-Crystallin polymers and polymerization: the view from down under

Several models have been proposed for the quaternary structure of -crystallin. Some suggest the subunits are arranged in concentric shells. Others propose that the subunits are in a micelle-like arrangement. However, none is able to satisfactorily account for all observations on the protein and the quaternary structure of -crystallin remains to be established. In this review, factors contributing to the assembly and polymerization are examined in order to evaluate the different models. Consideration of the variations in particle size and molecular weight under different conditions leads to the conclusion that -crystallin cannot be a micelle or a layered structure. Instead, it is suggested that the protein may be assembled from a ‘monomeric’ unit comprising eight subunits arranged in two tetramers with cyclic symmetry. The octameric unit is proposed to be disc-like particle with a diameter of 9.5 nm and a height of 3 nm. The larger particles, chains and sheet-like structures commonly observed are assembled from the octamers. Structural predictions indicate that the polypeptide may be folded into three independent domains which have different roles in the structural organization and functions of the protein. It is suggested that the tetramers are stabilized through interactions involving the second domain (residues 64–104) while assembly into the octamers and higher polymers requires hydrophobic interactions involving the N-terminal domain. Deletion of parts of this domain by site directed mutagenesis revealed that residues 46–63 play a critical role in the assembly. Current research aims to identify the specific amino acids involved.     

Direct visualization of the structure of the "20 S" aggregate of coat protein of tobacco mosaic virus. The "disk" is the major structure at pH 7.0 and the Proto-helix at lower pH.

We have employed the rapid-freeze technique to prepare specimens for electron microscopy of a coat protein solution of tobacco mosaic virus at equilibrium at pH 7.0 and 6.8, ionic strength 0.1 M and 20 degrees C. The former are the conditions for the most rapid assembly of the virus from its isolated protein and RNA. At both pH values, the equilibrium mixture contains approximately 80% of a "20 S" aggregate and 20% of a "4 S" aggregate (the so-called A-protein). The specimens were prepared either totally unstained or positively stained with methyl mercury nitrate, which binds to an amino acid residue (Cys27) internally located within the subunit, which we show not to affect the virus assembly. The images in the electron microscope are compatible only with the major structure for the "20 S" aggregate at pH 7.0 containing two rings of subunits and these aggregates display the same binding contacts as those seen between the aggregate that forms the asymmetric unit in the crystal, which has been shown by X-ray crystallography to be a disk containing two rings, each of 17 subunits, oriented in the same direction. In contrast, the images from specimens prepared at pH 6.8 show the major structure to be a proto-helix at this slightly lower pH, demonstrating that the technique of cryo-electron microscopy is capable of distinguishing between these aggregates of tobacco mosaic virus coat protein. The main structure in solution at pH 7.0 must therefore be very similar to that in the crystal, although slight differences could occur and there are probably other, minor, components in a mixture of species sedimenting around 20 S under these conditions. The equilibrium between aggregates is extremely sensitive to conditions, with a drop of 0.2 pH unit tipping the disk to proto-helix ratio from approximately 10:1 at pH 7.0 to 1:10 at pH 6.8. This direct determination of the structure of the "20 S" aggregate in solution, under conditions for virus assembly, contradicts some recent speculation that it must be helical, and establishes that, at pH 7.0, it is in fact predominantly a two-layer disk as it had been modelled before.     

Delayed Osmotic Effect on in Vitro Assembly of RuBisCO : Relationship to Large Subunit-Binding Protein Complex Dissociation

Higher plant ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) cannot reassociate after dissociation, and its subunits do not assemble into active RuBisCO when synthesized in Escherichia coli. Newly synthesized subunits of RuBisCO are associated with a high molecular weight binding protein complex in pea chloroplasts. The immediate donor for large subunits which assemble into RuBisCO is a low molecular weight complex which may be derived from the high molecular weight binding protein complex. When the high molecular weight binding protein complex is diluted, it tends to dissociate, forming low molecular weight complexes. When the large subunit-binding protein complexes were examined after in organello protein synthesis, it was found that the low molecular weight complexes were more abundant when protein synthesis was carried out under hypotonic conditions. This increase in the assembly competent population of low molecular weight large subunit complexes can account for the increased amount of in vitro RuBisCO assembly which occurs under these conditions. The data indicate that the assembly of large subunits into RuBisCO is a function of the aggregation state of the large subunit binding protein complex during protein synthesis. This implies that the binding protein exerts its effects during or shortly after large subunit synthesis.     

Proteolytic Cleavage of Subunits of the Nucleocapsid of the Paramyxovirus Simian Virus 5

The protein subunits of the nucleocapsid of the parainfluenza virus simian virus 5 isolated from infected cells after dispersion with trypsin, chymotrypsin, or ficin are cleaved proteolytically. The molecular weights of the subunits which result from cleavage depend on the enzyme used, but are around 43,000, compared to the native subunit of 61,000. In most instances cleavage of the subunit appears to be due to the protease used to disperse the cell, and follows cell disruption. Nucleocapsids composed of native, uncleaved subunits can frequently be obtained from infected cells dispersed without a proteolytic enzyme; however, cleavage occasionally occurs even under those conditions, indicating that cellular proteases can at times cleave this protein. Nucleocapsids containing uncleaved subunits can be isolated from cells persistently infected with simian virus 5, indicating that persistent infection is not invariably associated with intracellular cleavage of this protein. Nucleocapsids composed of native subunits are hydrophobic, whereas those composed of the cleaved subunit can be dispersed in aqueous solution. It is suggested that the portion of the molecule removed by cleavage may be responsible for a specific interaction during virus assembly between the nucleocapsid and those areas of plasma membrane which contain the non-glycosylated viral membrane protein, which is also hydrophobic. An amino acid analysis of native and cleaved subunits has been done. The portion of the subunit removed by cleavage does not have a high proportion of hydrophobic residues, suggesting that those present are arranged together to form a hydrophobic domain.The N termini of both the native and cleaved subunits are blocked. This suggests that the portion of the molecule which is externally disposed and removed by cleavage contains the C terminus, and the cleaved subunit which reacts with the viral RNA contains the N terminus.     

cAMP-dependent protein kinases I and II: divergent turnover of subunits

cAMP-dependent protein kinase subunits were isolated from livers of rats that had been subjected to biosynthetic labeling with radioactive leucine. By application of ligand and antibody affinity techniques pure regulatory (R I; R II) and catalytic (C) subunits could be obtained in high yields, which allowed measurement of the apparent degradation rate constants and half-lives following a double isotope labeling protocol. In this way marked differences of apparent half-lives of regulatory subunits R I (t1/2 = 31 h) and R II (t1/2 = 125 h) were observed. To avoid the negative influence of reutilization inherent in the decay experiments, specific radioactivities were determined after a short isotope pulse. This parameter, which under steady-state conditions reflects the fractional turnover rate of the subunits, was found to be different for all three protein kinase subunits. Relative to total liver protein, the ratios R I:R II:C corresponded to 3.9:0.6:2. Our data indicate that in each type of protein kinase isoenzymes regulatory and catalytic subunits turn over with similar rates. The type I isoenzyme, however, is renewed much faster than protein kinase II. Furthermore, our findings are consistent with the thesis that free subunits as generated by activation are more susceptible to degradation than the holoenzymes, leading under steady-state conditions to compensatory resynthesis. Since renewal of R I exceeded that of R II also in two other tissues, the elevated turnover of protein kinase I as an indicator of preferential activation appears to be a general phenomenon. The different turnover of the two isoenzymes, then, may relate to different cellular functions like modulation of enzyme activity vs. modulation of gene activity.     

The "gamma subunit" of Na,K-ATPase: a small, amphiphilic protein with a unique amino acid sequence

The "gamma subunit", or "proteolipid", of Na,K-ATPase is a small, membrane-bound protein that copurifies with the alpha and beta subunits of this enzyme. The importance of gamma in the function of Na,K-ATPase remains to be established, but some evidence indicates that it may be involved in forming a receptor site for cardiac glycosides. We have previously communicated [Reeves, A. S., Collins, J. H., & Schwartz, A. (1980) Biochem. Biophys. Res. Commun. 95, 1591-1598] the purification and amino acid composition of sheep kidney gamma, and in this paper we present the first available sequence information on this protein. Although the amino terminus of gamma seems to be blocked and it is resistant to proteolytic cleavage, we have determined approximately half of its amino acid sequence. Our results indicate that gamma contains a total of 68 amino acid residues, with a calculated Mr of 7675. The sequenced portion appears to be at the carboxyl terminus of the polypeptide chain. The gamma sequence is unique, providing strong evidence for its homogeneity and establishing for the first time that it is not a breakdown product of the alpha or beta subunits. gamma is not a true proteolipid, but rather it is an amphiphilic protein with two distinct structural domains. The amino-terminal domain (residues 1-49) is very hydrophilic, with many charged amino acid side chains, and must be extracellular. This domain includes a concentrated segment of four aromatic residues which may be involved in glycoside binding. The carboxyl-terminal domain (residues 50-68) is hydrophobic and probably spans the cell membrane.     

Mitochondrial ATP Synthase: A Bioinformatic Approach Reveals New Insights About the Roles of Supernumerary Subunits <Emphasis Type="Italic">g</Emphasis> and A6L

The mitochondrial ATP synthase is a membrane protein complex which couples the proton gradient across the mitochondrial inner membrane to the synthesis of ATP from ADP + P_i. The complex is composed of essential subunits for its motor functions and supernumerary subunits, the roles of which remain to be elucidated. Subunits g and A6L are supernumerary subunits, and the specific roles of these subunits are still matters of debate. To gain insight into the functions of these two subunits, we carried out the alignment and the homolog search of the protein sequences of the subunits and found the following features: Subunit g appears to have isoforms in animals, and the transmembrane domain of the animal subunit g contains a completely conserved acidic residue in the middle of a helix on the conserved side of the transmembrane helix. This finding implicates the conserved acidic residue as important for the function of subunit g. The alignment of A6L protein sequences shows a conserved aromatic residue at the N-terminal domain with which the N-terminal MPQL sequence comprises a unique MPQLX₄Ar motif that can signify the protein A6L. The conserved aromatic residue may also be important for the function of A6L.     

Structures and roles of the polymorphic forms of tobacco mosaic virus protein: VI. Assembly of the nucleoprotein rods of tobacco mosaic virus from the protein disks and RNA

The assembly of tobacco mosaic virus involves a preformed protein aggregate, the disk, which consists of two rings each of 17 protein subunits, as the sole protein source. The kinetics of this assembly have been studied, using both tobacco mosaic virus RNA, which causes a rapid initiation and so enables growth to be studied, and also polyadenylic acid, with which initiation is slowed down and thus can be partially resolved from growth. Two disks interact with a special nucleotide sequence at the 5′-hydroxyl end of a single tobacco mosaic virus RNA molecule to initiate the formation of the viral nucleoprotein helix, which then grows by the addition of further disks. All of the subunits from these further disks are incorporated into the helix, so that growth proceeds by the co-operative addition of 34 subunits at a time. Under the conditions used, rearrangement of each disk takes about six seconds, giving a total time for the growth of a complete virus particle of just over six minutes.     

Unfolding of C-phycocyanin followed by loss of non-covalent chromophore-protein interactions: 1. Equilibrium experiments

Optical spectroscopic properties of the covalently linked chromophores of biliproteins are profoundly influenced by the state of the protein. This has been used to monitor the urea-induced denaturation of C-phycocyanin (CPC) from Mastigocladus laminosus and its subunits. Under equilibrium conditions, absorption, fluorescence and circular dichroism of the chromophores were monitored, as well as the circular dichroism of the polypeptide. Treatment of CPC trimers (αβ)₃ resulted first in monomerization (αβ), which was followed by a complex unfolding process of the protein. Loss of chromophore fluorescence is the next process at increasing urea concentrations; it indicates increased flexibility of the chromophore while maintaining the native, extended conformation, and a less compact but still native-like packing of the protein in the regions sampled by the chromophores. This was followed by relaxation of the chromophores from the energetically unfavorable extended to a cyclic-helical conformation, as reported by absorption and CD in the visible range, indicating local loss of protein structure. Only then is the protein secondary structure lost, as reported by the far-UV CD. Sequential processes were also seen in the subunits, where again the chromophore-protein interactions were reduced before the unfolding of the protein. It is concluded that the bilin chromophores are intrinsic probes suitable to differentiate among different processes involved in protein denaturation.     

Energetic Calculations to Decipher pH-Dependent Oligomerization and Domain Swapping of Proteins

Domain swapping mechanism is a specialised mode of oligomerization of proteins in which part of a protein is exchanged in a non-covalent manner between constituent subunits. This mechanism is highly affected by several physiological conditions. Here, we present a detailed analysis ofthe effect of pH on different regions of the domain swapped oligomer by considering examples which are known to be sensitive to pH in transiting from monomeric to domain-swapped dimeric form. The energetic calculations were performed using a specialized method which considers changes in pH and subsequent changes in the interactions between subunits. This analysis provides definitive hints about the pH-dependence switch from monomer to domain-swapped oligomer and the steps that may be involved in the swapping mechanism.     

Differential detergent solubility investigation of thermally induced transitions in cytochrome c oxidase

The thermal denaturation of membrane-reconstituted cytochrome c oxidase (EC 1.9.3.1) occurs at approximately 63 degrees C as determined by high-sensitivity differential scanning calorimetry. The heat capacity profile associated with this process is characterized by the presence of two well-defined peaks, indicating that all the enzyme subunits do not have the same thermal stability. This thermal denaturation of the enzyme complex is coupled to a change in its solubility properties. This change in solubility allows separation of the native and denatured protein fractions by detergent solubilization followed by centrifugation under conditions in which only the native fraction is solubilized. Using this principle, it has been possible to study the denaturation of membrane-reconstituted cytochrome c oxidase and quantitatively identify the protein subunits undergoing thermal denaturation using computer-assisted gel electrophoresis analysis. This technique allows calculation of single-subunit thermal denaturation profiles within the intact enzyme complex, and as such, it can be used to obtain transition temperatures, molecular populations, and van't Hoff enthalpy changes for individual protein subunits, thus complementing results obtained by high-sensitivity differential scanning calorimetry.     

A functional protein pore with a "retro" transmembrane domain.

Extended retro (reversed) peptide sequences have not previously been accommodated within functional proteins. Here, we show that the entire transmembrane portion of the beta-barrel of the pore-forming protein alpha-hemolysin can be formed by retrosequences comprising a total of 175 amino acid residues, 25 contributed by the central sequence of each subunit of the heptameric pore. The properties of wild-type and retro heptamers in planar bilayers are similar. The single-channel conductance of the retro pore is 15% less than that of the wild-type heptamer and its current-voltage relationship denotes close to ohmic behavior, while the wild-type pore is weakly rectifying. Both wild-type and retro pores are very weakly anion selective. These results and the examination of molecular models suggest that beta-barrels may be especially accepting of retro sequences compared to other protein folds. Indeed, the ability to form a retro domain could be diagnostic of a beta-barrel, explaining, for example, the activity of the retro forms of many membrane-permeabilizing peptides. By contrast with the wild-type subunits, monomeric retro subunits undergo premature assembly in the absence of membranes, most likely because the altered central sequence fails to interact with the remainder of the subunit, thereby initiating assembly. Despite this difficulty, a technique was devised for obtaining heteromeric pores containing both wild-type and retro subunits. Most probably as a consequence of unfavorable interstrand side-chain interactions, the heteromeric pores are less stable than either the wild-type or retro homoheptamers, as judged by the presence of subconductance states in single-channel recordings. Knowledge about the extraordinary plasticity of the transmembrane beta-barrel of alpha-hemolysin will be very useful in the de novo design of functional membrane proteins based on the beta-barrel motif.     

An electron microscopic approach to the quaternary structure of mitochondrial F1-ATPase

The three-dimensional structure of F1-ATPase from beef heart mitochondria was investigated by electron microscopic techniques. The presence of high concentrations of nucleotides is essential for preservation of the quaternary structure. When investigated under such conditions, monodisperse F1-ATPase could not be distinguished from the membrane-bound enzyme. At low resolution, the particle shape resembles an oblate ellipsoid of revolution with an axial ratio of about 2:1. From several lines of evidence (including field micrographs at higher magnifications, Markham rotational analysis, and tilting experiments), two conclusions may be drawn concerning the three-dimensional fine structure of F1-ATPase. 1. At the periphery of the molecule, six globular protein masses are orientated in a way similar to the chair conformation of cyclohexane. This array is interpreted to be made up of an alternating sequence of alpha and beta subunits. 2. Part of the central space is occupied by a seventh protein mass, protrusions of which are likely to be in contact with some of the outer subunits. A gamma subunit is supposed to be constituent part of this central protein mass. As a consequence, this model favours a stoichiometry of alpha 3 beta 3 gamma for the large subunits of beef heart F1-ATPase.     

Hydrophobic properties of the beta 1 and beta 2 subunits of the rat brain sodium channel

Voltage-sensitive sodium channels purified from rat brain in functional form consist of a stoichiometric complex of three glycoprotein subunits, alpha of 260 kDa, beta 1 of 36 kDa, and beta 2 of 33 kDa. The alpha and beta 2 subunits are linked by disulfide bonds. The hydrophobic properties of these three subunits were examined by covalent labeling with the photoreactive hydrophobic probe 3-(trifluoromethyl)-3-(m-[125I]iodophenyl)diazirine [( 125I]TID) which labels transmembrane segments in integral membrane proteins. All three subunits of the sodium channel were labeled by [125I]TID when the purified protein was solubilized in mixed micelles of Triton X-100 and phosphatidylcholine (4:1). The half-time for photolabeling was approximately 7 min consistent with the half-time of 9 min for photolysis of TID under our conditions. Comparable amounts of TID per mg of protein were incorporated into each subunit. Purified sodium channels reconstituted in phosphatidylcholine vesicles were also labeled by TID with comparable incorporation per mg of protein into all three subunits. The efficiency of photolabeling of the three subunits was reduced from 39 to 44% by a 2-fold expansion of the hydrophobic phase of the reaction mixture but was unaffected by a 2-fold expansion of the aqueous phase, confirming that the photolabeling reaction took place in the lipid phase of the vesicle bilayer. The hydrophobic properties of the sodium channel subunits were examined further using phase separation in the nonionic detergent Triton X-114. Under conditions in which beta 1 is dissociated from alpha, the beta 1 subunit was preferentially extracted into the Triton X-114 phase, and the disulfide-linked alpha beta 2 complex was retained in the aqueous phase. When the disulfide bonds between the alpha and beta 2 subunits were reduced with dithioerythritol, the beta 2 subunit was also preferentially extracted into the Triton X-100 phase leaving the free alpha subunit in the aqueous phase. A preparative method for isolation of the beta 1 and beta 2 subunits was developed based on this technique. Considered together, the results of our hydrophobic labeling and phase separation experiments indicate that the alpha, beta 1, and beta 2 subunits all have substantial hydrophobic domains that may interact with the hydrocarbon phase of phospholipid bilayer membranes. Since the alpha subunit is known to be a transmembrane protein with many potential membrane-spanning segments, we conclude that the beta 1 and beta 2 subunits are likely to also be integral membrane proteins with one or more membrane-spanning segments.(ABSTRACT TRUNCATED AT 400 WORDS)     

Effect of acetic acid on phycocyanin-phycoerythrocyanin mixture extracted from Anabaena variabilis

Exposure of a phycocyanin-phycoerythrocyanin mixture extracted from Anabaena variabilis to sodium acetate, pH 3.8, ionic strength of 0.1, results in dissociation of the phycoerythrocyanin's beta subunit from its alpha subunit. The alpha subunit obtained by this method has a strong absorption transition at 508 nm. This transition is a consequence of the subunit's specific conformation, rather than of a new chromophore. The behavior of the phycocyanin-phycoerythrocyanin mixture in acetate buffers of variable compositions suggests that interactions which involve carboxylic amino acid residues play an important role, along with hydrophobic associations, in the association of phycoerythrocyanin subunits into monomers (alpha beta) and between this protein and phycocyanin. This work also indicates that the linkage between alpha and beta subunits of phycoerythrocyanin is labile and may be weaker than the association of these subunits with phycocyanin under acidic conditions.     

Stable association of G proteins with beta 2AR is independent of the state of receptor activation.

beta 2-Adrenergic receptors expressed in Sf9 cells activate endogenous Gs and adenylyl cyclase [Mouillac B., Caron M., Bonin H., Dennis M. and Bouvier M. (1992) J. Biol. Chem. 267, 21733-21737]. However, high affinity agonist binding is not detectable under these conditions suggesting an improper stoichiometry between the receptor and the G protein and possibly the effector molecule as well. In this study we demonstrate that when beta 2-adrenergic receptors were co-expressed with various mammalian G protein subunits in Sf9 cells using recombinant baculoviruses signalling properties found in native receptor systems were reconstituted. For example, when beta 2AR was co-expressed with the Gs alpha subunit, maximal receptor-mediated adenylyl cyclase stimulation was greatly enhanced (60 +/- 9.0 versus 150 +/- 52 pmol cAMP/min/mg protein) and high affinity, GppNHp-sensitive, agonist binding was detected. When G beta gamma subunits were co-expressed with Gs alpha and the beta 2AR, receptor-stimulated GTPase activity was also demonstrated, in contrast to when the receptor was expressed alone, and this activity was higher than when beta 2AR was co-expressed with Gs alpha alone. Other properties of the receptor, including receptor desensitization and response to inverse agonists were unaltered. Using antisera against an epitope-tagged beta 2AR, both Gs alpha and beta gamma subunits could be co-immunoprecipitated with the beta 2AR under conditions where subunit dissociation would be expected given current models of G protein function. A desensitization-defective beta 2AR (S261, 262, 345, 346A) and a mutant which is constitutively desensitized (C341G) could also co-immunoprecipitate G protein subunits. These results will be discussed in terms of a revised view of G protein-mediated signalling which may help address issues of specificity in receptor/G protein coupling.     

Electron microscopy and image analysis of the GroEL-like protein and its complexes with glutamine synthetase from pea leaves.

The molecular structure of GroEL-like protein from pea leaves has been studied by electron microscopy and image analysis of negatively stained particles. Over 1500 molecular projections were selected and classified by multivariate statistical analysis. It was shown that the molecule consists of 14 subunits arranged in two layers with 72 point group symmetry. Side view projections of the molecule show a four-striation appearance, which subdivides both layers of seven subunits into two halves; this may be explained by a two-domain structure of the subunits. The presence in protein preparations of projections corresponding to one layer of subunits or half-molecules is consistent with the molecular structure suggested. Electron microscopic evidence for a specific association of GroEL-like protein and octameric glutamine synthetase, which was co-purified with this protein, was obtained.     

Reversible dissociation of gamma-glutamylcysteine synthetase into two subunits

gamma-Glutamylcysteine synthetase (rat kidney; Mr approximately 104,000) is composed of 2 nonidentical subunits. In the present work, a procedure was developed for the reversible dissociation of the enzyme into its subunits (Mr = 73,000 and 27,700) under nondenaturing conditions. Students in which gel electrophoresis was used, in conjunction with an enzyme activity stain and elution and re-electrophoresis of protein bands, showed that the heavy subunit contains all of the structural requirements for enzymatic activity and also for feedback inhibition of the enzyme activity by glutathione. The light subunit, which may be formed from a precursor protein, has a significantly lower content of Trp, Phe, Tyr, Val, and Ala residues than the heavy subunit, while its content of Lys, His, Met, and Asx residues is higher.     

Analysis of the kinetic and equilibrium binding of Ku protein to DNA.

The loading of Ku onto a DNA end in a double-strand DNA break is thought to be one of the first steps in the non-homologous DNA end joining (NHEJ) pathway, giving it an essential role in the maintenance of genomic integrity. The binding of Ku to DNA is complicated since DNA can accommodate multiple Ku subunits, which can translocate on the DNA strand. Furthermore, Ku may exhibit cooperativity in the loading process. Therefore, simple one- to-one kinetic models are unable to adequately simulate the process. However, through the use of computer simulation and curve-fitting, we are able to provide a comprehensive mechanistic model and rate constants that closely approximate experimental data for DNA molecules that bind one, two, and three Ku molecules under both kinetic and equilibrium conditions. The model obtains a best fit with Ku having a roughly seven-fold preference to bind to DNA ends rather than internal positions and is consistent with Ku having a strong preference of which face of the protein loads onto the DNA end.     

Oligosaccharyltransferase: a complex multisubunit enzyme of the endoplasmic reticulum

The attachment of N-linked oligosaccharide chains to proteins is an important cotranslational process. These chains can, in some cases, serve to stabilize the protein, while in other cases they function as recognition elements. A key enzyme in the N-glycosylation process is oligosaccharyltransferase (OT). In yeast this enzyme, which is found in the endoplasmic reticulum, consists of nine different transmembrane protein subunits. Our general aim is to learn more about the functions of the multiple subunits of yeast OT and their mode of interaction with each other. Using a combination of biochemical and genetic techniques the subunit Ost1p has been shown to recognize Asn-X-Ser/Thr glycosylation sites. The principle tool used in the identification process was a benzophenone-based glycosylation site peptide that was shown to be crosslinked to Ost1p. Our current objective is to identify the domain in the primary structure that is involved in recognition of the glycosylation site sequence. By use of bifunctional crosslinkers, the possible interaction of Ost1p with other subunits of OT will be studied. This work and other studies on the OT subunits are concisely summarized.