Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mitochondria and microdomains in vascular smooth muscle cell Ca²⁺ signalling Poburko, Damon Todd 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2005-105550.pdf [ 19.82MB ]
JSON: 831-1.0099856.json
JSON-LD: 831-1.0099856-ld.json
RDF/XML (Pretty): 831-1.0099856-rdf.xml
RDF/JSON: 831-1.0099856-rdf.json
Turtle: 831-1.0099856-turtle.txt
N-Triples: 831-1.0099856-rdf-ntriples.txt
Original Record: 831-1.0099856-source.json
Full Text

Full Text

MITOCHONDRIA AND MICRODOMAINS IN VASCULAR SMOOTH MUSCLE CELL Ca 2 + SIGNALING By DAMON T O D D POBURKO B.Sc, The University of British Columbia, 2001 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Pharmacology and Therapeutics) THE UNIVERSITY OF BRITISH COLUMBIA November 2005 © Damon Todd Poburko, 2005 A B S T R A C T Contraction of vascular smooth muscle (VSM) is regulated by fluctuations in the intracellular concentration of free ionic calcium ([Ca ]j). The spatio-temporal regulation of [Ca 2 +]i relies on the sub-cellular architecture of the smooth muscle cell and the juxtaposition of opposing plasmalemma (PM), sarcoplasmic reticulum (SR) and mitochondria. This thesis addresses two related aspects of Ca2+-signaling in V S M : 1) basal Ca2+-entry across the P M and 2) mitochondrial Ca2+-uptake during agonist mediated stimulation in cultured rat aorta smooth muscle cells. Basal Ca2+-entry into resting cells, measured with radio-labeled 4 5 C a 2 + , was blocked (-80%) by organic inhibitors of L-type voltage-gated Ca2+-channels (nifedipine), store-/receptor-operated Ca2+-channels (SKF-96365) and inositol-1,4,5-trisphosphate receptors (IP3R) (2-APB). At increasing concentrations, gadolium (Gd 3 +) biphasically inhibited Ca2+-uptake. The maximal effect of the first phase (lOOuM Gd 3 + ) was equally as effective as combined treatment with 2-APB and SKF-96365. At 0.2-10mM, G d 3 + inhibited C a 2 + influx to a greater extent than the organic inhibitors. We concluded that basal C a 2 + entry primarily occurred via basal activity of excitable channels, and PCR analysis suggests this influx to involve L-type voltage-gated Ca 2 + -channels, and TRPC1, TRPC4 and TRPC6. Next, we used Ca2+-sensitive proteins (aequorins and pericams) and dyes (fura-2) to measure parallel [Ca ] changes in the mitochondrial matrix, subplasmalemmal cytosol and bulk cytosol. Replacing extracellular N a + with n-methyl-d-glucamine (NMDG) caused Ca2+-entry by reversal of the Na+/Ca2+-exchanger (NCX), which was selectively blocked by KB-R7943. N C X -reversal increased mitochondrial and subplasmalemmal but not bulk cytosolic [Ca 2 +] revealing a local interaction of the SR, N C X and mitochondria. Furthermore, NCX-reversal and mitochondrial Ca2+-uptake appear to occur during the [Ca2+]j plateau phase of the response to purinergic stimulation (ATP). However, this mitochondrial Ca2+-uptake does not increase [Ca 2 + _MT because of a compensatory stimulation of mitochondrial Ca2+-extrusion (blocked by CGP-37157). Finally, we dissected the [Ca 2 + _MT response to SR Ca2+-release in response to agonist-mediated stimulation. ATP and [ARG ]-vasopressin transiently increased [Ca ]MT by activating ii both IP3R and ryanodine receptors (RyR) (selectively inhibited by 2-APB and procaine). Image analysis of fluorescently labeled mitochondria, IP3R and RyR corroborated functional evidence that IP3R and RyR release C a 2 + from separate sub-compartments of the SR and that physiological [Ca 2 + ]MT elevations rely on H^R-RyR cross-talk. i i i TABLE OF CONTENTS: page A B S T R A C T » i i T A B L E OF CONTENTS iv LIST OF T A B L E S .viii LIST OF FIGURES ix LIST OF S Y M B O L S & ABBREVIATIONS xi LIST OF PUBLICATIONS & ABSTRACTS xii A C K N O L W L E D G E M E N T S xiv CHAPTER I 1 INTRODUCTION TO CALCIUM SIGNALING IN VASCULAR SMOOTH MUSCLE 1.1 THE BASICS OF SMOOTH M U S C L E : 1 1.1.1 Physiological role of smooth muscle 1 1.1.2 The mechanism of smooth muscle contraction 2 1.1.3 Mechanisms of C a 2 + homeostasis and signaling in V S M 2 1.2 LINKED C a 2 + TRANSPORT IN SMOOTH M U S C L E 5 1.2.1 Structural considerations of PM-SR junctions 6 1.2.2 Function of PM-SR junctional complexes 7 Dynamic Ca 2 + buffering and cycling in resting smooth muscle 7 Wave-like [Ca2+]j oscillations in activated smooth muscle 9 Ca 2 + spark/wave-STOC coupling in relaxing smooth muscle 10 CICR in activating smooth muscle 11 1.2.3 Structural considerations of mitochondrial junctions 11 1.2.4 Function of mitochondrial junctions 13 1.3 MITOCHONDRIA A T THE CROSSROADS OF C a 2 + SIGNALING 14 1.3.1 Pertinent mitochondrial ion transporters 16 Ca2+uptake mechanisms 16 Ca 2 + extrusion & release mechanisms 17 Outer mitochondrial membrane 18 1.3.2 Interpreting mitochondrial inhibition 18 Protonophores & Oligomycin 19 ETC inhibition alters ROS production, redox state and A\\im 20 1.3.3 Mitochondria buffer C a 2 + influx & release 20 Uniporter vs. RaM activation 21 Mitochondrial Ca2+-cycling at rest & [Ca2+]i set-point 22 Kinetics of mitochondrial versus cytosolic Ca 2 + elevations 23 1.3.4 Mitochondria modulate SRCa 2 +release channel gating 24 1.3.5 Mitochondrial modulation of P M channels 26 1.3.6 Mitochondria facilitate SR refilling 27 1.3.7 Mitochondria influence C a 2 + oscillations 28 1.3.8 Mitochondrial ROS production , O2 detection 29 1.3.9 Summary of mitochondria & C a 2 + signaling 30 1.4 PURPOSE A N D SPECIFIC AIMS 31 CHAPTER II 33 BASAL C a 2 + ENTRY IN VASCULAR SMOOTH MUSCLE CELLS 2.1 INTRODUCTION 33 2.2 METHODS 34 2.3 RESULTS 38 2.3.1 Washout of extracellularly bound tracer Ca 38 2.3.2 Rate of basal C a 2 + influx 38 2.3.3 Inhibition by gadolinium ion 40 2.3.4 Non-stimulated C a 2 + entry through excitable C a 2 + channels 41 2.3.5 Comparison of organic C a 2 + entry blockers with inorganic gadolinium 42 2.3.6 Expression of candidate genes for channels responsible for non-stimulated 43 C a 2 + entry 2.4 DISCUSSION ....44 2.5 APPENDIX 48 CHAPTER III 50 DIRECT COMMUNICATION BETWEEN PERIPHERAL MITOCHONDRIA AND THE Na + /Ca 2 + -EXCHANGER 3.1 INTRODUCTION. . . , 50 3.2 METHODS 51 3.3 RESULTS 53 3.3.1 Reversal of N C X stimulates mitochondrial Ca2+-uptake 53 3.3.2 C a 2 + influx through revNCX increases [Ca 2 +] in the sub-plasmalemmal space 54 3.3.3 0Na + causes a delayed increase in global cytosolic [Ca 2 +] 56 v 3.3.4 SR and mitochondria compete for uptake of reverse-NCX mediated Ca entry 56 3.3.5 Mitochondrial Ca buffering moderates reverse-NCX mediated 57 elevation of [Ca 2 + ] S U bPM 3.3.6 N C X reversal influences the tail of the mitochondrial response to ATP 58 3.3.7 Agonist-mediated up-regulation of mitochondrial C a 2 + extrusion 61 3.4 DISCUSSION 62 3.4.1 Functional and localized interaction of N C X , SERCA and C a 2 + uniporter 63 3.4.2 SERCA and the C a 2 + uniporter compete for C a 2 + entering the cell through 63 reverse-mode N C X 3.4.3 NCX-reversal stimulates mitochondrial C a 2 + uptake upon purinergic 65 receptor stimulation 3.4.4 Conclusion 66 CHAPTER IV 68 IP3 AND RYANODINE RECEPTORS LOCALIZE TO SEPARATE SR SUB-COMPARTMENTS THAT PERFERENTIALLY CO-LOCALIZE NEAR MITOCHONDRIA 4.1 INTRODUCTION 68 4.2 METHODS ....70 4.3 RESULTS 73 4.3.1 Mitochondria detect activation of IP3R and RyR by purinergic stimulation 73 4.3.2 Cytosolic detection of Ca2+-release from IP3R and RyR during purinergic 74 stimulation 4.3.3 Mitochondrial stimulation by I P 3 R and RyR as a common response to 75 vasopressor agonists 4.3.4 Imaging IP3R and RyR distribution in R A S M C 77 4.3.5 Localization of IP3R and RyR to separate SR elements 78 4.3.6 Visualizing the relationship of IP3R and RyR with mitochondria 80 4.3.7 Preferential co-localization of IP3R and RyR near mitochondria 81 4.4 DISCUSSION 84 4.4.1 Functional Localization of IP 3R and RyR in V S M 85 4.4.2 Functional interaction of IP3R and RyR 86 4.4.3 Spatial association of IP3R and RyR near mitochondria 89 4.4.4 Conclusion 90 vi CHAPTER V 91 GENERAL CONCLUSIONS & RECOMMENDATIONS FOR FUTURE WORK 5.1 IMPLICATIONS OF B A S A L C a 2 + E N T R Y 91 5.2 MITOCHONDRIA A N D L O C A L C a 2 + EVENTS 92 5.2.1 Mitochondria and plasmalemma Na /Ca -exchange. 92 5.2.2 Agonist-mediated stimulation of mitochondria by IP3R & RyR. 92 5.3 CLOSING COMMENTS 93 BIBLIOGRAPHY 94 vii LIST OF TABLES page Table 1.1. Intracellular linked Ca + and Na + transport 6 Table 1.2. Abridged list bf mitochondrial inhibitors 16 Table 2.1. Oligonucleotide sequences of RT-PCR primers 44 Table 4.1. Co-localization statistics of IP3R and RyR in dual-labeled cells. 79 Table 4.2. Co-localization statistics of IP3R and RyR with mitochondria. 81 viii LIST OF FIGURES page Figure 1.1 Structure of the blood vessel wall 1 Figure 1.2 Calcium and Excitation-contraction coupling 3 Figure 1.3 C a 2 + cycling at PM-SR junctions in smooth muscle 8 Figure 1.4 Mitochondrial junctions with the sarcoplasmic reticulum and plasmalema. 12 Figure 1.5 Mitochondrial 3-way regulation of C a 2 + signaling. 15 Figure 1.6 Inhibition of mitochondrial Ca2+-extrusion in SMC 23 Figure 2.1. Optimizing washout of excess tracer. 38 Figure 2.2. Exponential 4 5 C a 2 + uptake. 40 Figure 2.3. SNAP-GFP expression pattern and 3-D reconstruction of a representative cell. 40 Figure 2.4. Complex concentration-dependent inhibition of 4 5 C a 2 + uptake by gadolinium ion. 41 Figure 2.5. Excitable calcium channels mediate calcium leak. 42 Figure 2.6. A combination of SKF 96365 and 2-APB fully blocks resting influx through 43 membrane channels. Figure 2.7. TrpC mRNA expression profile in cultured rat aortic smc. 44 Figure 3.1 Reversal of Na+/Ca2+-exchange stimulates mitochondrial Ca2+-uptake in cultured 54 smooth muscle cells Figure 3.2 Reverse-NCX increases sub-plasmalemmal [Ca 2 +] 55 Figure 3.3 Effects of N a + substitution and SERCA blockade on [Ca 2 +]i 56 Figure 3.4 The sarcoplasmic reticulum and mitochondria buffer and compete for revNCX- 58 mediated C a 2 + influx Figure 3.5 C a 2 + influx contributes to the tail of the mitochondrial responses to ATP 59 Figure 3.6 NCX-reversal causes mitochondrial Ca2+-uptake during agonist stimulation 61 Figure 3.7 Reverse-mode N C X increases mitochondrial C a 2 + flux in ATP-stimulated cells 62 Figure 3.8 Model for interaction of N C X , SERCA and mitochondria in vascular smooth 64 muscle Figure 4.1. Purinergic stimulation increases [Ca 2 + ]MT by activation both I P 3 R and RyR 74 Figure 4.2. Both I P 3 R and RyR contribute to the ATP-mediated elevation of cytosolic C a 2 + 75 Figure 4.3. Activation of V i vasopressin receptors increases [ C a 2 + ] M T by activation both 76 IP 3 RandRyR Figure 4.4. ATP and A V P stimulate overlapping populations of mitochondria 77 Figure 4.5. Subcellular distribution of I P 3 R , RyR and actin in R A S M C 78 ix Figure 4.6. Co-localization analysis of I P 3 R & RyR 79 Figure 4.7. Mitochondrial association with I P 3 R and RyR 80 Figure 4.8. Modeling the effect of voxel clustering on random co-localization of 83 mitochondria associated I P 3 R and RyR Figure 4.9. Assessing the selectivity of procaine for R Y R versus mitochondria 84 x LIST OF SYMBOLS & ABBREVIATIONS [Ca2+] free ionic Ca2+ concentration mito-aeq mitochondrial targeted aequorin [Ca ] c y t 0 [Ca2+] in the bulk cytosol mito-GFP mitochondria targeted GFP [Ca 2 + ] M T [Ca2+] in the mitochondrial matrix mitoNCX mitochondrial Na+/Ca2+ [Ca 2 +] S R [Ca2+] in the sarcoplasmic exchanger; mNCX reticulum MLCK myosin light chain kinase [Ca2+] s ubPM [Ca2+] in the subplasmalemmal domain MLCP myosin light chain phosphatase 2-APB 2-aminoethyl diphenylborate MT mitochondria(l) ADP adenosine diphosphate mUni mitochondrial Ca2+ uniporter ANT adenine nucleotide translocase NA numerical aperture ATP adenosine 5'-trisphosphate NKA Na+/K+-ATPase AVP [ARG8]-vasopressin Na+/Ca2+ exchanger 1,2-Bis(2-amino-5- NCX BAPTA- bromophenoxy)ethane-Ar,A/,A?',A;- Pi phosphate AM tetraacetic acid acetoxy methyl PKC protein kinase C ester. A calcium chelator Cam-KII Ca2+-calmodulin dependent kinase II 2+ PM-aeq See SNAP-aeq CCE PM plasmalemma capacitative Ca entry PTP mitochondrial permeability transition pore CPA cyclopiazonic acid (SERCA blocker) CsA cyclosporin A ROK Rho kinase cytoC cytochrome C RyR ryanodine receptor Ca2+ channel DIDS 4,4'-diisothiocyanostifbene-2,2'-disulfonic Acid mitochondrial membrane potenial R membrane bound agonist receptor Av(/m SERCA sarco/endoplasmic reticulum ATPase EGTA Ethylene glycol-bis(2-aminoefhyl)-N,N,N',N'-tetraacetic acid SMC smooth muscle cell p membrane potential SNAP-aeq SNAP-25 tagged, sub-ER endoplasmic reticulum plasmalemmal targeted aequorin ETC electron transport chain SOC store-operated channel FCCP carbonyl cyanide 4- SR sarcoplasmic reticulum (trifluoromethoxy)phenylhydrazone SSR superficial SR GFP green fluorescent protein TrpC transient receptor potential gene H 2 0 2 hydrogen peroxide TRPC canonical transient receptor HXC H+/Ca2+ exchanger potential channe IPj inositol 1,4,5-trisphosphate U mitochondrial Ca2+ uniporter IP3R inositol 1,4,5-trisphosphate receptor VDAC voltage-dependant anion channel (aka -porin) L vasopressor receptor ligand VSMC VSM vascular smooth muscle cells vascular smooth muscle LIST PUBLICATIONS & ABSTRACTS Material from this dissertation has been published in: Poburko D, Lhote P, Szado T, Behra T, Rahimian R, McManus B, van Breemen C, Ruegg UT. Basal Calcium entry in vascular smooth muscle. European Journal of Pharmacology. 2004; 505(1-3): 19-29. • Poburko D, Lee C-H, van Breemen. Vascular smooth muscle mitochondria at the cross roads of Ca 2 + regulation. Cell Calcium. 2004; 35(6): 509-21. (Invited Review) • Poburko D, Kuo K-H, Dai J, Lee C-H, van Breemen C. Organellar junctions promote targeted Ca 2 + signaling in smooth muscle: Why two membranes are better than one. Trends in Pharmacological Sciences. 2004 Jan; 25(1): 8-15. (Invited Review) Material from this dissertation has been submitted for publication in: • Damon Poburko, Kathryn Potter, Edo van Breemen, Nicola Fameli, Olivier Basset, Urs T Ruegg, and Cornelis van Breemen. Mitochondria buffer NCX-mediated Ca2+-entry and limit its diffusion into vascular smooth muscle cells. Cell Calcium Material from this dissertation has been presented in poster format at the following international meetings: • Damon Poburko, Megan McLarnon, Eric Lin, Hubert Walinski, Edo van Breemen, Cornelis van Breemen. Activation of different G-protein linked receptors leads to convergent mitochondrial Ca 2 + signaling in aortic smooth muscle cells. FASEB Experimental Biology Congress, April, 2005, San Diego, California. FASEB JOURNAL 19 (5): A1084-A1084 Part 2 Suppl. S MAR 7 2005 • Damon Poburko, Nicola Fameli, Aaron Wong, Cornelis van Breemen. Impairing mitochondrial Ca2+ uptake mechanisms and collapsing of the TAm differentially affect subplasmalemmal Ca 2 + signaling. Biophysical Society Annual Meeting, February 12-16, 2005, Long Beach, California. BIOPHYSICAL JOURNAL 88 (1): 90A-91A Part 2 Suppl. S JAN 2005 • Damon Poburko, Kate Potte, Edo van Breemen, Megan McLarnon, Cornelis van Breemen. Linked Ca 2 + transport between smooth muscle mitochondria and the Na+/Ca2+-exchanger. FASEB Experimental Biology Congress, April, 2004, Washington, DC. FASEB JOURNAL 18 (4): A703-A703 Suppl. S MAR 23 2004 • Damon Poburko, Kathryn Potter, Edo van Breemen, Enrique Gallego, Urs Ruegg, Cornelis van Breemen. Privileged Ca 2 + communication between smooth muscle mitochondria and the Na+/Ca2+-exchanger. CIHR-ICRH Young Investigator's Forum, May 6-9, 2004, Winnipeg, Canada. • Poburko D, Gallego E, van Breemen C. Altered mitochondrial Ca 2 + transport could underlie cyclosporin A-induced hypertension. Canadian Society for Clinical Pharmacology 25th anniversary meeting. Sept. 19-21. Ottawa, Canada. • Poburko DT, Ruegg UT, van Breemen C. Mitochondrial calcium transport in resting & stimulated smooth muscle. FASEB JOURNAL 17 (4): A46-A46 Part 1 Suppl. S MAR 14 2003 (resulted in invitation to publish both a primary paper and scholarly review) • Damon Poburko, A. Mike Keep, Amrit Mahil, Megan McLarnon, Comelis van Breemen. Vascular Smooth Muscle Heterogeneity of Calcium Recycling & Junctional Composition. IUPHAR World Congress, July 12-15, 2002, San Fransisco, Ca. • Poburko D, van Breemen C. Mitochondrial calcium transport in resting & stimulated vascular smooth muscle. Frontiers in Cardiovascular Science. February 02-03, Vancouver, British Columbia • Damon Poburko, Tania Szado, Tasneim Behra, Roshanak Rahimian, Bruce McManus, Cornelis van Breemen, Urs T. Ruegg. Ca 2 + influx pathways mediating basal calcium entry in vascular smooth muscle: The calcium leak and excitable channels. IUPHAR satellite meeting: Vascular Neuroeffectors Meeting, Julyl2-15, 2002, Grenlibakken Resort, Lake Tahoe, California. Material from this dissertation has also been presented orally for the Graduate Student Seminars Series in the Department of Pharmacology and Therapeutics at U B C ACKNOWLEDGEMENTS M y time as a graduate student has been incredibly enjoyable and educational and has taught me many lessons. For this my deepest thanks go to Dr. Casey van Breemen, who embodies everything good that one could ask for in a scientist, a supervisor, and friend. To him I owe an immense debt of gratitude for the knowledge and experience that he shared, for his trust and encouragement, and his fathomless patience. Above all, he opened my eyes to the wonder of research and the value of a balanced life. Thank you. Over the last few years, I have had the opportunity to learn from the experiences of fellow trainees and to work with many students and technicians. I would especially like to thank Mike Keep, Megan McLarnon, Edo van Breemen, and Kate Potter for your hard work and dedication and for keeping everyone smiling. To Tania Szado, I owe a special thanks for establishing the aequorin technique in the lab. Thanks to Eric Lin for his patient lessons and assistance with deconvolution and confocal microscopy. To the rest of the gang at the BC Research Institute for Children's and Women's Health and at St. Paul's Hospital, thank you for your technical support, laughs and insightful discussions. I must also recognize two special individuals, Dr. Urs Ruegg and Mr. Philippe Lhote in Geneva & Lausanne, Switzerland for all their supervision and excellent technical assistance. I will never forget my time in Lausanne, and it would not have been so productive without the commitment of these two to our common goals. Thank you. To my most amazing family and friends, thank you for your encouragement, advice, love and support. Thank you for understanding when I couldn't be there, and thank you for being there when I needed you. Finally, I would like to acknowledge the Natural Sciences and Engineering Research Council of Canada, the Michael Smith Foundation for Health Research and the Killam Foundation for their financial support during my training. xiv CHAPTER I Introduction to calcium signaling in vascular smooth muscle 1.1 The basics of smooth muscle. 1.1.1 Physiological role of smooth muscle. Blood vessels provide the primary route of transportation for nutrients, immune cells, and signaling molecules in the body. Like most hollow organs in the body, the function of blood vessels is highly dependent on the layer of smooth muscle, or tunica media, that surrounds them. The vessel wall is further composed of an outer layer of connective tissue (tunica adventitia) that provides structural integrity to the vessel, and the endothelial layer (tunica intima) that lines the lumen of the vessel (figure 1.1"). In their contractile phenotype, the basic physiological role of the vascular smooth muscle (VSM) is to regulate the diameter of the vessel lumen in order to control perfusion pressure and direct regional blood flow. The physiological control of vascular tone is modulated by numerous factors, thus V S M C are responsive to changes in perfusion pressure, autonomic stimuli through the innervation of larger vessels, paracrine and autocrine receptor ligands, and oxygen tension. In damaged vessels however, V S M cells can differentiate to a secretary/ proliferative phenotype to assist with blood vessel repair. In both phenotypes, changes in the free ionic concentration of calcium in the cytosol ([Ca ],) play a central role in the regulation of multiple functions of the V S M . However, with increasing age, poor diet and a lack of exercise, the regulation of V S M becomes increasingly prone to failure. This failure can lead to hypertension, vasospasm and may contribute to the progression of atherosclerosis. smooth muscle vaso vasorurri (tunica media) fibroblasts nerve fibres i n t e r n a l e last ic l a m i n a endothelium {tunica intima) fibrous adventicia (tunica adventicia) Figure 1.1. Structure of the blood vessel wall. 1 1.1.2 The mechanism of smooth muscle contraction. Cytosolic free ionic C a 2 + is the primary second messenger linking membrane excitation or stimulation to the contraction of smooth muscle cells. Vascular smooth muscle cells can be described as tapering cylinders 3-10 um in diameter depending on the state of contraction and 50-500 urn in length (33). In most vessels, the V S M C are wrapped around the blood vessels in a slightly helical manner, while in some vessels a layer of smooth muscle is also oriented along the longitudinal axis of the cell. Within the smooth muscle cell is a contractile apparatus consisting of thick myosin filaments surrounded by thin cc-actin filaments that are anchored to dense bodies throughout the cytosol and dense bands (or plaques) at the plasma membrane (PM) (210). Elevation of [Ca 2 +]; promotes the binding of 4 C a 2 + ions to calmodulin (CaM), which then binds to and activates the myosin light chain kinase (MLCK) (Figure 1.2). Activated M L C K then phosphorylates the myosin light chain and stimulates acto-myosin cross-bridge cycling (188). The activity of M L C K is opposed by myosin light chain phosphatase (MCLP). The opposing actions of M L C L and M C L P on myosin phosphorylation can also be modulated by Ca 2 + -independent phosphorylation events, causing an apparent shift in the Ca2+-sensitivity of the myofilaments. For example, Rho-kinase (ROK) can phosphorylate M L C P and impair its dephosphorylation of myosin thereby favouring contraction (Figure 1.2) (218). Phosphorylation of the actin-associated proteins, caldesmon and calponin, also regulate their inhibitory effect on cross-bridge cycling (6;69). Relative to cardiac or skeletal muscles, smooth muscle can maintain tension with much less energy expenditure by virtue of a prolonged association of actin and myosin upon myosin dephosphorylation known as the "latch-state" (97;171). This model has recently been expanded into an 8-state model, which further accounts for the phosphor-regulation of the thin actin filament (96), and future iterations may include regulatory influences of proteins such as ROK. 1.1.3 Mechanisms of C a 2 + Homeostasis and Signaling in VSM. In order for [Ca2 +]j to increase and stimulate the contractile apparatus or other Ca 2 + -dependent processes, C a 2 + must enter the cytosol from the extracellular space or be released into the cytosol from the sarcoplasmic reticulum (summarized in Figure 1.2). 2 PM Figure 1.2 Calcium and Excitation-contraction coupling. A general scheme is depicted for the agonist dependent elevation of intracellular [Ca2+] via activation of phospholipase C. Ca 2 + release from the SR activates store-operated Ca2+-channels, while DAG activates non-selective cation channels that mediate the Na+-influx required for NCX reversal. Elevation of [Ca2+]s causes Ca 2 + to bind to calmodulin. Ca 2 + 4-CaM then activates MCLK, which phosphorylates myosin and stimulates acto-myosin cross-bridge cycling. MLCP-dependent dephosphorylation of myosin is impaired by Rho-kinase that is stimulated by receptor activation. Abbreviations: A, actin; CaM, calmodulin; cav, caveolae; DAG, diacylglycerol; IP3- inositol trisphosphate; I P 3 R , IP3 receptor; M, myosin; MLCK, myosin light chain kinase; MCLP, myosin light chain phosphatase; NCX, Na+/Ca2+-exchanger; PLC, phospholipase C; R, G-protein coupled receptor; ROC, receptor-operated channel; ROK, Rho kinase; RyR, ryanodine receptor; SERCA, sarco/endoplasmic reticulum Ca2+-ATPase; SOC, store-operated channel; SR, sarcoplasmic reticulum; VGCC, voltage-gated Ca 2 + channel. The permeation of Ca across the smooth muscle P M is mediated by several types of channels. Depolarization activates a family of voltage-gated C a 2 + channels, which are encoded by separate genes. It has also been known for at least 30 years that depletion of SR C a 2 + stores and activation of G-protein coupled receptors can activate C a 2 + influx through what have been termed store-operated channels (SOC) and receptor-operated channels (ROC). Only recently were the transient receptor potential proteins (TRPs) identified as molecular candidates for SOCs and ROCs (21). In smooth muscle it is currently debated whether SOCs are primarily Ca 2 + -3 selective or non-selective channels, but either way channels sensitive to SR Ca depletion likely contain some combination of TRPC1, TRPC4, and TRPC5. In contrast, ROCs appear to be activated by diacylglycerol (DAG) that is generated with inositol-1,4,5-trisphosphate (IP3) following hydrolysis of phosphatidyl inositol by phospholipase C (PLC) (225). These ROCs are non-selective cation channels (NSCC) and are likely composed of TRPC3 and TRPC6 (21 ;225). Since the extracellular concentration of N a + is ~ 100-fold higher than that of Ca 2 + , these channels mediate substantial Na + influx, which in turn causes the Na+/Ca2+-exchanger (NCX) to reverse and bring C a 2 + into the cell (11). Release of C a 2 + from the sarcoplasmic reticulum is mediated by two types of Ca 2 + -channels, ryanodine receptors (RyR) and in IP 3 receptors (IP3R) each of which is expressed in three isoforms. Both channels are essentially gated by local [Ca 2 +], such that cyclic-ADP-ribose (cADPr) and IP 3 alter the Ca2+-affinity of the RyR and D?3R, respectively. Release of SR C a 2 + induced by PLC-linked receptors, such as a-adrenergic receptors, is often associated with the production of IP3 and the activation of IP3R. In contrast, less is known about linkage between receptor activation and the production of cADPr by ADP ribose cyclase (aka CD38) (15). Rather the activation of RyR is most often attributed to Ca2+-induced Ca2+-release (CICR) in response to local elevation of [Ca 2 +] (254). Despite the fact that both receptors release C a 2 + into the cytosol, evidence suggests that they may be localized to separate elements of the SR (85;153), which may allow each receptor to regulate specific processes in the cell. Calcium acts is a second messenger for numerous signaling cascades, and cells possess sophisticated and redundant mechanisms to clear C a 2 + from the cytosol. Both the sarcoplasmic reticulum and mitochondria serve as rapid cytosolic buffers of Ca 2 + . However, for the intracellular C a 2 + content to be maintained at a steady-state level in the face of basal or stimulated C a 2 + entry, the C a 2 + that is taken up by these buffers must eventually be extruded from the cell. With a [Ca2+]j of ~100nM at rest and an extracellular [Ca 2 +] of ~1.5mM, Ca extrusion is an energy dependent process that is mediated by the plasma membrane Ca -ATPase (PMCA) and the N C X , which extrudes C a 2 + in exchange for N a + at an electrogenic ratio of lCa 2 +:3Na +(33). Having introduced the major players in V S M C a 2 + signaling, the next two sections discuss in detail how the behaviour of individual smooth muscle cells requires precise choreography between these ion translocators, with special emphasis the role of mitochondria. 4 1.2 LINKED Ca 2 + TRANSPORT IN VASCULAR SMOOTH MUSCLE 1 Calcium (Ca 2 +) regulates nearly all fast processes in the body, including contraction, chemotaxis, secretion, synaptic transmission, and several slower processes, including fertilization, proliferation, learning, memory and apoptosis. An important unresolved question is how the fluctuations of the concentration of a single inorganic ion, Ca 2 + , can regulate such a multitude of cellular processes. On the one hand, cells have developed a plethora of Ca 2 + -sensitive, signal-transducing proteins to tailor their cell-specific regulation of many diverse processes by C a 2 + (recently reviewed in (23) ). Yet this cannot entirely explain how multiple processes, such as cross-bridge cycling and myosin filamentogenesis in smooth muscle or endothelial nitric oxide (NO) and epoxyeicosatrienoic acid production, can be regulated simultaneously. It has been proposed that the solution lies in the physical and temporal separation of numerous targets for Ca 2 + , combined with the generation of localized cytoplasmic C a 2 + gradients (176). Clearly, mitochondrial dehydrogenases, voltage-gated C a 2 + channels (VGCCs), inositol (l,4,5)-trisphosphate (IP3) receptors (IP3R) and DNAses are localized in different sub-cellular regions and could be selectively activated by focal Ca signals. Moreover, activation of certain target proteins by Ca2 +-calmodulin could be site-specific, despite the wide-spread intracellular distribution of calmodulin (CaM), since CaM can be tethered to effector complexes, such as smooth muscle myofilaments (256). Analysis of the interaction of Ca2+-transport molecules in the plasmalemma, the sarco/endoplasmic reticulum (SR/ER), the nuclear envelope and mitochondria suggests that these interactions provide structural basis for the spatially and temporally encoded fluctuations in cytosolic concentration of C a 2 + ( [Ca 2 + ] i ) that are thought to mediate site-specific C a 2 + signaling. This interaction takes place in two fundamentally different ways: (i) directed C a 2 + supply to or removal from the Ca2+-sensing domains of signal-transducing molecules and C a 2 + translocators; and (ii) C a 2 + delivery from a transport site located in one membrane to a second C a 2 + transport site located in an apposing membrane. An example of the first type of interaction is the delay of Ca2+-mediated inhibition of VGCCs due to nearby mitochondria sequestering C a 2 + (156). The second type of interaction is exemplified by coupling of C a 2 + entry through the N C X to the sarco-endoplasmic reticulum Ca 2 +-ATPase (SERCA) during store refilling (133). The concept of preferential C a 2 + delivery from one transporter to the next is referred to as "linked C a 2 + 1 Aversion of section 1.2 has previously been published as: Poburko D, Kuo KH, Dai J, Lee CH, van Breemen C. Organellar junctions promote targeted Ca2+ signaling in smooth muscle: why two membranes are better than one. Trends Pharmacol Sci.. 25(1):8-15. Review. 5 transport." This process circumvents free diffusion throughout the cytoplasm and, typically, takes place at organellar junctions where physically restricted diffusion of Ca within the narrow cytoplasmic domain is further slowed down by a high density of fixed, negatively charged C a 2 + binding sites (24). Table 1.1 lists both known and hypothesized junctions within smooth muscle. Table 1.1. Intracellular linked C a 2 + and Na + transport a"c. Delivering Structures Acceptors Event Rpf Membrane Translocator Membrane Translocator Ca2+ PM VGCC iiiissRii: SERCA Buffering PM SOC SSR ' SERCA Refilling (133) PM SERCA Refilling (133) SSR RYR PM NCX Unloading (162) SSR IP3R PM NCX Unloading (162) SR IP3R & RYR SR SERCA Recycling (162) LR1 if (1 EK2 SERCA Redistribution I l l f f i l l l l SR/ER ( » ( > MT uniporter Mito. Ca2 + sig. (74;220) MT MNCX i i i i s i i i i SERCA ., buffering (10) MT MHCX SR/ER SERCA » » N/A MT PTP depol lllKiSR/ER; SERCA (36) SR/ER IP3R & RYR MT PTP pol >» ») N/A Na+ PM SOC PM NCX reversal (133)' PM SOC PM NKA<x3 Na+ extrusion (12) PM IIIIB^ ^ PM NKA u3 Na" extrusion (12) a The ion translocators delivering Ca 2 + or Na + are listed under "Delivering Structures," and the transport molecules accepting the ions, thus completing the linked transport, are listed under "Acceptors." b Although some of the above mechanisms remain hypothetical, various levels of supportive evidence has been presented for all cases (see main text). ' Abbreviations: ER & ER1, endoplasmic reticulum; ER2, adjacent but luminally distant endoplasmic reticulum; for standard abbreviations see main list of abbreviations 1.2.1 Structural considerations of PM-SR junctions. The SR of smooth muscle is a continuous network of membranous tubules and sheets that can be divided into superficial and deep SR and the nuclear envelope (164). Electron microscopy of serial sections shows that the superficial SR forms a flattened pedestal as it approaches the P M . The junctional cytoplasmic space between the P M and SR pedestals has a consistent width of 15-20 nm and a diameter of up to 500 nm. In vascular smooth muscle (VSM) cells of the rabbit inferior vena cava (JVC), 14.2 ± 0.7% of the P M (including the necks of caveolae) is closely associated (within 30 nm) with the superficial SR, the junctional width averages 19+1 nm (133), and the apices of the caveolae protrude through this thin fenestrated junctional SR (58;63). At present, it is unknown what causes a particular portion of the P M to form a junction, although such junctions are common in caveolae rich regions of the P M . The junctional P M might have specific chemical characteristics, such as those of cholesterol and sphingolipid-rich 6 lipid rafts, which form platforms for signaling and transport molecules (14), or plasmalemmal receptors may be physically linked to SR proteins such as the linkage of metabotropic glutamate receptors to SR release channels via Homer proteins (72). Interestingly, the alo-adrenoceptor contains a consensus sequence for Homer protein binding (266). The regularity of the width of the PM-SR junction both within a single junction and between junctions is similar to that seen in the triads and diads of skeletal and cardiac muscle. In these tissues junctophilins 1, 2 and 3 are candidates for the structural cement between the T-tubular and SR membranes. Junctophilins (JP) are SR/ER transmembrane proteins with a cytoplasmic region (MORN motifs) that bind to the plasma membrane, and JP-2 is critical to normal calcium signaling in cardiac myocytes (223). Unfortunately the difficulty in isolating smooth muscle PM-SR junctions has so far precluded identification of their chemical nature. 1.2.2 Function of PM-SR junctional complexes Dynamic C a 2 + buffering and cycling in resting smooth muscle. The "superficial buffer barrier" function of the superficial SR (SSR) in smooth muscle provided the first example of linked C a 2 + transport (4;233). In this case a fraction of C a 2 + entering the cell through plasmalemmal C a 2 + channels is taken up by SERCA of the superficial-SR before it has time to diffuse to the deeper myoplasm (Table 1.1 and Figure 1.3). The concept of active C a 2 + uptake by an organellar network to limit diffusion into other regions of the cell was subsequently applied to the "nuclear C a 2 + buffer barrier" (249) in coronary smooth muscle and the "mitochondrial C a 2 + buffer barrier" in pancreatic acinar cells (170). To maintain the superficial buffer barrier, C a 2 + accumulating in the superficial-SR is recycled to the extracellular space as it is released by the superficial SR in a vectorial manner into the PM-SR junctional space coupled to the forward mode of the N C X in the P M (162). The superficial buffer barrier occurs in many types of smooth muscle including coronary artery, airway and uterine smooth muscle (115;214;222), and it provides a mechanism to generate peripheral C a 2 + gradients and protect cells from C a 2 + overload. This protective role appears to be compromised in diabetic vascular disease and up-regulated as a result of exercise (249). Based on the morphology described above, Figure 1.3 illustrates that the superficial buffer barrier involves two types of linked C a 2 + transport: (i) "loosely coupled" transport, where SERCA captures part of the Ca entering the cell through the non-junctional P M , possibly including the caveolar apices; and (ii) "tightly coupled" transport within the PM-SR junction between SR release channels and the N C X . When the N C X is blocked, C a 2 + is conserved within the SR by SERCA mediated re-uptake, indicating 7 resting Ca 2 + cycle • N C X H S O C L A ca2<ct}—4== P M C A . S E R C A activated Ca 2 + cycle P M C A L S E R C A myofilament Figure 1.3. Ca 2 + cycling at PM-SR junctions in smooth muscle, (a) This electron micrograph shows superficial-SR (white arrow heads) associated with the PM (black arrow heads) and caveole (asterisk), (b) Ca2+ cycling in the superficial buffer barrier in non-stimulated VSM. The superficial-SR buffers Ca 2 + entry due to basal activity of excitable Ca 2 + channels, and cycles it, via release channels (RC: IP3R, RYR), to the NCX in the junctional space (JS) to be extruded. Ca 2 + is also extruded by the PMCA in non-junctional PM. (c) This EM depicts superficial and radial SR (white arrow heads) associated with a mitochondrion (MT), caveolae (asterisk) and plasmalemma (black arrow heads ). (d) Ca2+ fluxes associated with agonist-mediated smooth muscle stimulation. Ca 2 + is release from IP3R near the myofilaments, and SR depletions and receptor stimulation opens a Na+ and Ca 2 + permeable SOC. Na+ influx depolarizates the cell and opens VGCC, while the [Na+] builds in the junction and reverses NCX to bring Ca 2 + to SERCA for store refilling. Other abbreviations: IP3 - inositol(l,4,5)-trisphosphate, RC - release channel, RY -ryanodine,. Scale bar = 400 nm. ** electron micrograph courtesy of Dr. Kuo-hsing Kuo. 8 that SERCA are located in or very near the PM-SR junctions (162). Only when both N C X and SERCA are blocked is Ca2+-release from the SR extruded from the cell by the plasmalemmal Ca 2 +-ATPase (PMCA), which, in this instance, is assumed to be located in non-junctional P M . Co-localization of calsequestrin within the superficial-SR and N C X in the P M (157) provides morphological support for the transport scheme of Figure 1.3, but more precise imaging is required to test the rank ordering of linkage between HVryanodine (RY) receptors and N C X , SERCA and P M C A . Analogous to the superficial buffer barrier, Ca2+-uptake by SERCA located in the nuclear envelope, which displays extensive inwardly directed cristae, can regulate the flow of C a 2 + between the cytoplasm and nucleus. This mechanism, termed the "nuclear buffer barrier" is postulated to exert control over gene expression (249). In this context, it is most interesting that experimental separation of the inner and the outer membranes of the nuclear envelope from hepatocytes revealed that SERCA was specifically localized to the outer membrane of the nuclear envelope (107), and that SERCA inhibition with thapsigargin abolishes the C a 2 + gradient between nuclear and cytosolic [Ca 2 +] (103). Wave-like [Ca2+]j oscillations in activated smooth muscle. C a 2 + oscillations appear to be a common regulatory mechanism in several types of smooth muscle (2). In large veins and arteries, cc-adrenergic-mediated force development is regulated by recruitment of cells and the frequency of asynchronous, agonist-induced C a 2 + oscillations that are primarily generated by the release of C a 2 + from the SR (70; 199). Wave-like [Ca2 +]j oscillations in V S M are efficient in activating myofilaments because the SR network penetrates deep into the myoplasm (133), and the frequency-encoded Ca2+-signals provide more precise contractile modulation than global changes in average [Ca2+]j. In resistance arteries, asynchronous waves have a dilatory effect due to the opening of Ca2+-activated K+-channels (Kc a) (113); however, oscillatory vasomotion is induced when they become synchronized during exposure to agonists (172). While the exact mechanisms underlying vascular heterogeneity of smooth muscle C a 2 + signaling are unresolved, differences in the ion transporters found in junctional complexes could be important. Furthermore, [Ca 2 +]; oscillations could serve to modulate frequency-encoded Ca2+-sensitive mitochondrial dehydrogenases and gene expression. A detailed description of the ionic mechanism of smooth muscle Ca oscillation has been reported for adrenergicaly-stimulated rabbit IVC and is illustrated in Figure 1.3. First, a i-adrenergic receptor stimulation leads to saltatory SR Ca2+-release from IP3R, spreading across 9 the cell from Ca -wave initiation sites, which may be similar to so called "hot spots." Secondary activation of ryanodine receptors (RyR) also occurs in some tissues. Subsequently, a P M channel permeable to both N a + and C a 2 + is activated, either by store depletion or receptor activation (11). The resultant inward current depolarizes the P M activating VGCCs. In the TVC and trachea, where V G C C play a modulatory but not permissive role in contractile regulation, Na + is postulated to enter the PM-SR junctional space to increase the local N a + concentration ([Na+]pM-SR) sufficiently to cause N C X reversal (133). There is convincing evidence that [Na+]pM-sR is separately regulated by the al- and ct3-isoforms of the Na +/K +-ATPase (NKAcc2/3) (12). The reverse-mode N C X bring C a 2 + into the cell, which is taken up into the SR by SERCA in preparation for the next wave of Ca2+-release. Ca 2 +-influx associated with SR refilling also balances the loss of cell C a 2 + through augmented C a 2 + extrusion during elevated [Ca2+]j. In support of the above model, separation of PM-SR junctions with calyculin A has been shown to abolish C a 2 + oscillations (131; 132). Blockade of IP3R with 2-aminoethoxy-diphenylborate (2-APB) immediately terminates the [Ca 2 +]i oscillations, while blockade of SOCs and VGCCs allows continuation of a few oscillations before silencing the C a 2 + waves (132). Site directed mutagenesis of the IP3 receptor to reduce its C a 2 + sensitivity also abolished V S M Ca2+-waves, convincingly demonstrating the primary importance of the IPvmediated Ca release in this process (109). C a 2 + spark/wave-STOC coupling in relaxing smooth muscle. Spatially restricted membrane currents through K c a , or spontaneous-transient outward currents (stocs), are activated by local increases in sub-plasmalemmal [Ca 2 +] resulting from spontaneous RYR-mediated Ca2+-release events (sparks) (254). Although spark-mediated [Ca 2 +] elevations can spread over a much larger distance than the 20 nm PM-SR junctional width (113), kinetic considerations for the relationship between sparks and stocs indicate that the RyRs producing sparks are within 20 nm of the K c a in the P M (for current reviews on sparks see (32; 102). This suggests for instance that cerebral resistance vessels possess PM-SR junctions that are different from those in the IVC, being specifically designed to hyperpolarize the P M and mediate relaxation. The presence of Kc a , , as opposed to say N C X , in PM-SR junctions of cerebral resistance arteries further explains why Ca2+-waves in this tissue are linked to relaxation (112). CICR in activating smooth muscle. Calcium entering smooth muscle through V G C C can activate RyR, and the SR Ca2+-release amplifies the contraction-inducing C a 2 + signal. 10 Although there is a very close relationship between V G C C and RyR in cardiac muscle, a large body of evidence indicates that Ca2+-induced Ca2+-release (CICR) is not a common mechanism in most smooth muscle types. For example in the uterus, which is mainly dependent on L-type C a 2 + channels, spontaneous contractile activity is not accompanied by fluctuations of SR C a 2 + content (207). However, CICR has been demonstrated in myocytes of some blood vessels, vas deferens and urinary bladder (124). It has been argued that, since CICR in the bladder depends more on the net C a 2 + influx than on the open probability of individual V G C C , and because there is a delay between the current and the subsequent spark, the V G C C are loosely coupled to the RyR with a distance between them in excess of 100 nm (124). This is consistent with the aforementioned idea that the V G C C are not located within the 20 nm cleft of the junction, but rather in the tips of the caveolae. 1.2.3 Structural considerations of mitochondrial junctions Recent years have seen an resurgence of interest in mitochondrial (MT) C a 2 + transport with novel evidence for its influence on both plasmalemmal and SR/ER C a 2 + transport that compliments rather than detracts from more traditional roles for mitochondrial in ATP production and apoptosis (190). A popular view of mitochondrial Ca -uptake is that mitochondria form diffusionally restricted domains with the P M or SR/ER, as in Figure 1.4, that develop C a 2 + levels sufficiently high to overcome the low affinity of mitochondrial Ca 2 + -uniporter, as measured in isolated mitochondria. However, the Ca -affinity of the uniporter in situ is hotly debated. In H E L A and luteal cells, mitochondrial Ca -uptake occurs upon modest [Ca 2 +]i elevation (<300 nM) (47;219), but these reports do not preclude the involvement of MT-ER junctions. Indeed, sustained MT-ER interactions were recently reported in H E L A cells (74), and mitochondria are known to take up the C a 2 + sparks released by nearby RyR (168), which occur more frequently as [Ca 2 +]; increases (254). In smooth muscle, direct measurement of [ C a 2 + ] M T revealed substantial mitochondrial Ca2+-uptake in response to agonist-mediated SR-C a 2 + release, but only modest mitochondrial C a 2 + uptake in response to sustained influx as discussed below (220). The aforementioned report is one of very few direct investigations of V S M mitochondrial Ca2+-signaling. Consequently, much of the following discussion is based on evidence from non-muscle cells, and awaits verification in V S M . 11 Figure 1.4. Mitochondrial junctions with the sarcoplasmic reticulum and plasmalema. (a) This smooth muscle electron micrograph shows the close contact between the SR and mitochondria (MT) that can often enshroud the MT. (c) Close apposition between MT and PM or MT, SR and PM can also be observed. In restricted regions, caveolae (asterisk) and MT are separated by less than 200 nm. (c & d) Probable junctional complexes at MT-SR junctions (c) and PM-SR-MT junctions (d) drawn to relative scale based on current literature. (1 & 4) SR release channels (RY & IP3 receptors) cluster in regions facing MT, likely near contact sites of the inner (IMM) and outer (OMM) mitochondrial membranes. Ca 2 + released into the MT-SR junctional space crosses the OMM via the voltage-dependent anion channel (VDAC) and reaches sufficiently higher [Ca2+] to activate the MT uniporter (U). (2 & 5) The mitochondrial Na+/Ca2+-exchanger (mNCX) or H+/Ca2+-exchange (mHCX) extrude Ca 2 + and help to refill the SR. (3) VGCC and TRPC channels likely localize in caveole, and mitochondria can buffer Ca 2 + entry through these channels implying a close association. Other abbreviations: IP3 - inositol(l,4,5)-trisphosphate, mUni -mitochondrial Ca 2 + uniporter, RY - ryanodine. Scale bar = 400 nm The junctions between mitochondria and SR/ER are more diverse than those between P M and SR discussed above. The width of the MT-SR junctional space in V S M varies between 15 to 50 nm, both within one junction and amongst different junctions within a given cell. The extent of apposition ranges from minor to apparently complete mitochondrial envelopment by SR, and the degree of MT-SR association can vary significantly depending on the state of cellular activation (personal communication - Dr. Kuo-Hsing Kuo, 2005) (250). This implies that MT-SR junctions have a dynamic nature. Indeed, lateral movement of mitochondria relative to the SR has been described (142). This dynamic behaviour is reduced upon physiological elevations of 2"f" [Ca ]j, which has been attributed to inhibition of mechano-chemical enzymes that are proposed 12 to be involved in providing a framework for local MT-ER connections (241;250;262). Potentially important enzymes include dynamin GTPases, dynamin related proteins, microtubules, microfilaments and kynesins (241). Moreover, the long-lasting MT-ER interactions in H E L A were reported in the context of mitochondria exhibiting typical motion and fusion/fission behaviour (74). Consistent with these concepts, the autocrine motility factor receptor appears to be a marker of ER regions in close contact with mitochondria (250). Ion channels of the outer mitochondrial membranes cluster at contact sites between the two membranes (52), and IP3 and RyR have also been shown to localize at SR/ER regions facing the mitochondria hinting at the existence of an anchoring mechanism (98). Supporting the concept of intimate ER-MT signaling, mast cell ER release channels have been estimated to be no more than lOOnm from mitochondrial Ca2+-uptake sites (55). 1.2.4 Function of mitochondrial junctions Mitochondrial serve numerous functions in all cells, and not surprisingly the function of the close associations of mitochondria with the SR/ER and P M is a complex and multi-faceted story that can be examined from many perspectives. From the biosynthesis point of view, the close contacts of M T and ER has long been thought to permit the exchange of mitochondrial derived lipids to the ER (128). In terms of Ca signaling, it is now widely accepted that these close junctions are required to locally generate micromolar elevations of [Ca ], to activate the low-affinity Ca2+-uniporter (Ka ~10u,M) (54;94;179). As a consequence of the ensuing and rapid C a 2 + uptake into the mitochondrial matrix, mitochondria modulate the release and refilling of SR C a 2 + stores in addition to buffering the sub-plasmalemmal C a 2 + gradients that modulate the gating of Ca2+-sensitive P M ion channels. However, the role of mitochondria in V S M C a 2 + signaling extends far beyond the C a 2 + microdomain. Consequently, the function of PM-MT and SR-MT junctions is discussed below in the context of the integral role of mitochondria in V S M C a 2 + signaling. 13 1.3 MITOCHONDRIA AT THE CROSSROADS OF Ca 2 + SIGNALING2 In vascular smooth muscle, C a 2 + signaling regulates vascular tone by an integrated response to endogenous agonists, membrane potential, oxygen supply, and the cellular oxidation-reduction potential. Meanwhile, mitochondrial C a 2 + signaling in V S M is often viewed simply as a means to match ATP production to cellular energy demands by virtue of the Ca -dependence of mitochondrial dehydrogenases and the FiFo-ATPase (38;61). This has led to the proposition that V S M mitochondria may act as integrators of three aspects of Ca signaling seldom discussed together (Figure 1.5). First, mitochondria modulate the activation and inactivation of plasmalemmal ion channels and intracellular Ca2+-release channels by buffering localized C a 2 + gradients in the cytoplasmic microdomains formed by close mitochondrial associations with the plasmalemma and sarco/endoplasmic reticulum (SR/ER) (13;37;67;93;251). These microdomains, subsequently referred to as PM-MT and SR/ER-MT junctions, in combination with tissue-specific channel expression are proposed to contribute to the heterogeneity of C a 2 + signaling observed in different types of V S M (164). Second, mitochondria are thought to contain the V S M oxygen sensor acting through the production of radical oxygen species (ROS) that modulate ion channels and second messenger production, which are central to the regulation of vascular tone. Finally, mitochondria contribute to regulation of cellular redox-potential in response to changes in the availability of metabolic substrates (18). For example, changes in the NADFENAD ratio can alter the metabolism of cyclic-ADP ribose (cADPr) via NADH-mediated cADPr hydrolase inhibition (257). As cADPr alters ryanodine receptor (RyR) C a 2 + sensitivity, the subsequent effect will depend on the tissue-specific role of RyR-mediated Ca2+-release. This model provides a working hypothesis to describe many of the effects of mitochondrial inhibition reported in V S M , but requires further evidence to validate whether or not mitochondria are indeed a central axis in V M S C a 2 + signaling. Before the importance of mitochondria in C a 2 + signaling was clearly established (13; 193), the clinical relevance of mitochondrial function in V S M was illustrated by mitochondrial encephalomyopathies that are caused by specific insufficiencies within the electron transport chain and exhibit altered mitochondrial morphology and possibly Ca signaling (51). In spite of the extensive MT-SR associations first reported in V S M more than 10 2 A version of section 1.3 has been published as: Poburko D, Lee CH, van Breemen C. (2004) Vascular smooth muscle mitochondria at the cross roads of Ca(2+) regulation. Cell Calcium. 35 (6): 509-521. 14 years ago (164), few details are known of the mitochondrial role in V S M Ca signaling. Aspects 7+ like the mitochondrial contribution to cytosolic Ca clearance or the time courses of mitochondrial Ca transients are still unclear. At the same time, several studies clearly illustrate the importance of mitochondria in V S M C a 2 + signaling (139;217;231) and ROS production (154). Our current understanding of mitochondrial C a 2 + signaling in V S M is not only limited by the scarcity of studies employing organelle-specific Ca probes, but also by the potential for vascular heterogeneity to confound extrapolation of findings between different blood vessels. Furthermore, hypotheses based on data from experiments in V S M prior to recognition of the 7+ central role of cytosolic microdomains in mitochondrial Ca signaling have largely gone unchallenged. The few available studies using organelle-specific calcium probes report mitochondrial Ca2+-signals consistent with those reported in other cell types and underscore the need to define the detailed mitochondrial interactions with SR and P M transporters in V S M (160;220). Figure 1.5. Mitochondrial 3-way regulation of C a 2 + signaling. ( 1 ) Buffering of Ca2' microdomains. Mitochondria buffer Ca 2 + entry and release to modulate the local [Ca2+] near Ca2+-sensitive PM and SR ion channels and to help refill the SR. ( 2 ) Relaying P02 to Ca2+ signaling. Hypoxia or proximal electron transport chain inhibition influence mitochondria-derived H 2 0 2 production. 02" and H 2 0 2 act on K v , phospholipase C and SERCA in different blood vessels. ( 3 ) Relaying metabolic perturbations to Ca2+ signaling. Perturbations in oxidative respiration and metabolite recycling affects the cytosolic NADH:NAD ratio, which can inhibit cADPr hydrolysis and modulate RyR and B K Q , gating properties. Plus signs denote increased activity. Minus sign denotes reduced activity.Abbreviations: ; I - IV, electron transport chains complexes I - IV; ADPr, ADP ribose; BKca, large-conductance Ca2+-activated K+-channel; cADPr, cyclic ADP ribose; Cl C a , Ca2+-activated Cl"-channel; K v , voltage-gated K+-channel; P 0 2 , oxygen tension; SOD, superoxide dismutase. A word of caution is in order before delving into the available data. I have discussed the available data on smooth muscle mitochondria in the context of advances made in other cell types and with an emphasis on how "mitochondrial inhibition" can affect C a 2 + signaling 15 pathways in at least three different ways. Much of the evidence from V S M describes effects of "mitochondrial inhibition" on cytosolic [Ca 2 +] ([Ca2+]j) responses, and substantial confusion stems from the term "mitochondrial inhibitor," which describes a diverse group of drugs acting by very different mechanisms (see table 1). We begin with a brief review of Ca -related mitochondrial ion transporters before considering the immediate effects of "mitochondrial inhibition" that include, but are not limited to, altered C a 2 + buffering, ATP production and ROS generation. Table 1.2. Abridged list of mitochondrial inhibitors Class Drug (EC50) Comments Ref Protonophores FCCP (1 uM) CCCP (1 uM) DNP (3 mM) depolarize PM at ~10uM (73) (127) (205) Complex 1 inhibitors Rotenone (1 nM) capsaicin DPI (23 nmol/mg) depolarize A i | / m , alter radical production inhibits PM-bound oxidases (125) (59) (182) Complex II Inhibitors Carboxin TTFA (0.6 uM) antimycin A depolarizes Au/m, alters radical production (184) (265) (45) Complex III inhibitors HQNO (37 U M)) Myxothiazol Stigmatellin antimycin A (3 pM) benzoyl peroxide complex lllb/c(1) - Qi site complex lllb/c(1) - Qo site complex lllb/c(1) - Qo site complex lllb(562) (189) (189) (189) (104) (122) Cytochrome c DCCDOOuM) (239) Complex IV Cyanide (46) azide (10 nM) (242) F 0 subunit of ATPase Oligomycin A & B (3uM) Apoptolidin (5 uM) (239) F i subunit of ATPase Aurovertin (25uM) (200) Uniporter blockers ruthenium red (~0.2 nM) inhibits RyR at higher , concentration (260) . RU360 (0.2 nM) (150) RaM blockers ruthenium red (212) mNCX inhibitor CGP-37157 (0.4 uM) selective over PM NCX (50;111) mitoKATP activator Diazoxide (0.4 uM) Cromakalim (1 uM) selective over PM K A TP (81) (81) m i t 0 K A T P inhibitor Glibenclamide (60 nM) 5-hydroxydecanoate non-selective for m i t oK A TP (17) (111) Abbreviations: CCCP, carbonyl cyanide m-chlorophenylhydrazone; CN, cyanide; DCCD, dicyclohexyl carboniimide; DNP, dinitrophenol; DPI, diphenyleneiodonium; FCCP, carbonyl cyanide 4-trifluoromethoxyphenylhydrazone; HQNO, 2-heptyl-4-hydroxyquinoline-N-oxide; N 3 , azide; TTFA, theonyltrifluoroacetate; 1.3.1 Pertinent mitochondrial ion transporters C a 2 + uptake mechanisms. The mitochondrial C a 2 + uniporter (mUni) is activated by C a 2 + concentrations > 3uM ( K m ~ lOuM) in the inter membrane space (IMS) (55;94;179). Such concentrations are thought to result from C a 2 + release or influx into cytosolic 16 microdomains at SR-MT and PM-MT junctions, which exhibit restricted diffusion (193). Uniporter activation mediates Ca2+-uptake by virtue of the negative membrane potential across the inner mitochondrial membrane (Av|/m), and the mUni appears to possess a Ca -calmodulin-dependant mechanism that sustains some degree of Ca2+-uptake after [Ca2+]rMs has fallen below the threshold for activation indicated above (56;65). This is referred to as mitochondrial memory, plasticity or facilitation. Mitochondria experimentally loaded with high [Ca 2 +] are partially depolarized, in which case activation of the uniporter with a pulse of appropriate extra-mitochondrial [Ca 2 +]i will permit C a 2 + diffusion out of the mitochondria through the uniporter (155). Ruthenium red is the prototypical mUni blocker (Kj -0.2 u,M) (260), but it is poorly membrane permeable, can inhibit the ryanodine receptor (Kj ~0.5uM, block at ~10uM) and binds to numerous Ca2 +-binding proteins such as calsequestrin (Kd 0.7uM) (44;158;202). The ruthenium red derivative, RU360, appears to be specific for the mUni (Kj ~10nM), but it is also poorly membrane permeable and rapidly oxidized (264;267). The Rapid Mode of Calcium Uptake (RaM) is a Ca2+-uptake mechanism demonstrating a high initial uptake velocity, and its steady-state uptake is reduced at [Ca2+]j above 150nM and completely inhibited at [Ca2+]j > 180nM (93). This likely limits the RaM to a role in resting V S M . RaM is also inhibited by a [ruthenium red], but at ~ 10-fold greater concentration than required to inhibit the uniporter, which thus also blocks RyR. Currently, no direct investigations of RaM activity have been reported in SMC. C a 2 + extrusion & release mechanisms. Differences in the kinetics of C a 2 + extrusion from mitochondria in intact cells versus isolated mitochondria are difficult to reconcile given the potential for artifacts associated with each technique and the fact that these transporters have not been isolated or reconstitute in lipid bilayers. Calcium extrusion from polarized mitochondria is thought to occur via two pathways, a Na+/Ca2+-exchanger (mNCX) and a Ff7Ca2+-exchange (mHCX) (22; 174). The contribution of each transporter to mitochondrial C a 2 + extrusion varies in different cells. Early studies of isolated SMC mitochondria suggested a lack mNCX activity (53), but recent evidence clearly shows Na+-dependant C a 2 + release in SMC mitochondria (228). Using CGP-37157, a selective mNCX inhibitor, we have found a modest contribution of mNCX to mitochondrial C a 2 + extrusion in cultured V S M cells (Figure 2.2). Both extrusion mechanisms have eluded molecular characterization, and mHCX has also eluded 17 pharmacological characterization perhaps because of the large number of H -translocating proteins on the inner mitochondrial membrane. The permeability transition pore (PTP) is generally thought to be activated under pathophysiological conditions, and here I view it as a Ca -releasing mechanism acting as an emergency release valve at high [Ca2+]MT (174). It is important to note that Ca2+-release upon pore opening requires mitochondrial depolarization, which has been observed in stimulated V S M C (141). In cells other than smooth muscle, PTP-mediated Ca2+-release can stimulate ER Ca2+-uptake (36). The molecular nature, regulation and pharmacology of the PTP are a complicated matter more comprehensively reviewed in (174). Outer mitochondrial membrane. The outer mitochondrial membrane (OMM) is sometimes viewed as being freely permeable to Ca 2 + , but this permeability is mediated via porins, also known as voltage-dependant anion channels (VDAC). Mitochondrial Ca2+-uptake is thought to occur at junctions with the SR/ER and P M , and V D A C preferentially localizes to regions of the O M M closely apposing (<100nm) the ER such that the V D A C expression level limits mitochondrial Ca2+-uptake (174). Numerous pro- and anti-apoptotic proteins, such as members of the Bcl-2 family, also appear to regulate mitochondrial and ER Ca transport; however, these proteins will not be discussed due to a lack of information specific to V S M . For general information we suggest the following reference (60). 1.3.2 Interpreting mitochondrial inhibition The term "mitochondrial inhibitor" is often used to describe any drug that partially or completely collapses the A\ | / m , thereby preventing mitochondrial Ca2+-uptake. This term is applied to protonophores and inhibitors of the electron transport chain (ETC). The resulting At|/m depolarization can cause FiFn-ATPase reversal whereby the hydrolysis of ATP mediates proton pumping to sustain a moderate Av|/m . Accordingly, an Fn-proton channel blocker, oligomycin, is often used to prevent ATPase reversal. ETC complexes influence the oxidation state of metabolite redox pairs and contribute to cellular ROS production; consequently, interpretation of the effects of their inhibition on mitochondrial Ca2+-buffering and [Ca 2 +]j signaling warrants caution and suitable controls to account for altered radical production, ATP supply, SR refilling and SR release. Moreover, since mitochondria are thought to contain the oxygen-sensor mediating hypoxic pulmonary vasoconstriction, the majority of studies on mitochondrial radical 18 production in SMC are based in pulmonary vessels in which regulation of ROS differs from systemic vessels (8;154;251;252). Protonophores & Oligomycin. Protonophores like CCCP and FCCP are the most common agents used to depolarize mitochondria in attempts to prevent mitochondrial Ca -uptake. At appropriate concentrations these agents are selective for mitochondrial versus plasmalemmal depolarization (127). Again, protonophore-mediated Av|/m collapse commonly reverses FiFn-ATPase, which can alter local [ATP], ATP:ADP ratio, local pH, and sustain some degree of Ca2+-uptake by maintaining a moderate A i | / m via ATP hydrolysis. Oligomycin prevents these effects, and alone is an effective tool to assess the influence of mitochondrial ATP production on C a 2 + signaling mechanisms. Protonophores are sometimes used to "release" mitochondrial Ca 2 + , but release per se is unlikely without some kind of stimulation to open the uniporter or PTP. In smooth muscle it is unclear whether protonophores selectively inhibit mitochondrial Ca2+-buffering or i f the loss of oxidative ATP production also affects SERCA activity. On the one hand, ionic homeostasis in V S M is generally accepted to be sustained by glycolytic ATP production (100;213), and glycolysis has been shown to support sustained vascular contraction (130). Thus glycolysis should also be able to sustain SERCA activity given that SERCA has a higher affinity for ATP than does myosin ATPase (Ka ~2 p M vs. -50 uM) (65). On the other hand localized [ATP] has not yet been directly recorded, so indirect evidence is used to argue that A\ | / m collapse does not deprive SERCA of ATP. In electrophysiological experiments the typical pipette [ATP] of ~2-5mM is assumed to support SERCA activity (91), which is a reasonable assumption given that maximal SR-filling rate requires only 200uM ATP in permeabilized BHK-21 cells (129). Dialysis with millimolar [ATP] might abolish localized regulation of SERCA by [ATP] microdomains (129), in which case the effects of oligomycin 9+ reported in smooth muscle would be consistent with preventing modest mitochondrial Ca -buffering through inhibition of FiFo-ATPase reversal rather than oligomycin A depriving SERCA of ATP (121). Oligomycin additively inhibited C a 2 + clearance on top of FCCP in 94-pulmonary artery SMC (121) and increased Ca clearance in femoral artery in a manner consistent with Fn inhibition causing mitochondrial hyperpolarization (120). While these observations cannot exclude the involvement of mitochondria locally supplying ATP to SERCA, they do indicate that bulk SERCA activity does not explicitly rely on oxidative phosphorylation. 19 The appropriate application of RU360 could greatly simplify this type of experiment and clarify to what extent A i | / m collapse affects processes other than mitochondrial Ca -uptake. ETC inhibition alters ROS production, redox state and A \ | / m . The electron transport chain is central to mitochondrial function. It creates the pH gradient that supports the A i | / m as a result of complexes I, III and IV pumping protons across the inner mitochondrial membrane (IMM), and it produces oxygen radicals and recycles metabolites essential to aerobic respiration. Inhibition of any one of these complexes can depolarize mitochondria to some extent, while protonophores depolarize Av|/m without preventing electron transport (48;141;154). The ETC is sometimes divided into the proximal and distal ETC such that complexes I and II are proximal and supply complex III with reduced ubiquinol. Complex I recycles matrix N A D H and may be an important regulator of the PTP (75), while complex II recycles glycolytic N A D H via the malate-aspartate and glycerol-3-phosphate shuttles. Thus complex II inhibition, perhaps more so than complex I inhibition, could indirectly induce anaerobic respiration by increasing the cytosolic N A D H : N A D ratio (8; 19). N A D H additionally inhibits cADPr-hydrolase thus increasing [cADPr] and promoting SR Ca2+-release (257). None the less, complex I inhibition with rotenone prevented refilling of UVreleasable C a 2 + stores in cultured V S M C (231). While the underlying mechanism in this case is unclear, it illustrates the need to assess the tissue-specific and inhibitor-specific effects of mitochondrial inhibition in a given preparation. For instance, rotenone and antimycin A increased hydrogen peroxide (H2O2) levels in renal artery SMC resulting in voltage-gated K+-channel (K v ) activation while producing exactly the opposite effect in pulmonary artery SMC (154). In this case the differences are associated with differential expression levels of the ETC complexes. At the same time, complex IIIb/c and IV inhibition with myxothiazol and cyanide did not alter radical production in pulmonary and renal artery SMC (154). Thus the release of internal C a 2 + stores by cyanide in mesenteric artery could illustrate differential control of C a 2 + and ROS signaling by specific ETC complexes (251) (but see (65)). Clearly, "mitochondrial inhibition" refers to complex interventions with multiple effects that should be considered when investigating the role of mitochondria in smooth muscle C a 2 + signaling. 1.3.3 Mitochondria buffer C a 2 + influx & release Mitochondria can both accelerate and retard clearance of [Ca 2 +]i elevations depending on the cell type being studied (95;170;217). In smooth muscle, mitochondrial depolarization 20 consistently reduces the rate of cytosolic Ca clearance following Ca influx across the P M or release from internal stores (65;66;91;118;120;151). These altered kinetics are interpreted as the loss of mitochondrial Ca2+-buffering, which might occur at [Ca 2 + ] j as low as 300nM (120). In one case, ruthenium red (50-200 uM) mimicked the effect of FCCP on clearance of voltage-gated Ca 2 +-influx, strongly indicating that Av|/m collapse inhibited mitochondrial buffering (65). Increased [Ca2+]MThas also been directly observed in response to SR C a 2 + release in vascular and nonvascular SMC with indications that the pattern of mitochondrial stimulation could be agonist-specific (65;95;151;160;220). Subtle differences amongst these observations have lead to discordant views on the details of uptake into and efflux from V S M mitochondria and whether or not mitochondria modulate the resting [Ca 2 +], set-point, but there is little doubt that mitochondria do buffer C a 2 + i n V S M . Uniporter vs. R a M activation. Most of the evidence that uniporter activation depends on high [Ca 2 +] in PM-MT and ER/SR-MT junctions comes from non-smooth muscle cells (10;55;156;193). Electron micrography has revealed extensive, close association of mitochondria with SR in smooth muscle (164), and mitochondrial Ca2+-uptake in V S M appears to be dependant on localized [Ca 2 +] elevations in junctional microdomains. Mitochondrial-targeted aequorin in permeabilized A10 SMC was largely insensitive to low concentration (< 1.8uM) Ca2+-buffers i f SR release was prevented (160). This illustrates that calcium-induced Ca2+-release, not the global elevation of [Ca 2 + ] j expected during stimulation, activated mitochondrial Ca2+-uptake (160). In cultured aortic SMC, vasopressin and ATP stimulated similar increases in bulk [Ca 2 + ] j , while ATP produced a much greater response in [ C a 2 + ] M T , and in both cases the stimulated [Ca 2 + ] i plateau was not reflected in [ C a 2 + ]MT (160;220). These finding are consistent with activation of the mUni by focal elevations of [Ca 2 +] above average [Ca 2 + ] j . They also suggest that the uniporter inactivates in, and is insensitive to, moderate global elevations of [Ca 2 + ] j . As such, changes in average [Ca 2 + ] j do not necessarily predict changes in [Ca 2 + JMT, and measurement of bulk [Ca 2 + ] j does not detect the occurrence of elevated cytosolic [Ca 2 +] microdomains. Mitochondria mediate rapid clearance of high [Ca 2 + ] j as would be expected for activation of uniporters in close proximity to C a 2 + sources in the P M and SR (65; 120), but it is less clear how mitochondria-mediated C a 2 + clearance can also occur at [Ca2 +]i < 300 n M (120). This could be explained in several ways: 1) the C a 2 + affinity of mUni is significantly lower in situ, 2) the initial evoked [Ca 2 + ] j elevation activates a Ca 2 + -CaM dependent mechanism that sustains uniporter Ca2+-permeability at low [Ca 2 +] (56; 120), 3) the RaM is 21 mediating mitochondrial uptake at low [Ca or 4) the uniporter is activated by localized Ca elevations (such as C a 2 + sparks) that are not reflected in bulk [Ca2 +]j. The insensitivity of mitochondrial uptake to low microM C a 2 + buffers argues against a reduced for the mUni in situ (160), while the return of [Ca 2 + ]MT to baseline levels during sustained [Ca2 +]j elevation argues against RaM activation as considered below (120;220). This favors the last explanation (4). Mitochondria have been shown to take up C a 2 + in response to near-bye C a 2 + sparks (168), but these events were visualized by sub-cellular confocal analysis of [Ca2 +]j and [Ca 2 + ] M T at sub-second temporal resolution, which has not been performed in V S M . Mitochondrial Ca2+-cycling at rest & [Ca2+]j set-point. In most vascular smooth muscle protonophores (FCCP/CCCP), cyanide and oligomycin or rotenone increase resting [Ca 2 +] (95;220;231;251), as has also been reported in bladder SMC (127). However, in Bufo marinus stomach smooth muscle cyanide does not affect basal [Ca 2 +]i (65). Thus it appears that V S M C mitochondria help sustain low [Ca 2 +]i by buffering or regulating Ca 2 +-influx and release (see sections 5 and 6). CGP-37157 and PTP inhibition with cyclosporin A elevated resting [Ca 2 + ]MT in V S M C mitochondria indicating basal Ca2+-uptake and therefore Ca 2 +-cycling in resting V S M C (Figure 1.6). Similar observations have also been observed in HeLa cells (10). Direct observation of mitochondrial Ca 2 +-influx at very low [Ca2 +]j implies that either: 1) leak of PM/SR C a 2 + creates few, but significant, C a 2 + microdomains that permit basal Ca2+-uptake, or 2) that RaM is important for buffering of these basal fluxes. RaM is not inhibited by a [Ca 2 +] of 100 nM, so a physiological role for RaM in cells with resting [Ca ]j near 100 nM is possible. On the other hand, focal C a 2 + elevations at resting [Ca 2 +]i, termed "marks," occur in response to RyR-mediated C a 2 + sparks in cardiac myotubes (168). However, the increased resting [Ca 2 + ]MT upon inhibition of the mNCX and PTP does not prove or disprove either mechanism. In addition, removal of extracellular C a 2 + decreases resting [Ca 2 + ]MT, but this too could result from either reduced SR filling and C a 2 + spark frequency or reduced basal [Ca2 +]j (176). Resolution of this issue requires the use of RU360 to specifically inhibit the uniporter. 22 Figure 1.6. Inhibition of mitochondrial Ca """-extrusion in SMC. (top left ) In cultured aortic SMC expressing mito-aequorin, inhibition of the mitochondrial NCX with CGP-37157 (lOuM) reveals mitochondrial Ca 2 + cycling at rest, (top middle) The transient increase in resting [Ca 2 +]M T upon inhibition of the permeability transition pore with cyclosporin A (CsA, lOuM) is returned to resting levels by the mitoNCX (dotted trace), (top right) Removal of extracellular Ca 2 + also reduced resting [Ca 2 +]M T. (bottom left) CGP-37157 impaired extrusion of Ca 2 + from mitochondria in cells stimulated with ATP (ImM) as revealed by the increased peak [Ca 2 +]M T and a reduced rate of [Ca 2 +]MT recovery, (bottom right) CsA increased peak stimulated [Ca 2 +]MT without prolonging recovery, indicating that the PTP mediates mitochondrial Ca2+-release but not extrusion. Kinetics of mitochondrial versus cytosolic C a 2 + elevations. Two types of mitochondrial Ca responses have been observed with respect to their recovery kinetics in smooth muscle. In pulmonary artery smooth muscle cells Rhod-2, a cationic dye with good mitochondrial selectivity, reported [Ca ]MT transients that decayed slowly over minutes, long outlasting the cytosolic C a 2 + transients (66), whereas in colonic SMC Rhod-2 reported [Ca 2 + ] M T transients decaying with a half-time of 47 seconds (151). Only in the latter study cytosolic Rhod-2 dialyzed from the cell to reduce the likelihood of cytosolic artifacts. In aortic SMC loaded with 9 + "74-Mag-fura-2, RyR- and IP3R-mediated Ca release produced [Ca ] M T elevations that outlasted the cytosolic response by minutes. This time course of decay was suggested to match SR refilling (95). But in cultured aortic SMC, aequorin targeted to the mitochondrial matrix reported transient [Ca 2 + ] M T elevations in response to SR C a 2 + release that decayed with half-times of less than one-minute (160;220). To determine which trend more accurately describes the physiological response we must consider the characteristics of both the C a 2 + sensors as well as the experimental methodologies. Without sub-cellular analysis, non-specific localization of fluorescent dyes can confound interpretation. Dispersion of Rhod-2 from depolarized mitochondria is not uncommon, nor is its 23 nucleolar accumulation. Mag-fura-2 loads into both the mitochondria and SR, and reciprocal [Ca 2 +] changes within these organelles make it difficult to interpret the kinetics of cell-averaged mag-fura-2 responses. In addition the ultraviolet and or intense laser light used to excite these dyes can produce free radicals that can inhibit proteins of the ETC and alter the activity of SR C a 2 + transporters (1). These considerations are discussed at length in a recent review (68). The above reports using fluorescent dyes were performed at room temperature, which can slow C a 2 + handling mechanisms (149). In contrast, the aequorin experiments were performed at 37°C and do not require illumination of the cells. Consumption of aequorin in subpopulations of mitochondria could undermine accurate reporting of [Ca 2 + ]MT kinetics. However, we have observed that the kinetics of the agonist-mediated mitochondrial-aequorin response very closely match those of the cytosolic [Ca 2 +] transient measured with fiira-2 up to the onset of the cytosolic plateau at which point the mitochondrial signal continues to fall (see chapter III). The close temporal relationship of the mitochondrial and cytosolic signal indicates that aequorin consumption did not significantly affect the observed mitochondrial kinetics. The similarity of the aequorin and dialyzed Rhod-2 decay half-times points to fast [Ca 2 + ]MT recovery as the more physiological response. In cultured cardiomyocytes, [Ca 2 + ]MT oscillations followed [Ca2 +]i oscillations with a period of ~1 s (196). If V S M mitochondria are also capable of such rapid C a 2 + oscillations this would support rapid cycling of Ca between the mitochondria and SR, allowing the mitochondria to enhance the apparent SR buffer capacity by facilitating SR refilling. 1.3.4 Mitochondria modulate SR C a 2 + release channel gating Mitochondria mediate multifaceted regulation of SR release channels by buffering local [Ca 2 +] and controlling local [ATP] and possibly via ROS signaling and cADPr metabolism. RyR and i p 3 r are sensitive to cADPr and I P 3 / A T P , respectively (25;208), and both channels have separate Ca2+-sensitive activating and inactivating sites. By influencing channel modulators and buffering localized [Ca 2 +]i, mitochondria can modulate both the Ca2+-sensitivity and the duration of opening of RyR and IP3R. In various cell types, including V S M , mitochondria affect the frequency and amplitude of sparks (91; 168), and regions of ER in close proximity to M T paradoxically exhibit increased rates of ER C a 2 + release and reduced ER [Ca 2 +] depletion upon IP3R stimulation (10;99). While this later point at first appears paradoxical, the fact that mitochondria facilitate ER refilling may help to explain how mitochondria can both enhance the rate and reduce the extent of ER C a 2 + depletion. While these phenomena are at least in part a consequence of diffusional restriction and linked Ca2+-transport at MT-SR junctions (99; 176), 24 few details are known about the specific interactions between mitochondria and SR release channels in SMC. Moreover, reports in SMC rely primarily on altering [Ca 2 +]i signaling with pharmacological tools without monitoring subcellular events with confocal microscopy. Thus, these findings are subject to the uncertainties associated with the mitochondrial inhibitors. Below we summarize the available SMC data with reference to recent advances made in other cell types. V S M C mitochondria appear to regulate C a 2 + release events from both IP3R and RyR that are important in the propagation of oscillating C a 2 + waves and the generation of C a 2 + sparks (254). In pulmonary and mesenteric artery SMC, moderate inhibition of complex IV increased both spark frequency and amplitude, which translated to increased Ca2+-induced Cl"-current (Ici(Ca)) (133;251). Increased spark frequency could indicate increased [Ca2 +]sR, increased [Ca 2 +] 7+ near the channel activating site, or Ca -sensitization of the RyR (254). Since mitochondrial inhibition is unlikely to increase SR filling (see section VI), it is most likely that moderate complex IV inhibition impedes local mitochondrial C a 2 + buffering or increases cytosolic [cADPr] by preventing oxidation of N A D H . While the latter explanation is somewhat unorthodox, it is consistent with the fact that further inhibition of complex TV, but not Av|/m collapse, increased bulk [Ca2+]j primarily by releasing SR C a 2 + (251). An ROS mechanism could also be involved, but in pulmonary and renal arteries cyanide (a complex TV inhibitor) did not alter radical production (154). Thus mitochondria might tonically suppress Ca sparks, as seen in H9c2 cardiac myotubes (168), which is consistent with a report that mitochondrial 7+ depolarization by activation of mito-KATP increased [Ca ]; in a manner consistent with RyR activation (261). Similarly, depolarization of mitochondria with FCCP slowed the time to peak of caffeine responses in pulmonary artery SMC, consistent with notion that mitochondrial Ca 2 + -buffering delays Ca2+-mediated RyR-inactivation (66). We propose that these seemingly paradoxical observations illustrate separate mitochondrial mechanisms to modulate cell-wide RyR Ca2+-sensitivity and site-specific Ca2+-mediated RyR inhibition. In addition, mitochondrial regulation of SR filling creates an indirect mechanism by which mitochondria can influence RyR activity. Regulation of I P 3 R by mitochondria has been reported in colonic, but not yet vascular, smooth muscle (151). CCCP caused progressive loss of IP3-mediated [Ca 2 +]; transients, and oligomycin did not alter the effect of CCCP. This indicates that loss of oxidative-ATP production did not account for the effect of CCCP despite suggestions of IP3R sensitivity to [ATP] (129). Gross depletion of SR C a 2 + stores was not observed, thus CCCP was assumed to 25 prevent mitochondrial Ca buffering near IP3R resulting in greater Ca -mediated inactivation of the release channels. As discussed below, several mitochondrial inhibitors also impaired adrenergicaly-driven oscillating [Ca ]i waves, which presumably require Ca release from IP3R 9 + (217). However, it was not directly demonstrated that the effects of the [Ca ]\ signals were directly due to mitochondrial modulation of IP3R gating. The above studies clearly illustrate that mitochondrial Ca -buffering is important for physiological RyR and IP3R mediated C a 2 + signaling, but many details remain to be described. 1.3.5 Mitochondrial modulation of PM channels Mitochondria modulate the activity of plasmalemmal ion channels such as C R A C channels, store-operated channels (SOCs), voltage-gated calcium channels (VGCC), Ca 2 + -activated K+-channels (Kc a) and Ca2+-activated CP-channels (Clc a) by buffering local C a 2 + gradients (106;146;147;156;226). Varying degrees of evidence support mitochondrial regulation of P M ion channels in V S M , which can indirectly affect SR refilling (and consequently SR release) and global C a 2 + signaling. Calcium-activated K+-channels and Clc a are essential to the regulation of vascular tone in many small blood vessels, and are primarily activated by C a 2 + sparks from SR elements in close proximity to the P M (254). In endothelial cells, Mali et al. elegantly showed that mitochondria situated in close proximity to large-conductance K+-channels (BKc a ) reduced channel activation by buffering C a 2 + sparks (146), and similar findings are observed in SMC. In pulmonary and coronary artery, mitochondrial depolarization increases B K c a open probability up to nine-fold by enhancing the diffusion of C a 2 + released by the SR toward B K c a (126;261). Moreover, stimulation of SR C a 2 + release in aortic SMC with caffeine can activate both K c a and Clc a , and the observation that FCCP alters the ratio of CI" versus K+-channels open at any demonstrates the complexity of mitochondrial regulation of the spatio-temporal profile of sub-plasmalemmal [Ca 2 +] gradients (95). While mitochondria acutely buffer the SR C a 2 + release or influx through V G G C that activates Kc a and Clc a (91), they also facilitate the refilling of the SR that is required to sustain the generation of sparks (section 7). In cultured aortic SMC, in which voltage-gated C a 2 + influx plays a minor role in contractile regulation, we observed only modest [ C a 2 + ] M T increases upon depolarization with 80mM K + and no effect of nifedipine on basal [ C a 2 + ] M T (unpublished observations- Poburko & van Breemen). Paradoxically, in pulmonary artery, mitochondrial inhibition with cyanide increased spark and stic frequency (251). These findings highlight the heterogeneity of mitochondrial regulation of [Ca2+]j throughout the vascular tree, 26 and caution against extrapolation of findings between functionally different vessels. However, we must also acknowledge mechanistic differences between protonophores and cyanide. Though speculative, current results are consistent with the proposition that cyanide alters [cADPr], while CCCP inhibits mitochondrial Ca2+-buffering. In several cell types, mitochondria promote store-operated C a 2 + entry (SOCE) by preventing Ca2+-mediated channel inactivation (106;147;169;226), but current evidence in V S M provides few conclusions. One study in V S M C , mitochondria buffered SOCE but did not directly affect the gating of SOC according to Mn2 +-quench in Ca2+-free solution (121). Here it is important to consider that SOCs may not be permeable to M n 2 + . In V S M , store-depletion can also cause Ca 2 +-influx via reversal of the plasmalemmal N C X secondary to Na+-entry through ROC/SOC (134), and studies in cultured SMC indicate a functional linkage between the mitochondria and N C X via regulation of SR-Ca 2 + extrusion (220). As discussed in chapter III, N C X reversal can directly stimulate mitochondrial Ca2+-uptake, which but the physiological relevance of this is not yet clear. 1.3.6 Mitochondria facilitate SR refilling In addition to promoting opening of SOCE pathways, mitochondria facilitate refilling of SR Ca2+-stores (10;79;106;147;169;226). Direct measurement of [ C a 2 + ] M T and [Ca 2 + ] E R in HeLa has shown mitochondrial C a 2 + extrusion via mNCX to be crucial for refilling of the ER (10; 147). At the same time, localized regulation of [ATP] or ATP:ADP ratio may also be important for SERCA activity (129). As mentioned previously, in permeabilized cells the maximal rate of E R -refilling is achieved at only 200uM ATP, and FCCP does not affect this process. In intact cells, however, FCCP or oligomycin can dramatically inhibit SR refilling (129). Half-maximal SERCA ATPase activity is reported to occur at 335 nm Ca 2 + , and SERCA exposed to resting [Ca2+];near lOOnM may operate at only 30% of maximal capacity (16). Thus, it is tempting to postulate that mNCX and SERCA co-localize at MT-SR junctions where elevated [Ca 2 +] would increase SERCA activity and the rate of SR-refilling. Furthermore, the plasmalemmal Na +/K +-ATPase is believed to utilize a compartmentalized ATP supply (100), thus it is conceivable that SERCA could be driven by ATP from closely juxtaposed mitochondria. In cerebral and femoral artery SMC, recovery from [Ca2+]j elevations occurs in three stages. FCCP abolishes the fast first stage and latent third stage, such that clearance becomes monotonic. Presumably, this monotinic clearance is mediated by the P M C A and N C X , as SERCA blockade does not affect the first or second phase. SERCA blockade does however 27 abolish the third stage, typically occurring at [Ca ]; <300nM (119;120). Susceptibility of the third stage to both FCCP and SERCA reflects an interrelationship between SERCA and mitochondrial Ca2+-uptake, whether by directed C a 2 + transfer toward SERCA or by Ca 2 + -stimulated mitochondrial ATP production. The former mechanism is supported by reports of CCCP abolishing sties in rabbit portal vein in a manner consistent with depletion of SR-Ca and abolition of C a 2 + sparks (92). An earlier study in cultured V S M C , which employed calcium green and mag-fura-2 to simultaneously measure [Ca2 +]j, [Ca 2 +]sR and [Ca 2 +]MT, also indicated that mitochondria facilitate SR-refilling. In this case, FCCP reduced the peak [Ca 2 +]i response to caffeine, retarded [Ca 2 + ]i clearance and appeared to prevent recovery of [ C a 2 + ]sR subsequent to stimulation (95). Together, these reports strongly support a mitochondrial role in SR refilling in vascular smooth muscle, and mechanisms described in other cell types set the stage for future studies employing confocal microscopy and agents like CGP-37157 to further elucidate the underlying mechanisms. 1.3.7 Mitochondria influence C a 2 + oscillations Oscillations of [Ca 2 +]i are an important signaling modality in V S M (133). Repetitive intracellular wave-like Ca2 +-oscillation are usually driven by ER/SR Ca2+-release, whereas P M C a 2 + influx is more important in repetitive spatially uniform transient rises in [Ca2 +]j. Cytosolic [Ca 2 +] oscillations result in mitochondrial [Ca 2 +] oscillations in V S M C (66), non-vascular SMC (78) and other cells (29;195;196;245), where the oscillations rely upon close ER-MT coupling. The [Ca 2 +]MT oscillations are also associated with temporally-coupled oscillations in [NADH] via stimulation of intra-mitochondrial dehydrogenases that are regulated by the frequency of C a 2 + oscillations (195). Thus, mitochondrial metabolism appears to depend on repetitive SR/ER Ca2 +release. Just as [Ca 2 +]i oscillations modulate mitochondria, mitochondria also modulate [Ca 2 +]i oscillations. In preparations like rat tail artery V S M C that exhibit asynchronous SR-mediated wave-like [Ca 2 +]i oscillations, mitochondrial inhibitors (rotenone, FCCP, ruthenium red, cyanide, antimycin-A, dinitrophenol) increased the frequency and decreased the amplitude of [Ca ]j oscillations (217;269). This modulatory role for mitochondria C a 2 + oscillations in vascular smooth muscle contrasts with observations in chromaffin cells, where mitochondria provide a C a 2 + buffer barrier that limits the spread of oscillatory Ca2+-waves (170). Recent evidence further suggests that differences in mitochondrial C a 2 + buffering underlie the differences in propagation of C a 2 + cytosolic C a 2 + waves in ventricular versus atrial cardiomyocytes (204). 28 Results obtained in hepatocytes and HeLa cells shed some light on why mitochondrial inhibition reduces the amplitude while increasing the frequency of Ca2+-oscillations in V S M C . In hepatocytes, mitochondria decrease the apparent ITVsensitivity of nearby IP3R, and regions of ER not in close contact with mitochondria exhibit reduced rates of ER-release and more extensive [Ca 2 +]ER depletion (10;40;99). At a constant [IP3]j the activation of the IP3R depends on the local [Ca2+]j generated by spontaneous SR Ca2+-release and loss of mitochondrial Ca 2 + -buffering increases [Ca2+]j during the troughs between oscillations. This reduces the time required to achieve the threshold [Ca 2 +] for IP3R activation and thus shortens the interval between the transients. At the same time Ca -release from IP3R more rapidly reaches inhibitory concentrations (IC50 -250 nm for type 1 IP3R) thus shortening the time to peak of the transients. The combined effect will be an increase in the frequency of C a 2 + oscillations. In addition, impaired mitochondrial refilling of the SR coupled with the reduced inter-wave period would result in relative depletion of the SR, which may contribute to the reduced Ca2 +-waye amplitude following mitochondrial inhibition. 1.3.8 Mitochondrial ROS production , O2 detection Reactive oxygen species are important signaling molecules in V S M acting via a multitude of mechanisms (143;186;253;258;259), and mitochondrial ROS are important physiological regulators of [Ca2+]j signaling (43;143;186;253). This is especially true in pulmonary artery SMC where mitochondria are believed to contain the oxygen sensor underlying hypoxic pulmonary vasoconstriction (reviewed in (8)). The effects of ROS on smooth muscle K+-channels and SR C a 2 + stores are discussed in recent reviews (143;186;253). The prominent radical produced by mitochondria is super oxide (O2), generated primarily at complex III and I (82;130;186;253)Superoxide dismutase converts O2" to H2O2, which can activate phospholipase A, phospholipase-C y, increase cADPr levels, release intracellular Ca stores and activate voltage-gated and Ca2+-activated K+-channels (86;154;211;257). Enhanced O2" formation can also increase conversion of mitochondrial NO into reactive peroxynitrites that affect C a 2 + handling and energy homeostasis (180). Rotenone and myxothiazol, but not C N , attenuated flow-induced radical production and dilation of coronary resistance vessels (139). Since coronary V S M C originate from different precursor cells than most V S M C , these vessels might exhibit C a 2 + signaling mechanisms unique from other vessels (144). For example, (246) in renal artery SMC, rotenone and antimycin A increased [H2O2] and stimulated dilatory Kv currents, while the opposite occurred in pulmonary artery. Myxothiazol and C N had no effect in either tissue (154), yet others report that C N 29 increases [Ca ] in both pulmonary and mesenteric arteries (251). Clearly, radical-mediated signaling qualitatively differs between systemic and pulmonary vessels. Inhibition of complex I, II and III tends to alter radical production, while complex IV inhibition causes effects more consistent with N A D H : N A D regulation. These data offer a glimpse at the integration of ROS and C a 2 + signaling mechanisms, and illustrates the need to characterize the specific details of these mechanisms in any given preparation. 1.3.9 Summary of mitochondria & C a 2 + signaling Mitochondria are integral to Ca2+-transport, ROS production and metabolite recycling in V S M , and these three systems are amalgamated in the orchestration of C a 2 + signaling that is central to physiologically regulation of blood flow. By buffering Ca2+-elevations in cytoplasmic microdomains mitochondria can modulate Ca2+-mediated activation and inactivation of ion channels underlying control of membrane potential in resistance vessels and oscillatory SR Ca 2 + -release in larger conduit vessels, while rapid Ca 2 +-cycling between mitochondria and the SR effectively permits the mitochondria to extend the SR-buffering capacity. At a more cell wide level, modulation of membrane potential and second messenger pathways by mitochondrial ROS permits the vascular response to regional oxygen supply to be superimposed on responses to paracrine and endocrine factors and physical factors such as pressure and flow. Moreover, oxygen and metabolic substrates co-regulate recycling of metabolic intermediates and the cellular redox potential through both mitochondrial and cytosolic mechanisms from which the consequent modulation of N A D H : N A D and ATP:ADP can alter second messenger metabolism 2+ and ion channel gating. These mechanisms place mitochondria at the cross-roads of V M S Ca signaling and reconcile otherwise discordant observations in the V S M literature. However, this remains a working model and many of the details remain to be investigated. 30 1.4 PURPOSE AND SPECIFIC AIMS To better understand C a 2 + homeostasis in vascular smooth muscle, I studied C a 2 + movements and protein localization in cultured rat aorta cells using a variety of techniques. I had 3 specific aims. > To determine the basic mechanism of the C a 2 + leak and to estimate the molar flux of this basal C a 2 + entry, I: • optimized conditions for the specific measurement of 4 5 C a 2 + uptake into resting cells. • screened a series of pharmacological agents for their effect on the basal rate of C a 2 + influx. . estimated the average volume of the cultured cells. This was accomplished by capturing images of cells expressing SNAP-25 targeted GFP (targeting the plasma membrane), deconvolving the image stacks, and reconstructing the cell volumes. . determined whether these particular cells expressed L-type voltage-gated channels, and canonical transient receptor potential (TRPC1-7) proteins. > To characterize the interaction of the plasmalemmal Na+/Ca2+-exchanger and mitochondria and to determine whether peripherally localized mitochondria contribute to the superficial buffer barrier in vascular smooth muscle cells, I: . transfected cells with aequorin targeted to the mitochondrial matrix (or inner leaflet of the PM) to directly measure mitochondrial (or sub-plasmalemmal) C a 2 + levels upon reversal of the Na+/Ca2+-exchanger. • performed parallel fura-2 experiments to measure changes in bulk [Ca ]j in the same smooth muscle cell line. > To dissect the relative contributions of inositol-1,4,5-trisphosphate receptors (EP3R) and ryanodine receptors (RyR) to agonist-mediated mitochondrial [Ca 2 +] elevations, and to determine whether IP3R and RyR stimulated separate populations of mitochondria, I: . measured the effect of pharmacological inhibition IP3R and RyR on the ATP-mediated and [ARG8]-vasopressin-mediated [Ca 2 +]MT elevations reported by aequorin targeted to the mitochondria. . determined whether two separate agonists consumed aequorin from individual or overlapping populations of mitochondria. . measured the effect of pharmacological inhibition IP3R and RyR on the ATP-mediated cytosolic [Ca 2 +]MT elevations reported by untargeted inverse-pericams. 31 . semi-quantitatively analyzed the spatial relationships of immuno-fluorescently labelled IP3R and RyR and GFP-labeled mitochondria using digital reconstruction of deconvolved images acquired by confocal microscopy. 32 CHAPTER II Basal Ca 2 + entry in vascular smooth muscle cells3 2.1 I N T R O D U C T I O N Described in smooth muscle 30 years ago, the phenomenon of basal calcium entry has received little attention especially in smooth muscle cells, where it plays a substantial role in resting calcium homeostasis and the maintenance of vascular tone (71;166;236). For example exposure of "resting" cells to C a 2 + free ambient solution causes a loss of sarcoplasmic reticulum C a 2 + content. This is due to leakage of C a 2 + from the sarcoplasmic reticulum (40), and sarcoplasmic reticulum C a 2 + can be restored upon replenishment of C a 2 + without the development of force (42;62). Conversely blockade of C a 2 + extrusion in the presence of extracellular C a 2 + leads to a net gain in cellular Ca 2 + . Thus it has long been clear that the inactive smooth muscle is not static with respect to C a 2 + metabolism, but supports continuous physiological cycling of C a 2 + between the intra and extracellular compartments. While it is generally accepted that efflux is mediated by the plasma membrane Ca 2 +-ATPase and the Na+/Ca2+-exchanger we have little knowledge to date regarding the components of the resting Ca 2* influx (166;240). The objective of this investigation is therefore to characterize the nature of Ca2+transport across the plasma membrane of non-stimulated smooth muscle cells. Three different techniques are commonly employed for the measurement of cellular C a 2 + fluxes: 1) imaging of fluorescent C a 2 + indicators, 2) patch clamp electrophysiology and 3) radioactive tracer analysis, each of which has several advantages and disadvantages. While fluorescent C a 2 + indicators have allowed us to investigate C a 2 + signalling at the subcellular level with high temporal and spatial resolution (198), this technique is not suited to measurement of resting C a 2 + influx because fluorescent measurements do not differentiate between intra and extracellular sources of Ca 2 + . This is a complicating issue since release of intracellular C a 2 + stores often activates capacitative calcium entry through store-operated channels. Electrophysiology is an invaluable tool that has provided great insight into the nature of ion channel behaviour. While the technique is capable of detecting tiny single channel currents of spontaneous channel activity, it is unable to detect electro-neutral transport as seen in some ion-exchange mechanisms. Additionally, the leak current associated with the patch seal is difficult to adequately distinguish from resting C a 2 + influx. The third method for measuring C a 2 + influx is to 3 A version of this chapter has been published as: Poburko D, Lhote P, Szado T, Behra T, Rahimian R, McManus B, van Breemen C, Ruegg UT. (2004) Basal calcium entry in vascular smooth muscle. European Journal of Pharmacology. 505 (1-3): 19-29. 33 label the calcium of the bathing solution with Ca and measure the cellular uptake of radioactive label. This method is theoretically the most direct, but is subject to two drawbacks. First, its temporal resolution is relatively poor and therefore sub-optimal for measurement of stimulated C a 2 + fluxes. Second, a background signal is generated by non-specific adsorption of 4 5 C a 2 + to sites within the extracellular space as previously described (71). Nevertheless we chose this third method as the most likely to provide direct information on the nature of basal Ca entry in monolayers of cultured smooth muscle cells, a preparation that lends itself well to removal of extracellular radioactive label. In addition the fact that net non-stimulated C a 2 + fluxes have a relatively slow time course reduces the requirement for high temporal resolution in these experiments. Given that pure lipid bilayers are not inherently leaky to inorganic ions, several mechanisms have been proposed to account for the resting C a 2 + permeability of physiological membranes (39;232). The first possibility is that imperfect junctions between phospholipid domains and membrane proteins are permeable to small ions. Second, pinocytosis may bring significant amounts of calcium into the cell, considering the ~ 10,000-fold difference between extracellular and intracellular C a 2 + concentration (89). Third, excitable Ca2+-permeable membrane channels may exhibit a degree of basal activity, or flickering, and thereby contribute to basal calcium influx (165). Our results provide experimental support for the latter hypothesis, and reveal the complexity and importance of the phenomenon of basal calcium entry. 2.2 METHODS 2.2.1 Cultures of smooth muscle cells. Rat aortic smooth muscle cells were previously prepared from aortae of male Wistar Kyoto rats (200-300g) as described elsewhere (140). Cells were cultured in Dulbecco's Modified Eagles Medium supplemented with essential and non-essential amino acids, vitamins, 0.001% ciproxin and 10% fetal calf serum, and kept at 37°C in a humidified atmosphere of 5% CO2 in air. Cells were seeded at 20,000 cells (between passages 6 and 11) per 16 mm diameter culture well and grown to confluence (ca. 400,000 cells per well) for 7 to 9 days.. 2.2.2 4 5 C a 2 + influx measurement. 4 5 C a 2 + uptake was recorded as previously described with minor modifications (140). After two washes in physiological salt solution (physiological saline solution, in mM: NaCl 145, KC1 5, M g C l 2 1, Hepes [4-(2-hydroxyethyl)piperazine-l-ethanesulfonic acid ] 5, glucose 10 and CaCb 1.2, pH 7.6), the cellular monolayers were pre-34 incubated at 37°C for 15 min (20 min when 2-APB (2-aminoethoxy-diphenylborate) was used) in 200 u.1 physiological saline solution containing the inhibitors. Twenty micro- liters of 4 5 Ca 2 + a t 0.02 mCi/ml were added to this solution and cells were incubated for 10 min at 37°C, or as specifically indicated. Influx was stopped by washing the cells 4 times at 0.5 min intervals with 0.5 ml of ice-cold 3 m M LaCL; or 0.2 mM E G T A (ethylene glycol-bis(2-aminoethylether)-N,N,N0,N0-tetraacetic acid) in physiological saline solution without calcium, and cells were" detached with 50 ul of physiological saline solution containing trypsin (0.25%), EDTA (ethylenediaminetetraacetic acid, 0.1%) and lysed with 250 ul of SDS (sodium dodecyl sulfate, 1%>). Radioactivity in the lysates was assessed by scintillation counting (Ultima Gold™, Packard, Groningen, N L , and L K B Wallac 1217 Rackbeta™, Turku, Finland). See Appendix I for conversion of cpm to moles of 4 5 C a 2 + . 2.2.3 Curve fitting: 1. Wash out kinetics: Extracellular tracer was removed by sequential washes of the cells at times denoted in Fig. 2.1. We subtracted the sum of cpm collected in washes previous to each point from the sum of cpm in all washes collected and in the final cell lysate. This treatment depicts the 4 5 C a 2 + activity present in the wells immediately prior to each wash. These values were fitted to a single exponential decay using GraphPad Prism 3.0 to determine the rate of tracer removal. 2. Uptake kinetics: An initial fast component was observed in the raw uptake data. The magnitude of this component was estimated by back extrapolation of the curve to time zero based on the assumption that the rate of uptake would be relatively constant during the first three minutes of influx. Linear regression of these data points (0.5 - 3.0 min) indicated a y-intercept of 96 ± 12 cpm/well. This rapid uptake occurred within the first 30 seconds of exposure and was subtracted from each data point. The corrected data set was fitted to a single exponential association using GraphPad Prism 3.0. 3. Curve Peeling: Visual inspection of the gadolinium (Gd 3 +) concentration-response relationship suggested that G d 3 + had a biphasic effect but overlapping concentration-response relationships. GraphPad Prism does not contain an equation to fit such a curve or to estimate the parameters of these two effects (pICso, Hil l slope, and magnitude) and to peel the two curves apart. We wrote an equation describing the sum of two sigmoid curves such that the top of the combined curve would be the sum of the two separate curves each with a bottom equal to zero and each with independent IC50S and Hil l slopes. The equation is shown below: 35 Y = 2*Bottom+ (TI-bottom) + (Top-T 1-bottom) ( 1 + 1 0 A ( ( L o g l C 5 o ^XJ'Hill Slopel)^ ( i + j 0 A ( ( L ° g ' c 5 0 2 -X)»Hill Slope2)^  Top-Tl is the top of the second curve. Using GraphPad Prism, the experimental data were fitted to this equation to estimate the individual curve parameters with the initial values set based on visual inspection of the raw data: Bottom = 0, TI = 70%, Top = 100%, Log(IC5o-i) = -6.0, Log(IC5n-2)= -3.0, and Hi l l slope(land2) = -1.0. The individual curves were then simulated from the best-fit parameters (dotted curves, Fig. 2.4). 2.2.4 Confocal microscopy and 3-D reconstruction: Cells were transfected with a plasmid encoding a green fluorescent protein (GFP) construct that is targeted to the inner leaf of the plasmalemma with a SNAP-25 pre-sequence (148). Using an Olympus BX50WI microscope fitted with an Ultraview Nipkow confocal disk (Perkin-Elmer, location), z-series of images were captured with a 60x water-dipping lens (numerical aperture 0.90) at 200 nm intervals (Prior HI28 motor drive). Image stacks were deconvolved using a Nipkow-optimized classic maximum likelihood estimation algorithm (Huygens, Scientific Volume Imaging, Netherlands). We then reconstructed the image stacks into 3-dimensional volumes with Imaris 3.3 (Bitplane, Zurich, Switzerland) and estimated cell volume with the Surpass function using a voxel size of 0.24 x 0.24 x 0.20 nm (Fig. 2.3b). 2.2.5 RNA extraction: Total cellular R N A from low- and high- confluent rat aortic smooth muscle cell lysates were extracted using a RNeasy mini kit™ (Quiagen) according to manufacturer's instructions. RNA was quantified by measuring absorbance spectrophotometrically at 260 nm, and its integrity was assessed after electrophoresis in non-denaturating 1% agarose gels stained with ethidium bromide. 2.2.6 Semi-quantitative RT-PCR: Reverse transcription of 5 pg total RNA was performed in 60 pi reaction volumes containing 200 units of Superscript II™ reverse transcriptase, 60 units RNase inhibitor, 3 m M MgCl 2 , l x Buffer II (Sigma) and 0.3 pg of random primers and 1 m M dNTP for 50 min at 42°C. Contaminating genomic D N A present in the R N A preparations was removed by digesting the reaction with 5 units of DNase I for 45 min at 37°C prior to the addition of reverse transcriptase. RT product (5 ul) was used in each 100 ul PCR 36 reaction. The PCR mixture contained 250 uM dNTP, 2 m M MgCh, l x volume of Buffer and 2.5 units Hotstar™ Taq polymerase, and 1 ul of forward and 1 ul of reverse primers. Amplification consisted of 40 cycles of 1 min at 94°C, 1 min at 55°C and 1 min at 72°C. The final extension was completed at 72°C for 7 min. Ten u.1 of 6x loading buffer (containing 0.25% bromothymol blue, 0.25% xylene cyanol FF, and 15% Ficoll type 400 (Pharmacia) in distilled water treated with DEPC (diethyl pyrocarbonate)) was added to the PCR products. Twenty pi of PCR products were then analyzed by electrophoresis on 2% agarose gels stained with ethidium bromide and gels were photographed under ultraviolet light. 18S ribosomal R N A expression was used as an internal control. The exemplary gels shown in this report (fig. 2.6) represent findings from a minimum of six different low confluency and four high confluency lysates. Rat brain mRNA was used as a positive control for the expression of TrpCl , 2, 3, 4, 5, 6, 7. Primers used for different amplifications were designed from published reports (152;247) or sequences available in Genebank (Table 2.1). RT-PCR (reverse transcriptase polymerase chain reaction) reactions run in the absence of reverse transcriptase or cDNA were used as negative controls (data not shown). Amplified PCR products from cell lysates were isolated from agarose gel, sequenced and found to be 100%) identical to the authentic sequences of rat TrpCl-7. 2.2.7 Data analysis: Results are presented as the means of at least three independent experiments with vertical bars indicating standard error (S.E.). I C 5 0 values were calculated by non-linear regression using GraphPad Prism 3.0 (GraphPad Software, San Diego, USA). Statistical evaluation was performed using one-way analysis of variance (ANOVA) followed by Bonferroni or Dunnett post-tests. Differences with a value of P < 0.05 were considered significant. 37 2.2.8 Materials: Unless specified otherwise drugs and chemicals were purchased from Sigma Aldrich, Switzerland. 2-APB was purchased from Fluka, Switzerland. Ultima Gold™ scintillation cocktail was purchased from Packard, Groningen, N L , and isotopic 4 5 C a 2 + was purchased from N E N Life Sciences Products, Geneva. Superscriptll™ reverse transcriptase, RNase inhibitor and random primers were obtained from Gibco/BRL, Canada. Buffer II (10x) was obtained from Sigma/Aldrich, Canada. MgCl 2 , dNTP, lOx volume PCR Buffer, Hotstar™ Taq polymerase and RNeasy mini kit™ were purchased from Qiagen, Canada. Primers for ribosomal R N A (18S) and RNAseZap were purchased from Ambion Inc., T X , USA. 2.3. RESULTS 2.3.1 Washout of extracellularly bound tracer Ca 2 + : Figure 2.1 illustrates the process of washing out the surface bound 4 5 C a 2 + from the culture wells as described in the methods. The 4 5 C a 2 + content of the cellular monolayer falls initially very rapidly when exposed to ice cold solutions containing either L a 3 + (3 mM) or EGTA (0.2 mM) and then stabilizes with a very slow decay. It appeared that the displacement of 4 5 C a 2 + by L a 3 + (k = 3.113 ± 0.077) was not appreciably faster than chelation by E G T A (k = 3.323 ± 0.112). However, it is known that high concentrations of L a 3 + block C a 2 + extrusion (238), and this is most probably the reason for the significantly higher plateau following L a 3 + washes (5.28 x 103 ± 0.12 x 103 cpm-well"1) than when cells were washed with EGTA (4.46 x 103 + 0.22 x 103 cpm-well"1). On the basis of these results we subsequently use two minutes of washing with ice-cold L a 3 + solution to remove 4 5 C a 2 + from the wells and outer cell surfaces, thereby permitting the determination of 4 5 C a 2 + uptake into the cells. Figure 2.1. Optimizing washout of excess tracer. At each time point cells were washed with 0.5 ml ice-cold physiological saline solution containing LaCl 3 (3 mM) or EGTA (0.2 mM). 4 5 C a 2 + activity in the well at each time point was determined by subtracting the sum of 4 5 C a 2 + activity collected in previous 1 ml washes from the total 4 5 C a 2 + collected in all washes plus the cell lysate. The to point indicates the total 4 5 C a 2 + activity loaded onto the cells Wash out is best fitted by single exponential decay. Wash out with LaCl 3 was described by k = 3.11 ± 0.08 min'1 and plateau = 5.28 ± 0.12 x 103 cpm-well"1. Wash out with EGTA was described by k = 3.32 ± 0.11 min'1 and plateau = 4.46 + 0.22 x 103 cpm-well"1. Given the equivalent rates of tracer removal using EGTA or LaCl 3, LaCl3 was chosen as the superior washing agent in that it resulted in a higher plateau than EGTA. 38 2.3.2 Rate of basal Ca influx: The SMC were exposed to Ca -labelled physiological saline solution (3.9 x 104 cpm-well"1 in 0.22 ml) for various times before washout of the surface bound label followed by scintillation counting of cell lysates. Close inspection of the curve revealed an initial very fast component of 96 ± 12 cpm-well"1 that was determined by.back extrapolation of the linear portion of the curve. This was followed by a mono-exponential component reaching a steady state of 1.92 x 103 ± 0.04 x 103 cpm-well"1 after about 60 min (see methods). The size of the initial fast component increased considerably when the time period of the cold lanthanum (La 3 +) incubation was decreased to 30 sec (data not shown). This strongly indicated that the fast phase was due to extracellular binding of tracer, so it was subtracted from the curve. The corrected uptake of 4 5 C a 2 + into the cells (Figure. 2.2) is best fitted to a single exponential process with a rate constant of 0.084 ± 0.005 cpm-min"1. This suggests that, at rest, permeation through the plasma membrane is rate limiting. For the purpose of obtaining highly reproducible measurements of resting C a 2 + influx in the presence of a variety of C a 2 + transport inhibitors we exposed cells to 4 5 C a 2 + for 10 min after pre-incubation in absence or presence of inhibitors before washing with cold L a 3 + solution. Cells were exposed to a [ 4 5 Ca 2 + ] t r a cer of ~ 0.12 nM in a 1.2 m M solution of 4 0 C a 2 + resulting in a 4 5 C a 2 + : 4 0 C a 2 + ratio of 9.7 x 106. The initial rate of tracer influx is approximated by the slope or first derivative of the exponential uptake curve at time 0 (Figure. 2.2). This is equal to Y m a x (plateau) multiplied by the rate constant (k) giving an instantaneous influx rate of 161 cpm-min"1, which was then converted to moles of 4 5Ca 2 +-labeled C a 2 + (see Appendix I). This conversion gives the instantaneous influx rate of 4 5 C a 2 + , and the rate of 4 0 C a 2 + influx was assumed to be proportional to the ratio of 4 0 C a 2 + : 4 5 C a 2 + in the tracer solution. We estimate the instantaneous influx of C a 2 + into the SMC to be 5.9 x 10 1 4 Ca2+-min" ' •well 1 . Each well contained an average of 4.0 x 105 cells, so cellular C a 2 + influx was approximately 1. 5 x 109 Ca^-cell^-min"1. The average volume of 14 cells from 3 individual experiments was calculated to be 9.6 ± 1.2 picolitres (volumes ranged from 3.6 to 15.8 picolitres) (Figure 2.3). Therefore, we estimate the molar minute influx of C a 2 + in our cultured smooth muscle cells to be on the order of 2.5 x 10"4 M-min"1, which corresponds to a whole cell current of approximately 7.5 pA. 39. 0 10 20 30 40 50 60 Influx duration (min) Figure 2.2. Exponential 4 5 C a 2 + uptake. Cells were exposed to 4 5 C a 2 + (0.6 uCi, 0.9 uM) in the presence of 1.2 mM 4 0 C a 2 + for the indicated times, and excess tracer was removed with four 30-second washes in ice-cold LaCl3 (3 mM). Linear back extrapolation of the first four points to t = 0 revealed a y-intercept of 96 ± 12 cpm-well"' indicating an initial fast component of tracer uptake that was interpreted as extracellular binding (see discussion). Following subtraction of this fast component, data were well described by a single exponential process with a plateau of 1.92 x 103 ± 0.04 x 103 cpm-well"1 and a rate constant of 0.084 ± .005 (fitted parameters ± S.E. of estimates). Data points represent the mean ± S.E. (n = 6). Figure 2.3. SNAP-GFP expression pattern and 3-D reconstruction of a representative cell. A.The peripheral localization of SNAP-GFP is evident in a single image in the x-y plane (bird's eye) of a single rat aortic smooth muscle cell amongst a confluent lawn. This localization of also revealed in slices through the z-x and z-y planes that were produced from deconvolved z-stacks using Imaris 3.3 software (Bitplane, Zurich, Switzerland). Scale bar equals 15 uM. B. These image stacks were then reconstructed into fractal-based volumes using the Surpas feature of Imaris to approximate cell volume (voxel size of 0.24(x) x 0.24(y) x 0.2(z) um), in this case 9 pi. 2.3.3 Inhibition by gadolinium ion: Lanthanides have long been known to inhibit membrane Ca transport (248), and it has been suggested that Gd can selectively inhibit SOC (229). We investigated the concentration response relationship of G d 3 + on inhibition of resting 40 Ca influx (Figure. 2.4). The semi-log plot of this relation yields two well-separated inhibitory processes with Hi l l slopes for both of about -1. The first process accounts for 65-70 % of total inhibition and has an IC50 of 1 pM. The second process accounts for 30 % of total inhibition with an IC50 of 2 mM. As illustrated below the larger more sensitive component is the result of inhibition of various Ca channels, which are also sensitive to more selective organic channel blockers. The smaller less sensitive component may be related to competition between G d 3 + and C a 2 + for negative binding sites on the plasma membrane (see discussion). Log[Gd3+] Figure 2.4. Complex concentration-dependent inhibition of 4 5 Ca 2 + uptake by gadolinium ion. Cells were incubated for 15 minutes in GdCl3 then exposed to 4 5 C a 2 + tracer (0.4 uCi) for 10 minutes followed by removal of excess tracer as described above. The concentration-effect relationship was closely fit using a custom programmed bi-sigmoidal equation in GraphPad Prism (see methods) where the initial values for iteration of the fit were set at: Hill slopesland2 = -1, pICso-i = 6 (visually estimated), pIC50.2 = 3 (visually estimated), Topi = 70, Top2 = 30, both bottoms = 0). Two overlapping sigmoids were simulated from the resulting parameters (left dotted curve: pIC5 0 = 6.0, Hill slope = -0.95, right dotted curve: pIC5 0 = -2.7, Hill slope = -0.92). 2.3.4 Non-stimulated Ca 2 + entry through excitable C a 2 + channels: We tested the hypothesis that part of the "resting" C a 2 + influx was due to entry through the same types of C a 2 + channels that support stimulated C a 2 + entry. In other words we hypothesized that voltage-gated C a 2 + channels, receptor-opeated channels and store-operated channels display a background activity when the cells are not stimulated either electrically or chemically. Figure 2.5 shows the inhibition of 4 5 C a 2 + uptake by maximally effective concentrations of nifedipine (10 uM; for L-type calcium channels), SKF 96365 (l-[b-[3-(4-Methoxyphenyl)propoxy]-4-methoxyphenethyl]-lH-imidazole) (50 p M ; for L-type and receptor-operated channels) and 2-APB (75 pM; selective, but not specific for inisitol-l,4,5-trisphosphate (IP3) receptor and some store-operated channels in addition to partial inhibitory effect on L-type channels). From these results (Figure 2.5) we deduced that distinct pathways contribute to resting C a 2 + influx. Thus 45% of the resting C a 2 + influx is mediated by voltage-gated C a 2 + channels, 41 which are completely blocked by nifedipine and SKF 96365 and are inhibited 50% by 75 uM 2-APB. 7% of the resting C a 2 + influx is mediated by C a 2 + channels exclusively blocked by SKF 96365 and, 23% of the resting C a 2 + influx is mediated by channels blocked exclusively by 2-APB. This conclusion is dependent on the use of optimally blocking concentrations of the various agents as demonstrated for SKF 96365 and 2-APB in the presence of nifedipine (10 pM) (Fig. 2.5B & C). Note that effective concentrations of 2-APB for inhibition of resting Ca influx are more consistent with those reported for inhibition of the IP3R (IC50 = 42 uM) in the SR than SOC in the P M (reported to be 0.5 pM) (215). We have previously shown 10 p M to be a maximally effective concentration of nifedipine (173). o E ' "o . . <t> — CL O = 1 o 0 C 5-100i 75 50 £ - 2 25 JO c B m ^ k. .So*; Q.-S = = 3 8 O - ^ 10Ch 75 50 25 0 « S i = S 8 « c 0 O - ^ 100 75 50 25 0 [SKF] (jiM) 1 T 1 s 75 100 150 10 30 [2-APB] (fiM) + SKF (50nM) Figure 2.5. Excitable calcium channels mediate calcium leak. A) Investigation of known organic calcium channel blockers. Cells were pre-incubated with nifedipine (nif, 10 uM) or SKF 96365 (50 uM) for 15 minutes or 2-APB (75 uM) for 20 minutes before 4 5 C a 2 + tracer (0.4 uCi) was added to cells for a 10-minute incubation period. Analysis of additivity: SKF 96365 inhibits Cav1.2 at this concentration, so Cav1.2 carries -45% of the leak influx and channels exclusively sensitive to SKF 96365 carry ~7% of the influx. Channels uniquely sensitive to 2-APB carry -23% of the resting Ca 2 + influx, (n = 6-15). (* = different from nif, ** = different from SKF 96365, *** = different from 2-APB ± SKF 96365. Determined by ANOVA (p < 0.05) and pair-wise post-hoc analysis) B) Concentration-response relationship for SKF 96365: This compound shows a very steep concentration-response relationship with a Hill slope -3.0, and a maximal effective concentration of 50 uM. (n = 3) C) Concentration-response relationship for 2-APB: The concentration-response relationship of 2-APB was determined in the presence of SKF 96365 (50 uM) to negate any cross-reactivity with SKF 96365-sensitive channels (n = 3). Error bars represent SE. 2.3.5 Comparison of organic Ca 2 + entry blockers with inorganic gadolinium: Due to the clear separation of the two inhibitory components of Gd , we used a concentration of 100 p M G d 3 + to completely block the more sensitive phase of 4 5 C a 2 + up-take, while leaving the less sensitive component essentially untouched (Figure 2.4). This permitted us to address the question 42 of whether there is a C a 2 + channel that is uniquely and potently inhibited by G d 3 + . Three independent experiments showed that a combination of 75 uM 2-APB and 50 uM SKF 96365 exerts the same degree of inhibition (79.6 ± 3.5%) as 100 uM Gd 3 +(83.7 ± 1.2%). Furthermore, the C a 2 + leak is not additively inhibited by organic and inorganic blockade (83.8 ± 4.0%) (Figure 2.6). This indicates that at concentrations below 100 uM, G d 3 + blocks all three types of C a 2 + channels: L-type C a 2 + channels, "receptor-operated channels" and "store-operated channels". Moreover, the first phase of influx blockade by G d 3 + occurs with a Hil l Slope of ~1, which is commonly interpreted to indicate a single binding site. In this case we take this to indicate a lack of selectivity by G d 3 + for any one of these channels. o — m c 2 3 £ O + £ o CM r - H -o •£ 3 b 100 75 504 25 0 4? 0 Figure 2.6. A combination of SKF 96365 and 2-APB fully blocks resting influx through membrane channels. 100 uM Gd 3 + is an approximation of the maximal effective concentration for the first phase of Gd3+-mediated inhibition of resting 45Ca2+-influx, with little effect on the second phase of inhibition. SKF 96365 and 2-APB were used at their maximally effective concentrations. There is no significant difference between inhibition of resting Ca 2 + entry by organic inhibitors alone or in combination with Gd 3 + (100 uM), or by Gd 3 + (100 uM) alone (ANOVA with Bon Ferroni pair-wise comparison, p = 0.07 ). Cells were pre-treated with inhibitors for 20 minutes before addition of 4 5 C a 2 + (0.4 uCi) for 10 minutes. Experiments represent the mean + standard error of 8 replicates. 2.3.6 Expression of candidate genes for channels responsible for non-stimulated Ca 2 + entry: The above experiments have helped to functionally characterize the channels contributing to the spontaneous background C a 2 + leak, but do not identify the molecular identities involved with the exception of the L-type channel (Cav1.2), by virtue of the high degree of selectivity of nifedipine. Accordingly, RT-PCR detected expression of the L-type channel-specific a-subunit (data not shown). To date the most likely candidate molecules for receptor-operated and store-operated channels in mammalian cells are the cannonical transient receptor potential channels (TRPC channels) (137;163). Confluent cultured rat aortic smooth muscle cells expressed TrpCl , 4 and 6 43 mRNA, while homogenized rat brain, used as a positive control, revealed expression of TrpCl-7 mRNA (Figure 2.7). 600 bp 500 bp 400 bp 300 bp 200 bp TrpCI TrpC2 TrpC3 TrpC4 TrpC5 TrpC6 TrpC7 18 S 600 bp 500 bp 400 bp 300 bp 200 bp TrpCI TrpC2 TrpC3 TrpC4 TrpC5 TrpC6 TrpC7 18 S Figure 2.7. TrpC mRNA expression profile in cultured rat aortic smc. Upper panel shows cDNA amplified by RT-PCR of total RNA isolated from confluent rat aortic smooth muscle cells at 7-9 days in culture. Lower panel shows a positive control for the probes using total RNA isolated from rat brain homogenate. All primers used amplified rat brain transcripts, (primer sequences in Table 2.1) Table 2.1. Oligonucleotide sequences of RT-PCR primers Channel (u'li Hank Accession No. Predicted Size, bp Sense/Antisense Location, nucleotides mTrpcl U40980 372 5'-CAAGATTTTGGGAAATTTCTGG-3' 1-22 5'-TTTATCCTCATGATTTGCTAT -3' 352-372 rTrpc2 AF136401 487 5'-CAGTTTCACCCGATTGGCGTAT-3' 1606-1627 5'-CTTTGGGGATGGCAGGATGTTA -3' 2071-2092 hTrpc3 U47050 331 5-ATTATGGTGTGGGTTCTTGG-3' 1483-1502 5'-GAGAAGCTGAGCACAACAGC -3' 1795-1814 mTrpc4 U50922 265 5'-CAAGGACAAGAGAAAGAAT-3' 5'-CCTGTTGACGAGTAATTTCT -3' 2535-2553 2781-2800 mTrpc5 AF029983 419 5'-CCTCGCTCATTGCCTTATCA-3' 675-694 5-TGGACAGCATAGGAAACAGG -3' 1075-1094 mTrpc6 U49069 410 5 '-CTGCT ACTC AAG AAGG A AA AC-31 738-758 5'-TTGCAGAAGTAATCATGAGGC -3' 1128-1148 mTrpc7 AF139923 260 5'-TG AC AGCC AATAGC ACCTTCA-3' 2397-2417 5'-GCAGGTGGTCTTTGTTCAGAT -3' 2637-2657 Trpc, transient receptor potential; m, mouse; r, rat; h, human. 2.4 D I S C U S S I O N We have analyzed basal C a 2 + entry in non-stimulated, "resting" cells and have identified four pharmacologically distinct components: 1. voltage gated channels, 2. receptor operated channels, 3. store operated channels and 4. a smaller undefined components sensitive only to millimolar Gd . This complements recent electrophysiological and molecular studies (159;166;240), and we illustrate here that basal C a 2 + entry in smooth muscle cells is mainly 44 mediated by a background open probability of electrically- and chemically-sensitive Ca 2+ permeable channels, which may be of similar molecular composition to "Ca leak channels" described in skeletal myotubes (105;240). A large component, 70-80% of the total C a 2 + influx, is carried by L-type C a 2 + channels, receptor-operated channels and store-operated channels in proportions that will most probably vary markedly with cell type, culture conditions and TRPC expression profile. It is important to note that the cells used in this study do not necessarily reflect the properties of vascular smooth muscle under physiological conditions. For example, the large basal entry through the L-type C a 2 + channels may indicate that these cells are partly depolarized after reaching confluency (123). Indeed preliminary experiments indicated that nifedipine inhibited considerably less influx (about 25% of total influx) in younger, non-confluent cells (data not shown). However, the most surprising finding was the large inhibition by 2-APB, which indicates the presence of open store-operated channels under resting conditions even though the sarcoplasmic reticulum contains ample stored C a 2 + (209;220). This finding is compatible with constant cycling of C a 2 + between the sarcoplasmic reticulum and the 9+ extracellular space independent of changes in bulk cytoplasmic [Ca ]; (234). This illustrates that the plasmalemmal C a 2 + leak is to some extent linked to the well-known sarcoplasmic reticulum C a 2 + leak (40). The TRP family of proteins are likely candidates for the molecular constituents forming receptor-operated and store-operated channels, and these proteins can form non-selective cation channels (163). Therefore, this may explain the resting activity of voltage-gated C a 2 + channels, since opening of non-selective cation channels will tend to partially depolarize the cells (263). 94- 94-In addition to the Ca leak carried by excitable Ca channels, a smaller component of 20-30% of basal 4 5 C a 2 + uptake is only inhibited by high concentrations of inorganic polyvalent cations such as L a 3 + and G d 3 + . At present we do not know the mechanism for this uptake nor its physiological significance, but the concentrations of L a 3 + or G d 3 + required to inhibit this uptake are far greater than the ICso's reported for inhibition of activated C a 2 + channels (227;248). Close inspection of the uptake kinetics reveals a small component of uptake with a Ua of approximately 7 seconds, which is faster than would be energetically favourable for channel permeation. In addition, the magnitude of this rapid component was dramatically reduced after enhancing the stringency of the 4 5 C a 2 + washout protocol (Fig. 2.1). This suggests the presence of a protected extracellular C a 2 + pool that is not readily displaced from its binding sites such as described Darby et al. (58). If extracellular binding sites are involved, the 15-minute G d 3 + pre-45 incubation used to create the Gd concentration-effect curve may effectively act as a blocking step, or alternatively may block internalization of the tracer. However, this conclusion remains speculative. It is particularly interesting that the C a 2 + channels contributing to basal C a 2 + entry that are sensitive to low micromolar concentrations of lanthanides appear to be the same channels already functionally identified in smooth muscle: voltage-gated, receptor-operated and store-operated channels. Other candidates for channels contributing to basal C a 2 + entry are stretch-activated channels and the "leak channels" recorded in skeletal muscle (105). Setoguchi et al. showed that stretch-activated channels are inhibited by Gd with an IC50 of 14 uM and are completely blocked by 100 uM Gd3 +(206). We cannot confirm or dispute a role for stretch-active channels in these cultured cells, but the lack of additivity of Gd on top of SKF 96365 and 2-APB would imply that i f stretch-activated channels are expressed and are contributing to the C a 2 + leak then they are also sensitive to these organic compounds. The skeletal muscle "leak channels," which may also be stretch-sensitive, are activated by nifedipine and inhibited by two unique dihydropyridines, AN-1043 (dimethyl 2,6-dimethyl-4-(4-bromophenyl)-l,4-dihydropyridine-3,5- dicarboxylate) and AN-406 (dimethyl 2,6-dimethyl-4-(4-trifluoromethylphenyl)-l,4-dihydropyridine-3,5-dicarboxylate) (5;105). However there is no evidence to date as to their sensitivity to SKF 96365, 2-APB or Gd 3 + . On the other hand Obejero-Paz et al. have electrophysiologically characterized two distinct "leak channels" in A7r5 cultured smooth muscle cells(166). One channel was divalent cation-selective (6pS), and the other was a relatively non-selective channel (17 pS), and these channels are inhibited by 50 uM Gd . Moreover, the 17 pS channel shares some electrophysiological characteristics with the skeletal muscle leak channels and the drosphila TRPC-L channel. Of the known excitable C a 2 + channels, current literature overwhelmingly points towards the TRPC channels as mediating the phenomena of receptor-activated and store-activated calcium influx. Generally, TRPC6 and 7 are thought to be activated by diacylglycerol (79; 117;243), while different reports show TRPC1, 3, 4 and 5 to be activated by both receptor-activation or store-depletion depending on the experimental conditions (137;163;167;175). This variable activation most likely reflects extensive species- and tissue-specific signaling, which makes it difficult to definitively label these channels as receptor-operated or store-operated channels (163). More importantly however, Gailly's group showed TRPC1 and 4 to be 46 constitutively active in A7r5 smooth muscle cells (240), and TRPC7 has been implicated as contributing to basal plasmalemmal permeability to divalent cations by virtue of it's activation by intracellular Mg-ATP and Mg-GTP (159). On the other hand Vanderbrouke et al. found that TRPC6 was not constitutively active, which is consistent with its specific roll as a receptor-operated channel. TRPC1 is linked to the type-2 IP3 receptor in some cells, which might confer sensitivity to the release of intracellular C a 2 + stores (145). Furthermore, it is important to note that heteromeric TRPC channels often exhibit radically different characteristics from those reported for the respective homomeric channels. (83). Thus since the cultured rat aortic smooth muscle cells in our study expressed TRPC1, 4 and 6, we propose that these cells express at least two forms of store-sensitive cation channel (TRPC1 and TRPC4) and at least one of type of channel that is selectively regulated by receptor activation (TRPC6). Although the pharmacology of TRPC channel modulation is still in its infancy, initial reports indicate that TRPC1 and 6 are inhibited by L a 3 + (2-50uM) and are also sensitive to SKF 96365, which incidentally has been shown to inhibit TRPC3 (35;268). Several reports have shown 2-APB to inhibit store-operated C a 2 + entry, presumably carried by TRPC channels (34). To be specific 2-APB can inhibit TRPC3 (229;230). In rabbit inferior vena cava, which expresses only TrpCl mRNA, SKF 96365, but not 2-APB, inhibited store refilling (135). Thus we hypothesize that the portion of the C a 2 + leak that is selectively blocked by 2-APB is carried 9 + by TRPC4. However, the potency that we observed for 2-APB's action on Ca leak inhibition is more consistent with inhibition of IP3 receptors than direct inhibition of a plasmalemmal channel (215). This implies: 1) that the IP3 receptor channels have a basal open probability, which could represent one pathway for the sarco/endoplasmic reticulum C a 2 + leak, and 2) that the resting C a 2 + leak from the sarcoplasmic reticulum is required and sufficient for the activation of TRPC4. This 9 + activity of 2-APB is in addition to partial inhibition of L-type Ca channels at the concentrations used, which as mentioned above may be an indirect effect. The fact that TRPC6 was shown to not be active in resting A7r5 smooth muscle cells (240), further indicates that 2-APB was selectively inhibiting TRPC4 in our cultured smooth muscle cells. By corollary, this would infer that the portion of the Ca leak that was specifically blocked by SKF 96365 may have been purely mediated by T R P C L This would be consistent with the report from Scarpa's group that only two single channel currents were active in the A7r5 cell C a 2 + leak (166). 47 Having characterized the nature of the calcium leak in the R A S M C , we must question the physiological significance of 2.4 femtomoles of C a 2 + permeating the cell membrane per minute. Is this influx sufficient to maintain intracellular calcium stores? In relation to cell volume, this corresponds to a turnover of 250 uM per minute, which at first seems excessive given a resting [Ca 2 +]i of-100 nM. Yet for comparison, Ganitkevich calculated a single episode of C a 2 + release 9+ 9+ from the endoplasmic reticulum to be 680 attomoles of Ca (80). Thus, resting Ca influx could provide sufficient C a 2 + to compensate for spontaneous C a 2 + sparks and puffs in resting cells, and in fact may be stimulated by these quantal release events from sarcoplasmic reticulum Ca On the other hand, this rate of resting calcium leak is equivalent to an inward current of -^7.5 pA, compared to a single open L-type channel earring 0.3 pA (84), and this magnitude of unstimulated influx is amenable to the concept of "leaky" or "flickering" excitable channels, being equivalent to 250 channels with a basal opening probability 0.1. However, it is important to consider how these findings in cultured cells relate to intact vascular smooth muscle. Given that removal of extracellular C a 2 + depletes intracellular C a 2 + stores (162), which implies that basal C a 2 + influx is required to maintain the resting C a 2 + concentration in the sarcoplasmic reticulum of intact vascular smooth muscle. The compounds used herein to inhibit the C a 2 + leak are also known to inhibit vascular contraction, thus it is of immediate interest to assess whether this is in part due to a loss of reticular C a 2 + stores in a manner analogous to removing 9+ extracellular Ca . 2.5 A P P E N D I X I: Converting tracer uptake to 4 0 C a 2 + influx. ^Ca2*]stock: Stock tracer solution was diluted 1000-fold, and the activity in six 20 u.1 alliquots was counted giving an average diluted activity of 3.90 x 104 cpm. Accordingly, the stock contained 1.95 x 109 cpm/ml. Given a half life of 163 days and 90% counting efficiency, the concentration of 4 5 C a 2 + ion required to give this activity is 7.31 x 10 1 4 ions/ml based on simple exponential decay. This is equivalent to 1.22 x 10"6 moles/liter, and gives a conversion factor of 3.76 x 105 4 5Ca 2 + /cpm. This values was then corrected to account for decay of the stock from the time when the uptake measurements were performed (30 days prior) to give a [ 4 5 Ca 2 + ] s t 0 ck of 1.37x 10" 6 M. ^Ca^Jon ceils and 40Ca2+-to-45Ca2* ratio : Upon addition of tracer to the experimental wells, the [ 4 5 Ca 2 + ] s t 0 C k had been diluted 11 000-fold giving a [ 4 5 Ca 2 + ] o n ceils of 1.24 x 10"10 M . The physiological saline solution used contained nominally 1.2mM C a 2 + giving a 4 0 Ca 2 + - to- 4 5 Ca 2 + ratio of 9.68 x 106. 48 Basal Ca2+ permeability: the instantaneous 4 5 C a 2 uptake rate of 161 cpm-min"1 was calculated as the first derivative of the influx kinetics curve at t = 0, which is equivalent to a rate constant multiplied b y the curve plateau (k = 0.084, y m a x ^ 1920). Converting cpm to 4 0 C a 2 + ions gives 5.86 x 10 1 4 4 0Ca 2 +min" 1-weH" 1. Given that each well contained 4.04 x 105cells on average, the basal cellular influx of C a 2 + was 1.45 x 109 Ca^-min"1. 49 C H A P T E R III Direct communication between peripheral mitochondria and the Na+/Ca2+-exchanger4 3.1 I N T R O D U C T I O N It has long been known that high salt intake may lead to hypertension. Demographic studies have shown that populations with limited dietary sodium chloride (NaCl) have lower blood pressures, while lowering one's salt intake to 75 mmoles/day can reduce systolic blood pressure by 5 mm Hg in ~50% of hypertensive patients (101). Blaustein and co-workers originally suggested that the vascular smooth muscle (VSM) Na /Ca -exchanger (NCX) provides the mechanistic basis for the above clinical correlation between blood pressure and [NaCl]piasma (27). This hypothesis is based on the fact that the plasmalemmal N a + gradient governs the rate and direction of the C a 2 + flux mediated by the N C X (27). While early mechanistic investigations failed to provide a consistent correlation between extracellular Na + concentration and cytoplasmic [Ca 2 +]j in V S M , it was later recognized that the N C X communicated more efficiently with the sarcoplasmic reticulum (SR) than with the myoplasm (3;28). The concept of preferential C a 2 + transport between the extra-cellular space and SR (or ER in non-muscle cells) was first proposed by van Breemen and co-workers (3;233;235) and is highly dependant on junctional complexes between the plasmalemma (PM) and the peripheral membranes of the superficial SR observed to cover about 10% of the inner P M surface (176;210). These junctional complexes consist of patches of P M and SR membranes separated by a narrow cytoplasmic space -20 nm in width and are often neighboured by mitochondria. Both the N C X and the ouabain-sensitive Na+/K+-ATPase-oc2 have been reported to co-localize with the superficial SR, which is enriched in the low affinity C a 2 + binding protein calsequestrin (116; 157). This coupling of proteins is thought to provide local regulation of the sub-plasmalemmal [Na+] that regulates the activity of the N C X and its interaction with the junctional SR (28), but the consequence or function of the mitochondria neighbouring these junctions has yet to be investigated. Traditionally the N C X has been studied in resting cells and during recovery from elevations of [Ca 2 +]i, at which time the N C X operates in the forward mode to unload the SR and to extrude C a 2 + from the cytoplasm (28; 161). Recently, several studies have revealed that the N C X can also operate in reverse mode during receptor mediated activation (11) to refill the SR 4 A version of this chapter has been submitted for publication in Cell Calcium as: Damon Poburko, Kathryn Potter, Edo van Breemen, Nicola Fameli, Olivier Basset, Urs Ruegg, Cornelis van Breemen. (2005) Mitochondria buffer NCX-mediated Ca2+-entry and limit its diffusion into vascular smooth muscle cells. 50 between recurrent waves of SR Ca release ((134) - also see (64). These oscillating Ca waves are also dramatically affected by mitochondrial C a 2 + transport (217), but the underlying mechanism has not been completely defined. The development of aequorin targeted to the M T matrix by Pozzan and co-workers has led to the discovery of linked C a 2 + transport between SR/ER and mitochondria (194). We recently reported that such linked transport between M T and the peripheral SR led to indirect 94-N C X control over mitochondrial Ca signaling (220). In this communication we employ aequorin targeted to either the M T or the P M to demonstrate a direct interaction between the N C X and the mitochondrial uniporter, which may be relevant to agonist-mediated C a 2 + signaling in vascular smooth muscle. 3.2 METHODS 3.2.1 Smooth muscle cell culture: Cells were cultured as described previously (220). A proprietary line of cultured rat aortic smooth muscle cells (RASMC), stored in 90% DMEM/10% DMSO in liquid nitrogen, were thawed and used between passages 8-12. Cells were incubated at 37°C in a humidified atmosphere of 5% C O 2 . 3.2.2 Expression of aequorins and green fluorescent proteins (GFP): RASMCs were transiently transfected with one of two pcDNAI expression vectors encoding apo-aequorin containing the amino terminal targeting sequences for human cytochrome oxidase subunit VIII (mito-aequorin) or SNAP-25, which localizes to the inner leaflet of the plasmalemma (PM-aequorin) (148;194). We also transfected cells with SNAP-targeted GFP (PM-GFP) to confirm the fidelity of the SNAP-25 localization sequence. R A S M C were seeded onto 13 mm Thermanox™ cover slips coated with Matrigel™ for 30 minutes at 37°C (Nunc, Life Technologies). After 1 day in culture, culture dishes were washed with Ca^ /Mg -free PBS and refilled with 500 pL of D M E M (10% FCS) before transfecting cells with Effectene® (Qiagen) as per manufacturer's instructions (1 pg D N A per 2.77 pL Effectene). Cells were used for experiments on the following day. 3.2.3 Immuno-labeling & imaging PM-targeted proteins: The localization PM-GFP was directly imaged in live cells, and the localization of PM-aequorin assessed in fixed cells by immunocytochemistry. Cells were fixed in 4% paraformaldehyde for 10 minutes in Tris-buffered solution (TBS). Excess fixative was quenched for 5 minutes in glycine (0.2 M) prior to 51 permeabilizing cells Triton X-100 (0.1%, 10 min.). Following a block-step (1% goat serum, 1% bovine serum albumin, 1 hr), cells were exposed to monoclonal anti-haemaglutinin (HA) antibody (1:400, Boehringer Ingelheim, 12CA5 clone) over-night at 4°C. Cells were washed in TBS 3 times and incubated with a goat anti-mouse F(ab2)' Alexa-488 antibody (1:100, Molecular Probes, Oregon) for 1 hour (23°C). Control cells were treated identically but were not exposed to primary antibody. Images were acquired on an upright Olympus BX50WI microscope with a 60x water-dipping objective (NA 0.9) and equipped with an Ultraview confocal imaging system (Perkin Elmer). Images were collected at 400 nm z-steps (Prior H-128 motor drive) and were reconstructed into 3D volumes using Imaris software (Bitplane, Switzerland). Immuno-fluorescence images were thresholded based on the intensity of non-specific binding in control images. 3.2.4 Measurement of mitochondrial & sub-plasmalemmal [Ca2+]: Mito-aequorin was reconstituted (coelenterazine 5 pM) in serum free D M E M for 2—4 hrs before experiments. SNAP-aequorin was reconstituted in serum free D M E M with 0.1 m M C a 2 + with coelenterazine (5 pM) for 2-4 hrs. Cells were superfiised at 1 mL/min with physiological salt solution (PSS, in mM: NaCl 145, KC1 5, M g C l 2 1, HEPES 5, glucose 10, and CaCl 2 1.2, pH 7.4). Luminescence was detected by photomultiplier tubes (PMTs) (EMI 9789 and P25232, Electron Tubes Inc, USA) and photon emission was recorded 1 Hz with EM6 photo-counting software (Electron Tubes Inc, USA). The EMI 9789 was coupled to an AD6 analog-digital converter and a CT-2 counting module (Electron Tubes Inc.). The P25232 is a self-contained counting-photon counting system. PMTs were cooled to 4-10°C to reduce dark counts. The off-line calibration of photon emission to [Ca 2 +] has been described in detail previously (148;220). 3.2.5 Measurement of cytosolic [Ca2+]: Cells (30,000 per cover slip) were seeded onto ethanolized and flame-sterilized 12 mm culture-coated glass cover slips (VRW Scientific) treated with Matrigel™. After 3 days in culture, cover slips were washed 6x with warm PSS and incubated with fura-2AM (5 pM, 0.05% DMSO, 0.05% pluronic acid F-127) for 45 min at room temperature (18-20°C). Cells were washed 6x with PSS (37°C) and allowed to equilibrate for 10 min at 33-35°C on the microscope stage. Ratiometric image capture and analysis are described elsewhere (221). Background fluorescence was measured in cell-free regions of interest and subtracted in a frame-wise manner. At the end of each experiment fura-2 fluorescence was 52 quenched with Mn (15 mM) in the presence of 20 u M ionomycin to determine auto-fluorescence values. In experiments with KB-R7943, fura-2 responses were expressed as the change in ratio from the pre-stimulatory baseline in order to compensate for a fluorescent artefact caused by absorption of U V light by KB-R7943 between 300-390 nm that produced a parallel shift in the absolute F340 and F380 values. 3.2.6 Statistical Analysis: Values are expressed as mean ± standard error (SE) with the number of replicates indicated for each experiment. Means were compared using the most robust test appropriate to each experimental design. Groups of three or more means were compared by A N O V A with pair-wise comparisons made by Bonferroni post-hoc tests. Data were compiled and analyzed using GraphPad Prism 4.0 in coordination with Microsoft Excel, and NCSS was used to perform statistical tests. 3.3 RESULTS 3.3.1 Reversal of NCX stimulates mitochondrial Ca2+-uptake: Acute removal of extracellular N a + reverses the plasmalemmal Na+-gradient causing transient reversal of the N C X and C a 2 + entry (3). Although removal of extracellular Na + creates an un-physiological condition, it provides an effective means to study local interactions between the plasmalemmal N C X , mitochondria and the SR. Iso-osmotic replacement of extracellular N a + with N-methyl-D-+ 2+ glucamine (NMDG), referred to as "ONa ", caused a rapid and transient increase maximal value of 2-4 u M (Figure 3.1 A), which declined to a plateau 1.5- to 2-fold above resting levels (~700 nM). The mitochondrial response to 0Na + was entirely dependent on extracellular C a 2 + (data not shown) indicating that the elevation of [ C a 2 + ]MT is due to C a 2 + influx rather than C a 2 + release from intracellular stores (220). The selective blocker of reverse-mode N C X , K B -R7943 (5-10 u.M), completely inhibited the transient phase of the mitochondrial response to 0Na + without affecting the plateau phase of the response (plateau: control - 662 ± 46 uM, n = 14; KB-R7943 - 761± 49 uM, n = 8, p=0.18, 2-sample t-test) (Figure 3.1A). This indicated that the [Ca 2 + ]MT plateau, in the absence or the presence of KB-R7943, was due to inhibition of forward mode Na+/Ca2+-exchange and a reduction in the extrusion of basal C a 2 + entry. 53 A B Time (min) Figure 3.1. Reversal of Na+/Ca2+-exchange stimulates mitochondrial CaJ+-uptake in cultured smooth muscle cells. A. Elevation of mitochondrial [Ca2+] upon external Na+ replacement with NMGD (black trace, n = 14). KB-R7943 (10 uM) inhibits the transient phase of this response (gray trace, n = 8). B. Selective inhibition of the mitochondrial Na+/Ca2+-exchanger with CGP-37157 (10-20 uM, black trace, n = 8) demonstrates that 0Na+ treatment alone (gray trace, n = 8) does not block the mitochondrial Na+/Ca2+-exchanger. C. Inhibition of NCX reversal by KB-R7943 (black and grey traces) does not alter the effect of CGP-37157 (black trace, n = 10) Traces are averaged responses with standard error shown for selected points. It is also possible that the 0Na+-mediated [Ca 2 +]MT plateau was due to loss of intracellular N a + and subsequent inhibition of the mitochondrial Na+/Ca2+-exchanger (mitoNCX) (147). Incubation of the cells with a selective inhibitor of the mitoNCX, CGP-37157 (10-20 uM), caused a state-state increase in [Ca ] M T that was additive with the plateau caused by ONa (Figure 3.IB). This additivity was also observed in cells that were pre-treated with KB-R7943 to reduce the depletion of intracellular Na + (Figure 3.1C). Thus it is unlikely that the steady-state elevation of [Ca 2 + ] M T in 0Na + is due to inhibition of the mitoNCX. 3.3.2 C a 2 + influx through revNCX increases [Ca2+] in the sub-plasmalemmal space: It is often assumed that rapid increases in [Ca ]MT require mitochondria to be closely apposed to their Ca source. We therefore investigated whether ONa caused an increase in [Ca ]SU.DPM in parallel with the [Ca ]MT elevation using PM-aequorin, which localizes to the inner leaflet of the plasmalemma and has previously been used A7r5 cells (148). We observed pronounced plasmalemmal accumulation and a variable degree of cytosolic localization of SNAP-25-tagged proteins in live cells expressing PM-GFP (Figure 3.2A) and in fixed cell that were immuno-54 fluorescently labeled at the HA-epitope on the recombinant PM-aequorin construct (Figure 3.2B). Localization was examined in deconvolved image stacks both in individual images and in xz and yz cross-sections reconstructed cells. The high density of SNAP-targeted proteins at the cell periphery supports the assumption that PM-aequorin preferentially reported changes in [Ca 2 + ] S U bPM-Figure 3.2. Reverse-NCX increases sub-plasmalemmal [Ca ]. A. Localization of transiently expressed PM-GFP shown in an x-y mid-cell slice and cross-sections in the x-z and y-z planes. B. Localization of transiently expressed PM-aequorin by immuno-fluorescent labelling of aequorin at the recombinant HA-epitope. Cells are amongst a confluent lawn of cells. Scale bars are in microns. C. The PM-aequorin response to Na+-removal (black trace, n = 12) was similar in shape to that reported by mito-aequorin. KB-R7943 (10 uM) attenuated the transient phase of the response (grey trace, n = 12). Traces are averaged responses with data points showing standard error. The PM-aequorin response to 0Na + was similar in nature to the mitochondrial response exhibiting a transient increase in [Ca 2 + ] S U bPM (to 1.2 - 2.5 pM) that fell to a plateau of-100-300 nM above resting levels (Figure 2C). The transient phase of the PM-aequorin response was also inhibited by KB-R7943 (10 pM) indicating that revNCX caused the transient elevation of [Ca 2 + ] S U bPM upon removal of extracellular Na + . Thus it was likely that mitochondria in the 9 + periphery of the cell could have responded to this elecation of [Ca ] s u bPM- To determine further whether this elevation of cytosolic [Ca ], and subsequent mitochondrial stimulation was restricted to the subplasmalemmal microdomain, we investigated the global change in cytosolic [Ca 2 +] in response to revNCX. 55 3.3.3 ONa+ causes a delayed increase in global cytosolic [Ca2+]: Since the fura-2 ratio is linearly related to [Ca 2 +]; it is ideally suited for measurement of average [Ca2 +]j in the bulk of the myoplasm (see discussion) (187). 0Na + caused a delayed and tonic increase in the fura-2 ratio (Figure 3.3A). Assuming that changes in the fura-2 ratio represent changes in the average cytosolic [Ca 2 +], we hypothesized that the delay in the [Ca 2 +]i elevation might be due the superficial SR buffering revNCX-mediated C a 2 + influx (187;237). As predicted, blocking SERCA with cyclopiazonic acid (CPA, 30 uM) abolished the delay in the response to 0Na + and increased the amplitude of the fura-2 response upon N C X reversal with 0Na + (Figure 3.3B). To a lesser extent mitochondrial inhibition has a similar effect, allow the transient phase of the response to be translated to the bulk cytosol (data not shown). A B Figure 3.3. Effects of Na+ substitution and SERCA blockade on [Ca2+]j. A. Fura-2 reported changes in cytosolic [Ca2+] ([Ca2 +]i) in response to 0Na+ (Average trace of 4 independent experiments). B. Effect of SERCA inhibition with cyclopiazonic acid (CPA, 30 uM) average cytosolic response to 0Na+ (Average trace of 13 independent experiments). Data points show mean ± standard error. 3.3.4 SR and mitochondria compete for uptake of reverse-NCX mediated Ca 2 + entry: C a 2 + released from the peripheral SR under resting conditions is extruded from the cell by 2+ forward-mode N C X in a vectorial manner (161). In the absence of extracellular Ca this extrusion process reduces the amount of SR C a 2 + available to stimulate mitochondria (220). Previous reports combined with our current findings with fura-2 and CPA suggest that the superficial SR may play an important role in buffering revNCX-mediated Ca 2 +-infiux (28). Inhibition of SERCA with CPA, which depletes releasable SR C a 2 + (220), increased the peak 56 mitochondrial response to 0Na + by -20% (from 3.10 ± 0.20 to 3.76 + 0.16 pM, p < 0.01, 2-sample t-test) (Figure 3.4A) and slowed the decay of [Ca 2 + ]MT following stimulation. This indicated that SR C a 2 + buffering influences the amplitude and kinetics of [Ca 2 + ]MT elevations, presumably by competing for C a 2 + influx upon N C X reversal. Fitting the decays of [Ca 2 + ]MT in the absence or presence of CPA to double exponential equations (R 2 > 0.99) and plotting the instantaneous rate of decay (-d[Ca2+]/dt) as a function of [Ca 2 + ]MT (Figure 3.4B) showed that CPA slowed the decay of [Ca 2 +]MT- Such an effect would indicate that inhibition of SERCA-mediated C a 2 + buffering slowed the decay of the ambient [Ca 2 + ] s u bPM near the mitochondria upon 9+ N C X reversal or that SERCA blocked somehow impaired mitochondrial Ca extrusion. In parallel experiments using PM-aequorin, SERCA inhibition increased the peak [Ca 2 + ] s u b p M response to 0Na + from 1.27 ± 0.15 p M to 2.13 ± 0.17 p M (n = 6 & 7, p = 0.01, 2-sample t-test), and caused a notable increase in the steady-state [Ca 2 + ] s u bPM (Figure 3.4C). This observation does not exclude the possibility that CPA impaired mitochondrial C a 2 + extrusion (see ref), but parallel changes in [Ca 2 + ] \ iT and [Ca 2 + ] s u bPM are indicative of a causal relationship. Moreover, 0Na + and CPA produced synergistic increases in steady-state [Ca 2 + ] S U bPM (0Na+ 112 ± 77 nM; CPA 146 ± 46 nM; combined 390 ± 86 nM above resting levels), which led us to propose that the peripheral mitochondria are sensitive to steady-state changes in [Ca 2 + ] s u bPM that reflect inhibition of mechanisms that buffer basal C a 2 + influx into the sub-plasmalemmal cytosol. 3.3.5 Mitochondrial C a 2 + buffering moderates reverse-NCX mediated elevation of [Ca 2 + ] s „bPM: Mitochondria clearly take up C a 2 + upon revNCX-mediated C a 2 + influx, but it was 94- 9+ unclear i f the amount of Ca uptake was sufficient to moderate changes in [Ca ] s u bPM (see (146)), which are also buffered by the superficial SR. Collapse of the mitochondrial membrane potential with FCCP (2 pM) impairs mitochondrial C a 2 + uptake (220), and FCCP increased the elevation of [Ca 2 + ] s u bPM mediated by 0Na + from 1.42 ±0.13 p M to 2.15 ± 0.11 p M (n = 6 pairs, p < 0.01 by paired t-test) (Figure 3.4D). The selective blocker of the C a 2 + uniporter, ruthenium red (2 pM, 30 min), reduced stimulated mito-aequorin responses by 35 ± 5% (n = 6, p < 0.01, data not shown) and increased the PM-aequorin response to 0Na + from 1.87 ± 0.13 p M to 2.26 ± 0.25 p M (n = 8 pairs, p < 0.05 by paired t-test) (Figure 3.4E). Thus, we conclude that peripheral mitochondria buffer revNCX-mediated C a 2 + influx, and by doing so regulate [Ca 2 + ] S U bPM in co-ordination with the superficial SR. 57 5000 30txM CPA B — 6000 c 1 4000 2000 O 7 0 2000 4000 [Ca'larrfnM) 3000 2500 1500 O 1000 500 r u t h e n i u m r e d 1 min 3000 ^ 2500-^  0 J 3Qu:M CPA ONa* I 2000- A r150°" i \ « 1 0 0 0 f \ " 500 t & V 4 ^ ^ > l 1 min 3000 Figure 3.4. The sarcoplasmic reticulum and mitochondria buffer and compete for revNCX-mediated Ca influx. A. Cyclopiazonic acid (CPA, 30 uM) increased the peak [Ca 2 + ] M T response to 0Na+. B. The instantaneous slope of the down stroke of the traces in A plotted as a function of [Ca 2 + ] M T (black trace - control; gray trace -CPA). C. CPA increased the peak [Ca 2 +] s u b P M response to 0Na+ (black trace - control, n = 6; CPA - grey trace, n = 7). D & E. Mitochondria depolarization with FCCP (2 uM, n = 4) or inhibition of the Ca 2 + uniporter with ruthenium red (2 uM, n = 9) increased the [Ca 2 + ] s u b PM response to 0Na+. F. Comparison of treatments in D & E with paired controls. Error bars show standard error. * - p-value < 0.05; ** - p-value < 0.01. 3.3.6 NCX reversal influences the tail of the mitochondrial response to ATP: In light of reports that revNCX contributes to agonist-induced Ca -entry (64; 134), we investigated whether revNCX occurred during agonist-induced stimulation and whether this caused mitochondrial C a 2 + uptake. In fura-2 loaded cells, activation of purinergic receptors with adenosine 5'-trisphosphate (ATP, 1 mM) produced a transient [Ca ] i peak and a subsequent 9 + plateau typical of agonist-mediated responses (Figure 3.5A). Incubation in Ca free PSS did not significantly reduce the peak stimulated [Ca2+]j elevation (without C a 2 + 2.56 ± 0.30 R340/380, n = 6; with C a 2 + 2.58 ± 0.14 R340/380, n = 15; p > 0.90), but the [Ca2 +]j plateau was abolished (data not shown). When the mitochondrial and cytosolic responses were scaled and superimposed to compare the time course of the responses, the transient [Ca 2 + ]MT elevation closely matched the transient phase of the fura-2 response, but [Ca 2 + ]MT continued to decline to 58 resting levels as the fura-2 signal reached a plateau. This initially indiciated that mitochondria were insensitive to agonist-induced Ca influx. However, when Ca was removed from the super-perfusate ~10 seconds before stimulation with ATP, the tail of the [Ca 2 + ]MT response declined faster than in the presence of extracellular C a 2 + and established a steady-state level below resting [Ca 2 + ]MT (Figure 3.5B). This brief 0Ca 2 + treatment did not affect the [Ca 2 + ]MT peak height (Figure 3.5Ci), but reduced the area under the curve to 84 ± 5% (n = 11, p = 0.01, paired t-test) of paired control responses (Figure 3.5Cii) and caused a significant suppression in the tail of the [ C a 2 + ] M T response (682 ± 121 nM vs 350 ± 84 nM, p < 0.01, n = 11) at the point of maximum separation between the tails of the responses in the presence and absence of extracellular C a 2 + (Figure 3.5Ciii). Apparently, ATP-stimulated C a 2 + entry had a small, but measurable, effect on mitochondrial C a 2 + uptake. A C Time (min) External [Ca2*] (mM) Figure 3.5. Ca 2 + influx contributes to the tail of the mitochondrial responses to ATP. A. Mitochondrial (grey dotted trace) and cytosolic (solid black trace) responses to ATP. B. Superposition of averaged mito-aequorin responses to ATP with extracellular Ca 2 + present (black trace) or removed from the bathing solution (dotted gray trace) 10-30 seconds before stimulation. C. Paired analysis of the effect of brief Ca 2 + removal: (i) peak stimulated [Ca 2 +]M T, (ii) area under the curved, (iii) [Ca 2 + ] M T at point of maximum separation between the traces. * p < 0.05. 59 Blockade of revNCX with KB-R7943 revealed a similar, subtle reduction in the tail of the [Ca 2 + ]MT response to ATP in addition to a small decrease in the peak [ C a 2 + ]MT elevation (Figure 3.6A). Paired analysis revealed that KB-R7943 reduced the area under the curve by 22 ± 5% (1.72 ± 0.13 uM-min control 1.35 ± 0.10 uM-min KB-R7943, p < 0.001, paired t-test) and the peak [ C a 2 + ] M T elevation by 11 ± 5 % (4.51 ± 0.28 uM control vs. 4.00 ± 0.28 uM KB-R7943, p = 0.001, paired t-test) (Figures 3.6Bi & Bii). Inhibition of revNCX also caused a significant separation in the tails of the [ C a 2 + ] M T traces (612 ± 59 nM control vs. 421 ± 24 nM KB-R7943, p < 0.01, n = 15 pairs) (Figure 3.6Biii). In contrast, the transient phase the ATP-mediated increase in average cytosolic [Ca 2 +] was unaffected by inhibition of revNCX (KB-R7943: 2.56 ± 0.33 R340/380, n = 12; control: 2.44 ± 0.14, n = 14 R340/380; p > 0.05, 2-sampled t-test), but the plateau phase was abolished (KB-R7943: 0.92 ± 0.04 R340/380, n = 12; control: 1.11 ± 0.06, n = 14 R340/380; p < 0.05, 2-sampled t-test) (Figures 3.6C & D). Similar to treatment with C a 2 + free solution, KB-R7943 caused both mitochondrial and cytosolic traces to' diverge from their respective controls 40 - 45 seconds after stimulation. Collectively, these observations provided compelling evidence that revNCX contributed to C a 2 + entry during the sustained cytosolic [Ca2 +] plateau stimulated by ATP. This revNCX-mediated C a 2 + entry also caused a detectable mitochondrial C a 2 + uptake. 60 A B 5000 O.Sh Figure 3.6. NCX-reversal causes mitochondrial Ca -uptake during agonist stimulation. A. KB-R7943 (lOuM) significantly reduced the mitochondrial response to ATP (1 mM) (gray dotted trace) compared to control experiments (black trace). Traces are average of 17 replicates for each treatment. Error bars show mean ± standard error. B . Quantitatifive analysis of KB-R7943 effect on (i) peak [Ca 2 + ] M T, the (ii) area under the curve and (iii) the point of maximal separation in the tail of the responses to ATP. C. Average cytosolic [Ca2+] responses to ATP in the presence of KB-R7943 (grey dotted trace) or absence of KB-R7943 (black trace) D . Analysis of KB-R7943 effects on (i) peak [Ca2+]j response to ATP rand the subsequent (ii) [Ca2+]j plateau. 3.3.7 Agonist-mediated up-regulation of mitochondrial C a 2 + extrusion: While the arguably modest effect of KB-R7943 on the mitochondrial response to ATP could be attributed to the fact that only peripheral mitochondria are expected to respond to N C X reversal, the return of [Ca 2 +]MT to resting levels in the sustained presence of the agonist suggested that stimulation of mitochondria by revNCX was only transient. However, it has been reported that both mitochondrial C a 2 + uptake and extrusion are stimulated during receptor activation (147). We tested whether receptor stimulation enhanced mitochondrial C a 2 + turnover by inhibiting the mitochondrial N C X with CGP-37157 (20 uM) prior to and during purinergic stimulation. CGP-37157 caused a steady-state [Ca 2 +]MT elevation that was significantly larger in the presence of ATP compared to resting cells (ATP = 184 ± 23 nM; control = 72 ± 31 nM; n = 14 pairs; p < 0.01, paired t-test) (Figure 3.7). Similar findings have been reported in endothelial cells by Malli and co-workers (146; 147). Pre-incubating cells with KB-R7943 (10 uM) reduced the height of 61 the CGP-mediated [Ca ]MT plateau following ATP-stimulation from 235 ± 22 nM to 139 ± 18 nM (p < 0.005, Wilcoxon signed rank t-test, n = 11 pairs). This effect of KB-R7943 was similar to the inhibition of the ATP-mediated cytosolic [Ca 2 +] plateau illustrated in Figure 3.6. Together these findings strongly suggest that during ATP-mediated stimulation of vascular smooth muscle cells reverse-mode N C X contributes to agonist-induced C a 2 + influx across the P M and stimulates mitochondrial C a 2 + turnover. A B 2000 1500 rA 10<XH + CM 5001 oJ C G P 2000 1500 1000 500 1 min. I 1 WW A T P 1 min. I 2000 1500 1000 500 K B R i ; 5 I I C G P 1 \ i\, /•'«*-**"*• A T P 1 min Figure 3.7. Reverse-mode N C X increases mitochondrial C a 2 + flux in ATP-stimulated cells. A Inhibition of the mitochondrial Na+/Ca2+-exchanger with CGP-37157 (20 uM) causes a steady-state increase of [Ca 2 + ] M T in resting cells that is larger in the presence of ATP (B). C . In an independent series of experiments, the addition of CGP in the presence of ATP caused a smaller increase in [Ca 2 + ] M T in cells pre-treated wuth KB-R7943 (10 uM). Grey dotted traces shows average response in absence of KB-R7943. 3.4 D I S C U S S I O N Mitochondria interactions with plasmalemmal ion channels have been demonstrated in numerous cell types. In smooth muscle, mitochondrial buffering of [Ca 2 + ] S U bPM was first indicated using the activity of Ca2+-sensitive K + - and Cl"-channels as an indirect measure of [Ca 2 + ] s u bPM (91 ;261). Rembold and Chen later took advantage of the fact that untargeted aequorin and fura-2 are differentially sensitive to focal changes in [Ca 2 +] to demonstrate that the superficial buffer 94-barrier creates a standing [Ca ] gradient between the subplasmalemmal space and the bulk cytosol in V S M (187;235). While untargeted (i.e. cytosolic) aequorin can detect [Ca 2 +] changes in the subplasmalemmal microdomain, the PM-aequorin provides more accurate and robust 9 + measures of [Ca ] S U bPM (20). Using an approach similar to Rembold, we find that mitochondria .2+ buffer Ca entering the sub-plasmalemmal cytoplasm upon N C X reversal. 62 Traditionally, the Na+/Ca2+-exchanger is viewed as an extrusion mechanism relying on release of C a 2 + from closely associated superficial SR to locally generate the [Ca 2 +] required to overcome its low Ca2+-affinity (161;220), but this close association also allows that SR to buffer C a 2 + influx upon N C X reversal (131). Our current results further illustrate that the superficial SR buffers revNCX-mediated C a 2 + entry in competition with the peripherally located mitochondria, and these processes are active during purinergic receptor stimulation. 3.4.1 Functional and localized interaction of NCX, SERCA and Ca 2 + uniporter: Electron microscopy studies in V M S have revealed sub-plasmalemmal mitochondria neighbouring junctions of the plasmalemma and superficial SR (176). In adrenal chromaffin cells, subplasmalemmal mitochondria mediate specialized C a 2 + handling (7); however, little is known of the specific C a 2 + handling of these peripheral mitochondria in V M S (reviewed in (177)). However, the N C X is thought to be localized at PM-SR junctions. The current data provide strong evidence that peripheral mitochondria directly communication with the N C X . Furthermore, the co-operative [Ca ] s ubPM buffering in stimulated and resting cells illustrates a functional C a 2 + signaling complex consisting of the N C X , SERCA and the mitochondrial 94-uniporter. The process of direct Ca transfer between organelles, especially mitochondria, is thought to rely on their close spatial association (98; 193). In this case the close linkage of the plasmalemmal N C X and the mitochondrial uniporter is supported by the immediate and parallel changes in [Ca 2 +]MT (Figure 3.1) and [Ca 2 +] S UbPM (Figure 3.2) upon extracellular Na+-removal. The restricted localization of these C a 2 + signals to the sub-plasmalemmal cytosol, and thus a sub-population of mitochondria, is further supported by the delayed and modest changes in the fura-2 response to 0Na +, which reports predominantly bulk cytosolic [Ca 2 +] changes (Figure 3.3). 3.4.2 SERCA and the C a 2 + uniporter compete for C a 2 + entering the cell through reverse-mode NCX: The increased responses in [Ca 2 +] s ubPM and [Ca 2 +]MT Na+-removal by inhibition of SERCA with CPA indicated that both the peripheral SR and mitochondria take up C a 2 + that enters the cell through reverse-mode N C X . Moreover, the fact that CPA enhanced the mitochondrial response (Figure 4) under conditions that deplete releasable SR Ca stores 2+ 2"t" indicates that the mitochondrial response to N C X reversal was not due to Ca -induced-Ca -release (220). Rather, SERCA-mediated C a 2 + buffering moderated the [ C a 2 + ] M T elevations upon N C X reversal. The simplest explanation of these findings is that the loss of SR C a 2 + buffering of 63 leads to increased and prolonged elevations of [Ca ] S U DPM and consequently [Ca ] M T indicating competitive or co-operative buffering of [Ca ] s u bPM elevations by mitochondria and the 9 + superficial SR. This model is in line with reports that mitochondria buffer capacitative Ca entry and facilitate refilling of the endoplasmic reticulum {(106; 147; 169) and reviewed in (176)}. RESTING forward ACTIVE Figure 3.8. Model for interaction of NCX, SERCA and mitochondria in vascular smooth muscle. Left. In resting smooth muscle, SERCA buffers resting Ca 2 + influx and SR Ca 2 + release channels deliver this Ca 2 + to the NCX operating in forward-mode. NCX is also directly activated by the elevated [Ca 2 + ] s u b PM independent of SR Ca 2 + release. Mitochondria also take-up basal Ca 2 + influx, but Ca 2 + removal from the sub-plasmalemmal space by NCX and SERCA reduces the exposure of mitochondria to this Ca 2 + entry. Right. Stimulation with ATP causes NCX reversal and immediate elevation of [Ca 2 +] s u b P M. This is due to opening of Na+-permeable channels in response to receptor activation or SR Ca2+-depletion. The resulting Ca2+-influx is taken up into the SR and mitochondria by SERCA and the Ca 2 + uniporter, respectively. Abbreviations: RC, Ca2+-release channel; SERCA, sarco-endoplasmic reticulum Ca2+-ATPase; Ca 2 + /X + , mitochondria Na+/Ca2+-exchanger and or H+/Ca2+-exchanger; U, Ca 2 + uniporter; VDAC, voltage-dependant anion channel. Direct inhibition of mitochondrial C a 2 + uptake further demonstrate that mitochondria sufficiently buffer revNCX-mediated Ca influx to regulate subplasmalemmal Ca signaling (Figure 3.4D & E). Indirect evidence has previously pointed to such a role for mitochondria in smooth muscle cells (91), and a recent report by Frieden et al. in endothelial cells elegantly demonstrated that mitochondria regulate the local [Ca 2 + ] S U bPM near Ca2+-activated K+-channels (76). We used two independent means of inhibiting mitochondrial C a 2 + uptake to confirm that 9 + mitochondria buffer [Ca ] S U bPM as While both FCCP and ruthenium red can have distinct actions 9 + other than inhibiting mitochondrial Ca uptake, FCCP is commonly used to inhibit mitochondrial C a 2 + uptake by depolarizing mitochondria, and previous reports in smooth muscle have provided indirect evidence indicating that FCCP increases [ C a 2 + ] s u b p M (91;261). While it is 64 possible that mitochondrial depolarization can cause confounding effects on Ca signaling that are not the direct results of the loss of mitochondrial C a 2 + uptake (i.e. altered SR C a 2 + release, ATP production and membrane potential (41;48;129), the fact that ruthenium red also enhanced the PM-aequorin response to N C X reversal further supports the conclusion that the Ca buffering capacity of peripheral mitochondria is sufficient to moderate [Ca ] s ubPM fluctuations in smooth muscle cells. By inducing robust NCX-reversal, we have demonstrated direct C a 2 + crosstalk between the N C X , SERCA and the C a 2 + uniporter. This crosstalk is likely restricted to a specific population of mitochondria that are exposed to local elevations of [Ca 2 +] s ubPM that are not readily translated into the bulk cytosol. Such a subplasmalemmal microdomain was postulated as an essential part of the "superficial buffer barrier" (233), and we hypothesize that peripheral mitochondria constitute a rapidly acting element of the superficial buffer barrier (illustrated in Figure 3.8). Such superficial C a 2 + buffering by mitochondrial is highlighted in studies of the regulation of capacitative C a 2 + entry by local [Ca 2 + ] s u b pM (106; 147; 169). In smooth muscle, store/receptor-operated C a 2 + entry likely occurs via the functional linkage of N C X with non-selective cation channels of the TRP family (11;21). 3.4.3 NCX-reversal stimulates mitochondrial C a 2 + uptake upon purinergic receptor stimulation: We have shown that reverse-mode N C X is required to sustain the tonic [Ca2 +]j elevation that follows purinergic receptor activation by ATP (Figure 3.6), similar to recent findings in duodenal mucosa (64). Presumably, this N C X reversal is due to the opening of Na + -permeable store-operated channels (SOC) associated with N C X and the local elevation of [Na+]SUbPM causing to reverse the N C X as proposed by Blaustein and co-workers (11). The molecular basis of this model was subsequently supported by functional linkage of N C X and TRPC3 in HEK293 cells (197). We have previously shown that such N C X reversal mediates refilling of SR C a 2 + stores in adrenergicaly-stimulated inferior vena cava via specialized PM-SR junctions that couple the N C X to SERCA (131; 134), and Hellstrand's group found that mitochondrial inhibition alters the frequency and amplitude of adrenergicaly-driven Ca -oscillations (217). This regulation of C a 2 + oscillations by mitochondria likely reflects the mitochondrial modulation of Ca2+-dependent gating of IP3R, but it is tempting to speculate that the interaction of mitochondria with N C X described herein also affects the nature of C a 2 + oscillations and SR refilling. 65 In line with a recent report in endothelial cells (147), we propose that peripherally located mitochondria co-operate with the P M and SR ion transporters in the junctional Ca signaling complexes thought to facilitate SR refilling during agonist-mediated C a 2 + signaling in some V S M (176). Central to this argument, the first detectable effect of KB-R7943 on the down stroke of the agonist-mediated [Ca 2 +]MT response coincided with the onset of the [Ca 2 +]; plateau (Figure 6). While KB-R7943 affected only a small portion of the total mitochondrial response to ATP, this must be considered in context. The transient phase of the mitochondrial response to ATP is mediated primarily by SR C a 2 + release (220), which should stimulate mitochondria throughout the cell (Figure 3.5). In contrast, the agonist-induced NCX-reversal that followed SR C a 2 + release and coincided with the tail of the [Ca 2 + ] M T elevation (Figure 3.6) should stimulate only peripheral mitochondria. Moreover, it is the trans-mitochondrial flux, rather than Ca accumulation, that may be important to allow sustained agonist-induced C a 2 + entry (147). One of the most interesting finds in the current study is the demonstration of agonist-induced up-regulation of mitochondrial NCX-mediated C a 2 + extrusion (Figure 3.7). This in part masked the elevated mitochondrial C a 2 + flux during purinergic stimulation. Similar elevation of mitochondrial C a 2 + flux despite [Ca 2 +]MT returning to resting levels in the sustained presence of agonist was also observed in endothelial cell (147), and we further demonstrate that C a 2 + flux was facilitated by revNCX-mediated C a 2 + entry. If in fact the plasmalemmal Na+/Ca2+-exchanger associates with Na+-permeable TRPC channels in these cells (expressing TRPCI, 4 and 6 (178)), it is conceivable that elevation of [Na+] s ubPM facilitates the increase in mitochondrial Na + /Ca 2 + -exchange. In doing so Na+-influx would mediate both C a 2 + influx via N C X reversal support the continuous mitochondrial C a 2 + flux that Malli et al. find to be required for sustained capacitative C a 2 + entry in endothelial cells (147). In light of the literature cited above, our current findings are likely to be of relevance in a number of different cell types. 3.4.4 Conclusion: Investigations of the role of mitochondria in the generation of localized C a 2 + gradients near the plasmalemma and SR have offered exciting insights into site-specific C a 2 + signaling, and we are only beginning to understand how mitochondria modulate these localized signals in V S M . Here we report the localized interaction of a sub-population of peripheral mitochondria, the Na+/Ca2+-exchanger and SERCA in V S M . These mitochondria assist the SR in buffering agonist-mediated C a 2 + influx before it can diffuse deeper into the 66 cytosol. The agonist-mediated C a 2 + influx is mediated by reversal of the Na + /Ca 2 + exchanger, which attributed to localization of N C X near Na+-permeable TRPC channels. 67 CHAPTER IV IP3 and ryanodine receptors localize to separate SR sub-compartments that preferentially co-localize near mitochondria5 4.1 I N T R O D U C T I O N : The regulation of calcium signaling in vascular smooth muscle (VSM) is highly dependent on the ultra-structural architecture of the cell. Contraction is one of many processes, including secretion, proliferation and apoptosis that are controlled by fluctuations in the cytosolic concentration of free ionic calcium ([Ca2+];). In order for this single ion to simultaneous control multiple processes, cells have developed highly ordered systems to regulate the spatial and temporal patterns of [Ca 2 +]; fluctuations resulting from C a 2 + influx across the plasmalemma and release from the sarcoplasmic reticulum (SR) (23; 176). While C a 2 + influx plays an important role in refilling SR C a 2 + stores and modulating [Ca 2 +]; signals, in this report we focus on the release of SR C a 2 + during agonist mediated [Ca 2 +]i elevation in the vascular smooth muscle. Unlike striated muscle, smooth muscle utilizes both inositol-1,4,5-trisphosphate receptors (IP3R) and ryanodine receptors (RyR) to release C a 2 + from the SR and generate the increase in global [Ca 2 +]i that stimulates myofilament contraction (32;87;255). The gating of IP3R and RyR is modulated by C a 2 + in addition to inositol 1,4,5-trisphosphate and cyclic-ADP ribose, respectively (244;254), and it is thought that localized C a 2 + gradients in the cytosolic microdomains between the SR and closely associated plasmalemma and mitochondria facilitate localized feedback control of I P 3 R and RyR (23-176). Because of the localized nature of these interactions, the precise sub-cellular distribution and spatial relationship of IP3R and RyR with each other, the plasmalemma and mitochondria are highly relevant to the regulation of cytosolic C a 2 + signaling and contraction in V S M . A long-standing question in V S M physiology is whether IP3R and RyR are expressed on and release C a 2 + from separate elements (or sub-compartments) of the SR or whether they are expressed on the same SR elements and share a common store of Ca 2 + . Fluorescent labeling of the SR of colonic smooth muscle revealed that IP3R and RyR are largely localized to same elements of SR (255), which is consistent with a recent review concluding that IP3R and RyR release C a 2 + from a common or overlapping pool of SR in most smooth muscle (153). However, 5 A version of this chapter has been prepared for publication in the Journal of Cell Biology as: IP3 and ryanodine receptors localize to separate SR sub-compartments that preferentially co-localize near mitochondria in aorta smooth muscle cells. Damon Poburko, Kathryn Potter, Eric Lin, Megan McLarnon, Nicola Fameli, Hubert Walinski, Tania Szado, Cornelis van Breemen 68 direct imaging of changes in SR [Ca ] in V S M shows that Ca can be differentially released from closely associated but spatially and pharmacologically distinct sub-compartments of the SR (85). This strongly suggests that IP3R and RyR are expressed on and release C a 2 + from separate SR elements in V S M . In portal vein, immuno-labeling has shown IP3R and RyR to be localized throughout cross sections of myocytes (31), but detailed analysis of their co-localization has yet to be reported in vascular smooth muscle cells. The subcellular localization of IP3R and RyR is also directly relevant to the stimulation of mitochondrial C a 2 + uptake because almost all mitochondria in smooth muscle cells are found in close association with SR (57;131;210). Release of SR C a 2 + into these SR-MT junctions generates the micromolar [Ca 2 +] required to activate the mitochondrial C a 2 + uniporter (160;192;220), which allows mitochondria to take up C a 2 + and thereby: 1) assist with SR refilling, 2) modulate the gating of IP3R and RyR by buffering local C a 2 + gradients, and 3) to serve as sensitive reporters of localized SR C a 2 + release. While in some types of smooth muscle, mitochondrial C a 2 + uptake may not be highly dependent on close mito-SR coupling (151), both the IP3R and RyR are closely coupled with the mitochondria in aorta myocytes (160;220). Moreover, in conduit artery smooth muscle, mito-SR cross talk is thought to play an important role in regulating the frequency and amplitude of agonist-induced C a 2 + oscillations (131;217). However, the relative contributions of IP3R and RyR to agonist-mediated mitochondrial Ca uptake and the relevance of their cross-talk (and therefore sub-cellular localization) is poorly characterized in V M S . Recently, we reported that distinct SR sub-compartments stimulate mitochondria in cultured aorta SMC (220), and in our present study we have again used aequorin targeted to the mitochondria to detect localized SR C a 2 + release (191) in combination with immuno-cytochemistry to demonstrate that IP3R and RyR are localized to, and release C a 2 + from, separate SR sub-compartments. Activation of two phospholipase-C coupled receptors caused increased [Ca 2 +]MT by similar patterns of SR C a 2 + release. Furthermore, comparison of agonist-mediated increases in [Ca 2 +]j and [Ca 2 +]MT with the sub-cellular distribution of IP3R and RyR has provided novel insights into how IT^R-RyR cross talk differentially influences cytosolic and mitochondrial C a 2 + signaling. 69 4.2 MATERIALS & METHODS: 4.2.1 Smooth muscle cell culture: Cells were cultured as described previously (220). A proprietary line of cultured rat aortic smooth muscle cells (RASMC), stored in 90% DMEM/10% DMSO in liquid nitrogen, were thawed and used between passages 8-12. Cells were incubated at 37°C in a humidified atmosphere of 5% CO2. 4.2.2 Expression of mito-aequorin and inverse pericam: Transfection of R A S M C with mitochondrial aequorin has been previously described (mito-aequorin) (148;194). Inverse pericam (Riken Institute) was transiently transfected into R A S M C on Thermanox™ cover slips coated with l/30x Matrigel™ (30 min, 37°C) (Nunc, Life Technologies). 24 hours after plating, cells were washed with Ca 2 +/Mg 2 +-free PBS and fed with D M E M (10% FCS) before transfection with Effectene® (Qiagen) as per manufacturer's instructions (1 pg D N A per 2.77 pL Effectene). Cells were used for experiments on the following day. 4.2.3 Measurement of mitochondrial [Ca2+]: Changes in mitochondrial [Ca 2 +] were reported by mito-aequorin. On the day of experiment, mito-aequorin was reconstituted with coelenterazine (5 pmol/L, serum free D M E M ) for 2—4 hrs. Experiments were performed in thermostated perfusion chambers (37°C) that were positioned ~5 mm below photon detectors. Cells were superfused (1 mL/min) with physiological salt solution (PSS, in mmol/L: NaCl 145, KC1 5, M g C l 2 1, HEPES 5, glucose 10, and CaCl 2 1.2, pH 7.4). Luminescence was detected by photomultiplier tubes (B2F-RFI/9789A and P25232, Electron Tubes Inc, USA), and photon emission was sampled at 1 Hz with EM6 photo-counting software (Electron Tubes Inc, USA). Off-line calibration of photon emission to [Ca 2 +] has been previously described (148;220). 4.2.4 Measurement of cytosolic [Ca2+]: Cover slips with pericam-expressing cells were mounted in a thermostated (35-37°C) perfusion chamber and perfused with PSS (~ 1.5 ml/min). Pericam fluorescence was imaged on an Olympus BX50WI microscope (lOx UMPlanFl objective, 0.30 NA) with an Ultraview Nipkow disc (Perkin Elmer). Image acquisition and analysis was performed with Ultraview 4.0 software (Perkin Elmer). Cells were excited at 488 nm (Argon/Krypton laser, Melles Griot), and fluorescence emission was collected at 525/530 nm. Fluorescence intensity was corrected for linear decay before data was inverted and expressed in the form of (F-F 0)/(Fm a x-F 0) where F 0 is the baseline fluorescence prior to stimulation and F m a x 70 is the maximal change in fluorescence after permeabilizing cells with digitonin (lOOpM) in the 2+ presence of 2.5mM Ca . 4.2.5 Immuno-cytochemistry: Cells grown on cover slips were washed with Ca 2 + /Mg 2 + -containing phosphate buffered solution (2 x 5min) followed by 0Ca 2 + /0Mg 2 + -PBS (PBS, GIBCO) immediately prior to fixation with paraformaldehyde (PFA, 4%, lOmin). Subsequent solutions were made with PBS (pH 7.4). After fixation, cells were washed 3 times for 5 min with PBS, followed by 5 min in 0.2 M glycine and two more PBS washes. Cells were permeabilized with Triton X-100 (0.1%, 10 min) and washed 3 x 5 min with PBS. Cell were incubated for 1 hr in blocking solution-step (PBS, 1% goat serum, 1% bovine serum albumin) before incubation with primary antibodies (1:100 mouse anti-RyR (PA3-925) and 1:200 rabbit anti-IP3R-l (PA3-901), Affinity Bioreagents) over-night at 4°C. Excess primary antibody was removed with three 20 min washes in blocking solution prior to a 1 hr incubation at room temperature with fluorescent labels, which included goat anti-mouse F(ab2)' Alexa-fluor 488 or 546 (1:100), goat anti-rabbit highly cross-adsorbed Alexa-fluor 594 (1:100) and Alex-fluor 488 conjugated phalloydin (1:200) (Molecular Probes, Oregon) as described. Cover slips were then mounted on slides with Prolong Gold (Molecular Probes) and cured overnight before acquiring images. For each set of labeled cells, images were taken with the same acquisition parameters from control cells that were treated identically but were not exposed to primary antibody. 4.2.6 Confocal image acquisition & deconvolution: Images were acquired on a Leica SP2 inverted confocal microscope with an 63x H C X PL APO objective (1.4 NA). In order to maximize image resolution several key factors were taken into account. Images were super-sampled at a pixel size of 71nm by 71nm and an z-interval of 150nm to ensure anti-aliasing and to supersede the Nyquist Sampling criteria (i.e. pixels size was than one-half of the theoretical limit of diffraction limited optical resolution). Images were sampled using 8 line-averaging to increase the signal to noise ratio. Dual labeled cells were illuminated at 488 nm and 594 nm simultaneously, and emission was collected on separate photomultiplier tubes at 495-580 nm and 605-720 nm. For triple labeled cells, excitation was switched from 488/594 nm to 543 nm between scan lines, and emission bands were collected on three separate photomultiplier tubes at 495-550, 555-605 and 620-720 nm. Due to the resolution dependencies of co-localization analysis, our images were processed offline to maximize image resolution. Images were deconvolved with a Maximum Likelihood Estimation algorithm using Huygens Professional 71 (Version 2.18 Scientific Volume Imaging) (see (203)). This algorithm takes into account the following imaging parameters: N A of the microscope objective (1.4); refractive index of medium (1.515); refractive index of sample (1.43); excitation wavelength (488/543/594); emission wavelength (525/585/625); confocal pinhole radius (212/237/258); pixel size (71 nm); z-axis interval (150 nm); microscope type (confocal); and number of excitation photons (1). 4.2.7 Co-localization analysis: Co-localization analysis, 3D visualization and reconstruction of 3D models were performed with Imaris (Bitplane, Switzerland). The intensity threshold values for co-localization analysis were determined in a two-step process. A minimal background was determined as the mean plus 2 standard deviations of intensity level that eliminated all but the brightest random-shot noise (not greater than 2x2x2 voxels) from 4-5 controls images. Due to the dense packing of the IP3R and RyR labels and the limits of axial resolution, we determined an additional threshold to limit the contribution of out of focus light, which blurs binarized images of the IP3R and RyR labels. In each cell, 10 instances of what appeared to be SR elements labeled for each of IP3R and RyR were sampled independently in 3-dimensions to determine the threshold required to limit the width of the SR label in the x-y plane to 3-4 pixels (~200-300nm). The mean intensity value of the 10 samples was used as the threshold for co-localization analysis. Since the GFP-labeled mitochondria are expected to have a width of 300-400 nm, GFP-images were subject to a median filter (3x3x1) and thresholded as above using 3-dimensional visualization to limit the size of binarized mitochondria to -400 nm. Co-localization was calculated as the percentage of pixels (i.e., volume) of one label that were also occupied by the other label being analyzed. Co-localization was treated as a binary event, thus all mito-IT^R and mito-RyR co-localization events were used to calculate the co-localization of IP3R and RyR that were also associated with the mitochondria. The 3-dimensional models were generated using the Surpass module of Imaris. The spatial relationship of the IP3R and RyR are depicted as space filling iso-surfaces (Figure 4.6G-I). The relationships of IP3R and RyR with mitochondria are depicted by maximal intensity projections of the mitochondria merged with the mito-IP3R, mito-RyR and mito-IP3R-RyR co-localization events (Figure 4.7D-F). 4.2.8 Modeling the effect of voxel clustering on random co-localization: For this analysis custom routines were written in IDL (IDL 5.6, RSI Inc.). Two 3-dimensional binary arrays were generated containing 115,000 elements, representing the average number of mito-GFP labeled 72 voxels per cell after thresholding for co-localization analysis. 41% and 34% of the elements in the two arrays (i.e. the mean co-localization of IP3R and RyR with mitochondria, respectively) were set to a value of 1 and randomly distributed. The extent of random co-localization was calculated as the number of elements retaining a value of 1 when the two arrays were multiplied. Co-localization was then assessed when the random distribution of voxels (or elements) was constrained by grouping voxels into 7-element star-shaped kernels or 27-element cube shaped kernels to loosely mimic the reticular clustering of IP3R and RyR labeled voxels in the deconvolved images. 4.2.9 Statistical Analysis: Values are expressed as mean ± standard error (SE) with the number of replicates indicated for each experiment. Means were compared with the most robust test appropriate to each experimental design subject to validation of statistical assumptions. For comparison of greater than two means A N O V A was used with pair-wise comparisons made by Bonferroni planned comparisons (i.e., p-values use Bonferroni corrected error). Data was compiled and analyzed using GraphPad Prism 4.0 in coordination with Microsoft Excel. NCSS was used to perform statistical tests. Significance was determined by an ct-value of 0.05. 4.3 RESULTS: 4.3.1 Mitochondria detect activation of I P 3 R and RyR by purinergic stimulation. To investigate the contributions of IP3R and RyR to ATP-mediated SR C a 2 + release, we.measured [Ca 2 +]MT responses to ATP following pharmacological blockade of IP3R or RyR. Under our experimental conditions, C a 2 + release from IP3R is selectively blocked by 2-APB (220) and C a 2 + release from RyR is blocked by procaine (108; 110) (for control experiments see Figure 4.9). Pre-incubation with 2-APB (75 pM) reduced the peak ATP-mediated [Ca 2 +]MT elevation to 67 ± 4% of controls values, and procaine (lOmM) reduced the [Ca 2 +]MT response to 57 ± 6% (Figure 4.1). Together these two blockers reduced the mitochondrial response to 14 ± 6% of controls. These mean values of inhibition suggested that this effect was additive, but careful analysis showed the inhibitory effect of 2-APB and procaine together to be slightly greater than the sum of their individual effects indicating a subtle synergism (p = 0.049, 2-sample t-test). We have previously shown that incubation of these cells in Ca2+-free media (0Ca 2 +) selectively depletes a superficial element of the SR that is preferentially associated with the N C X (220). We now find that incubation in 0Ca 2 + had an additive effect with 2-APB, reducing the ATP-mediated responses 73 from 67 ± 4% to 51 ± 5% of controls. In contrast, the effect of OCa was not additive with the effect of procaine, as procaine and 0Ca 2 + together reduced the ATP-mediated responses to 51 ± 7%. This effect was not statistically different from the inhibition seen with procaine alone (Figure 4. ID), which suggested that 0Ca 2 + caused the depletion of a superficial sub-compartment of the SR that was distinct from the sub-compartment released by the IP3R and indistinguishable from the sub-compartment released by RyR. Figure 4.1. Purinergic stimulation increases [Ca2+]MT by activation both IP3R and RyR. A-C. Average [Ca 2 + ] M T responses to ATP show the individual and combined effects of IP3R inhibition with 2-APB (75 uM), RyR inhibition with procaine (10 mM) and depletion of peripheral SR Ca 2 + stores in Ca2+-free solution (0Ca2+, 0.1 mM EGTA). The average control response to ATP is shown as the faint dotted trace. D. The effects of each treatment were compared to paired control responses with only ATP, and means of normalized responses were compared by ANOVA. P-values from planned Bonferroni comparison are show between specific treatments. 4.3.2 Cytosolic detection of Ca2+-release from IP3R and RyR during purinergic stimulation. Changes in [Ca 2 + ] i were measured under conditions parallel to those used to measure [Ca 2 + ] M T-2-APB inhibited the cytosolic response to 69 ± 7% of matched controls, similar to its inhibitory effect on the mitochondrial responses (Figure 4.2A). Procaine, however, only reduced the cytosolic response to 80 ± 5% of control responses, much less inhibition than was observed for the mitochondrial response (Figure 4.2B). In contrast to the effects observed on the mitochondrial responses, the inhibition of the [Ca 2 + ] i response to ATP by 2-APB and procaine 74 together was highly synergistic (Figure 4.2C). This synergism was consistent with reports that cross-talk between IP3R and RyR is an important factor in the generation of physiological cytosolic C a 2 + signals (30;87;114). A . B . _ 0.5 o ii- 0.4 x ro p. 0.3 o LL 2-APB procaine ALE. 0.1 -0.1 x ro E LL^ 'o LL 1 min. 1 min. —* 0-P]procaine & 2-APB D . 0.0039 +2APB +proc +2APB & proc 1 min. Figure 4.2. Both IP3R and RyR contribute to the ATP-mediated elevation of cytosolic Ca 2 +. A-C. Changes in [Ca 2 +]i in response to ATP (1 mM) were reported by transiently expressed inverse-pericams in the absence (black traces) or presence of 2-APB (75 uM) and/or procaine (10 mM) (grey traces) as indicated. On average, 2-APB and procaine reduced responses to 71% and 79% of their respective controls. D. Quantification of the effects of 2-APB and procaine by two sample t-tests. 2-APB (n = 25) reduced responses to 71% of respective controls (n = 32). Procaine (n = 36) reduced responses to 79% of respective controls (n = 49). 2-APB and procaine together (26) reduced responses to -0.5% of controls (31). Data show as mean ± SE. n equals number of cells analyzed from at least 3 independent experiments. 4.3.3 Mitochondrial stimulation by IP3R and RyR as a common response to vasopressor agonists. The above results obtained with ATP led us to test whether stimulation of mitochondria by IP3R and RyR is a general phenomenon common to phospholipase lipase de-coupled vasopressor agonists in V S M . [ARG8]-vasopressin (AVP) was used to stimulate SR C a 2 + release and the contributions of IP 3R and RyR to [Ca 2 + ] M T elevations were dissected as for ATP-mediated stimulations. A VP stimulated C a 2 + release from both channel types (Figure 4.3A&B), but the 2-APB/procaine synergy was more obvious in the mitochondrial response to A V P (Figure 4.3C). Statistical analysis showed that 0Ca 2 + had an additive effect with 2-APB but not procaine (Figure 4.3D). Given the similarities in the mitochondrial responses to ATP- and 75 AVP-mediated SR C a 2 + release, we further investigated whether the two PLC-coupled receptors stimulated the same population of mitochondria by analyzing their cross-consumption of aequorin (156). Aequorin luminescence generated in response to A V P or ATP was reduced if cells were first stimulated with the other agonist (Figure 4.4A). Comparison of the raw aequorin luminescence with the calibrated [ C a 2 + ] M T elevations (Figure 4.4B) confirmed that vasopressin could reduce the amount of aequorin available for ATP to activate, without changing the amplitude of the [Ca 2 + ]MT- This indicated that in addition to stimulating similar patterns of IP3R and RyR activation, the two agonists stimulated overlapping populations of mitochondria. 3000 2500 s 2000 r5 1500 + ra o 1000 500 0 2APB/procaine AVP (1(iM) p = 0.27 1 min Figure 4.3. Activation of V"i vasopressin receptors increases [Ca + ] M T by activation both IP3R and RyR. A-C. Average [Ca 2 +]M T responses to AVP show the individual and combined effects of IP3R inhibition with 2-APB (75 uM), RyR inhibition with procaine (10 mM) and depletion of peripheral SR Ca 2 + stores in Ca2+-free solution (0Ca2+, 0.1 mM EGTA). The average control response to AVP is shown as the faint dotted trace. D. The effects of each treatment were normalized against paired control responses with only AVP, and means of normalized responses were compared by ANOVA. P-values from planned Bonferroni comparison are show between specific treatments. 76 A. B • 40001 p = 0.0585 , CO O 2 3 4 » 59 Time (min) 1 ^ 10 11 12 13 Figure 4.4. ATP and AVP stimulate overlapping populations of mitochondria. Averaged mitochondrial responses are shown from cells that were consecutively stimulated with ATP and then AVP (black traces, n = 4) or AVP followed by ATP (grey traces, n = 6). A. Analysis of aequorin luminescence by two sample t tests. B. Analysis of calibrated changes in [Ca 2 + ] M T by two sample t tests. In both cases, average traces are shown with mean ± SE at peaks. 4.3.4 Imaging IP 3 R and RyR distribution in R A S M C . In cells dual-labeled for IP 3R and RyR, both release channels were widely expressed throughout the cell (Figure 4.5A&B), but direct comparison of the two labels indicated several consistent differences in their subcellular distribution. Ryanodine receptors were diffusely expressed throughout the cell with some degree of peri-nuclear accumulation (Figure 4.5B), while high concentrations of IP3R were observed at the cell periphery and in striations running across the cell (Figure 4.5A). These striations were reminiscent of actin stress fibers, which we subsequently labeled with fluorescently-conjugated phalloidin (Figure 4.5C). In merged images of IP3R, RyR and phalloidin, the IP3R, but not RyR, were preferentially distributed with actin filaments (Figure 4.5D). This gross difference in the distribution of IP 3R and RyR at the level of the whole cell provided direct evidence for the localization of I P 3 R and RyR to different SR elements in V M S C . 77 p h a l l o i d i n a n t i - I P 3 R / p h a l l o i d i n Figure 4.5. Subcellular distribution of IP3R, RyR and actin in RASMC. A representative cell is shown in which (A) RyR-1/2 and (B) IP3R were immuno-fluorescently labeled and (C) F-actin was labeled with phalloidin. (D) Merging the IP3R (red) and actin images (green) confirmed that IP3R are extensively associated with actin filaments (yellow). Images representative of 9 cells imaged from >3 independent experiments. IP3R striations were also seen in the absence of phalloidin. N - nucleus. 4.3.5 Localization of IP 3 R and RyR to separate SR elements. With the assistance of digital deconvolution and 3D-reconstruction, analysis of the image volumes at higher magnification revealed that IP3R and RyR were largely localized to separate reticular structures (Figure 4.6A-F). The two SR labels were highly interlaced but often separated by 300 - 400 nm, and analysis of their co-localization showed that the IP3R and RyR were not extensively co-localized (Table 4.1, see methods). Moreover, co-localization events often occurred between perpendicular strands of IP 3R and RyR label (Figure 4.6F), which were likely due to separate elements of the SR passing over and under each other. The IP 3R- and RyR-labeled reticular networks and their minimal overlap are more clearly illustrated in 3D reconstructions (Figure 4.6G-I). This distribution of IP3R and RyR provided direct evidence to support the hypothesis that the two types of release channels are physically localized to separate SR elements, which raised the 78 question of why procaine inhibited the mitochondrial responses to a greater extent than cytosolic responses when 2-APB had a very similar effect on the two responses. To rule out the possibility that this was due to a preferentially localization of RyR with mitochondria, we analyzed mito-RyR and mito-IP3R co-localization. a n t i - R y R a n t i - IP3R c o - l o c a l i z a t i o n Figure 4.6. Co-localization analysis of IP3R & RyR. (A-C) Single deconvolved images at the middle of a representative cell are shown for RyR (A, green), IP3R (B, red) and their co-localization (C, white). The nucleus is prominently labeled by the IP3R anti-body. Scale bar is 5 um. (D-F) Higher magnification of the boxed region from A-C reveals the distinct separation of RyR (green) and IP3R (red) labeled elements on separate reticular networks. (F) Co-localization events (white) often occurred at orthogonal intersections (blue arrows) of RyR andIP3R. The closely inter-twined reticular RyR (G, green) and IP3R (H, red) networks are more evident in 3-D reconstructions. (I) 3-D reconstruction of c-localization events (white) is shown relative to the RyR (transparent green). Table 4.1. Co-localization statistics of IP3R and RyR in dual-labeled cells. I. % co-localization Label A Label B A with B B with A IP3R-I RyR-1/2 19± 1 34 ± 3 79 4.3.6 Visualizing the relationship of IP3R and RyR with mitochondria. We investigated the spatial relationships between the two release channels and mitochondria by dual-labeling the RyR and IP3R in cells expressing mito-targeted GFP. These images revealed extensive association of the mitochondria with IP3R and RyR, but they did not indicate a preferential association of mitochondria with RyR (Figure 4.7A-C). Co-localization analysis confirmed that similar portions of the RyR and IP3R labeled SR were associated with the mitochondria (14 ± 1% and 17 ± 2% respectively, n = 17 cells). To better understand the spatial relationship of the RyR and IP3R with the mitochondria, we generated 3D-models of the mitochondria in each cell and used the co-localization results to paint the portions of the mitochondria that were closely associated with either IP3R (Red), RyR (yellow) or both (white) (Figure 4.7D-F). This method provides a more detailed visualization than can obtained from single images and allowed us to observe that most of the mitochondria labeled voxels were associated with an IP3R or RyR label (41 ± 2% and 34 ± 3% of mitochondrial surface, respectively). In fact, many mitochondria appeared to be associated with both SR labels. anti-RyR anti-IP3R co-localization Figure 4.7. Mitochondrial association with IP3R and RyR. (A-C) Mitochondria are shown as maxim intensity projections (green), upon which the co-localization of RyR (yellow), IP3R (red) or both (white) is painted. (D-F) At higher magnification the association it can be seen that most mitochondria associate with RyR (D, yellow) and or IP3R (E, ref). (F) For mitochondria associated with both RyR (yellow) and IP3R (red), co-localization (white) of the two SR labels is common. 80 4.3.7 Preferential co-localization of IP3R and RyR near mitochondria. Based on these observations and previous reports that IP3R and RyR can be found in high concentrations near mitochondria, we hypothesized that H^R-RyR co-localization occurs more frequently for SR elements neighbouring the mitochondrial surface than in the bulk cytosol. On a cell by cell basis, we statistically analyzed the co-localization between the I P 3 R , RyR and mitochondria (see Table 4.2). The I P 3 R - and RyR-labeled voxels that were co-localized with the mitochondria were then analyzed for co-localization with each other to isolate the JT^R-RyR co-localization events associated with the mitochondria (Figure 4.7C&F). This analysis revealed that co-localization was significantly more frequent (1.4-fold) for I P 3 R and RyR specifically associated with the mitochondria than the average JT^R-RyR co-localization across the entire cell. Table 4.2. Co-localization statistics of I P 3 R and RyR with mitochondria. I. % co-localization Label A Label B A with B B with A mitochondria IP3R-I 41 ± 2 (IPjRmito) 14 ± 1 mitochondria RyR -1 /2 33 ± 3 (RyRmito) 17 ± 2 IP3R-I RyR -1 /2 33 ± 3 a 45 ± 2 b IP3Rmito-C RyRmito- c 45 ± 3 a 56 ± 3 b II. % of mitochondrial voxels with D ^ R - R y R co-localization 1 9 ± 2 d a & b For values with matched superscripts paired t-test gives p < 0.0001, n = 16-18 pairs, c mito-IP3R & m i t o - R y R co-localization events were co-localized with each other. d = (IPaRnto) X (IPjRmito With R y R m t o ) = (RyRmito) X (RyRmito With IPjRmUo) This mitochondria-associated IT^R-RyR co-localization was compared with that which might be expected by random co-incidence of the two labels given the total number of mitochondrial voxels and percent of those voxels that were also labeled for the I P 3 R or RyR. If the IP 3R and RyR voxels were distributed across the mitochondria completely randomly, then the product of the percent of mitochondrial voxels that co-localized with I P 3 R (41 ± 2%) and RyR (33 ± 3%) should predict the percent of the mitochondrial voxels in which H^R-RyR co-localize (14 ± 2%, mean of 16 cells). However, the mitochondria associated IT^R-RyR co-localization measured from the images was 19 ± 2% (mean of same 16 cells), which was significantly (1.35 times) more frequent than would predicted for a purely randomly con-incidence of a similar number of voxels (paired t-test, p < 0.0001, n = 16 cells). However, fluorescently labeled voxels do not distribute purely randomly. Rather, they cluster into the shape of the labeled structure or into small kernels in the case of sub-resolution structures such as ion channels. We created an algorithm to generate 3D-arrays (representing the mitochondrial volume) and calculate the co-localization of randomly placed "IP 3R" and "RyR" voxels (see methods). When individual IP 3R 81 and RyR voxels (41% and 33% of the model volume, respectively) were randomly distribute, the predicted co-localization was simply the product of the percent of I P 3 R - and RyR-labeled voxels (-14%) (Figure 4.8A). However, when voxels were clustered into simple 7-voxel star-shaped kernels or 27-voxel cube-shaped kernels, the extent of "random" co-localization increased dramatically, with the actual shape of the kernel having little effect (Figure 4.8A&C). The extent of random co-localization was also sensitive to the cross-sectional area of the modeled volume. When this cross-section approximated the size of imaged mitochondria, the random co-localization of clustered voxels closely matched the mitochondria-associated rP 3R-RyR co-localization measured from our images (1.35 - 1.45-fold greater than for individually distributed voxels). Thus the IP 3R-RyR co-localization associated with the mitochondria was likely no greater than could be expected by random co-incidence given the density of the two labels within the mitochondrial volume. We must stress however, that this issue is distinct from the observation that the incidence of rP 3R-RyR co-localization near the mitochondria was greater than the average incidence across the entire cell. 82 0.25-1 0.20-X a> ind 0.15-o o 0.10-o o 0.05-0.00-41%IP3R, 33% RyR V 27-voxel cube-shaped cluster • 7-voxel star-shaped cluster • random distrubttion of voxels 4? B " Q • -B • -Q-0.06-1 0.05-0.04-0.03-0.02-0.01 0.00 20%IP3R, 15% RyR V • •Q • -O —• - • -Q—-o o o u •o Q) S V) 3 O 1.8 1.7 1.6 8 1-5 E 1.4 O "O 1.3 2 1.2 1.1 1.0 S 1.8-1 1.7-1.6-1.5-1.4-1.3-1.2-1.1-1.0-V V • V • tS>+ S V • array shape array shape Figure 4.8. Modeling the effect of voxel clustering on random co-localization of mitochondria associated IP3R and RyR. Co-localization was assessed after randomly filling two computer-generated 3D arrays of 115,00 voxels (i.e., mean mitochondrial volume analyzed) with IP3R and RyR at densities equal to (A,C) or half of that which was measured during image analysis (B, D). (A, B) Co-localization was assessed when "IP3R" and "RyR" voxels were randomly distributed as individual voxels, 7-voxel star-shaped clusters or 27-voxel cubic clusters. The dotted line shows the "random" co-localization predicted by multiplying the fraction of voxels labeled for IP3R and RyR. (C, D) Co-localization of clustered voxels was normalized against co-localization of individually distributed voxels. Error associated with repeated simulations is less than 0.1%. 83 A.i A.i 1000n 75<H I-S 500 O 2 5 0 control )6^^^ Cocaine 1 min 4-CmC (500 nM) B . i 1000-, s 800 600 +" CM 400 o 200 0Na+ + procaine (10 mM) 1 min Ci. controh- proc. Figure 4.9. Assessing the selectivity of procaine for RYR versus mitochondria. A) Representative traces showing procaine-mediated inhibition of the [Ca 2 + ] M T elevation upon activation of RyR with 4-chloro-/?i-cresol (4-CmC). Aii) The effect of procaine was significant as determined by a two-sample t-test (n = 4 control, 6 procaine). B) Iso-osmotic replacement of extracellular Na+ with n-methyl-D-glucamine (145 mM) causes Ca 2 + influx via the Na+/Ca2 +-exchanger which is taken up by mitochondria. Procaine did not inhibit the mitochondrial Ca 2 + uptake in response to 0Na+ (n = 3 control, 3 procaine). C) When IP3R are inhibited by 2-APB, ryanodine (lOOpM, 30 min) further reduces the mitochondrial response to ATP-mediated SR Ca + release. Bii) The effect of ryanodine was assessed by a paired t-test (n = 9 pairs). control + ryan. 4.4 DISCUSSION: Ca + signaling in smooth muscle is highly dependent on the SR, which is composed of membrane bound elements exhibiting heterogeneous expression of C a 2 + pumps, channels and buffer proteins (136;157;164;181). In this investigation we address a long standing question in smooth muscle physiology, which is whether RyR and I P 3 R are functionally and structurally localized to the same or separate elements of the SR. In vascular smooth muscle this issue is complicated in 1) that the relative contribution of RyR and I P 3 R to C a 2 + signaling varies between different blood vessels (176) and 2) that the fundamental principles governing the interaction of IP3R and RyR in V S M are not completely elucidated. Here we present functional and imaging data indicating that IP3R and RyR are localized to separate elements of the SR in RASMC. Moreover, we propose that the extensive mitochondrial association with the SR locally increases the juxtaposition of I P 3 R and RyR on separate SR elements such the IP 3R-RyR cross-talk • 2+ • • differentially influences cytosolic and mitochondrial Ca signaling. 84 4.4.1 Functional Localization of IP3R and RyR in VSM. Despite the fact that RyR can be widely distributed through V S M (30;31), they are best known for their peripheral juxtaposition near and modulation of Ca2+-sensitive CT- and K+-channels on the adjacent plasmalemma (254). 9 + In some V S M the peripheral SR also interacts with the N C X to mediate preferential Ca 94-extrusion by the N C X over the plasmalemmal Ca ATPase (3; 162). We have previously shown this SR-NCX coupling to cause the N C X to extrude C a 2 + released from a specific sub-compartment of the SR in the absence of extracellular C a 2 + (220), and our current data indicate that this SR sub-compartment releases C a 2 + via RyR, rather than IP3R (also see (138)). This conclusion is based on the observation that 2-APB and 0Ca 2 + treatment additively reduced the [Ca 2 +]MT elevation in response to ATP and A V P . In contrast, removal of extracellular C a 2 + did not have an additive effect with procaine and caused much less inhibition of the [Ca 2 + ] M T elevations than did procaine. This indicated that mitochondria took-up C a 2 + released from RyR-associated SR throughout the cell, while only a peripheral portion of the RyR-associated SR was directly coupled to the N C X . This latter point is corroborated by direct observation of widespread RyR distribution throughout the cell, consistent with previous investigations in V S M (30;49;136). Further more, this underscores an important principle that the SR is composed of functionally distinct sub-domains (85). The IP3R can also be found throughout the cell. Immuno-gold labeling of the IP3R in aortic smooth muscle revealed gold particles localized to both central and peripheral SR (164). Immunocytochemistry has revealed a uniform distribution of IP3R in a fine network of deep and peripheral SR in portal vein myocytes (31), with some evidence for isoform specific patterns of localization (77;164;224;244). We also observed that TP3R were localized to a fine network in both the peripheral and central SR, and we found that IP3R were highly concentrated along actin stress fibers traversing the cell, which was consistent with previous reports (90;216). In terms of structure-function relationships, this pattern of IP3R distribution suggests that IP3R may be strategically localized to regulate myofilament contraction. Furthermore, despite extensive localization of IP3R to the sub-plasmalemmal SR, the aequorin data suggests that IP3R do not interact with the N C X , which emphasizes the notion that functional coupling of the SR and N C X requires highly localized interactions (23; 176). This observation and the strikingly different patterns of IP3R and RyR distribution at the level of gross subcellular organization (figure 4.5) strongly indicate that the IP3R and RyR are localized to separate elements of the SR. 85 To pursue this issue in greater detail, we have provided a detailed co-localization analysis of IP3R and RyR in vascular SMC. At high magnification, the distribution of IP3R and RyR in fine reticular networks was consistent with previous reports (30;31;88).By examining the LP3R and RyR in 3-D volumes of digitally deconvolved images at high magnification, we found that the IP3R and RyR labels were largely distributed on separated but interlaced reticular networks that run throughout the cell. Statistical analysis confirmed that IP3R and RyR were not highly co-localized throughout the cytosol (see table 4.1), but without a reliable marker of the cell membrane it is difficult to estimate the density of the two labels within the cell volume and therefore to mathematically assess whether the extent of co-localization was greater than would be expected by chance given the density and clustering of the two SR labels. We can however assume the co-localization analysis likely overestimates the incidence of co-localization since the width of the SR (-50-100 nm in diameter) would be exaggerated by the limited resolution of optical microscopy (full-width half-max: -300 nm x-y, -700 nm axial). Moreover, co-localization often occurred at orthogonal intersections of the IP3R/RyR labels indicating that some co-localization events were due to separate elements of the SR passing over and under each other at distances below the axial resolution of the deconvolved volumes. Overall the evidence strongly argues in favour of IP3R and RyR being localized to closely neighbouring but separate SR elements with distinct C a 2 + stores. While we cannot exclude the possibility that IP3R and RyR labeled SR elements might be joined by unlabeled sections of SR, our current evidence is consistent with functional studies from a variety of V S M , which are considered below in the context of functional interactions between IP3R and RyR observed in this study. 4.4.2 Functional interaction of IP3R and RyR. Three basic models have been presented for the functional interaction of IP3R and RyR in smooth muscle (153). The two channels could communicate 1) through a shared SR lumen, 2) through an interaction of SR membrane proteins, and or 3) through diffusible cytosolic factors (such as Ca ). If we were to consider only the changes in [Ca 2 + ] j and the synergistic effect of 2APB and procaine, we might conclude that the IP3R and RyR release C a 2 + from a single store such that inhibition of either type of release channel would increase the C a 2 + that was available for the other channel to release. Indeed, this model is consistent with the existing data from most types of smooth muscle (153). However, when we also consider the distinct patterns of IP3R and RyR localization discussed above and the near lack of synergy between 2APB and procaine on SR-dependent changes in [ C a 2 + ] M T reported here, then we would conclude that IP3R and RyR in these cultured rat aortic smooth muscle cells 86 release Ca from separate SR sub-compartments and interact via allosteric modulation of channel gating by locally generated [Ca ]\ gradients. The open probability of IP3R and RyR are regulated by the [Ca 2 +] in the neighbouring cytosol and the SR lumen. Type 1 IP3R (the predominant V S M isoform) have an EC50 of -300 nM Ca 2 + , and C a 2 + exerts an inhibitory effect at concentrations greater than - l u M (26; 183). In contrast, Ca2+-induced activation of RyR in situ is triggered by [Ca 2 +] > - l u M and plays an important role in the stimulation of rapid elevations of [ C a 2 + ] M T (108;160). This threshold is reduced experimentally by 4-chloro-m-cresol or caffeine and endogenously by cyclic-ADP ribose, which likely accounts for the elevation of [ C a 2 + ] M T elicited by both ATP and A V P in the presence of 2-APB. In rat portal vein, photolysis of caged-LP3 activates IP3R at global [Ca 2 +]; < 200 nM, suggesting that the micromolar [Ca2+]j threshold to activate RyR is restricted to cytosolic microdomains between neighbouring IP 3R and RyR (9;30;31;114;160). This local cross-talk is required for the initiation of both agonist-induced and spontaneous cytosolic C a 2 + waves, in which IP3R provide the trigger C a 2 + to stimulate RyR and convert C a 2 + sparks into waves (87;134;199). Along the C a 2 + wave front, propagating activation of C a 2 + release units has been resolved at -1-2 pm separation, but the trigger C a 2 + released from IP3R has yet be resolved from the initiating C a 2 + spark (87). Here we found that IP3R and RyR in deconvolved images of fixed cells were often separated by less than 400 nm, which provides a structural basis for the technical difficulty of imaging the local IT^R-mediated [Ca2 +]j elevation that is thought to convert C a 2 + sparks (averaging 1.5-2 um in width) into C a 2 + waves (9;88;114). Mitochondrial aequorin provides an alternative and sensitive, i f indirect, indicator of localized SR C a 2 + release (191; 192), and it has previously been used in aorta smooth muscle cells to demonstrate that IP3R and RyR are functionally coupled with mitochondria (160;220). In one of these studies it was concluded that mitochondria did not detect IP3R-RyR cross-talk in R A S M C because [ C a 2 + ] M T responses to IP3 and caffeine persisted in the presence of ryanodine and heparin, respectively (160). However, in similar rat aorta SMCs we find that mitochondria do detect IT^R-RyR cross-talk based on: 1) the synergistic inhibition of ATP-mediated elevations of [ C a 2 + ] M T and 2) the preferential co-localization of EP3R and RyR near mitochondria. Moreover, novel insights into localized IP^R-RyR crosstalk can be drawn from the differences in the extent of 2-APB/procaine synergy on the mitochondrial versus cytosolic responses. Upon C a 2 + release into the cytosolic microdomain between the SR and closely apposed mitochondria 87 the local [Ca 2 +]j is thought to exceed 10 p M (55), which facilitates activation of the mitochondrial C a 2 + uniporter and mitochondrial C a 2 + buffering (98; 192). This same local elevation of [Ca 2 +]i will also cause IP3R facing the mito-SR junctional space to be more rapidly inactivated relative to IP3R localized elsewhere on the SR. Consistent with this model, we found that IP3R alone (i.e., in the presence of procaine) generated -80% of the parallel ATP-mediated [Ca 2 +]i elevations, and only -55% of the normal mitochondrial response. In contrast, the elevation of [Ca 2 +]i by RyR alone (i.e., in the presence of 2APB) was well correlated with the parallel elevation in [Ca 2 +]MT (both -65% of respective controls), which is consistent with reports that inactivation of RyR in V S M is governed by time-dependent inactivation more than by local [Ca 2 +]; elevations or decreases in [Ca2 +]sR(9;88). Thus, the pronounced 2-APB/procaine synergy on [Ca 2 +]i elevations likely indicates that C a 2 + release from IP3R normally activates neighbouring RyR, from which C a 2 + release causes feedback inhibition of the IP3R (30;31;114). This provides a simple explanation as to why IP3RS alone (in the presence of procaine) were able to generate of -80% of normal cytosolic C a 2 + elevations. At the Mito-SR junctions, however, impaired diffusion of C a 2 + released from IP3R likely causes sufficient auto-inhibition such that RyR do not cause detectable inhibition of IP3R. However, IP3R-RyR co-localization is significantly (1.4-times) more likely near mitochondria, and IP3R alone were only capable of generating -65% of normal [Ca 2 +]MT responses to ATP or A V P . Thus, we propose that IP3R-2+ mediated activation of RyR plays a critical role in generating physiological [Ca ] M T signals in vascular smooth muscle cells, which is currently under further investigation (160). The model described above is based on the comparison of aequorin and pericam responses and the selectivity of the pharmacological agents used, which warrants consideration. Since aequorin and the inverse pericam have different mathematical relationships to changes in [Ca 2 +], the effects of 2-APB and procaine were normalized against control responses. 2-APB inhibited the cytosolic and mitochondrial responses to similar extents suggesting that the relative response of the two reporters can be reliably compared. Since the transient phase of ATP- and AVP-mediated responses is primarily due to SR C a 2 + release (220) and 2-APB and 0Ca 2 + additively reduced the mitochondrial responses, the action of 2-APB as used here was likely on IP3R rather than store-operated C a 2 + channels (34). While procaine is commonly used as a Na + -channel blocker in non-excitable cells it has been used to inhibit RyR in muscle (108). Procaine did not directly impair mitochondrial C a 2 + uptake as the [Ca 2 + ] M T elevation in response to N C X -88 mediated Ca influx was not affected by procaine (Figure 4.9), and the mitochondrial effects of procaine were mimicked by ryanodine (100 uM) (Figure 4.9). 4.4.3 Spatial association of IP3R and RyR near mitochondria. Important and unresolved physiological questions regarding rapid C a 2 + transfer between the SR/ER and mitochondria are whether SR/ER-mitochondria associations occur at specific sites on the opposing membranes and whether the C a 2 + handling machinery is localized to these sites. Immuno-labeled electron micrographs show IP3R and RyR on SR/ER elements neighbouring mitochondria in V M S consistent with our current imaging data (136; 164), but whether the concentration of IP3R and RyR is higher at the SR-mito junctions relative to the rest of the SR remains to be quantified in V S M (see (201)). Since the co-localization analysis used here disregards intensity information after the images are thresholded, we cannot draw inferences on the local concentrations of IP3R or RyR on the SR, but we can draw inferences on localized concentrations of IP3R or RyR expressing SR. Specifically, co-localization of IP3R (-17%) and RyR (14%) voxels associated with the mitochondria was 1.4-times more frequent than the average IF^R-RyR co-localization throughout the cell. This could be explained by two probable mechanisms: 1) both SR Ca release channels could be preferentially associated with common sites on the mitochondria, or 2) an increased density of SR near the mitochondria could increase the random co-incidence of the two labels. In smooth muscle (vascular and other), mitochondria are often found wrapped in SR (63; 176). Quantification of this phenomenon from electron micrographs revealed that 82% of mitochondria are completely enwrapped in SR and that 48%. of the average O M M is within 30 nm of the SR in resting tracheal SMC (57). Given the limits of optical resolution, this enwrapping of the mitochondria with SR would be manifest as an increased density or concentration of LP3R and RyR labeled SR neighbouring the mitochondria. This provides an ultra-structural basis for the increased co-localization of IP3R and RyR near the mitochondria assuming that overlap of IP3R and RyR labeled elements of the SR would be more likely where the concentration of SR is increased. However, evidence from a number of cell types indicates that SR/ER-mitochondria associations occur at specific sites on the SR and outer mitochondrial membranes (OMM). In M D K cells a sub-compartment of the ER containing the autocrine-motility factor receptor (AMF-R) preferentially associates with mitochondria in a Ca -dependent manner (250), while voltage-dependent anion channels (VDAC), the primary route for 89 Ca permeation across the O M M , appear to be preferentially distributed at close contacts with the ER in HeLa cells (185). Thus it was necessary to determine in our aortic smooth muscle cells whether IT^R-RyR co-localization events associated with the mitochondria occurred more frequently than might be expected by chance. We developed a simple, but robust, algorithm that randomly distributes two labels within a specified volume and calculates their co-localization. This algorithm indicated that co-localization of two labels that were "randomly" distributed as clusters (9 or 27 voxels in size) was much higher than when the labels were distributed as individual voxels. When the minor cross section of the model mitochondria approximated that of the GFP-labeled mitochondria, the random co-localization closely matched the measured incidence of mito-associated IP^R-RyR co-localization, which indicated that the incidence of IP 3R-RyR co-localization near mitochondria was not greater than could be expected by chance. Thus the preferential co-localization of IP3R and RyR neighbouring the mitochondria was not a direct indication of an association with common loci on the mitochondria, but was likely due to an increased concentration of SR near the mitochondria. 4.4.4 Conclusion: In summary, the work presented herein demonstrates a method to assess the localization of IP3R and RyR to separate SR sub-compartments, which takes into account both structural and functional evidence. Our results further support the functional relevance of localized interactions between mitochondria, IP3R and RyR, and demonstrate differential roles of IP3R-RyR crosstalk in the generation of mitochondrial and global C a 2 + signals in vascular smooth muscle cells. While the separation of IP3R and RyR should be further assessed with immuno-electron microscopy, our current findings provide as starting point for functional and electron microscopy studies to further investigate the dynamic nature mito-SR interactions during stimulation of smooth muscle. 90 CHAPTER V GENERAL CONCLUSIONS & RECOMMENDATIONS FOR FUTURE WORK Vascular smooth muscle is essential to the regulation of blood pressure and responds to the neuro-humoural signals that control regional blood flow. It must be able to respond rapidly to the moment-to-moment metabolic demands of specific tissues, while maintaining global vascular tone, which requires constant and exquisite control of cytosolic [Ca 2 +]i. This thesis focuses on the notion that intracellular C a 2 + signals are spatially and temporally partitioned, such that C a 2 + can simultaneously modulate contraction and processes such as gene transcription and oxidative metabolism. Great effort has been devoted to understanding how the subcellular architecture of the SR contributes to this partitioning of cytosolic C a 2 + gradients. By comparison, mitochondria have received much less attention than the P M and SR regarding the establishment and maintenance of local C a 2 + gradients and feedback mechanisms in V S M , despite the fact that mitochondria are able to rapidly sequester Ca 2 + . This thesis presents novel information regarding 9 + the nature of agonist-mediated mitochondrial [Ca ] elevations specifically in V S M and exploits mitochondria as indicators of localized cytosolic C a 2 + elevations. 5.1 IMPLICATIONS OF BASAL C a 2 + ENTRY In recent years basal C a 2 + influx into V S M C was attributed to the basal flicker of at last two distinct C a 2 + permeable channels. The pharmacological dissection of the leak reported herein 9 + suggests that these channels are likely the same channels that are activated by SR Ca -depletion and or PLC-coupled receptor activation. Moreover, we provide numerous indirect lines of evidence that this basal C a 2 + influx may result from basal SR C a 2 + release (from both RyR and IP 3R) and or basal activity of PLC. 9 + Based on our estimate of the basal rate of Ca influx, it appears as though V S M C are capable of maintaining low [Ca 2 +]j despite considerable C a 2 + influx at rest. Given that [Ca 2 + ] S U D pM was higher than most estimates of bulk [Ca 2 +]i at rest, our current findings further support the 9 + notion that the superficial SR and mitochondria are able to establish a standing Ca gradient at the periphery of the cell. Two important implications of such a standing gradient are that: 1) basal C a 2 + influx is 9 + taken up by mitochondria and is a critical determinant of basal [Ca ]MT, and 2) that studies of store-operated Ca2+-entry (SOCE) induced by SERCA inhibition must take into consideration 91 that loss of SR-mediated C a 2 + buffering will greatly confound potential increases in [Ca2+]j due to SOCE. 5.2 MITOCHONDRIA AND LOCAL C a 2 + EVENTS 5.2.1 Mitochondria and plasmalemma Na+/Ca2+-exchange. We have utilized mitochondria and other targeted C a 2 + indicators to examine the interaction of the mitochondria, SR and N C X . We found that rapid buffering of C a 2 + entering the cytosol from the extracellular space by the SR and mitochondria can limit the detection of such C a 2 + influx by fluorescent cytosolic C a 2 + indicators. We observed that mitochondria readily buffer C a 2 + entry due to reversal of the plasmalemmal N C X , and that reverse-mode N C X mediates considerable Ca 2 + -entry during stimulation of V S M C with PLC-coupled agonists. These observations were consistent with the hypothesis that reverse-mode N C X is functionally coupled with the agonist-induced opening non-selective cation channels (likely TRPs). This represents an important paradigm shift in the study of agonist-induced Ca -influx and has met with considerable resistance. Not only do our current results support the physiological existence of such a mechanism, but they suggest that the activation of non-selective cation channels may provide an important source of Na + to permit the up-regulation of mitochondrial Na+/Ca2+-exchange during agonist stimulation (else where suggest to occur in endothelial cells). Concrete demonstration of such a mechanism will benefit from; 1) detailed characterization of changes in intracellular Na + , and 2) in situ characterization of TRPC proteins as constituents of C a 2 + channels or non-selective cation channels. 5.2.2 Agonist-mediated stimulation of mitochondria by IP3R & RyR. Mitochondria have proven to be excellent indicators of localized C a 2 + events in a number of cells. Here, pharmacological dissection of agonist-induced [Ca 2 +]MT elevations revealed that PLC-coupled agonists evoke C a 2 + release from and mitochondrial stimulation by both IP3R and RyR, which appear to release C a 2 + from separate sub-compartments of the SR. 3-Dimensional analysis of cells in which the mitochondria, IP3R and RyR were fluorescently labeled suggested that the extensive association of mitochondria and SR influenced the stimulation of mitochondria by IP3R and RyR. Previous studies of mitochondrial C a 2 + signaling in V S M have largely relied on mitochondrial accumulation of the cationic, Ca2+-sensitive dye Rhod-2 and have suggested that 92 SR Ca release results in prolonged elevations of [Ca ]MT- In contrast, our studies with mito-aequorin indicate that agonist-induced elevations of [Ca ]MT are transient and completely return to resting levels due to up-regulation of mitochondria Ca2+extrusion. Given that the kinetics of the [Ca 2 + ]MT elevation reported with aequorin closely resembled those of the [Ca 2 +]; elevations (reported with fura-2, fluo-3 and cytosolic pericams), it is unlikely that the transient nature of the [Ca 2 + ]MT was due to excessive consumption of aequorin. Rather, this is likely because 1) only sub-population of M T are exposed to the localized supra-micromolar [Ca ] elevations caused by C a 2 + influx, and 2) mitochondrial C a 2 + extrusion is up-regulated to compensate for the increased C a 2 + uptake. Thus we would conclude that transient elevations of [Ca 2 + ]MT likely represent the physiological mode of mitochondrial stimulation in V S M . 5.3 CLOSING COMMENTS The impetus behind these studies is ultimately to gain a comprehensive understanding of C a 2 + signaling in smooth muscle of blood vessels; healthy and diseased, young and old. The contributions made in this thesis to further defining the nature of [Ca ] M T elevations resulting' from "physiological" stimulation of V S M bring us that much closer to being able to identify the nature of the [Ca 2 + ]MT elevations that lead to activation of apoptotic cell death when C a 2 + homeostasis is compromised. With the proliferation of targeted indicators and the development of technologies to deliver intact proteins into living tissue, our ability to address this issue is primarily limited by: 1) our creativity in designing new probes and 2) our ability to extract meaning.full data from the images that we are now readily able to capture. While cultured cells provide a ready means to develop principles, lessons from previous studies underscore the heterogeneity exhibited in the behaviour of mitochondria in different blood vessels and tissues. Thus in closing, we emphasize the importance of characterizing the detailed behaviour and role of mitochondria in the context of the specific tissue under investigation. Such studies should culminate in the unraveling of the relationship between mitochondrial C a 2 + transport and smooth muscle heterogeneity. 93 B I B L I O G R A P H Y 1. (1995) Handbook of Biological Confocal Microscopy, 2nd Ed., Plenum Press, New York 2. Aalkaer C and Nilsson H (2005) Vasomotion: cellular background for the oscillator and for the synchronization of smooth muscle cells. Br. J Pharmacol. 144 (5), 605-616 3. Aaronson P and van Breemen C (1981) Effects of sodium gradient manipulation upon cellular calcium, 45Ca fluxes and cellular sodium in the guinea-pig taenia coli. J. Physiol 319,443-461 4. Adams DJ, Barakeh J, Laskey R, and van Breemen C (1989) Ion channels and regulation of intracellular calcium in vascular endothelial cells. FASEB J. 3 (12), 2389-2400 5. Alderton JM and Steinhardt RA (2000) Calcium influx through calcium leak channels is responsible for the elevated levels of calcium-dependent proteolysis in dystrophic myotubes. J. Biol. Chem. 275 (13), 9452-9460 6. Allen BG and Walsh MP (1994) The biochemical basis of the regulation of smooth-muscle contraction. Trends Biochem. Sci. 19 (9), 362-368 7. Alonso MT, Montero M, Carnicero E, Garcia-Sancho J, and Alvarez J (2002) Subcellular Ca(2+) dynamics measured with targeted aequorin in chromaffin cells. Ann. N. Y. Acad. Sci. 971:634-40., 634-640 8. Archer SL, Weir EK, Reeve HL, and Michelakis E (2000) Molecular identification of 02 sensors and 02-sensitive potassium channels in the pulmonary circulation. Adv. Exp. Med. Biol. 475:219-40., 219-240 9. Arnaudeau S, Boittin FX, Macrez N, Lavie JL, Mironneau C, and Mironneau J (1997) L-type and Ca2+ release channel-dependent hierarchical Ca2+ signalling in rat portal vein myocytes. Cell Calcium 22(5), 399-411 10. Arnaudeau S, Kelley WL, Walsh JV, Jr., and Demaurex N (2001) Mitochondria recycle Ca(2+) to the endoplasmic reticulum and prevent the depletion of neighboring endoplasmic reticulum regions. J Biol. Chem. 276 (31), 29430-29439 11. Arnon A, Hamlyn JM, and Blaustein MP (2000) Na+ entry via store-operated channels modulates Ca2+ signaling in arterial myocytes. Am. J. Physiol Cell Physiol 278 (1), C163-C173 12. Arnon A, Hamlyn JM, and Blaustein MP (2000) Ouabain augments Ca2+ transients in arterial smooth muscle without raising cytosolic Na+. Am J Physiol Heart Circ Physiol 279 (2), H679-H691 13. Babcock DF and Hille B (1998) Mitochondrial oversight of cellular Ca2+ signaling. Curr. Opin. Neurobiol. 8 (3), 398-404 14. Babiychuk VS, Draeger A, and Babiychuk EB (2000) Smooth muscle actomyosin promotes Ca2+-dependent interactions between annexin VI and detergent-insoluble glycosphingolipid-enriched membrane domains. Acta Biochim. Pol. 47 (3), 579-589 15. Bai N, Lee HC, and Laher I (2005) Emerging role of cyclic ADP-ribose (cADPR) in smooth muscle. Pharmacol. Ther. 105 (2), 189-207 16. Balaban RS, Bose S, French SA, and Territo PR (2003) Role of calcium in metabolic signaling between cardiac sarcoplasmic reticulum and mitochondria in vitro. Am J Physiol Cell Physiol 284 (2), C285-C293 17. Barrett-Jolley R and Davies NW (1997) Kinetic analysis of the inhibitory effect of glibenclamide on KATP channels of mammalian skeletal muscle. J Membr. Biol. 155 (3), 257-262 18. Barron JT, Gu L, and Parrillo JE (1999) Relation of NADH/NAD to contraction in vascular smooth muscle. Mol Cell Biochem. 194 (1-2), 283-290 19. Barron JT, Gu L, and Parrillo JE (1998) Malate-Aspartate Shuttle, Cytoplasmic NADH Redox Potential, and Energetics in Vascular Smooth Muscle* 1. Journal of Molecular and Cellular Cardiology 30 (8), 1571-1579 20. Basset O, Boittin FX, Dorchies OM, Chatton JY, van BC, and Ruegg UT (2004) Involvement of inositol 1,4,5-trisphosphate in nicotinic calcium responses in dystrophic myotubes assessed by near-plasma membrane calcium measurement. J Biol. Chem. 279 (45), 47092-47100 21. Beech DJ, Muraki K, and Flemming R (2004) Non-selective cationic channels of smooth muscle and the mammalian homologues of Drosophila TRP. J. Physiol 559 (Pt 3), 685-706 22. Bernardi P (1999) Mitochondrial Transport of Cations: Channels, Exchangers, and Permeability Transition. Physiol. Rev. 79 (4), 1127-1155 23. Berridge MJ, Bootman MD, and Roderick HL (2003) Calcium signalling: dynamics, homeostasis and remodelling. Nat. Rev. Mol Cell Biol. 4 (7), 517-529 94 24. Bers, D. M. (2001) Exitation-Contraction Coupling and Cardiac Contractile Force, 2nd Ed., Kluwer Academic Publishers, 25. Bezprozvanny I and Ehrlich BE (1995) The inositol 1,4,5-trisphosphate (InsP3) receptor. J Membr. Biol. 145 (3), 205-216 26. Bezprozvanny I, Watras J, and Ehrlich BE (1991) Bell-shaped calcium-response curves of Ins(l,4,5)P3- and calcium-gated channels from endoplasmic reticulum of cerebellum. Nature 351 (6329), 751-754 27. Blaustein MP (1977) Sodium ions, calcium ions, blood pressure regulation, and hypertension: a reassessment and a hypothesis. Am J Physiol 232 (5), C165-C173 28. Blaustein MP (1993) Physiological effects of endogenous ouabain: control of intracellular Ca2+ stores and cell responsiveness. Am. J Physiol 264 (6 Pt 1), C1367-C1387 29. Boitier E, Rea R, and Duchen MR (1999) Mitochondria exert a negative feedback on the propagation of intracellular Ca2+ waves in rat cortical astrocytes. J Cell Biol. 145 (4), 795-808 30. Boittin FX, Coussin F, Macrez N, Mironneau C, and Mironneau J (1998) Inositol 1,4,5-trisphosphate- and ryanodine-sensitive Ca2+ release channel-dependent Ca2+ signalling in rat portal vein myocytes. Cell Calcium 23 (5), 303-311 31. Boittin FX, Macrez N, Halet G, and Mironneau J (1999) Norepinephrine-induced Ca(2+) waves depend on InsP(3) and ryanodine receptor activation in vascular myocytes. Am. J Physiol 277 (1 Pt 1), C139-C151 32. Bolton TB, Gordienko DV, Pucovsky V, Parsons S, and Povstyan O (2002) Calcium release events in excitation-contraction coupling in smooth muscle. Novartis. Found. Symp. 246,154-168 33. Bolton TB, Prestwich SA, Zholos AV, and Gordienko DV (1999) Excitation-contraction coupling in gastrointestinal and other smooth muscles. Annu. Rev. Physiol 61, 85-115 34. Bootman MD, Collins TJ, Mackenzie L, Roderick HL, Berridge MJ, and Peppiatt CM (2002) 2-aminoethoxydiphenyl borate (2-APB) is a reliable blocker of store-operated Ca2+ entry but an inconsistent inhibitor of InsP3-induced Ca2+ release. FASEB J 16 (10), 1145-1150 35. Boulay G, Zhu X, Peyton M, Jiang M, Hurst R, Stefani E, and Birnbaumer L (1997) Cloning and expression of a novel mammalian homolog of Drosophila transient receptor potential (Trp) involved in calcium entry secondary to activation of receptors coupled by the Gq class of G protein. J. Biol. Chem. 272 (47), 29672-29680 36. Bowser DN, Petrou S, Panchal RG, Smart ML, and Williams DA (2002) Release of mitochondrial Ca2+ via the permeability transition activates endoplasmic reticulum Ca2+ uptake. FASEB J 16 (9), 1105-1107 37. Brini M (2003) Ca2+ signalling in mitochondria: mechanism and role in physiology and pathology. Cell Calcium 34 (4-5), 399-405 38. Brown GC (1992) Control of respiration and ATP synthesis in mammalian mitochondria and cells. Biochem. J 284 (Pt 1), 1-13 39. Busselberg D, Piatt B, Michael D, Carpenter DO, and Haas HL (1994) Mammalian voltage-activated calcium channel currents are blocked by Pb2+, Zn2+, and A13+. J. Neurophysiol. 71 (4), 1491-1497 40. Camello C, Lomax R, Petersen OH, and Tepikin AV (2002) Calcium leak from intracellular stores— the enigma of calcium signalling. Cell Calcium 32 (5-6), 355-361 41. Campbell JD and Paul RJ (1992) The nature of fuel provision for the Na+,K(+)-ATPase in porcine vascular smooth muscle. J. Physiol 447, 67-82 42. Casteels R and Droogmans G (1981) Exchange characteristics of the noradrenaline-sensitive calcium store in vascular smooth muscle cells or rabbit ear artery. J Physiol 317, 263-279 43. Chakraborti T, Das S, Mondal M, Roychoudhury S, and Chakraborti S (1999) Oxidant, Mitochondria and Calcium: An Overview. Cellular Signalling 11 (2), 77-85 44. Charuk JH, Pirraglia CA, and Reithmeier RA (1990) Interaction of ruthenium red with Ca2(+)-binding proteins. Anal. Biochem. 188 (1), 123-131 45. Chen Q, Vazquez EJ, Moghaddas S, Hoppel CL, and Lesnefsky EJ (2003) Production of reactive oxygen species by mitochondria: central role of complex III. J Biol. Chem. %19;278 (38), 36027-36031 46. Clark JF, Matsumoto T, and Nakayama S (2000) Intact smooth muscle metabolism: its responses to cyanide poisoning and pyruvate stimulation. Front Biosci. 5:A18-23., A18-A23 47. Collins TJ, Lipp P, Berridge MJ, and Bootman MD (2001) Mitochondrial Ca(2+) uptake depends on the spatial and temporal profile of cytosolic Ca(2+) signals. J Biol. Chem. 276 (28), 26411-26420 48. Collins TJ, Lipp P, Berridge MJ, Li W, and Bootman MD (2000) Inositol 1,4,5-trisphosphate-induced Ca2+ release is inhibited by mitochondrial depolarization. Biochem. J 347 (Pt 2), 593-600 95 49. Coussin F, Macrez N, Morel JL, and Mironneau J (2000) Requirement of ryanodine receptor subtypes 1 and 2 for Ca(2+)-induced Ca(2+) release in vascular myocytes. J Biol. Chem. 275 (13), 9596-9603 50. Cox DA, Conforti L, Sperelakis N, and Matlib MA (1993) Selectivity of inhibition of Na(+)-Ca2+ exchange of heart mitochondria by benzothiazepine CGP-37157. J Cardiovasc. Pharmacol. 21 (4), 595-599 51. Crimmins D, Morris JG, Walker GL, Sue CM, Byrne E, Stevens S, Jean-Francis B, Yiannikas C, and Pamphlett R (1993) Mitochondrial encephalomyopathy: variable clinical expression within a single kindred. J Neurol. Neurosurg. Psychiatry 56 (8), 900-905 52. Crompton M, Barksby E, Johnson N, and Capano M (2002) Mitochondrial intermembrane junctional complexes and their involvement in cell death. Biochimie 84 (2-3), 143-152 53. Crompton M, Moser R, Ludi H, and Carafoli E (1978) The interrelations between the transport of sodium and calcium in mitochondria of various mammalian tissues. Eur. J Biochem. 82 (1), 25-31 54. Csordas G, Thomas AP, and Hajnoczky G (2001) Calcium signal transmission between ryanodine receptors and mitochondria in cardiac muscle. Trends Cardiovasc. Med. 11 (7), 269-275 55. Csordas G, Thomas AP, and Hajnoczky G (1999) Quasi-synaptic calcium signal transmission between endoplasmic reticulum and mitochondria. EMBO J 18 (1), 96-108 56. Csordas G and Hajnoczky G (2003) Plasticity of Mitochondrial Calcium Signaling. J. Biol. Chem. 278 (43), 42273-42282 57. Dai J, Kuo KH, Leo JM, van BC, and Lee CH (2005) Rearrangement of the close contact between the mitochondria and the sarcoplasmic reticulum in airway smooth muscle. Cell Calcium 37 (4), 333-340 58. Darby PJ, Kwan CY, and Daniel EE (2000) Caveolae from canine airway smooth muscle contain the necessary components for a role in Ca(2+) handling. Am J Physiol Lung Cell Mol Physiol 279 (6), L1226-L1235 59. Dedov VN, Mandadi S, Armati PJ, and Verkhratsky A (2001) Capsaicin-induced depolarisation of mitochondria in dorsal root ganglion neurons is enhanced by vanilloid receptors. Neuroscience 103 (1), 219-226 60. Demaurex N and Distelhorst C (2003) Cell biology. Apoptosis—the calcium connection. Science 300 (5616), 65-67 61. Denton RM, McCormack JG, and Edgell NJ (1980) Role of calcium ions in the regulation of intramitochondrial metabolism. Effects of Na+, Mg2+ and ruthenium red on the Ca2+-stimulated oxidation of oxoglutarate and on pyruvate dehydrogenase activity in intact rat heart mitochondria. Biochem. J 190 (1), 107-117 62. Deth R and van Breemen C (1974) Relative contributions of Ca2+ influx and cellular Ca2+ release during drug induced activation of the rabbit aorta. Pflugers Arch. 348 (1), 13-22 63. Devine CE, Somlyo AV, and Somlyo AP (1972) Sarcoplasmic reticulum and excitation-contraction coupling in mammalian smooth muscles. J Cell Biol. 52 (3), 690-718 64. Dong H, Sellers ZM, Smith A, Chow JY, and Barrett KE (2005) Na(+)/Ca(2+) exchange regulates Ca(2+)-dependent duodenal mucosal ion transport and HCO(3)(-) secretion in mice. Am. J Physiol Gastrointest. Liver Physiol 288 (3), G457-G465 65. Drummond RM and Fay FS (1996) Mitochondria contribute to Ca2+ removal in smooth muscle cells. Pflugers Arch. 431 (4), 473-482 66. Drummond RM and Tuft RA (1999) Release of Ca2+ from the sarcoplasmic reticulum increases mitochondrial [Ca2+] in rat pulmonary artery smooth muscle cells. J Physiol 516 (Pt 1), 139-147 67. Duchen MR (1999) Contributions of mitochondria to animal physiology: from homeostatic sensor to calcium signalling and cell death. J Physiol 516 (Pt 1), 1-17 68. Duchen MR, Surin A, and Jacobson J (2003) Imaging mitochondrial function in intact cells. Methods Enzymol. 361:353-89., 353-389 69. EL-Mezgueldi M (1996) Calponin. Int. J Biochem. Cell Biol. 28 (11), 1185-1189 70. Evans JH and Sanderson MJ (1999) Intracellular calcium oscillations induced by ATP in airway epithelial cells. Am J Physiol 277 (1 Pt 1), L30-L41 71. Fayazi AH, Lapidot SA, Huang BK, Tucker RW, and Phair RD (1996) Resolution of the basal plasma membrane calcium flux in vascular smooth muscle cells. Am J Physiol Heart Circ Physiol 270 (6), H1972-H1978 72. Feng W, Tu J, Yang T, Vernon PS, Allen PD, Worley PF, and Pessah IN (2002) Homer Regulates Gain of Ryanodine Receptor Type 1 Channel Complex. J. Biol. Chem. 277 (47), 44722-44730 73. Fiekers JF, Gelbspan D, and Heppner TJ (2001) Calcium homeostasis in a clonal pituitary cell line of mouse corticotropes. Can. J Physiol Pharmacol. 79 (6), 502-511 96 74. Filippin L, Magalhaes PJ, Di Benedetto G, Colella M, and Pozzan T (2003) Stable interactions between mitochondria and endoplasmic reticulum allow rapid accumulation of calcium in a subpopulation of mitochondria. J Biol. Chem.. 75. Fontaine E and Bernardi P (1999) Progress on the mitochondrial permeability transition pore: regulation by complex I and ubiquinone analogs. J Bioenerg. Biomembr. 31 (4), 335-345 76. Frieden M, Malli R, Samardzija M, Demaurex N, and Graier WF (2002) Subplasmalemmal endoplasmic reticulum controls K(Ca) channel activity upon stimulation with a moderate histamine concentration in a human umbilical vein endothelial cell line. J Physiol 540 (Pt 1), 73-84 77. Fujimoto T, Miyawaki A, and Mikoshiba K (1995) Inositol 1,4,5-trisphosphate receptor-like protein in plasmalemmal caveolae is linked to actin filaments. J Cell Sci. 108 (Pt 1), 7-15 78. Fukuta H, Kito Y, and Suzuki H (2002) Spontaneous electrical activity and associated changes in calcium concentration in guinea-pig gastric smooth muscle. J Physiol 540 (Pt 1), 249-260 79. Gamberucci A, Innocenti B, Fulceri R, Banhegyi G, Giunti R, Pozzan T, and Benedetti A (1994) Modulation of Ca2+ influx dependent on store depletion by intracellular adenine-guanine nucleotide levels. J. Biol. Chem. 269 (38), 23597-23602 80. Ganitkevich YY (1996) The amount of acetylcholine mobilisable Ca2+ in single smooth muscle cells measured with the exogenous cytoplasmic Ca2+ buffer, Indo-1. Cell Calcium 20 (6), 483-492 81. Garlid KD, Paucek P, Yarov-Yarovoy V, Sun X, and Schindler PA (1996) The Mitochondrial KflMAGE] Channel as a Receptor for Potassium Channel Openers. J. Biol. Chem. 271 (15), 8796-8799 82. Genova ML, Pich MM, Biondi A, Bernacchia A, Falasca A, Bovina C, Formiggini G, Castelli GP, and Lenaz G (2003) Mitochondrial production of oxygen radical species and the role of Coenzyme Q as an antioxidant. Exp. Biol Med. (Maywood.) 228 (5), 506-513 83. Goel M, Sinkins WG, and Schilling \VP (2002) Selective association of TRPC channel subunits in rat brain synaptosomes. J Biol. Chem. 277 (50), 48303-48310 84. Gollasch M, Hescheler J, Quayle JM, Patlak JB, and Nelson MT (1992) Single calcium channel currents of arterial smooth muscle at physiological calcium concentrations. Am J Physiol 263 (5 Pt 1), C948-C952 85. Golovina VA and Blaustein MP (1997) Spatially and functionally distinct Ca2+ stores in sarcoplasmic and endoplasmic reticulum. Science 275 (5306), 1643-1648 86. Gonzalez-Pacheco FR, Caramelo C, Castilla MA, Deudero JJ, Arias J, Yague S, Jimenez S, Bragado R, and Alvarez-Arroyo MV (2002) Mechanism of vascular smooth muscle cells activation by hydrogen peroxide: role of phospholipase C gamma. Nephrol. Dial. Transplant. 17 (3), 392-398 87. Gordienko DY and Bolton TB (2002) Crosstalk between ryanodine receptors and IP(3) receptors as a factor shaping spontaneous Ca(2+)-reIease events in rabbit portal vein myocytes. J Physiol 542 (Pt 3), 743-762 88. Gordienko DV, Greenwood IA, and Bolton TB (2001) Direct visualization of sarcoplasmic reticulum regions discharging Ca(2+)sparks in vascular myocytes. Cell Calcium 29 (1), 13-28 89. Grasso P, Santa-Coloma TA, and Reichert LE, Jr. (1992) Correlation of follicle-stimulating hormone (FSH)-receptor complex internalization with the sustained phase of FSH-induced calcium uptake by cultured rat Sertoli cells. Endocrinology 131 (6), 2622-2628 90. Grayson TH, Haddock RE, Murray TP, Wojcikiewicz RJ, and Hill CE (2004) Inositol 1,4,5-trisphosphate receptor subtypes are differentially distributed between smooth muscle and endothelial layers of rat arteries. Cell Calcium 36 (6), 447-458 91. Greenwood IA, Helliwell RM, and Large WA (1997) Modulation of Ca(2+)-activated CI- currents in rabbit portal vein smooth muscle by an inhibitor of mitochondrial Ca2+ uptake. J Physiol 505 (Pt 1), 53-64 92. Greenwood IA, Miller LJ, Ohya S, and Horowitz B (2002) The large conductance potassium channel beta-subunit can interact with and modulate the functional properties of a calcium-activated chloride channel, CLCA1. J Biol. Chem. 277 (25), 22119-22122 93. Gunter TE and Gunter KK (2001) Uptake of calcium by mitochondria: transport and possible function. IUBMB. Life 52 (3-5), 197-204 94. Gunter TE, Gunter KK, Sheu SS, and Gavin CE (1994) Mitochondrial calcium transport: physiological and pathological relevance. Am J Physiol 267 (2 Pt 1), C313-C339 95. Gurney AM, Drummond RM, and Fay FS (2000) Calcium signalling in sarcoplasmic reticulum, cytoplasm and mitochondria during activation of rabbit aorta myocytes. Cell Calcium 27 (6), 339-351 96. Hai CM and Kim HR (2005) An expanded latch-bridge model of protein kinase C-mediated smooth muscle contraction. J Appl. Physiol 98 (4), 1356-1365 97 97. Hai CM and Murphy RA (1988) Cross-bridge phosphorylation and regulation of latch state in smooth muscle. Am. J Physiol 254 (1 Pt 1), C99-106 98. Hajnoczky G, Csordas G, Madesh M, and Pacher P (2000) The machinery of local Ca2+ signalling between sarco-endoplasmic reticulum and mitochondria. J. Physiol 529 Pt 1, 69-81 99. Hajnoczky G, Hager R, and Thomas AP (1999) Mitochondria suppress local feedback activation of inositol 1,4, 5- trisphosphate receptors by Ca2+. J Biol. Chem. 274 (20), 14157-14162 100. Hardin CD, Raeymaekers L, and Paul RJ (1992) Comparison of endogenous and exogenous sources of ATP in fueling Ca2+ uptake in smooth muscle plasma membrane vesicles. J Gen. Physiol 99 (1) , 21-40 101. He FJ and MacGregor GA (2004) Effect of longer-term modest salt reduction on blood pressure. The Cochrane Database of Systematic Reviews (1) 102. Herrera GM and Nelson MT (2002) Sarcoplasmic reticulum and membrane currents. Novartis. Found. Symp. 246,189-203 103. Himpens B, De Smedt H, Droogmans G, and Casteels R (1992) Differences in regulation between nuclear and cytoplasmic Ca2+ in cultured smooth muscle cells. Am J Physiol 263 (1 Pt 1), C95-105 104. Ho SH, Das GU, and Rieske JS (1985) Detection of antimycin-binding subunits of complex III by photoaffinity-labeling with an azido derivative of antimycin. J Bioenerg. Biomembr. 17 (5), 269-282 105. Hopf FW, Reddy P, Hong J, and Steinhardt RA (1996) A capacitative calcium current in cultured skeletal muscle cells is mediated by the calcium-specific leak channel and inhibited by dihydropyridine compounds. J. Biol. Chem. 271 (37), 22358-22367 106. Hoth M, Fanger CM, and Lewis RS (1997) Mitochondrial regulation of store-operated calcium signaling in T lymphocytes. J. Cell Biol. 137 (3), 633-648 107. Humbert JP, Matter N, Artault JC, Koppler P, and Malviya AN (1996) Inositol 1,4,5-trisphosphate receptor is located to the inner nuclear membrane vindicating regulation of nuclear calcium signaling by inositol 1,4,5-trisphosphate. Discrete distribution of inositol phosphate receptors to inner and outer nuclear membranes. J Biol. Chem. 271 (1), 478-485 108. lino M (1989) Calcium-induced calcium release mechanism in guinea pig taenia caeci. J Gen. Physiol 94 (2), 363-383 109. lino M (2002) Molecular basis and physiological functions of dynamic Ca2+ signalling in smooth muscle cells. Novartis. Found. Symp. 246,142-146 110. Iizuka K, Yoshii A, Dobashi K, Horie T, Mori M, and Nakazawa T (1998) InsP3, but not novel Ca2+ releasers, contributes to agonist-initiated contraction in rabbit airway smooth muscle. J Physiol 511 (Pt 3), 915-933 111. Jaburek M, Yarov-Yarovoy V, Paucek P, and Garlid KD (1998) State-dependent inhibition of the mitochondrial KATP channel by glyburide and 5-hydroxydecanoate. J Biol. Chem. 273 (22), 13578-13582 112. Jaggar JH and Nelson MT (2000) Differential regulation of Ca(2+) sparks and Ca(2+) waves by UTP in rat cerebral artery smooth muscle cells. Am J Physiol Cell Physiol 279 (5), C1528-C1539 113. Jaggar JH, Porter VA, Lederer WJ, and Nelson MT (2000) Calcium sparks in smooth muscle. Am J Physiol Cell Physiol 278 (2), C235-C256 114. Janiak R, Wilson SM, Montague S, and Hume JR (2001) Heterogeneity of calcium stores and elementary release events in canine pulmonary arterial smooth muscle cells. Am. J Physiol Cell Physiol 280 (1), C22-C33 115. Janssen LJ, Betti PA, Netherton SJ, and Walters DK (1999) Superficial buffer barrier and preferentially directed release of Ca2+ in canine airway smooth muscle. Am J Physiol 276 (5 Pt 1), L744-L753 116. Juhaszova M and Blaustein MP (1997) Distinct distribution of different Na+ pump alpha subunit isoforms in plasmalemma. Physiological implications. Ann. N. Y. Acad. Sci. 834, 524-536 117. Jung S, Strotmann R, Schultz G, and Plant TD (2002) TRPC6 is a candidate channel involved in receptor-stimulated cation currents in A7r5 smooth muscle cells. Am J Physiol Cell Physiol 282 (2) , C347-C359 118. Kamishima T, Davies NW, and Standen NB (2000) Mechanisms that regulate [Ca2+]i following depolarization in rat systemic arterial smooth muscle cells. J Physiol 522 Pt 2:285-95., 285-295 119. Kamishima T and McCarron JG (1998) Ca2+ removal mechanisms in rat cerebral resistance size arteries. Biophys. J. 75 (4), 1767-1773 120. Kamishima T and Quayle JM (2002) Mitochondrial Ca2+ uptake is important over low [Ca2+]i range in arterial smooth muscle. Am J Physiol Heart Circ Physiol 283 (6), H2431-H2439 98 121. Kang TM, Park MK, and Uhm DY (2003) Effects of hypoxia and mitochondrial inhibition on the capacitative calcium entry in rabbit pulmonary arterial smooth muscle cells. Life Sci. 72 (13), 1467-1479 122. Kennedy CH, Winston GW, Church DF, and Pryor WA (1989) Benzoyl peroxide interaction with mitochondria: inhibition of respiration and induction of rapid, large-amplitude swelling. Arch. Biochem. Biophys. 271 (2), 456-470 123. Knot HJ, de Ree MM, Gahwiler BH, and Ruegg UT (1991) Modulation of electrical activity and of intracellular calcium oscillations of smooth muscle cells by calcium antagonists, agonists, and vasopressin. J. Cardiovasc. Pharmacol. 18 Suppl 10, S7-14 124. Kotlikoff ML, Wang YX, Xin HB, and Ji G (2002) Calcium release by ryanodine receptors in smooth muscle. Novartis. Found. Symp. 246,108-119 125. Kotlyar AB and Gutman M (1992) The effect of delta mu H+ on the interaction of rotenone with complex I of submitochondrial particles. Biochim. Biophys. Acta 1140 (2), 169-174 126. Krick S, Platoshyn O, Sweeney M, Kim H, and Yuan JX (2001) Activation of K+ channels induces apoptosis in vascular smooth muscle cells. Am J Physiol Cell Physiol 280 (4), C970-C979 127. Kubota Y, Hashitani H, Fukuta H, Kubota H, Kohri K, and Suzuki H (2003) Role of mitochondria in the generation of spontaneous activity in detrusor smooth muscles of the Guinea pig bladder. J Urol. 170 (2 Pt 1), 628-633 128. Kuge O and Nishijima M (2003) Biosynthetic regulation and intracellular transport of phosphatidylserine in mammalian cells. J Biochem. (Tokyo) 133 (4), 397-403 129. Landolfi B, Curci S, Debellis L, Pozzan T, and Hofer AM (1998) Ca2+ homeostasis in the agonist-sensitive internal store: functional interactions between mitochondria and the ER measured In situ in intact cells. J. Cell Biol. 142 (5), 1235-1243 130. Leach RM, Hill HM, Snetkov VA, Robertson TP, and Ward JP (2001) Divergent roles of glycolysis and the mitochondrial electron transport chain in hypoxic pulmonary vasoconstriction of the rat: identity of the hypoxic sensor. J Physiol 536 (Pt 1), 211-224 131. Lee CH, Kuo KH, Dai J, Leo JM, Seow CY, and Breemen C (2005) Calyculin-A disrupts subplasmalemmal junction and recurring Ca2+ waves in vascular smooth muscle. Cell Calcium 37 (1), 9-16 132. Lee CH, Poburko D, Kuo KH, Seow C, and van Breemen C (2002) Relationship between the sarcoplasmic reticulum and the plasma membrane. Novartis. Found. Symp. 246:26-41; discussion 41-7,48-51., 26-41 133. Lee CH, Poburko D, Kuo KH, Seow CY, and van Breemen C (2002) Ca(2+) oscillations, gradients, and homeostasis in vascular smooth muscle. Am J Physiol Heart Circ Physiol 282 (5), H1571-H1583 134. Lee CH, Poburko D, Sahota P, Sandhu J, Ruehlmann DO, and van Breemen C (2001) The mechanism of phenylephrine-mediated [Ca(2+)](i) oscillations underlying tonic contraction in the rabbit inferior vena cava. J Physiol 534 (Pt 3), 641-650 135. Lee CH, Rahimian R, Szado T, Sandhu J, Poburko D, Behra T, Chan L, and van Breemen C (2002) Sequential opening of IP(3)-sensitive Ca(2+) channels and SOC during alpha-adrenergic activation of rabbit vena cava. Am J Physiol Heart Circ Physiol 282 (5), H1768-H1777 136. Lesh RE, Nixon GF, Fleischer S, Airey JA, Somlyo AP, and Somlyo AV (1998) Localization of ryanodine receptors in smooth muscle. Circ. Res. 82 (2), 175-185 137. Li SW, Westwick J, and Poll CT (2002) Receptor-operated Ca2+ influx channels in leukocytes: a therapeutic target? Trends Pharmacol. Sci. 23 (2), 63-70 138. Liang W, Buluc M, van BC, and Wang X (2004) Vectorial Ca2+ release via ryanodine receptors contributes to Ca2+ extrusion from freshly isolated rabbit aortic endothelial cells. Cell Calcium 36 (5), 431-443 139. Liu Y, Zhao H, Li H, Kalyanaraman B, Nicolosi AC, and Gutterman DD (2003) Mitochondrial sources of H202 generation play a key role in flow-mediated dilation in human coronary resistance arteries. Circ Res. %19;93 (6), 573-580 140. Lo Russo A., Passaquin AC, Andre P, Skutella M, and Ruegg UT (1996) Effect of cyclosporin A and analogues on cytosolic calcium and vasoconstriction: possible lack of relationship to immunosuppressive activity. Br. J. Pharmacol. 118 (4), 885-892 141. Loew LM, Carrington W, Tuft RA, and Fay FS (1994) Physiological cytosolic Ca2+ transients evoke concurrent mitochondrial depolarizations. Proc. Natl. Acad. Sci. U. S. A %20;91 (26), 12579-12583 142. Loew LM, Tuft RA, Carrington W, and Fay FS (1993) Imaging in five dimensions: time-dependent membrane potentials in individual mitochondria. Biophys. J 65 (6), 2396-2407 99 143. Lounsbury KM, Hu Q, and Ziegelstein RC (2000) Calcium signaling and oxidant stress in the vasculature. Free Radic. Biol. Med. 28 (9), 1362-1369 144. Luttun A and Carmeliet P (2003) De novo vasculogenesis in the heart. Cardiovasc. Res. 58 (2), 378-389 145. Ma HT, Patterson RL, van Rossum DB, Birnbaumer L, Mikoshiba K, and Gill DL (2000) Requirement of the inositol trisphosphate receptor for activation of store-operated Ca2+ channels. Science 287 (5458), 1647-1651 146. Malli R, Frieden M, Osibow K, and Graier WF (2003) Mitochondria efficiently buffer subplasmalemmal Ca2+ elevation during agonist stimulation. J Biol. Chem. 278 (12), 10807-10815 147. Malli R, Frieden M, Osibow K, Zoratti C, Mayer M, Demaurex N, and Graier WF (2003) Sustained Ca2+ transfer across mitochondria is Essential for mitochondrial Ca2+ buffering, sore-operated Ca2+ entry, and Ca2+ store refilling. J Biol. Chem. 278 (45), 44769-44779 148. Marsault R, Murgia M, Pozzan T, and Rizzuto R (1997) Domains of high Ca2+ beneath the plasma membrane of living A7r5 cells. EMBO J. 16 (7), 1575-1581 149. Marshall C, Elias C, Xue XH, Le HD, Omelchenko A, Hryshko LV, and Tibbits GF (2002) Determinants of cardiac Na+/Ca2+ exchanger temperature dependence: NH2-terminal transmembrane segments. Am. J. Physiol Cell Physiol 283 (2), C512-C520 150. Matlib MA, Zhou Z, Knight S, Ahmed S, Choi KM, Krause-Bauer J, Phillips R, Altschuld R, Katsube Y, Sperelakis N, and Bers DM (1998) Oxygen-bridged dinuclear ruthenium amine complex specifically inhibits Ca2+ uptake into mitochondria in vitro and in situ in single cardiac myocytes. J Biol. Chem. 273 (17), 10223-10231 151. McCarron JG and Muir TC (1999) Mitochondrial regulation of the cytosolic Ca2+ concentration and the InsP3-sensitive Ca2+ store in guinea-pig colonic smooth muscle. J Physiol 516 (Pt 1), 149-161 152. McDaniel SS, Platoshyn O, Wang J, Yu Y, Sweeney M, Krick S, Rubin LJ, and Yuan JX (2001) Capacitative Ca(2+) entry in agonist-induced pulmonary vasoconstriction. Am. J. Physiol Lung Cell Mol. Physiol 280 (5), L870-L880 153. McGeown JG (2004) Interactions between inositol 1,4,5-trisphosphate receptors and ryanodine receptors in smooth muscle: one store or two? Cell Calcium 35 (6), 613-619 154. Michelakis ED, Hampl V, Nsair A, Wu X, Harry G, Haromy A, Gurtu R, and Archer SL (2002) Diversity in mitochondrial function explains differences in vascular oxygen sensing. Circ Res. 90 (12), 1307-1315 155. Montero M, Alonso MT, Albillos A, Garcia-Sancho J, and Alvarez J (2001) Mitochondrial Ca(2+)-induced Ca(2+) release mediated by the Ca(2+) uniporter. Mol. Biol. Cell 12 (1), 63-71 156. Montero M, Alonso MT, Carnicero E, Cuchillo-Ibanez I, Albillos A, Garcia AG, Garcia-Sancho J, and Alvarez J (2000) Chromaffin-cell stimulation triggers fast millimolar mitochondrial Ca2+ transients that modulate secretion. Nat. Cell Biol. 2 (2), 57-61 157. Moore ED, Etter EF, Philipson KD, Carrington WA, Fogarty KE, Lifshitz LM, and Fay FS (1993) Coupling of the Na+/Ca2+ exchanger, Na+/K+ pump and sarcoplasmic reticulum in smooth muscle. Nature 365 (6447), 657-660 158. Moore RA, Nguyen H, Galceran J, Pessah IN, and Allen PD (1998) A transgenic myogenic cell line lacking ryanodine receptor protein for homologous expression studies: reconstitution of RylR protein and function. J Cell Biol. 140 (4), 843-851 159. Nadler MJ, Hermosura MC, Inabe K, Perraud AL, Zhu Q, Stokes AJ, Kurosaki T, Kinet JP, Penner R, Scharenberg AM, and Fleig A (2001) LTRPC7 is a Mg.ATP-regulated divalent cation channel required for cell viability. Nature 411 (6837), 590-595 160. Nassar A and Simpson AW (2000) Elevation of mitochondrial calcium by ryanodine-sensitive calcium-induced calcium release. J Biol. Chem. 275 (31), 23661-23665 161. Nazer MA and van Breemen C (1998) A role for the sarcoplasmic reticulum in Ca2+ extrusion from rabbit inferior vena cava smooth muscle. Am. J. Physiol 274 (1 Pt 2), H123-H131 162. Nazer MA and van Breemen C (1998) Functional linkage of Na(+)-Ca2+ exchange and sarcoplasmic reticulum Ca2+ release mediates Ca2+ cycling in vascular smooth muscle. Cell Calcium 24 (4), 275-283 163. Nilius B (2003) From TRPs to SOCs, CCEs, and CRACs: consensus and controversies. Cell Calcium 33 (5-6), 293-298 164. Nixon GF, Mignery GA, and Somlyo AV (1994) Immunogold localization of inositol 1,4,5-trisphosphate receptors and characterization of ultrastructural features of the sarcoplasmic reticulum in phasic and tonic smooth muscle. J Muscle Res. Cell Motil. 15 (6), 682-700 100 165. Nowycky MC, Fox AP, and Tsien RW (1985) Long-opening mode of gating of neuronal calcium channels and its promotion by the dihydropyridine calcium agonist Bay K 8644. Proc. Natl. Acad. Sci. U. S. A 82 (7), 2178-2182 166. Obejero-Paz CA, Jones SW, and Scarpa A (1998) Multiple channels mediate calcium leakage in the A7r5 smooth muscle- derived cell line. Biophys. J. 75 (3), 1271-1286 167. Obukhov AG and Nowycky MC (2002) TRPC4 can be activated by G-protein-coupled receptors and provides sufficient Ca(2+) to trigger exocytosis in neuroendocrine cells. J Biol. Chem. 277 (18), 16172-16178 168. Pacher P, Thomas AP, and Hajnoczky G (2002) Ca2+ marks: miniature calcium signals in single mitochondria driven by ryanodine receptors. Proc. Natl. Acad. Sci. U. S. A 99 (4), 2380-2385 169. Parekh AB (2003) Store-operated Ca2+ entry: dynamic interplay between endoplasmic reticulum, mitochondria and plasma membrane. J Physiol 547 (Pt 2), 333-348 170. Park MK, Ashby MC, Erdemli G, Petersen OH, and Tepikin AV (2001) Perinuclear, perigranular and sub-plasmalemmal mitochondria have distinct functions in the regulation of cellular calcium transport. EMBO J. 20 (8), 1863-1874 171. Paul RJ (1990) Smooth muscle energetics and theories of cross-bridge regulation. Am. J Physiol 258 (2 Pt 1), C369-C375 172. Peng H, Matchkov V, Ivarsen A, Aalkjaer C, and Nilsson H (2001) Hypothesis for the initiation of vasomotion. Circ Res. 88 (8), 810-815 173. Perez-Vizcaino F, Tamargo J, Hof RP, and Ruegg UT (1993) Vascular selectivity of seven prototype calcium antagonists: a study at the single cell level. J. Cardiovasc. Pharmacol. 22 (5), 768-775 174. Pfeiffer DR, Gunter TE, Eliseev R, Broekemeier KM, and Gunter KK (2001) Release of Ca2+ from mitochondria via the saturable mechanisms and the permeability transition. IUBMB. Life 52 (3-5), 205-212 175. Plant TD and Schaefer M (2003) TRPC4 and TRPC5: receptor-operated Ca2+-permeable nonselective cation channels. Cell Calcium 33 (5-6), 441-450 176. Poburko D, Kuo KH, Dai J, Lee CH, and van Breemen C (2004) Organellar junctions promote targeted Ca2+ signaling in smooth muscle: why two membranes are better than one. Trends Pharmacol. Sci. 25 (1), 8-15 177. Poburko D, Lee CH, and van Breemen C (2004) Vascular smooth muscle mitochondria at the cross roads of Ca2+ signaling. Cell Calcium 35 (6), 509-521 178. Poburko D, Lhote P, Szado T, Behra T, Rahimian R, McManus B, van Breemen C, and Ruegg UT (2004) Basal calcium entry in vascular smooth muscle. European Journal of Pharmacology 505 (1-3), 19-29 179. Pozzan T, Rizzuto R, Volpe P, and Meldolesi J (1994) Molecular and cellular physiology of intracellular calcium stores. Physiol Rev. 74 (3), 595-636 180. Radi R, Cassina A, Hodara R, Quijano C, and Castro L (2002) Peroxynitrite reactions and formation in mitochondria. Free Radic. Biol. Med. 33 (11), 1451-1464 181. Raeymaekers L, Verbist J, Wuytack F, Plessers L, and Casteels R (1993) Expression of Ca2+ binding proteins of the sarcoplasmic reticulum of striated muscle in the endoplasmic reticulum of pig smooth muscles. Cell Calcium 14 (8), 581-589 182. Ragan CI and Bloxham DP (1977) Specific labelling of a constituent polypeptide of bovine heart mitochondrial reduced nicotinamide-adenine dinucleotide-ubiquinone reductase by the inhibitor diphenyleneiodonium. Biochem. J 163 (3), 605-615 183. Ramos-Franco J, Fill M, and Mignery GA (1998) Isoform-specific function of single inositol 1,4,5-trisphosphate receptor channels. Biophys. J 75 (2), 834-839 184. Ramsay RR, Ackrell BA, Coles CJ, Singer TP, White GA, and Thorn GD (1981) Reaction site of carboxanilides and of thenoyltrifluoroacetone in complex II. Proc. Natl. Acad. Sci. U. S. A 78 (2), 825-828 185. Rapizzi E, Pinton P, Szabadkai G, Wieckowski MR, Vandecasteele G, Baird G, Tuft RA, Fogarty KE, and Rizzuto R (2002) Recombinant expression of the voltage-dependent anion channel enhances the transfer of Ca2+ microdomains to mitochondria. J Cell Biol. 159 (4), 613-624 186. Rathaus M and Bernheim J (2002) Oxygen species in the microvascular environment: regulation of vascular tone and the development of hypertension. Nephrol. Dial. Transplant. 17 (2), 216-221 187. Rembold CM and Chen XL (1998) The buffer barrier hypothesis, [Ca2+]i homogeneity, and sarcoplasmic reticulum function in swine carotid artery. J Physiol 513 (Pt 2), 477-492 188. Rembold CM and Murphy RA (1990) Latch-bridge model in smooth muscle: [Ca2+]i can quantitatively predict stress. Am. J Physiol 259 (2 Pt 1), C251-C257 101 189. Rich PR, Jeal AE, Madgwick SA, and Moody AJ (1990) Inhibitor effects on redox-linked protonations of the b haems of the mitochondrial bcl complex. Biochim. Biophys. Acta 1018 (1), 29-40 190. Rizzuto R, Bernardi P, and Pozzan T (2000) Mitochondria as all-round players of the calcium game. J Physiol 529 Pt 1,37-47 191. Rizzuto R, Brini M, and Pozzan T (1993) Intracellular targeting of the photoprotein aequorin: a new approach for measuring, in living cells, Ca2+ concentrations in defined cellular compartments. Cytotechnology 11 Suppl 1, S44-S46 192. Rizzuto R, Duchen MR, and Pozzan T (2004) Flirting in little space: the ER/mitochondria Ca2+ liaison. Sci. STKE. 2004 (215), rel 193. Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, and Pozzan T (1998) Close contacts with the endoplasmic reticulum as determinants of mitochondrial Ca2+ responses. Science 280 (5370), 1763-1766 194. Rizzuto R, Simpson AW, Brini M, and Pozzan T (1992) Rapid changes of mitochondrial Ca2+ revealed by specifically targeted recombinant aequorin. Nature 358 (6384), 325-327 195. Robb-Gaspers LD, Rutter GA, Burnett P, Hajnoczky G, Denton RM, and Thomas AP (1998) Coupling between cytosolic and mitochondrial calcium oscillations: role in the regulation of hepatic metabolism. Biochim. Biophys. Acta 1366 (1-2), 17-32 196. Robert V, Gurlini P, Tosello V, Nagai T, Miyawaki A, Di Lisa F, and Pozzan T (2001) Beat-to-beat oscillations of mitochondrial [Ca2+J in cardiac cells. EMBO J. 20 (17), 4998-5007 197. Rosker C, Graziani A, Lukas M, Eder P, Zhu MX, Romanin C, and Groschner K (2004) Ca2+ Signaling by TRPC3 Involves Na+ Entry and Local Coupling to the Na+/Ca2+ Exchanger. J. Biol. Chem. 279 (14), 13696-13704 198. Rudolf R, Mongillo M, Rizzuto R, and Pozzan T (2003) Looking forward to seeing calcium. Nat. Rev. Mol Cell Biol. 4 (7), 579-586 199. Ruehlmann DO, Lee CH, Poburko D, and van Breemen C (2000) Asynchronous Ca(2+) waves in intact venous smooth muscle. Circ Res. 86 (4), E72-E79 200. Salomon AR, Voehringer DW, Herzenberg LA, and Khosla C (2001) Apoptolidin, a selective cytotoxic agent, is an inhibitor of FOFl-ATPase. Chem. Biol. 8 (1), 71-80 201. Satoh T, Ross CA, Villa A, Supattapone S, Pozzan T, Snyder SH, and Meldolesi J (1990) The inositol 1,4,5,-trisphosphate receptor in cerebellar Purkinje cells: quantitative immunogold labeling reveals concentration in an ER subcompartment. J. Cell Biol. I l l (2), 615-624 202. Schumacher C, Konigs B, Sigmund M, Kohne B, Schondube F, Vob M, Stein B, Weil J, and Hanrath P (1995) The ryanodine binding sarcoplasmic reticulum calcium release channel in nonfailing and in failing human myocardium. Naunyn Schmiedebergs Arch. Pharmacol. 353 (1), 80-85 203. Sedarat F, Lin E, Moore ED, and Tibbits GF (2004) Deconvolution of confocal images of dihydropyridine and ryanodine receptors in developing cardiomyocytes. J Appl. Physiol 97 (3), 1098-1103 204. Seguchi H, Ritter M, Shizukuishi M, Ishida H, Chokoh G, Nakazawa H, Spitzer KW, and Barry WH (2005) Propagation of Ca(2+) release in cardiac myocytes: Role of mitochondria. Cell Calcium 38 (1), 1-9 205. Sensi SL, Ton-That D, and Weiss JH (2002) Mitochondrial sequestration and Ca(2+)-dependent release of cytosolic Zn(2+) loads in cortical neurons. Neurobiol. Dis. 10 (2), 100-108 206. Setoguchi M, Ohya Y, Abe I, and Fujishima M (1997) Stretch-activated whole-cell currents in smooth muscle cells from mesenteric resistance artery of guinea-pig. J Physiol 501 (Pt 2), 343-353 207. Shmigol AV, Eisner DA, and Wray S (2001) Simultaneous measurements of changes in sarcoplasmic reticulum and cytosolic. J Physiol 531 (Pt 3), 707-713 208. Sitsapesan R, McGarry SJ, and Williams AJ (1995) Cyclic ADP-ribose, the ryanodine receptor and Ca2+ release. Trends Pharmacol. Sci. 16 (11), 386-391 209. Skutella M and Ruegg UT (1996) Increase of empty pool-activated Ca2+ influx using an intracellular Ca2+ chelating agent. Biochem. Biophys. Res. Commun. 218 (3), 837-841 210. Somlyo AP (1985) Excitation-contraction coupling and the ultrastructure of smooth muscle. Circ. Res. 57 (4), 497-507 211. Sotnikova R (1998) Investigation of the mechanisms underlying H202-evoked contraction in the isolated rat aorta. Gen. Pharmacol. 31 (1), 115-119 212. Sparagna GC, Gunter KK, Sheu SS, and Gunter TE (1995) Mitochondrial calcium uptake from physiological-type pulses of calcium. A description of the rapid uptake mode. J Biol. Chem. 270 (46), 27510-27515 102 213. Steenbergen JM and Fay FS (1996) The quanta! nature of calcium release to caffeine in single smooth muscle cells results from activation of the sarcoplasmic reticulum Ca(2+)-ATPase. J Biol. Chem. 271 (4), 1821-1824 214. Sturek M, Kunda K, and Hu Q (1992) Sarcoplasmic reticulum buffering of myoplasmic calcium in bovine coronary artery smooth muscle. J Physiol 451,25-48 215. Su Z, Barker DS, Csutora P, Chang T, Shoemaker RL, Marchase RB, and Blalock JE (2003) Regulation of Ca2+ release-activated Ca2+ channels by INAD and Ca2+ influx factor. Am J Physiol Cell Physiol 284 (2), C497-C505 216. Sugiyama T, Matsuda Y, and Mikoshiba K (2000) Inositol 1,4,5-trisphosphate receptor associated with focal contact cytoskeletal proteins. FEBS Lett. 466 (1), 29-34 217. Sward K, Dreja K, Lindqvist A, Persson E, and Hellstrand P (2002) Influence of mitochondrial inhibition on global and local [Ca(2+)](I) in rat tail artery. Circ Res. %19;90 (7), 792-799 218. Sward K, Mita M, Wilson DP, Deng JT, Susnjar M, and Walsh MP (2003) The role of RhoA and Rho-associated kinase in vascular smooth muscle contraction. Curr. Hypertens. Rep. 5 (1), 66-72 219. Szabadkai G, Pitter JG, and Spat A (2001) Cytoplasmic Ca2+ at low submicromolar concentration stimulates mitochondrial metabolism in rat luteal cells. Pflugers Arch. 441 (5), 678-685 220. Szado T, Kuo KH, Bernard-Helary K, Poburko D, Lee CH, Seow C, Ruegg UT, and van Breemen C (2003) Agonist-induced mitochondrial Ca2+ transients in smooth muscle. FASEB J 17 (1), 28-37 221. Szado T, McLarnon M, Wang X, and van Breemen C (2001) Role of sarcoplasmic reticulum in regulation of tonic contraction of rabbit basilar artery. Am J Physiol Heart Circ Physiol 281 (4), H1481-H1489 222. Taggart MJ and Wray S (1997) Agonist mobilization of sarcoplasmic reticular calcium in smooth muscle: functional coupling to the plasmalemmal Na+/Ca2+ exchanger? Cell Calcium 22 (5), 333-341 223. Takeshima H, Komazaki S, Nishi M, lino M, and Kangawa K (2000) Junctophilins: a novel family of junctional membrane complex proteins. Mol Cell 6 (1), 11-22 224. Tasker PN, Taylor CW, and Nixon GF (2000) Expression and distribution of InsP(3) receptor subtypes in proliferating vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 273 (3), 907-912 225. Thebault S, Zholos A, Enfissi A, Slomianny C, Dewailly E, Roudbaraki M, Parys J, and Prevarskaya N (2005) Receptor-operated Ca(2+) entry mediated by TRPC3/TRPC6 proteins in rat prostate smooth muscle (PS1) cell line. J Cell Physiol. 226. Thyagarajan B, Poteser M, Romanin C, Kahr H, Zhu MX, and Groschner K (2001) Expression of Trp3 Determines Sensitivity of Capacitative Ca2+ Entry to Nitric Oxide and Mitochondrial Ca2+ Handling. EVIDENCE FOR A ROLE OF Trp3 AS A SUBUNIT OF CAPACITATIVE Ca2+ ENTRY CHANNELS. J. Biol. Chem. 276 (51), 48149-48158 227. Tiruppathi C, Freichel M, Vogel SM, Paria BC, Mehta D, Flockerzi V, and Malik AB (2002) Impairment of store-operated Ca2+ entry in TRPC4(-/-) mice interferes with increase in lung microvascular permeability. Circ Res. 91 (1), 70-76 228. Tokunaga H, Hollenberg NK, and Graves SW (2000) Sodium-dependent calcium release from vascular smooth muscle mitochondria. Hypertens. Res. 23 (1), 39-45 229. Trebak M, Bird GS, McKay RR, and Putney JW, Jr. (2002) Comparison of human TRPC3 channels in receptor-activated and store- operated modes. Differential sensitivity to channel blockers suggests fundamental differences in channel composition. J Biol. Chem. 277 (24), 21617-21623 230. Trebak M, Vazquez G, Bird GS, and Putney JW, Jr. (2003) The TRPC3/6/7 subfamily of cation channels. Cell Calcium 33 (5-6), 451-461 231. Vallot O, Combettes L, and Lompre AM (2001) Functional coupling between the caffeine/ryanodine-sensitive Ca2+ store and mitochondria in rat aortic smooth muscle cells. Biochem. J 357 (Pt 2), 363-371 232. van Breemen C (1968) Permselectivity of a porous phospholipid-cholesterol artificial membrane. Calcium and lanthanum effects. Biochem. Biophys. Res. Commun. 32 (6), 977-983 233. van Breemen C (1977) Calcium requirement for activation of intact aortic smooth muscle. J. Physiol 272 (2), 317-329 234. van Breemen C, Cauvin C, Johns A, Leijten P, and Yamamoto H (1986) Ca2+ regulation of vascular smooth muscle. Fed. Proc. 45 (12), 2746-2751 235. van Breemen C, Chen Q, and Laher I (1995) Superficial buffer barrier function of smooth muscle sarcoplasmic reticulum. Trends Pharmacol. Sci. 16 (3), 98-105 236. van Breemen C, Farinas BR, Casteels R, Gerba P, Wuytack F, and Deth R (1973) Factors controlling cytoplasmic Ca 2+ concentration. Philos. Trans. R. Soc. Lond B Biol. Sci. 265 (867), 57-71 103 237. van Breemen C, Lukeman S, Leijten P, Yamamoto H, and Loutzenhiser R (1986) The role of superficial SR in modulating force development induced by Ca entry into arterial smooth muscle. J. Cardiovasc. Pharmacol. 8 Suppl 8, S111-S116 238. Van Breemen D and van Breemen C (1969) Calcium exchange diffusion in a porous phospholipid ion-exchange membrane. Nature 223 (209), 898-900 239. Van Dyke RW, Scharschmidt BF, and Steer CJ (1985) ATP-dependent proton transport by isolated brain clathrin-coated vesicles. Role of clathrin and other determinants of acidification. Biochim. Biophys. Acta 812 (2), 423-436 240. Vandebrouck C, Martin D, Schoor MC-V, Debaix H, and Gailly P (2002) Involvement of TRPC in the abnormal calcium influx observed in dystrophic (mdx) mouse skeletal muscle fibers. J. Cell Biol. 158 (6), 1089 241. Vandecasteele G, Szabadkai G, and Rizzuto R (2001) Mitochondrial calcium homeostasis: mechanisms and molecules. IUBMB. Life 52 (3-5), 213-219 242. Vasilyeva EA, Minkov IB, Fitin AF, and Vinogradov AD (1982) Kinetic mechanism of mitochondrial adenosine triphosphatase. Inhibition by azide and activation by sulphite. Biochem. J 202 (1), 15-23 243. Vennekens R, Voets T, Bindels RJM, Droogmans G, and Nilius B (2002) Current understanding of mammalian TRP homologues. Cell Calcium 31 (6), 253-264 244. Vermassen E, Parys JB, and Mauger JP (2004) Subcellular distribution of the inositol 1,4,5-trisphosphate receptors: functional relevance and molecular determinants. Biol. Cell 96 (1), 3-17 245. Voronina S, Sukhomlin T, Johnson PR, Erdemli G, Petersen OH, and Tepikin A (2002) Correlation of NADH and Ca2+ signals in mouse pancreatic acinar cells. J Physiol 539 (Pt 1), 41-52 246. Wada AM, Smith TK, Osier ME, Reese DE, and Bader DM (2003) Epicardial/Mesothelial cell line retains vasculogenic potential of embryonic epicardium. Circ Res. 92 (5), 525-531 247. Walker RL, Hume JR, and Horowitz B (2001) Differential expression and alternative splicing of TRP channel genes in smooth muscles. Am. J. Physiol Cell Physiol 280 (5), C1184-C1192 248. Wallnofer A, Cauvin C, Lategan TW, and Ruegg UT (1989) Differential blockade of ago. Am. J. Physiol 257 (4 Pt 1), C607-C611 249. Wamhoff BR, Bowles DK, Dietz NJ, Hu Q, and Sturek M (2002) Exercise training attenuates coronary smooth muscle phenotypic modulation and nuclear Ca2+ signaling. Am J Physiol Heart Circ Physiol 283 (6), H2397-H2410 250. Wang HJ, Guay G, Pogan L, Sauve R, and Nabi IR (2000) Calcium regulates the association between mitochondria and a smooth subdomain of the endoplasmic reticulum. J Cell Biol. 150 (6), 1489-1498 251. Wang YX, Zheng YM, Abdullaev I, and Kotlikoff MI (2003) Metabolic inhibition with cyanide induces calcium release in pulmonary artery myocytes and Xenopus oocytes. Am J Physiol Cell Physiol 284 (2), C378-C388 252. Waypa GB, Marks JD, Mack MM, Boriboun C, Mungai PT, and Schumacker PT (2002) Mitochondrial reactive oxygen species trigger calcium increases during hypoxia in pulmonary arterial myocytes. Circ Res. 91 (8), 719-726 253. Weir EK, Hong Z, Porter VA, and Reeve HL (2002) Redox signaling in oxygen sensing by vessels. Respir. Physiol Neurobiol. 132 (1), 121-130 254. Wellman GC and Nelson MT (2003) Signaling between SR and plasmalemma in smooth muscle: sparks and the activation of Ca(2+)-sensitive ion channels. Cell Calcium 34 (3), 211-229 255. White C and McGeown JG (2002) Carbachol triggers RyR-dependent Ca(2+) release via activation of IP(3) receptors in isolated rat gastric myocytes. J Physiol 542 (Pt 3), 725-733 256. Wilson DP, Sutherland C, and Walsh MP (2002) Ca2+ activation of smooth muscle contraction: evidence for the involvement of calmodulin that is bound to the triton insoluble fraction even in the absence of Ca2+. J Biol. Chem. 277 (3), 2186-2192 257. Wilson HL, Dipp M, Thomas JM, Lad C, Galione A, and Evans AM (2001) Adp-ribosyl cyclase and cyclic ADP-ribose hydrolase act as a redox sensor, a primary role for cyclic ADP-ribose in hypoxic pulmonary vasoconstriction. J Biol. Chem. 276 (14), 11180-11188 258. Wolin MS (2000) Interactions of oxidants with vascular signaling systems. Arterioscler Thromb Vase Biol 20 (6), 1430-1442 259. Wolin MS, Gupte SA, and Oeckler RA (2002) Superoxide in the vascular system. J Vase Res. 39 (3), 191-207 260. Wood JN, Winter J, James IF, Rang HP, Yeats J, and Bevan S (1988) Capsaicin-induced ion fluxes in dorsal root ganglion cells in culture. J Neurosci. 8 (9), 3208-3220 104 261. Yamamura H, Sakamoto K, Ohya S, Muraki K, and Imaizumi Y (2002) Mechanisms underlying the activation of large conductance Ca2+-activated K+ channels by nordihydroguaiaretic acid. Jpn. J Pharmacol. 89 (1), 53-63 262. Yi M, Weaver D, and Hajnoczky G (2004) Control of mitochondrial motility and distribution by the calcium signal: a homeostatic circuit. J Cell Biol. 167 (4), 661-672 263. Zakharov SI, Mongayt DA, Cohen RA, and Bolotina VM (1999) Monovalent cation and L-type Ca2+ channels participate in calcium paradox-like phenomenon in rabbit aortic smooth muscle cells. J. Physiol 514 ( Pt 1), 71-81 264. Zazueta C, Sosa-Torres ME, Correa F, and Garza-Ortiz A (1999) Inhibitory properties of ruthenium amine complexes on mitochondrial calcium uptake. J Bioenerg. Biomembr. 31 (6), 551-557 265. Zhang JG and Fariss MW (2002) Thenoyltrifluoroacetone, a potent inhibitor of carboxylesterase activity. Biochem. Pharmacol. 63 (4), 751-754 266. Zhong H and Minneman KP (1999) Alphal-adrenoceptor subtypes. Eur. J Pharmacol. 375 (1-3), 261-276 267. Zhou Z and Bers DM (2002) Time course of action of antagonists of mitochondrial Ca uptake in intact ventricular myocytes. Pflugers Arch. 445 (1), 132-138 268. Zhu X, Jiang M, and Birnbaumer L (1998) Receptor-activated Ca2+ influx via human Trp3 stably expressed in human embryonic kidney (HEK)293 cells. Evidence for a non-capacitative Ca2+ entry. J. Biol. Chem. 273 (1), 133-142 269. Zimmermann B (2000) Control of InsP3-induced Ca2+ oscillations in permeabilized blowfly salivary gland cells: contribution of mitochondria. J Physiol 525 Pt 3:707-19., 707-719 105 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items