A novel function of calcitonin gene-related peptide in body fluid Cl⁻ homeostasis

Abstract

Vertebrates need to maintain extracellular chloride (Cl⁻) concentrations to ensure the normal operation of physiological processes; the transition from aquatic to terrestrial environments necessitated the development of sophisticated mechanisms to ensure Cl⁻ homeostasis in the face of fluctuating Cl⁻ levels. Zebrafish calcitonin gene-related peptide (CGRP), unlike its splice variant calcitonin, does not respond to environmental Ca²⁺ levels. This study aimed to test the hypothesis that CGRP is involved in the control of body fluid Cl⁻ homeostasis. Acclimation to high-Cl⁻ artificial water stimulated the mRNA expression of cgrp and the receptor (crlr1) when compared with low-Cl⁻. CGRP knockdown induced upregulation of the Na⁺-Cl⁻ co-transporter (ncc2b), while overexpression of CGRP resulted in the downregulation of ncc2b mRNA synthesis and a simultaneous decrease in Cl⁻ uptake in embryos. Consistent with these findings, knockdown of either cgrp or crlr1 was found to increase the density of NCC2b-expressing cells in embryos. This is the first demonstration that CGRP acts as a hypochloremic hormone through suppressing NCC2b expression and the differentiation of NCC-expressing ionocytes. Elucidation of this novel function of CGRP in fish body fluid Cl⁻ homeostasis promises to enhance our understanding of the related physiology in vertebrates.

1. Introduction

Chloride (Cl⁻) has various important roles in vertebrates, including regulating the excitability of nerves and muscles, facilitating transepithelial solute transport, and controlling blood pressure [1,2]. Therefore, the maintenance of extracellular Cl⁻ concentration is crucial for many vital functions in vertebrates; such organisms originated in aquatic environments (marine and freshwater (FW)) and thereafter became terrestrial, which necessitated the development of sophisticated mechanisms for body fluid Cl⁻ homeostasis during their evolution from aquatic to terrestrial environments with diverse Cl⁻ levels. Elucidating the ion transport mechanisms and hormone-mediated control of body fluid Cl⁻ homeostasis is not only important for our understanding of physiology, but also has evolutionary significance.

Our current knowledge of body fluid Cl⁻ homeostasis is mostly derived from studies of mammals. In the distal convoluted tubule (DCT) cell model of Cl⁻ reabsorption in mammalian kidneys, Cl⁻ enters the cell through an apical membrane via the sodium chloride co-transporter (NCC) and is ultimately extruded to the blood through a CLC-K chloride channel [3–5]. Previous studies in zebrafish identified NCC2b-expressing cells in the gills and skin as the major ionocytes responsible for Cl⁻ uptake and thus these cells act in a similar manner to DCT cells in the mammalian kidney [6]. In zebrafish NCC2b-expressing cells, Cl⁻ is absorbed through an apical membrane via NCC2b and is extruded to the blood through a CLC-2c chloride channel at the basolateral side of gill epithelial cells. Other than the NCC2b-mediated CLC-2c Cl⁻ uptake, pathways through SLC26 anion transporters might also participate in Cl⁻ uptake in zebrafish [7] although the detailed mechanisms are still missing a lot of information and further investigation is needed. The ion transport mechanism for body fluid Cl⁻ homeostasis appears to be conserved from fish to mammals [3–5,8].

Different hormones participate in the control and regulation of NCC function in mammalian kidneys. Aldosterone, angiotensin II, vasopressin, parathyroid hormone (PTH), oestradiol, progesterone, and prolactin are known to regulate the expression and activity of NCC in the renal DCT in mammals. Aldosterone induces enhanced phosphorylation of NCC, thereby increasing individual transporter activity without affecting NCC protein expression [9]. On the other hand, angiotensin II stimulates NCC apical membrane surface protein expression [10]. Phosphorylation of NCC is activated by vasopressin through the Ste20-like kinase, 'sucrose-phosphate synthase (SPS)-related proline/alanine-rich kinase (SPAK), and oxidative stress responsive kinase (OSR1) pathways [11]. Furthermore, oestradiol, progesterone, and prolactin, three hormones involved in the sexual cycle, pregnancy, and lactation, upregulate the activity of NCC through the SPAK pathway [12]. To date, only PTH has been identified as a negative regulator of NCC function; it acts via a phospholipase C (PLC)/Ras-guanyl releasing protein 1 (GRP1)/extracellular signal-regulated kinase (ERK) pathway [13]. In fish, only prolactin and isotocin have been demonstrated to regulate the expression and function of NCC [14,15].

The calcitonin gene (CALC) undergoes alternative splicing to give rise to two different peptide hormones: calcitonin (CT) and calcitonin gene-related peptide (CGRP) [16]. Zebrafish only have one CGRP [17], which is an orthologue of human αCGRP according to NCBI and Ensembl databases. In zebrafish, CT acts through the calcitonin receptor (CTR) as a hypocalcemic factor for body fluid Ca²⁺ homeostasis, while CGRP acts through the calcitonin receptor-like receptor 1 (CRLR1) without any effect on Ca²⁺ regulation [17]. In mammals, CGRP is known to be involved in the control of cardiovascular functions through acting as a hypotensive factor on heart and blood vessels [18,19]. Extracellular Na⁺/Cl⁻ homeostasis is well known to be associated with blood pressure [2,20]; however, there is very limited information on the involvement of CGRP in mammalian iono- or osmoregulation. CGRP was found to activate the function of the cystic fibrosis transmembrane conductance regulator (a Cl⁻ channel) in human airway epithelia cells, which is associated with epithelial fluid transport [21]. On the other hand, it is known that the plasma CGRP concentration in rainbow trout is upregulated by environmental salinity [22,23], implying a possible role of CGRP in osmoregulation and ionic (Na⁺ and/or Cl⁻) homeostasis. Furthermore, our preliminary studies revealed that zebrafish CGRP is upregulated by a high Cl⁻ environment. These findings inspired us to test the hypothesis that CGRP is involved in the control of fish body fluid Cl⁻ homeostasis. The expression and function of Cl⁻ transporters have been well studied in zebrafish, because of the comprehensive genetic databases and well-developed molecular physiological approaches available for this organism [6,8]. In zebrafish embryos, the skin ionocytes are responsible for the majority of ion regulation until 7–14 days post-fertilization (dpf), at which time the gills are fully developed and functional [24]. In this work, we used zebrafish embryos as a model to investigate the roles of CGRP and its receptor, CRLR1, in ion uptake regulation. Our results indicate that CGRP acts as a negative regulator to suppress Cl⁻ uptake through the downregulation of NCC2b.

2. Material and methods

(a). Experimental animals

The AB strain of zebrafish was obtained from stocks of the Institute of Cellular and Organismic Biology, Academia Sinica. Fish were kept in local tap water (normal FW) with a circulating system at 28.5°C under a 14 light (L); 10 dark (D) photoperiod. Fish were fed artificially bred brine shrimp. The experimental protocols were approved by the Academia Sinica Institutional Animal Care and Utilization Committee (approval no. RFIZOOHP220782).

(b). Acclimation experiment

High-Na⁺ (10 mM), low-Na⁺ (0.04 mM), high-Cl⁻ (10 mM), low-Cl⁻ (0.04 mM), high-Ca²⁺ (2.00 mM), and low-Ca²⁺ (0.02 mM) artificial water were prepared as described previously [8]. Zebrafish embryos were acclimated to artificial water for 30 hour(s) post-fertilization (hpf) or 4 dpf, and the artificial waters were replaced twice a day to maintain stable ion concentrations and pH.

(c). Preparation of total RNA and reverse transcription

In order to obtain a sufficient quantity of RNA, 30 embryos were pooled as a sample. Samples were homogenized in 0.8 ml Trizol Reagent (Invitrogen, Carlsbad, CA). After chloroform extraction, the total RNA samples were purified and treated with DNase1 to remove the genomic DNA using an RNeasy Mini Kit (Qiagen, Huntsville, AL). For cDNA synthesis, 5 µg of total mRNA was reverse-transcribed in a final volume of 20 µl containing 0.5 mM deoxyribonucleoside triphosphates (dNTPs), 2.5 µM oligo(dT)₁₈, 5 mM dithiothreitol, and 200 units PowerScript reverse transcriptase (Invitrogen) for 2 h at 50°C, followed by a 15 min incubation at 70°C.

(d). Quantitative real-time PCR

Real-time PCR was performed with a LightCycler real-time PCR system (Roche, Penzberg, Germany); the reaction mixture had a final volume of 10 µl and contained 5 µl 2X SYBR Green I Master (Roche Applied System), 300 nM of the primer pairs, and 20–30 ng cDNA. The standard curve for each gene was checked to be in a linear range using rpl13a as an internal control. The primer sets for real-time PCR are shown in table 1.

Table 1.

Specific primer sets used for real-time quantitative PCR analysis.

gene	sequence
cgrp forward	5′-CGACTACGAGGCGAGAAGATTG-3′
cgrp reverse	5′-CTCAGAAAGTCTGCCAGGCGAT-3′
crlr1 forward	5′-AGCAGTGGCCAACAATCAAGA-3′
crlr1 reverse	5′-CAAACACTGCCACAACAATGAG-3′
ncc2b forward	5′-GCCCCCAAAGTTTTCCAGTT-3′
ncc2b reverse	5′-TAAGCACGAAGAGGCTCCTTG-3′
clc-2c forward	5′-ATTGAGAAATGGGAGGAGCA-3′
clc-2c reverse	5′-GGCATGGAGCCTGTGATG-3′
nbce1b forward	5′-TGTTCCTCTACATGGGCGTCG-3′
nbce1b reverse	5′-TGGCGGCTACTGTTGACTTG-3′
rpl13a forward	5′-CCTCGGTCGTCTTTCCGCTATTG-3′
rpl13a reverse	5′-CAGCCTGACCCCTCTTGGTTTTG-3′

(e). Immunocytochemistry and cell counting

Immunocytochemistry (ICC) analysis of NCC2b was performed as described previously [25]. Embryos were incubated with home-made NCC2b antibody (diluted 1 : 100; customization produced by Genomics, Taipei, Taiwan) against the N-terminal domain: IKKSRPSLDVLRNPPDD. Embryos were washed twice with 1× phosphate-buffered saline with Tween 20 (PBST) and incubated with an Alexa Fluor 488 goat anti-rabbit immunoglobulin G (IgG) antibody (1 : 300 dilution with phosphate-buffered saline (PBS); Molecular Probes, Invitrogen, Carlsbad, CA) for 2 h at room temperature. Finally, embryos were washed three times with 1× PBST (15 min per wash). Images were acquired with an inverted microscope (Axioplan 2 Imaging; Carl Zeiss). The cell number of ionocytes at one side of the whole yolk-sac region was counted in 4 dpf control embryos and CALC and crlr1 morphants using freehand selections made with an image processing program (ImageJ 1.45 s; Wayne Rasband, NIH), as previously described [26]. The total area of the yolk sac (one side for each embryo) was calculated using ImageJ based on the designated scale bar size of the image acquired from the microscope.

(f). Microinjection of antisense morpholino oligonucleotides and capped mRNA (cRNA)

The morpholino-modified antisense oligonucleotide (MO) sequences were obtained from Gene Tools (Philomath, OR). The 25 nucleotide MOs designed against the zebrafish CALC gene (5′-CATGGTCCCCTTAAGATGCTCAGCT-3′) or crlr1 gene (5′-CTCGCTGTCATCTTCTTTGGCATTT-3′) were diluted in sterile water to obtain a 3 mM stock solution. A 25 nucleotide standard control oligo was used as a control (5′-CCTCTTACCTCAGTTACAATTTATA-3′). This standard control oligo has no target and no significant biological activity. The entire coding region of CGRP from the zebrafish CALC gene cDNA was inserted into a pCS2+ or a PCS2+ green fluorescent protein (GFP) XLT expression vector. Capped cgrp mRNAs (cRNAs) were synthesized using a mMessage mMachine kit (Ambion, Austin, TX) from a linearized vector containing the respective cDNAs and were then checked for their concentration and quality. Control, control MO, CALC MO, crlr1 MO, and cgrp cRNA solutions containing 0.1% phenol red were injected into one- to two-cell stage zebrafish embryos at a dose of 4 ng per embryo and 250 pg per embryo, respectively, using an IM-300 microinjector system (Narishige, Tokyo, Japan).

(g). ³⁶Cl⁻ influxes

Tracer media were prepared by adding appropriate amounts of Na³⁶Cl (Amersham, Piscataway, NJ) to normal FW to generate a final working specific activity of 210 000–260 000 counts per minute (cpm) mol⁻¹ of ³⁶Cl⁻. Zebrafish embryos were transferred to the tracer medium and incubated for 4 h. After incubation, embryos were washed several times in isotope-free FW. For ³⁶Cl⁻, 15 embryos were first incubated in 200 µl Solvable (Packard) at 50°C for 16 h; next, 1 ml scintillation cocktail (Hionic-Fluor, Packard) was added and then the embryos were analysed with a beta counter (1211 Rackbeta; LKB, Turku, Finland). Unidirectional Cl⁻ influx was calculated using the following formula: J_in = Q_embryo X ⁻¹_outt⁻¹, where J_in is the influx (pmol per embryo h⁻¹), Q_embryo is the radioactivity of the embryo (cpm per individual) at the end of the incubation period, X_out is the specific activity of the incubation medium (cpm per pmol) and t is the incubation time (h).

(h). Western blot analysis

Forty larvae were pooled and homogenized as one sample. Protein samples were separated via 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (100 µg per well). After transfer of the proteins to a polyvinylidene difluoride membrane (Millipore), the blots were incubated with antibodies against CRLR1 (dilute 1 : 500). The membranes were subsequently incubated with an alkaline-phosphatase-conjugated goat anti-rabbit IgG (dilute 1 : 2 500, Jackson Laboratories) and then developed with 5-bromo-4-chloro-3-indolyphosphate/nitroblue tetrazolium. β-actin was used as an internal control. Western blot images were captured using the UVP BioSpectrum 600 Image System (UVP Inc., CA). The intensities of CRLR1 protein and β-actin bands were quantified using the software Image-Pro Plus (Media Cybernetics, MD).

(i). Statistical analyses

Values are presented as means ± s.e.m. and were compared by Student's t-test.

3. Results

(a). Effects of environmental Ca²⁺, Na⁺, and Cl⁻ on the expression of cgrp mRNA and its receptor, crlr1

To determine whether CGRP is involved in zebrafish ion regulation, we first examined expression of the encoding mRNA following acclimation to artificial water containing different concentrations of Ca²⁺, Na⁺, and Cl⁻. The expression levels of cgrp and crlr1 were no different between zebrafish embryos acclimated to artificial water containing high- or low-Ca²⁺ (figure 1a), but were significantly greater in embryos acclimated to high-Na⁺/high-Cl⁻ compared with those acclimated to high-Na⁺/low-Cl⁻ or low-Na⁺/low-Cl⁻ water (figure 1b). Cl⁻ appears to be the major ion affecting the expression of cgrp and crlr1. We previously showed that expression of ncc2b, which is primarily responsible for Cl⁻ uptake, is significantly decreased in embryos acclimated to high-Cl⁻ water [8]. Therefore, high-Na⁺/high-Cl⁻ artificial water decreases expression of ncc2b, which is accompanied by upregulation of cgrp and crlr1. This suggests a potent role of CGRP and CRLR1 in Cl⁻ regulation.

Figure 1.

The effects of different Ca²⁺, Na⁺, and Cl⁻ concentrations in artificial FW on cgrp and crlr1 mRNA expression. (a) Zebrafish embryos were maintained in high-Ca²⁺ (2.00 mM) or low-Ca²⁺ (0.02 mM) artificial water. The expression of cgrp and crlr1 mRNA was analysed by quantitative PCR (QPCR) at 30 h post-fertilization (hpf) and 4 days post-fertilization (dpf). (b) Zebrafish embryos were acclimated to high-Na⁺/high-Cl⁻ (NaCl: 10.00 mM), high-Na⁺/low-Cl⁻ (Na⁺: 10.00 mM; Cl⁻: 0.04 mM), or low-Na⁺/low-Cl⁻ (NaCl: 0.04 mM) artificial water for 30 h or 4 days and the expression of cgrp and crlr1 mRNA was examined by QPCR. Values are the mean ± s.e.m. (n = 6). The values were normalized to rpl13a. Different letters (a–h) indicate a significant difference (one-way ANOVA, Tukey's pairwise comparisons).

(b). Injection of green fluorescent protein-cgrp cRNA with or without morpholino-modified antisense oligonucleotide

To further investigate the role of CGRP, we blocked the synthesis of CGRP in zebrafish using a specific CALC MO and examined the effect on ncc2b and clc-2c synthesis. The efficiency of CALC MO was tested by injecting embryos with a cgrp-GFP fusion cRNA. The signals of the cgrp-fused GFP construct were observed at 4 dpf (figure 2a,b). Co-injection of CALC MO abolished the GFP signals (figure 2c,d), suggesting that CALC MO targets cgrp mRNA and blocks translation. The CALC MO injection resulted in significant increases in ncc2b and clc-2c synthesis (1.80- and 1.74-fold, respectively) at 4 dpf (figure 2e). This result suggests that CGRP exerts an inhibitory effect on ncc2b and clc-2c expression.

Figure 2.

The effects of CALC MO injection on ncc2b, clc-2c, and nbce1b mRNA expression in FW. (a–d) Zebrafish embryos were injected with cgrp-green fluorescent protein (GFP) cRNA and GFP expression was observed in the head (a) and the tail (b). When embryos were co-injected with cgrp-GFP cRNA and CALC MO, GFP expression in the head (c) and tail (d) was abolished. (e) QPCR was performed to detect the expression of ncc2b, clc-2c, and nbce1b in CALC MO-injected embryos and control at 4 dpf. Values are the mean ± s.e.m. (n = 6). The values were normalized to rpl13a. Asterisks indicate a significant difference from the control (Student's t-test, p < 0.05).

(c). Effect of CALC morpholino-modified antisense oligonucleotide injection on ncc2b and crlr1 mRNA expression in fish acclimated to high-Na⁺ high-Cl⁻ media

We proceeded to examine whether calc underlies the observed decrease of ncc2b and increase of crlr1 in embryos acclimated to high-Na⁺/high-Cl⁻ artificial water (figure 1b) by injecting embryos with CALC MO. Embryos injected with CALC MO exhibited a significant decrease in crlr1 and a significant increase in ncc2b expression in high-Na⁺ high-Cl⁻ artificial water, when compared with controls injected with control MO (figure 3). These data suggest that blocking CGRP synthesis induces downregulation of its specific receptor, crlr1, and prevents the inhibition of ncc2b synthesis.

Figure 3.

The effects of CALC MO injection on crlr1 and ncc2b mRNA in embryos acclimated to a high-Na⁺/high-Cl⁻ medium. Zebrafish embryos were maintained in a high-Na⁺/high-Cl⁻ artificial medium and then injected with CALC MO. Expression of crlr1 and ncc2b mRNA was quantified by QPCR at 30 hpf and 4 dpf. Values are the mean ± s.e.m. (n = 6). The values were normalized to rpl13a. Asterisks indicate a significant difference from the control (Student's t-test, p < 0.05).

(d). Effect of cgrp overexpression on ncc2b gene expression

To investigate the mechanisms underlying the role of CGRP in Cl⁻ regulation, we observed the effect of cgrp cRNA overexpression on ncc2b gene expression (figure 4). Expression of ncc2b was significantly reduced by 1.57- and 1.71-fold at 30 hpf and 4 dpf, respectively, in embryos injected with cgrp cRNA and this may be presumed to affect Cl⁻ uptake.

Figure 4.

Effect of cgrp cRNA overexpression on ncc2b mRNA expression, as quantified by QPCR. One- to two-cell stage zebrafish embryos were injected with cgrp cRNA. Expression of ncc2b mRNA in embryos injected with cgrp cRNA was examined by QPCR at 30 hpf and 4 dpf. Values are the mean ± s.e.m. (n = 6). The values were normalized to rpl13a. Asterisks indicate a significant difference from the control (Student's t-test, p < 0.05).

(e). Effect of CALC morpholino-modified antisense oligonucleotide injection and cgrp overexpression on Cl⁻ influx in zebrafish embryos

To confirm the role of CGRP in Cl⁻ homeostasis, we examined the effects of CALC knockdown and cgrp overexpression on whole-mount Cl⁻ influx (figure 5). At 4 dpf, overexpression of cgrp induced a significant decrease in Cl⁻ influx when compared with the control group. By contrast, CALC knockdown induced a significant increase in Cl⁻ influx, further supporting the hypothesis that CGRP may be involved in Cl⁻ homeostasis.

Figure 5.

Effects of CALC morpholino-modified antisense oligonucleotides (MOs) and cgrp cRNA overexpression on Cl⁻ influx in 4 dpf zebrafish embryos. Values are the mean ± s.e.m. (n = 10). Different letters (a–c) indicate a significant difference (one-way ANOVA, Tukey's pairwise comparisons).

(f). Effect of crlr1 morpholino-modified antisense oligonucleotide injection on ncc2b, clc-2c, and nbce1b mRNA expression

To further investigate the role of CGRP in Cl⁻ regulation, we observed the effect of crlr1 MO knockdown on the expression of genes implicated in Cl⁻ homeostasis (figure 6). As shown by Western blot analysis, CRLR1 protein was reduced in crlr1 morphants when compared with that in control morphants at 4 dpf (figure 6a,b). Furthermore, the crlr1 MO injection significantly increased synthesis of ncc2b and clc-2c (1.37- and 2.10-fold increase, respectively), but did not affect the expression of sodium bicarbonate exchanger 1b (nbce1b) at 4 dpf (figure 6c). These results further support the hypothesis that CGRP exerts an inhibitory effect on Cl⁻ uptake.

Figure 6.

Effects of crlr1 MO injection on ncc2b, clc-2c, and nbce1b mRNA expression in embryos acclimated to FW. (a) CRLR1 protein levels in crlr1 MO-injected embryos and control at 4 dpf were analysed by Western blot with an anti-CRLR1 antibody. (b) The histogram shows band densitometric values. Data are shown as means ± s.e.m. (n = 3). Asterisks indicate a significant difference from the control (Student's t-test, p < 0.05). (c) Expression of ncc2b, clc-2c, and nbce1b mRNA in the indicated embryos, as quantified by QPCR at 4 dpf. Values are the mean ± s.e.m. (n = 6). The values were normalized to rpl13a. Asterisks indicate a significant difference from the control (Student's t-test, p < 0.05).

(g). Effect of crlr1 morpholino-modified antisense oligonucleotide injection on NCC2b-expressing ionocytes

To determine if CGRP affects the cell density of NCC2b-expressing ionocytes in zebrafish embryonic skin, we examined the effects of either CALC or crlr1 MO knockdown on NCC2b-expressing ionocytes (figure 7). Injection with CALC (figure 7b) or crlr1 (figure 7c) MO induced a 1.98- and 2.62-fold increase in NCC2b-expressing ionocyte density, respectively, when compared with the control (figure 7a, quantified in figure 7d). These findings suggest that CGRP negatively regulates Cl⁻ uptake, probably through suppressing the differentiation of NCC2b-expressing ionocytes.

Figure 7.

Effects of CALC and crlr1 MO injection on NCC2b-expressing ionocyte densities in embryos acclimated to FW. NCC2b-expressing ionocytes in control embryos (a), and embryos injected with CALC (b) or crlr1 MO (c) at 4 dpf. (d) Comparison of the numbers of NCC2b-expressing ionocytes in 4 dpf zebrafish embryos. Values are the mean ± s.e.m. (n = 9–10). Different letters (a–c) indicate a significant difference (one-way ANOVA, Tukey's pairwise comparisons).

4. Discussion

These data provide the first molecular physiological evidence of a novel function of CGRP in regulating body fluid Cl⁻ homeostasis in FW teleost fishes. CGRP, unlike its splice variant CT (a hypocalcemic hormone), does not respond to environmental Ca²⁺ levels, but rather acts as a negative regulator of Cl⁻ uptake function through suppressing the expression of NCC2b and the differentiation of NCC-expressing ionocytes in the gills and embryonic skin of zebrafish.

Mammalian CGRP is known to be a vasodilator, which manipulates cAMP and nitric oxide through an endothelial receptor [27], and facilitates the reduction of blood pressure in both normotensive and hypertensive rats [28]. However, the mechanism by which CGRP reduces hypertension is unclear. A recent study suggested that CGRP has a protective effect against angiotensin II-induced hypertension [29]. Most studies on CGRP in mammals have not been directly related to body fluid ionic homeostasis [30]. Angiotensin II and aldosterone exert their actions on blood pressure though control of NCC in the renal DCT epithelia cells [9,10,31,32] and thus the possibility that CGRP also targets NCC cannot be ruled out. On the other hand, some recent studies in fish suggested a possible involvement of CGRP in iono- and osmoregulation. The plasma concentrations of CGRP in trout increased after transfer from FW to seawater (SW). Moreover, the increase in CGRP is related to an increase in the specific binding affinity of CGRP to its branchial membrane receptors [22]. Similar increases in plasma CGRP levels and gill-specific CGRP binding sites were also observed in eel transferred from FW to SW [23]. On the other hand, the mRNA encoding the CGRP receptor decreased and that encoding CGRP was absent after transfer of flounder from SW to FW [33]. These data suggest that CGRP may be involved in ion regulation or osmoregulation. However, there was no convincing molecular physiological evidence for a possible mechanism until this study.

To precisely identify the ion transport pathways targeted by CGRP, we compared the regulation of cgrp mRNA expression in zebrafish acclimated to different ionic concentrations or injected with the CALC gene morpholino. Exposure to high-Cl⁻ environments, which is known to suppress the expression of Cl⁻ transporters (NCC2b and CLC-2c) and Cl⁻ uptake function [6,8], resulted in the upregulation of cgrp mRNA expression (figure 1b); in addition, CGRP loss-of-function stimulated the expression of Cl⁻ transporters (figure 2e) and Cl⁻ influx (figure 5). As the use of CALC morpholinos does not allow us to distinguish between the effects of CT and CGRP loss-of-function, we also performed specific CGRP gain-of-function experiments: cgrp overexpression suppressed both the expression of ncc2b mRNA (figure 4) and Cl⁻ influx (figure 5). These findings demonstrate, for the first time, that CGRP has a negative effect on transepithelial Cl⁻ absorption in the context of body fluid ionic homeostasis.

We previously proposed that CRLR1 acts as a specific receptor of the CGRP peptide hormone, based on our comparison of the effects of cgrp overexpression on different receptors [17]. Here, we further confirmed the involvement of CRLR1 in the negative effect of CGRP on transepithelial Cl⁻ uptake function, by demonstrating that NCC2b and clc-2c transcripts are upregulated in crlr1 morphants (figures 6c and 7). This suggests that CGRP-CRLR1 signalling suppresses Cl⁻ uptake function by downregulating the mRNA expression of Cl⁻ transporters.

PTH is the only hormone known to exert hypochloremic effects on mouse DCT cells, an effect mediated through the suppression of NCC function via Ras-GRP1 and the ERK1/2 mitogen-activated protein kinase pathway [13]. However, CGRP, which is also a hypochloremic hormone, controls NCC function via transcriptional and/or translational regulation. Loss of function of either CGRP or CRLR1 affected mRNA and/or protein expression of Cl⁻ transporters, as described above. Furthermore, knockdown of either CGRP or CRLR1 resulted in an increased density of NCC2b-expressing cells (figure 7). The quantities and densities of ionocytes are related to the functional regulation of ion uptake and acid/base balance mechanisms in fish during acclimation to environmental changes [34–36]. Taking all of the above into consideration, it appears that CGRP-CRLR1 signalling is involved in body fluid Cl⁻ homeostasis. Interestingly, Nicoli and co-workers [37] demonstrated that CRLR1 is involved in arterial differentiation via the sonic hedgehog-vegf-notch signalling cascade in zebrafish [37]. CGRP-CRLR1 signalling appears to affect cell differentiation in diverse tissues. Investigation of the role of this signalling pathway in other organisms may yield findings of evolutionary importance.

Interaction of CRLR with different receptor activity-modifying proteins (RAMPs) determines the ligand specificity and affinity of CRLR in mammals [38]. The CRLR family has undergone considerable diversification in teleost fishes. In pufferfish (Takifugu obscurus), CRLR1 combined with only certain types of RAMPs shows specific association with CGRP according to the transient expression experiments [39]. Precise identification of the specific associations of CRLR1 with CGRP in zebrafish will require further transient expression of different combinations in future. In a previous study, overexpression of CGRP was found to increase only the expression of CRLR1 in zebrafish embryos [17]. Moreover, expression of both CGRP and CRLR1 were stimulated by high-Na⁺ high-Cl⁻ artificial water (figure 1b) and loss-of-function of either gene had similar effects on the Cl⁻ transporter in zebrafish embryos in this study (figures 2e and 6c). These data provide some evidence in support of our proposal that CGRP-CRLR1 signalling is involved in body fluid Cl⁻ homeostasis.

Feedback and mutual balance of the activities of positive and negative regulatory hormones are important for body fluid ionic homeostasis. CGRP and PTH are negative regulators of the function and/or expression of NCC as described above (this study; [13]). Stanniocalcin-1, much like CGRP, inhibits the differentiation of NCC2b-expressing cells and thereby regulates Cl⁻ uptake [26]. Several hormones with positive regulatory actions on NCC have also been reported. Angiotensin II, aldosterone, and vasopressin stimulate NCC function at the transcriptional, translational, or posttranslational level in mammals [9–11,40–42] and zebrafish [43]. Isotocin, prolactin, and cortisol enhance the expression of ncc2b via the upregulation of cell differentiation in zebrafish [14,44]. It remains to be established how these hormones work together to regulate the function of NCC2b and other Cl⁻ transporters in the context of body fluid Cl⁻ homeostasis in different vertebrate species.

Vertebrates originated in marine environments, before invading freshwater habitats and subsequently adapting to a terrestrial existence; concurrently, the strategies employed to deal with body fluids evolved from osmoconforming to osmo- and ionoregulating [45]. Ion regulation mechanisms and hormonal control of body fluid Cl⁻ homeostasis may have developed in aquatic fishes; these mechanisms are conserved in mammals, presumably due to their importance during the evolution from aquatic (marine and freshwater) to terrestrial habitats, which have diverse ambient Cl⁻ levels. Our finding that CGRP affects Cl⁻ transport provides new opportunities for future studies and new insights into our understanding of an issue of not only physiological, but also evolutionary importance.

Ion transport and cell differentiation mechanisms appear to be conserved in zebrafish skin/gill ionocytes and mammalian renal tubular cells [46]. The regulatory action of CGRP on body fluid Cl⁻ homeostasis through control of NCC2b-expressing cell differentiation may be also conserved in mammals. CGRP is known to control cardiovascular functions through hypotensive action on heart and blood vessels [18,19,47,48]. Extracellular Cl⁻ acts as another important and independent player in blood pressure regulation [20,49]. As such, future studies into the regulatory action of CGRP on Cl⁻ uptake should provide new insights into the molecular mechanisms of blood pressure regulation, an important issue for human physiology and disease.

Ethics

The experimental protocols were approved by the Academia Sinica Institutional Animal Care and Utilization Committee (approval no. RFIZOOHP220782).

Authors' contributions

Y.-F.W. and P.-P.H. conceived and designed the research. Y.-F.W., A.-G.L., and Y.-C.L. carried out the molecular laboratory work. Y.-F.W. and P.-P.H. wrote the manuscript. P.-P.H. supervised the project. All authors approved the manuscript for publication.

Competing interests

The authors have no conflicts of interest to declare.

Funding

This study was financially supported by grants to P.-P.H. from Academia Sinica and the Ministry of Science and Technology, Taiwan, ROC.