The amino acid sequence of human insulin-like growth factor I and its structural homology with proinsulin

The complete amino acid sequence of human insulin-like growth factor I (IGF-I), a polypeptide isolated from serum, has been determined. IGF-I is a single chain polypeptide of 70 amino acid residues cross-linked by three disulfide bridges. The calculated molecular weight is 7649. IGF-I displays obvious homology to proinsulin: positions 1 to 29 are homologous to insulin B chain and positions 42 to 62 to insulin A chain. A shortened "connecting" peptide with 12 residues (positions 30 to 41) compared to 30 to 35 in proinsulins shows no homology to proinsulin C peptide. An octapeptide sequence at the COOH-terminal end is also a feature not found in proinsulins. The number of differences in amino acid positions between IGF-I and insulins suggests that duplication of the gene of the common ancestor of proinsulin and IGF occurred before the time of appearance of the vertebrates. Of the 19 residues known to be invariant in all insulins so far sequenced, only glutamine A5 and asparagine A21 are replaced in IGF-I by glutamic acid and alanine, respectively. The fact that all half-cystine and glycine residues and most nonpolar core residues of the insulin monomer are conserved is compatible with a three-dimensional structure of IGF-I similar to that of insulin.     

Sequence-specific 1H-NMR assignments for the aromatic region of several biologically active, monomeric insulins including native human insulin

The aromatic region of the 1H-FT-NMR spectrum of the biologically fully-potent, monomeric human insulin mutant, B9 Ser----Asp, B27 Thr----Glu has been investigated in D2O. At 1 to 5 mM concentrations, this mutant insulin is monomeric above pH 7.5. Coupling and amino acid classification of all aromatic signals is established via a combination of homonuclear one- and two-dimensional methods, including COSY, multiple quantum filters, selective spin decoupling and pH titrations. By comparisons with other insulin mutants and with chemically modified native insulins, all resonances in the aromatic region are given sequence-specific assignments without any reliance on the various crystal structures reported for insulin. These comparisons also give the sequence-specific assignments of most of the aromatic resonances of the mutant insulins B16 Tyr----Glu, B27 Thr----Glu and B25 Phe----Asp and the chemically modified species des-(B23-B30) insulin and monoiodo-Tyr A14 insulin. Chemical dispersion of the assigned resonances, ring current perturbations and comparisons at high pH have made possible the assignment of the aromatic resonances of human insulin, and these studies indicate that the major structural features of the human insulin monomer (including those critical to biological function) are also present in the monomeric mutant.     

Perturbation of insulin-receptor interactions by intramolecular hormone cross-linking. Analysis of relative movement among residues A1, B1, and B29

We have evaluated, by use of isolated canine hepatocytes, the importance of intramolecular hormone cross-linking (and of concomitant changes in molecular flexibility) to the interaction of insulin with its plasma membrane receptor. Cross-linked hormone analogs were prepared by reacting porcine insulin, N alpha A1-t-butyloxycarbonyl insulin or N alpha A1-t-butyloxycarbonyl [D-LysA1]insulin with various dicarboxylic acid active esters to obtain alpha-GlyA1/epsilon-LysB29-, alpha-PheB1/epsilon-LysB29-, and epsilon-D-LysA1/epsilon-LysB29-cross-linked insulins, respectively. In the aggregate, insulin analogs cross-linked by groups containing 2-12 atoms retained 1.4-35% of the receptor binding potency of native insulin. Analysis of our results suggests that: (a) loss of chemical functionality, steric interference, and restriction of potential intramolecular movement can all play roles in determining the receptor binding potencies of cross-linked insulin analogs; (b) restriction of intramolecular movement between residues A1 and B29 affects negatively the binding of insulin to its receptor (but accounts for only a fraction of the conformational change which insulin must undergo to achieve a high affinity state of ligand-receptor interaction); and (c) introduction of a cross-link between residues B1 and B29 (residues that are in fact in proximity in one crystalline form of the hormone) decreases markedly the receptor binding potencies of the corresponding analogs. The importance of these findings is discussed in relation to the potential structure of insulin when it is bound to its plasma membrane receptor.     

Soluble, prolonged-acting insulin derivatives. II. Degree of protraction and crystallizability of insulins substituted in positions A17, B8, B13, B27 and B30

It has previously been found that insulins, to which positive charge has been added by substitutions in position B30, thus raising the isoelectric point towards pH 7, had a prolonged action when injected as slightly acidic solutions because such derivatives crystallize very readily upon neutralization. Positive charge has now been added by substituting the B13 and A17 glutamic acid residues with glutamines and B27 threonine with lysine or arginine. These substitutions were introduced by site-specific mutagenesis in a gene coding for a single-chain insulin precursor. By tryptic transpeptidation the single-chain precursors were transformed to the double-chain insulin structure, concomitantly with incorporation of residue B30. Thus insulins combining B13 glutamine, A17 glutamine and B27 lysine or arginine with B30 threonine, threonine amide or lysine amide were synthesized. The time course of blood glucose lowering effect and the absorption were studied after subcutaneous injection in rabbits and pigs. The prolonged action of B30-substituted insulins was markedly enhanced by B27 lysine or arginine substitutions and by B13 glutamine. The B27 residue is located on the surface of the hexamer, so a basic residue in this position presumably promotes the packing of hexamers at neutral pH. The B13 residues cluster in the centre of the hexamer. When the electrostatic repulsive forces from six glutamic acid residues are abolished by substitution with glutamine, a stabilization of the hexamer can be envisaged.(ABSTRACT TRUNCATED AT 250 WORDS)     

Isolation, primary structure, and biological and immunological properties of pink and chum salmon insulins

1. Insulins have been isolated from islet tissue of pink (Oncorhynchus gorbuscha) and chum (Oncorhynchus keta) salmon. The primary structure of chum and pink salmon insulins was found to be identical. Compared to the amino acid sequence of human insulin, the salmon insulins under study differed at 14 positions. 2. Biological activity of pink salmon insulin was 83% of that of standard porcine insulin. 3. The immunological properties of fish insulins were investigated in specific radioimmunoassay (RIA) systems, based on porcine and pink salmon insulins. 4. A significant difference in the antigenic determinants of these fish and mammalian hormones was revealed.     

Preparation of N,N-bis(methylsulphonylethoxycarbonyl)insulins (author's transl)]

The preparation of N,N-bis(methylsulfonylethoxycarbonyl)insulins is described. In an aequeous buffer at pH 5.8 selectivity of the reaction of insulin with 20 equivalents of N-(methysulfonylethoxycarbonyloxy)succinimide (Msc-ONSu) leads very specifically to N alpha A 1,-N alpha B 1-(Msc)2 - insulin. The product can be isolated in a yield of 60%. Using N alpha A 1-citraconylinsulin the N alpha B 1, NEB29-(Msc)2 -insulin can be prepared in a yield of 40% based on insulin.     

The effect of chronic ethanol ingestion on synthesis and degradation of soluble, contractile and stromal protein fractions of skeletal muscles from immature and mature rats.

Both insulin and glucagon from the pancreas of the holocephalan cartilaginous fish Callorhynchus milii (elephantfish) have been isolated and purified. Two reverse-phase h.p.l.c. steps enabled recovery of sufficient material for gas-phase sequencing of the intact chains as well as peptide digestion products. The elephantfish insulin sequence shows 14 differences from pig insulin, including two unusual substitutions, Val-A14 and Gln-B30, though none of these is thought likely to influence receptor binding significantly. The insulin B-chain contains 31 residues, one more than mammalian insulins, but markedly less than that of the closely related ratfish with which it otherwise exhibits high sequence similarity. Elephantfish and pig glucagons differ at only four positions, but there are six changes from the ratfish glucagon-36 (normal glucagon contains 29 residues) sequence. It is apparent that different prohormone proteolytic processing mechanisms operate in the two holocephalan species.     

X-ray analysis of the single chain B29-A1 peptide-linked insulin molecule. A completely inactive analogue.

A crystal structure of a totally inactive insulin molecule has been determined. For this insulin molecule, the first without detectable activity to be characterized, the A and B-chains are linked by a peptide bond between A1 Gly and B29 Lys. The molecule has retained all its normal self-association properties and it can also accommodate the two different conformations designated T and R, as seen in 4Zn native pig insulin crystals. The hexamers of the crosslinked insulin molecule were crystallized using the 4Zn insulin recipe of Schlichtkrull. The structure has been crystallographically refined with data extending to 2 A using restrained least-square methods. Comparison of the B29-A1 peptide crosslink insulin and the 4Zn native insulin reveals close structural similarities with the native dimer. The analysis of the structure confirms the earlier hypothesis that insulin structures in crystals are not in an active conformation and that a separation of N-terminal A-chain and C-terminal B-chain is required for interaction with the insulin receptor.     

Comparative studies on the dynamics of crosslinked insulin

Molecular dynamics simulations were carried out on an insulin crosslinked between the N-terminal A chain and the C-terminal B chain to form a so-called mini-proinsulin: N -^A1-N -^B29-diaminosuberoyl insulin (DASI). To investigate the influence of crosslinking on the dynamics of the insulin moiety, the bridge was removed from a transient DASI structure and simulation was carried on independently with the then unlinked (ULKI) as well as with the crosslinked species. The effects of crystal packing and quaternary interactions were checked by simulating both types of monomers and dimers known from the hexamer structure. All simulations were compared to previous ones of native insulin. DASI shows general similarity to the native simulations in most parts of the structure. Deviations are visible in the segments to which the bridge is directly connected, i.e. their flexibility is reduced. Upon removal of the bridge the ULKI simulations reapproach those of native insulin. The influence of the bridge spreads over the whole molecule, but all of its main structural features remain intact. The simulations suggest that the displacement of the C-terminal B chain of native insulin, considered important for receptor interaction, is prevented by the bridge, which also partially shields some binding residues. This is in accordance with the poor biological potency of A1-B29-crosslinked insulins.Abbreviations DASI-insulin(DASI) bovineN -^A1-N -^B29-di-aminosuberoyl insulin - ULK-insulin (ULKI) Native beef insulin with the bridge of DASI removed     

Binding of insulin to bovine liver plasma membrane. Use of insulin analogues modified at the A1 residue

Bovine liver plasma membranes [Rösen, Ehrich, Junger, Bubenzer & Kühn (1979) Biochim. Biophys. Acta 587, 593–605] show similar insulin-binding characteristics, as evaluated by Scatchard analysis, to those of membrane systems from other species. However, the dissociation rate of bound insulin cannot be accelerated by the addition of insulin, in contrast with membranes isolated from rat liver. The dissociation rate is strongly dependent on the pH. Although dependent on temperature, the total capacity of binding sites is minimally changed, but the number of high-affinity sites is increased 2–3-fold, by lowering the incubation temperature. These data might be interpreted by assuming a single population of receptors whose distribution between different affinity states depends on temperature. In competition studies, most of the modified insulins examined show a close correlation between binding, determined in plasma membranes from bovine liver, and biological activity, measured in adipocytes. The hypothesis that a positive charge on the A1 residue may be favourable for binding is supported by experiments with an isosteric pair of insulins modified at this residue ([carbamoyl-Gly^A1]- and [amidino-Gly^A1]insulin) and with modified insulins carrying one or more positive charges on the A1 residue ([Arg-Gly^A1]-, [Arg-Arg-Gly^A1]-, [Arg-Arg-Arg-Gly^A1]- and [Lys-Arg-Gly^A1]insulin). The latter insulin derivatives show a higher binding activity for plasma membranes from bovine, porcine and rat liver than expected from their biological activities in adipocytes.     

Soluble, prolonged-acting insulin derivatives. III. Degree of protraction, crystallizability and chemical stability of insulins substituted in positions A21, B13, B23, B27 and B30

It was previously demonstrated that insulins to which positive charge has been added by substituting B13 glutamic acid with a glutamine residue, B27 threonine with an arginine or lysine residue, and by blocking the C-terminal carboxyl group of the B-chain by amidation, featured a prolonged absorption from the subcutis of rabbits and pigs after injection in solution at acidic pH. The phenomenon is ascribed to a low solubility combined with the readiness by which these analogs crystallize as the injectant is being neutralized in the tissue. However, acid solutions of insulin are chemically unstable as A21 asparagine both deamidates to aspartic acid and takes part in formation of covalent dimers via alpha-amino groups of other molecules. In order to circumvent the instability, substitutions were introduced in position A21, in addition to those in B13, B27 and B30, challenging the fact that A21 asparagine has been conserved in this position throughout the evolution. Biological potency was retained when glycine, serine, threonine, aspartic acid, histidine and arginine were introduced in this position, although to a varying degree. In the crystal structure of insulin a hydrogen bond bridges the alpha-nitrogen of A21 with the backbone carbonyl of B23 glycine. In order to investigate the importance of this hydrogen bond for biological activity a gene for the single-chain precursor B-chain(1-29)-Ala-Ala-Lys-A-chain(1-21) featuring an A21 proline was synthesized. However, this single-chain precursor failed to be properly produced by yeast, pointing to the formation of this hydrogen bond as an essential step in the folding process. The stability of the A21-substituted analogs in acid solutions (pH 3-4) with respect to deamidation and formation of dimers was approximately 5-10 times higher than that of human insulin in neutral solution. The rate of absorption of most insulins is decreased by increasing the Zn2+ concentration of the preparation. However, one analog with A21 glycine showed first-order absorption kinetics in pigs with a half-life of approximately 25 h, independent of the Zn2+ concentration. The day-to-day variation of the absorption of this analog was significantly lower than that of the conventional insulin suspensions, a property that might render such an insulin useful in the attempts to improve glucose control in diabetics by a more predictable delivery of basal insulin.     

Insulin analogues with permuted A chain N-terminus (author's transl)]

By partial synthesis insulin analogues were prepared in which the amino acid in position 1 of the A chain was permuted. Glycine in position A 1 was exchanged for leucine, tert.- butyloxycarbonylvaline, valine, proline, lysine as well as glutamic acid. Two pathways of partial synthesis were followed: Firstly, des-1-glycine-A-chain S-sulfonate was reacted with active esters of tert.-butyloxycarbonylamino acids. The ensuing modified A-chains were combined with natural B-chain to give A1-permuted insulins. In the second procedure, the preparation of tris-Boc-[A1-leucine]insulin was accomplished by reaction of Boc-leucine N-hydroxysuccinimide ester with NalphaB1,NepsilonB29-bis(tert.-butyloxycarbonyl)-des-A1-glycine-insulin. The protected insulin derivative had been prepared by combination of des-glycine-A-chain with Nalpha1,Nepsilon29-bis(tert.-butyloxycarbonyl)-B-chain. The deprotected analogues differed considerably in their CD-spectra from insulin and possessed low in vitro biological activities of 2.5-17%. Crystallization attempts failed. Thus, the introduction of side chains in position A1 distorts the conformation sterically and decreases the biological activity.     

Insulins with built-in glucose sensors for glucose responsive insulin release.

Derivatization of insulin with phenylboronic acids is described, thereby equipping insulin with novel glucose sensing ability. It is furthermore demonstrated that such insulins are useful in glucose‐responsive polymer‐based release systems. The preferred phenylboronic acids are sulfonamide derivatives, which, contrary to naïve boronic acids, ensure glucose binding at physiological pH, and simultaneously operate as handles for insulin derivatization at LysB29. The glucose affinities of the novel insulins were evaluated by glucose titration in a competitive assay with alizarin. The affinities were in the range 15–31 mM (K_d), which match physiological glucose fluctuations. The dose‐responsive glucose‐mediated release of the novel insulins was demonstrated using glucamine‐derived polyethylene glycol polyacrylamide (PEGA) as a model, and it was shown that Zn(II) hexamer formulation of the boronated insulins resulted in steeper glucose sensitivity relative to monomeric insulin formulation. Notably, two of the boronated insulins displayed enhanced insulin receptor affinity relative to native insulin (113%–122%) which is unusual for insulin LysB29 derivatives. Copyright © 2004 European Peptide Society and John Wiley & Sons, Ltd.     

cDNAs from neurosecretory cells of brains of Locusta migratoria (Insecta, Orthoptera) encoding a novel member of the superfamily of insulins.

From neurohaemal lobes of corpora cardiaca of Locusta migratoria a 5-kDa peptide has been isolated and its sequence established [see the accompanying paper, by Hietter et al. (1990) Eur. J. Biochem. 187, 241-247]. We have designed oligonucleotide probes from the peptide sequence of this molecule and screened a library prepared from mRNA of the neurosecretory cell region of the brain of this insect. Several positive cDNAs were isolated, the combined nucleotide sequences of which predict a large precursor of 145 residues (15770 Da) containing the newly isolated 5-kDa peptide. The peptide is flanked by regions homologous to the A and B chains of the superfamily of insulins. The overall organization of the precursor is as follows: signal peptide/domain homologous to the B chain of insulins/C (connecting)-peptide (corresponding to the newly isolated 5-kDa peptide)/domain homologous to the A chain of insulins. The numbers and relative positions of the cysteines of the Locusta peptide are equivalent to those of the other members of the insulin superfamily and most of the hydrophobic core residues are conserved.     

Soluble, prolonged-acting insulin derivatives. I. Degree of protraction and crystallizability of insulins substituted in the termini of the B-chain

Hydrophilic insulins, more positively charged than human insulin at neutral pH, have been prepared by substitution with basic amino acids at the termini of the B-chain and by blocking the C-terminal carboxyl group of the B-chain. The isoelectric pH of the insulin is thereby moved from 5.4 towards physiological levels. Slightly acid solutions of derivatives, in which charge has been added in the C-terminus of the B-chain, have a prolonged action in vivo, in particular if the carboxyl group is blocked. It is found that the prolonged-acting hydrophilic insulins crystallize instantly when the pH is adjusted to 7. The prolonged action is ascribed to this readiness to crystallization combined with a low solubility, which may be further decreased by increased concentration of zinc ions. Hydrophobic insulins have a prolonged action independent of the site of substitution even if the derivative is soluble at physiological pH. Some derivatives were prepared from porcine insulin by tryptic transpeptidation. N-terminal B-chain substituted insulins were prepared by alkylation of a biosynthetic single-chain insulin precursor, followed by tryptic transpeptidation rendering the double chain insulin derivative. The observed blood glucose lowering in the rabbits implies that neither N- nor C-terminal B-chain substitution results in substantial deterioration of biological potency. An index for the degree of protraction based on the blood glucose data is used to compare the insulins.     

Atomic force microscopy of crystalline insulins: the influence of sequence variation on crystallization and interfacial structure.

The self-association of proteins is influenced by amino acid sequence, molecular conformation, and the presence of molecular additives. In the presence of phenolic additives, LysB28ProB29 insulin, in which the C-terminal prolyl and lysyl residues of wild-type human insulin have been inverted, can be crystallized into forms resembling those of wild-type insulins in which the protein exists as zinc-complexed hexamers organized into well-defined layers. We describe herein tapping-mode atomic force microscopy (TMAFM) studies of single crystals of rhombohedral (R3) LysB28ProB29 that reveal the influence of sequence variation on hexamer-hexamer association at the surface of actively growing crystals. Molecular scale lattice images of these crystals were acquired in situ under growth conditions, enabling simultaneous identification of the rhombohedral LysB28ProB29 crystal form, its orientation, and its dynamic growth characteristics. The ability to obtain crystallographic parameters on multiple crystal faces with TMAFM confirmed that bovine and porcine insulins grown under these conditions crystallized into the same space group as LysB28ProB29 (R3), enabling direct comparison of crystal growth behavior and the influence of sequence variation. Real-time TMAFM revealed hexamer vacancies on the (001) terraces of LysB28ProB29, and more rounded dislocation noses and larger terrace widths for actively growing screw dislocations compared to wild-type bovine and porcine insulin crystals under identical conditions. This behavior is consistent with weaker interhexamer attachment energies for LysB28ProB29 at active growth sites. Comparison of the single crystal x-ray structures of wild-type insulins and LysB28ProB29 suggests that differences in protein conformation at the hexamer-hexamer interface and accompanying changes in interhexamer bonding are responsible for this behavior. These studies demonstrate that subtle changes in molecular conformation due to a single sequence inversion in a region critical for insulin self-association can have a significant effect on the crystallization of proteins.     

Agonism and Antagonism at the Insulin Receptor

Insulin can trigger metabolic as well as mitogenic effects, the latter being pharmaceutically undesirable. An understanding of the structure/function relationships between insulin receptor (IR) binding and mitogenic/metabolic signalling would greatly facilitate the preclinical development of new insulin analogues. The occurrence of ligand agonism and antagonism is well described for G protein-coupled receptors (GPCRs) and other receptors but in general, with the exception of antibodies, not for receptor tyrosine kinases (RTKs). In the case of the IR, no natural ligand or insulin analogue has been shown to exhibit antagonistic properties, with the exception of a crosslinked insulin dimer (B29-B’29). However, synthetic monomeric or dimeric peptides targeting sites 1 or 2 of the IR were shown to be either agonists or antagonists. We found here that the S961 peptide, previously described to be an IR antagonist, exhibited partial agonistic effects in the 1–10 nM range, showing altogether a bell-shaped dose-response curve. Intriguingly, the agonistic effects of S961 were seen only on mitogenic endpoints (³H-thymidine incorporation), and not on metabolic endpoints (¹⁴C-glucose incorporation in adipocytes and muscle cells). The agonistic effects of S961 were observed in 3 independent cell lines, with complete concordance between mitogenicity (³H-thymidine incorporation) and phosphorylation of the IR and Akt. Together with the B29-B’29 crosslinked dimer, S961 is a rare example of a mixed agonist/antagonist for the human IR. A plausible mechanistic explanation based on the bivalent crosslinking model of IR activation is proposed.     

N epsilonB29-(+)-biotinylinsulin and its complexes with avidin. Synthesis and biological activity

Insulin was modified with d-biotin-N-hydroxysuccinimide ester in dimethylformamide. Mono-, di-, and triacylated insulins were separated by preparative isoelectric focusing. Monoacylated derivatives (isoelectric point 5.1) were fractionated twice on DEAE-cellulose to yield pure N epsilonB29-biotinylinsulin. The structure of the product was established by amino acid analysis before and after deamination. N epsilonB29-biotinylinsulin had biological activity indistinguishable from insulin on glucose oxidation and lipid synthesis assays using isolated rat epididymal fat cells. Complexes of N epsilonB29-biotinylinsulin with avidin, having essentially all but one binding site filled with biotin, were prepared in order to obtain a 1:1 insulin:avidin ration. The elicited identical maximal biological responses, but showed a potency decreased to 5% of that of insulin. Such complexes conjugated with ferritin will provide a useful tool in the development of electron microscopic stains of insulin receptors.     

The C domain of translation termination factor eRF1 is close to the stop codon in the A site of the 80S ribosome

The arrangement of the stop codon and its 3′-flanking codon relative to the components of translation termination complexes of human 80S ribosomes was studied using mRNA analogs containing the stop signal UPuPuPu (Pu is A or G) and the photoreactive perfluoroarylazido group, which was linked to a stop-signal or 3′-flanking nucleotide (positions from +4 to +9 relative to the first nucleotide of the P-site codon). Upon mild UV irradiation, the analogs crosslinked to components of the model complexes, mimicking the state of the 80S ribosome at translation termination. Termination factors eRF1 and eRF3 did not change the relative arrangement of the stop signal and 18S rRNA. Crosslinking to eRF1 was observed for modified nucleotides in positions +5 to +9 (that for stop-codon nucleotide +4 was detected earlier). The eRF1 fragments crosslinked to the mRNA analogs were identified. Fragment 52–195, including the N domain and part of the M domain, crosslinked to the analogs carrying the reactive group at A or G in positions +5 to +9 or at the terminal phosphate of nucleotide +7. The site crosslinking to mRNA analogs containing modified G in positions +5 to +7 was assigned to eRF1 fragment 82–166 (beyond the NIKS motif). All but one analog (that with modified G in position +4) crosslinked to the C domain of eRF1 (fragment 330–422). The efficiency of crosslinking to the C domain was higher than to the N domain in most cases. It was assumed that the C domain of eRF1 bound in the A site is close to nucleotides +5 to +9, especially +7 and +8, and that eRF1 undergoes substantial conformational changes when binding to the ribosome.     

Photoreactive insulin analogues used to characterise the insulin receptor

Three photoreactive insulin analogues (“photoprobes”) have been prepared in which an aryl azide group was substituted at either the A1, B1 or B29 positions of the insulin molecule. When incubated with rat liver plasma membranes and irradiated all three photoprobes covalently labelled specific insulin binding sites within the membrane. SDS-polyacrylamide gel electrophoresis of plasma membranes covalently tagged with either of the three ¹²⁵I-photoprobes resolved one major specifically labelled polypeptide with an apparent molecular weight of 130,000. The labelled polypeptide migrated anomalously in SDS-polyacrylamide gels and a molecular weight of 90,000 for the polypeptide was determined from a ‘Ferguson’ plot using the combined results from gels of different acrylamide concentrations. Column chromatography of detergent solubilised photoprobe-labelled membranes indicated that the labelled polypeptide may be a subunit of a larger protein complex.