- Email: [email protected]
Connexin40 regulates renin production and blood pressure
original article
http://www.kidney-international.org & 2007 International Society of Nephrology
see commentary on page 781
Connexin40 regulates renin production and blood pressure N Krattinger1, A Capponi2, L Mazzolai3, J-F Aubert3, D Caille4, P Nicod1, G Waeber1, P Meda4 and J-A Haefliger1 1
Department of Medicine, University Hospital, Lausanne, Switzerland; 2Department of Internal Medicine, School of Medicine, University of Geneva, Gene`ve, Switzerland; 3Department of Medicine, Service d’Angiologie, University Hospital, Lausanne, Switzerland and 4 Department of Cell Physiology and Metabolism, School of Medicine, University of Geneva, Gene`ve, Switzerland
Renin secretion is regulated by coordinated signaling between the various cells of the juxtaglomerular apparatus. The renin-secreting cells (RSC), which play a major role in the control of blood pressure, are coupled to each other and to endothelial cells by Connexin40 (Cx40)-containing channels. In this study, we show that Cx40 knockout (Cx40/) mice, but not their heterozygous littermates, are hypertensive due to the increase in the number of RSC, renin biosynthesis, and plasma renin. Treatment with the angiotensin II receptor AT1 antagonist candesartan or the angiotensin II-converting enzyme inhibitor ramipril reduced the blood pressure of the Cx40/ mice to the same levels seen in wild-type (WT) mice. The elevated blood pressure of the knockout mice was not affected by clipping one renal artery (2K1C, renindependent model of hypertension) or after a high salt diet. Under these conditions, however, Cx40/ mice showed an altered production and release of renin. The renin mRNA ratio between the clipped and the non-clipped kidney was lower in the knockout than in the WT 2K1C mice. This indicates that the response to a change in blood pressure was altered. The RSC of the Cx40/ mice did not have a compensatory increase in the levels of either Cx43 or Cx37. Our data show that renin secretion is dependent on Cx40 and suggest the Cx40/ mice may be a genetic model of renin-dependent hypertension. Kidney International (2007) 72, 814–822; doi:10.1038/sj.ki.5002423; published online 11 July 2007 KEYWORDS: gap junctions; connexins; transgenic mice; hypertension; secretion
Correspondence: J-A Haefliger, Department of Internal Medicine, Laboratory of Molecular Medicine 19-135S, University Hospital, Lausanne CHUV-1011, Switzerland. E-mail: [email protected] Received 7 March 2007; revised 3 May 2007; accepted 22 May 2007; published online 11 July 2007 814
The control of blood pressure, which is of paramount importance for the organism, is mostly achieved within the kidney cortex via the release of renin by the juxtaglomerular apparatus (JGA).1 Renin is normally secreted in a regulated manner, as a function of the integration of various mechanical, osmotic, and electrochemical stimuli, acting on the different cell types which collectively form the JGA.2 This regulation implies that JGA cells function in coordination. One of the mechanisms whereby this coordination may be achieved is via gap junctions,3–6 the membrane regions where connexin (Cx) channels for the intercellular sharing of ions and signaling molecules are localized.7 A functional gap junction is composed of two opposing hexamers, each contributed by one of two adjacent cells, and consisting of an assembly of six connexin proteins.7 Accordingly, gap junctions made of Cx37, Cx40, and Cx438–10 link the cells of the JGA.11,12 Furthermore, gap junctions made of Cx40 also link the endothelial cells of the afferent arteriole to the reninsecreting cells (RSC), which form the media of the juxtaglomerular part of this arteriole.8,13 Gap junctional communication via Cx40 channels appears to play a central role in the function of the vascular system,13,14 as exemplified by the finding that loss of Cx40 results in chronic hypertension.15–17 While the mechanism of this effect is still unknown, the observation that changes in Cx40 are associated with altered vasomotion and signal conduction along arterioles has led to propose that Cx40 contributes to control blood pressure by mediating the adaptive changes of the vascular walls, that are induced by increased pressure and mechanical stress.15–17 However, since connexins control secretion of various hormones,3,18 and since the arterial expression of Cx40 is increased in rodents featuring a renin-dependent hypertension,8,13 we postulated that loss of Cx40 could also alter the coordinated functioning of the RSC, hence accounting for increased blood pressure. Here, we report on the direct testing of this possibility in mice lacking Cx40 (Cx40/),19,20 and demonstrate that the hypertension of these mice is mostly due to an altered production of renin. Thus, these mice represent a valuable genetic model of renin-dependent increase in blood pressure, the most common situation in human hypertension. Kidney International (2007) 72, 814–822
original article
N Krattinger et al.: Renin secretion and Cx40
d
e
g
f b
***
Volume of cytoplasma (a.u.)
3 2 1 0
h
i Irregular renin-containing granules (%)
Number of renin cells per glomerlus (a.u.)
c
Number of granules per cell profile
a
20 10
***
0
30 25 20 15 10 5 0
***
70 60 50 40 30 20 10 0
***
Figure 1 | Loss of Cx40 alters the organization of the RSC. (a) In control mice, in situ hybridization shows that Cx40 mRNA is expressed along the endothelial cell (e) that line the lumen of the afferent arteriole (aa), as well as between the RSC (rc) that form the media of the juxtaglomerular (gl) portion of this vessel. (b) Immunocytochemistry shows a similar localization of the cognate protein. Bar ¼ 20 mm. (c) Immunofluorescence reveals the alignment of RSC in three afferent arterioles of a Cx40/ mouse. (d and e) Mice lacking Cx40 had more renin-containing cells per glomerule than WT controls. (f–i) When compared to controls (f), the RSC of Cx40/ mice (g) were smaller (e) and contained fewer (h), and more round secretory granules (i). Data are mean7s.e.m. values. ***Po0.001 Cx40/ (solid bars) vs WT values (open bars). Bar ¼ 30 mm in (c) and 2 mm in (f) and (g).
In agreement with previous reports, in situ hybridization and immunofluorescence showed the expression of Cx40 in the endothelial cells of the afferent arterioles of wild-type mice (WT), as well as in RSC that formed the media of the juxtaglomerular portion of this vessel (Figure 1a and b). No Cx40 was found in the corresponding regions of homozygous Cx40/ mice (data not shown). When compared to the RSC of WT, which expressed Cx40, the RSC of Cx40/ littermates (Figure 1c), which lacked Cx40, were more numerous per glomerule (Figure 1d), featured a smaller cytoplasmic volume (Figure 1e) and decreased (Po0.003) numbers of secretory granules (Figure 1h). Furthermore, whereas in control cells, most of the renincontaining granules had an irregular outline (Figure 1f), in cells lacking Cx40, most of these granules were round (Figure 1g and i). Thus, loss of Cx40 alters the cellular composition of the afferent arteriole as well as the intracellular storage of renin. Cx40/ mice are hypertensive due to activation of the renin–angiotensin system
WT (n ¼ 9) and Cx40 þ / mice (n ¼ 8), which expressed Cx40, were normotensive (Figure 2a). In contrast, Cx40/ littermates (n ¼ 10) featured a 40% increase (Po0.001) in blood pressure (Figure 2a). This hypertension was associated with a large increase (Po0.001) in the levels of circulating renin (Figure 2b). Since no significant difference was observed between the first two sets of mice, only the data for the WT controls are shown hereafter. Kidney International (2007) 72, 814–822
70 WT
160
Cx40+/–
140
Cx40–/–
120 100 80 60
***
60 PRA (ng Ang l/ml h)
8–13
180
Mean blood pressure (mm Hg)
RESULTS The RSC express Cx40, and are altered after loss of this connexin
***
50 40 30 20
40 20 0
10 0
Figure 2 | Cx40/ mice are hypertensive due to increased levels of circulating renin. (a) Mean blood pressure was higher in Cx40/ mice (solid bar) than in heterozygous ( þ /; shaded bar) and WT littermates (open bar). (b) Plasma renin activity (PRA) was markedly increased in the hypertensive Cx40/ mice. Values are mean7s.e.m. ***Po0.001 Cx40/ vs WT levels.
Treatment with candesartan, an antagonist of the angiotensin II receptor AT1, or ramipril, an inhibitor of the angiotensin II converting enzyme (Table 1), reduced the blood pressure of Cx40/ mice to the levels observed in control littermates (Figure 3a), in spite of a persistent elevation of plasma renin (Figure 3b). In contrast, plasma renin and the expression of renin mRNA in kidneys were increased in control mice after both treatments (Figure 3b and c). Thus, the hypertension of Cx40/ mice is due to an activation of the renin–angiotensin system (RAS), which results in elevated levels of circulating renin, and that cannot be properly regulated. 815
original article
N Krattinger et al.: Renin secretion and Cx40
Table 1 | Characteristics of the mice treated with inhibitors of the renin–angiotensin system Group
Treatment
n
Body weight (g)
Heart rate (beats/min)
LKi (mg/g)
RKi (mg/g)
WT
Untreated Candesartan Ramipril
9 10 12
28.471.7 30.171.0 26.570.9
657714 628731 634757
6.670.4 6.970.4 6.870.3
6.870.4 6.870.4 6.870.3
Cwi (mg/g) 6.470.3 6.170.2 5.370.3*
Cx40/
Untreated Candesartan Ramipril
10 12 15
25.971.0 31.672.5 27.571.2
648722 610737 68278
6.570.2 6.770.3 7.270.2
6.570.2 6.570.3 7.370.2
8.670.7111 7.670.6*11 7.170.3*11
Cwi, cardiac weight index; LKi, left kidney index (ratio between the left kidney weight and body weight); n, number of mice; RKi, right kidney index. Values are expressed as mean7s.e.m. *Po0.05 vs untreated mice; 11Po0.01 vs WT mice; 111Po0.001 vs WT mice.
70
***
Cx40 –/– 50
***
160
PRA (ng Ang l/ml h)
Mean blood pressure (mm Hg)
180
140 120 100 80
***
***
40
*** 30
20
60 40
10
20 0
0
ol rtan ipril ol rtan ipril ntr ntr Co desa Ram Co desa Ram n n Ca Ca
n n ol ril ril ol ntr rta ip ontr sarta amip C de Co desa Ram R n n Ca Ca
1.2
§
** **
**
**
2.0
** 1.5 1.0
1.0
AT1A/GAPDH mRNA
2.5
Renin/GAPDH mRNA
***
60
WT
0.8 0.6
*** *** ***
0.4
0.5
0.2
0
0
n ipril ol ol rtan ipril ntr rta ntr Co desa Ram Co desa Ram n n Ca Ca
*** ***
Ctrl
LK RK 2K1C
Ctrl
LK RK 2K1C
Figure 3 | High blood pressure and renin levels of Cx40/ mice are caused by an activation of the RAS. (a) Inhibitors of the RAS lowered blood pressure of Cx40/ mice, but did not affect that of WT controls. (b) Circulating renin was not significantly affected by these treatments in Cx40/ mice, but was increased in WT controls (WT). (c) Basal expression of renin mRNA was higher in Cx40/ than in WT mice. Candesartan and ramipril significantly increased the expression of renin in WT animals. (d) Expression of the angiotensin 1 receptor subtype (AT1A) mRNA was lower in Cx40/ than in WT mice. After induction of hypertension in the 2K1C model, the AT1A levels of WT mice decreased to the values observed in Cx40/ mice. Values are mean7s.e.m. **Po0.01, ***Po0.001 Cx40/ (solid bars) vs WT levels (open bars). y Po0.05 vs levels in untreated Cx40/ mice.
In control mice, increased release of renin leads to enhanced generation of angiotensin II which, in turn, decreases the expression of the AT1A receptor subtype, that 816
ultimately controls the RAS.21,22 The expression of this receptor was significantly lower in Cx40/ mice (n ¼ 10) than in WT littermates (n ¼ 9) (Figure 3d), in agreement Kidney International (2007) 72, 814–822
original article
N Krattinger et al.: Renin secretion and Cx40
To identify the underlying mechanism, we first investigated the effect of diets known to increase (0.03% NaCl) or decrease (4% NaCl) plasma renin in control mice, under conditions of constant blood pressure (Figure 4a and b). Three weeks after exposure to these diet, the blood pressure of 9 (low-salt diet) to 11 (high-salt diet) Cx40/ mice was also unchanged and remained significantly higher than that of WT (Figure 4a). The circulating levels of renin of Cx40/ mice also remained markedly elevated compared to physiological levels, and did not show the changes observed in WT controls. Thus, whereas these levels were slightly reduced (about half the control change) after the low-salt diet, they did not increase, as anticipated by the physiology, after the high salt intake (Figure 4b). These differences were seen in spite of a similar regulation of the kidney expression of the renin transcript in the Cx40/ and the WT mice, during both the low and high-salt diets (Figure 4c). Thus, loss of Cx40 did affect selective aspects of the regulation of renin secretion dependent on the macula densa. To evaluate how this dysregulation affected the RSC, WT and Cx40/ mice were submitted to a surgical 2K1C procedure.10,23 The left, hypoperfused kidney of all 2K1C animals weighed less than that of the shamoperated mice, whereas the contralateral kidney was significantly larger than that of these controls (Table 2). Also, the heart of hypertensive Cx40/ mice was hypertrophied compared to that of WT controls, as indicated by a 45% increase in cardiac weight index (Table 2). Four weeks after clipping one renal artery, we observed an increase in the blood pressure of WT mice (Figure 5a), due to elevated (Po0.01) levels of circulating renin (Figure 5b), as a result of a larger production of the enzyme in the hypoperfused left kidney (Figure 5c and d). When submitted to the same protocol, Cx40/ mice retained unchanged their high blood pressure (Figure 5a) and circulating levels of renin (Figure 5b), in spite of increased renin expression within the clipped kidney (Figure 5c and d). However, this change was only half that observed in WT mice, indicating that the regulation of renin secretion under conditions of altered kidney perfusion was altered in Cx40/ mice. Consistent with this finding, in situ hybridization demonstrated that the levels of renin mRNA were also significantly less reduced in the unclipped kidney of Cx40/ mice than in that of WT littermates (Figure 5d). Kidney International (2007) 72, 814–822
b
70
*** WT Cx40–/–
***
60
PRA (ng Ang l/ml h)
180
*** *** ***
160 140 120 100 80
50 40
§ ***
30 20
60 40
*
10
20 0
* 0 NS LS HS NS LS HS
c
NS LS HS NS LS HS
WT NS
LS
LK RK LK RK
Cx40–/– HS
NS
LK RK
LK RK
LS
HS
LK RK LK RK
Renin GAPDH
§ **
4 Renin/GAPDH mRNA
The regulation of renin production and secretion is altered in Cx40/ mice
a
Mean blood pressure (mm Hg)
with the much higher levels of circulating renin of the former mice. The WT levels of AT1A decreased up to the values observed in Cx40/ animals when WT mice were submitted to an experimental procedure (2K1C) that causes hypertension by elevating the circulating levels of renin. Under the same conditions, the levels of AT1A did not change in Cx40/ mice (Figure 3d).
3
**
2
* 1 0
*
NS LS HS
§ *
NS LS HS
Figure 4 | The regulation of renin secretion by salt is altered in Cx40/ mice. (a) Mean blood pressure of WT (WT) and Cx40/ mice was not altered after feeding a normal (NS), a low (LS), or a highsalt diet (HS). (b) Under the latter diet, the circulating levels of renin decreased in both WT and Cx40/ mice. In contrast, a low salt diet increased renin secretion in WT but not Cx40/ mice. (c) Northern blots showed that the low and the high-salt diets had opposite effects on the expression of the renin gene. Data are mean7s.e.m. values. *Po0.05, **Po0.01, ***Po0.001 vs the NS value of WT mice (open bars); yPo0.05 vs. NS values of Cx40/ mice (solid bars).
Loss of Cx40 is not compensated by other connexins in the RSC
In the 2K1C model, the levels of Cx40 mRNA were found increased in the hypoperfused kidney of WT mice and reduced in the contralateral kidney, in agreement with the differential regulation of renin expression in the two organs (Figure 6a). Hypertension, was also associated with increased expression of Cx43 mRNA in both WT and Cx40/ mice kidneys (Figure 6a), in keeping with the previously reported changes of this connexin in the vascular wall of hypertensive animals.8,10,13 The renal mRNA levels of Cx37, another connexin of the vasculature which is found in the JGA,9 were not modified by the 2K1C procedure in WT controls (Figure 6a). In contrast, these levels were elevated in the 817
original article
N Krattinger et al.: Renin secretion and Cx40
Table 2 | Characteristics of the 2K1C and sham-operated mice Group
Treatment
n
Body weight (g)
Heart rate (beats/min)
LKi (mg/g)
RKi (mg/g)
Cwi (mg/g)
WT
Sham 2K1C Sham 2K1C
9 9 11 11
30.570.6 28.870.8 26.371.6 28.270.6
654718 641724 655718 669713
7.370.2 6.370.2* 7.370.4 5.970.2**
7.370.2 8.570.3** 7.270.3 8.470.5**
6.570.3 7.870.3* 9.671.211 9.070.611
Cx40/
Cwi, cardiac weight index; LKi, left kidney index; n, number of mice; RKi, right kidney index. Values are expressed as mean7s.e.m. *Po0.05, **Po0.01 vs 2K1C sham-operated animals; 11Po0.01 vs WT animals.
180 160
b
WT Cx40–/–
***
**
140
***
d
70 60
PRA (ng Ang l/ml h)
Mean blood pressure (mm Hg)
a
120 100 80 60
***
50
aa
aa
40
G
G
30 20
**
40
WT LK
WT RK
10
20 0
***
Sham 2K1C 2K
0
Sham 2K1C 2K
c
WT
aa
Cx40–/–
Sham
2K1C
LK RK
LK RK
Sham
G
aa
G
Sham 2K1C Sham 2K1C 2K 2K
2K1C
LK RK
LK RK
Renin
WT LK 2K1C
WT RK 2K1C
GAPDH
aa G G
3.0
§ **
Renin/GAPDH mRNA
2.5 2.0
aa Cx40–/– RK
Cx40–/– LK
**
* ## §
1.5
G
1.0
G
0.5
***
0 Sham LK
RK 2K1C
aa Sham LK RK 2K1C
Cx40–/– LK 2K1C
aa Cx40–/– RK 2K1C
Figure 5 | Renin secretion and hypertension depend on the expression of Cx40. (a) The 2K1C procedure increased the blood pressure of WT but not of Cx40/ mice. (b) Plasma renin followed the same pattern model. (c) Changes of renin expression took place in both the hypoperfused (LK) and the controlateral (RK) kidneys of both WT and Cx40/ mice. However, the effects on renin mRNA levels were smaller in Cx40/ than in WT animals. (d) In situ hybridization showed that the levels of this transcript in the afferent arterioles (aa) also decreased much less in the kidneys of 2K1C Cx40/ mice than in those of WT controls (right panels). G, glomerulus. Bar ¼ 80 mm. Data are mean7s.e.m. values of the number of mice indicated in Table 2. *Po0.05, **Po0.01, ***Po0.001 vs the cognate sham value in WT mice (open bars); yPo0.05 vs sham value of Cx40/ mice (solid bars); ##Po0.001 vs the value in the right kidney (RK) of WT mice.
hypertensive, sham-operated Cx40/ mice, and in the hypoperfused left kidney of the Cx40/ mice which were submitted to the 2K1C procedure (Figure 6a). To assess whether these renal changes in the mRNA levels of Cx43 and Cx37 provided a compensation for the loss of Cx40 in RSC, we immunostained kidney sections with 818
specific antibodies to these two connexins. In agreement with previous reports,9,10 we identified Cx43 and Cx37 in the interlobular vessels of WT controls, as well as in the wall of their afferent arterioles, and glomerular vessels (Figure 6b and c). In contrast, the RSC of WT mice were immunostained for Cx37 (Figure 6b), but not Cx43 (Figure 6c). In Kidney International (2007) 72, 814–822
original article
N Krattinger et al.: Renin secretion and Cx40
3.5
WT Cx40–/–
Cxs/GAPDH mRNA
3.0 2.5
˚˚
2.0
* *
* 1.5
* * *
* *
1.0
*
0.5 0
SK LK RK SK LK RK SK LK RK SK LK RK SK LK RK SK LK RK Sham 2K1C Sham 2K1C Sham 2K1C Sham 2K1C Sham 2K1C Sham 2K1C
Cx40
Cx43
Cx37
Figure 6 | Cx37 and Cx43 are not over expressed in the RSC of the hypertensive Cx40/ mice. (a) Analysis of total kidney RNA revealed different changes of the transcripts of Cx40, Cx43 and Cx37 in both the hypoperfused (LK) and the controlateral (RK) kidneys of WT (open bars) and Cx40/ mice (solid bars) submitted to the 2K1C procedure. Data are mean7s.e.m. values of the number of mice indicated in Table 2. *Po0.05, vs sham value in WT mice. 11Po0.01. (b) Antibodies detected Cx37 in the intra-glomerular vessels of WT mice (top panels), as well as in the media of the juxta-glomerular portion of the afferent arterioles, where RSC are located (arrows in middle top panel), and between the endothelial cells of interlobular arteries (arrows in left bottom panel). A similar distribution was observed in Cx40/ mice, even though the levels of Cx37 appeared decreased in both glomeruli (middle bottom panel) and interlobular vessels (right bottom panel). (c) Cx43 was expressed by the endothelial cells of renal vessels, including the afferent arteriole (left top panel). However, and at variance with Cx37, Cx43 was not found in the RSC of this vessel (arrow in left top panel). A similar distribution and expression levels were observed in the afferent arterioles (aa) of Cx40/ mice (middle and right top panels; arrows point to putative RSC). In these mice, Cx43 was also observed between the endothelial cells of interlobular arteries (ila, arrows in left bottom panel), at the low levels observed in controls. Cx43 appeared increased between the smooth muscle cells of the media of large arcuate arteries (arrows in right bottom panel). gl, glomerulus; aa, afferent arteriole; ila, interlobular artery; md, macula densa; ara, arcuate artery. Bar ¼ 25 mm.
Cx40/ mice, a similar distribution was observed for both Cx37 and Cx43. However, Cx37 appeared reduced in vessels and RSC (Figure 6b), whereas Cx43 appeared increased (Figure 6c) in the intimal endothelium and in smooth muscle cells of the intima and media of large interlobular vessels, respectively. As in controls, RSC of Cx40/ mice expressed some Cx37 (Figure 6b) but no Cx43 (Figure 6c). Thus, loss of Cx40 in RSC is not compensated by an obvious change in the expression of other connexin proteins, which are expressed in renal vessels and JGA. DISCUSSION
The gap junction protein Cx40 has been implicated in the function of the vascular system, notably because knockout mice lacking this connexin are hypertensive. Previous studies have concluded that this hypertension Kidney International (2007) 72, 814–822
was mostly due to an abnormal adaptation of the arterial walls to altered hemodynamic conditions,15–17 implying that Cx40 contributes to the detection and propagation of the blood-borne signals that are elicited by increased blood pressure. Here, we show that the hypertension of Cx40 mice mostly results from alterations in the number and function of the RSC, and the ensuing activation of the RAS. We further document that, in the absence of Cx40, the regulation of this system is altered both under a low-salt regimen and under conditions of changed renal blood flow, two situations that are sensed by different cell types of the JGA.1,2 The data show that connexins play a central role in the signaling from these cells, that converge on the myo-epithelial cells of the juxtaglomerular portion of the afferent arterioles, which secrete renin. 819
original article
These data have two implications. First, they extend to the RSC, the increasing evidence pointing to connexin signaling as a major, obligatory player in the control of secretory systems.3,18 Our data document for the first time that loss of Cx40 increases the number of RSC, the renal production of the hormone and its release, indicating that the signaling dependent on this protein natively downregulates the function of RSC. Second, our data show that Cx40 plays a central role in the regulatory function of the JGA. Thus, loss of this connexin markedly activates the RAS, altering its normal regulation of blood pressure, and resulting in hypertension. In view of its distribution in the membrane of both endothelial cells and RSC of the afferent arteriole, and its mechano-sensitive characteristics,24 Cx40 is ideally located to detect bloodborne signals and to propagate them across the vascular and the RSC of the JGA, thus permitting the integration of the mechanical, hormonal, and ionic inputs that control blood pressure.2,25 By comparing WT and Cx40/ mice in a renin-dependent (2K1C) model of hypertension our experiments show for the first time that the hypertension of Cx40/ animals is essentially due to an excessive secretion of renin, due to increased numbers of RSC, which feature alterations of their enzyme-containing granules. Most probably, the signaling mechanism of Cx40 is integrated with that of both Cx43 and Cx37, two other connexins that are also expressed within the JGA,8,10 and that are affected by hypertension.8,10,13,26 With regard to Cx43, we have recently provided evidence that replacement of this protein by the Cx32 isoform, which cannot form channels with Cx40,27 downregulates the secretion of RSC, protecting transgenic mice against the hypertension which is expected in a model of renovascular hypertension.10 Cx43 is not detected in the RSC of WT animals, but is expressed by rare endothelial cells of arteries, as well as by most smooth muscle cells of these vessels,13 including the afferent arteriole. Consistent with previous reports of elevated vascular levels of Cx43 in different models of experimental hypertension,8,10,13,26 we show here that the renal levels of Cx43 were higher in the hypertensive Cx40/ mice than in the cognate, normotensive controls. We further show that, when the latter animals were made hypertensive by the 2K1C procedure, the levels of Cx43 expression rose in the kidneys of controls, up to the levels observed in the Cx40/ mice that featured a similar degree of hypertension, due to excessive renin secretion. We have not observed an induction of Cx43 in the RSC of the Cx40/ animals. Therefore, the contribution of this connexin to the dysregulation of renin secretion may imply the mediation of an extracellular signal initiated in the endothelial and/or smooth muscle cells of the afferent arteriole, since Cx40 and Cx43 appear unable to form functional heterotypic gap junction channels.28 The close apposition of RSC, endothelial, and smooth muscle cells of the JGA provides an ideal setting for paracrine effects within the afferent arteriole.8 An analogous, connexion-dependent 820
N Krattinger et al.: Renin secretion and Cx40
paracrine mechanism has been recently postulated in the stria vascularis epithelium of the inner ear.29 With regard to Cx37, its coexpression with Cx40 in the RSC9 raises the question of which of the two connexin isoforms is actually responsible for the increased renin secretion of Cx40/ mice. Several observations indicate that Cx40 largely predominates in this role. Thus, Cx37/ mice, which still express Cx40, are normotensive,30 and no obvious change was observed in the expression of Cx37 in hypertensive 2K1C WT mice, which feature a striking upregulation of renin expression. Still, and in agreement with a previous suggestion in the vascular compartment,30 we observed that the levels of the Cx37 protein were decreased in the kidneys of the Cx40/ mice, indicating that the expression of Cx40 and Cx37 are somehow interdependent. Our findings are consistent with a model in which proper renin secretion depends both on the homotypic signaling ensured by Cx40 between the RSC themselves, and on a heterotypic signaling, mediated by Cx40, and possibly Cx37, between RSC and the endothelial and/or smooth muscle cells of the afferent arteriole. Full validation of these tentative conclusions requires that the paracrine signal(s) and mechanism dependent on Cx43 be identified, and that the incidence and extent of the homotypic RSC coupling dependent on Cx40 be compared, within the intact kidney, to those of the heterotypic communication between RSC and vascular cells mediated by Cx37–Cx40 channels. Both are formidable tasks in view of the small number and dispersion of RSC within the kidney cortex, and the tight intermingling of the different cell types that collectively form the JGA. Therefore, at least two complementary, if not alternative, possibilities should be considered at this stage. The first possibility finds its basis on recent studies that have documented that specific connexins selectively affect the expression of a variety of genes, by a mechanism that remains to be elucidated and that may be independent of the establishment of cell-to-cell communication.29,31,32 Thus, large-scale transcriptome analysis of different cell and animal models have shown that Cx43 and Cx32 regulate an array of genes involved in the control of cell growth,31,32 whereas Cx30 appears to preferentially control genes implicated in the metabolism of homocysteine.29 Strikingly, the tissue accumulation of this amino acid, which is seen in the absence of Cx30, triggers a local, presumably paracrine mechanism which eventually results in the altered function of endothelial cells that do not express this connexin isoform.29 The second more intriguing possibility is that, as in any other study involving knockout mice,33 the physiological relevance of the effects we observed may be overestimated, because of the compounding effect of experimentally induced alterations, that, in our case, could be largely due to over time compensation of the missing Cx40 by other connexin isoforms or some connexin-dependent gene. However, the finding that both the pharmacological (RAS inhibitors, salt load, or depletion) and surgical interventions (2K1C) elicited similar, and anticipated, blood pressure and renin changes in WT and Cx40 þ / mice, Kidney International (2007) 72, 814–822
N Krattinger et al.: Renin secretion and Cx40
whereas selective alterations of these physiological changes were solely observed in Cx40/ mice, plaids against this possibility. Furthermore, while we have documented changes in the transcripts of other connexin isoforms in the kidney of the hypertensive animals, we have also shown that the cognate proteins were either not ectopically expressed (Cx43) or at best somewhat decreased (Cx37) in RSC. The latter change is unlikely to have cause the pressure perturbations we observed, inasmuch as Cx37/ mice, which still express Cx40, are normotensive.30 Still, what is the actual physiological role of Cx40 in the kidney of living WT mice remains to be validated in other models, including cell- and timespecific knockout and siRNA interference studies. By acutely decreasing the expression of Cx40 in selected cell types, these approaches should avoid the conundrum of compensation and mechanistic adaptations that probably occur over time in any knockout animal. Awaiting the completion of such studies, our findings open new paradigms to envision the still obscure pathogenesis of hypertension. Since polymorphisms of the Cx40 gene have been recently associated with human hypertension,34,35 our data call for an evaluation of renin function in the affected patients. They also point to the Cx40/ mice as a pertinent genetic model to unravel further the molecular mechanisms of renin-dependent hypertension, a most common situation in the human clinic. MATERIALS AND METHODS Knockout mice The generation of Cx40/ mice and their genotypic characterization have been described.19 Breeding pairs (kindly provided by Dr K Willecke) were crossed with control C57BL6 partners. Identification of the WT, heterozygous ( þ /), and homozygous Cx40/ mice was performed by PCR of genomic DNA using the following primers: for the Cx40 WT allele (314 bp), 50 -GGGAGATG AGCAGGCCGACTTCCGGTGCG-30 (sense) and 50 -GTAGGGTGC CCTGGAGGACAATCTTCCC-30 (antisense); for the Cx40 recombinant allele (494 bp), 50 -GGATCGGCCATTGAACAAGATGGAT TGCAC-30 (sense) and 50 -CTGATGCTCTTCGTCCAGATCATCCTG ATCG-30 (antisense). Three 4-month-old animals, from litters obtained after at least 10 backcrosses into the C57/BL6 background, were used in all the experiments. Morphometric and ultrastructural analysis To analyze the number and distribution of RSC, kidneys were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, embedded in paraffin, and serially sectioned at 5 mm thickness. Sections were immunostained using a rabbit polyclonal antibody against renin (2D12, courtesy of Professor P Corvol, Paris), diluted 1:200, and evaluated morphometrically as per standard methods.36 Animal experiments Mouse care, surgery, and euthanasia procedures were approved by our Institutional Committee for Animal Experiments and the Veterinary Office (Lausanne, Switzerland). Modulating RAS. Blockade of the RAS was performed by adding to the drinking water either 3 mg/kg day the AT1-receptor antagonist candesartan (Cilexetils, Astra Zeneca, Zoug, SwitzerKidney International (2007) 72, 814–822
original article
land) or 10 mg/kg day the angiotensin-converting enzyme inhibitor ramipril (Aventis, Geneva, Switzerland) for 3 weeks. To test the effects of a salt load, another series of mice was fed for 3 weeks with a diet (Kliba, Gossau, Switzerland) containing either 0.03% (low-salt group), 4% (high-salt group), or 0.3% NaCl (control group). All mice had free access to tap water. Inducing a renin-dependent hypertension. For the twokidney, one clip model (2K1C), the left renal artery was clipped with a 0.12 mm internal diameter U-shaped silver clip, as described previously.10,23 Measurement of blood pressure. One day before killing, the right carotid artery was penetrated with a catheter connected to a pressure transducer and a computerized data acquisition system, that allowed for recording heart rate and both diastolic and systolic intra-arterial pressure in conscious mice every 20 s.10 Data are shown as the mean (MBP) of the diastolic and systolic values. After a 30-min recording period, 300 ml blood was collected in ethylenediaminetetraacetic acid-coated tubes and centrifuged at 41C. Plasma was stored at 701C until use. Measurements of plasma renin Plasma renin activity was measured by a microassay based on angiotensin I (Ang I) trapping by an antibody.10 Controls showed that endogenous angiotensinogen was not a limiting factor for the measurement of the high levels of renin found in the hypertensive mice (data not shown). RNA analysis RNA isolation and Northern blot analysis were performed as described10 using cDNA probes specific for either renin10 or a 1.1 kb (HindIII–EcoRI) fragment of GAPDH cDNA.8 Quantitative realtime PCR was performed as reported10,37 using the following primers: for renin (158 bp), -50 -TCTCTGGGCACTCTTGTTGC TCTG-30 - (sense) and -50 -ATACGTCCCATTCAGCACTGAGCC-30 (antisense); for GAPDH (334 bp), -50 -gactccacgacatactcagc-30 (sense) and -50 -gtcggtgtgaacggatttgg-30 - (antisense); for AT1a (110 bp), 50 -GTCGCACTCAAGCCTGTCTA-30 (sense); for Cx40 (310 bp), -50 -TTTGGCAAGTCACGGCAGGG-30 (sense) and 50 -TG TCACTATGGTAGCCCTGAG-30 (antisense);38 for Cx37 (144 bp), 50 -GGGAGATAAAGGCACGAAGG-30 (sense) and 50 -ATGAGAGC CCGTTGTAGGTG-30 (antisense); for Cx43 (255 bp), 50 -GATTGAA GAACACGGCAAGG-30 (sense) and 50 -AGAGCGAGAGACACCA AGGA-30 (antisense). Analysis of data was performed using the 3.5 version of the Light Cycler software (Roche Diagnostics, Zong, Switzerland, GmbH, Germany). For the in situ hybridization of the renin and Cx40 transcripts, kidneys sections were hybridized as previously described.8,10 Analysis of connexins Immunodetection of connexins was performed as previously reported.10 Cryostat sections of kidneys (5 mm thickness) were permeabilized 10 min in cold (201C) acetone, and incubated 2 h at room temperature, using the following rabbit polyclonal antibodies: for Cx40, the antibody described by Haefliger et al.,8 diluted 1:250; for Cx37, the antibodies described by Haefliger et al.39 and Delorme et al.,40 diluted 1:100; for Cx43, the antibodies 71-0700 (Zymed, Basel, Switzerland), diluted 1:100, C6219 (Sigma) diluted 1:200, or that described,41 diluted 1:50. At the end of the immunolabeling, the sections were counterstained 3 min in 0.03% Evans blue (in phosphate-buffered saline), which gives the tissues a red background 821
original article
staining that enhances the visualization of the fluorescein isothiocynate immunosignals.36
N Krattinger et al.: Renin secretion and Cx40
16.
17.
Statistical analysis Data were expressed as mean7s.e.m. Differences between means were assessed using Student’s t-test. Statistical significance was defined at a value of Po0.05 (* or, 1 or #, or y), Po0.01 (**, or 11, or ## , or yy), Po0.001 (***, or 111, or ###, or yyy). ACKNOWLEDGMENTS
This work was supported by grants from the Swiss National Science Foundation (Grant nos. 310000-109530, 3100A0-100797, and 310000-109402), the Juvenile Diabetes Foundation International (Grant nos. 1-2005-46 and 1-2007-158), the Placide Nicod Foundation, the Octav and the Marcella Botnar Foundation, the Novartis Foundation, The Emma Muschamp Foundation, Novo Nordisk, and the Geneva Program for Metabolic Diseases.
18.
19.
20.
21.
22.
23.
NOTE ADDED IN PROOF After the submission of our study, the following papers independently confirmed that the hypertension of Cx40/ mice is largely due to altered function (Circ Res 2007; 100: 556) and organization of renin-producing cells (J Am Soc Nephrol 2007; 18: 1103).
24. 25. 26. 27.
REFERENCES 1. Mullins LJ, Bailey MA, Mullins JJ. Hypertension, kidney, and transgenics: a fresh perspective. Physiol Rev 2006; 86: 709–746. 2. Schweda F, Kurtz A. Cellular mechanism of renin release. Acta Physiol Scand 2004; 181: 383–390. 3. Michon L, Nlend Nlend R, Bavamian S et al. Involvement of gap junctional communication in secretion. Biochim Biophys Acta 2005; 1719: 82–101. 4. Le Gurun S, Martin D, Formenton A et al. Connexin-36 contributes to control function of insulin-producing cells. J Biol Chem 2003; 278: 37690–37697. 5. Calabrese A, Caton D, Meda P. Differentiating the effects of Cx36 and E-cadherin for proper insulin secretion of MIN6 cells. Exp Cell Res 2004; 294: 379–391. 6. Ravier MA, Guldenagel M, Charollais A et al. Loss of connexin36 channels alters beta-cell coupling, islet synchronization of glucose-induced Ca2+ and insulin oscillations, and basal insulin release. Diabetes 2005; 54: 1798–1807. 7. Sohl G, Willecke K. Gap junctions and the connexin protein family. Cardiovasc Res 2004; 62: 228–232. 8. Haefliger JA, Demotz S, Braissant O et al. Connexins 40 and 43 are differentially regulated within the kidneys of rats with renovascular hypertension. Kidney Int 2001; 60: 190–201. 9. Zhang J, Hill CE. Differential connexin expression in preglomerular and postglomerular vasculature: accentuation during diabetes. Kidney Int 2005; 68: 1171–1185. 10. Haefliger JA, Krattinger N, Martin D et al. Connexin43-dependent mechanism modulates renin secretion and hypertension. J Clin Invest 2006; 116: 405–413. 11. Guo R, Liu L, Barajas L. RT-PCR study of the distribution of connexin 43 mRNA in the glomerulus and renal tubular segments. Am J Physiol 1998; 275: R439–R447. 12. Mink D, Schiller A, Kriz W et al. Interendothelial junctions in kidney vessels. Cell Tissue Res 1984; 236: 567–576. 13. Haefliger JA, Nicod P, Meda P. Contribution of connexins to the function of the vascular wall. Cardiovasc Res 2004; 62: 345–356. 14. Figueroa XF, Isakson BE, Duling BR. Connexins: gaps in our knowledge of vascular function. Physiology (Bethesda) 2004; 19: 277–284. 15. de Wit C, Roos F, Bolz SS et al. Impaired conduction of vasodilation along arterioles in connexin40-deficient mice. Circ Res 2000; 86: 649–655.
822
28.
29.
30.
31. 32. 33.
34.
35. 36.
37. 38.
39.
40.
41.
de Wit C, Roos F, Bolz SS et al. Lack of vascular connexin 40 is associated with hypertension and irregular arteriolar vasomotion. Physiol Gen 2003; 13: 169–177. Figueroa XF, Paul DL, Simon AM et al. Central role of connexin40 in the propagation of electrically activated vasodilation in mouse cremasteric arterioles in vivo. Circ Res 2003; 92: 793–800. Klee P, Boucard N, Caille D et al. Connexins modulators of endocrine function. Winterhager E(ed). Springer Verlag, Berlin, Heldelberg. Gap Junctions in Development and Disease 2005: 197–221. Kirchhoff S, Nelles E, Hagendorff A et al. Reduced cardiac conduction velocity and predisposition to arrhythmias in connexin40-deficient mice. Curr Biol 1998; 8: 299–302. Simon AM, Goodenough DA, Paul DL. Mice lacking connexin40 have cardiac conduction abnormalities characteristic of atrioventricular block and bundle branch block. Curr Biol 1998; 8: 295–298. Haefliger JA, Bergonzelli G, Waeber G et al. Renin and angiotensin II receptor gene expression in kidneys of renal hypertensive rats. Hypertension 1995; 26: 733–737. Wang ZQ, Millatt LJ, Heiderstadt NT et al. Differential regulation of renal angiotensin subtype AT1A and AT2 receptor protein in rats with angiotensin-dependent hypertension. Hypertension 1999; 33: 96–101. Wiesel P, Mazzolai L, Nussberger J et al. Two-kidney, one clip and onekidney, one clip hypertension in mice. Hypertension 1997; 29: 1025–1030. Bao L, Sachs F, Dahl G. Connexins are mechanosensitive. Am J Physiol Cell Physiol 2004; 287: C1389–C1395. Kurtz A, Wagner C. Cellular control of renin secretion. J Exp Biol 1999; 202: 219–225. Haefliger JA, Castillo E, Waeber G et al. Hypertension increases connexin43 in a tissue-specific manner. Circulation 1997; 95: 1007–1014. Elfgang C, Eckert R, Lichtenberg-Frate H et al. Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J Cell Biol 1995; 129: 805–817. Rackauskas M, Kreuzberg MM, Pranevicius M et al. Gating properties of heterotypic gap junction channels formed of connexins 40, 43, and 45. Biophys J 2007; 92: 1952–1965. Cohen-Salmon M, Regnault B, Cayet N et al. Connexin30 deficiency causes instrastrial fluid-blood barrier disruption within the cochlear stria vascularis. Proc Natl Acad Sci USA 2007; 104: 6229–6234. Simon AM, McWhorter AR. Decreased intercellular dye-transfer and downregulation of non-ablated connexins in aortic endothelium deficient in connexin37 or connexin40. J Cell Sci 2003; 116: 2223–2236. Iacobas DA, Iacobas S, Spray DC. Connexin43 and the brain transcriptome of newborn mice. Genomics 2007; 89: 113–123. Kardami E, Dang X, Iacobas DA et al. The role of connexins in controlling cell growth and gene expression. Prog Biophys Mol Biol 2007. Insel PA, Patel HH. Do studies in caveolin-knockouts teach us about physiology and pharmacology or instead, the ways mice compensate for ‘lost proteins’? Br J Pharmacol 2007; 150: 251–254. Firouzi M, Kok B, Spiering W et al. Polymorphisms in human connexin40 gene promoter are associated with increased risk of hypertension in men. J Hypertens 2006; 24: 325–330. Grayson TH. Is Cx40 a marker for hypertension? J Hypertens 2006; 24: 279–280. Charollais A, Gjinovci A, Huarte J et al. Junctional communication of pancreatic beta cells contributes to the control of insulin secretion and glucose tolerance. J Clin Invest 2000; 106: 235–243. Allagnat F, Martin D, Condorelli DF et al. Glucose represses connexin36 in insulin-secreting cells. J Cell Sci 2005; 118: 5335–5344. Simon AM, McWhorter AR. Vascular abnormalities in mice lacking the endothelial gap junction proteins connexin37 and connexin40. Dev Biol 2002; 251: 206–220. Haefliger JA, Polikar R, Schnyder G et al. Connexin37 in normal and pathological development of mouse heart and great arteries. Dev Dyn 2000; 218: 331–344. Delorme B, Dahl E, Jarry-Guichard T et al. Expression pattern of connexin gene products at the early developmental stages of the mouse cardiovascular system. Circ Res 1997; 81: 423–437. el Aoumari A, Fromaget C, Dupont E et al. Conservation of a cytoplasmic carboxy-terminal domain of connexin 43, a gap junctional protein, in mammal heart and brain. J Membr Biol 1990; 115: 229–240.
Kidney International (2007) 72, 814–822
http://www.kidney-international.org & 2007 International Society of Nephrology
see commentary on page 781
Connexin40 regulates renin production and blood pressure N Krattinger1, A Capponi2, L Mazzolai3, J-F Aubert3, D Caille4, P Nicod1, G Waeber1, P Meda4 and J-A Haefliger1 1
Department of Medicine, University Hospital, Lausanne, Switzerland; 2Department of Internal Medicine, School of Medicine, University of Geneva, Gene`ve, Switzerland; 3Department of Medicine, Service d’Angiologie, University Hospital, Lausanne, Switzerland and 4 Department of Cell Physiology and Metabolism, School of Medicine, University of Geneva, Gene`ve, Switzerland
Renin secretion is regulated by coordinated signaling between the various cells of the juxtaglomerular apparatus. The renin-secreting cells (RSC), which play a major role in the control of blood pressure, are coupled to each other and to endothelial cells by Connexin40 (Cx40)-containing channels. In this study, we show that Cx40 knockout (Cx40/) mice, but not their heterozygous littermates, are hypertensive due to the increase in the number of RSC, renin biosynthesis, and plasma renin. Treatment with the angiotensin II receptor AT1 antagonist candesartan or the angiotensin II-converting enzyme inhibitor ramipril reduced the blood pressure of the Cx40/ mice to the same levels seen in wild-type (WT) mice. The elevated blood pressure of the knockout mice was not affected by clipping one renal artery (2K1C, renindependent model of hypertension) or after a high salt diet. Under these conditions, however, Cx40/ mice showed an altered production and release of renin. The renin mRNA ratio between the clipped and the non-clipped kidney was lower in the knockout than in the WT 2K1C mice. This indicates that the response to a change in blood pressure was altered. The RSC of the Cx40/ mice did not have a compensatory increase in the levels of either Cx43 or Cx37. Our data show that renin secretion is dependent on Cx40 and suggest the Cx40/ mice may be a genetic model of renin-dependent hypertension. Kidney International (2007) 72, 814–822; doi:10.1038/sj.ki.5002423; published online 11 July 2007 KEYWORDS: gap junctions; connexins; transgenic mice; hypertension; secretion
Correspondence: J-A Haefliger, Department of Internal Medicine, Laboratory of Molecular Medicine 19-135S, University Hospital, Lausanne CHUV-1011, Switzerland. E-mail: [email protected] Received 7 March 2007; revised 3 May 2007; accepted 22 May 2007; published online 11 July 2007 814
The control of blood pressure, which is of paramount importance for the organism, is mostly achieved within the kidney cortex via the release of renin by the juxtaglomerular apparatus (JGA).1 Renin is normally secreted in a regulated manner, as a function of the integration of various mechanical, osmotic, and electrochemical stimuli, acting on the different cell types which collectively form the JGA.2 This regulation implies that JGA cells function in coordination. One of the mechanisms whereby this coordination may be achieved is via gap junctions,3–6 the membrane regions where connexin (Cx) channels for the intercellular sharing of ions and signaling molecules are localized.7 A functional gap junction is composed of two opposing hexamers, each contributed by one of two adjacent cells, and consisting of an assembly of six connexin proteins.7 Accordingly, gap junctions made of Cx37, Cx40, and Cx438–10 link the cells of the JGA.11,12 Furthermore, gap junctions made of Cx40 also link the endothelial cells of the afferent arteriole to the reninsecreting cells (RSC), which form the media of the juxtaglomerular part of this arteriole.8,13 Gap junctional communication via Cx40 channels appears to play a central role in the function of the vascular system,13,14 as exemplified by the finding that loss of Cx40 results in chronic hypertension.15–17 While the mechanism of this effect is still unknown, the observation that changes in Cx40 are associated with altered vasomotion and signal conduction along arterioles has led to propose that Cx40 contributes to control blood pressure by mediating the adaptive changes of the vascular walls, that are induced by increased pressure and mechanical stress.15–17 However, since connexins control secretion of various hormones,3,18 and since the arterial expression of Cx40 is increased in rodents featuring a renin-dependent hypertension,8,13 we postulated that loss of Cx40 could also alter the coordinated functioning of the RSC, hence accounting for increased blood pressure. Here, we report on the direct testing of this possibility in mice lacking Cx40 (Cx40/),19,20 and demonstrate that the hypertension of these mice is mostly due to an altered production of renin. Thus, these mice represent a valuable genetic model of renin-dependent increase in blood pressure, the most common situation in human hypertension. Kidney International (2007) 72, 814–822
original article
N Krattinger et al.: Renin secretion and Cx40
d
e
g
f b
***
Volume of cytoplasma (a.u.)
3 2 1 0
h
i Irregular renin-containing granules (%)
Number of renin cells per glomerlus (a.u.)
c
Number of granules per cell profile
a
20 10
***
0
30 25 20 15 10 5 0
***
70 60 50 40 30 20 10 0
***
Figure 1 | Loss of Cx40 alters the organization of the RSC. (a) In control mice, in situ hybridization shows that Cx40 mRNA is expressed along the endothelial cell (e) that line the lumen of the afferent arteriole (aa), as well as between the RSC (rc) that form the media of the juxtaglomerular (gl) portion of this vessel. (b) Immunocytochemistry shows a similar localization of the cognate protein. Bar ¼ 20 mm. (c) Immunofluorescence reveals the alignment of RSC in three afferent arterioles of a Cx40/ mouse. (d and e) Mice lacking Cx40 had more renin-containing cells per glomerule than WT controls. (f–i) When compared to controls (f), the RSC of Cx40/ mice (g) were smaller (e) and contained fewer (h), and more round secretory granules (i). Data are mean7s.e.m. values. ***Po0.001 Cx40/ (solid bars) vs WT values (open bars). Bar ¼ 30 mm in (c) and 2 mm in (f) and (g).
In agreement with previous reports, in situ hybridization and immunofluorescence showed the expression of Cx40 in the endothelial cells of the afferent arterioles of wild-type mice (WT), as well as in RSC that formed the media of the juxtaglomerular portion of this vessel (Figure 1a and b). No Cx40 was found in the corresponding regions of homozygous Cx40/ mice (data not shown). When compared to the RSC of WT, which expressed Cx40, the RSC of Cx40/ littermates (Figure 1c), which lacked Cx40, were more numerous per glomerule (Figure 1d), featured a smaller cytoplasmic volume (Figure 1e) and decreased (Po0.003) numbers of secretory granules (Figure 1h). Furthermore, whereas in control cells, most of the renincontaining granules had an irregular outline (Figure 1f), in cells lacking Cx40, most of these granules were round (Figure 1g and i). Thus, loss of Cx40 alters the cellular composition of the afferent arteriole as well as the intracellular storage of renin. Cx40/ mice are hypertensive due to activation of the renin–angiotensin system
WT (n ¼ 9) and Cx40 þ / mice (n ¼ 8), which expressed Cx40, were normotensive (Figure 2a). In contrast, Cx40/ littermates (n ¼ 10) featured a 40% increase (Po0.001) in blood pressure (Figure 2a). This hypertension was associated with a large increase (Po0.001) in the levels of circulating renin (Figure 2b). Since no significant difference was observed between the first two sets of mice, only the data for the WT controls are shown hereafter. Kidney International (2007) 72, 814–822
70 WT
160
Cx40+/–
140
Cx40–/–
120 100 80 60
***
60 PRA (ng Ang l/ml h)
8–13
180
Mean blood pressure (mm Hg)
RESULTS The RSC express Cx40, and are altered after loss of this connexin
***
50 40 30 20
40 20 0
10 0
Figure 2 | Cx40/ mice are hypertensive due to increased levels of circulating renin. (a) Mean blood pressure was higher in Cx40/ mice (solid bar) than in heterozygous ( þ /; shaded bar) and WT littermates (open bar). (b) Plasma renin activity (PRA) was markedly increased in the hypertensive Cx40/ mice. Values are mean7s.e.m. ***Po0.001 Cx40/ vs WT levels.
Treatment with candesartan, an antagonist of the angiotensin II receptor AT1, or ramipril, an inhibitor of the angiotensin II converting enzyme (Table 1), reduced the blood pressure of Cx40/ mice to the levels observed in control littermates (Figure 3a), in spite of a persistent elevation of plasma renin (Figure 3b). In contrast, plasma renin and the expression of renin mRNA in kidneys were increased in control mice after both treatments (Figure 3b and c). Thus, the hypertension of Cx40/ mice is due to an activation of the renin–angiotensin system (RAS), which results in elevated levels of circulating renin, and that cannot be properly regulated. 815
original article
N Krattinger et al.: Renin secretion and Cx40
Table 1 | Characteristics of the mice treated with inhibitors of the renin–angiotensin system Group
Treatment
n
Body weight (g)
Heart rate (beats/min)
LKi (mg/g)
RKi (mg/g)
WT
Untreated Candesartan Ramipril
9 10 12
28.471.7 30.171.0 26.570.9
657714 628731 634757
6.670.4 6.970.4 6.870.3
6.870.4 6.870.4 6.870.3
Cwi (mg/g) 6.470.3 6.170.2 5.370.3*
Cx40/
Untreated Candesartan Ramipril
10 12 15
25.971.0 31.672.5 27.571.2
648722 610737 68278
6.570.2 6.770.3 7.270.2
6.570.2 6.570.3 7.370.2
8.670.7111 7.670.6*11 7.170.3*11
Cwi, cardiac weight index; LKi, left kidney index (ratio between the left kidney weight and body weight); n, number of mice; RKi, right kidney index. Values are expressed as mean7s.e.m. *Po0.05 vs untreated mice; 11Po0.01 vs WT mice; 111Po0.001 vs WT mice.
70
***
Cx40 –/– 50
***
160
PRA (ng Ang l/ml h)
Mean blood pressure (mm Hg)
180
140 120 100 80
***
***
40
*** 30
20
60 40
10
20 0
0
ol rtan ipril ol rtan ipril ntr ntr Co desa Ram Co desa Ram n n Ca Ca
n n ol ril ril ol ntr rta ip ontr sarta amip C de Co desa Ram R n n Ca Ca
1.2
§
** **
**
**
2.0
** 1.5 1.0
1.0
AT1A/GAPDH mRNA
2.5
Renin/GAPDH mRNA
***
60
WT
0.8 0.6
*** *** ***
0.4
0.5
0.2
0
0
n ipril ol ol rtan ipril ntr rta ntr Co desa Ram Co desa Ram n n Ca Ca
*** ***
Ctrl
LK RK 2K1C
Ctrl
LK RK 2K1C
Figure 3 | High blood pressure and renin levels of Cx40/ mice are caused by an activation of the RAS. (a) Inhibitors of the RAS lowered blood pressure of Cx40/ mice, but did not affect that of WT controls. (b) Circulating renin was not significantly affected by these treatments in Cx40/ mice, but was increased in WT controls (WT). (c) Basal expression of renin mRNA was higher in Cx40/ than in WT mice. Candesartan and ramipril significantly increased the expression of renin in WT animals. (d) Expression of the angiotensin 1 receptor subtype (AT1A) mRNA was lower in Cx40/ than in WT mice. After induction of hypertension in the 2K1C model, the AT1A levels of WT mice decreased to the values observed in Cx40/ mice. Values are mean7s.e.m. **Po0.01, ***Po0.001 Cx40/ (solid bars) vs WT levels (open bars). y Po0.05 vs levels in untreated Cx40/ mice.
In control mice, increased release of renin leads to enhanced generation of angiotensin II which, in turn, decreases the expression of the AT1A receptor subtype, that 816
ultimately controls the RAS.21,22 The expression of this receptor was significantly lower in Cx40/ mice (n ¼ 10) than in WT littermates (n ¼ 9) (Figure 3d), in agreement Kidney International (2007) 72, 814–822
original article
N Krattinger et al.: Renin secretion and Cx40
To identify the underlying mechanism, we first investigated the effect of diets known to increase (0.03% NaCl) or decrease (4% NaCl) plasma renin in control mice, under conditions of constant blood pressure (Figure 4a and b). Three weeks after exposure to these diet, the blood pressure of 9 (low-salt diet) to 11 (high-salt diet) Cx40/ mice was also unchanged and remained significantly higher than that of WT (Figure 4a). The circulating levels of renin of Cx40/ mice also remained markedly elevated compared to physiological levels, and did not show the changes observed in WT controls. Thus, whereas these levels were slightly reduced (about half the control change) after the low-salt diet, they did not increase, as anticipated by the physiology, after the high salt intake (Figure 4b). These differences were seen in spite of a similar regulation of the kidney expression of the renin transcript in the Cx40/ and the WT mice, during both the low and high-salt diets (Figure 4c). Thus, loss of Cx40 did affect selective aspects of the regulation of renin secretion dependent on the macula densa. To evaluate how this dysregulation affected the RSC, WT and Cx40/ mice were submitted to a surgical 2K1C procedure.10,23 The left, hypoperfused kidney of all 2K1C animals weighed less than that of the shamoperated mice, whereas the contralateral kidney was significantly larger than that of these controls (Table 2). Also, the heart of hypertensive Cx40/ mice was hypertrophied compared to that of WT controls, as indicated by a 45% increase in cardiac weight index (Table 2). Four weeks after clipping one renal artery, we observed an increase in the blood pressure of WT mice (Figure 5a), due to elevated (Po0.01) levels of circulating renin (Figure 5b), as a result of a larger production of the enzyme in the hypoperfused left kidney (Figure 5c and d). When submitted to the same protocol, Cx40/ mice retained unchanged their high blood pressure (Figure 5a) and circulating levels of renin (Figure 5b), in spite of increased renin expression within the clipped kidney (Figure 5c and d). However, this change was only half that observed in WT mice, indicating that the regulation of renin secretion under conditions of altered kidney perfusion was altered in Cx40/ mice. Consistent with this finding, in situ hybridization demonstrated that the levels of renin mRNA were also significantly less reduced in the unclipped kidney of Cx40/ mice than in that of WT littermates (Figure 5d). Kidney International (2007) 72, 814–822
b
70
*** WT Cx40–/–
***
60
PRA (ng Ang l/ml h)
180
*** *** ***
160 140 120 100 80
50 40
§ ***
30 20
60 40
*
10
20 0
* 0 NS LS HS NS LS HS
c
NS LS HS NS LS HS
WT NS
LS
LK RK LK RK
Cx40–/– HS
NS
LK RK
LK RK
LS
HS
LK RK LK RK
Renin GAPDH
§ **
4 Renin/GAPDH mRNA
The regulation of renin production and secretion is altered in Cx40/ mice
a
Mean blood pressure (mm Hg)
with the much higher levels of circulating renin of the former mice. The WT levels of AT1A decreased up to the values observed in Cx40/ animals when WT mice were submitted to an experimental procedure (2K1C) that causes hypertension by elevating the circulating levels of renin. Under the same conditions, the levels of AT1A did not change in Cx40/ mice (Figure 3d).
3
**
2
* 1 0
*
NS LS HS
§ *
NS LS HS
Figure 4 | The regulation of renin secretion by salt is altered in Cx40/ mice. (a) Mean blood pressure of WT (WT) and Cx40/ mice was not altered after feeding a normal (NS), a low (LS), or a highsalt diet (HS). (b) Under the latter diet, the circulating levels of renin decreased in both WT and Cx40/ mice. In contrast, a low salt diet increased renin secretion in WT but not Cx40/ mice. (c) Northern blots showed that the low and the high-salt diets had opposite effects on the expression of the renin gene. Data are mean7s.e.m. values. *Po0.05, **Po0.01, ***Po0.001 vs the NS value of WT mice (open bars); yPo0.05 vs. NS values of Cx40/ mice (solid bars).
Loss of Cx40 is not compensated by other connexins in the RSC
In the 2K1C model, the levels of Cx40 mRNA were found increased in the hypoperfused kidney of WT mice and reduced in the contralateral kidney, in agreement with the differential regulation of renin expression in the two organs (Figure 6a). Hypertension, was also associated with increased expression of Cx43 mRNA in both WT and Cx40/ mice kidneys (Figure 6a), in keeping with the previously reported changes of this connexin in the vascular wall of hypertensive animals.8,10,13 The renal mRNA levels of Cx37, another connexin of the vasculature which is found in the JGA,9 were not modified by the 2K1C procedure in WT controls (Figure 6a). In contrast, these levels were elevated in the 817
original article
N Krattinger et al.: Renin secretion and Cx40
Table 2 | Characteristics of the 2K1C and sham-operated mice Group
Treatment
n
Body weight (g)
Heart rate (beats/min)
LKi (mg/g)
RKi (mg/g)
Cwi (mg/g)
WT
Sham 2K1C Sham 2K1C
9 9 11 11
30.570.6 28.870.8 26.371.6 28.270.6
654718 641724 655718 669713
7.370.2 6.370.2* 7.370.4 5.970.2**
7.370.2 8.570.3** 7.270.3 8.470.5**
6.570.3 7.870.3* 9.671.211 9.070.611
Cx40/
Cwi, cardiac weight index; LKi, left kidney index; n, number of mice; RKi, right kidney index. Values are expressed as mean7s.e.m. *Po0.05, **Po0.01 vs 2K1C sham-operated animals; 11Po0.01 vs WT animals.
180 160
b
WT Cx40–/–
***
**
140
***
d
70 60
PRA (ng Ang l/ml h)
Mean blood pressure (mm Hg)
a
120 100 80 60
***
50
aa
aa
40
G
G
30 20
**
40
WT LK
WT RK
10
20 0
***
Sham 2K1C 2K
0
Sham 2K1C 2K
c
WT
aa
Cx40–/–
Sham
2K1C
LK RK
LK RK
Sham
G
aa
G
Sham 2K1C Sham 2K1C 2K 2K
2K1C
LK RK
LK RK
Renin
WT LK 2K1C
WT RK 2K1C
GAPDH
aa G G
3.0
§ **
Renin/GAPDH mRNA
2.5 2.0
aa Cx40–/– RK
Cx40–/– LK
**
* ## §
1.5
G
1.0
G
0.5
***
0 Sham LK
RK 2K1C
aa Sham LK RK 2K1C
Cx40–/– LK 2K1C
aa Cx40–/– RK 2K1C
Figure 5 | Renin secretion and hypertension depend on the expression of Cx40. (a) The 2K1C procedure increased the blood pressure of WT but not of Cx40/ mice. (b) Plasma renin followed the same pattern model. (c) Changes of renin expression took place in both the hypoperfused (LK) and the controlateral (RK) kidneys of both WT and Cx40/ mice. However, the effects on renin mRNA levels were smaller in Cx40/ than in WT animals. (d) In situ hybridization showed that the levels of this transcript in the afferent arterioles (aa) also decreased much less in the kidneys of 2K1C Cx40/ mice than in those of WT controls (right panels). G, glomerulus. Bar ¼ 80 mm. Data are mean7s.e.m. values of the number of mice indicated in Table 2. *Po0.05, **Po0.01, ***Po0.001 vs the cognate sham value in WT mice (open bars); yPo0.05 vs sham value of Cx40/ mice (solid bars); ##Po0.001 vs the value in the right kidney (RK) of WT mice.
hypertensive, sham-operated Cx40/ mice, and in the hypoperfused left kidney of the Cx40/ mice which were submitted to the 2K1C procedure (Figure 6a). To assess whether these renal changes in the mRNA levels of Cx43 and Cx37 provided a compensation for the loss of Cx40 in RSC, we immunostained kidney sections with 818
specific antibodies to these two connexins. In agreement with previous reports,9,10 we identified Cx43 and Cx37 in the interlobular vessels of WT controls, as well as in the wall of their afferent arterioles, and glomerular vessels (Figure 6b and c). In contrast, the RSC of WT mice were immunostained for Cx37 (Figure 6b), but not Cx43 (Figure 6c). In Kidney International (2007) 72, 814–822
original article
N Krattinger et al.: Renin secretion and Cx40
3.5
WT Cx40–/–
Cxs/GAPDH mRNA
3.0 2.5
˚˚
2.0
* *
* 1.5
* * *
* *
1.0
*
0.5 0
SK LK RK SK LK RK SK LK RK SK LK RK SK LK RK SK LK RK Sham 2K1C Sham 2K1C Sham 2K1C Sham 2K1C Sham 2K1C Sham 2K1C
Cx40
Cx43
Cx37
Figure 6 | Cx37 and Cx43 are not over expressed in the RSC of the hypertensive Cx40/ mice. (a) Analysis of total kidney RNA revealed different changes of the transcripts of Cx40, Cx43 and Cx37 in both the hypoperfused (LK) and the controlateral (RK) kidneys of WT (open bars) and Cx40/ mice (solid bars) submitted to the 2K1C procedure. Data are mean7s.e.m. values of the number of mice indicated in Table 2. *Po0.05, vs sham value in WT mice. 11Po0.01. (b) Antibodies detected Cx37 in the intra-glomerular vessels of WT mice (top panels), as well as in the media of the juxta-glomerular portion of the afferent arterioles, where RSC are located (arrows in middle top panel), and between the endothelial cells of interlobular arteries (arrows in left bottom panel). A similar distribution was observed in Cx40/ mice, even though the levels of Cx37 appeared decreased in both glomeruli (middle bottom panel) and interlobular vessels (right bottom panel). (c) Cx43 was expressed by the endothelial cells of renal vessels, including the afferent arteriole (left top panel). However, and at variance with Cx37, Cx43 was not found in the RSC of this vessel (arrow in left top panel). A similar distribution and expression levels were observed in the afferent arterioles (aa) of Cx40/ mice (middle and right top panels; arrows point to putative RSC). In these mice, Cx43 was also observed between the endothelial cells of interlobular arteries (ila, arrows in left bottom panel), at the low levels observed in controls. Cx43 appeared increased between the smooth muscle cells of the media of large arcuate arteries (arrows in right bottom panel). gl, glomerulus; aa, afferent arteriole; ila, interlobular artery; md, macula densa; ara, arcuate artery. Bar ¼ 25 mm.
Cx40/ mice, a similar distribution was observed for both Cx37 and Cx43. However, Cx37 appeared reduced in vessels and RSC (Figure 6b), whereas Cx43 appeared increased (Figure 6c) in the intimal endothelium and in smooth muscle cells of the intima and media of large interlobular vessels, respectively. As in controls, RSC of Cx40/ mice expressed some Cx37 (Figure 6b) but no Cx43 (Figure 6c). Thus, loss of Cx40 in RSC is not compensated by an obvious change in the expression of other connexin proteins, which are expressed in renal vessels and JGA. DISCUSSION
The gap junction protein Cx40 has been implicated in the function of the vascular system, notably because knockout mice lacking this connexin are hypertensive. Previous studies have concluded that this hypertension Kidney International (2007) 72, 814–822
was mostly due to an abnormal adaptation of the arterial walls to altered hemodynamic conditions,15–17 implying that Cx40 contributes to the detection and propagation of the blood-borne signals that are elicited by increased blood pressure. Here, we show that the hypertension of Cx40 mice mostly results from alterations in the number and function of the RSC, and the ensuing activation of the RAS. We further document that, in the absence of Cx40, the regulation of this system is altered both under a low-salt regimen and under conditions of changed renal blood flow, two situations that are sensed by different cell types of the JGA.1,2 The data show that connexins play a central role in the signaling from these cells, that converge on the myo-epithelial cells of the juxtaglomerular portion of the afferent arterioles, which secrete renin. 819
original article
These data have two implications. First, they extend to the RSC, the increasing evidence pointing to connexin signaling as a major, obligatory player in the control of secretory systems.3,18 Our data document for the first time that loss of Cx40 increases the number of RSC, the renal production of the hormone and its release, indicating that the signaling dependent on this protein natively downregulates the function of RSC. Second, our data show that Cx40 plays a central role in the regulatory function of the JGA. Thus, loss of this connexin markedly activates the RAS, altering its normal regulation of blood pressure, and resulting in hypertension. In view of its distribution in the membrane of both endothelial cells and RSC of the afferent arteriole, and its mechano-sensitive characteristics,24 Cx40 is ideally located to detect bloodborne signals and to propagate them across the vascular and the RSC of the JGA, thus permitting the integration of the mechanical, hormonal, and ionic inputs that control blood pressure.2,25 By comparing WT and Cx40/ mice in a renin-dependent (2K1C) model of hypertension our experiments show for the first time that the hypertension of Cx40/ animals is essentially due to an excessive secretion of renin, due to increased numbers of RSC, which feature alterations of their enzyme-containing granules. Most probably, the signaling mechanism of Cx40 is integrated with that of both Cx43 and Cx37, two other connexins that are also expressed within the JGA,8,10 and that are affected by hypertension.8,10,13,26 With regard to Cx43, we have recently provided evidence that replacement of this protein by the Cx32 isoform, which cannot form channels with Cx40,27 downregulates the secretion of RSC, protecting transgenic mice against the hypertension which is expected in a model of renovascular hypertension.10 Cx43 is not detected in the RSC of WT animals, but is expressed by rare endothelial cells of arteries, as well as by most smooth muscle cells of these vessels,13 including the afferent arteriole. Consistent with previous reports of elevated vascular levels of Cx43 in different models of experimental hypertension,8,10,13,26 we show here that the renal levels of Cx43 were higher in the hypertensive Cx40/ mice than in the cognate, normotensive controls. We further show that, when the latter animals were made hypertensive by the 2K1C procedure, the levels of Cx43 expression rose in the kidneys of controls, up to the levels observed in the Cx40/ mice that featured a similar degree of hypertension, due to excessive renin secretion. We have not observed an induction of Cx43 in the RSC of the Cx40/ animals. Therefore, the contribution of this connexin to the dysregulation of renin secretion may imply the mediation of an extracellular signal initiated in the endothelial and/or smooth muscle cells of the afferent arteriole, since Cx40 and Cx43 appear unable to form functional heterotypic gap junction channels.28 The close apposition of RSC, endothelial, and smooth muscle cells of the JGA provides an ideal setting for paracrine effects within the afferent arteriole.8 An analogous, connexion-dependent 820
N Krattinger et al.: Renin secretion and Cx40
paracrine mechanism has been recently postulated in the stria vascularis epithelium of the inner ear.29 With regard to Cx37, its coexpression with Cx40 in the RSC9 raises the question of which of the two connexin isoforms is actually responsible for the increased renin secretion of Cx40/ mice. Several observations indicate that Cx40 largely predominates in this role. Thus, Cx37/ mice, which still express Cx40, are normotensive,30 and no obvious change was observed in the expression of Cx37 in hypertensive 2K1C WT mice, which feature a striking upregulation of renin expression. Still, and in agreement with a previous suggestion in the vascular compartment,30 we observed that the levels of the Cx37 protein were decreased in the kidneys of the Cx40/ mice, indicating that the expression of Cx40 and Cx37 are somehow interdependent. Our findings are consistent with a model in which proper renin secretion depends both on the homotypic signaling ensured by Cx40 between the RSC themselves, and on a heterotypic signaling, mediated by Cx40, and possibly Cx37, between RSC and the endothelial and/or smooth muscle cells of the afferent arteriole. Full validation of these tentative conclusions requires that the paracrine signal(s) and mechanism dependent on Cx43 be identified, and that the incidence and extent of the homotypic RSC coupling dependent on Cx40 be compared, within the intact kidney, to those of the heterotypic communication between RSC and vascular cells mediated by Cx37–Cx40 channels. Both are formidable tasks in view of the small number and dispersion of RSC within the kidney cortex, and the tight intermingling of the different cell types that collectively form the JGA. Therefore, at least two complementary, if not alternative, possibilities should be considered at this stage. The first possibility finds its basis on recent studies that have documented that specific connexins selectively affect the expression of a variety of genes, by a mechanism that remains to be elucidated and that may be independent of the establishment of cell-to-cell communication.29,31,32 Thus, large-scale transcriptome analysis of different cell and animal models have shown that Cx43 and Cx32 regulate an array of genes involved in the control of cell growth,31,32 whereas Cx30 appears to preferentially control genes implicated in the metabolism of homocysteine.29 Strikingly, the tissue accumulation of this amino acid, which is seen in the absence of Cx30, triggers a local, presumably paracrine mechanism which eventually results in the altered function of endothelial cells that do not express this connexin isoform.29 The second more intriguing possibility is that, as in any other study involving knockout mice,33 the physiological relevance of the effects we observed may be overestimated, because of the compounding effect of experimentally induced alterations, that, in our case, could be largely due to over time compensation of the missing Cx40 by other connexin isoforms or some connexin-dependent gene. However, the finding that both the pharmacological (RAS inhibitors, salt load, or depletion) and surgical interventions (2K1C) elicited similar, and anticipated, blood pressure and renin changes in WT and Cx40 þ / mice, Kidney International (2007) 72, 814–822
N Krattinger et al.: Renin secretion and Cx40
whereas selective alterations of these physiological changes were solely observed in Cx40/ mice, plaids against this possibility. Furthermore, while we have documented changes in the transcripts of other connexin isoforms in the kidney of the hypertensive animals, we have also shown that the cognate proteins were either not ectopically expressed (Cx43) or at best somewhat decreased (Cx37) in RSC. The latter change is unlikely to have cause the pressure perturbations we observed, inasmuch as Cx37/ mice, which still express Cx40, are normotensive.30 Still, what is the actual physiological role of Cx40 in the kidney of living WT mice remains to be validated in other models, including cell- and timespecific knockout and siRNA interference studies. By acutely decreasing the expression of Cx40 in selected cell types, these approaches should avoid the conundrum of compensation and mechanistic adaptations that probably occur over time in any knockout animal. Awaiting the completion of such studies, our findings open new paradigms to envision the still obscure pathogenesis of hypertension. Since polymorphisms of the Cx40 gene have been recently associated with human hypertension,34,35 our data call for an evaluation of renin function in the affected patients. They also point to the Cx40/ mice as a pertinent genetic model to unravel further the molecular mechanisms of renin-dependent hypertension, a most common situation in the human clinic. MATERIALS AND METHODS Knockout mice The generation of Cx40/ mice and their genotypic characterization have been described.19 Breeding pairs (kindly provided by Dr K Willecke) were crossed with control C57BL6 partners. Identification of the WT, heterozygous ( þ /), and homozygous Cx40/ mice was performed by PCR of genomic DNA using the following primers: for the Cx40 WT allele (314 bp), 50 -GGGAGATG AGCAGGCCGACTTCCGGTGCG-30 (sense) and 50 -GTAGGGTGC CCTGGAGGACAATCTTCCC-30 (antisense); for the Cx40 recombinant allele (494 bp), 50 -GGATCGGCCATTGAACAAGATGGAT TGCAC-30 (sense) and 50 -CTGATGCTCTTCGTCCAGATCATCCTG ATCG-30 (antisense). Three 4-month-old animals, from litters obtained after at least 10 backcrosses into the C57/BL6 background, were used in all the experiments. Morphometric and ultrastructural analysis To analyze the number and distribution of RSC, kidneys were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, embedded in paraffin, and serially sectioned at 5 mm thickness. Sections were immunostained using a rabbit polyclonal antibody against renin (2D12, courtesy of Professor P Corvol, Paris), diluted 1:200, and evaluated morphometrically as per standard methods.36 Animal experiments Mouse care, surgery, and euthanasia procedures were approved by our Institutional Committee for Animal Experiments and the Veterinary Office (Lausanne, Switzerland). Modulating RAS. Blockade of the RAS was performed by adding to the drinking water either 3 mg/kg day the AT1-receptor antagonist candesartan (Cilexetils, Astra Zeneca, Zoug, SwitzerKidney International (2007) 72, 814–822
original article
land) or 10 mg/kg day the angiotensin-converting enzyme inhibitor ramipril (Aventis, Geneva, Switzerland) for 3 weeks. To test the effects of a salt load, another series of mice was fed for 3 weeks with a diet (Kliba, Gossau, Switzerland) containing either 0.03% (low-salt group), 4% (high-salt group), or 0.3% NaCl (control group). All mice had free access to tap water. Inducing a renin-dependent hypertension. For the twokidney, one clip model (2K1C), the left renal artery was clipped with a 0.12 mm internal diameter U-shaped silver clip, as described previously.10,23 Measurement of blood pressure. One day before killing, the right carotid artery was penetrated with a catheter connected to a pressure transducer and a computerized data acquisition system, that allowed for recording heart rate and both diastolic and systolic intra-arterial pressure in conscious mice every 20 s.10 Data are shown as the mean (MBP) of the diastolic and systolic values. After a 30-min recording period, 300 ml blood was collected in ethylenediaminetetraacetic acid-coated tubes and centrifuged at 41C. Plasma was stored at 701C until use. Measurements of plasma renin Plasma renin activity was measured by a microassay based on angiotensin I (Ang I) trapping by an antibody.10 Controls showed that endogenous angiotensinogen was not a limiting factor for the measurement of the high levels of renin found in the hypertensive mice (data not shown). RNA analysis RNA isolation and Northern blot analysis were performed as described10 using cDNA probes specific for either renin10 or a 1.1 kb (HindIII–EcoRI) fragment of GAPDH cDNA.8 Quantitative realtime PCR was performed as reported10,37 using the following primers: for renin (158 bp), -50 -TCTCTGGGCACTCTTGTTGC TCTG-30 - (sense) and -50 -ATACGTCCCATTCAGCACTGAGCC-30 (antisense); for GAPDH (334 bp), -50 -gactccacgacatactcagc-30 (sense) and -50 -gtcggtgtgaacggatttgg-30 - (antisense); for AT1a (110 bp), 50 -GTCGCACTCAAGCCTGTCTA-30 (sense); for Cx40 (310 bp), -50 -TTTGGCAAGTCACGGCAGGG-30 (sense) and 50 -TG TCACTATGGTAGCCCTGAG-30 (antisense);38 for Cx37 (144 bp), 50 -GGGAGATAAAGGCACGAAGG-30 (sense) and 50 -ATGAGAGC CCGTTGTAGGTG-30 (antisense); for Cx43 (255 bp), 50 -GATTGAA GAACACGGCAAGG-30 (sense) and 50 -AGAGCGAGAGACACCA AGGA-30 (antisense). Analysis of data was performed using the 3.5 version of the Light Cycler software (Roche Diagnostics, Zong, Switzerland, GmbH, Germany). For the in situ hybridization of the renin and Cx40 transcripts, kidneys sections were hybridized as previously described.8,10 Analysis of connexins Immunodetection of connexins was performed as previously reported.10 Cryostat sections of kidneys (5 mm thickness) were permeabilized 10 min in cold (201C) acetone, and incubated 2 h at room temperature, using the following rabbit polyclonal antibodies: for Cx40, the antibody described by Haefliger et al.,8 diluted 1:250; for Cx37, the antibodies described by Haefliger et al.39 and Delorme et al.,40 diluted 1:100; for Cx43, the antibodies 71-0700 (Zymed, Basel, Switzerland), diluted 1:100, C6219 (Sigma) diluted 1:200, or that described,41 diluted 1:50. At the end of the immunolabeling, the sections were counterstained 3 min in 0.03% Evans blue (in phosphate-buffered saline), which gives the tissues a red background 821
original article
staining that enhances the visualization of the fluorescein isothiocynate immunosignals.36
N Krattinger et al.: Renin secretion and Cx40
16.
17.
Statistical analysis Data were expressed as mean7s.e.m. Differences between means were assessed using Student’s t-test. Statistical significance was defined at a value of Po0.05 (* or, 1 or #, or y), Po0.01 (**, or 11, or ## , or yy), Po0.001 (***, or 111, or ###, or yyy). ACKNOWLEDGMENTS
This work was supported by grants from the Swiss National Science Foundation (Grant nos. 310000-109530, 3100A0-100797, and 310000-109402), the Juvenile Diabetes Foundation International (Grant nos. 1-2005-46 and 1-2007-158), the Placide Nicod Foundation, the Octav and the Marcella Botnar Foundation, the Novartis Foundation, The Emma Muschamp Foundation, Novo Nordisk, and the Geneva Program for Metabolic Diseases.
18.
19.
20.
21.
22.
23.
NOTE ADDED IN PROOF After the submission of our study, the following papers independently confirmed that the hypertension of Cx40/ mice is largely due to altered function (Circ Res 2007; 100: 556) and organization of renin-producing cells (J Am Soc Nephrol 2007; 18: 1103).
24. 25. 26. 27.
REFERENCES 1. Mullins LJ, Bailey MA, Mullins JJ. Hypertension, kidney, and transgenics: a fresh perspective. Physiol Rev 2006; 86: 709–746. 2. Schweda F, Kurtz A. Cellular mechanism of renin release. Acta Physiol Scand 2004; 181: 383–390. 3. Michon L, Nlend Nlend R, Bavamian S et al. Involvement of gap junctional communication in secretion. Biochim Biophys Acta 2005; 1719: 82–101. 4. Le Gurun S, Martin D, Formenton A et al. Connexin-36 contributes to control function of insulin-producing cells. J Biol Chem 2003; 278: 37690–37697. 5. Calabrese A, Caton D, Meda P. Differentiating the effects of Cx36 and E-cadherin for proper insulin secretion of MIN6 cells. Exp Cell Res 2004; 294: 379–391. 6. Ravier MA, Guldenagel M, Charollais A et al. Loss of connexin36 channels alters beta-cell coupling, islet synchronization of glucose-induced Ca2+ and insulin oscillations, and basal insulin release. Diabetes 2005; 54: 1798–1807. 7. Sohl G, Willecke K. Gap junctions and the connexin protein family. Cardiovasc Res 2004; 62: 228–232. 8. Haefliger JA, Demotz S, Braissant O et al. Connexins 40 and 43 are differentially regulated within the kidneys of rats with renovascular hypertension. Kidney Int 2001; 60: 190–201. 9. Zhang J, Hill CE. Differential connexin expression in preglomerular and postglomerular vasculature: accentuation during diabetes. Kidney Int 2005; 68: 1171–1185. 10. Haefliger JA, Krattinger N, Martin D et al. Connexin43-dependent mechanism modulates renin secretion and hypertension. J Clin Invest 2006; 116: 405–413. 11. Guo R, Liu L, Barajas L. RT-PCR study of the distribution of connexin 43 mRNA in the glomerulus and renal tubular segments. Am J Physiol 1998; 275: R439–R447. 12. Mink D, Schiller A, Kriz W et al. Interendothelial junctions in kidney vessels. Cell Tissue Res 1984; 236: 567–576. 13. Haefliger JA, Nicod P, Meda P. Contribution of connexins to the function of the vascular wall. Cardiovasc Res 2004; 62: 345–356. 14. Figueroa XF, Isakson BE, Duling BR. Connexins: gaps in our knowledge of vascular function. Physiology (Bethesda) 2004; 19: 277–284. 15. de Wit C, Roos F, Bolz SS et al. Impaired conduction of vasodilation along arterioles in connexin40-deficient mice. Circ Res 2000; 86: 649–655.
822
28.
29.
30.
31. 32. 33.
34.
35. 36.
37. 38.
39.
40.
41.
de Wit C, Roos F, Bolz SS et al. Lack of vascular connexin 40 is associated with hypertension and irregular arteriolar vasomotion. Physiol Gen 2003; 13: 169–177. Figueroa XF, Paul DL, Simon AM et al. Central role of connexin40 in the propagation of electrically activated vasodilation in mouse cremasteric arterioles in vivo. Circ Res 2003; 92: 793–800. Klee P, Boucard N, Caille D et al. Connexins modulators of endocrine function. Winterhager E(ed). Springer Verlag, Berlin, Heldelberg. Gap Junctions in Development and Disease 2005: 197–221. Kirchhoff S, Nelles E, Hagendorff A et al. Reduced cardiac conduction velocity and predisposition to arrhythmias in connexin40-deficient mice. Curr Biol 1998; 8: 299–302. Simon AM, Goodenough DA, Paul DL. Mice lacking connexin40 have cardiac conduction abnormalities characteristic of atrioventricular block and bundle branch block. Curr Biol 1998; 8: 295–298. Haefliger JA, Bergonzelli G, Waeber G et al. Renin and angiotensin II receptor gene expression in kidneys of renal hypertensive rats. Hypertension 1995; 26: 733–737. Wang ZQ, Millatt LJ, Heiderstadt NT et al. Differential regulation of renal angiotensin subtype AT1A and AT2 receptor protein in rats with angiotensin-dependent hypertension. Hypertension 1999; 33: 96–101. Wiesel P, Mazzolai L, Nussberger J et al. Two-kidney, one clip and onekidney, one clip hypertension in mice. Hypertension 1997; 29: 1025–1030. Bao L, Sachs F, Dahl G. Connexins are mechanosensitive. Am J Physiol Cell Physiol 2004; 287: C1389–C1395. Kurtz A, Wagner C. Cellular control of renin secretion. J Exp Biol 1999; 202: 219–225. Haefliger JA, Castillo E, Waeber G et al. Hypertension increases connexin43 in a tissue-specific manner. Circulation 1997; 95: 1007–1014. Elfgang C, Eckert R, Lichtenberg-Frate H et al. Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J Cell Biol 1995; 129: 805–817. Rackauskas M, Kreuzberg MM, Pranevicius M et al. Gating properties of heterotypic gap junction channels formed of connexins 40, 43, and 45. Biophys J 2007; 92: 1952–1965. Cohen-Salmon M, Regnault B, Cayet N et al. Connexin30 deficiency causes instrastrial fluid-blood barrier disruption within the cochlear stria vascularis. Proc Natl Acad Sci USA 2007; 104: 6229–6234. Simon AM, McWhorter AR. Decreased intercellular dye-transfer and downregulation of non-ablated connexins in aortic endothelium deficient in connexin37 or connexin40. J Cell Sci 2003; 116: 2223–2236. Iacobas DA, Iacobas S, Spray DC. Connexin43 and the brain transcriptome of newborn mice. Genomics 2007; 89: 113–123. Kardami E, Dang X, Iacobas DA et al. The role of connexins in controlling cell growth and gene expression. Prog Biophys Mol Biol 2007. Insel PA, Patel HH. Do studies in caveolin-knockouts teach us about physiology and pharmacology or instead, the ways mice compensate for ‘lost proteins’? Br J Pharmacol 2007; 150: 251–254. Firouzi M, Kok B, Spiering W et al. Polymorphisms in human connexin40 gene promoter are associated with increased risk of hypertension in men. J Hypertens 2006; 24: 325–330. Grayson TH. Is Cx40 a marker for hypertension? J Hypertens 2006; 24: 279–280. Charollais A, Gjinovci A, Huarte J et al. Junctional communication of pancreatic beta cells contributes to the control of insulin secretion and glucose tolerance. J Clin Invest 2000; 106: 235–243. Allagnat F, Martin D, Condorelli DF et al. Glucose represses connexin36 in insulin-secreting cells. J Cell Sci 2005; 118: 5335–5344. Simon AM, McWhorter AR. Vascular abnormalities in mice lacking the endothelial gap junction proteins connexin37 and connexin40. Dev Biol 2002; 251: 206–220. Haefliger JA, Polikar R, Schnyder G et al. Connexin37 in normal and pathological development of mouse heart and great arteries. Dev Dyn 2000; 218: 331–344. Delorme B, Dahl E, Jarry-Guichard T et al. Expression pattern of connexin gene products at the early developmental stages of the mouse cardiovascular system. Circ Res 1997; 81: 423–437. el Aoumari A, Fromaget C, Dupont E et al. Conservation of a cytoplasmic carboxy-terminal domain of connexin 43, a gap junctional protein, in mammal heart and brain. J Membr Biol 1990; 115: 229–240.
Kidney International (2007) 72, 814–822