A novel 43-kDa protein as a negative regulatory component of phenoloxidase-induced melanin synthesis.

The melanization reaction induced by activated phenoloxidase in arthropods is important in the multiple host defense innate immune reactions, leading to the sequestration and killing of invading microorganisms. This reaction ought to be tightly controlled because excessive formation of quinones and systemic hypermelanization are deleterious to the hosts, suggesting that a negative regulator(s) of melanin synthesis may exist in hemolymph. Here, we report the purification and cloning of a cDNA of a novel 43-kDa protein, from the meal-worm Tenebrio molitor, which functions as a melanization-inhibiting protein (MIP). The deduced amino acid sequence of 352 residues has no homology to known sequences in protein data bases. When the concentration of the 43-kDa protein was examined by Western blot analysis in a melanin-induced hemolymph prepared by injection of Candida albicans into T. molitor larvae, the 43-kDa protein specifically decreased in the melanin-induced hemolymph compared with control hemolymph. Recombinant MIP expressed in a baculovirus system had an inhibitory effect on melanin synthesis in vitro. RNA interference using a synthetic 445-mer double-stranded RNA of MIP injected into Tenebrio larvae showed that melanin synthesis was markedly induced. These results suggest that this 43-kDa MIP inhibits the formation of melanin and thus is a modulator of the melanization reaction to prevent the insect from excessive melanin synthesis in places where it should be inappropriate.

The melanization reaction induced by activated phenoloxidase in arthropods is important in the multiple host defense innate immune reactions, leading to the sequestration and killing of invading microorganisms. This reaction ought to be tightly controlled because excessive formation of quinones and systemic hypermelanization are deleterious to the hosts, suggesting that a negative regulator(s) of melanin synthesis may exist in hemolymph. Here, we report the purification and cloning of a cDNA of a novel 43-kDa protein, from the mealworm Tenebrio molitor, which functions as a melanization-inhibiting protein (MIP). The deduced amino acid sequence of 352 residues has no homology to known sequences in protein data bases. When the concentration of the 43-kDa protein was examined by Western blot analysis in a melanin-induced hemolymph prepared by injection of Candida albicans into T. molitor larvae, the 43-kDa protein specifically decreased in the melanininduced hemolymph compared with control hemolymph. Recombinant MIP expressed in a baculovirus system had an inhibitory effect on melanin synthesis in vitro. RNA interference using a synthetic 445-mer double-stranded RNA of MIP injected into Tenebrio larvae showed that melanin synthesis was markedly induced. These results suggest that this 43-kDa MIP inhibits the formation of melanin and thus is a modulator of the melanization reaction to prevent the insect from excessive melanin synthesis in places where it should be inappropriate.
The innate immune system is a host defense mechanism that is evolutionarily conserved from plants to humans and is mainly involved in the recognition and control of the early stage of infection in all animals (1,2). It is activated by a group of germ line-encoded receptors and soluble proteins, conceptu-ally termed pattern recognition receptors and proteins, which recognize microbial surface determinants that are conserved among microbes but absent in the host, such as lipopolysaccharide, peptidoglycan, 1,3-␤-D-glucan, and mannan. Upon recognition, these receptors activate distinct signaling cascades that regulate specific immune-related proteins aimed at the aggressors. Recently, our knowledge of innate immunity in mammals and insects has increased dramatically (1)(2)(3)(4)(5)(6). The recruitment of similar proteins and pathways in both insects and mammals in the fight against infection suggests that they have developed similar mechanisms and molecular pathways to recognize and eliminate pathogenic invaders (4,6).
The prophenoloxidase (pro-PO) 1 system, like the vertebrate complement system, is a proteolytic cascade comprising pattern recognition proteins, serine proteases, their inhibitors, and terminates with the zymogen, pro-PO (7)(8)(9)(10). The pro-PO system is an important non-self-recognition system present in most invertebrates. Microbial polysaccharides lipopolysaccharide, peptidoglycan, and 1,3-␤-D-glucan first react with pattern recognition proteins, which then induce activation of several serine proteases within the pro-PO system. Determining the molecular mechanism, by which pattern recognition molecules differentiate non-self from self and how to transduce signals that stimulate defense responses, is a key for understanding the ways in which innate immune systems are regulated. The pro-PO-activating enzymes or factors, which all are similar to Drosophila easter-type serine protease (clip domain-containing trypsin-like serine protease), cleave pro-PO to generate the active enzyme, phenoloxidase (PO) (11)(12)(13)(14). This enzyme produces toxic compounds to microorganisms by oxidizing phenols to form the melanin pigment, and it also participates in the scleralization of the cuticle, which is vital for the survival of insects (8,9).
Melanin synthesis is essential for defense and development but must be tightly controlled because systemic hyperactivation of the pro-PO system, excessive formation of quinones, and inappropriate excessive melanin synthesis are also deleterious to the hosts, suggesting that the pro-PO activation system and melanin formation are tightly regulated by melanization-regulatory molecules. Only a few inhibitors of the pro-PO system have been identified from insects and crustaceans (15)(16)(17)(18). Previously, it has been shown that Drosophila serpin-27A specifically inhibited the pro-PO-activating enzyme and prevented the melanin synthesis induced by activated PO (19) and that the proteinase inhibitor, pacifastin, efficiently inhibited the pro-PO-activating enzyme in a crayfish (18). However, the specific inhibitor protein for melanin formation induced by activated PO was not found. The identification of a specific melanization-inhibitory protein (MIP) will provide important information to clarify how an arthropod's innate immune reaction such as the melanization response is regulated and controlled.
We reported previously that a 160-kDa vitellogenin-like protein was involved in melanin synthesis as a melanizationenhancing protein (MEP) in the beetle Tenebrio molitor (20). In our preliminary experiments we observed that a 43-kDa protein specifically disappeared from the hemolymph during melanin synthesis. Because the melanization-regulatory proteins usually disappear or are degraded during the melanization reaction, it is not easy to purify these proteins from crude hemolymph. Therefore, to purify a novel melanization-regulatory protein(s) of T. molitor larvae, we assumed that the components of pro-PO system should be in their nonactive form, and hence we prepared a nonactive pro-PO system by treating hemolymph with an irreversible serine protease inhibitor, diisopropyl fluorophosphate (DFP) to prevent activation of pro-PO-activating enzyme(s) (11). By using DFP-treated hemolymph, we isolated a novel 43-kDa protein to homogeneity, cloned its cDNA, and examined its function by using RNA interference (RNAi) experiments in vivo and biochemical analysis in vitro. The recombinant 43-kDa protein was found to decrease the melanization reaction, but not PO activity in vitro. We here report for the first time a novel function of this 43-kDa protein as a negative regulator of the melanization reaction, which is an important innate immune reaction of arthropods.

EXPERIMENTAL PROCEDURES
Animals and Collection of Hemolymph-The methods for raising the insects and collecting the hemolymph were as described previously (19). Briefly, T. molitor larvae (mealworms) were maintained on a laboratory bench in terrarium containing wheat bran. Vegetables were placed on top of the bran to provide water. To harvest the hemolymph, the head of each larva was pricked with a 25-gauge needle. The extruding crude hemolymph from 25 larvae (about 20 l of hemolymph/larva) was collected in 1 ml of a modified anticoagulation buffer (136 mM trisodium citrate, 26 mM citric acid, 20 mM EDTA, and 15 mM NaCl, pH 5.5) and then was centrifuged again at 203,000 ϫ g for 4 h at 4°C. The resulting supernatant was named hemolymph and stored at Ϫ80°C until use. Hemocytes were collected from the extruding crude hemolymph by centrifugation at 200 ϫ g for 10 min at 4°C, washed with anticoagulation buffer, and stored at Ϫ80°C. About 3 ϫ 10 6 packed cells were obtained from 500 l of the crude hemolymph.
Assay of PO Activity and Ability of Melanin Synthesis-An assay of PO and the preparation of G-100 solution were carried out according to our previously described method (20). Briefly, to obtain the solution showing PO activity in the presence of Ca 2ϩ and 1,3-␤-D-glucan, 50 ml of hemolymph was concentrated by ultrafiltration through a membrane filter (YM10, Amicon). Approximately 3 ml of the concentrated solution was applied to a Sephadex G-100 column (1 ϫ 50 cm) equilibrated with anticoagulation buffer and eluted with the same buffer at a flow rate of 12 ml/h. Fractions showing specific PO activity in the presence of Ca 2ϩ and 1,3-␤-D-glucan were pooled and named G-100 solution. This solution was used to examine PO activity and melanin synthesis. To measure PO activity, 30 l of G-100 solution (3.5 mg/ml proteins) was preincubated with 10 l of 1,3-␤-D-glucan (1 g) for 5 min at 30°C, and then 460 l of the substrate solution (1 mM 4-methylcatechol, 2 mM 4-hydroxyproline ethylester in 20 mM Tris-HCl buffer, pH 8.0, containing 10 mM CaCl 2 ) was added to the reaction mixture. After incubation at 30°C for 1 h, the increase in absorbance at 520 nm (A 520 ) was measured using a spectrophotometer (Shimazu). One unit of PO activity was defined as the amount of enzyme causing an increase in absorbance of 0.1 at 520 nm/10-min incubation.
Melanin synthesis was measured according to a method published previously (20). Briefly, 30 l of G-100 solution (3.5 mg/ml proteins) was preincubated with 10 l of 1,3-␤-D-glucan (1 g) at 30°C for 5 min. After incubation, 460 l of the substrate solution (20 mM Tris-HCl, pH 8.0, containing 1 mM dopamine and 10 mM CaCl 2 ) was added to the reaction mixture and then incubated at 30°C for 1 h. The increase in absorbance at 400 nm, which records melanin formation, was measured. To examine the effects of purified native or recombinant MIP on the melanization reaction, the native MIP or recombinant MIP was preincubated with 20 l of hemolymph (100 g of proteins) in the presence or absence of Ca 2ϩ and 1,3-␤-D-glucan at 30°C for 5 min. The melanization reaction was measured after incubation at 30°C for 1 h as described above.
Measurement of Amidase Activity during Pro-PO Activation-To determine amidase activity in the samples, commercially available trypsin substrate (t-butyloxycarbonyl-benzyl-L-phenylalanyl-L-seryl-L-arginine 4-methylcoumaryl-7-amide (Boc-Phe-Ser-Arg-MCA)) was used. In our previous studies, we observed that this substrate was mostly hydrolyzed during insect pro-PO activation in the presence of 1,3-␤-D-glucan and calcium ion (11,21). This substrate was dissolved in dimethylformamide according to the instructions from the manufacturer. 100 l of reaction mixture for measuring PO activity was incubated with 490 l of substrate solution, which contains 40 M substrate in 20 mM Tris-HCl buffer, pH 8.0. After incubation of the mixture in the presence or the absence of MIP at 30°C for 1 h, 500 l of 17% (v/v) acetic acid was added to terminate the enzyme reaction. The specific amidase activity of the eluate solution can be detected by a fluorescence spectrophotometer at ex ϭ 380 nm and (g) em ϭ 460 nm. As a control, 100 l of buffer A was added to check amidase activity as above. One unit of the amidase activity was defined as the amount that liberated 1 nmol of 7-amino-4-methylcoumarin/min. SDS-PAGE Analysis of Hemolymph during Melanin Synthesis-To examine which bands of the hemolymph disappeared during PO-induced melanization, we induced melanin formation as described above. When melanin pigments were generated, the reaction mixture was centrifuged at 28,500 ϫ g at 4°C for 10 min. The proteins in the supernatant were precipitated with trichloroacetic acid and subjected to SDS-PAGE under reducing or nonreducing conditions. To reconfirm whether the disappeared bands on SDS-PAGE were specifically related to melanin synthesis, a well known PO inhibitor, phenylthiourea (PTU), was added to the same reaction mixture under the same conditions. After incubation, the proteins of the supernatant were precipitated with trichloroacetic acid and analyzed on SDS-PAGE under reducing conditions.
Purification of Denatured 43-kDa MIP from Hemolymph-The denatured 43-kDa MIP was purified to homogeneity from polyacrylamide gel using the micro-Electroeluter (Centriluter, Amicon) according to the manufacturer's instructions. To determine the internal amino acid sequences, the purified protein (25 g) was reduced, alkylated, and digested with 2 g of lysylendopeptidase (Wako) at 37°C for 12 h. The resulting peptides were separated through high performance liquid chromatography on a reverse phase C 18 column (Waters) and applied to an amino acid sequencer (Applied Biosystem Procise Automated Gas Phase Sequencer). To determine the amino-terminal sequence, the purified protein (1 g) transferred to a polyvinylidene difluoride filter was applied directly to the amino acid sequencer.
Antibody and Immunoblotting-Antibodies against the 43-kDa protein were raised by injecting 10 g of the denatured 43-kDa protein into a male albino rabbit with complete Freund's adjuvant, and 14 days later two booster injections with the same amount of protein were done (22). The resulting antibody was affinity purified as described previously (23) and used for the purification of the native 43-kDa protein. For immunoblotting, the proteins separated by SDS-PAGE were transferred electrophoretically to a polyvinylidene difluoride membrane, which was blocked by immersion in 5% skimmed milk solution containing 1% horse serum for 12 h. The membrane was then transferred to rinse solution I (20 mM Tris-HCl, pH 7.5, containing 150 mM NaCl, 0.1% Tween 20, and 2.5% skimmed milk) containing the affinity-purified 43-kDa protein antibody (50 ng/ml) and incubated at 4°C for 2 h. The amounts of the bound antibodies were determined using ECL Western blotting reagent kit (Amersham Biosciences).
Detection of MIP-binding Protein in Hemolymph-The proteins of control G-100 solution (20 g of proteins) and the supernatant of melanininduced G-100 solution (20 g) separated on SDS-PAGE were transferred electrophoretically to a nitrocellulose membrane, and the filter was blocked by immersion in 5% skim milk solution containing 1% horse serum for 12 h. The membrane was then incubated with the purified recombinant MIP (5 g of protein) at 4°C for 2 h. The membrane was then transferred to rinse solution I (20 mM Tris-HCl, pH 7.5, containing 150 mM NaCl, 0.1% Tween 20, and 2.5% skimmed milk) containing the affinity-purified 43-kDa protein antibody (50 ng/ml) and incubated at 4°C for 2 h. The amounts of the bound antibodies were determined using ECL Western blotting reagent kit (Amersham Biosciences).
cDNA Cloning and Nucleotide Sequencing of 43-kDa MIP-A cDNA library from T. molitor larvae was constructed as described previously (24). Among chemically determined four partial amino acid sequences of the purified 43-kDa protein, DNA oligonucleotide corresponding to YE-ANDDMI was synthesized as follows: 5Ј-TA(T/C)GA(A/G)GCIAA(T/ C)GA(T/C)GA(T/C)ATGAT-3Ј, and it was labeled with [␥-32 P]-ATP by a previously described method (25). For the initial screening, ϳ50,000 recombinants of T. molitor larvae cDNA library were used and the membranes were prehybridized at 65°C for 1 h in 4 ϫSSC (600 mM NaCl, 60 mM trisodium citrate, pH 7.0), 10 ϫ Denhardt's solution (50 ϫ Denhardt's solution: 1% (w/v) bovine serum albumin, 1% (w/v) Ficoll, and 1% (w/v) polyvinylpyrrolidone) and 25 g/ml salmon sperm DNA. The membranes were then hybridized at 51.5°C for 12 h in the same solution during prehybridization. We obtained 10 hybridization-positive clones and analyzed two plasmids containing four chemically determined partial amino acid sequences and amino-terminal sequence. The deduced amino acid sequence of the 43-kDa protein was compared with the protein sequence data base of the National Center for Biotechnology Information (NCBI) using the Genetyx system (Software Development Co., Ltd., Tokyo).
Purification of the Native 43-kDa MIP (nMIP) from Tenebrio Hemolymph-To prevent the activation of Tenebrio pro-PO-activating enzyme(s), 600 l of 0.5 M DFP was added to 150 ml of hemolymph (2 mM DFP as a final concentration), and the mixture was incubated at 4°C for 2 h, and then the hemolymph was dialyzed at 4°C for 12 h against 5 liters of buffer A (50 mM Tris-HCl, 3 mM EDTA, pH 6.0). After centrifugation at 3,000 rpm at 4°C for 10 min to remove the precipitated proteins, the supernatant was loaded onto a Sepharose CL-4B column (2 ϫ 10 cm) preequilibrated with buffer A at a flow rate of 0.5 ml/min. The column was washed with the same buffer until the A 280 nm was 0 and then eluted with 1 M NaCl in buffer A. Fractions containing the 43-kDa protein were confirmed by Western blotting analysis and collected. The pooled fractions were concentrated by ultrafiltration through a filter and Centricon (Amicon) to 1 ml and then loaded to a Sephacryl S-200 column (2 ϫ 120 cm) preequilibrated with buffer B (50 mM Tris-HCl, 2 mM EDTA, 150 mM NaCl, pH 6.0) at a flow rate of 0.2 ml/min at 4°C. Fractions containing the 43-kDa protein were pooled and diluted 10 times with buffer C (50 mM Tris-HCl, 2 mM EDTA, pH 7.5) and then loaded onto a Mono Q FPLC column (Amersham Biosciences) preequilibrated with buffer C at a flow rate of 0.4 ml/min. After washing until the A 280 nm was 0, the column was eluted with a linear gradient of NaCl from 0 to 0.75 M over 90 min. Fractions containing the 43-kDa protein from the Mono Q FPLC column were pooled and diluted with buffer C and then reloaded to the Mono Q FPLC column once again. The column was eluted with a linear gradient of NaCl from 0.25 to 0.75 M over 120 min. Finally, fractions containing the purified 43-kDa protein were pooled and concentrated to 0.1 mg/ml by Centricon. The identity and purity of the native 43-kDa protein were confirmed both by the determination of amino-terminal sequence and by the analysis of SDS-PAGE under reducing and nonreducing conditions.
Expression and Purification of Recombinant MIP (rMIP)-The construction of the recombinant baculovirus vector and the expression of the rMIP were performed according to manufacturer's instructions (Invitrogen). The cDNA encoding the mature MIP was subcloned into pFastBacDUAL vector (Invitrogen) using BamHI and HindIII enzyme sites. The recombinant virus for MIP expression was generated according to the manufacturer's instruction manual (Bac-to-Bac Baculovirus expression system, Invitrogen). The recombinant virus was amplified using Spodoptera frugiperda 9 (Sf9, Invitrogen) cells in SF-900II serum-free medium (Invitrogen) at 27°C. To produce the protein, Trichoplusia ni BTI-TN-5B1-4 (High-five, Invitrogen) cells were grown at 27°C in 1 liter of Grace medium (Invitrogen) supplemented with 20% Sf-900II serum-free medium and 0.1% Pluonic F-68 reagent (Invitrogen). The cells were infected with a cell density of 1 ϫ 10 6 cells/ml at a multiplicity of infection of 10 and were incubated for 5 days.
For purification of rMIP from Hi-5 cells medium, 800 ml of medium was first concentrated into ϳ100 ml by ultrafiltration through a membrane filter (Amicon, YM10) and then dialyzed against 10 liters of buffer D (20 mM Tris-HCl, 150 mM NaCl, pH 8.0) at 4°C for at least 18 h. After centrifugation at 3,000 ϫ g at 4°C for 5 min to remove precipitates, the supernatant was mixed with 2 ml of precharged nickel-nitrilotriacetic acid-agarose (Qiagen) preequilibrated with buffer D and gently rotated at 4°C for 2 h. The unbound proteins were allowed to pass through the resin, and the resin was washed with 50 ml of buffer D containing 20 mM imidazole. The proteins bound to the resin were eluted with 200 mM imidazole in buffer D and then analyzed by SDS-PAGE. The fractions containing rMIP were pooled and applied to a Mono Q FPLC column preequilibrated with buffer E (20 mM Tris-HCl, pH 8.0). After washing with buffer E until the A 280 nm was 0, the column was eluted with a linear gradient of NaCl from 0 to 0.9 M over 120 min, and the fractions containing rMIP were pooled and concentrated with Centricon to a final concentration of 0.1 mg/ml. The amino-terminal amino acid sequence analysis was carried out to verify the sequence of rMIP.
RNAi Experiments-The template (spanning from nucleotide 502 to 946) for synthesis of double-stranded RNA (dsRNA) for MIP was amplified from cDNA by PCR using the forward primer TAATACGACTCAC-TATAGGCTGATTTAGGCATTCCTGGC and the reverse primer TAAT-ACGACTCACTATAAAGAAGAGGGGTTGACCAAGACTC. The dsRNA was synthesized using the T7 MEGAscript kit (Ambion) according to the manufacturer's instructions. The resulting RNA was extracted with phenol/chloroform and ethanol precipitated and finally dissolved in diethyl pyrocarbonate-treated water. Annealing of the complementary strands was performed by heating to 70°C for 10 min and cooling overnight in water bath at room temperature. The template for synthesis of dsRNA for green fluorescent protein (dsGFP) was the PD2EGFP-1 plasmid (Clontech) was amplified by PCR using the forward primers TAATACGACT-CACTATAGGGCGACGTAAACGGCCACAAGT and the reverse primer TAATACGACTCACTATAGGGTTCTTGTACAGCTCGTCCATGC, and dsRNA were then synthesized as described above. Tenebrio larvae were injected with 50 l of dsRNA in 50 ng/l phosphate-buffered saline using a 25-gauge needle. After the injection the larvae were kept at room temperature for 24 h.
Reverse Transcription-PCR and Western Blot Analysis of dsRNAtreated Larvae-To detect MIP transcript, total RNA was isolated from the fat body of dsRNA-injected T. molitor larvae using the TRIzol LS regent (Invitrogen) according to the manufacturer's instructions. Contaminating genomic DNA was removed by treatment with RNase-free DNase I (1 unit/l), and the reaction was stopped by the addition of 20 mM EDTA, pH 8.0. After purification by phenol/chloroform extraction and ethanol precipitation the RNA (400 ng) was reverse-transcribed using Thermoscript (Invitrogen) according to the manufacturer's instructions. The cDNA was analyzed by PCR using the following primers: MIP (spanning from nucleotides 101 to 428) forward, CGTCGTC-ACTCCTCTCCAAAAG; reverse, TCGTCATCATCGTCGTCGTCGTAG. For controls (ribosomal protein L27a, accession number X99204) forward, GCATGGCAAACACAGAAAGCATC; and reverse, ATGACAGG-TTGGTTAGGCAGGC. The PCR products were resolved on 1.2% agarose gel electrophoresis. For Western blotting analyses 40 g of hemolymph proteins from MIP-dsRNA-treated larvae or GFP-dsRNAtreated larvae were transferred onto polyvinylidene difluoride membranes, and the immunoblotting was carried out as described above.
Determination of MIP Localization-To determine the localization of the purified 43-kDa protein, we prepared plasma, extracts of hemocyte, and fat bodies as described previously (23). The hemocyte and plasma were obtained as described above. The precipitated hemocytes (3 ϫ 10 8 cells) were washed with 500 l of buffer A, suspended with 500 l of buffer A, sonicated for 15 s at 4°C, and then centrifuged at 22,000 ϫ g at 4°C for 10 min. The supernatant was used as a source of hemocyte lysate. The soluble proteins of hemocyte lysate and fat bodies were precipitated with trichloroacetic acid, subjected to SDS-PAGE, and then immunoblotting with the affinity-purified 43-kDa protein antibody was carried out.

Characterization of Proteins Associated with Melanization-
To identify melanin-regulatory proteins in insect hemolymph, we first prepared G-100 solution from Tenebrio hemolymph. This solution showed PO activity by incubation with 1,3-␤-D-glucan and Ca 2ϩ . Melanin pigments were induced when the G-100 solution was incubated with dopamine in the presence of 1,3-␤-D-glucan and Ca 2ϩ . After induction of melanization, the supernatants of the reaction mixtures were analyzed by SDS-PAGE under reducing conditions. As shown in Fig. 1A, six proteins clearly disappeared after a 30-min incubation (lane 8) compared with the G-100 solution (lane 4), indicating that these proteins are associated with melanin synthesis. It is known that melanins induced by arthropod PO are closely linked to protein matrices in the form of melanoproteins (7)(8)(9). We assumed that if some proteins in hemolymph were used in melanin synthesis, they would disappear as positive regulatory proteins such as MEPs during the melanization reaction. If some proteins functioned as negative regulatory proteins such as MIPs, they should be generated or degraded from the hemolymph when the melanization reaction was started.
To characterize further these six disappearing bands, we determined their amino-terminal sequences. Bands 1 and 2 showed the same sequences as FNTLSPWDKEVIYNW, which perfectly matched with the Tenebrio vitellogenin-like protein that was reported as a MEP (20). The amino-terminal sequence of band 3 was blocked. The sequence of band 4 was determined as SDKVVCYYNSKSHFRE, which showed high homology (64%) with that of silkworm imaginal disc growth factor-like protein (26) or human chitinase 3-like protein (27). It was reported that silkworm imaginal disc growth factor-like protein is known as ecdysteroid-inducible genes in the programmed cell death during insect metamorphosis (26). A partial amino acid sequence of band 5 was identified as DAEESAEPTSGE-VIS, which showed no sequence homology with known proteins in the data base, and the function of this protein is not determined yet. The amino-terminal sequence of band 6 was determined as SDSDEVLEKVKSKHL, which perfectly matched with the Tenebrio desiccation protein that was reported to be engaged in developmental and environmental regulation of thermal hysteresis (28).
To examine how these six proteins were engaged in the melanization reaction, we carried out melanin synthesis experiments in the presence of PTU, a well known PO inhibitor, so that melanin formation is blocked. As expected, all bands did not disappear in the presence of PTU even after a 1-h incubation (lane 8 in Fig. 1B), indicating that six proteins are involved in the melanization reaction induced by PO activity. Among them, we extracted the 43 kDa band by electroelution from the polyacrylamide gels using a micro-Electroeluter and purified the 43 kDa band to homogeneity (lane 2 in Fig. 1C). The yield of the denatured 43-kDa protein was 20 g from 40 mg of the crude hemolymph.
Cloning and Characterization of the 43-kDa Band-We determined four internal amino acid sequences of the 43-kDa protein as follows: EMFEPLADLGIPGSVTAEEFN, KGELETVIYLDKDELYG, VQPHLLGTLSDYYEANDDMI-LSG, and WRVLVNPSSYYV. To determine the whole amino acid sequence of the 43-kDa protein, we synthesized a degenerate oligonucleotide probe for YEANDDMI and screened a cDNA library of T. molitor larvae, and then obtained 15 positive clones. The nucleotide sequence of the longest insert and the deduced amino acid sequence encoded in the open reading frame of this clone are shown in Fig. 2. The open reading frame encodes a protein consisting of 352 amino acid residues. The five peptide sequences derived from the 43-kDa protein are all found within the complete sequence, indicating that the cDNA encodes for the 43-kDa protein. The deduced amino acid sequence of the 43-kDa protein has no significant homology to other proteins reported so far, but it contains an Asp-rich region, including 11 contiguous Asp residues, in the central part of the 43-kDa protein, which shows a high degree of homology to several unrelated proteins with different functions. As shown in Fig. 2B, the Asp-rich sequence of the 43-kDa protein shows high homology with the RNA polymerase II subunit 5 (RPB5)-mediating protein, which is an RPB5interacting protein and which also counteracts transactivation

FIG. 1. 12% SDS-PAGE analysis of G-100 solution after induction of melanin synthesis (A) and the effects of PTU on melanin synthesis (B) and the purified denatured 43-kDa protein (C).
A, the melanin synthesis reaction was induced as described under "Experimental Procedures." After incubation, the supernatants of reaction mixtures were precipitated with trichloroacetic acid and analyzed in 12% SDS-PAGE under reducing conditions. Lanes 1, 2, 3, and 4  by hepatitis B virus X protein (29). This Asp-rich motif is also present in another protein such as Xenopus laevis RNA polymerase 1 transcript factor (Xl-RPTF, 30) or Mus musculus upstream binding factor (Mm-UBF), which is involved in the proliferation and differentiation of murine myeloid cells (31), and Drosophila melanogaster Gp150 protein (Dm-Gp150), which is related to early ommatidial development through modulation of Notch signaling (32). Also, the 11-Asp stretch is encoded by two different stretches of the triplets GAC and GAT, indicating that the Asp-rich region seems to avoid destabilizing the DNA structure by a long triplet repeat (33).
To ascertain the 43-kDa protein as a melanization-related protein, we examined the changes in concentration of the 43-kDa protein in vivo during melanin synthesis. As shown in Fig.  3C, Tenebrio larvae injected with Candida albicans cells (5 ϫ 10 7 cells) turned black after 24 h, whereas no change of color occurred in control larvae (Fig. 3A). When the concentration of the 43-kDa protein was examined by Western blot in the melanin-induced hemolymph by the injection of C. albicans, we found that the 43-kDa protein decreased but not in the control larvae (Fig. 3D), suggesting that the 43-kDa may be related with melanin synthesis. However, we could not determine whether the 43-kDa protein is a MEP or MIP.
Both Purified Native and Recombinant 43-kDa Proteins Inhibited Melanin Synthesis in Vitro-Because the 43-kDa protein disappeared during melanin synthesis, the Tenebrio pro-PO system must be in its inactive form to allow purification of the 43-kDa protein from hemolymph. We reported previously that insect pro-PO-activating factors involved in the pro-PO  (29), and Drosophila Gp150 (30). Gaps have been introduced for proper arrangement of the conserved Asp residues. system were strongly inhibited by DFP (10). As expected, the DFP-treated hemolymph showed very low PO activity compared with the crude hemolymph, indicating that Tenebrio pro-PO-activating factors were inactivated by DFP treatment (data not shown). To gain insights into the function of the 43-kDa protein in melanin synthesis induced by dopamine, we purified native and recombinant 43-kDa protein to homogeneity (lane 2 in Fig. 4A and lane 2 in Fig. 4B). To address whether the purified native or recombinant 43-kDa protein functions as MEP or MIP in vitro, we added the purified 43-kDa recombinant protein to the crude hemolymph solution and examined the effects on melanin synthesis using dopamine as substrate. As shown in Fig. 5A, the recombinant 43-kDa protein had an inhibitory effect on melanin synthesis in a dose-dependent manner. Under the same conditions, when we checked the effects of PO activity by addition of recombinant 43-kDa protein to the hemolymph solution, PO activity induced by synthetic substrates 4-methylcatechol and 4-hydroxyproline ethylester as PO substrates was not affected (Fig. 5B).
Previously it was reported that insect pro-PO system induced the activation of serine protease zymogen to active serine protease during 1,3-␤-D-glucan-dependent pro-PO activation (11,21). To explore the relationship between serine protease activity and PO activity in the presence of MIP, we compared the amidase activity in the presence and in the absence of MIP with 1,3-␤-D-glucan and Ca 2ϩ by using commercially available trypsin substrate, Boc-Phe-Ser-Arg-MCA. As shown in Fig. 5C, when Ca 2ϩ ion and 1,3-␤-D-glucan were added to G-100 solution (column 2), amidase activity increased compared with the G-100 solution alone (column 1). By addition of MIP to the G-100 solution in the presence of Ca 2ϩ and 1,3-␤-D-glucan, amidase activity was not changed even though increased amounts of MIP (columns 3, 4, and 5). These results suggest that MIP does not inhibit the amidase activity of any yet unidentified pro-PO-activating serine protease(s). This indicates that MIP is not an inhibitor of the pro-PO-activating cascade itself, but rather that MIP inhibits the formation of melanin from PO-oxidized phenol substrates.
To characterize further the biochemical properties of MIP during melanin synthesis, we tried to find the MIP-binding protein(s) from G-100 solution and melanin-induced G-100 solution by Western blotting analysis. As shown in Fig. 5D, ϳ47-kDa protein of G-100 solution but not melanin-induced G-100 solution was specifically recognized by the MIP antibody, indicating that MIP might bind to this 47-kDa protein in G-100 solution. When melanin synthesis induced by the activated PO has been started, MIP might be dissociated from the 47-kDa protein and then degraded from hemolymph.

RNAi Experiments Reveal That the 43-kDa Protein Functions as a Negative Regulatory Protein during Melanin Synthesis in
Vivo-We also performed in vivo gene inactivation experiments to confirm further the function of MIP during melanin synthesis. dsRNA of 445 bp of MIP was found to silence MIP mRNA expression efficiently (lanes 1-3 in Fig. 6A), whereas control GFP dsRNA was without effect (lanes 4-6). Under the same conditions, MIP protein expression also decreased in the dsRNA-MIP-treated larvae (lanes 1-3 in Fig. 6B) but not in the dsRNA-GFP-treated larvae (lanes 4-6), indicating that dsRNA of MIP silenced the mRNA and thus also decreased the expression of the MIP protein. Finally the melanin synthesis abilities of MIP-dsRNA-injected larvae and control GFP-dsRNA-injected larvae were compared. Interestingly, the hemolymph of MIP-dsRNA-treated larvae showed a strong melanin-synthesizing activity (Fig. 6C, upper panel), whereas in the control silenced animals the melanin synthesis was not induced (Fig.  6C, lower panel). These results clearly demonstrate that melanin synthesis was induced when the mip gene was silenced, indicating that MIP functions as a negative regulatory protein during melanin synthesis, compatible with the biochemical studies above.
The 43-kDa MIP Is Localized in Plasma-As shown in Fig. 7, no 43-kDa MIP was detected in the fat body or hemocyte lysate (lanes 3 and 4), but a significant amount of the 43-kDa protein was detected in the hemolymph and plasma (lanes 2 and 5). This result indicates that the purified 43-kDa MIP is localized in plasma. DISCUSSION We report here the purification, molecular cloning, and biochemical characterization of a novel 43-kDa protein that was found to be a negative regulator of the melanization reaction. This conclusion was made after obtaining results from these two experiments: 1) the recombinant MIP specifically inhibits melanin synthesis in vitro; 2) the hemolymph of larvae in which the mip gene was silenced by RNAi induces a strong melanization reaction in vivo. Although there are several proteins that inhibit the enzyme activity of PO or inhibit the activation of pro-PO by making a complex with pro-PO-activating enzymes (15)(16)(17)(18)(19), this is the first report of a novel 43-kDa protein that functions as a negative regulator of the melanization reaction.
To maintain defense to microbial invasion and proper development of T. molitor larvae, positive and negative regulators that control melanin synthesis are essential components for their survival. It is well known that activated PO can generate reactive quinone compounds and free radicals and subsequently induce the melanization reaction by nonspecific crosslinking of hemolymph proteins such as MEPs (7-9, 20, 34). A plausible function of MIP in naive larvae is that MIP may function as a scavenger-like compound against reactive quinone products and free radicals and then prevent the diffusion of these harmful substances to the host. This hypothesis is supported by our in vivo RNAi studies. When we silenced the mip gene by RNAi, we observed a strong melanin formation, suggesting that if MIP is not available to trap the melanininducing substances, such as excessive quinones or reactive radicals, this will lead to unlimited melanin synthesis. Further studies are needed to elucidate the molecular mechanism of 43-kDa MIP in the melanization reaction.
To date, several melanization-inhibitory molecules have been isolated and characterized from several animals, and these molecules have been shown to be inhibitors of melanin synthesis induced by the activated PO of invertebrates or by the tyrosinase of vertebrates. Tyrosinase is a copper-containing enzyme widely distributed in nature which catalyzes two key reactions in the melanin biosynthetic pathway (35). A search of inhibitors for tyrosinase is one of the major strategies in developing new whitening agents. Many tyrosinase inhibitors are polyphenol derivatives of flavonoids or of trans-stilbene (t-stilbene), such as resveratrol and its derivatives, which have been investigated intensively (36,37). They are usually constructed from one of two distinct substructures, which dictate their mechanism of tyrosinase inhibition: containing either a 4-substituted resorcinol moiety or catechol. It was suggested that 4-substituted resorcinol-type inhibitors are potent in tyrosinase inhibition and bind to the enzyme binuclear active site (36). The catechol structure, with two OH groups at o-positions, may behave as a chelator to the copper ions in the tyrosinase (38). However, no report has shown that a protein, such as the 43-kDa MIP described in this study, has an ability to inhibit the melanization reaction step.
From invertebrates, an endogenous PO inhibitor (POI) containing a dopa-containing 38-mer peptide from Musca domestica was purified and characterized (15). It was suggested that this inhibitor, POI, plays an important role in smoothing the way of emergence of the adult through hindering excessive melanization, as well as hardening, of cuticular proteins under the epicuticle. However, the amino acid sequence of 43-kDa MIP did not show any homology with that of POI. Another negative regulator of the pro-PO cascade in Drosophila is the blood serine protease inhibitor serpin-27A (17,19). Serpin-27A has been shown to regulate the melanization cascade through the specific inhibition of the terminal protease of the pro-POactivating cascade, which also was shown for the pacifastin proteinase inhibitor in a crustacean (18). Also, it was reported that the melanization reaction requires Toll pathway activation and depends on the removal of the Drosophila serpin-27A (17). In addition, the Kanost group (16) reported that serpin-3, one of the serpin variants, is a physiological regulator of the Manduca sexta pro-PO activation reaction and makes a complex with purified pro-PO-activating proteinases in the hemolymph. However, there is no report that a protein functions as a negative regulator during melanin formation.
Recently we showed that several pro-PO-activating enzymes and other factors are necessary to activate pro-PO by in vitro reconstitution experiments in coleopteran insects (39). Based on these results, the pro-PO cascade can be divided into three steps: 1) recognition with pattern recognition proteins, 2) activation consisting of pro-PO-activating enzymes and their adaptor proteins, and 3) finally melanin synthesis induced by PO, MEP, MIP, and unidentified regulatory protein(s). Even though several pattern recognition proteins for 1,3-␤-D-glucan and peptidoglycan have been identified (21,40) from our insects, the molecular mechanism of the recognition part is poorly understood. For instance, we do not know how lipopolysaccharide, peptidoglycan, 1,3-␤-D-glucan recognition proteins can induce activation of the pro-PO-activating enzymes or other factor(s). The activation part of the pro-PO system is quite complex and involves limited proteolysis of several factors and pro-POs. The cross-talks between proteins in this step were partially determined by using the purified factors and pro-POs (39), but the molecular mechanisms and structural basis of the melanization reaction still remain to be clarified. Here, we show the existence of a novel negative regulator of melanization with a previously reported positive regulator protein. The continued and ongoing analysis of the relationship among MIP, MEP, and PO will eventually offer clues to the function of melanogenesis as an innate immune reaction in arthropods.