Immunoadsorbents with synthetic oligosaccharide hapten

Oct 10, 1990 - the end of the spacer arm or indirectly via a protein conjugated to the hapten. .... Immunoadsorbents with Synthetic Blood Group Substa...
0 downloads 0 Views 781KB Size
Bioconjugate Chem. 1991, 2,32-37

32

Immunoadsorbents with Synthetic Oligosaccharide Hapten Representing Blood Group A Substances M. Abdul Mazid' and Marie Kaplant CHEMBIOMED LTD., Edmonton Research Park, 2011-94 Street, Edmonton, Alberta, Canada T6H 4N9. Received October 10, 1990

Immunoadsorbents with a synthetic oligosaccharide hapten representing human blood group A specific substances are prepared. The synthetic hapten, known as A-trisaccharide, which carries a space arm, is chemically attached to various solid supports, either directly through a suitable functional group at the end of the spacer arm or indirectly via a protein conjugated to the hapten. The preparation involves simple and mild procedures for the activation and/or derivatization of the supports. The latter includes naturally occurring polyhydroxy materials such as agarose, cellulose, or cellulose derivatives, and other particulate materials such as inorganic diatomites and a synthetic organic copolymer. The methods used for the coupling concern specifically the preparation of controlled-capacity and high-efficiency immunoadsorbents, with limited incorporations, which may be prepared easily and used for the selective removal, or affinity chromatographic separation, of specific antibodies from plasma environment or blood. It has been found that while hapten incorporation to the support may be varied rather easily, the physical nature of the support as well as the form of the hapten is important in determining the efficiency of an immunoadsorbent.

INTRODUCTION Selective removal of specific substances from blood has been described variously since therapeutic intervention with an immunoadsorbent was first reported in the mid1970s. Considerable attention has been devoted to the development of adsorbents for the removal of unwanted plasma protein components, especially for the treatment of autoimmune disorders. Several experimental in vivo systems for the elimination of specific antigens or antibodies have been reported. In one example, blood flows through a tube coated with entrapped immunoadsorbents (I). In another, antigen is bound to a derivatized nylon tube which is used to remove specific antibody during blood circulation (2). Yet, in a different system, antigen is immobilized on collodion adherent to activated charcoal particles and placed into a column to remove antibodies from dogs by perfusion of plasma through the column (3). The feasibility of the foregoing approach has been demonstrated for the removal of anti-A and anti-B antibodies using formalin-fixed erythrocytes as immunoadsorbents (4). Subsequently, synthetic trisaccharides representing blood group A and B substances chemically coupled to silica particles have been used in a number of studies (5-16) for the selective removal of A and B isoagglutinins directed against the corresponding blood groups. Some of these studies have been applied in humans and it has been possible to successfully avoid the effects of AB0 incompatibility in bone marrow and renal transplants. Another therapeutic approach has been the use of protein A bound to charcoal-collodion for the treatment of solid tumors (17 ) and heat-killed Staphylococcus aureus sorbent to remove serum IgG in autoimmune hemolytic anemia syndrome (18). Recently, Behm et al. (19) reviewed selective and specific adsorbents for medical therapy, while Margel and Marcus (20) described the

* Author to whom

all correspondence should be addressed. Present address: Department of Medical Microbiology, University of Alberta, Edmonton, Alberta, Canada. f

1043-1802/91/2902-0032$02.50/0

potential uses of a specific sorbent system consisting of cross-linked agarose-polyacrolein microsphere beads (so called agarose acrobeads) for specific hemoperfusion in various systems. Affinity adsorbents using agarose as a carrier have been prepared mostly by two methods: the cyanogen bromide (CNBr) method and the epoxy method. Immunoadsorbents with blood group substance A, in particular, coupled to cross-linked agarose by the CNBr method has been described by Kristiansen (21). The adsorbent prepared by this method is reported to have a number of disadvantages, including low ligand concentration, instability, nonspecificadsorption, and excess chemicalreactivity (22). With the epoxy method, nonderivatized sugars have been used directly to obtain a stable linkage, a higher degree of substitution, and an adsorbent with less nonspecific adsorption (22-25). However, this method has some disadvantages such as the requirement of high pH and a large excessof ligand during the coupling step. Therefore, some epimerization and degradation of sugars are considered inevitable in this step (24, 26). To avoid these severe conditions for coupling, the conversion of epoxy groups on agarose gel into more reactive groups has been reported (27). Matsumoto et al. (22) described the derivatization of amino and carboxyl agarose, prepared from epoxyactivated agarose by simple procedures, with a variety of sugars under mild conditions. Two types of affinity adsorbents for the purification of lectins were prepared. These include coupling of oligosaccharide ligands (lactose and maltose) by reductive amination with cyanoborohydride to amino-Sepharose 4B prepared by amination of epoxyactivated Sepharose 4B. The second type of affinity adsorbent involves coupling of hexosamines (GalN and GlcN) by the aid of carbodiimide to carboxy-Sepharose 4B prepared by succinylation of amino-Sepharose4B. Both types of adsorbents have high ligand concentrations and chemically stable linkage between ligand and Sepharose 4B; however, the coupling of oligosaccharides to amino@ 1991 American Chemical Society

Immunoadsorbents with Synthetic Blood Group Substances

Sepharose 4B by reductive amination is very slow, while the coupling of amino sugars with carboxy-Sepharose-4B via succinylation also requires a large excess of ligand during the coupling step and does not necessarily produce a good adsorbent with high capacity. This fact alone is a major drawback to the use of synthetic oligosaccharides such as the blood group determinants for the removal of specific antibodies by extracorporeal immunoadsorption. It is notable that most immunoadsorbents so far have been reported with arbitrary amounts of ligands contained therein or coupled thereto, possibly limited by the efficiency or conditions of the coupling method used and, perhaps, without due consideration to the utilization of optimum capacity of the immunoadsorbents. We have recently reported on the flow-rate dependence of in vitro removal of anti-A and anti-B antibodies by immunoadsorbents with synthetic oligosaccharides representing blood group substances (28). However,a systematic study of the relationship between ligand density and specific binding or selective removal capacity of an immunoadsorbent during therapeutic intervention is not possible, nor is it easy to evaluate the efficacy of different immunoadsorbents using various solid supports even with the same ligand. This paper attempts to describe a comparative study of a number of immunoadsorbents under laboratory conditions. The immunoadsorbents are prepared by covalently linking chemically synthesized A-trisaccharide hapten with a spacer arm (abbreviated together as ATS), representing human blood group A substances, to a variety of commercially available support materials, their capacities being evaluated by in vitro hemagglutination techniques following batch absorptions. In particular, the methods used for the chemical coupling of the synthetic hapten concern its limited incorporation to the support material and relate specifically to the preparation of controlled-capacity and efficient immunoadsorbents for the selective removal of anti-A antibodies from human plasma. EXPERIMENTAL PROCEDURES

Materials and Methods. General. Immunoadsorbents with chemically synthesized human blood group A-trisaccharide, an 8-(azidocarbonyl)octylderivative of cuGalNAcl,3[c~Fucl,2]/3Gal or a similar derivative of the trisaccharide but with another functional group at the end of the spacer arm, covalently attached to natural polyhydroxy supports and other particulate materials, including inorganic silica and a synthetic copolymer,are prepared by published methods. The natural polyhydroxy supports are of the types of agarose and cellulose, which include commercially available Sepharose 4B (Pharmacia), (aminoethyl)cellulose, (carboxymethy1)cellulose(AE-and CMcellulose, Sigma), beaded Excellulose (Pierce), and ordinary cotton. The synthetic support material used is the aminomethylated styrene/divinylbenzene (DVB) copolymer, available under the trade name of Lewatit (Bayer), while the inorganic silica particle is of the diatomaceous type Chromosorb P (30/60 mesh) obtained from Manville Corp. The synthetic blood group substance, i.e. the A-trisaccharide with the linking spacer arm, and the trisaccharidederivatized silica particles, known as Synsorb A, are the products of Chembiomed Ltd. (Edmonton, Canada). The latter is surface-modified with a synthetic polymeric membrane-type coating by a proprietary technique (29) to provide a better surface resistance and therefore adequate physical and mechanical stability to the immunoadsorbent particles (otherwise known as Biosynsorb A).

Bioconjugate Chem., Vol. 2, No. 1, 1991 33

This coating prevents the release of potentially harmful fines during clinical application, while the original biological activity and efficacy is maintained. The derivatization of silica particles follows the introduction of reactive amino groups using (3-aminopropy1)triethoxysilane under standard conditions (30) as described elsewhere (29,31,32). The silane-treated diatomite itself containing approximately 5-10 pmol of NH2/g, also coated by the above mentioned proprietary technique, are additionally coupled to the synthetic trisaccharide indirectly via a conjugated protein such as bovine serum albumin (BSA) in various amounts. The aminoethyl and carboxymethyl derivatives of cellulose and aminomethylated styrene/divinylbenzene supports are used without further derivatization. In general, the coupling procedure involves treatment of the carboxyl derivative of the support or the synthetic ligand with di-N-succinimidyl carbonate (DSC), followed by reaction with an amino-terminated ligand or support, respectively. The coupling of amino groups is performed via activation with glutaraldehyde and that of amino groups and carboxyl groups between the support and the ligand are achieved in the presence of a water-soluble carbodiimide, while underivatized polyhydroxy supports are utilized after activation with divinyl sulfone or epoxy reagents. Immobilization of A-Trisaccharide. Di-N-succinimidyl Carbonate Activation. Di-N-succinimidyl carbonate is dissolved in 0.2-0.5 mL of dimethylformamide (DMF) to which is added a molar equivalent of carboxylterminated A-trisaccharide (ATS-acid). Alternatively, the solution of DSC in DMF is added to a suspension of the carboxyl-derivatized support CM-cellulose (thoroughly washed and preswelled) in DMF. The activation of the ligand or the support is carried out for 2-3 h, followed by addition of the amino-derivatized support AE-cellulose or the ligand ethylenediamine (EDA) terminated A-trisaccharide, ATS-EDA,or ATS-BSA conjugate. The coupling reaction is continued overnight (12-16 h) at room temperature with gentle agitation, and then the support is washed thoroughly with water, the washings being saved for the estimation of the trisaccharide and therefore its incorporation to the support. Finally, the haptenated support is treated with a solution of 10% glycine in 0.5 M Na2C03 to block any residual activated carboxyl group on the support. 1-Ethyl-3-[3’-(dimethylamino)propyl]carbodiimide Activation. The covalent attachment of amino group and carboxyl group between the support and the ligand is performed in the presence of a water-soluble carbodiimide such as 1-ethyl-3-[3’- (dimethylamino)propyllcarbodiimide hydrochloride (EDAC). About 10 mg of the reagent is dissolved initially in a buffer under acidic conditions, 0.1 M phosphate at pH between 4.0 and 5.0, and the carboxyl group containing ligand ATS-acid or the support CM-celluloseis mixed to it. The mixture is rotated for 30 min at room temperature, and then the support AE-cellulose or the ligand containing an amino group is added. The coupling reaction is continued for 3-4 h at room temperature after which the haptenated support is washed thoroughly with water. Glutaraldehyde Activation. The coupling reaction between amino groups in supports, including the silanetreated and surface-modified diatomite, and ligands via activation with glutaraldehyde is carried out under the following conditions. The support is suspended in a buffer, 0.1 M borate or phosphate at a pH in the range of 7-9, containing 1-10 % glutaraldehyde, which is precooled to

34

Bioconjugate Chem., Vol. 2, No. 1, 1991

a temperature below 10 "C. The activation of the support is carried out a t this temperature for 1h withgentle stirring, and then washed extensively with water or buffer until a negative Schiff s test is obtained. The washed support is kept suspended in the same buffer, pH -7, to which the amino-terminated ligand ATS-EDA or ATS-BSA conjugate is added. The coupling reaction is carried out initially a t room temperature for 1h and then continued overnight at temperatures between 0 and 4 "C. The haptenated support is washed thoroughly with cold water, followed by washing with phosphate buffer containing 1 M NaCl in the case of coupling with ATS-BSA conjugate, and finally suspended in water or the buffer after washing off the NaCl. The washings are saved for the estimation of trisaccharide by the phenol-sulfuric acid assay (33)or protein by the standard Bio-Rad procedures (34). Epoxy Activation. The hydroxyl groups occurring in agarose and cellulose are utilized for chemical coupling of the A-trisaccharide, following initial activation and functionalization by published methods (22,26). These involve the formation of epoxy derivatives with epichlorohydrin (ECH)or the long-chain bisoxirane 1,4-n-butanediol diglycidyl ether (BDDE) under basic conditions, e.g. 5% (v/v) in 0.4 N NaOH solution containing 20% dioxane, to provide for a covalently linked epoxide. The epoxide group is then opened with 6 M ammonia or 5 7% ethylenediamine to provide for terminal primary amines. This, in turn, is utilized for coupling of carboxyl-terminated hapten using methods previously described, involving di-N-succinimidyl carbonate activated ATS-acid in DMF. After washing with methanol, the remaining amine groups are blocked by acetylation in 5% acetic anhydride in methanol. Divinyl Sulfone Activation. The activation of hydroxyl group containing natural polymers (agarose and cellulose) is performed with divinyl sulfone (DVS) in alkaline solutions (35). The reaction is carried out for 3 h at room temperature or 1 h a t 40 "C using 0.1-0.5 mL of the reagent in a suspension of 5-10 g suction-dried support in 10-20 mL solution of 0.5 M Na2C03. The amino-terminated ATS-EDA or ATS-BSA conjugate is coupled by mixing overnight at room temperature, also in 0.5 M Na2C03, but after washing the activated support free of divinyl sulfone. Any unreacted vinyl group is blocked by reaction with a solution of glycine (10% w/v) in 0.5 M NaZC03 or 1.5 M 2-mercaptoethanol or sodium thiosulfate (35). Determination of Biological Activity of Immunoadsorbents. The immunoadsorbents prepared by various methods are tested for their biological activity by in vitro hemagglutination technique. Generally, 25 mg of the immunoadsorbent material (suction-dried) is incubated with 0.5 mL of 0-plasma or 0-serum by rotating in a hematology mixer for 1 h at room temperature (21 OC). The supernatant is then removed, and antibody titers of IgM and IgG are determined by saline agglutination and indirect antiglobulin (Coombs) tests, respectively, at 37 "C from serial dilutions with human AlRBC (red blood cells) by using standard procedures (36). The titers are expressed as the reciprocal of the highest dilution that produces macroscopically visible agglutination. It is to be noted that some of the antibody titers (indicated with a footnote in tables of results) have been obtained with 0-serum/ plasma of which preadsorption titers are lower by a factor of 2 or 4 (one or two reciprocal dilutions) than the most commonly observed preadsorption titers shown in the top of a table. These titers are therefore multiplied by the corresponding factor for the purpose of a more direct comparison.

Mazid and Kaplan

Table I. Biological Activity of Immunoadsorbents Based on Agarose Couded with A-Trisaccharide hapten incorporated, pmol/g of human A1 titers material tested suction-dried sumort IaM IeG ._ 0-serum 32 256 agarose control, 0 32 128 ECH/EDA treated and acetylated agarose- ATS-acid 0.12 8 32 0.36 8 16 8 0.68 4 1.26 2 4 1 4 3.02 agarose-(ATS-BSA) 0.46 4 32O I

I

I

Titer multiplied by four to conform with that of the starting serum. (I

Table 11. Biological Activity of Immunoadsorbents Based on Cellulosic Supports Coupled with A-Trisaccharide hapten incorporated,

0-plasma cellulose fiber, untreated control cellulose-ATS-acid cellulose-(ATS-BSA) beaded cellulose, untreated control beaded cellulose-(ATS-EDA) beaded cellulose-ATS-acid beaded cellulose-(ATS-BSA)

0

1.84 0.14 0

0.46 0.51 0.07

64 64

128 64

32 64 32

64 128 64

32 32 16

64 64 32O

Titer multiplied by two to conform with that of the starting plasma. Table 111. Biological Activity of Immunoadsorbents Based on Derivatized Cellulosic Supports with A-Trisaccharide hapten incorDorated.

0-plasma AE-cellulose, washed control AE-cellulose-ATS-acid AE-cellulose-(ATS-BS A) CM-cellulose, washed control CM-cellulose-(ATS-BSA)

0

1.67 0.19 0

0.31

64 64 8 2 64 8

128 128 32 4 64 16

RESULTS AND DISCUSSION

The biological activity of various immunoadsorbents as well as the amount of hapten incorporated therein are shown in Tables I-V. Included in Tables I and V are also the results for agarose- and diatomite-based immunoadsorbents a t different levels of haptenation with ATS-acid and ATS-BSA conjugates, respectively. Table I shows that the efficiency of antibody removal with the agarose-based immunoadsorbents increases significantly as the amount of hapten incorporation is increased. This would suggest that a certain amount of hapten may be attached to a smaller quantity of agarose gel and yet obtain the same level of efficiency. There is, however, no significant advantage in coupling an antigen such as ATS-BSA conjugate to agarose-type supports since the efficiencies of these immunoadsorbenta with equivalent amounts of A-trisaccharide hapten are about the same. This may be accounted for by the fact that the porosity of the agarose beads (exclusion limit of 20 X 106 Da) is sufficient to allow equally efficient interaction of antibodies with haptens coupled to the support either directly or

Bioconjugate Chem., Vol. 2, No. 1, 1991

Immunoadsorbents with Synthetic Blood Group Substances

Table IV. Biological Activity of Immunoadsorbents Based on Synthetic Styrene/DVB Copolymers Coupled with A-Trisaccharide hapten incorporated,

0-plasma aminomethylated (AM) styrene/ DVB control AM-styrene/DVB, acetylated AM-styrene/DVB-ATS-acid

AM-styrene/DVB-ATS-acid,

0 0

1.13 0.38

64 32'

128 128"

32 64O 64O

32 128O 128O

acetylated

AM-styrene/DVB-(ATS-BSA)

0.13 84 16" AM-styrene/DVB-BSA (0.44 0 64 128 mg/d a Titers multiplied by a factor of 4 to conform with that of the starting plasma. Table V. Biological Activity of Immunoadsorbents Based on Inorganic Diatomaceous Supports (30/60 Mesh) Coupled with A-Trisaccharides hapten incorporated,

0-plasma 0.5'Y polystyrene-coated

diatomite precoated diatomite coupled with ATS-BSA after glutaraldehyde activation

precoated diatomite coupled with ATS-BSA in the presence of glutaraldehyde uncoated diatomite-ATS uncoated diatomite-(ATS-BSA)

128 64

256

0.05

32

64

0.10 0.20 0.30 0.40

16 16 16 16 16

32 32 32 32 64

0.10 0.20

8

0.7

16

0.6

8ll

32 16 32 8"

0

0.05

8

128

Titer adjusted by a factor of 2 to conform with that of the starting plasma. (I

with conjugated ATS-BSA. It is also possible that the small size of agarose beads (50-140 pm) provides a relatively larger external surface area which is readily available for hapten incorporation and, therefore, more amenable or predisposed to interaction with neutralizing antibodies present in the plasma. In contrast to agarose-based immunoadsorbents, chemical coupling of haptens to other polyhydroxy supports such as the cellulosic materials in various forms show different results. These supports include fibrous and beaded cellulose as well as cellulose functionalized with aminoethyl and carboxymethyl groups (Tables I1 and 111). Haptenation of cellulose fibers with EDA-terminated A-trisaccharide via activation with the long-chain bisoxirane 1,4-n-butanediol diglycidyl ether appeared unsuccessful (data not shown), while ATS-BSA incorporation yields an equivalent of 0.14 pmol of ATS/g of suctiondried support (1g dry weight corresponds to 2 g of suctiondried support). Also, incorporation of DSC-treated ATSacid t o ECH-activated cotton fiber gives a n immunoadsorbent with 1.84 pmol of ATS/g of suction-dried support. However, none of these immunoadsorbents shows any considerable biological activity (see Table 11)despite the fact that reasonable levels of hapten incorporation is achieved. Thus, it would appear that small oligosaccha-

35

ride haptens attached to cellulose are not accessible to the antibodies since cellulose in its fibrous form is relatively nonporous. Alternatively, haptenation of beaded cellulose was achieved both with ATS-EDA (0.46 pmol/g of suctiondried support) via DVS activation and also with DSCtreated ATS-acid (0.51 pmol/g of suction-dried support) via ECH activation. Again, there is little biological activity as noted in Table 11, suggesting that the physical nature of the support is not appropriate or suitable for obtaining an efficient immunoadsorbent. On the other hand, haptenation of beaded cellulose with ATS-BSA conjugate via ECH activation (0.07 pmol of ATS/g of suction-dried support) shows some biological activity, which is consistent with the specified exclusion limit of 5000 Da for this support. However, the overall performance of all the beaded cellulosic supports is generally poor, indicating that the porosity of the beads is not sufficient to permit interaction of antibodies with haptens which may be attached to the interior of the support, similar to the fibrous cellulose. The effect of the nature of immunoadsorbents on biological activity is more pronounced with derivatized cellulosic supports (see Table 111). For instance, haptenation with ATS-BSA was achieved both for AE-cellulose via glutaraldehyde activation (0.19 pmol of ATS/g of suction-dried support) and for CM-cellulose via EDAC activation (0.31 pmol of ATS/g of suction-dried support) or via DSC activation (0.11 pmol of ATS/g of suctiondried support). In terms of biological activity, however, both CM-cellulose supports coupled with ATS-BSA conjugate appear to be equally efficient as the AE-cellulose coupled with DSC-treated ATS-acid which contains about 6-15 times more hapten (1.67 pmol of ATS/g of suctiondried support), while the best efficiency is observed with AE-cellulose-(ATS-BSA) containing only 0.19 pmol of ATS/g of suction-dried support. This implies that the physical nature of the support, including its morphology, as well as the form of the hapten is important in determining the efficiency of an immunoadsorbent. It would seem that haptenation of an appropriate support with an antigen such as ATS-BSA conjugate is a logical approach to the preparation of an efficient immunoadsorbent. This fact is also corroborated by the results obtained with immunoadsorbentsbased on styrene/ divinylbenzenecopolymer coupled with the hapten. Table IV shows that aminomethylated styrene/divinylbenzene copolymer haptenated with ATS-BSA conjugate (0.13 pmol of ATS/g) via glutaraldehyde activation has reasonablygood biological activity, while the untreated control support or the one haptenated with a considerable amount of ATS-acid (0.38 pmol/g) has practically no activity. It is also notable that increasing the level of haptenation does not necessarily yield an improved immunoadsorbent, as evident from the results of another styrene/divinylbenzene support coupled with ATS-acidvia DSC activation which yields an incorporation of 1.13 pmol/g. Tests were carried out to determine if acetylation of aminomethylated styrene/DVB copolymer itself or blocking of residual amino groups after haptenation and BSA incorporation alone would show any difference in activity. Incorporation of BSA was done chemically via glutaraldehyde activation of unacetylated support. The results of biological activity assays for styrene/DVB-based immunoadsorbents are shown in Table IV, which includes an untreated support as well. It would appear that there is very little or no effect of the various treatments, in general, while acetylation of the supports, haptenated or

36 Bioconlugate Chem., Vol. 2, No. 1, 1991

not, enhances the reduction of IgG titers to some extent. This may be explained by increased nonspecific interactions due to blocking of amino groups which would increase the hydrophobicity of the support. However, considering the level of hapten incorporation, this effect is not an important factor in determining the overall efficiency of the immunoadsorbent. It may be noted that the use of different blocking reagents following immobilization of ATS via DVS activation would not influence the immunoadsorbent activity considerably since the blocking involves similar inactivation of otherwise same reactive group on the support. Table V shows results for immunoadsorbents prepared with aminosilated diatomite (30/60mesh) which was precoated with a polymer, as described elsewhere (29). Haptenation was done by coupling of various amounts of ATSBSA conjugate via glutaraldehyde activation. Included in the table are also results for an immunoadsorbent prepared by the same procedure but using uncoated diatomite and for one that is haptenated with ATS-acyl azide. The results indicate that a saturating level of efficiency could be reached with a hapten incorporation equivalent to 0.1 pmol of ATS/g of immunoadsorbent, and that haptenation of the coated support in presence of glutaraldehyde (in situ coupling) shows relatively better activity than the support preactivated with glutaraldehyde. In both cases, however, the maximum activity appears to be observed with immunoadsorbents containing only a fraction of the hapten compared to the uncoated supports haptenated either with ATS-acyl azide or with ATS-BSA conjugate, all of which show more or less similar activities. Further work along this line would seem necessary to confirm this finding. The improved efficiency with precoated immunoadsorbents may be accounted for by the fact that the activation of the supports with glutaraldehyde, and therefore chemical coupling of the ATS-BSA conjugate thereto, would occur rather preferentially on the coated surface and also involve simultaneous intramolecular cross-linking of the protein on the surface of the coated support. This implies that the protein portion of the antigen may be unfolded and hence the haptens are more completely or better exposed to yield maximum biological activity. The significance of these results is that physically stable immunoadsorbents with improved efficiency may be prepared by coating the support first with a suitable polymeric membrane-type film (29)followed by chemical attachment or immobilization of an antigenic determinant such as a hapten-protein conjugate by an appropriate method. ACKNOWLEDGMENT

We wish to thank Dr. R. M. Ippolito for valuable discussion and for his interest in the work. LITERATURE CITED (1) Schenkein, I., Bystryn, J.-C., and Uhr, J. W. (1971) Specific removal of in vivo antibody by extracorporeal circulation over an immunoadsorbent in gel. J . Clin. Invest. 50, 1864-1868. (2) Lyle, L. R., Parker, B. M., and Parker, C. W. (1974) The use of protein-substituted nylon catheters for selective immunoadsorption in vivo. J. Immunol. 113, 517-521. (3) Terman, D. S., Tavel, T., Petty, D., Racic, M. R., and Buffaloe, G. (1977) Specific removal of antibody by extracorporeal circulation over antigen immobilized in collodion-charcoal. Clin. Exp. Zmmunol. 28, 180-188. (4) Buskard, N. A., Williams, B., Terman,D. S., Buckner, C. D., Clift, R. A., Gray, M., and Thomas, E. D. (1978) Antibody

Mazid and Kaplan

adsorption on formalinized red cell ghosts: A potential method for the in vivo removal of isohemagglutinins. Blood 52 (suppl. I), 296. (5) Bensinger, W. I. (1981) Plasma exchange and immunoadsorption for removal of antibodies prior to ABO-incompatible bone marrow transplantation. Znt. Soc. Artif.Organs 51,254258. (6) Bensinger, W. I., Baker, D. A., Buckner, C. D., Clift, R. A., and Thomas, E. D. (1981) Immunoadsorption for removal of A and B blood group antibodies. N . Engl. J. Med. 304, 160162. (7) Bensinger, W. I., Baker, D. A., Buckner, C. D., Clift, R. A., and Thomas, E. D. (1981) In vitroand in vivo removal of anti-A erythrocyte antibody by adsorption to a synthetic immunoadsorbent. Transfusion 21, 335-342. (8) Bensinger, W. I., Buckner, C. D., Baker, D. A., Clift, R. A., and Thomas, E. D. (1982) Removal of specific antibody in vivo by whole blood immunoadsorption: Preliminary results in dogs. J . Clin. Apheresis I , 2-5. (9) Bensinger, W. I., Buckner, C. D., Williams, B., and Clift, R. A. (1982) Immune adsorption of anti-A and anti-B antibodies. In Cytapheresis and Plasma Exchange: Clinical Indications (W. R. Volger, Ed.) pp 295-300, Alan R. Liss, Inc., New York. (10) Watanabe, K., Kobayaski, N., Nakamura, K., Kate, S.,Hosi, N., Hisazumi, T., and Hida, M. (1983) An example of ABOincompatible bone marrow transplant by use of immunoadsorbent. Jpn. Clin. Haematol. Conf. Abstr. 219. (11) Bensinger, W. I. (1984) Selective removal of A and B isoagglutinins. In Selective Plasma Component Removal (A. A. Pineda, Ed.) pp 43-70, Futura Publishing Co., Mount Kisco, NY. (12) Bensinger, W. I., Buckner, C. D., and Clift, R. A. (1985) Whole blood immunoadsorption of anti-A or anti-B antibodies. Vox Sang 48, 357-361. (13) Bussel, A., Schenmetzler, C., Gluckman, E., and Reviron, J. (1985) Removal of recipient allo-agglutinins in ABOincompatible bone marrow transplantation: Comparative study between plasma exchange and immunoadsorption. Plasma Ther. Transfus. Technol. 6 , 461-464. (14) Osterwalder, B., Gratwohl, A., Nissen, C., and Speck, B. (1986) Immunoadsorption for removal of anti-A and anti-B blood group antibodies in ABO-incompatible bone marrow transplantation. Blut 53, 379-390. (15) Raja, R., Bannett, A., Caruana, R., and Baquero, A. (1986) Removal of antibodies with immunoadsorption from an ABOincompatible recipient prior to renal transplant. Trans. Am. SOC.Artif. Intern. Organs 32, 102-103. (16) Bannett, A. D., Bensinger, W. I., Raja, R., Baquero, A., and McAlack, R. F. (1987) Immunoadsorption and renal transplant in two patients with a major ABO-incompatibility. Transplantation 43, 909-910. (17) Terman, D. S., Young, J. B., Shearer, W. T., Ayus, C., Lehane, D., Mattioli, C., Espada, R., Howell, J. F., Yamamoto, T., Zaleski, H. I., Miller, L., Frommer, P., Feldman, L., Henry, J. F., Tillquist, R., Cook, G., and Daskal, Y. (1981) Preliminary observations of the effects on breast adenocarcinoma of plasma perfusion over immobilized protein A. N . Engl. J . Med. 305, 1195-1200. (18) Besa, E. C., Ray, P. K., Swami, V. K., Idiculla, A., Rhoades, J. E., Jr., Bassett, J. G., Joseph, R. R., and Cooper, D. R. (1981) Specific immunoadsorption of IgG antibody in a patient with chronic lymphocytic leukemia and autoimmune hemolytic anemia: A new form of therapy for the acute critical stage. Am. J . Med. 71, 1035-1040. (19) Behm, E., Ivanovich, P., and Klinkmann, H. (1989)Selective and specific adsorbents for medical therapy. Int. J . Artif. Organs 12, 1-10, (20) Margel, S., and Marcus, L. (1986) Specific Hemoperfusion through agarose acrobeads. Appl. Biochem. Biotechnol. 12, 37-66. (21) Kristiansen, T. (1974) Studies on blood group substances. V. Blood group substance A coupled to agarose as an immunosorbent. Biochim. Biophys. Acta 263, 567-574. (22) Matsumoto, I., Kitagaki, H., Akai, Y., Ito, Y., and Seno, N. (1981) Derivatization of epoxy-activated agarose with various carbohydrates for the preparation of stable and high-capacity

Immunoadsorbents wkh Synthetic Blood Group Substances

affinity adeorbents: Their use for affinity chromatography of carbohydrate-bindingproteins.Anal. Biochem. 116,103-110. (23) Vretblad, P. (1979) Purification of lectine by biospecific affinity chromatography. Biochim. Biophys. Acta 434,169176. (24) Uy, R., and Wold, F. (1977) 1,4-Butanediol diglycidyl ether coupling of carbohydrates to Sepharose: Affinity adsorbents for lectins and glycosidases. Anal. Biochem. 81, 98-107. (25) Gurtler, L. G. (1978) The fucosyl specific lectins of Ulex europaeus and Sarothamnus scoparius: Biochemical characterization and binding properties to human B-lymphocytes. Biochim. Biophys. Acta 544, 593-604. (26) Matsumoto, I., Mizuno, Y., and Seno, N. (1979) Activation of Sepharose with epichlorohydrin and subsequent immobilization of ligand for affinity adsorbent. J . Biochem. 85,10911098. (27) Matsumoto, I., Seno, N., Golovtchenko-Matsumoto, A. M., and Osawa, T. (1980) J . Biochem. 87, 535-540. (28) Kaplan, M., and Mazid, M. A. (1989) Flow rate dependence of in vitro removal of anti-A and anti-B antibodies by immunoadsorbents with synthetic oligosaccharidesrepresenting blood group substances. Znt. J . Artif. Organs 12, 799-804.

Bioconjugate Chem., Vol. 2, No. 1, 1991 37

(29) Mazid, M. A. (1990) Improved affinity supports for hemoperfusion. European Pat. Appln. E P 0 371 636 A2. (30) Weetall, H. H., and Filbert, A. M. (1974) Porous glass for affinity chromatography applications. Methods Enzymol. 34B, 59-72. (31) Lemieux, R. U., Bundle, D. R., and Baker, D. A. (1979) Glycoside-ether-ester compounds. US.Patent 4 137 401. (32) Lemieux, R. U., Bundle, D. R., and Baker, D. A. (1980) Artificial oligosaccharideantigenic determinants. U.S.Patent 4 238 473. (33) Dubois, M., Gilles, K., Hamilton, J. K., Rebere, P. A., and Smith, F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356. (34) Bradford, M. M. (1979) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72,248254. (35) Porath, J., La&,T., and Janson, J.C. (1975)Agar derivatives for chromatography, electrophoresis and gel-bound enzymes. 111. Rigid agarose gels cross-linkedwith divinyl sulfone (DVS). J. Chromatogr. 103, 49-62. (36) Issitt, P. D. (1970) Applied Blood Group Serology, Chapter 4, Spectra Biologicals, Oxnard, CA.