Update on the source of phenoloxidase activity in the hemolymph of kuruma prawn Marsupenaeus japonicus

Crustacean phenoloxidase (PO) and hemocyanin (Hc) are classified as type 3 copper proteins. PO catalyzes the oxidation of mono- and di-phenol compounds, which is the rate-limiting step of melanization, while Hc generally functions as a dioxygen-transporting protein in the hemolymph of arthropods. To date, many studies have shown PO activity in Hc, which is inspired by their structural similarity. Here, the source of PO activity in crustaceans was re-examined by purifying Hc and PO exclusively from the hemolymph of kuruma prawn. The conventional procedure for the preparation of arthropod Hc, which includes precipitation of Hc by ultracentrifugation and subsequent purification by size exclusion chromatography, was not able to completely remove hemolymph-type PO from Hc. In contrast, fractionation with 50% saturation of ammonium sulfate and subsequent hydrophobic chromatography yielded sufficiently pure Hc, which contained no detectable PO protein and virtually no PO enzymatic activity. These results indicate that the main source of PO activity in the hemolymph of kuruma prawn is hemolymph-type PO and that the improved purification method of Hc is preferable for evaluating the PO activity of Hc.


Introduction
Melanization in crustaceans is a critical problem in the preservation and processing of shrimp and crabs (Gonçalves and de Oliveira 2016). It causes the deterioration of food value by forming black spots on the body of crustaceans after harvest and food distribution. Phenoloxidase (PO) is a type 3 copper protein, which is characterized by bi-nuclear copper atoms at the active site. PO, also known as tyrosinase in mammals, plants, fungi, and bacteria, catalyzes the hydroxylation and oxidation of mono-phenols (monophenolase activity; E.C. 1.14.18.1) and the oxidation of o-diphenol substrates to the corresponding quinone [diphenolase (catecholoxidase) activity (E.C. 1.10.3.1)], both of which are rate-limiting reactions of melanin formation; it is designated as the key enzyme for melanization (Solomon et al. 1996). This reaction produces toxic quinone species as an intermediate that attack pathogens, and the final product, melamin, facilitates the encapsulation of pathogens and wound healing in arthropods (Cerenius et al. 2008). Due to the high reactivity of PO, its activity must be tightly regulated. Thus, PO is synthesized as an inactive pro-enzyme, prophenoloxidase (proPO), whose activation is regulated by a protease cascade composed of several steps (Cerenius and Söderhäll 2004). Hemocyanin (Hc), another member of the type 3 copper protein, functions as a dioxygen carrier protein in the hemolymph of arthropods and mollusks. Crustacean PO and Hc are derived from a common ancestral protein (Burmester 2002), whereas molluscan Hc has a structure that is different from that of crustaceans (Markl 2013).
Whether Hc can acquire PO activity due to the closely related properties of PO and Hc from arthropods remains an intriguing question. Since the 1990s, the ability of Hc to gain PO activity has been revealed in chelicerates (Decker and Rimke 1998;Decker et al 2001;Nagai et al. 2001;Nagai and Kawabata 2000) and crustaceans (Zlateva et al 1 3 1996), as well as mollusks (Salvato et al. 1998;Morioka et al. 2006;Suzuki et al. 2008). Thereafter, the PO activity of Hc has been reported, not only in chelicerates, but also in crustaceans. In earlier studies, Hc was usually purified to a dissociated subunit from a multimeric complex in the hemolymph of arthropods for characterization. However, over the past two decades, simplified purification methods have been adopted in most studies on the activation of Hc in crustaceans. This conventional purification method for Hc from hemolymph is composed of two steps; the first one is ultracentrifugation (UCF) by which Hc is precipitated due to its large molecular size, and the second one is size exclusion chromatography (SEC), which is also a purification method based on the apparent molecular weight of the candidate protein. This purification procedure seems to work well and produces Hc with sufficient purity, because Hc is extremely abundant in the hemolymph plasma of arthropods, that is, Hc is suggested to account for 90-95% of total protein in crustacean hemolymph plasma (van Holde and Miller 1995). In contrast to chelicerates that have no intrinsic gene encoding PO (Nellaiappan and Sugumaran 1996;Nagai and Kawabata 2000), there is a possible source of PO activity in the hemolymph of crustaceans, a hemolymphtype PO. Initially, arthropod PO molecular species were identified in insects (Hall et al. 1995;Fujimoto et al. 1995;Kawabata et al. 1995) andcrayfish (Aspan andSöderhäll 1991;Aspán et al. 1995). These well-characterized POs are expressed in hemocytes and remain proPO under normal conditions (Ashida and Söderhäll 1984;Aspan and Söderhäll 1991). However, another type of PO molecular species that is distributed in the hemolymph of crustaceans has been found and identified in the last decade (Olianas et al. 2005;Masuda et al. 2012Masuda et al. , 2018. Furthermore, the high-resolution crystal structures of crustacean Hc and PO revealed that they share common structural characteristics (Li et al. 2009;Masuda et al. 2014Masuda et al. , 2020Hu et al. 2016), such as the hexameric state of their quaternary structure, even though the precise assembly of the constituent subunits is different (Masuda et al. 2014(Masuda et al. , 2020. It is possible that this kind of hemolymph-type proPO (proPOβ in kuruma prawn, Marsupenaeus japonicus) could contaminate the purified sample of Hc, and could in turn contribute to misunderstanding of the source of PO activity.
To investigate exactly whether and how Hc is converted to an active form with PO activity, one has to remove the intrinsic hemolymph-type proPO from the purified sample of Hc. Here, the author evaluated the conventional Hc purification strategy in terms of the contamination of intrinsic PO in the hemolymph of crustaceans, and shows a suitable and simple purification method for the preparation of Hc without hemolymph-type PO.

Preparation of hemolymph plasma
Live kuruma prawns (M. japonicus) were purchased from a local market (Sankatsu Shoten) in the Kyoto City Central Wholesale Market. Hemolymph was withdrawn from live kuruma prawns according to a previously described method (Masuda et al. 2012). The composition of the anti-coagulation buffer used was 10 mM Tris-HCl pH 7.5, 0.45 M NaCl, 10 mM KCl, and 10 mM ethylenediaminetetraacetic acid (EDTA).

Purification of Hc from hemolymph of kuruma prawn
Conventional purification for Hc was carried out according to a previously reported method (van Holde and Miller 1995;Decker and Rimke 1998;Adachi et al. 2001). In brief, the extracted hemolymph was centrifuged at 500 × g for 30 min to remove hemocytes. Subsequently, the supernatant was further centrifuged at 10,000 × g for 15 min to obtain the cleared hemolymph supernatant. Each 1-ml aliquot from the resultant supernatant was subjected to ultracentrifugation (UCF) at 200,000 × g for 3 h using an ultracentrifuge (CP 90NX, HiMac, Tokyo, Japan). The dark blue precipitant in each tube was dissolved in an equal volume of the buffer (20 mM Tris-HCl, pH 7.5 containing 0.15 M NaCl and 1 mM EDTA), followed by SEC using Superdex 200 pg 16/60 column (GE Healthcare, Chicago, IL). The purified Hc obtained by this conventional method is designated as Hc_cv here.
Another method for purification of Hc was performed as follows: ammonium sulfate was added to the supernatant of 10,000 × g centrifugation to achieve 50% saturation (2.05 M). The solution was mixed gently at 4 °C overnight, followed by centrifugation at 12,000 × g for 15 min. The resulting supernatant was subjected to hydrophobic chromatography using butyl Toyopearl M resin (Tosoh, Tokyo, Japan). The butyl Toyopearl resin was packed in an Econo column (1.5 × 20 cm, BioRad), followed by equilibration with buffer containing 10 mM Tris-HCl pH 7.5 and 50%-saturated ammonium sulfate (2.05 M). Then, the supernatant was applied to the column and washed with the equilibration buffer until the unbound proteins did not elute anymore. The proteins were eluted by linear gradient of ammonium sulfate from 50 to 0% saturation, or 2.05-0 M. An aliquot of the rest was desalted using a PD MiniTrap G-25 desalting column (GE Healthcare) for SDS-PAGE and western blot analysis. The Hc preparation purified by this new method is designated as Hc_nv here.

SDS-PAGE separation and immunoblotting
Proteins were visualized and analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting using a rabbit antiserum raised against proPOβ (Masuda et al. 2018). For activity staining on the gel, samples were prepared using sample buffer without 2-mercaptoethanol and not heat-treated (non-reducing SDS-PAGE). After the non-reducing SDS-PAGE, the gel was dipped in a substrate solution containing 5 mM L-DOPA and 0.1% SDS as an activator. Otherwise, each sample was mixed with an equal volume of 2 × SDS-PAGE sample buffer [125 mM Tris-HCl pH 6.8, 4% SDS (w/v), 20% (v/v) glycerol, 10% (v/v) 2-mercaptoethanol] and subjected to heat treatment (5 min in boiling water) before SDS-PAGE (reducing SDS-PAGE). For western blot analysis, proteins on the gel were transferred to a PVDF membrane and probed with anti-proPOβ antiserum (Masuda et al. 2012(Masuda et al. , 2018. Horseradish peroxidase-conjugated anti-rabbit IgG (Promega, Madison, WI) was used as the secondary antibody for chemiluminescence detection. Chemiluminescence signals were visualized using a set of reagents (Chemilumi-One L, Nacalai, Kyoto, Japan) and a laser imager (Odyssey Fc, LI-COR Bioscience, Lincoln, NE). Unstained and Dual color Precision Plus Protein Standards (BioRad, Hercules, CA, USA) were used in Fig. 1a and b, respectively, while the prestained protein size marker (DynaMarker Protein MultiColor III, BioDynamics Laboratory, Tokyo, Japan) was used in Figs. 3 and 4. DynaMarker Protein MultiColor III was visualized by fluorescence excited by near-infrared laser light in cases of western blotting.

N-terminal sequence analysis and cDNA cloning of Hc subunits
Purified Hc from kuruma prawn hemolymph plasma was separated using SDS-PAGE, followed by electroblotting to a PVDF membrane. The bands were visualized by Ponceau S and subjected to protein sequence analysis using a protein sequencer (Procise 494 system, Applied Biosystems, Waltham, MA) based on the Edman degradation method.

Assay for PO activity
PO is usually synthesized as an inactive pro-enzyme (proPO). Hence, the in vitro activation of proPO requires the addition of detergents such as SDS or proteases (Hall et al. 1995;Yasuhara et al. 1995;Masuda et al. 2012 To determine the yield of protein purification and to detect the di-PO activity of each preparation after each purification step, the enzymatic activities were assayed using a microplate reader (Sunrise, Tecan, Männedorf, Switzerland) at 25 °C in a buffer containing 10 mM Tris-HCl (pH 7.7), 0.15 M NaCl, 0.1% SDS, and 5 mM L-DOPA. Among these components, L-DOPA was a substrate for di-PO activity, while 0.1% SDS was added as an activator of proPO and Hc. The formation of dopachrome was monitored by measuring the absorbance at 492 nm every 15 s for 5 min. The protein concentration of each purification step was determined using a protein assay (BioRad). The specific PO activity of proteins from each purification step was calculated by dividing the initial rate of dopachrome formation (A492 min −1 ml −1 ) by the protein concentration (mg ml −1 ). PO activity for each preparation was calculated from five replicates. R (version 4.0.3) software package was used for statistical analyses.

Distribution of Hc and proPOβ after UCF precipitation
The first step of the conventional purification method for Hc is UCF. The UCF-precipitated dark blue Hc was dissolved in an equal volume of buffer. Figure 1 shows the distribution of Hc (Fig. 1a) and proPOβ (Fig. 1b) after UCF precipitation. As shown in Fig. 1, both Hc and proPOβ were distributed almost exclusively in the precipitation fraction. The di-PO activity of hemolymph supernatant (crude), UCF supernatant (Sup), and UCF precipitant (Ppt) were assessed using L-DOPA as a substrate (Fig. S1). As shown in Fig. S1, PO activity was almost exclusively detected in the precipitant of UCF (Ppt) that included Hc and proPOβ. The supernatant was transparent and contained virtually no Hc, as determined by SDS-PAGE analysis (Fig. 1).

cDNA sequences and N-terminal amino acid sequences of major Hc subunits
The amino acid sequences deduced from the identified cDNA sequences encoding MjHcY and MjHcL (GenBank accession nos. MZ014374 and MZ014373, respectively) were almost the same as previously reported (GenBank Accession Nos. hemocyanin Y: EF375712; hemocyanin L: EF375711). The N-terminal sequences of MjHcY and MjHcL were NH 2 -DDVQK and NH 2 -DAGGV, respectively, indicating that both Hc subunits possess a signal peptide at the N-terminus. Thus, their deduced molecular weights were 73,216 and 74,906, respectively.

Protein separation ability of SEC
The elution profile of each protein separated with the SEC column is shown in Fig. 2. The deduced molecular weights of the mature region of the two major Hc subunits were 73,216 (HcY) and 74,906 (HcL), while that of proPOβ was 78,773 (GenBank No. AB617654). Under these experimental conditions, both Hc and proPOβ form hexamers according to our previous study (Masuda et al. 2014(Masuda et al. , 2020. These differences in molecular size reflected the retention time of Hc and proPOβ, that is, proPOβ being larger eluted earlier (65.3 min) than Hc (67.8 min) (Fig. 2). According to the elution times in the calibrated SEC separation, the estimated molecular weights of Hc and proPOβ were 407,517 and 483,426, respectively. These values correspond well with the molecular weights of assembled hexamers deduced from Fig. 2 The elution profiles of Hc (solid line) and proPOβ (dashed line). Hc and proPOβ were loaded on a SEC column (Superdex 200 pg) and eluted by the same buffer as the equilibration buffer (10 mM Tris-HCl containing 0.15 M NaCl) at a constant flow-rate of 0.8 ml/min. The absorbance at 280 nm was measured to be an arbitrary unit (A.U.) using a UV detector immediately after elution from the SEC column. For direct comparison of their elution times, the protein concentrations were adjusted before separation amino acid sequences. However, the resolution of the SEC was not sufficient to separate them exclusively.

Source of PO activity
To identify the source of PO activity in the Hc preparations, the gel was stained with Coomassie Brilliant Blue R-250 (CBB) (Fig. 3a) and substrate solution (activity staining) (Fig. 3b) after separation by non-reducing SDS-PAGE. This activity staining can visualize the source of PO activity, because the enzymatic reaction produces the chromogenic product dopachrome (Fig. 3b). The presence of proPOβ was also tested using western blotting (Fig. 3c). When proteins on the gel were stained by CBB after nonreducing SDS-PAGE separation, a band of abundant protein Hc was observed between the 137-and 214-kDa markers (Fig. 3a). On the other hand, the activity staining showed a distinct chromogenic signal on the gel far above Hc's band (lane 6, Fig. 3b). These signals of activity staining correspond to the bands detected in western blotting probed with anti-proPOβ-specific antiserum (Fig. 3c). Hc was most abundant in fraction 7 of SEC after UCF precipitation under both conditions of SDS-PAGE (Fig. 3a, d), whereas the most distinct band of activity staining was observed in fraction 6 ( Fig. 3b). Furthermore, western blot analysis showed that proPOβ was the most abundant in fraction 6 (Fig. 3c, e), which corresponds to PO activity. Images of whole gels and membranes in Fig. 3

Purification of Hc by surface hydrophobicity
As mentioned above, Hc and proPOβ are difficult to separate from each other by the purification method based on molecular size. Accordingly, I attempted another method for the purification of Hc. Since our previous research showed that proPOβ can be precipitated by 40% saturation (1.64 M) of ammonium sulfate (Masuda et al. 2012), Hc was purified from the supernatant with 50% saturation (2.05 M) of ammonium sulfate. Subsequently, the supernatant was further purified via hydrophobic interaction chromatography using butyl Toyopearl resin. Hc was eluted from the hydrophobic chromatography column as a single elution peak by the linear gradient of ammonium sulfate concentration (Fig. 4a). The summary of this purification method is shown in Table 1. The procedure of salting-out [50% saturation (2.05 M) of ammonium sulfate] lost approximately 70% of protein, because a significant amount of Hc was also precipitated with the proPOβ. However, due to the extreme abundance of Hc, more than 100 mg of pure Hc can be obtained by this procedure from 20 ml of hemolymph solution, which contained an equal amount of hemolymph and lanes 2-10, fraction numbers of SEC separation after UCF precipitation. After non-reducing SDS-PAGE on 7.5% acrylamide gel, the gel was stained with CBB (a), 5 mM L-DOPA, and 0.1% SDS (activity staining) (b), and electroblotted onto a PVDF membrane followed by immunodetection with anti-proPOβ antiserum (c). After reducing SDS-PAGE on a 10% acrylamide gel, the gel was stained with CBB (d) and western blotting was carried out with proPOβ antiserum (e)  (Table 1). The yield of this method was 16%, assuming the protein in hemolymph is composed of Hc only. Figure 4b shows the purity of Hc preparations from each purification step. Each Hc preparation was composed of two subunits, HcL (upper band) and HcY (lower band), according to the protein sequence. Estimated from the Coomassie Brilliant Blue R-250 (CBB) stain, the purity seems almost equal among the samples. The di-PO activity of Hc preparations was also assayed (Online Resource, Fig. S1, Figs. 5a, and 6). PO activity was detected in the hemolymph supernatant (crude), precipitant of UCF (UCF), and samples that were further purified by SEC (UCF_SEC) (Fig. 5a). However, the specific activity became weaker during purification of Hc. In contrast, the specific PO activity increased more than 100 times in the final purification step of proPOβ when compared with the activity in the initial crude hemolymph supernatant (Fig. 5b). On the other hand, after the purification by salting-out and hydrophobic chromatography, residual proPOβ was almost undetectable on immunoblotting (Fig. 4c), and PO activity reached a level of reaction without any protein (control) (Fig. 6). These results suggest that the PO activity in Hc_cv prepared from hemolymph of kuruma prawn was mostly derived from proPOβ.

Discussion
The PO activity of Hc was shown in the 1990s for the first time (Zlateva et al. 1996;Decker and Rimke 1998); thereafter, the closely related common properties shared by these proteins were revealed, implying Hc's ability to gain PO activity. The active sites of Hc and PO, which contain binuclear copper atoms, are almost identical (Masuda et al. 2014(Masuda et al. , 2020. A pioneering demonstration of Hc's PO activity was performed using a chelicerate (tarantula) and crustaceans (green shore crab and American lobster) (Decker and Rimke 1998; Zlateva et al. 1996). In these studies, the authors commonly used the dissociated single subunit for the investigation of PO activity in Hc. Since then, the ability of Hc to gain PO activity has been investigated mainly in chelicerates, such as horseshoe crabs and tarantula (Nagai and Kawabata 2000;Decker et al. 2001;Nagai et al. 2001). These studies have also shown the activity of a certain subunit of Hc by dissociating it from an oligomeric form. In contrast, whole oligomeric Hcs have been predominantly evaluated without elaborate purification procedures in crustaceans, except for some detailed studies (Zlateva et al. 1996;Fujieda et al. 2010a, b).
There is no doubt that proPO also exists in the hemolymph of crustaceans (Olianas et al. 2005;Masuda et al. 2012Masuda et al. , 2018, even though its abundance is far lower than that of Hc. Generally, the molecular sizes of a monomer of crustacean Hc and PO are similar (Burmester and Scheller  Fig. 4 a Elution profile of the hydrophobic chromatography. Each fraction was taken every 5 min after starting the elution with a linear gradient. The absorbance at 280 nm of each fraction is shown as a dot, while the molar concentration of ammonium sulfate is indicated by a dashed line (2.05-0 M, secondary y axis). b Purity of Hc in each purification stage and purified proPOβ. Each sample was separated on a 10% SDS polyacrylamide gel followed by CBB staining. Protein concentration was adjusted and protein sample (2.5 μg) was loaded on each lane. Lane 1, hemolymph supernatant of Marsupenaeus japonicus; lane 2, the dissolved precipitation of UCF; lane 3, SEC separation after UCF precipitation (Hc_cv); lane 4, supernatant of 50%-saturated ammonium sulfate; lane 5, after subsequent purification with hydrophobic chromatography (Hc_nv); lane 8, purified proPOβ from M. japonicus. c Western blot analysis of each preparation in b. After SDS-PAGE separation with 10% gel, the proteins were electroblotted onto a PVDF membrane. Proteins on the membrane were probed with anti-proPOβ (from M. japonicus) antiserum. The purified proPOβ used in b was diluted 100 times. Signals were visualized using chemiluminescence. At the same time, the size marker was visualized using a near-infrared laser on the same membrane  Volbeda and Hol 1989;Masuda et al. 2020). Furthermore, the hemolymph-type proPOs form a hexamer (Masuda et al. 2014(Masuda et al. , 2018, which is also a common oligomeric form in crustacean Hc. These observations indicate that Hc and hemolymph-type PO have similar molecular weights in their native forms. In fact, the present results show that proPOβ precipitated with Hc on UCF and they co-existed in the Hc_cv preparation (Figs. 3 and 4). Subsequent SEC did not have enough resolution to separate Hc completely from the residual proPO (Figs. 2 and 3). On the other hand, the purification procedure using hydrophobic chromatography after salting out the residual proPO made the Hc preparation purer (Figs. 4 and 5a). The surface hydrophobicity of these proteins is different from each other, that is, proPOβ has a more hydrophobic surface than Hc, judging from their behavior in ammonium sulfate precipitation. Thus, the difference in surface hydrophobicity of these proteins is enough to separate them from each other. This simple purification method produced Hc without hemolymph-type proPO, which represents a promising avenue for future research on Hc activation.
Crustaceans have both proPO and Hc in their hemolymph plasma, whereas chelicerates have no specific PO molecular species, rendering some Hc subunits bifunctional. Moreover, a certain kind of subunit can be activated in some crustaceans under certain conditions; sophisticated studies have demonstrated that purely prepared and dissociated subunits of crustacean Hc can obtain PO activity (Zlateva et al. 1996;Fujieda et al. 2010b). The closely related structures of coppercontaining active sites of Hc and PO from various origins only allows researchers to imagine the bifunctionality of crustacean Hc. The extreme abundance of Hc would cause detectable PO activity in crustacean hemolymph if it could obtain PO activity, even though the activity would be slight. To elucidate which Hc can obtain PO enzymatic activity and how, removal of residual hemolymph-type proPO is the most critical step. The method for Hc preparation presented here enables this kind of investigation without dissociating the oligomer that is the natural form of Hc in crustacean hemolymph. The mechanisms and conditions through which Hc gains PO activity is currently under investigation by using such Hc preparations. The main source of PO activity in crustaceans remains a significant problem in the field of fisheries science.

Fig. 5
Transition of specific PO activity of Hc (a) and proPOβ (b) during the purification. The PO activity was calculated from the changes in absorbance at 492 nm in the first 1 min (A492 min −1 ml −1 ). Specific activity indicates activity per mg protein. Each purification step in panels (a) and (b) are as follows: crude hemolymph (Crude), the precipitant by UDF (UCF), SEC separation after UCF precipitation (UCF_SEC), the supernatant of hemolymph after 50%-saturation of ammonium sulfate (50%_S) and subsequent butyl Toyopearl separation (Butyl) in panel (a); and crude hemolymph (Crude), salting-out with 20-40%-saturated ammonium sulfate (20-40%), Q-sepharose separation after salting out (Q), and subsequent SEC separation (Q_SEC) in panel (b). The lower (Q1) and upper (Q3) quartiles represent observations outside the 9-91 percentile range. The diagram also shows the median (bold line), mean (crosses), and observed data points (light gray points) for the specific PO activity (n = 5). Data falling outside the Q1-Q3 range are plotted as outliers of the data (dark gray spots)