Prokaryotic Expression, Purification, and Functional Characterization of the Large Yellow Croaker (Larimichthys crocea) Mannose Receptors Subunits (MRC1 and MRC2)

Xiangli Dong1,2, Shilin Mikhail Borisovich2, Jiji Li1,*, Jianyu He1, Zeqin Fu1, Yingying Ye1, Julia N. Lukina3, Olga V. Apalikova4 and Jianshe Zhang1 1National Engineering Research Center for Marine Aquaculture, Zhejiang Ocean University, Haida South Road 1, 316022, Zhoushan, Zhejiang, PR China 2Department of Ecology, Russian State Hydro-meteorological University, Petersburg, Malookhtinskij Street 98, 195196, St. Petersburg, Russia 3Saint-Petersburg State Academy of Veterinary Medicine, Chernigovskaya Street 5, 196084, St. Petersburg, Russia 4State Research Institute of Lake and River Fisheries, Embankment Makarova Street 26, 199004, St. Petersburg, Russia Article Information Received 20 January 2020 Revised 01 March 2020 Accepted 09 March 2020 Available online 22 January 2022 (early access)


INTRODUCTION
M arine fish play a significant role in maintaining the stability of marine ecosystems and provide a high quality protein source for human beings (He et al., 2014). There are many marine fish species with high economic value and methods have been developed to grow them in artificial marine culture conditions . For instance, the large yellow croaker (Larimichthys crocea), which belongs to the Larimichthys genus and Sciaenidae family, is a common commercial marine-cultured fish highly valued in China since the 1960s due to its high nutritional quality, palatability,

O n l i n e F i r s t A r t i c l e
understand the anti-infective immune response in croaker by characterizing mannose receptor genes and their expression profiles (Dong et al., 2016). According to the experimental results, before the V. anguillarum infection, mRNAs of the mannose receptor gene were expressed in 8 tissues including liver, kidney, spleen, intestine, and heart, especially in liver, kidney and spleen. After V. anguillarum infection, the expression levels in liver, kidney and spleen were significantly higher than before infection, which proves that it is possible that the mannose receptor gene can regulate the infection of V. anguillarum. The mannose receptor (MR) is part of the C-type lectin glycoprotein superfamily, which has three other members: the M-type receptor for secretory phospholipases A 2 (PLA 2 R), DEC-205/gp200-MR6, and Endo180/uPARAP (Boskovic et al., 2006). All 4 members are type I transmembrane receptors, which contain an N-terminal cysteine-rich domain and a single fibronectin type II (FNII) domain. They differ from other superfamily members in that they have multiple C-type lectin-like domains (CTLDs) within a single polypeptide backbone (eight in the case of MR (Miron, 1992;Taylor et al., 1990)), PLA 2 R, and Endo180, and ten in the case of DEC205 (Zheng et al., 2015;Jiang et al., 1995). These proteins are able to cycle between the plasma membrane and the endosomal apparatus due to discrete motifs present within their cytoplasmic domains. Despite their overall structural similarity, these four receptors have evolved to use different domains to interact with discrete ligands (East and Isacke, 2002). In addition, they differ in their ability to mediate endocytic and phagocytic events and in their intracellular destinations. Together, they represent a unique group of multi-domain, multi-functional receptors (East and Isacke, 2002).
MR was first recognized as a receptor involved in the clearance of endogenous glycoproteins in the late 1970s (Pontow, 1991). It obtained its name because its lectin activity terminated in mannose, fucose, or N-acetyl glucosamine (Kilpatrick, 2010). Since its initial description, it has been the focus of significant structural and functional characterization. It has been implicated in the recognition of exposed mannose residues on the surface of certain pathogens, and the internalization of mannosylated antigens results in enhanced T cell presentation (Taylor et al., 2005). MR is a pattern recognition receptor (PRR), a group of receptors that play a significant role in innate immunity responses through binding to pathogen-associated molecular patterns (PAMPs) (Gazi and Pomares, 2009). MR is primarily expressed in macrophages and dendritic cells (East and Isacke, 2002).
Two subunits (MRC1 and MRC2) of MR were isolated in the large yellow croaker (Larimichthys crocea) (Dong et al., 2016). Both subunits' structures contain extracellular, transmembrane, and cytoplasmic regions. MRC1 of L. crocea (L.c-MRC1) has an extracellular region that consists of three domains: an N-terminal cysteinerich (CR) domain, a fibronectin type II (FN II) domain, and eight tandemly arranged C-type lectin-like domains (CTLDs). L.c-MRC2 is similar to L.c-MRC1, but lacks the CR domain (Dong et al., 2016). Based on their secondary structures, the tertiary structures of the two proteins could be different, suggesting that each subunit plays different functional roles in the croaker. We previously performed theoretical analysis of MRC1 and MRC2 gene expression levels (Dong et al., 2016). Here, we expressed the two proteins in prokaryotes and purified them. Our findings could help to develop an antibacterial vaccine, which would avoid the side effects of antibiotics and provide a greener treatment for farmed fish.

PCR reaction conditions
The PCR reaction conditions were established using DH5α-pMD18-T-MRC1-C and DH5α-pMD18-T-MRC2-C as templates, and the MRC1-C and MRC2-C gene fragments were amplified. The reaction was performed in a 50 µL volume, including 25 µL PreMix Taq, 1.6 µL dNTPs, 1 µL primer-F, 1 µL primer-R, 1 µL template cDNA, and 22 µL H 2 O. The PCR amplification O n l i n e

F i r s t A r t i c l e
was conducted in a thermal cycler (Bio-Rad, USA) using the following amplification conditions: 5 min at 94°C, followed by 35 cycles of 30 s at 94°C, 30 s at 58°C, and 60 s at 72°C, with a final extension of 5 min at 72°C. After double digestion with EcoRI/XhoI, the target fragment was recovered using a DNA gel recovery kit. Preparation of competent E. coli BL21 (DE3) using CaCl 2 and transformation of E. coli BL21 (DE3) were performed using methods published in the "Guidelines for Molecular Cloning" (Nguyen et al., 2002). One microliter of culture was used to screen for positive clones by PCR. The PCR was performed in a 15 µL volume, including 1 µL template pMD18-T-MRC1 / 2-C, 1 µL MRC-R, 1 µL MRC-F, 7.5 µL PreMix Taq, and 4.5 µL H 2 O. The PCR amplification was conducted in a thermal cycler (Bio-Rad, USA) using the following conditions: 5 min at 94°C followed by 35 cycles of 30 s at 94°C, 30 s at 58°C and 60 s at 72°C, with a final extension of 10 min at 72°C. Plasmids were extracted from PCR-positive bacterial cultures using a small volume extraction kit (Kehaojia Biological Technology Co., Ltd., OMEGA, Wuhan) and subjected to double restriction digestion. Electrophoresis was performed and the results were examined under UV light. Positive clones were identified by PCR and restriction enzyme digestion. The recombinant plasmids were then sequenced.

Expression of L.c-MRC1-C and L.c-MRC2-C in E. coli BL21 (DE3)
We chose E. coli as the host bacterium. The recombinant plasmids MRC1 (1044 bp) -pET32A and MRC2 (993 bp) -pET32A were transformed into BL21 (DE3) E. coli. Protein expression was induced by incubating at 37°C on a 220 rpm shaker until the OD600 = 0.4−0.6. IPTG was added to a final concentration of 0.5 mM and the cells were incubated at 18°C for 10 h. The bacteria were precipitated and collected.

Protein purification and detection
A fresh recombinant colony was inoculated in 10 mL of LB medium and grown on a shaker to obtain a saturated culture. 500 µL of the overnight culture was then inoculated into 50 mL of LB medium, shaking at 37°C. When the culture OD600 reached 0.6, it was centrifuged at 10000 × g at room temperature. The supernatant was discarded and the samples were heated for SDS-PAGE analysis. SDS-PAGE was performed to analyze the expression of the exogenous genes in E. coli. The identified BL21 bacteria were inoculated into 40 mL of LB medium and expression was induced with 0.5 mM IPTG at 18°C for 10 h. After the medium was induced, the bacteria were collected by centrifugation. The cell pellet was resuspended and ultrasonic cracking was conducted. After centrifuging at 10000 rpm for 20 min at 4°C, the supernatant and precipitate were collected and sampled. The cells were resuspended in 40 mL of inclusion body solution. The supernatant was collected and filtered.

Target fragment determination
Membrane protein analysis showed two transmembrane domains in MRC1 and MRC2: a fibronectin type-II domain and a C-type lectin domain (Fig. 1). We then used Epitlot software (http://tools.iedb. org/bcell/) for epitope analysis; a higher amino acid score indicates that the epitope is more likely to form antibodies. Figure 2 shows regions with high antigenic epitope scores (yellow). We found two epitope concentrated areas, which corresponded to the Fibronectin type-II domain and the C-terminal side of the C-type lectin domain.

O n l i n e F i r s t A r t i c l e
Expression, Purification and Characterization of the Large Yellow Croaker

5
The later region is highly specific and would not lead to cross reactivity between MRC1 and MRC2. Thus, the C-type lectin domain region was selected as the target immunogenic region. In the above analysis, two fragments were selected for MRC1 and MRC2 ( Fig. 3A; Table II). Cross analysis of the two fragments identified MRC1 amino acids 795 to 1143 and MRC2 amino acids 681 to 991 for the generation of antibodies. For MRC1 and MRC2 (681 ~ 991), the fragment homology was 6.11%. For MRC2 (681 ~ 991) and MRC1 (795 ~ 1143), the fragment homology was 14.07%. The corresponding antibodies produced by these two antigens would not lead to cross recognition.

Construction of MRC1 and MRC2 cloning vectors L.c-MRC1 and L.c-MRC2
were PCR amplified and run on an agarose gel. The L.c-MRC1 amplicon was about 1000 bp, which was consistent with the predicted size of L.c-MRC1 (1044 bp). The L.c-MRC2 amplicon was also around 1000 bp, which was also consistent with the predicted size of L.c-MRC2 (933bp) (Fig. 3). The target fragments were then ligated into the pET32A cloning vector. MRC1 (1044bp) -pET32A and MRC2 (993bp) -pET32A clones were preliminarily verified by bacterial PCR and double restriction digests (Fig. 3B). Bacterial PCR analysis showed 1000 bp and 2000 bp bands from the MRC1 (1044 bp) -pET32A culture and a 1000 bp band from the MRC2 (993 bp) -pET32A culture. The 1000 bp bands were likely the target fragments (Fig. 4). The MRC1-pET32A and MRC2-pET32A plasmids were then extracted and subjected to double restriction digests. As shown in Figure 3C and D, there were bands between 2000 bp and 1000 bp, which correspond to the pET32A fragment and the MRC1 (1044 bp) or MRC2 (993 bp) fragments. The preliminary results showed successful construction of the MRC1 (1044 bp) -pET32A and MRC1 (993 bp) -pET32A cloning vectors. Subsequent digestion verified that the insert was correct.

Protein expression and purification
As shown in Figure 4A, protein expression from the recombinant MRC1 and MRC2 vectors was induced and yielded high expression of proteins between 45 and 66.2 kDa. This was consistent with the predicted MRC1 (1044 bp) protein size (60 kDa) and the predicted MRC2 (993 bp) protein size (54 kDa) (Table III). An empty vector-induced control showed no such protein (Fig. 4B). We also tested for protein in the precipitate (Table III), further confirming protein expression. After purification, the protein fragments were between 45 kDa and 66.2 kDa (Fig. 4C), this result is consistent with the predicted protein length.
We analyzed the purity of the recovered proteins and found single bands in between 45 kDa and 66.2 kDa (Fig. 4C). L.c-MRC1 and L.c-MRC2 proteins were predicted to be 60 kDa and 54 kDa, respectively. This indicated that the proteins were of excellent purity and could be used for immunization and preparation of antibodies.

DISCUSSION
The use of the large yellow croaker (L. crocea) has been compromised in recent years by widespread disease outbreaks due to various bacterial pathogens (Taylor et al., 2005;Zheng et al., 2015;Jiang et al., 1995;East and Isacke, 2002). In particular, Vibrio anguillarum is a major pathogen of farmed fish that causes severe damage and heavy economic losses to the fish farming industry. The most susceptible fish species include both marine and freshwater fish such as Japanese flounder, European sea bass (Aamri et al., 2015;Zlotkin et al., 1998), Asian seabass (Kayansamruaj et al., 2015), rainbow trout (Lahav et al., 2004;Safari et al., 2016), zebrafish (Membrebe et al., 2016), Nile tilapia (Membrebe et al., 2016), channel catfish (Xia et al., 2008;, and large yellow croaker. In this study, we tested the prokaryotic expression and purification of two L. crocea mannose receptor gene subunits (MRC1 and MRC2). The mannose receptor (MR) is a pattern recognition receptor (PRR), a type of receptor that plays a significant role in innate immunity immune responses through binding to pathogenassociated molecular patterns (PAMPs) (Gazi and Pomares, 2009;Chen et al., 2019). MR recognizes surface polysaccharides of various pathogens, such as viruses, bacteria, yeasts, and parasites, including HIV (Nguyen and Hildreth, 2010), Dengue virus (Miller et al., 2008), Candida albicans (Marodi et al., 1991), Mycobacterium tuberculosis (Lailleux et al., 2003;Rajaram et al., 2010), Pneumocystis carinii (Ezekowitz et al., 1991;Swain et al., 2003), Cryptococcus neoformans (Dan et al., 2008), Klebsiella pneumonia (Zamze et al., 2002), Streptococcus pneumonia (Zamze et al., 2002) and Leishmania spp. (Chakraborty et al., 1998(Chakraborty et al., , 2001. In recruited inflammatory peritoneal macrophages, MR levels were increased in response to interleukin-4 (IL-4), IL-13, and IL-10 (Chakraborty et al., 2001). Owing to its crucial role in innate immunity responses, MR has been extensively studied in humans and mice. However, there have been few studies of MR in fish. Additionally, prokaryotic MR protein expression levels were undetermined. Therefore, we chose to study the mannose receptor gene. Prior to this study, we have studied the quantitative expression of the mannose receptor gene. Based on this, we continued to study the prokaryotic expression and purification of the gene of the L. crocea, and obtained good expression and higher purification results.
The MRC1 subunit contains four domains: an extracellular region containing a cysteine-rich (CR) domain, a domain containing fibronectin type two repeats (FNII) and multiple C-type lectin-like carbohydrate recognition domains (CTLDs), a transmembrane domain, and a short cytoplasmic tail. MRC2 is similar to MRC1, but lacks the extracellular CR domain (Dong et al., 2016). We performed membrane protein analysis in the process of antigen preparation and found two conserved domains of MRC1 and MRC2: the fibronectin type-II domain and the C-type lectin domain. Further, epitope analysis identified two concentrated epitope areas in the fibronectin type-II region and the C-type lectin region (Fig. 2). The C-terminal side of the C-type lectin region is highly specific, would not cause cross-reactivity between MRC1 and MRC2, and would lead to better clones. Thus, the C-type lectin region was selected as the immunogen region. We performed cross identification analysis on the selected target fragments (Fig. 2). Homology comparison of MRC1 (795 ~ 1143 bp) and MRC2 (681 ~ 991 bp) suggests that the corresponding antibodies produced by these two antigens would not be cross reactive. The MRC1 fragment (amino acids 795 to 1143) and the MRC2 fragment (amino acids 681 to 991) would generate antibodies against the C-type lectin regions of each respective protein subunit. These two regions are within the CTLDs 5−7. Although both MRC1 and MRC2 have 8 CTLDs, CTLD 4−8 have higher glycoprotein binding capacity (Dong et al., 2016). The 8 CTLDs domains of L.c-MRC1 are: residues 212 to 341, 360 to 484, 503 to 625, 645 to 770, 791 to 913, 933 to 1070, 1085to 1203, and 1220to 1346, and the 8 CTLDs of L.c-MRC2 are: residues 97 to 222, 245 to 368, 383 to 507, 531 to 666, 683 to 809, 830 to 965, 982 to 1110 to 1247) (Dong et al., 2016). Comparison of the tertiary structures of the two proteins (Fig. 5) indicated that the MRC1-C and MRC2-C subunits may have different functions. Our findings further our understanding of innate immunity in L. crocea and will enable future studies of its antimicrobial defense mechanisms. Our results could help to establish effective disease control measures and potentially pave the way for developing an MR-based vaccine against bacteria, viruses, and other infections in L. crocea. Previously, MRC1 and MRC2 genes were found to be expressed in various L. crocea tissues at different tissuespecific expression levels. Bacterial challenge experiments showed that MRC1 and MRC2 mRNA expression levels were upregulated in the liver, spleen, and kidney of young fish, suggesting that MRC1 and MRC2 contribute to the defense against pathogenic bacterial infections. Prokaryotic expression and antibody generation will provide necessary tools for further study of L. crocea MR.
As opposed to the traditional use of antibiotics and antimicrobial compounds, vaccination is an effective green intervention that can help to control V. anguillarum infection in aquaculture . Genetically engineered vaccines are safer and more serotype-independent . However, efforts to develop subunit vaccines are limited by the lack of the conserved protective antigens. Thus, the identification and screening of novel and conservative immunogenicity antigens is crucial for the development of an effective subunit vaccine (Tian et al., 2011). In this study, we evaluated the immunogenicity of L.c-MRC1 and L.c-MRC2 and determined that these two candidate immunogens could potentially be used as an effective vaccine against V. anguillarum infection in L. crocea.

CONCLUSION
Here, we report the cloning of partial L. crocea mannose receptor subunit (MRC1 and MRC2) cDNAs. The amino acid sequences of MRC1 and MRC2 were highly conserved with other vertebrates. Using membrane protein analysis, epitope analysis, and target fragment cross identification analysis, the C-type lectin region was identified to be highly specific and not cross reactive between MRC1 and MRC2. Thus, it was selected as the immunogen region. These fragments were then cloned into a prokaryotic expression vector. Expression was induced in E. coli and the proteins were purified. We achieved high protein expression levels and excellent protein purity. The purity and expression levels were suitable for antibody preparation.