Expression of Ca_V3.2 T-Type Ca²⁺ Channels in a Subpopulation of Retinal Type-3 Cone Bipolar Cells

Abstract

Retinal bipolar cells and ganglion cells are known to possess voltage-gated T-type Ca²⁺ channels. Previous electrophysiological recording studies suggested that there is differential expression of different T-type Ca²⁺ channel α1 subunits among bipolar cells. The detailed expression patterns of the individual T-type Ca²⁺ channel subunits in the retina, however, remain unknown. In this study, we examined the expression of the Ca_v3.2 Ca²⁺ channel α1 subunit in the mouse retina using immunohistochemical analysis and patch-clamp recordings together with a Ca_v3.2 knock out (KO) mouse line. The specificity of a Ca_v3.2 Ca²⁺ channel antibody was first confirmed in recombinant T-type Ca²⁺ channels expressed in HEK (human embryonic kidney) cells and in Ca_v3.2 KO mice. Our immunohistochemical analysis indicates that the Ca_v3.2 antibody labels a subgroup of type-3 cone bipolar cells (CBCs), the PKAβII-immunopositive type-3 CBCs. The labeling was observed throughout the cell including dendrites and axon terminals. Our patch-clamp recording results further demonstrate that Ca_v3.2 Ca²⁺ channels contribute to the T-type Ca²⁺ current in a subpopulation of type-3 CBCs. The findings of this study provide new insights into understanding the functional roles of T-type Ca²⁺ channels in retinal processing.

Introduction

T-type Ca²⁺ channels play a variety of roles in the central nervous system, including controlling membrane excitability, pacemaker activity, intracellular Ca²⁺ concentration, and hormone secretion (Huguenard, 1996; Perez-Reyes, 2003; Carbone et al., 2006). Three isoforms of poreforming T-type Ca²⁺ channel α1 subunits, Ca_v3.1(α1_G), Ca_v3.2(α1_H), and Ca_v3.3(α1_I), have been cloned and characterized (Perez-Reyes, 2003; 2006). T-type Ca²⁺ currents with different α1 subunits exhibit distinct biophysical properties (McRory et al., 2001). T-type Ca²⁺ channels are also subject to subunit-specific modulations (Todorovic et al., 2001; Chemin et al., 2006; Traboulsie et al., 2007; Hildebrand et al., 2007; Nelson et al., 2007; Perez-Reyes, 2010). Heterogeneous expression of different isoforms of T-type Ca²⁺ channels has been reported in the CNS (Talley et al., 1999) and contributes to distinct neuronal physiological functions (Chemin et al., 2002; Molineux et al., 2006; Cain and Snutch, 2010).

T-type Ca²⁺ currents have been well documented in mammalian retinal neurons, including retinal bipolar cells (Kaneko et al., 1989; Protti & Llano, 1998; Pan, 2000) and ganglion cells (Sherwin et al., 2003). Previous studies reported the evidence of a differential expression of different isoforms of T-type Ca²⁺ channel α1 subunits among retinal bipolar cells (Hu et al., 2009). The expression patterns of the individual T-type Ca²⁺ channel subunits in the retina, however, remain unknown. In this study, using immunohistochemical analysis and electrophysiological recordings together with Ca_v3.2 knockout (KO) mice, we investigated the expression of Ca_v3.2 T-type Ca²⁺ channels in the mammalian retina. We found that Ca_v3.2 Ca²⁺ channels are expressed in a subtype of retinal cone bipolar cells (CBCs).

Methods:

HEK cell culture and DNA transfection

HEK-293 cells were maintained in Dulbecco's modified Eagle's medium (Gibco/BRL, Grand Island, NY) supplemented with 10% fetal bovine serum, 100 U/ml penicillin G, and 100 μg/ml streptomycin at 37°C in a humidified atmosphere of 95% air and 5% CO₂. For DNA transfection experiments, the cells were seeded in 35-mm dishes and transfected with cDNAs for rat Cav3.1, human Cav3.2, or rat Cav3.3 (kindly provided by Dr. Perez-Reyes) using Lipofectamine (Invitrogen, San Diego, CA). Immmunostaining was performed ~24 h after the transfection.

Animals

C57BL/6J mice were purchased from Jackson Laboratory (Bar Harbor, ME). A Ca_v3.2 KO mouse line, B6.129-Cacna1h^tm1Kcam, which developed in the Campbell laboratory at the University of Iowa (Chen et al., 2003), was obtained from the Mutant Mouse Regional Research Center (#009979-MU). The Ca_v3.2 KO mice were generated on a mixed C57BL/6J and 129 background and backcrossed to C57BL/6J mice for at least 6 generations. Mice were genotyped by PCR using the following primers: wild type forward primer, 5'- ATT CAA GGG CTT CCA CAG GGT A-3', mutant forward primer, 5'-GCT AAA GCG CAT GCT CCA GAC TG-3'; common reverse primer 5'-CAT CTC AGG GCC TCT GGA CCA C-3'. The PCR conditions were 94°C for 3 min, followed by 35 cycles of 94°C for 45 sec, 66°C for 45 sec, 72°C for 45 sec, and 72°C for 10 min. The PCR product sizes were as follows: mutant = 330 bp; heterozygote = 330 bp and 480 bp; wild type = 480 bp. All animal-handling procedures were approved by the Institutional Animal Care and Use Committee at Wayne State University and were in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Immunohistochemistry

Mice at 1–2 months of age were deeply anesthetized with CO₂ and decapitated. The retinas in the eyecup were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB) for 20 minutes. The isolated retinas were successively cryoprotected in a gradient of varying sucrose concentrations (10%, 20%, and 30% w/v in PB), and vertical sections (20 μm thick) were cut with a cryostat.

Retinal sections were blocked for 1 hour in PB containing 5% Chemiblocker (membrane-blocking agent; Chemicon), 0.5% Triton X-100 and 0.05% sodium azide (Sigma). The primary antibodies were goat polyclonal anti-T-type Ca²⁺ Ca_v3.2 (1:100; Cat # sc-16263; Santa Cruz), mouse anti-protein kinase A regulatory subunit IIβ (PKAIIβ, 1:80000; Cat # 610625; BD), rabbit anti-hyperpolarization-activated and cyclic nucleotide-gated channel 4 (HCN4, 1:500, Cat # APC-052, Alomone), rabbit anti-recoverin (1:2000; Cat # AB5585; Chemicon), mouse anti-calsenilin (1:2000; kindly provided by Dr. Wasco, Harvard Medical School), and mouse anti-calretinin (1:10000; Cat # MAB1568; Chemicon). The primary antibodies were diluted in the same PB solution and applied overnight, followed by incubation (1–2 h) in the secondary antibodies that were conjugated to Alexa 555 or Alexa 594 (1:600; red fluorescence; Invitrogen) or Alexa 488 (1:600; green fluorescence; Invitrogen). All steps were performed at room temperature.

All images were acquired using a Zeiss Axioplan 2 microscope with an Apotome oscillating grating to reduce out-of-focus stray light. Image projections were constructed by collapsing individual z-stacks of optical sections onto a single plane. The brightness and the contrast were adjusted using Adobe Photoshop.

Patch-clamp recordings

Retinal slices were prepared according to previously described procedures (Cui et al., 2003). The recording chamber was continuously superfused with bubbled Ames solution at a rate of ~2 ml/min. The electrode solution contained (in mM): CsCl 120, TEA-Cl 20, MgCl₂ 1, CaCl₂ 0.5, EGTA 5, HEPES 10, Na-GTP 0.5 and Mg-ATP 6; pH 7.3. The fluorescent dye Alexa 568 was added to the electrode solution at a concentration of 100 μM. The extracellular solution contained (in mM) 95 NaCl, 5 KCl, 10 CaCl₂, 10 CsCl, 20 TEA-Cl, 1 MgCl₂, 10 HEPES, and 22.2 glucose, with phenol red, 0.001% v/v, pH 7.2. In all recordings, 100 μM bicuculline, 100 μM TPMPA, 2 μM strychnine, and 1 μM tetrodotoxin (TTX) were included in the extracellular solutions. After the recordings, the retinal slices were fixed with 4% paraformaldehyde for 30 min, followed by blocking for 2 hrs in a solution containing 5% Chemiblocker. The slices were incubated with the primary antibody anti-calretinin at a dilution of 1:5000 for 3–4 days at 4°C. After washing with PB, the slices were treated with a secondary antibody (Donkey anti-mouse Alexa 488) at a dilution of 1:500 for 1 hr in the dark at RT. Specimens were viewed using a Zeiss Axiophot microscope.

Results:

The expression of Ca_v3.2 T-type Ca²⁺ channels in the retina was investigated by immunostaining. The specificity of the Ca_v3.2 Ca²⁺ channel antibody (Santa Cruz; Cat. 16263) was first examined against recombinant Ca_v3.2 T-type Ca²⁺ channels expressed in HEK cells. As shown in Figure 1, immunoreactivity of the Ca_v3.2 antibody was observed in HEK cells that were transfected with the Ca_v3.2 subunit (Fig. 1A) but not in those that were either untransfected (Fig. 1B) or transfected with Ca_v3.1 or Ca_v3.3 subunits (data not shown), indicating that the Ca_v3.2 antibody is specific for Ca_v3.2 channels. In retinal vertical sections, the Ca_v3.2 Ca²⁺ channel antibody was found to label a population of cells with their somata located in the inner nuclear layer (Fig. 2A). No staining was observed in retinal sections from Ca_v3.2 KO mice (Fig. 2B). The latter finding further confirmed the specificity of the Ca_v3.2 antibody in the mouse retina.

Figure 1. Confirmation of the specificity of a Ca_v3.2 Ca²⁺ channel antibody in an HEK cell expression system.

HEK cells transfected with (A) or without (B) Ca_v3.2 Ca²⁺ channel subunits and labeled with the goat anti-Ca_v3.2 antibody conjugated to Alexa 595. (A) Positive immunolabeling of Ca_v3.2 was observed in the Ca_v3.2 Ca²⁺ channel subunit-transfected HEK cells. (B) No immnunolabeling of Ca_v3.2 was observed in the untransfected HEK cells. The left and right panel show the transmission and fluorescence images, respectively. Scale bar: 100 μm.

Figure 2. Immunolabeling of Ca_v3.2 in retinal vertical sections of wild-type and Ca_v3.2 KO mice.

(A) The fluorescence signal of immunolabeling of Ca_v3.2 in the retinal vertical section of wild-type mouse. The Ca_v3.2 immunolabeled cells are bipolar cells with their dendrites located in the outer plexiform layer and axon terminals narrowly stratified in the inner plexiform layer. (B) No immunoactivity to Ca_v3.2 was detected in the retinas of Ca_v3.2 KO mice. (C) The colabeling of Ca_v3.2 (green) and calretinin (red) in the retinal vertical section of wild-type mouse. The axon terminals of Ca_v3.2-immunopositive bipolar cells were located between the two distal calretinin bands. Scale bar: 50 μm.

Based on their morphological properties, the Ca_v3.2-immunopositive cells were found to be bipolar cells because they exhibited dendrites located in the outer plexiform layer (OPL) and axon terminals that narrowly stratified in the inner plexiform layer (IPL) (Fig. 2A). To visualize the axon terminal stratification of the Ca_v3.2-immunopositive cells in the IPL, we performed double-labeling for Ca_v3.2 and calretinin, a commonly used marker for the stratification of the IPL (Haverkamp and Wässle, 2000). As shown in Figure 2C, the axon terminals of the Ca_v3.2-immunopositive cells were located between the two distal calretinin bands, which are known to define the stratum 2 of the IPL. In the rodent, only type-3 CBCs have their axon terminals stratified in the stratum 2 of the IPL (Euler and Wässle, 1995; Ghosh et al., 2004; Pignatelli and Strettoi, 2004). These results suggest that the Ca_v3.2-immunopositive cells are type-3 or a subpopulation of type-3 CBCs.

Two subtypes of type-3 CBCs, termed type-3a and type-3b, have been reported, which can be labeled by antibodies against HCN4 and PKAIIβ, respectively (Mataruga et al., 2007). To confirm that the Ca_v3.2-immunopositive cells are in fact type 3 CBCs and also to determine whether they could be subtype of type-3 CBCs, we performed double- labeling of Ca_v3.2 with HCN4 or PKAIIβ. As shown in Figures 3A–D, the Ca_v3.2-positive cells did not show colabeling with HCN4 as indicated by the lack of overlap staining both in the somata (Fig. 3C) and the axon terminals (Fig. 3D). However, the Ca_v3.2-positive cells showed colabeling with PKAIIβ (Fig. 3E–H), as demonstrated by the overlap of Ca_v3.2 and PKAIIβ staining in the somata (Fig. 3G) and the axon terminals (Fig. 3H). In addition, the Ca_v3.2-positive cells were colabeled neither with recoverin (Fig. 3I–L) nor calsenilin (Fig. 3M–Q), the markers for type-2 CBCs and type-4 CBCs, respectively. The latter finding ruled out the possibility of these cells being type-2 or type-4 CBCs. Taken together, our results indicate the expression of Ca_v3.2 Ca²⁺ channels in a subtype of type 3 CBCs – the PKAIIβ-positive type-3 or type-3b CBCs.

Figure 3. Double-labeling analysis of Ca_v3.2-positive bipolar cells in wild-type mouse retinas.

(A–D) Ca_v3.2-positive bipolar cells (A, green, asterisks) do not show labeling by an antibody against HCN4 (B, red, arrows) as indicated by the lack of colabeling of Ca_v3.2 and HCN4 in somata and axon terminals (D). To obtain a better view of the axon terminals, the rectangular region in (C) is shown in (D) at higher magnification. (E–H) Ca_v3.2-positive bipolar cells (E, green, asterisks) were also labeled by an antibody against PKAIIβ (F, red, arrows), as demonstrated by the co-labeling of Ca_v3.2 and PKAIIβ both in the soma and axon terminals (G and H). Ca_v3.2-positive bipolar cells did not show colabeling with recoverin (I–L) and calsenilin (M–Q). Scale bar: 50 μm for the images in the first three columns and 10 μm for the images in the last column.

We next determined whether Ca_v3.2 Ca²⁺ channels contribute to the T-type Ca²⁺ current in type-3 CBCs. For this purpose, whole-cell patch-clamp recordings were performed to record Ca²⁺ currents from type-3 CBCs in retinal slices. Type 3 CBCs were identified based on their axon terminal stratification in stratum 2 of the IPL with dye filling through the recording electrode following by immunolabeling for calretinin (Fig. 4A and B). We first examined the properties of T-type Ca²⁺ current in type-3 CBCs in wild-type animals (C57BL/6J mice). Because bipolar cells are known to possess both T- and L-type Ca²⁺ currents (de la Villa et al., 1998; Pan, 2000), to separate the T-type Ca²⁺ current from the L-type, the Ca²⁺ currents elicited at each test potential (ranging from −60 mV to +10 mV) were depolarized from two different pre-pulses (1 sec) at −80 mV and −50 mV (see left panel in Fig. 4C). The currents evoked from −50 mV should be the L-type Ca²⁺ current, while those evoked from −80 mV should represent a mixture of T- and L-type (the total currents). Therefore, the subtraction of the L-type current from the total current was used to estimate the T-type current (right panel in Fig. 4C). Figure 4D shows the I–V relationship for the total current (black), L-type (blue), and the estimated T-type (red). The T-type Ca²⁺ current was activated at a potential >−60 mV and reached its peak around −30 mV, while the L-type Ca²⁺ current was activated at a potential of >−40 mV and reached to its peak around −30 to −20 mV. It should note that, due to the different activation kinetics of these two currents, the sum of the T-type and L-type currents is greater than the total current for the step to >−40 mV.

Figure 4. Voltage-gated Ca²⁺ currents of the type-3 cone bipolar cells in wild-type mice.

(A) A recorded type-3 cone bipolar cell in a retinal slice preparation filled with the fluorescent dye Alexa 568 through the recording electrode. (B) The retinal slice was immunolabeled with an antibody against calretinin (green). (C) A representative recording of Ca²⁺ currents from type-3 cone bipolar cells (left panel). To separate T- and L-type Ca²⁺ currents, prepulses of −80 and −50 mV (1 s) were applied before a brief (5 ms) depolarization to −50 mV followed by 200 ms test pulses ranging from −60 to −30 mV. The subtracted currents (right panel) represent T-type currents. (D) The I–V relationships of the T-, L-, and the total Ca²⁺ currents (mean ± SE; n = 6).

Because the L-type Ca²⁺ current was not activated at a potential of −40 mV under our recording conditions, the T-type Ca²⁺ currents evoked by the test pulse to −40 mV from the holding potential of −80 mV in type-3 CBCs were compared in wild-type and Ca_v3.2 KO mice to determine the contribution of Ca_v3.2 Ca²⁺ channels to the T-type Ca²⁺ current. Because the two types of type-3 CBCs were unable to be distinguished morphologically, the recordings were made from all morphologically identified type-3 CBCs. Figure 5A shows the distribution plot of the current amplitude for the recorded type-3 CBCs in control (n = 29) and Ca_v3.2 KO (n = 25) mice. As shown in the Figure 5A, a portion of the recorded type-3 CBCs in Ca_v3.2 KO mice exhibited a marked decrease in the T-type Ca²⁺ current. The average of the T-type Ca²⁺ currents from all recorded type-3 CBCs in wild-type and Ca_v3.2 KO mice was 89.9 ± 28.2 (mean ± SD; n = 25) and 59.1 ± 34.5 (n = 29), respectively (Fig. 5B). These two values are significantly different (P<0.005).

Figure 5. Comparison of T-type currents in type-3 cone bipolar cells between wild-type and Ca_v3.2 KO mice.

The T-type Ca²⁺ currents were evoked by depolarizing the membrane potential to −40 mV from the holding potential of −80 mV. (A) The current amplitude distribution for all recorded cells from wild-type and Ca_v3.2 KO mice. (B) The average values of the T-type Ca²⁺ currents for wild-type and Ca_v3.2 KO mice. These two values are significantly different (mean ± SE; P < 0.005).

Discussion:

In this study, we investigated the expression of Ca_v3.2 Ca²⁺ channels in the mouse retina. We first validated the specificity of a Ca_v3.2 Ca²⁺ channel antibody in the HEK cell expression system and a Ca_v3.2 KO mouse line. Our immunostaining results showed that the Ca_v3.2 antibody labeled a population of bipolar cells with their axon terminals narrowly stratified in stratum 2 of the IPL, suggesting that they are type-3 CBCs according to the common classification scheme of bipolar cells in the rodent retina. Our detailed immunohistochemistry analysis with bipolar cell markers indicated that Ca_v3.2- immunopositive bipolar cells belong to a subtype of type-3 CBCs, the PKAβII- immunopositive type-3 or termed type-3b CBCs (Mataruga et al., 2007). Using whole-cell patch-clamp recordings and comparing the currents between wild-type and Ca_v3.2 KO mice, we provided evidence that Ca_v3.2 Ca²⁺ channels contribute to the T-type Ca²⁺ current in a subpopulation of type-3 CBCs. Taken together, our studies revealed for the first time the expression pattern of the Ca_v3.2 Ca²⁺ channel in the mouse retina.

Our immunostaining results show that labeling of Ca_v3.2 is observed throughout the Ca_v3.2-postive cell, indicating that the T-type Ca²⁺ channels are localized in the soma, dendrites, and axon terminals of type-3b CBCs. The immunostaining results are consistent with the findings of previous electrophysiological recordings that T-type Ca²⁺ channels are located both in the soma and at the axon terminal of bipolar cells (Pan, 2000; Pan et al., 2001). The subcellular localization pattern could be important for understanding the potential physiological functions of T-type Ca²⁺ channels in bipolar cells. In general, Ca²⁺ channels in dendrites can facilitate synaptic signal integration while Ca²⁺ channels at the axon terminals can facilitate or directly trigger transmitter release (Huguenard 1996; Dunlap et al., 1995).

Our electrophysiological results suggest that the T-type Ca²⁺ currents in the Ca_v3.2-expressing type-3b CBCs are likely to be mainly composed of Ca_v3.2 Ca²⁺ channels because a marked reduction of the T-type Ca²⁺ current was observed in a subgroup of type-3 CBCs in Ca_v3.2 KO mice. The results appear to also suggest that the type-3b T-type Ca²⁺ current is formed by other T-type Ca²⁺ channel α1 subunits in addition to Ca_v3.2 because the T-type Ca²⁺ current was not totally absent in Type-3 CBCs from KO mice. However, it should be noted that the Cav3.2 KO mouse line used in this study is a global knockout model (Chen et al., 2003). Therefore, it is possible that the remaining T-type Ca²⁺ current observed in type-3b CBCs from Cav3.2 KO mice is the result of functional compensation by other T-type Ca²⁺ channel α1 subunits due to knockout of Cav3.2. This question can be addressed when conditional knockout animal models become available. Nevertheless, our results suggest that the T-type Ca²⁺ currents in the PKAβII-positive type-3b CBCs are at least mainly composed of Ca_v3.2 channel subunits.

The finding in this study of the expression of Ca_v3.2 Ca²⁺ channels in a specific bipolar cell type supports previous reports of the heterogeneous properties of T-type Ca²⁺ currents among bipolar cells and a differential expression of three T-type Ca²⁺ channel subunits among different retinal bipolar cells (Pan, 2000; Hu et al., 2009). What would be the possible functional implications of the expression of Ca_v3.2 Ca²⁺ channels in a specific bipolar cell type in particular and the heterogeneous expression of different T-type Ca²⁺ channels in different bipolar cells in general? In the mammalian retina, there is one type of rod bipolar cells and more than 10 types of CBCs (Euler and Wässle, 1995; Ghosh et al., 2004; Pignatelli and Strettoi, 2004; Mataruga et al., 2007; Fyk-Kolodziej and Pourcho, 2007; Cui and Pan, 2008). Bipolar cells are commonly believed to play an important role in segregating visual information into multiple parallel information pathways in the retina (Wässle, 2004). T-type Ca²⁺ currents were previously shown to be essential for the initiation of spontaneous activity in bipolar cells and to contribute to regenerative spike-like potentials (Ma and Pan, 2003; Hu et al., 2009). Increasing evidence suggests that regenerative potentials or spiking activity may play a role in bipolar cell processing (Dreosti et al., 2011; Saszik and DeVries, 2012). Furthermore, T-type Ca²⁺ channels at the axon terminal of bipolar cells are capable of triggering transmitter release (Pan et al., 2001; Singer and Diamond, 2003). Because T-type Ca²⁺ currents with different α1 subunits exhibit different activation and inactivation kinetics and could be subjected to distinct regulation by endogenous substances (Todorovic et al., 2001; Chemin et al., 2006; Traboulsie et al., 2007; Hildebrand et al., 2007; Nelson et al., 2007; Perez-Reyes, 2010), a differential expression of multiple T-type Ca²⁺ channels in different bipolar cell types could result in diversified signal processing and thus contribute to parallel visual information processing in the retina. In particular, the fast inactivation of T-type Ca²⁺ currents could be used for transmitting high frequency visual signals, possibly related to processing temporal visual information, such as motion and directional selectivity. In this regard, it is interesting to note that the axon terminals of type-3 CBCs show overlap with OFF cholinergic band. Therefore, it is possible that the Cav3.2-positive cells form synaptic contact with cholinergic amacrine cells, the cells underlying the generation of direction selectivity in the retina (Lee and Zhou, 2006; Demb, 2007), and/or direction- selective retinal ganglion cells. Thus, further studies are needed to investigate whether the Cav3.2 Ca²⁺ channels in specific and the T-type Ca²⁺ currents in general can contribute to temporal information processing in the retina.

Highlights

• Ca_v3.2 Ca²⁺ channel antibody was confirmed in HEK cell expression system and in Ca_v3.2 KO mice.

• Immunostaining shows the expression of Ca_v3.2 Ca²⁺ channels in a subgroup of type-3 bipolar cells.

• The immunostaining results were confirmed by patch-clamp recordings.

Abbreviations

KO

knock out

CBC

cone bipolar cell

HEK

human embryonic kidney

PB

phosphate buffer

OPL

outer plexiform layer

IPL

inner plexiform layer

HCN4

hyperpolarization-activated and cyclic nucleotide-gated channel 4