CA1144858A

CA1144858A - Blood samples or stroma bonded through a diazoamino bond to a polymer

Info

Publication number: CA1144858A
Application number: CA000352730A
Authority: CA
Inventors: Albert Venturini
Original assignee: Ortho Diagnostic Systems Inc
Current assignee: Ortho Clinical Diagnostics Inc
Priority date: 1979-06-13
Filing date: 1980-05-27
Publication date: 1983-04-19
Also published as: DE3022278A1

Abstract

ABSTRACT
PURIFICATION OF BLOOD GROUP ANTIBODIES
A method for covalently bonding red blood cells or red blood cell membranes (stroma) to a polymer containing a benzenoid nucleus via a diazoamino bridge and the product of said method. There are further disclosed methods of employing these products for the detection or removal of antibodies and the like from biological fluids. The preferred embodiments are a polystyrene/stromata con-jugate useful for removal or detection of blood group antibodies in serum and methods using this conjugate.

Description

-~ ORD 22 ~4~S~

PURIFICATION OF BLOOD GROUP ANTIBODIES

BACKGROUND OF THE INVENTION
The present invention relates to attachment of red blood cells or red blood cell membranes (stroma) to polymers containing a benzenoid nucleus and more particularly the attachment of said materials to polystyrene. The invention further relates to the compositions so pro-duced and to methods of employing these compositions for removing or detecting antibodies and the like in biological fluids.
DISCUSSION OF THE PRIOR ART
The typing of blood, whether to determine suitability for transfusion or for the diagnosis or prevention of disease (e~g., hemolytic disease of the newborn) is an important part of modern medicine. The purpose of the procedure is to determine the presence of antigens (so-called blood group antigens) on red blood cell surfaces and to determine the presence of antibodies to certain of these substances in the patient's serum.
Because of the danger of death or serious injury from the transusing of incompatible blood, blood typing is routinely done on all donated blood. Typing of the blood of prospective parents is also routinely done to determine the likelihood of the fetus being killed or injured by a blood incompatibility with the mother.
~.

., .

S~3 In this typing process, the presence of blood group antigens (e.g., A, B, or Rh) is determined by reacting a sample of the patient's blood with a purified anti-body to the antigen in question. If the addition of S this specific antibody causes either agglutination or cell lysis, the test is considered positive for the antigen in question.

It can thus be seen that blood typing requires a con~
siderable ~uantity of specific antibody to each of the blood group antigens to be detected. Currently, these specific antibodies are produced from human serum con-taining a mi~ture of antibodies by reacting the serum ~i~h purified human red blood cells having antigens to the unwanted antibodies. The unwanted antibodies will react with the antigens on the red blood cells and can thus be removed from the solution by e.g., filtration or centrifugation.

Thus, to prepare a serum containing anti-D antibody according to the prior art technique, one would react a serum containing the desired antibody ~anti-~) in admixture with an undesired antibody (e.g., anti-A) with group A red cells to remove the undesired antibody.
The red blood cells having the undesired anti~A anti-body combined with them must be discarded and cannot be reused. A method which could remove these unwanted antibodies from serum without consuming human red blood cells would be a significant technical advance.
The attachment of a biological material to a solid substrate while preserving its activity has been re-ported previously. Examples of prior art teachings in this regard are (e.g.) United States Patents Nos.
3,788,948; 3,555,143; 3,806,417; 4,059,685; and 4,046,723.
Campbell, _ al., Proc. Natl. Acad. Sci., U.S. 37, 575 (1951), teach the preparation of a diazoben~yl cellulose derivative and the coupling of bovine serum albumin to this solid substrate. The review article "Water-Insoluble ' S~

Derivatives of Enz~mes, Antigens, and Antibodies" by I.H. Silman and E. Katchalski in Annual Review of Bio-chemistry, Volume 35 (II), 873 (1966), P.D. ~oyer, et al., editors, provides an excellent summary of work in this area up to the mid-1960's.

Several workers have employed red blood cell membranes (stroma) in various experiments.

Landsteiner and Shear, J. Exptl. Med., 63, 325 (1936), disclose the use of stroma as an insoluble substrate for attachment of haptens.

Arquilla and Coblence, Anat. Record, 138, 203 (1960), disclose the use of a mixture of stroma and cellulose to adsorb insulin, and the use of this adsorbed conjugate to isolate rabbit insulin antibodies.

Arquilla and Finn, J. Exptl. Med., 118, 55 (1963), teach the use of stroma as a solid substrate for supporting insulin and rendering it insoluble.

Henry C. Isliker, Ann. N.Y. Academ. of Sci., 57, (2):
226-232 (November 1953), teaches the attachment of anti-gens ~including stroma) to ion exchange resins andsuggests the use of these materials for removal of antibodies from (e.g.) human plasma. The use of a sulfonated aromatic amine as an intermediate link is also suggested. Nevertheless, this method suffers from several disadvantages and has not been put into practical use in the twenty-five years since it was published. Specifically, the paper discloses that a good deal of non-specific adsorption of antibodies is occurring and that it was generally not possible to elute more than about 55% of the adsorbed antibodies under mild conditions. Moreover, the regenerated resin had diminished adsorption capacity as compared with the S~

unused material. A separated Rh antigen was combined with resin and used for adsorption of complete and incomplete antibodies. No evaluation of the purity or specificity of any serum treated with a resin-supported antigen was disclosed.

The preparation of diazonium salts of polystyrene and other polymers containing a benzenoid nucleus is well-known in the art, being disclosed, for e~ample, in a United States Patent No. 2, 274, 551.

It has now been discovered that stroma may be covalently bonded through a diazo amino bridge to polystyrene diazonium salt or the diazonium salt of other polymers containing a benzenoid nucleus. This resulting conjugate may be used and reused to prepare highly purified blood group antibodies in high yield.

UMM~RS? OF THE INVENTION
It is accordingly one object of the present invention to provide a method for covalently bonding red blood cells or stroma via a diazoamino bridge to a polymer containing a benzenoid nucleus.

25 It is a further object to provide a method for covalently binding stroma to polystyrene.

It is a still further object to provide the compounds prepared by this method.
It is a still further object ~o provide a method for specifically removing or detecting an antibody in a mixture (e.g., serum) to an antigen on the surface of the bound red blood cells or stroma.
Further objects will become apparent on a reading of the following disclosure.

~. -.
, ~ ORD 22 ..
5The above objects are achieved by the subject process which comprises the following steps. First, a polymer containing a benzenoid nucleus i5 nitrated on the benze-noid nucleus by procedures known per se, such as by treatment with fuming nitric acid under nitrating con-ditions. Second, the nitrated polymer is then reduced to form an amino-substituted polymer, again by methods well-known in the art such as by reaction with sodium hydrosulfite and sodium hydroxide under reducing con-ditions. Third, the amino-substituted polymer is diazotized, preferably by use of the well-known nitrous acid technique in which the nitrous acid is formed _ situ by the reaction of sodium nitrite and hydro-chloric acid. Fourth, this polymer diazonium salt is reacted with red blood cells or stroma under conditions whereby these materials will conjugate to the diazonium radical of the polymer to form a diazoamino bridge be-tween the two. This conjugation has been found to occur when the polymer diazonium salt is mixed with a solution or suspension of red blood cells or stroma in saline at a pH from about 6 to about 7.5 at room temperature.
The temperature is not believed to be critical and may be adjusted as desired. For polystyrene and stroma, two grams of polystyrene diazonium chloride per one gram of lyophilized stromata has been found to afford a satisfactory polystyrene/stromata conjugate. However, this weight ratio is not seen to be critical. It is contemplated that those skilled in the art could vary - the proportions of polystyrene diazonium salt and the red blood cells or stroma so as to obtain a satisfactory conjugate.

Reference is made to United 5tates Patent No. 2,274,551 for a description of one method for preparing diazonium salts of polymers containing a benzenoid nucleus.
Whlle the present invention is illustrated below by reference to polystyrene (which is the preferred polymer), it is ... .
.
' .

:~ ~f~ 5~

contemplated that any polymer having a benzenoid ring therein may be used, including those in which the benzenoid ring is attached to the hydrocarbon polymer chain in different ways. For example, hydrocarbon polymers with benzene rings attached to the chain by carbon to carbon linkages (e.g., polystyrene or polydihydronaphthalene) may be employed. Also, polymers with the benz~noid group as a part of the polymer chain (e.g., phenol-formaldehyde polymer) or polymers where the benzenoid ring is connected to the polymer chain by amide formation (e.g., polyacr~lanilide) may be employed. A11 of these types of resins may be nitrated on the benzenoid nucleus followed by reduction and diazotization to produce a polymer diazonium salt. Polymers containing a benzenoid nucleus are represented by the general formula ~ (I), which is intended to include all of the types of polymers exemplified aboveO

Included within the present inventlon are the materials produced by the above method, which are polymers con-taining a benzenoid nucleus having either red blood cells or stroma bonded thereto by a diazoamino bridge. These materials are described herein by the term "polymer/red blood cell conjugate" or polymer/stroma conjugate." The specific polymer used may be specified, such as "poly-styrene/stroma conjugatel' or the like~ These conjugates are useful for the specific remo~al or det.ection of blood groups antibodies in body fluids. For example, antibody to an antigen on the bound stroma (present in a mixture;
e.g., a biological fluid) will be specifically removed (without removal of other antibodies) when the fluid is contacted by a sufficient quantity of the conjugate.
In this way, the antibody may be specifically removed from the biological fluid if this is desired. Alternate-ly, by determining the percentage of the polymer-bound stroma antigen sites to which antibody has coupled, the presence and concentration of antibody in the biological fluid may be determined.

In view of the above discussion, the invention further comprises methods for specifically detecting or removing a blood group antibody from a mixture (e.g., a biological fluid), which comprises contacting this mixture with polymer-bound red blood cells or stroma (represented as "S") having surface antigens reactive with the antibody, whereby the antibody is coupled to the con-jugate, and removing the conjugate from the mixture.
The above process will remove the specific free antibody from the mixture without removal of any other antibody.
This mixture is preferably a biological fluid such as serum.

If it is desired to known the concentration of this free antibody in the biological fluid, the percentage of conjugated sites to which the antibody has bonded may be determined in a manner known per se, e.g., radio-immunoassay, enzyme immunoassay, and the like. Alter-natively, the bound antibody may be cleaved from the conjugate as described below, and the concentration or amount of cleaved antibody detexmined.

This process may be illustrated diagrammatically as follows. A polymer containing benzenoid nuclei is converted to its diazonium salt as described above, the polymer diazonium salt being represented by formula ~II), wherein X is a suitable anion. Reaction of this polymer diazonium salt with the red blood cells or stroma (S) forms the conjugate of formula (III) a~
shown below:

1~ ~
(II) (III) :L~4~

It should be understood the red blood cells or stroma are bonded to the polydiazonium salt through either a primary or a secondary amine group to produce a diazo-amino bond.

When the conjugate is contacted by the biological fluid, the antibody to be removed (Ab) is specifically bound as represented by formula (IV) below:

~ / rc ~\~ N=N-S . Ab (IV) This specific binding removes this antibody without removing an~ other antibody in the fluid.

DETAILED DESCRIPTION OF THE INVENTION
One particularly useful application of the present in-vention is for the production of pure antibody (e.g.,anti A, anti-B, anti-D, etc.) for the detection of blood group antigens (e.g., A, B, D, etc.) in humans. The production of pure anti-D will be used as an example.
It should be understood that this antibody is produced from a human source from which unwanted antibodies to the major blood groups (i.e., anti-A and anti-B) must be removed, while allowing the anti-D to remain. This is presently achieved by adsorbing the anti-D positive serum with corxesponding group A or group B, D-negative red blood cell~. The anti-A or anti-B antibody will bind to these red blood cells, leaving the anti-D in the serum. Removal o~ the red blood cells with the bound antibody is the first step in the preparation of a pure antiserum suitable for diagnostic testing for Rh factors. While this prior art antibody puriication method is illustrated by reference to detection of Rh factors, it should be understood that the same princi-pal is applicable for the preparation of any pure , .

.~ .

.

85i~

antiserum. The major drawback in the prior art method of purification is that fresh red blood cells (which can be used only once) are necessary for the purification.
Not only are these red blood cells very costly/ but their use is depleting a valuable natural resource.

Using the present invention, the outer membrane of suitable red blood cells (which outer membrane contains the antigenic determinants responsible for, among other thin~s, the major blood groups) may be covalently bonded to a suitable polymer support such as polystyrene beads, test tube, slide, strip, or the like, using the present method. Other polymers having a benzenoid nucleus may also be used~ Once this polymer-bonded stroma (membrane) is prepared, it may be employed to specifically remove the corresponding antibodies from serum. That is, polymer/stroma conjugate rom A-positive, D-negative red blood cells will remove anti-A but not anti-D, while polymer/stroma con~ugate from B-positive, D-negative red blood cells will remove anti-B but not anti-D, and so on.

Onesignificant advantage of the polymer-bound stroma over unbound red blood cells is that the former can be reused, while the latter cannot. By placing the polymer-bound stroma having antibodies bound thereto in an acidic solution, essentially all the adsorbed antibody is released and the polymer-bound stroma is ready for reuse. Moreover, the polymer-bound stroma has been found in many instances to be more efficient in the removal of unwanted antibodies than unbound red blood cells.

The possibility of regenerating the polymer-bound stroma by cleaving of the antibody in an acidic solution has the additional advantage that the cleaved antibody may be recovered in substantially pure form. Thus, using the present invention, a given antibody may be obtained .

c~ ORD 22 ~) in substantially pure orm ~y either: 1) specific re-moval of contaminating antibodies from serum by adsorbing them on polymer-bound stroma; or 2) specific adsorption of the desired antibody onto the polymer-bound stroma followed by cleavage and recovery. Essentially all of the bound antibod~ can be eluted from the polymer-bound stroma in this fashion.

If the prior art non-covalent bonding technique had been used, treatment with an acidic medium would have cleaved not only the stroma/antibody bond, bu-t also the poly-styrene/stroma bond as well. Thus, this prior art material could not be successfully regenerated.

Not only anti-D antibodies, but also antibodies to any other antigenic determinant may be isolated from a biological fluid (e.g., serum) containing them.
Examples of such blood group antibodies are (ln addition to anti-D), ABO and Rh antibodies such as anti-A, anti-B, anti~H, anti-C, anti-E, anti-c, anti-e, anti-f, anti-Ce, anti-CW, anti-V, anti-G, anti-hrS, anti-VS, and the like. Other examples of blood antibodies which the present method is useful to isolate are anti-Gamma, anti(Non-Gamma), anti-Lewis, and the like.
The invention is illustrated by the following examples:

E~AMPLE I
Polystyrene beads (5.0 g; 200-100 mesh; Eastman) were cleaned by dissolving them in N,N-dimethylformamide (50 ml) for about 10-15 minutes at ambient temperature, after which the solution was poured into distilled water to precipitate the polystyrene. The polystyrene was then separated from the supernatant and placed in an oven to dry at about 100C for approximately 90 minutes. The dried polystyrene was then pulverized to a fine powder. This cleaning step removes impurities such as detergents and tne like which are left on the surface of the particles during manufacture.

The cleaned polystyrene prepared above was added slowly to 50 ml of fuming nitric acid in an ice bath at a rate to keep ~he temperature of the reaction mixture below about 50C. The mixtuxe was stirred for 60 minutes, and was then poured into cold distilled water to stop the reaction. The light yellow nitrated polystyrene was filtered from the supernatant liquid, washed with water, and dried.

The nitrated polystyrene was then combined with 200 ml of 2 N aqueous sodium hydroxide, followed by 12 grams of sodium hydrosulfite, and the whole was stirred at about 100C for about three days to produce amino polystyrene. At the end of this time, the reaction mix-ture was cooled to 0C in ice and an excess of sodiumnitrite (10 grams) was added with stirring. Then 3 N
hydrochloric acid was titrated into the reaction flask carefully so as to keep the temperature of the reaction mixture as close to 0C as possible. The reaction is complete when nitrous acid is no longer generated in the flask upon addition of hydrochloric acid ~as shown by the cessation of yellowish nitrous oxide gas evolution~. The pH of the suspension of polystyrene diazonium chloride was then adjusted to 7.5 with aqueous sodium bicarbonate solution and stored. The polystyrene diazonium chloride was filtered off, washed with deionized water and dried just before preparation of the conjugate in Example II below.

EXAMPLE II
Two grams of polystyrene diazonium chloride prepared in Example I were placed in 20 ml of 0.9~ saline and the p~ adjusted to 7.0 ~ 0.2 with either 1 N hydrochloric acid or 1 N sodium bicarbonate~ To the suspension of polystyrene diazonium chloride was added one gram of lyophilized B-positive stroma slowly with mixing and the suspension was allowed to mix for 30 minutes. The , : .

5~

polystyrene-bound stroma was then washed with 0.9~
saline fxom 3 to about 5 times until the wash solution is clear. The polystyrene-bound stroma (conjugate) is now ready for use.

EXAMPLE III
Three grams of stroma/polystyrene conjugate produced in Example II was added to 30 ml of antisera containing anti-B and anti-D antibodies, after which the mixture was cooled to 5C and stirred for about one hour. The mixture was then centrifuged at 2,000 RPM for ten minutes and the antiserum was decanted. The pre-adsorption titre for anti-B was 1:64; all anti-B
activity was removed by one adsorption. The anti-D
activity o the antiserum was unchanged by the removal of the anti-B antibody.

To regenerate the polystyrene/stroma conjugate, it was washed about five times with 0.9% saline to remove any free protein. The conjugate was then resuspended in 0.9% saline and the p~ of the suspension was lowered to between 2.7 and 3.3 with 1 N hydrochloric acid to cause the release of the anti-B antibody which had been adsorbed. After mixing for about 90 minutes at room temperatur~, the suspension was then centrifuged at

2,000 RPM for 10 minutes and the supernatant decanted.
The conjugate was then washed five times with 0.9%
saline, after which it was suspended in 0.9% saline and the pH adjusted to 7.0 with sodium bicarbonate.
The conjugate was then stored at 5C. This regeneration has been performed 27 times without diminution in adsorption activity of the conjugate.

The anti-B antibody which was removed from the conjugate 3S by centriugation as described above may then be used as desired.

:: :
', ~

.: ~

:~4~ V ~

EXAMPLE IV
The purified anti-D antiserum produced in Example III
was bonded to a number of polys~yrene diazonium chloride tubes following the technique of Example II. The polystyrene diazonium chloride tubes had been prepared from polystyrene tubes (Falcon Plastics~ following the procedure of Example I, except that no cleaning of the tubes was necessary prior to nitration. Into separate tubes were introduced D~ red blood cells and D red blood cells. The former adsorbed onto the bonded anti-D
on the tube wall, while the latter was not adsorbed.
The adsorbed D~ cells were lysed and the amount of hemoglobin determined spectrophotometrically to quan-titate the number of D+ cells.
EXAMPLE V
The ability of polystyrene/stroma conjugate to remove anti-A,B was demonstrated by the following test. The conjugate was prepared as in Example II but using a mixture of group A and group B stroma instead of just group B stroma. The adsorption was performed as in Example III except that the one-hour incubation prior to centrifuging (which was at 3,400 RPM) was performed at 25C.
Anti-A,B serum having a strong reaction at a titre of 1:32 and a granular reaction at a titre of 1:1024 was used. Adsorption of 1.0 ml serum with 10 ml of stroma conjugate reduced the titre to granular at 1-8. A
second adsorption using fresh stroma conjugate removed all anti-A,B activity.

This example demonstrates the efficiency of the subject method for removal of anti-A,B antibodies from serum.
For comparison, adsorption with an equivalent amount of unconjugated mixed A~ and B+ red blood cells yielded equivalent results.

* Trademark ... ..

., ~

, ORD ~2 .~4~8S~

EXAMPLE VI
Following the procedure of Example V, adsorptions were performed on anti-gamma, anti-non-gamma r and anti-Lewis sera containing anti-A and anti-B as contaminants. The conjugate removed an equivalen~ amount of non-specificity for the first two sera compared to the red blood cells.
One adsorption with the conjugate removed the same amount of non-speci~icity rom the anti-Lewis serum as three adsorptions with red blood cells.
EXAMPI.E VII
Following the procedure of Example III but using the A, B stroma conjugate of Example V, adsorptions were per-formed on anti-D serum containing anti-A and anti-B as contaminants. The pre-adsorption titres were strong through 1:8. When the adsorption was conducted at 25C, the resulting serum had a strong anti-A reaction only at a titre of 1:1 and a strong anti-B reaction only at a total of 1:2. An adsorption at 5C resulted in serum which had no detectable anti-A or anti-B.

EXAMPLE VIII
Following the procedure of Example III but using stroma conjugate of the appropriate specificity, the following results were obtained. The contaminant strength was measured at 1:1 concentration by two methods before and after adsorption with the conjugate. In the first method (,designated "Imm"), the strength was determined on the mixture of antiserum and red blood cells imme-diately after mixing, whereas in the second method(designated "Inc"), the mixture is incubated at 37C
prior to the reading. The latter method is more sensi-tive. The strength is given on a scale of 0 (no reaction) to 5 ~strongest reaction). In several in-stances the presence of post-absorption activity indicated the necessity of readsorption.

.

.
- . .

O ¦ r-l r-l ~I r~ r-l r~ H ~ 1 ~1 0 ~I r-l O O O O ~1 r-l ~1 ~1 ~
O

~r~ O ~
~} 1~ H O ~ r r~ I O O O O O O O O O O O O O O
o ~ ~
O r~ H
~1 O
~ U~
a) r~
U~ ~ ~
In U~ U~ ~ LnLr) ~r ~ ~ o ~r ~1 H

O ~
m ~m~m~m ~m~m m - m - m m ~ ~m P~ to E~ O
o ~o ~
r~
'I ~ R m ,C m ~ m ~ m m a: m m ~ ~ m C) ~ r~ r~ r~ r~ ~1 r~ ~r~ r~ ~r~
~r1 ~r~ ~r~ ~ ~ ~ ~) ~) ~ ~ I I I I ~1 _ ~) - - ~ ~ ~\ ~ J_) _:
a~ o m h R R~:1 R R .4 ,~ o o ' o o o o o o o ~ ~ a ~ a u~
r~ ~ I I ~ R 1 ~ RI r4 ~ I I I I ,q ~:1 ~1 ~1 U~ ~r~ rl ~ fd ~ r~ 1 aS

~ m t~ a ~ ~ V 5: H

5~3 EXAMPLE IX
Following the procedure of Example V but using stroma conjugate of the appropriate specificity, the following comparative results were obtained. The data indicate that equivalent results (absence of contaminant) were obtained with both the prior art red blood cell ad-sorption method and the present method.

Strength of Contaminant Specificity of Type of @ 1:8 after Adsorption Specificity of Contaminating Conjugate Red Blood Cell Conjugate Antiserum Antibody Used Adsorption Adsorption A. Anti Human Anti species A O O

Gamma B O O

O O O

20 B. Anti Human Anti species A O O

Non-Gamma B

O O O

EX~MPLE X
To compare the efectiveness of the present method with the prior art method for removing undesired contaminants from anti-D antiserum while retaining anti-D activity, the following test was conduc~ed. Three lots of anti-D
antisera containing anti~A and/or anti-B were purified using: 1) first group A stroma conjugate and second group B stroma conjugate ~two lots of anti-D antisera);
and 2) the prior art red cell method (one lot of anti-D
antisera). After removal of all anti-A and anti-B
activity, the antisera were tested for anti-D activity.
The conjugate-purified lots exhibited the same anti-D
titre endpoint (64-128) as the red cell-purified lot, but were slightly more active in quantitative evaluation following the immediate determination technique of Example VIII.

. ~

.. . :

~ 58 ORD 22 As used herein, the term "essentially all" in reference to elution of anti~odies from the conjugate means greater than a~out 95~ of the bound antibody. The term "specifically" In reference to removal of an antibody means that only that antibody corresponding to the stroma antigen Ce.g., anti-A) is removed, without removing an antibody ~e.g., anti-~) which is to be left in the serum. Of course, if stroma having more than one antigen is used (e.g~, A~, B+~, removal of anti-A and anti-B ~but not anti-D) from the serum is specific removal.

: ' :

Claims

WHAT IS CLAIMED IS:

1. A method for covalently bonding red blood cells or stroma to a polymer containing a benzenoid nucleus which comprises the steps of:
a) nitrating the benzenoid nucleus of the polymer to produce a nitrated polymer;
b) reducing the nitrated polymer to produce an amino polymer;
c) diazotizing the amino polymer to produce a polymer diazonium salt; and d) reacting the polymer diazonium salt with the red blood cells or stroma to form a covalent diazoamino bond between the diazo group of the polymer diazonium salt and an amino group of the. red blood cells or stroma.

2. The method of Claim 1 wherein the polymer containing a benzenoid nucleus is polystyrene.

3. A chemical composition comprising a polymer containing benzenoid nuclei having red blood cells or stroma bound thereto by a diazoamino bond.

4. A chemical composition comprising polystyrene having red blood cells or stroma bonded thereto by a diazoamino bond.

5. A method for specifically removing an antibody (Ab) from a mixture of said antibody with other materials which comprises:
a) contacting said mixture with a chemical com-position of Formula:

wherein represents a polymer containing a benzenoid nucleus, and S is red blood cell or stroma having an antigen for said antibody whereby the antibody is adsorbed onto said bonded S to yield a second chemical composition of Formula:

and b) separating said second chemical composition from said mixture.

6. The method of Claim 5 wherein is a polystyrene/stroma conjugate.

7. The method of Claim 6 which further comprises the steps of:
c) reacting said chemical composition with acid at a pH of from about 2.7 to about 3.3 to cleave essentially all of said antibody from said polystyrene/
stroma conjugate; and d) separating said cleaved antibody from the polystyrene/stroma conjugate.