The Role of Chloride Channels in Regulation of Pulmonary Artery Smooth Muscle Cell Proliferation

Wenbin Liang

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Physiology University of Toronto

Wenbin Liang

Doctor of Philosophy

Department of Physiology University of Toronto

2011 Abstract

Pulmonary arterial hypertension (PAH) is a rare but fatal disease with an annual mortality rate of

15% despite current therapies. Uncontrolled proliferation of pulmonary artery smooth muscle cells (PASMCs) results in adverse vascular remodeling contributing to PAH. Understanding the mechanisms of PASMC proliferation may identify new targets for treatment. Chloride currents/channels (ICl) are expressed in PASMCs and their roles in proliferation have been suggested based on their importance in resting membrane potential and cell volume regulation.

The present study explored the role of ICl in proliferation in rat and human PASMCs. We found that either nonspecific ICl inhibitors (DIDS or NPPB) or a putative specific blocker of swelling- activated ICl (ICl,swell) reduced proliferation of PASMCs cultured in serum-containing media.

Patch-clamp studies showed that proliferating PASMCs had increased baseline ICl and ICl,swell in association with depolarized membrane potentials. Quantitative real-time RT-PCR studies identified expressions of CLC-3, a candidate gene of ICl,swell, and several other CLC genes in proliferating PASMCs. While selective knockdown of CLC-3 with lentiviral shRNA reduced

PASMC proliferation, it had no effect on ICl,swell. These findings are consistent with the conclusion that ICl regulate proliferation of PASMCs and suggest that selective ICl inhibition may be useful in treating pulmonary arterial hypertension. ii

Acknowledgments

It has been a great experience to study in the Ph.D program at the Department of Physiology. The training is a critical part of my life and I am very proud of it. I am indebted to countless people who taught me many things that deeply affect my research and life.

My supervisor Dr. Peter Backx is a knowledgeable scientist and inspirational mentor. His enthusiasm in ion channels has a profound influence on this work and my research career. This thesis is dedicated to my co-supervisor Dr. Michael Ward (1956~2009), an eminent scientist and extraordinary mentor, for showing me the joy of doing science. The guidance from my supervisory committee members, Drs. Christine Bear, Steffen-Sebastian Bolz and Scott Heximer, helped me overcoming many hurdles in this work. I would also like to thank my former advisors, Drs. Rui Wang, Lingyun (Lily) Wu and Shizhong Jiang, for introducing me to the exciting field of cardiovascular research. I also thank Dr. Zhong-Ping Feng for help with coursework and Ms. Rosalie Pang, Sandra Monkewich and Dr. Marc Perry for administrative supports.

It was my pleasure to know so many great people during my study in the Backx and Ward laboratories. Their knowledge and kindness contributed a lot to my research work. I would like to take this opportunity to thank Drs. B-G. Kerfant, R. A. Rose, R. G. Tsushima, G. Y. Oudit, H. Sun, K-H Kim, N. D'Avanzo, R. Pekhletski, N. Gong, D. Zhao, M. Cieslak, M. M. Patel, M. G. Kabir, M. G. Trivieri, P. C. Papageorgiou, R. D. Vanderlaan, I. Lorenzen-Schmidt, D. Sednev, K. Ban, A. J. Cooper, J. B. Ray, J. Z. He, P. Plant, S. Beca, P. B. Helli, R. Sobbi, J. Simpson, G. P. Farman, J. Liu, Y. Wang, L. Huang, N. Bousette, Y. Fedyshyn, T. Ketela, A. Rosen, M. Mirkhani, B. K. Panama, W. Yang, M. Sellan, and F. Izaddoustdar. I sincerely apologize for those people who helped me but their names are not mentioned here.

Finally, I want to thank my family for their encouragements and continuous supports during my study. Without their supports and understanding, I would never finish this work.

iii

Table of Contents

Acknowledgments ...... iii

Table of Contents ...... iv

List of Tables ...... vii

List of Figures ...... viii

List of Appendices ...... x

Chapter 1 General Introduction ...... 1

1 Introduction ...... 2

1.1 Pulmonary Arterial Hypertension (PAH) ...... 2

1.1.1 PAH: A Disease of Pulmonary Vasculature ...... 2

1.1.2 Hypoxia and PAH ...... 4

1.1.3 Phenotypes of Vascular Smooth Muscle Cells ...... 5

1.2 Cellular Proliferation ...... 6

1.2.1 Cell Cycle ...... 6

1.2.2 Cell Volume Regulation and its Role in Proliferation ...... 8

1.2.3 Membrane Potential and Ca2+ in Proliferation ...... 10

1.3 Mammalian Cl- Channels ...... 11

1.3.1 Cl- Equilibrium and Cl- Channels in Mammalian Tissues ...... 11

1.3.2 Swelling-Activated Cl- Currents ...... 13

1.3.3 Ca2+-Activated Cl- Currents ...... 16

1.3.4 CLC Family ...... 19

1.3.5 Cystic Fibrosis Transmembrane Conductor Regulator (CFTR) ...... 26

1.3.6 Bestrophin Family ...... 27

1.3.7 TMEM16 Family ...... 27

1.3.8 Ligand-Gated Cl- Channels ...... 28

1.4 Aims of Present Studies ...... 28

1.5 RNA Interference ...... 29

1.5.1 siRNA-Mediated Gene Silencing ...... 29

1.5.2 siRNA Duplex Designing ...... 31

1.5.3 Validation of Knockdown ...... 36

Chapter 2 Materials and Methods ...... 38

2 Materials and Methods ...... 39

2.1 Animals ...... 39

2.2 Detection of PASMC Proliferation by BrdU Uptake In Vivo ...... 39

2.3 Acute Dissociation of Rat PASMCs ...... 39

2.4 Primary Culture of Rat PASMCs ...... 40

2.5 Effect of Cl- Channel Inhibitors on Cell Number of Cultured Rat PASMCs ...... 40

2.6 Effect of Cl- Channel Inhibitors on BrdU Incorporation and DNA Content in Cultured Rat PASMCs ...... 41

2.7 Effect of Cl- Channel Inhibitors on Annexin V and PI Staining in Cultured Rat PASMCs ...... 41

2.8 Culture and Hypoxia Exposure of Human PASMCs ...... 42

2.9 Immunocytostaining of CLC-3 in Human PASMCs ...... 43

2.10 Quantitative Real-Time Reverse Transcription-PCR (qRT-PCR) ...... 43

2.11 Lentivirus Generation ...... 44

2.12 Transduction of Human PASMCs ...... 45

2.13 Recording of Membrane Potential by Whole-Cell Patch-Clamp Technique in Rat PASMCs ...... 45

2.14 Recording of Cl- Currents by Whole-Cell Patch-Clamp Technique in Rat and Human PASMCs ...... 46

2.15 Antibodies and Reagents ...... 47

2.16 Statistical Analysis ...... 47

Chapter 3 ...... 48

3 Regulation of Proliferation and Membrane Potential by Cl- Currents in Rat Pulmonary Artery Smooth Muscle Cells ...... 49

3.1 Abstract ...... 49

3.2 Introduction ...... 50

3.3 Summary of Materials and Methods ...... 50

3.4 Results ...... 51

3.5 Discussion ...... 67

Chapter 4 ...... 71

4 Swelling-Activated Cl- Currents and Intracellular CLC-3 are Required for Proliferation of Human Pulmonary Artery Smooth Muscle Cells ...... 72

4.1 Abstract ...... 72

4.2 Introduction ...... 73

4.3 Summary of Materials and Methods ...... 74

4.4 Results ...... 75

4.5 Discussion ...... 90

Chapter 5 ...... 93

5 Summary and Future Directions ...... 94

5.1 Summary ...... 94

5.2 Future Directions ...... 95

5.2.1 Molecular Identification of ICl in Proliferating PASMCs ...... 95

5.2.2 A More Detailed Role of ICl in PASMC Proliferation ...... 95

5.2.3 In Vivo Studies on Effects of ICl Blockade on PASMC Proliferation and PAH ... 96

5.3 Concluding Remarks ...... 97

Appendix A: Publication List ...... 123

Appendix B: Published Paper 1 ...... 124

Appendix C: Published Paper 2 ...... 133 vi

List of Tables

Table Number Table Title Page

Table 1.1 Intracellular Cl- concentrations in smooth muscles. 12

Table 4.1 Oligonucleotide primers for CLC family genes in quantitative real-time 78 reverse transcription-PCR.

Table 4.2 Oligonucleotide primers for Bestrophin family genes in quantitative 79 real-time reverse transcription-PCR.

Table 4.3 shRNAs used in CLC-3 knockdown studies. 80

vii

List of Figures

Figure Number Figure Title Page

Figure 1.1 The CLC family of Cl- channels in mammals. 21

Figure 1.2 A continuous single-channel recording of CLC-0. 22

Figure 1.3 Structure of CLC-ec1 (an E. coli CLC channel). 23

Figure 1.4 A scheme for siRNA-mediated gene silencing. 33

Figure 1.5 siRNA and target mRNA structures. 33

Figure 1.6 Steps in shRNA-mediated RNAi. 34

Figure 1.7 microRNA-based shRNA-mediated RNAi. 35

Figure 3.1 DIDS (an ICl inhibitor) inhibits proliferation, BrdU incorporation, 52 and cell cycle progression in rat PASMCs.

Figure 3.2 DIDS at low dosages has no effects on rat PASMC death or 53 apoptosis.

Figure 3.3 CFTR inhibitor (CFTRinh-172) does not affect rat PASMC 54 proliferation.

Figure 3.4 DIDS causes greater membrane hyperpolarization in cultured than 56 in acutely dissociated rat PASMCs.

Figure 3.5 Iodide substitution causes greater membrane hyperpolarization in 57 cultured than in acutely dissociated rat PASMCs.

Figure 3.6 Cl- currents recorded with nystatin-perforated method in rat 59 PASMCs.

viii

Figure 3.7 Baseline and hypotonic solution-induced Ca2+-independent Cl- 60 currents in cultured rat PASMCs.

Figure 3.8 Effects of anion substitutions on reversal potentials of swelling- 61 activated currents and baseline currents in rat PASMCs.

Figure 3.9 Hypotonic solution perfusion does not activate Ca2+-independent Cl- 64 currents in acutely dissociated rat PASMCs.

Figure 3.10 Ca2+-activated Cl- currents in cultured and acutely dissociated rat 65 PASMCs.

Figure 3.11 Ca2+-independent Cl- currents in PASMCs acutely dissociated from 66 hypoxic rats.

Figure 4.1 DIDS-sensitive, swelling-activated Cl- currents in human PASMCs. 81

Figure 4.2 DCPIB inhibits swelling-activated Cl- currents in human PASMCs. 82

Figure 4.3 DIDS and DCPIB inhibit human PASMC proliferation. 83

Figure 4.4 Standard curves of quantitative real-time RT-PCR primers. 84

Figure 4.5 Relative mRNA amounts for CLC and Bestrophin Cl- channel genes 85 in human PASMCs.

Figure 4.6 CLC-3-shRNAs selectively reduce CLC-3 expression in human 86 PASMCs.

Figure 4.7 CLC-3 knockdown inhibits human PASMC proliferation. 87

Figure 4.8 Swelling-activated Cl- currents in human PASMCs expressing 88 control-shRNA or CLC-3-shRNA.

Figure 4.9 Immunocytostaining of human PASMCs with anti-CLC-3 antibody. 89

List of Appendices

Appendix A Publications during Ph.D study Page 123

Appendix B Regulation of Proliferation and Membrane Potential by Chloride Page 124 Currents in Rat Pulmonary Artery Smooth Muscle Cells. Wenbin Liang, Julie Basu Ray, Jeff Z. He, Peter H. Backx, and Michael E. Ward. Hypertension, 2009;54:286-293.

Appendix C Role of Phosphoinositide 3-Kinase α, Protein Kinase C, and L-Type Page 133 Ca2+ Channels in Mediating the Complex Actions of Angiotensin II on Mouse Cardiac Contractility. Wenbin Liang, Gavin Y. Oudit, Mikin M. Patel, Ajay M. Shah, James R. Woodgett, Robert G. Tsushima, Michael E. Ward, and Peter H. Backx. Hypertension, 2010;56:422- 429.

Chapter 1

General Introduction

2 1 Introduction 1.1 Pulmonary Arterial Hypertension (PAH)

1.1.1 PAH: A Disease of Pulmonary Vasculature

Pulmonary hypertension (PH) refers to the condition of increased blood pressure in pulmonary arteries. The pulmonary circulation is a low pressure system with a mean arterial pressure of 9-18 mmHg, and PH is diagnosed when mean pulmonary arterial pressure (PAP) is more than 25 mmHg in patients at rest (Barst et al., 2004). Pulmonary arterial hypertension (PAH) is a subgroup of PH (category I) and involves increased pulmonary vascular resistance (PVR, >3 Wood units (Archer et al., 2010b)) and a normal pulmonary capillary wedge pressure (<15 mmHg). The prevalence of PAH is 15-26 cases per million (Humbert et al., 2006; Peacock et al., 2007) with heterologous etiologies. PAH is rare in its idiopathic form and is more common in association with other diseases including collagen vascular disease, congenital systemic-to- pulmonary shunts, portal hypertension, HIV infection, and anorexigen usage (Archer et al., 2000; Simonneau et al., 2004). PAH is usually progressive and elevated PAP increases right ventricular afterload resulting in right heart failure, a major cause of death in PAH patients. If untreated, idiopathic PAH is highly fatal with a median survival of 2.8 years (Rich et al., 1987). Recent studies in PAH pathogenesis have led to several new treatments, including prostacyclin analogs, endothelin antagonists, and phosphodiesterase-5 inhibitors. As a result, the prognosis of PAH has been much improved but the mortality rate still remains high (15% annually) (Thenappan et al., 2007; Humbert et al., 2010).

PAH is primarily a disease of the pulmonary arterial vasculature which determines the resistance to blood flow. Excess pulmonary vasoconstriction is found in PAH and contributes to the development of PAH. Moreover, histopathological studies revealed obstructive structural remodeling of pulmonary arteries in both PAH patients and experimental PAH animals (Taichman et al., 2006). Remodeled pulmonary arteries in PAH have decreased luminal diameter and increased wall thickness, resulting from intimal hyperplasia, medial hypertrophy and adventitial fibrosis, as well as muscularization of small precapillary vessels that are normally non-muscular. These constrictive structural changes are believed to be the major cause of increased PVR in PAH because more than 80% of the PAH patients do not respond to acute vasodilator therapies (Sitbon et al., 2005). PAH vascular remodeling is caused by enhanced proliferation/reduced apoptosis of vascular cells (particularly smooth muscle cells) (Humbert et

3 al., 2004; Hassoun et al., 2009) and increased extracellular matrix production (Cowan et al., 2000).

Multiple factors contribute to the proliferation of pulmonary vascular cells in PAH. Endothelial dysfunction, platelet activation and inflammation, commonly implicated in PAH, increase levels of pro-proliferative factors (e.g. serotonin (Herve et al., 1995), endothelin-1

(Giaid et al., 1993) and thromboxane A2 (Christman et al., 1992)) and reduce levels of anti- proliferative factors (e.g. prostacyclin (Christman et al., 1992) and nitric oxide (NO) (Giaid et al., 1995)) in pulmonary vascular wall. This imbalance of the pro-proliferative/anti-proliferative factors is a major stimulus for cellular proliferations. New therapies designed to correct some of these abnormalities, including prostacyclin analogs (Sitbon et al., 2002), endothelin antagonists (Channick et al., 2001), and phosphodiesterase-5 inhibitors (Galie et al., 2005; Archer et al., 2009) (which enhance the NO-cGMP pathway), have proven useful for treating PAH.

Alterations in the activity of ion channels and transporters in pulmonary artery smooth muscle cells (PASMCs), the dominant cell type in vascular wall, can also promote proliferation. PASMCs from PAH patients or experimental animals are associated with decreased activities (Yuan et al., 1998a) and expression (Yuan et al., 1998b) of voltage-gated K+ channels (Kv, especially Kv1.5) and increased activities of transient receptor potential channels (TRP) (Yu et al., 2004) and Na+/Ca2+ exchanger (Zhang et al., 2007a). These changes in ion channels and transporters lead to membrane depolarization and increased Ca2+ influx, which promotes proliferation partly by Ca2+/calcineurin-mediated activation of NFAT (Bonnet et al., 2007). These ion channels and transporters represent novel therapeutic targets. In fact, increasing PASMC Kv activities with Kv1.5 gene therapy (Pozeg et al., 2003) has been shown to be helpful in a rat PAH model.

Genetic factors also play a role in some forms of PAH. Loss-of-function mutations of bone morphogenetic protein receptor 2 (BMPR2) are found in ~80% of the familial PAH (a subtype of idiopathic PAH) (Deng et al., 2000; Lane et al., 2000) and promote PASMC proliferation because the normal BMPR2-Smad pathway inhibits proliferation (Yang et al., 2008a). Single nucleotide polymorphisms (SNPs) of ion channel gene promoters (Kv1.5 (Remillard et al., 2007) and TRPC6 (Yu et al., 2009)) identified in some idiopathic PAH patients may cause the abnormal ion channel expression, which promotes proliferation via mechanisms mentioned above. Corrections of genetic defects with gene therapy are being tested in

4 experimental animals (Reynolds et al., 2007; Reynolds et al., 2010) and may lead to novel therapies.

1.1.2 Hypoxia and PAH

Hypoxia plays a critical role in the pathogenesis of a subtype of PAH that is associated with preexisting hypoxic lung diseases, such as chronic obstructive pulmonary disease (COPD) (Naeije, 2005) and sleep apnea (Dempsey et al., 2010), or that occurs in patients living at high altitudes (Arias-Stella et al., 1963). In fact, long-term exposure to a low-oxygen environment is sufficient to cause PAH. For example, the mean pulmonary arterial pressure was doubled

(Rabinovitch et al., 1979) in rats after breathing air with a 50% reduction in O2 partial pressure for 2-3 weeks (using either hypobaric air or nitrogen-diluted normobaric air). This has been the most commonly used animal model in PAH studies. Hypoxia causes PAH by stimulating vascular constriction and remodeling. Acute hypoxia leads to hypoxic pulmonary vasoconstriction (Weir et al., 1995) and plays a key role in the early stage of PAH, while long- term hypoxia induces vascular cell (particularly PASMC) proliferation, contributing to the obstructive vascular remodeling in PAH (Niedenzu et al., 1981; Stenmark et al., 2006).

The chronic hypoxia-induced PASMC proliferation in intact animals is not fully understood and several mechanisms have been suggested. Hypoxia decreases PASMC activities and expression of voltage-gated K+ channels (Smirnov et al., 1994; Wang et al., 1997a), and enhances capacitative Ca2+ entry (via TRP channels) (Fantozzi et al., 2003; Lin et al., 2004) 2+ 2+ leading to elevated cytosolic free Ca level ([Ca ]i) that is a potent stimulator for proliferation (see discussion on page 10). These changes in PASMC ion channels result both from direct effect of hypoxia and indirectly from hypoxia-induced secretions of mitogenic factors, such as endothelin-1 and serotonin. Hypoxia may directly regulate K+ channel and TRP channel expressions by hypoxia-inducible factor-1 (HIF-1)-mediated mechanisms since the effect of hypoxia was not observed in mice with partial HIF-1α deficiency (Shimoda et al., 2001; Wang et al., 2006). In addition, hypoxia increases endothelin-1 (ET-1) level and ET-1 receptor expression in pulmonary artery walls (Li et al., 1994; Aversa et al., 1997), and ET-1 has been suggested to reduce the expression of voltage-gated K+ channels in PASMCs (Whitman et al., 2008). Hypoxia also increases lung serotonin level (Esteve et al., 2007), which inhibits voltage-gated K+ channels in PASMCs (Cogolludo et al., 2006). Several other growth factors are also affected by

5 hypoxia and may contribute to PASMC proliferation. For example, hypoxia-induced pulmonary vasoconstriction increases local shear stress, which activates endothelial transcription and synthesis of TGF-ß and PDGF whose promoters contain shear stress response element (Resnick et al., 1993; Resnick et al., 1995; Voelkel et al., 2000). Another mechanism for hypoxia-induced PASMC proliferation is cellular alkalization (Quinn et al., 1996; Madden et al., 2001). Hypoxia + + increases PASMC intracellular pH (pHi) by enhancing expression and activity of Na /H exchanger (NHE) (Rios et al., 2005; Shimoda et al., 2006), which uses the Na+ gradient to extrude H+ (Quinn et al., 1991). In support of this conclusion, NHE1 deficient mice had less pulmonary artery remodeling after hypoxia exposure (Yu et al., 2008).

1.1.3 Phenotypes of Vascular Smooth Muscle Cells

Vascular smooth muscle cells (VSMCs) in mature mammals are highly differentiated, show very low rates of replication (Owens et al., 2004), and their major function is contraction that determines the vascular diameter and resistance to blood flow. However, contractile VSMCs retain the capability to switch to a proliferative, synthetic phenotype (Owens, 1995; Miguel- Velado et al., 2005) in response to increased levels of growth factors, cytokines and inflammatory mediators, as occur in vascular diseases and injury (Clowes et al., 1983). This phenotypic transition and subsequent proliferation of VSMCs play a key role in the adverse vascular remodeling in many vascular diseases, including PAH (Stenmark et al., 2006) and atherosclerosis.

Contractile and proliferative VSMCs have different gene expression patterns that fit their respective functions. Contractile VSMCs have a high expression level of contractile proteins, including smooth muscle (SM) α-actin and SM myosin heavy chain (SMMHC), two differentiation marker genes (Owens et al., 2004). CArG element [CC(A/T-rich)6GG] is present in the promoter-enhancer regions of these differentiation marker genes and is involved in gene expression regulation. Binding of serum response factor (SRF, a transcription factor) to CArG element, with myocardin as a coactivator, is a critical step for transcription of differentiation marker genes and maintenance of a contractile phenotype. In contrast, proliferative VSMCs have decreased expressions of differentiation marker genes and increased expressions of genes required for proliferation and synthesis (Kawai-Kowase et al., 2007). Several transcription factors are involved in the gene expression alterations in the transition of VSMCs from the

6 contractile to the proliferative phenotype (Owens et al., 2004; Wamhoff et al., 2006). For example, Krüppel-like factor 4 is important for the suppression of differentiation marker genes (Liu et al., 2005), while Krüppel-like factor 5 is involved in the expression of embryonic isoform of smooth muscle myosin heavy chain (SMemb/NMHC-B) (Nagai et al., 2000; Suzuki et al., 2009), which is a marker gene of proliferative VSMCs.

Of interest, some ion channels have been found to be differentially expressed in contractile and proliferative VSMCs. For example, contractile VSMCs have a higher expression 2+ 2+ + of L-type Ca channels (LTCC) and large-conductance, Ca -activated K channels (BKCa), both of which play key roles in regulating vascular tone. LTCC is the major pathway for Ca2+ entry triggering vasoconstriction, while activation of BKCa hyperpolarizes membrane potential leading to vasodilation (Amberg et al., 2003). In contrast, proliferative VSMCs have greater activities of T-type Ca2+ channels (TTCC) (Kuga et al., 1996; Pluteanu et al., 2011) and TRP channels (Golovina et al., 2001; Yu et al., 2003; Bergdahl et al., 2005; Kumar et al., 2006), which mediate Ca2+ entry important for proliferation. Upregulation of TTCC in PASMCs by IGF-1 stimulation is mediated by PI3K/Akt pathway (Pluteanu et al., 2011) and RNAi inhibition of TTCC reduces proliferation (Rodman et al., 2005; Pluteanu et al., 2011). Upregulation of TRPC6 channels in PASMCs upon PDGF stimulation is mediated by c-Jun/STAT3 (signal transducer and activator of transcription 3) and is required for proliferation (Yu et al., 2003).

1.2 Cellular Proliferation

1.2.1 Cell Cycle

Cell cycle, the growth and division of a cell to give two daughter cells, is the basic unit of cellular proliferation. The duration of a cell cycle for a typical mammalian cell is about 24 hours, and the growth phase (interphase, including G1, S, and G2 phases) occupies 23 hours while the division (M phase) takes 1 hour (Alberts et al., 2008). In the growth phase cells are prepared for the division. Chromosomes are accurately duplicated in S phase (which takes 10-12 hours)

(Alberts et al., 2008), and the gap phases (G1 between M and S phase; G2 between S and M phases) allow more time for doubling of other cell contents, including proteins and organelles (Alberts et al., 2008). Cell division occurs in M phase and involves the splitting of both the

7 nucleus (mitosis) and cytoplasm (cytokinesis), giving two genetically identical daughter cells.

The daughter cells, depending on the external conditions, can either enter the G1 phase of a new cell cycle or stop cycling by entering a specialized G0 phase, in which it can stay for a varied period of time (days to years) before re-entering a new cell cycle under the stimulation of growth factors (Alberts et al., 2008).

Cell cycle progression requires both external stimulation signaling (growth factors) and internal cell cycle signaling (cyclins and cyclin-dependent protein kinases). Growth factor stimulation is required for a cell to go through the G1 phase, particularly for passing the restriction point at the end of G1. Once the restriction point is passed, the cell is independent of growth factors and committed to complete the rest of the cycle driven by the internal cell cycle control signaling (Alberts et al., 2008). The key component of the internal cell cycle control system is cyclin-dependent protein kinases (CDKs). As their name implies, activities of CDKs are dependent on the binding of cyclins, which exposes the active sites of CDKs causing partial activation, while full activation requires phosphorylation at the active sites by CDK-activating kinase (CAK). Cyclins are so named because their levels undergo cyclic changes in each cell cycle (while CDK protein levels remain relatively constant in a cell cycle), leading to cyclical assembly and activation of different cyclin-CDK complexes mediating distinct cell-cycle events (Alberts et al., 2008). In addition to the regulation by cyclins, CDK activities are also fine-tuned by other mechanisms (Alberts et al., 2008). Phosphorylation of their inhibitory sites by Wee1 (a Polo-like kinase) inhibits CDK activities, while dephosphorylation by Cdc25 (a phosphatase) increases activities. CDK activities are also inhibited by binding of CDK inhibitors (CKIs, including p21 and p27).

The key events in a typical cell cycle are summarized as follows (Alberts et al., 2008).

Stimulation of growth factor receptors on the plasma membrane of a cell in early G1 phase increases the nuclear Myc level (a transcription factor) mainly via the Ras/mitogen-activated protein kinase (MAPK) pathway and the Wnt/β-catenin pathway. Myc promotes transcription of Cyclin D which, together with CDK 4/6, phosphorylates retinoblastoma protein (Rb) and removes its inhibition of E2F (a key transcription factor at restriction point) (Dyson, 1998). Activated E2F increases the transcription of other cell cycle regulators required for S phase entry and DNA synthesis, including Cyclin E, Cylin A and Cdc25A (Sears et al., 2002; Trimarchi et al., 2002; Dimova et al., 2005). In S phase, chromosomes (both DNA and packaging proteins) are duplicated. Cyclin A/CDK 2 are required for the assembly of preinitiation complex at the

8 origins of replication on chromosomes, a key step in the initiation phase of DNA replication. Cyclin A/CDK 2 also increase histone protein synthesis for packaging the newly synthesized DNA strands. A cell divides in M phase which consists of 3 steps: nuclear envelope breakdown and spindle assembly, chromosome separation in anaphase, and cytokinesis. In M phase, Cyclin B/CDK 1 activities are increased in the nucleus, mainly as a result of Cdc25-mediated dephosphorylation of their inhibitory sites. Cyclin B/CDK 1 initiate nuclear envelope breakdown by phosphorylation and disassembly of nuclear pore complexes and nuclear laminas (the framework beneath the envelope). Breakdown of the nuclear envelope allows the assembly of spindle between centrosomes in the cytoplasm and chromosomes in the nucleus. Chromosome segregation is triggered by anaphase-promoting complex (APC). APC activation requires Ca2+/calmodulin-dependent protein kinase II (CaMKII)-mediated phosphorylation of Emi2 and subsequent degradation of Emi2 (by ubiquitination complex SCF) removes its inhibition on APC. APC activates separase which hydrolyzes cohesin molecules that hold together the two sister chromatids, initiating chromosome separation. Cytokinesis is the last step in M phase and involves the “cutting” of the cytoplasm in the middle via the contraction of a contractile ring. The assembly and contraction of the contractile ring (mainly composed of actin and myosin II), triggered by RhoA (a small GTPase) and Ca2+, cause the cleavage furrow observed in a dividing cell. Increases of plasma membrane area, via insertion of membrane vesicles from the intracellular side, are required for the formation of cleavage furrow that cuts the cell into two halves completing a cell cycle.

The orderly progression of a cell cycle is monitored by the p53 surveillance system. Errors such as DNA damage due to cell stress can occur during cell cycling, and are detected by p53 surveillance system which either arrests the cycle progression until errors are corrected or induces apoptosis if the errors are severe and cannot be righted.

1.2.2 Cell Volume Regulation and its Role in Proliferation

The ability of a cell to regulate its volume is critical for survival and function since uncontrolled volume changes jeopardize the delicate cellular structure and metabolism (Lang et al., 1998). The volume of a cell is determined by its content of water, which easily permeates the plasma membrane because of the expression of aquaporin water channels (Borgnia et al., 1999) in most cells. A stable cell volume is maintained by the balance between the osmolarities of

9 intracellular and extracellular fluids. Because cells contain impermeant macromolecules (e.g. proteins) that are negatively charged, more permeant ions would stay in the inside than in the outside of the cells (the Gibbs–Donnan effect). This is counteracted by the active ion extrusion mechanisms (e.g. the Na+/K+ pump) maintaining a constant cell volume (Hoffmann et al., 1989). However, this balance can be disturbed by physiological cell activities (transport of ions and organic osmolytes across cell membrane; synthesis and breakdown of macromolecules in cells) or under pathological conditions (e.g. ischemia (Jennings et al., 1991) which alters cellular metabolism and causes waste accumulation in interstitial space). In face of these challenges, volume regulation mechanisms are activated to counteract unfavorable volume changes (Lang et al., 1998). Cell swelling (volume increase) activates membrane transport systems leading to loss of ions (mostly K+ and Cl- via ion channels) and organic osmolytes, a process known as regulatory volume decrease (RVD) (Lang et al., 1998; Hoffmann et al., 2009). In contrast, cell shrinkage (volume decrease) activates the regulatory volume increase (RVI) mechanism, leading to accumulation of Na+, K+ and Cl- via activities of Na+/K+/2Cl- cotransporter, Na+/H+ - - exchanger, and Cl /HCO3 exchanger (Lang et al., 1998; Hoffmann et al., 2009).

Cell cycle progression is associated with volume changes (Bussolati et al., 1996). Changes in volume are not an epiphenomenon, but rather appear obligatory for a cell cycle (Matsumura et al., 2010). A general increase in cell volume is obvious in each cycle that produces two daughter cells, each of which has a volume similar to that of its parent cell.

Specifically, the volume of a cell increases in G1 phase (Pendergrass et al., 1991) and, in yeasts, cell volume has to increase and reach a threshold level before entering S phase (Johnston, 1977). Cell volume is further increased to a higher level in S phase (Bussolati et al., 1996). In addition, cell volume is also fine-tuned at certain stages. In glial cells, volume has been shown to be actively adjusted (reduced) in M phase, leading to chromatin condensation before cell division (pre-mitotic condensation, PMC) (Habela et al., 2007; Habela et al., 2008). A role of volume changes in cell cycle is suggested by the observations that proliferation was inhibited when cells were cultured in a hypertonic medium (Michea et al., 2000) that limits volume increase, and that increased proliferation was observed when cells were cultured in a slightly hypotonic medium (Zhang et al., 2007b).

Changes in cell volume during cell cycling are primarily caused by alterations in the number of osmotically active molecules in the cells, since osmolarities of extracellular fluids (cell culture media in vitro or interstitial fluids in vivo) are relatively stable for most cells.

10 Indeed, volume increases during cell cycling have been suggested to be caused by accelerated transport of ions (mostly K+ and Cl-) and amino acids (mostly glutamine and glutamate via the Na+-driven “system A” mechanism) into the cells (Bussolati et al., 1996). Growth factor- stimulated Na+/K+/Cl- cotransport plays a major role in volume increase. Inhibition of Na+/K+/Cl- cotransport led to a 50% reduction in cell volume increases in association with a 60% decrease in cell proliferation (Bussolati et al., 1996). Other factors affecting intracellular osmolarities include synthesis of macromolecules (DNA, proteins, and spindle) and breakdown of cellular structures (nuclear membrane and spindle).

1.2.3 Membrane Potential and Ca2+ in Proliferation

Mammalian cells have a negative plasma membrane potential (Em) (ranging from -10 to -90 mV), which is determined by their membrane conductances to different permeant ions + + 2+ - + (K , Na , Ca , and Cl ), with K conductance dominating in most cases. Em is an important 2+ 2+ regulator of cytosolic free Ca level ([Ca ]i) (Nelson et al., 1990). Membrane depolarization leads to Ca2+ influx by increasing activities of voltage-gated Ca2+ channels. Membrane depolarization also favors the reverse mode (Ca2+ entry mode) operation of the Na+/Ca2+ 2+ exchanger (Blaustein et al., 1999), which contributes to elevated [Ca ]i. Studies have suggested 2+ a role of Em and [Ca ]i in regulation of proliferation (Platoshyn et al., 2000). Cells undergoing active cycling generally have a more depolarized Em than growth-arrested cells. For example, proliferating Chinese hamster ovary cells have an Em of -10 to -20 mV, which was decreased to -

60 mV when proliferation was stopped by contact inhibition (Cone et al., 1973). Em of proliferating VSMCs is also more depolarized than that of non-proliferating VSMCs (Platoshyn et al., 2000; Miguel-Velado et al., 2005). Moreover, proliferating cells in different cycle phases also have different Em (Sachs et al., 1974). The Em changes appear to play a role in proliferation since forced Em depolarization is able to induce DNA synthesis and mitosis in otherwise nonproliferating mature neurons (Cone et al., 1976). Ca2+ signaling is critical for cell cycle 2+ progression. In rat liver epithelial cells, epidermal growth factor stimulation increases [Ca ]i via 2+ 2+ plasma membrane Ca influx, and this [Ca ]i increase is necessary for DNA synthesis (Hill et al., 1988). Ca2+ levels in culture medium are positively related with proliferation rate (Hickie et al., 1983), and rat PASMC proliferation was completely stopped when culture medium Ca2+ level was reduced to ~500 nM (Platoshyn et al., 2000). In addition, depletion of IP3-sensitive

11 intracellular Ca2+ store with SERCA pump inhibitors (thapsigargin) inhibited DNA synthesis and stopped proliferation in smooth muscle cells (Short et al., 1993). Thus, both extracellular Ca2+ 2+ 2+ influx and intracellular Ca release are required for proliferation. Elevated [Ca ]i may stimulate proliferation by regulating Ca2+-dependent cyclin/CDK activities and expression of cell cycle- promoting transcription factors (Hardingham et al., 1997; Husain et al., 1997; Choi et al., 2006). 2+ As discussed above (page 8), an elevated [Ca ]i is also critical for chromosome segregation during mitosis and contraction of the contractile ring during cytokinesis.

1.3 Mammalian Cl- Channels

1.3.1 Cl- Equilibrium and Cl- Channels in Mammalian Tissues

Chloride ions (Cl-) are the most abundant anion in mammalian tissues. Blood and interstitial - - - fluid contain 108 mM Cl (and 28 mM HCO3 ) (Guyton et al., 2006), while intracellular Cl - - levels ([Cl ]i) are tissue specific. In many tissues, including skeletal muscle and neurons, Cl is primarily passively distributed across cell membrane according to the Donnan equilibrium with - [Cl ]i in the range of 5-10 mM (Kerkut et al., 1966; Hironaka et al., 1980). The equilibrium - potential of Cl (ECl) in these tissues is close to resting membrane potential (Em) and opening of - Cl channels stabilizes Em by counteracting depolarization. In contrast, vascular smooth muscle has a high activity of Cl- transport mechanisms that accumulate Cl- in cells (Chipperfield et al., 2000). These mechanisms include the Na+-K+-2Cl- co-transporter, the Na+-dependent and Na+- - - - independent Cl /HCO3 exchangers (Aickin, 1990; Madden et al., 2001). Studies with Cl - - sensitive microelectrode and fluorescent probes showed that [Cl ]i is in the range of 31–51 mM in different smooth muscle tissues (Table 1.1) (Aickin, 1990; Davis et al., 1997). Accordingly, the - ECl (-18 to -35 mV) is above the resting Em (-48 to -67 mV, Table 1.1) and opening of Cl channels generates depolarizing currents in smooth muscle. Activities of smooth muscle Na+-K+- 2Cl- co-transporter are under regulations of hormones (e.g. angiotensin II, atrial natriuretic peptide and catecholamines) and growth factors (e.g. epidermal growth factor) (Owen, 1984; O'Donnell et al., 1986; O'Donnell et al., 1987; Owen et al., 1989; Davis et al., 1997; - - Chipperfield et al., 2000), which may affect [Cl ]i and thus ECl and Cl currents.

Table 1.1 Intracellular Cl- concentrations in smooth muscles. Reproduced from (Chipperfield et al., 2000) with permission.

Cl- channels are lipid membrane proteins that allow passive diffusion of Cl- ions down their electrochemical gradients across the membrane. Compared to the well-studied cation channels, Cl- channels have been less understood and channel-specific antagonists are generally not yet available. Nevertheless, the crucial roles of Cl- channels in diverse physiological functions and diseases have been proven. Cl- channels in plasma membrane regulate membrane excitability, transepithelial transport (secretion and absorption), cell volume and ion homeostasis, cell proliferation and apoptosis (Jentsch et al., 2002; Nilius et al., 2003). Cl- channels are also present in the membrane of intracellular organelles, including endosomes, lysosomes, sarcoplasmic reticulum, mitochondria, and secretory/synaptic vesicles, and regulate organelle volume and electrical potential (Jentsch et al., 2002; Nilius et al., 2003). Abnormalities in Cl- channel functions have been associated with myotonia, cystic fibrosis, renal diseases (kidney stones and proteinuria), osteopetrosis, and lysosomal storage disease. Although several families of Cl- channel genes have been cloned, the genes for some Cl- currents have not been identified. Cl- channels/currents are categorized by their gating (activation) mechanisms: swelling-activated Cl- currents, Ca2+-activated Cl- currents, voltage-gated CLC Cl- channels, cAMP-activated CFTR Cl- channel, and ligand-gated (glycine- or GABA-activated) Cl- channels.

1.3.2 Swelling-Activated Cl- Currents

- Swelling-activated Cl currents (ICl,swell) are found in almost every cell type examined and play an important role in cell volume regulation, proliferation and apoptosis (Kelly et al., 1994; Okada et al., 1997; Chen et al., 2007; Lang et al., 2007; Okada et al., 2009). The properties of

ICl,swell have been well characterized in electrophysiological studies and include activation by cell swelling, outward rectification, dependency on cytosolic ATP, anion permeability sequence of I- > Cl- > gluconate, a single-channel conductance of 10-100 pS, and inhibitions by DIDS, NPPB, tamoxifen, and extracelullar ATP (Okada, 1997). However, the gene encoding ICl,swell remains unknown, although several candidates have been suggested (Okada et al., 1998; Okada, 2006;

Okada et al., 2009). Moreover, the signaling pathway from volume sensor to activation of ICl,swell is poorly understood (Okada et al., 2009).

The current-voltage curves of ICl,swell show modest or moderate outward rectification in whole-cell patch-clamp recordings from a variety of mammalian cells, including epithelial,

14 endothelial, and muscle cells, fibroblast and osteoclast (Kelly et al., 1994; Okada, 1997). The outward rectification was also observed in single-channel recordings (Okada et al., 1994; Jackson et al., 1995), suggesting that it is an intrinsic property of the channel pore rather than resulting from voltage-dependent gating. ICl,swell also show time-dependent inactivation at high depolarizing voltages (>50 mV) with faster kinetics at higher voltages. The inactivation is affected by cytosolic free Mg2+ level, suggesting Mg2+ blockage as a possible mechanism

(Okada, 1997). ICl,swell channels are permeable for most small anions and negatively charged amino acids with a permeability sequence of SCN- > I- > Br- > Cl- > F- > amino acids (aspartate, glutamate, gluconate, lactate, and glycine). ICl,swell are sensitive to several ICl blockers including

DIDS (IC50, 10-50 µM), NPPB (IC50, 5-10 µM), and tamoxifen (IC50, 1-5 µM), as well as heavy metal ions (Gd3+ and La3+) (d'Anglemont de Tassigny et al., 2003). These blockers are, however, not specific for ICl,swell. DCPIB has been shown to be a relatively specific blocker for ICl,swell with - an IC50 of 4 µM, at which concentration it has little or no effects on many other Cl channels or K+ channels (Decher et al., 2001).

Despite the well-characterized biophysical properties of ICl,swell, the gene encoding the channel has been unclear. Several candidate genes have been proposed, but conflicting results have been reported over the last two decades. The difficulty in molecular identification of ICl,swell channel may result from the lack of highly potent and specific inhibitors/ligands for the channel, which hampers the purification of ICl,swell protein, and the expression of the current in almost all cell types including the commonly used heterologous expression systems making the strategy of expression cloning less useful. P-glycoprotein (P-gp, encoded by MDR1 gene), a transporter of the ATP-binding cassette (ABC) family, was first suggested to be a candidate for ICl,swell channel based on observations that expression of MDR1 in fibroblasts generated Cl- currents that have biophysical properties similar to those of native ICl,swell (Valverde et al., 1992). However, this conclusion was later corrected by the same research group that P-gp itself does not mediate

ICl,swell but is a regulator of ICl,swell based on further observations that native ICl,swell were found in HeLa cells (a cervical cancer cell line) that do not express MDR1 and that MDR1 overexpression conferred PKC sensitivity of ICl,swell in HeLa cells and increased ICl,swell activation rate in CHO cells (Hardy et al., 1995; Bond et al., 1998). The conclusion that P-gp is not the native ICl,swell channel is further supported by several other studies (Dong et al., 1994; Morin et al., 1995). pICln (protein for new ICl) was also suggested to be a candidate for ICl,swell channel based on expression cloning studies showing that pICln expression in xenopus oocytes elicited Cl- currents

15 with biophysical and pharmacological properties similar to those of native ICl,swell (Paulmichl et al., 1992). This conclusion was later questioned by findings that native ICl,swell are present in xenopus oocytes (Krapivinsky et al., 1994) and that pICln is primarily a cytoplasmic protein (Krapivinsky et al., 1994; Emma et al., 1998). Heterologously expressed CLC-2 channel is activated by cell swelling but its biophysical properties are distinct from those of native ICl,swell

(Grunder et al., 1992), making it unlikely a candidate for ICl,swell.

CLC-3 as a candidate for ICl,swell remains a subject of current debate. CLC-3 was first proposed to encode ICl,swell in studies showing that CLC-3 expression in NIH/3T3 cells generated - Cl currents that are sensitive to cell volume and have properties of ICl,swell (Duan et al., 1997;

Duan et al., 1999). Intracellular dialysis of an anti-CLC-3 antibody prevented ICl,swell activation in cardiomyocytes and PASMCs, as well as in CLC-3-transfected NIH/3T3 cells (Duan et al.,

2001). Inhibition of CLC-3 with antisense oligonucleotides reduced activation of ICl,swell in epithelial cells (Wang et al., 2000), HeLa cells and xenopus oocytes (Hermoso et al., 2002).

Moreover, inhibition of ICl,swell was also observed with RNAi inhibition of CLC-3 in mesenchymal stem cells (Tao et al., 2008) and nasopharyngeal carcinoma cells (Xu et al., 2010).

Reduced ICl,swell have also been reported in cardiomyocytes isolated from inducible cardiac- specific CLC-3 knockout mice (Xiong et al., 2010). However, there is also considerable evidence suggesting that CLC-3 does not encode ICl,swell. First, immunohistochemistry studies showed that CLC-3 protein is primarily located in intracellular organelles in native cells (Stobrawa et al., 2001; Miller et al., 2007), and only ~6% of the CLC-3 protein resides in plasma membrane in CLC-3 overexpressed cells (Weylandt et al., 2001; Zhao et al., 2007). Second, some groups reported that heterologous expression of CLC-3 either failed to generate detectable Cl- currents (Friedrich et al., 1999; Weylandt et al., 2001), or generated Cl- currents that had properties distinct from those of ICl,swell (Li et al., 2000; Matsuda et al., 2008). Finally, ICl,swell were not altered in cells isolated from CLC-3 knockout mice (Stobrawa et al., 2001; Arreola et al., 2002; Gong et al., 2004).

The mechanism of ICl,swell activation is not fully clear, partly because their molecular identity is unknown. ICl,swell can be activated by cell swelling induced either by bathing cells in a hypotonic solution (Okada, 1997) or by dialyzing cell interior with a hypertonic solution (Best et al., 2004). ICl,swell can also be activated in the absence of cell swelling by other stimuli, including a reduction in cytosolic ionic strength (Nilius et al., 1998) and intracellular dialysis of GTPγS (Nilius et al., 1999; Nilius et al., 2003). Cell swelling is accompanied by a reduction in cytosolic

16 ionic strength, which possibly contribute to the activation of ICl,swell during osmotic swelling

(Sabirov et al., 2000). Activation of ICl,swell requires the presence of intracellular ATP, but its hydrolysis is not needed since ATP can be replaced with a non-hydrolysable analogue (Sakaguchi et al., 1997). Tyrosine phosphorylation may be involved in the signaling pathway of

ICl,swell activation, since inhibitors of protein tyrosine kinase (PTK) prevent ICl,swell activation and inhibitors of protein tyrosine phosphatase (PTP) accelerate ICl,swell activation (Nilius et al., 2003).

The activation of ICl,swell by intracellular GTPγS dialysis suggests the involvement of G protein activations. Consistent with this concept, inhibitors of the small GTPase Rho or its downstream kinase ROCK impair activation of ICl,swell (Nilius et al., 1999), suggesting a role of the Rho-

ROCK pathway in ICl,swell activation. Other studies also suggested the involvements of angiotensin II receptors and MEK/ERK in ICl,swell activation (Okada et al., 2009).

1.3.3 Ca2+-Activated Cl- Currents

2+ - Ca -activated Cl currents (ICl,Ca) are broadly expressed and play roles in smooth and cardiac muscle contractions, airway and intestinal transepithelial transports, neuronal excitability, and olfactory sensation (Janssen et al., 1995; Large et al., 1996; Hartzell et al., 2+ 2005). ICl,Ca are dependent on an increase in [Ca ]i and have outward rectification, time- dependent activation at high depolarizing voltages, an anion permeability sequence of SCN- > - - - - - NO3 > I > Br > Cl >F , and sensitivity to niflumic acid and DIDS (Janssen et al., 1995; Hartzell et al., 2005). The molecular identity of ICl,Ca is not known, but TMEM16a has been recently suggested as a strong candidate (see page 27).

2+ ICl,Ca in vascular smooth muscle are activated by cytosolic Ca released from IP3- sensitive stores or Ca2+ entered via plasma membrane Ca2+ channels (Large et al., 1996).

Because smooth muscle ECl is more depolarized that resting Em (see pages 11-12), ICl,Ca generate 2+ inward currents that depolarize Em and activate voltage-gated Ca channels promoting vasoconstriction (Large et al., 1996). This effect of ICl,Ca contributes to the vasoconstrictive actions of PLC-linked GPCR agonists, such as norepinephrine, angiotensin II and endothelin

(Pacaud et al., 1991; Takenaka et al., 1992; Carmines, 1995). ICl,Ca in cardiomyocytes shorten action potential duration (Hirayama et al., 2002) by contributing to early repolarization (Hiraoka 2+ et al., 1998). Cardiac ICl,Ca also promote arrhythmogenesis in Ca -overloaded conditions by inducing after-potentials (Hiraoka et al., 1998). ICl,Ca are also present on the apical membrane of

17 airway epithelial cells and, together with CFTR, mediate the Cl- flux from cytoplasm to extracellular space hydrating the airway surface (Tarran et al., 2002). In cystic fibrosis, the function of ICl,Ca is preserved (Tarran et al., 2002) and therefore may be a potential target for therapy (Wagner et al., 1991). ICl,Ca also play a role in the transepithelial transports of intestinal 2+ tissue and exocrine glands and are regulated by [Ca ]i–elevating neurotransmitters and cytokines (Hartzell et al., 2005). ICl,Ca are also important for olfactory sensation (Schild et al., 1998). Binding of odorants to G protein-coupled receptors in olfactory receptor neurons leads to 2+ 2+ activation of cAMP-gated Ca permeable channels and Ca influx. The activation of ICl,Ca by 2+ increased [Ca ]i depolarizes membrane potential facilitating action potential generation for olfactory signaling (Schild et al., 1998).

2+ Activities of ICl,Ca are dependent on [Ca ]i. In rat portal vein SMCs, a minimum of 180 2+ 2+ nM [Ca ]i is required for ICl,Ca activation, and full activation is achieved at 600 nM [Ca ]i with 2+ half-maximum activation at 360 nM [Ca ]i (Pacaud et al., 1992). In single-channel studies, 2+ activation curve of ICl,Ca was shifted to more negative potentials with increased [Ca ]i and an estimated Hill coefficient of 2-5 suggested multiple Ca2+ bindings for each channel (Arreola et 2+ al., 1996; Kuruma et al., 2000). Activation of ICl,Ca by Ca is not fully understood and can be mediated by direct Ca2+ binding to channel or by indirect actions via Ca2+-dependent enzymes

(e.g. CaMKII, see below) (Wagner et al., 1991; Nilius et al., 1997b; Hartzell et al., 2005). ICl,Ca activation at high depolarizing voltages is time dependent and activation rate is positively 2+ 2+ affected by [Ca ]i with disappearance of time dependence at >1 µM [Ca ]i (Hartzell et al., 2+ 2005). The current-voltage relationships of ICl,Ca show outward rectification at <500 nM [Ca ]i 2+ but the rectification become less with higher [Ca ]i (Kuruma et al., 2000). As with ICl,swell, the anion permeabilities (ease to enter the channel) of ICl,Ca are related to anion dehydration energies (determined primarily by anion radii) with larger anions have greater permeabilities (i.e. I- > Br- > Cl- > F-) (Hartzell et al., 2005). But anion conductances (ease to pass through the channel) of

ICl,Ca have a bell-shaped relationship to dehydration energies. Although ICl,Ca in whole-cell recording studies showed similar properties in different tissues, single-channel studies suggested that there may be at least four different subtypes of ICl,Ca with unitary conductances reported to be 1-3 pS in cardiomyocytes (Collier et al., 1996) and VSMCs (Klockner, 1993; Piper et al., 2003), 8 pS in pulmonary artery endothelial cells (Nilius et al., 1997a), 15 pS in colon cancer cells (Morris et al., 1993) and biliary epithelial cell (Schlenker et al., 1996), and 50 pS in airway epithelial cells (Frizzell et al., 1986; Hartzell et al., 2005). ICl,Ca can be blocked by several non-

18 selective ICl blockers, such as niflumic acid, DIDS, and NPPB. In addition, some novel ICl,Ca blockers (CaCCinh-A01 and CaCCinh-B01, IC50 ~1 µM) have been recently identified in drug screening studies (De La Fuente et al., 2008).

ICl,Ca activities are regulated by CaMKII-mediated phosphorylation. But the effects of 2+ CaMKII are tissue-dependent. CaMKII is required for (and may mediate) the Ca -induced ICl,Ca activation in airway and biliary epithelial cells (Wagner et al., 1991; Schlenker et al., 1996), colonic tumor cells (Chan et al., 1994), T lymphocytes (Nishimoto et al., 1991), and macrophages (Holevinsky et al., 1994). In contrast, CaMKII inhibits ICl,Ca in rabbit pulmonary and coronary artery SMCs (Greenwood et al., 2001), and airway SMCs (Wang et al., 1997b). In 2+ addition, the Ca -dependent phosphatase calcineurin activates ICl,Ca in rabbit coronary artery

SMCs (Ledoux et al., 2003), suggesting phosphorylation-dependent ICl,Ca inhibition. ICl,Ca are also regulated by CFTR. Expression of CFTR in pulmonary artery endothelial cells inhibits ICl,Ca

(Wei et al., 1999), while loss of CFTR in epithelial cells leads to increased activities of ICl,Ca

(Grubb et al., 1994). The inhibition of ICl,Ca by CFTR is possibly mediated by the interaction between ICl,Ca channels and the CFTR R-domain (see page 26) (Wei et al., 2001). A type of cGMP-dependent ICl,Ca is found in rat mesenteric artery SMCs (Matchkov et al., 2004; Piper et al., 2004) and can be inhibited by PKG antagonists, suggesting the requirement of PKG- mediated phosphorylation for channel activity (Matchkov et al., 2004).

The genes encoding ICl,Ca channels are not known and several candidates have been proposed. CLCA (named after Ca2+-activated Cl- channel) family genes were first discovered in bovine airway epithelium (Cunningham et al., 1995), and their expression in HEK293 cells elicited Cl- currents that were Ca2+-dependent (Gandhi et al., 1998; Gruber et al., 1998). However, CLCA-associated currents have low Ca2+ sensitivity (2 mM for current activation), very weak outward rectification, time-independent activation, and sensitivity to dithiothreitol (a reducing agent), all of which are different from those of classical ICl,Ca (Gandhi et al., 1998;

Gruber et al., 1998; Papassotiriou et al., 2001; Eggermont, 2004). Moreover, ICl,Ca were found in Ehrlich cells that have no CLCA expression (Papassotiriou et al., 2001). Therefore, CLCA as a candidate for classical ICl,Ca is highly contentious (Eggermont, 2004). Bestrophin family genes have also been shown to encode Ca2+-regulated Cl- channels (see page 27) but their biophysical properties (with near linear current-voltage relationship or inward rectification) are different from those of classical ICl,Ca. Nevertheless, one member of the Bestrophin family (Best-3) has been suggested to encode the cGMP-dependent ICl,Ca in VSMCs (Matchkov et al., 2008). TMEM16a

19 has been recently shown to encode Ca2+-activated Cl- channels by three research groups and is a strong candidate for classical ICl,Ca because of their similar biophysical and pharmacological properties and tissue distribution (see page 27) (Caputo et al., 2008; Schroeder et al., 2008; Yang et al., 2008b).

1.3.4 CLC Family

The CLC family Cl- channel genes are highly conserved in species from bacteria to mammals. The first member of the CLC genes (CLC-0) was identified by expression cloning from the electric organ of Torpedo Marmorata (marbled electric ray) by Dr. Jentsch and colleagues (Jentsch et al., 1990). Later work showed that mammals have 9 CLC genes (CLC-1 to -7, CLC-Ka and -Kb) that encode either plasma membrane Cl- channels or intracellular vesicular Cl-/H+ exchangers (Figure 1.1) (Dhani et al., 2006; Jentsch, 2008). Mammalian CLC genes play important roles in physiology and disease (Jentsch et al., 2002). CLC-1 is expressed in skeletal muscle and regulates excitability. Loss-of-function mutation of CLC-1 causes myotonia congenita (Koch et al., 1992). CLC-2 is broadly expressed and plays roles in cell volume regulation (Xiong et al., 1999; Roman et al., 2001), neural excitability regulation (Rinke et al., 2010), and transepithelial Cl- transport (Gyömörey et al., 2000). CLC-3 is widely expressed and has been suggested to encode (Duan et al., 1997) or regulate (Matsuda et al., 2010) ICl,swell. CLC-Ka and -Kb are important for transepithelial transport in renal tubules, and their mutations are linked to renal diseases. Other CLC genes (CLC-4 to -7) are mainly present in the membrane of intracellular vesicles, including endosomes, lysosomes and synaptic vesicles, and may play roles in acidification and Cl- accumulation of the vesicular lumen.

1.3.4.1 General Properties of CLC Channels

CLC channel consists of two identical subunits, each of which contains an ion permeation pathway (pore) (Middleton et al., 1994; Middleton et al., 1996). Each subunit has 17 helices that are inside (but not necessarily span) the lipid bilayer (Dutzler et al., 2002). Eukaryotic CLCs have a large cytosolic C-terminus that contains two CBS domains, which have been shown to modulate channel gating when bound to ATP in skeletal muscle CLC-1 (ATP binding inhibits channel activities by shifting activation voltage to more positive values)

20 (Bennetts et al., 2005). CLC-7 and CLC-Ka/Kb also require an auxiliary β subunit (Ostm1 and barttin, respectively) for channel function.

CLC channel gating is controlled by at least two mechanisms: fast gating and slow gating, which can be appreciated from the single-channel recordings of reconstituted CLC-0 as shown in Figure 1.2 (page 22). Fast gating refers to the rapid transition (in milliseconds) among the 3 current levels (labeled 0, 1, and 2) within each activity burst, corresponding to the independent opening and closing of the two pores in a CLC channel. The appearance and disappearance of bursts indicate slow gating (in seconds), which controls both pores simultaneously and involves interactions between the two subunits (Miller, 2006). The fast gating involves the side chain of an external glutamate residue (Figure 1.3) which, in a closed channel, is negatively charged and blocks a Cl- binding site preventing entry of external Cl- into the pore. Binding of a H+ (protonation) causes the outward rotation of the glutamate side chain, allowing access of external Cl- into the pore. CLC fast gating is dependent on voltage, Cl- concentration ([Cl-]) and pH. CLC channel open probability is increased with more depolarized potentials. However, unlike voltage-gated cation channels, CLC channel does not have a charged domain to serve as voltage sensor. The voltage dependence of gating has been suggested to rely on the charged permeant Cl- ions (Pusch et al., 1995; Chen et al., 2001), because a reduction in external [Cl-] shifts the open probability-voltage curve to more positive membrane potentials (Pusch et al., 1995). Fast gating is also affected by external pH with a lower pH favoring channel opening (Chen et al., 2001). The slow gating of CLC channels are less understood but are also affected by voltage, [Cl-] and pH, as well as temperature.

- - - - CLC channels conduct most small monovalent anions, including Cl , Br , NO3 , I , and - - 2- SCN , but exclude cations and large anions, such as H2PO4 and SO4 . Among the permeant small anions, only Cl- is physiologically relevant and strong selectivity among these anions is not required for CLC channels. The permeability ratios among permeant anions are usually no more ------than 10, and the general permeability sequence is: SCN ≈ NO3 > Cl > Br > NO3 > I . 2- 2- CLC channels are not blocked by intracellular impermeant anions, such as HPO4 and SO4 , possibly due to their lack of wide vestibules preventing the binding of these large anions.

Figure 1.1 The CLC family of Cl- channels in mammals. Based on homology, the nine mammalian CLC proteins can be grouped into three branches, as shown by the dendrogram (left). Channels of the first branch predominantly reside in the plasma membrane, whereas channels from the two other branches are thought to be predominantly intracellular. The localization on human chromosomes is indicated below the channel names. The next columns indicate the most important features of their tissue distribution, their presumed functions, the phenotype of the corresponding knock-out (KO) mouse model, and the name of the human disease associated with the channel, respectively. The asterisk indicates that mutations in barttin, a β-subunit for CLC-Ka and CLC-Kb, cause Bartter syndrome with sensorineural deafness and kidney failure because it compromises the function of both CLC-Ka and CLC-Kb in the kidney and the inner ear. Reproduced from (Jentsch et al., 2002) with permission from APS.

Figure 1.2 A continuous single-channel recording of CLC-0. The horizontal bar over the upper trace marks 1 second. Fast gating is seen as the rapid fluctuations among three levels (representing 0, 1 or 2 pores open) within a 'burst' of activity. Common gating (slow gating) is the appearance and disappearance of the bursts. Holding voltage was -100 mV and single- channel current is 0.9 pA. Reproduced from (Miller, 2006) with permission.

Figure 1.3 Structure of CLC-ec1 (an E. coli CLC channel). a, The protein is depicted in a lipid membrane, with the two subunits differently coloured. Chloride ions are shown as green spheres, and the two proton-transfer glutamate residues, Gluex and Gluin, are spacefilled, with carboxyl oxygen atoms in red. b and c, Diagrammatic representation of Cl--binding region in

'closed' and 'open' conformations defined by the position of the side chain of Gluex. Closed conformation shows two Cl- ions in the 'central' and 'internal' positions; a third Cl- ion, in the 'external' position appears in the open conformation. Reproduced from (Miller, 2006) with permission.

24 1.3.4.2 Properties and Physiology of Individual CLC Channels

CLC-1 channel is expressed in the plasma membrane of skeletal muscle and has a single- channel conductance of 1.5 pS. As with other CLC channels, CLC-1 gating is dependent on voltage, [Cl-] and pH (Rychkov et al., 1996). As mentioned above, intracellular ATP also regulates CLC-1 gating by binding to cytosolic CBS domains (Bennetts et al., 2005) and channel activities can be inhibited by protein kinase C (Rosenbohm et al., 1999). CLC-1 is more sensitive to 9-anthracene-carboxylic acid (9-AC) than other CLC channels. CLC-1 constitutes a majority (~80%) of the resting conductance in skeletal muscle plasma membrane and plays an important role in stabilizing membrane potential. Mice with mutated CLC-1 have increased skeletal muscle excitability (Gronemeier et al., 1994) and CLC-1 mutations are associated with human myotonia (Koch et al., 1992).

CLC-2 is an inwardly rectifying plasma membrane Cl- channel that is ubiquitously expressed (Thiemann et al., 1992) and has a single-channel conductance of 2-3 pS. CLC-2 is activated by hyperpolarizing voltages (Thiemann et al., 1992), cell volume increases (Grunder et al., 1992), and low external pH (Jordt et al., 1997). All these activation mechanisms require the cytoplasmic amino terminus, removing of which generates a constitutively open channel without rectification (Grunder et al., 1992). CLC-2 gating is also affected by intracellular Cl- - - concentration ([Cl ]i). Increases in [Cl ]i shift CLC-2 activation voltages to more positive values (Pusch et al., 1999), facilitating channel activation in physiological voltage ranges. CLC-2 is 2+ 2+ more potently blocked by Zn (IC50 ~40 µM), Cd , and 5-nitro-2-(3-phenylpropylamino)- benzoatic acid (NPPB) (Clark et al., 1998; Furukawa et al., 1998), than other Cl- channel blockers (DIDS and 9-AC). Consistent with biochemistry evidence showing a broad expression of CLC-2 gene, hyperpolarization-activated, CLC-2-like Cl- currents are found in a variety of tissues and play diverse roles in regulation of cell volume (Xiong et al., 1999; Roman et al., 2001), neural excitability (Rinke et al., 2010), and transepithelial Cl- transport (Gyömörey et al., 2000). CLC-2 mutations were associated with epilepsy (Haug et al., 2003), but this conclusion is currently controversial. Mice lacking CLC-2 show degenerations of male germ cells and retinal tissues, suggesting a role of CLC-2 in transepithelial transport and cell-cell interactions (Bosl et al., 2001).

CLC-3, a broadly expressed CLC gene, has been extensively studied but considerable controversies remain about its properties. Kawasaki et al. expressed rat CLC-3 in xenopus

25 oocytes and recorded Cl- currents that have slight outward rectification, sensitivity to DIDS (90% inhibition at 1 mM), and inhibition by PKC activation (Kawasaki et al., 1994). Duan et al. showed that CLC-3 expressed in NIH/3T3 cells generated Cl- currents, which were activated in the absence of cell swelling and resembled classical ICl,swell in terms of biophysical and pharmacological properties (outward rectification, volume sensitivity, and inhibition by DIDS) (Duan et al., 1997). In contrast, Li et al. reported that CLC-3 expressed in CHO cells gave Cl- currents that had strong outward rectification, and insensitivity to PKC activation or DIDS (0.3 mM) (Li et al., 2000). Furthermore, several other groups found that heterologous expression of CLC-3 failed to generate detectable Cl- currents (Friedrich et al., 1999; Decher et al., 2001; Weylandt et al., 2001), suggesting that CLC-3 may not be a plasma membrane channel. CLC-3 knockout mice have a smaller body size (which implies possible metabolism alterations) and severe degeneration of hippocampus and retinal tissues (Stobrawa et al., 2001). A role of CLC-3 in cellular proliferation has been suggested in aortic SMCs (Wang et al., 2002), mesenchymal stem cells (Tao et al., 2008) and nasopharyngeal carcinoma cells (Xu et al., 2010), based on observations that CLC-3 inhibition (with antisense oligonucleotide or RNAi) reduced mitogen- stimulated proliferation. However the mechanisms by which CLC-3 affects proliferation are not clear. Both ICl,swell and proliferation were reduced by CLC-3 inhibition in mesenchymal stem cells (Tao et al., 2008) and nasopharyngeal carcinoma cells (Xu et al., 2010), suggesting a possible role of ICl,swell in mediating the effects of CLC-3. In contrast, Chu et al. suggested that CLC-3 regulates VSMC proliferation by facilitating NOX1 (a NADPH oxidase)-mediated generation of reactive oxygen species in endosomes in response to plasma membrane mitogenic stimulation (Chu et al., 2011).

CLC-K channels (-Ka and -Kb in humans; -K1 and -K2 in rodents) are expressed in the plasma membrane of renal tubular cells, as well as in inner ear. Unlike other CLC channels, CLC-K channels do not have the gating glutamate and their current-voltage curves are near linear or slightly outwardly rectifying (Waldegger et al., 2000). CLC-K channels require a β subunit barttin for function (Scholl et al., 2006), and play a key role in renal transepithelial transport. Loss-of-function mutations of either CLC-K or barttin cause Barter Syndrome (renal salt loss).

CLC-4 to -7 are mainly present in the membrane of intracellular vesicles, including endosomes, lysosomes and synaptic vesicles, and may play roles in acidification and Cl- accumulation of the vesicular lumen. CLC-4 and -5 are believed to function as Cl-/H+ exchangers

26 in intracellular vesicles (Scheel et al., 2005), although plasma membrane Cl- currents with strong outward rectification can be recorded when heterologously overexpressed. CLC-4 is broadly expressed, but its physiological role remains unclear. CLC-5 is expressed in kidney and other tissues and is required for renal tubular endocytosis (Piwon et al., 2000). CLC-5 mutations impair renal tubule reabsorption and are associated with Dent’s disease (proteinuria and kidney stones) (Wrong et al., 1994; Piwon et al., 2000; Yamamoto et al., 2000). CLC-6 and -7 are highly homologous. CLC-6 is highly expressed in nervous tissues and CLC-6 knockout mice have lysosomal storage disease (Poet et al., 2006). CLC-7 is broadly expressed and CLC-7 knockout mice show severe osteopetrosis (Kornak et al., 2001).

1.3.5 Cystic Fibrosis Transmembrane Conductor Regulator (CFTR)

CFTR, a cAMP-activated Cl- channel (Bear et al., 1992), is expressed in cell membranes of epithelia in lung airways, intestine, bile ducts and secretory glands (Nilius et al., 2003). CFTR plays a key role in transepithelial transport and its mutations cause cystic fibrosis, an inherited disease with defective secretory functions of exocrine organs leading to thick mucus and chronic infection in the lumens of lung airways, intestines, and pancreas. CFTR channel consists of two transmembrane domains, each of which contains 6 membrane-spanning helices and a cytoplasmic nuclear binding domain (NBD). A regulatory domain (R-domain) links the two transmembrane domains. Activation of CFTR channel requires phosphorylations of its R-domain by cAMP-dependent kinase (PKA), and is voltage-independent with a linear current-voltage relationship. CFTR channel has a unitary conductance of 6-10 pS and an anion permeability - - - - sequence of Br > Cl > I >F (Nilius et al., 2003). CFTR is blocked by glybenclamide (IC50, 1-10 µM) (Schultz et al., 1999; d'Anglemont de Ta et al., 2003), and DIDS blocks CFTR only from the cytoplasmic side (Linsdell et al., 1996). In addition to being a Cl- channel, CFTR also regulates functions of other membrane channels and transporters. For example, expression of

CFTR inhibits activation of ICl,swell (Vennekens et al., 1999), possibly by the interaction between the CFTR NBD2 domain and the ICl,swell channel (Ando-Akatsuka et al., 2002).

27 1.3.6 Bestrophin Family

Mammalian Bestrophin family genes have four members (Best-1 to -4) and are shown to encode Cl-channels (Sun et al., 2002; Hartzell et al., 2008). Heterologous expression studies suggested that human Best-1 (hBest-1) generates Ca2+-dependent Cl- currents with time- independent activation and weak outward rectification; hBest-2 generates Cl- currents with time- independent activation and no rectification; hBest-3 generates Cl- currents with time-dependent activation and strong inward rectification; hBest-4 generates Cl- currents with time-dependent inactivation and without rectification (Sun et al., 2002; Tsunenari et al., 2003). Mouse Best-2 (mBest-2) generates Cl- currents that are Ca2+-dependent, have an anion permeability sequence - - - - - of SCN > I > Br > Cl > F , and are sensitive to DIDS (IC50, 3.1 μM) (Qu et al., 2004). The tissue expression patterns of Bestrophin genes are not fully clear. hBest-1 protein was found to be exclusively expressed in retinal pigment segment, while mBest-1 was also found in kidney, colon, and airways (Barro Soria et al., 2009). Both hBest-2 and mBest-2 are expressed in colon and testes, while no information is available for Best-3 or Best-4 protein expression patterns. Physiological functions of Bestrophin proteins are unknown. Mutations of hBest-1 are associated with an eye disease (Best vitelliform macular dystrophy) (Marquardt et al., 1998), but it remains unclear if the disease is related to the Cl- channel function of hBest-1.

1.3.7 TMEM16 Family

TMEM16a, a member of the TMEM16 gene family (transmembrane proteins with unknown functions), has been recently shown to confer a Cl- current that is Ca2+-dependent (Caputo et al., 2008; Schroeder et al., 2008; Yang et al., 2008b). TMEM16a-associated currents have many properties that are similar to those of classical ICl,Ca found in many tissues, including outward rectification, time-dependent activation at high depolarizing voltages, an anion - - - - - permeability sequence of NO3 > I > Br > Cl > F , and sensitivity to niflumic acid, NPPB, and

DIDS. The cell surface expression of TMEM16a protein is consistent with its role as a component of native ICl,Ca (Schroeder et al., 2008). TMEM16a is expressed in many tissues that have native ICl,Ca, including lung and renal epithelial cells, pancreatic and salivary gland acinar cells, and sensory neurons (Schroeder et al., 2008; Yang et al., 2008b). In addition, RNAi inhibition of TMEM16a suppressed native ICl,Ca in submandibular gland acinar cells (Yang et al., 2008b), and reduced UTP-stimulated I- flux and short-circuit currents in epithelial cells (Caputo

28 et al., 2008). A role of TMEM16a in encoding rat PASMC ICl,Ca has also been suggested

(Manoury et al., 2010). Interestingly, TMEM16a was shown to be important for ICl,swell activation (Almaca et al., 2009). Another member of the TMEM16 family, TMEM16b, was also suggested to contribute to ICl,Ca (Schroeder et al., 2008). Clearly, further studies are needed to understand the physiological role of TMEM16 proteins.

1.3.8 Ligand-Gated Cl- Channels

Mammalian ligand-gated Cl- channels include type A γ-aminobutyric acid receptors

(GABAA) and glycine receptors which are found in central nervous system (CNS). In adult CNS - - that has a low [Cl ]i, opening of GABAA or glycine receptors mediates Cl influx generating hyperpolarizing currents. The resulting membrane hyperpolarization mediates the inhibitory actions of GABA and glycine, two important inhibitory neurotransmitters in CNS. In CNS still - in early stages of development, however, the [Cl ]i is high and GABAA or glycine receptors mediate Cl- efflux causing depolarization and neuronal excitation. These excitatory actions of GABA and glycine may be important for development (Cherubini et al., 1991). Ligand-gated Cl- channel consists of 5 subunits, each of which (either α, β, or γ isoform) has 4 transmembrane domains and extracellular N- and C-terminal domains. GABAA receptors have an anion permeability sequence of SCN- > I- > Br- > Cl- > F-, multiple unitary conductance levels of 12-30 pS, and are blocked by bicuculline and picrotoxin. GABAA receptors are expressed in the brain and defective GABAA receptors are associated with epilepsy (Olsen et al., 1999). Glycine receptors have a similar anion permeability sequence to that of GABAA receptors. Glycine receptors are expressed in spinal cord and brain stem and defective glycine receptors are associated with startle disease (hyperekplexia) (Rees et al., 1994).

1.4 Aims of Present Studies

As mentioned above, PASMC proliferation plays a key role in PAH pathogenesis; cellular - proliferation is associated with obligatory changes in both Em and cell volume; Cl channels, via - mediation of Cl flux across plasma membrane, regulate both Em and cell volume. In the present studies, we tested the hypothesis that Cl- channels play a role in PASMC proliferation by - examining effects of Cl channel inhibition on PASMC proliferation and by characterizing the ICl

29 expressed in proliferating PASMCs. These studies may identify new targets for therapeutic interventions that can reduce PASMC proliferation and ameliorate PAH progression.

1.5 RNA Interference

Among the various techniques used in the present study, RNA interference (RNAi) is relatively new and I here summarize our current understanding of this technique. Other techniques are briefly described in the Method section (Chapter 2). RNAi refers to a mechanism of gene silencing at the post-transcriptional level, in which double-stranded small interfering RNA (siRNA) triggers degradation of targeted mRNA preventing protein translation (Fire et al., 1998). RNAi is an endogenous process of eukaryotes and presumably functions as a defense mechanism by suppressing unwanted foreign gene expression as in viral infection (Milhavet et al., 2003). The importance of RNAi in biomedical research was first recognized by the demonstration that the RNAi machinery can be harnessed to selectively suppress endogenous gene expression by introducing artificial siRNA designed to bind and degrade target mRNA (Fire et al., 1998; Elbashir et al., 2001). RNAi is a simple but powerful research tool now widely used in loss-of-function studies.

1.5.1 siRNA-Mediated Gene Silencing

RNAi can be experimentally induced by various strategies (see below), all of which converge to a common step in the RNAi pathway: the generation of small interfering RNAs (siRNA), which are double-stranded, ~20-nucleotide (nt) small RNAs with 2-nt 3’ overhangs (Figures 1.4 and 1.5, page 33) (Zamore et al., 2000; Zeng, 2009). siRNAs bind to a group of proteins to form the RISC complex (RNA-induced silencing complex) (Hammond et al., 2000). The sense strand of the siRNA is removed from the RISC complex exposing the antisense strand (also known as guide strand) that binds to the complementary sequence on target mRNA. mRNA th th st is cleaved between the 10 and 11 nts (upstream from the nucleotide bound by the 1 5’ nt of the antisense strand, Figure 1.5) by an endoribonuclease (Argonaute), a core component of the RISC complex. The cut mRNA is degraded by other nucleases in cytosol preventing protein translation of target gene. This mRNA degradation pathway is the classical RNAi mechanism and a second mechanism, inhibition of translation, is less understood and may also play a role in

30 RNAi under certain circumstances (Rossbach, 2010). The siRNA-RISC complex may bind to 3’ UTR (untranslated region) of target RNA with incomplete sequence complementarities, which does not trigger cleavage of the mRNA but instead inhibits its translation (Rossbach, 2010).

Transfection of cells with synthetic siRNA duplexes provides a simple and fast approach to induce RNAi. The algorithms for designing effective siRNA have been developed, and a large reservoir of designed siRNAs and transfection reagents is commercially available. Moreover, some siRNAs have been experimentally validated for effective knockdown of target genes. Thus, siRNA transfection studies can be quickly set up and changes in target mRNA level can be checked 24-48 hours post-transfection for validation (Zeng, 2009). However, the siRNA duplex transfection method has several limitations. First, this approach only leads to transient gene knockdown since introduced siRNA duplexes are diluted when cell grows and divides, limiting its application in rapidly dividing cells (Taxman, 2009). This will also pose a problem when the target protein has a slow turnover rate, making it difficult to achieve an efficient knockdown at the protein level. One solution to this problem may be multiple transfections of cells with the same siRNA duplex every 2-3 days. The second limitation of siRNA transfection method is the low transfection efficiencies in many primary cells with current transfection techniques. Since most commercial siRNA duplexes do not have a fluorescent maker for identification of successfully transfected cells, phenotypic studies at the single-cell level including path-clamp and cell imaging studies are difficult and unreliable. The low transfection efficiency also makes it difficult for knockdown validation studies with quantitative real-time RT-PCR and western blotting studies. The third limitation of the siRNA transfection method is increased off-target effects when a high intracellular siRNA level is needed for efficient target gene knockdown (Taxman, 2009). Cellular immune responses may be activated by certain siRNAs present in endosomal vesicles (Judge et al., 2005), which may occur when lipid-based methods are used to deliver siRNA duplexes.

Another commonly used strategy to selectively activate RNAi process is transduction of cells with viruses engineered to express a transcript that forms a short hairpin (shRNA) which is then processed by cellular endonucleases to release the siRNA duplexes (Figure 1.6, page 34). Lentiviral vectors have been engineered and widely used for this purpose because lentiviruses are able to transduce both dividing and non-dividing cells and integrate the transgenes into the host genome. Therefore, lentiviral vector-mediated RNAi can be used in a wide range of cells and the induced RNAi effects are long-term because of the stable expression of the shRNA

31 (Stewart et al., 2003). Two types of lentiviral shRNA-encoding vectors have been developed. The first type uses a pol III promoter, such as U6 (promoter of the U6 small nuclear RNA gene), to drive the expression of shRNA which consists of a double-stranded stem connected by a loop (Figure 1.6). The property of pol III transcripts to terminate at a repeated U sequence (Campbell et al., 1992) is exploited to control the length of transcribed shRNA, so as to avoid the long RNA duplex-induced interferon response in mammalian cells (Manche et al., 1992; Elbashir et al., 2001; Taxman, 2009). As shown in Figure 1.6, the shRNA transcribed in nucleus is first exported to the cytoplasm with the help of Exportin-5 (Yi et al., 2003) and then cleaved by Dicer (an endonuclease) to release the stem region, which is the siRNA duplex for triggering RNAi. The second type of lentiviral vector is designed to express a microRNA (miRNA) that contains the siRNA duplex in its stem sequence. As shown in Figure 1.7 (page 35), the artificial miRNA is first cleaved by Drosha in the nucleus generating a transcript that mimics a shRNA, which is then exported to cytoplasm and further processed by Dicer to generate siRNA duplex. Transcription of the artificial miRNA is generally driven by a pol II promoter. Since it is difficult to control pol II transcript termination, a modified CMV (cytomegalovirus) promoter in combination with a poly (A) sequence is commonly used in these vectors. Although the stronger U6 promoter leads to a higher level of shRNA expression, the siRNA duplexes produced from the artificial miRNA are more efficiently loaded to the RISC complex inducing more potent RNAi (Cullen, 2005). One disadvantage of shRNA-encoding lentiviral vectors is that the ends of some siRNA duplexes generated from Dosha/Dicer cleavage of precursors may be slightly different from those desired sequences that favor incorporation of the antisense strand into the RISC complex (see below) (Zeng, 2009).

1.5.2 siRNA Duplex Designing

A reliable RNA interference study requires that the siRNA-induced gene knockdown is both effective and specific, i.e. it causes a high level of target transcript reduction while having minimal unwanted effects. Although any parts of the target transcript, including both protein- coding region and UTRs, can be targeted in a knockdown experiment, different siRNAs targeting the same mRNA have been shown to have distinct knockdown efficiencies. Many siRNA design algorithms, although some are controversial, have been developed and some general rules are summarized here. First, regions of the transcript that have known complex secondary structure or

32 protein/RNA interactions should be avoided. Second, since only one strand (antisense) of the siRNA duplex has complementary sequence to target mRNA, it is important to design a siRNA duplex that favors the incorporation of the antisense strand into the RISC complex. Thus, it is preferable for the designed siRNA duplex to have asymmetric sequences because the strand with a weaker binding nucleotide at its 5’ end is more likely to be incorporated into the RISC complex (Schwarz et al., 2003). Accordingly, the 5’ end of the antisense strand should be an A or U, while the 5’ end of the sense strand should be a G or C (Zeng, 2009). This designing, therefore, promotes the incorporation of the antisense strand (and prevents the incorporation of the sense strand) into the RISC complex increasing efficiency of target transcript degradation while reducing sense strand-mediated unwanted effects. Third, some other sequence characteristics of the antisense strand (guide strand) known to favor the RNAi process include an A- and U-rich sequence among the first 7 nts in the 5’ end (Figure 1.5) and A or U at the 10th nucleotide (which is the cleavage site and thus may positively affect transcript cleavage by Argonaute). Fourth, a lower stability (<50% G/C content) of the siRNA duplex, particularly in the center region (9th to 14th nucleotides in antisense strand, Figure 1.5), also tends to increase knockdown efficiency (Pei et al., 2006). Finally, some nucleotides on the antisense strand (positions 1st , 20th and 21st, Figure 1.5) play little, if any, role in recognition of target mRNA and mismatches at these positions can be tolerated (Pei et al., 2006).

A major concern of gene knockdown studies is the unwanted off-target effects and non- specific effects. Although siRNA duplexes that have complete sequence complementarity with any unwanted transcripts can be easily identified and excluded by a simple BLAST search, off- target effects may still be caused by partial sequence complementarities between siRNA duplex (especially the 2nd-8th nucleotides in antisense strand which is known as the seed region (Birmingham et al., 2006), Figure 1.5) and some non-targeted transcripts, whose translation can be blocked by binding of the RISC complex to their 3’ UTR. Therefore, siRNA candidates with significant similarities to unrelated transcripts in a BLAST search should be excluded. In addition, cellular immune responses can be activated by RNA molecules that contain putative immunostimulatory motifs, including U- and G-rich motifs (e.g.UGUGU). Therefore, such RNA motifs should be avoided in both strands of the siRNA duplexes (Marques et al., 2005).

Figure 1.4 A scheme for siRNA-mediated gene silencing. The primary sequence asymmetry of duplex determines which strand is preferentially assembled into RISC (see explanations on page 32). Reproduced from (Pei et al., 2006) with permission.

Figure 1.5 siRNA and target mRNA structures. Upper panel: Standard siRNA duplex. Lower panel: Target mRNA specificity. The cleavage site is indicated by scissors in the target mRNA. Target recognition and off-target activity can occur in two modes, the catalytic siRNA- guided cleavage reaction requiring extensive complementarity in the region surrounding the cleavage site (blue) and the miRNA-like destabilization of mRNAs requiring pairing of the siRNA 5' end (green). Reproduced from (Pei et al., 2006) with permission.

Figure 1.6 Steps in shRNA-mediated RNAi. 1) Plasmid-expressed short hairpin RNA (shRNA) requires the activity of endogenous Exportin 5 for nuclear export. 2) Ago2 (Argonaute 2) is recruited by TRBP, that forms a dimer with Dicer, and then receives the shRNA. 3) The shRNA is cleaved in one step by Dicer generating a 19-23 nt duplex siRNA with 2 nt 3’overhangs. 4) After identification of the “guide strand” in the siRNA duplex, the “passenger strand” is cleaved by Ago2. 5) The “guide strand” is released. 6) The “guide strand” is integrated in the active RISC complex that contains different argonautes and argonaute-associated proteins. 7) The siRNA guides RISC to the target mRNA. 8) RISC delivers the mRNA to cytoplasmic foci named processing bodies (P-bodies or GW- bodies) wherein mRNA decay factors are concentrated. 9) The target mRNA is cleaved by Ago2 and degraded.

Figure and legend are reproduced with permission from: http://www.invivogen.com

Figure 1.7 microRNA-based shRNA-mediated RNAi. Left side: Artificial siRNAs can enter the RNAi pathway as synthetic siRNA duplexes, as shRNAs transcribed by a Pol III promoter or as artificial pri-miRNAs (shRNA-mir) transcribed by a Pol II promoter. Right side: Steps of microRNA-based shRNA-mediated RNAi. Ago2, Argonaute-2; Exp5, Exportin-5. Reproduced from (Cullen, 2005) with permission.

1.5.3 Validation of Knockdown

Highly efficient and specific knockdown of target genes is of great importance for a successful RNAi study. However, none of current algorithms for designing siRNA/shRNA can guarantee knockdown efficiency and specificity since the RNAi mechanism has not been fully understood (Pei et al., 2006). Therefore, it is important to examine experimentally the function of a given siRNA/shRNA. Even previously validated siRNA may still need to be tested, particularly when used in a different cell type.

A typical study starts with three or more siRNA/shRNAs targeting different regions of a given transcript, together with one or two control siRNA/shRNAs that do not target any known transcripts in the cells studied. Quantitative real-time RT-PCR can be used to check the reduction of target transcript in cells after siRNA/shRNA treatment. This method, however, requires that the majority of the cells have taken up the siRNA/shRNA, because the target transcript reduction in transfected cells may be masked by a large number of untransfected cells in the culture. Therefore, it is critical to obtain a high transfection efficiency for experiments using synthetic siRNA duplexes, while cells successfully transduced with shRNA-encoding lentivirus can be selected by their antibiotic resistance gene contained in most lentiviral vectors. Confirmation of gene knockdown at protein level with western blotting is also important for many studies. Changes in protein level after siRNA/shRNA treatment are affected by the turnover rate of the target protein. While rapid turnover proteins can be inhibited 2-3 days after siRNA delivery or shRNA expression, some stable proteins may require up to 7 days before a reduction can be observed (Zeng, 2009).

Off-target effects and non-specific effects of siRNA/shRNAs are a concern in gene knockdown studies, although they are already reduced in the siRNA/shRNA design step that eliminates candidates bearing complementarities with unwanted transcripts or containing immunostimulatory motifs, as discussed in page 32. When the reduction of a target transcript is accompanied by phenotypic changes, it is necessary to examine if other highly related transcripts, e.g. other members of the same gene family that may perform similar functions, are also affected which may cause same changes in phenotype. In addition, the effects of siRNA/shRNAs are likely specific when knockdown of a target transcript by two or more different siRNA/shRNAs produce same phenotypic changes, since siRNA/shRNAs with different sequences are less likely to have overlapping non-specific effects. Another assay for

37 assessing knockdown specificity is to introduce to siRNA/shRNA-treated cells a gene that encodes the protein of interest but can avoid the siRNA/shRNA-induced suppression, by using either a sequence-modified version of the gene or a gene of a different species. If this approach can rescue the phenotypic changes induced by siRNA/shRNA treatment, the specificity of knockdown is highly suggested.

Chapter 2

Materials and Methods

39 2 Materials and Methods

2.1 Animals

All procedures were compliant with the guidelines of the Canadian Council on Animal Care and approved by the institutional animal care committee. Adult male Sprague-Dawley rats (200-250 g, Charles River Laboratories, Montreal, Quebec, Canada) were anesthetized by inhalation of isoflurane. Lungs were excised and PASMCs were isolated (see below) for immediate use in electrophysiological studies (acutely dissociated cells) or else maintained in primary culture for 3 to 5 days. Rats exposed to hypoxia to induce PASMC proliferation in vivo breathed a gas mixture containing 10% O2. Normoxic rats breathed room air under otherwise identical conditions. The arterial O2 partial pressure in rats exposed to hypoxia by this method averaged 38 Torr (range 35-42 Torr) (Auer et al., 1998).

2.2 Detection of PASMC Proliferation by BrdU Uptake In Vivo

To examine rat PASMC proliferation in vivo, bromodeoxyuridine (BrdU) uptake study was performed. BrdU is incorporated in newly synthesized DNA and can be used as an index of cell proliferation. This experiment was conducted by Dr. Julie B. Ray and the method has been described previously (Ray et al., 2008). Briefly, rats were infused subcutaneously with BrdU at ~0.4 mg/h for 48 hours (h) using osmotic pumps before euthanasia. Pulmonary artery segments were collected, fixed, and cut into sections (5 µm thick). Incorporated BrdU was detected with fluorescein-labeled anti-BrdU antibody and examined with microscopy. Significant PASMC BrdU incorporation (proliferation) was observed after 7 days of hypoxia.

2.3 Acute Dissociation of Rat PASMCs

In some electrophysiological studies, single SMCs were isolated on the day of experiments from adult rat pulmonary arteries with a previously described protocol (Archer et al., 1996). Lungs were placed in ice-cold Ca2+-free Hank’s balanced salt solution, containing (in mM): NaCl 140, KCl 4.2, KH2PO4 1.2, MgCl2 0.5, HEPES 10, and EGTA 0.1 (pH=7.4 with

40 NaOH). Small intralobar arteries (3rd and 4th order branches) were dissected and stripped of connective tissue under a dissecting microscope. Arterial segments were placed in Ca2+-free Hank’s solution at room temperature for 10 min and then transferred into Hank’s solution (without EGTA), containing 1 mg/ml papain and 0.2 mg/ml bovine serum albumin at 4C for 20 min. Dithiothreitol (0.85 mg/ml) was added and the enzyme solution kept at 4C for another 50 min and then at 37C for 15 min. The segments were then washed and maintained in Hank’s solution (with 6 mM glucose) at room temperature for 10 min. Single PASMCs were released by gentle trituration through a fire-polished Pasteur pipette. Most cells were relaxed (spindle shape) in Ca2+-free solution, but became partially contracted (oval shape) when 1.8 mM Ca2+ was added. Their identity as smooth muscle cells was confirmed in initial studies by immunostaining with anti-α smooth muscle actin antibody. PASMC suspension was kept at 4C and used for experiments within 8 h of dissociation.

2.4 Primary Culture of Rat PASMCs

In some proliferation and electrophysiological studies, rat PASMCs were isolated and cultured in the presence of serum according to a previously described protocol (Yuan, 1995). Pulmonary arteries were dissected and the adventitia and endothelium removed after incubation in Hank’s solution–Based Minimal Essential Medium (HB-MEM) containing 1 mg/ml collagenase type II at 37°C for 15-20 min. Segments were then cut into small pieces, incubated for 45 min at 37°C in HB-MEM containing 1 mg/ml albumin, 1 mg/ml collagenase type II, and 0.5 mg/ml elastase and PASMCs were dissociated by trituration. Dissociated cells were cultured at 1.4×103 cells/cm2 in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 1 µg/ml fungizone and 50 µg/ml gentamycin in an incubator infused with of 5% CO2/95% air at 37°C. Subconfluent cells at day 3-5 were used for electrophysiology studies. Purity of cultured cells was corroborated, in initial studies, by immunofluorescent staining using anti-α smooth muscle actin antibody.

2.5 Effect of Cl- Channel Inhibitors on Cell Number of Cultured Rat PASMCs

To examine the effects of Cl- channel inhibitors on cell proliferation, rat PASMCs were prepared as above and cultured at 1.4×103 cells/cm2 in DMEM containing 10% FBS in the

41 presence of either 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS, final concentration =10, 30, or 100 μM), 5-Nitro-2-(3-phenylpropylamino) benzoic acid (NPPB, final concentration

=100 μM), CFTRinh-172 (final concentration=0.1, 1, or 10 μM) or vehicle (DMSO). The culture medium was exchanged every 2 days. On days of cell number assessment, cells were washed with PBS and detached with 0.1% trypsin, collected and counted using a standard hematocytometer (Sigma). Trypan blue exclusion was used to test cell viability.

2.6 Effect of Cl- Channel Inhibitors on BrdU Incorporation and DNA Content in Cultured Rat PASMCs

To explore if the effect of DIDS on PASMC numbers was due to altered cell proliferation, DNA synthesis was assessed by BrdU incorporation and DNA content (an index of cell cycle phases) was quantified by propidium iodide (PI) staining. Rat PASMCs were cultured as above, and when at 30~40% confluency were serum deprived (0% FBS) for 24 h to synchronize the cells. The culture medium was changed to DMEM containing 10% FBS and 10 μM BrdU as well as DIDS (30 μM) or vehicle (DMSO, 0.06% v/v), and incubated for 16 h. Cells were trypsinized, pelleted, and washed once in cold PBS then fixed with 70% ethanol and maintained at 4°C for 24 h. The cells were then washed with 1 ml washing buffer (PBS+0.5% BSA) and centrifuged at 230×g for 10 min. Pellets were resuspended in 500 μL denaturing solution (2 M HCl) and incubated for 20 min at room temperature. After washing again with washing buffer, cell pellets were resuspended in 500 μL borate buffer (0.1M Na-borate, pH=8.5) and incubated for 5 min at room temperature. The cells were washed and incubated for 1 h at room temperature in 72.5 μL FITC-anti BrdU loading solution (0.5% Tween 20, 1% BSA/PBS, 0.7 mg/ml RNase A, and 20 μL FITC-anti BrdU solution) then washed and incubated in 100 μL PI solution (10 μg/ml PI in PBS) for 1 h at room temperature. 400 μL PBS was added, and cells were analyzed by a flow cytometer (Cytomics FC500, Beckman Coulter Inc., Fullerton, CA). In each experiment, 10,000 cells were counted.

2.7 Effect of Cl- Channel Inhibitors on Annexin V and PI Staining in Cultured Rat PASMCs

To determine whether the effect of DIDS on cell number can also be attributed to changes in the rate of cell death, a BD Pharmingen Annexin V-FITC apoptosis detection kit was used to detect apoptosis. Phosphatidylserine (PS), a phospholipid present primarily in the inner

42 leaflet of plasma membrane (Op den Kamp, 1979), is translocated to the outer leaflet in early stages of apoptosis (Fadok et al., 1992). Annexin V is a potent and specific ligand of PS, and was used in our study to detect externalized PS in apoptotic cells (Vermes et al., 1995). Propidium iodide (PI), a fluorescent dye, is extruded by viable cells and was used in our study to detect membrane integrity and cell viability. In the present study, rat PASMCs were cultured as above in the presence of DIDS (30 μM) or vehicle (DMSO, 0.06% v/v) for 4 days, a time point at which the effect of DIDS on cell number is readily apparent (Figure 3.1A). Cells were trypsinized and centrifuged at 230×g for 10 min. The pellet was resuspended in 5 ml of cold PBS and centrifuged again. The pellet was then resuspended in 110 µL of Annexin V-FITC labeling solution, containing 100 μL binding buffer, 5 µL Annexin V-FITC solution and 5 µL PI solution, and incubated for 15 min at room temperature in the dark. 400 μL binding buffer was added into each sample. Labeled cells were analyzed by flow cytometry, and the numbers of cells positive for either Annexin V, PI, or both were calculated.

2.8 Culture and Hypoxia Exposure of Human PASMCs

Human PASMCs were obtained from Cambrex Incorporation (East Rutherford, NJ) and cultured at 1×105 cells/dish in 60-mm culture dishes (BD Biosciences, Franklin Lakes, NJ) in smooth muscle growth medium (SMGM-2, Lonza, Walkersville, MD) consisting of smooth muscle basal medium (SMBG) supplemented with 5% fetal bovine serum, 0.1% insulin, 0.2% human fibroblast growth factor-B, 0.1% human epidermal growth factor, and 0.1% GA-1000 (gentamicin and amphotericin B). Human PASMCs were kept in an incubator infused with

5% CO2/95% air at 37°C, and passaged when at 60-80% confluency. Human PASMCs at passage 5-12 were used in the present study. In proliferation studies, human PASMCs were cultured in 12-well plates (1×104 cells/well) in SMGM-2 containing either DIDS (final concentration=1, 3, 10, 30, 100, or 300 μM), DCPIB (final concentration= 0.3, 1, 3, 10, 30, or 100 μM), or vehicle (DMSO or ethanol). After culture for 4 days, cell number in each well was determined by cell counting. In hypoxia study, human PASMCs were seeded in 12-well plates 4 (1×10 cells/well) in SMGM-2 and initially maintained in 5% CO2/95% air at 37°C for overnight. Cells were then transferred to a humidified Plexiglas chamber (Billups Rothberg, San

Diego, CA) maintained at 37 °C and continuously flushed with a gas composed of 1% O2/5%

CO2/balance N2 (Ray et al., 2008). Normoxic control cells were maintained in 5% CO2/95% air incubator under otherwise identical conditions.

2.9 Immunocytostaining of CLC-3 in Human PASMCs

To examine the subcellular localization of CLC-3 proteins, immunocytostaining studies were performed in human PASMCs with anti-CLC-3 antibodies. Human PASMCs were seeded on laminin-coated coverslips and cultured in SMGM-2 for overnight. Cells were washed with phosphate-buffered solution (PBS) and then fixed and permeabilized by incubation in 3.7% paraformaldehyde solution (in PBS) supplemented with 1.8% Triton-X at 37°C for 30 min. Cells were washed with PBS, and incubated in blocking buffer (1% bovine serum albumin in PBS) at 37°C for 30 min. Blocking buffer was removed, and cells were incubated with anti-CLC-3 antibody (Alamone Labs, Jerusalem, Israel) at 1 µg/ml in blocking buffer at 4°C for overnight. Anti-CLC-3 antibody solution was removed and cells were washed with blocking buffer at room temperature for 5 min × 3 times. Cells were incubated with a Fluor 480-congugated goat anti- rabbit antibody solution (Molecular Probes, 1:500 dilution) at room temperature for 1 h. The antibody solution was removed and cells were washed with blocking buffer at room temperature for 5 min × 3 times. The coverslip with cells was inverted onto a slide containing 10 µL mounting media and mounted on the stage of an Olympus microscope (IX51). Fluorescent cell images (excitation 488 nm, emission 525±25 nm) were obtained with a confocal unit consisting of a spinning disk (CSU-X1, YOKOGAWA), a QuantEM 512 camera (Photometrics) and a 100× oil immersion objective (Olympus).

2.10 Quantitative Real-Time Reverse Transcription-PCR (qRT- PCR)

qRT-PCR is a technique for detection and quantification of gene expressions (mRNA abundance) in tissue or cell samples. In this technique, RNA is first extracted and purified from samples and reverse transcribed (RT) to complementary DNA (cDNA). A segment of the cDNA of interest is amplified in a PCR reaction with specific oligonucleotide primers. The amount of PCR product is monitored with a fluorescent dye after each PCR cycle (i.e. in real time). The number of PCR cycles required for the fluorescence to reach a set level is related to the input

44 amount of template cDNA in each sample. This information is used for derivation of the absolute copy number or relative amount of mRNA of interest in test samples when proper control samples are included. In our studies, total RNA was extracted from control or lentivirus- transduced human PASMCs at 60-80% confluency with the Protein and RNA Isolation System (Applied Biosystems, Foster City, CA) (He et al., 2010). RNA samples were treated with DNase (TURBO DNA-free, Applied Biosystems) to eliminate residual genomic DNA. RNA samples were quantified with a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific Inc., Wilmington, DE) and 1 µg RNA was used to synthesize the first-strand cDNAs with the iScript cDNA Synthesis Kit containing both oligo(dT) and random primers (Bio-Rad Laboratories, Hercules, CA). Quantitative real-time PCR was performed in triplicate for each sample using SYBR green reagent (Applied Biosystems) with the ABI PRISM 7900 HT sequence detection system (Applied Biosystems). Primers for Cl- channel genes were designed using Primer3 program (http://frodo.wi.mit.edu/primer3/) and sequences are given in Tables 4.1 and 4.2. Data were normalized to 28S ribosomal RNA levels.

2.11 Lentivirus Generation

pGIPZ constructs encoding microRNA-adapted shRNAs targeting human CLC-3 were obtained in E. coli stocks from Open Biosystems (Huntsville, AL) and purified using EndoFree Plasmid Maxi Kit (Qiagen Inc., Mississauga, Ontario). Control shRNAs included a non-silencing shRNA in a pGIPZ vector (Open Biosystems) and a shRNA targeting firefly luciferase in a pPRIME vector (generously provided by Dr. S. J. Elledge). The target sequences of these shRNAs were given in Table 4.3. Lentiviruses were generated by transfecting HEK293T cells (passage 4-9, 1×106 cells/60-mm dish) with 2 µg shRNA-encoding plasmids, 2 µg pMD2.G and 0.2 µg pCMV-dR8.91 using 12 µL FuGENE6 transfection reagent (Roche Applied Science, Indianapolis, IN). Lentiviruses were collected 48 h and 72 h post-transfection and stored at -80°C or immediately used for transducing human PASMCs. Titers were determined by counting number of GFP-positive HEK293T cells infected with diluted lentiviruses. Lentivirus generation and transduction of human PASMCs (see below) were conducted in a Biosafety Level 2 facility.

45 2.12 Transduction of Human PASMCs

Human PASMCs were cultured in 12-well plates (2×104 cells/well) for overnight. Cells were treated with lentivirus at a MOI (multiplicity of infection) of 10-20 together with 4 µg/ml polybrene (Sigma), centrifuged at 1,800 RPM at 32°C for 45 min, and then incubated at 37°C. Virus medium was replaced with fresh virus-free medium at 24 h post-infection. Puromycin (2 µg/ml, Sigma) was added to medium at 48 h post-infection to kill uninfected cells. After 4 days of puromycin treatment, ~100% of the cells were GFP-positive and cells were maintained in puromycin-free medium for 3-5 days before RNA extraction, proliferation, and patch-clamp studies.

2.13 Recording of Membrane Potential by Whole-Cell Patch- Clamp Technique in Rat PASMCs

The whole-cell patch-clamp technique using the nystatin-perforated recording configuration was used in current-clamp mode to record membrane potential (Jiang et al., 2003). In this technique, the current through the cell membrane is kept at zero by the varying output voltage of an amplifier and this voltage is recorded as membrane potential (Em). The nystatin- perforated method was used because it does not disturb cell interior and allows Em recording in a more physiological condition. In our studies, rat PASMCs were superfused with physiological salt solution (PSS), containing (in mM): NaCl 135, KCl 5, CaCl2 1.8, MgCl2 1.2, HEPES 10, and glucose 10 (pH=7.4 with NaOH), at a rate of 2 ml/min at 22oC. Pipettes were pulled from borosilicate glass tubes (World Precision Instruments Inc., Sarasota, Florida, U.S.A.) with a Flaming/Brown micropipette puller (P-97, Sutter Instrument Company, Novato, CA, U.S.A.) and had a resistance of 2-4 M Ohms when filled with pipette solution (in mM: KCl 30, K+-gluconate

118, MgCl2 1, HEPES 10 and nystatin 200 μg/ml, pH=7.2 with KOH). Fresh nystatin-containing pipette solution was made every 2 h. A reference Ag-AgCl electrode was embedded in a 3 M KCl agar bridge located in the bath solution. Membrane potential was recorded using an AxoPatch 200B amplifier (Axon Inc., CA) and pClamp 8.0 software (Axon Inc., CA). Data were sampled at 100 Hz and filtered at 1 kHz. To determine the contribution of Cl- conductance to - resting Em, the effect of Cl channel inhibition was assessed by perfusing cells with bath solutions containing DIDS (1-300 μM), made from a 50 mM stock solution (in DMSO), after Em had reached a steady state. Similarly, the effect of extracellular Cl- replacement with iodide (I-)

46 on Em was evaluated by replacing 135 mM NaCl with equimolar NaI in bath solution. The voltage errors due to bath solution exchanges were minimized by the use of a 3 M KCl agar bridge. Liquid junction potential (LJP) between pipette solution and bath solution, calculated with pClamp 8.0 software, was 13 mV and membrane potential data were presented as LJP- corrected values (Em=Epipette–13 mV).

2.14 Recording of Cl- Currents by Whole-Cell Patch-Clamp Technique in Rat and Human PASMCs

- The Cl currents (ICl) active in rat PASMCs were examined in nystatin-perforated cells with patch-clamp technique in the voltage-clamp mode using bath and pipette solutions designed to minimize cation–mediated currents. In voltage clamping, the cell membrane voltage is kept at fixed values by injecting currents with a resistive feedback circuit of the patch-clamp amplifier, and the injected currents are recorded as the membrane ionic currents (but opposite in polarity). In our studies, the bath solution contained (in mM): N-methyl-D-glucamine-HCl (NMDG-HCl)

132, MgCl2 1.5, HEPES 10, and glucose 10 (pH=7.4 with NMDG). The pipette solution + contained (in mM): CsCl 30, Cs -gluconate 118, MgCl2 1, HEPES 10 and nystatin 200 μg/ml (pH=7.2 with CsOH). Holding potential was set at –40 mV, which was the equilibrium potential for chloride (ECl) in the study. Voltage steps (1 second duration) from –100 mV to +60 mV were applied in 20-mV increments. Series resistance was compensated by 70-80%. Data were sampled at 10 kHz and filtered at 1-2 kHz.

2+ Ca -independent, baseline and swelling-activated ICl were recorded in rat and human PASMCs with the conventional ruptured-patch configuration, using a pipette solution containing (in mM): NMDG-Cl 30, NMDG-gluconate 93, HEPES 5, EGTA 10, and Mg-ATP 5 (pH=7.2 2+ with NMDG, free Ca <1 nM). The bath solution contained (in mM): NMDG-HCl 100, MgCl2 1.5, HEPES 10, glucose 10, and D-mannitol 70 (pH=7.4 with NMDG, estimated osmolarity

=295 mOsm). To record swelling-activated ICl, a hypotonic bath solution lacking D-mannitol (estimated osmolarity=225 mOsm) was used to induce cell swelling. In rat PASMCs, ion selectivity of swelling-activated ICl was assessed using a pipette solution containing (in mM): NMDG-Cl 123, HEPES 5, EGTA 10, and Mg-ATP 5 (pH=7.2 with NMDG), and bath solutions containing 100 mM gluconate or 100 mM I- made by equimolar replacement of Cl-. Currents were elicited by voltage ramps (1 second duration) from –120 to +100 mV. Anion permeability

47 ratios of gluconate or I- versus Cl- were estimated from the Goldman-Hodgkin-Katz voltage equation with shifts of reversal potential.

2+ 2+ To stimulate Ca -activated ICl in rat PASMCs, a pipette solution containing a free Ca concentration of 500 nM was used. The standard ICl recording bath solution (containing 132 mM NMDG-HCl and 0 mM D-mannitol) was used while the pipette solution contained (in mM):

NMDG-Cl 30, NMDG-gluconate 93, HEPES 5, EGTA 10, CaCl2 7.7, MgCl2 0.7, and Mg-ATP 5 (pH=7.2 with NMDG, free Ca2+ concentration=500 nM at 22C, calculated with winmaxc32 2+ 2+ software). Ca -activated ICl were taken as the difference between ICl recorded with 500 nM Ca and <1 nM Ca2+ pipette solutions.

2.15 Antibodies and Reagents

Anti-α smooth muscle actin antibody was from Abcam Inc. (Cambridge, MA); anti-BrdU antibody was from Roche Inc. (Mississauga, ON, Canada); anti-CLC-3 antibody was from Alomone Labs (Jerusalem, Israel); isoflurane was from Halocarbon Laboratories (River Edge, NJ); HB-MEM, fungizone and gentamycin were from GIBCO (Invitrogen Canada Inc., Burlington, ON, Canada); collagenase and elastase were from Worthington Biochemical Inc.(Lakewood, NJ); FBS was from Hyclone (Logan, UT); DCPIB (4-[(2-Butyl-6,7-dichloro-2- cyclopentyl-2,3-dihydro-1-oxo-1H-inden-5-yl)oxy] butanoic acid) was from Tocris Bioscience (Ellisville, MO). Unless otherwise indicated, all other chemicals were from Sigma Chemicals Inc. (St. Louis, MO).

2.16 Statistical Analysis

Data are expressed as mean ± SEM for n number of observations with p<0.05 considered significant. Differences between two means were evaluated by two-tailed Student’s t-test. Differences among multiple means were assessed by one-way analysis of variance (ANOVA). When significance was detected by ANOVA, differences among individual means were evaluated post hoc by Bonferroni’s test.

Chapter 3

Regulation of Proliferation and Membrane Potential by Cl- Currents in Rat Pulmonary Artery Smooth Muscle Cells

A version of this appeared in:

Wenbin Liang, Julie B. Ray, Jeff Z. He, Peter H. Backx and Michael E. Ward. Regulation of proliferation and membrane potential by chloride currents in rat pulmonary artery smooth muscle cells. Hypertension, 2009;54:286-293

Attributions:

Histological experiments were performed by Dr. Julie B. Ray; flow cytometric experiments were performed in collaboration with Dr. Jeff Z. He; other experiments were performed by W. Liang.

3 Regulation of Proliferation and Membrane Potential by Cl- Currents in Rat Pulmonary Artery Smooth Muscle Cells

3.1 Abstract

Pulmonary artery smooth muscle cell (PASMC) proliferation contributes to increased pulmonary vascular resistance in pulmonary arterial hypertension. Since proliferation is dependent on - membrane potential (Em) and since Em is, in part, determined by Cl currents (ICl), we examined the effects of ICl inhibition with DIDS on cultured rat PASMCs. DIDS (30 μM) reduced cell numbers, decreased BrdU incorporation and delayed cell cycle progression. ICl inhibition with NPPB (100 μM) also reduced cell numbers of cultured rat PASMCs. To test the possible involvement of ICl in the regulation of PASMC proliferation, we measured Em and ICl in both cultured (proliferating) and acutely dissociated (non-proliferating) rat PASMCs. Em (–39.3±1.4 - mV) was close to equilibrium potential of Cl (ECl) (–39 mV) in proliferating PASMCs but differed from ECl in acutely dissociated cells (–45.3±0.9 mV). DIDS and substitution of - - extracellular Cl with I induced Em hyperpolarization in proliferating but not non-proliferating 2+ PASMCs. Consistent with Em recordings, DIDS-sensitive baseline and swelling-activated (Ca - 2+ independent) ICl, recorded with low Ca (<1 nM) pipette solutions, were ~5-fold greater in 2+ proliferating than in non-proliferating PASMCs. By contrast, Ca -activated ICl did not differ 2+ between proliferating and non-proliferating PASMCs. Ca -independent ICl were also increased in proliferating PASMCs acutely dissociated from rats exposed to hypoxia (10% O2, 7 days).

These findings are consistent with the conclusion that ICl regulate proliferation of PASMCs and suggest that selective ICl inhibition may be useful in treating pulmonary arterial hypertension.

Key Words: Ca2+-independent chloride currents; pulmonary arterial hypertension; hypoxia; cell cycle; phenotypic transition

3.2 Introduction

Pulmonary arterial hypertension, uniformly fatal in its primary idiopathic form and a cause of premature mortality when it complicates cardiopulmonary disease, remains a difficult management problem. Pulmonary artery smooth muscle cell (PASMC) proliferation is a hallmark of the disease and contributes to increased vascular resistance to blood flow (Humbert et al., 2004; McLaughlin et al., 2006). Understanding the mechanisms that regulate PASMC proliferation is necessary in order that new targets for therapeutic intervention may be identified.

- 2+ Swelling-activated Cl currents (ICl) (Zhong et al., 2002) and Ca -activated ICl (Angermann et al., 2006) have been identified in PASMCs, and PASMC expression of CLC-3, a candidate for mediation of swelling-activated ICl (Duan et al., 1997) and important for cell proliferation (Wang et al., 2002; Habela et al., 2008), is upregulated in the pulmonary hypertensive rats (Dai et al., 2005). Because of the transmembrane Cl- concentration gradients that exist in vascular smooth muscle, activation of Cl- channels causes Cl- efflux, cell volume decrease (Kunzelmann, 2005) and plasma membrane depolarization (Aickin, 1990; Davis et al., 1997). Moreover, previous studies have identified that cytoplasmic volume must be adjusted during cell cycle progression (Kunzelmann, 2005) and that membrane potential and intracellular free Ca2+ levels are important regulators of proliferation (Platoshyn et al., 2000; Kunzelmann,

2005). These observations suggest a role for ICl in PASMC proliferation. The present study was, therefore, carried out to test the hypothesis that ICl participate in regulation of PASMC proliferation and to compare the ICl active in proliferating versus contractile PASMCs.

3.3 Summary of Materials and Methods

Please refer to Chapter 2 for detailed methods.

All procedures were approved by the institutional animal care committee. Single pulmonary artery smooth muscle cells (PASMCs) were isolated (Archer et al., 1996) from small intralobar arteries (3rd and 4th order branches) of normoxic rats (adult, male, Sprague-Dawley) and rats exposed to hypoxia (10% O2, 7 days). These cells were cultured in the presence of 10%

FBS as described (Yuan, 1995). The effects of an ICl inhibitor (DIDS, 30 µM) on BrdU incorporation and DNA content were assessed by flow cytometry as described (Ray et al., 2008).

Resting membrane potential (Em) of rat PASMCs was measured using the whole-cell patch- clamp technique in current-clamp mode using the nystatin-perforated recording configuration

(Jiang et al., 2003) with bath solution containing (in mM): NaCl 135, KCl 5, CaCl2 1.8, MgCl2 1.2, HEPES 10, and glucose 10 (pH=7.4) and pipette solution containing (in mM): KCl 30, K+- gluconate 118, MgCl2 1, HEPES 10 and nystatin 200 μg/ml (pH=7.2). ICl were recorded with whole-cell patch-clamp technique in voltage-clamp mode in nystatin-perforated or conventional ruptured recording configuration, with bath solution containing (in mM): N-methyl-D-glucamine

(NMDG)-HCl 132, MgCl2 1.5, HEPES 10, and glucose 10 (pH=7.4 with NMDG) and pipette solution containing (in mM): NMDG-HCl 30, NMDG-gluconate 93, HEPES 5, EGTA 10, and Mg-ATP 5 (pH=7.2 with NMDG). In some studies, NMDG-HCl in the bath solution was reduced to 100 mM (estimated osmolarity=225 mOsm) to induce cell swelling. In other studies 2+ Ca -activated ICl were stimulated by adding 7.7 mM CaCl2 to the pipette solution (free Ca2+=500 nM). Series resistance was compensated by 70-80%. Data were sampled at 10 kHz and filtered at 1-2 kHz. Liquid junction potential between pipette solution and bath solution in Em recordings was calculated and corrected. Differences between two means were evaluated by two- tailed Student’s t-test. Differences among multiple means were assessed by one-way ANOVA and evaluated post hoc by Bonferroni’s test.

3.4 Results

To determine if ICl are involved in smooth muscle cell proliferation, we examined the effects of an ICl inhibitor, DIDS, on cell numbers in cultured rat PASMCs. As summarized in Figure 3.1A, DIDS reduced (p<0.01, n=6) cell numbers in a dose- and time-dependent manner. Cell viability, as assessed using trypan blue exclusion, was not affected by DIDS at 10 or 30 μM, but was reduced (p<0.01) by ~10% in cells treated with 100 μM DIDS (Figure 3.2A). DIDS (30 μM) had no effect on the percentage of cells staining positive for Annexin V (Figure 3.2B), suggesting the lack of effect of DIDS on apoptotic cell death at this concentration. To further confirm that DIDS inhibits cell replication, BrdU incorporation was measured. After treatment for 16 hours, DIDS decreased (p<0.05, n=3) the percentage of BrdU positive cells by 36%

(Figure 3.1B). Importantly, DIDS also increased (p<0.05, n=3) the percentage of cells in G0 and

G1 phases while decreasing (p<0.05) the percentage of cells in G2/M phase (Figure 3.1C), indicating that DIDS delayed G1/S phase transition. Thus, the changes in cell numbers of

Figure 3.1 DIDS (an ICl inhibitor) inhibits proliferation, BrdU incorporation, and cell cycle progression in rat PASMCs. A. Left panel, Effects of incubation with DIDS for 4 days on cell numbers of cultured rat pulmonary artery smooth muscle cells (PASMCs); Cell numbers were reduced by 20.4%, 37.8% and 92.3% by DIDS at 10, 30 and 100 µM, respectively. Right panel, Effects of incubation with DIDS (30 µM) for 3, 4, and 5 days on cell numbers of cultured rat PASMCs. B and C. Effects of incubation with DIDS (30 µM) for 16 h on numbers of BrdU positive cells (B) and cell cycle phase distribution (C) in cultured rat PASMCs. *p<0.05; #p<0.01 vs. corresponding values in vehicle-treated cells.

Vehicle 10 M DIDS A 30 M DIDS 100 M DIDS 100 *

50 Viable cells (%)

Figure 3.2 DIDS at low dosages has no effects on rat PASMC death or apoptosis. A. Viability of rat PASMCs assessed with trypan blue exclusion method after 4-day culture in the presence of 10, 30, or 100 μM DIDS or vehicle (DMSO). *p<0.01 vs. vehicle-treated cells. B. Flow cytometric analysis of rat PASMCs for Annexin V staining (an index of apoptosis) and propidium iodide staining after 4-day culture in the presence of DIDS (30 μM) or vehicle (DMSO).

2.5 ) 2 2.0 /cm 4

1.5

1.0

0.5 Cell number (x10 Cell number

0.0 Vehicle 0.1 M 1M 10 M [CFTR -172] inh

Figure 3.3 CFTR inhibitor (CFTRinh-172) does not affect rat PASMC proliferation. Effects of incubation with a CFTR inhibitor, CFTRinh-172 (IC50, 0.3 µM), for 4 days on cell numbers of cultured rat pulmonary artery smooth muscle cells (PASMCs). Cell numbers were not different among the groups (p=0.28, n=4 with ANOVA).

PASMCs treated with DIDS at concentration of 30 μM (or less) were due to effects on PASMC proliferation and not changes in cell survival. Since DIDS affects multiple ion flux mechanisms, we confirmed the effects of ICl blockade on rat PASMC proliferation by testing another ICl blocker, NPPB, which is structurally different from DIDS. NPPB (100 μM) treatment reduced (p<0.01, n=5) cell numbers of rat PASMCs by 68.9% (data not shown) without affecting cell viability (97.4±0.5% vs. vehicle-treated 97.2±1.0%, p=0.87). In contrast, a specific inhibitor of - CFTR Cl channel, CFTRinh-172 (IC50, 0.3 µM) (Ma et al., 2002), did not affect (p=0.28, n=4) cell numbers of rat PASMCs at concentrations up to 10 µM (Figure 3.3), suggesting that CFTR- independent Cl- conductances are involved in rat PASMC proliferation (although it remains to be determined whether CFTR is expressed and contributes to plasma membrane Cl- conductance in rat PASMCs).

To explore the ionic mechanisms responsible for reduced proliferation of DIDS-treated rat PASMCs, we used nystatin-perforated whole-cell recordings to investigate the effects of

DIDS on membrane potential (Em) and ICl in PASMCs. In cultured rat PASMCs, the resting Em was –39.3±1.4 mV (n=78 cells) which was close to the equilibrium potential of Cl- in the study - (ECl = –38.9 mV, calculated with Nernst equation from [Cl ] in pipette and bath solutions).

Figure 3.4 shows that, DIDS addition induced a concentration-dependent Em hyperpolarization in cultured rat PASMCs. At maximum levels of DIDS Em was shifted (p<0.05, n=6) to more negative potentials by ~30 mV and the estimated DIDS concentration eliciting a 50% change in

Em was ~20 μM. By comparison, the effect of DIDS on acutely dissociated (non-proliferating) cells was small, with the maximal Em shift to more negative potentials being ~6 mV at saturating DIDS levels.

The results above indicate that the resting Em is more strongly regulated by ICl in cultured PASMCs than in acutely dissociated PASMCs. Consistent with this suggestion, Figure 3.5 shows that replacing extracellular Cl- with I- (an anion that is typically more permeant than Cl- for most VSMC Cl- channels (Large et al., 1996; Yamazaki et al., 1998)) induced a ~10 mV hyperpolarization shift (p<0.01, n=6) of Em in cultured PASMCs. In contrast, in acutely - - - dissociated cells, Cl replacement with I did not alter (p=0.46) Em (–48.5±2.2 mV before I , and - –50.7±1.9 mV after I , n=11). These findings indicate that ICl are larger in proliferating than in non-proliferating PASMCs.

A Cultured Acutely dissociated DIDS (M) DIDS (M) 30 100 10 30 100 300

60 mV 60 mV

10 mV 2 min B

Cultured Acutely dissociated -40 * *

-20 *

0 1 10 100 Membrane potential changes (mV) [DIDS] (M)

Figure 3.4 DIDS causes greater membrane hyperpolarization in cultured than in acutely dissociated rat PASMCs. A. Representative recordings, showing membrane hyperpolarization after treatment with DIDS in cultured (left panel) and acutely dissociated (right panel) rat pulmonary artery smooth muscle cells (PASMCs). B. Concentration-response relationship for DIDS-induced membrane hyperpolarization in cultured (filled circles, n=4–7 cells) and acutely dissociated (open circles, n=4–11 cells) rat PASMCs. Solid and dashed lines represent fitted relationships from which maximum hyperpolarization (29.3±0.2 mV for cultured and 6.3±1.0 mV for acutely dissociated cells) and IC50 values were calculated. *p<0.01 for difference from corresponding values in acutely dissociated cells.

Cultured I- substitution Acutely dissociated

I- substitution 40mV 40mV

10 mV

1 min B

-15 Cultured * Acutely dissociated

-10 al changes(mV)

-5 #

0 Membrane potenti

Figure 3.5 Iodide substitution causes greater membrane hyperpolarization in cultured than in acutely dissociated rat PASMCs. A. Representative recordings showing membrane potential changes after substitution of external Cl- with iodide (I-) in cultured (left panel) and acutely dissociated (right panel) rat PASMCs. B. I- substitution-induced membrane potential changes in cultured (–10.2±4.2 mV, n=6 cells) and acutely dissociated (n=11 cells) rat PASMCs. *p<0.01 vs. values before I- substitution, #p<0.01 vs. changes in cultured cells.

To more directly investigate ICl in these cells, nystatin-perforated patch-clamp recordings of PASMCs were made under conditions where cation currents were minimized (see Methods). As shown in Figure 3.6A, voltage steps from a holding potential of –40 mV induced currents with "instantaneous" activation kinetics and with minimal inactivation during a one-second voltage step. The relationship between current density and voltage in cultured PASMCs showed prominent outward rectification with an estimated reversal potential (Erev) of –32.9±1.9 mV

(corrected for liquid junction potential, n=15, Figure 3.6B), which was similar to ECl (–38.9 mV). The addition of 100 μM DIDS reduced (p<0.05) the current density at +60 mV by 86.0±7.7% in cultured PASMCs (from 12.2±2.9 to 1.69±0.96 pA/pF, n=4), confirming that the currents are generated by DIDS-sensitive Cl- channels. By comparison, in non-proliferating PASMCs acutely dissociated from normoxic rats (Figure 3.6A, right panel) the current-voltage relationship showed little rectification, Erev (–56.1±1.8 mV) differed (p<0.01) from ECl and the current density at +60 mV (1.46±0.16 pA/pF, n=11) was lower (p<0.01) than that in cultured PASMCs.

The currents recorded above were measured under basal (i.e. isotonic) conditions.

Previous studies have shown that swelling-activated ICl, which are sensitive to DIDS (Okada, 1997), are important for cell cycle progression (Wondergem et al., 2001), suggesting that blockade of swelling-activated ICl may underlie the reduced proliferation observed in DIDS- treated cultured PASMCs. Therefore, ICl recordings were made in response to cell swelling induced by the application of hypotonic solutions using the ruptured patch-clamp configuration and the pipette solutions contained <1 nM free Ca2+. The current density at +60 mV of the baseline currents recorded in cultured PASMCs in isotonic conditions (without cell swelling) was 2.2±0.7 pA/pF (n=7). Switching to hypotonic bath solutions caused swelling of these cells 2+ and induced large outwardly rectifying (Ca -independent) ICl (Figure 3.7A middle panel) with a density at +60 mV of 65.6±20.4 pA/pF (n=6). Consistent with the conclusion that these swelling- activated currents were ICl, DIDS (100 μM) reduced (p<0.05) the current density at +60 mV to 5.0±1.7 pA/pF (n=4, Figure 3.7A). Moreover, analysis of the DIDS-sensitive currents, calculated as the difference between the currents before and after DIDS, showed an Erev of –29.5±2.2 mV

(n=4), which was close to ECl (–29.2 mV).

To further establish that the swelling-activated currents recorded in cultured PASMCs - were ICl, we measured the changes in Erev with external Cl replacements.

59 A

Cultured 250 pA Cultured Acutely 200 ms + DIDS dissociated

B I(pA/pF) 15 Cultured * Cultured + DIDS Acutely dissociated 10 * 5 * *

-120 -80 -40 40 80 * V (mV) * -5 * *

Figure 3.6 Cl- currents recorded with nystatin-perforated method in rat PASMCs. A. representative Cl- currents in cultured (left panel) and acutely dissociated (right panel) rat PASMCs; middle panel shows effect of DIDS (100 µM) on currents in cultured PASMC. Dashed lines indicate the position of zero current. Holding potential was set at –40 mV, and voltage steps (1 second duration) from –100 mV to +60 mV were applied in 20-mV increments. Series resistance was compensated by 70-80%. B. Current-voltage relationships for Cl- currents in cultured (filled circles, n=15 cells, capacitance=30.3±3.7 pF) and acutely dissociated (triangles, n=11 cells, capacitance=10.4±1.3 pF) rat PASMCs as well as for currents in cultured cells in the presence of 100 µM DIDS (open circle, n=4 cells). *p<0.01 vs. corresponding values in acutely dissociated, untreated cells.

60 A cell683/4154 cell683/4160 cell683/4162 1000 pA Hypotonic solution 200 ms Hypotonic solution Isotonic solution + DIDS

B I (pA/pF) 150 *

Isotonic * 100 Hypotonic * Hypotonic + DIDS

50 * * * -120 -80 -40 40 80 120 * * V (mV) * -50

Figure 3.7 Baseline and hypotonic solution-induced Ca2+-independent Cl- currents in cultured rat PASMCs. Currents were recorded with ruptured whole-cell configuration with pipette solutions containing <1 nM free Ca2+. A. Cl- currents recorded in isotonic (left panel) and hypotonic (middle panel) bath solutions and in hypotonic solution in the presence of 100 µM DIDS (right panel). Dashed lines indicate the position of zero current. B. Current-voltage relationships for Cl- currents in isotonic (filled circles, n=7 cells) and hypotonic (open circles, n=6 cells) solutions and in hypotonic solution in the presence of 100 µM DIDS (squares, n=4 cells). *p<0.01 vs. values in isotonic solution.

A: Swelling-activated currents

I (pA) 2000 I (pA) 8000 - Cl I-

1000 4000 Cl- Gluconate

-120 -80 -40 40 80 120 -120 -80 -40 40 80 120 V (mV) V (mV) -1000 -4000

-2000 -8000

B: Baseline currents

I (pA) 300 I (pA) 300 I- Cl- 150 150 Cl- Gluconate

-120 -80 -40 40 80 120 -120 -80 -40 40 80 120 V (mV) V (mV) -150 -150

-300 -300

Figure 3.8 Effects of anion substitutions on reversal potentials of swelling-activated currents and baseline currents in rat PASMCs. Representative swelling-activated currents (panel A) and baseline currents (panel B) recorded in cultured rat PASMCs with pipette solutions containing <1 nM free Ca2+ in hypotonic (panel A) or isotonic (panel B) external solutions that were Cl--rich (all panels, black traces), gluconate-rich (left panels, blue traces), or I--rich (right panels, red traces). Currents were elicited by voltage ramps from –120 to +100 mV. The permeability ratio of I- versus Cl-, estimated from the Goldman-Hodgkin-Katz voltage equation with shifts of reversal potential, for the baseline currents (7.7) was greater (p<0.01) than that for swelling-activated currents (2.8).

When external Cl- was replaced with gluconate, an anion that is relatively impermeant to Cl- channels (Okada, 1997; Chen et al., 2004), Erev shifted (p<0.01) to more positive potentials (by

35.8±2.7 mV, n=8) (Figure 3.8A, left panel). Conversely, Erev shifted (p<0.01) to more negative potentials (by 23.3±4.2 mV, n=7) when external Cl- was replaced with I- (Figure 3.8A, right panel). The anion permeability ratios for gluconate and I- versus Cl- were 0.23±0.03 and 2.8±0.47, respectively, consistent with the sequence of anion permeability (i.e. I->Cl->gluconate) - reported for swelling-activated ICl (Okada, 1997). Cl replacements also caused similar shifts in

Erev in baseline currents recorded in isotonic bath solutions (Figure 3.8B).

In PASMCs acutely dissociated from normoxic rats, switching to hypotonic bath solutions also caused cell swelling but did not increase (p=0.17) the currents (Figure 3.9A): current density at +60 mV was 0.33±0.04 pA/pF (n=26) in isotonic solutions and 0.23±0.03 pA/pF (n=13) in hypotonic solutions. Moreover, the currents recorded in hypotonic solutions were not affected (p=0.80, n=5) by DIDS (data not shown). The above results establish that proliferating rat PASMCs display prominent baseline and swelling-activated Ca2+-independent

ICl which are virtually absent in acutely dissociated, non-proliferating PASMCs.

2+ PASMCs also possess prominent Ca -activated ICl which appear to play roles in cell 2+ proliferation (Klausen et al., 2007) and in regulating Em (Yuan, 1997). To assess Ca -activated

ICl in our rat PASMCs, ruptured patch-clamp recordings were made using pipette solutions containing 500 nM free Ca2+ and using isotonic bath solutions. As shown in Figure 3.10A, prominent currents were observed under these conditions in both cultured and acutely 2+ 2+ dissociated PASMCs. These currents contained both Ca -activated ICl and baseline Ca - independent ICl. The current density at +60 mV was greater (p<0.01) in cultured PASMCs (10.4±2.8 pA/pF, n=15) than that in acutely dissociated PASMCs (3.5±0.7 pA/pF, n=24) (Figure 2+ 3.10B). However, subtraction of the baseline Ca -independent ICl (ICl recorded in isotonic bath solutions in Figures 3.7 and 3.9) from these currents recorded with high pipette Ca2+ (500 nM) 2+ revealed that the Ca -activated ICl were not different (p>0.30, Figure 3.10C) as assessed using the parameters originated from Boltzmann fits.

Since hypoxia is a potent stimulus for PASMC proliferation in vivo (Niedenzu et al., 1981; Stenmark et al., 2006), we hypothesized that exposure of rats to hypoxic conditions will 2+ induce elevation in Ca -independent ICl in PASMCs. Consistent with previous studies

(Niedenzu et al., 1981), exposure of rats to hypoxia (10% O2) for 7 days induced proliferation of

63 PASMCs, as demonstrated by the increased BrdU uptake of PASMCs in hypoxic rats compared to normoxic rats (data collected by Dr. Julie B. Ray, not presented here). Consistent with our 2+ 2+ hypothesis, Ca -independent ICl recorded with low Ca (<1 nM) pipette solutions in PASMCs acutely dissociated from hypoxia-exposed rats (Figure 3.11) had current densities at +60 mV of 0.9±0.2 pA/pF (n=12) in isotonic bath solutions and 4.3±0.3 pA/pF (n=7) following the application of hypotonic solutions to induce cell swelling. Both baseline and swelling-activated 2+ Ca -independent ICl were greater (p<0.01) than the corresponding currents recorded in PASMCs acutely dissociated from normoxic rats (Figure 3.9).

64 A 676/4120

25 pA 200 ms Isotonic solution Hypotonic solution

B I (pA/pF)

Isotonic 0.5 Hypotonic

-120 -80 -40 40 80 120 V (mV)

-0.5

Figure 3.9 Hypotonic solution perfusion does not activate Ca2+-independent Cl- currents in acutely dissociated rat PASMCs. Currents were recorded with ruptured-patch whole-cell configuration with pipette solutions containing <1 nM free Ca2+ in PASMCs acutely dissociated from normoxic rats. A. Representative currents recorded in isotonic (left panel) and hypotonic (right panel) bath solutions. Dashed lines indicate the position of zero current. B. Current-voltage relationships of currents in isotonic (filled circles, n=26 cells) and hypotonic (open circles, n=13 cells) solutions.

cell570/3812_filtered cell663/4078_filtered A Cultured Acutely dissociated 200 pA 200 ms

B C I (pA/pF) 15 I (pA/pF) * 4

Cultured 10 Cultured Acutely dissociated Acutely dissociated

* 2 5 * -120 -80 -40* 40 80 -80 -40 40 80 V (mV) Corrected V (mV) * *

Figure 3.10 Ca2+-activated Cl- currents in cultured and acutely dissociated rat PASMCs. A. Representative Cl- currents recorded with ruptured-patch whole-cell patch-clamp method with pipette solutions containing 500 nM free Ca2+ in cultured (left panel) and acutely dissociated (right panel) rat PASMCs. Dashed lines indicate the position of zero current. B. Current-voltage relationships for Cl- currents recorded with pipette solutions containing 500 nM free Ca2+ in cultured (filled circles, n=15 cells) and acutely dissociated (open circles, n=24 cells) rat PASMCs. *p<0.01 vs. values in acutely dissociated cells. C. Current-voltage relationships of Ca2+-activated Cl- currents in cultured (filled circles) and acutely dissociated (open circles) rat PASMCs, which were derived by subtracting currents recorded with <1 nM Ca2+ pipette solutions and isotonic bath solutions (Figures 3.7 and 3.9) from currents recorded with 500 nM Ca2+ pipette solution (panel B in this Figure). Solid and dashed lines represent the fitted relationships, obtained with Bolzmann function, for cultured and acutely dissociated PASMCs, respectively.

66 A ell727/4359 cell727/4361

Isotonic solution 25 pA Hypotonic solution 200 ms

B I(pA/pF) 15 * Isotonic * Hypotonic 10 *

5 *

-120 -80 -40 40 80 120 V (mV) * -5 * * -10

Figure 3.11 Ca2+-independent Cl- currents in PASMCs acutely dissociated from hypoxic rats. Currents were recorded with ruptured whole-cell configuration with pipette solutions 2+ containing <1 nM free Ca in PASMCs from rats exposed to hypoxia (10% O2, 7 days). A. representative Cl- currents recorded in isotonic (left panel) and hypotonic (right panel) bath solutions. Dashed lines indicate the position of zero current. B. Current-voltage relationships for Cl- currents in isotonic (filled circles, n=12 cells) and hypotonic (open circles, n=4 cells) solutions. Dashed line indicates current-voltage relationship of currents recorded in hypotonic solution in PASMCs acutely dissociated from normoxic rats (same data as shown in Figure 3.9). *p<0.01 vs. values in isotonic solution.

67 3.5 Discussion

Vascular smooth muscle cells in mature mammals are highly differentiated and under normal physiological conditions show low rates of replication (Owens et al., 2004). However, these cells can transit from a non-proliferative, contractile phenotype to a proliferative "synthetic" phenotype (Owens, 1995; Miguel-Velado et al., 2005) and this phenotypic transition and subsequent proliferation in pulmonary arteries contribute to pulmonary arterial hypertension (Stenmark et al., 2006; Hassoun et al., 2009). Our study is the first to demonstrate that transition of PASMCs to a proliferative phenotype is associated with upregulation of DIDS-sensitive, Ca2+- independent ICl, and the application of DIDS reduces cell numbers of proliferating PASMCs. These effects on cell numbers by DIDS appeared to result from decreased proliferation rate, specifically from delayed cell cycle progression through the G1/S transition, since cultured proliferating rat PASMCs treated with DIDS accumulated in G0/G1 phase of the cell cycle and had reduced BrdU incorporation.

We conclude, for reasons outlined below, that the effects of DIDS on PASMC 2+ proliferation result from the blockade of ICl, and likely from blockade of DIDS-sensitive, Ca - independent "baseline" and "swelling-activated" ICl. First, the resting Em of proliferating

PASMCs was close to the ECl and DIDS induced a ~30 mV hyperpolarization of the Em, causing

Em to become closer to EK (–85 mV). By contrast, DIDS had very little effect on non- proliferating PASMCs. These results suggest that ICl play a more dominant role in setting Em in proliferating than in non-proliferating PASMCs. Consistent with this suggestion, Cl- replacement with I-, an anion with a higher permeability than Cl- for both baseline and swelling-activated 2+ Ca -independent ICl in rat PASMCs, induced a ~10 mV shift in Em to more hyperpolarized potentials while Cl- replacement with gluconate, an anion with a lower permeability than Cl-, shifted Em to more depolarized potentials (data not shown). These findings are also consistent with the anion permeability properties of swelling-activated ICl recorded in other tissues as reviewed (Nilius et al., 1996; Okada, 1997). Conversely, anion substitutions had virtually no effect on Em in acutely dissociated non-proliferating PASMCs. These results establish that ICl active in cultured proliferating PASMCs promote membrane depolarization, which can serve as a mitogenic stimulus (see below) (Kunzelmann, 2005).

Further support for a connection between elevated ICl and PASMC proliferation is provided by the results of our voltage-clamp studies which were designed to isolate ICl in whole-

68 cell recordings (see Methods). The ICl recorded in both isotonic bath solutions (i.e. baseline ICl) and hypotonic (osmolarity decrease of 70 mOsm) bath solutions (i.e. swelling-activated ICl) in 2+ ruptured-patch recordings with low pipette Ca levels (<1 nM), as well as ICl recorded in isotonic bath solutions with nystatin-patch method, were much larger in cultured proliferating PASMCs compared to acutely dissociated non-proliferating PASMCs. Moreover, consistent with our Em recordings, DIDS application reduced these currents by ~90% in cultured PASMCs, while having no effect in non-proliferating PASMCs. Our conclusion that these DIDS-sensitive

(baseline and swelling-activated) currents in proliferating PASMCs were indeed ICl is bolstered - - by the large shifts in Erev observed when extracellular Cl was replaced with I (negative shifts in

Erev) or gluconate (positive shifts in Erev).

Another important finding supporting the conclusion that ICl can regulate PASMC proliferation is our observation that exposure of rats to hypoxia for 7 days induced the proliferation of SMCs in the intact pulmonary artery and PASMCs acutely dissociated from these hypoxia-exposed rats showed prominent baseline and swelling-activated Ca2+-independent

ICl, which were virtually absent in non-proliferating PASMCs acutely dissociated from normoxic rats. Thus, ICl expression strongly correlates with the proliferation rates of PASMCs, whether the stimulus for proliferation is hypoxia in vivo or mitogens present in the serum of culture medium.

These findings also suggest that the reduced ICl in PASMCs acutely dissociated from normoxic rats are unlikely due to damage of ICl channel proteins during enzymatic isolation.

The mechanism whereby baseline and swelling-activated ICl regulate PASMC proliferation is unclear. Because ECl is typically above Em in SMCs (Aickin, 1990; Davis et al.,

1997), the presence of larger ICl in proliferating PASMCs is expected to promote depolarization.

Indeed, in the present study, the resting Em in proliferating cultured PASMCs is more depolarized (by ~6 mV) than in acutely dissociated PASMCs. Similar membrane depolarizations were observed in proliferating cultured (as compared to acutely dissociated) human uterine artery SMCs (Miguel-Velado et al., 2005), in PASMCs acutely dissociated from hypoxia- exposed rats (compared to cells from normoxic rats) (Osipenko et al., 1998) and in proliferating (as compared to growth arrested) rat PASMCs in culture (Platoshyn et al., 2000). Membrane 2+ depolarization as seen in proliferating PASMCs is predicted to increase [Ca ]i, via the presence of a “window current" for voltage-dependent Ca2+ channels which in SMCs has a steep voltage- dependence between –50 to –30 mV (Fleischmann et al., 1994). This Ca2+ entry has been shown to promote PASMC proliferation (Platoshyn et al., 2000), possibly as a result of the Ca2+-

69 dependent regulation of cell cycle regulators (Hardingham et al., 1997; Husain et al., 1997; Choi et al., 2006). In addition to inducing membrane depolarization, the larger swelling-activated ICl in proliferating PASMCs may facilitate the obligatory changes in cell volume during cell cycling - (Okada, 1997; Ullrich et al., 1997; Kunzelmann, 2005). Cl efflux via swelling-activated ICl is thought to protect against excessive cell swelling due to active uptake of amino acids and other substrates during G1 phase (Kunzelmann, 2005) and may play a similar role in proliferating PASMCs.

2+ Inhibition of Ca -activated ICl with DIDS (Large et al., 1996) may also contribute to the 2+ reduced proliferation in our cultured rat PASMCs, since Ca -activated ICl are expressed in these 2+ cells. The Ca -activated ICl were not different between proliferating and non-proliferating contractile PASMCs, suggesting that proliferation of PASMCs may not require elevation in these currents. However, because cytosolic [Ca2+] is higher in cultured PASMCs (Platoshyn et al., 2+ 2000), it is conceivable that Ca -activated ICl will be more active in proliferating cells. Although 2+ Ca -activated ICl regulate resting Em and vascular tone (Large et al., 1996; Yuan, 1997) in contractile PASMCs, their role in proliferation of PASMCs remains unclear due to the lack of specific antagonists. The recent identification of TMEM16a as a strong candidate gene for 2+ encoding Ca -activated ICl (Caputo et al., 2008; Schroeder et al., 2008; Yang et al., 2008b) may 2+ motivate further studies using molecular tools to clarify the role of Ca -activated ICl.

The underlying molecular basis for, as well as the mechanisms for the upregulation of, 2+ the Ca -independent ICl observed in our proliferating PASMCs are unknown. In cultured rat 2+ PASMCs, the baseline and swelling-activated Ca -independent ICl displayed similar biophysical and pharmacological properties (baseline ICl were also sensitive to DIDS, data not shown), suggesting that the same Cl- channel might generate both currents. The permeability ratio of I- versus Cl- for the baseline currents (7.7), however, was greater than that for swelling-activated currents (2.8). Therefore, baseline and swelling-activated ICl in proliferating rat PASMCs may be mediated by different channels, although it is possible that channel properties may be altered by increased membrane tension associated with cell swelling (Dai et al., 1998). Although several candidate genes have been suggested (Paulmichl et al., 1992; Duan et al., 1997), the gene encoding swelling-activated ICl remains unclear (Okada et al., 1998) and additional work will be necessary to determine the molecular basis for the increased ICl in proliferating PASMCs. Several transcription factors have been identified to play a role in the altered gene expressions during the transition of VSMCs from the contractile to proliferative phenotype (Owens et al.,

70 2004; Wamhoff et al., 2006). For example, Krüppel-like factor 4 is involved in the suppression of differentiation marker genes (Liu et al., 2005), while Krüppel-like factor 5 is important for the expression of embryonic isoform of smooth muscle myosin heavy chain (SMemb/NMHC-B) (Nagai et al., 2000; Suzuki et al., 2009), which is a marker gene of proliferative VSMCs. It is possible that these transcription factors may also regulate other aspects of the VSMC phenotypic modulation, including the upregulation of ICl in PASMCs as observed in the present study. Accordingly, further studies may be needed to test this possibility.

One limitation of the present study is the poor selectivity of DIDS, which blocks several types of ICl (d'Anglemont de Tassigny et al., 2003). Thus it remains to be determined which types of ICl are required for PASMC proliferation. Besides its inhibitory effects on ICl, DIDS has - - - - also been shown to inhibit HCO3 /Cl exchanger. However, because inhibition of the HCO3 /Cl exchanger leads to intracellular alkalization (Wu et al., 1994), which increases cell proliferation (McLean et al., 2000; Putney et al., 2003) and reduces apoptosis (Rich et al., 2000), the reduced - - PASMC proliferation by DIDS treatment cannot be explained by its action on HCO3 /Cl exchanger.

In summary, the results of the current study establish that the phenotypic transition of rat PASMCs from the contractile state to the proliferative state is associated with upregulation of 2+ DIDS-sensitive, Ca -independent, baseline and swelling-activated ICl and blockade of these ICl reduces proliferation rates of PASMCs. These observations may have direct and immediate relevance to pathological conditions associated with excessive PASMC proliferation, such as pulmonary arterial hypertension. The development of pharmacological tools that permit selective blockade of ICl may be useful in ameliorating PASMC proliferation and thus the progression of pulmonary arterial hypertension and right heart decompensation.

Chapter 4

Swelling-Activated Cl- Currents and Intracellular CLC-3 are Required for Proliferation of Human Pulmonary Artery Smooth Muscle Cells

Attributions:

W. Liang performed the experiments in this study; Dr. Jeff He contributed the qRT-PCR technique; Ms. Dongling Zhao contributed the technique of lentivirus generation.

4 Swelling-Activated Cl- Currents and Intracellular CLC-3 are Required for Proliferation of Human Pulmonary Artery Smooth Muscle Cells

4.1 Abstract

Proliferation of pulmonary artery smooth muscle cells (PASMCs) leads to adverse vascular remodeling and contributes to pulmonary arterial hypertension, a fatal disease with a 15% annual - mortality rate despite current therapies. Our previous study suggested a role of Cl currents (ICl) - in PASMC proliferation. However, it remains unknown which type of ICl or which Cl channel - gene is involved. In the present study, swelling-activated Cl currents (ICl,swell) were identified in proliferating human PASMCs (hPASMCs) and their sensitivity to DCPIB (IC50, 2.1 µM), a specific blocker of ICl,swell, was confirmed. Serum-stimulated increases of hPASMC number were dose-dependently inhibited by DCPIB (IC50, 6.3 µM), suggesting a role of ICl,swell in hPASMC proliferation. Previous studies showed that CLC family of Cl- channel genes play important roles in physiology and disease. Our real-time qRT-PCR studies demonstrated the expression of 6 CLC genes in proliferating hPASMCs with CLC-3 as the most abundant. Since

CLC-3 has also been suggested to encode ICl,swell, we examined effects of CLC-3 inhibition on proliferation and ICl,swell in hPASMCs. Stable expression of a microRNA-adapted shRNA targeting CLC-3 in hPASMCs via lentiviral transduction reduced CLC-3 mRNA by 82% (with no inhibition of other CLC genes) and reduced the serum-stimulated cell number increases by 45% compared to hPASMCs expressing a control shRNA. However, shRNA inhibition of

CLC-3 had no effect on ICl,swell. Consistent with this observation, immunocytostaining with anti-CLC-3 antibody showed that CLC-3 proteins are primarily located in intracellular areas.

These findings are consistent with the conclusion that both ICl,swell and CLC-3 are required for

PASMC proliferation, but CLC-3 is not the channel mediating ICl,swell.

Key Words: swelling-activated Cl- currents; DCPIB; CLC-3; RNAi; proliferation; pulmonary arterial hypertension

73 4.2 Introduction

Pulmonary arterial hypertension (PAH) is a rare but fatal disease of the pulmonary vasculature (Thenappan et al., 2007; Humbert et al., 2010). PAH is associated with proliferation of pulmonary artery smooth muscle cells (PASMCs) (Humbert et al., 2004) that leads to obstructive vascular remodeling and contributes to PAH. Inhibition of PASMC proliferation has been proven to be a valid strategy to treat PAH, but current therapies are not fully satisfactory and the PAH mortality rate remains high (15% annually) (Thenappan et al., 2007; Humbert et al., 2010). A better understanding of PASMC proliferation mechanisms is necessary and may identify new targets for more efficient treatments.

- Cl currents (ICl) have been suggested to play roles in proliferation of vascular smooth muscle cells (VSMCs) (Xiao et al., 2002; Cheng et al., 2007) and other cells (Wondergem et al., 2+ 2001), and PASMC expression of swelling-activated ICl (ICl,swell) and Ca -activated ICl (ICl,Ca) have been demonstrated in previous studies of ours (Liang et al., 2009) and others (Zhong et al.,

2002; Angermann et al., 2006). The PASMC activities of ICl,swell are associated with cell proliferative status with proliferating PASMCs having greater ICl,swell amplitude (Liang et al.,

2009). In addition, PASMC proliferation is inhibited by ICl blockers (Liang et al., 2009). These observations support a role of ICl (especially ICl,swell) in PASMC proliferation. However, since the

ICl blockers used in these studies (e.g. DIDS and NPPB) are of poor specificity and block several types of ICl (d'Anglemont de Tassigny et al., 2003), it remains unknown if blockade of ICl,swell alone is sufficient to suppress PASMC proliferation.

DCPIB, a derivative of ethacrynic acid, has been shown to be a more selective blocker of

ICl,swell (Decher et al., 2001). It potently inhibits native ICl,swell (IC50, 4.1 µM) in pulmonary artery endothelial cells (PAECs) (Decher et al., 2001). DCPIB at 10 µM, causing 84% reduction of

PAEC ICl,swell, has no or little effects (<10% inhibition) on other ICl, including baseline ICl (active in resting cells), ICl,Ca, CFTR, CLC-1, CLC-2, CLC-4, CLC-5, and CLC-K1 (Decher et al., 2001). Lack of effects on cation currents has also been shown for DCPIB (Decher et al., 2001). In order to explore the role of ICl,swell in PASMC proliferation, the present study was carried out to examine the effects of DCPIB on ICl,swell and proliferation in human PASMCs. Our studies showed that both PASMC ICl,swell and proliferation were dose-dependently inhibited by DCPIB with close IC50 values, suggesting a role of ICl,swell in PASMC proliferation. Although the - molecular identity of ICl,swell has been elusive, a member of the CLC family Cl channel genes,

CLC-3, has been proposed to mediate ICl,swell (Duan et al., 1997; Xiong et al., 2010). In addition, expression of CLC-3 has been shown in both VSMCs (Lamb et al., 1999) and PASMCs (Dai et al., 2005) and CLC-3 inhibition with antisense oligonucleotides reduces VSMC proliferation

(Wang et al., 2002). In the present study, a possible role of CLC-3 in PASMC ICl,swell and proliferation was explored in human PASMCs with CLC-3 knockdown using lentiviral shRNAs. Our study suggests that CLC-3 is required for PASMC proliferation but it does not contribute to

ICl,swell.

4.3 Summary of Materials and Methods

Please refer to Chapter 2 for detailed methods.

Human PASMCs were obtained from Cambrex Incorporation and cultured in smooth muscle growth medium (SMGM-2). Cells at passage 5-12 were used in the present study.

ICl were recorded with whole-cell patch-clamp technique in voltage-clamp mode in conventional ruptured recording configuration, with a bath solution containing (in mM): N-methyl-D- glucamine(NMDG)-HCl 100, MgCl2 1.5, HEPES 10, D-mannitol 70, and glucose 10 (pH=7.4 with NMDG, estimated osmolarity=295 mOsm) and a pipette solution containing (in mM): NMDG-HCl 30, NMDG-gluconate 93, HEPES 5, EGTA 10, and Mg-ATP 5 (pH=7.2 with NMDG). Hypotonic bath solution was made by omitting D-mannitol (estimated osmolarity=225 mOsm). Series resistance in voltage-clamp studies was compensated by 70-80%. In proliferation studies, cells were cultured in 12-well plates at 1×104 cells/well in the presence of either DIDS, DCPIB, or vehicle (DMSO or ethanol). After 4 days in culture, cell numbers in each well were determined by cell counting with a hemocytometer. Total RNA was extracted from cells at 60- 80% confluency with the Protein and RNA Isolation System (Applied Biosystems, Foster City, CA) (He et al., 2010) and treated with DNase to eliminate residual genomic DNA. To examine expression of Cl- channel genes in human PASMCs, qRT-PCR was performed in triplicate for each sample using SYBR green reagent (Applied Biosystems) with the ABI PRISM 7900 HT sequence detection system (Applied Biosystems). Primers for Cl- channel genes were designed using Primer3 program (http://frodo.wi.mit.edu/primer3/) and sequences are given in Tables 4.1 and 4.2. Data were normalized to 28S ribosomal RNA levels. CLC-3-shRNA lentiviruses were generated by transfecting HEK293T cells (passage 4-9, 1×106 cells/60-mm dish) with 2 µg shRNA-encoding plasmid, 2 µg pMD2.G and 0.2 µg pCMV-dR8.91 using 12 µL FuGENE6

75 transfection reagent (Roche Applied Science, Indianapolis, IN). Human PASMCs were transduced with lentivirus at a MOI (multiplicity of infection) of 10-20 together with 4 µg/ml polybrene (Sigma). Infected cells were selected with puromycin (2 µg/ml, Sigma) for 4 days when ~100% of the cells were GFP-positive. Differences between two means were evaluated by two-tailed Student’s t-test. Differences among multiple means were assessed by one-way ANOVA and evaluated post hoc by Bonferroni’s test.

4.4 Results

- In order to examine if swelling-activated Cl currents (ICl,swell) are required for proliferation of human PASMCs with small molecule inhibitors, we first determined if ICl,swell are present in these cells and if they can be blocked by Cl- channel inhibitors. In whole-cell patch-clamp studies, small baseline currents (current density at +60 mV, 1.8±0.2 pA/pF, n=5, Figure 4.1) were recorded in human PASMCs when perfused with an isotonic bath solution and, similar to our previous observations in rat PASMCs (Liang et al., 2009), large currents (current density at +60 mV, 41.6±14.2 pA/pF, n=5, Figure 4.1) were activated after switching to a hypotonic solution (designed to isolate ICl in whole-cell recordings, see Methods) for 5~10 min when cell swelling was visually observed. These swelling-activated currents had outward rectification, instantaneous activation, and time-dependent inactivation at high voltages (> +60 mV), all of which are typical properties of ICl,swell found in rat PASMCs (Liang et al., 2009) and other cell types (Okada, 1997). Addition of a classical ICl,swell blocker DIDS (100 µM) to the hypotonic solution inhibited (p<0.05) the swelling-induced currents by 91.6% (current density at +60 mV with DIDS, 3.5±1.3 pA/pF, n=4, Figure 4.1), suggesting that these swelling-activated - currents in human PASMCs are ICl,swell. We next tested effects of another Cl channel blocker

DCPIB, which has been shown to be a potent (IC50, 4.1 µM) and specific inhibitor of ICl,swell in other cell types (Decher et al., 2001), on ICl,swell in human PASMCs. Consistent with previous studies, DCPIB inhibited ICl,swell in human PASMCs in a dose-dependent manner with an estimated IC50 of 2.1 µM (Figure 4.2). Furthermore, we found that the effects of DCPIB were partly reversible with recovery of ICl,swell in human PASMCs after drug washout (Figure 4.2).

After knowing that DIDS- and DCPIB-sensitive ICl,swell are present in human PASMCs, we examined if these inhibitors affect proliferation in these cells. Consistent with our observations in rat PASMCs (Liang et al., 2009), addition of DIDS in the culture media reduced

(p<0.01) cell number of human PASMCs in a dose-dependent manner with an estimated IC50 of 29.3 µM (Figure 4.3A). More importantly, cell number of human PASMCs was also reduced by the ICl,swell specific inhibitor DCPIB in a dose-dependent manner and the estimated IC50 (6.3 µM,

Figure 4.3B) was close to that of DCPIB-mediated ICl,swell inhibition, suggesting a role of ICl,swell in human PASMC proliferation.

Despite their well-characterized properties, the molecular identity of ICl,swell remains unclear. Members of the CLC Cl- channel family (especially CLC-3) have been proposed to encode ICl,swell (Duan et al., 1997; Xiong et al., 2010). We, therefore, examined if CLC family genes are expressed in human PASMCs with quantitative real-time RT-PCR (qRT-PCR). The oligonucleotide primers designed to detect CLC gene transcripts (Table 4.1) in these studies had similar amplification efficiencies, as reflected by the close slope values (-3.0 to -3.2) of their standard curves (Figure 4.4). Thus, it was possible to compare the relative mRNA amounts of different CLC genes. As shown in Figure 4.5, qRT-PCR identified the expressions of 6 CLC genes (CLC-2 to -7) in proliferating human PASMCs, while mRNA of CLC-1, -Ka or -Kb could not be detected. The relative mRNA amounts of the 6 CLC genes in human PASMCs were CLC- 3(17.5) > CLC-7(5.3) > CLC-5(2.8) ≈ CLC-6(2.6) > CLC-4(2.0) > CLC-2(1.0) (n=4). This is consistent with previous reports that CLC-3 is the most highly expressed CLC gene in VSMCs (Lamb et al., 1999). Bestrophin family genes (BEST-1 to -4) also encode Cl- channels (Hartzell et al., 2004) and we further examined their expressions in human PASMCs. As shown in Figure 4.5, only BEST-1 mRNA was detected in human PASMCs.

Although our above observations with DCPIB suggest a role of ICl,swell in human PASMC proliferation, we cannot exclude the possibility that the DCPIB-induced proliferation inhibition is mediated by its yet unknown targets other than ICl,swell. Because CLC-3 has been shown to be important for both ICl,swell and cellular proliferation in several cell types (Wang et al., 2002; Tao et al., 2008; Xu et al., 2010), we next tested if CLC-3 inhibition affects proliferation and ICl, swell in human PASMCs. Stable expression of two different microRNA- adapted shRNAs targeting CLC-3 in human PASMCs, via lentiviral transduction, reduced (p<0.01, n=3) CLC-3 mRNA by 56.7% and 82.1% respectively (Figure 4.6). In addition, the expressions of other CLC genes were not reduced by CLC-3-shRNAs (Figure 4.6), suggesting CLC-3 specific knockdown. However, expressions of CLC-4 and CLC-5, two intracellular Cl- /H+ exchangers (Jentsch, 2008), showed a trend of increase (p=0.07) in the CLC-3 knockdown human PASMCs (Figure 4.6). Since CLC-3, -4 and -5 are in a same subgroup of CLC family and

77 have higher sequence homology than other CLC genes (Jentsch et al., 2002), the selective increases in CLC-4 and -5 expressions may reflect a compensatory mechanism in the deficiency of CLC-3.

To test if shRNA-mediated CLC-3 knockdown affects proliferation in human PASMCs, we compared cell numbers after culture in serum-containing medium between cells expressing either a CLC-3-shRNA (the one that caused an 82.1% reduction in CLC-3 mRNA) or a control- shRNA. As shown in Figure 4.7, cell number of CLC-3-shRNA group was 45% less (p<0.01, n=4) than that of control-shRNA group after culture in normal O2 (21%) condition. Since hypoxia is known to increase PASMC proliferation, we also explored if CLC-3 knockdown reduces human PASMC proliferation in a low O2 condition. When cultured in hypoxic (1%O2) environment, cell number of CLC-3-shRNA group was still 30% less (p<0.01, n=4) than that of control-shRNA group (Figure 4.7). These results suggest that CLC-3 is important for human PASMC proliferation under both normoxic and hypoxic conditions.

CLC-3, although controversial (Li et al., 2000; Weylandt et al., 2001), has been proposed to encode ICl,swell (Duan et al., 1997; Xiong et al., 2010), which is important for human

PASMC proliferation. We, therefore, examined if CLC-3 knockdown cells have altered ICl,swell that may mediate their reduced proliferation. Patch-clamp studies showed that hypotonic solution perfusion induced DIDS-sensitive ICl,swell in human PASMCs expressing either control-shRNA or CLC-3-shRNA (Figure 4.8A and B). However, neither the outward currents (+60 mV, p=0.47) nor the inward currents (-60 mV, p=0.27) were different between cells expressing control- shRNA or CLC-3-shRNA (n=16, Figure 4.8C). These results suggest that CLC-3 knockdown does not affect osmotically activated ICl,swell in human PASMCs and therefore the reduced proliferation of CLC-3 knockdown cells is not due to alterations in ICl,swell.

In order to examine the subcellular localizations of CLC-3 proteins in human PASMCs, we conducted immunocytostaining experiments with anti-CLC-3 antibodies. As shown in Figure 4.9, CLC-3-associated fluorescence was primarily in perinuclear area and cytoplasm with little, if any, in plasma membrane. These results suggest that CLC-3 is mainly an intracellular protein in human PASMCs, consistent with the unaltered plasma membrane ICl,swell in CLC-3 knockdown cells.

Table 4.1 Oligonucleotide primers for CLC family genes in quantitative real-time reverse transcription-PCR.

Human Genes Sequences Location Product Size (kb) Forward: (5’-3’)

CLCN1 TTTCCCACACAGTCTCCACA exon 15 91 Reverse: (5’-3’)

GCCAAGATAACAGCCACCAT exon 15 Forward: (5’-3’) TGCCTGTCTTTGTCATTGGA exons 13-14 CLCN2 97 Reverse: (5’-3’) TAGGTGCTGCTGTCCGTATG exon 14 Forward: (5’-3’) GGGCTGATTGCATTACACCT exon 10 CLCN3 90 Reverse: (5’-3’) CCACCAGGGAGACAGTCATT exon 11 Forward: (5’-3’) GAGGGTGCCAGTGCTTACAT exons 5-6 CLCN4 92 Reverse: (5’-3’) AATACACGCACCAGGGAGAC exon 6 Forward: (5’-3’) TTGGGCTTTTATCAGGTTCG exons 3-4 CLCN5 132 Reverse: (5’-3’) CATGCTCAGAGTTCCAGCAA exon 4 Forward: (5’-3’) ACAGACATCGGTGGAGGAGT exons 5-6 CLCN6 93 Reverse: (5’-3’) GCCAGGAAGACAAAGGTGAG exon 6 Forward: (5’-3’) GGAGAAAATGGCCTACACGA exons 12-13 CLCN7 84 Reverse: (5’-3’) TTGAACACTGCTCCAAGCAC exon 14 Forward: (5’-3’) CACCAGGCACAAAACCTCTT exons 18-19 CLCNKa 100 Reverse: (5’-3’) TCTTCATCTCCACCCAGGAC exons 19-20 Forward: (5’-3’) AGGAGGTGGTCAAGGTTGTG exon 2 CLCNKb 75 Reverse: (5’-3’) GGATCTGGGACTCTGTGCTC exons 2-3 Forward: (5’-3’) TTGAAAATCCGGGGGAGAG 28S Ribosomal 100 RNA Reverse: (5’-3’) ACATTGTTCCAACATGCCAG

Table 4.2 Oligonucleotide primers for Bestrophin family genes in quantitative real-time reverse transcription-PCR.

Human Genes Sequences Location Product Size (kb) Forward: (5’-3’)

BEST1 TGCTGAACGAGATGAACACC Exons 5-6 109 Reverse: (5’-3’)

CTGTACACCGCCACAGTCAC exon 7 Forward: (5’-3’) AAACTTCCAGGTGTCCATGC exons 7-8 BEST2 128 Reverse: (5’-3’) GCAGCTGGAAGACAGTAGCC exon 8 Forward: (5’-3’) TGGGGAAACAGATGCCTAAG exons 9-10 BEST3 78 Reverse: (5’-3’) CACACTGGAATTGAGGCTGA exon 10 Forward: (5’-3’) GGTGGCTGAACAGATCATCA exons 6-7 BEST4 81 Reverse: (5’-3’) TGCAAGTTGCGGTCTATGAG exon 7

Table 4.3 shRNAs used in CLC-3 knockdown studies.

Target Regions in Name of shRNAs Plasmids Hairpin Types Target Sequences (5’-3’)* Transcript

Human CLC-3, CLC-3-shRNA-1 pGIPZ microR30-based CUACCUCUUUCCAAAGUAU Exon 7

Human CLC-3, CLC-3-shRNA-2 pGIPZ microR30-based CCUGGAAGAGGUUAGCUAU Exons 8-9

Control-shRNA-1 pPRIME microR30-based Targets firefly luciferase transcript Do not target known mammalian transcripts Control-shRNA-2 pGIPZ microR30-based UCUCGCUUGGGCGAGAGUAAG

*Target sequence refers to the segment of mRNA that is complementary to the mature antisense strand (guide strand) of the shRNA (see Figure 1.5, page 33).

A Hypotonic solution 1000 pA

Isotonic solution 200 ms Hypotonic + DIDS

B I (pA/pF) 100

Isotonic solution Hypotonic solution Hypotonic + DIDS 50

-120 -80 -40 40 80 120 V (mV) -50

Figure 4.1 DIDS-sensitive, swelling-activated Cl- currents in human PASMCs. A. Cl- currents recorded in isotonic (left panel) and hypotonic (middle panel) bath solutions and in hypotonic solution in the presence of 100 µM DIDS (right panel). Dashed lines indicate the position of zero current. B. Current-voltage relationships for Cl- currents in isotonic (filled circles, n=5 cells) and hypotonic (open circles, n=5 cells) solutions and in hypotonic solution in the presence of 100 µM DIDS (triangles, n=4 cells).

82 A

Hypotonic solution

1000 pA Washout with

Isotonic solution 200 ms Hypotonic + DCPIB Hypotonic solution

I (pA/pF) 150 Isotonic solution Hypotonic solution 100 Hypotonic + DCPIB Washout 50

-120 -80 -40 40 80 120 V (mV) -50 B Inhibition of I at +60mV Inhibition of I at -60mV Cl,swell Cl,swell 100 100

80 80

 IC = 2.0 M 60 IC50 = 2.1 M 60 50

40 40

Current inhibition (%) 20 20 Current inhibition (%) inhibition Current

0 0 -6.5 -6.0 -5.5 -5.0 -4.5 -4.0 -6.5 -6.0 -5.5 -5.0 -4.5 -4.0 Log [DCPIB] Log [DCPIB]

Figure 4.2 DCPIB inhibits swelling-activated Cl- currents in human PASMCs. A. Upper panels, Cl- currents recorded in a human PASMC bathed in different solutions as labeled (10 µM DCPIB was used in the 3rd panel). Dashed lines indicate the position of zero current. Lower panel, current-voltage relationships in isotonic (filled circles, n=10 cells) and hypotonic (open circles, n=11 cells) solutions, as well as in hypotonic solution with 10 µM DCPIB (filled squares, n=11 cells) and after DCPIB washout (open squares, n=4 cells). B. Dose- response curves for DCPIB-induced decreases in swelling-activated Cl- currents at +60 mV (left, n=4-7 cells) and -60 mV (right, n=4-7 cells). Red lines indicate dose-response fittings with (LOGx0-x)*p) formula: y = A1 + (A2-A1)/(1 + 10 ), from which IC50 values were derived.

100

80 IC50 = 29.3 M 60

20 Cell number reduction (%) reduction number Cell 0 -6.0 -5.5 -5.0 -4.5 -4.0 -3.5 Log [DIDS] B

100

IC50 = 6.3 M 60

20 Cell number reduction (%) reduction number Cell 0 -6.5 -6.0 -5.5 -5.0 -4.5 -4.0 Log [DCPIB]

Figure 4.3 DIDS and DCPIB inhibit human PASMC proliferation. Dose-response curves for DIDS-induced (A, n=4-8) and DCPIB-induced (B, n=4-8) decreases in human PASMC numbers after 4-day culture in serum-containing media. Red lines indicate dose-response fittings with (LOGx0-x)*p) formula: y = A1 + (A2-A1)/(1 + 10 ), from which IC50 values were derived.

21 28S rRNA 31 CLC-2 20 2 30 2 19 R = 0.999 R = 0.992 29 18 Ct Ct 28 17 Slope = -3.2 Slope = -3.0 Intercept = 14.9 27 Intercept = 30.1 16 26 15 -1.6 -1.2 -0.8 -0.4 0.0 -0.4 0.0 0.4 0.8 1.2 Log RNA input (ng) Log RNA input (ng)

29 CLC-3 CLC-4

30 28 R2 = 0.995 R2 = 0.997

27 28 Ct Ct

Slope = -3.2 26 Slope = -3.0 26 Intercept = 26.7 Intercept = 27.9 25 24 -1 0 1 -0.4 0.0 0.4 0.8 Log RNA input (ng) Log RNA input (ng)

CLC-5 29 CLC-6 27 28 R2 = 0.999 R2 = 0.990 26 27 Ct Ct 25 Slope = -3.1 26 Slope = -3.0 Intercept = 26.8 Intercept = 28.0 24 25

-0.4 0.0 0.4 0.8 -0.4 0.0 0.4 0.8 Log RNA input (ng) Log RNA input (ng)

32 CLC-7 34 BEST-1

30 32 R2 = 0.996 R2 = 0.989 28 30 Ct Ct 26 28 Slope = -3.1 Slope = -3.2 26 24 Intercept = 25.3 Intercept = 26.7

24 -2 -1 0 1 -2 -1 0 1 Log RNA input (ng) Log RNA input (ng)

Figure 4.4 Standard curves of quantitative real-time RT-PCR primers. Oligonucleotide primer sequences are shown in Tables 4.1 and 4.2. Red lines indicate linear fittings (y=ax+b), from which slope (a) and intercept (b) values were derived. Ct means threshold cycle.

4 Relative mRNA amount mRNA amount Relative 0 -3 -4 -6 b 1 2 3 4 C -Ka T- T- T- LC-2 LC LC-5 LC LC-7 S S S CLC-1 C C CL C C C E E E CLC CLC-K B B BEST- B

Figure 4.5 Relative mRNA amounts for CLC and Bestrophin Cl- channel genes in human PASMCs. Oligonucleotide primer sequences are shown in Tables 4.1 and 4.2. The primers had similar amplification efficiencies, as reflected by the close slope values (-3.0 to -3.2) of their standard curves as shown in Figure 4.4. The relative mRNA amounts of the CLC genes are CLC-3(17.5) > CLC-7(5.3) > CLC-5(2.8) ≈ CLC-6(2.6) > CLC-4(2.0) > CLC-2(1.0) (n=4). mRNA of CLC-1, -Ka, and -Kb, and BEST-2 to -4 could not be detected.

Control-shRNA-1 Control-shRNA-2 30 CLC-3-shRNA-1 CLC-3-shRNA-2

10 * Relative mRNA amount *

2 -4 -5 -7 C C C L L CLC-3 CLC- C CL CLC-6 C

Figure 4.6 CLC-3-shRNAs selectively reduce CLC-3 expression in human PASMCs. *p<0.01, vs control-shRNA. n=5 for control-shRNA-1; n=2 for control-shRNA-2; n=3 for both CLC-3-shRNA-1 and -2. Target sequences of shRNAs are shown in Table 4.3.

) Control-shRNA 6 CLC-3-shRNA cells/well 4 4 * x10

( * 2

Cell number number Cell 0

21% O2 1% O2

Figure 4.7 CLC-3 knockdown inhibits human PASMC proliferation. Cell numbers were determined by cell counting after 4-day culture in serum-containing medium in a normoxic (21%

O2) or hypoxic (1% O2) environment. *p<0.01 vs. corresponding control-shRNA group, n=4.

A Control-shRNA I (pA/pF) 80

Hypotonic solution Isotonic solution 60 30 pA/pF Hypotonic solution Hypotinic + DIDS 200 ms Isotonic solution Hypotonic + DIDS 40 20

-120 -80 -40 40 80 120 -20 V (mV)

B CLC-3-shRNA I (pA/pF) 80

Hypotonic solution Isotonic solution 60 30 pA/pF Hypotonic solution Hypotinic + DIDS 40 Isotonic solution 200 ms Hypotonic + DIDS 20

-120 -80 -40 40 80 120 V (mV) -20

C I (pA/pF) 80

Control-shRNA 60 CLC-3-shRNA

-120 -80 -40 40 80 120 V (mV) -20

Figure 4.8 Swelling-activated Cl- currents in human PASMCs expressing control-shRNA or CLC-3-shRNA. A. Control-shRNA: Cl- currents recorded in isotonic (left 1st panel) and hypotonic (2nd panel) bath solutions, as well as in hypotonic solution in the presence of 100 µM DIDS (3rd panel). Dashed lines indicate the position of zero current. Rightmost panel: current- voltage relationships for Cl- currents in isotonic (filled circles, n=10 cells) and hypotonic (open circles, n=10 cells) solutions and in hypotonic solution in the presence of 100 µM DIDS (triangles, n=6 cells). B. CLC-3-shRNA: Cl- currents (left 3 panels) obtained in the same conditions as those described for control-shRNA. Rightmost panel: current-voltage relationships for Cl- currents in isotonic (filled circles, n=16 cells) and hypotonic (open circles, n=16 cells) solutions and in hypotonic solution in the presence of 100 µM DIDS (triangles, n=11 cells). C. Comparison of current-voltage relationships of swelling-activated Cl- currents between human PASMCs expressing control-shRNA (filled circles) or CLC-3-shRNA (open circles). Neither the outward current density (at +60 mV, p=0.47) nor the inward current density (at -60 mV, p=0.27) was different between the two groups. Lines were fitted with Bolzmann function.

Figure 4.9 Immunocytostaining of human PASMCs with anti-CLC-3 antibody. Confocal images of two human PASMCs showing the intracellular localizations of CLC-3. Negative control group (without anti-CLC-3 antibody in staining process) had weak diffusive background fluorescence. Images were obtained with a 100× objective. Bars indicate 30 µm.

4.5 Discussion

A role of ICl,swell in cell proliferation has been proposed based on inhibition of proliferation by pharmacological ICl blockers (e.g. DIDS and NPPB) (Wondergem et al., 2001; Liang et al., 2009) and by siRNA inhibition of CLC-3 (Tao et al., 2008; Xu et al., 2010).

However, a definitive role of ICl,swell in proliferation could not be established because these ICl blockers also inhibit several other ICl, including native ICl,Ca (Large et al., 1996), and expressed

TMEM16a (Yang et al., 2008b) and Bestrophin-2 (Qu et al., 2004) (two candidates for ICl,Ca), as well as CFTR (Linsdell et al., 1996; Zhang et al., 2000). In addition, the role of CLC-3 in ICl,swell has been questioned (Stobrawa et al., 2001; Arreola et al., 2002; Gong et al., 2004), making it uncertain whether proliferation inhibition by CLC-3 repression is mediated by ICl,swell reductions. The major findings of the present study are that proliferation of human PASMCs was dose- dependently inhibited by DCPIB, which blocks ICl,swell but not ICl,Ca or CFTR (Decher et al.,

2001), suggesting a critical role of ICl,swell in PASMC proliferation. In addition, shRNA inhibition of CLC-3 suppressed human PASMC proliferation but did not affect ICl,swell, suggesting a role of

CLC-3 in PASMC proliferation but not in ICl,swell.

Results of the present study are consistent with the conclusion that human PASMC proliferation requires ICl,swell. This conclusion is supported by the observations that serum- stimulated PASMC number increases were dose-dependently inhibited by either DIDS

(a classical but non-specific ICl,swell blocker) or DCPIB (a novel and specific ICl,swell blocker).

The expression of DIDS-sensitive and DCPIB-sensitive ICl,swell in human PASMCs, and the close

IC50 values for DCPIB-induced inhibitions of PASMC ICl,swell (2.1 µM) and proliferation

(6.3 µM) further support a role of ICl,swell in PASMC proliferation. In addition, our previous observation that activities of ICl,swell are greater in proliferating than in non-proliferating rat

PASMCs (Liang et al., 2009) also implies the importance of ICl,swell for PASMC proliferation.

However, it is not clear which cell cycle steps require ICl,swell activation or what roles ICl,swell play in cell cycle progression. Activities of ICl,swell have been shown to be cell cycle-dependent with activities being lower in G1 phase and higher in S phase (Shen et al., 2000; Klausen et al., 2007), suggesting a role of ICl,swell in S phase or G1/S transition. The inhibition of BrdU incorporation

(reflecting DNA synthesis in S phase) by DIDS (Liang et al., 2009) and arrest of cells in G0/G1 phase by NPPB (Shen et al., 2000) also support the requirement of ICl,swell activation for S phase + + - or G1/S transition. Growth factor-stimulation activates Na /K /2Cl transport and uptake of

91 amino acids causing cell swelling (Bussolati et al., 1996; Jiang et al., 2001). Activation of ICl,swell by cell swelling, as a component of the regulatory volume decrease (RVD) mechanism that also involves K+ channels, has been proposed to be important for proliferation by limiting drastic changes in cell volume (Okada, 1997; Nilius, 2001). Further studies are needed for understanding a more detailed role of ICl,swell in cell cycling.

The mammalian CLC gene family has 9 members that encode either Cl- channels or Cl-/H+ exchangers (Jentsch, 2008) and play a variety of roles in physiology and disease (Jentsch et al., 2002). The present study showed the expression of 6 CLC genes (CLC-2 to -7) in human PASMCs, consistent with their broad tissue distributions (Jentsch, 2008). Although their physiological roles in PASMCs remain to be determined, these CLC genes have been shown to play roles in fundamental cell functions (Jentsch, 2008). CLC-2 is involved in cell volume regulation (Xiong et al., 1999; Roman et al., 2001), neural excitability regulation (Rinke et al., 2010), and transepithelial Cl- transport (Gyömörey et al., 2000). CLC-3 has been suggested to encode ICl,swell (Duan et al., 1997). CLC-4 to -7 are mainly present in the membrane of intracellular vesicles, including endosomes, lysosomes and synaptic vesicles, and may be important for acidification and Cl- accumulation of the vesicular lumen (Jentsch, 2008). The lack of expressions of CLC-1, -Ka and -Kb in human PASMCs is in agreement with previous findings that CLC-1 is primarily expressed in skeletal muscle, and CLC-Ka and -Kb are selectively expressed in renal tissues and inner ears (Jentsch, 2008).

CLC-3 as a candidate for ICl,swell remains a subject of current debate. Results of the present study are consistent with the conclusion that CLC-3 does not play a role in human

PASMC ICl,swell. This is supported by the unaltered properties and amplitudes of ICl,swell in CLC-3-shRNA-expressing human PASMCs. Moreover, the intracellular localization of CLC-3 in human PASMCs does not support it as a plasma membrane Cl- channel. In addition, the trend for increased CLC-4 and CLC-5, two intracellular Cl-/H+ exchangers (Picollo et al., 2005; Suzuki et al., 2006), in the deficiency of CLC-3 implies that these three CLC channels, which share 80% sequence identity and belong to the same homology branch of the CLC family (Jentsch, 2008), may play similar roles in intracellular organelles. Indeed, recent studies suggest that CLC-3 is important for functions of several intracellular vesicles, including endosomal generation of reactive oxygen species (ROS) in VSMCs (Miller et al., 2007), insulin granule maturation in pancreatic β cells (Deriy et al., 2009), and presynaptic vesicular accumulation of neurotransmitters in hippocampus (Riazanski et al., 2011).

92 Consistent with previous observations in VSMCs (Wang et al., 2002; Tang et al., 2008; Chu et al., 2011), our results suggest that CLC-3 is required for proliferation of human PASMCs. However, in contrast to previous observations in mouse mesenchymal stem cells (Tao et al., 2008) and human nasopharyngeal carcinoma cells (Xu et al., 2010), CLC-3 knockdown in human PASMCs affected only proliferation but not ICl,swell, suggesting an ICl,swell-independent mechanism for the altered proliferation in CLC-3 deficient cells. As mentioned above, CLC-3 has been suggested to be required for mitogenic stimulation-induced endosomal generation of ROS (Miller et al., 2007), which may be important for subsequent activation of extracellular signal-regulated kinase (ERK)1/2 and VSMC proliferation (Chu et al., 2011). In addition, siRNA knockdown of CLC-3 in rat VSMCs has been shown to impair the G1/S phase transition, accompanied by decreased expression of cyclins D1 and E and increased expression of cyclin- dependent kinase inhibitors p27 and p21 (Tang et al., 2008). Further studies are needed to determine if CLC-3 plays a similar role in human PASMC proliferation.

In summary, the present study suggests that both ICl,swell and CLC-3 are important for human PASMC proliferation, but they are not correlated and may regulate proliferation by different mechanisms. Since PASMC proliferation contributes to PAH, selective inhibition of

ICl,swell or CLC-3 may be useful for treating PAH.

Chapter 5 Summary and Future Directions

5 Summary and Future Directions 5.1 Summary

Smooth muscle cells (SMCs) in healthy pulmonary artery are differentiated, contractile and rarely proliferate. Under pathological conditions, however, these SMCs can switch to a proliferative phenotype and their proliferation contributes to adverse vascular remodeling in pulmonary arterial hypertension (PAH), a fatal disease with an annual mortality rate of 15% despite current therapies. Our studies demonstrated that proliferative pulmonary artery SMCs - (PASMCs) have greater activities of baseline and swelling-activated Cl currents (ICl) than contractile, non-proliferative PASMCs. Inhibition of these ICl with pharmacological blockers reduces serum-stimulated PASMC proliferation, suggesting that these ICl are required for proliferation. We further explored the mechanisms underlying ICl regulation of PASMC proliferation. Increased baseline ICl in proliferating PASMCs lead to membrane depolarization, which may contribute to their proliferation. Since swelling-activated ICl are important for cell volume regulation, our observation that selective inhibition of these ICl is sufficient to reduce PASMC proliferation suggests a role of cell volume regulation in PASMC proliferation.

Although electrophysiological studies have identified several types of ICl in PASMCs, little is known about the PASMC expression of Cl- channel genes. Our quantitative real-time RT-PCR studies showed expressions of 6 members of the CLC family Cl- channel genes (including CLC-3), and 1 member of the Bestrophin family Cl- channel genes in proliferating PASMCs.

Although it remains unclear which genes encode the baseline and swelling-activated ICl in

PASMCs, CLC-3 has been proposed to be a key component of swelling-activated ICl. In order to examine if CLC-3 is important for swelling-activated ICl in PASMCs, we generated CLC-3 knockdown PASMCs with lentiviral vectors encoding shRNAs targeting CLC-3. While selective knockdown of CLC-3 reduced PASMC proliferation, it had no effect on swelling-activated ICl, suggesting that CLC-3 is not related to these ICl. Furthermore, our immunocytostaining studies showed that CLC-3 proteins are primarily present in intracellular areas, which is also inconsistent with their previously proposed role as cell membrane ion channels. Our findings are consistent with the conclusion that baseline and swelling-activated ICl play important roles in promoting PASMC proliferation and suggest that development of tools that allow selective inhibition of PASMC ICl is necessary, as they may be novel treatments for PAH.

95 5.2 Future Directions

Our studies have demonstrated a correlation between ICl expression/activity and PASMC proliferation rate. However, further studies are needed for identification of the genes that encode

PASMC ICl, for elucidation of a more detailed role of ICl in PASMC proliferation, and for determination of effects of ICl blockade on PASMC proliferation in vivo and PAH development in experimental animals.

5.2.1 Molecular Identification of ICl in Proliferating PASMCs

It is of great importance to understand the molecular identities of the ICl that are upregulated in PASMCs after the switching from the contractile to the proliferating state.

Knowing the genes that encode these ICl would be critical for elucidating their gating mechanisms at the molecular level, for understanding the transcriptional regulations of their expression during PASMC phenotypic transition, and for the designing of PASMC-specific ICl modulators (either small chemical molecules or genetic tools) for PAH treatments. Although several families of mammalian genes have been proposed to encode Cl- channels (see section - 1.3, page 11), the properties of these Cl channels are not compatible with those of PASMC ICl. - This suggests that PASMC ICl may be encoded by other unknown Cl channel genes. Since it is known that proliferating PASMCs have increased ICl as compared to contractile PASMCs, a microarray comparison of gene expression patterns between these two types of PASMCs may identify genes that have increased expression in proliferating PASMCs. A further analysis of these selected genes may identify a subset of genes that encode membrane proteins with unclear functions. Immunocytostaining studies may identify a smaller subset of genes that encode plasma membrane proteins. Electrophysiological studies, in combination with heterologous expression and shRNA inhibition studies, may identify genes that encode PASMC Cl- channels. A similar strategy has been successfully used to identify the TMEM16a gene as a candidate for Ca2+-activated Cl- channels in bronchial epithelial cells (Caputo et al., 2008).

5.2.2 A More Detailed Role of ICl in PASMC Proliferation

The present study suggested a role of ICl in PASMC proliferation based on observations that ICl are increased in PASMCs undergoing cell cycling and ICl inhibitor treatments reduce

PASMC proliferation. Our studies also showed that increased ICl lead to membrane

96 depolarization in proliferating PASMCs and specific blockade of swelling-activated ICl inhibits

PASMC proliferation, suggesting that ICl-mediated membrane potential and cell volume alterations may promote PASMC proliferation. However, it is unclear which phases of the

PASMC cell cycle require ICl or what roles the ICl play in these phases. Since our current studies examined non-synchronized cells, further electrophysiological studies in combination with cell synchronization technique may be needed to determine if ICl activities and membrane potential are cell cycle-dependent in PASMCs, as has been shown for swelling-activated ICl in other cell types (Shen et al., 2000; Klausen et al., 2007). Future studies on cell volume regulation of synchronized PASMCs by measuring cell volume changes in response to osmotic challenges may reveal which cell cycle phases are associated with greater volume regulation ability, and if they are correlated with activities of swelling-activated ICl. A cell cycle-dependent role of ICl may be further examined by studies on cell cycle progression in the presence of ICl inhibition using either specific pharmacological blockers (e.g. DCPIB for swelling-activated ICl) or shRNA knockdown (if the PASMC ICl genes have been identified). These studies may identify a more detailed role of ICl in PASMC proliferation and provide further evidence supporting the use of ICl inhibition as a means for inhibiting PASMC proliferation and for treating PAH.

5.2.3 In Vivo Studies on Effects of ICl Blockade on PASMC Proliferation and PAH

The present study has demonstrated an important role of ICl in proliferation of cultured

PASMCs. Animal studies will be necessary to determine if ICl inhibition is able to suppress PASMC proliferation in the context of experimental pulmonary hypertension. Commonly used

PAH animal models include chronic hypoxia-treated rats (10% O2, 2-3 weeks) (Pozeg et al., 2003), monocrotaline-treated rats (Rosenberg et al., 1988), and fawn-hooded rats (that develop spontaneous PAH in adulthood) (Sato et al., 1992; Archer et al., 2010a). These animal models have been shown to develop PAH in association with pulmonary vascular remodeling and

PASMC proliferation (Csiszar et al., 2009; Stenmark et al., 2009). Accordingly, ICl blockers (e.g.

DCPIB) or virus vectors expressing shRNA targeting ICl genes (if the genes for PASMC ICl are known) may be tested in these animal models to see if PASMC proliferation, vascular remodeling, and PAH development can be suppressed. Since ICl are also expressed in other tissues, specific pulmonary artery delivery of ICl inhibitors is preferred. This may be partially achieved by using local drug delivery methods (Pozeg et al., 2003; Chen et al., 2011) or using

97 virus vectors that allow regulated shRNA expression (driven by VSMC-specific promoters or hypoxia-sensitive promoters) (Rhee et al., 2009). Alternatively, shRNAs can be designed to target specifically the PASMC isoforms of Cl- channels. These animal studies would be critical for assessing ICl inhibition as a strategy for PAH treatment.

5.3 Concluding Remarks

Our studies demonstrated the expression of ICl in proliferating PASMCs and the requirement of ICl activity for PASMC proliferation, providing the proof-of-principle for ICl inhibition as a strategy for suppressing PASMC proliferation and treating PAH. Our findings provide the rationale for future studies on molecular characterizations of PASMC ICl and in vivo examination of ICl inhibition as a approach for treating PAH.

References Aickin C (1990). Chloride channels and carriers in nerve, muscle, and glial cells. edn. Plenum Press: New York.

Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P (2008). The cell cycle. In: Molecular Biology of the Cell, 5th edn. New York: Garland Science. Page:1053-1114.

Almaca J, Tian Y, Aldehni F, Ousingsawat J, Kongsuphol P, Rock JR, et al. (2009). TMEM16 proteins produce volume-regulated chloride currents that are reduced in mice lacking TMEM16A. J Biol Chem 284(42): 28571-28578.

Amberg GC, Bonev AD, Rossow CF, Nelson MT, Santana LF (2003). Modulation of the molecular composition of large-conductance, Ca2+-activated K+ channels in vascular smooth muscle during hypertension. J Clin Invest 112(5): 717-724.

Ando-Akatsuka Y, Abdullaev IF, Lee EL, Okada Y, Sabirov RZ (2002). Down-regulation of volume-sensitive Cl- channels by CFTR is mediated by the second nucleotide-binding domain. Pflugers Arch 445(2): 177-186.

Angermann JE, Sanguinetti AR, Kenyon JL, Leblanc N, Greenwood IA (2006). Mechanism of the inhibition of Ca2+-activated Cl- currents by phosphorylation in pulmonary arterial smooth muscle cells. J Gen Physiol 128(1): 73-87.

Archer S, Rich S (2000). Primary pulmonary hypertension: A vascular biology and translational research "work in progress". Circulation 102(22): 2781-2791.

Archer SL, Huang JMC, Reeve HL, Hampl V, Tolarova S, Michelakis E, et al. (1996). Differential distribution of electrophysiologically distinct myocytes in conduit and resistance arteries determines their response to nitric oxide and hypoxia. Circ Res 78(3): 431-442.

Archer SL, Marsboom G, Kim GH, Zhang HJ, Toth PT, Svensson EC, et al. (2010a). Epigenetic attenuation of mitochondrial superoxide dismutase 2 in pulmonary arterial hypertension: a basis for excessive cell proliferation and a new therapeutic target. Circulation 121(24): 2661- 2671.

Archer SL, Michelakis ED (2009). Phosphodiesterase type 5 inhibitors for pulmonary arterial hypertension. N Engl J Med 361(19): 1864-1871.

Archer SL, Weir EK, Wilkins MR (2010b). Basic science of pulmonary arterial hypertension for clinicians: New concepts and experimental therapies. Circulation 121(18): 2045-2066.

Arias-Stella J, Saldana M (1963). The terminal portion of the pulmonary arterial tree in people native to high altitudes. Circulation 28(5): 915-925.

99 Arreola J, Begenisich T, Nehrke K, Nguyen H-V, Park K, Richardson L, et al. (2002). Secretion and cell volume regulation by salivary acinar cells from mice lacking expression of the CLCN3 Cl- channel gene. J Physiol 545(1): 207-216.

Arreola J, Melvin JE, Begenisich T (1996). Activation of calcium-dependent chloride channels in rat parotid acinar cells. J Gen Physiol 108(1): 35-47.

Auer G, Ward ME (1998). Impaired reactivity of rat aorta to phenylephrine and KCl after prolonged hypoxia: role of the endothelium. J Appl Physiol 85(2): 411-417.

Aversa CR, Oparil S, Caro J, Li H, Sun S-D, Chen Y-F, et al. (1997). Hypoxia stimulates human preproendothelin-1 promoter activity in transgenic mice. Am J Physiol Lung Cell Mol Physiol 273(4): L848-L855.

Barro Soria R, Spitzner M, Schreiber R, Kunzelmann K (2009). Bestrophin-1 enables Ca2+- activated Cl- conductance in epithelia. J Biol Chem 284(43): 29405-29412.

Barst RJ, McGoon M, Torbicki A, Sitbon O, Krowka MJ, Olschewski H, et al. (2004). Diagnosis and differential assessment of pulmonary arterial hypertension. J Am Coll Cardiol 43(12_Suppl_S): 40S-47.

Bear CE, Li CH, Kartner N, Bridges RJ, Jensen TJ, Ramjeesingh M, et al. (1992). Purification and functional reconstitution of the cystic fibrosis transmembrane conductance regulator (CFTR). Cell 68(4): 809-818.

Bennetts B, Rychkov GY, Ng H-L, Morton CJ, Stapleton D, Parker MW, et al. (2005). Cytoplasmic ATP-sensing domains regulate gating of skeletal muscle CLC-1 chloride channels. J Biol Chem 280(37): 32452-32458.

Bergdahl A, Gomez MF, Wihlborg A-K, Erlinge D, Eyjolfson A, Xu S-Z, et al. (2005). Plasticity of TRPC expression in arterial smooth muscle: correlation with store-operated Ca2+ entry. Am J Physiol Cell Physiol 288(4): C872-880.

Best L, Yates AP, Decher N, Steinmeyer K, Nilius B (2004). Inhibition of glucose-induced electrical activity in rat pancreatic beta-cells by DCPIB, a selective inhibitor of volume- sensitive anion currents. Eur J Pharmacol 489(1-2): 13-19.

Birmingham A, Anderson E, Reynolds A, Ilsley-Tyree D, Leake D, Fedorov Y, et al. (2006). 3' UTR seed matches, but not overall identity, are associated with RNAi off-targets. Nat Methods 3(3): 199-204.

Blaustein MP, Lederer WJ (1999). Sodium/calcium exchange: Its physiological implications. Physiol Rev 79(3): 763-854.

Bond TD, Valverde MA, Higgins CF (1998). Protein kinase C phosphorylation disengages human and mouse-1a P-glycoproteins from influencing the rate of activation of swelling- activated chloride currents. J Physiol 508 ( Pt 2): 333-340.

100 Bonnet S, Rochefort G, Sutendra G, Archer SL, Haromy A, Webster L, et al. (2007). The nuclear factor of activated T cells in pulmonary arterial hypertension can be therapeutically targeted. Proc Natl Acad Sci U S A 104(27): 11418-11423.

Borgnia M, Nielsen S, Engel A, Agre P (1999). Cellular and molecular biology of the aquaporin water channels. Annu Rev Biochem 68: 425-458.

Bosl M, Stein V, Hubner C, Zdebik A, Jordt S, Mukhopadhyay A, et al. (2001). Male germ cells and photoreceptors, both dependent on close cell-cell interactions, degenerate upon CLC-2 Cl- channel disruption. EMBO J 20(6): 1289-1299.

Bussolati O, Uggeri J, Belletti S, Dall'Asta V, Gazzola GC (1996). The stimulation of Na,K,Cl cotransport and of system A for neutral amino acid transport is a mechanism for cell volume increase during the cell cycle. FASEB J 10(8): 920-926.

Campbell FE, Jr, Setzer DR (1992). Transcription termination by RNA polymerase III: uncoupling of polymerase release from termination signal recognition. Mol Cell Biol 12(5): 2260-2272.

Caputo A, Caci E, Ferrera L, Pedemonte N, Barsanti C, Sondo E, et al. (2008). TMEM16A, a membrane protein associated with calcium-dependent chloride channel activity. Science 322(5901): 590-594.

Carmines PK (1995). Segment-specific effect of chloride channel blockade on rat renal arteriolar contractile responses to angiotensin II. Am J Hypertens 8(1): 90-94.

Chan HC, Kaetzel MA, Gotter AL, Dedman JR, Nelson DJ (1994). Annexin IV inhibits calmodulin-dependent protein kinase II-activated chloride conductance. A novel mechanism for ion channel regulation. J Biol Chem 269(51): 32464-32468.

Channick RN, Simonneau G, Sitbon O, Robbins IM, Frost A, Tapson VF, et al. (2001). Effects of the dual endothelin-receptor antagonist bosentan in patients with pulmonary hypertension: a randomised placebo-controlled study. Lancet 358(9288): 1119-1123.

Chen B, Nicol G, Cho WK (2004). Electrophysiological characterization of volume-activated chloride currents in mouse cholangiocyte cell line. Am J Physiol Gastrointest Liver Physiol 287(6): G1158-1167.

Chen L, Nakano K, Kimura S, Matoba T, Iwata E, Miyagawa M, et al. (2011). Nanoparticle- mediated delivery of pitavastatin into lungs ameliorates the development and induces regression of monocrotaline-induced pulmonary artery hypertension. Hypertension 57(2): 343-350.

Chen LX, Zhu LY, Jacob TJ, Wang LW (2007). Roles of volume-activated Cl- currents and regulatory volume decrease in the cell cycle and proliferation in nasopharyngeal carcinoma cells. Cell Prolif 40(2): 253-267.

Chen M-F, Chen T-Y (2001). Different fast-gate regulation by external Cl− and H+ of the muscle- type CLC chloride channels. J Gen Physiol 118(1): 23-32.

101

Cheng G, Kim M-J, Jia G, Agrawal DK (2007). Involvement of chloride channels in IGF-I- induced proliferation of porcine arterial smooth muscle cells. Cardiovasc Res 73(1): 198- 207.

Cherubini E, Gaiarsa JL, Ben-Ari Y (1991). GABA: an excitatory transmitter in early postnatal life. Trends Neurosci 14(12): 515-519.

Chipperfield A, Harper A (2000). Chloride in smooth muscle. Prog Biophys Mol Biol 74(3-5): 175-221.

Choi J, Chiang A, Taulier N, Gros R, Pirani A, Husain M (2006). A calmodulin-binding site on 2+ cyclin E mediates Ca -sensitive G1/S transitions in vascular smooth muscle cells. Circ Res 98(10): 1273-1281.

Christman BW, McPherson CD, Newman JH, King GA, Bernard GR, Groves BM, et al. (1992). An imbalance between the excretion of thromboxane and prostacyclin metabolites in pulmonary hypertension. N Engl J Med 327(2): 70-75.

Chu X, Filali M, Stanic B, Takapoo M, Sheehan A, Bhalla R, et al. (2011). A critical role for chloride channel-3 (CLC-3) in smooth muscle cell activation and neointima formation. Arterioscler Thromb Vasc Biol 31(2): 345-351.

Clark S, Jordt S-E, Jentsch TJ, Mathie A (1998). Characterization of the hyperpolarization- activated chloride current in dissociated rat sympathetic neurons. J Physiol 506(3): 665-678.

Clowes AW, Reidy MA, Clowes MM (1983). Kinetics of cellular proliferation after arterial injury. I. Smooth muscle growth in the absence of endothelium. Lab Invest 49(3): 327-333.

Cogolludo A, Moreno L, Lodi F, Frazziano G, Cobeno L, Tamargo J, et al. (2006). Serotonin inhibits voltage-gated K+ currents in pulmonary artery smooth muscle cells: Role of 5-HT2A receptors, caveolin-1, and KV1.5 channel internalization. Circ Res 98(7): 931-938.

Collier ML, Levesque PC, Kenyon JL, Hume JR (1996). Unitary Cl- channels activated by cytoplasmic Ca2+ in canine ventricular myocytes. Circ Res 78(5): 936-944.

Cone C, Cone C (1976). Induction of mitosis in mature neurons in central nervous system by sustained depolarization. Science 192(4235): 155-158.

Cone C, Tongier M (1973). Contact inhibition of division: involvement of the electrical transmembrane potential. J Cell Physiol 82(3): 373-386.

Cowan KN, Heilbut A, Humpl T, Lam C, Ito S, Rabinovitch M (2000). Complete reversal of fatal pulmonary hypertension in rats by a serine elastase inhibitor. Nat Med 6(6): 698-702.

Csiszar A, Labinskyy N, Olson S, Pinto JT, Gupte S, Wu JM, et al. (2009). Resveratrol prevents monocrotaline-induced pulmonary hypertension in rats. Hypertension 54(3): 668-675.

Cullen B (2005). RNAi the natural way. Nat Genet 37(11): 1163-1165.

102

Cunningham SA, Awayda MS, Bubien JK, Ismailov, II, Arrate MP, Berdiev BK, et al. (1995). Cloning of an epithelial chloride channel from bovine trachea. J Biol Chem 270(52): 31016- 31026. d'Anglemont de Ta A, Souktani R, Ghaleh B, Henry P, Berdeaux A (2003). Structure and pharmacology of swelling-sensitive chloride channels, ICl,swell. Fundam Clin Pharmacol 17(5): 539-553.

Dai J, Sheetz MP, Wan X, Morris CE (1998). Membrane tension in swelling and shrinking molluscan neurons. J Neurosci 18(17): 6681-6692.

Dai YP, Bongalon S, Hatton WJ, Hume JR, Yamboliev IA (2005). CLC-3 chloride channel is upregulated by hypertrophy and inflammation in rat and canine pulmonary artery. Br J Pharmacol 145(1): 5-14.

Davis J, Harper A, Chipperfield A (1997). Stimulation of intracellular chloride accumulation by noradrenaline and hence potentiation of its depolarization of rat arterial smooth muscle in vitro. Br J Pharmacol 122(4): 639-642.

De La Fuente R, Namkung W, Mills A, Verkman AS (2008). Small-molecule screen identifies inhibitors of a human intestinal calcium-activated chloride channel. Mol Pharmacol 73(3): 758-768.

Decher N, Lang HJ, Nilius B, Bruggemann A, Busch AE, Steinmeyer K (2001). DCPIB is a novel selective blocker of ICl,swell and prevents swelling-induced shortening of guinea-pig atrial action potential duration. Br J Pharmacol 134(7): 1467-1479.

Dempsey JA, Veasey SC, Morgan BJ, O'Donnell CP (2010). Pathophysiology of sleep apnea. Physiol Rev 90(1): 47-112.

Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G, et al. (2000). Familial primary pulmonary hypertension (gene PPH1) is caused by mutations in the bone morphogenetic protein receptor-II gene. Am J Hum Genet 67(3): 737-744.

Deriy LV, Gomez EA, Jacobson DA, Wang X, Hopson JA, Liu XY, et al. (2009). The granular chloride channel CLC-3 is permissive for insulin secretion. Cell Metab 10(4): 316-323.

Dhani SU, Bear CE (2006). Role of intramolecular and intermolecular interactions in CLC channel and transporter function. Pflugers Arch 451(6): 708-715.

Dimova DK, Dyson NJ (2005). The E2F transcriptional network: old acquaintances with new faces. Oncogene 24(17): 2810-2826.

Dong Y, Chen G, Duran GE, Kouyama K, Chao AC, Sikic BI, et al. (1994). Volume-activated chloride current is not related to P-glycoprotein overexpression. Cancer Res 54(19): 5029- 5032.

103 Duan D, Cowley S, Horowitz B, Hume JR (1999). A serine residue in CLC-3 links phosphorylation-dephosphorylation to chloride channel regulation by cell volume. J Gen Physiol 113(1): 57-70.

Duan D, Winter C, Cowley S, Hume JR, Horowitz B (1997). Molecular identification of a volume-regulated chloride channel. Nature 390(6658): 417-421.

Duan D, Zhong J, Hermoso M, Satterwhite CM, Rossow CF, Hatton WJ, et al. (2001). Functional inhibition of native volume-sensitive outwardly rectifying anion channels in muscle cells and Xenopus oocytes by anti-CLC-3 antibody. J Physiol 531(Pt 2): 437-444.

Dutzler R, Campbell E, Cadene M, Chait B, MacKinnon R (2002). X-ray structure of a CLC chloride channel at 3.0 A reveals the molecular basis of anion selectivity. Nature 415(6869): 287-294.

Dyson N (1998). The regulation of E2F by pRB-family proteins. Genes & Dev 12(15): 2245- 2262.

Eggermont J (2004). Calcium-activated chloride channels: (un)known, (un)loved? Proc Am Thorac Soc 1(1): 22-27.

Elbashir S, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T (2001). Duplexes of 21- nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411(6836): 494-498.

Emma F, Breton S, Morrison R, Wright S, Strange K (1998). Effect of cell swelling on membrane and cytoplasmic distribution of pICln. Am J Physiol Cell Physiol 274(6): C1545- 1551.

Esteve J, Launay J, Kellermann O, Maroteaux L (2007). Functions of serotonin in hypoxic pulmonary vascular remodeling. Cell Biochem Biophys 47(1): 33-44.

Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM (1992). Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J Immunol 148(7): 2207-2216.

Fantozzi I, Zhang S, Platoshyn O, Remillard CV, Cowling RT, Yuan JXJ (2003). Hypoxia increases AP-1 binding activity by enhancing capacitative Ca2+ entry in human pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol 285(6): L1233-L1245.

Fire A, Xu S, Montgomery M, Kostas S, Driver S, Mello C (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391(6669): 806- 811.

Fleischmann B, Murray R, Kotlikoff M (1994). Voltage window for sustained elevation of cytosolic calcium in smooth muscle cells. Proc Natl Acad Sci U S A 91(25): 11914-11918.

Friedrich T, Breiderhoff T, Jentsch TJ (1999). Mutational analysis demonstrates that CLC-4 and CLC-5 directly mediate plasma membrane currents. J Biol Chem 274(2): 896-902.

104

Frizzell RA, Rechkemmer G, Shoemaker RL (1986). Altered regulation of airway epithelial cell chloride channels in cystic fibrosis. Science 233(4763): 558-560.

Furukawa T, Ogura T, Katayama Y, Hiraoka M (1998). Characteristics of rabbit CLC-2 current expressed in Xenopus oocytes and its contribution to volume regulation. Am J Physiol Cell Physiol 274(2): C500-512.

Galie N, Ghofrani HA, Torbicki A, Barst RJ, Rubin LJ, Badesch D, et al. (2005). Sildenafil citrate therapy for pulmonary arterial hypertension. N Engl J Med 353(20): 2148-2157.

Gandhi R, Elble RC, Gruber AD, Schreur KD, Ji HL, Fuller CM, et al. (1998). Molecular and functional characterization of a calcium-sensitive chloride channel from mouse lung. J Biol Chem 273(48): 32096-32101.

Giaid A, Saleh D (1995). Reduced expression of endothelial nitric oxide synthase in the lungs of patients with pulmonary hypertension. N Engl J Med 333(4): 214-221.

Giaid A, Yanagisawa M, Langleben D, Michel RP, Levy R, Shennib H, et al. (1993). Expression of endothelin-1 in the lungs of patients with pulmonary hypertension. N Engl J Med 328(24): 1732-1739.

Golovina VA, Platoshyn O, Bailey CL, Wang J, Limsuwan A, Sweeney M, et al. (2001). Upregulated TRP and enhanced capacitative Ca2+ entry in human pulmonary artery myocytes during proliferation. Am J Physiol Heart Circ Physiol 280(2): H746-755.

Gong W, Xu H, Shimizu T, Morishima S, Tanabe S, Tachibe T, et al. (2004). CLC-3- independent, PKC-dependent activity of volume-sensitive Cl- channel in mouse ventricular cardiomyocytes. Cell Physiol Biochem 14(4-6): 213-224.

Greenwood IA, Ledoux J, Leblanc N (2001). Differential regulation of Ca2+-activated Cl- currents in rabbit arterial and portal vein smooth muscle cells by Ca2+-calmodulin-dependent kinase. J Physiol 534(Pt. 2): 395-408.

Gronemeier M, Condie A, Prosser J, Steinmeyer K, Jentsch TJ, Jockusch H (1994). Nonsense and missense mutations in the muscular chloride channel gene CLC-1 of myotonic mice. J Biol Chem 269(8): 5963-5967.

Grubb BR, Vick RN, Boucher RC (1994). Hyperabsorption of Na+ and raised Ca2+-mediated Cl- secretion in nasal epithelia of CF mice. Am J Physiol 266(5 Pt 1): C1478-1483.

Gruber AD, Elble RC, Ji HL, Schreur KD, Fuller CM, Pauli BU (1998). Genomic cloning, molecular characterization, and functional analysis of human CLCA1, the first human member of the family of Ca2+-activated Cl- channel proteins. Genomics 54(2): 200-214.

Grunder S, Thiemann A, Pusch M, Jentsch T (1992). Regions involved in the opening of CLC-2 chloride channel by voltage and cell volume. Nature 360(6406): 759-762.

105 Guyton AC, Hall JE (2006). Textbook of medical physiology. 11 edn. Elsevier Saunders: Philadelphia.

Gyömörey K, Yeger H, Ackerley C, Garami E, Bear CE (2000). Expression of the chloride channel CLC-2 in the murine small intestine epithelium. Am J Physiol Cell Physiol 279(6): C1787-C1794.

Habela CW, Olsen ML, Sontheimer H (2008). CLC3 is a critical regulator of the cell cycle in normal and malignant glial cells. J Neurosci 28(37): 9205-9217.

Habela CW, Sontheimer H (2007). Cytoplasmic volume condensation is an integral part of mitosis. Cell Cycle 6(13): 1613-1620.

Hammond S, Bernstein E, Beach D, Hannon G (2000). An RNA-directed nuclease mediates post- transcriptional gene silencing in Drosophila cells. Nature 404(6775): 293-296.

Hardingham G, Chawla S, Johnson C, Bading H (1997). Distinct functions of nuclear and cytoplasmic calcium in the control of gene expression. Nature 385(6613): 260-265.

Hardy SP, Goodfellow HR, Valverde MA, Gill DR, Sepulveda V, Higgins CF (1995). Protein kinase C-mediated phosphorylation of the human multidrug resistance P-glycoprotein regulates cell volume-activated chloride channels. EMBO J 14(1): 68-75.

Hartzell C, Putzier I, Arreola J (2005). Calcium-activated chloride channels. Annu Rev Physiol 67: 719-758.

Hartzell H, Qu Z, Fischmeister R (2004). Volume-sensitivity of the bestrophin family of chloride channels. Invest Ophthalmol Vis Sci 45(5): 1758.

Hartzell HC, Qu Z, Yu K, Xiao Q, Chien L-T (2008). Molecular physiology of bestrophins: multifunctional membrane proteins linked to Best disease and other retinopathies. Physiol Rev 88(2): 639-672.

Hassoun PM, Mouthon L, Barbera JA, Eddahibi S, Flores SC, Grimminger F, et al. (2009). Inflammation, growth factors, and pulmonary vascular remodeling. J Am Coll Cardiol 54(1 Suppl): S10-19.

Haug K, Warnstedt M, Alekov A, Sander T, Ramirez A, Poser B, et al. (2003). Mutations in CLCN2 encoding a voltage-gated chloride channel are associated with idiopathic generalized epilepsies. Nat Genet 33(4): 527-532.

He JZ, Ho JJD, Gingerich S, Courtman DW, Marsden PA, Ward ME (2010). Enhanced translation of heme oxygenase-2 preserves human endothelial cell viability during hypoxia. J Biol Chem 285(13): 9452-9461.

Hermoso M, Satterwhite CM, Andrade YN, Hidalgo J, Wilson SM, Horowitz B, et al. (2002). CLC-3 is a fundamental molecular component of volume-sensitive outwardly rectifying Cl- channels and volume regulation in HeLa cells and Xenopus laevis oocytes. J Biol Chem 277(42): 40066-40074.

106

Herve P, Launay JM, Scrobohaci ML, Brenot F, Simonneau G, Petitpretz P, et al. (1995). Increased plasma serotonin in primary pulmonary hypertension. Am J Med 99(3): 249-254.

Hickie R, Wei J, Blyth L, Wong D, Klaassen D (1983). Cations and calmodulin in normal and neoplastic cell growth regulation. Can J Biochem Cell Biol 61(8): 934-941.

Hill TD, Kindmark H, Boynton AL (1988). Epidermal growth factor-stimulated DNA synthesis requires an influx of extracellular calcium. J Cell Biochem 38(2): 137-144.

Hiraoka M, Kawano S, Hirano Y, Furukawa T (1998). Role of cardiac chloride currents in changes in action potential characteristics and arrhythmias. Cardiovasc Res 40(1): 23-33.

Hirayama Y, Kuruma A, Hiraoka M, Kawano S (2002). Calcium-activated Cl- current is enhanced by acidosis and contributes to the shortening of action potential duration in rabbit ventricular myocytes. Jpn J Physiol 52(3): 293-300.

Hironaka T, Morimoto S (1980). Intracellular chloride concentration and evidence for the existence of a chloride pump in frog skeletal muscle. Jpn J Physiol 30(3): 357-363.

Hoffmann EK, Lambert IH, Pedersen SF (2009). Physiology of cell volume regulation in vertebrates. Physiol Rev 89(1): 193-277.

Hoffmann EK, Simonsen LO (1989). Membrane mechanisms in volume and pH regulation in vertebrate cells. Physiol Rev 69(2): 315-382.

Holevinsky KO, Jow F, Nelson DJ (1994). Elevation in intracellular calcium activates both chloride and proton currents in human macrophages. J Membr Biol 140(1): 13-30.

Humbert M, Morrell NW, Archer SL, Stenmark KR, MacLean MR, Lang IM, et al. (2004). Cellular and molecular pathobiology of pulmonary arterial hypertension. J Am Coll Cardiol 43(12_Suppl_S): 13S-24.

Humbert M, Sitbon O, Chaouat A, Bertocchi M, Habib G, Gressin V, et al. (2010). Survival in patients with idiopathic, familial, and anorexigen-associated pulmonary arterial hypertension in the modern management era. Circulation 122(2): 156-163.

Humbert M, Sitbon O, Chaouat A, Bertocchi M, Habib G, Gressin V, et al. (2006). Pulmonary arterial hypertension in France: Results from a national registry. Am J Respir Crit Care Med 173(9): 1023-1030.

Husain M, Bein K, Jiang L, Alper SL, Simons M, Rosenberg RD (1997). c-Myb-dependent cell cycle progression and Ca2+ storage in cultured vascular smooth muscle cells. Circ Res 80(5): 617-626.

Jackson PS, Strange K (1995). Single-channel properties of a volume-sensitive anion conductance. Current activation occurs by abrupt switching of closed channels to an open state. J Gen Physiol 105(5): 643-660.

107 Janssen LJ, Sims SM (1995). Ca2+-dependent Cl- current in canine tracheal smooth muscle cells. Am J Physiol Cell Physiol 269(1): C163-C169.

Jennings RB, Reimer KA (1991). The cell biology of acute myocardial ischemia. Annu Rev Med 42: 225-246.

Jentsch T, Steinmeyer K, Schwarz G (1990). Primary structure of Torpedo marmorata chloride channel isolated by expression cloning in Xenopus oocytes. Nature 348(6301): 510-514.

Jentsch TJ (2008). CLC chloride channels and transporters: from genes to protein structure, pathology and physiology. Crit Rev Biochem Mol Biol 43(1): 3-36.

Jentsch TJ, Stein V, Weinreich F, Zdebik AA (2002). Molecular structure and physiological function of chloride channels. Physiol Rev 82(2): 503-568.

Jiang G, Klein JD, O'Neill WC (2001). Growth factors stimulate the Na-K-2Cl cotransporter NKCC1 through a novel Cl--dependent mechanism. Am J Physiol Cell Physiol 281(6): C1948-1953.

Jiang J, Backx P, Teoh H, Ward M (2003). Role of Cl- currents in rat aortic smooth muscle activation by prostaglandin F2 alpha. Eur J Pharmacol 481(2-3): 133-140.

Johnston GC (1977). Cell size and budding during starvation of the yeast Saccharomyces cerevisiae. J Bacteriol 132(2): 738-739.

Jordt S, Jentsch T (1997). Molecular dissection of gating in the CLC-2 chloride channel. EMBO J 16(7): 1582-1592.

Judge A, Sood V, Shaw J, Fang D, McClintock K, MacLachlan I (2005). Sequence-dependent stimulation of the mammalian innate immune response by synthetic siRNA. Nat Biotechnol 23(4): 457-462.

Kawai-Kowase K, Owens GK (2007). Multiple repressor pathways contribute to phenotypic switching of vascular smooth muscle cells. Am J Physiol Cell Physiol 292(1): C59-69.

Kawasaki M, Uchida S, Monkawa T, Miyawaki A, Mikoshiba K, Marumo F, et al. (1994). Cloning and expression of a protein kinase C-regulated chloride channel abundantly expressed in rat brain neuronal cells. Neuron 12(3): 597-604.

Kelly ME, Dixon SJ, Sims SM (1994). Outwardly rectifying chloride current in rabbit osteoclasts is activated by hyposmotic stimulation. J Physiol 475(3): 377-389.

Kerkut GA, Meech RW (1966). Microelectrode determination of the intracellular chloride concentration in nerve cells. Life Sci 5(5): 453-456.

Klausen TK, Bergdahl A, Hougaard C, Christophersen P, Pedersen SF, Hoffmann EK (2007). Cell cycle-dependent activity of the volume- and Ca2+-activated anion currents in Ehrlich lettre ascites cells. J Cell Physiol 210(3): 831-842.

108 Klockner U (1993). Intracellular calcium ions activate a low-conductance chloride channel in smooth-muscle cells isolated from human mesenteric artery. Pflugers Arch 424(3-4): 231- 237.

Koch M, Steinmeyer K, Lorenz C, Ricker K, Wolf F, Otto M, et al. (1992). The skeletal muscle chloride channel in dominant and recessive human myotonia. Science 257(5071): 797-800.

Kornak U, Kasper D, Bosl M, Kaiser E, Schweizer M, Schulz A, et al. (2001). Loss of the CLC-7 chloride channel leads to osteopetrosis in mice and man. Cell 104(2): 205-215.

Krapivinsky GB, Ackerman MJ, Gordon EA, Krapivinsky LD, Clapham DE (1994). Molecular characterization of a swelling-induced chloride conductance regulatory protein, pICln. Cell 76(3): 439-448.

Kuga T, Kobayashi S, Hirakawa Y, Kanaide H, Takeshita A (1996). Cell cycle-dependent expression of L- and T-type Ca2+ currents in rat aortic smooth muscle cells in primary culture. Circ Res 79(1): 14-19.

Kumar B, Dreja K, Shah SS, Cheong A, Xu SZ, Sukumar P, et al. (2006). Upregulated TRPC1 channel in vascular injury in vivo and its role in human neointimal hyperplasia. Circ Res 98(4): 557-563.

Kunzelmann K (2005). Ion channels and cancer. J Membr Biol 205(3): 159-173.

Kuruma A, Hartzell HC (2000). Bimodal control of a Ca2+-activated Cl- channel by different Ca2+ signals. J Gen Physiol 115(1): 59-80.

Lamb FS, Clayton GH, Liu BX, Smith RL, Barna TJ, Schutte BC (1999). Expression of CLCN voltage-gated chloride channel genes in human blood vessels. J Mol Cell Cardiol 31(3): 657- 666.

Lane KB, Machado RD, Pauciulo MW, Thomson JR, Phillips JA, Loyd JE, et al. (2000). Heterozygous germline mutations in BMPR2, encoding a TGF-beta receptor, cause familial primary pulmonary hypertension. Nat Genet 26(1): 81-84.

Lang F, Busch GL, Ritter M, Volkl H, Waldegger S, Gulbins E, et al. (1998). Functional significance of cell volume regulatory mechanisms. Physiol Rev 78(1): 247-306.

Lang F, Foller M, Lang K, Lang P, Ritter M, Vereninov A, et al. (2007). Cell volume regulatory ion channels in cell proliferation and cell death. Methods Enzymol 428: 209-225.

Large WA, Wang Q (1996). Characteristics and physiological role of the Ca2+-activated Cl- conductance in smooth muscle. Am J Physiol Cell Physiol 271(2): C435-454.

Ledoux J, Greenwood I, Villeneuve LR, Leblanc N (2003). Modulation of Ca2+-dependent Cl- channels by calcineurin in rabbit coronary arterial myocytes. J Physiol 552(Pt 3): 701-714.

Li H, Chen SJ, Chen YF, Meng QC, Durand J, Oparil S, et al. (1994). Enhanced endothelin-1 and endothelin receptor gene expression in chronic hypoxia. J Appl Physiol 77(3): 1451-1459.

109

Li X, Shimada K, Showalter LA, Weinman SA (2000). Biophysical properties of CLC-3 differentiate it from swelling-activated chloride channels in Chinese hamster ovary-K1 cells. J Biol Chem 275(46): 35994-35998.

Liang W, Ray JB, He JZ, Backx PH, Ward ME (2009). Regulation of proliferation and membrane potential by chloride currents in rat pulmonary artery smooth muscle cells Hypertension 54: 286-293.

Lin M-J, Leung GPH, Zhang W-M, Yang X-R, Yip K-P, Tse C-M, et al. (2004). Chronic hypoxia-induced upregulation of store-operated and receptor-operated Ca2+ channels in pulmonary arterial smooth muscle cells: A novel mechanism of hypoxic pulmonary hypertension. Circ Res 95(5): 496-505.

Linsdell P, Hanrahan JW (1996). Disulphonic stilbene block of cystic fibrosis transmembrane conductance regulator Cl- channels expressed in a mammalian cell line and its regulation by a critical pore residue. J Physiol 496(Pt 3): 687-693.

Liu Y, Sinha S, McDonald OG, Shang Y, Hoofnagle MH, Owens GK (2005). Kruppel-like factor 4 abrogates myocardin-induced activation of smooth muscle gene expression. J Biol Chem 280(10): 9719-9727.

Ma T, Thiagarajah JR, Yang H, Sonawane ND, Folli C, Galietta LJ, et al. (2002). Thiazolidinone CFTR inhibitor identified by high-throughput screening blocks cholera toxin-induced intestinal fluid secretion. J Clin Invest 110(11): 1651-1658.

Madden JA, Ray DE, Keller PA, Kleinman JG (2001). Ion exchange activity in pulmonary artery smooth muscle cells: the response to hypoxia. Am J Physiol Lung Cell Mol Physiol 280(2): L264-L271.

Manche L, Green SR, Schmedt C, Mathews MB (1992). Interactions between double-stranded RNA regulators and the protein kinase DAI. Mol Cell Biol 12(11): 5238-5248.

Manoury B, Tamuleviciute A, Tammaro P (2010). TMEM16A/Anoctamin1 protein mediates calcium-activated chloride currents in pulmonary arterial smooth muscle cells. J Physiol 588(13): 2305-2314.

Marquardt A, Stohr H, Passmore LA, Kramer F, Rivera A, Weber BH (1998). Mutations in a novel gene, VMD2, encoding a protein of unknown properties cause juvenile-onset vitelliform macular dystrophy (Best's disease). Hum Mol Genet 7(9): 1517-1525.

Marques J, Williams B (2005). Activation of the mammalian immune system by siRNAs. Nat Biotechnol 23(11): 1399-1405.

Matchkov VV, Aalkjaer C, Nilsson H (2004). A cyclic GMP-dependent calcium-activated chloride current in smooth-muscle cells from rat mesenteric resistance arteries. J Gen Physiol 123(2): 121-134.

110 Matchkov VV, Larsen P, Bouzinova EV, Rojek A, Boedtkjer DMB, Golubinskaya V, et al. (2008). Bestrophin-3 (vitelliform macular dystrophy 2-like 3 protein) is essential for the cGMP-dependent calcium-activated chloride conductance in vascular smooth muscle cells. Circ Res 103(8): 864-872.

Matsuda JJ, Filali MS, Moreland JG, Miller FJ, Lamb FS (2010). Activation of swelling- activated chloride current by tumor necrosis factor-alpha requires CLC-3-dependent endosomal reactive oxygen production. J Biol Chem 285(30): 22864-22873.

Matsuda JJ, Filali MS, Volk KA, Collins MM, Moreland JG, Lamb FS (2008). Overexpression of CLC-3 in HEK293T cells yields novel currents that are pH dependent. Am J Physiol Cell Physiol 294(1): C251-262.

Matsumura K, Yagi T, Hattori A, Soloviev M, Yasuda K (2010). Using single cell cultivation system for on-chip monitoring of the interdivision timer in Chlamydomonas reinhardtii cell cycle. J Nanobiotechnology 8(1): 23.

McLaughlin VV, McGoon MD (2006). Pulmonary arterial hypertension. Circulation 114(13): 1417-1431.

McLean LA, Roscoe J, Jorgensen NK, Gorin FA, Cala PM (2000). Malignant gliomas display altered pH regulation by NHE1 compared with nontransformed astrocytes. Am J Physiol Cell Physiol 278(4): C676-688.

Michea L, Ferguson DR, Peters EM, Andrews PM, Kirby MR, Burg MB (2000). Cell cycle delay and apoptosis are induced by high salt and urea in renal medullary cells. Am J Physiol Renal Physiol 278(2): F209-F218.

Middleton R, Pheasant D, Miller C (1996). Homodimeric architecture of a CLC-type chloride ion channel. Nature 383(6598): 337-340.

Middleton R, Pheasant D, Miller C (1994). Purification, reconstitution, and subunit composition of a voltage-gated chloride channel from Torpedo electroplax. Biochemistry 33(45): 13189- 13198.

Miguel-Velado E, Moreno-Dominguez A, Colinas O, Cidad P, Heras M, Perez-Garcia MT, et al. (2005). Contribution of Kv channels to phenotypic remodeling of human uterine artery smooth muscle cells. Circ Res 97(12): 1280-1287.

Milhavet O, Gary DS, Mattson MP (2003). RNA interference in biology and medicine. Pharmacol Rev 55(4): 629-648.

Miller C (2006). CLC chloride channels viewed through a transporter lens. Nature 440(7083): 484-489.

Miller FJ, Jr., Filali M, Huss GJ, Stanic B, Chamseddine A, Barna TJ, et al. (2007). Cytokine activation of nuclear factor kappa B in vascular smooth muscle cells requires signaling endosomes containing Nox1 and CLC-3. Circ Res 101(7): 663-671.

111 Morin XK, Bond TD, Loo TW, Clarke DM, Bear CE (1995). Failure of P-glycoprotein (MDR1) expressed in Xenopus oocytes to produce swelling-activated chloride channel activity. J Physiol 486 ( Pt 3): 707-714.

Morris AP, Frizzell RA (1993). Ca2+-dependent Cl- channels in undifferentiated human colonic cells (HT-29). II. Regulation and rundown. Am J Physiol 264(4 Pt 1): C977-985.

Naeije R (2005). Pulmonary hypertension and right heart failure in chronic obstructive pulmonary disease. Proc Am Thorac Soc 2(1): 20-22.

Nagai R, Kowase K, Kurabayashi M (2000). Transcriptional regulation of smooth muscle phenotypic modulation. Ann N Y Acad Sci 902: 214-222; discussion 222-213.

Nelson MT, Patlak JB, Worley JF, Standen NB (1990). Calcium channels, potassium channels, and voltage dependence of arterial smooth muscle tone. Am J Physiol Cell Physiol 259(1): C3-18.

Niedenzu C, Grasedyck K, Voelkel NF, Bittmann S, Lindner J (1981). Proliferation of lung cells in chronically hypoxic rats. An autoradiographic and radiochemical study. Int Arch Occup Environ Health 48(2): 185-193.

Nilius B (2001). Chloride channels go cell cycling. J Physiol 532(Pt 3): 581.

Nilius B, Droogmans G (2003). Amazing chloride channels: an overview. Acta Physiol Scand 177(2): 119-147.

Nilius B, Eggermont J, Voets T, Droogmans G (1996). Volume-activated Cl- channels. Gen Pharmacol 27(7): 1131-1140.

Nilius B, Prenen J, Szucs G, Wei L, Tanzi F, Voets T, et al. (1997a). Calcium-activated chloride channels in bovine pulmonary artery endothelial cells. J Physiol 498 ( Pt 2): 381-396.

Nilius B, Prenen J, Voets T, Eggermont J, Droogmans G (1998). Activation of volume-regulated chloride currents by reduction of intracellular ionic strength in bovine endothelial cells. J Physiol 506 ( Pt 2): 353-361.

Nilius B, Prenen J, Voets T, Van den Bremt K, Eggermont J, Droogmans G (1997b). Kinetic and pharmacological properties of the calcium-activated chloride current in macrovascular endothelial cells. Cell Calcium 22(1): 53-63.

Nilius B, Voets T, Prenen J, Barth H, Aktories K, Kaibuchi K, et al. (1999). Role of Rho and Rho kinase in the activation of volume-regulated anion channels in bovine endothelial cells. J Physiol 516 ( Pt 1): 67-74.

Nishimoto I, Wagner JA, Schulman H, Gardner P (1991). Regulation of Cl- channels by multifunctional CaM kinase. Neuron 6(4): 547-555.

112 O'Donnell M, Bush E, Holleman W, Owen N (1987). Biologically active atrial natriuretic peptides selectively activate Na/K/Cl cotransport in vascular smooth muscle cells. J Pharmacol Exp Ther 243(3): 822-828.

O'Donnell ME, Owen NE (1986). Atrial natriuretic factor stimulates Na/K/Cl cotransport in vascular smooth muscle cells. Proc Natl Acad Sci U S A 83(16): 6132-6136.

Okada Y (2006). Cell volume-sensitive chloride channels: phenotypic properties and molecular identity. Contrib Nephrol 152: 9-24.

Okada Y (1997). Volume expansion-sensing outward-rectifier Cl- channel: fresh start to the molecular identity and volume sensor. Am J Physiol Cell Physiol 273(3): C755-789.

Okada Y, Oiki S, Hazama A, Morishima S (1998). Criteria for the molecular identification of the volume-sensitive outwardly rectifying Cl- channel. Letter to the editor. J Gen Physiol 112(3): 365-367.

Okada Y, Oiki S, Tominaga M, Kubo M, Miwa A, Tominaga T, et al. (1997). Volume-sensitive Cl- channel in human epithelial cells: regulation by ATP and relation to P-glycoprotein. Jpn J Physiol 47 Suppl 1: S19-20.

Okada Y, Petersen CC, Kubo M, Morishima S, Tominaga M (1994). Osmotic swelling activates intermediate-conductance Cl- channels in human intestinal epithelial cells. Jpn J Physiol 44(4): 403-409.

Okada Y, Sato K, Numata T (2009). Pathophysiology and puzzles of the volume-sensitive outwardly rectifying anion channel. J Physiol 587(Pt 10): 2141-2149.

Olsen RW, DeLorey TM, Gordey M, Kang MH (1999). GABA receptor function and epilepsy. Adv Neurol 79: 499-510.

Op den Kamp JA (1979). Lipid asymmetry in membranes. Annu Rev Biochem 48: 47-71.

Osipenko ON, Alexander D, MacLean MR, Gurney AM (1998). Influence of chronic hypoxia on the contributions of non-inactivating and delayed rectifier K currents to the resting potential and tone of rat pulmonary artery smooth muscle. Br J Pharmacol 124(7): 1335-1337.

Owen NE (1984). Regulation of Na/K/Cl cotransport in vascular smooth muscle cells. Biochem Biophys Res Commun 125(2): 500-508.

Owen NE, Ridge KM (1989). Mechanism of angiotensin II stimulation of Na-K-Cl cotransport of vascular smooth muscle cells. Am J Physiol Cell Physiol 257(4): C629-636.

Owens GK (1995). Regulation of differentiation of vascular smooth muscle cells. Physiol Rev 75(3): 487-517.

Owens GK, Kumar MS, Wamhoff BR (2004). Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev 84(3): 767-801.

113 Pacaud P, Loirand G, Baron A, Mironneau C, Mironneau J (1991). Ca2+ channel activation and membrane depolarization mediated by Cl- channels in response to noradrenaline in vascular myocytes. Br J Pharmacol 104(4): 1000-1006.

Pacaud P, Loirand G, Gregoire G, Mironneau C, Mironneau J (1992). Calcium-dependence of the calcium-activated chloride current in smooth muscle cells of rat portal vein. Pflugers Arch 421(2-3): 125-130.

Papassotiriou J, Eggermont J, Droogmans G, Nilius B (2001). Ca2+-activated Cl- channels in Ehrlich ascites tumor cells are distinct from mCLCA1, 2 and 3. Pflugers Arch 442(2): 273- 279.

Paulmichl M, Li Y, Wickman K, Ackerman M, Peralta E, Clapham D (1992). New mammalian chloride channel identified by expression cloning. Nature 356(6366): 238-241.

Peacock AJ, Murphy NF, McMurray JJV, Caballero L, Stewart S (2007). An epidemiological study of pulmonary arterial hypertension. Eur Respir J 30(1): 104-109.

Pei Y, Tuschl T (2006). On the art of identifying effective and specific siRNAs. Nat Methods 3(9): 670-676.

Pendergrass WR, Angello JC, Kirschner MD, Norwood TH (1991). The relationship between the rate of entry into S phase, concentration of DNA polymerase alpha, and cell volume in human diploid fibroblast-like monokaryon cells. Exp Cell Res 192(2): 418-425.

Picollo A, Pusch M (2005). Chloride/proton antiporter activity of mammalian CLC proteins CLC-4 and CLC-5. Nature 436(7049): 420-423.

Piper AS, Large WA (2004). Direct effect of Ca2+-calmodulin on cGMP-activated Ca2+- dependent Cl- channels in rat mesenteric artery myocytes. J Physiol 559(Pt 2): 449-457.

Piper AS, Large WA (2003). Multiple conductance states of single Ca2+-activated Cl- channels in rabbit pulmonary artery smooth muscle cells. J Physiol 547(Pt 1): 181-196.

Piwon N, Gunther W, Schwake M, Bosl M, Jentsch T (2000). ClC-5 Cl- -channel disruption impairs endocytosis in a mouse model for Dent's disease. Nature 408(6810): 369-373.

Platoshyn O, Golovina VA, Bailey CL, Limsuwan A, Krick S, Juhaszova M, et al. (2000). Sustained membrane depolarization and pulmonary artery smooth muscle cell proliferation. Am J Physiol Cell Physiol 279(5): C1540-1549.

Pluteanu F, Cribbs LL (2011). Regulation and function of Cav3.1 T-type calcium channels in IGF-1 stimulated pulmonary artery smooth muscle cells. Am J Physiol Cell Physiol 300(3): C517-525.

Poet M, Kornak U, Schweizer M, Zdebik AA, Scheel O, Hoelter S, et al. (2006). Lysosomal storage disease upon disruption of the neuronal chloride transport protein CLC-6. Proc Natl Acad Sci U S A 103(37): 13854-13859.

114 Pozeg ZI, Michelakis ED, McMurtry MS, Thebaud B, Wu X-C, Dyck JRB, et al. (2003). In vivo gene transfer of the O2-sensitive potassium channel Kv1.5 reduces pulmonary hypertension and restores hypoxic pulmonary vasoconstriction in chronically hypoxic rats. Circulation 107(15): 2037-2044.

Pusch M, Jordt S-E, Stein V, Jentsch TJ (1999). Chloride dependence of hyperpolarization- activated chloride channel gates. J Physiol 515(2): 341-353.

Pusch M, Ludewig U, Rehfeldt A, Jentsch T (1995). Gating of the voltage-dependent chloride channel CLC-0 by the permeant anion. Nature 373(6514): 527-531.

Putney LK, Barber DL (2003). Na-H Exchange-dependent Increase in Intracellular pH Times G2/M Entry and Transition. J Biol Chem 278(45): 44645-44649.

Qu Z, Fischmeister R, Hartzell C (2004). Mouse bestrophin-2 is a bona fide Cl- channel: identification of a residue important in anion binding and conduction. J Gen Physiol 123(4): 327-340.

Quinn D, Dahlberg C, Bonventre J, Scheid C, Honeyman T, Joseph P, et al. (1996). The role of Na+/H+ exchange and growth factors in pulmonary artery smooth muscle cell proliferation. Am J Respir Cell Mol Biol 14(2): 139-145.

Quinn DA, Honeyman TW, Joseph PM, Thompson BT, Hales CA, Scheid CR (1991). Contribution of Na+/H+ exchange to pH regulation in pulmonary artery smooth muscle cells. Am J Respir Cell Mol Biol 5(6): 586-591.

Rabinovitch M, Gamble W, Nadas AS, Miettinen OS, Reid L (1979). Rat pulmonary circulation after chronic hypoxia: hemodynamic and structural features. Am J Physiol Heart Circ Physiol 236(6): H818-H827.

Ray JB, Arab S, Deng Y, Liu P, Penn L, Courtman DW, et al. (2008). Oxygen regulation of arterial smooth muscle cell proliferation and survival. Am J Physiol Heart Circ Physiol 294: H839-852.

Rees MI, Andrew M, Jawad S, Owen MJ (1994). Evidence for recessive as well as dominant forms of startle disease (hyperekplexia) caused by mutations in the alpha 1 subunit of the inhibitory glycine receptor. Hum Mol Genet 3(12): 2175-2179.

Remillard CV, Tigno DD, Platoshyn O, Burg ED, Brevnova EE, Conger D, et al. (2007). Function of Kv1.5 channels and genetic variations of KCNA5 in patients with idiopathic pulmonary arterial hypertension. Am J Physiol Cell Physiol 292(5): C1837-1853.

Resnick N, Collins T, Atkinson W, Bonthron DT, Dewey CF, Gimbrone MA (1993). Platelet- derived growth factor B chain promoter contains a cis-acting fluid shear-stress-responsive element. Proc Natl Acad Sci U S A 90(10): 4591-4595.

Resnick N, Gimbrone M (1995). Hemodynamic forces are complex regulators of endothelial gene expression. FASEB J 9(10): 874-882.

115 Reynolds AM, Holmes M, Morrell NW, Danilov S, Reynolds PN (2010). Gene delivery of bone morphogenetic protein receptor type-2 attenuates established hypoxic and monocrotaline- induced pulmonary hypertension. Am J Respir Crit Care Med 181(1_MeetingAbstracts): A6333.

Reynolds AM, Xia W, Holmes MD, Hodge SJ, Danilov S, Curiel DT, et al. (2007). Bone morphogenetic protein type 2 receptor gene therapy attenuates hypoxic pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol 292(5): L1182-1192.

Rhee SW, Stimers JR, Wang W, Pang L (2009). Vascular smooth muscle-specific knockdown of the noncardiac form of the L-type calcium channel by microRNA-based short hairpin RNA as a potential antihypertensive therapy. J Pharmacol Exp Ther 329(2): 775-782.

Riazanski V, Deriy LV, Shevchenko PD, Le B, Gomez EA, Nelson DJ (2011). Presynaptic CLC- 3 determines quantal size of inhibitory transmission in the hippocampus. Nat Neurosci 14(4): 487-494.

Rich IN, Worthington-White D, Garden OA, Musk P (2000). Apoptosis of leukemic cells accompanies reduction in intracellular pH after targeted inhibition of the Na+/H+ exchanger. Blood 95(4): 1427-1434.

Rich S, Dantzker DR, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, et al. (1987). Primary pulmonary hypertension: A national prospective study. Ann Intern Med 107(2): 216-223.

Rinke I, Artmann J, Stein V (2010). CLC-2 voltage-gated channels constitute part of the background conductance and assist chloride extrusion. J Neurosci 30(13): 4776-4786.

Rios EJ, Fallon M, Wang J, Shimoda LA (2005). Chronic hypoxia elevates intracellular pH and activates Na+/H+ exchange in pulmonary arterial smooth muscle cells. Am J Physiol Lung Cell Mol Physiol 289(5): L867-874.

Rodman DM, Reese K, Harral J, Fouty B, Wu S, West J, et al. (2005). Low-voltage-activated (T- type) calcium channels control proliferation of human pulmonary artery myocytes. Circ Res 96(8): 864-872.

Roman RM, Smith RL, Feranchak AP, Clayton GH, Doctor RB, Fitz JG (2001). CLC-2 chloride channels contribute to HTC cell volume homeostasis. Am J Physiol Gastrointest Liver Physiol 280(3): G344-G353.

Rosenberg HC, Rabinovitch M (1988). Endothelial injury and vascular reactivity in monocrotaline pulmonary hypertension. Am J Physiol 255(6 Pt 2): H1484-1491.

Rosenbohm A, Rudel R, Fahlke C (1999). Regulation of the human skeletal muscle chloride channel hCLC-1 by protein kinase C. J Physiol 514(3): 677-685.

Rossbach M (2010). Small non-coding RNAs as novel therapeutics. Curr Mol Med 10(4): 361- 368.

116 Rychkov GY, Pusch M, Astill DS, Roberts ML, Jentsch TJ, Bretag AH (1996). Concentration and pH dependence of skeletal muscle chloride channel CLC-1. J Physiol 497(Pt_2): 423- 435.

Sabirov RZ, Prenen J, Tomita T, Droogmans G, Nilius B (2000). Reduction of ionic strength activates single volume-regulated anion channels (VRAC) in endothelial cells. Pflugers Arch 439(3): 315-320.

Sachs HG, Stambrook PJ, Ebert JD (1974). Changes in membrane potential during the cell cycle. Exp Cell Res 83(2): 362-366.

Sakaguchi M, Matsuura H, Ehara T (1997). Swelling-induced Cl- current in guinea-pig atrial myocytes: inhibition by glibenclamide. J Physiol 505 ( Pt 1): 41-52.

Sato K, Webb S, Tucker A, Rabinovitch M, O'Brien RF, McMurtry IF, et al. (1992). Factors influencing the idiopathic development of pulmonary hypertension in the fawn hooded rat. Am Rev Respir Dis 145(4 Pt 1): 793-797.

Scheel O, Zdebik A, Lourdel S, Jentsch T (2005). Voltage-dependent electrogenic chloride/proton exchange by endosomal CLC proteins. Nature 436(7049): 424-427.

Schild D, Restrepo D (1998). Transduction mechanisms in vertebrate olfactory receptor cells. Physiol Rev 78(2): 429-466.

Schlenker T, Fitz JG (1996). Ca2+-activated Cl- channels in a human biliary cell line: regulation by Ca2+/calmodulin-dependent protein kinase. Am J Physiol 271(2 Pt 1): G304-310.

Scholl U, Hebeisen S, Janssen AGH, Muller-Newen G, Alekov A, Fahlke C (2006). Barttin modulates trafficking and function of CLC-K channels. Proc Natl Acad Sci U S A 103(30): 11411-11416.

Schroeder BC, Cheng T, Jan YN, Jan LY (2008). Expression cloning of TMEM16A as a calcium-activated chloride channel subunit. Cell 134(6): 1019-1029.

Schultz BD, Singh AK, Devor DC, Bridges RJ (1999). Pharmacology of CFTR chloride channel activity. Physiol Rev 79(1 Suppl): S109-144.

Schwarz D, Hutvagner G, Du T, Xu Z, Aronin N, Zamore P (2003). Asymmetry in the assembly of the RNAi enzyme complex. Cell 115(2): 199-208.

Sears RC, Nevins JR (2002). Signaling networks that link cell proliferation and cell fate. J Biol Chem 277(14): 11617-11620.

Shen M-R, Droogmans G, Eggermont J, Voets T, Ellory JC, Nilius B (2000). Differential expression of volume-regulated anion channels during cell cycle progression of human cervical cancer cells. J Physiol 529(2): 385-394.

117 Shimoda LA, Fallon M, Pisarcik S, Wang J, Semenza GL (2006). HIF-1 regulates hypoxic induction of NHE1 expression and alkalinization of intracellular pH in pulmonary arterial myocytes. Am J Physiol Lung Cell Mol Physiol 291(5): L941-949.

Shimoda LA, Manalo DJ, Sham JSK, Semenza GL, Sylvester JT (2001). Partial HIF-1 alpha deficiency impairs pulmonary arterial myocyte electrophysiological responses to hypoxia. Am J Physiol Lung Cell Mol Physiol 281(1): L202-208.

Short AD, Bian J, Ghosh TK, Waldron RT, Rybak SL, Gill DL (1993). Intracellular Ca2+ pool content is linked to control of cell growth. Proc Natl Acad Sci U S A 90(11): 4986-4990.

Simonneau G, Galie N, Rubin LJ, Langleben D, Seeger W, Domenighetti G, et al. (2004). Clinical classification of pulmonary hypertension. J Am Coll Cardiol 43(12_Suppl_S): 5S- 12.

Sitbon O, Humbert M, Jais X, Ioos V, Hamid AM, Provencher S, et al. (2005). Long-term response to calcium channel blockers in idiopathic pulmonary arterial hypertension. Circulation 111(23): 3105-3111.

Sitbon O, Humbert M, Nunes H, Parent F, Garcia G, Herve P, et al. (2002). Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: Prognostic factors and survival. J Am Coll Cardiol 40(4): 780-788.

Smirnov SV, Robertson TP, Ward JP, Aaronson PI (1994). Chronic hypoxia is associated with reduced delayed rectifier K+ current in rat pulmonary artery muscle cells. Am J Physiol Heart Circ Physiol 266(1): H365-370.

Stenmark KR, Fagan KA, Frid MG (2006). Hypoxia-induced pulmonary vascular remodeling: Cellular and molecular mechanisms. Circ Res 99(7): 675-691.

Stenmark KR, Meyrick B, Galie N, Mooi WJ, McMurtry IF (2009). Animal models of pulmonary arterial hypertension: the hope for etiological discovery and pharmacological cure. Am J Physiol Lung Cell Mol Physiol 297(6): L1013-1032.

Stewart SA, Dykxhoorn DM, Palliser D, Mizuno H, Yu EY, An DS, et al. (2003). Lentivirus- delivered stable gene silencing by RNAi in primary cells. RNA 9(4): 493-501.

Stobrawa SM, Breiderhoff T, Takamori S, Engel D, Schweizer M, Zdebik AA, et al. (2001). Disruption of CLC-3, a chloride channel expressed on synaptic vesicles, leads to a loss of the hippocampus. Neuron 29(1): 185-196.

Sun H, Tsunenari T, Yau KW, Nathans J (2002). The vitelliform macular dystrophy protein defines a new family of chloride channels. Proc Natl Acad Sci U S A 99(6): 4008-4013.

Suzuki T, Rai T, Hayama A, Sohara E, Suda S, Itoh T, et al. (2006). Intracellular localization of CLC chloride channels and their ability to form hetero-oligomers. J Cell Physiol 206(3): 792-798.

118 Suzuki T, Sawaki D, Aizawa K, Munemasa Y, Matsumura T, Ishida J, et al. (2009). Kruppel-like factor 5 shows proliferation-specific roles in vascular remodeling, direct stimulation of cell growth, and inhibition of apoptosis. J Biol Chem 284(14): 9549-9557.

Taichman D, Snow J, Pietra G (2006). Histopathology of pulmonary arterial hypertension. In: Mandel J, Taichman D (editors). Pulmonary vascular disease, edn. Philadelphia: Saunders. Page: 20-32.

Takenaka T, Epstein M, Forster H, Landry DW, Iijima K, Goligorsky MS (1992). Attenuation of endothelin effects by a chloride channel inhibitor, indanyloxyacetic acid. Am J Physiol 262(5 Pt 2): F799-806.

Tang YB, Liu YJ, Zhou JG, Wang GL, Qiu QY, Guan YY (2008). Silence of CLC-3 chloride channel inhibits cell proliferation and the cell cycle via G/S phase arrest in rat basilar arterial smooth muscle cells. Cell Prolif 41(5): 775-785.

Tao R, Lau C-P, Tse H-F, Li G-R (2008). Regulation of cell proliferation by intermediate- conductance Ca2+-activated potassium and volume-sensitive chloride channels in mouse mesenchymal stem cells. Am J Physiol Cell Physiol 295(5): C1409-1416.

Tarran R, Loewen ME, Paradiso AM, Olsen JC, Gray MA, Argent BE, et al. (2002). Regulation of murine airway surface liquid volume by CFTR and Ca2+-activated Cl- conductances. J Gen Physiol 120(3): 407-418.

Taxman D (2009). siRNA and shRNA design. In: Doran T, Helliwell C (editors). RNA interference: Methods for plants and animals, edn. Wallingford, UK: CAB International. Page: 228-253.

Thenappan T, Shah SJ, Rich S, Gomberg-Maitland M (2007). A USA-based registry for pulmonary arterial hypertension:1982-2006. Eur Respir J 30(6): 1103-1110.

Thiemann A, Grunder S, Pusch M, Jentsch T (1992). A chloride channel widely expressed in epithelial and non-epithelial cells. Nature 356(6364): 57-60.

Trimarchi JM, Lees JA (2002). Sibling rivalry in the E2F family. Nat Rev Mol Cell Biol 3(1): 11- 20.

Tsunenari T, Sun H, Williams J, Cahill H, Smallwood P, Yau KW, et al. (2003). Structure- function analysis of the bestrophin family of anion channels. J Biol Chem 278(42): 41114- 41125.

Ullrich N, Sontheimer H (1997). Cell cycle-dependent expression of a glioma-specific chloride current: proposed link to cytoskeletal changes. Am J Physiol Cell Physiol 273(4): C1290- 1297.

Valverde MA, Diaz M, Sepulveda FV, Gill DR, Hyde SC, Higgins CF (1992). Volume-regulated chloride channels associated with the human multidrug-resistance P-glycoprotein. Nature 355(6363): 830-833.

119 Vennekens R, Trouet D, Vankeerberghen A, Voets T, Cuppens H, Eggermont J, et al. (1999). Inhibition of volume-regulated anion channels by expression of the cystic fibrosis transmembrane conductance regulator. J Physiol 515(1): 75-85.

Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C (1995). A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. J Immunol Methods 184(1): 39-51.

Voelkel N, Tuder R (2000). Hypoxia-induced pulmonary vascular remodeling: a model for what human disease? J Clin Invest 106(6): 733-738.

Wagner JA, Cozens AL, Schulman H, Gruenert DC, Stryer L, Gardner P (1991). Activation of chloride channels in normal and cystic fibrosis airway epithelial cells by multifunctional calcium/calmodulin-dependent protein kinase. Nature 349(6312): 793-796.

Waldegger S, Jentsch TJ (2000). Functional and structural analysis of CLC-K chloride channels involved in renal disease. J Biol Chem 275(32): 24527-24533.

Wamhoff BR, Bowles DK, Owens GK (2006). Excitation-transcription coupling in arterial smooth muscle. Circ Res 98(7): 868-878.

Wang G-L, Wang X-R, Lin M-J, He H, Lan X-J, Guan Y-Y (2002). Deficiency in CLC-3 chloride channels prevents rat aortic smooth muscle cell proliferation. Circ Res 91(10): e28- 32.

Wang J, Juhaszova M, Rubin LJ, Yuan XJ (1997a). Hypoxia inhibits gene expression of voltage- gated K+ channel alpha subunits in pulmonary artery smooth muscle cells. J Clin Invest 100(9): 2347-2353.

Wang J, Weigand L, Lu W, Sylvester JT, L. Semenza G, Shimoda LA (2006). Hypoxia inducible factor 1 mediates hypoxia-induced TRPC expression and elevated intracellular Ca2+ in pulmonary arterial smooth muscle cells. Circ Res 98(12): 1528-1537.

Wang L, Chen L, Jacob TJ (2000). The role of CLC-3 in volume-activated chloride currents and volume regulation in bovine epithelial cells demonstrated by antisense inhibition. J Physiol 524(Pt 1): 63-75.

Wang YX, Kotlikoff MI (1997b). Inactivation of calcium-activated chloride channels in smooth muscle by calcium/calmodulin-dependent protein kinase. Proc Natl Acad Sci U S A 94(26): 14918-14923.

Wei L, Vankeerberghen A, Cuppens H, Cassiman JJ, Droogmans G, Nilius B (2001). The C- terminal part of the R-domain, but not the PDZ binding motif, of CFTR is involved in interaction with Ca2+-activated Cl- channels. Pflugers Arch 442(2): 280-285.

Wei L, Vankeerberghen A, Cuppens H, Eggermont J, Cassiman JJ, Droogmans G, et al. (1999). Interaction between calcium-activated chloride channels and the cystic fibrosis transmembrane conductance regulator. Pflugers Arch 438(5): 635-641.

120 Weir E, Archer S (1995). The mechanism of acute hypoxic pulmonary vasoconstriction: the tale of two channels. FASEB J 9(2): 183-189.

Weylandt K-H, Valverde MA, Nobles M, Raguz S, Amey JS, Diaz M, et al. (2001). Human CLC-3 is not the swelling-activated chloride channel involved in cell volume regulation. J Biol Chem 276(20): 17461-17467.

Whitman EM, Pisarcik S, Luke T, Fallon M, Wang J, Sylvester JT, et al. (2008). Endothelin-1 mediates hypoxia-induced inhibition of voltage-gated K+ channel expression in pulmonary arterial myocytes. Am J Physiol Lung Cell Mol Physiol 294(2): L309-L318.

Wondergem R, Gong W, Monen SH, Dooley SN, Gonce JL, Conner TD, et al. (2001). Blocking swelling-activated chloride current inhibits mouse liver cell proliferation. J Physiol 532(3): 661-672.

Wrong OM, Norden AGW, Feest TG (1994). Dent's disease: a familial proximal renal tubular syndrome with low-molecular-weight proteinuria, hypercalciuria, nephrocalcinosis, metabolic bone disease, progressive renal failure and a marked male predominance. QJM 87(8): 473-493.

Wu M, Tsai M, Tseng Y (1994). DIDS-sensitive pHi regulation in single rat cardiac myocytes in - nominally HCO3 free conditions. Circ Res 75(1): 123-132.

Xiao GN, Guan YY, He H (2002). Effects of Cl- channel blockers on endothelin-1-induced proliferation of rat vascular smooth muscle cells. Life sciences 70(19): 2233-2241.

Xiong D, Heyman NS, Airey J, Zhang M, Singer CA, Rawat S, et al. (2010). Cardiac specific, inducible CLC-3 gene deletion eliminates native volume-sensitive chloride channels and produces myocardial hypertrophy in adult mice. J Mol Cell Cardiol 48(1): 211-219.

Xiong H, Li C, Garami E, Wang Y, Ramjeesingh M, Galley K, et al. (1999). CLC-2 activation modulates regulatory volume decrease. J Membr Biol 167(3): 215-221.

Xu B, Mao J, Wang L, Zhu L, Li H, Wang W, et al. (2010). CLC-3 chloride channels are essential for cell proliferation and cell cycle progression in nasopharyngeal carcinoma cells. Acta Biochimica et Biophysica Sinica 42(6): 370-380.

Yamamoto K, Cox JPDT, Friedrich T, Christie PT, Bald M, Houtman PN, et al. (2000). Characterization of renal chloride channel (CLCN5) mutations in Dent's disease. J Am Soc Nephrol 11(8): 1460-1468.

Yamazaki J, Duan D, Janiak R, Kuenzli K, Horowitz B, Hume JR (1998). Functional and molecular expression of volume-regulated chloride channels in canine vascular smooth muscle cells. J Physiol 507(3): 729-736.

Yang J, Davies RJ, Southwood M, Long L, Yang X, Sobolewski A, et al. (2008a). Mutations in bone morphogenetic protein type II receptor cause dysregulation of Id gene expression in pulmonary artery smooth muscle cells: Implications for familial pulmonary arterial hypertension. Circ Res 102(10): 1212-1221.

121

Yang YD, Cho H, Koo JY, Tak MH, Cho Y, Shim WS, et al. (2008b). TMEM16A confers receptor-activated calcium-dependent chloride conductance. Nature 455(7217): 1210-1215.

Yi R, Qin Y, Macara IG, Cullen BR (2003). Exportin-5 mediates the nuclear export of pre- microRNAs and short hairpin RNAs. Genes & Dev 17(24): 3011-3016.

Yu L, Quinn DA, Garg HG, Hales CA (2008). Deficiency of the NHE1 gene prevents hypoxia- induced pulmonary hypertension and vascular remodeling. Am J Respir Crit Care Med 177(11): 1276-1284.

Yu Y, Fantozzi I, Remillard CV, Landsberg JW, Kunichika N, Platoshyn O, et al. (2004). Enhanced expression of transient receptor potential channels in idiopathic pulmonary arterial hypertension. Proc Natl Acad Sci U S A 101(38): 13861-13866.

Yu Y, Keller SH, Remillard CV, Safrina O, Nicholson A, Zhang SL, et al. (2009). A functional single-nucleotide polymorphism in the TRPC6 gene promoter associated with idiopathic pulmonary arterial hypertension. Circulation 119(17): 2313-2322.

Yu Y, Sweeney M, Zhang S, Platoshyn O, Landsberg J, Rothman A, et al. (2003). PDGF stimulates pulmonary vascular smooth muscle cell proliferation by upregulating TRPC6 expression. Am J Physiol Cell Physiol 284(2): C316-330.

Yuan JX-J, Aldinger AM, Juhaszova M, Wang J, Conte JV, Jr., Gaine SP, et al. (1998a). Dysfunctional voltage-gated K+ channels in pulmonary artery smooth muscle cells of patients with primary pulmonary hypertension. Circulation 98(14): 1400-1406.

+ 2+ Yuan X-J (1995). Voltage-gated K currents regulate resting membrane potential and [Ca ]i in pulmonary arterial myocytes. Circ Res 77(2): 370-378.

Yuan XJ (1997). Role of calcium-activated chloride current in regulating pulmonary vasomotor tone. Am J Physiol Lung Cell Mol Physiol 272(5): L959-968.

Yuan XJ, Wang J, Juhaszova M, Gaine SP, Rubin LJ (1998b). Attenuated K+ channel gene transcription in primary pulmonary hypertension. Lancet 351(9104): 726-727.

Zamore P, Tuschl T, Sharp P, Bartel D (2000). RNAi: double-stranded RNA directs the ATP- dependent cleavage of mRNA at 21 to 23 nucleotide intervals. Cell 101(1): 25-33.

Zeng Y (2009). RNAi in mammalian cells. In: Doran T, Helliwell C (editors). RNA interference: Methods for plants and animals, edn. Wallingford, UK: CAB International. Page: 205-227.

Zhang S, Dong H, Rubin LJ, Yuan JXJ (2007a). Upregulation of Na+/Ca2+ exchanger contributes to the enhanced Ca2+ entry in pulmonary artery smooth muscle cells from patients with idiopathic pulmonary arterial hypertension. Am J Physiol Cell Physiol 292(6): C2297-C2305.

Zhang W, Zhu Z, Chao Y, Lei Y, Cui J (2007b). [Effects of hypotonic solution on the 2+ proliferation, alkaline phosphatase activity and [Ca ]i concentration of osteoblast-like cells] (Article in Chinese). Sichuan Da Xue Xue Bao Yi Xue Ban 38(4): 649-652.

122

Zhang ZR, Zeltwanger S, McCarty NA (2000). Direct comparison of NPPB and DPC as probes of CFTR expressed in Xenopus oocytes. J Membr Biol 175(1): 35-52.

Zhao Z, Li X, Hao J, Winston JH, Weinman SA (2007). The CLC-3 chloride transport protein traffics through the plasma membrane via interaction of an N-terminal dileucine cluster with clathrin. J Biol Chem 282(39): 29022-29031.

Zhong J, Wang G-X, Hatton WJ, Yamboliev IA, Walsh MP, Hume JR (2002). Regulation of volume-sensitive outwardly rectifying anion channels in pulmonary arterial smooth muscle cells by PKC. Am J Physiol Cell Physiol 283(6): C1627-1636.

123

Appendix A: Publication List

Publications during Ph.D study (2006-2011)

1. Ashok Jadhav,* Wenbin Liang,* Guillaume Bastin, Jeff Kroetsch, John Balsevich, Scott Heximer, Peter H. Backx, and Venkat Gopalakrishnan. L-Tryptophan esters dilate rat mesenteric arterioles by inhibiting smooth muscle L-type Ca2+ channels. British Journal of Pharmacology (In Revision). *Authors who made equal contributions. 2. Wenbin Liang, Gavin Y. Oudit, Mikin M. Patel, Ajay M. Shah, James R. Woodgett, Robert G. Tsushima, Michael E. Ward and Peter H. Backx. Role of PI3Kα, PKC and L-type Ca2+ channels in mediating the complex actions of angiotensin II on mouse cardiac contractility. Hypertension, 2010;56:422-429. Comment in Hypertension, 2010;56:349-350. 3. Wenbin Liang, Julie B. Ray, Jeff Z. He, Peter H. Backx and Michael E. Ward. Regulation of proliferation and membrane potential by chloride currents in rat pulmonary artery smooth muscle cells. Hypertension, 2009;54:286-293. 4. Jing Li, Wanli Xuan, Ran Yan, Michael B. Tropak, Emilie Jean-St-Michel, Wenbin Liang, Rachel Gladstone, Peter H. Backx, Rajesh K. Kharbanda, and Andrew N. Redington. Remote preconditioning provides potent cardioprotection via PI3K/Akt activation and is associated with nuclear accumulation of β-catenin. Clinical Science (London), 2011;120:451-462. 5. Huijun Hu, Xiaowen Pan, Yi Wan, Qi Zhang, and Wenbin Liang. Prognostic factors of delayed encephalopathy after acute carbon monoxide poisoning. American Journal of Emergency Medicine, 2011;29:261-264. 6. Danny Guo, Zamaneh Kassiri, Fung L. Chow, Xiuhua Wang, Wenbin Liang, Emilio Hirsch, Josef M. Penninger, Peter H. Backx, and Gavin Y. Oudit. Loss of PI3Kγ enhances cAMP-dependent MMP remodeling of the myocardial N-cadherin adhesion complexes and extracellular matrix in response to early biomechanical stress. Circulation Research, 2010;107:1275-1289. 7. Changbin Yang, Shu Zhang, Yu Zhang, Bing Wang, Yongjie Yao, Yongchun Wang, Yanhong Wu, Wenbin Liang, Xiqing Sun. Combined short-arm centrifuge and aerobic exercise training improve cardiovascular function and physical working capacity in humans. Medical Science Monitor, 2010;16:CR575-583. 8. Xianlong Qi, Jian Gao, Dejie Sun, Wenbin Liang, Yi Wan, Chunying Li, Xiuli Xu, Tianwen Gao. Biofilm formation of the pathogens of fatal bacterial granuloma after trauma: Potential mechanism underlying the failure of traditional antibiotic treatments. Scandinavian Journal of Infectious Diseases, 2008;40:221-228. 9. Fujun Shang, Lianyou Zhao, Qiangsun Zheng, Jiepin Wang, Zhe Xu, Wenbin Liang, Hui Liu, Shaowei Liu, and Lijuan Zhang. Simvastatin inhibits lipopolysaccharide-induced TNF-α expression in neonatal rat cardiomyocytes: The role of reactive oxygen species. Biochemical and Biophysical Research Communication, 2006;351:947-52.

124

Appendix B: Published Paper 1

Regulation of Proliferation and Membrane Potential by Chloride Currents in Rat Pulmonary Artery Smooth Muscle Cells. Wenbin Liang, Julie Basu Ray, Jeff Z. He, Peter H. Backx, and Michael E. Ward. Hypertension, 2009;54:286-293.

Regulation of Proliferation and Membrane Potential by Chloride Currents in Rat Pulmonary Artery Smooth Muscle Cells

Wenbin Liang, Julie Basu Ray, Jeff Z. He, Peter H. Backx, Michael E. Ward

Abstract—Pulmonary artery smooth muscle cell (PASMC) proliferation contributes to increased pulmonary vascular

resistance and pulmonary hypertension. Because proliferation depends on membrane potential (Vm) and because Vm is, Ϫ Ј in part, determined by Cl currents (ICl), we examined the effects of ICl inhibition with 4,4 -diisothiocyanatostilbene- 2,2Ј-disulfonic acid (DIDS) on cultured rat PASMCs. DIDS (30 ␮mol/L) reduced cell numbers, decreased

5-bromodeoxyuridine incorporation and delayed cell cycle progression. ICl inhibition with 5-Nitro-2-(3- phenylpropylamino) benzoic acid (100 ␮mol/L) also reduced cell numbers of cultured rat PASMCs. To test the possible

involvement of ICl in the regulation of PASMC proliferation, we measured Vm and ICl in both cultured (proliferating) Ϫ Ϯ and acutely dissociated (nonproliferating) rat PASMCs. Vm ( 39.3 1.4 mV) was close to the equilibrium potential of ClϪ (Ϫ39 mV) in proliferating PASMCs but differed from equilibrium potential of ClϪ in acutely dissociated cells Ϫ Ϯ Ϫ Ϫ ( 45.3 0.9 mV). DIDS and substitution of extracellular Cl with I induced Vm hyperpolarization in proliferating but not nonproliferating PASMCs. Consistent with Vm recordings, DIDS-sensitive baseline and swelling-activated 2ϩ 2ϩ Ͻ Ϸ (Ca -independent) ICls, recorded with low Ca ( 1 nmol/L) pipette solutions, were 5-fold greater in proliferating 2ϩ than in nonproliferating PASMCs. By contrast, Ca -activated ICl did not differ between proliferating and nonprolif- 2ϩ erating PASMCs. Ca -independent ICls were also increased in proliferating PASMCs acutely dissociated from rats exposed to hypoxia (10% O2; 7 days). These findings are consistent with the conclusion that ICls regulate proliferation of PASMCs and suggest that selective ICl inhibition may be useful in treating pulmonary hypertension. (Hypertension. 2009;54:286-293.) Key Words: Ca2ϩ-independent chloride currents Ⅲ pulmonary hypertension Ⅲ hypoxia Ⅲ cell cycle Ⅲ phenotypic transition

ulmonary hypertension, uniformly fatal in its primary plasma membrane depolarization.10,11 Moreover, previous Pidiopathic form and a cause of premature mortality when studies have identified that cytoplasmic volume must be it complicates cardiopulmonary disease, remains a difficult adjusted during cell cycle progression9 and that membrane ϩ management problem. Pulmonary artery smooth muscle cell potential and intracellular free Ca2 levels are important (SMC; PASMC) proliferation is a hallmark of the disease and regulators of proliferation.9,12 These observations suggest a 1,2 contributes to increased vascular resistance to blood flow. role for the ICl in PASMC proliferation. The present study Understanding the mechanisms that regulate PASMC prolif- was, therefore, carried out to test the hypothesis that ICls eration is necessary in order that new targets for therapeutic participate in the regulation of PASMC proliferation and to intervention may be identified. compare the ICls active in proliferating versus contractile Ϫ 2ϩ PASMCs. Swelling-activated chloride (Cl ) currents (ICls) and Ca - 3,4 activated ICls have been identified in PASMCs, and PASMC expression of ClC-3, which is a candidate for Materials and Methods 5 mediation of swelling-activated ICl and is important for cell Detailed methods are available in the online data supplement proliferation,6,7 is upregulated in the pulmonary hypertensive (http://hyper.ahajournals.org). 8 Ϫ All of the procedures were approved by the institutional animal rats. Because of the transmembrane Cl concentration gra- 13 Ϫ care committee. Single PASMCs were isolated from small intralo- dients that exist in vascular smooth muscle, activation of Cl bar arteries (third- and fourth-order branches) of normoxic rats (adult Ϫ 9 channels causes Cl efflux, cell volume decrease, and male Sprague-Dawley) and rats exposed to hypoxia (10% O2;7

Received February 2, 2009; first decision February 20, 2009; revision accepted May 28, 2009. From the Departments of Physiology and Medicine (W.L., P.H.B.), Division of Respirology, Department of Medicine (J.B.R., J.Z.H., M.E.W.), Division of Cardiology, University Health Network (W.L., P.H.B.), University of Toronto; and Li Ka Shing Knowledge Institute (J.B.R., J.Z.H., M.E.W.), St Michael’s Hospital, Toronto, Ontario, Canada. Correspondence to Peter H. Backx, Heart and Stroke/Richard Lewar Centre, 150 College St, Toronto, Ontario, Canada M5S 3E2. E-mail [email protected] © 2009 American Heart Association, Inc. Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.109.130138

Downloaded from hyper.ahajournals.org at286 University of Toronto on September 15, 2010 Liang et al Chloride Currents in Pulmonary Artery SMCs 287

A Vehicle ) ) 6 Vehicle 2 10 µmol/L DIDS 2 DIDS 2 30 µmol/L DIDS * 100 µmol/L DIDS cells/cm cells/cm 4 4 # 4 # 1 2 # # # Cell number (x10Cell 0 Cell number (x10 0 0 543 Time (days) B Vehicle DIDS 100 Vehicle Figure 1. A (left), Effects of incubation 60 DIDS with DIDS for 4 days on cell numbers of cultured rat PASMCs. Cell numbers were 80 reduced by 20.4%, 37.8%, and 92.3% by 40 E E DIDS at 10, 30, and 100 ␮mol/L, respec- 11.8% 56.4% N N 60 * tively. Right, Effects of incubation with 86.7% 41.2% DIDS (30 ␮mol/L) for 3, 4, and 5 days on 20

t BrdUrd positive cells cell numbers of cultured rat PASMCs. B Cell number 40 and C, Effects of incubation with DIDS (30 ␮mol/L) for 16 hours on numbers of

Percen 0 10 0 10 1 10 0 10 1 BrdUrd-positive cells (B) and cell-cycle BrdUrd fluorecence (arbitrary units) phase distribution (C) in cultured rat PASMCs. *PϽ0.05; #PϽ0.01 vs corresponding values in vehicle-treated cells. C G0/G1 S G2/M %cells

G0/G1 S G2/M Vehicle Vehicle 47.3±0.8 14.1±2.1 38.6±1.6

DIDS 61.3±4.9 * 15.8±4.1 22.9±4.8 *

DIDS

0 20 40 60 80 100 % cells in cell cycle stage

days). These cells were cultured in the presence of 10% FBS, as Results 14 Ј described previously. The effects of the ICl inhibitor (4,4 - To determine whether I s are involved in SMC prolifer- diisothiocyanatostilbene-2,2Ј-disulfonic acid [DIDS]; 30 ␮mol/L) on Cl 5-bromodeoxyuridine (BrdUrd) incorporation and DNA content ation, we examined the effects of the ICl inhibitor DIDS on were assessed by flow cytometry as described previously.15 Resting cell numbers in cultured rat PASMCs. As summarized in Ͻ ϭ membrane potential (Vm) of rat PASMCs was measured using the Figure 1A, DIDS reduced (P 0.01; n 6) cell numbers in whole-cell patch-clamp technique in current-clamp mode using the a dose- and time-dependent manner. Cell viability, as nystatin-perforated recording configuration16 with bath solution assessed using trypan blue exclusion, was not affected by containing the following (in mmol/L): NaCl 135.0, KCl 5.0, CaCl2 ϭ DIDS at 10 or 30 ␮mol/L but was reduced (PϽ0.01) by 1.8, MgCl2 1.2, HEPES 10.0, and glucose 10.0 (pH 7.4) and pipette solution containing the following (in mmol/L): KCl 30, Kϩ- Ϸ10% in cells treated with 100 ␮mol/L of DIDS (Figure ␮ gluconate 118, MgCl2 1, and HEPES 10, as well as nystatin 200 S1). DIDS (30 mol/L) also had no effect on the percent- ␮ ϭ g/mL (pH 7.2). ICls were recorded with a whole-cell patch-clamp age of cells staining positive for Annexin V and/or technique in voltage-clamp mode in nystatin-perforated or conven- propidium iodide (data not shown), suggesting the lack of tional ruptured recording configuration, with bath solution contain- effect of DIDS on apoptotic cell death at this concentra- ing the following (in mmol/L): N-methyl-D-glucamine (NMDG)- ϭ tion. To further confirm that DIDS inhibits cell replication, HCl 132.0, MgCl2 1.5, HEPES 10.0, and glucose 10.0 (pH 7.4 with NMDG) and pipette solution containing the following BrdUrd incorporation was measured. After treatment for (in mmol/L): NMDG-HCl 30, NMDG-gluconate 93, HEPES 5, 16 hours, DIDS decreased (PϽ0.05; nϭ3) the percentage ϭ EGTA 10, and Mg-ATP 5 (pH 7.2 with NMDG). In some of BrdUrd-positive cells by Ϸ36% (Figure 1B). Impor- studies, NMDG-HCl in the bath solution was reduced to Ͻ ϭ 100 mmol/L (osmolarityϭ225 milliosmoles) to induce cell swell- tantly, DIDS also increased (P 0.05; n 3) the percentage 2ϩ of cells in G and G phases while decreasing (PϽ0.05) the ing. In other studies, Ca -activated ICls were stimulated by 0 1 adding 7.7 mmol/L of CaCl2 to the pipette solution (free percentage of cells in the G2/M phase (Figure 1C), indi- Ca2ϩϭ500 nmol/L). The liquid junction potential between pipette cating that DIDS delayed the G1/S-phase transition. Thus, solutions and bath solutions was calculated and corrected. The the changes in cell numbers in PASMCs treated with DIDS current density of IC1 was measured in picoampere/picofarad Յ ␮ (pA/pF). Differences between 2 means were evaluated by 2-tailed at concentrations of 30 mol/L were attributed to effects Student t test. Differences among multiple means were assessed on PASMC proliferation and not changes in cell survival. by 1-way ANOVA and evaluated posthoc by Bonferroni’s test. Because DIDS affects multiple ion flux mechanisms, we

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 288 Hypertension August 2009

A Cultured Acutely dissociated B Cultured Acutely dissociated DIDS (µmol/L) DIDS (µmol/L) I- substitution 30 100 3001003010 I- substitution −40mV −40mV −60 mV −60 mV

10 mV 10 mV 1 min 2 min

Cultured -15 Cultured Acutely dissociated Acutely dissociated -40 * * * -10 * -20 -5 #

0 0

Membrane potential changes (mV) 110100 [DIDS] (µmol/L) Membrane potential changes (mV) Figure 2. A (top), Representative recordings showing membrane hyperpolarization after treatment with DIDS in cultured (left) and acutely dissociated (right) rat PASMCs. Bottom, Concentration-response relationship for DIDS-induced membrane hyperpolarization in cultured (F nϭ4 to 7 cells) and acutely dissociated (E nϭ4 to 11 cells) rat PASMCs. Solid and dashed lines represent ﬁtted relation- Ϯ Ϯ ships from which maximum hyperpolarization (29.3 0.2 mV for cultured and 6.3 1.0 mV for acutely dissociated cells) and IC50 values were calculated. *PϽ0.01 for difference from corresponding values in acutely dissociated cells. B (top), Representative recordings of membrane potential changes after substitution of external ClϪ with IϪ in cultured (left) and acutely dissociated (right) rat PASMCs. Bot- tom, IϪ substitution–induced membrane potential changes in cultured (Ϫ10.2Ϯ4.2 mV; nϭ6 cells) and acutely dissociated (nϭ11 cells) rat PASMCs. *PϽ0.01 vs values before IϪ substitution; #PϽ0.01 vs changes in cultured cells.

confirmed the effects of ICl blockade on rat PASMC ICls are larger in proliferating than in nonproliferating proliferation by testing another ICl blocker, 5-Nitro-2-(3- PASMCs. phenylpropylamino) benzoic acid (NPPB), which is struc- To more directly investigate ICls in these cells, nystatin- turally different from DIDS. NPPB (100 ␮mol/L) treat- perforated patch-clamp recordings of PASMCs were made ment reduced (PϽ0.01; nϭ5) cell numbers of rat PASMCs under conditions where cation currents were minimized (see by 68.9% without affecting cell viability (97.4Ϯ0.5% Methods). As shown in Figure 3A, voltage steps from a versus vehicle-treated 97.2Ϯ1.0%; Pϭ0.87). holding potential of Ϫ40 mV induced currents with “instan- To explore the ionic mechanisms responsible for re- taneous” activation kinetics and with minimal inactivation duced proliferation of DIDS-treated rat PASMCs, we used during a 1-second voltage step. The relationship between nystatin-perforated whole-cell recordings to investigate the current density and voltage in cultured PASMCs showed effects of DIDS on membrane potential (V ) and I sin prominent outward rectification with an estimated reversal m Cl Ϫ Ϯ ϭ PASMCs. In cultured rat PASMCs, the resting V was potential (Erev)of 32.9 1.9 mV (n 15; Figure 3B), which m Ϫ ␮ Ϫ39.3Ϯ1.4 mV (nϭ78 cells), which was close to the was similar to ECl ( 38.9 mV). The addition of 100 mol/L Ϫ PϽ ϩ equilibrium potential of Cl (E ; Ϫ38.9 mV) in the study. of DIDS reduced ( 0.05) the current density at 60 mV by Cl 86.0Ϯ7.7% in cultured PASMCs (from 12.2Ϯ2.9 to Figure 2A shows that DIDS additions induced a 1.69Ϯ0.96 pA/pF; nϭ4), confirming that the currents are concentration-dependent V hyperpolarization in cultured m generated by DIDS-sensitive ClϪ channels. By comparison, rat PASMCs. At maximum levels of DIDS, V was shifted m in nonproliferating PASMCs acutely dissociated from nor- (PϽ0.05; nϭ6) to more negative potentials by Ϸ30 mV. moxic rats (Figure 3), the current-voltage relationship showed The estimated DIDS concentration eliciting a 50% change in Vm Ϫ Ϯ Ͻ little rectification, Erev ( 56.1 1.8 mV) differed (P 0.01) was estimated to be Ϸ20 ␮mol/L. By comparison, the effect of ϩ Ϯ from ECl, and the current density at 60 mV (1.46 0.16 DIDS on acutely dissociated (nonproliferating) cells was small, pA/pF; nϭ11) was lower (PϽ0.01) than that in cultured Ϸ with the maximal Vm shift to more negative potentials being 6 PASMCs. mV at saturating DIDS levels. The currents recorded above were measured under basal The results above indicate that the resting Vm is more (ie, isotonic) conditions. Previous studies have established strongly regulated by I s in cultured PASMCs than in acutely 17 Cl that swelling-activated ICls, which are sensitive to DIDS, dissociated PASMCs. Consistent with this suggestion, Figure are important for cell cycle progression,9,19,20 suggesting Ϫ Ϫ 2B shows that replacing extracellular Cl with I (an anion that blockade of swelling-activated I s may underlie the Ϫ Ϫ Cl that is typically more permeant than Cl for most Cl reduced proliferation observed in DIDS-treated cultured 17,18 Ͻ channels ) induced a hyperpolarization shift (P 0.01; PASMCs. Therefore, ICl recordings were made in response ϭ Ϸ n 6) of Vm by 10 mV in cultured PASMCs. In contrast, in to cell swelling induced by the application of hypotonic Ϫ Ϫ acutely dissociated cells, Cl replacement with I did not solutions using the ruptured patch-clamp configuration, ϭ Ϫ Ϯ Ϫ alter (P 0.46) Vm ( 48.5 2.2 mV before I and and the pipette solutions contained Ͻ1 nmol/L of free Ϫ50.7Ϯ1.9 mV after IϪ;nϭ11). These findings indicate that Ca2ϩ. The current density at ϩ60 mV of the baseline currents

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 Liang et al Chloride Currents in Pulmonary Artery SMCs 289

Figure 3. A, Representative ClϪ currents recorded with the nystatin-perforated conﬁguration in cultured (left) and acutely dissociated (right) rat PASMCs. Middle, Effect of DIDS (100 ␮mol/L) on currents in cultured PASMCs. Dashed lines indicate the position of 0 current. B, Current-voltage relationships for ClϪ currents in cultured (F nϭ15 cells) and acutely dissociated (Œ nϭ11 cells) rat PASMCs, as well as for currents in cultured cells in the presence of 100 ␮mol/L of DIDS (E nϭ4 cells). *PϽ0.01 vs corresponding values in acutely dissociated, untreated cells. recorded in cultured PASMCs in isotonic conditions (without sequence of anion permeability (ie, IϪϾClϪϾgluconate) Ϫ Ϯ ϭ 17 cell swelling) was 2.2 0.7 pA/pF (n 7). Switching to reported for swelling-activated ICl. Cl replacements also hypotonic bath solutions caused cell swelling in these cells caused similar shifts in Erev in baseline currents recorded in and induced large outwardly rectifying (Ca2ϩ-independent) isotonic bath solutions (Figure S2B). ϩ ICls (Figure 4A, middle, and Figure 4B) with a density at 60 In PASMCs acutely dissociated from normoxic rats, mV of 65.6Ϯ20.4 pA/pF (nϭ6). Consistent with the conclu- switching to hypotonic bath solutions also caused cell swell- ϭ sion that these swelling-activated currents were ICls, DIDS ing but did not increase (P 0.17) the currents (Figure 5A): (100 ␮mol/L) reduced (PϽ0.05) the current density at ϩ60 current density at ϩ60 mV was 0.33Ϯ0.04 pA/pF (nϭ26) in mV to 5.0Ϯ1.7 pA/pF (nϭ4; Figure 4A). Moreover, analysis isotonic solutions and 0.23Ϯ0.03 pA/pF (nϭ13) in hypotonic of the DIDS-sensitive currents, calculated as the difference solutions. Moreover, the currents recorded in hypotonic ϭ ϭ between the currents before and after DIDS, showed an Erev of solutions were not affected (P 0.80; n 5) by DIDS (data not Ϫ Ϯ ϭ Ϫ 29.5 2.2mV(n 4), which was close to ECl ( 29.2 mV). shown). The above results establish that proliferating rat To further establish that the swelling-activated currents PASMCs display prominent baseline and swelling-activated 2ϩ recorded in cultured PASMCs were ICls, we measured the Ca -independent ICls, which are virtually absent in acutely Ϫ changes in Erev with external Cl replacements. When dissociated nonproliferating PASMCs. Ϫ 2ϩ external Cl was replaced with gluconate, an anion that is PASMCs also possess prominent Ca -activated ICls, Ϫ 17,21 22 relatively impermeant to Cl channels, Erev shifted which appear to play roles in cell proliferation and in ϩ Ͻ Ϯ ϭ 23 2 (P 0.01) to more positive potentials (by 35.8 2.7 mV; n 8; regulating Vm. To assess Ca -activated ICls in our rat Ͻ Figure S2A, left). Conversely, Erev shifted (P 0.01) to more PASMCs, ruptured patch-clamp recordings were made using negative potentials (by 23.3Ϯ4.2 mV; nϭ7) when external pipette solutions with a free Ca2ϩ concentration of 500 ClϪ was replaced with IϪ (Figure S2A, right). The anion nmol/L and using isotonic bath solutions. As shown in Figure permeability ratios for gluconate and IϪ versus ClϪ were S3A, prominent currents were observed under these condi- 0.23Ϯ0.03 and 2.8Ϯ0.47, respectively, consistent with the tions in both cultured and acutely dissociated PASMCs.

A B I (pA/pF) 150 *

1000 pA * Hypotonic solution Isotonic 200 mS 100 Hypotonic solution Hypotonic * Isotonic solution + DIDS Hypotonic + DIDS * 50 * *

-120 -80 -40 40 80 120 * * * V (mV) -50 Figure 4. Ca2ϩ-independent ClϪ currents recorded with ruptured whole-cell conﬁguration with pipette solutions containing Ͻ1 nmol/L of free Ca2ϩ in cultured rat PASMCs. A, ClϪ currents recorded in isotonic (left) and hypotonic (middle) bath solutions and in hypotonic solutions in the presence of 100 ␮mol/L of DIDS (right). Dashed lines indicate the position of 0 current. B, Current-voltage relationships for ClϪ currents in isotonic (F nϭ7 cells) and hypotonic (E nϭ6 cells) solutions and in hypotonic solutions in the presence of 100 ␮mol/L of DIDS (f nϭ4 cells). *PϽ0.01 vs values in isotonic solution.

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 290 Hypertension August 2009

A Acute PASMCs from normoxic rats B Acute PASMCs from hypoxia-exposed rats

25 pA Isotonic solution 25 pA Hypotonic solution 200 mS Hypotonic solution Isotonic solution 200 mS

I (pA/pF) I(pA/pF) 15 *

* Isotonic 0.5 Isotonic 10 * Hypotonic Hypotonic 5 *

-120 -80 - 40 40 80 120 -120 -80 -40 40 80 120 V (mV) V (mV) * -5 * -0.5 * -10

Figure 5. Ca2ϩ-independent ClϪ currents recorded with ruptured whole-cell conﬁguration with pipette solutions containing Ͻ1 nmol/L of free Ca2ϩ in PASMCs acutely dissociated from normoxic or hypoxic rats. A, PASMCs acutely dissociated from normoxic rats. Top, Representative currents recorded in isotonic (left) and hypotonic (right) bath solutions. Dashed lines indicate the position of 0 current. Bottom, Current-voltage relationships of currents in isotonic (F nϭ26 cells) and hypotonic (E nϭ13 cells) solutions. B, PASMCs acutely Ϫ dissociated from rats exposed to hypoxia (10% O2; 7 days). Top, Representative Cl currents recorded in isotonic (left) and hypotonic (right) bath solutions. Dashed lines indicate the position of 0 current. Bottom, Current-voltage relationships for ClϪ currents in isotonic (F nϭ12 cells) and hypotonic (E nϭ4 cells) solutions. *PϽ0.01 vs values in isotonic solution.

2ϩ 26 These currents contained both Ca -activated ICls and base- rates of replication. However, these cells can transit from 2ϩ ϩ line Ca -indepedent ICls. The current density at 60 mV a nonproliferative contractile phenotype to a proliferative was much larger (PϽ0.01) in cultured PASMCs (10.4Ϯ2.8 “synthetic” phenotype,27,28 and this phenotypic transition pA/pF; nϭ15) than that in acutely dissociated PASMCs and subsequent proliferation in pulmonary arteries contrib- (3.5Ϯ0.7 pA/pF; nϭ24; Figure S3B). However, subtraction ute to pulmonary hypertension.24 Our study is the first to 2ϩ of the baseline Ca -indepednet ICls(ICls recorded in isotonic demonstrate that transition of PASMCs to a proliferative bath solutions in Figures 4A and 5A) from these currents phenotype is associated with upregulation of DIDS- 2ϩ recorded with high pipette Ca (500 nmol/L) revealed that sensitive Ca2ϩ-independent I s, and the application of 2ϩ Ͼ Cl the Ca -activated ICls were not different (P 0.30; Figure DIDS reduces cell numbers of proliferating PASMCs. S3C) as assessed using the parameters originated from These effects on cell numbers by DIDS appeared to result Boltzmann fits. from a decreased proliferation rate, specifically from Because hypoxia is a potent stimulus for PASMC delayed cell cycle progression through the G1/S transition, proliferation in vivo,24,25 we hypothesized that exposure of 2ϩ because cultured proliferating rat PASMCs treated with rats to hypoxic conditions will induce elevation in Ca - DIDS accumulated in the G /G phase of the cell cycle and independent I s in PASMCs. Consistent with previous 0 1 Cl had reduced BrdUrd incorporation. studies,25 exposure of rats to hypoxia (10% O ) for 7 days 2 We conclude, for the reasons outlined below, that the induced proliferation of PASMCs, as demonstrated by the effects of DIDS on PASMC proliferation result from the increased BrdUrd uptake of PASMCs in hypoxic rats blockade of ICls and likely from the blockade of DIDS- compared with normoxic rats (data not shown). Consistent 2ϩ 2ϩ sensitive Ca -independent “baseline” and “swelling-acti- with our hypothesis, Ca -independent ICls recorded with low Ca2ϩ (Ͻ1 nmol/L) pipette solutions in PASMCs vated” ICls. First, the resting Vm of proliferating PASMCs Ϸ acutely dissociated from hypoxia-exposed rats (Figure 5B) was close to the ECl, and DIDS induced an 30-mV had current densities at ϩ60 mV of 0.9Ϯ0.2 pA/pF (nϭ12) hyperpolarization of the Vm, causing Vm to become closer Ϫ in isotonic bath solutions and 4.3Ϯ0.3 pA/pF (nϭ7) after to EK ( 85 mV). By contrast, DIDS had very little effect the application of hypotonic solutions to induce cell on nonproliferating PASMCs. These results suggest that 2ϩ swelling. Both baseline and swelling-activated Ca - ICls play a more dominant role in setting Vm in prolifer- independent I s were greater (PϽ0.01) than the corre- ating than in nonproliferating PASMCs. Consistent with Cl Ϫ Ϫ sponding currents recorded in PASMCs acutely dissoci- this suggestion, Cl replacement with I , an anion with a Ϫ ated from normoxic rats (Figure 5A). higher permeability than Cl for both baseline and 2ϩ swelling-activated Ca -independent ICls in rat PASMCs Ϸ Discussion (Figure S2), induced an 10 mV shift in Vm to more Vascular SMCs in mature mammals are highly differenti- hyperpolarized potentials, whereas ClϪ replacement with ated and, under normal physiological conditions, show low gluconate, an anion with a lower permeability than ClϪ

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 Liang et al Chloride Currents in Pulmonary Artery SMCs 291

32–34 (Figure S2), shifted Vm to more depolarized potentials regulators. In addition to inducing membrane depolar- (data not shown). These findings are consistent with the ization, the larger swelling-activated ICls in proliferating anion permeability properties of swelling-activated ICls PASMCs may facilitate the obligatory changes in cell Ϫ recorded in other tissues as reviewed.17,29 Conversely, volume during cell cycling.9,17,35 Cl efflux via swelling- these anion substitutions had virtually no effect on Vm in activated ICls is thought to protect against excessive cell acutely dissociated nonproliferating PASMCs. These re- swelling because of active uptake of amino acids and other 9 sults establish that ICls active in cultured proliferating substrates during the G1 phase and may play a similar role PASMCs promote membrane depolarization, which can in proliferating PASMCs. 2ϩ 36 serve as a mitogenic stimulus (see below).9 Inhibition of Ca -activated ICls with DIDS may also

Further support for a connection between elevated ICls contribute to the reduced proliferation in our cultured rat 2ϩ and PASMC proliferation is provided by the results of our PASMCs, because Ca -activated ICls are also expressed in these cells. The Ca2ϩ-activated I s were not different voltage-clamp studies, which were designed to isolate ICls Cl between proliferating and nonproliferating contractile in whole-cell recordings (see Methods). The ICls recorded PASMCs, suggesting that proliferation of PASMCs may in both isotonic bath solutions (ie, baseline ICls) and not require elevation in these currents. However, because hypotonic (osmolarity decrease of 70 milliosmoles) bath ϩ cytosolic [Ca2 ] is higher in cultured PASMCs,12 it is solutions (ie, swelling-activated ICls) in ruptured-patch ϩ 2ϩ 2 recordings with low pipette Ca levels (Ͻ1 nmol/L), as conceivable that Ca -activated ICls will be more active in 2ϩ proliferating cells. Although Ca -activated ICls regulate well as ICl recorded in isotonic bath solutions with the 18,23 nystatin-patch method, were much larger in cultured pro- resting Vm and vascular tone in contractile PASMCs, liferating PASMCs compared with acutely dissociated their role in proliferation of PASMCs remains unclear be- nonproliferating PASMCs. Moreover, consistent with our cause of the lack of specific antagonists. The recent identification of TMEM16A as the gene for encoding Ca2ϩ-activated Vm recordings, DIDS application reduced these currents by 36–38 Ϸ ICls may motivate further studies using molecular tools to 90% in cultured PASMCs while having no effect in 2ϩ nonproliferating PASMCs. Our conclusion that these clarify the role of Ca -activated ICls. DIDS-sensitive (baseline and swelling-activated) currents The underlying molecular basis for, as well as the mechanism for the upregulation of, the Ca2ϩ-independent in proliferating PASMCs were indeed ICls is bolstered by Ϫ ICls observed in our proliferating PASMCs is unknown. In the large shifts in Erev observed when extracellular Cl Ϫ Ϫ cultured rat PASMCs, the baseline and swelling-activated was replaced with I (negative shifts in E ) or gluconate rev Ca2ϩ-independent I s displayed similar biophysical and (positive shifts in E ; Figure S2). Cl rev pharmacological properties (baseline I s were also sensi- Another important finding supporting the conclusion Cl tive to DIDS; data not shown), suggesting that the same that I s can regulate PASMC proliferation is our obser- Ϫ Cl Cl channel might generate both currents. The permeabil- vation that exposure of rats to hypoxia for 7 days induced Ϫ Ϫ ity ratio of I versus Cl for the baseline currents (7.7), the proliferation of SMCs in the intact pulmonary artery, however, was higher than that for swelling-activated cur- and PASMCs acutely dissociated from these hypoxia- rents (2.8; Figure S2). Therefore, baseline and swelling- exposed rats showed prominent baseline and swelling-ac- activated I s in proliferating rat PASMCs may be medi- tivated Ca2ϩ-independent I s, which were virtually absent Cl Cl ated by different channels, although it is possible that in nonproliferating PASMCs acutely dissociated from channel properties may be altered by increased membrane normoxic rats. Thus, I expression strongly correlates Cl tension associated with cell swelling.39 Although several with the proliferation rates of PASMCs, whether the candidate genes have been suggested,40,41 the genes encod- stimulus for proliferation is hypoxia in vivo or mitogens 42 ing swelling-activated ICls remain unclear, and additional present in the serum of culture medium. work will be necessary to determine the molecular basis The mechanism whereby baseline and swelling-acti- for the increased ICls in proliferating PASMCs. Many vated ICls regulate PASMC proliferation is unclear. Be- transcription factors are involved in the altered gene 10,11 cause ECl is typically above Vm in SMCs, the presence expressions during the transition of VSMCs from the of larger ICls in proliferating PASMCs is expected to contractile to proliferative phenotype.26,43 For example, promote depolarization. Indeed, in the present study, the Kru¨ppel-like factor 4 is involved in the suppression of resting Vm in proliferating cultured PASMCs is more differentiation marker genes,44 whereas Kru¨ppel-like fac- Ϸ depolarized (by 6 mV) than in acutely dissociated tor 5 is involved in the expression of the embryonic PASMCs. Similar membrane depolarizations were ob- isoform of smooth muscle myosin heavy chain (SMemb/ served in proliferating cultured human uterine artery NMHC-B),45,46 which is a marker gene of proliferative SMCs,28 PASMCs acutely dissociated from hypoxia- VSMCs. It is possible that these transcription factors may exposed rats,30 and proliferating cultured rat PASMCs.12 also regulate other aspects of the VSMC phenotypic Membrane depolarization, as seen in proliferating modulation, including the upregulation of ICls in PASMCs, 2ϩ PASMCs, is predicted to increase [Ca ]i via the presence as observed in the present study. Accordingly, further 2ϩ of a “window current” for voltage-dependent Ca chan- studies may be needed to test this possibility. nels, which, in SMCs, has a steep voltage dependence ϩ between Ϫ50 and Ϫ30 mV.31 This Ca2 entry has been Perspectives shown to promote PASMC proliferation,12 possibly as a The results of the current study establish that the phenotypic result of the Ca2ϩ-dependent regulation of cell cycle transition of rat PASMCs from the contractile state to the

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 292 Hypertension August 2009 proliferative state is associated with upregulation of DIDS- 13. Archer SL, Huang JMC, Reeve HL, Hampl V, Tolarova S, Michelakis E, sensitive, Ca2ϩ-independent, baseline and swelling-activated Weir EK. Differential distribution of electrophysiologically distinct myocytes in conduit and resistance arteries determines their response to ICls, and blockade of these ICls reduces proliferation rates of nitric oxide and hypoxia. Circ Res. 1996;78:431–442. PASMCs. These observations may have direct and immediate 14. Yuan X-J. Voltage-gated Kϩ currents regulate resting membrane 2ϩ relevance to pathological conditions associated with exces- potential and [Ca ]i in pulmonary arterial myocytes. Circ Res. 1995;77: sive PASMC proliferation, such as pulmonary hypertension. 370–378. 15. Ray JB, Arab S, Deng Y, Liu P, Penn L, Courtman DW, Ward ME. The development of pharmacological tools that permit selec- Oxygen regulation of arterial smooth muscle cell proliferation and tive blockade of ICls may be useful in ameliorating PASMC survival. Am J Physiol Heart Circ Physiol. 2008;294:H839–H852. Ϫ proliferation and the progression of pulmonary hypertension 16. Jiang J, Backx P, Teoh H, Ward M. Role of Cl currents in rat aortic smooth muscle activation by prostaglandin F alpha. Eur J Pharmacol. and right heart decompensation. 2 2003;481:133–140. 17. Okada Y. Volume expansion-sensing outward-rectifier ClϪ channel: fresh Acknowledgments start to the molecular identity and volume sensor. Am J Physiol Cell We thank Drs Jason X.-J. Yuan and Carmelle V. Remillard (Uni- Physiol. 1997;273:C755–C789. 18. Large WA, Wang Q. Characteristics and physiological role of the Ca2ϩ- versity of California San Diego) for help with primary culture of rat Ϫ pulmonary artery SMCs. We also thank Dongling Zhao and Dr Peter activated Cl conductance in smooth muscle. Am J Physiol Cell Physiol. Helli (University of Toronto) for technical assistance. 1996;271:C435–C454. 19. Chen LX, Zhu LY, Jacob TJ, Wang LW. Roles of volume-activated ClϪ currents and regulatory volume decrease in the cell cycle and Sources of Funding proliferation in nasopharyngeal carcinoma cells. Cell Prolif. 2007;40: This study was funded by a research grant from the Canadian 253–267. Institutes of Health Research (to M.E.W. and P.H.B.) and a Canadian 20. Wondergem R, Gong W, Monen SH, Dooley SN, Gonce JL, Conner TD, Institutes of Health Research grant to P.H.B. W.L. was a recipient of Houser M, Ecay TW, Ferslew KE. Blocking swelling-activated chloride a Doctoral Research Award from the Heart and Stroke Foundation of current inhibits mouse liver cell proliferation. J Physiol. 2001;532: Canada. P.H.B. holds a Career Investigator Award from the Heart 661–672. and Stroke Foundation of Ontario. 21. Chen B, Nicol G, Cho WK. Electrophysiological characterization of volume-activated chloride currents in mouse cholangiocyte cell line. Am J Disclosures Physiol Gastrointest Liver Physiol. 2004;287:G1158–G1167. 22. Klausen TK, Bergdahl A, Hougaard C, Christophersen P, Pedersen SF, None. Hoffmann EK. Cell cycle-dependent activity of the volume- and Ca2ϩ- activated anion currents in Ehrlich lettre ascites cells. J Cell Physiol. References 2007;210:831–842. 1. Humbert M, Morrell NW, Archer SL, Stenmark KR, MacLean MR, Lang 23. Yuan XJ. Role of calcium-activated chloride current in regulating pul- IM, Christman BW, Weir EK, Eickelberg O, Voelkel NF, Rabinovitch M monary vasomotor tone. Am J Physiol Lung Cell Mol Physiol. 1997;272: Cellular and molecular pathobiology of pulmonary arterial hypertension. L959–L968. J Am Coll Cardiol. 2004;43:13S–24S. 24. Stenmark KR, Fagan KA, Frid MG. Hypoxia-induced pulmonary 2. McLaughlin VV, McGoon MD. Pulmonary arterial hypertension. vascular remodeling: cellular and molecular mechanisms. Circ Res. 2006; Circulation. 2006;114:1417–1431. 99:675–691. 3. Zhong J, Wang G-X, Hatton WJ, Yamboliev IA, Walsh MP, Hume JR. 25. Niedenzu C, Grasedyck K, Voelkel NF, Bittmann S, Lindner J. Pro- Regulation of volume-sensitive outwardly rectifying anion channels in liferation of lung cells in chronically hypoxic rats: an autoradiographic pulmonary arterial smooth muscle cells by PKC. Am J Physiol Cell and radiochemical study. Int Arch Occup Environ Health. 1981;48: Physiol. 2002;283:C1627–C1636. 185–193. 4. Angermann JE, Sanguinetti AR, Kenyon JL, Leblanc N, Greenwood IA. 26. Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular Mechanism of the inhibition of Ca2ϩ-activated ClϪ currents by phosphor- smooth muscle cell differentiation in development and disease. Physiol ylation in pulmonary arterial smooth muscle cells. J Gen Physiol. 2006; Rev. 2004;84:767–801. 128:73–87. 27. Owens GK. Regulation of differentiation of vascular smooth muscle cells. 5. Hermoso M, Satterwhite CM, Andrade YN, Hidalgo J, Wilson SM, Physiol Rev. 1995;75:487–517. Horowitz B, Hume JR. ClC-3 is a fundamental molecular component of 28. Miguel-Velado E, Moreno-Dominguez A, Colinas O, Cidad P, Heras M, Ϫ volume-sensitive outwardly rectifying Cl channels and volume regu- Perez-Garcia MT, Lopez-Lopez JR. Contribution of Kv channels to phe- lation in HeLa Cells and Xenopus laevis oocytes. J Biol Chem. 2002; notypic remodeling of human uterine artery smooth muscle cells. Circ 277:40066–40074. Res. 2005;97:1280–1287. Ϫ 6. Habela CW, Olsen ML, Sontheimer H. ClC3 is a critical regulator of the 29. Nilius B, Eggermont J, Voets T, Droogmans G. Volume-activated Cl cell cycle in normal and malignant glial cells. J Neurosci. 2008;28: channels. Gen Pharmacol. 1996;27:1131–1140. 9205–9217. 30. Osipenko ON, Alexander D, MacLean MR, Gurney AM. Influence of 7. Wang G-L, Wang X-R, Lin M-J, He H, Lan X-J, Guan Y-Y. Deficiency chronic hypoxia on the contributions of non-inactivating and delayed in ClC-3 chloride channels prevents rat aortic smooth muscle cell prolif- rectifier Kϩ currents to the resting potential and tone of rat pulmonary eration. Circ Res. 2002;91:e28–e32. artery smooth muscle. Br J Pharmacol. 1998;124:1335–1337. 8. Dai YP, Bongalon S, Hatton WJ, Hume JR, Yamboliev IA. ClC-3 31. Fleischmann B, Murray R, Kotlikoff M. Voltage window for sustained chloride channel is upregulated by hypertrophy and inflammation in rat elevation of cytosolic calcium in smooth muscle cells. Proc Natl Acad Sci and canine pulmonary artery. Br J Pharmacol. 2005;145:5–14. USA. 1994;91:11914–11918. 9. Kunzelmann K. Ion channels and cancer. J Membr Biol. 2005;205: 32. Hardingham G, Chawla S, Johnson C, Bading H. Distinct functions of 159–173. nuclear and cytoplasmic calcium in the control of gene expression. 10. Aickin C. Chloride Channels and Carriers in Nerve, Muscle, and Glial Nature. 1997;385:260–265. Cells. New York, NY: Plenum Press; 1990. 33. Husain M, Bein K, Jiang L, Alper SL, Simons M, Rosenberg RD. c-Myb 11. Davis J, Harper A, Chipperfield A. Stimulation of intracellular chloride dependent cell cycle progression and Ca2ϩ storage in cultured vascular accumulation by noradrenaline and hence potentiation of its depolar- smooth muscle cells. Circ Res. 1997;80:617–626. ization of rat arterial smooth muscle in vitro. Br J Pharmacol. 1997;122: 34. Choi J, Chiang A, Taulier N, Gros R, Pirani A, Husain M. A calmodu- 2ϩ 639–642. lin-binding site on cyclin E mediates Ca -sensitive G1/S transitions in 12. Platoshyn O, Golovina VA, Bailey CL, Limsuwan A, Krick S, Juhaszova vascular smooth muscle cells. Circ Res. 2006;98:1273–1281. M, Seiden JE, Rubin LJ, Yuan JX-J. Sustained membrane depolarization 35. Ullrich N, Sontheimer H. Cell cycle-dependent expression of a glioma- and pulmonary artery smooth muscle cell proliferation. Am J Physiol Cell specific chloride current: proposed link to cytoskeletal changes. Am J Physiol. 2000;279:C1540–C1549. Physiol Cell Physiol. 1997;273:C1290–C1297.

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 Liang et al Chloride Currents in Pulmonary Artery SMCs 293

36. Yang YD, Cho H, Koo JY, Tak MH, Cho Y, Shim WS, Park SP, Lee J, 41. Paulmichl M, Li Y, Wickman K, Ackerman M, Peralta E, Clapham D. Lee B, Kim BM, Raouf R, Shin YK, Oh U. TMEM16A confers receptor- New mammalian chloride channel identified by expression cloning. activated calcium-dependent chloride conductance. Nature. 2008;455: Nature. 1992;356:238–241. 1210–1215. 42. Okada Y, Oiki S, Hazama A, Morishima S. Criteria for the molecular 37. Schroeder BC, Cheng T, Jan YN, Jan LY. Expression cloning of identification of the volume-sensitive outwardly rectifying ClϪ channel TMEM16A as a calcium-activated chloride channel subunit. Cell. 2008; [letter]. J Gen Physiol. 1998;112:365–367. 134:1019–1029. 43. Wamhoff BR, Bowles DK, Owens GK. Excitation-transcription coupling 38. Caputo A, Caci E, Ferrera L, Pedemonte N, Barsanti C, Sondo E, Pfeffer in arterial smooth muscle. Circ Res. 2006;98:868–878. U, Ravazzolo R, Zegarra-Moran O, Galietta LJV. TMEM16A, a 44. Liu Y, Sinha S, McDonald OG, Shang Y, Hoofnagle MH, Owens GK. membrane protein associated with calcium-dependent chloride channel Kruppel-like factor 4 abrogates myocardin-induced activation of activity. Science. 2008;322:590–594. smooth muscle gene expression. J Biol Chem. 2005;280:9719–9727. 39. Dai J, Sheetz MP, Wan X, Morris CE. Membrane tension in swell- 45. Nagai R, Kowase K, Kurabayashi M. Transcriptional regulation of smooth ing and shrinking molluscan neurons. J Neurosci. 1998;18:6681– muscle phenotypic modulation. Ann N Y Acad Sci. 2000;902:214–222. 6692. 46. Suzuki T, Sawaki D, Aizawa K, Munemasa Y, Matsumura T, Ishida J, 40. Duan D, Winter C, Cowley S, Hume JR, Horowitz B. Molecular identi- Nagai R. Kruppel-like factor 5 shows proliferation-specific roles in fication of a volume-regulated chloride channel. Nature. 1997;390: vascular remodeling, direct stimulation of cell growth, and inhibition of 417–421. apoptosis. J Biol Chem. 2009;284:9549–9557.

Downloaded from hyper.ahajournals.org at University of Toronto on September 15, 2010 133

Appendix C: Published Paper 2

Role of Phosphoinositide 3-Kinase α, Protein Kinase C, and L-Type Ca2+ Channels in Mediating the Complex Actions of Angiotensin II on Mouse Cardiac Contractility. Wenbin Liang, Gavin Y. Oudit, Mikin M. Patel, Ajay M. Shah, James R. Woodgett, Robert G. Tsushima, Michael E. Ward, and Peter H. Backx. Hypertension, 2010;56:422-429.

Role of Phosphoinositide 3-Kinase ␣, Protein Kinase C, and L-Type Ca2؉ Channels in Mediating the Complex Actions of Angiotensin II on Mouse Cardiac Contractility

Wenbin Liang, Gavin Y. Oudit, Mikin M. Patel, Ajay M. Shah, James R. Woodgett, Robert G. Tsushima, Michael E. Ward, Peter H. Backx

Abstract—Although angiotensin II (Ang II) plays an important role in heart disease associated with pump dysfunction, its direct effects on cardiac pump function remain controversial. We found that after Ang II infusion, the developed ϩ pressure and dP/dtmax in isolated Langendorff-perfused mouse hearts showed a complex temporal response, with a rapid transient decrease followed by an increase above baseline. Similar time-dependent changes in cell shortening and L-type Ca2ϩ currents were observed in isolated ventricular myocytes. Previous studies have established that Ang II signaling involves phosphoinositide 3-kinases (PI3K). Dominant-negative inhibition of PI3K␣ in the myocardium selectively eliminated the rapid negative inotropic action of Ang II (inhibited by Ϸ90%), whereas the loss of PI3K␥ had no effect on the response to Ang II. Consistent with a link between PI3K␣ and protein kinase C (PKC), PKC inhibition (with GF 109203X) reduced the negative inotropic effects of Ang II by Ϸ50%. Although PI3K␣ and PKC activities are associated with glycogen synthase kinase-3␤ and NADPH oxidase, genetic ablation of either glycogen synthase kinase-3␤ or p47phox (an essential subunit of NOX2-NADPH oxidase) had no effect on the inotropic actions of Ang II. Our results establish that Ang II has complex temporal effects on contractility and L-type Ca2ϩ channels in normal mouse myocardium, with the negative inotropic effects requiring PI3K␣ and PKC activities. (Hypertension. 2010;56:422-429.) Key Words: cardiac contraction Ⅲ Langendorff heart Ⅲ patch clamp Ⅲ glycogen synthase kinase-3␤ Ⅲ NADPH oxidase

ngiotensin II (Ang II) plays an important role in cardio- (PI3K) in cardiac myocytes13,14 and vascular smooth mus- Avascular physiology and pathology. Circulating Ang II cle.15 Because the class IA PI3K, PI3K␣, enhances cardiac 16 17,18 levels and the activity of the local cardiac renin-angiotensin contraction strength and ICa,L while the class IB PI3K, system are increased in heart disease, including cardiac PI3K␥, decreases cardiac contractility, accelerates cAMP hypertrophy1 and failure.2 Chronic Ang II stimulation has degradation,19,20 and increases ␤-adrenergic receptor down- been linked to cardiac remodeling (characterized by intersti- regulation,21 we examined the involvement of PI3K␣ and tial fibrosis, myocyte hypertrophy and death, and metabolism PI3K␥ in mediating the inotropic effects of Ang II in mouse alterations),3,4 which contributes to decreased mechanical myocardium. Our results show that Ang II has complex function in heart disease. However, the acute actions of Ang effects on mouse cardiac contractility and ICa,L, and that II on cardiac function remain unclear and are controversial PI3K␣, but not PI3K␥, is required for the negative inotropic with positive,5–8 negative,9 or no10,11 effects on cardiac effects of Ang II. These actions of Ang II are independent of contractility reported. Moreover, the signaling pathways me- glycogen synthase kinase-3␤ (GSK3␤) or NOX2–NADPH diating the inotropic effects of Ang II are not fully clear. The oxidase activity. positive inotropic effects of Ang II have been previously ϩ linked to protein kinase C (PKC),8 L-type Ca2 currents Materials and Methods 6 ␤ 8 Detailed methods are available in the online Supplement (available (ICa,L), and -Arrestin-2, as well as secondary endothelin-1 7 online at http://hyper.ahajournals.org). release, whereas negative inotropic effects have been asso- All procedures were approved by the local animal care committee ciated with PKC9,12 and p38 MAPK9 activities. Ang II has at the University of Toronto. C57BL/6 mice (male, 8–12 weeks) also been shown to activate phosphoinositide 3-kinases were obtained from Charles River Laboratories (Montreal, Canada).

Received December 23, 2009; first decision January 14, 2010; revision accepted July 8, 2010. From the Departments of Physiology and Medicine (W.L., M.M.P., P.H.B.), Division of Cardiology (W.L., M.M.P., P.H.B.), University Health Network (W.L., M.M.P., P.H.B.), University of Toronto, Toronto, Canada; Division of Cardiology (G.Y.O.), Department of Medicine, and Mazankowski Alberta Heart Institute (G.Y.O.), University of Alberta, Edmonton, Canada; Cardiovascular Division (A.M.S.), King’s College London British Heart Foundation Centre of Research Excellence, London, United Kingdom; Samuel Lunenfeld Research Institute (J.R.W.), Mount Sinai Hospital, Toronto, Canada; Department of Biology (R.G.T.), York University, Toronto, Canada; Division of Respirology (M.E.W.) and the Li Ka Shing Knowledge Institute (M.E.W.), St. Michael’s Hospital, University of Toronto, Toronto, Canada. Correspondence to Peter H. Backx, Room 68, Fitzgerald building, 150 College Street, Toronto, Ontario, Canada. E-mail [email protected] © 2010 American Heart Association, Inc. Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.109.149344 422 Liang et al Ang II Has Complex Inotropic Effects 423

A 3 nmol/L Ang II -5 min 0 min 5 min 8 min 6000 3 nmol/L Ang II 110

4000 (mmHg/s) max

2000 +dP/dt

LV pressure (mmHg) 0 0 -5 0 5 10 15 Time (min) B Figure 1. A, Representative left ventricle (LV) pres- 5000 5000 ϩ ϭ 3 nmol/L Ang II 30 nmol/L Ang II sure traces (left) and dP/dtmax (right; n 4) of mouse hearts during infusion of angiotensin (Ang) P1075 P1075 II (3 nmol/L). Hearts were perfused using the Lan- gendorff method at a constant perfusion pressure. 4000 4000 ϩ B, The dP/dtmax time course during Ang II infu- (mmHg/s) sion at 3 nmol/L (top left; nϭ5) and 30 nmol/L (top max right; nϭ4). Hearts were perfused at a constant 3000 3000 coronary ﬂow rate in the presence of a vasodilator (P1075, 100 nmol/L). Summary of peak early- +dP/dt 0 0 phase decreases (bottom left) and peak late-phase ϩ -5 0 5 10 15 -5 0 5 10 15 increases (bottom right) in dP/dtmax of hearts Time (min) Time (min) treated with 3, 30, or 300 nmol/L (nϭ4) Ang II. 20 30 max max

15 20

10 10 (% of baseline)

(% of baseline) 5

0 0 Peak increase in +dP/dt Peak decrease in +dP/dt 3 30 300 3 30 300 [Ang II] (nmol/L) [Ang II] (nmol/L)

Mice lacking PI3K␣ activity in the myocardium (dominant-negative function. Figure 1A shows typical left ventricle pressure traces Ϫ Ϫ [DN]-PI3K␣)22 and mice lacking PI3K␥ activity (PI3K␥ / )23 or Ϫ Ϫ recorded at the indicated times after Ang II (3 nmol/L) infusion. NOX2-NADPH oxidase activity (p47phox / )24 have been de- Ang II caused complex temporal changes in pressure develop- scribed. To generate conditional cardiac-specific GSK3␤ knockout flox/flox Ͻ ϭ mice, GSK3␤ mice25 were crossed with mice expressing ment characterized by rapid reductions (P 0.01; n 4) of the ␣ 26 ␣ ϩ tamoxifen-inducible -myosin MerCreMer. The -myosin Mer- peak rate of left ventricle pressure development ( dP/dtmax)by CreMer/GSK3␤flox/flox mice (12 weeks old) were treated with 32.0%Ϯ4.7% below baseline (from 3154Ϯ175 to 2206Ϯ tamoxifen citrate (20 mg/kg per day for 4 days), which reduced Ϸ ␤ Ͼ 215 mm Hg/sec) at 5 minutes after Ang II. After this rapid cardiac GSK3 protein levels by 90%. A total of 90 mice were ϩ Ͻ used in Langendorff studies. In Langendorff studies, hearts were reduction, dP/dtmax increased (P 0.01) and peaked at perfused via the aorta with a modified Krebs solution containing the 69.8%Ϯ4.5% above (PϽ0.01; nϭ4) baseline (ie, 5336Ϯ following (in mmol/L): 118 NaCl, 23 NaHCO , 3.2 KCl, 1.2 Ϸ ϩ 3 121 mm Hg/sec) after 8 minutes of infusion. The dP/dtmax KH2PO4, 2.0 CaCl2, 1.2 MgSO4, 0.5 Na2-EDTA, 11 glucose, and 2 ϩ declined thereafter to a plateau above (PϽ0.05; nϭ4) baseline. Na-pyruvate (free Ca2 ϭ1.5 mmol/L), which was bubbled with 95% ϭ Similar patterns of change (PϽ0.05;nϭ4) in peak pressure O2–5% CO2 (pH 7.40) and kept at 37°C. A balloon was inserted into the left ventricle to record pressure. Single ventricular myocytes (Ppeak) and the peak rate of left ventricle pressure decline Ϫ were isolated from mouse hearts with collagenase as previously ( dP/dtmin) were also observed with Ang II infusion. As described.17 In cell shortening studies, myocytes were field- expected from its vasoconstrictor action, Ang II infusion caused stimulated at 1 Hz in Tyrode solution (containing 1.2 mmol/L 2ϩ 27 a decrease of 46.9%Ϯ4.0% (PϽ0.01; nϭ4) in coronary artery Ca ). ICa,L were recorded from myocytes with patch-clamp technique in amphotericin B-perforated configuration. Differences flow rate at Ϸ5 minutes, which returned to baseline levels at Ϸ8 between 2 means were assessed using paired or unpaired Student t minutes (Supplemental Figure IA available online at tests. Differences among multiple means were assessed by 1-way http://hyper.ahajournals.org). analysis of variance. PϽ0.05 was considered significant. Group data are expressed as meanϮSEM. It is conceivable that the negative inotropic effects of Ang II were mediated by changes in coronary vascular resistance, Results possibly leading to metabolic changes or perfusion-related The effects of Ang II on cardiac contractility were examined in changes in contractility (ie, Gregg’s phenomenon).28 How- isolated Langendorff-perfused mouse hearts. For these studies, ever, when hearts were perfused at a constant coronary flow hearts were initially equilibrated at a constant coronary perfusion rate to achieve a perfusion pressure of Ϸ80 mm Hg at pressure of 80 mm Hg and ventricular end-diastolic pressures baseline, Ang II (3 nmol/L) caused early decrease were set at Ϸ5 mm Hg (online Supplement) to establish baseline (12.6%Ϯ2.5%) followed by a late increase (18.9%Ϯ2.3%) in 424 Hypertension September 2010

ϩ Ͻ ϭ dP/dtmax (P 0.01; n 5) over baseline (Supplemental Fig- A Ang II ure IB). Consistent with its vasoconstrictor action, Ang II also caused time-dependent increases (PϽ0.01; nϭ5) in perfusion 0 min 5 min 15 min

pressure when perfusion rate was fixed (Supplemental Figure (-) Irbersartan IB). Because vascular effects of Ang II could modulate the inotropic actions of Ang II, hearts were pretreated with 5%

P1075, a vasodilator that opens plasmalemmal KATP channels 1 Second Ϸ preferentially (by 20-fold) in vascular smooth muscle (+) Irbersartan compared to myocardium.29 As expected, pretreatment with P1075 (100 nmol/L), at fixed coronary flows, decreased Ͻ ϭ Ϯ (P 0.01; n 4) the perfusion pressure from 79.4 1.4 to B 64.2Ϯ4.5 mm Hg and eliminated the effects of Ang II on 0 min coronary perfusion pressure (Supplemental Figure IC). Con- 15 5 min 15 min sistent with previous reports showing P1075 dose- # 30,31 dependently affects cardiac function, P1075 slightly re- 10 duced contractility (ie, reduction of 9.4%Ϯ1.5%; PϽ0.01; nϭ4), probably as a result of action potential abbreviation.32 More importantly, P1075 did not influence the actions of Ang 5 *

II. Specifically, Ang II (3 nmol/L) infusion in the presence of Cell shortening (%) P1075 still induced (PϽ0.01; nϭ5) a rapid decline of 0 Ϯ ϩ 12.3% 1.7% in dP/dtmax relative to baseline, followed by (-) Irbersartan (+) Irbersartan an increase that peaked at 13.4Ϯ3.1% above (PϽ0.01) Figure 2. A, Representative cell shortening traces of ventricular baseline at Ϸ10 minutes after Ang II infusion and, thereafter myocytes without (0 minutes) and with (5 and 15 minutes) Ang II PϽ treatment (30 nmol/L) in the absence (top) and presence (bot- (at 16 minutes after infusion), remained above ( 0.05) ␮ tom) of an AT1 receptor inhibitor (irbersartan, 10 mol/L). The baseline (Figure 1B). In separate experiments using the same downward deﬂections from baselines indicate cell length ϩ conditions, we found that the dP/dtmax remained elevated decrease (shortening). Cell shortening was expressed as a per- (PϽ0.01; nϭ6) above baseline for 30 minutes after Ang II centage of the baseline diastolic length. Recordings were performed at 36°C and myocytes were stimulated at 1 Hz. B, Sum- treatment. Similar temporal changes in Ppeak (not shown) as ϭ Ϫ mary of Ang II on cell shortening (n 13–14) in the absence (left) well as dP/dtmin and the time constant (Tau) for pressure and presence (right) of irbersartan. *PϽ0.01, #PϽ0.05 vs control relaxation (Supplemental Figure II available online at http:// (0 minutes). hyper.ahajournals.org) were also observed. In all remaining ␮ studies, hearts were pretreated with P1075 before Ang II (AT1R) blocker irbersartan (10 mol/L), supporting the infusion to prevent the effects of Ang II on coronary artery conclusion that AT1Rs in cardiomyocytes mediate the con- constriction. tractile effects of Ang II (Figure 2). To further characterize the inotropic actions of Ang II, we Previous studies have reported that Ang II can increase 33,34 compared the responses of hearts to different concentrations cardiac ICa,L, which are key elements in cardiac excita- (3, 30, and 300 nmol/L) of Ang II. As shown in Figure 1B and tion–contraction coupling and contraction. To examine if

Supplemental Table I (available online at http://hyper. ICa,L is affected by Ang II in mouse myocardium, we recorded ϩ ahajournals.org), the magnitude of changes in dP/dtmax ICa,L using whole-cell patch-clamp technique with the am- showed little dose-dependence, and a near-maximal response photericin B-perforated configuration, which appears to be 7 33,34 was obtained at 30 nmol/L, consistent with previous studies. necessary to identify the effects of Ang II on ICa,L. As ϩ It is notable that dP/dtmax showed a small, but highly shown in Figure 3, Ang II perfusion of isolated ventricular reproducible, increase before the rapid decline phase, myocytes caused (PϽ0.05; nϭ7) rapid transient Ϸ10% ϩ Ϸ whereas the late-phase increases in dP/dtmax occurred decreases in ICa,L, followed by 13% increases over baseline earlier (PϽ0.05; nϭ5) with higher Ang II levels. These values, suggesting that the negative and positive inotropic findings suggest that the positive and negative inotropic effects of Ang II are, at least partially, mediated by alterations

actions of Ang II involve separate processes with different in ICa,L. kinetics. Similar temporal changes were also observed in We next attempted to determine the signaling pathways Ϫ dP/dtmin during Ang II infusion, as summarized in Supple- involved in the biphasic effects of Ang II on myocardial mental Table I. contractility. It has been shown that Ang II activates Although the myocardial effects of Ang II could be G-protein-coupled receptors, which can signal through 15,35 36 mediated by Ang II receptors in nonmyocyte cells of the PI3K. Because PI3K␥ is G␤␥-dependent and regulates heart, Ang II also produced biphasic alterations in cell cardiac contractility,19,37 we initially examined the effects of shortening amplitudes of isolated myocytes (Figure 2). Spe- Ang II on mouse hearts lacking PI3K␥ (PI3K␥Ϫ/Ϫ). As cifically, the percent cell shortening was decreased (PϽ0.01; expected,37,38 Supplemental Figure IIIA (available online at ϭ Ϸ ϩ n 13) by 50% at 5 minutes after Ang II treatment, http://hyper.ahajournals.org) shows that baseline dP/dtmax followed by Ϸ40% increase (PϽ0.05) above control values at of PI3K␥Ϫ/Ϫ hearts was Ϸ30% higher (Pϭ0.01; nϭ5) than 15 minutes. Ang II had no effect (Pϭ0.98; nϭ14) on percent that of wild-type littermate hearts. However, similar to cell shortening in the presence of the type 1 Ang II receptor wild-type hearts, Ang II (30 nmol/L) infusion into PI3K␥Ϫ/Ϫ Liang et al Ang II Has Complex Inotropic Effects 425

A B 0 Control 140 Baseline Ang II Peak decrease Peak increase Ang II 120 #

-500 b currents 2+ 100 b 500 pA * a a currents (pA) 80 2+ c 40 mS -1000 (% of baseline) Ca c 60 Normalized Ca 0 024681012 Time (min) Figure 3. A, Time course (left) and representative traces (right) of L-type Ca2ϩ currents recorded at 0 mV from ventricular myocytes with Ang II (30 nmol/L) perfusion (white circles) or without Ang II (black circles). Capacitance currents in right panel were removed for clarity. Currents were recorded at 22°C with patch-clamp technique using amphotericin B-perforated conﬁguration. B, Summary of Ang II-induced changes in amplitude of L-type Ca2ϩ currents recorded at 0 mV (nϭ7). *PϽ0.05, #PϽ0.01 vs baseline currents. hearts induced (PϽ0.01; nϭ5) an early-phase reduction compared to wild-type hearts (Figure 4A), consistent with a Ϯ ϩ 18 (13.1% 3.4%) in dP/dtmax, followed by a late-phase in- recent study. More important, Ang II infusion resulted in a Ϯ Ϯ ϩ crease (27.3% 6.5%) over baseline levels (Supplemental modest and insignificant 2.3% 0.7% decline in dP/dtmax in Ϫ ␣ Ͻ Figure III), with similar changes in dP/dtmin (not shown). DN-PI3K hearts (Figure 4), which was far less (P 0.01; The relative changes in dP/dt induced by Ang II in PI3K␥Ϫ/Ϫ nϭ7) than the 19.6%Ϯ1.6% decline seen in wild-type litter- Ͼ ϭ ϩ hearts were not different (P 0.15; n 5) from those observed mate hearts. However, the late-phase increase in dP/dtmax in wild-type littermate hearts (Supplemental Figure III). was unaffected (Pϭ0.80; nϭ7) by PI3K␣ inhibition (Figure Thus, PI3K␥ does not mediate the actions of Ang II on mouse 4). Ang II infusion of DN-PI3K␣ hearts also caused much Ͻ ϭ Ϫ Ϯ myocardial contractility. smaller (P 0.01; n 7) decreases in dP/dtmin (5.7% 1.1%) Ϯ Because AT1Rs are coupled to Gq protein, which can compared to wild-type littermates (23.8% 1.6% decrease) regulate PI3K␣,39 we tested the actions of Ang II in hearts without causing notable differences (Pϭ0.40; nϭ7) in the ␣ ␣ Ϫ with dominant-negative inhibition of PI3K (DN-PI3K ). late-phase increases of dP/dtmin (data not shown). These ␣ ϩ ␣ DN-PI3K hearts had slightly lower baseline dP/dtmax data demonstrate that PI3K plays an important role in the

A 5000 Wild- type DN-PI3Kα 5000

Ang II Ang II

4000 4000 (mmHg/s) max

3000 3000 +dP/dt

0 0 -5 0 5 10 15 -5 0 5 10 15 Time (min) Time (min) B Wild -type 40 α

max DN-PI3K

30 max 30

20 20

(% of baseline) 10 (% of baseline) 10 * Peak decrease in +dP/dt Peak increase in +dP/dt 0 0 ϩ ϭ ␣ Figure 4. A, Time courses of dP/dtmax during Ang II (30 nmol/L) infusion in wild-type hearts (left; n 5) and DN-PI3K hearts (right; nϭ7). Mouse hearts were perfused at a constant ﬂow rate in the presence of a vasodilator (P1075, 100 nmol/L). B, Summary of peak ϩ ␣ decreases (left) and peak increases (right) of dP/dtmax during Ang II infusion in wild-type hearts (white bars) and DN-PI3K hearts (black bars). *PϽ0.01 vs wild-type. 426 Hypertension September 2010

A Ang II 5000 5000 Ang II GF 109203X

4000 4000 (mmHg/s) max ϩ Figure 5. A, Time courses of dP/dtmax of wild- 3000 3000 type C57BL/6 mouse hearts during Ang II (30 +dP/dt nmol/L) infusion in the absence (left; nϭ10) and 0 0 ϭ -5 0 5 10 15 -5 0 5 10 15 presence (right; n 7) of a PKC inhibitor (GF 109203X, 1 ␮mol/L). Hearts were perfused at a Time (min) Time (min) constant ﬂow rate in the presence of a vasodila- B tor (P1075, 100 nmol/L). B, Summary of peak Control decreases (left) and peak increases (right) of max max 30 ϩ 15 GF 109203X dP/dtmax during Ang II infusion in the absence (white bars) or presence (black bars) of GF 20 109203X. *PϽ0.05 vs control. 10 * * 5 10 (% of baseline) (% of baseline)

0 0 Peak increase in +dP/dt Peak decrease in +dP/dt early-phase negative inotropic and lusitropic effects of Ang II littermate hearts. These data establish that NOX2 is not required in mouse hearts. for the inotropic effects of Ang II in mouse hearts. PKC activity has been recently linked to PI3K␣.40 In addition, PKC has been shown to be involved in either the Discussion negative9,12 or the positive8 inotropic effects of Ang II. Despite Ang II being an important regulator of the cardio- Despite these observations, Ang II (30 nmol/L) infusion still vascular system, its direct effects on cardiac contractility are Ͻ ϭ ϩ 5–8,48 9,12,49 10,11 caused biphasic changes (P 0.01; n 7) in dP/dtmax in the unclear, with positive, negative, and no effects ␮ ϭ presence of GF 109203X (1 mol/L; IC50 32 nmol/L for being reported. Our study shows that Ang II produces PKC inhibition), with a rapid transient 5.7%Ϯ1.1% decline negative and positive effects on contractility in mouse hearts. followed by a 12.0%Ϯ3.0% increase above baseline (Figure Although the transient negative inotropic actions of Ang II in 5). However, PKC inhibition reduced (PϽ0.05; nϭ7) the our studies at constant perfusion pressure could arise from early-phase decreases and the late-phase increases in ϩdP/ compromised myocardial metabolism resulting from reduc- Ϸ dtmax by 50% compared to changes caused by Ang II in the tions in coronary flow, secondary to vasoconstriction, de- absence of PKC inhibition (Figure 5). PKC inhibition also creased contractility also was observed when coronary flow reduced (PϽ0.01; nϭ7) the early-phase decrease in ϪdP/ was kept constant in the absence or presence of P1075 to Ϸ ϭ ϭ dtmin by 50% without affecting (P 0.91; n 7) the late- dilate the coronary arteries. These findings establish that the Ϫ phase increase in dP/dtmin induced by Ang II (not shown). contractile effects of Ang II on the myocardium are not These results suggest that PKC plays a role in the negative mediated indirectly by actions of Ang II on coronary blood and positive inotropic effects of Ang II. vessels. Consistent with this conclusion, Ang II treatment of Both PKC41 and PI3K42 (particularly PI3K␣)43 have been cardiomyocytes induced a transient impairment of cell short- shown to regulate GSK3␤, which plays an important role in ening amplitude that was followed by a brisk increase in heart pump function.44 However, Ang II infusion into hearts single cell contractility, mimicking the pattern seen in Lan- with conditional knockout of GSK3␤ induced biphasic gendorff hearts. These inotropic effects of Ang II were Ͻ ϭ ϩ changes (P 0.05; n 5) in dP/dtmax (Supplemental Figure eliminated by AT1R-specific blockade in single cardiomyo- IVA available online at http://hyper.ahajournals.org) that cytes but were unaffected by pretreatment with ␤-adrenergic were not different (PϾ0.19, nϭ5) from changes induced by receptor blockers (propranolol) in isolated hearts (data not Ang II in wild-type littermate hearts. These data show that shown), establishing that enhanced local norepinephrine re- GSK3␤ does not play a role in the inotropic effects of Ang II lease from autonomous nerve terminals50 by Ang II or ␤ in mouse hearts. receptor cross-talk between AT1R and -adrenergic recep- PKC can be activated by reactive oxygen species.45 The tor49 does not contribute to the actions of Ang II. p47phox-dependent NADPH oxidase (NOX2) is a key enzyme Taken together, our studies show that Ang II produces for cardiac generation of reactive oxygen species46 and can be complex contractile responses by direct actions on myocar- 47 activated by PI3K. Therefore, we tested whether NOX2 dial AT1R. This conclusion is consistent with the observation activity is required for the inotropic effects of Ang II by using that Ang II altered ICa,L in mouse ventricular myocytes, with mice lacking p47phox (p47phoxϪ/Ϫ). Ang II infusion into a time course that mirrored precisely the changes in heart p47phoxϪ/Ϫ hearts induced biphasic changes (PϽ0.05; nϭ5) in contractility. The potential mechanistic link between cardiac ϩ dP/dtmax (Supplemental Figure IVB) that were not different contractility and ICa,L changes is further supported by our (PϾ0.34; nϭ5) from changes induced by Ang II in wild-type results in DN-PI3K␣ hearts as discussed below. However, it Liang et al Ang II Has Complex Inotropic Effects 427 is conceivable that Ang II also affects other end-effectors PKC. PKC activation by Ang II could also activate Naϩ/Hϩ ϩ involved in Ca2 regulation. Support for this possibility is exchanger,56 leading to elevations in contractility via elevated Ϫ ϩ provided by the indices of ventricular relaxation ( dP/dtmin Na loading or increases in intracellular pH. Clearly, addi- and the time constant for pressure relaxation, Tau; Supple- tional studies are required to further determine the cellular mental Figure IIB), which also showed biphasic alterations mechanism whereby Ang II alters cardiac contractility. after Ang II treatment, consistent with possible changes in sarcoplasmic reticulum Ca2ϩ ATPase activity or myofila- Perspectives ment Ca2ϩ sensitivity. However, the interpretation of these In contrast to previous studies, our studies establish that Ang studies is complicated by complex interdependence of the II activation of AT1R leads to complex temporal decreases 2ϩ many processes involved in Ca cycling and contractile and increases in ICa,L and contractility in “healthy” mouse protein activation/relaxation. Clearly, additional studies will myocardium, with the decreases requiring PI3K␣ and PKC. be required to discern the potential role of other functional The relevance of our studies to larger mammals, including proteins and processes in transducing the actions of Ang II. humans wherein elevations in Ang II are linked to decreased Regardless of the functional mechanisms involved in contractility in advanced heart disease,2 is unclear. Neverthe- mediating contractility changes induced by Ang II, we found less, it is conceivable that the PI3K␣-dependent and PKC- that the negative inotropic actions of Ang II were largely dependent reductions in ICa,L and contractility, induced by abolished in DN-PI3K␣ hearts, but not in PI3K␥Ϫ/Ϫ hearts. Ang II, may contribute to the impaired function in heart The involvement of PI3K␣, but not PI3K␥ (which is disease. Moreover, these and other potential mechanisms for 36 G␤␥-dependent and a negative regulator of cardiac con- the effects of Ang II on heart function in mice might also be tractility19,37), is somewhat surprising because Ang II signals relevant to humans, even though the relative contribution of through G-protein-coupled receptors. These findings, to- various downstream factors and signaling pathways may be gether with observations that genetic inhibition of GSK3␤ or modified in large mammals. Further studies will be required NADPH oxidase did not affect the inotropic effects of acute to assess whether targeting pathways mediating the negative Ang II stimulation, highlight the divergent mechanisms inotropic actions of Ang II is a valid strategy for treating required for the acute vs long-term regulations of contractility patients with heart disease. by Ang II.4,19,37,44 Regardless, it is not clear whether PI3K␣ activity is increased or decreased by elevated Ang II levels. It Acknowledgments was previously concluded that G␣q protein, which is linked The authors thank Drs Steffen-Sebastian Bolz and Rudolf Schubert for assistance. The authors dedicate this study in the memory of Dr to AT R, inhibits PI3K␣.39,51 Thus, because PI3K␣ positively 1 Michael E. Ward (deceased November 2009). 17 18 regulates cardiac ICa,L and contractility, our results are consistent with the conclusion that Ang II mediates its rapid Source of Funding negative inotropic actions on hearts by inhibiting PI3K␣, This study was funded by a research grant from Canadian Institutes leading to decreases in cardiac ICa,L. This is supported by the of Health Research (153103 to P.H.B.). P.H.B. holds a Career Investigator Award from the Heart and Stroke Foundation (HSF) of rapid transient reduction in ICa,L after Ang II perfusion in our studies. Involvement of I in the PI3K␣-dependent nega- Ontario. W.L. received a Doctoral Research Award from HSF of Ca,L Canada. tive contractility effects of Ang II is also consistent with our observation that pretreatment of heart with a PKC inhibitor Disclosures abolished a large fraction of the reductions in cardiac con- None. tractility induced by Ang II because PKC activation can 52 inhibit cardiac ICa,L. References In addition to interfering with Ang II-mediated impairment 1. Sadoshima J, Xu Y, Slayter HS, Izumo S. Autocrine release of angioten- of contractility, PKC inhibition also blocked Ϸ50% of the sin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro. Cell. 1993;75:977–984. positive inotropic actions of Ang II. This dual action of PKC 2. Serneri GGN, Boddi M, Cecioni I, Vanni S, Coppo M, Papa ML, is consistent with previous studies showing that PKC activa- Bandinelli B, Bertolozzi I, Polidori G, Toscano T, Maccherini M, Modesti tion can have positive53 or negative54 inotropic effects on PA. Cardiac angiotensin II formation in the clinical course of heart failure myocardium. Moreover, because loss of the ␣ isoform of and its relationship with left ventricular function. Circ Res. 2001;88: ␣ 961–968. PKC (PKC ) in myocardium enhances contractility and 3. Sadoshima J, Izumo S. Molecular characterization of angiotensin protects against heart failure,55 our data support the possibil- II-induced hypertrophy of cardiac myocytes and hyperplasia of cardiac ity that the rapid negative inotropic actions of Ang II involve fibroblasts. Critical role of the AT1 receptor subtype. Circ Res. 1993;73: PI3K␣-dependent PKC␣ activation, leading to I reduc- 413–423. Ca,L 4. Gusev K, Domenighetti AA, Delbridge LMD, Pedrazzini T, Niggli E, tions. On the other hand, global PKC inhibition reduces the Egger M. Angiotensin II-mediated adaptive and maladaptive remodeling positive inotropic actions of Ang II, suggesting that Ang II of cardiomyocyte excitation-contraction coupling. Circ Res. 2009;105: may enhance contraction by activating other PKC isoforms 42–50. ␣ 5. Moravec C, Schluchter M, Paranandi L, Czerska B, Stewart R, (besides PKC ) leading to increased ICa,L as suggested Rosenkranz E, Bond M. Inotropic effects of angiotensin II on human previously.34,48 This suggestion is consistent with studies cardiac muscle in vitro. Circulation. 1990;82:1973–1984. 6. Petroff MGV, Aiello EA, Palomeque J, Salas MA, Mattiazzi A. Sub- showing that the positive effects of Ang II on cardiac ICa,L are only observed in perforated-patch voltage-clamp record- cellular mechanisms of the positive inotropic effect of angiotensin II in cat myocardium. J Physiol. 2000;529:189–203. 33,34 2ϩ ings when intracellular Ca levels are not buffered, 7. Cingolani HE, Villa-Abrille MC, Cornelli M, Nolly A, Ennis IL, Gar- ϩ thereby allowing activation of Ca2 -dependent isoforms of ciarena C, Suburo AM, Torbidoni V, Correa MV, Camilionde Hurtado 428 Hypertension September 2010

MC, Aiello EA. The positive inotropic effect of angiotensin II: role of tractility via a direct inhibition of Ik,slow currents. Cardiovasc Res. 2004; endothelin-1 and reactive oxygen species. Hypertension. 2006;47: 62:509–520. 727–734. 28. Gregg DE. Effect of coronary perfusion pressure or coronary flow on 8. Rajagopal K, Whalen EJ, Violin JD, Stiber JA, Rosenberg PB, Premont oxygen usage of the myocardium. Circ Res. 1963;13:497–500. RT, Coffman TM, Rockman HA, Lefkowitz RJ. ␤-Arrestin2-mediated 29. Sato T, Sasaki N, Seharaseyon J, O’Rourke B, Marban E. Selective inotropic effects of the angiotensin II type 1A receptor in isolated cardiac pharmacological agents implicate mitochondrial but not sarcolemmal

myocytes. Proc Natl Acad Sci U S A. 2006;103:16284–16289. KATP channels in ischemic cardioprotection. Circulation. 2000;101: 9. Palomeque J, Sapia L, Hajjar RJ, Mattiazzi A, Vila Petroff M. Angio- 2418–2423. tensin II-induced negative inotropy in rat ventricular myocytes: role of 30. Sargent CA, Sleph PG, Dzwonczyk S, Normandin D, Antonaccio MJ, reactive oxygen species and p38 MAPK. Am J Physiol Heart Circ Grover GJ. Cardioprotective effects of the cyanoguanidine potassium Physiol. 2006;290:H96–H106. channel opener P-1075. J Cardiovasc Pharmacol. 1993;22:564–570. 10. Lefroy DC, Crake T, Del Monte F, Vescovo G, Dalla Libera L, Harding 31. Jilkina O, Kuzio B, Grover GJ, Kupriyanov VV. Effects of KATP channel S, Poole-Wilson PA. Angiotensin II and contraction of isolated myocytes openers, P-1075, pinacidil, and diazoxide, on energetics and contractile from human, guinea pig, and infarcted rat hearts. Am J Physiol Heart Circ function in isolated rat hearts. J Mol Cell Cardiol. 2002;34:427–440. Physiol. 1996;270:H2060–H2069. 32. Xu X, Tsai TD, Lee KS. A specific activator of the ATP-inhibited Kϩ 11. Mercure C, Yogi A, Callera GE, Aranha AB, Bader M, Ferreira AJ, channels in guinea pig ventricular cells. J Pharmacol Exp Ther. 1993; Santos RAS, Walther T, Touyz RM, Reudelhuber TL. Angiotensin(1-7) 266:978–984. blunts hypertensive cardiac remodeling by a direct effect on the heart. 33. Ichiyanagi O, Ishii K, Endoh M. Angiotensin II increases L-type Ca2ϩ Circ Res. 2008;103:1319–1326. current in gramicidin D-perforated adult rabbit ventricular myocytes: 12. Sakurai K, Norota I, Tanaka H, Kubota I, Tomoike H, Endo M. Negative comparison with conventional patch-clamp method. Pflugers Arch. 2002; inotropic effects of angiotensin II, endothelin-1 and phenylephrine in 444:107–116. indo-1 loaded adult mouse ventricular myocytes. Life Sci. 2002;70: 34. Aiello EA, Cingolani HE. Angiotensin II stimulates cardiac L-type Ca2ϩ 1173–1184. current by a Ca2ϩ- and protein kinase C-dependent mechanism. Am J 13. Wenzel S, Abdallah Y, Helmig S, Schafer C, Piper HM, Schluter K-D. Physiol Heart Circ Physiol. 2001;280:H1528–H1536. ␣ Contribution of PI3-kinase isoforms to angiotensin II- and -adrenoceptor- 35. Oudit GY, Kassiri Z, Patel MP, Chappell M, Butany J, Backx PH, mediated signalling pathways in cardiomyocytes. Cardiovasc Res. 2006;71: Tsushima RG, Scholey JW, Khokha R, Penninger JM. Angiotensin 352–362. II-mediated oxidative stress and inflammation mediate the age-dependent 14. Rabkin SW, Goutsouliak V, Kong JY. Angiotensin II induces activation cardiomyopathy in ACE2 null mice. Cardiovasc Res. 2007;75:29–39. of phosphatidylinositol 3-kinase in cardiomyocytes. J Hypertens. 1997; 36. Brock C, Schaefer M, Reusch HP, Czupalla C, Michalke M, Spicher K, 15:891–899. Schultz G, Nurnberg B. Roles of G␤␥ in membrane recruitment and 15. Quignard JF, Mironneau J, Carricaburu V, Fournier B, Babich A, activation of p110␥/p101 phosphoinositide 3-kinase ␥. J Cell Biol. 2003; Nurnberg B, Mironneau C, Macrez N. Phosphoinositide 3-kinase ␥ 160:89–99. mediates angiotensin II-induced stimulation of L-type calcium channels 37. Crackower M, Oudit G, Kozieradzki I, Sarao R, Sun H, Sasaki T, Hirsch in vascular myocytes. J Biol Chem. 2001;276:32545–32551. E, Suzuki A, Shioi T, Irie-Sasaki J, Sah R, Cheng H, Rybin V, Lembo G, 16. Yano N, Tseng A, Zhao TC, Robbins J, Padbury JF, Tseng YT. Tem- Fratta L, Oliveira-dos-Santos A, Benovic J, Kahn C, Izumo S, Steinberg porally controlled overexpression of cardiac-specific PI3K␣ induces S, Wymann M, Backx P, Penninger J. Regulation of myocardial contrac- enhanced myocardial contractility—a new transgenic model. Am J tility and cell size by distinct PI3K-PTEN signaling pathways. Cell. Physiol Heart Circ Physiol. 2008;295:H1690–H1694. 2002;110:737–749. 17. Sun H, Kerfant BG, Zhao D, Trivieri MG, Oudit GY, Penninger JM, 38. Kerfant BG, Gidrewicz D, Sun H, Oudit GY, Penninger JM, Backx PH. Backx PH. Insulin-like growth factor-1 and PTEN deletion enhance ϩ Cardiac sarcoplasmic reticulum calcium release and load are enhanced by cardiac L-type Ca2 currents via increased PI3K␣/PKB signaling. Circ subcellular cAMP elevations in PI3K␥-deficient mice. Circ Res. 2005; Res. 2006;98:1390–1397. 96:1079–1086. 18. Lu Z, Jiang YP, Wang W, Xu XH, Mathias RT, Entcheva E, Ballou LM, 39. Ballou LM, Lin HY, Fan G, Jiang YP, Lin RZ. Activated G␣q inhibits Cohen IS, Lin RZ. Loss of cardiac phosphoinositide 3-kinase p110␣ p110␣ J Biol Chem results in contractile dysfunction. Circulation. 2009;120:318–325. phosphatidylinositol 3-kinase and Akt. . 2003;278: 19. Kerfant B, Rose R, Sun H, Backx P. Phosphoinositide 3-kinase ␥ reg- 23472–23479. ulates cardiac contractility by locally controlling cyclic adenosine mono- 40. Rigor DL, Bodyak N, Bae S, Choi JH, Zhang L, Ter-Ovanesyan D, He Z, phosphate levels. Trends Cardiovasc Med. 2006;16:250–256. McMullen JR, Shioi T, Izumo S, King GL, Kang PM. Phosphoinositide ␤ 20. Kerfant BG, Zhao D, Lorenzen-Schmidt I, Wilson LS, Cai S, Chen SRW, 3-kinase Akt signaling pathway interacts with protein kinase C 2inthe Maurice DH, Backx PH. PI3K␥ is required for PDE4, not PDE3, activity regulation of physiologic developmental hypertrophy and heart function. in subcellular microdomains containing the sarcoplasmic reticular Am J Physiol Heart Circ Physiol. 2009;296:H566–H572. calcium ATPase in cardiomyocytes. Circ Res. 2007;101:400–408. 41. Fang X, Yu S, Tanyi JL, Lu Y, Woodgett JR, Mills GB. Convergence of 21. Nienaber JJ, Tachibana H, Naga Prasad SV, Esposito G, Wu D, Mao L, multiple signaling cascades at glycogen synthase kinase-3: Edg receptor- Rockman HA. Inhibition of receptor-localized PI3K preserves cardiac mediated phosphorylation and inactivation by lysophosphatidic acid ␤-adrenergic receptor function and ameliorates pressure overload heart through a protein kinase C-dependent intracellular pathway. Mol Cell failure. J Clin Invest. 2003;112:1067–1079. Biol. 2002;22:2099–2110. 22. Shioi T, Kang P, Douglas P, Hampe J, Yballe C, Lawitts J, Cantley L, 42. Tong H, Imahashi K, Steenbergen C, Murphy E. Phosphorylation of ␤ Izumo S. The conserved phosphoinositide 3-kinase pathway determines glycogen synthase kinase-3 during preconditioning through a heart size in mice. EMBO J. 2000;19:2537–2548. phosphatidylinositol-3-kinase-dependent pathway is cardioprotective. 23. Sasaki T, Irie-Sasaki J, Horie Y, Bachmaier K, Fata J, Li M, Suzuki A, Circ Res. 2002;90:377–379. Bouchard D, Ho A, Redston M, Gallinger S, Khokha R, Mak T, Hawkins 43. Braz JC, Gill RM, Corbly AK, Jones BD, Jin N, Vlahos CJ, Wu Q, Shen P, Stephens L, Scherer S, Tsao M, Penninger J. Colorectal carcinomas in W. Selective activation of PI3K␣/Akt/GSK-3␤ signalling and cardiac mice lacking the catalytic subunit of PI(3)K␥. Nature. 2000;406:897–902. compensatory hypertrophy during recovery from heart failure. Eur 24. Tkalcevic J, Novelli M, Phylactides M, Iredale JP, Segal AW, Roes J. J Heart Fail. 2009;11:739–748. Impaired immunity and enhanced resistance to endotoxin in the absence 44. Michael A, Haq S, Chen X, Hsich E, Cui L, Walters B, Shao Z, Bhat- of neutrophil elastase and cathepsin G. Immunity. 2000;12:201–210. tacharya K, Kilter H, Huggins G, Andreucci M, Periasamy M, Solomon 25. Patel S, Doble BW, MacAulay K, Sinclair EM, Drucker DJ, Woodgett RN, Liao R, Patten R, Molkentin JD, Force T. Glycogen synthase JR. Tissue-specific role of glycogen synthase kinase 3␤ in glucose kinase-3␤ regulates growth, calcium homeostasis, and diastolic function homeostasis and insulin action. Mol Cell Biol. 2008;28:6314–6328. in the heart. J Biol Chem. 2004;279:21383–21393. ϩ 26. Sohal DS, Nghiem M, Crackower MA, Witt SA, Kimball TR, Tymitz 45. Gopalakrishna R, Anderson WB. Ca2 - and phospholipid-independent KM, Penninger JM, Molkentin JD. Temporally regulated and tissue- activation of protein kinase C by selective oxidative modification of the specific gene manipulations in the adult and embryonic heart using a regulatory domain. Proc Natl Acad Sci U S A. 1989;86:6758–6762. tamoxifen-inducible Cre protein. Circ Res. 2001;89:20–25. 46. Bendall JK, Cave AC, Heymes C, Gall N, Shah AM. Pivotal role of a 27. Sun H, Oudit GY, Ramirez RJ, Costantini D, Backx PH. The phospho- gp91phox-containing NADPH oxidase in angiotensin II-induced cardiac inositide 3-kinase inhibitor LY294002 enhances cardiac myocyte con- hypertrophy in mice. Circulation. 2002;105:293–296. Liang et al Ang II Has Complex Inotropic Effects 429

47. Yamamori T, Inanami O, Nagahata H, Kuwabara M. Phosphoinositide 52. McHugh D, Sharp EM, Scheuer T, Catterall WA. Inhibition of cardiac 3-kinase regulates the phosphorylation of NADPH oxidase component L-type calcium channels by protein kinase C phosphorylation of two sites p47phox by controlling cPKC/PKC ␦ but not Akt. Biochem Biophys Res in the N-terminal domain. Proc Natl Acad Sci U S A. 2000;97:12334- Commun. 2004;316:720–730. 12338. 48. Salas M, Vila-Petroff M, Palomeque J, Aiello E, Mattiazzi A. Positive 53. Pi Y, Sreekumar R, Huang X, Walker JW. Positive inotropy mediated by inotropic and negative lusitropic effect of angiotensin II: intracellular diacylglycerol in rat ventricular myocytes. Circ Res. 1997;81:92–100. mechanisms and second messengers. J Mol Cell Cardiol. 2001;33: 54. Yuan SH, Sunahara FA, Sen AK. Tumor-promoting phorbol esters inhibit 1957–1971. cardiac functions and induce redistribution of protein kinase C in perfused 49. Meissner A, Min J, Simon R. Effects of angiotensin II on inotropy and beating rat heart. Circ Res. 1987;61:372–378. intracellular Ca2ϩ handling in normal and hypertrophied rat myocardium. 55. Liu Q, Chen X, MacDonnell SM, Kranias EG, Lorenz JN, Leitges M, J Mol Cell Cardiol. 1998;30:2507–2518. Houser SR, Molkentin JD. Protein kinase C␣, but not PKC␤ or PKC␥, 50. Schumann H, Starke K, Werner U. Interactions of inhibitors of noradren- regulates contractility and heart failure susceptibility: implications for aline uptake and angiotensin on the sympathetic nerves of the isolated ruboxistaurin as a novel therapeutic approach. Circ Res. 2009;105:194–200. rabbit heart. Br J Pharmacol. 1970;39:390–397. 56. Gunasegaram S, Haworth RS, Hearse DJ, Avkiran M. Regulation of 51. Lu Z, Jiang YP, Ballou LM, Cohen IS, Lin RZ. G␣q inhibits cardiac sarcolemmal Naϩ/Hϩ exchanger activity by angiotensin II in adult rat 2ϩ L-type Ca channels through phosphatidylinositol 3-kinase. J Biol ventricular myocytes: opposing actions via AT1 versus AT2 receptors. Chem. 2005;280:40347–40354. Circ Res. 1999;85:919–930.