The novel acidophilic structure of the killer toxin from halotolerant yeast demonstrates remarkable folding similarity with a fungal killer toxin.

BACKGROUND: Several strains of yeasts and fungi produce proteinous substances, termed killer toxins, which kill sensitive strains. The SMK toxin, secreted by the halotolerant yeast Pichia farinosa KK1 strain, uniquely exhibits its maximum killer activity under conditions of acidic pH and high salt concentration. The toxin is composed of two distinct subunits, alpha and beta, which tightly interact with each other under acidic conditions. However, they are easily dissociated under neutral conditions and lose the killer activity. The three-dimensional structure of the SMK toxin will provide a better understanding of the mechanism of toxicity of this protein and the cause of its unique pH-dependent stability. RESULTS: Two crystal structures of the SMK toxin have been determined at 1.8 A resolution in different ionic strength conditions. The two subunits, alpha and beta, are jointly folded into an ellipsoidal, single domain structure belonging to the alpha/beta-sandwich family. The folding topology of the SMK toxin is essentially the same as that of the fungal killer toxin, KP4. This shared topology contains two left-handed split betaalphabeta motifs, which are rare in the other proteins. Many acidic residues are clustered at the bottom of the SMK toxin molecule. Some of the carboxyl sidechains interact with each other through hydrogen bonds. The ionic strength difference induces no evident structural change of the SMK toxin except that, in the high ionic strength crystal, a number of sulfate ions are electrostatically bound near the basic residues which are also locally distributed at the bottom of the toxin molecule. CONCLUSIONS: The two killer toxins, SMK and KP4, share a unique folding topology which contains a rare structural motif. This observation may suggest that these toxins are evolutionally and/or functionally related. The pH-dependent stability of the SMK toxin is a result of the intensive interactions between the carboxyl groups. This finding is important for protein engineering, for instance, towards stabilization of the toxin molecule in a broader pH range. The present crystallographic study revealed that the structure of the SMK toxin itself is hardly affected by the ionic strength, implying that a high salt concentration affects the sensitivity of the cell against the toxin.

Model for the complex between protein G and an antibody Fc fragment in solution.

BACKGROUND: Streptococcal protein G and staphylococcal protein A are bacterial antibody-binding proteins, widely used as immunological tools, whose antibody-binding domains are structurally quite different. The binding of protein G to Fc fragments is competitive with respect to protein A, suggesting that the binding sites for protein A and protein G on Fc overlap, notwithstanding the fact that they lack sequence or structural similarity. RESULTS: To resolve this issue, the residues involved in the interaction between an IgG-binding domain of protein G (domain II) and the Fc fragment of mouse IgG2a have been identified by use of 13C and 15N NMR. Binding of protein G domain II selectively perturbed resonances from residues between the CH2 and CH3 domains of Fc, whereas in domain II the residues affected are primarily those on the alpha-helix and the third strand of the beta-sheet. This information was used, together with the structures of the two uncomplexed proteins, to construct a model of the complex, using Monte Carlo minimization techniques. In this model, the alpha-helix of protein G lies in the same position as helix 1 of protein A in the crystal structure of the protein A:Fc complex, but its orientation differs from the latter by 180 degrees. CONCLUSIONS: The interactions of the bacterial antibody-binding proteins with their 'target' immunoglobulins involve a very versatile set of protein-protein interactions. First, the IgG-binding domains of protein A and protein G have quite different three-dimensional structures, but bind to sites on the Fc fragment that overlap extensively. Secondly, protein G employs two quite different regions of its surface to bind to the Fab and Fc regions of IgG.

Effect of the tertiary structure alteration by ligation on the interface contacts between subunits of hemoglobin.

In order to obtain knowledge of molecular mechanism underlying the cooperative ligand binding to hemoglobin (Hb), atomic contacts between subunits in the hypothetical intermediate quaternary structures have been investigated using the atomic coordinates of O2-liganded subunits' within the T-state. The atomic coordinates are obtained by energy minimization of hypothetical atomic coordinate sets in which an O2 molecule is placed near the iron atom of a deoxy subunit. In an 'O2-liganded beta-subunit', drastic atomic displacements occur at the distal side amino-acid residues E11Val and E12Leu in order to produce sufficient space for the O2 molecule, while this kind of atomic displacement cannot be seen in an 'O2-liganded alpha-subunit'. Therefore the features of structural alterations at heme surroundings by O2 ligation are markedly different between alpha- and beta-subunits. When an alpha-subunit in deoxy Hb is replaced by the corresponding 'O2-liganded subunit', tight inter-subunit contacts between alpha 1FG4(92)Arg and the two residues beta 2C3(37)Trp and beta 2C6(40)Arg strengthen greatly, while, when a beta-subunit is replaced by a corresponding O2-liganded one, serious steric hindrances occur between beta 2FG4(97)His and alpha 1CD2(44)Pro. The characteristics of structural alteration confirm that the effect of O2-ligation at alpha-subunit is transmitted to the other subunit through the alpha FG4Arg. However, the role of FG4His in beta-subunit is not as clear as that of the alpha FG4Arg.

Calculation of the energy difference between the quaternary structures of deoxy- and oxyhemoglobin.

The energy difference between the quaternary structures of deoxy- and oxyhemoglobin is evaluated on the basis of the atomic coordinates determined by X-ray diffraction analysis. Calculation of the van der Waals interaction between subunits shows that in a hemoglobin molecule as a whole, the interaction is more attractive in the oxy form than in the deoxy form by about 8 kcal/mol, and that in each pair of two subunits except the pair alpha1alpha2, the interaction energy varies by about 15 kcal/mol. The electrostatic interactions originating in the partial charges on all constituent atoms of hemoglobin and in the polar residues on the surface of hemoglobin make only a small contribution to the energy difference between the quaternary structures of deoxy- and oxyhemoglobin. Thus, the contribution of the clusters of the polar residues in the internal cavity between like subunits and also of the freedom of rotation of the C-terminal of each subunit in oxyhemoglobin may be important energetically in the transition from deoxy to oxy quaternary structure. In this point, the present calculation supports Perutz' model, but suggests necessity of further investigations on the transitional characteristics of the quaternary structure in the intermediate steps of oxygenation. The discussion on the transitional characteristics is given in the last section.

A study on the quaternary structure change of hemoglobin in the ligation process.

In order to inquire into the molecular mechanism underlying the cooperative ligand binding to hemoglobin (Hb), conformational interaction at the interfaces between subunits are investigated on the basis of the atomic coordinates of human deoxy and human carbonmonoxy Hbs. Hypothetical intermediate structures are used, each of which is obtained from the procedure where one or more subunits in deoxy Hb are replaced by the corresponding CO-liganded subunits in carbonmonoxy Hb using the method of superimposition of two sets of atomic coordinates. When either alpha or beta subunit is substituted with the corresponding subunit in carbonmonoxy Hb, serious steric hindrances are produced between alpha 1FG4(92)Arg and beta 2C3(37)Trp or between alpha 1C6(41)Thr and beta 2FG4(97)His, all of which belong to the allosteric core affected directly by ligand binding. These steric hindrances become more serious when both alpha 1(alpha 2) and beta 2(beta 1) subunits are substituted. Therefore the change in the relative distance between iron atom and porphyrin by ligation results in strain in the C-terminal residues as an effect of the steric hindrance between the FG and C segments. However, no steric hindrance can be seen between subunits when the subunits in carbonmonoxy Hb are substituted with the corresponding subunits in deoxy Hb. The nature of the quaternary structural change from liganded to deoxy Hb seems to be different from that from deoxy to liganded Hb.

An electrochemical study of the iron storage protein, ferritin.

The electrochemical behavior of iron storage protein, ferritin (molecular weight 700 000) was studied by pulse polarography, cyclic voltammetry and 'potentiometry' monitored by colorimetry. Comparison between in vivo and voltammetric redox reactions of the ferritin Fe(III) core and thermodynamic E'0 values for native ferritin are described.

Fluorescence quenching measurements of the membrane bound lipid haptens with different length spacers.

Dansyl lipid haptens with three different length (short, intermediate and long) spacers have been incorporated into DMPC or DPPC liposomes. Anti-DNS-IgG bound most efficiently to these liposomes containing lipid haptens with an intermediate length spacer, although the binding efficiency became more increased when liposomes were made of a mixture of phospholipid (DMPC or DPPC) and cholesterol. To explain these results we have measured the accessibility of dansyl lipid haptens in liposomal membranes by the fluorescence quenching method. It was found that the dansyl haptens located on the surfaces of DMPC (or DPPC) membranes and fluorescence quenchers (iodide ions) possessed almost similar accessibility for the dansyl haptens with different length spacers. However, in the DMPC (or DPPC) membranes with 50% cholesterol, a part of the dansyl haptens became buried into the interior of the liposomal membranes depending on the length of the spacer and another part removed into the aqueous solution with greater affinity for antibody. These results were explained well by our recent model for antibody binding to the membrane-bound lipid haptens.

Proton nuclear magnetic resonance study of human interleukin 6: chemical modifications and partial spectral assignments for the aromatic residues.

Partial assignments for the 1H-NMR resonances of the aromatic residues in human interleukin 6 (IL-6) are reported. The homonuclear Hartmann-Hahn spectrum clearly shows all connectivities for the histidine, tyrosine and tryptophan residues that exist in IL-6. Using a deuterium exchange method, the imidazole proton resonances of His-16 and His-165 have been assigned. Iodination of the tyrosine residues led to the assignment of Tyr-32. Photo-chemically induced dynamic nuclear polarization data have shown that His-16, Tyr-32 and Trp-158 are exposed to solvent, whereas His-165, Tyr-98 and Tyr-101 are buried. Iodination of Tyr-32 gave no significant effect on IL-6 activity, suggesting that Tyr-32 is not responsible for IL-6 activity.

Characterization of partially purified prostaglandin E2 receptor from the canine renal medulla: evidence for physical association of the receptor protein with the inhibitory guanine nucleotide-binding protein.

Prostaglandin (PG) E2 binding protein, a putative PGE2 receptor, was purified 26-fold with 0.4% recovery from canine renal outer medullary membranes solubilized with 12% digitonin with the sequential use of a Superose 12, Wheat Germ Agglutinin (WGA) Affigel 10, DEAE-5PW and Ampholine column chromatographies. The final preparation retained the binding activity specific for PGE2, but lost most of the sensitivity to guanosine-5'-(gamma-thio)triphosphate (GTP gamma S). An antibody against alpha subunit of the inhibitory guanine nucleotide-binding protein (alpha Gi)1 and alpha Gi2 or that against common sequences of alpha subunit of guanine nucleotide-binding proteins (alpha G(common)) reacted at 41 kDa protein in the sample of each step of purification, but failed to do so in the final preparation. An antibody against alpha Gi3 or alpha Go had no effect. In fact, peaks of the binding activity and immunoreactivity for alpha Gi1,2 were chromatographically separated by isoelectric focusing. Moreover, antibodies against alpha G(common) or alpha Gi1,2, but not that against alpha Gi3 and alpha Go, precipitated PGE2 binding activity in the active fractions of WGA-Affigel 10 column chromatography. These results suggest that the PGE2 receptor is an acidic glycoprotein and that Gi1 or Gi2 is physically associated with the PGE2 receptor and dissociates from the receptor protein during purification procedures.

Role of the domain-domain interaction in the construction of the antigen combining site. A comparative study by 1H-15N shift correlation NMR spectroscopy of the Fv and Fab fragments of anti-dansyl mouse monoclonal antibody.

A comparative NMR structural study of anti-dansyl Fv and Fab fragments is reported. Both of these antigen binding fragments have been prepared using antibodies that originate from the identical anti-dansyl switch variant cell lines. It has been confirmed that the Fv and Fab fragments possess the identical binding property. The antigen binding fragment analogs labeled with 15N of the main chain amide group of the aromatic residues (His, Phe, Trp, and Tyr) were used. The chemical shift and hydrogen-deuterium exchange rate of the amide protons are compared for the Fv and Fab fragments. On the basis of the NMR data obtained, we have concluded that (1) the structural change induced in the VH domain upon antigen binding significantly affects the dynamical structure of the VL domain and (2) the existence of the constant regions affects the fluctuation of the VL domain, increasing the thermal stability of the variable region.

Dynamical structure of the hinge region of immunoglobulin G as studied by 13C nuclear magnetic resonance spectroscopy.

The segmental flexibility of the hinge region of immunoglobulin G (IgG) molecules has been considered to play important roles in the mediation between antigen recognition and effector functions. Here we report the result of a 13C nuclear magnetic resonance study of the dynamical structure of the hinge region of a mouse monoclonal IgG2a antibody. In the present work, a variety of IgG2a analogs selectively labeled with 13C at the carbonyl carbon atom were prepared. The flexibility of the hinge region was discussed on the basis of the linewidth data of 13C resonances obtained using a Carr-Purcell-Meiboom-Gill pulse train. It was concluded that the hinge region of the mouse IgG2a molecule has a mosaic structure with heterogeneous nature of flexibility with the rigid core part flanked by upper and lower hinges, which are highly flexible peptide segments. Cleavage of the inter-chain disulfide bridges led to homogeneous flexibility of the hinge region, except for the N-terminal side of the upper hinge. It was suggested that the rigidity in the N-terminal side of the upper hinge is due to the interaction with the CH1 domain. Biological significance of the mosaic structure of the hinge region was briefly discussed.

The pH-dependent structural variation of complementarity-determining region H3 in the crystal structures of the Fv fragment from an anti-dansyl monoclonal antibody.

The Fv fragment from an anti-dansyl antibody was optimally crystallized into two crystal forms having slightly different lattice dimensions at pH 5.25 and 6.75. The two crystal structures were determined and refined at high resolution at 112 K (at 1.45 A for the crystal at pH 5.25 and at 1.55 A for that at pH 6.75). In the two crystal structures, marked differences were identified in the first half of CDRH3 s having an amino acid sequence of Ile95H-Tyr96H-Tyr97H-His98H-Tyr99H-Pro1 00H-Trp100aH-Phe100bH-Ala101H- Tyr102H. NMR pH titration experiments revealed the p Kavalues of four histidine residues (His27dL, His93L, His55H and His98H) exposed to solvent. Only His98H (p Ka=6.3) completely changed its protonation state between the two crystallization conditions. In addition, the environmental structures including hydration water molecules around the four histidine residues were carefully compared. While the hydration structures around His27dL, His93L and His55H were almost invariant between the two crystal structures, those around His98Hs showed great difference in spite of the small conformational difference of His98H between the two crystal structures. These spectroscopic and crystallographic findings suggested that the change in the protonation state in His98H was responsible for the structural differences between pH 5.25 and 6.75. In addition, the most plausible binding site of the dansyl group was mapped into the present structural models with our previous NMR experimental results. The complementarity-determining regions H1, H3 and the N-terminal region in the VH domain formed the site. The side-chain of Tyr96H occupied the site and interacted with Phe27H of H1, giving a clue for the binding mode of the dansyl group in the site.

NMR analysis of the interaction between protein L and Ig light chains.

Protein L is a cell wall protein expressed by some strains of the anaerobic bacterial species Peptostreptococcus magnus. It binds to immunoglobulin (Ig) light chains predominantly of the kappa subtype from a wide range of animal species. This binding is mediated by five highly homologous repeats designated as B1-B5, each of which comprises 72 to 76 amino acid residues. The fold of the Ig light chain-binding B1 domain of protein L has previously been shown to comprise an alpha-helix packed against a four-stranded beta-sheet. The Ig-binding region of the protein L domain involves most of the residues in the second beta-strand, the C-terminal residues of the alpha-helix, and residues in the loop connecting the alpha-helix with the third beta-strand. In the present study, we have identified the protein L-binding site of an Ig light chain by use of stable isotope-assisted NMR spectroscopy. The light chain of a murine monoclonal anti-17alpha-hydroxy-progesterone Fab fragment (IgG2b, kappa) was selectively labeled with 13C at carbonyl groups of Ala, Arg, Cys, Ile, Lys, Met, Phe, Trp, or Tyr. The residues in which the carbonyl 13C chemical shift was significantly perturbed upon binding of the protein L B1 domain were preferentially found in the second beta-strand of the variable kappa domain and parts of its flanking beta-strands. None of these residues were affected by the addition of the antigen against which the monoclonal Fab fragment is directed. Therefore, we conclude that protein L binds to the outer surface of the framework region of the V(L) domain, primarily involving the V(L) second strand, and that this binding is independent of antigen-binding. The present NMR data, in combination with sequence comparisons between kappa light chains with and without protein L affinity, suggest that the amino acid substitutions at positions 9, 20, and/or 74 of the kappa light chains could crucially affect the interaction between protein L and the V(L) domain.

Three-dimensional structures of the Fab fragment of murine N1G9 antibody from the primary immune response and of its complex with (4-hydroxy-3-nitrophenyl)acetate.

The three-dimensional structures of the Fab fragment, in its unliganded and liganded crystals, of mouse anti-(4-hydroxy-3-nitrophenyl)acetate (NP) antibody N1G9 have been determined by the molecular replacement method. The unliganded and NP-liganded structures were refined at 2.4 A resolution to crystallographic R-factors of 0.194 and 0.196, respectively. Antibody N1G9 bears lambda light chains, and is one of the primary immune response antibodies. Fab N1G9 exhibits an elbow angle of 197 degrees in both structures. This large angle is ascribed to the VL-CL interface formed by lambda-chain residues. A hydrophobic pocket surrounded by the complementarity-determining regions except L2 is identified as a hapten-binding site. Between the liganded and unliganded structures, root-mean-square (r.m.s.) positional deviations are 0.42 A for the main-chain atoms, and 0.74 A for all the protein atoms. The major structural differences between these structures are localized in the hapten-binding site, and yield an r.m.s. deviation of 1.03 A for the side-chain atoms. The soaked NP ligand is in van der Waals contact with the aromatic side-chains of Tyr32L and Trp91L of the light chain, and Trp33H and Tyr97H of the heavy chain, and is hydrogen-bonded to the side-chains of Trp96L, His35H, Arg50H, Tyr95H, and Ser100aH. The side-chain of Lys58H is salt-bridged to the NP hydroxyl group. The side-chains of Arg50H, Trp33H, and Tyr97H are shifted toward the NP carboxyl group. The side-chain of Trp33H, whose replacement to Leu increases affinity by tenfold, is sandwiched between the Arg50H and Tyr97H side-chains, and is in cramped contact both with the ligand and with these side-chains. Affinity increases in the maturation of the anti-NP antibodies are ascribable to conformational relief of these cramped contacts through the replacement of Trp33H or through suitable structural alterations in the H3 region.

Three-dimensional solution structure of the calcium channel antagonist omega-agatoxin IVA: consensus molecular folding of calcium channel blockers.

The three-dimensional solution structure of omega-agatoxin IVA, which is a specific blocker of the P-type calcium channel isolated from funnel web spider venom and has a molecular mass of 5.2 kDa, was determined by two dimensional 1H NMR spectroscopy, combined with simulated annealing calculations. On the basis of 563 experimental constraints, including 516 distance constraints obtained from the nuclear Overhauser effect, 21 torsion angle (phi, chi 1) constraints, and 26 constraints associated with hydrogen bonds and disulfide bonds, a total of 14 converged structures were obtained. The atomic root mean square difference for the 14 converged structures with respect to the mean coordinates is 0.42 (+/- 0.07) A for the backbone atoms (N, C alpha, C) and 0.95 (+/- 0.15) A for all heavy atoms of the central part (residues 4 to 38) constrained by four disulfide bonds. The N- and C-terminal segments (residues 1 to 3 and 39 to 48, respectively) have a disordered structure in aqueous solution. The molecular structure of omega-agatoxin IVA is composed of a short triple-stranded antiparallel beta-sheet, three loops, and the disordered N- and C-terminal segments. The overall beta-sheet topology is +2x, -1, which is the same as that reported for omega-conotoxin GVIA, an N-type calcium channel blocker. Irrespective of differences in the number of disulfide bonds and low primary sequence homology, these two peptide toxins show a significant structural similarity in three dimensions. The whole-cell voltage-clamp recording using rat cerebellar slices suggests that the hydrophobic C-terminal segment of omega-agatoxin IVA, which does not exist in omega-conotoxin GVIA, plays a crucial role in the blocking action of omega-agatoxin IVA on the P-type calcium channel in rat cerebellar Purkinje cells. The present study provides a molecular basis for the toxin-channel interaction, and thereby provides insight into the discrimination of different subtypes of calcium channels.

Structural basis of the interaction between IgG and Fcgamma receptors.

The binding of multivalent antigen-antibody complexes to receptors for the Fc portion of IgG (FcgammaR) induces the clustering of the FcgammaR and triggers cell activation leading to defence reactions against pathogens. The Fc portion of IgG consists of two identical polypeptide chains which are related to each other by a 2-fold axis and are folded in two structural domains, the C(H)2 domain, near the flexible hinge region of the IgG molecule, and the C(H)3 domain. We studied the interaction in solution between the Fc fragment of mouse IgG2b and the extracellular region of mouse FcgammaRII. We find that one Fc molecule binds one FcgammaRII molecule only. Using NMR spectroscopy, we show that FcgammaRII binds to a negatively charged area of the C(H)2 domain, corresponding to the lower hinge region, and that the binding of FcgammaRII onto one of the two symmetrically related sites on the Fc induces a conformational change in the other site. We therefore propose a model that explains why IgG molecules are unable to trigger FcgammaR-mediated cellular responses spontaneously in the absence of crosslinking by multivalent antigens.

Recommendations for the presentation of NMR structures of proteins and nucleic acids.

The recommendations presented here are designed to support easier communication of NMR data and NMR structures of proteins and nucleic acids through unified nomenclature and reporting standards. Much of this document pertains to the reporting of data in journal articles; however, in the interest of the future development of structural biology, it is desirable that the bulk of the reported information be stored in computer-accessible form and be freely accessible to the scientific community in standardized formats for data exchange. These recommendations stem from an IUPAC-IUBMB-IUPAB inter-union venture with the direct involvement of ICSU and CODATA. The Task Group has reviewed previous formal recommendations and has extended them in the light of more recent developments in the field of biomolecular NMR spectroscopy. Drafts of the recommendations presented here have been examined critically by more than 50 specialists in the field and have gone through two rounds of extensive modification to incorporate suggestions and criticisms.

Visualization of Freezing Behaviors in Leaf and Flower Buds of Full-Moon Maple by Nuclear Magnetic Resonance Microscopy.

1H-Nuclear magnetic resonance (NMR) microscopy was used to study the freezing behavior of wintering buds of full-moon maple (Acer japonicum Thunb.). The images obtained predominantly reflected the density of mobile (i.e. non-ice) protons from unfrozen water. A comparison of NMR images taken at different subfreezing temperatures revealed which tissues produced high- and low-temperature exotherms in differential thermal analyses. In leaf and lower buds of A. japonicum, the scales and stem bark tissues were already frozen by -7[deg]C, but the primordial inflorescence and terminal primordial shoots remained supercooled at -14[deg]C, and the lateral primordial shoots were unfrozen even at -21[deg]C. The freezing of these supercooled tissues was associated with their loss of viability. The size of the supercooled primordial shoots and inflorescences was gradually reduced with decreasing temperature, indicating extraorgan freezing in these tissues. During this process the formation of dark regions beneath the primordia and subsequent gradual darkening in the basal part of supercooled primordia were visible. As the lateral shoot primordia were cooled, the unfrozen area was considerably reduced. Since the lateral primordia remained viable down to -40[deg]C, with no detectable low-temperature exotherms, they probably underwent type I extraorgan freezing. Deep supercooling in the xylem was clearly imaged. NMR microscopy is a powerful tool for noninvasively visualizing harmonized freezing behaviors in complex plant organs.

Mn2+-probe ESR method for the analyses of the dissociation of charged residues on the surface of immunoglobulins.

Dissociation of charged residues on the surface of immunoglobulins was analysed by an Mn2+ probe ESR method that has been developed in our previous work. Several kinds of IgG proteins and their Fab and Fc fragments were used for the experiments. The pH dependence of the intensity of ESR signals was analysed. It was shown that the number of Asp, Glu and His residues on the surface of Fc is about twice as many as that of Fab. The accessible surface area of amino acid residues calculated using X-ray crystallographic data is quite consistent with the present ESR experiments. This indicates that the number of the Asp, Glu and His residues on the surface of IgG molecules in solution is similar to that in the crystal. The Mn2+-probe ESR method was also applied to other classes of immunoglobulins, i.e. IgA and IgM. It was demonstrated that the IgA protein, which is known to lack the ability to bind Clq, has on the surface of it a smaller number of Asp, Glu and His residues as compared to IgG and IgM proteins. On the basis of these results obtained by the Mn2+-probe ESR method, we suggest that the Clq molecule, which is a basic protein, interacts favorably with the Fc portion whose surface is more negatively charged with Asp and Glu residues, compared to the Fab portion. Fine adjustment of fitting of the head of the Clq molecule into the CH2 domain of the Fc portion presumably follows for optimum binding. It was also demonstrated that Ser and Thr residues are much more abundant on the surface of Fab than in the case of Fc. We suggest that the Ser and Thr residues on the surface of Fab play an important role for binding of C4b upon activation of the complement system.

Molecular structural studies of effector functions of a mouse immunoglobulin G that lacks the entire CH1 domain: small-angle X-ray scattering, nanosecond fluorescence depolarization and stable isotope-aided NMR analyses.

Molecular structural studies are reported of a short-chain mouse IgG2a antibody that lacks the entire CH1 domain. We have recently shown that (1) this short-chain antibody comprises two components in which the inter light-chain disulfide bridge does and does not exist, and (2) these two components are different in the constitutive complement-activating activity [Mizutani et al. (1993) J. Immunol. 150, 131-138]. Structures were compared for these two components on the basis of small-angle X-ray scattering, nanosecond fluorescence depolarization and isotope-aided NMR data. It has been discussed how the presence and absence of the inter light-chain disulfide bridges affect the complement-activating activity of the two components of the short-chain antibody.