Conformation of Immunoglobulin M. III. Structural requirements of antigen for complement fixation by equine IgM.

Complexes of IgM equine anti-dansyl antibodies and different dansyl substituted carriers were tested for their ability to fix complement (C). Only dansyl92-Ficoll and dansyl12-poly-L-lysine were found to be effective. Dansyl13-bovine serum albumin, dansyl127-keyhole limpet hemocyanin, and reduced and alkylated dansyl10-ribonuclease were all ineffective. Lack of C fixation by the dansyl-ribonuclease was not due to lack of antibody-antigen complex formation, since binding at the concentrations employed for C fixation was established. However, in contrast, polymerized dansyl-ribonuclease (polydisperse, with m.w. = 74,000 to 230,000) was very effective in inducing C fixation. These results suggest that large antigen size is necessary for IgM to bind in a multivalent fashion to provide the correct conformation for C fixation. A similar conclusion had been made in earlier studies on rabbit IgM by Cunniff and Stollar. Since optimal C fixation occurred at lower antigen concentrations than maximal precipitation, it would appear that complexes in which several combining sites within a given IgM molecule may be bound to the same antigenic surface may be the most effective. The observation that the amount of C1q bound to antibody was the same in the presence and absence of antigen suggests that enhanced C fixation by antibody-antigen complexes is due to additional C component interactions such as C1r or C1s.

Nanosecond fluorescence spectroscopy of pyrenebutyrate anti-pyrene antibody complexes.

The utility of the long-lived fluorophore, pyrene, as a probe in nanosecond fluorescence depolarization measurements was investigated using pyrenebutyrate bound in the combining sites of rabbit antipyrenebutyrate immunoglobulin G. The time dependent anisotropy decay data points showed very little scatter in the time interval 0-350 ns, which is more than three times the comparable time interval observed with epsilon-1-dimethylamino-5-naphthalenesulfonyllysine (DNS-lysine) bound in the combining sites of anti-DNS antibodies [Holowka, D.A., and Cathou, R.E. (1976), Biochemistry 15, 3379]. Thus, the use of pyrene can significantly extend the range of macromolecular rotational correlation times that can be measured by the single photon technique. In the present investigation, we confirmed the presence of Fab segmental fexibility in immunoglobulin G molecules specific for a hapten different from DNS-lysine. We obtained a value of about 135 ns for the longer rotational correlation time which probably represents global rotation of the entire molecule. In the course of these experiments, we have also found that the combining sites of antipyrenebutyrate antibodies are, as expected, relatively nonpolar.

Fine structure of three different anti-fluorescein combining sites: induced circular dichroism of hapten bound to autologous and heterologous recombinants.

We have previously reported the sequential appearance in hyperimmunized rabbits of three distinct types of antifluorescein-combining sites that can be distinguished by the characteristic induced circular dichroism (CD) of the bound hapten, fluorescein. Such induced CD depends on the configuration of the surrounding residues, and, in the case of anti-fluorescein, can be localized to the configuration of the sub-site which binds the hydroxyxanthenone moiety of fluorescein. In the present investigation, we have studied the fine structure of both autologous and heterologous recombinant sites and of the free chains by measuring the induced CD of bound hapten. Heavy chain dimers at pH 5.4 bound fluorescein that displayed a weak negative CD band which is different from that observed in any of the native antibodies. No induced CD was observed with light chains. Thus, the hydroxyxanthenone subsite does not exist intact in any of the isolated chains. Fluorescein bound to purified autologous recombinants, when studied at pH 7.5, exhibited CD spectra very similar to those observed for the original antibodies. Most significantly, fluorescein bound to purified heterologous recombinants, prepared in all possible combinations, showed CD spectra most similar to those exhibited by sites from which the heavy chain was derived. Thus, the microenvironment of the subsite appears to be due primarily to the heavy chain.

Conformation of immunoglobulin M. I. Characterization of anti-epsilon-1-dimethylamino-5-naphthalenesulfonyl-L-lysine immunoglobulin M antibodies from horse, pig, and shark.

IgM antibodies specific for the fluorophore epsiolon-1-dimethylamino-5-naphthalenesulfonyl-L-lysine(DNS-lysine) were elicited in the horse and nurse shark by immunization with a DNS-lysine streptococcal conjugate; the antibodies were purified by specific adsorption with an immunoadsorbent followed by gel filtration to select the IgM class (molecular weight 900 000). About 90% of the equine anti-DNS was IgM.DNS-Lysine, when bound in the combining sites of a population of these anti-DNS IgM antigodies from horse and nurse shark, as well as from pig, exhibited a marked fluorescence enhancement and shift of the emission spectrum to shorter wavelengths compared with emission in aqueous solution; these results indicate that the environments of the anti-DNS combining sites of this population were relatively hydrophovic. Approximately one-third of the ten possible combining sites in each of these anti-DNS IgM species bound DNS-lysine in this manner with an average intrinsic association constant (Ko) of greater than 10(6) M-1. Small differences were noted in binding behavior among the three species of antibodies. The enzymatic susceptibility of equine IgM was similar to that of human IgM. (Fab')2mu, Fab'mu, and Fabmu fragments were prepared following digestion with pepsin. These fragments could be clearly differentiated on the basis of molecular size. They bound DNS-lysine with the same affinity as intact IgM and the DNS-lysine-fragment complexes exhibited the same spectral properties as the parent IgM. It was concluded that the anti-dNs IgM antibodies from all three species, as well as the enzymatic fragments, were suitable for nanosecond depolarization studies which are reported in the accompanying paper Holowka, D.A., and Cathou, R.E. (1976), Biochemistry, the following papter in this issue.

Conformation of immunoglobulin M. 2. Nanosecond fluorescence depolarization analysis of segmental flexibility in anti-epsilon-l-dimethylamino-5-naphthalenesulfonyl-L-lysine anti-immunoglobulin from horse, pig, and shark.

The rotational motions of immunoglobulin M (IgM) were investigated by the nanosecond fluorescence depolarization technique. The fluorophore epsilon-1-dimethylamino-5-naphthalenesulfonyl-L-lysine (DNS-lysine) was specifically bound in the combining sites of anti-DNS IgM antibodies from the horse, pig, and nurse shark. Fluorescence lifetime analysis showed the presence of a long lifetime component (21-27 ns) with antibodies from all three species. With the mammalian antibodies, the fluorophore appeared to be rigidly bound in the combining sites as judged by the presence of induced circular dichroism of DNS-lysine (equine antibodies) and single exponential anisotropy decay of the isolated Fabmu fragments (equine and porcine antibodies). The small amount of available purified nurse shark antibody did not allow preparation of fragments or induced circular dichroism measurements to directly determine rigidity of fluorophore binding. However, at least some of the hapten must have been rigidly bound since long rotational correlation times were measured for the shark DNS-lysine-anti-DNS complexes. When the emission anisotropy of the fluorophore-anti-DNS IgM complexes was measured as a function of time, it was found that all three antibody species exhibited restricted segmental flexibility in the nanosecond time range. Moreover, when the equine anti-DNS IgM was exposed to 1 M acetic acid for 1 h, the antibody underwent a conformational change which resulted in an increase in its overall flexibility. Comparison of the rotational correlation times of native equine IgM and of proteolytic fragments indicated that flexibility of IgM consists of either hindered rotation of the Fab'mu segment or a combination of at least two modes of motion: rotation of Fabmu and/or Fab'mu and bending of the entire (Fab')2mu region as a unit. Similar modes of flexibility also occur in native porcine IgM. In acid exposed equine IgM, the major contribution to depolarization is from independent rotation or wagging of the Fab'mu segments. Thus, acid apparently causes a conformational change in or near the Cmu2 domains. In contrast, flexibility in nurse shark IgM appears to involve only bending of (Fab')2mu as a unit. Our results suggest that segmental flexibility is an essential functional feature of all IgM antibodies and that control of this flexibility through domain interactions may play an important role in such conformationally sensitive functions as complement fixation.

Changes in intrinsic circular dichroism of several homogeneous anti-type 3 pneumococcal antibodies on binding of a small hapten.

Three homogeneous antibodies against the capsular polysaccharide of Type III pneumococci of similar specificities and affinities were purified from a single bleeding of an individual rabbit and fractionated by isoelectric focusing. A comparison of the circular dichroic spectra of the three antibodies revealed differences among them, although the spectra were generally similar to those obtained previously for heterogeneous rabbit antibodies [Cathou, R. E., Kulcycki, A., Jr. & Haber, E. (1968) Biochemistry 7, 3958]. On binding of the hexasaccharide, -[-->3)-beta-D-glucuronic acid-(1-->4)-beta-D-glucose-(1-](3)-->, significant changes in all three circular dichroic spectra were observed. Since the oligosaccharide alone shows no transitions above 220 nm, these spectral changes can be attributed to changes in the intrinsic optical activity of the antibodies. Calculated difference circular dichroism spectra (of antibody minus that of antibody-hapten complex) of the three antibodies are different from each other, and resemble spectra of tryptophan and tyrosine derivatives. These changes in optical activity can be ascribed to changes in the asymmetric environments of aromatic chromophores directly in the combining site and/or to changes in orientation inside or beyond the site. Since the hapten-antibody interactions are different in the three antibodies, as shown by the difference spectra, the structures of the combining sites are presumably also different. We have interpreted these observations to mean that a relatively simple ligand may be bound by several different complementary sites.

The shape of immunoglobulin G molecules in solution.

We have studied the shape of rabbit Immunoglobulin G molecules in solution by using singlet-singlet energy transfer to determine the minimum distance between the two hapten binding sites. A hybrid antibody was prepared in which one site specifically bound the energy donor, epsilon-dansyl-lysine, and the other site bound the energy acceptor, fluorescein. For this donor-acceptor pair, R(0) was calculated to be 4.8 +/- 0.2 nm (48 +/- 2 A). From a comparison of the lifetime of the donor's excited state in the presence or absence of acceptor, it was found that no energy transfer had occurred in the hybrid. Since the maximum distance over which transfer is measurable was 8.2 nm (82 A; 1.7 R(0)), and since the Fab moieties exhibit segmental flexibility, the average distance between the two hapten-binding sites was estimated to be 9.2-10 nm (92-102 A). If one assumes that the length of the Fab fragment is 7 nm (70 A), the corresponding minimum angle between Fab moieties, alpha(M), would be 80-95 degrees . The molecules in solution, thus, have an open Y- or T-shaped configuration in which the hapten binding sites are not more than 2.5 nm (25 A) from the extreme ends of the Fab fragments. The existence of conformations in which alpha(M) is less than 80 degrees , as has been observed in some antibody-antigen complexes, must therefore be the result of definite conformational changes.