4568 [CONTRIBUTIOS FROM THE
JONATHAN
s. DIXONAND C H O H HA0 LI
HORXONE RESEARCH LABORATORY, UNIVERSITY
Vol. 82
OF CALIFORNIA,
BERKELEY, CALIFORNIA]
The Isolation and Structure of a-Melanocyte-stimulating Hormone from Horse Pituitaries BY JONATHAN S . DIXOKAND CHOHHim LI RECEIVED FEBRUARY 23, 1960 11 has been piissible to isolate a,->ISH in highly purified form from the posterior lohcs of horse pituit >implified method which makes use of counter-current distribution in a solvent systein consisting o f 0. butanol and pyridine, mixed in the volume ratio of 11 : A : 8 . In this solvent system. the hormone distribu coefficient of 1.30. By the frog-skin assay method, a , - M S H possesses an activity of 1.0 X 10'0 units per gram. In addition, this molecule has been characterized b>-means of terininnl residue and amino acid analyses a5 well as on the basis of the results of enzymic digestion. In every respect, a,.-MSH was slioivn t o be identical with CY-MSH from beef pituitaries. A partial amino-acid sequence has been proposed for CY,-MSH;it appears very probable t h a t the equine a-MSHhas a structure identical to t h a t previously proposed for o->ISH isolated from beef and pig pituitaries.
The achievements of a variety of workers leading to the isolation and structural elucidation of the various melanocyte-stimulating polypeptide hormones from porcine and bovine pituitary glands have recently been reviewed. -A melanocytestimulating hormone, B - l Z S I I , from pig pituitaries, has been shown to coiisist of a single peptide chain2-j containing 18 amino acid residues in the sequence H-dsp.Glu.Gly.Pro.Tyr.Lys.hlet .Glu.His.Phe.Akg.Try.G1y.Ser.Pro, P r o , L y s . A s p - 0 H. Bovine 13-1ISHti.'has a composition identical with that of the hormone from pig glands except that a glutamyl residue (position 2 ) is replaced by a seryl residue. &Inother melanocyte-stimulating peptide (oc-XSH), also in the form of a single polypeptide chain but containing only 13 amino acid residues, has been isolated froin both pigsJ and beef pituitary glands,l o The complete removal of contaminating material from L U , - ~ I S has H ' ~presented considerable difficulties (see, for example, ref.l*). The amino sequence of cyb-SISH was shown to be identical with that cietermined for ap-MSH,I;], l6 namely, acetyl.Ser.Tyr.Ser.~fet.Glu.His.Phe.Xrg.Try . G1y.Lys.Pro.T'al.SH2. The adrenocorticotropic horinone (-ACTH) preparations from pig, beef and sheep glands" -I3 also possess a me!anocyte-stimuI I ) C .€1, Li, Aii8. PioC. Chen! , 12, '269 (19.57). ( 2 ) I . I. Geschwind. C . H. Li a n d I,. Barnafi, 'Trrrs .ll!N (196fi). (3) J . I . Harris and 1'. R w ( 4 ) I. I . Geschaind. C . H
JOURNAI..
73,
ti20 ( l ' l z i ) . ' 5 ) J . I . H a i r i s and 1' R w ( 6 ) I . I Geschmitid, C. I1 100.3 ( 1 9 R i ) . ( 7 ) I . I . Gescliwiiid, C. 11. 1.i a n < l I, I3arn;iii thz $ I ) ] ,we refer i o b o v i n c wXISH as olpI!SH a n d p o r c i n e ,r-MSH a s a , , ~ . l I S I Ithrnfihi,ut , (hi, p i p e r , and w e desi,qn:tte t h e equine hormone iis a,..\ISH (14) S. L. Stee!man I? S . Andersen a n d I
lating activity but of a lower magnitude than that of the MSH peptides.1,20 AU1these polypeptide hormones that exert melanocyte-stimulating activity contain an identical internal amino acid sequence which extends through se\ren amino acid residues (Met.Glu.His,Phe.Xrg .Try.Gly) . The isolation and structural investigation of the LISH polypeptides lrom horse pituitaries have been undertaken in an effort to determine whether the similarity hitherto found to exist among the melanocyte-stimulating hormones extends t o yet another species. Details of the study2' on a-MSH from horse glands ( ae-1lSH) are presented herein.
Experimental Xssays for melanocyte-stimulating activit>- were performed either on hypophysectomized Rana pipicizs" or oii isolated frog skins.Z* T h e early steps employed in isohting cr,-MSH from tlie posterior lobes of horse pituitctries followed closely those previously published for the X S H preparations from the other species.' The nielallocytestimulating activity was adsorbed onto iixycellulose iti thc iisual maimer aiid was in turn eluted from tlie cisycellulosc with 0.1 N HCI. Deacidification of this eluate IV:LS accomplished by batchwise treatment with t h e Xnilxrlite anion exchange resin IR-4B iii the hydroxide frirrn. Lyophylization of the deacidified solution yielded :Lwhite priwtler which possessed the MSH activi Countercurrent distribution d.1 was carried out i l l a solvent system consisting of acetic acid, t-but:Lnd and pyridine mixed in the \-olume ratios of 11: ,j:3.*5 lie:itelti with t h e s a m e partition coefficient a s t h a t of ue XIbH
4570
JONATHAN
Bovine
s. DIXONAND C H O H HA0 LI
Vol. 52 TABLE I1
Equine
O F ~ I , - > ~ S I IA N D ae-&fSH PREPARATIOSS
THE h f I S 0 ACID COMPOSITION
Ch-l
Amino acid
ab-MSH"
Glu
1
-4sp Ser Pro Ala
2 1
(:I?.
1
Ch-2
cyss
His Met 1.al Im1
1 1
1
Purified cr,-MSII h
1.0 0.3 2.1 1.3 0.1 1.4 O.06
.s. I ,
9
ae-hl SH
1.1 2.1 0.9 1.2
0.5 1.3 1.0
>
.I
Phe 1 1.B 1 1) Lys 1 1.1 08 Tyr 1 1 . !I 0 8 Arg 1 0.7 I .2 Try 1 1 1 a Taken from (11,12); results are expressed as molar ratios. Obtained after first countercurrent distribution. ,
Ch-3
Terminal Residue Analysis.-Digestion of aeMSH with carboxypeptidase for 24 hr. a t 38" failed t o release any amino acids; however, digestion with carboxypeptidase carried out for 6 hr. on a sample of the equine hormone that had stood for 70 I:ig. 2.-Resolution of chymotryptic digests ( i f a,-hISH minutes in 1 N HCl a t 80' released 0.47 mole of and ab-hlSH by zone electrophoresis on paper; buffer, -,- valine per mole of the peptide as well as apprecicdlidine-acetic acid-water at pH i . O and 4'; 11 volts/cni. able quantities of other amino acids. Thus, it may for 6 hr. be assumed that the C-terminal valine is in amide peptides.1q12,20From 1 kg. of posterior lobes of form, similar to oxytoxin and the vasopressins (in horse pituitaries, 0.038 g. of the hormone was iso- all of which the C-terminal amino acid is glycinalated. Table I summarizes the yield and estimated mide) . 3h MSH potency of various fractions obtained from The intact hormone did not give a positive nineach step of the procedure for the isolation of hydrin reaction,33 indicating that it probably conae-MSH. tains no free a-amino group. In addition, when TAIlLIl 1 cre-lISH was allowed to react with fluoodinitroYIELDASD POTENCY OF I-ARIOUS I~RACTICJSS OIITAISEI) benzene, nfi no a-dinitrophenyl amino acids were formed ; the only product was e-rlitiitrol)he~~ylFROM P R O C E D U R E F O R ~ S O I , h l ' I O N O F cYe-A1sIJ Yield from 1 k ~ . l i i t i m : i t c < l lysine. I t was also shown that among t h e pepf r r s h posterior hISH tide fragiiients obtainctl from either the chymotrypliixse ~ ~ i t i i i t a r i c > , i>cbtc,ncy, l'rt>ccdiirc i: tic or tryptic digestions. there \?;as cine which dit1 Acc.totie-dricd pci\vtlcr IRiJ 1 . 1 x 111: iiot react wit11 ninhylriii. Thew observatic>1lsnlkly (;larial acetic :tcitl e1tr:ict (i,,-Ix10; be taken to mean that the 0-imino group in the O\yccllulISH’Sisolated from the two species (horse and ox) are identical. Tryptic Digest.-The products resulting from the tryptic digestion of an ae-MSH sample were Tr-2 separated by the same procedures as those employed for the chymotryptic digests. Electrophoresis on paper of these tryptic digests gave the patterns shown in Fig. 3. Only two main bands were ob;at hode tained. Band Tr-1 was ninhydrin negative but gave a strong Pauly reaction, indicating the presFig. 3.-Resolution of tryptic digests of a,-MSH and ence of tyrosine and/or histidine, whereas Band ab-MSH by zone electrophoresis on paper; conditions same Tr-2 gave a strong color with ninhydrin and a posi- as in Fig. 2 . tive Ehrlich test for tryptophan. N-Terminal It is evident t h a t the amino acid sequence of residue analysis by the phenyl isothiocyanate the equine a-MSH has been almost completely (37) P. Edman, A c f a Chem. Scand., 4, 283 (1950). determined. It would appear from the identical (38) H. Fraenkel-Conrat and J. I. Harria, THIS JOURNAL, 7 6 , 6058 behavior manifested by it and the bovine a-MSH (1954).
1 4
that these two peptide hormones are also structurally identical, a t least with respect to amino acid sequence. I n addition, it can be said that the carboxyl group of the C-terminal amino acid residue, valine, is blocked, probably by an amide structure, since very inild x i d hydrulysis leaves the valine susceptible to rernoval by carboxypeptidase digestion I>oth tu,,-lISK’’Ih a r i d abZIS€I” ” 1 1 , l ~ IW*II e \ ~ I O L V I I to l~ l ~ l o c h ~ t:Itl t h e
[ C O Y T R I n l T I O S FROM THE
h‘-terminus by an acetyl residue The failure to react with ninhydrin on the part of the S terminal amino group of a,-MSH indicates similar blocking a t this terminus of the equine hormone as well, possibly also by an wet$ group. Acknowledgment -1Ye wish to thank the S a tional Institutes of Health of the United States Public Health Service for , I grant ft;?O07) whicli iii:icle this w \ r k pi)sGlilt..
BIOCHEMISTRY S E C T I O S , A R M O U R AND CO?,IPASY, CHICAGO 9, JLLISOIS]
CESTRAI. R E S E A R C H I , A D O R A T U R I E S ,
The Fractionation of Insulin on Diethylaminoethylcellulose 317 MARGUERITE VOLINI AND
MILTON
MITZ
RECEIVED JANUARY 29, 1960 Insulin was separated into two components on the anion exchanger diethylaminoethylcellulose, using stepwise elution with carbon dioxide-Tvater solution and 0.1 J1 ammonium phosphate. T h e fractions were compared as to biological proper ties, crystallizability, solubility, amino-acid content, optical rotation and electrophoretic mobility. They were identical in qualitative amino-acid content but displayed significant differences in physical configuration as well as in net charge. Their apparent interrelationship is t h a t of native and partially denatured protein forms. It is postulated t h a t the two insulins are composed of polymeric forms each containing a different monomeric species.
Introduction Heterogeneity in purified insulin preparations has been demonstrated by means of several criteria.’-’ In his sedimentation studies, Fredericq’ has obtained evidence that the component differences lie in the elementary sub-units of the insulin polymer. The classical work of Sanger8 shows that these differences are not concerned with the nature or sequence of the constituent amino acids; the fractionation studies of Fredericq,‘ Harfenist and Craig2, and Timasheff, Brown and Kirkwood3 collectively suggest that the differences can be attributed to variations in a few chemical groups, resulting in slight differences in net molecular charge. Indeed, Harfenist and Craig2 have actually showed that their A and B components, isolated by countercurrent distribution, differ by only a single amide group per 6000 niolecularweight unit. The purpose of the present paper is to report the separation of purified insulin into two components apparently unlike those obtained in prior investigations. Fractionation was accomplished on the anion exchanger diethylaminoethylcellulose (DEAIEcellulose) ‘3 by using stepwise elution with carbon dioxide-water solution’” and 0.1 -11 ammoniuin phosphate. The relation of the two fractions to ( I ) E. Fredericq, “Ciba Foundatiun Collaquia on Dndocrinology: Internal Secretions of the Pancreas,” 9, .Iand A. Churchill 1,td.. I,rindon, l H > ( i , p . SS-lO‘r. ( 2 ) E. J. I-Iarfenist a n d 1.. C Craig, THISJ O I T R N A L , 7 4 , 3083 (1932); ibid.,7 5 , 5,528 (l9>:3) (:i) S. S. Timasheff, I
