Properties of penicillin amidase covalently bound to cellulose matrices]

Properties of penicillinamidase (PA) covalently bound with the cellulose matrix were studied. The efficiency of the binding depended on the bind type and purity of the native enzyme taken for binding. Stability of the immobilized PA (IPA) was studied at wide pH ranges. The effect of the ion strength, substrate concentration and purity of the native PA on stability of IPA was also investigated. The maximum stability of the enzyme was observed at pH 6.5-7.0 Stability of IPA depended on the purity of the native enzyme. When PA of the diazotized ether of cellulose containing amino groups was used, the enzyme was destabilized. IPA prepared on chlortriazinylcellulose was more stable than the respective native PA almost by I order.     

Immobilization of immunoglobulins on silica surfaces. Stability.

The development of new immunosensors based on surface-concentration-measuring devices requires a stable and reproducible immobilization of antibodies on well-characterized solid surfaces. We here report on the immobilization of immunoglobulin G (IgG) on chemically modified silica surfaces. Such surfaces may be used in various surface-oriented analytical methods. Reactive groups were introduced to the silica surfaces by chemical-vapour deposition of silane. The surfaces were characterized by ellipsometry, contact-angle measurements and scanning electron microscopy. IgG covalently bound by the use of thiol-disulphide exchange reactions, thereby controlling the maximum number of covalent bonds to the surface, was compared with IgG adsorbed on various silica surfaces. This comparison showed that the covalently bound IgG has a superior stability when the pH was lowered or incubation with detergents, urea or ethylene glycol was carried out. The result was evaluated by ellipsometry, an optical technique that renders possible the quantification of amounts of immobilized IgG. The results outline the possibilities of obtaining a controlled covalent binding of biomolecules to solid surfaces with an optimal stability and biological activity of the immobilized molecules.     

Dimeric galectin-1 binds with high affinity to alpha2,3-sialylated and non-sialylated terminal N-acetyllactosamine units on surface-bound extended glycans

Galectin-1 is a member of the galectin family of glycan-binding proteins and occurs as an approximately 29.5-kDa noncovalent homodimer (dGal-1) that is widely expressed in many tissues. Here, we report that human recombinant dGal-1 bound preferentially and with high affinity (apparent K(d) approximately 2-4 microM) to immobilized extended glycans containing terminal N-acetyllactosamine (LN; Galbeta1-4GlcNAc) sequences on poly-N-acetyllactosamine (PL; (-3Galbeta1-4GlcNAcbeta1-)(n)) sequences, complex-type biantennary N-glycans, or novel chitin-derived glycans modified to contain terminal LN. Although terminal Gal residues are important for dGal-1 recognition, dGal-1 bound similarly to alpha3-sialylated and alpha2-fucosylated terminal LN, but not to alpha6-sialylated and alpha3-fucosylated terminal LN. The binding specificity of human recombinant dGal-1 was similar to that observed with purified bovine heart-derived dGal-1. Unexpectedly, dGal-1 bound free ligands in solution with relatively low affinity and displayed no preference for extended glycans, indicating that dGal-1 preferentially recognizes extended glycans only when they are surface-bound, such as found on cell surfaces. Human dGal-1 also bound to both native and desialylated human promyelocytic HL-60 cells with similar affinity as observed for immobilized long chain PL. Binding to these cells was reduced upon treatment with endo-beta-galactosidase, which cleaves PL sequences, indicating that cell-surface PLs are ligands. To test the role of dimerization in dGal-1 binding, we examined the binding of a mutated form of dGal-1 that weakly dimerizes (monomeric Gal-1 (mGal-1)) and a covalently dimerized (chemically cross-linked) form of mGal-1 (cd-mGal-1). dGal-1 and cd-mGal-1 had similar affinities that were both approximately 3.5-fold higher for immobilized PL than observed for mGal-1, suggesting that dGal-1 acts as a dimer to cross-link terminal LN units on immobilized PL. These results indicate that dGal-1 functions as a dimer to recognize LN units on extended PLs on cell surfaces.     

A two-dimensional support for selective binding of polyhistidine-tagged proteins: identification of a proliferating cell nuclear antigen point mutant with altered function in vitro

Whatman 3MM paper was chemically modified to generate nickel-charged iminodiacetic acid paper (Ni2+-IDA paper). Bacteria were transformed with Escherichia coli expression plasmids coding for either unmodified proliferating cell nuclear antigen (PCNA) or PCNA containing a genetically engineered polyhistidine tract (his-tag) located at its NH2 terminus. They were then grown, induced, and lysed, and macromolecules were transferred to Ni2+-IDA paper. After exhaustive washing, his-tagged PCNA but not unmodified PCNA remained bound to the paper. Moreover, bound his-tagged PCNA was biochemically active in an in situ DNA synthesis assay with exogenous template-primer and purified calf thymus DNA polymerase delta (pol delta). Ni2+-IDA paper was used to identify a PCNA- point mutant that, relative to wild-type PCNA, promotes increased DNA synthesis by pol delta beyond a model abasic template site. In addition, metal-charged IDA paper promises to be generally useful for functional screening of cells expressing cloned proteins.     

Selective immobilization of multivalent ligands for surface plasmon resonance and fluorescence microscopy

Cell surface multivalent ligands, such as proteoglycans and mucins, are often tethered by a single attachment point. In vitro, however, it is difficult to immobilize multivalent ligands at single sites due to their heterogeneity. Moreover, multivalent ligands often lack a single group with reactivity orthogonal to other functionality in the ligand. Biophysical analyses of multivalent ligand-receptor interactions would benefit from the availability of strategies for uniform immobilization of multivalent ligands. To this end, we report the design and synthesis of a multivalent ligand that has a single terminal orthogonal functional group and we demonstrate that this material can be selectively immobilized onto a surface suitable for surface plasmon resonance (SPR) experiments. The polymeric ligand we generated displays multiple copies of 3,6-disulfogalactose, and it can bind to the cell adhesion molecules P- and L-selectin. Using SPR measurements, we found that surfaces displaying our multivalent ligands bind specifically to P- and L-selectin. The affinities of P- and L-selectin for surfaces displaying the multivalent ligand are five- to sixfold better than the affinities for a surface modified with the corresponding monovalent ligand. In addition to binding soluble proteins, surfaces bearing immobilized polymers bound to cells displaying L-selectin. Cell binding was confirmed by visualizing adherent cells by fluorescence microscopy. Together, our results indicate that synthetic surfaces can be created by selective immobilization of multivalent ligands and that these surfaces are capable of binding soluble and cell-surface-associated receptors with high affinity.     

Activity and specificity of covalently immobolized wheat germ agglutinin toward cell surfaces

Wheat germ agglutinin protein, which is able to agglutinate tumor cells better than normal cells, was covalently bound to polyacrylamide gel beads. The specific binding activity of the protein was preserved on these beads and was expressed heterogeneously by the binding of mouse leukemia cells (L1210) to the protein coupled gels. The selective activity of the immobilized protein was maximal when the number of sites available to covalently couple the protein was lowest. The application of this observation to the general field of covalent immobilization of proteins and enzymes may be of considerable utility.     

A novel microtiter plate based method for identification of B‐cell epitopes

A new type of microtiter plate capable of binding biomolecules covalently in a one step procedure was used to map linear B‐cell epitopes in two different proteins using a peptide‐based solid phase immunoassay. The method was compared with a conventional immobilization method using passive adsorption to microtiter plates. An array of 15‐mer peptides, overlapping by five amino acids, representing the entire sequences of ubiquitin and murine tumor necrosis factor‐α, respectively, was synthesized. The peptides were immobilized covalently using the new, specialized microtiter plates or non‐covalently using conventional ELISA microtiter plates of the high binder type. Subsequently, specific antisera to ubiquitin or murine tumor necrosis factor‐α were added to identify potential linear B‐cell epitopes. All peptides, which were recognized on the conventional microtiter plates, were also recognized on the plates with the covalently bound peptides. In addition, the covalent immobilization method revealed epitopes that were not identified using the method for non‐covalent binding although the peptides were in fact present on the non‐covalent binding surface. The interaction with the hydrophobic surface of the conventional microtiter plate apparently interfered negatively with antibody recognition. The covalently binding microtiter plates described here could be useful for identification of new B‐cell epitopes in protein antigens. Copyright © 1999 European Peptide Society and John Wiley & Sons, Ltd.     

C-terminal truncation and histidine-tagging of cytochrome c oxidase subunit II reveals the native processing site, shows involvement of the C-terminus in cytochrome c binding, and improves the assay for proton pumping

To enable metal affinity purification of cytochrome c oxidase reconstituted into phospholipid vesicles, a histidine-tag was engineered onto the C-terminal end of the Rhodobacter sphaeroides cytochrome c oxidase subunit II. Characterization of the natively processed wildtype oxidase and artificially processed forms (truncated with and without a his-tag) reveals Km values for cytochrome c that are 6-14-fold higher for the truncated and his-tagged forms than for the wildtype. This lowered ability to bind cytochrome c indicates a previously undetected role for the C-terminus in cytochrome c binding and is mimicked by reduced affinity for an FPLC anion exchange column. The elution profiles and kinetics indicate that the removal of 16 amino acids from the C-terminus, predicted from the known processing site of the Paracoccus denitrificans oxidase, does not produce the same enzyme as the native processing reaction. MALDI-TOF MS data show the true C-terminus of subunit II is at serine 290, three amino acids longer than expected. When the his-tagged form is reconstituted into lipid vesicles and further purified by metal affinity chromatography, significant improvement is observed in proton pumping analysis by the stopped-flow method. The improved kinetic results are attributed to a homogeneous, correctly oriented vesicle population with higher activity and less buffering from extraneous lipids.     

Direct electrochemistry of novel affinity-tag immobilized recombinant horse heart cytochrome c

During the last decade protein electrochemistry at miniaturized electrodes has become important not only for functional studies of the charge transfer properties of redox proteins but also for fostering the development of sensitive biosensor and bioelectronic devices. One of the major challenges in this field is the directed coupling between electronic and biologically active components. A prerequisite for a fast and reversible electron transfer between electrode and protein is that the protein can be bound to the electrode in a favourable orientation. We examined electrostatic and bioaffinity-tag binding strategies for the directed immobilization of horse heart cytochrome c (cytc) on gold electrode surfaces to achieve this goal. Horse heart cytc was expressed in E. coli either as non-modified or genetically modified, i.e. histidine (his)-tag containing protein. The his-tags were introduced at defined positions at the N- or C-terminus of the polypeptide. It was our aim to generate tagged-versions of cytc that facilitate strong electronic coupling between protein and electrode and, at the same time, retain their catalytic and regulatory properties. The combination of different immobilization strategies, e.g. his-tag and electrostatic immobilization also opens new avenues for bivalent immobilization of proteins. This is of interest for molecular bioelectronic and biosensing applications where the proteins are immobilized between two crossing electrodes.     

Development and characterization of Ni-NTA-bearing microspheres

BACKGROUND: For ease of purification, proteins are often expressed with a short affinity sequence of five or six adjacent histidine residues (His-tag). This His-tag binds to the metal of metal chelator complexes such as Ni(2+)-nitrilotriacetic acid (Ni-NTA) or -iminodiacetic acid (Ni-IDA). Chromatography resins bearing covalently attached metal chelator complexes are used widely for the easy affinity purification of His-tagged proteins or peptides. Because Ni-NTA microspheres were not commercially available at the beginning of our studies, we prepared and characterized such microspheres to immobilize His-tagged proteins and study their interactions. Our microspheres are of three types: (a) metal chelator complexes bound covalently to polystyrene microspheres, (b) metal chelator complexes bound covalently to silica microspheres, and (c) lipid-linked metal chelator complexes adsorbed to silica microspheres forming self-assembled bilayer membranes where the metal chelators have lateral mobility. METHODS: The microspheres bearing covalently attached Ni-chelator were synthesized by reacting a primary amine-bearing Ni-NTA ligand with carboxy-functionalized microspheres and then loading with Ni(2+). Microspheres with laterally mobile metal chelator were made by incubating glass microspheres with liposomes containing phosphatidylcholine (PC) and the metal chelating lipid 1,2-dioleoyl-sn-glycero-3-[(N (5-amino-1-carboxypentyl)iminodiacetic acid)succinyl]. Binding of a His-tagged enhanced green fluorescent protein (EGFP) was used to characterize these microspheres by flow cytometry for their specificity, sensitivity, capacity and stability. RESULTS: While all micospheres specifically bind His-tagged proteins, the conditions to achieve this are different for the polystyrene- and silica-based spheres. All three types of microspheres bind His-EGFP with saturation occurring at 30-50 nM and an apparent avidity (concentration of half-maximal binding) of approximately 1 to 2 x 10(-8) M at pH 7.4. Binding of His-EGFP is inhibited by imidazole or ethylene-diaminetetraacetic acid (EDTA). Polystyrene Ni-NTA microspheres showed significant nonspecific binding as measured by binding in the presence of imidazole or EDTA or by binding of fluorescent proteins lacking a His-tag. This nonspecific binding of proteins to and aggregation of polystyrene spheres could only be prevented by the inclusion of low concentrations of Tween 20, but not by including bovine serum albumin (BSA), polyethylene glycols, or polyvinylpyrrolidones as blocking agents. In contrast, silica-based microspheres with covalently attached Ni-NTA or silica microspheres bearing adsorbed bilayers that contain Ni-NTA-lipid showed little nonspecific binding in the presence of BSA. Our results on the stability of immobilization indicate that washing destabilizes the binding of His-tagged proteins to Ni-NTA microspheres. This binding consists of two interactions of different affinities. We also demonstrate that limited multiplexed analysis with differently sized silica microspheres bearing the Ni-NTA-lipid is feasible. CONCLUSIONS: The microspheres described are well suited to selectively immobilize His-tagged proteins to analyze their interactions by flow cytometry. The affinity and kinetic stability of the interaction of His-tagged proteins with Ni-NTA are insufficient to use Ni-NTA microspheres in multiplexed analysis formats where different His-tagged proteins are bound to distinct microspheres. Improvements towards this end (improved chelators and/or improved affinity tags) are critical for extending the use of this method. We are currently working on novel chelators to strengthen the stability of immobilization of His-tagged proteins to surfaces. Such improvements would greatly enhance the analysis of interactions of immobilized His-tagged proteins and could make the development of microsphere-based arrays with His-tagged protein/antibody possible.     

Binding and signaling of surface-immobilized reagentless fluorescent biosensors derived from periplasmic binding proteins

Development of biosensor devices typically requires incorporation of the molecular recognition element into a solid surface for interfacing with a signal detector. One approach is to immobilize the signal transducing protein directly on a solid surface. Here we compare the effects of two direct immobilization methods on ligand binding, kinetics, and signal transduction of reagentless fluorescent biosensors based on engineered periplasmic binding proteins. We used thermostable ribose and glucose binding proteins cloned from Thermoanaerobacter tengcongensis and Thermotoga maritima, respectively. To test the behavior of these proteins in semispecifically oriented layers, we covalently modified lysine residues with biotin or sulfhydryl functions, and attached the conjugates to plastic surfaces derivatized with streptavidin or maleimide, respectively. The immobilized proteins retained ligand binding and signal transduction but with adversely affected affinities and signal amplitudes for the thiolated, but not the biotinylated, proteins. We also immobilized these proteins in a more specifically oriented layer to maleimide-derivatized plates using a His(2)Cys(2) zinc finger domain fused at either their N or C termini. Proteins immobilized this way either retained, or displayed enhanced, ligand affinity and signal amplitude. In all cases tested ligand binding by immobilized proteins is reversible, as demonstrated by several iterations of ligand loading and elution. The kinetics of ligand exchange with the immobilized proteins are on the order of seconds.     

Enhanced molecular chaperone activity of the small heat-shock protein alphaB-cystallin following covalent immobilization onto a solid-phase support

The well-characterized small heat-shock protein, alphaB-crystallin, acts as a molecular chaperone by interacting with unfolding proteins to prevent their aggregation and precipitation. Structural perturbation (e.g., partial unfolding) enhances the in vitro chaperone activity of alphaB-crystallin. Proteins often undergo structural perturbations at the surface of a synthetic material, which may alter their biological activity. This study investigated the activity of alphaB-crystallin when covalently bound to a support surface; alphaB-crystallin was immobilized onto a range of solid material surfaces, and its characteristics and chaperone activity were assessed. Immobilization was achieved via a plasma-deposited thin polymeric interlayer containing aldehyde surface groups and reductive amination, leading to the covalent binding of alphaB-crystallin lysine residues to the surface aldehyde groups via Schiff-base linkages. Immobilized alphaB-crystallin was characterized by X-ray photoelectron spectroscopy, atomic force microscopy, and quartz crystal microgravimetry, which showed that 300 ng cm(-2) (dry mass) of oligomeric alphaB-crystallin was bound to the surface. Immobilized alphaB-crystallin exhibited a significant enhancement (up to 5000-fold, when compared with the equivalent activity of alphaB-crystallin in solution) of its chaperone activity against various proteins undergoing both amorphous and amyloid fibril forms of aggregation. The enhanced molecular chaperone activity of immobilized alphaB-crystallin has potential applications in preventing protein misfolding, including against amyloid disease processes, such as dialysis-related amyloidosis, and for biodiagnostic detection of misfolded proteins.     

The effect of cellular receptor diffusion on receptor-mediated viral binding using Brownian adhesive dynamics (BRAD) simulations

Brownian adhesive dynamics (BRAD) is a new method for simulating the attachment of viruses to cell surfaces. In BRAD, the motion of the virus is subject to stochastic bond formation and breakage, and thermal motion owing to collisions from the solvent. In the model, the virus is approximated as a rigid sphere and the cell surface is approximated as a rigid plane coated with receptors. In this article, we extend BRAD to allow for the mobility of receptors in the plane of the membrane, both before and after they are ligated by viral attachment proteins. Allowing the proteins to move within the membrane produced several differences in behavior from when the receptors are immobilized. First, the mean steady-state bond number is unaffected by changes in cellular receptor density because proteins are now free to diffuse into the contact area, and the extent of binding is dictated by the availability of viral attachment proteins. Second, the time required to reach steady-state binding increases as both the cellular receptor number decreases and the receptor mobility decreases. This is because receptor diffusion is a slower process than the binding kinetics of the proteins. Decreasing the rate of protein binding was found to decrease the fraction of viruses bound to steady state, but not the extent of binding for those viruses that were bound. Increasing the binding rate increased the fraction of viruses bound, until no further viruses could bind. Alterations in receptor binding kinetics had no discernable effect on the mean steady-state bond number between virus and cell, because interactions were of sufficiently high affinity that all available receptor-viral attachment proteins were destined to bind at steady state.     

Protein-coated agarose surfaces for attachment of cells

Plastic dishes were coated with an agarose layer. The layer was modified by covalently binding proteins to it, using the CNBr-method. Cells were seeded on the dishes and the number of attached cells was evaluated. The specificity of the attachment was demonstrated by showing that cells, carrying specific membrane-bound immunoglobulins, attached only to the corresponding anti-immunoglobulins. This indicated that the method could be used for cell sorting. The attachment of cells to proteins was influenced by the amount of bound protein, incubation time, temperature and the degree of trypsinization. Most attached cells were viable for several days and when dying they detached. Detailed morphological and cytochemical analyses of the dynamics of attachment and cytoplasmic spreading on the chemically well-defined surfaces were possible using the new method.     

Simultaneous determination of kinetic parameters for the binding of cholera toxin to immobilized sialic acid and monoclonal antibody using an array biosensor

Interactions between protein toxins and carbohydrate receptors are often semi-selective processes and the kinetic parameters that define the binding of a receptor to different toxins may vary with each interaction. In this study, we have determined the affinity constants for binding of cholera toxin (CT) to immobilized sialic acid and to anti-CT antibody (as a simultaneous reference) by measuring real-time binding processes using an array biosensor. N-Acetylneuraminic acid (Neu5Ac), a member of the sialic acid family, was covalently immobilized onto maleimide-activated planar waveguides via a thiol-terminated linker attached to the anomeric carbon of the sugar. Control antibodies were immobilized using two different approaches: covalent attachment onto maleimide-activated slides via the thiol on cysteine residues and non-covalent attachment using a biotin-NeutrAvidin linkage. Cy5-labeled CT was flowed over the immobilized receptors and the fluorescent intensity of the bound CT-receptor complex was recorded as a function of time. The association constants for CT binding to covalently attached Neu5Ac, to covalently attached anti-CT monoclonal antibody, and to antibody tethered by biotin-NeutrAvidin interactions were determined to be 1.3 x 10(8), 2.1 x 10(8) and 5.7 x 10(8)M(-1), respectively.     

Immobilization of glycoenzymes through carbohydrate side chains.

Glucoamylase, peroxidase, glucose oxidase, and carboxypeptidase Y were covalently bound to water-insoluble supports through their carbohydrate side chains. Two approaches were used. First, the carbohydrate portions of the enzymes were oxidized with periodate to generate aldehyde groups. Treatment with amines (ethylenediamine or glycyltyrosine) and borohydride provided groups through which the protein could be immobilized. Ethylenediamine was attached to glucoamylase, peroxidase, glucose oxidase, and carboxypeptidase Y to the extent of 24, 20, 30, and 15 mol/mol of enzyme, respectively. These derivatives were coupled to an aminocaproate adduct of CL-Sepharose via an N-hydroxysuccinimide ester or to CNBr-activated Sepharose. Coupling yields were in the range of 37–50%. Retained activities of the bound aminoalkyl-enzymes were 41% (glucoamylase), 79% (peroxidase), 71% (glucose oxidase), 83% (carboxypeptidase Y). A glycyltyrosine derivative of carboxypeptidase Y was bound to diazotized arylamine-glass. Coupling yield was 42% and retained esterase activity was 84%. In the second approach, the enzyme was adsorbed to immobilized concanavalin A and the complex was crosslinked. Adsorption of carboxypeptidase Y on immobilized concanavalin A followed by crosslinking with glutaraldehyde was also effective. The bound enzyme retained 96% of the native esterase activity and showed very good operational stability.     

Kinetic study of site directed and randomly immobilized his-tag alkaline phosphatase by flow injection chemiluminescence

We immobilized his-tag alkaline phosphatase (ALP) randomly and with the desirable orientation (site directed) to compare the effects of the enzyme activity on the beads. The chemiluminescence was employed to increase the sensitivity of enzyme labelled assays. Flow injection was also carried out for the detection of chemical and biological molecules in flow solutions. The Vmax of randomly immobilized his-tag ALP was 1.2 and the Vmax of site directed immobilized his-tag ALP was 1.5. In other words, the activity of site directed immobilized his-tag ALP was about 1.3-folds increased. The detection limit was detected to be 6 x 10(-6) M for the flow injection system.     

Direct detection of the binding of avidin and lactoferrin fluorescent probes to heparinized surfaces

We describe the use of two heparin-binding proteins, avidin and lactoferrin, as probes for monitoring the amount of heparin immobilized to plastic surfaces. The proteins were derivatized with either fluorescent labels or europium chelates, enabling sensitive, fast, reproducible, and robust assays, and were used to measure the amount of protein bound to heparinized microplates, with particular attention to plates that have been coated with bovine serum albumin (BSA)-heparin conjugate. This direct method unequivocally shows that BSA-heparin affords an economical, convenient, and reliable method for coating both polystyrene microtiter plates and magnetic beads with heparin. We demonstrate that assays using directly labeled proteins overcome the problems of dissociation of the heparin-protein complex, which can occur during incubation and washing steps associated with antibody-based detection methods, and the loss in binding capacity caused by certain blocking regimes. We suggest that labeled avidin and lactoferrin are convenient probes for heparinized surfaces with the potential for much wider applicability than that presented here.     

Characterization of thin layers of glucose oxidase

Glucose Oxidase (GOD) has been covalently bound to functionalized glass cover slips. The surface density of immobilized GOD molecules was measured by a method based on the amperometric determination of Flavin Adenine Dinucleotide (FAD). Atomic Force Microscopy (AFM) images, obtained in aqueous solution for the covalently bound enzyme, show a monomolecular layer of the enzyme on a functionalized glass surface. The catalytic constants were measured for the immobilized GOD and compared with those of the free enzyme.     

Protein-coated agarose surfaces for attachment of cells

Summary Plastic dishes were coated with an agarose layer. The layer was modified by covalently binding proteins to it, using the CNBr-method. Cells were seeded on the dishes and the number of attached cells was evaluated. The specificity of the attachment, was demonstrated by showing that cells, carrying specific membrane-bound immunoglobulins, attached only to the corresponding anti-immunoglobulins This indicated that the method could be used for cell sorting. The attachment of cells to proteins was influnced, by the amount of bound protein, incubation time temperature and the degree of trypsinization. Most attached cells were viable for several days and when dying they detached. Detailed morphological and cytochemical analyses of the dynamics of attachment, and cytoplasmic spreading on the chemically well-defined surfaces were possible using the new method. The work has been supported financially by the Swedish Cancer Society and the Swedish Natural Science Research Council. Box 531. Box 533.