Simply the Best?
Identity of Vascular cGMP-Dependent Cl− Current Revealed
See related article, pages 864–872
Proteins encoded by best1 to -3 genes are implicated as molecular correlates of calcium-activated chloride channels in epithelia. In this issue of Circulation Research, Matchkov et al present compelling evidence that best-3 expression is essential for the generation of calcium-sensitive cGMP-dependent chloride channels in rat mesenteric artery.1
Chloride channels are enigmatic beasts. Numerous phenotypes exist, as determined by their mode of activation, channel kinetics, and pore properties, but the molecular identity has only really been identified for the voltage-dependent Cl− channels (CLCs) and the cAMP-dependent, cystic fibrosis transmembrane regulator (CFTR) channels. For the other types of Cl− channels, there is far less certainty about the molecular identity. Many candidates for the swelling-activated Cl− channel have been promulgated that ultimately have been repudiated.2 Similarly, the molecular identity of the Ca2+-gated Cl− channel, common to vascular smooth muscle cells, secretory cells, and cardiomyocytes remains to be elucidated.3 In this issue of Circulation Research, Matchkov et al suggest that we can add 1 more Cl− current to the list of identified species.1 Perversely, the calcium-sensitive cGMP-dependent Cl current, which is the focus of the work by Matchkov et al, is among the newest additions to the Cl− current family. This article completes a body of work by this group that begins and ends with Circulation Research. In 2001 Peng et al, proposed that the oscillatory contractile activity driven by membrane receptor agonists (vasomotion) in rat mesenteric arteries was generated by Cl− current activation mediated by a rise in cGMP within the smooth muscle cell.4 Later studies characterized the biophysical and pharmacological properties of the cGMP-dependent Cl− current (IClcGMP) in isolated smooth muscle cells both at macroscopic5 and single-channel level.6 These studies showed that IClcGMP was activated by a rise in Ca2+ but differed from the classic Ca2+-gated Cl− channel currents (IClCa) recorded in most smooth muscle cells. Thus, IClCa exhibits distinctive voltage-dependent kinetics, a small unitary conductance, and high thiocyanate permeability, and activation does not require cGMP.7 In contrast, IClcGMP is time-independent, has an intermediate conductance with a thiocyanate permeability just greater than chloride, and an obligatory requirement for cGMP (Kd, ≈5 μmol/L). Moreover, IClcGMP is abolished by 10 μmol/L ZnCl2, which does not affect IClCa. Matchkov et al5 commented that many of the characteristics of IClcGMP were similar to currents generated by the then newly identified bestrophins encoded by the best1 to -3 genes. There is now a considerable amount of evidence based on expression profiling and interference RNA work that bestrophins underlie Ca2+-activated CLCs in epithelial cells.8 It would seem from the present work by Matchkov et al that their comment 4 years ago5 may have been rather insightful.
In the present study, Matchkov et al conclude that proteins encoded by best-3 constitute the cGMP-dependent Cl− channel in rat mesenteric artery.1 This was based on bioinformatics studies that revealed that the C terminus of best-3 proteins contains cGMP phosphorylation motifs (see the online data supplement of the article1), comparison of the best-3 expression profile and the IClcGMP distribution profile, consolidated by the use of interference RNA. Thus, abundant best-3 message was detected by RT-PCR in rat aorta and mesenteric artery, as well as A7r5 cultured myocytes, in which robust IClcGMP was recorded but was less abundant in pulmonary artery myocytes, in which IClcGMP is negligible. Best-1 and best-2 message was also detected in these cells but at lower levels. Interestingly, Leblanc et al7 readily detected message for all 3 best genes in rabbit pulmonary artery, perhaps suggesting a species difference. Two different short interfering (si)RNA oligonucleotides directed to different exons of best-3 produced ≈80% and 55% reduction in best-3 RNA and protein, respectively, in both mesenteric arteries and A7r5 cells, which was not mimicked by a mutated form of siRNA containing 2 mismatched nucleotides. Crucially targeted siRNA ablated the IClcGMP in both freshly dispersed and cultured types of myocytes. No effects with the best-3 siRNA were observed on best-1 or best-2 mRNA. Moreover, the classic IClCa was purportedly unaffected by best-3 siRNA. These data seem to suggest unequivocally that best-3 encoded proteins constitute the Ca2+-sensitive, cGMP-dependent Cl− channel in vascular smooth muscle. However, there are some considerations. First of all, although many of the properties of IClcGMP are similar to bestrophin currents, there are some disparities. Although the single channel conductance of best-3 has not been determined, channels formed by the overexpression of best-1 have a unitary conductance of ≈2 pS,9 whereas cGMP-dependent Cl− channels have multiple conductances that range from 15 to 55 pS.6 In addition, the thiocyanate/chloride permeability ratio for bestrophins and IClcGMP differs (8 compared to 2, respectively),6 and the latter has a rather steep Ca2+ dependence compared to the bestrophins. Moreover, there is no experimental data to show that currents generated by the overexpression of best genes can be activated by a rise in cGMP. Another consideration is that bestrophins often form heteromultimers and can also regulate other ion channels, especially voltage-dependent calcium channels. The data of the present study do not rule out a role for best-3–encoded proteins as a regulatory subunit or as a constituent of a multimeric protein complex (something acknowledged by the authors). It is still unclear how a cGMP-dependent Cl− channel functions in a coordinated response like vasomotion, especially because the involvement of this channel as an excitatory mechanism seems paradoxical in the light of cGMP being a relaxant molecule. Consequently, regardless of what membrane events occur NO released from the endothelium is likely to vasorelax. These concerns are not, however, the focus of this report, and despite the earlier considerations, it would be churlish to not take the data of Matchkov et al at face value and to celebrate the identification of another Cl− channel family member. Logically, the next step would be to apply the siRNA technology to the whole tissue to see whether contractile oscillations are terminated by best-3 knockdown. This would complete the story from functional response to molecular candidate.
Sources of Funding
Work in the laboratory of the author is supported by the British Heart Foundation.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
Matchkov VV, Larsen P, Bouzinova EV, Rojek A, Briggs Boedtkjer DM, Golubinskaya V, Pedersen FS, Aalkjær C, Nilsson H. Bestrophin-3 (vitelliform macular dystrophy 2-like 3 protein) is essential for the cGMP-dependent calcium-activated chloride conductance in vascular smooth muscle cells. Circ Res. 2008; 103: 864–872.
Peng H, Matchkov V, Ivarsen A, Aalkjaer C, Nilsson H. Hypothesis for the initiation of vasomotion. Circ Res. 2001; 88: 810–815.
Matchkov VV, Aalkjaer C, Nilsson H. A cyclic GMP-dependent calcium-activated chloride current in smooth-muscle cells from rat mesenteric resistance arteries. J Gen Physiol. 2004; 123: 121–134.
Chien LT, Zhang ZR, Hartzell HC. Single Cl− channels activated by Ca2+ in Drosophila S2 cells are mediated by bestrophins. J Gen Physiol. 2006; 128: 247–259.