|Molecular Vision 2001;
Received 26 February 2001 | Accepted 17 August 2001 | Published 22 August 2001
Expression of the a1F calcium channel subunit by photoreceptors in the rat retina
Morgans,1 Phil Gaughwin,1
1John Curtin School of Medical Research and 2Research School of Biological Sciences, Australian National University, Canberra, Australia
Correspondence to: Dr. C. W. Morgans, Synaptic Biochemistry Group, Division of Neuroscience, John Curtin School of Medical Research, Canberra, ACT, 0200, Australia; Phone: +61 (06) 6249-2149; FAX: +61 (06) 6249-2687; email: email@example.com
Purpose: The CACNA1F gene encodes a voltage-gated calcium channel a1 subunit, a1F, which is expressed in the human retina. Mutations in this gene cause incomplete X-linked congenital stationary night blindness (CSNB2). The aim of this study was to obtain the sequence of the rat a1F cDNA and localize the encoded polypeptide in the rat retina.
Methods: The full-length rat a1F sequence was compiled from sequencing of overlapping a1F PCR fragments amplified from rat retinal cDNA. Antiserum was raised against a human a1F peptide. It was found that the human a1F peptide used to generate the antiserum was conserved at only 11 out of 19 residues in the cloned rat sequence. Therefore, antibodies were affinity purified against either the human a1F peptide or the equivalent rat peptide and used for immunofluorescent staining of rat retina sections.
Results: The rat a1F amino acid sequence was found to be 91% and 95% identical to the human and mouse a1F sequences, respectively. Antibodies affinity purified against the human a1F peptide stained both the outer nuclear layer (ONL) and outer plexiform layer of rat retina sections. In contrast, staining with antibodies affinity purified against the corresponding rat a1F peptide was restricted to the ONL.
Conclusions: The rat a1F amino acid sequence is highly homologous to the human and mouse sequences. The immunohistochemical results indicate the existence of distinct a1F isoforms or a1F-like channels, which are differentially distributed in the cell bodies and synaptic terminals of photoreceptors in the rat retina.
Incomplete X-linked congenital stationary night blindness, CSNB2, is a recessive non-progressive visual disease characterized by poor night vision and decreased visual acuity. The electroretinograms (ERGs) of patients with CSNB2 display a normal a-wave under scotopic conditions, but a reduced b-wave . The phenotype of CSNB2 is consistent with a defect in neurotransmission within the retina between the photoreceptors and second order neurons. The gene for CSNB2 was recently identified as the CACNA1F gene, which encodes the alpha-1 subunit of a voltage gated calcium channel, a1F [2,3].
Voltage activated calcium channels are heteromeric protein complexes containing a principal pore-forming a1 subunit in association with a b subunit and a disulfide linked a2/d subunit. The molecular identity of the a1 subunit largely determines the pharmacological sensitivity of the channel. For example, a1C and a1D subunits give rise to L-type channels that are sensitive to dihydropyrridines (DHPs) . Sequence comparisons show that human a1F is also a member of the L-type family of a1 subunits, displaying greatest amino acid identity, 62%, to the a1D subunit of brain L-type calcium channels [2,3]. Physiological studies show that tonic glutamate release from photoreceptors and bipolar cells is supported by non-inactivating, L-type calcium channels [5-7]. Hence, it has been suggested that calcium entry into the photoreceptor terminals is mediated by the a1F calcium channel .
Because of the prevalence of the rat model in retinal neuroanatomy and physiology, we have cloned and sequenced the rat a1F cDNA and localized the polypeptide in the rat retina. We show that the encoded a1F channel is localized to the cell bodies but not the synaptic terminals of photoreceptors in the rat retina, arguing against a direct role for this channel in synaptic transmission. We present evidence, however, for an additional a1F isoform (product of alternate mRNA splicing) or a1F-like channel (product of a gene closely related to CACNA1F) localized to photoreceptor synaptic terminals, and therefore more likely to directly mediate transmitter release.
Total RNA was purified from isolated rat retinas (described below) using the phenol-guanidine-isothiocyanate-chloroform extraction protocol (TRIzol Reagent, Life Technologies, Melbourne, Australia). Synthesis of first-strand cDNA from total RNA was conducted using a modified version of the Superscript II reverse transcriptase preamplification system (Life Technologies) in a reaction volume of 60 ml. Total RNA (10 ng) and RNAse inhibitor (5 U) in 36 ml, the first strand synthesis reagents (kit buffer [50 mM Tris-HCl, 75 mM KCl, 3 mM MgCl2, pH 8.3], dithiothreitol [DTT, 10 mM] and dNTPs [0.5 mM each of dATP, dGTP, dCTP and dTTP]) were treated separately with DNAse I (RNase-free; Boehringer Mannheim, Germany) to prevent amplification of genomic sequences. Following heat inactivation of the DNAse (2 min at 70 °C), the RNA mix was incubated for a further 10 min at 70 °C with 1.5 mg oligo dT primers. The RNA and reagent mixes were chilled briefly, then combined and incubated for 50 min at 42 °C in the presence of Superscript II MuLV reverse transcriptase (600 U). The reaction was heat inactivated (70 °C, 10 min), chilled and incubated a further 20 min at 37 °C with 10 U RNAse H, in order to remove all remaining RNA prior to PCR.
Retina cDNA was used as a template for PCR amplification. PCR primers (shown in Table 1) were designed against the human a1F cDNA sequence presented by Strom et al. (1998) . All primer pairs were presumed to span at least one intron based on the assumption that the rat and humans introns have the same positions in the orthologous genes. Primer sequences are given in Table 1. PCR reactions in a total volume of 20 ml contained: serial dilutions of template cDNA, 0.1 mM dNTPs, 2.0 ml 10X buffer, 0.25 U Taq polymerase, 10 pmol each forward and reverse primers, H2O to 20 ml. The MgCl2 concentration (2.5-4 mM) was optimized for each primer pair. Cycle conditions were the following: denaturation at 94 °C, 15 s; annealing at an optimized temperature (55-58 °C); extension at 72 °C for 60-90 s depending on the expected product length. The reaction mixes were subjected to 35 cycles followed by a final 5 min extension at 72 °C. All PCR reactions were conducted in duplicate. No PCR products were generated from tissues other than retina, indicating that genomic DNA was not amplified under the conditions described.
Subcloning and sequencing of rat a1F cDNA fragments
Amplification products were gel purified and subcloned into a p-GEM-T vector (Promega, Madison, WI). Inserts were sequenced in duplicate in both the forward and reverse directions using the BigDye chain terminator cycle sequencing kit (Perkin Elmer, Norwalk, CT) The construction of a contiguous sequence was accomplished using WebANGIS (Australian National Genomic Information Service). The extraction of the amino acid sequence was accomplished using Map and Extractpeptide. A contiguous amino acid sequence was constructed from overlapping PCR fragments using a combination of Bestfit, Pileup and Multalign. The GenBank accession number for the rat a1F cDNA sequence is AF365975.
The production of anti-a1F antiserum was performed by Chiron Technologies (Melbourne, Australia) as described previously . Sheep were immunized with a peptide corresponding to amino acids 777 to 795 of human a1F (peptide sequence: SNEKDLPQENEGLVPGVEK) coupled through an additional C-terminal cysteine to diptheria toxoid (DT). Immune serum was collected following two immunizations and stored at -20 °C.
Affinity purification of the a1F antibody: Sequence analysis of a rat a1F cDNA fragment obtained by PCR revealed limited conservation between the human a1F peptide and the equivalent rat peptide. Only 11 out of 19 residues were conserved. The sequence of the rat peptide is the following: SSEGNPPQENKVLVPGGEN (residues conserved between the rat and human sequences are in bold type). The rat and human peptides were synthesized, and 1 mg of each coupled to HiTrap NHS-activated affinity columns (Amersham Pharmacia Biotech; Sweden) and used according to the manufacturer's protocol to affinity purify antibodies from the anti-a1F antiserum. The material affinity purified on the human peptide was called anti-a1F(human), and the material purified on the rat peptide, anti-a1F(rat).
Wistar Kyoto rats of 6-8 weeks old were anaesthetized with a lethal intraperitoneal injection of Nembutal and their retinas removed. Subcellular retina fractions were prepared by adapting the protocol of Muresan et al.  as follows. All steps were done at 0-4 °C. Two rat retinas were immersed in ice-cold 750 ml PB buffer (15 mM phosphate buffer pH 7.4, 1 mM MgCl2, 1 mM EGTA, 0.025% NaN3). A protease inhibitor cocktail (50 ml; AEBSF, pepstatin A, E-64, bestatin, leupeptin, aprotinin; Sigma) was added to the buffer (PB*) and the tissue was homogenized by hand with 40 strokes in a 1 ml glass homogenizer with a Teflon pestle. The homogenate was carefully layered over 500 ml of 50% (w/v) sucrose in PB buffer in a 1.5 ml microcentrifuge tube and centrifuged for 10 min at 15,000x g, 4 °C in a Hereaus microcentrifuge. The following fractions were collected: the supernatant above the buffer/sucrose interface (cytosol), the material at the buffer/sucrose interface (membranes) and the pellet (nuclei). All fractions were brought to the same volume in a final concentration of 1X SDS sample buffer and stored at -20 °C. Equal volumes of each fraction were subjected to SDS-polyacrylamide gel electrophoresis on pre-cast 4-12% Bis-Tris gels (Novex; San Diego, CA) and immunoblotted using chemiluminescent detection (ECL; Amersham International, England) as previously described . Antibodies were used at the following concentrations: affinity purified anti-a1F(rat), 1:100; anti-a1F(human), 1:1000; donkey anti-sheep IgG coupled to horseradish peroxidase (HRP; Jackson ImmunoResearch Laboratories, West Grove, PA), 1:5000. For blocking experiments, rat or human a1F peptides at a concentration of 1 mg/ml were preincubated with diluted primary antibodies for 30 min at room temperature (RT) before applying the antibodies to the blots.
Preparation of retina sections
Eyes were removed from anaesthetized rats and enucleated, hemisected at the junction of the sclera and cornea, and the lens and vitreous removed from the posterior eyecup. Eyecups were fixed by immersion in 4% (w/v) paraformaldehyde in phosphate buffered saline (PBS) for 5 to 15 min, then washed in PBS and cryoprotected in 30% (w/v) sucrose. The eyecups were embedded in Tissue-Tek O.C.T. Compound (Sakura Finetek, Torrence, CA) and sectioned vertically at 12 mm thickness on a cryostat. Sections were collected on gelatin-coated slides, air-dried and stored at -20 °C. For the detergent extraction experiment, an unfixed retina was cryoprotected and sectioned as described above. Unfixed sections were washed in PBS and then immunostained as described below.
Retina sections were blocked by incubation for 30 min at RT in Antibody Incubation Solution (AIS; 0.5% Triton X-100, 5% horse serum, 0.05% NaN3 in PBS). Sections were then incubated in the primary antibody diluted in AIS for either 2 h or ON at RT. The sections were washed three times in PBS, then incubated for 1 h at RT in the secondary antibody diluted in AIS. Sections were again washed in PBS and then coverslipped with Mowiol (Hoechst, Germany). The primary antibodies were used at the following concentrations: anti-a1F(rat) and anti-a1F(human) 1:102 and anti-synaptotagmin, 1:103. For blocking experiments, diluted a1F antibodies were preincubated with either the human or rat a1F peptides at a concentration of 1 mg/ml. The following secondary antibodies were used: donkey-anti-sheep-IgG coupled to carboxymethylindocyanine (Cy3; Jackson ImmunoResearch Laboratories) diluted 1:500 and goat-anti-rabbit-IgG coupled to fluorescein isothiocyanate (FITC; Jackson ImmunoResearch Laboratories) diluted 1:50. The sections were analyzed with a Leica TCS 4D confocal laser scanning microscope using a 40x/1.4 N.A. oil immersion objective (Leica, Germany). Confocal images were imported into Adobe Photoshop for editing. Image enhancement was strictly limited to minor adjustments to image brightness done uniformly over the entire image.
Molecular cloning of the rat a1F cDNA
Oligonucleotides corresponding to the human a1F cDNA sequence were used to amplify a series of overlapping rat a1F cDNA fragments from rat retina (Figure 1B). All primer pairs yielded products of the predicted length based on the human sequence (Figure 1A). PCR fragments of expected size were subcloned and sequenced to yield a contiguous sequence encoding rat a1F (base pairs 1 to 5964). The rat nucleotide sequence is 86% and 94% identical to the human (GenBank accession number AJ224874) and mouse sequences (GenBank accession number AF192497), respectively; and the translated rat amino acid sequence is 91% and 95% identical to the human and mouse polypeptides, respectively [2,3,10]. An amino acid alignment of the rat and human a1F amino acid sequences (Figure 2) shows that the transmembrane alpha helices are almost perfectly conserved. Eight of the nine residues required for full DHP sensitivity of L-type channels  are conserved in a1F. The same substitution of a phenylalanine for a tyrosine in the second alpha helix of the fourth transmembrane domain (IVs2) occurs in both the human and rat sequences. This substitution may account for the relatively low sensitivity of mammalian photoreceptor and bipolar cell calcium currents to dihydropyridines [12-15].
The most divergent regions between rat and human a1F, are the carboxy terminal cytoplasmic tail and the putative cytoplasmic loop between membrane domains II and III (Figure 2). This cytoplasmic loop is also the least conserved region between different calcium channel a1 subunits, and for this reason a peptide sequence from this loop was chosen to raise a1F-specific antibodies. The a1F polypeptide is localized to photoreceptor cell bodies and excluded from photoreceptor synaptic terminals in the rat retina. Antiserum was raised against a peptide derived from the human a1F sequence [2,3] lying in the cytoplasmic loop between membrane domains II and III. This is the least conserved region between different calcium channel a1 subunits, and therefore unlikely to generate antibodies that cross-react with other calcium channels. Upon completion of the rat sequence, it was found that only 11 out of 19 amino acids of the human a1F peptide are conserved in the rat (Figure 2, sequence denoted by asterisks). Therefore, the corresponding rat peptide was synthesized and used to affinity purify antibodies, referred to as anti-a1F(rat), from the anti-a1F serum. The anti-a1F(rat) antibodies thus recognize epitopes in common between the rat and human peptides and were used to localize rat a1F in the retina. Figure 3 shows a sequence alignment of the human and rat a1F peptides with the equivalent peptides from rat a1C and a1D, the a1 subunits bearing the greatest sequence identity to a1F. The degree of sequence identity between the human and rat a1F peptides is significantly greater than between the human a1F peptide and either the a1C or a1D sequences.
Horizontal sections of rat retina were labeled by immunofluorescence with affinity purified anti-a1F(rat) to reveal the cellular distribution of the cloned rat a1F channel. Strong staining of the outer nuclear layer (ONL) was observed outlining the photoreceptor somata (Figure 4). In addition to photoreceptor cell bodies, the ONL contains Müller cell processes, which extend from the outer plexiform layer (OPL) to the outer limiting membrane . The a1F staining in the ONL is very similar to that for the photoreceptor glutamate transporter, GLT-1 , and unlike the staining for the Müller cell marker, glutamine synthetase . This indicates the presence of a1F on the plasma membrane of photoreceptor cell bodies rather than on Müller cell processes. Photoreceptors in the rat retina are approximately 99% rods and 1% cones , thus the staining in the ONL is a clear indication of the presence of a1F around rod photoreceptor cell bodies. No gaps in the ONL staining were observed on any sections, so the small cone population may also be labeled. On some sections, the occasional outer segment was intensely labeled (Figure 4B and Figure 5A). Faint, diffuse staining of the ganglion cell layer (GCL) was also seen (Figure 4B). All staining was completely blocked by pre-incubation of the antibody with either the rat a1F peptide (Figure 4C) or the human a1F peptide (not shown) demonstrating the specificity of the staining.
The localization of a1F in photoreceptors was compared to that of a synaptic marker (Figure 6). Rat retina sections were double labeled with anti-a1F(rat) and an antibody against synaptotagmin, an integral membrane protein of synaptic vesicles. Figure 6 (left panel) shows strong a1F staining in the ONL, whereas synaptotagmin is predominantly localized to the outer plexiform layer, the site of photoreceptor synapses (middle panel). Overlay of the two staining patterns (right panel) shows little overlap in the OPL. The double labeling experiment demonstrates that a1F is confined to photoreceptor cell bodies and is excluded from photoreceptor synaptic terminals.
Evidence for additional a1F-like channels in the rat retina
In contrast to the anti-a1F(rat) staining, antibodies affinity purified against the human a1F peptide (see Methods), referred to here as anti-a1F(human), gave strong punctate labeling of the OPL in addition to the ONL staining (Figure 5A). By comparing the staining patterns in Figure 5A (left panel) and Figure 4, it appears that anti-a1F(rat) labels a subset of epitopes recognized by anti-a1F(human). Preincubation of anti-a1F(human) with the human a1F peptide completely blocks all staining (not shown), whereas preincubation with the rat a1F peptide completely blocks the ONL staining but has little effect on the OPL staining (Figure 5A, right panel). A possible interpretation of these data is that the anti-a1F(human) antibody fraction recognizes additional a1F-like channels present in the synaptic terminals of photoreceptors which are not recognized by the anti-a1F(rat) antibody fraction.
Western blots of retinal proteins with anti-a1F(rat) and anti-a1F(human) support this interpretation. Both antibody fractions label bands of the same apparent molecular weights on western blots, but they are differentially blocked by preincubation with the rat and human a1F peptides. On blots of rat retinal membrane proteins, both anti-a1F(rat) and anti-a1F(human) label two major bands at 142 and 150 kD (Figure 5B). Two additional higher molecular weight bands at 170 and 190 kD were occasionally observed (Figure 5C). These bands are likely to be proteolytic products of full-length a1F, which has a predicted molecular weight of 220 kD (see discussion). No bands were detected in a sample of brain membranes with either anti-a1F(human) (Figure 5B, lane 3) or anti-a1F(rat) (not shown), and no bands were detected in retina cytosolic fractions (not shown). Preincubation of the primary antibody with the human a1F peptide completely eliminates detection of all four bands for both anti-a1F(rat) and anti-a1F(human). In contrast, preincubation with the rat a1F peptide almost entirely blocks detection of the bands with anti-a1F(rat) but only slightly lowers the intensity of labeling with anti-a1F(human). The differential blocking effect of the two peptides supports the existence of more than one a1F isoform or a1F-like channel in the rat retina. This is further supported by differences in the relative intensity of the four bands detected by anti-a1F(rat) and anti-a1F(human).
Physiological and pharmacological studies indicate that photoreceptor calcium currents are mediated by L-type calcium channels, yet the properties of the calcium currents in mammalian photoreceptors and bipolar cells differ somewhat from other brain L-type calcium currents. The retinal L-type currents have a lower activation threshold (~50 mV) and display a lower sensitivity to DHPs [12-15]. It is likely that photoreceptor and bipolar cell calcium channels contain an a1 subunit distinct from the a1C or a1D subunits of brain L-type calcium channels. Recently, a human gene, CACNA1F, encoding a novel human calcium channel, a1F, was identified [2,3], which appears to be expressed only in the retina . Sequence analysis indicates that a1F belongs to the L-type family of calcium channel a1 subunits. In this study, the rat a1F primary sequence is presented. The rat a1F polypeptide encoded by the cloned sequence was found to be localized to the outer nuclear layer of the retina where it is distributed around photoreceptor cell bodies and excluded from photoreceptor synaptic terminals in the OPL. Immunohistochemical evidence suggests that an additional a1F isoform or a1F-like channel is present at photoreceptor synapses in the OPL.
It is unlikely that the a1F immunoreactivity in either the ONL or OPL is due to cross-reactivity of the a1F antibodies with one of the other known L-type a1 subunits (a1C, a1D, a1S). There is a modest degree of sequence similarity between the rat and human a1F peptides used to affinity purify the antibodies and the equivalent a1C and a1D peptides (Figure 3). However, no specific bands were detected on western blots of brain membranes (Figure 5B), which contain both a1C and a1D subunits. Furthermore, immunostaining of rat retina sections with antibodies against a1C and a1D results in a qualitatively different staining pattern to that obtained for a1F (Morgans, unpublished). Western blots of retinal proteins for a1F invariably resulted in bands between 140-190 kD, significantly smaller than the predicted molecular weight of 220 kD. It is likely that these bands represent a1F polypeptides truncated in their C-terminal cytoplasmic domain. Proteolytic cleavage of the C-terminal domain of L-type a1 subunits is well documented. For example, more than 80% of the a1C subunits in cardiac myocytes were found to migrate approximately 50 kD smaller than their predicted molecular weight on SDS gels due to cleavage of their C-terminal cytoplasmic tail . Despite cleavage, the C-terminal fragment remains functionally associated with the channel . Similar C-terminal cleavage of a1C in brain and a1S in skeletal muscle has also been reported [20,21].
The complex phenotype of CSNB2 suggests that multiple cellular functions are affected. This is supported by our evidence for at least two differentially distributed forms of a1F, one around photoreceptor cell bodies in the ONL and the other localized to photoreceptor synapses in the OPL. The synaptically localized a1F channel is likely to be required for glutamate release from photoreceptors by mediating voltage-gated calcium entry into the terminals. The influx of calcium through the ONL a1F calcium channel may influence photoreceptor function by its effect on calcium-activated currents and by directly influencing the membrane potential. In order to better understand the genetic and physiological basis of CSNB2, it will be important to determine whether the differentially distributed a1F channels identified immunohistochemically are the results of alternate splicing of the same gene or are products of distinct genes.
Many thanks to Dr. John Bekkers and Dr. Rowland Taylor for helpful discussions and critical reading of the manuscript. This work was supported in part by grants from the Ramaciotti Foundation, the Australian Retinitis Pigmentosa Association, and the Centre for Visual Sciences at the Australian National University.
1. Miyake Y, Yagasaki K, Horiguchi M, Kawase Y, Kanda T. Congenital stationary night blindness with negative electroretinogram. A new classification. Arch Opthalmol 1986; 104:1013-20.
2. Bech-Hansen NT, Naylor MJ, Maybaum TA, Pearce WG, Koop B, Fishman GA, Mets M, Musarella MA, Boycott KM. Loss-of-function mutations in a calcium-channel alpha1-subunit gene in Xp11.23 cause incomplete X-linked congenital stationary night blindness. Nat Genet 1998; 19:264-7.
3. Strom TM, Nyakatura G, Apfelstedt-Sylla E, Hellebrand H, Lorenz B, Weber BH, Wutz K, Gutwillinger N, Ruther K, Drescher B, Sauer C, Zrenner E, Meitinger T, Rosenthal A, Meindl A. An L-type calcium-channel gene mutated in incomplete X-linked congenital stationary night blindness. Nat Genet 1998; 19:260-3.
4. Fox AP, Nowycky MC, Tsien RW. Kinetic and pharmacological properties distinguishing three types of calcium currents in chick sensory neurones. J Physiol 1987; 394:149-72.
5. Tachibana M, Okada T, Arimura T, Kobayashi K, Piccolino M. Dihydropyridine-sensitive calcium current mediates neurotransmitter release from bipolar cells of the goldfish retina. J Neurosci 1993; 13:2898-909.
6. Rieke F, Schwartz EA. Asynchronous transmitter release: control of exocytosis and endocytosis at the salamander rod synapse. J Physiol 1996; 493:1-8.
7. Schmitz Y, Witkovsky P. Dependence of photoreceptor glutamate release on a dihydropyridine-sensitive calcium channel. Neurosci 1997; 78:1209-16.
8. Morgans CW, Brandstatter JH, Kellerman J, Betz H, Wassle H. A SNARE complex containing syntaxin 3 is present in ribbon synapses of the retina. J Neurosci 1996; 16:6713-21.
9. Muresan V, Lyass A, Schnapp BJ. The kinesin motor KIF3A is a component of the presynaptic ribbon in vertebrate photoreceptors. J Neurosci 1999; 19:1027-37.
10. Naylor MJ, Rancourt DE, Bech-Hansen NT. Isolation and characterization of a calcium channel gene, Cacna1f, the murine orthologue of the gene for incomplete X-linked congenital stationary night blindness. Genomics 2000; 66:324-7.
11. Sinnegger MJ, Wang Z, Grabner M, Hering S, Striessnig J, Glossmann H, Mitterdorfer J. Nine L-type amino acid residues confer full 1,4-dihydropyrridine sensitivity to the neuronal calcium channel alpha1A subunit. Role of L-type Met1188. J Biol Chem 1997; 272:27686-93.
12. Yagi T, Macleish PR. Ionic conductances of monkey solitary cone inner segments. J Neurophysiol 1994; 71:656-65.
13. Taylor WR, Morgans C. Localization and properties of voltage-gated calcium channels in cone photoreceptors of Tupaia belangeri. Vis Neurosci 1998; 15:541-52.
14. de la Villa P, Vaquero CF, Kaneko A. Two types of calcium currents of the mouse bipolar cells recorded in the retinal slice preparation. Eur J Neurosci 1998; 10:317-23.
15. Protti DA, Llano I. Calcium currents and calcium signaling in rod bipolar cells of rat retinal slices. J Neurosci 1998; 18:3715-24.
16. Haverkamp S, Wassle H. Immunocytochemical analysis of the mouse retina. J Comp Neurol 2000; 424:1-23.
17. Szel A, Rohlich P, Caffe AR, van Veen T. Distribution of cone photoreceptors in the mammalian retina. Microsc Res Tech 1996; 35:445-62.
18. Gao T, Puri TS, Gerhardstein BL, Chien AJ, Green RD, Hosey MM. Identification and subcellular localization of the subunits of L-type calcium channels and adenylyl cyclase in cardiac myocytes. J Biol Chem 1997; 272:19401-7.
19. Gao T, Cuadra AE, Ma H, Bunemann M, Gerhardstein BL, Cheng T, Eick RT, Hosey MM. C-terminal fragments of the alpha 1C (CaV1.2) subunit associate with and regulate L-type calcium channels containing C-terminal-truncated alpha 1C subunits. J Biol Chem 2001; 276:21089-97.
20. Hell JW, Yokoyama CT, Wong ST, Warner C, Snutch TP, Catterall WA. Differential phosphorylation of two size forms of the neuronal class C L-type calcium channel alpha 1 subunit. J Biol Chem 1993; 268:19451-7.
21. De Jongh KS, Warner C, Colvin AA, Catterall WA. Characterization of the two size forms of the alpha 1 subunit of skeletal muscle L-type calcium channels. Proc Natl Acad Sci U S A 1991; 88:10778-82.