Biophysical Characteristics of Gap Junctions in Vascular Wall Cells: Implications for Vascular Biology and Disease

Brazilian Journal of Medical and Biological Research (2000) 33: 415-422 Gap junctions in vascular wall cells 415 ISSN 0100-879X

Biophysical characteristics of gap junctions in vascular wall cells: implications for vascular biology and disease

P.R. Brink1, Departments of 1Physiology and Biophysics, Institute for Molecular Cardiology, and J. Ricotta2 and 2Surgery, SUNY at Stony Brook, Stony Brook, NY, USA G.J. Christ3,4 Departments of 3Urology, and 4Physiology and Biophysics, Institute for Smooth Muscle Biology, Albert Einstein College of Medicine, Bronx, NY, USA

Abstract

Correspondence The role gap junction channels play in the normal and abnormal Key words P.R. Brink functioning of the vascular wall is the subject of much research. The · Vascular smooth muscle Department of Physiology and biophysical properties of gap junctions are an essential component in · Gap junctions Biophysics understanding how gap junctions function to allow coordinated relax- SUNY at Stony Brook ation and contraction of vascular smooth muscle. This study reviews Stony Brook, NY 11794 USA the properties thus far elucidated and relates those properties to tissue function. We ask how biophysical and structural properties such as Presented at the Meeting gating, permselectivity, subconductive states and channel type (het- “Gap Junctions in the Nervous and eromeric vs homotypic vs heterotypic) might affect vascular smooth Cardiovascular Systems: Clinical muscle tone. Implications”, Rio de Janeiro, RJ, Brazil, June 6-11, 1998.

Research supported by NIH grant GM55263. Introduction subject of modeling strategies which rely on gap junction-mediated communication to In vascular smooth muscle the role that produce tissue responses to stimuli which Received August 13, 1999 gap junctions play is integral among a num- mimic reality (8,9). For this reason, the bio- Accepted November 23, 1999 ber of other processes and parameters influ- physical characteristics of gap junction chan- encing tissue function (1-4). The ultimate nels in vascular smooth muscle and associ- tissue function, however, is coordinated re- ated vascular wall cells, such as endothelial laxation and contraction of the many cells cells, should be elucidated. In particular, for within the wall of any particular vessel. Ulti- example, are there gap junctional properties mately, the major factors which influence which lend themselves to dynamic regula- such coordinated tissue behavior are neu- tion? How many gap junction channels are ronal innervation density, cell to cell com- needed to allow cells to act syncytially? What munication and cellular excitability (1,5-7). types of gap junction channels are present in As vascular smooth muscle cells do not tend the vascular wall? Is it possible that hybrid to generate action potentials, innervation channel types exist in vascular smooth density and gap junction-mediated commu- muscle, and if so, would their presence affect nication become paramount as potential regu- tissue function? These issues are considered latory sites. The former parameter is the herein.

Braz J Med Biol Res 33(4) 2000 416 P.R. Brink et al.

Results and Discussion tion channels and subsequently compare their behavior to Cx37, Cx40 and Cx45 whenever Generic gap junction channel types in possible. In this regard, the instantaneous vascular tissues junctional current for Cx43 is linear over a large voltage range. Transjunctional voltage The subunit protein of any gap junction steps of sufficient amplitude and duration channel is the connexin and in vascular cause junctional currents to decline with smooth muscle connexin43 (Cx43) is the time, resulting in a steady-state conductance most abundant, followed by connexin40 which is a fraction of the instantaneous con- (Cx40), connexin45 (Cx45) and possibly ductance. The time course of the voltage- connexin37 (Cx37) (1,10-12). Endothelial dependent behavior is of the order of hun- cells are also able to co-express Cx43, Cx40 dreds of milliseconds to seconds (13). Cx43 and Cx37, the latter being the most abundant shows relatively symmetric voltage depend- in situ (see 1 for review). ence but has different voltage sensitivity

The putative presence of multiple con- relative to Cx37. That is, the point on the Vj nexin subtypes in a given vascular smooth axis where junctional conductance is 50% of

muscle cell has potentially important physi- the maximum (i.e., Vo) is approximately 25- ological implications. For example, there are 30 mV for Cx37 and 70-85 mV for Cx43

three types of gap junction channels possible (12,14,15). Cx40 has a Vo in the 40 mV when more than one connexin is being ex- range, while the Vo for Cx45 has been pressed in the same cell. They are as follows: reported to be 20 mV (16). Such observa- 1) a homotypic gap junction channel com- tions emphasize that, with the possible ex- posed of 12 identical connexin subunits, 2) a ception of Cx45, the voltage sensitivity of heterotypic gap junction channel where all connexins in vascular wall cells is not likely the connexins of one cell are identical but to effect dynamic changes in intercellular different from the connexins of the adjacent coupling. cell, and 3) the heteromeric gap junction Analysis of the junctional gating over channel where the connexins in any one cell many seconds to minutes has revealed, in the are not identical within a hemichannel. In case of Cx43, another wrinkle to gating, and the case of vascular smooth muscle cells the that is mode shifting. The process has been dominant connexin is Cx43. Therefore, we described in detail (13). Two important fea- will focus our discussion on properties of tures must be considered here. First, the time homotypic gap junction channels composed course to reach steady state was of the order of Cx43, Cx40, Cx45 and Cx37. Further we of 60 s or longer, and second, the mechanism will compare homotypic Cx43 and Cx37 to could not be explained as a simple reduction the in vitro documentation of Cx43-Cx37 of mean open time or increase in mean closed heterotypics and heteromerics. The relevant time for an identical population of channels. properties to be described are gating and Rather, a large portion of the channel popu- permselectivity. lation (90%) underwent a shift to prolonged silence (prolonged closure) leaving only a Homotypic gap junction membrane voltage small population of functional channels. The dependence process was reversible. Again, this is a voltage-dependent process and the magnitude In general, all mammalian homotypic gap and duration of the voltage steps (30-50 mV, junction channels display symmetric voltage with durations of many seconds) seem to be dependence. Because of its ubiquity we will outside the physiologically useful range. first characterize Cx43 homotypic gap junc- However, it should be pointed out that if this

Braz J Med Biol Res 33(4) 2000 Gap junctions in vascular wall cells 417 process is modified or driven in the presence that the permselective properties of the sub- of physiologically important ligands/intra- conductive states are very different from the cellular messengers, such as cAMP or IP3, it main state conductance has given rise to could well be a regulatory point with great much speculation. For example, if the sub- dynamic range. This latter postulate has yet conductive state is the dominant one in terms to be experimentally tested, let alone shown of dwell time, then it would necessarily fol- to be functional under physiological condi- low that the permeation rates of solutes seek- tions. Mode shifting has not been character- ing transit from cell to cell would be quite ized for Cx37 but has been demonstrated in dependent on the nature of the subconducting A7r5 cells which co-express Cx40 and Cx43 state(s). However, before even addressing (He DS, personal communication). Thus, the permselective properties, the first ques- voltage-dependent gating, a common prop- tion answered must be how much of the open erty of all the vertebrate connexins thus far time is dominated by subconductive states. studied, seems to be an unlikely tool for cells If the dwell time in any given subconductance to utilize in order to affect dynamic alter- state is significant (10-20% or greater?), then ations in cell to cell coupling mediated by the permeation characteristics of the distinct gap junctions. Whether ligand gating, linked subconducting states become of interest. For with processes such as mode shifting, might Cx43, Christ and Brink (24) have deter- be able to provide dynamic behaviors that mined that the occupancy time of the sub- are physiologically relevant to the vascular conductive state(s), for a transjunctional volt- wall cell has yet to be determined. age of 40 mV, is less than 2% for a channel whose open probability is on the order of 70- Homotypic gap junction channel 90%. In this case, even if the permselectivity conductance of the subconductive state is dramatically different from the main state it would seem The unitary conductance for Cx43 is 90- hardly relevant to macroscopic tissue behav- 120 pS (13,17- 20). Table 1 lists the unitary ior with such a low occupancy time. Con- conductances for all four connexin types versely, if gap junction channels are driven found in the vascular system. The different into subconductive states by interaction with unitary conductance values imply that dif- ligands, or alterations in second messenger ferent permselectivity properties may exist levels, such that for the majority of the open for distinct connexins. What becomes im- time the conductive and permeation rates portant then, is determining if differential are dictated by the subconductive state, conductance values translate into vessel-spe- then this property of gap junction channels cific behaviors of the vascular wall, such as, would have very telling affects on tissue for example, propagated vasodilation or con- function. striction (21).

Subconductive states: are they Table 1 - Unitary conductance of the four connexin physiologically relevant? types found in the vascular system.

All gap junction channels thus far stud- Connexin Unitary conductance Reference ied have displayed subconductive states. One (picoSiemens, pS) common feature is the predominance of sub- Cx37 350-400 22,14 conductive states with large transjunctional Cx40 160 23 voltages, the classic example of this being Cx43 90-115 13,19 Cx45 25-30 16,18 Cx37 (22) and Cx43 (24). The observation

Braz J Med Biol Res 33(4) 2000 418 P.R. Brink et al.

Channel permselectivity: its role in tissue Figure 1A shows two data sets taken physiology from Wang and Veenstra (20) for Cx43 where unitary conductance is plotted against the The homotypic Cx43 gap junction chan- cation mobility of the major cation in the nel has been studied the most with respect to pipette solution when Cl is the major anion permselectivity. A recent study by Wang and (circles). Unitary conductance was also plot- Veenstra (20) delineated the permeability of ted against anion mobility (triangles) where Cx43 to monovalent cations. The analysis the major cation is K. The solid lines repre- revealed that homotypic Cx43 is non-selec- sent linear regression fits of the data. The tive, and based on comparison of unitary thicker of the two lines represents the anion conductance, follows an Eisenmann series I data where the Y-intercept denotes the re- or II sequence (25), i.e., Rb = Cs>K>Na> sidual cation conductance of 41 pS. The Li>TMA>TEA. The anion series also shows thinner line represents the cation data where a proportionality between mobility and uni- the Y-intercept is the residual anion conduc- tary conductance. Hence Cx43 appears to be tance of 27 pS. Figure 1B represents the very poorly selective and falls into the cat- same analysis for Cx40. The circles repre- egory of Porin OmpF (26-28) which itself is sent cation data for Cx40 and the triangles permeable to both cations and anions. How- anion data. The homotypic Cx43 seems to be ever, it should be noted that the porins have a highly non-selective channel where ionic no sequence homology with the connexins. mobility is the rate limiting factor for perme- Cx40 follows the same cation sequence, but ation. For homotypic Cx40 the data imply an has a very different anion sequence as that interaction between highly mobile anion spe- described for Cx43 and Cx37. In fact, uni- cies and cations which hampers cation tran- tary conductance was slightly decreased with sit through the channel. In fact, as the anion increased ionic mobility, indicating that Cx40 mobility declines the unitary conductance is much less anion permeable and/or that the increases, indicating that some interaction cation-anion environment surrounding or between cation and anion in Cx40 channels within the channel affects the selectivity could be occurring (23). filter(s). If this latter case is in fact true, then Data for Cx37 are shown in Figure 2 more detailed analysis will clearly be neces- which reveals a potential hazard to the con- sary. ventional interpretation for such data. Spe-

Figure 1 - A, Selectivity sequence of Cx43 for cations and anions 100 ABCx43 Cx40 taken from Wang and Veenstra 150 (20). Unitary conductance is plot- 80 ted against equivalent conductiv- ity (cm2/Ohm). The triangles rep- 60 100 resent unitary conductances measured with different anions, 40 + + 50 + with K as the cation. The circles Conductance (pS) Anion varied with constant K Conductance (pS) Anion varied with constant K 20 are unitary conductance meas- Cation varied with constant Cl- Cation varied with constant Cl- ured with different cations, with 0 0 Cl- as the anion. The lines repre- 020406080 020406080 sent the linear regression fits to Equivalent conductivities Equivalent conductivities the data points. The thick solid line is the fit for the anions while the thinner line is for the cations. The y-intercept for the anions represents the conductance due to K+ (41 pS) while the y-intercept for the cation data represents the Cl- conductance (27 pS). B, Cx40 taken from Beblo and Veenstra (23). The analysis is the same as that for Cx43. With constant K+ as the anion species, increases in size and decreases in mobility resulting in increased channel conductance imply a cation/anion interaction for the more mobile anions which reduces conductance. The y-intercepts again represent the cation conductance (171 pS) and anion conductance (8 pS), respectively.

Braz J Med Biol Res 33(4) 2000 Gap junctions in vascular wall cells 419 cifically, note that the y-intercept is -49 pS, the adjacent cell or not. The junctional con- the theoretical Cl conductance. Such behav- ductance was also determined. For all the ior predicts that the unitary conductance vs connexin types studied, the minimal con- cation/anion concentration will be non-lin- ductance for observing dye transfer was in ear in the mM range. This has been shown to the 2-5 nS range. These data indicate that be the case for Cx37 (29) and is a ready many, if not all, connexins are capable of explanation for the aberrant Y-intercept. passing both cations and anions of some size For some connexins the permselectivity and charge. At issue is not the poor selectiv- properties might be dramatically altered by ity but rather the permeation rates necessary the ionic environment. Gap junction mem- to allow specific messenger molecules to brane conductance is known to be modu- activate/inhibit whole groups of cells. This lated by a number of molecules which also latter point is very important to answer if we are able to diffuse through the gap junction are going to be able to understand the role of channel (30). For example, hydrogen and gap junctions in non-excitable tissue rather calcium ions are two species which cause than continue to have speculation as the sole the channel to gate closed in sufficiently basis of further investigation. high concentration, but in turn both are able to diffuse through the channel at concentra- Are heterotypic or heteromeric channel tion ranges normally found in cells (31,32). forms important to function? It could be that cation and anion species also affect gap junction channel gating as well as An interesting issue which has yet to be the permeation path. There is now evidence addressed with regard to permselectivity is for the supposition that the concentration of whether or not heteromeric and/or hetero- monovalents affects the gating of Cx43 typic biophysical properties are sufficiently through a screening mechanism (29). But for different from homotypic channels. Suffi- Cx37 homotypic gap junction channels the ciently different, that is, to affect the cell to monovalent cation species has no detectable cell diffusion of critical substances able to effect. Rather, Mg2+ is effective in altering trigger/alter cellular responses. In the vascu- the kinetics of the voltage dependence (33). lar wall this is of particular interest because However, with respect to Cx37, there are of the connexin co-expression that occurs in still no data available on the screening ef- endothelial and vascular smooth muscle cells. fects and/or the binding of specific sites on Figure 2 - Measured unitary con- the permeation path. 400 Cx37 K ductance for Cx37 with various cations, with Cl- again remain- Another technical approach which has Cs been utilized to document the poor selectiv- ing constant. The data indicate 300 Rb that the Cl- conductance is a ity characteristics of gap junctions is the negative number. This illustrates transfer of fluorescent probes. Few attempts that the analytical approach used has limits. It implies that chan- have been made to quantify transfer rates, 200 Na nel ion interactions could play a but dye transfer has been demonstrated for Li role in determining the nature of the majority of the mammalian connexins. In TMA the permselectivity and that dif- 100 one study the transfer of carboxyfluorescein ferent ionic conditions will yield Conductance (pS) different permeation rates for and dichlorofluorescein was shown for Cx45, the same ion. TMA, Tetramethyl- Cx43, Cx40 and Cx37 (18). Millimolar a- 0 ammonium. mounts of dye were introduced into one cell Cation varied with constant Cl- of a pair via a whole cell patch. After 10 min, -100 fluorescence was monitored. Cell pairs were 020406080 scored either as having transferred probe to Equivalent conductivities

Braz J Med Biol Res 33(4) 2000 420 P.R. Brink et al.

A question which has yet to be answered is tify the role that the three generic gap junc- the gap junction types which are present in tion types might play in normal and abnor- co-expressing cells in situ. Heterotypic and mal tissue physiology. heteromeric channel types have been demonstrated in vitro, implying that the same Conclusions and clinical relevance types of channels could form in situ (14,34). In this regard, it becomes obvious that The major role played by intercellular permselectivity measurements of the hetero- communication through gap junctions in typic, heteromeric and, to a lesser extent, modulating vasomotor tone throughout the homotypic channel types in situ are not go- vascular tree has been discussed in detail ing to be readily available in the near future. elsewhere (37,38). The implication for this Dye transfer between cells co-expressing report is that it follows naturally that since connexins has been documented in situ (35) gap junctions play such a critical role in but quantitative analysis has not been forth- maintaining circulatory homeostasis, they coming to allow comparison with cells ex- are logical molecular targets for vascular pressing a single connexin. An interesting disease, and as such, for vascular therapy. observation made by Little et al. (35) was the However, it should be emphasized that there lack of dye transfer from smooth muscle are currently no published reports that have cells of the tunica media to endothelial cells either confirmed or denied a direct proximal in the tunica intima. The reverse transfer role for gap junctions in vascular disease. (from endothelial to smooth muscle cells) More specifically, it is easy to imagine was demonstrable. This type of result might how increased intercellular coupling, for well be best explained as one of the two cell example, among vascular smooth muscle types representing an infinite sink, the clas- cells might contribute to the pathology of sic example of this phenomenon being that diseases that involve heightened contractil- illustrated by Rae et al. (36). ity/impaired relaxation of the vascular smooth In vitro studies of Cx43 and Cx37 have muscle cell (i.e., hypertension, Raynaud’s shown that Cx43 and Cx37 form heterotypic disease, coronary artery disease, erectile dys- and heteromeric channels along with homo- function, etc.). In other words, an increase in typic forms (14). The most notable param- intercellular coupling could lead to greater eter which was different in cells co-express- stimulus/impulse propagation among some ing Cx34 and Cx37 was a broadening of the population of coupled smooth muscle cells,

Vo. The second feature was the occurrence thus leading to exaggerated contractile re- of channel conductances that could not be sponses, and perhaps hypertension (39,40). explained as heterotypic or homotypic. The The opposite could be true for reduced endo- different conductive states imply, as indi- thelium-dependent relaxation responses. In cated before, different permselectivity prop- addition, restenosis and atherosclerosis (41), erties. This observation points to the issue disease states which result in significant re- already raised, i.e., that even if the permse- modeling of the vascular wall, might be ex- lectivity is different, is it different enough to pected to profoundly alter the degree of in- affect tissue physiology? There is a second tercellular coupling and/or be affected by perplexing problem. In a given vascular tis- altered cell-to-cell communication among sue co-expressing distinct connexins, what vascular wall cells. The implication of such population of channel types are there? What changes to vascular function have hardly fraction of the total number of channels is been investigated. Moreover, it should be homotypic, heterotypic or heteromeric? With- emphasized that the role of endothelial- out this information it is impossible to quan- smooth muscle cell coupling in the regula-

Braz J Med Biol Res 33(4) 2000 Gap junctions in vascular wall cells 421 tion of vascular tone is just beginning to be cal alterations in second messenger mol- studied. There are many possibilities, and it ecules/ions that result in dramatic changes in is clear the real work has just begun. the documented biophysical behaviors of On the other hand, theoretical studies connexins, then the role of gap junctions suggest that the degree of coupling among could be drastically altered. Dynamic ligand vascular smooth muscle cells may already be gating of substate dwell times or permselec- such that 10-fold or greater increases in cou- tivity properties, for example, or the pres- pling would be required for physiologically ence of significant heteromeric or hetero- relevant changes in vascular contractility typic channel types, could have profound (8,9). This is not to say that altered connexin implications for normal vascular physiology expression would play no role in vascular and disease. Certainly the potential clinical pathology, but that intercellular communi- correlates of altered intercellular communication is so critical to vascular function, that cation can be visualized, but have not yet there is significant plasticity or reserve to been documented. On this basis, this would prevent it from becoming the focal point for seem to be a very fruitful area of future tissue pathology. Of course all of the latter research. considerations are based on the known biophysical characteristics of Cx43 in human Acknowledgments vascular smooth muscle, i.e., long open times, high open probabilities, physiologically ir- The authors would like to acknowledge relevant subconductance states, and lack of Dr. S.V. Ramanan, SUNY at Stony Brook, selectivity. In this regard, however, if there NY, USA, who kindly donated the data for are indeed physiological or pathophysiologi- Figure 2.

References

1. Christ GJ, Spray DC, ElSabban M, Moore 6. Hirst GDS & Edwards FR (1989). Sympa- gap junctions. In: Perrachia C (Editor), Gap LK & Brink PR (1996). Gap junctions in thetic neuroeffector transmission in ar- Junctions. Academic Press, New York. vascular tissues: Evaluating the role of teries and arterioles. Physiological Re- 13. Brink PR, Ramanan SV & Christ GJ (1996). intercellular communication in the modu- views, 69: 546-604. Human connexin43 gap junction channel lation of vasomotor tone. Circulation Re- 7. Burnstock G (1970). Structure of smooth gating evidence for mode shifts and/or search, 79: 631-646. muscle and its innervation. In: Bulbring E, heterogeneity. American Journal of Phys- 2. Segal SS & Kurjiaka DT (1995). Coordina- Brading AF, Jones AW & Tomita T (Edi- iology, 271: C321-C331. tion of blood flow control in the resis- tors), Smooth Muscle. Williams & Wilkins, 14. Brink PR, Cronin K, Banach K, Peterson E, tance vasculature of skeletal muscle. Baltimore, MD, 1-69. Westphale EM, Seul KH, Ramanan SV & Medicine and Science in Sports and Exer- 8. Christ GJ, Brink PR & Ramanan SV (1994). Beyer EC (1997). Evidence for hetero- cise, 27: 1158-1164. Dynamic gap junctional communication: a meric gap junction channels formed from 3. Beach JM, McGahren ED & Duling BR delimiting model for tissue responses. rat connexin43 and human connexin37. (1998). Capillaries and arterioles are elec- Biophysical Journal, 67: 1335-1344. American Journal of Physiology, 273: trically coupled in hamster cheek pouch. 9. Ramanan SV, Brink PR & Christ GJ (1998). C1386-C1396. American Journal of Physiology, 275 (Part Neuronal innervation, intracellular signal 15. Valiunas V, Bukaukas F & Weingart R 2): H1489-H1496. transduction and intercellular coupling: a (1997). Conductances and selective per- 4. Pacicca C, Schaad O & Beny J (1996). model for syncytial tissue responses in meability of connexin43 gap junction Electrotonic propagation of kinin-induced, the steady state. Journal of Theoretical channels examined in neonatal rat heart endothelium-dependent hyperpolariza- Biology, 193: 69-84. cells. Cirulation Research, 80: 708-719. tions in pig coronary smooth muscles. 10. Beyer EC (1993). Gap junctions. Interna- 16. Moreno AP, Liang JG, Beyer EC & Spray Journal of Vascular Research, 33: 380- tional Review of Cytology, 137: 1-37. DC (1995). Properties of gap junction 385. 11. Brink PR (1998). Gap junctions in smooth channels formed of connexin 45 endog- 5. Bevan JA & Torok J (1970). Movement of muscle. Acta Physiologica Scandinavica, enously expressed in human hepatoma norepinephrine through the media of rab- 164: 349-356. (SKHep1) cells. American Journal of Phys- bit aorta. Circulation Research, 27: 325- 12. Brink PR, Valunias V & Christ GJ (2000). iology, 268: C356-C365. 331. Homotypic, heterotypic and heteromeric 17. Moreno AP, Campos de Carvalho AC,

Braz J Med Biol Res 33(4) 2000 422 P.R. Brink et al.

Christ GJ & Spray DC (1993). Gap junc- muscle cells. Circulation Research, 84: form heteromeric hemichannels. Journal tional communication between human 797-803. of Biological Chemistry, 270: 6768-6772. corpus cavernosum smooth muscle cells 25. Eisenmann G & Horn R (1983). Ionic se- 35. Little TL, Xia J & Duling BR (1995). Dye in culture: Gating behavior and single lectivity revisited: the role of kinetic and tracer defines differential endothelial and channel events. American Journal of equilibrium processes in ion permeation smooth muscle coupling patterns within Physiology, 264: C80-C92. through channels. Journal of Membrane the arteriolar wall. Circulation Research, 18. Campos de Carvalho AC, Moreno AP, Biology, 331: 599-635. 76: 498-504. Christ GJ, Melman A, Roy C, Hertzberg 26. Benz R (1986). Analysis and chemical 36. Rae J, Bartling C, Rae J & Mathias RT EL & Spray DC (1993). Gap junctions modification of bacterial porins. In: Miller (1996). Dye transfer between cells of the formed of connexin43 interconnect C (Editor), Ion Channel Reconstitution. lens. Journal of Membrane Biology, 150: smooth muscle cells of the human corpus Plenum Publishing Corp., New York. 89-103. cavernosum. Journal of Urology, 149: 27. Benz R, Kanko K & Langer P (1980). Pore 37. Christ GJ, Brink PR, Xhao W, Moss J, 1568-1575. formation by the matrix protein (porin) of Gondra J, Roy C & Spray DC (1993). Gap 19. Veenstra RD, Wang HZ, Beblo DA, Chilton Escherichia coli in planar bilayer mem- junctions modulate tissue contractility and MG, Harris AL, Beyer EC & Brink PR branes. Annals of the New York Academy a1-adrenergic agonist efficacy in isolated (1995). Selectivity of connexin-specific of Sciences, 258: 13-24. rat aorta. Journal of Pharmacology and gap junctions does not correlate with 28. Brink PR & Fan SF (1989). Patch clamp Experimental Therapeutics, 266: 1054- channel conductance. Circulation Re- recordings form membranes which con- 1065. search, 77: 1156-1165. tain gap junction channels. Biophysical 38. Christ GJ (1995). Modulation of a1-adre- 20. Wang H-Z & Veenstra RD (1997). Monova- Journal, 56: 579-593. nergic contractility in isolated vascular tis- lent ion selectivity sequences of the rat 29. Banach K, Ramanan SV & Brink PR (1998). sues by heptanol: a functional demonstra- connexin 43 gap junction channel. Journal Homotypic hCx37 and rCx43 and their tion of the potential importance of inter- of General Physiology, 109: 491-507. heterotypic form. In: Werner R (Editor), cellular communication to vascular re- 21. Segal SS, Welsh DG & Kurjiaka DT (1999). Gap Junction. IOS Press, Amsterdam, 76- sponse generation. Life Sciences, 56: Spread of vasodilation and vasoconstric- 80. 709-721. tion along feed arteries and arterioles of 30. Burt J & Spray DC (1988). Inotropic agents 39. Watts SW, Tsai ML, Loch-Caruso R & hamster skeletal muscle. Journal of Phys- modulate gap junctional conductance be- Webb RC (1994). Gap junctional commu- iology, 516: 283-291. tween cardiac myocytes. American Jour- nication and vascular smooth muscle re- 22. Veenstra RD, Wang HZ, Beyer EC, nal of Physiology, 254: H1206-H1210. activity: use of tetraethylammonium chlo- Ramanan SV & Brink PR (1994). Con- 31. Spray DC, Harris AL & Bennett MVL ride. Journal of Vascular Research, 31: nexin37 forms high conductance gap junc- (1981). Equilibrium properties of a volt- 307-313. tion channels with subconductance state age-dependent junctional conductance. 40. Watts SW & Webb RC (1996). Vascular activity and selective dye and ionic per- Journal of General Physiology, 77: 77-93. gap junctional communication is in- meabilities. Biophysical Journal, 66: 1915- 32. Loewenstein W (1981). Junctional inter- creased in mineralocorticoid-salt hyper- 1928. cellular communication: The cell to cell tension (see comments). Hypertension, 23. Beblo DA & Veenstra RD (1997). Monova- membrane channel. Physiological Re- 28: 888-893. lent cation permeation through the con- views, 61: 829-913. 41. Polacek D, Bech F, McKinsey JF & Davies nexin 40 gap junction channel. Journal of 33. Ramanan SV, Brink PR, Varadaraj K, PF (1997). Connexin43 gene expression General Physiology, 109: 509-522. Peterson E, Schirrmacher K & Banach K in the rabbit arterial wall: effects of hyper- 24. Christ GJ & Brink PR (1999). An analysis (1999). A three-state model for con- cholesterolemia, balloon injury and their of the presence and physiological rel- nexin37 gating kinetics. Biophysical Jour- combination. Journal of Vascular Re- evance of subconducting states of con- nal, 76: 2520-2529. search, 34: 19-30. nexin43-derived gap junction channels in 34. Stauffer KA (1995). The gap junction pro- cultured human corporal vascular smooth teins B1-connexin and B2-connexin can

Braz J Med Biol Res 33(4) 2000