Báo cáo Y học: Function and cellular localization of farnesoic acid O -methyltransferase (FAMeT) in the shrimp, Metapenaeus ensis

doi:10.1046/j.1432-1033.2002.03048.x

Eur. J. Biochem. 269, 3587–3595 (2002) (cid:1) FEBS 2002

Function and cellular localization of farnesoic acid O-methyltransferase (FAMeT) in the shrimp, Metapenaeusensis

Y. I. N. Silva Gunawardene1, S. S. Tobe2, W. G. Bendena3, B. K. C. Chow1, K. J. Yagi2 and S.-M. Chan1 1Department of Zoology, The University of Hong Kong, Pokfulam Road, Hong Kong, China; 2Department of Zoology, University of Toronto, Ontario, Canada; 3Department of Biology, Queen’s University, Kingston, Ontario, Canada

eyestalk and the ventral nerve cord. To show that our cloned gene product had FAMeT activity, we demon- strated that expressed rFAMeT gene product catalyzed the conversion of FA to MF in a radiochemical assay. The ubiquitous distribution of FAMeT suggests that this enzyme is involved in physiological processes in addition to gametogenesis, oocyte maturation and development and metamorphosis of the shrimp. We hypothesize that FAMeT directly or indirectly (through MF) modulates the reproduction and growth of crustaceans by interacting with the eyestalk neuropeptides as a consequence of its presence in the neurosecretory cells of the X-organ-sinus gland.

farnesoic acid O-methyltransferase; Keywords: shrimp; methyl farnesoate; neuropeptides; juvenile hormone.

The isoprenoid methyl farnesoate (MF) has been impli- cated in the regulation of crustacean development and reproduction in conjunction with eyestalk molt inhibiting hormones and ecdysteroids. Farnesoic acid O-methyl- transferase (FAMeT) catalyzes the methylation of farne- soic acid (FA) to produce MF in the terminal step of MF synthesis. We have previously cloned and characterized the shrimp FAMeT. In the present study, recombinant FAMeT (rFAMeT) was produced for bioassay and anti- serum generation. FAMeT is widely distributed in shrimp tissues with the highest concentration observed in the ventral nerve cord. Interestingly, an additional larger protein in the eyestalk also showed immunoreactivity to anti-FAMeT serum. FAMeT was localized in the neuro- secretory cells of the X-organ-sinus gland complex of the eyestalk. As shown by RT-PCR, FAMeT mRNA is constitutively expressed throughout the molt cycle in the

Methyl farnesoate (MF), the sesquiterpenoid precursor of the insect juvenile hormone III (JH III), is produced and released by the mandibular organs of decapod crustaceans [1]. The physiological function for MF is not well understood in crustaceans. However, by analogy to the established functions of JH III in insects, MF has been suggested to play an important role in the regulation of growth and reproduction in crustaceans [2–4]. In some crustaceans, circulating titer and biosynthesis of MF appear to be positively correlated with the maturation of the ovary [5,6]. MF has also been suggested to play a role in delaying the onset of molting in larval crustaceans [7,8]. This evidence implicates MF in both crustacean growth and reproduction. Farnesoic acid O-methyltransferase

(FAMeT; also known as S-adenosyl-methionine:farnesoic acid O-methyl- transferase) is the enzyme that catalyses the final step in

Correspondence to S. M. Chan, Department of Zoology, The University of Hong Kong, Pokfulam Road, Hong Kong, China. Fax: + 852 2299 0864, E-mail: chansm@hkucc.hku.hk Abbreviations: MF, methyl farnesoate; FAMeT, Farnesoic acid O-methyltransferase; FA, farnesoic acid; JH III, insect juvenile hormone III; IPTG, isopropyl thio-b-D-galactoside; CHH, crustacean hyperglycemic hormone; MIH, molt inhibiting hormone; GIH, gonad inhibiting hormone. (Received 7 March 2002, revised 4 June 2002, accepted 17 June 2002)

the

the MF biosynthetic pathway in crustaceans [9,10]. FAMeT, also present in the insect corpora allata, carries out the methylation of farnesoic acid (FA) to yield MF using the cofactor S-adenosyl-L-methionine [9,11]. Varia- tions in activity of the O-methyltransferase during devel- opment appear to be an important component in the regulation of JH biosynthesis in insects [10,12]. The rate- limiting step controlling JH III biosynthesis in the cock- roach larva is catalyzed by FAMeT [13]. Neuropeptides released from the eyestalk X-organ sinus gland complex appear to modulate the MF synthesis in the crustaceans similar to the allatostatins repressing JH biosynthesis by the corpora allata in insects [14–16]. This negative regulation of MF biosynthesis occurs in part through the inhibition of FAMeT activity [9,17]. Furthermore, studies in insects have also demonstrated that allatostatins are able to suppress the production of JH by modulating O-methyltransferase and/or epoxidase activity [10]. Thus, in both insects and crustaceans, O-methyltransferase plays a role in the regulation of MF synthesis and thereby mediates an effect on vitellogenesis and metamorphosis of these animals through MF. In the shrimp, Metapenaeus ensis, FAMeT mRNA is expressed throughout ovarian maturation in the nerve and eyestalk suggesting a possible role of FAMeT in the regulation of reproduction [18]. Coincidentally, the high levels of FAMeT mRNA tran- scripts and protein in the eyestalk and the ventral nerve cord in shrimp parallels the expression of some neuro- peptides of crustacean hyperglycemic hormone (CHH)/gonad inhibiting hormone (GIH)/ molt inhibiting hormone (MIH) family [18,19].

3588 Y. I. N. Silva Gunawardene et al. (Eur. J. Biochem. 269)

(cid:1) FEBS 2002

binding buffer (8 M urea in binding buffer) and agitated for 2 h at room temperature. Following centrifugation as above, the supernatant was collected and loaded onto a Ni-nitrilotriacetic acid–agarose (Qiagen, Hilden, Germany) affinity column (pre-equilibrated with denatured binding buffer). Denatured binding buffer was used to wash the column until zero absorbency (280 nm) was observed in the eluate. The fusion protein was eluted with an elution buffer (6 mM Tris/HCl, pH 7.9, 0.15 M NaCl and 300 mM imidazole, 8 M urea) and dialyzed in a buffer (0.01% SDS, 0.1% Tween 20, 0.1 · NaCl/Tris (0.8 g NaCl, 0.02 g KCl and 0.3 g Tris base in 1 L of water, pH 7.4) at 4 (cid:3)C overnight. A second and a third dialysis were performed for 10 h in 0.1 · NaCl/Tris and 0.1 · NaCl/Pi (0.8 g NaCl, 0.02 g KCl, 0.144 g Na2HPO4 and 0.024 g KH2O4 in 1 L of water, pH 7.4), respectively. A protein assay kit (Bio-Rad) was used to determine the protein concentration.

To understand the role of MF in the growth and reproduction of crustaceans, considerable effort has been invested in studying the regulation of the biosynthesis of MF [1,4,6]. Although FAMeT has been implicated as a potential point in the regulation of MF biosynthesis, the biochemical and biological properties of the gene encoding the FAMeT have not been characterized. Thus, the study of the enzymes of the MF biosynthetic pathway will enable us to better understand the role of MF in shrimp as well as its probable structural and functional relationship with the crustacean neuropeptides. We have previously reported on the expression, cloning and characterization of the cDNA encoding the FAMeT in the shrimp Metapenaeus ensis [18]. In the lobster, putative FAMeT clones have also been isolated (GenBank no. U25846 and GenBank no. AF249871) but the function of these genes has not yet been demonstrated. Database searches also suggest that a homologue for this gene exists in the Drosophila genome [18]. Although there is a report on the cytosolic nature of O-methyltransferase in the locust corpora allata, the cellular localization/distribution of the enzymes of the MF biosyn- thetic pathway remain unidentified [11]. In, this study, we expressed recombinant (r) FAMeT in bacteria and deter- mined its biological function by a radiochemical assay. Antiserum raised against rFAMeT was used to localize native FAMeT at the cellular level.

E X P E R I M E N T A L P R O C E D U R E S

Construction of the shrimp FAMeT expression plasmid

Production of anti-rFAMeT serum Purified rFAMeT (50 mgÆmL)1 in NaCl/Pi) was mixed with equal volume of complete Freund’s adjuvant (Gibco) by homogenization with a Polytron. The emulsion was injected into a New Zealand white rabbit subcutaneously at three different sites. During the injections and subse- quent handling of the animal, principles of laboratory animal care and specific national laws were followed. The rFAMeT used for the second and third booster injections were mixed with an equal volume of incomplete Freund’s adjuvant. The second and third injections (100 lg) were given at 10-day intervals. A week after the third injection, the antibody titer was monitored by SDS/PAGE and Western blot analysis. The anti-rFAMeT serum was prepared from the rabbit whole blood by centrifugation at 800 g for 15 min and the supernatant was stored at )80 (cid:3)C in aliquots.

Detection of FAMeT in shrimp tissue protein extracts

The DNA fragment encoding the FAMeT coding sequence (nucleotides 75–917; GenBank accession no. AF 333042) was amplified using the primers FAOMX-F4 (forward: 5¢-C CGGGATCCATGGCTGACAACTGGCCTGCC-3¢) and FAOMX-R6 (reverse: 5¢-CGGGGTACCTTAGAATTCG AACTTCCACTT-3¢) by PCR. Following amplification the cDNA fragment was digested with BamHI and KpnI and ligated into the BamHI, KpnI-digested vector pREST-A. (Invitrogen, Groningen, the Netherlands). The ligated product (pREST-A/rFAMeT) was transformed into E. coli XL1-blue for DNA sequence determination. The pREST-A/ rFAMeT construct was transformed into BL21(DE3) cells for protein expression. Overnight culture of a single colony was diluted (1 : 200) with Luria–Bertani medium and incubated at 37 (cid:3)C with vigorous shaking to an D600 of 0.3–0.5. IPTG was added at a final concentration of 1 mM to induce the protein expression for a period of 4 h. The protein expression and cell growth was monitored by taking 1 mL of the bacterial culture at 1-h intervals. The protein extract was analyzed by 12% SDS/PAGE and Western blot analysis using Ni-nitrilotriacetic acid–alkaline phosphatase conjugate or mouse anti-histidine IgG as the first antibody and a goat anti-mouse IgG–alkaline phosphatase conjugate as the secondary antibody.

Purification of rFAMeT

serum (1 : 1000) as

Bacteria were pelleted by centrifugation and resuspended (20 mM Tris/HCl pH 7.9, in 15 mL binding buffer 0.5 mM NaCl and 5 mM imidazole). The cells were homogenized with a Polytron and centrifuged at 5700 g for 15 min. The pellet was suspended in 15 mL denatured Shrimp (8–10 g) were purchased from a local sea food market and immediately transferred to sea water aquaria held at ambient photoperiod and temperature. The molt the animals was performed by pleopod staging of setogenesis [20] and the female reproductive stage was determined by the gonadal-somatic index. Different tissues from the shrimps were dissected under NaCl/Pi and homogenized in 1 mL per 50 mg (wet tissue weight) of homogenization buffer (0.1 M NaCl, 0.05 M Tris, 0.1% Tween-20, 1 mM phenylmethanesulfonyl fluoride and 1 lgÆmL)1 aprotinin; pH 7.8). The homogenate was centrifuged at 3000 g for 5 min The supernatant was collected and centrifuged at 17 000 g for 40 min. This successive supernatant containing the total protein extract was analyzed by 10% SDS/PAGE and Western blot detection with anti-rFAMeT serum as the primary antibody (1 : 20 000) and the goat anti-rabbit IgG– alkaline phosphatase conjugate as the second antibody (1 : 5000). The negative control was detected with preimmune rabbit the primary antibody. To test for specificity of the anti-rFAMeT serum duplicate Western blots of shrimp protein extracts were detected with anti-rCHH and anti-rMIH serum [23].

Shrimp farnesoic acid O-methyltransferase (Eur. J. Biochem. 269) 3589

(cid:1) FEBS 2002

Signals were visualized by adding 5-bromo-4-chloro- 3-indolyl phosphate/Nitro Blue tetrazolium/alkaline phos- phatase detection buffer and the colour reaction was terminated with running tap water. Protein concentration in the shrimp extracts was determined using a Bio-Rad protein assay kit.

Immunocytochemical detection of FAMeT in shrimp eyestalk denatured salmon and

included denaturation at 95 (cid:3)C for 1 min, annealing at 55 (cid:3)C for 1 min and extension at 72 (cid:3)C for 1 min for 35 cycles and an extension at 72 (cid:3)C for 10 min for the completion of the DNA synthesis. PCR products were analyzed on a 1% agarose gel and blotted onto a nylon membrane (Amersham) and cross-linked with UV irradi- ation. The membrane was hybridized overnight at 42 (cid:3)C to a DIG-labeled nonradioactive cDNA probe of 843 bp in a buffer (5 · NaCl/Cit, 0.5% SDS, 5 · Denhardt’s solution sperm DNA (100 lgÆmL)1) with 50% formamide). The membrane was then washed twice in 2 · NaCl/Cit with 0.1% SDS for 5 min at room temperature and finally at 68 (cid:3)C in 0.5 · NaCl/Cit containing 0.1% SDS twice for 15 min. Positive signals were detected with DIG–alkaline phosphatase alkaline conjugate by adding Nitro Blue tetrazolium (5%) and 5-bromo-4-chloro-3-indolyl phos- phate (5%) to the detection buffer (0.1 M Tris/HCl, pH 9.5, 0.1 M NaCl).

Enzyme preparation

Transformed bacteria BL21 (DE3) containing the expres- sion plasmid construct was cultured as stated above. Following induction with isopropyl thio-b-D-galactoside (IPTG) the bacterial culture was grown at 37 (cid:3)C with shaking for 2.5 h. The bacterial cells were pelleted by centrifugation at 800 g for 20 min. The bacteria pellet was then resuspended in NaCl/Pi buffer containing 1 mM phenylmethanesulfonyl fluoride and 1 lgÆmL)1 aprotinin and the cells lysed by a Polytron homogenizer. Following centrifugation at 10 000 g for 30 min at 4 (cid:3)C, the superna- tant was used to test for O-methyltransferase activity. The BL21 (DE3) with expression vector only was used as a negative control to check for any bacterial O-methyltrans- ferase activity.

Methyltransferase assay

Different tissues from the shrimp were dissected under cold NaCl/Pi. The rigid cuticle containing most of the retina and the lamina layer of the eyestalk was removed and the tissues were immediately fixed in Bouin’s fixative at 4 (cid:3)C for 24 h, dehydrated and embedded in paraffin. Consecutive 7-lm sections were mounted onto slides and dried at 37 (cid:3)C overnight. Tissue sections were deparaf- finized in xylene (twice for 5 min) and rehydrated through an ethanol series. The endogenous peroxidase was removed by incubating the tissue sections in 0.3% H2O2 in methanol for 30 min at room temperature and rinsed briefly with NaCl/Pi. The immunocytochemical staining of the tissue sections were performed with a Vectastain Elite ABC kit (Vector Laboratories, CA, USA) according to the manufacturers instructions. Tissue sec- tions were blocked in normal blocking serum (provided with the kit) for 30 min at room temperature. After blotting the excess blocking serum, the sections were incubated in the anti-rFAMeT serum (1 : 3000 dilution in NaCl/Pi buffer) overnight at 4 (cid:3)C. The control sections were incubated in preimmunized rabbit serum (1 : 400 dilution). The slides were washed with a large volume of NaCl/Pi under gentle agitation twice for 5 min each and incubated in a biotinylated secondary antibody solution (diluted according to the manufacturers instructions) for 30 min at room temperature. After the washing of the slides for 5 min in a large volume of NaCl/Pi, the sections were incubated for 30 min with Vectastain Elite ABC reagent according to the manufacturers instructions. The slides were washed with NaCl/Pi for 5 min and incubated in the peroxidase substrate solution containing 0.01% H2O2 and 1.39 mM 3,3¢-diaminobenzidine (Sigma, St Louis, MO, USA). The colour reaction was terminated by rinsing the slides in running tap water. The sections were dehydrated through increasing concentrations of ethanol, cleared with xylene and mounted with coverslips in DPX mountant (Sigma).

RT-PCR and Southern blot detection of FAMeT

acetate/acetic system of

The O-methyltransferase activity of rFAMeT was assayed according to the procedure of Reibstein & Law and Feyereisen et al. [11,22] with some modifications. The enzyme preparations were incubated in a final volume of 100 lL of 100 mM phosphate buffer (pH 7.2, containing 1 mM EDTA, 1 mM 2-mercaptoethanol and 1% BSA) with 0.4 lM [3H]S-adenosyl-L-methionine (SAM), 30 lM farnesoic acid and 1.0 mM unlabeled SAM for 45 min at 37 (cid:3)C. The reaction was stopped by the addition of 200 lL methanol and 100 lL 1% Na-EDTA. Cold carriers (20 lg JH III and 20 lg MF) were added to the reaction mixture prior to extraction with chloroform (2· 750 lL). After passing the chloroform phase through anhydrous Na2SO4, the extracts were dried under N2, redissolved in diethyl ether and applied to TLC plates (Merck silica gel 60 F254 plastic sheet). The samples were then focused twice with methanol. After development in a solvent acid toluene/ethyl (85 : 15 : 1, v/v/v), the bands corresponding to MF and JH III were cut out and assayed for radioactivity by liquid scintillation spectrometry. The absolute enzyme activity of rFAMeT was obtained by subtracting background and bacterial activity from the above results of the rFAMeT assay. Nervous tissue and eyestalks were extracted for total RNA [21]. RNA quality was monitored by agarose gel electrophoresis. The first strand cDNA was synthesized by reverse transcription in a buffer containing 1–5 lg total RNA, 2 pmol of gene specific primer, 2 mM dNTP mix, 2.5 mM MgCl2 and 1 U of Superscript II reverse tran- scriptase (Life Technologies, USA) at 42 (cid:3)C for 3 h. A pair of primers (forward: FAOM F4 5¢-CCGGGATCCA TGGCTGACAACTGGCCTGCC-3¢; reverse: FAOM R6 5¢-CGGGGTACCTTAGAATTCGAACTTCCACTT-3¢) PCR mix consisted of 2 lL of the RT reaction mix, 2 mM MgCl2, 2 mM of dNTP, 1 U of Taq DNA polymerase and 10 pmol of each primer in 1· PCR buffer. PCR

3590 Y. I. N. Silva Gunawardene et al. (Eur. J. Biochem. 269)

(cid:1) FEBS 2002

R E S U L T S

to remove the detergent. The yield for rFAMeT in the E. coli pRSET expression system was approximately 3 mgÆL)1. Bacterial expression of the recombinant FAMeT

Antiserum against Ni-nitrilotriacetic acid affinity-purified rFAMeT was raised by injecting rabbits with the renatured protein. The titer and specificity of the antiserum were determined by Western blot analysis (Figs 1B, I,II). The results indicated that the antiserum was highly specific as it detected only the rFAMeT in the cell lysate (Fig. 1B, II) with a dilution of 1 : 160 000.

Tissue distribution and immunolocalization of native FAMeT in shrimp

The rFAMeT protein was detected by using either the Ni-nitrilotriacetic acid–alkaline phosphatase conjugate or the mouse anti-(His-tag) IgG [23]. At different time points (0–4 h) after induction with IPTG, the recombinant protein of the expected size was detected on Western blots of the protein samples obtained from the cell pellets (Fig. 1A, II). The maximal expression of the rFAMeT in E. coli was observed 2.5 h after the IPTG induction (Fig. 1A, I,II). Furthermore, the expression of rFAMeT in the positive expression constructs is specific because no positive bands were detected in the bacterial controls (Fig. 1A, II). The the detected rFAMeT band was molecular mass of (cid:1) 35 kDa, which was the expected size of the fusion protein.

Purification of rFAMeT and production and specificity of anti-rFAMeT serum

Fig. 1. SDS/PAGE and Western blot analysis of the expression of rFAMeT protein in pREST-A/BL21 (DE3) (A) and titering of rabbit antiserum against rFAMeT protein by Western Blot (B). (A) (I) Cell lysates were prepared at different induction time points and proteins analyzed on a 12% SDS/PAGE that was stained with Commassie Blue. (II) A 1 : 1000 dilution of the Ni-nitrilotriacetic acid–alkaline phosphatase conjugate was used in the Western blot analysis to detect the His-tag rFAMeT fusion protein. The numbers on top indicate the induction time. All lanes show the soluble fraction except 4P, which indicate the insoluble fraction of the lysed cell pellet. The lysate of only bacteria BL21 (DE3) or pREST-A vector transformed BL21 (DE3) were used as controls. (B) (I) 200 ng of rFAMeT proteins (lanes 1 and 4–10) and whole bacterial lysate (lane 2) were analyzed by 12% SDS/PAGE and transferred to Hybond-C membrane. (II) Individual lanes were immunostained with different dilutions of rabbit antiserum against rFAMeT. PM, prestained standard protein marker. The last lane was the Western blot with preimmune rabbit antiserum.

The results of Western blot analysis indicated the presence of FAMeT in the ventral nerve cord, epidermis, eyestalk, heart, tail muscle, mandibular organ, ovary and testis (Fig. 2A,B). The strongest signal was observed for the ventral nerve cord whereas no positive signals were detected for the hepatopancreas and the mid gut (Fig. 2A,B). The molecular mass (32 kDa) of the protein was in accordance with that predicted from the amino-acid sequence deduced from the cDNA [18]. In addition, a larger protein of approximately 44 kDa unique to the eyestalk was detected in both the male and the female shrimps (Fig. 2A,B). To determine whether the distribution of the FAMeT varied between adult and juvenile shrimp, total protein of the eyestalk and ventral nerve cord of the juvenile shrimps were analyzed. The results indicate that FAMeT is present in both developmental stages in the eyestalk and the ventral nerve cord (Fig. 2C). Furthermore, the larger protein (i.e. 44 kDa) was also observed in the juvenile eyestalk (Fig. 2C). Due to the presence of expressed rFAMeT in inclusion bodies, the protein remained insoluble following lysis of the bacterial pellet. At high concentrations of denaturing reagents, such as urea and guanidine thiocyanate, most of the inclusion bodies were soluble in solution. Western blot analysis at different steps of the purification process indicated that this method enables the complete purification of rFAMeT with no contaminants (data not shown). To renature the protein following elution, the protein was first dialyzed in NaCl/Tris to remove urea and imidazole and the protein was further dialyzed against NaCl/Tris and NaCl/Pi

Shrimp farnesoic acid O-methyltransferase (Eur. J. Biochem. 269) 3591

(cid:1) FEBS 2002

Fig. 3. Cellular localization of FAMeT in shrimp. (A) Immunolocal- ization of FAMeT in eyestalk sections. (B) Higher magnifications (see rectangle in A). XO, X-organ; SG, sinus gland; ME, medulla externa; MI, medulla interna; MT, medulla terminalis; LAM, lamina gangli- onaris; NGF, nerve and glial fibers; AT, axonal tract; ON, optic nerve; OB, onion bodies of the X-organ; NSC, neurosecretory cells of the X-organ.

Fig. 2. Western blot analysis of proteins from different tissues of the M. ensis. (A) Detection of FAMeT in M. ensis. Lanes are protein samples from different tissues of the Female shrimp. (B) Detection of M. ensis FAMeT native protein in different tissues of the male shrimp. The last lane indicates protein sample from testis. (C) Detection of M. ensis FAMeT native protein in ventral nerve cord and eyestalk tissues of the juvenile shrimp.

Expression of the FAMeT at different molting stages

The expression of the FAMeT mRNA during the molt cycle was analyzed to ascertain whether the expression of FAMeT gene is related to the molting cycle in the shrimp. Because of the limited quantity of RNA obtained from ventral nerve cord and eyestalk of a single shrimp, RT-PCR was employed to study the expression of FAMeT. The results revealed that FAMeT mRNA is expressed throughout the intermolt (stage C), premolt (stage D) and postmolt (stages A, B) stages in both adult and the juvenile shrimp in the eyestalk tissue (Figs 4A,B). Similarly, FAMeT is also expressed in the ventral nerve cord of both juvenile and adult animals during all stages of the molt cycle (Fig. 4C,D). However, there is no significant variation in the level of expression in the ventral nerve cord and the eyestalk during the molt cycle. Further analysis indicated that the FAMeT mRNA is expressed in adults of both sexes throughout the molting cycle (Fig. 4B). However, in the eyestalk of juvenile premolt animals the FAMeT mRNA is expressed only in some shrimps (Fig. 4A). To further confirm the RT-PCR results, a Southern hybrid- ization was carried out on the RT-PCR products with a cDNA-specific probe (Fig. 4A–D, III) and b-actin RT-PCR amplification was used as an internal control to show the integrity of the RNA samples (Figs 4A–D, II). Immunocytochemistry of serial sections of the eyestalk revealed the presence of the FAMeT in the sinus gland, the X-organ neurosecretory cell clusters together with the lamina ganglionaries and the nerve fibers of the three ganglia of the eye: the medulla externa, medulla interna and the medulla terminalis (Fig. 3A). The sinus gland, located between the medulla externa and interna of shrimp eyestalk show strong immunoreactivity to anti-rFAMeT serum (Fig. 3A). The neurosecretory cells and the laminar struc- ture of the onion bodies of the X-organ [24] show the highest immunoreactivity to the anti-rFAMeT serum in the M. ensis eyestalk (Fig. 3B).

3592 Y. I. N. Silva Gunawardene et al. (Eur. J. Biochem. 269)

(cid:1) FEBS 2002

Fig. 4. RT-PCR detection of the FAMeT transcripts at different molting stages in M. ensis eyestalk (ES) and ventral nerve cord (VNC). (A) Expression of the FAMeT of M. ensis at different molting stages in the juvenile eyestalk. (B) Expression of the FAMeT of M. ensis in the eyestalk at different molting stages in the adult males and females. (C) Expression of the FAMeT of M. ensis at different molting stages in the adult ventral nerve cord. (D) Expression of the FAMeT of M. ensis at different molting stages in the juvenile ventral nerve cord. (A–D, I), RT-PCR detection of M. ensis FAMeT transcripts at different molting stages of the shrimp; (A–D, II), RT-PCR detection of the shrimp b-actin gene using gene-specific primers; and (A–D, III) Southern blot analysis of FAMeT RT-PCR products with cDNA coding region probe.

Radiochemical assay for the functional analysis of rFAMeT

To obtain a large quantity of FAMeT for the functional assay, we expressed the rFAMeT in a pRSET bacterial expression system. The O-methyltransferase assay was carried out with the rFAMeT to check for presence of enzyme activity. As shown in Fig. 5, MF production increased in relation to the increasing amounts of rFAMeT. However, when the protein quantity was increased beyond 17.5 lg, the enzyme activity was relatively constant (Fig. 5). The data shown in Fig. 5 indicated the absolute enzyme activity for each of the samples following the subtraction of bacterial or background activity.

D I S C U S S I O N

Fig. 5. Methyl farnesoate production by different quantities of rFA- MeT. Each point represents the mean ± STDEV of five determina- tions.

The insect JH-related compound MF is produced and released by the mandibular organ of crustaceans [5,25,26].

MF is thought to be involved in both reproduction and molting in the crustaceans [5,7]. FAMeT catalyzes the methylation of FA to MF as the terminal step in the MF biosynthetic pathway. Changes in O-methyltransferase activity during development in insect corpora allata have been suggested to be important components in the regula- tion of JH biosynthesis [9]. Hence, the present study was carried out to determine the biochemical function of the shrimp FAMeT gene and cellular distribution of native FAMeT enzyme, in an attempt to investigate the biochemi- cal and biological properties and the regulation of MF production in M. ensis. Using the bacterial expression system, a recombinant protein of (cid:1)35 kDa was produced (Fig. 1A, II). This protein is 3 kDa larger than the predicted molecular mass of the FAMeT protein (32 kDa) as a result of the addition of several amino acids at the N-terminal end of the fusion protein. The rFAMeT was purified through an affinity column with the aide of the series of His6-tag residues in the fusion protein, which function as a metal binding domain. The His6-tag residues further assist the subsequent analysis of the purification by Western detection of the fusion protein with Ni-nitrilotriacetic acid conjugate. The results from the Western blot indicate the anti-rFAMeT serum specifically detected rFAMeT in the cell lysate demonstrating the high specificity of the antiserum (Fig. 1B, II) and the wide distribution in nature of the FAMeT in shrimp (Fig. 2). The anti-rFAMeT serum detects the highest level of the FAMeT in ventral nerve cord tissue, whereas FAMeT level is lower in the epidermis, eyestalk, heart, tail muscle, mandibular organ, ovary and testis (Figs 2A,B). This observation agrees with the results of our earlier study on the expression of the FAMeT mRNA [18]. However, in contrast to results from RT-PCR [18], FAMeT is not present in the hepatopancreas by Western blot analysis (Figs 2A,B). Northern blot analysis [18] suggests that FAMeT mRNA level is much lower in hepatopancreas than ventral nerve cord, tail muscle or testes and thus, FAMeT expression in this tissue may be below the sensitivity of this assay. Alternatively, FAMeT may be

Shrimp farnesoic acid O-methyltransferase (Eur. J. Biochem. 269) 3593

(cid:1) FEBS 2002

intermolt, increases rapidly during premolt and drops precipitously prior to ecdysis in crustaceans [20,34,35]. According to our observations, FAMeT mRNA is expressed in the ventral nerve cord and eyestalk throughout the molt cycle in both the adult and juvenile shrimp (Fig. 4A–D). Furthermore, analysis of FAMeT mRNA in the eyestalk of the adult shrimp from both sexes gives similar results (Fig. 4B). Because MF appears to be involved in the regulation of molting, the expression of FAMeT mRNA throughout the molting cycle is required for the synthesis of MF. We have observed that FAMeT is expressed in a sex-specific manner in the juvenile shrimp [18]. Therefore the differential expression of FAMeT during the premolt stage in juveniles (Fig. 4A) is likely to be related to sex-specific expression in juvenile shrimps.

secreted from the hepatopanceas. However, extensive work will be required to determine the definite role of FAMeT in the shrimp hepatopancreatic tissue. Surprisingly, the man- dibular organ elicits a weak signal for FAMeT in compar- ison to ventral nerve cord and other tissues (Fig. 2A,B). This suggests that the last step of MF biosynthesis catalysed by FAMeT enzyme occurs in many tissues. FA has been implicated as a major product of release from the mandib- ular organ in the crustaceans [26]. Similar reports have also been documented in the insects where the JH-acids released from the corpora allata are converted to the corresponding JHs in the male accessory reproductive glands [27]. The wide distribution of the FAMeT in different shrimp tissues further confirms our earlier speculation that the major part of the FA is secreted from the mandibular organ and transported to target tissues for final conversion to MF [18]. However, it is also possible that FAMeT methylates substrates other than FA in tissues apart from mandibular organ. In M. ensis, FAMeT is also found in the juvenile ventral nerve cord and eyestalk tissues (Fig. 2C). In insects, O-methyltransferase activity in the corpora allata of the D. punctata has been reported to be high during the early stages of development suggesting a possible regulatory role for O-methyltransferase on JH biosynthesis at these stages [10]. Similarly, the presence of FAMeT in these juvenile tissues suggests that this protein could be an important component in the regulation of MF synthesis during the early developmental stages of the shrimp.

Our data from the assay of rFAMeT demonstrates that the enzyme is able to produce a significant quantity of MF (Fig. 5). This indicates that the FAMeT gene codes for a functional farnesoic acid O-methyltransferase in M. ensis and is the first demonstration of the biological function of this gene. The refolding of the protein during renaturation is important for the function of the enzyme. When we used recombinant protein renatured following purification under denaturing conditions, a very low enzyme activity was observed. This is presumably a consequence of the incom- plete refolding of the fusion protein. However, when we use a partially purified rFAMeT extract in the subsequent functional assays a significantly higher level of O-methyl- transferase activity was observed (Fig. 5). Because the enzyme activity attributable to the bacterial proteins or the vector proteins (background very low and insignificant) were subtracted from the final data of the rFAMeT enzyme activity, the data shows the absolute level of O-methyl- transferase activity of the recombinant protein. Although the activity of rFAMeT is comparatively low when com- pared to the native O-methyltransferase activity of crusta- ceans and insects, the actual amounts of FAMeT in these studies were unknown [10,25,26]. Moreover, it is well known that bacterial expression systems commonly fail to process the disulfide bonds and the correct refolding as a consequence of the lack of post-translational modifications in prokaryotes. Therefore, recombinant proteins from bacterial systems usually have much lower activity than the native proteins. This has been observed in other recombinant proteins of the shrimp and also in other recombinant proteins expressed in bacteria [19,36,37]. Hence, only a small fraction of the rFAMeT was expected to fold in the correct conformation and consequently low FAMeT activity in the functional assay.

FAMeT was localized in the neurosecretory cells of the X-organ sinus gland complex by immunocytochemical detection (Fig. 3A). The X-organ-sinus gland complex, a neurohemal organ of decapod crustaceans, is the site of synthesis, storage and release of the neuropeptides of the CHH/MIH/GIH family [19,28]. We have previously shown that expression of the FAMeT in the eyestalk and ventral nerve cord coincides with the expression of some of these eyestalk neuropeptides [18,19]. Eyestalk neuropeptides of the CHH/MIH/GIH family are also located in the neu- rosecretory cells of the X-organ-sinus gland [19,23]. More- over the results from previous studies indicate that in the crab, the biosynthesis of MF in the mandibular organ is regulated by an eyestalk neuropeptide, mandibular organ inhibiting hormone [9,16,17]. Interestingly, a protein of a larger molecular mass ((cid:1) 44 kDa) was also detected in the eyestalk of both the adult and juvenile M. ensis (Fig. 2A– C). Analysis of FAMeT for protein modification signatures (PROSITE) suggest that modification is unlikely to be responsible for the shift in protein size observed. Therefore, the occurrence of the larger protein in the eyestalk may be attributable to binding of FAMeT with the eyestalk factors related to MF biosynthesis in the shrimp. Another alterna- tive is that the antiserum is simply recognizing in eyestalk tissue, another protein with similar antigenic determinants. However, further studies are required to ascertain the significance of this larger protein.

In conclusion, the wide tissue distribution of the FAMeT in shrimp implies that this gene may also be involved in other physiological processes apart from its role in repro- duction and metamorphosis. Further, the specific cellular localization of FAMeT in the X-organ-sinus gland complex suggests that FAMeT may interact with the eyestalk neuropeptides and thereby regulate the synthesis and role of MF. All this suggests that FAMeT is a key regulatory enzyme in the MF biosynthetic pathway of the shrimp. In addition, the occurrence of FAMeT mRNA throughout the molting cycle suggests that FAMeT may be involved in molting or related processes in the shrimp. Finally, the present work provides a framework for study of the regulation of biosynthesis and function of MF by FAMeT Molting, a prerequisite for growth in arthropods, is under direct control of neurohormones in the Crustacea [29,30]. There is also evidence supporting a regulatory role of MF in the molting process of the crustaceans [31–33]. Therefore, to ascertain the expression of FAMeT in relation to the molting process of the shrimp, we determined the level of expression of this gene during the molt cycle. In general, the circulating hemolymph ecdysteroid titer remains low during

3594 Y. I. N. Silva Gunawardene et al. (Eur. J. Biochem. 269)

(cid:1) FEBS 2002

of an allatostatin from the tobacco hornworm Manduca sexta. Proc. Natl Acad. Sci. USA 88, 9458–9462.

16. Tsukimura, B. & Borst, D.W. (1992) Regulation of methyl far- nesoate in the hemolymph and mandibular organ of the lobster, Homarus americanus. Gen.Comp. Endocrinol. 86, 297–303.

and its mode of interaction with eyestalk neuropeptides to establish definitive roles of MF in shrimp and all crusta- ceans.

A C K N O W L E D G E M E N T S

This work was supported by a RGC earmarked grant (# HKU 7227/ 00M) awarded to S. M. C. by the Hong Kong SAR Government and by operating grants from the Natural Sciences and Engineering Research Council of Canada. Y. S. G. is the recipient of a Presidential scholarship from the Sri Lankan Government and the J. G. Phillips Memorial Scholarship from the University of Hong Kong.

17. Wainwright, G., Webster, S.G. & Rees, H.H. (1999) Involvement of adenosine cyclic-3¢,5¢-monophosphate in the signal transduc- tion pathway of mandibular organ-inhibiting hormone of the edible crab, Cancer pagurus. Mol. Cell. Endocrinol. 154, 55–62. 18. Silva Gunawardene, Y.I.N., Chow, B.K.C., He, J.G. & Chan, S.M. (2001) The shrimp FAMeT cDNA is encoded for a putative enzyme involved in the methylfarnesoate (MF) biosynthetic pathway and is temporally expressed in the eyestalk of different sexes. Insect Biochem. Mol. Biol. 31, 1115–1124.

R E F E R E N C E S

19. Gu, P.L., Yu, K.L. & Chan, S.M. (2000) Molecular character- ization of an additional shrimp hyperglycemic hormone: cDNA cloning, gene organization, expression and biological assay of recombinant proteins. FEBS Lett. 472, 122–128.

1. Tamone, S.L., Prestwich, G.D. & Chang, E.S. (1997) Identification and characterization of methyl farnesoate binding proteins from the crab, Cancer magister. Gen. Comp. Endocrinol. 105, 168–175. 2. Chaix, J.C. & De Reggi, M. (1982) Ecdysteroid levels during ovarian development and embryogenesis in the spider crab Acanthonyx lunulatus. Gen. Comp. Endocrinol. 47, 7–14.

20. Chan, S.M., Rankin, S.M. & Keeley, L.L. (1988) Characterization of the molt stages in Penaeus vannamei: setogenesis and hemo- lymph levels of protein, ecdysteroid and glucose. Biol. Bull. 175, 185–192.

3. Laufer, H., Landau, M., Borst, D.W. & Homola, E. (1986) Advances in Reproduction, 4, (Andries, J.C. & Dhainaut, A., eds), pp. 135–143. Amsterdam, Elsevier Science Publishers.

21. Chomczynski, P. & Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159.

4. Laufer, H., Borst, D.W., Baker, F.C., Carrasco, M., Sinkus, M., Reuter, C.C., Tsai, L.W. & Schooley, D.A. (1987) Identification of a juvenile hormone-like compound in crustacean. Science 235, 202–205.

22. Reibstein, D. & Law, J.H. (1973) Enzymatic synthesis of insect juvenile hormones. Biochem. Biophys. Res. Commun. 55, 266–272. 23. Gu, P.L., Chu, K.H. & Chan, S.M. (2001) Bacterial expression of the shrimp molt-inhibiting hormone (MIH): antibody production, immunocytochemical study and biological assay. Cell Tissue Res. 303, 129–136.

5. Laufer, H., Landau, M., Homola, E. & Borst, D.W. (1987) Methyl farnesoate; its site of synthesis and regulation of secretion in a juvenile crustacean. Insect Biochem. 17, 1129–1131.

24. Bell, T.A. & Lightner, D.V. (1988) A Handbook of Normal Penaeid Shrimp Histology, 1st edn. pp. 26–33. The World Aquaculture Society and Allen Press, Inc. Lawrence, Kansas, USA.

6. Sagi, A., Homola, E. & Laufer, H. (1991) Methyl farnesoate in the prawn Macrobrachium rosenbergii: synthesis by the mandibular organ in vitro, and titers in the hemolymph. Comp. Biochem. Physiol. B. 99, 879–882.

25. Tobe, S.S., Young, D.A. & Khoo, H.W. (1989) Production of methyl farnesoate by the mandibular organs of the mud crab, Scylla serrata: validation of a radiochemical assay. Gen.Comp. Endocrinol. 73, 342–353.

7. Borst, D.W., Laufer, H., Landau, M., Chang, E.S., Hertz, W.A., Baker, F.C. & Schooley, D.A. (1987) Methyl farnesoate and its role in crustacean reproduction and development. Insect Biochem. 17, 1123–1227.

26. Tobe, S.S., Young, D.A., Khoo, H.W. & Baker, F.C. (1989b) Farnesoic acid as a major product of release from crustacean mandibular organ in vitro. J. Exp. Zool. 249, 165–171.

27. Weirich, G.F. & Culver, M.G. (1979) S-Adenosylmethionine: Juvenile hormone acid methyltransferase in male accessory reproductive glands of Hyalophora cecropia (L.). Arch. Biochem. Biophys. 198, 175–181.

28. Liu, L., Laufer, H., Wang, Y. & Hayes, T. (1997) A neuro- hormone regulating both methyl farnesoate synthesis and glucose metabolism in a crustacean. Biochem. Biophys. Res. Commun. 237, 694–701. 29. Webster, S.G.

8. Smith, P.A., Clare, A.S., Rees, H.H., Prescott, M.C., Wainwright, G. & Thorndyke, M.C. (2000) Identification of methyl farnesoate in the cypris larva of the barnacle, Balanus amphitrite, and its role as a juvenile hormone. Insect Biochem. Mol. Biol. 30, 885–890. 9. Wainwright, G., Webster, S.G. & Rees, H.H. (1998) Neuropeptide regulation of biosynthesis of the juvenoid, methyl farnesoate, in the edible crab, Cancer pagurus. Biochem. J. 334, 651–657. 10. Wang, Z., Ding, Q., Yagi, K.J. & Tobe, S.S. (1994) Terminal stages in juvenile hormone biosynthesis in Corpora Allata of Diploptera punctata: development changes in enzyme activity and regulation by allatostatins. J. Insect Physiol. 40, 217–223.

11. Feyereisen, R., Friedel, T. & Tobe, S.S.

(1998) Neuropeptides inhibiting growth and reproduction in Crustaceans. In Resent Advances in Arthropod Endocrinology (eds G. M. Coast & S. G. Webster), pp. 33–52. Cambridge University Press, UK.

(1981) Farnesoic acid stimulation of C16 juvenile hormone biosynthesis by corpora allata of adult female Diploptera punctata. Insect Biochem. 11, 401–409.

30. Ding, Q. & Tobe, S.S. (1991) Production of farnesoic acid and the cray fish,

methyl farnesoate by mandibular organs of Procambarus clarkii. Insect Biochem. 21, 285–291.

31. Byard, E., Shivers, R. & Aiken, D.E. (1975) The mandibular organ of the lobster, Homarus americanus. Cell Tissue Res. 162, 13–22.

12. Tobe, S.S. & Clarke, N. (1985) The effect of 1-methionine con- centration on juvenile hormone biosynthesis by corpora allata of the cockroach Diploptera punctata. Insect Biochem. 15, 175–179. 13. Yagi, K.J., Konz, K.G., Stay, B. & Tobe, S.S. (1991) Production and utilization of farnesoic acid in the juvenile hormone biosyn- thetic pathway by corpora allata of larval Diploptera punctata. Gen. Comp. Endocrinol. 81, 284–294.

32. Taketomi, Y. & Kawano, Y. (1985) Ultrastructure of the man- dibular organ of the shrimp, Penaeus japonicus, in untreated and experimentally manipulated individuals. Cell Biol. Int. Report 9, 1069–1074.

14. Woodhead, A.P., Stay, B., Seidel, S.L., Khan, M.A. & Tobe, S.S. (1989) Primary structure of four allatostatins: neuropeptide inhibitors of juvenile hormone synthesis. Proc. Natl Acad. Sci. USA 86, 5997–6001.

33. Yamamoto, H., Okino, T., Yoshimura, E., Tachibana, A., Shimizu, K. & Fusetani, N. (1997) Methyl farnesoate induces larval metamorphosis of the barnacle Balanus amphitrite via pro- tein kinase C activation. J. Exp. Zool. 278, 349–355.

15. Kramer, S.J., Toschi, A., Miller, C.A., Kataoka, H., Quistad, G.B., Li, J.P., Carney, R.L. & Schooley, D.A. (1991) Identification

Shrimp farnesoic acid O-methyltransferase (Eur. J. Biochem. 269) 3595

(cid:1) FEBS 2002

regions of the lobster (Panulirus argus) nervous system. Glia 20, 275–283.

34. Chang, E.S. & Bruce, M.J. (1980) Ecdysteroid titers of juvenile lobsters following molt induction. J.Exp. Zool. 214, 157–160. 35. Tamone, S.L. & Chang, E.S. (1993) Methyl farnesoate stimulates ecdysteroid secretion from crab Y-organs in vitro. General Comp Endocrinol. 89, 425–432.

36. Linser, P.J., Trapido-Rosenthal, H.G. & Orona, E.

(1997) Glutamine synthetase is a glial-specific marker in the olfactory

37. Ren, X., Gao, S., You, D., Huang, H., Liu, Z., Mu, Y., Liu, J., Zhang, Y., Yan, G., Luo, G., Yang, T. & Shen, J. (2001) Cloning and expression of a single-chain catalytic antibody that acts as a glutathione peroxidase mimic with high catalytic efficiency. Biochem. J. 359, 369–374.