Approximately 65% of the calcium filtered by the kidney is reabsorbed along the segments of the PT, largely via passive paracellular transport. ^, Up to 10% of calcium absorption in the straight portion of the proximal tubule may occur by the transcellular pathway, via a voltage-gated calcium channel-dependent mechanism. PT calcium loss contributes to the development of hypercalciuria and kidney stone formation, ^, underscoring the importance of understanding the molecular pathways responsible for calcium reclamation in the PT.

The reabsorption of sodium and resulting reabsorption of water driven by the osmotic force generated are the primary mechanisms allowing the transport of calcium across the PT. Sodium entry into the PT is primarily driven via the sodium-hydrogen exchanger isoform 3 (NHE3), encoded by the SLC9A3 gene. This sodium transport is facilitated by an inward-directed sodium gradient, which is generated by sodium efflux from the cell by the basolaterally expressed Na ⁺ -K ⁺ -ATPase ( Fig. 7.5 ). The sodium and subsequent water reclamation facilitates passive paracellular calcium reabsorption by creating a calcium concentration gradient from the tubular lumen to the blood. Aquaporin-1 water channels are highly expressed in the apical and basolateral membranes of the PT. ^, Consequently, transcellular water reabsorption predominates in the PT (∼70%), but there is also considerable reabsorption of water via the paracellular pathway. ^, This paracellular flux may promote the process of solvent drag, which is the convective movement of calcium through the tight junction. ^, Calcium transport through the tight junction of the PT may be facilitated by either the calcium concentration gradient or solvent drag, and the relative contribution of each has not been established. Consistent with this, micropuncture data show a close relationship between sodium, fluid, and calcium reabsorption. ^,

Mechanisms by which calcium, magnesium, and phosphate are transported in the proximal tubule (PT).

Within the proximal tubule, the sodium proton exchanger isoform 3 (NHE3) facilitates the inward transport of sodium, while the sodium-potassium-ATPase (Na ⁺ -K ⁺ -ATPase) is responsible for the efflux of sodium. Additionally, the electrogenic sodium bicarbonate cotransporter (NBCe1) is responsible for the efflux of sodium and bicarbonate. Paracellular calcium reabsorption is facilitated by the reabsorption of sodium and water across the PT and occurs primarily through claudin-2 (CLDN2) and also claudin-12 (CLDN12). Paracellular magnesium permeation is less than calcium in the PT. The insert serves as a magnified representation of the tight junction, indicating the preferential permeability of calcium to magnesium in the PT. For the sake of clarity, paracellular sodium and water fluxes have been omitted. The process of phosphate uptake is facilitated by the SLC34A1 and SLC34A3 transporters, driven by the sodium gradient. SLC34A1 is retained on the brush border membrane by NHERF-1. Ezrin promotes the binding of NHERF-1 to actin. The xenotropic and polytropic retroviral receptor (Xpr1) may contribute in the process of phosphate extrusion at the basolateral membrane.

The paracellular reabsorption of divalent cations may also be impacted by the transepithelial voltage gradient across the PT. In the early PT, a small lumen negative potential difference of about–2 mV is generated by sodium-coupled glucose reabsorption. This transepithelial voltage gradient gradually turns positive along the length of the PT (∼ +2 mV), due to the preferential paracellular reabsorption of chloride. The relative ion selectivity of the proximal tubular tight junction pores also changes along the length of this segment. In the early PT, the pore is slightly cation selective, whereas in the late PT, the pore is anion selective. These changes may contribute to the shift in charge of the transepithelial voltage gradient along the PT. In the absence of experimental data, modeling studies have been completed on calcium transport across the PT, which suggest that the transepithelial voltage gradient has a marginal contribution to the transport of calcium across the segment.

The PT is a leaky epithelium with a transepithelial resistance of only 5 to 15 Ω · cm². ^, The permeability of the PT, as in other epithelia, is conferred by the claudin family. Claudins are four-pass transmembrane proteins. Their expression in the tight junction allows them to interact in both the same cell (cis-interaction) and between neighboring cells (trans-interaction). This facilitates the formation of pores or barriers between cells that determines the permeability of the paracellular shunt, thereby enabling the paracellular movement of specific ions. In adult animals, the PT expresses claudin (CLDN)-2, CLDN12, and CLDN10a, ^{, ,} with CLDN2 and CLDN12 contributing to the formation of a cation permeable pore, whereas CLDN10a forms an anion permeable pore. ^{, ,}

PT permeability to calcium is contributed by CLDN2 and CLDN12 (see Fig. 7.5 ). When overexpressed in cell culture, permeation of calcium is increased by CLDN2 in monolayers from both MDCK I, OK cells, and CaCo2 cells determined using either radioactive flux studies or bi-ionic diffusion potential measurements. Both sodium, calcium, and water permeate the paracellular shunt via CLDN2, ^, and water flux can be driven by sodium through this pathway. ^, It is not yet clear whether similar electrolyte and water flux can occur through CLDN12. Nevertheless, CLDN12 increases paracellular calcium permeability when expressed in CaCo2 cells, as determined by measurements of bi-ionic diffusion potential and radioactive calcium fluxes. ^,

Only the PT and initial part of the thin descending limb in long looped nephrons express CLDN2. Cldn2 -deficient mice exhibit both hypercalciuria and nephrocalcinosis. ^, Although Cldn12 -deficient mice do not exhibit hypercalciuria, ex vivo microperfusion of PTs from this strain demonstrates that CLDN12 confers permeability to both calcium and sodium in the PT as well. Data from Cldn2 and Cldn12 double knockout mice provide more evidence for the significance of CLDN12 in PT calcium transport. Compared with Cldn2 -deficient mice, these animals show exaggerated renal calcium wasting leading to hypocalcemia and decreased bone mineral density. ^, CLDN2 and CLDN12 form independent, partially redundant paracellular pores, each contributing to calcium permeability.

The most common metabolic abnormality associated with kidney stone disease is hypercalciuria. ^, Decreased calcium reabsorption from PT is a mechanism causing hypercalciuria in some cases. ^{, ,} In line with these findings is the observation that Slc9a3 null animals exhibit elevated urinary calcium excretion. Children with pathogenic variants in SLC9A3 experience sodium-losing diarrhea, but the excretion of calcium in urine has not been investigated. ^, Furthermore, single nucleotide polymorphisms (SNPs) in AQP1 correlate with stone disease in humans, underscoring the importance of the reclamation of water to the reabsorption of calcium in the PT. However, the precise mechanism by which these SNPs contribute to an increased risk of stone formation is currently unknown. Noncoding SNPs in the CLDN2 gene that associate with either increased or reduced expression of CLDN2 correlate with reduced or increased risk of kidney stone disease, respectively. Furthermore, in an Iranian family, kidney stones appear to be caused by coding pathogenic variants in CLDN2. SNPs in CLDN12 have not been linked to the development of kidney stones; however, this is not unexpected considering that Cldn12 knockout mice do not exhibit hypercalciuria.

As with calcium, magnesium is reabsorbed from the proximal tubule via the paracellular route (see Fig. 7.5 ). Detailed micropuncture studies demonstrate that the flux of magnesium across the PT increases in a nonsaturating manner as luminal magnesium concentrations rise. Furthermore, when punctured tubules are perfused with a magnesium-free solution, backflux of magnesium occurs. ^, However, in comparison with calcium, only up to 30% of the filtered magnesium is reclaimed from the PT in adult animals ^, (see Fig. 7.4 ). The kinetics of paracellular magnesium transport differ markedly from those of both sodium and calcium. In fact, the magnesium concentration in the lumen needs to rise markedly (TF/UF >1.9) for uptake of magnesium to occur. ^, This is in line with findings from flux studies, which suggested that the PT permeability to magnesium is substantially lower than that of calcium (1 × 10 ^–5 cm/s for magnesium, 7–27 × 10 ^–5 cm/s for calcium), although a direct comparison of permeabilities was not performed. It is of interest to note that younger animals display higher PT magnesium reabsorption than adults.

Since the two main cation-permeable claudins expressed in the adult PT are CLDN2 and CLDN12, these may allow magnesium permeation across the epithelium. Consistent with this, magnesium permeability was higher in MDCK cell monolayers when CLDN2 was overexpressed. However, Cldn2 -deficient mice or Cldn2 and Cldn12 double-deficient mice do not exhibit either hypomagnesemia or urinary magnesium wasting. ^, This may be explained by the comparatively small amount of magnesium transport in the PT and the ability of both the TAL and distal convolution to compensate PT losses. In fact, following deletion of Cldn10a, the junctional patches that would have expressed CLDN10a along the PT now become populated by CLDN2. These Cldn10a -deficient animals do display both lower urinary magnesium excretion and hypermagnesemia, as well as a higher relative permeability for magnesium in the S1 and S3 segments. Cldn6 and Cldn9 expression has been documented in the PT of younger animals with higher PT magnesium transport, though they are unlikely to confer permeability to magnesium since these claudins are barrier forming. Ultimately, the identity of the PT magnesium pore needs to be clearly delineated.

The PT is the main site of phosphate reabsorption, reclaiming at least 80% of the filtered load. There is significant variation in the transport of phosphate along the length of the PT. Detailed micropuncture measurements have determined that the majority of filtered phosphate is reclaimed in the convoluted portion of the PT, with the bulk of this in the early convoluted portion. ^, In the PT, phosphate influx is largely transcellular and mediated by the sodium-coupled phosphate transporters expressed in this segment. These include the sodium-dependent phosphate cotransporter 2A (NaPi-2a, SLC34A1) encoded by the SLC34A1 gene, the sodium-dependent phosphate cotransporter 2C (NaPi-2c, SLC34A3 ) encoded by the SLC34A3 gene, and PIT2 encoded by the SLC20A2 gene. SLC34A1 and SLC34A3 mediate most, if not all, of the phosphate reclamation in the PT (see Fig. 7.5 ), and a role for PIT2 remains to be established. Importantly, the SLC34 transporters are most abundantly expressed in the proximal convoluted tubule, where the highest reclamation of phosphate is observed. ^, SLC34A1 drives electrogenic sodium and phosphate cotransport ^, with a stoichiometry of 1 divalent phosphate ion to 3 sodium ions. In contrast, SLC34A3 is electroneutral and transports 1 divalent phosphate ion and 2 sodium ions.

SLC34A1 is the main isoform in mice, accounting for the majority of phosphate reclamation by the PT. ^{, ,} Renal-specific deletion of Slc34a3 in mice does not affect renal phosphate transport. This may result from SLC34A1 compensating for this loss. Consistent with this hypothesis, double Slc34a1- and Slc34a3 -deficient mice display significantly reduced plasma phosphate and elevated fractional phosphate excretion, in comparison with Slc34a1 -deficient animals. In humans, SLC34A3 may have a more prominent role, as inactivating pathogenic variants in SLC34A3 cause the rare disease hereditary hypophosphatemic rickets with hypercalciuria (HHRH), a syndrome with renal phosphate wasting and high 1,25-dihydroxyvitamin D levels, with the latter augmenting intestinal absorption of calcium and causing hypercalciuria. ^, Inactivating pathogenic variants in SLC34A1 can cause idiopathic infantile hypercalcemia, a rare childhood disease. Affected patients display renal phosphate wasting and hypophosphatemia at presentation. ^, In Slc34a1 -deficient mice, and potentially also patients, the hypercalcemia results from suppression of FGF23 due to lowered phosphate levels and subsequent enhanced synthesis of 1,25-dihydroxyvitamin D. SLC20A2 does not appear important for PT phosphate transport, despite being present in the apical membrane and its expression being altered in response to changes in dietary phosphate intake. As such, patients with pathogenic variants in the SLC20A2 gene exhibit idiopathic basal ganglia calcification but do not have alterations in phosphate homeostasis. ^,

A number of interactions between the SLC34A1 transporter and binding proteins have been described to occur via its C-terminal PDZ domain ^, including with the sodium/hydrogen exchanger regulatory factor (NHERF) family, which are expressed in or near the brush border membrane ^, (see Fig. 7.5 ). The inability to properly interact with NHERF1 via the PDZ domain impairs the proper expression of SLC34A1 in the brush border membrane, leading to disassembly, subsequent endocytosis of SLC34A1, and eventual degradation in lysosomes. This is exemplified in mice with target deletion of the Slc9a3r1 gene encoding NHERF1. These animals display a loss of SLC34A1 from the brush border, hypophosphatemia, and renal phosphate wasting. Furthermore, genetic variants found in the human NHERF1 gene are associated with nephrolithiasis or bone demineralization and were identified only in patients with low tubular phosphate reclamation. While NHERF3 shows a similar interaction, phosphate handling in the proximal tubule and SLC34A1 regulation seem marginally affected in Slc9a3r3- deficient mice. In contrast, NHERF3 may play a role in maintaining SLC34A3 in the brush border membrane. When wild-type and Slc9a3r3- deficient mice were maintained on a low-phosphate diet, wild-type animals displayed an increased SLC34A1 and SLC34A3 abundance at the brush border, while only SLC34A1 appeared to increase to the wild-type level in Slc9a3r3 deficient mice. Phosphate exits across the basolateral membrane. Although a potential contribution from the xenotropic and polytropic retroviral receptor 1 (XPR1) has been described, the basolateral efflux mechanisms for phosphate in the PT are still not fully delineated (see Fig. 7.5 ).

Available studies suggest that calcium and magnesium are reclaimed from the filtrate via the same paracellular pathway in the TAL, and therefore the transport of both electrolytes is discussed together in this section. Micropuncture studies have demonstrated that the TAL is responsible for reabsorbing up to 60% of magnesium filtered by the glomerulus and approximately 25% of filtered calcium ^{, ,} (see Fig. 7.4 ). This is not surprising given that reclamation of calcium is higher than for magnesium in the PT. Reclamation of calcium and magnesium in the TAL is largely paracellular. Some reports indicate the presence of a transcellular transport pathway for calcium in the TAL; however, the contribution of this to TAL calcium transport and the identity of the pathway are not well understood.

Paracellular calcium and magnesium reabsorption is passive and therefore dependent on the electrochemical gradients. This is illustrated in perfusion experiments of isolated segments from cortical TAL, where applying a lumen-negative voltage facilitates diffusion of divalent cations into the lumen of the tubule, in comparison with directional transport from lumen to bath under normal conditions. The majority of paracellular divalent cation transport takes place in the cortical TAL and outer stripe of outer medulla (OSOM), whereas limited calcium and magnesium reclamation occurs from the inner stripe of outer medulla (ISOM). ^, This is largely due to the permeability characteristics of the shunt, as is discussed in detail later.

The lumen-positive potential ranges from +5 to +13 mV in the early part of the medullary TAL and rises as the segment ascends such that in the cortical TAL the lumen-positive potential may reach up to +30 mV.

Divalent cation reclamation is highly dependent on the lumen-positive electrochemical gradient across the TAL. ^, The transepithelial voltage gradient is formed by the asymmetric transport of sodium, potassium, and chloride across the TAL epithelium. NKCC2, the furosemide-sensitive cotransporter, drives transport of sodium, potassium, and two chloride ions from the lumen into the epithelial cell, ^{, ,} with sodium being effluxed by the Na ⁺ -K ⁺ -ATPase at the basolateral surface. ROMK channels in the apical membrane allow the recycling of potassium back into the tubular lumen, ^, whereas CLC-K channels and the Barttin subunit, which is necessary for CLC-K channel function, allow the exit of chloride ions across the basolateral membrane (see Chapter 6 ) ( Fig. 7.6 ). Loop diuretics act by potently inhibiting NKCC2 transport, ^, which effectively collapses the voltage gradient and hence the transepithelial movement of divalent cations via the paracellular shunt. ^{, , ,}

Mechanisms by which calcium and magnesium are transported in the thick ascending limb (TAL).

In normocalcemic conditions, the TAL exhibits asymmetric electrolyte secretion, resulting in the generation of a lumen-positive voltage that facilitates the reabsorption of calcium and magnesium via the paracellular shunt. This requires NKCC2 (furosemide-sensitive sodium, potassium, 2 chloride cotransporter), the renal outer medulla potassium channel (ROMK), and the chloride channel Kb (ClC-Kb) with its subunit Barttin, and the sodium-potassium-ATPase (Na ⁺ -K ⁺ -ATPase). The interaction between CLDN16 and CLDN19 creates a cation-permeable pore that allows for both paracellular calcium and magnesium reabsorption. CLDN10 forms a monovalent cation pore driving sodium reabsorption; however, the transtubular potential difference can be further augmented by the backflux of sodium toward the top of the TAL.

Loop diuretics, such as furosemide, are primarily used in pediatric and adult patient populations for the treatment of edema. As they reduce the transepithelial voltage gradient in the TAL, they also limit the reabsorption of divalent cations. The results of this are observed in experimental animal models and humans treated with loop diuretics, which cause urinary calcium and magnesium wasting. ^,

Pathogenic variants in the SLC12A1 gene encoding NKCC2, the KCNJ1 gene encoding ROMK, the CLCNK2 gene encoding CLC-Kb (CLC-K2 in mice), or BSDN encoding Barttin cause various types of Bartter syndrome that strongly inhibit or abolish TAL electrolyte transport and hence the lumen-positive voltage critical for paracellular divalent cation reclamation. ^{, , ,} Hypotension, hypokalemic metabolic alkalosis, excessive renal urinary sodium and chloride excretion, and hyperreninemia are hallmark features of Bartter syndrome. The wasting of calcium and magnesium into the urine is also observed in Bartter syndrome, but the severity and affected divalent cation differ between genetic defects.

Pathogenic variants in SLC12A1 and KCNJ1 cause the antenatal forms of Bartter syndrome that present with polyhydramnios, which is frequently accompanied by premature birth. Furthermore, affected individuals experience severe hypercalciuria and early-onset nephrocalcinosis, while hypomagnesemia and hypermagnesuria are not commonly observed in antenatal Bartter syndrome. ^{, , ,} A newly identified transient form of antenatal Bartter syndrome caused by pathogenic variants in MAGED2 also presents with hypercalciuria. MAGED2 appears important for regulating NKCC2 expression, which leads to this antenatal Bartter form. However, the syndrome, including hypercalciuria, resolves spontaneously as the infant ages, though nephrocalcinosis may persist. Pathogenic variants in the CLCNKB gene cause the classical form of Bartter syndrome. Here, larger phenotypic variability is seen at presentation, with a number of less severe cases. As such, hypercalciuria and nephrocalcinosis are less frequently observed, whereas hypomagnesemia develops frequently. The presence of other efflux mechanisms enabling chloride exit from the TAL epithelial cell may explain why calcium wasting is not frequently encountered in patients with classical Bartter syndrome. Furthermore, the expression of CLC-Kb in the distal convoluted tubule (DCT) could influence the development of hypomagnesemia. ^, Clcnk2 -deficient mice do not display hypercalciuria consistent with the traditional Bartter phenotype, which may result from the retained expression of the CLC-K1 chloride channel (CLC-Ka in humans), although this is predominantly found in the medullary TAL. Increased urinary magnesium excretion was also found in these mice, which may be caused by reduced transcellular magnesium reabsorption from the DCT rather than the TAL since Clcnk2 deletion causes atrophy of the early DCT segment.

The TAL is a rather leaky epithelium with a transepithelial resistance ranging from 4 to 34 Ω · cm², ^, which is comparatively higher in the cortical segments. ^{, ,} Hence the transepithelial resistance found in the TAL is comparable to the PT. ^{, ,} The TAL exhibits significant variation of permeability characteristics along its length. As such, the permeability of sodium (PNa) may exceed that of chloride (PCl) by 2-fold to 10-fold. ^{, , ,} In the TAL there is a higher PNa/PCl ratio in the ISOM that decreases toward the cortical TAL, while the opposite is found for calcium (PCa) and magnesium (PMg) permeabilities, with both divalent cation permeabilities being comparably higher as the TAL ascends toward the cortex. ^, Limited data are available, and direct comparisons of PCa and PMg are lacking from flux experiments under normal conditions, in part due to the lack of radioactive tracers for magnesium and suitable selective dyes that are only sensitive to magnesium (discussed in detail here ). Bi-ionic diffusion potential measurements in isolated mouse TALs, which require the establishment of large concentrations across the epithelial layer, suggest that PCa exceeds that of PMg. ^{, ,} The rate of paracellular transport does not depend on the permeability characteristics of the epithelium alone but on the electrochemical gradients. However, as discussed elsewhere, direct measurements of the concentration of calcium and magnesium delivered to the loop bend have not been systematically performed, and the scarcity of data poses a challenge in determining the apparent differences. Furthermore, concentration measurements of cations at various segments in the TAL and peritubular capillaries at the same site would be needed but are not easily obtained. Overall, divalent cation permeability is highest along the OSOM and the cortical TAL.

The expression of select claudins and their restriction to specific tight junctions in the TAL likely accounts for the axial heterogeneity of the relative permeability differences observed for monovalent and divalent cations in the TAL. As such, while all cells in the TAL appear to express NKCC2, a number of distinct cell types have been identified by single-cell RNA sequencing, which may enable them to contribute unique transport characteristics. In human and mouse kidneys, two distinct cell populations can be differentiated by their expression of CLDN16 and CLDN10. ^, CLDN16 and CLDN19 are primarily responsible for permeability to calcium and magnesium and hence the movement of divalent cations across the TAL. In contrast to the anion-selective CLDN10a predominantly expressed in the PT, it is the cation-selective CLDN10b that predominates in the TAL. CLDN10b is therefore important for the paracellular movement of sodium and other monovalent cations across the TAL epithelium. Although this distribution of isoform expression is well established from expression studies in various Cldn10 -deficient mouse strains, ^, currently available antibodies recognize epitopes in both, making them unsuitable for discerning their relative expression. It is assumed that the CLDN10 recognized in the TAL is primarily detecting the CLDN10b variant.

Both CLDN16 and CLDN19 are expressed in a subset of TAL cells, where they colocalize to the basolateral membrane domains and tight junction ^, (see Fig. 7.6 ). CLDN16 expression is concentrated to TAL segments in the OSOM and cortex, whereas CLDN19 is also found in the ISOM. However, there, it is not expressed in the tight junction. ^{, ,} Tight junctions not occupied by CLDN16 and CLDN19 in the OSOM and cortex are populated by CLDN10, giving rise to a mosaic expression pattern in the junctions along the length of these segments ^, (see Fig. 7.6 ). In addition, CLDN10 is abundantly expressed in the basolateral membranes and tight junctions of the ISOM. It is the junctional localization of the claudins that most likely determines the permeability characteristics of the epithelium. Overexpression of claudins in cell monolayers has produced variable results with respect to permeability characteristics, especially for CLDN16 and CLDN19. The reasons for that remains unclear but could relate to whether the protein reaches the tight junction or its interaction with claudins already expressed in the cell model (discussed in detail by Gunzel and Yu ). In mice, a detailed study linked the segmental permeability characteristics of the TAL epithelium with the junctional localization of claudins in the same tubule. Here, perfused cortical TAL tubules showed a three times higher PMg/PNa permeability ratio compared with tubules from the ISOM, which coincided with a higher abundance of CLDN16 in the tight junction. Furthermore, the junctions mainly composed of CLDN10, which are highly abundant in the ISOM, where CLDN16 is absent, showed a higher PNa/PCl ratio, in line with CLDN10 forming a sodium-permeable pore, whereas CLDN16 and CLDN19 contribute to the formation of a divalent cation-permeable pore.

These observations are well supported by human disease, where pathogenic variants in the CLDN16 and CLDN19 genes cause the autosomal recessive disorder of familial hypomagnesemia with hypercalciuria and nephrocalcinosis (FHHNC). ^, Loss of function of either gene causes significant urinary losses of divalent cations and, as a consequence, the development of hypomagnesemia and nephrocalcinosis, which can lead to renal insufficiency. ^{, , ,} In contrast, pathogenic variations in CLDN10 cause the HELIX syndrome, which is a rare autosomal recessive disorder characterized by Hypohidrosis, Electrolyte imbalance, Lacrimal gland dysfunction, Ichthyosis, and Xerostomia. ^, These patients present with a sodium chloride–losing phenotype with higher fractional excretion of sodium chloride. Furthermore, these patients display hypermagnesemia, which appears more pronounced in children compared with adults and is accompanied by inappropriately normal urinary magnesium excretion. Furthermore, urinary calcium excretion was normal, but blunted fractional calcium excretion was observed after furosemide injection.

The reason for these findings can be explained on the basis of a detailed analysis of the tubular transport in animals lacking the Cldn16 and Cldn10 genes, or both. Mice with targeted deletion of Cldn16 recapitulated parts of the FHHNC phenotype, displaying excessive losses of urinary calcium and magnesium but without nephrocalcinosis. ^, Furthermore, microperfused cortical TAL tubules from Cldn16 -deficient mice displayed decreased permeability ratios of PCa/PNa and PMg/PNa, whereas no effect of gene deletion was observed on the permeability ratio of PNa/PCl, in line with CLDN16 forming a calcium and magnesium permeable paracellular pore. Divalent cation permeability ratios remained unchanged in perfused TALs from the medullary portion, where CLDN16 is largely absent. ^, Deficient Cldn19 mice have been generated, but the permeability characteristics of the TAL have not been determined. However, in these animals, CLDN19 is required to keep CLDN16 in the tight junction and loss of Cldn19 thus causes the removal of both from the junction, while CLDN16 is retained in the basolateral membrane. ^,

Mice with targeted deletion of Cldn10 in the distal nephron including the TAL recapitulate some features of HELIX syndrome, with higher magnesium concentrations in blood. These mice also display decreased fractional magnesium excretion and marginally decreased fractional calcium excretion, in line with TAL hyperabsorption of these ions. Furthermore, Cldn10 -deletion facilitated the development of nephrocalcinosis, which is absent in patients. ^{, ,} In perfused tubule experiments, distal tubular deletion of Cldn10 leads to a reduction in the PNa/PCl in the TAL but with a concomitant increase in the permeability ratios of PCa/PNa and PMg/PNa. Furthermore, the transepithelial voltage gradient is markedly increased in Cldn10 -deficient animals. The high PCa/PNa and PMg/PNa in Cldn10 -deficient animals result from repopulation of the CLDN10-deficient junctions with CLDN16, while the opposite is not seen in Cldn16 -deficient animals. ^, In combination with the higher transepithelial voltage gradient following Cldn10 deletion, this is a key driver for hyperabsorption of divalent cations along the TAL. In fact, Cldn10 – and Cldn16 -double-deficient mice display reduced PCa/PNa and PMg/PNa in perfused TAL from the cortex in comparison with Cldn10-deficient mice. The development of nephrocalcinosis was also absent in double-deficient animals. Whether the same claudin redistribution occurs in the human kidney remains to be determined in detail for CLDN16, but CLDN19 was found redistributed to junctions in the TAL of a patient with mutations in CLDN10. In addition to these pore-forming claudins that predominate in the TAL is the pore-blocking CLDN14, which is largely absent during baseline conditions but strongly stimulated by the CaSR (see section on CaSR effects on the kidney later).

While phosphate is not reclaimed along the TAL, it is notable that the sodium-phosphate cotransporter SLC34A2 has been found to be expressed in the TAL. Its contribution to phosphate transport or other cellular mechanisms remains to be defined.

The distal convolution reclaims 4% to 8% of the calcium filtered by the glomerulus (see Fig. 7.4 ). Calcium is reabsorbed by an active transcellular process, allowing the reabsorption of the ion against a lumen-negative transepithelial voltage gradient. ^, The Transient Receptor Potential Vanilloid 5 (TRPV5) is the main uptake mechanism for calcium in the distal convolution. ^, TRPV5 is part of the multigene TRP superfamily, which encodes a membrane protein that assembles into a homotetrameric ion channel. TRPV5 was cloned from primary rabbit CNT and CD isolates but only localized to the CNT in rabbit. ^, In mice, TRPV5 is expressed in the DCT2, CNT, and cortical CD; however, its expression in humans remains to be determined ^{, ,} (see Fig. 7.7A and 7.7B ). Differences in this segmentation have been described between species, with the DCT2 segment being absent in rabbits, shortened in humans, and more prominent in mice and rats. ^, Massive renal calcium wasting, osteopenia, and compensatory intestinal hyperabsorption are observed in mice with targeted ablation of Trpv5. These Trpv5 -deficient mice showed normal proximal tubular calcium handling, but calcium reabsorption along the distal convolution was markedly altered (see Fig. 7.7C ). As potassium secretion along the distal convolution increased, potassium concentration was used as an indication of distance at various puncture sites along the distal convolution. Wild-type mice display a reduction in luminal calcium delivery along the length of the micropuncturable distal convolution, in accordance with tubular reclamation of calcium along the segment. However, Trpv5 -deficient mice showed persistently elevated calcium concentrations along the length of the distal convolution, consistent with a failure to reabsorb calcium from the DCT/CNT and indicating that this is the segment responsible for their hypercalciuria (see Fig. 7.7C ). In humans, pathogenic variants that cause loss of function have not been identified, but a missense SNP, albeit rare, has been described to correlate with kidney stone disease. In addition, TRPV6 may provide an additional contribution to this transport, but animals lacking Trpv6 have not been described to display significant calcium wasting ^, (see Fig. 7.7A and 7.7B ).

Calbindin-D _28K is expressed along with TRPV5 in the distal convolution (see Fig. 7.7A and 7.7B ). Calbindin-D _28K is an intracellular protein that can bind calcium with high affinity. It aids in transferring calcium across the cell and buffers intracellular calcium concentrations. This is thought to contribute to high vectorial flux across the cell, without substantially elevating intracellular calcium concentrations. Furthermore, a high intracellular calcium concentration also inhibits TRPV5 activity. However, hypercalciuria has not been observed in animals with deletion of the Calb1 gene encoding Calbindin-D _28K. Whether the absence of hypercalciuria results from compensation from other calcium-binding proteins expressed in the segment remains to be determined. Insights from a number of different knockout mouse studies suggest this may be the case. The targeted ablation of the Vdr gene leads to large disturbances in calcium balance, including a 90% reduction in Calbindin-D _9K in kidney; however, this is largely without changes in Calbindin-D _28K . Double knockout mice with Vdr and the Calb1 gene had more severe hypercalciuria and mineral disturbances in comparison with Vdr -deficient animals. Further, the deletion of both the S100g gene encoding Calbindin-D _9K along with Calb1 does not result in a change in serum calcium. However, it has not been reported whether hypercalciuria develops. In addition, other yet-to-be-identified calcium-binding proteins could also compensate for this function in the distal nephron.

While the electrochemical gradient drives the influx of calcium from the lumen into the cell, calcium needs to be removed from the cell against this gradient. This is largely facilitated by the plasma membrane calcium ATPase PMCA4 and the sodium/calcium exchanger type 1 (NCX1) ^{, ,} (see Fig. 7.7A and 7.7B ). They drive the secretion of calcium across the basolateral membrane by use of ATP or via a secondarily active process dependent on the sodium gradient, respectively. The PMCAs are high-affinity, low-capacity pumps that maintain low calcium concentrations inside the cytosol, while NCXs are low-affinity, high-capacity carriers. Isolation of tubules from the distal convolution from mice revealed that sodium-dependent basolateral extrusion facilitated approximately half of transcellular efflux with low vertorial flux. Targeted deletion of the Atp2b4 gene encoding PMCA4 does not lead to noticeable alterations in calcium balance, suggesting that NCX1 is the predominant efflux mechanism in this segment. Furthermore, NCX2 may also contribute to basolateral calcium efflux, although the exact contribution of this exchanger has not been defined. The expression of NCX1 is highest in the DCT2 and CNT region of the distal convolution, with lower expression of both PMCA4 and NCX1 in the DCT1. However, an apical calcium channel enabling calcium influx has not been identified in the DCT1 ^, (see Fig. 7.7A ).

Detailed micropuncture experiments have demonstrated that 5% to 6% of the filtered magnesium is reabsorbed along the distal convolution. ^{, ,} The measurement of magnesium delivery to early and late puncture sites along the distal convolution of the same tubule suggests that the early convolution, which likely represents the DCT, is where the majority of magnesium is reabsorbed. The luminal concentration of magnesium ranges between 0.2 and 0.7 mM in this segment, whereas intracellular concentrations of magnesium are between 0.2 and 1.0 mM. ^, As such, the membrane potential is likely a key contributor to determining the apical entry of magnesium. ^, Inward transport of magnesium has generally been assumed to be mediated by the Transient Receptor Potential Melastin 6 channel (TRPM6), which is a channel permeable to divalent cations. Patients with mutations of TRPM6 suffer from the autosomal recessive disorder hypomagnesemia with secondary hypocalcemia (HSH). HSH is characterized by markedly impaired absorption of magnesium from the intestinal lumen. ^{, ,} This severe disease typically presents in the first few months of life with seizures or general signs of neuromuscular excitability including tetany or muscle spasm. TRPM6 is highly expressed in the large intestine of experimental animals, where the transcellular flux of magnesium is high, explaining impaired intestinal magnesium absorption in HSH patients. TRPM6 is also expressed in the apical membrane of the DCT, as identified by immunohistochemical staining ( Fig. 7.8 ). In line with these findings is the identification of a higher-than-normal fractional excretion of magnesium in HSH patients given their blood concentration of magnesium, ^, a finding indicative of renal tubular magnesium wasting. This renal wasting of magnesium is especially pronounced in HSH patients receiving intravenous administration of magnesium. In contrast to the suggestion that TRPM6 contributes to both renal and intestinal magnesium reabsorption are the findings of Chubanov and colleagues, who generated transgenic mouse lines with targeted deletion of either intestinal or renal Trpm6 . While intestinal deletion of the channel resulted in a phenotype with lower serum magnesium levels, the renal deletion of Trpm6 using the Ksp-Cre driver line that targets the distal tubule did not yield any changes in serum magnesium or urinary magnesium excretion. As the Ksp-Cre driver line appears to target the DCT with varying efficiency, with activity reported to be less than half in this segment, this might explain the absence of marked changes in magnesium homeostasis reported for this strain. In contrast and consistent with a role for TRPM6 in renal tubular magnesium reabsorption, renal tubular Trpm6- deficient animals generated on a six2-cre background displayed hypomagnesemia and higher magnesium in the urine. In addition to these findings, TRPM6 may not be the sole pathway for magnesium entry into the DCT, and the contribution of the close homolog TRPM7 remains to be fully delineated in this process ^, (see Fig. 7.8 ). However, using the Ksp-Cre driver line, Trpm7 deletion in the kidney was not found to cause alterations in mineral balance, either with respect to urinary excretion or serum concentration. Deletion of Trpm7 using the Pax3-Cre, which excises genes in the metanephric mesenchyme, leads to an early developmental defect. Additional studies with DCT-specific Cre lines are required to firmly conclude its contribution to renal magnesium transport in mice. In contrast, Trpm7 deletion in the intestine caused a severe phenotype with disturbed mineral balance and early lethality after birth, suggesting that it does play an important role in the intestine. Two patients with pathogenic variants in the TRPM7 gene have been reported. They display an autosomal dominant type of HSH, indicating that loss of a single allele can lead to disturbed intestinal mineral transport in humans.

Tubular transport of magnesium in the distal convolution.

Schematic representation of the magnesium transport machinery in the distal convoluted tubule. The apical entry of magnesium is thought to occur through transient receptor potential Melastin 6 (TRPM6) and TRPM7 channels. The basolateral exit pathway for magnesium remains to be determined, but CNNM2 or SLC41A1 could contribute. The membrane voltage is adjusted by the potassium channel, Kv1.1, and influenced by the Na ⁺ -K ⁺ -ATPase, the accessory subunit, FXYD2, and Kir4.1, a basolateral potassium channel that recycles potassium ions that enter the cell through the Na ⁺ -K ⁺ -ATPase. Mutations in HNF1B and PCBD1 cause hypomagnesemia and may influence FXYD2 expression. Epidermal growth factor (EGF) and epidermal growth factor receptor (EGFR) are involved in the regulation of transcellular magnesium transport in the segment.

It currently remains to be determined in detail whether TRPM6 is required for TRPM7 to function. While some groups have shown that TRPM6 can form currents in heterologous expression systems, ^, others have not. ^{, , ,} A TRPM7-like current has been detected in mouse trophoblast stem cells that express both TRPM6 and TRPM7 channels. When these cells are isolated from mice with Trpm7 -deficiency, meaning only functional TRPM6 is left in the cell, the measured current is absent. In contrast, Trpm6 -deficient cells that contain only TRPM7 display a reduced current amplitude but are more sensitive to magnesium inhibition, indicating that TRPM6 requires TRPM7 to function, whereas TRPM6 may alter the function of the TRPM6-TRPM7 channel complex.

A number of pathogenic variants in genes expressed in the distal convolution cause hypomagnesemia. These gene defects have provided insights into mechanisms important for magnesium transport in the distal convolution. ^, Patients with pathogenic variants in the Na ⁺ -K ⁺ -ATPase γ subunit display autosomal dominant hypomagnesemia, which is secondary to urinary loss of magnesium. This disorder is often associated with hypocalciuria. Children may have muscle cramps and seizures or may be asymptomatic. This variant of hypomagnesemia was identified in two families. The cause was subsequently identified as a G41R mutation in the FXYD2 gene, encoding the γ subunit of the Na ⁺ -K ⁺ -ATPase. The FXYD proteins are a group of transmembrane proteins that have the ability to regulate the function of Na ⁺ -K ⁺ -ATPase. The localization of the γ subunit in kidney remains to be fully delineated; however, reports suggest expression in the DCT (see Fig. 7.8 ). The γ subunit is important for modulating the kinetics of the Na ⁺ -K ⁺ -ATPase ^{, ,} ; however, the protein with the pathogenic variant G41R does not localize to the plasma membrane in comparison with the wild type when expressed in cells, suggesting that Na ⁺ -K ⁺ -ATPase activity could be reduced. ^, Changes in the transepithelial voltage, resulting from this, could lower the electrical gradient for apical uptake of magnesium. However, no change in serum magnesium or urinary magnesium excretion is seen in Fxyd2 -deficient mice or humans lacking one allele of the FXYD2 gene, ^, suggesting that the G41R variant, rather than gene deletion, is causative of the phenotype. Another study has found that the γ subunit may form an inward rectifying channel, and the G41R variant changes channel properties. The exact role of the Na ⁺ -K ⁺ -ATPase γ subunit in magnesium transport in the kidney remains to be understood in detail. The importance of adequate Na ⁺ -K ⁺ -ATPase activity for renal magnesium reabsorption has been highlighted in patients with de novo mutations in the Na ⁺ -K ⁺ -ATPase α subunit encoded by the ATP1A1 gene, which leads to renal hypomagnesemia, refractory seizures, and intellectual disability (see Fig. 7.8 ).

Another cause of renal magnesium wasting and hypomagnesemia is due to mutations in the transcription factor hepatocyte nuclear factor 1B (HNF1B) (see Fig. 7.8 ). Pathogenic variations in the HNF1B gene do not universally cause hypomagnesemia, but they can be part of a complex clinical picture that includes renal cysts and diabetes. In 66 patients, 44% of carriers were found to have hypomagnesemia, hypermagnesuria, and hypocalciuria. The nephron segment and exact cause of the renal magnesium wasting is not clear, but studies have demonstrated that the transcription of the FXYD2 gene is regulated by HNF1B. Furthermore, pathogenic variants in the PCBD1 gene can cause hypomagnesemia and modify HNF1B-dependent transcription of FXYD2 (see Fig. 7.8 ).

Patients with pathogenic variations in the KCNJ10 gene, which encodes the inwardly rectifying potassium channel (Kir4.1, KCNJ10), have a complex syndrome that includes epilepsy, ataxia, sensorineural deafness, and tubulopathy (EAST, SeSAME). The tubulopathy manifests with hypokalemic metabolic alkalosis, hypomagnesemia, and hypocalciuria, similar to the clinical presentation of Gitelman syndrome (discussed later). ^, KCNJ10 is located in the basolateral membrane domains of the DCT, and the channel plays a role in recycling potassium ions across the basolateral membrane to maintain the high Na ⁺ -K ⁺ -ATPase activity of the DCT (see Fig. 7.8 ). Without KCNJ10, a decrease in Na ⁺ -K ⁺ -ATPase activity is expected to cause cell depolarization, which in turn reduces the reabsorption of magnesium.

A rare autosomal recessive disorder that results in isolated renal hypomagnesemia has been described due to a pathogenic variant in the KCNA1 gene encoding the Kv1.1 voltage-gated potassium channel, situated in the apical DCT membrane. These patients present with muscle weakness, cerebellar atrophy, myokymia, muscle cramps, and tetany starting from early childhood. The channel appears to play an important role in maintaining membrane voltage. Hypomagnesemia is thought to be caused by mutations in the channel, which likely decreases the inward driving force for magnesium (see Fig. 7.8 ).

CNNM2 is highly expressed in the distal nephron and brain. Pathogenic variants in this gene result in renal magnesium loss, hypomagnesemia, seizures, and impaired intellectual development. ^, Although this protein appears to mediate sodium-dependent magnesium flux, its specific role in DCT magnesium transport is not fully understood (see Fig. 7.8 ). In contrast, another member of the family, CNNM4, plays an important role in transcellular efflux of magnesium from the intestine in mice. Pathogenic variants in the SLC41A1 gene encoding a magnesium transporter result in a nephronophthisis-like phenotype in humans. ^, Deletion of the Slc41a3 gene, a mitochondrial magnesium transporter, results in hypomagnesemia in mice (see Fig. 7.8 ). However, double knockout of both the Slc41a1 and Slc41a3 genes does not cause exaggerated hypomagnesemia. The molecular physiology as to the etiology of these disturbances is unclear. Other causes of renal-mediated magnesium wasting include pathogenic variations in the SLC12A3 gene, which causes Gitelman syndrome and is discussed later in the section on diuretics and pathogenic variants in the EGF gene (see Fig. 7.8 ), which is discussed in the section on antineoplastic drugs.

The contribution of the distal convolution to the reabsorption of phosphate is debated. Multiple studies have been unable to detect phosphate transport in the distal convolution in animals receiving a phosphate-replete standard diet. However, there could be some distal tubular phosphate reabsorption following phosphate deprivation. ^, If this occurs, the cellular mechanism remains to be defined.

The majority of data does not support a role for the CD in the reclamation of filtered calcium; however, some reports have suggested a small amount of calcium might be reabsorbed in the collecting system. Micropuncture studies have established that the delivery of calcium to the end of the distal convolution is low, in line with the absence of significant calcium transport along the remainder of the CD. ^{, ,} Immunohistochemical staining of the kidney shows that TRPV5, NCX1, and PMCA are expressed in the CNT but undetectable in the cortical CD. In contrast, mice with targeted expression of green fluorescent protein driven by the Trpv5 promoter revealed expression along the cortical CD. In the rabbit, conflicting results have been presented. While some microperfusion studies find significant calcium flux across the cortical CD, ^, others have failed to do so. ^, In the inner medullary portion of the CD in rat, microcatheterization revealed a gradual but significant decline in calcium along its length. Therefore the overall contribution of the CD to calcium transport seems insignificant, but it remains to be determined whether the segment contributes a small amount of calcium transport.

No magnesium reabsorption has been detected in the collecting duct, consistent with transcellular flux occurring primarily in the DCT. In addition, microcatheterization has been unable to identify magnesium reabsorption in the CD in the inner medulla.

While some microcatheterization studies have described a small amount of phosphate transport in the collecting duct under some conditions, the contribution appears small and the potential molecular mechanism mediating this is not well understood. ^,

Lowered blood calcium levels relieve the CaSR-dependent inhibition of PTH release from the parathyroid, which results in the secretion of PTH into the circulation. The pre-prohormone PTH is processed and ultimately released into the circulation as an 84-amino acid hormone, with the first 34 amino acids sufficient to confer biological activity via binding to a PTH receptor, PTH1R or PTH2R. PTH1R is the main receptor responsible for the renal actions of PTH and is expressed along the renal tubular epithelium, as well as in bone, whereas the PTH2R expression is restricted to extrarenal sites including the pancreas, testis, central nervous system, and placenta. In situ hybridization and single-cell RNA sequencing have primarily found PTH1R to be expressed in podocytes and along the PT, TAL, and distal convolution, with little or no expression in the remainder of the collecting system. ^, Immunohistochemistry confirmed PTH1R localization to the PT, TAL, and distal convolution, where PTH1R is found in the basolateral membrane. ^, These segments also show enhanced adenylate cyclase activity in the presence of PTH, resulting in elevated cAMP formation. Additionally, PTH1R has been localized to the brush border membrane of the PT in rodents. ^, The overall effect of PTH is to reduce urinary calcium excretion, thereby conserving calcium in the blood.

Phosphate

PTH increases the urinary excretion of phosphate ^, ( Fig. 7.9 ). PTH reduces proximal tubule phosphate reabsorption by decreasing the brush border membrane abundance of SLC34A1 and SLC34A3, as well as PIT2. The mechanisms whereby this occurs have been studied in detail for SLC34A1. PTH activation of the PTH1R expressed in the basolateral membrane of the PT results in the activation of the protein kinase A–dependent pathway and downstream extracellular-signal-regulated kinases (ERK) signaling leading to the phosphorylation of NHERF1. ^, NHERF1 is a scaffolding protein important for tethering SLC34A1 to the brush border membrane (discussed earlier). PTH-dependent downstream phosphorylation of NHERF1 leads to dissociation from SLC34A1 and subsequent endocytosis via clathrin-coated pits, ultimately resulting in its degradation in lysosomes. When transfecting variants in NHERF1 from patients with nephrolithiasis or bone mineralization defects, a potentiated response in cAMP formation is seen after PTH administration to cells expressing the NHERF1 variants. Furthermore, while administration of PTH inhibited phosphate uptake in all transfected cells, PTH had a greater ihibitory effect on phosphate transport in cells transfected with the NHERF1 variants than the wild-type variant.

Changes in phosphate absorption along the proximal tubule (PT) in various conditions.

The effects of dietary phosphate loading or deprivation, parathyroid hormone (PTH) infusion, or parathyroidectomy on phosphate absorption along the PT. Proximal tubule %, Distance along the PT as a percentage of total length; PTH, parathyroid hormone; TF/UFPi , ratio of tubular fluid-to-ultrafiltrate phosphate concentration.

In addition to this regulation via basolateral PTH1R signaling, apical receptors have also been identified in the PT. Activation of these receptors also causes internalization of SLC34A1, although this has been described to occur through a phospholipase C and a protein kinase C–dependent pathway. Activation of this pathway via PTH1R in the apical membrane also activates downstream ERK signaling, releasing SLC34A1 from NHERF1 and hence its tethering to the brush border membrane. ^{, ,} While PTH also causes internalization of SLC34A3, unlike SLC34A1, it does not undergo subsequent degradation. ^, The precise molecular mechanisms underlying the internalization of SLC34A3 have not been completely elucidated, although there is evidence suggesting the involvement of PKC. There is limited knowledge regarding the impact of PTH on PiT-2, even though the hormone decreases the abundance of the transporter at the brush border.

It is not surprising then that primary hyperparathyroidism causes hypercalcemia and symptoms associated with this electrolyte disturbance including GI upset, polyuria, polydipsia, muscle fasciculations, kidney stones, or bone pain. This is typically due to parathyroid hyperplasia, adenoma, or less commonly a malignant tumor of the parathyroid glands. Conversely, primary hypoparathyroidism presents with symptoms of hypocalcemia including muscle aches/spasms, tetany, and seizures. Hypoparathyroidism is most commonly the result of damage from thyroid surgery but can be due to genetic mutations, or autoimmune disease. Due to the effect of PTH on renal phosphate excretion, phosphate levels are typically lower in hyperparathyroidism, whereas they are often high or in the high end of the normal range in hypoparathyroidism.

1,25-dihydroxyvitamin D (calcitriol) plays a critical role in the maintenance of calcium balance. Vitamin D ₃ is produced in the skin. In keratinocytes, UVB light converts 7-dehydrocholesterol to pre–vitamin D ₃ and this precursor is then converted to vitamin D ₃ ( Fig. 7.10 ). Vitamin D ₃ (cholecalciferol) and vitamin D ₂ (ergocalciferol) can also enter the circulation via dietary consumption. Vitamin D ₂ is obtained from plants, whereas vitamin D ₃ is found in meat and milk products. After entering the circulation, these compounds can be hydroxylated to form 25-hydroxy-vitamin D (calcifediol) by cytochrome P450 enzymes in the liver. ^, Subsequent conversion of 25-hydroxy-vitamin D to the active hormone, 1,25-dihydroxyvitamin D, is mediated by the cytochrome P450 enzyme CYP27B1, which is also known as the 1-α-hydroxylase (1α-hydroxylase). ^, 1α-hydroxylase is strongly stimulated by PTH and inhibited by 1,25-dihydroxyvitamin D, FGF23, and calcium. The effect of 1,25-dihydroxyvitamin D is dependent on binding to the vitamin D receptor (VDR), which, in complex with the retinoid X receptor (RXR), drives gene expression via the VDR response element (VDRE) in the promoters of target genes. The VDR is present in mineral-transporting epithelia such as the kidney and intestine, as well as bone, in addition to a number of other organs, where it relays the effect of 1,25-dihydroxyvitamin D on target tissues.

Formation and metabolism of vitamin D3.

1-hydroxylase (1-α hydroxylase, CYP27B1, cytochrome P450 family 27 subfamily B member 1). 24-hydroxylase (CYP24A1, Cytochrome P450 family 24 subfamily A member 1).

FGF23 is a peptide hormone consisting of 251 amino acids that is produced and secreted by osteocytes and osteoblasts in response to increased levels of phosphate or 1,25-dihydroxyvitamin D in blood. Proteolytic cleavage of the active hormone is essential to its biological activity. A furin protein convertase cleaves intact FGF23 at amino acid 179 into two fragments that may or may not have plasma phosphate-lowering activity. ^, Full-length FGF23 primarily functions to decrease the reabsorption of phosphate in the PT and inhibit the synthesis of calcitriol, but it also has effects on tubular calcium reabsorption in the distal nephron. ^{, ,} Klotho is produced at high levels in the distal convolution, and circulating klotho cleavage products can be detected in the peripheral circulation. Animals lacking the Kl gene encoding klotho exhibit both calcium wasting and nephrocalcinosis. Klotho was first recognized for its ability to catalyze the hydrolysis of oligosaccharides linked to N-glycan, increasing the abundance of TRPV5 channels situated at the apical membrane, and thus has been implicated in calcium homeostasis. ^, Yet subsequent research has shown that klotho functions as a nonenzymatic scaffold that is essential for FGF23 receptor activation, which is likely its more important role. In line with this, mice lacking Fgf23 exhibit a comparable phenotype to mice lacking klotho in terms of TRPV5 membrane abundance and renal calcium wasting.

SGLT2 inhibitors are frequently employed to treat type 2 diabetes, confer protective effects on renal function, and prevent cardiovascular events. An increased fracture rate has been associated with these compounds. ^, Increased serum phosphate, PTH, and FGF23 have been observed to occur with treatment, although altered renal excretion of calcium and phosphate has not been observed. It is speculated that increased PT phosphate reabsorption drives hyperphosphatemia and subsequent increases in PTH and FGF23 causing bone fragility. SGLT2 inhibitors have been reported in several case series to increase serum magnesium levels when given to patients with hypomagnesemia. This seems to be a class effect, but the mechanism is currently unknown. Fractional excretion of magnesium was reported to be reduced in two cases but unchanged in another two. Post-hoc analyses and meta-analyses of trials of SGLT2 inhibitors have also shown small increases in magnesium levels.

Loop diuretics, such as furosemide, are primarily used in pediatric and adult patient populations for the treatment of edema. As they reduce the transepithelial voltage gradient in the TAL, they limit the reabsorption of divalent cations. This is mirrored in experimental animal models, where loop diuretics can cause urinary calcium and magnesium wasting. ^, Similar findings have also been observed in patients and healthy volunteers receiving loop diuretics. Hypomagnesemia has been reported following loop diuretic treatment in humans; in contrast, studies on the occurrence of hypocalcemia are limited ^, and direct comparisons of these side effects following the administration of loop diuretics are lacking. The appearance of symptoms related to overall mineral balance may also depend on the dose and frequency of administration, as well as preexisting mineral imbalances. Loop diuretics did not alter urinary phosphate excretion in healthy humans acutely. Loop diuretics are discussed further in Chapter 49 .

Thiazides and thiazide-like diuretics remain the most frequently prescribed hypertension medication. The class of diuretics acts by blocking the NCC cotransporter. ^{, ,} In addition, they also have a net positive effect on calcium balance by reducing urinary calcium excretion, ^, an important feature of the drug that also limits hypercalciuria in kidney stone patients. ^, The hypocalciuric effect appears independent from changes in PTH, and a similar phenomenon is observed in patients with the salt-wasting Gitelman syndrome due to pathogenic variants in the SLC12A3 gene. The hypocalciuric effect is also documented in animals following thiazide administration or genetic mouse models deficient in Slc12a3. ^, The hypocalciuric effect remains to be fully clarified and may result from altred tubular calcium transport in both the PT and distal convolution (see reviews for additional details ^, ). The most explored pathway for hypocalciuria is compensatory hyperabsorption of calcium in the PT. This follows thiazide-induced volume contraction, which augments PT reabsorption of sodium and hence also calcium. As such, the hypocalciuric effects of thiazides are blunted when salt loss is compensated in thiazide-treated individuals. Furthermore, volume depletion by salt restriction does not induce hypocalciuria in Slc12a3 –deficient mice but it does in wild-type littermates. Micropuncture experiments have confirmed this in animals after chronic thiazide administration and in Slc12a3 -deficient animals. The hypocalciuria was still present in Trpv5 -deficient mice following thiazide administration, suggesting that the PT contributes importantly to the hypocalciuric effect. A distal mechanism to drive hypocalciuria may also exist in addition to the PT mechanism ; however, this pathway may primarily contribute in the absence of volume contraction and the exact nature of this pathway remains elusive. ^,

In patients, chronic thiazide treatment may lead to the appearance of hypomagnesemia, albeit infrequently in some populations. ^, In contrast, hypomagnesemia is a characteristic feature observed in patients with Gitelman syndrome. The differences in the appearance of hypomagnesemia may depend on the degree of NCC transporter inhibition. The hypomagnesemia results from renal magnesium wasting. In experimental animals, both Slc12a3-deficiency and chronic thiazide administration lead to atrophy of the DCT1 segment in the distal convolution and hypertrophy of the DCT2 and CNT. ^{, , , ,} This may depend on the degree of inhibition and length of treatment, as other studies find no significant changes in the volume of the DCT. ^, This segment plays a key role in renal magnesium transport, which likely explains the observed hypomagnesemia and renal magnesium losses following thiazides and Gitelman syndrome. Furthermore, direct downregulation of TRPM6 has also been reported without significant alterations in DCT morphology after thiazide administration in experimental animals. Thiazide diuretics were not found to alter urinary phosphate excretion acutely. Thiazide diuretics are discussed further in Chapter 49 .

Potassium-sparing diuretics such as spironolactone and amiloride may have mild effects on the urinary excretion of both calcium and magnesium, acting to increase reabsorption of both in experimental animals and humans ^{, ,} ; however, the underlying mechanisms remain to be explored in detail. For further discussion on the effects of aldosterone, see Chapter 12 . The potassium-sparing diuretic, triamterene, did not alter urinary phosphate excretion in healthy volunteers.