Bioconjugate Chem. 1995, 6, 139-144
139
Recombinant Metallothionein-Conjugated Streptavidin Labeled with lssRe and 99mTc F. Virzi, P. Winnard, Jr., M. Fogarasi, T. Sano,? C. L. Smith,? C. R. Cantor,? M. Rusckowski, a n d D. J. Hnatowich* Department of Nuclear Medicine, University of Massachusetts Medical Center, Worcester, Massachusetts 01655, and Center for Advanced Biotechnology and Departments of Biomedical Engineering and Biochemistry, Boston University, Boston, Massachusetts 02215. Received June 27, 1994@
Consideration is now being given to the use of avidin (or streptavidin) and biotin for radiotherapy of tumor. Accordingly, the goal of this study was to radiolabel a mouse metallothionein-streptavidin fusion protein with lssRe and to compare its properties to those of the same fusion protein radiolabeled with 99mT~. A recombinant metallothionein-streptavidin fusion protein was radiolabeled by transchelation with 99mTc-and lS8Re-glucoheptonate. Labeling efficiency, which was not optimized for either radionuclide, was approximately 60% for 99mTcand 20% for lseRe. Radiochemical purity was demonstrated by size exclusion HPLC both by nearly quantitative shifts of the lssRe label to higher molecular weight upon the addition of biotinylated antibody and by the absence of a shift with biotinsaturated 188Re-metallothionein-streptavidin.Stability of the labels in 37 "C serum was evaluated by comparing the HPLC radiochromatograms of serum samples both before and after the addition of biotinylated antibody. The lasRe label behaved like 9 9 m Tin ~ that the same peaks were evident, including one prominent peak due to labeled cysteine. Recoveries during HPLC analysis of serum samples showed that oxidation rates to perrhenate and pertechnetate were identical. However, instability to cysteine challenge was greater for lSsRe; for example, the loss of label to cysteine after 24 h under one set of conditions was 41% for ls8Re and 22% with 9 9 m T ~Analysis . by HPLC of liver and kidney homogenates from mice administered the labeled antibodies were qualitatively and, in large measure, quantitatively independent of label. Biodistributions a t 5 h in normal mice were statistically identical between the two labels in blood and in most tissues. In conclusion, streptavidin may be radiolabeled with radiorhenium using recombinant mouse metallothionein as a bifunctional chelator, and under one set of labeling conditions a t least, lssRe showed similar in vitro and in vivo behavior to that of 9 9 m Tlabeled ~ to the same fusion protein.
INTRODUCTION
Pretargeting of tumor in which unlabeled but biotinylated antitumor antibodies are administered prior to the administration of radiolabeled avidin or streptavidin has been considered as one promising approach toward improving the tumorlnormal tissue ratios for diagnosis (1-5). Results have been generally favorable, and as such, consideration is now being given to the use of this and related pretargeting approaches for radiotherapy (6). Accordingly, this investigation was concerned with radiolabeling streptavidin with rhenium-188 (lssRe), a therapeutic radionuclide, via mouse metallothionein serving as a bifunctional chelator. The metallothioneins are a series of single-chain proteins, each of about 6-7 kDa, found in many species and believed to function as agents of metal detoxification (7). Mammalian metallothioneins contain about 20 cysteine residues distributed fairly uniformly along the peptide chain (8). As a result, normally seven divalent metal ions (usually cadmium or zinc) are bound per metallothionein molecule (7). The number of cysteine residues in metallothionein and, in particular, the large number which are present as next-to-nearest neighbors, has led to investigations of this protein as a bifunctional chelating group for the labeling of antibodies with tech-
* To whom correspondenceshould be addressed. Tel: (508) 856-4256. Fax: (508) 856-4572. + Boston University. Abstract published in Advance ACS Abstracts, December 1, 1994. @
netium-99m (99mTc) (9, 10). Thus, metallothionein, free from cadmium and saturated with zinc to protect the free sulfhydryls, was conjugated to the B72.3 IgG antibody using a heterobifunctional crosslinking agent and effectively radiolabeled with 99mTc(11). Metallothionein may therefore be a n attractive chelator to label streptavidin with lesRe for pretargeting applications for two reasons. Because metallothionein is itself a protein, it is possible to fuse it to other proteins through recombinant DNA technologies. Mouse and human metallothionein genes have been fused to genes coding for human growth hormone (12),somatostatin (13),and the Fab' domain of the S107 antibody (9). Furthermore, the chemical properties of rhenium are similar to those of technetium since both elements are members of the same family in the periodic table of the elements. Thus, it may be possible to radiolabel metallothionein-containing fusion proteins with lsSReby methods which are similar to that developed for 99mTc. Rhenium-188 is an attractive radionuclide for antibodymediated radiation therapy (14). In addition to a P-ray of intermediate energy (2.12 MeV) and a suitable physical half-life (0.70 days), this radionuclide is available as a radionuclide generator product by decay of its 70-day parent, tungsten-188 (ls8W). Recently, an expression system for a cloned streptavidin gene was developed in which active streptavidin may be efficiently expressed in Escherichia coli (15). Thereafter, it was possible to fuse a coding sequence for mouse metallothionein to the streptavidin gene such that this target protein was expressed as a streptavidinmetallothionein chimera. The purified chimera consists
1043-1802/95/2906-0139$09.00/0 0 1995 American Chemical Society
Virzi et al.
140 Bioconjugate Chem., Vol. 6,No. 1, 1995
of four identical subunits; each subunit was shown to bind one biotin and approximately seven cadmium ions, thus demonstrating that both the streptavidin and metallothionein moieties are fully functional (16). We report herein on a method for radiolabeling the recombinant streptavidin-metallothionein fusion protein with laaReand on the properties of the label in vitro and in vivo in mice. EXPERIMENTAL PROCEDURES
All reagents used in this investigation, sodium citrate, acetic acid (Aldrich, Milwaukee, WI), stannous chloride dihydrate, sodium glucoheptonate, and horse kidney metallothionein (Sigma Chemical Co., St. Louis, MO), were used without purification. Technetium-99m pertechnetate was obtained from a 9 9 M ~ - 9 9 mgenerator T~ (NEN Dupont, Billerica, MA). A 100 mCi radionuclide laeWla8Regenerator was kindly provided by Dr. F. F. Knapp, Jr., Oak Ridge National Laboratory. The B72.3 IgG antibody was a gift from Dr. John Rodwell (Cytogen Corp., Princeton, NJ). The recombinant metallothionein-streptavidin fusion protein was prepared as described previously (16).
Radiolabeling of Uncoqjugated Metallothionein. The conditions required to radiolabel metallothioneinstreptavidin with laaRewere first investigated using free, unconjugated, metallothionein. The metallothionein was of horse kidney origin and was used as obtained (i.e., without purification from its zinc and cadmium content, listed by the manufacture to be 4-7% by weight). The protein was radiolabeled by transchelation from laaRelabeled citrate. The laaRewas obtained by elution of the laaW-laaRe generator with 5-10 mL of 0.15 M NaCl solution. The citrate complex of laaRewas prepared by adding 50 pL (180 pCi) of lsaRe-perrhenate in saline to 50 pL of a 80 mg/mL solution of sodium citrate, pH 6.0, to which was added 20 pL of a fresh 20 mg/mL solution of stannous chloride dihydrate in 10% acetic acid. The solution was left undisturbed for 30 min a t room temperature and was then adjusted to pH 4.8 by the addition of 10 pL of a 1M solution of sodium acetate, pH 6.0. Strip chromatography with ITLC-SG (Gelman Sciences, Ann Arbor, MI) using acetone eluant, in which perrhenate migrates, and using saline eluant, in which both labeled citrate and perrhenate migrate, was used to determine the radiochemical purity of the labeled citrate complex. Ten microliters of a fresh la8Re-citrate solution was added to an equal volume of metallothionein in saline to give a final protein concentration in the range 0-2.5 mgl mL. Aliquots were removed a t various time points over 24 h for analysis by strip chromatography.
Radiolabeling of the Metallothionein-Streptavidin Fusion Protein. Each fusion protein consisted of four identical subunits of 190 amino acids with one metallothionein bound to each subunit by a 10-amino acid tether (16). The fusion protein was saturated with zinc ions to protect the sulfhydryls from oxidation. Attempts to label the construct with laaRevia the citrate complex were unsuccessful; therefore, the glucoheptonate complex was investigated for this purpose. The glucoheptonate complex was prepared by dissolving 250 mg of sodium glucoheptonate in 1.0 mL of saline and adjusting the pH to 4.5 with 6 M HC1. To 25 pL of la8Re-perrhenate generator eluant (90-160 pCi) was added 40 pL of the freshly prepared glucoheptonate solution, and the pH was adjusted to 5.0 with 10 pL of 1 M sodium bicarbonate, pH 10. To this was added 10 pL of a freshly prepared 25 mg/mL solution of stannous chloride dihydrate in 10% acetic acid. The solution was
incubated a t room temperature for 1h and analyzed by strip chromatography as for laaRe-citrate. Procedures for the preparation of 99mTc-labeledglucoheptonate were similar. Several vials, each containing 200 mg of sodium glucoheptonate and 0.14 mg of stannous chloride dihydrate adjusted to pH 5.5 with dilute HC1, were purged with nitrogen, sealed, and freeze dried. The vials were stored a t -20 "C until needed o r for up to 60 days. To one vial was added 3.0 mL of nitrogenpurged saline, and 25 pL of this solution was added to 50 pL (1.4-2.6 mCi) of 99mTc-pertechnetategenerator eluant and, after 15 min a t room temperature, analyzed by strip chromatography as for lsaRe-glucoheptonate. The recombinant metallothionein- streptavidin fusion protein was concentrated by ultrafiltration to 0.45 mgl mL in 0.1 M sodium phosphate buffer, pH 6.5 (Centriprep-30, Amicon, Beverly, MA). A 25 pL aliquot of the metallothionein-streptavidin fusion protein solution was added to a n equal volume of either 9 9 m T(460-860 ~pCi) or la8Re-glucoheptonate(26-47 pCi). The 99"Tc preparation was purified over a 0.6 x 21 cm column of Sephadex G-50 after 1h of incubation at room temperature, while the lsaRe preparation was purified in the same manner but after 2 h of incubation. In both cases, the radiolabeled fusion protein appeared in the column void volume. Radiochemical purity of the labeled protein was determined by high-performance liquid chromatography (HPLC) analysis using a single 1.6 x 29 cm column of Superose-12 (Pharmacia-LKB, Piscataway, NJ) and an in-line UV and radioactivity monitors. The column was eluted at 0.5 m u m i n with 0.1 M phosphate-buffered saline, pH 7 eluant. Analysis was performed on the protein with and without the addition of a large (12:l) molar excess of biotinylated B72.3 IgG antibody. Radioactivity shifting to higher molecular weight after the addition of the biotinylated antibody indicates the binding of the labeled metallothionein-streptavidin fusion protein to the biotinylated antibody. The biotinylated antibody was prepared as previously described with an average of six biotin groups per molecule as determined spectrophotometrically by measuring the displacement of 2-(4'-hydroxyazobenzene)benzoic acid (HABA) from avidin (1). In Vitro Serum Stability Studies. Preparations of 9 9 m T ~and - la*Re-labeled metallothionein-streptavidin fusion protein, after Sephadex G-50 purification, were incubated a t 37 "C in fresh human serum a t a concentration of 10 pglmL. In both cases, stability was assessed by size exclusion HPLC analysis with and without the addition of biotinylated B72.3 antibody. Mouse Studies. Each normal CD-1 mouse, 28-37 g (Charles River Laboratories, Wilmington, MA), received via a tail vein 11pg of the fusion protein labeled with either *Tc (365 pCi) or lSsRe(33 pCi). The animals were sacrificed 5 h later by spinal dislocation after metofane anesthesia. Tissues were removed, rinsed with saline, and after weighing, were counted along with blood samples in a NaI(T1) well counter against standards of the injectates. The radioactivity counts were converted into percentage of the injected dose per gram of blood or tissue corrected to a 25 g whole body weight. Samples of urine and serum were analyzed by HPLC using the Superose-12 column in which 0.3 mL fractions were collected for counting in the well counter. RESULTS
Radiolabeling of Unconjugated Metallothionein. Following the procedure described in the previous section, it was possible to achieve labeling efficiencies of greater than 95% in the preparation of laaRe-citrate. Further-
Technical Notes more, as determined by strip chromatography, the radiochemical purity was still greater than 90% after 2 h at room temperature, with the decrease due entirely to the accumulation of perrhenate. Horse kidney metallothionein was successfully radiolabeled with ls8Reusing the citrate complex. Labeling efficiencies varied from about 30% at 1 h to 60% a t 24 h of incubation a t metallothionein concentrations of about 1 mg/mL or greater. Radiolabeling of Metallothionein-Streptavidin Fusion Protein. Although it was possible to radiolabel unconjugated metallothionein with ls8Re-citrate, no labeling of the metallothionein-streptavidin fusion protein was achieved using citrate as transchelator. Therefore, glucoheptonate was considered as a n alternative. Radiochemical purities of both 99mTc-and ls8Re-labeled glucoheptonate, prepared as described, were routinely greater than 90%. The 99mTc-glucoheptonatepreparations showed no change in radiochemical purity after 2 h a t room temperature. In the lssRe case, the radiochemical purity was still 90% after 2 h at room temperature with the decrease due entirely to the accumulation of perrhenate. Possibly because streptavidin does not possess cysteine residues (16 ) ,the “nonspecific”transfer of reduced 9 9 m Tor ~ lssRe to streptavidin was not a concern; labeling studies under identical conditions showed less than 3%of lssRe bound to unconjugated streptavidin. Under the conditions described for the radiolabeling of the fusion protein with 99mT~and lssRe-glucoheptonate, the labeling efficiencies, determined by Sephadex G-50 chromatography, averaged 61 f 10% (standard deviation, N = 5) and 19 f 3% (standard deviation, N = 5) for 99mTcand le8Re, respectively. The ls8Re-metallothionein-streptavidin fusion protein complex was found to be stable to storage a t room temperature for a t least 2 h. Figure 1 presents radiochromatographic profiles obtained by size exclusion HPLC of 9 9 m T (Figure ~1A) and ls8Re-labeled metallothionein-streptavidin (Figure 1B) with (Figure 1A2 and 1B2) and without (Figure 1Al and 1B1) the addition of a large molar excess of biotinylated B72.3 antibody. In the case of both radiolabels, the profiles of the labeled proteins show several radiolabeled species but one prominent peak in the absence of the biotinylated antibody. Following addition of the antibody, however, a large and almost quantitative shift to the void is evident for both 9 9 m Tand ~ lssRe. As such, the majority of both radiolabels must be on streptavidin, most probably via the metallothionein groups. No shift was apparent when the same study was repeated with either 99”Tc-labeledmetallothionein-streptavidin saturated after preparation with a 10-fold molar excess of biotin (unpublished observations) or the lssRe-labeled fusion protein prepared in the identical fashion (Figure 1B3). Figure 2 shows the W profile of metallothioneinstreptavidin a t 280 nm (Figure 2A) which has the same general shape as the radioactivity profiles (Figures 2, l A l , l B l , and 1B3) and that the identical shift to higher molecular weight is evident upon the addition of biotinylated antibody. In Vitro Serum Stability Studies. Purified samples of 99mT~and 188Re-labeledmetallothionein-streptavidin fusion proteins were incubated a t 37 “C in fresh serum, and aliquots were removed periodically for analysis by HPLC with and without the addition of biotinylated antibody. Figure 3 presents radiochromatograms of the purified preparations themselves and of serum incubates after 1 and 24 h of incubation with and without the addition of biotinylated antibody (B-IgG). The percent
Sioconjugate Chem., Vol. 6,No. 1, 1995 141 A
B
I\
i:, I
\
3 _ _ I _
ELUTION VOLUME
Figure 1. Radiochromatographic profiles obtained by size exclusion HPLC analysis of 9gmTc- and 188Re-labeledmetallothionein-streptavidin fusion proteins without (99mT~ (Al); ls8Re (Bl)) and with PgmTc(A.2);ls8Re (B2)) the addition of biotinylated antibody. In the case of lSsRe,the radiochromatogram obtained afler adding biotinylated antibody to biotinsaturated metallothionein-streptavidin fusion protein is also presented (B3).
ELUTION VOLUME
Figure 2. W absorbance profiles obtained by size exclusion HPLC of the metallothionein-streptavidin fusion protein without (A) and with (B) the addition of biotinylated antibody.
recovery of radioactivity is listed in each chromatogram. In all chromatograms other than those of the preparation itself (i.e., injectate) in saline, a peak appears in fraction 76. In the case of 99mTcand, most probably lssRe as well, this peak is due to radiolabeled cysteine produced by transchelation of the label from the fusion protein to this thiol (17-19). It may be estimated from Figure 3 that 22% of the 9 9 m Tand ~ 41%of ls8Rewere present as labeled cysteine at 24 h. The addition of biotinylated antibody to aliquots of the serum incubates in the case of both labels resulted in a shift to higher molecular weight. The
Bioconjugafe Chem., Vol. 6,No. 1, 1995 Tc- 9 9 m
Virzi et al. Re- 188
injectate
Re- 188 injectate
Tc-99m
n
n
T = 1 HOURS
/\ f4
Irerum 89%
A
L i
98'1,
serum
T = 24 HOURS
93% I
0
" 0 20 30 40 5G 6 0 7 0 80
0
10
20 30 40 50 60 70 8 0
FRACTION NUMBER
2
10 2 0
30
40
50 5C 70 8 0 0
10 2 0 30 40 50 50 70 80
NUMBER Figure 3. Radiochromatographic profiles obtained by size exclusion HPLC analysis of 99mT~(left column) and ISSRelabeled (right column) metallothionein-streptavidin. Chromatograms are presented for the preparations themselves (injectates)and serum incubates after 1 and 24 h at 37 "C. In the case of both radionuclidesand both times, a chromatogram is also presented showing the change in profile following the addition of biotinylated antibody (B-I&) to the serum incubates. The recoveries are listed for each chromatogram. FRACTION
complexity of these radiochromatograms and the poor resolution do not permit a n accurate estimate of the extent of these shifts. In Vivo Mouse Studies. Samples of urine and serum obtained from mice administered 99mTc-and lsERe-labeled metallothionein-streptavidin were obtained during sacrifice a t 5 h post administration. The serum radiochromatograms (Figure 4) show a shift to higher molecular weight, especially in the case of 99mTc,from that of the injectates, although the profiles are generally similar. The recoveries are high indicating the absence of significant concentrations in serum of pertechnetate or perrhenate, both of which are retained by the column. A small peak centered a t fraction 75 is present in both radiochromatograms and most probably is due to labeled cysteine. The urine radiochromatograms show a prominent cysteine peak in both cases and several smaller, unidentified peaks centered in fractions 51 and 58 which may be identical to peaks observed previously in similar studies of 99mTc-labeledantibodies (17). The recovery in the 99mTccase is low which indicates the presence in this urine sample of about 40% pertechnetate. The biodistribution results a t 5 h post administration of 99mT~and lSsRe-labeledmetallothionein-streptavidin fusion proteins are listed in Table 1. There are no significant differences (Student's T test, p < 0.05) between these two labels in blood and all tissues except for heart, lung and spleen ( p < 0.05).
Figure 4. Radiochromatographic profiles for 99mTc (left column) and lS%e (right column) obtained by size exclusion HPLC analysis of the injectates, one urine and one serum sample obtained at 5 h post administrationof labeled metallothioneinstreptavidin fusion protein. "he recoveries are listed for each
chromatogram.
Table 1. Biodistribution Results in Normal Mice at 5 h Post Administrationa organ 99mT~ ls8Re p value liver 6.5 (0.7) 6.0 (0.9) 0.37, NS.* 2.8 (0.5) heart 1.7 (9.3) 0.006 5.9 (1.5) 6.7 (1.1) 0.38, N.S. kidneys 3.5 (0.6) lung 2.3 (0.2) 0.006 stomach 0.9 (0.2) 0.7 (0.1) 0.11, N.S. spleen 4.8 (0.7) 0.017 12 (5) muscle 0.5 (0.2) 0.3 (0.03) 0.07, N.S. blood 0.94, N.S. 14 ( 7 ) 13 (3) N=6 N=5
Percent Injected Dose Per Gram with 1 s.d. of the mean in Parentheses N.S. = not significant (p > 0.05, Student's T test). DISCUSSION The object of this investigation was to develop a suitable means of labeling a recombinant mouse metallothionein-streptavidin fusion protein with lE8Re, a radionuclide with attractive properties for radiotherapy of cancer. This anticipates a need for a radiolabeled streptavidin which may be used in a pretargeting approach to deliver sterilizing radiation doses to tumor. In this strategy, a biotinylated antitumor antibody would comprise the first injectate and would be followed a t the proper time with the radiolabeled streptavidin (I, 3 , 4 ) . The high affinity of streptavidin for biotin immobilized a t the tumor site via the antitumor antibody could offer high tumorhormal tissue ratios (5). Like any protein, a large number of labeling methods can be considered for radiolabeling streptavidin. However, labeling streptavidin with metallothionein as a bifunctional chelator has several clear advantages. Metallothionein has been used successfully to radiolabel antibodies with 99mTc(IO, 11) and thus is likely to be successful in radiolabeling with lE8Reas well considering the similarities in chemical properties of these two
Technical Notes
radionuclides. Furthermore, it is now possible to use recombinant DNA techniques to fuse metallothionein to proteins. Indeed, this has already been accomplished for streptavidin (161,and a recombinant metallothioneinstreptavidin fusion protein was used in this investigation. Although glucoheptonate was used successfully to transchelate reduced 99mTcto a metallothionein-conjugated antibody (111, in this investigation, citrate was first considered as a transchelator because it has been used successfully in other thiol-based labeling procedures using radiorhenium (21, 22). Whereas it was possible to radiolabel with lssRe unconjugated horse-kidney metallothionein in this manner, citrate failed to provide a label on metallothionein-streptavidin. This failure may be related to variations in properties toward radiolabeling among the different metallothioneins. In addition to the difference of horse kidney metallothionein vs streptavidin-fused murine metallothionein, we have observed greater instabilities of 99mTcwhen chelated to the fusion protein relative to a n antibody radiolabeled with 9 9 m T ~ via a rabbit metallothionein (unpublished observations). These differences may not be surprising considering the large variations in the amino acid sequences of the metallothioneins from species to species. Although each contains 20 cysteine residues, there can be as much as a 20% difference in the remaining residues. In particular, mouse metallothionein differs in amino acid composition from that of horse and rabbit metallothioneins by 10 and 2 residues, respectively (7). Furthermore, in the construct used in this investigation, the metallothionein was attached to the streptavidin core via a 10-amino acid tether (16)which may not be the optimal length to bring the metal binding sites on metallothionein into a sterically favorable position for chelation. Consequently, it may not be surprising that the coordination properties for metal binding may differ significantly among the metallothioneins. While it was not possible to radiolabel the fusion protein with ls8Revia citrate, we did achieve about a 20% labeling efficiency in 2 h at room temperature with glucoheptonate. In order to maximize the labeling efficiency, the effect of such variables as time, temperature, and protein concentrations will need to be investigated. That the metallothionein-streptavidin fusion protein was radiolabeled is evident by the quantitative shift to higher molecular weight which followed the addition of biotinylated antibody (Figures 1and 2) and the absence of a shift when the labeled metallothionein-streptavidin is saturated with biotin. The radiolabel is most probably on the metallothionein moiety since streptavidin itself cannot be radiolabeled with lssRe by this approach. The stability of the lssRe in fresh human serum a t 37 "C was compared to 99mTcand shown to be similar (Figure 3). Both labels were unstable toward transchelation to cysteine. From Figure 3, 22% of the 9 9 m Tand ~ 41% of lssRe were present as labeled cysteine a t 24 h. The shift of radioactivity to high molecular weight was also similar between labels both a t 1 and 24 h of incubation (Figure 3). The two labels also exhibited strong similarities when biodistributions were compared in normal mice (Table 1). Likewise, the radiochromatographic profiles obtained by size exclusion HPLC analysis of urine and serum samples obtained a t sacrifice also show similarities (Figure 4). Both labels show increased levels of higher molecular weight species in serum in comparison to the injectates, and the retention times of the major low molecular weight peaks in the urine profiles were the same for both labels although the peak intensities were occasionally different. In both cases, however, a major
Bioconjugafe Chem., Vol. 6,No. 1, 1995 143
fraction of the label in urine was present as the cysteine complex, probably generated elsewhere and excreting through the kidneys (19). In summary, it has been possible to radiolabel streptavidin through murine metallothionein moieties fused via recombinant DNA techniques. Relative to 9 9 m Tlabeled ~ to the same fusion protein, the ls8Re label behaved in vitro and in vivo in a similar and, occasionally, identical manner. ACKNOWLEDGMENT
The authors wish to thank Drs. F. F. Knapp, Jr., and A. P. Callahan of the Oak Ridge National Laboratory for providing the 18sW-188Reradionuclide generator and Dr. John Rodwell of Cytogen Corp. for providing the B72.3 antibody. This work was supported in part by DE-FGO293ER61656 from the U.S. Department of Energy and by CA59785 from the National Cancer Institute. LITERATURE CITED (1) Hnatowich, D. J., Virzi, F., Rusckowski, M. (1987)Investigations of avidin and biotin for imaging applications. J . Nucl. Med. 28, 1294-1302. (2) Kalofonos, H. P., Rusckowski, M., Siebecker, D. A., et al. (1990) Imaging of tumor in patients with indium-111-labeled biotin and streptavidin conjugated antibodies: preliminary communication. J . Nucl. Med. 31, 1791-1796. (3) Rmm, M. V., Fells, H. F., Perkins, A. C., et al. (1988) Iodine131 and indium-111 labeled avidin and streptavidin for pretargeted immunoscintigraphy with biotinylated anti-tumor monoclonal antibody. Nucl. Med. Commun. 9, 931-941. (4) Khawli, L. A., Alauddin, M. M., Miller, G. K., et al. (1993) Improved immunotargeting of tumors with biotinylated monoclonal antibodies and radiolabeled streptavidin. Antib. Immunoconj. Radionucl. 6, 13-27. (5) Sung, C., van Osdol, W. W., Saga, T., et al. (1994) Streptavidin distribution in tumors pretargeted with a biotinylated monoclonal antibody: theoretical and experimental pharmacokinetics. Cancer Res. 54, 2166-2175. (6) Paganelli, G., Magnani, P., Meares, C., et al. (1993)Antibody guided therapy of CEA positive tumors using biotinylated monoclonal antibodies, avidin and goY-DOTA-biotin:initial evaluation. J . Nucl. Med. 34, 94P. (7) Kagi, J. H. R., Kojima, Y. (1987)Chemistry and biochemistry of metallothionein. Experientia Suppl. 52, 25-61. (8) Nordberg, M., Kojima, Y. (1979) Metallothioinein and other low molecular weight metal-binding proteins. Experientia S ~ p p l34,41-116. . (9) Das, C . , Kulkarni, P. V., Constantinescu, A., et al. (1992) Recombinant antibody-metallothionein: design and evaluation for radioimmunoimaging. Proc. Natl. Acad. Sci. U S A . 89, 9749-9753. (10) Burchiel, S. W., Hadjian, R. A., Hladik, W. B., et al. (1989) Pharmacokinetic evaluation of technetium-99m-metallothionein coaugated mouse monoclonal antibody B72.3 in rhesus monkeys. J . Nucl. Med. 30, 1351-1357. (11) Brown, B. A., Drozynski, C. A. Dearborn, C. B., et al. (1988) Conjugation of metallothionein to a murine monoclonal antibody. Anal. Biochem. 172,22-28. (12) Palmiter, R. D., Norstedt, G., Gelinas, R. E., et al. (1983) Metallothionein-human GH fusion genes stimulate growth of mice. Science 222, 809-814. (13) Low, M. J., Hammer, R. E., Goodman, R. H., et al. (1985) Tissue-specific posttransitional processing of pre-somatostatin encoded by a metallothionein-somatostatin fusion gene in transgenic mice. Cell 41, 211-219. (14) Grifiths, G. L., Goldenberg, D. M., Knapp, F. F., Jr., et al. (1991) Direct radiolabeling of monoclonal antibodies with generator-produced rhenium-188 for radioimmunotherapy: labeling and animal biodistribution studies. Cancer Res. 51, 4594-4602.
Virzi et al.
144 Simnjugate Chem., Vol. 6,No. 1, 1995 (15) Sano, T., Cantor, C. R. (1990) Expression of a cloned streptavidin gene in Escherichia coli. Proc. Nut. Acad. Sci. U S A A87, . 14'i-146. (16) Sano, T., Glazer, A. N., Cantor, C. R. (1992) A streptavidin-metallothionein chimera that allows specific labeling of biological materials with many different heavy metal ions. Proc. Nut. h a d . Sci. U S A . 89, 1534-1538. (17) Mardirossian, G., Wu, C., Rusckowski, M., et al. (1992) The stability of 99Tcm directly labeled to an Fab' antibody via stannous ion and mercaptoethanol reduction. Nucl. Med. Commun. 13, 503-512. (18) Hnatowich, D. J.,Mardirossian, G., Rusckowski, M., et al. (1993) Pharmacokinetics of the F023C5 antibodv f r a m e n t iabelled with 99Tcm and ll11n: a comparison in patients- NUC. Med. Commun. 14, 52-63. (19) Hnatowich, D. J., Mardirossian, G., Rusckowski, M., et al. (1993) Directly and indirectly technetium-99m-labeled anti-
bodies - a comparison of in vitro and animal in vivo properties.
J.Nucl. Med. 34, 109-119. (20) Mardirossian, G., Bushe, H., Hnatowich, D. J. (1992) Pharmacokinetic modeling of Tc-99m labeled antibody in patients. J . Nucl. Med. 33, 958. (21) Najafi, A., Alauddin, M. M., Sosa, A., et al. (1992) The evaluation of 186Re-labeled antibodies using N2S4 chelate in vitro and in vivo using tumor-bearing nude mice. Nucl. Med. Biol. 19, 205-212. (22) Goldrosen, M. H., Biddle, W. C., Pancook, J., et al. (1990) Biodistribution, pharmacokinetics, and imaging studies with lS6Re-labeledNR-LU-10 whole antibody in LS174T colonic tumor-bearing mice. Cancer Res. 50, 7973-7978. BC940085R
