Functional analysis of congenital stationary night blindness type-2 CACNA1F mutations F742C, G1007R, and R1049W

Neuroscience 150 (2007) 335–345

FUNCTIONAL ANALYSIS OF CONGENITAL STATIONARY NIGHT BLINDNESS TYPE-2 CACNA1F MUTATIONS F742C, G1007R, AND R1049W J. B. PELOQUIN,1 R. REHAK,1 C. J. DOERING AND J. E. McRORY*

ake et al., 1986; Boycott et al., 2000; Haeseleer et al., 2004). CSNB-2 patients display a negative-type SchubertBornschein mixed cone–rod electroretinogram (ERG) signal in which the amplitude of the b wave is reduced. ERG recordings also display low amplitude photopic and scotopic b waves, indicating that cone and rod transmission to second order retinal neurons is affected (Miyake et al., 1986, 1987; Ruether et al., 1993; Tremblay et al., 1995; Lorenz et al., 1996). A 16 exon region corresponding to the 3= distal end of the CACNA1F gene was first identified by sequencing the Xp11.23-p11.22 interval (Fisher et al., 1997). Two subsequent reports identified the full length CACNA1F gene and showed that various mutations distributed throughout this gene are linked to the X-linked recessive disorder CSNB-2 (Bech-Hansen et al., 1998; Strom et al., 1998). The CACNA1F gene consists of 48 exons and encodes the ␣1 subunit of the Cav1.4 L-type voltage-gated calcium channel (Bech-Hansen et al., 1998; Strom et al., 1998). Cav1.4 is expressed in retina, as well as spleen, spinal cord, bone marrow and thymus (Kotturi et al., 2003; McRory et al., 2004). In the retina, immunohistochemical experiments using an antibody specific for Cav1.4 showed immunoreactivity in both rod and cone terminals of the macaque retina (Morgans et al., 2005). In the dark, the resting membrane potential of photoreceptors is depolarized (Witkovsky et al., 1997) and supports tonic glutamate release at the ribbon synapse (Morgans, 2000). Given that Cav1.4 is expressed in the synaptic terminals of photoreceptors, activates quickly, and has a large window current, this channel may be ideally suited for its function in the retina under constantly depolarized conditions (Koschak et al., 2003; McRory et al., 2004). To date, there have been a total of 80 CACNA1F mutations observed in non-related patients diagnosed with CSNB-2, 57 of which are unique and include: 20-missense, 12-deletion/insertion, 15-nonsense and 10-splice site mutation (Bech-Hansen et al., 1998; Strom et al., 1998; Boycott et al., 2001; Nakamura et al., 2001; Weleber, 2002; Wutz et al., 2002; Jacobi et al., 2003; Zito et al., 2003; Zeitz et al., 2005). Recently, a number of these mutations have been incorporated into the wild-type (WT) Cav1.4 construct and functionally examined using heterologous expression systems (reviewed in Doering et al., 2007). Some of these mutations display altered biophysical properties, and were gain or loss of function, while others have no effect on the biophysical properties of the channel (McRory et al., 2004; Hoda et al., 2005, 2006; Singh et al., 2006). In this study we characterize the bio-

Department of Physiology and Biophysics, Hotchkiss Brain Institute, University of Calgary, HMRB 172b, 3330 Hospital Drive Northwest, Calgary, Alberta, Canada T2N 4N1

Abstract—Congenital stationary night blindess-2 (incomplete congenital stationary night blindness (iCSNB) or CSNB-2) is a nonprogressive, X-linked retinal disease which can lead to clinical symptoms such as myopia, hyperopia, nystagmus, strabismus, decreased visual acuity, and impaired scotopic vision. These clinical manifestations are linked to mutations found in the CACNA1F gene which encodes for the Cav1.4 voltage-gated calcium channel. To better understand the physiological effects of these mutations, three missense mutants, F742C, G1007R and R1049W, previously shown to be mutated in patients with CSNB-2, were transiently expressed in human embryonic kidney (HEK) tsA-201 cells and characterized using whole-cell patch clamp. The G1007R mutation is located in transmembrane segment 5 (S5) of domain III and R1049W is located in the extracellular linker between S5 and the P-loop of domain III. Both mutants produced full length proteins that targeted to the membrane but did not support ionic currents. In 20 mM Ba2ⴙ, F742C (S6 domain II) produced a ⬃21 mV hyperpolarizing shift in half activation potential (Va[1/2]) and a ⬃23 mV hyperpolarizing shift in half inactivation potential (Vh[1/2]). Additionally, F742C displayed slower inactivation kinetics and a smaller whole cell conductance (Gmax). In physiological 2 mM Ca2ⴙ, F742C produced a ⬃19 mV hyperpolarizing shift in Va[1/2]. These findings suggest that the pathology of CSNB-2 in patients with these missense mutations in the Cav1.4 calcium channel is the result in either a gain of function (F742C) or a loss of function (G1007R, R1049W). © 2007 IBRO. Published by Elsevier Ltd. All rights reserved.

Congenital stationary night blindness-2 (CSNB-2) is clinically (Miyake et al., 1986; Boycott et al., 2000) and genetically heterogeneous (Bech-Hansen et al., 1998; Strom et al., 1998; Zeitz et al., 2006), characterized as a nonprogressive X-linked retinal disorder with an ophthalmologically normal fundus and symptoms that may include nystagmus, strabismus, myopia, hyperopia, decreased visual acuity, and variable degrees of night blindness (Miy1

Both authors contributed equally to this work. *Corresponding author. Tel: ⫹1-403-210-8553; fax: ⫹1-403-283-7137. E-mail address: [email protected] (J. E. McRory). Abbreviations: CSNB-2, congenital stationary night blindness type 2; Erev, reversal potential (mV); ERG, electroretinogram; Gmax, whole cell conductance (nS); HEK, human embryonic kidney; S, slope factor (mV); Va[1/2], half-activation potential (mV); Vh[1/2], half-inactivation potential (mV); WT, wild type; Z, slope factor reflecting charge movement (mV); ␶a, time constant for activation (ms).

0306-4522/07$30.00⫹0.00 © 2007 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2007.09.021

335

336

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

physical properties of three additional CSNB-2 missense mutations using whole cell patch clamping and immunocytochemistry. We provide further evidence that CSNB-2 mutations can result in either gain (F742C) or loss (G1007R and R1049W) of function of the biophysical properties of the Cav1.4 channel.

ramping cells from ⫺100 to ⫹80 mV over a time of 180 ms (1 mV/ms). Voltage-dependence of inactivation was assessed by holding cells at various conditioning potentials, ranging from ⫺60 mV to ⫹30 mV for 10 s before and after an application of a test depolarization to 0 mV.

EXPERIMENTAL PROCEDURES

Data analysis and offline leak subtraction was carried out using Clamp-fit 9.2 (Axon Instruments) and SigmaPlot 2000 (SPSS Inc.). Current–voltage relations were fitted according to a modified Boltzmann equation (Equation 1), where Gmax is the maximum slope whole cell conductance, I is peak current amplitude, Va[1/2] is the half-activation potential, S is the slope factor, Erev is the reversal potential, and V is the test potential. Erev was extrapolated from fitting the chord conductance of the current–voltage relations, and not the physiological Erev per se. Activation curve was plotted using Va[1/2] and S parameters obtained from the Boltzmann equation.

Cloning of the human Cav1.4 calcium channel mutations To generate the three mutations a 1.8 kb fragment of the full length human Cav1.4 calcium channel ␣1 subunit (McRory et al., 2004) was subcloned into pGEM-T Easy (Promega, Fitchburg, WI, USA). Site-directed mutagenesis of this construct was carried out using the QuickChangeII Site-Directed Mutagenesis kit (Stratagene) according to manufacturer’s instructions. After confirmation of the appropriate mutations by direct DNA sequencing, the fragments were transferred back into the full length Cav1.4 calcium channel ␣1 subunit in pcDNA3.1-ZEO using AgeI and ClaI, and the sequence was re-confirmed.

Transient transfection of tsA-201 cells and electrophysiological characterization The Cav1.4 calcium channel was transfected in a heterologous expression system and characterized using various whole cell patch-clamp recording techniques as previously described (McRory et al., 2004). Briefly, human embryonic kidney (HEK)-293 tsA-201 cells were cultured at 37 °C in a humidified environment of 5% CO2 and bathed in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 200 units/ml penicillin, and 0.2 mg/ml streptomycin (Invitrogen). Cells were grown to a confluency of 80%–90%, enzymatically dissociated using trypsin-EDTA (0.25%), and dissociated cells were plated on 10 mm glass coverslips in 110 mm dishes. Following a 12 h recovery time, cells were washed and transfected using a standard calcium phosphate method. In each transfection, 6 ␮g of pore forming WT or mutant Cav1.4 human cDNA was included along with 6 ␮g of the auxiliary subunits rat-␤2a and rat ␣2␦ cDNA. Additionally, 2 ␮g of cDNA encoding enhanced green fluorescent protein was also transfected as a transfection marker. Transfected cells were incubated at 37 °C (5% CO2) for 12 h, washed, and then transferred to 28 °C (5% CO2) for at least 24 h prior to electrophysiological recordings. For whole-cell recordings, coverslips were transferred into a recording chamber containing (mM): 20 BaCl2, 1 MgCl2, 10 Hepes, 40 tetraethylammonium chloride (TEA-Cl), 10 glucose, and 65 CsCl, pH 7.2 adjusted with tetraethylammonium-hydroxide (TEA-OH) (in some experiments, 2 mM CaCl2 was replaced with 20 mM BaCl2). Physiological recordings were obtained by perfusing 2 mM Ca2⫹ onto cells bathed in 20 mM barium. Borosilicate glass micropipettes were pulled on a Sutter microelectrode puller and fire polished (Narashige microforge) to a resistance between 2 and 5 M⍀ when filled with intracellular solution (in mM: 108 CsCH3SO3, 4 MgCl2, 9 EGTA, 9 Hepes, pH 7.2 adjusted with cesium-hydroxide (Cs-OH). Pharmacological recordings were performed in 20 mM Ba2⫹ with 10 ␮M (⫾) Bay K8644 (SigmaRBI). All data were acquired at room temperature. Recordings were performed using an Axopatch 200B amplifier (Axon Instruments, Union City, CA, USA) linked to a personal computer with Digidata 1322A interface. Currents were filtered at 1 kHz and digitized at 4 kHz. Series resistance was compensated by 85%. Current–voltage relations were acquired by holding cells at ⫺100 mV and stepping to various test potentials ranging from ⫺60 mV to ⫹60 mV for 150 ms. For paired 20 mM Ba2⫹ and 2 mM Ca2⫹ experiments, current–voltage relations were obtained by

Data analysis

I⫽

V⫺Erev 1⫹e

Va 关1⁄2兴⫺V S

Gmax

(1)

Inactivation curves after a 10 s conditioning pulse were fitted with a modified Boltzmann relationship in Equation 2, where Inormalized is the fractional current available following a given conditioning prepulse, x is the fraction of non-inactivating current, V is the potential of the conditioning pulse, Vh[1/2] is the half-inactivation potential, and Z is a S reflecting the effective gating charge. Inormalized⫽x⫹

1⫺x 1⫹e

⫺z(Vh 关1⁄2兴⫺V) 25.6

(2)

Time constants for activation (␶a) and inactivation (␶h) at various voltages were obtained from monoexponential fits to raw current data. Normalized current depicted in some figures was determined by calculating the fraction of current relative to maximal peak current for a given cell. Values listed are mean⫾S.E.M., with cell numbers indicated in parentheses. All statistical analyses were performed with SigmaStat 2.03 (SPSS Inc.), using one-way analysis of variance Bonferroni test (with the exception of ␶a where a Student’s t-test was performed). * Denotes a significance level set at Pⱕ0.05 and ** at Pⱕ0.001.

Immunohistochemical analysis of WT and mutant Cav1.4 channels To confirm plasma membrane distribution of the mutant channels we stained transiently transfected tsA-201 cells with a polyclonal antibody specific for Cav1.4 (McRory et al., 2004). Briefly, transfected cells were grown on poly-L-coated coverslips for 2 days at 37 °C. The coverslips were washed with PBS, fixed in 4% paraformaldehyde (5 min), and then washed 2⫻ with PBS. The cells were permeabilized with 3% BSA/PBS containing 0.1% Triton X-100 for 30 min at room temperature. The coverslips were incubated overnight at 4 °C in 3% BSA/PBS containing the Cav1.4 antibody (1:10,000 dilution). Coverslips were washed (three times) for 10 min with 3% BSA/PBS. Cy3-conjugated secondary sheep-anti rabbit antibody (Sigma; 1:1000 dilution) was applied and hybridized for 30 min in 3% BSA/PBS. The coverslips were washed (three times) with 3% BSA/PBS and mounted on slides with VECTASHIELD containing DAPI (Vector). Images were acquired using a Zeiss LSM-510 META confocal microscope.

Western blot analysis of WT and mutant Cav1.4 channels For Western blots, transfected HEK-293 tsA-201 cells in 10 cm dishes were lysed on ice with lysis buffer (300 mM NaCl, 50 mM

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345 Tris pH 7.5, 0.1% Triton X-100) containing a protease inhibitor cocktail (complete Mini EDTA-free, Roche). Standard polyacrylamide gel electrophoresis (6% resolving gel) techniques were used, using 10 ␮l of cell lysate per well. Proteins were transferred to PVDF membranes, and Cav1.4 protein detected using Cav1.4 antibody (1:5000 dilution) and ECL anti-rabbit secondary (1:5000, GE Healthcare).

RESULTS Wutz et al. (2002) reported 20 novel mutations found in the CACNA1F gene of families diagnosed with CSNB-2. Eleven of these mutations were missense mutations, two of which we have chosen to characterize here: F742C occurs in IIS6 segment and G1007R located in the IIIS5 (Fig. 1A). Strom et al. (1998) reported nine different mutations in the CACNA1F gene from 13 different families diagnosed with CSNB-2, and we characterized R1049W located in the extracellular linker between the IIIS5 and the P-loop. Sequence alignment of residues F742, G1007 and GR1049 reveals a high level of conservation among L- and P/Q-type channels (Fig. 1A) in these regions. Immunocytochemistry and immunoblotting of G1007R and R1049W The average peak current in mock-transfected cells transfected with cDNAs coding for only ␤2a and ␣2-␦ subunits,

337

as well as eGFP was ⫺10⫾3 (n⫽8) (current density⫽ 0.5⫾2 pA/pF, n⫽8), and increased to ⫺12⫾2 (n⫽24) (current density⫽0.6⫾0.1 pA/pF, n⫽24) in the presence of (⫾) Bay K8644. Therefore, only cells with currents exceeding these values could be attributed to the presence of the channel and were used for analysis. Despite several attempts, we were unable to record currents from transfections with G1007R and R1049W that exceeded endogenous levels in mock-transfected cells (n⫽43 G1007R, n⫽37 R1049W, n⫽9 mock transfected; combined data from three transfections; summarized in Table 1), although we were able to record currents attributable to the WT clone within the same transfection. To determine if the mutant clones are expressed and targeted to the plasma membrane, we used immunohistochemistry and Western blots with an antibody specific to Cav1.4 (McRory et al., 2004). As shown in Fig. 2A, G1007R and R1049W mutants show staining in the plasma membrane. In addition protein was detected using Western blot analysis; mutant protein migrated with similar molecular weights as WT, suggesting that mutant proteins are full length channels (Fig. 2B). The L1068P CSNB-2 mutant reported by Hoda et al., 2005 yielded current, but only in the presence of the channel activator Bay K8644. G1007R and R1049W were also tested in the presence of 10 ␮M (⫾) Bay K8644, but did not

Fig. 1. CSNB-2 mutations F742C, G1007R and R1049W alter channel gating. (A) Schematic representation of the Cav1.4 ␣ subunit with the location of missense mutations F742C (gray circle), G1007R (gray triangle) and R1049W (black triangle) depicted on the channel. A ClustalW sequence alignment using accession numbers for Cav1.4␣1 (NM_005183), Cav1.3␣1 (NM_000720), Cav1.2␣1 (NM_000719), Cav1.1␣1 (NM_000069) and Cav2.1␣1 (NM_023035) was performed for a small region upstream and downstream from each mutation. The numbering of mutation is according to Strom et al. (1998) (AJOO6216). Residues with a light-gray background indicate the location of the CSNB-2 mutation. In the alignment for the F742C mutant, black squares indicate a cluster of four critical residues important for gating (Hohaus et al., 2005). The black circle indicates a naturally occurring mutation discovered from molecular cloning of the human fibroblast Cav1.2 ␣1 channel (Soldatov, 1992; Soldatov et al., 2000). The black triangle represents a CACNA1F mutation linked to an X-linked retinal disorder, similar but distinct from CSNB-2 (Hemara-Wahanui et al., 2005).

0.6⫾0.1 (n⫽24)

⫺12.4⫾2.0 (n⫽24)

produce an inward current that was significantly greater than mock-transfected cells (n⫽21 G1007R, n⫽21 R1049W, n⫽24 mock transfected; data from three transfections; Table 1). In contrast, all Cav1.4 WT cells in 10 ␮M (⫾) Bay K8644 produced robust current in 11 of 11 cells from the same transfections (Fig. 2C). Taken together, these results suggest that G1007R and R1049W mutants are expressed and targeted to the plasma membrane but are incapable of supporting currents detectable above background endogenous currents even in the presence of L-type agonist (⫾) Bay K8644. F742C mutation in 20 mM Ba2ⴙ affects whole-cell conductance, Va[1/2], Erev, and open probability

ND

Fig. 3A illustrates ensemble waveforms for F742C and WT Cav1.4 channels acquired by step I–V protocol. In 20 mM Ba2⫹ the F742C mutant began to activate at ⬃⫺45 mV, peaked ⬃⫺5 mV, and displayed very slow inactivation kinetics; WT began to activate ⬃⫺25 mV, peaked ⬃⫹15 mV, and inactivated with faster kinetics. These differences are further emphasized in Fig. 3 B, which depicts mean normalized I–V relations for the F742C mutant and WT constructs. The F742C mutation resulted in a hyperpolarization of the voltage dependence of activation by 21 mV (step I–V protocol). Mean values (taken from individual Boltzmann fits) were: WT, Va[1/2]⫽⫹5.2⫾1.0 mV and S⫽6.1⫾0.4 mV (n⫽24); for F742C, Va[1/2]⫽⫺15.9⫾ 1.1 mV and S⫽6.4⫾0.3 mV (n⫽16). These values were used to generate the open probability plots in Fig. 3C, which displays the leftward shift of F742C relative to WT. The mean normalized current as a function of voltage illustrated in Fig. 3B was fitted with a Boltzmann equation and yielded similar parameters: WT Va[1/2]⫽3.7 mV, Erev 53 mV, S⫽6.3 mV (n⫽18 –24), and the F742C mutant Va[1/2]⫽⫺16.8 mV, Erev 33.6 mV, S 6.7 mV (n⫽11–16). Fig. 3D emphasizes the differences in the Gmax and Erev between WT and F742C mutant (step I–V protocol). F742C was characterized by significantly smaller wholecell conductance (Gmax⫽3.4⫾0.6 nS (n⫽16)) and hyperpolarized Erev (Erev⫽30.8⫾1.9 (n⫽24)) compared with WT (Gmax⫽6.5⫾0.4 (n⫽24), 49.2⫾1.0 (n⫽24). The mutant also displayed significantly different time constants of activation relative to WT at several voltages, as depicted in Fig. 3E. Values were determined from monoexponential fits to raw currents elicited by stepping to various test voltages. These values were: ␶a (ms)⫽7.5 (⫺20 mV), 5.9 (⫺10 mV), 4.8 (0 mV), 3.3 (⫹20 mV), 3.4 (⫹30 mV) for WT; ␶a (ms)⫽5.2 (⫺20 mV), 4.6 (⫺10 mV), 3.7 (0 mV), 3.8 (⫹20 mV), 3.9 (⫹30 mV) for F742C. F742C mutation in physiological 2 mM Ca2ⴙ affects Gmax, Va[1/2], Erev, and current density

* P ⬍ 0.05. ** P ⬍ 0.001.

ND ND

ND

⫺414⫾107** (n⫽11) 13.8⫾3.4 (n⫽11) ND ND ND

ND

⫺21.4⫾2.7 (n⫽21) 0.8⫾0.1 (n⫽21) ND ND

ND

ND

⫺8.7⫾1.3 (n⫽21) 0.4⫾0.1 (n⫽21) ND ND

ND

ND

⫺13.3⫾2.8** (n⫽6) 0.6⫾0.1** (n⫽6) ⫹19⫾4** (n⫽6) ⫺18.0⫾2.0** (n⫽6)

⫹0.8⫾0.3* (n⫽6)

⫹6.5⫾0.8 (n⫽6)

⫺58.4⫾10.6** (n⫽6) 2.9⫾0.6** (n⫽6) ⫹6.1⫾0.5 (n⫽6) ⫹36⫾2** (n⫽6) ⫺9.4⫾1.0** (n⫽6)

⫹2.2⫾0.8** (n⫽6)

ND ND ⫹6.4⫾0.3 (n⫽16) ⫹3.4⫾0.6* (n⫽16) ⫹31⫾2** (n⫽16) ⫺15.9⫾1.1** (n⫽16)

⫺47⫾10 (n⫽6) 2.0⫾0.5 (n⫽6) ⫹45⫾3 (n⫽6) ⫹0.6⫾2.0 (n⫽6)

1.6⫾0.3 (n⫽6)

7.0⫾0.5 (n⫽6)

⫺155⫾39 (n⫽6) 6.4⫾1.9 (n⫽6) 7.3⫾0.6 (n⫽6) 5.6⫾1.2 (n⫽6) ⫹55⫾2 (n⫽6) ⫹11.0⫾1.5 (n⫽6)

ND ⫹6.1⫾0.4 (n⫽24) ⫹6.5⫾0.9 (n⫽24) ⫹49⫾1.0 (n⫽24) ⫹5.2⫾1.0 (n⫽24)

Cav1.4 WT 20 mM Ba2⫹ (step protocol) Cav1.4 WT 20 mM Ba2⫹ (ramp protocol) Cav1.4 WT 2 mM Ca2⫹ (ramp protocol) Cav1.4 F742C 20 mM Ba2⫹ (step protocol) Cav1.4 F742C 20 mM Ba2⫹ (ramp protocol) Cav1.4 F742C 2 mM Ca2⫹ (ramp protocol) Cav1.4 G1007R 20 mM Ba2⫹ (⫾)BAY-K 8644 (ramp protocol) Cav1.4 R1049W 20 mM Ba2⫹ ⫹ (⫾)BAY-K 8644 (ramp protocol) Cav1.4 WT 20 mM Ba2⫹ (⫾)BAY-K 8644 (ramp protocol) Mock 20 mM Ba2⫹ (⫾)BAY-K 8644 (ramp protocol)

Current density (pA/pF) S (mV) Gmax (nS) Erev (mV) Va1/2 (mV)

Table 1. Biophysical properties of Ba2⫹ and Ca2⫹ currents through heterologously expressed Cav1.4 WT and CSNB-2 mutants F742C, G1007R, and R1049W

ND

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

Mean current (pA)

338

Figs. 4A and B are representative traces of inward Ba2⫹ and Ca2⫹ currents elicited from a ramp protocol for Cav1.4 WT and F742C respectively. For each ramp experiment, cells were recorded in 20 mM Ba2⫹ external solution and subsequently perfused with 2 mM Ca2⫹. Switching from 20 mM Ba2⫹ to 2 mM Ca2⫹ decreased current amplitude and hyperpolarized the current–voltage relation. Fig. 4C

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

G1007R

250 kDa

150 kDa

Current Density (pA/pF)

F742C

G1007R

R1049W

Control

C WT

B

R1049W

10 µM Bay K8644 20

**

(11)

16 12 8 4 (21)

(21)

(24)

MOCK

F742C

R1049W

WT

G1007R

A

339

WT

0

Fig. 2. (A) Immunohistochemistry of representative tsA-201 HEK cells expressing WT and mutant Cav1.4 calcium channels. G1007R and R1049W are non-functional channels despite being well targeted and expressed at the plasma membrane. WT and mutant ␣1 subunit were co-expressed with ␤2a, ␣2␦, and GFP. Top panel depicts fluorescent cells overlaid with transmitted cells; bottom panel displays only transmitted cells. (B) Western blot of Cav1.4 WT and the CSNB-2 mutants. Cells were transfected with WT or mutant cDNA and auxiliary subunits as described in Experimental Procedures. Cav1.4 subunits were detected using anti Cav1.4 Ab and migrated with the expected molecular weight of ⬃220 kDa. (C) Bar graph depicting mean current density (J) upon exposure to 10 ␮M of (⫾) Bay K8644. In the presence of (⫾) Bay K8644, current density for G1007R and R1049W was not significantly larger than endogenous current observed in mock transfected cells (␣2␦⫹␤2a⫹GFP only). In contrast, WT cells consistently displayed robust current in the presence of (⫾) Bay K8644 (11 of 11 cells).

illustrates the hyperpolarizing shift in Va1/2 for F742C relative to WT recorded in 2 mM Ca2⫹and 20 mM Ba2⫹. From ramp protocol, the F742C mutant shifted the voltage dependence of activation by 19 mV and 20 mV in the hyperpolarizing direction in 2 mM Ca2⫹ and 20 mM Ba2⫹ respectively. Mean values (taken from individual Boltzmann fits) recorded in 2 mM Ca2⫹ were: WT, Va[1/2]⫽⫹0.6⫾2.0 mV (n⫽6); for F742C, Va[1/2]⫽⫺18.0⫾2.0 mV (n⫽6). Addition-

ally, in 2 mM calcium the F742C caused the Erev to be significantly shifted by 26 mV in the hyperpolarized direction, and the Gmax was decreased twofold (WT, Erev⫽⫹45⫾3 mV and Gmax⫽1.6⫾0.3 (n⫽6); for F742C, Erev⫽⫹19⫾4 mV and Gmax⫽0.8⫾0.3 (n⫽6)). The S for F742C was not significantly different from WT in 2 mM Ca2⫹. Fig. 4D depicts the effect of F742C on average current density (n⫽6 cells) in both 20 mM Ba2⫹ and 2 mM Ca2⫹ as compared with WT. Mean average

340

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

Fig. 3. Functional analysis of F742C, G1007R and R1049W CACNA1F missense mutations linked to CSNB-2 in 20 mM Ba2⫹. (A) Representative traces for WT and the F742C Cav1.4 mutant, elicited by various step depolarization from a holding potential of ⫺100 mV. F742C is shown to reach peak current at more negative step depolarization as compared with WT. (B) Normalized current–voltage relation for WT and all three mutants. G1007R and R1049W mutants produced nonfunctional channels (n⫽43 and n⫽37 respectively), while F742C drastically altered the voltage dependence of activation (n⫽16) as compared with WT (n⫽24). (C) Activation plots (generated from mean Va[1/2], and S parameters) exemplifying the large hyperpolarizing shift in the open probability. F742C had no significant change in S, but a large significant ⬃⫺21 mV change in Va[1/2] illustrated in the inset. (D) Effects of the F742C mutant on the Erev and Gmax. F742C; Gmax⫽3.4⫾0.6 nS (n⫽16) and Erev⫽30.8⫾1.9 (n⫽24). In contrast, WT; Gmax⫽6.5⫾0.4 (n⫽24), 49.2⫾1.0 (n⫽24). (E) ␶a As a function of voltage with the F742C mutant having faster activation kinetics than WT.

current densities in 20 mM Ba2⫹ were: WT⫽6.4⫾1.9 pA/pF (n⫽6) and F742C⫽2.9⫾0.6 pA/pF (n⫽6), and in 2 mM Ca2⫹: WT⫽2.0⫾0.5 pA/pF (n⫽6), and F742C⫽0.6⫾0.1 pA/pF (n⫽6). F742C mutation affects inactivation properties Next, we tested to see if the F742C mutation also affects inactivation properties. Fig. 5A is a plot of inactivated current after a 10 s conditioning pulse to various test potentials recorded in 20 mM Ba2⫹. The F742C mutant hyperpolarizes

Vh[1/2] by a ⬃23 mV from ⫺6.5⫾3.9 mV (n⫽8, WT) to ⫺29.9⫾2.7 mV (n⫽12, F742C). Additionally, the Z factor was determined to be more positive for the F742C mutant (4.7⫾0.8 mV) as compared with WT (2.7⫾0.2 mV). Fig. 5A also illustrates that F742C mutant had a larger fraction of current that would not inactivate compared with WT current during the 10 s inactivating test pulse. Fig. 5B depicts superimposed normalized representative traces obtained at a test potential of 0 mV. Here, the F742C mutant displays slower inactivation kinetics as compared with the WT clone. Similar

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

341

Fig. 4. Functional analysis of F742C in physiological recording solution of 2 mM Ca2⫹. (A, B) Representative I–V relation obtained from ramp protocol in paired 20 mM Ba2⫹ and 2 mM Ca2⫹ experiments for both WT and F742C. (C) Mean Va[1/2] (taken from individual Boltzmann fits) for WT and F742C in 20 mM Ba2⫹ and 2 mM Ca2⫹ acquired from ramp protocol (1 mV/ms). (D) Effects of F742C on mean current density recorded in 20 mM Ba2⫹ and 2 mM Ca2⫹ as compared with WT.

observation can be observed in Fig. 5C, D and E which compare the percent of remaining current over a time of 1 s, 5 s, and 10 s at 0 mV, ⫺10 mV, and, Vmax, respectively.

DISCUSSION To date, a total of nine mutations in the CACNA1F gene linked to the human retinal disease CSNB-2 have been functionally characterized in expression systems and are hypothesized to induce their pathology through different molecular mechanisms (McRory et al., 2004; Hoda et al., 2005, 2006; Singh et al., 2006; Doering et al., 2007). Here, we show three CSNB-2 mutations severely alter the gating properties resulting in gain or loss of function mutations. G1007R and R1049W mutations, despite being expressed and targeted to the plasma membrane, did not elicit detectable currents (even in the presence of 10 ␮M (⫾) Bay K8644). Alternatively, the F742C mutant did support ionic currents with different biophysical properties from WT.

G1007R and R1049W mutants produce non-functional channel In this study we show that the G1007R and R1049W mutants are unable to support ionic currents. These mutations could disrupt the functionality of the conducting pore, alter channel gating, or targeting of the channel to the plasma membrane. Immunohistochemistry and Western blot analyses (Fig. 2A and 2B) suggest the loss of function effect mediated by these mutants is not a direct result of protein synthesis failure, as the channels are expressed and targeted to the plasma membrane in our expression system. Previous reports identified two CSNB-2 mutants, S229P and L1068P that were also unable to support calcium channel activity, despite protein expression, in HEK tsA-201 cells (Hoda et al., 2005). These results suggest that G1007W and R1049W mutations are critical to channel gating but still allow for protein expression and targeting to the plasma membrane.

342

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

Fig. 5. The effects of the F742C mutant on voltage dependence and kinetics of inactivation. (A) Voltage dependence of inactivation of IBa through WT and F742C mutant channel, derived from 10 s conditioning pulsed to various step depolarizations. Note the large fraction of current unable to inactivate and the negative shift in the Vh1/2 for the F742C mutant. (B) Two representative traces evoked by a 10 s test depolarization to 0 mV, normalized and superimposed to highlight the effect of the F742C on the inactivation kinetics (note the slower inactivation kinetics of the F742C). (C–E) Percent of remaining IBa current after 1 s, 5 s and 10 s observed at test depolarizations to 0 mV, ⫺10 mV, and Vmax, respectively.

F742C mutants dramatically alters various aspects of gating properties F742C mutation produced a large shift in the voltage dependence of activation and inactivation (⬃⫺21 mV and ⬃⫺23 mV, respectively, in 20 mM Ba2⫹). The F742C mutant also displayed slower inactivation kinetics and had a larger fraction of noninactivating current compared with WT current with the 10 s conditioning pulse protocol tested. This noninactivating current could reflect a slower time course of inactivation compared with WT within the 10 s conditioning pulse, and hence inactivation was not at steady-state. A similar phenomenon was also reported for the I745T CSNB-2 mutation (Hemara-Wahanui et al., 2005). It has also previously been shown that increasing the length of the conditioning pulse from 10 s to 20 s or

30 s does not statistically alter the Vh[1/2] for murine Cav1.4 channels, (Baumann et al., 2004), even though the 10 s pulse was insufficient to reach a steady-state inactivation level. Hence, increasing the duration of the pulse may affect the fraction of non-inactivating current we observed, but would not affect the Vh[1/2]. F742C also shifted the Erev, similar to previously described CSNB-2 mutants G369D (Hoda et al., 2005) and R508Q and L1364H (Hoda et al., 2006), indicative of a change in the cation selectivity (Yang et al., 1993). Under our experimental conditions the Erev is determined by the relative outflux of the intracellular Cs1⫹ ions and the influx of the extracellular 20 mM Ba2⫹ or 2 mM Ca2⫹ions (Fenwick et al., 1982; Lee and Tsien, 1982, 1984). The large hyperpolarizing shift in the Erev may be indicative of an increase in the outflux of Cs1⫹, and/or a

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

decrease in Ba2⫹ influx (Hagiwara and Byerly, 1981). The F742C leads to a charge substitution, replacing a neutral phenylalanine for a more positively charged cysteine residue. This newly added positively charged residue is found in S6 domain II, relatively close to the selectivity filter. The appearance of larger shift in the Erev caused by a point mutation could be the result of two factors: a direct effect on the selectivity filter, or allosteric conformational changes in the channel, that manifest themselves as altered selectivity between divalent and monovalent ions. Under our experimental conditions the Erev is determined by the relative outflux of the intracellular Cs1⫹ ions and the influx of the extracellular 20 mM Ba2⫹ or 2 mM Ca2⫹ ions. The large hyperpolarizing shift in the Erev may be indicative of an increase in the outflux of Cs1⫹, and/or a decrease in Ba2⫹ influx. The F742C leads to a charge substitution, replacing a neutral phenylalanine for a more positively charged cysteine residue. This newly added positively charged residue is found in S6 domain II, relatively close to the selectivity filter. The appearance of a larger shift in the Erev caused by a point mutation could be the result of two factors: a direct effect on the selectivity filter, or allosteric conformational changes in the channel, that manifest themselves as altered selectivity between divalent and monovalent ions. Lastly, F742C is the first reported functional CACNA1F CSNB-2 mutant to show a significant decrease in Gmax/current density, as well as a significantly different time constants of activation relative to WT. Next, we studied the effects F742C in a more physiological relevant recording solution of 2 mM Ca2⫹. I–V relations in 2 mM Ca2⫹ were acquired by ramp protocol and the effects of the F742C were shown to be consistent with those recorded in 20 mM Ba2⫹. In physiological 2 mM Ca2⫹, F742C displayed a large 19 mV hyperpolarizing shift in the voltage dependence of activation, an ⬃threefold decrease in current density, an ⬃twofold decrease in Gmax, and a 25 mV hyperpolarizing shift in the Erev. Irrespective of external ion (i.e. 20 mM Ba2⫹ or 2 mM Ca2⫹) F742C yields smaller and more hyperpolarized I–V relations as compared with WT. Inactivation protocols and step I–V protocols were to long and cumbersome to record F742C in 2 mM Ca2⫹ given the extremely small current size normally accompanying each experiment (mean peak current; F742C⫽⫺13.3⫾2.8 pA compared with WT⫽⫺47⫾ 10 pA for WT). Results obtained by ramp protocol nonetheless demonstrate that many of the effects observed by F742C in 20 mM Ba2⫹ are synonymous to the effects shown in physiological 2 mM Ca2⫹. Physiological consequences of the F742C gain of function mutation F742C shifts the voltage dependence of activation to more negative potentials, speeds voltage dependent activation and slows voltage dependent inactivation kinetics, changes which promote channel activity. In contrast, Gmax was decreased thus supporting a decrease in calcium entry. Hoda et al. (2005) suggested the gain of function G369D CSNB-2 mutation, with an ⬃⫺11 mV shift in the activation curve, would reduce the dynamic

343

range of tonic glutamate release when the plasma membrane is hyperpolarized from ⫺35 mV (dark) to ⫺50 mV (light) by ⬃sixfold. This decrease in dynamic range may explain how a gain of function mutation compromises signaling in the retina. Using a similar approach, our data revealed that the F742C mutation with a Va[1/2] shift of ⬃⫺19 mV (2 mM Ca2⫹), reduced the dynamic range of tonic glutamate release by ⬃sevenfold. Therefore CSNB-2 mutations with a hyperpolarizing shift in their Va[1/2] (i.e. G369D ⬃11 mV (Hoda et al., 2005), K1591X ⬃13 mV (Singh et al., 2006) and F742C ⬃19 mV) allow for a greater influx of calcium at any given voltage within the operating range of a photoreceptor, but the dynamic range can be dramatically reduced when compared with physiological conditions. F742 is part of an important cluster of residues in IIS6 of L-type calcium channels The F742C mutant is located immediately adjacent to a cluster of four critical amino acids (Cav1.4 Leu743-Ala-IleAla) previously shown to have a large effect on the biophysical properties of other L-types when these amino acids are sequentially mutated (Hohaus et al., 2005) (see Fig. 1A). This important region was initially identified by functional analysis of the I745T mutation identified in a New Zealand family diagnosed with mental defects and CSNB-2-like phenotypes (Hemara-Wahanui et al., 2005). Substituting residues of different hydrophobicity, size, and polarity along the IIS6 region of the Cav1.2 channel (Hohaus et al., 2005) hyperpolarized the voltage dependence of activation and inactivation by as much as ⬃37 mV. Interestingly, the Cav1.2 F778 residue (analogous to F742 in Cav1.4) was also tested and substituted with a proline, but no statistical difference in the activation profile was observed (Cav1.2 F778P⫽Va[1/2] ⫺9.3⫾0.8 mV, where Cav1.2 WT⫽Va[1/2] ⫺9.9⫾1.1) (Hohaus et al., 2005). Our results show that replacing a non-polar hydrophobic phenylalanine with a small polar cysteine immediately adjacent to the critical L-A-I-A residues of IIS6, dramatically alters the gating properties of the channel. It is therefore likely that Phe-742 is critical to gating, just as is the case for the adjacent residues L-A-I-A. It is not surprising, given the close proximity, that the F742C mutant displayed very similar changes in gating properties to the I745T upon functional characterization in tsA-201 HEK cells. Both mutations dramatically shifted the voltage dependence of activation in the hyperpolarizing direction, had similar voltage dependence of inactivation profiles, and shared slower inactivation kinetics (HemaraWahanui et al., 2005). Interestingly, carriers afflicted with the F742C mutation have been clinically diagnosed with CSNB-2 (Wutz et al., 2002). In contrast carriers of the I745T have been diagnosed with a more severe retinal disease. The clinical manifestation of the I745T mutant in males appears to be more severe than CSNB-2, affects heterozygous female carriers, and may be linked to intellectual disability (Hemara-Wahanui et al., 2005). It would be interesting to re-analyze and compare phenotypes currently observed in members of the I745T New Zealand

344

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345

family with carriers of the F742C in the Belgium family B27 reported by Wutz et al. (2002). Perhaps, common phenotypes may be associated with mutations located in the IIS6 region, particularly in regard to the well-conserved F-L-AI-A domain.

REFERENCES Baumann L, Gerstner A, Zong X, Biel M, Wahl-Schott C (2004) Functional characterization of the L-type Ca2⫹ channel Cav1.4alpha1 from mouse retina. Invest Ophthalmol Vis Sci 45:708 –713. Bech-Hansen NT, Naylor MJ, Maybaum TA, Pearce WG, Koop B, Fishman GA, Mets M, Musarella MA, Boycott KM (1998) Loss-offunction mutations in a calcium-channel alpha1-subunit gene in Xp11.23 cause incomplete X-linked congenital stationary night blindness. Nat Genet 19:264 –267. Boycott KM, Maybaum TA, Naylor MJ, Weleber RG, Robitaille J, Miyake Y, Bergen AA, Pierpont ME, Pearce WG, Bech-Hansen NT (2001) A summary of 20 CACNA1F mutations identified in 36 families with incomplete X-linked congenital stationary night blindness, and characterization of splice variants. Hum Genet 108:91–97. Boycott KM, Pearce WG, Bech-Hansen NT (2000) Clinical variability among patients with incomplete X-linked congenital stationary night blindness and a founder mutation in CACNA1F. Can J Ophthalmol 35:204 –213. Doering CJ, Peloquin JB, McRory JE (2007) The Cav1.4 calcium channel: More than meets the eye. Channels 11:6 –11. Fenwick EM, Marty A, Neher E (1982) Sodium and calcium channels in bovine chromaffin cells. J Physiol 331:599 – 635. Fisher SE, Ciccodicola A, Tanaka K, Curci A, Desicato S, D’Urso M, Craig IW (1997) Sequence-based exon prediction around the synaptophysin locus reveals a gene-rich area containing novel genes in human proximal Xp. Genomics 45:340 –347. Haeseleer F, Imanishi Y, Maeda T, Possin DE, Maeda A, Lee A, Rieke F, Palczewski K (2004) Essential role of Ca2⫹-binding protein 4, a Cav1.4 channel regulator, in photoreceptor synaptic function. Nat Neurosci 7:1079 –1087. Hagiwara S, Byerly L (1981) Calcium channel. Annu Rev Neurosci 4:69 –125. Hemara-Wahanui A, Berjukow S, Hope CI, Dearden PK, Wu SB, Wilson-Wheeler J, Sharp DM, Lundon-Treweek P, Clover GM, Hoda JC, Striessnig J, Marksteiner R, Hering S, Maw MA (2005) A CACNA1F mutation identified in an X-linked retinal disorder shifts the voltage dependence of Cav1.4 channel activation. Proc Natl Acad Sci U S A 102:7553–7558. Hoda JC, Zaghetto F, Koschak A, Striessnig J (2005) Congenital stationary night blindness type 2 mutations S229P, G369D, L1068P, and W1440X alter channel gating or functional expression of Ca. (v) 1.4 L-type Ca2⫹ channels. J Neurosci 25:252–259. Hoda JC, Zaghetto F, Singh A, Koschak A, Striessnig J (2006) Effects of congenital stationary night blindness type 2 mutations R508Q and L1364H on Cav1.4 L-type Ca2⫹ channel function and expression. J Neurochem 96:1648 –1658. Hohaus A, Beyl S, Kudrnac M, Berjukow S, Timin EN, Marksteiner R, Maw MA, Hering S (2005) Structural determinants of L-type channel activation in segment IIS6 revealed by a retinal disorder. J Biol Chem 280:38471–38477. Jacobi FK, Hamel CP, Arnaud B, Blin N, Broghammer M, Jacobi PC, Apfelstedt-Sylla E, Pusch CM (2003) A novel CACNA1F mutation in a French family with the incomplete type of X-linked congenital stationary night blindness. Am J Ophthalmol 135:733–736. Koschak A, Reimer D, Walter D, Hoda JC, Heinzle T, Grabner M, Striessnig J (2003) Cav1.4alpha1 subunits can form slowly inactivating dihydropyridine-sensitive L-type Ca2⫹ channels lacking Ca2⫹-dependent inactivation. J Neurosci 23:6041– 6049. Kotturi MF, Carlow DA, Lee JC, Ziltener HJ, Jefferies WA (2003) Identification and functional characterization of voltage-dependent

calcium channels in T lymphocytes. J Biol Chem 278:46949 – 46960. Lee KS, Tsien RW (1982) Reversal of current through calcium channels in dialysed single heart cells. Nature 297:498 –501. Lee KS, Tsien RW (1984) High selectivity of calcium channels in single dialysed heart cells of the guinea-pig. J Physiol 354:253–272. Lorenz B, Andrassi M, Miliczek KD (1996) [Incomplete congenital stationary night blindness (CSNB). An important differential diagnosis of congenital nystagmus.] Klin Monatsbl Augenheilkd 208:48 –55. McRory JE, Hamid J, Doering CJ, Garcia E, Parker R, Hamming K, Chen L, Hildebrand M, Beedle AM, Feldcamp L, Zamponi GW, Snutch TP (2004) The CACNA1F gene encodes an L-type calcium channel with unique biophysical properties and tissue distribution. J Neurosci 24:1707–1718. Miyake Y, Horiguchi M, Ota I, Shiroyama N (1987) Characteristic ERG-flicker anomaly in incomplete congenital stationary night blindness. Invest Ophthalmol Vis Sci 28:1816 –1823. Miyake Y, Yagasaki K, Horiguchi M, Kawase Y, Kanda T (1986) Congenital stationary night blindness with negative electroretinogram. A new classification. Arch Ophthalmol 104:1013–1020. Morgans CW (2000) Neurotransmitter release at ribbon synapses in the retina. Immunol Cell Biol 78:442– 446. Morgans CW, Bayley PR, Oesch NW, Ren G, Akileswaran L, Taylor WR (2005) Photoreceptor calcium channels: insight from night blindness. Vis Neurosci 22:561–568. Nakamura M, Ito S, Terasaki H, Miyake Y (2001) Novel CACNA1F mutations in Japanese patients with incomplete congenital stationary night blindness. Invest Ophthalmol Vis Sci 42:1610 –1616. Ruether K, Apfelstedt-Sylla E, Zrenner E (1993) Clinical findings in patients with congenital stationary night blindness of the SchubertBornschein type. Ger J Ophthalmol 2:429 – 435. Singh A, Hamedinger D, Hoda JC, Gebhart M, Koschak A, Romanin C, Striessnig J (2006) C-terminal modulator controls Ca2⫹-dependent gating of Ca. (v) 1.4 L-type Ca2⫹ channels. Nat Neurosci 9:1108 – 1116. Soldatov NM (1992) Molecular diversity of L-type Ca2⫹ channel transcripts in human fibroblasts. Proc Natl Acad Sci U S A 89: 4628 – 4632. Soldatov NM, Zhenochin S, AlBanna B, Abernethy DR, Morad M (2000) New molecular determinant for inactivation of the human L-type alpha1C Ca2⫹ channel. J Membr Biol 177:129 –135. Strom TM, Nyakatura G, Apfelstedt-Sylla E, Hellebrand H, Lorenz B, Weber BH, Wutz K, Gutwillinger N, Ruther K, Drescher B, Sauer C, Zrenner E, Meitinger T, Rosenthal A, Meindl A (1998) An L-type calcium-channel gene mutated in incomplete X-linked congenital stationary night blindness. Nat Genet 19:260 –263. Tremblay F, Laroche RG, De Becker I (1995) The electroretinographic diagnosis of the incomplete form of congenital stationary night blindness. Vision Res 35:2383–2393. Weleber RG (2002) Infantile and childhood retinal blindness: a molecular perspective (the Franceschetti lecture). Ophthalm Genet 23: 71–97. Witkovsky P, Schmitz Y, Akopian A, Krizaj D, Tranchina D (1997) Gain of rod to horizontal cell synaptic transfer: relation to glutamate release and a dihydropyridine-sensitive calcium current. J Neurosci 17:7297–7306. Wutz K, Sauer C, Zrenner E, Lorenz B, Alitalo T, Broghammer M, Hergersberg M, de la Chapelle A, Weber BH, Wissinger B, Meindl A, Pusch CM (2002) Thirty distinct CACNA1F mutations in 33 families with incomplete type of XLCSNB and Cacna1f expression profiling in mouse retina. Eur J Hum Genet 10:449 – 456. Yang J, Ellinor PT, Sather WA, Zhang JF, Tsien RW (1993) Molecular determinants of Ca2⫹ selectivity and ion permeation in L-type Ca2⫹ channels. Nature 366:158 –161. Zeitz C, Kloeckener-Gruissem B, Forster U, Kohl S, Magyar I, Wissinger B, Matyas G, Borruat FX, Schorderet DF, Zrenner E, Munier FL, Berger W (2006) Mutations in CABP4, the gene encoding the

J. B. Peloquin et al. / Neuroscience 150 (2007) 335–345 Ca2⫹-binding protein 4, cause autosomal recessive night blindness. Am J Hum Genet 79:657– 667. Zeitz C, Minotti R, Feil S, Matyas G, Cremers FP, Hoyng CB, Berger W (2005) Novel mutations in CACNA1F and NYX in Dutch families with X-linked congenital stationary night blindness. Mol Vis 11:179–183.

345

Zito I, Allen LE, Patel RJ, Meindl A, Bradshaw K, Yates JR, Bird AC, Erskine L, Cheetham ME, Webster AR, Poopalasundaram S, Moore AT, Trump D, Hardcastle AJ (2003) Mutations in the CACNA1F and NYX genes in British CSNBX families. Hum Mutat 21:169.

(Accepted 3 October 2007) (Available online 14 September 2007)