Insulin - Like Growth Factors. Somatomedins: Basic Chemistry, Biology and Clinical Importance. Proceedings of a Symposium on Insulin-Like Growth Factors, Somatomedins, Nairobi, Kenya, November 13–15, 1982 9783110866490, 9783110095623

Citation preview

SYMPOSIUM ON INSULIN-LIKE GROWTH FACTORS/ SOMATOMEDINS NAIROBI, KENYA - NOVEMBER 13-15, 1982

, r

w. X

i

^»tp*

w »

W

Wm '

/. , »

i r

/*

/f«

Is

7 Am '

«ii&K

3»

„ M rl^i, " ' 'W • \ -» Wf^Sfe

¡ ¡ M S *

)

f

»J

• W ' C ^ u i

»

«a

• P 1«

SfSPKL

H

m i «

' "

K

i

W f m fcUf

' .

rl

»••if

y ' V ' t •r' ¡T

•J*-

h r O

1 HgiiilfcTPt

W*« 1 **

0

»• 11 *

t

•* Immunoperoxidase staining of insulin-like growth factor-l (Somatomedin-C) in the articular cartilage of a growth hormone deficient Little (lit/lit) mouse treated with growth hormone. The reddish color indicates positive staining, except in the bone marrow which contains a high amount of endogenous peroxidase.

Insulin-Like Growth Factors Somatomedins Basic Chemistry • Biology Clinical Importance Proceedings of a Symposium on Insulin-Like Growth Factors/Somatomedins Nairobi, Kenya, November 13-15,1982 Editor E. Martin Spencer

W DE G Walter de Gruyter • Berlin • New York 1983

Editor: E. Martin Spencer, M. D. Ph. D. Laboratory of Growth and Development Children's Hospital of San Francisco San Francisco, Ca. 94119 U.S.A.

CIP-Kurztitelaufnahme

der Deutschen

Bibliothek

Insulin-like growth factors, somatomedins: bas. chemistry • biology clin. importance; proceedings of a Symposium on Insulin-Like Growth Factors, Somatomedins, Nairobi, Kenya, November 13-15,1982 / ed. E. Martin Spencer. - Berlin; New York: de Gruyter, 1983. ISBN 3-11-009562-9 NE: Spencer, E. Martin [Hrsg.]; Symposium on Insulin-Like Growth Factors, Somatomedins «1982, Nairobi»

Library of Congress Cataloging

in Publication

Data

Symposium on Insulin-Like Growth Factors / Somatomedins (1982: Nairobi, Kenya) Insulin-Like growth factors, somatomedins. "Satellite symposium to the 11th Congresses of the International Diabetes Federation" - Pref. Includes bibliographies and indexes. 1. Somatomedin—Congresses. I. Spencer, E. Martin, 1929-. II. International Diabetes Federation. Congress (11th: 1982: Nairobi, Kenya) III. Title. [DNLM: 1. Somatomedins— Congresses. 2. Nonsuppressible insulin-like activity—Congresses. WH 400 S9885i 1982] Q P 5 5 2 . S 6 5 S 9 4 1 9 8 2 599'.031 83-7838 ISBN 3-11-009562-9

Copyright © 1983 by Walter de Gruyter & Co., Berlin 30. All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced in any form - by photoprint, microfilm or any other means - nor transmitted nor translated into a machine language without written permission from the publisher. Printing: Gerike GmbH, Berlin. - Binding: Dieter Mikolai, Berlin. Printed in Germany.

Preface

The Symposium on Insulin-Like Growth Factors/ Somatomedins was held in Nairobi, Kenya on November 13-15, 1982, as a satellite symposium to the 11th Congress of the International Diabetes Federation. The last international symposium on insulin-like growth factors had been held 4 1/2 years previously. During this time many major discoveries were made culminating in the direct demonstration that pure insulin-like growth factors promote growth in vivo. The increasing importance of this area of research is attested to by the large number of new workers who have entered the field and the exponential increase in papers. Because of the truly multinational character of research in this area, an international forum was desired for this meeting. The genetic relationship of the insulin-like growth factors to insulin made it appropriate to hold this meeting in conjunction with the International Diabetes Federation meeting in Nairobi, Kenya. The Symposium was held during the weekend recess of the parent conference. Representatives were present from many countries, including Switzerland, Great Britain, Australia, United States, Japan, Canada, France, West Germany, Kenya, Hong Kong, Denmark, South Africa, and Nigeria. The format consisted of invited lecturers, oral communications, and two poster sessions. The surroundings afforded a conducive environment for interaction between investigators, both during the conference and afterwards on safari. The organizers are particularly indebted to the following sponsors, without whose generous contributions the Symposium could not have been held: Hoffman-La Roche, Inc., 11th Congress of the International Diabetes Federation, Shionoge Corporation, International Mineral & Chemical Corporation, Pfizer Central Research, Monsanto Company, Smith, Kline Clinical Laboratories,

VI

KabiVitrum, Nichols Institute, Sumitomo Corporation, Mead Johnson & Company, Miles/Bayer Laboratories, Sandoz Pharmaceuticals, Children's Hospital of San Francisco, Hoechst A.G., Genentech, Inc., Connaught, Labs, Serono Laboratories, Inc., Pharmacia Laboratories, Inc., Dako Corporation, Beckman Instruments, Swissair, Speywood Labs. Contributions were also obtained from: AMGen, LKB Produkter AB, Endocrine Sciences, Cetus, Upjohn Company, Adria Laboratories, Lilly Research Laboratories, Becton Dickinson, Ross Laboratories, Merck Sharp & Dohme Research Laboratories, Schering Corporation, New England Nuclear. San Francisco, May 1983 E. Martin Spencer

CONTENTS

THE THREE HISTORICAL ORIGINS OF INSULIN-LIKE GROWTH FACTOR RESEARCH

The Somatomedin Hypothesis: Developments W.H. Daughaday From NSILA to IGF: and Breakthroughs E.R. Froesch

Origins and Recent 3

A Look Back on the Major Advances

Multiplication Stimulating Activity for Cells in Culture S.P. Nissley, S.O. Adams, A.M. Acquaviva, Y.W.-H. Yang, C.B. Bruni, G.P. August, R.M. White, T.P. Foley, Jr., A.C. Moses, K.L. Cohen and M.M. Rechler

13

31

IN VIVO ACTION OF INSULIN-LIKE GROWTH FACTORS

Long-Term in Vivo Effects of Insulin-Like Growth Factors (IGF) I and II on Growth Indices: Direct Evidence in Favor of the Somatomedin Hypothesis E. Schoenle, J. Zapf and E.R. Froesch

51

125x_iGF Binding Patterns in Serum and Glucose Transport in Fat Cells from Hypox Rats after Long-Term Treatment with IGF I, IGF II or Growth Hormone (GH): Evidence for Effects of GH Not Mediated by IGF J. Zapf, E. Schoenle and E.R. Froesch

57

Insulin-Like Growth Factors : Direct CNS Effects on Pulsatile Growth Hormone Secretion and Body Weight Regulation G.S. Tannenbaum, H.J. Guyda and B.I. Posner

63

The Synlactin Hypothesis: Prolactin's Mitogenic Action May Involve Synergism with a Somatomedin-Like Molecule T.R. Anderson, J. Rodriguez, C.S. Nicoli and E.M. Spencer

71

VIII STRUCTURE AND PURIFICATION OF INSULIN-LIKE GROWTH FACTORS

The Identity of Human Insulin-Like Growth Factors I and II w i t h Somatomedins C and A and Homology w i t h Rat IGF I and II E.M. Spencer, M. Ross and B. Smith

81

A Computer Graphics Study of Insulin-Like Growth Factors and Their Receptor Interactions A. Honegger and T. Blundell

97

Evidence for Proteolytic Conversion of Insulin-Like Growth Factors to a Biologically Active Acidic Form A.C. Herington and A.D. Kuffer

113

IGF-Like Characteristics of a n Acidic Non-Suppressible Insulin-Like Activity A.C. Herington and A.D. Kuffer

121

Somatomedin-Like Activity in Bovine Serum K. Ray, M. Wallis and A. Holder

127

CARRIER PROTEIN FOR INSULIN-LIKE GROWTH FACTORS

Serum Forms of Insulin-Like Growth Factors and Their Carrier Proteins R.L. Hintz and F. L i u

133

Characterization of the IGF Binding Proteins (BPs) Produced by the Liver in Organ Culture. Their Relations w i t h Serum BPs and Cerebrospinal Fluid BPs P. Hossenlopp, S. Hardouin, C. Lassarre, B. SegoviaQuinson and M. Binoux

139

A Human Hepatoblastoma-Derived Cell Line (HEP G2) Secretes a Specific IGF Carrier Protein A.C. Moses, A.J. Freinkel, B.B. Knowles and D.P. A d e n ..

145

MEASUREMENT OF INSULIN-LIKE GROWTH FACTORS

Determination of Insulin-Like Growth Factors: of Methods J. Zapf

A Survey

The Use of Synthetic Peptides for the Development of Radioimmunoassays for the Insulin-Like Growth Factors R.L. Hintz, F. Liu, D. Chang and E.R. Rinderknecht

155

169

IX

Measurement of Insulin-Like Growth Factors: Special Considerations Related to Basic Somatomedin in Serum R.M. Bala, B. Bhaumick and M.S. Sheppard

177

REGULATION OF PLASMA LEVELS OF INSULIN-LIKE GROWTH FACTORS

Nongrowth Hormone Dependent Hormonal Regulation of Plasma Somatomedin Levels R.W. Furlanetto

197

Undernutrition and Inhibitors as Regulators of IGF Plasma Levels and Cellular A c t i o n H.D. Mosier, Jr. and D.J. Knauer

211

Thymidine Inhibitory Activity of R.at Serum: Its Influence o n Cornea and Cartilage in Starved and Hypophysectomized Rats H.D. Mosier, Jr., M.A. Mosier and R.A. Jansons

223

Effect of Cold Stress on Plasma Somatomedin Activity (SM) and Growth in Rats G.S.G. Spencer and G.J. Garssen

229

CLINICAL USES OF PLASMA INSULIN-LIKE GROWTH FACTOR LEVELS

Plasma Immunoreactive Somatomedin-C/IGF I in the Evaluation of Short Stature L.E. Underwood, D.R. Clemmons, J.J. Van Wyk, P.G. Chatelain and K.C. Copeland

235

Age Related Variations of IGF (Insulin-Like Growth Factor) and IGF BP (IGF Binding Protein) Serum Levels in Normal Children and Adolescents. Comparison w i t h Levels in Children w i t h Constitutional Short Stature M. Gourmelen, F. Girard and M. Binoux

255

Somatomedin Activity in Patients w i t h Growth Retardation Due to Hypopituitarism or Familial-Constitutional Growth Delay H. Jasper, A. Martinez and J. Heinrich

259

Insulin-Like Growth Factors in Pygmies: Characterization of the Metabolic Actions of IGF I and IGF II in M a n T.J. Merimee, J. Zapf and E.R. Froesch

263

X Somatomedins in the AKA Pygmies from "Basse-Lobaye" R.-M. Schimpff, A.-M. Repellin, B. Leduc, P. Gamier, J.-C. Job and G. Jaeger

271

Somatomedin and GH Measurements in Acromegaly W.H. Daughaday, P.E. Cryer and I.K. Mariz

277

Comparison of Somatomedin C with Growth Hormone Levels in Evaluating Therapeutic Response in Treated Acromegaly L.D. Stonesifer, R.M. Jordan and P.O. Kohler

285

Plasma Somatomedin in Diabetics with Retinopathy and Joint Contractures I.K. Ashton, T.L. Dornan, B. Haitas and R.C. Turner ....

289

Insulin-Like Growth Factors in Adult Diabetics T.J. Merimee, J. Zapf and E.R. Froesch

295

Evolution of Serum IGF (Insulin-Like Growth Factor) Levels in Patients with Insulin-Dependent Diabetes during Severe Ketosis and Reequilibration M. Rieu, G. Tchobroutsky and M. Binoux

299

Is C-Peptide a Marker for Retinal Angiogenesis Factor? M.A. Mosier

303

The Role of Somatomedins in Psychiatric Disorders V.R. Sara, K. Hall and L. Wetterberg

311

Insulin-Like Growth Factor (IGF) Levels Measured by Radioimmunoassay (RIA) and Radioreceptorassay (RRA) in Various Forms of Tumor Hypoglycemia U. Widmer, J. Zapf, E.R. Froesch and M.C. Kew

317

Hypoglycaemia in Primary Hepatoma R.T.T. Yeung, D.C.Y. Yeung and S.S.C. Wong

325

INSULIN-LIKE GROWTH FACTORS IN FETAL GROWTH A N D DEVELOPMENT

Role of Somatomedins/Insulin-Like Growth Factors in the Regulation of Fetal Growth L.E. Underwood, P.B. Kaplowitz, A.J. D'Ercole

331

Reduced Plasma Somatomedin Activity during Experimental Growth Retardation in the Fetal and Neonatal Rat D. Hill, M. Fekete, D. Milner, F. De Prins and A. Van As sehe

345

XI

Stimulation of Thymidine Incorporation into Fetal Rat Cartilage In vitro by Human Somatomedin, Epidermal Growth Factor and Other Growth Factors D. Hill, D. Milner, J. Seid, S. Tomlinson, A. Holder and M. Preece

353

The Potential of Insulin as a Regulator of Fetal Somatomedin Production I. Fennoy, H.J. Eisen and R.M. White

357

Increased Somatomedin Activity (SM) Following Chronic Hyperinsulinaemia in Fetal Pigs G.S.G. Spencer, G.J. Garssen, D.J. Hill, B. Colenbrander and A.A. Macdonald

365

Serum Growth-Promoting Activity of Human Newborns and Mothers Measured as 3H-Thymidine Incorporation into Human Activated Lymphocytes R.-M. Schimpff, J.-C. Job, M. Bozzola, G. Mingrat, M. Ghini, E. Polito and F. Severi

373

BIOLOGICAL ACTIONS OF INSULIN-LIKE GROWTH FACTORS

Role of Somatomedins in the Regulation of the Animal Cell Cycle B.J. Bockus, M.A. Chaikin and C.D. Stiles

381

IGF-Effects on and Binding to Rat Calvaria Cells in Culture C. Schmid, T. Steiner and E.R. Froesch

421

Action of Growth Factors on Chondrocytes: Discovery of Local Somatomedins in Fetal Bovine Cartilage F. Suzuki, Y. Kato, Y. Hiraki, E. Canalis and L. Raisz . 431 Mitogenic Action of Somatomedin Peptides on Human Cartilage and Chondrocytes I.K. Ashton

437

Stimulation of Glycogen Synthesis in Osteoblast-Like Cells by PTH and IGF C. Schmid, T. Steiner and E.R. Froesch

441

Serum Somatomedin Bioactivities: Interrelations between 35s04 -- and 3H-Thymidine Uptakes in Cartilage and 3H-Thymidine Incorporated in Activated Lymphocytes, in Chickens and Humans J. Charrier, M. Bozzola and F. Severi

447

Insulin and Somatomedin C as Growth Promoters of Cells in Serum-Free Medium J.P. Mather and R. Furlanetto

451

XII RECEPTORS FOR INSULIN-LIKE GROWTH FACTORS

Properties of Insulin-Like Growth Factor Receptor Subtypes M.M. Rechler, M. Kasuga, N. Sasaki, M.A. De Vroede, J.A. Romanus and S.P. Nissley

459

Receptors for Insulin-Like Growth Factors: Basic Somatomedin Preceptors in Human and Rodent Tissues R.M. Bala, B. Bhaumick, G.H. Armstrong and M.D. Hollenberg

491

Regulation of Somatomedin-C/Insulin Like Growth Factor-I Receptors R. Rosenfeld and L. Dollar

509

Somatomedin and Insulin Receptors in Rat Chondrocytes K. Asakawa, K. Takano, N. Hizuka, M. Kogawa and K. Shizume

523

IGF-II Receptor Expression in Developing Tissues: Models In vivo and In vitro J.R. Gavin, III and B. Trivedi

531

Regulation of Binding of Insulin and Insulin-Like Growth Factor by Cell Growth Status B. Pfeifle, V. Maier and H. Ditschuneit

539

Somatomedin Receptors in the Human Brain Throughout Life V.R. Sara and K. Hall

545

MOLECULAR BIOLOGY OF INSULIN-LIKE GROWTH FACTORS

Ectopic Growth Factor Production by Tumor Cells and Their Role in the Expression of the Transformed Phenotype J.E. De Larco

551

Immunoperoxidase Localization of Insulin-Like Growth Factor-I Containing Tissues J. Bennington, E.M. Spencer and K. Reber

563

Production of Insulin-Like Growth Factors (IGFs) and Their Binding Proteins (IGF BPs) by the Pituitary Gland and the Nervous Tissue in Culture M. Binoux, P. Hossenlopp, C. Lassarre, A. Barret, A. Faivre-Bauman, C. Loudes and A. Tixier-Vidal

571

XIII M o n o c l o n a l A n t i b o d i e s that Inhibit the Activity of Human Serum D.C. W a t k i n s , M. W a l l i s a n d J. Ivanyi

Sulphation 577

Interactions of U l t r a f i l t r a b l e Factors P r e s e n t in the Human Serum with Somatomedin Like Peptides M . - H . Heulin, M. Artur, F. Sarem, J. Straczek, A. L a s b e n n e s , F. Belleville, M. Pierson, J.-F. Stoltz, P. N a b e t a n d J.-C. J o b

581

Synthesis a n d S e c r e t i o n of I n s u l i n - L i k e G r o w t h F a c t o r a n d of Its B i n d i n g P r o t e i n b y the P e r f u s e d Rat Liver: D e p e n d e n c e o f G r o w t h H o r m o n e Status J.C. Schwander, C. Hauri, J. Zapf a n d E.R. F r o e s c h

585

Influence of N u t r i t i o n o n S o m a t o m e d i n I n s u l i n - L i k e G r o w t h F a c t o r II S y n t h e s i s a n d R e l e a s e f r o m C u l t u r e d Buffalo Rat L i v e r Cells D.S. S c h a l c h a n d P.W. M a y e r

591

B i o s y n t h e s i s of M u l t i p l i c a t i o n S t i m u l a t i n g A c t i v i t y (MSA) in Rat L i v e r Cells: D e m o n s t r a t i o n of P r e - P r o - M S A and Pro-MSA Y . W . - H . Yang, A . M . A c a u a v i v a , C.B. Bruni, J.A. Romanus, S.P. N i s s l e y a n d M.M. "Rechler

603

P u b e r t a l R i s e of I m m u n o r e a c t i v e S o m a t o m e d i n a n d Its E v e n t u a l Source K. Hall, E.M. Ritzen, R.E. J o h n s o n b a u g h a n d M. P a r v i n e n

611

H i g h M o l e c u l a r W e i g h t S o m a t o m e d i n - C / I G F - I from T47D H u m a n M a m m a r y C a r c i n o m a Cells: Immunoreactivity and Bioactivity R.C. Baxter, J.E. M a i t l a n d , R.L. Raison, R.R. R e d d e l a n d R.L. S u t h e r l a n d

615

B i g u a n i d e s Inhibit S o m a t o m e d i n A c t i o n In vitro A . M . Taylor, M.A. K h o k h e r a n d P. D a n d o n a

619

The Insulin (IGF) Gene Family

W.J. Rutter, G.I. Bell and 0. Laub

629

Subject Index

641

Author Index

663

The Three Historical Origins of Insulin-Like Growth Factor Research

THE SOMATOMEDIN HYPOTHESIS: ORIGINS AND RECENT DEVELOPMENTS

William H. Daughaday Washington University School of Medicine St. Louis, Missouri. USA

The roots of the somatomedin hypothesis can be traced to the recognition by workers in Herbert Evans1 laboratory in the 1940's that cartilage is the main target organ for growth hormone stimulated skeletal growth. Most of the early studies were carried out on epiphysial plate cartilage using tedious histologic morphometric techniques. After World War II, there was a boom in interest in radioactive isotopes as markers of metabolic processes and a great improvement in availability. Among the isotopes in the hands of investigators was 35S sulfur. H. Bostrom in Stockholm (1) and D.D. Dzlewietkowski (2) of the Rockefeller Institute were pioneers in the use of sulfate uptake in cartilage as a marker of chondroitin-sulfate synthesis. In this pre-RIA era, I was attracted to the possibility that

35

S-sulfate uptake in

cartilage could be used as an end point for an in vitro bioassay of GH by the observation of Murphy in my laboratory that there was a defect in sulfate uptake in cartilage of hypophysectomized rats which could be corrected with 24 hours by GH administration (3). At this time I was fortunate to have a very bright, hard-working but extremely modest and retiring young research fellow, Dr. William D. Salmon, Jr., join my laboratory. He rapidly established the conditions for in vitro incubaton of cartilage and a simplified method of measurement of 35s sulfate uptake.

He found that GH

given in vivo to hypophysectomized rats restored the in vitro uptake of 35s-sulfate by nasal, xiphoid and costal cartilage to the level of normal cartilage but when he added GH to the incubation medium either alone or with hypophysectomized rat serumto hypox rat cartilage, he observed little or no stimulation of 35s-sulfate uptake (4). This observation greatly disappointed us and aborted our plans to develop an in vitro

Insulin-Like Growth Factors / Somatomedins © 1983 Walter d e Gruyter & Co., Berlin • New York

4

bioassay.

We could not attribute our failure to metabolic deterioration of cartilage

during incubation because cartilage has no intrinsic blood supply and is normally adapted to anoerobic metabolism.

Functional viability of cartilage in tissue culture

media is maintained for long periods. We next considered the possibility that GH could be acting indirectly through some serum component.

Such a mechanism had been proposed earlier by Bornstein and

Park (5) to explain the inhibitor of glucose uptake which appeared in hypophysectomized alloxan diabetic rat serum a f t e r GH treatment.

As I had previously

worked with Rollo Park, I was familiar with this hypothesis. Our initial experiments showed that normal serum stimulated two-fold increase in 35 S -sulfate uptake but serum from hypox rats was virtually inactive.

The e f f e c t was not immediate but

increased progressively over 24 hours of study.

Treatment of hypox rats with GH

restored the sulfate uptake stimulatory activity of serum to nearly normal levels. In these GH treated hypox rats there was a good temporal correlation between increase in 35 S -sulfate uptake of their cartilage with the appearance of the S^s-sulfate uptake stimulatory activity of their serum.

We showed that the stimulatory e f f e c t of GH

was not attributable to insulin or glutamine.

We went on to show that a difference

between hypox and hormonal rat serum persisted a f t e r dialysis. The term "sulfation factor" was proposed for the activity. A quantitative bioassay for s u l f a t i o n tor w a s

fac-

proposed.

The assay for sulfation factor developed by Salmon for rat serum was applicable to human serum (6). Extremely low concentrations of sulfation factor were observed in patients a f t e r total hypophysectomy and in pituitary dwarfs.

Most patients with

acromegaly had elevations of serum sulfation factor. Our observations of sulfation factor attracted little attention except in Stockholm where Professor Rolf Luft recognized their potential importance and interested one of his doctoral candidates, Sven Almqvist, in sulfation factor.

Almqvist made

improvements inthe statistical design of the hypox rat costal cartilage assays and by personally hypophysectomizing his rats and dissecting their costal cartilage he was able to minimize assay variance which has been the bane of this assay. In a series of 7 papers, Almqvist confirmed the basic findings of my laboratory and went on to

5 describe the changes of serum SF with age (7). He was the first to recognize that sulfation factor was low during the first 4 years of life. He also described a fall in sulfation factor concentration of two acromegalic patients treated with estrogens and the kinetics of decline of sulfation factor a f t e r hypophysectomy of acromegalic patients. After this brilliant start in somatomedin research, it was decided by Professor Luft that the Department of Endocrinology and Metabolism at the Karolinska Hospital was more in need of a thyroidologist than a sulfation factorologist and Almqvist was sent to the NIH in Bethesda to become a thyroidologist. The next years brought conclusive evidence that sulfation factor effects were not limited to hormonal sulfation of proteoglycans but included stimulation of collagen synthesis (8), synthesis of non-collagen protein (9), DNA synthesis (10) and RNA synthesis (9). A clinical study of my laboratory in collaboration with Zvi Laron and associates at Petah Tikva Israel in patients with dwarfism and high serum growth hormone, commonly referred to as Laron dwarfism, attracted general interest in sulfation factor (11). We found that sulfation factor levels were as low in these patients as were found in patients with hypopituitarism but

treatment with human growth

hormone failed to restore normal serum factor activity . This dissociation of growth hormone levels and sulfation factor levels supported the essentiality of sulfation factor in human growth but other interpretations are possible. The hiatus of interest in sulfation factor in Stockholm ended when Kerstin Hall (12) began her doctoral studies in the late 1960's in Professor Luft's Department.

She

developed an embryonic chick cartilage bioassay which had virtues of economy and convenience at the price of some loss of specificity. With Judson Van Wyk, who spent a sabbatical year in Stockholm, and others, she undertook a full scale attempt at isolation of the sulfation factor. An initial acid ethanol extraction was utilized to free the active peptide from binding proteins.

Subsequent purification steps were

monitored by measuring sulfate uptake, thymidine uptake in cartilage and insulin-like e f f e c t s on epididymal f a t . It was observed that throughout the various purification

6 steps sulfation factor, thymidine factor and insulin-like activity all co-purified and it was suggested that they were properties of the same molecules. It is notable that isoelectric separation provided clear evidence of separate basic and neutral peptides with growth factor activity. Subsequent purification of the neutral peptide was pursued in Stockholm by Kerstin Hall and the Kabi group and the basic peptide in Chapel Hill by Van Wyk and associates. The Zurich group with Rene Humbel was following a parallel duality in their isolation and characterization of IGF-I and IGF-II. At this stage it was recognized that the operational name of sulfation factor was too restrictive for the multiple actions of the GH dependent tissue growth factors. The term somatomedin was arrived at by consensus of most of the investigators of the time (13). This Greco-Roman hybrid has been useful as a generic term for hormonal peptides mediating GH action. The neutral peptide under study in Stockholm was designated somatomedin A, an acidic peptide subsequently shown to contain EGF as a contaminant was called somatomed B and the basic peptide studied in Chapel Hill was named somatomedin C. The changing of names from sulfation factor to somatomedin can also be looked upon as the coming of age of the somatomedin hypothesis. It marked the time when many new investigators entered the field and progress became more rapid with further characterization of the chemical nature of these substances culminating in the accomplishments of Rinderknecht and Humbel in establishing the sequences of IGF-I and II (14, 15). Our knowledge of the serum binding proteins for somatomedin increased. New radioreceptor and radioimmunoassays for somatomedins largely replaced bioassays. Separate receptors for IGF-I/Sm C and IGF-II were recognized and characterized. Seious study was undertaken of inhibitors of somatomedin actions. The hypothesis that the somatomedins are important regulators of skeletal growth has not gone unchallenged.

The in vitro biological activity of somatomedin

complexed to its binding protein has been questioned.

It has been impossible to

demonstrate unequivocally the presence of unbound somatomedins in serum.

All

detectable somatomedin exists in specific binding protein complexes. In a number of test systems such as the isolated rat heart (16), rat adipocytes (17) and chick embryo

7

fibroblast (18) bound somatomedin is virtually inactive. We have observed that less than one-third of the somatomedin in whole serum has access to the somatomedin receptor on human placental membranes (19). In addition, the large complexes of protein bound somatomedins must be greatly hindered in crossing the capillary epithelium.

Despite these limitations of action of protein bound somatomedins on

certain tissues, somatomedin, in whole serum can effectively stimulate cartilage metabolism in vitro even at high dilutions. With our current in vitro conditions, 100200% stimulation of sulfate incorporation by hypophysectomized rat cartilage is achieved with only 2% rat serum in the incubation medium. It is unknown whether the extremely small concentration of unbound somatomedin which escapes detection could be responsible for receptor activation or whether one or another species of bound somatomedin can activate the receptor directly. properties of the serum are altered at high dilution.

We find that the binding This could act to increase

dissociation of bound somatomedin in interstitial tissues. Whatever the explanation, somatomedin can reach cartilage in sufficient concentration and availability to exert its stimulatory effects. While I do not wish to reject the insulin-like effects of somatomedins on non skeletal tissues, cartilage is the only mammalian tissue which has been studied in vitro which so specifically responds to somatomedin containing serum as compared to somatomedin poor serum. A second major objection to the somatomedin hypothesis has been the lack of confirmation of the growth promoting actions of somatomedins in vivo.

Until

recently, investigators have been handicapped by limitations in availability of highly purified somatomedin peptides. Relatively large amounts of the purified somatomedins must be given to restore and maintain the serum concentration of somatomedin of hypophysectomized animals to normal. This is a consequence of the high serum concentration of the somatomedin peptides as compared to other hormones and the rapid clearance of administered somatomedin when not bound in the normal binding protein complex. In contrast to the need for relatively large amounts of somatomedin for replacement treatment of hypophysectomized animals, smaller amounts of GH are required. GH acts on liver and perhaps other tissues to produce somatomedins. It is likely that a single molecule of GH can promote the secretion of multiple somatomedin molecules.

8 The first positive demonstration that somatomedins can stimulate growth in vivo was provided by Van Buul-Offers et al. (20) who injected

partially

purified human

somatomedin into immature, Snell dwarf mice. Growth in length and weight occurred and sulfate and thymidine uptake in isolated cartilage was stimulated. This study was not conclusive because the preparation administered, although devoid of significant GH or insulin contamination, was admittedly crude.

This criticism cannot be applied to the important observations of Schoenle et al. (21) who obtained unequivocal stimulation of growth of hypophysectomized rats with IGFI infused continuously by implanted osmotic minipumps. similar fashion was much less e f f e c t i v e (22).

IGF-II administered in a

Dr. Zapf will describe these studies in

greater detail in his text.

The demonstration that IGF-I is more potent than IGF-II in stimulating growth in vivo and cartilage metabolism in vitro and the recognition that growth hormone dependence of IGF-I is much greater than IGF-II all lead to the conclusion that it is the major somatomedin of serum.

IGF-II, possessing a separate dedicated

receptor,

probably will be shown to have different physiologic roles.

Isaksson et al. (23) have challenged the somatomedin hypothesis by demonstrating that

GH

can

stimulate

longitudinal

bone growth

directly.

These

investigators

injected the epiphysial growth plates of hypophysectomized rats with 10 ug of GH on three occasions over a f i v e day period.

Appositional bone growth measured by a

tetracycline labeling technique, demonstrated a 44% increase on the injected side as compared to the uninjected side. hypothesis.

A t f a c e value this contradicts the somatomedin

There are certain aspects of the experiment that need to be considered.

The injection of 10 y g of hormone into an avascular tissue undoubtedly unphysiologically high concentrations of hormone.

created

The response was relatively small

compared to a 227% stimulation of growth induced by 5 pg/day of GH subcutaneously in a similar experimental system by Thorngren and Hansson (24). associates

certainly

have

not

shown

that

exposure

of

the

Isaksson and

epiphyseal

plate

to

physiologic concentrations of GH can restore normal appositional bone growth in the absence of somatomedins.

9 Another challenge to the hormonal role of somatomedin exists. It has been observed by Atkison et al. (25) and Clemmons et al. (26) that certain fibroblasts release RIA detectable Sm C/IGF-I-like peptides and that this release is stimulated by GH. These same cells are capable of being stimulated by Sm C/IGF-I.

If this type of local

production of somatomedins and their paracrine action is important in vivo, the somatomedins might not be true hormonal agents.

These experiments would not

explain the lack of e f f e c t of GH on isolated cartilage and the exquisite sensitivity of this tissue to somatomedin. In conclusion I have reviewed the genesis of the somatomedin hypothesis and some of the early evidence on which it was founded.

The years have brought additional

clinical and experimental evidence in its support. The recent demonstration of the in vivo growth promoting potency of IGF-I has provided a long awaited and welcome addition to the evidence supporting the hypothesis.

As of 1982, the role of

somatomedins in mediating some or all of GH action on skeletal tissue remains an attractive and viable hypothesis.

References 1.

Bostrom, H.: On the metabolism of the sulfate group of chondroitinsulfuric acid. J . Biol. Chem. 196, 477 (1952).

2.

Dziewiatkowski, D.D.: Effect of age on some aspects of sulfate metabolism in the r a t . J . Exper. Med. 99, 273 (1954).

3.

Murphy, W.R., Daughaday, W.H., Hartnett, C.: The e f f e c t of hypophysectomy and growth hormone on the incorporation of labeled sulfate into tibial epiphyseal and nasal cartialage of the rat. J . Lab. Clin. Med. £7, 715-722 (1956).

4.

Salmon, W.D., Jr., Daughaday, W.H.: A hormonally controlled serum factor which stimulates sulfate incorporation by cartilage in vitro. J . Lab. Clin. Med. 49, 825-836 (1957).

5.

Bornstein, J., Park, C.R.: Inhibition of glucose uptake by the serum of diabetic rats. J . Biol. Chem. 205, 503 (1953).

6.

Daughaday, W.H., Salmon, W.D., Jr., Alexander, F.: Sulfation factor activity of sera from patients with pituitary disorders. J . Clin. Endocrinol. Metab. 19, 743758 (1959).

10 7.

Almqvist, S.: Studies on sulfation factor activity of human serum. Doctoral Thesis, Department of Endocrinology and Metabolism, Karolinska Sjukhuset, Zeteerlund and Thelanders Boktryckeri AB, Stockholm (1961).

8.

Daughaday, W.H., Mariz, I.K.: Conversion of proline-U-C 14 to labeled hydroxyproline by rat cartilage in vitro: Effects of hypophysectomy, growth hormone, and Cortisol. J . Lab. Clin. Med. 59, 741-752 (1962).

9.

Salmon, W.D., Jr., DuVall, M.R.: A serum fraction with "sulfation factor activity" which stimulates in vitro incorporations of leucine and sulfate into protein-polysaccharide complexes, uridine into RNA and thymidine into DNA of costal cartilage from hypophysectomized rats. Endocrinology 86, 721-727 (1970).

10.

Daughaday, W.H., Reeder, C.: Synchronous activation of DNA synthesis in hypophysectomized rat cartilage by growth hormone. J . Lab. Clin. Med. 68, 357-368 (1966).

11.

Daughaday, W.H., Laron, Z., Pertzeland, A., Heins, J.N.: Effective sulfation factor generation: A possible etiological link in dwarfism. Trans. Assoc. Am. Phys. 82, 129-138 (1969).

12.

Hall, K.: Human somatomedin, determination occurrence, biological activity and purification. Acta Endocrinol. Suppl. 163, (1972).

13.

Daughaday, W.H., Hall, K., Raben, M.S., Salmon, W.D., Jr., Vanden Brande, J.L., Van Wyk, J.J.: Somatomedin: A proposed designation for the "sulfation factor". Nature 235, 107 (1972).

14.

Rinderknecht, E., Humbel, R.E.: The amino acid sequence of human insulin-like growth factor I and its structural homology with proinsulin. J . Biol. Chem. 253, 2769-2776 (1978).

15.

Rinderknecht, E., Humbel, R.E.: Primary structure of human insulin-like growth factor II. FEBS Lett. 89, 283-286 (1978).

16.

Meuli, C., Zapf, J., Froesch, E.R.: NSILA-carrier protein abolishes the action of non-suppressible insulin-like activity (NSILA-S) on perfused rat heart. Diabetologia 14, 255-259 (1978).

17.

Zapf, J., Schoenle, E., Jagars, G., Grunwald, J., Froesch, E.R.: Inhibition of the action of nonsuppressible insulin-like activity on isolated rat f a t cells by binding to its carrier protein. J . Clin. Invest. 63, 1077-1084 (1979).

18.

Knauer, D.J., Smith, G.L.: Inhibition of biological activity of multiplicationstimulating activity by binding to its carrier protein. Proc. Natl. Acad. Sci. USA 77, 7252-7256 (1980).

19.

Daughaday, W.H., Mariz, I.K., Blethen, S.L.: Inhibition of access of bound somatomedin to membrane receptor and immunobinding sites - a comparison of radioreceptor and radioimmunoassay of somatomedin in native and acidethanol-extracted serum. J . Clin. Endocrinol. Metab. 51, 781-788 (1980).

11

20.

Van Buul-Offers, S., Van den Brande, J.L.: Effect of growth hormone and peptide fractions containing somatomedin activity on growth and cartilage metabolism. Acta Endocrinol. 92, 242-257 (1979).

21.

Schoenle, E., Zapf, J., Humbel, R.E., Froesch, E.R. Insulin-like growth factor I stimulates growth in hypophysectomized rats. Nature 296, 252 (1982).

22.

Schoenle, E., Zapf, J., Froesch, E.R.: Insulin-like growth factors I and II stimulate growth of hypophysectomized rats. Program and Abstracts, 64th Annual Meeting, The Endocrine Society, San Francisco, CA. June 1982.

23.

Isaksson, O.G.P., Jansson, J-O., Gause, I.A.M.: Growth hormone stimulates longitudinal bone growth directly. Science 216, 1237-1238 (1982).

24.

Thorngren, K.-G., Hansson, L.I.: Bioassay of growth hormone. II. Determination of longitudinal bone growth with tetracycline in thyroxine-treated hypophysectomized rats. Acta Endocrinol. 75:669, (1974).

25.

Atkison, P.R., Weidman, E.R., Bhaumick, B., Bala, R.M.: Release of somatomedin-like activity by cultured WI-38 human fibroblasts. Endocrinology 106, 20062012 (1980).

26.

Clemmons, D.R., Underwood, L.E., Van Wyk, J.J.: Hormonal control of immunoreactive somatomedin production by cultured human fibroblasts. J . Clin. Invest. 67, 10-19 (1981).

F R O M N S I L A TO IGF: A L O O K B A C K O N T H E M A J O R A D V A N C E S

AND

BREAKTHROUGHS

E.R.

Froesch

M e t a b o l i c U n i t , D e p a r t m e n t of M e d i c i n e , U n i v e r s i t y of

Zurich

The d i s c o v e r y of i n s u l i n - l i k e a c t i v i t y of s e r u m and of nonsuppressible

insulin-like

The d i s c o v e r y of i n s u l i n - l i k e

activity

activity

(ILA) of s e r u m d a t e s

back b e f o r e the time w h e n r a d i o i m m u n o a s s a y s

for the m e a s u r e -

m e n t of i n s u l i n in s e r u m b e c a m e a v a i l a b l e . T h e m a i n

observa-

t i o n s w e r e the f o l l o w i n g : W h e n the d i a p h r a g m or a d i p o s e of the rat are i n c u b a t e d tissues

is s t i m u l a t e d

approximately

200

in s e r u m , g l u c o s e u p t a k e of

as if they w e r e of i n s u l i n per ml

tissue

these

in the p r e s e n c e of (1,2). T h e s e

findings

w e r e f o l l o w e d up by i n c u b a t i o n e x p e r i m e n t s w i t h a d i p o s e

tissue

and s e r u m in the p r e s e n c e of a n t i - i n s u l i n s e r u m f r o m g u i n e a pigs which

i n h i b i t s the a c t i o n of i n s u l i n . It w a s found

that

90 % of the i n s u l i n - l i k e e f f e c t of s e r u m o n a d i p o s e t i s s u e not s u p p r e s s e d by a n t i - i n s u l i n s e r u m and it w a s r e a s o n e d the

insulin-like

substance

cally identical with

in s e r u m could not be

insulin

activity

immunologi-

(3). Rat a d i p o s e t i s s u e w a s

m o s t l y used for the d e t e c t i o n and m e a s u r e m e n t of ible i n s u l i n - l i k e

was

that

nonsuppress-

(NSILA) of s e r u m and t h i s

p r o v e d to be v e r y r e p r o d u c i b l e .

bioassay

It s e r v e d as the m a j o r bio -

a s s a y for N S I L A until t h e s e s u b s t a n c e s w e r e p u r i f i e d c h e m i c a l l y c h a r a c t e r i z e d . T h e a c t i v i t y of N S I L A of b e f o r e and after e x t r a c t i o n w a s a l w a y s e x p r e s s e d

serum

in terms of

m i c r o u n i t s of i n s u l i n w h i c h s e r v e d as a w e l l d e f i n e d hormone.

Insulin-Like Growth Factors / Somatomedins © 1983 Walter de Gruyter& Co., Berlin • New York

and

standard

14 Extraction procedures

for N S I L A - s

In a n a l o g y to the e x t r c a t i o n of i n s u l i n f r o m p a n c r e a s ,

acid/

e t h a n o l w a s o r i g i n a l l y used to e x t r a c t N S I L A f r o m s e r u m . A small p o r t i o n of total N S I L A of s e r u m w a s found to be in a c i d / e t h a n o l .

soluble

It w a s c a l l e d N S I L A - s , the £ s t a n d i n g

for

s o l u b l e . T h e m o l e c u l a r w e i g h t of N S I L A - s w a s e s t i m a t e d to be a r o u n d 7'500 w h i c h w a s in s h a r p c o n t r a s t to the m o l e c u l a r w e i g h t of N S I L A

in n a t i v e s e r u m w h i c h lies in the o r d e r

m a g n i t u d e of 200 '000 daltons (4) .

During acid/ethanol

of s e r u m m o s t of the N S I L A r e m a i n s

in the p r e c i p i t a t e .

This

f r a c t i o n of s e r u m N S I L A w a s c a l l e d N S I L A - P , P s t a n d i n g precipitated

(5). T h e s e f i n d i n g s w e r e e r r o n e o u s l y

of

extraction for

interpreted

to m e a n that the a c i d / e t h a n o l e x t r a c t i o n led to a d e n a t u r a t i o n and p r e c i p i t a t i o n of the m a j o r p a r t of s e r u m N S I L A . was biologically

characterized

and it w a s f o u n d to be

o n b o t h a d i p o s e t i s s u e and d i a p h r a g m in v i t r o and intraperitoneal

i n j e c t i o n into the rat

ments with intravenous

NSILA-P active

after

(6). In v i v o

experi-

i n j e c t i o n s of impure p r e p a r a t i o n s of

N S I L A - P e n d e d w i t h the r a p i d d e a t h of the r a t s

(unpublished

o b e r s v a t i o n ) . N S I L A of a h i g h m o l . wt. w a s also p r e p a r e d

by

Dowex-chromatography

steps

and s e v e r a l s u b s e q u e n t p u r i f i c a t i o n

by P o f f e n b a r g e r w h o c a l l e d the s u b s t a n c e N S I L P , P s t a n d i n g protein

(7). A n t i b o d i e s

and a r a d i o i m m u n o a s s a y to some r e p o r t s

against NSILP were produced for N S I L P w a s d e v e l o p e d

in r a b b i t s

(8). A c c o r d i n g

in the l i t e r a t u r e N S I L P m a y be e l e v a t e d

p a t i e n t s w i t h tumor h y p o g l y c e m i a . H o w e v e r , N S I L P l e v e l s also found to be e l e v a t e d

in

for

in were

patients with tumors which did

not g o along w i t h h y p o g l y c e m i a . T h u s , the

physiological

s i g n i f i c a n c e of N S I L P s t i l l is in the d a r k . It is also le to s e p a r a t e

large m o l . wt. N S I L A f r o m N S I L A - s by

chromatography

(9). T h e s e large m o l e c u l a r

possib-

Sephadex

f o r m s of N S I L A ,

i.e.

N S I L A - P , N S I L P and large m o l . wt. N S I L A c a n n o t be c o n v e r t e d N S I L A - s and p r o b a b l y are not r e l a t e d to N S I L A - s . T h e y

have

to

15

not b e e n f u r t h e r c h a r a c t e r i z e d , n e i t h e r c h e m i c a l l y ,

nor

b i o l o g i c a l l y , nor p h y s i o l o g i c a l l y . T h e s e forms of N S I L A

are

not to be c o n f o u n d e d w i t h N S I L A - s or IGF I and I G F II b o u n d t h e i r c a r r i e r p r o t e i n s b e c a u s e acid t r e a t m e n t of any k i n d not d i s s o c i a t e wt.

IGF I or I G F II from t h e s e forms of large

to

does mol.

NSILA.

For the a n a l y t i c a l m e a s u r e m e n t of N S I L A - s

in i n d i v i d u a l

sera

the d i s s o c i a t i o n of N S I L A - s from their b i n d i n g p r o t e i n s m a n d a t o r y . T h i s w a s r e a l i z e d by acid c h r o m a t o g r a p h y S e p h a d e x by S c h l u m p f et al. in o u r l a b o r a t o r y one-step procedure NSILA-s

is

over

(10). By

this

is r e p r o d u c i b l y d i s s o c i a t e d

from

the c a r r i e r p r o t e i n and can t h e n be d e t e r m i n e d a s s a y s y s t e m s w h i c h are now a v a i l a b l e

in any of

(bioassays using

c e l l s or fat p a d s , c h i c k e m b r y o or o t h e r f i b r o b l a s t s ,

fat sul-

f a t i o n of c a r t i l a g e of v a r i o u s a n i m a l s , p r o t e i n b i n d i n g using the c a r r i e r p r o t e i n of h u m a n or o t h e r s e r a , a s s a y for IGF I and IGF II),

the

assay

radioimmuno-

(for d e t a i l s see Zapf,

this

issue).

L a r g e s c a l e p r o d u c t i o n of N S I L A - s Many tissues were extracted with acid/ethanol that

in the

hope

1) l a r g e a m o u n t s of N S I L A - s m i g h t be o b t a i n e d and 2)

t h a t the p r o b l e m of the o r i g i n of this h o r m o n e m i g h t

be

r e s o l v e d . H o w e v e r , we and o t h e r s found t h a t m o r e N S I L A - s present

in s e r u m p e r mg of p r o t e i n t h a n in any o t h e r

T h e r e f o r e , the large scale p r o d u c t i o n of N S I L A - s had

is

tissue. to s t a r t

f r o m s e r u m as raw m a t e r i a l . P r e c i p i t a t e B w h i c h is s i m i l a r C o h n f r a c t i o n IV and w h i c h

is a b y - p r o d u c t of the

preparation

of h u m a n a l b u m i n and h u m a n g a m m a g l o b u l i n s and w h i c h be used for any b e t t e r p u r p o s e w a s found to c o n t a i n a m o u n t s of N S I L A - s b o u n d to its b i n d i n g p r o t e i n s

to

cannot large

(5). A m e t h o d

w a s d e v i s e d in our l a b o r a t o r y to e x t r a c t and p u r i f y N S I L A - s

in

16

small amounts

from serum

and

p r o c e d u r e w a s then addopted Roche who extracted and

acid

ethanol

Department NSILA-s

using

procedure tides,

by Dr. Richard

6 tons of Cohn

and sent

and

a total of about finally managed IGF

at

fraction

IV

IV with

acetone

to H u m b e l

in the (11).

this crude p r e p a r a t i o n

6 steps

for their

to identify

. This

Hoffmann-La

of the U n i v e r s i t y of Zurich

and H u m b e l purified

IGF I and

fraction

the a c e t o n e powder

of B i o c h e m i s t r y

Rinderknecht

from this Cohn

of

purification

two pure

polypep-

II.

P U R I F I C A T I O N S C H E M E L E A D I N G TO THE ISOLATION OF I G F I A N D I I

Specific biological activity: mU/mg protein (fat pad assay)

Purification step N a t i v e serum Precipitate B Acid ethanol extract (acetone powder) Acetic acid 800)

Rat V 2 5 5 1(764)

(747)f

|(498)

t(466)

1(718)

ILAs icv

4(570)

(CMC) Rat V 2 7 4

1200

1300 Time (Hours)

1400

1200

1300 Time (Hours)

Figure 1. Effect of intracerebroventricular ( i c v ) administration of either normal s a l i n e (A, C) or ILAs (CMC preparation, B; Sephadex preparation, D) on i n d i v i d u a l , representative six-hour GH secretory p r o f i l e s i n 2 r a t s . Central administration of both ILAs preparations caused a dramatic suppression in amplitude of GH secretory bursts after an interval of approximately 2 h and plasma GH l e v e l s remained markedly suppressed for up to 6 h after i n j e c t i o n . Curved arrows indicate time of i n j e c t i o n .

66 which was in striking contrast to normal saline-treated control

animals

whose peak GH values ranged from 230-764 ng/ml during this time.

Analysis

of the time course of effect of ILAs revealed no significant difference in mean plasma GH levels between ILAs- and normal saline-treated groups during the first 2 h after injection.

However mean plasma GH levels were sig-

nificantly depressed during the remaining 4-h sampling period (Fig. 2). The finding that the ability to suppress GH release was very similar for the CMC-ILAs and Sephadex-ILAs despite an approximately 3-fold greater purity of the former suggests that the biological ILAs.

activity inheres in the

Specificity of the GH response to ILAs is indicated by the findings

that neither BSA, a protein control, nor porcine insulin, another growth factor not directly stimulated by GH, significantly altered plasma GH levels (Fig. 2).

In a second study, four additional groups of rats (350-425 g) were used to assess the role of ILAs in nutritional

regulation.

Food intake, measured

in terms of 24-h intake of pelleted Purina rat chow, and body weight change 150,-

Normal

Saline

ILAs

(Sephadex)

ILAs

(CMC)

BSA

Insulin

Figure 2. Effect of icv administration of ILAs and control materials on mean plasma GH levels 2-6 h post injection. Each bar represents the mean + SEM, and the number of animals in each group is shown in parentheses. **Significantly different from all other groups, P

QC

•


2, 5% CC>2 atmosphere for 6 hours. The in vitro response of the3 tissue to hormonal treatment was expressed as the counts of

H-TdR incorporated

into TCA-insoluble material per mg dry weight of tissue.

Fig-

ure 3, panel A, shows the results of an experiment with cropsac tissue taken from untreated birds. without effect.

PRL and PI were both

However, if the experiment was done with

74

Figure 3. Incorporation of tritiated thymidine into crop-sac tissue in 6 hr. incubation in vitro. Tissue was taken from birds and cut into 1cm squares, which were then incubated in 5 ml of Waymouth's medium containing lyCi of 3 H-TdR, with or without addition of lpg/ml of PRL or PI. The tissue was then lyophilized, weighed, and the amount of 3h incorporated into TCA insoluble material determined. In Expt. A, the tissue was from birds given no prior treatment. In panel B, birds were pretreated with lOmg/day of a sheep pituitary powder, suspended in 0.9% NaCl and injected into the loose skin between a leg and the body cavity.

crop-sac tissue from birds injected with a suspension of a sheep pituitary powder prior to incubation, PI stimulated TdR incorporation (panel B).

3

H-

PRL was again without effect.

These results suggest that the PRL in the sheep pituitary powder sensitized the crop-sac epithelium to the actions of synlactin, and that PI is acting as a synlactin agonist. In a separate series of experiments, we studied the possibility that various pituitary hormones might increase the sensitivity of the crop-sac to the direct mitogenic effects of PRL. The experiment was prompted by the observations that PRL-induced

proliferation of the crop-sac is reduced in hypophys-

ectomized (HX) birds (6), and that systemic injections of pituitary extracts are more potent stimulators of cell prolifer-

75

SYSTEMIC

TREATMENT

Figure 4. Effects of systemic injections of various hormones on the response of the pigeon crop-sac to direct application of PRL. On the first day of the experiment, systemic injections of hormones or of the pituitary powder were made at the following doses: Pituitary powder, lOmg; TSH, 50 yg; GH, lmg; PRL, 0.4mg; PI, lmg/kg body weight. On the following 3 days, similar injections were made at doses one-tenth of those used on day 1. On days 2-4, birds were treated with 2.5yg of PRL dissolved in 0.25ml of saline over one hemicrop, while the other side was given control injections of an equal volume of saline. The cross-hatched portion of each bar represents the mucosal dry weight of the saline-injected hemicrop, while the total height of the bar represents the response of the contralateral PRL-treated hemicrop. The difference between the two responses (the stippled portion of each bar) indicates the effect of the systemic treatment on the responsiveness of the crop-sac to the direct action of PRL.

proliferation in the crop-sac than are purified PRL preparations (7).

Furthermore, PRL, acting systemically, causes much

steeper dose-response slopes than it does when injected directly over the crop-sac (5,8).

These observations suggest that a

pituitary factor(s) has indirect effects on crop-sac proliferation.

We found that systemic injections of a sheep pituitary

powder into pigeons increased the responsiveness of the

76

crop-sac to direct application of PRL.

Systemic injections of

ACTH or TSH, or a combination of LH and FSH had no effect on the direct local response to PRL.

However, PRL injected sys-

temically caused a dramatic augmentation of the local response to PRL (Fig. 4).

GH acted as a mimic of PRL in this regard.

These results suggest that PRL has at least two modes of action as a mitogen on this epithelium—a direct effect, as well as an indirect one, possibly mediated by increased secretion of synlactin into the bloodstream.

Systemic injections of PI had

effects similar to those of the pituitary powder and of PRL or GH, lending further credence to the suggestion that the mechanism of the enhanced responsiveness involves increased secretion of synlactin.

Discussion In these studies, we first demonstrated that SM-C, relaxin, insulin and PI all acted synergistically with PRL to promote proliferation of the pigeon crop-sac epithelium, while they had little or no growth-promoting activity alone.

It is not sur-

prising that these 4 molecules should act as mutual agonists, because they have similar conformations (9).

Their potency or-

der suggests that their action is not a function of their activity as insulin analogs. Secondly, treatment of birds with a sheep pituitary suspension sensitized their crop-sacs to the mitogenic actions of PI upon subsequent incubation in vitro. Finally, systemic treatment of birds with sheep pituitary powder led to an increased responsiveness of the crop-sac to diect application of PRL.

This effect could be mimiced by PRL

or GH, but not by other pituitary hormones.

Furthermore, the

observation that systemic treatment with PI was just as effective in this regard as other systemic treatments is consistent with the notion that the effect of the pituitary agent is mediated by a Pi-like substance.

We are suggesting, then, that

the mitogenic actions of PRL on the pigeon crop-sac mucosal

77

epithelium involves sensitization of target cells to the actions of a SM-like molecule, and that PRL also causes increased secretion of this molecule into the bloodstream.

We

have tentatively named this synergist "synlactin". Our synlactin hypothesis may apply to PRL's action on the mammary gland.

Although PRL induces mammary gland proliferation

in vivo, it generally does not have mitogenic activity in vitro on mammary epithelial cells (10).

However, PRL injec-

tions into virgin female mice sensitized their mammary glands to the in vitro actions of insulin or to insulin-free serum (11).

In other studies, it has been shown that relaxin pro-

motes the growth of mammary glands in rodents (12) .

Relaxin

had no such effect in hypophysectomized and ovariectomized rats rats, but was active in these animals if co-injected with PRL (13).

These observations are consistent with the suggestion

that one of PRL's actions is to sensitize its target cells to the mitogenic actions of an insulin-like molecule (synlactin?). Our hypothesis that hormone-induced cell proliferation involves altered target cell sensitivity to SM-like molecules may have utility in understanding the mechanism of action of GH.

It has

recently been shown that SM-C injections into hypophysectomized rats stimulated their growth as measured by tibial width in3 crease, H-TdR incorporation into costal cartilage, or body weight increase (14).

However, their data show that the dose-

related effects of GH on these parameters was greater than was the slope in response to SM, particularly for body weight gain. Thus, SM is not sufficient to completely duplicate the growthpromoting effect of GH in vivo.

It has also been reported that

injections of GH directly into the tibial epiphysis of hypophysectomized rats caused cartilage growth (15).

Taken together,

these data suggest that the mechanism of the growth-promoting action of GH in vivo may be analogous to that of PRL on the crop-sac.

In both systems, the pituitary hormone may have a

direct action on the target cells to induce sensitivity to an

78 i n s u l i n - l i k e molecule, increase

secretion

In c o n c l u s i o n ,

of

as well that

in this

from o t h e r

paper,

actions

their

blood-borne messengers.

data are c o n s i s t e n t with the

target

named

as

hor-

their cells

idea t h a t PRL's mitogenic

mitoto

Specifically,

i s m e d i a t e d i n p a r t y by a S M - l i k e m o l e c u l e , tatively

as well

that pituitary

i n v i v o may e x e r t

i n p a r t by s e n s i t i z i n g

of other

to

laboratories,

indicate

mones w h i c h a r e g r o w t h - p r o m o t i n g genic actions

action

molecule.

the r e s u l t s

those presented

as a systemic

the

our action

w h i c h we h a v e

ten-

"synlactin".

References 1.

Salmon, W.D., Daughaday, W.H.:

J . Lab. C l i n . Med. 49, 824-836

2.

Clemmons, D.R., Van Wyk, J . J . : 161-208 ( 1 9 8 1 ) .

Handbook of E x p t l . Pharmacol. 57,

3.

Daughaday, W.H.: In Endocrine Control of Growth (ed. W.H.Daughaday), E l s e v i e r P r e s s , New York, 1981.

4.

Rothstein, J . :

5.

Nicoll, C.S.:

6.

Schooley, J . P . , Riddle, P . , Bates, R.W.: (1941).

7.

Raud, J . R . , Odell, W.D.:

8.

Nicoll, C.S.:

9.

Blundell, T . L . , Bedarkar, S . , Rinderknecht, E. , Humbel, R . E . : Natl. Acad. S c i . USA 75, 180-184 (1978)

I n t . Rev. Cytol. 78, 127-232 Endocrinology 80, 641-655

(1957).

(1982).

(1967). Am. J . Anat. 69, 123-154

Endocrinology 88, 991-1002

Acta Endocrinol. 60, 91-100

(1971).

(1969).

10.

Topper, Y . J . ,

Ereeman, C . S . : P h y s i o l . Rev. 60, 1049-1106

11.

Oka, T . , Topper, Y . J . : (1972).

12.

Harness, J . R . , Anderson, R.R. Proc. Soc. Exp. B i o l . Med. 148, 933-936 ( 1 9 7 5 ) .

13.

Harness, J . R . , Anderson, R . R . : 354-358 ( 1 9 7 7 ) .

14.

Schoenle, E . , Zapf, J . , Humbel, R . E . , 252-253 ( 1 9 8 2 ) .

15.

Isaksson, O.G.P., Jansson, J . - O . , Gause, I.A.M.: 1237-1239 ( 1 9 8 2 ) .

Proc. Natl. Acad. S c i .

Proc.

(1980).

USA 69, 1693-1696

Proc. Soc. Exp. B i o l . Med. 156, Hroesch, E . R . :

Nature 296,

Science 216,

Structure and Purification of Insulin-Like Growth Factors

THE WITH

IDENTITY

OF HUMAN

SOMATOMEDINS

INSULIN-LIKE

C AND

GROWTH

A AND H O M O L O G Y

FACTORS

WITH RAT

I AND

IGF

II

I AND

II.

E. M A R T I N

SPENCER,

CHILDREN'S

HOSPITAL

4 DEVELOPMENT,

Because

the

be

SAN

indications

that

IGF-II and

been

established.

definitive

Rinderknecht and

IGF-II

fraction

IV

bioassays,

have

SM-A,

and Humbel acid

(1, 2 ) .

ethanol

pad and m i t o g e n i c i t y

Cohn fraction

incorporation

into

the

(the

and

Insulin-Like Growth Factors / Somatomedins © 1983 Walter d e Gruyter & Co., Berlin • N e w York

are

strong possibly not

IGF-I

from

Cohn by

two

epididymal

fat

fibroblasts. acid

ethanol

different

stimulation

later

have

sequenced

purification

but

they

by

in the rat

IV,

the

properties,

of h u m a n

isolated

and assays

cartilage

GROWTH

been done

chicken embryo

extracts

schemes

and

extracts

originally

and

relationships

purified

The SMs also w e r e

purification

OF

same and

IGFs has

activity

toward

there

the

these

They m o n i t o r e d

insulin-like

of h u m a n

are

on the

(IGFs)

biologic

Although

far

who

LABORATORY

factors

identical

thus

SMITH.

CALIFORNIA.

growth

identical.

chemistry

from

AND B E T H

IGF-I and SM-C

also

The

FRANCISCO,

(SMs)

chemically

ROSS

OF SAN F R A N C I S C O ,

insulin-like

somatomedins may

MAUREEN

the

human

of

sulfate

placental

82 membrane neither

radioreceptorassay) of t h e

different

have

SM's,

procedures

purification, SMs

two

the

A and

suspected

on chemically

sufficiently

to c o m p a r e

been

the use

on their the

pi

peptide

peptide

found

a single

to be

closely

studies,

chemical

similar

identity

suggested greater

related

that

extension.

to

partial has

not

of

C-terminal

to p r o v e

of S M s

sequence

that does

(8,300

not -

SM-C has

this

for

peak see

been

shown

antisera However,

Svoboda

of S M - C w a s

et

al.

slightly

daltons),

carboxy-terminal point.

the

binding

(3).

vs. 7,300

for

confusion.

histidine-containing the

each

different

data

based

decision

competitive four

we

has

A,

C, f o r

been established.

IGF-I

Unfortunately,

determined

with

and

SMs

considerable

purified.

I G F - I by

identity

the

been named

A peak

the m o l e c u l a r w e i g h t

than that

to a p u t a t i v e

highly

IGFs

By c o n v e n t i o n ,

not g u a r a n t e e

to

the

in this work

pH r e g i o n a n d

and the

and

their

This operational

has led

the m o s t

antigenic

(4, 5) a n d

not

specie,

- a point which

has been

does

in

purification

SMs have

in a neutral

sequenced

IGFs.

focusing.

in the b a s i c r e g i o n .

contains

SM-C

and

the

because

between

Therefore,

them w i t h

point),

however,

used

characterizing

isoelectric

(isoionic

Thus, been

identities

nomenclature

of

nomenclature,

below

C, h a v e

not b e e n e s t a b l i s h e d .

to t h e

used.

and assays were

concentrated

Central

were

due

peptide sequence

was

83 SM-A, which comprises the neutral SM-containing region on isoelectric focusing, has not been adequately characterized.

However, it can be shown that this peak

also contains variable amounts of IGF-I by two different antisera (5 and see below). zonal electrophoresis continued

Fryklund et al.

substituted

for isoelectric focusing and

to call the material isolated A in spite of the

fact that isoionic points cannot be accurately characterized A1

by zonal electrophoresis.

and A,,, that they isolated had a free

The two peptides, cysteine

and no disulfide bonds. On structural grounds, it would be highly unlikely that they purified insulin-like without disulfide bonds.

Their preparations

molecules

probably

contained small amounts of SM(s) to account for their activity.

Based on the immunological behavior to IGF-I and

-II antisera, their preparation appears to be a mixture and Zapf, this volume). finding in A 1

(6,

This is supported by their own

of 5% N-terminal glycine, the same

N-terminal residue as in IGF-I.

The latter would

explain

why their antisera developed against "SM-A" is 10 times more sensitive to IGF-I/SM-C than to

n

SM-An

(7).

Since

SM-A has no unique assay, the only way to answer the riddle of the identity of SM-A is to characterize the neutral material on isoelectric

focusing.

Thus, we adopted

the

original purification procedure which consisted of acid ethanol extraction of Cohn fraction IV, Sephadex

84 chromatography The

SM focusing

traditionally by h i g h

Our

in the

acid

have

liquid

predominantly

isoelectric

pH r a n g e ,

SM-A,

was

SM-C

in the SM-A IGF-II with

focusing.

which

purified

chromatography

shown that

the material

and

neutral

been called

perfomance

studies

that is

in a c e t i c

to

and

homogeneity

characterized.

is i d e n t i c a l

peak

has

to

IGF-I

on isoelectric

variable

amounts

of

and

focusing IGF-I.

METHODS

Acid

ethanol

performed Basel,

extraction

by

Ritschard

Switzerland.

chromatographed provided

human

and

(8)

to u s .

In our

Cohn fraction

Roncari

at

The acetone

on Sephadex

rechromatographed The

of

G-75

on Sephadex

G-50

purification was monitored

by

was

Hoffmann-LaRoche, precipitate

and

laboratory,

IV

active these

in 0.5

was

fractions

fractions M acetic

the h u m a n

were

were acid.

placental

1 25 radioreceptorassay active and

fractions

0.75 were

isoelectric were

chromatography

lyophilized

from

chromatography

and

a t a Kp b e t w e e n

then subjected

on Sephadex

G-75.

the a m p h o l y t e s

in 0.5

length

then purified

I - I G F - I as a tracer.

which migrated

focusing

separated

effective

using

M acetic

of 2 0 0

cm.

to h o m o g e n e i t y (HPLC)

using

acid

The by

by

0.55

to f l a t

SM-containing Sephadex

peaks

performance

an Beckman

C „

bed peaks

G-50

on a column with

lyophilized high

The

an

were

liquid

reverse

85 phase

column

i n 0.1 2.1.

M potassium The

Sephadex

Amino acid

eluting

with a stepwise

phosphate/phosphoric

SM-containing

acid

composition was

analyzer.

Cystine

performic

a Beckman

sequenator.

C-terminal

liberated

Y and

C-terminal

o n the a m i n o

of

solvent

acid

systems,

analysis.

in embryonic

was

and,

pH

on Biogel

or

Durrum

by

amino

cysteic

Sequencing

acid

was

done

on

consecutive

B digestions.

determined

activity chick

c a s e of

was established

cartilage

by H e r i n g t o n ) ,

fibroblasts

Immunologic

SDS-PAGE,

The

directly

by

activity

by N i s s l e y

activity identity

of

sulfation by G a r l a n d

activity and

adipocytes

cultured and our

in B a l b / c - 3 T 3 to

in 2

C-terminal

in i s o l a t e d

mitogenesis

HPLC

N-terminal

SM-C,

(performed

(performed

progression

by

composition,

in the

(performed

Stiles).

acids were

acid

insulin-like

cycle

as

determined

determined

Jennings),

cell

determined

oxidation.

amino

amino

sequence

biologic

with

carboxypeptidase

peptides was

analysis

embryo

solution

analyzer.

sequence

The

desalted

determined

was

acid

sequence

carboxypeptidase

Purity

peaks were

acid

gradient

columns.

following

The

acetonitrile

chicken lab),

and

(performed

IGF-I was determined

by

with

the a n t i s e r u m

developed

by

Heber

and Liske.

(8,

9)

RESULTS

Somatomedin-C peak

from

peaks The

.

Isoelectric

Sephadex

designated

G50

A and

prestained

band which retained gel.

Sequence

done

at a level

contamination

C according

immunoreactivity

and only

agreement

with

of

have

acid

(Table

The

of

values

the N - t e r m i n a l

could

be d e t e c t e d

sequence

of

IGF-I

(Figure

residues

are

IGF-I

consistent sequence

sequence method. -Lys-Ser-Ala

by

carboxypeptidase

The

2).

with

which

purified was

active

and

for

only

a

of

single

elution from residues

greater

is i n

1).

SDS-PAGE

the

was

t h a n a 2%

found at

each

excellent

IGF-I and

contains

of

two

presence

of

IGF-I

sequence

was

by

our

found

to

Y followed

by

sequence,

(Figure 2).

cartilage

known

half-cystines

c a n n o t be d e t e c t e d

This

the

unidentified

carboxypeptidase

in the

determined.

all a g r e e d w i t h

The the

B digestion.

to t h a t

(Figure

19 r e s i d u e s w a s

C-terminal

initial

is i d e n t i c a l

after

major

I).

17 r e s i d u e s

in the

showed

residue was

composition

two

by H P L C .

detected

a single

SM-containing

pi

the 5 N - t e r m i n a l

the i n t e g r a l

no h i s t i d i n e

to t h e i r

fluorescein

that w o u l d

The amino

sequence

with

the

revealed

to h o m o g e n e i t y

analysis

position.

of

chromatography

C peak was purified

an aliquot

focusing

The

be

-Lys-Ser-Ala, SM-C

sulfation assay,

fat

87

RRA (jug/Fx) 800-

60040020024

20

/AV * i

16 12 8 CM FROM CATHODE

F i g u r e 1. I s o e l e c t r i c F o c u s i n g of S o m a t o m e d i n s o n a G - 7 5 Sephadex Flatbed. A l i q u o t s f r o m e a c h 1 cm s e c t i o n w e r e t e s t e d by t h e h u m a n p l a c e n t a l m e m b r a n e r a d i o r e c e p t o r a s s a y t o m e a s u r e a l l S M s a n d by r a d i o i m m u n o a s s a y f o r S M - C i n w h i c h t h e c r o s s - r e a c t i v i t y f o r I G F - I I i s l e s s t h a n 3%•

88 TABLE I A C O M P A R I S O N OF THE A M I N O ACID C O M P O S I T I O N OF

SOMATOMEDIN-C

W I T H THE K N O W N V A L U E S OF I N S U L I N - L I K E G R O W T H AMINO ACID

SOMATOMEDIN-C

INSULIN-LIKE GROWTH

(Experimental)

FACTOR-I

(Integrals)

Lys

3-35

3

Arg

5.98

6

Asx

5.27

5

Thr

3.11

3

Ser

5.27

5

Glx

7.18

6

Pro

it. 86

Gly

7.04

7

Ala

6.54

6

HCys

5.1

6

Val

3.00

3

Met

0.66

1

He

0.67

1

Leu

5.37

6

Tyr

2.5

3

Phe

3.52

4

His

0.24

0

Trp

ND

0

^ D e t e r m i n e d as c y s t e i c

acid

FACTOR-I

5

89 N-TERMINAL IGF-I SM-C

1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 Gly-Pro-Glu-Thr-Leu-Cys-Gly-Ala-Glu-Leu-Val-Asp-AlaGly-Pro-Glu-Thr-Leu-( )-Gly-Ala-Glu-Leu-Val-Asp-Ala14 15 16 17 18 19 Leu-Gln-Phe-Val-Cys-GlyLeu-Glx-Phe-Val-( )-Gly-

Figure

C - T E R M INAL 68 69 70 -Lys-Ser-Ala -Lys-Ser-Ala

2. A C o m p a r i s o n of the S e q u e n c e s of S o m a t o m e d i n - C (SM-C) and Insulin-Like G r o w t h F a c t o r - I (IGF-I)

N-TERMINAL 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 Ala-Tyr-Arg-Pro-Ser-Glu-Thr-Leu-Cys-Gly-Gly-GluA1a-Tyr-Arg-Pro-Ser-Glu-Thr-Leu-( )-Gly-Gly-Glu-

IGF-II SM-A

13 14 15 16 17 18 19 20 C-TERMINAL -Leu-Val-Asp-Thr-Leu-Glu-Phe-Val-Lys-Ser-Glu -Leu-Val-Asp-( )-Leu-Glu-Phe-ValFigure

3.

A C o m p a r i s o n of t h e S e q u e n c e of S o m a t o m e d i n (SM-A) and I n s u l i n - l i k e G r o w t h F a c t o r - I I (IGF-II)

A

90 TABLE

II.

C O M P A R I S O N OF THE AMINO ACID C O M P O S I T I O N

OF

SOMATOMEDIN-A

W I T H THE K N O W N V A L U E S OF I N S U L I N - L I K E G R O W T H AMINO ACID

SOMATOMEDIN-A (Experimental)

FACTOR-II

INSULIN-LIKE GROWTH

FACTOR-II

(Integrals) Lys

1.55

1

Arg

7.82

8

Asx

3.85

3

Thr

3-75

4

Ser

7.6

7

Glx

7.65

7

Pro

3.1

3

Gly

7.0

5

Ala

5.25

5 1

HCys

5-34

6

Val

3.5

4

Met

0.15

0

He

0.85

1

Leu

6.0

6 2

Tyr

2.68

Phe

3.6

4

His

0.382

0

Trp

ND

0

^

c o r r e c t e d for 20?

2

one

determination

decomposition

3

91 pad

assay

for

and

chick

embryonic

Somatomedin focusing reacted

insulin-like

A.

fibroblast

The SM-A

contained

peak

at least

immunologically component,

was

to be i d e n t i c a l

composition with

values

N-terminal

residue

at e a c h

the

terminal level

sequence

that would

identical

to

undetermined expected

for

residues 3).

The

with

that

of

A more

a 5%

(Figure

A

single

IGF-I was complete

done

gave

one

present

Nat

a

residues

of t h e 2

cystine

would

be

t h a n 3%

radioimmunoassay

membrane

HPLC,

acid

II).

impurity

a half

by

1).

agreement

residues

had less

IGF-I/SM-C

placental

component,

amino

(Table

of 20

being where The SM-A

isoelectric

in e x c e l l e n t

18 p o s i t i o n s w i t h

in the

in the h u m a n

from

purification

IGF-II

assay

assay.

IGF-I antisera

was

detected

progression

the minor

IGF-II.

determination have

(Figure

the

8 positions.

IGF-II at

crossreactivity active

to

identical

first

obtained

further

( m e a n of 2 r u n s )

the i n t e g r a l

of

after

cell

mitogenic

2 SM's;

with

The m a j o r found

activity,

and

was

radioreceptorassay.

DISCUSSION

SM-C was and

purified

C-terminal

indicated does

that

not have

Svoboda

et a l .

to h o m o g e n e i t y

regions. SM-C

The

(3)

Minor

sequenced

sequence

is c h e m i c a l l y

the C - t e r m i n a l

and

data

identical

extension

impurities

at the

(Figure

2)

to I G F - I

postulated

probably

N-

and

by

accounted

for

92 their

results.

system gave compared

We have

an unacceptably

to t h e

improved

Until

a single

would

be S M - C / I G F - I

The m a j o r

noted

of t h e

SMs

predominantly

sequence

analysis,

material

(Figure

from

the

should

A peak.

be u s e d

The m i n o r

immunoreactivity Zapf,

immunoreactivity (7)

The

material

nature

Thus

if w e w i s h

causing

unknown

but

peptide

cleavage,

other

with

the

neutral

region

a mixture

of

by

IGF-I/SM-C

like

of

to u s e

the

isolated

term

SM-A,

it

IGF-II.

for

the a l t e r a t i o n

the

of

in the of

IGF-I/SM-C

by o t h e r s ,

"A" a n t i s e r u m

genetic

designation

SM p e p t i d e s w e r e

accounts

be a r e s u l t

lab.

demonstrated

in the A peak w i t h

it to m i g r a t e

could

best

contained

been suspected

of t h e of

that

amounts

No

and

peak,

in our

the

IGF-II,

3)-

has

this volume)

employed

SM-A

of

and lesser

contamination

solvent

asymmetrical

was

called

synonomous

HPLC

versa).

study

focusing

their

is d e v i s e d

(or v i c e

on i s o e l e c t r i c consisting

broad

systems

nomenclature

finding

that

or

and

IGF-I/SM-C

raised the

by H a l l

et

al.

IGF-I-like

neutral

deamidation,

variants

(5,

pro

region

is

partial

forms

of

the

h o r m o ne.

Other

small

isoelectric find

SM

containing

focusing

a t pH 4 . 8

has

peaks were

(Figure

been

shown

1).

observed

The m i n o r

to r e s u l t

from

on

peak which an

we

enzymatic

93 modification

during

purification

m a k e s the p e p t i d e s m o r e acidic The pi 9.5

too could variant

except

result

for

or be a pro h o r m o n e

We believe

IGF-I/SM-C

circulating

& Humbel

is

immunoreactivity.

It

confirm

IGFs/SMs

(i.e.

the existence

not derived

IGF-II/SM-A),

sequence

a peak

the f i n d i n g s

forms of IGF/SM in adults.

do not preclude

or a

appears

experiment.

(1, 2) that there are

however,

volume).

et al. and

form. F r e q u e n t l y

that these r e s u l t s

Rinderknecht

by Svoboda

from a peptide m o d i f i c a t i o n

at pi 7 - 5 , but not in this

and

( H e r i n g t o n - this

peak has b e e n observed

uncharacterized

of IGF-I & IGF-II which

two

These

but their l e v e l s would

to

have

major

findings,

of other

from or p r e c u r s o r

of

unique IGF-I/SM-C

to be very

small.

The r e l a t i o n s h i p structural

of SMs A & C to IGF-II & I simplifies

homologies b e t w e e n species.

26 of the 29 N - t e r m i n a l IGF-I/SM-C w i t h

H o w e v e r , rat because

The basic rat SM has

r e s i d u e s identical

three r e s i d u e s u n d e t e c t e d

the b a s i s of this homology

with

human

(Figure it).

it can be called

rat

immunological

On

IGF-I/SM-C.

IGF-I is not the same as h u m a n in its

it has d i f f e r e n t

the

determinants

entirety (9,

11).

BRL-MSA

(multiplication-stimulating

conditioned

b u f f a l o rat liver

activity,

isolated

culture m e d i a ) w a s

from

sequenced

94 1 RAT I G F - I HUM I G F - 1

5

10

15

20

G P E T L C G A E L V D A L Q F V C G() G P E T L C G A E L V D A L Q F V C G D 25 ( ) G F T F N K ( ) T R G F T F N K Q T

Figure 4.

RAT I G F - I I HUM I G F - I I

A Comparison o f t h e S e q u e n c e s o f R a t I n s u l i n L i k e Growth F a c t o r - I ( I G F - I ) ( R u b i n e t a l ) w i t h Human I G F - I ( R i n d e r k n e c h t and Humbel).

1

10 20 A Y R P S E T L C G G E L V D T L Q F V A Y R P S E T L C G G E L V D T L Q F V

30 C S D R G F Y F S R P C G D R G F Y F S R P

40 S G R A N R R S R A_S R V_S R R S R

50 60 G I V E E C C F R S C D L A L L E T Y C G I V E E C C F R S C D L A L L E T Y C A T P A K S E A T P A K S E

Figure 5.

A Comparison o f t h e S e q u e n c e s o f R a t and Human I n s u l i n - L i k e Growth F a c t o r - I I s A f t e r Marquardt e t a l and R i n d e r k n e c h t and Humbel. Different residues are underlined.

by M a r q u a r d t human

et al and

IGF-II/SM-A

found

to be

(Figure 5) (2).

homologous Four

s u b s t i t u t i o n s w e r e in or i m m e d i a t e l y peptide

of the

adjacent

with

five "Cn

to the

region.

CONCLUSION

SM-C is i d e n t i c a l as S M - C / I G F - I . 'The A peak amounts

to IGF-I and can correctly

No s e p a r a t e , u n i q u e

in i s o e l e c t r i c

of modified

focusing

SM-A h a s been is a m i x t u r e

IGF-I and m a j o r a m o u n t s of

Thus it would

seem a p p r o p r i a t e

SM-A/IGF-II.

Other m i n o r p e a k s isolated

p u r i f i c a t i o n appear

be referred

to refer

to be m o d i f i e d

to SM-A

t

found.

of

small

IGF-II. as

during

forms of

IGF-I and

-II.

ACKNOWLEDGEMENT Support for these studies was provided by the National Institute of Health HD 14506. REFERENCES

1.

Rinderknecht,

2769-2776 2.

E., Humbel

R.E.:

J. Biol.

Chem.

253,

89,

283-28

(1978).

Rinderknecht,

E., Humbel

R.E.: F E B S Letter

(1978) 3.

S v o b o d a , M . E . , Van Wyk,

J.J., K l a p p e r ,

R.E., G r i s s o m , F.E., S c h l u e t e r , 790-797,

(1980).

R.J.:

D.G.,

Biochem.

Fellows, 19,

96 4.

Van Wyk,

Clin. 5.-

Endocrinol. Hintz,

Pediatric

68,

Hall,

Zapf,

10. R.A.:

J., Morell,

Rubin,

12.

J.S.,

R.C., Metab.

Marquardt,

Oroszlan,

S.:

in

J.

(1980). Seegan,

G.,

on Recent

Advances

Clin.

in

press. Froesch,

E.R.:

J.,

Engberg,

G., Fryklund,

48,

R.:

H.,

271-278 Horm.

J.

Invest.

95,

I.K.,

Axiak, 54,

S.,

474-476

Todaro,

J. B i o l .

Chem.

Z.,

(1976). Froesch,

W.H.,

Bradshaw,

(1982).

Raison,

R.L.:

J.

Clin.

(1982).

G.J., 256,

J.

(1980).

Daughaday,

110, 734-740

201-213

H., L a r o n ,

505-517

L.:

(1979).

Res. 7,

B., W a l t e r ,

Mariz,

Endocrinology

Endocrinol.

D.,

Symposium

L.E.:

H.,

Endocrinol.

Baxter,

206-208

Chang,

Metab.

K., Liske,

Acta

Dnderwood,

(1981).

Endocrinol.

9. E.R.:

F.,

K., Brandt.

Reber,

M.E.,

50,

Serono

J., W a l t e r ,

8.

11.

E.:

1321-1330

Clin.

Metab.

Endocrinology,

Zapf,

7.

Svoboda,

R., Leu,

Rinderknecht,

6.

J.J.,

Henderson, 1859-1865

L.E., (1981).

A COMPUTER GRAPHICS STUDY OF INSULIN-LIKE GROWTH FACTORS AND THEIR RECEPTOR INTERACTIONS

Annemarie Honeggert and Tom Blundell Laboratory of Molecular Biology, Department of Crystallography, Birkbeck College, University of London, London WC1E 7HX, UK ^Present address : Biochemisches Institut der Universität Zürich, CH-8028 Zürich, Switzerland

Introduction The close homology of the insulin-like growth factors to proinsulin, and their ability to bind insulin receptors, provide good evidence that they are all members of a family of growth factors and hormones derived by divergent evolution from a common ancestral polypeptide (1, 2, 3).

The evidence is

strengthened by the observation that the sequences of insulins and insulin-like growth factors (IGF I and II)

are compatible

with similar three-dimensional structures each comprising not only identical main-chain conformations in large regions of the polypeptides, but also an identical arrangement of conserved residues including the disulphide bridges and the hydrophobic core (3, 4).

The availability of powerful interactive com-

puter calligraphic systems now allows precise models of the insulin-like growth factors (5) to be constructed from the coordinates of porcine insulin defined by high resolution X-ray analysis (6).

These models form the basis for a systematic

study of the relation of structural differences to variations in receptor affinity and antigenicity.

In this paper we

briefly review the models for IGF I and II, which are detailed elsewhere.

We then describe the use of a new computer program.

Insulin-Like Growth Factors / Somatomedins © 1983 Walter de Gruyter & Co., Berlin • New York

98 BILBO (7) in a study of the putative receptor binding surfaces of the insulin-like growth factors.

Computer Models of IGF Structures Figures 1 and 2 show equivalent stereo views of IGF I and II (4, 5). These models were constructed using an interactive computer program, FRODO (8) rewritten and extended by I. J. Tickle and T. A. Jones. The general tertiary structures are indicated in simplified diagrams in Figure 3 and they are shown schematically in comparison to insulin and proinsulin in Figure 4.

Figure 1 : A stereo view of IGF I (all atoms) with numbering as in Table 1

Figure 2 : A stereo view of IGF II (all atoms) with numbering as in Table 1

99

Figure 3 : IGF I and IGF II (only Ca positions shown) viewed from the same direction as those of Figures 1 and 2 and showing the positions equivalent to the receptor binding (•) and antigenic (•) sites of insulin.

Figure 4 : Schematic diagrams of insulin, proinsulin, IGF I and IGF II to demonstrate the family relationships. The main-chains of IGF I and IGF II between residues B4 and B27 and between residues A1 and A20 (insulin numbering, see Table 1) are constructed with torsion angles identical to those of molecule 2 of the insulin dimer as defined by the high resolution refinement of porcine insulin (6).

The conserva-

tion of the glycines at B8, B20 and B23 allows the unusual torsion angles of insulin to be attained in both growth factors.

The side-chains of the invariant residues of the hydro-

phobic core - B6 Leu, Bll Leu, B12 Val, B15 Leu, B18 Val, B19 Cvs, B24 Phe, A2 lie, A3 Val, A6 Cys, All Cys, A16 Leu, A19 Tyr and A20 Cys - and the invariant residues on the sur-

100

face - B7 Cys, A7 Cys, B22 Arg, A13 Leu and All Glu - have positions identical to those in insulin.

The conservatively

varied residues in IGF I and IGF II such as B4 Glu (Gin), B13 Asp (Glu), B21 Asp (Glu), B25 Tyr (Phe), B26 Phe (Tyr), A4 Asp (Glu), A5 Glu (Gin), occupy positions close to those of insulin.

Where side-chains vary, like charges are kept apart,

hydrophobic groups are buried where possible and contacts less than van der Waal's distances avoided. The conformations of the N-termini of the B-chains and the Cand D-peptides are based upon predictions of the secondary structure using Chou and Fasman (9) techniques, optimisation of tertiary interactions both within the peptides and between them and the remaining tertiary structure, and an attempt to conserve the secondary structure between IGF I and IGF II. Apart from the regions B28 - C5 which have a high potential for 3-turns, there is no evidence that the C-peptides have any regular secondary structure.

Their highly polar nature is

consistent with positions on the surface of the molecules with side-chains exposed to solvent.

The deletions in the IGF II

sequence indicated by the alignment in Table 1 are coincident with bends in the IGF I structure and are therefore easily accommodated in a similar tertiary structure.

D-peptides can

also have a conformation with a g-turn in IGF I where a deletion occurs in IGF II (see Figure 3).

The g-turns

allow the a-carboxylate of D8 to play the role of the acarboxylate at A21 of insulin and form an ion pair with B22 Arg.

The conserved A17 Glu may also form an ion pair

with B22 Arg so that the positively-charged guanidinium group is placed between two negatively-charged carboxylate groups. The N-terminal residues of the B-chain (B2 - B3) of IGF I can occupy positions equivalent to those of insulin, but the extension in IGF II allows some flexibility in the model.

It may

extend into the solvent or alternatively fold back towards the molecular surface in the region of A13 Leu and A14 Ala.

Both

101

I

i

i í è

\\ S -S o o2 o £ * 3 «

s

-

i

È e

•

tf a

o> S « pin

tP

p>

- a a ••S S « s * * a 3 1 ™

£

'

"

5 S

© s i - á a : £ a 3 - â a - « a

il a à s\3| a

© * h 2 ä 8 h o> » « a

fi « » a

i i

e i a

§ •

n" Q

P

eu

&

5

3&e a >o«ig

35a a 3>»«

fi.

$

«

2

a

a3^ a G «I >1 j«|

3 3 3 ) o

102

arrangements allow B1 Arg to form an ion pair with A12 Asp, but the second arrangement allows hydrophobic interactions between B-l Tyr and B-2 Ala, and A13 Leu and A14 Ala. We note that in IGF I where the N-terminus is shorter, A14 is an arginine. The models have been subjected to energy minimisation using the program of Levitt (10). The coordinates of the models are deposited with the Brookhaven Data Bank. The molecular structures here have many attractive features, one of which is the extensive system of charge interactions linking C7 Arg, A4 Asp, A9 Arg, A5 Glu, A15 Arg, A12 Asp and A14 Arg in IGF I.

In insulins, A5 and A15 are both gluta-

mines, but here they are changed complementarily. Much of the region equivalent to that involved in insulin dimer formation (see 11, 12 for reviews) is conserved in IGFs, for example B12 Val and B2 4 Phe, or conservatively varied, for example B25 Phe (Tyr) and B26 Tyr (Phe), where IGF residues are shown in parenthesis. It is therefore possible that IGFs might dimerise. However, much of the region involved in zinc hexamer formation, including BIO His which binds zinc in insulin, is quite different in the growth factors. Some residues are more polar, for example B4 Gin (Glu), B14 Ala (Thr) and A14 Tyr (Arg, Ala) in the IGFs,and the B-chain Nterminus is variable. It is therefore unlikely that IGFs form hexamers. However, the unusual hydrophobic patch involving B14 Ala, B17 Phe, B18 Val and A13 Leu of IGF I would be available for interaction with a binding or carrier protein. In IGF II the surface would look a little different as the extension of B-2 Ala and B-l Tyr may lie over A13, and B14 becomes a threonine as is often found in hystricomorphs which do not form hexamers. The fact that IGF II has a higher affinity to the carrier protein than IGF I (13) may indicate that some of these changes enhance the binding with the carrier protein.

103

With respect to the antigenicity of the insulin-like growth factors, the region of insulin which is most often involved in binding antiporcine/bovine insulin antibodies is indicated in Figure 3.

It includes B2, B3, B4, A8, A9 and A10 which are

adjacent in the three-dimensional structure.

These regions

are quite different in IGFs from insulins either of pig or of guinea pig, in which antibodies are often raised.

Thus it is

expected that neither IGF I nor IGF II bind anti-insulin antibodies - that they are not suppressible.

Computer Simulation Using BILBO, of Receptor Binding Regions The residues which have been considered important to the receptor binding and biological activity of insulin are centred around B22 - B25 and are indicated in Figure 3. Much of the rather hydrophobic surface involved in formation of insulin dimers appears to be important for receptor binding. These residues include B12 Val (Val), B16 Tyr (Gin), B24 Phe (Phe), B25 Phe (Tyr), B26 Tyr (Phe) where the insulin-like growth factor residues are shown in parentheses. Polar residues around the periphery of this region may also be of importance including A1 Gly (Gly), A19 Tyr (Tyr), A21 Asn (Ala), B13 Glu (Asp), B21 Glu (Asp), B22 Arg (Arg), B23 Gly (Gly). Many of the residues are conserved between insulins and insulin-like growth factors and these may explain the ability of insulinlike growth factors to bind insulin receptors (13). However, there are also many changes in this surface region, most notably in the addition of the C- and D-peptides. To consider these changes we need to model the receptor region and generate complementary surfaces. We also need to consider the regions of the insulin-like growth factors which may bind their own receptors.

104

For this purpose we have used BILBO (7), written for the Evans and Sutherland PSII, an interactive calligraphic (line drawing) system, and designed for the study of protein surface topography and protein-protein interactions. We first identify a fragment of the insulin or IGF molecule which includes the amino acid residues involved in receptor binding. This is defined in BILBO by a least-squares plane through up to ten points on the protein backbone entered by pointing at the appropriate atoms on the display. The plane thus generated can be shifted into a better position if desired. Only the surface points distal (away from the centre of the molecule) to the plane are generated, although atoms up to one atomic radius from the plane can contribute to the surface and therefore have to be included in the calculations. For the insulin receptor binding region the plane can be positioned so that the fragment includes the side-chains of all those residues listed above, or alternatively a more restricted set such as B24, B25 and B26. The surface is defined by a fast bitmapping procedure (14) which calculates 6000 surface points in one minute and which has been modified to allow the retention of a connection between the surface point information and the atomic centre. The space occupied by the molecule is 15 mapped on to a 2 bit, three-dimensional binary array. A flexible spacing of the grid is used to avoid having a limit to the size of the molecular fragment. Atom by atom, the molecular fragment is then mapped on to a binary array, setting to 'l" each bit which is within a distance of one atomic radius The rounded to the nearest grid unit of the atomic centre. resulting volume bitmap can then be converted to a surface bitmap by eliminating all 11' bits which do not have at least one '0' bit nearest neighbour, or a variety of logical operations can be performed between the volume bitmaps of different molecules, eg, 'and', 'exclusive or 1 , 'inclusive or', 'and not' to highlight differences and similarities in the spacial requirements of these molecules. The composite bitmaps thus

105

obtained can then be converted to surface bitmaps, using the same method as for single maps. The surface bitmaps are read off atom by atom in the order in which they were created, to retain a connection in the form of a pointer between the surface points and the atoms concerned. This pointer allows one to look up atomic properties and to selectively display for example, charged groups, hydrophobic parts of the surface or the part of the surface belonging to a given residue.

Surface points generated by the bitmapping

procedure can be displayed directly, but at most orientations of the display object, lining up of the points create undesirable visual effects.

For this reason each point of the

surface is shifted slightly along a line through the centre of the atom to which it belongs until its distance from the centre equals the atomic radius (a shift of a maximum of 0.5 grid units).

This procedure improves the accuracy of the

surface representation and destroys the regularity of the point distribution. Figure 5 shows a van der Waals surface of part of IGF I in the region which is topologically equivalent to the receptor binding region of insulin. Even using the corrections to the point coordinates described above, and using intensity depth cueing, point representations do not give a very good spacial impression of the threedimensional shape of an object, especially in photographs, where motion as an additional depth cue is missing.

Other

ways of representing a surface include giving a clue to the slope of the surface at each point; this can be achieved by drawing vectors perpendicular to the surface instead of points giving a fur-like effect.

Varying the length of these vectors

results in different surface textures which can be used to code for different surface properties in the absence of colours. Such a "fur" representation is shown in Figure 5(b).

This is

106

(b)

Figure 5 :

van der Waals surface with "fur" representation; "stars" represent positive charges

The surface of IGF I, topologically equivalent to

108

equivalent to that shown in Figure 5(a) but emphasises the positively-charged lysine and arginine groups by extending the length of the vectors to the atomic centres so giving a starlike effect. To generate water accessible (15) surfaces, a water radius (1.48) is added to the van der Waals radii of the atoms before mapping them on to the bitgrid. Thus the water accessible surface constitutes the geometrical location of the centres of water molecules in van der Waals contact with the protein. Figure 5(c) shows the water accessible surface for the IGF I fragment shown in Figure 5(a) and (b). An alternative surface - the so-called re-entrant surface (16) - can also be usefully displayed using BILBO. This is calculated by removing bits within 1. 48 of each point on the water accessible surface. It therefore constitutes contact points of the water molecules on the van der Waals surface but "fills in" inaccessible clefts. Finally, it is often useful to display the surface by joining up the surface points by lines, either contouring or drawing a net over the surface. This is illustrated in Figure 5(d). The net may also be considered as a model of the surface complementary to the hormone or growth factor, and may thus be considered as a crude model of the receptor. BILBO can also be used to compare the receptor binding sites of insulins and insulin-like growth factors. this the molecules must be first aligned.

In order to do This can be

achieved by displaying more than one molecule on the screen at the same time.

One molecule is rotated and translated until

a good visual fit is achieved and this is then optimised by using a least-squares procedure (17, 18).

The choice of

atoms to be fitted may be determined by the computer on the basis of proximity or they may be identified by the user. Once the two molecules are aligned, locations of charged groups, potential hydrogen bonds and hydrophobic patches may

109

be compared.

The two molecules can be bitmapped and logical

operations performed between the bitmaps.

For example, the

use of 'inclusive or1 for a large number of fully active hormones or growth factors will map out the volume or space required at the receptor to accommodate them all.

Other

operations such as 'and not1 or 'exclusive or' may identify those regions which might sterically prevent the hormone or growth factor binding to the receptor with full affinity. Figure 6(a) shows the result of an 'and' operation on IGF I and porcine insulin. Those parts shared by both are identified. More interestingly. Figure 6(b) shows an 'IGF I and not insulin' operation so identifying differences in IGF I which would interfere with receptor binding at an insulin receptor. These show clearly that although part of the insulin receptor binding region is conserved in the insulinlike growth factors, the C- and D-peptides would tend to interfere with the formation of a strong complex between the growth factor and the insulin receptor. It is also clear from comparisons of the two insulin-like growth factors that IGF II should bind more strongly to the insulin receptor as the deletions at C2 and C3 and at D2 and D3 make the receptor binding region more available for interaction. Nevertheless, it is clear that both growth factors might bind the insulin receptor but with reduced affinity. The methods described here can also be used to define the IGF receptor binding regions. However, in order to do this we need to have available structure activity data for each of the receptors for a range of analogues. The synthesis of these analogues, perhaps using recombinant DNA technology, must now be a first priority.

Figure 6 : Comparison of surfaces of insulin and IGF I. (a) IGF I 'and' insulin; (b) IGF I 'and not' insulin.

Acknowledgements We are grateful to Dr S. Bedarkar for making available coordinates for the models of IGF I and IGF II, and to Dr I. J. Tickle and L. H. Pearl for help and discussions on computer program development. We thank the US Science and Engineering Research Council for their support of the Evans and Sutherland Picture System.

References 1. 2. 3. 4. 5.

Rinderknecht, E., Humbel, R.E.; J. Biol. Chem, 253, 2769-2776 (1978). Rinderknecht, E., Humbel, R.E.: FEBS Letters 89^, 283-289 (1978). Blundell, T.L., Humbel, R.E.: Nature 287, 781-787 (1980). Blundell, T.L., Bedarkar, S., Rinderknecht, E., Humbel, R.E.: Proc. Natl. Acad. Sei. USA 75, 180-184 (1978). Blundell, T.L., Bedarkar, S-, Humbel, R.E.: Fed. Proc., in Press

6.

Dodson, E.J., Dodson, G.G., Hodgkin, D.C., Reynolds, C.D. Can. J. Biochem. _57, 469-479 (1979) .

7.

Honegger, A.: BILBO, an Interactive Computer Graphics Program for the Study of Protein Surface Topography and Protein-Protein Interactions, Birkbeck College, London University, 1982

8.

Jones, T.A.: J. Appl. Cryst. 1J., 268-272 (1978).

9.

Chou, P-Y., Fasman, G.D.: Biochemistry L3, 222-224 (1974)

10. Levitt, M.: J. Mol. Biol. 82, 393-420 (1974). 11. Blundell, T.L., Wood, S.P: Ann. Rev. Biochem. jU, 123-154 (1982) . 12. Blundell, T.L., Pitts, J.E., Wood, S.P.: Crit. Revs. Biochem. ¿3, 141-213 (1982) . 13. Zapf, J., Schoenle, E., Froesch, E.R.: Eur. J. Biochem. 87, 285-296 (1978) . 14. Pearl, L.H., Honegger, A.: J. Mol. Graphics, in Press 15. Lee, D., Richards, F.M.: J. Mol. Biol. 55, 379-400 (1971)

112

16. Greer, J., Bush, B.L.: Proc. Natl. Acad. Sci. 75, 303-307 (1978). 17. McLachlan, A.D.: J. Mol. Biol. 128, 49-79

(1979).

18. Remington, S.J., Matthew, B.W.: J. Mol. Biol. 140, 77-79 (1980) .

EVIDENCE FOR PROTEOLYTIC CONVERSION OF INSULIN-LIKE GROWTH FACTORS TO A BIOLOGICALLY ACTIVE ACIDIC FORM

Adrian C. Herington and Adrien D. Kuffer Medical Research Centre, Prince Henry's Hospital, Melbourne, Australia 3004.

In addition to the widely recognized biologically active forms of human insulin-like growth factors (IGF)/somatomedins (Sm) (viz. IGF-I/SmC pi ^ 8.4; SmA pi ^ 7.4; IGF-II pi ^ 6.5) (1) a fourth acidic form of nonsuppressible insulin-like activity (ILA pi 4.8) has recently been described (2).

This latter species, which possesses many of the biolog-

ical characteristics and/or activities of the other IGFs (3), has not been well studied despite being recognized more than 10 years ago (4,5). However, the recent identification in human serum of a specific IGF inhibitor (6), which has a pi ^ 4.4, provides an explanation for the apparently low serum concentration of ILA pi 4.8. More recently we have also shown that an IGF purification procedure based on the pH 5.5 ion exchange (SP Sephadex) method of Cockerill et al (2) , together with a step designed to remove IGF inhibitor (6) , results in a very high yield of ILA pi 4.8 with concomitant loss of IGF-I and IGF-II (7).

In addition, in comparing purification protocols used by different

laboratories we demonstrated that the relative yields of the various IGF/ Sm species were dependent principally on the initial approach used (7). The main protocols studied, together with their corresponding yields of ILA pi 4.8, IGF-II, SmA and IGF-I/SmC are summarized in Fig 1.

Of

particular interest is the relatively high proportion (^21%) of ILA pi 4.8 obtained by the "standard" methods (A & C) once the IGF inhibitor is removed by the Biogel P-30 chromatography step.

A similar pattern, but

with a lower overall yield, was obtained using an initial acid-ethanol extraction (data not shown).

The distribution pattern with the pH 5.5

SP Sephadex procedure (method B) , however, is the most striking with only

Insulin-Like Growth Factors / Somatomedins © 1983 Walter de Gruyter & Co., Berlin • New York

114

Methods and Results The isolation procedures A and B (Fig 1) were used primarily for these studies.

Following flat bed isoelectric focusing (IEF) the pooled pH

regions were dialysed and assayed at 3 doses in the isolated rat adipocyte bioassay in the presence of excess anti-insulin antiserum, as previously described (8).

Potencies of all fractions were computed

relative to an insulin standard (assayed without antiserum) by accepted methods for parallel line bioassays.

All data are expressed as mU

insulin-like activity recovered/100 gm Cohn IV-1 or as a % of total activity recovered/IEF plate. Since the major procedural differences between methods A and B exist only in the initial steps, these differences (asterisked in Fig 1) were assessed for a possible role in the conversion of the IGFs to the acidic ILA pi 4.8 form. a)

The effect of differences in pH conditions was tested by a series of cross-over experiments:

i)

Cohn IV-1 was extracted at pH 5.5 (as in method B) and then acidified to pH < 2.8 with acetic acid (as in method A).

Following

"reneutralization" (with NH^OH) to pH 5.5 it was subjected to the normal SP Sephadex procedure (method B) . A typical distribution pattern for method B (as shown in Fig lb) was obtained i.e. isolation of only ILA pi 4.8 ii)

Conversely, as shown in Fig 2a, extraction of Cohn IV-1 at pH 5.5 (as in method B) , prior to acidification (pH < 2.8) and subsequent acid gel filtration (as in method A), gave the distribution pattern expected for method A (as shown in Fig la).

Thus it appears from i) and ii) that exposure to different initial pH conditions per se (pH < 2.8 or 5.5) plays no role in the conversion process. iii) The involvement of the second major pH difference between methods A and B (i.e. the pH 9.7 elution step of the SP Sephadex procedure) was tested as follows; Cohn IV-1 was extracted at pH 5.5 followed by adjustment to, and maintenance at, pH 9.7 for 48 hrs (a duration experienced in the large-scale batch procedure of method B) prior to

115 METHOD A Acidified Oohn IV-1 Oohn IV-1 dissolved in 1» Formic Acid (pH 2.3)*

Acidified Serum Serum made 1% Formic Acid

SP Sephadex Ion Exchange Cohn IV-1 extracted in 0.05H NH 4 OAc pH 5.5* SP Sephldex C-25 pH Step Elution pH 5 . 5 — 6 . 4 — 9 . 7 * IGF Active Fraction

J

. Sephadex G-75 1* Formic Acid |

«

(kav 0.4 - 0.8)

Biogel P30 1% Formic Acid | (kav 0.5 - 0.9) Flat-Bed Isoelectric Focusing pH 3.5 - 10 j pi 4.8 ± 0.5 (ILA pi 4.8)

pi 6.2 ± 0.5 pi 7.4 ± 0.5 (IGF II) j (Sm A)

pi 8.5 ± 0.5 (IGF 1/SmC)

IGF Bioassay 80 r . < a. w £ 60 Z o

ACIDIFIED

COHNS

SP SEPHADEX

YIELD :96 8±I4 0 (SEM) n-3 mU/IOOgm

YIELD 87 5 ± 27 0 n-3 mU/l00gm

20

SERUM YIELD : 32-9 + 7 7 (SEM) n-3 mU/litre

a. 4 0
II)

placental membranes have most

polypeptides:

IGF II

NOT APPLICABLE

as t h e p r e s e n c e o f s p e c i f i c

human placental

(I > II)

NOT APPLICABLE TO RAT SERUM

The crossreactivity of various

AFTER ACIDIC GEL FILTRATION ON SEPH. G-50 (SMALL MOL. WT. FRACTION) OR ACID/ETWANOL EXTRACTION (SUPERNATE) TOTAL IGF

for I G F R R A s .

frequently

I (26,28) or SM-A (30) a n d M S A

In c o n t r a s t ,

liver and rat

membranes

to be a m o s t

suitable matrix

appear

s p e c i f i c d e t e r m i n a t i o n of I G F II II-like polypeptides. several

inhibitory

bioassays hormones

(like P D G F

and p r o v i d e v a l u a b l e

serum

(or o t h e r b i o l o g i c a l

should h o w e v e r , be

for

the

(33),

animal

to

assay or

in

thyroid tool

in

fluids or tissue are u s e d w h e r e

at

IGF

information

RRA results obtained

species

(27).

that

assay) do not crossreact

F o r e x a m p l e , R R A s in w h o l e

interpret

IGF

corticosteroids,

in

in

extracts) the

and affinity of IGF carrier p r o t e i n may e s s e n t i a l l y from human serum.

is

interpreted w i t h c a u t i o n , above all

sera from different

difficult

i.e.

IGF activity

supplementary

and R I A s .

these

of R R A s

T h u s , R R A s are a n i m p o r t a n t

addition to bioassays whole

(like

simulate

in t h e t h y m i d i n e

in t h e c h i c k c a r t i l a g e

the IGF receptor. research

serum components

the

placental

(32) a n d M S A

O n e of t h e a d v a n t a g e s

etc.) or those which

to

considerable

crossreactivity.

oestrogens,

Human

(29). H o w e v e r ,

(31) s h o w

well

membranes

been used

RRA does not specifically measure

I G F II

as

if amount

differ

rat serum

are

162 ii) C o m p e t i t i v e p r o t e i n b i n d i n g a s s a y The p a r t i a l l y p u r i f i e d

(table 2)

IGF c a r r i e r p r o t e i n f r o m h u m a n

h a s b e e n u s e d as a " r e c e p t o r "

for t h i s r a d i o l i g a n d

(34,35). A s a m a t t e r of fact, the assay c a n n o t be o u t in w h o l e s e r u m b e c a u s e s e r u m itself c o n t a i n s

assay carried

carrier

p r o t e i n . T h e r e f o r e , I G F h a s to be s e p a r a t e d from the protein before

it can be q u a n t i t a t e d . T h e a s s a y

serum

carrier

measures

t o t a l IGF. H o w e v e r , s i n c e the p a r t i a l l y p u r i f i e d

carrier

p r o t e i n f r o m h u m a n s e r u m h a s a g r e a t e r a f f i n i t y for I G F II t h a n for I G F I this a s s a y o v e r e s t i m a t e s

I G F II r e l a t i v e

I G F I. If I G F II is used as a t r a c e r the r e s u l t s

to

reflect

p r e d o m i n a n t l y the I G F II c o n t e n t of the s a m p l e . T h e

assay

is u s u a l l y c a r r i e d o u t w i t h I G F I t r a c e r , and a p a r t i a l l y p u r i f i e d IGF p r e p a r a t i o n c o n t a i n i n g

a 1 : 1 m i x t u r e of IGF I

and II is used as a s t a n d a r d . T h e s t a n d a r d "standardized" calculated pad insulin

itself

in the fat p a d a s s a y so t h a t the

is

first

results

f r o m the s t a n d a r d c u r v e c o r r e s p o n d to uU of

fat

equivalents.

C a r r i e r p r o t e i n from r a t liver e x p l a n t s h a s also b e e n in a c o m p e t i t i v e p r o t e i n b i n d i n g assay

used

(36). S i n c e the

rat

c a r r i e r p r o t e i n has a h i g h e r a f f i n i t y for IGF I t h a n for II, this a s s a y m e a s u r e s p r e d o m i n a n t l y iii) R a d i o i m m u n o a s s a y s

IGF

I G F I.

(RIAs)

To d a y , s e p a r a t e d e t e r m i n a t i o n of IGF I and II is o n l y p o s s i b l e by R I A . S p e c i f i c a n t i b o d i e s t o w a r d s v a r i o u s

SMs

h a v e b e e n p r o d u c e d d u r i n g r e c e n t y e a r s . T h e first I G F RIA d e v e l o p e d by R e b e r and L i s k e

(37) w a s c a r r i e d o u t in w h o l e

serum under equilibrium conditions. Therefore, results

are

d i f f i c u l t to i n t e r p r e t . In o r d e r to p e r f o r m the I G F RIA under equilibrium conditions serum

I G F h a s to be e x t r a c t e d

from

(38). F u r l a n e t t o has c i r c u m v e n t e d t h i s d i f f i c u l t y

by

163 a p p l y i n g disequilibrium c o n d i t i o n s serum

for the S M - C R I A in w h o l e

(39). T h e R I A for S M - A of H a l l et al. a l s o

disequilibrium conditions

(40) in w h o l e s e r u m .

uses

However,

l i m i t e d a c c e s s i b i l i t y to the a n t i b o d i e s of the IGF to the c a r r i e r as w e l l as the p o s s i b l e

interference

nonsaturated binding protein render quantitative

complexed of

determination

d i f f i c u l t . A c i d i f i c a t i o n and s u b s e q u e n t l y o p h i l i z a t i o n serum samples, which dissociates

IGF f r o m the

of

carrier

complex and d e s t r o y s m o s t of the c a r r i e r p r o t e i n , a l l o w s more quantitative

assessment

P r e t r e a t m e n t of s e r u m by a c i d i c gel f i l t r a t i o n acid/ethanol extraction

a

(41).

(27) p r i o r to the RIA

(8,42) or by eliminates

m o s t of the p r o b l e m e s e n c o u n t e r e d w i t h IGF RIAs in w h o l e serum. S p e c i f i c a n t i b o d i e s a g a i n s t IGF I and II have b e e n in r a b b i t s

produced

(8). W i t h these a n t i b o d i e s the c r o s s r e a c t i v i t y

of

IGF II in the IGF I R I A is 1 %, t h a t of IGF I in the IGF II RIA is 10 % (fig. 3). T h e c r o s s r e a c t i v i t y o f S M - C in the

IGF

I and II R I A is m o r e or less i d e n t i c a l to t h a t of IGF I (8)). S i m i l a r l y , the c r o s s r e a c t i v i t i e s of IGF I at the S M - C a n t i b o d y or at the a n t i b o d y to b a s i c SM are i d e n t i c a l t h o s e of S M - C or b a s i c S M , r e s p e c t i v e l y

to

(30,41).

T h e c r o s s r e a c t i v i t y of SM-A in the IGF I and II RIA is 10 % each

(8). R a t SM c r o s s r e a c t s

30 % w i t h the I G F I a n t i b o d y ,

w h e r e a s t h e r e is no s i g n i f i c a n t c r o s s r e a c t i o n w i t h the II a n t i b o d y

IGF

(unpublished).

IGF f r a g m e n t s w h i c h are i n a c t i v e

in the v a r i o u s

bioassays,

in the rat liver R R A and in the p r o t e i n b i n d i n g

assay, may

be " a c t i v e "

in R I A s . T h e s y n t h e t i c IGF II f r a g m e n t 27 -

45 is an e x a m p l e

(fig. 4): it is as p o t e n t an i n h i b i t o r

IGF II itself o n a w e i g h t b a s i s

as

in the I G F II R I A , w h e r e a s

it

164 Competitive inhibition of the binding of « v - labeled IGFJl CBJ to IGF I - a n d J I - a n t i s e r u m i final dilution labeled IGF I , 1G FIT, somatomedin A, somatomedin norma! serum.

is c o m p l e t e l y above

IGF I 1:2000) C and

(A) and by unstripped

i n a c t i v e in all the o t h e r a s s a y s

mentioned

(43). S u r p r i s i n g l y , the IGF I f r a g m e n t 25 - 41

some c r o s s r e a c t i v i t y

in the IGF II R I A , w h e r e a s

shows

it is

i n a c t i v e in the IGF I R I A as w e l l as in all the o t h e r assays

( u n p u b l i s h e d r e s u l t s ) . T h u s the R I A m a y p i c k

i n a c t i v e IGF f r a g m e n t s p o s s i b l e y o c c u r r i n g other biological

above

up

in s e r u m o r

f l u i d s . T h i s w o u l d also be c o m p a t i b l e

the f i n d i n g that the a m o u n t of i m m u n o r e a c t i v e

IGF (I + II)

in h u m a n s e r u m is g r e a t e r t h a n the a m o u n t m e a s u r e d b i o a s s a y s or r a d i o l i g a n d

assays

with

by

(8).

H i n t z et al. h a v e used the s y n t h e s i z e d C - p e p t i d e and

the

D - r e g i o n of I G F I (42,44) and the C - p e p t i d e of IGF II

(45)

to p r o d u c e s p e c i f i c a n t i b o d i e s w h i c h a l l o w d e t e r m i n a t i o n of I R - I G F I and II. T h e s e a n t i b o d i e s show less s e n s i t i v i t y the n a t i v e p o l y p e p t i d e s t h a n the IGF I and II

antibodies,

but o n the o t h e r hand the c r o s s r e a c t i v i t y b e t w e e n the p o l y p e p t i d e s I G F I and II is m u c h s m a l l e r t h a n in R I A s a n t i b o d i e s a g a i n s t IGF I or

II.

for

native using

165 Crossreactivit/es of various synthetic fragments /n the IGF H PI A.

1

IGF

-

HI

1

1

10°

10'

'

1

10 2

ng of peptide

added/

'10*

10 1

Figure 4

0.4m!

G e n e r a l l y , R I A s and c o m p e t i t i v e p r o t e i n b i n d i n g a p p e a r to be t e c h n i c a l l y

less d e m a n d i n g

and m o r e

assays precise

t h a n R R A s or b i o a s s a y s . H o w e v e r , the a b i l i t y o f a c i r c u l a t i n g p o l y p e p t i d e h o r m o n e to b i n d to a s p e c i f i c m e m b r a n e m a y b e t t e r r e f l e c t its b i o l o g i c a l

receptor

a c t i v i t y t h a n a RIA

d e t e r m i n a t i o n , a b o v e all if the b i o l o g i c a l l y a c t i v e site

is

r e m o t e f r o m the i m m u n o l o g i c a l d e t e r m i n a n t . The e x p e r i m e n t fig. 4 m a y be t a k e n as an e x a m p l e : T h e i m m u n o r e a c t i v e

in

IGF II

f r a g m e n t 27 - 45 d o e s n o t b i n d to the c a r r i e r p r o t e i n nor to the r a t l i v e r m e m b r a n e r e c e p t o r , nor is it b i o l o g i c a l l y a c t i v e . T h i s i n d i c a t e s t h a t the b i n d i n g s i t e of n a t i v e II for the a n t i b o d y is d i f f e r e n t

the c a r r i e r p r o t e i n o r to the m e m b r a n e r e c e p t o r and m a y t h e r e f o r e be r e s p o n s i b l e

for b i o l o g i c a l

which

activity.

a g a i n u n d e r l i n e s the c o n t e n t i o n t h a t d e t e r m i n a t i o n immunoreactive

IGF

from t h a t w h i c h b i n d s to This

of

IGF m a y s o m e t i m e s r e q u i r e v a l i d a t i o n by

bioassay or receptorassay before conclusive statements p h y s i o l o g i c a l or p a t h o p h y s i o l o g i c a l

issues can be m a d e .

on

166 REFERENCES 1.

R i n d e r k n e c h t , E., H u m b e l , R . E . : J. B i o l . C h e m . 2 7 6 8 - 2 7 7 6 (1978)

253,

2.

Svoboda, M.E., van Wyk, J.J., Klapper, D.G., Fellows, R.E., Grissom, F.E., Schlueter, R.J.: B i o c h e m i s t r y J_9, 790-797 ( 1980)

3.

B a l a , R . M . , B h a u m i c k , B.: C a n . J. 57: 1 2 8 9 - 1 2 9 8 (1979)

4.

R u b i n , J . S . , M a r i z , I., J a c o b s , J . W . , D a u g h a d a y , B r a d s h a w , R.: E n d o c r i n o l o g y 110, 7 3 4 - 7 4 0 (1982)

5.

Rinderknecht, E., Humbel, R.E: FEBS-Letters 283-286 (1978)

6.

R e c h l e r , M . M . , N i s s l e y , S . P . , K i n g , G . L . , et al.: J. S u p r a m o l . S t r u c t , and C e l l B i o c h e m . J_5, 253-286 (1981 )

7.

F r y k l u n d , L., U t h n e , K . , S i e v e r t s s o n , H . : Biophys. Res. Comm. 9 5 7 - 9 6 2 (1974)

8.

Z a p f , J . , W a l t e r , H . , F r o e s c h , E . R . : J. I n v e s t . 68, 1321-1330 (1981)

9.

Posner, B.I., Guyda, H.J., Corvol, M.T., Rappaport, R., H a r l e y , C . , G o l d s t e i n , S.: J. C l i n . E n d o c r i n o l . M e t a b . 47, 1240-1250 (1978)

Biochem: W.H.,

89:

Biochem.

Clin.

10. M o s e s , A . C . , N i s s l e y , S . P . , S h o r t , P . A . , et al.: N a t l . A c a d . Sei (USA) 77, 3 6 4 9 - 3 6 5 3 (1980)

Proc.

11. M a r q u a r d t , H . , T o d a r o , E . J . , H e n d e r s o n , L . E . , O r o s z l a n , S.: J. B i o l . C h e m . 256, 6 8 5 9 - 6 8 6 5 (1981) 12. M o s e s , A . C . , N i s s l e y , S . P . , S h o r t , P . A . , R e c h l e r , M . M . , P o d s k a l n y , J . M . : Eur. J. B i o c h e m . 103, 387-400 (1980) 13. S a l m o n , W . D . jr, D a u g h a d a y , W . H . : J. Lab. C l i n . £ 9 , 825-836 (1957)

Med.

14. F r o e s c h , E . R . , B ü r g i , H . , R a m s e i e r , E . B . , B a l l y , P . , L a b h a r t , A . : J. C l i n . I n v e s t , 42, 1816-1834 (1963) 15. R o d b e l l , M. : J. B i o l . C h e m . 239 , 3 7 5 - 3 8 0

( 1964)

16. Z a p f , J . , S c h o e n l e , E., F r o e s c h , E . R . : E u r . J. 87, 2 8 5 - 2 9 6 (1978)

Biochem.

17. P i e r s o n , R . W . , T e m i n , H . M . : J. C e l l . P h y s i o l . 79, 330 (1972) 18. T h i e r o t - P r e v o s t , G . , S c h i m p f f , R. : A c t a E n d o c r . 98, 3 5 8 - 3 6 3 (1981) 19. G r a n t , D . B . : C l i n . E n d o c r i n o l .

1, 3 8 7 - 3 9 8

(1972)

319-

(Kbh.)

167 20. A u d h y a , T . K . , G i b s o n , K . D . : P r o c . N a t l . A c a d . (USA) 72, 604 (1975)

Sei.

21. Z a p f , J . , J a g a r s , G . , S a n d , I., F r o e s c h , E . R . : L e t t e r s 90: 135-140 (1978)

FEBS-

22. J a k o b , A . , H a u r i , C h . , F r o e s c h , E . R . : J. C l i n . £ 7 , 2 6 7 8 - 2 6 8 8 (1968)

Invest.

23. P o f f e n b a r g e r , P . L . : J. C l i n . I n v e s t . 56, (1975)

1455-1463

24. Z a p f , J . , S c h o e n l e , E . , J a g a r s , G . , S a n d , I., G r u n w a l d , J . , F r o e s c h , E.R. : J. C l i n . I n v e s t . 6_3, 10771084 (1979) 25. D a u g h a d a y , W . H . , M a r i z , I.K. B l e t h e n , S . L . : J. C l i n . E n d o c r i n o l . M e t a b . 5J_, 7 8 1 - 7 8 8 (1980) 26. H o r n e r , J . M . , L i u , F., H i n t z , R . L . : J. C l i n . M e t a b . £ 7 , 1 2 8 7 - 1 2 9 5 (1978)

Endocrinol.

27. D a u g h a d a y , W . H . , P a r k e r , K . A . , B o r o w s k i , S . , T r i v e d i , B . , K a p a d i a , M . : E n d o c r i n o l o g y 110, 575-581 (1982) 28. M a r s h a l l , R . N . , U n d e r w o o d , L . E . , V o i n a , S . J . , F o u s h e e , D . B . , v a n W y k , J . J . : J. C l i n . E n d o c r i n o l . M e t a b . 39, (1974) 29. H a l l , K . , T a k a n o , K . , F r y k l u n d , L.: J. C l i n . M e t a b . 39_, 9 7 3 - 9 7 6

Endocrinol.

30. V a n W y k , J . J . , S v o b o d a , M . E . , U n d e r w o o d , L . E . : J. C l i n . E n d o c r i n o l . M e t a b . J50, 2 0 6 - 2 0 8 ( 1980) 31. R e c h l e r , M . M . , F r y k l u n d , L., N i s s l e y , S . P . , H a l l , K., P o d s k a l n y , J . M . , S k o t t n e r , A . , M o s e s , A . C . : E u r . J. B i o c h e m . 82, 5 - 1 2 (1978) 32. D a u g h a d a y , W . H . , T r i v e d i , B . , K a p a d i a , M . : J. E n d o c r i n o l . M e t a b . 53, 289-294 (1981)

Clin.

33. R e c h l e r , M . M . , Z a p f , J . , N i s s l e y , S . P . , F r o e s c h , E . R . , Moses, A.C., Podskalny, J.M., Schilling, E.E., H u m b e l , R . E . : E n d o c r i n o l o g y 107, 1451-1459 (1980) 34. Z a p f , J . , W a l d v o g e l , M . , F r o e s c h , E . R . : A r c h . B i o p h y s JM58, 6 3 8 - 6 4 5 (1975)

Biochem.

35. S c h a l c h , D . S . , H e i n r i c h , U . E . , K o c h , J . G . , J o h n s o n , C . J . S c h l u e t e r , R . J . : J. C l i n . E n d o c r i n o l . M e t a b . 46^, 6 6 4 671 (1978) 36. R i e u , M . , G i r a r d , F., B r i c a i r e , H . , B i n o u x , M . : J. C l i n . E n d o c r i n o l . M e t a b . ^ 5 , 147-153 (1982) 37. R e b e r , K . , L i s k e , R. : H o r m o n e R e s . 1_, 2 0 1 - 2 1 3

( 1976)

38. Z a p f , J . , M o r e l l , B . , W a l t e r , H . , L a r o n , Z., F r o e s c h , E . R : A c t a E n d o c r . (Kbh.) 95, 5 0 5 - 5 1 7

(1980)

168 39. F u r l a n e t t o , R . W . , U n d e r w o o d , L . E . , v a n W y k , J . J . , D ' E r c o l e , A . J . : J. C l i n . I n v e s t . 60^ 6 4 8 - 6 5 7 (1977) 40. H a l l , K . , B r a n d t , J . , E n b e r g , G . , F r y k l u n d , L.: J. C l i n . E n d o c r i n o l . M e t a b . 4j?, 271-278 (1979) 41. B a l a , R . M . , B h a u m i c k , B.: J. C l i n . E n d o c r i n o l . 49, 7 7 0 - 7 7 7

Metab.

42. H i n t z , R . L . L i u , F., M a r s h a l l , L . B . , C h a n g , D.: J. C l i n . E n d o c r i n o l . M e t a b . 50, 4 0 5 - 4 0 7 (1980) 43. W i d m e r , U . , Z a p f , J . , F r o e s c h , E . R . : J. C l i n . E n d o c r i n o l . M e t a b 55, 8 3 3 - 8 3 9 (1982) 44. H i n t z , R . L . , L i u , F., R i n d e r k n e c h t , E.: J. Endocrinol. Metab. 6 7 2 - 6 7 3 ( 1980)

Clin.

45. H i n t z , R . L . , L i u , F.: J. C l i n . E n d o c r i n o l . 54, 4 4 2 - 4 4 6 (1982)

Metab.

S u p p o r t e d b y g r a n t N o . 3.167-0.81 National Science

Foundation.

of the

Swiss

THE USE OF SYNTHETIC PEPTIDES FOR THE DEVELOPMENT OF RADIOIMMUNOASSAYS FOR THE INSULIN-LIKE GROWTH FACTORS

R.L. Hintz and F. Liu Stanford University Medical Center Stanford, CA, USA

D. Chang Peninsula Laboratories Belmont, CA, USA

E.R. Rinderknecht Genentech, Incorporated So. San Francisco, CA, USA

Introduction

The insulin-like growth factors (IGF) or somatomedins are a group of peptides found in plasma which share both physiological and structural characteristics. These peptides have their historical roots in three different sets of observations. First, it was discovered that the direct in vitro action of growth hormone on cartilage could not explain the observed in vivo activities. Instead there was a factor or factors under growth hormone control in the plasma which appeared to have direct action on sulfate uptake by cartilage. This factor was originally entitled sulfation factor (1). Later the name somatomedin was proposed when it became obvious that this factor had many actions other than the increase in sulfate uptake by cartilage (2).

Insulin-Like Growth Factors / Somatomedins © 1983 Walter d e Gruyter & Co., Berlin • N e w York

170

The second line of research leading to the discovery of these IGF peptides was the observation that there was much more insulin-like activity measureable in plasma by bioassay than there was by radioimmunoassay. This biological activity could not be supressed by anti-insulin antibodies. Thus the name of nonsuppressible insulin-like activity (NSILA) was proposed for this phenomenon (3). Similar to sulfation factor, this original name was later abandoned when more information became available and these peptides were renamed the insulin growth factors (IGF) (4). A third line of observation was the experiments conducted on tissue culture in which it was shown that a plasma factor was necessary for the growth of chick fibroblast in culture. This factor was originally isolated from bovine fetal serum, and was entitled multiplication stimulating activity (MSA) (5). The MSA has more recently been isolated from the supernatant of buffalo rat liver cultures, which were found to make substances very similar to that found in fetal calf serum (6). It is now clear that these three sets of biologically active peptides, the somatomedins, the IGF's and the MSA's are all related both biologically and structurally. They all are strongly insulin-like in biological action, and stimulate cell growth and division in many in vitro systems. In addition they all appear to be roughly 5 to 10,000 molecular weight when in their purified form, but to be bound to larger plasma proteins in the natural state. Finally they all show some degree of growth hormone dependence. Three of these peptides have been sequenced. IGF-I was the first one isolated in pure form and sequenced (7). The structure of IGF-I has strong homologies to human pro-insulin and it was on the basis of this data that the proposed name changed to insulin-like growth factor or IGF was proposed. Soon after the isolation and sequencing of IGF-I the sequence for IGF-II became available (8). This sequence shows strong homolgy to both IGF-I and human proinsulin. In the past year Marquardt, working in Tadoro's lab at NIH, has isolated one of the buffalo rat liver MSA peptides and sequenced it (9). This MSA peptide is only 5 amino acids different from the sequence of human IGF-II. Because the IGF peptides have had to be purified from dilute sources such as plasma or tissue culture media, the amount of pure mateial that has been available for the development and performance of radioimmunoassays has been very constrained.In this paper we describe our efforts to solve the severe shortage of purified IGF peptides by the synthesis of peptide fragments of the segments of the IGF-I and IGF-II sequences. By making antisera specific against

171

these synthetic segments, we have been able to develop several immunoassays that are specific for IGF-I or IGF-II. In some instances it has allowed us to avoid the need for pure IGF-I or IGF-II even as radioligand and standards.

Materials and Methods

Synthetic IGF-I peptide regions 24-41, 30-41 (C region), 63-70 (D region) and 66-70 were synthesized by the solid phase technique (10). Their purity was checked by amino acid analysis and end group analysis. In addition, IGF-II region 57-66 (C region) was also synthsized by solid phase technique. Somatomedin-C was prepared by our previously described techniques (11), and was a gift from Dr. J.J. van Wyk. Pure IGF-I and IGF-II standards were a gift from Dr. Rene Humbel of Zurich, Switzerland. Iodination of peptides was performed by the chloraminine T method (12). The peptide fragments were coupled to bovine thyroglobulin by the carbodiimide method (13). The usual ratio of incorporation was such that an average of 60 molecules of IGF-I or IGF-II peptide fragments were incorporated per molecule of bovine thyroglobulin. These conjugates were emulsified in complete Freund's adjuvant and injected in multiple subcutaneous sites in to young male rabbits. The usual dose of antigen was the equivalent of 500 micrograms of pure peptide fragment per rabbit. Boosters doses were given every 4 to 6 weeks. Bleedings from each rabbit were screened for their ability to bind the iodinated peptide fragment injected and their ability to bind iodinated IGF-I or IGF-II. Antisera were selected on their basis to bind radioligand for further characterization. All radioimmunoassays were conducted in a pH 7.4 0.04 M phosphate buffer with 0.5% BSA and 0.15 M NaCl. Incubation was at 4 C overnight. The separation bounded free was carried out by polyethyleneglycol method (14).

Results

W e have developed successful radioimmunoassays

against

172

several of the synthetic IGF peptide fragments. We have previously published our data utilizing the IGF-I C-peptide segment (15), the IGF-I D-peptide segments (16), and the IGF-II peptide segments (17). The sensitivity and specificity of these immunoassays is summarized in Table I. For comparison the placental membrane radioreceptorassay is included. Table I Assay

Semsitivity (half displ.)

Specificity (IGF-I/IGF-II)

IGF-I C region

60 ng/ml

>1000/1

IGF-I D region

80 ng/ml

non-parallel

IGF-II C region

50 ng/ml

0 A • AB

•

• •

0.4

0.0

HEPARINPLASMA

CITRATEPLASMA

EDTAPLASMA

SERUM

Fig.3. Comparison of IRSM concentrations in similar aliquots of plasma and serum obtained at the same time from the same ten normal subjects. Results are shown in the same symbols for each subject along with mean ± S.E.M. for each group. All samples were acidified and lyophilized prior to RIA. study (15).

(In other studies comparing different heparin

preparations the differences persisted but were less than 2fold).

Addition of citrate or EDTA to the serum samples, in

amounts identical to those present in the corresponding plasma, did not change the measured amount of IRSM.

Whereas

addition of heparin to serum resulted in measurement of apparent amounts of IRSM similar to that in heparinized plasma.

The apparent amounts of IRSM measured in heparinized

plasma, in contrast to serum, were approximately similar with or without acidification and lyophilization prior to RIA. Addition of heparin in buffer alone did not affect the binding of radiolabeled SM by the SM antiserum.

To further study the

effects of heparin on this RIA similar aliquots of heparinized plasma and serum obtained at the same time from the same subjects were fractionated by gel filtration under neutral or acidic conditions as shown in Fig. 4.

After acidic gel

185 1.2

a. PLASMA - NEUTRAL CHROMATOGRAPHY

b. PLASMA-ACIDIC CHROMATOGRAPHY

1.0 0.8

0.6

c lo -

0.4

§ 0.2

• •11,000• I

I 0.0 g

e

•0,000

•TtT

1.0

c.

SERUM-NEUTRAL CHROMATOGRAPHY

• • • •

•0,000 IS £00 J0.000 «,000 • • • •

d. SERUM - ACIDIC CHROMATOGRAPHY

0.8 0.6

0.4

STORTING a O 02 0 4 0.6 0 8 1.0 MATERIAL ^ „ j q ^

STARTING 0 0 0 2 0 4 0.6 OA 1.0 MATERIAL ^ | N J E R V A L S

Fig.4. Amounts of immunoreactive somatomedin in pooled heparin-plasma (a and b) and serum (c and d), derived from the same blood samples, before and after gel filtration on columns of Sephadex G-75 with neutral (0.05 M sodium phosphate, pH 7.5) or acidic (1% formic acid, pH 2.3) eluant. Amounts of immunoreactive somatomedin in starting materials (heparin-plasma and serum) were measured without or with acidification prior to radioimmunoassay. After gel filtration the fractions eluted under acidic conditions were lyophilized and reconstituted with the assay buffer, whereas the fractions eluted under neutral conditions were not altered prior to the assay. (Reproduced with permission from Clin.Invest.Med. 5:60, 1982) . filtration of serum (Fig. 4(d)) the amount of total IRSM activity, recovered in the smaller molecular weight fractions, was similar to that measured in the acidified starting sample. After neutral gel filtration of serum, as shown in Fig. 4(c), a lesser amount of IRSM was measured, in the starting serum and eluted fraction after neutral gel filtration without

186

acidification of samples prior to RIA.

In comparison, a

greater amount of IRSM was measured in a similar aliquot of heparinized plasma before and after neutral gel filtration (Fig. 4(a)).

In contrast, as shown in Fig. 4(b) after acidic

gel filtration of heparinized plasma the amounts of recovered IRSM in the eluted fractions was essentially identical to that recovered after acidic chromatography of serum but much less than that measured in the heparinized starting plasma. These results suggest that heparin bound to plasma proteins, but not heparin alone, artefactually affects the B-SM RIA in measurement of greater apparent amounts of IRSM.

Similar

artefactual effects of heparin have been reported in measurement of insulin (16).

The artefactual effects of heparin on

the B-SM RIA can be eliminated by acidic gel filtration of heparinized plasma prior to assay. The serum concentrations of immunoreactive SM-C, IGF-I, and B-SM are greater than normal in patients with acromegaly and very much le.ss than normal in patients with hypopituitarism or growth hormone deficiency (4-6).

Serum concentrations of

immunoreactive IGF-II are not significantly greater than normal in patients with acromegaly and are less subnormal than IGF-I in patients with hypopituitarism (6).

As shown

in Fig. 5, the low levels of B-SM in serum of patients with GH deficiency are increased after hGH injections in a dose responsive manner.

Zapf et al (6) reported that serum levels

IGF-II increased after hGH treatment of patients with GH deficiency whereas Daughaday et al (17), using an IGF-II radioreceptor assay, reported that most GH deficient patients did not show an increase in IGF-II levels in serum after hGH treatment.

These overall observations suggest that the Type

I SM (basic group of SM) are highly GH dependent whereas Type II SM (neutral, IGF-II, group of SM) have less GH dependence and may be significantly more dependent on other factors in regulation of their concentrations in serum.

187 I N I T I A L IRSM

RESPONSE TO

hGH

Fig.5. Mean increments in serum IRSM in patients with GH deficiency at various time intervals after intramuscular injection of various doses of hGH. Number of patients in each group shown in brackets. Prior to more detailed clinical investigative studies in patients with GH deficiency we measured the concentrations of immunoreactive B-SM in a large number of normal males and females at different ages from birth to adults (18).

As

shown in Fig. 6 the serum levels of B-SM in cord blood were approximately one-third of normal adult levels.

These levels

further dropped to very low levels within several days after birth and then gradually increased towards normal near the age of puberty in both males and females.

During the ages

usually corresponding to puberty the mean serum levels of B-SM were approximately two-fold higher than that for normal adults.

Similar findings were reported for SM-A (19) and

IGF-I (6).

These observations suggest that clinical investi-

gative studies based on measurement of serum levels of the basic group of SM require careful comparison with age matched normal controls.

The similar concentrations of B-SM in

venous and arterial cord serum is suggestive evidence against placental production of B-SM or transfer of B-SM from the

188

AGE OF SUBJECTS (years)

Fig.6. Comparison of IRSM levels in normal male and female newborns, children at various ages, and adults. Sera were obtained from the same newborns at da Livery (CAS and CVS) and again 1 and 3 days after birth. The number of subjects in each group is indicated in parenteses. *,Significant difference between the mean levels of IRSM in the sera of males and females in each age group, as determined by one-way analysis of variance (P^0.001), followed by Duncan's multiple range test (P< 0.05) . (Reproduced with permission from J.Clin.Endocrinol. Metab. 52:508, 1981). maternal to the fetal circulation.

However, the significant

drop of B-SM in serum after birth might suggest that the placenta was producing some factors that stimulated fetal production of B-SM.

The low levels of B-SM in serum of

young children, when growth rates are maximal, requires

189

further explanation if the basic group of SM are important skeletal growth factors.

However, the measurement of total

B-SM in serum, the majority of which is bound to serum proteins , rather than free or non-protein bound SM in serum or interstitial fluid, presents a significant handicap in further more concise hypotheses regarding the biological role of B-SM in serum.

Serum levels of immunoreactive IGF-II in normal

humans have been reported to be approximately half-normal during the first one year of life and are relatively constant thereafter (6).

These observations might suggest that meas-

urements of IGF-II rather than IGF-I would be more useful in diagnosis of GH deficiency particularly if it was confirmed that IGF-II levels in serum in these patients were highly GH dependent. We have carried out further studies in GH deficiency patients to determine whether serum SM levels were diagnostic of GH deficiency and whether the serum SM response could be used as a predictor of growth rate response to hGH therapy (20). Serum samples were obtained before and during hGH therapy of 177 GH deficiency patients as part of the Medical Research Council of Canada collaborative study. assayed for B-SM by radioimmunoassay

Serum samples were

(R.M.Bala) and for ILAs

(an IGF-II-like SM)(10) by radioreceptor assay (H.Guyda). As shown in Fig. 7 the pre-treatment mean concentrations of serum B-SM were lower than age matched normal control subjects.

The overall mean (± S.D.) pre-treatment IRSM con-

centration in serum was 0.21 ± 0.30 U/ml compared to 1.0 U/ml B-SM activity designated for normal adult reference serum. When compared to age matched normal control subjects approximately one-sixth of all children with GH deficiency short stature had serum levels of B-SM which were in the normal range.

This observation significantly limits the usefulness

of RIA measurement of serum concentrations of B-SM in diagnosis of GH deficiency particularly in children less than 8 years of age.

Serum concentrations of B-SM, 24-48 hr after

190

CHRONOLOGICAL AGE (YR)

Fig. 7. The relationship between IRSM levels (mean ± SEM) and age in GH deficient children, before and during hGH therapy. The shaded area shows the smoothed data, mean ± SD, for normals from Fig.6. N for each group is indicated in parenthesis. (Reproduced with permission from J.Clin.Endocrinol.Metab. In press). hGH injection, obtained after 1-6 months of treatment with 2 units of hGH intracmuscularly three times weekly were increased above basal pre-treatment levels in approximately two-thirds of these GH deficient patients.

As shown in Fig.

7, however, the mean B-SM levels in serum were lower than that of the normal age matched controls.

The mean increments

in serum IRSM levels after hGH therapy was neligible in GH deficient patients less than 10 years of age.

The mean lev-

els of serum B-SM in GH deficient patients were correlated

191

with chronologic age before and after hGH therapy.

A similar

positive correlation existed between bone age and pre- and post-treatment serum levels of B-SM.

Rosenfeld at al (21)

have reported a similar correlation of post-hGH treatment concentrations of serum SM-C and bone age.

These investiga-

tors did not find a similar correlation before therapy which might be explained by lack of sufficient numbers of patients to achieve statistical significance.

These age related

changes of serum B-SM levels may suggest an underlying nonGH dependent developmental pattern of SM synthesis, metabolism, or protein binding rather than a lesser degree of GH deficiency in the older patients in these studies.

The mean pre-

treatment levels of ILAs in serum of these patients

(0.39 ±

0.25 U/ml) were less subnormal than the B-SM levels compared to 1.0 U/ml designated for normal adult reference serum. Somewhat similar to the results for B-SM, even though the mean levels of ILAs in serum of GH deficient patients were lower than those in age matched normal controls approximately one-third of the ILAs levels, compared to one-sixth of the B-SM levels, in GH deficient patients were in the normal range. This would suggest that there is no overall advantage in measurements of serum concentrations of IGF-II rather than IGF-I in diagnosis of GH deficiency.

After hGH treatment, the

levels of ILAs in serum were increased in approximately fourfifths of these patients.

Serum levels of ILAs showed a

similar, even though less significant, age relationship with chronologic and bone age in GH deficient patients noted for B-SM before but not after treatment.

In children less than

8-10 years of age the relative increments in serum concentrations of ILAs were greater than those noted for B-SM. These observations might suggest that patients with GH deficiency have a delayed theoretical transition in predominance of IGF secretion from the Type II to the Type I SM which has been postulated to occur during the late fetalinfancy age in normals

(22).

192

Even though our results showed an increase in serum concentrations of B-SM or ILAs with an increae in height velocity in the majority of GH deficient children treated with hGH, other combinations occurred including increased SM with decreased height velocity (HV), decreased SM with increased HV and decreased SM with a decreased HV.

There was no over-

all correlation between the change in growth velocity and the change in serum concentrations of B-SM or ILAs after therapy.

Normalized growth velocity for age, using SDS,

similarly did not show correlation with serum SM levels. Rosenfeld et al (21) reported a similar lack of correlation between growth rate and serum SM-C levels after GH therapy. We are not able to reconcile these observations with those of Rudman et al (23) who showed a high degree of correlation between growth velocity and serum SM-C levels in normals and during GH therapy of GH deficient children. The overlap of serum SM levels in GH deficient patients imto the range of age matched normal controls decreases the dianostic usefulness of serum SM measurements in diagnosis of GH deficiency in an individual patient with short stature. Since growth hormone therapy does increase growth rate and serum SM levels in the majority of GH deficient patients, the lack of significant correlation of serum SM levels with growth rate before or after GH therapy requires further explanation.

It is possible that there would be a greater

correlation between growth rate and levels of free or nonprotein bound SM in serum or in interstitial fluid.

Other-

wise these results might significantly question the hypothesis that SM is the principle mediator of growth hormone action in stimulating skeletal growth.

193

Summary Despite minor structural differences it would appear that the basic group of SM (Type I), which include IGF-I, SM-C and B-SM may be considered functionally identical.

Radioimmuno-

assays based on any one of these basic SM can be presumed to measure the same SM in serum.

Accurate measurements of SM

concentrations in serum require separation of SM from the serum binding proteins by acidic gel filtration or acidethanol extraction prior to RIA.

Availability of a highly

potent SM antiserum enables RIA measurements of total immunoreactive basic somatomedin in serum which has been acidified and lyophilized prior to RIA.

Heparin in plasma may arte-

factually increase measured amounts of B-SM.

Serum con-

centrations of B-SM vary with age in normal subjects and growth hormone deficient patients.

Measurements of Type I

or II SM concentrations in serum may not be diagnostic of growth hormone deficiency in individual young children. Growth hormone therapy increases growth rate and serum concentrations of Type I and II SM in the majority of patients. The lack of significant correlations of growth rate and serum SM levels, in GH deficient patients before and after therapy, noted in our studies and those reported by others, requires further explanation if SM is the principle mediator of growth hormone action on skeletal growth.

References 1.

Marshall, R.N., Underwood, L.E., Voina, S.J., Foushee, D.B., Van Wyk, J.J.: J. Clin. Endocrinol. Metab. _39, 283 (1974) .

2.

Hintz, R.L., Liu, F.: J. Clin. Endocrinol. Metab. 45, 988 (1977) .

3.

Zapf, J., Kaufmann, U., Eigenmann, E.J., Froesch, E.R.: Clin. Chem. 47, 677 (1977).

194

4.

Furlanetto, R.W., Underwood, L.E., Van Wyk, J.J., D'Ercole, A.J.: J. Clin. Invest. 60, 648 (1977).

5.

Bala, R.M., Bhaumick, B.L J. Clin. Endocrinol. Metab. j49, 770 (1979).

6.

Zapf, J., Walter, H., Froesch, E.R.: J. Clin. Invest. 68, 1321 (1981).

7.

Hall, K., Brandt, J., Enberg, G., Fryklund, L.: J. Clin. Endocrinol. Metab. 48, 271 (1979).

8.

Daughaday, W.H., Hall, K., Raben, M., Salmon, W.D. Jr., Van Den Brande, J.L., Van Wyk, J.J.: Nature (Lond) 235, 107 (1972). Bala, R.M., Bhaumick, B.: Can. J. Biochem. 57, 1289 (1979)

9. 10.

Posner, B.I., Guyda, H.J., Corval, M.T., Rappaport, R., Harley, C., Goldstein, S.: J. Clin. Endocrinol. Metab. 47, 1240 (1978).

11.

Horner, J.M., Liu, F., Hintz, R.L.: J. Clin. Endocrinol. Metab. 47, 1287 (1978).

12.

Daughaday, W.H., Mariz, I.K., Blethen, S.L.: J. Clin. Endocrinol. Metab. 51, 781 (1980) .

13.

Furlanetto, R.W.: J. Clin. Endocrinol. Metab. 51, 12 (1980)

14.

Blethen, S.L., Van Wyk, J.J., Underwood, L.E., Copeland, K.C., Chatelain, P.G., Chapin, D.C.: Program of the 61st Annual Meeting of the Endocrine Society, 220 (1979).

15. 16.

Bhaumick, B., Bala, R.M.: Clin. Invest. Med., 5, 57 (1982) Henderson, J.R.: Lancet 2, 545 (1970).

17.

Daughaday, W.H., Brivedi, B., Kapadia, M.: J. Clin. Endocrinol. Metab. 53, 289 (1981).

18.

Bala, R.M., Lopatka, J., Leung, A., McCoy, E., McArthur, R.G.: J. Clin. Endocrinol. Metab. 52, 508 (1981).

19.

Hall, K., Enberg, G., Ritzen, M., Svan, H., Fryklund, L., Takano, K.: Acta Endocrinol. 94, 155 (1980). Dean, H.J., Kellett, J.G., Bala, R.M., Guyda, H.J., Bhaumick, B., Posner, B.I., Friesen, H.G.: Clin. Invest. Med. 4, 13B (1981) and J. Clin. Endocrinol. Metab. (In press).

20.

21.

Rosenfeld, R.G., Kemp, S.F., Hintz, R.L.: J. Clin. Endocrinol. Metab. 53, 611 (1981).

22.

d'Ercole, J.A., Wilson, D.F., Underwood, L.E.: J. Clin. Endocrinol. Metab. 5J., 674 (1980).

23.

Rudman, D., Moffitt, S., Fernhoff, P.M., McKenzie, W.J., Kenny, J.M., Bain, R.P.: J. Clin. Endocrinol. Metab. 52, 622 (1981).

Regulation of Plasma Levels of Insulin-Like Growth Factor Levels

NONGROWTH

HORMONE

SOMATOMEDIN

DEPENDENT

REGULATION

OF

PLASMA

LEVELS

Richard W.

Furlanetto,

Department

of E n d o c r i n o l o g y

Hospital

HORMONAL

of

Ph.D.,

Philadelphia

Pennsylvania,

& Metabolism,

and

Philadelphia,

M.D.

Children's

the U n i v e r s i t y

of

PA.

INTRODUCTION

A number

of

addition

to g r o w t h h o r m o n e ,

steroids

are all

now well

established

hormones

important

the e f f e c t s

of g r o w t h

these

hormones

other

One m e c h a n i s m plasma

these of

a number

hormones

these

this

thyroxine, for

the

normal

regulate they

of

form

growth

somatomedin

that

changes

have

of

reference

levels.

are

While

mediate by

It is n o t

will

it

is

many

of

understood. regulating

surprising

the

levels.

this

sex

which

is by

examined

In

and the

poorly

function

somatomedin

the b a s i s

chapter

growth.

the m e c h a n i s m s

could

studies

growth.

Cortisol

somatomedins

concentrations.

in plasma such

to i n f l u e n c e

hormone,

on plasma

studies

Throughout

that

by w h i c h

somatomedin

then that

are k n o w n

effect

The

of

results

chapter.

be m a d e

to

changes

It is i m p o r t a n t

to

realize

do n o t n e c e s s a r i l y

Insulin-Like Growth Factors / Somatomedins © 1983 Walter d e Gruyter& Co., Berlin • N e w York

require

a change

in

198 the

rate

of

somatomedin

synthesis.

the m e t a b o l i c

clearance

concentration

or m o l e c u l a r

protein

could

also result

levels.

With

few

exceptions,

the

hormones

regulate

somatomedin

levels

Prolactin.

The

regulation

of

has

effects

of s t r u c t u r a l l y The

structural by

lead

homology

a number

that

human

finding activity

are

Clemmons

et

that

al

hormone

suggesting

that

activity.

However,

findings

for of

normal

they

found

growth

and Friesen

which

the

the

somatomedin with been

a

close

reported

agreement

somatogenic on

plasma

the

somatomedin

hyperprolactinemia.

controlled

study,

in p a t i e n t s w i t h

found growth

hyperprolactinemia,

normal

hormone

suggesting

by

study

is b a s e d

total

with

binding

Sm-C/IGF-I

is g e n e r a l

does have weak

this hormone.

Carr

and

normal

extreme

prolactin

hyperprolactinemia, potential

and

to

on

have

if a n y ,

in a carefully

levels were

deficiency

patients with

levels

in

somatomedin

in

a hormone

conclusion

in p a t i e n t s

(3).

Sm-C/IGF-I

little,

This

Sm-C/IGF-I normal

hormones

There

in

unknown.

hormone

hormone,

of l a b o r a t o r i e s .

that

is

investigators

to g r o w t h

(1,2,3,^)

somatomedin

particularly

prolactin,

prolactin has

activity.

of

changes

in plasma

of g r o w t h

related

of

the

or

exact mechanism

levels,

a number

effects

of

in c h a n g e s

somatomedin

alterations

of s o m a t o m e d i n

form

importance

levels,

levels.

rate

Indeed,

at

These

somatogenic

Sm-C

levels

secretion

best a minor results

(5) w h i c h

show

in

and somatogenic

agree with that

human

the

199 prolactin liver Sm-A

is a v e r y

growth hormone and

IGF-II

Placental another

elevated

correlate

et al

has

range

hP1

the rare

hormone

Sm-C/IGF-I

with

and

during

the

third

trimester

rise

this

While

to o c c u r . also

on

(hP1)

homology

is

to

growth

of p r e g n a n c y

and

to the

(6)

suggestion

Recently

to s t u d y In this

necessary

Merimee

pregnant,

patient

into

thereby

a

that

the

both

normal

establishing

that

for

the

third earlier

The

authors,

noting

the

speculated

t h a t hP1

may

mediating

be

increase.

the

studies

cited

in the r e g u l a t i o n conclusive changes

studied

since

occur

evidence

using

prolactin

are

increased

is not

human

levels

led

patient.

trimester,

growth hormone

(6),

has

activity.

IGF-II levels

lactogen

Sm-C/IGF-I

trimester

of

to

reported.

structural

opportunity

pituitary

findings

placental

levels

deficient

binding

The e f f e c t s

that

third

somatogenic

(7) h a d

growth

the

for

not b e e n

a close

observation

during

t h a t hP1

have

Human

hormone with The

competitor

receptor.

levels

Lactogen.

hormone.

they

weak

for the

of

a large

during

number

lactogen was

rats.

such

a role

studies

are

and

of h P 1 ,

equipotent

found

Hurley

with

that

bovine

hP1

not

direct

lactogen on Sm-C They

for

metabolic

To o b t a i n m o r e

effect

of p l a c e n t a l

suggest

of h o r m o n a l

pregnancy.

a somatogenic effect

(6,7)

somatomedin,

hypophysectomized

placental

above

et al levels

ovine growth

(8)

200 hormone

in s t i m u l a t i n g

placental with

lactogen was

the f i n d i n g s

oP1

is e q u i p o t e n t

hGH

receptor

as h i g h

balance

with

and Blizzard

as 200 m g / d ,

suggests

somatomedin

levels.

Sm-C/IGF-I which pregnancy

that

may

did

(9)

humans. hP1

during

be t h e r e s u l t

below,

the

of

nitrogen

evidence

plasma

the rise

third

in

trimester

of i n c r e a s e d

liver

hP1 , i n

positive

not r e g u l a t e

As d i s c u s s e d

occurs

that

The w e i g h t

does

that

competitor.

found

not induce

agree

showed

to the h u m a n

a very weak

human

results

(5) w h i c h

binding

is only

in hypopituitary

therefore

for

system but

These

and Friesen

hGH

hP1

Schultz

in this

not effective.

of C a r r

while

Furthermore, doses

Sm-C/IGF-I

of

progesterone

secretion.

Thyroid are

Hormone.

dramatic.

investigating somatomedin (12,13). (10,11)

This has

levels

of

(12)

levels

hypothyroidism

appears

to b e

the r e s u l t

thryoid

hormone.

of

In part,

decrease

in growth hormone

However,

thyroid

effects

hormone

synthesis

of g r o w t h

by

hormone

and

hormone of

hormone

that

two this

I).

decrease in

directly

the liver

growth

plasma

synthesis Sm-C/IGF-I

reduced

This

independent

secretion

also

on

both

are moderately (Table

on

studies

somatomedin

agreement

patients with

somatomedin

thyroid

thyroid

(10,11,12)

is g e n e r a l

Sm-A

of

l e d to a n u m b e r

the e f f e c t s

There and

The e f f e c t s

in

decrease

effects

is d u e

to

of a

hypothyroidism. stimulates

(13) a n d

on somatomedin

potentiates

synthesis

(12).

the

201

TABLE I SOMATOMEDIN LEVELS IN HYPOTHYROIDISM REFERENCE

SOMATOMEDIN LEVEL (NORMAL)

HORMDNE Sm-C/IGF-I

35

Furlanetto et al (10)

Sm-C/IGF-I

67

Baxter et al (11)

Sm-A

39

Takano et al (12)

IGF-II

45

Baxter et al (11)

Nonetheless does

this

moderate

not e x p l a i n

deficiency effect

of

gr o w t h .

thyroid

ho n o n e

somatomedins

but

metabolism.

In

hormone

on

Froesch

et al

biological

levels

is the r e s u l t

levels

are

(12). the

normal

Although

t h e d a t a of B a x t e r also

low

of

e t al

a l t e r ed

major

are

thyroxine (11)

the

thyroid

as reported

by

important

deficiency.

in p a t i e n t s there

of

the more

h o r m o ne

by

cellular

synthesis,

in patients w i t h

hormone

is n o t m e d i a t e d

to be

thyroid

effect

the

direct effect

the

seem

c o n s e q u e n ce of

measuring

are

of

levels

thyroid

implies that

on growth

(14), w o u l d

hyperthyroidism

levels,

This

this regard

th a t

effect

cartilage proteoglycan

Somatomedin-A

directly

the d r a m a t i c

has on

in s o m a t o m e d i n

decrease

with

no

studies

on

suggests

IGF-II that

hypothyroidism.

IGF-II

202 Glucocorticoids.

Glucocorticoid

growth

and a number

impairment

effects While

of g l u c o c o r t i c o i d s

initial

results

studies

(17,18)

techniques

have

levels

normal

are

It w o u l d this

appear

disorder

levels

but

then that

is

the

to a d i r e c t

cellular

metabolism.

Such

observed

in bioassay

systems

available in

on the

that

and

failure

changes

effect

a direct

in

Sm-A

(15)

excess.

observed

in

somatomedin

of g l u c o c o r t i c o i d s effect

(19,20).

has

There

of g l u c o c o r t i c o i d s

above

adult

Rosenfeld pubertal

and

on

on

been

is no

data

IGF-II

are age and

Somatomedin

levels

increase

during

two y e a r s

events

(21)

to t h e s e

to t h r e e

achieved

years

earlier

they after

with

changes

levels

that

maximal

and remains above

adult

to four rise

t h a n in

in S m - C / I G F - I age,

bone

shown

dependent

boys.

and

of found

age and

correlation with Sm-C/IGF-I

fold

occurring

the r e l a t i o n s h i p

but a poor found

have

sex

two the

in girls

chronologic

development In fact,

puberty,

investigated

correlation with

velocity.

of s t u d i e s

levels

e t al

pubertal

A number

Sm-A

levels

approximately

good

Androgens.

Sm-C/IGF-I

(21,22,23).

two

effect

and

binding

humans.

Estrogens

of

by

conflicting

protein

(1,16)

the

(1 , 1 5 , 1 6 ) .

glucocorticoid

growth

not m e d i a t e d

is due

gave

competitive

in patients w i t h

reported

levels

bioassays

Sm-C/IGF-I

severe

have

on somatomedin

using

shown that

causes

of s t u d i e s

employing

studies

excess

stage growth

levels

peak

linear

growth velocity

levels

until

growth

a

has

is

203 virtually to o t h e r

ceased. hormonal

correlation with In g i r l s

the

to 70

Sm-C/IGF-I

increase

of

pg/ml

synthesis

consistent

wth

somatomedin and

also

androgen,

Turner's

increased

effect was above

pretreatment

estrogen

Sm-C/IGF-I

on

syndrome.

observed

In

boys,

level The

estrogen

although on

effect

a

direct

somatomedin

of e s t r o g e n

with

in

concentrations

effect

is

of e s t r o g e n

by W i e d m a n n

and

on

Schwartz

(25).

the e f f e c t s

However,

adult

declined.

The decrease

as r e p o r t e d

levels.

of t h e

Sm-C/IGF-I They

the S m - C / I G F - I

alone.

interaction

level

to

of a s t i m u l a t o r y

inhibitory

influence

oxandrolone,

estrogen

secretion,

possible.

e t al

further.

of e s t r o g e n

in higher

production

(26) h a v e r e p o r t e d

given

hormone

girls.

estrogen

declined

at m o d e s t

the r e s u l t

a direct

Clemmons

Androgens

with

be

is a l s o

observed

a s the

Sm-C/IGF-I

boys and

plasma

increased

in

significant

in both

as the

occurring

or a s y n e r g i s t i c

hormone

a

then gradually

concentration

Sm-C/IGF-I

(24)

level

and

on growth

low

found

levels

increased

could

changes

and then gradually

in Sm-C/IGF-I

estrogen

growth

pg/ml

these

they

increased

level

to 20

concentrations

effect

estrogen

as the e s t r o g e n

increased

of

changes,

Sm-C/IGF-I

increased levels

In r e l a t i n g

level

when

found

al

non-aromatizable levels

that

in five

combined with levels

girls

oxandrolone

by a p r r o x i m a t e l y

and Sm-C/IGF-I levels.

R u d m a n n et

hGH a

30?

when

synergistic

increased

350?

204 IGF-II

levels

show

no p u b e r t a l

sex s t e r o i d s

do n o t r e g u l a t e

This

suggests,

finding

Sm-C/IGF-I result

also

cause

a rise

Progestins.

in

Recent

also

regulate

plasma

occurs

during

have

reported

that

increase

plasma

levels.

studies

suggest

pharmacologic

secretion, direct this

Since

this

effect

MPA

finding

occurs

throughout ng/ml

during

at t e r m

occurs

i n hP1

Additional

and

studies

progesterone interaction

(29).

This

aimed at

on somatomedin are

binding

of

the

(MPA)

et al

increase

in

with

has

important

levels

the rise

clearly

in

the

productionand

on the h o r m o n a l

regulation

of

of

160

which

Sm-C/IGF-I.

effects

their

of

possible

warranted.

plasma

of

rise

SUMMARY Data

a

somatomedin

the rise

elucidating

in

hormone

a mean concentration parallels

the

40$

steroid The

may (28)

increase

growth

this

to

synthetic

approximately

that

the

be e x p e c t e d

Progesterone

rise

correlates

is n o t

Meyer

to t h e r i s e

reaching

in pregnancy

by

that

plasma

progesterone

generation.

pregnancy.

pregnancy

that

acetate

suggests

relate

would

doses

not

on somatomedin

o b s e r v a t i o n may

which

does

in

puberty

levels.

Sm-C/IGF-I concentration T a b l e II.

adult m a l e s A

rise

in s o m a t o m e d i n

IGF-II

progestin medroxyprogesterone

implying

concentration.

the

such an increase

Sm-C/IGF-I

(27),

plasma that

of a n o n s p e c i f i c since

its

however,

levels which

protein levels

increase

somatomedin

205

TABLE II EFFECT OF MEDROXYPROGESTERONE ACETAIELON PLASMA Sm-C LEVELS Subject

Sm-C(U/ml) Before RX

During Rx

1

0.92

1.55

2

1.02

1.46

3

0.65

1.35

4

1.16

1.25

5

0.74

1.18

6

1.01

1.06

7

0.87

1.76

8

0.84

1.12

9

1.60

1.63

10

0.69

1.13

11

1.16

1.69

Mean + 1SD

.97 + .27

Significance p 2.0 U/ml Sm-C after incubation* at pH 7.4

0.91 (72)**

0.96 (10)

Sm-C after incubation* at pH 3.6

0.89 (72)

0.87 (20)

Sm-C after acid gel chromatography***

0.92 (all values)

*incubation at 37C for 24 hr ••numbers in parentheses are number of samples tested •••using method of Zapf et al, J. Clin. Invest. 68, 1321 (1980).

Somatomedin-C in normal individuals The concentrations of Sm-C in EDTA plasma from 220 normal individuals between 18 and 64 years conform to a log-normal distribution, with a mean value of 0.90 units/ml (95% confidence limits, 0.4-2.0 units/ml).

During adult life, values tend to

decline with age, and it has been reported that the relatively low Sm-C of older adults can be increased by administration of GH (10).

In 122 women, the mean Sm-C was 1.06 units/ml, while

in 98 men, this value was 0.87 units/ml.

The mean Sm-C in cord

blood is about 0.35 units/ml and values remain relatively low until 3-5 years of age.

A cross-sectional study involving over

800 normal children, done in conjunction with Wayne Moore, (University of Kansas), Michael Preece (University of London) and workers at Nichols Institute, indicates that Sm-C concentrations have an accelerated increase around 6 years of age and reach peak levels around the time that stage II (Tanner), genital,

245

pubic hair and breast development occur.

Even in childhood,

values in girls are higher than those in boys. We have measured Sm-C in samples drawn approximately every 30 minutes from an adult male and a female.

Values changed little

over a 48 hour period and do not appear to be altered by eating or other routine activities. during sleep.

A modest decline seemed to occur

In another study (11), in which blood was col-

lected by a portable constant flow, withdrawal pump over 24 hours, 16 normal subjects showed a small but definite decline in Sm-C during periods of sleep (mean for waking hours = 1.13 ± 0.09 U/ml; for sleep = 0.85 ± 0.08).

Regulation of Sm-C by factors other than GH To interpret the Sm-C value in individuals in whom GH deficiency or excess is suspectecd, it is necessary to have an understanding of the factors other than GH which may raise or lower the circulating levels of this peptide.

We have observed that a

single injection of ovine placental lactogen (oPL) raises the serum somatomedin of hypophysectomized rats (12) and that the SmC dose-response to oPL is parallel to and equipotent with ovine GH (unpublished).

Furthermore, in a cross-sectional study of

pregnant women, Sm-C concentrations were found to be raised after the 20th week of gestation, to be highest in the last few weeks of pregnancy, to correlate with serum levels of hPL and to fall promptly following delivery of the placenta (13).

While

the pregnancy-related elevation in Sm-C has not been proven to be secondary to placental lactogen, the findings in rats and the close temporal relationships between the rise and fall of Sm-C and the r ise and fall of hPL suggest that this is a possibility. We also have shown that secretion of excessive amounts of prolactin raises Sm-C values to normal in patients with GH deficiency secondary to pituitary tumors (14).

In 23 patients

246

with large pituitary-region tumors, GH deficiency and normal prolactin levels, the mean Sm-C concentration was only 0.23 ± 0.10 units/ml (1SD).

On the other hand, 20 patients with large

prolactin secreting pituitary tumors and GH deficiency, had normal Sm-C concentrations (mean =1.0 ± 0.44 U/ml).

In patients

who do not have GH deficiency, increased prolactin did not raise Sm-C above the normal range.

These data suggest that in humans,

prolactin is a weak stimulator of somatomedin secretion which produces a detectable effect only when GH deficiency is present. The difference in potency between GH and prolactin for Sm-C induction might be explained by the differential receptor specificity of each hormone for binding to receptors.

In IM-9 lympho-

cytes and liver membranes, human prolactin is a weak competitor for binding to human GH receptors, and ¿hese relative potencies correlate with their respective potencies for Sm-C secretion in vivo.

The important practical implication of the finding that

excessive prolactin raises Sm-C is that hyperprolactinemia must be excluded in patients with pituitary tumors before Sm-C can be used as a screening test for GH deficiency.

FIGURE 5. Nitrogen balance and immunoreactive plasma somatomedin-C levels during fasting and refeeding of 7 slightly obese adults. Nitrogen balance (top panel) was determined as the nitrogen intake minus daily urinary urea nitrogen plus 2 g nitrogen (2 g nitrogen were estimated to be the loss in stool, skin, and urinary nonurea nitrogen). The mean (± SEM) balance values are depicted in the upper panel and the mean (± SEM) plasma somatomedin-C is depicted in the lower panel. The control day sample represents the mean values for all subjects on 3 consecutive control days. Reproduced from Clemmons et al: J. Clin. Endocrinol. Metab. 53^, 1247 (1981), with permission.

247 Recent studies from our laboratory using the Sm-C RIA have confirmed the bioassay results of many other workers, showing that nutritional status is an important determinant of plasma Sm concentrations.

We assessed the effect of fasting for 10 days

on plasma concentrations of immunoreactive Sm-C and urinary urea nitrogen excretion in 7 obese male volunteers (15).

From a mean

prefast value of 0.83 units/ml, plasma Sm-C fell to 0.21 units/ml after 10 days of fasting (Fig. 5). ed with refeeding.

A prompt increase was observ-

The change in Sm-C during fasting showed a

highly significant correlation with the change in urinary urea nitrogen excretion (Fig. 6).

These results suggest that plasma

IR-Sm-C is a sensitive indicator of nitrogen loss and may be useful in monitoring the changes in protein metabolism that occur during alterations in nutritional status.

FIGURE 6. Correlation between the percent control urea nitrogen excretion and the percent control plasma somatomedin-C for 36 fasting days in 7 slightly obese adults. Reproduced from Clemmons et als J. Clin. Endocrinol. Metab. 53, 1247 (1981), with permission.

01 z o LU S o

100

-

90

-

• • r = 0.74 p< 0.001

80

.• .• •

70


00 0 1 O m • CM 00 m o i rH

^ oo

^ O O

CD - H

e

O in

ai i—i

(D 73 fH • H

>>

Ol o •Ñt< O CO o CD i n CO 0 1 1—I CO m o i CM c ~

i

a o

>>

rH

>> M

rH fH

0 ft

W

W

öS >)

00 ^ rH t> r H i—l CM CO

328 the phosphorylated and dephosphorylated forms.

Of these two

forms, only the dephosphorylated form is active (7,8).

Since

serum insulin and C-peptide levels are subnormal in these patients (9), the decrease in cyclic AMP level leads to a lowering of the activity of cyclic AMP-dependent protein kinase which catalyses the phosphorylation of both enzymes. Thus both enzymes will be present in the active dephosphorylated form resulting in an increase in the activity.

However, the possibility that the enzyme changes

reported in the present and previous studies (1,2) may be secondary to the effect of IGF produced by the tumour (10) cannot be excluded. Increased triglyceride content and Acetyl-CoA Carboxylase activity may also be encountered in the tumour of patients with terminal hypoglycaemia.

References 1.

McEadzean, A.J.S., Yeung, R.T.T.: Am. J. Med. 47, 220-235 (1969).

2.

Yeung, R.T.T., Yeung, D.C.Y., Au, K.S.: Cancer 32, 14821489 (1973).

3.

Hugget, A. St. G., Nixon, D.A.: Lancet 2, 368-370 (1957).

4.

Henry, R.J.: Clinical Chemistry (1966); Harper and Row.

5.

Thomas, J.A., Schlinder, K.K., Larner, J.: Anal. Biochem. 25, 486-499 (1968).

6.

Inoue, H., Lowenstein, J.M.: Methods in Enzymology (S.P. Colowick and N.O. Kaplan eds)., 35 (B) p.3 Academic Press, New York (1975).

7.

Newsholme, E.A., Start, C.: Regulation in Metabolism 4, 146-194, John Wiley Sons (1973).

8.

Hardie, D.G., Guy, P.S.: Eur. J. Biochem. 110, 167-177 (1980).

9. 10.

Yeung, R.T.T., Teng, C.S. (to be published). Megyesi, K., Kahn, C.R., Roth, R., Gorden, P.: J. Clin. Endocrinol. Metab. 38, 931-934 (1974).

Insulin-Like Growth Factors in Fetal Growth and Development

ROLE OF SOMATOME DINS/INSULIN—LIKE GROWTH FACTORS IN THE REGULATION OF FETAL GROWTH Louis E. Underwood, Paul B. Kaplowitz, A. Joseph D'Ercole Department of Pediatrics, University of North Carolina, Chapel Hill, NC USA

Although the somatomedins have not been proven to be stimulators of fetal growth, they appear to be capable of such activity, since they have been shown to cause mitosis of cultured fetal cells.

A stimulatory role in the fetus also

is supported by the observations that fetal tissues from many species possess specific somatomedin receptors, cultured fetal cells can synthesize somatomedins, and the somatomedins are present in the fetal circulation (1).

In this presentation

we will review some of the evidence supporting the involvement of the somatomedins in fetal growth.

Particular

emphasis will be placed on our studies of somatomedin-C in the fetus.

Biological Actions of Somatomedins on Fetal Tissues Because of the limited quantities of pure somatomedins, no in vivo studies of the action of these substances have been carried out in the fetus.

All somatomedins, however, stimul-

ate mitosis of embryonic chick fibroblasts and human foreskinderived fibroblasts ^Ln vitro (2-4).

In addition human embry-

onic lung fibroblasts (WI-38) undergo mitosis in response to somatomedin-C (5) and fetal rat glial cells respond to somatomedin-A (6).

Although it remains to be proven that the stimu-

latory effect of plasma on fetal cartilage growth is due to the somatomedin contained in the plasma, there are studies

Insulin-Like Growth F a c t o r s / S o m a t o m e d i n s © 1983 Walter de Gruyter & Co., Berlin • New York

332 which suggest that the somatomedins in plasma may play a special role in proliferation of fetal cartilage.

Specifically

Ashton and Francis have shown that plasma stimulates thymidine uptake in isolated human fetal chondrocytes (7).

Other studies

show that fetal and neonatal tissues are particularly sensitive to stimulation by plasma (8, 9). We have recently reported on the effect of somatomedin-C and other peptide growth factors on the jjn vitro growth of mesenchymal cells derived from embryonic mouse limb buds (10).

Fore

and hind limbs of 11 day mouse embryos were dispersed by incubation at 37°C in 0.1% Trypsin, and plated as monolayer cultures (2.5-3x10^ cells/2.1 cm^ ), or as micromass cultures.

For the

latter, 2-2.5x10^ cells were spotted in a 15 ul drop, and after attachment, the high density culture was flooded with minimal essential medium (MEM) containing 5% baby calf serum.

After

incubation for 20-24 hours, media of both monolayer and micromass cultures were changed to MEM containing 0.2% baby calf serum, and the growth factors to be tested were added.

Response

to growth factors was assessed by cell count after 2 additional days of culture. In monolayer cultures, proliferation of limb bud mesenchymal cells was greatly favored by high cell density.

When plated at

5x10^ cells (a number sufficient to coat the culture surface), cell number more than doubled after 3 days.

When 2.5x10^ cells

were plated, cell number changed little for 3 days, then doubled by the fifth day.

When 1.25 x 10 5 cells were plated, cell num-

ber declined over the 5 day culture period.

Also in monolayer

cultures the individual addition of 0.2% serum, EGF (10 ng/ml), FGF (150 ng/ml), MSA (250 ng/ml) porcine insulin (1 ug/ml) or somatomedin-C had no significant effect on cell number.

However,

medium conditioned by mouse fetal liver explants stimulated mesenchymal cell growth at concentrations as low as 2% (V/V). Maximal effects of fetal liver conditioned medium were observed

333 at 5-10% (cell number, 150-170% of control).

While EGF and in-

sulin had no effect alone, both enhanced the stimulatory effect of liver medium (cell number increased to 250% of control). In high density micromass cultures, incubation in 0.2% serum alone resulted in a 50% increase in cell number between days 1 and 3 (Fig. 1).

During this period, localized areas of cell

aggregation are formed and chondrogenesis begins.

Addition of

EGF, FGF, MSA, insulin and somatomedin-C all caused significant stimulation of cell growth (Fig. 1).

rV X

cc UJ m 4

X

I r - r l n

Z3

2

EGF FGF In« MSA Sm-C lOng/ml 200ng/ml l>jg/ml 200ng/ml 20 ng/ml

Figure 1. Effect of growth factors on micromass cultures. Cultures were established for 20-24 hours. After this (day 1), growth factors were added to medium containing 0.2% serum. Data are presented only for the lowest concentrations which resulted in maximal stimulation of cell growth. The range of concentration tested was as follows: liver conditioned medium (LM) 2 to 10 %, EGF 5 to 50 ng/ml, FGF 50 to 300 ng/ml, insulin 0.3 to 10 ug/ml, MSA 100 to 500 ng/ml, and somatomedin-C (Sm-C) 5 to 40 ng/ml. Cell counts were done on day 3 except for day 1 control. Results are expressed as mean ± S.D. All cultures with additions of purified growth factors have a significant increase in cell number (P< 0.0025). Reproduced from Kaplowitz, P.B. et al, J. Cell Physiol. 112, 353 (1982), with permission.

334

X A

o 7 Figure 2. Effect of combinations of three and four growth factors on micromass cultures. See Figure 1 for experimental details. Reproduced from K a p l o w i t z , P.B. et al, J. Cell Physiol. 112, 353 (1982), with permission.

LLI

6

LM EGF FGF Ins -

In the m i c r o m a s s system, the g r e a t e s t stimulation of limb bud cell growth was obtained with a combination of tioned medium, EGF, FGF and insulin

centrations at which they had a m a x i m a l e f f e c t (Fig. 2).

liver-condi-

(LEFI), all used in conindividually

O n day 3, the cell number in cultures treated with

these 4 factors was 82% greater than controls, and the rate of increase in cell number between days 1 and 3 was more than 3 fold greater than in controls, and exceeded that produced by an optimal concentration mine the relative

(5%) of baby calf serum alone.

To deter-

importance of each component of LEFI, the

e f f e c t of omitting one factor at a time was determined. out EGF, stimulation was greatly reduced.

O m i s s i o n of

m e d i u m had a m a j o r effect, while o m i s s i o n of FGF or caused a smaller but significant reduction in g r o w t h tion (p < 0.001).

Withliver

insulin stimula-

Since insulin and somatomedin-C m a y exert

their growth-promoting effects on some cells by the same m e c h a n ism, we examined their interaction in this system. at optimal c o n c e n t r a t i o n s neither insulin and

W h e n added

somatomedin-C,

nor insulin and MSA produced additive effects on growth. atomedin-C

(20 ng/ml) could replace insulin

Som-

(1 ug/ml) in the

LEFI combination, and addition of insulin to m e d i u m

containing

335

somatomedin-C plus liver conditioned medium, EGF, and FGF did not further stimulate limb bud cell g r o w t h

(Table 1).

These

results suggest that either insulin, MSA or somatomedin-C

can

satisfy the growth requirement of limb bud cells for somatomedin-like

peptides.

T a b l e 1. E f f e c t s of S o m a t o m e d i n - C a n d limb bud c e l l s in m i c r o m a s s c u l t u r e s ^ Additions

i n s u l i n o n q r o w t h of

Cell Number

None

(xlo-5) on day 3

5.44

Insulin

6.93

Somatomedin-C

6.92

I n s u l i n 4- S o m a t o m e d i n - C

6.82

LEP2

8.03

Insulin + LEF

9.49

Somatomedin-C + LEF

9.51

Insulin + Somatomedin-C + LEF

9.67

^ M i c r o m a s s c u l t u r e s w e r e e s t a b l i s h e d a s in P i q û r e s 1 a n d 2 e x c e p t t h a t the c e l l n u m b e r o n d a y 1 w a s g r e a t e r 2.8x10s).

(3.9

I n s u l i n (1 uq/ml) a n d / o r S o m a t o m e d i n - C

{20

a n d , w h e r e i n d i c a t e d l i v e r m e d i a , E G F , and F G F (LEF)

vs ng/ml) were

a d d e d o n d a y 1 a n d c e l l s w e r e c o u n t e d o n d a y 3. 2

L E F = Liver conditioned medium

(200

Reproduced 353

(10%) + E G F (10 n g / m l ) +

FGF

ng/ml). f r o m K a p l o w i t z , p . B . e t a l , J. C e l l P h y s i o l .

(1982), w i t h

112,

permission.

12

Figure 3. Time course of m i c r o m a s s culture growth. From day 1 (0 hr), half of the cultures were incubated w i t h MEM containing 0.2% serum only (control) and the remainder with liver conditioned m e d i u m (10%) + EGF (10 mg/ml) + Insulin (1 ug/ml) + F G F (200 ng/ml). Cell counts were perfromed at the intervals indicated. Reproduced from Kaplowitz, P.B. et al, J. Cell Physiol. 112, 353 (1982), with permission.

LM+EGF+Ins + FGF^P /

uT 10

/

'o

ui m s u

/

/

X

a:

/

8

„

CONTROL

0

24

36 48 HOURS

72

The time course of g r o w t h factor-stimulated micromass growth is depicted in Fig. 3.

By 72 hours of incubation the cultures in

LEFI m e d i u m had twice as many cells as controls and 3.5 times as many as had been plated.

From this we infer that on the

336 average, these cells undergo 2 consecutive mitotic cycles within 3 days.

This rate of growth compares favorably with the

increase in limb bud cell number _in vivo between days 11-13 (10). The liver conditioned medium growth-stimulating activity has an apparent molecular weight by Sephacryl -200 chromatography of 30,000-40,000, is non-dialyzable, is stable at 56°C for 30 min, but is destroyed by boiling for 10 min.

When liver medium was

exposed to glass beads coated with somatomedin-C anti-serum, there was no diminution in its growth-promoting activity, despite the fact that virtually all the immunoreactive somatomedin-C had been removed.

These findings, along with the observation that

none of the previously characterized growth factors can fully duplicate its activity, suggests that liver medium contains a distinct growth factor.

These studies, taken together with our

previous findings that mesenchymal cells synthesize somatomedinlike growth factors and possibly other growth factors, make it clear that complex mechanisms are involved in the growth of limb bud cells, and that several peptides may be involved.

Somatomedin Binding by Fetal Tissues Following the observation that human placenta could bind 125j_ somatomedin-C, we showed that in the fetal pig somatomedin-C was bound at all gestational ages and in all tissues lung, kidney, liver and heart) studied (11).

(placenta,

Almost without

exception, the specific binding of 125j_somat-0niedin-C exceeded binding of 125j-insulin.

In some tissues (late gestation fetal

placenta and fetal lung), somatomedin-C receptor number and/or affinity were significantly increased over early gestation tissues and tissues from adult animals.

We subsequently have

observed that membranes prepared from fetal, placental, and decidual tissues of the mouse also possess somatomedin-C binding sites (12).

Daughaday et al have described

somatomedin

337 receptors in fetal rats (18).

Owens et al, studying the bind-

ing of MSA also have shown that somatomedin receptors are widespread in tissues of the fetal sheep (14).

Increased somato-

medin binding has been observed in human cord blood monocytes (15) and in membranes prepared from brains of human abortuses (16).

While none of these studies secure a role for somato-

medin in fetal growth, the demonstration that somatomedins bind to fetal cell membranes suggests that such a role is possible since the presence of specific receptors is thought to be an essential prerequisite to biologic action.

Origin of somatomedin in the fetus After injecting 125 I _ s o m a t o m e ( 3i r l _c

into the circulation of the

pregnant dog, sheep and rat, we observed absolutely no placental transfer of labeled hormone and concluded that fetal somatomedin is made either by the placenta or the fetus (17).

Rechler

et al (18) then reported that fetal rat liver synthesizes MSA in an organ explant system.

Using a similar explant system, we

studied the appearance in incubation medium of immunoreactive and membrane receptor reactive somatomedin-C.

We observed that

multiple tissues from fetal mice (limb bud mesenchyme, intestine, heart, brain, kidney, liver, lung) but not placenta, are capable of producing somatomedin (19).

This somatomedin appears to be

synthesized de novo by the fetal liver from as early as 11 days of gestation.

It has chemical characteristics which are similar

to human somatomedin-C and is the same size as one of the molecular forms of somatomedin-C found in human serum.

Haselbacher

et al (10) also have reported that IGF-I is synthesized by chick embryo liver cells.

Our finding that multiple fetal tissues synthesize somatomedinC, as well as the reports of Atkison et al (21) and Clemmons et al (22) that cultured human fibroblasts synthesize somatomedins, has led to the postulate that the primary actions of somatomedin

338

might be exerted locally at its sites of origin.

Although a

function of this type has not been proven, it seems possible that fetal cells are capable of producing local mitogens which act in a paracrine fashion according to the scheme proposed by Sporn and Todaro (23).

Somatomedins in fetal blood:

Concentrations and molecular

forms Compared to adult or maternal sera, cord serum somatomedin concentrations are low when measured by bioassays, radioreceptor assays, and radioimmunoassays (see ref 1).

Using bioassays, a

positive correlation has been observed between somatomedin concentrations and birth size (24, 25), and depressed

somatomedin

concentrations in small for gestational age infants have been reported

(26).

We have reported

(1) that immunoreactive somato-

medin-C in 145 cord sera obtained from full-term infants correlated with birth weight, birth length, and placental weight (r=0.39, p < 0.001; r=0.38, p < 0.001; and r=0.34, p < 0.001, respectively).

Small for gestational age infants ( < 2500 gms)

had significantly lower cord somatomedin-C concentrations than those of appropriately grown infants (p 3800 gms) were significantly higher (p < 0.05). Moses et al (27) have reported that immunoreactive MSA in fetal rat serum is 20-100 fold higher than in maternal sera. Daughaday et al using a rat placental membrane binding assay for IGF II and MSA, also observed more activity in fetal and early postnatal rat serum (13).

These observations raise the possibility

that the neutral somatomedins such as MSA might be more important in the fetus than basic somatomedin.

On the basis of

these findings one might expect relatively high levels of IGFII in human fetal serum.

This, however, has not been found to

be the case, at least for cord serum (28).

Further studies are

339 needed to determine whether serum IGF II concentrations are elevated earlier in human fetal life.

Sara et al (16), using

125j_gra_A and membranes derived from the brain of second trimester human abortuses, have reported that human fetal plasma somatomedin concentrations are elevated from as early as the second trimester.

These investigators have proposed that

unlike term placental or adult tissues, receptors derived from human fetal brain recognize an "embryonic somatomedin".

This

finding has not yet been confirmed. This puzzling relationship between fetal blood somatomedin concentrations and somatomedin action in the fetus will remain unsolved until several important questions are answered. Specifically, it needs to be determined:

(a) Whether the

neutral somatomedins are the fetal-active form, and if so why bioassayable somatomedin activity is low in the fetus, in the face of relatively large quantities of immuno and receptor reactivity.

(b)

Whether there is an "embryonic somatomedin"

distinct from those which have been defined.

(c)

Whether or

not circulating levels of somatomedin are of physiological relevance or whether fetal tissues are unusually sensitive to low circulating concentrations of somatomedin.

In addition to quantitation of immunoreactive somatomedin-C in fetal blood we have carried out studies of the circulating forms of this growth factor during fetal life (29).

The study

was done using blood samples collected prior to delivery of the placenta of 23 infants delivered between 20-43 weeks of gestation.

Serum was chromatographed at neutral pH on a Sephacryl

200 column.

Two distinct elution patterns of immunoreactive

somatomedin-C were observed.

Pattern I sera were characterized

by a single discrete peak of somatomedin-C with an apparent

340

molecular weight of approximately 150,000 daltons (150 K) (Fig. 4, top panel), and was observed exclusively in sera from third trimester infants.

The immunoreactive somatomedin-C in Pattern

II sera migrated only at

-40,000 daltons (40 K) (Fig. 4,

bottom panel) and was observed in sera from fetuses up to 27 weeks gestation. Figure 4. Sepharcyl 200 chromatography of cord sera from infants of different gestational ages. Sera from a full term and two premature infants were chromatographed after preincubation with 125j-somatomedin-C at 4°C for 30-60 min. The bars represent immunoreactive somatomedin-C calculated for each fraction as the percent of the total immunoreactive somatomedin-C measured in all fractions. In fractions where there are no bars, the somatomedin-C content was below assay limits. Sums of the somatomedin-C of the chromatographed fractions ranged between 58 and 123% of that which was loaded on the column. The open circles depict the migration of 1 2 5 j _ s o m a t o m e d i n - C . The closed circles represent OD 280. Vg is the elution volume of a particular fraction. Vg is the void volume or the volume at which blue dextran elutes from the column. Reproduced from D'Ercole et al, J. Clin. Endocrinol. Metab. 51, 674 (1980), with permission.

4 0 W e e k s Gestation

VE/V0

In all infants studied, the major portion of protein bound 125 somatomedin-C eluted at 40 K, suggesting the presence of unsaturated binding sites.

Pattern I sera, however, consistently

showed a small discrete peak of 1 2 5 I _ s o r a a t o m e ( 3 i n _ c binding at 150 K.

The pattern I samples resembled those of normal adult

sera except that in the latter, there is a small peak of immunoreactive somatomedin-C at 40 K and 1 2 5 j _ s o m a t o m e d i n - C

binding

341

at 150 K represents a greater portion of the bound l 2 5i-Somatomedin-C.

Since the 150 K binding protein is believed to be

under growth hormone control, the absence of the 150 K peak in mid-trimester fetuses might reflect fetal growth hormone resistance.

Alternatively, its absence may be due to immaturity of

the mechanisms involved in the synthesis of somatomedin binding proteins.

In a 43 week anencephalic fetus, a pattern II elution

profile was observed.

The absence of 150 K somatomedin-C in

this infant supports the hypothesis that 150 K proteins are acquired in response to growth hormone or other pituitary hormones.

It also suggests that human placental lactogen lacks

the capacity to stimulate synthesis of proteins necessary for the appearance of 150 K somatomedin-C.

Conclusions Evidence is mounting from _in vitro studies that the somatomedins stimulate fetal growth, but proof of such a function awaits direct JJI vivo studies of the effects of addition and removal of somatomedins.

Reports that purified somatomedins stimulate in

vivo postnatal growth opens the way for similar studies during prenatal life, once adequate quantities of growth factor are available.

It remains to be determined whether an "embryonic

somatomedin" is the primary insulin-like growth factor of the fetuses, or whether the relatively low serum concentrations of the known somatomedins are sufficient to stimulate fetal growth. Finally, the question of the physiological importance of circulating somatomedin needs to be addressed, in light of the hypothesis that the somatomedins may act primarily in a paracrine fashion.

342

Acknowledgements Research on the role of somatomedin-C in fetal growth was made possible by USPHS-NIH Research grant #HD08299 and #AM01022; USPHS-NIH Research Career Development Award # HD00435 to A. Joseph D'Ercole; USPHS-NIH training grant #AM07129; NIH Research Fellowship HD05982 to Paul B. Kaplowitz; a grant from the National Foundation March of Dimes #5-188; and a grant from the Human Growth Foundation. The authors wish to thank Judson J. Van Wyk, M.D., Department of Pediatrics, University of North Carolina for his continuing encouragement and support of this research, Drs. Marjorie Svoboda, and Douglas F. Willson for aid with portions of these studies; Ms. Mary Murphy, Billie M. MoatsStaats, and Evyonne Bruton for technical assistance; and Ms. Christine Silva for help in preparing the manuscript. References 1.

D'Ercole, A.J., Underwood, L.E.: Growth Factors in Fetal Growth and Development, in Fetal Endocrinology: Symposium of the Oregon Regional Primate Center Series, M.J. Novy, J.A. Resko (eds), Academic Press, Inc., New York 1981

2.

Van Wyk, J.J., Underwood, L.E.: In Biochemical Actions of Hormones (G. Litwack, ed) Vol. V, p 102, Academic Press, New York 1981

3.

Zapf, J., Rinderknecht, E., Humbel, R.E., Froesch, E.R.: Metabolism 27, 1803 (1978).

4.

Rechler, M.M., Podskalny, J.M., Nissley, S.P.: 259, 134 (1976).

5.

Weidman, E.R., Bala, R.M.: 92, 577 (1980).

6.

Sara, V.R., Hall, K., Ottosson-Seeberger, A., Wetterberg, L.: in Endocrinology 1980 (I.A. Cumming, J.W. Funder, F.A.O. Mendelsohn, eds) p 453, Elsevier/North Holland, New York 1980

7.

Ashton, I. K. , Francis, M.J.O.: (1978).

8.

Ashton, I.K., Matheson, J.A.: Calcif. Tissue Int. ^9, 89 (1979).

9.

Hill, D.J., Milner, R.D.G.: Ciba Foundation Symposium 86 on The Fetus and Independent Life, p 124, Pittman, London, 1981

Nature

Biochem. Biophys. Res. Com.

J. Endocrinol. 76^, 473

10. Kaplowitz, P.B., D'Ercole, A.J., Underwood, L.E.: Cell Physiol. 112, 353 (1982).

J.

343 11. D'Ercole, A.J., Foushee, D.B., Underwood, L.E.: Endocrinol. Metab. 43, 1069 (1976). 12. D'Ercole, A.J. , Underwood, L.E.: (1980).

J. Clin.

Develop. Biol. 79^, 33

13. Daughaday, W.H., Parker, K.A., Borowsky, S., Trivedi, B. , Kapadia, M.: Endocrinology 110, 575 (1982). 14. Owens, P.C., Brinsmead, M.W., Waters, M.J., Thornburn, G.D. : Biochem. Biophys. Res. Com. 96, 1812 (1980). 15. Rosenfeld, R. Thorsson, A.V., Hintz, R.L.: J. Clin. Endocrinol. Metab. 48, 456 (1979). 16. Sara, V.R., Hall, K., Rodeck, C.H., Wetterberg, L.: Human Embryonic Somatomedin. Proc. Natl. Acad. Sei. USA 28, 3175 (1981). 17. Underwood, L.E., D'Ercole, A.J., Furlanetto, R.W., Handwerger, S., Hurley, T.W.: In Somatomedin and Growth (G. Giordano, J.J. Van Wyk, F. Minuto, eds) p 215 Academic Press, New York 1979 18. Rechler, M.M., Eisen, H.J., Higa, O.Z., Nissley, S.P., Moses, A.C., Schilling, F.E., Fennoy, I., Bruni, C.B., Phillips, L.S., Baird, K.L.: J. Biol. Chem. 254, 7942 (1979). 19. D'Ercole, A.J., Applewhite, G.T., Underwood, L.E.: Develop. Biol. 75, 315 (1980). 20. Haselbacher, G.K., Andres, R.V., Humbel, R.E.: Eur. J. Biochem. Ill, 245 (1980). 21. Atkison, P.R., Weidman, E.R., Bhaumick. B., Bala, R.M.: Endocrinology 106, 2006 (1980). 22. Clemmons, D.R., Underwood, L.E., Van Wyk, J.J.: Invest. 67, 10 (1981). 23. Sporn, M.B., Todaro, G.T.: (1980).

J. Clin.

N. Engl. J. Med. 303, 878

24. Gluckman, P.D., Brinsmead, M.W.: J. Clin. Endocrinol. Metab. 43, 1378 (1976). 25. Ashton, I.K., Vesey, J.: Ear. Human Devel. 2, 115 (1978). 26. Foley, T.P., DePhilip, R., Perricelli, A., Miller, A.: J. Pediat. 605 (1980). 27. Moses, A.C., Nissley, S.P., Short, P.A., Rechler, M.M., White, R.M., Knight, A.B., Higa O.Z.: Proc. Natl. Acad. Sei. USA 77, 3649 (1980). 28. Zapf, J., Walter, J., Froesch, E.R.: 68, 1321 (1981).

J. Clin. Invest.

29. D'Ercole, A.J., Willson, D.F., Underwood, L.E.: Endocrinol. Metab. 51, 674 (1980).

J. Clin.

REDUCED PLASMA SOMATOMEDIN ACTIVITY DURING EXPERIMENTAL GROWTH RETARDATION IN THE FETAL AND NEONATAL RAT

David Hill, Miklos Fekete, David Milner Department of Paediatrics, University of Sheffield, Children's Hospital, Sheffield S10 2TH, U.K. Frans De Prins, Andre Van Assche The Unit for the Study of Reproduction, Department Ontwikkelingsbiologie, Katholicke Universiteit, Leuven, Belgium

Introduction The somatomedins are present in the fetus and neonate and a positive correlation with body size in utero has suggested that these hormones are closely implicated in fetal growth and development (1). Factors controlling the circulating levels of somatomedins in early life are poorly understood but nutritional availability is likely to play an irrportant role. In this study three experimental models have been utilized to produce a relative growth retardation in the fetal and neonatal rat. These were; 1) a limitation of the maternal blood supply to the gravid uterus; 2) rratemal fasting during late gestation and 3) the creation of large and small litter sizes on the day of birth. The consequences of these procedures on circulating levels of bioassayable somatomedin activity, cartilage metabolic activity and other hormones implicated in early body growth were assessed.

Methods The artery and vein to one uterine horn of the pregnant Wistar rat were closed with a single ligation positioned at the cervical end beyond the last fetus on day 16 of gestation (day of mating taken as day 0) by the

Insulin-Like Growth Factors / Somatomedins © 1983 Walter d e Gruyter & Co., Berlin • New York

346

technique of Wigglesworth (2). The opposite, non-ligated horn contained the control animals. Fetuses were delivered by Caesarian section on day 20. In the second study pregnant rats were fasted from days 10 to 14 or days 17 to 21 of gestation. All fetuses were delivered on day 21. Two separate pools of fetal plasma were collected from the animals in the ligated uterine horn and from those in the control horn following uterine vessel ligation, and one pool was collected from all animals in a litter following fasting. Litters of 4 or 14 neonatal rats were created on the day of birth as described by Widdowson and PfcCance (3) and returned to a lactating rat. Animals were killed at days 4, 14 and 21 of life and a pool of plasma created for each litter. Fetal plasma somatomedin activity was determined by bioassay utilizing the incorporation of [35S] sulphate in vitro by costal cartilage from normal rat fetuses of 20 or 21 days gestational age as described previously (4). Postnatal rat plasma somatomedin activity was measured firstly by the postnatal pig cartilage assay and secondly using rat costal cartilage from animals of the same age as those which provided the plasma (1,4) . Activity was expressed as a potency relative to a standard pool of plasma from normal adult male rats (1 Unit/ml). Rat plasma insulin and growth hormone were measured by radioimmunoassay and glucose by a glucose oxidase technique. After measurement of body and organ weights, and nose-tail tip length, costal cartilage was taken from each fetus after uterine vessel ligation; from the four fetuses nearest the mean litter weight following maternal fasting; and from each neonatal rat in litters of 4 animals and the four rats nearest the mean litter body weight in litters of 14. The metabolic activity of cartilage in vitro was assessed by the uptake of [35S] sulphate in the presence or absence of 10% (v/v) standard adult rat plasma, called stimulated and basal activity respectively (4).

Results 1) Uterine vessel ligation. The body weight (mean + S.E.M.) of 108 fetuses (37 litters) in the ligated uterine horn was significantly lower than that of 146 control fetuses (2.82 + 0.05 g vs 3.18 ± 0.05 g; p2 and air at 37 °C for 3 days.

After that the

cells were grown in Eagle's minimum essential medium

(EMEM)

containing 10% FCS and antibiotics, and fed every other day. 2)

Measurement of glycosaminoglycan synthesis.

Glycosaminoglycan

(GAG) synthesis was monitored by measuring 35 incorporation of Na 2 SO^ using the following the procedure as described by Suzuki et al. (1). When the cells reached confluence, the media was changed to EMEM without FCS. After 24 hr incubation, the cells were washed three times with MgSO^ free Earle's solution and incubated for35 15 min at 37°C with the same buffer.

Then 1.5 pCi/dish Na^

SO^ was added to the

cells with test samples and incubated for 24 hr at 37°C in 5% CC>2 and air.

The reaction was stopped by 5% trichloroacetic

acid (TCA) and, the precipitant and cells were washed several times with 5% TCA.

They were digested in 1 mg/ml Pronase E in

0.2 M tris buffer for 10 hr at 55°C.

Then 1 mg/ml chondroitin

sulphate and 1% cetylpyridinium chloride were added and the mixture was incubated for 1 h at 37°C. washed with cold water and dissolved emulsifier and counted. 3)

Binding study to chondrocytes.

The precipitant was

in 10 ml of Insta-Gel

525

Chondrocytes which became confluent were detached from flask by dispase, washed and resuspended in buffer G. counted trypan

in a hemocytometer blue

exclusion.

exceeded 90%.

In

all

experiments

of 1-4 x 10

cell s/ml, either

viability

jil of the incubated containing

final concentration

I-somatomedin or

(10000 cpm) and unlabelled hormones. tubes

cell

The incubation mixture had a final volume of

0.5 ml and consisted of chondrocytes at 6

Cells were

and viability was determined by

suspension was aliquoted

200

pi

buffer

I-insulin

After incubations, 200 G.

into microfuge

These

tubes

were

centrifuged at 8000 x g for 1 min, supernatants were discarded and the cell pellets were counted.

Results 1)

Effects of somatomedin A and insulin on glycosaminoglycan

(GAG) synthesis (Fig. 1). Somatomedin

stimulated

GAG

synthesis

manner between 1.25 and 20 ng/ml. GAG

synthesis

control.

At

significantly 20 ng/ml of

179.6 ± 3.4 % of control.

in

a

dose

dependent

At 5 ng/ml of somatomedin A

increased

to

somatomedin GAG

151.3 ±. 4.6 synthesis

% of

reached

On the other hand, insulin also

stimulated GAG synthesis in a dose dependent manner between 50 and

1000

ng/ml.

significantly order

to

At

obtain

concentration

500

increased was

the

ng/ml

to 138.7 same

required

of

insulin,

±. 11.2

stimulatory

more

than

% of

GAG

effect,

100-fold

synthesis

control.

In

insulin

greater

than

that of somatomedin. 2)

Association

of

somatomedin

A

and

insulin

with

rat

cultured chondrocytes. 125 125 The time course of I-somatomedin A and I-insulin binding to rat cultured chondrocytes at 4, 15 and 37°C are shown in Fig. 2.

At 15°C, somatomedin A binding increased gradually

and reached a steady state in 2 hr.

The binding at 4°C was

526 EFFECT OF SMA AND INSULIN ON GAG SYNTHESIS

SOMATOMEDIN F i g .

1

>

> 200

*

150

D UJ


?'} XßmfäMßJ?-

/ Figure 2.

-V f f l ^ W « '«•• f./

4

•

Differentiation of BC3H1 cells. Day 1 cells reveal large, poorlydifferentiated myoblasts. Day 9 cells show confluent, elongated forms characteristic of the mature myocyte forms.

535 reached a peak a t day 5 post-partum, and subsequently achieved baseline levels by day 9. On t h e other hand, maternal levels of serum IGF-II remained relatively stable throughout gestation in t h e f a c e of marked variations in hepatic IGF-II binding. In the current study, we have found t h a t serum insulin levels in mothers and pups remained unchanged through gestation and post-partum, a p a t t e r n similar to t h a t observed for insulin receptor binding. IGF-II binding was f u r t h e r evaluated in an in vitro model using a mouse cloned muscle cell line in monolayer tissue culture (BC3H] cells).

IGF-II binding to these

cells was time, t e m p e r a t u r e , concentration, and pH-dependent.

Degradation of

labelled IGF was less than 20% of total t r a c e r added a f t e r periods of up to 2 hrs incubation a t 37° C.

125i_iqf-II binding was not displaced by insulin, IGF-I, or any

non-insulin-like peptides.

The specificity of binding was virtually indistinguishable

f r o m that seen in the hepatic membranes in the in vivo study. Of interest was t h e observation t h a t microgram per ml concentrations of

highly-purified MSA, an

amount sufficient to inhibit t r a c e r binding to cells by more than 90%, caused no down-regulation of homologous receptors following periods of incubation up to 16 hrs.

Ontogeny of IGF-II receptors in these cells revealed the highest

concentrations

during days 1 - 5 of d i f f e r e n t i a t i o n (Fig. 2), at which time they were primarily myoblasts.

By days 6 - 10, cells were essentially confluent and fully d i f f e r e n t i a t e d

t o myocytes (Fig. 2), whereupon IGF-II receptors reached their nadir.

In c o n t r a s t ,

insulin receptor levels were minimal in the early phases of d i f f e r e n t i a t i o n , and gradually increased to peak levels by day 6 - 7 .

Discussion The results of this investigation add to a growing body of data suggesting t h a t IGFII may have unique and important regulatory roles in f e t a l growth and development (1, 2, 3, 9).

In contrast to the stable p a t t e r n of insulin receptor binding and serum

536 insulin levels in fetal and neonatal rats, IGF-II and its receptors appear uniquely regulated during periods of maximum growth and development.

Thus, the striking

differences between IGF-II and its receptor versus insulin and its receptor in late fetal and early neonatal life suggest that these growth factors may have different roles in fetal development, with a predominant influence of IGF-II appearing most likely in the rat. It would be of considerable advantage to have an appropriate in vitro model in which to evaluate correlations between emergence of IGF receptors and regulation of IGFsensitive biologic responses during development. provide such a model.

The BC3H] cultured muscle cells

These cells have been shown to possess functional insulin

receptors, with biologic responses to physiologic concentrations of hormone (10). These cells have also been shown to possess specific IGF-I receptors (11). Our results demonstrate the presence of specific IGF-II receptors, and we have found these binding sites to be functionally linked to a variety of biologic responses (unpublished observations). The striking profile observed in the ontogenesis of IGF-II receptors in these cells and its quite different pattern from that found in the insulin receptors suggest that these receptors may have quite different roles during cellular differentiation. Thus, the availability of such an in vitro model provides a tool for investigating the various questions we have raised during this discussion.

Such studies are currently

underway and may provide important insight on the roles of various IGFs in growth and development both in vitro and in vivo.

References 1.

Kelley, P.A., Posner, B.I., Tsushima, T., Friesen, H.G.: Endocrinology 95:532 (1971).

2

D'Ercole, A.J., Foushee, D.B., Underwood, L.E.: J . Clin. Endocrinol. Metab. 13:1069 (1976).

537 3.

Daughaday, W.H., Parker, K.A., Borowsky, S., Trivedi, B., Kapadia, M: Endocrinology 110, 575 (1982). Cuatrecasas, P . : Proc. Natl. Acad. Sci. USA. 69, 318 (1972).

5.

Lowry, O.H., Rosebrough, N.3., Farr, A.L., Randall, R.3.: J . Biol. Chem. 265 (1951).

193,

6.

Tait, 3 . F . , Weinman, S.A., Bradshaw, R.A.: 3. Biol. Chem. 256, 11086 (1981).

7.

Gavin, 3 . R . , III, Gorden, P., Roth, 3., Archer, 3.A., Buell, D.N.: J . Biol. Chem. 248, 2202 (1973).

8.

Hizuka, N., Takano, K., Shizume, K., Hasumi, Y . : A c t a endocr. 97, 352 (1981).

9.

Moses, A.C., Nissley, S.P., Short, P.A., Rechler, M.M., White, R.M., Knight, A.B., Higa, O.Z.: Proc. Natl. Acad. Sci. USA 77, 3649 (1980).

10.

Pollet, R . 3 . , Standaert, M.L.: Clin. Res. 29, fl8A (1981).

11.

De Vroede, M.A., Rechler, M.M., Standaert, M.L., Pollet, R . 3 . : 64th Program of The Endocrine Society, Abstr. 327, 161 (1982).

REGULATION OF BINDING OF INSULIN AND INSULINLIKE GROWTH FACTOR BY CELL GROWTH STATUS

Beate Pfeifle, Volker Maier , Hans Ditschuneit Zentrum für Innere Medizin, Innere Medizin I und II, Universität Ulm, D-7900 Ulm

Introduction Insulin and insulinlike growth factor (IGF) are related polypeptides with similar biological activities (1,2). Insulin has a more potent metabolic effect than IGF and IGF has a more potent growth-promoting effect than insulin. Both factors are able to act via the binding sites for insulin and IGF (1,3). The metabolic effect may by mediated by the insulin receptor and the growth-promoting effect may by mediated by the IGF receptor. We examined the binding of insulin and IGF to cultured arterial smooth muscle cells in various growth state of the cells. Material and Methods IGF was isolated from lyophilized human serum by acidethanol extraction and acetone-ethanol precipitation. The precipitate was purified by Sephadex G-50 chromatography in 0.1 mol/l acetic acid and further purified by preparative isoelectric focusing. IGF with an isoelectric point of 8.5 (+0.2) contained a = 3:1 mixture of IGF I and IGF II according to a radioimmunological determination in the laboratories of Prof. E.R.Froesch, Zürich. The specific insulinlike activity was 10 mU/ml measured by stimulation 14 of ( Cj-glucose uptake into lipids of isolated fat cells of rat. Radioimmunoassay revealed that less than 1 ng/mg

Insulin-Like Growth Factors / Somatomedins © 1983 Walter d e Gruyter & Co., Berlin • N e w York

540

of the protein could be accounted for by immunoreactive insulin. IGF was iodinated to a specific activity of 200 yuCi/yug according to the method of Hunter & Greenwood (4). Porcine insulin was purchased from Eli Lilly, Bad Homburg, Germany, and was iodinated to a specific activity of 180 yuCi/yug. Smooth muscle cells from the intima and media of the rat aorta were cultivated by explantation in modified Dulbecco's Modified Eagle Medium containing 10% fetal calf serum (5). Binding studies (2) were carried out with cells which were growing in petri-dishes for 1, 2, 3, 4 and 5 days, in 0.1 mol/l Hepes-buffer, pH 7.5, containing 0.2% human serum albumin (HSA) and

125

I - I G F (20 nCi/ml) or

125

I-insulin

(6 nCi/ml) and increasing concentrations of insulin and IGF. Incubation time was 90 min at 20° C for and 120 min at 12° C for

125

125

I - I G F binding

I - i n s u l i n binding.

Results The growth state of the cells regulated the binding of 125 125 I-insulin and I-IGF to arterial smooth muscle cells. 125 I-IGF binding to the cells decreased from days 1-5 in culture. 125 I-insulin binding increased. The interaction among the binding sites for IGF and insulin varied with the growth state of the cells. 125 IGF competed with I-IGF for its binding sites in the logarithmic and stationary phase of growth. Half-maximal 125 inhibition of I-IGF binding was produced by IGF between 10 and 100 nmol/l (Fig.1). Insulin competed only weakly 125 with I-IGF for its binding sites in the logarithmic phase of growth. In the stationary phase of growth, insulin 125 displaced ^I-IGF from its binding sites with a half-maximal inhibition at 1 ^umol/l (Fig. 1). 125 I-insulin binding could be displaced by insulin in the

541

22 wks). Within 9 hours of death, the whole brain was removed at autopsy from a 76 year old woman without any signs of neurological disorders. The brain was dissected into various regions. Frontal lobe biopsy samples showing no sign of disease, were removed at surgery. All tissues were stored at -80°C until plasma membranes were prepared by ultracentrifugation (3, 4). IGF-1 and IGF-2 were kindly provided by René Humbel. MSA II, used in fetal studies, was kindly provided by S. Peter Nissley and Mathew Rechler. MSA, obtained from Collaborative Research Inc., was used for displacement from adult brain membranes. Porcine insulin (25 IU/mg) and proinsulin were supp-

Insulin-Like Growth Factors / Somatomedins © 1983 Walter de Gruyter & Co., Berlin • N e w York

546 l i e d by the Nordic I n s u l i n L a b o r a t o r i e s . Hormones were l a b e l l e d by the lactoperoxidase method and p u r i f i e d on carboxymethylcellulose on a pH g r a dient i n ammonium acetate b u f f e r (0.1 M). Binding s t u d i e s were performed as described in d e t a i l e a r l i e r (3, 4) and the number and a f f i n i t y of binding s i t e s were c a l c u l a t e d by Scatchard anal y s i s using the MLAB program.

R e s u l t s and D i s c u s s i o n Displacement s t u d i e s revealed the presence of IGF-1 and IGF-2 binding 125 s i t e s on b r a i n membranes ( 3 ) . In the youngest f e t a l group, I - I G F - 1 was p r e f e r e n t i a l l y displaced by IGF-2 whereas i n the 17-22 week g e s t a t i o n a l age group IGF-1 and IGF-2 are equipotent. A f t e r 22 weeks of g e s t a t i o n a l

age

however, IGF-1 was more potent than IGF-2 and the order of c r o s s r e a c t i o n ,k I-iof-I TOTAL BOUND

IS.»-/.

Hormone Concentration (ng/ml)

125 Figure 1. Displacement of I - I G F - 1 from a d u l t human b r a i n membranes (660 pg membrane protein/ml) by d i f f e r e n t concentrations of unlabelled IGF-1, IGF-2, MSA ( C o l l a b o r a t i v e Res. I n c ) , i n s u l i n and p r o i n s u l i n . No d i s placement of T25j_jqp_i w a s observed with nerve growth f a c t o r (NGF), somatomedin B (SMB), growth hormone (hGH) or t h y r o i d hormones (T3, T4). Total s p e c i f i c binding using an excess of 1 yg IGF-1 /ml was 13.4°2. Membranes were prepared from f r o n t a l lobe biopsy m a t e r i a l .

547 was identical to that found in the adult brain (4). Figure 1 shows the dis125 placement of

I-IGF-1 from adult brain membranes by increasing concentra-

tion of IGF-1, IGF-2, MSA, insulin and proinsulin. 125 Other hormones and I-IGF-1 from the brain

growth factors did not show any displacement of

membranes. These results suggest alterations in the characteristics of the brain IGF-1 receptor during development. This is apparent from the Scatchard analysis (Table 1) which shows a higher concentration of a lower affinity IGF-1 binding site prior to 17 weeks gestational age. With advancing maturation however, a higher affinity IGF-1 binding site appears. The presence of an IGF-2 receptor early1?5in development1?5is suggested by the preferential displacement of both I-IGF-1 and I-IGF-2 by IGF-2 in 125 the youngest fetal group (3). In spite of this, specific I-IGF-2 bin125 ding was lower than specific I-IGF-1 binding leading us to suspect that iodination had altered the binding region of IGF-2 and invalidating Scatchard analysis (3). Table 1. Calculated affinity constant and concentration of IGF-1 binding sites on human brain plasma membranes. IGF-1 BINDING SITES Affinity constant (Mole-1)

Concentration (Mole/g)

4 0 0 1 - 4 0 0 5 .

(1 9 7 3 ) (1978)

(1974)

Nature 272, (1962)

J. Cell

356-358.

Expl.

Cell

P h y s i o l . 81_,

Proc. N a t l . Acad.

Sei.

A n z a n o , M. A . , R o b e r t s , A . B . , S m i t h , J . M . , Lamb, L . C . and S p o r n , M. B . (1982) A n a l . B i o c h e m . J_25, 21 7 - 2 2 4 . R o b e r t s , A . B . , A n z a n o , M. A . , Lamb, L . C . , S m i t h , J . M. and S p o r n , M. B . (1981) P r o c . N a t l . A c a d . S e i . USA 7 8 , 5 3 3 9 - 5 3 4 3 . A n z a n o , M. A . , R o b e r t s , A . B . , M e y e r s , C. A . , K o m o r i y a , A . , Lamb, L . C . , S m i t h , J . M . , and S p o r n , M. B. (1982) Cancer Res. 42, 4776-4778. A n z a n o , M. A .

Personal

communication.

IMMUNOPEROXIDASE GROWTH FACTOR-I

J. B e n n i n g t o n , Children's K.

LOCALIZATION CONTAINING

E. M.

OF

INSULIN-LIKE

TISSUES

Spencer

Hospital,

San

Francisco

Reber+

Hoffmann-LaRoche,

Basel,

Switzerland

Introduction Insulin-like of

animal

those

growth.

that

portance liver

All

the

kidney, creas, IGFs

lung,

are

6).

the

and

to

that of

their

derived

in

tissues

entire

site

range

In some

from

for IGFs

of

are

Insulin-Like Growth Factors / Somatomedins © 1983 Walter de Gruyter & Co., Berlin • New York

and

primary

im-

regulation.

of

IGF

known

there

do

tissues,

is e v i d e n c e

normally

tissues

other

gland,

neoplasms

not

(1,2).

but

mesenchymal

thought

+Deceased

IGFs

growth

not

malignant

target

of

salivary

addition,

regulators

produce

synthesis

muscle,

normally

major

are

in

include

tissues

that

that

interested

brain,

are

stimulation

production

(3).

(IGFs)

tissues

produce

gut,

testis

from

The

those

primary

produced

neoplasms IGF

is

sites

and

factors

investigators

tissues

suggested

Thus

respond

to

The

growth

panthat

including to

produce

susceptible

produce IGF

(4-

to

IGF

564 stimulation

is not

known

the m u s c u l o s k e l e t a l plicating

tissues

portance

of

IGFs

the

at

but

certainly

is b r o a d e r

s y s t e m and p r o b a b l y and

many

identifying cellular

includes most

neoplasms.

sites level,

than

Because

of

production

we

have

of

and

just

rethe

im-

action

employed

an

of

immuno-

h i s t o c h e m i c a l m e t h o d for the l o c a l i z a t i o n of IGF I.

The m e t h o d we used for these s t u d i e s was the antiperoxidase primary

antibody

(bridging

or

globulin; radish

(PAP)

immunoperoxidase

was

rabb it

linking)

and

the

peroxidase

bridging

body

antibody

and the

tissue tion

of of

(which

a

can

chromogenic hydrogen

cellular

be

hydrogen

in i d e n t i f y i n g

ther

refinements make

tween

cells

it

IGF-I

tissue

antirabbit

immuno-

rabbit

anti-horse-

peroxidase.

antigen

with b o t h

the

by

by d e v e l o p i n g

primary

the

the

various

technique

in

the

will and

synthesis

to of

t i s s u e cells that b i n d but do not s y n t h e s i z e

with

(DAB)

distinguish

IGF.

and

and

fur-

necessary

IGF

a

success-

however,

be

in

complex

were

tissues;

anti-

demonstra-

sections

studies

and

presence

diaminobenzadine

preliminary in

the

The

(IGF-I)

The

is e s t a b l i s h e d

semiquantitative

involved

was

second

peroxidase.

donor, Our

of

sheep

The

the

peroxidase-antiperoxidase

visualized

ful

to

combines

(IGF-I)

peroxide).

order

the

with

method.

IGF-1;

to h o r s e r a d i s h

anti-horseradish

antigen

was

antibody

complexed

primary antibody reacts the

antihuman

antibody

third

staining

peroxidase-

in be-

target

565

Materials and Methods

The

growth

hormone

(lit/lit) used was Harbor, has

Maine.

bioassayable

obtained

In

negligible

deficient

this

serum

IGF-I

and

(7)

growth.

The heterozygote

and

exhibits

growth

mone

daily

trols

for

given

erozygous

six

saline

of

same

time

and

and

given

with

initial weight of 25 grams. the

GH,

very

by

administration

For

this 5 ug

Normal

normal All

processed

tissue

low

retarded stimulates

levels

study of

litter-mate

injections.

litter-mates

of

Bar

homozygote

characterized

serum

were

days

mouse

(lit/+) mouse has adequate growth

growth.

homozygous Little mice

the

levels

is

Little

Laboratories,

of mouse

hormone

normal

normal

Jackson

tissue

and

However,

production,

from

strain

growth.

hormone

homozygous

of

IGF

15-16

gram

growth

hor-

homozygous

con-

rat

controls

phenotype

were

which

het-

had

an

animals were sacrificed

at

histologically

and

immuno-

chemically in the same batches.

Human

tissues

were

obtained

from

surgical

All mouse and human tissues were

fixed

specimens.

in neutral

buffered

formalin, embedded in paraffin, and cut at 6 to 10 wm.

The human IGF-I used for immunization was prepared Cohn Roche,

Fraction Basel

IV (8)

by

Ritschard

and

who

estimated

a

Roncari,

purity

of

from

Hoffmann-Laat

least

90%

566

based on physiochemical and biologic methods. was established also indicated M.

by N-terminal

embryo

Cal

Tech).

fibroblasts

Bioactivity

agreed

in

human

IGF-I,

(8).

The IGF-I antibody had less than human

which

the

primary

IGF-II

and

antibody,

rat

stimulating

with that for pure IGF-I

determined by Rinderknecht and Humbel (9).

with

analysis

a minimum purity of 905& (kindly performed by

Hunkapiller,

chicken

microsequence

Its identity

IGF-II

was

The rabbit antiprepared

by

3% cross

(BRL-MSA

Reber

reactivity

kindly

supplied

by Nissley) but a 30-50% cross reactivity with rat IGF-I/SMC

(kindly

supplied

by Daughaday).

The second

linking) antibody used was sheep anti-rabbit (Dako).

The

soluble rabbit

plex was obtained modification

of

(bridging or

immunoglobulin

antiperoxidase-peroxidase

from Immulok. the procedure

The method of

followed

Sternberger,

et

comwas a

al.

Controls for the immunoperoxidase technique included

(10)

1) the

use of rabbit non-immune serum in place of the primary antibody, 2) use of a rabbit antibovine serum albumin of the primary second

antibody,

antibody,

and

4)

3) elimination incubation

of the

with

the

in

place

linking PAP

or

complex

alone.

Results

IGF-I was successfully localized in human and mouse tissues by the PAP immunohistochemical method.

Since the con-

trols for all tissues staining positive for IGF-I were nega-

tive, the staining was presumed to be specific for IGF-I although IGF-I

controls were

not

specificity. hepatoma

done. 1)

specific

However,

Normal

scattered

articular

absorption

further

hepatocytes

cells were negative.

mice showed and

involving

positive

growth hormone deficient

the

evidence

stained

strongly

in the tibial

for

Little mice

anti-

supported

2) While heterozygous

chrondrocytes

cartilage

of

but

Little

epiphysis

IGF-I, the homozygous, showed

no

evidence

of

staining for IGF-I in the tibial epiphysis or articular

car-

tilage.

with

However,

growth hormone physis

and

for

in

mice

cartilage

stained

treated

of the tibial epi-

strongly

for

IGF-I.

frontispiece).

Subsequently tal, infant

Little

six days chondrocytes

articular

(Fig. 1 - see

homozygous

and

in the tissues

studi es were done on a var iety of human adult

tissues.

A

sampled by age are

summary

of

the

fe-

findings

indicated in Table 1.

Conclus ions

This

study

demonstration tissues IGF-I

in

represents of

cellularly

the mouse

interaction

the

and

at

the

first

localized

human

and

tibial

immunohistochemical IGF-I

in

various

growth

hormone

epiphysis

and

induced

articular

cartilage of the mouse.

The

peroxidase-antiperoxidase

immunohistochemical

(PAP)

568 technique lizing

appears

at the

sues.

As

result

of

IGF-I.

allow

is

Liske

be

to

and

Reber

an

indirect of

these

the

be

of

of

uses

manyfold.

It

subcellular in

of

mice

staining

and/or

binding

technique

and

provide

should

semiquan-

are

at v a r i a n c e

using

the

same

antibody or

muscle

can

be

PAP

and

identifying

used

to

antibody

suited

for

the

of

insulin-like

but

showed

(4).

cannot

We

The

difference

sensitivities

procedure IGF-I

in the both

is a n t i c i p a t e d those

tissues

elucidate

above two p r o c e s s e s . ly

those

technique

satisfactorily. relative

with

of

the

two

IGFs

are

cellular

and

study

at

the

to be of

which

of

considerable

produce

insulin-

like g r o w t h factors and those w h i c h serve as t a r g e t It

tis-

content.

lung

localize

levels

and

loca-

procedures.

the

can

the

for

positive

the

made

fluorescent

differences

a function

The

who

liver,

human

production

tissues

(11)

in

constituted,

refinements

in h u m a n

immunohistochemical

value

IGF-I

cellular

distinction

staining

explain may

sensitive method

a n a l y s i s of IGF c e l l u l a r

employing no

level

from

Further

Our r e s u l t s of

a highly

presently

either

this

titative

cellular

it

can

to be

the

factors

F i n a l l y , this t e c h n i q u e

demonstration

growth

e m b r y o g e n e s i s and fetal

factors

of

the

in

various

maturation.

time

tissues.

regulating appears of

the

ideal-

appearance

tissues

during

569 Table 1 TISSUES

SAMPLED

Human

Mouse

Kidney Fetal

(1st

Trimester)

N.S.

Fetal

(Mid

Trimester)

N.S. +

Adult

(tubules)

+

(tubules)

Liver Fetal

(1st

Trimester)

N.S.

Fetal

(Mid

Trimester)

N.S. +

Adult Salivary

Gland

+

(ducts)

N.S.

E n d o m e t r ium N.S.

Non-Pregnant Pregnant

+

(decidua)

Pancreas

+

(islets)

Thyroid

0

0

Muscle

+

+

+

+

N.S. +

Smooth Striated Cartilage Infant Adult

N.S. - not

sampled

(islets)

570 Acknowledgement Support for these studies was provided by the West Coast Cancer Foundation and the National Institutes of Health HD 14506.

References 1. McConaghey, R., Sledge, L.H.: (1970) 2.

Spencer, E.M.:

Nature 225 , 1294-1250

FEBS Letters 99 , 157-163

(1979)

3. D'Ercole, A.J., Applewhite, G.T., Underwood, L.E.:Dev. Biol. 75 , 315 (1980) 4.

DeLarco, J.E., Todaro, G.J.:

Nature 272 , 356-358

(1978)

5. Knauer, D.J., Iyer, A.P., Banerjee, M.R., Smith, G.L.: Cancer Res. 40 , 4368-4372 (1980) 6. Baxter, R.C., Maitland, J.E., Raison, R.L., Reddel, R.R., Sutherland, R.L.: This volume (1983) 7. Nissley, S.P., Knazek, R.A., Wolff, G.L.: Res. J_2 , 158-164 ( 1980) 8.

Reber, K., Liske, R.:

Hormone Res. 7 , 201-214

9. Rinderknecht, E., Humble, R.E.: 2769-2776 (1978) 10. Sternberger, L.A., Joseph, S.A.: Cytochem. 27 ,1424-1429 (1979) 11. Liske, R., Reber, K.:

Horm. Metab. (1976)

J. Biol. Chem. 253 , J. Histochem.

Hormone Res. 7 ,215-217 (1976)

PRODUCTION OF INSULIN-LIKE GROWTH FACTORS (IGFs) AND THEIR BINDINC PROTEINS (IGF BPs) BY THE PITUITARY GLAND AND THE NERVOUS TISSUE IN CULTURE. M. B i n o u x ^ P. Hossenlopp + , C. Lassarre + , A. Barret + + , A. Faivre-Bauman ++ , C. Loudes and A. Tixier-Vidal +

INSERM U 14-2, Hôpital Trousseau, ++ Lab. Neuroendocrinologie Cellulaire, Collège de France, Paris, France. Two years ago, we reported that expiants from pituitary gland and

various cerebral tissues of young rats release in the culture medium IGFs and their BPs (1). Here we report new results obtained with dissociated hypothalamic and cerebral cells of mouse fetuses. Moreover, a physicochemical characterization of the |IGF-BP| complexes has been undertaken. METHODS 1. Organ culture (see Réf. 1) Expiants of the various organs excised from if-6 weeks old rats were cultured in Mc Coy's 5a medium without serum (A- hemi-anterior pituitary lobes; 2 neuro-intermediate lobes; 4 hemihypothalamus per dish in 2 ml medium). 24 hours later, the medium was discarded and replaced, then the culture was allowed to developfor3 more days. The media from each set of expiants were then pooled, desalted through Sephadex G 25 columns and put aside in a lyophilized state for subsequent studies. 2. Dissociated cell culture

The technique was previously described

(2). Hypothalamus and cerebral hemispheres were excised from mouse fetuses on the 16 th day of gestation and mechanicallydissociated. The cells were cultured in the "N2" synthetic medium of Bottenstein and Sato (3) supplemented with 1 0 - 1 2 M 17

(3-estradiol (^ 1.5 x 10 6

cells per dish in 2ml

medium). At day 5, the medium was renewed. The media of the next three days were collected, desalted and lyophilized. The evolution of the cell

culture has been described elsewhere (2).

After a week, we observe a basal layer of flat and pale cells, mainly astrocytic; over this basal layer, one can see clusters of small dense cells and many refringent cells displaying neuron-like features at different stages of differentiation (Fig. 1). Neurites are clearly visualized using tetanus toxin-binding and immunofluorescence.Thyroliberin and neurotransmitters are synthesized, and synapses are formed during the second week (4).

Insulin-Like Growth Factors / Somatomedins © 1983 Walter de Gruyter & Co., Berlin • New York

572 3. IGF and BP radioligand assays (see R e f. 1) The lyophilized samples corresponding to 5-10 ml culture medium were gel filtered on Ultrogel AcA5^ in 1 M acetic acid in order to separate IGFs and their BPs. IGF content was determined by a competitive protein-binding assay using specific BPs produced by rat liver in culture and IGF I (generous gift of Dr. Humbel, Zurich) as tracer. The BP content was estimated by titration, after incubation with

i 2 5 l IGF I, in comparison with a reference prepa-

ration (BPs extracted from rat serum). Results were expressed in serum units (with respect to a rat serum pool with an assigned potency of 1 U IGF and 1 U BP per ml).

Fig. 1 Morphological effect of T3. Fetal hypothalamic cells were grown for 8 days in regular serum-free medium (a) or in presence of T3 10~9 M (b). Living cells, photographed in phase contrast microscopy (x 180). The physico-chemical characterization of the

IGF-BP complexes

was performed, after incubation of the culture medium with using three different methods : - gel filtration on Ultrogel AcA5^ at pH 7.4-,

12 si IGF I,

573 - sedimentation on a 4-15% sucrose gradient at pH 7.4-^ - electrophoresis (SDS-PAGE) after covalent binding of the

IGF-BP com-

plexes with dimethylsuberimidate. RESULTS 1. Results of IGF and BP measurements in various culture media are summarized in table I. By way of comparison are shown those previously reported for the rat liver (5). The concentrations of IGF in the culture media from mouse or rat were in the same range. The concentrations of BP always exceeded those of IGF. The concentrations of IGF and BP per ml of medium were much lower than in the serum, but related to the total amount of proteins, these concentrations were much greater (this has not been checked in the case of cell culture because of the large amounts of insulin and transferrin added to the medium).

CULTURE

MOUSE (cell cult.)

IGF

Hypothalamus (ft exper.)

«DIUM BP

mU/ml

mU/ml

5.8

93

-v. 3.0

98

provins mU/mg

proteins mU/mg

Cerebral hemisph. (2 exper.)

RAT (organ culture)

Adenohypophysis (10 exper.)

3.20 + 0.63

18.5 ± 2.1

21 ± 3.8

128 ± 1ft

Neurointerm. lobe (10 exper.) Hypothalamus (7 exper.) Liver (7 exper.)

Rat serum

2.38 ± 0.73

ft.99 + 0.57 162 ± 33

2.30 + 0.50

3.1ft ± 0.58

ft.7 ± 0.55

1000

|

50 ± 10

1000

(arbitrary ref. values)

Table I. Production

"in vitro" of IGF and IGF BP.

37 ± 9

378 ± 55

53 ± 1ft

91 ± 11 1026 ± 2ft8

•v. 1ft

1ft

574 2. In view of the well known effect of thyroid hormones on nervous maturation processes, experiments were done with addition of Ts (10~ 12 to 1CT9 M) to the culture medium of hypothalamic cells. There was an increase in the neuronal cell body size and an enhaucement of the neurite length and arborization (Fig. 1). This effect could be measured using morphometric analysis (6). In contrast, the number of astrocytic cells was decreased (6). The release of IGF was

stimulated in a dose-dependent manner, rea-

ching more than twice the control values at 1 nanomolar concentration of T3. This stimulatory effect was observed in three different experiments. No significant variations of the BPs were seen. 3. Gel filtration studies. In contrast with the elution profile of the liver |IGF-BP| complex which forms a single peak with an apparent mol. wt. % W

K, those synthesized by the pituitary and the nervous tissues

eluted in a wide asymetrical peak, the heterogeneity of which was demonstrated by rechromatography of the two parts of the peak (1). In some expe riments the heterogeneity was obvious showing two peaks with an apparent mol. wt. of % 53 and 38 K. 4. On sucrose gradient¿the sedimentation coefficient of the 53 K pituitary

|IGF-BP | complex was estimated to be 3.3 s. That of the 38 K

complex was 3.0 s (The liver |IGF-BP| complex sedimented at 2.9 s). 5. analysis

af the

|IGF-BP | complexes by SDS-PAGE confirmed their

heterogeneity. An example of migration patterns is shown on Fig. 3. One can

see for most of the preparations two zones of migration around 40 K

and 50 K. It is worth nothing that the small differences between the migration profiles were reproducible

in several experiments using different

culture media and therefore seem to be characteristic of a given tissue. Moreover the 40 K region is heterogeneous and contains at least two bands (this is true also in the case of the liver). The specificity of the binding was proved by addition of cold IGF before cross-linking. CONCLUSIONS The synthesis of IGFs and BPs by pituitary and nervous cells, and the regulatory effect of

on hypothalamic cells, suggest a local role

for these components in the growth and/or the maintenance of cerebral tis sues.

575 Determination of the types of IGF released into culture media remains to be clarified. The BPs of pituitary and nervous origin have physical properties similar but not identical to those of liver BPs. Isolation of the various BPs will be necessary to determine their relative affinities for IGF I and for IGF II.

RAT (organ culture)

MOUSE (cell culture)

F i g . 2 Autoradiography of the | 1 2 5 J IGF I-BP | complexes separated on SDSPAGE (5-17% gradient). The culture media were incubated overnight at 4°C with 1 2 5 J IGF I, then with dimethylsuberimidate for 3 h at 20°C, and ana1 lyzed by slab gel electrophoresis (Laemmli). fe proteins were used as markers.

REFERENCES 1. Binoux, M., Hossenlopp, P., Lassarre, C., and Hardouin, s . ; FEBS Letters 124, 178-183 0.981). 2. Faivre-Bauman, A., Rosenbaum, E., Puymirat, 3., Grouselle, D., Tixier-Vidal, A.: Develomental Neurosciences 118-129 (1981). 3. Bottenstein, 3.E., Sato, G.H.: Proc. Natl. Acad. Sci 76, 514-517 (1979) Puymirat, 3., Loudes, C., Faivre-Bauman, A., Bourre, J.M., Tixier-Vidcl In : Growth of cells in hormonally defined media (Sirbatsku D. & SatoG. ed.). Cold Spring Harbor conferences on cell proliferation 9, 10331051 (1982).

576 5. Binoux, M., Lassarre, C. , Hardouin, S. : Acta Endocrinol. 99, 4-22-MO (1982). 6. Puymirat, 3., Barrett, A., Picart, R., Vigny, A., Loudes, C., FaivreBauman, A., Tixier-Vidal, A.: Neuroscience, in press, (1983).

MONOCLONAL ANTIBODIES THAT INHIBIT THE SULPHATION ACTIVITY OF HUMAN SERUM

David C. Watkins and Michael Wallis School of Biological Sciences, University of Sussex, Falmer, Brighton, BN1 9QG, England. Juraj Ivanyi Department of Experimental Immunobiology, Wellcome Research Laboratories, Langley Court, Beckenham, Kent, BR3 3BS, England.

Introduction Monoclonal antibodies (McAb) produced by the spleen cell fusion technique of Kohler and Milstein (1) have the advantage, compared with conventional antisera, that highly specific antibodies can be produced using relatively impure antigens (e.g. ref.2). Pure preparations of somatomedins (Sm), suitable for the preparation of conventional antisera are not widely available and so the possibility of raising McAb to partially purified Sm was investigated using ability to block sulphation factor activity in human serum as an assay for anti-Sm activity.

An alternative approach has been described

recently (3), in which a McAb to human Sm C was identified on the basis of 125 ability to bind the I-labelled peptide.

Materials and Methods Culture medium (RPMI/FCS), HAT selective medium and HT medium was as described in ref.l, except that RPMI 1640 was used instead of DMEM. were performed as described in ref.l.

Fusions

Mice (C57-BL-6J strain) were immu-

nized with a partially pure preparation of human Sm C (12 U/mg; obtained from Dr. A.T. Holder of the Institute of Child Health, University of London). Spleen cells were harvested 4 days after a final intraperitoneal injection 7 and fused with 10 myeloma cells (P3-NSl/l-Ag4-l line) using polyethylene glycol. Samples of approximately 10® cells were then cultured in wells of

Insulin-Like Growth Factors/Somatomedins © 1983 Walter de Gruyter & Co., Berlin • New York

578

tissue culture plates containing 2 ml HAT medium.

Media from wells contain-

ing macroscopically visible colonies were assayed for anti-Sm activity. Colonies from positive wells were separated and cultured in HT medium and later RPMI/FCS medium and cloned by the technique of limiting dilution (4). The presence of anti-Sm was detected by its ability to lower the apparent bioassayable Sm in human serum using the porcine costal cartilage assay system of Spencer and Taylor (5), except that Tris-HCl buffer supplemented with amino acids (Tris-aa; 6) was used.

Culture medium from wells to be 35 tested was added to Tris-aa, then serum (20%) and Na^ SO^ ( 4 nCi/ml) were added.

The results of the bioassays are expressed as the percentage of

label which is incorporated into the cartilage discs and all values represent the mean (± S.E.M.) of 4 or 6 discs. The number of hybridoma cells per well may vary and consequently deplete nutrients to different extents, which may in turn affect the level of silphation factor activity measured.

To overcome this two or more concentra-

tions of test medium were compared for their effects on sulphation factor activity.

Media were only considered to contain anti-Sm activity if (1)

their presence caused a reduction in incorporation of label compared with control (in which the test medium was replaced with RPMI/FCS medium) and (2) the higher dose level gave less apparent sulphation factor activity than the lower dose (Fig.l).

Results From a single fusion 23 colonies of hybridoma cells producing antibodies which blocked sulphation factor activity were obtained. and stored in liquid N .

These were frozen

During cloning of these lines it was necessary to

take colonies growing at limiting dilution (4) and to grow substantial numbers of cells in larger wells (2 ml) in order to provide sufficient medium for testing in the bioassay.

Of 5 of these lines investigated to date, only

one (F7/47-B1) retained anti-Somatomedin-like activity (anti-SmLA) after cloning.

F7/47-B1 was shown to represent a homogeneous population after a

second cloning step and was subsequently grown in larger amounts and the

579

o

4I< -J CC LU o 5 3-

DC

F7/47-B1

RPMI/FCS

f

•f

-1

O li. o o

2 1-

10

20

40

10

20

40

CONC. MEDIUM(%)

Fig.l. Effect of several dose levels of culture medium from hybridoma line F7/47-B1 and the equivalent doses of RPMI/FCS medium on the incorporation of 35-S-sulphate in the presence of 20% human serum. *indicates significant difference from effect with 10% F7/47-B1 (P