. .A. Cover Sheet FINAL REPORT Covering Period: October 2000-March 2004 Submitted to the U.S. Agency for International Development: Bureau for Economic G r o u ~ . .4piculture. and Trade IlLPROVEMENT OF GROU'TH .AND SURVIVAL IN CULTURED RABBITFISH (SIG.4.Yl :T GL'7T.4T L . 9 Principal Investigator: Dr. Bruria Funkenstein Grantee Institution: Israel Oceanographic % Limnological Research Tel-Shihona P.0.B 8030. Haifa 3 1080. Israel Tel: 9724-8565233. e-mail: bruria~ocran.org.il Collaborator: Prof. h i e h Genler Institution: The Hebrew l'niversiv of Jerusalem Collaborator: Dr. Felix G. Ayson Institution: Southeast Asian Fisheries Development Center Project Number: C 19-001 Grant Number: TA-MOU-99-C 19-00 1 Grant Project Officer: Mr. William Crane Project Duration: April 1.2000-March 3 I. 2004 + B . Table of Contents .A . Cover Page .............................................................. 1 B . Table of Content ........................................................ 2 C. Executive Summary.................................................... 3 . . D . Research Object~ves.................................................. -4 E . Methods and Results ................................................... 4 F. Impact. Relevance.and Technolog?. Transfer.......................6 G . Project .4ctivitiedOutputs............................................. 6 .. H . Project Product~ulty.................................................... 6 I . Future Work ........................................................... . -7 . .A ppendis . Publ~caoonNo .1 ......................................... 8 Publicalion No . :!....................................... 17 Publication No. 3 ...................................... 27 i C. Executive Summan.: 2 The purpose of this project was to improve the grouzh and sumival of the rabhitfish. Sipnus guffuftu. cultured in Southeast Asia by underslanding the rate-limiting componenqs) in the grouth hormone-insulin-like growth factor system. which might explain the slow -gonth m e during the early juvenile stage and use this information to enhance p u t h rate by a d m i n i d o n of groulh hormone using various methodologies. During the project. grouth hormone (GH) from rabbitfish uas purified and antiserum \\as raised in rabbits. The native protein purified at he nart of the project \\as used in p w z h enhancement experiments of rabbitfish kept under various salinities. The cDK";As for rabbitfish GH and insulin-like grouth factors I and II (IGF-I and IGF-11) were also cloned. Recombinant GH protein was expressed in bacteria in large quantities and purified. The e f f e c t i v m of he recombinant hormone to enhance grouth u ~ a stested in rirw by injecting he hormone to rabbitfish and monitoring grouth (increases in body weight and body length). For testing he biological activit!- in rirro. GH was added to liver pieces removed from rabbitfish and l i e for shon incubation periods. Levels of IGF-I mRNA were determined. Experiments were also carried oul (by using molecular biology methods) in u-hich he level of expression of two genes for p u t h factors (IGF-I and IGF-11) was determined during earl>-stages of development of he &bitfish. The results obtained suggest that administration of a little exogenous GH to he fr?- significantly improved its g o n t h and rearing fr?. in dilute seawater likeuise improved its grouth performance. The collaboration during the project included a visit of the Philippine PI to he Israeli Pls laboratories. During the visit experiments were initiated to test various uals of introdwing hormones into microdiet of fish Ian-ae. In addition the Philippine PI and he Israeli PI met during the Fish Endocrinoloa Meeting and took this opportunip to discuss h e mul& of he project and planning future work. The funding provided by the project. enabled Dr. A y n to acquire imponant pieces of equipment needed for the production of recombinant f f i H in his laboraton-. In addition. a fermentor. freeze-drier and a big refrigerated centrifuge were provided by a pant from the Japanese Gowrnment. . D. Research Obiectives: 4 1. Purification of S. gullarus GH polypeptide and raising antibodies. .7 Molecular cloning of S. prrlufzrs GH and IGF-I cDNAs. 3. Cloning of GH and IGF-I in expression vectors for producing large quantities. 4. Effect of exogenous natitr S guttatus GH on growIII. 5. Effect of exogenous bovine GH and low salinity on growIII. 6. Production of large quantities of S prrranls recombinant GH polypeptide. 7. Production of S. gutratus recombinant IGF-I poljpeptide. 8. Assays to rest the biological activities of recombinant S. gutfarus GH. 9. Extending the half-life of the recombinant S. gutratus GH. E. Methods and Results: I lsolarion o/-growthhormone lGH)from rabbi!fish GH was emacted from p i t u i w glands under alkaline conditions. tiactionated by gel filtration chromatography on Sephadex G-100. and purified b?; high performance liquid chromampphy (The results are summarized in publication No. I J. 2. Cloning ofrahhi!fishGH cD.X4 A combination of RT-PCR and RACE (rapid amplification of cDNA ends) resulted in a cDN.4 of 860 nt. excluding the poly(A) tail. The unmslated regions are 94 and 175 bp in he 5' and 3' ends. respectively. It has an open reading frame of 588 bp coding for a signal peptide of 18 amino acids and a mature protein of 178 amino acid residues (The results are summarized in Publicarion No. I ). 3 Cloning ojrubbiesh IGF-I cD.V.4 Rabbitfish IGF-I cDNA was successfully cloned from the liver. The preprolGF-l is composed of 186 amino acids (aa): this is composed of 4 aa for the signal peptide. 68 aa for he mature peptide and 71 aa for the E domain. Only one form of Edomain is found in rabbitfish IGF-I (The results see summarized in publication No. 2). 4. Cloning 01-rabbiifishIGF-II cD.Y4 (nor in the original proposall Rabbitfish IGF-I1 cDNA was also successfully cloned from the liver. The preprolGF-ll is 215 aa long: this is composed of 1 7 aa for the signal peptide. 70 aa for he mature peptide and 98 aa for the E domain (The results are summarized in publication No. 2). 3. Drvelopnrenral espression o f IGFs in rubhir/ish In order to examine if there is differential regulation of both lGFs during early development in rabbitfish. primers were designed for both rabbitfish IGF-I and I1 and used it for RT-PCR of . 3 samples from different stages of embnogenesis and different l m a l stages. IGF-II transcripl uas e\pressed in unfertilized eggs (although \ e n ueak) and a \en. strong expression uas alrpad? ohsened during morula stage (about 3 hours after fertil~zation)and this uas maintained until hatching. On the other hand. IGF-I transcript \\as observed only in embpos near hatching. In he lanae. ho\ve\-er. both lGFs u-ere more or less equally expressed (The results are summarized in publication No. 2). 6 Erperinrenrs to enhance growrh in rabbi@sh a ) Treatment with bovine GH to improve p u t h in rabbitfish 6y: Rahbittish f n u-ere injected uith bovine GH at doses of 0.1 and 0.01 p@g BW. GH injection \\as gken weekl!- for 4 weeks and the experiment lasted for 8 weeks. Significant increase in p a t h \\-as observed starting on the 5' week in the treated groups. b) Rabbitfish fn. grow faster in dilute SW than in full-strength SW. .4n esperiment was conducted to find out whether rearing h) in dilute SW and treating h e m with bovine GH u ~ u l d have an additive effect on growth. Fry were reared for 8 weeks in dilute SW (20 ppli and botine GH (0.01 pp'g BBW) was injected u~eeklyin the fim 4 weeks. Rearing the h)-in dilute SW and treating them also uith bGH did not further improw the growh. c) Preliminary experiments urre performed to test the effect of he purified GH to promote grouth in rabbitfish fn. Weekly intraperitoneal injections of he GH significantl) a c c e h e d grouth. The effect uas e\ident 3 weeks after the starl of he treatment. and w a s still significant1 \\eelis afier the treatment u-asterminated (The results are summarized in publication So. I). -. Producrion of recombinanr rabbitfih GH Recombinant rabbirfish GH was cloned in the expression vector PET-3d and used to aansform BL?lDEI; bacteria. Large amounts of the recombinant protein were produced: he prorein uas refolded and purified. The refolding and purification steps were monitored by SDS-PAGE and HPLC. (The results are summarized in publication No. 3). X Biological acrivir) ofrecombinan/ rabbitfish GH The biological activity of 2 preparations of recombinant rabbitfish GH (&H1 using rabbittish fr?-. Groups of rabbitfish Fr?. were uas tested in r i w injected with rrfGH-1 (refolded in 0.1 mM c)-steine) and rrfGH-2 (refolded in 0.2 mM cysteine) at doses of 0.01 (low dose). 0.1 (medium dose) or 0.5 (high dose) pg'g BW once a week for five weeks. Increases in body aright and body length wrre monitored during each injection schedule. which is at I-week interval for 5 weeks. Rabbitfish fn. injected uith the highest dose (0.5 pg/g BBW) of &H refolded in 0.1 m5i c!-steine \\ere bigger than h) in all other ~ m u p s The . size difference w a s apparent 2 \\-eels afier he stan of the treatment (The results are summarized in publication No. 3). 6 F. Impact. Relevance and Technoloev Transfer Throughout the project. the research involved a close collaboration beheen he scientists participating in the project. Recombinant GH was re-folded and purified b!- he Israeli collaborators and transferred to the Philippine partner for testing its ability to enhance -gonth in rabbitfish fi)-. The Philippine Fanner came for a \isit to Israel in order to familiarize himself with the equipment needed for large-scale production of recombinant proteins as well as to learn new ways for incorporating the GH in micmdiet of fish larvae. The recombinant rfGH nas used for enhancement of p w l h of cultured rabbitfish. Dr. Felix Ayson came to Israel in he second >-ear of the project for training in methods of incorporating recombinant GH polypeptide in micmdiet of fish larvae as one nay of administrating the hormone. He also n-as p m e n t d to he equipment needed for production of recombinant proteins so that he mill be able to acquire these items using pan of the budget allocated for this purpose. G. Proiect .4ctivities/Oumuts I . Dr. A>-sonvisited Israel for 2 \veeks in November 2002. 2 . Publications: Ayson. F.G.. de Jesus E.G.T.. Arnemiya Y.. Moriyama. S.. Hirano. T.. and Kan;auchi. H. (2000). Iwlation. cDNA cloning. and p n l h promoting activin- of rabbitfish (Siganus gunatus) gronfi hormone. Gen. Comp. Endocrinol. 117: 251259. Ayson. F.G.. de Jesus E.G.T.. Moriyama S.. Hoyodo. S. Funkenstein B.. <krrler. .and I. L. u a u c h i . H. (2002). Differential espression of insulin-like p n t h factor I and 11 mRNXs during emb~ogenesisand early l m a l development in rabbifish Siganus gzmarils. Gen. Comp. Endocrinol. 116: 165-174. Funkenstein. 8.. m-man. A.. Lapidot. Z.. de Jesus-Ayson. E.G.. Gertler. A. and .4ywn. F.G. (2005). Expression and purification of a biologically a d \ - e recombinant rabbitfish (Sigonus g~~frrrro~ns) gronlh hormone. Aquaculture 250: 504515. H. Project pmductiviIy \lost of o w goals \\-ere ache\-ed throughout the project although h - o aims were not obtained. In the original proposal we intended to tv to extend the half-life of the recombinant rabbitfish GH that \ve produced. This pan was not accomplished for lack of time. Also. in our original plan.^ 7 intended to produce r e c o r n b i t rabbitfish IGF-I. Again. due to the short time and limited budgel provided for this project. we have only cloned h e IGF-I cDNA and characterized in sequence hut did not produce h e protein. I. F u m Work In the laboraton of the Philippine PI. activities toward developing s m e g i e s for incorporating ihe recombinant protein into formulated feeds for mass application will be continued. Some important pieces of laboratop equipment necessw for this kind of work ha\-e already been purchased using the funds provided by USAID. However. outside funding sources will S i l l be a c t i \ ~ I ysought h>-the Philippine PI in order for him to continue the work we narted using M from USAID. s General ard Cmparauve G d o u u o b 1117.251-259 (i000) dot 10 I W p 1999 7409. a ~ l l a b l e m l m at h n p ~ j v v x tdeal%m u.l i t tl' Isolation, cDNA Cloning, and Growth Promoting Activity of Rabbitfish (Siganus guttatus) Growth H ~ r m o n e ' . ~ Felix G. Ayson,' Evelyn Grace T. de Jesus,' Yutaka Amemiya.t S h u n s u k e Moriyama.t Tetsuya Hiran0.S and Hiroshi Kawauchit 'Aquarullure Depanmenr. burheart Asian Fisherks kebpmenr Center BEAFDEC AQD). Tmusn 5021 Ibib. The Philiooinerr ~ d Fkheria Snenm. ~.~ &laem Cnh-w. ~.? Labmatom - ,of h4olmhr E d m w x h Sdml Sanriku. {warp 022-0101, Japan: and ?Hawaii lnstiruteofMm.m Bielw, L ' f i ~ ~ d ~ Ht ayw a ~ iC . m u t Irhnd. K-, Hawaii 96744 - ~.. Accepted October 25. 1999 We repon the isolation. cDNA cloning, and gmwth promoting activity of rabbitfish (Siganmguftahls;Teleostei; Perciformes; Siganidae) growth hormone (GH). Rabbidish GH was -ed horn pituitary glands under alkaline conditions, fractionated by gel filtration chromatography on Sephadex C-100, and p u d k d by highperformance liquid chromatography. The bactions containing GH were identified by immunobloning with bonito GH antiserum. Under m d u c i n g conditions. the molecular weight of rabbidish GH is about 19 kDa as estimated by SDS-PAGE. The purir~edhormone war potent in promoting growth in rabbidish fry. Weekly intraperitoneal injecrions of the hormone signlficanlly accelerated growth. This was evident 3 weeks aRer the start of the treabnent. and i t s eRect was still s i m ~ c a n2t weeks aRer the treatment was terminated. Rabbidish GH cDNA was cloned to determine its nudeotide sequence. Excluding the poly (A) tail, rabbidiih GH cDNA is 860 base pairs (bp) long. It contained unbanslated regions of 94 and 175 bp in the 5' and 3' ends. respectively. 11 ha% an open reading lrame of 588 bp coding f w a signal peptide of 18 amim acids and a mature protein of 178 amino acid residues. Rabbidish GH ha% 4 cysteine p a p ha= k n data w p m * in &psiled in ihe DDBJ/EMBL/GenBank dalabavr urder Accession No. AEQ31298. Aoalentkoendin~flFQ 1-199941457)for thecDNXseoueweo1 rabbithsh G H reponed in h i s paper. I The n d m * vq- . 0 . vesidues. On the amlna d kvel. nbbidish GH shorn high Identity (71-74%) with GHs d ocha perifornu such as tuna. sea laass. ~ U O W tail, bmim. d 6LpL. a d l e u (47-49%) identity with sllmonid d carp Gk. a m - h Key H b k rabbitfish; growthhonnne:gmth pn*oo tion Growth hormone (GH) therapy isarmma\fy used lo enhance growth in domesric animals and it has also become increasingly popular in fish in aquacufture.In hsh. approaches to accelerate grow* by GH adminktration vary from injection and oral administration of the hormone to uansgenio. Results from several of these ~riakwere. however. incomiaenL bxpl for some cases of transgenia. where the fish p m d d were exuaordinarily large, albeit few (Devlin H d. 1994). intraperitoneal injection of either m ~ r a or l recombinant GH yielded s i g n i h t results 4 if the weekly injections were for more than 5-9 week. (Kawauchi eta/.. 1986: Sakata et d.. 1993: Guillen U d.. 1998). Oral administration of the hormone \vould be the most practical for mass application in hsh hushandrv al&ueh the oueslion of* effective deli\-rr. 3. of the hormone wit&,ut it being degraded in the stomach should be carefully awessed O f o r i y H ~ dl.. ~ ~ 1993: McLean lgg9). Slo\v-growjng and economically important fish spe- cies are good candidates for growth enhancement by GH therapy. Such is the case for rabbitfish. Siganus guttatus Perciformes: Siganidae), an important aquaculture species in Southeast Asia and the Arab Gulf region. A full understanding of the factors that regulate growth in this species is necessary if we are to design strategies to accelerate its growth. As a first step to understanding how growth is regulated in rabbifih. we purified rabbitfish GH, examined i t . growth pmmoting activity in the fry. and cloned its cDNA. MATERIALS AND METHODS Fish Immature rabbifih (Siganus gunatus: 150-200 g body wt) were purchased from fishpond owners in the northern Philippines. Pituitary glands were collected from heshly killed fish and immediately frozen in liquid nitrugen and stored at -80'C until use. Isolation and Purification Pituitary glands ( I g) were homogenized in 5 ml of 50 mM ammonium acetate buffer (pH 9.0) containing 5 mM phenylmethylsuUonyl fluoride. The homogenates were stirred for I h at 4°C and centrifuged at 12.000g for 30 min. The supernatant was applied on a Sephadex GI00 column (1.9 X 100 cm). Proteim wereeluted with 50 m M ammonium bicarbonate buITer (pH 9.0) at a flow rate of 15 d / h . Three-milliliter fractions were collected and the absorbance was measured at 280 nm. Appropriate fractions were pooled and lyophilued. and 10 pg of each fraction was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Laemmli. 1970). Resolving and stacking gel concentrations were IS and 7.5%. respectively. The gel was stained with 0.25% Coomassie brilliant blue (R-250) and the protein bands in the gel were visualized after destaining in methanol:acetic acid:water (2.5:1:6.5). For immunobloning, protein bands from SDS-PAGE were t r a n s f e d into nitrocellulose membrane and incubated overnight at 4'C with bonito GH antiserum (diluted 2000~).Immunoreaaion war detected using the ABC kit (Vector Labs. CA).Tn-o immunoreactive fractions ( G 3 and G 4 ) were further purifd by reversed-phase high-performance liquid chromatography (HPLC). Five hundred miaugrams of G 3 and C-4 faclions war dissolved in 0.1% vinuoroamic acid tTFA) and war analyzed by HPLC (LC800. Jawo. Tokyo. Japan) on a TSK gel ODS120T column (0.46 X 25 cm. panicle size 5 pm) with a linear gradient of 20-8046 acetonicrile in 0.1% TFA for 60 rnin at a flow rate of 1 ml'min and aolumn tempramre of l0%. Abmhnm was monitored at 220 nrn Individual peaks HW collected. lyphhed. and analyzed by SDS-PAGE and immunobloning. The N-term& sequence of rabbitfish G H m a s determined by automatic q-ing on a S h i i PSQl gas-phase protein sequencer as described by Rand-Weaver and Kawauchi (1992). cDNA Cloning Total R&A was extracted from I00 mg of frozen pimitary with Isogen (Nippon Gene. Tokp. Japan) following the manufacturer's protocol. The concentration of total RKA was estimated by masuing the absorbance at 260 nm (conversion fanor. I OD = IO pg RNA/d). First-suand cDRA was re\-tramoibed from total W A using the F i t S m n d cDSASynthesis kit (Pharmacia Biotech. Sweden). Rererse namaip uon was carried out in a 15.~1reaction mixture containing 2.8 p1 of total RNA (5 pg). 5 pI of bulk first-wand cDKA reaction mu. I pl of IS m41 dithie threitol. I pI (5 pg/pl) of Noddm:, b i n a l primer, and 5.2 pl of DEPC-treated water. S u degenerate sense primen and two anti-seme primers were synthesized (Espec Oligo Senice. Tsukuba. Japan) based on the mast w n w ~ ~ regions ed of the amino acid sequence of knoun t e l a GHs to clone the internal region of rabbitfish GH c D S A During PCR. the 50-pl reaaion mixture I I pl h t suand cDNA (template). 2 pl each of sense and anti-sense primers (final conc. 0.4 pMJ. I pI nucleotide mixture (0.2 mM). 5 pl of IOX PCR buffer (final corn.: 10 m!! Tris HCI. pH 8.8: 50 d l KCI: 1.5 1nk4SlgCI,). 1.25 units of Taq DNA polymerasel war subjected to 35 cycles of amplification: each cycle m i s t e d of I min and 30 s denaturation (94-C). I min primer annealing (40'C). and I rnin and 30 s primer extension (72'C!. Final emension (72-C) was for 7 min. . b o n g the 12 primer combinations rested. the sense primer GH-8 15'-CC(ACG'IlAT(ACn GA(CT)AA(AG) CA(Cn GA3'1 and the anti-sense primer GH-2R 15'-TC(AGTC) ACVC) TT(AG) TGC AT(AG) TC-3'1 yielded a PCR product of the expected size after gel electrophoresis. The rapid amplification of cDNA ends (R4CE) method of Frohman (1990) using the 5' RACE system kit. Ver. 2.0 (Gibco BRL, Gaithersburg. MD).was used to clone the 5' end. Two anti-sense primers; RF-GH-I (5'-AGA CTC AAC CAG TCG ATA GG-3') and RFGH-2 (5'-GAC AGC AGC TTC ATC ACA GA-3'). were prepared based on the nucleotide sequence of the internal region of the rabbitfish GH cDNA. These two gene-specific primers and an abridged anchor primer provided in the kit were used for reverse transcription and PCR. PCR conditions were the same as described above. Two sense primers from the 5' end of rabbitfish GH cDNA were designed. RF-GH-3 (5'-ATC ACA GCT GAA CTG CAC AA-3') and RF-GH-4 (5'-AGC AAG GAC TAG GAC CAG AA-3'). and used with Nodd m l s bifunctional primer as the antisense primer in cloning the FuU-length rabbitfish GH cDNA. Multiple clones were examined to guard against PCR errors. PCR produns were eleamphoresed on agamce gels and visualized by ethidium bromide staining. The DNA was extracted and purified from agarose gels using QIAEX II gel extraction kit (Qiagen. Germany). inserted and Ligated into pT7 Blue T-Vector. and t m formed in NovaBlue competent cells (Fr'ovagen. WI). The plasmid DNA was then exuacted from the bacterial cells using QIAprep Spin Plasmid Kit (Qiagen). Nucleotide sequencing was performed using an ABI Prism DNA Sequencer 373 (Perkin-Elmer. CA). Biological Activity The growth promoting activity of purified rabbitfish GH was tested in the rabbitfish fry. Forty-five fry (3-4 g BW) were randomly divided into three groups (15 fish/group) and stocked in separate 250-literfiberglass tanks. Two groups were given weekly intraperitoneal injections of rabbitfish GH at doses of 0.01 and 0.1 pg/g body wt. The hormone was first dissolved in a small volume of0.01 N NaOH and diluted with saline (0.9% NaCI). The third group was injected with saline solution and served as a contml. Fish in each group were individually marked by a small cut in the operculum lo follow growth in each fish. The hormone injection was given only until the fifth week and the fish were reared for 1more weeks. Feeding and water managements were the same in all groups.-~hangesin fish body weight and fork length were measured every week. Significant differences in body weight and fork length between the control and the hormone-maled groups at each sampling time were determined by r test after one-way ANOVA (Sokal and Rohlf. 1995). Percentage data were transformed by arc sine or square r w t transformations. where appropriate. before the analysis. RESULTS kolation and Purification The elution panern of the alkaline emwrs of rabbitfish pituitary glands on Sephadex Glal column has already been reported in a separate paper (Ay3on rraaL. 1999). From the elution profile. fraaiom G 3 and C-4 contained protein h d s of molecular weights becum 18 and 22 kDa. and these bands =me immunoreactiw with bonito GH antiserum. The HPLC profiles of fractions G-3 and C-4 are shown in Fi. I. \hen stained with Coomassie's brilliant Mue after SDSPAGE. the peak eluting at 32 rnin contained a band with molecular weight of about 19 kDa (Fig. W : rhts band cmss-reacted with bonito GH antisemm Fig. 2B). No immunoreaction was detected with antisalmon PRL and anti-flounder SL (data m shoam). The yields of pure GH from G-3and C-4 -ions \'ere 0.9 and 1.0 mg/g of pituitary weight). rpspectively. N-terminal amino acid sequencing of intM pmein gave no sequence indicating that the amino terminal is blocked. cDNA Cloning A cDNA fragment of 267 bp was amplified in the first PCR using the degenerate primers. GH-8 and GH-2R. This fragment corresponds to nucleotides 347613 of the cDSA (Fig. 3). The second PCR using primers RF-GH-2and 5' RACE abridged anchor primer yielded a product of 351 bp which spans from the 5' end and overlaps with the kno\%nsequence of the OM- ~ P 0 r, CI :: 5 9 SI 9 ~ n RG. I . HPLC c h w M t q l n m r of haaiom G-3 (A) and C4 (B). The lracrimn (0 5 mgl were dimlvcd tn 0 I'bTFA k d d a pminm ~ dGH and somatolarin 6L) is indicavd G 3 lraion was hry subjened to affinity chromatography to removt ihc $>+opmeirs FIG. 2. SDS-PAGE panern o l rabbihh CH aher staining with Cmmassie's brilliant blue (A) and aha M antirrmm (B). >I. molecular weight marker OrDa). I . nonteduced: 2. r e d u d uith p-mercaptoethaml. m p with barito CH GH-S ffiH-3 -* + 347 613 + GH-2R pajd irrerrrd don: I 860 t f i H full I& .5brMr1!~ dom FlC. 3. Sdpmaticdiagramshowing he p i t i o n d i h e IhmFCR poducu I p m i a l a d hll-kglhdmerl.TheNkd b m Rprrraurht@on for he m r w peptide and he open box. he signal peptide (sp) The unvlmlatd RT regions in the 5' and 3' ends m h n & primers d in ~ e r FCR y arr also indicated AAP Is 5' RACE abridged anchor primer 569 141 G A T C C U G A C A W I T G I A C G M C T ~ ~ C U ~ G U U T ~ C ~ D P R Q M T E L L A C F R K D M H K V E FlC. 4. The nudmtide sequence d rabbilhsh C H cDNA and ils deduced amino and seqThe signal pepode 118 Mluo U is underlined.The amino add residws are numbered Lu'ginning from he m a r w peptide. The IernWdOn mdan i s indjolEd b? an lnstrt T k Iw qsleine residues and ihe pol.yadenylation signal are in bold. 40 rabbltfish tuna Aslan seabass yellow tail bonito tilapia flounder chum ralmn I rainbow trout carp NSUUD.2 ----- (IT-E----- (IT-E---T-g--Proe- ----- -----CT-I--N--QL.--OX-LPER -&T-LPDER --D--LP-EI w rabbitflsh tuna Asian seabars yellow tail bonito tilapia flounder chum salmn I rainbau crout carp :a0 SCNSDYIRSP r-..--> ALIG'GS... s-s-.--.-. s-s-.---. I-- ..--.. ....... -...... . r--.. F------I-F------I-F------I-F------I.P-H- -NFL-F-.--S-V-F----S-V-F----.m- F-S 3.5 S-S F-VX'SF... ' YSL i-*SL S-S-N-YS:. 7--...- rabblcfish tuna Aslan seabssr yellow tail bonl to cilapia flounder c h m salmn I rainbox trout carp rabbitfish tuna Asian seabans yellow tail bonito tilapla flounder chum salmon I rainbad trout carp FIG.5. r) Alignmcnl of th? amino acid v q u m of rabbilhsh GH W b uhu leleas CHs Caps ur intiodurrd w squtnm alignmnL Dashes (-) indicate amim acid reridus identid lo ihm o l nhbitkh CH.The l o a r i m ofth? b e amino ddtl.F6ar unique lo r a b b i w GH u boxed. internal region (nucleotides 1-354) (Fig. 3). To obtain the full sequence. a third PCR was carried out using newly designed primers (RF-GH-3 and RF-GH-4) and Nofldm,, bifunctional primer. Excluding the p l y (A) tail. the rabbifish GH cDNA consisted of 860 bp. This was compased of 94 bp in the 5' untranslated region. 54 bp for the signal sequence. 534 bp coding for the hormone. and 175 bp in the 3' unbarslated region including the hexanucleotide M T M . a typical pol!adenylation site located 13 nucleotides upstreamof h e p l y (A) tail (Fig. 4). Thus. an open reading frame of rabbitfish GH cDh'A encodes a prehormone of 196 amino acid residues with a s i p 1 peptide of 18 amino acids and a mature protein of 178 amino acid residues. TABLE I Nuclemide and .Amino Acid S e q u m Identity olRabbiu%h CH Compared w i h Those of Chher Fish Spedes Sequence identity (%) Fish species Tuna ~TiluNlmL h m m ) h i a n seabas (Llsaim+aj Yellow lail lWda mmownd~~at ~ o n i t o(l(lk&& Xlapia ((hgcbnmirnilolicur) Flounder ( P a d d q s dinaur) Rainbow m r ( ~ a f n m gaid-ud Chum salmon (Oncwhww k-) Carp ( C p h n wpid Nuclemides Amino Id& 73.W 73 40 72 88 72.53 70.63 W.27 59.06 58.37 54-57 74.09 72 51 72.51 71.50 69.94 57.51 58.03 58.03 47.15 Nme. Tuna (Sato a d..1988): k i m seaban W b b a d..1991): y e U m tail Wlahiki rt d..1988): m u ,(Naro a 11. 1988); lilapia (Render-Delrueet d..1989): flounder PV~tahikia 11.1989): rainbow tmut (Agellon and Chen. 1986): chum salmon (Ekine el 11. 1985): rarp ( K m n ad..1989). followed by gel liltranon chromatography on Sephadex G-100 and rpHPLC. In each step in the p u r i h . tion process, rabbitfish GH was monitored by SDSPAGEand immunoblorting with bonito GH antiserum. The immunoreaction with bonito GH antiserum was specific: no immunoreaction was dececled with antisalmon PRL and anti-flounder SL. and the intensity of the immunoreaction was sh-ong. indicating a high concentration of GH in the e m . This can be exp&ed since GH cells are major constituents of the pituitary gland. The purified protein was ah confirmed to be GH bared on its molecular weight. Under nonreducing conditions. rabbitfish GH has a molecular weight of about 19 kDa. as estimated from SDS-PAGE This agrees with the mokular weight of M.4 W)a. From .-DI\'A The yield d GH 1.9 mg/g wl of gknd ann;at. the yield PRL was only half that of GH (unpublished resultsl. We tried to determine the amino acid sequence of the N-tmnind region d the intaa p u e i n but found it blaked indicating that the hm anino add ha the N-lennims is ppglummine This was cmfirmed w* the deduced amino acid sequence of the GH cDNA. We have also cloned the rabbitfish GH cDSA. The prehormone is composed of 1% amino aidr with a signal peptide of 18 amino acids. The marure hormone has 178 amino acids. slighrly larger (by 5 amino acids) than flounder GH, which is the smallest teleow GH (173 amino acids) known so far (H$tahiki a 11.. 1989). In rabbitfish GH. there is a deletion of 5 amino acids (Pro-TvyrGly-hpTv) in position 143-117 immediately preceeding the I 1 amino acid deletion in llounder GH. This deletion is unique to rabbidish GH since no such deletion was observed in any known fkh GHs (Fig. 5). This stretch of amino acid midues is ouuide the five conserved domains of GHs as denribed by Watahii ct d.(1989).The biological significance of this deletion is presently unknown. Single or double amino acid deletions are ah found in rabbitfish GH: amino acids 9.103- 104. and 157 (Fig. 5). As in the orher teleost GHs (Rand-Waver a d..1993). rabbidish GH has four cysteine midues located in highly conserved positions. Rabbitfish GH is closer in identity to GHs of other perciforms like tuna. sea bass. yellow tail. bonito. and tilapia than to those of salmonid and carp GHs. As is evident in Fig. 5. except for the highly co-pd amino acids in 170-180. the sequences of salmonids ,, > Sequence Compariwn The deduced amino acid sequence of rabbitfish GH was compared with those of other fish species (Fig. 5). The sequence identity is Listed in Table 1. Rabbitfish GH has four cysteine residues loeted at highly m d positions aa 51.151.168. and 176 (Fig. 4). Rabbitfish GH is highly identical (>70%) with GHs of more advanced telensu.suchasseabass.tunayeUowQil.andbonito.and less identialwith thcseof d m m i d s and carp GHr Biological Activity Rabbitfish Fry inpcted with the hormone at a dose of 0.1 pg/g body w? showed sigruficantly higher increase in body weight and fork length compared to the control starting on the third week of treatment (Fig. 6). T h e e inpcledwithO.01 pg/g body w d t h e hormone showed a siuficantly higher immwaRer the fiRh week Although the treatment was stopped aRer the fiRh week the siplicant immw in body weight and fork length in the hormoneaeated p p s we. mainrairrd until the seventh week or for 2 weeks aRer the last injection. DISCUSSION We have isolated and purified GH from rabbitfish pituitary glands by extraction under alkaline condition A -8- -E+ 25 i wnuol f f i H (0.01 pug B Y f f i H (0.1 pug BW) f f i H (0.0 1 BW) -E+ f f i H (0. I pug BW) 5 weeks FIG. 6. E R e n r o i f a b b i ~CH on growh in body weight (A) and iork length (B)dherabbithrh fry Fish rcrriwd a r o l a l dsix inmp&~oml injecilornevery 1 d a y . V e n i d lines represent SEhf 'Signifiwntd i l l e r e m IP < 0.05) compared lo ~k crmrml and carp GHs are clearly different fmm that of perciforms where rabbitfkh is classified. The purified rabbitfish GH was effective in promoting growth in rabbitfish hy. Although the hormone was given only every 7 days. its gmwth promoting effect was already evident after the third week in the fry treated with 0.1 pg/g body r~ and starting on the fifth week in those treated with 0.01 pg/g body wt. Interestingly. the significant effect of the hormone was still observed 2 weeks after the final injection. Considering that the hormone was given only once a week. administering the hormone daily using appropriate routes such as feeding would probably banslate into much larger gmu& increments. ACKNOWLEDGMENTS HC are grateful to Dr A. T a k M of Kitarato University for his invaluable suggesionr and technical suppon. ThanLs are alro due to Roger E M P. Mamauag and Pauor Jones Denusla of SEAFDEC AQD for help in pituitary coUenion. REFERENCES Agellon. L 6.. and Chen. T. T (1986). Rainbow vout gmwth honnm: Md& cloning ol cDNA and -ion in Ewhsirhi3 mLi m A 5.46-77. A p n . F C.. de Jezur.E C. T.Amemiya. Y. Morlyrma S.. Hinm. T . and Kawauchi. H. (1999). lrolation and cDNA cloning of vrmawlacdn i n rabbi& (Siganlrr g u r m d . Csl 6 m p mml115.292-300. Devlin. R. H.. Y d . T.Y. Biagi. C. X.. Donaldron. E M.. Sruamon. P. and Cham W K. (1994). E . u a u d i M ~ yYLmOn p*. Nan"? 371.209-210. Fmhmar~M. A (1990). RACE: Rapid ampliiauon ofc0NAMdr. la 'VCR Rolaolr: A Guide w M e h & and Applianom" N.A Innis. D. H. Celfand. J. J. Snimky. and T. J. While. -1. pp 2&38 Acackmic Rm.San Diego. Cuillm. I.. UR.. Agnmonte. A . M a a l a . R.. M a a l a A. Hernandez C. A.. hzquer M. M.. Diaz. M.. H m . M. T.. Alvarez-Lapnchere. L. Hemandm 0..and de la Fuente. J. (19983. Physiological changes in lhcjuvrnile ~lryhalinet e l ~ n tIhe . nhpia Ommhnmb homorwn. injected wilh E &erind homolqpm gmwlhhornme.1 Mar Bar& 6.142-151. ffiwauchi. H.. Ymyama. 5.. Ymda. A. Yamaguchi. K.. Yliraham. K.. Kubom. J.. and Himno. T (1986). lsolalion and d u r a s i n l i o n of chwn A w n gmwth hormone. Arch Biochm. B m . 214. 542-552. Knibb. W . Robins. A. C-, L. &zm J . H+n& A. md HCILt J. (1991) M o l r m k r c l m q and d Amanban black beam AmrlropqFa burmsiand b a m m d h fish gmwh hornme cDSA t&kg pol\chak~nrcm D N A S ~ W N ~ J ~ D M A ~2 121-1U. .\+ K m . Y. Sarid. S.. Ber. R. and Dank!. V (1989) Cap -rh hormane: Molecular doning and sqtanq d cDSX Gsr 77. Y)9-3IS. Laemmli. U. K. (1970) Cleavage d rvuctlrnl pmam d m q lhc assembly d k head d b a u i o p h q g T 4 . h Z Z 7 . 6 S l 4 S 5 Mdran. E. Romhddt. B.. %en C . and NyMuddin kIPP9)~ Canroinles~delireryd~andpaMh~w.qurult d telemo. A ~ N In. R ~1.247 Mmyarna. S.. Yamamoco. H. Sugimno. 5. Abc. T.W.T. and Kawauchi. H. (1993). O d adrmruYRnm d mnnbinn wlmDn g m w t h h o r m a n e ~ r a i n b m v ~ u ~ ~ & ~ ~ lurr 112 s 1 m . Noso. T.Yasuda A. Ka-. 1 .-T H.Tabhashi. .A. W, K.. and Kamauchi. H. (1988). b h 3 m and durxtarmian d @mwhhwmone fmm a nwim Tub bonito iKamwmn plunirt. Inr J~Rpti& RaWl R n 32.579-589 Rand-Weavu. M.. and Kauawhi. H. (19923 A Rpid pacdrrr for Ihe idation of bioaaiw p hhannrr Gsl Lw+ m~ 85.34-345. R a n d - h v u . M.. Kawauchi H. and Om M (1993) E w i m m d I h e ~ t u r e d i h e g ~ ~ w t h h o r n m e a &The nd~f~ E n d o c r i ~ d C m u t h D r n b p n a r m d ~ m N r w bates.' W.P Schreibmam C. C. S a m 6 md K T P.ng Eds). pp. 13-42, Aedanic Rar San Drgo. R a k - D d r u r . F . SuD.. P h i h w J. C. C k . C ~ Lm . M. Benrubl. 0.. and h*mul. J A (1989) T iw t h hmmm: M d r m k r d w of c0SA md upasa,in Gchf&a d i DNAbnl-nB H~ -la S.. Nmo. T . M a i y S. ~ Hiruo. ~ T. rd (1993). noundamhornme.lodmndrm-rd d e c L s ' m g r o v a h d j u u s l i * r a i n b m ~ a r6 .~ .e w. m. M. Sato. 5.. L 4 b ~ a d ~K. t M u n u K. M. Y. Kimura S. X w u b M.. and KimA. II-. Y o k c u h c b v l l g and nwleotide sqequurr drum p hhornme rDNA Bda A N 909.3542 S e h . 5. b h k a m i T. Nirhi T. Kuwnu. Y_ Luo A. h n U. Itoh. S.. and Kauawhi H (1985) C h q rd u p a r ~ a nd cDSA for A w n @mwh hnmm in Errkr*Bu miY ld. .W A d Sci US4 82. 4)(M310~ Sokd. R. R.. and Rohlf. F J (19951 "Bianmy- 3rd cd. F m m a n New Y a k U$W h4 . . Tanaka. M.. Wzsda X . Y d m M. Ymrda Y and Nahhirna. K. (19861 d J S A w a n d pi--umumd y*llw. la11(Saidr -a l1 pqmwlh homo"t Gsl Guq Guiarmd 70. IOI-406 Walahi*. .U.. Y m w m ~ o .M . Yanukana U . Tanalra M. and Yakarhirna. K 1140I ComaMd and u n t unimr p d rrvdus 1nlhedomuaoIihegmu%~hamMsJ&dChZLI312-316 .General and Comparauw E n d a m l o g y 126. 165- 174 (2002) do1 10 1006gcen 2002 7788. awlable onllne ar hrrpll-w ~ddnbrar)corn on 1BELL@ Differential Expression of Insulin-like Growth Factor I and I1 mRNAs during Embryogenesis and Early Larval Development in Rabbitfish, Siganus guttatus E v e l y n Grace T. de Jesus.' Shunsuke M0riyama.t Susumu Hyodo.S Bruria F u n k e n s t e i n . 5 Arieh Gertler.1 and Hiroshi Kawauchif Felix G. Ayson.* 'Aquaculrum Depanment. Soulheau kian Fisheries Lkdopmenr Center (SEAFDEC AQDj. Timuan 5021. h k . Philippine: thborarory of Molecular EndocMolw. .k/xml dFirheries Sciences. I(I~arat0U~JWNN. hnnlru. Inare 022-0101. J a m !hborato,y dPhjsiorOSy. &an Research ~ m l u t c univcmy . ofrohp. %kam. Tokw. Japan. §Dppanmenr d.Uark B i o l w and Biorechnalogy, brael Oceanographk and Limnobpcal Rsearch. Tel-Shiknmna. Haifa 3 1080. luael: and lDeparmwn1 oflioc'lemiruy. Fmd Science and Numuon. Famiry of A ~ M t u r e The . Hebrew UniveWr). ofjenrralem. Rehovor 76100. hael Accepted January 17,2002 In rodents. the expression of insulin-like growth factor I1 OGF-11) is higher than that of insulin-like growth factor I OGF-I) during fetal life while the reverse is true after birth. We wanted to examine whether this is also true in fish and whether ICF-I and IGF-Il are differentially regulated duringdifferent stages of embryogenesis and early larval development in rabbitfish. We first cloned the cDNAs of rabbitfish IGF-1 and ICF-11 from the liver. R a b bitfish IGF-I has an open reading frame of 558 bp that codes for a signal peptide of 44 amino acids (aa). a mature protein of MI aa. and a single form of E domain of 74 aa. Rabbitfish IGF-II, on the other hand, has an open reading frame of 645 bp that codes for a signal peptide of 47 aa. a mature protein of 70 aa, and an E domain of 98 aa. On the amino acid level. rabbitfish IGF-I sharer 68% similarity with IGF-n. We then examined the relative expression of the two lCFs in unfertilized eggs, during different stages of embryogenesis. and in early larval stages of rabbitfish by a semiquantitative reverse transaiption-polymerase chain reaction. Primen that amplify the mature peptide region of both lGFs were used and PCR for both peptides was done simultaneously. with identical PCR conditions for both. The identity of the PCR products was confirmed by direct sequencing. Contrary to published reports for seabream and rainbow trout. IGF-I mRNA was not de- tected in rabbitfish unfenilized egkr; it wns hexpressed in larvae soon after hatching. ICF-n m m A . hever. was expressed in unfertilized cggs. a b i t w d l y . and was already strongly expressed during the c l u n g e stage. m R N k for both peptides upre expmsed in the larvae. although IGF-U mRWA expression was i. .s-r *A higher than IGF-I expression. c ~ cb Key Words: rabbitfish: IGF 1: IGF-Il; cDNA clcming: mRVA expression. INTRODUCTION Insulin-like grow.th factors I and Il OGF-I and IGFUJare two highly homologous mitogenic peptides that are expressed ubiquitously and show diverse effects o n development. growth. a n d meabolism (reviewed by Rosenfeld and Roberts. 1999; Werner and LeRoith. 2000). The c e n m l role of ICF-I is that it regulates postnatal growh and mediates the growth-promoting hormone (GH). although smdies aka action of g r o ~ t h show that it is important for normal embvonic growth (Baker er a/.. 1993; Powel!-Branon u al. 1993). The physiological function of IGF-U. on the other hand, is not yet definitely known, although it is considered a fetal growth factor based on the evidence in mice. where a significant growth retardation was observed in pups born with the IGF-I1 gene disrupted or knocked-out @echiara et al.. 1990). The biological actions of lGFs are mediated by IGF cype I (IGF-I-R) and II (IGF-ll/mannase 6-phosphate-R) cell surface receptors. Interaction with IGF-I-R is mainly responsible for the mitogenic effects of IGF (LeRoith et al., 1995) and there is enough consensus at present that much of the biological action of lGFs on growth and dilferentiation is mediated by this receptor. The role of type 11 IGF receptor in IGF signal transduction remains controversial (Kiess. 1999: Nissley. 1999). In addition. there is the insulin receptor (lR)/IGF-I-R hybrid which seems to be widely expressed and even appears to be the most abundant form of receptor in certain tissues (Baiyles et aL. 1997) but the specific differences in signaling characteristics in the IGF-I-R and the IR/ IGF-I-R hybrids have not yet been established. Furthermore, at least six insulin-like growth factor binding proteins (lGFBPs) are already found in circulation and they modulate the biological activity of lCFs by controlling the availability of free lCFs and by prolonging their half-life (Hwa et d.. 1999). These IGFBPs show contrasting effects on IGF function. IGFBP-I (Van Buul-Offen et al.. 2000). ICFBP-2 @uan et al.. 1999: Hoeflich et al., 1999: Wolf et al.. 2000). and ICFBP-3 (Hockheid et al., 2000) have inhibitory effects on IGF actions, whereas there is evidence that IGFBP-5 functions as a growth factor (biiyakoshi et al.. 2001). To accelerate development and improve growth of rabbitfish (Siganusgunatus), an important aquaculture species in the Southeast Asian region. we have been studying the hormones in the pituitary-liver axis of this fish. We have isolated and cloned rabbitfish GH and found it potent in accelerating growth in the fry (Ayson et al.. 2000). In this study. we cloned the cDNAs for rabbifish IGF-l and IGF-I1 from the liver and examined the expression of the mRNA of both lCFs during different stages of embryogenesis and during early larval development. We report here that IGF-l mRNA was not detected in unfertilized eggs but was fist expressed in the larvae scan after hatching. ICF-11 mRNA, however. was detected in unfertilized eggs. although very weakly expressed, and was already strongly expressed staning from the cleavage TABLE I Primers Used in Cloning Rabbi& ICF-I CORA I. Degenerate primers for doning he lrvcmal rrlpon IGF-IF 5 ' -GG (CTG) AT G T C G !TC: GAGTGCTGC- 3. IGF-ZF 5' CAGAC (CCll CT {CGi TC in,CCCX LC?.-\>-3' IGF-IR 5' -CC (AC) CA (GA) TACAT ICT! TC LCCn .AC- 3' 11. P r i m for cloning he 5' end' rllCFI-IR 5' -CAXACTGCACCCTGTCCACC- 3' rflCFI-2R 5 ' -IUACCCTCTCTCTCCAC.ACAC. 3' UI. Primm for cloning ihc 3. end' rllCF1-IF 5' - C A T C T C T A G C G C T C m C C m t - 3' rflCFI-ZF 5' -CTCATCTCTTCAACAGTCCG 3' rflGFI J F 5' -ACCAGTGCTCCTTCCAAAGCTG 3' N. Prime, used for RT-PCR K R pmduct size = 151 bp! rflGFln-F 5' -TCTCACTACTUTTCTCCATC- 3' rflGFlfl-R 5' -CTCTTCCCTCGACTTCACTT. 3' (m ' The abridged a n c k prima pmvikl in he 5 - RACE kmt w a s used as ihc forward primer Nod-dm,, b~functionalprimer w a r used as ihc mau prum ' stage. Similar to those observed in other venebrates. IGF-U is expressed more than IGF-1 in rabbi& embryos. MATERIALS AND Ml3HODS Cloning of Rabbitfish ICF-1 and lGF-U d ) N h The procedure of Ayson ef al. (2000) in cloning rabbitfish GH was followed with some modilkations Briefly, total RNA uas emacted Fmm the liver with Isogen (Nippon Gene. Tokyo) and h - s p a n d cDN.4 was reverse-transcribed from 5 pg of coral RNA using a First strand cDXA synthesis kir (.Arnersham Phar. macia Biotech. Sweden). Degenerate sense and antisense primers (Tables I and 2) designed based on the consened sequences of the B and A domains of known lCFs were qnlhesized and used to clone the internal regions of rabbifish IGF cDNAs. A 50+1 PCR mixture using G e d m p ' s .Amplifaq Gold kit (Perkin-Elmer. Palo Alto. CA) icas subjected to 15-min hot start at 95' and 40 qcles of amplification: each cycle consisted of I min 30 s denaturation (94"). I min primer annealing (45'). 1 min 30 s primer extension (72'). and a final extension of 7 min at 72'. The 5' rapid amplification of cDNA ends (RACE) method of Frohman (1990). using a 5' RACE s>Ttem ' ICFs during Early Dwelopmenr In Rabbilfish TABLE 2 Primers Used in Cloning Rabbidish IGF-I1 cDNA I. Degemrate primers for cloning the internal region IGF2-IF 5' -GA (AC) ACGCT (AG) TCTCC (AG) GCA- 3' IGF2-IR 5' -CCCACAGTACTGTCCA (AG)- 3' ICF2-2R 5' -ACAACACTCCTC TTC) ACCAT (AC) CC- 3' 11. Primers for cloning the 5' end' rflCF2-IR 5' -TTCACCTCACAGCTACGGAA- 3' rflCF2-2R 5' -CTCCTCCTTGGCCT.4CTGAA- 3' UI. Primers for doning the 3' end' rflCF2-IF 5' -TCACTACTCCCATCTCACATG 3' rflCF2-2F 5' -ACTCACmCCCACACCTCCCG 3' rflCF2-3F 5' -CACAACCGTCCCATTCTACAAC- 3' N.Primers used f w RT-PCR (PCR pmducl size = 452 bp) rflGF2R-F 5' -TCACCCTCTACCTTCTGCAA- 3' rflCF2fl-R 5' -CTGTGCAACATTCCCTCCTC- 3' 'The abridged anchor primer provided in the 5' RACE lut war used ar the forward pnmer. Mod-dm,, biiunctional primer was used as the R V e m primer. ' kit, ver 2.0 (Gibco BRL,Gaithersburg. MD),was used to clone the 5' end of the cDNAs. Antisense primers (Tables 1 and 2) designed based on the known nucleotide sequence of the internal region of rabbifih IGF-I and ICF-11were synthesized. These gene-specific antisense primers were used together with the abridged anchor primer provided in the 5' RACE kit in reverse transcription and PCR. PCR conditions were the same as described except for the annealing temperature (50'). Sense primers (Tables 1 and 2) were designed based on the known nudmtide sequences of the 5' end of rabbitfish IGF-I and IGF-11. and these were used with Nod adapter primer as the antisense primer in cloning the 3' end of IGF cDNAs. Takara's LA Tag kit (Takara. Japan) was used during PCR. PCR conditions were the same as described except that the 15-min hot scan (95') was omitted and the annealing temperature was raised to 55'. PCR produrn were electrophoresed in agarose gels and visualized by ethidium bromide staining. The DNA was extracted and purified from agarose gels using QIAEX 11 gel extraction kit (Qiagen. Germany). inserted and ligated into pT7 Blue T-vector (Novagen. WI: for partial clones) or TOP0 XL cloning vector flnvitrogen. CA: for full-length clones). and transformed in JM 109 competent cells (Takara: for partial clones) or TOP 10 chemically competent cells (Invitrogen; for full-length clones). The plasmid DNA was then extracted from the bacterial cells using QlAprep Spin Plasmid kit (Qiagen). lrjudmtide sequencing was performed using an AB1 Prism 377 DSA Sequencer Perkin-Elmer). Expression of IGF-I and IGF-U mRV& during Embryogenesis and Early Larval Der~lopmentin Rabbitfish Unfertilized eggs, embrym at different stages of stage). gastrula (508 development (cleavage (32-1~11 epiboly), neurula. during tail bud elongation. and embryos with heart beat), and larvae from hatching until 6 days old were collected. Unfertilized eggs meereremoved from the ovary by cannulation I day before the expected date of spawning. Because of the v q small size of the eggs and larvae of rabbifih, samples from each developmental stage were pooled. immediately put in dry ice. and later stored at -80' until analysis. Total RNA from the pooled samples (100-300 mg) was extracted and first-strand cDN.4 was reversetranscribed from a b e d amount of total RNA (5 rg) During PCR, I pl each of first-strand cD%A was used as the template in all samples. Primers (Tables I and 2) that amplify the mature peptide region of rabbitfish IGF-I and IGF-I1 were synrhesized and used. PCR conditions were as described except for the annealing temperature (58') A PCR using first-strand cDNA from the liver as the template and one without a template were also included as positive and negatiw control. respectively. The identity of the PCR produns was confirmed by direct sequencing. Rabbitfish pactin was used as the internal standard during quantification of IGF-I and IGF-U mkVA expression. Using degenerate primers designed based on the sequences of known teleost pactin. rabbitfish pactin was first cloned and. from its partial sequence. sense (5' -GCC-AAC-ACC-CA%-.4.4G-AC- 3') and antisense (5' -RG-CCA-ATGGTG-ATG-r\CCTG- 3') primers were designed. PCR conditions were optimized by first checking the Linearity of the PCR and. based on the results of the linearity check.a PCR of 40 cycles was used for all samples. PCR products were loaded into the gel and stained with erhidium bromide. Quantification of PCR products m-as done using GeneGenius gel documentation and analysis system (Syngene. CK) \'slues for IGF-I and IGF-I1 were normalized with that of 8-acu'n. A U G D C r r r U W r U G ~ W O U ; t T A r U X T - ~ R G F Y F S K P T G Y G P N S R R P R G ---.----------------------! I - - - - - - - -'-,,-in .. , 4>5 4; XAGAZZ~GTCTr~CMAU:TC'17~:~TCC-XWC-;T&77A TtTT?TiYXTIUGGC&TiA?G4TGRA - r e s t o f 3 ' -Itrai.sla:e2 re;l=r 5: 5 - 8 56 316 -------- . -- FIG. 1. Nuckolide and d e d u d amim acid sequencpi drabbitfuh prrpmlGF-I. The signal peptide (44a)is underlined. the mwurr ~ g a n (68 a)mnsisung d B . C. A. and D domaim. is in boldfaced kom. and the sir Cys m i d u a d l h e nrature peptide am double undcrlmcd. T k E domain is composed of 74 a.The amim acids are numbered waning from the malure p p t i d e ~ RESULTS determined by 5' and 3' RACE. respectively. By connecting- the sequences of the overlapping- clones. the nuclmtide sequence from the start of the 5' end through some portion of the 3' unuanslated region of rabbitfish IGF'I cDNA war obtained. From sequence. rabbitfish preprolcf-l war found ro consist of 186 amino acids iaa) (558 bp) (Fig. 1). This is composed of 44 aa for the signal peptide. 68 aa for the mature peptide comprising B. C. A. and D domains. and i 4 aa ~~ Structurr of Rabbihh IGF-I and IGF-U cDNAs The nuclmtide sequence of rabbitfish ICF-1 cDNA was obtained in three steps. The sequence of the B and A domains was detemined using degenerate primers based on the highly consewed region of the peptide. The unknown sequences in the 5' and 3' regions were this ICFs during Early Development m Rabb~lhrh 803 865 -iGAG-~T~TTAT-T5TCT,7WGTUTTTCrCT~ GTCCICTTTGACCTCCCTGC - r e s t of 2' c : r u s l a t e d regicr. 564 FIG. 2. Nudeolide and deduced amim acid wguenm olrabbilfish preprolGF-11. The s i p a l pptide (47 aa) u undaiincd. he r m m Rgm (70 aa) is in boldlacld leners El. C. A and D domains). and the six CF midues d ihe malure ppide are doubk u r d u i d . The E d o m is compovd o l 98 aa. The amirm x i & are numbered sraning from thc mature peptide. for the E domain. compared with the ICFs from 10 fish species (Fig. 3). Similarly. we found that rabbitfish preprolCF-I1 is 215 aa long (Fig. 2). The 645-bp open reading frame is composed of 47 aa for the signal peptide. 70 aa for the mature peptide (B. C. A, and D domains), and 98 aa for the E domain. compared with the ICF-11s from six fish species (Fig. 4). Expression of1GF-1and IGF-U mRNAs during Embryogenesis and Early Larval Development in Rabbitfish mRNA for IGF-I was first detected in lanae soon after hatching but not in unfertilized eggs and during embryogenesis (Fig. 5A). In conoast. IGF-n mRVA was already detected in unfenilized eggs. albeit with rabbrtf~sh glltkead seasream black seabrear barranundi sculpro f launder Lllapia salmcr. ra:nbcu trout carp goldfish rabbrtfrsh grlthead seabceam black se&reaa bar:zc"tbra sculp-n flounder' trlapra salacn* :ark% troui* carp gold€ rsh -. - - - - - -- - - - - - - - - - -. FIG. 3. Alignment of the amim acid sequence of nbbilhsh ICF-I with the lCF-Lr of 10 fch rpecia Caps 0 are mmdxal to n u n m m the sequence alignment. Dashes (-) indicate amino aid residues identical to Ihose of rabbllfish IGF-I. *: gmhkd SFMU w (Duguay s 11. 1996): black seabream. kmlhoplgna schkgcli (GenBanL A m i o n No. AFUW.73). barnmundi. b olcrilCI 1Xmh.h Stahlborn a d . 1999):d p i n . Cot- saxpius RoFIbg.Cueni U A. 1998): flounder. Rntichrb~dio~ivmafTawdka a 11- 1998): dhpu Ordmmsb m s m b i m Reinecke s 11. 1997): salmon ~ y n c h u Ururch s ( G o U 11.1989): rainbow ~ r o O um ~ rhp3m IPltmbbtt a d thcn 1992):carp. Crprinur orpio (Liang s 11.1996);goldhsh. Clnoiw aunm (Kermwnl s 11. 1998). ' Domaim B. C. A and D OmsNuIc I hMlW peptide: ' more than one form of E domain is found. low expression; the expression was already stmng starting at the cleavage stage and remained so throughout embryonic development (Figs. 5A and 58). In larvae up to 6 days old. both lCFs were stmngly expressed, although ICF-I1 is expressed more than ICF-I (Figs. 6A and 6B). DISCUSSION Rabbitfish prolCF-I and prolCF-I1 cDNAs contained the B. C. A. D. and E domains. typical of ICFs. The mature peptide of rabbitfish ICF-l and ICF-TJ have the six conserved Cys residues: Rvo in the B domain and four in the A domain. The B and A domains of ICF-l have the highest sequence homology in several fish species (Fig. 3). Except for a single amino acid substitution in tilapia. the A domain of the 10 fish species in Fig. 3 has 100% sequence homology. The B domain of rabbittish ICF-l has a rhreonine that is lacking in other advanced fish species (Order Perdfo-) and its C domain lacks histidine and asparagine. similar to the C domains of the perciforms. .Amino acid 9 in the C domain of rabbidrsh ICF-I is proline: this is serine in all other hsh species. The mature peptide region of rabbitM ICF-Il (70 aa) is 2 amino acids longer rhan rabbidish IGF-1 (68 aa). After aligning rhe six curnerved Cys residua of both ICFs, the amino acid sequence homology of rhe mature peptide region of rabbidish ICF-I and IGF-U is 68%. Like ICF-I. the B and A domains of ICF-U in most species are highly conserved (Fig. 4) but. in addition. the D domain of ICF-Il is also highly conserved. The amino acid sequences of both A and D domains of ICF-I1 in hsh species known to date haw 100%homolowThe E domain of ICF-l is, however. highly variable and more than one form can be found in some species: two in humans (Rotwein. 1986). four in salmonids u! pauawp l w y s e . ~r h y w 1-331 .uo!le~y![dure j o 00)Jaye paualap IOU 5e.w I! l e w ~ o ha^ l os SEA\ s o ~ q w apue ~ 8 % paz!lwajun qsy~!qqer u! V ~ ~ U 1 - 3 3 1 ~ 0la.ra1 a q 'Llqeqad ~ -qsaua8ohqwa j o sa8els waJaU!p aql u! pue q q v q q w j o sBBa p a q w a j u n u! p a w l a p IOU s e . ~v h w 1-331 ' ~ 8 %pazgwajun u! pauawp 1-331a J a w 3 6 6 1 'uq3 pm auaaJ3) ~non .noqlrleJ pue (9661 ' - pIJ uialsuayund weaJqeas p e w [ @ u! a w u n v y ! q q e r j o a w l w u! pue . q s a u a W q w a j o sa8ers IuaJaNp 8 ~ n ' p s a a paz! -1rslapn u! ldl~suen 1 - 3 3 1 ~ 0masa~d aql pau!urexa a.n 'Xpnls luasa~d41UI q n p e a w U! ueql s o h q -wa a c ( ~u! ram01 Llleraua8 F! U O ~ M U ~ J U O J !SI q8noql -It!.luawdola.\ap J ~ 0 h q w ap W O U J O j 1lmuodUlj osle q 1-331 amp .+\w d d e 11 .(EM[ '.re JJ u o l x e ~ ~ .Ila.\\od :EM[' . ~ I n! J ?) ~ u a u r d o l a ~ a pue p qwo.18 lmaj p a ~ d w !peq 1-331 8 q ~ e al ! w l e w mnoqs sew I! lpun qswa%o.hqwa 8 u w p ~ueuodwIOU q 1-331leql 1q8noq1 LlmowaJd mm I! '(6661 "re JJ rnvu asnow u! ~ u a u ~ d o l wpue p qw0.18 l e u o u JOJ Iuaplyns F! s a w ~pedaqucww a y 1-331 1mp pamoqs &rus IUe @ n q p 'uaw w ~ W T P u!~1-331 J >pedal( uo ~3 j o a~uapuadapaq, j o ammag 'qsy u! osle l n q slewwew u! Lluo IOU Kuew u! pauuyuo3 uaaq a w %u!le!paw U! Jseq ~3 j o uaya ~U!IO~OJ~-~IMOJS pue q w o B jeletumd 8upelnBaJ u! 1-331j o aloJ a u -ap~daa dnleu a w w o ~ paAeap j s! I! Jalp r ( l ~ p ~ le~@olo!q e ssmd 01 I! ~ oJOj t g w o p 3 a w ir! s u o j aldglnw j o w o d a ~ ou aJe aJaw .JaAamoq '11-331 JOJ xwaaks sno%lo - m a q u! uaAa r(lmp~eJ N ~ ~ O I ! ~m o d u p m o p 3 j o s ~ u ~aoaj ~ ws.lnon w o q w j o sappdad ~ u e u ! q w o m I ~ p a uIo d a ~(6661) .re JJ ue!L pue 'palep!we uaqm ~!ua8ol!ws! 91-331u m n q o fflewop 3 a w j o appdad ~ p a w u K se l e w umoqs aAeq (1661) 'p JJ p a u j 8 a t ~ 'r(lmp~ep~@olo!qawos ssassod osle L a u .(g661 "re JJ LenBna) 8u!ploj IJ~JJOJ 10'8u!ssa>o~d'uodnren Jelnl -laJesu! J O ~~ u e u o d w !aq Lew u!ewop 3 aql .appdad a w j o ~ s atp a ~w 0 4 a8meap !SI 01 ~ o u dimp pz!s - a w d q s! 11 .sanp!saJ ee p ~seq s q pue qsyl!qqel u! punoj q u!ewop 3 j o u o j auo ,41110 '(8661 " p JJ q e u q :6661 'PJJ w o q l v l s ilnl(ulX) p q pue z q uourles 01 8u!puodsauo~ JapunoU pue Ipunureueq ir! o w Pue ' ( ~ 6 6 1' U ! I A ~ PW ~!I@M :6861 '.pJJ oe3) .apodad un~m a(, amlmum nnewop a pcre 'v '3 ' 8 , -(z&I 'uaq3 pue 5 m n v :(SZZL~X -oh: u o F w a v i(uesu3) e u l W n w g .uourles w n v :(MI 'PP " ~ ~ ~ - ~ u w o - I ) 1rqq-61 mdma -3 u d m -us1'Pr w3)"Qs?m.pcw . q d e n ~:Us61 '.P P ~119) m j w l o S J J '!prmurerreq ~ :(9661'-FP ken%) 01 pnuap! s n p ! s l p!x w ! w a l w w (-) a q s w 1 v a d ! p amanbx np~yllls :nmqI I - ~ W!WU ~I p -arwdzynl ns JO ~ I I Jw ~qllm I n-rl31w ! q q = JO axanbas PW w.wa(,JO IU~&!IV -b atp anumvl m panparrn u ~~ :no;a n c q 3 r e 1 u q e s mq3 --q-m-------------------------------------------~-3--d-q-q---~-----------. u1a:n3s .-------~------~-yi-y-------------------------------~-~~--~-~---------------e r d e j?a -----.---------------------------------------------O--d-q-l----.----------:pKnlrrr=eq --------------.---i----..----------------------.---~----4-~d----l---------w a r w a s peag2:yb ~ W ~ ~ ~ ~ ~ : ) I T I C I ~ . W W ' . I Y ~ S W ~ W ~ X ~ W S U ~ ~ ~ S A ~ ~ XY s~? iYl ? ~ q qW Pl ~ I A ~ S ----~--------A------.;-------------------y----------.--d+~-~~-~--I--------- ----~--------.~------~-------------------~----------.--d-ao-~i-~~-~--------- . .. II ILI qsy~!qqqu! luaudo~a~aa 4lqj h m p 1-31 A IGF-I FIG. 5. ICF-I and ICF-U mRNA expression during rabbi* embygenesis determined by semiquanlilalive KT-PCR. (A) Agamse gel elecuuporesis of RT-PCR produrn lor ICF-U and Baclin. No PCR product for ICF-I: m. marker 4. u n f e n i l i eggs ( v e q faint band la IGF-In; c. cleavage sage (32ceII wage): & g m k : n. neurula: [be. tail bud elogation: hb. embryo wilh heart k t : nh. newly hatched larvae. El) ICF-I1mRNA levek exprezvd a the ralio of ICF-U and Banin. Labels are similar to chat in A. rabbitfish Larvae soon after hatching. In their studies. Funkenstein et al. (1996) and Greene and Chen (19971 used a higher concentration of total RNA during the reverse transcription step--20 pg. which is fourfold higher than that used in the present study. On the other hand. IGF-I1 has been postulated as a fetal growth factor. Support for this hypothesis came when targeted disruption of the IGF-11 gene resulted in fetal growth retardation-mice lacking the IGF-U gene were 40% smaller at b i i than their normal counterparts Oechiara et al.. 1990). The reduction in the growth rate in mice lacking the IGF-I1 gene is confined only in utero, clearly demonstraring the dependence of embryonic development on IGF-11. In addition. postnatal overexpression of IGF-II does not result in increased somatic growth, whereas mice overexpressing IGF-I are larger than the normal type and showed signs of overgrowth after 3-4 weeks of postnatal life Wanhews et al.. 1988). These findings and the generally higher levels of IGF-Il in the embryos compared to adulrs in many species argue for irs importance in early lie. The observation that IGF-Il expression is higher than IGF-I expression during early development seems to be true also in fish. as shoum by our present resulrs. IGF-I1 mRNA was strongly expressed in rabbitfish embryos during the cleavage stage. whereas IGF-l mRhlA was first detected only in the lawae soon after hatching. Furthermore. IGF-II mRXA was also detected in rabbitfish unfedized eggs. albeit weakly. whereas ICF-I was nor. Different obsenations were made by Greene and Chen (1997.1999) who found not only IGF-11. but also IGF-I. mRNA in rainbow trout unfertilized eggs. Nevenheless. the absence of PCR products for IGF-I in unfertilized eggs and during embryogenesis in rabbitfish does not imply pmr quality of RiVA from these samples or low efficiency in reverse transcription because a PCR product for rabbitfish pactin was amplified in all samples and there was no apparent difference in their expression in all stages. In the larvae. although ICF-I1 mRUA Iebvls R u i n 1ISb- FIG. 6. ICF-I and ICF-II mRWA uprrnm during ca+ Land development in rabbilfuh d n e n n i d by mqwutatn~ RTKR (A1 A ~ a m w gel elecvopmis 01 KT-PCR poducu lor ICF.1. ICF-U. and baain; nc. ncgatlr? conlrol: liver. ~ I U M & IB &F-I and ICF-I1 mRYA lewk expressed as ihc raD u1lh Bxtin were also higher than those of ICF-I during the first 6 days after hatching. the reverse is true in the adult, as shown in the higher level of ICF-I mRNA in the liver. Overall. the findings of the present study seem to agree with the concepts that ICF-II is probably more important than ICF-I during embryogenesis and that IGF-I exerts more influence on growth and development in later life stages. ACKNOWLEDGMENTS We lhanlr Dr. A. Takahashi of Kitasato University and Dr. Y. Takei of Orran Research Institute. University d Tokyo for heir helpful advice and s u g g m n s . F.G.A. is grateful to the Japan Society for the Promotion of Science for the powdatoral fellowship and for the financial suppon in the conduct ofthe wudy. Thu work was supponed in pan by USAID (Grant No. TA-MOU-99C19-001. Program in U S - h l Cooperative Development Research h gram. Emmmic Gmwrh). We also lhanlr his. F. P e d m and E. M. w a g of SEAFDEC AQD for their arskunce. REFERENCES Ayson. F. G.. de Jesus. E. C. T.. Amemiya. Y.. Moriyama. 5.. Hirano. T.. and Kawauchi H. (2000). Isolalion. cDNA don@ and gmwth pmmxing aaivity of rabbirhsh ( S i g a , ~gumno) gmwrh hormone. G n Cmnp. E d u f n d 117. 251-259. Bailyes. E. M.. Nave. B. T.. Som. M. A.. Orr. S. R.. Hayward. A. C.. and Siddk. K (199T). Insulin rccopior/lGF-l rcmp~orhybrids are widely disuibuted in mammalian h a : Quancitielon of indlvidud -pa species by deaiw immunopreripilation and immumbbtting. B a h . 1. 327.21?-215. Faker. 1.. Liu. J. P.. Robenson E. J.. and Efwnriads. A. 11993). Rok of insulin-like growrh factors in embryonic and post-natal growth. Cell 75.7382. Cao. Q. P.. Duguay. 5. J.. Pliselskaya. E.. Steiner. D. F.. and Chan. 5. J. (1989). Nucleocide sequence and gmwrh hormone-regulated expression of salmon imulimlike g m w h f m o r I mRNA. Mol. ~ - 3.200s-2010. d . Chen.1. Y.. Chang.C. Y.. Chen.J. C.. Shen.S. C.. and Wu.1. L (1997). Production of biologically aclive recombinant tilapia imulin-like gmwrh factor 11 polypeplidc in Esdvichia mli celh and characterization of thc gemmic wrucrure of the coding region. DNA GI1 Biol. 16. 683-892. Collet. C.. Candy. 1.. Richardson. N.. and ha. V. (1993. Organization. sequem. and expmsian of the gene encoding ICF-II from barrarnundi vel-teii: Lars darifed. Biahem. G e m . 35. 211224. X. and Robatrmr € J. (1990) A gmwThdekienr). p h e m ~ in p heluoqgour mice caming an insulinlike groulh faaor 11 gax dipUped by t a r p i n g .blurt 345. 78-80~ Duan. C.. Ding. 1.. L.Q.. Trai. W.. and PK (1999) hs&rAhc gmnlh facror btnding prolcin 2 is a v l h inhibitcry prolnn conserved in zbrafuh. h ,Vad A d Sd C S 4 S l5?i1-I5279 Duguay, 5. 1.. Lai-Zhang. J.. Stelm. D F~.Funkcrwem. B. and Chan S.J (1996) Dev~lopmntaland -ted upuskm of ICF-I and ICF.11 m R V h in Spuur .unuI .Wd EndDoud 1 6 123-132. Frohman. M. A. (1990). R4CE Rapid amplikakm d c D N A mbr In 'PCR R o l a o h : A Crude to .W&ods and A p p k a l m ' 1x1A. Innis. D. H.Celfand. 1. J. Sninsky. and T~J~ Blulc. Edrl. pp 28-38. Academic Rm San DKgo. Funkenstein B.. S h u n r . R. Amuly. R. and Cohcn 1 (1996). Nucleotide s e q u e m of the pomu ~ g d mSpana wb i m ~ I u r WregmwrhfacurIgaxandorpmmdLGF-Im-and embqm. .Wd. M u Bid Bkmaolnd 5.1351 Gmne. hi. W.. and Chen T T 11993Ternponleqnxmn1 patern of i n s u b l i k e gmnlh fmor mRFA d w m g ernbr)au &*bp ment in a telearl rainbow uout 1 ~ n c I+Sh A M .\k Bid BLurdupI. 6. 144-151. Creene. M. W.. and C h T. T. (1999). Q u a ~ u i m dlGF.1. IGF.11. and mulliplc insulin rrrrpa family manber msarga RV.b during embrpnic devehpmn~in nidmw u o u ~S4d F+d Dn. 51.348-361. Hahscheid. R. Jaqua. G.. and W e g n u n n B. 0 . T d e a b md h u m n imulin-like growh f a n a - b u d i n g parin 3 gene inhhu cell grmnh and tumorigenicir)-. A cell culnne m d for b g cancer. 1. EndDcriml. I& 5-563 Hoeflich. A.. Wu. bl.. Mohan 5.. F d J ~ .Wankr. R. F MT. h l d . G. I.. Lahm. H. Kolb. H J ~ .and WOK E. 11999) arm. -inn d i d i n - l i k e pmmh f a a - b * -2 in O a m . L I genic mice r e d u a r p o w ~ body d -ugh gain 5188 -54%. Hwa V.. O h Y.. and R d c l d . R G. 11999). Tk ~ ~ I i k c j p m r I h factor binding prolein (IGFBP) suprfamih.. F&DI b-. M 761787. K e m i . A.. Mahmoud. 5.5.. W q S . Mokney. 51. and Habib. H. R (199g). Cloning d a full-kqfhinsulin-i&e gmwlh fanor ! compkmen~aryDNA in thc goldfsh liwr and m-. and dcrrlopment of a quanlitalivt PCR m c h d for n u mtarunanau On C a p . m - d .111. 51-60. K i e s W (1999).Mo*cuLr b w b - d t h c ICF.lllnunnoad-phmphate rcmp~or.In -TkIGF S y a m - Mdrmiar Biobg\. Ph>uol. ogy and Clinial A p p k a i o m - I R G. RaserJcld and C. T. R m . Jr.. E&.). p p 89-109. H u x m a Rm TOIOU~.NJ. Kinhult Slahlbxn. A . . h a . V. R..and Hoebm. P. (1999). ImYn.lllrr gmulh factor mRVA in b u r a m d (Iws a*;nkrl:.Altrmai\r splicing and non-respondto gmwrh hamm Cerrr 37. 69-93. LeRoith. D . Werner, H.. Bcimer-Johnson. D.. and Robtlll C T (1995) Molecukr and cellular spms of ihe imulin-like factor I receptor. Endmx Rev,. 16. 143-163 kchiara. T. >I.. Efsnatiabr Liang. Y. H.. Cheng. C. H. K.. and Chan K. M. (1996). h l i n - l i k e growth l m o r lea2 is the prrdominanlly expressed form of ICF-I in common carp (Cypinus carpa). Mol Mar B i d B i o t h l 5. 145152. Liu. J. P.. Baker. J.. Perkins. A. S. Robenson. E. 1.. and Efslratiadis. A. (1993). Mice carrying null murariom of the gener encoding insulin-like gmwlh factor Ifl@-ll and type IICF receptor flgflr). CLU 75. 59-72. Loffing-Cueni. D.. Schmid. A C.. Craf. H.. and Re~necke.M. (1998). ICF-I in the bony hsh Cormr wwpiln cDNA. e x p s i o n and dillerenlial localization in brain and irlerr. .%id.WI EndDoiml 141. 181-191. Loffing-Cueni. D.. Schmid. A. C.. and Reinecke. M. (1999). Molecular cloning and tisue expression of ik Wulin-like growlh factor 11 pmhormone in the bony hsh Comn urnpius. Cen. Camp. Em+ aim1 113. 32-37. Mathew. L. S. Hammer. R. E. Behringer. R R. D'Ercole. A. 1.. Bell. C. I.. Brinster. R. L.. and Palmiter. R D. (1988). Cmwth enhance ment of lranrgenic mice expreuing human imulimlike gmwth fator I.E n d x r i d o g y 123. 2827-2833. Miyakarhi. N.. Richman. C.. KanJrawa. Y.. Linkhart.T. A.. Baylink. D. 1.. and Mohan. 5. (2001). Evidence that ICF-binding pmteiw5 functiorn as a growth factor. I . Cljn Inw. 107. 72-81 Nissley. S. P. (1999). Type 2 ICF receptor-mediated evenn. I n -The ICF System: Molecular Biology. Physiology and Clinical Appl~catiom* IT. C. Rosenfeld and C. T. Robens. Jr.. Eds.).pp. 165197. Hurnana Pren. Tolowa NJ. Powell-Branon. L. Hollinghead. P. Warbunon. C.. Dowd. hi.. Pitts-hleek. 5.. Dalton. D.. Cillett. N.. and Stewan. T. A. (1993). ICF-I is q u i r e d f w normal embryonic gmwh in mice. Gnu LLv. 7.2509-2617. Reinecke. M.. Schmid. A.. Ennatinger. R. and Loffing-Cueni. D. (199T). Insulin-like growh k t w I in the telem ORalrmmir nrssambin~the lilapia: Gcne sequena. t i m K w o n . and cellular bcali+alion. Endmumlogy 138. 3612-3619. R d e l d . R. C.. and R o h C. T.. Jr. (Eds.1 (1999). -The ICF Syscem: Molecular Bidogy. Physiology and Clinical Applicatiom.' Hurnana Ren. Totowa. NJ. Rotwein. P. (19861. Two m l i m l i k e gmuzh f x ~ u I RV.& are exin human L b u . R u i ? l A d Sa LS.\ 8 3 77-81. Shamblolt M. J.. and Chen. T. T. (19921. !dmtITwMm d a m d insulin-like grow* f a n u in a firh rpcser Pm .klA d Sa L:SA 89. 8913-8917. Siegfried. J~ XI.. Kaspnyk. P C.. T r m o n A. Y . Mubhur. I L. Qu~M. K. A.. and Cuaina. F (1992). A mtogaur ppldc am& encoded wilhin the E p e p d e domain d ik tmulif-hke gmn* f m o r IB pmhomarr. Pm .%d~ A d So.L-$4 8%3107-8111 Tanaka. M.. Taniguchi. T.. Yamarnolo. L. Sxkagm!!. K ~ Yoshuato. . H.. 0hkubo.T.. and N K (1998). Gcne and DZIA wruc. huvs of flounder insulimhke gromzh f m u I flCF-0: Yuhrpk mRUA species encode a single s h n l mature ICF-I. DSA Cdl &rl 17.859-868 Ti.X. C.. Chen. M. 1.. Panuchenko. A. C- Y a w T.J. and C h m T. T. (1999). Recmnbirwt E-pepidcr d p l C F - I h a w m t o g a u aaiviy. 110. 3387-3X. Van Buul-Otierr S. C.. Van Kletiens Y.. Kons. J C . LudmbsghKonleve. D. 1.. Cresni~.51. C.. Dmp. 5. L 5 . HoogcrbrugsF. C . h i . . B l a m e n R J . . K o e d a m J A . . a n d \ ' m W J . W 11(ml Human insulin-Like gmu* factor UCFI hndirg pmarr I h b ~ u ICF-I rtimulaled body growlh but somukvs ~ I I dI the Mney i n v ~ ld lu a r f m m 141. 1493-1493 Wallir. A. E.. and k l i n R H 119931D u p l x r e d i m l * r v t h l m r - l g e m in salmon display al-r rpbcmg p a t h n a ? ~ ~ .Ud. E n h r i m L 7. 409-422. Werner. H.. and LeRoith D. (m00) .New 0 0 1 1 ~ i ~n(mguhnm ~ and function of the irnulirrlike gmu* fLmpliobDm lu undemanding mrmal gmu* and rropIasu Cdl .%kd Lik Sb 57. 932-942. WOW. E. Lahm. H . WU. M.. Wanke. R and HoeX maOl~ Etiecu of ICFBP-2 o v w e q x e s z ' ~in \lm, d m virv Phbtr Ncphml 14. 572-578. Yakr.S..Liu.J.L..~B..&I~.A..AmhD.Lus.B.d LeRoith. D. (1999). N m l grmzh and k d o p m a r m lk ab vnce of hepatic insulwlike grou* f a a u I Pm .W .4d 50 USA %. 7324 -7329. Available online at ww*vxiencedirect.com Expression and purification of a biologically active recombinant rabbitfish (Siganus guttatus) growth hormone Bruria Funkensteina-*,Anna Dymana, Ziva Lapidota, Evelyn Grace de ~esus-&son~. .Geh Gertlerc, Felix G. son^ and E ~ o t c c h n o l .Vanonal ~: lmnrvv 4? -. l r d ad T e I - % h a . PO &u &l30. Ho./a 3la60. l r w l ' l q u u m l m L k p m m c S o u h r .am Fishme brl0p"m Cenm 6E4FDECIQDl. 7igbrmul .W2l l h i b 'lnnrruw o j E ; a h m u n ) : F o d S ~ ? F . ~ 18nd I HUmlm. Fa+ 0J.AFood d fiwoowwmd a The Hebrp* L'nr,m#nqTJemalem R c b w - 6 l M l r d 'Lkprrnnenr o j . L h i n e E i o l w Rccrlred 6 A* 2001; rrcclred m r e d form I ? .-I ?LW. R m a n amprsd 20 Apnl ? W 5 + Rmmblnanl rabbi& p t h homune(&if)pvlein was in EicherdP col;. B L ' I (Dm) alk Tbe CDXA a u o d m p rhc marure protein ofrRjH was fim cloned in pGEhl-Teasy vector and then t r a n s f e d w PET-3daprssim %amEqmsmn m E~coli cells was then induced by IPTG (0.4 d l ) . Inclusion bodies (18) containing thc expressed polein nut pnhl* n u M g bacterial cells pellet with l)mzyme followed by repealed washings in cold upler containing T r i m S-100. sauarru~lad cenmhgalion. 10 were tben mlubilued in 4.5 h l uru, mfolded a1 pH 11.3 in thcpaena o f a n l y t u mounts o-.f lai purified by @Sepharoscmlumn. Gel filtrationon Superdex column shou-edtbe pmfied w n i w be a man- GH. M on SDS-P.4GE, the puriry ofthe recombinant rRjH prrparadon i s appmumawly Ba;. The rocanbinanl f f i H n u tcncd k w IB btological aniviry both i n vim. by its ability u, nimulaw IGF-I mRNA cxpazicm in tbc liva. lad in \i\o. by lo cb~llc) I@ accelerale g m w h in rabbifish fry injccled with the hormone. A sign~fican~ inm gmwlh was bedm nbbilfid fr? pbm the recombinan1 h o m n e . P o l y c l o d antibody raised against the native f f i H immumrurod -4th dx r o ~ l b iffiH ~ l ha bloa and in ELISA. indicating thc suibbiliy o f thcw reagents for fuiurc q w t i h o n of C H in rabbiIfifish p h t 0 0 5 Elwvier B.V All righb r m d . -' Concspad~ng a u h Tcl: 4 7 2 4 8565233. fax: -72 8211911~ E-moil d h : [email protected]~[B F~mkcnsldn). Gmuth hormone (GH) is a singlcchain polgepti& 4 of about tt kDa prodwed and w e d by sonwcmphs in the antenor pad o f the pituitary gland. GH plays an essential mle i n stimula~ingromalic _mum in vertebrates. Numerous studies have demonstrated that injection of narive or recombinant GH to different fish species resulted in -mwh enhancement (runkclljrzin. 2'H10), indicating thar increased levels of GH have the potenrial to improve ,guuzh in fish. The rabbitfishes (Perciformes. Siganidae) have long been considered a good candidate species for mariculNre because of their capacity to breed in captivity consume a wide variety ofalgae in the natural environment. can be readily weaned to accept formulated feeds in confinement, and high tolerance to environnlental conditions (L3111, 107-1). The 26 species of rabbitfishes found in the Indo-Pacific area (\lwud;l cr 31.. I'J8-i) are all excellent food fish and among them, the S~ganur g~~rrarus is best suited for mariculhlre because of their large size. Consequently. lhis species has been the subject of many research studies in SEAFDEC AQD resulting in the development of technologies for c a p tive breeding and hatchery production (Du?. I ' M ) . We have been trying to accelerate grouzh in S. gurrarus in order to reduce the culhlre durat~onto reach marketable size. Towards this end. we med to identify the limiting component(s) in the g r o u ~ haxis (GH-IGF-I axis) in this species in order to be able to recommend to the fish farmers some interventions to make this species prow faster. .As a first step. we have cloned and charac~erizedthe GH family of proteins (GHprolactimsomatolactin) and insulin-like grouzh factors (lGFs) in S gvrrorur (-4yson ct 21.. IqW. :In.)l1. :111)3). This allowed us to look at gene expression panems of these hormones during various developmental stages. In those smdies, we found that mulriple injections of purified native rabbitfish GH (&H) ro rabbitfish fry stimulated u-eight gain, indicat~ngthe porential of GH therapy to enhance prouzh in th~sfish species. However, obtaining even a small amounr of narive GH requires the sacrifice of enormous mlmben of fish. In lhis paper, we Rpon on cloning. expression and purification of a biologically active recombinant rKjH in E coli cells. 2. Materials and methods Oligonucleorides were prepared by Sigma (Israel). Resmction and mod~kingenzymes were purchased from Xew England Biolabs (Beverly. %fan) and Promega (Madison. W. All other chemicals were o i analytical p d e . 2.2. Construction of rJ@H erpression wctor A fragment of rfGH cDNA codins for manue GH protein was amplified usxng rfGH cDS.4 (A~solrer al.. X W t ) as the templae and the pnmcrs shown in Fie. I . The upmeam PCR prima cornsponded to amino acids I 4 of &H and COnlauKd an additional alanine and the .Vml resnicrion site. which donated the ATG codon. in the same ding frame as the rfGH. The downmeam PCR p k corresponded to amino acids 173-178 and included a stop codon immediately aAer the last mdon. iollowed by a BamHl restriction site. PCR amplificb lion was conducted in 50 111 using 2.5 L' Taq Polymerase (Sigma. luael) and 30 cycles. as fob lows: I min at 94 'C. 1.30 min at 56 'C. 2 min at 72 'C except for the first cycle for 3 min af 94 'C and the last cycle of 10 min at 72 'C. Thc amplihal fragment was purified by gel elecoophorsis. followed by elunon using QLAquick Gel Exnacbon Kir (Hilden. Germany) and cloned in pGE.\i-Te-vecror (Pmmega Madison, W.1)using JM109 E cob as the hog bacteria. R c ~ a of an LllM w m verified by digestion uith EcoRI rcmiction -me. One positive clone was seqmced with SP6 and r RKA pmmoter primas to verify IIW m smrs occumd during PCR amplification and cbning. The insen of the malure GH vas cut ow fmn dm plasmid using .%I and h H 1 &aim q m e s and cloned in the expression vector PET-3d d i g c s d also with >VcoIand B m H l and used to tnrrsfmncd &I109 E coli bacteria For expression, the plasmid PET-3d containing rfGH from three clones uas solated and used to transform the hoa bactena B U I p E 3 ) E coll. b l a b x i colonies that e x p r e s d the protein after indwnon by IPTG were s e w for purification. The p m c e of rfGH p t e i n in the colonies uas checked by SDS-PAGE and \Vestem blotting using homologous anrivnun (uc below). The DNA plasmid of rfGH-PET-3d clone that g x e h e best expression uas sequenced using F RXA promoter primer to confirm drar the frdgnmt of f f i H disated with Sdel BamHI mas cloned axrrclIy inro PET-3d digested uith Sdel BamHI. Rabbitfish G H cDSA - .YCDl 5 CGCGCC ATG GCG CAG CCA ATG ACG GAC ACT- 3 Ma Ala Gln Pro Ma Thr Sn 1 2 3 4 6 &.-H I 3 ETT CCC TTG ACG TGG AAC ATC C ~ K G C G 5 Glu 4h Asn C ~ JThr Lcu slop 1-3 174 1-5 1-6 17- 178 2.3. Erpression. re/olding and purification o/ rfGH ln a preliminary experiment, the lhree recombinant bacleria were g o w n in 5 m l L B at 37 'C lo an OD, of 0.8 and then IF'TG was added to a final concenmtion o f 0.4 mM. Bac~criawere p u n Tor an additional 3 h. centrifuged for 5 min at 11,000 rpm in a chilled microcentrifuge and h e pellet was resuspended in 500 p l TE, pH 8, washed twice in the same solution, sonic& twice for 30 son ice, cenmfuged again and the resullanl pellet of inclusion bodies was resuspendmi in 100 o f Laemmli's sample buffer containing SDS and &mercaptoethanol. Samples were heated at 100 'C for 5 min. chilled on ice and then cenmfuged briefly. .An aliquot o f I 0 p l was analyzed by SDSPAGE, on a gadienl gel 5% lo 15% One gel ua stained with Coomasie Brilliant Blue R-250 (Sigma, Israel) and the second was subjected lo a U'eslem blot using homologousantif f i H antibody (see below) raised againsl native f f i H (\!.,!n s[ 31.. 20()0). Two clones expressed f i H (-2 I kDa). which reacted with the antibody. Clone =2 u-as chosen for large-scale prepantion. 0 pGEM-T(3018 bps I For large-uale prcparation. B G I ( D E 3 I E d i clone 4 was gmun in 20 x 500 ml LB in dw pmsmx~ o f ampicillin (100 pgtiml) at 37 'C in 2-L flasks lo anof ODm 0.8. Then U'TG u-as added lo 0.4 mu.The a l l s were p u n for an additional 4 h. p e l l e d for 5 mm a 10.000xg in a Sowall RC5C cenmfuge at 4 'C and s t o d at - 20 'C. Fmzen pellets o b i n e d from 10-L culture (-30 g) u-ere hawed and &bided into L allEach pellet uas f i t s 1 musquou. in w o I-L Ma. pended thoroughly i n -20 ml m l d distilled uatcr. hen 380 ml o f 10 mW EDT.4. 10 mhf Tris-HCI. pH 8.0 were added while mixing conunwusly with a glass rod. Lysoqme (0.5 mg'ml) was dissolved in asmall amount of HzO (200 me400 m l o f baclerial soluoon) and added. The soluuon uas ~ncubavdfor 30 min at 4 'C with continuous nirring. The e l l s were thcn sonicaud for 6 min wilh onoffpulscs of0.5 s and pelleted f o r j 0 min. 25.000xg. at 4 'C. The inclusion bodies ilBI pellet conraining h e f i H was rrsuspended in cold dislilled water containing 0.1?& Triton X-100 and sonica~ed.pelleled and resuspended again in Tnlon S-100 containing cold uater. This pmedure u-a-led several times, until a clean white pellet was obtained. Last two washes were carried out without Triton. The final 1B pellet was rauspended in 100 ml water and divided into 20 m l batches and kept at -20 'C. I n order to optimize refolding conditions, the IB pellet obtained From 2 L fermentation culture (20 ml) was divided into 4 aliquots. Each o f the aliquon was solubilized in 4.5 M urea buffered with 40 rmU Trisma base ina total volume o f 14 ml. The pH was adjusted to 11.3 with 2-3 drops o f 2 N NaOH, resulting in cleanng the solution and then cysteine was added to 0.1 mu, 0.2 mM, I mM and 5 mM. The solution was stirred at 4 'C for I h, diluted with 2 volumes o f cold water and dialyed for 48 h against 5 x ? L o f 10 mhl Tris, pH 9 and applied to a Q-Sepharose (fast flow) (Pharmacia LKB Biotechnology AB, Uppsala Sweden) column 0.8 x 4 cm, pre-quilibrated with the same buffer at 4 'C. Elution was carried out using a&wontinuous NaCl gradient in the same buffer and I-ml fractions were collected. I n thae experiments most ofthe &H eluted with 100 m.U NaCI. I n subsequent refolding experiments the protein was refolded in the presence ofO.2 mU cysteine but dial@ against 10 mU Tris, pH 8. The elution was carried out by discontinuous KaCI gradlenr and the f f i H was eluted with 50 mll NaCI. Protein concentration was monitored at 280 nm and monomer content was determined by gel filmtion chromatography on a Superdex 75 column ( I x 30 cm) (Pharmacia L K B Biotechnology .a) In. addition, aliquots ofselected fractions were analyzed by SDSPAGE (see below). 2.4. SDSI14GE and lYesrern bloning The presence. purily and the amount o f protein recovered from the IB and during the puritiwtion steps were monitored by SDS-PAGE using 5-15?6 gradient gels. Samples were mked in Laemmli sample buffer with or without i5-merCapt0~hol. Gels were stained with Coommie Brilliant Blue R-250. For Westem blot analysis. proteins were elecno-transferred to nitrocellulose membranes (Schleicher and Schuell, Dassel. Germany), blocked in 0.5% BSA in TBS (100 mX1Tns. O.%NaCI, pH 7.5) and immunoreactedwith a 1 :5000 dilution ofspecific polyclonal antibody nised in rabbits (see below) against native purified &H \ I 1 I ! . fulll).Membranes were then incubated with biotinylated goat anti-rabbit $,I (H- L). affinity punfied (Vector Labs. Burlmgame. C.4) &luted to a final concenuatlon o f 0.15 p g ml. k e c n o n was x compl~shedby the akidln4lonn reapent Weemlam. .XBC k~t.Vector Labs) and >,Ydlam~nobemdlne In H202 Antiserum against &H was r a i d in rabbits. ?;alive rfGH (100 pgml) purified from the p i l u i ~ gland (;\yx,n el 21.. 3Mh')) was dissolved in d i e solution and was emulsified in an qua1 volume o f Freund's complete adjuk-ant. One milliliter o f the emulsion containing 50 % o f the antigen was injmed inuadermally into the exposed skin at the back ofthe nbbi~ Booster injections o f the antigen (50 ug m l o f emulsion in Freund's incompleie adjuvant) m-ere then given inbadennally into the b s k o f the rabb~nfor four timer at intervals o f 3 week. .4 mu& afier the last booster injectioh blood was colkcted from the heart Serum was sepameb lyophilued and stored u - 20 'C until use. Two rabbits w a e used to pcneraa the antiserum. Wells ofELIS.4 plates a!-shapedm M u r %Z?iC. Denmark) were coated with 0.1 m l o f raanblnant f f i H (20. 10. 5, 1, 0.1 and 00.01 llg'ml in 0.1 M castmnate buffer, pH 9.8). incubated o k m i g h l at 4 'C. bed with 1 0 . formaldehyde for 10 mln u 4 'C. and then washed with PBS containing O.OS?%Twsm 20 (PBS-T) four t i m a for 5 min och. Wells wae p a coated with 0.15 m l o f 49.0 bovlne serum albumtn (BS.4) in PBS-T. Folloming overnight incubation at 4 'C and washing. 0.1 ml o f rabbit anti-native punlied f f i H was added ( I : 1000. 1:5000. 1: 10.000, 1 :20.000. I:40.000, 1 :80.000. I:100.000 dilutions in PBS-TI and incubated for I h at 37 ' C Afier mashing. goat ant!rabbit IgG conjugated toalkaline phosphamse(Car KO. IIS-055-144. Jackson ImrnunoR-h Laboratories. Inc. W a t Gmve. PA) was added to each well and incubated for I h at 37 T. Following washing. 0 ~ 1 ml o f subsmte for alkaline phosphawu. p-nitrophenylphasphates (pNPP; Cat. No. 104-105. Sigma) mas added at I mg'ml to each wrII and incat 3' 'C. Contml wells did not contain rtGH or no anti--. The optical densit). (OD) was determicd using a micro- ELIS.4 reader at 405 nm. Each ELlSA result was expressed as the ratio between B e mean absorbance o f tested sample in positive (P) and the mean absorbance in negative (h?wells. 2.7. Biological acnviy Two batches o f f f i H preparations were used sep arately for the i n vivo and in v i m bioassaj-s. 2 . 1 1 . In wrm The procedure o f \loriyalla er 31. I1lXl21 was followed with some modifications. Rabbitfish fr). u i t h body weight ranging from 13 lo 15 g were dissected and the livers removed. The livers were washed twice u i t h sterile water and three times with the sterile medium before cuning into pieces (roughly 0.5 cm size) and incubated in v ~ t r oin Medium 199 (with Hanks' salts and L-glutamine, without sodium bicarbonate) with or without recombinant f f i H . The medium (pH 7.4) was enriched u i t h oxygen and supplemented with I00 U'ml penicillin. Liver slices were distributed individually in 24-well culture dish containing I ml ofthe medium. They were pre-incubated for 24 h in the medium without the hormone. Incubation was catried out at 18 "C with slow shaking. The medium u;rr changed the following day with the same medium containing recombinant f f i H at the following concentrations: 0 (control). I.10. 100 and 500 ng!ml. Incubation was continued for another 24 h after which the liver slices were immediately put into RNAlateF solution (Ambion, USA) and stored a1 -20 'C for RT-F'CR. Liver slices h m three individual fishes were used for each concenhalion o f the hormone tested. Increased expression o f IGF-I mRY.4 h m the liver slices incubated with the medium containing recombinant G H war d e n as a measure for the bioactiuir). o f the recombinant GH preparation. Tow1 RXA from the liver was extracled using T R l z o l ~Reagent (lnviuogen, Carlsbad, CA) and first strand cDNAu.as reversed uanscribed !?om equal amounts o f total RKA (5 pg) using a cDKA synthesis kit ( h e r s h a m P h a c i a , Sweden). PCR was performed using an aliquot o f I *I o f frst strand cDWA as template and the following primers (forward: 5'-TCTCACT.4CTGCTGTGCATC3'; reverse; j%TGTTGCCTCGAClTGhGTT-3) specific for B G F - l (.X><nner 31.. ?l)i12). .4 PCR product o f 454 bp i s expected using these primer pain. Th~ny-tive cycles o f F'CR were performed u - ~ t hthc fullouing conditions: hot stan 9J 'C 30 s. denamation 9J 'C 30 s, annealing 58 'C 60 s, exunsion 72 'C ik) s. final extension 72 'C 7 min. The idmtiry o f !he PCR prcducts was confirmed by d i m ! sequencins. Rabbitfish pactin uzs used as the Internal d a r d during quantification. Primasspecific for rabbltfish $actin u-ere used: s e w (S%CCA4C.4G<jCl.l.14AGATGAC-3) and anti(5'-TTGCC.AATGGTGATGACCTG-3) (A:-son e l ~ i . . 001). F'CR pFJducts were loaded onto 1% agarox gel and named u ~ t h ethidium bromide. Quantification o f PCR +-ls was done using GeneGenius gel documen~ationand analysis system (Syngene. fK).Valf w IGF-I uar normalized with rhal o f pactin. 2.72. In v i w A total o f 120 tabbitfish fr). u i t h body wright ranging h m 3 lo 5 g u-ere randomly divided Into four groups with 30 fry per group. The fn- in m h group were stocked i n thrrc vparate libaplass mnks of25O-L capacily or an equivalent o f I 0 tiy per lank. A total o f I 2 tanks were used for the four groqxs A11 the tanks were filled u.ith manual seawiuer and u w e held indoors. Three groups were given intqmiloneal iqection once a week for 4 uacks o f recombinant f f i H a dosesofO.01.0.IOand0.5~gBW.Thehormoneu~ dissolved in saline solution (0.9% SaCI). The fourth group (conml) was injected u i t h an equal volume o f saline solution. The fry were fed tabbiduh anirutal diet at ja% BBW per day. The feeds were given at duct rations at 9:00 AM, 11:00 AH and 2 9 0 PM. Feedins and waler management were h e same for all mamwnn. Individual fish were measured every week for changes i n body weight (BW) and body length IBLi. Significant d ~ t f e r e ~inn the values between the c o n m l and hormone mated groups w m dermlned by DMRT afier A N O N . 3.1. Erpression. pun:fuation and choracteri-anon o i recomhinanr cjGH Thrce bacterial clones were found to express &H. o f which expressed large amounts o f the protein as judged from small-scale induct~onexpmimcnts. NO SDSPAGE detected a major band with a similar mobilir). lo that of human GH (F:?. >A). \'estern blot analysis revealed that this band reacted with an antibody raised against native punfied rfGH which had a similar mobility to that o f GH detecred in pituil;u). exmcts o f the gilthead seabream (Sprur o~rroro)(Fig. 2B). indicating that the clones indeed produced recombinant f f i H protein. Furthermore, the Western blot analysis indicated that antibodies nised against the homologous native peptide recognized the recombinant peptide produced in E coli. IB conmining rtGH were prepared from -30 g wet weight bacterial pellet obuined from 10 L as described in Materials and methods. An aliquot o f 100 p I was laken for analyzing the amount o f rfGH in the I B by SDSPAGE using recomb in an^ bovine G H (bGHI as the standard. The yield o f f f i H in the IB was calculated to be -60 me. The protein contained in Ihe I6 was refolded in 4.5 M urea as described in Materials and methods. To determine the optimal concentration o f cysteine needed for proper refolding. the purified rfGH was refolded i n h e presence o f several camlyric concentrations o f cysteine: 0.1 mM, 0.2 mU. and 1 mM.The refolded rfGH was purified on a Q S e p h a w column by stepwise elution wirh incmasing concentrations o f NaCl In 10 mU Tris pH 9. The purified rfGH. which was eluted i n I 0 0 mU F t g 2 SDS-PAGE and W a v m Mm d ? ~ of a lopl crlluLr pmtetns from E roll cdlr Lane I. R r v r w a c x m hum Spvur d u m ~ rlane ~ : 2. Clone n2 oipET &H. lane 3. rnwmbmull M H A. C o u m s t c blue stalnnng: B. W a r m btof u ~ r ha n n w ntd aqannw Mtrre punlied &H. SaCl (Fig. IA). was anahzed by elecmphoress u n d n non-reducing conditions and is shown in Fig 56. In all three cyxteine c o n c e n ~ i o n smed for refolding. the purified rfGH appeared as a monomer. Aliquols o f the peak fraclions, shown in Fig. 5.k were analyzed for monomer content by gel filmicn lhrough Superdex 75 column and wem found lo contain a monomenc rfGH (dam nor shown). The elution profile obtained with a discontinuous -madimt o f SaCI In 10 m\1 Tris pH S and in the presence o f 0.2 mS1 cysteine is shown in Fie. 4.4. .4nalysis o f aliquols bkSDS-PAGE with or without @maapmerhanol re\.ealed monomeric rfGH (dam not shown). F m h a analysis o f the peak fractions eluted uith SO n\l SaCl and I 0 0 mu NaCl (shown in F i g 1A) for monomer content by gel f i l u a ~ ~ orhrmgh n S175 column is shown in Fig 4B. The rfGH eluted i n -W mu SaCl were almost pure monomeric form ullile with 100 m.U KaCI impwities appeared. Companwn o f elecmphoretic mobilin. o f rfGH eluted m 0 m\l NaCl by gel electrophoresis with or withoul Ci-mercaptoe~hanol from a thii preparauon o f rfCH revealed monomeric form. Furrhenmrr under mnreducing conditions the mobility uas s h i M slightly forward. indicating a globular s n u m i F ~ g -4C) Fractions h a t mntained monomeric pmuin s shown by SDSP.4GE and HPLC anal)= wur p o o l 4 dialjzed against 0.05% ic'aHCOj and Iyophilized The monomeric smclwe and idenl~ryof ihc dried-rauspended rfGH was furlher anal:& by. SDSPAGE under non-ducing condition. followed by Westem blot. The ccsults. shown m F I ~5.4. revealed over 98% pure band o f - ? I kDa. u-hich is recognized by anti-native 6 H antibody (Fig SB). The yield oblained from Z L o f E cuh fulm was calculated as -5 mg o f pure m o m c &H. The molar extinction u r f f i c i e n l al ZBl om was calculated using the molar extinction o f I)nuwle (1 197) and ryptophan (5559) yielding a kalue o f II , e n , 19hh). The spac~fic for a O.l?-6 solution (ms'ml) was 05-89. assuming a molar mass o f 21.000 Da 3.2. Chmc~erirarion of onn-fiH rjc~ ~m~ mo,&;nanl anrhodie b~ As s Ant step towards developing a quanoIatikx assay (ELISA) for determination o f GH levels in BEST AVAILABLE COPY Tube m. - - Fig 1 (.+I Elutm profile of reFomblnant ffiH WfKd on Q - S c p W mi-. Tbc OScphmv m b ra e g u W d 10 mtt TmHCI pH 9 The pmvln eluted r l l h a d ~ ~ o a m u x gradnslr s of S K I in k ~m bullcr. of -%3M mM SdX Foi I ml rm collated. The pmwin conrnmrian was mmmd by lbwrbvlo u ?80 m.(8)SDSPAGE ofprnhcd &H. dDLdcd urn pacm at dlllirsll conmumam of w n n c and clecaophoraed m ~ 1 5 % gmd~rnls ~ ~ p o t ? m - k . dgrl c I. ?bc of 7 LI and 10. MU' d c r . Ian?.I mhf .ylc1nc(5 111): Lvrr 3 d i 0 I mM qtlrnnc(5 ard 10 )ill. Lurr 5 m d 6.6.2 mU 9%- ( I mt I0 plb lams 7. 8 d 9. reccmbnmr ovme GH a 10. 20 lad 30 115The gel uar mrned wlh -C B d h Blur R 2% s e m o f rabbitfish. the antibody raised against mtive purified f f i H (:\)son el AIL.30(W)) and the recombinant f f i H purified in the current swdy were charactenzed. The results o f the ELISA, summarized i n Table I, showed concentrationdependent and dilutiondependent reaction. For f u w determinalions. the antibody can be diluted benveen 1:40.000 and 1:100,000 and ELISA plates should be coaled with I p_e ml o f pure recombinant f f i H , which gave 12.5 to 21.I P - 3 values The 10-fold dilution o f anlibodies for ELISA compared to the dilution used ior \\&tern * blot (see Fig. SB)is mithin the accepted diludons u d for rhae d~ffermt&says. 3.3. Assnvjor the b i o l q ~ m actirin l ofr~rombinant r/GH Ln an in bim e q x r i m e n ~liver uplants fmm rabbitfsh were incubated in medium containing recombinant f f i H at doses o f 1-500 ng'ml for 24 h. A slight but s~aristicallynot si@~ficanriin h e levels o f ICF-I mRYA was noted in explan~sincubar- BEST AVAILABLE COPY " , 0 10 20 30 40 Tube no. 50 60 70 B. 50 mM NaCl I?== 524-xm CHI 100 mM NaCl S15-rm MI a FIE 4 ,A) E l u l m p o f i k o l d m t &H plnficd on QSrphamv c o h m T h i QSqbmemh.nn wm&hb.ad nlb 10 m u TrsHCI pH 8 The w o n u-rr e l u d wrh a darunnnuur g d m l o l NaCl m rhe ramc bufkr. of 50-?00 mM S K I -F or I ml *a mllec:ed Thi polm co-IROM mild by a k ~ c b m a t 280 nm (Bi Lkmmmlm of ihc momma C:of mH b? & filna~~on chrnmmgrqhy an a Sup&. 75 m h . RJr karuunr fmm Q ScphmY i o h cluoed n&50 mM SKI a d lW d lY L l rhoun in .A. mpecavrly Thi nvmkrr above rhe p s k rhow IIhc mmlion mm m m u l c r fCi SDEPAGE d r u ol &TjH d w d I. -X m\l naCl irasloon. Lmr I. \I\\. marker lam I. ~T*lmiwn.i?du~tag i~dnimm( u ~ i b.I-i ? LUY 5. & ~ v d m I& p-nxrraptcdhanol) The gel u= rlrsned ~ 8 t hC,r,mairlc Bnllranr Blvc K 2 i i l ~ :* BEST AVAILABLE COPY Fd_e 5. SDSP.4GE and W e r m b l a a n a l p i of pure rrxmbml &H. lam I. CNdc pepmoo. of boma 2. h k d p-tlon af~nclusionbodis; lux 3. Rm &H ~ f o l d c dIn the -c 010.2 mu w u r and dud m l lm n\l ?i.CI lac L.g L: e l c c m @ o d wnhout ~mmaprurhual;15. Ruc f f i H d i d > m l 10.05% SaHCOj. I!ophtlual a d V cli S d c HQ. elcc?mphaacd wilhout &-macaptahaml; bne 6. Raombmmt o v m GH (10 MI (A) C o m m s t c B l u samms I B i ua;m Nx ~mmwurcactedr t r h mtr-native &H antlkdy ed with I00 ng'ml for 24 h (Fie. 6). Other doses were not effective in stimulating IGF-I gene expression in the liver explants compared 10 the controls. In vivo, however, a significan~increase in BW gain was o b m e d starting on the 2nd week in rabbitfish fry injec~ed with the recombinant f f i H a1 a dose of 0.5 p g g BW compared to the control and those injected with 0.1 and 0.01 p g g BW (Ftg. 'A). This significant increase in the BW gain was observed until the 4th week of the experiment. No significanl difference in BW gain was oberved berueen the fry in ihe control p u p and those mated with 0.01 and 0. I p g g BW during the entire duration of the experiment. Similarly, a significant SWO 10.000 Z0.W 4O.W so.m IMl.WO 0 >4? >42 >42 >42 >42 >42 33 4 30 1 I . 18.1 17.7 I 1.0 1.0 >42 >4Z >42 U.5 0.2 IS.? 1.0 W.0 24 5 I4 I I 81 I 8.2 28 I 47 4.2 10 zo 80 4.5 2.7 1.7 1.2 17 1.0 1.0 >42 >4Z 40.4 12.5 10 4.5 increase in body length WBS observed m @ tnjmed with 0.5 pg'g BW aRer the 2nd week; this *as no1 o b s m e d in those that w m inuith ihe lou-er doses and the control ( F I ~-B). In the presenl srudy, we rep^ on t k crprnrian, refolding and purifidon of recambinant f f i H . which is conlainad in inclusion bodia. Thc yeld war -5 mg of close 10 98% pure moanncric &H fmm 2 L of bacterial culrurc. The pollam found in the 'ac~ion eluted nith 50 mW SaCI iuhcn 1.1 - The optical denrmry (OD) u.%sd+rcrm~ncduang a m E L I S A reader at 405 nm. Each ELlSA -11 *rzr c r p d a Ihc w o berwcen the mean absorbanre of tested ramplc in pmaorc IPI and the mean rbsorbancc in Depmc 1x1 wells. -y u x per- h g 6 IGF I m R t 4 b c k LI .1&h 11,s d r a mo&ad m wtrm m m d u m c m r a ~ g mnrmmmm o i W*If i H IFF-I Y*l Frnn pcne crvm c k m m d f o d in mpllcats. hb RT-PCR BEST A V A I U B U COPY A +0l~ggBW 0 1 2 3 weeks 4 5 B c 20 7- 0 01 pgfg BW s 0 0 1 2 3 weeks 4 5 Fug 7 Thc e f f m or r o o D m b l ~ n 1f i H on l"cmaw m bod, *rnght I A l dI d kagm ( B Iolrabb~rfishfry a h w d l ) h n r m a n c m ofrhc h o r n Thc .auk M pevnrcd a -I VJua w ~ l h atensk art sngnnfiirly dnff-I imm rhc control ... the refolding was carried out in 10 m\1 Tris, pH 8) but in the 100 mll KaCI fractions when the refolding was carried out in 10 mM Tris, pH 9. .4nalysis by HPLC and SDSP.4GE under non-reducing conditions revealed a monomeric form. indicating a c o m t refolding of the polypeptide under both pmtocols. Furthermore, comparison of the mobility of the purified refolded f f i H by SDSPAGE under reducing and non-reducing conditions showed a slight shift forward in the mobility in the absence or pmercaptoethanol. Even after lyophilization and reupcnsion. the peptide retained its monomeric form. The protocol for IB preparation. refolding and purification of rtGH followed essentially that described for S. awara GH IsaGH. B e n - A l i ~et 3i.. IQW). although in the laner repon an additional protein (placramsse) was coproduced. co-purified and co-eluted with saCH by 100 dl NaCI. The recombinant f f i H WTS tested for its suitability for the development o i m ELISA. using polyclonal a n u i i e s raised in rabbits against the purified native f f i H (4ywr: et 31. r)iJnl. which was also used for Wenern b l o ~to v m @ the identir?.o i the expressed ffiH. The results of the ELIS.4 demonstrated a high titer for the antibdies. based on their ability to detect recombinant f f i H at \-- IOU- concentrations (0.01 pg'ml) when used in a high dtluhon ( 1:80.000). Although numemus sndia ha\.e k e n published on the expression and purification of recumbinant tish GH. in this paper we describe the refolding of a fish GH that carries unusual features compared IO o h fish GHs (.Aywn et al.. 10(X)). Rabbitfsh GH is larger by 5 amino acids than the flounder GH. uhich is the smallest teleon GH (173 amino acids) makine it h e second smallat GH molecule. It has a unquc characteristic: 5 amino acid deletion in paritions I4SI47. immediately preceding the I4 amino acid deletim m flounder GH. In addition, it has single a double amino acid deletions. This deletion of 5 amino aciris is unique to rabbifih GH. These 5 amino a-ids am even present in flounder GH. At this point we do nol know the biological significance of this dcktion. Despite this f e a t ~our ~ , data suggest that the peptide uas folded correctly, using the proroml dccailed in the ma nu scrip^ In addition. it could be argued lhat the inuinsic activity of GH is IOU- in rabb~rfirhand t h ~ s probably explains the slow groulh of f a b b ~ d s h(pmb ably one of the factors). In order IO pnaw thrs unique chvacteristic of rabbitfish GH. it baame n e e s a y ar a f i s t step IO expms, purify and fold cum+ rabbitli~hGH. The biological acn\.it). of the mombinaot rRjH was tested both in \IUO and in \ivo by employing homologous systems. In vivo. rabbirf~hfr). injected intra-peritoneal (IP) with the recumbinant f f i H at a dose of 0.5 pg'g BW already showed significant increase in growth starting on the 2nd u.eek of treatment and thls u-as maintained as long as the hormone injection was given (4 weeks). The lower doses (0.01 and 0.1 pg;g BW) of the hormone used did not eltcit a faster growh. A slight difference in the bioadvir). was found between the gmu+ expenmenu conducted u1th the recombinant f f i H tlhis study) and our previous repon using the native f f i H that was purified fmm the pimilary gland (A?wn ct 11.. ?(.<I)).\\hie in in gmu* our previous work a significant inwas already observed in f r j injected u-ith a dose of 0.1 pg's BW, this dose was withoul effect in the present study. This difference can be amibuted to a lower specific activity o f the recombinant rRiH. Although by SDSPAGE and HPLC analysis the recombinant rfCH appeared as a monomer, the amount o f purified protein wed i n the nvo systems could have been overor under-estimated, depending on the way the amount was calculated. Nevertheless, a positive effect o f the recombinant rRiH on growth, even with a higher dose. can be considered as an indication for the b i e logical activity o f the refolded, purified f i H . Similar results were reported for several other fish species. in which an increased growlh was also observed following injecuon o f recombinant fish GH. Thw, in mullet (.\tug;/ cepholur), fish receiving tP injections o f recombinant porgy G H every 2 weeks for 16 weeks al a dose o f 0.1 p p _e B W showed significant increase in gmw-th. starting on the 6th week with a lesser effect seen in those injected with a higher dose o f I p g g B W and no effect was seen when a high dose o f 2 pg' g B W was used (Tsai ct at.. 9 9 ello on et 31. I I')SR:I also reported increased growth rap in rainbow trout injected with recombinant h u t GH at a dose o f 0.2 p g g BLV. Likewise, slower growth was observed in those injected with 2 pp'g BM'. In the present study. we did not use a dose higher than 0.5 pg'g BH_ Slightly different results were found i n sea b m juveniles. i n which a dose o f 2 pg'g EN' eliciled an increase i n growth rare (Bell-Ati:! r t al., IWU). By contrast. the i n v i m experiments for testing the biological activity were less promising as only a m a ~ i n a effect l o f the recombinant rfCH on IGF-I eene expression in liver explanls tiom rabbidish Fry was noted after a 24-h incubation with 100 n p m l o f the hormone. I n previous studies. primary cultures o f hepatocytes, rather than tissue explants, were used for testing G H activity. Thus, Duan et al. ilo9;l observed a dosedependent increase in IGF-I m W . 4 levels in coho salmon a k r incubation with native coho salmon GH at a dose ranging fmm 0.01 to I p g m l . Sharnblon sf JI. l19')51 also observed an increase in IGF-I f i . 4 levels i n primary cultures o f hepatocytes in rainbow trout after incubation with bovine G H at similar doses with that o f Duan al. I 1Q9.il. X dosedependent increase i n IGF-1 mRNA determined by RT-PCR was reported also in p r i m q culrures o f tilapia hepatocy-tes cultured in the presence o f recombinant tilapia GH (Scl1m131LA 31.. 20011). However in - the laner study no correction *as nudr to a conwltutive &%A like Factin. for example. and therciore the apparent effect could be over-shrnated A daudependent inin IGF-I mRSA levels i n thc l i b e r o f coho salmon was also o b e d after a slngle LP injection o f coho salmon GH ( h u r c: a:.. 109:). Several possible explanations could be o f f d i;u the lack o f significant induction o f IGF-I mRXA in liver explanls and Ihe d i f f e m b e e m h e m vivo and in v i m assays. Smce i n the c m t study ur uud rwo different preparations for thc in b i ~ v and m w m bioassays, bariabiliv in the potency o f the nr-o prrparations or inaccurale d e t e r m i d o n o f pmein concentration could be one reason for low aclivit). In the in v i m experiments. Another pcssibility is poor pcnetration o f the r f c H lo Ihe tissue explants. Sum Ihe [issue explants were maina~nedin Ihe a b m x o f serum or other pmtein substitution. 11 could have affected also the integrity o f he tl+aw although this is unlikely since bypan blue exclusion tra of thc tissues showed ne-mive results. Numerous srudies repolled on poduwn o f recombinant fish GH pmltins. Exccpl for few cavr where the pmtein w w p r o d u d in a a - o b c expression sptem (TSI el al . :W; Ho d a!.. IQQ*). most o f tbese mornbinant fi+h GHs hd been expressed in bacterial cells (rebxU.ed in Fmkmdrcr.. 2MN) Polypeptide expression i n bacrenal cells is Ihe method o f choice for the production o f simpk pwcins ltke GH that needs no p o s t - m s l a M m o d ~ t i a a o m because i t is easv. m o v e r is fas~s i r n b a u r r ~arc -mnn for only a day. and the yield is hi&. S i m h to carp. sea bream and dolphin fish rccombinant GHs. rfCH also conmined an a m i n e m i n a l mcchionine residue. Thc law nucleotide donaud by the .Vml resuiction site (uud as one o f the cloning sim in the p m d vector used in the cumnl smdy) a d h e sequence of the fim amino acid residue o i t h e nwum f f i H made 11 m x s a r y to add N;O nuckotidcs m order lo keep h e reading frame. t b u creating an additional codon for tbe amino acid alanine. R e v ~ o u studies have shown IJW i n some casa the additional amino-terminal methionine is almost completely being processed and he pmltin is alanyIGH. ltke in the case o f S aurora G H (Ben-.Atla 31. IUW) while in other cavs the additional methion~ne no1 cleaved durine processing by E coli. l ~ k ein Ihe case c f dolphin fish I P n d ~ c l e: n l . :MI. I n Ihe sre o f c v l r w 250 12W5i SW-515 carp GH, about IZ-13% were Met-GH (Fine e l al.. 14951 The three preparations, which were folded using a similar protocol to that described in the current study, retained their biological activity in several assays. There is an increasing interest in the application o f G H therapy in fish in aquaculture since this possesses the potential to increase fish production. While there are several ways to administer the hormone IO the tish (injection, immersion), incorporating the protein in the diet is the most practical way for its mass application. T h i s method, however. will become etTective only if fundamental questions surrounding the method like, avoidance of enzymatic degradation in the gastrointestinal bact, quantification of the hormone concentralion in the feeds, efficient mixing o f the protein in the feeds without losing its activity, among others, can be addressed effectively. $15 mbr?og& a d culy hnd k d q - Z . 6 u M A Lncbxrul. 126. 165- 1 - 1 Beo-Atla. I..Fmc. H.. T d s . i. FlmLmrtah B. S. Cabmi. B.. G m k . .i IPYP . Pqwzmr~ ot ;suxsrbnuu pit. hcadsabrcam(Spnrur~~lllgmr*~r*mrar:br mulation o f lmr gm*-h b) onl mirdnlnnm C q Endocrinol. 113. 155-IM Duan. C.. hp). S.J.. Pl-~a EM. IW3 W m - l a c p l h facm I (IGF-II mRXA exparm d m dmcm (kCIIh m c h u hrurch: Ih h m m md csarco of p a b-prolam ftmll? provlm F& P b > d B b :I. 3?l-379 h?. M 3..I W 8 E l ~ o b pmd Cuhnc of S q m d s SLiFDEC Acpcuhtrc Drpumm~ Ilab. Phd53 pp Fmc. H.. Sdrd. L. \'&I. D . IBnd. V. I n A . LprhrP 0. Gmls. A.. 1993 R e c a n b ~ worp fQpaw c e p l p w i h hom*mc aprcsm p u i F a 3 - a 4rhdllwvnua of b w l o g d uurrr). ul v ~ m ml m uro Gsl Curp & mml 89. 51-61~ bxF d m m n . 8. ?MO P a m d l p p l r m a o f . m o n a d p n 7 h -m $&& b. Faapmmx \t. Sagabhmhansn. R (Edr.). Raceru Ad-. c \(rZ Bka ucbmlw.Aqwmhue. Rn B F* ~ 4L -1 ss. k..E n f a SH. p p 57-76 6cn C-. .- HO.WK.K..M~LQ.~HR.P~CT.-YK.Y)YL We thank Dr Sara ~Uauriceo f the Hebrew Universin. of Jerusalem for assistance in recombinant G H purification. We also thank Fiona Pedrmo. Edgel May Bayag, Hannah Cordero and the staff o f the Fish Hatchery of SEAFDEC AQD for assistance in various aspects of the project. This work is supported by a grant from US.UD (Grant No. TA-MOU-99-C 1940I. Program in USlsrael Cooperative Development Research. Economic Growh). References .igclbn. L.B.. E-. CJ.. J-. J.M.. D a v k S.L.. h g k . A.D. TT. 1988. Rwnodun of npd w u l h of mnbou UaoI j S d m gurmhm) by a raromhmc fah p u l h h m n e Can. J Flrh Aqua El.41. I*-151. .A,pon. FG. dr JE.G.T. A m i s Y . >tonyS.. H m . T. hwauchn. H . IPYP lvlbtan imd cDS.i cbntng of mtolaxm m &bluish !Signnu g u m 1 Gm. Comp. Edc-znml. l IS. 292-300. i j w n , F.G. & Jculr. E.G.T. . h y a Y. M o n y m u 5 . HI-. T. Kawrrh~.H.. 2aT). hlarim cDNA clonrng imd gruulh p m i n g acnr-~;yof rabbidlrh ( S i p w gumnu) w u 7 h bmnc G m C o w . Erdocmol 117. 3 1 -259~ A!ron. FG.. de Jslri, E G T.. M o t i ~ w S . Hyodo. S.. F d m stein. B.. Gmla. hwauchr. H.. 2002. Dnllmual a p r a rwn or inrul~n-liic grouih b a o r I d 11 rnR5.Q dunnp .. KT.. D l s s S . O r . A . P K . L q J . ~ W M . % \ ~ . Wong A.OL~. 199? hpcmca o l g p r cmp m b by -nna m h r El.Khm Ek#B.rr .is 1381. 331-33% Lam TJ. 1974 Sagvudr a d dm mancuhm .Aapxd rn 3. 3U-3.S M* H., -r K.. %C.. L ~ o m T. Yobpo. T. IW The Flrher o r k- J . b j q e @ m T& CM-~ Raa J a w 437 p p ~ MotiyS.. TJ;lutr T. Y a m q d u . K . b& H . M o h k c h m g of d m Ikc ~ - f w chD S A s rd b uprcrwn la Jlpaaac scl Fah Sa 68 tSlpp( rr 5 " - - 5 MwI. A.. Clupn+Cohe4 5 . Gsrb. A. Ellop. I . 1999 R e p u a t k d charmrnnan of mamkew AaW= Wi (C-pkem h,-I p 7 h hmma Rm b p Ant' 14 11--423 Pnttn\.SJ.1986 R a c l . v o h n n s m l h y h 5 m & t . J E l l o c h . IS?. 169-180 W m d . A.C. R&+ H . KLoa. U-. Nh.al heptqm or k bm? fah. am~ T Y Idr ~ P . ubpu a rdd :ml i c r p h + a ~adp a IGF-I m lur= J Eodoouol 166. 265-?i3 C.L. BOIL D . C?a TT. 1995 .-ShamMoq M J . a ol~uulin-Ilk p ~d I InamR\A o dr ir-w rd c?h or a ldmn i"rspomc D aogoou, W%? kmxxe Roc N d .idSc: C S i91.W:-tA% TsaH.J..ChcRH.M..La,CF. 1% -s?nthaoolrinr ;Mmb,w:k\ p n * A b) d k .arr'g a h w bvraror Cm. J Frrh A q u Sca S!. I-Tw. HJ.. Lan. K.L. Kuo. J C. Chm.S W . IWS Hlg!!i> c&ffrrcnc a-boa of Tuh p ! h hmnvlc b) G d m - h ~ ar d p p l En-iron \lvirobwi 61. 1 1 :6-2119 * w- - -
© Copyright 2024