The octapeptide, Z-Gln-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-NHEt, is an effective inhibitor of LHRH both in vitro and in vivo5. We prepared the 6-D-Lys analogue of this peptide to allow easy substitution at a central position in the molecule; this in turn would allow the formation of active macromolecular conjugates with the agonist2. The peptide was prepared from the protected heptapeptide described previously2 (H-Trp-Ser-TyKBu^-D-LystBOQ-Leu-Arg-Pro-NHEt). This was coupled, in dimethylformamide, with the activated derivative Z-Gln-OCP, and the product purified by chromatography on Sephadex LH-20 in methanol. The side chain protecting groups were removed with trifluoroacetic acid/water (9:1) and the peptide was isolated by chromatography on Sephadex G-50 in acetic acid/water (1:4). Analysis of an acid hydrolysate showed good amino acid ratios and the peptide was revealed as a single spot by high voltage electrophoresis at pH2.1, 6.5 and 8.9 using ninhydrin, Ehrlich's and Sakaguchi visualization reagents and by TLC on silica gel plates in n-butanol/acetic acid/water (4:1:5). Table 1 Secretion of luteinizing hormone from porcine pituitary cells in culturePeptideLHRH (10~9 M) LHRH (10~8 M) Inhibitor (10~9M) Inhibitor (1(T6M) LHRH (1(T9 M) 4- inhibitor (1(T6 M) Inhibitor (10~8 M) Inhibitor (1(T6M) Inhibitor (10~8 M) LHRH (10~8 M) AntiserumRabbit anti-inhibitor (1 : 1,000 dilution) Rabbit anti-inhibitor (1 : 1,000 dilution) Rabbit anti-D-Lys agonist (1 : 1,000 dilution) Rabbit anti-inhibitor (1 : 1,000 dilution) LH released18524 30034 295 495 3512 21536 11216 298 36151 Cells were incubated at 37 °C for 24 h after dissociation at 4x 106 cells per 2.5-cm diameter Petri dish; before each experimental incubation they were washed three times in serum-free medium. The cells were then incubated in 1 ml medium for 30 min with additions as shown. Values are ng ml"1s.e.m. luteinizing hormone. The pure peptide was then coupled through the side chain of the central Lys residue to bovine serum albumin (BSA) using l-ethyl-3(3-dimethylaminopropyl)carbodiimide. Antisera to the peptide were raised in rabbits by injection of conjugate (1.0 mg) in Freund's complete adjuvant subcutaneously (s.c.), followed by two booster injections of conjugate in saline at intervals of 4 weeks. The antisera were able to bind iodine-labelled LHRH, indicating that there were features common to the hormone and antagonist.Dissociated porcine pituitary cells were prepared as described previously6 and luteinizing hormone (LH) release measured by radioimmunoassay for the various treatments (Table 1). Clearly, the 6-D-Lys analogue of the antagonist5 retained similar activity in vitro as gonadotropin release by the native hormone was suppressed completely at concentrations 100 times that of the agonist. The analogue alone gave no gonadotropin release at concentrations up to 10"6 M, indicating that the occupied receptor complex lacked the structure required for a biological response. However, the presence of antibodies converted this antagonist complex into a functional agonist by aggregating occupied receptors as a first stage. Lower levels of gonadotropin were secreted at higher concentrations of peptide, presumably because of extensive binding of antibody by free peptide in solution. The specificity of the effect was demonstrated bythe fact that the antiserum raised against the peptide having ^Glu-His instead of Z-Gln was unable to evoke a response. Although this indicated that antibody binding occurred through the N-terminal region, the anti-inhibitor antiserum may have multiple specificity and could also contain different immunoglobulin isotypes. Thus, although aggregation is involved, its scale and specificity is unclear. From the rabbit antiserum to the inhibitory peptide, we prepared pure IgG having specificity for the N-terminus of the peptide (for experimental details see Table 2). The purified specific IgG and Fab proteins were then used to study LH release from pituitary cells (Table 2).For this particular cell preparation, LHRH at 10~7 M stimulated LH release 3-4-fold. The inhibitor alone was ineffective at 10~9M but in conjunction with purified specific IgG at ~KrnM, LH release occurred. That this required divalency was apparent from the inability of similar amounts of Fab fragment to evoke similar effects and by the ability of goat anti-rabbit anti-serum to induce the Fab-antagonist complex to stimulate LH release. The effect of all agents together (Table 2) indicated that LH release was only mediated via the LHRH receptor. Table 2 Sequential experiments with porcine pituitary cells in culture showing LH release for each treatment period.Control period Period 1 Period 2 25.02.4 LHRH(10"7M) 94.44.7 30.34.2 Inhibitor (10~9M) 28.21.8 30.51.8 Inhibitor (10~9M) + 1.3 ng ml'1 IgG 78.44.3 20.04.7 Inhibitor (10~9M) + 0.13 ngmF1 IgG 26.42.6 30.53.73 Inhibitor (10-9M) + 0.8 ngmF1 Fab 37.68.3 Inhibitor (10~9M) + 0.8 ngmF1 Fab + GaRS (1:20) 15020.3 23.62.8 Inhibitor (10~9M) + 0.8 ngmF1 Fab Inhibitor (10~9 M) + 0.8 ng mF1 Fab + LHRH(10~7M) 120.56.2 + LHRH (10~7M) + GaRS (1:20) 126.416.2 Pituitary cells were prepared and plated out as described in Table 1. Between each 30-min incubation period, the cells were rinsed with serum-free medium for three 1-min periods. Values are ng mF1s.e.m. luteinizing hormone. The goat anti-rabbit serum (GaRS) had no significant effect on LH release when used alone at 1:20 dilution. The purified specific IgG and Fab proteins used in the study were prepared as follows. Rabbit antiserum (1 ml) to the inhibitory peptide, diluted with phosphate-buffered saline (PBS), was recycled through an affinity column comprising 2 ml Sepharose coupled to 20 mg of the C-terminal peptide-sequence (Leu-Arg-Pro-NHEt) at 4 °C for 48 h. The non-retarded serum solution was recycled through a 5 ml column of Sepharose-Protein A (Pharmacia) for 24 h. The column was washed with 5 vol PBS, eluted with 5 ml 0.1 M citrate buffer, pH 3.5 and the eluate neutralized immediately with 0.5 ml M Tris-HCl buffer, pH 8.0. The lyophilized eluate was taken up into water and applied to a column of Ultrogel ACA 34 (60 x 0.9 cm) in 0.05 M ammonium bicarbonate. The fractions from the single peak in the precalibrated IgG position (2.6 mg) were combined and lyophilized. Half the sample was taken up into 200 jxl buffer (0.1 M phosphate pH 7.0, 0.01 M cysteine, 0.002 M EDTA) and treated with 15 jxg papain at 37 °C for 16 h. In the subsequent Ultrogel chromatography designed to obtain Fab fragments, we observed no intact IgG. Previously, we have observed aggregation and internalization of bound LHRH2 and it has been shown that receptor aggregation is essential for the action of some polypeptide hormones. Thus, antibodies to the insulin receptor produced insulin-like responses directly7'8 and a degraded EGF analogue became mitogenic when aggregated by antibody9. However, for these larger molecules, it is possible that the antisera have multiple specificity; the simpler octapeptide used here allows clearer definition of the effects.Since our initial report of the conversion of a pure antagonist to agonist10, we have shown that a different antagonist when converted to a dimer, could be activated using anti-serum11, thus confirming the phenomenon. Here, use of purified immunoglobulin and peptide monomer allows firm conclusions to be made concerning the nature of the hormone-receptor interactions. The high specificity of the hormone for the receptor presumably involves several amino acids, and the antigenic determinants of proteins and polypeptides have also been shown to encompass several amino acids12. Thus, it is unlikely that residues contained in the antigen-binding site are available for receptor binding at the same time and vice versa; the logical conclusion is that the octapeptide cannot bind more than one antibody molecule concurrent with the receptor site being occupied by this inhibitor. The IgG preparation, which is adivalent protein specific to the peptide N-terminus, can produce a biological response from an inactive hormone complex. This indicates that formation of an occupied receptor dr (and no greater aggregation is required) by lateral movement through the membrane is a necessary step in the response pathway after binding of hormone to the receptor. Accepting that the octapeptide is only large enough to bind to the receptor and to one antibody molecule, it follows that the C-terminal region must have a dominant role in binding to the receptor, and this is common to both agonist and antagonist. That the anti-antagonist antibody recognizes the N-terminal region is supported by the fact that the anti-agonist antibody (Table 1) was unable to induce LH release, thus it is proposed that the ^Glu-His-Trp-Ser sequence is important in the dimerization of occupied receptors. We conclude, therefore, that this small peptide hormone is able to exert an effect on a normal cell population by having a sequence allowing specific binding, followed by a second sequence that causes dimerisation of the receptors, thus leading to biological expression by an as yet undefined series of stages.