BIOLOGICAL ACTIONS OF HUMAN SOMATOTROPHIN AND ITS DERIVATIVES ON MOUSE MAMMARY GLAND AND TELEOST URINARY BLADDER B. A.

DONEEN,

H. A. BERN AND CHOH HAO LI

Department ofZoology and Cancer Research Laboratory, University of California, Berkeley, California 94720, and ^Hormone Research Laboratory, University of California, San Francisco, California 94143, U.S.A. *

(Received 7 September 1976) SUMMARY

These studies are concerned with the structural and functional evolution of the ancestrally related pituitary prolactins and somatotrophins. Prolactin-like biological activities of human somatotrophin (hGH) and its peptide fragments were bioassayed in vitro on the mouse mammary gland and the teleost urinary bladder. Plasmin modified-hGH was as active as hGH in both bioassays. The NH2-terminal 134-residue fragment possessed about 10 % of the lactogenic and urinary bladder potency of hGH, whereas the CO2H-terminal 51-residue fragment was inactive at the concentrations observed. These results suggest that the same regions of primary structure are responsible for the prolactin-like actions of hGH on the target organs of lower and higher vertebrates. Alteration of the tertiary structure of hGH, human chorionic somatomammotrophin, and ovine prolactin by performic acid oxidation destroys the mammary gland activities of these hormones.

INTRODUCTION

somatotrophin (hGH) is known to possess lactogenic and luteotrophic actions (Forsyth, Folley & Chadwick, 1965; Lyons & Dixon, 1966; Li, 1972). It also displays prolactin-like activities in lower vertebrates: in promotion of larval amphibian growth (Clarke, Bern, Li & Cohen, 1973) and in teleost osmoregulation (Pickford, Robertson & Sawyer, 1965; Clarke et al. 1973; Doñeen, 1976). Its potency on crop-sac and mammary gland and on prolactin target organs in lower vertebrates distinguishes hGH from other mammalian somatotrophins which, however, share the somatotrophic property of hGH. It has been proposed that separate structural features specify the prolactin-like and growth-promoting activities of hGH (Sherwood, Handwerger & McLaurin, 1972; Kostyo, 1974). The unequal loss of crop-sac and rat tibial activities shown by hGH altered at its tryptophan residue, supports this hypothesis (Bewley, Brovetto-Cruz & Li, 1969; Brovetto-Cruz & Li, 1969). However, comparison of the amino acid sequences of homologous hormones hGH, human chorionic somatomammotrophin (HCS), bovine growth hormone (BGH) and ovine prolactin does not unambiguously disclose mutually exclusive sites defining each type of activity (Bewley, Dixon & Li, 1972). Moreover, Li & Graf (1974) showed that the structural requirements for both rat tibial activity and crop-sac function are contained in the NH2-terminal 134 amino acid residues of hGH. A smaller C02H-terminal fragment (residues 141-191) was half as active as the NH2-terminal fragment on a weight basis (Li & Graf, 1974). Human

The prolactin activity of hGH contrasts with inactivity or weak activity of its homologous HCS in poikilotherms (Clarke et al. 1973; Doñeen, 1976). Inasmuch as HCS, like hGH, has activity on mammary gland and crop-sac (Josimovich & MacLaren, 1962; Handwerger, Pang, Aloj & Sherwood, 1972), prolactin target tissue (receptor) interactions in lower vertebrates may be specified by structural features different from those bestowing activity in homeotherms (Bern & Nicoli, 1968; Ensor & Ball, 1968). This paper reports the lactogenic and teleost osmoregulatory activities of hGH in vitro and its peptide fragments derived from digestion with human plasmin. By comparing the activities of hGH fragments on mouse mammary gland and on Gillichthys urinary bladder with those previously observed in the crop-sac (Li & Graf, 1974), definition of some structural features conferring prolactin action to hGH may be possible. The extent to which segments of hGH primary structure specifying prolactin function can be distinguished from those sufficient for somatotrophic action can also be assessed. In addition, the effect of abolishing the native tertiary structure of the lactogenic hormones on mammary gland activity was judged by comparing the potencies of performic acid oxidized-hGH, -HCS, and -ovine prolactin with that of native hGH. MATERIALS AND METHODS

HGH and its fragments Human GH was isolated

by the method of Li, Liu & Dixon (1962). Plasmin modified-hGH, Cys (Cam)53 [carbamidomethylated]-hGH-(l-134) and Cys (Cam)1«5·182· 189-hGH-(141-191) were prepared as described earlier (Li & Graf, 1974). Performic acid oxidation of hGH, HCS and ovine prolactin was performed essentially as described by Li (1957 a) for ovine prolactin. Hormone or hormone fragment concentrations were calculated from careful determinations of weight (Cahn M-10 Electrobalance), and were uncorrected for water content. Proteins were dissolved in 001 M-NaOH and diluted with culture

bioassay (see below)

medium for

Bioassays

Methods for measuring the lactogenic activity of hormones on midpregnant mouse (BALB/c) mammary gland have been described (Juergens, Stockdale, Topper & Elias, 1965; Doñeen, 1976). Briefly, the dose-dependent incorporation of 3H-labelled amino acids into casein-like protein (rennin + calcium-precipitable) was determined after explantation of tissues for 3 days in Waymouth's medium supplemented with crystalline insulin (5 /¿g/ml) and cortisol (1 /ig/ml) and with graded concentrations of the preparation being assayed. Teleost urinary bladder water permeability was estimated from the reduction of osmotically driven net mucosal to serosal water flux in bladders organ-cultured for 3 days in Medium 199 containing 1 /¿g cortisol/ml and various concentrations of hGH or hGH fragments; water movement was determined gravimetrically ( % weight loss/h). Bladders for culture were dissected from Gillichthys mirabilis 1 day after transfer of the fish from sea water to 10% sea water

(Doñeen, 1976).

Statistical analysis

Data are presented as the mean ± standard error. Bioassay results were analysed for variance, non-parallelism, curvature and opposed curvature according to Finney(1964). Experimental means were compared with control values by Dunnett's multiple comparison test (twosided; Dunnett, 1970). To determine whether responses to hGH fragments were significantly different from standard (hGH or plasmin modified-hGH) Duncan's multiple range test (Dunnett, 1970) was used. Where appropriate, tables of results include the bioassay index of precision ( ) and the geometric mean potency relative to hGH or plasmin modified-hGH. Potencies were calculated for preparations showing statistically significant responses and

insignificant non-parallelism to standard. Potencies of materials not meeting these criteria were defined as 000. Computer calculations were performed by the EXBIOL program (Sakiz, not dated). RESULTS

Plasmin modified-hGH, consisting of residues 1-134 and 141-191 joined by the Cys53-Cys165 disulphide bond, exhibited mammary lactogenic activity equivalent to that of intact hGH (Table 1). The calculated geometric mean potency of plasmin modified-hGH relative to hGH was 87 % (Table 1). However, mammary gland responses to plasmin modified-hGH and hGH were not significantly different (at 005; Duncan's test) in this assay. In three additional bioassays, plasmin modified-hGH again displayed lactogenic activity similar to that of hGH (data not shown). Thus, within the precision of the mammary gland bioassay ( 0-20-0-35), plasmin modified-hGH and hGH appeared to have equal potency. =

of human somatotrophin (hGH), plasmin modified-hGH (PL-hGH), and fragments of hGH

Table 1. Lactogenic activities

Geometric mean

Response IO-3 Concentration c.p.m./mg protein (/ig/ml) (Mean + s.E.M.)

Peptide Control hGH

4-14 + 0-87 6-67 ±0-92 10-35 + 0-91**

0075 0-225 0-675 0075 0-225

PL-hGH

Cys (Cam)165-

182·

18e-hGH-(141-191)

relative to hGH

95

% confidence limits

100

14-54±l-20**

6-12±0-65

0-87

0-52-1-44

003

002-005

1113± 1 06** 12-84 + 0-75** 700 ±0-94 10-40 + 0-86** 12-73 + 0-73** 5-28 ±0-62 6-31+0-91 5-52 + 0-74

0-675

Cys(Cam)53-hGH-(l-134)

potency

0-675 2025 6075 2025

6-075 18-225

000

0-25. Four replicates in each group (n 4); ** < 001 ; Dunnett's test, compared with control value. =

=

reduced but significant activity on the mammary 3 from (Table 1) to 10 % of that of hGH and plasmin potency gland, exhibiting modified-hGH. Moreover, the response of the mammary gland to hGH-(l-134) paralleled those obtained with hGH and plasmin modified-hGH. In contrast, high concentrations of the C02H-terminal fragment did not show statistically significant lactogenic action (Table 1). Unlike native hGH, HCS and ovine prolactin (Doñeen, 1976), the performic acid-oxidized hormones did not stimulate mammary casein synthesis, even when concentrations exceeding the minimal effective hGH concentration by 50-fold were tested (Table 2). As with the mammary gland assay, plasmin modified-hGH was as effective as hGH in lowering the water permeability of Gillichthys urinary bladder (Table 3). Moreover, the teleost bladder potency of the NH2-terminal fragment (9 % of hGH, Table 3), was similar to its relative potency on the mammary gland (Table 1). Finally, the teleost bladder, again like the mammary gland, failed to respond to the C02H-terminal fragment (141-191) at the one concentration studied (Table 4). The

NH2-terminal fragment displayed a mean

Table 2. Lactogenic activities of human somatotrophin (hGH) and performic acid-oxidized (PO)-hGH, human chorionicsomatomammotrophin(PO-HCS), and ovine prolactin (PO-oPRL)

Response 10~3 c.p.m./mg protein (Mean + s.E.M.)

Concentration

Peptide

(,"g/ml)

Control hGH

PO-HCS PO-oPRL

3;

=

NS

203 ±0-36 4-5810-48

010 0-50 0-50 5-00 0-50 500 0-50 5 00

PO-hGH

=

005 001 NS NS NS NS NS NS

7-25 + 0-56 2-15 + 0-49 2-04 + 0-23 219 + 019

2-11+0-28 2-38 + 0-34 2-22 ± 0-29

0-31.

=

not

value

(Dunnett's test)

significantly different (

>

005) from controls.

Gillichthys urinary bladder activities of human somatotrophin (hGH), plasmin modified-hGH (PL-hGH) and ^-terminal fragment of hGH

Table 3.

Concen¬ tration

0ig/ml)

Peptide

Response

(% weight loss/h)

Potency

26-4 + 0-2 21-5 + 2-8 17-8 + 3-3

100

Control hGH

2-5 7-5 22-5 2-5 7-5 22-5 7-5 22-5 67-5

PL-hGH

Cys (Cam)53-hGH-(l-134) „ *

Table 4.

=

4; =S

**

^

% confidence limits

11-1+2-3**

18-3 + 1-2 15-2 + 2-9*

10-7+1-3** 20-9+3-5

1-81

0-50-11-80

009

002-0-35

16-9 + 2-1 14-2 + 2-8*

-0-44.

=

005;

95

(Mean + s.E.M.)

001; Dunnett's

test

compared with control value.

Gillichthys urinary bladder activities of human somatotrophin (hGH) and its carboxyl terminal fragment Response

Concentration

G"g/ml)

Peptide

(% weight loss/h) (Mean + s.E.M.) 31-9 + 3-8

Control hGH

Cys (Cam)165·182· 189-hGH-(141-191) -0-37 / 4; =

**

100 400 400

18-1+2-2** 12-3 + 1-0**

31-2+1-9

=

001 ; Dunnett's test,

compared with control

value.

DISCUSSION

Organ-cultured mouse mammary gland and Gillichthys urinary bladder share sensitivity to mammalian prolactins and human somatotrophin (Doñeen, 1976). Because the measurement of the biological activities of hormone fragments and derivatives may be complicated by their variable half-lives in vivo these in-vitro bioassays appear especially useful for structurefunction investigations.

Plasmin modified-hGH, lacking residues 135-140, was as potent on the mouse mammary gland and teleost urinary bladder as native hGH (Tables 1 and 3). However, within the precision of these bioassays, and in contrast to the observations of Yadley, Rcdbard & Chramback (1973) and Singh, Seavey, Rice, Lindsey & Lewis (1974), plasmin modified-hGH did not show a statistically significant increase in activity with respect to hGH (see also Li & Graf, 1974; Reagen, Mills, Kostyo & Wilhelmi, 1975). The NH2-terminal 134-residue fragment of hGH had significant activity on the mamma¬ lian and teleostean target organs, exhibiting (on a weight basis) about 10 % of the potency of hGH and plasmin modified-hGH (Tables 1 and 3). Inasmuch as this fragment also showed 10-20% of the hGH and plasmin modified-hGH potency on the pigeon crop-sac (Li & Graf, 1974), the NH2-terminal two-thirds of its primary structure is apparently sufficient for these prolactin activities of hGH. Its reduced potency suggests that Cys (Cam)53-hGH(1-134) is deficient in structural features (primary, secondary or tertiary) necessary for full biological activity. The C02H-terminal fragment did not exhibit significant lactogenic (Table 1) or urinary bladder (Table 4) activities at the concentrations tested. Because Gillichthys bladder is rather insensitive to mammalian hormones (Doñeen, 1976) and because high concentrations of Cys (Cam)165·182· 18e-hGH-(141-191) possessed weak crop-sac and rat tibial activity (Li & Graf, 1974), higher concentrations should be tested to establish firmly that teleost bladder is insensitive to this fragment. Treatment with performic acid substantially alters the native structure of proteins. Performic acid cleaves disulphide bonds by oxidation and also oxidizes tryptophan and methionine residues (Hirs, 1956). Discrepant results have been reported for the effect of performic acid on the biological activities of lactogenic hormones. Performic acid oxidation of hGH destroys its somatotrophic activity in vivo (Trenkle, Li, Sadri & Robertson, 1962). Oxidized ovine prolactin lacks crop-sac activity (Li, 19576). These observations are in contrast with those of Handwerger et al. (1972) who reported full retention of mammary gland activity by performic acid-oxidized hGH and HCS in vivo and in vitro. The present investigations (Table 2) indicate a substantial, perhaps complete, loss of lactogenic activity by performic acid-treated hGH, HCS and ovine prolactin in vitro. Bewley et al. (1969) showed that reduction and alkylation of the disulphide bonds of hGH with either iodoacetamide or iodoacetic acid caused no loss of crop-sac activity. However, reduction and alkylation were also shown (Bewley et al. 1969) to produce only minor changes in the secondary and tertiary structure of hGH as measured by circular dichroism, spectrophotometric titration and viscosity. In contrast, total loss of the immunoreactivity and large changes in the circular dichroism spectra (Aloj, Edelhoch, Handwerger & Sher¬ wood, 1972; Handwerger et al. 1972) of performic acid-oxidized hGH and HCS, denote profound changes in the tertiary structures of these hormones. Thus cleavage of disulphide bonds by performic acid oxidation and by reduction and alkylation do not produce com¬ parable structural changes. Tryptophan and methionine residues are also altered by performic acid. As the integrity of the tryptophan residue of hGH appears to be important for its crop-sac activity (BrovettoCruz & Li, 1969), the loss of mammary gland action by oxidized hGH (and HCS and ovine prolactin) observed herein may be attributable to modification of this residue. Our results with the performic acid-treated hormones, in summary, do not confirm the idea that all disulphide bonds or native tertiary structure are dispensable for the mammary gland activity of hGH and other lactogenic hormones (Aloj et al. 1972; Handwerger et al. 1972). With two exceptions, hormones active in the mammary gland assay are also active in the fish bladder assay; these exceptions are the weak activity of HCS on the fish bladder and the inactivity of teleost prolactins on the mammary gland. The unequal lactogenic and teleost osmoregulatory potencies of HCS and fish prolactin suggested that target organs of lower

and higher vertebrates might recognize different structural aspects of these ancestrally related hormones (Nicoli, Bern & Brown, 1966; Bern & Nicoli, 1968; Clarke et al. 1973). However, the present study indicates parallel patterns of activity for hGH, plasmin modifiedhGH and Cys (Cam)53-hGH-(l-134) on mammary gland and Gillichthys bladder. Compared with hGH, the potency of the NH2-terminal fragment was similar in the mammalian and teleostean systems. Moreover, Cys (Cam)165· 182· 189-hGH-(141-191) was inactive in both bioassays. These data, therefore, do not support the contention that separate structural features of hGH are recognized by mammary gland and teleost bladder. These studies also show that fragments of hGH having significant somatotrophic potency, plasmin modifiedhGH and Cys (Cam)53-hGH-(l-l 34) (Li & Graf, 1974; Reagan et al. 1975), are also active on mammary gland and fish bladder as well as on pigeon crop-sac (Li & Graf, 1974). Therefore, the prolactin-like and growth-promoting properties of hGH have not yet been separated as

discrete peptides.

We thank Ms Karen T. Mills, Ms Ann Mos and Mr Jimmy Louie for expert assistance. Dr G. Shyamala and Dr H. Papkoff offered helpful suggestions. Computer time was provided by the Computer Center, University of California, Berkeley. This study was aided by NIH Grants CA-05388 and CA-05045. REFERENCES

Aloj, S. M., Edelhoch, H., Handwerger, S. & Sherwood, L. M. (1972). Correlations in the structure and function of human placental lactogen and human growth hormone. II. The effect of disulfide bond modification on the conformation of human placental lactogen. Endocrinology 91, 728-737. Bern, H. A. & Nicoli, C. S. (1968). The comparative endocrinology of prolactin. Recent Progress in Hormone Research 24, 681-720. Bewley, . ., Brovetto-Cruz, J. & Li, C. H. (1969). Human pituitary growth hormone. Physiochemical investigation of the native and reduced-alkylated protein. Biochemistry 8, 4701-4708. Bewley, . ., Dixon, J. S. & Li, C. H. (1972). Sequence comparison of human pituitary growth hormone, human chorionic somatomammotropin and ovine pituitary growth and lactogenic hormones. Inter¬ national Journal of Peptide and Protein Research 4, 281-287. Brovetto-Cruz, J. & Li, C. H. (1969). Human pituitary growth hormone. Studies of the tryptophan residue. Biochemistry 8, 4695-4701. Clarke, W. C, Bern, . ., Li, C. H. & Cohen, D. C. (1973). Somatotropic and sodium-retaining effects of human growth hormone and placental lactogen in lower vertebrates. Endocrinology 93, 960-964. Doñeen, . . (1976). Biological activities of mammalian and teleostean prolactins and growth hormones and of human placental lactogen in mouse mammary gland and in teleost urinary bladder. General and Comparative Endocrinology 30, 34-42. Dunnett, C. W. (1970). Multiple comparisons. In Statistics in Endocrinology, pp. 79-100. EdsJ. W. McArthur & T. Colton. Cambridge, Mass: MIT Press. Ensor, D. M. & Ball, J. N. (1968). A bioassay for fish prolactin. General and Comparative Endocrinology 11, 104-110. Finney, D. M. (1964). Statistical method in biological assay, edn. 2, London: Griffin. Forsyth, . ., Folley, S. J. & Chadwick, A. (1965). Lactogenic and pigeon crop-stimulating activities of human pituitary growth hormone preparations. Journal of Endocrinology 31, 115-126. Handwerger, S., Pang, E. C, Aloj, S. M. & Sherwood, L. M. (1972). Correlations in the structure and func¬ tion of human placental lactogen and human growth hormone. I. Modification of the disulfide bonds. Endocrinology 91, 721-727. Hirs, C. W. (1956). Oxidation of ribonuclease with performic acid. Journal of Biological Chemistry 219, 611-621. Josimovich, J. B. & MacLaren, J. A. (1962). Presence in the human placenta and term serum of a highly lactogenic substance immunologically related to pituitary growth hormone. Endocrinology 71, 209-220. Juergens, W. G., Stockdale, F. E., Topper, Y. J. & Elias, J. J. (1965). Hormone-dependent differentiation of mammary gland in vitro. Proceedings of the National Academy of Sciences of the U.S.A. 54, 629-634. Kostyo, J. L. (1974). The search for the active core of pituitary growth hormone. Metabolism 23, 885-899. Li, C. . (1957 a). Studies on pituitary lactogenic hormone. XVIII. Oxidation of the hormone with performic acid. Journal of Biological Chemistry 229, 157-183.

Li, C. H. (1957¿>). Hormones of the anterior pituitary gland. II. Melanocyte-stimulating and lactogenic hormones. Advances in Protein Chemistry 12, 269-317. Li, C. H. (1972). Recent knowledge of the chemistry of lactogenic hormones. In Lactogenic hormones, pp. 7-26. Eds G. E. W. Wolstenholme & J. Knight. London: Churchill Livingstone. Li, C. H. & Graf, L. (1974). Human pituitary growth hormone: Isolation and properties of two biologically active fragments from plasmin digests. Proceedings of the National Academy of Science of the U.S.A. 71, 1197-1201.

Li, C. H., Liu, W.- K. & Dixon, J. S. (1962). Human pituitary growth hormone. VI. Modified procedure for isolation and NH2-terminal acid sequence. Archives of Biochemistry and Biophysics Suppl. 1, 327-332. Lyons, W. R. & Dixon, J. S. (1966). The physiology and chemistry of the mammotropic hormone. In The pituitary gland, vol. 1, pp. 527-581. Eds G. W. Harris & B. T. Donovan. London: Butterworth. Nicoli, C. ., Bern, H. A. & Brown, D. (1966). Occurrence of mammotrophic activity (prolactin) in the vertebrate adenohypophysis. Journal of Endocrinology 43, 343-354. Pickford, G. E., Robertson, E. E. & Sawyer, W. H. (1965). Hypophysectomy, replacement therapy and the tolerance of the euryhaline killifish, Fundulus heteroclitus, to hypotonie media. General and Comparative Endocrinology 5, 160-180. Reagan, C. R., Mills, J. B., Kostyo, J. L. & Wilhelmi, A. E. (1975). Isolation and biological characterization of fragments of human growth hormone produced by digestion with plasmin. Endocrinology 96, 625-636. Sakiz, E. Statistical analysis in experimental biology with EXBIOL, a single multiphase program. Common Research Computer Facility, Texas Medical Center, Houston, not dated. Sherwood, L. M., Handwerger, S. & McLaurin, W. D. (1972). The structure and function of human placental lactogen. In Lactogenic hormones, pp. 27-44. Eds G. E. W. Wolstenholme & J. Knight. London : Churchill

Livingstone. Singh, R. N. P., Seavey, B. K., Rice, V. P., Lindsey, T. T. & Lewis, U. J. (1974). Modified forms of human growth hormone with increased biological activities. Endocrinology 94, 883-891. Trenkle, ., Li, C. H., Sadri, S. K. & Robertson, H. (1962). Effects of chemical treatment on immunochemical properties of human growth hormone and sheep interstitial cell-stimulating hormone. Archives of Biochemistry and Biophysics 99, 288-293. Yadley, R. ., Rodbard, D. & Chramback, A. (1973). Isohormones of human growth hormone. III. Isolation by preparative polyacrylamide gel electrophoresis and characterization. Endocrinology 93, 866-873.

Biological actions of human somatotrophin and its derivatives on mouse mammary gland and teleost urinary bladder.

BIOLOGICAL ACTIONS OF HUMAN SOMATOTROPHIN AND ITS DERIVATIVES ON MOUSE MAMMARY GLAND AND TELEOST URINARY BLADDER B. A. DONEEN, H. A. BERN AND CHOH H...
NAN Sizes 0 Downloads 0 Views