letter
1
2
3
1
4
4
3
1
Color vision is facilitated by distinct populations of cone photoreceptors in the retina. In rodents, cones expressing different opsin photopigments are sensitive to middle (M, ‘green’) and short (S,
‘blue’) wavelengths, and are differentially distributed across the retina 1,2 . The mechanisms that control which opsin is expressed in a particular cone are poorly understood 2,3 , but previous in vitro studies implicated thyroid hormone in cone differentiation 4,5
Thyroid hormone receptor
β
2 (TR
β
2) is a ligand-activated tran-
.
scription factor that is expressed in the outer nuclear layer of the embryonic retina 6,7 . Here we delete Thrb (encoding Tr
β
2) in mice, causing the selective loss of M-cones and a concomitant increase in S-opsin immunoreactive cones. Moreover, the gradient of cone distribution is disturbed, with S-cones becoming widespread across the retina. The results indicate that cone photoreceptors throughout the retina have the potential to follow a default Scone pathway and reveal an essential role for Tr
β
2 in the commitment to an M-cone identity. Our findings raise the possibility that
Thrb mutations may be associated with human cone disorders 8 .
a b
The thyroid hormone receptors Tr
α
1 and Tr
β are encoded by the genes Thra and Thrb, respectively. Thrb, through differential promoter and exon usage, encodes two receptor variants,
Tr
β
1 and Tr
β
2, which have identical DNA binding and thyroid hormone (T3) binding domains, but divergent amino termini 6,7,9 (Fig. 1d). Tr
β
1 and Tr
β
2 have subtle differences in their transactivation properties in vitro 10–12 . Moreover, Tr
β
1 is expressed in a range of tissues whereas Tr
β
2 is restricted to the outer nuclear layer (ONL) of the retina (which contains the developing photoreceptors ear 6,13
7 ), the pituitary gland and the inner
, indicating that Tr
β
2 has specialized functions.
To determine the role of Tr
β
2, we deleted part of the Tr
β
2-specific exon and replaced it with a lacZ cassette (Fig. 1a). This
Thrb tm2Df
Thrb tm1Df mutant allele is distinct from the previously described allele, which deleted both Tr
β
1 and Tr
β
2 (ref. 14).
Thrb tm2Df/tm2Df (hereafter Thrb –/– ) mice were viable. Southern-blot analyses confirmed the presence of the mutation in Thrb –/–
Thrb +/tm2Df (hereafter Thrb +/– and
) mice, whereas the intact exon was detected in only wild-type and Thrb +/– mice (Fig. 1b). To demonstrate the deletion of Tr
β
2, we analysed pituitary and eye RNA samples by RT–PCR using primers specific for Tr
β
1 and Tr
β
2 transcripts (Fig. 1d). In wild-type tissue, two bands corresponding to Tr
β
2 and Tr
β
1 were detected, whereas in Thrb –/– mice, only the
Tr
β
1-specific band was present, confirming the loss of Tr
β
2 mRNA.
Northern-blot analysis with a Tr
β
1-specific probe showed no obvious perturbation of Tr
β
1 mRNA expression in eye (Fig. 1c).
Northern-blot analysis revealed prominent expression of Tr
β
2 mRNA in mouse eye development (Fig. 2a), similar to the pattern described in chick 7,15 . Tr
β
2 mRNA levels peaked at approximately embryonic day (E) 17.5 and declined in the postnatal period. Tr
β
1 and Tr
α
1 mRNA levels increased progressively over the same c
Fig. 1 Targeted disruption of the Tr β 2 exon. a , The targeting vector deleted
165 bp of the Tr β 2 exon, including the splice donor (SD) site, which was replaced with a lacZ-neo cassette. This Thrb tm2Df mutant allele generated a product of 54 amino acid residues of Tr β 2 N terminus fused to β -galactosidase.
B, BamHI; B2, BglII; H, HindIII. b , Southern-blot analyses of genomic DNA showed correct targeting of Thrb. The origin of probes X, Y and Z is shown in
(a). In BamHI digests, probe Y detected a 9.0-kb wild-type band and a 5.5-kb mutant band. Probe Z was specific for lacZ in the mutant allele and demonstrated a single-copy integration of the targeting vector. Probe X was specific for the wild-type allele and did not detect a band in Thrb –/– mice, demonstrating deletion of the exon. Sizes of bands are indicated (kb). c , Northern blot showing that Tr β 1 mRNA expression was not altered in eyes of Thrb –/– mice.
Samples of total RNA were analysed with a Tr β 1-specific probe. d , Deletion of
Tr β 2 and retention of Tr β 1 mRNA in pituitary and eye in Thrb –/– mice. Left, the primers specific for Tr β 2 (b2/c) and Tr β 1 (b1/c) that were used for RT–PCR analysis shown (right). Both Tr β 2 (688 bp) and Tr β 1 (608 bp) products were detected in wild-type tissue, whereas only the Tr β 1 product was present in Thrb –/– tissue.
The RT–PCR products were cloned and sequenced to confirm their identity.
DBD, DNA-binding domain; LBD, ligand-binding domain.
1 Department of Human Genetics, Box 1498, Mount Sinai School of Medicine, New York, New York, USA. 2 Howard Hughes Medical Institute and
Department of Biochemistry, Box 357370, 3 Department of Biological Structure, Box 357420, Health Sciences Center, University of Washington, Seattle,
Washington, USA. 4 Laboratory of Developmental Biology, CMB, Karolinska Institute, Stockholm, Sweden. Correspondence should be addressed to
D.F.(e-mail: douglas.forrest@mssm.edu) or T.A.R. (e-mail: tomreh@u.washington.edu).
nature genetics • volume 27 • january 2001 1

2 letter
Fig. 2 Expression of Tr
β
2 in eye development.
a , Northern-blot analysis of Thrb and Thra mRNA expression in wild-type mouse eye. Poly(A)-selected mRNA (5
µ g/lane) was analysed with probes specific for Tr
β
2, Tr
β
1, the Tr
α
1 and Tr
α
2 Cterminal variant products of Thra, and, as a control for RNA integrity, G3pdh.
Eyes were studied together with newborn (P0) mouse brain. Autoradiographs were exposed for 2–3 days.
b , Detection of
β
-galactosidase activity indicating
Tr
β
2-lacZ expression in cells around the periphery of the immature outer nuclear layer in Thrb +/– and Thrb –/– embryos at E17.5. No activity was detected in wild-type embryos. INL, inner nuclear layer; ONL, outer nuclear layer; RPE, retinal pigmented epithelium.
a period. Staining for
β
-galactosidase activity in Thrb +/– and Thrb –/– embryos at E17.5 localized expression to a band of cells around the immature ONL of the retina, indicating that Tr
β
2 was expressed in the layer that contained the prospective photoreceptors (Fig. 2b).
We detected stained cells in the formative ONL as early as E13.5, whereas staining diminished in the postnatal retina.
The retina of Thrb –/– mice lacked M-opsin immunoreactive cones. In retina, we detected all cones with peanut agglutinin
(PNA)-fluoroscein (green fluorescence), and cones expressing Mopsin with antibodies that gave red fluorescence (Fig. 3a). In adult wild-type mice, M-opsin immunoreactive cones were present in a gradient with 100% of PNA-labeled cones in the superior retina being positive for M-opsin, whereas approximately 78% were positive in the inferior retina. In Thrb –/– mice, we did not detect any b
M-opsin-expressing cells in any retinal region. This phenotype probably reflected a developmental failure to form M-cones rather than a degenerative loss, as a similar phenotype was detected as early as postnatal day (P) 14, only a few days after M-opsin expression normally begins 16 . The absence of M-cones was unlikely to be due to excessive cell death, as TUNEL staining revealed no abnormal incidence of apoptotic cells in Thrb –/– mice at P14 (data not shown). Cell counts showed only slightly reduced total numbers of cones in the superior or inferior retina, suggesting that Thrb –/– a mice generate approximately normal numbers of cones.
Also, the gradient of S-cones across the retina was disturbed in
Thrb –/– mice. In the most superior part of the retina of wild-type mice, only approximately 3% of PNA-positive cones were immunoreactive for S-opsin, whereas in inferior regions almost all were positive. In Thrb –/– mice, S-opsin immunoreactivity was detected in 100% of PNA-labeled cones in both superior and infeb c
Fig. 3 Absence of M-cones and altered distribution of S-cones from Tr β 2-deficient mice. Consistent results were obtained by analysis of either retinal flatmounts
(shown) or serial retinal sections (data not shown).
a , M-opsin immunoreactive cones were present in the wild-type retina, but were not detetected in either superior or inferior retina of Thrb –/– mice. Flat mounts of retinas of wild-type and
Thrb –/– mice were analysed with PNA to detect both M- and S-cone photoreceptors (green fluorescence) and with antibodies against M-opsin (red fluorescence), and were imaged by confocal microscopy.
b , Abnormal presence of S-opsin cones in both the superior and inferior retina of Thrb –/– mice. In wild-type mice, many
PNA-labeled cones (green) co-labeled for S-opsin immunoreactivity (red) in the inferior retina, but very few did so in the superior retina. In contrast, in
Thrb –/– mice, all PNA-labeled cones in both superior and inferior regions were immunoreactive for S-opsin. Methods were similar to those in (a).
c , Anomalous S-opsin immunoreactive cells in
Thrb –/– mice were detected in retinal sections near the outer plexiform layer and in the INL (white arrows). These cells were not typical of cones based on their location and failure to colabel with PNA. Approximately three such cells were observed per section, whereas these were not detected in wild-type retina. IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; OS, outer segments.
nature genetics • volume 27 • january 2001

letter a b Fig. 4 BRIEF TITLE PLEASE a , Northern blot showing severe reduction of Mopsin mRNA levels in the eye of Tr
β
2-deficient mice (
≤
10% of wild-type levels).
S-opsin mRNA was precociously expressed, being fivefold above wild-type levels at E17.5. Mice lacking both Tr β 1 and Tr β 2 ( β 1/ β 2–/–; ref. 14) showed at least as severe a phenotype as Tr β 2-deficient ( β 2–/–) mice, with no detectable Mopsin mRNA. Lanes contained 15 µ g total RNA (wild-type and β 2–/–) or 5 µ g poly(A)-selected RNA (wild-type and β 1/ β 2–/–). Opsin mRNA levels were normalized relative to G3pdh signals detected on the same filters.
b , Premature appearance of S-opsin immunoreactive cells in the ONL in retinal sections of
Tr β 2-deficient (–/–) embryos at E17.5. Examples indicated by arrows. S-opsin immunoreactivity was undetectable in wild-type embryos.
rior regions. Mice normally possess S- and M-opsin–specific cones as well as many cones that co-express both opsins 17–19 sitivity to longer waves (600 nm; Fig. 5). This defect was consistent with a significant loss of functional M-cones. The results indicated that the remaining S-cones were functional and that their widespread distribution did not overtly perturb the electroretinogram. Dark-adapted rod responses from
Thrb –/– mice had normal kinetics and sensitivities.
. The detection of S-opsin in all cones in Thrb –/– mice indicates that in the absence of Tr
β
2, all cones follow a default S-cone pathway. Tr
β
2deficient mice also had some anomalous cells that were S-opsin immunoreactive, but which resided outside the cone layer in, for example, the inner part of the ONL and the inner nuclear layer (Fig.
3c). These cells were also PNA-negative, indicating that the major changes in the M- and S-cone populations in Thrb –/– mice may have perturbed the differentiation of some other retinal cell types.
Tr
β
2-deficient mice prematurely expressed S-opsin mRNA with levels fivefold greater than normal at E17.5 (Fig. 4a). Tr
β
2deficient embryos also had S-opsin immunoreactive cells, when normally no S-opsin was detectable (Fig. 4b). Levels of M-opsin mRNA were severely reduced (
≤
10% of normal), whereas Sopsin mRNA levels were slightly increased in adult Thrb –/– mice. These results suggest a transcriptional defect in cone differentiation and indicate a role for Tr
β
2 not only in inducing
M-opsin, but also in suppressing S-opsin until the developmentally appropriate stage 16,20 . As would be predicted, mice lacking both Tr
β
2 and Tr
β
1 displayed at least as severe a phenotype as
Tr
β
2-deficient mice, with reduced M-opsin mRNA and loss of
M-cones (Fig. 4a, and data not shown).
Electroretinograms of Thrb –/– mice revealed the presence of short-wave (450 nm) responses, but substantially reduced sen-
Genotype wild type
Thrb –/–
Table 1 • TSH and thyroid hormone levels in
Tr
β
2-deficient mice
TT3
(ng/dl)
81.2±6.4
118.8±10.1*
(
TT4
µ g/dl)
5.6±0.2
6.1±0.4
TSH
(ng/ml)
121±5
145±6*
% TSH
β
positive cells
5.1±0.7
5.6±0.7
Values are means±s.e.m. Groups contained n = 6–15 mice at 7–8 weeks of age.
TSH, T3 and T4 levels were increased in Thrb
–/– mice by
∼
20% (P < 0.01), 45% (P
< 0.01) and 8% (P > 0.05), respectively. TSH is shown for males (females have lower values). The percentage of TSH β -positive cells was determined from immunostained serial sections of the pituitary gland.
The role of Tr
β
2 in the pituitary and in auditory function was investigated. Serum levels of thyrotropin (TSH) and T3 were slightly elevated in Thrb –/– mice (Table 1). As thyroid hormone increases would normally suppress pituitary TSH, the results indicate a role for Tr
β
2 in the feedback regulation of the pituitary-thyroid axis. This phenotype represents a milder form of the pituitary resistance caused by deletion of both Tr
β
1 and Tr
β
2 (ref. 14), and indicates that Tr
β
2 cooperates with
Tr
β
1 in pituitary control 21 . Thrb is essential for the development of hearing
Thrb –/–
22 . Although Tr
β
2 is expressed in the cochlea 13 , mice at young (
∼
4 weeks) or adult (9–14 weeks) ages displayed normal auditory-evoked brainstem responses (data not shown). Therefore, the major function of Thrb in the development of hearing is likely to be mediated by Tr
β
1, or the role of Tr
β
2 may be substituted by Tr
β
1.
Our results reveal a unique role for Tr
β
2 in cone photoreceptor development that cannot be compensated for by Tr
β
1. Thus, Thrb, through differential expression of receptor variants, has a critical role in the development of two sensory systems, hearing and color vision. The retinal phenotype is consistent with the demonstration that T3 promotes the differentiation of cones in cultures of rat or human retinal progenitor cells 4,5 and precociously induces cone pigment expression in chick eye cultures 23 . Thyroxine also regulates development of ultraviolet-sensitive cones in the trout smolt 24 .
The role of Tr
β
2 in retina is likely to be conserved given its similar expression in chick and mouse. In primates, M- and S-cones represent distinct populations with separate bipolar cell pathways 25 , whereas in mouse, this system may be less highly developed, as many cones express both M- and S-opsins 17,19
Nonetheless, a related role for human TR
β
2 is a likely explanation
.
for the monochromacy in an individual with recessive resistance to thyroid hormone associated with deletion of the receptor gene 8,26 .
It remains to be determined whether the point mutations in the carboxy-terminal T3 binding domain in the dominant form of this syndrome result in cone disorders [AUTHORS: CORRECT?] .
Our findings indicate that immature cones throughout the retina have the potential to follow a default S-cone pathway and that Tr
β
2 is required for a subpopulation to form M- or dual (M and S) opsin cones. This is in accord with the expression of S-opsin preceding that of M-opsin, and with a proposal that many cones express both opsins before some cones transform into M-cones 16,18 . In a simple model, Tr
β
2 may induce
M-opsin and repress S-opsin. Nevertheless, the peak of Tr
β
2 expression precedes by many days the expression of M-opsin and corresponds with the prenatal generation of immature cones 3 . Thus, rather than directly inducing M-opsin, Tr
β
2 may act at an earlier stage of commitment to an M-cone identity.
The normal postnatal induction, or de-repression, of S-opsin might be achieved through diminished expression of Tr
β
2
(Fig. 2a), overriding activation by other transcription factors, or by a switch from T3-independent to T3-dependent functions of the Tr triggered by thyroid hormone increases in development 7 . Finally, the disturbed gradient of distribution of S-cones in Thrb –/– mice indicates that the generation and distribution of distinct cone populations are closely coupled nature genetics • volume 27 • january 2001 3

letter
Fig. 5 Defective cone responses in
Tr
β
2-deficient mice. Cone ERG responses were recorded using a paired flash protocol
19
. A bright conditioning flash of white light (100 ms at
1 mW) was first used to saturate rod photoreceptors for several seconds.
Because cones recover more quickly than rods, cone b-wave responses were then recorded 1.5 s after the conditioning flash using 450 nm or 600 nm flash stimuli. The flash intensities at the cornea were 1.5
×
10
20
, 2.1
×
10
19
,
3.0
×
10
18 tons/cm
2
4.6
×
10
19
9.2
×
10
16
, 2.0
×
10
17 and 4.1
×
10
16 phoat 450 nm, and 3.2
×
10
20
,
, 6.1
×
10
18
, 4.9
×
10
17 photons/cm
2 and at 600 nm. The traces shown are averages of 8–10 recordings from 3 wild-type mice and
6–7 recordings from 4 Thrb
–/– mice.
Responses to 600 nm light were reduced in amplitude. Means±s.d. for the peak amplitudes of wild-type and
Thrb
–/– mice were 278±44 and and
87±45
µ
V, respectively, for the maximum 600-nm response. Although the maximum amplitude of the 450-nm response from Thrb
–/– mice is normal, the sensitivity was found to be slightly reduced. The mutant requires 450-nm flashes
∼
3 times brighter than normal to elicit the same response as wild-type mice.
processes. Another nuclear receptor that might influence this same pathway is Nr2e3. Although the phenotypes differ somewhat from that of Tr
β
2-deficient mice, Nr2e3 mutations cause retinal degeneration in rd7 mice 27 , and are thought to disrupt long- and medium-wavelength cone photoreceptors in human enhanced S-cone syndrome 28 .
antibody; sections were then labeled with FITC-PNA (Sigma) to detect the entire population of cones. Sections were viewed on a Zeiss epifluorescence microscope. For P14 and E17 eyes, all analyses were preformed on cryosections. Retinas from other adult animals were dissected from extraocular tissue, flat-mounted and fixed in 4% PFA for 1–2 h. The flatmounts were washed in PBS, then immunostained and PNA-labeled as described above.
Flat mounts were viewed with a Zeiss PASCAL confocal microscope. We analyzed some sections from P14 retinas for the presence of apoptotic cells using the TUNEL method. Sections were incubated in Cy3-dUTP (Amersham) in the presence of 3´ TdT (Promega) for 1.5 h at 37 °C and the sections were then viewed with epifluorescence.
Targeted mutagenesis.
Mouse genomic DNA clones spanning the Tr
β
2 exon were isolated with a chick TR
β
2 probe 7 and used to construct the targeting vector. No Tr
β
1 sequences were present in these genomic clones, indicating that the Tr
β
2- and Tr
β
1-specific exons are separated by large
(>10 kb) distances. We used a linker to introduce a KpnI site at a StuI site in the Tr
β
2 exon, allowing the in-frame insertion of a lacZ-neo cassette 29 .
The mutation was introduced into W9.5 embryonic stem cells derived from 129/Sv-+ p + Tyr-c Mgf Sl-J /+ mice as described 14 . The resultant mutant mice had a mixed background of 129/Sv and C57BL/6J strains. The mutation was studied after backcrossing for 2–3 generations onto C57BL/6J. All animal experiments followed approved institutional protocols.
Histochemistry and immunostaining.
Tissues were fixed in 4% paraformaldehyde (PFA) in phosphate buffered saline (PBS). Cryosections were stained for
β
-galactosidase activity using 5-bromo-4-chloro-3indolyl-
β
-D-galactopyranoside as substrate and counterstained lightly with neutral red. TSH
β cells were counted after immunostaining of serial sections of the pituitary from three Thrb –/– and three wild-type mice, using antibodies against TSH
β
(AFP-66P9986), as described 30 . For immunostaining of adult eyes, in some cases the entire eye was cryo-sectioned and alternate sections were labeled for either S-opsin or M-opsin using specific antibodies JH455 and JH492, respectively, at a dilution of 1:5,000 (gifts from J. Nathans). Following overnight incubation at 4 °C, the sections were incubated with AlexaFluor 568 (Molecular Probes) to reveal the primary
Cone-cell counts.
We determined the number of cone photoreceptors in
≥
3 wild-type and 3 Thrb –/– mice by imaging two 0.0625 mm 2 fields in each retina; fields were chosen to be midway from the center to either the superior or inferior periphery. The number of cones that expressed either of the two opsins was determined as a percentage of the total
PNA-labeled cones. Total numbers of cone photoreceptors per 100
µ m 2 of retina were determined for 3 wild-type and 3 Thrb –/– mice for both inferior and superior retinal quadrants using PNA labeling to identify cones. Although in the wild-type mouse many cones express both Mand S-opsins 17 , it was not possible to demonstrate dual-opsin cell types directly because both opsin antibodies used were raised in rabbit, thus precluding differential labeling.
RNA analysis.
We used total RNA from eyes and pituitaries to analyse Tr
β
1 and Tr
β
2 mRNAs by RT–PCR as described 14 . Primers were based on mouse cDNA sequences 9 as follows: b1, Tr
β
1 nt 92–116; b2, Tr
β
2 nt
113–143; c, common antisense primer, nt 676–700.
We analyzed poly(A)-selected eye mRNA from BALB/c mouse embryos and pups using probes for mouse Tr
β
2, Tr
β
1, Tr
α
1, Tr
α
2 and G3pdh as described 15 . Tr
β probes were derived from cloned mouse cDNAs (ref. 9).
Mouse S- and M-opsin probes were made by RT–PCR from eye RNA and subcloned into pGEM-T Easy vectors (Promega). All PCR products were sequenced to confirm identity. On northern blots, opsin mRNA levels were quantified using a phosphorimager (Molecular Dynamics).
Hormone radioimmunoassays.
We measured TT4 and TT3 as described 30 . The TSH assay used a mouse LH/TSH reference
(AFP51718mp), a mouse TSH antiserum (AFP98991) and a rat TSH tracer (NIDDK-rTSH-I-9; provided by A. Parlow).
Electroretinogram (ERG) and auditory-evoked brainstem response
(ABR).
ERGs were recorded using a gold wire/contact lens electrode 31 placed on the cornea together with a drop of 2–3% methylcellulose. A reference electrode was placed in the mouth. Mice were anesthetized with a mixture of xylazine and ketamine, and maintained on a heating block at
37 °C throughout the experiment. Light flashes were delivered through a bifurcated fiber optic bundle with a lens focused on the cornea. One arm of the bundle delivered light from a Canon 540EZ Speedlite flash unit filtered with either 450 or 600 nm interference filters and focused onto the end of the bundle. The other arm of the bundle delivered white light from a 100 W halogen source equipped with a computer-controlled shutter.
Stimuli were attenuated with neutral density or interference filters. The unattenuated output from the 450-nm flash measured at the cornea was
66
µ
J/cm 2 and at 600 nm was 104
µ
J/cm 2 . The unattenuated power from the halogen lamp was 1.0 mW/cm 2 . ERG potentials were amplified
4 nature genetics • volume 27 • january 2001

letter
10,000 times with a Dagan model 2400 amplifier, filtered between 1 Hz and 3,000 Hz and digitized at 5,000 Hz. Data were collected in Igor Pro
(Wavemetrics) with an Instrutech ITC-16 analog to digital interface using a library of custom acquisition routines written by F. Rieke.
Thresholds were determined as described 22 using an apparatus from
Intelligent Hearing Systems. Groups of 10–12 mice were anesthetized with
Avertin and ABR thresholds determined in response to a click and pure tone pips at 8, 16 and 32 kHz.
Acknowledgments
We thank J. Nathans for opsin antibodies; C. Stewart for W9.5 ES cells; W.
Wood for Tr
β cDNAs, A.F. Parlow for reagents for TSH studies; I. Lisoukov for assistance with hormone assays; and the ES Cell Facility at the
Department of Human Genetics, Mount Sinai School of Medicine, and the
Transgenic Facility at the Karolinska Institute for assistance. This work was supported in part by grants from the National Institutes of Health (D.F.,
J.B.H., T.A.R.), the Swedish Medical Research Council and Swedish Cancer
Fund (B.V) and by the Human Frontiers Science Program (D.F., B.V.).
Received 18 August; accepted 14 November 2000.
[AUTHORS: NOTE REF. 23 QUERY]
1.
Wang, Y. et al. A locus control region adjacent to the human red and green visual pigment genes. Neuron 9 , 429–440 (1992).
2.
Szel, A., Lukats, A., Fekete, T., Szepessy, Z. & Rohlich, P. Photoreceptor distribution in the retinas of subprimate mammals. J. Opt. Soc. Am. A. Opt. Image Sci. Vis. 17 ,
568–579 (2000).
3.
Cepko, C.L., Austin, C.P., Yang, X., Alexiades, M. & Ezzeddine, D. Cell fate determination in the vertebrate retina. Proc. Natl Acad. Sci. USA 93 , 589–595 (1996).
4.
Kelley, M.W., Turner, J.K. & Reh, T.A. Ligands of steroid/thyroid receptors induce cone photoreceptors in vertebrate retina. Development 121 , 3777–3785 (1995).
5.
Kelley, M.W., Turner, J.K. & Reh, T.A. Regulation of proliferation and photoreceptor differentiation in fetal human retinal cell cultures. Invest.
Ophthalmol. Vis. Sci.
36 , 1280–1289 (1995).
6.
Hodin, R.A. et al. Identification of a thyroid hormone receptor that is pituitaryspecific. Science 244 , 76–78 (1989).
7.
Sjöberg, M., Vennström, B. & Forrest, D. Thyroid hormone receptors in chick retinal development: differential expression of mRNAs for
α and N-terminal variant β receptors. Development 114 , 39–47 (1992).
8.
Newell, F.W. & Diddie, K.R. Typical monochromacy, congenital deafness, and resistance to intracellular action of thyroid hormone. Klin. Monatsbl.
Augenheilkd.
171 , 731–734 (1977).
9.
Wood, W.M., Ocran, K.W., Gordon, D.F. & Ridgway, E.C. Isolation and characterization of mouse complementary DNAs encoding
α and
β thyroid hormone receptors from thyrotrope cells: the mouse pituitary-specific β 2 isoform differs at the amino terminus from the corresponding species from rat pituitary tumor cells. Mol. Endocrinol. 5 , 1049–1061 (1991).
10. Ng, L., Forrest, D., Haugen, B.R., Wood, W.M. & Curran, T. N-Terminal variants of thyroid hormone receptor β : differential function and potential contribution to syndrome of resistance to thyroid hormone. Mol. Endocrinol. 9 , 1202–1213
(1995).
11. Sjöberg, M. & Vennström, B. Ligand-dependent and -independent transactivation by thyroid hormone receptor
β
2 is determined by the structure of the hormone response element. Mol. Cell. Biol. 15 , 4718–4726 (1995).
12. Langlois, M.-F. et al. A unique role of the
β
-2 thyroid hormone receptor isoform in negative regulation by thyroid hormone. J. Biol. Chem. 272 , 24927–24933 (1997).
13. Bradley, D.J., Towle, H.C. & Young, W.S. III
α and
β thyroid hormone receptor (TR) gene expression during auditory neurogenesis: evidence for TR isoform-specific transcriptional regulation in vivo. Proc. Natl Acad. Sci. USA 91 , 439–443 (1994).
14. Forrest, D. et al. Recessive resistance to thyroid hormone in mice lacking thyroid hormone receptor
β
: evidence for tissue-specific modulation of receptor function.
EMBO J.
15 , 3006–3015 (1996).
15. Forrest, D., Sjöberg, M. & Vennström, B. Contrasting developmental and tissuespecific expression of α and β thyroid hormone receptor genes. EMBO J. 9 ,
1519–1528 (1990).
16. Szel, A., Rohlich, P., Mieziewska, K., Aguirre, G. & van Veen, T. Spatial and temporal differences between the expression of short- and middle-wave sensitive cone pigments in the mouse retina: a developmental study. J. Comp. Neurol. 331 ,
564–577 (1993).
17. Rohlich, P., van Veen, T. & Szel, A. Two different visual pigments in one retinal cone cell. Neuron 13 , 1159–1166 (1994).
18. Szel, A., van Veen, T. & Rohlich, P. Retinal cone differentiation. Nature 370 , 336
(1994).
19. Lyubarsky, A.L., Falsini, B., Pennesi, M.E., Valentini, P. & Pugh, E.N., Jr. UV- and midwave-sensitive cone-driven retinal responses of the mouse: a possible phenotype for coexpression of cone photopigments. J. Neurosci. 19 , 442–455
(1999).
20. Chiu, M.I. & Nathans, J. A sequence upstream of the mouse blue visual pigment gene directs blue cone-specific transgene expression in mouse retinas. Vis.
Neurosci.
11 , 773–780 (1994).
21. Abel, E.D. et al. Divergent roles for thyroid hormone receptor
β isoforms in the endocrine axis and auditory system. J. Clin. Invest. 104 , 291–300 (1999).
22. Rüsch, A., Erway, L., Oliver, D., Vennström, B. & Forrest, D. Thyroid hormone receptor β -dependent expression of a potassium conductance in inner hair cells at the onset of hearing. Proc. Natl Acad. Sci. USA 95 , 15758–15762 (1998).
23. Sjöberg, M. TITLE?
Thesis, Karolinska Institute (1994).
24. Browman, H.I. & Hawryshyn, C.W. The developmental trajectory of ultraviolet photosensitivity in rainbow trout is altered by thyroxine. Vision Res. 34 ,
1397–1406 (1994).
25. Dacey, D.M. Parallel pathways for spectral coding in primate retina. Annu. Rev.
Neurosci.
23 , 743–775 (2000).
26. Refetoff, S., Weiss, R.E. & Usala, S.J. The syndromes of resistance to thyroid hormone. Endocrine Rev. 14 , 348–399 (1993).
27. Akhmedov, N.B. et al. A deletion in a photoreceptor-specific nuclear receptor mRNA causes retinal degeneration in the rd7 mouse. Proc. Natl Acad. Sci. USA 97 ,
5551–5556 (2000).
28. Haider, N.B. et al. Mutation of a nuclear receptor gene, NR2E3, causes enhanced S cone syndrome, a disorder of retinal cell fate. Nature Genet. 24 , 127–131 (2000).
29. Chomez, P. et al. Increased cell death and delayed development in the cerebellum of mice lacking the rev-erbA( α ) orphan receptor. Development 127 , 1489–1498
(2000).
30. Göthe, S. et al. Mice devoid of all known thyroid hormone receptors are viable but exhibit disorders of the pituitary-thyroid axis, growth and bone maturation.
Genes Dev.
13 , 1329–1341 (1999).
31. Bayer, A. et al. Comparisons of the amplitude size and the reproducibility of three different electrodes to record the corneal flash electroretinogram in rodents.
Doc. Ophthalmol.
98 , 233–246 (2000).
nature genetics • volume 27 • january 2001 5