Functional characterization of a Na+-phosphate cotransporter (NaPi-II) from zebrafish and identification of related transcripts

Abstract

1. We report the molecular identification of a Na+-Pi (inorganic phosphate) cotransport system of the NaPi-II protein family from zebrafish intestine. Following a PCR-related strategy, a DNA fragment from intestine-derived RNA was isolated. Rapid amplification of cDNA ends (3'- and 5'-RACE) resulted in the complete sequence (2607 bp) containing an open reading frame of 1893 bp. 2. The NaPi-II-related protein was expressed in Xenopus laevis oocytes and the resulting transport activity was analysed by electrophysiological means. The apparent Km for Pi was 250 microM (96 mM Na+, -60 mV), and voltage-dependent binding of Na+ exhibited a Km of 67.1 mM (1 mM Pi, -60 mV). 3. Interestingly, the overall transport activity was almost insensitive to changes in the holding potential. The apparent affinity for Na+ decreased under hyperpolarizing conditions, whereas Pi binding showed no voltage dependence. Transport activity was inhibited at low pH, which is characteristic for renal NaPi-II isoforms. 4. The expression of the NaPi-II-related isoform was addressed by reverse-transcription PCR. The mRNA could be detected in intestine, liver, eye and kidney. Unexpectedly, a second NaPi-II-related isoform was identified and found to be expressed in kidney, intestine, liver, brain, eye and prominently in testis. In addition, a shorter amplicon was demonstrated to be an antisense transcript related to the NaPi-II intestinal isoform.

Figure 1. Model of NaP_i-II-related proteins

A, hydrophobicity plot of the zebrafish NaP_i-II-deduced amino acid (aa) sequence according to the Kyte-Doolittle algorithm (Kyte & Doolittle, 1982). Positive values (arbitrary units) indicate an increase in hydrophobicity. B, hypothetical model of the NaP_i-II protein family-related proteins showing eight transmembrane-spanning domains, as proposed by Murer et al. (1998). The location of the N- and C-terminus (intracellular) as well as the glycosylated loop (extracellular) have been confirmed experimentally (Hayes et al. 1994; Kohl et al. 1998). The putative N-glycosylation sites (♦), the protein kinase C site (▴), and the protein kinase A recognition sequence (⋆) are indicated. C, schematic representation of the zebrafish intestine cDNA including the indicated restriction sites used for cloning. In addition, the binding sites of the different NaP_i-II-specific primers for screening, RACE and cloning, respectively, are indicated (arrows). The numbers above and underneath the arrows indicate the position of the first base relative to the 5′ end of the entire cDNA (5′ for the sense primers, 3′ for the antisense primers). f, forward; r, reverse.

Figure 2. Nucleotide and protein sequence of intestinal zebrafish NaP_i-II

The cloned cDNA which was used for the expression experiments is indicated by upper case letters. Lower case letters indicate RACE-derived sequences. The sequence of the amplified antisense fragment is represented in bold with the 6 bp overhang in italics. Note that the gap within this emboldened sequence would introduce a frame shift if read in the sense direction. The GenBank accession number of zebrafish NaP_i-II is AF121796.

Figure 3. Electrophysiological characterization of zebrafish NaP_i-II expressed in Xenopus laevis oocytes

The oocytes were kept in frog Ringer solution containing 96 mm Na⁺ at a holding potential (V_h) of −60 mV. Currents (I_P) were induced by 1 mm P_i in the superfusate. These standard conditions were changed according to the parameters assayed. A, P_i concentration dependence of I_P. Data were normalized to the quasi-maximal current obtained with 3 mm P_i. The curve was fitted according to the Michaelis-Menten equation with a resulting K_m of 250 μm. n = 9 for each data point. B, I_P as a function of the extracellular Na⁺ concentration. Data were obtained at four different holding potentials: −120 mV (□), −90 mV (○), −60 mV (▵), and −30 mV (▿). The calculated K_m values and Hill coefficients, respectively, were 74.3 ± 4.0 mm and 2.0 ± 0.1 (−120 mV), 69.8 ± 4.4 mm and 2.1 ± 0.2 (−90 mV), 67.1 ± 5.3 mm and 2.1 ± 0.2 (−60 mV), and 63.8 ± 5.5 mm and 2.2 ± 0.3 (−30 mV). Data were normalized to the current at V_h =−120 mV and 150 mm Na⁺. n = 13 from 4 different batches of oocytes for each data point; s.e.m. values were smaller than 10% of the maximal current. C, current-voltage relationship of I_P of zebrafish NaP_i-II compared with that of flounder NaP_i-II. Data were normalized to the current at V_h =−120 mV. D, zebrafish NaP_i-II-related I_P at various pH values, as indicated. Data were normalized to the current at pH 8.0. n = 3 for each condition.

Figure 4. Tissue distribution of NaP_i-II-related transcripts

mRNA was isolated from different zebrafish tissues and 0.5 μg was used for RT-PCR. The resulting DNA fragments were visualized on an agarose gel. The upper panel represents the expression pattern of the intestine-derived NaP_i-II isoform. The bands show the expected size of 665 bp. The smaller fragments derived from liver and heart were sequenced. The lower panel shows amplicons obtained with primers specific for zebrafish kidney-derived NaP_i-II. The fragments of 660 bp are of the expected size.

Figure 5. Strand-specific amplification of zebrafish NaP_i-II-related fragments

The left panel shows the experiment performed with primers specific for the intestinal isoform. The starting material was either liver- or intestine-derived mRNA. In the liver sample the short fragment is amplified only from sense-primed reverse transcription. RT primed with antisense oligonucleotides is required for the amplification of the 665 bp fragment from liver and intestine. A comparable experiment using primers specific for the ‘renal’ isoform is shown in the right panel. The reverse transcription was primed with either sense or antisense primers. Only samples derived from antisense-primed reverse transcription gave rise to distinct bands.