ROLE OF CLIC1 AND L-TYPE CALCIUM CHANNELS IN THE PATHOPHYSIOLOGY OF GLIOBLASTOMA AND VENTRICULAR ARRHYTHMIAS

Ion channels are transmembrane proteins that allow and control the flux of ions (sodium, potassium, calcium, and chloride) across the plasma membrane. They are present in all cell types and play critical roles in a variety of biological processes. Historically, ion channels have always been an attractive target for the treatment of different pathologies mainly because numerous drugs can specifically bind ion channels modifying their functional activity. My PhD thesis addresses the role of two different ion channels in the two leading causes of death in the modern society: heart disease and cancer. The first part of this PhD dissertation, developed at University of Milan under the mentorship of Prof. Michele Mazzanti, focuses on understanding the role of CLIC1 channels in glioblastoma cancer stem cells (CSCs). Glioblastoma is the most lethal among brain tumors. As other solid tumors, this cancer is composed of two cell types: a small population of cells able to self-renew and generate progeny (CSCs) and a larger population of differentiated cells (bulk cells). Glioblastomas are very aggressive tumors because of CSCs brain infiltration efficiency and resistance to chemotherapies. CLIC1 is a metamorphic protein mainly present as a soluble form in the cytoplasm that is able to translocate to the plasma membrane in response to oxidative stimuli where it acts as a Cl- channel. Several forms of glioblastomas show a high level of expression of CLIC1 compared to normal brains tissue. In electrophysiological experiments, overexpression of CLIC1 in murine CSCs were associated with a specific increase of the protein at the plasma membrane compared to normal stem cell (NSC). To study the relevance of CLIC1 we used CSCs isolated from human glioblastoma biopsies. By knocking down CLIC1 protein using siRNA viral infection (siCLIC1), we found that CLIC1-deficient cells proliferate less efficiently than control cells infected with siRNA for luciferase (siLUC). Since CLIC1 is a dimorphic protein we asked whether the reduction in proliferation was due to CLIC1 as ion channel. We performed perforated patches electrophysiological experiments for both siLUC and siCLIC1 cells. Cl- currents mediated by CLIC1 were isolated using IAA94, a CLIC1 ion channel inhibitor. The results showed that siCLIC1 cells did not display IAA94-sensitive currents, while siLUC cells presented the CLIC1-mediated chloride current. These findings strongly suggest that CLIC1 ion channel activity is required in the proliferation activity of CSCs, and therefore represents a promising target direct in the reduction of CSC gliomagenesis. To target CLIC1 ion channel, the only effective drug so far identified is IAA94 which seems to be rather specific but toxic. For this reason we sought non-toxic drugs that could interact with CLIC1 ion channel. Epidemiological and preclinical studies propose that Metformin, a first-line drug for type-2 diabetes, exerts direct antitumor activity specifically on CSCs. Although several clinical trials are ongoing, the molecular mechanisms of this effect are unknown. To study Metformin's effect on CLIC1-mediated current (isolated with IAA94), we performed electrophysiological experiments from perforated patches using a glioblastoma U87 cell line. We constructed a dose response curve comparing Metformin effect on CLIC1 maximum current (isolated using IAA94) from which we calculated EC50 2.1 ± 0.4 mM. To validate the specificity of Metformin for CLIC1, we compared the extent of block of this drug to the one of IAA94 by sequentially adding the two inhibitors with either order. If the second drug perfusion did not show additional block, we can confirm that the two inhibitor share the same target. Interestingly, experimental data show that Metformin-mediated inhibition of CLIC1 is similar to IAA94 block, suggesting that they both act on CLIC1. Metformin displays antiproliferative activity mainly acting on CSC and not on the differentiated cells. Is this phenotype the result of different expression of CLIC1 in plasma membranes? Interestingly, the relative abundance of CLIC1-mediated current in CSCs was about three fold bigger than in differentiated cells, suggesting that CLIC1 inhibition is relevant in the antiproliferative activity of Metformin. We sought to understand how Metformin bind the channel. Taking advantage of computational modeling and the available CLIC1 crystal structure, our collaborator predicted that the arginin 29 (R29) in the CLIC1 transmembrane domain may be part of the Metfomin binding site. We test this hypothesis in perforated patch clamp experiment using CHO cells stably transfected with CLIC1 wild type or CLIC1 R29A. The substitution of Arg29 in the putative CLIC1 pore region impairs Metformin modulation of channel activity. These results demonstrate that CLIC1 is required for human glioblastoma cell proliferation. Furthermore, we identified CLIC1 as direct target of Metformin antiproliferative activity in human glioblastoma cells. These findings are paving the way for novel and needed pharmacological approaches to glioblastoma treatment. The second part of this PhD dissertation, focusing on cardiac arrhythmia, was developed at University of California, Los Angeles under the mentorship of Prof. Riccardo Olcese. The electrical activity of the heart originates from the rhythmic activity in the sinoatrial node (SAN) and spreads across the heart as a wave of depolarization (the cardiac action potential). While the normal ventricular cardiac action potential (AP) repolarizes monotonically, returning to the diastolic membrane potential, under certain pathological conditions the repolarization can be interrupted by sudden depolarizations called early afterdepolarizations (EADs), occurring during phase 2 or phase 3 of the AP. These events, observable at the cellular and tissue level, are recognized triggers of cardiac arrhythmias. In fact, EADs can generate a new AP that propagates across the heart disrupting the propagation of normal AP wave leading to ventricular tachycardia (VT) and ventricular fibrillation (VF). Ventricular fibrillation is the most commonly identified arrhythmia in sudden cardiac death (SCD), one of the leading causes of death in the United States. This project investigated the relevance of the voltage gated L-type calcium channel (CaV1.2) in the etiology of EADs of the cardiac action potential. EADs are largely induced by the reactivation of L-type Ca2+ currents (ICa,L) that occurs at the range of membrane potential from -40 to 0 mV, called window current region. To study the dependence of EADs on the biophysical properties of L-type Ca2+ current (ICa,L) we adopted a hybrid biological–computational approach: the dynamic clamp technique. Under dynamic clamp it was possible to replace the native ICa,L of a ventricular myocyte with a computed ICa,L defined by programmable parameters. We previously identified three L-type Ca2+ channel (LTCC) biophysical parameters that effectively suppress EADs induced by oxidative stress or hypokalemia by preventing ICa,L reactivation in the window current region. Specifically, EADs were potently suppressed by: i) a ~5 mV depolarizing shift of the steady-state activation curve, ii) a ~5 mV hyperpolarizing shift of the steady-state inactivation curve or iii) a reduction of the non-inactivating pedestal component. Importantly, these changes did not significantly alter the peak ICa,L or Ca2+ transient amplitude during the action potential. Since LTCCs are multiprotein complexes in which CaVβ subunits modulate the gating properties and voltage dependence of the pore-forming CaV1.2 α1C subunit, we explored whether modifying LTCC β subunit composition is a suitable therapeutic strategy to suppress EADs. Voltage clamp experiments, in which we expressed Cav1.2 α1c with different β subunits, showed that subunit subtypes β2a and β2b, which are abundantly expressed in ventricular myocytes, give rise to LTCCs with voltage-dependent properties favoring EADs formation. Accordingly, we tested an adenovirus-based shRNA delivery strategy to reduce β2 expression in primary ventricular myocyte cultures; the rationale being that a LTCC population in a cell with a smaller proportion of β2a- and β2b-containing channels should generate ICa,L with an overall voltage dependence disfavorable to EADs emergence. The simultaneous partial knock down of β2a and β2b shifted the whole-cell ICa,L steady-state activation curve to more depolarized potentials by ~4 mV without significantly affecting peak ICa,L. A “narrower” window current could diminish the probability of EADs formation by preventing channel reopening. In congruence with the dynamic clamp results, EADs occurrence under oxidative stress (H2O2) was potently prevented in rabbit ventricular myocytes with β2 knock-down (no EADs observed). Conversely, control myocytes from the same batches exhibited significant action potential prolongation and all cells developed EADs after H2O2 exposure. These findings demonstrate that manipulation of the subunit composition can be an effective strategy for modifying the steady-state properties of ICa,L. Thus, our results highlight the use of genetic engineering as a therapeutic avenue for the treatment of EADs-related cardiac arrhythmias.