Calciotropic and Magnesiotropic TRP Channels

Joost G. J. Hoenderop, René J. M. Bindels

Abstract

Significant progress has been made into our understanding of the molecular mechanisms responsible for Ca2+ and Mg2+ homeostasis. Members of the transient receptor potential channel (TRP) superfamily proved essential to the maintenance of divalent cation levels by regulating their absorption from renal and intestinal lumina. This review highlights the molecular and functional aspects of these new calciotropic and magnesiotropic TRPs in health and disease.

Physiology of Ca2+ and Mg2+ Homeostasis

Ca2+ and Mg2+ are of central importance for multiple essential physiological functions, not the least of which include muscle contraction, neuronal excitability, bone formation, and enzymatic activity (25). Consequently, their concentration throughout the body is tightly regulated by an efficient feedback control system through the concerted action of kidney, intestine, and bone. Specifically, they are absorbed by the gastrointestinal tract and filtered and reabsorbed in the kidney, whereas the bone acts as a dynamic storage compartment, releasing divalent cations into blood when needed, i.e., on deprivation. Ca2+ and Mg2+ (re)absorption by these organs is mediated by paracellular as well as transcellular transport processes. The paracellular component of epithelial Ca2+ and Mg2+ transport is passive, directly connecting the luminal compartment with the blood compartment. By contrast, transcellular transport is active, involving the passage of two plasma membrane barriers (FIGURE 1). The latter is specifically regulated by the combined action of various calciotropic and magnesiotropic hormones.

FIGURE 1.

Transcellular Ca2+ and Mg2+ transport across renal and intestinal epithelia
A: luminal Ca2+ influx is mediated by TRPV5 in the renal distal convoluted and connecting tubules and TRPV6 in the duodenum. Subsequently, Ca2+ is transported across the cell to the basolateral membrane by Ca2+-binding proteins calbindin-D28K (kidney) and calbindin-D9K (intestine). Extrusion into the blood takes place via the plasma membrane Ca2+-ATPase (PMCA1b) in the intestine and Na/Ca2+ exchanger (NCX1) and PMCA1b in the kidney. B: luminal Mg2+ influx is mediated by TRPM6 in the renal distal convoluted tubules and the intestine. Subsequently, Mg2+ is transported across the cell to the basolateral membrane and extruded into the blood. The molecular identity of the latter players remains unknown at present. In this way, there is net Ca2+ and Mg2+ (re)absorption from the luminal space to the extracellular compartment.

Recently, significant progress has been made into our understanding of the molecular mechanisms responsible for transepithelial transport of divalent cations. This process can be divided into three consecutive steps. First, the ions enter the epithelial cell from the luminal compartment via specific influx proteins. Second, Ca2+ and putatively Mg2+ are bound to specialized intracellular carrier proteins that diffuse to the basolateral membrane. Third, the ions are transported into the interstitial fluid by extrusion mechanisms, including ion exchangers and pumps (FIGURE 1). Importantly, the polarized distribution of the influx and efflux mechanism(s) ensures their net transport from the luminal to the basolateral compartment.

There is ample evidence that cytosolic diffusion of Ca2+ ions inside of epithelial cells is mediated by specific Ca2+-binding proteins, known as calbindins. These proteins greatly enhance the diffusion rate of Ca2+. Two separate proteins have been described to date: calbindin-D28K, which is primarily present in renal epithelia, and calbindin-D9K, which is predominantly located in the intestine (20). Specific Mg2+-binding proteins have not been identified thus far, although the calbindins have been shown to bind Mg2+ as well. The physiological relevance of this is uncertain since the cytosolic Mg2+ concentration is typically in the submillimolar range, questioning the need for specific binding proteins to enhance the cytosolic diffusion rate of Mg2+ (1). Once at the basolateral membrane, Ca2+ is extruded into the blood by the Na+/Ca2+ exchanger (NCX1) and the plasma membrane Ca2+-ATPase (PMCA1b). Both these proteins are exclusively localized to the basolateral membrane of epithelial cells (FIGURE 1) (25). The characteristics of potential Mg2+ extrusion proteins are lacking to date, and new initiatives should be focussed in this direction. Regulation of active transcellular absorption occurs predominantly at the luminal membrane. The tightly regulated, apically expressed Ca2+ and Mg2+ channels have been identified in the last decade through expressional cloning and gene linkage analysis, respectively (26, 55, 61, 82). The identified channels belong to the superfamily of the transient receptor potential channels (TRPs) and are the topic of this review.

Introduction to Calciotropic and Magnesiotropic TRP Channels

Mammalian TRP channels are cation-permeable channels that have been subgrouped into six categories on the basis of sequence homology (i.e., TRPC, TRPV, TRPM, TRPA, TRPP, and TRPML) (7, 49). They are among the largest family of ion channels known, with representative members in a diverse range of species from yeast to humans. TRP channels are involved in a variety of physiological processes ranging from sensory functions to smooth muscle proliferation, endothelial permeability, gender-specific behavior, and epithelial divalent ion transport (25, 49, 64, 77). It is generally postulated that TRP channels span the membrane six times, have a hydrophobic pore loop between transmembrane regions 5 and 6, and contain large intracellular amino- and carboxyl-termini (FIGURE 2, A AND B). TRPs typically assemble into (hetero)tetramers to form a functional cationic channel, which permits the influx of Na+, Ca 2+, or Mg2+ ions. The stoichiometry and composition of the heterotetramer is determined by the ability of homologous members to interact with each other (21).

FIGURE 2.

Characteristics of calciotropic and magnesiotropic TRP channels
The proposed topology of the TRP channel includes six transmembrane (TM)-spanning domains and a pore-forming unit between TM5 and TM6. A: proposed structure of calciotropic TRPV5 and TRPV6 channels, including regulatory domains. Ankyrin repeats present in the amino-terminus of TRPV5 and TRPV6 are depicted with ANK. TRPV5 contains two protein kinase C (indicated by C) phosphorylation sites that are important for tissue-kallikrein-stimulated channel activity. In addition, the channel is N-glycosylated at position N358 (numbering of rabbit species) between TM1 en TM2. B:proposed structure of the magnesiotropic TRPM6 channel. TRPM6 is a protein that is both an ion channel and a kinase. The fused kinase domain at the carboxyl-terminus belongs to the α-kinase. The kinase is capable of phosphorylating itself and other potential substrates. The kinase activity is dependent on the intracellular ATP concentration that can bind the kinase domain. C:channel regulation can take place at various levels in the epithelial cell. First, the abundance of TRP channels is regulated by transcriptional and translational processes in the epithelial cell (1). The cell surface expression of TRPs is controlled in a dynamic way by insertion and retrieval pathways adjusting the requested functional activity at the plasma membrane (2). Finally, these TRP channels are gated at the cell surface by intracellular Ca2+ and Mg2+, providing a negative feedback regulation (3).

The Calciotropic TRP Channels

Pioneering work from the laboratories of Bindels and Hediger identified two closely related TRP channels, categorized into the TRPV subgroup, that are responsible for Ca2+ reabsorption in kidney (TRPV5) and intestine (TRPV6) (26, 55). These homologous channels are composed of ~730 amino acids (FIGURE 2A). The corresponding genes consist of 15 exons juxtaposed on human chromosome 7q35. They contain amino-terminal ankyrin repeats implicated in their assembly into functional tetrameric channels (4, 13). TRPV5 and TRPV6 are the most Ca2+-selective channels within the TRP superfamily. The permeability ratio of PCa:PNa calculated from the reversal potential at physiological Ca2+ concentrations is >100 (78). This unique permeation property is not conserved in the other TRPV members, including those with the highest homology to TRPV5 and TRPV6 (25). The molecular determinant of their Ca2+ selectivity and permeation is a single aspartate residue (TRPV5D542 and TRPV6D541, respectively) present in the pore-forming region (51, 80). The pore diameter of TRPV5 and TRPV6 is ~5.4 A (80), which is consistent with the pore size of other highly selective voltage-operated Ca2+ channels (3). Mutating TRPV6D541, a residue implicit to high-affinity binding of Ca2+, alters the apparent pore diameter (80). The external entrance in TRPV5 and TRPV6 may build up the three structural domains consisting of a coiled structure that is connected to a short pore helix followed by the selectivity filter (with D542 and D541 forming the narrowest part) and another coiled structure before the beginning of TM6, representing the first structural models of a TRP channel pore.

TRPV5 is predominantly expressed in the distal convoluted (DCT) and connecting (CNT) tubule of the kidney. These nephron segments are known to be involved in transcellular Ca2+ reabsorption. Here, the channel co-localizes with the previously described Ca2+ transporters, calbindin-D28K, NCX1, and PMCA1b. TRPV5 is also expressed, albeit at a lower level, in osteoclasts where it contributes to Ca2+ resorption (Table 1) (76). In contrast to TRPV5, TRPV6, is more ubiquitously expressed with significant expression levels in small intestine, mammary gland, and also prostate (52). Intestinal TRPV6 expression is mainly restricted to proximal duodenum where it is co-expressed with calbindin-D9K and PMCA1b (FIGURE 1) (25, 53).

View this table:
Table 1.

Localization of TRPV5 and TRPV6 in Ca2+-transporting tissues

Lessons from TRPV5 and TRPV6 Knockout Mice Models

Generation and characterization of TRPV5 and TRPV6 knockout mice elaborated the physiological role of these TRP channels. TRPV5 knockout (TRPV5−/−) mice exhibit a significant calciuresis; approximately sixfold more Ca2+ is excreted in their urine compared with wild-type littermates (27). Active Ca2+ reabsorption from DCT and CNT is impaired in these animals, confirming that apical Ca2+ reabsorption is mediated by TRPV5 in these nephron segments. Surprisingly, the knockout mice are normocalcemic, which is achieved through significant elevation in levels of bioactive vitamin D [1,25(OH)2D3] compared with control animals. This results in increased Ca2+ absorption from the small intestine of TRPV5−/− mice. Ultimately, this leads to a severe bone phenotype with low bone mineral densities. TRPV5 is predominantly localized to the ruffled border membrane of the osteoclast, and is not expressed in osteoblasts (76). TRPV5−/− mice show increased osteoclast numbers and osteoclast area, whereas the urinary bone resorption marker deoxypyridinoline is reduced compared with control mice. Bone resorption is diminished in osteoclast cultures from TRPV5−/− mice, supporting the impaired resorption observed in vivo. The contribution of each of these TRPV5-dependent processes, bone resorption and renal Ca2+ reabsorption, to overall Ca2+ homeostasis will be evaluated in the near future by the analysis of osteoclast and kidney-specific TRPV5 knockout mouse models.

A disturbed Ca2+ homeostasis was also observed in mice lacking TRPV6 (TRPV6−/−). They display defective intestinal Ca2+ absorption, low bone mineral density, decreased weight gain, and reduced fertility (2). Furthermore, TRPV6−/− mice display increased urinary Ca2+ excretion, although at low levels this Ca2+ channel is also expressed in the kidney, suggesting that it plays a role in renal Ca2+ handling (2). Finally, the skin of TRPV6−/− mice has fewer and thinner layers of stratum corneum, and ~20% percent of these animals develop alopecia and dermatitis (2). The applied TRPV6 knockout targeting strategy affected, in addition to the TRPV6 gene, also the closely adjacent EphB6 gene that might have contributed to the observed phenotype. Taken together, TRPV6−/− mice exhibit an array of abnormalities in multiple organs, reflecting an important role in Ca2+ homeostasis and beyond.

Pathophysiology of TRPV5 and TRPV6

Numerous clinical disorders and side effects from commonly prescribed medication cause electrolyte disturbances, resulting in the dysregulation of whole body Ca2+ balance. TRPV5 and TRPV6 have been suggested to contribute to the pathogenesis of altered Ca2+ homeostasis observed in both kidney diseases and from drug treatment therapies (29, 64). Because the abnormalities observed in TRPV5−/− and TRPV6−/− mice are also reported in patients with idiopathic hypercalciuria and/or Ca2+ malabsorption, TRP channel dysfunction may contribute to the pathogenesis of such disorders in humans. Despite several efforts, mutations in these channels have, however, not yet been identified (43). Commonly prescribed medications, including diuretics, immunosuppressant agents, and vitamin D analogs, may cause TRP channel-associated mineral (dys)regulation. Immunosuppressant drugs, such as tacrolimus and the glucocorticoids dexamethasone and prednisolone, are prescribed for a wide array of renal diseases and themselves affect the expression of TRPV5 and TRPV6 (Table 2). In addition, acid-base status affects the handling of Ca2+ (47). Chronic metabolic acidosis, a complication of renal failure, distal renal tubular acidosis, and chronic diarrhea are associated with increased renal Ca2+ excretion. An acid load in mice is accompanied by decreased renal TRPV5 expression, explaining the calciuresis observed under these conditions. A detailed overview of the (dys)regulation of TRPV5 and TRPV6 under various circumstances is summarized in Table 2.

View this table:
Table 2.

Regulation of calciotropic and magnesiotropic TRP channels

The Magnesiotropic TRP Channel

TRPM6 was identified as the molecule central to transcellular Mg2+ transport by two independent research groups, who isolated the causative genetic mutation from patients with hypomagnesemia with secondary hypocalcemia (HSH) (61, 82). TRPM6 contains a unique carboxyterminal serine/threonine protein kinase domain, homologous to the α-kinase family (FIGURE 2B) (11, 58). This distinct combination suggests intriguing possibilities for the regulation of either the channel or kinase activity. TRPM6 is a large protein, ~2,000 amino acids, encoded by a gene containing 39 exons (60). The channel is highly homologous (~55% at the protein level) to the previously characterized TRPM7 channel that has been implicated in cellular Mg2+ homeostasis (62). In kidney, TRPM6 is predominantly expressed along the apical membrane of DCT cells, the segment known to be responsible for transepithelial Mg2+ transport (81). Maintenance of total body Mg2+ homeostasis occurs principally within the kidney by tightly controlling renal Mg2+ excretion to parallel intestinal Mg2+ absorption. Approximately 20% of Mg2+ is reabsorbed by the proximal tubule. The majority of Mg2+ (50–70%) is, however, reabsorbed by the thick ascending limp of Henle’s loop. In both these segments, Mg2+ reabsorption occurs via paracellular transport. The DCT reabsorbs 5–10% of filtered Mg2+, and the reabsorption rate in this segment defines the final urinary Mg2+ concentration since virtually no reabsorption takes place beyond this segment (10). Expression profiling of TRPM6 indicates its presence, other than in kidney, also in colon, ceacum, and lung (18). TRPM6 is a Mg2+-permeable channel that is tightly regulated by intra-cellular levels of Mg2+, thereby providing a feedback mechanism to regulate its activity. This suggests that intracellular Mg2+ buffering capacity and Mg2+ extrusion mechanisms will strongly impact channel functioning and consequently transcellular transport. TRPM6 is permeable to all divalent cations tested (permeation rank order determined from the inward current amplitude at −80 mV is Ba2+ > Ni2+ > Mg2+ > Ca2+). It is thought that the luminal free Mg2+ concentration, in the Mg2+-absorbing part of the nephron, is in the range of 0.2–0.7 mM (FIGURE 1B) (9). To preferentially conduct Mg2+ in the presence of a greater extracellular concentration of Ca2+ (~1 mM), the luminal Mg2+ influx pathway must exhibit a higher affinity for Mg2+ than for Ca2+. Indeed, TRPM6 expressed in HEK293 has a four times higher affinity for Mg2+ compared with Ca2+ (81). This property, increased Mg2+ over Ca2+ permeability, is unique to TRPM6. All known Ca2+-permeable channels, including members of the TRP superfamily, display a 10–1,000 times lower affinity for Mg2+ than for Ca2+.

Pathophysiology of TRPM6

HSH is an autosomal recessive disorder that manifests in early infancy with generalized convulsions and symptoms of increased neuromuscular excitability, including muscle spasms and tetany (33). The pathophysiology of HSH is largely unknown, but physiological studies indicate a primary defect in intestinal Mg2+ transport. Previously, a gene locus for HSH was mapped to chromosome 9q22, the locus of TRPM6. Subsequently Konrad et al. and Sheffield et al. independently identified truncation and missense mutations in TRPM6 as the likely causative genetic defect in HSH (60, 61, 82). Consistent with their observations, the identified mutations in TRPM6 abolish channel activity (6, 81). Localization of TRPM6 to the DCT confirmed the earlier hypothesis of Cole and Quamme that renal Mg2+ wasting contributes significantly to the pathogenesis of HSH (8). This is further supported by intravenous Mg2+-loading tests in HSH patients, who exhibit a considerable renal Mg2+ leak during hypomagnesemia. As already discussed for disturbances in Ca2+ homeostasis, commonly prescribed medication used to treat renal and intestinal diseases affect whole body Mg2+ status through an alteration in the expression of TRPM6. An overview of our current view with respect to this regulation is summarized in Table 2.

Controlling TRP Channel Activation

Calciotropic and magnesiotropic TRP channels are integral membrane proteins that mediate the apical influx of Ca2+ and Mg2+. When open, these TRP channels, provide a diffusion pathway across the apical membrane, permitting the reabsorption of these solutes. Typically, they have a high throughput, allowing the flux of millions of ions per second and therefore require tight regulatory control of the channel’s open state. This positions the calciotropic and magnesiumtropic TRPs to be ideal targets for control of divalent cation reabsorption. Minute-to-minute whole body Ca2+ and Mg2+ status is adjusted by the regulation of these channels. The whole cell patch-clamp configuration enables the measurement of the current present at the plasma membrane. These currents (I) correspond to the formula I = i · N · P, where i represents the single-channel current (in pA), N is the number of ion channels in the plasma membrane, and P is the open probability of the channels. Trafficking of the channel into and out of the apical plasma membrane alters the number of surface-exposed channels (FIGURE 2C). Further fine tuning occurs through channel gating once inserted into the plasma membrane. The most basic aspect of channel regulation takes place by feedback control through the intracellular concentration Ca2+ and Mg2+ itself. Once the Ca2+ concentration in close vicinity to the channel pore of TRPV5 and TRPV6 reaches a threshold concentration, it acts as a negative feedback regulator, inactivating the channel at the cell surface (50, 78) (FIGURE 2C). Similar observations were made for intracellular Mg2+ with respect to TRPM6 channel activity. These TRPs rapidly inactivate during hyperpolarizing voltage steps in a Ca2+- or Mg2+-dependent fashion, respectively (78). In addition, TRP channels are regulated by a wide variety of physical and chemical factors. Recently, several members of the TRP channel family were reported to be regulated by phosphatidylinositol 4,5-bisphosphate (PIP2), including TRPV and TRPM channels (57). Future studies will hopefully address the physiological relevance of PIP2 in controlling calciotropic and magnesiotropic TRP channel activity.

Extracellular protons have a profound and diverse effect on TRP channel activity. Protons inhibit the calciotropic TRP channels by binding to glutamate-522 (E522) within the extracellular domain of the channel (84). E522 has been postulated to be an extracellular “pH sensor,” and its titration by extracellular protons reduces TRPV5 channel activity via conformational changes of the pore helix (84). In addition, the extra-cellular pH determines the cell surface expression of these TRP channels. An alkaline environment rapidly increases the number of channels at the plasma membrane, whereas an acidic milieu produces the opposite effect (36). Conversely, the magnesiotropic TRPs are stimulated by extracellular protons. Here, the small inward current of TRPM6 and TRPM7 is significantly enhanced by a decrease in extracellular pH. The ability of protons to potentiate inward currents in TRPM6 is lost in the E1024Q mutant, suggesting that this amino acid is critical to both Mg2+ permeability and its pH sensitivity (39).

Interacting Proteins

Several interesting regulatory proteins of these TRPs have now been described in detail (for an overview, see Table 2 and Ref. 71). Most of these regulators alter channel trafficking (FIGURE 2C). The first such interacting protein identified is the S100A10 protein. It forms a heterotetrameric complex with annexin 2 and associates specifically with the conserved sequence (VATTV) located in the carboxyl-terminal tail of TRPV5 and also TRPV6 (73). This associated S100 protein complex facilitates the movement of the calciotropic TRPs toward the luminal plasma membrane. Subsequently, other interacting protein candidates have been pulled out of yeast-two hybrid analysis utilizing kidney cDNA libraries and confirmed by pull-down assays using renal cell lysates. NHERF2, rab11a, and S100A10 regulate the transfer of TRPV5 and/or TRPV6 toward the plasma membrane, whereas the EF-hand-containing proteins calmodulin, calbindin-D28K, and 80K-H are thought to be regulatory factors of channel activity at the plasma membrane. In addition, other interacting partners such as FKBP52, NHERF2 and 4, and BSPRY have a significant impact on channel activity; however, their molecular mechanism of action is not yet known (Table 2). Future experiments should address in detail the cell biological aspects of vesicular TRPV5/6 transport and the regulation of channel gating kinetics by these and other associated proteins. To date, there is no information with respect to regulatory proteins for the magnesiotropic channel TRPM6.

“Extracellular protons have a profound and diverse effect on TRP channel activity."

Novel Calciotropic and Magnesiotropic Hormones

Our understanding of the calciotropic and magnesiotropic hormones has evolved over the last years (Table 2). Ours and other laboratories have demonstrated stimulatory effects of the classical calciotropic hormones, vitamin D and parathyroid hormone (PTH), on transepithelial Ca2+ transport. In vitro and in vivo studies show that 1,25-dihydroxivitamin D3 [1,25(OH)2D3] stimulates Ca2+ (re)absorption from the small intestine and renal DCT and CNT cells. The mechanism of action is via an upregulation of mRNAs encoding TRPV6, TRPV5, the calbindins, and NCX1, which translates into increased Ca2+ (re)absorption. More recently, new hormones have been identified that regulate the Ca2+ balance, including the female sex hormone estrogen (67). Estrogen deficiency causes a negative Ca2+ balance and bone loss in post-menopausal women. Estrogen increases renal expression of TRPV5 in a 1,25(OH)2D3-independent manner (67). This was confirmed in ovariectomized vitamin D-deficient mice where 17β-estradiol replacement therapy elevated renal TRPV5 expression, leading to the normalization of serum Ca2+ levels. By altering TRPV5 expression, the rate-limiting step in transcellular Ca2+ transport, estrogen might positively regulate Ca2+ reabsorption and homeostasis (see Table 1 for complete overview).

Recently, the family of the calciotropic hormones has been expanded to include a new member, the anti-aging hormone klotho. Klotho is a type I trans-membrane protein with β-glucuronidase activity. This activity is contained in the extracellular domain that can be secreted into the blood, urine, and cerebrospinal fluid. Klotho-deficient mice suffer from premature aging and display altered Ca2+ homeostasis resulting in osteopenia (34). The nephrology community has shown significant interest in klotho due to its predominant renal expression, its co-localization with TRPV5 in the DCT, and its interaction with fibroblast growth factor-23 to regulate phosphate homeostasis (66). Klotho modifies the N-glycan of TRPV5 via its β-glucuronidase activity, resulting in increased channel abundance at the plasma membrane (Table 2) (5). This is the first example of a calciotropic hormone operating from the urinary compartment to stimulate Ca2+ channel activity. Subsequently, Imura and coworkers further strengthen the role of klotho in Ca2+ homeostasis. They described an association of klotho with the ubiquitous Na+-K+-ATPase that augments the cell surface expression of this pump. This would increase the driving force for Ca2+ efflux, as Ca2+ is exchanged for Na+ down an increased concentration gradient, by the basolateral NCX1 primarily present in the DCT cell (31). Overall the combined action of klotho on the DCT enhances here the Ca2+ reabsorption.

Another hormone affecting Ca2+ balance is tissue kallikrein. This is the main mammalian kinin-forming enzyme that is produced by the kidney, where it co-localizes with TRPV5 (16). Other than being a major protein synthesized in the distal part of the nephron, this hormone is also abundantly secreted into tubular fluid. Tissue kallikrein displays serine protease activity that converts kininogen to kinin; this protein then acts by stimulating kinin receptors such as the bradykinin receptor type 2. The first observation that tissue kallikrein plays a role in Ca2+ homeostasis came from tissue kallikrein knockout mice, which display significant hypercalciuria (56). These findings suggest that tissue kallikrein deficiency disturbs renal tubular Ca2+ reabsorption. Experimental evidence indicates that tissue kallikrein directly activates TRPV5 by protein kinase C-mediated phosphorylation (FIGURE 2A). In a similar fashion to klotho, this factor enhances channel surface expression and consequently increased channel activity. Unraveling the molecular mechanisms of TRPV5 channel regulation by klotho, tissue kallikrein, and other extracellular factors has demonstrated the existence of novel regulatory mechanisms for active trancellular Ca2+ reabsorption that act from the pro-urine.

In the past, several hormones have been postulated to be significant players in the maintenance of Mg2+ homeostasis; however, convincing data of their effect on epithelial Mg2+ transport or TRPM6 activity is lacking. It has been shown that 17β-estradiol but not 1,25(OH)2D3 or parathyroid hormone regulates TRPM6 renal mRNA levels (18). TRPM6 expression in kidneys of ovariectomized rats is significantly reduced, whereas 17β-estradiol treatment normalizes TRPM6 mRNA levels. Groenestege et al. recently identified epidermal growth factor (EGF) as a new magnesiotropic hormone regulating renal handling of Mg2+ by unraveling the mechanism responsible for a rare inherited Mg2+ wasting disorder in a consanguineous kindred (19). Two affected sisters in a large Dutch family displayed increased urinary Mg2+ excretion, despite having low serum Mg2+ levels (0.53–0.66 mM), consistent with a disturbance in tubular reabsorption of Mg2+. The causative mutation was identified to be the pro-EGF gene, which was shown to be a novel magnesiotropic hormone (19). Pro-EGF exists as a membrane-bound molecule that is proteolytically cleaved to generate a 53-amino acid peptide hormone, EGF. The processed EGF is detectable in serum, cerebrospinal fluid, and urine, indicating that mature EGF might be important for maintaining Mg2+ balance. In kidney, pro-EGF is primarily expressed in the DCT where it is secreted from the apical and baso-lateral membranes (19). The EGFR is localized baso-laterally in most epithelial cells and can be found throughout the entire nephron with a predominant localization to the basolateral membrane of the DCT (14, 59). Groenestege et al. demonstrated that EGF stimulates TRPM6 through activation of the EGFR in the DCT cell. The mutation observed is a substitution of a highly conserved proline in pro-EGF that prevents the normal processing and secretion of EGF. As a consequence, the EGFR cannot be activated, and TRPM6-mediated Mg2+ influx is diminished, causing the renal Mg2+ wasting observed in isolated autosomal recessive renal hypomagnesemia (19).

These findings contribute to our increasing understanding of the kidney’s role in endocrinology as a production site of hormones. The calciotropic hormone 1,25(OH)2D3 is formed in the proximal tubule, whereas klotho and tissue kallikrein are produced in the DCT and the CNT and then secreted into the urine. Finally, EGF is synthesized in the DCT and released into the urine and blood stream to act as an autocrine/paracrine magnesiotropic hormone in the control of Mg2+ reabsorption.

Further Questions

It is apparent that the calciotropic and magnesiotropic TRP channel members are central to transepithelial Ca2+ and Mg2+ transport, thereby regulating whole body Ca2+ and Mg2+ balance. Although several studies have provided new information with regard to the regulation of these TRPs, the molecular mechanisms responsible for their trafficking and activation within epithelial cells is largely unknown. The kinetics of channel internalization and their fate once internalized is poorly understood. Other than TRPM6, the epithelial Mg2+ transporters that facilitate transcellular Mg2+ reabsorption remain to be identified. Channel interacting proteins that affect the activity of this magnesiotropic channel need to be elucidated. In addition, the role of the α-kinase domain in the carboxy terminus of TRPM6 is not well understood. Furthermore, other TRP channels could contribute to epithelial Ca2+ and Mg2+ transport like the nonselective cation channel TRPV4 that is highly expressed in kidney (65). Future studies should address these questions. This will provide a better understanding of the function and regulation of these calciotropic and magnesiotropic TRPs. In turn, this will confer greater insight into the maintenance of Ca2+ and Mg2+ homeostasis.

Acknowledgments

We thank Dr. T. Alexander for critical reading of this manuscript.

This work was financially supported in part by grants from the Dutch Kidney Foundation (C03.6017 and C06.2170), Human Frontiers Science Program (RGP32/2004), the Netherlands Organization for Scientific Research (Zon-Mw 016.006.001, Zon-MW 9120.6110, NWO-ALW 814.02.001, TOPCW 05.B.012), and the Dutch Stomach-Intestine-Liver Foundation (MWO 03-19). J. Hoenderop is supported by an European Young Investigator Award.

References

View Abstract