The transient receptor potential vanilloid 5 (TRPV5) channel determines urinary Ca2+ excretion, and is therefore critical for Ca2+ homeostasis. Interestingly, mice lacking the serine protease tissue kallikrein (TK) exhibit robust hypercalciuria comparable to the Ca2+ leak in TRPV5 knockout mice. Here, we delineated the molecular mechanism through which TK stimulates Ca2+ reabsorption. Using TRPV5‐expressing primary cultures of renal Ca2+‐transporting epithelial cells, we showed that TK activates Ca2+ reabsorption. The stimulatory effect of TK was mimicked by bradykinin (BK) and could be reversed by application of JE049, a BK receptor type 2 antagonist. A cell permeable analog of DAG increased TRPV5 activity within 30 min via protein kinase C activation of the channel since mutation of TRPV5 at the putative PKC phosphorylation sites S299 and S654 prevented the stimulatory effect of TK. Cell surface labeling revealed that TK enhances the amount of wild‐type TRPV5 channels, but not of the TRPV5 S299A and S654A mutants, at the plasma membrane by delaying its retrieval. In conclusion, TK stimulates Ca2+ reabsorption via the BK‐activated PLC/DAG/PKC pathway and the subsequent stabilization of the TRPV5 channel at the plasma membrane.
The overall Ca2+ balance is regulated by a homeostatic mechanism tightly controlling the concerted actions of intestinal Ca2+ absorption, exchange of Ca2+ from bone and renal Ca2+ reabsorption. In kidney, Ca2+ can cross the epithelial cells and reach the blood compartment via two pathways: passive (paracellular) and active (transcellular) Ca2+ reabsorption. Active Ca2+ reabsorption takes place in the distal convoluted (DCT) and the connecting (CNT) tubules (Hoenderop et al, 2005). Although it accounts only for ∼15% of total renal Ca2+ reabsorption, it is generally considered the site for fine‐tuning of urinary Ca2+ excretion. Furthermore, active Ca2+ reabsorption is the primary target for regulation by calciotropic hormones, including 1,25‐dihydroxyvitamin D3 and parathyroid hormone (PTH), enabling the organism to regulate Ca2+ reabsorption to the body's demand.
The transient receptor potential vanilloid 5 (TRPV5) channel is expressed along the apical membrane of DCT and CNT and represents the rate‐limiting step in renal transcellular Ca2+ reabsorption (Hoenderop et al, 2005). Inactivation of TRPV5 in mice (TRPV5−/−) abolishes active Ca2+ reabsorption in kidney resulting in severe hypercalciuria (Hoenderop et al, 2003). To compensate this renal Ca2+ leak, TRPV5−/− mice exhibit intestinal Ca2+ hyperabsorption. In addition, the bone structure of these mice is significantly disturbed as illustrated by a reduced trabecular and cortical bone thickness (Hoenderop et al, 2003). Hence, these data demonstrate the key function of TRPV5 in active Ca2+ reabsorption and its essential role in the body Ca2+ homeostasis.
Interestingly, hypercalciuria was recently observed in tissue kallikrein‐deficient (TK−/−) mice (Picard et al, 2005). TK is a serine protease produced in CNT (Figueroa et al, 1988), where it co‐localizes with TRPV5 (Hoenderop et al, 2003). Proteolytic enzymes such as TK are synthesized as inactive precursors or zymogens, to prevent protein degradation and to enable spatial and temporal regulation of enzymatic activity. The precursor of TK is converted to the mature active form before entering the luminal tubular compartment. Trypsin, plasma kallikrein, plasmin and thermolysin can cleave in vitro the TK precursor, but the endogenous activator is still unknown (Margolius, 1995). Once activated, TK is excreted and can process low molecular weight kininogen to release kinin, which acts through kinin receptors such as the bradykinin (BK) 2 receptor (B2R) (Bhoola et al, 1992). Remarkably, recent studies show that TK can directly activate the B2R independently of BK release (Hecquet et al, 2000). However, the molecular events that link TK to Ca2+ balance are at present unknown.
The aim of this study was, therefore, to elucidate the molecular mechanism of hypercalciuria observed in TK−/− mice. To this end, the relation between TK expression and hypercalciuria was investigated in vivo using TK−/− and TRPV5−/− mice. Subsequently, the effect of TK on transcellular Ca2+ transport was examined in primary cultures of renal CNT/cortical collecting duct (CCD) cells. Finally, the signaling pathway through which TK acts on Ca2+ reabsorption and its effect on TRPV5 surface expression were delineated in TRPV5‐expressing cells.
TK expression levels inversely correlates with urinary Ca2+ excretion
The relation between urinary TK expression and urinary Ca2+ excretion was investigated by using wild‐type (WT) (TK+/+) and TK−/− mice fed a 0. 3% w/w (normal) and 3% w/w (high) Na+ diet. Urinary Ca2+ excretion was significantly increased in TK+/+ mice on high Na+ diet compared to mice on normal diet (P<0.05, n=12 mice) (Figure 1A). Conversely, Ca2+ excretion in TK−/− mice on normal Na+ diet was significantly higher (140±13%) than in TK+/+ mice (P<0.05, n=12 mice), but remained constant on a high Na+ diet (P>0.2, n=12 mice) (Figure 1A). On the high Na+ diet, there was no significant difference in urinary Ca2+ excretion between TK+/+ and TK−/− mice. Body weight, food intake, urinary creatinine as well as plasma values of Ca2+, creatinine and protein concentrations did not differ between mice genotypes on either diet (Table I).
TK is upregulated in TRPV5−/− mice
The amount of TK excreted in the urine was determined by trichloroacetic acid (TCA) protein precipitation of 24 h urine collected from WT (TRPV5+/+) and TRPV5−/− mice fed either a 0.02% w/w (low) or a 2% w/w (high) Ca2+ diet. Immunoblot analysis of the TCA precipitate showed the specific TK band migrating at ∼40 kDa (Figure 1B). Semiquantitative densitometry of the TK corresponding protein band demonstrated that urinary TK excretion increased in TRPV5−/− mice compared to TRPV5+/+ littermates on low Ca2+ diet, whereas the high Ca2+ diet did not affect TK excretion between TRPV5−/− and TRPV5+/+ genotypes (Figure 1C).
TK stimulates Ca2+ transport in renal primary cell cultures
The effect of TK on transcellular Ca2+ transport was evaluated in primary cultures of renal CNT/CCD cells. Application of 100 nM TK to the apical side of the monolayer significantly stimulated Ca2+ transport compared to nontreated monolayers, while no effect was observed when TK was added to the basolateral side only (Figure 2A). Importantly, apical addition of TRPV5 channel blocker ruthenium red (10 μM) completely inhibited baseline transepithelial Ca2+ transport and abolished the stimulatory effect of TK (Figure 2A). TK action was mimicked by apical addition of 100 nM trypsin, which like TK displays serine protease activity (Figure 2B), as well as 1 μM BK (Figure 2C). Interestingly, the stimulatory effect on Ca2+ reabsorption of both BK and TK was inhibited by apical application of 1 μM JE049, a B2R antagonist (Figure 2C). Basolateral treatment of trypsin, BK or JE049 did not affect transcellular Ca2+ transport across the monolayer (data not shown).
TK enhances TRPV5 channel activity
The stimulatory effect of TK on Ca2+ transport was further examined in TRPV5‐transfected human embryonic kidney (HEK) 293 cells. Incubation of these cells with TK concentrations ranging from 0.05 to 5000 nM enhanced 45Ca2+ influx in a dose‐dependent manner with a maximal stimulation at 500 nM and an EC50 of 6.00±0.04 nM (Figure 3A), whereas no effect was observed in mock‐transfected cells (data not shown). Further, TK action on TRPV5 channel activity was investigated by the whole‐cell patch‐clamp technique in TRPV5‐expressing HEK293 cells. Preincubation with 100 nM TK increased the inward Ca2+ current in response to the step protocol, but the Ca2+‐dependent inactivation of the TRPV5 currents remained unchanged (Figure 3B). The current–voltage (I–V) relation of the currents in response to a voltage ramp did not alter after preincubation with 100 nM TK (Figure 3C), but the inward Ca2+ current at −80 mV increased by 167±14% compared to non‐treated control (CTRL) cells (P<0.05, n=15 cells) as depicted in Figure 3D. Likewise, in divalent free (DVF) solution, a significant increase of 149±17% in Na+ inward currents at −80 mV (P<0.05, n=15 cells) was observed, while the I–V relation remained unchanged (Figure 3E and F).
TK increases TRPV5‐mediated currents via the B2R
Since BK mimicked the TK action on Ca2+ transport in primary cultures of CNT/CCD cells, the effect of BK on TRPV5‐expressing HEK293 cells was further studied. First, the expression of the B2R in HEK293 cells was confirmed by immunoblot analysis. B2R‐transfected HEK293 cells were used as a positive control. In both mock and B2R‐transfected HEK293 cells, the receptor was detected as a single immunopositive band of ∼69 kDa (Figure 4A), confirming the presence of the endogenous B2R in this cell line. Subsequently, TRPV5‐transfected HEK293 cells were incubated for 1 h with 1 μM BK and functionally characterized by patch‐clamp measurements in comparison to non‐treated (CTRL) cells. Interestingly, BK treatment resulted in a significant increase of in both Ca2+ (Figure 4D) and Na+ (Figure 4F) currents (P<0.05, n=12 cells) compared to CTRL cells, without affecting the Ca2+‐dependent inactivation (Figure 4B) or the I–V relation of the currents (Figure 4C and E). Because the different treatments of TRPV5‐expressing HEK293 cells affected both Ca2+ and Na+ current amplitudes leaving unaffected the other current properties, only the Ca2+ currents were depicted to demonstrate the effect of particular compounds. Incubation of TRPV5‐transfected cells with 1 μM JE049 reversed the stimulatory effect of TK on TRPV5‐mediated currents. The Ca2+‐dependent inactivation of Ca2+ currents remained unchanged for cells treated with TK or TK and JE049 (Figure 4G). Ca2+ current amplitudes measured from the step protocol were significantly lower, 285±50 pA/pF, for the JE049‐treated cells (P<0.05, n=11 cells) compared to TK‐treated cells (Figure 4H).
TK stimulates TRPV5‐mediated currents via the PLC‐dependent PKC pathway
To elucidate in detail the mechanism through which TK enhances TRPV5 activity, the following strategies were followed. First, pretreatment with 10 μM of the PLC inhibitor U73122 for 10 min abolished the TK stimulatory effect on TRPV5‐mediated Ca2+ currents, whereas 10 μM of its inactive analog U73343 had no effect (P<0.05, n=10 cells) (Figure 5A). Then, TRPV5‐transfected cells were incubated for 1 h with 10 μM 1‐oleoyl‐acetyl‐sn‐glycerol (OAG), a synthetic DAG analog. OAG mimicked the stimulatory effect of TK on TRPV5‐mediated Ca2+ currents (P<0.05, n=10 cells) (Figure 5B). The stimulatory effect of OAG was observed within 30 min after addition (Figure 5C). Next, the effect of TK on the sextuple PKC phosphorylation‐deficient mutant was tested. In HEK293 cells expressing this mutant, the stimulatory effect of TK was abolished (Figure 5D). Subsequently, the PKC phosphorylation sites were individually mutated into an alanine residue. TK increased Ca2+ currents of all single PKC mutants, except of S299A and S654A (P<0.05, n=12 cells) (Figure 5E). Furthermore, to determine which PKC isoforms are involved in this process, cells were incubated for 24 h with 1 μM of phorbol ester (phorbol 12‐myristate 13‐acetate (PMA)) to downregulate the PMA‐sensitive PKC isoforms expressed in HEK293 cells as previously described (Camden et al, 2005). Then, these PMA‐pretreated cells were treated as aforementioned with 100 nM TK for 1 h. TK was still able to significantly increase (164±14% of control, P<0.05, n=14 cells) the TRPV5‐mediated Ca2+ currents (Figure 5F). Vehicle‐incubated (NT) cells were used as control for the effect of PMA on TRPV5 currents and PMA incubation left TRPV5‐mediated currents unaffected.
Subsequently, the possible involvement of other G‐protein coupled receptors in addition to the B2R was investigated. To this end, purinergic receptors were activated by the addition of extracellullar ATP (1 mM for 1 h) to TRPV5‐expressing HEK293 cells. Indeed, as shown in the Figure 6, ATP treatment enhanced the inward Ca2+ currents in these cells, while leaving the I–V relationship and Ca2+‐dependent inactivation unchanged. Moreover, after pretreatment with the B2R antagonist JE049, ATP was still able to stimulate the TRPV5 currents, indirectly demonstrating the specificity of TK stimulation via the B2R.
TK increases cell surface expression of TRPV5 by delaying channel retrieval
The effect of TK on the amount of TRPV5 channel expressed at the plasma membrane was investigated in HEK293 cells. TRPV5‐transfected cells were incubated with 100 nM TK alone or in combination with 1 μM JE049 for 1 h and then subjected to cell surface biotinylation. Biotinylated cell lysates were precipitated with neutravidin‐agarose beads and immunoblotted for TRPV5. Mock‐transfected and non‐biotin‐treated cells were used as negative controls. TK treatment enhanced TRPV5 expression in the biotinylated fraction, whereas JE049 abolished this effect (Figure 7A, left panel). Importantly, TRPV5 was equally expressed in total cell lysates of all tested conditions (Figure 7A, right panel). The observed increase in cell surface expression of TRPV5 by TK could be due to either an enhanced trafficking from the Golgi apparatus to the cell surface or a reduction in channel retrieval from the plasma membrane. The kinetics of cell‐surface retrieval was measured in HEK293 cells using half‐life cell‐surface biotinylation. After a 1 h TK treatment, the retrieval of TRPV5 channel from the plasma membrane was decreased by 71, 46 and 17% (P<0.05, n=3 blots) at the time points of 1, 3 and 6 h, respectively (Figure 7B). TRPV5 expression in cell lysates was identical in all tested conditions (Figure 7C). These results suggested that TK increases cell surface expression of TRPV5 by delaying channel retrieval from the plasma membrane.
TRPV5 PKC mutants S299A and S654A are insensitive to TK treatment
The effect of TK on the TRPV5 PKC mutants S299A and S654A was investigated in HEK293 cells by half‐life analysis. TRPV5 WT, S299A and S654A transfected cells were incubated with or without 100 nM TK for 1 h and then the kinetics of cell‐surface retrieval was measured in HEK293 cells using half‐life cell‐surface biotinylation. Without TK treatment (Figure 8A), the plasma membrane expression of TRPV5 WT, S299A and S654A was decreased to a similar extent at the time points of 1, 3, 6 and 12 h. Interestingly, with TK treatment (Figure 8B), the retrieval of TRPV5 WT channel from the plasma membrane was decreased, but retrieval of S299A and S654A was not altered. TRPV5 WT, S299A and S654A expression in cell lysates was identical in all tested conditions (data not shown). These results indicated that the TK effect of delaying channel retrieval from the plasma membrane was abolished in TRPV5 PKC mutants S299A and S654A.
The present study showed that extracellular TK stimulates TRPV5‐mediated Ca2+ reabsorption by activating the B2R and subsequently the DAG/PKC pathway resulting in accumulation of TRPV5 channels at the cell surface. This conclusion is based on the following experimental observations: (i) Urinary Ca2+ excretion in mice is inversely related to the expression of TK; (ii) TK stimulates Ca2+ reabsorption in primary cultures of renal CNT/CCD cells, which can be blocked by the B2R antagonist JE049; (iii) the stimulatory effect of TK is mediated by the PLC/DAG/PKC pathway and requires the two conserved PKC sites S299 and S654 in TRPV5; and (iv) TK enhances Ca2+ transport by increasing TRPV5 abundance at the plasma membrane.
TK stimulated TRPV5‐mediated Ca2+ transport via the B2R receptor. The physiological relevance of this latter finding is substantiated by the co‐localization of B2R with TK (Figueroa et al, 1995) and TRPV5 (Hoenderop et al, 2003) in DCT and CNT. B2R belongs to the seven‐transmembrane domain G protein‐coupled receptor superfamily (Hess et al, 1992) and signals via Gαq protein with consequent activation of PLC. PLC catalyzes the hydrolysis of phosphatidylinositol‐4,5‐biphosphate (PIP2) in inositol trisphosphate (IP3) and DAG (Blaukat, 2003). Commonly, IP3 releases Ca2+ from the endoplasmic reticulum stores, while DAG activates PKC (Blaukat, 2003). Certainly, our results show that TRPV5 activation requires PIP2 breakdown following B2R stimulation and relies on DAG as a downstream effector. Subsequently, DAG increases TRPV5 activity via PKC phosphorylation of the channel since potential PKC phosphorylation deficient mutants of TRPV5 at positions S299 and S654 lack the TK effect. This is the first study demonstrating that TRPV5 is activated through its PKC phosphorylation sites. Among the six predicted phosphorylation sites in the TRPV5 sequence, two serines, S299 and S654, are critical for TRPV5 activation by TK. Importantly, TRPV5 closest homolog, TRPV6, did not respond to TK (data not shown). Since the second serine (S654) is not conserved in TRPV6, both serines are apparently critical for stimulation of TRPV5 by TK. In addition, 24‐h preincubation with PMA downregulates α, β and ε isoforms of PKC, but not PKC ζ (Frecker et al, 2005). This latter PKC isoform is presumably involved in TK increase of TRPV5 currents since the stimulatory effect of PKC was similar in PMA‐pretreated and control cells. Interestingly, stimulation of active Ca2+ transport in primary cultures of renal CNT/CCD also involves PMA‐insensitive PKC isotypes (Hoenderop et al, 1999) among which the isotype PKC ζ can be activated by B2R (Christopher et al, 2001). However, B2R knockout mice (B2R−/−) mice showed no change in urine Ca2+ excretion (Picard et al, 2005), in contrast with TK−/− mice, suggesting that compensation mechanisms could mask the TK effect on the Ca2+ balance in these knockout mice. For instance, it has been shown that in these knockout mice a compensatory induction of the B1 receptor occurs that could theoretically be a new target for TK (Duka et al, 2001). Alternatively, TK exerts its in vivo effect via a mechanism independent of the B2R.
Previous studies showed that TRP channels can be activated by different signaling molecules of the PLC pathway. For example, the three canonical TRP members TRPC3, 6 and 7 are characterized by their sensitivity to DAG (Hofmann et al, 1999). Alternatively, the vanilloid and melastatine members of TRP family, TRPV1 (Chuang et al, 2001), TRPV5 (Lee et al, 2005), TRPM4 (Nilius et al, 2006), TRPM5 (Liu and Liman, 2003), TRPM7 (Runnels et al, 2002) and TRPM8 (Liu and Qin, 2005), can be directly regulated by PIP2. Thus, in vivo TRPV5 activation by PIP2 (Lee et al, 2005) or DAG/PKC could adjust TRPV5 activity in response to physiological fluctuations. In addition, Ca2+ entering the cell through TRPV5 could prevent the electrostatic interaction between the negatively charged PIP2 and the channel by screening the negative charge on the lipid head group, as proposed for Mg2+ and TRPM7 (Kozak and Cahalan, 2003).
Stimulation of Ca2+ transport through TRPV5 upon TK action could be the result of an increase in either open probability of the channel or in expression of TRPV5 at the plasma membrane. Surface biotinylation analysis showed that TK increases the amount of TRPV5 channels at the plasma membrane. Thus, PKC activation of TRPV5 following the TK application would regulate the balance between constitutive exocytosis and endocytosis in favor of the former leading to the accumulation of TRPV5 at the cell surface. Similarly, the epidermal growth factor prevents the internalization of plasma membrane TRPC3 (Smyth et al, 2006). The role of the cytoskeleton in this translocation process is presently unknown. It is possible that PKC‐dependent phosphorylation of TRPV5 leads to activation of motor proteins that transport the channels towards the plasma membrane. This process could involve the FKBP52 protein, characterized previously as a TRPV5 regulatory protein (Gkika et al, 2006), since FKBP52 is known to interact with the motor protein dynein (Czar et al, 1995; Silverstein et al, 1999). Interestingly, TRPM7 associates to the actomyosin cytoskeleton upon BK stimulation regulating cell adhesion (Clark et al, 2006). Remarkably, the protein synaptotagmin has been proposed to regulate the exocytosis of TRPC5 (Strubing et al, 2001). However, the cytoskeletal elements and the motor proteins participating in the incorporation of TRPV5 in the plasma membrane remain to be identified. Furthermore, accumulation of channels at the cell surface can also occur by increased incorporation into the plasma membrane. Indeed, PKC potentiation of TRPV1 promotes the recruitment of a channel vesicular pool to the cell surface (Morenilla‐Palao et al, 2004). This exocytosis process of TRPV1 is dependent on the soluble N‐ethylmaleimide‐sensitive‐factor attachment proteins receptor (Morenilla‐Palao et al, 2004), known to act as membrane recognition molecules and acceptors for vesicle trafficking, docking and fusion (Duman and Forte, 2003). It would be interesting to investigate the role of similar scaffold proteins in the assembly of TRPV5 and PKC upon TK treatment, considering that such signaling pathways are currently indicated to function in spatially distinct microdomains (Ambudkar, 2006). For instance, the A‐kinase‐anchoring protein has been described to coordinate the subcellular localization of second messenger‐regulated enzymes, such as PKC in order to modulate the activity of K+ channel, KCNQ/M, upon agonist stimulation via Gq‐coupled pathway (Hoshi et al, 2005).
The aforementioned data are in support of plasma membrane recycling of TRPV5 protein as mechanism of channel regulation. A tight control of TRPV5 activity is of primordial importance for body Ca2+ homeostasis, since TRPV5 constitutes the fine‐tuned Ca2+ entry step in active Ca2+ reabsorption (Hoenderop et al, 2005). Besides the hormonal regulation of TRPV5, accessory proteins play a role in modulating channel trafficking and activity (van de Graaf et al, 2003; Gkika et al, 2004). Together with a recent report on the β‐glucuronidase klotho (Chang et al, 2005), our study introduces a new mechanism of TRPV5 regulation, which is based on extracellular enzymatic activation. TK and klotho co‐localize with TRPV5 in the distal part of the nephron where they activate the channel from the luminal side. A comparable enzymatic regulation is described for the epithelial Na+ channel, ENaC, present in CNT, by the channel activating proteases, CAP‐1, CAP‐2 and CAP‐3 (Rossier, 2004). Unlike the protease CAP and the β‐glucuronidase klotho, TK stimulates TRPV5 indirectly via activation of the B2R that then induces a redistribution of TRPV5 channels towards the plasma membrane. On the contrary, CAPs, which are membrane‐bound serine proteases, act directly on the channel gating by enhancing the open probability of ENaC (Vuagniaux et al, 2002) to stimulate epithelial Na+ absorption. Likewise, klotho directly activates TRPV5 by enzymatic modification of the N‐glycan to stabilize TRPV5 channels in the plasma membrane (Chang et al, 2005).
Interestingly, we showed that an increase in Na+ supply induced a Ca2+ wasting in WT mice. It is well know that a high Na+ diet increases the plasma volume that triggers the renin–angiotensin–aldosterone system (Nijenhuis et al, 2005). These hormones will reduce the reabsorption of Na+ and consequently Ca2+ in the proximal tubules as a negative feedback response. Ultimately, this will increase the urinary excretion of Ca2+. However, the anticipated calciuria was not observed in TK−/− mice, possibly these mice cannot further increase their urinary Ca2+ excretion. In addition, the rates of urinary Ca2+ excretion in WT and TK−/− mice were similar on a high Na+ diet, whereas they markedly differed on a normal Na+ diet. This strongly supports the fact that TK participates in the renal response to a high Na+ diet and to the attendant decrease in urinary Ca2+ reabsorption.
Here, we demonstrated that TK displays autocrine or paracrine stimulation of active Ca2+ reabsorption in a dose‐dependent manner. As shown in the mice models, reduction in TK expression increases urinary Ca2+ excretion due to impaired tubular Ca2+ reabsorption. Conversely, genetic ablation of TRPV5 doubled the renal expression of TK possibly due to counterbalance the significant Ca2+ wasting observed in the TRPV5−/− mice by maximally stimulating the process of Ca2+ reabsorption. Indeed, an increase in dietary Ca2+ intake as an alternative maneuver to offset the renal Ca2+ loss in TRPV5−/− mice prevented the compensatory increase in TK expression. These results underline the role of TK during Ca2+ restriction. It is, therefore, interesting to study the Ca2+ excretion in humans with a genetically reduced TK activity. In this respect, a loss‐of‐function polymorphism of the TK gene in which the active‐site arginine at position 53 is changed into a histidine (R53H), reduces by 50–60% TK activity is interesting (Slim et al, 2002). Partial genetic deficiency in TK activity of the R53H subjects is associated with a form of arterial dysfunction (Azizi et al, 2005). TK is well known as a vasodilator factor of renal cortical blood vessels through BK production, while abnormalities of TK levels has long been documented in the pathogenesis of hypertension (Chao and Chao, 2005). It is therefore tempting to speculate that TK stimulation of TRPV5‐mediated Ca2+ reabsorption plays a role in TK hypotensive action, since urinary Ca2+ leak has been correlated with higher baseline systolic and diastolic blood pressures (McCarron and Reusser, 1999).
In conclusion, our data demonstrated that TK reduces urinary Ca2+ excretion by an autocrine and paracrine stimulation of TRPV5‐mediated Ca2+ reabsorption explaining the hypercalciuria in TK−/− mice. Figure 9 illustrates the molecular pathway linking TK to stimulation of TRPV5, as elucidated in the present study. TK activates the B2R and through the PLC derived‐messenger DAG initiates the phosphorylation of the TRPV5 PKC sites S299 and S654. Subsequently, Ca2+ influx through TRPV5 is enhanced by the accumulation and stabilization of the channel at the plasma membrane. Thus, TK‐directed translocation of TRPV5 channels constitutes a mechanism by which renal cells can fine‐tune the Ca2+ reabsorption.
Materials and methods
TK−/− mice were produced as previously described (Meneton et al, 2001; Picard et al, 2005) and fed ad libitum two Na+ diets containing 0.3% w/w or 3% w/w for 14 days. Subsequently, 24 h urine was collected from TK−/− and TK+/+ mice housed in metabolic cages and blood was obtained by orbital puncture. Measurements of biological parameters in the animals plasma and urine were performed as previously described (Picard et al, 2005). Urinary Na+, Ca2+ and Mg2+ excretion was expressed as ratios to urinary creatinine excretion to take into account the variations in urine collection. TRPV5−/− mice were generated as described previously (Hoenderop et al, 2003). At the age of 4 weeks, mice were fed ad libitum two diets containing 0.02% w/w or 2% w/w Ca2+ for 5 weeks and were subsequently placed in metabolic cages. At 24 h urine samples were collected and subjected to TCA precipitation. The animal ethics board of the Radboud University Nijmegen approved all animal experimental procedures.
Urine protein precipitation by TCA
Of the 24 h urine volume collected of TRPV5−/− mice, 10% v/v was centrifuged at 200 g for 10 min at 4°C to remove cell debris. Proteins in the urine were precipitated using ice‐cold TCA (Acros organics, NJ, USA). The TCA mixture was centrifuged at 13 000 g for 10 min at 4°C, and the pellet was subsequently washed with 300 μl acetone. Next, the pellet was air‐dried for approximately 3 min, dissolved in Laemmli buffer containing 0.1 M DTT and 150 mM Tris (pH 8.8), and analyzed by immunoblot analysis for the expression of TK using a rabbit anti‐TK antibody (Calbiochem, San Diego, CA, USA). Immunopositive bands were scanned using an imaging densitometer to determine pixel density (Molecular Analyst Software, BioRad Laboratories, Hercules, CA).
DNA constructs and cell culture
The TRPV5 pCINeo/IRES‐GFP constructs were generated as described previously (van de Graaf et al, 2003). Single and combined PKC mutants were generated by alanine substitution of the six putative phosphorylation sites of TRPV5 (S144A, S299A, S316A, S654A, S664A, S698A) using in vitro mutagenesis (QuickChange Site‐Directed Mutagenesis kit, Stratagene, La Jolla, CA, USA). The B2R pcDNA3 was a kind gift from Professor G Erdös, MD (Department of Pharmacology, University of Illinois, Chicago, USA). HEK293 cells were transfected at 70% of confluency using polyethylenimine (Polysciences, Inc., Warrington, USA) or Lipofectamin 2000 (Invitrogen Life Technologies, Breda, The Netherlands). After 48–60 h, cells were used for 45Ca2+ uptake assays, patch‐clamp and/or biotinylation experiments. Prior to the assays, cells were incubated for 1 h in serum‐free medium containing the particular compound.
Transcellular Ca2+ transport in renal primary cultures
CNT/CCD tubules were immunodissected from kidney cortex of New Zealand White rabbits (∼0.5 kg) using antibody R2G9 and then placed in primary culture on permeable filters (0.33 cm2; Costar, Cambridge, MA, USA) as described before (Bindels et al, 1991). At confluence, monolayers were used for the transepithelial Ca2+ transport assay as previously described (Bindels et al, 1991). Transepithelial potential difference and resistance were checked before and after transport measurement to confirm the integrity of the monolayer.
45Ca2+ uptake assay
HEK293 cells were transfected with TRPV5 pCINeo/IRES‐GFP or pCINeo/IRES‐GFP cDNA. Ca2+ uptake was determined in uptake medium (110 mM NaCl, 5 mM KCl, 1.2 mM MgCl2, 0.1 mM CaCl2, 10 mM Na‐acetate, 2 mM NaH2PO4, 20 mM HEPES–Tris, pH 7.4 supplemented with 10 μM felodipine, 10 μM methoxy‐verapamil, 1 mM BaCl2 and 1 μCi/ml 45CaCl2) for 10 min at room temperature (20–25°C). Each well was washed extensively with stop buffer (110 mM NaCl, 5 mM KCl, 1.2 mM MgCl2, 0.5 mM CaCl2, 1.5 mM LaCl3, 10 mM Na‐acetate, 20 mM HEPES–Tris, pH 7.4) at 4°C, incubated with 0.05% w/v SDS and the lysates were counted for radioactivity using liquid scintillation.
Electrophysiology and solutions
Patch‐clamp experiments were performed as described previously (Vennekens et al, 2000) in the tight seal whole‐cell configuration at room temperature using an EPC‐9 patch‐clamp amplifier computer controlled by the Pulse software (HEKA Elektronik, Lambrecht, Germany). Two voltage protocols were used: a ramp, to establish the I–V relation in nominally DVF or in 10 mM Ca2+‐containing extracellular solutions, and a hyperpolarizing step protocol to measure the Ca2+‐dependent inactivation. Na+ current densities were calculated from the current at −80 mV during the ramp protocols, while the Ca2+ currents were calculated from the current at −80 mV during the ramp protocols or from Ca2+ peak values were extracted from the current at −100 mV during the step protocol. The analysis and display of patch‐clamp data were performed using Igor Pro software (WaveMetrics, Lake Oswego, USA).
Cell surface labeling with biotin
HEK293 cells were transfected with 15 μg HA‐TRPV5, TRPV5‐S299A or TRPV5‐S654A pCINeo/IRES‐GFP or pCINeo/IRES‐GFP in poly‐l‐lysine (Sigma, St Louis, MO, USA) coated 10 cm dishes. At 48 h after transfection, cells were incubated for 1 h with 100 nM TK and 1 μM JE049. The biotinylation assay was performed, cells were homogenized in 1 ml lysis buffer as described previously (Chang et al, 2005) using the NHS‐LC‐LC‐biotin (Pierce, Etten‐Leur, The Netherlands). Finally, biotinylated proteins were precipitated using neutravidin‐agarose beads (Pierce). TRPV5 expression was analyzed by immunoblot for the precipitates (plasma membrane fraction) and for the total cell lysates using the guinea‐pig anti‐TRPV5 antibody (Hoenderop et al, 2000). For the half‐life assay, the biotinylation assay was pursued as described above. For time point 0 h, cells were collected from plates and lysed, immediately after biotinylation. Other plates of cells were further cultured after biotinylation for 1, 3, 6 or 12 h, then washed once with ice‐cold PBS (pH 7.4), and subsequently homogenized in lysis buffer. All samples were processed as described above.
TK, BK, PMA and OAG were purchased from (Sigma, St Louis, MO, USA). U73122 and U73343 were purchased from Upjohn Laboratories (Kalamazoo, MI, USA). JE049, formerly known as icatibant or HOE140, which was a kind gift of Dr J Pünter (Aventis Pharma Deutschland GmbH, Frankfurt, Germany).
In all experiments, the data are expressed as mean±s.e.m. Overall statistical significance was determined by analysis of variance (ANOVA). In case of significance, differences between the means of two groups were analyzed by unpaired t‐test, while multiple comparisons between groups were performed by Bonferroni post hoc tests. P<0.05 was considered significant. The statistical analyses were performed using the SPSS software (SPSS Inc., Chicago, IL, USA).
We thank Mrs A van der Kemp, Mr D van den Berg and Mr R Janssen for expert technical assistance. This work was supported by the Dutch Organization of Scientific Research (Zon‐Mw 016.006.001, Zon‐Mw 902.18.298, NWO‐ALW 810.38.004, NWO‐ALW 805.09.042), Dutch Kidney Foundation (C03.6017), the European Molecular Biology Organization (long‐term fellowship ALTF 727‐2005) and the Human Frontiers Science Program (RGP0032/2004).
- Copyright © 2006 European Molecular Biology Organization