Novel globular C1q domain-containing protein (PmC1qDC-1) participates in shell formation and responses to pathogen-associated molecular patterns stimulation in Pinctada fucata martensii

The C1q protein, which contains the globular C1q (gC1q) domain, is involved in the innate immune response, and is found abundantly in the shell, and it participates in the shell formation. In this study, a novel gC1q domain-containing gene was identified from Pinctada fucata martensii (P. f. martensii) and designated as PmC1qDC-1. The full-length sequence of PmC1qDC-1 was 902 bp with a 534 bp open reading frame (ORF), encoding a polypeptide of 177 amino acids. Quantitative real-time PCR (qRT-PCR) result showed that PmC1qDC-1 was widely expressed in all tested tissues, including shell formation-associated tissue and immune-related tissue. PmC1qDC-1 expression was significantly high in the blastula and gastrula and especially among the juvenile stage, which is the most important stage of dissoconch shell formation. PmC1qDC-1 expression was located in the outer epithelial cells of mantle pallial and mantle edge and irregular crystal tablets were observed in the nacre upon knockdown of PmC1qDC-1 expression at mantle pallial. Moreover, the recombined protein PmC1qDC-1 increased the rate of calcium carbonate precipitation. Besides, PmC1qDC-1 expression was significantly up-regulated in the mantle pallial at 6 h and was significantly up-regulated in the mantle edge at 12 h and 24 h after shell notching. The expression level of PmC1qDC-1 in mantle edge was significantly up-regulated at 48 h after LPS stimulation and was significantly up-regulated at 12 h, 24 h and 48 h after poly I:C stimulation. Moreover, PmC1qDC-1 expression was significantly up-regulated in hemocytes at 6 h after lipopolysaccharide (LPS) and poly I:C challenge. These findings suggest that PmC1qDC-1 plays a crucial role both in the shell formation and the innate immune response in pearl oysters, providing new clues for understanding the shell formation and defense mechanism in mollusk.

Multiple sequence alignment and phylogenetic tree. The deduced amino acid sequence of PmC-1qDC-1 was homologous to the C1q family. NCBI BLAST analysis revealed that the deduced amino acid sequence of PmC1qDC-1 shared similarity with C1qDCs from other organisms, such as from Crassostrea virginica (28.57% identity with C1qDC), from C. gigas (26.20% identity with C1qDC), from Ostrea edulis (32.20% identity with C1qDC), and from Mizuhopecten yessoensis (29.7% identity with C1qDC). The gC1q region of these C1qDCs was aligned. The 10-stranded β-sandwich in the gC1q domain of PmC1qDC-1 and the eight conserved residues in the human gC1q domain had been labeled (Fig. 1). The result of multiple alignments indicates that some of the eight amino acid residues conserved in the human C1q domain have been mutated in PmC1qDC-1. Phylogenetic tree was constructed to analyze the relationship between PmC1qDC-1 and C1qDC in other species (Fig. 2). The result indicates that PmC1qDC-1 clustered together with other C1qDC proteins from invertebrates, while the C1qDC in vertebrates were clustered together.
Expression pattern of PmC1qDC-1 in different tissues and in larvae development. We detected the expression level of PmC1qDC-1 in the mantle central, mantle pallial, mantle edge, foot, hemocytes, adductor muscle, and gill. PmC1qDC-1 was constitutively expressed in all examined tissues, with the highest expression in the mantle edge (P < 0.05) and the lowest expression in hemocytes (Fig. 3a). We also detected the expression level of PmC1qDC-1 at various larvae development stages (Fig. 3b). The result showed that PmC1qDC-1 expression was significantly high in the blastula, and gastrula and among juveniles (P < 0.05). PmC1qDC-1 increased calcium carbonate precipitation and contributed to shell formation. In vitro, the effect of PmC1qDC-1 on the precipitation rate was determined via a calcium carbonate precipitation experiment. Compared with the control, PmC1qDC-1 increased the precipitation rate, and this effect was concentration dependent (Fig. 6).
In vivo, we knocked down PmC1qDC-1 expression by injecting ds_PmC1qDC-1. The mRNA expression levels of PmC1qDC-1 in the mantle pallial significantly decreased after treatment (Fig. 7a). We observed the inner   www.nature.com/scientificreports/ surface microstructure of the nacre layer under a scanning electron microscope. Disordered crystal growth was observed in the experimental group (Fig. 7b), while the crystal growth on the control group was normal (Fig. 7c).
Expression pattern of PmC1qDC-1 after shell notching. We investigated the expression patterns of PmC1qDC-1 in shell damage repair. PmC1qDC-1 expression in the mantle pallial was gradually up-regulated after shell injury and significantly up-regulated at 6 h and 12 h compared with that at 0 h (Fig. 8a). PmC1qDC-1 gene was significantly overexpressed at the mantle edge 12 h and 24 h after shell notching (Fig. 8b).
PmC1qDC-1 response to PAMPs stimulation. The response of the PmC1qDC-1 gene to LPS and poly I:C stimulation was detected by quantitative real-time PCR (qRT-PCR) (Fig. 9). β-actin was used as the referent gene to determine relative expression levels. The expression level of PmC1qDC-1 in hemocytes was significantly up-regulated at 6 h after injecting LPS (Fig. 9a). After challenge with poly I:C for 6 h, PmC1qDC-1 expression in hemocytes was significantly up-regulated (Fig. 9b). We also detected the expression pattern of PmC1qDC-1 in mantle edge after LPS and poly I:C stimulation. The expression level of PmC1qDC-1 in mantle edge was significantly up-regulated at 48 h after LPS stimulation (Fig. 9c). After challenge with poly I:C for 12 h, 24 h, and

Discussion
C1q is the key component of complement system containing the typical C-terminal globular C1q domain. C1qDC protein can recognize many ligands via the gC1q domain with the common binding sites 21,38 . In the present study, we identified a novel gC1q gene from P. f. martensii and designated it as PmC1qDC-1. PmC1qDC-1 contains a typical C-terminal globular C1q domain with 10 β-stand sheets and is considered as a member of the C1qDC protein family. All invertebrate C1qDCs, including those from C. gigas, M. yessoensis, C. virginica, O. edulis, and A. irradians, were grouped into a single branch. Thus, PmC1qDC-1 and other invertebrate C1qDCs may have evolved from the same ancestral gene, and subsequent selection pressure then led to the diversification of those genes within the invertebrate lineage. Multiple sequence alignment also supports this hypothesis. PmC1qDC-1 has low sequence similarity to C1qDC in other mollusks (26.2-32.2%), and some of the eight conserved amino acids 39 in the human C1q domain had been mutated in PmC1qDC-1; thus these amino acids may have mutated to accommodate versatile attributes of function. We detected the expression pattern of PmC1qDC-1 in seven tissues. Constitutive expression in all of these tissues suggested the possibility of PmC1qDC-1 performing various functions in different tissues. Although the C1q domain is an immune-related domain, it also functions as SMPs existing in shell 35 . The high expression of PmC1qDC-1 in the mantle suggests that it may be involved in shell formation. The highest expression of PmC1qDC-1 in juveniles indicats that PmC1qDC-1 participats in the formation of dissoconch shell, which is the normal adult shell structure 40 . The mantle is the most important tissue involved in biomineralization and mantle pallial is responsible for the formation of nacreous layer. The strong hybridization signals located in the outer epithelial cells of mantle pallial indicates that PmC1qDC-1 participated in nacreous formation. Moreover, in vitro, PmC1qDC-1 can increase the calcium carbonate precipitation rate. Especially, disordered crystal growth was observed in the nacre upon knockdown of PmC1qDC-1 expression at the mantle pallial. This result  www.nature.com/scientificreports/ indicats that PmC1qDC-1 is directly involved in nacre formation. The C1q globular domain consists of spherical heterotrimeric with Ca 2+ ion bound at the top 2 . Therefore, PmC1qDC-1 may participate in nacre formation by accumulating Ca 2+ ions. However, the mechanism through which PmC1qDC-1 participates in nacre formation requires further investigation. gC1q protein is common in bivalves. For example, 167 of the 168 C1qDC gene models in Mediterranean mussel are gC1q proteins 28 . Numerous invertebrate gC1q proteins are involved in the immune process. For example, MgC1q acts as a pattern-recognition molecule that can recognize pathogens during innate immune responses in M. galloprovincialis 19 . AiC1qDC-1 of scallops can agglutinate fungi and has mannose and PGN binding sites at its gC1q domain 41 . PmC1qDC-1 is extremely highly expressed at the mantle edge, indicating that it may be involved in resisting bacterial and pathogenic invasion. Besides the key roles of the mantle edge in shell formation, it also plays important roles in direct immune defense during exposure to seawater condition and even in shell damage and parasitic invasion in bivalves. For example, the mantle edge of C. gigas can secrete defensin in response to pathogen colonization 42 . The expression level of PmC1qDC-1 in mantle edge was significantly up-regulated at 48 h after LPS stimulation. After challenge with poly I:C for 12 h, 24 h, and 48 h, PmC1qDC-1 was significantlyhigh expressed in mantle edge compared with control group. The expression fluctuation of PmC1qDC-1 after LPS and poly I:C stimulation indicates that it is involved in immune defense. PmC1qDC-1 was significantly up-regulated in the mantle pallial at 6 h and 12 h post of shell notching while it was significantly up-regulated at mantle edge at 12 h and 24 h after shell damage. At the shell regeneration stage, shell damage will accelerate SMPs secretion 43 , and the organic membrane where the initiation of crystal deposition occurrs has formed near the nick at 6 h after shell notching 43,44 . Moreover, the mantle tissue will retract and increase the zone of tissue exposed to seawater due to the "V" nick, which will easily result in a second injury due to bacterial and pathogen infection 43 . Thus, PmC1qDC-1 may participate in shell regeneration and in the immune process of shell damage.
C1qDC proteins, which are pattern recognition molecules, rely on their gC1q domain to recognize a variety of self and non-self ligands, including a vast range of PAMPs of bacteria, yeast, viruses, and parasites 20,45 . The www.nature.com/scientificreports/ shell notching experiment indicates that PmC1qDC-1 participates in the immune process. The hemocytes, the most critical immune organ, play a central role in the recognition of exogenous agents and in the defense against bacterial invasion in mollusks 46 . However, the expression level in hemocytes was relatively low. We speculate that, similar to HcC1qDC6 20 , PmC1qDC-1 expression in hemocytes may be induced under stress. As such, the challenge experiment of LPS and poly I:C was performed to further understand the immune-related functions of PmC1qDC-1 in pearl oysters. PmC1qDC-1 was significantly up-regulated in the hemocytes at 6 h after injecting LPS, and the same expression pattern was present in the poly I:C treatment group. This finding indicates that PmC1qDC-1 is involved in the immune response against pathogen invasion.
In this work, we identified a novel gC1q gene named PmC1qDC-1. PmC1qDC-1 participates in nacre formation by increasing the calcium carbonate precipitation. PmC1qDC-1 may be involved in shell regeneration and immune response after shell injury. PmC1qDC-1 gene can quickly respond to LPS and poly I:C challenge. These findings revealed that it participates in shell formation and the innate immune response in P. f. martensii.

Materials and methods
Experimental animals. Adult pearl oysters (2 years old) were obtained from Houhong Xuwen, Zhangjiang,Guangdong Province, China. The animals were cultured at 25-28 ℃ indoor breeding tanks with seawater for 2 days before the experiment.
The sample of larvae at different development stages was the same used in previous genomic research 32 .
RNA isolation and cDNA synthesis. Various tissues containing the mantle central, mantle pallial, mantle edge, foot, hemocytes, adductor muscle, and gill were separated from the pearl oysters. Total RNA was isolated through TRIzol method following the protocol we submitted to protocols.io before (https ://dx.doi.org/10.17504 /proto cols.io.9qgh5 tw). Absorbance at OD260/OD280 was measured using a NanoDrop ND 1000 spectrophotometer (ThermoFisher Scientific Inc, Waltham, MA, USA) to confirm the RNA quantity. RNA integrity was determined by fractionation on 1.0% agarose gel. cDNA was synthesized using an M-MLV reverse transcriptase (Promega, USA). In brief, 500 ng of RNA and 1 μL of random primers were mixed, and RNase free water was added until 6 μL was reached, and the mix was incubated at 70 ℃ for 10 min. Gene cloning of PmC1qDC-1 and sequence analysis. In the present study, the full-length cDNA of PmC1qDC-1 was obtained using RACE. The procedure for PCR is listed as follows: 95 ℃ for 5 min, 38 cycles at 98 ℃ for 15 s, 60 ℃ for 30 s, 72 ℃ for 2 min, and 72 ℃ for 10 min. The PCR product was gel purified and cloned into a PMD 18-T simple vector (TaKaRa, Dalian, China). The base sequence was obtained via Sanger sequencing. All the primers used in the study were listed in the Table 1.
We used the ORF Finder tool (https ://www.ncbi.nlm.nih.gov/orffi nder/) and SignalP-5.0 Server (http://www. cbs.dtu.dk/servi ces/Signa lP/) to obtain the ORF region and signal peptide of PmC1qDC-1. Domain information of PmC1qDC-1 was obtained through the Simple Modular Architecture Research Tool (http://smart .embl-heide lberg .de/smart /show_motif s.pl). Clustal Omega website tool (https ://www.ebi.ac.uk/Tools /msa/clust alo/) was used to align the protein sequences. Evolutionary relationship of PmC1qDC-1 and other orthologs was build up by MEGA 7. qRT-PCR and statistical analysis. We used the qRT-PCR to test PmC1qDC-1 expression levels. The mix reagent was from DyNAmo Flash SYBR Green qPCR kit (ThermoFisher Scientific). The qRT-PCR experiment of PmC1qDC-1 was carried out on Applied Biosystems 7500/7500 Fast Real-Time System (Applied Biosystems, Foster City, CA, USA). The PCR program was as follows: 95 °C for 5 min, 40 cycles at 95 °C for 30 s, 60 °C for 15 s, 72 °C for 15 s. The relative expression levels of reference genes (β-actin and GAPDH) and PmC1qDC-1 were calculated through the 2 −ΔCT method.
Significance was analyzed using SPSS 22.0 (IBM, Chicago, IL, USA). The expression levels of PmC1qDC-1 at the tissues, different development stages, and different time points of shell notching were analyzed using one-way ANOVA. Differences in PmC1qDC-1 expression between the two groups were evaluated using the T-test. The significant level for these analyses was set at P < 0.05. Fluorscent in situ hybridization experiment. RNA probes of PmC1qDC-1 were synthesized in vitro by using T7 RNA polymerase and digoxigenin (DIG) RNA Labeling Mix. Integrity of PmC1qDC-1 RNA probes was confirmed by using 1% agarose gel electrophoresis. Concentration and purity of PmC1qDC-1 RNA probes were detected by using a nucleic acid quantifier. The mantle of P. f. martensii was separated and fixed 2 h in 4% paraformaldehyde containing 0.1% diethyl dicarbonate (Sangon Biotech). Then the mantle tissues were washed with phosphate buffer saline (PBS) for three times and embedded in the medium called optimal cutting temperature (O. C. T.). Finally, the embedded samples were made into slices with 10 μm thickness via the instrument of LEICA CM3050 S. Fluorescent in situ hybridization was constructed through the instructions of Enhanced Sensitive ISH Detection Kit IV (CY3) (BOSTER). The detailed protocol was referred to the file we submitted to protocols.io before (https ://doi.org/10.17504 /proto cols.io.9qhh5 t6). The fluorescence signals were observed under microscope (Nikon ECLIPSE Ni, DS-Ri2). www.nature.com/scientificreports/

PmC1qDC-1 function interference experiment. RNAi was used to knockdown the expression level of
PmC1qDC-1 following the protocol of Hao et al. 47 . In brief, double-stranded RNA (dsRNA) of PmC1qDC-1 (ds_ PmC1qDC-1) and Red Fluorescent Protein gene (ds_RFP) were synthesized and purified by using a T7 High-Efficiency Transcription kit (TransGen Biotech, JT101) and EasyPure RNA Purification kit (TransGen Biotech, ER701) correspondingly. 100 µL ds_PmC1qDC-1 were injected into the adductor muscle of P. f. martensii, while 100 µL ds_RFP were injected at the control group. Mantle pallial was collected, put in liquid nitrogen at the 4 days post second injection and then frozen samples were transferred in − 80 °C freezer. The shells of all groups were cut into a small piece (0.5 cm × 1.5 cm) containing the transition region of the nacre and prismatic layer. The shell samples were clean with ultrapure water and air dried. The inner surface of nacre layer near the transition was observed under a scanning electron microscope (JSM-6300 LV) to obtain the image of ultrastructure.
Prepared recombinant protein of PmC1qDC-1. Recombinant PmC1qDC-1 protein was prepared in cooperation with Abmart. PmC1qDC-1 gene sequence without signal peptide was subcloned to the vector modified from (pET30a) by Abmart, and the recombinant plasmid was transformed using Rosetta Competent Cell for expression. Transformed cells were cultured in Luria-Bertani medium (with 50 mg/mL kana + ) at 37 ℃ 200 rpm, induced with 1 mM IPTG when OD600 reached 0.5-0.8, and cultured at the same condition for another 4 h. The cell pellet was collected after centrifugation at 6000g for 5 min at 4 ℃ and washed three times with PBS. The harvested cells were resuspended with PBS and subjected to ultrasonication. The protein was purified using Amylose Resin (NEB). The wash fractions with 10 mM maltose were collected, boiled and loaded onto an SDS-PAGE gel. The purified protein was desalted by dialysis and freeze dried for storage. The protein concentration was measured by using BCA assay kit (Sangon Biotech).
Calcium carbonate precipitation assay. The effect of PmC1qDC-1 on the rate of calcium carbonate precipitation was tested following the method of Dong et al. 48 . The control was 20 μg/mL MBP. The 10 μL of sample solution and 100 μL of calcium chloride (100 Mm, pH 8.5) were added to 96-well plates and mixed completely. Then, 100 μL of 100 mM sodium bicarbonate was added to the mixed solution quickly. The formation of calcium carbonate precipitate was tested by recording the absorbance at 570 nm every 1 min by using a multimode plate reader (EnSpire, PerkinElmer).
Shell notching and PAMP stimulation. The notching assays were performed on 35 normal pearl oysters.
Thirty normal pearl oysters were selected and a "V" shaped notch was cut on the shell until the nacreous layer was reached. The mantle edge and mantle pallial of every five pearl oysters were collected at 2 h, 4 h, 6 h, 12 h, and 24 h after damage, and the mantle edge of five pearl oysters (no notching) was harvested at 0 h. The pearl oysters were randomly divided evenly into three groups (36 individuals for each group), and 100 μL (10 μg/mL) each of LPS, poly I:C, and PBS were injected into the adductor muscle. The pearl oysters were cultured Table 1. Names and sequences of primers used in this study. Note: The sequences underlined are the sequence of the T7 promoter.