CA2158056A1 - Method of coupling ligands within porous supports (p.e. azlactone) and uses thereof - Google Patents

Abstract

A method is disclosed for coupling a ligand within a porous support. The method involves mixing ligand and porous support under conditions sufficient to suppress coupling conditions of the ligand to the porous support while enhancing the relative rate of diffusion, to the rate of reaction, of the ligand into and within the porous support, and then altering the conditions to enhance rapid coupling of the ligand within the porous support. The alteration from diffusion conditions to coupling conditions involves a change in the reaction solution of pH, ionic strength, temperature, or coupling competitor, such that a relatively lower Thiele Modulus during diffusion conditions changes to a relatively higher Thiele Modulus during coupling conditions. Derivatized porous supports produced according to the method are also disclosed. The derivatized porous support has enhanced functional efficiency. Derivatized porous supports prepared from azlactone-functional porous supports are also disclosed.

Description

~ W O 94122918 21 ~ 8 ~ 6 PCTAUS94/03927 METHOD OF COUPLING LIGANDS WITHIN POROUS SUPPORTS (P.E. AZLACTONE) AND USES
THEREOF

Field of the Invention This invention relates to an improved method of covalently r immobilizing ligands to su~ and products produced from the methocl.

10 Background of the Invention The use of biologically active subst~nces, such as proteins, is çnh~ncell when such substances are covalently immobilized, i.e., coupled as lig~n-l~, ontosupports. Separation techniques such as affinity chro.,latogl~phy are based on the ability of the coupled ligand to bind specific, targeted biologically active substances 15 from a mixture of other materials. Common examples of affinity chromatographytechniques include the binding of immunoglobulins using coupled proteins and thebinding of antigens using coupled antibodies.
Successful ligand coupling is based on two factors: quantity immobilized and quality of immobilization. Quantity immobilized, ~Ap~essed as 20 density per unit volume of support, is an in-lic~tor of the amount of ligand coupled regardless of the quality of that immobilization. In fact, most protein coupled in highly dense regions of a support is biologically inactive. That is a waste.
Quality of immobilization, e,Ll)ressed as bound specific biological activity, is an in~i~tor of the amount of ligand coupled onto a support in a manner 25 that causes the ligand to retain its biologically activity. Maximi7in~ bound specific biological activity is desirable. However, there must be enough ligand density to achieve practical utility.
The optimal condition in ligand coupling would be the maximum amount of ligand that is coupled with maximum bound specific biological activity.
30 That results in optimal ligate binding or functional efficiency of the coupled ligand on the support.
Thus, for purposes of this application, "functional efficiency" means the combination of acceptable quantity of ligand coupling with acceptable bound specific biological activity.

wo 94/2~9~g PCT/US94/03927 Most ligand candidates are large molecules that have specific conformations necessary to retain biological activity. In the case of antibody binding of antigen, low antigen binding efficiencies have been attributed to the concerted actions of surface density of antibody, multi-point ~tt~chment of antibody to porous S supports, undesirably restrictive conform~tiQn~ imposed by covalent ~tt~chmtont steric effects, and orientation effects. See Velander et al., "The Use of Fab-Masking Antigens to Fnh~nce the Activity of Immobilized Antibodies", Biotechnology and Bioengineering, Vol. 39, 1013-1023 (1992) which ~i~los~s how enh~nced functi~n~lefficiency was achieved when the Fab portion of a monoclonal antibody was masked10 with synthetic antigens prior to covalent immobilization of the antibody on the support, followed by unm~cking.
F.nh~nced bound specific biological activity was disclosed in PCT
Publication WO 92/07879 (May 14, 1992) by the use of polyanionic salts in concentrations of greater than about 0.5M during covalent immobilization of 15 biologically active materials to azlactone functional ~u~ . Preferably, in addition to polyanionic salts, amounts of azlactone quencher were added during covalent immobilization to compete with the biologically active material during covalent immobilization .
Notwithct~n-ling these advances in the field of ligand coupling 20 technique, there remains the problem of optimi7ing functional efficiency of coupling expensive and precious biologically active substances as ligands to supports.
The problem of optimal functional efficiency has not been solved by others. For example, U.S. Pat. No. 4,968,742 (Lewis et al.) uses an elaborate, stepwise method to couple ligands involving derivatizing a polymer with an activating 25 agent to introduce a couplable functional group, with the derivatization ~lro~,l-ed in the presence of a blocking agent which is reactive with the same functionality on the polymer as the activating agent, in order to control the number of couplable functional groups for covalent immobilization of ligand.
U.S. Pat. No. 4,839,419 (Kramer et al.) discloses a method for 30 adsorbing a protein onto a support and then crosslinkin~ the protein to the support where the reaction conditions for coupling do not differ from those of the adsorption.
U.S. Pat. No. 4,775,714 (Hermann et al.) discloses a two-step process of immobilization of biologically efficient compounds on a carrier involving the steps 2 ~ ~ 8 ~ 5 6 PCT/US94/03927 of hydrophobic interaction and covalent immobilization. Examples 5-7 therein disclose the stepwise addition of a solution of inorganic salt, in a concentration of between 0.5 M to 3.0 M, to a reaction vessel cont~ining the biologically efficiçnt compound and the carrier, followed by a slow reaction (40 hours at 40C under moderate sh~king) in order to produce an immobilized, biologically active col-,poùlld.
There have been efforts to experim~nt~lly determine how coupled protein was distributed to activated porous su~lls. See Stage et al. Biochimica et Biophysica 343, 382-391, (1974), where uniform distribution was reported for immunoglobulins coupled to cyanogen bromide-activated Sel)harose branded beads.
See also, Lasch et al. Eur. J. Biochem. 60, 163-167 (1975) which re~lled that uniform distributions of ferritin were found, except when the CNBr activation ofSepharose branded beads was very high ( > 50 mg/ml) and/or when coupling efficiency was higher than 90~. Thus, non-uniform distribution resulted from high coupling efficiencies.
While non-uniform distribution of immobilized enzymes on porous ~u~polls has been studied using alterations in reaction conditions, and, in some cases, found preferable, others have warned of the possibility that non-uniform antibody immobilization on Sepharose branded beads was responsible for the loss of binding activity with increase of average antibody density. See Tharakan et al. J. Chrom.
522, 153-162 (1990).
Another sought to improve the ~;,rol"-allce of an immobilized enzyme by "kinetic control" during the immobilization by using salt concentration and time to control the enzyme distribution to an ionic support and chPrnic~lly couple with a reagent added to fix this distribution. See Borchert et al. Biotechnolo~y and Bioen~in~ring (26)7, 727-736 (1984). Another has studied reaction conditions andproposed a restriction effect at openings of pores in the porous s~lp~lL that could prevent further protein from being coupled to the support. See Clark et al.
Biotechnology and Bioengineering 26(8) 892-900 (1984).
Generally, the art has found that one can achieve coupling efficiency by reacting at conditions to achieve a high coupling capacity at the expense of loss of bound specific biological activity. Alternatively, the art has found that one can achieve high bound specific biological activity with lower coupling efficiency (because the total amount of ligand coupled is lowered.) In both of these _~ :19- 4-~5: ~o ~3a ~ '8 '~1Yy- +~ )Yl~
~ 2i5805~

circu~s~nces, the ~action candilio~s w~re noe a~ter~d durin~ thc re~;tion process ~7 e~pt as dcscribed in U.S. Pat. ~o. 4,77~,714 (~lerm~nn et al.) in ~mr~ 7. In that circumstanu~, salt was allded stepw~c afLer the Li,gand solution and the support ~erc combined hut ~fore tl~ r~tion commenced for 40 hours at 40C
Ptlbli~;ation WO-A-~ 9Q7 618 di~closes a mcthod of prepa~ing an afflnity tnatri~ of mod~fied polysa~ch~ support~ wher~ a "on~-po~" sysl~m is u~cd to po3yn en~e monome~s in lhc pres~nce of polys~h~n~e follQwed by lirl~dng of ~y~ro~y group~ of the polys~c~h~ritt~ to lh~ ~orm~d po1ym~ tempcraturc used in the Iinking is hi~her ~an the t~m~erature used in polyn~e~ iun of the morloJnen~.
1~ Summar~ of the l~vention ~ he present ~nven~ion pro~ s a rapid m~hod of co~/~lenl.
immobili~alion ~a~ ~urpr~singly enh~nce~ ~unctional c~fi~;icncy ~f a biologied~ly active substance ~ a ligall~ coupled 10 ~urfacos ~itl~in a p~rous suppurt. 'rhe method effectiYcIy distributes the ~gand within a support prio~ to coupl~g o~ lig~nd o~to 15 the supp~rt. Ihe m~thod employs a two-step coupling procc~,s wher~ reacti~n conAi~ion~ are al~red ~;I.Wee~1 st~P,ps, a~ld pref~rably whcre no ~nmobili~t~ age?~ i~
added IxLwee~l st~ps.
~ result? thc supporl ~l~ri~d~ d wi'~h Lh~ ligan,l opL~ functionaleffic~:~cy with a tar~ d biologically acti~ su~s~nce.
~) The ~Lr.st s1,~p of lhc method ~uppresses con(li~olls or ~ r re~c~on belween ligand alld a porou~, support~ wa~rh P.nh51n~P~ the rcla~ rate of di~usion, to ~ r~te of r~action, ~ the li~nd into ~d w~thin th~ support The sec~nd RtCp of thc method e--h~ s coup3ing condi~ns, such Lhat t~e 1i~and couE~1es t~ the ~3upport rapidly" i.e., wilhin about fou~ hours st~ that the l~ d ~ouples to ~e support be~orc :he ligand has a~ oyp~J~ y ~ ~a~ate ~e desi~d loca~an for c~upli~g, ~ e m~thod achie~es ~ lig~~ ;csup1~g that a~roids surfacc c~ow~ g that wo~l1d othe~se inhibit or lawsr ~hc bio1ogical activily of ~e ligand. Couple~ ~igand on a d~iva1i~cd s~pport ~Lep&~d u~ hc melhod of the presc~lt in~fention ~s rem~ h]y mv~ llniron~ spa~ially ~istributed than a deriYati7Pd ~upport usin~
30 prs~iouAly knuwn methods. The r~n~ g d~n~ ed sup~rt a~hie~ ahout a 1.2~-fold to lO-fu1d in~,ase in fi~nCti~n~1 Pff;(;iPnry than se~n when pre.viously l~nown ~oupl~ng m~ods haw b(:en eInployed.

2~ 2~81'~9- +-~9 89 2399~465:# 1 :19- 4-~5 : 2~:4~ : . 2~2824819~ +49 89 2~99446h:~ 7 ~ 8056 A reat~lre of Ihc ~rese~t invention is th~ ease of ligand coupl~ng in a ~i mannor ~hat Optill~i~S ~unccion~l ~fficien~y.
Ano~er ~ ure o t}le przsent i~v~ti~ is the increagcd efficiency of ~se ~f pr~cious or expensi~re bio]uglcally ac~ivc s~l~sta~ccs such a~ a~ds ~r S soupli~g onto a suppcrt.

-4a-~WO 94/22918 2 ~ 5 8 0 ~ 6 PCT/US94/03927 An advantage of the present invention is a dramatic reduction of the amount of derivatized support needed to achieve a given bioch~mic~1 procçs~ing capacity provided by the immobilized biologically active substance for any givenamount of ligand coupled per volume of support. Conversely expressed, the advantage is that the use of a given amount of derivatized support ~r~aled according to the present invention will dr~m~tir~lly increase the biochPmir~1 procç~ing capacity. Either way, dramatic improvement to biosh~mic~1 pr~ces~ing capacity isrealized by the present invention.
Another advantage of the present invention is the savings in related costs (e.g., the expense of proce~ing fluids such as aqueous buffers) of bioch~mic~1 proces~ing capacity due to the unexpectedly enh~nced functional efficiency of the derivatized support.
Another advantage of the present invention is that the method provides a more uniform spatial distribution of ligand coupled to support. This advantagelS results in a decrease in spatial density of ligand coupled to the support, making greater use of electrophilic function~1ity residing at the surfaces of the support. This advantage also ~ll~ s a larger average density of ligand coupled to the support with increased amounts of ligand, without overcrowding of ligand at or near the exterior surfaces of the support.
Thus, the invention provides a method for coupling a ligand within a porous support that comprises the steps of mixing ligand and porous support under conditions sufficient to suppress coupling of the ligand to the porous support while enhancing the relative rate of diffusion, to the rate of reaction, of the ligand into and within the porous support, and altering conditions to enhance rapid coupling of the ligand within the porous support.
In another aspect of the invention, the method improves functional efficiency of binding a ligate otherwise deleteriously affected by restricted diffusion into a porous support or by steric effects of binding to a ligand coupled to the porous ~uppo. L. The steps of the method comprise coupling a ligand to the porous support according to the two-step diffusion/coupling method in~ ~ted above, whereby spatial distribution of ligand coupled to surfaces of the supports çnh~nces functional efficiency of ligates otherwise affected by restricted diffusion or steric effects, and binding the ligate to the spatially distributed ligand, such that functional efficiency of WO 94/22918 2 ~ 5 8 ~ 5 ~ PCT/US94/03927 ~
the ligand is greater than the functional efficiency of the ligand coupled in circumstances where either restricted diffusion into the porous support, or steric effects of the ligate binding, or both, are present.
The invention also provides a derivatized porous support produced according to the method of the present invention.
The invention also provides a derivatized porous support comprising Protein A coupled to the porous support in a coupling efficiency of at least about 80% and having a functional efficiency relative to binding immunoglobulins of greater than about 3.0 bound IgG/coupled Protein A. It is contemplated that an increase of greater than 15% fi~nction~l efficiency can be achieved at coupling den~isies of from about 6 mg to about 15 mg of Protein A coupled per ml of swollen or hydrated support.
The invention also provides a derivatized porous support comprising an antibody against ligate protein (e.g., Protein C) coupled to the porous support in a coupling efficiency of greater than about 70% and having a functional efficiencyrelative to binding ligate protein (e.g., Protein C) to the coupled anti-ligate protein (e.g., Protein C antibody) of greater than about 20% at a molar ratio of 2:1 of ligate protein to antibody. It is contemplated that greater than 15% functional efficiency can be achieved at coupling den~ities of from about 3 mg to about 10 mg of antibody coupled per ml of swollen or hydrated support.
The invention also provides a derivatized porous support comprising ligand coupled to the porous support in a manner that at least 30% of the amount of coupled ligand is coupled to internal surfaces, i.e., those surfaces that are within 70%
of the geometric center of the support. Expressed ~ltern~tively for a spherical particle, the internal surfaces are those surfaces within a sphere having a radius of 70% of the total radius of the particle.
The invention also provides a derivatized porous support comprising ligand coupled to the porous support in a manner that has a percentage of permeation of coupled ligand of about 30% to the internal surfaces of a porous support.
For a greater appreciation of the invention, embodim~nt~ of the invention are described following a Brief Description of the Drawings.

Brief Description of the Drawings ~WO 94/22918 2 1 5 8 ~ 5 ~ PCT/US94/03927 Fig. 1 is a com~alison fluorescence micrograph of a derivatized porous support, shown in cross-section, produced according to prior art methods, that shows an uneven distribution of Protein A ligand coupled principally at outer surfaces of the support, demonstrating an undesired "halo effect".

Fig. 2 is a fluorescence mi~;,ogldph of a derivatized porous support, shown in cross-section, produced according to the present invention, that shows a more even distribution of Protein A ligand coupled throughout all surfaces of the support,enh~ncing functional efficiency.
Fig. 3 is a comparison fluorescçnce micrograph of the same derivatized porous support as in Fig. 1, shown in cross-section, that shows an uneven distribution of immunoglobulin bound to the Protein A principally at outer surfaces of the support, demonstrating an undesired "halo effect" for functional efficiency.
Fig. 4 is a fluorescence micrograph of the same derivatized porous support as in Fig.
2, shown in cross-section, that shows a more even binding of immunoglobulin to Protein A ligand coupled throughout all surfaces of the ~u~", proving enh~n~Rcl functional efficiency.
Embodiments of the Invention Porous Supports Acceptable porous supports for use in the present invention include 25 those commercially available for affinity chru,l~at~g,dphy techniques. The porous support can be any porous solid, whether natural or synthetic, organic or inorganic, having a porous structure and which is insoluble in water or aqueous solutions.
Suitable solids with a porous structure have pores of an average ~ meter of at least 30 nanometers and a pore volume of over 0.1 cm3/g. Preferably, the average pore 30 ~i~meter is at least 50 nm because larger pores will be less restrictive to diffusion.
Preferably, the pore volume is at least 0.5 cm3/g for greater potential capacity due to greater surface area surrounding the pores.
Nonlimiting examples of such porous supports are naturally or WO 94/22918 2 ~ PCT/US94/03927 synthetically-modified natural compositions such as polysaccharides, celluloses, and agaroses. An example of a commercially available agarose is Sepharose branded beads from Pharmacia AB of Uppsala, Sweden. Such porous su~ require activation with a composition, such as cyanogen bromide or cyanotransfer agents, in S order to covalently immobilize lig~n-l~. Such activation is re~uired prior to using the method of the present invention.
Other nonlimiting examples of such porous supports are synthetic homopolymers and copolymers of acrylates, methacrylates, acryl~mi-les, vinyl aromatics, and vinyl alcohols. Desirably, such homopolymers or copolymers have a10 functionality (e.g., azlactone, aldehyde, or the like) or are modified to provide a functionality to permit rapid, direct covalent reaction with ligands to form derivatized supports. Oxirane or epoxide functionality is not suitable for methods of the present invention because such groups are not rapid in coupling reaction or require adversely high pH for rapid coupling. See Yarmush et al., Biotech Adv. 10, 413-446 (1992).As seen in Example S of U.S. Pat. No. 4,775,714 (Hermann et al.), the coupling reaction time was 40 hours at 40C under moderate ~h~king. That is not a rapid coupling condition, i.e., within about four hours.
Other nonlimiting .oy~mplçs of such porous sup~lLs are porous inorganic particles such as porous glass, silica, ~hlmin~, ~ir~onilll,l oxides, and other 20 metal oxides.
Porous supports can be membranes, porous fibers, webs, or particles, such as beads. Preferably, porous supports useful in the present invention are reactive porous particles so that the ligand can be covalently coupled to the support.
In addition to these supports described above that require activation 25 prior to use according to the present invention, other sul ~lls have covalently reactive surfaces without need for an interme~i~te activation step. Preferably, such directly reactive porous supports are particles.
Directly reactive porous particles useful in the present invention are generally of two broad types: chemically modified inorganic particles and organic, 30 polymeric particles.
The inorganic particles may be, for example, metal oxides such as alumina, silica, and zirconia; glass beads, glass bubbles, and controlled pore glass;
and the like. These particles are chemic~lly modified by methods such as coating ~VO 94122918 21~ ~ ~ 5 6 PCT/US94/03927 with a polymer (usually organic) which contains a reactive functional group or by reaction with a suitable reagent (e.g. an alkoxy silane coupling agent) containing the reactive functional group.
The organic particles may be crosclink~d or noncrocclink~d polymers which have been pr~aled, for example, by polymerization or copolymerization of amonomer con~ ng the appro~liate reactive functional group, by coating a particlesupport as described above, or by çhemic~l mo lifi~tiQn of another polymer to introduce the reactive functional group.
A number of useful particles are commercially available or can be prepalt;d by techniques well known in the art, a partial listing of which can be found below in Table 1.
Directly reactive particles useful in the present invention can have a spherical shape, a regular shape, or an irregular shape. Size of reactive particles can vary widely within the scope of the invention and will depend to some extent upon the intended use of the particles.
Generally size of reactive particles ranges from 0.1 micrometers to 5 millimeters in average ~ meter.
Whether directly covalently reactive or indirectly activated via a composition such as cyanogen bromide, the covalent reactive functional groups which are useful for the purposes of the invention can be cl~ccifi~ in general as electrophiles. Reaction with a nucleophile (e.g. amine, alcohol, or me~
produces a covalent chemical bond either by an addition reaction or by a displ~emPnt or substitution type reaction (in which a byproduct molecule is released). Addition type reactions are pre~lr~d.
Examples of useful directly and indirectely reactive functional groups and examples of commercially available particles cont~ining them are listed in Table 1.

WO 94/22918 2 ~L 5 8 ~ 5 6 PCT/US94/03927 c C -S '' ~ ~ E

C. o o .~
æ

V ~
E E

K ',~ , C

Z V~ ~ ¢

~WO 94/22918 2~ 5 8 ~ ~ 6 PCT/US94/03927 Particularly preferred as reactive particles useful in the present invention are particles having azlactone-functional groups on internal and/or external surfaces of such particles. Thus, such reactive particles have an azlactone-functional group of Formula I:
S

~/ ~(CH2~ I

S~ ;

wherein:
R' and R2 independently can be an alkyl group having 1 to 14 carbon atoms, a cycloalkyl group having 3 to 14 carbon atoms, an aryl group having 5 to 12 ring atoms, an arenyl group having 6 to 26 carbon atoms and 0 to 3 S, N, and nonyeloxidic O heteroatoms, or R' and R2 taken together with the carbon to whichthey are joined can form a carbocyclic ring cont~ining 4 to 12 ring atoms, and n is an integer 0 or 1.
Azlactone-functional reactive particles are particularly yre~lled in the present invention because such particles rapidly and directly covalently couple ligands better than commercially available reactive functional groups shown in Table 1.
Further, such azlactone-functional groups are quite stable prior to covalent coupling with a ligand. Further, covalent coupling of a ligand with an azlactone-functional group causes no displacement of a byproduct molecule, which avoids undesired purification of the composite article after covalent coupling of the ligand.
Also, azlactone-functional groups are known to possess high covalent coupling capacities with biologically active materials such as Protein A. Further, such high covalent coupling capacities with Protein A also yield high specific bound biological activity of Protein A as the coupled ligand. Thus, an azlactone-functional reactive particle is particularly ylefell~d for use in the present invention.
Azlactone-functional polymeric particles can be made, for example, by wo 94,229l8 2 ~ 5 8 n ~ 6 PCT/US94/03927 copolymerization of a (meth)acryloylamino acid with a variety of other free radically polymerizable comonomers followed by reaction with a cyclizing agent, as described in U.S. Patent Nos. 4,737,560 and 4,871,824 or by copolymerization of an alkenylazlactone with other comonomers as described in Euro~ Patent Publication 0 392 5 735, which are both incorporated herein by reference. Azlactone-functional particles can also be p,epared by solution coating an azlactone-functional polymer onto anorganic or inorganic particle, also as described in above mentioned Eul~an Patent Publication 0 392 735.
Azlactone-functional reactive particles can also be made from azlactone graft copolymers which are disclosed in U.S. Patent 5,013,795 and Eur~l Patent Publication 0 392 783.
Size of particles of azlactone-functional particles can be from about 0.1 to 1,000 micrometers and preferably from 0.5 to 250 micrometers. Dry azlactone-functional particles can have an average pore size ranging from about 1 to about 300 15 nanometers and preferably from 5 to about 200 nanometers. Azlactone-functional particles can have an average pore volume of at least 1.0 cm3/g of particle. In a particle having a size of 50-80 micrometers, a pore volume of at least 1.2 cm3/gprovides a pore volume of about 60% of the particle volume. In the same particle, the surface area is at least 50 m2/g. Thus, there is substantial surface area within an 20 azlactone-functional particle available for covalent immobilization according to the present invention.
Most preferably, porous supports useful for the present invention are EmphazeTM brand porous azlactone-functional activated affinity chrolllalog~d~hy beads commercially available from Minnesota Mining and Manufacturing Company of St.
25 Paul, MN.

Ligands for Covalent Immobilization As stated above, reactive functional groups on porous :~Up~)lLS are desirably electrophiles. Thus, for direct covalent immobilization, ligands useful in 30 the present invention contain nucleophiles.
Nonlimiting examples of ligand functional groups include primary and secondary amines, alcohols, and melcal)lans. Of these, amine-functional ligands are especially preferred.

~WO 94/22918 21 5 8 0 ~ G PCT/US94/03927 T.ig~n~ls useful for the preparation of adduct composite articles can also vary widely within the scope of the present invention. Preferably, a ligand is chosen based upon the contemplated end use of the derivatized porous support.
Once ligands are coupled according to m~tho.1s of the present invention, such ligands are available for biological or chemical interaction with an enhanced functional efficiency, such as adsorbing, complexing, catalysis, or reagent end use.
Derivatized porous supports are useful as adsorbants, complexing agents, catalysts, reagents, as enzyme and other protein-bearing supports, and as chromatographic articles.
In a plefelled aspect of the present invention, the ligand desired for covalent immobilization is a biologically active substance or compound having nucleophilic-functional groups. Nonlimiting examples of biologically active materials are substances which are biologically, immllnochemic~lly, physiologically, or pharmaceutically active. Examples of biologically active materials include proteins, peptides, polypeptides, antibodies (monoclonal or polyclonal), antigenic subst~nces, enzymes, cofactors, inhibitors, lectins, hormones, ~cel)lo,s, coagulation factors, amino acids, histones, vit~mins, drugs, cell surface lllalkel~, and substances which interact with them.
Of the biologically active materials, proteins, enzymes and ~ntigenic substances are desired for covalent immobilization. Nonlimiting examples of proteins, enzymes, and antigenic substances include natural and recombinant Protein A (ProtA), Immunoglobulins such as rat (rIg), human (hIg), bovine (bIg), rabbit (rbIg), and mouse (mIg), Concanavalin A (ConA), Bovine Serum Albumin (BSA), Thyroglobulin (TG), Apoferritin (Af), Lysozyme (Ly), Carbonic Anhydrase (CA), and Bacterial Antigen (BA). Nonlimifing examples of coagulation factors include Protein C (ProtC), heparin, fibrinogen, and thrombin.
Uses for coupled proteins, enzymes and antigenic substances are disclosed in European Patent Publication 0 392 735. Uses for coagulation factorsinclude activation of zymogens to active proteins, such as Protein C to Active Protein C by coupled thrombin.
The presently preft;lled biologically active substances are antibodies and ProtA.

~:~58g~6 WO 94/22918 . - PCT/US94/03927 Alternatively, a derivatized porous support of the present invention can comprise a coupled enzyme to catalyze a chemical transformation of substances recognized by the enzyme. Also, a derivatized porous support comprising a coupled antigenic substance can be utilized for affinity purification of a correspondingS antibody from a complex biological fluid.
In another example, porous particles having Protein A coupled to internal and external surfaces according to the method of the present invention can adsorb biologically active materials such as Immunoglobulin G for affinity separations processes. In another example, a derivatized porous support can be used for immobilization of antibodies or be used for immuno li~gnosti~s or for Western blotting.
Presently plerell~d azlactone-functional groups will undergo nucleophilic attack by amines, thiols, and alcohols. Thus, ligands having at least one amine, thiol, or alcohol group thereon are candidates for covalent immobilization in an azlactone-functional porous support.
The method of the present invention is particularly useful for binding large ligates that are otherwise affected by restricted diffusion into a porous support or by steric effects of binding to a densely coupled ligand, or both.
Without being limiting, large ligates can be ch~ terized as those ligates which would be unable to traverse a pore already having ligate bound to coupled ligand at opening surfaces of the pore. Thus, coupled ligand within the derivatized support are under-utilized or possibly un-utilized during binding ifopening surfaces of the pore coupled with ligand bind ligates and block passage to derivatized surfaces within the porous support.
Also, large ligates can be characterized as those ligates which adversely alter the binding of additional ligates due to steric effects. Thus, binding of the ligate can occur but inefficiently (e.g., in a manner that causes a waste of coupled ligand). If that ligand were more uniformly spatially distributed, the support could accommodate both a greater binding capacity and a greater functional efficiency.
The "largeness" of the ligate is a relative consequence of the pore size of the porous support, the reactivity and density of coupled ligand, and other factors.
Nonlimiting examples of large ligates are those biologically active materials i(lçntifie~l ~VO 94/22918 2 ~ 5~ ~ 6 PCT/US94/03927 above as ligand c~nr~ tPs, except for small moleculçs such as peptides, polypeptides, amino acids, and drugs.

Method of Covalent Immobilization S The method of the present invention involves two steps. The first step brings the ligand in close proximity to surfaces of the porous support. The second step causes rapid covalent immobili7~tion, i.e., coupling within about four hours, of the ligand to surfaces of the porous support. An advantage of the present invention is the superior functional efficiency achieved by coupling ligands according to themethod of the present invention.
The first step employs, in a single reaction vessel, conditions sllfficient to ~Up~lG~S coupling of ligand to the suppo.L while enh~ncing the relative rate of diffusion, to the rate of reaction, of the ligand into and within the porous support.
Thus, the first step suppresses the rate of ligand coupling relative to the rate of ligand diffusion.
Nonlimiting examples of su~ essing coupling conditions in the reaction vessel include use of specific ranges of conditions in one or more of acombination of pH, ionic strength, le",pe~ture, and coupling colllpeLi~ . P.ese"tly pr~îe..ed as con-lition~ to ~uppr~ss coupling are a control of pH of a reaction solution and/or control of ionic strength, in which the ligand and the porous support aremixed and otherwise in condition for coupling. The pH can be controlled to be in a range from about 3 to about 7. A pH of that range provides diffusion conditions that minimi7~ the reaction of a nucleophilic group on a ligand with an electrophilic group on the surface of a porous support.
More preferably, the pH during diffusion should be from about 4 to about 6 because of the stability of the electrophilic functional groups to hydrolysis is greater as pH increases.
Most preferably, the pH during diffusion is about 5, because electrophilic functional groups are more stable to possible hydrolysis.
When pH of the reaction solution ranges from about 3 to about 7, other reaction conditions can be as conventionally employed in the art of immobilization. In other words, adjustm~nt of the pH alone can be sl-fficient tosuppress coupling conditions while also enhancing the relative rate of diffusion, to the WO 94/22918 2 ~ ~ g ~ ` PCT/US94/03927 rate of reaction, of the ligand into and within the porous support.
Ionic strength due to the presence of polyanionic salts (e.g., sulf~t~s, phosphates, citrates, tartrates, ana the like), during diffusion is minimi7~ to ~uppl~SS
coupling conditions. Preferably, the molarity of the polyanionic salts in the reaction 5 solution during diffusion is from about 0.01 to about 0.4 M.
Coupling competitors, described in greater detail below, which would otherwise compete for nucleophilic reaction with the functional groups can be 0 to 2 Molarity.
Te~ re of the reaction solution during diffusion needs to be 10 controlled to slow the reaction rate of ligand coupling. Preferably, the t~lllpel~tllle can be from about the freezing point of the aqueous solution to about 25C.
The reaction solution usually includes buffering agents. Buffering agents for aqueous media include acetate, phosphate, pyrophosphate, borate, and other salts known to those skilled in the art, such as those buffering agents disclosed in Good et al., Biochemistry, 5, (1966) p. 467 et seq.
The concentration of buffering agents in aqueous media can range from about 10 mM to about 750 mM and desirably from about 50 mM to about 200 mM, inclusive, depending on the concentration of biologically active substance chosen for coupling and the concentrations of other optional ingredients that can affect the ionic 20 strength of the reaction solution.
The duration of the first step should be sufficient in length to assure diffusion of the ligand into proximity with all surfaces of the porous support. The amount of time can vary according to the type of porous support employed, the pore size and pore volume of the porous support, the size and conformation of the ligand 25 to diffuse through the pore volume of the porous support and other physical considerations. Generally, a diffusion duration of at least 5 minutes is sufficient to accomplish acceptable diffusion. Desirably, diffusion lasts at least 10 min~ltes to improve diffusion. Preferably, diffusion lasts at least lS minutes to assure diffusion in most porous supports in the case of smaller support geometries. For larger 30 support geometries, the diffusion lasts to at least the characteristic diffusion time, which is the mean support thickness divided by the diffusivity of the ligand of interest in aqueous solution.
The second step causes coupling of ligand to surfaces of the porous ~WO 94/22918 ~15 ~ ~ 6 PCT/US94/03927 support to occur rapidly and assuredly. Thus, reaction conditions in the second step are abruptly altered from that of the first step and preferably occur in the absence of adding any coupling agent to the reaction solution. A "coupling agent" means a reagent that reacts with either the ligand or the support to improve the coupling of the S ligand to the support but does not mean a coupling competitor, i.e., a reagent that competes for reaction sites on the porous support.
The rate of coupling is a function of the rate constant of coupling, the concentration of ligand, reactivity of the functional groups per unit area of the support, the rate of diffusion of ligand into the support, and the te~ dlure.
When diffusion conditions are established using pH of the reaction solution, a change in pH comprises the second step.
Coupling conditions are enh~nced when pH of a reaction solution is changed to a pH within one pKa of the nucleophilic ligand, usually within a range from about 7 to about 10. This range causes rapid and assured coupling of ligand to the porous support.
A pH of this range provides coupling conditions that maximize the reaction of a nucleophilic group on a ligand with an electrophilic group on the surface of a porous support.
More preferably, the pH of the reaction solution for coupling should be from about 7.5 to about 9.5 so that hydrolysis or other reaction with solvent islessened.
Most preferably, the pH of the reaction solution for coupling is about 8.5 to m~int~in the biological activity of ligands, especially protein~ eous lig~n-lc.
The alteration in reaction conditions from the diffusion step to the coupling step can be limited to a change in pH or can also include other changes.
In addition to, or in substitution for a change in pH, a change in ionic strength of the reaction solution can also be used to enh~nce functional efficiency of the ligand coupled to the porous support. The amount of change in ionic strength can be in an amount sufficient to enhance coupling of ligand and to enhance functional efficiency of the coupled ligand.
The amount of change in ionic strength from the diffusion step to the coupling step can range from about 0.5M to about 1.5M. Preferably, the amount of wo 94/22918 215 8 ~ ~ 6 PCT/USg4/03927 ~
change in ionic strength can range from about 0.6M to about 1.2M to m~int~in thesolubility of ligands, especially protein~eous lig~n~c. Most preferably, the change in ionic strength of the solution in the coupling step can be about 1.0-1.2M.
The change in ionic strength is induced by an addition of polyanionic 5 salt or salts to the reaction solution. Suitable polyanionic salts, both inorganic and organic, are i~entifi~l in PCT WO 92/07879 and U.S. Pat. No. 5,200,471 (Coleman et al.).
As explained in Coleman et al. regarding plerelled azlactone-functional supports, using an inorganic polyanionic salt in aqueous bllrrered media to covalently 10 immobilize protein to azlactone-functional polymeric supports more than doubles the bound specific biological activity of the biologically active m~teri~l, when co~ ~ed to using an inorganic monoanionic salt, such as NaCl.
This enhanced efficiency of immobilization is achieved in a very rapid and facile reaction. The use of an inorganic polyanionic salt in high concentrations 15 does not disrupt other valued aspects of using ~7l~ctone-functional polymeric supports such as the very rapid covalent immobilization rates achievable at ambient ~c;ldtures.
Of the inorganic polyanionic salts, sulfates are desired because of increased bound specific biological activity relative to the molar concentration of 20 inorganic polyanion in the aqueous media. If inorganic polyanionic salts are used, use of Na2SO4 is presently plel~l-ed when coupling proteins (that have activity unaffected by metallic cations) in an aqueous medium buffered at a pH from aboutpH 4 to about pH 9. Sulfates are also plerelled to phosphates because a lower molar concentration of slllf~tes than phosphates is n~ceC~ry to achieve the same density of 25 coupled ligand on the azlactone-functional polymeric support. Evidence of this advantage may be found in Coleman et al. J. Chromatogr. 512 (1990) 345-363.
Preferably, organic polyacids and their salts can provide even more productive and efficient coupling of ligands on preferled azlactone-functional polymeric supports than inorganic polyanionic salts. Organic polyanionic salts are 30 more con~i~ten~ly ionic than inorganic polyanionic salts in a pH range of pH 7 to pH
9 where most covalent immobilizations are conducted and in the plefelled range of pH alteration from the diffusion step to the coupling step of the present invention.
Thus, organic polyanionic salts have a higher ionic strength per mole of polyanion.

O 94/22918 ~ ~ ~i 8 ~ 5 6 PCT/US94/03927 ~W
Consequently, fewer moles of organic salt are frequently required for covalent immobilization. Further, a greater variety of organic polyanionic salts than inorganic polyanionic salts is sufficiently soluble in buffered aqueous media employed in covalent immobilizations. Thus, organic polyanionic salts are presently plefe,led to 5 inorganic polyanionic salts.
Of the organic polyacid c~n~ tPs, di-acids, tri-acids, and tetra-acids, or their salts are desired. Nonlimiting examples of salts of such acids include malonate, malate, and tartrate di-acid salts of alkali metals, citrate tri-acid and nitrilo-tri-acetic acid (NTA) salts of alk;lli metals, and ethyl~o-nYli~min~tçtraacetic acid 10 (EDTA) tetra-acid salts of alk~li metals. The presently prefel,~;d organic polyanionic salt is sodium citrate.
In addition to, or in substitution for, a change in pH and/or a change in ionic strength, an addition of a coupling co~l~pelitor to the reaction solution can also be used to enhance functional efficiency of the ligand coupled to the porous support.
The coupling competitor can be added in an amount sufficient to enh~nce bound specific biological activity of ligand (but not in an amount whichsubst~nti~lly reduces the amount of ligand coupled) in a manner resulting in ~onh~nced functional efficiency of the coupled ligand.
The type of coupling co",pelilor can vary according to the nature of 20 the ligand to be coupled to the porous support. The kinçtic~ of reaction (as influenced by pH, ligand concentration, l~m~;,~ re, ionic strength of the reaction solution, among other factors) between the porous support and the ligand determine the amount and type of coupling co",~lilor to be used.
While not being limited to a particular theory, a coupling competitor 25 competes for the reactive sites on a porous support where ligand would otherwise couple. The reduction in the number of reactive sites can limit the possibility that ligand couples in a manner that alters its conformation and reduces or elimin~t~s its biological activity. Unexpectedly, it is believed that a coupling co",pelilor enh~n-~s functional efficiency of the coupled ligand by providing a sparcity of reactive sites 30 without elimin~ting too many reactive sites for coupling. This also tends to result in a more uniform or effective distribution of coupled ligand.
When ~lcrel,ed azlactone-functional porous supports are used, coupling competitors are azlactone quenchers. Suitable azlactone quenchers are also Wo 94/22918 2 ~ ~ 8 ~ 5 6 pcT/uss4lo3927 ~
identif1ed in PCT WO 92/07879 and U.S. Pat. No. 5,200,471 (Coleman et al.).
Nonlimiting examples of azlactone quenchers for use include ethanolamine, bovine serum albumin, casein lysate, hydroxylamine, ethylamine, ammonium hydroxide, glycine, ammonium sulfate, butylamine, glycinamide, TRIS, S gelatin, lysozyme, non-fat dry milk, beta-mt;,caploelhanol, mercaptoethylether, dithiothreitol, glutathione, arginine, gl-~ni~ine, lysine, ~ min~s~ and combinations thereof. Some of these nonlimiting examples include proteins "irrelevant" to theimmobilization desired.
The co~ ntration of azlactone quencher to be added with a change in 10 pH and/or an increase of ionic strength for the second step of the method can range from about 0.1 M to about lO M. Desirably, the range may be between about 0.5 M
to about 2 M. When ethanolamine serves as azlactone quencher, the concentration may range from about 0.1 M to about 1 M. The presently preferred concentration of ethanolamine as azlactone quencher is about 0.5 M to about 1 M.
In addition to, or in substitution for, a change in pH, a change in ionic strength, or the addition of a coupling co",~litor, or a combination of them, anincrease in the lt;l~ ture of the reaction solution can also be used to çnh~nce functional efficiency of the ligand coupled to the porous support. The amount oftemperature change can be an amount sufficient to çnh~nce coupling of ligand, so20 long as the speed of the reaction is rapid, and to enh~n~e functional efficiency of the coupled ligand.
The temperature increase can be about 10C to about 35C, and preferably about 20C to about 30C because an increase in temp~ re of that amount increases reactivity of the ligand to the support without deleteriously affecting 25 the biological activity of the ligand.
While not being limited to a particular theory, one or more of an increase of pH, an increase in ionic strength, an increase of le",~~ re, or an addition of a coupling competitor to the reaction solution causes ligand to couple in a manner which retains bound specific biological activity of the ligand due to a 30 resulting more uniform or effective distribution of ligand which minimi7es restricted diffusion or steric effects of ligates aLl~ )ting to bind to the coupled ligand. This improves functional efficiency of the res-llting coupled ligand by minimi7ing the number of coupled ligands that are not biologically active and ma~imi7ing the amount :=~

~WO 94/22918 ~ ~ 5 8 ~ 5 ~ PCT/US94/03927 of ligand coupled.
Coupling during the second step is rapid and assured. The duration of the step can range from 0.5 to 4 hours.
Coupling is completed by quenchin~ any rem~ining reactive sites with 5 an addition, in excess, of a quencher that couples to effectively all of the rem~ini~g reactive sites on the porous support.
When the porous support is ælactone-functional, the qllen~h~r employed can be any of the azlactone qlle-n~hers identifie l above in concentrations in excess.
While not being limited to a particular theory, the method of the present invention utilizes the advantages of controlling the Thiele modulus during the diffusion and the coupling steps of the mPthod Thiele modulus broadly expresses in dimçn~ionless terms the ratio of the rate of reaction to the rate of diffusion and can be expressed using the following equation (I):
~ = Rp (ka/De)l'2 (I) where Rp is the Reactivity of the ligand, k is the first-order rate constant, a is the internal surface area, and De is the effective Diffusivity of the ligand.
Another expression of Thiele modulus is the following equation (II):
~ = 1/2(dp) (V~"p/KmDe)l'2 (II) where dp is the carrier di~meter, V~x is the activity, p is density, Km is kinetic parameter for coupling, and De is effective Diffusivity. See Borchert et al.
publication identified above.
The first step of the present invention is to SLI1~PIeSS coupling conditions to enhance the relative rate of diffusion to the rate of coupling. The first step utilizes a lower Thiele modulus (where rate of diffusion is appreciably greater than rate of reaction) relative to the second step. Conversely, the second step utilizes a higher Thiele modulus (where rate of reaction is appreciably greater than rate of diffusion) than found in the first step.
Thus, expressed alternatively, the method of the present invention provides a permeation step for more uniform spatial distribution having conditions using relatively low Thiele modulus followed by a coupling step for rapid coupling wo 94/22918 2 l ~ 8 ~56 PCTtUS94/03927 ~
which uses conditions having a relatively high Thiele modulus.
Relatively low Thiele modulus conditions are achievable with a low pH, low ionic strength reaction medium, lower le~ ture, coupling competitor, and combinations of them.
Relatively high Thiele mod~ ls conditions are achievable with a high pH, high ionic strength reaction meAillm, higher le~ dture, coupling co",~ to~
and combinations of them. ~
The impact of these factors upon Thiele modulus for ligand diffusion and coupling ~epen-1s on the ligand chosen. For example, functional efficiency of Protein A coupled to an azlactone-functional bead can be enhanced with a change in Thiele modulus to initiate the coupling step using an increase in ionic strength, an addition of coupling coll,~litor, a combination of those two factors. However, achange in pH alone will result in çnh~nced functional efficiency of anti-Protein C
antibody on a porous azlactone-functional bead.
At any given ligand concentration range in solution, one of the results in controlling the difference in Thiele modulus betwe~en the first and second step is a change in the activation energy necç~,y for a coupling of ligand to a function~lgroup on a surface within a porous support. The con-lition~ of the lower Thiele modulus conditions raise the activation energy required for the coupling re~ction7 while the conditions of the higher Thiele modulus lower the activation energy required for the coupling reaction.
The abruptness of the step change in Thiele modulus conditions after the diffusion step and to initiate the coupling step of the method results in minim~l back-diffusion of ligand from the support. This ,.,~ much of the dispersed spatial distribution achieved during the diffusion step. Further, the reaction kin~ics of the functional group on the surface of the porous support çnh~nces the rapidity of the coupling reaction before the desired more uniform spatial distribution of the ligand within the support is lost. Rapidity of the coupling reaction should be less than about 4 hours and preferably less than about 1 hour.
Usefulness of the Invention The method of the present invention provides a process of immobilizing biologically active substances on porous supports in such a fashion as to ~WO 94/22918 215 8 0 5 6 PCT/US94103927 avoid surface crowding which inactivates a ~ignifir~nt portion of the coupled ligand.
The derivatized support has coupled ligand with a molecular sparcity which has afunctional efflciency that is ~ignific~ntly higher than ligand coupled with a higher surface density.
S Figs. 1 and 2 provide a direct colllp~ on of the the advantages of the present invention.
Fig. 1 is a co,l,l)~ison fluoresc~nre micrograph of cross-sections of a derivatized porous support produced according to a method used in the prior art and i(ientifi~l in Comparison Example 6 below, where there is no al~",pl to provide a two step reaction of diffusion and then coupling. The cross-section shows an uneven distribution of ligand coupled at outer surfaces of the SU~)~)ll. This is evidence of overcrowding that the method of the present invention avoids.
Fig. 2 is a fluoresc~nre microgld~ of cross-sections of a derivatized porous s~,l produced according to the method of present invention and spe~ific~lly according to Example 7 below where there was a change in pH betweenthe diffusion step and the coupling step. Throughout the porous cross-section of the bead, a significantly more even distribution of ligand coupled throughout all surfaces of the sup~.L is found. With a molecular ~ily of coupled ligand at the outer surfaces, and a distribution of coupled ligand throughout all surfaces of a porous suppo,l, an optimum of coupling is achieved.
As seen in a co",pa,ison of Figs. 1 and 2, a more uniro~lll spatial distribution of coupled ligand is achieved such that at least 30Yo of ligand is coupled to internal surfaces within 70% of the geo",el,ic center of the su~ll. Thus, themethod of the present invention provides a controllable, more uniform spatial distribution of coupling of the ligand to the porous suppo,l.
Derivatized porous sul)l)o, l~ of the present invention have a sparcity of coupling of the ligand to outer surfaces of the porous support and have an enh~n~ ed coupling of the ligand to inner surfaces of the porous ~upl)oll.
Functional efficiency of coupled ligand can increase as much as 1.1 to 10 fold using the method of the present invention over functional efficiency achieved using prior m~tho ls. As such an unexpectedly superior derivatized support is achieved.

WO 94/22918 2 ~ 5 8 ~ 5 6 PCT/US94/03927 ~
Fig. 3 is a l;o~ alison fluorescence micrograph of cross-sections of a derivatized porous support shown in Fig. 1 with binding of immllnoglobulin, produced according to a method used in the prior art and icl~ontifi~d in Comparison Example 8 below. The cross-section shows an u-neven distribution of ligate binding 5 to ligand coupled at outer surfaces of the support. This is evidence of restricted diffusion and steric effects that demonstrates the "halo effect" of inadequate functional efficiency.
Fig. 4 is a fluoresc~nce micn)gldph of cross-sections of a derivatized porous support shown in Fig. 2, produced accordillg to the method of present 10 invention and specifically according to Example 9 below where there was a ligate binding more uniformly throughout the bead and an avoidance of restricted diffusion and steric effect. Throughout the porous cross-section of the bead, a ~i~nifi~ntly more even distribution of ligate bound to coupled ligand throughout all surfaces of the support is found. With binding of ligate distributed on coupled ligand throughout 15 all surfaces of a porous support, an optimum of functional effi~iency is achieved.
For a greater underst~n-ling of the scope of the invention, the following examples are provided.

Examples 20 Com~arison Example 1.
This example describes the coupling of Protein A to EmphazeTM
Biosupport Medium AB1 using a one step method of the prior art. To triplicate 15ml screw capped polypropylene tubes con~ g 50 milligr~m~ each of dry ElllphazeTM beads (Minnesota Mining and ~nllf~ctllring Company, St. Paul, MN) 25 2.5 milliliters of a solution of natural Protein A (Ferment~h, Ltd., Edinburgh, UK) at 1.6 milligrams per millilitt-.r in 1.2 M Na2SO4, 0.1 M NaH2PO4 pH 7.5 was added and mixed to wet all of the dry beads. The reaction Illix~ule was then ~git~ted by end-over-end rotation for a total of 75 Ill;rl~les at room t~",pe,d~ure. After centrifugation at 1200 x g for 10 minutes, the supernatant was removed and analyzed 30 for rs~"~ining Protein A by colorimetric analysis (BCA method from Pierce Chemic~l Company, Rockford, IL). The amount of Protein A coupled and the efficiency of coupling was det~llllined by a difference calculation. The results of the coupling are shown in Table A. The beads were ~uenched with 4.0 millilitPrs of 3 ~WO 94/22918 21 ~ 8 ~ ~ 6 PCT/US94/03927 M ethanol amine at pH 9.0 for about 18 hours- (overnight). The beads from the combined triplicates were then sequentially washed with phosphate buffered saline (PBS: 0.025 M NaH2PO4, 0.15 M NaCl pH 7.4), 0.2 M sodium acetate pH 5.0, 0.5 M sodium bic~bollate pH 8.5, and PBS on a sintel~d glass fritted funnel S (porosity D) using 20-30 volumes of each solution.
The washed beads were then tested for their immunoglobulin binding capacity by packing them into a 3 x 50 mm Omni glass column and running a chr~.l,atog,dm with purified human IgG (Sigma Ch~mic~1 Colllpal y, St. Louis, MO) as the test solution. A total of 48 milligrams (16 mil1ilittors of 3 milligrams per milliliter in 0.0l M NaH2PO4 pH 7.5) was loaded at a flow rate of 0.57 mi11i1iters per minute followed by a wash of 6.8 millilit~rs of loading buffer, 6.8 mi11i1iters of 2 M NaCl, 0.0l M NaH2PO4 pH 7.5, 4.6 milliliters of loading buffer, and the bound IgG was eluted with 4.6 milliliters of 0. l M glycine, 2% acetic acid pH 2.2. Elution fractions were collected and the amount of IgG present determined by absorbance at lS 280 nm. The results for these beads is shown in Table A.

Comparison Example 2.
This example describes the coupling of Protein A to Emphaze beads in a manner similar to that of Comparison Example l according to the prior art with the 20 exception that the coupling solution con~ a coupling co,.,~lilor, 0.5 M
Tris(hydro~Lylllethyl)aminometh~ne (TRIS) in addition to sodium sulfate and phosphate buffer at pH 7.5.
The results of the coupling and the IgG capacity evaluation (as described in Comparison Example l) are shown in Table A.
Fxample 3.
This example describes a two step method of the present invention of coupling Protein A to Emphaze beads as a direct colllp~ison to Comparison Example l.
Thus, the final solution conditions are the same as those of Comparison Example l 30 but there is a dirrerel t ionic strength of the reaction Illixlult; during the first step. In the second step the ionic strength of the reaction llli.XLU~ iS increased. To triplicate 50 milligram samples of dry Emphaze beads in lS milliliter screw capped polypropylene tubes 0.5 milliliters of Protein A solution (8 milligrams per milliliter WO 94/22918 ~ ~ ~ 8 ~ ~ G ~ ; PCT/US94/03927 in 0.05 M NaH2PO4) was added at room tempelature and allowed to hydrate the beads for 15 minutes. Then in a second step 2.0 millilit~rs of 1.5 M Na2SO4 0.125 M NaH2SO4 pH 7.5 was added and the reaction continued with agitation as in Example 1 for an additional 60 mim~t~. The coupling analysis and IgG capacity were carried out as in Comparison Example 1 and the results are shown in Table A.

Example 4.
This example describes the coupling of Protein A to Emphaze beads using a two step method of the present invention as a direct co",p~ison to Comparison Example 2. In the second step, the ionic strength of the reaction mi~lure is increased, a coupling co",~lilor that q~lencl~s azlactone is added, and the pH is increased. The method is similar to that of Example 3 with the exceptions that the Protein A is in 0.05 M sodium acetate pH 5.0 and the solution added in the second step contains 0.5 M TRIS in addition to the sodium sulfate and phosphate buffer at pH 7.5. The Protein A coupling and IgG binding capacities were determined as in Comparison Example 1 and those results are shown in Table A.

Comparison Example 5.
This example describes the coupling of Protein A to Emphaze beads using a two step method of the present invention where only the pH ch~ng~s between the first and second step. This is a co."~uison example because it shows that use of a different pH from that of Examples 3 and 4, and a change in pH alone from the first step to the second step does not work for the ligand Protein A. The method is similar to that of Example 3 with the exception that the Protein A was in 0.05 Msodium acetate pH 5.0 and the second step uses 0.125 M boric acid solution pH 9.5.
Protein A coupling was determined as in Comparison Example 1 and the results areshown in Table A.

wo 94/22918 215 g ~ ~ 6 PCT/US94/03927 TABLE A

Results of Protein A Coupling to r ,h~.e Bead~
Protein A Coupling IgGFunctional S Example Coupled Efficiency Capacity Efficiency (mg/ml) (~) (mg/ml) (IgG/PA) 110.3 (0.2)99.0 24.0 2.44 29.4 (0.2)97.9 26.2 2.43 39.7 (0.1)99.2 29.1 2.99 0 49.9 (0.1)98.6 38.3 3.96 0.0 0.0 The functional efficiency of both two step methods (Examples 3 and 4) of the present invention increased when coll,p~cd with the one step methods of the prior art (Comparison Examples 1 and 2, respectively). The increases range from 22%-62%.
There is also an additive effect of an increase in ionic ~ nglll and addition of an azlactone quencht,r in the second step. That increase is 32%.

Comparison Example 6 and Example 7.
These examples describe the p,c;~ ion of Fmph~7~ beads coupled to fluorescently labeled Protein A for use in deLell"il1ing the distribution of the protein that results from dirrele,lt coupling m~tho ls according to the prior art for Comp~ri~on Example 6 and according to the present invention for Example 7.
Protein A was tagged with Texas Red (Sigma Ch~mi~l Company) by the method of Titus et.al. in J. Immunological Methods ~Q, 193-204 (1982) such that a solution of 5 mg/ml in PBS resl-lt~. This was mixed with a unlabeled Protein A such that a final concentration of 25 mg/ml re,s-llte~. This solution was then diluted for coupling to F.mph~7~ beads according to the method as in Comparison Example 1 for Comparison Example 6 and according to the method as in Example 4 for Example 7 using 20 milligrams of beads and 1.0 milliliters of total solution. Coupling effiçiencies and the amounts of Protein A coupled (norm~li7ecl to 1 ml of beads) were found to be es~nti~lly the same as those for the unlabeled Protein A in Comparison Example 1 and Example 4.

WO 94/22918 2 ~ 5 g ~ 5 6 PCT/US94103927 Samples of these beads were then cast in JB4-Plus embedding me~ m (PolySciences, Warrington, PA) and sectioned with a glass knife into sections about 4 microns in thickness. These were then mounted, e-~mined under fluorescent conditions at a m~gnific~tiQn of 500 X, and photographed. Figures 1 and 2 show 5 ~i~nific~ntly different spatial distributions of the Protein A coupled to E~ ha~
beads. Since both ~r~aldtions have the same total Protein A content, the "halo"
distribution of Protein A in Fig. 1 (Comparison Example 6 beads) in~ tes that the ligand is concenl,~ted into ~ignific~ntly less volume than the more uniform distribution of Fig. 2 (Example 7 beads). The Protein A is coupled to the bead in a 10 manner that at least 70% of the amount of coupled Protein A is coupled to intern~l surfaces of the bead that are within 35% of the geometric center of the bead.

Comparison Example 8 and Example 9.
These examples describe a method of delellllining the spatial distribution of fluorescently labeled human IgG bound to Emphaze beads with Protein A coupled. Aliquots of about 100 microliters of beads from Examples 1 and 4 wereincub~t~ with FITC-labeled human IgG (Sigma Ch~mi~l Company), about 1.5 milli~r~m~ in 0.5 milliliters of PBS, overnight at room l~lll~,alur~. The bead 20 samples were then washed five times with 2.0 milliliters of PBS each to remove unbound antibody.
The beads were then pç~aled for microscopy as in Comparison Example 6 and Example 7. The photomicrographs of these beads can be seen in Figs. 3 and 4 where the distributions of bound ligate are similar to the coupled ligand 25 in Figs. 1 and 2. This in-~ic~tes that the more uniform distribution of Protein A
ligand seen in Fig 2 results in a more uniform distribution of bound IgG ligate (Fig 4) and a higher capacity than the "halo" distribution of Protein A (Figs. l and 3) and the lower IgG capacity of Comparison Example 1 beads. This also conL,,Ils that the method of coupling of the present invention can have an effect on the spatial 30 distribution of ligand which can determine the functional efficiency of the r~sl-lting biologically active support. Thus, in this example where the support is a subst~nti~lly spherical particle and the ligate is bound to ligand coupled to intern~l surfaces within a radius of 70% of the total radius of the particle. The ligand is ~WO 94/22918 ~ 1 ~ 8 ~ ~ 6 PCT/US94/03927 coupled to the porous bead in a manner that has a ~ereel,tdge of permeation of coupled ligand of about 70% to the internal surfaces of a porous support.

Comparison Example 10.
This example describes the coupling of Protein A to cyanogen bromide activated S~harose branded agarose using a one step method of the prior art.
CNBr-Activated Sepharose 4B (Pharmacia LKB, Biotechnology AB, Uppsala, Sweden) was prel)ar~d for reaction according to the mAnllf~c~turer's instructions and aliquots of slurry equivalent to 0.6 milliliters of gel were placed into 15 milliliter 10 screw capped polypropylene tubes. After removing the ~ ~, A~ t buffer solution, 3.75 millilit~rs of 1.0 milligrams of Protein A per milliliter in 0.5 M NaHCO3 pH
8.5 was added and the mixture ~git~t~d by end-over-end rotation for a total of 75 minutes at room le,l,~ldlllre. At that time the slurry was cen~iruged at 1200 x g for 5 min~ltes and the supernatant removed for de~llllinalion of the amount of coupling of Protein A as in Comparison Example 1. Results are shown in Table B. The gel was resu~pended in 4.0 milliliters of 3 M ethanol amine pH 9.0 and ~git~ted overnight (about 18 hours) at room lelll~ldtuf~. The gel was then sequentially washed with PBS, 1 M NaCl in PBS and PBS using 8 milliliters of each solution.
These beads were then evaluated for the binding capacity of the coupled Protein A by the method of Comparison Example 1. Results are shown in Table B.

Comparison Example 11.
This example describes the coupling of Protein A to cyanogen bromide activated Sepharose using a two step method where pH alone was adjusted from thefirst step to the second step. Slurry aliquots (0.6 milliliters) of CNBr-Activated Sepharose 4B were ~r~ d as in Comparison Example 10 and reacted with 0.75 milliliters of 5.0 milligram Protein A per millilit~r, and 0.05 M sodium acetate pH
5.0 at 4 degrees C for 15 minutes. In a second step 3.0 milliliters of 0.625 M
sodium bicarbonate pH 8.5 was added and the reaction Illixlul~ ~git~ted by end-over-end rotation at room te"~ ture for an additional 60 minutes. The lllixlur~ was then treated as in Comparison Example 10 to determine the amount of Protein A coupledand its binding capacity. Results are shown in Table B.

WO 94/22918 2 ~ 5 ~ ~ ~ 6 PCT/US94/03927 ~
Comparison Example 12.
This example describes the coupling of Protein A to cyanogen bromide activated Sepharose using a one step method of the prior art similar to that of Comparison Example 10 with the exception that the coupling solution contained 1.0 M Na2SO4, 0.5 M NaHCO3 pH 8.5. Protein A coupling results and binding capacities were determined as in CQmparison Example 1 and are shown in Table B.

Example 13.
This example describes the coupling of Protein A to cyanogen bromide activated Sepharose using a two step method of the present invention similar to that of Example 11 with the exception that the solution used in the second step colllailled 1.25 M Na2SO4, 0.625 M NaHCO3 pH 8.5. Protein A coupling results and binding capacities were determined as in Comparison Example 1 and are shown in Table B.

Example 14.
This example describes the coupling of Protein A to cyanogen bromide activated Sepharose using a one step method similar to that of Comparison Example 10 with the exception that the coupling solution contained 1.0 M Na2SO4, 0.5 M
NaHCO3, 0.4 M TRIS pH 8.5. Protein A coupling results and binding c~p,~citi~oc were determined as in Comparison Example 1 and are shown in Table B.

Example 15.
This example describes the coupling of Protein A to cyanogen bromide activated Sephal()se using a two step method of the present invention similar to that of Comparison Example 11 with the exception that the solution used in the secondstep contained 1.25 M Na2SO4, 0.625 M NaHCO3, 0.5 M TRIS pH 8.5. Protein A
coupling results and binding capacities were determined as in Comparison Example 1 and are shown in Table B.

~WO 94/22918 21~ 8 0 S 6 PCT/US94/03927 TABLE B

Re~ult~ of Protein A Couplin~ to CNBr-Activated Beads Protein A Coupling IgGFunctional S Example Coupled Efficiency Capacity Efficiency (mg/ml) (%)(mg/ml) (IgG/PA) 5.6 89.631.0 5.54 11 5.6 89.630.9 5.52 12 5.9 100.034.8 5.90 0 13 5.9 100.037.4 6.34 14 2.8 48.3 19.4 6.93 4.0 69.031.1 7.78 Functional efficiency increases for the two step m~thods (Examples 13 and 15) when individually co~ ued to their one step counlel~ (Comparison Examples 12 and 14, respectively). The functional efficiency increases for F.Y~mrles 13 and 15 are 7% and 12%, r~s~li~ely, over Comr~ri~on Examples 12 and 14. These two step methods of the present invention also can be seen to produce a superior result to coupling under the m~nl~f~tllrer's l~coullllended con.iitiQn~ with either a one or two step method (Comparison Examples 10 and 11). The coupling efficiency of the one step method with TRIS present (Comparison Example 14) is ~i~nific~ntly less thanthe two step method with TRIS present (Example 15), but both are less than the other examples of coupling Protein A to CNBr-Activated Sepharose 4B even though functional efficiency is greater from Example 15 over ComI ~ri~on Example 14.

~xam~le 16 1 mg of 7D7B10-Mab (obtained from American National Red Cross) is incllb~t~d with 125 mg EmphazeTM beads at pH 4.0 and the solution is allowed to permeate the beads for 10 mins at 4C in the presence of 0.5M Tris. After the first 10 minute incubation with 0.5 M Tris, the salt concentration is raised to 0.8 M
Na2SO4 at pH 4.0 and allowed to permeate the beads for 10 mins at 4C. The pH isthen increased to pH 9.0 with several drops of lN NaOH. The reaction at pH 9.0 is WO 94/22918 215 ~ ~ 5 6 PCT/US94/03927 ~
allowed to proceed for 40 mins at 4C. The total permeation/diffusion and reaction time is 60 mins at 4 C. The supernatant is pipetted off. Residual reactive sites are blocked with 4 ml of 1.0 M ethanolamine in 0.05 M sodium pyrophosphate, pH 9.3 for 30 mins at RT. Beads are allowed to settle and the supern~t~nt is pipetted off.

5 An additional 4 mls of blocking solution is combilled with the beads and incub~t~A
for 60 mins at RT. Upon completion of the second blocking step, the beads are washed with four column volumes of 0.5 M NaCI and equilibrated with loading buffer for protein immunosorption. The 7D7B10-Mab coupling efficency to the azlactone is greater than 70%. 0.07 or 0.7 mgs of recombinant human Protein C in3 ml of 0.125 M Tris, 0.1 NaCl, 25 mM EDTA at pH 6.5 is loaded batchwise and eluted columnwise at a linear velocity of 1 cm/min into a 1.0 cm by 10.0 cm length glass chromalo~l~hy column cont~inin~ the 7D7Bl0-Mab:~7l~ctone beads at 4C.
The protein C is eluted from the immunosorbent at 1 cm/min linear velocity with 4.0 ml of 0.125 M Tris, 0.1 NaCl, 25 mM CaCl2 at pH 6.5. The immunosulbellt 15 functional efficiency is about 16%. The results of this Example 16 can be comp~ed with the results of Comparison Example S (Protein A) where an adj~ of pH
between the first and second steps did not yield coupling.

~xample 17 10 mg of 7D7B10-Mab is in~;ub~t~ with 125 mg r......... --l-h~7eTM beads at pH 4.0 and the solution is allowed to permeate the beads for 10 mins at 4C in the presence of 0.5M Tris. After the first 10 minute incubation with 0.5 M Tris, the salt concentration is raised to 0.8 M Na2SO4 at pH 4.0 and allowed to permeate the beads for 10 mins at 4C. The pH is then increased to pH 9.0 with several drops of lN
NaOH. The reaction at pH 9.0 is allowed to proceed for 40 mins at 4C. The totalpermeation/diffusion and reaction time is 60 mins at 4 C. The ~llpell,at~t is pipetted off. Residual reactive sites are blocked with 4 ml of 1.0 M ethanolamine in 0.05 M sodium pyrophosphate, pH 9.3 for 30 mins at RT. Beads are allowed to settle and the supernatant is pipetted off. An ~ lition~l 4 mls of blocking solution is combined with the beads and incub~ted for 60 mins at RT. Upon completion of the second blocking step, the beads are washed with four column volumes of 0.5 M NaCl and equilibrated with loading buffer for protein immnnQSorption. The 7D7B10-Mab coupling efficency to the azlactone is greater than 70%. 0.07 or 0.7 mgs of ~WO 94122918 2 ~ 5 8 ~ 5 6 PCT/US94/03927 recombinant humdn Protein C in 3 ml of 0.125 M Tris, 0.1 NaCl, 25 mM EDTA at pH 6.5 is loaded batchwise and eluted columnwise at a linear velocity of 1 cm/min into a 1.0 cm by 10.0 cm length glass chlonlatography column co,~li.ini,~g the 7D7B10-Mab~ tone beads at 4C. The protein C is eluted from the immunosorbent at 1 cm/min linear velocity with 4.0 ml of 0.125 M Tris, 0.1 NaCl,25 mM CaC12 at pH 6.5. The immunosorbent functional efficiency is about 16%.

Example 18 1 mg of 7D7B10-Mab is incub~t~d with 125 mg E.,.phazeTM beads at pH 4.0 for 10 mins at 4C in the presence of 0.5M Tris. After the first 10 minute incubation with 0.5 M Tris, the pH is then increased to pH 9.0 with several drops of lN NaOH.
The reaction at pH 9.0 is allowed to proceed for 50 mins at 4C. The total permeation/diffusion and reaction time is 60 mins at 4 C. The ~u~llat~lt is pipetted off. Residu~l reactive sites are blocked with 4 ml of 1.0 M ethanolamine in 0.05 M sodium pyrophosphate, pH 9.3 for 30 mins at RT. Beads are allowed to settle and the Su~llld~l is pipetted off. The 7D7B10-Mab coupling efficency is greater than 50%. An additional 4 mls of blocking solution is combined with the beads and incubated for 60 mins at RT. Upon completion of the second blocking step, the beads are washed with four column volumes of 0.5 M NaCl and equilibrated with loading buffer for protein immllnosorption. 0.07 or 0.7 mg of recombinant human Protein C in 2.0 ml of 0.125 M Tris, 0.1 NaCl, 25 mM EDTA at pH 6.5 is loaded batchwise and eluted columnwise at 1 cm/min linear velocity into a 1.0 cm by 10.0 cm length glass chr~"lato~,~dphy column con~ g the 7D7Bl0-Mab:~7l~ctone beads at 4C. The protein C is eluted from the immunosorbent at 1 cm/min linear velocity with 4.0 ml of 0.125 M Tris, 0.1 NaCl, 25 mM CaCl2 at pH 6.5. The immunosorbent functional efficiency is about 14%.

Bxample 19 10 mg of 7D7B10-Mab is incub~t~d with 125 mg azlactone beads at pH 4.0 for 10 mins at 4C in the presence of 0.5M Tris. After the first 10 minute incubation with 0.5 M Tris, the pH is then increased to pH 9.0 with several drops of lN NaOH.
The reaction at pH 9.0 is allowed to proceed for 50 mins at 4C. The total permeation/diffusion and reaction time is 60 mins at 4 C. The ~upelllal~lt is WO 94/22918 2 ~ ~ 8 ~ ~ ~ PCT/US94/03927 ~
pipetted off. ~Ç~idl-~l reactive sites are blocked with 4 ml of 1.0 M ethanolamine in 0.05 M sodium pyrophosphate, pH 9.3 for 30 mins at RT. Beads are allowed to settle and the supernatant is pipetted off. The 7D7B10-Mab coupling efficency isgreater than 50%. An ~ddition~l 4 mls of blochng solution is combined with the 5 beads and incubated for 60 mins at RT. Upon completion of the second blocking step, the beads are washed with four colum~ volumes of 0.5 M NaCl and equilibrated with loading buffer for protein immllnosorption. 0.07 or 0.7 mg of recombinant human Protein C in 2.0 ml of 0.125 M Tris, 0.1 NaCl, 25 mM EDTA at pH 6.5 is loaded batchwise and eluted columnwise at 1 cm/min linear velocity into a 1.0 cm by 10.0 cm length glass cl.ru~ og,~l-hy column cont~ining the 7D7B10-Mab:azlactone beads at 4C. The protein C is eluted from the immunosorbent at 1 cm/min linear velocity with 4.0 ml of 0.125 M Tris, 0.1 NaCl, 25 mM CaCl2 at pH 6.5. The immunosorbent functional efficiency is 14%.

Ex~mple 20 20 mg of 12A8-Mab is inc~lb~t~d with 350 mg EmphazeTM beads at pH 4.0 for 10 mins at 4C in the presence of 0.5M Tris. After the first 10 minute incubation with 0.5 M Tris, the salt co~centration is raised to 0.8 M Na2SO4 at pH 4.0 and allowed to permeate the beads for 10 mins at 4C. The pH is then increased to pH9.0 with several drops of lN NaOH. The reaction at pH 9.0 is allowed to proceed for 40 mins at 4C. The total permeation/diffusion and reaction time is 60 mins at 4 C. The supernatant is pipetted off. Residual reactive sites are blocked with 10 ml of 1.0 M ethanolamine in 0.05 M sodium pyrophosphate, pH 9.3 for 30 mins at RT.
Beads are allowed to settle and the s~ ldt~nt is pipetted off. The 12A8-Mab coupling efficency is greater than 70%. An additional 10 mls of blocking solution is combined with the beads and incubated for 60 mins at RT. Upon completion of the second blocking step, the beads are washed with four column volumes of 0.5 M NaCl and equilibrated with loading buffer for protein immunosorption. 3.0 mg of recombinant human Protein C in 30 ml of 0.125 M Tris, 0.1 NaCl, 25 mM EDTA at pH 6.5 is loaded columnwise at 1 cm/min linear velocity into a 1.0 cm by 10.0 cmlength glass chromalogldphy column conl~inil-g the 12A8-Mab ~7l~ tone beads at 4C. The protein C is eluted from the immunosorbent at 1 cm/min linear velocity ~WO 94122918 2 15 8 0~ PCT/US94/03927 with 15.0 ml of 0.1 M NaHCO3, 0.15 M NaCl at pH 10Ø The immlln~sorbent functional efficiency is 25 %.

FY~mple 21 750 mg of 12A8-Mab (obtained from the ~meric~n National Red Cross) is incllb~t~ with 24 g azlactone beads at pH 4.0 for 10 mins at 4C in the presence of 0.5M Tris. After the first 10 minute in.;ub~ n with 0.5 M Tris, the salt concentration is raised to 0.8 M Na2SO4 at pH 4.0 for 10 mins at 4C. The pH is then increased to pH 9.0 dropwise with lN NaOH. The reaction at pH 9.0 is allowed to proceed for 40 mins at 4C. The total diffusion and reaction time is 60 mins at 4 C. The supernatant is pipetted off. R~id--~l reactive sites are blocked with 240 ml of 1.0 M ethanolamine in 0.05 M sodium pyrophosphate, pH 9.3 for 30 mins at RT. Beads are allowed to settle and the ~u~l~-atalll is pipetted off. The 12A8-Mab coupling efficency is greater than 70%. An additional 240 mls of blocking solution is combined with the beads and incl-b~ted for 60 mins at RT. Upon completion of the second blocking step, the beads are washed with four column volumes of 0.5 M NaCl and equilibrated with loading buffer for protein immllnosorption. 125 mg of reco",binallt human Protein C in 600 ml of 0.125 M
Tris, 0.1 NaCl, 15 mM MgCl2 at pH 8.0 is loaded columnwise at 1 cm/min linear velocity into a 5.0 cm by 50.0 cm length glass chr~ lo~.,.phy column con~ -g the 12A8-Mab:~7l~ctone beads at 4C. The protein C is eluted from the immunosorbent at 1 cm/min linear velocity with 210 ml of 0.1 M NaHCO3, 0.15 M
NaCl at pH 10Ø The immunosoll.ent functional efflciency is 25 %.
While embo~im~nt~ have been identified and exemplified, the following claims and their equivalents provide the scope of the present invention.

Claims (4)

5. The method of Claim 1, wherein the method provides means for controlling spatial distribution of coupling of the ligand to the porous support and wherein the method provides covalent coupling of the ligand to inner surfaces and outer surfaces of the porous support.

6. The method of Claims 1-5, wherein the coupling conditions of step (b) employ a coupling competitor in a concentration ranging from 0.1 M to 10 M.

7. The method of Claim 1, wherein the porous support is azlactone-functional.

8. The method of Claim 1, wherein step (a) has a lower Thiele modulus than step (b).

9. A method of binding a ligate otherwise deleteriously affected by restricted diffusion into a porous support or by steric effects of binding to ligand coupled to the porous support, comprising steps of:
(1) coupling the ligand to the porous support according to the steps (a) and (b) of the method of Claim 1-8 to provide spatial distribution of ligand coupled to surfaces of the support; and (2) binding the ligate to the spatially distributed ligand.

10. A derivatized porous support produced according to the method of Claims 1-8.

11. A derivatized porous support comprising ligand coupled to an azlactone-functional porous support in a manner that at least 30% of the amount of coupled ligand is coupled to internal surface of the porous support that are within 70% of the geometric center of the support.

What is claimed is:

1. A method for coupling a ligand within a porous support, comprising the steps of:

(a) diffusing ligand into a porous support, and (b) altering conditions to provide rapid covalent coupling of the ligand to the porous support, wherein a rate of ligand coupling during step (a) is less than a rate of ligand coupling during step (b), and wherein the conditions altered from step (a) to step (b) comprise increasing pH, increasing ionic concentration, or increasing temperature, or combinations thereof.

2. The method of Claim 1, wherein pH of the diffusion conditions of step (a) is less than pH of the coupling conditions of step (b), wherein the pH of the diffusion conditions of step (a) ranges from about 3 to about 7, and wherein the pH of the coupling conditions of step (b) ranges from about 7 to about 10.

3. The method of Claim 1, wherein ionic concentration in the diffusing step (a) is less than ionic concentration of the coupling conditions of step (b), and wherein the amount of change in ionic concentration from step (a) to step (b) ranges from about 0.5 M to about 1.5 M.

4. The method of Claim 3, wherein the temperature of step (a) is less than the temperature of step (b), and wherein amount of change in temperature step (a) to step (b) ranges from about 10°C to about 35°C.