Although TiO.sub.2 is preferred as a support and was one of the carriers used in the illustrative examples below, inorganic supports consisting of silica, alumina, and porous glass were also used successfully to bond biologically active proteins via the polymeric isocyanate. Among the biologically active proteins successfully coupled to inorganic according to the methods disclosed herein were the enzymes papain, urease, glucoamylase, and lactase. A representative antibody (anti-digoxin) was also successfully coupled. The procedures used to couple the papain and anti-digoxin antibodies are described, for illustrative purposes, in the examples below.
EXAMPLES I AND II
Papain Coupled to Porous TiO.sub.2 and SiO.sub.2
In these examples, the polymeric isocyanate consisted of a 1% (v/v) solution of the PAPI 901 polymer. To prepare the solution, 0.25 ml of the PAPI-901 was diluted to 25 ml with acetone. The papain (0.2 Hoover Ball Milk Clot) was obtained from Nutritional Biochem. Co. The titania support consisted of porous TiO.sub.2 particles (30/80 mesh) having an average pore diameter of about 420A and a surface area of 35 m.sup.2 /g. The porous silica particles (25/60 mesh) had an average pore diameter of 500A and a surface area of 50 m.sup.2 /g.
Preparation of Carrier Surfaces
Two 500 mg quantities of each titania and silica support (carrier) were placed in four 25 ml Erlenmeyer flasks. To each flask was added 1 ml of the coupling solution which was allowed to react with the carrier at room temperature for 5 minutes. The titania immediately reacted with the coupling agent to form an orange-coated porous body. The silica was somewhat slower to react but a yellow coating of coupling agent formed on the silica bodies. The color due to the coupling agent was completely removed from solution in the case of the titania porous bodies. The residual acetone solvent was decanted from the flask and the porous bodies were then ready for reaction with the enzyme.
The enzymes were then coupled to the treated carriers in two separate series (Series 1 under alkaline conditions and Series 2 under acidic conditions). The reactions were as follows:
Coupling of the Enzymes
Series 1 -- immobilization from alkaline, pH 8.0, enzyme solution. One gram of the papain was diluted to 10 ml with 0.2 M Na.sub.2 HPO.sub.4.
To one flask containing isocyanated titania and one flask containing isocyanated silica was added 10 ml of the above enzyme solution and the flasks were placed in a 23.degree. C water bath with shaking for 7 hours. The flasks were then removed from the shaking water bath, the unreacted enzyme solution was decanted, and the immobilized enzymes were washed first with water, followed by 0.5 M sodium chloride and finally followed by water again.
Series 2 -- immobilization from mildly acidic solutions, pH 5.2.
Enzyme solution -- 1 gram of the papain was diluted to 10 ml with distilled water.
To each of the remaining 2 flasks, one containing the isocyanated titania and one containing the isocyanated silica, were added 10 ml of papain solution and the flasks were placed in a shaking water bath at 23.degree. C. The flasks were shaken for 3 hours. The flasks were then removed from the bath and the unreacted enzyme solutions were decanted. The immobilized enzymes were first washed with water and then 0.5 M sodium chloride solution, followed by a final wash with water.
The series 1 and 2 immobilized enzymes were stored at room temperature in water between assays. The immobilized enzymes were assayed with 0.2% egg albumin solution in acetate buffer containing cysteine and EDTA at pH 5.0 at 37.degree. C. The optical density of the trichloroacetic acid extract of the reacted substrate was determined at 280 nm and compared to a curve for the free enzyme to give the milligram of active papain per gram of carrier. These results are recorded below.
In previous immobilization studies that were performed with porous silica and porous glass, it was noted that most of the activity (99% or more) of papain was lost during the first 21 days of storage. In addition, on a per gram basis of carrier, the loadings were usually under 30 mg of papain. It is significant to note from the results above that there was little difference between the loading and stability on the titania of the alkaline and acid series. It would appear as though higher loadings were achieved under the acidic condition in the silica series.
COMPARISON WITH ADSORBED ENZYME
To have a basis for comparing the efficiency of the polymeric isocyanate as a coupling agent in a practical application (e.g. flow through column), the coupled enzyme composite was compared with an adsorbed enzyme system.
In preparing both the adsorbed (Column #1) and chemically coupled papain (Column #2) for comparison studies, the above described porous titania particles were used.
Column #1 -- Adsorbed Immobilized Papain
Preparation of Enzyme for Immobilization
40 gms of the papain was dispersed to 400 ml with distilled water and transferred to dialysis tubing. The papain was dialyzed against 4 liters of distilled water at room temperature for 30 minutes. The dialysis water was then changed and dialysis was continued for 30 mins. Every half hour the dialysis water was changed and the dialysis was repeated. This was repeated for a total of 6 times. The final volume of the papain solution after dialysis was 485 ml.
Preparation of Carrier for Immobilization
400 ml of water was added to 20 gms of the porous titania particles in a flask. The flask was placed in a shaking water bath at room temperature for 1 hour and 20 mins. with shaking. The water was then decanted and 400 ml of 0.1 M Na.sub.2 S.sub.2 O.sub.3 (pH 7.9) was added. The flask was shaken for 10 mins. at room temperature and the solution was then decanted. 400 ml of water was then added to the flask and shaking was repeated for 1 hour at room temperature. The water was decanted and the carrier was ready for immobilization.
Preparation of the Immobilized Enzyme
The 485 ml of dialyzed papain was added to the 20 gms of pretreated titania in a flask. The flask was immersed in a water bath at room temperature and allowed to shake overnight (approximately 17 hours). The enzyme solution was decanted and washed first with 500 ml of distilled water by shaking in a water bath for 1/2 hr. at room temperature and then with 500 ml of 0.5 M NaCl solution by shaking in a water bath for 1 hour at room temperature. After decantation of the sodium chloride solution, the immoblized (adsorbed) enzyme was finally washed with 3 washes of 500 ml of distilled water, each wash comprising 1 hour of shaking in the water bath at room temperature. The immobilized enzyme was then transferred to a water jacketed column which was controlled to 37.degree. C. A 0.5% casein substrate containing cysteine and EDTA buffered to pH 6.1 with sodium acetate was pumped through the column at approximately 40 ml per hour.
Column #2 -- Immobilized Papain by Chemically Coupling to Titania with Polymeric Isocyanate
Preparation of Carrier for Coupling
40 ml of the PAPI-901 was added to 20 gms of the titania particles in a 500 ml Erlenmeyer flask. This was followed by the immediate addition of 500 ml of acetone. The sample was mixed and allowed to stand for 30 mins. at room temperature. The acetone containing the PAPI was decanted and the porous bodies were washed with 400 ml of distilled water. The water wash was immediately decanted and the carrier was utilized for coupling.
Immobilization of Papain by Chemically Coupling
A papain solution containing 40 gms of papain in 400 ml of water was immediately added to the polymeric isocyanate carrier and the flask with contents was placed in a water bath at room temperature with shaking for 3 hrs. The flask was then removed from the bath and allowed to stand at room temperature overnight (approximately 17 hours). The enzyme solution was decanted and samples were washed with distilled water several times. The immobilized enzyme was transferred to a water jacketed column at 37.degree. C and the same substrate utilized for the adsorbed enzyme was pumped through the immobilized enzyme at 40 ml per hour.
Monitoring of Immobilized Enzyme Columns
Samples of the product collected from the columns were precipitated with trichloroacetic acid. The soluble extract was separated from the precipitate and analyzed at 280 nm. The % hydrolysis was obtained by comparing the optical density with that of a TCA extract of a 0.5% casein solution which had been completely hydrolyzed by papain with no subsequent reaction with papain. The % hydrolysis figure was then utilized for determining the grams of casein converted per hour. The results of these column studies are recorded in the following table.
These results were plotted and a linear regression was determined before the column was clogged up and then after the immobilized enzymes in each column were washed with water.
It is apparent from the results that the adsorbed enzyme exhibited an activity that was 22% greater than the coupled enzyme at the third day of operation. However, after 20 days of operation, the adsorbed enzyme exhibited an activity of 20-30% lower than that of the coupled enzyme. It is further apparent that after washing the coupled enzyme with water, this preparation was fully restored to its third day activity, while that of the adsorbed enzyme recovered only 70-75% of its third day level with washing. If one extrapolates back to zero time, it may be noted that the coupled enzyme is loaded only to approximately 80% of that of the adsorbed enzyme. However, after about 10 days of operation, these columns are about equal in activity. The half-life of the adsorbed column, column #1, is approximately 16 days, while that of the coupled enzyme, column #2, is about 26 days.
The half-lives discussed are apparent half-lives, since these columns can be restored to higher levels of activity with water washing. In an attempt to gain information with respect to the true half-life, one would have to again extrapolate back to zero time for each column and determine the losses of activity at a point when the columns were thoroughly washed. By utilizing this procedure, the calculated half-life for the adsorbed column is approximately 41 days, while that of the coupled column is 107 days.
EXAMPLE III
Anti-digoxin Antibodies Coupled to Titania Particles
It is well known that immobilized antibodies (IMA) can be used in a variety of applications requiring a high degree of chemical specificity (e.g. antigen or hapten purification from a mixture of molecules and solid-phase radioimmunoassays (SPRIA) to detect and determine concentrations of various substances). In this example, goat anti-digoxin antibodies which had been dialyzed against 0.01 M phosphate buffer at pH 7.4 were coupled via the PAPI 901 to titania particles having an average particle size of 320A and a surface area of 48 m.sup.2 /g. The anti-digoxin antiserum contained 66.5 mg protein per ml by the Folin Phenol method and 42.75 mg protein N/ml by the hydrolysis-Ninhydrin determination. The antiserum bound 50% of 0.8 mg of .sup.3 H digoxin at a dilution of 1:45,000. The polymeric isocyanate solution consisted of 0.1 ml of the PAPI-901 diluted to 20 ml with acetone (50 mg of the isocyanate per 10 ml). The titania carrier in this example was non-porous to eliminate or minimize diffusion effects in the subsequent assays.
Coupling Procedure
1 gm quantities of the titania were transferred to each of two Erlenmeyer flasks (25 ml) to prepare 2 samples, IMA #1 and IMA #2. To each flask 10 ml of coupling solution (50 mg PAPI 901) was added. The contents were mixed and the flasks were stoppered and permitted to stand at room temperature for 1 hour and 10 minutes. During this time the coupling solution, which had been yellow due to the presence of the coupling agent, turned colorless. The colorless acetone solvent was decanted and 1.25 ml of the anti-digoxin was added to each flask containing the isocyanated titania followed by the addition of 5 ml of 0.01 molar phosphate buffer, pH 7.4, to each flask. The flasks were placed in a reciprocal shaker and shaken at room temperature for 1 hour and 10 minutes. The samples were then transferred to respective 30 ml centrifuge tubes with the aid of water addition to about a 25 ml total level. The contents of the centrifuge tubes were mixed with a vortex mixer followed by stirring with a Teflon.TM. stirring rod. The tubes with contents were centrifuged in a clinical centrifuge at top speed for 2 minutes. The clear supernatant fluid was decanted and saved from each tube for protein analysis. The sediments in the centrifuge tubes were then washed with 15 ml of buffer, mixed, and recentrifuged as above. This wash with buffer was repeated two more times and all of the washes were pooled from each sample for protein analysis.
Control
An identical experiment was performed as above utilizing an anti-rabbit gamma globulin from the goat which contained no anti-digoxin activity as a control. The antiserum showed no reaction in a soluble system with digoxin. It contained 49.90 mg protein N/ml by the Ninhydrin method and 77.0 mg/ml by the Folin Phenol determination.
Blank
The blank carrier was prepared with the same quantity of coupling agent utilizing the same procedure except that no protein was added to the carrier.
Dry Weight Determination
The covalently coupled immobilized antiserum wet cakes (IMA #1 and IMA #2) were analyzed first for their dry weight and protein nitrogen content. Portions of the wet cake were dried in hot air ovens (110.degree. C) overnight and the weight difference calculated to obtain dry weight.
Protein Nitrogen Determination by Ninhydrin Method
A portion of the wet cake was prepared as a 1% suspension in phosphate buffered saline (PBS). Ninhydrin protein nitrogen determinations were performed on 1 ml portions of these IMA suspensions. All washes following the coupling reaction were also assayed for protein nitrogen.
Biological Activity Determination
The same 1% IMA suspension was assayed also for its biological activity. Dilutions of the IMA's were prepared in stabilized phosphate buffered saline (SPBS) so that duplicate ml aliquots were reacted with 0.050 ml of .sup.3 H-Digoxin (0.8 ng digoxin). The PBS consisted of 0.01 M phosphate in 0.15 M NaCl. The stabilized PBS (SPBS) was prepared by the addition of 1 mg of bovine serum albumin (BSA) and 0.02% NaN.sub.3.
The reaction was performed at room temperature for 10 min. The reaction tubes, with total count and blank controls, were centrifuged at 4500 rpm and 5.degree. C for 30 min. The supernatant solutions containing the free (unbound) .sup.3 H-Digoxin were decanted into liquid scintillation vials containing 10 ml of Aquasol (New England Nuclear) and counted for 1 min. in a Packard TriCarb Liquid Scintillation Spectrometer, Model 3320. The percent .sup.3 H-Digoxin bound was determined by calculating the ratios between total free counts in the tests and the counts after reaction with IMA. The results were made into a graph and the 50% binding points determined.
Solid-Phase Radioimmunoassay
The IMA's were used to perform a SPRIA. The dilutions from a 1% suspension of IMA calculated to bind 50% of .sup.3 H-Digoxin were added to 0.2 ml volumes of human plasma containing digoxin concentrations of 0.5 to 10 ng/ml. The total volume was 1 ml to which was added 0.05 ml of .sup.3 H-Digoxin. The reagents were mixed and reacted for 10 min. at room temperature, and the bound and unbound .sup.3 H-Digoxin separated and calculated as described above. The results were analyzed statistically by a computerized two variable linear regression.
Results
Dry Weight and Protein Analysis
The IMA complexes were relatively uniform with respect to protein N/g dry weight of support. The procedure coupled an average of 97.5% of the protein N offered, and an average of 102.88% of protein N was recovered. The 26.2 mg of protein N/g dry weight of carrier (4.2 mg N/g) on the TiO.sub.2 -polyisocyanate control can be attributed to the presence of ligand itself (i.e., PAPI, polymethylene polyphenyl isocyanate). The ninhydrin nitrogen determination was performed with aqueous reagents which produce a primary amine in place of the residual isocyanate group.
Biological Activity
Only those IMA complexes prepared with digoxin specific goat anti-digoxin serum demonstrated activity. The control TiO.sub.2 support-ligand and goat anti-rabbit gamma globulin IMA complexes were completely negative.
Solid-Phase Radioimmunoassay
Representative curves of IMAs #1 and #2 were prepared. Analyses of these curves indicate correlations of 97.4%. These curves could be used to detect and quantitate 0.5 to 10.0 ng of digoxin/ml of human plasma. The results and properties of the two samples (IMA #1 and IMA #2) are compared with the control (anti-rabbit gamma globulin) and blank (no antibody) in Table III.
From the above, it was found that the procedures for covalently coupling antiserum to TiO.sub.2 -polymeric isocyanate (TiO.sub.2 -PAPI) derivative demonstrated several advantages: (1) The TiO.sub.2 -PAPI derivative is fast and easy to prepare. (2) The speed of the antiserum coupling step rivals those of previously used Schiff-base and thiourea reactions using Silanized inorganics. (3) The efficiency of antiserum protein N coupling averaged 97.5% (93.3-104.2%) which compared favorably to an average of 81.7% (73.48-91.25%) for azo-linkage to arylamine-(organic solvent silanization) bonding techniques, 70.5% (67.5-73.5%) for Schiff-base using glutaraldehyde and alkylamine-porous glass carriers, and 107.1% (87.5-121.5%) for the thiourea bonding method.
The percent of activity recovered was a low average of 11.675%, compared to a range of 17% to 49.5% by azo-linkage to arylamine CPG. The difference may result from the isocyanate derivatization step which had not been optimized and which may have been cross-linking the anti-serum protein excessively. The overall procedure has the following advantages: (1) low cost and ready availability of materials and reagents; (2) rapidity of producing the polyisocyanate derivative; (3) the ease of coupling antiserum to the derivative; and (4) the high rate of coupling efficiency.
Other Ways of Utilizing Polymeric Isocyanate Coupling Agent
It may be desirable to couple an enzyme through other than an isocyanate group. It this were the case, then an organic amine or hydroxyl compound with another functional group could be directly attached to the isocyanate. Another approach could be to react the isocyanate after it is coupled to the carrier with water. Water reacts with the isocyanate to produce carbon dioxide and results in the production of an amine on the original coupling agent in place of the isocyanate group. This amine could be treated in the same fashion as the amine on silane coupling agents for either modification or attachment to other functional molecules such as glutaraldehyde, etc.
Inasmuch as the above-described procedures for bonding biologically active proteins to inorganics are subject to modification, it is intended that the examples should be construed as illustrative and that the invention should be limited only by the appended claims.