L-type Calcium Channel Blockers Enhance Trafficking and Function of Epilepsy-associated α1(D219N) Subunits of GABA(A) Receptors.

Page 1 of 52

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 α1(D219N) 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Figure

ER

Enhanced binding to lectin chaperones Ca 2+

Enhanced assembly

CCB

Ca2+

α1(D219N)

Channels

ERAD

Cytoplasm

Enhanced folding?

CANX

α1(D219N)

Misfolded domain

proteasome

N- linked glycan

ACS Paragon Plus Environment

Pentamer

Enhanced trafficking to plasma membrane


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 52

L-type Calcium Channel Blockers Enhance Trafficking and Function of Epilepsy-associated α1(D219N) Subunits of GABAA Receptors

Dong-Yun Han,1 Bo-Jhih Guan,2 Ya-Juan Wang,3 Maria Hatzoglou,2 Ting-Wei Mu 1,*

1

Department of Physiology and Biophysics,

2

Department of Pharmacology,

3

Center for

Proteomics and Bioinformatics and Department of Epidemiology and Biostatistics, Case Western Reserve University School of Medicine, Cleveland, Ohio 44106, USA.

1

To whom correspondence should be addressed.

Telephone: 216-368-0750; Fax: 216-368-5586; E-mail: [email protected]

1 ACS Paragon Plus Environment

Page 3 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ABSTRACT Gamma-aminobutyric acid type A (GABAA) receptors are the primary inhibitory ion channels in mammalian central nervous system and play an essential role in regulating inhibition-excitation balance in neural circuits. The α1 subunit harboring the D219N mutation of GABAA receptors was reported to be retained in the endoplasmic reticulum (ER) and traffic inefficiently to the plasma membrane, leading to loss of function of α1(D219N) subunits and thus idiopathic generalized epilepsy (IGE). We present the use of small molecule proteostasis regulators to enhance the forward trafficking of α1(D219N) subunits to restore their function. We showed that treatment with verapamil (4 µM, 24 h), an L-type calcium channel blocker, substantially increases the α1(D219N) subunit cell surface level in both HEK293 cells and neuronal SHSY5Y cells and remarkably restores the GABA-induced maximal chloride current in HEK293 cells expressing α1(D219N)β2γ2 receptors to a level that is comparable to wild type receptors. Our drug mechanism study revealed that verapamil treatment promotes the ER to Golgi trafficking of the α1(D219N) subunits post-translationally. To achieve that, verapamil treatment enhances the interaction between α1(D219N) subunit and β2 subunit and prevents the aggregation of the mutant protein by shifting the protein from the detergent-insoluble fractions to detergent-soluble fractions. By combining

35

S pulse-chase labeling and MG-132 inhibition

experiments, we demonstrated that verapamil treatment does not inhibit the ER-associated degradation of the α1(D219N) subunit. In addition, its effect does not involve a dynamin-1 dependent endocytosis. To gain further mechanistic insight, we showed that verapamil increases the interaction between the mutant protein and calnexin and calreticulin, two major lectin chaperones in the ER. Moreover, calnexin binding promotes the forward trafficking of the mutant subunit. Taken together, our data indicate that verapamil treatment enhances the 2 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

calnexin-assisted forward trafficking and subunit assembly, which leads to substantially enhanced functional surface expression of the mutant receptors. Since verapamil is an FDAapproved drug that crosses blood-brain barrier and has been used as an additional medication for some epilepsies, our findings suggest that verapamil holds great promise to be developed to ameliorate IGE resulting from α1(D219N) subunit trafficking deficiency.


Page 4 of 52

Page 5 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION In normal physiology, the proteome needs to maintain a delicate balance between protein synthesis, folding, trafficking, degradation and aggregation. Excessive degradation of a protein, often the result of the inheritance of a misfolding-prone mutation, causes loss of its presence in its final destination and thus functional deficiency and corresponding diseases (1-3). To correct such diseases, an emerging therapeutic strategy is to adapt the protein homeostasis (proteostasis) network using small molecules or biologicals to prevent the degradation and/or enhance the folding and/or trafficking, and thus restore the function of the mutant protein (4-6). Such small molecules or biologicals are named proteostasis regulators; examples include heat shock response activators (7), unfolded protein response activation (8, 9), histone deacetylase inhibitors (10), and calcium signaling regulators (11, 12). In other cases, aggregation or excess of a protein is associated with diseases; therefore, it would be desirable to promote the degradation of the target protein to ameliorate such disease phenotypes (13, 14). It appears that a proteostasis regulator can only work on a number of proteins, presumably because a protein utilizes a subset of proteostasis network components (15, 16), which can be specifically targeted by this proteostasis regulator. Idiopathic epilepsies have a strong genetic association with loss of function of γaminobutyric acid type A (GABAA) receptors (17-21). As the primary inhibitory neurotransmitter-gated ion channels in mammalian central nervous system (22), GABAA receptors provide most of the inhibitory tone to balance the tendency of excitatory neural circuits to induce hyperexcitability, and thus maintain the excitatory-inhibitory balance (23, 24). It was reported that numerous point mutations in GABAA receptors are associated with a variety of idiopathic epilepsies (18). We focus our study on trafficking deficiency mutations, which lead to



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

reduced cell surface expression and thus loss of function of GABAA receptors. Two prominent examples in the α1 subunit are the A322D mutation (25) and D219N mutation (26). The A322D mutation was first identified in an French Canadian family with autosomal dominant juvenile myoclonic epilepsy (ADJME) (25), and the D219N mutation was first found in an French Canadian family with idiopathic generalized epilepsy or febrile seizures (26). GABAA receptors are heteropentamers, belonging to the Cys-loop superfamily of ligandgated ion channels (27). There are at least 19 GABAA receptor subunit isoforms in human: six α subunits (α1-6), three β subunits (β1-3), three γ subunits (γ1-3), one δ subunit, one ε subunit, one θ subunit, one π subunit, and three ρ subunits (ρ1-3). The most common pentameric GABAA receptors in the human brain are composed of two α1, two β2, and one γ2 subunits. A recent crystal structure of human homopentameric β3 subunit (28) confirms the long-predicted topology of GABAA receptors: each subunit has four transmembrane (TM) (TM1-4) helices, a large extracellular (or the endoplasmic reticulum (ER) luminal) N-terminus, a large intracellular loop domain between TM3 and TM4 (ICD), and a short extracellular (or the ER luminal) Cterminus. The A322D mutation in TM3 results in the misfolding of the α1(A322D) subunit, leading to a rapid degradation of most of this mutant protein, mainly by the ER-associated degradation (ERAD) pathway (29). The D219N mutation in ER luminal domain (ERD) results in its ER retention and reduced trafficking to the cell surface to a less extent than the A322D mutation (26). Consequently, the A322D mutation generates much smaller GABA-induced chloride current than the D219N mutation in patch-clamping experiments, and both mutations lead to substantially reduced currents than WT receptors (26). To ameliorate the syndrome due to such α1 subunit mutations, it is crucial to restore the surface trafficking and function of the mutant receptors. 5 ACS Paragon Plus Environment

Page 6 of 52

Page 7 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


We hypothesize that we can use proteostasis regulators to enhance the surface trafficking of

α1

subunit

variants

to

restore

their

function.

Previously,

we

showed

that

suberanilohydroxamic acid (SAHA) and trichostatin A (TSA), both potent histone deacetylase inhibitors, restored proteostasis of α1(A322D) subunits partially by promoting the calnexin and BiP-assisted folding (30). In search of more proteostasis regulators to restore α1 subunit variant function, L-type calcium channel blockers (CCBs) came to our attention. Our prior studies demonstrated that several L-type CCBs restored proteostasis of misfolding-prone lysosomal enzymes, such as glucocerebrosidase harboring the N370S or L444P mutation, partially by enhancing calnexin-assisted folding, but such positive effects were not observed for rescuing ∆F508 CFTR (cystic fibrosis transmembrane conductance regulator) (11, 12). The effectiveness of L-type CCBs on trafficking-deficient ion channels with integral membrane components has not yet been established. We consider α1 subunit variants as a potentially novel target for L-type CCBs because both α1 subunit and glucocerebrosidase variants, but not ∆F508 CFTR, possess large ER luminal components, whose folding and/or trafficking could be facilitated by ER chaperones, such as calnexin. Here, we revealed that representative L-type CCBs, including verapamil, diltiazem, and nitrendipine, increased the total protein level of α1(D219N) subunits. We further focused on studying how verapamil enhanced the function of α1(D219N)β2γ2 GABAA receptors in HEK293 cells.

RESULTS AND DISCUSSION L-type CCBs increase the total protein level of α1(D219N) subunits of GABAA receptors To evaluate whether representative L-type CCBs increased the total protein level of misfolding-prone α1 subunits, we applied verapamil (4 µM), diltiazem (10 µM), or nitrendipine 6 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(10 µM) for 24 h to HEK293 cells stably expressing α1(D219N)β2γ2 GABAA receptors. They all increased the total protein level of α1(D219N) subunits (Figure 1A). In the following experiments, we chose to use verapamil as a representative L-type CCB, to study its effect on α1(D219N) subunits (see Figure 1B for verapamil’s chemical structure). To determine whether verapamil had a selective effect on mutant receptors, HEK293 cells expressing wild type (WT) α1β2γ2 or mutant α1(D219N)β2γ2 GABAA receptors were treated with verapamil (4 µM, 24 h). Immunoblotting analysis showed that drug treatment substantially increased the total protein level of mutant α1(D219N) subunit, but only slightly for WT α1 subunit (Figure 1C), confirming verapamil’s effect in increasing the steady-state protein level was more dramatic for mutant proteins. The effect of verapamil on WT α1 subunit was probably due to the inefficient ER to Golgi trafficking of this WT protein (31, 32). Consistent with this idea, GABA was reported to act as a pharmacological chaperone to enhance the inefficient forward trafficking of the WT receptors in HEK293 cells (33). Time-course study demonstrated that verapamil treatment (4 µM) increased the total α1(D219N) subunit protein level as early as 5 h, and the effect plateaued at 24 h (Figure 1D). The incubation period on the time scale of hours was necessary probably because verapamil’s effect required the folding of newly synthesized α1(D219N) subunits in the ER and their subsequent trafficking from the ER to the Golgi and to the plasma membrane. Dose-response analysis revealed that verapamil treatment (24 h) increased the total α1(D219N) subunit protein level starting at 0.5 µM and achieved its best efficacy at 4 µM (Figure 1E; see quantification in Figure 1F). Remarkably, the total α1(D219N) subunit protein level after verapamil treatment (2 µM to 10 µM) was comparable to the WT α1 subunit (Figure 1F). No apparent cell toxicity was observed for up to 10 µM verapamil treatment (24 h) according to trypan blue staining (data not 7 ACS Paragon Plus Environment

Page 8 of 52

Page 9 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


shown). We also examined verapamil’s effect on the misfolding-prone α1(A322D) subunit: verapamil treatment did not increase its total protein level (data not shown). The different effect of verapamil on the α1(D219N) subunit and α1(A322D) subunit could be due to that α1(A322D) subunits were degraded much faster than α1(D219N) subunits.

Verapamil treatment increases functional surface expression of the α1(D219N) subunit Because GABAA receptors need to reach the plasma membrane to function as a chloride channel in response to GABA, we evaluated the effect of verapamil treatment on the functional surface expression of the α1(D219N) subunit. To determine whether the α1(D219N) subunits reached the plasma membrane, we employed surface biotinylation assay to quantify cell surface α1 subunit level in HEK293 and human neuronal SH-SY5Y cells stably expressing α1(D219N)β2γ2 GABAA receptors. Remarkably, verapamil treatment (4 µM, 24 h) significantly increased the surface α1(D219N) subunit level 3.4-fold in HEK293 cells and 3.6-fold in SHSY5Y cells (Figures 2A and 2B, quantification shown on the bottom), indicating that verapamil treatment enhanced the proper trafficking of the α1(D219N) subunit to the plasma membrane. To investigate the functional consequence of this enhanced surface trafficking, we performed whole-cell patch-clamping experiments to record GABA-induced chloride currents. The D219N mutation in the α1 subunit reduces GABAA receptor function by decreasing the total surface expression of mature protein (26). As a result, the mutant α1(D219N)β2γ2 GABAA receptors generate weaker GABA-induced chloride ion currents compared to WT receptors. As revealed by whole-cell patch-clamping electrophysiology recording, the peak current was 184 ± 27 pA in response to GABA (3 mM) in HEK293 cells expressing α1(D219N)β2γ2 GABAA receptors, and 720 ± 134 pA in response to GABA (1 mM) in HEK293 cells expressing WT 8 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 52

α1β2γ2 GABAA receptors (Figure 2C, see Figure 2D for quantification), which is consistent with the literature (26). Strikingly, verapamil treatment (4 µM, 24 h) significantly increased the GABA-induced peak current to 653 ± 77 pA in HEK293 cells expressing α1(D219N)β2γ2 GABAA receptors (Figure 2C, see Figure 2D for quantification), amounting to 90% of the GABA-induced peak current in HEK293 cells expressing WT GABAA receptors, indicating that verapamil restored the function of epilepsy-associated α1(D219N)β2γ2 GABAA receptors to similar level to that for WT receptors. We further evaluated whether verapamil acted as a pharmacological chaperone to increase the surface expression of α1(D219N). Pharmacological chaperones are cell-permeant small molecules that bind directly to and stabilize the target proteins (34-36). To explore whether it bound α1 subunits to regulate their function, we directly applied verapamil (4 µM) into the external perfusion recording buffer for HEK293 cells expressing α1β2γ2 GABAA receptors for only 1 min during the whole-cell patch-clamping experiment. This operation did not induce chloride current (data not shown) or change the GABA-induced peak current (Figure S1), indicating that verapamil did not act as an agonist or antagonist for GABAA receptors. Therefore, verapamil does not bind directly to GABAA receptors, precluding its effect as a pharmacological chaperone. The above data indicate that verapamil treatment enables substantially more α1(D219N) subunits to reach the plasma membrane. As a result, the function of the α1(D219N) subunit is almost completely rescued by verapamil treatment considering the peak GABA-induced chloride current. Because verapamil crosses the blood-brain barrier (37), the use of verapamil to ameliorate α1(D219N)-associated epilepsy could be further developed as a promising therapeutic strategy. Intriguingly, verapamil has been used effectively as an additional 9 ACS Paragon Plus Environment

Page 11 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


medication for the treatment of severe myoclonic epilepsy in infancy, or Dravet syndrome, which is caused by mutations in the voltage-gated sodium channel neuronal type α1 subunit (SCN1A) (38). The proposed mechanisms include P-glycoprotein inhibition to overcome drug resistance and recovery of neuronal membrane equilibrium due to diminished Ca2+ entry into the cytosol. Although Dravet syndrome is different from idiopathic generalized epilepsy resulting from GABAA receptor trafficking deficiency, this case strongly strengthens the relevance of using verapamil to treat idiopathic generalized epilepsy.

Verapamil treatment enhances the ER-to-Golgi trafficking of the α1(D219N) subunit without an apparent effect on its ERAD or endocytosis To examine the molecular mechanism of verapamil in increasing α1(D219N) subunit protein level, we performed the following experiments to evaluate individual biogenesis steps of the α1(D219N) subunits. Quantitative RT-PCR analysis demonstrated that verapamil treatment (4 µM, 24 h) did not change the mRNA level of the α1(D219N) subunit significantly, confirming the post-transcriptional effect of verapamil treatment on the α1(D219N) (Figure 3A). This result is consistent with the effect of L-type CCBs in other ERAD substrates: the mRNA level of L444P glucocerebrosidase was not changed by diltiazem treatment (11). Moreover, polysome fractionation by sucrose density gradient centrifugation allows us to evaluate the translational efficiency of the α1(D219N) mRNA (39). Changes in the distribution of mRNAs on polyribosomes in response to different cues reflect a change in the efficiency of their translation. We therefore determined if verapamil treatment caused any changes in the distribution of the α1(D219N) mRNA on polyribosomes. We observed no change (Figure 3B) and concluded that



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 52

verapamil treatment did not change the translational efficiency of the α1(D219N) mRNA. Therefore, verapamil increased the α1(D219N) protein levels post-translationally. We next investigated whether verapamil inhibited the ERAD of the α1(D219N) subunit as a way to increase its total protein level. The D219N mutation resulted in the ER retention and reduced surface trafficking of the α1(D219N) subunit (26). Treatment with MG-132 (10 µM, 2 h), a potent proteasome inhibitor, increased the total α1(D219N) subunit in HEK293 cells stably expressing α1(D219N)β2γ2 receptors, indicating that this mutant subunit was degraded by the ERAD pathway (Figure 3C, cf. lane 3 to 1). In the presence of MG-132, verapamil treatment still significantly increased the total α1(D219N) protein level by 2.6-fold (Figure 3C, cf. lane 4 to 3), indicating that proteasome inhibition did not compromise verapamil’s effect. Therefore, verapamil treatment per se did not reduce the ERAD of the α1(D219N) subunit. Furthermore, we determined whether verapamil treatment decreased the degradation rate of newly synthesized α1(D219N) subunits using

35

S pulse-chase experiments. HEK293 cells stably expressing

α1(D219N)β2γ2 receptors were treated with verapamil (4 µM, 24 h) or DMSO vehicle control and then followed by chases for 0, 0.5, 1, 2, 3, 4 h. The cell lysates were immunoprecipitated using anti-α1 subunit antibody and detected by autoradiography. The half-life of α1(D219N) in DMSO vehicle control-treated cells was 1.6 h when fitted to a single exponential function, whereas its half-life in verapamil-treated cells was 1.7 h (Figure 3D). This finding indicates that verapamil treatment did not reduce the degradation rate or the ERAD of the α1(D219N) subunit. Because GABAA receptors are known to undergo dynamin-1-dependent endocytosis on the plasma membrane (40, 41), we next evaluated whether verapamil treatment inhibited this process as a way to elevate the plasma membrane expression of the α1(D219N) protein. Surface biotinylation assay was used to quantify cell surface α1 subunit level in HEK293 cells. In the 11 ACS Paragon Plus Environment

Page 13 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


presence of a specific, potent dynamin-1 inhibitor, dynole 34-2 (2.5 µM, 24 h) (42), verapamil treatment (4 µM, 24 h) still significantly increased the surface α1(D219N) protein level (Figure 3E), indicating that verapamil’s effect did not rely on dynamin-1-dependent endocytosis. We next determined whether verapamil treatment enhanced the ER-to-Golgi trafficking of the α1(D219N) subunit as a way to increase its total and surface protein level by carrying out endoglycosidase H (endo H) enzyme digestion assay (29, 30). This method monitors the movement of glycosylated proteins from the ER to the Golgi. The endo H enzyme selectively cleaves after asparaginyl-N-acetyl-D-glucosamine in the N-linked glycans incorporated on the α1 subunit in the ER, but after the glycans are further modified in the Golgi, endo H cannot remove the oligosaccharide chain from the α1 subunit. Therefore, endo H resistant α1 subunit bands represent properly folded, post-ER α1 subunit glycoforms that traffic at least to the Golgi compartment, whereas endo H sensitive α1 subunit bands represent immature α1 subunit glycoforms that are retained in the ER. The peptide-N-glycosidase F (PNGase F) enzyme cleaves between the innermost N-acetyl-D-glucosamine and asparagine residues from N-linked glycoproteins, serving as a control for unglycosylated α1 subunits (Figure 4A, lane 3). After endo H digestion, subunits with a molecular weight equal to unglycosylated α1 subunits were considered endo H-sensitive, whereas those with higher molecular weight were considered endo H-resistant (Figure 4A, lanes 2 and 5). There are two endo H-resistant bands because the α1 subunit has two glycosylation sites (Asn38 and Asn138) in the ER. Verapamil treatment (4 µM, 24 h) clearly increased the upper two endo H-resistant α1(D219N) subunit bands (Figure 4A, cf. lane 5 to 2), indicating that verapamil treatment increased properly folded, post-ER glycoforms of the α1(D219N) subunit. The ratio of endo H resistant/total α1(D219N) serves as a measure of trafficking efficiency of the α1(D219N) subunit. Verapamil treatment significantly increased this 12 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 52

ratio (Figure 4A, cf. lane 5 to 2, see the right panel for quantification), and thus the trafficking efficiency of the α1(D219N) subunit. Interestingly, we observed distinct effect of verapamil on the maturation of other important misfolding-prone ion channels. Verapamil (4 µM, 24 h) was applied to HEK293 cells stably expressing hERG (human Ether-à-go-go-Related Gene) channel variants (T65P or N470D) (43). Verapamil treatment did not increase total protein level or top mature band (~155 kD) of hERG channel variants (Figure S2). Also L-type CCBs did not enhance the maturation of ∆F508 CFTR (cystic fibrosis transmembrane conductance regulator) (12). These findings suggest that verapamil has certain level of selectivity among ion channels. One possible cause for that specificity is that GABAA receptors have different topology compared to hERG channels and CFTR: GABAA receptors have large defined ER lumen structures, which can be targeted by ER chaperones, whereas hERG channels and CFTR only have relatively short ER lumen components. In addition, the D219N mutation is located in the ER lumen. Consequently, verapamil’s effect could depend on the activity of ER chaperones (also see below).

Verapamil treatment enhances the interaction of the α1(D219N) subunit with the β2 subunit and shifts more mutant proteins into detergent-soluble fractions To determine how verapamil treatment increased the ER-to-Golgi transport of the α1(D219N) subunit, we first evaluated whether this operation enhanced the subunit assembly process because subunit assembly in the ER membrane is an important process in forming the heteropentamic GABAA receptors for their forward trafficking (44). When expressed alone, the α1 subunits are retained in the ER and rapidly degraded (32). Coexpression of β subunits with α subunits is indispensable for their assembly in the ER membrane and their subsequent trafficking


Page 15 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to the plasma membrane (45, 46), and addition of a γ subunit will further confer the sensitivity to benzodiazepines (47). Because the D219N mutation is located close to the α1-β2 subunit interface, we reasoned that verapamil treatment could potentially influence the interaction between α1 and β2 subunits as a way to enhance their assembly. Therefore, we carried out the following co-immunoprecipitation experiments to examine the inter-subunit interactions (48). Verapamil treatment significantly increased the associated β2 subunit that was pulled down by the α1(D219N) subunit in HEK293 cells (Figure 4B, cf. lane 6 to 5, see the right panel for quantification), indicating that verapamil promoted the interaction between the α1(D219N) subunit and β2 subunit, and thus the subunit assembly for subsequent forward trafficking. When only the α1(D219N) subunit was expressed in HEK293 cells, verapamil treatment did not increase its total protein level, whereas MG-132 treatment did (Figure S3), indicating that verapamil treatment did not prevent the ERAD of the α1(D219N) subunit, consistent with our results in Figures 3C and 3D. Instead, verapamil’s function depended on promoting the assembly of the mutant receptors. The α1(D219N) subunits are retained in the ER (26), and it has been reported that other mutant α1 subunits could form aggregates (49). Therefore, we reasoned that if verapamil influenced ER chaperones (see below), it could reduce the detergent-insoluble form of the α1(D219N) subunit in cells. We observed a defined α1(D219N) band using the 1% NP-40insoluble pellet fraction in SDS-PAGE (Figure 4C, lane 3). Clearly, verapamil treatment led to a drop of the mutant subunit in the detergent-insoluble fraction but an increase in the detergentsoluble fraction (Figure 4C), indicating that this operation shifted the mutant protein from an aggregation-prone state to a folding-prone state.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 52

Taken together, our findings demonstrated that verapamil treatment increases the population of functional surface expression of the α1(D219N) subunit post-translationally. To achieve that, it shifts the α1(D219N) subunit to detergent-soluble fractions and promotes the subunit assembly. As a result, properly folded and assembled α1(D219N) subunits exit the ER and traffic more efficiently to the plasma membrane.

Verapamil treatment enhances the interaction between calnexin/calreticulin and α1(D219N) subunits, and calnexin and calreticulin facilitate α1(D219N) subunit folding To gain further mechanistic insight, we asked how verapamil treatment influences the ER folding environment. The ER is a dynamic Ca2+ store. Verapamil presumably increases the ER lumen Ca2+ concentrations by directly inhibiting ER Ca2+ efflux by targeting ryanodine receptors on the ER membrane (50) and indirectly inhibiting L-type Ca2+ channels to antagonize ryanodine receptor-mediated Ca2+-induced Ca2+ release from the ER (11, 12). There are several Ca2+buffering chaperones in the ER, including calnexin, calreticulin, BiP and Grp94 (51). The lectinlike chaperones, calnexin and calreticulin, bind the N-glycans in glycoproteins to facilitate their post-translational folding in the ER (52, 53). Therefore, we examined whether verapamil administration regulated the expression and/or activity of Ca2+-binding chaperones in the ER. Verapamil treatment (4 µM, 24 h) did not alter the total protein level of all the tested Ca2+buffering chaperones in the ER (Figure 5A, cf. lane 2 to 1). Since the activity of chaperones might depend on their interactions with their substrates, we further tested whether verapamil influenced the binding between the α1(D219N) subunit and major ER chaperones. HEK293 cells expressing α1(D219N)β2γ2 GABAA receptors were treated with verapamil (VPM) (4 µM, 24 h) or DMSO vehicle control. Cells were then lysed and immunoprecipitated with an anti-α1 subunit 15 ACS Paragon Plus Environment

Page 17 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


antibody before being subjected to SDS-PAGE and Western blot analysis. Verapamil treatment significantly increased the ratio of immunoprecipitated calnexin/α1 by 1.6-fold and the ratio of immunoprecipitated calreticulin/α1 by 1.3-fold (Figure 5A, cf. lane 6 to 5, see Figure 5B for quantification), indicating that verapamil enhanced α1(D219N) subunit’s interaction with calnexin and calreticulin. In contrast, the binding of the mutant subunit with BiP was not changed, whereas that interaction with Grp94 was significantly reduced (Figures 5A and 5B). The role of Grp94 in regulating α1 subunit folding/degradation requires future investigation. Calnexin and calreticulin are both lectin-like Ca2+-binding chaperones in the ER although calnexin is an integral ER membrane protein, whereas calreticulin is an ER luminal protein. Transient overexpression of calnexin significantly increased the ratio of endo H-resistant/total α1 subunit by 3.5-fold in HEK293 cells expressing α1(D219N)β2γ2 GABAA receptors (Figure 5C, cf. lane 4 to 2, see the right panel for quantification), indicating that calnexin overexpression increased the trafficking efficiency of the α1(D219N) subunit. We observed much smaller effect of calreticulin overexpression in increasing the trafficking efficiency of the mutant protein (Figure 5D, see the right panel for quantification and Figure S4 for calreticulin overexpression confirmation), suggesting that calreticulin can only partially substitute the function of calnexin. Similar case was reported in the context of misfolding-prone L444P glucocerebrosidase (GC): overexpression of calnexin but not calreticulin increased the activity of L444P GC in Gaucher patient-derived fibroblasts (12). Therefore, we next focused on studying calnexin’s effect on promoting the α1(D219N) subunit folding. Because verapamil enhanced the interaction between calnexin and α1(D219N) subunit, we continued to test the hypothesis that this association enhanced the maturation of α1(D219N) subunits. We first examined whether N-glycosylation was required for the α1(D219N) folding. 16 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 52

Upon entering the ER, at two glycosylation sites (Asn38 and Asn138), the α1(D219N) subunit is attached with the core oligosaccharide Glc3Man9GlcNAc2 (Glc: glucose; Man: mannose; GlcNAc: N-acetylglucosamine). The outermost two glucoses are removed by glucosidase I and II, which generates the monoglucosylated oligosaccharide Glc1Man9GlcNAc2, a substrate for calnexin

(53).

The

N38Q/N138Q

mutations

significantly

decreased

the

ratio

of

immunoprecipitated calnexin / α1(D219N) in HEK293 cells (Figure 6A, cf. lane 7 to 6, see the right panel for quantification), indicating that loss of N-glycans in the α1(D219N) subunit decreased the interaction between calnexin and α1(D219N) subunit. The N38Q/N138Q double mutations in α1(D219N) subunit also resulted in a lower molecular weight band and reduced the total protein level for the α1(D219N) subunit (Figure 6A, cf. lane 2 to 1), indicating that glycosylation is crucial for α1(D219N) protein maturation. Moreover, verapamil treatment (4 µM, 24 h) did not increase the total protein level of the N38Q/N138Q mutant α1(D219N) subunit (Figure 6A, cf. lane 3 to 2) or change the ratio of immunoprecipitated calnexin / N38Q/N138Q mutant α1(D219N) subunit (Figure 6A, cf. lane 8 to lane 7, see the right panel for quantification), indicating that a glycan-binding activity of calnexin is critical in regulating verapamil’s function to promote α1(D219N) subunit folding. We next investigated whether decreasing the interaction between calnexin and α1(D219N) subunit reduced the maturation of α1(D219N) subunit. Castanospermine (CST), a glucosidase inhibitor, inhibited the production of monoglucosylated N-linked oligosaccharide and thus prevented an N-linked glycoprotein from entering the calnexin folding cycles (54). Indeed, CST administration decreased the ratio of immunoprecipitated calnexin / α1(D219N), indicating that this treatment reduced the association of calnexin with the α1(D219N) protein (Figure 6B, cf. lane 6 to 5). As a result, CST treatment decreased the total level (Figure 6C, cf. lane 3 to 1) as well as the endo-H resistant band (Figure 17 ACS Paragon Plus Environment

Page 19 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


6C, cf. lane 4 to 2) of the α1(D219N) protein. In contrast, CST’s such effects were eliminated when the N-glycosylation of the α1(D219N) was abolished with the N38Q/N138Q double mutations (Figure 6C, lanes 5 to 9). The CST results clarified that the association between calnexin and the α1(D219N) protein through the N-glycans positively regulated the maturation of α1(D219N) subunit.

Figure 7 illustrates a mechanism model for verapamil’s effect on promoting the functional surface expression of the α1(D219N) subunit. The D219N mutation in the ER luminal domain of the α1 subunit leads to its misfolding and subsequence ERAD. Treatment with L-type CCBs, such as verapamil, increases the ER luminal Ca2+ concentrations by antagonizing Ca2+ channels on the ER membrane, such as ryanodine receptors and inositol trisphosphate receptors, and/or inhibiting the Ca2+-induced Ca2+ release signaling pathway (11, 12). Verapamil administration enhances the interaction between the α1(D219N) subunit and Ca2+-sensing lectin chaperones, including calnexin and calreticulin, which bind N-glycans in the α1(D219N) subunit. Previously, we demonstrated that depleting Ca2+ using EGTA, a specific Ca2+ chelator, reduced the association between calnexin and mutant L444P glucocerebrosidase (12). Therefore, it is possible that the increased interaction between calnexin and the mutant α1(D219N) protein also results from the increased Ca2+ in the ER lumen afforded by verapamil treatment. Calcium signaling has been identified as an important factor in epileptogenesis, and elevated Ca2+ influx into neurons is believed to cause membrane hyperexcitability and increase epilepsy risks (55). In addition, potent T-type CCBs have been reported to suppress absence seizures in a Genetic Absence Epilepsy Rats from Strasbourg (GAERS) model (56). Therefore, it is tempering to postulate that the D219N mutation in the α1 subunit might lead to decreased Ca2+ concentration 18 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 52

in the ER lumen and thus verapamil treatment might correct such a potential Ca2+ imbalance. However, this possibility requires further validation. We further demonstrated that the interaction between calnexin and α1(D219N) protein positively regulates the forward trafficking of the mutant protein. The enhanced forward trafficking could come from the enhanced folding of the α1(D219N) protein and/or enhanced intersubunit assembly. Due to the prominent role of calnexin in assisting glycoprotein folding, our findings strongly support that verapamil treatment promotes the calnexin-assisted folding of the α1(D219N) protein although a direct demonstration of this conclusion requires further experiments. Our results revealed that verapamil treatment increased the interaction between α1(D219N) subunit and β2 subunit and thus promoted the subunit assembly. Consequently, verapamil enabled more α1(D219N) subunits to reach the plasma membrane and corrected the functional deficiency caused by the mutant protein. Therefore, our study provides a promising way to intervene the Ca2+ signaling pathway to restore function of misfolding-prone GABAA receptors and thus the excitatory-inhibitory balance in neural circuits. This strategy could be further developed as an antiepileptic treatment since many CCBs, such as verapamil, cross the blood-brain barrier. It is desirable to test the in vivo effect of verapamil in an α1(D219N) knockin mouse model in future studies.

METHODS Reagents Verapamil (VPM), diltiazem, nitrendipine, and MG-132 were obtained from Sigma-Aldrich, castanospermine from Cayman Chemical, and dynole 34-2 from Tocris Bioscience. The pCMV6 plasmids containing human GABAA receptor α1, β2 (isoform 2), and γ2 (isoform 2) subunits, 19 ACS Paragon Plus Environment

Page 21 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


human KCNH2 plasmid and pCMV6 Entry Vector plasmid (pCMV6-EV) were obtained from Origene. The GFP plasmid was obtained from Addgene. The pCR(calreticulin)-HA plasmid was kindly provided by Professor Tohru Mizushima (Kumamoto University), and the Apr-M8-CNX plasmid by Professor Michael Brenner (Harvard Medical School). The human GABAA receptor α1 subunit missense mutation D219N, A322D or N38Q/N138Q and human KCNH2 missense mutation T65P or N470D were constructed using QuickChange II site-directed mutagenesis Kit (Agilent Genomics), and the cDNA sequences were confirmed by DNA sequencing. The mouse monoclonal anti-α1 (clone BD24) antibody was obtained from Millipore. The mouse monoclonal anti-FLAG M2 and anti-β-actin antibodies came from Sigma-Aldrich. The rabbit polyclonal anti-calnexin, mouse monoclonal anti-calreticulin, and rat monoclonal anti-Grp94 antibodies were obtained from Enzo Life Sciences. The rabbit monoclonal anti-sodium potassium ATPase and rabbit monoclonal anti-BiP antibodies came from Abcam. The rabbit polyclonal anti-HA tag antibody was obtained from Abgent. Cell Culture and Transfection HEK293 cells and SH-SY5Y cells came from ATCC and were maintained in Dulbecco’s Modified Eagle Medium (DMEM) (Hyclone) with 10% heat-inactivated fetal bovine serum (Sigma-Aldrich) and 1% Pen-Strep (Hyclone) at 37°C in 5% CO2. Monolayers were passaged upon reaching confluency with TrypLE Express (Invitrogen). Cells were grown in 6-well plates or 10-cm dishes and allowed to reach ~70% confluency before transient transfection using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instruction. Stable cell lines for α1β2γ2, α1(D219N)β2γ2 and α1(A322D)β2γ2 receptors were generated using the G-418 selection method. Briefly, cells were transfected with α1:β2:γ2 (1:1:1), α1(D219N):β2:γ2 (1:1:1), and α1(A322D):β2:γ2 (1:1:1) plasmids, and then maintained in DMEM supplemented



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 52

with 0.8 mg/mL G418 (Enzo Life Sciences) for 15 days. G-418 resistant cells were selected for follow-up experiments. Trypan blue stain (Hyclone) was used to evaluate cell viability. Western Blot Analysis Cells were harvested with TrypLE Express and then lysed with lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, and 1% Triton X-100) supplemented with Roche complete protease inhibitor cocktail. Lysates were cleared by centrifugation (16,000 × g, 10 min, 4 °C). Protein concentration was determined by MicroBCA assay (Pierce). Endoglycosidase H (endo H) and Peptide-N-Glycosidase F (PNGase F) (New England Biolabs) digestion was performed according to published procedure (29, 30). Aliquots of cell lysates were separated in an 8% SDSPAGE gel, and Western blot analysis was performed using the appropriate antibodies. Band intensity was quantified using Image J software from the NIH. Biotinylation of Cell Surface Proteins HEK293 cells and human neuronal SH-SY5Y cells stably overexpressing α1(D219N)β2γ2 receptors were plated in 10-cm dishes for surface biotinylation experiments. Intact cells were washed twice with ice-cold PBS and incubated with the membrane-impermeable biotinylation reagent Sulfo-NHS SS-Biotin (0.5 mg ⁄ mL; Pierce) in PBS containing 0.1 mM CaCl2 and 1 mM MgCl2 (PBS+CM) for 30 min at 4 °C to label surface membrane proteins. To quench the reaction, cells were incubated with 10 mM glycine in ice-cold PBS+CM twice for 5 min at 4 °C. Sulfhydryl groups were blocked by incubating the cells with 5 nM N-ethylmaleimide (NEM) in PBS for 15 min at room temperature. Cells were solubilized for 1 h at 4 °C in lysis buffer (Triton X-100, 1%; Tris-HCl, 50 mM; NaCl, 150 mM; and EDTA, 5 mM; pH 7.5) supplemented with Roche complete protease inhibitor cocktail and 5 mM NEM. The lysates were cleared by centrifugation (16,000 × g, 10 min at 4 °C) to pellet cellular debris. The supernatant contained


Page 23 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the biotinylated surface proteins. The concentration of the supernatant was measured using microBCA assay (Pierce). Biotinylated surface proteins were affinity-purified from the above supernatant by incubating for 1 h at 4 °C with 100 µL of immobilized neutravidin-conjugated agarose bead slurry (Pierce). The samples were then subjected to centrifugation (16,000 ×g, 10 min, at 4 °C). The beads were washed six times with buffer (Triton X-100, 0.5%; Tris–HCl, 50 mM; NaCl, 150 mM; and EDTA, 5 mM; pH 7.5). Surface proteins were eluted from beads by boiling for 5 min with 200 µL of LSB ⁄ Urea buffer (2x Laemmli sample buffer (LSB) with 100 mM DTT and 6 M urea; pH 6.8) for SDS-PAGE and Western blotting analysis. Whole-Cell Patch Clamp Electrophysiology Recording Whole-cell currents were recorded 48 h post transfection using HEK293 cells. The glass electrodes were pulled from thin-walled borosilicate capillary glass (Kimble-Chase) and firepolished on a DMZ Universal puller (Zeitz Instruments), having a tip resistance of 3-5 MΩ. The internal solution contains 153 mM KCl, 1 mM MgCl2, 5 mM EGTA, 10 mM HEPES, and 2 mM MgATP (pH 7.3). The external solution contains 142 mM NaCl, 8 mM KCl, 6 mM MgCl2, 1 mM CaCl2, 10 mM glucose, 10 mM HEPES, and 120 nM Fenvalerate (pH 7.4). Coverslips containing HEK293 cells were placed in a RC-25 recording chamber (Warner Instruments) on the stage of an Olympus IX-71 inverted fluorescence microscope and perfused with external solution. Fast GABA application was accomplished with a pressure-controlled perfusion system (Warner Instruments) positioned within 50 µm of the cell utilizing a Quartz MicroManifold with 100-µm inner diameter inlet tubes (ALA Scientific). The whole cell GABA-induced currents were recorded at a holding potential of -60 mV in voltage clamp mode using an Axopatch 200B amplifier (Molecular Devices). The signals were filtered at 2 kHz and detected at 10 kHz using pClamp10 acquisition software.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 52

Quantitative RT-PCR The relative expression levels of target genes were analyzed using quantitative RT-PCR according to published procedure (30). Briefly, the cells were incubated with drugs at 37 °C for the indicated amount of time before total RNA was extracted from the cells using RNeasy Mini Kit (Qiagen #74104). cDNA was synthesized from 500 ng of total RNA using QuantiTect Reverse Transcription Kit (Qiagen #205311). Quantitative PCR reactions (45 cycles of 15 s at 94°C, 30 s at 57°C, and 30 s at 72°C) were performed using cDNA, QuantiTect SYBR Green PCR Kit (Qiagen #204143) and corresponding primers in the StepOnePlus system (Applied Biosystems) and analyzed using StepOne v2.2 software (Applied Biosystems). The forward and reverse

primers

for

GABRA1

are

5'-GTCACCAGTTTCGGACCCG-3'

and

5'-

AACCGGAGGACTGTCATAGGT-3'; the forward and reverse primers for GAPDH (housekeeping

gene

control)

are

5'-GTCGGAGTCAACGGATT-3'

and

5'-

AAGCTTCCCGTTCTCAG-3'. Threshold cycle (CT) was extracted from the PCR amplification plot, and the ∆CT value was defined as: ∆CT = CT (target gene) - CT (housekeeping gene). The relative mRNA expression level of target genes of drug-treated cells was normalized to that of untreated cells: Relative mRNA expression level = 2 exp [- (∆CT (treated cells) - ∆CT (untreated cells))]. Each data point was evaluated in triplicate and measured three times. Polysome Profile Analysis and mRNA Distribution Polysome profiling and mRNA distribution analysis was described previously (39). Briefly, cells were seeded in 150 mm culture dishes and grown up to 70% confluence (~1.5 × 107 cells). Following the DMSO or VPM treatment for 24 h, cycloheximide (CHX) was added to cells in 100 µg/mL for 10 min at 37°C. Cells were collected at 4,000 rpm for 10 min and washed twice with cold PBS containing 100 µg/mL CHX. The cell pellets were suspended in 500 µL of lysis 23 ACS Paragon Plus Environment

Page 25 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


buffer (10 mM HEPES-KOH (pH 7.5), 2.5 mM MgCl2, 100 mM KCl, 0.25% NP-40, 100 µg/mL CHX, 1 mM DTT, 200 unit/mL RNase inhibitor (RNaseOUT, Invitrogen) and EDTA-free protease inhibitor (Roche Applied Science)), kept on ice for 20 min and then passed 15 times through a 23-gauge needle. Lysates were cleared at 14,000 rpm for 15 min and supernatants were collected and measured in absorbance of 260 nm. Twelve ODs of lysates were layered over 1550% of cold sucrose gradients in buffer (10 mM HEPES-KOH (pH 7.5), 2.5 mM MgCl2, 100 mM KCl). Gradients were centrifuged at 17,000 rpm in a Beckman SW28 rotor for 13.5 h at 4°C. After centrifugation, 12 fractions (1.2 mL/fraction) were collected. RNA from each fraction was isolated using TRIzol LS reagent (Life Technologies) and equal volume of RNA from each fraction was used for cDNA synthesis. The relative quantities of α1(D219N) mRNA were measured by quantitative RT-PCR. Non-polysome fractions (#1-6) and polysome fractions (#7-12) were pooled and plotted. Pulse-Chase Labeling of Cells with [35S]Methionine/Cysteine For measurement of newly synthesized α1(D219N) protein half-life, pulse-chase labeling was analyzed as described previously (57). Briefly, cells grown in 100 mm culture dishes up to 70% confluence were incubated for 1 h in label medium (Met/Cys-free DMEM supplemented with 10% dialyzed FBS). Cells were then pulse-labeled 1 h with 0.1 mCi/ml (Express [35S] protein labeling mix, PerkinElmer Life Sciences) in the label medium. For chasing, cells were washed twice with cold chase medium (label medium with 3 mM L-Met and 1 mM L-Cys) and then incubated for the indicated time at 37°C. Throughout the pulse-chase process, DMSO or VPM was included in all media. Cells were then washed twice in cold PBS buffer and protein extracted in lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% NP-40, 1% Triton X-100) supplemented with protease inhibitor (Roche Applied Science). For



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 52

immunoprecipitation, lysates (1 mg of protein) were pre-cleared followed by overnight incubation at 4 °C with anti-α1 antibody (2.5 µg, clone BD24, Millipore) and Dynabeads Protein G (Life Technologies). Samples were separated by 8% SDS-PAGE, dried, and analyzed with autoradiography. The intensity of specific α1 band was scanned and quantified by ImageJ software. The half-life was calculated by non-linear regression. Immunoprecipitation Cell lysates (500 µg) were pre-cleared with 30 µL of protein A/G plus-agarose beads (Santa Cruz) and 1.0 µg of normal mouse IgG for 1 hour at 4°C to remove nonspecific binding proteins. The pre-cleared cell lysates were incubated with 2.0 µg of mouse anti-α1 antibody (clone BD24, Millipore) or normal mouse IgG (negative control for nonspecific binding) for 1 hour at 4°C, and then with 30 µL of protein A/G plus agarose beads overnight at 4°C. The beads were collected by centrifugation at 8000 ×g for 30 s, and washed three times with lysis buffer. The α1 subunit complex was eluted by incubation with 30 µL of SDS loading buffer in the presence of DTT. The immunopurified eluents were separated in 8% SDS-PAGE gel, and Western blot analysis was performed. Statistical analysis All data are presented as mean ± SEM, and any statistical significance was calculated using twotailed Student’s t-Test.

SUPPORTING INFORMATION Supporting information includes four Supplemental figures. This material is available free of charge via the Internet at http://pubs.acs.org.


Page 27 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ACKNOWLEDGEMENTS This work was supported by the Research Startup Fund from Case Western Reserve University School of Medicine and Epilepsy Foundation of America 225243 (to T. Mu), NIH T32 HL007567 (to Y. Wang), and NIH R01 DK53307 (to M. Hatzoglou).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 52

FIGURE LEGENDS Figure 1. Verapamil treatment increases the total protein level of mutant α1(D219N) subunits in HEK293 cells. (A) Verapamil (4 µM, 24h), diltiazem (10 µM, 24h), and nitrendipine (10 µM, 24h) increase the protein level of α1(D219N) subunits in HEK293 cells stably expressing α1(D219N)β2γ2 GABAA receptors (n = 2 biological replicates). (B) The chemical structure of verapamil (VPM). (C) Verapamil treatment (4 µM, 24h) increases the total protein levels of α1 subunit variants in HEK293 cells expressing WT α1β2γ2 or α1(D219N)β2γ2 GABAA receptors (n = 3 biological replicates). (D) Time-course study of verapamil’s effect (4 µM) on total α1(D219N) subunits (n = 2 biological replicates). (E) Dose-response analysis of verapamil’s effect (24 h treatment) on total α1(D219N) subunits (n = 3 biological replicates). (F) Quantification of the α1(D219N) subunit intensity in (E) is shown. The WT α1 subunit quantification data is from (C). β-actin serves as a loading control. Each data point is reported as mean ± SEM. * P < 0.05, ** P < 0.01 compared with α1(D219N) subunits in the presence of DMSO vehicle control. WT: wild type; IB: immunoblotting.

Figure 2. Verapamil treatment increases the functional surface expression of the α1(D219N) subunit. (A & B) Verapamil (VPM) treatment (4 µM, 24 h) increases the surface α1(D219N) subunit in HEK293 cells (A) and human neuronal SH-SY5Y cells (B) stably expressing α1(D219N)β2γ2 27 ACS Paragon Plus Environment

Page 29 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


GABAA receptors using cell surface biotinylation assay (n = 3 biological replicates). The intensity of the surface α1(D219N) protein was quantified using Image J software, normalized to the DMSO vehicle control treatment, and shown below. Na+/K+ ATPase α chain serves as a loading control for biotinylated membrane proteins. IB, immunoblotting. (C & D) The GABA-induced chloride currents in HEK293 cells expressing WT or α1(D219N)β2γ2 GABAA receptors with or without verapamil treatment (4 µM, 24 h) were recorded using whole-cell patch-clamping setup with a holding potential of -60 mV. Representative traces are shown in (C), and quantification of the peak currents (Imax) is shown in (D). The number of patched cells in each group is shown at the top of the bar. pA, picoampere. Each data point in (A), (B), and (D) is reported as mean ± SEM. **p < 0.01.

Figure 3. Verapamil treatment has no apparent effects on transcription, translation, or degradation of the α1(D219N) subunit in HEK293 cells. (A) Verapamil (VPM) treatment (4 µM, 24 h) has no significant effects on the mRNA level of the α1(D219N) subunit using quantitative RT-PCR analysis in HEK293 cells expressing α1(D219N)β2γ2 GABAA receptors. The experiments were done using three biological replicates with triplicate each time. (B) The protein translation efficiency of α1(D219N) mRNA determined by polysome profile is not changed with VPM treatment. HEK 293 cells expressing α1(D219N)β2γ2 GABAA receptors were treated with DMSO or VPM (4 µM) for 24 h, and the distribution of α1(D219N) mRNA with polysomes was analyzed by polysome profile in 15-50% sucrose density. Equal volume of each fraction was used for RNA extraction and then reverse transcription. The fractions #1-6 (8

2

50%

15%

3

4

5 6 7 >8

Sucrose density

50%

RNA quality

1.0

1 2 3 4 5 6 7 8 9 10 11 12

α1(D219N) mRNA Distribution (%)

80

0.5

0.0

DMSO

D219N

VPM

VPM MG-132

− −

+ −

− +

E

+ +

DMSO

NS

VPM

NS

60 40 20 0

#1-6 #7-12 Polysome fractions

Surface + −

+ +

IB: α1 IB: Na+/K+ ATPase *

600

*

400

* 200

DMSO

300 200 100 0

VPM MG-132 MG-132 +VPM

DMSO

D219N Chase time (h)

0

0.5

1

2

3

4

0

0.5 1

2

Remaining of S35[Met/Cys]labeled α1(D219N) (%)

DMSO VPM Expon. (DMSO) Expon. (VPM)

100

0

1

Dynole

VPM

IB: α1

10

**

400

Surface α1(D219N) Quantity

*

0

1 2 3 4 5 6 7 8 9 10 11 12

D219N Dynole VPM

IB: β-actin Normalized α1 Intensity (% of α1(D219N) + DMSO)

Polysome

40S 60S

Sucrose density

IB: α1

D

VPM 24h 80S

80S

O.D. (254 nm)

D219N Normalized α1(D219N) mRNA Level (fold change to DMSO)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

A

Page 48 of 52

2 3 Chase time (h)


Figure 3

4

3

4

Dynole + VPM

Page 49 of 52

D219N

DMSO Buffer

EndoH

Ratio of the EndoH Resistant / Total α1(D219N)

A VPM Pf

Buffer

EndoH

}

IB: α1

EndoH Resistant EndoH Sensitive

IB: β-actin

*

0.3 0.2 0.1 0

DMSO

VPM

HEK

D219N

D219N

−

−

+

−

−

+

IB: α1 IB: FLAG IB: β-actin

D219N VPM

2

1

0

Sol −

Insol +

−

**

3

D219N

VPM

IP: α1

D219N

Input

C

0.4

α1(D219N)β2FLAGγ2

Normalized β2 / α1 Post IP

B

HEK

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


+

IB: α1 IB: β-actin

Figure 4


DMSO

VPM

A D219N

Input

IgG

Page 50 of 52

IP: α1 2.0

DMSO VPM DMSO VPM DMSO VPM IB: CANX IB: CRT IB: BiP IB: Grp94 IB: α1

Ratio of ER Chaperones to α1 (fold change to DMSO) post IP

*

1.5

NS

1.0

0.5

CANX/α1 CRT/α1

GFP

−

CANX

+

−

+

IB: CANX

}

IB: α1 IB: β-actin

EndoH Resistant EndoH Sensitive


Plasmid

D

BiP/α1

Grp94/α1

CANX/α1

D219N

EndoH

**

0.0

IB: β-actin

C

DMSO VPM

*

**

0.8 0.6 0.4 0.2 0.0

Plasmid

GFP

CANX

D219N Plasmid EndoH

IB: α1

GFP

−

CRT

+

−

+ EndoH Resistant EndoH Sensitive

IB: β-actin


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

B


0.4 0.3 0.2 0.1 0.0

Plasmid

Figure 5


*

GFP

CRT

Page 51 of 52

−

D219N

D219N-N38Q -N138Q

+

+

−

+

−

*

1.2 Ratio of CANX / α1 post IP (fold change to D219N)

−

D219N-N38Q -N138Q

−

IP: α1 D219N-N38Q -N138Q

VPM

IgG D219N-N38Q -N138Q

Plasmid

D219N-N38Q -N138Q

Input

D219N-N38Q -N138Q

A D219N

IB: α1 IB: CANX

NS

0.9 0.6 0.3 0.0 D219N

IB: β-actin Lane

B

1

2

D219N

3

4

Input

CST

−

5

IgG

+

−

+

6

7

8

IP: α1 −

+

IB: α1 IB: CANX IB: β-actin

1.5

D219N CST

1.0

0.5

0.0

D219N-N38Q-N138Q

−

−

+

+

−

−

+

+

+

Ctl

Eh

Ctl

Eh

Ctl

Eh

Ctl

Eh

Pf

1

2

3

4

5

6

7

8

9

IB: α1

IB: β-actin Lane


Figure 6

−

*

DMSO

C

D219N-N38Q D219N-N38Q -N138Q -N138Q

−

VPM

Ratio of CANX / α1 post IP (fold change to DMSO)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CST

+


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ER α1(D219N)

Enhanced Enhanced folding? α1(D219N) assembly Pentamer

Enhanced binding to lectin chaperones Ca2+

α1(D219N) CCB

Ca2+ Channels ERAD

Cytoplasm

Page 52 of 52

CANX Misfolded domain

proteasome

N-linked glycan

Figure 7


Enhanced trafficking to plasma membrane

Effects of calcium channel blockers on renal function.

Effects of calcium channel blockers on renal function in mice.

Calcium-channel blockers and anaesthesia.

Discovery and Development of Calcium Channel Blockers.

Regulation of high-voltage-activated Ca2+ channel function, trafficking, and membrane stability by auxiliary subunits.

Antiatherosclerotic effects of calcium channel blockers.

Calcium channel blockers and excitatory amino acids.

Calcium channel blockers and gingival overgrowth.

Calcium channel blockers and congestive heart failure.

Calcium channel blockers in cardiovascular pharmacotherapy.

Calcium channel blockers in myocardial infarction.

The effect of subunit composition of rat brain GABAA receptors on channel function.

Calcium channel blockers for primary Raynaud's phenomenon.

Effect of calcium-channel blockers on renal function in renal-graft recipients treated with cyclosporine.

Comparative Effect of Calcium Channel Blockers on Glomerular Function in Hypertensive Patients with Diabetes Mellitus.

Calcium-Channel Blockers Attenuate the Antiplatelet Effect of Clopidogrel.

Natriuretic effect of calcium-channel blockers in hypertensives.

Antiatherogenic effects of calcium-channel blockers: possible mechanisms of action.

Calcium channel blockers for inhibiting preterm labour and birth.

Pharmacologic management of ischemic heart disease with beta-blockers and calcium channel blockers.

Effect of calcium channel blockers on intraocular pressure.

Assembly, trafficking and function of α1β2γ2 GABAA receptors are regulated by N-terminal regions, in a subunit-specific manner.

Interaction between ethanol and calcium channel blockers in humans.

Calcium blockers enhance cisplatin-induced nephrotoxicity in rats.