Conformation of biologically active derivatives of human placental lactogen.

Cleavage of human placental lactogen (hPL) by plasmin, which removes the hexapeptide containing residues 135 to 140, has allowed us to prepare biologically active derivatives. In order to study the relationship between chemical structure and biologic activity, we have compared the conformation of two active derivatives, plasmin-cleaved hPL and (1-134)2 (a disulfidelinked dimer of Fragment 1-134), with native hPL and reduced and alkylated 1-134, using circular dichroism, intrinsic  fluorescence, and immunologic measurements. Circular dichroism in the region of peptide bond activity showed that hPL, plasmin-cleaved hF’L, and (1-134)2 had comparable degrees of (Y helix, while reduced and alkylated 1-134 had much less. In the nearultraviolet region of the spectrum circular dichroism measurements of hPL and plasmin-cleaved hPL were nearly identical. The spectra of (1-134)2 and reduced and alkylated 1-134 were considerably different from hPL and from each other. A red shift to tryptophanyl fluorescence was produced by titration with increasing alkali or urea. The amount of urea needed for this transition, in decreased order, was: W L == plasmincleaved hPL > (1-134)2 > reduced and alkylated 1-134. Two radioimmunoassays, one using antibody raised against  native  hPL  and  the  other  using  antibody against reduced and alkylated 1-134, were used. The specificity of these assays for the conformation of hPL was indicated by the virtual lack of cross-reactivity of Fragment 1-134 in the native hPL radioimmunoassay and the marked decrease in activity of native hPL in the assay for reduced and alkylated 1-134. Comparison of displacement curves in these assays showed that plasmin-treated hPL was different from, but closely related to, native hPL while (1-134)2 had immunologic features of both native and unfolded hPL. We conclude that the biologically active derivatives of hPL contain important elements of the native configuration as well as varying amounts of features characteristic of unfolded hPL.

Cleavage of human placental lactogen (hPL) by plasmin, which removes the hexapeptide containing residues 135 to 140, has allowed us to prepare biologically active derivatives. In order to study the relationship between chemical structure and biologic activity, we have compared the conformation of two active derivatives, plasmin-cleaved hPL and (1-134)2 (a disulfidelinked dimer of Fragment 1-134), with native hPL and reduced and alkylated 1-134, using circular dichroism, intrinsic fluorescence, and immunologic measurements. Circular dichroism in the region of peptide bond activity showed that hPL, plasmin-cleaved hF'L, and (1-134)2 had comparable degrees of (Y helix, while reduced and alkylated 1-134 had much less. In the nearultraviolet region of the spectrum circular dichroism measurements of hPL and plasmin-cleaved hPL were nearly identical. The spectra of (1-134)2 and reduced and alkylated 1-134 were considerably different from hPL and from each other. A red shift to tryptophanyl fluorescence was produced by titration with increasing alkali or urea. The amount of urea needed for this transition, in decreased order, was: W L == plasmincleaved hPL > (1-134)2 > reduced and alkylated 1-134.
Two radioimmunoassays, one using antibody raised against native hPL and the other using antibody against reduced and alkylated 1-134, were used. The specificity of these assays for the conformation of hPL was indicated by the virtual lack of cross-reactivity of Fragment 1-134 in the native hPL radioimmunoassay and the marked decrease in activity of native hPL in the assay for reduced and alkylated 1-134. Comparison of displacement curves in these assays showed that plasmin-treated hPL was different from, but closely related to, native hPL while (1-134)2 had immunologic features of both native and unfolded hPL. We conclude that the biologically active derivatives of hPL contain important elements of the native configuration as well as varying amounts of features characteristic of unfolded hPL.
A series of biologically active derivatives of hPL' has been prepared by limited plasmin digestion. These have been useful publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 The abbreviations used are: hPL, human placental lactogen; PL-. hPL, plasmin-treated hPL; 1-134 and 141-191, hPL fragments containing the indicated residues; RCAM, reduced and S-carbamidomethylated; SDS, sodium dodecyl sulfate. in studying the relationship between the primary structure of the hormone and its biologic activity. Plasmin digestion of hPL removes a single hexapeptide (residues 135 to 140) and leaves two fragments (residues 1 to 134 and 141 to 191) joined by a disulfide bond between Cys5:3 and Cys~a (1). This derivative (PL-hPL) has increased potency in both receptor binding and in vitro biologic assays. Two 1-134 fragments can be joined by a disulfide bond between Cysn;~ residues to form (1-134)*. This derivative retains the ability to bind to lactogenic receptors and is fully active in vitro ( 2 ) . In the accompanying paper we describe the production and characterization of active derivatives produced by the recombination of plasmingenerated fragments of hPL with fragments of hGH (3).
Our earlier reports indicated that the conformation of the hPL derivatives was important in determining their biologic activity (1-3). For example, RCAM-1-134 was shown to be biologically inactive while the dimer (1-134)g was active. This difference could only be accounted for by the refolding of the dimer to an active conformation. In this manuscript we describe studies of the conformation of hPL and its derivatives using circular dichroism, intrinsic fluorescence, and immunologic measurements. The results indicate that the structure of PL-hPL is similar to hPL, while (1-134)2 contains structural features of both native and unfolded hPL.2

DISCUSSION
Three methods were employed to give complementary information about the structure of hPL and its derivatives. The far-UV circular dichroism spectrum was used to measure secondary structure, two radioimmunoassays were used to distinguish native and unfolded conformational states, and intrinsic fluorescence and near-UV circular dichroism were used to monitor the environments of the aromatic chromophores (9-17).
RCAM-hPL and RCAM-I-134-All of the analytical methods indicate that the structures of RCAM-hPL and RCAM-1-134 were very different than that of the native hPL. Aloj et al. (9) showed that the structure of RCAM-hPL is not random but is different than that of native hPL. Therefore, the structure of the RCAM-hPL derivatives will be referred to as "unfolded" forms, recognizing that they are not entirely random.
PL-hPL-The secondary structure and environment of the aromatic groups of hPL and PL-hPL were very similar since their far-UV and near-UV circular dichroic spectra were nearly identical. Their immunologic reactivities were also very similar since PL-hPL reacted much better in the assay directed against native hPL than in the assay directed against RCAM-1-134. PL-hPL was more potent than hPL in the RCAM-1-134 assay, indicating some degree of unfolding. The fluorescence titrations with urea and alkali indicated that PL-hPL and hPL were similar, but that the transition of the tryptophanyl residue of PL-hPL to longer wavelength, and therefore a more polar environment occurred more easily. The observations emphasize the similarities between the conformation of hPL and PL-hPL while the differences indicate that PL-hPL has a small amount of structure characteristic of the unfolded form.
(1-134)z-The structure of (1-134)2 was of great interest since one-third of the native molecule is missing, yet it retains the structure needed for biologic activity. The far-UV circular dichroism spectrum indicated that the dimer contained approximately the same per cent a helix as the native molecule. However, it cannot be concluded that the same a helices are present since the ellipticities are calculated for mean residue weights and the molecular weights of hPL and (1-134)? are not the same. Since it is not known how much of the a helix in hPL involves the first 134 residues and how much is in the remaining residues, it is not possible to compare quantitatively the per cent of a helix in (1-134h with the per cent of a helix in the same residues contained in native hPL. The immunologic studies showed that (1-134)~ combines features of the native and unfolded forms. In the radioimmunoassay for native hPL, (1-134)2 regained substantial activity. This indicated that the structure induced by dimerization through the disulfide bond was similar and possibly identical with structural elements in native hPL. In the radioimmunoassay directed against RCAM-1-134, (1-134)2 was more potent than the reduced and alkylated fragment itself. Thus, in addition to assuming a native structure, the dimer also retains elements characteristic of the unfolded form. Why (1-134):! was more potent than RCAM-1-134 is not clear.
Holladay and Puett (12) have shown that the negative CD peak of hPL at 283 nm is the most intense (resolved intensity of -134 deg cm'/dmol) and is made up of contributions from both tyrosine and tryptophan transitions (13). The bands at 293 nm and 276 nm, which are not fully resolved in the hPL spectrum, have been assigned to tryptophan and tyrosine, respectively, and are considerably weaker (-17 and -40 deg cm'/dmol). The two disulfide bands do not have resolved bands of C D activity, but rather contribute a broad background of activity from below 250 nm to 290 nm (18). With respect to hPL, the magnitude of the (1-134h bands at 282 and 277 nm were reduced slightly more than 50%. Since each of the aromatic residues potentially can contribute either positive or negative ellipticity it is not possible to assign changes to the individual residues. The spectra are consistent with a model in which some of the aromatic residues are in enviroments similar to those in native hPL while the other residues are devoid of activity in the near-UV region. The intrinsic fluorescence titration curves of (1-134)2 with alkali and urea both indicate that the tryptophan transition to longer wavelengths occurs more easily than for native hPL and PL-hPL.
Conclusion-Each of the biologically active derivatives of hPL contains elements of native conformation as well as features characteristic of unfolded hPL. There are two possibilities that could account for the simultaneous presence of elements of native and unfolded structure in the same molecule which cannot be distinguished by the methods employed. These are as follows: 1) each molecule has a stable conformation which retains both native and unfolded elements and 2) there is rapid equilibrium between conformations, with one form closely related to the native molecule and the other related to the unfolded species. Theoretically, these possibilities could be distinguished by kinetic measurements using a structural parameter which could be obtained more rapidly than the potential rates of interconversion of two species. It will probably be necessary for x-ray crystallographic data to define more precisely the three-dimensional structure necessary for lactogenic activity in hPL and, in the homologous proteins growth hormone and prolactin, the structure necessary for both growth-promoting and lactogenic effects.