I. INTRODUCTION: OVERALL MECHANISM

Top Previous Next References

Renal handling of P_i determines its concentration in the extracellular space, the "traffic" place between the two major body compartments: skeleton and intracellular space (37, 46, 101, 102, 216-218, 374). In cells phosphate participates in energy metabolism and is a constituent of signaling molecules, lipids, and nucleic acids. Under "normal" ("steady-state") physiological conditions, urinary P_i excretion corresponds roughly to phosphate intake in the alimentary tract, mainly via upper small intestine (37, 94, 101, 218). To fulfill the "homeostatic" function, i.e., keeping extracellular P_i concentration within a narrow range, urinary P_i excretion must be (and is) under strong physiological control (37, 101, 102). In contrast to intestinal P_i absorption, which adjusts rather "slowly" (for review, see Refs. 94, 290), renal P_i excretion can "adjust" very fast to altered physiological conditions.

Renal P_i excretion is the balance between free glomerular filtration and regulated tubular reabsorption. Under normal physiological conditions, ~80-90% of filtered load is reabsorbed; renal tubular reabsorption occurs primarily in proximal tubules, with higher rates at early segments (S₁/S₂ vs. S₃) and in deep nephrons (e.g., Refs. 24, 142, 146, 159, 203, 232, 318; for review, see Refs. 37, 218, 374). A small fraction of filtered P_i seems to be reabsorbed in the distal tubule (13), but the apparent loss of P_i observed after proximal tubular micropuncture sites could be most likely explained by the higher reabsorption in proximal tubules of deep nephrons (for review, see Ref. 37). Therefore, a study/analysis of mechanisms participating at the level of the kidney in control of P_i excretion can be reduced to phenomena occurring in the proximal tubule.

The cellular mechanisms involved in proximal tubular P_i reabsorption have been studied by a variety of techniques including in vivo and in vitro microperfusions (e.g., Refs. 24, 49, 99, 100, 142, 144, 402), tissue-culture techniques (e.g., Refs. 41, 43, 64, 66, 67, 116, 261, 264), and studies with isolated brush-border and basolateral membrane vesicles (e.g., Refs. 18, 22, 23, 33-35, 44, 52, 53, 55, 76, 78, 80, 88, 95, 98, 118, 127, 143, 148, 149, 152, 153, 155, 179-182, 204, 239-241, 245, 246, 255, 256, 278, 291, 321, 324, 328, 352, 355, 356, 360, 370-372, 375, 392, 400, 401, 410, 429-434). We and others have written previously several comprehensive reviews on cellular mechanisms participating in renal tubular handling of P_i and summarized the experiments with above-mentioned techniques (e.g., Refs. 37, 46, 100, 101, 138, 149, 278, 282, 283, 291). From these studies a secondary active transport scheme emerged (see Fig. 1, left). P_i is taken up from the tubular fluid by (a) brush-border membrane sodium/phosphate (Na-P_i) cotransporter(s) and leaves the cell via basolateral transport pathways. The brush-border entry step is the rate-limiting step and the target for almost all physiological (and pathophysiological) mechanisms altering P_i reabsorption (see below). Basolateral exit is ill defined, and several P_i transport pathways have been postulated including Na-P_i cotransport, anion exchange, and even an "unspecific" P_i leak (channel?). Basolateral P_i transport has to serve at least two functions: 1) complete transcellular P_i reabsorption in a case where luminal P_i entry exceeds the cellular P_i requirements and 2) guarantee basolateral P_i influx if apical P_i entry is insufficient to satisfy cellular requirements. The second can be considered as a "house-keeping" function and might not be specific for (re)absorptive cells. In this review we summarize the present knowledge on the key transporter molecules involved in proximal tubular transmembrane P_i movement (apical and basolateral; see Fig. 1, right).

View larger version (14K): [in this window] [in a new window]   Fig. 1. Scheme for proximal tubular Pi reabsorption. Left: concept of secondary active transport as evidenced by microperfusion studies and studies on isolated membrane vesicles. Right: Na-Pi cotransporter molecules in the proximal tubular epithelial cell. For further details and references, see text. [Adapted from Murer et al. (288).]

	II. PHYSIOLOGICAL REGULATION

Top Previous Next References

As already indicated, regulation of proximal tubular P_i reabsorption and thus of brush-border membrane Na-P_i cotransport codetermines overall P_i homeostasis. Again, many reviews summarizing the regulation of proximal tubular P_i reabsorption at the organ, tubule, cell, and membrane levels have been written (37, 46, 138, 282-289, 292). This information is briefly presented here. In this review we focus on the molecular mechanisms underlying these regulations.

For a brief overview on regulatory events, we focus on major factors and other factors, and the latter is subdivided into hormonal and nonhormonal factors controlling proximal tubular P_i reabsorption (see Ref. 37). For each of these regulatory phenomena, a "memory" effect exists, i.e., the changes are induced by adequate pretreatment, and after characterization (e.g., by clearance techniques) in vivo can then be further analyzed in vitro, e.g., in microperfusion studies or in studies with isolated membrane vesicles (for review, see Refs. 37, 46, 138, 282, 283, 287). This memory effect can at present easily be understood, as physiological regulation of P_i reabsorption involves, as far as they have been studied at the molecular level, an altered expression of a brush-border Na-P_i cotransporter protein (type IIa Na-P_i cotransporter; for review, see Refs. 39, 242, 284-289, 292; see below). Therefore, they are in all cases, with the possible exception of "fasting" (204), related to changes in maximum velocity (V_max) of brush-border membrane Na-P_i cotransport activity in isolated brush-border membrane vesicles (for review, see Refs. 37, 46, 107, 206, 283).

A.  Major Factors

1.  Dietary P_i intake

A low dietary P_i intake can lead to an almost 100% reabsorption of filtered P_i, whereas a high dietary P_i intake leads to a decreased proximal tubular P_i reabsorption (for review, see Refs. 37, 218). These changes can occur independent of changes in the plasma concentration of different phosphaturic hormones (for review, see Refs. 37, 218; see also Refs. 7, 9, 316). Thus an "unknown" humoral factor may be involved in the mediation of these effects. However, as evidenced by studies on cultured renal proximal tubular epithelial cells (e.g., OK cells), a direct effect ("intrinsic") of altered P_i concentration in the extracellular fluid (plasma, glomerular filtrate, culture media) also elicits changes in apical (brush-border) membrane Na-P_i cotransport activity (e.g., Refs. 41, 43, 64, 309, 339).

Parathyroid hormone (PTH) induces phosphaturia by inhibiting brush-border membrane Na-P_i cotransport activity; removal of PTH (parathyroidectomy) leads to an increase in Na-P_i cotransport activity (e.g., Refs. 108, 120, 153; for review, see Refs. 37, 46, 101, 138, 206, 218, 283, 292). These effects can also be analyzed in a tissue-culture model to study cellular/molecular mechanisms involved in proximal tubular P_i handling, in opossum kidney cells (OK cells; Refs. 67, 85, 86, 261-264, 267, 268, 307, 308, 310, 311, 341-345). This in vitro model also provided evidence for cAMP-dependent and cAMP-independent signaling mechanisms in PTH action (see below; see also Refs. 85, 86, 235, 264, 308, 329-333; for review, see Refs. 280, 283, 288, 289).

Vitamin D is suggested to increase/stimulate proximal tubular P_i reabsorption. 1,25-Dihydroxycholecalciferol treatment of rats was found to stimulate brush-border membrane Na-P_i cotransport (226, 227). It is, however, difficult to discriminate between direct versus indirect effects, as in vivo the vitamin D status is closely associated with alterations in plasma calcium and PTH concentrations (for review, see Refs. 37, 46, 101, 107). Thus, at present, it is not clear whether 1,25-dihydroxyvitamin D₃ [1,25(OH)₂D₃] directly regulates mammalian brush-border membrane Na-P_i cotransport. This is in contrast to the upper small intestine where 1,25(OH)₂D₃ stimulates brush-border membrane Na-P_i cotransport (for review, see Refs. 94, 290). In chicken tubular preparations, administration of 1,25(OH)₂D₃ increased P_i uptake, an effect prevented by inhibition of protein synthesis (249, 250). However, in these studies in suspended cells, it is not clear whether the stimulation is related to an increased uptake across the brush-border membrane. It has been suggested that the effects of 1,25(OH)₂D₃ are related to changes in the lipid characteritsics of the membrane (114; for review, see Refs. 21, 37). A stimulatory effect of 1,25(OH)₂D₃ was also observed in a subclone of OK cells and in studies on promoter activation (see sect. VC; Refs. 8, 380).

B.  Other Factors

1.  Hormonal factors

There are additional hormonal factors (e.g., insulin, growth hormone/insulin-like growth factor I/other growth factors, thyroid and other lipophilic hormones, calcitonin, glucocorticoids, atrial natriuretic peptide, nerve transmitters, prostaglandins, parathyroid hormone-related peptide, phosphatonin, and stanniocalcin) with reported effects on proximal tubular P_i reabsorption, i.e., brush-border membrane Na-P_i cotransport (for review, see Refs. 37, 101, 107, 206, 283).

A) INSULIN. Insulin enhances proximal tubular P_i reabsorption by stimulation of brush-border membrane Na-P_i cotransport and prevents the phosphaturic action of PTH (e.g., Ref. 155; for review, see Refs. 37, 150, 206). Specific binding sites for insulin have been identified in basolateral membranes of proximal tubular epithelial cells (154, 155; for review, see Ref. 150).

B) GROWTH HORMONE/INSULIN-LIKE GROWTH FACTOR I/OTHER GROWTH FACTORS. Growth hormone, at least in part mediated by insulin-like growth factor I (IGF-I; locally produced in the kidney), stimulates proximal tubular Na-P_i cotransport (e.g., Refs. 65, 153, 281, 335; for review, see Refs. 37, 150, 206), an effect also observed in OK cells (63, 193). Receptors for growth hormone have been identified on the basolateral membrane of proximal tubular cells and appear to activate the phospholipase C pathway (350). Receptors for IGF-I have also been identified in proximal tubular cell membranes, and associated effects may involve tyrosine kinase activity (151, 154; for review, see Ref. 150).

Epidermal growth factor (EGF) stimulates P_i reabsorption in perfused proximal tubules (336, 337) but inhibits P_i transport in LLC-PK₁ and OK cells (15, 140, 314). These effects are independent of cAMP and may involve tyrosine kinase activity and/or phospholipase C activation (see below; for review, see Ref. 206).

Transforming growth factors [i.e., transforming growth factor- (TGF-)] decrease Na-P_i cotransport activity in OK cells (233, 314). These effects are independent of cAMP, and the mechanisms might be similar to those in EGF action, sharing the same receptor (TGF- and EGF; for review, see Refs. 150, 206).

C) THYROID HORMONE/LIPOPHILIC HORMONES. Thyroid hormone stimulates proximal tubular P_i reabsorption via a specific increase in brush-border membrane Na-P_i cotransport (31, 118, 213, 433, 434; for review, see Refs. 37, 107). The effect of thyroid hormone can also be observed in primary cultured chick renal cells and in OK cells and is dependent on protein synthesis (298, 367).

There are additional lipophilic hormones with reported effects on "renal tubular" P_i transport. All-trans-retinoic acid (CatRA) specifically increases Na-P_i cotransport in OK cells (30; for review, see Ref. 107). On the other hand, -estradiol specifically decreases Na-P_i cotransport in brush-border membranes from adequately pretreated rats (32; for review, see Ref. 107).

D) CALCITONIN. Calcitonin reduces proximal tubular brush-border membrane Na-P_i cotransport in a PTH- and cAMP-independent manner (36, 430, 436; for review, see Refs. 37, 206). This effect might be mediated by a rise in intracellular calcium concentration (for review, see Ref. 37).

E) GLUCOCORTICOIDS. Glucocorticoids increase phosphate excretion by an inhibition of proximal tubular brush-border membrane Na-P_i cotransport (47, 127; see also Ref. 411); this effect can occur independent of an increase in PTH (for review, see Ref. 37). The effects of glucocorticoids are also apparent in vitro, in primary chick proximal tubular cells (299), and in OK cells (192; see also Refs. 156a, 319, 320). An increase in plasma glucocorticoid levels may mediate the phosphaturic response in chronic metabolic acidosis (11, 47, 127; for review, see Ref. 37).

F) ATRIAL NATRIURETIC PEPTIDE. Atrial natriuretic peptide (ANP) also inhibits proximal tubular brush-border membrane Na-P_i cotransport (156, 429). Although a small effect of ANP, mediated by a rise in cGMP, was observed on OK cell Na-P_i cotransport (294), a direct effect on proximal tubular cells is questionable, since receptors for ANP were not identified in proximal tubular epithelial cells (for review, see Ref. 37). An increase in renal dopamine production (see below) could mediate, in the intact organ, the effect of ANP on brush-border membrane Na-P_i cotransport (for review, see Ref. 37; see also Ref. 419).

G) PTH-RELATED PEPTIDE. PTH-related peptide produced by tumors causes phosphaturia. This "PTH analog" causes phosphaturia by mechanisms identical to that involved in PTH action (for review, see Refs. 37, 206; see also Refs. 315, 349).

H) PHOSPHATONIN. Studies in patients with tumor-induced osteomalacia, with associated hypophosphatemia and renal P_i wasting, led to the hypothesis that there is an additional humoral factor controlling serum P_i concentration and renal P_i handling (for review, see Refs. 37, 111, 224, 225). This as yet unidentified factor was named phosphatonin and is suggested to inhibit proximal tubular P_i reabsorption (60). It was observed that conditioned culture media from tumor cells derived from patients inhibited OK cell Na-P_i cotransport. This factor (phosphatonin?) was suggested to have a proteinous nature and a molecular weight between 8,000 and 25,000. The inhibition of Na-P_i cotransport occurred independently of changes in cellular cAMP content. Also, a PTH-receptor antagonist was found (but not identified; PTH related) in these culture media; it interfered with PTH inhibition of OK cell Na-P_i cotransport but not with the inhibitory effect of phosphatonin (for review, see Refs. 37, 111, 224, 225).

I) GLUCAGON. Glucagon administration increases P_i excretion. It was suggested that the effect of pharmacological doses of glucagon is indirect and related to an increase in plasma concentration of liver-derived cAMP (3).

J) STANNIOCALCIN. Two different isoforms of stanniocalcin (STC) were identified and suggested to be involved in calcium and phosphate homeostasis in fish and in mammals. STC-1 was originally identified in fish and later in rat kidney, in more distally located nephron segments (420). STC-2, ~34% amino acid similarity to STC-1 (189), was identified from an osteosarcoma library, and related transcripts were found in different tissues including kidney (72, 105, 189). STC-1 stimulates proximal tubular brush-border membrane Na-P_i cotransport (409); STC-2 has at least in vitro (OK cells), the opposite effect, by a suppression of the type IIa Na-P_i cotransporter (189). Thus STC-1/2 may serve paracrine modulators of P_i reabsorption.

K) PROSTAGLANDIN. Prostaglandins, produced intrarenally, also modulate renal P_i handling. PGE₂ antagonizes the phosphaturia observed under different physiological conditions, e.g., increased PTH levels. This effect is in part, but not fully, explained by effects on the cAMP signaling cascade. The latter is illustrated by the observation that inhibition of renal prostaglandin synthesis (by indomethacin) potentiates the cAMP-independent phosphaturic action of calcitonin (36; for review, see Ref. 37).

L) NERVE TRANSMITTERS. Nerve transmitters also appear to control renal proximal tubular Na-P_i cotransport. Acute renal denervation increases renal P_i excretion, independent of the PTH status (for review, see Ref. 37). These effects can be related to the production of dopamine and/or reduced - or -adenoreceptor activity. Dopamine and its precursor L-dopa increase P_i excretion (104, 187, 188) and inhibit Na-dependent P_i transport in OK cells as well as in isolated rabbit proximal tubules (19, 79, 104, 129, 137, 196). Dopamine can be generated from L-dopa after brush-border membrane uptake of -glutamyl-L-dopa and leads in an autocrine/paracrine manner via a stimulation of adenylate cyclase to the inhibition of brush-border membrane Na-P_i cotransport (104). Stimulation of -adenoreceptors might interfere with hormone-dependent stimulation of adenylate cyclase activity (e.g., by PTH) and might therefore lead to an apparent increase in Na-P_i cotransport activity, and explain a hypophosphaturic action of -agonists (70, 403, 422, 423; see also Ref. 234). In addition, stimulation of -adenoreceptors in OK cells blunted the actions of PTH on cAMP production and inhibition of Na-P_i cotransport (77; see also Ref. 103). Serotonin is also synthesized in the proximal tubules and is antiphosphaturic; it stimulates proximal tubular P_i reabsorption (103, 128, 129, 147).

Adenosine infusion in rats stimulates renal P_i reabsorption (312).

A) FASTING. Fasting may result in phosphaturia and reverse the effects of a low-P_i diet (for review, see Ref. 37; see also Ref. 28). This effect relates also to a change in brush-border membrane Na-P_i cotransport (204). In contrast to dietary P_i-induced changes and other regulatory conditions, the lowered P_i uptake under fasting conditions might be explained by an increase in the apparent Michaelis constant (K_m) value for P_i (204). The effect of fasting may involve, but cannot be explained by, an increase in glucagon levels (for review, see Ref. 37).

B) PLASMA CALCIUM. Changes in plasma calcium lead to changes in renal proximal tubular P_i reabsorption that are primarily associated with the corresponding changes in PTH concentration (12, 421; for review, see Ref. 37). However, in vitro data also suggest a direct cellular effect of extracellular calcium on proximal tubular brush-border membrane Na-P_i cotransport (e.g., Ref. 301). In isolated perfused convoluted rabbit proximal tubules, an increase in bath and perfusate calcium concentration provoked an increase in P_i reabsorption (351). In studies on OK cells, opposite data were obtained: a decrease in medium calcium concentration stimulated Na-P_i cotransport (62). These differences are not understood but might be related to the time scale used in the experiments. The effects in OK cells required prolonged exposure, were dependent on protein synthesis, and may be related to changes in intracellular Ca²⁺ concentration (see sect. VC4; see also Ref. 353).

C) ACID BASE. The influence of changes in systemic acid-base status on renal proximal tubular Na-P_i cotransport are rather complex and are summarized only briefly. The effects on the kinetic properties of the carrier are discussed in section VA4; in brief, an alkaline intratubular pH leads to a stimulation of Na-P_i cotransport (14, 328, 334, 352; for review, see Refs. 37, 138, 216-218, 283). Acute metabolic acidosis does not significantly interfere with P_i reabsorption. In contrast, chronic metabolic acidosis leads to a decrease in Na-P_i cotransport, most likely related to the evaluated glucocorticoid levels (11, 47, 127). These effects are also apparent in OK cells following appropriate changes in media pH conditions (192, 194). Respiratory acidosis leads to phosphaturia involving corresponding changes in Na-P_i cotransport. In contrast, respiratory alkalosis stimulates proximal tubular P_i reabsorption (for review, see Ref. 37).

D) VOLUME EXPANSION. Volume expansion of animal increases P_i excretion and decreases Na-P_i cotransport rates in isolated brush-border membrane vesicles and in isolated perfused proximal tubules (74, 313, 317, 323, 324; for review, see Ref. 37). It is assumed that the effect of volume expansion on proximal tubules is indirect (i.e., via some humoral factors, in part ANP and/or dopamine).

	III. PATHOPHYSIOLOGICAL ALTERATIONS

Top Previous Next References

In addition to the above briefly discussed physiological regulatory mechanisms, that adjust brush-border Na-P_i cotransport to the needs of body P_i homeostasis, there are genetically determined alterations in renal P_i handling and "acquired" alterations in renal P_i reabsorption.

The genetic aspects of proximal tubular Na-P_i cotransport have been covered in many reviews (e.g., Refs. 338, 384, 390), and we only mention those disorders that have been characterized at the molecular level. Several genetic defects resulting in isolated renal phosphate wasting have been described, such as X-linked hypophosphatemic rickets (XLH; e.g., Refs. 295, 384, 385), autosomal dominant hypophosphatemic rickets not associated with hypercolcinuria (ADHR, 110, 113), and hereditary hypophosphatemic rickets with hypercalciuria (HHRH; 136, 394). The first is caused by mutations in the PHEX gene, which has homology to neutral endopeptidase genes and is hypothesized to process or degrade a circulating factor that regulates by an unknown mechanism renal brush-border membrane Na-P_i cotransport (see below; for review, see Refs. 110, 112, 384, 390). A candidate gene for ADHR and/or HHRH could be the brush-border membrane Na-P_i cotransporter (see below). However, the gene involved in ADHR was recently mapped to chromosome 12p13 (113), a gene locus different from the brush-border Na-P_i cotransporter (5q35; 222, 223; see below). Although HHRH has the biochemical features of mice with a gene deletion for the brush-border membrane Na-P_i cotransporter (25; see below), recent studies on a bedouin kindred with HHRH do not support the hypothesis of a direct involvement of the transporter gene in HHRH (A. O. Jones, I. Tzenova, T. M. Fujiwara, D. Frapier, M. Tieder, K. Morgan, and H. S. Tenenhouse, unpublished data). An interesting form of a genetically determined reduction in renal P_i handling is in Dent's disease, where mutations in a chloride channel (CLC5) lead to an apparent P_i transport defect (252; for review, see Ref. 391). How the loss of function of an endosomal chloride channel leads to a decreased brush-border Na-P_i cotransport needs to be determined. Other genetic defects in renal P_i handling are secondary to changes in vitamin D, PTH, or acid/base metabolism or are a consequence of more general metabolic disorders (for review, see Refs. 109, 216, 217, 338, 390).

Disturbances in proximal tubular P_i transport seem to be an early indicator of "nonspecific" proximal tubular alterations, occurring as a consequence of "unphysiological" extrarenal factors (for review, see Refs. 216, 217). This may be explained by the specific kinetic properties of the brush-border membrane Na-P_i cotransporter (see sect. VH). An example of this may be the observed phosphaturia when the filtered load of glucose is augmented (in diabetes mellitus), where a "competition" for driving force will reduce Na-P_i cotransport rate (22, 392). More generally speaking, when driving forces across the brush-border membrane (Na⁺ gradient and/or membrane potential) are altered, the transport of phosphate will be reduced and thus phosphaturia will occur. Furthermore, as part of its physiological regulation, the transporter protein mediating the rate-limiting Na-P_i cotransport has a high turnover. Therefore, "damage" to the brush-border membrane or the transporter protein itself will result in a massive reduction in the brush-border membrane content of Na-P_i cotransporters and thus reduce P_i transport leading to phosphaturia. This may explain, for example, the sensitivity of renal P_i reabsorption to heavy metal intoxication (see Refs. 4, 141, 169).

Diuretics may inhibit proximal tubular P_i reabsorption when administered to animals or intact tubular preparations (for review, see Ref. 37). Because the greatest effect is produced by acetazolamide, it is assumed that inhibition is related to an inhibition of carbonic anhydrase; therefore, the effect is also dependent on the presence of bicarbonate. The effect of other diuretics on proximal tubular P_i reabsorption correlates to some extent with their potency to inhibit carbonic anhydrase. Inhibition of carbonic anhydrase leads to acute and/or chronic changes in systemic and/or tubular pH, which in turn causes the changes in P_i reabsorption.

	IV. PHOSPHATE TRANSPORT MOLECULES IN PROXIMAL TUBULAR CELLS

Top Previous Next References

The cellular scheme for proximal tubular P_i reabsorption given above includes three Na-P_i cotransporters (Fig. 1). They have been molecularly identified and have been named type I, type II, and type III Na-P_i cotransporters (175; for review, see Refs. 284-289, 377). However, there may be additional pathways in the brush-border and basolateral membranes that have not yet been defined at the molecular level. In heterologous expression systems (e.g., Xenopus laevis oocytes), the corresponding cRNA/proteins augment highly Na⁺-dependent P_i uptake. The three families of Na-P_i cotransporters share no significant homology at the level of their primary amino acid sequence (Fig. 2 and Table 1). We discuss the structural properties, tissue expression, and functional characteristics of these three families of Na-P_i cotransporters. Because the type IIa Na-P_i cotransporter is the key player (see sect. V), the kinetic properties and the regulatory behavior of the type IIa transporter are then covered separately and in more detail.

View larger version (12K): [in this window] [in a new window]   Fig. 2. Hydrophobicity predictions for the type I, type II, and type III Na-Pi cotransporters. Shaded areas indicate potential transmembrane segments. Sequence similarities are indicated by the vertical bars. For further discussion and references, see text.

A cDNA related to the type I Na-P_i cotransporter was initially identified by screening a rabbit kidney cortex library for expression of P_i transport activity in X. laevis oocytes (416). Homologous cDNA (and in part proteins) were then found in human, mouse, and rat kidney cortex, in cerebellar granular cells, and in Caenorhabditis elegans (81, 82, 247, 248, 276, 277, 296, 297; for review, see Ref. 414).

The gene encoding the type I cotransporter (NPT 1) maps in humans to chromosome 6 p21.3-p23 (82, 223), in mouse to chromosome 13 close to the Tcrg locus (81, 437), and in rabbits to chromosome 12p11 (223). The promoter organization of NPT 1 has been characterized; 104 bp upstream of exon 1, a single transcription start site was found and a TATA-like sequence at 41 (378).

Type I transporter mRNA has been detected by in situ hybridization in mouse kidney proximal tubules and to a lesser extent also in distal tubules (81). In rabbits, RT-PCR of microdissected tubular segments localized type I mRNA to the proximal tubules (92). Immunohistochemical experiments and studies with isolated membranes localized the type I transporter protein to the proximal tubular brush-border membrane in rabbits and in mice (40; M. Lötscher, J. Biber, and H. Murer, unpublished observations). Studies on brush-border membranes provided evidence for a higher expression in "deep" juxtamedullary compared with superficial nephrons (97).

On the basis of hydropathy predictions, the type I Na-P_i cotransporter protein may contain six to eight transmembrane regions (Fig. 2); it contains three N-glycosylation motifs of which some are used as indicated by immunoblotting studies with isolated brush-border membrane vesicles and by in vitro translation experiments (416; for review, see Refs. 285, 284, 414).

The induction of increased Na-P_i cotransport activity after injection into X. laevis oocytes was the basis for the expression cloning of the cDNA encoding the type I Na-P_i cotransporter cDNA (416). Stable transfection of type I transporter cDNA into Madin-Darby canine kidney (MDCK) and LLC-PK₁ cells resulted also in an increased cellular uptake of P_i (325). Na⁺-dependent P_i uptake, induced after expression of the type I transporter in oocytes, has been extensively characterized (50, 56, 276, 297, 416). The apparent K_m for P_i was ~0.3 mM for expression of the human and ~1 mM for the rabbit type I Na-P_i cotransporter. The apparent K_m value for Na⁺ interaction was ~50 mM,with a Hill coefficient exceeding unity. Furthermore, no pH dependence of type I transporter-mediated Na⁺-dependent P_i uptake could be observed in oocytes. In electrophysiological studies in oocytes, evidence was obtained that the type I transporter protein might be multifunctional, since evidence for anion channel function with permeability for chloride and different organic anions was obtained (50, 57). In the oocyte experiments it was observed that the induction of chloride conduction by expression of the type I transporter cDNA was time and dose dependent, in contrast to Na⁺-dependent P_i uptake, which was maximally increased at low doses of injected cRNA and after short time periods of expression (50). This could suggest that the type I transporter protein may modulate an intrinsic "oocyte" Na-P_i uptake activity, present not only in oocytes, and that the type I transporter protein may or may not be a Na-P_i cotransporter itself but rather an anion channel protein with expression in renal brush-border membrane. Its role in proximal tubular secretion of anions (e.g., organic anions, xenobiotics) needs to be determined. Certainly, the above-described characteristics of type I transporter-induced Na⁺-dependent P_i uptake does not resemble the characteristics of Na⁺-dependent P_i uptake in brush-border membrane vesicles (e.g., Ref. 14; for review, see Refs. 138, 283). Therefore, the type I transporter is not a major player in mediating or controlling brush-border membrane Na-P_i cotransport.

Yabuuchi et al. (426) have studied in more detail the anion conductive properties of the type I Na-P_i cotransporter (human Npt 1). In oocytes, benzylpenicillin, -lactam antibiotics, probenecid, foscarnet, and melavonic acid were transport substrates. In the hepatocytes, Npt 1 was located on the sinusoidal membrane (426).

To establish the physiological role of NPT 1 in above anion secretion (as well as in renal P_i handling), gene deletion experiments are required (H. S. Tenenhouse and I. Soummounou, personal communication).

The cDNA encoding the type II (type IIa) Na-P_i cotransporter was identified by expression cloning in X. laevis oocytes, from rat and human kidney cortex libraries, respectively (260). Homology-based approaches then led to the identification of type II-related transporters in kidneys from different species including flounder and zebrafish, in opossum kidney cells (OK cells), and in a bovine epithelial cell line (NBL-1; Refs. 87, 88, 163, 168, 219, 294a, 366, 405, 417; see also Fig. 3). A type II-related Na-P_i cotransporter was identified in apical membranes of mammalian small intestine and type II pneumocytes (121, 175, 396) and has been designated type IIb Na-P_i cotransporter (Table 1). The regions with highest homology between type IIa and type IIb transporters are in transmembrane domains, and regions with no or little homology are at the cytoplasmic NH₂ and COOH termini (Fig. 4; Ref. 175).

View larger version (101K): [in this window] [in a new window]   Fig. 3. Proximal tubular expression of the type IIa Na-Pi cotransporter protein (left) and mRNA (right). The protein shows a luminal and some distinct intracellular localization. The mRNA is also exclusively located in the proximal tubules. P, proximal tubules; G, glomeruli; D, distal tubules. For comparison, see also Figures 11 and 12. For details, see text and references given in the text.

View larger version (105K): [in this window] [in a new window]   Fig. 4. Sequence comparisons between type IIa and type IIb Na-Pi cotransporters. Predicted transmembrane areas are given by black bars above the sequences and show high sequence homologies. For further discussion, see text and included references. [Adapted from Hilfiker et al. (175).]

Werner et al. (414) compared the sequences of the type II Na-P_i cotransporters (Fig. 5) and three "families" were identified. Interestingly, the type IIa transporter is preferentially expressed in kidney, with a proximal tubular apical location (see sect. VB1). The type IIb transporter can have multiple locations; in mammals, it is expressed in the small intestine, type II pneumocytes, and other tissues, whereas in nonmammalian vertebrates it can be either in the kidney and/or small intestine (175, 190, 219, 294a, 396). A type IIa Na-P_i cotransporter appears to be expressed also in osteoclasts and may play a role in bone resorption (145). A type II Na-P_i cotransporter seems also to be expressed in brain, where the function is not yet established (177). Finally, type II related proteins appeared very early in evolution, and related genes were found in Vibrio cholerae and C. elegans (see Fig. 5; for review, see Ref. 414).

View larger version (26K): [in this window] [in a new window]   Fig. 5. Evolutionary tree of type II Na-Pi cotransporters identified in different species. The database accession numbers are as follows: mouse kidney, L33878 and U22465; rat kidney, L13257; human kidney, L13258; sheep, AJ001385; rabbit kidney, U20793; opossum kidney cells, L26308; mouse intestine, AF081499; human intestine, AF111856; bovine kidney cells, X81699; chicken kidney (A. Werner, unpublished data); carp kidney (Werner, unpublished data); zebrafish kidney (Werner, unpublished data), flounder kidney and intestine, U13963; Xenopus intestine, L78836; zebrafish intestine (Werner, unpublished data); Vibrio cholerae, AJ010968; Caenorhabditis elegans, AF095787. A type II Na-Pi cotransporter was also identified from sheep kidney (S. P. Shirazi-Beechey, unpublished data; accession number AJ001385). [Dendrogram is extended from that published by Werner et al. (414) and given to us courtesy of Andreas Werner.]

The human type IIa cotransporter gene (NPT 2) maps to chromosome 5q35 (Fig. 6; Refs. 222, 223, 269, 277) and the murine (Npt 2) gene to chromosome 13B (437). The human type IIb Na-P_i cotransporter maps to chromosome 4p15-16 (418a).

View larger version (31K): [in this window] [in a new window]   Fig. 6. Chromosomal location and genomic organization of human type IIa Na-Pi cotransporter. The gene is localized on chromosome 5 (5q35) and has 12 introns and 13 exons. The position of the introns and their size is given with respect to the transcription initiation site. For further discussion, see text and included references. [Adapted from Hartmann et al. (162) and Kos et al. (222).]

The genomic structure of NPT 2 (human IIa) and Npt 2 (murine IIa) has been determined; they are ~16 kb in length and consist of 13 exons and 12 introns (Fig. 6; Refs. 162, 379). In the promoter region of the human, murine and OK cell NPT 2/Npt 2 gene, a TATA box is present 31 bp upstream from the transcription start sites. A GCAAT element and several AP-1 sites may control promoter activity (162, 379). The NPT 2/Npt 2 promoter is active only in a proximal tubular environment, i.e., in OK cells (162, 174, 176). 5'-Flanking sequences of the OK cell type II Na-P_i cotransporter gene contain elements mediating transcriptional control under different bicarbonate/carbon dioxide tensions (194). For the rat Npt 2 promoter, an important role of repeating AP-2 consensus sites in regulating cell-specific expression was documented (359). In reporter gene studies, no physiological regulation (e.g., by low-P_i medium, PTH, thyroid hormones, and growth factors) was observed using a short promoter (327 bp) fragment (174, 176). However, in COS-7 cells expressing the human vitamin D receptor, a vitamin D response element was observed at ~2 kb upstream from the transcription initiation site of the NPT 2/Npt 2 gene (380). Furthermore, regulatory sequences within the NPT 2 gene, ~1 kb upstream of the transcription start site, were identified as binding sites to nuclear proteins upregulated in kidneys of weaning mice fed a low-P_i diet (212a). The corresponding DANN-binding protein could be identified; it corresponds to a known transcription factor (TFE3) that activates transcription through the µE3 site of the immunoglobin heavy chain enhancer (212a). The mRNA encoding TFE3 was found to be significantly increased in kidney tissues of weaning mice fed a low-P_i diet (212a).

In situ hybridization of renal sections (Fig. 3) and nephron microdissection, followed by RT-PCR, documented that type IIa mRNA expression is restricted to the kidney proximal tubule (87, 91, 348, 388). Therefore, in mouse kidney, the type IIa is by far the most abundant of known Na-P_i cotransporters (388). Type IIa Na-P_i cotransporter protein is found in the brush-border membrane of proximal tubules (see Fig. 3; Refs. 91, 348). Inter- and intranephron distribution type of IIa Na-P_i cotransporter highly depends on the physiological requirements within overall P_i homeostasis, (see Figs. 3 and 11; Refs. 91, 210, 243, 348; for review, see Refs. 242, 284-287). The type IIa Na-P_i cotransporter is also expressed in OK cells but not in other renal cell lines (310, 311, 366, 386, 424; and J. Forgo, G. Strange, J. Biber, and H. Murer, unpublished observations). Recently, a type IIa transporter protein-related immunoreactivity was observed in membrane fractions isolated from nontransformed immortalized mice kidney cortex epithelial cells (71). There is no evidence that the type IIa Na-P_i cotransporter is expressed in primary renal proximal tubular epithelial cell cultures (Forgo et al., unpublished observations). Its expression in OK cells is the basis for the use of this cell line as an in vitro model for the study of cellular mechanisms involved in regulation of type IIa Na-P_i cotransport activity (see Refs. 192-194, 234-236, 263, 307-311; for review, see Refs. 283-288).

The related type IIb Na-P_i cotransporter is found in the apical membrane of upper small intestinal enterocytes and type II pneumocytes (see Refs. 175, 396); type IIb transcripts have been found in a variety of other tissues (175).

Hydropathy analysis predicted eight transmembrane segments for the type IIa cotransporter protein (Fig. 2; Ref. 260; for review, see Refs. 284, 285). This membrane topology was supported by several experimental findings: 1) insertion of FLAG epitopes and accessibility of the epitope to antibodies (231); 2) lack of accessibility of antibodies directed against COOH- and NH₂-specific amino acid sequences (231); 3) identification of two glycosylation sites in the second "suggested" extracellular loop (166); and 4) accessibility to membrane-impermeant sulfhydryl reagents after insertion of cysteine residues at specific sites of the protein (228, 229). Two regions may penetrate partially into the lipid bilayer (Fig. 7).

View larger version (54K): [in this window] [in a new window]   Fig. 7. Secondary structure (membrane topology) of type IIa Na-Pi cotransporter (rat, NaPi-2). The model is derived from hydropathy predictions (Fig. 2; Ref. 260) and is experimentally supported by studies on N-glycosylation (166), accessibilities of specific antibodies to either the NH2 or COOH terminus (231), FLAG-epitope insertion (231), and cysteine insertions/deletions (228, 229). For further details, see text and included references.

Type IIa Na-P_i cotransporters contain numerous potential phosphorylation sites for protein kinase C and casein II kinases (165, 260). The role of these sites in physiological control of transport activity is not clear (see sect. VC4).

On immunoblots of brush-border membrane proteins performed under nonreducing conditions, the type IIa Na-P_i cotransporter shows an apparent molecular mass of 80-90 kDa; under reducing conditions two bands of ~45-50 kDa appear (39, 48, 91, 425). The latter suggests that the transporter might be proteolytically cleaved between the two glycosylation sites at positions N298 and N328 (39, 48, 228, 229, 305). It is not known whether this proteolytic cleavage occurs in situ or whether it is experimentally induced. Site-directed mutagenesis studies documented an S-S bridge in the second extracellular loop (Fig. 7; Refs. 228, 229). It is of interest that separate oocyte injections of cRNA encoding NH₂- and COOH-terminal fragments of the flounder type IIb Na-P_i cotransporter resulted in induction of P_i uptake activity only if both "parts" of the proteins were "present" (220).

The question of a multimeric structure of the type IIa cotransporter has been addressed mainly in radiation inactivation studies (33, 98, 195). The size of the functional unit of brush-border membrane Na-P_i cotransport (mostly type IIa Na-P_i cotransporter mediated) was found to be between 170 and 200 kDa, suggesting a multimeric structure. Recent experiments in oocytes expressing wild-type and mutant (inactivatable, cysteine insertion) type IIa Na-P_i cotransporters suggested that each individual wild-type cotransporter molecule within an assumed homomultimeric complex is functional (220a). The apparent high functional molecular mass observed in brush-border membranes could also be due to a heteromultimeric complex (see below). The experiments in different heterologous expression systems (e.g., in Sf9 cells, Refs. 134, 135; in MDCK cells and LLC-PK₁ cells, Refs. 325, 326; and in oocytes, Ref. 260) suggest that an unknown additional protein within the functional complex is not an obligatory requirement for the type IIa Na-P_i cotransporter-mediated P_i uptake activity or, rather unlikely, is present as an intrinsic protein (to serve as a transporter subunit) in different expression systems.

Tatsumi et al. (381) have identified type IIa Na-P_i cotransporter-related cDNA (named NaPi-2, NaPi-2, and NaPi-2). The NaPi-2-encoded protein (355 amino acids) has a high homology to the NH₂-terminal half of the type IIa cotransporter, NaPi-2 encodes for 327 amino acids identical to the NH₂-terminal part of type IIa cotransporter with a completely different 146-amino acid COOH-terminal end, and NaPi-2 encodes a 268-amino acid protein from the COOH-terminal end of the molecule (381). It seems that the related mRNA are formed by alternative splicing of the type IIa cotransporter gene (381). Isoform specific mRNA were found on Northern blots of rat kidney cortex mRNA. With the use of a full-length type IIa Na-P_i cotransporter cDNA probe, the major transcript detected was ~2.6 kb (260, 381). Additional bands (9.5, 4.6, and 1.2 kb) were seen, although in our experience, these bands are not abundant (260, 381). The NaPi-2 probe hybridizes with transcripts of 9.5 and 4.6 kb, the NaPi-2 probe with a transcript of 1.2 kb, and the NaPi-2 probe with transcripts of 9.5 and 2.6 kb (381). In Western blots, with the use of NH₂- or COOH-terminal type IIa Na-P_i cotransporter specific antibodies, proteins of 45, 40, and 37 kDa were observed, corresponding approximatively to the size of the in vitro translated proteins (NaPi-2, NaPi-2, NaPi-2; Ref. 381). The full-length type IIa Na-P_i cotransporter protein is recognized in Western blots from brush-border membrane as a 80- to 90-kDa protein in its glycosylated form (87). In our hands, the lower molecular mass bands are not detected in the absence of reducing agents (see Refs. 39, 91). Because they are only visible under reducing conditions, type 2, 2, and/or 2 related proteins might be linked to the full size type IIa Na-P_i cotransporter via S-S bridges. Alternatively, the possibility exists that the smaller proteins, apparent after reduction of S-S bridges, are a product of proteolytic cleavage of the full size type IIa Na-P_i cotransporter protein (see above). Based on coexpression experiments in oocytes, Tatsumi et al. (381) postulated that the smaller isoforms might regulate, in a dominant negative manner, the function of the type IIa Na-P_i cotransporter protein. However, this interpretation requires further studies to, for example, document the coexistence, within a "heterologous" complex, of the different proteins at the brush-border membrane. Furthermore, quantitative aspects are crucial, since in our experience NaPi-2, -2, and -2 can only be present in rather small amounts relative to the full size type IIa cotransporter. Thus the role of the small type IIa Na-P_i cotransporter-related proteins in brush-border membrane Na-P_i cotransport in vivo is not clear.

An antisense type IIb Na-P_i cotransporter transcript was detected in different nonmammalian tissues. It was postulated that it might be involved in the control of cotransporter protein expression (physiological control; tissue specificity; Ref. 184).

When expressed in X. laevis oocytes, the type IIa Na-P_i cotransporter mediates Na-P_i cotransport activity with functional characteristics identical to those observed in isolated brush-border vesicles (87, 163, 260, 366, 405). A 3:1 stoichiometry (Na⁺:P_i) is the basis for its membrane potential sensitivity (electrogenicity; e.g., Refs. 59, 123). As discussed in section V, the transport characteristics and kinetic behavior of the type IIa transporter have been studied in great detail. Similar transport characteristics were also observed in different other heterologous expression systems such as insect Sf9 cells, fibroblasts, and MDCK cells (134, 135, 326, 395).

Surprisingly, the receptor for gibbon ape leukemia virus (Glvr-1) and the receptor for the mouse amphotropic retrovirus (Ram-1) have been shown to mediate Na-P_i cotransport activity after their expression in X. laevis oocytes (201, 202, 302). The transporter proteins have been named PiT-1 and PiT-2 and are now classified as type III Na-P_i cotransporters (Table 1).

Expression of type III Na-P_i cotransporters seems to be ubiquitous, and related mRNA have been identified in kidney, parathyroid glands, bone, liver, lung, striated muscle, heart, and brain (Table 1; Refs. 84, 201, 202, 302, 303, 362, 382). In mouse kidney, transcripts of type III cotransporters are found throughout the different structures (362, 388). Immunofluorescence studies showed in the proximal tubule a basolateral location (C. Silve, personal communication). Based on mRNA levels, type III Na-P_i cotransporters are two orders of magnitude less abundant than type IIa transporters (388). Its role in the proximal tubule seems not to be in transcellular P_i transport but rather in cell P_i uptake if luminal P_i entry is insufficient for cell metabolic functions. Type III transporter expression seems not to be altered by PTH (386).

The type III transporters show some homology to a Neurospora crassa gene (Pho^-4+) involved in transmembrane P_i movements (302). Hydropathy analysis suggests 10 transmembrane regions (Fig. 2; Refs. 201, 202).

PiT-1- and PiT-2-mediated Na-P_i cotransport has been studied by expression in X. laevis oocytes or in fibroblast transfection (201, 202, 302). Transport is characterized by a K_m for P_i in the order of 20-30 µM and a K_m for Na of 40-50 mM. pH dependence of type III Na-P_i cotransporter is opposite to the type IIa cotransporter, i.e., decreased activity by increasing pH. Similar to the type IIa, type III-mediated transport of P_i is electrogenic with a net influx of a positive charge during the transport cycle, suggesting also a 3:1 stoichiometry .

	V. TYPE IIA SODIUM-PHOSPHATE COTRANSPORTER: THE KEY PLAYER IN BRUSH-BORDER MEMBRANE PHOSPHATE FLUX

Top Previous Next References

The tissue expression, the relative renal abundance, and overall transport characteristics of type I, II (IIa), and III Na-P_i cotransporters suggest that the type IIa transporter plays a key role in brush-border membrane P_i flux. As discussed in this section, changes in expression of the type IIa Na-P_i cotransporter protein parallel alterations in proximal tubular P_i handling, documenting its physiological importance (for review, see Refs. 242, 284-289). In addition, experiments on molecular (genetic) suppression of the type IIa Na-P_i cotransporter support its role in mediating brush-border membrane Na-P_i cotransport. 1) Intravenous injection of specific antisense oligonucleotides led to reduced brush-border membrane Na-P_i cotransport activity that was associated with a decrease in type IIa cotransporter protein (300). 2) Disruption of the type IIa Na-P_i cotransporter gene (Npt 2) in mice led to an ~70% reduction in brush-border Na-P_i cotransport rate and complete loss of the protein (25, 178; see also below). The molecular basis for the remaining brush-border membrane Na-P_i cotransport after Npt 2 gene disruption is unclear. Either the type I transporter protein or another not yet identified Na-P_i cotransporter could account for residual transport activity. 3) Injection of type IIa antisense oligonucleotides in oocytes completely inhibited Na-P_i cotransport mediated by kidney cortex mRNA, confirming its major role in brush-border membrane Na-P_i cotransport (275, 276, 389, 415).

As already indicated, the transport characteristics of the type IIa cotransporter heterologously expressed in different cellular systems (mainly X. laevis oocytes) resembles closely those of Na-P_i cotransport activity observed in isolated brush-border membranes (e.g., Refs. 87, 88, 163, 168, 219, 260, 366, 405, 417). In particular, in all expression systems studied thus far, type IIa-mediated Na-P_i cotransport activity increased with increasing media pH values, a "signature" for proximal tubular brush-border membrane Na-P_i cotransport (e.g., Refs. 14, 260).

The simplest experimental technique to analyze the transport characteristics of a Na-substrate cotransporter is by studying Na⁺ gradient-driven tracer substrate influx under different conditions. For the type IIa cotransporter, this has been done already in 1976 in isolated rat brush-border membrane vesicles (179); obviously, it was then not known that the brush-border Na-P_i cotransport activity is mostly associated with the type IIa cotransporter protein (25, 260). The studies with isolated membrane vesicles provided, however, significant insights into the mechanism/kinetic of brush-border membrane Na-P_i cotransport (e.g., Refs. 14, 34, 35, 55, 80, 352). A detailed kinetic characterization of type IIa-mediated Na-P_i cotransport activity was performed after its expression in X. laevis oocytes (e.g., Ref. 260). The first characterization, performed using tracer techniques, suggested a Na⁺:P_i stoichiometry exceeding unity (see sect. VA1; Ref. 260). These data and the evidence for electrogenicity of Na-P_i cotransport across the brush-border membrane from microperfusion experiments in vivo (133) and from studies with isolated vesicles (35, 55) were the rationale for an electrophysiological characterization of the type IIa Na-P_i cotransporter after its expression in oocytes. The electrophysiological studies, performed under steady-state conditions, complemented the tracer uptake study, whereas pre-steady-state measurements provided new insights into individual steps within the transport cycle (see below).

Under voltage-clamp conditions, superfusion of oocytes expressing the rat type IIa cotransporter with 1 mM P_i in the presence of Na⁺ (100 mM) elicits an inward current, the magnitude of which depends on the holding potential (Fig. 8, A and B; Refs. 56, 58, 59, 122, 123, 125, 126, 407, 408). This observation indicates that a P_i-induced inward movement of positive charge(s) occurs during the transport cycle. At a given membrane potential, dose-response relationships can then be obtained for both Na⁺ and P_i. Furthermore, the interdependence of the apparent affinity for either Na⁺ or P_i and/or the applied membrane potential was studied (e.g., Ref. 122). For P_i, a hyperbolic saturation curve is observed, whereas for Na⁺, the saturation curve is sigmoidal (Fig. 8, C and D). At 100 mM Na⁺, the apparent K_m for P_i interaction is ~0.1 mM and shows little dependency on the holding potential (Fig. 8E). At 1 mM P_i, the apparent K_m for Na⁺ interaction is ~50 mM (Fig. 8F). The concentration dependence of P_i-induced current depends on the external Na⁺ concentration with a dual effect: increasing Na⁺ leads to a decrease in the apparent K_m for P_i and to an increase in the apparent V_max (Fig. 8C). On the other hand, increasing P_i also leads to an increase in affinity for Na⁺ (Fig. 8D). At the lower Na⁺ concentrations, the apparent K_m for P_i interaction shows a marked dependence on the holding membrane potential (Fig. 8E); this is not observed for the K_m for Na⁺ interaction (Fig. 8F). Finally, these saturation experiments provide some information with respect to the stoichiometry (P_i :Na⁺). Hill coefficients calculated on the basis of the P_i saturation curves were always close to 1, whereas those calculated for Na⁺ saturation were always close to 3 (e.g., Refs. 59, 122 123). A 3:1 stoichiometry explains the positive inward current (59, 122, 123). The stoichiometry (Na⁺:P_i) has been determined more directly by simultaneous measurements of substrate flux and charge movement under voltage-clamp conditions in the same oocytes (123). It was found that translocation of a positive charge into the oocyte is associated with the transfer of 1 P_i and 3 Na⁺. These experiments also provided evidence for the preferential transport of divalent P_i anions (123).

View larger version (27K): [in this window] [in a new window]   Fig. 8. Electrogenic behavior of type IIa Na-Pi cotransporter (rat, NaPi-2). A: basic scheme for the two-electrode voltage-clamp system as used in oocytes expressing the transporter. Vc, current potential; Im, membrane current. B: basic experimental recording. When Pi is suppressed in the presence of Na+, an inward current is recorded; its amplitude depends on the transmembrane holding potential (steady-state current recordings). Vh, holding potential. C: dose-response data obtained at 50 mV (holding potential) by altering Pi concentrations in the superfusate containing either 96 mM Na+ () or 50 mM Na+ (), resulting in Michaelis constant (Km) values for Pi interaction (see E). Ip, pipette current. D: dose-response data obtained at 50 mV (holding potential) by altering Na+ concentration at either 1.0 mM Pi () or 0.1 mM Pi (), resulting in Km values for Na+ interaction (see F). E: dependence of Km values for Pi interaction on different holding potentials (see C). F: dependence of Km values for Na+ interaction on different holding potentials (see D).

The antiviral agent foscarnet (phosphonoformic acid, PFA) is a known competitive inhibitor of brush-border membrane Na-P_i cotransport (e.g., Refs. 10, 205, 256, 375, 404). In electrophysiological studies, PFA inhibited P_i-induced inward currents but did not elicit PFA-induced currents (59, 122). Thus PFA interferes with P_i binding but is not a transported substrate. In addition, arsenate is a competitive inhibitor of brush-border membrane and oocyte type IIa Na-P_i cotransporter-mediated P_i uptake (179). In contrast to PFA, arsenate induces inward currents and is thus a transported substrate (e.g., Ref. 163). Recently, a "slippage current" associated with the transfer of the partially loaded type IIa cotransporter was identified (only with Na⁺, see below; Fig. 9; Ref. 122). This current was blocked by PFA and showed a dose dependence suggesting interaction with only one Na⁺. The slippage current accounts for ~10% of maximally induced current of the fully loaded carrier (122). Although this slippage current is of little functional significance, it is important in our understanding of the transporter cycle (see sect. VA3).

View larger version (15K): [in this window] [in a new window]   Fig. 9. Pre-steady-state charge movements related to type IIa Na-Pi cotransporter (rat NaPi-2). A: voltage step between 100 to 0 mV applied to oocyte and corresponding membrane currents superimposed for 96 mM Na+ with and without 1 mM Pi. The small displacement in the baseline preceding the step represents the steady-state induced current at 100 mV. B: magnified view of current records in A showing a clear difference in the relaxations depending on the presence or absence of substrate. Thin trace is for 1 mM Pi; bold trace is for 0 mM Pi. C: difference between records in B, showing the relaxations superimposed on the Pi-induced steady-state currents. Note that the rate of relaxation for the forward step (100 to 0 mV) is faster than the return relaxation, indicating that the relaxation time constant depends on the target potential that is a characteristic of voltage-dependent charge movements. The area under each relaxation, determined by numerical integration, is a measure of the apparent charge translocated and should be equal for the depolarizing and hyperpolarizing steps. For further discussion, see text and included references.

Pre-steady-state relaxation's resulting from the application of voltage steps to the voltage-clamped cell have been first reported for the cloned Na⁺-glucose cotransporter (SGLT-1; Refs. 45, 75, 167) expressed in oocytes and subsequently for many other Na⁺-solute cotransport systems (e.g., Ref. 117). They permit an identification of partial reactions within the transport cycle. Such measurements were also performed with the rat (types IIa and IIb) and flounder isoforms (type IIb) of the type II Na-P_i cotransporters expressed in oocytes (122, 124, 125, 126). Figure 9 provides an example from a study on the rat type IIa cotransporter (122). Application of a voltage step in the presence of 96 mM Na⁺, and in the presence or absence of saturating P_i, leads to current transients that are primarily due to charging oocyte membrane capacitance (Fig. 9A). Recording at a higher gain results in a slower relaxation to the steady state in the absence of P_i (Fig. 9B). Subtraction of the curves obtained in the presence/absence of P_i shows transporter cycle-dependent relaxation currents (Fig. 9C), whose magnitude could be directly related to the magnitude of P_i-induced steady-state currents in oocytes expressing different amounts of cotransporters at their surface (122). Furthermore, the P_i-induced effects on the pre-steady-state relaxation shows the same saturation characteristics (apparent K_m) as that observed for P_i interaction in steady-state measurements (122). These current transients are consistent with the translocation of charged entities within the transmembrane electric field. The voltage dependence of the time constants of relaxation and equivalent charge associated with the relaxation can be obtained from measur