Analysis of the streptococcal hyaluronic acid synthase complex using the photoaffinity probe 5-azido-UDP-glucuronic acid.

The mucopolysaccharide, hyaluronic acid, is an important component of both mammals and pathogenic streptococci. This high molecular weight polymer is synthesized by a membrane-associated, multisubunit hyaluronate synthase which utilizes UDP-glucuronic acid and UDP-N-acetylglucosamine as substrates. Using the photoaffinity probe, [beta-32P]5-azido-UDP-glucuronic acid, three streptococcal membrane proteins (42, 33, and 27 kDa) specifically photoincorporated this probe. Labeling of these proteins was enhanced in the presence of UDP-N-acetylglucosamine, whereas UDP-galactose or UDP-glucose had no effect on incorporation. UDP-glucuronic acid inhibited the labeling of the three proteins in a dose-dependent manner. Detergent-solubilized membrane proteins from transposon-inactivated hyaluronic acid capsule mutants no longer incorporated the probe. This was also the case when membranes from stationary phase organisms were tested. Finally, glucuronic acid no longer was incorporated into high molecular weight hyaluronic acid with either the mutant or stationary phase preparations. Further biochemical analysis will be required to demonstrate the exact role each of the proteins play in hyaluronic acid biosynthesis.     

Glycoprotein hormone assembly in the endoplasmic reticulum: II. Multiple roles of a redox sensitive beta-subunit disulfide switch

All three human glycoprotein hormone heterodimers are assembled in the endoplasmic reticulum by threading the glycosylated end of alpha-subunit loop two (alpha2) beneath a disulfide "latched" strand of the beta-subunit known as the "seatbelt." This remarkable event occurs efficiently even though the seatbelt effectively blocks the reverse process, thereby stabilizing each heterodimer. Studies described here show that assembly is facilitated by the formation, disruption, and reformation of a loop within the seatbelt that is stabilized by the most easily reduced disulfide in the free beta-subunit. We refer to this disulfide as the "tensor" because it shortens the seatbelt, thereby securing the heterodimer. Formation of the tensor disulfide appears to precede and facilitate seatbelt latching in most human choriogonadotropin beta-subunit molecules. Subsequent disruption of the tensor disulfide elongates the seatbelt, thereby increasing the space beneath the seatbelt and the beta-subunit core. This permits the formation of hydrogen bonds between backbone atoms of the beta-subunit cystine knot and the tensor loop with backbone atoms in loop alpha2, a process that causes the glycosylated end of loop alpha2 to be threaded between the seatbelt and the beta-subunit core. Contacts between the tensor loop and loop alpha2 promote reformation of the tensor disulfide, which explains why it is more stable in the heterodimer than in the uncombined beta-subunit. These findings unravel the puzzling nature of how a threading mechanism can be used in the endoplasmic reticulum to assemble glycoprotein hormones that have essential roles in vertebrate reproduction and thyroid function.     

Occurrence of a bacterial membrane microdomain at the cell division site enriched in phospholipids with polyunsaturated hydrocarbon chains

In this study, we found that phospholipids containing an eicosapentaenyl group form a novel membrane microdomain at the cell division site of a Gram-negative bacterium, Shewanella livingstonensis Ac10, using chemically synthesized fluorescent probes. The occurrence of membrane microdomains in eukaryotes and prokaryotes has been demonstrated with various imaging tools for phospholipids with different polar headgroups. However, few studies have focused on the hydrocarbon chain-dependent localization of membrane-resident phospholipids in vivo. We previously found that lack of eicosapentaenoic acid (EPA), a polyunsaturated fatty acid found at the sn-2 position of glycerophospholipids, causes a defect in cell division after DNA replication of S. livingstonensis Ac10. Here, we synthesized phospholipid probes labeled with a fluorescent 7-nitro-2,1,3-benzoxadiazol-4-yl (NBD) group to study the localization of EPA-containing phospholipids by fluorescence microscopy. A fluorescent probe in which EPA was bound to the glycerol backbone via an ester bond was found to be unsuitable for imaging because EPA was released from the probe by in vivo hydrolysis. To overcome this problem, we synthesized hydrolysis-resistant ether-type phospholipid probes. Using these probes, we found that the fluorescence localized between two nucleoids at the cell center during cell division when the cells were grown in the presence of the eicosapentaenyl group-containing probe (N-NBD-1-oleoyl-2-eicosapentaenyl-sn-glycero-3-phosphoethanolamine), whereas this localization was not observed with the oleyl group-containing control probe (N-NBD-1-oleoyl-2-oleyl-sn-glycero-3-phosphoethanolamine). Thus, phospholipids containing an eicosapentaenyl group are specifically enriched at the cell division site. Formation of a membrane microdomain enriched in EPA-containing phospholipids at the nucleoid occlusion site probably facilitates cell division.     

A direct fluorescence-based assay for RGS domain GTPase accelerating activity

Diverse extracellular signals regulate seven transmembrane-spanning receptors to modulate cellular physiology. These receptors signal primarily through activation of heterotrimeric guanine nucleotide binding proteins (G proteins). A major determinant of heterotrimeric G protein signaling in vivo and in vitro is the intrinsic GTPase activity of the Galpha subunit. RGS (regulator of G protein signaling) domain-containing proteins are GTPase accelerating proteins specific for Galpha subunits. In this article, we describe the use of the ribose-conjugated fluorescent guanine nucleotide analog BODIPYFL-GTP as a spectroscopic probe to measure intrinsic and RGS protein-catalyzed nucleotide hydrolysis by Galphao. BODIPYFL-GTP bound to Galphao exhibits a 200% increase in fluorescence quantum yield. Hydrolysis of BODIPYFL-GTP to BODIPYFL-GDP reduces the quantum yield to 27% above its unbound value. We demonstrate that BODIPYFL-GTP can be used as a rapid real-time probe for measuring RGS domain-catalyzed GTP hydrolysis by Galphao. We demonstrate the effectiveness of this assay in the analysis of loss-of-function point mutants of both Galphao and RGS12. This assay should be useful in screening for and analyzing RGS protein inhibitory compounds.     

ZPRED: predicting the distance to the membrane center for residues in alpha-helical membrane proteins

MOTIVATION: Prediction methods are of great importance for membrane proteins as experimental information is harder to obtain than for globular proteins. As more membrane protein structures are solved it is clear that topology information only provides a simplified picture of a membrane protein. Here, we describe a novel challenge for the prediction of alpha-helical membrane proteins: to predict the distance between a residue and the center of the membrane, a measure we define as the Z-coordinate. Even though the traditional way of depicting membrane protein topology is useful, it is advantageous to have a measure that is based on a more "physical" property such as the Z-coordinate, since it implicitly contains information about re-entrant helices, interfacial helices, the tilt of a transmembrane helix and loop lengths. RESULTS: We show that the Z-coordinate can be predicted using either artificial neural networks, hidden Markov models or combinations of both. The best method, ZPRED, uses the output from a hidden Markov model together with a neural network. The average error of ZPRED is 2.55A and 68.6% of the residues are predicted within 3A of the target Z-coordinate in the 5-25A region. ZPRED is also able to predict the maximum protrusion of a loop to within 3A for 78% of the loops in the dataset. AVAILABILITY: Supplementary information and training data is available at http://www.sbc.su.se/~erikgr/.     

Structure-based rational design of streptavidin mutants with pseudo-catalytic activity

Introduction of enzymatic activity into proteins or other types of polymers by rational design is a major objective in the life sciences. To date, relatively low levels of enzymatic activity could be introduced into antibodies by using transition-state analogues of haptens. In the present study, we identify the structural elements that contribute to the observed hydrolytic activity in egg white avidin, which promote the cleavage of active biotin esters (notably biotinyl p-nitrophenyl ester). The latter elements were then incorporated into bacterial streptavidin via genetic engineering. The streptavidin molecule was thus converted from a protector to an enhancer of hydrolysis of biotin esters. The conversion was accomplished by the combined replacement of a "lid-like loop" (L3,4) and a leucine-to-arginine point mutation in streptavidin. Interestingly, neither of these elements play a direct role in the hydrolytic reaction. The latter features were thus shown to be responsible for enhanced substrate hydrolysis. This work indicates that structural and non-catalytic elements of a protein can be modified to promote the induced fit of a substrate for subsequent interaction with either a catalytic residue or water molecules. This approach complements the conventional design of active sites that involves direct modifications of catalytic residues.     

Interaction of hexane phosphonic acid diethyl ester with phospholipids in hepatic microsomes and reconstituted liposomes as studied by 31P-NMR

By use of 31P-NMR, quasi-elastic light scattering and freeze-fracture electron microscopy it is shown that hexane phosphonic acid diethyl ester (PAE) is incorporated in hepatic microsomes without any alteration of the bilayer structure at two different sites. These findings proved that PAE can be used as molecular 31P-NMR probe in microsomes to get information about lipid-protein interactions. Extensive studies on reconstituted liposomal systems which contained cytochrome P-450 and cytochrome P-450 reductase showed that both proteins influence the localization of incorporated PAE. The results indicate a specific interaction of phosphatidylethanolamine (PE) with cytochrome P-450 in microsomes.     

Features of a Spatially Constrained Cystine Loop in the p10 FAST Protein Ectodomain Define a New Class of Viral Fusion Peptides

The reovirus fusion-associated small transmembrane (FAST) proteins are the smallest known viral membrane fusion proteins. With ectodomains of only ∼20–40 residues, it is unclear how such diminutive fusion proteins can mediate cell-cell fusion and syncytium formation. Contained within the 40-residue ectodomain of the p10 FAST protein resides an 11-residue sequence of moderately apolar residues, termed the hydrophobic patch (HP). Previous studies indicate the p10 HP shares operational features with the fusion peptide motifs found within the enveloped virus membrane fusion proteins. Using biotinylation assays, we now report that two highly conserved cysteine residues flanking the p10 HP form an essential intramolecular disulfide bond to create a cystine loop. Mutagenic analyses revealed that both formation of the cystine loop and p10 membrane fusion activity are highly sensitive to changes in the size and spatial arrangement of amino acids within the loop. The p10 cystine loop may therefore function as a cystine noose, where fusion peptide activity is dependent on structural constraints within the noose that force solvent exposure of key hydrophobic residues. Moreover, inhibitors of cell surface thioreductase activity indicate that disruption of the disulfide bridge is important for p10-mediated membrane fusion. This is the first example of a viral fusion peptide composed of a small, spatially constrained cystine loop whose function is dependent on altered loop formation, and it suggests the p10 cystine loop represents a new class of viral fusion peptides.     

Solution structure of the cyclotide palicourein: implications for the development of a pharmaceutical framework

The cyclotides are a family of disulfide-rich proteins from plants. They have the characteristic structural features of a circular protein backbone and a knotted arrangement of disulfide bonds. Structural and biochemical studies of the cyclotides suggest that their unique physiological stability can be loaned to bioactive peptide fragments for pharmaceutical and agricultural development. In particular, the cyclotides incorporate a number of solvent-exposed loops that are potentially suitable for epitope grafting applications. Here, we determine the structure of the largest known cyclotide, palicourein, which has an atypical size and composition within one of the surface-exposed loops. The structural data show that an increase in size of a palicourein loop does not perturb the core fold, to which the thermodynamic and chemical stability has been attributed. The cyclotide core fold, thus, can in principle be used as a framework for the development of useful pharmaceutical and agricultural bioactivities.     

Use of membrane spin label spectra to monitor rates of reaction of partitioning compounds: hydrolysis of a local anesthetic analog

ESR spectra of membrane spin probes are conventionally used to obtain structural information. Here we show, for the first time, that when a membrane-soluble compound undergoes a chemical reaction, time-dependent changes in the ESR spectra of membrane spin probes can yield information about the kinetics of reaction. A benzoic acid ester, analog of the local anesthetic tetracaine, partitions between aqueous and membrane phases, causing changes in membrane structure as monitored by the ESR spectra of a probe. At alkaline pH, the lineshapes are time-dependent and the spectra go back to that in the absence of drug. The changes follow pseudo-first order kinetics. This effect is due to drug hydrolysis leading to water-soluble products, as confirmed by direct spectrophotometric measurements of the reaction. The pseudo-first order rate constants found by the latter method are in very good agreement with those calculated by ESR. The rate of hydrolysis decreases with increasing membrane concentration. This phenomenon accounts in part for the increased potency and toxicity of the more membrane-soluble local anesthetics.     

(1)H/(15)N heteronuclear NMR spectroscopy shows four dynamic domains for phospholamban reconstituted in dodecylphosphocholine micelles

We report the backbone dynamics of monomeric phospholamban in dodecylphosphocholine micelles using (1)H/(15)N heteronuclear NMR spectroscopy. Phospholamban is a 52-amino acid membrane protein that regulates Ca-ATPase in cardiac muscle. Phospholamban comprises three structural domains: a transmembrane domain from residues 22 to 52, a connecting loop from 17 to 21, and a cytoplasmic domain from 1 to 16 that is organized in an "L"-shaped structure where the transmembrane and the cytoplasmic domain form an angle of approximately 80 degrees (Zamoon et al., 2003; Mascioni et al., 2002). T(1), T(2), and (1)H/(15)N nuclear Overhauser effect values measured for the amide backbone resonances were interpreted using the model-free approach of Lipari and Szabo. The results point to the existence of four dynamic domains, revealing the overall plasticity of the cytoplasmic helix, the flexible loop, and part of the transmembrane domain (residues 22-30). In addition, using Carr-Purcell-Meiboom-Gill-based experiments, we have characterized phospholamban dynamics in the micros-ms timescale. We found that the majority of the residues in the cytoplasmic domain, the flexible loop, and the first ten residues of the transmembrane domain undergo dynamics in the micros-ms range, whereas minimal dynamics were detected for the transmembrane domain. Hydrogen/deuterium exchange factors measured at different temperatures support the existence of slow motion in both the loop and the cytoplasmic helix. We propose that these dynamic properties are critical factors in the biomolecular recognition of phospholamban by Ca-ATPase and other interacting proteins such as protein kinase A and protein phosphatase 1.     

Only a portion of the small seatbelt loop in human choriogonadotropin appears capable of contacting the lutropin receptor

Twenty residues of the human choriogonadotropin (hCG) beta-subunit that are wrapped around alpha-subunit loop 2 like a "seatbelt" stabilize the heterodimer and enable the hormone to distinguish lutropin (LHR), follitropin, and thyrotropin receptors. The N-terminal portion of the seatbelt contains a small disulfide-stabilized loop needed for heterodimer assembly and is thought to mediate hCG-LHR interactions. To test the latter notion, we compared the LHR binding and signal transduction activities of hCG analogs in which the alpha-subunit C terminus (alphaCT) was cross-linked to residues in the small seatbelt loop. Analogs having an intersubunit disulfide between a cysteine in place of alphaCT residue alphaSer-92 and cysteines substituted for loop residues betaArg-94, betaArg-95, or betaSer-96 had high activities in LHR binding and signaling assays despite the fact that both portions of the hormone are thought to be essential for hCG activity. Use of a larger probe blocked hormone activity when the alphaCT was cross-linked to cysteines in place of residues betaArg-95 and betaAsp-99, but not to cysteines in place of residues betaArg-94, betaSer-96, or betaThr-97. This suggested that the side chains of residues betaArg-95 and betaAsp-99, which face in the same outward direction from the heterodimer, are nearer than the others to the LHR interface. The finding that residue 95 can be cross-linked to small alphaCT probes without eliminating hormone activity indicates its side chain does not participate in essential LHR contacts. We suggest that contacts between the small seatbelt loop and the LHR, if any, involve its backbone atoms and possibly the side chain of residue betaAsp-99.     

A biomimetic strategy in the synthesis and fragmentation of cyclic protein.

This paper describes a simple biomimetic strategy to prepare small cyclic proteins containing multiple disulfide bonds. Our strategy involves intramolecular acyl transfer reactions to assist both the synthesis and fragmentation of these highly constrained cyclic structures in aqueous solution. To illustrate our strategy, we synthesized the naturally occurring circulin B and cyclopsychotride (CPT), both consisting of 31 amino acid residues tightly packed in a cystine-knot motif with three disulfide bonds and an end-to-end cyclic form. The synthesis of these small cyclic proteins can be achieved by orthogonal ligation of free peptide thioester via the thia zip reaction, which involves a series of reversible thiol-thiolactone exchanges to arrive at an alpha-amino thiolactone, which then undergoes an irreversible, spontaneous ring contraction through an S,N-acyl migration to form the cyclic protein. A two-step disulfide formation strategy is employed for obtaining the desired disulfide-paired products. Partial acid hydrolysis through intramolecular acyl transfer of X-Ser, X-Thr, Asp-X, and Glu-X sequences is used to obtain the assignment of the circulins disulfide bond connectives. Both synthetic circulin B and CPT are identical to the natural products and, thus, the total synthesis confirms the disulfide connectivity of circulin B and CPT contain a cystine-knot motif of 1-4, 2-5, and 3-6. In general, our strategy, based on the convergence of chemical proteolysis and aminolysis of peptide bonds through acyl transfer, is biomimetic and provides a useful approach for the synthesis and characterization of large end-to-end cyclic peptides and small proteins.     

Synthesis of stable C‐linked ferrocenyl amino acids and their use in solution‐phase peptide synthesis

Incorporation of ferrocenyl group to peptides is an efficient method to alter their hydrophobicity. Ferrocenyl group can also act as an electrochemical probe when incorporated onto functional peptides. Most often, ferrocene is incorporated onto peptides post‐synthesis via amide, ester or triazole linkages. Stable amino acids containing ferrocene as a C‐linked side chain are potentially useful building units for the synthesis of ferrocene‐containing peptides. We report here an efficient route to synthesize ferrocene‐containing amino acids that are stable and can be used in peptide synthesis. Coupling of 2‐ferrocenyl‐1,3‐dithiane and iodides derived from aspartic acid or glutamic acid using n‐butyllithium leads to the incorporation of a ferrocenyl unit to the δ‐position or ε‐position of an α‐amino acid. The reduction or hydrolysis of the dithiane group yields an alkyl or an oxo derivative. The usability of the synthesized amino acids is demonstrated by incorporating one of the amino acids in both C‐terminus and N‐terminus of tripeptides in solution phase. Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.     

The spin label amino acid TOAC and its uses in studies of peptides: chemical, physicochemical, spectroscopic, and conformational aspects

We review work on the paramagnetic amino acid 2,2,6,6-tetramethyl-N-oxyl-4-amino-4-carboxylic acid, TOAC, and its applications in studies of peptides and peptide synthesis. TOAC was the first spin label probe incorporated in peptides by means of a peptide bond. In view of the rigid character of this cyclic molecule and its attachment to the peptide backbone via a peptide bond, TOAC incorporation has been very useful to analyze backbone dynamics and peptide secondary structure. Many of these studies were performed making use of EPR spectroscopy, but other physical techniques, such as X-ray crystallography, CD, fluorescence, NMR, and FT-IR, have been employed. The use of double-labeled synthetic peptides has allowed the investigation of their secondary structure. A large number of studies have focused on the interaction of peptides, both synthetic and biologically active, with membranes. In the latter case, work has been reported on ligands and fragments of GPCR, host defense peptides, phospholamban, and β-amyloid. EPR studies of macroscopically aligned samples have provided information on the orientation of peptides in membranes. More recent studies have focused on peptide-protein and peptide-nucleic acid interactions. Moreover, TOAC has been shown to be a valuable probe for paramagnetic relaxation enhancement NMR studies of the interaction of labeled peptides with proteins. The growth of the number of TOAC-related publications suggests that this unnatural amino acid will find increasing applications in the future.     

Assignment of disulfide bonds in proteins by partial acid hydrolysis and mass spectrometry

A new method is described for locating disulfide bonds in proteins which cannot be cleaved between half-cystinyl residues by enzymic methods, as is often the case for tightly coiled proteins, or for proteins in which half-cystinyl residues are not separated by residues required for enzymic cleavage. Partial acid hydrolysis of a model protein, hen egg-white lysozyme, produces a mixture of disulfide-containing peptides from which the disulfide connections may be deduced. The usefulness of a combination of HPLC, fast atom bombardment mass spectrometry, and computer-assisted analysis to identify disulfide-containing peptides present in the partial acid hydrolysate of the model protein is demonstrated. Chromatographic fractions of the hydrolysate were analyzed by mass spectrometry before and after chemical reduction of the disulfide bonds to determine the molecular weights of disulfide-containing peptides. Computer-assisted analysis was then used to relate the molecular weights of these peptides to specific segments of the protein from which the disulfide connectivities could be determined. Partial acid hydrolysis of proteins, which is attractive because it proceeds relatively independent of the amino acid sequence and structure, and because disulfide interchange is unlikely to occur in dilute acid, has become practical because disulfide-containing peptides present in complex mixtures can be identified rapidly and definitively by this method.     

Binding of phlorizin to the C-terminal loop 13 of the Na(+)/glucose cotransporter does not depend on the [560-608] disulfide bond

The disulfide bonds of the Na(+)/glucose cotransporter (SGLT1) are believed to participate in the binding of the transport inhibitor phlorizin. Here, we investigated the role of the [560-608] disulfide bond on the phlorizin-binding function of the C-terminal loop 13 of SGLT1 using 3-iodoacetamidophlorizin (3-IAP) as a probe. The reactivity of 3-IAP to the fully reduced loop 13 was competitively inhibited by phlorizin, as evident from the MALDI mass spectra. It indicates that the disulfide bond is not mandatory for phlorizin binding. CD and equilibrium unfolding studies showed that the secondary structure and conformation stability of loop 13 were not affected by removing the disulfide bond. Furthermore, we generated a series of loop 13 mutants to assess the contribution of the disulfide bond to phlorizin binding. A positive correlation between the stability and phlorizin affinity of the mutant proteins was observed, implying that the protein stability, rather than the disulfide bond, is relevant to the phlorizin-binding function of loop 13.     

Improved Fmoc‐solid‐phase peptide synthesis of an extracellular loop of CFTR for antibody selection by the phage display technology

Cystic fibrosis (CF), a life‐shortening genetic disease, is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene that codes for the CFTR protein, the major chloride channel expressed at the apical membrane of epithelial cells. The development of an imaging probe capable of non‐invasively detect CFTR at the cell surface could be of great advantage for the management of CF. With that purpose, we synthesized the first extracellular loop of CFTR protein (ECL1) through fluorenylmethyloxycarbonyl (Fmoc)‐based microwave‐assisted solid‐phase peptide synthesis (SPPS), according to a reported methodology. However, aspartimide formation, a well‐characterized side reaction in Fmoc‐SPPS, prompted us to adopt a different side‐chain protection strategy for aspartic acid residues present in ECL1 sequence. The peptide was subsequently modified via PEGylation and biotinylation, and cyclized through disulfide bridge formation, mimicking the native loop conformation in CFTR protein. Herein, we report improvements in the synthesis of the first extracellular loop of CFTR, including peptide modifications that can be used to improve antigen presentation in phage display for selection of novel antibodies against plasma membrane CFTR.     

An azidoaryl thioglycoside of sialic acid. A potential photoaffinity probe of sialidases and sialic acid-binding proteins

An azidoaryl thioglycoside of sialic acid was prepared, as a potential photoaffinity probe reagent for the analysis of sialidases and sialic acid-binding proteins, by treatment of the glycosyl chloride of N-acetylneuraminic acid methyl ester with potassium thioacetate to give, in 70% yield, methyl 5-acetamido-4,7,8,9-tetra-O-acetyl-2-S-acetyl-2,3,5-trideoxy-2-thio-alph a-D- glycero-D-galacto-2-nonulopyranosonate. Selective hydrolysis of the thioacetate ester, followed by condensation with 4-fluoro-3-nitrophenyl azide, O-deacetylation, and hydrolysis gave (4-azido-2-nitrophenyl)- 5-acetamido-2,3,5-trideoxy-2-thio-alpha-D-glycero-D-galacto-2- nonulopyranosidonic acid.     

Glycoprotein hormone assembly in the endoplasmic reticulum: IV. Probable mechanism of subunit docking and completion of assembly

The unique structures of human choriogonadotropin (hCG) and related glycoprotein hormones make them well suited for studies of protein folding in the endoplasmic reticulum. hCG is stabilized by a strand of its beta-subunit that has been likened to a "seatbelt" because it surrounds alpha-subunit loop 2 and its end is "latched" by an intrasubunit disulfide bond to the beta-subunit core. As shown here, assembly begins when parts of the NH(2) terminus, cysteine knot, and loops 1 and 3 of the alpha-subunit dock reversibly with parts of the NH(2) terminus, cystine knot, and loop 2 of the hCG beta-subunit. Whereas the seatbelt can contribute to the stability of the docked subunit complex, it interferes with docking and/or destabilizes the docked complex when it is unlatched. This explains why most hCG is assembled by threading the glycosylated end of alpha-subunit loop 2 beneath the latched seatbelt rather than by wrapping the unlatched seatbelt around this loop. hCG assembly appears to be limited by the need to disrupt the disulfide that stabilizes the small seatbelt loop prior to threading. We postulate that assembly depends on a "zipper-like" sequential formation of intersubunit and intrasubunit hydrogen bonds between backbone atoms of several residues in the beta-subunit cystine knot, alpha-subunit loop 2, and the small seatbelt loop. The resulting intersubunit beta-sheet enhances the stability of the seatbelt loop disulfide, which shortens the seatbelt and secures the heterodimer. Formation of this disulfide also explains the ability of the seatbelt loop to facilitate latching during assembly by the wraparound pathway.