CA1083057A

CA1083057A - Pressure driven enzyme coupled membranes

Info

Publication number: CA1083057A
Application number: CA264,995A
Authority: CA
Inventors: Harry P. Gregor
Original assignee: Individual
Current assignee: Individual
Priority date: 1975-11-07
Filing date: 1976-11-05
Publication date: 1980-08-05
Also published as: JPS5283995A; DE2650920A1; FR2330695A1; GB1519676A; IT1121686B; FR2330695B3; JPS5950317B2; US4033817A

Abstract

ABSTRACT OF THE DISCLOSURE

Pressure-driven enzyme-coupled membranes are prepared starting with a membrane filter having pores of appropriate, molecular dimensions and composed of a polymeric matrix, the activation of that matrix (where neces-sary) under pressure-driven conditions to impart to it appropriate groups capable of coupling to enzymes and similar substances of biological activity under pressure-driven conditions, and then employing this system under pressure-driven conditions to effect conversion processes which include the use of the coupled enzymes as the catalysts.

Description

~3~5~
The present invention relates to matrix membranes~ coupling reactions and enzymes. It is directed to the preparation of a broad class of enzymes coupled with ultrafiltration membrane filters in which enzymes and other molecules of similar catalytic activity are attached by chemical bonds, where the process of activation of the pore surfaces of these membranes for purposes of subsequent coupling (where such is needed) is carried out under pressure-driven conditions, and where the coupled enzyme system is used under similar pressure-driven conditions, namely by forcing the substrate to be treated through the membrane pores under pressure.
The conventional uses of immobilized enzymes are well known in the scientific, industrial and patent literature. The advantages of this general technique are, accordingly, well known. The usual technologies employed heretofore have been to use small particles of either natural or synthetic polymers or inorganic, porous materials. The enzymes are coupled thereto by chemical bonds or by the process of forming an insoluble gel in the presencè of these enzymes, or by encapsulating the enzymes within small beads or within hollow fibers of porous nature, and the like. All of these conventional processes have their advantages and also their disadvantages.
The purpose of the pressure-driven enzyme-coupled membranes of the present invention is directed to a novel method of preparing such membranes and to the membranes produced thereby, and to their use for a variety of applications.
Their principal use is as a catalyst for carrying out chemical reactions.
In the form which is described herein, the systems display a high capacity in terms of amount of enzyme contained therein and a high enzymatic activity, thus making these systems particularly useful for large-scale industrial processes. The enzymes so stabilized have, in addition to the before mentioned advantages, the usual, intrinsic advantages of stabilized enzymes in terms of their chemical and thermal stability.
This invention seeks to overcome these difficulties by providing a ~ -: -., ~ .

. :
: - 1 -~83~

process for producing an enzyme-coupled ultrafiltration membrane comprising forcing a solution of an enzyme through an activated porous ultrafiltrakion membrane under a pressure equivalent to at least about 10 psig, the activated sites being reactive with said enzyme thereby to couple said enzyme to said membrane.
In another aspect, there is a process for producing an enzyme-coupled ultrafiltration membrane comprising forcing a solution of chymotrypsin or ~-galactosidase under a pressure equivalent to at least about 10 psig through a porous cellulosic ultrafiltration membrane activated at various sites with cyanogen bromide, the activated sites being reactive with said enzyme thereby to couple said enzyme to said membrane.
In yet another aspect, the invention provides an ultrafiltration membrane having pores whose surfaces are lined wi~h an enzyme coupled thereto, said pores having an average size from about 3 to 5 times the diameter of said enzyme. Preferably said enzyme is selected from a cellulosic ultra-filtration membrane having pores whose surfaces are lined with an enzyme coupled thereto selected from the group consisting of chymotrypsin or ~-galactosidase.
Furthermore, the invention provides a method of using the ultra-filtration membrane defined above for effecting an enzymatic reactioncomprising forcing a substrate solution through said membrane under a pressure equivalent to at least about S psig, said substrate solution containing a material which is capable of being converted by said enzyme, whereby said enzyme-coupled membrane effects s~id enzymatic conversion of said material.
In accordance with the present invention, I have found that the enzyme reactors of high capacity and long term stability, of a kind which is particularly suitable for large-scale industrial use, can be prepared as follows. First, reasonably homoporous matrix membranes are prepared. These matrix membranes can be prepared from either homopolymers~ from copolymers or from interpolymer mixtures. ~hen prepared from homopolymers they must have-an appropriate chemical and mechanical stability so that they can be used for periods of months in the presence of a solvent ~usually water) without ~ - 2 -C;~ . ,' ' suffering significant changes in their porosity (volume fraction which ispores), pore diameter or dimensions. Accordingly, homopolymers which are employed must be insoluble. They can swell to but a limited extent in the solvent such as water. There are a limited number of homopolymers which are useful for the purposes of this invnntion. Cellulose is one homopolymer, and polyvinylbenzylchloride is another. The usual means of preparing membranes of these kinds is to cast them as a film from an appropriate solvent, then after part of the solvent has evaporated to form a gel-like film, the system is coagulated so as to produce a film having at least 50% of the volume of the film as pores of appropriate molecular dimensions. While there are many techniques which can be employed for the preparation of membranes of these kindsl a preferred procedure is to allow enough of the solvent to evaporate so that one forms a gel-like structure and then to introduce via the vapor phase a second solvent which acts to precipitate or coagulate the polymer.

- 2a -:

5t7 Introduction of the second solvent by the vapor phase helps to control the rate and extent of coagulation such as to produce highly homoporous membranes of desirable properties. Examples of films which can be made in this manner include those from cellulose dissolved in a mixture of dimethylsulfoxide and paraformaldehyde, prepared in accordance with the formulations of D.C.
Johnson et al, (Institute of Paper Chemistry Technical Paper Series No. 5, April~ 1975, Appleton, Wisconsin~. Matrix polymer precipitation in this case is achieved by the use of either water or me~hanol vapor, and a film of de-sirable characteristics is produced along with the regeneration of the cellu-lose. Alternately, one can use a homopolymer or copolymer of vinylbenzyl-chloride, in this case polyvinylben~ylchloride or a copolymer of vinylbenzyl-chloride with styrene. These polymers are dissolved in a suitable solvent such as methylene chloride, the film is cast therefrom and then coagulation in the vapor phase employing methanol produces a membrane of the desired characteristics .
An alternate and a usually preferred method of preparing matrix membranes is to employ the general procedures prescribed by H.P. Gregor in United States Patent 3,808,305 wherein membranes are cast from interpolymer mixtures. In a typical example, two parts of polyacrylic acid are dissolved in a suitable solvent such as dimethylformamide with one part of a film form-ing matrix polymer such as polyvinylidinefluoride (Kynar, Pennwalt Co.) to-gether with a suitable epoxide cross-linking agent as described in United States Patent 3,808,305 and this is then cast in the usual manner. In order to make these membranes of suitable porosity, while the cast membrane is par-tially dry~ it is then coagulated by the introduction of a vapor which causes ;
a partial coagulation but one which does not cause a loss of the structural integrity of the membrane. A suitable solvent in this case is water or ~ -another substance which causes the ionogenic polymer to become ionic, such as ammonia or triethylamine, following which final drying and curing by cross-linking of the film renders a membrane which has pores of appropriate , .: , .
,: : . . . . ..
. . - , ': ' ' . ': :

~3~

dimensions while still containing a high concentration of carboxyl groups to which coupling reactions can take place. Similarly, amines such as poly-vinylimidazole or poly N-methylethylenimine can be employed in this formula-tion. Polymer mixtures which include polystyrenesulfonic acid provide for a highly polar negative charge within the membrane, and can be used together with a group capable of coupling such as polyacrylic acid. Accordingly, it is evident that by the judicious selection of matrix polymers and polymers to which coupling can take place, one can prepare a wide range of suitable matrix membranes. Thus, it is evident that a range of matrix polymers which have a wide range of hydrophilicity or hydrophobicity can be prepared, rang-ing from those which incorporated polystyrenesulfonic acid as a stronglg hydrophilic polymer to those which contain polyvinylbenzylchloride as a high-ly hydrophobic polymer.
Similarly, a wide range of coupling chemistries are available. In some cases activation is required as in the treatment of cellulose by cyano-gen bromide or the treatment of a polyvinylalcohol-containing membrane with `
cyanuric chloride. All of these procedures are well known to those versed in the art and are described in the literature, particularly in the book by O.R. Zaborskg (~'Immobilized En~ymes~, C.R.C. Press, Cleveland, Ohio, 1974).
It is not the purpose of this invention to teach new chemical reactions for these activation reactions or the coupling reactions which follow, but rather to stress the unique combination of materials and procedures as taught here-in, namelg that the coupling and activation procedures can be carried out on a suitable membrane matrix under pressure-driven conditions.
The size of the enzyme molecules being coupled and the size of the membrane pores are important considerations in this invention. The si~e of enzyme molecules can be dete~nined bg man~ techniques, of which calculations based upon their rates of diffusion are but one. The size of membrane pores are usuallg determined bg measuring the relative rates of diffusion or fil-tration of molecules of different si&e through the membrane pores and then 5~7 employing hydrodynamic equations to calculate effective pore diameters. The procedures which are described by Kawabe et al (J~ Colloid Interface Sci. 21, 79 (1966)) are useful for this purpose. I have found that calculations which employ the parallel plate rather than the cylindrical pore models are more valid for the purposes of this invention. For example, I have found that when the diameter of the enzymes being coupled are one-third the diameter of the membrane pore, and when driving pressures not so high as to cause stress denaturation of the protein are employed, that one can effect a high degree of enzymatic coupling, such that an appreciable fraction (at least 30% of the total pore surface) is occupied by enzyme. I have found, accordingl~, that ~; -a simple mechanical model suffices as a guide for these preparative proced-ures. If the pore walls are lined with enzymes by pumping in enæyme solu-tions from one side of the membrane to the other, the pore ~ntrance will be blocked unless the pore diameter is at least three times that of the enzyme molecule itself. The application of this rule is not rigid because enzyme molecules are deformable and it is for this reason the pore should be at least twice the diameter of the enzyme molecule being attached to it, prefer-abl~ some three to five times as large.
Another size consideration is that ~ich arises due to the size of the substrate molecule. For example, if a large substrate molecule such as a protein is to be digested by a protease which is coupled to the membrane -pores, then it is important that the large enzyme does not give rise to shear denaturation of the coupled enzymes while being forced through the pore.
In order to make appropriate use of the pressure-driven enzyme-coupled membrane concept, it is important that the matrix membrane have cer-tain characteristics. First, its pores must be of molecular dimensions, at ~ -least twice the size of the enzymes to be accommodated therein and preferably 3 to 10 times the diameters of such enzyme molecules, but less than 20 times these diameters so that the membranes can have a high internal surface for coupling. Membranes havi~g macroscopic pores of the dimensions of microns are not useful for the purposes of this application because their internal pore surface will be considerably smaller. The matrix membranes must be capable of withstanding the pressure gradient across them without collapse.
Even at the relatively low pressure differences imposed in this system, namely from 10 to 150 psig, on the average, very soft gel-like structures collapse and thus the advantages of the method are lost. It is not necessary that the membrane withstand the pressure unsupported because the available membrane technologies allow one to support thin and fragile membranes on a variety of support materials. It is an important criterion, however, that the membrane pores remain substantially constant in size under conditions of use. An auxiliar~ re~uirement is that a substantial part of the interior pore surface be available for the coupling of enzyme molecules. Having a large number of very fine pores which are permeable to water but not to the enz,yme molecule is not desirable.
The number of homopolymers which can be used for these purposes is limited. For example, the homopolymer polyvinylidenefluoride (Kynar, Pennwalt) and also the copolymer of acrylonitrile-vinylchloride (~ynel, Union Carbide) are both excellent film formers and have many of the requisite pro-perties desired. However, they do not possess of themselves groups which afford the convenient coupling of enzymes, and are difficult to activate because of their high chemical resistances. On the other hand~ cellulose nitrate is an excellent film former and has been used traditionally to form matrix membranes of the kind desirable for this process, but ~his polymer is not of sufficient chemical stability as such for use. The homopolymer cel-lulose does have the requisite chemical stability, it can be formed into membranes of appropriate properties either through its regeneration from cellulose acetats, or it can be formed directly by dissolution of cellulose in the dimethylsulfoxide-paraformaldehyde solvent system. Cellulose also is capable of being activated by a number of different agents for enzymatic coupling, and has been used widely for this purpose in its various forms.

~ rr~d~r~ - 6 - ~-, ~

33~5~

However, cellulose has a limited chemical stability. In addition, because of its nature, it appears to be capable of a limited reaction or coupling with carbohydrases and similar en3ymes and thus may not be the preferred matrix polymer for enzyme reactors involving these en3ymes. Further, cellulose itself is subject to decomposition by cellulases and for this reason is unde-sirable in certain speciali~ed applications.
The interpolymer system, on the other hand, affords one the chem-ical stability of an insoluble and chemically resistant film-forming polymer when combined with a second polymer capable of chemical coupling to enzymes:
where needed, one can also employ cross-linking agents to produce a material of the requisite properties. This interpolymer technique has already been highl~ developed by Gregor (United States Patent 3,808,305).
The reactive polymers which can be part of an interpolymer mixture inc:Lude cellulose, polyvinylbenzylchloride and its copolymers, poly-p-amino-styrene and its copolymers, copolymers of methacrylic acid with the fluor-anilide adduct of methacrylic acid, polythiols~yrene, polyacrylic acid and -its copolymers, the homopolymers and copolymers of maleic anhydride, poly-~
vin~limidazole, poly-N-methylethyleneImine and linear polyethyleneimine. In additionJ one can also use polymers in the interpolymer mixture, ones which act to control the en~ironment within the membrane pore, the so-caIled micro-environment of the coupled en~yme. Here one can employ the strongly acidic polymer polystyrenesulfonic acid, the strongly basic polymer poly-N-methyl-imida~olium chloride, as examples. The use of weak base and weak acid poly-mers have been enumerated above. Finally, if one wishes to make the micro-environment more hydrophobic, addition of acyl substituted polystyrene or its copolymers affords an excellent means to accomplish this, where the length of the hydrocarbon chain can be controlled at will.
The coupling of insoluble polymers to en~ymes can occur through the amino group of the enz~ne, through its carboxyl groups, through its tyrosine residues or through its mercapto groups. The most common means of coupling 3~3S~7 are achieved through the amino group of the en~yme, and for this the cyanogen bromide, anhydride, cyanuric chloride, and glutaraldehyde coupling procedures are well known. The dia~onium coupling procedure is employed for coupling through the tyrosine residues and the isocyanide or Ugi reaction couples through the carboxyl groups. Thiol groups couple through the mer~apto resi-dues of the en~yme. On the other hand, by the use of Woodward~s reagent K
one can couple either with the amino group or the carbaxyl group of the en~yme. Also, carboxylic groups on the matrix can be treated with carbodi-imide to couple an enzyme either through the amino or carboxylic group.
One of the principal areas of application of these pressure-driven enzyme-coupled membrane systems are those to the cane and beet sugar indus-tries, the corn processing industries, and their employment in the pressure-driven enzyme-coupled form is particularly advantageous. Certain en~ymes of this clAss are usually adaptable to coupling by the use of a copolymer of methacrylic acid-3-fluoranilide which is subsequently nitrated, as described by G. Manecke (Pure ~ Applied Chemistry, 4, 507, (1962)). This polymer can be dissolved along with polyvinylidenefluoride (Kynar~, Pennwalt) in a sol-vent mixture, a film is then cast, partially d:ried and then coagulated to form the desirable homoporous film. This membrane is already activated so that the direct coupling of enzymes to it follows, in this case throu~h the amino group of the en3yme. Or, a copolymer of methacrylic acid and methacry-lic acid-3-fluoranilide is prepared, combined with the same matrix polymer to form a film of appropriate porosity and then subsequent~y nitrated, activat-ing the fluoranilide groups followed by coupling. The en~ymes which are par-ticularly amenable to this coupling procedure include dextranase, invertase, cellulase, dextran sucrase, alpha galactosidase, alpha amylase and glucose oxidase.
Procedures particularly amenable to employment in pressure-driven sn~yme-coupled systems where coupling is through the dia~onium linkage can employ interpolymer mixtures of Kynar and polyaminostyrene. Interpolymer ~Trade mark ~ ;

..

films which contain cellulose can use cyanogen bromide to react with the cel-lulosic portion o~ the matrix film, and then treatment with p-phenylenedia-mine or benzidine, followed by treatment to form the dia~onium salt allows for direct coupling. The enæymes which are particularly amenable to diazon-ium coupling include invertase, alpha amylase, beta amylase, glucose isomer-ase, amyloglucosidase and beta galactosidase. The reaction of cellulose with titanium tetrachloride can be employed to couple invertase, alpha amylase, anyloglucosidase and glucose oxidase~ Finally, interpolymer membranes con-taining polyamines can be treated with glutaraldehyde to couple beta galacto-sidase and glucose oxidase. Accordingly, coupling of these carbohydrases and related enzymes follows directly and is particularly convenient when employ-ing the pressure-driven en~yme-co-upled membrane systems.
Another particularly useful embodiment of this invention is that wherein several enzymes are coupled to the same membrane or a series of mem-bra;nes, used one after the other to effect a specific conversion. In this case 4 en~yme-coupled membranes were prepared, one with trypsin, a second with chymotrypsin, a third with aminopepsidase M and a fourth with prolidase.
All of these membranes were prepared using cyanogen bromide activation fol-lowed by coupling. Upon the serial recirculation of solutions of polypep-tides or proteins through this series of membranes, a complete amino acid hydrolysis resulted. Since the enzymes were entirely immobilized, since they could not attack one another and would not leach from the membranes, a com-plete digestion was effected without contamination. This form of enzymatic hydrolysis is particularly useful for analytical purposes because the amino acids that are usually completely or partially destroyed by conventional acid hydrolysis prior to their determination, namely tryptophan, tyrosine, serine, asparagine and glutamine are not destroyed in this procedure. In subsequent experiments, mixtures of these 4 enzymes were coupled simultaneously and, in yet others~ dilute solutions of each of the en~ymes were serially coupled to the same membrane. Since the procedures of this inven-tion involve a rapid .. -.': "

_ g _ .:

3~

coupling to a rigid matri~ which acts to isolate the enzymes and inhibit au~-digestion and mutual digestion, a particularly useful and compact polypep tide and protein digestion system results.
Many applications of this multi-enzyme system are available. For example, it has been established that many substances which give rise to allergies and are thus antigenic do so because they are contaminated with small traces of foreign proteins. For example, penicillin and related anti-biotics are often contaminated with a small amount of protein arising either from the initial enzymatic method of production or the conversion of penicil-lin to related compounds. Thus, the treatment of penicillin solutions to remove all traces of proteins can be effected by this multiple enzymatic reactor. Similarly, many other substances which contain small amounts of protein, ones which are difficult to remove by conventional means are amen-able to treatment by this procedure. They can be used ~o decompose traces of protein present in commercial preparations of the alginates to produce com-pounds which do not possess a protein antigenicity.
The pressure-driven en~yme-coupled membranes of this invention are useful also for a number of important industrial processes in the drug and fermentation industries. One important area o~ application involves the use of such coupled enzymes to produce valuable derivatives from a variety of synthetic and natural products. A specific example is the use of membranes incorporating an acylase enzyme capable of the conversion of penicillin to 6-aminopenicillanic acid, also referred to as 6-AMP or 6-APA. This valuable derivative is produced from penicillin obtained by fermentation by the use of conventional fermentation processes. Pressure-driven enzyme-coupled reactors are important here because of their high capacity, reactivity and stability and the high purity of the products prepared thereby.
The examples given below are taken from laboratory studies and are representative of what can be obtained also under pilot plant and full-scale plant conditions. Since the process, in essence, consists o-f pumping a . .

~3~

solution containing a given substrate through the membrane to which the en~yme is attached, the same process takes place on a large scale as on a small one. Accordingly, the scaling up of these processes offers no problem, another one of its advantages. In conventional fermentation, as an example, the prediction of the parameters of a large-scale fermentation process from those obtained in small laboratory fermenters is hazardous.
Many mechanical configurations are available for use with the objects of this invention. The devices employed are those which are cammonly employed in conventional ultrafiltration and reverse osmosis processes.
These include the conventional plate and frame devices, the spiral wound de-vices, the hollow fiber devices, the tube devices and the tubelet devices, all of which are well-known in the technical and patent literature. Since the purpose in any commercial fermenter is to obtain the highest capacity in terms of catalytic activity at the lowest cost, and since the cost of the en~yme is dependent upon other considerations, it is obviously advantageous to minimize the cost of the remainder of the equipment. The enzyme-coupled pressure-driven ultrafiltration membranes which are the objects of these in-ventions are unique in that devices can be made wherein the cost of the mem-branes, the cost of coupling and the cost of the mechanical equipmen~ will be, it is estimated, substantially less than the cost of the en~yme itself unless one is dealing with a particularly inexpensive enzyme. Since one can include in such systems large membrane area and volume in a relatively small and simple device, the pressure-driven system has many and obvious advantages over all other configurations.
One of the obvious disadvantages, if such it is, of the pressure- `
driven ultrafi~tration process involving coupled en~ymes is that solutions containing particulate matter, suspended solids and the like, whose diameters are larger than the diameter of the pore openings, cannot be processed. It is necessary, as a consequence, to have available a process which will render these crude process streams capable of treatment. For this purpose the - 11 - ,, . - . . . .

33~57 fixed-charge ultrafiltration membranes of Gregor (United States Patent 3,808,305) offer an ideal pretreatment and have been used for such purposes.
Another advantage of the teachings of this invention is that one can obtain a high catalytic capacity per unit volume of reactor. Membranes have been prepared where at least 20% and as high as 60% of the dry weight of the finished membrane is the weight of the enzyme itself. This high capa-city rises from the very large inner pore surface of the membrane and the pressure-driven method of coupling. The conventional microporous filter media which have pores in the micron Si7'e do not possess this large capacity and consequently systems made with their employment are not comparable to those of this invention.
The following examples are provided to more fully illustrate the invention. It will be understood that, because the examples are illustra-tive, they are not to be construed as limiting the invention, e~cept as de-fined by the appended claims. All compositions are parts by weight, except where otherwise expressly stated.
EXA~LE 1 A membrane was prepared from a 1:1 mixture of Kynar~ and cellulose in M~0-paraformaldehyde solution, allowed to dry for 2 min. in air~ covered and allowed to equilibrate in the solvent vapor for 1 hour and then coagulat-ed with water vapor and washed to have a ~hickness of 20 microns, a water content of 90% and a hydraulic permeability of 4.5 liters per hour at 50 psig pressure for an 11.3 cm2 area. Activation by cyanogen bromide (40 mg CNBr in pH 11 water) was performed at 50 psig for 25 minutes. The membrane was wash-ed for one minute in a 0.1 M phosphate buffer at pH 7.5 and then a solution consisting of 50 mg of the enzyme horse liver alcohol dehydrogenase in 100 ml of the same buffer was passed through at 50 psig. The membrane was then stored in this effluent for 24 hours at 4C, then washed with water and its protein content was fo~md to be 18% of the dry weight of the membrane.
The activity of this en~yme was determined using the standard Trade mark - 12 _ ... . . . . . . . .. . . .

~33~7 procedures described in the Worthington Biochemical Corporation Manual (Freehold, New Jersey, 1972). The free solution activity of this horse liver alcohol dehydrogenase was determined at pH 7.5 in the phosphate buffer em-ploying NAD as a coen~yme and with ethanol as substrate. The free solution activity of this enzyme was 0.31 units/mg and at 100 psig substrate pressure ~ -the activity of the coupled enz~ne was 0.19 units/mg.

Jack bean urease (Worthington Biochemical) was coupled to a Kynar~-cellulose membrane ~s prepared in Example 1 using cyanogen bromide activation ~mder 50 psig pressure. The membrane had a hydraulic perm~ability corre-sponding to an average pore diameter of 300 AU. This en~yme has a molecular weight of 483,000 and is sensitive to traces of metal being present, so all experiments were carried out in the presence of 0.001 M E~TA. After the mem-brane was washed with cold water for 1 minute after activation, a solution of 50 mg of urease in 100 ml of 0.02 M phosphate buffer at pH 7.0 was passed through at 25 psig. The membrane was then stored in the effluent from the coupling reaction at 4C overnight, then washed thoroughly with buffer. Its protein content showed that 55% of the weight of the dry membrane was pro-tein. Its activity was determined by titration for the ammonia formed as a result of the decomposition of urea following the procedure in the Worthing-ton Biochemical Corporation Catalogue (Freehold, New Jersey, 1972). Using this test, it was found that the activity of urease in the coupled state was 60% that of the free en~yme.
E~AMPLE 3 Glucose Qxidase from A. niger (l~orthington) was coupled to the cellulose-Kynar~ membrane of Example 1. After activation with cyanogen bromide and washing with cold water for one minute, 40 mg of glucose oxidase in 100 ml distilled water was passed through at 70 psig. The membrane was then stored in this effluent for 24 hours at 4C, then washed thoroughly.
Its protein content as determined by Lowry was 4.4 mg or 42% by weight of the Trade mark -~ . ~ , ' . . .

~ ,.......................... ~

~33~'~7 dry membrane. Its activity was determined by the Worthington procedure but in the absence of horseradish peroxidase because the hydrogen peroxide formed would not inhibit the enzyme but be removed under the pressure driven con-ditions employed. Accordingly, using the standard 18% solution of glucose as substrate and employing pure oxygen as pressurization gas so the oxygen con-tent would not be rate-limiting, the rate of production of peroxide was measured and found to be 80% of that of the native en~yme.

In another modification of the present invention, a penicillin acylase was coupled to a membrane containing maleic anhydride groups under pressure-driven conditions. A film was prepared from 2 parts of a copolymer 7 of vinylmethylether and maleic anhydride in equal molar amounts and combined with one part of Kynar dissolved in a mixture of hexamethylenephosphoramide and DMF together with an epoxide cross-linking agent, and the membrane cast therefrom was dried partially. After 10 minutes of drying in dry air at 60Cg the membrane was sprayed with a fine mist of toluene which caused a partial coagulation of the film, following which drying to effect a final cure was completed to produce a highly porous film. The coupling of penicil-lin ac~lase was achieved by dissolving 1 gram of this enzyme having a specif-ic activity of 15 units/mg in 100 ml of a pH 6 buffer, wetting the membrane wi.th this buffer which caused it to swell and then recirculating the enzy-matic solution through the membrane at 4C for a period of two hours at ~0 psig. The pH was kept constant by the addition of dilute base. After a few moments the membrane had swelled sufficiently so that the solution of en~yme could be pumped through the membrane readily. The membrane continued to swell and its hydraulic permability increased. The final membrane has a protein content of 30% and its enzymatic activity was 45% of the activity of the same amount of native en~yme. For the production of 6-AMP from penicil-lin G, 100 g of the substrate in 1 liter of water at constant pH 7.8 was passed through the membrane at 70 psig and at 38C for 2 hours. It was found that the yield of product was 88% of theoretical.

A penicillin acylase derived from ~ coli was coupled to the cellu-lose-Kynar'~ membrane of Example 1. This membrane was activated by treatment with cyanogen bromide (40 mg in 100 ml at pH 11 kept constant by the addition of base) at 70 psig, following which the membrane was washed for one minute, and was treated with a solution of the enzyme in a phosphate buffer at pH 5.5, and containing 5 grams of enzyme in 200 n~ of this solution. After pumping this solution through the membrane at 70 psig, followed by incubation over-night at room temperature and washing, the film obtained had 31% of its dry weight as en~yme protein. In a typical application of this membrane, 1 gram of the sodium salt of 7-(2~thienyl)-acetamidocephalosphoranic acid was treat- -ed at pH 7.5 at 37 with this en~yme~coupled membrane at 50 psig to produce 7-~SP in a yield of 98% a~`ter 10 minutes of reaction. In this experiment the effective area of the membrane was 11 cm2, the amount of enzyme bound was 3.5 mg and its activity was 60% that of the native enzyme in th~s reaction.

A carboxylic membrane was prepared by the Aissolution of 2 parts of polymethylacrylic acid to one part of polyv:inylidenefluoride (Kynar~, ~ennwalt~ together with an epoxide (0.1 part) cross-linking agent in a 10 solution of hexamethylenephosphoramide and DMF. The membrane was cast, allowed to dry partially at 60C, then coagulated with water vapor to in-crease its porosity and finally cured at 120C after drying at 60C was com-pleted. When ~etted with water, the membrane had a water content of 80% and a hydraulic permeability corresponding to a pore diameter of 180 A.U. To prepare the enzyme-coupled membrane, 5 volumes of a 25% aqueous glutaralde_ hyde solution and 5 volumes of diaminopentane were added to 50 ml of water, the total volume made up to 100 ml and the pH adjusted to 6. This solution was pumped through the membrane at room temperature at 70 psig for 2 hoursg ths membrane then washed with purs pH 6 buffsr, and then a 0.5% solution of Trade mark 3~

an acylase preparation obtained from ~. coli dissolved in 1 liter of pH 5 buffer was passed through the membrane for 3 hours at room temperature and at 60 psig. After washing with water it was found that the protein content of the membrane corresponded to 20% by weight of the dry membrane. The activity of the membrane was such as to correspond to a conversion rate of 200 micromoles per minute per gram of dry membrane of a 5% solution of K
benzylpenicillin at pH 7.8 at 37C, carried out at 30 psig~ and with greater than 95% conversion.

A membrane was prepared as in Example 1 but with a 2:1 ration of Kynar to cellulose to obtain a film of 180 AU pore diameter. After activa-tion with cyanogen bromide and one minute of washing with water, all at 70 psig, the membrane was treated with a solution of p-phenylenedia~ine (100 mg in 100 ml) at pH 8.0 for 10 minutes, also at 70 psig, then incubated over-night, all at room temperature and, in the case of the last treatment, in the dark. Then after washing with water and 1 M sodium chloride, and then water again, the membrane was diasotiæed by treatment with 2 N hydrochloric acid at 0C for 30 minutes. Then 5 ml of 14% sodiwn nitrite was added to the acid solution slowly, the membrane kept in this solution for one hour and then washed with 0.3% sulfamic acid at 0C. Then the membrane is treated with a solution of the enzyme glucose isomerase (40 mg in 100 ml of pH 8.5 buffer) at 0C at 30 psig for 1 hour of recirculation, followed by overnight incuba-tion at 4C. After washing with distilled water at 50 psig, the enzyme-coupled membrane was then treated with 0.5 sodium chloride solution. In use, the membrane was subjected to a 0.5 M solution of glucose at pH 8.0 at 60C
and at 30 psig to produce a fructose-containing efPluent. The original ac-tivity of this en~yme was 15 units/mg and of the coupled enæyme was 13 units/mg. The loading of enæyme was 18% by weight of the dry membrane.
In the process of the invenkion as illustrated hereinabove, any superatmospheric pressure can be employed to accelerate the passage of .;' .. .:' ' .

activating liquid and/or of enzyme through the membrane although a pressure of at least about 10 psig, especially about 30 to 120 psig, gives particular-ly good results. For the solution of material acted upon by the en~yme-coupled membrane, a pressure of at least about S psig is desirable. The pressure or potential, instead of being of a pneumatic or hydraulic type, can be of an electrical nature, i.e. as in the well-known phenomenon of electro-osmosis and/or electrophoresis wherein an electrical potential is imposed across a membrane or filter and combinations of the activation-coupling and use operations effected in that manner. Thus, for example, with chymotrypsin coupled following activation with cyanogen bromide a current which produces a flow of solution through the membrane comparable to that due to a pressure gradient of 70 psi produces a system whose en~ymatic activity is nearly the same as one prepared under 70 psig.
It will be appreciated that the instant specification and examples are set forth by way of illustration and not limitation, and that various -modifications and changes may be made without departing from the spirit and scope of the present invention.

- 17 - ~

,:
' ' .. . . . .

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for producing an enzyme-coupled ultrafiltration membrane comprising forcing a solution of an enzyme through an activated porous ultrafiltration membrane under a pressure equivalent to at least about 10 psig, the activated sites being reactive with said enzyme thereby to couple said enzyme to said membrane.

2. The process of claim 1, wherein the activated membrane is produced by forcing an activating agent through a membrane under a pressure equivalent to at least about 30 psig.

3. The process of claim 1, wherein said membrane has pores averaging in size from about 3 to 5 times the diameter of said enzyme.

4. The process of claim 3, wherein the activated membrane is produced by forcing an activating agent through said membrane under a pressure equivalent to about 70 to 120 psig, and the enzyme is forced through said membrane under a pressure equivalent to about 25 to 120 psig.

5. A process for producing an enzyme-coupled ultrafiltration membrane comprising forcing a solution of chymotrypsin or .beta.-galactosidase under a pressure equivalent to at least about 10 psig through a porous cellulosic ultrafiltration membrane activated at various sites with cyanogen bromide, the activated sites being reactive with said enzyme thereby to couple said enzyme to said membrane.

6. The process of claim 5 wherein said membrane has pores averaging in size from about 3 to 5 times the diameter of said enzyme, and wherein the activated membrane is produced by forcing an activating agent through said membrane under a pressure equivalent to about 70 to 120 psig, and the enzyme is forced through said membrane under a pressure equivalent to about 30 to 120 psig.

7. An ultrafiltration membrane having pores whose surfaces are lined with an enzyme coupled thereto, said pores having an average size from about 3 to 5 times the diameter of said enzyme.

8. A cellulosic ultrafiltration membrane having pores whose surfaces are lined with an enzyme coupled thereto selected from the group consisting of chymotrypsin or .beta.-galactosidase, said pores having an average size from about 3 to 5 times the diameter of said enzyme.

9. A method of using the ultrafiltration membrane of claim 7 for effecting an enzymatic reaction comprising forcing a substrate solution through said membrane under a pressure equivalent to at least about 5 psig, said substrate solution containing a material which is capable of being converted by said enzyme, whereby said enzyme-coupled membrane effects said enzymatic conversion of said material.

10. The process of claim 9, wherein said enzyme contains a plurality of proteases, said material being contaminated with at least one protein or polypeptide, whereby said contaminant is enzymatically converted.

11. The process of claim 9, wherein said enzyme is an amylase, and said material is soluble starch, whereby said starch is enzymatically converted into glucose.

12. The process of claim 9, wherein said enzyme is glucose isomerase, and said material is glucose, whereby said glucose is partially enzymatically converted into fructose.

13. The process of claim 9, wherein the solution of glucose is produced by passing soluble starch through a membrane coupled to an amylase.

14. The process of claim 9, wherein said enzyme is a penicillin-acylase, and said material is penicillin, whereby the penicillin is converted into penicillanic acid.