467

INVITED REVIEW Paracellular calcium transport across renal and intestinal epithelia1 Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

R. Todd Alexander, Juraj Rievaj, and Henrik Dimke

Abstract: Calcium (Ca2+) is a key constituent in a myriad of physiological processes from intracellular signalling to the mineralization of bone. As a consequence, Ca2+ is maintained within narrow limits when circulating in plasma. This is accomplished via regulated interplay between intestinal absorption, renal tubular reabsorption, and exchange with bone. Many studies have focused on the highly regulated active transcellular transport pathways for Ca2+ from the duodenum of the intestine and the distal nephron of the kidney. However, comparatively little work has examined the molecular constituents creating the paracellular shunt across intestinal and renal epithelium, the transport pathway responsible for the majority of transepithelial Ca2+ flux. More specifically, passive paracellular Ca2+ absorption occurs across the majority of the intestine in addition to the renal proximal tubule and thick ascending limb of Henle's loop. Importantly, recent studies demonstrated that Ca2+ transport through the paracellular shunt is significantly regulated. Therefore, we have summarized the evidence for different modes of paracellular Ca2+ flux across renal and intestinal epithelia and highlighted recent molecular insights into both the mechanism of secondarily active paracellular Ca2+ movement and the identity of claudins that permit the passage of Ca2+ through the tight junction of these epithelia. Key words: kidney, NHE3, claudins, solvent drag. Résumé : Le calcium (Ca2+) est un important constituant d’une myriade de processus physiologiques allant de la signalisation intracellulaire a` la minéralisation de l’os. Conséquemment, le Ca2+ est maintenu a` l’intérieur de limites étroites lorsqu’il circule dans le plasma. Cela s’accomplit grâce a` l’influence réciproque régulée entre l’absorption intestinale, la réabsorption rénale et l’échange avec l’os. Plusieurs études se sont concentrées sur la voie hautement régulée du transport actif trans-cellulaire du Ca2+ du duodénum et du néphron distal. Cependant, en comparaison, peu d’études ont examiné les constituants moléculaires qui créent une dérivation para-cellulaire a` travers l’épithélium intestinal et rénal, la voie de transport responsable de la majorité du flux de Ca2+ trans-épithélial. Plus spécifiquement, l’absorption para-cellulaire passive de Ca2+ survient le long de la plus grande partie de l’intestin en plus du tubule proximal du rein et de la branche ascendante large de l’anse de Henle. Fait important, des études récentes démontrent que le transport de Ca2+ a` travers la dérivation para-cellulaire est sujet a` une importante régulation. Les auteurs ont donc fait la synthèse des indices qui suggèrent qu’il existe différents modes de flux para-cellulaires de Ca2+ a` travers l’épithélium du rein et de l’intestin, et souligné les récentes percées sur les plans du mécanisme moléculaire du mouvement para-cellulaire de Ca2+ secondairement actif et de l’identité des claudines qui permettent le passage du Ca2+ a` travers les jonctions serrées de ces épithéliums. [Traduit par la Rédaction] Mots-clés : rein, NHE3, claudines, déplacement de solvant.

Introduction (Ca2+)

Calcium participates in a diverse array of physiological processes, including neural transmission, muscle contraction, blood coagulation, intracellular signal transduction as a second messenger, and the mineralization of bone as part of the hydroxylapatite crystal. Consequently, the concentration of free ionized Ca2+ is tightly regulated in plasma. Hypocalcemic or hypercalcemic states can cause severe neurological and cardiovascular sequelae. In fact, disturbances in Ca2+ balance may in some instances produce tetany, carpopedal spasms, life-threatening arrhythmias, coma, or cardiac arrest (Bilezikian 1993; Guise and Mundy 1995). Globally, Ca2+ homeostasis is maintained via regulated interplay between 3 organ systems: intestinal absorption of

Ca2+ from the diet, storage, and exchange of Ca2+ with bone, and changes in the renal reabsorptive capacity for Ca2+. These processes effectively stabilize systemic Ca2+ concentrations within normal limits. Parathyroid glands play a central role in sensing changes in serum Ca2+ concentrations and respond by amending secretion of parathyroid hormone (PTH). Increased intestinal absorption or augmented resorption from bone increases serum Ca2+ typically leading to increased levels of Ca2+ in the urine because of compensatory reductions in renal transport capacity (Pak et al. 1975). Similarly, dysregulation of select renal transporters within the kidney results in hypercalciuria. Importantly, hypercalciuria remains the single greatest risk factor for developing a Ca2+-based kidney stone (Levy et al. 1995), a disease with high and

Received 9 May 2014. Revision received 30 August 2014. Accepted 14 September 2014. R.T. Alexander. Department of Pediatrics, The University of Alberta, 4-585 Edmonton Clinic Health Academy, 11405 – 87 Ave, Edmonton, AB T6G 2R7, Canada; Membrane Protein Disease Research Group, The University of Alberta, Edmonton, AB T6G 2H7, Canada. J. Rievaj. Membrane Protein Disease Research Group, The University of Alberta, Edmonton, AB T6G 2H7, Canada; Department of Physiology, The University of Alberta, Edmonton, AB T6G 2H7, Canada. H. Dimke. Department of Cardiovascular and Renal Research, Institute of Molecular Medicine, University of Southern Denmark, Odense, Denmark. Corresponding author: R. Todd Alexander (e-mail: [email protected]). 1This review is part of a Special Issue commemorating The Canadian Society for Molecular Biosciences 57th Annual Meeting – Membrane Proteins in Health and Disease, held in Banff, Alberta, 9–13 April 2014. Biochem. Cell Biol. 92: 467–480 (2014) dx.doi.org/10.1139/bcb-2014-0061

Published at www.nrcresearchpress.com/bcb on 17 September 2014.

468

Biochem. Cell Biol. Vol. 92, 2014

Fig. 1. Modes of transepithelial Ca2+ flux. Ca2+ is absorbed down its electrochemical gradient via passive diffusion (top tight junction) or with water following the osmotic gradient created by solute absorption (often sodium, bottom junction). Transcellular Ca2+ flux (bottom cell) occurs via apical entry through a channel down the electrochemical gradient. Ca2+ then moves to the basolateral membrane bound to a calcium-binding protein and then is extruded across the basolateral membrane via a calcium pump or in exchange for sodium.

VTE

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

Paracellular Diffusion

[Ca2+]

Na+

[Ca2+]

Na+ Na+

Ca2+ H2O

Solvent Drag

H2O

Ca2+

Ca2+ Calb

Ca2+

2+

Ca

Ca2+

Na+

Ca2+

Transcellular Flux Lumen increasing prevalence (Stamatelou et al. 2003; Coe et al. 2005). Therefore, it is important to establish the physiological regulation of these epithelial Ca2+ transport pathways and the molecular constituents driving these processes in an effort to treat or prevent these diseases. Movement of Ca2+ from the lumen of the intestine or renal tubule into the blood can proceed via 1 of 2 pathways: transcellular (i.e., the ion moves through the cell) or paracellular (i.e., the ion is transported between epithelial cells; Fig. 1) (Dimke et al. 2010, 2011). Transcellular Ca2+ absorption has been reported from the duodenum, jejunum, cecum, and proximal colon in the intestine as well as the renal distal convoluted and connecting tubules (Table 1). This process occurs via apical influx of Ca2+ through selective Ca2+ channels, largely predicted to be of the transient receptor potential of the vanilloid subtype (either TRPV5 or 6). Ca2+ is then shuttled to the basolateral membrane typically bound to a Ca2+ binding protein, such as a calbindin, and then secreted across the basolateral membrane via a Ca2+-dependent ATPase or

in exchange for Na+ by a Na+/Ca2+ exchanger (Hoenderop et al. 2005; Dimke et al. 2011). Transcellular Ca2+ absorption from the renal tubule and the intestine is subject to intense regulation, which has been the focus of a large amount of work and the topic of a number of reviews (Hoenderop et al. 2005; Dimke et al. 2010, 2011). Paracellular Ca2+ flux purportedly accounts for the majority of Ca2+ absorption from the intestine and the renal tubule. Although once thought to be a static unregulated process, emerging evidence suggests otherwise. Similar to ion channels, tight junctions are characterized by differing size and charge selectivity, which controls the permeation of ions. Recent evidence has greatly expanded our knowledge about the constituents of the tight junction, which is composed of a family of proteins called the claudins. These proteins determine the paracellular permeability characteristics of a given epithelial type (Simon et al. 1999; Wilcox et al. 2001; Konrad et al. 2006; Hou et al. 2008b; Markov et al. 2010) and regulation of these pore-forming claudins can dramatically Published by NRC Research Press

Alexander et al.

469

Table 1. Type of transintestinal Ca2+ absorptiona reported by segment.

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

Segment

Transcellular flux

Paracellular flux, passive

Paracellular flux, secondarily active b

Duodenum

Yes

Yes

Yes, secretion

Jejunum

Yesc

Yes

Yes, secretion

Ileum

No

Yes

Yes, secretion

Cecum and proximal colon

Yes

Yes

Yes, absorption

Distal colon

Yes

Yes

Yes, secretion

References (Charoenphandhu et al. 2001, 2006; Karbach 1992; Pansu et al. 1983; Tudpor et al. 2008) (Karbach 1992; Karbach and Feldmeier 1993; Pansu et al. 1983; Rievaj et al. 2013; Walling and Kimberg 1973) (Hu et al. 1993; Karbach 1992; Karbach and Feldmeier 1993; Nellans and Kimberg 1979; Pansu et al. 1983; Walling and Kimberg 1973) (Karbach 1992; Karbach and Feldmeier 1993; Karbach and Rummel 1987; Kraidith et al. 2009; Nellans and Goldsmith 1981; Rievaj et al. 2013) (Karbach et al. 1986)

aTranscellular

Ca2+ flux is always absorption, the direction of passive paracellular flux is dependent on the electrochemical gradient, and the direction of secondarily active paracellular Ca2+ flux is segment dependent. bThere is some debate as to whether there is secondarily active Ca2+ absorption from the duodenum. Decreased flux after the apical removal of glucose supports absorption; however, both voltage clamp experiments and analysis of concentration dependence supports a slight asymmetry mediating secretion. cTranscellular flux has been observed across the proximal jejunum of rats.

affect the paracellular permeability characteristics of the aforementioned epithelia (Gong et al. 2012; Dimke et al. 2013; Gong and Hou 2014). This review aims to summarize current knowledge of Ca2+ transport occurring via the paracellular pathway in intestine and kidney. In particular, we highlight the molecular constituents involved in setting the driving forces and permeability characteristics that allow passive paracellular Ca2+ flux across renal and intestinal epithelia. Moreover, we describe how regulated changes in specific tight junction proteins alter paracellular Ca2+ absorption from these epithelia.

Intestine Daily dietary intake of Ca2+ in the Western diet should be approximately 1 g (Bailey et al. 2010), although a minority of this (20%–40% of intake) is actually absorbed (Brine and Johnston 1955; Nordin et al. 1979). Uptake of dietary Ca2+ from the intestinal lumen can occur via active transcellular or passive paracellular transport (Bronner 1998). Active transcellular Ca2+ flux is saturable and predominates when dietary Ca2+ is low. This process permits the uptake of Ca2+ into the enterocyte when low intraluminal Ca2+ concentrations are present (Bronner and Pansu 1999). In contrast, absorption of Ca2+ via the passive paracellular pathway requires a sufficiently high luminal Ca2+ concentration and, hence, adequate dietary Ca2+ intake (Bronner and Pansu 1999); the molecular details of both pathways are depicted in Fig. 2. The small intestine is thought to be responsible for most Ca2+ uptake, with the colon absorbing less than 10% under normal conditions (Marcus and Lengemann 1962a, 1962b; Cramer 1965). The relative amounts of Ca2+ taken up paracellularly in different intestinal segments relate to several variables. The most important are the intraluminal Ca2+ concentration, the solubility of Ca2+, the permeability of the tight junction, and sojourn time (Duflos et al. 1995). Sojourn time is the time that the intestinal content spends transiting each segment. Combined, these factors determine the paracellular transport rate of Ca2+ across a given intestinal segment. The sojourn time varies considerably between individual segments, with very short transit times in the duodenum and much longer times in the distal small intestine, cecum, and colon (Marcus and Lengemann 1962b; Duflos et al. 1995). In rats fed a high Ca2+ diet, sojourn times were measured to be 3 min in duodenum, 43 min in jejunum, 141 min in ileum, 92 min in cecum, and 92 min in colon (Bronner and Pansu 1999). The dissimilarity between sojourn times in segments of the intestine is reflected in the total Ca2+ absorption measured, with 7%–8% of Ca2+ uptake occurring from the duodenum, 4%–17% from the jejunum, and 65%–88% from the ileum (Cramer and Copp 1959; Marcus and

Lengemann 1962a, 1962b; Cramer 1965; Wasserman 2004). Some of these studies measured the disappearance of radiotracers from the intestinal lumen and did not consider Ca2+ flux in the opposite direction (i.e., secretion). Consequently, they likely underestimated the relative contribution of more distal segments; radiotracer absorption from distal segments may be falsely underrepresented due to proximal secretion. Significant distal intestinal Ca2+ absorption has also been observed in humans (Birge et al. 1969). Passive paracellular intestinal Ca2+ absorption Most intestinal Ca2+ absorption, on a normal Ca2+-containing diet, is reported to occur via passive, or secondarily active, paracellular transport (Bronner 2003). This assumption is largely based on in vivo experiments employing the ligated loop technique. The relationship between luminal Ca2+ concentration and rate of Ca2+ absorption obtained by this method can be conceptualized as the sum of 2 processes: a saturable component, which is observed in the duodenum and to a lesser extent the jejunum, and an unsaturable process with linear dependency observed in all segments of the small intestine (Pansu et al. 1983). The saturable process is assumed to represent active transcellular absorption, whereas the unsaturable process is interpreted as passive paracellular absorption (Bronner et al. 1986; Bronner 2003). However, this interpretation is limited by a paucity of data supporting the assumption that the entire unsaturable component occurs in a paracellular fashion; that is, one may envisage transcellular mechanisms that may not display saturable kinetics under the experimental conditions (McCormick 2002; Kellett 2011). Further, as the relative permeability for Ca2+ (per unit of intestinal length) is similar for all segments of the small intestine (Pansu et al. 1983), the amount and direction of passive paracellular Ca2+ flux would be strongly dependent on the electrochemical gradient for Ca2+ across intestinal epithelium. The potential difference across the whole length of the intestine is lumen negative (Geall and Summerskill 1969). Measurements of the transepithelial potential difference in vivo across the small intestine, regardless of segment, found a difference of approximately –5 mV (lumen negative), whereas a transepithelial voltage potential of approximately –20 mV was measured across the large intestine (although measurements of up to –60 mV have been reported) (Geall et al. 1969; Geall and Summerskill 1969). Similar values are reported from Ussing chamber studies, although these reports typically do not exceed –20 mV even for large bowel preparations (Clarke 2009). Consequently, the energy for paracellular Ca2+ absorption must be derived from the chemical concentration gradient. Published by NRC Research Press

470

Biochem. Cell Biol. Vol. 92, 2014

Fig. 2. Intestinal transepithelial Ca2+ flux. The vectorial movement of Na+ creates an osmotic gradient for water reabsorption from the intestine, which can either create a concentration gradient for Ca2+ to diffuse down (top junction) or drive paracellular Ca2+ flux via convection/solvent drag (bottom junction). There is evidence supporting a role for NHE3 in paracellular Ca2+ flux across some intestinal segments and sodium-glucose transport has also been implicated (Charoenphandhu et al. 2001). The claudins (CLDN) implicated in creating paracellular Ca2+ permeability are depicted. In the duodenum and cecum, there is significant transcellular Ca2+ absorption. The players here include apical entry via TRPV6, shuttling to the basolateral membrane via calbindin-D9K (Calb-D9K), and then efflux across the basolateral membrane via the calcium pump, PMCA1b. Note, we have depicted a generic intestinal epithelium; however, in different parts of the intestine, different pathways will function to a greater or lesser extent (please refer to the text and Table 1 for details).

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

paracellular diffusion CLDN 2

CLDN 15?

Ca2+

Ca2+ CLDN CLDN 12 2

Glucose SGLT1 +

Na

Na a+

Na+

Na+

NaK ATPase

Na a+ NHE3

Ca

H+

CLDN 2

2+ H2O

K+

CLDN 15? H2O

solvent drag

CLDN CLDN 12 2

Ca2+ Calb D9K

Ca2+ TRPV6

Ca2+

Ca2+ Ca2+

PMCA1b

transcellular flux

Employing the Nernst equation and assuming a free Ca2+ concentration in the blood of 1.2 mmol/L, one can calculate that the concentration of free Ca2+ at the surface of the small bowel required to overcome a –5 mV transepithelial potential difference must be at least 1.74 mmol/L. Yet, to our knowledge, the free Ca2+ concentration on the apical surface of intestinal epithelium has not been investigated. The free Ca2+ concentration in the whole lumen of different intestinal segments was found to be dependent on the Ca2+ content of ingested food. When food with a “low” Ca2+ content (0.15% w/w) was ingested, the intraluminal free Ca2+ concentration was always observed lower than the concentration in blood throughout the intestine (Sernka and Borle 1969). However, when ingested Ca2+ content was increased to 1%–1.5%, an intraluminal concentration of 1.5–8 mmol/L was observed. For food containing approximately 3% Ca2+, the intraluminal free Ca2+ concentration in the small intestine increased along its course and

peaked at approximately 10–45 mmol/L by the terminal ileum (Cramer 1965; Duflos et al. 1995). This increase in the luminal Ca2+ concentration along the course of the intestine occurs because of Na+ and, consequently, osmotically driven water absorption (Cramer 1964, 1965). This secondary concentration gradient is one purported mechanism connecting paracellular Ca2+ flux to water absorption. The second is solvent drag, which is the movement of Ca2+ together with paracellular water flow via convection (Diamond and Bossert 1967; Pappenheimer and Reiss 1987; Larsen et al. 2000). Distinguishing between a secondary concentration gradient and solvent drag is not easily accomplished because both are considered secondarily active processes and are, in principle, able to move Ca2+ ions against the observed electrochemical gradient. Further, the value of the transintestinal potential difference and luminal concentration of Ca2+ detailed previously Published by NRC Research Press

Alexander et al.

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

are influenced by the methods employed and model systems studied and, therefore, should be considered estimates. Evidence in support of secondary active paracellular intestinal Ca2+ absorption Evidence for the existence of secondary active paracellular Ca2+ flux is largely derived from Ussing chamber studies. This in vitro technique benefits from being able to control both the electrical and chemical transepithelial gradients. The relationship between extracellular Ca2+ concentration in the apical compartment and rate of Ca2+ flux from apical to basolateral side of the epithelium reveals a saturable and a linear unsaturable component to total Ca2+ flux, likely reflecting transcellular and paracellular transport routes, respectively. The unsaturable component contributes significantly to total Ca2+ flux in all parts of the small and large intestines (Karbach et al. 1986; Nellans 1990; Karbach 1992; Karbach and Feldmeier 1993). These data obtained in Ussings chambers are comparable to results obtained by the ligated loop technique described previously. Ussing chamber experiments also permit the measurement of Ca2+ flux in the presence of divergent voltage clamps, permitting the separation of Ca2+ flux into either voltagedependent (presumably paracellular) or voltage-independent (presumably transcellular flux) components (Frizzell and Schultz 1972). A detailed discussion of the merits of these different techniques is beyond the scope of this review. Importantly, fluxes assumed to be “paracellular” by the aforementioned techniques are not always equivalent when measured in opposite directions (i.e., apical to basolateral vs. basolateral to apical) even in the absence of an electrochemical gradient. This difference favors paracellular Ca2+ absorption (i.e., higher apical to basolateral than basolateral to apical flux) from the cecum and ascending colon and paracellular Ca2+ secretion from all segments of the small intestine and the descending colon. This net paracellular Ca2+ movement has been attributed to solvent drag (Nellans 1990; Karbach and Feldmeier 1993; Rievaj et al. 2013). Because water flux is assumed (but rarely measured under the same experimental conditions) to be in the apical to basolateral direction, the term “anomalous solvent drag” was coined to explain paracellular Ca2+ secretion from the small intestine (Nellans and Kimberg 1979). However, data from Ussing chamber studies measure much higher fluxes in the basolateral to apical direction than what is generally assumed from in vivo studies. This led to the reporting of net Ca2+ secretion in the jejunum and no net transport from the duodenum when symmetrical Ca2+ concentrations were employed in Ussing chamber experiments under voltage clamp (i.e., in the absence of an electrochemical gradient) (Walling and Kimberg 1973, 1975; Karbach and Feldmeier 1993; Rievaj et al. 2013). Although net Ca2+ secretion has also been observed during in vivo studies, this always occurred down its electrochemical gradient (Krawitt and Schedl 1968; Sernka and Borle 1969). We can only speculate about the reasons for the discrepancy in measurement between methodologies, with possible explanations including selective damage to water absorptive cells during prolonged in vitro measurements (Inagaki et al. 2005; Inagaki-Tachibana et al. 2008a, 2008b), the presence of substances in vivo which prevent anomalous solvent drag such as bile salts (Hu et al. 1993), or the presence of greater unstirred layers on the basolateral side of epithelia leading to underestimates of apical to basolateral flux in Ussing chamber studies, especially when performed on unstripped epithelium. The other way to estimate the involvement of secondarily active paracellular Ca2+ absorption is by altering the driving force of water movement. All intestinal segments demonstrate both water absorptive and water secretive properties. Under physiological conditions, water absorption prevails. Water can move across an epithelium via either a transcellular or paracellular pathway. Although both mechanisms of water flux could generate a second-

471

ary concentration gradient for Ca2+ absorption, paracellular water movement is necessary for solvent drag. However, the relative contribution of each pathway to water absorption is unknown. Although several aquaporins have been identified in the epithelia of small and large intestines (Laforenza 2012), their role in transepithelial water flux has not been examined with the exception of aquaporin-4. This water channel plays a minor role in colonic water absorption, as evinced by a knockout animal that has mildly increased water content in defecated stool (Wang et al. 2000). Similarly, little is known about paracellular intestinal water movement. Although the tight junction protein claudin-2 (Cldn2) has been implicated in mediating paracellular water flux across a renal epithelial cell line (Rosenthal et al. 2010) and Cldn2 is expressed in mouse intestine (Fujita et al. 2008), Cldn2 knockout mice do not display a phenotype consistent with decreased intestinal water absorption (Tamura et al. 2011). Regardless of the pathway, water movement passively follows the movement of ions (Curran and Macintosh 1962). Consequently, water secretion has been attributed to electrogenic anion secretion through apical anion channels, whereas water absorption is coupled to epithelial Na+ uptake. The mechanisms mediating Na+ uptake can be divided into electroneutral and electrogenic transport. Electrogenic transport occurs via apical Na+ channels expressed mainly in the distal colon (Garty and Palmer 1997) and Na+/nutrient-linked cotransporters, in particular Na+-dependent glucose cotransporters (SGLTs) expressed in the small intestine. Electroneutral Na+ absorption occurs through the apically expressed Na+/H+ exchangers, where NHE3 appears to play the predominant role in Na+ absorption from both the small and large intestine (Schultheis et al. 1998; Gawenis et al. 2002). Na+ efflux occurs across the basolateral membrane via the Na+/K+ ATPase for both pathways. The exact mechanism coupling water flux to sodium transport is still debated. Among multiple different theories, the revised standing gradient model (Diamond and Bossert 1967) and the Na+ recirculation theory, which is summarized in detail in Larsen and Mobjerg (2006) and Larsen et al. (2007), are probably the most developed. Both theories are based on the observation that Na+/K+ ATPase is predominantly localized to the lateral membrane of intestinal epithelial cells (DiBona and Mills 1979). This lateral localization of the Na+/K+ ATPase and the vastly different resistance between the tight junction and the basal junction would generate a hyperosmotic and hyperbaric lateral intercellular space, which would provide the conditions for fluid uptake in the absence of an electrochemical gradient or even against a concentration gradient. Additionally, the Na+ recirculation theory assumes influx of Na+ from the basolateral membrane and hence “recirculation” of Na+. Although this would be energetically unfavourable for a tissue, there is experimental evidence consistent with the existence of this phenomenon (Nedergaard et al. 1999). Given the dependence of secondarily active paracellular intestinal Ca2+ absorption on water flux, which is driven by intestinal Na+ absorption, it can be hypothesized that molecules implicit to intestinal Na+ absorption would also play a role in intestinal Ca2+ absorption. This was tested by removal of glucose from the mucosal side of intestinal epithelium, thereby inhibiting Na+-coupled glucose transport by depriving the cotransporter of substrate (Charoenphandhu et al. 2001). However, this method has been criticized due to its effect on the polarity of the apical membrane (Kellett 2011). To test whether electroneutral NHE3-mediated Na+ flux contributes to intestinal Ca2+ uptake, we measured intestinal 45Ca2+ uptake after gastric gavage in wild-type and Nhe3 knockout mice (Pan et al. 2012). Importantly, Ca2+ uptake from the intestine was reduced in Nhe3−/− mice. Direct measurements of Ca2+ flux across duodenum and cecum (the 2 parts of intestine with known net Ca2+ absorption) in Ussing chambers confirmed decreased luminal to basolateral transepithelial Ca2+ flux (Pan et al. 2012; Rievaj et al. 2013), confirming that NHE3 contributes to paracelluPublished by NRC Research Press

472

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

lar Ca2+ uptake. Finally, to avoid potential compensatory changes resulting from an altered Ca2+ balance in Nhe3−/− mice, we confirmed the findings using a NHE3 inhibitor in cecum of wild-type mice (Rievaj et al. 2013). Together these data support a role for NHE3 in transepithelial intestinal Ca2+ absorption, likely via driving paracellular Ca2+ flux either via convection or by allowing water removal and thereby generating a favourable concentration gradient for Ca2+ to follow. Molecules setting the paracellular permselectivity across intestinal epithelia For paracellular ion flux to occur, there must be both a driving force and a Ca2+-permeable pore. The discovery and investigation of a family of tight junction proteins, termed claudins, has contributed significantly to our understanding of paracellular pores. Claudins are 4-pass transmembrane proteins with both the amino and carboxy termini resident in the cytosol. They form homomeric and heteromeric interactions with one another in the same tight junction and in the tight junction of the adjacent epithelial cell. Although the structure of this complex is poorly understood, it clearly plays a role in determining the permeability of the tight junction (for recent reviews see Gunzel and Fromm (2012) and Gunzel and Yu (2013)). Evidence supports a role for Cldn2 and Cldn12 in creating cation-permeable paracellular pores, hence permitting Ca2+ flux between intestinal epithelial cells (Fujita et al. 2008). Cldn2, when overexpressed in epithelial cell culture models, creates a paracellular cation-selective pore permeable to Ca2+ (Furuse et al. 2001; Amasheh et al. 2002; Fujita et al. 2008; Yu et al. 2009, 2010). Examination of the Cldn2 knockout mouse confirmed that it forms a paracellular cation-selective pore across the intestine and renal proximal tubule (Muto et al. 2010; Tamura et al. 2011). However, direct Ca2+ flux studies have not been reported from Cldn2−/− mice to date. Overexpression and siRNA-mediated knockdown of Cldn12 in intestinal epithelial cells demonstrate that this claudin also creates a cation-selective pore permeable to Ca2+ (Fujita et al. 2008). However, the phenotype of a Cldn12 knockout animal has not been reported. Overexpression studies also demonstrate the ability for Cldn15 to form a paracellular cation-selective pore (Colegio et al. 2002; Van Itallie et al. 2003). Interestingly, deletion of this gene in mice leads to the development of megaintestine (Tamura et al. 2011). The knockout mice also display decreased intestinal Na+ and water absorption, which appears exacerbated in the Cldn2 and Cldn15 double knockout (Wada et al. 2013). Given these findings, it is tempting to speculate that Cldn15 also contributes to the formation of cation-selective paracellular pores permeable to Ca2+ along the course of the intestine. Again, direct measurements of Ca2+ flux across the intestine of Cldn15 knockout mice have not been reported. The localization of Cldn2, Cldn12, and Cldn15 has been examined along the course of the mouse intestine (Fujita et al. 2006, 2008). Consistent with a role in paracellular transport, Cldn2 and Cldn12 expression is greatest in the distal segments of the small bowel. This expression pattern differs slightly in rats; however, significant expression of both isoforms is seen in the distal small bowel (Markov et al. 2010). Expression of Cldn15 in mice is greater in more proximal intestinal segments, although there is significant expression along the course of the bowel consistent with a possible role in paracellular Ca2+ reabsorption (Fujita et al. 2006). The expression of CLDN12 and CLDN15 has been observed along the course of small and large bowel in humans, although CLDN2 has only been reported in the small bowel as summarized in Lu et al. (2013). Regulation of paracellular Ca2+ absorption in intestine The calciotropic hormone 1,25-dihydroxyvitamin D3 acts to increase intestinal Ca2+ absorption. Much research has focused on how 1,25-dihydroxyvitamin D3 increases the expression and activ-

Biochem. Cell Biol. Vol. 92, 2014

ity of channels and transporters that govern transcellular Ca2+ reabsorption as recently reviewed in Fleet and Schoch (2010) and Christakos (2012). A role for 1,25-dihydroxyvitamin D3 in regulating paracellular Ca2+ flux has been debated; however, recent evidence is consistent with 1,25-dihydroxyvitamin D3 also increasing paracellular Ca2+ flux (Wasserman 2004; Fujita et al. 2008). In-line with a role for 1,25-dihydroxyvitamin D3 in altering paracellular Ca2+ flux, the expression of Cldn2 and Cldn12 was reduced in vitamin D–receptor knockout mice (Fujita et al. 2008). Moreover, treatment of the intestinal Caco-2 cell line with 1,25-dihydroxyvitamin D3 increased Cldn2 and Cldn12 expression, and increased paracellular Ca2+ flux (Fujita et al. 2008). Further, it has been speculated that increased 1,25-dihydroxyvitamin D3–responsive paracellular Ca2+ flux compensates, at least in part, for decreased transepithelial Ca2+ flux in TRPV6 and calbindin-D9K double knockout mice (Christakos et al. 2010). Although this discussion is consistent with the possibility of claudin pores contributing to regulated paracellular Ca2+ absorption from the intestine, much work still needs to be done to firmly establish this. This includes direct measurements of Ca2+ flux across the intestine of claudin-specific knockout animals. Active transcellular Ca2+ absorption occurs from both the duodenum and proximal large bowel Active transcellular Ca2+ absorption is predominantly studied in duodenum, whereas active paracellular Ca2+ absorption has only been observed in the cecum and proximal colon. The cecum and proximal colon are also sites of significant transcellular Ca2+ absorption. As in the duodenum, the same transcellular Ca2+ absorption machinery is present and measurements in Ussing chambers have confirmed both significant transcellular and secondarily active paracellular Ca2+ flux across these segments, in rodents at least (Nellans and Goldsmith 1981; Karbach and Feldmeier 1993; Kraidith et al. 2009; Zhang et al. 2010; Rievaj et al. 2013). Consistent with a role of proximal large bowel in Ca2+ homeostasis, resection of the cecum in rats leads to decreased bone mineral density secondary to enhanced trabecular bone resorption (Charoenphandhu et al. 2012; Jongwattanapisan et al. 2012). These results are not likely directly translatable to humans given the relatively larger size of the cecum in rodents vs. the cecum in humans (Casteleyn et al. 2010). The role of the proximal large bowel in Ca2+ homeostasis in humans has been relatively unexplored; we are aware of only a single study directly confirming active transcellular flux from this segment after vitamin D3 administration (Grinstead et al. 1984). Kinetic studies suggest that under physiological conditions, there is a small but measurable contribution of colon/cecum to total Ca2+ absorption (Barger-Lux et al. 1989). Thus, the large intestine seems to provide a reserve, which can be used in situations when absorption from the small intestine is inadequate (Hylander et al. 1980, 1990; Abrams et al. 2007). Whether increased Ca2+ absorption from the cecum/colon plays a role in situations of increased Ca2+ requirements, such as childhood growth, pregnancy, lactation, or when Ca2+ intake is reduced, remains to be determined. Given the frequency of colonic resection, it would be important to determine the effect of this procedure on Ca2+ homeostasis in humans.

Kidney Ca2+ filtered by the kidney is reclaimed by the nephron to maintain serum levels within normal limits. Like the intestine, the majority of Ca2+ is reabsorbed via passive paracellular processes. The bulk, approximately 70% of filtered Ca2+, is reabsorbed in this fashion from the proximal tubule and approximately 25% via passive paracellular fluxes from the thick ascending limb. The remainder of Ca2+ reabsorption occurs via active transcellular flux from the distal convoluted and connecting tubules. This latter process appears to occur via an analogous process in the duodenum and cecum. Here, TRPV5 channels mediate apical uptake of Published by NRC Research Press

Alexander et al.

473

Fig. 3. Proximal tubular Ca2+ reabsorption. The majority of Ca2+ flux across the proximal tubule occurs in a paracellular fashion, driven by the transepithelial flux of sodium (Na+), which in turn drives water (H2O) flux. Water removal from the proximal tubular lumen could act to concentrate Ca2+ providing a concentration gradient (bottom junction) or paracellular water movement may drive Ca2+ flux via convection (i.e., solvent drag) (top junction). The epithelial Na+/H+ exchanger (NHE3) is the primary Na+ influx pathway, whose activity is driven by the Na+/K+-ATPase (NaK ATPase). Note that two-thirds of water flux across the proximal tubule occurs in a transcellular fashion mediated by aquaporin-1 pores (not depicted here). There is also significant paracellular Na+ flux.

CLDN 2

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 12/02/14 For personal use only.

2+

Ca H O 2

H2O CLDN 2

Na+ NHE3

Na+

H+

Na+ NaK ATPase

K+

CLDN 2

Ca2+

Ca2+ CLDN 2

the ion, with calbindin-D28K buffering intracellular Ca2+ and extrusion across the basolateral membrane occurs via the plasma membrane Ca2+-dependent ATPase and the Na+/Ca2+ exchanger (NCX) (Hoenderop et al. 2005; Dimke et al. 2010, 2011). The remainder of the review will focus on the mechanisms mediating Ca2+ flux across the proximal tubule and thick ascending limb. Proximal tubule Most (approximately 55%–60%) Ca2+ filtered by the glomerulus is reabsorbed by the proximal convoluted tubule (Duarte and Watson 1967; Suki 1979). A further 10% of filtered Ca2+ is reabsorbed from the straight portion of the proximal tubule (Suki 1979; Seldin 1999). Therefore, most (i.e., at least two-thirds) of Ca2+ in the ultrafiltrate is reclaimed by the proximal tubule. Evidence from micropuncture studies performed on multiple species is consistent with Ca2+ uptake from this nephron segment occurring in a passive paracellular fashion (Duarte and Watson 1967; Morel et al. 1969; Sutton and Dirks 1975). The ratio of Ca2+ concentration in tubular fluid to the ultrafiltrate, [(TF/UF)Ca], is consistently reported to be between 1.0 and 1.2. This ratio was also consistently similar, if not identical, to the ratio for Na+, inferring that active Na+ reabsorption provides the driving force for Ca2+ reabsorption. Consistent with this, attempts to uncouple proximal Na+ reabsorption from Ca2+ reabsorption by the administration of parathyroid hormone, furosemide, hydrochlorothiazide, or acetazolamide were unsuccessful (Agus et al. 1973; Beck and Goldberg 1973; Edwards et al. 1973; Sutton and Dirks 1975). That Ca2+ reabsorption from the proximal tubule occurs in a paracellular fashion is consistent with microperfusion studies of the proximal tubule in the absence of an electrochemical gradient, which demonstrate predominant passive paracellular flux (Ng et al. 1984). These authors did detect a small (

Paracellular calcium transport across renal and intestinal epithelia.

Calcium (Ca(2+)) is a key constituent in a myriad of physiological processes from intracellular signalling to the mineralization of bone. As a consequ...
811KB Sizes 0 Downloads 5 Views