Insights into Wnt binding and signalling from the structures of two Frizzled cysteine-rich domains

Members of the Frizzled family of seven-pass transmembrane proteins serve as receptors for Wnt signalling proteins. Wnt proteins have important roles in the differentiation and patterning of diverse tissues during animal development, and inappropriate activation of Wnt signalling pathways is a key feature of many cancers. An extracellular cysteine-rich domain (CRD) at the amino terminus of Frizzled proteins binds Wnt proteins, as do homologous domains in soluble proteins-termed secreted Frizzled-related proteins-that function as antagonists of Wnt signalling. Recently, an LDL-receptor-related protein has been shown to function as a co-receptor for Wnt proteins and to bind to a Frizzled CRD in a Wnt-dependent manner. To investigate the molecular nature of the Wnt signalling complex, we determined the crystal structures of the CRDs from mouse Frizzled 8 and secreted Frizzled-related protein 3. Here we show a previously unknown protein fold, and the design and interpretation of CRD mutations that identify a Wnt-binding site. CRDs exhibit a conserved dimer interface that may be a feature of Wnt signalling. This work provides a framework for studies of homologous CRDs in proteins including muscle-specific kinase and Smoothened, a component of the Hedgehog signalling pathway.

A mammalian expression vector for expression and purification of secreted proteins for structural studies

A mammalian expression vector with features optimized for simple expression and purification of secreted proteins has been developed. This vector was constructed to facilitate X-ray crystallographic studies of cysteine-rich glycoproteins that are difficult to express by other means. Proteins expressed with this vector possess an N-terminal human growth hormone domain and an octahistidine tag separated from the desired polypeptide sequences by a tobacco etch virus protease recognition site. Advantages of this vector are high levels of expression, simple detection and purification of expressed proteins, and reliable cleavage of the fusion protein. Cotransfection of this vector with a dihydrofolate reductase gene allows amplification of expression levels with methotrexate. Over one dozen cysteine-rich secreted proteins have been expressed in sufficient quantity for structural studies using this vector; the structure of at least one of these proteins has been determined.

Structure of the Homer EVH1 domain-peptide complex reveals a new twist in polyproline recognition

Homer EVH1 (Ena/VASP Homology 1) domains interact with proline-rich motifs in the cytoplasmic regions of group 1 metabotropic glutamate receptors (mGluRs), inositol-1,4,5-trisphosphate receptors (IP3Rs), and Shank proteins. We have determined the crystal structure of the Homer EVH1 domain complexed with a peptide from mGluR (TPPSPF). In contrast to other EVH1 domains, the bound mGluR ligand assumes an unusual conformation in which the side chains of the Ser-Pro tandem are oriented away from the Homer surface, and the Phe forms a unique contact. This unusual binding mode rationalizes conserved features of both Homer and Homer ligands that are not shared by other EVH1 domains. Site-directed mutagenesis confirms the importance of specific Homer residues for ligand binding. These results establish a molecular basis for understanding the biological properties of Homer-ligand complexes.

Sonic hedgehog protein signals not as a hydrolytic enzyme but as an apparent ligand for patched

The amino-terminal signaling domain of the Sonic hedgehog secreted protein (Shh-N), which derives from the Shh precursor through an autoprocessing reaction mediated by the carboxyl-terminal domain, executes multiple functions in embryonic tissue patterning, including induction of ventral and suppression of dorsal cell types in the developing neural tube. An apparent catalytic site within Shh-N is suggested by structural homology to a bacterial carboxypeptidase. We demonstrate here that alteration of residues presumed to be critical for a hydrolytic activity does not cause a loss of inductive activity, thus ruling out catalysis by Shh-N as a requirement for signaling. We favor the alternative, that Shh-N functions primarily as a ligand for the putative receptor Patched (Ptc). This possibility is supported by new evidence for direct binding of Shh-N to Ptc and by a strong correlation between the affinity of Ptc-binding and the signaling potency of Shh-N protein variants carrying alterations of conserved residues in a particular region of the protein surface. These results together suggest that direct Shh-N binding to Ptc is a critical event in transduction of the Shh-N signal.

Implications of atomic-resolution structures for cell adhesion

Molecules involved in cell adhesion processes are often both structurally and functionally modular, with subdomains that are members of large protein families. Recently, high-resolution structures have been determined for representative members of many of these families including fragments of integrins, cadherins, fibronectin-like domains, and immunoglobulin-like domains. These structures have enhanced our understanding of cell adhesion processes at several levels. In almost all cases, ligand-binding sites have been visualized and provide insight into how these molecules mediate biologically important interactions. Metal-binding sites have been identified and characterized, allowing assessment of the role of bound ions in cell adhesion processes. Many of these structures serve as templates for modeling homologous domains in other proteins or, when the structure of a fragment consisting of more than one domain is determined, the structure of multidomain arrays of homologous domains. Knowledge of atomic structure also allows rational design of drugs that either mimic or target specific binding sites. In many cases, high-resolution structures have revealed unexpected relationships that pose questions about the evolutionary origin of specific domains. This review briefly describes several recently determined structures of cell adhesion molecules, summarizes some of the main results of each structure, and highlights common features of different systems.

Crystal structure of a Hedgehog autoprocessing domain: homology between Hedgehog and self-splicing proteins

The approximately 25 kDa carboxy-terminal domain of Drosophila Hedgehog protein (Hh-C) possesses an autoprocessing activity that results in an intramolecular cleavage of full-length Hedgehog protein and covalent attachment of a cholesterol moiety to the newly generated amino-terminal fragment. We have identified a 17 kDa fragment of Hh-C (Hh-C17) active in the initiation of autoprocessing and report here its crystal structure. The Hh-C17 structure comprises two homologous subdomains that appear to have arisen from tandem duplication of a primordial gene. Residues in the Hh-C17 active site have been identified, and their role in Hedgehog autoprocessing probed by site-directed mutagenesis. Aspects of sequence, structure, and reaction mechanism are conserved between Hh-C17 and the self-splicing regions of inteins, permitting reconstruction of a plausible evolutionary history of Hh-C and the inteins.

The role of the divalent cation in the structure of the I domain from the CD11a/CD18 integrin

BACKGROUND

The integrin family of cell-surface receptors mediates a wide variety of cell-cell and cell-extracellular matrix interactions. Integrin-ligand interactions are invariably dependent on the presence of divalent cations, and a subset of integrins contain a approximately 200 amino acid inserted (I) domain that is important for ligand binding activity and contains a single divalent cation binding site. Many integrins are believed to respond to stimuli by undergoing a conformational change that increases their affinity for ligand, and there is a clear difference between two crystal structures of the CD11b I domain with different divalent cations (magnesium and manganese) bound. In addition to the different bound cation, a 'ligand mimetic' crystal lattice interaction in the CD11b I domain structure with bound magnesium has led to the interpretation that the different CD11b I domain structures represent different affinity states of I domains. The influence of the bound cation on I domain structure and function remains incompletely understood, however. The crystal structure of the CD11a I domain bound to manganese is known. We therefore set out to determine whether this structure changes when the metal ion is altered or removed.

RESULTS

We report here the crystal structures of the CD11a I domain determined in the absence of bound metal ion and with bound magnesium ion. No major structural rearrangements are observed in the metal-binding site of the CD11a I domain in the absence or presence of bound manganese ion. The structures of the CD11a I domain with magnesium or manganese bound are extremely similar.

CONCLUSIONS

The conformation of the CD11a I domain is not altered by changes in metal ion binding. The cation-dependence of ligand binding thus indicates that the metal ion is either involved in direct interaction with ligand or required to promote a favorable quaternary arrangement of the integrin.

2.0 A crystal structure of a four-domain segment of human fibronectin encompassing the RGD loop and synergy region

We have determined the 2.0 A crystal structure of a fragment of human fibronectin encompassing the seventh through the RGD-containing tenth type III repeats (FN7-10). The structure reveals an extended rod-like molecule with a long axis of approximately 140 A and highly variable relationships between adjacent domains. An unusually small rotation between domains 9 and 10 creates a distinctive binding site, in which the RGD loop from domain 10 and the "synergy" region from domain 9 are on the same face of FN7-10 and thus easily accessible to a single integrin molecule. The cell-binding RGD loop is well-ordered in this structure and extends approximately 10 A away from the FN7-10 core.

A potential catalytic site revealed by the 1.7-A crystal structure of the amino-terminal signalling domain of Sonic hedgehog

Within the past few years, members of the hedgehog (hh) family of secreted signalling proteins have emerged as the primary signals generated by certain embryonic patterning centres. In vertebrate embryos, for example, sonic hedgehog expression in the notochord appears to be responsible for the local and long-range induction of ventral cell types within the neural tube and somites (reviewed in refs 1, 2). Protein products encoded by hh family members are synthesized as precursors that undergo autoprocessing to generate an amino-terminal domain that appears to be responsible for both local and long-range signalling activities, and a carboxy-terminal domain that contains the autoprocessing activity. As part of an effort to understand how hh family members participate in cell-to-cell signalling, we have determined and report here the crystal structure at 1.7 A of the amino-terminal domain of murine Sonic hedgehog (Shh-N). The structure revealed a tetrahedrally coordinated zinc ion that appears to be structurally analogous to the zinc coordination sites of zinc hydrolases, such as thermolysin and carboxypeptidase A. This previously unsuspected catalytic site represents a distinct activity from the autoprocessing activity that resides in the carboxy-terminal domain.

Interaction between CD8 and major histocompatibility complex (MHC) class I mediated by multiple contact surfaces that include the alpha 2 and alpha 3 domains of MHC class I

The cell surface glycoprotein CD8 functions as a coreceptor with the TCR on cytotoxic T lymphocytes. Mutational analysis of the binding site of CD8 for MHC class I predicted that distinct surfaces of CD8 would interact with both the alpha 2 and alpha 3 domains of class I. Using a cell-cell adhesion assay, we identified three residues Q115, D122, and E128 in the alpha 2 domain of class I critical for interaction with CD8. The side chains of these residues point towards a cavity formed by the alpha 1/alpha 2 platform, the alpha 3 domain and beta 2-microglobulin (beta 2m) of class I. These residues were predicted to contact CD8 based on a bivalent model of interaction between one CD8 alpha/alpha homodimer and two MHC class I molecules. These results therefore provide support for the model.

Crystal structure of the I-domain from the CD11a/CD18 (LFA-1, alpha L beta 2) integrin

We report the 1.8-A crystal structure of the CD11a I-domain with bound manganese ion. The CD11a I-domain contains binding sites for intercellular adhesion molecules 1 and 3 and can exist in both low- and high-affinity states. The metal-bound form reported here is likely to represent a high-affinity state. The CD11a I-domain structure reveals a strained hydrophobic ridge adjacent to the bound metal ion that may serve as a ligand-binding surface and is likely to rearrange in the absence of bound metal ion. The CD11a I-domain is homologous to domains found in von Willebrand factor, and mapping of mutations found in types 2a and 2b von Willebrand disease onto this structure allows consideration of the molecular basis of these forms of the disease.

A structural view of CD4 and CD8

CD4 and CD8 are cell-surface glycoproteins that participate in molecular complexes involved in both T cell development and antigen recognition by T cells. CD4 and CD8 interact with nonpolymorphic regions of class II and class I major histocompatibility complex (MHC) molecules, respectively, and these interactions result in increased intercellular adhesion and enhanced stimulation of T cells. A src-like tyrosine kinase, p56lck, is associated with the cytoplasmic domain of both CD4 and CD8 and may be involved in transmembrane signaling. Crystal structures of extracellular regions of CD4 and CD8 have been determined and have provided a basis for understanding and probing CD4 and CD8 function. The structures of CD4 and CD8 are reviewed here, along with the implications of these structures for CD4 and CD8 function.

Crystallization of a fragment of human fibronectin: introduction of methionine by site-directed mutagenesis to allow phasing via selenomethionine

Crystals of a fragment of human fibronectin encompassing the 7th through the RGD-containing 10th type III repeats (FN7-10) have been produced with protein expressed in E. coli. The crystals are monoclinic with one molecule in the asymmetric unit and diffract to beyond 2.0 A Bragg spacings. A mutant FN7-10 was produced in which three methionines, in addition to the single native methionine already present, have been introduced by site-directed mutagenesis. Diffraction-quality crystals of this mutant protein have been grown in which methionine was replaced with selenomethionine. The introduction of methionine by site-directed mutagenesis to allow phasing from selenomethionyl-substituted crystals is shown to be feasible by this example and is proposed as a general approach to solving the crystallographic phase problem. Strategies for selecting propitious sites for methionine mutations are discussed.

The role of charge and multiple faces of the CD8 alpha/alpha homodimer in binding to major histocompatibility complex class I molecules: support for a bivalent model

The CD8 dimer interacts with the alpha 3 domain of major histocompatibility complex class I molecules through two immunoglobulin variable-like domains. In this study a crystal structure-informed mutational analysis has been performed to identify amino acids in the CD8 alpha/alpha homodimer that are likely to be involved in binding to class I. Several key residues are situated on the top face of the dimer within loops analogous to the complementarity-determining regions (CDRs) of immunoglobulin. In addition, other important amino acids are located in the A and B beta-strands on the sides of the dimer. The potential involvement of amino acids on both the top and the side faces of the molecule is consistent with a bivalent model for the interaction between a single CD8 alpha/alpha homodimer and two class I molecules and may have important implications for signal transduction in class I-expressing cells. This study also demonstrates a role for the positive surface potential of CD8 in class I binding and complements previous work demonstrating the importance of a negatively charged loop on the alpha 3 domain of class I for CD8 alpha/alpha-class I interaction. We propose a model whereby residues located on the CDR-like loops of the CD8 homodimer interact with the alpha 3 domain of MHC class I while amino acids on the side of the molecule containing the A and B beta-strands contact the alpha 2 domain of class I.

Structure of a fibronectin type III domain from tenascin phased by MAD analysis of the selenomethionyl protein

Fibronectin type III domains are found in many different proteins including cell surface receptors and cell adhesion molecules. The crystal structure of one such domain from the extracellular matrix protein tenascin was determined. The structure was solved by multiwavelength anomalous diffraction (MAD) phasing of the selenomethionyl protein and has been refined to 1.8 angstrom resolution. The folding topology of this domain is identical to that of the extracellular domains of the human growth hormone receptor, the second domain of CD4, and PapD. Although distinct, this topology is similar to that of immunoglobulin constant domains. An Arg-Gly-Asp (RGD) sequence that can function for cell adhesion is found in a tight turn on an exposed loop.

Crystal structure of a soluble form of the human T cell coreceptor CD8 at 2.6 A resolution

A secreted fragment of the extracellular portion of human CD8 alpha has been expressed in CHO cells, and a deglycosylated and proteolyzed form of this fragment has been crystallized. We report here the crystal structure of this fragment as refined at 2.6 A resolution. The structure was solved by molecular replacement using a superposition of ten variable domains from immunoglobulin light chains as the search model. Only the N-terminal 114 amino acids of CD8 alpha are visible in the electron density maps. The domain formed by these residues possesses a fold typical of immunoglobulin variable domains and associates to form Fv-like homodimers.

2.9 A resolution structure of an anti-dinitrophenyl-spin-label monoclonal antibody Fab fragment with bound hapten

The crystal structure of the Fab fragment of the murine monoclonal anti-dinitrophenyl-spin-label antibody AN02 complexed with its hapten has been solved at 2.9 A resolution using a novel molecular replacement method. Prior to translation searches, a large number of the most likely rotation function solutions were subjected to a rigid body refinement against the linear correlation coefficient between intensities of observed and calculated structure factors. First, the overall orientation of the search model and then the orientations and positions of the four Fab domains (VH, VL, CH1 and CL) were refined. This procedure clearly identified the correct orientation of the search model. The refined search model was then subjected to translation searches which unambiguously determined the enantiomer and position in the unit cell of the crystal. The successful search model was refined 2.5 A crystal structure of the Fab fragment of HyHel-5 from which non-matching residues in the variable domains had been removed. HyHel-5 is a murine monoclonal antibody whose heavy and light chains are of the same subclass (gamma 1, kappa, respectively) as AN02. After molecular replacement the structure of the AN02 Fab has been refined using simulated annealing in combination with model building and conjugate gradient refinement to a current crystallographic R-factor of 19.5% for 12,129 unique reflections between 8.0 and 2.9 A. The root-mean-square (r.m.s.) deviation from ideal bond lengths is 0.014 A, and the r.m.s. deviation from ideal bond angles is 3.1 degrees. The electron density reveals the hapten sitting in a pocket formed by the loops of the complementarity determining region. The dinitrophenyl ring of the hapten is sandwiched between the indole rings of Trp96 of the heavy-chain and Trp91 of the light-chain. The positioning of the hapten and general features of the combining site are in good agreement with the results of earlier nuclear magnetic resonance experiments.

Structural and kinetic studies of the Fab fragment of a monoclonal anti-spin label antibody by nuclear magnetic resonance

Nuclear magnetic resonance has been used to study the structure of the anti-spin label antibody AN02 combining site and kinetic rates for the hapten-antibody reaction. The association reaction for the hapten dinitrophenyl-diglycine (DNP-diGly) is diffusion-limited. The activation enthalpy for association, 5.1 kcal/mol, is close to the activation enthalpy for diffusion in water. Several reliable resonance assignments have been made with the aid of recently reported crystal structure. Structural data deduced from the nuclear magnetic resonance (n.m.r.) spectra compare favorably with the crystal structure in terms of the combining site amino acid composition, distances of tyrosine residues from the unpaired electron of the hapten, and residues in direct contact with the hapten. Evidence is presented that a single binding site region tyrosine residue can assume two distinct conformations on binding of DNP-diGly. The AN02 antibody is an autoantibody. Dimerization of the Fab fragments is blocked by the hapten DNP-diGly. The n.m.r. spectra suggests that some of the amino acid residues involved in the binding of the DNP-hapten are also involved in the Fab dimerization.

Crystallization of an anti-2,2,6,6-tetramethyl-1-piperidinyloxy-dinitrophenyl monoclonal antibody Fab fragment with and without bound hapten

The Fab fragment of a monoclonal antibody, AN02, specific for a 2,2,6,6-tetramethyl-1-piperidinyloxy-dinitrophenyl hapten was crystallized both with and without bound hapten. Both crystals formed in phosphate-buffered saline (150 mM-NaCl, 10 mM-Na2PO4, 0.02% (w/v) NaN3 (pH 7.4) at 4 degrees C and diffracted beyond 2.2 A resolution (1A = 0.1 nm). Fab with bound hapten crystallizes in space group P6(1)22 or P6(5)22, with cell dimensions a = b = 73.23 A, c = 373.8 A, alpha = beta = 90 degrees and gamma = 120 degrees. Unoccupied Fab also crystallizes in space group P6(1)22 or P6(5)22 with cell dimensions a = b = 73 A, c = 377 A, alpha = beta = 90 degrees and gamma = 120 degrees.

Sequences of 12 monoclonal anti-dinitrophenyl spin-label antibodies for NMR studies

Eleven monoclonal antibodies specific for a spin-labeled dinitrophenyl hapten (DNP-SL) have been produced for use in NMR studies. They have been named AN01 and AN03-AN12. The stability constants for the association of these antibodies with DNP-SL and related haptens were measured by fluorescence quenching and ranged from 5.0 X 10(4) M-1 to greater than 1.0 X 10(8) M-1. cDNA clones coding for the heavy and light chains of each antibody and of an additional anti-DNP-SL monoclonal antibody, AN02, have been isolated. The nucleic acid sequence of the 5' end of each clone has been determined, and the amino acid sequence of the variable regions of each antibody has been deduced from the cDNA sequence. The sequences are relatively heterogeneous, but both the heavy and the light chains of AN01 and AN03 are derived from the same variable-region gene families as those of the AN02 antibody. AN07 has a heavy chain that is related to that of AN02, and AN09 has a related light chain. AN05 and AN06 are unrelated to AN02 but share virtually identical heavy and light chains. Preliminary NMR difference spectra comparing related antibodies show that sequence-specific assignment of resonances is possible. Such spectra also provide a measure of structural relatedness.

Homovanillic acid concentrations in brain, CSF and plasma as indicators of central dopamine function in primates

In a large number (91) of vervet monkeys, correlation coefficients were determined between homovanillic acid (HVA) concentrations in four brain areas. Significant correlations existed between dorsal frontal cortex and orbital frontal cortex and between putamen and caudate nucleus. However, no significant correlations existed between either cortical area and the basal ganglia areas. Correlations were tested between CSF and plasma HVA and between these fluids and brain regions. The only significant relationship found was between CSF and dorsal frontal cortex, after possible treatment effects were statistically removed. The assumption that primate CSF HVA concentration necessarily reflects basal ganglia HVA concentration is questioned and furthermore, the results suggest that HVA from cortex contributes significantly to that in cisternal CSF. Raw plasma HVA measurements (even when uninfluenced by diet or anesthetic) appear to be of limited value in gauging central dopamine metabolism and turnover.