BACKGROUND OF THE INVENTION

Field of the Invention

The present invention relates to the photoactivation o proteins by ultraviolet radiation, so that they may be reacte with other chemical entities, including radio-isotopes, to for useful conjugates.

Description of the Background Art A. Radio label ling of Antibodies The conventional method of labeling antibodies wit technetium (and similar radiometals) is to reduce the disulfid bonds of the antibody with a chemical reducing agent and the react the reduced antibodies with reduced pertechnetate.

Chemical reducing agents are difficult to handle due t their susceptibility to oxidation. They may cause unwanted sid reactions on proteins (for example, by reducing carbonyl groups) , and it may be difficult to control the extent of the reductio reaction given the need to use strong reducing reagents.

Prior to radiolabelling, the reduced antibodies must b separated from the remaining reducing reagent to stop th reaction, to remove potentially toxic substances (e.g., DTT, i used for reduction) , and to avoid complexing technetium an thereby preventing it from labeling the protein. Thi purification process is time-consuming and may lead to loss o protein, or to re-oxidation of the protein.

Since a source of reduced technetium is required fo radiolabeling, a certain amount of reducing agent must be adde back to either the reduced protein or to the technetium sourc itself. The latter implies that the labeling process i necessarily at least two steps.

Rhodes, et al., J. Nucl. ed. , 27:685 (1986 and Crockfor and Rhodes, US 4,424,200 (1984) teach a method of labellin

F(ab\') ₂ fragments of antibodies with the radioisotope Tc-99m The fragments were incubated overnight with stannous ions (e.g. SnCl ₂ ) , a reducing agent, in a phthalate-tartrate buffer. Thi "pretinning" process converts the dimeric F(ab\') ₂ to monomeri fragments through reduction of the disulfide bonds connecting th heavy chains. The fragments are then reacted with Tc-99m.

Unfortunately, the Rhodes method has several disadvantages First, it requires a high concentration of stannous ion to reduc the antibody. Nonetheless, Rhodes desires that about two-third of the stannous ions be oxidized in the reaction with th protein, the remainder serving to reduce the pertechnetate whe it is added later. However, it is difficult to control th extent to which the stannous ions are involved in the reductio of the antibody. If too little stannous ion was provided, th pertechnetate will not be completely reduced by the residuum If too much stannous ion was furnished, some will complex th pertechnetate so that it will not bond to the antibody Consequently, it is customary, after reducing the antibody, t purify out the residual stannous ion, and then add back a small controlled amount to reduce the pertechnetate. The purificatio step of course add to the time and expense of the process. See e.g., the discussion in Bremer, et al., EP Appl. 271,806 "Process for Producing an Organ-Specific Substance Labeled wit Technetium-99m"

Rhodes, U.S. 5,078,985 acknowledge that his origina pretinning method had "problems associated with lack of purit of the protein, continued reduction of disulfide bonds in th protein, and the formation of additional reduced protein specie prior to admixture with sodium pertechnelate. Fewer stannou ions are needed to reduce pertechnetate than to reduce disulfid bonds in proteins; with some proteins, excess stannous ions caus reduction of disulfide bonds and fragmentation of the protein t less than the desired size." Consequently, Rhodes \'985, taugh that after the initial reduction, whether with stannous ion o a different agent such as dithiothreitol (DDT) , the reduce protein should be purified to substantially remove the firs

reducing agent and impurities, and stannous ion added in sufficient amount to reduce the radionuclide (the latter to b added in subsequent step) .

Because the stannous ion concentration is high, th likelihood of undesired fragmentation of the antibody i increased. This is because a strong reducing agent, such a stannous ion, will reduce disulfide bonds, and the extent of suc reduction, and therefore of fragmentation, is a function o concentration. The Crockford and Rhodes process is not readil adjusted to reduce unwanted fragmentation.

While the \'200 patent asserts that a pH of 4.5 to 8.5 i "preferred", in our experience the upper limit on pH is actuall 6. With a higher pH, stannous ion is precipitated out o solution, and therefore is unavailable for reduction. This i unfortunate, as some proteins, including certain antibodies, ar more stable at a higher pH.

Rhodes\' method requires long incubation times; the \'20 patent calls for reaction for at least 15 hours, more preferabl 21 hours. Such long reactions are inherently more vulnerable t mishaps, such as loss of temperature control or nitrogen supply contamination, etc. Finally, its yields are lackluster; th yield reported in the Crockford \'200 patent was 73% radiolabele IgG. A more typical yield is a litte over 50% as seen in Tabl 2A. About 17% of Crockford\'s Tc was in colloidal form Moreover, about 28% of the Tc on the antibody wa "exchangeable" .

Shochat, U.S. 5,061,641 uses the same reducing agent an buffer, but suggests that it may be helpful to contact th resulting technetium-labeled antibody with an "exogenous cappin ligand", such as a mercaptoalkane or phosphane, to fill th remaining coordination sites of the radiometal and thereb stabilize it. But this patent does not address any of the othe disadvantages of the Rhodes method.

B. Photoaf finity Labeling of Proteins

Photoactivatable agents, including arylazides, have been used to immobilize an antigen or antibody on a support. See Kramer, U.S. 4,689,310; Scheebers, U.S. 4,716,122; AU-A-47690/85 (organogen) . Dattagupta, U.S. 4,713,326 photochemically coupled a nucleic acid to a substrate. Other molecules have also been insolubilized by this technique. Lingwood, U.S. 4,597,999. Pandey, et al. , J. Immunol. Meth., 94:237-47 (1986) conjugated a haptenic primary aromatic amine (3-azido-N-ethylcarbazole) to various proteins to create a synthetic antigen.

Furthermore, one may prepare a "prodrug" or "protoxin" which is converted to the active drug or toxin by photochemical cleavage of a bridging group. Zweig, U.S. 4,202,323; Senter,

U.S.4,625,014; Edelson, U.S.4,612,007; Reinherz, U.S.4,443,427 (col. 4) .

Photoaffinity labeling of enzymes is also known. Chowdry, Ann. Rev. Biochem., 48:293-305 (1979). In this technique, probe compound is allowed to interact with a biomatrix in whic a specific receptor or binding site is envisaged. Upon formatio of the specific complex by their affinity for one another th pair are chemically linked via a reactive group on the probe. This greatly facilitates the identification and characterizatio of the receptor itself. The great advantage of probes incorporating a photoactive group is that the linkage process ma be activated after all non-specific interactions of the prob have been eliminated and thus little or no indiscriminate bindin or loss of probe through hydrolysis takes place. Thes photoaffinity reagents may be prepared either via syntheti routes for simple molecules or via conjugation of a photoactiv ligand for more complex macromolecules.

Thus, photoaffinity labeling of peptide hormone bindin sites was reported by Galardy, et al., J. Biol. Chem. , 249: 35

(1974) . The probes employed were aryl azides, which, whe photoactivated to yield aryl nitrenes, can label any binding sit containing carbon-hydrogen bonds by insertion into the C-H bond.

Biomira (Noujaim) , EP Appl. 354,543 teaches th photochemical attachment of chelating groups to biomolecules, such as proteins (especially antibodies) . This may be used t indirectly radiolabel a biomolecule with a chelatable ion. chelating group is first coupled to a photoactivatabl functionality. If reacted with an antibody in the presence o light, the "photochelate" will label the antibody. The disclose activated species are carbenes, nitrenes and free radicals, though only nitrenes are exemplified.

C. Studies of Effect of Ultraviolet Radiation on Proteins

Low intensity ultraviolet irradiation of human gamm globulin has been shown to result in the generation of fre sulfhydryl groups and in the aggregation of the protein. Wickens, et al., "Free radical-mediated aggregation of huma gamma globulin, Agents and Actions, 11:650 (1981) . Disulfid bonds are reduced at random and the resulting free sulfhydryl can re-combine with free thiol groups on other molecules t generate aggregates. Later experiments showed that other source of oxygen free-radicals, such as a mixture of copper salts an hydrogen peroxide, had similar effects. Wickens, et al. Biochim. Biophys. Acta, 742: 607 (1983). The intrachain, rathe than the interchain, disulfide bonds were deemed to be th primary targets since further fragmentation was not seen According to Lunec, et al., J. Clin. Invest. 76: 2084 (1985), th aromatic amino acids, especially tryptophan and tyrosine, ar also attacked. These researchers teach that free radical damag of IgG, through the mechanism of aggregate formation, resulte in the production of rheumatoid factors, i.e., in an autoimmun response to the aggregated antibodies. Thus, it would normall be considered undesirable to expose antibodies, for long enoug periods to reduce disulfide bonds, to ultraviolet radiation.

The aforementioned findings with respect to antibodies ar consistent with studies of the inactivation of disulfide enzyme (e.g., trypsin) by ultraviolet radiation; cystine disruption i

likely to be a major contributing factor. Moreover, the presence of tryptophan or tyrosine residues in a given protein may contribute to the destruction of the cystine residues. K. Dose, "Theoretical Aspects of the U.V. Inactivation of Proteins containing Disulfide Bonds", Photochem. Photobiol. 6., 437-443 (1967) ; K. Dose, "The Photolysis of Free Cystine in the Presence of Aromatic Amino Acids", Photochem. Photobiol. 8 _. , 331-335 (1968) . As a result, ultraviolet irradiation of antibodies, which contain disulfides, is contraindicated, as it may result in loss of activity if disulfide bonds essential to antigen- binding and/or effector function are ruptured.

No admission is made that any reference or statement herein constitutes prior art, and Applicants reserve the right to challenge the accuracy of the nominal publication date of any cited publication, or of the contents thereof.

SUMMARY OF THE INVENTION

The present invention overcomes the above-noted deficiencies of the background art. More particularly, it contemplates the photochemical activation of proteins, especially disulfide- containing proteins, (e.g., antibodies) , for conjugation to other chemical entities ("partners") . In one preferred embodiment, the partner is a radioisotope, such as a radiometal, especially technetium. In another preferred embodiment, the partner is another protein, a toxin, or a chelate.

The methods describe herein make possible a simple, "one pot" preparation of an activated protein. The specific examples included refer, in a non-limiting manner, to the activation and radiolabeling of proteins with technetium. The radiolabeling is rapid, with high radiolabeling yields being obtained in less than an hour. The yield, indeed, is higher than that obtained by the pretinning method with even an overnight incubation. The extent of the reaction is readily controlled, and it is not necessary to remove any of the reagents. The medium may be optimized for antibody stability, including use of buffers other than saline, and its pH may be greater than 6. In using this method to label

antibodies with technetium, we typically see less than 5 colloidal Tc and less than 2% exchangeable Tc.

Surprisingly, the irradiation of the antibody does no inactivate it, or result in the formation of potentiall immunogenic aggregates, though these problems would have bee expected in view of the background art.

One way in which immunogenic aggregates could have bee expected to form is through the reaction of the free thiol formed by the photochemical reaction to form new disulfide bond connecting two different antibody molecules.

Another effect of the The photoche ically-induced reductio of the disulfide bonds which could have been thought capable o increasing antigenicity is that it can result in an equivalen oxidation elsewhere in the antibody molecule. While the Wicken paper (Wickens, et al., Agents and Actions, ϋ, 650, 1981) doe not identify the sites of oxidation in the antibody molecule, i nonetheless presents evidence (fluorescence data) that indicate that aromatic amino acid side chains such as those of tryptopha or tyrosine are oxidized during irradiation. In classica organic chemistry, such concomitant oxidation/reduction reaction are referred to as disproportionations. We are thus looking a an intramolecular disproportionation of the antibody. Th oxidized tryptophan or tyrosine derivatives of separate antibod molecules can subsequently react to form crosslinks which ma increase antigenicity or decrease antigen-binding activity.

Given the frequency of Tyr, Phe, and Trp residues i complementarity-determining regions of antibodies, one would hav been led to expect that the photochemical reduction of th antibodies\' disulfide bonds could have produced oxidative change in the CDRs with the potential for considerable loss of antige binding capacity. Moreover, certain disulfide bonds, such a those local to the variable domain, may make importan contributions to antigen binding activity, and it was no possible, a priori, to state whether photochemical reductio would have cleaved these disulfide bonds.

The fact that Noujaim did not experience loss of antige binding capacity as a result of UV irradiation of certai derivatized antibodies would not have overcome the negativ

teachings of Wickens et al. Noujaim did not determine whether, as a result of the exposure to UV, his antibody had become more antigenic. Consequently, a person of ordinary skill in the art, aware of the teachings of both Wickens and Noujaim, would still have been concerned that the irradiated antibody would form immunogenic aggregates or would otherwise be altered in a manner which would activate the immune system.

In the Noujaim case, irradiation is used to effect conversion of a photosensitive molecule, such as an aromatic azido derivative, into a highly reactive intermediate, such as a nitrene. Both the reaction generating the nitrene and the subsequent attack of the carbene or nitrene are extremely rapid, so that a typical reaction time for the procedure of Noujaim is 5 min. While Noujaim teaches that the azide may be photolyzed in the presence of an antibody, Noujaim does not contemplate that the irradiation will have any effect on the antibody. Rather, the antibody is seen as a passive substrate for a chemical attack by the nitrene resulting from the photoactivation of the azide.

In contrast, in the present invention, the antibody itself is photoactivated by UV irradiation. It then reacts with partner molecules through its free thiol groups, or possibly, with reactive derivatives of its aromatic amino acids. These moieties are much less reactive than are the nitrenes (or carbenes) of the Noujaim disclosure. It should further be pointed out that Noujaim\'s findings regarding retention of antigen-binding activity would not necessarily be extrapolated, by persons skilled in the art, to the preferred embodiments of the present invention. Noujaim\'s photolysis apparatus was a water-cooled, quartz-jacketed Hanovia UV lamp (254 nm) . The reaction mixture was placed 10 cm from the center of the lamp, and the irradiation was for 1-10 minutes. In Example 2 of the present application, the source was apparently the same, but the reaction mixture was closer to the lamp (10 cm) and the exposure time was longer (30 min.). It therefore appears reasonable to surmise that the irradiation conditions of our Example 1 were more drastic than those taught by Noujaim. While Applicants do not wish to be limited to the conditions of Example 1, a person skilled in the art would be

hesitant to assume that simply because the diminution of antige binding activity observed by Noujaim was small, that this woul hold true if the irradiation were longer or more intense, or wit wavelengths to which antibodies are more sensitive. Wickens suggests that damage to the antibodies will increase wit increasing irradiation time.

Surprisingly, despite the prolonged irradiations contemplated by the present invention, neither increases i antigenicity (whether through dimer formation or throug modification of aromatic residues) nor decreases in antige binding capacity have proven to be serious concerns. While th inventors do not wish to be bound to any theory, they believ that disulfide bonds differ in their vulnerability t photochemical reduction, and that the disulfide bonds most directly involved in antigen binding activity are also the ones least susceptible to attack. As for the aromatic residues, it appears that whatever modifications are occurring, that these d not enhance immunogenicity or interfere with antigen binding.

The appended claims are hereby incorporated by referenc into this specification as a further enumeration of the preferre embodiments .

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows a radiochemical chromatograph of Tc-99 labeled, irradiated MAb 170. The main peak (1) is monomeric Tc- 99m MAb 170. The minor shoulders (2) and (3) are, respectively, Tc-99m Low MW protein and Tc-99m buffer complex. The ordinat is radioactivity in relative units, as measured by a flow throug Nal(Tl) crystal detector, and the abscissa is time, in minutes, after injection onto the SE-HPLC column.

Figure 2 shows the effect of UV irradiation on radiolabelin of Mab 170 (white squares) and Mab B43 (black squares) . Thes are compared with the radiolabeling of MAb 170, non-irradiated, when labeled by the Rhodes method (white circles) . The ordinat is the % Tc-99m which is protein bound, as measured by a G50 spi column, and the abscissa is time or irradiation in minutes fo the irradiated antibodies, or the pretinning time for the non

irradiated antibody.

Figure 3 shows an UV280 nm chromatogram of irradiated, Tc-

99m labeled MAb 170. The main peak is monomeric MAb 170 with no detectable aggregate or fragment formation. The ordinate is UV absorbance at 280 nm in relative units and the abscissa is time, in minutes, after injection onto the SE-HPLC column.

Figure 4 shows the effect of photoactivation wavelength on radiolabeling yield.

Figure 5 depicts sulfhydryl generation as a function of the duration of the photoactivation step.

Figure 6 shows the effect of sulfhydryl blocking agents on radiolabeling yield.

Figure 7 considers the effect of a cysteine challenge on the photoactivation reaction, as indicated by the percentage of cysteine which is radiolabeled.

Figure 8 compares the radiolabeling (Tc-99m) yield of stannous ion-reduced and photoactivated MAb-170 after cysteine challenge.

Figure 9 plots the effect of stannous source and concentration on radiolabeling yield.

Figure 10 depicts the effect of stannous concentration and irradiation time on radiolabeling yield.

Figure 11 is a study of the effect of irradiation volume and irradiation time on radiolabeling yield.

Figures 12A-12D compare biodistribution of Tc-99m MAb-170, in mice, depending on the method used to prepare the radiolabeled antibody.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The Photoactivatable Protein

The protein to be photochemically activated for conjugati purposes is preferably one containing disulfide bonds no essential for the intended use of the protein. In such protein, a disulfide bond is present as a result of the oxidati of the thiol (-SH) side groups of two cysteine residues. Thes residues may lie on different polypeptide chains, or on the sa polypeptide chain. As a result of the oxidation, a disulfi bond (-S-S-) is formed between the beta carbons of the origin cysteine residues. After reduction, the residues shoul technically speaking, be termed half-cystines, but the ter cysteine, cystine and half-cystine are often us interchangeably, the correct meaning being apparent from context The effect of the radiation, as previously stated, is t reductive cleavage of the disulfide bonds to leave free thi groups.

Examples of disulfide bonded proteins include antibodie many enzymes, and certain hormones. Antibodies . The immunoglobulins may be used in vit diagnosis, in vivo imaging, or therapy of diseases or conditio with distinctive antigens. The basic unit of immunoglobul

(antibody) structure is a complex of four polypeptides -- t identical low molecular weight ("light") chains and two identic high molecular weight ("heavy") chains, linked together by bo non-covalent associations and by disulfide bonds. Differe antibodies will have anywhere from one to five of these bas units. The immunoglobulin unit may be represented schematical as a "Y" . Each branch of the "Y" is formed by the amino termin portion of a heavy chain and an associated light chain. The ba of the "Y" is formed by the carboxy terminal portions of the t heavy chains. The node of the "Y" is the so-called hinge regio and is quite flexible. Five human antibody classes (IgG, Ig IgM, IgD and IgE) , and within these classes, various subclasse are recognized on the basis of structural differences, such the number of immunoglobulin units in a single antibody molecul the disulfide bridge structure of the individual units, a differences in chain length and sequence. The class and subcla

of an antibody is its isotype.

The amino terminal regions of the heavy and light chains are far more diverse in sequence than the carboxy terminal regions, and hence are termed the variable domains. This is the part of the antibody whose structure confers the antigen-binding specificity of the antibody. A heavy variable domain and. a light variable domain together form a single antigen-binding site, thus, the basic immunoglobulin unit has two antigen-binding sites. The walls of the antigen-binding site are defined by hypervariable segments of the heavy and light variable domains. Binding site diversity is generated both by sequence variation in the hypervariable region and by random combinatorial association of a heavy chain with a light chain. Collectively, the hypervariable segments are termed the paratope of the antibody; this paratope is essentially complementary to the epitope of the cognate antigen.

The carboxy terminal portion of the heavy and light chains form the constant domains. While there is much less diversity in these domains, there are, first of all, differences from one animal species to another, and secondly, within the same individual, there will be several different isotypes of antibody, each having a different function.

The IgG molecule may be divided into homology units. The light chain has two such units, the V _L and C _L , , and the heavy chain has four, designated V _H , C _H 1, C _H 2 and C _H 3. All are about 110 amino acids in length and have a centrally located intrachain disulfide bridge that spans about 60 amino acid residues. The sequences of the two V-region homology units are similar, as are the sequences of the four C-region homology units. These homology units in turn form domains. The two variable domains have already been mentioned; there are also four constant domains. Mild proteolytic digestion of IgG results in the production of certain fragments of interest. V-Cl is Fab; C _H 2- C _H 3 is Fc; (V-Cl)j is (Fab\') ₂ , V-C1-C2 is Fabc, and V alone is Fv. While the variable domains are responsible for antigen binding, the constant domains are charged with the various effector functions: stimulation of B cells to undergo proliferation and differentiation, activation of the complement

cell lysis system, opsonization, attraction of macrophages t ingest the invader, etc. Antibodies o iifferent isotypes hav different constant domains and therefore have different effecto functions. The best studied isotypes are IgG and IgM. The term "antibody", when used herein without furthe qualification, is intended to include both "intact" antibody an its various proteolytic derivatives. Antibodies may b conjugated to other molecules to produce conjugates useful i diagnosis, therapy, etc. The antibody\'s variable domain give the conjugate the ability to bind specifically to particula antigenic targets.

The antibodies may be directed against antigens of medica interest, such as those associated with pathogens (incl. viruses bacteria, fungi, and protozoa) , parasites, tumor cells, o particular medical conditions. In the case of a tumor-associate antigen (TAA) , the cancer may be of the lung, colon, rectum breast, ovary, prostt e gland, head, neck, bone, immune system or any other anatomical location. Antigens of particula interest are carcinoma-embryonic antigen (CEA) , human chorioni gonadotropin (hCG) , alpha-fetoprotein (AFP) , ferritin, Thomsen Friedenreich antigen (TF-alpha; TF-beta) , stage-specifi embryonic antigen-1 (SSEA-1) , human mammary tumor-associate antigen (hMTAA) , oncomodulin, malignin, human placental lactoge (hPL) , prostatic antigen (PA) , prostatic acid phosphatase (PAP) high molecular weight-melanoma-associated antigen (HMW-MAA) thyroglobulin (Tg) , tyrosine phosphokinase (TPK) , epiderma growth factor (EGF) , neuron-specific enolase (NSE) , tissu polypeptide antigen (TPA) , beta-2 microglobulin (B2M) phosphohexose isomerase (PHI), fibrin, Tn, sialyl Tn, CA-19.9 CA-125, and CA-15.3.

The term "tumor-specific antigen" as used herein will b understood to connote an antigen characteristic of a particula tumor, or strongly correlated with such a tumor. However, th current understanding in the art with respect to tumor-specifi antigens is that they are not necessarily unique to tumor tissue i.e., that antibodies to them may cross-react with antigens o normal tissue. Even where tumor-specifc antigens are not uniqu to tumor cells, it frequently occurs that, as a practical matter

antibodies binding to tumor-specific antigens are sufficiently specific to tumor cells to carry out the desired procedures without unwarranted risk or interference due to cross-reactions. Many factors contribute to this practical specificity. For example, the amount of antigen on the tumor cell may greatly exceed the amount of the cross-reactive antigen found on normal cells, or the antigen on the tumor cells may be more effectively presented. Therefore the term "tumor-specific antigen" relates herein to a specificity of practical utility, and is not intended to denote absolute specificity or to imply an antigen unique to the tumor.

MAb 170 (more accurately, MAbl70H.82) is a murine monoclonal antibody of the IgGj kappa isotype that was produced by immunizing BALB/c mice with a synthetic glycoconjugate consisting of a Thomsen-Friedenreich (TF) beta (Galbetal->3GalNAc) disaccharide hapten coupled to an immunologically suitable carrier (serum albumin) . It was selected based on its reactivity with human adenocarcinoma tissue in vitro. It clearly reacts with adenocarcinomata of the breast, ovary, endometrium, colon, prostate and some bladder. It is described in more detail in copending Ser. No. 07/153,162, filed May 12, 1988, incorporated by reference herein, which is a continuation of Ser. No. 06/927,277, filed Oct. 27, 1986. MAb 170 has been formulated into a Tc-99m radiolabeled antibody kit (TRUSCINT AD, Biomira, Inc., Edmonton, Alberta, Canada) for radioimmunodiagnosis of adenocarcinomas. See McEwan, et al., Nuclear Medicine Communications, 13: 11-19 (1992). A hybridoma (170H82.R1808) secreting MAb 170 was deposited on July 16, 1991 with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 USA, an International Depositor Authority under the Budapest Treaty, and assigned the accession number HB 10825. This deposit should not be construed as a license to make, use or sell the hybridoma or MAb 170.

MAb B43 (more accurately, B43.13) is a murine monoclonal antibody of the IgGj kappa isotype that was produced b immunizing mice with high molecular weight mucins partiall purified from ovarian ascites, and selected for its reactivit to CA 125, an ovarian carcinoma-associated antigen. It inhibits

the binding of MAb OC125 to CA125. MAb B43 is reactive wit CA125 antigen in biopsy tissue and in serous and endometroi carcinomas of the ovary. It has been formulated into a Tc99m- radiolabeled antibody kit (TRUSCINT OV, Biomira, Inc. Edmonton, Alberta, Canada) for radioimmunodiagnosis of ovarian carcinomas. See Capstick, et al., Int. J. Biol. Markers, 6: 129-135 (1991). Reference to these two antibodies should not be construe as a limitation on the generality of the present invention.

The antibody may be a polyclonal antibody or a monoclona antibody. When the subject is a human subject, the antibody ma be obtained by immunizing any animal capable of mounting a usabl immune response to the antigen. The animal may be a mouse, rat, goat, sheep, rabbit or other suitable experimental animal. Th antigen may be presented in the form of a naturally occurrin immunogen, or a synthetic immunogenic conjugate of a hapten an an immunogenic carrier. In the case of a monoclonal antibody, antibody producing cells of the immunized animal may be fuse with "immortal" or "immortalized" human or animal cells to obtai a hybridoma which produces the antibody. If desired, the gene encoding one or more of the immunoglobulin chains may be clone so that the antibody may be produced in different host cells, an if desired, the genes may be mutated so as to alter the sequenc and hence the immunological characteristics of the antibod produced. Disulfide -Bonded Enzymes . Disulfide bonded enzymes includ trypsin, chymotrypsin, aldolase, papain, and glyceraldehyd phosphate dehydrogenase. These enzymes may be conjugated t other molecules for use, e.g., as labels.

Other disulfide -bonded proteins . These include albumin transferrin, and somatostatin. Albumin can be conjugated t radioisotopes and used as a blood pool agent. Transferrin coul be labeled with radioisotopes and used to image transferri receptors; some, tumours are known to have large amounts o transferrin receptors. Somatostatin could be labeled wit radioisotopes and used for tumor imaging.

The sulfhydryl/disulfide composition of several proteins i given in Table 109A.

Non-Disulfide Bonded Proteins . The present invention ma

also be used to photoactivate non-disulfide bonded proteins, such as pre-albumin and protein A, that contain residues (e.g., aromatic residues) which are made more reactive by reaction with free radicals generated by photolysis.

Irradiation

The protein to be photoactivated is irradiated with ultraviolet radiation. Ultraviolet radiation is generally defined as that portion of the electromagnetic spectrum running from 10 to 820 nm wavelength. However, the quanta of shorter wavelengths are more energetic and therefore more potentially damaging to the protein. Preferably, the protein is irradiated with ultraviolet radiation which is primarily, e.g., at least 90%, more preferably at least 99%, of wavelengths in the range of 250-320 nm. Desirably, the radiation includes the wavelengths 270-320 nm, and, more desirably, is primarily of those wavelengths.

Ultraviolet radiation may be generated by any convenient source, such as a hydrogen or deuterium discharge lamp, a xenon arc lamp or a mercury vapour lamp. The lamp may be provided with a fluorescent coating so as to alter the effective wavelength of the light emitted by the lamp. For example, the internal output may be 254 nm, but may cause a coating to fluoresce at a longer wavelength, which is what is externally emitted.

To filter out undesired wavelengths, a suitable filter may be employed. Quartz transmits light of 190-820 nm, while borosilicate glass has a transmission spectrum of 300-820. Other materials which selectively transmit ultraviolet radiation include window glass, optical (white crown) glass, Vycor, quartz crystal, clear fused quarts, suprasil, synthetic sapphire, natural fluorite, synthetic lithium fluoride and plexiglass (polymethylmethacrylate) . Both broad and narrow band filters are known in the art.

The filter may be incorporated into the lamp unit, or place in the light path between the vessel and the source. Alternatively, the protein may be placed in a vessel forme wholly or partially of the filter material. It is possible tha the disulfide bonds are especially sensitive to highly specifi

wavelengths, in which case a narrow band pass filter may b useful to minimize radiation damage to the protein. For furthe information on ultraviolet filters, see Calvert and Pitts, Jr., Photochemistry, pp. 686-798, Chapter 7, "Experimental Methods i Photochemistry" (John Wiley & Sons, N.Y. : 1966).

The effective intensity of the radiation at the target sit is a function of the source intensity, the distance from th source to the reaction vessel, and the degree of absorbance o the radiation by the lamp filter and vessel walls. It i generally most convenient to adjust the effective intensity b moving the subject closer to or farther from the source, rathe than by changing the source or the thickness of the filters i the light path. However, any of these parameters may be modifie as desired. The overall degree of photoactiv-a ion is a functio of the intensity of the radiation and- the irradiation time, a well as the susceptibility of the protein to photoactivation wit the wavelengths employed. Like the effective intensity, th irradiation time is readily modified. For any given protein, i is contemplated that the intensity and irradiation time will b systematically varied to identify the optimum--value.

If a protein is particularly sensitive to irradiation wit particular UV wavelengths, the reaction can be slowed down b filtering out, at least to some degree, those wavelengths. Contrariwise, if a protein is refractory, the intensity an duration of the irradiation at the wavelengths to which th protein is most sensitive may be increased.

Preferably, the source is placed 1-10 cm, more preferabl about 5 cm, from the reaction vessel. This vessel is desirabl formed of quartz or of borosilicate glass. The irradiation tim is typically 5-100 minutes, preferably over 10 minutes and unde 60 minutes, more preferably over 15 minutes, still mor preferably at least about 20 minutes. The protein concentratio is usually 1-10 mg/ml, more preferably about 6 mg/ l. The pH i typically 4-9, more preferably 6-7. When pertechnetate is th partner, it is preferably reduced with 5-100 μg Sn, mor preferably about 10-30 μg.

The reaction medium may also affect the progress of th photoactivation. In particular,--.the pH of the medium, and th

presence or absence of sensitive amino acids such as tryptophan and tyrosine, of free radical inhibitors such as glutathione, dithiothreitol and the like, or free radical generators like peroxides, may have a positive or negative effect. It is know that the disulfide bond is more susceptible to reduction at p 8. Around this pH the effect of UV radiation should be more effective than at lower pHs. If the UV energy is absorbed b adjacent amino acids such as tyrosine or tryptophan an transferred down to the disulfide bond of cystine, then the absence of these amino acids would reduce the number of SH groups generated. If the UV energy is producing free radicals (as is thought) , then free radical inhibitors will decrease the numbe of SH groups formed; likewise free radical activators o generators should generate more SH groups. The addition of ascorbic acid (a free radical scavenger) to the protein solutio prior to UV irradiation drastically reduces the Tc-99 radiolabeling yields, implying a free radical mechanism may be involved.

Photoactivation of production quantities of material can be accomplished by irradiating a vessel holding a homogeneous solution of the MAb or other target, with continuous stirrin during irradiation, using commercially acquired UV sources. Alternatively, the bulk solution can be circulated through a appropriate glass tube passing around or through the UV sourc (must be custom made) . The material can also be vialled and the irradiated via on-line (i.e., assembly line through one U source) or bulk (irradiate all vials at once) irradiation.

Once the protein has been photoactivated by irradiation, i may be conjugated immediately, or it may be stored for futur use. The conjugation reactions may be carried out in any buffe appropriate for maintaining the desired pH range. The stabilit of the material may be increased, if desired, by lyophilization, or by addition of thiol stabilizers.

The Conjugation "Partner" The photoactivated protein may be conjugated with an "partner" of interest, including a radiometal salt, drug, toxin, chelate, etc.

When photoactivation of the protein results in formation of free thiol groups, the protein may be conjugated with an sulfhydryl reactive agent. Preferably, the agent is one whic is substantially specific for free thiol groups. It is certainl not necessary that the agent be an extremely reactive species such as a nitrene or carbene.

Sulfhydryl -Reactive Radiometals . Radiometals which can b bound to proteins using the present method are those which bin tightly to sulfhydryl groups. Generally, these will be metal ions which form relatively insoluble sulϊides in the conventional qualitative analysis schemata. These include, but are no limited to, ions of Tc-99m, Re-186, Re-188, Cu-64, Cu-67, Hg-195, Hg-197, Hg-203, Pb-203, Pb/Bi-212, Zn-72, Ag-105, Ag-111, Au-198, Au-199, Cd-115, Cd-115m, Sn-117, Sn-125, and the like. Radiometals having a gamma emission energy in the range of about 50-500 KeV are useful for scintigraphy. Positron emitters ca also be used for imaging applications. Beta and alpha emitter are useful for therapy. Preferably, the radiolabeling yield i greater than 80%, and, more preferably, is greater than 90%. Desirably, these preferred yields are achievable with protein which were irradiated for not more than about two hours, and mor preferably not more than about one he. r.

In one embodiment, the partner is a pertechnate, rhennate, or other radioisotopic agent of similar chemistry. In general, the pertechnetate or rhennate will be reduced so that it wil react with the free thiol groups of the protein. Suitabl reducing agents include sources of stannous ion, such as stannou chloride and stannous tartrate; stannous tartrate is preferre because the tartrate anion stabilizes the Sn-Tc complex. Othe reducing agents known in the art include 2-mercaptoethanol,1,4 dithiothreitol,2,3-dihydroxybutane-1.4-dithiol,2-aminoethane thio HCl, 2-mercaptoethylamine, thioglycolate, cyanide and cysteine The amount of the reducing agent, and the incubation time, ar adjusted in the light of the reducing agent employed. Th stannous ion may be added to the protein prior t photoactivation, and the pertechnetate added afterward, or th protein may first be photoactivated, and the stannous ion an pertechnetate added after irradiation. Typically only smal

quantities of reducing agent are required, as it is used only to reduce the radiometal and not the antibody or other partner, so that purification to remove excess tin is unnecessary.

Technetium-99m is a preferred radiolabel for scintigraphy because of its ready availability and ease of preparation from commercial pertechnetate generators.

Technetium labeling of the sulfhydryl-containing protein is generally effected by conventional methods. Pertechnetate is obtained from a commercially available generator, most commonly in the form of NaTc0 ₄ , normally in saline solution. Other forms of pertechnetate may be used, with appropriate modification of the procedure, as would be suggested by the supplier of a new form of generator or as would be apparent to the ordinary skilled artisan. Pertechnetate is generally used at an activity of about 0.2- 10 mCi/ml in saline, e.g., 0.9% ("physiological") saline, buffered at a pH of about 3-7, preferably 3.5-5.5, more preferably about 4.5-5.0. Suitable buffers include, e.g., acetate, tartrate, phthalate, citrate, phosphate and the like. Rhenium is found just below technetium in the periodic table and has the same outer shell electronic configuration. Rhenium and its compounds are expected to have very similar chemical properties to technetium and its analogous compounds. In fact, rhenium compounds behave similarly to technetium compounds insofar as reduction and chelation are concerned but their greater susceptibility to oxidation requires greater care in handling.

The radioisotope Re-186 is attractive for both imaging and therapy. It has a half-life of about 3.7 days, a high LET beta emission (1.07 MeV) and a convenient gamma emission energy (0.137 MeV) . Rhenium may be produced from perrhenate, and the reduced rhenium ions can bind non-specifically to protein. Accordingly, a method for Re-186 labeling of proteins, wherein the reduce perrhenate is bound to sulfhydryl groups of a protein molecul such as an antibody, would be advantageous. Re-188 is generator-produced beta and gamma emitter with a half-life o about 17 hours and could be useful for imaging and therapy.

Rhenium labeling will be effected in substantially the sam

manner as technetium labeling, with special care being taken t ensure the absence of air or other source of oxygen from th system. Re-186 is produced in the form of sodium perrhenate b use of a generator analogous to currently available technetiu generators.

By "reduced pertechnetate" or "reduced perrhenate" is mean the species of technetium or rhenium ion formed by chemica reduction of pertechnetate or perrhenate and chelated by th thiol group(s). It is generally thought that reduce pertechnetate is in the form of Tc(III) and/or Tc(IV) and/o Tc(V) in such chelates and that reduced perrhenate is in the for of Re(III) and/or Re (IV) and/or Re(V) , but higher or lowe oxidation states and/or multiple oxidation states cannot b excluded and are within the scope of the invention. Copper wil normally be in the form of Cu(II), although Cu(I) and/or Cu(II) are not excluded. Mercury will normally be in the form of Hg(I) and/or Hg(II). Lead/bismuth will normally be in the form o Pb(II) or Pb(IV) .

Reduction is effected by any of a variety of conventiona reducing agents, preferably stannous ion generally in aqueou solution. Other suitable reducing agents include, e.g. dithionite, borohydride, ferrous ion, formadine sulfonic acid and the like. It will be appreciated that stannous ion can b generated in situ from tin metal, e.g., foil, granules, powder turnings and the like, by contact with aqueous acid, e.g., HCl

Copper ions are also tightly chelated by sulfur chelators

Cu-67 is another attractive radionuclide for imaging and therapy

It has a half-life of about 2.6 days, and is a beta (0.570 MeV and gamma emitter (0.185 MeV), although the beta energy i relatively low. Cu-67 is relatively expensive and not readil available at present, although such conditions can change a demand develops. It has the advantage that it forms tigh chelates with thiols. The labeling is simple and rapid, an requires no reducing agent for the radiometal. Copper labeling will be effected by reaction of a thiol containing protein with a solution of copper ions, normall Cu(II) ions, in the form of a convenient salt, e.g., chloride citrate, tartrate or the like, either as available or by mixin

of e.g., the chloride with, e.g., sodium, potassium or ammoniu citrate, tartrate or the like. Cu-67 is currently available a CuCl ₂ from Oak Ridge National Laboratories, Tennessee, or fro Los Alamos National Laboratories, N. Mex. Zinc, silver, gold an cadmium isotopes would chelate SH groups in a manner similar t copper.

Other radionuclides with similar chelation behavior t copper, e.g., mercury and lead, also could be bound to thiol containing compounds according to the method of the invention. Hg-197 has a half-life of about 1.5 days, and emits gamm radiation in an energy range of 78-268 KeV, and Pb-203 is strong gamma-emitter at about 275 KeV, with a half-life of abou

51 hr, making them suitable for gamma scintigraphy. Bi-212 i an alpha emitter with a half-life of about 1 hr and an energy o 6.09 MeV, making it of considerable interest for in vivo therapy.

It is produced in situ from a Pb-212 precursor with emission o gamma radiation of 239 KeV, with a half-life of about 10.6 hr

Thus, antibody conjugates for Bi-212 therapy will be Pb-21 labeled conjugates, and the short-hand notation lead/bismuth o Pb/Bi is used herein to indicate this. Chelation to the antibod protein is effected analogously to Cu-67 labeling.

Mercury radioisotopes are normally available as HgCl ₂ or a Hg(N0 ₃ ) ₂ , e.g., from Oak Ridge National Laboratories.

Lead/bismuth radioisotopes are normally available fro Argonne National Laboratories in the form of supported rado generator.

It will be understood that the invention is not limited t the exemplified radiometal ions, but is generally applicable t ions that bind tightly to sulfhydryl groups. Stable isotopes may also be conjugated to proteins fo therapeutic (e.g., Au for arthritis) or diagnostic (e.g. colloidal Au compounds for electron microscopy) purposes.

Chelates . Chelates may be used to associate an antibod with chelatable substances, such as certain radioisotopes, whic cannot be directly reacted with the photoactivated antibody. Th present invention is not limited to any particular chelatin agent. While, EDTA and DTPA derivatives are preferred, man chelating agents are known. See Mears, U.S. 4,678,667; Wieder

U.S. 4,352,751; Hnatowich, U.S. 4,479,930; Meares, U.S. 4,043,998; Ueda, U.S. 4,564,742; Davidson, U.S. 4,673,562; Hnatowich, U.S. 4,668,503; Arano, U.S. 4,559,221 and Costa, U.S. 3,809,632. Besides EDTA and analogous polycarboxylic acids, the macrocyclic chelators are of particular interest. The chelating agent is derivatized, if necessary, so as to react with the free thiols of the photoactivated protein while retaining its chelating function. The following table shows ions chelated by various agents.

⁵³ Sm ¹⁵³ Sm

153 Sm

EDTA = ethylenediaminetetraacetic acid DTPA = diethylenetriaminepentaacetic acid

MA-DTPA and CA-DTPA = Methylene Adduct and Cyclic

Anhydride. TETA and DOTA = not acronyms for chemical name; just the abbreviation given for these two chelates

DADS = diaminodisulfur

Drugs . Suitable drugs include antibiotics such as adriamycin, antitumor agents such as methotrexate, 5- fluorouracil and cis-platinum, and antiparasitic agents such as pentamidine isethionate. When an antibody is conjugated to suc a drug, it serves to direct the drug to the sites where th corresponding antigen occurs.

Toxins . Toxins are usefully conjugated to antibodie specific for antigens associated with tumor, parasite o microbial cells. The toxin may be e.g., a plant (e.g., ricin o abrin) , animal (e.g., a snake venom), or microbial (e.g., diphtheria or tetanus toxin) .

Besides antibodies, the drugs or toxins may be conjugate

to other carrier proteins, such as albumin.

Sulfhydryl Reactive Agents

A molecule which is not inherently sulfhydryl reactive may still be conjugated to the photactivated proteins of the present invention by means of a bifunctional crosslinking agent whic bears both a group reactive with the molecule of interest and a sulfhydryl reactive group. This agent, may be reacted simultaneously with both the molecule of interest (e.g., through an amino, carboxy or hydroxy group) and the photoactivate protein, or it may be used to derivatize the molecule of interest to form a partner molecule which is then sulfhydryl reactive b virtue of a moiety derived from the agent, or it may be used to derivatize the photoactivated protein to make it reactive wit the molecule of interest. Sulfyhdryl reactive agents include alpha-haloacetyl compounds such as iodoacetamide, maleimides such as N- ethylmaleimide, mercury derivatives such as 3,6-bis- (mercurimethyl)dioxane with counter ions of acetate, chloride o nitrate, and disulfide derivatives such as disulfide dioxid derivatives, polymethylene bismethane thiosulfonate reagents an crabescein (a fluorescent derivative of fluorescein containin two free sulfhydryl groups which have been shown to add acros disulfide bonds of reduced antibody) .

Alpha-haloacetyl compounds such as iodoacetate readily reac with sulfhydryl groups to form amides. These compounds have bee used to carboxymethylate free thiols. They are not strictly S specific and will react with amines. The reaction involve nucleophilic attack of the thiolate ion resulting in displacement of the halide. The reactive haloacetyl moiety, X- CH ₂ CO- , has been incorporated into compounds for variou purposes. For example, bromotrifluoroacetone has been used fo F-19 incorporation, and N-chloroacetyliodotyramine has bee employed for the introduction of radioactive iodine int proteins. Maleimides such as N-ethylmaleimide are considered to b fairly specific to sulfhydryl groups, especially at pH value below 7, where other groups are protonated. Thiols underg

Michael reactions with maleimides to yield exclusively the adduct to the double bond. The resulting thioet-her bond is very stable and cannot be cleaved under physiological conditions. They also react at a much slower rate with amino and imidazoyl groups. At pH 7, for example, the reaction with simple thiols is about 1,000 fold faster than with the corresponding amines. The characteristic absorbance change in the 300 nm region associated with the reaction provides a convenient method for monitoring the reaction. These compounds are stable at low pH but are susceptible t hydrolysis at high pH.

See gene ,lly Won-" Chemistry of Protein Conjugation and Cross-linkin CRC Prest, Inc., Boca Raton, 1991: Chapters 2 and 4) .

Conjugates and Their Uses In vitro Immunodiagnosis . In one embodimt \';, an antibody is conjugated to a detectable label for use in in vitro immunodiagnosis. The label may be a radiolabe.. fluorophore, or enzyme which is directly or indirectly conjugatable to a free thiol group of the photoactivated antibody. The sample may be of clinical (e.g., blood, urine, semen, or cerebrospinne fluid, or a solid tissue or organ or nonclinical soil, water, food) nature. The assay may be qualitative or quantitative, and in any desired format, including sandwich and competitive formats. Numerous immunoassay formats, labels, immobilization techniques, etc., are disclosed in the following publications, hereby incorporated by reference herein: 0\'Sullivan, Annals Clin.

Biochem., 16:221-240 (1976); McLaren, Med. Lab. Sci., 38:245-51

(1981); Ollerich, J. Clin. Chem. Clin. Biochem., 22:895-904

(1984); Ngo and Lenhoff, Mol. Cell. Biochem., 44:3-12 (1982). Immunoimaging. An immunoconjugate may also be used for in vivo immunoimaging. For this purpose, the antibody must be labeled by means which perm external visualization of its position. Typically, an immunoimaging agent will be an antibody labeled directly (as with Technetium) or indirectly (as with chelated Indium) with a suitable radioisotope. After injection into the patient, the location of the conjugate may be tracked by a detector sensitive to-particles emitted by the radiolabel,

e.g., a gamma-scintillation camera in the case of a gamma emitter.

Immunotherapy . For immunotherapy, the antibody may be conjugated to a suitable radioisotope, drug or toxin. In Vivo Use, Generally. Whether for immunoimaging or for immunotherapy, the conjugate must be introduced into the patient. Preferably, it is introduced by injection. Typically, the agent is administered intravascularly (intravenously or intraarterially) or intrathetically, often by infusion. In additon, in appropriate cases the conjugate may be introduced subcutaneously, submucosally, intramuscularly, intracranially, or by other accepted routes of drug administration.

Further discussion of techniques of immunoimaging and immunotherapy is found in standard works such as Chatal, Monoclonal Antibodies in Immunoscintography (CRC Press: 1989) ; Magerstadt, Antibody Conjugates and Malignant Disease (CRC Press 1981); and Burchiel et al., Radioimmunoimaging and Radioimmunotherapy (Elsevier) . A more general review of immunological methods appears in Harlow and Lane, Antibodies: A Laboratory Manual (Cold Spring Harbor Laboratory: 1988) .

Miscellaneous . Other proteins may be used in a similar manner to an antibody as a targeting agent for diagnosis, imaging or therapy, when they have sufficiently specificity and affinit for an appropriate marker. See, e.g., Bakker, et al., Recepto scintigraphy with a Radioiodinated somatostatin analog: Radiolabeling, Purification, Biologic Activity and in Vivo Application in Animals. J. Nucl., Med., ϋ: 1501-1509, 1990. Enzymes, lectins, and various biological receptors can serve as binding agents. TPk, EGF, NSE, TPA, B2M, PHI and fibrin ar known to be taken up preferentially by certain tumors.

Other Activation Methods

Lunec, et al. report the formation of IgG aggregates b three methods: (a) a mixture of copper sulfate and hydroge peroxide, (b) arachidonic acid and (c) photolysis. The mechanis is thought to be through the generation of free radicals tha attack disulfides and cause disruption of the molecular structur of the molecule. They also suggest the use of enzymati

processes (e.g., superoxide dismutase, catalase and glutathione peroxidase) to generate free radicals. The present inventi extends to the enhancement of photoactivation of disulfide containing proteins by methods, other than photolysis, which generate free radicals which attack disulfides.

EXAMPLES

MATERIALS AND METHODS FOR EXAMPLES 1-11

A. UV Apparatus

A Conrad-Hanovia low pressure quartz mercury-vapor lamp (Ace Cat # 12128) with a water cooled quartz jacket was used. Samples were irradiated for various lengths of time at a distance of 3 cm from the water jacket (5 cm from the source) .

B. Radiochemical Analysis

The radiolabelled proteins were analyzed by size exclusion high pressure liquid chromatography (SE-HPLC) using a 7.8 x 300 mm TSK-3000SW analytical column and modified soft gel chromatography using a Sephadex G-50 spin-column. The SE-HPLC column is useful for quantitating relative amounts of aggregates, monomeric- protein, fractionation products, unreacted reduced technetium complex, free pertechnetate and bound reduced technetium species. The Sephadex G-50 spin column is a simple and fast method for determining the percent stable protein bound radioactivity.

C. Radioimmunoassay (RIA) To evaluate the immunoreactivity of Mab 170 after photolysis and subsequent radiolabelling, an RIA utilising a rabbit anti-MAb 170 antibody was immobilised on polystyrene tubes. Immunoreactivity indices were obtained by analysing the competition of the various dilutions of the test or reference sample with 1-125 MAb 170. The index was derived from the ratio of the concentration of reference standard that gave 50% inhibition to the concentration of the test sample that gave 50% inhibition:

Immunoreactivity Index = IC-50 Standard

IC-50 Sample

Example 1

Radiolabelling of Bovine Serum Albumin (BSA) and Huma Transferrin

The first radiolabelling experiment using UV irradiation wa carried out on BSA and human transferrin. Three differen concentrations of BSA or human transferrin solution (1, 5, 1 mg/ml) in 0.05 M phosphate buffer pH 7.0 were prepared. Tw hundred μl of the protein solution was pipetted into a 12x75 m quartz test-tube and irradiated as described in method (A) abov for 0, 5, 15 or 30 minutes. After the irradiation, lOOμl o stannous tartrate solution (>20μg of Sn ⁺² ) was added, followed b 300μl of Tc-99m sodium pertechnetate. The mixture was allowe to incubate for 30 minutes and was then analysed by Sephadex G-5 spin column to determine the % stable protein radiolabelling. The results are summarised in table 1.

The results show that there is an increase in the percen radiolabelling of the proteins with reduced pertechnetate afte UV irradiation. At higher protein concentrations, the reactio is more efficient, resulting in higher percent radiolabellin with a shorter irradiation time.

Example 2

UV Irradiation of MAb 170 in quartz test tubes MAb 170 (Biomira, Inc., Edmonton) at ~6 mg/ml in 0.05M PB pH 7.0 was irradiated under the same setup as in method (A above. The MAb (200 μl) was irradiated in a quartz test tube (1 x 75 mm) for 30 minutes and then 100 μl of stannous tartrate wa added followed by 300 μl of Tc-99m sodium pertechnetate. Th reaction mixture was assayed by SE-HPLC 30 minutes later. Th percentage of radiolabelling species included 18% MAb aggregates 49.3% monomeric MAb, 9.0% low molecular weight protein specie and 23.0% Tc-99m buffer complex and Tc-99m pertechnetate.

Example 3

UV irradiation of MAb 170 in borosilicate glass vial

A nitrogen-purged 2 ml borosilicate glass tubular vial containing 200 μl of MAb 170 (6 mg/ml) and 10 μl of 0.1 M tartrate with 5mM stannous chloride was irradiated for 30 minutes under the same setup as in method (A) above. An equal volume of Tc-99m sodium pertechnetate was then added to the vial. After 30 minutes, the reaction mixture was analysed by Sephadex G-50 spin column. The SE-HPLC chromatogram obtained is shown in figure 1.

The results indicate that over 95% of the Tc-99m radioactivity is bound to the protein in the MAb 170 reaction.

The process is mild and does not result in excessive radiolabeled fragments or aggregates in the preparation, as shown by figure 1.

Example 4

UV absorption spectrum of quartz test tube and borosilicate glass vial

The UV absorption spectra of quartz and borosilicate glass test tubes were measured by an HP8452 UV/Vis spectrophotometer. The quartz glass showed good transmittance of UV in the 190 to 820 nm range. The borosilicate glass test tube had high absorbance below 300 nm so the bulk of the irradiation is being carried out with wavelengths greater than 320 nm UV. The difference in UV adsorption seen with the two types of glass indicate that the overall effects on the protein may be variable since shorter wavelengths may produce more damage than longer wavelengths. The preferred vessel for irradiation is a borosilicate glass test tube or tubular vial.

Example 5

Optimisation of amount of stannous ion for Tc-99m radiolabelling of MAb 170 using UV irradiation

MAb 170 (6 mg/ml) was mixed with different amounts of stannous tartrate (containing 5 to 4 μg of Sn ⁺² ) in a 2 ml nitrogen purged borosilicate glass tubular vial. The vial was

then irradiated under the same setup as in method (A) above for 60 minutes. An equal volume of Tc-99m sodium pertechnetate was added and the percent radiolabelling was measured by Sephadex G- 50 spin-column after 30 minutes. The results are summarised in table 2.

It is obvious from the data that the percent radiolabelling decreases with increasing amount of Sn ⁺² . The optimum amount of Sn ⁺² is in the range of 5 to 20 μg. The decrease in radiolabelling yield is probably due to the competition between the irradiated protein and the Tc-99m-Sn complex formed at high Sn concentrations.

Example 6

Effect of duration of UV irradiation on Tc-99m radiolabelling of MAb 170 and MAb B43 Mab 170 was pretinned by the Rhodes method and then reacted with pertechnetate. Pretinning buffer solution comprised 0.005 M Sn ⁺² in 40 mM KH phthalate, 10 mM NaK tartrate, pH 5.6. The antibody had a starting concentration of 1.7 mg/ml. Three parts antibody solution were mixed with two parts pretinning buffer solution to give a final protein concentration of 1 mg/ml antibody and a final Sn ⁺² concentration of 237 mg/ml. The reaction vial was purged with nitrogen, sealed and incubated at room temperature up to 24 hours. Sampling was done at 1, 3, 6, 12 and 24 hours. At the end of the 24 hour incubation, the remaining pretinned antibody solution was aliquoted in 2 mg aliquots and stored at -20 °C until used. Tc-99m labeling was accomplished by adding 2 mCi of Tc-99m in 0.5 ml of saline. After 30 minutes reaction time, the vial was analysed using SE- HPLC and Sephadex G-50 spin column. This conventional labeling method was compared with photoactivation.

MAb 170 (6 mg/ml, 200 μl) with 5 μg of Sn ⁺² ion in a 2 ml nitrogen purged borosilicate glass tubular vial was irradiated under the same setup as in method (A) above for 5 to 120 minutes. Another monoclonal antibody, MAb B43 (Biomira, Inc., Edmonton), was alsotested for suitability for Tc-99m radiolabelling with U irradiation. MAb B43 (5 mg/ml) was irradiated for different

periods of time under the same setup as in method (A) above (page 4) with 5 μg of stanr s ion. An equal volume of Tc-99m sodium pertechnetate was adαcd after the irradiation. The results are shown in figure 2 and Table 2A.

The results indicate that high radiolabelling yields can be obtained with UV irradiation. SE-HPLC on the 60 minute irradiation sample shows "95% monomeric IgG, confirming the results of the soft gel (Sephadex G 50 Spin) column. The optimal conditions for radiolabelling of B43 appear to be different from that of MAb 170 and this is probably due to the difference in amino acid composition of the antibodies. Note that the duration can be further reduced by increasing the intensity of irradiation

(e.g., by using a higher power output UV source, by moving the reaction solution closer to the source or decreasing the thickness of the glass irradia ion vessel walls) .

Example 7

Effect of protein concentration on Tc-99m radiolabelling of MAb 170 A series of samples of MAb 170 at 1 to 10 mg/ml with 5μl of stannous ion in 2 ml nitrogen-purged borosilicate glass tubular vials were irradiated for 30 minutes. An equal volume of Tc-99m sodium pertechnetate was added to each vial at the completion of irradiation and the percent radiolabelling assayed by Sephadex G-50 spin column. The percent radiolabelling yields are shown in table 3.

The results show that the radiolabelling yield is concentration dependent and increasing the amount of protein increases the amount of radiolabelling obtained.

Example 8

Effect of Irradiation on Apparent Size of Irradiate Antibody Molecule

SE-HPLC analysis (UV 280 nm trace) (Figure 3) of radiolabelled, irradiated MAb 170 indicates that the antibod

does not undergo any alteration in apparent size.

Example 9

Evaluation of immunoreactivity of MAb 170 after irradiatio and radiolabeling The MAb was irradiated and radiolabeled under the condition illustrated in Example 2. The immunoreactivity of untreated MAb, irradiated MAb and irradiated and radiolabeled MAb were the compared using a standard anti-idiotype RIA (described unde Materials and Methods, C. RIA) . The results are summarized i Table 4.

MAb 170 preparations treated by irradiation and subsequen radiolabelling with Tc-99m retain greater than 90% of thei immunoreactivity indicating that the two-step proces

(irradiation with subsequent radiolabelling) incurs minima damage to the antibody.

Example 10

Formation of Aggregates During Photolysis

MAb 170 was irradiated and radiolabeled as described i Example 2. A sample of the preparation was injected onto a SE HPLC column and the elution profile was followed using absorbanc at 280 nm. Figure 3 shows the UV 280 nm chromatogram of this MA and indicates that the antibody shows no aggregation (absence o a peak to the left of the main monomeric IgG peak) and minima fragmentation (presence of a minute peak to the right of the mai monomeric IgG peak) .

Example 11

Effect of pH

Two experiments have been done comparing MAb 170 in eithe PBS (10 mM phosphate buffered saline, pH 7) or 0.1 M sodiu potassium tartrate. pH 6 with irradiation at 3 cm from the wate jacket for 1.5 hour and 1 hour respectively. The percen labelling efficiency for the MAb in PBS is 92.2% and for the MA in tartrate is 91.6%. The difference is not significant Although we have not established an absolute preference for p

it appears that for MAb 170 at least, pH is not a factor in obtaining high labelling yit_--.d.

Example 101

Studies of the effect of UV wavelength on yield Different UV light sources have been used for the photoactivation reaction. A simple Hanovia UV lamp placed inside a quartz water jacket was used for the initial experiment. The principal emission from this light source was in the form of 254 nm wavelength. The Rayonet photochemical reactor was chosen for subsequent development studies, as it offers certain advantages over the Hanovia system. The Rayonet photoreactor can be fitted with 8 UV lamps inside the reactor chamber. The user can choose to turn on either 2, 4, 6 or 8 lamps, thus controlling the average intensity during the photoactivation. There is also a choice of 3 types of UV lamps covering different portion of the UV spectrum. This, coup ^*1 - with the use of different types of UV filters, offers a means for studying the effect of UV wavelengths on the efficiency of the photoactivation reaction.

The photoactivation reaction is carried out in the Rayonet photochemical reactor using a quartz test-tube as the irradiation vessel. The UV light from the reactor has to pass through a 2" x 2" narrow band pass-filter (254-365 nm) before it reaches the quartz tube. 0.5 mL of MAb-170 (5mg/mL in 50 mM PBS) is placed inside the quartz test-tube with 30 μg Sn ⁺² /mg MAb added. The solution is irradiated using different wavelength filters and then radiolabelled with Tc-99m sodium pertechnetate. The relative radiolabelling yield at each wavelength is normalize with the relative irradiation intensities at the different wavelengths.

Six experiments are describ- , in Table 101. For eac experiment, it states the filter\'s transmission wavelength, the nominal lamp wavelength, the lamp output, the relative lam intensity at the transmission wavelength, and the exposure time. It should be noted that the exposure time has been chosen t compensate for differences in intensity at the transmissio

wavelength.

Figure 4 is a bar chart depicting the yield of radiolabele antibody against the wavelength transmitted. Yield rise steadily from 254 to 313, and then dropped off sharply with a increase in wavelength to 334. Therefore, the most desirabl wavelengths are in the range of 250-320 nm.

Example 102

Sulfhydryl generation during photoactivation

The amount of free sulphydryl generated durin photoactivation is measured by reaction with Ellman\'s reagent. MAb-170 is photoactivated in a quartz test-tube using the Hanovi UV system. At different times post irradiation, a small aliquo is removed and allowed to react with Ellman\'s reagent. The U absorbance at 412 nm is measured and the molar cystein concentration is calculated suing a extinction coefficient o 13600.

The results are shown in Figure 5. The number of SH group per antibody molecule increases to almost 4 in about 15 minutes. Further increases are at a slower rate, e.g., to about 5.5 in th interval from (t + 15) to (t + 60) mins.

Example 103

Effect of sulfhydryl blocking on labeling yield

Free sulphydryl groups can be readily blocked by sulphydryl reactive chemicals such as iodoacetamide and N- ethylmaleimide A drop in radiolabelling yield after blocking the sulphydryl i an indication that the sulphydryl plays a role in th radiolabelling reaction.

MAb-170 (5 mg/mL in 0.1 M tartrate buffer) is placed in 2 mL borosilicate glass vial. The vial was capped, crimped an purged with nitrogen for 1 minute and irradiated at 300 nm fo 10 to 15 minutes. The radiolabelling was then performed at a 1: ratio with Tc-99m sodium pertechnetate using 5-10 μg Sn/mg MAb For the blocking, either 20 mM of iodoacetamide or 20 mM of N

ethylmaleimide was added immediately after irradiation and allowed to incubate for 15 to 30 minutes before radiolabelling. The radiolabelling yield was determined with SE-HPLC.

The results of sulfhydyl blocking with iodacetamide (IA) or N-ethylmaleimide (NEN) are shown in Table 102 and Figure 6.

The data clearly indicates the importance of the generation of free sulfhydryls to radiolabelling.

Example 104

Effect of sulfhydryl exchange on labeling yield Another indication of the involvement of the sulphydryl group in the radiolabelling reaction can be demonstrated by the interaction with cysteine. Sulphdryl exchange reaction may interfere with stability of the Tc-SH complex.

Tc-99M MAb-170 is prepared by either using stannous- reduction or photoactivation (30 minutes at 300nm) . A 6.6 mM stock solution of cysteine in dd H ₂ 0 is prepared. The Tc-99m MAb-170 is mixed with the cysteine solution to give MAb:cysteine molar ratio ranging from 1:10 to 1:900. The mixture is then analyzed by SE-HPLC at different times post incubation and the % of non-protein bound radioactivity measured.

Figure 7 shows the yield of radiolabeled cysteine on cysteine challenge at 5 minutes and 2 hours post-incubation for four different MAb:Cysteine ratios (1:10, 1:50, 1:300, 1:900) . Figure 8 compares the yield of radiolabeled antibody, despite cysteine challenge, for both stannous ion reduced an photoactivated antibody, at two (1:10, 1:50) different MAb:cysteine ratios. There does not appear to be a significant difference between the results for the stannous ion reduce antibody and the photoactivated antibody, suggesting that the two will have similar in vitro stability with respect to cysteine transchelation.

Example 105

Effect of stannous sources chosen as pertechnetate reducin agent after photoactivation

In conventional methods in which stannous ion is used t reduce both the pertechnetate and the disulfide bonds of th antibody to be labeled, the choice of stannous source is of grea importance. For example, in Rhodes\' method, stannous phosphat cannot be used, as the complex would be too stable and th antibody therefore would not be reduced. However, we have foun that a variety of stannous sources with varying levels o stannous complexing capacities can be used as the pertechnetat reducing agent after photoactivation. The amount of the stannou species should be optimized for optimal Tc-99m radiolabelling.

Several different stannous source were used as th pertechnetate reducing agent after photoactivation. A stoc solution of the stannous species was prepared and assayed for it stannous content using an iodometric assay. Different levels o stannous species were added to MAb 170 and irradiated for 3 minutes at 300 nm. The Tc-99m radiolabelling yield of th photoactivated MAb is then determined by SE-HPLC.

Figure 9 shows that superior radiolabeling yields ar obtained with Sn Tartrate and Sn pyrophosphate. The yield in S methylene diphosphonate buffer is high for a relatively narro range of Sn:MAb (w/w) ratios, which, however, is within th preferred range for this ratio. The yield for the reaction i Sn EDTA is consistently lower than when the reaction occurs i the two preferred buffers. While not shown on this Figure, th yield with stannous phosphate lies in between the methylen diphosphonate and the pyrophosphate data.

Example 106 Effect of MAb buffer on yield

MAb-170 was diafiltered into different buffer systems usin the Amicon ultrafiltration cell with YM30 membrane. The buffe used included 0.1 M Tris, pH 7.5, 0.1M tartrate buffer, pH 6. and 0.05 M sodium acetate, pH 5.5. 500 μL of the MAb is pu

inside the glass with 25 μg Sn/mg MAb. The solution is irradiated for 30 minutes at 300 nm. The solution is then radiolabelled with a 1:4 (MAb:pertechnetate) ratio and analyzed by SE-HPLC.

The results are shown in Table 106. High yields were obtained with all three buffers. Given that the stannous sources mentioned in the preceding example also did not interfere with the photoactivation of the antibody, it can be surmised that any buffer which does not substantially abso ^~ UV radiation of the preferred wavelengths can be used for the contemplated photochemical reaction. When the protein is to be labeled with technetium, and the technetium is provided in the form of pertechnate reduced with stannous ion, the buffer will also need to be compatible with the stannous ion reduction system (note that in the preceding example, the problem with EDTA was that it formed too strong a complex with the stannous ion) .

Example 107

Effect of stannous ion concentration on yield

The amount of stannous ion present in the photoactivated formulation can affect the radiolabelling yield of the final product due to the competition for reduced-Tc species with the photoactivated protein. Therefore, the amount of stannous ion in the preparation should be optimized so that there is enough to carry out complete pertechnetate reduction without adversely affecting the radiolabelling yield.

Stannous phosphate solution is prepared by dissolving the stannous chloride crystals in 0.5M PBS pH 7.4. The solution is then assayed by iodometric stannous assay to determine the stannous level. The required amount of stannous phosphate solution to be added to the MAb-170 (5 mg/mL) is calculated (to give final stannous concentration of 15-30 μg Sn/mg MAb) and transferred to the MAb solution. 0.5 mL aliquots of the MAb solution is then injected into 2mL nitrogen purged glass vials. The vials are irradiated at 300 nm for different period of time and assayed for radiolabelling yield.

Figure 10 shows the radiolabeling yield, plotted agains time, for four different stannous ion concentrations.

Example 108

Effect of irradiation volume on yield The kinetics of the photoactivation and subsequen radiolabelling appears to have a biphasic pattern. There is a initial rapid increase in radiolabelling yield, followed by more gradual phase until it reaches a plateau. The change i irradiation volume during a bulk vessel irradiation is expecte to alter the kinetics of the reaction due to a lowering o average irradiation intensity/unit volume.

A 50 mL borosilicate glass vial (3.8 x 7.5 cm) is used a the irradiation vessel. Different volumes (5, 20, 40 mL) of MA 170 containing 25-30 μg Sn/mg MAb is added to the vial. The via is then capped, crimped and purged with nitrogen. The vial i irradiated at 300 nm and samples are removed at different tim period to assay for radiolabelling yield.

The results are shown in Figure 11. As expected, smalle irradiation volumes result in higher yields.

Example 109

Labeling of proteins. other than antibodies, b photoactivation

The use of a photoactivation reaction for Tc-99 radiolabelling is not limited to monoclonal antibodies; othe cystine containing proteins have also been radiolabelled with Tc 99m using this method. Thus, photoactivation has the potentia for use with a variety of proteins and peptide structures. Th conditions to obtain optimal radiolabelling yield may b different for different proteins due to the variability in th cystine and other amino acid contents and also the tertiar structure of the molecule.

In the comparative labeling experiment of Table 109, al proteins are in 50 mM PBS buffer, pH 7.4. 0.2 to 0.25 mL o protein solution containing 5 to 10 micro-g of Sn/mg protein wer

injected into a 2 mL glass vial. The samples were irradiated at 300 nm and subsequently radiolabelled with Tc-99m sodium pertechnetate. The protein solution in the control vial has the same amount of stannous ion but was not exposed to UV irradiation.

The results are shown in Table 109. The yield of radiolabeled transferrin and BSA is comparable to that for the three radiolabeled monoclonal antibodies.

Example 110 Small scale photoactivation by individual vial irradiation

Due to the simplicity of the method, photoactivation can be performed in a variety of configurations, from the simple individual vial irradiation using less than 1 mg of material up to grams scale for production purposes using either a bulk-vessel or a flow-throug} irradiation system. The sterility of the product can be easily maintained since a close irradiation system can be used during the reaction and no complicated post-reaction purification is required. Consequently, this method can be easily adapted for use in routine research laboratory or for scale-up production in the pharmaceutical industries.

The photoactivation is performed using a Rayonet photochemical reactor with 8 x 300 nm UV lamps installed. Precrimped borosilicate glass sterile, empty vials (2 mL) were purchased from Hollister Stier (Canada) . The vials are purged with nitrogen to maintain an inert atmosphere. A Rayonet merry- go-round unit is used as the sample holder which can accomodate up to 8 vials at a time. The unit is lowered into the cavity of the photoreactor and rotates inside the chamber to ensure uniform irradiation of all vials.

MAb-170 (5 mg/mL) is mixed with stannous phosphate (30 μg Sn ⁺² /mg MAb in 0.5 M phosphate buffer) and 1 mg aliquot (-0.2 mL) is injected into the sealed vials using a sterile disposable syringe. The vials are loaded onto the merry-go-round unit and lowered into the chamber of the photoreactor. The vials are

allowed to irradiate for 45 minutes rotating inside the chambe cavity. At the end of the irradiation time, the vials ar removed from the chamber, labelled with the contents and the store frozen at -20°C.

A control set is also prepared with the same amount of MAb 170 and stannous phosphate, incubated for the same period of tim but without any photoactivation. The radiolabelling yields fo both sets of MAb-170 are determined by SE-HPLC using on-lin radioactivity detection.

Example 111 photoactivation by bulk irradiation

For reactions requiring a larger scale, single bul irradiation vessel can be used as the reaction vessel. Efficien mixing of the solution inside the vessel is required to ensur uniform irradiation. The size of the reaction vessel i determined by the space available inside the photoreacte chamber. With our current photoreactor, a 300 mL vessel can b efficiently irradiated inside the chamber giving us a scale o up to 1.5 grams (using MAb at 5 mg/mL) .

(a) Photoactivation of MAb-170 using bulk vessel irradiatio

A 300 mL quartz bottle with 2 side arms ports is used as th reaction vessel. MAb-170 (60 mL at 5 mg/mL, 300 mg total) ar mixed with stannous phosphate (30 μg Sn ⁺² /mg MAb in 0.5 phosphate buffer) . The MAb solution is poured into the quart vessel with a magnetic stir bar and lowered onto a magneti stirrer located inside the photoreactor. The headspace of th vessel is purged with nitrogen during the irradiation, using th two side-arm ports. The solution is stirred during th irradiation and samples are withdrawn from the vessel at regula time-intervals and radiolabelled with Tc-99m sodiu pertechnetate. The radiolabelling yield is determined by SE HPLC. The results are shown in Table 111.

(b) Photoactivation of Human IgG (Gaminune N) with bul vessel irradiation

Human IgG (Gamimune N, Miles, Canada) is diluted to 5 mg/mL with 50 mM PBS pH 7.4. The human IgG is subjected to the same treatment as in (a) above, with the results illustrated in Table 111A.

Example 112

Photoactivation in recirculating flow-through system

Another option for the scale-up of the photoactivation reaction involves the use of a flow-through system for the irradiation of the protein. A glass or quartz coil can be prepared which will be located in the center of the photoreactor chamber. The protein solution can be stored in a vessel outside of the reaction chamber and the irradiation performed by pumping the solution through the coil, which is exposed to the UV light. Two variations of the format has been attempted. The first variation involves the recirculation of MAb solution through the coil back into the storage vessel. Efficient mixing of the solution is required for this procedure. Another variation involves a "one-pass" condition whereby the solution is pumped at a controlled flow-rate through the coil and collected in a separate collection bottle as the final product. The flow through irradiation allows for almost unlimited scale-up potential and eliminates the "volume effect" shown using a single irradiation vessel.

A PBS buffer (1.8 mL, 0.5 M, pH 7.4) was added to a stirred solution of MAb-170 (110 mL, 5 mg/mL). Then, 3.7 mL of stannous phosphate solution (3.3 mg stannous/mL solution; 12.1 mg stannous) was added. The material was irradiated at roo temperature in a borosilicate glass coil passing through Rayonet mini-reactor with 8 (300 nm) UV lamps. The pump spee was 20 RPM. Sample were removed at different time inteval during the irradiation, reacted with pertechnetate, and assaye for Tc-99m radiolabelling yield. The results are shown in Tabl 112.

Example 113 Photoactivation in one-pass flow-through system

Sample were removed at different time intervals during th irradiation and assayed for Tc-99m radiolabelling yield. Greate than 90% radiolabelling yields are routinely obtained.

Example 114 Preparation of lyophilized antibody

A lyophilized MAb kit can offer certain advantages such a better long-term product stability and ease of storage an shipment over a frozen formulation. The photoactivated MAb ca be lyophilized using a suitable freeze-drying cycle and th product obtained retains is biological and radiochemica characteristics. An inert bulking agent is usually added to th formulation to improve the appearance of the cake structure.

MAb-170 (110 mL, 550 mg) is photoactivated according to th recirculating flow-through design. At the end of the process inositol concentration of 0.5% (Final MAb concentration -= 2 mg 0.6 mL) . 2 mg dose of the MAb solution is dispensed into 5 m empty vial using an automatic dispenser and partially capped The vials are then transferred to a Virtus lyophilizer an lyophilized according to the following cycle: Pre-chill shelves for about 1.5 hours at -45°C Primary drying: Shelf temp. = -20 ± 1°C

Ramp time = 18 mins. Dwell time = 15 hours Vacuum level = 180 mT Secondary drying: Shelf temp. = 25 ± 1°C

Ramp time = 45 mins. Dwell time = 13 hours Vacuum level = higher than 180 mT

After lyophilization, the vials are capped automaticall inside the lyophilizer and crimped manually. The radiolabellin yield after lyophilization is measured by SE-HPLC. (Table 114

Example 115

Effect of photoactivation on antibody conformation an activity

For biologically active or receptor specific molecules, it is important that the process utilized for radiolabelling should not alter the reactivity of the molecule. Furthermore, the radiolabelled molecule should be both biochemically and radiochemically pure. For routine radiopharmaceutical production, it is desirable that the product will retain stable for a reasonable period of time and that the radiolabel remain stable both in-vitro and in-vivo.

For Tc-99m radiolabelled protein using the indirect approach, the conjugation reaction either targets specific amino acids or is nonspecific (e.g. photoactivatable azido chelates) . If the amino acids or conjugation targets happen to lie in the active sites of the molecules, the immunoreactivity of the molecule could be compromised. Additional manipulation of the molecule post reaction, e.g., post-reaction purification, will not only lower the overal yield of the product but also increase the risk of further reduction in immunoreactivity.

(a) SDA-PAGE and IEF profile of photoactivated MAb-170 The photoactivated MAbs are routinely monitored for its biochemical characteristics. SDS-PAGE (under reducing conditions) and isoelectric focusing are used to monitor the change in product qulaity. HPLC using a size exclusion column is used to quantify the extent of aggregation or fragmentation in the product.

MAb-170 is photoactivated using the individual-vial irradiation. The photoactivated MAb is then frozen at -20°C and an aliquot of the sample is removed for the analysis. SDS-PAGE and IEF are performed according to established procedures, with the results depicted in Table 115.

(b) SE-HPLC (UV) Analysis of photoactivated MAb=170

MAb-170 is photoactivated using the individual-vial irradiation. The photoactivated MAb is then frozen at -20°C and an aliquot of the sample is removed for teh analysis. SE-HPLC is performed using a TSK SW3000XL analytical column (7.8 x 100

mm) and a TSK SW3000 guard column. The eluate is monitored fo its UV absorbance at 280 nm using a Beckman programmable UV/Vis detector and integrated using the Beckman system gold software. The results are shown in Table 115A.

(c) SE-HPLC profiles of photoactivated MAb-170 and MAb-17

The radiochemical profile of the Tc-99m radiolabelle protein is routinely monitored by SE-HPLC which is capable o quantitating relative amounts of aggregates, nomomeric protein, fractionation products, unreacted reduced technetium complex, free pertechnetate and bound reduced technetium species.

MAb-170 and MAb-174 are photoactivated using individual vial irradiation. The photoactivated MAb are radiolabelled at specific activity of -30 mCi/mg with Tc-99m sodium pertechnetate. The SE=HPLC radiochemical profile are exhibited in Table 115B.

(d) Immunoreactivity of UV irradiated and radiolabelled MAb 170

The immunoreactivity of UV photoactivated MAb-170 i measured using two techniques. An anti-idiotype RIA is use whereby the UV-treated MAb-170 will compete with 1-125 MAb-17 standard for binding sites. A cell-line bioassay is also use to study the binding to MAb-170/bound antigen-expressing cel lines. The results for both assays on photoactivated Tc-99m MAb 170 is summarized in Table 115C.

(e) Stability sutdies on lyophilized photoactivated MAb-17 The change in biochemical profile of a lyophilize formulation of MAb-170 labeled by photoactivation is summarize in Table 115D.

(f) Stability of photoactivated radiolabelled MAb versu serum challenge Tc-99m MAb-170 is prepared by individual vial irradiatio and then radiolabelled with Tc-99m sodium pertechnetate. 6 μ of the radiolabelled MAb-170 is mixed with 300 μL of human seru and incubated at 37°C. At 24 and 48 hours post incubation, th

mixture is analyzed by SE-HPLC and the MAb-associated peak is collected by a fraction collector. The amount of radioactivit in the MAb fraction is counted and expressed as a % of total radioactivity. (Table 115E) .

Example 116

Effect of photoactivation on biodistribution of radiolabele antibody

MAb-170 is radiolabelled with Tc-99m using three different techniques. A stannous reduction technique is used according t a modified Rhodes method (Sn-treated) . MAb-170 is als radiolabelled using the flow-through photoactivation techniqu from a frozen formulation (UV-Flow) and also from a lyophilize formulation prepared using the individual vial irradiation (UV- Lyo) . 20 μg of the radiolabelled MAb are injected IV through th tail vein into normal Balb/C mice. Group of mice are sacrifice and the organ of interest dissected at different time period pos injection. The biodistribution of the three different compound are summarized in Figures 12A-12D.

All patents, patent applications, and publications cited i this specification, including Applicants \' prior applications, ar hereby incorporated by reference.

Table 1: Radiolabelling of BSA and Human Transferrin by UV Irradiation

% Protein Radiolabelling Protein/Concentration Irradiation Time (minutes)

0 5 15 30

Table 2: Effect of amount of Sn ⁺² ion on UV radiolabelling of protein

Amount of Sn ⁺² % Radiolabelling

5 μg 96.9

10 μg 95.5

20 μg 92.1 40 μg 84.8

ND - Not Done

For Rhodes method, time indicates time for pretinning process

For Irradiation method, time indicates amount of time exposed to UV prior to addition of Tc-99m.

Table 3: Effect of MAb 170 concentration on UV induced radiolabelling

Concentration

(mg/ml) % Radiolabelling

1 63.8

3 96.2

6 98.0

10 99.0

Table 4: Immunoreactivity of Irradiated MAb 170

MAb Treatment Process Radioactive Immuno-

Time reactivity (minutes) Indexl

170 None None No 1.00 (Ref.)

170 Irradiation 30 No 0.97

170 Irradiation 30 Yes 0.91

1 Value obtained from anti-idiotype RIA. RRIIAA ccoonnssiissttss o tubes coated with rabbit anti-MAb 170 and subsequen competition between 1-125 labeled MAb 170 and test o reference standard sample. Immunoreactivity Index derive from:

IC-50 Standard IC-50 Sample

Table 51s MAJOR UV ABSORBING AMINO ACIDS

Molar Extinction Coefficient Values*

*Ref.: E. Mihalyi, J. Chem. Eng. Data 13:179-182, 1968

Table 52: MAJOR UV ABSORBING AMINO ACIDS

Relative Absorption Properties in Immunoglobulins*

^♦ Assuming TRP = 3.83%; TYR = 5.97%; PHE = 4.29%; CYS = 2.20\'

Table 53: MAJOR UV ABSORBING AMINO ACIDS

Relative Absorption Properties*

* Assuming equimolar \'concentrations a e value at 254 nm + 258 nm used for PHE e value at 240 nm used for CYS b e value at 266 nm estimated as 150 for CYS c e value at 290 nm + 310 nm estimated at 75 for CYS

Table 101: WAVELENGTH EFFECTS FOR PHOTOACTIVATION WITH USE OF NARROW BAND-PASS (INTERFERENCE) FILTERS

Table 102: ROLE OF SULPHYDRYL GROUPS IN PHOTOACTIVATION OF MAb-170 FOR Tc-99m RADIOLABELLING

Sulphydryl Blocking Agents

Protein Photoactivation

MAb-170 10 min./300nm

10 min./300nm

MAb-170 10 min./300nm 10 min./300nm

a IA = Idoacetamide b EM = N-Ethylmaleimide

Table 106: EFFECT OF MAb BUFFERS ON PHOTOACTIVATION

Buffer Radiolabelling

0.1 M Tris, pH 7.5 88, 6% Control* 5, 3%

0.1 M Tartrate, pH 6.0 84, Control* 28, 0.05 M Na acetate, pH 5.5 93.7^

* Control = without photoactivation

Table 109: PHOTOACTIVATION OF PROTEINS FOR DIRECT Tc-99 RADIOLABELLING

Protein Concentration Duration of reaction % Radiolabelle Protein

BSA 5.0

Table 109A: Cysteine Content of Selected Proteins

Albumin 5.09g/100g protein ("28 cysteines [as disulfides or free sulfhydryls] /molecule of albumin)

Transferrin 5.07g/100g protein ("36 cysteines [as disulfides or free sulfhydryls] /molecule of transferrin)

IgG 2.20g/100g protein ("27 cysteines [as disulfides or free sulfhydryls] /molecule of IgG)

IgM 1.58g/100g protein ("122 cysteines [as disulfides or free sulfhydryls] /molecule of IgM)

IgA 2.10g/100g protein ("31 cysteines [as disulfides or free sulfhydryls] /molecule of IgA)

Table 110: Radiolabelling yield for Tc-99m radiolabelled MAb=170 pre ^p ared by individual vial irradiation

Sample ID % Radiolabelling Yield

Photoactivated > 90? MAb-170

Control MAb-170 < 10?

Table 111: Radiolabelling yield of Tc-99m MAb-170 using bulk vessel photoactivation

Irradiation Time % Radiolabelling Yield

15 minutes 81.5.

30 minutes 92.83

45 minutes 95.43

60 minutes 95.83

75 minutes 96.6?

Table 111A: Radiolabelling yield of Tec-99m Human IgG using bulk vessel photoactivation

Irradiation Time % Radiolabelling Yield

5 mins. 54 . 9%

15 mins. 77 . 4%

30 mins. 85 . 7%

60 mins. 90 . 7%

120 mins. 90 . 9%

Table 112: Radiolabelling yield of Tc-99m MAb-170 using recirculation flow-through photoactivation

Irradiation Time % Radiolabelling Yield

Time 0 2.1% 15 minutes 93.0% 30 minutes 98.4% 60 minutes 95.8% 90 minutes 96.7%

Table 114

Sample Radiolabelling

Photoactivated Pre-Lyophilized 94.73

Photoactivated Lyopholized 91.3?

Table 115A

Sample #1

Peak# Retention Time Identification % UV 280nm Absorbanc

1-3 5.8-6.8 min. Aggregates 2.64 ± 0.04%

4 8.3 min. Tc-99m MAb-170 96.29 ± 0.19%

5-6 9.5 - 10.1 Low M.W. Species 1.07 ± 0.16%

Sample #2

Peak# Retention Time Identification % UV 280nm Absorbanc

1-2 5.7-6.9 min. Aggregates 1.63 ± 0.54% 3 8.3 min. Tc-99m MAb-170 97.67 + 0.96%

4-5 9.4-10.2 min. Low M.W. Species 0.70 ± 0.42%

Sample #3

Peak# Retention Time Identification % UV 280nm Absorbanc

1-2 5.9-7.5 min. Aggregates 4.42 ± 0.45%

3 9.0 min. Tc-99m MAb-174 95.19 ± 0.92%

4 10.0 min. Low M.W. Species 0.39 ± 0.69%

Table 115B

Sample #1

Peak # Retention Time Identification %UV 280nm Absorbance

1-2 5.8-7.0 min. Aggregates 3.21 ± 0.75% 3 8.3 min. Tc-99m MAb-170 92.12 ± 1.40%

4-6 9.5-11.7 min. Low M.W. Species 4.00 ± 0.67% 7 20.7 min. Retained Tc-99m Species 0.67 ± 0.02%

Sample #2

Peak# Retention Time Identification UV 280nm Absorbance

1-2 5.9-7.1 min. Aggregates 3.82 ± 0.22% 3 8.4 min. Tc-99m Mab-170 90.11 ± 0.07%

4-6 9.5-11.6 min. Low M.W. Species 4.55 ± 0.03% 7-8 20.2-22.9 min. Retained Tc-99m

Species 1.52 ± 0.18?

Sample #3

Peak # Retention Time Identification UV 280nm Absorbance

1-2 5.9-7.5 min. Aggregates 6.03 ± 1.25% 3 9.2 min. Tc-99m MAb-174 87.09 ± 2.60%

4-5 10.2-11.6 min. Low M.W. Species 5.01 ± 1.06% 6-7 21.4-24.0 min. Retained Tc-99m

Species 1.87 ± 0.58?

Table 115C

Parameter Test Method Result Comment

Immuno- Immuno- Anti- >0.8 A---CφtafcO chemical reactivity idiotype Reference

Quality RIA Ratio Value

Immuno- >0.9 AcceptabL histo- Reference chemical Ratio Value

Table 115D

Table 115E

Incubation Time % MAb Associated Radioactivity