One year old individuals of H. cumingii, with an average shell length of 65 mm were collected from Weiwang Pearl Farm of Jinhua, Zhejiang Province, China, which were transported to the laboratory and kept in a PVC tank with aerated pond water at 25˚C for 48 h. After that, the mollusks were sampled, and five tissue samples were isolated for RNA extraction, including gonad, hepatopancreas, foot, mantle and gill. All tissue samples were immediately washed with sterile PBS, stored in liquid nitrogen and stored at −80˚C.

Total RNA from the gill was extracted using TRIzol Reagent (Invitrogen, USA) according to the manufacturer’s protocol. The quality and quantity of total RNA were then checked using gel electrophoresis system (Bio-Rad, USA) and Spectrophotometer (Thermo, USA). First-strand cDNA was synthesized with 5 μg DNA-free total RNA as the PCR template using a RevertAid First Strand cDNA Synthesis Kit (Thermo, USA), according to the manufacturer’s instructions.

A partial cDNA sequence of PMCA from H. cumingii had been harvested via high-throughput transcriptome sequencing using the mRNA extracted from pooled six tissues before this study [7]. Here, a rapid amplification of cDNA ends (RACE) cDNA library was constructed with a Clontech Smart cDNA Amplification kit using RNA from the gill, according to the manufacturer’s instructions. To obtain the 3' terminal sequence of the PMCA end, two rounds of the forward primers, including RC146-R1 and RC146-R2 was designed based on the EST sequence obtained from our previous transcriptomic analysis of H. cumingii [1] (Table 1). The 25-μL reaction mixture consisted of 12.5 μL of 2 × GC buffer I, 4.0 μL of dNTP Mix (2.5 Mm), 1.0 μL of template, 0.5 μL of universal primer mix, 0.5 μL of RC146-R1 or RC146-R2, 0.2 μL of Polymerase Mix (5 U/μL) and 6.3 μL ddH₂O was programmed as follows: 3 min at 95˚C; 40 cycles of 30 s at 94˚C, 30 s (and 58 s in the 2^nd round) at 58˚C and 60 s at 72˚C; 7min at 72˚C. The amplified fragment was then sub-cloned into a pMD18-T vector prior to sequencing by a commercial company (Sangon, China). Based on the 3'-cDNA end sequence of HcPMCA, gene-specific primers were designed for two rounds of 5'-RACE PCR and listed in Table 1. PCR of 5'-cDNA ends and cloning of the products were carried out as described above for 3'-cDNA ends. The complete HcPMCA cDNA sequence was obtained by overlapping these three fragments.

All recombinant DNA was sequenced, and determination of the gene, ORF, and protein sequence was performed using the Expert Protein Analysis System (ExPASy, http://au.expasy.org). Multiple sequence alignments and the phylogenetic tree were analyzed using the DNAMAN 8. Domain prediction was undertaken using the simple modular architecture research tool (SMART, http://smart.embl-heidelberg.de/). Analysis of the C-terminal sequence of the protein with the Calmodulation Database and Meta-Analysis Predictor (CDMAP) server (http://cam.umassmed.edu), was performed to search the putative CaM- binding motifs within HcPMCA. HMMTOP [8] and TMHMM (http://www.cbs.dtu.dk/services/TMHMM), were used to predict transmembrane helices and to perform a topology analysis of HcPMCA.

The tissue-specific expression analysis was performed by qRT-PCR in a real-time thermal cycler (ABI, USA) using synthesized cDNA as the template, based on a previous protocol [9]. Here, the sampled tissues were gonad, hepatopancreas, foot, mantle and gill. RNA preparation and cDNA synthesis were conducted using the methods mentioned above. Gene-specific primers for qRT-PCR were designed (Table 1). The 25-μL reaction mixture consisted of 12.5 μL of 2× qPCR Mix, 2.0 μL of 7.5 μM primers, 2.5 μL cDNA and 8.0 μL of dd H₂O was programmed as follows: 10 min at 95˚C; 40 cycles at 95˚C for 15 s, 60˚C for 60 s and 72˚C for 30 s; and a melt from 75˚C to 95˚C.

A series of increasing Ca²⁺ concentrations, 0 (control group), 40, 60, 80 and 100 mg/L, were established by adding CaCl₂ into 20L-tanks. 15 mollusks were collected and averagely divided into these five groups. After a 7-days’ culture with aerated pond water at 25˚C, mantle tissue were sampled.

According to the former result, a calcium stimulation test with Ca²⁺ concentration of 60 mg/L was designed for a 7-days’ culture. Mantle samples of three individual mollusks were isolated at 0 h, 24 h, 48 h, 96 h and 168 h, respectively. All samples were immediately washed with sterile PBS, frozen in liquid nitrogen and stored at −80˚C. The expression profiles of HcPMCA mRNA were detected with the qRT-PCR method mentioned above.

When using qRT-PCR method, the β-actin was used as an internal control. All qRT-PCR tests were conducted three times using individual templates. The 2^−ΔΔCt method was used to calculate the relative expression level of HcPMCA [10]. Significant differences were determined by one-way analysis of variance and the Tukey’s HSD test for multiple-range comparison using SPSS 16.0 software, with significant levels accepted at p < 0.05.

The HcPMCA cDNA had a full-length of 5230 bp (GenBank accession No. KR080192.1), which comprised of a 417-bp 5'-UTR, a 3588-bp ORF, and a 1225-bp 3'-UTR. Also, it had a stop codon TAA and a poly (A) tail, but it had no poly (A) addition signal.

The deduced HcPMCA protein contained 1195 amino acids, within which Ile had the highest content (8.5%). The inferred HcPMCA protein had a theoretical molecular mass of 131.64 kDa and an isoelectric point (pI) of 6.05, but no putative signal peptide was predicted.

SMART analysis showed that the deduced amino acid sequence of HcPMCA presented typical characteristics of Ca²⁺-ATPase, which was constitute by five domains, including Cation transporter/ATPase, N-terminus (Cation ATPase N) domain (residues 40 - 115), E1-E2 ATPase domain (residues 136 - 288 and residues 345 - 460), haloacid dehalogenase-like hydrolase (HAD) domain (residues 468 - 811), Cation transporting ATPase, C-terminus (Cation ATPase C) domain (residues 884 - 1066) and Plasma membrane calcium transporter ATPase C terminal (ATP Ca trans C) domain (residues 1106 - 1166). These domains contained common structures for PMCAs, such as Ca²⁺ translocation site, phosphorylation site, ATP binding site (Figure 1). In addition, the analysis of the C-terminal region from HcPMCA (1106-1166 residues) predicts a potential CaM-BD based on the density of canonical binding motifs and a number of patterns, including 1 - 10, 1 - 12, 1 - 14, 1 - 16, 1 - 5 - 8 - 14, 1 - 8 - 14 and IQ-LIKE Ca²⁺/CaM-binding motifs were predicted (Table 2). Generally, PMCA consists of 10 transmembrane α-helices [11], which is also consistent with our result of HcPMCA.

The BLASTp search analyses demonstrated that HcPMCA shared the highest identity (74%) with T. squamosa PMCA (AML22897.1), followed by M. yessoensis PMCA 2-like (XP_021343203.1), Crassostrea virginica PMCA 2-like (XP_022299339.1), P. fucata PMCA (ABL63470.1) and Crassostrea gigas PMCA 2 (XP_019926614.1) successively with 71%, 69%, 68% and 66% identity respectively. In contrast, although the deduced amino acid sequence of HcPMCA has 60% similarity with human PMCA4b (P23634-6), which was a little lower than that between HcPMCA and other mollusk PMCA sequences, their functional domains are highly conserved (Figure 1).

Based on the BLASTp results and the reported PMCA homologs in mollusks and arthropods, a phylogenetic tree was constructed to show their evolutionary relationships (Figure 2). In this tree, PMCA homologs were grouped into two clusters. HcPMCA together with TsPMCA, PfPMCA, CgPMCA, CvPMCA and MyPMCA formed a cluster, and the other PMCA homologs from insects belonged to another cluster, indicating that HcPMCA had a closer evolutionary relationship with other mollusks’ PMCAs.

The tissue-specific expression of HcPMCA in the gonad, hepatopancreas, foot, mantle and gill was analyzed by qRT-PCR using β-actin as an internal control. HcPMCA mRNA was detected in all the tested tissues, and expressed the highest level in the gill, whereas the lowest level in the foot (Figure 3).

The HcPMCA transcripts in the mantle were significantly decreased under environmental calcium stimulation, with the highest expression level at the Ca²⁺ concentration of 60 mg/L, whereas the lowest level at the 40 mg/L (Figure 4(a)). Also, The temporal expression profile of HcPMCA at Ca²⁺ concentration of 60 mg/L was examined, which showed that mRNA expression of HcPMCA gene in mantle was bell-shaped with the exposure time increased. Detailly, the HcPMCA transcripts in mantle were slightly decreased at 24 h after treatment, but were upregulated from 24 to 48 h post-treatment, peaking at 48 h post-treatment (Figure 4(b)). Moreover, the HcPMCA transcripts were downregulated at the following period, with the expression amounts at 96h and 168 h post-treatment were nearly in line with that at 24 h.

Compared with PMCA homologs from seawater mollusks, HcPMCA had high similarity with them in both se­quence and structure. Despite some differences in amino acid sequence, our results suggest that HcPMCA folds in a similar manner to “P-type” Ca²⁺-ATPase proteins from higher eukaryotes, and contains all the conserved protein domain found in other P-type ATPases [12] [13], suggesting that HcPMCA might possess the same physiological function as other PMCA homologs. Meanwhile, it’s found that a series of predicted patterns, such as 1 - 10, 1 - 12, 1 - 14, 1 - 16 and IQ-LIKE motifs at the C-terminal region from HcPMCA were the canonical classical Ca²⁺/CaM-binding ones that have been described in other CaM-modulated proteins such as CaM-dependent protein-kinase II, MLCK and KCNQ channels [14] [15] [16], indicating HcPMCA may be one CaM-modulated protein.

As we all know, Mammalian PMCA exists in four isoforms, with each isoform possessing multiple splice variants [17]. PMCA1 and 4 are generally universally expressed and serve housekeeping functions. Whereas, PMCA2 and 3 are differentiated by cell type [18]. Here, aignment between HcPMCA and Eukaryota PMCAs using SMART revealed that HcPMCA had the same domain with many PMCAs, such as PMCA1 from Homo sapiens, PMCA1 from Sus scrofa, PMCA1 from Rattus norvegicus, PMCA2 from Homo sapiens, PMCA2 from Mus musculus, PMCA3 from Homo sapiens and PMCA4 from Homo sapiens. Obviously, it is necessary to carry out further exploration that involves the separate isoforms of this gene and concrete functions in the calcium homeostasis of H. cumingii.

Both freshwater and seawater mollusks absorb calcium actively from environment, but their calcium metabolism mechanisms are significant differences. Some of the Ca²⁺ absorbed by freshwater mollusks can be stored as calcium spheres for later usage [19], while seawater mollusks are opposite. In freshwater mollusks, foot is the main locomotion organ, gonads are an important part of reproductive system, hepatopanpancreas are closely related to digestive and immune functions, mantle is an important tissue secreting calcium to the mineralization site [20], and gill is a pivotal tissue for calcium uptake from water [21]. Tissue-specific expression analysis revealed that HcPMCA mRNA was detected in all these five tissues, indicating HcPMCA may be a housekeeping PMCA isoform. Among these sampled tissues, mantle is the site of pearl formation, and is highly permeable to Ca²⁺ [22] [23]. Both of its inner and outer epidermis have the functions of absorbing and storing Ca²⁺ actively with a high utilization rate of Ca²⁺ [24]. Gill, as respiratory and filter feeding organ of mussels, is another important tissue for calcium metabolism with a strong affinity and a high metabolic rate of Ca²⁺. Therefore, calcium metabolism in the mantle and gill is vigorous, and the maintenance of calcium homeostasis is especially vital. The highest expression level of HcPMCA in the gill and mantle indicates that the gene plays an important role in Ca²⁺ transportation and extruding, engaging in calcium homeostasis and biomineralization processes, such as pearl formation. In addition, the lower mRNA expression of this gene in hepatopancreas, gonad and foot, may indicate that the abilities of extruding calcium from the cytosol into the extracellular space of these there tissues were weak. Further studies on the mechanisms of calcium absorption, storage and transport are still needed.

Environmental calcium concentration is one of the important factors affecting calcium metabolism of mollusk. when Ca²⁺ was added, the flow direction of extracellular Ca²⁺ gradually changed from efflux to internal flow, and the flow velocity of Ca²⁺ increased with the increase of Ca²⁺ concentration, which resulted in the enhancement of intracellular fluorescence signal [19]. A similar view was drawn in this study. Here, the HcPMCA transcripts in the mantle were significantly decreased after calcium was added into the aquatic water, suggesting the existence of free enviromemtal Ca²⁺ may inhibit the expression of this gene, accumulating more cytosolic calcium concentration in cells in order to guarantee future use.

The regulatory factors related to calcium metabolism may participate in or regulate the formation of shell/pearl through calcium absorption, transport, storage and deposition. The absorption and transport of Ca²⁺ play an important role in improving the yield of pearls in the freshwater pearl cultivation process. Accelerating the deposition of Ca²⁺ in the pearl sac is an important way to promote the growth of pearls rapidly. However, the calcium content is very low in natural water. Thus, maintaining appropriate Ca²⁺ concentration in water is a prerequisite for the growth of pearls. A appropriate concentration of Ca²⁺ could promote the absorption, transport and storage of Ca²⁺ in the mantle of mussels, promoting the calcium metabolism process and accelerating the formation of pearls, while excessive content of Ca²⁺ would inhibit the calcium metabolism of mantle [25] [26].

In this study, the expression level of HcPMCA gene in the mantle reached the highest expression levels at the Ca²⁺ concentration of 60 mg/L, and the lowest level at the 40 mg/L. It was speculated that the mantle secreted calcium actively to the biomineralization site to participate in pearl growth at the the Ca²⁺ concentration of 60 mg/L, resulting in enhancement of extruding function. While, the result was opposite when the Ca²⁺ concentration was 60 mg/L. In addition, the present study also found that the mRNA expression of HcPMCA gene in mantle was bell-shaped with the exposure time increased at Ca²⁺ concentration of 60 mg/L, and peaked at 48 h post-treatment, indicating that the mantle of molusck had a strong ability of transporting Ca²⁺ at 48 h, and then decreased.