Quarterly Reviews of Biophysics 34, 3 (2001), pp. 325–472. " 2001 Cambridge University PressDOI: 10.1017/S0033583501003705 Printed in the United Kingdom325Emerging issues of connexin channels:biophysics fills the gapAndrew L. HarrisDepartment of Pharmacology & Physiology, New Jersey Medical School, University of Medicine andDentistry of New Jersey, 185 South Orange Ave, University Heights, Newark, NJ 07103, USA1. Introduction 3261.1 What? Terminology and general properties 3271.2 Why? Reasons for biophysical study 3291.3 How? Special issues for study of connexin channels 3302. Molecular and structural context 3312.1 Biochemical features 3312.2 Structures 3342.2.1 Junctional channels 3352.2.2 Hemichannels 3382.2.3 Heteromeric channels 3422.2.4 Junctional plaques 3473. Experimental approaches and issues specific to study of connexin channelphysiology 3493.1 Macroscopic currents 3493.1.1 Junctional channels 3493.1.2 Hemichannels 3543.2 Single-channel currents 3553.2.1 Junctional channels 3553.2.2 Hemichannels 3583.3 Molecular permeability 3613.3.1 A selection of tracers 3613.3.2 Junctional channels 3623.3.3 Hemichannels 3663.4 Other 3674. Structural issues 3684.1 What lines the pore? 3684.2 Docking between hemichannels 3734.2.1 Structural and molecular basis 3744.2.2 Determinants of specificity of interaction 3805. Permeability and selectivity 3815.1 Among the usual ions 3835.1.1 Unitary conductance 3835.1.2 Selectivity 3845.1.3 Nonlinear single-channel I–V relations and their molecular determinants 3865.2 Among large permeants 391Tel.: (973) 972-1620 ; Fax: (973) 972-7950 ; E-mail: aharris!umdnj.edu326 Andrew L. Harris5.2.1 Uncharged molecules 3925.2.2 Charged molecules 3935.2.3 Cytoplasmic/signaling molecules 3966. Voltage sensitivity 3996.1 Macroscopic transjunctional voltage sensitivity 4046.2 Microscopic voltage sensitivity – Vj-gating 4076.2.1 Molecular basis – voltage sensor 4076.2.2 Molecular basis – transduction and/or state stability 4096.3 Microscopic voltage sensitivity – loop gating 4126.4 Vm-gating 4147. Direct chemical modulation 4157.1 Phosphorylation 4177.2 Cytoplasmic pH and aminosulfonates 4197.3 Calcium ion 4247.4 Lipophiles 4247.4.1 Long chain n-alkyl alcohols 4257.4.2 Fatty acids and fatty acid amides 4267.4.3 Halothane 4267.5 Glycyrrhetinic acid and derivatives 4277.6 Cyclic nucleotides 4287.7 Other candidates 4308. Connexinopathies 4319. Summary 43510. Acknowledgements 43811. References 4381. IntroductionConnexins are the proteins that form the intercellular channels that compose gap junctions invertebrates. Connexin channels mediate electrotonic coupling between cells and serveimportant functions as mediators of intercellular molecular signaling. Convincingdemonstration of the latter function has been elusive, as have the experimental tools requiredfor detailed functional study of the channels. Recently, substantial progress has been made onboth fronts. Connexin channels are now known to be dynamic, multifunctional channelsintimately involved in development, physiology and pathology, and amenable to study bystate-of-the-art approaches. A host of developmental and physiological defects are caused bydefects in connexin channels, and therefore in the intercellular molecular movement theymediate. The channel structure has been determined to 7n5AHresolution within the plane ofthe membrane. Experimental paradigms have been developed that enable application of thetools of modern channel biophysics to study connexin channel structure–function. As aresult, the biophysical mechanisms and biological functions of connexin channels now enjoya vigorous and expanding experimental interest. This article focuses on the former, but withattention to issues likely to have biological consequences.327Connexin channelsConnexin channels have historically been regarded by channel physiologists as poorrelations of ‘ real ’ channels. In part, this view developed because the chemical signalingfunction of connexin channels (its function in most contexts) is less easily defined and lesslikely to depend crucially on the intrinsically compelling properties of rapid channel gatingand selectivity among atomic ions that tend to motivate channel biophysicists. Moreover, withthe exception of electrotonic coupling, the study of connexin channels emerged from cellbiology and anatomy rather than physiology.On the pragmatic side, the intercellular structure of gap junction channels makes themdifficult to study and requires analytical paradigms different from those applied to otherchannels. Single-channel recordings of connexin channels were unavailable for many years(and remain technically challenging), as were specific blockers and affinity reagents, whichhave been immensely beneficial in the study of other channels. In addition, much of thephysiological data seemed contradictory or ambiguous, something we now know largelyresults from diverse connexin channel physiologies, arising from approximately 20 connexinisoforms and their ability to form heteromeric channels.Much has changed in the last several years. Homomeric channels formed by severalconnexin isoforms have been characterized in detail using variants of commonly appliedmolecular and physiological approaches. In addition, new approaches have beendevised. The information coming from these studies provides a growing, credible and largelyself-consistent framework for understanding connexin channel structure–function. Anintriguing set of biophysical issues and questions are emerging from the recent data.It is hoped that this review will address the (usually) unspoken questions that arise in theminds of most channel physiologists when they hear about connexin channels: Are they realchannels? Do they do anything important? Do they do anything interesting from abiophysical perspective? Is there an understanding of their structure–function analogous tothat for other channels? What does the terminology mean? What are the key issues? Why isit so difficult to get answers to these questions?1.1 What? Terminology and general propertiesGap junctions are ubiquitous aggregates of intercellular channels found between most cellsacross many phyla, from metozoa to chordates (Bennett et al. 1994 ; Becker et al. 1998). They(a) span closely apposed plasma membranes, (b) provide pathways for direct current flowbetween cytoplasms of coupled cells, and (c) can be permeable to molecules the size ofcytoplasmic second messengers. Each gap