PfmPif97-like regulated by Pfm-miR-9b-5p participates in shell formation in Pinctada fucata martensii

Mollusk shell matrix proteins are important for the formation of organic frameworks, crystal nucleation, and crystal growth in Pinctada fucata martensii (P. f. martensii). MicroRNAs (miRNAs) are endogenous small non-coding RNAs that play important roles in many biological processes, including shell formation. In this study, we obtained the full-length sequence of Pif97-like gene in P. f. martensii (PfmPif97-like). PfmPif97-like was mainly distributed in mantle pallial and mantle edge. Correlation analysis indicated that the average shell thickness and weight showed a positive correlation with PfmPif97-like expression (P < 0.05). The inner surface of the nacreous layer and prismatic layer showed atypical growth when we knocked down the expression of PfmPif97-like by RNA interference (RNAi). We used a luciferase reporter assay to identify that miR-9b-5p of P. f. martensii (Pfm-miR-9b-5p) downregulated the expression of PfmPif97-like by interacting with the 3′-untranslated region (UTR) while we obtained the same result by injecting the Pfm-miR-9b-5p mimics in vivo. After injecting the mimics, we also observed abnormal growth in nacre layer and prismatic layer which is consistent with the result of RNAi. We proposed that PfmPif97-like regulated by Pfm-miR-9b-5p participates in shell formation of P. f. martensii. These findings provide important clues about the molecular mechanisms that regulate biomineralization in P. f. martensii.

SMPs are not only participants of organic framework but also crystal regulators [9]. On vertebrates, miRNAs act as regulators of extracellular accumulation, osteoclast and osteoblast differentiation, transcription factor expression, and growth factor secretion [25]. miRNA regulators exist not only in vertebrates but also in invertebrates. Several miRNAs in mollusk were identified by using Solexa deep sequencing or bioinformatics analysis [26]. miR-29a [27], miR-183 [28], and miR-2305 [29] are involved in shell formation via the downregulation of matrix protein gene expression. These pieces of evidence signify that miRNAs commonly regulate the matrix protein gene to participate in shell formation.
Although some SMPs have been identified, the detailed molecular mechanisms of shell biomineralization remain poorly understood. In this study, we identified a matrix protein gene PfmPif97-like and focused on the function of biomineralization and miRNA regulation.

Experimental materials
The experimental animals P. f. martensii (5-6 cm shell length) were sampled from Liushawan, Zhanjiang, in the South China Sea. The animals were temporarily farmed with circulating seawater until use.

RNA isolation, cDNA synthesis, gene cloning, and real-time PCR assay
Total RNA was prepared using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions with some modification (https://dx.doi.org/10.17504/ protocols.io.9qgh5tw) and cDNA was synthesized by M-MLV reverse transcriptase (Promega, USA). miRNAs were extracted by using SanPrep Column microRNA Extraction Kit (Sangon Biotech) and miRNA First-Strand was synthesized by using Mir-X miRNA First-Strand Synthesis Kit (TaKaRa). The 3 0 and 5 0 ends of the PfmPif97-like gene were cloned by using rapid amplification of cDNA ends (RACE) technology. The expression level was detected by Realtime PCR (RT-PCR) with DyNAmo Flash SYBR Green qPCR kit (Thermo Scientific) on a Roche LightCycler 480 (Roche, Switzerland). The PCR program was conducted as follows: 5 min at 95˚C and 40 cycles (each cycle was for 30 s at 95˚C, 15 s at 60˚C, and 15 s at 72˚C) The relative expression level of the target genes was calculated through the 2 (CT β-actin-CT Target gene) method, and β-actin was used as the reference gene. All primers and mimics sequences used in this study are listed in Table 1.

Expression distribution pattern of PfmPif97-like and Correlation analysis
In situ hybridization was used to determine the expression distribution of PfmPif97-like in the mantle. RNA probes were prepared by in vitro transcription using T7 RNA polymerase and digoxigenin (DIG) RNA Labeling Mix. RNA probe integrity was detected by using 1% agarose gel electrophoresis, and the quality of probes was analyzed in conjunction with the RNA concentration and purity determined by using a nucleic acid quantifier. The mantle tissues were fixed with 4% paraformaldehyde for 2 hours at 4˚C. Fixative volume is over 20 times that of tissue on a weight per volume. Then the tissue was dehydrated through a series of graded ethanol baths to displace the water, and then infiltrated with wax. The infiltrated tissues were then embedded into wax blocks. Then the mantle tissue were cut into 7 μm in thickness via the instrument of LEICA RM2235. In situ hybridization was carried out according to the instructions of Enhanced Sensitive ISH Detection Kit I (POD) (BOSTER) with some modification. This protocol has been deposited in protocols.io (dx.doi.org/10.17504/protocols.io.9qhh5t6).
Thirty-five normal pearl oysters were select to perform the notching assays. Thirty normal pearl oysters were select and cut a "V" shaped notch until the nacreous layer of the shell. Collected the mantle edge of every five pearl oysters at 2h, 4h, 6h,12h, 24h and 36h after damage and harvested the mantle edge of five pearl oysters (no notching) at 0 h. The larval sample is the same sample as the sample of the developmental transcriptome in the previous genomic article.

Primer
Sequence Application The bases of lowercase letters are restriction enzyme sites. The underlined bases are the T7 promoter sequence. https://doi.org/10.1371/journal.pone.0226367.t001 The Pearson correlation coefficient between PfmPif97-like expression in mantle edge and growth traits was examined using Pearson test in SPSS 22.0. A total number of 21 normal pearl oyster were used for Pearson correlation coefficient analysis. Shell parameter contains shell length, shell width, shell height, shell weight and shell thickness.

PfmPif97-like function interference experiment in vivo
Double-stranded RNA (dsRNA) was synthesized using the T7 High-Efficiency Transcription Kit (TransGen Biotech, JT101) and purified using EasyPure RNA Purification Kit (TransGen Biotech, ER701). Fifteen P. f. martensii individuals of similar size (5-6 cm shell length) were randomly divided into three groups. Double-stranded RNA (dsRNA) (60 μg/100 μL) and diethyl pyro carbonate (DEPC) water were injected into the adductor muscle. 100 μL of dsRNA-PfmPif97-like was injected as the experimental group. For control groups, 100 μL of DEPC water was injected and 100 μL of dsRNA-Red Fluorescent Protein (RFP) was injected. On the sixth day after injection, the mantle tissue was collected, frozen rapidly in liquid nitrogen, and stored at −80˚C. The shells were cut into 0.5 cm × 1.5 cm size and washed with ultrapure water. The shell samples were air dried, and the inner surface of the shell was observed by using scanning electron microscope (SEM) in 15 kV.

Target verification between Pfm-miR-9b-5p and PfmPif97-like in vitro and in vivo
The PfmPif97-like 3 0 UTR containing the target site was cloned by using PrimeSTAR HS DNA Polymerase (purchased from takara). Then the sequence was inserted into a pMIR-reporter plasmid (purchased from Ambion) (pMi-UTR) by using restriction enzyme. We also constructed a mutant plasmid (pMi-MUTR) as the control. The Pfm-miR-9b-5p mimics and negative control mimics (N. C.) were compounded from GenePharma. The pMi-UTR and pMi-MUTR groups were blank control group. The N. C. and pMi-UTR co-transfection group, N. C. and pMi-MUTR co-transfection group, and Pfm-miR-9b-5p mimics and pMi-MUTR cotransfection group were negative control group. The Pfm-miR-9b-5p mimics and pMi-UTR co-transfection group was the experimental group. Each group was co-transfected with 22.5 ng of the pRL-TK vector as internal quality control to determine the relative activity in a dualluciferase report system. Luciferase activity was detected using the dual-luciferase assay kit (Promega) with a microplate reader at 48 h after transfecting to HEK-293T cells cultured in 48-well plate. HEK-293T cells were cultured at 37˚C in DMEM/HIGE GLUCOSE medium containing 10% fetal bovine serum in a CO 2 incubator with 5% CO 2 .
Pfm-miR-9b-5p mimics and N. C. mimics were diluted to 0.1 μg/μL of RNase-free water, and 100 μL was injected into the adductor muscles. The N. C. group served as the negative control. Five P. f. martensii individuals (5-6 cm shell length) were used for each group. The mantle tissue and shell of each sample were collected correspondingly and the expression level was detected using RT-PCR. The inner surface of the nacre and prism layer was observed by using an SEM.

Cloning and characterization of PfmPif97-like
The full-length cDNA of PfmPif97-like (GenBank accession numbers: MK962312) was 2356 bp, containing an 1821 bp ORF encoding a putative protein of 606 aa, 5 0 UTR of 47 bp, and 3 0 UTR of 458 bp (Fig 1). PfmPif97-like contains a VWA domain and two ChtBD type 2 (ChtBD2). The local BLASTp search indicated that the amino acid sequence of PfmPif97-like showed 26%-41% identity to those of PfmPif97, PmPif97, PmxPif97, PpPif97, and CgPif97 (S1 File). Sequence alignment of VWA and ChtBD2 domains from PfmPif97-like, Pif, and Pif homologs from several mollusks revealed the conservation functional site in the domain (Fig  2).

Expression distribution, expression pattern and correlation analysis
Result of in situ hybridization demonstrated that PfmPif97-like was mainly distributed in the mantle pallial, mantle edge and was little distributed in the central zone of mantle (Fig 3). PfmPif97 expression was significantly increased at 2 h, 4 h and 36 h after shell damage (Fig 4A) while PfmPif97-like expression was significantly increased at 6 h, 12 h and 36 h (Fig 4A). In larval development stage, PfmPif97 was highly expressed in gastrula, eye-spotted larvae and postveliger stage (Fig 4B) while PfmPif97-like were highly expressed in blastula and eye-spotted larvae ( Fig 4B).
The Pearson correlation coefficient showed PfmPif97-like expression exhibited significant correlation with wet shell weight (R = 0.753, P < 0.05), dry shell weight (R = 0.762, P < 0.05), the average thickness of the left shell (R = 0.751, P < 0.05) and the average thickness of the right shell (R = 0.762, P < 0.05) ( Table 2). However, no significant correlation was observed between PfmPif97-like expression and shell width, height, and length.

PfmPif97-like interference disrupted shell biomineralization
The interference experiment showed PfmPif97-like expression levels in the mantle pallial, mantle edge and central zone of mantle were significantly downregulated compared with those of the control group (P < 0.05) (Fig 5A). The nacre of the corresponding shell grew disorderly and slowly (Fig 5B), and the prismatic layer changed from angular to fragmented ( Fig  5B).

Pfm-miR-9b-5p negatively regulated PfmPif97-like expression
The potential target site between the 3 0 UTRs of PfmPif97-like and Pfm-miR-9b-5p was obtained by using miRanda and RNAhybrid software (Fig 6A). In vitro, luciferase activity was downregulated in Pfm-miR-9b-5p mimics and pMi-UTR co-transfection group compared with that in the control groups (P < 0.05) (Fig 6B). The expression of Pfm-miR-9b-5p was significantly upregulated after injecting the Pfm-miR-9b-5p mimics, whereas that of PfmPif97-like was significantly downregulated (Fig 6C). The surface of the nacre and prism layer was disordered (Fig 6D).

Discussion
SMPs are considered structural and functional macromolecules [6,7,11]. The function of matrix protein in shell formation has been extensively studied, including nacrein [5], Pif [9], ACCBP [32], PNU7 [19], Pfy2 [18], and N19 [7]. However, the mechanism of shell formation is not fully understood. In this study, we obtained the full-length sequence of PfmPif97-like which is similar to PfmPif97. Whether in the nacre or in the prismatic layer, chitin serves as the key component of the organic framework [3], building compartment structures and The amino acid sequence with the black bold font is signal peptide and with gray background is VWA domain. The underlined amino acid sequences are two ChtBD2 domains. The nucleotide with a frame represents the start and stop codons. The sequence before ATG is the 5 0 UTR region. The sequence after TAA is the 3 0 UTR region.
https://doi.org/10.1371/journal.pone.0226367.g001 linking with matrix protein to the morphology of calcium carbonate crystals [33,34]. In addition to chitinase [35], the matrix protein with the chitin-binding domain is related to chitin [22]. The chitin-binding domain is an extracellular domain that contains six conserved cysteines that probably form three disulfide bridges [36], which are also found in PfmPif97-like. The recombinant PfmPif97 is a framework protein for the association of chitin-aragonite [22]. Similar to PfmPif97, PfmPif97-like has two ChtBD2, indicating that PfmPif97-like may be involved in the formation of the organic framework by binding β-chitin. PfmPif97-like also has a VWA domain which is a family of 200-amino-acid residues and works as an interaction module [37]. VWA-containing proteins are widely found in P. margaritifera [38], Mytilus coruscus [39], Crassostrea gigas [40], Lottia gigantea [41], and P. f. martensii [20]. In molluscks, VWA-containing proteins may be involved in protein-protein interactions, providing initial hydrogel properties for biomineralization [20,37]. Thus, PfmPif97-like may act as a medium for connecting chitin and SMP like PfmPif97.
We found PfmPif97-like was mainly expressed in the mantle pallial, mantle edge and had a small distribution in the central zone of mantle, which are regions responsible for the nacreous layer and prismatic layer formation. And the PfmPif97-like expression level in mantle edge was significantly correlated with shell weight and thickness. This showed that PfmPif97-like may participate in oyster shell formation. PfmPif97 expression was significantly increased at 2 h, 4 h, and 36 h after shell notching. However, PfmPif97-like expression was significantly increased at 6 h, 12 h and 36 h. The expression pattern of PfmPif97-like in the notching experiment is similar to that of alveoline-like protein [42], which plays essential role in shell formation. Shell notching causes mantle tissue retraction, causing an immune response and disturbing shell protein secretion. [43]. After that, the shell regeneration process is mainly carried out [43]. Thus, PfmPif97-like and PfmPif97 participate in shell regeneration, and Pfmpif97-like might be less related to immune response after shell notching than PfPif97. PfmPif97-like were highly expressed in the blastula and PfmPif97 was highly expressed in the gastrula. This indicates that they have the different function at the early larval development stage. Li found that a large number of genes involved in the calcium signaling pathway and synchronization with the shell formation were upregulated during the eye-spotted stage [44]. Thus, PfmPif97-like and PfmPif97 are highly expressed in the eye-spotted larvae suggesting that they play a role in shell formation. To directly determine whether PfmPif97-like is involved in shell formation, we knocked down the expression level of PfmPif97-like by using RNAi and found that the nacre and prism layers grew disorderly. Therefore, PfmPif97-like participates in shell formation, possibly by linking chitin and other SMPs.
Mollusk shell formation is a complex process that requires precise regulation [3,26]. miRNA is a negative regulator [45]. miRNAs target approximately 60% of genes of mammals [46] showing their important biological functions. miRNAs, such as miR-29a, miR-183, and miR-2305, were found to participate in shell formation by targeting biomineralization genes [27][28][29]. These show that miRNAs are generally involved in shell formation. In the present study, we found that Pfm-miR-9b-5p may be targeting PfmPif97-like 3 0 UTR, and the relationship of regulation was verified by dual-luciferase system in vitro. In vivo, upon the overexpression of Pfm-miR-9b-5p via injection with Pfm-miR-9b-5p mimics, the expression of PfmPif97-like regulated by Pfm-miR-9b-5p participates in shell formation PfmPif97-like was downregulated, and the nacre and prism layers exhibited atypical growth, similar to the results of RNAi. These show that PfmPif97-like is downregulated by Pfm-miR-9b-5p and participates in shell formation.

Conclusion
We discover a matrix protein gene similar to PfmPif97, named PfmPif97-like, from P. f. martensii and we ascertained that PfmPif97-like is regulated by Pfm-miR-9b-5p and participates in