Proc. Nadl. Acad. Sci. USA Vol. 82, pp. 277-281, January 1985 Biochemistry

Processing of an anglerfish somatostatin precursor to a hydroxylysine-containing somatostatin 28 (pancreatic islets/hydroxylation/thermolysin/automatic Edman degradation/reversed-phase high-pressure liquid chromatography) JOACHIM SPIESS* AND BRYAN D. NOEtf *Peptide Biology Laboratory, Thb Salk Institute, 10010 North Torrey Pines Road, La Jolla, CA 92037; tDepartment of Anatomy, Emory University School of Medicine, Atlanta, GA 30322; and tMarine Biological Laboratory, Woods Hole, MA 02543 Communicated by Hans Neutath, August 29, 1984

ABSTRACT A novel 28-residue somatostatin (SS) has ref. 16 for a review) and considered as precursors to the been isolated from anglerfish pancreatic islets and character- smaller forms. ized by complete Edman degradation, peptide mapping, and For the investigation of SS precursor processing, the pan- amino acid analysis. The primary structure of this anglerfish creatic islet (Brockmann body) of anglerfish (Lophius ameri- SS-28 (a9S-28) containing hydroxylysine (Hyl) was established canus) has been especially attractive because of its low con- to be H-Ser-Val-Asp-Ser-Thr-Asn-Asn-Leu-Pro-Pro-Arg-Glu- tent of nonspecific proteases (17) and its relatively high con- Arg-Lys-Ala-Gly-Cys-Lys-Asn-Phe-Tyr-Trp-Hyl-Gly- tent of SS-containing D cells (18). Phe-Thr-Ser-Cys-OH. This sequence (with the exception of hy- We have demonstrated earlier in pulse-chase experiments droxylysine-23, which is replaced by lysine) is identical to the (17) and by chemical characterization (6) that SS-14 cleaved sequence of the COOH-terminal 28 residues of prepro-SS H from 12,000- to 14,000-dalton polypeptide precursors repre- predicted on the basis of cDNA analysis [Hobart, P., Craw- sents the predominant small SS produced by anglerfish pan- ford, R., Shen, L., Pictet, R. & Rutter, W. J. (1980) Nature creatic islets. (London) 288, 137-141J. This is the first instance in which By cloning and sequence analysis of cDNA generated hydroxylysine (to date characteristically observed in collagen from anglerfish pancreatic islet mRNA, the amino acid se- or collagen-like structures) has been found in a potential regu- quences of two distinct SS precursors were predicted (19). latory peptide. Chromatographic characterization of peptides, These precursors, prepro-SS I and II, contained at their radiolabeled in islet culture, revealed that aSS-28 contained COOH termini the sequences of oSS-14 or [Tyr7, GlylloSS- 10-12% of the radioactivity incorporated into the 8000- to 14, respectively. A third precursor was predicted (20, 21) to 1000-dalton SSlike polyjeptides, whereas 88-90% of this ra- differ from prepro-SS I by only two amino acids in the cen- dioactivity was detected in ahglerfish SS-14. It appears proba- tral part of the sequence. SS precursors corresponding to ble that aSS-28 represents the predominant primary cleavage anglerfish prepro-SS I also have been predicted from mam- product derived from prepro-SS II by cleavage at the COOH- malian (21-23) and catfish (24) mRNAs. terminal side of a single arginine. Based on knowledge of the In peptide mapping experiments, we demonstrated (25) collagen biosynthesis, it is speculated that hydroxylation may that the primary cleavage product of Pro-SS II is not [Tyr7, take place as an early post-translational event. Gly10]SS-14. We present here the complete chemical charac- terization of the main primary cleavage product, a 28-residue Somatostatin-14 (SS-14), a regulatory peptide discovered on SS-like polypeptide containing hydroxylated [Tyr7, Gly'0]- the basis of its inhibition of the secretion of hypophysial oSS-14. growth hormone (1), was characterized by sequence analysis after isolation first from ovine hypothalamus (2, 3) and sub- EXPERIMENTAL PROCEDURES sequently from porcine hypothalamus (4), pigeon (5), angler- fish (6), and channel catfish (7) pancreas. Thus, the primary Extraction and Purification of aSS-28 After Radiolabeling. structure of SS-14 was found to be completely conserved in After preincubation for 30 min at 20'C in Krebs-Ringer bi- different species and tissues. carbonate buffer, islets were incubated for 5 hr in the pres- It was observed that synthetic SS-14 riot only inhibited the ence of pairs of amino acids labeled with 3H and with either secretion of growth hormone with High potency but also the 4C or 3 S as described (17). secretion of many other hormones (see ref. 8 for a review), After radiolabeling, approximately 110 mg (wet weight) of including insulin and glucagon (9). pancreatic islet tissue was homogenized in 2 M acetic acid Based on immunologic and chromatographic evidence (8), with a Potter-Elvehjem homogenizer (300 rpm, 10 strokes). multiple forms of SS were recognized to occur besides SS-14 After centrifugation (800 x g for 10 min at 230C), the super- in various tissues of different species. One of these forms, natant was desalted with a Bio-Gel P-2 (100-200 mesh) col- porcine intestinal SS-28 (pSS-28) (10), also identified in ex- umn (2.5 x 18 cm) equilibrated with 2 M acetic acid. The tracts of porcine (11) and ovine (12, 13) hypothalamus, was material (2.24 mg of protein), which was eluted in the void determined by sequence analysis (10-13) to contain the volume, was subjected to gel filtration through a Bio-Gel P- structure of ovine hypothalamic SS-14 (oSS-14) at its COOH 30 column as described in Fig. 1. Approximately 125 ,g of terminus. Another form, catfish pancreatic SS-22 (cSS-22), protein was eluted with partition coefficients of0.3-0.5, cor- was found by sequence analysis (14, 15) to be significantly responding to a molecular weight range of 8000-2500, con- different from SS-14 and pSS-28. centrated by lyophilization, dissolved in 3 M acetic acid, and Besides the forms of SS mentioned above, 12,000- to filtered through 0.22-tim filters. The filtered peptides were 14,000-dalton SS-like polypeptides have been observed (see Abbreviations: SS, somatostatin; oSS-14, ovine hypothalamic SS- 14; aSS-28, anglerfish pancreatic islet SS-28; pSS-28, porcine intes- The publication costs of this article were defrayed in part by page charge tinal SS-28; cSS-22, catfish pancreatic SS-22; > PhNCS, phenyl- payment. This article must therefore be hereby marked "advertisement" thiohydantoin; PhNHCS, phenylthiocarbamyl; RP-HPLC, re- in accordance with 18 U.S.C. M734 solely to indicate this fact. versed-phase HPLC; CM, carboxymethylated.

277 Downloaded by guest on September 29, 2021 278 Biochemistry: Spiess and Noe Proc. NatL Acad. Sci. USA 82 (1985) usually S-carboxymethylated (13) and subsequently purified in two steps with reversed-phase HPLC (RP-HPLC) using 32.0 Vydac C18 columns (0.46 x 25 cm; particle size, 5 gm; pore I-- size, 330 A) and an aqueous mixture of 0.1% trifluoroacetic acid and acetonitrile for elution. Extraction and purification 25.6 were carried out at 230C and monitored by protein determi- 0 nation as described by Bradford (26). 19.2 Amino Acid Analysis. Peptides (0.3-5.0 pug) were hydro- x x lyzed for 22-24 hr at 110'C with 25 of 4 12.8 E ,ul M methanesulfonic _0 E -o acid, 0.2% tryptamine, or 2 M NaOH in the presence of 1 -o oi nmol of norleucine as internal standard. The hydrolysates 6.4 were treated with 25 ,ul of 3.5 M NaOH or 2.3 M methanesul- I fonic acid plus 0.12% tryptamine, respectively, and applied 0.0 to a Beckman 121 MB amino acid analyzer using sodium or 30 50 70 90 110 130 lithium citrate buffers for elution and ninhydrin for detec- Fraction tion. The coefficient of variation was usually between 1% and 10%. More details are published elsewhere (13). FIG. 1. Gel filtration of radiolabeled islet peptides. Approxi- Edinan Degradation. Peptides (0.5-2.0 nmol) were degrad- mately 105 mg (wet weight) of freshly dissected islet tissue was incu- ed in the presence of 3-4 mg of Polybrene in an automatic bated for 5 hr at 20'C with [3H]tryptophan (4.5 Ci/mmol; 1 Ci = 37 spinning cup sequencer (27) with a single-coupling single- GBq) and ['4C]isoleucine (345 mCi/mmol). The tissue was subse- cleavage program as described (28, 29). Phenylthiohydantoin quently washed, extracted, and applied to a (1.6 x 95 cm) column packed with Bio-Gel P-30 (100-200 mesh) and equilibrated with 2 M (> PhNCS)-conjugated amino acids were determined with acetic acid. Arrows mark the void volume and the elution volumes RP-HPLC (coefficient of variation, 1-6%; minimal detect- of anglerfish proinsulin (aP-I), insulin (al), glucagon (aG)/aSS-28 able amount, <10 pmol) (28). The main > PhNCS derivative (and [Lys23]aSS-28), and SS-14. The bar indicates the pool of the of 5-hydroxylysine (Calbiochem) subjected to automatic Ed- 8000- to 2500-dalton polypeptides subjected to RP-HPLC for further man degradation and RP-HPLC was presumably > PhNCS- purification. N8-phenylthiocarbamyl (PhNHCS)-hydroxylysine, which was eluted between > PhNCS-dehydrothreonine and > was detected in the 28-residue SS-like polypeptides. In view PhNCS-methionine with an absorption maximum at 262 nm. of the duration (5 hr) of the pulse period used here, it Was Back hydrolysis of the collected > PhNCS derivative of 5- estimated that the radioactivity distribution probably reflect- hydroxylysine with constant boiling HCl containing 0.1% ed the relative abundance of SS-14, aSS-28, and [Lys23]aSS- SnCl2 (30), 3 1.d of thioglycol/ml, and norleucine as internal 28 in anglerfish islets. standard (for 15-27 hr at 1400C) yielded a product that could - From one Brockmann body, approximately 11 pug of pep- not be distinguished from 5-hydroxylysine with the amino tide was purified with gel filtration and RP-HPLC. This acid analyzer using sodium or lithium citrate buffers. amount corresponded to -3.0 nmol of carboxymethylated aSS-28 (CM-aSS-28). No evidence for major contaminants RESULTS was provided by RP-HPLC under highly resolving condi- tions (Fig. 2). As mentioned above, we had established (25) that [Tyr7, Amino acid analysis with a sodium citrate program indi- Gly'0IoSS-14, which was predicted (19) to represent the cated that the amino acid composition of aSS-28 was identi- COOH-terminal sequence of prepro-SS II, was not ex- cal to the composition of the COOH-terminal 28 residues of pressed as such. Therefore, we screened pancreatic extracts prepro-SS 11(19) with the exception of the replacement of (after tissue culture in the presence of various radioactive one lysine residue by one hydroxylysine residue (Table 1). amino acids) for 8000- to 1000-dalton polypeptides contain- These data were confirmed by amino acid analyses with lith- ing most of the amino acids constituting [Tyr7, Glyl"]oSS-14. ium citrate buffers. Complete Edman degradation was per- This screening was carried out by gel filtration of the extract- formed with 2.0 nmol of natural CM-aSS-28. Only one major ed polypeptides and subsequent RP-HPLC of pools of the > PhNCS-amino acid was detected per Edman cycle. The 8OW- to 2500-dalton and 2500- to 1000-dalton polypeptides. relative > PhNCS-amino acid yields of consecutive cycles In the pool of the 2500- to 1000-dalton polypeptides, we did were in agreement with the assumption that only one major not find a species that was labeled with radioactive tyrosine, peptide was degraded in the spinning cup. Minor amounts of glycine, cysteine, and tryptophan as expected for an extend- additional > PhNCS-amino acids were identified in some cy- ed sequence of [Tyr7, Glyl0]oSS-14. However, the pool of cles. The yields of these > PhNCS-amino acids never ex- the 8000- to 2500-dalton polypeptides contained a species ceeded 10% of the major > PhNCS-amino acid yields. The that could be radiolabeled with tyrosine, glycine, trypto- yields of > PhNCS-amino acids derived from CM-aSS-28 phan, cysteine, asparagine, threonine, valine, arginine, ly- ranged from 420 pmol (serine-i) to 1600 pmol (valine-2) to 60 sine, and leucine. Chromatographic results of one labeling pmol (CM-cysteine-28). In cycle 23, the > PhNCS derivative experiment in which the pair [3H]tryptophan and [14C]isoleu- of hydroxylysine was determined as the major > PhNCS- cine Was used is presented as an example for this approach conjugated amino acid. No evidence for > PhNCS-N8- (Fig. 1). PhNCS-lysine was found in this cycle, whereas in cycles 14 In subsequent experiments with RP-HPLC, it was re- and 18, > PhNCS-Ne-PhNHCS-lysine represented the major Vealed that the species discovered in the pool of the 8000- to > PhNCS-amino acid. No > PhNCS derivative of hydroxy- 2500-dalton polypeptides was mainly composed of two com- lysine was found in these latter cycles. ponents, which later were identified as the 28-residue SS- The data derived from Edman degradation were con- like polypeptides aSS-28 and [Lys23]aSS-28. aSS-28, which firmed and supplemented by peptide mapping of CM-aSS-28 eluted before [Lys23JaSS-28, represented the predominant fragments obtained by digestion with thermolysin (Fig. 3). fdrm, carrying >95% of the total radioactivity of the 28-resi- Based on the absorbance profile (Fig. 3) and amino acid due SS-like polypeptides. Analysis of the distribution of ra- analysis (Table 1), the major fragments cleaved from natural dioactivity over the 8000- to 1000-dalton SS-like polypep- CM-aSS-28 were identified to be CM-aSS-28-(1-19)-OH, tides showed that -88-90% of the radioactivity found in SS- aSS-28-(20-24)-OH and CM-aSS-28-(25-28)-OH. The frag- like peptides was incorporated in SS-14, whereas 12-10% mentation pattern matched with the substrate specificity of Downloaded by guest on September 29, 2021 Biochemistry: Spiess and Noe Proc. NatL Acad Sci. USA 82 (1985) 279

CM-aSS-28 CM-[Tyr7Gly1O]oSS- 14 100

0II

.00 5000 0.025 0 8 -00

O

0.000 0 10 20 30 Time, min

FIG. 2. RP-HPLC of -0.4 1tg (0.13 nmol) of purified CM-aSS-28. The peptide was applied to a Vydac C18 column (0.46 x 25 cm; particle size, 5 /Am; pore size, 330 A) and eluted with a mixture of aqueous 0.05% trifluoroacetic acid and acetonitrile. RP-HPLC was performed with a Hewlett-Packard 1084B liquid chromatograph equipped with a variable wavelength detector. Arrows mark the retention times of synthetic peptides. The absorbance increase immediately after sample application was not caused by peptide as indicated by amino acid analysis. thermolysin, which has been observed to cleave preferential- acid ratios from integers in amino acid analysis (Table 1), ly peptide bonds at the NH2-terminal side of hydrophobic Edman degradation, and peptide mapping on RP-HPLC amino acids as long as they are not COOH-terminally linked (Fig. 3). to proline (31). The retention time ofCM-aSS-28-(25-28)-OH Since hydroxylysine of collagen has often been found to did not deviate significantly from the retention time of CM- be linked O-glycosidically to glucose or galactose or gluco- oSS-14-(11-14)-OH cleaved from synthetic CM-oSS-14 with sylgalactose (32), the possibility that hydroxylysine-23 of thermolysin. However, aSS-28-(25-28)-OH was eluted sig- CM-aSS-28 might be glycosylated was examined. We fol- nificantly later than its COOH-terminally amidated analog lowed closely a procedure that had been used by Brownell CM-oSS-14-(11-14)-NH2 cleaved from synthetic CM-oSS- and Veis (33) in the characterization of carbohydrate deriva- 14-NH2. Therefore, it was concluded that CM-aSS-28, and tives of hydroxylysine of collagen. This procedure is based thus aSS-28, carry a free COOH-terminal carboxylic group. on the higher stability of acetalic bonds under alkaline condi- aSS-28-(20-24)-OH containing hydroxylysine-23 was re- tions compared to acidic conditions. When purified human solved from its nonhydroxylated analog [Tyr7, Gly'0-oSS-14- collagen, type II, was hydrolyzed with 4 M methanesulfonic (6-10)-OH cleaved from synthetic [Tyr7, Glyl0]oSS-14. acid and 2 M NaOH, the amino acid ratios of hydroxylysine The purity of natural CM-aSS-28 was estimated to be to the sum of hydroxylysine and lysine were 1/2.2 and 1/3.3, >90%o on the basis of RP-HPLC (Fig. 2), deviation of amino respectively. The finding of a lower ratio after alkaline hy- drolysis indicated the presence of carbohydrate linked to hy- Table 1. Amino acid composition of CM-aSS-28 and droxylysine. For CM-aSS-28, a ratio of 1/3.1 was indepen- its fragments % dent of the type of hydrolysis applied. Thus, it was conclud- Acid hydrolysis* ed that hydroxylysine-23 of CM-aSS-28 probably was not Edman glycosylated. Partial glycosylation appeared highly improba- Amino degradation of Fragment of CM-aSS-28 ble in view of the peptide mapping experiment (Fig. 3), acid CM-aSS-28 CM-aSS-28 1-19 20-24 25-28 which provided evidence for only one hydroxylysine-con- taining fragment. CM-Cys 2 1.9 (2) 1.0 (1) 1.1 (1) On the basis of the data presented here, Edman degrada- Asx 4 4.0 (4) 3.9 (4) tion, peptide mapping, and amino acid analysis, the primary Thr 2 1.9 (2) 0.9 (1) 0.9 (1) structure of aSS-28 was established (see Fig. 4). With a simi- Ser 3 2.7 (3) 1.8 (2) 0.9 (1) lar approach, the minor 28-residue SS-like polypeptide of Glx 1 1.1 1.1 0.1 (1) (1) (0) pancreatic islets was completely characterized to be [Lys23]- Pro 2 2.3 (2) 2.1 (2) aSS-28 (37). Gly 2 2.1 (2) 1.3 (1) 1.2 (1) 0.2 (0) Ala 1 1.0 (1) 0.9 (1) Val 1 0.9 (1) 0.9 (1) DISCUSSION Leu 1 1.1 (1) 1.1 (1) In the process of screening for an SS-like polypeptide con- Tyr 1 1.0 (1) 1.0 (1) taining the sequence of [Tyr7, Glyl0]oSS-14, we discovered Phe 2 2.1 (2) 1.0 (1) 1.0 (1) and completely characterized a novel somatostatin, aSS-28, Hyl 1 1.1 (1) 1.2 (1) the sequence of which corresponds to the COOH-terminal 28 Lys 2 2.2 (2) 2.0 (2) residues of prepro-SS II except for lysine-23, which is re- Trp 1 1.1 (1) 0.9 (1) placed by hydroxylysine. Arg 2 2.0 (2) 2.1 (2) The finding of hydroxylysine in aSS-28 is surprising be- *Approximately 0.1-1.1 ,ug of peptide was hydrolyzed with 4 M cause to date this amino acid has only been observed as a methanesulfonic acid and analyzed as described. Amino acid ratios component of collagen and collagen-like structures (32). In are presented; the nearest integers are given in parentheses. collagen, hydroxylysine is located NH2-terminally to glycine Downloaded by guest on September 29, 2021 280 Biochemistry: Spiess and Noe Proc. NatL Acad ScL USA 82 (1985)

1 5 10 15 1 20 W K 125 S VD S T N N L PP R E R K A G C K N F 4Y W K!IG F T S C 100 20-24 FTSCN 0.030 25-28 IV -l

I

50 a, 0. E 0ac a) 0.015 F 0 Nsc-

Jo

0.000 L

0 10 20 30 40 Time, min

FIG. 3. Peptide mapping of natural CM-aSS-28 on RP-HPLC. CM-aSS-28 (- 2.0 nmol) was digested in 50 Al of 0.2 M N-ethylmorpholine (adjusted to pH 8.0 with trifluoroacetic acid)/2 mM CaCl2/5% (vol/vol) 1-propanol containing thermolysin (substrate-to-enzyme weight ratio, 20:1) for 1 hr at 30'C. Subsequently, the mixture was applied to RP-HPLC with a Perkin-Elmer series 4 liquid chromatograph equipped with a Kratos Spectroflow 773 monitor. Chromatography was carried out on a Vydac C18 column as described in the legend to Fig. 2. All products detected in the eluate on the basis of their absorbance were subjected to amino acid analysis. Products I-IlI contained <0.1 pg of peptide. The identity of major products [CM-aSS-28-(25-28)-OH, 1.0 ,ug; CM-aSS-28-(1-19)-OH, 2.0 ,ug; aSS-28-(20-24)-OH, 1.2 gg] is indicated (see also Table 1). Products IV (0.26 jig of peptide) and V (0.13 ,ug) probably represent aSS-28-(21-24)-OH and CM-aSS-28-(1-20)-OH, respectively. The retention times of CM-oSS-14-(11-14)-NH2 (FTSC*) and [Tyr7, Gly10]SS-14-(6-10)-OH (FYWKG) are also denoted by arrows. The fragmenta- tion is illustrated at the top of the figure with major ( ) and minor ( ) cleavage sites. A, alanine; C, cysteine (all cysteine residues shown here were carboxymethylated); D, aspartic acid; E, glutamic acid; F, phenylalanine; G, glycine; H, histidine; I, isoleucine; K, lysine; K*, hydroxyly- sine; L, leucine; M, methionine; N, asparagine; Q, glutamine; R, arginine; S, serine; T, threonine; V, valine; W, tryptophan; Y, tyrosine; m, amide.

residues. It has been shown that collagen hydroxylysine pro- part, oSS-14-(6-11) which (with the exception of threonine- vides sites for glycosylation and crosslinking (32). For the 10) has been recognized (34) to be essential for the potent latter reaction, oxidative desamination of the hydroxylysine inhibition of growth hormone secretion from cultured rat an- side chain seems to be required. Since hydroxylysine-23 of terior pituitary cells. Thus, aSS-28 is not expected to be a aSS-28 was neither observed to be glycosylated nor desa- potent inhibitor of hypophyseal growth hormone secretion, minated, its presence may be suggestive for a spectrum of as has been shown already for the synthetic analog [Tyr7, (pancreatic) biologic activities of aSS-28 significantly differ- Gly'0]oSS-14 (Wylie Vale, personal communication). ent from the spectrum of SS-14 activities. It has been demonstrated in experiments with free and This view is supported also by a structural comparison of membrane-bound ribosomes (35) that hydroxylation of ly- aSS-28 with other peptides of the somatostatin family (Fig. sine residues in collagen is apparently catalyzed by mem- 4). Because of the significant size variation within this pep- brane-bound lysyl hydroxylase and probably represents a tide family, the sequences of the smaller peptides have to be cotranslational event. A similar mechanism may hold for the compared with the corresponding COOH-terminal parts of hydroxylation of aSS-28. In view of the location of the hy- the larger peptides. Thus, the sequence homologies of aSS- droxylation site in the COOH-terminal sequence of aSS-28, 28 with [Lys23]aSS-28, pSS-28 (10-13), oSS-14 (2-7), and and in view of the size of the participating subcellular struc- cSS-22 (14, 15) were calculated to be 96%, 61%, 79%, and tures such as ribosomes and membranes, hydroxylation may 23%, respectively. A sequence comparison between aSS-28 take place in this instance as an early post-translational and oSS-14 reveals that the primary structure of aSS-28 devi- event (Fig. 5). ates significantly from the sequence of oSS-14 in the central Based on the pattern of the radioactivity distribution in the

1 5 10 15 20 25

aSS-28 SVDST NNL P P RE R KAGC KN FYWK G F TS C [Lys23]aSS-28 SVDSTNNL P P RERRkAGCKN F YWK G F T S C pSS-28 S A N S N P AMAPRERK A G C K N F FWK T F T S C 1 5 10 oSS-14 A G C K NF FWK T F T S C 1 5 10 15 20 cSS-22 D N T V T S K P L N C M N Y F W K SR T A C

FIG. 4. Amino acid sequence homologies (shaded) between aSS-28 and the other members of the SS family. Downloaded by guest on September 29, 2021 Biochemistry: Spiess and Noe Proc. NatL Acad Sci USA 82 (1985) 281

ANGLERFISH 3. Burgus, R., Ling, N., Butcher, M. & Guillemin, R. (1973) Proc. Nadl. Acad. Sci. USA 69, 684-688. 1_ 4. Schally, A. V., Dupont, A., Arimura, A., Redding, T. W., Ni- shi, N., Linthicum, G. L. & Schlesinger, D. H. (1976) Bio- PEANGLERFISH chemistry 15, 509-514. 2 tHyI23]-Pro-SS II tWS-aSSS I 5. Spiess, J., Rivier, J., Rodkey, J. A., Bennett, C. D. & Vale, W. (1979) Proc. Natl. Acad. Sci. USA 76, 2974-2978. LH \\ ,/PCE 6. Noe, B. D., Spiess, J., Rivier, J. & Vale, W. (1979) Endocri- nology 105, 1410-1415. Pro-SS I} 7. Andrews, P. C. & Dixon, J. (1981) J. Biol. Chem. 256, 8267- (SP 8270. 8. Vale, W., Rivier, C. & Brown, M. (1977) Annu. Rev. Physiol. Prepro-SS II (AF) 39, 473-527. 9. Koerker, D., Ruch, W., Chideckel, E., Palmer, J., Goodner, C. J., Ensinck, J. & Gale, C. C. (1974) Science 184, 482-484. Prepro-SS I (MAM, AF, CF) 10. Pradayrol, L., Jornvall, H., Mutt, V. & Ribet, A. (1980) FEBS {SP Lett. 109, 55-58. 11. Schally, A. V., Huang, W. Y., Chang, L. C., Arimura, A., Pro-SS I Redding, T. W., Millar, R. P., Hunkapiller, M. W. & Hood, L. E. (1980) Proc. Natl. Acad. Sci. USA 77, 4489-4493. 12. Esch, F., Bohlen, P., Ling, N., Benoit, R., Brazeau, P. & Guil- lemin, R. (1980) Proc. Natl. Acad. Sci. USA 77, 6827-6831. 13. Spiess, J., Villarreal, J. & Vale, W. (1981) Biochemistry 20, MAMMALS MAMMALS, 1982-1988. ANGLERFISH, 14. Oyama, H., Bradshaw, R. A., Bates, 0. J. & Permutt, A. CATFISH (1980) J. Biol. Chem. 255, 2251-2254. 15. Magazin, M., Minth, C. D., Funckes, C., Deschenes, R., Ta- FIG. 5. Enzymatic processing of SS precursors to various forms vianini, M. & Dixon, J. (1982) Proc. Natl. Acad. Sci. USA 79, of somatostatin. AF, anglerfish; CF, catfish; LH, lysyl hydroxylase; 5152-5156. MAM or m, mammalian; PCE, prohormone converting enzyme; SP, 16. Noe, B. D. (1984) in Current Trends in Tumour Pathology, signal peptidase. eds. Polak, J. M. & Bloom, S. R. (Churchill Livingstone, Ed- inburgh), in press. gel filtration and RP-HPLC of radiolabeled islet peptides, it 17. Noe, B. D., Fletcher, D. J. & Spiess, J. (1979) Diabetes 28, has been concluded that aSS-28 represents the predominant 724-730. smallest SS-like polypeptide containing the (modified) se- 18. Johnson, A., Torrence, J., Elde, R., Bauer, G., Noe, B. & quence of [Tyr7, Gly10]oSS-14 derived from prepro-SS II. In Fletcher, D. (1976) Am. J. Anat. 147, 119-122. 19. Hobart, P., Crawford, view of the appearance of aSS-28 in the early chase period of R., Shen, L.-P., Pictet, R. & Rutter, W. J. (1980) Nature (London) 288, 137-141. pulse-chase experiments (unpublished results), it seems 20. Goodman, R. H., Jacobs, J. W., Chin, W. W., Lund, P. K., probable that aSS-28 represents the predominant primary Dee, P. C. & Habener, J. F. (1980) Proc. Natl. Acad. Sci. cleavage product in the processing of anglerfish prepro-SS II USA 77, 5869-5873. (Fig. 5). 21. Goodman, R. H., Jacobs, J. W., Dee, P. C. & Habener, J. F. It is interesting that anglerfish prepro-SS I is processed to (1982) J. Biol. Chem. 257, 1156-1159. SS-14 by proteolytic cleavage at the COOH-terminal side of 22. Shen, L.-P., Pictet, R. L. & Rutter, W. J. (1982) Proc. Nail. a pair of basic residues (-Arg-Lys-), whereas prepro-SS II, Acad. Sci. USA 79, 4575-4579. which also contains these residues and is structurally ho- 23. Funckes, C. L., Minth, C. D., Deschenes, R., Magazin, M., Tavianini, M. H. mologous in the environment of this dibasic pair, was not A., Sheets, M., Collier, K., Weith, L., Aron, D. C., Roos, B. A. & Dixon, J. E. (1983) J. Biol. Chem. 258, observed to be cleaved at this site. Instead cleavage was 8781-8787. found to occur between a single arginine and serine-1 of aSS- 24. Taylor, W. L., Collier, K. J., Deschenes, R. J., Weith, H. L. 28. An equivalent cleavage of an SS-28 from anglerfish pre- & Dixon, J. (1981) Proc. Natl. Acad. Sci. USA 78, 6694-6698. pro-SS I has not been detected. In contrast with these find- 25. Noe, B. D. & Spiess, J. (1983) J. Biol. Chem. 258, 1121-1128. ings, both mammalian SS-28 and SS-14 may be cleaved from 26. Bradford, M. (1976) Anal. Biochem. 72, 248-254. the mammalian SS precursor corresponding to anglerfish 27. Wittmann-Liebold, B. (1980) in Polypeptide Hormones, eds. prepro-SS I (Fig. 5). To date, a mammalian equivalent of Beers, R. F., Jr., & Bassett, E. G. (Raven, New York), pp. prepro-SS II has not been discovered. 87-120. 28. Spiess, J., Rivier, J., C. & W. in Methods The observation of different processing pathways is sug- Rivier, Vale, (1982) in Protein Sequence Analysis, ed. Elzinga, M. (Humana, Clif- gestive for either a high substrate specificity of the proteases ton, NJ), pp. 131-138. involved in the SS precursor conversion (36) or the localiza- 29. Spiess, J., Rivier, J. & Vale, W. (1983) Biochemistry 22, 4341- tion of precursors or proteases in different cells, or both. 4346. 30. Mendez, E. & Lai, C. Y. (1975) Anal. Biochem. 68, 47-53. 175-189. We thank Dr. J. Rivier for providing synthetic peptides, Wilson 31. Heinrikson, R. L. (1977) Methods Enzymol. 47, Woo, David Karr, Gail Debo, and Patricia Moxley for excellent 32. Prockop, D. J., Kivirikko, K. I., Tuderman, L. & Guzman, N. A. (1979) N. Engl. J. Med. 301, 13-23. technical assistance, and Susan Logan and Dorothy Weigand for 33. Brownell, A. G. & Veis, A. (1975) Biochem. Res. manuscript preparation. This research was supported by National Biophys. Institutes of Health Grants AM26378 and AM16921 and the Clayton Commun. 63, 371-377. 34. Vale, W., Rivier, J., Ling, N. & Brown, M. (1978) Metabolism Foundation for Research-California Division. J.S. is a Clayton 27, 1391-1401. Foundation Investigator. 35. Harwood, R., Grant, M. E. & Jackson, D. S. (1975) Biochem. J. 152, 291-302. 1. Krulich, L., Dhariwal, A. P. S. & McCann, S. M. (1968) En- 36. Noe, B. D., Debo, G. & Spiess, J. (1984) J. Cell Biol. 99, 578- docrinology 83, 783-790. 587. 2. Brazeau, P., Vale, W., Burgus, R., Butcher, M., Rivier, J. & 37. Spiess, J. & Noe, B. (1984) in Somatostatin, eds. Patel, Y. & Guillemin, R. (1973) Science 179, 77-79. Tannenbaum, G. (Plenum, New York), in press. Downloaded by guest on September 29, 2021