Developments in Bioenergetics and Biomembranes Volume 6 Other volumes in this series: Volume 1 Bioenergetics of Membranes Lester Packer, George C. Papageorgiou and Achim Trebst editors, 1977 Volume 2 The Proton and Calcium Pumps G. F. Azzone, M. Avron, J.C. Metcalfe, E. Quagliariello and N. Siliprandi editors, 1978 Volume 3 Function and Molecular Aspects of Biomembrane Transport E. Quagliariello, F. Palmieri, S. Papa and M. Klingenberg editors, 1979 Volume 4 Hydrogen Ion Transport in Epithelia I. Schulz, G. Sachs, J.G. Forte and K.J. Ullrich editors, 1980 Volume 5 Vectorial Reactions in Electron and Ion Transport in Mitochondria and Bacteria F. Palmieri, E. Quagliariello, N. Siliprandi and E.C. Slater editors, 1981 STRUCTURE AND FUNCTION OF MEMBRANE PROTEINS Proceedings of the International Symposium on Structure and Function of Membrane Proteins held in Selva di Fasano (Italy), May 23-26, 1983. Editors E. Quagliartene P. Palmieri 1983 ELSEVIER SCIENCE PUBLISHERS AMSTERDAM · NEW YORK · OXFORD © 1983 Elsevier Science Publishers B.V. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. Published by: Elsevier Science Publishers B.V. P.O. Box 211 1000 AE Amsterdam, The Netherlands Sole distributors for the USA and Canada: Elsevier Science Publishing Company Inc. 52 Vanderbilt Avenue New York, N.Y. 10017 ISBN: for this volume: 0-444-80540-0 ISBN: for the series: 0-444-80015-8 Library of Congress Cataloging in Publication Data International Symposium on Structure and Function of Membrane Proteins (1983, : Selva di Fasano, Italy) Structure and function of membrane proteins. (Developments in bioenergetics and biomembranes ; v. 6) Bibliography: p. Includes index. 1. Membrane proteins—Congresses. I. Quagliariello, Ernesto. II. Palmieri, F. (Ferdinando) III. Title. IV. Series. CDNLM: 1. Membrane proteins—Congresses. 2. Structure-activity relationship—Congresses. Wl DE99TVK v.6 / QU 55 l6T^s 19833 QP552.MMtI57 1983 57^.87!5 83-16565 ISBN 0-UMi-805UO-0 (U.S.) Printed in The Netherlands V PREPACE The purpose of this International Sympsoium on "Structure and Function of Membrane Proteins" at Selva di Fasano was to review the present status of research into the structure-function relatioship of membrane proteins. Structure was what the organizers had primarily in mind, although they were aware that knowledge in this field is still embryonic. However, membrane proteins are no longer those oily creatures floating around in the lipid bilayer or in ill-defined detergent micelles. Mem­ brane proteins can be put into well-ordered arrays of either two-or three-dimensional order, reflecting the well-defined monodisperse structure of these entities. Wereas, prokaryotic membrane proteins, particularly of the exotic varieties, are often highly expressed, making solubilization and purification less important, more difficulties have been encountered in the isolation of intact membrane proteins from eukaryotic membranes. Only in the last ten years have these problems been solved, in principle, through our understanding of the appropriate handling of detergents, in particular of the non-ionic variety of these. As a result, numerous sol ubi li zed membrane proteins are available in both monodisperse and native forms. It should now be possible to obtain more information on the structure of these membrane proteins. Structural research comprises, first of all, the elucidation of the amino acid sequence, the folding and the overall shape, leading up to ultimate, atomic, resolution which requires crystals suitable for X-ray crystallography. Progress in elucidating the primary structure of membrane proteins has come from improved handling of the hydrophobic peptides, and also from the use of cloned c-DNA sequences. Two-dimensional arrays constituted the first step in ordering membrane proteins and their successful application is now relatively widespread. Three-dimensional crystals for X-ray crystallography are available in only two instances, both of which have been discussed at this meeting. From primary sequences, hypothetical folding predictions are made, with the focus on the ubiquitous transmembrane α-helical segment. These predominantly theoretical models can be useful for explaining sidedness and assigning active sites, but will ultimately be replace by physical data. The other approach to structure determination is based on VI monodisperse, soluble proteins or protein micelles using scattering methods for X-ray, light and neutron beams. Structural information can also be obtained by using fluorescence probes and spin labels as long as specific localization and assignments at the protein can be made. Structure and function are reciprocally linked. We wish to determine the structure in order to understand the function, and the mechanism of action will be understood only by our knowledge of the atomic structure. The gathering of data on the function of membrane proteins prior to knowledge of their structure is valuable for characterizing and defining the proteins. Once the structure is known, another stage of research will penetrate to the functional assignments of the structure. We have witnessed in this symposium the various stages of this very lively and effective research on biomembrane proteins. New enthusiasm and scope for research has been created through obtaining the first crystals of membrane proteins and it is to be hoped that some years from now another meeting, possibly in the same room, might witness again the great progress which we expect in this exciting field. 0 1983 Elsevier Science Publishers B.V. Structure and Function of Membrane Proteins, E. (luagliariello and F. Palmieri editors. 3 GRAMICIDIN A TRANSMEMBRANE CHANNEL: KINETICS OF PACKAGING IN LIPID MEMBRANES. MA SOT TI^, 2 , LANFFWNCO PAOLO CAVATORTA ALBERTO SPISNI~,E MANUELA , 4 CASALIl, GIORGIO SARTORl, IVONNE PASQUALI-RONCHETT13 ARTHUR SZABO. 'Institute of Biological Chemistry, University of Parma, Viale Gramsci, 14, 43100 Parma, (Italy); 21nstitute of Physics-GNCB, University of Parma, Via D'Azeglio, 85, 43100 Parma, (Italy); 31nstitute of General Pathology, University of Modena, Via Campi, 165, 41100 Modena, (Italy); 'Division of Biological Sciences, National Research Council, 100 Sussex Dr., Ottawa, Ontario K1A OR6, (Canada). INTRODUCTION Biological membranes are recognized to act as anchoring point for bound enzymes, to provide a medium where proteins, substrates and products of enzyme reactions of limited solubility in water become soluble, and where multienzyme complexes are organized. Fur- thermore they define cellular compartments where substrates, prod- ucts and effectors of metabolic reactions are separated and control their selective transport between compartments. In particular, se- lective ion movement across membraneis essential for the processes of energy transduction and cell excitability. Several attempts have been made to characterize native channels in membranes in terms of structure and mechanism (1,2), but because of their complexity, attention has also been focused on model sys- tems capable to function as selective channels for ions. Gramicidh A'(GA'), a hydrophobic polypentadecapeptide, has been shown to be 0 able to form helical structures with an internal pore of about 4 A + in diameter, that can accomodate monovalent cations such as Na or . K+ (3-10) It has been demonstrated (11) that GA' can be incorporated into lysolecithin micelles and that it assumes a left-handed helical configuration able to transport monovalent cations selectively (12). Moreover we have shown that the formation of the channel is associ- ated with a reorganization of the lipid phase in a bilayer struc- ture, wherein channel aggregates are embedded (13,14). Having recognized the importance of a supramolecular organization for the channel activity (15) our attention has been focused on the 4 studies concerning the kinetics of the incorporation and organiz­ ation of GA1 in the lipid phase. MATERIALS AND METHODS Lysolecithin was obtained from Sigma Chemical Company, St. Louis, Mo., and checked for purity by nuclear resonance and thin layer chromatography, Gramicidin was purchased from ICN Pharmaceuticals, Cleveland, Ohio, as a mixture of 80% Gramicidin A, 6% Gramicidin B, and 14% Gramicidin C, and was used without further purification. The mixture is referred to as Gramicidin A1 (GA1). Phospholipid was suspended in water and heated at 70°C up to 22 hours, as previously reported (15), either in the presence or in the absence of GA1 . Two samples with different GA'/üpid ratios were prepared: the first with 6 mg GA', the second with 3 mg GA' per 25 mg of lysolecithin. Aliquots were withdrawn at times : 0, 10, 30, 60, 120, 180, 270 minutes and 21 hours. Static fluorescence measurements were carried out as previously described (15) . Lifetime measurements were performed using a Spectra Physics apparatus with a pulsed laser source (1 psec). Absorption measurements were performed using a Perkin Elmer 576 Spectrophotometer, equipped with a termostatically controlled cuvette holder at temperature of 30±0.5°C. A molar extinction coefficient of 22500 mol cm at 281 nm in CH^OH was used to calculate the GA" concentration. Circular Dichroism measurements were performed on a JASCO J-500 Spectropolarimeter equipped with a microprocessor unit for spectra accumulation. The samples were diluted with water and run using cuvettes with 0.2 mm pathlength. The ellipticities were calculated using a mean molecular weight per residue of 124.5. For electron-microscopic studies, the sample were left to equil­ ibrate at room temperature before being processed. The specimens were examined in a Siemens Elmiskop 1A and Philips 410 electron microscopes. The magnifications were calibrated by optical dif­ fraction of catalase crystals. For negative staining see caption to Fig.4. RESULTS AND DISCUSSION In order to verify if the initial amount of GA1 used in the 5 experiments can influence the velocity of incorporation of the polypeptide into the lipid system, the two concentrations re­ ported in the Materials and Methods section were chosen. The percent of GA' incorporated, taking as 100% the actual initial amount of the polypeptide, shows that the rate of incorporation is somewhat proportional to the initial amount of GA1. The fluorescence emission maximum, as shown in Fiq. 1, is shifted from 342±1 nm at the time t=0 to 328+1 nm at the time t=22 hours of incubation. Such a blue shift tipically indicates a change in the polarity of the environment of tryOtophane residues, from a polar to a non polar one. — — — time zero 1 y^S. -'*-'- \\ after 22 h ^ / ·;/ ■\\ / ·' ' b / ; ' v \\ < / ; ' \ *N V-* / ·' ' \ "Λ LU / ·* ; \ *A Ü / ; ' zLU 1 : ; 1' \ \ *-*x·χ ωÜ 0,5 / : / \ Λ / : ; \ #,·Ν ce / .* ' \ *A o / ; ; \ *"Λ D / ; ' \ * -J / : ' \ Λ u. / ; f \ *Λ ; 1 · ' / ·: ; X *·.\ /·"" ' // / ^ "***. £·> _j i ■*·■■ i 1 300 350 400 450 X(nm) Fig. 1. Fluorescence spectra of GA1 incorporated into the lipid system at different times of incubation. The time course of absorbance, corrected for light scattering, 6 during incorporation, Fig. 2, clearly shows that the GA' incor­ porates into the lipid system. In fact at time t=0', the amount of GA1 incorporated is just 6% of the total amount added and reaches the 80% at t=22 hours. At the same time the quantum yield decreases from a value of 0.7 for both the samples at time t=0', to 0.06 at t=22 hours. However, the high values of the quantum yield at short times might be overestimated, being the absorbance not entirely cor­ rected for light scattering and "obscured" absorbance (16). O· 6 mg/ml DB 3 mg/ml 60 -j 50 1.0 J 40 0.8 ■ / < Q.Y. T / oo cc 30 0.6 O —a </> m 1 ö ^ ^~ J 20 -\ 0.4 J 10 0.2 σ · 1 n 200 400 600 800 1000 1200 TIME (min) Fig. 2. Absorbance (open symbols)and quantum yield (Q.Y.)(full symbol) dependence on time of incubation. The CD spectra, shown in Fig. 3, demonstrate that also at the time t=0', some of the polypeptide has already adopted an helical structure. At subsequent times the CD pattern is similar to that D 1.0 0.5 t- 0c la a 4 9 -H 0.5 I B 15 1.0 C 30 D 60 , , , I I ? , , 190 210 230 2 50 2 70 290 h tnm~ Fig. 3. Circular Dichroism spectra of G.A' incorporated into the lipid systems at different times of incubation. reported previously (111, until after 2 hours is consistent with that of a left-handed helix, and it is stable with time. The lifetime measurements (see Table 1) seem to indicate that at the time t=O' two tvpes of tryptophane are detected, of which one predominates. The lifetime does not change with wavelength indicating that both types of tryptophane experience the same environment. Up to 90' the lifetimes change very little. At longer incubation times, shorter lifetimes are measured at 335 nm, while at longer wavelenqth again no change is observed. The ultrastructural studies of the samnles durina incorworation