Vascular conducted responses are believed to play a central role in controlling the microcirculatory blood flow. The responses most likely spread through gap junctions in the vascular wall. At present, four different connexins (Cx) have been detected in the renal vasculature, but their role in transmission of conducted vasoconstrictor signals in the preglomerular arterioles is unknown. Connexin mimetic peptides were previously reported to target and inhibit specific connexins. We, therefore, investigated whether conducted vasoconstriction in isolated renal arterioles could be blocked by the use of mimetic peptides directed against one or more connexins. Preglomerular resistance vessels were microdissected from kidneys of Sprague-Dawley rats and loaded with fura 2. The vessels were stimulated locally by applying electrical current through a micropipette, and the conducted calcium response was measured 500 μm from the site of stimulation. Application of connexin mimetic peptides directed against Cx40, 37/43, 45, or a cocktail with equimolar amounts of each, did not inhibit the propagated response, whereas the nonselective gap junction uncoupler carbenoxolone completely abolished the propagated response. However, the connexin mimetic peptides were able to reduce dye coupling between rat aorta endothelial cells shown to express primarily Cx40. In conclusion, we did not observe any attenuating effects on conducted calcium responses in isolated rat interlobular arteries when exposed to connexin mimetic peptides directed against Cx40, 37/43, or 45. Further studies are needed to determine whether conducted vasoconstriction is mediated via previously undescribed pathways.
- conducted responses
- intercellular communication
within the vascular wall, endothelial cells and vascular smooth muscle cells (VSMC) are coupled together by low-resistance gap junctions (myo-endothelial, endothelial-endothelial, and myocyte-myocyte), allowing transmission of both electrical signals (44) and small signaling molecules between cells (10). This intercellular communication is believed to play an important role in the regulation of blood flow in the microcirculation.
Vascular conducted responses (VCR) coordinate microcirculatory control in several vascular beds, and these responses are the consequence of direct intercellular communication between cells of the vascular wall. As a consequence of intercellular communication, localized stimulation of a vessel will result in a local response (dilation or constriction) and an additional remote response, which may be detected several millimeters from the site of stimulation. Vascular conducted responses most likely travel through gap junctions (GJ).
VCRs may play a particularly important role in the renal microcirculatory control. Both the tubuloglomerular feedback response and renin secretion are controlled by signals from the macula densa (MD) cells in the thick ascending limb in the loop of Henle. However, MD is only in proximity with the very last part of the afferent arteriole. An efficient control of preglomerular vascular tone and renin secretion requires that signals from MD reach the relevant cell population [renin-containing juxtaglomerular cells and/or vascular smooth muscle cells]. This most likely involves a VCR in preglomerular vessels, traveling from cell to cell through GJs (34, 38).
In the renal vasculature, several isoforms of connexins have been found both as mRNA and protein, namely, Cx37, Cx40, and Cx43 (1, 21, 45). Furthermore, studies from Cx45-deficient mice with targeted replacement of the Cx45-coding region with the lacZ reporter gene indicate that Cx45 may also be present in the renal vasculature (28). It should, however, be noted that immunostaining of renal preglomerular arterioles from rats and mice show detectable gap junctions formed by Cx37, Cx40, and Cx43, primarily in the endothelial cells (1, 21, 24, 45) and not in the vascular smooth muscle cells.
Inhibition of gap junctions by use of connexin mimetic peptides designed to inhibit specific connexin isoforms has been reported by several authors (6, 17, 23, 36). These peptides are homologs to specific sequences on the extracellular loops of the targeted connexin. Gap26 peptides are homologs to sequences on the first extracellular loop, whereas Gap27 peptides are homologs to sequences on the second loop. These areas are believed to be important for two connexons to dock and form a functional channel (17). Thus, Gap26 and Gap27 peptides are believed to either inhibit the formation of new functional gap junctions between neighboring cells or to affect the gating properties of already established gap junctions (17).
The information on the role of gap junctions in conducted vasoconstriction is sparse. In Cx40 knockout mice, conduction of vasoconstriction induced by local application of high K+ in arterioles from the cremaster muscle was observed to be unchanged compared with wild-type mice, whereas the conduction of vasodilation induced by ACh or bradykinin was severely attenuated (12). In contrast, in Cx37 knockout mice K+-induced conducted vasoconstriction was significantly attenuated (32).
Regarding local responses, the nonselective gap junction uncouplers heptanol and 18α-glycyrrhetinic acid (18α-GA) and the more specific connexin mimetic peptide directed against Cx37 and Cx43 (37/43Gap27) abolished pressure-induced vasoconstriction (myogenic response) (15, 29). However, vascular constriction in response to K+ or phenylephrine was not affected.
The aim of the present investigation was to examine the effect in vitro of impairment of gap junctional communication on conducted calcium responses in renal arterioles. To block renal vascular gap junctions, we used connexin mimetic peptides designed to specifically inhibit Cx37 and Cx43 (37/43Gap27), Cx40 (40Gap27), and Cx45 (45Gap27). As Cx40 seems to be the most abundantly expressed isoform in renal arterioles (1), the function of the peptide 40Gap27 was verified by blocking dye transfer in cultured rat endothelial aorta cells expressing Cx40. We tested the hypothesis that these peptides can reduce the propagation of locally induced calcium responses in isolated preglomerular renal arterioles.
Endothelial cell experiments.
Rat aorta endothelial cells (RAOEC; R304-05, Cell Applications, San Diego, CA) cells were grown at 37°C in RAOEC growth medium (Cell Applications).
Immunostaining and confocal microscopy.
To test the specificity of the antibodies used in this study, we transfected plasmids encoding either Cx37, Cx40, Cx43, or Cx45 into connexin-free HeLa cells using the transfection reagent Fugene 6 (Roche, Diagnostics, Vienna, Austria). Every antibody was checked against cells transfected with either of the four connexin isoforms, as well as nontransfected cells.
To test the antibody specificity or the expression of connexins in RAOEC, cells were fixed in 2% paraformaldehyde/PBS for 15 min. After 20 min in 50 mM NH4Cl2 to stop the fixation, the cells were permeabilized in 0.2% Triton X-100 in PBS/4% BSA for 15 min. Then, the cells were incubated with a primary antibody [Cx37 1:50 (Alpha Diagnostics International, San Antonio, TX), Cx40, 1:500 (AB1726; Chemicon, Temecula, CA), Cx43, 1:500 35-5000; Zymed, San Francisco, CA) or Cx45, 1:500 (Chemicon, MAB3100)] overnight at 4°C. The cells were washed in PBS and incubated for 45 min at room temperature with a secondary antibody (ALEXA 488 conjugated) and phalloidin-rhodamine (both Molecular Probes). A Leica TCS SP2 confocal microscope using a 63 × 1.2 NA water immersion objective was used to generate the images.
RAOEC were seeded on collagen-coated coverslips (10 mg/16 ml EtOH, Sigma, type VI) in dishes 3 cm in diameter two days before the experiment (∼500,000 cells/dish) to form a nearly confluent layer on the day of the experiment. The cells were used in up to seven passages from delivery, and expression of Cx40 was monitored throughout the experiments by immunostaining. At the day of the experiment, RAOECs were incubated in RAOEC growth medium with or without connexin mimetic peptide (40Gap27, 600 μM or 40Gap27, 37/43Gap27, 45Gap27, 300 μM each) for 1 h. As a positive control, we applied carbenoxolone (Sigma-Aldrich, St. Louis, MO; 100 μM), which is a nonspecific blocker of gap junctions, that is, it blocks gap junctions regardless of their isoform composition. RAOEC were incubated for 30 min in media containing carbenoxolone. In all experiments, 4% BSA (Sigma-Aldrich) was added to the medium, and the pH was adjusted to 7.4.
The cover slips were mounted in a chamber containing Tyrode's buffer (in mM): 136 NaCl, 4 KCl, 5 HEPES, 5 MES, 0.8 MgCl2, and 1.8 CaCl2, pH 7.4.
For dye injection, pipettes were pulled from glass capillaries (GC150F-15; Harvard Apparatus, Edenbridge, UK) using a Flaming-Brown microelectrode puller (P-87, Sutter Instruments, Novato, CA). The electrodes were filled with Lucifer yellow solution (10 mM LY-dilithium salt, L0259; Sigma-Aldrich). When a cell was impaled, dye was allowed to diffuse into the cell for 30 s, and the pipette was then withdrawn. Three cells on each coverslip were impaled, and dye was allowed to diffuse to the neighboring cells for 5 min. Cells were then illuminated by a 480-nm light from a monochromator (J&M, Aalen, Germany), and the emitted fluorescence image at 510 nm was collected using a SensiCam charge-coupled device camera (PCO, Kelheim, Germany). The setup was controlled by Imaging Workbench 4 (INDEC BioSystems, Santa Clara, CA). For each image, the number of cells receiving dye from the injected cell was counted.
The experimental conditions for the individual images were blinded to the operator during the experiment and the subsequent analysis.
Isolated Arteriole Experiments
Isolation and fura-2 loading of the preparation.
Preglomerular resistance vessels were microdissected from Sprague-Dawley rats (250–300 g body wt; Taconic, Lille Skensved, Denmark), as previously described (37). Briefly, thin slices (0.5–1 mm) of the kidney were cut from the midregion and transferred to a dissection dish containing a physiological salt solution (PSS) (in mM): 135 NaCl, 5.0 KCl, 1.0 CaCl2, 1.0 MgCl2, 10 HEPES, and 5.0 d-glucose, pH 7.38–7.42, with 0.5% BSA (Sigma). All solutions were equilibrated with atmospheric air. The microdissection was performed under a microscope using sharpened forceps. An interlobular artery was localized at its origin from an arcuate artery, and tubular structures were removed. Only the most-distal part of the interlobular artery and cortical afferent arterioles were used. The kidney were discarded if no preparation were obtained during the first 90 min of dissection. After completion of dissection, the arteriole was loaded with 5 μM fura-2-AM for 60 min or with fura 2-AM and connexin mimetic peptide (600 μM) in the dark at room temperature. Fura-2 loading was facilitated with 0.01% Pluronic F-127 (Sigma-Aldrich). The vessel was transferred to a PSS-containing chamber on the stage of an inverted microscope (Olympus IX 50). The ends of the vessel were aspirated into glass holding pipettes to obtain mechanical stability. For the carbenoxolone experiments, carbenoxolone (100 μM) was added directly to the chamber after control electrical stimulation of the vessel. After 10 min of incubation, the vessel was stimulated again.
The peptides were dissolved and used in the following concentrations: 40Gap27 (600 μM, sequence SPRTEKNVFIV, n = 11), 37/43Gap27 (600 μM, sequence SRPTEKTIFII, n = 7), 45Gap27 (600 μM, sequence SPRTEKTIFLL, n = 10), or 40Gap27, 37/43Gap27, 45Gap27 (300 μM each, n = 7). The concentration of peptides was chosen from previous experiments showing inhibitory effect of the peptides on conducted vasodilatation in isolated arterioles (13, 18, 39). 40Gap27 and 45Gap27 were both dissolved in PSS containing 0.5% BSA. 37/43Gap27 was dissolved as a 6 mM stock in distilled water and further dissolved to 600 μM in PSS containing 0.5% BSA. Carbenoxolone (100 μM, n = 5) was dissolved in PSS containing 0.5% BSA. The concentration of carbenoxolone was chosen from previous studies showing inhibitory effect of carbenoxolone in isolated arterioles (14, 25, 26, 43). All solutions were adjusted to pH 7.4.
Measurements of intracellular Ca2+ concentrations ([Ca2+]i).
A 40× quartz oil immersion objective was used for measurements of [Ca2+]i. A digital video camera (PCO, Kelheim, Germany) and Image Workbench software (INDEC Biosystems, Santa Clara, CA) were used. The vessel was visualized on a computer screen, and an area for measurement of [Ca2+]i was encircled using a software-based routine. UV light of alternating 340- and 380-nm wavelengths was provided from a monochromator controlled by the software. The fluorescent emission was detected by the digital video camera, and the ratio of fluorescence obtained with 340-nm excitation to that obtained with 380-nm excitation (R340/380) was calculated. Changes in this ratio were used as an index for changes in [Ca2+]i (19). In preliminary experiments, autofluorescence was measured in nonloaded preparations and was found to be <10% of the fluorescence of the fura 2-loaded vessels.
The experimental solutions (room temperature) were added in a volume large enough to allow total exchange of the fluid in the experimental chamber. The volume in the experimental chamber was kept constant during replacement of fluids by a vacuum suction system. The viability of each preparation was assessed by stimulation with a norepinephrine (1 μM) and a 50 mM K+ solution. Vessels not responding promptly were discarded. The increase in [Ca2+]i was accompanied by a visible contraction of the vessel. Thereafter, the [Ca2+]i response to electrical stimulation was studied. At the end of each experiment, the viability of the vessel was reassessed by the addition of 50 mM K+.
Local electrical stimulation was performed as previously described (20, 38, 41). Glass pipettes, pulled to an outer tip diameter of 8–10 μm, were filled with 2 M NaCl (resistance 0.7–0.8 MΩ) and were placed in an electrode holder attached to a micromanipulator. The vessel was stimulated by a train of continuous unipolar current pulses (2.5 Hz frequency, 200-ms pulse duration, +90 V amplitude), which were obtained from an isolation unit controlled by a Grass stimulator. The pipette tip was positioned close to the vessel, ∼500 μm proximal to the area where [Ca2+]i was measured. Control experiments verified that when the electrode was placed at the same distance (500 μm) from the area of measurement but removed from the vessel, the [Ca2+]i response to electrical stimulation was abolished. This indicates that the distant response is due to the spread along the vessel wall and not to a generalized electrical field in the experimental chamber.
The dye injection results are presented as a mean from each coverslip (i.e., a mean of injections in 3 different cells). The experiments were compared using unpaired two-tailed Student's t-test. Results from the isolated renal arterioles are presented as the increase in the ratio of fluorescence (R340/380) from the control situation to the experimental situation (Δ340 nm/380 nm). The results were normalized to the ratio measured 30 s before stimulation. Results are means ± SE of original data. Because of variance inhomogeneity (Levene's test) and a nonnormal distribution of the fluorescence data, the data were analyzed by the Wilcoxon or the Mann-Whitney U-test as appropriate. In cases in which multiple groups were compared, the data were initially evaluated by the Kruskal-Wallis test. If this were significant, individual means were then compared by the Mann-Whitney U-test with Bonferroni correction. P < 0.05 was considered to be statistically significant.
Endothelial Cell Experiments
To test for specificity of the applied antibodies (Ab), we transfected HeLa cells, which are devoid of connexins with constructs encoding the different connexins. We tested each antibody against HeLa cells transfected with Cx37, Cx40, Cx43, or Cx45. The Cx37, Cx40, and Cx45 antibodies only reacted with cells transfected with the corresponding connexin (Fig. 1A, B, and D, respectively). The Cx43 Ab reacted with Cx43-transfected cells (Fig. 1C) but also showed a faint signal in some Cx40-transfected cells (Fig. 1E). This signal did not resemble normal connexin distribution and seemed to originate from either preprocessed connexins in the endoplasmic reticulum or connexins located in lysosomal or proteasomal compartments.
To confirm the presence of connexins in cultured rat aorta endothelial cells, we tested the endogenous RAOEC connexin expression histochemically. The endothelial cells showed low expression of Cx37 localized both intracellularly and in the plasma membrane (Fig. 2A). Connexin 40 was clearly localized to the plasma membrane between juxtaposed cells and with only weak intracellular signals (Fig. 2B). The cells also showed a weak Cx43 signal, which was located primarily intracellularly (Fig. 2C). A somewhat stronger signal was seen in cells stained for Cx45. This connexin seemed to be located mainly intracellularly in the RAOEC cells (Fig. 2D).
To validate the functionality of the connexin mimetic peptides, Lucifer yellow was injected into a single cell, and the number of neighboring cells receiving the dye was counted (Fig. 3). The experimental settings were blinded to the experimenter. In untreated RAOEC cells (controls), 6.2 ± 0.9 cells received the dye (Fig. 3B, n = 9). After 1 h of incubation with 600 μM 40Gap27, the number of cells receiving the dye after 5 min was reduced to 3.5 ± 0.7 (P < 0.05, Fig. 3D, n = 7) or by 44 ± 10%. To test whether the remaining dye transfer could be due to connexins other than Cx40, we performed an additional series of experiments in which RAOEC cells were incubated for 1 h with a cocktail of all three peptides (40Gap27, 37/43Gap27, and 45Gap27; 300 μM, n = 8). Again, the peptides significantly reduced the number of cells receiving dye from 4.2 ± 0.7 (control, n = 9) to 2.1 ± 0.4 (P = 0.01, data not shown) or by 50 ± 9%, but the inhibition of dye transfer between the cells was not significantly different from that seen when the cells were incubated only with 40Gap27. Treatment for 30 min with carbenoxolone further reduced the number of cells containing the dye to 1.4 ± 0.2 (P < 0.01, Fig. 3F; n = 6) or by 78 ± 3%.
Isolated Arteriole Experiments
To examine the effect of gap junction mimetic peptides on conducted vascular calcium signals, renal arterioles were incubated in PSS (control vessels), PSS containing one or more mimetic peptides, or PSS containing carbenoxolone. Both the peak response obtained during the first 5 s of the registration and the sustained response after 30 s of stimulation were examined.
Because of variation in baseline 340 nm/380 nm ratio (R340/380) between the different groups, the responses were normalized to the ratio obtained 30 s before stimulation.
In control arterioles (n = 16) electrical stimulation elicited a significant conducted increase in [Ca2+]i 500 μm from the stimulation site (P < 0.05) (Fig. 4). The response reached a sustained plateau that was constant for at least 30 s. In arterioles treated with peptide against Cx40 (n = 11), Cx37/43 (n = 7), Cx45 (n = 10), or a cocktail of all three peptides (n = 7) electrical stimulation also induced a significant (P < 0.05) conducted increase in [Ca2+]i. The conducted increases in [Ca2+]i were not reduced by any of the peptides compared with the control group (Fig. 4). In arterioles treated with 37/43Gap27, there was a nonsignificant tendency to a larger response. Treatment with 100 μM of the gap junction uncoupler carbenoxolone for 10 min (n = 5) abolished the conducted peak increase in [Ca2+]i in response to electrical stimulation (NS compared with 0%). In addition, it was significantly smaller than the response in the control group (P < 0.01) (Fig. 4). The sustained increase in [Ca2+] (30-s response) was also abolished by carbenoxolone (NS compared with 0%). However, compared with the response in the control group, the difference did not reach statistical significance (Fig. 4). Nevertheless, 50 mM K+ added to the bath still elicited an increase in Ca2+ concentration in all arterioles (cf. example in Fig. 5), verifying that the vessels were still responsive. Specifically, 50 mM K+ in the bath increased the 340 nm/380 nm ratio by 18.7 ± 4.9% in the control group, and by 17.8 ± 5.3% in the carbenoxolone group, values that were not significantly different. Baseline 340 nm/380 nm ratio did not change during the 10-min incubation period with carbenoxolone (before 0.64 ± 0.15 vs. after 0.62 ± 0.16, NS), nor in time control vessels incubated in PSS for 10 min (before 0.91 ± 0.17 vs. after 0.88 ± 0.16, NS).
Little is known about the role of connexins in the propagation of vasoconstriction and even less is known about the role of specific connexin isoforms. Therefore, we have, in this study, focused on the propagation of calcium responses and the role of specific connexin isoforms in the conduction. We have used connexin mimetic peptides to inhibit intercellular communication conducted through specific subtypes of connexins.
Several studies have shown a coupling between nephrons deriving their afferent arterioles from the same interlobular artery (9, 22). This coupling appears to be mediated by a conducted vascular response in the preglomerular vessels, whereby a local diameter change may travel several hundred micrometers along the preglomerular arterioles (22, 41, 44). These findings show that the afferent arteriolar cells are extensively coupled and capable of conducting vasoconstrictor signals. Hence, several lines of evidence suggest that conducted vasoconstrictor responses, most likely mediated via gap junctions, have a functional importance in the regulation of renal function.
In the present study, we investigated the effect of specific inhibition of four different connexin isoforms (Cx37, Cx40, Cx43, and Cx45) on the conduction of calcium responses in renal arterioles. The different connexins were inhibited by connexin mimetic peptides specifically targeted against the corresponding isoforms. The functionality of the peptides was examined in RAOEC cells shown to express Cx37, Cx40, Cx43, and Cx45 (Fig. 2). The connexin mimetic peptide against Cx40 significantly reduced transfer of Lucifer yellow between the endothelial cells by 44 ± 10% compared with control cells not treated with the peptide (Fig. 3). Combining all three peptide isoforms to obtain inhibition of the four connexins expressed in RAOEC cells did not reduce dye transfer further (50 ± 9%, NS). This is consistent with the fact that RAEOC cells only express small to moderate amounts of Cx37, Cx43, and Cx45 in addition to Cx40 (Fig. 2) and that the expression is primarily intracellular. The majority of the functional gap junctions is therefore composed by Cx40. The inhibitory effect of connexin mimetic peptides was only intermediate of the effect of the nonselective gap junction uncoupler carbenoxolone. There is evidence to suggest that connexin mimetic peptides block formation of new gap junctions but do not impair the function of already existing channels (11). Connexins in the plasma membrane have a half-life of 1.5–3 h (3, 30). A reduction of ∼50% in dye transfer is, therefore, within the range of values to be expected following 1 h of incubation with the connexin mimetic peptides. Carbenoxolone, on the other hand, reduces intercellular coupling significantly within 6–7 min (33) and targets already existing gap junctions.
Others have also shown an inhibitory effect of the mimetic peptides on dye transfer in endothelial cells (5, 16, 31), confirming that the connexin mimetic peptides are able to reduce intercellular coupling in endothelial cells. Also, the specificity of the peptides has been tested in nonvascular cell systems. In confluent COS fibroblasts expressing Cx43, intercellular dye transfer of Lucifer yellow was impaired by 37/43Gap27, but not by 40Gap27, despite the sequences of these peptides only differing by three amino acids (8).
We next examined the effect of the connexin mimetic peptides on conducted calcium responses induced by electrical stimulation in isolated renal arterioles. Our laboratory has previously reported that local electrical stimulation of the rat interlobular artery in vitro causes a distant increase in vascular smooth muscle [Ca2+]i (38). The mechanism most likely involves an electrotonic spread of the locally induced membrane depolarization along the vessel causing depolarization, opening of voltage-dependent calcium channels, and calcium influx and release at the remote sites (38). Also, in the rabbit microdissected juxtaglomerular apparatus, stimulation of the MD elicited a calcium wave that spread upstream toward proximal segments of the afferent arteriole and the adjacent glomerulus (34). The distal increase in [Ca2+]i appears to be mediated by connexins, since the calcium wave was inhibited by the nonselective gap junction blockers 18α-GA and heptanol (34).
It is, therefore, likely that the conduction of vasoconstriction is mediated via gap junctions, but there is little information on which connexin isoform may mediate the conduction. In Cx37 knockout mice, the conducted vasoconstriction induced by application of KCl in cremaster arterioles was significantly reduced compared with wild-type mice (32). Interestingly, Cx37 is upregulated in aortic endothelial cells from Cx40 knockout mice (27), which maintain normal conducted vasoconstriction, further suggesting a role for Cx37 in the conduction of vasoconstriction in mouse arterioles.
The acute effect of a decrease in intercellular communication exerted by connexin mimetic peptides on conducted calcium responses has, to our knowledge, not been examined before. In the acute setting, as opposed to the condition in genetic knockout mice, there is limited time for upregulation or downregulation of other connexins to take over the function of the connexin chosen for examination (e.g., upregulation of Cx37 in Cx40 knockout mice).
In the present study, treatment with peptides against Cx40, Cx37/43, or Cx45 did not attenuate the conducted calcium response to electrical stimulation in rat renal arterioles (Fig. 4). Chaytor et al. (7) found that a combination of peptides (40Gap27 + 43Gap26, 300 μM of each) was far more effective in inhibiting endothelium-derived hyperpolarizing factor- mediated vasodilatation in rat arterioles than one peptide alone. However, in our experiments, the combination of the three different peptides (40Gap27 + 37/43Gap27 + 45Gap27, 300 μM of each) had no effect on conducted calcium responses in isolated renal arterioles (Fig. 4). On the other hand, treatment with carbenoxolone completely inhibited the conducted response to electrical stimulation (Fig. 4). This was not due to nonspecific effects of carbenoxolone on the local Ca2+ entry and/or release mechanisms, since high K+ elicited a significant increase in [Ca2+]i in the presence of carbenoxolone (cf. Fig. 5 and results).
The present results indicate that in renal arterioles from rats, connexin 37, 40, 43, and 45 do not seem to play a central role in the conduction of vasoconstriction. Studies in arterioles from the hamster cheek pouch show that, at least in this preparation, the conduction of vasoconstriction takes place exclusively within the vascular smooth muscle cells (2). We and others have consistently found either no expression or a very low expression of Cx37, Cx40, and Cx43 in the smooth muscle cells of renal preglomerular vessels (1, 21, 24, 42, 45). This is consistent with the observed lack of effect of connexin mimetic peptides on the conducted Ca2+ response. Whether Cx45 forms functional gap junctions within the vascular wall remains undetermined. In the present study, we found staining for Cx45 in the RAOEC cells, but the staining appeared to be located mainly intracellularly. This does not exclude the possibility that small amounts of Cx45 gap junctions are present between the cells. However, the present results suggest that even if this is the case, these gap junctions may not be essential for the conducted Ca2+ response, since it persisted even in the presence of a combination of connexin mimetic peptides (40Gap27 + 37/43Gap27 + 45Gap27). In contrast, the nonselective gap junction blocker carbenoxolone completely abolished the conducted Ca2+ response without affecting the local Ca2+ response to membrane depolarization. At present, more than 20 isoforms of connexins are known in mammals (40), and it is possible that connexins other than Cx37, Cx40, Cx43, and Cx45 are present in the vascular smooth muscle cells.
These results should, however, be considered with caution because of possible unspecific actions of carbenoxolone. It has been shown that carbenoxolone may affect ion channels and pumps, e.g., K+ channels (35), Cl− channels (4), and the Na+/K+-ATPase (46) and therefore has the potential to change cellular membrane potentials. We cannot exclude an unspecific non-gap junctional effect of carbenoxolone on the vessels, resulting in an uncoupling of the cells, although the persistence of the local Ca2+ response to depolarization makes this a less likely possibility. The unspecific effect of nonselective gap junction uncouplers such as carbenoxolone, heptanol, and halothane further emphasize the need for more specific uncouplers to investigate the physiological role of gap junctions.
Perspectives and Significance
No attenuating effect was seen in conducted Ca2+ responses elicited by local electrical stimulation when interlobular arteries from rats were exposed to connexin mimetic peptides against Cx40, Cx37/43, and Cx45. We were, however, able to verify that 40Gap27 had an inhibitory effect on the dye transfer in cultured endothelial cells expressing connexin 40. Future experiments are needed to clarify whether renal vascular conducted vasoconstriction is mediated via still unknown pathways.
The present study was supported by grants from the Danish Medical Research Council, the Danish National Research Foundation, the Novo-Nordisk Foundation, the Danish Heart Foundation, and the König-Petersen Foundation.
The technical assistance of Anni Salomonsson, Trine Eidsvold, and Ian Godfrey is gratefully acknowledged. The plasmid encoding Cx37 was a kind gift from Professor Steven M. Taffet, Upstate Medical University, Syracuse, NY.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society