Calcineurin subunit B is involved in shell regeneration in Haliotis diversicolor

Experimental animals, RNA isolation and cDNA synthesis

Adult abalone H. diversicolor, about 2 years of age, 55.0 mm in size and 10.0 g in weight, were reared in polyethylene tanks at 23−25 °C with a salinity of 28–30 ppt at Phuket Abalone Farm, Phuket, Thailand and fed daily with fresh kelp as previously described (Buddawong et al., 2020). Animal handling followed the guideline of Animal Care and Use Committee (SCMU-ACUC, Protocol Number MUSC60-040-390), Faculty of Science, Mahidol University. Prior to sampling, animals were anesthetized with 0.1 M KCl until no movement was detected. The mantle tissues were carefully dissected and stored in RNAlater RNA stabilization reagent (Ambion, Austin, TX, USA) prior to RNA extraction. Total RNA was extracted with Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol and genomic DNA was eliminated by DNAse treatment (Thermo Fisher Scientific, Carlsbad, CA, USA). RNA was reverse transcribed into cDNA using the SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen) with random hexamers following the manufacturer’s instructions.

Shell notching experiment

To investigate the expression patterns of HcCNA and HcCNB in shell damaged abalone, the shell notching experiment was performed according to the method of (Mount et al., 2004) with some modifications. The V-shaped notch (approximately 10 mm²) was cut on the right margin of the abalone shells (n = 21) without damaging the mantle tissues. Animals were randomly divided into seven groups, each group containing three individuals. The seven groups were returned to seawater tanks and the abalones were sacrificed at 0, 3, 6, 12, 24, 36, and 48 h after shell notching. Approximately three cm² area of the mantle tissue around the notching area was collected and immediately kept in RNAlater reagent for further extraction with Trizol reagent and DNAse treatment under the same condition described above. The mantle tissues of three abalones without shell notching were also collected as a control.

Gene expression analysis by quantitative real-time PCR (qPCR)

qPCR was used to quantify the expression levels of HcCNA and HcCNB. As described previously (Buddawong et al., 2020), 1 µg of the total RNA from mantle was reverse transcribed into cDNA and qPCR was performed in triplicate using a Bio-Rad CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) with Luna Universal qPCR Master Mix (New England Biolabs, Ipswich, MA, USA). The 20 µl reaction mixture contained 0.5 µl (20 ng) of cDNA, 10 µl of Luna Universal qPCR mix with Hot Start Taq DNA Polymerase, 0.5 µl of 10 µM of each primer (for primer details see Table 1), and 8.5 µl of PCR grade water. The cycling parameters were 95 °C for 1 min, 45 cycles of 95 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s. A melting curve analysis was performed at a final single cycle to determine the specificity of PCR amplification by increasing the temperature from 60 °C to 95 °C in the rate of 0.05 °C/sec. The baseline was set automatically by Bio-Rad CFX manager software (version 3.1). PCRs with no template controls were also performed for each primer pair. Relative expression levels were calculated by the Livak (2^−ΔΔCq) method using β-actin as a reference gene because of its previously determined stable expression in H. diversicolor (Li et al., 2012).

Expression and purification of recombinant HcCNA (rHcCNA) and HcCNB (rHcCNB) in E.coli

To express the recombinant HcCNA and HcCNB in E.coli, the coding regions of HcCNA and HcCNB obtained from the previous study (Buddawong et al., 2020) were amplified by PCR using 2 pairs of primers (expressed CNA-F and expressed CNA-R for HcCNA and expressed CNB-F and expressed CNB-R for HcCNB) which contained the positions of NcoI/BamHI digestion and NdeI/BamHI digestion for HcCNA and HcCNB, respectively. The PCR products were purified with FavorPrep GEL/PCR purification kit (Favorgen, Ping-Tung, Taiwan) and digested with NcoI/BamHI and NdeI/BamHI for HcCNA and HcCNB, respectively. The digested HcCNA was inserted into pET-16b while that of digested HcCNB was inserted into pET-15b (Novagen, Darmstadt, Germany). The recombinant plasmids were confirmed by restriction analysis and DNA sequencing and transformed into E. coli strain BL21 (DE3, Novagen). Protein expression was induced with 1 mM isopropylthiogalactopyranoside (IPTG) at 37 °C when the optical density (600 nm) of the culture had reached 0.6. Bacterial cells were harvested by centrifugation (5,000× g, 10 min, 4 °C) after 4 h of induction.

For protein purification, the bacterial pellets were washed twice with PBS, centrifuged, and the bacterial pellets were suspended in ice-cold lysis buffer (50 mM Tris-HCl, 140 mM NaCl, 5 mM dithiothreitol (DTT), and 2 mM phenylmethylsulfonyl fluoride (PMSF), pH 8.0). Cell disruption was achieved by sonication on ice. After centrifugation (10,000× g, 30 min, 4 °C), crude proteins in supernatant were filtered through 0.45 µm sieve and loaded onto a nickel affinity chromatographic column. The column was washed with washing buffer (50 mM Tris-HCl, 140 mM NaCl, 20 mM imidazole, pH 8.0) and then eluted with elution buffer (50 mM Tris-HCl, 300 mM NaCl, 300 mM imidazole, pH 8.0).

Protein profiling and Western blot analysis of rHcCNA and rHcCNB

The obtained fractions of rHcCNA and rHcCNB from the affinity column were analyzed by sodium dodecyl sulfate-polyacrylamine gel electrophoresis (SDS-PAGE) and stained with Coomassie Brilliant Blue. The dissolved proteins were transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Bedford, MA). The membranes were submersed in 10% skim milk with 0.05% Tween 20 in TBS for overnight at 4 °C and were incubated in mouse anti-His polyclonal antibody (GE Healthcare, Chicaco, IL) at a dilution of 1:3,000 (2 h, room temperature). After washing with TBS-0.05% Tween (TBST), the membrane was incubated with goat anti-mouse IgG-HRP (Sigma Co., St. Louis, MA) at a concentration of 1:5,000 (1 h, room temperature). The reactivity of the antibody was detected by an enhanced chemiluminescence method using an ECL kit (Amersham Biosciences, Buckinghamshire, UK).

Calcineurin enzymatic activity assay

The recombinant calcineurin enzymatic activity was performed using a colorimetric calcineurin phosphatase activity assay kit (Abcam, Cambridge, MA) based on a Malachite green assay. Briefly, the rHcCNA, rHcCNB, calmodulin and calcineurin assay buffer (100 mM Tris, pH 7.5, 1 mM DTT, 0.05% NP-40, 1 mM CaCl₂) were mixed, the RII calcineurin substrate (Enz et al., 1994) was added, and the reaction was terminated by adding a Malachite green reagent. After allowing color to develop for 20 min, optical density (OD 620 nm) was read on a Versamax tunable microplate reader (Molecular Devices, San Jose, CA). The absorbance data were converted into released phosphate amount, and calcineurin activity was calculated as a ratio of phosphate amount/reaction time.

Analysis of calcium-binding activity

The properties of the Ca²⁺-binding activity of the rHcCNB were investigated by ruthenium red staining. The rHcCNB was resolved by SDS-PAGE and was then transferred to a PVDF membrane prior to staining with Ponceau S or ruthenium red (25 mg/l of ruthenium red in 60 mM KCl, 5 mM MgCl₂, 10 mM Tris–HCl, pH 7.5) for 48 h (Will et al., 2007). Bovine serum albumin (BSA) was used as a negative control.

The recombinant calcineurin treatment in vivo

To investigate the role of the recombinant calcineurin in shell regeneration, the shell notching experiments were performed according to the method described above. After notching, the abalones were randomly divided into 3 groups, each contained 9 individuals. In all experimental groups, test substances were injected into foot tissue. Group 1 animals (control) were injected with 50 µl of sterile saline solution (0.85% NaCl). Group 2 animals were injected with 100 µg of rHcCNB (about 10 µg/g animal) in 50 µl sterile saline solution. Group 3 animals were injected with 200 µg of rHcCNB (about 20 µg/g animal) in 50 µl sterile saline solution. All animals were maintained in seawater tanks and a subset of abalones from each group were sacrificed after 1, 2 and 3 weeks to collect their shells for measurements of the length and area of regenerated shells and to observe ultrastructure. The length of regenerated shell was measured perpendicular to the shell circumference, from the proximal apex of the original notch to the edge of the new shell growth. The area of regenerated shell represented the area of the original triangular notch that had been infilled with new shell growth (Fig. S1). Regenerated length and area were analyzed by using ImageJ software (http://rsb.info.nih.gov/ij/), and were expressed as a percentage of the original notch length and area to account for minor differences in original notch size.

Scanning electron microscopy (SEM)

To visualize the growth of 27 regenerated shells, the internal connective tissues of the shell were removed and the shells were washed extensively with distilled water. Thereafter, the shells were left to air-dry in petri dishes at room temperature for several days, cut into 1 cm ×1 cm pieces, and sputter coated with gold using a SPI gold coater (West Chester, PA). The ultrastructural features of the inner surface of the regenerated shells were observed by a Jeol Neoscope JCM-5000 at 15 kV.

Statistical analysis

Statistical analyses were performed using a one-way ANOVA test followed by a post hoc analysis (Tukey’s multiple comparison test) using IBM SPSS Statistics Processor (IBM, Armonk, NY). Data were expressed as mean ± standard deviation. A P-value (p) of <0.05 was considered statistically significant.

HcCNA and HcCNB expression after shell notching

To monitor the response of HcCNA and HcCNB genes in the shell regeneration process, the expression levels of HcCNA and HcCNB genes were examined by quantitative real-time PCR at each time point after shell notching (Fig. 1). All PCR assays displayed efficiency between 91–97% and R² values were more than 0.995. It was noted that the expression level of HcCNA was up-regulated at 6 h after shell notching and reached the highest level of about 2.5 fold at 12 h. Thereafter, it was down-regulated at 24, 36, and 48 h. In contrast, the expression level of HcCNB significantly increased to almost 4 times higher than HcCNA at 6 h and peaked at approximately 3 times that of the HcCNA level at 12 h after shell notching. Thereafter, expression of HcCNB gradually decreased, however expression levels remained higher than that of the levels of HcCNA at the same time intervals.

Purified recombinant HcCNA and HcCNB proteins in E. coli exerted their phosphatase activity

We produced the recombinant proteins, rHcCNA and rHcCNB, in E. coli and initially tested the general phosphatase activity as well as the property of Ca²⁺-binding activity of the rHcCNB. Upon its purification with Ni-NTA chromatography, the rHcCNA protein was visible as a single band with an approximate purity of >95%. The protein yield per one purification was approximately 260 µg/mg bacterial extract. Its apparent molecular mass was approximately 59 kDa which was consistent with the predicted molecular mass of HcCNA (Fig. 2A). In a Western blot using an anti-His antibody, only a single immunoreactive band at 59 kDa was observed (Fig. 2B). Similar results were obtained for purification of rHcCNB, where >95% purity and a yield of about 300 µg/mg crude protein was obtained. As also shown in Figs. 2A and 2B, the expressed rHcCNB was shown as a single-band purified protein at about 19 kDa which was consistent with the predicted molecular mass of HcCNB and a single band at 19 kDa was observed by a Western blot analysis with monoclonal anti-His antibody.

Generally, the phosphatase function of calcineurin is dependent on dimerization of the calcineurin subunits as well as the presence of a calmodulin catalytic counterpart (Rusnak & Mertz, 2000). In this study, we tested this enzymatic combination followed by the detection with a Malachite green RII calcineurin substrate to measure the amount of phosphate released at different reaction time points. Approximately 0.05 nmol of phosphate was detected at 10 min. This level gradually increased at 30, 40, 60 min and the highest amount of phosphate released was approximately 0.2 nmol at 90 min of the reaction time (Fig. 3A). Together, the results suggested that purified rHcCNA and rHcCNB produced individually from E. coli exerted comparable phosphatase activity to that previously reported for CN (Rusnak & Mertz, 2000).

In addition to examination of the phosphatase activity of both subunits, the Ca²⁺-binding activity of rHcCNB was investigated by colorimetric assays using ruthenium red (Fig. 3B). Ruthinium red which has been widely used for analysis of Ca²⁺-binding of proteins, was found to achieve the same results as the ⁴⁵Ca²⁺ overlay technique, and also stained known calcium-binding proteins such as the EF hand Ca²⁺-binding proteins (Charuk, Pirraglia & Reithmeier, 1990; Corbalan-Garcia, Teruel & Gomez-Fernandez, 1992). Here, after the proteins were separated on SDS-PAGE, they were transferred into a PVDF membrane and stained with Ponceau S or ruthenium red. The rHcCNB was stained by ruthenium red, but BSA (the negative control) was not stained, suggesting that rHcCNB had Ca²⁺-binding activity.

The effect of rHcCNB on the regenerated shell length and area

Since our qPCR results indicated that HcCNB was significantly up-regulated more than HcCNA after shell notching, and that the role of CNB as an independent modulator has been well documented, we decided to further investigate the independent effect of rHcCNB on shell regeneration in vivo. The animals were injected with 0.85% NaCl, 100 µg, or 200 µg of rHcCNB as described above and their shells were collected at 1, 2, and 3 weeks after the treatment. There was no abalone mortality observed and all animals showed evidence of shell regeneration. The regenerated shell length and area were shown as the percentages of the notched region that had been refilled. The percent shell length from three groups were shown in Fig. 4A and their representative images were shown in Figs. S1A–S1C. At one and two weeks after treatment, the percentages of the regenerated shell lengths of all groups showed no significant difference. At three weeks after treatment, the regenerated shell lengths of abalones in Group 3 were significantly higher than that of control groups. The percent shell growth at week 3 (Group 3) exceeded 100% of the original length of the notch. On the other hand, the percentages of regenerated areas showed no significant difference compared to the control at all weeks after treatment (Fig. 4B).

The effect of rHcCNB on the ultrastructure of the regenerated shell

We further investigated ultrastructure of the regenerated shell upon treatment with rHcCNB. Scanning electron microscopy of the inner surface of the regenerated shell revealed three distinct zones (Fig. S1D). The outermost, or prismatic (P), layer was located at the edge of the regenerated shell. The innermost layer, the nacreous (N) layer, was located proximally to the prismatic layer. A band with distinct ultrastructure that was located between the prismatic and nacreous layers was named transition zone (T).

Upon treatment with rHcCNB after shell notching, the inner surfaces of the regenerated shells changed their structures drastically compared with the control group. At weeks 1 and 2 after shell notching and treatment, the ultrastructure of prismatic layers of all groups contained numerous small fine-grained crystals that aggregated to form various sized prolate spheroids (Figs. 5A–5F). At week 3, the ultrastructure of the crystals differed significantly between control (Fig. 5G) and rHcCNB injected animals (Figs. 5H–5I). The structure at the transition zone (T) at weeks 1 and 2 was not different from the control group (Figs. 6A–6F). The transition zones were easily identifiable, containing fine-grained crystals that aggregated to form larger lattices. Of particular interest, the transition zone was not observed in the rHcCNB injected animals at week 3 after treatment (Figs. 6H–6I). In the nacreous layer, the polygonal crystal tablets of regenerated shell generally resembled the pyramidal stacks of the control group at week 1 (Figs. 7A–7C). However at week 2 and 3, the organization of tablets did not appear as stacks in Group 3, but rather as discontinuous brick columns. The diameter of the columns in Group 3 was much larger than that of the stacked tablets, resulting in the reduction of the space between units (Figs. 7F and 7I).