The A563T variation of the renal epithelial calcium channel TRPV5 among African Americans enhances calcium influx

Abstract

The transient receptor potential cation channel, subfamily V, member 5 (TRPV5) gene, which encodes the Ca²⁺ channel in the apical membrane of distal convoluted tubule and connecting tubule of the kidney, exhibits an unusually high frequency of nonsynonymous single nucleotide polymorphisms (SNPs) among African Americans. To assess the functional impacts of the nonsynonymous SNP variations in TRPV5, these variants were analyzed with radiotracer ⁴⁵Ca²⁺ influx assay and the voltage-clamp technique using Xenopus laevis oocytes. Among the variations tested, including A8V, R154H, A563T, and L712F, the latter two significantly increased TRPV5-mediated Ca²⁺ influx. The A563T variant, which exists in African Americans with relative high frequency, exhibited increased Ca²⁺ influx at extracellular Ca²⁺ from 0.01 to 2 mM despite a lower expression level at the plasma membrane. This variant also exhibited a reduction in Na⁺ current as a result of increased sensitivity to extracellular Mg²⁺. By substituting threonine-563 (Thr⁵⁶³) with serine or valine residue, the bulky side chain of Thr⁵⁶³ was shown to facilitate Ca²⁺ transport, whereas the hydroxyl group of Thr⁵⁶³ is likely related to Mg²⁺ sensitivity. The A563T variant was capable of increasing TRPV5-mediated Ca²⁺ influx, even when it was expressed under conditions mimicking heterozygous or compound state with other variants. In conclusion, the A563T variant of TRPV5 significantly increased Ca²⁺ influx by affecting the Ca²⁺ permeation pathway. Thus the A563T variation in TRPV5 may contribute to the superior ability of renal Ca²⁺ conservation in African Americans.

ca²⁺ balance of the body is maintained by intestinal absorption, renal reabsorption and excretion, as well as disposition of Ca²⁺ in and mobilization from the bone. The distal convoluted tubule (DCT) and connecting tubule (CNT) are critical segments of the nephron for regulating Ca²⁺ reabsorption (21, 30). Ca²⁺ is actively reabsorbed in this portion of the nephron by a vitamin D-regulated transcellular pathway. The renal epithelial Ca²⁺ channel transient receptor potential cation channel, subfamily V, member 5 (TRPV5) is expressed in the apical membrane of DCT and CNT and plays a gatekeeper role for active Ca²⁺ reabsorption (15, 16, 27). Mice lacking Trpv5 resulted in a 6- to 10-fold increase in urinary Ca²⁺ excretion, and ultimately in defects in bone mineralization (16).

Urinary Ca²⁺ excretion is an important factor for kidney stone formation and bone health. For instance, urine Ca²⁺ excretion correlates with bone loss in calcium-stone-forming patients with idiopathic hypercalciuria (3). Interestingly, African Americans exhibit lower urinary Ca²⁺ excretion than whites (8, 25, 29, 34, 35, 40, 45), and the risk of kidney stones in African American is lower than that in whites (32, 36, 38). Furthermore, African Americans have higher bone mass (5, 41) and lower incidence of osteoporosis-related fractures than whites (4, 6, 7). The mechanism underlying the lowered urinary Ca²⁺ in African Americans is not well understood. A lower 25-hydroxyvitamin D concentration in African Americans could result in an increased parathyroid hormone level and in turn renal Ca²⁺ conservation. However, Braun and colleagues (8) observed significantly lower urinary Ca²⁺ excretion in adolescent African American girls than white girls in a wide range of controlled Ca²⁺ intake, whereas the 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D and parathyroid hormone values were not significantly different between the two groups. The difference in bone mass and urinary Ca²⁺ excretion between blacks and whites is not restricted to Americans; a similar finding was reported in South Africa (10). Thus genetic rather than social and environmental factors may play a major role in the superior renal Ca²⁺ conservation mechanism in African descendents.

Because TRPV5 is a key protein that regulates Ca²⁺ reabsorption, genetic variations of TRPV5 may influence urinary Ca²⁺ excretion. Interestingly, TRPV5 gene is one of the four contiguous genes, including EPHB6, TRPV6, TRPV5, and KEL, in chromosome 7q34-35 with striking evidence of a recent selective sweep in European Americans based on the analysis on the single nucleotide polymorphisms (SNPs) of over 100 genes from 24 African Americans and 23 European Americans provided by SeattleSNPs program (1, 37). Akey and colleagues (2) further showed that the TRPV6 haplotype, defined by three nonsynonymous SNPs, is nearly fixed in populations outside Africa, suggesting that these variations may confer a selective advantage, e.g., efficiency in Ca²⁺ absorption from dairy products or resistance to a pathogen, after early humans migrated out of Africa. The TRPV6 variant with the three nonsynonymous SNPs exhibited increased Ca²⁺ transport ability and may play a role in absorptive hypercalciuria and kidney stone formation (39). However, TRPV5 variations were not investigated in these studies. According to the data from SeatleSNPs, there are four nonsynonymous SNPs in TRPV5 gene; three of them are only present in African Americans. Eleven of 24 African Americans (45.8%) surveyed carry one or more of these African American specific nonsynonymous SNPs. Although further population studies are needed to confirm these observations, it is likely that these nonsynonymous SNPs in TRPV5 are common in African Americans. A large portion of African descents may be affected if these nonsynonymous SNPs have functional consequences. Thus functional characterization of these African American specific nonsynonymous SNPs in TRPV5 gene may help in understanding their potential impacts on Ca²⁺ reabsorption among African descents.

In the present study, we analyzed the functional significance of the four nonsynonymous SNPs of TRPV5 gene using the Xenopus laevis oocyte expression system. Two of them, A563T and L712F, exhibited significant increases in Ca²⁺ transport ability that may contribute to the lower level of urinary Ca²⁺ excretion in African Americans.

MATERIALS AND METHODS

cDNA constructs.

Human TRPV5 cDNA was prepared as described previously (26). For experiments using X. laevis oocytes, the cDNA was subcloned into the X. laevis oocyte expression vector pNWP as previously described (19). TRPV5 SNP variants and other mutations at amino acid 563 of TRPV5 were generated using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) following the manufacturer's instruction. Constructs containing double variations (A8V-R154H, R154H-A563T, R154H-L712F) were constructed by substitution of the EcoR V-Nde I fragment containing R154H, with the same fragment from constructs with A8V, A563T, or L712F variation. All of the constructs containing variations generated with the QuikChange site-directed mutagenesis kit were confirmed by sequencing.

Ca²⁺ uptake in X. laevis oocytes and determination of the concentration dependence of Ca²⁺ uptake and Mg²⁺ sensitivity.

In vitro transcription, injection of the capped synthetic RNAs (cRNAs) into oocytes, and Ca²⁺ uptake assay in oocytes were performed as described previously (19, 28). The animal protocol was approved by the Institutional Animal Care and Use Committee of the University of Alabama at Birmingham. cRNAs for TRPV5 and its different mutations were injected at 12.5 ng/oocyte. When a combination of two cRNAs was required, the cRNAs were mixed, and the total concentration of the cRNAs was maintained.

For measuring the concentration dependence of Ca²⁺ uptake, groups of control oocytes and oocytes expressing TRPV5 (A⁵⁶³) or A563T (T⁵⁶³) were cultured at 18°C in 0.5× L-15 medium (Invitrogen, Carlsbad, CA) supplemented with 10 mM HEPES (pH 7.6), 5% heat-inactivated horse serum, penicillin (10,000 U/l), streptomycin (10 mg/l), and amphotericin B (25 μg/l) for 2 days after cRNA injection. Each group of seven to nine oocytes was washed three times with nonradioactive standard uptake solution with different Ca²⁺ concentrations ([Ca²⁺]) before uptake experiments. Ca²⁺ uptake was initiated by immersion of the oocytes in different [Ca²⁺] (0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1, and 2 mM) with ⁴⁵Ca²⁺ tracer in standard uptake solution that contained (in mM) 100 NaCl, 2 KCl, 1 MgCl₂, and 10 mM HEPES (pH 7.5). After a 30-min incubation at room temperature (∼24°C), oocytes were washed six times with ice-cold standard uptake solution without ⁴⁵Ca²⁺ and were dissolved by 10% SDS solution, and the incorporated ⁴⁵Ca²⁺ was determined using a scintillation counter. Data of 14 oocytes in each group from two frogs were expressed as means ± SE. The kinetic parameters of concentration dependence of Ca²⁺ uptake were determined by fitting the Ca²⁺ uptake data to the equation V = V_maxS/(K_m + S) + NS, where V is the Ca²⁺ transport, V_max is the derived saturable Ca²⁺ transport maximum, S is the [Ca²⁺], K_m is the [Ca²⁺] at which saturable Ca²⁺ transport is half-maximal, and N is the slope of the nonsaturable Ca²⁺ transport. Parameter values (V_max, K_m, and N) were obtained using Graphpad Prism 5.0 software (La Jolla, CA). For measuring Mg²⁺ sensitivity, the standard uptake solution with [Ca²⁺] at 1 mM (including ⁴⁵Ca²⁺ tracer) was used. Oocytes were washed with standard uptake solution without ⁴⁵Ca²⁺ tracer and Mg²⁺ three times before adding uptake solution with varied Mg²⁺ concentration (0, 0.5, 1.0, 2.0, 5.0, or 10 mM). Data of three independent experiments with a total of 21 oocytes in each group were expressed as means ± SE.

Two-microelectrode voltage-clamp technique.

The two-microelectrode voltage-clamp experiments were performed as described previously (19, 28) with a GeneClamp 500 amplifier and pCLAMP software (version 9) (Axon Instruments, Foster City, CA). The resistance of microelectrodes filled with 3 M KCl was 0.5–2 MΩ. In experiments involving voltage jumps, the oocyte was clamped at the holding potential of −50 mV. Voltage pulses (100 ms) between −160 and +60 mV, in increments of 20 mV, were then applied, and steady-state currents were obtained as the average values in the interval from 60 to 95 ms after the initiation of the voltage pulses. The standard perfusion solution used contained (in mM) 100 choline chloride, 2 KCl, 1 MgCl₂, and 10 HEPES, pH 7.5 (adjusted using Tris base and HCl). Choline chloride was substituted with NaCl when Na⁺ current was tested. CaCl₂ (1 mM) was added to the standard perfusion solution when Ca²⁺-evoked current was recorded. For determining Mg²⁺ dose-dependent effect on Na⁺ current, oocytes need to be exposed to divalent cation free standard perfusion solution. Under this condition, the endogenous Ca²⁺-inactivated Cl⁻ current (44), which is likely formed by connexin38 hemi-gap-junctional channels (13), was induced. To completely block the Ca²⁺-inactivated Cl⁻ current in X. leavis oocyte, flufenamic acid at 200 μM was included in solutions (44). Flufenamic acid at this concentration has negligible effects on TRPV5-mediated Na⁺ current and Ca²⁺ uptake. Na⁺ currents were recorded in the presence of varied concentrations of Mg²⁺ (0, 0.1, 0.2, 0.5, 1, 2, and 5 mM) in response to voltage pulses between −160 and +60 mV, in increments of 20 mV. Net Na⁺ currents were obtained by subtracting adjacent recordings in the absence of Na⁺ (choline substitution).

Western blotting analysis.

For detection of TRPV5, oocytes were lysed in extraction buffer (100 mM NaCl, 1% Triton X-100, 20 mM Tris·HCl, pH 7.6) plus protease inhibitor cocktail and centrifuged at 5,000 g for 15 min to remove the cellular debris and yolk proteins. In general, extract supernatant corresponding to one oocyte was loaded per lane. After SDS-PAGE, the proteins were transferred to polyvinylidene difluoride membranes. The membrane was blocked with Tris-buffered saline (TBS) containing 5% nonfat milk for 1∼2 h at room temperature. Primary antibodies (TRPV5; Alpha Diagnostic Intl, CAT21-A) were incubated overnight at 4°C (1:3,000), followed by multiple washes in TBS-Tween 20. The appropriate horseradish peroxidase-conjugated secondary antibodies (1:3,000; Pierce Biotechnology, Rockford, IL) were incubated in the blocking solution for 1 h at room temperature, followed by multiple washes with TBS-Tween 20. Chemiluminescence was detected using a SuperSignal West Femto Maximum Sensitivity Substrate kit (34095; Pierce) in accordance with the manufacturer's protocol.

Oocyte surface biotinylation.

After cRNA injection (2 days), oocytes were washed with cold PBS three times and incubated for 15 min at room temperature in PBS with 0.005% subtilisin A (Sigma-Aldrich, St. Louis, MO) under mild end-to-end shaking to partially digest the vitelline membranes of oocytes. The oocytes were then incubated with Sulfo-NHS-SS-Biotin (1 mg/ml; Pierce) in PBS for 1 h at 4°C with end-to-end shaking. After incubation, oocytes were washed five times with cold PBS and then quenched with 100 mM glycine for 1 h at 4°C with end-to-end shaking. Oocytes were lysed, and the yolk was removed by centrifugation at 5,000 g for 15 min at 4°C. The supernatant was incubated with immobilized NeutrAvidin beads (Pierce) for 2 h at 4°C. The beads were centrifuged, and the supernatants were removed. The beads were then washed three times with TBS. Biotinylated proteins were eluted from the beads by 1× SDS-PAGE loading buffer with 50 mM dithiothreitol at 65°C for 10 min. Input cell lysates and biotinylated sample were subjected to SDS-PAGE and Western blot analysis as described above.

RESULTS

Nonsynonymous SNPs in TRPV5 gene prevalently exist in African Americans.

TRPV5 gene is one of the four genes in a contiguous region of chromosome 7 that exhibits evidence of selective sweep in recent human history by Akey and colleagues (1). Similar findings were reported by Stajich and Hahn (37). These studies utilized SNP genotype data for 132 (1) or 151 (37) genes sequenced in 24 African Americans and 23 European Americans from the SeattleSNPs program (9). Akey and colleagues further demonstrated the allele frequencies of the 3 nonsynonymous SNP variations of TRPV6 in 1,064 individuals from 52 populations, suggesting that the derived TRPV6 allele, which is nearly fixed in populations outside Africa, conferred a selective advantage after early humans migrated out of Africa (2). Although detailed TRPV5 nonsynonymous SNP data were not provided in these publications, they are available to public from the SeattleSNPs website (http://pga.mbt.washington.edu). Four nonsynonymous SNPs were identified by SeattleSNPs; they are summarized in Table 1. Interestingly, out of the four nonsynonymous SNPs, three of them (A8V, A563T, and L712F) were only present in African Americans, not in European Americans. A8V, R154H, and L712F are localized in the intracellular terminal regions; A563T, however, is in the last transmembrane domain (Fig. 1A).

Table 1.

Nonsynonymous SNPs in TRPV5 identified by SeattleSNPs

dbSNP rs No.	dbSNP Allele	Amino Acid Residue	Amino Acid Position	AA Freq	EA Freq
rs4252372	T	Val (V)	8	0.12	0.00
	C	Ala (A)		0.88	1.00
rs4236480	A	His (H)	154	0.44	0.28
	G	Arg (R)		0.56	0.72
rs4252499	A	Thr (T)	563	0.17	0.00
	G	Ala (A)		0.83	1.00
rs4252509	C	Phe (F)	712	0.02	0.00
		G	Leu (L)		0.98	1.00

SNP, single nucleotide polymorphism; dbSNP, Single Nucleotide Polymorphism database; rs, Reference SNP ID; TRPV5, transient receptor potential cation channel, subfamily V, member 5; AA and EA Freq, allele frequency in African Americans (European Americans).

Fig. 1.

Transient receptor potential cation channel, subfamily V, member 5 (TRPV5) nonsynonymous single nucleotide polymorphisms (SNPs) revealed by SeatleSNPs Project. A: position of nonsynonymous SNP variations in TRPV5 model. The cylinders represent transmembrane domains; the ellipses represent ankyrin repeats in the intracellularly localized NH₂-terminal region of TRPV5. B: alignment of amino acid sequences surrounding the nonsynonymous SNPs of TRPV5 from various species.

The variations in TRPV5 are unique, since they are not commonly present in other species surveyed, including chimpanzee, canine, rat, and mouse, with the exceptions of Thr⁵⁶³ in dog and Val⁸ in rat (Fig. 1B). This is in striking contrast to the nonsynonymous SNP variations in TRPV6, which are conserved among these species (not shown).

The A563T variation significantly increases TRPV5-mediated Ca²⁺ transport activity.

We next evaluated the functional impacts of these newly derived variations in TRPV5, including A8V, R154H, A563T, and L712F, using the X. laevis oocyte expression system. Among the TRPV5 variants evaluated, A563T and L712F exhibited a 54.1 ± 5.1% (n = 9) and 26.2 ± 4.5% (n = 9) increase in Ca²⁺ uptake, respectively, over the reference form of TRPV5 (Fig. 2, top). No significant changes were observed for the A8V or R154H variant, although the latter exhibited a slightly lower activity. With the exception of A563T, which exhibited a decrease in the fully glycosylated band (also see Fig. 3), the total protein expression levels of the other three variants were not significantly different from the reference TRPV5 as determined by Western blotting analysis using antibody against TRPV5 (Fig. 2A, bottom). Because the L712F variant only promoted a moderate increase in TRPV5-mediated Ca²⁺ uptake, and it existed only in one allele among 23 European and 24 African Americans examined in the SeatleSNPs program, we focused mainly on the difference between A563T variant and the reference form of TRPV5 in the following studies. For simplicity, when only these two forms of TRPV5 were compared, they were referred to as T⁵⁶³ and A⁵⁶³, respectively.

Fig. 2.

The A563T variant of TRPV5 increased Ca²⁺ influx. A: Ca²⁺ influx mediated by TRPV5 nonsynonymous SNP variants. Xenopus laevis oocytes were injected with 12.5 ng of cRNA of TRPV5 and its nonsynonymous SNP variants (A8V, R154H, A563T, and L712F), and radiotracer ⁴⁵Ca²⁺ uptake and protein expression level were determined 2 days later. A representative Western blotting result using antibody against TRPV5 is shown at bottom. Ca²⁺ uptake data from 63 oocytes in each group from 9 independent experiments were pooled and are shown as means ± SE. *P <0.01 vs. TRPV5 group by Student's t-test. B: concentration dependence of Ca²⁺ influx mediated by TRPV5 and the A563T variant. Data at each point are shown as means ± SE of 14 oocytes from 2 frogs.

Fig. 3.

The T⁵⁶³ variant exhibited a decreased plasma membrane expression level relative to the A⁵⁶³ variant. A: the scar-like mark was more prominent in X. laevis oocytes expressing the T⁵⁶³ variant than those expressing the A⁵⁶³ variant. Shown are representative X. laevis oocytes 18 h after injection with cRNA of T⁵⁶³, A⁵⁶³, or water (as a control). B: biotinylated TRPV5 protein level was decreased in oocytes expressing T⁵⁶³ than those expressing A⁵⁶³, as shown in a representative Western blot analysis. X. laevis oocytes were injected with T⁵⁶³ or A⁵⁶³ cRNA at 12.5 ng/oocyte, and biotinylation experiments were performed 2 days later. Biotinylated proteins corresponding to 5 oocytes and total proteins corresponding to of an oocyte were loaded. C: intensity ratio between T⁵⁶³ and A⁵⁶³ variant in core-glycosylated and fully glycosylated bands (Fig. 3B, bands A and B, respectively). Data from 8 independent experiments are shown as means ± SE.

Because [Ca²⁺] gradually decreases along the distal tubule (43), we tested the ability of T⁵⁶³ to increase Ca²⁺ uptake at various extracellular [Ca²⁺]. T⁵⁶³ appeared to be able to increase Ca²⁺ uptake at all [Ca²⁺] tested from 0.01 to 2 mM. TRPV5-mediated Ca²⁺ uptake exhibited a Michaelis-Menten type of saturation kinetics at extracellular [Ca²⁺] between 0 to 0.1 mM; at higher concentrations (0.2–2.0 mM), TRPV5-mediated Ca²⁺ uptake exhibited a more linear relationship to extracellular [Ca²⁺].

The saturation kinetics may be a result of the Ca²⁺-dependent feedback inhibition mechanism of TRPV5. Once the intracellular [Ca²⁺] exceeds the limit of Ca²⁺-dependent feedback inhibition, a linear relationship becomes dominant. The overall Ca²⁺ transport kinetics fit well with the equation V = V_maxS/(K_m + S) + NS. T⁵⁶³ exhibited a higher apparent K_m, a higher V_max, and a higher N value than A⁵⁶³ (K_m 0.024 ± 0.010 vs. 0.019 ± 0.009 mM; V_max 84.6 ± 11.6 vs. 75.4 ± 10.4 pmol·oocyte⁻¹·30 min⁻¹; N 182.2 ± 7.3 and 122.8 ± 6.3 pmol·oocyte⁻¹·30 min⁻¹·mM⁻¹). When the three parameters of the fitted curves were compared with GraphPad Prism 5.0 software, there was a significant difference between the two sets (P < 0.0001).

The increased Ca²⁺ influx mediated by T⁵⁶³ variant is not a result of increased expression at the plasma membrane.

Because the overall activity of TRPV5 is largely determined by its abundance at the plasma membrane, we next evaluated the level of T⁵⁶³ at the plasma membrane. Oocytes expressing TRPV5 exhibited a scar-like mark, which is likely formed by redistribution of pigments in the oocyte due to Ca²⁺-induced aggregation of cytoskeleton components under the pigments (46). Oocytes expressing T⁵⁶³ exhibited a more prominent mark at the surface than those expressing A⁵⁶³, consistent with the increased Ca²⁺ uptake mediated by T⁵⁶³ (Fig. 3A). The increase in the scale of the mark correlates with the increased Ca²⁺ influx in TRPV5-expressing oocytes, and often the increased Ca²⁺ influx is a result of increased surface expression of TRPV5 (46). However, this is not the case for the T⁵⁶³ variant. In fact, a 30.0 ± 6.9% (n = 8) reduction of fully glycosylated T⁵⁶³ protein (band B), which corresponds to the mature TRPV5 at the plasma membrane (19), was observed (Fig. 3, B and C). The biotinylated T⁵⁶³ protein at the oocyte surface was also significantly decreased (Fig. 3B). Thus the increased scale of the scar-like mark in T⁵⁶³-expressing oocytes reflects the increased level of Ca²⁺ influx, not the level of protein expression at the oocyte surface.

The T⁵⁶³ variant exhibits a reduction in Na⁺ current.

T⁵⁶³ exhibited increased Ca²⁺ influx without increased surface expression. This suggested that the channel properties are altered by the A563T substitution, which is in the last transmembrane domain of TRPV5. We next used the voltage-clamp approach to examine other alterations in TRPV5 properties. In the absence of Ca²⁺ in the extracellular solution, TRPV5 allows Na⁺ to pass through the channel (23). The amplitude of TRPV5-mediated Na⁺ current correlates with the amplitude of Ca²⁺ uptake and is often used as a measure of the channel activity of TRPV5 (19). Most of the SNP variants tested exhibited similar levels of Na⁺ current. However, in contrast to the increased Ca²⁺ influx, T⁵⁶³ exhibited a significant reduction in Na⁺ current (Fig. 4).

Fig. 4.

The A563T variant exhibited a reduction in Na⁺ current. A: representative current traces for TRPV5 and its nonsynonymous SNP variants at membrane potentials varied from −160 to +60 mV at 20-mV intervals. Na⁺-evoked currents were obtained by subtracting currents recorded in the choline solution (0 mM of Na⁺) from those recorded from Na⁺ solution (100 mM of Na⁺). Water-injected control oocytes (Water) exhibited negligible currents. B: current-voltage (I-V) plots of TRPV5 and its nonsynonymous SNP variants. Data are expressed as means ± SE from 3 experiments with a total of 17 oocytes in each group. C: external application of 100 mM Na⁺ or 1 mM Ca²⁺ evoked inward current when the oocyte expressing A⁵⁶³ or T⁵⁶³ variant held at −50 mV was perfused with standard solution without Na⁺ and Ca²⁺. The peak current amplitudes of Ca²⁺-evoked currents in oocytes expressing A⁵⁶³ and T⁵⁶³ were comparable; in contrast, the amplitude of Na⁺-evoked current in the oocyte expressing A⁵⁶³ was much higher than that in the oocyte expressing T⁵⁶³. Water-injected control oocytes exhibited negligible Na⁺- or Ca²⁺-evoked current. D: the ratio between the peaks of Ca²⁺- and the Na⁺-evoked currents was significantly increased for T⁵⁶³. Data from 14 oocytes of two batches of oocytes are expressed as means ± SE. *P <0.05 vs. A⁵⁶³ group by Student's t-test.

To confirm that the A563T variation has different effects on Na⁺ current and Ca²⁺ influx, we next compared the Na⁺- and Ca²⁺-evoked currents in the same oocytes (Fig. 4C). When an oocyte expressing A⁵⁶³ was held at −50 mV in Na⁺- and Ca²⁺-free solution, external application of Na⁺ at 100 mM induced an inward current (Fig. 4C). After removal of Na⁺ solution, subsequent application of Ca²⁺ at 1 mM induced an inward current (Fig. 4C) that was due to the activation of Ca²⁺-activated Cl⁻ channels endogenously expressed in X. laevis oocytes (11, 22) likely mediated by TMEM16A (33). The amplitude of the Ca²⁺-activated Cl⁻ current reflected the increase in intracellular Ca²⁺ resulting from TRPV5-mediated Ca²⁺ influx. In Fig. 4C, the Ca²⁺-evoked currents in an oocyte expressing T⁵⁶³ was slightly larger than that in an oocyte expressing A⁵⁶³; in contrast, the Na⁺ current of T⁵⁶³ was approximately one-half that of A⁵⁶³. As a result, the ratio between the peak levels of Ca²⁺- and Na⁺-evoked currents in the same oocytes expressing T⁵⁶³ was significantly increased (Fig. 4D).

Increased sensitivity to extracellular Mg²⁺ in T⁵⁶³ variant leads to the reduction in Na⁺ current.

The increased Ca²⁺ influx with a reduction of Na⁺ current suggested that the pore of TRPV5 is altered by the A563T substitution. Because Ca²⁺ permeation and Mg²⁺ blockade of TRPV5 are linked to a single Asp⁵⁴² residue at the pore (18, 24), we examined whether the sensitivity to extracellular Mg²⁺ is altered in T⁵⁶³. When measured without extracellular Mg²⁺, Na⁺ current of T⁵⁶³ was only slightly lower than that of A⁵⁶³ (Fig. 5A). In contrast, in the presence of 0.5 mM extracellular Mg²⁺, the Na⁺ current of T⁵⁶³ was significantly reduced compared with that of A⁵⁶³ (Fig. 5A). T⁵⁶³ exhibited a lower IC₅₀ of Mg²⁺ blockade than A⁵⁶³ (0.42 vs. 0.72 mM), as deduced from the dose-response curve of extracellular Mg²⁺ concentration on Na⁺ current amplitude (Fig. 5B). Therefore, the reduction in Na⁺ current of T⁵⁶³ regularly recorded in the presence of 1 mM Mg²⁺ was the result of the increased sensitivity to extracellular Mg²⁺. The sensitivity of Ca²⁺ influx to extracellular Mg²⁺ was also increased in T⁵⁶³, albeit to a much lesser extent (Fig. 5C). Extracellular Mg²⁺ is expected to affect Na⁺ permeation more than Ca²⁺ permeation, because although Mg²⁺ affinity is much higher than Na⁺ affinity, it is ∼100 times lower than Ca²⁺ affinity of TRPV5 (18).

Fig. 5.

Decreased Na⁺ current of T⁵⁶³ variant was due to increased sensitivity to extracellular Mg²⁺. A: representative Na⁺ current traces of A⁵⁶³ and T⁵⁶³ variants in the presence of 0 and 0.5 mM extracellular Mg²⁺ are shown in top. The I-V curves of Na⁺ current for A⁵⁶³ and T⁵⁶³ were parallel in the absence of extracellular Mg²⁺ and departed in the presence of 0.5 mM extracellular Mg²⁺ (bottom). B: Na⁺ current of the T⁵⁶³ variant was more sensitive to extracellular Mg²⁺ concentration than that of the A⁵⁶³ variant. Currents at −160 mV from two independent experiments of 18 oocytes/group are expressed as means ± SE. Flufenamic acid at 200 μM was included in the solutions to block the Ca²⁺-inactivated Cl⁻ current induced by the removal of both Ca²⁺ and Mg²⁺ in the extracellular solution. C: effect of extracellular Mg²⁺ on ⁴⁵Ca²⁺ uptake mediated by oocytes expressing A⁵⁶³, T⁵⁶³, or by water-injected control oocytes (water). Mg²⁺ (10 mM) in the extracellular medium led to a stronger reduction in Ca²⁺ uptake in oocytes expressing T⁵⁶³ than those expressing A⁵⁶³ (49.4 ± 2.2 vs. 38.5 ± 2.5%). Data from 3 independent experiments are expressed as means ± SE.

Roles of the hydroxyl and methyl group in the side chain of Thr⁵⁶³ in TRPV5-mediated cation transport.

The above observation indicated that the A563T substitution in TRPV5 significantly altered the cation transport properties of TRPV5. The side chain of threonine has an additional hydroxyl (·OH) and a methyl (·CH₃) group compared with that of alanine (Fig. 6A, top); we thus examined the roles of the hydroxyl and the methyl group of Thr⁵⁶³ by serine or valine substitution (Fig. 6). When Thr⁵⁶³ was replaced by serine, which lacks a methyl group compared with threonine, the Na⁺ current remained low at the level similar to T⁵⁶³ (Fig. 6, A and B); however, the increase in Ca²⁺ uptake was abolished to the level comparable to A⁵⁶³ (Fig. 6C). When Thr⁵⁶³ was replaced by valine, which has a methyl group in the place of the hydroxyl group in threonine, the Na⁺ current was restored to some extent toward A⁵⁶³ (Fig. 6, A and B) while the Ca²⁺ uptake remained elevated to the level not significantly different from that of T⁵⁶³ (Fig. 6C). These results indicate that the methyl group of Thr⁵⁶³ is responsible for the increase in Ca²⁺ transport, whereas the hydroxyl group may contribute to the increased Mg²⁺ sensitivity that causes the reduction of Na⁺ transport.

Fig. 6.

Roles of the side chain in Thr⁵⁶³ in the transport property of A563T variant. A: Na⁺-evoked current of TRPV5 (Ala), A563T (Thr), A563S (Ser), and A563V (Val). Shown are representative Na⁺ currents in response to different membrane potentials from −160 to +60 mV. The side chains of Ala, Thr, Ser, and Val are also shown on the top of the respective groups. B: I-V relationships of Na⁺ current mediated by TRPV5 variants with different amino acid residues at position 563. C: Ca²⁺ uptake mediated by TRPV5 variants carrying different amino acid residues at position 563. Data from 3 independent experiments are expressed as means ± SE. *P <0.01 vs. Ala group by Student's t-test. NS, not statistically significant.

The A563T variation is capable of increasing TRPV5-mediated Ca²⁺ uptake when present as one of two alleles.

Among 24 African American subjects studied in the SeattleSNPs, six of them contained A563T variation, but only two of the six were homozygous, whereas four of them had A563T as one of two alleles. The nonsynonymous SNP R154H existed in both African Americans and European Americans at high allele frequencies (44 and 28%, respectively); thus, R154H variant should be considered as another reference form of TRPV5. Four of the African American individuals carrying A563T were in compound status with R154H. Because TRPV5 forms tetramers as a functional unit (14, 17), we mimicked transcripts from the two alleles in the cases found in African American subjects by coinjection of an equal amount of two cRNAs containing A563T and/or R154H in different combinations to assess whether the A563T variation is capable of increasing TRPV5-mediated Ca²⁺ transport in these situations (Fig. 7A). In each group, we combined an equal amount (6.25 ng/oocyte) of each of the two cRNA species encoding identical or different TRPV5 variants. We assumed this is equivalent to the mRNA species derived from the two alleles carrying identical or different variations. It was evident that A563T derived from only one allele was sufficient to increase Ca²⁺ uptake mediated by TRPV5 tetramer that contained half reference TRPV5 and half A563T. No significant difference was observed between the group that consisted of all A563T and that consisted of half A563T and half TRPV5. When coexisting with R154H, either in cis or in trans, the A563T variation increased TRPV5-mediated Ca²⁺ uptake to a similar level as that in the situation without R154H (Fig. 7A). An exceptional case observed, when the A563T variation was in homozygous and the R154H variation was in heterozygous condition, Ca²⁺ uptake was increased significantly over other combinations tested (Fig. 7A). It requires further investigation to understand the nature of this apparent synergistic action between the two variations.

Fig. 7.

The A563T variation was capable of increasing TRPV5-mediated uptake when present in a single allele and/or with other nonsynonymous SNP variations. A: the A563T variation increased TRPV5-mediated Ca²⁺ uptake in the presence of the R154H variation. The TRPV5 proteins encoded by 2 cRNA species mimicking the mRNAs derived from 2 identical or different alleles are shown using 2 adjacent vertical lines under the x-axis. Open, gray, and filled circles represent A8V, R154H, and A563T variation, respectively. The 2 cRNA species indicated in different groups were coinjected at an equal amount (6.25 ng/oocyte) to mimic the TRPV5 mRNA transcripts derived from the 2 alleles with different forms of variations in TRPV5. Data from 5 independent experiments are expressed as means ± SE. B: the A563T variation was capable of increasing TRPV5-mediated Ca²⁺ influx in the presence of A8V variation. Data from 3 independent experiments are expressed as means ± SE. *P <0.01 vs. control group (TRPV5). #P <0.01 vs. * groups by Student's t-test.

There was one case studied in the SeattleSNPs program that in which three nonsynonymous SNPs (A8V, R154H, and A563T) coexisted in one subject. We mimicked the situation and found the presence of the A8V variation did not significantly alter the ability of A563T to increase TRPV5-mediated Ca²⁺ uptake (Fig. 7B). The combinations tested in Fig. 7 include all cases of TRPV5 variation found in the SeattleSNPs.

DISCUSSION

Although the frequencies of the TRPV5 variations identified in the SeatleSNPs project need to be tested in larger population studies, the chance that these genetic variations in TRPV5 present at relative high prevalence among African Americans is high. Two of the three African-specific TRPV5 variants identified from the SeattleSNPs project, A563T and L712F, significantly increase TRPV5-mediated Ca²⁺ influx. This suggests that these variations may contribute to African Americans' superior ability in Ca²⁺ reabsorption. Among the two variants, A563T exhibits a significantly higher frequency. In addition, A563T is capable of increasing Ca²⁺ influx in the heterozygous state; thus, it is expected that the percentage of African Americans affected by this variation would be higher than its allele frequency.

The ∼50 or 25% increase in TRPV5-mediated Ca²⁺ influx caused by the A563T or L712F variation should have a significant impact on Ca²⁺ reabsorption, taking into account that the Ca²⁺ reabsorption is a continuing process. The DCT and CNT, where TRPV5 is expressed, are important nephron segments for regulating the level of Ca²⁺ excretion in the urine. In the rat, luminal [Ca²⁺] was 0.9 mM in the early distal tubule and dropped to 0.04 mM in the late distal tubule as measured by free-flow micropuncture experiments (43). TRPV5 plays a key role as a gatekeeper of this active Ca²⁺ reabsorption process in these nephron segments, as demonstrated by a 6- to 10-fold increase in urinary Ca²⁺ excretion in mice lacking Trpv5 (16). Thus it is reasonable to predict a significant reduction of urinary Ca²⁺ excretion in individuals carrying the A563T or L712F variation.

During the review of this manuscript, three nonsynonymous SNPs (A8V, R154H, and A561T) in TRPV5 were reported in a study of 20 French patients with renal hypercalciuria (31). All of these variations exhibited no significant difference in TRPV5 activity relative to the reference TRPV5 (31). The lack of significant functional changes in A8V and R154H is consistent with our results in this study. Interestingly, the A561T variation is just two amino acids apart from the A563T variation in our study, yet it has no significant effect on TRPV5 function, suggesting a crucial role of position 563 in TRPV5 function.

The location of A563T variation in the last transmembrane domain of TRPV5 implicates the possible alteration of cation transport properties of the channel. Indeed, increased Ca²⁺ influx with the reduction of surface expression in the A563T variant indicates an increased Ca²⁺ permeation in the A563T variant. This increase in Ca²⁺ permeation is accompanied by an increased apparent K_m for Ca²⁺ and an increased Mg²⁺ affinity. The consequence of the increased Mg²⁺ affinity is a reduced permeation to Na⁺ in the presence of extracellular Mg²⁺. The increased Mg²⁺ affinity does not affect Ca²⁺ influx very much, likely because the Mg²⁺ affinity of TRPV5 is much higher than Na⁺ affinity but much lower than the Ca²⁺ affinity (18). TRPV5 shares many permeation properties with the classical voltage-gated Ca²⁺ channels that choose Ca²⁺ over Na⁺ at a ratio of over 100:1 (24, 42). Four negatively charged aspartate residues (Asp⁵⁴²) in tetrameric TRPV5 form the main selective filter for Ca²⁺ (24). Because the 563 position in the last transmembrane domain of TRPV5 is close to Asp⁵⁴², it is likely that A563T substitution may affect the architecture of the pore and make it more accommodating for Ca²⁺ ion permeation, although the lack of an accurate three-dimensional structure of TRPV5 precludes a precise prediction how this happens. It has been recently demonstrated that the substitution of a hydrophobic amino acid residue with a positively charged lysine in the last transmembrane domain of TRPM2 and TRPM8 converts the channels from cation selective to anion selective (20). This suggests that amino acid residues in the last transmembrane domain may participate in defining ion permeation properties directly or indirectly by affecting the conformation of the amino acid residues in the cation filter at the adjacent pore region. Because both Ca²⁺ affinity and Mg²⁺ affinity are determined by the Asp⁵⁴² residue in the pore region of TRPV5 (18), it is not surprising that both Ca²⁺ permeation and Mg²⁺ affinity of the A563T variant are affected. The Mg²⁺ affinity of TRPV5 is ∼100 times lower than the Ca²⁺ affinity (18); therefore, the relatively small increase in Mg²⁺ affinity of T⁵⁶³ only has robust effects on Na⁺ influx but not on Ca²⁺ influx. Ca²⁺ permeation is inversely related to Ca²⁺ affinity in cyclic nucleotide-gated channels, likely because Ca²⁺ influx is retarded by stronger binding (12). If this is applicable to TRPV5 channel, the increased Ca²⁺ permeation of T⁵⁶³ could be a result of a decreased Ca²⁺ affinity as implied by the increased apparent K_m value of Ca²⁺ uptake, although we were unable to determine the apparent Ca²⁺ affinity due to the Ca²⁺-induced Cl⁻ current in X. laevis oocytes under our experimental condition.

By substitution of Thr⁵⁶³ with serine or valine, we pinpointed that the methyl group in the side chain of Thr⁵⁶³ is responsible for the increase in Ca²⁺ uptake. It is likely that the bulky side chain of threonine relative to alanine supports a favorable translocation pathway for Ca²⁺. The hydroxyl group of Thr⁵⁶³, on the other hand, is likely responsible for the increase in Mg²⁺ affinity and does not appear to affect Ca²⁺ transport. These substitutions further confirm the importance of amino acid residue 563 in the cation permeation properties of TRPV5.

If the urinary calcium excretion data were compared between adolescent African American and white girls in a well-controlled Ca²⁺ balance study (8), it is noticeable that ∼¼ to ⅓ of African Americans showed a very low value of urinary calcium excretion while other individuals of African Americans had values not much different from the white subjects. This small portion of African Americans contributed significantly to the lowered urinary Ca²⁺ excretion in African Americans. It is possible that these African Americans may carry gene variations that provide superior ability to conserve Ca²⁺ over other African Americans and whites. It is of great interest to examine if the A563T variation in TRPV5 is among those genetic factors that contribute to the superior ability to conserve Ca²⁺ in these African Americans. Our result also warrants future studies on the potential association between A563T variation in TRPV5 gene and the lower prevalence of osteoporosis and kidney stones among African Americans.

GRANTS

Research in our laboratory is supported by the National Institute of Diabetes and Digestive and Kidney Diseases Grant R01DK-072154.