Identification of a coproporphyrinogen-III oxidase gene and its correlation with nacre color in Hyriopsis cumingii

Pearl color is an important factor influencing pearl value, and is affected by the nacre color of the shell in Hyriopsis cumingii. Coproporphyrinogen-III oxidase (CPOX) is a key enzyme in porphyrin synthesis, and porphyrins are involved in color formation in different organisms, including in the nacre color of mussels. In this study, a CPOX gene (HcCPOX) was identified from H. cumingii, and its amino acid sequence was found to contain a coprogen-oxidase domain. HcCPOX mRNA was expressed widely in the tissues of white and purple mussels, and the highest expression was found in the gill, followed by the fringe mantle. The expression of HcCPOX in all tissues of purple mussels (except in the middle mantle) was higher than that of white mussels. Strong hybridization signals for HcCPOX were observed in the dorsal epithelial cells of the outer fold of the mantle. The activity of CPOX in the gill, fringe mantle, and foot of purple mussels was significantly higher than that in white mussels. Moreover, the expression of HcCPOX and CPOX activity were decreased in RNA interference experiments. The findings indicate that HcCPOX might contributes to nacre color formation in H. cumingii by being involved in porphyrin synthesis.


Introduction
Hyriopsis cumingii is a freshwater mussel that is widely distributed in China and produces high-quality freshwater pearls [1]. Pearl size, shape, color, and luster influence the value of pearls, with color being the primary factor [2]. Some studies have indicated that the pearl color of host mussels is affected by the nacre color of the donor mussels [3] and, therefore, nacre color has a substantial impact on the value of pearls. The nacre color of a shell is not only influenced by water environmental factors, but also by genetic factors [4]. Studies have reported that melanin and carotenoids are involved in nacre color formation, and related genes (e.g., HcTyr, HcMitf, hcApo) have also been found to play a role in nacre color in H. cumingii [5][6][7][8][9]. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 The pearl oyster shells are usually composed of nacre, prismatic layer and periostracum. The nacre is the inner layer of a shell, and its formation occurs as the shell is made, which involves a complex biomineralization process. Briefly, the gill, foot, and mantle epidermis absorb Ca 2+ and HCO 3 − from the water, which is transported into the mantle through hemolymph or other mechanisms. These compounds are then secreted by epithelial cells between the hemolymph and mantle cavity, and deposited on the shell. However, some specific mechanisms of this process occur remain unclear [10][11][12].
Porphyrins are common natural pigments in multiple organisms. Different porphyrins show different colors, such as uroporphyrin is red and protoporphyrin is purple [13]. Moreover, they usually exist in organisms in the form of metal chelates, and the different metal ions they bind will result in different colors. For example, manganese porphyrin and magnesium porphyrin are green, while iron porphyrin is red [14]. Based on the organic chromophore theory and the chemical structure characteristics of porphyrins, Hao et al. [15] revealed that there are multiple conjugated double bonds in porphyrins, and its structure belongs to a cyclic conjugated system. The π-π conjugated structure of porphyrin will affect its spectral absorption band. When porphyrin is combined with different metal ions, the original conjugated structure of the porphyrin is distorted or deformed due to differences in the size of the metal ions, which cause the absorption band to shift, resulting in color differences.
Porphyrins are important for the color of organisms. In fact, a large number of studies have shown that porphyrins are the key substances that determine the color of bird eggshells, and that different metal porphyrins can lead to different shell colors [16][17][18][19]. Mussel shells are similar to eggshells in that they are both comprised of calcium carbonate. Williams et al. found that uroporphyrin I and uroporphyrin III are involved in shell color formation in marine snails [20]. Bonnard et al. indicated that the shell color is also related to uroporphyrin in Crassostrea gigas oyster [21]. Moreover, studies found that the color of freshwater pearls produced by H. cumingii related to porphyrins [22]. It is widely known that the nacre formation is similar to the pearl formation, so the effect of porphyrins on the nacre color formation is worth studying.
The synthesis of porphyrin in organisms is a complex process in which coproporphyrinogen-III oxidase (CPOX) is a key enzyme. CPOX catalyzes coproporphyrin III to synthesize protoporphyrinogen IX, and then further synthesis protoporphyrin IX under the action of protoporphyrinogen oxidase, and finally combines with metal atoms to synthesize metalloporphyrin. However, the CPOX effect on shell color has only been reported in Meretrix meretrix [23]. Therefore, the effect of CPOX on nacre color in H. cumingii requires further investigation.
In this paper, the full-length cDNA of the H. cumingii coproporphyrinogen III oxidase gene (HcCPOX) was obtained. The expression of HcCPOX and the activity of CPOX in different tissues were also determined. The distribution of HcCPOX in the mantle was detected using in situ hybridization. RNA interference (RNAi) was performed in purple mussels and the changes in HcCPOX expression and CPOX activity were evaluated in different tissues. The findings indicated that HcCPOX may affect nacre color by affecting the synthesis of porphyrins.

RNA extraction and full-length cDNA cloning
First, mantle tissue was collected and immediately frozen in liquid nitrogen and stored at −80˚C until use. Then, total RNA was extracted using TRIZOL reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. The quality of RNA was examined by a spectrophotometer (NanoDrop 2000c, Thermo, USA) and electrophoresis on 1% agarose gels. The SMART RACE cDNA amplification kit (Clontech, Dalian, China) was used to obtain mantle cDNA, and then 5 0 -and 3 0 -RACE was performed to clone the full-length cDNA of HcCPOX. The specific primers (Table 1) were designed based on the HcCPOX expressed sequence tag obtained from the H. cumingii mantle transcriptome library [24]. The PCR parameters were as follows: 5 min at 94˚C; 30 cycles at 94˚C for 30 s, 58˚C for 30 s, and 72˚C for 1 min, and a 10 min final extension at 72˚C. The PCR products were sequenced using the Sanger method (Sangon, Shanghai, China).

Bioinformatics analysis
The open reading frame (ORF) of HcCPOX was predicted using the NCBI ORF Finder tool (http://www.ncbi.nlm.nih.gov/projects/gorf/) [25]. The BLAST program (http://www.ncbi. nlm.gov/blast) was used to perform homology analyses of the nucleotide and protein sequences [25]. The Simple Modular Architecture Research Tool (SMART, (http://smart. embl-heidelberg.de/)) was used to analyze the structural domains contained in the amino acid sequence. The online Protparam (http://www.expasy.org/tools/protparam) tool was used to obtain amino acid sequence composition, molecular weight, isoelectric point, and other physical parameter information [26]. Clustalx software was used to perform multiple sequence comparisons [27]. MEGA 5.2 was used to construct a phylogenetic tree using the maximum likelihood method, and the aligned sequences were bootstrapped till 1000 replicates were obtained [28].

Tissue-specific expression analysis of HcCPOX
Quantitative RT-PCR was used to detect the tissue-specific expression of HcCPOX in different tissues. Different tissues (fringe mantle, middle mantle, adductor muscle, gill, foot, hepatopancreas, and hemolymph) were obtained from 12 healthy H. cumingii (six white mussels and six purple mussels), and total RNA extraction was performed as described above. The PrimeScriptTM RT reagent kit with gDNA Eraser (Takara, Dalian, China) was used to synthesize cDNA for qPCR analysis. Specific primers (Table 1) were designed based on the sequence of HcCPOX, and EF-1α (Table 1) was used as the internal reference [29]. qPCR was carried out in a quantitative thermal cycler (Bio-Rad CFX-96; Bio-Rad, USA), with the following parameters: 3 min at 95˚C; then 40 cycles of 95˚C for 5 s, 57.5˚C for 30 s and 72˚C for 30 s. Finally, tissue-specific expression of HcCPOX was analyzed using the comparative CT method (2 −ΔΔCt ) [30].

In situ hybridization of HcCPOX in the mantle
In situ hybridization was used to detect the precise locations of expression of HcCPOX in the mantle of purple mussels. First, the RNA probes were synthesized using the T7 in vitro Transcription Kit (Promega, USA), according to the manufacturer's instructions. The mantle samples (including outer fold, middle fold, inner fold of the mantle) were then collected and fixed in 4% paraformaldehyde for 6 h at 4˚C, and then placed in 20% sucrose overnight at 4˚C [5,7]. A freezing microtome (Leica CM 1950; Wetzlar, Germany) was used to slice mantle samples into 13-μm thicknesses and stored at −80˚C. Finally, in situ hybridization was performed according to the manufacturer's instructions (Enhanced Sensitive ISH Detection Kit, Boster, Switzerland).

Analysis of the coproporphyrinogen-III oxidase activity
Eight mussels (four white mussels and four purple mussels) were used to detect the coproporphyrinogen-III oxidase activity in H. cumingii. Tissues (gill, fringe mantle, middle mantle, Table 1. Primers used in this study.

RNAi assays
Specific primers (Table 1) were designed for double-stranded RNA (dsRNA) synthesis based on the sequence of HcCPOX. T7 RNA polymerase (TAATACGACTCACTATAGGG) was added to the 5 0 end of the upstream and downstream primers. The PCR product was used for the synthesis of dsRNA using the T7 in vitro Transcription Kit (Promega, USA), according to the manufacturer's instructions. Green fluorescent protein (GFP) dsRNA was synthesized as described above and used as the control group. Twenty-nine healthy purple mussels were divided equally into three groups (experimental group, negative control group, and blank group), and injected with dsHcCPOX (80 μg/100 μL), dsGFP (80 μg/100 μL) and 1 × PBS (100 μL) via the adductor muscle. Tissues (gill, fringe mantle, middle mantle, and hemolymph) from each mussel were collected after culturing in freshwater for one week. Finally, the expression of HcCPOX and the activity of CPOX in each tissue were detected as described above.

Statistical analysis
Data are shown as the mean ± SD, and a one-way ANOVA was performed using SPSS 17.0 (SPSS, Chicago, IL, USA) to determine whether there were any significant differences. A pvalue < 0.05 was considered statistically significant.

Cloning and sequence analysis of HcCPOX
The full-length cDNA of HcCPOX (Accession No. The ORF also contained a coprogen-oxidase domain (PF01218). A phylogenetic tree was constructed and showed that the amino acid sequence of HcCPOX belonged to the same clade as CPOX from Meretrix meretrix, Mizuhopecten yessoensis, and Crassostrea gigas, with a support of 62% (Fig 2).

qPCR analysis and in situ hybridization
The relative expression level of HcCPOX in the tissues of white and purple mussels was detected by qPCR. As shown in Fig 3, HcCPOX was expressed in all tissues of white and purple mussels. In white mussels, the tissue with the highest expression was the gill, followed by the fringe mantle. The level of expression in gill and fringe mantle were significantly higher than that in other tissues (p < 0.05). No significant differences in the expression levels of HcCPOX in hemolymph, foot, and middle mantle were observed (p > 0.05). The expression of HcCPOX in the adductor muscle was significantly lower than that in the hepatopancreas (p < 0.05). In purple mussels, the highest level of HcCPOX expression was also in the gill, followed by the fringe mantle. The level of HcCPOX expression in the gill and fringe mantle was significantly higher than that in other tissues (p < 0.05). No significant differences were observed between HcCPOX expression in the foot and the hemolymph (p > 0.05). The levels of expression in the remaining tissues followed the order from high to low of middle mantle > hepatopancreas > adductor muscle (p < 0.05). Finally, comparing the expression levels of HcCPOX in different tissues of the two types of mussels, it was found that the expression of HcCPOX was higher in all tissues from purple mussels (with the exception of the middle mantle) than white mussels (p < 0.05).
In situ hybridization revealed the specific locations of HcCPOX expression in the mantle. As shown in Fig 4, strong hybridization signals appeared in the dorsal epithelial cells of the outer fold of the mantle (arrowed), while no positive signals were detected in the negative control.

Coproporphyrinogen-III oxidase activity
As shown in Fig 5, CPOX activity in mussel tissues was consistent with the HcCPOX expression levels. CPOX activity was the highest is the gill, followed by the fringe mantle. Additionally, the activity of CPOX in the gill, fringe mantle, and foot of purple mussels was significantly higher than that of white mussels (p < 0.05), however, no significant differences were observed in the hemolymph (p > 0.05).

The impact of RNAi
HcCPOX expression levels and CPOX activities were detected after RNAi treatment. As shown in Fig 6A, the expression of HcCPOX in the gill, fringe mantle, foot, and hemolymph of the mussels injected with dsHcCPOX was significantly decreased compared to mussels injected with dsGFP and 1 × PBS (p < 0.05). However, the expression of HcCPOX in the tissues of mussels injected with dsGFP was not significantly different than that of mussels injected with 1× PBS (p > 0.05). Furthermore, CPOX activities in those tissues were also similar to the HcCPOX expression levels. After injection of dsHcCPOX, CPOX activities in all tissues were inhibited compared to that of mussels injected with dsGFP and 1 × PBS (p < 0.05). Additionally, no significant difference in CPOX activity was observed between mussels injected with dsGFP and 1 × PBS (p > 0.05) (Fig 6B).

Discussion
CPOX is an indispensable enzyme that is required for the synthesis of heme in different organisms. Some studies found that many different pigments (e.g., uroporphyrin, cobalamin, phorcabilin, biliverdin, and bilirubin) are produced in the synthesis and decomposition pathways of heme, which affect the shell color in Pinctada margaritifera [31]. If the heme synthesis pathway is affected, it may also lead to the formation of different shell colors in marine snails [20].
In this study, the CPOX gene (HcCPOX) from H. cumingii was successfully cloned. The amino acid sequence contained a coprogen-oxidase domain, which belongs to the conserved sequence of the coproporphyrinogen oxidase superfamily. Phylogenetic analysis also revealed that the HcCPOX amino acid sequence was in the same clade as CPOX sequences from shellfish (M. meretrix, M. yessoensis, and C. gigas). Therefore, the data suggested that HcCPOX belongs to the CPOX superfamily.
The CPOX gene encodes porphyrinogen III oxidase in organisms, which catalyzes the formation of porphyrin substances via an oxidation reaction. Studies have found that porphyrin plays an important role in the synthesis of haem in the blood of mammals and mollusks [32,33]. Williams et al. identified some genes of the haem pathway in marine snails, which affect shell color by participating in the synthesis of porphyrin [20]. Moreover, similar results were also found in P. margaritifera [31].
In a study in which the CPOX gene was knocked out in zebrafish, it was found that the hemolymph changed from red to nearly colorless, thus, indicating that CPOX gene is involved in the formation of hemolymph color (heme) in zebrafish [34]. H. cumingii is a bivalve mollusk, and its hemolymph is typically nearly colorless, which occasionally appears light blue after being fully oxidized. This color change is mainly caused by the small amount of hemocyanin contained within its hemolymph [35].
Combined with the qPCR results, we found that HcCPOX was expressed in hemolymph, but only a low levels. This may be due to its colorless hemolymph. In addition, there was no significant difference in CPOX activity in the hemolymph of white and purple mussels, and was much lower than that in the gill and fringe mantle.
The expression of HcCPOX in white and purple mussels was highest in the gills. Similar results were also obtained for CPOX activities. Cooper et al. found that the gills of mussels were the main locations for the uptake of metal ions from water environments [36]. The high levels of HcCPOX expression in the gill may be due to high level of metal ions contained therein, and gill may provide a suitable environment for the synthesis of metalloporphyrins, and then porphyrins transferred to the mantle. Comfort et al. indicated that metalloporphyrin is a key substance in the formation of color in shells [37]. Shi et al. found that a variety of metal ions exist in the pearls of H. cumingii by Raman spectroscopy, and speculated that they might affect the shell color in the form of metalloporphyrin [22]. The mantle is the tissue responsible for the nacre formation of shells and pearls. The qPCR results found that the levels of HcCPOX in the fringe mantle of white and purple mussels was high, and only lower than that of the gill. Moreover, positive signals were detected in the dorsal epithelial cells of the outer fold of the mantle using in situ hybridization. Studies indicated that outer fold of the mantle was directly involved in the nacre formation of shells [38]. Thus, these results suggested that HcCPOX may be involved in the formation of nacre in H. cumingii.
Zhan et al. found that the CPOX gene (MmCPOX) in M. meretrix was highly expressed in the mantle, and expression was positively correlated with shell color [23]. The uroporphyrinogen III synthetase (an important enzyme involved in the structural formation of porphyrin) gene was found to be involved in nacre color formation in H. cumingii [39]. In this study, the expression of HcCPOX in tissues of white and purple mussels were compared. The result revealed that the expression levels of HcCPOX were significantly higher in all tissues of purple mussels (except the middle mantle) than in white mussels. We also found that the CPOX activity in the tissues (fringe mantle, gill, and foot) of purple mussels were also significantly higher than that in white mussels.
The RNAi experiments revealed that the activity CPOX was decreased following the inhibition of HcCPOX expression in all tissues. Therefore, we hypothesize that HcCPOX might affect nacre color formation by participating in the synthesis of CPOX. In conclusion, the results of this study indicate that HcCPOX contributes to nacre color formation in H. cumingii. Such findings can help understand the mechanisms underlying the formation of pearl color, and assist in the cultivation of high quality pearls.