Modulation of Cav1.3 Ca2+ channel gating by Rab3 ...

Molecular and Cellular Neuroscience 44 (2010) 246–259

Contents lists available at ScienceDirect

Molecular and Cellular Neuroscience j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y m c n e

Modulation of Cav1.3 Ca2+ channel gating by Rab3 interacting molecule Mathias Gebhart a,b,1, Gabriella Juhasz-Vedres a,b,1, Annalisa Zuccotti c, Niels Brandt d, Jutta Engel d, Alexander Trockenbacher a,b, Gurjot Kaur a,b, Gerald J. Obermair e, Marlies Knipper c, Alexandra Koschak a,b, Jörg Striessnig a,⁎ a

Institute of Pharmacy, Pharmacology and Toxicology, University of Innsbruck, Peter-Mayr-Strasse 1/I, A-6020 Innsbruck, Austria Center of Molecular Biosciences Innsbruck (CMBI), Innsbruck, Austria Department of Otolaryngology, Tübingen Hearing Research Centre, Molecular Physiology of Hearing, University of Tübingen, Elfriede-Aulhorn-Str. 5, D-72076 Tübingen, Germany d Department of Biophysics, Saarland University, Building 76, D-66421 Homburg/Saar, Germany e Institute of Physiology and Medical Physics, Medical University of Innsbruck, Fritz-Preglstr. 3, A-6020 Innsbruck, Austria b c

a r t i c l e

i n f o

Article history: Received 15 February 2010 Revised 19 March 2010 Accepted 25 March 2010 Available online 2 April 2010 Keywords: Calcium channels Channel gating Calcium current inactivation Inner hair cells Hearing

a b s t r a c t Neurotransmitter release and spontaneous action potentials during cochlear inner hair cell (IHC) development depend on the activity of Cav1.3 voltage-gated L-type Ca2+ channels. Their voltage- and Ca2+-dependent inactivation kinetics are slower than in other tissues but the underlying molecular mechanisms are not yet understood. We found that Rab3-interacting molecule-2α (RIM2α) mRNA is expressed in immature cochlear IHCs and the protein co-localizes with Cav1.3 in the same presynaptic compartment of IHCs. Expression of RIM proteins in tsA-201 cells revealed binding to the β-subunit of the channel complex and RIM-induced slowing of both Ca2+- and voltage-dependent inactivation of Cav1.3 channels. By inhibiting inactivation, RIM induced a non-inactivating current component typical for IHC Cav1.3 currents which should allow these channels to carry a substantial window current during prolonged depolarizations. These data suggest that RIM2 contributes to the stabilization of Cav1.3 gating kinetics in immature IHCs. © 2010 Elsevier Inc. All rights reserved.

Introduction Depolarization-induced Ca2+ entry through voltage-gated Ca2+ channels (VGCC) into electrically excitable cells is a key process regulating numerous physiological processes. Ten Ca2+ channel isoforms within three classes (Cav1–3) with different biophysical properties and subcellular localizations (Catterall et al., 2005) accomplish these diverse functions. Among isoforms gating is further fine-tuned by alternative splicing (Lipscombe and Raingo, 2007; Singh et al., 2008), accessory α2-δ and β-subunits (Davies et al., 2007; Dolphin, 2003) as well as by other channel associated proteins (CalinJageman and Lee, 2008; Dai et al., 2009). Among the high voltage activated Ca2+ channels Cav2 channels predominantly control presynaptic neurotransmitter release in neurons whereas postsynaptic Ca2+ influx through Cav1 (L-type) Ca2+ channels (LTCCs) modifies gene transcription and synaptic plasticity (Gomez-Ospina et al., 2006; Zhang et al., 2006). However, presynaptic neurotransmitter release at ribbon synapses from sensory cells of retinal photoreceptors and the cochlea is under the control of Cav1 rather than Cav2 channels. Tonic neurotransmitter release in response to light- or sound-evoked graded changes in membrane potential between −60 and −40 mV ⁎ Corresponding author. E-mail address: [email protected] (J. Striessnig). 1 These authors contributed equally to this work. 1044-7431/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.mcn.2010.03.011

requires unusually slow inactivation of L-type Ca2+ currents in photoreceptors (Rabl and Thoreson, 2002) and cochlear inner hair cells (Grant and Fuchs, 2008; Johnson and Marcotti, 2008; Lee et al., 2007) and maintenance of window currents over prolonged time periods (McRory et al., 2004). Slow inactivation in IHCs is also a prerequisite to produce Ca2+ signals and spontaneous action potentials during IHC development (Marcotti et al., 2003). In VGCCs inactivation during depolarizations is driven by the Ca2+ concentration sensed at the inner channel mouth (Ca2+-dependent inactivation, CDI) and by transmembrane voltage (voltage-dependent inactivation, VDI). To prevent efficient inactivation by these processes, LTCCs in photoreceptors developed special strategies. Photoreceptor L-type currents are largely carried by Cav1.4 LTCCs which auto-inhibit their own calmodulin (CaM)-dependent CDI by an intramolecular protein interaction within their C-terminus and their remaining VDI is intrinsically slow (Singh et al., 2006). Cav1.3 channels in the heart and brain display pronounced CDI and fast VDI (Koschak et al., 2001; Mangoni et al., 2003; Yang et al., 2006) but both processes are very slow in Cav1.3 channels of cochlear inner hair cells (Grant and Fuchs, 2008; Johnson and Marcotti, 2008; Marcotti et al., 2003). Strongly reduced CDI in IHCs can be explained by Ca2+ binding proteins, such as CaBP1 and CaBP4 (Cui et al., 2007; Lee et al., 2007; Striessnig, 2007; Yang et al., 2006), which compete with CaM binding and Ca2+ sensing to the Cav1.3 α1 subunit. Cavβ2 was recently shown to slightly affect CDI in Cavβ2 deficient IHCs, however VDI remained unaltered (Neef

M. Gebhart et al. / Molecular and Cellular Neuroscience 44 (2010) 246–259

et al., 2009). Even if CDI is completely inhibited by these proteins, the remaining VDI of Cav1.3 currents in IHCs is still much slower than the VDI of Cav1.3 currents in heart, brain or heterologously expressed

247

Cav1.3 channel complexes (Koschak et al., 2001; Mangoni et al., 2003; Platzer et al., 2000). So far the molecular basis for this physiologically relevant difference is unclear. The underlying mechanisms could be

Fig. 1. A flag-tagged I–II-linker of Cav1.3 channels targets β subunits to the plasma membrane in tsA-201 cells. Immunofluorescence images of C-terminally V5-tagged β subunits and C-terminally flag-tagged I–II-linker of Cav1.3 (I–IIflag) heterologously expressed in tsA-201 cells. (A–C) Expression of I–IIflag (A), palmitoylated β2aV5 (B), and the palmitoylation deficient C3S/C4Sβ2aV5 (C). (D) Expression of only β1aV5, β3V5, β4V5, and C3S/C4Sβ2aV5. Note that β4V5 shows nuclear localization as expected from previous studies. (E) Co-expression of the flag tagged I–II-linker and the indicated C-terminally V5 tagged β-subunits. (F) Expression of either I–IIflag W441A alone or together with C3S/C4Sβ2aV5, respectively. Representative cells from 3 independent experiments are shown. Scale bar = 10 µm.

248


the expression of Cav1.3 α1 subunit splice variants with slow VDI in IHCs, stabilization of slow VDI by accessory subunits or by proteins associated with the presynaptic signaling complex at ribbon synapses. So far, no splice variants slowing the VDI have been reported in IHCs or in other adult tissues (Klugbauer et al., 2002) and none of the accessory subunits, not even β2a can explain this effect (Neef et al., 2009). Some modest reduction of VDI has been observed for CaBP1 (Cui et al., 2007) and syntaxin (Song et al., 2003) when co-expressed with the channel complex in HEK-293 cells but neither of them can account for the slow inactivation time constants observed in IHCs. Recently the presynaptic Rab3 interacting molecule (RIM; Wang et al., 1997), a scaffold protein at the presynaptic active zone involved in Ca2+-induced neurotransmitter release in neurons (Sudhof, 2004; Schöch et al., 2005) has been identified as a Ca2+ channel modulator. It markedly suppressed the inactivation time course of presynaptic Cav2 channels and shifted the voltage-dependence of inactivation to more depolarized voltages (Kiyonaka et al., 2007). The modulatory effects of RIM are mediated via its tight binding to the β-subunit of the Ca2+ channel complexes. Due to its presynaptic location RIM effects

were extensively studied on presynaptically localized Cav2 channels. Considering that Cav1.3 also represents a presynaptic VGCC in IHCs, RIM may be co-localized with LTCCs in the active zone of IHCs. In the present study we therefore investigated if RIM proteins are expressed in IHCs, if they are capable of modulating Cav1.3 function and to which extent they could contribute to the slow VDI and CDI of Cav1.3 currents in cochlear IHCs. Results Using a yeast two hybrid approach with the I–II linker (cytoplasmic I–II linker of Cav1.3 α1 subunit) together with β2a as the bait complex, a human fetal brain cDNA library expressed as N-terminally myristoylated prey peptides was screened (see Experimental methods). RIM1α (NM_014989) was identified as a potential interaction partner of the Cav1.3α1/β2a complex. The RIM clone identified encoded the C-terminus (aa 1503 to 1692) of human RIM1α containing the entire C2B domain, a highly conserved structural domain present in all known RIM isoforms (Wang and Sudhof, 2003). Because RIM

Fig. 2. C-terminus of RIM1 interacts with β subunits in tsA-201 cells. Immunofluorescence images of the N-terminally HA-tagged C-terminal fragment of RIM1 (aa 1503–1692) (HARIM1C) co-expressed with palmitoylation deficient V5 labeled β2a (C3S/C4Sβ2aV5, A) or β3 (β3V5, B) and the untagged I–II-linker. β-subunits and the I–II-linker were expressed from the same plasmid (pBudCE4.1). Immunofluorescence images show HARIM1C at the plasma membrane. Inset: HARIM1C is distributed throughout the cell when expressed alone. (C) C-terminally V5 tagged β2a (β2V5) subunit co-expressed with HARIM1c in the absence of the I–II-linker. (D) Co-expression of HARIM1C with I–IIflag. Representative cells from a least 3 independent experiments are shown. Scale bar = 10 µm.


proteins act as scaffold proteins in presynaptic active zones not only in neurons but also in the ribbon synapses of photoreceptors (tom Dieck et al., 2005; Wang et al., 1997) they provide attractive candidates as modulatory proteins in ribbon synapses of cochlear IHCs that could participate in the stabilization of slow VDI of Cav1.3 channels, a role which has not yet been studied.

The Cav1 I–II linker targets β-subunits and the β/RIM complex to the plasma membrane For RIM1 a modulatory role for Ca2+ channels has been shown through interaction between the RIM C2B domain and β subunits (Kiyonaka et al., 2007). To prove that RIM binds to β subunits in living cells, we established a biochemical assay that exploited the finding that the I–II linker (i.e. the intracellular sequence connecting homologous domains I and II) of Cav1.3 alone can target to the plasma membrane in tsA-201 cells (shown for the flag-tagged I–II-linker, I–IIflag, in Fig. 1A; Takahashi et al., 2005). This is similar to the well-known plasma membrane targeting of the palmitoylated splice variant of β2, β2a (as shown for β2aV5 in Fig. 1B). Because the I–II-linker contains the major β-subunit interaction site AID (alpha-interaction domain), we reasoned that the linker should also be able to translocate β subunits to the plasma membrane. As expected, the V5-tagged β-subunits β1aV5, β3V5, and the palmitoylation-deficient β2a mutant (C3S/C4Sβ2aV5) were distributed evenly in the cytoplasm (Fig. 1C, D). In agreement with previous findings (Colecraft et al., 2002; Hibino et al., 2003; Subramanyam et al., 2009), β4V5 showed nuclear targeting (Fig. 1D, β4). However, when β subunits were co-expressed together with I–IIflag each β-subunit was efficiently targeted to the plasma membrane (Fig. 1E), including the non-palmitoylated C3S/C4Sβ2aV5 mutant (Fig. 1E). Plasma membrane targeting was not only observed for the Cav1.3 I–II linker but also found for the corresponding linkers of Cav1.1, Cav1.2 and Cav1.4 LTCCs (n = 3, data not illustrated).

249

The membrane targeting of I–IIflag could have been due to the interaction with a β-subunit endogenously expressed in tsA-201 cells. To test for this possibility we introduced a mutation within the AID of I–IIflag (I–IIW441A ) which has been shown to prevent β-AID interaction flag and therefore membrane targeting (Leroy et al., 2005; Obermair et al., 2010). I–IIW441A was still targeted to the plasma membrane when flag expressed alone but was unable to alter the cytoplasmatic distribution of co-expressed β subunits (shown for C3S/C4Sβ2aV5 in Fig. 1F). Therefore our data suggest that the I–II linker can target to the plasma membrane in a β-subunit independent manner but that its ability to induce plasma membrane targeting of β-subunits requires high affinity binding to the AID (Pragnell et al., 1994). To confirm an interaction between RIM protein and the β-subunits, we investigated, if RIM cotargets with β subunits to the plasma membrane. We expressed the N-terminally HA-tagged C-terminus of human RIM1 (residues 1503– 1692 of RIM1α; HARIM1C) with or without the untagged Cav1.3 I–II linker and the C-terminally V5-tagged β subunits, which were both expressed from the same vector (pBudCE4.1, see Experimental methods). Transfection of HARIM1C alone resulted in a distribution throughout the cell (Fig. 2A, middle, inset). Likewise, the co-expressed I–IIflag linker targeted to the membrane but did not redistribute HARIM1C (Fig. 2D). In contrast, HARIM1C co-localized at the plasma membrane when C3S/C4Sβ2aV5 (Fig. 2A) or β3V5 (Fig. 2B) were also expressed in the presence of the I–II linker. As expected, plasma membrane targeting of HARIM1C was also seen upon co-expression with the palmitoylated wild-type β2aV5 in the absence of I–II (Fig. 2C). This shows that the RIM1 protein binds to β-subunits rather than to the I–II linker. In a similar manner a full-length human RIM2β construct N-terminally tagged with the monomeric red fluorescent protein (mRFPRIM2β) was shuttled from the cytoplasm (Fig. 3A inset) to the plasma membrane in the same β-subunit-dependent manner (Figs. 3B, C). Taken together these results clearly demonstrate that RIM interacts with β-subunits but not with the I–II linker under experimental conditions used for our further functional analysis.

Fig. 3. Full-length RIM2β interacts with β subunits C-terminus of RIM1 interacts with β subunits. (A) Immunofluorescence images of N-terminally monomeric red fluorescent protein labeled RIM2β (mRFPRIM2β) co-expressed with V5 tagged β2a (β2V5) subunit in tsA-201 cells. Inset: mRFPRIM2β expressed alone. B, C: mRFPRIM2β co-expressed with C3S/C4Sβ2aV5 (B) or β3V5 (C) together with the untagged I–II-linker from the same plasmid. Representative cells from at least 3 independent experiments are shown. Scale bar = 10 µm.

250


Fig. 4. Expression of RIM transcripts in the total organ of Corti and IHCs. All experimental details and primers are given in Experimental methods. Adult mouse whole brain cDNA (b) was used as positive control. In the negative control (−), no cDNA was added. (A) RT-PCR experiments detecting RIM isoform transcripts in mouse organ of Corti at developmental stages P4 and P20. (B) RT-PCR (nested approach, see Experimental methods) experiments showing RIM2 isoform expression in preparations from pools of IHCs at P4 and P20.

RIM2 is expressed in the mouse cochlea and IHCs As it was unknown if RIM isoforms were expressed in IHCs we performed RT-PCR analysis with RNA isolated from whole organ of Corti

and IHCs using specific primers for RIM1α, RIM2α, RIM2β and RIM3γ mRNAs. Mouse whole brain cDNA was used as a positive control. As illustrated in Fig. 4A, RIM1α (3 out of 3), RIM2α (7 out of 7) and RIM2β isoforms (5 out of 5) were reproducibly detected in independent RNA

Fig. 5. Immunolocalization of RIM2, Cav1.3 α1 and CtBP2 in mouse IHCs. Immunohistochemical staining of mouse cochlear sections at P6 using anti-RIM2, anti-Cav1.3 α1 and antiCtBP2/RIBEYE antibodies. Puncta with immunoreactivity for RIM2 (A, red), Cav1.3 (E, red) and CtBP2/RIBEYE (B, F, green) were observed at the basal pole of IHCs. The partial colocalization of RIM2 and CtBP2/RIBEYE proteins (C) and Cav1.3 and CtBP2/RIBEYE proteins (G) becomes evident upon merging corresponding images. Images in D and H correspond to the boxed regions of interest in C and G, respectively. Cell nuclei were counterstained with DAPI (blue). Scale bar indicates 10 μm. Numbers of immunopositive spots for CtBP2/ RIBEYE, RIM2 and Cav1.3 were measured from eight immature IHCs. Data were normalized to the number of CtBP2/RIBEYE-positive spots (100%).


Fig. 6. Expression of Cav1.3 N-terminal splice variant transcripts α1a and α1b in the mouse organ of Corti. RT-PCR analysis of Cav1.3 α1a (A) and α1b (B) splice variants, in organ of Corti preparations from P4 mice (n = 8). All experimental details and primers are given in Experimental methods. Total embryonic (E11,e) and brain (b) cDNA were used as positive controls. In the negative control (−), no cDNA was added.

preparations of total organ of Corti at developmental stage P4. None of the RIM transcripts were detected at a later developmental stage (P20) when hearing is fully developed (Fig. 4A, P20, 3 independent RNA preparations). In immature IHCs RIM2α (12 out of 14 different IHC preparations) but no RIM2β transcripts (6 independent RNA preparations) could be detected by standard RT-PCR. Also, when using a nested PCR approach RIM2α was only detected at an early developmental stage (Fig. 4B, n = 3) and RIM2β was undetectable in IHCs (Fig. 4B, n = 3). We further detected RIM3γ in three independent whole organ of Corti preparations at P4 but not at P20 (Fig. 4A). RIM1α and RIM3γ were absent in nested PCR experiments with IHCs at P4 (n = 3, not illustrated). These data demonstrate that RIM1 and RIM2 proteins are expressed in the mouse organ of Corti but that RIM2α is the predominant isoform in immature IHCs.

RIM2 and Cav1.3 co-localize with CtBP2 in IHCs Due to the prevalent expression of RIM2 in immature IHCs in RTPCR experiments, we investigated the subcellular distribution of RIM2 and Cav1.3 α1-subunit protein in immature IHCs in parallel experiments. To provide evidence for co-localization of these proteins we had to rely on an indirect approach through the ribbon marker CtBP2/ RIBEYE (C-terminal binding protein 2) that is known to co-localize with Cav1.3 (Brandt et al., 2005). Direct double staining experiments could not be performed as suitable anti-Cav1.3 and anti-RIM2 antibodies were available only from the same species (see Experimental methods). The specificity of the Cav1.3 antibody was verified in staining experiments using Cav1.3 knockout mice and preincubating the antibody with the antigenic peptide (not illustrated; Platzer et al., 2000; Zampini et al., 2010).

251

We have recently reported that in the immature IHCs ∼50% of ribbons co-localize with Ca2+ channels at the mouse IHC presynaptic region (Zampini et al., 2010). Focusing restrictively on the basolateral part of IHCs, RIM2 (Fig. 5A, red), Cav1.3 α1-subunits (Fig. 5E, red) and CtBP2/RIBEYE (Fig. 5B, F, green) proteins were localized in a prominent dot-like staining around the basolateral pole of the IHCs. The merged signal within the CtBP2/RIBEYE immunopositive region revealed a prominent co-localization of RIM2 (Fig. 5C, D white arrows) and Cav1.3 (Fig. 5G, H white arrows) with IHC synaptic ribbons. In immature (P6) mouse cochlear sections, a total number of 35.6 ± 5.1 (mean± SEM) CtBP2/RIBEYE immunopositive ribbons was detected, in agreement with previous findings (Zampini et al., 2010). 63% of ribbons were colocalized with RIM2 and 48% with Cav1.3 (Fig. 5J). These data imply that at least 11% of Cav1.3 channels co-localize with RIM2 at presynaptic release sites in immature IHCs. However, considering an approximate size of ∼ 376 nm of a ribbon and our optical resolution of ∼257 nm, this criterion for “overlap” (see Experimental methods) rather underestimates the extent of co-localization. Thus co-localization of these two proteins should allow RIM2 to functionally modulate a significant portion of Cav1.3 channels. Interestingly, co-localization of RIM2 (but not RIM1) with LTCCs has recently also been found at ribbon synapses of retinal photoreceptors (tom Dieck et al., 2005; Regus-Leidig et al., 2009). Slow Cav1.3 current inactivation does not arise from an embryonic slowly inactivating Cav1.3 (Cav1.3α1b) splice variant Next we determined if a slowly inactivating N-terminal Cav1.3 α1subunit splice variant is expressed in IHCs. Cav1.3 α1a and α1b splice variants have previously been described in embryonic mouse heart resulting from alternative usage of exons 1a and 1b (Klugbauer et al., 2002). VDI of α1b channels is slower than of α1a and, hence, could already contribute to the slow VDI observed in IHCs. Using a nested PCR approach we reproducibly detected α1b transcripts in control experiments with mouse embryonic RNA at developmental stage E11 (Fig. 6B, 8 of 8 preparations) but reproducibly failed to detect such transcripts in the whole organ of Corti at P4 (Fig. 6B), a stage at which RIM2α was expressed. In contrast, the α1a variant was reproducibly found in all organ of Corti preparations tested (Fig. 6A, 6 of 6) as well as in mouse embryonic cDNA at E11 (Fig. 6A, 5 of 5) and adult mouse brain (Fig. 6A, 5 of 5). We therefore concluded that at P4, Cav1.3α1a is the predominant N-terminal splice variant with no major contribution of the embryonic α1b variant. RIM2 affects Cav1.3 channel gating To address the question whether RIM modulates Cav1.3 channels we co-expressed N-terminally GFP-labeled RIM C-termini (RIM1C,

Fig. 7. Effects of RIM2C on Cav2.1 channel function. Cav2.1 α1 subunits were co-expressed with β3 and α2δ1 in tsA-201 cells in the absence (control) or presence of RIM2C (grey). 15 mM Ca2+ was used as charge carrier. (A) Inactivation time course for normalized ICa during 2.5-s depolarizing pulses from a holding potential of −80 mV to Vmax is depicted. (B) Percentage of inactivation of ICa was analyzed at the indicated pre-specified time periods. **p b 0.01 (Mann–Whitney test). Error bars reflect SEM for the indicated number of experiments. (C) Voltage-dependence of ICa inactivation. Inactivation curves were obtained by test potentials to Vmax after 5 s pulses to the indicated voltages. The solid line is a fit to the Boltzmann relationship. Inactivation parameters for Cav2.1 in the absence and presence of RIM2C are given in Table 3.

252


Fig. 8. Effects of RIM2C on Cav1.3 channel function. Cav1.3 α1 subunits were co-expressed with β3 and α2δ1 in tsA201 cells in the absence (black) or presence of RIM2C (grey). Either 15 mM Ca2+ or 15 mM Ba2+ was used as charge carrier. (A) Peak current–voltage (I–V) relations for ICa were measured as described in Experimental methods. For activation parameters see Table 1. (B) Voltage dependence of inactivation (5 s preconditioning pulses to the indicated potentials). Inactivation parameters were obtained from fits to a Boltzmann function and are given in Table 3. (C) Inactivation time course of Cav1.3 ICa during 5-s depolarizing pulses from a holding potential of −80 mV to Vmax. (D, E) Percentage of inactivation of ICa (D) and IBa (E) calculated at the indicated pre-specified time points. Statistically significant differences (one-way ANOVA followed by Bonferroni post-test) are indicated: with vs. without RIM: ***p b 0.001; **p b 0.01; differences IBa vs. ICa) +++p b 0.001; ++p b 0.01. Error bars reflect SEM for the indicated number of experiments. (F,G) Mean inactivation of ICa and IBa from several experiments (7–25 experiments) was fitted to a bi-exponential decay. The obtained parameters for fast and slowly inactivating components are listed in Table 4. Inset: Quantification of CDI by plotting the fraction of ICa and IBa remaining after 50 ms (r50) depolarizations to the indicated voltages; f is the maximal difference between r50 values of ICa and IBa at + 10 mV.

and RIM2C) containing the interacting C2B domain with channel complexes in tsA-201 cells. In control experiments (Fig. 7), RIM2C coexpressed with Cav2.1 channel complexes (Cav2.1α1 + α2δ1 + β3)

modulated gating as previously described for C2B containing RIM1 fragments (Kiyonaka et al., 2007). Inactivation of ICa during 5-s depolarizing pulses was strongly reduced (Fig. 7A, B Table 2) and

Table 1 Effect of RIM1C and RIM2C on Cav1.3-mediated Ca2+ currents (ICa). Channel Cav1.3+ Cav1.3+ Cav1.3+ Cav1.3+ Cav1.3+ Cav1.3+ Cav1.3+ Cav2.1+ Cav2.1+

β3 β3+ RIM1C β3+ RIM2C β2a β2a + RIM2C C3S/C4Sβ2a C3S/C4Sβ2a + RIM2C β3 β3 RIM2C

V0.5,act (mV)

kact (mV)

Vmax (mV)

Vrev (mV)

Activation threshold (mV)

n

% inact. 0.25 s

% inact. 1 s

% inact 5 s

n

−0.9 ± 0.9 9.6 ± 1⁎⁎⁎ 7.2 ± 1⁎⁎ 2.3 ± 1.8 0.94 ± 2.1 −2.7 ± 1.5 3.9 ± 1.3⁎

9 ± 0.2 10.3 ± 0.3⁎⁎ 11.2 ± 0.4⁎⁎⁎ 10.15 ± 0.3 10.1 ± 0.3 9.3 ± 0.3 9.7 ± 0.4 3.4 ± 0.3 3.5 ± 0.2

14.4 ± 0.9 24.2 ± 0.6⁎⁎⁎ 22.5 ± 0.8⁎⁎⁎ 18.0 ± 1.5 71.6 ± 1.7⁎

71.7 ± 0.8 76.6 ± 1.4⁎⁎ 75.9 ± 1.4⁎ 76.5 ± 1 16.0 ± 1.9 75 ± 1.6 74 ± 0.4 67.5 ± 0.3 66.3 ± 0.6

32.1 ± 0.6 28.2 ± 0.9⁎⁎ 32.0 ± 1 32.9 ± 0.8 34.3 ± 1.2 34.6 ± 1.1 30.5 ± 1.1⁎

29 12 8 9 9 10 9 7 8

66.6 ± 2.5 27 ± 1.6⁎⁎⁎ 28.7 ± 2.5⁎⁎⁎ 36.4 ± 4.0+++ 25.7 ± 7.0⁎

89.0 ± 1.0 57.0 ± 2.9⁎⁎⁎ 52.2 ± 2.4⁎⁎⁎ 56.3 ± 4.7+++ 49.85 ± 6.3⁎

95.5 ± 0.7 82.7 ± 3.5⁎⁎⁎ 78.6 ± 2.4⁎⁎⁎ 76.0 ± 4.1+++ 74.8 ± 6.0⁎

27 9 7 8 5 8 6 9 8

8.7 ± 1 7.3 ± 1.3

13.0 ± 1.3 18.5 ± 1.2⁎ 17.5 ± 1.4 16.4 ± 1.4

3.9 ± 0.7 4.8 ± 1.0

64.7 ± 2.6+ 24.4 ±2.9⁎⁎⁎ 45.8 ± 4 3.8 ± 0.8⁎⁎⁎

86 ± 2.2+ 47.2 ± 3.7⁎⁎⁎ 88.2 ± 2.2 15.1 ± 1.5⁎⁎⁎

96.3 ± 0.6+ 74.4 ± 3.8⁎⁎⁎ 98.2 ± 0.8 56.7 ± 2.4⁎⁎

The indicated α1 and β-subunits were co-expressed with α2δ-1 in tsA-201 cells, and the biophysical properties were determined using 15 mM Ca2+ as a charge carrier. V0.5,act, kact, and Vmax were obtained by fitting the data as described in Experimental methods. The activation threshold was determined as the test potential at which 5% of the maximal current was activated. Data are given as means ± SEM. Statistically significant differences between data obtained in the absence or presence of RIM constructs for a given subunit composition (calculated by one-way ANOVA, followed by Bonferroni test) are indicated: *p b 0.05; **p b 0.01; ***p b 0.001.


253

Table 2 Effects of RIM2C on Cav1.3-mediated Ba2+ currents (IBa). Channel

V0.5,act (mV)

kact (mV)

Vmax (mV)

Vrev (mV)

Activation threshold (mV)

n

% inact. 0.25 s

% inact. 1 s

% inact 5 s

n

Cav1.3+ β3 Cav1.3+ β3 + RIM2C Cav1.3+ β2a Cav1.3 + β2a + RIM2C Cav2.1+β3 Cav2.1+ β3 + RIM2C

−10.0 ± 1.3 −9.5 ± 1.9 2.3 ± 1.8 0.9 ± 2.1 0.07 ± 0.9 −2.9 ± 1.8

−7.7 ± 0.3 −9.3 ± 0.3 −10.15 ± 0.3 −10.1 ± 0.3 −4 ± 0.1 −3.8 ± 0.3

5.0 ± 1.25 5.7 ± 1.5 18.1 ± 1.5 16.0 ± 1.9 9.6 ± 1.1 6.1 ± 1.9

65.4 ± 0.9 66 ± 1.4 76.5 ± 1.03 71.6 ± 1.7* 59.4 ± 1.2 56.3 ± 0.7

−35.6 ± 0.6 −41.2 ± 0.7** −32.9 ± 0.8 −34.3 ± 1.2 −14.6 ± 0.7 −16.5 ± 1.4

7 10 9 9 8 8

30.2 ± 6.5 12.9 ± 2.8*** 13.8 ± 1.5+++ 7.7 ± 2.4* 62.9 ± 7.4 3.8 ± 0.8***

53.8 ± 6.1 32 ± 4.8*** 26.3 ± 2.9+++ 25.6 ± 4.2* 92.6 ± 1.9 14.7 ± .0***

77.1 ± 3.3 69.4 ± 5.9*** 56.8 ± 3.8+++ 62.4 ± 3.8* 97.3 ± 0.9 50.8 ± 4.2***

7 7 10 10 5 6

The indicated α1 and β-subunits were co-expressed with α2δ-1 in tsA-201 cells and the biophysical properties were determined using 15 mM Ba2+ as a charge carrier. V0.5,act, kact, and Vmax were obtained by fitting the data as described in Experimental methods. The activation threshold was determined as the test potential at which 5% of the maximal current was activated. Data are given as means ± SEM. Statistically significant differences between data obtained in the absence or presence of RIM2c for a given subunit composition (calculated by one-way ANOVA, followed by Bonferroni test) are indicated: *p b 0.05; **p b 0.01; ***p b 0.001.

steady-state inactivation curves (5-s conditioning pulses) were shifted to more positive voltages (Fig. 7C, Table 3). RIM2C and RIM1C also affected the activation and inactivation gating properties of Cav1.3 (Fig. 8, Tables 1–4). When co-expressed with Cav1.3 channel complexes (Cav1.3α1 + α2δ1 + β3) both RIM constructs significantly shifted the V0.5,act of ICa activation to more depolarized potentials by 8–10 mV. This was mainly due to an increased slope factor (Fig. 8A, Table 1). As shown for RIM2C, this shift was only observed with Ca2+, but not with Ba2+ as a charge carrier (Table 2). The half maximal voltage of ICa steady-state inactivation (V0.5,inact) was also significantly shifted to more positive voltages by about 10 mV by RIM (Fig. 8B, Table 3). This effect was most pronounced at −20 mV. Note that the voltage dependence of activation and inactivation is shifted by about 15 mV to more positive voltages under our recording conditions (15 mM Ca2+ as charge carrier, Xu and Lipscombe, 2001) corresponding to about −35 mV at physiological Ca2+ concentrations. The operating voltage range of IHC is between approximately −35 and −60 mV. Association of Cav1.3 with RIM therefore slows its inactivation at the positive end of IHCs operating voltage-range. These findings imply that a substantially larger window Ca2+ current would be maintained over time at synaptic ribbons in which Cav1.3 channels are associated with RIM2. Both RIM constructs slowed the time course of ICa inactivation during 5-s depolarizing test pulses to Vmax, with significant differences at all pre-specified time points analyzed (Figs. 8C, D and Table 1). During 5-s depolarizations to even very positive voltages (Vmax and beyond, Fig. 8B, D), RIM co-expression also increased the fraction of current not inactivating by 4- to 5-fold (Table 1). To distinguish whether the slower ICa inactivation induced by RIM arose from a moderation of CDI or a slowing of VDI or both, we also analyzed RIM effects on Cav1.3 currents using Ba2+ as charge carrier. IBa inactivated significantly slower than ICa at all time points (Fig. 8D, E

Table 3 Effect of RIM2C on the voltage-dependent inactivation properties of Cav1.3 channel complexes. Constructs

V0.5,inact

kinact

n

Cav1.3 + β3 Cav1.3 + β3+ RIM2C Cav1.3 + β3+ RIM2C Cav1.3 + β2a Cav1.3 + β2a + RIM2C Cav1.3 + C3S/C4Sβ2a Cav1.3 + C3S/C4Sβ2a + RIM2C Cav2.1 + β3 Cav2.1 + β3 + RIM2C

−22.4 ± 0.8 −11.4 ± 1.9** −9.9 ± 0.8*** −13.6 ± 1 −19.9 ± 0.8** −25.4 ± 1 −14.3 ± 1.1*** −15.8 ± 2.5 5.6 ± 1.0**

−5.9 ± 0.2 −13.8 ± 0.7*** −12.1 ± 0.95*** −9.9 ± 0.9 −9.9 ± 1.0 −5.8 ± 0.4 −10.5 ± 0.7*** −4.6 ± 0.2 −2.5 ± 0.3**

13 6 8 5 6 7 7 7 5

The indicated α1 and β-subunits were co-expressed with α2δ-1 in tsA-201 cells, and the biophysical properties were determined using 15 mM Ca2+ as a charge carrier. V0.5,inact, and kinact were obtained by fitting the data as described in Experimental methods. Data are given as means± SEM. Statistically significant differences between data obtained in the absence or presence of RIM2c for a given subunit composition (calculated by one-way ANOVA, followed by Bonferroni test) are indicated: *p b 0.05; **p b 0.01; ***p b 0.001.

and Table 2) which indicates CDI. RIM2C slowed IBa inactivation during 5-s depolarizations. This effect was statistically significant and maximal at 1 s (Fig. 8E). To further quantify the effects of RIM2C on the Cav1.3 inactivation time course we fitted the mean of the inactivation curves for ICa and IBa to a bi-exponential current decay (Fig. 8F, G). For control Cav1.3, the slower inactivation time course of IBa as compared to ICa was mostly due to a 3-fold higher slow inactivation time constant (τslow), a 1.6-fold higher contribution of τslow to total IBa inactivation and a more pronounced non-inactivating component (%slow) (Table 4). Co-expression of RIM2C slowed the inactivation time course of IBa, mainly by increasing the contribution of τslow (Table 4) without a further increase of the non-inactivating component. In contrast, the main effect of RIM2C on ICa was to triple the contribution of the non-inactivating component (to values very similar to IBa), and to also increase τslow and its fractional contribution to inactivation (Fig. 8F, G Table 4). These results indicate that RIM2C not only slows VDI but also CDI of Cav1.3 currents in particular by stabilization of a slowly inactivating current component of ICa but not IBa. In contrast to slow inactivation, CDI occurring during the first 50 ms was not affected by RIM1C. For this time period we could measure the typical U-shaped voltage-dependence of CDI (Fig. 8F, inset) with an f-value (maximal difference in residual current of ICa vs. IBa, see Experimental methods) of 0.33 ± 0.03. This was not affected by co-expression of RIM2C (f = 0.34 ± 0.02, not illustrated). It is known that the modulation of Ca2+ channel inactivation can vary with the type of β subunit forming part of the channel complex (Dolphin, 2003). In particular, the N-terminally palmitoylated β2a subunits slows the inactivation of different types of Ca2+ channels, including Cav1.3 (Koschak et al., 2001) and this effect is strictly dependent on palmitoylation. We tested if the RIM2C effect on Cav1.3 channel inactivation is additive to those of β2a. As shown in Fig. 9 and Tables 1–3 this was not the case. Instead, the modulatory action of RIM2C on the ICa inactivation time course and on the voltagedependence of steady-state inactivation was essentially absent when Cav1.3 channel complexes contained β2a. To determine if this was due to differences in the protein component or due to palmitoylation we co-expressed the channel with the palmitoylation-deficient β2a subunit (C3S/C4Sβ2a). As shown in Fig. 2, C3S/C4Sβ2a

Table 4 Effect of RIM2C on the time course of Cav1.3 IBa and ICa inactivation. Channel*

Current

%fast

τfast (ms)

%slow

τslow (ms)

%non-inact

n

Cav1.3 + β3

ICa IBa ICa IBa

66.8 35.8 21.2 9.9

203.9 310.4 399.7 351.6

27.9 43.6 60.7 72.2

830 2616.8 2577.3 3117.7

5.23 20.6 18.1 17.9

27 5 7 7

Cav1.3 + β3+ RIM2C

The indicated α1 and β-subunits were co-expressed with α2δ-1 in tsA-201 cells, and the biophysical properties were determined using 15 mM Ca2+ or Ba2+ as charge carrier. Normalized inactivation curves (5-s depolarization to Vmax) from the indicated number of experiments were pooled and fitted to a bi-exponential decay. The fraction of current (%fast, %slow) decaying with fast and slow time constants (τfast, τslow) and the fraction of a non-inactivating current component (%non-inact) are given.

254


Fig. 9. Effects of RIM2C on the inactivation properties of Cav1.3 channel complexes containing β2a. Cav1.3 α1 subunits were co-expressed with β2a (above) or the palmitoylationdeficient C3S/C4Sβ2a mutant (below) and α2δ1 in tsA-201 cells in the absence (black) or presence of RIM2C (grey). (A) Inactivation time course of Cav1.3 ICa during 5-s depolarizing pulses from a holding potential of −80 mV to Vmax is depicted. (B) Percentage of inactivation of ICa was analyzed at the indicated pre-specified time periods. Statistically significant differences (one-way ANOVA followed by Bonferroni post-test) are indicated: with vs. without RIM: ***p b 0.001; difference vs. control: +++p b 0.0001. Error bars reflect SEM for the indicated number of experiments. (C) Voltage-dependence of ICa inactivation curves were obtained by test potentials to Vmax after 5-s pulses to the indicated voltages and fitted to the Boltzmann relationship. Inactivation parameters are given in Table 3.

is still able to bind both RIM and the channel's I–II-linker in tsA-201 cells. C3S/C4Sβ2a still supported robust Cav1.3 currents but no longer slowed its inactivation and thus stabilized ICa properties essentially indistinguishable from β3. Co-expression of C3S/C4Sβ2a fully restored the modulatory properties of RIM2C indicating that palmitoylation itself diminished the modulation by RIM2C (Fig. 9A–C). Discussion Here we show that RIM2 transcripts are expressed in cochlear IHCs, where RIM protein is targeted to presynaptic release sites and colocalizes with Cav1.3 channels. Expression of Cav1.3 channel complexes together with RIM constructs in tsA-201cells revealed that RIM slows both CDI and VDI through binding to β-subunits. By slowing inactivation over a large voltage range and by substantially increasing the noninactivating component of ICa, RIM-associated channel complexes should carry larger Cav1.3 window currents during prolonged depolarization. In this respect, RIM mimicked the modulatory properties of the β2a-subunit splice variant on channel inactivation. Another novel observation was that the cytoplasmic I–II linker of Cav1 channels is sufficient to specifically target β-subunits to the plasma membrane when expressed in tsA-201 cells. This property then allowed us to demonstrate RIM binding to β-subunits under our recording conditions. Cav1.3 VGCCs fulfill different functional requirements at different developmental stages. In mature IHCs they trigger sound-evoked exocytosis of closely co-localized readily releasable vesicles in response to rapid and graded receptor potentials. Before the onset of hearing, Cav1.3 channels enable the firing of spontaneous Ca2+ action potentials which drive afferent synaptic transmission and also seem to contribute to maturation of the IHCs and to pruning of synaptic connections of the auditory pathway (Moody and Bosma, 2005). Both functions rely on the channels negative activation threshold, their rapid activation and deactivation (within less than 1 ms, Koschak et al., 2001; Platzer et al., 2000) and their slow and incomplete inactivation during prolonged depolarization. Although negative activation thresholds and fast

activation are typical for all Cav1.3 channels, the slow inactivation of ICa is a special feature exclusively of Cav1.3 channels in IHCs. Cav1.3 Ca2+ currents in heterologous expression systems as well as in the heart and brain inactivate rapidly due to the presence of pronounced CDI (Mangoni et al., 2003; Song et al., 2003). CDI of Cav1.3 channels can be almost completely inhibited in heterologous expression systems by coexpression of the Ca2+ binding proteins CaBP1 and CaBP4 which are present in IHCs (Cui et al., 2007; Yang et al., 2006). This provides one possible molecular explanation for the moderate CDI observed in hair cells (Platzer et al., 2000; Schnee and Ricci, 2003). Although CDI is much weaker in hair cells than in Cav1.3 currents recorded from sinoatrial node cells (Mangoni et al., 2003) or from heterologously expressed channels (Koschak et al., 2001) it is still present with its size being dependent on intracellular Ca2+ buffering and recording temperature (Grant and Fuchs, 2008; Martini et al., 2004; Song et al., 2003). Independent of recording conditions, weak CDI in combination with slow VDI prevents Cav1.3 currents in hair cells from complete inactivation even after prolonged depolarization to positive voltages. This ensures continuous spontaneous activity during early development but also provides an adequate dynamic range for sound-evoked gating of the limited number of Ca2+ channels (∼180) thought to operate at each ribbon synapse (Zampini et al., 2010; Brandt et al., 2005). Here we demonstrate that the association with RIM2 protein represents another possible molecular mechanism able to inhibit CDI and VDI of Cav1.3 channels in IHCs. By slowing the inactivation, coexpression of RIM increased the ICa component persisting even after 5-s depolarizations to Vmax. Such slowly inactivating current components are characteristic for native IHC Cav1.3 currents, although their amplitudes reported in the literature are quite variable even under conditions of strong intracellular Ca2+ buffering (Koschak et al., 2001; Lee et al., 2007; Marcotti et al., 2003; Tarabova et al., 2007). The pronounced effect of RIM on inactivation is also evident at early inactivation time points. During the first 0.25 s, RIM reduced inactivation to 20–30%, values close to those reported for native IHC currents (Kong et al., 2008; Neef et al., 2009). RIM2 also slowed VDI with IBa


kinetics falling within the inactivation time courses reported by others in mouse (18% after 0.5 s; Cui et al., 2007) and chicken hair cells (Lee et al., 2007), although much slower IBa inactivation has also been reported in IHCs (9% of peak IBa during 5 s; Koschak et al., 2001). Our data do not rule out the possibility that other presynaptic proteins known to interact with the Cav1.3 channel complex (Song et al., 2003) or even other mechanisms (Shen et al., 2006) contribute to or fully account for the slow inactivation kinetics of IHC Cav1.3 Ca2+ currents. However, we provide evidence that RIM2 also represents an attractive candidate protein. Important differences are found in the synaptic and channel function between immature and mature IHCs. CDI and peak current densities are significantly more pronounced in pre-hearing as compared to posthearing when recorded at 37 °C (Grant and Fuchs, 2008). In addition, in immature IHC exocytosis depends non-linearly (power of 3) on ICa, whereas it becomes a linear function in mature cells (Johnson et al., 2005). This developmental change appears not to be associated with differences in ribbon morphology (Johnson and Marcotti, 2008) but rather with a switch in the Ca2+ sensor(s) for vesicle fusion. Recently, synaptotagminIV expressed after the onset of hearing has been shown to linearize the synaptic transfer function in mature IHCs (Johnson et al., 2010). These differences may be important for the specific signaling functions of Cav1.3 channels during different developmental stages with immature mammalian IHCs generating spontaneous, regenerative Ca2+ action potentials (Kros et al., 1998; Marcotti et al., 2003) important for normal IHC development. RIM2 appears to be one of the proteins involved in changes of biochemical ribbon remodeling during development because we repeatedly failed to detect RIM transcripts in mature IHCs. Based on this observation it is tempting to speculate that also different molecular mechanisms may be involved in the stabilization of slow ICa inactivation in developing and mature IHCs. This could also hold true for CaBPs. Since CaBP4-deficient mice exhibit normal hearing, CaBP1 appears to be the main CaBP mediating CDI inhibition. However, this appears to be limited to mature IHCs because only there CaBP1 co-localizes with Cav1.3 channels in the presynaptic clusters (Cui et al., 2007). The small amount of IHC tissue makes direct biochemical proof of the physical interaction between RIM2 and the Cav1.3 associated β subunits by biochemical analysis technically highly challenging. We therefore employed immunohistochemistry to provide strong indirect evidence for their association within the same presynaptic domains of IHCs. Data in the present study show that in the basolateral pole of IHCs ∼50% of ribbons are co-localized with Cav1.3 and an even higher amount of ribbons (∼60%) with RIM2. A recent study in rodent photoreceptors at a developmental stage comparable to our study revealed that Cav1/RIM2 scaffolds aggregate at the active zone only gradually with RIBEYEcontaining assemblies (Regus-Leidig et al., 2009; tom Dieck et al., 2005) This fits very nicely to the present observation of only partial overlap of ribbons, Ca2+ channels and RIM2 in immature IHCs. We also found that a palmitoylated splice variant of Ca2+ channel β2-subunits, β2a, prevents the modulatory effect of RIM. This was evident from the fact that the slowing of inactivation kinetics induced by β2a was not additive to the effect of RIM. These data imply that the modulatory role of RIM is absent in Cav1.3 channels associated with β2a subunits. Indeed, β2 subunits are the prominent β-subunit isoform in IHC ribbon synapses (Kuhn et al., 2009; Neef et al., 2009). However, there is indirect experimental evidence that the palmitoylated splice variant does not participate significantly in the stabilization of current kinetics in IHCs because voltage-dependent inactivation kinetics are not changed after genetic ablation of β2 (Neef et al., 2009). This also indicates that slow inactivation of native IHC Cav1.3 channels must be stabilized by another mechanism, including RIM as a candidate in immature IHCs. The prevention of the RIM effect was not seen for the non-palmitoylated mutant. From the β-subunit three dimensional structure it has been predicted that anchoring of the acylated N-terminus of β2a in the lipid bilayer reduces its flexibility (Opatowsky et al., 2004) and through its

255

interaction with the I–II linker thereby also restricts the conformational flexibility of the channels gating machinery (Findeisen and Minor, 2009; Vitko et al., 2008). We therefore speculate that RIM binding to non-palmitoylated β-subunits also decreases their conformational flexibility to an extent similar to palmitoylation. Although RIM has always been regarded to be only located presynaptically, a short RIM3γ isoform has recently been detected postsynaptically throughout the rat central nervous system (Liang et al., 2007). RIM3γ also contains the highly conserved C2B domain (Wang and Sudhof, 2003) required for channel modulation. Therefore RIM proteins may also modulate Cav1.3 channels in neurons, where Cav1.3 channels are located postsynaptically.

Experimental methods Cloning of cDNA constructs Numbering of amino acids for human RIM1α, RIM2β and Cav1.3α1 constructs refers to Genbank accession numbers NM_014989, NM_014677 and EU363339, respectively. Rat RIM2α was kindly provided by Susanne Schoch (Wang et al., 2000). The integrity of all cDNA constructs was confirmed by sequencing (MWG Biotech, Martinsried, Germany). HARIM1C was generated by PCR amplification of the nucleotides corresponding to aa 1503 to 1692 of RIM1α, thereby adding an Nterminal HA-tag and artificial 5’-NheI and 3’-EcoRI sites. The resulting PCR fragment was then cloned into the mammalian expression vector pCINeo (Promega, Mannheim, Germany). To generate the mammalian expression vector pmRFP-C1, first EYFP was removed from the vector pEYFP-C1 (Clontech, Saint-Germain-en-Laye, France) by cutting with AgeI and BglII and ligation with monomeric red fluorescent protein (mRFP) which was PCR amplified from vector pCX-mRFP1 (Long et al., 2005, kindly provided by Anna-Katerina Hadjantonakis with permission of Roger Y. Tsien). For generation of mRFPRIM2β full-length RIM2β was cloned from the EST clone EHS1001-7376976 (Open Biosystems) and fused at its N-terminus with mRFP by cloning into the HindIII and SalI restriction sites of pmRFP-C1. The flag-epitope tagged cytoplasmic I–II linker of the Cav1.3α1 subunit (I–IIflag) was cloned by PCR amplifying the nucleotide sequence corresponding to aa 407 to 523 of the human Cav1.3α1 subunit thereby adding a C-terminal flag tag and subsequent ligation into the NheI and NotI restriction sites of pCINeo. The W441A mutation was introduced by SOE-PCR with a silent BglII restriction site at nucleotide position 84 of the I–II linker sequence. For Cterminal V5 tagging of the β subunits the whole open reading frames were PCR amplified, thereby introducing artificial restriction sites (given in parentheses). The resulting PCR products were then cloned in frame into the appropriate cloning sites of the multiple cloning site (MCS) of the EF-1α promoter of the mammalian expression vector pBudCE4.1 (Invitrogen). The β subunits employed were as follows: rabbit β1a (Genbank accession number: M25514): 5’-NotI; 3’-BglII; rat βa: M80545 (5’-NotI, 3’-BglII); rat β3: NM_012828 (5’-NotI, 3’-BglII); rat β4: XM_215742 (5’-NotI, 3’-XhoI). To generate the palmitoylation deficient mutant C3S/C4Sβ2a the cystein residues 3 and 4 of β2a were exchanged by serines with PCR. For immunocytochemical experiments of β1a and β3, the C-terminally V5 tagged constructs were expressed from the expression vector pβA-PL (Obermair et al., 2004, Schlick et al., 2010). β1aV5 and β3V5 were PCR amplified from the corresponding pBudCE4.1 constructs, thereby inserting appropriate restriction sites for insertion into pbA-PL. To express each of the β subunits together with the untagged linker I–II simultaneously from the same plasmid the β subunits were cloned as above into the MCS of the EF-1α promoter of pBudCE4.1/I–II. The latter was generated by cloning the PCR amplified I–II linker into the HindIII and SalI restriction sites of the MCS of the CMV promoter of pBudCE4.1.

256


Constructs for electrophysiological recordings The C-terminal peptides of RIM1α (aa 1503–1692) or RIM2α (aa 1168–1503) were fused at their N-termini with GFP to generate the expression constructs RIM1C and RIM2C. The nucleotides of the corresponding aa were PCR amplified and then cloned into the HindIII and EcoRI sites of the vector pGFP+ (Grabner et al., 1998). Yeast expression constructs To generate the yeast expression vector pGBTK7-GAL the GAL4-DNA binding domain of pGBTK7 (Clontech, Saint-Germain-en-Laye, France) was removed: pGBTK7 was cut with DraIII and EcoRI, the vector backbone and the 1571 bp fragment were isolated and the latter used as template for PCR with: fwd primer: 5'-CCCACTACGTGAACCATCACC-3', rev primer: 5'-GCGAATTCAGTTGATTGTATGCTTGGTATAGC-3'. The resulting PCR product was cut with EcoRI and DraIII and then ligated with the vector back-bone of pGBTK7. C3S/C4Sβ2a was then PCR amplified with: fwd primer: 5'-CGGGATCCATGCAGTCCTCCGGGCTG-3', rev primer: 5'-TTCTGCAGTCATTGGCGGATGTATACATC-3' and cloned into the BamHI and PstI sites of pGBTK7-GAL. To generate an N-terminal fusion construct with human Sos, the linker I–II was PCR amplified with: fwd primer: 5'CATGCCATGGCCGGAGAATTCTCAAAGGAAAGAG-3', rev primer: 5'CGGTCGACTCAGACAGACTTCACGGCGG-3' and then cloned into the NcoI and SalI sites of the yeast expression vector pSos (Stratagene). Cell culture and transfection HEK293 (tsA-201) cells were cultured and transfected as described previously (Koschak et al., 2001). For electrophysiological recordings Cav1.3α1 or CaV2.1 α1 (Watschinger et al., 2008) was always coexpressed together with β3 (or β2a) and α2δ1-subunits as described (Koschak et al., 2001) and either GFP-RIM2C or GFP alone to identify transfected cells. For immunocytochemistry tsA-201 cells were transfected as indicated in the results using 2–3 µg of plasmid DNA for each construct adjusted to a total amount of 5 µg with pUC cDNA. Immunocytochemical experiments Transiently transfected tsA-201 cells were seeded onto poly-Llysine coated microscopy cover-slips 24 hrs after transfection and after 48 hrs fixed for 15 min with 4% (w/v) paraformaldehyde at room temperature (RT) and then washed thoroughly with PBS. Cells were blocked for 30 min at RT with 5% (w/v) normal goat serum (Gibco) in 1% (v/v) Triton/PBS for cell permeabilization. Primary antibodies were incubated over night at 4 °C in washing buffer containing 1% Triton and 1% (w/v) BSA (immunoglobulin free, Sigma) in PBS. After washing the cells (3 times briefly and 2 times for 15 min) they were incubated with the secondary antibody for 1 h at RT in the dark. Washing was repeated, the cover-slips mounted with Vectashield mounting medium (Vector Laboratories) and sealed with nail polisher on microscope slides. The following antibodies were employed: rat-monoclonal anti-HA clone 3F10, working dilution 1:1000 (Roche, No.1867423); mouse-monoclonal anti-V5, working dilution 1:500 (Invitrogen, No. R960-25); rabbit-polyclonal antiFLAG, working dilution 1:500 (Sigma, No. F7425); Alexa-488conjugated goat-anti-rat and goat-anti-rabbit and Alexa-488- and Alexa-594-conjugated goat-anti-mouse antibodies, working dilution 1:4000 (Molecular Probes, Invitrogen). Images were captured with an Axiophot microscope (Carl Zeiss, Jena, Germany) using a cooled CCD camera and Meta View image processing software (Universal Imaging Corporation, West Chester, PA). Images were manually adjusted with Adobe Photoshop 7.0. Animals, tissue preparation, immunohistochemistry Care and use of the animals as well as the experimental protocol were done in accordance with the ethical guidelines approved by the

University of Tübingen, the Tierschutzgesetz (Germany) and the Austrian Bundesministerium für Wissenschaft und Forschung (Kommission für Tierversuchsangelegenheiten). For organ of Corti and IHC preparation NMRI mice were used, whole-brain tissue was obtained from C57B/6 N wild-type mice. Immature mouse (P8) cochleae were isolated, fixed, cryosectioned, and stained as described (Knipper et al., 2000 Knirsch et al., 2007). Shortly cochleae were quickly fixed by injection with 2% paraformaldehyde, 125 mM sucrose in 100 mM phosphate buffered saline (PBS), pH 7.4. After short injection of 25% sucrose in PBS, pH 7.4, cochleae were embedded in O.C.T. compound (Miles Laboratories, Elkhart, Ind., USA). Tissues were then cryosectioned at 10 µm thickness mounted on SuperFrost*/plus microscope slides, dried for 1 hour and stored at –20 °C before use. Mouse cochlear sections were thawed and permeabilized with 0.5% Triton X100 for 10 min at room temperature, pre-blocked with 4% normal goat serum (NGS) in PBS, and incubated overnight at 4 °C with primary antibodies. IHCs were stained using rabbit polyclonal anti-RIM2 (Synaptic Systems; Goettingen; Germany; 140103; 1:150), rabbit polyclonal anti-Cav1.3 (Alomone Labs, APC-014; 1:50) and mouse monoclonal anti-CtPB2/RIBEYE (BD Transduction Laboratories, CA, USA, 612044; 1:50) antibodies. Direct double staining experiments of Cav1.3 α1-subunits and RIM2 protein were impossible due to the lack of suitable RIM antibodies from a second animal species. A mouse monoclonal anti-RIM2 antibody (Abcam) did not yield specific staining in the cochlea. Therefore neighboring sections of the same cochleae were used and IHCs from the same cochlear turn compared for either immunostaining of CtBP2/RIBEYE and Cav1.3 or CtBP2/ RIBEYE and RIM2. Antibodies were diluted in PBS containing 0.1% Triton X-100/1% NGS and were detected with Cy3-conjugated (Jackson ImmunoResearch Laboratories, USA) or Alexa Fluor 488-conjugated antibodies (Molecular Probes, USA). Sections were embedded with Vectashield mounting medium containing DAPI (Vector Laboratories, USA). Sections were viewed using an Olympus AX70 microscope equipped with epifluorescence illumination and a motorized z-axis. Images were acquired using a CCD camera and the imaging software Cell F (OSIS GmbH, Münster, Germany). Typically z-stacks consisted of 30 layers with a z-increment of 0.276 µm, for each layer one image per fluorochrome was acquired. Z-stacks were 3-dimensionally deconvoluted using Cell F's RIDE module with the Nearest Neighbour algorithm (OSIS GmbH, Münster, Germany). Figures are illustrated as a composite image, which represents the maximum intensity projection over all layers of the z-stack. Recently in order to better define the actual resolution of our system, we measured the pixel intensity of beads with known diameter (175 nm—green light emission; PS-Speck Microscope Point Source Kit, Molecular Probes) as recently described (for details see Zampini et al., 2010). Briefly, the diameter of ribbons we measured from immature IHCs was on average 376 nm (using Alexa 488 as secondary antibody), which is in the range of that previously reported by transmission electron microscopy (about 300 nm; Sobkowicz et al., 1982) and larger than the diffraction pattern of the beads. As a conservative approach we used a minimum overlap between ribbon and Ca2+ channel immunospots of 46% as a criterion for co-localization. This would probably underestimate the co-localized clusters, since the spread of the diffraction is constant (does not change relative to the object's size). Yeast two hybrid screen The Cytotrap (Stratagene) yeast two hybrid system was used and carried out according to the manufacturer's instructions with the following modifications. In brief: the yeast strain cdc25H was cotransformed with C3S/C4Sβ2a in pGBTK7-GAL together with human Soslinker-I–II and a human fetal brain cDNA library in pMyr (Stratagene). Transformants were selected on appropriate glucose yeast drop-out


plates at the permissive temperature (25 °C). Growing colonies were then replica plated onto galactose drop-out plates to induce library expression and incubated at the none-permissive temperature (37 °C) to screen for interacting proteins. Putative positives were rescued on glucose plates and grown at the permissive temperature, re-plated onto glucose and galactose plates and incubated at 37 °C. pMyr plasmids were only isolated from colonies that grew again on galactose plates at the none-permissive temperature. Interaction of bait and prey was then verified in a second round of the interaction assay. Proper controls excluded the possibility that the prey complex or the prey library themselves showed a transactivation activity. RT-PCR Total RNA was isolated from whole-brain tissue of C57B/6 N wildtype mice with the RNAqueusR-4PCR Kit (Ambion, Foster City, CA). After DNAseI treatment the integrity of the RNA was checked by the quality of the 28 S and 18 S rRNA bands on a denaturing agarose gel. organ of Corti and IHC mRNA was prepared from P4 or P20 mice as described (Michna et al., 2003). Mouse embryonic (E11) total RNA was purchased from Clontech (Saint-Germain-en-Laye, France). One µg of total RNA was used for first strand cDNA synthesis by reverse transcriptase (RevertAid™ H Minus First Strand cDNA Synthesis Kit, MBI; Fermentas, Hanover, USA) and random hexamer primers. To detect RIM2α and RIM2β in mouse cochlea a standard PCR approach was sufficient and to investigate RIM1α, RIM2α, RIM2β and RIM3γ expression in IHCs and Cav1.3 (α1b) in embryonic cDNA a nested PCR approach was chosen. The following primers were used: RIM2α (Genbank accession number: NM_053271): fwd primer: 5'-TGCAGCAACCTGATCAAAAG-3', rev primer: 5'-GTAGGGTTGTGGCTTTACGG-3' (572 bp), for the nested PCR in addition the inner primers: fwd primer: 5'AGCCCCTCAGGAGAAGAAAG3' and rev primer: 5'-TAGCGTGCCTGGTATTCCTC-3' (477 bp) were used; RIM2β (theoretical sequence was assembled from NM_053271 and NW_001030570.1): fwd primer: 5'-ATGCAATTTGAGACGTTGCG-3', rev primer: 5'-GTAGGGTTGTGGCTTTACGG-3' (471 bp), for nested PCR in addition the inner primers: fwd primer: 5'-TGCAATTCTGTTTTATCTCATTTCC-3', rev primer: 5'-TAGCGTGCCTGGTATTCCTC-3' (395 bp) were used; RIM1α (NM_053270): outer primers: fwd primer: 5'-CCGACGTGTGGAATCTGTC-3’' rev primer: 5'-CTTGGTGGCTCACTTCTTGAC-3' (464 bp); inner primers: fwd primer: 5'-GGCCATCTCTGCTCCTATTG-3', rev primer: 5'-CTTCTGTTCAGGACCCAAGG-3' (387 bp); RIM3γ (NM_182929): outer primers: fwd primer: 5'-AGGGCCTCCAGGAATGTAG-3', rev primer: 5'-TCTCCAGCAGGTAAGCCTTG-3' (569 bp); inner primers: fwd primer: 5'-TAGCTATCGTCGGCTTGACC-3’, rev primer: 5'TTCAATCACTTCCACCTCCAG-3' (368 bp); Cav1.3(α1b) (NM_028981) : outer primers: fwd primer: 5'-ATGAACCTTCCGACATTTTCTAG-3', rev primer: 5'-ACACAATTGGCAAAAATAGCC-3' (488 bp); inner primers: fwd primer: 5'-CAAGAGACTGATGCCCGATATAAAG-3', rev primer: 5'CTCTTCGGATGGGGTTATTG-3' (364 bp); Cav1.3(α1a) (NM_001083616): fwd primer: 5'-GCAACATCAACGGCAGCACC-3', rev primer: 5'-ACACAATTGGCAAAAATAGCC-3' (393 bp). As a positive control mouse GAPDH was used (NM_008084): fwd primer: 5'-ACTCCACTCACGGCAAATTC-3', rev primer: 5'-CACATTGGGGGTAGGAACAC-3' (572 bp); as negative control instead of cDNA nuclease free water was added to the PCR reactions. The resulting PCR products were sub-cloned into pGEM-T Easy vectors (Promega, Mannheim, Germany) and their identity verified by sequencing (MWG Biotech AG, Martinsried, Germany). Electrophysiological recordings Inward Ba2+ (IBa) and Ca2+ (ICa) currents through Cav1.3 channels were measured using whole-cell patch-clamp techniques in transiently transfected tsA-201 cells as described (Koschak et al., 2001; Singh et al., 2008). All electrophysiological recordings were carried out at room temperature using an Axopatch 200B amplifier (Axon Instruments, Foster City, CA) linked to a personal computer equipped

257

with pClamp version 9.0. Currents were recorded at sampling rates of 5 - 25 kHz and low-pass filtered at 2 - 5 kHz with a Digidata 1322A analog-to-digital board (Axon Instruments). Pipettes (Borosilicate glass, 64-0792, Harvard apparatus, USA) were pulled on a P-97 microelectrode puller (Sutter Instruments) and fire polished with an MF-830 microforge (Narishige, Japan), showing typical resistances of 2–3 MΩ when filled with internal solution. Series resistance compensation of 60 % was used. The voltage error due to uncompensated series resistance was always below 5 mV. Recording solutions for whole-cell measurements were as follows (in mM): Internal solution: 135 CsCl, 10 Cs-EGTA, 10 HEPES, and 1 MgCl2, Na2-ATP adjusted to pH 7.4 with CsOH; Bath solution: 15 BaCl2 or CaCl2, 10 HEPES, 150 Choline Cl, 1MgCl2, adjusted to pH 7.4 with CsOH. The holding potential (HP) was −80 mV. All voltages were corrected for a liquid junction potential of the respective recording solutions (– 8.5 mV). Leak currents were subtracted offline whenever online P/4 leak subtraction was not included in the protocol. Transfected cells were visualized by co-expression of green fluorescent protein (GFP) or GFP-labeled RIM constructs. Protocols used and data analysis The voltage dependence of activation was determined from current–voltage (I–V) curves obtained by step depolarization from the HP to various test potentials. I–V curves were fitted according to the following equation: I = Gmax(V − Vrev)/{1 + exp[(V − V0.5act)/k]} where Vrev is the extrapolated reversal potential of IBa or ICa respectively, V is the membrane potential, I is the peak current, Gmax is the maximum conductance of the cell, V0.5 is the voltage for halfmaximal activation, and k is the slope factor. Percent inactivation was measured at various time points holding the cell at Vmax for 5 s. In steady-state inactivation experiments, a control test pulse (20 ms to Vmax) was followed by a 5 s conditioning step (−70 to + 50 in 20 mV increments), and a subsequent 20 ms test pulse to Vmax. Curves were fitted with the Boltzmann relationship: I = ISS + (1 - ISS) / (1 + exp (V-V0.5,inact/kinact) where I is the peak current amplitude, ISS is the non-inactivating fraction, V is the membrane potential, V0.5inact is the half-inactivation potential, and kinact is the slope factor. To quantify CDI, f-values were calculated and defined as the maximal difference between r50 values of IBa versus ICa. The r50 value describes the fraction of peak current remaining at the end of a 50 ms depolarization and was plotted as a function of test potential. The time course of current inactivation was fitted to the following biexponential functions yielding time constants for a fast (τfast) and a slow (τslow) component. I(t) = A fast [exp(t/τfast)] + A slow [exp(t/τslow)] + C where I(t) is the current at time t after the depolarization, A is the steady state current amplitude with the respective time constant of activation, t, and C the remaining steady state current. Statistics Data were analyzed using Clampfit 9.0 (Axon Instruments) and Origin 5.0 (Microcal Software, Northampton, MA). All data are presented as mean ± SEM (standard error of the mean) for the indicated number of experiments. Statistical significance was determined using one-way ANOVA followed by Bonferroni post-test. Statistical significance was set at p b 0.05. Unpaired Student's t-test or Mann–Whitney test was used for comparisons of two groups for parametric and nonparametric data, respectively. Acknowledgments We thank Jennifer Müller and Sabrina Hassler for expert technical assistance, Susanne Schoch for the RIM1α expression construct and Martina J. Sinnegger-Brauns and Bernhard E. Flucher for helpful discussions. This work was supported by the Marie Curie Research Training

258


Network CavNET (MRTN-CT-2006-035367), the Austrian Science Funds (P20760) and the University of Innsbruck (Austria). References Brandt, A., Khimich, D., Moser, T., 2005. Few Cav1.3 channels regulate the exocytosis of a synaptic vesicle at the hair cell ribbon synapse. J. Neurosci. 25, 11577–11585. Calin-Jageman, I., Lee, A., 2008. Cav1 L-type Ca2+ channel signaling complexes in neurons. J. Neurochem. 105, 573–583. Catterall, W.A., Perez-Reyes, E., Snutch, T.P., Striessnig, J., 2005. International Union of Pharmacology. XLVIII. Nomenclature and structure–function relationships of voltage-gated calcium channels. Pharmacol. Rev. 57, 411–425. Colecraft, H.M., Alseikhan, B., Takahashi, S.X., Chaudhuri, D., Mittman, S., Yegnasubramanian, V., Alvania, R.S., Johns, D.C., Marban, E., Yue, D.T., 2002. Novel functional properties of Ca2+ channel beta subunits revealed by their expression in adult rat heart cells. J. Physiol. 541, 435–452. Cui, G., Meyer, A.C., Calin-Jageman, I., Neef, J., Haeseleer, F., Moser, T., Lee, A., 2007. Ca2+binding proteins tune Ca2+-feedback to Cav1.3 channels in mouse auditory hair cells. J. Physiol. 585, 791–803. Dai, S., Hall, D.D., Hell, J.W., 2009. Supramolecular assemblies and localized regulation of voltage-gated ion channels. Physiol. Rev. 89, 411–452. Davies, A., Hendrich, J., Van Minh, A.T., Wratten, J., Douglas, L., Dolphin, A.C., 2007. Functional biology of the α2 subunits of voltage-gated calcium channels. Trends Pharmacol. Sci. 28, 220–228. Dolphin, A.C., 2003. Beta subunits of voltage-gated calcium channels. J. Bioenerg. Biomembr. 35, 599–620. Findeisen, F., Minor Jr., D.L., 2009. Disruption of the IS6-AID linker affects voltage-gated calcium channel inactivation and facilitation. J. Gen. Physiol. 133, 327–343. Gomez-Ospina, N., Tsuruta, F., Barreto-Chang, O., Hu, L., Dolmetsch, R., 2006. The C terminus of the L-type voltage-gated calcium channel CaV1.2 encodes a transcription factor. Cell 127, 591–606. Grabner, M., Dirksen, R.T., Beam, K.G., 1998. Tagging with green fluorescent protein reveals a distinct subcellular distribution of L-type and non-L-type Ca2+ channels expressed in dysgenic myotubes. Proc. Natl. Acad. Sci. U. S. A. 95, 1903–1908. Grant, L., Fuchs, P., 2008. Calcium- and calmodulin-dependent inactivation of calcium channels in inner hair cells of the rat cochlea. J. Neurophysiol. 99, 2183–2193. Hibino, H., Pironkova, R., Onwumere, O., Rousset, M., Charnet, P., Hudspeth, A.J., Lesage, F., 2003. Direct interaction with a nuclear protein and regulation of gene silencing by a variant of the Ca2+-channel β4 subunit. Proc. Natl. Acad. Sci. U. S. A. 100, 307–312. Johnson, S.L., Franz, C., Kuhn, S., Furness, D.N., Ruttiger, L., Munkner, S., Rivolta, M.N., Seward, E.P., Herschman, H.R., Engel, J., Knipper, M., Marcotti, W., 2010. Synaptotagmin IV determines the linear Ca2+ dependence of vesicle fusion at auditory ribbon synapses. Nat. Neurosci. 13, 45–52. Johnson, S.L., Marcotti, W., 2008. Biophysical properties of Cav1.3 calcium channels in gerbil inner hair cells. J. Physiol. 586, 1029–1042. Johnson, S.L., Marcotti, W., Kros, C.J., 2005. Increase in efficiency and reduction in Ca2+ dependence of exocytosis during development of mouse inner hair cells. J. Physiol. 563, 177–191. Kiyonaka, S., Wakamori, M., Miki, T., Uriu, Y., Nonaka, M., Bito, H., Beedle, A.M., Mori, E., Hara, Y., De Waard, M., Kanagawa, M., Itakura, M., Takahashi, M., Campbell, K.P., Mori, Y., 2007. RIM1 confers sustained activity and neurotransmitter vesicle anchoring to presynaptic Ca2+ channels. Nat. Neurosci. 10, 691–701. Klugbauer, N., Welling, A., Specht, V., Seisenberger, C., Hofmann, F., 2002. L-type Ca2+ channels of the embryonic mouse heart. Eur. J. Pharmacol. 447, 279–284. Knipper, M., Zinn, C., Maier, H., Praetorius, M., Rohbock, K., Kopschall, I., Zimmermann, U., 2000. Thyroid hormone deficiency before the onset of hearing causes irreversible damage to peripheral and central auditory systems. J. Neurophysiol. 83, 3101–3112. Knirsch, M., Brandt, N., Braig, C., Kuhn, S., Hirt, B., Munkner, S., Knipper, M., Engel, J., 2007. Persistence of Cav1.3 Ca2+ channels in mature outer hair cells supports outer hair cell afferent signaling. J. Neurosci. 27, 6442–6451. Kong, J.H., Adelman, J.P., Fuchs, P.A., 2008. Expression of the SK2 calcium-activated potassium channel is required for cholinergic function in mouse cochlear hair cells. J. Physiol. 586, 5471–5485. Koschak, A., Reimer, D., Huber, I., Grabner, M., Glossmann, H., Engel, J., Striessnig, J., 2001. α1D (Cav1.3) subunits can form L-type Ca2+ channels activating at negative voltages. J. Biol. Chem. 276, 22100–22106. Kros, C.J., Ruppersberg, J.P., Rusch, A., 1998. Expression of a potassium current in inner hair cells during development of hearing in mice. Nature 394, 281–284. Kuhn, S., Knirsch, M., Ruttiger, L., Kasperek, S., Winter, H., Freichel, M., Flockerzi, V., Knipper, M., Engel, J., 2009. Ba2+ currents in inner and outer hair cells of mice lacking the voltage-dependent Ca2+ channel subunits beta3 or beta4. Channels (Austin) 3. Lee, S., Briklin, O., Hiel, H., Fuchs, P., 2007. Calcium-dependent inactivation of calcium channels in cochlear hair cells of the chicken. J. Physiol. 583, 909–922. Leroy, J., Richards, M.W., Butcher, A.J., Nieto-Rostro, M., Pratt, W.S., Davies, A., Dolphin, A.C., 2005. Interaction via a key tryptophan in the I–II linker of N-type calcium channels is required for beta1 but not for palmitoylated beta2, implicating an additional binding site in the regulation of channel voltage-dependent properties. J. Neurosci. 25, 6984–6996. Liang, F., Zhang, B., Tang, J., Guo, J., Li, W., Ling, E.A., Chu, H., Wu, Y., Chan, Y.G., Cao, Q., 2007. RIM3 is a postsynaptic protein in the rat central nervous system. J. Comp. Neurol. 503, 501–510. Lipscombe, D., Raingo, J., 2007. Alternative splicing matters: N-type calcium channels in nociceptors. Channels (Austin) 1, 225–227.

Long, J.Z., Lackan, C.S., Hadjantonakis, A.K., 2005. Genetic and spectrally distinct in vivo imaging: embryonic stem cells and mice with widespread expression of a monomeric red fluorescent protein. BMC Biotechnol. 5, 20. Mangoni, M.E., Couette, B., Bourinet, E., Platzer, J., Reimer, D., Striessnig, J., Nargeot, J., 2003. Functional role of L-type Cav1.3 Ca2+ channels in cardiac pacemaker activity. Proc. Natl. Acad. Sci. U. S. A. 100, 5543–5548. Marcotti, W., Johnson, S.L., Rusch, A., Kros, C.J., 2003. Sodium and calcium currents shape action potentials in immature mouse inner hair cells. J. Physiol. 552, 743–761. Martini, M., Rispoli, G., Farinelli, F., Fesce, R., Rossi, M.L., 2004. Intracellular Ca2+ buffers can dramatically affect Ca2+ conductances in hair cells. Hear. Res. 195, 67–74. McRory, J.E., Hamid, J., Doering, C.J., Garcia, E., Parker, R., Hamming, K., Chen, L., Hildebrand, M., Beedle, A.M., Feldcamp, L., Zamponi, G.W., Snutch, T.P., 2004. The CACNA1F gene encodes an L-type calcium channel with unique biophysical properties and tissue distribution. J. Neurosci. 24, 1707–1718. Michna, M., Knirsch, M., Hoda, J.C., Muenkner, S., Langer, P., Platzer, J., Striessnig, J., Engel, J., 2003. Cav1.3 (alpha1D) Ca2+ currents in neonatal outer hair cells of mice. J. Physiol. 553, 747–758. Moody, W.J., Bosma, M.M., 2005. Ion channel development, spontaneous activity, and activity-dependent development in nerve and muscle cells. Physiol. Rev. 85, 883–941. Neef, J., Gehrt, A., Bulankina, A.V., Meyer, A.C., Riedel, D., Gregg, R.G., Strenzke, N., Moser, T., 2009. The Ca2+ channel subunit β2 regulates Ca2+ channel abundance and function in inner hair cells and is required for hearing. J. Neurosci. 29, 10730–10740. Obermair, G.J., Szabo, Z., Bourinet, E., Flucher, B.E., 2004. Differential targeting of the Ltype Ca2+ channel α1C (CaV1.2) to synaptic and extrasynaptic compartments in hippocampal neurons. Eur. J. NeuroSci. 19, 2109–2122. Obermair, G.J., Schlick, B., Di Biase, V., Subramanyam, P., Gebhart, M., Baumgartner, S., Flucher, B.E., 2010. Reciprocal interactions regulate targeting of calcium channel β subunits and membrane expression of α1 subunits in cultured hippocampal neurons. J. Biol. Chem. 285, 5776–5791. Opatowsky, Y., Chen, C.C., Campbell, K.P., Hirsch, J.A., 2004. Structural analysis of the voltage-dependent calcium channel beta subunit functional core and its complex with the α1 interaction domain. Neuron 42, 387–399. Platzer, J., Engel, J., Schrott-Fischer, A., Stephan, K., Bova, S., Chen, H., Zheng, H., Striessnig, J., 2000. Congenital deafness and sinoatrial node dysfunction in mice lacking class D L-type Ca2+ channels. Cell 102, 89–97. Pragnell, M., De Waard, M., Mori, Y., Tanabe, T., Snutch, T.P., Campbell, K.P., 1994. Calcium channel beta-subunit binds to a conserved motif in the I–II cytoplasmic linker of the α1-subunit. Nature 368, 67–70. Rabl, K., Thoreson, W.B., 2002. Calcium-dependent inactivation and depletion of synaptic cleft calcium ions combine to regulate rod calcium currents under physiological conditions. Eur. J. NeuroSci. 16, 2070–2077. Regus-Leidig, H., Tom Dieck, S., Specht, D., Meyer, L., Brandstatter, J.H., 2009. Early steps in the assembly of photoreceptor ribbon synapses in the mouse retina: the involvement of precursor spheres. J. Comp. Neurol. 512, 814–824. Schlick, B., Flucher, B., Obermair, G.J., 2010. Voltage-activated calcium channel expression profiles in mouse brain and cultured hippocampal neurons. Neuroscience. 167, 786–798. Schnee, M.E., Ricci, A.J., 2003. Biophysical and pharmacological characterization of voltage-gated calcium currents in turtle auditory hair cells. J. Physiol. 549, 697–717. Schoch, S., Mittelstaedt, T., Kaeser, P.S., Padgett, D., Feldmann, N., Chevaleyre, V., Castillo, P.E., Hammer, R.E., Han, W., Schmitz, F., Lin, W., Sudhof, T.C., 2006. Redundant functions of RIM1α and RIM2α in Ca2+-triggered neurotransmitter release. EMBO J. 25, 5852–5863. Shen, Y., Yu, D., Hiel, H., Liao, P., Yue, D.T., Fuchs, P.A., Soong, T.W., 2006. Alternative splicing of the Cav1.3 channel IQ domain, a molecular switch for Ca2+-dependent inactivation within auditory hair cells. J. Neurosci. 26, 10690–10699. Singh, A., Gebhart, M., Fritsch, R., Sinnegger-Brauns, M.J., Poggiani, C., Hoda, J.C., Engel, J., Romanin, C., Striessnig, J., Koschak, A., 2008. Modulation of voltage- and Ca2+dependent gating of Cav1.3 L-type calcium channels by alternative splicing of a Cterminal regulatory domain. J. Biol. Chem. 283, 20733–20744. Singh, A., Hamedinger, D., Hoda, J.C., Gebhart, M., Koschak, A., Romanin, C., Striessnig, J., 2006. C-terminal modulator controls Ca2+-dependent gating of Cav1.4 L-type Ca2+ channels. Nat. Neurosci. 9, 1108–1116. Sobkowicz, H.M., Rose, J.E., Scott, G.E., Slapnick, S.M., 1982. Ribbon synapses in the developing intact and cultured organ of Corti in the mouse. J. Neurosci. 2, 942–957. Song, H., Nie, L., Rodriguez-Contreras, A., Sheng, Z.H., Yamoah, E.N., 2003. Functional interaction of auxiliary subunits and synaptic proteins with Cav1.3 may impart hair cell Ca2+ current properties. J. Neurophysiol. 89, 1143–1149. Striessnig, J., 2007. C-terminal tailoring of L-type calcium channel function. J. Physiol. 585, 643–644. Subramanyam, P., Obermair, G.J., Baumgartner, S., Gebhart, M., Striessnig, J., Kaufmann, W.A., Geley, S., Flucher, B.E., 2009. Activity and calcium regulate nuclear targeting of the calcium channel β4b subunit in nerve and muscle cells. Channels (Austin) 3. Sudhof, T.C., 2004. The synaptic vesicle cycle. Annu. Rev. Neurosci. 27, 509–547. Takahashi, S.X., Miriyala, J., Tay, L.H., Yue, D.T., Colecraft, H.M., 2005. A Cavβ SH3/ guanylate kinase domain interaction regulates multiple properties of voltage-gated Ca2+ channels. J. Gen. Physiol. 126, 365–377. Tarabova, B., Lacinova, L., Engel, J., 2007. Effects of phenylalkylamines and benzothiazepines on Cav1.3-mediated Ca2+ currents in neonatal mouse inner hair cells. Eur. J. Pharmacol. 573, 39–48. tom Dieck, S., Altrock, W.D., Kessels, M.M., Qualmann, B., Regus, H., Brauner, D., Fejtova, A., Bracko, O., Gundelfinger, E.D., Brandstatter, J.H., 2005. Molecular dissection of the photoreceptor ribbon synapse: physical interaction of Bassoon and RIBEYE is essential for the assembly of the ribbon complex. J. Cell. Biol. 168, 825–836.

M. Gebhart et al. / Molecular and Cellular Neuroscience 44 (2010) 246–259 Vitko, I., Shcheglovitov, A., Baumgart, J.P., Arias II, O., Murbartian, J., Arias, J.M., Perez-Reyes, E., 2008. Orientation of the calcium channel beta relative to the α12.2 subunit is critical for its regulation of channel activity. PLoS ONE 3, e3560. Wang, Y., Okamoto, M., Schmitz, F., Hofmann, K., Sudhof, T.C., 1997. Rim is a putative Rab3 effector in regulating synaptic-vesicle fusion. Nature 388, 593–598. Wang, Y., Sudhof, T.C., 2003. Genomic definition of RIM proteins: evolutionary amplification of a family of synaptic regulatory proteins( small star, filled ). Genomics 81, 126–137. Wang, Y., Sugita, S., Sudhof, T.C., 2000. The RIM/NIM family of neuronal C2 domain proteins. Interactions with Rab3 and a new class of Src homology 3 domain proteins. J. Biol. Chem. 275, 20033–20044. Watschinger, K., Horak, S.B., Schulze, K., Obermair, G.J., Wild, C., Koschak, A., SinneggerBrauns, M.J., Tampe, R., Striessnig, J., 2008. Functional properties and modulation of

259

extracellular epitope-tagged CaV2.1 voltage-gated calcium channels. Channels (Austin) 2, 461–473. Xu, W., Lipscombe, D., 2001. Neuronal Cav1.3α1 L-type channels activate at relatively hyperpolarized membrane potentials and are incompletely inhibited by dihydropyridines. J. Neurosci. 21, 5944–5951. Yang, P.S., Alseikhan, B.A., Hiel, H., Grant, L., Mori, M.X., Yang, W., Fuchs, P.A., Yue, D.T., 2006. Switching of Ca2+-dependent inactivation of Cav1.3 channels by calcium binding proteins of auditory hair cells. J. Neurosci. 26, 10677–10689. Zampini, V., Johnson, S.L., Franz, C., Lawrence, N.D., Munkner, S., Engel, J., Knipper, M., Magistretti, J., Masetto, S., Marcotti, W., 2010. Elementary properties of Cav1.3 Ca2+ channels expressed in mouse cochlear inner hair cells. J. Physiol. 588, 187–199. Zhang, H., Fu, Y., Altier, C., Platzer, J., Surmeier, D.J., Bezprozvanny, I., 2006. Ca1.2 and Cav1.3 neuronal L-type calcium channels: differential targeting and signaling to pCREB. Eur. J. NeuroSci. 23, 2297–2310.