Nut!. Med.BioLVol. 18, No. 7, pp. 719-726, 1991 Inl. J. Radial.Appl.Insrrum.ParrB Printed in Great Britain. All rights reserved

0883-2897/91$3.00+ 0.00 Copyright 0 1991Pergamon Press plc

New Indium- 111 Labeled Biotin Derivatives for Improved Immunotargeting F. VIRZI,

B. FRITZ,

M. RUSCKOWSKI, and

M. GIONET,

H. MISRA

D. J. HNATOWICH*

Department of Nuclear Medicine, University of Massachusetts Medical Center, Worcester, MA 01655. U.S.A. (Received 12 December 1990) Investigations into the use of streptavidin-conjugated antibodies and labeled biotin to improve radioimmunotargeting have shown background levels drastically reduced over the conventional approach. Nevertheless, accumulation of “‘In-biotin in normal tissue as well as streptavidin-independent accumulation in tumor, was observed. In this work, the effect of altering the biotin molecule to reduce this nonspecific uptake without decreasing specific localization has been investigated. Three EDTA and DTPA derivatives of biotin have been synthesized and investigated along with a commercial biotin derivative (DTPA-B,). The labeled biotin chelates were administered i.p. to normal mice implanted with avidin beads in one thigh. A wide variation in biodistribution was seen among the biotin derivatives. The most favorable results were obtained with biotinyl-hydrazino-EDTA (EDTA-B,), which showed the lowest accumulation in normal tissues but equivalent uptake in the target with respect to the other compounds. Averaged over 8 tissues sampled, the target-to-nontarget ratio was 140 vs 9 for EDTA-B, vs DTPA-B, (N = 6) at 24 h post administration. Similar observations have been made in culture with two tumor cell lines: positive accumulation of both DTPA-B, and EDTA-B, was measured in tumor cells independent of streptavidin-antibody conjugate, however in the case of the latter derivative, this accumulation was 3-5 fold lower. These studies show that modification of the biotin species can alter accumulation in normal tissues as well as the antibody-streptavidin independent accumulation in tumor tissue.

clonal antibodies in which accumulations of radiolabel in normal tissue can be problematic. Initially (Hnatowich et af., 1987) we chose to attach “‘In to biotin using DTPA-ar,w-bis (biocytinamide) (DTPA-B2)3. While studies using “‘In-DTPA-B2 have demonstrated that background levels may be drastically reduced over conventional targeting (Kalofonos et al., 1990), some accumulation in normal tissue is observed. In an attempt to improve the in oiuo behavior of labeled biotin, we have synthesized three additional biotin-chelator molecules for “‘In-labeling and have examined their behavior in vitro and in vivo.

Introduction Avidin (streptavidin) and biotin have been used in a wide spectrum of bioanalytical applications as described in a recent review (Wilchek and Bayer, 1988). In particular, reports have appeared which suggest that avidin and biotin may be of value for drug (Rosenberg et al., 1987; Philpott et al., 1980; Hashimoto et al., 1984; Urdal and Hakromori, 1980) or radionuclide (Hnatowich et al., 1987; Forster er al., 1989; Khawli et al., 1988; Sinitsyn ef al., 1989; Paganelli et al., 1988; Pimm et al., 1988) targeting. This laboratory is investigating the delivery of radiolabeled biotin to sites which are pretargeted by a preliminary administration of a streptavidinmonoclonal antibody conjugate and we have seen encouraging results for abscess (Rusckowski and Hnatowich, 1988) and tumor (Kalofonos et at., 1990) imaging. An important advantage of this approach is the potential for rapid whole body clearance of the radiolabel in contrast to the situation usually found in the use of conventionally radiolabeled mono-

Experimental Section The structures of the biotin-chelators considered in this investigation appear in Fig. 1. DTPA-a,w-bis(biocytinamide) (DTPA-B2), was obtained commercially (Sigma Chemical Co., St Louis, MO.). Biotinyl-hydrazino-DTPA

*Author to whom all correspondence and reprints requests should be addressed.

(DTPA-B,)

To a stirred solution of biotin hydrazide (Sigma) (22.0 mg, 0.085 mmol) in 10 mL of DMF at 65°C was added 1OmL of CHCI, followed by a solution containing DTPA bicyclic anhydride (Sigma) in 20mL

719

720

F. VIRZIet al. and then refluxed briefly (l-2 min), cooled to room temperature, and the white solid was filtered, immediately washed with CH,CN, dried, redissolved in 1 mL of 0.2 M sodium bicarbonate pH 8.5, and isolated as above for DTPA-B, to yield EDTA-B, (38 mg, 37% yield) melting at 196-199°C (dec.). The stoichiometric ratio of EDTA-to-biotin was determined as described below to be 1.00 (+0.06):0.93 (f0.12). EDTA-u,w-bis(biocytinamide) m-PA

-

Ba

(EDTA-83

This synthesis was performed following the identical procedure for DTPA-B, described previously (Hnatowich et al., 1987) using a 5 : 1 molar excess of biocytin to EDTA anhydride which results in preferential production of the disubstituted species. Purification was accomplished by the methods described above for DTPA-B, and EDTA-B,.

LWPA-B,

Biotin quantitation

BDTA - 8,

Fig. 1. Structures of biotin derivatives considered in this investigation.

of DMF/CHClr (1: 1, v/v) (50 mg, 0.14 mmol). The precipitate was stirred for 1.5 h at 65”C, then 0.2 mL of water was added to hydrolyze any remaining anhydride and, after another 10 min, 300mL of CH,Cl, was added. The precipitate was filtered and washed immediately with CH,CN, air dried, redissolved in 1 mL of 0.2 M sodium bicarbonate pH 8.5, and passed through a 1.4 x 120cm gel filtration column of P-2 (Bio-Rad, Richmond, Calif.) with water as eluant. An aliquot of the first peak (by U.V. detection at 220 nm) was labeled with “‘In (DuPont/NEN, Boston, Mass.) in the acetate form, combined with excess avidin (Sigma), and analyzed by G-50 Sephadex chromatography. Greater than 95% of the radiolabel eluted in the void volume (and was therefore bound to avidin) to demonstrate the presence of the product. Lyophilization yielded DTPA-B, (6.0 mg, 11% yield) which melted at 21 l-213°C (dec.). The stoichiometric ratio of DTPAto-biotin in the product was determined as described below to be 0.88 (kO.07): 1.00 (kO.14). Biotinyl-hydrazino-EDTA

(EDTA-B,)

To a stirred solution of biotin hydrazide (50 mg, 0.194 mmol) in 20 mL of DMF/CHCl, (1: 1, v/v) at 65°C was added 20mL of DMF/CHCl, (1: 1, v/v) containing EDTA bicyclic anhydride (65.0 mg, 0.254 mmol) prepared as previously described (Sosnovsky et al., 1985). The mixture was stirred for 30min, hydrolyzed by the addition of 0.5 mL of water and stirred for an additional 10min. The reaction mixture was poured into 300 mL of CHr Cl,

The spectrophotometric assay developed by Green (1965) based on the displacement of HABA (Sigma) from avidin by biotin was used in this work. A solution of HABA-saturated avidin (1 mg/mL avidin, 16.2 pg/mL HABA, in 0.1 M sodium acetate buffer, pH 6.0) was prepared and the absorption at 500 nm determined. The decrease in absorption was then determined following successive additions of 50 ,uL of a solution containing 27.8 pg/mL of d-biotin (Sigma) in 0.1 M sodium acetate buffer, pH 6.0. The decrease in absorbance was also measured following additions of biotin solutions of different concentrations (55.6 and 83.4 pg/mL) and for acetate buffer as a control. The biotin content of both EDTA-B, and DTPA-B, was then determined by measuring the rate of decrease in absorbance following successive 50 FL additions of solutions of these species at a fixed concentration (100 pg/mL) in the same buffer. Chelator quantitation

The assay procedure of Van der Walt et al. (1989) for quantitation of polyaminopolycarboxylic acid chelators was used in the present study. Solutions of EDTA-B, and DTPA-B, at 1.0 mg/mL in 0.1 M sodium acetate, pH 6.0, were added in 20 PL aliquots to a solution containing 18.4 p L of 2.0 mg/mL copper sulfate pentahydrate (Fisher Scientific Co., Fairlawn, N.J.) in 0.01 M HCl and 3OpL of 4.6 mg/mL eriochrome cyanine R (Sigma), and were then diluted to 1 mL with the acetate buffer. After overnight equilibration, the absorbance was measured at 567 nm, and the concentration of chelator was determined by comparison with calibration curves obtained using EDTA and DTPA. Labeling with “‘In and quality assurance

Solutions of biotin chelates at 1 mg/mL in 0.5 M sodium acetate, pH 6.0, were labeled with “‘In as the acetate at a specific activity of up to 125 mCi/mg. No attempt was made to maximize the specific activity. The avidin binding ability of the “‘In-labeled biotin

In- 1 11-biotin immunotargeting chelates was measured after adding a IO-fold molar excess of avidin to the labeled biotins and analyzing the samples by Sephadex G-50 to determine the percent radioactivity in the protein fractions and therefore bound to avidin. Binding to serum proteins

The four biotin chelates were labeled at SO-75 pCi/pg with “‘In as described above. Samples containing 0.1 ,~g of radiolabeled biotin chelate in 10 pL saline were combined with 0.1 mL of fresh sterile-filtered human serum, and the tubes were incubated at 37°C for 1 h. Aliquots of each sample were then analyzed by Sephadex G-50 chromatography using saline as the eluant to determine the percentage bound to serum proteins. Serum stability

Each of the biotin chelators was labeled at 75 pCi/pg as above and then 0.2 pg of the radiolabeled biotins in 100 PL saline were combined with 100 PL of a suspension of avidin-agarose beads (Sigma). The beads were incubated at room temperature for 15 min and then washed 3 times with sterile saline. Sterilefiltered human serum (0.2 mL) was added to each tube, and the suspensions were incubated for 24 h at 37°C with vigorous shaking. The beads were then recovered, washed 3 times with saline and the percentage of added activity bound to the beads was measured. Electrophoresis

Solutions of the biotin chelators and free-DTPA were radiolabeled with “‘In in 0.1 M sodium acetate buffer, pH 6.0 at a specific activity of 1 gCi/gg and, after 1 h, were applied to strips of Whatman No. 1 paper. The samples were run for 2 h at 200 V in 0. I M HEPES. pH 7.0 and the strips were cut into 5 mm segments and counted. Extraction

into butanol

To 1 pg of each of the “‘In-labeled biotin chelates in 4 p L of 0.1 M sodium acetate buffer, pH 6.0 was added 1 mL of distilled H,O and 1 mL of n-butanol or n-octanol. After vigorous shaking, the organic and aqueous phases were allowed to separate for 2 h, and 0.5 mL aliquots of both layers were counted. Competitive

binding L’Sd-biotin

The biotin compounds were labeled with “‘In as above at 100 pCi/pg. 800 pmol of each derivative was diluted with solutions of d-biotin in 0.05 M sodium acetate buffer pH 6.0 such that the stoichiometric ratios of biotin to derivatized biotin ranged from 0.0 to 10.0. To these mixtures was added 20 pmol of avidin (0.1 mg/mL) in saline, and the solutions were quickly vortexed. After incubation at 20°C for 24 h, the percent activity bound to avidin was determined by paper chromatography using Whatman No. 1 paper in 0.05 M sodium acetate, pH 6.0. In this

721

system, labeled avidin remains at or near the origin while the labeled biotins migrate to the solvent front. Cell binding studies LS-174T and H-MESO cells were a gift from Dr Thomas W. Griffin of this institution and were grown under conventional tissue culture techniques. The cells were harvested with 0.05% trypsin containing 0.50 mM EDTA and then washed twice with RPM1 1640 media (Gibco, Grand Island, N.Y.). The cells were washed and resuspended in biotin free Dulbecco’s modified essential media, high glucose (DMEM) (Gibco) to approx. lo6 cells per mL. All reagents were maintained at 37°C. The four biotin chelates were labeled with “‘In at SOpCi/pg and then diluted with DMEM to a final concentration of lOpg/mL. Each of the four labeled chelates was added to an aliquot of the LS- 174T or H-MESO cells at 250 ng of chelator per IO6 cells. The tubes were gently vortexed and incubated at 4 or 37°C in a water bath with shaking for 1.5 h. After incubation, an aliquot of cells was added to 500 PL of media in an Eppendorf tube precoated with 2% HSA in saline at room temperature for 2 h to minimize nonspecific binding. The cells were pelleted, washed twice, then counted in a y well counter and the percent of bound activity calculated. To investigate whether the incorporation of the biotins was receptor mediated, the cell suspensions were prepared as described but with the prior addition of a lOO-fold excess of d-biotin (250 pg). To determine if the incorporation was ATPase dependent, the cell suspension was pre-equilibrated with lo-” to 10-4M ouabain (Skou, 1960) (Sigma) for 30 min at 37°C before the addition of the radiolabeled biotin chelate. The samples were then treated as described previously. Animal studies

A suspension of avidin-agarose beads (0.2 mL) was injected into the left thigh of four sets of six normal male CD-l mice (Charles River Supply, Wilmington, Mass.). After 24 h each animal received an i.p. injection of 1 pg of one of the “‘In-labeled biotin chelates in 0.1 mL saline. Two pairs of mice receiving either “‘In-DTPA-B2 or “‘In-EDTA-B, were imaged with a gamma camera 2 h after injection of radioactivity and 1 day after injection of the avidin beads. Twenty-four hours after injection of the radioactivity, all animals were sacrificed by spinal dislocation, dissected, and the tissues counted in a ‘; counter. The streptavidin-independent uptake of “‘InDTPA-B, by H-MESO tumors in nude mice (Taconic Labs, Germantown, N.Y.) was also examined. Two sets of 6-week-old Swiss male nude mice were injected i.p. with 1 mL of RPM1 media containing approx. 1.5 x 10’ H-MESO cells per mL. During the course of tumor development, the ascite was drained twice to maximize growth. At 29 days post inoculation the

F. VIRZIet al.

722

Table 1. Properties of the radiolabeled biotin derivatives Compound

Percent binding to serum proteins*

Percent stability in serumt

Migration (mm)S

DTPA-B, DTPA-B, EDTA-B, EDTA-B, DTPA

29 (26-34) 46 (4448) 9 (8-9) 20 (la-33) ND

94+ 1 94+ 1 95 * 2 99* 1 ND

45 15 60 35 105

*Binding of “‘In-labeled biotin chelates to serum proteins at I h (N = 3) range in parentheses. TActivity in serum as “‘In-labeled biotin chelates bound to avidin after 24 h at 37°C (N = 6, *SD). SElectrophoretic migration of “‘In-labeled biotin chelates. Distance of peak from origin towards anode. ND = not done.

D-BIOllN/BIOllN CHEIATEMOLARRATIO

Fig. 2. Competitive binding of “‘In-labeled biotin derivatives with d-biotin (semi log scale). In each case binding in the absence of d-biotin adjusted to 100%. Error bars show

mice all had palpable tumors. At this time one set of mice received an i.p. injection of 200ng of “‘InDTPA-B, labeled at a specific activity of 125 pCci/pg in 100 PL of saline. A control set of mice received 64 ng of i”In-DTPA labeled at a specific activity of 375pCi/pg. Three hours after injection of radioactivity, the mice were imaged on an Elscint y camera. The animals were then sacrificed by spinal dislocation and a biodistribution performed which showed evidence of diffuse tumor proliferation within the abdomen.

SD of repeat measurements (N = 5).

of solvent extraction measurements in which all four labeled biotin derivatives showed less than 2% extraction into butanol or octanol from aqueous phase.) The results of a competition between each biotin derivative and d-biotin for avidin is shown in Fig. 2. Equivalent affinities for avidin would result in 50% activity bound to avidin at a 1: 1 molar ratio. Only EDTA-B, approaches this value. All four biotin derivatives, therefore, show some impaired ability to bind to avidin with the extent of impairment varying in the order EDTA-B, > DTPA-B, > DTPA-B, > EDTA-B,. Figure 3 presents the biodistribution results obtained at 24 h post administration of the labeled biotin derivatives in normal mice with avidin-agarose beads implanted in the left thigh. This bead model was used by us previously (Hnatowich et al., 1987) and was selected for this investigation to provide a highly reproducible target which may be superior in this with regard to tumor models. It is clear from the figure that both the DTPA-B2 and the EDTA-B,

Results Each of the three new biotin derivatives could be radiolabeled with “‘In to a specific activity of at least 125 pCi/pg and, when added to an excess of avidin in solution, greater than 90% of the “‘In was bound to avidin. The absence of contaminating radiolabeled species was also apparent from the electrophoresis results in which 90% of the radiolabel migrated to the anode as a single peak for each species (Table 1). All four biotins showed minimal (~6%) dissociation of the radiolabel after 24 h of serum incubation at 37°C and between 9 and 46% binding to serum proteins. (Not shown are the results

0.08

3

0.8

T

HEART

LUNG

ST0

CH SPLEEN

fllcg

BLOOD

KIDNEY

LEFT THIGH

Fig. 3. Biodistribution of “‘In-labeled biotin chelates in normal mice with avidin-agarose beads implanted in the left thigh. Error bars show SD (N = 6). Note change of scale.

Fig. 4. Images obtained 2 h post i.p. administration of either “‘In-DTPA-B, (left) or “‘In-EDTA-B, (right) in mice previously implanted in their left thighs with avidin-agarose beads. Both images were taken at the identical same upper and lower camera threshold settings.

5 Fig. 5. Images obtained 2 h post i.p. injection of “‘In-DTPA-B, (top) or “‘In-DTPA (bottom) in nude mice implanted with H-MESO tumor cells in the peritoneum. All mice were imaged in duplicate simultaneously in the same field of view.

723

In-l 11-biotin immunotargeting

725

Table 2. Ratio of “‘In-activity in target (left thigh) to normal tissues in mice implanted with avidin-aaarose beads (N . = 6),

Organ

DTPA-B,

EDTA-B,

Liver Heart Kidney

5 14 0.5

12 340

Lung

II

1:;

8 6

121 91

Stomach Spleen Left thigh Right thigh Blood

I

1

10 15

148 213

0

20

40

00

M)

100

120

TIME(minuba)

show the highest accumulation in normal tissues and that EDTA-BI shows the lowest accumulation in these same tissues. In addition to this favorable property, EDTA-B, shows accumulation in the target (left thigh) which is comparable to that of the other biotin derivatives. The ratio of “‘In activity in the left thigh to normal tissues for labeled EDTA-B, vs DTPA-B, is presented in Table 2 and illustrates the superior properties of the former derivative. As a control, three mice with underivatized agarose beads implanted in the left thigh showed negligible accumulation of activity in this region 24 h after i.p. administration of “‘In-EDTA-B, (data not shown). The improved properties of “‘In-EDTA-B, over “‘In-DTPA-B, are also apparent from the images presented in Fig. 4. These images were obtained at 2 h post administration of the labeled biotin (and at 1 day post administration of the avidin-agarose beads). Two animals are shown for each injectate. Both sets of images were acquired at the same upper and lower threshold settings. In each case, a large fraction of the whole body activity has cleared through the kidney even at this early time. However, background activity levels in liver and other tissues are clearly lower in the case of the animals receiving labeled EDTA-B, vs those receiving DTPA-B, . Figure 5 illustrates the streptavidin-independent accumulation of “‘In-labeled DTPA-B, in tumor tissues. Two mice implanted with H-MESO tumor cells in the peritoneum received “‘In-DTPA-B, and two identical mice received “‘In-DTPA. At 2 h post administration, bladder activity is observed in all four images, however, accumulation in diffuse tumor is apparent only in the case of animals receiving the biotin derivative. The nature of this apparent nonspecific accumulation of biotin derivatives in tumor cells has been investigated. Table 3 presents the results of incubating the labeled biotins with two tumor cell lines. Table 3. Percent binding of “‘In-labeled biotin chelates to LS-174T and H-MESO cells (N = 6) in culture Comvound

LS-174T

DTPA-B, DTPA-B, EDTA-B, EDTA-B,

1.14~0.10 1.05 +0.10 0.70 f 0.04 0.21 f 0.03

H-MESO 0.43 0.51 0.35 0.12

* f f *

0.03 0.01 0.04 0.01

,I-‘-‘-’

06. 0

/

.

100

.

ZOO

I Jo0

400

500

a00

700

NANOGRAMSOF BIOTINCHELATEADDED

Fig. 6. (A) Percent incorporation of “‘In-DTPA-B,

in H-MESO cells vs time at 37°C and 4°C with and without ouabain; (B) accumulation in H-MESO cells of “‘InDTPA-B, vs chelator concentration.

Although under the conditions of the study, only a small percentage of added activity accumulated, the differences among the derivatives are significant (Student r-test, P < 0.001) except for DTPA-B, vs DTPA-B, in the LS- I74T cells (P < 0.1) and DTPAB, vs EDTA-B, in the H-MESO cells (P < 0.01). The order of accumulation, namely DTPA-B2 = DTPAB, > EDTA-B2 > EDTA-B, , appears to follow the same trend with both tumor cell lines. The nature of “‘In-DTPA-B, accumulation in the H-MESO cell line was investigated. As shown in Fig. 6, the incorporation is insensitive to high concentrations of ouabain and is inhibited at 4°C relative to 37°C [Fig. 6(A)]. In addition, the incorporation is not saturable or inhibited by excess biotin [Fig. 6(B)]. Therefore, the accumulation of “‘In-DTPA-B, in the H-MESO cell type appears to result from passive diffusion alone.

Discussion One of the incentives for pursuing this research arose from a collaborative investigation in cancer patients in which “‘In-DTPA-B, was introduced after the administration of streptavidin-conjugated HMFGl antibody (Kalofonos et al., 1990). Although the accumulation of activity in normal tissues was drastically reduced over that observed following administration of radiolabeled antibodies, these levels were potentially problematic, especially in the kidneys. It now seems likely that replacement of the DTPA-BI derivative with EDTA-B, may improve

F. VIRZIet al.

126

the image quality by reducing background activity without reducing levels in tumor. A second significant observation from this clinical trial was the accumulation of “‘In-DTPA-B, in tumor even without the prior administration of streptavidin-conjugated antibody (Kalofonos et al., 1990). Although the streptavidin-independent accumulation may have been due simply to nonspecific diffusion, “‘In-DTPA-B, may show accumulation in certain tumors. This possibility has been investigated in this work by incubating the labeled biotins with two tumor cell lines grown in culture. All four biotin derivatives were found to accumulate in both tumor types, although to varying degrees. In the case of DTPA-B, and the H-MESO cell line, the accumulation is most likely due to passive diffusion. The fact that EDTA-B, shows the lowest tumor accumulation in oitro compared to the other three derivatives also suggests that this derivative may be the most sensible choice in future studies with streptavidin-conjugated antibodies. In conclusion, it is apparent that relatively minor changes in the molecular structure of biotin chelators can have a significant effect on in vitro and in vivo properties. The accumulation of “‘In-EDTA-B, in normal tissues in mice was found to be the lowest of the derivatives tested while, in the bead model, this derivative did not show reduced accumulations in the target. Finally, in two tumor tissues, “‘In-EDTA-B, showed the lowest streptavidin-independent accumulation. The basis for these favorable properties was not established in this work. Acknowledgements-The

authors wish to thank MS Stephenie Roy for assistance with the imaging studies. This research was supported in part by NIH Grants CA 33029, RFP NCI-CM-97570 and AI 25570.

References Forster T., Soskuthy G. and Bursics L. (1989) Tc-99mchelating biotin derivatives for immunoscintigraphic applications. Eur. J. Nucl. Med. 15, 514 (Abstract). Green N. M. (1965) A spectrophotometric assay for avidin and biotin based on binding of dyes by avidin. Biochem. J. 94, 23C-24C.

Hashimoto N., Takatsu K., Masuho Y. et al. (1984) Selective elimination of a B cell subset having acceptor site(s)

for T cell-replacing factor (TRF) with biotinylated antibody to the acceptor site(s) and avidin-ricin A-chain conjugate. J. Immunol. 132, 129-135. Hnatowich D. J., Virzi F. and Rusckowski M. (1987) Investigations of avidin and biotin for imaging appli: cations. J. Nucl. Med. 28. 1294-1302. Kalofonos H. P., Rusckowski M., Siebecker D. A. et al. (1990) Imaging of tumor in patients with “‘In labeled biotin and streptavidin-conjugated antibodies: preliminary communication. J. Nucl. Med. 31, 1791-1796. Khawli L. A., Kinsey B. M., Adelstein S. J. and Kassis A. I. (1988) The facile preparation of a radiolabeled biotin derivative. J. NucI.~ Med. 29, 923 (Abstract). Paaanelli G.. Riva P.. Deleide G. et al. (1988). In viuo l&eling of biotinylated monoclonal antibodies by radioactivite avidin: a strategy to increase tumor radiolocalization. Int. J. Canc&2, (Suppl.) 121-125. Philnott G. W.. Kulczvcki A.. Grass E. H. and Parker C. W. (1980) Selectivd binding and cytotoxicity of rat basophilic leukemia cells (RBL-1) with immunoglobulin E-biotin and avidin-glucose oxidase conjugates. J. Immunol. 125, 1201-1209. Pimm M. V., Fells H. F., Perkins A. C. and Baldwin R. W. (1988) Iodine-131 and indium-1 11 labeled avidin and streptavidin for pre-targeted immunoscintigraphy with biotinylated anti-tumor monoclonal antibody. Nucl. Med. Commun. 9, 931-941. Rosenberg M. B., Breakefield X. 0. and Hawrot E. (1987) Targeting of liposomes to cells bearing nerve growth factor. J. Neurochem. 48, 865-875. Rusckowski M. and Hnatowich D. J. (1988) Streptavidin conjugated monoclonal antibodies for imaging abscess with indium-111 labeled biotin. J. Nucl. Med. 28, 889 (Abstract). Sinitsyn V. V., Mamontova A. G., Chekneva Y. Y. ef al. (1989) Rapid clearance of biotinylated IgG after infusion of avidin. J. Nucl. Med. 30, 66-69. Skou J. C. (1960) Further investigations on a Mg++ + Na+activated adenosine triphosphatase, possibly related to the active, linked transport of Na+ and K+ across the nerve membrane. Biochim. Biophys. Acta 42, 6-23. Sosnovsky G., Li S. W. and Rao N. U. M. (1985) Spin labeled chelating agents and their gadolinium complexes as contrast enhancing agents for NMR imaging. Z. Naturforsch. 4Ob, 1558-1562. Urdal D. L. and Hakromori S. (1980) Tumor-associated ganglio-N-triosylceramide: target for antibody dependent, avidin-mediated drug killing of tumor cells. J. Biol. Chem. 255, 10509-10516. Van der Walt T. N., Coetzee P. P. and Fourie P. J. (1989) Indirect spectrophotometric determination of BIDA, DISIDA, DTPA and MDP in labeled compounds. Appl. Radiat. Isot. 40, 525-529.

Wilchek M. and Bayer E. A. (1988) The avidin-biotin complex in bioanalytical applications. Analyt. Biochem. 171, l-32.