect their function. An effect of Crx on these cells may be indirect. It remains possible, therefore, that neither rods nor cones are required for photoentrainment and that the critical effect of the Crx mutation is on other cells. Recent reports on photoentrainment of non-rod, non-cone mice may support this idea20,21.
Methods Generation of Crx-mutant mice. We obtained the Crx genomic clone from a 129SVJ mouse library (Stratagene). We subcloned 5.5-kb HindIII-EcoRI and 7.8-kb EcoRI-EcoRV fragments into the pPNT vector and transfected the TC1 ES cell line22 with the Crx targeting vector. The culture, electroporation and selection of TC1 were carried out as described22. Embryonic stem cells heterozygous for the targeted mutation were microinjected into C57BL/6 blastocysts to obtain germline transmission. Genotype analysis. For PCR genotype analysis, we detected the wild-type Crx allele using a sense primer (5´–AGCGGCGGGAGCGGACCACAT TCA–3´) and an antisense primer (5´–TGGACCCTGGACTCAGGCAGATTG–3´). These primer pairs amplify a 135-bp fragment from wild-type Crx. DNA was also amplified using primers (sense primer, 5´–GCC GGGGCAGGATCTCCTGTCATCTCACCT–3´; antisense primer, 5´–CAG GCATCGCCATGGGTCACGACGAGATCCT–3´) to detect neo in the mutant Crx allele. In this case, we detected a 285-bp fragment in mice heterozygous or homozygous for the mutant Crx allele, whereas no signal was detected in wild-type mice. PCR conditions were 94 °C, 2 min followed by 94 °C, 1 min; 60 °C, 1 min; 72 °C, 1 min; for 36 cycles. RT-PCR for retina and pineal gland. Total RNA was extracted from retina or pineal gland (collected between 18:00 and 19:00) of one-month Crx mice. After RNA was treated with RNase-free DNase (Boehringer), cDNAs were synthesized and RT-PCR was conducted. RT-PCR primers for detection of Crx mRNA were 5´–TGGAGGAGCTGGAGGCCCTGTTTGCCAAGAC–3´ and 5´–CCAAAGGATCTGTACAAACATCTGTAGAG–3´. Product size was 258 bp. PCR conditions were 94 °C, 1 min followed by 94 °C, 30 s; 60 °C, 50 s; 72 °C, 50 s; 40 cycles. We amplified Aanat by nested RT-PCR. The first set of primer pairs for Aanat transcripts was 5´–CCATGTTGAACATCAACTCCCTGA–3´ and 5´–GTCAGCAGCCGCTGTTCCTGCGCAG –3´. The second set of primers was 5´–TTCCTAGGCTGCCAGCGGCG CCACA–3´ and 5´–GAGCCCACGGTGATGGCAC ATGGGCCCAC–3´, which amplified a 488-bp product. The first set of primer pairs for mouse 469
letter
© 1999 Nature America Inc. • http://genetics.nature.com
rhodopsin transcripts was 5´–GTGGTCCTGGCCATTGAGCGCTACG–3´ and 5´–TCGTCATCTCCCAGTGGATTCTTGCC–3´. The second set of primers was 5´–ATCCCTGAGGGCATGCAATGTTCATG–3´ and 5´–CTG CTTGTTCAACATGATGTAGATGAC–3´, which amplified a 518-bp product. The first set of primer pairs for mouse blue cone opsin transcripts was 5´–TCACAGTCTTCATCGCCAGCTGTCAC–3´ and 5´–TGAGAGCCAGACACGTCAGA TTCGTC–3´. The second set of primers was 5´–ATCCGCTTCAACTCC AAGCATGCAC–3´ and 5´–CGATTGTTGACCATGTACATGGCCAG–3´, which amplified a 401-bp product. PCR conditions for first- or second-round PCR were as described above. We confirmed the amplified products by subcloning and sequencing. Histology. We perfused mice whose eyes were to be used for light and EM microscopy with a mixture of 2.5% glutaraldehyde, 2% paraformaldehyde in phosphate buffer (0.1 M, pH 7.4). Eyes were removed, additionally fixed overnight at 4 °C and, after trimming, processed for light microscopy. Eyes were post-fixed in osmium tetroxide and embedded in Epon. We cut sections (0.4 µm) from these blocks and stained with toluidine blue for light microscopy.
© 1999 Nature America Inc. • http://genetics.nature.com
described23.
At least three wild-type, ERG. We conducted ERG testing as heterozygous and homozygous Crx mice were evaluated, respectively. For each animal analysed, four separate responses were generated and computer-averaged. Representative tracings for each genotype are shown. Northern-blot analysis. We prepared total RNA from P10 retina and electrophoresed total RNA (5 µg). These experiments were carried out at least two times. We measured the intensity of each band by densitometer and averaged the values. Some probes were rat or human cDNA that crosshybridize specifically with mouse counterparts8.
1.
2. 3.
4.
5.
6. 7.
8.
9. 10. 11. 12.
13. 14.
Furukawa, T., Morrow, E.M. & Cepko, C.L. Crx, a novel otx-like homeobox gene, shows photoreceptor-specific expression and regulates photoreceptor differentiation. Cell 91, 531–541 (1997). Chen, S. et al. Crx, a novel Otx-like paired-homeodomain protein, binds to and transactivates photoreceptor cell-specific genes. Neuron 19, 1017–1030 (1997). Freund, C.L. et al. Cone-rod dystrophy due to mutations in a novel photoreceptorspecific homeobox gene (CRX) essential for maintenance of the photoreceptor. Cell 91, 543–553 (1997). Swain, P.K. et al. Mutations in the cone-rod homeobox gene are associated with the cone-rod dystrophy photoreceptor degeneration. Neuron 19, 1329–1336 (1997). Sohocki, M.M. et al. A range of clinical phenotypes associated with mutations in CRX, a photoreceptor transcription-factor gene. Am. J. Hum. Genet. 63, 1307–1315 (1998). Freund, C.L. et al. De novo mutations in the CRX homeobox gene associated with Leber congenital amaurosis. Nature Genet. 18, 311–312 (1998). Li, X. et al. A pineal regulatory element (PIRE) mediates transactivation by the pineal/retina-specific transcription factor CRX. Proc. Natl Acad. Sci. USA 95, 1876–1881 (1998). Blackshaw, S. & Snyder, S.H. Developmental expression pattern of phototransduction components in mammalian pineal implies a light-sensing function. J. Neurosci. 17, 8074–8082 (1997). Humphries, M.H. et al. Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nature Genet. 15, 216–219 (1997). Sanyal, S. & Jansen, H.G. Absence of receptor outer segments in the retina of rds mutant mice. Neurosci Lett. 21, 23–26 (1981). Reuter, J.H. & Sanyal, S. Development and degeneration of retina in rds mutant mice: the electroretinogram. Neurosci. Lett. 48, 231–237 (1984). Liou, G.I. et al. Retina-specific expression from the IRBP promoter in transgenic mice is conferred by 212 bp of the 5-’flanking region. Biochem. Biophys. Res. Commun. 181, 159–165 (1991). Boatright, J.H. et al. A major cis activator of the IRBP gene contains CRX-binding and Ret-1/PCE-I elements. Mol. Vis. 3, 15 (1997). Bovolenta, P., Mallamaci, A., Briata, P., Corte, G. & Boncinelli, E. Implication of
470
Analysis of circadian rhythms. We monitored activity rhythms as described24. We kept 12 mice (6 wild type and 6 Crx–/–, 3 of each sex per group, 4 months old) on L:D, 12:12 (lights on at 0700 EST), the same cycle on which they were born and raised. After 12 d, we advanced the light/dark cycle 4 h so that lights were on at 0300 and off at 1500. The number of days required to re-entrain to the shifted light/dark cycle was determined as the number of activity onsets before onsets again consistently occurred with a period of 24 h, rather than shorter than 24 h as occurred during the shift forward. This analysis was performed without knowledge of the genotypes of experimental animals. GenBank accession numbers. Gnat1, U38504; Gnat2, L10666; Pdeb, X87952; Cncg, M84742; CNGA3, AF065314; Rcvrn, X66196; Rhok, U63971; Sag, M24086; rat cone arrestin, U03628; Rbp3, X69523; Neurod, U28888; Rom1, M96760; Prph2, X14770; human cone phosphodiesteraseα’, U31973; Abc10, AF000149. Acknowledgements
We thank P. Leder and C. Doherty for helping us generate mutant mice; E. Raviola and S. Ito for advice on light and electron microscopy of retinal sections; J. Nathans and T. Shinohara for cDNAs encoding mouse cone opsin (blue, green/red) and rod arrestin, respectively; and S. Fields-Berry, J. Zitz, M. Samson and H. Regan for technical assistance. This work was supported by the Howard Hughes Medical Institute.
Received 27 April; accepted 28 September 1999.
15.
16. 17.
18.
19. 20.
21. 22. 23.
24.
25.
26. 27.
OTX2 in pigment epithelium determination and neural retina differentiation. J. Neurosci. 17, 4243–4252 (1997). Foxman, S.G., Heckenlively, J.R., Bateman, J.B. & Wirtschafter, J.D. Classification of congenital and early onset retinitis pigmentosa. Arch. Ophthalmol. 103, 1502–1506 (1985). Foster, R.G. Shedding light on the biological clock. Neuron 20, 829–832 (1998). Provencio, I., Jiang, G., De Grip, W.J., Hayes, W.P. & Rollag, M.D. Melanopsin: an opsin in melanophores, brain, and eye. Proc. Natl Acad. Sci. USA 95, 340–345 (1998). Sun, H., Gilbert, D.J., Copeland, N.G., Jenkins, N.A. & Nathans, J. Peropsin, a novel visual pigment-like protein located in the apical microvilli of the retinal pigment epithelium. Proc. Natl Acad. Sci. USA 94, 9893–9898 (1997). Soni, B.G., Philp, A.R., Foster, R.G. & Knox, B.E. Novel retinal photoreceptors. Nature 394, 27–28 (1998). Lucas, R.J., Freedman, M.S., Munoz, M., Garcia-Fernandez, J.M. & Foster, R.G. Regulation of the mammalian pineal by non-rod, non-cone, ocular photoreceptors. Science 284, 505–507 (1999). Freedman, M.S. et al. Regulation of mammalian circadian behavior by non-rod, non-cone, ocular photoreceptors. Science 284, 502–504 (1999). Deng, C., Wynshaw-Boris, A., Zhou, F., Kuo, A. & Leder, P. Fibroblast growth factor receptor 3 is a negative regulator of bone growth. Cell 84, 911–921 (1996). Li, T. et al. Effect of vitamin A supplementation on rhodopsin mutants threonine17→methionine and proline-347→serine in transgenic mice and in cell cultures. Proc. Natl Acad. Sci. USA 95, 11933–11938 (1998). Grosse, J. & Davis, F.C. Melatonin entrains the restored circadian activity rhuthms of syrian hamsters bearing fetal suprachiasmatic nucleus grafts. J. Neurosci. 18, 8032–8037 (1998). Goldberg, A.F. & Molday, R.S. Defective subunit assembly underlies a digenic form of retinitis pigmentosa linked to mutations in peripherin/rds and rom-1. Proc. Natl Acad. Sci. USA 93, 13726–13730 (1996). Sun, H. & Nathans, J. Stargardt’s ABCR is localized to the disc membrane of retinal rod outer segments. Nature Genet. 17, 15–16 (1997). Morrow, E.M., Furukawa, T., Lee, J.E. & Cepko, C.L. NeuroD regulates cell fate determination in the developing neural retina. Development 126, 23–36 (1999).
nature genetics • volume 23 • december 1999
