Introduction

Superoxide dismutases (SODs) are a group of antioxidant enzymes that serve as the first line of defense against oxidative stress by catalyzing the conversion of the superoxide anion (O2•⁻), a primary reactive oxygen species (ROS), into molecular oxygen or hydrogen peroxide1. SODs utilize a transition metal ion in their active sites as the catalyst, and in eukaryotes, this ion is either copper (Cu²⁺) or manganese (Mn²⁺)2. Most copper-dependent SODs also contain a zinc ion to stabilize the structural conformation of the active sites and are thus named copper-zinc superoxide dismutases (Cu/Zn-SODs). This type of SOD is highly conserved in both structure and function across species3,4. In addition to mitigating oxidative stress, SODs also play a role in the regulation of signaling in cells via the modulation of redox-sensitive proteins such as protein phosphatases (PTPs), nonreceptor protein tyrosine kinases (PTKs), protein kinases C (PKC), mitogen-activated protein kinases (MAPKs) and transcriptional factors (TFs) in cellular signaling pathways by H2O2 at low and moderate concentrations5,6.

Cu/Zn-SODs are present in animals in two forms: intracellular SODs (icCu/Zn-SODs) and extracellular SODs (EC-SODs)7. While evolutionarily related and structurally similar, intracellular and extracellular SODs locate in different cellular compartments and play distinct roles in host antioxidant defense8. Intracellular Cu/Zn-SODs are present in the cytoplasm and mitochondria acting to protect host cells from oxidative damage and in the nucleus functioning to regulate gene expression9. In contrast, EC-SODs exist on cell surfaces and within the extracellular matrix playing a role of scavengers of extracellular superoxide radicals10.

It was until the discovery of MtSodC, a copper-dependent SOD from the bacterium Mycobacterium tuberculosis, that all copper-dependent SODs had been believed to be bimetallic, containing a copper for catalysis and a zinc for conformation stabilization in the active sites11,12. MtSodC holds a copper ion only in the active site13. The protein retains the general conformation of regular Cu/Zn-SODs including the electrostatic loop (ESL), but lacks zinc-binding residues His115 and His123 and the replacement of His123 by an alanine leads to the formation of a stable and zinc-independent dimeric interface14. Later, another copper-dependent SOD that contains only a copper ion in the active site (SOD5) is identified in the pathogenic fungus Candida albicans15. Unlike MtSodC, the C. albicans SOD5 does not have a whole ESL in the active site, but the replacements of His110 and His113 by glutamic acid and aspartic acid respectively result in the loss of zinc-binding site and the formation of an open-accessed copper site15. These studies on copper-dependent SODs in M. tuberculosis and C. albicans have led to the proposal of a new type of SOD, copper-only superoxide dismutases (Cu-SODs)13,16. Despite lacking a zinc ion in the active site, Cu-SODs exhibit comparable catalytic efficiency as regular Cu/Zn-SODs. In addition, they are reported to accumulate first in inactive form and then be activated by acquiring copper ions extracellularly and to play a crucial role in pathogenesis by protecting the pathogenic microorganisms from the damage by host ROSs generated in respiratory burst17,18. In addition to M. tuberculosis and C. albicans, typical Cu-SODs of 20–30 kDa single-domain proteins are identified in some other unicellular microbes including bacteria, fungi, and oomycetes through homologous sequence searches16.

In animals, a class of proteins that consist of tandem repeat Cu-SOD domains with the predicted molecular weight of around 100 kDa is believed to be homologous to the fungal Cu-SODs and named copper-only SOD repeat proteins (CSRPs)14. CSRPs are identified at the sequence level to be present in the unicellular protist Capsaspora owczarzaki and animal species, including the placozoan Trichoplax adhaerens, the Pacific oyster Crassotrea gigas, the mosquito Anopheles gambiae and several bony fish species such as the zebrafish Dania rerio, the red piranha Pygocentrus nattereri, the common carp Cyprinus carpio, and the Atlantic salmon Salmo salar. In addition, CSRPs are predicted to be extracellular with glycosylphosphatidylinositol (GPI) anchors14. These structural similarities between CSRP domains and the C. albicans SOD5 have warranted a speculation that the CSRPs may have the potentials of acting in disproportating superoxide like the fungal Cu-SOD. However, the possible functions of these multidomain proteins in animal antioxidant defense remain to be investigated.

Oysters are bivalve mollusks living in the intertidal and subtidal zones on the fluctuating marine coasts and estuaries. Owing to the sessile lifestyle, they routinely face the environmental harshness and cope with oxidant stress caused by a variety of biotic and abiotic stressors in their habitats19. Cu/Zn-SOD coding genes are identified in multiple oyster species20. The presence and antioxidant functionality of an intracellular type of Cu/Zn-SOD has been confirmed in Pacific oysters C. gigas with recombinantly expressed21. In the meantime, a homolog of extracellular SODs has been identified as the most abundant protein in the oyster hemolymph based on its general sequence similarities with known EC-SODs and was initially proposed to be an oyster EC-SOD22,23. However, activity assays of proteins purified from oyster hemolymph plasma or produced by recombinant expression show that the EC-SOD homolog in oysters does not have the SOD activity. Amino acid sequence comparisons and protein structural modeling analysis also reveal the absence of the binding sites for the copper and zinc ions in the region corresponding to the active site of regular EC-SODs, justifying the loss of the enzyme activity during molecular evolution21,24. On the other hand, oyster genomes are found to contain CSRP coding genes; in addition to the gene identified by Robinett et al. in 2017, we have identified more homologous genes in oyster genomes recently20.

The intriguing concurrency of lacking regular EC-SODs and having multiple CSRP coding genes in the genome had thus led us to speculate the possibilities of CSRPs functioning as special types of extracellular SODs in oysters. To explore the speculation, we set the following research objectives: (1) analyze the phylogenetic relationships and gene structures of CSRPs across species, (2) examine the expression patterns of CSRP genes in two oyster species, the Pacific oyster Crassostrea gigas and the the Kumamoto oyster Crassostrea sikamea, under abiotic stress using transcriptomic data, and (3) investigate the potential of CSRPs functioning as SODs in oysters by modeling their structure and testing SOD enzyme activity using recombinant proteins from a representative CSRP of C. sikamea. Through this study, we aim to gain insights into the functions of these largely unexplored proteins and the mechanisms of antioxidant defense in the extracellular matrix of oysters.

Materials and methods

Oyster collections and maintenance

Kumamoto oysters, 45 ± 7.2 mm in shell height, were collected from Ningbo, Zhejiang Province, China, in March 2024. The oysters were acclimated in the laboratory for one week in 20 L tanks containing artificial seawater with the temperature of 22 ± 0.5 °C and salinities chosen based on documented habitat conditions of the studied species25,26which were 30 ± 0.5 ppt for C. gigas or 25 ± 0.5 ppt for C. sikamea. During acclimation, oysters were fed daily with filtered Chlorella autotropica powder, and the rearing water was fully changed daily.

Generation of oyster transcriptomics data

The transcriptomes of different tissues of the acclimated normal oysters and the gill tissues from the oysters under the stresses were prepared, and transcriptomic data were generated for the assessment of CSRP gene expressions. To prepare the tissue transcriptomes of normal oysters, 9 acclimated Kumamoto oysters were sampled individually for the tissues of adductor muscle, digestive glands, gills, mantle, and gonads using ethanol-cleaned scissors and forceps. The sampled tissues were snap-frozen separately in liquid nitrogen and processed for transcriptomic analysis as described below.

To prepare the gill transcriptomes of the stressed oysters, the oysters were experimentally challenged by salinity variations, air exposure, or chronic heating. For the salinity challenge, 200 oysters were randomly divided into 4 tanks containing artificial seawater at the salinity of 0, 5, 30, and 40 ppt, respectively, 50 oysters per tank. The oysters were otherwise maintained as for the acclimation, and 9 oysters per tank each time were sampled for gills on the days of 1, 3, 5, and 7 after the treatment. Nine oysters maintained at 20 ppt were sampled to prepare the transcriptome representing the start point of all 4 salinity treatments. For the air exposure challenge, 150 oysters were divided into 3 groups, 50 oysters each, and placed correspondingly in air at 20 °C, in seawater at 35 °C, and in air at 35 °C. The oysters were otherwise maintained as for acclimation, and 9 oysters per group each time were sampled for gills in the hours of 1, 3, 6, 12, and 24 after the treatments. Nine oysters kept in water at 20 °C were sampled to prepare the transcriptome that served as the starting point of all the 3 treatments. For the chronic heating challenge, 60 oysters were divided into 3 groups, 20 oysters each, and placed in seawater at 20 °C, 28 °C, or 35 °C. The oysters were otherwise maintained as for acclimation, and 9 oysters per group each time were sampled for gills on the days of 5 and 15 after treatment. All gill samplings were done individually as described above.

The 9 individual samples from each sampling point were pooled to make 3 pools, 3 individuals per pool, for transcriptomic sequencing and data analysis as reported previously27. Each pooled transcriptomic sample was sequenced. The final sequence data of each pooled transcriptomic sample were generated from two runs of sequencing (technical replicate) by an Illumina Hiseq platform and analyzed as one biological replicate of a given sampling point.

CSRP identification and sequence analyses

Copper-containing SODs were identified using BLAST searches across species in the database of NCBI with the amino acid sequence of Candida albicans SOD5 (KAG8203737.1) as the query. Sequences with an E-value ≤ 1e-10 were retrieved, and all the sequences that showed an E-value > 1e-10 were completely excluded from further analyses. The retrieved sequences were put through domain identification using the Conserved Domain Search tool28 (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi), and those that contained more than one Sod_Cu domain (PF00080) were selected as CRSPs. The selected CRSP sequences were aligned with the C. albicans SOD5 and the C. gigas CSRP (EKC41617.1) reported by Robinett et al.14 for further confirmation. Multiple sequence alignments were performed in Clustal Omega with default parameters29 (https://www.ebi.ac.uk/Tools/msa/clustalo/).

Sequence-based molecular characterization involved the prediction of multiple molecular characteristics using different bioinformatic tools. Amino acid sequences were deduced using Translate30 (https://www.expasy.org/resources/translate). Signal peptides were predicted with SignalP 5.031 (http://www.cbs.dtu.dk/services/SignalP/). Protein domains were identified using the Conserved Domain Search tool28 (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). Protein molecular weight (MW) and theoretical isoelectric point (pI) were calculated with ProtParam32 (https://web.expasy.org/protparam/). Individual Cu-SOD domains were compared with the representative of the Cu-SOD family (C. albicans SOD5) using multiple sequence alignments as described above.

Protein structures were modeled using SwissModel33 (https://swissmodel.expasy.org/) and visualized with PyMOL34. Phylogenetic analysis of representative sequences from major animal phyla was performed in MEGA1135 using the Neighbor-Joining algorithm with branch bootstrap testing based on 1000 replicates.

CSRP sequence verification

A representative CSRP sequence identified from C. sikamea was verified for genomic sequencing accuracy via PCR and sequencing. Total RNA was extracted from the gill and mantle tissues collected from an acclimated oyster using Trizol™ reagent (OMEGA, USA) following the manufacturer’s protocol. RNA concentrations were measured with a microspectrophotometer (ALLSHENG, Nano-300), and integrity and purity were assessed via 1.0% agarose gel electrophoresis. cDNA synthesis was performed using the PrimeScript RT kit with gDNA Eraser (Takara, Japan). The extracted RNAs and synthesized cDNA were subjected to PCR assays. Primers targeting the gene Cs-CSRP1, annotated as superoxide dismutase in the C. sikamea genome36were designed to amplify the full-length mRNA (Table S1). PCR amplification was performed in a 25 µL reaction containing 1 µL cDNA, 1 µL of each primer (10 µM), 12.5 µL 2× Taq Master Mix, and 9.5 µL ddH₂O. Thermal cycling conditions were: initial denaturation at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 2 min, with a final extension at 72 °C for 10 min. PCR products were purified using a PCR Product Purification Kit (Sangon Biotech, China), cloned into the pEASY-Blunt Zero vector (TransGen Biotech, China), and sequenced at Sangon Biotech (Shanghai, China).

Assessment of expression levels of CSRP genes

The expression of the CSRP genes in 2 oyster species, C. sikamea and C. gigas, was assessed using transcriptomic data generated in the present research for C. sikamea and those retrieved from the NCBI database for C. gigas according to the procedure described in our previous study37. The retrieved C. gigas sequence data were under the project number of PRJNA146329 in the dataset of GCA_902806645.1 and the analyzed tissues included the mantle (SRR334212), the digestive glands (SRR334213), the gills (SRR334215), the adductor muscle (SRR334217), and the gonads (SRR334218). The levels of expression were measured with the expression abundance of the analyzed CSRP genes in the transcriptome with normalization by TPM (Transcripts Per Million) values38.

Recombinant protein production and SOD activity assay

The coding region of the verified C. sikamea CSRP gene (Cs-CSRP1) was cloned into the pSmart-1 plasmid (TransGen Biotech, China) using a One-Step Cloning Kit (Vazyme, China). The constructed plasmids were transformed into DH5α E. coli cells to produce recombinant plasmids, and the insertion of the Cs-CSRP1 coding sequence was confirmed by PCR amplification and sequencing. The produced recombinant plasmids were extracted from the host E. coli cells using the Plasmid DNA Mini Kit (OMEGA, USA) and transformed into E. coli BL21 (DE3) cells for recombinant protein production (Sangon Biotech, China).

The recombinant plasmid transformed E. coli BL21 (DE3) cells were cultured with shaking in LB medium supplemented with 50 µg/mL kanamycin at 37 °C. When OD600 reached 0.6–0.8, Isopropyl β‑D‑thiogalactoside (IPTG) was added to a final concentration of 0.2 mM to induce protein expression for 16 h at 15 °C. The induced bacterial cells were then harvested by centrifugation at 4000 rpm for 10 min and lysed in 50 mM Tris-HCl with 300 mM NaCl and 5 mM imidazole (pH 7.9), followed by centrifugation at 12,000 rpm for 20 min. After centrifugation, recombinant proteins in the supernatants were purified using Ni-NTA columns, and the inclusion bodies were washed and solubilized in urea-containing buffers, followed by purification in affinity chromatography with Ni-NTA Sefinose™ Resin (Sangon Biotech, China). Purified proteins were desalted, concentrated, and analyzed via SDS-PAGE. Protein concentrations were determined using the BCA Protein Assay Kit (Thermo Fisher, USA).

SOD activity of recombinant proteins was measured using an SOD Assay Kit (Nanjing Jiancheng Bioengineering Institute, China) based on the xanthine oxidase method39. One unit (U) of enzyme activity was defined as the amount of enzyme required to inhibit 50% of substrate oxidation by xanthine oxidase.

Statistics

Data was presented as means ± standard error (SE) (n = 3). CSRP gene expression differences in response to stresses were analyzed using one-way analysis of variance (ANOVA) followed by Duncan’s post hoc test to identify statistically significant differences. Statistical analyses were done with the software Origin, and significance levels were set at p < 0.05, with highly significant differences at p < 0.01.

Results

Phylogenetic relationships of CSRP homologous proteins

BLAST search into the NCBI protein databases with the C. albicans SOD5 identified 4389 homologous proteins that included both Cu/Zn-SODs and Cu-SODs. Further sequence comparisons revealed that 2446 sequences that showed E-value ≤ 1e-10 had conserved key amino acid residues of C. albicans SOD5 and were thus identified as Cu-SOD homologs. Among them, 2003 were single-domain proteins and 443 were CSRPs. Single-domain Cu-SOD homologs were primarily found in fungi, monocotyledon plants, unicellular eukaryotes, and some crustaceans. CSRPs, on the other hand, were found to exist only in the Kingdom Animalia, involving 11 Phyla that ranged from the Phylum Placozoa to the Phylum Chordata (Table 1). The Phyla Mollusca, Arthropoda, and Chordata encompass most species that were found to have CSRPs and many of the identified species were aquatic animals. Phylogenetic and conserved domain analyses of CSRPs from the representative species of the 4 Phyla revealed the molecular diversity of the protein class (Fig. 1). In the constructed phylogenetic tree, the selected CSRPs were grouped into five distinct clades. Clade I (light blue) included three-domain CSRPs from fish species. Clade II (green) comprised CSRPs from five insect species and one coelenterate. Clade III (purple) contained two-domain CSRPs from spiral shells. Clade IV (black) consisted of four-domain CSRPs from the lancelet. Clade V (red) included 11 four-domain CSRPs from bivalves and one three-domain CSRP from the bootlace worm. The number of conserved Cu-SOD domains contained in a CSRP molecule differed with the host species. In addition, the 2 oyster species, C. gigas and C. sikamea, were both identified to have 3 CSRP coding genes in the genome.

Table 1 The distribution of the CSRPs in eukaryotes.
Full size table
Fig. 1
figure 1

Phylogenetic and conserved domain analysis of the CSRPs from the representative species of Phyla Mollusca, Arthropoda, and Chordata. The CSRPs were identified from cnidarias (Allacma fusca [Afu]), nemerteans (Lineus longissimus [Ll]), molluscans (Haliotis rufescens [Hr], Batillaria attramentaria [Ba], Crassostrea giga [Cg], Crassostrea sikamea [Cs], Mytilus edulis [Me], Azumapecten farreri [Afa], Mizuhopecten yessoensis [My], Argopecten irradians [Ai], Mytilus galloprovincialis [Mg]), arthropods (Folsomia candida [Fc], Eciton burchellii [Eb], Eufriesea Mexicana [Em], Vespa mandarinia [Vm], Polistes canadensis [Pc]), and chordates (Danio rerio [Dr], Sinocyclocheilus rhinocerous [Sr], Heptranchias perlo [Hp], Branchiostoma lanceolatum [Bl]). Afu (Cnidaria-hydra), Ll (Nemertea-bootlace worm), Hr (Mollusca-gastropod), Ba (Mollusca-gastropod), Cg (Mollusca, Bivalvia-scallop), Cs (Mollusca, Bivalvia-scallop), Me (Mollusca, Bivalvia-scallop), Afa (Mollusca, Bivalvia-scallop), My (Mollusca, Bivalvia-scallop), Ai (Mollusca, Bivalvia-scallop), Mg (Mollusca, Bivalvia-scallop), Fc (Arthropod), Eb (arthropod-ant), Em (arthropod-bee), Vm (arthropod-hornet), Pc (arthropod-wasp), Dr (Chordata/Vertebrata-finfish), Sr (Chordata/Vertebrata-finfish), Hp (Chordata/Vertebrata-shark), Bl (Chordata/Cephalocordata-amphioxus). Three CSRP homologous genes were screened out from both C. sikamea and C. gigas, and one CSRP homologous gene was selected from each of the remaining species. Values on the evolutionary branches represent bootstrap values. Gray shading indicates the presence of signal peptides, white indicates the absence of signal peptides, and red-highlighted fonts represent CSRP homologous proteins in C. sikamea and C. gigas. “I” to “V” represented the five clades. The “i”, “ii”, “iii”, and “iv” in the domain section were the numeric of conserved Cu/Zn-SOD domains. The rest of the shapes represent the domain superfamily. The MDR superfamily is a zinc-dependent alcohol dehydrogenase-like family, and the VSP superfamily is a Giardia variant-specific surface protein.

Full size image

Predicted molecular characteristics of CSRPs from 2 oyster species

The six CSRPs from C. gigas and C. sikamea were identified as extracellular hydrophilic proteins of > 100 kDa with predicted isoelectric points ranging from 5.63 to 7.58. SignalP analysis detected a 22-amino acid signal peptide in Cs-CSRP1 and Cg-CSRP3 (Table 2). Amino acid sequence alignment of the six CSRP genes showed identity ranging from 56 to 98% (Table 3). They all contained 4 conserved Cu-SOD domains. Sequence comparisons at the domain level indicated that the similarity was enormously diverse between the domains within the same molecule, as compared to that in different molecules. In addition, the domain sequence similarity ranges from 38 to 56% within the same molecule. There are significant differences in the sequence similarity at the domain level between different molecules of the same species and between different species, both of which range from 30 to 100% (Fig. 2). According to the CSRPs of two species of oysters, it is concluded that the sequence similarity between Cg-CSRP1 and Cg-CSRP3 and the Cs-CSRP1 sequence is higher, and the sequence similarity between Cg-CSRP2 and the Cs-CSRP2 and Cs-CSRP3 sequences is higher.

Table 2 Predicted properties of CSRP homologs in C. gigas and C. sikamea.
Full size table
Table 3 Percent amino acid sequence identity between CSRPs from C. gigas and C. sikamea.
Full size table
Fig. 2
figure 2

Sequence similarity comparison of the four domains of oyster CSRPs. The percent similarity was shown.

Full size image

Domain comparison and structure modeling

Multiple sequence alignments of the oyster CSRP domains with human Cu/Zn-SODs (cytoplasmic SOD1 and extracellular SOD3), C. gigas Cu/Zn-SOD (XP_034334952.1), C. albicans SOD5, and C. gigas dominin3 (XP_011439797.2) revealed conserved amino acid residues essential for Cu²⁺ coordination in the enzyme’s active sites. However, two key histidine residues for Zn²⁺ binding were mutated, with one replaced by glutamate, mirroring the sequence of C. albicans SOD5 (Fig. 3).

Fig. 3
figure 3

Comparison of Cu-SOD domains of the oyster CRSPs. At the active site, the red letters on the top lines denote the Cu-binding sites, yellow for the dynamic histidine, blue for Zn-binding sites, and black for mutated Zn-binding sites. The similarity between the four domains of CSRPs ranged from 32–55%, with 35.67% identity and 52.23% similarity to C. albicans SOD5. The identity with Cg-dominin3 was 27%, with a similarity of 34%.

Full size image

The predicted structure of Cs-CSRP1 was modeled using Cg-Cu/Zn-SOD as a template in the SwissModel database. The active site and metal-binding conformation of Cs-CSRP1 were conserved across its four domains, with Cu²⁺ positioned similarly to C. albicans Cu-SODs (Fig. 4). Similar to C. albicans SOD5, the predicted domain structure retained the Greek key β-barrel fold characteristic of Cu/Zn-SODs but lacked two zinc-binding histidines and the extensive portion of the electrostatic loop (ESL).

Fig. 4
figure 4

Four conserved domains of the Cs-CSRP1 protein structure and their active sites. From left to right are the overall structure of Cs-CSRP1 protein, the single functional domain structure of Cs-CSRP1 protein, and the active sites of Cs-CSRP1 protein. Key structural features of the single functional domain: Greek key β-barrel core (purple), partial ESL (yellow), active site arginine (Arg-388), disulfide ring with cysteine (light blue), copper is a pink sphere, and its coordinating residues histidine (His-291, His-293, His-381). Dynamic histidine (His-316) is blue, and Glu-339 and Asp-342 are orange and green. The dashed lines in yellow indicate the hydrogen bond network.

Full size image

CSRP gene expressions in oyster tissues and oysters under stress

CSRP gene expression varied between tissues and genes in both C. sikamea and C. gigas. In C. sikamea, Cs-CSRP1 showed a significantly higher level in expression than the other 2 genes (Fig. 5a). In C. gigas, Cg-CSRP1 and Cg-CSRP3 expressed than Cg-CSRP2 (Fig. 5b). In both species, all the highly expressed genes were detected to have the highest expression level in the mantle.

Fig. 5
figure 5

Expression patterns of CSRPs in different tissues of C. sikamea and C. gigas. Amu: Adductor muscle, Dgl: digestive gland, Gil: gill, Man: mantle, Gon: gonads. n = 3.

Full size image

In C. sikamea, Cs-CSRP2 was excluded from stress analyses due to low expression levels (< 0.01). Under salinity stress, Cs-CSRP1 expression remained stable, while Cs-CSRP3 was significantly upregulated at 0 ppt after 5 days (p < 0.001) (Fig. 6a). During acute air exposure, Cs-CSRP1 and Cs-CSRP3 expression levels decreased overall but peaked at 6 h for Cs-CSRP1 and 12 h for Cs-CSRP3 (Fig. 6b). Under acute 35 °C heat stress, both Cs-CSRP1 and Cs-CSRP3 peaked at 6 h. During combined air exposure and heat stress, Cs-CSRP1 was downregulated at 12 h (p < 0.001), while Cs-CSRP3 was significantly upregulated. Under chronic heat stress, Cs-CSRP1 showed higher expression than Cs-CSRP2 and Cs-CSRP3, with significant differences across temperature treatments (Fig. 6c).

Fig. 6
figure 6

Expression patterns of CSRPs in C. sikamea under abiotic stress. (a) S_0, S_5, S_30, and S_40 indicate salinity of 0, 10, 30, and 40, respectively. (b) A, H35, and A_H35 represent air exposure stress, 35℃ heat stress, and combined air exposure with heat stress, respectively. (c) H20, H28, and H35 represent 20℃, 28℃, and 35℃ heat stress, respectively. “*” indicates significant differences: “*” (p ≤ 0.05), “**” (p < 0.01), “***” (p < 0.001), “****” (p < 0.0001), and “ns” indicates not significant (p > 0.05). n = 3.

Full size image

Recombinant protein expression and SOD activity

The recombinant Cs-CSRP1 plasmid was successfully constructed and expressed in E. coli. SDS-PAGE analysis confirmed significant expression of Cg-Cu/Zn-SOD at approximately 34 kDa (Fig. S1a, b). Cs-CSRP1 was primarily expressed as inclusion bodies, with clear bands around 110 kDa (Fig. S1c). Both recombinant proteins were purified via Ni²⁺ affinity chromatography, yielding concentrations of 0.5 to 1 mg/mL. Western blotting verified successful purification.

SOD activity assays revealed Cs-CSRP1 activity at 45.123 ± 4.8 U, compared to 86.194 ± 24.1 U for Cg-Cu/Zn-SOD (positive control). The experiment was repeated three times under the same conditions for detection. Higher activity was observed in Cs-CSRP1 inclusion bodies than in the soluble supernatant (Table 4).

Table 4 SOD activity of Recombinant proteins.
Full size table

Discussion

Oysters frequently encounter oxidative stress due to their intertidal habitats and filter-feeding lifestyle. Extreme fluctuations in environmental conditions such as salinity, temperature and the combination of drying in air and heat (air exposure) pose tremendous challenges to the robustness of the antioxidant defense of these species. As a result, SODs, which are believed to function as the most important frontline antioxidant enzymes, are likely under constant pressure of evolution. Our recent research revealed the presence of a cytoplasmic Cu/Zn-SOD in Pacific oysters21. However, results of multiple studies indicate that the oyster homolog of the extracellular Cu/Zn-SOD is characterized to have lost the SOD activity in evolution40raising a question if there are distinct molecules playing the role of EC-SOD in extracellular antioxidant defense. CSRPs that were predicted to have the potential of acting as an extracellular SOD are present with multiple coding genes in the genome of oysters20. In the present research, we gained more insights into the potential of CSRPs acting as new type of SODs in oysters. While it remains to be investigated what specific ecological factors have been driving these molecules’ evolution, our findings indicate that CSRPs likely play a role of extracellular SOD in oyster antioxidant defense.

Several lines of evidence support the speculation that CSRPs can function as extracellular SODs. CSRPs are identified as a class of multiple tandem Cu-SOD domain proteins14. They are found in aquatic organisms and insects and are predicted to be extracellular14,20,41. Our analysis reveals the structural basis for the SOD activity in the oyster CSRPs. Like fungal Cu-SODs, CSRPs in C. sikamea contain signal peptides, suggesting the extracellular presence of the mature proteins. Structural modeling predicts the conformational similarities between the oyster CSRP domain and the C. albicans Cu-SOD, which include proper folding and the copper ion presence in the active sites (Fig. 4). More importantly, recombinant Cs-CSRP1 demonstrates significant SOD activity, providing the essential attribute of CSRPs acting as a SOD enzyme. Studies of C. albicans Cu-SOD have shown superoxide catalysis rates comparable to Cu/Zn-SODs13,15,17. In the parallel experiment of expression and activity assay, we observed a level of enzyme activity in the recombinant icCg-Cu/Zn-SOD comparable to that in the previous study21indirectly validating the recombinant expression and activity assay of the CSRP in the present research. Notably, the inclusion body-expressed protein showed stronger activity and stability than the supernatant-expressed protein. It is likely that most of the secreted CSRPs are not folded correctly in the prokaryotic expression system due to their molecular size of > 100 kDa. Instead, the renaturation buffer with changes in pH and ionic strength helps the polypeptides to achieve correct folding. Future studies using different expression systems, such the yeast or other eukaryotic systems, will generate more information for the explanation of this phenomenon.

The similarities in amino acid sequence and conformation between the fungal Cu-SOD and the CSRP domains may place these proteins in the same family. The protein family is, however, extremely diverse in its molecular properties. For example, a single domain protein has a molecular weight of 20–30 kDa, but a CSRP is around 100 kDa4,16. There are also differences in the distribution of single domain Cu-SODs and CSRPs, with the former being found mainly in unicellular microorganisms such as bacteria, fungi and oomycetes, and the latter being present in animals, including some mollusks, chordates, and insects14. In addition, CSRPs differ significantly in molecular size and the number of functional domains. Phylogenetic analysis grouped CSRP homologs into five branches based on domain sequence similarity. However, Cs-CSRP1 from C. sikamea contains four Cu/Zn-SOD domains that are predicted to have the SOD activity as they all show conformational similarities with the C. albicans Cu-SOD. In contrast, a CSRP from the zebrafish D. rerio is found to contain three repeat domains that possess the copper-binding site as that of the fungal Cu-SOD14. Although it is immature to specifically speculate on the biological significance of these molecular diversities, the structural differences likely entail functional diversities. Multi-domain proteins are believed to offer significant advantages over single-domain proteins, such as enhanced functional versatility42improved structural stability and dynamics43superior signaling and allosteric regulation44and evolutionary advantages45. Some of these features of multi-domain proteins may be particularly significant for species in fluctuating habitats such as aquatic animals, which conforms with the species distribution of CSRPs16. The molecular diversities of the Cu-SOD protein family also bring about the mechanisms of molecular evolution. One possibility could be that a transfer of the genetic materials between a single-domain Cu-SOD holding species and the ancestors of the animals via horizontal gene transfer (HGT) has led to the independent evolution of the single-domain protein in the unicellular species and the CSRPs in animals. However, any speculations about the evolutionary mechanism of the diverse protein family will require assessments based on more future studies.

Gene expression profiling of CSRPs in C. sikamea and C. gigas revealed high expression variations among the analyzed tissues and between genes in the same species. In both species, mantles exhibit the highest levels of expression of the CSRP genes, suggesting the functional significance of the coded proteins in the tissues. Mantles are membranous tissues wrapping the most part of oyster soft body and being liable to the impacts of environmental conditions and thereby oxidative stress46. The higher levels of expression of CSRP genes in oyster mantles are thus in conformation with the predicted role of the coded proteins in functioning as a SOD. In the meantime, variations in levels of expression between different genes are also observed in the present research. In addition to the different expression patterns of different genes across tissue types in both oyster species, we found that Cs-CSRP1 constantly exhibit higher levels of expression than the other two homologous genes in oysters under chronic heating stress and at certain point (e.g., 20 °C) the difference has been determined to be statistically significant. Studies in the Zhikong scallop Chlamys farreri have revealed that different families of SOD exhibit different patterns of expression in response to stressors and CSRP genes significantly overexpress in scallops under toxic algae stress47. Gene duplication represents the most important driving force in evolution to generate paralogous genes and duplicate genes usually diverge in expression pattern after subfunctionalization or neofunctionalization48,49. Therefore, the variations in expression patterns between the CSRP genes in the same species likely reflect the functional diversification of CSRPs in oysters and the scallop. In turn, this functional diversification of CSRP paralogues may strengthen the antioxidant defense of the poikilothermal and metamorphic marine bivalves during stress response. More studies are needed before any conclusive speculations regarding the specific functional significance of this expression divergence and the related evolutionary mechanism can be made.

In conclusion, our study demonstrates that C. sikamea CSRPs possess extracellular SOD activity, elucidating the molecular diversity of Cu/Zn-SOD homologs in oysters. Gene expression profiling revealed significant differences in CSRPs between C. gigas and C. sikamea. While their exact physiological roles remain speculative, the presence of signal peptides and SOD activity suggests involvement in extracellular ROS regulation and redox homeostasis. Whether CSRPs function similarly to fungal Cu-SODs warrants further investigation.