Paracellular calcium transport in the proximal tubule and the formation of kidney stones

Abstract

The proximal tubule (PT) is responsible for the majority of calcium reabsorption by the kidney. Most PT calcium transport appears to be passive, although the molecular facilitators have not been well established. Emerging evidence supports a major role for PT calcium transport in idiopathic hypercalciuria and the development of kidney stones. This review will cover recent developments in our understanding of PT calcium transport and the role of the PT in kidney stone formation.

INTRODUCTION

Bulk reabsorption of about two-thirds of sodium, chloride, water, and calcium occurs within the proximal tubule (PT) of the kidney. For decades, it has been known that a substantial proportion of calcium reabsorption is passive and tightly coupled to the reabsorption of sodium, including active sodium bicarbonate reabsorption and sodium-coupled cotransport of glucose, amino acids, and other osmotically active solutes (13). The PT has a very low transepithelial resistance (TER) and is highly permeable to water and small ions, especially calcium, sodium, and chloride (3, 37). In the early PT, sodium-coupled transport drives water reabsorption, leading to the generation of a gradient for the diffusion of calcium and chloride in later segments of the PT (22). Additionally, the paracellular pathway is relatively impermeable to bicarbonate, allowing for passive chloride reabsorption and generation of a positive lumen potential (13). Thus, the result of PT sodium and chloride reabsorption is a favorable electrochemical gradient for the passive reabsorption of calcium. A recent mathematical model of PT transport predicts that the major driving force for calcium reabsorption in the PT is the concentration gradient, with the positive lumen potential of the late PT playing a lesser role (5). Figure 1 shows our current model of paracellular calcium transport in the PT.

Fig. 1.

In the early portion of the proximal tubule (early PT), reabsorption of sodium, bicarbonate, and water occur alongside sodium-coupled cotransport of glucose (Glu) and amino acids (AA). As a result, the intraluminal chloride concentration increases above that of plasma. Additionally, much of early PT sodium transport is electrogenic, thus generating a negative lumen potential (approximately −2 mV). This favorable electrochemical gradient drives the paracellular reabsorption of chloride. In the later proximal tubule (mid-late PT), paracellular chloride transport generates a positive lumen potential (approximately +2 mV). These processes generate the electrochemical driving forces for proximal tubule sodium and calcium transport (33).

ASSOCIATION BETWEEN HYPERCALCIURIA AND RENAL STONES

The association between hypercalciuria and renal stones was first established many decades ago (8). In the majority of cases, this hypercalciuria is identified in the absence of hypercalcemia or other metabolic defects and thus considered idiopathic (29). Compared with controls, calcium stone formers have increased natriuresis in response to hydrochlorothiazide and decreased natriuresis in response to acetazolamide, suggesting defective PT reabsorption as a general mechanism for idiopathic hypercalciuria in stone formers (38). This is particularly true for patients with fasting hypercalciuria (35, 38). By measuring endogenous lithium clearance, a well-recognized marker of PT sodium reabsorption, more recent studies have bolstered the evidence that calcium stone formers with idiopathic hypercalciuria have reduced PT reabsorption (19, 43). Despite this, the majority of patients have normal fractional sodium excretion, presumably due to compensatory transport mechanisms in the downstream nephron (43).

PT CALCIUM TRANSPORT

The PT is highly permeable to calcium (26), and PT calcium transport is closely coupled to sodium transport (1, 6, 37). Microperfusion of isolated PT segments has demonstrated net calcium transport of zero in the absence of osmotic or potential differences and stimulation of net absorption in response to a favorable electrochemical gradient (28). Together, these data suggest that the majority of PT calcium transport is paracellular. Approximately 10% of PT calcium transport may also occur via a transcellular pathway (39). The claudin family of proteins are major determinants of paracellular ion selectivity (16). In 1998, Furuse et al. (10) first sequenced 22-kDa peptides found in tight junction (TJ) membrane fractions and identified two distinct four-pass transmembrane proteins that localize to the TJ of epithelial cells. These proteins were named claudin-1 and claudin-2 after the Latin word “claudere,” or “to close” (10). Ever since the initial discovery of claudins, at least 24 different claudin isoforms have been described (16). Mice with genetic deletion of claudin-1 exhibit severe dehydration and death without morphological changes to the TJ ultrastructure (12). Despite the absence of ultrastructural abnormalities to the TJ, loss of claudin-1 in mice allowed diffusion of a molecular tracer to the surface of the skin, suggesting a role in barrier formation between cells (12). Whereas claudin-1 is expressed in a multitude of tissues and appears to function as an intercellular barrier (10), claudin-2 is expressed only in leaky epithelial tissues such as the PT and intestinal crypts (7, 31). In contrast to claudin-1, which increases TER in vitro (17), claudin-2 decreases the TER of epithelial monolayers by preferentially increasing cation permeability (2, 11, 44). Consequently, the first two claudin proteins to be identified nicely illustrate the diverse function of the claudin family of proteins to create either extracellular ion barriers or pores.

In the PT, the most highly expressed claudin isoforms are claudin-10a, claudin-17, claudin-3, and claudin-2 (7, 14, 20, 21). Claudin-10 is abundantly expressed in the PT (18, 21). Transcription of claudin 10 gene can result in multiple splice variants, depending on the cell type, with the two major variants differing in their charge selectivity (41). In the PT of mice, the anion-selective variant claudin-10a is expressed (14) and may provide the paracellular route for chloride reabsorption. As an anion-selective claudin isoform, claudin-10a expression is unlikely to have a large effect on calcium transport, but selective deletion of claudin-10a in an animal model has not yet been described. Similarly, claudin-17 has been shown to increase anion permeability in vitro and is not likely to play a major role in calcium reabsorption (20). In the rat, claudin-3 is highly expressed in the late, straight segments of PT (21). Claudin-3 appears to function as a barrier isoform, with overexpression leading to reduced cation permeability and increased TER in vitro (24).

CLAUDIN-2

The cation-selective isoform claudin-2 has been shown to increase calcium permeability in vitro (44), and emerging evidence suggests that it plays a role in intestinal calcium absorption (9). Muto et al. (27) showed that loss of claudin-2 (Cldn2 gene) in mice leads to reduced sodium permeability in isolated PT segments. Additionally, claudin-2 knockout mice exhibit a threefold increase the fractional excretion of calcium compared with wild-type mice and excrete more sodium after being loaded with hypertonic sodium chloride (27). Although the calcium permeability of isolated PTs from claudin-2 knockout mice has not been measured directly, these data do suggest a major role for claudin-2 in PT paracellular calcium reabsorption. The pattern of elevated urinary calcium and normal sodium excretion in claudin-2 knockout mice mimics findings in patients with idiopathic hypercalciuria and nephrolithiasis (19). Our laboratory’s initial findings suggest that loss of claudin-2 in mice causes the development of papillary nephrocalcinosis in addition to hypercalciuria. Papillary nephrocalcinosis is a common finding in human renal tissue and is associated with the formation of kidney stones (32). Among patients with kidney stones and idiopathic hypercalciuria, defective PT reabsorption appears to be most pronounced in men (19). Recent work in mice suggests that renal transport of sodium differs between male and female mice, with male mice having a greater abundance of claudin-2 and other known PT transport proteins (42). We hold that this evidence warrants further investigation of the role of claudin-2 in kidney stone formers.

Although clinical data implicates PT calcium transport as a common factor in the development of kidney stones, no genetic associations have been identified thus far to directly link PT calcium transport and kidney stone disease. Although several genetic factors link PT transport to nephrolithiasis risk, none appear to directly involve PT calcium transport mechanisms. For instance, Dent disease type I (chloride voltage-gated channel 5 gene) and Dent disease type II (inositol polyphosphate 5-phosphatase gene) involve generally defective PT solute reabsorption and include hypercalciuria, nephrocalcinosis, and nephrolithiasis (15). In addition, mutations in the genes for the PT sodium-phosphate cotransporters SLC34A1 and SLC34A3 have been linked to nephrolithiasis (4, 30). Aquaporin-1 (AQP1) is a water transport protein found at the apical and basolateral membranes of PT epithelial cells (23). In 2012, a nephrolithiasis risk variant near the gene for AQP1 was identified, thus implicating PT water transport in the development of nephrolithiasis (40). Aqp1 knockout mice exhibit severe dehydration upon water restriction (23), and dehydration is a major risk factor in kidney stone disease (25). As noted above, PT water reabsorption generates the major driving force for paracellular calcium transport in the segment. Hypercalciuria in Aqp1 knockout mice has not been reported, and it remains unknown whether this AQP1 variant or other AQP1 mutations in humans are associated with increased urine calcium excretion. In addition to its apparent role in calcium homeostasis, it is interesting to note that claudin-2 has also been demonstrated to function as a water pore (34) and contributes to PT water reabsorption in vivo (36).

CONCLUSIONS

Recent work highlights a potential role for the PT and claudin-2 in human idiopathic hypercalciuria and nephrolithiasis. Our future studies aim to analyze these associations and further characterize the contribution of PT claudin-2 to hypercalciuria in vivo.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.N.C. prepared figures; J.N.C. drafted manuscript; J.N.C. and A.S.Y. edited and revised manuscript; A.S.Y. approved final version of manuscript.