The melting behavior of the homopolymers poly[A5-(3-hydroxypropyl)-L-glutamine] and poly[(V6-(4-hydroxybutyl)-L-glutamine], and copolymers of these two amino acids, was determined in water. The homopolymer data were treated by the Zimm-Bragg theory and the copolymer data were analyzed with the lowest order approximation of the theory discussed in the previous paper. Within the experimental error, it was not possible to detect any temperature dependence of the parameter or of the parameters AH and AS for the homopolymers. Using these temperature-independent parameters for one of the homopolymers, it was possible to compute those for the other homopolymer by applying the host-guest technique (and associated theory) to the copolymer data. Good agreement was obtained between the parameters computed directly from homopolymer data and those obtained by the host-guest technique, thus establishing the validity of the latter method for future use in obtaining the helix-coil stability constants for other amino acids.

The two-dimensional statistical mechanical equations necessary to analyze the phase diagram of a nonideal lipid mixture are derived. The Prigogine approximation of the combinatorial term is used in the maximum term of the partition function which contains two parameters that describe the degree of nonideality in the liquid crystal and gel phases. These parameters are used to calculate the mole fraction of like-like and like-unlike lipid interfaces which describe the lateral species separation of like lipids in each phase. Data for several temperatures and compositions of the DMPC/DPPC, DPPC/DSPC, and DMPC/DSPC mixtures are analyzed.

Since it is necessary to have the Zimm-Bragg parameters and s of all naturally occurring amino acids (in water) for studies of protein conformation, and since these cannot be obtained from studies of the helix-coil transition in homopolymers, because of experimental difficulties, a technique has been developed to circumvent these problems. It involves the study of the thermally induced transition curves for random copolymers of "guest" amino acid residues in a water-soluble "host" poly(amino acid). The data may be interpreted with the aid of suitable theories for the helix-coil transition in random copolymers to obtain and s for the "guest" residues. While exact theories are available (for the one-dimensional nearestneighbor Ising model), the computer costs involved in the computations are prohibitively high. Therefore, resort is had to approximate theories. It is shown here that, for the usual ranges of parameters found for poly(amino acids), one of the two lowest order approximations (corresponding to earlier treatments by Lifson and Allegra) is completely adequate, i.e., gives results essentially identical with the exact result. In essence, the low-order approximations hold if and s for the two constituents of the copolymer do not differ appreciably from each other. If they do, then higher order approximations (which become exact in highest order) are required. Calculations are also reported for both regular-sequence and random copolymers in order to demonstrate how the amino acid sequence and composition of the copolymer, and also the values of and s of the constituents, affect the transition temperature, the breadth of the transition, and the most probable length of a helical sequence in the copolymers. It is shown that small departures from a random sequence do not affect the melting curve significantly.

The 1H nuclear magnetic resonance (NMR) spectra of four fibrinogen-like oligopeptides (H-Gly-Pro-Ala-NH2, H-Arg-Gly-Pro-Ala-NH2, H-Val-Arg-Gly-Pro-Ala-NH2, and H-Gly-Val-Arg-Gly-Pro-Ala-NH2) in dimethyl-d6 sulfoxide (Me2SO-d6), and of the hexapeptide in water, and the 13C NMR spectrum of H-Gly-Pro-Ala-NH2 in Me2SO-d6, were recorded and interpreted in terms of preferred conformations in solution. Each peptide exists in Me2SO-d6 as in a 30:70 mixture of cis and trans isomers about the Gly-Pro bond, and the hexapeptide in water is solely the trans isomer. For the trans isomers in Me2SO-d6, there is a hydrogen bond between the Gly CO group and one of the C-terminal primary amide hydrogens, and a beta turn involving the Gly-Pro-Ala-NH, section of the molecules. A strong NOE between Pro CalphaH and Ala NH for the trans isomer of the tripeptide in Me2SO-d6 completes the identification of this structural feature as a type II beta turn.

The NMR spectra at 220 MHz of iysinevasopressin and its precursors were measured in dimethyl sulfoxide, and the peaks were assigned to specific protons. Information about hydrogen bonding was obtained from the temperature coefficients of the chemical shifts. With these data, and with several chemical-shift positions and coupling constants, structural information was derived for this polypeptide hormone.High-resolution nuclear magnetic resonance (NMR) studies offer the possibility of determining the conformations of macromolecules in solution (1). This method is applied here to the well-known octapeptide hormone lysine-vasopressin (LysVP) in deuterated dimethyl sulfoxide [2H]6Me2SO solution. The interpretation of the NMR spectrum of this polypeptide was aided materially by the availability of the blocked precursors (dipeptide to nonapeptide) used in the synthesis of LysVP, i.e., the changes in the spectrum in the progression from dipeptide to nonapeptide to LysVP provided information on the chemical shift of specific protons, as well as on changes in conformation as each amino acid residue is added.LysVP (I) has the following structure, in which the numbers indicate the positions of the individual amino acid residues:Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Lys-Gly-NH2. The half-cys-1 2 3 4 5 6 7 8 9 tine residues in positions 1 and 6 are referred to as Cys-1 and Cys-6, respectively. MATERIALS AND METHODSLysVP was a purified synthetic preparation (2, 3) that possessed about 250 U/mg of rat pressor activity.t The following protected peptide intermediates were also used in this study:Abbreviations for amino acid residues and protecting groups are in accordance with the IUPAC-IUB Tentative Rules [J. Biol. Chem., 24, 2491Chem., 24, (1966 The proton NMR spectra were recorded on a Varian HR-220 spectrometer, and on a Varian HA-100 spectrometer using an internal lock and frequency sweep mode. The sample concentrations were 4-10 g/100 ml in (100 mol %) [2H]6Me2SO, that had been stored over molecular sieves. When the polypeptide concentration was varied in this range, no change in the spectra were detected. All chemical shifts are downfield from the internal standard tetramethylsilane. The temperatuie of the sample was controlled to ±2°C. Homonuclear spin-decoupling on the HR-220 spectrometer was done by the field sweep method, using a General Radio Co. 1107-A interpolation oscillator, and on the HA-100 by the frequency sweep method. The spectra were calibrated by the side-band technique.After the NMR measurements were completed, the LysVP solutions were bioassayed. No appreciable inactivation of the LysVP was detected. RESULTS AND DISCUSSION Assignment of peaksThe peaks of the NMR spectrum of LysVP were assigned by making use of the spectra of the blocked precursors, from the di-to the nonapeptide, together with information on the spectra of other oligopeptides and of the isolated amino acids. The latter could not be used by themselves because of the influence of charges on the a-NH3+ and a-COO-groups. However, an amino aci...