Failure of radiolabeled monoclonal antibodies (MAbs) in the treatment of solid tumors, for the most part, is a result of undesirable pharmacokinetics that lead to significant radiation exposure of normal tissues and an inadequate delivery of radiation doses to tumors. Using genetic engineering, antitumor MAbs can be optimized for desirable clinical applications. In the present study, we report the generation of a tetravalent single-chain Fv {[sc(Fv)2]2}of the murine MAb CC49 that recognizes the tumor-associated glycoprotein, TAG-72. [Sc(Fv)2]2 was expressed as a secreted soluble protein in Pichia pastoris under the regulation of alcohol oxidase 1 promoter. The in vitro binding properties of the tetravalent construct were analyzed by solid-phase RIA and surface plasmon resonance studies using BIAcore. The binding affinity constant(KA) for the[sc(Fv)2]2 and CC49 IgG were similar, i.e., 1.02 × 108m−1 and 1.14 × 108m−1, respectively, and were 4-fold higher than its divalent scFv [sc(Fv)2; 2.75 × 107m−1]. At 6 h postadministration, the percentage of injected dose accumulated/g of LS-174T colon carcinoma xenografts was 21.3 ± 1.3,9.8 ± 1.3, and 17.3 ± 1.1 for radioiodinated [sc(Fv)2]2,sc(Fv)2, and IgG, respectively. Pharmacokinetic analysis of blood clearance studies showed the elimination half-life for[sc(Fv)2]2, sc(Fv)2, and IgG as 170, 80, and 330 min, respectively. The gain in avidity resulting from multivalency along with an improved biological half-life makes the tetravalent construct an important reagent for cancer therapy and diagnosis in MAb-based radiopharmaceuticals.

MAbs3used to be referred to as magic bullets because of their ability to target malignant cells for delivery of cytocidal agents like radionuclides, enzymes, genes, drugs, and toxins (1, 2). Currently, some MAbs are in clinical trials for both diagnostic and therapeutic purposes (3, 4, 5, 6). The major limitation of using intact MAbs for cancer radioimmunotherapy is because of the long persistence in circulation, thereby exhibiting side effects that are significantly deleterious to the normal tissues with only limited quantities delivered to tumors (7, 8). Furthermore, intact MAbs show poor diffusion from the vasculature into and through the tumor (9).

Immunoglobulins have been engineered to retain only the domains that are required for antigen binding and/or effector functions and have also been rebuilt into multivalent, high-affinity reagents (10, 11). In scFvs, the variable regions of heavy and light chains are joined covalently by either a polypeptide linker or a disulfide bond (12, 13). As monomers(Mr ∼30,000), these reagents are ideal for diagnostic applications because of their excellent tumor penetration, high RI (ratio of the %ID/g in the tumor:%ID/g in normal tissue), and low backgrounds (14, 15, 16, 17, 18). However, the absolute amounts of scFv uptake by the tumors remain low, mainly because of their monovalent nature and fast elimination (14, 15, 17, 18, 19, 20). The lower tumor:normal tissue ratio makes monovalent scFvs inefficient for radioimmunotherapeutic applications (8, 18, 21).

Many of the therapeutically important tumor-associated antigens are either glycolipids or glycoproteins with highly repetitive structures(22). Antibodies and antibody fragments with multiple valencies therefore represent an enormous gain in the functional affinity attributable to multiple interactions within a single antigen-antibody complex (10). Indeed, various divalent scFvs revealed improved antigen affinity in vitro when compared with the monovalent forms. Also, in animal models, divalent scFvs exhibited a significant improvement in tumor targeting over monovalent species like scFv and Fab because of their higher avidity and slower clearance properties rendered by their larger size(17, 18, 23, 24, 25). To improve further the in vitro and, ultimately, the in vivo performance of scFvs, the valency of scFv has been increased by designing trivalent and tetravalent scFvs (26, 27, 28, 29, 30, 31, 32, 33, 34, 35). However, the utility of scFv multimers in vivo for tumor targeting has not been adequately investigated. There are only a few studies where biodistribution studies were performed with trivalent antigen-binding constructs F(ab′)3 generated by chemical linkages (36, 37, 38).

In the present study, we report for the first time genetic engineering and in vivo evaluation of a tetravalent scFv construct of MAb CC49. The tetravalent scFv was formed by a noncovalent association of the covalent dimer sc(Fv)2.[Sc(Fv)2]2 with four potentially active antigen-binding sites showed improved in vitro binding properties as compared with sc(Fv)2 and CC49 IgG.[sc(Fv)2]2 exhibited>2-fold increase in the absolute tumor uptake (from 4 h after injection onwards). The larger molecular size of[sc(Fv)2]2, which exceeds the renal threshold for the first pass elimination, translated into an improved biological half-life of[sc(Fv)2]2. We believe that the tetravalent scFv, with higher avidity and prolonged pharmacokinetics in blood, meets the prerequisites of an optimum tumor-targeting reagent in radionuclide-mediated therapy and diagnosis.

Vector Construction.

The divalent CC49 scFv gene(VL-linker-VH-linker-VL-linker-VH)was constructed using the linker designated as 205C (39)and was cloned into the Pichia pastoris expression vector,pPICZαA (Invitrogen, Carlsbad, CA) as described earlier (25, 40). Briefly, competent P. pastoris KM71 cells(his4arg4aox1Δ::ARG4) were transformed with 1–3 μg of plasmid DNA linearized with SacI and selected on yeast extract peptone dextrose plates with 100 μg/ml Zeocin. The clones were screened for the secreted recombinant protein by solid-phase competition ELISA using biotinylated CC49 IgG.

Protein Expression and Purification.

Yeasts were grown at 30°C (A600 nm of 2–6)in buffered glycerol-complex medium containing 100 μg/ml zeocin. To induce expression, the cells were shifted to the buffered methanol-complex medium. Methanol was added to a final concentration of 0.5% every 24 h. At day 4, the culture was centrifuged, and the supernatant containing the scFv was dialyzed against 50 mm sodium phosphate (pH 7.2), 300 mm NaCl. Purification of the monovalent and divalent forms of scFv was performed by immobilized metal affinity chromatography using the Ni-Nitrilotriacetic acid Superflow (Qiagen Inc., Valencia, CA) as the chelating resin. Bound fragments were eluted in dialysis buffer containing 250 mm imidazole. The protein was analyzed by SDS-PAGE and by solid-phase competition ELISA. Size exclusion chromatography was used to separate the tetravalent and divalent scFvs from aggregated and breakdown products. For this, a Superdex-200 column (Pharmacia, Piscataway, NJ,1.6 × 60 cm) was equilibrated and run at 1 ml/min in PBS (pH 7.4). Protein concentrations were determined by the method of Lowry et al.(41).

SDS-PAGE.

SDS-PAGE was performed according to the method of Laemmli(42); the proteins were evaluated with or without reduction by βmercaptoethanol. The gels were stained with Coomassie Blue R-250. For radiolabeled protein, gels were exposed to a phosphor screen and analyzed using the ImageQuant software of the PhosphorImager (Molecular Dynamics, Sunnyvale, CA).

Labeling of CC49 scFv Forms.

The scFv forms were labeled with Na125/131I using 1,3,4,6-tetrachloro-3α,6α-diphenylglycoluril (Iodo-Gen;Pierce Chemical, Rockford, IL) as the oxidant (43). Briefly, 20–100 μg of protein in 0.1 m sodium phosphate buffer (pH 7.2) was transferred into a 12 × 75 mm glass tube coated with 20 μg of Iodo-Gen. The protein was incubated for 3 min with 0.5–1 mCi of Na125/131I (NEN, Boston,MA) followed by gel filtration. The specific activity of 125I- and/or 131I-labeled scFv molecules was about 3–9 mCi/mg.

HPLC Analyses.

Gel filtration on HPLC was used to analyze the radiolabeled scFvs. Samples were injected onto TSK G2000SW and TSK G3000SW (Toso Haas, Tokyo, Japan) connected in series with 67 mm sodium phosphate buffer (pH 6.8), 0.1 m KCl as the mobile phase. The columns were calibrated using the Gel Filtration Calibration Kit(Bio-Rad, Hercules, CA). The elution was monitored by an in-line UV detector at 280 nm, and the radioactivity was determined in a Packard Minaxi Auto-Gamma 5000 gamma counter (Meriden, CT).

Binding Analyses.

The immunoreactivity was assessed by a solid phase ELISA using BSM(Sigma Chemical Co., St. Louis, MO) as the antigen (18). Test samples were incubated for 2 h at room temperature in 3-fold serial dilutions with 6 ng of biotinylated CC49 IgG followed by incubation with alkaline phosphatase-conjugated streptavidin (Jackson ImmunoResearch Lab, West Grove, PA) for 1.5 h at room temperature. p-Nitrophenyl phosphate was used as the substrate, and absorbance was read at 410 nm using a Dynatech MR 5000 automatic 96-well microtiter reader (Chantilly, VA).

The quality control test of radiolabeled CC49 scFv forms was performed using a RIA where BSM or BSA (positive and negative controls,respectively) was attached to a solid-phase matrix (Reacti-Gel HW-65F;Pierce Chemical; Ref. 18). Binding was allowed to proceed for 1 h at room temperature. The unbound radiolabeled protein was removed by repeated washing with PBS containing 1% BSA and 0.1% Tween 20. The pellet was counted in a gamma scintillation counter, and the total percentage bound was calculated.

The affinities of[(scFv)2]2,sc(Fv)2, and CC49 IgG for BSM were determined by SPR using the upgraded version of BIAcore 1000 (Pharmacia Biosensor,Uppsala, Sweden). Approximately 400 resonance units of BSM or BSA were coupled to a CM5 sensor chip as described earlier (18, 44). Binding interaction was performed in HBS buffer [10 mm HEPES (pH 7.4) containing 150 mm NaCl, 3 mm EDTA, 0.005% surfactant, Nonidet P20] at a flow rate of 65 μl/min at 25°C. Analyses of scFvs were done in the concentration range of 50–500 nm, run in duplicate for each sample. The surface was regenerated with 1 m3-[cyclohexylamino]-1-propanesulfonic acid at a flow rate of 5μl/min with no loss of activity. IgG used for binding analysis was purified on Protein G Sepharose Fast Flow resin (Pharmacia Biotech.,Uppsala, Sweden) and dialyzed into HBS buffer. The kinetic rate constants (kon and koff), as well as equilibrium association constant (KA; kon/koff)and equilibrium dissociation constant(KD;1/KA), were determined using BIAevaluation 3.0.2 software (Pharmacia Biosensor) where the experimental design correlated with the Langmuir 1:1 interaction model(45).

Biodistribution and Pharmacokinetics Studies.

Female athymic mice (nu/nu; 4–6 weeks old) were used for the in vivo studies (Charles River, Wilmington, MA). LS-174T cells, a human colon carcinoma cell line (46), were implanted s.c. (4 × 106), and the mice were used 8 days (tumor volume, ∼200–300 mm3) after the injection of cells. Dual-label biodistribution studies were performed after a simultaneous i.v. injection via the tail vein of 125I-sc(Fv)2 (5 μCi) and 131I-[sc(Fv)2]2(2.5 μCi) or 125I-[sc(Fv)2]2(5 μCi) and 131I-CC49 IgG (2.5 μCi). At designated times, groups (n = 6) were sacrificed, and the tumor, blood, and major organs were removed,weighed, and counted in a gamma scintillation counter to determine the%ID/g for each labeled protein. For the whole body retention studies,mice bearing the LS-174T xenografts (three/group) were injected via the tail vein with 1.5 μCi of radiolabeled scFvs. Each scFv was evaluated separately. The whole body radioactivity was determined at various times after injection in a custom-built NaI crystal.

The blood clearance studies were done as described previously(18, 25). Blood samples were obtained from the tail vein at various times after the injection of 10 μCi of the individual radioiodinated scFvs. The half-lives were calculated using a numerical module of the SAAM II computer program (SAAM Institute, University of Washington, WA) for kinetic analysis. The data were fitted into a biexponential equation with a bolus injection as an experimental model. The clearance rates were compared using a two-sample Student’s t test for differences between means.

Stability Studies.

The stability of multivalent scFvs was assessed both in vitro and in vivo. For the in vitrostability study, the radioiodinated[sc(Fv)2]2 and sc(Fv)2 (0.05–1 mg/ml) were mixed with 1% BSA and incubated at 37°C for varying time intervals. The proteins were analyzed by HPLC. In vivo stability studies were performed using the tumor-bearing mice that received injections with 10 μCi of radioiodinated (125I and 131I) scFvs. Blood was collected at various time points as in biodistribution studies, and plasma was separated. The plasma samples (approximately 1 × 106 cpm) were run on HPLC, and the radioactivity associated with the high molecular weight proteins, scFvs, and dissociated forms was determined.

Construction, Expression, and Purification of Tetravalent scFv.

A noncovalent tetrameric CC49 scFv was expressed as a secreted protein in the methylotrophic yeast, P. pastoris. Earlier, we constructed divalent scFv of the MAb CC49 using the 205C helical linker(Fig. 1,A; Ref. 25, 40). The divalent form of scFv was found to noncovalently associate into [sc(Fv)2]2(Fig. 1,B). Purification of scFvs was expedited using the hexahistidine tag attached to the COOH-terminal of the construct (Fig. 1,A). The typical yields for covalent dimer were 15–20 mg/liter with 20–30% of the scFv population forming tetramers and/or higher aggregates. The percentage yield of tetravalent scFv and aggregates varied between batches with the typical yields of[sc(Fv)2]2 as 2.0–3.5 mg/liter. The purity and integrity of scFv preparations were analyzed by SDS-PAGE. As shown in Fig. 2,A,sc(Fv)2 appeared as a single band of Mr ∼58,000 under both nonreducing (Lane 1) and reducing (Lane 3)conditions. [sc(Fv)2]2also migrated as a single band of Mr∼58,000, indicating that the two polypeptide chains of tetramer were noncovalently linked (Fig. 2,A, Lanes 2 and 4). Both sc(Fv)2 and[sc(Fv)2]2 were ≥95%pure when observed by Coomassie Blue staining (Fig. 2 A).

Radiolabeling and Quality Control of scFvs.

The tetravalent scFv was radiolabeled with Na125I or Na131I, and protein stability was demonstrated by analyzing the radioactivity on SDS-PAGE and HPLC. For reference, the divalent scFv was also radiolabeled and run on the same gel. Electrophoresis under nonreducing (Fig. 2,B, Lanes 1 and 2) and reducing conditions (Fig. 2,B, Lanes 3 and 4) showed approximately ≥95% of the total radioactivity associated with the protein band of Mr ∼58,000 for both sc(Fv)2 (Fig. 2,B, Lanes 1and 3) and[sc(Fv)2]2 (Fig. 2,B, Lanes 2 and 4). Minor bands at Mr 30,000 and 45,000 with about ≤5%of the total incorporated radioactivity were also seen. In HPLC analysis, scFvs eluted as single peaks at Mr 120,000 and 60,000 for[sc(Fv)2]2 and sc(Fv)2, respectively (Fig. 3). The HPLC profile of CC49 IgG is also included in Fig. 3. HPLC profiles of[sc(Fv)2]2 were consistent therefore with the theoretical molecular weight and were not altered by the iodination procedure.

In solid-phase RIA, the specific bindings for CC49 IgG,[sc(Fv)2]2, and sc(Fv)2 were 85–95% with the nonspecific bindings between 0.8% and 1.5% (data not shown). Real-time kinetic analysis (Fig. 4) using BIAcore exhibited the rate of association (kon) for[sc(Fv)2]2 as ∼2-fold slower than IgG (9.05 × 104m−1s−1and 2.2 × 105m−1s−1,respectively), which was compensated by a dissociation rate(koff) also about 2–3-fold slower than IgG (8.91 × 10−4s−1 and 2.07 × 10−3s−1, respectively). Dimer had similar koff values as tetramer but showed a 4–5-fold slower kon. The association constant(KA) as determined by BIAcore analysis for [sc(Fv)2]2,sc(Fv)2, and CC49 IgG was 1.02 × 108m−1, 2.75 × 107m−1, and 1.14 × 108m−1, respectively.

Pharmacokinetic Studies with Tetravalent and Divalent scFvs.

As seen in Fig. 5, blood clearance of[sc(Fv)2]2 was faster than IgG but slower than sc(Fv)2. The data were analyzed using a biexponential model showing an α-phase T1/2 (the clearance of molecules from the blood to the extravascular space) and a β-phase T1/2(the clearance of molecules from blood to the nonextravascular space or out of the body). The elimination T1/2 of[sc(Fv)2]2,sc(Fv)2, and CC49 IgG was calculated as 170, 80,and 330 min, respectively. The overall clearance of radioactivity from the blood pool appeared to be a triphasic process with the third phase(24 and 48 h) being essentially the clearance of free radioiodide.[sc(Fv)2]2, therefore,exhibited almost a 2-fold increase in the biological half-life in blood as compared with divalent scFv. Similar trends were observed in whole body clearance studies with T1/2 values of 8.9 ± 1.3 and 5.1 ± 0.7 h for[sc(Fv)2]2 and sc(Fv)2, respectively (data not shown). At 48 h, ∼95% of the radiolabeled tetravalent and divalent scFvs were cleared from the body, indicating that these were not retained in the extravascular space or in any specific organ but were eliminated apparently through the urine. The whole body clearance for CC49 IgG was slower with only 75% clearing in 48 h.

Biodistribution Studies.

In dual-label biodistribution studies, a higher level of tumor targeting was observed with[sc(Fv)2]2 than with sc(Fv)2, with the %ID/g of 21.3 ± 1.3 and 10.5 ± 1.1 for[sc(Fv)2]2 and 9.8 ± 1.3 and 5.1 ± 0.7 for sc(Fv)2 at 6 and 24 h, respectively (Table 1). At the same time points, the %ID/g for CC49 IgG in tumor were 17.3 ± 1.1 and 28.4 ± 1.7, respectively. Elevated uptake was also detected in normal tissues, particularly the spleen and the liver with CC49 IgG (Table 1). By 24 h,[sc(Fv)2]2 exhibited 10.5 ± 1.1%ID/g in tumor with ≤2% uptake by normal organs. At 24 h, the retention of[sc(Fv)2]2 in the liver was about 3-fold higher than sc(Fv)2, indicating that the [sc(Fv)2]2,which was Mr ∼120,000, might be clearing through the liver. Although CC49 IgG showed 28.4 ± 1.7%ID/g in tumors at 24 h postadministration, the nonspecific retention of radiolabeled IgG in the liver, spleen, blood,and kidneys was noticeably higher than that of[sc(Fv)2]2 (Table 1). The RI (%ID/g of tumor divided by %ID/g of normal tissue) for well-vascularized organs such as the liver and spleen were determined;the tumor:liver and tumor:spleen ratios were 6.2:1 and 10.5:1 for[sc(Fv)2]2 and 8.5:1 and 8.5:1 for sc(Fv)2 at 24 h postadministration(Fig. 6). To investigate the occurrence of any specific or nonspecific accumulation of divalent and tetravalent scFvs in tissues, the tissue:blood ratio was calculated. At 6 h postadministration, both[sc(Fv)2]2 and sc(Fv)2 exhibited a 3–4-fold higher tumor accumulation than CC49 IgG (Fig. 7,A). The liver and spleen retention of [sc(Fv)2]2and CC49 IgG were similar at 6 h (Fig. 7,A). At 24 h, the specific accumulation of scFvs and CC49 IgG in tumors with the tumor:blood values were 35 for[sc(Fv)2]2, 25.5 for sc(Fv)2, and 2.3 for CC49 IgG (Fig. 7 B).

Stability of Tetravalent and Divalent scFv.

Both radiolabeled tetrameric and dimeric forms of scFvs were found to be stable in vitro up to 3 days at 37°C as observed by HPLC (Table 2). For both[sc(Fv)2]2 and sc(Fv)2, ≥90% radioactivity was found associated with the protein until 24 h. At 72 h,[sc(Fv)2]2 demonstrated 5.6% of the radioactivity in sc(Fv)2 and 2.9%radioactivity in low molecular weight proteins(Mr 45,000 and 30,000). This was attributed to radiolysis because the unlabeled proteins can be stored at −70°C for 6 months without any loss in immunoreactivity or the evidence of breakdown products (data not shown).

The in vivo stability of[sc(Fv)2]2 and sc(Fv)2 in blood plasma was analyzed in biodistribution studies. HPLC studies showed that at 1 h after injection, >70% of the radiolabel was associated with the tetravalent scFv corresponding to the protein peak of Mr 120,000 (Fig. 8). However, at 2 h after injection only 47% of the radioactivity was found to be associated with the[sc(Fv)2]2, with 38% in higher molecular weight forms probably attributable to aggregation with serum proteins. The divalent scFv demonstrated only 20% of the total radioactivity with the protein peak at 1 h postadministration,with the rest as free iodine (data not shown). Our earlier studies with noncovalent dimers of MAb CC49 have indicated that the noncovalent interactions between scFvs were stable when incubated in vitro with 1% mouse serum at 37°C for up to 4 days with the appearance of higher molecular weight forms at later time points, 24 and 48 h (18, 25).

Tumor-specific MAbs have been used as carrier vehicles for the targeted delivery of cytocidal agents to cancer cells (8, 11). For better outcomes of radioimmunotherapy, it has remained a challenge to optimize the therapeutic index by improving the tumor localization and reducing the uptake by the normal organs(47). Larger molecules like IgG and F(ab′)2 exhibit high tumor uptake, yet the slow blood clearance of these forms results in low tumor specificity with significant toxicity to the normal tissues (15, 17, 18). Smaller fragments like Fab and scFvs clear rapidly from circulation thereby showing improved tumor specificity but resulting in poor tumor accretion (8, 14, 15, 17, 44).

The murine MAb CC49 is one of the most extensively studied MAbs for cancer therapy. It has shown efficient targeting of radionuclides to human colon carcinoma xenografts in nude mice and is currently in clinical trials (48, 49, 50, 51). We have previously constructed and characterized the monovalent CC49 scFv (14, 52). The monovalent scFv offered two major advantages: (a) the rate of clearance of radioiodinated scFv from blood pool and normal tissues was much more rapid than that seen with intact IgG,F(ab′)2 fragments, or Fab fragments, which resulted in the possibility of earlier imaging times and less radiation toxicity to normal tissues in therapy (44); and(b) improved autoradiographic studies were feasible because of homogeneous tumor penetration (16, 20). However,monovalent scFv showed a lower percentage of injected dose that localized in the tumor when compared with the divalent IgG molecule,because they depend on intrinsic affinity rather than functional affinity to remain bound to the antigen (18). Noncovalent and covalent dimeric CC49 scFvs were constructed subsequently to improve the tumor accretion for radioimmunotherapy (21, 25, 40). In biodistribution studies, at 6 h postadministration the tumor localization of radioiodinated dimers was ∼5-fold higher than with monomers with a 3-fold increase in the serum half-life(18). As CC49 recognizes sialyl-Tn epitope, a unique disaccharide present in multiple copies of TAG-72, additional increase in valency by generating tetravalent scFv, without compromising much on pharmacokinetics advantage inherent to the scFv, should provide a better reagent for radioimmunotherapeutic applications.

We describe here the generation and characterization of tetravalent CC49 scFv formed by the noncovalent interactions of covalent dimers(Fig. 1 B). Such noncovalent association of scFvs to yield multimers has been reported by several investigators (18, 53, 54). The multimers have been shown to exhibit stable thermodynamic characteristics rather than being associated with a simple equilibrium (18, 23). Once purified, the tetravalent scFv was found to be stable and did not dissociate upon dilution.

Both intrinsic affinity and antibody valency are known to contribute to the overall antigen binding and subsequent immunoreactivity of antibodies (10). ScFv multimers (trivalent and tetravalent) have demonstrated, therefore, significantly increased binding affinity in vitro as compared with the monovalent forms, probably attributable to increased functional affinity through multiple antigen-antibody interactions (26, 27, 28, 29, 33, 34). It has been reported (55) that IgG dimers (tetravalent)exhibited a 150-fold higher KA as compared with monomer by SPR studies. A comparative SPR study with scFv monomers, dimers, and tetramers formed by the streptavidin moiety showed tetramers with the highest association rate constant with no measurable dissociation over a period of 15 min (26, 56). Santos et al.(33) reported that the tetravalent CC49 scFvs have a significantly higher relative antigen-binding affinity than the parent IgG by competitive ELISA.

The functional affinity of tetravalent scFv was compared with divalent scFv and IgG based on SPR measured with the BIAcore system (Fig. 4). Muller et al.(57) showed that accurate kinetic rate constants for multivalent binding interactions that use the solid-phase antigen molecules cannot be determined. Therefore, for comparing the functional affinity of multivalent antibody fragments, it is essential to determine the dissociation rates(koff) and the KA of its monovalent fragment(58). The SPR studies revealed the KA value for[sc(Fv)2]2 was similar to CC49 IgG and was 3–4-fold higher than sc(Fv)2and ∼100-fold higher than the monovalent CC49 scFv(KA = 1.4 × 106m−1; Ref.59). Compared with the CC49 IgG, the tetramer showed a 2–3-fold slowed koff rate, which suggested a stronger association with the antigen (Fig. 4). It has been shown that multiple bindings can effectively reduce the off rates thereby increasing the retention time of the antibody bound to the target antigen (29, 30). A direct extrapolation of the results of in vitro affinity measurements to the in vivo pharmacokinetics of a bivalent/multivalent antigen-antibody interaction remains difficult because of variability in antigen density, endocytosis and clustering of certain cellular antigens, and stearic constraints of both antibody and antigen (60, 61). However, in biodistribution studies, a significant increase in tumor retention was observed with the increase in the functional affinity of the scFvs from monovalent to divalent interactions (18, 25).

An improved tumor targeting over monovalent scFvs has been achieved using scFv dimers (17, 18, 24, 25, 62, 63, 64). Tetravalent immunoglobulin dimers (55, 65, 66) and F(ab′)n where n = 3, 4(36, 37, 67) have also been constructed in the search for better tumor-targeting reagents. However, the in vivotumor-targeting efficiency of genetically engineered multivalent(trivalent and tetravalent) scFvs has not been investigated. A direct comparison of in vivo biodistribution and tumor localization of radioiodinated tetravalent scFv was performed with CC49 IgG and sc(Fv)2 in xenograft-bearing mice.[sc(Fv)2]2 showed a rapid tumor accretion with low nonspecific uptake by normal organs. The nonspecific uptake of[sc(Fv)2]2 was ∼4-fold lower than that of CC49 IgG in the organs tested. With divalent sc(Fv)2, the absolute tumor uptake was ∼2-fold lower than [sc(Fv)2]2after 4 h of administration, which suggested that the tetravalent scFv would be a better candidate for therapeutic applications. At 16 h, the tumor uptake of[sc(Fv)2]2 was 2-fold greater than sc(Fv)2, thereby reaching values like that of IgG, with significantly lower nonspecific localization. Monovalent scFvs exhibited tumor uptakes of only 2.0 and 1.1%ID/g at 6 h and 24 h, respectively (8).

In animal models, an increase in MAb affinity has been correlated to greater tumor localization (68, 69). However, with an increase in binding affinity the phenomenon of “binding site barrier” occurs (70). This results in the localization of high affinity antibody/fragments to only the well-vascularized regions of the tumor with a poor interstitial penetration(71). In the present study, we have not performed a quantitative autoradiography to define the tumor penetration of radiolabeled tetravalent scFv as compared with other immunoglobulin forms or sc(Fv)2. Although the tetravalent scFv has a similar binding affinity(KA) as that of prototype CC49 MAb,the advantage of being a smaller molecule than IgG can favor both tumor vessel and interstitial diffusion of[sc(Fv)2]2. A quantitative autoradiographic study by Yokota et al.(16) has demonstrated the impact of the size of an antibody fragment on tumor penetration and distribution. Nevertheless,it is important to maintain a balance between total percentage of dose accumulating in solid tumors versus interstitial penetration of the antibody-based molecules.

An important parameter in evaluating the therapeutic efficacy of the radiolabeled MAb constructs is the RI. At 24 h, the tetravalent scFv exhibited a tumor:blood ratio that was about 15-fold higher than IgG and 1.5-fold higher than sc(Fv)2. Similarly,[sc(Fv)2]2 demonstrated a higher RI than sc(Fv)2 and intact IgG in all of the other organs tested with the exception of the liver, which could be the possible site for the elimination of tetravalent scFv as evidenced by the lower kidney uptake of[sc(Fv)2]2.

Pharmacokinetic studies showed a 2-fold increase in the circulating half-life of [sc(Fv)2]2as compared with sc(Fv)2. This increased persistence of the tetramer in the blood allows increased uptake by the tumor. The blood T1/2 for CC49 IgG was 2-fold longer than [sc(Fv)2]2,which explained the higher nonspecific uptakes. Whole body clearance studies showed that the blood clearance of radiolabeled proteins was indeed because of the removal of the radionuclide from the body and not because of the accumulation in some specific organ or extravascular space. [sc(Fv)2]2displayed a rapid clearance from the whole body, similar to the blood clearance pattern.

In summary, the tetravalent scFv of MAb CC49 was constructed to improve the tumor targeting. The tetramer was found to be stable with better functional affinity than the divalent forms (IgG and covalent scFv dimer) and the monovalent scFvs. Biodistribution studies in xenografted mice demonstrated that the construct showed higher tumor accretion with low uptake by the normal organs than it did with IgG and divalent scFv. We conclude that the tetravalent scFv will be ideal for radioimmunotherapy because of the longer half-life in blood than divalent scFv and overall clearance that provides low radiation doses to the normal organs.

Fig. 1.

Schematic drawing of the: A, covalent dimer expressed in P. pastoris. L designates the 205C linker consisting of 25 amino acids, and VL and VH are the variable light and heavy domains of MAb CC49, respectively; and B, schematic model of tetravalent scFv. Noncovalent interactions are depicted by dotted lines.

Fig. 1.

Schematic drawing of the: A, covalent dimer expressed in P. pastoris. L designates the 205C linker consisting of 25 amino acids, and VL and VH are the variable light and heavy domains of MAb CC49, respectively; and B, schematic model of tetravalent scFv. Noncovalent interactions are depicted by dotted lines.

Close modal
Fig. 2.

SDS-PAGE analysis of purified[sc(Fv)2]2 and sc(Fv)2 under nonreducing and reducing conditions; A, gel showing the Coomassie Blue R-250 staining of unlabeled scFvs; B,phosphorescence image of the 125I-labeled scFvs.

Fig. 2.

SDS-PAGE analysis of purified[sc(Fv)2]2 and sc(Fv)2 under nonreducing and reducing conditions; A, gel showing the Coomassie Blue R-250 staining of unlabeled scFvs; B,phosphorescence image of the 125I-labeled scFvs.

Close modal
Fig. 3.

HPLC size-exclusion of the 125I-CC49 IgG, 125I-[sc(Fv)2]2, and 125I-sc(Fv)2. The IgG (○),[sc(Fv)2]2 (•), and sc(Fv)2(□) were eluted as single peaks with a Mrof 150,000, 120,000, and 60,000, respectively.

Fig. 3.

HPLC size-exclusion of the 125I-CC49 IgG, 125I-[sc(Fv)2]2, and 125I-sc(Fv)2. The IgG (○),[sc(Fv)2]2 (•), and sc(Fv)2(□) were eluted as single peaks with a Mrof 150,000, 120,000, and 60,000, respectively.

Close modal
Fig. 4.

BIAcore analysis of CC49 antibody constructs. Sensogram demonstrates the binding and dissociation of IgG,[sc(Fv)2]2, and sc(Fv)2.

Fig. 4.

BIAcore analysis of CC49 antibody constructs. Sensogram demonstrates the binding and dissociation of IgG,[sc(Fv)2]2, and sc(Fv)2.

Close modal
Fig. 5.

Pharmacokinetics of blood pool clearance of[sc(Fv)2]2, sc(Fv)2, and CC49 IgG in the athymic mice bearing LS-174T colon carcinoma xenografts. Each data point represents the average of two experiments.

Fig. 5.

Pharmacokinetics of blood pool clearance of[sc(Fv)2]2, sc(Fv)2, and CC49 IgG in the athymic mice bearing LS-174T colon carcinoma xenografts. Each data point represents the average of two experiments.

Close modal
Fig. 6.

Comparative biodistribution studies of[sc(Fv)2]2, sc(Fv)2, and CC49 IgG(RI; %ID/g of tumor divided by %ID/g of normal tissue) at 24 h postadministration. Iodinated proteins were injected into athymic mice bearing LS-174T tumors. Mice were killed at indicated times, and the RI for each organ was determined.

Fig. 6.

Comparative biodistribution studies of[sc(Fv)2]2, sc(Fv)2, and CC49 IgG(RI; %ID/g of tumor divided by %ID/g of normal tissue) at 24 h postadministration. Iodinated proteins were injected into athymic mice bearing LS-174T tumors. Mice were killed at indicated times, and the RI for each organ was determined.

Close modal
Fig. 7.

Biodistribution studies (tissue:blood ratio) with[sc(Fv)2]2, sc(Fv)2, and CC49 IgG. Radioiodinated proteins were injected into athymic mice bearing LS-174T tumors. Mice were sacrificed (A) 6 h and(B) 24 h postadministration, and the ratio of%ID/g of tissue:%ID/g of blood was calculated.

Fig. 7.

Biodistribution studies (tissue:blood ratio) with[sc(Fv)2]2, sc(Fv)2, and CC49 IgG. Radioiodinated proteins were injected into athymic mice bearing LS-174T tumors. Mice were sacrificed (A) 6 h and(B) 24 h postadministration, and the ratio of%ID/g of tissue:%ID/g of blood was calculated.

Close modal
Fig. 8.

In vivo stability study of radioiodinated[sc(Fv)2]2. Serum samples were obtained at different postadministration times from athymic mice bearing LS-174T tumors. Samples were analyzed by HPLC, and the percentage of radioactivity associated with the scFv was determined. The fraction numbers corresponding to the[sc(Fv)2]2 and free iodine peak were 31 and 71, respectively.

Fig. 8.

In vivo stability study of radioiodinated[sc(Fv)2]2. Serum samples were obtained at different postadministration times from athymic mice bearing LS-174T tumors. Samples were analyzed by HPLC, and the percentage of radioactivity associated with the scFv was determined. The fraction numbers corresponding to the[sc(Fv)2]2 and free iodine peak were 31 and 71, respectively.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by a Grant from the United States Department of Energy (DE-FG02-95ER62024).

3

The abbreviations used are: MAbs, monoclonal antibodies; scFv, single chain Fv; sc(Fv)2, covalent dimeric scFv; [sc(Fv)2]2, noncovalent tetrameric scFv; HPLC, high performance liquid chromatography; SPR,surface plasmon resonance; RI, radiolocalization index; %ID/g, % of injected dose/g; BSM, bovine submaxillary gland mucin; VL,variable light; VH, variable heavy.

Table 1

Biodistribution of CC49 IgG, [sc(Fv)2]2, and sc(Fv)2 (%ID/g) in athymic mice bearing LS-174T xenografts

TissueTime (h)
0.5146162448
[sc(Fv)2]2        
Tumor 6.2 ± 1.0 9.0 ± 1.2 12.3 ± 1.6 21.3 ± 1.3 17.0 ± 1.6 10.5 ± 1.1 8.2 ± 0.1 
Blood 22.0 ± 0.6 12.2 ± 1.4 8.5 ± 0.7 4.5 ± 0.4 1.3 ± 0.9 0.3 ± 0.1 0.2 ± 0.0 
Liver 7.9 ± 0.6 6.0 ± 0.8 7.0 ± 1.2 4.9 ± 0.6 3.4 ± 0.8 1.7 ± 0.3 1.2 ± 0.1 
Spleen 9.0 ± 1.4 7.0 ± 0.6 6.4 ± 1.2 5.1 ± 1.5 2.5 ± 0.2 1.0 ± 0.2 0.4 ± 0.0 
Kidneys 8.5 ± 0.6 5.4 ± 0.4 3.6 ± 0.1 2.9 ± 0.2 1.5 ± 0.1 0.8 ± 0.1 0.3 ± 0.0 
Heart 4.7 ± 0.3 4.8 ± 0.3 2.5 ± 0.2 1.7 ± 0.1 0.4 ± 0.0 0.2 ± 0.0 0.1 ± 0.0 
Lungs 5.6 ± 0.1 5.1 ± 0.4 2.6 ± 0.7 0.7 ± 0.2 0.5 ± 0.1 0.3 ± 0.0 0.1 ± 0.0 
sc(Fv)2        
Tumor 6.8 ± 1.0 7.1 ± 1.0 8.0 ± 1.0 9.8 ± 1.3 7.3 ± 0.8 5.1 ± 0.7 4.0 ± 0.2 
Blood 19.2 ± 0.8 12.0 ± 0.5 4.1 ± 0.4 2.9 ± 0.3 0.3 ± 0.0 0.2 ± 0.0 0.1 ± 0.0 
Liver 5.9 ± 0.8 4.9 ± 1.2 2.0 ± 1.2 1.7 ± 0.4 1.2 ± 0.6 0.6 ± 0.2 0.2 ± 0.1 
Spleen 7.1 ± 0.7 3.9 ± 0.5 2.8 ± 0.8 1.6 ± 0.6 1.2 ± 0.1 0.6 ± 0.1 0.2 ± 0.0 
Kidneys 21.1 ± 1.2 13.0 ± 0.7 2.9 ± 0.3 1.9 ± 0.2 1.1 ± 0.0 0.4 ± 0.0 0.2 ± 0.0 
Heart 4.2 ± 0.2 3.4 ± 0.2 1.0 ± 0.1 0.7 ± 0.1 0.2 ± 0.0 0.1 ± 0.0 0.1 ± 0.0 
Lungs 5.5 ± 1.4 2.1 ± 0.3 1.6 ± 0.3 1.3 ± 0.2 0.6 ± 0.0 0.2 ± 0.1 0.1 ± 0.1 
CC49 IgG        
Tumor 5.2 ± 0.8 7.7 ± 0.9 12.5 ± 0.8 17.3 ± 1.1 19.5 ± 1.3 28.4 ± 1.7 34.3 ± 2.4 
Blood 43.7 ± 2.7 31.9 ± 1.1 28.1 ± 1.3 18.6 ± 1.3 13.7 ± 1.5 12.5 ± 0.9 9.8 ± 1.1 
Liver 21.2 ± 1.6 19.5 ± 2.4 18.7 ± 1.2 15.5 ± 1.6 13.9 ± 1.4 12.5 ± 0.6 11.4 ± 0.6 
Spleen 28.2 ± 2.9 24.2 ± 3.3 18.0 ± 1.9 16.9 ± 0.8 14.8 ± 1.2 13.8 ± 1.0 12.5 ± 0.7 
Kidneys 8.0 ± 0.4 5.8 ± 0.5 4.7 ± 0.5 3.9 ± 0.6 3.1 ± 0.5 2.8 ± 0.3 2.4 ± 0.1 
Heart 7.0 ± 0.4 5.5 ± 0.6 4.4 ± 0.5 3.9 ± 0.7 3.2 ± 0.7 3.3 ± 0.3 2.5 ± 0.2 
Lungs 14.8 ± 1.0 10.2 ± 1.2 9.7 ± 1.1 7.7 ± 0.3 5.7 ± 0.1 4.3 ± 0.3 3.4 ± 0.1 
TissueTime (h)
0.5146162448
[sc(Fv)2]2        
Tumor 6.2 ± 1.0 9.0 ± 1.2 12.3 ± 1.6 21.3 ± 1.3 17.0 ± 1.6 10.5 ± 1.1 8.2 ± 0.1 
Blood 22.0 ± 0.6 12.2 ± 1.4 8.5 ± 0.7 4.5 ± 0.4 1.3 ± 0.9 0.3 ± 0.1 0.2 ± 0.0 
Liver 7.9 ± 0.6 6.0 ± 0.8 7.0 ± 1.2 4.9 ± 0.6 3.4 ± 0.8 1.7 ± 0.3 1.2 ± 0.1 
Spleen 9.0 ± 1.4 7.0 ± 0.6 6.4 ± 1.2 5.1 ± 1.5 2.5 ± 0.2 1.0 ± 0.2 0.4 ± 0.0 
Kidneys 8.5 ± 0.6 5.4 ± 0.4 3.6 ± 0.1 2.9 ± 0.2 1.5 ± 0.1 0.8 ± 0.1 0.3 ± 0.0 
Heart 4.7 ± 0.3 4.8 ± 0.3 2.5 ± 0.2 1.7 ± 0.1 0.4 ± 0.0 0.2 ± 0.0 0.1 ± 0.0 
Lungs 5.6 ± 0.1 5.1 ± 0.4 2.6 ± 0.7 0.7 ± 0.2 0.5 ± 0.1 0.3 ± 0.0 0.1 ± 0.0 
sc(Fv)2        
Tumor 6.8 ± 1.0 7.1 ± 1.0 8.0 ± 1.0 9.8 ± 1.3 7.3 ± 0.8 5.1 ± 0.7 4.0 ± 0.2 
Blood 19.2 ± 0.8 12.0 ± 0.5 4.1 ± 0.4 2.9 ± 0.3 0.3 ± 0.0 0.2 ± 0.0 0.1 ± 0.0 
Liver 5.9 ± 0.8 4.9 ± 1.2 2.0 ± 1.2 1.7 ± 0.4 1.2 ± 0.6 0.6 ± 0.2 0.2 ± 0.1 
Spleen 7.1 ± 0.7 3.9 ± 0.5 2.8 ± 0.8 1.6 ± 0.6 1.2 ± 0.1 0.6 ± 0.1 0.2 ± 0.0 
Kidneys 21.1 ± 1.2 13.0 ± 0.7 2.9 ± 0.3 1.9 ± 0.2 1.1 ± 0.0 0.4 ± 0.0 0.2 ± 0.0 
Heart 4.2 ± 0.2 3.4 ± 0.2 1.0 ± 0.1 0.7 ± 0.1 0.2 ± 0.0 0.1 ± 0.0 0.1 ± 0.0 
Lungs 5.5 ± 1.4 2.1 ± 0.3 1.6 ± 0.3 1.3 ± 0.2 0.6 ± 0.0 0.2 ± 0.1 0.1 ± 0.1 
CC49 IgG        
Tumor 5.2 ± 0.8 7.7 ± 0.9 12.5 ± 0.8 17.3 ± 1.1 19.5 ± 1.3 28.4 ± 1.7 34.3 ± 2.4 
Blood 43.7 ± 2.7 31.9 ± 1.1 28.1 ± 1.3 18.6 ± 1.3 13.7 ± 1.5 12.5 ± 0.9 9.8 ± 1.1 
Liver 21.2 ± 1.6 19.5 ± 2.4 18.7 ± 1.2 15.5 ± 1.6 13.9 ± 1.4 12.5 ± 0.6 11.4 ± 0.6 
Spleen 28.2 ± 2.9 24.2 ± 3.3 18.0 ± 1.9 16.9 ± 0.8 14.8 ± 1.2 13.8 ± 1.0 12.5 ± 0.7 
Kidneys 8.0 ± 0.4 5.8 ± 0.5 4.7 ± 0.5 3.9 ± 0.6 3.1 ± 0.5 2.8 ± 0.3 2.4 ± 0.1 
Heart 7.0 ± 0.4 5.5 ± 0.6 4.4 ± 0.5 3.9 ± 0.7 3.2 ± 0.7 3.3 ± 0.3 2.5 ± 0.2 
Lungs 14.8 ± 1.0 10.2 ± 1.2 9.7 ± 1.1 7.7 ± 0.3 5.7 ± 0.1 4.3 ± 0.3 3.4 ± 0.1 
Table 2

Stability of divalent and tetravalent CC49 scFvs at 37°C

Radioiodinated scFvs (0.05 mg/ml) were mixed with 1% BSA and incubated. Samples were run on size exclusion chromatography and the percentage radioactivity associated with various forms was calculated. Proteins categorized under high- and low-molecular weights were≥130,000 and ≤50,000 respectively.

Immunoglobulin form% radioactivity inTime (h)
06244872
[sc(Fv)2]2 High Mr proteins 0.4 0.6 0.9 1.1 1.6 
 [sc(Fv)2]2 96.2 95.2 90.2 89.2 85.6 
 sc(Fv)2 1.1 1.4 3.8 4.2 5.6 
 Low Mr proteins 0.9 1.5 1.9 2.1 2.9 
 Iodine 1.4 1.3 3.2 3.4 4.3 
sc(Fv)2 High Mr proteins 0.5 0.8 1.4 1.6 1.9 
 sc(Fv)2 97.2 95.7 92.9 90.8 88.4 
 Low Mr proteins 1.2 1.9 2.5 3.7 4.5 
 Iodine 1.1 1.6 3.2 3.9 5.2 
Immunoglobulin form% radioactivity inTime (h)
06244872
[sc(Fv)2]2 High Mr proteins 0.4 0.6 0.9 1.1 1.6 
 [sc(Fv)2]2 96.2 95.2 90.2 89.2 85.6 
 sc(Fv)2 1.1 1.4 3.8 4.2 5.6 
 Low Mr proteins 0.9 1.5 1.9 2.1 2.9 
 Iodine 1.4 1.3 3.2 3.4 4.3 
sc(Fv)2 High Mr proteins 0.5 0.8 1.4 1.6 1.9 
 sc(Fv)2 97.2 95.7 92.9 90.8 88.4 
 Low Mr proteins 1.2 1.9 2.5 3.7 4.5 
 Iodine 1.1 1.6 3.2 3.9 5.2 

We thank H. Conway, K. Devish, J. Jokerst, and Erik Moore for expert technical assistance. We acknowledge the Molecular Interaction Facility and the Molecular Biology Core Lab at the University of Nebraska Medical Center for the BIAcore and sequencing studies. The CC49 scFv construct was a generous gift from the National Cancer Institute Laboratory of Tumor Immunology and Biology and the Dow Chemical Company.

1
Chester K. A., Hawkins R. E. Clinical issues in antibody design.
Trends Biotechnol.
,
13
:
294
-300,  
1995
.
2
Meredith R. F., LoBuglio A. F., Spencer E. B. Recent progress in radioimmunotherapy for cancer.
Oncology (Huntingt)
,
11
:
979
-984,  
1997
.
3
Murray J. L., Macey D. J., Kasi L. P., Rieger P., Cunningham J., Bhadkamkar V., Zhang H. Z., Schlom J., Rosenblum M. G., Podoloff D. A. Phase II radioimmunotherapy trial with 131I-CC49 in colorectal cancer.
Cancer (Phila.)
,
73
:
1057
-1066,  
1994
.
4
Juweid M. E., Zhang C. H., Blumenthal R. D., Hajjar G., Sharkey R. M., Goldenberg D. M. Prediction of hematologic toxicity after radioimmunotherapy with (131)I-labeled anticarcinoembryonic antigen monoclonal antibodies.
J. Nucl. Med.
,
40
:
1609
-1616,  
1999
.
5
Press O. W. Radiolabeled antibody therapy of B-cell lymphomas.
Semin. Oncol.
,
26
:
58
-65,  
1999
.
6
Akabani G., Cokgor I., Coleman R. E., Gonzalez Trotter D., Wong T. Z., Friedman H. S., Friedman A. H., Garcia-Turner A., Herndon J. E., DeLong D., McLendon R. E., Zhao X. G., Pegram C. N., Provenzale J. M., Bigner D. D., Zalutsky M. R. Dosimetry and dose-response relationships in newly diagnosed patients with malignant gliomas treated with iodine-131-labeled anti-tenascin monoclonal antibody 81C6 therapy.
Int. J. Radiat. Oncol. Biol. Phys.
,
46
:
947
-958,  
2000
.
7
Waldmann T. A. Monoclonal antibodies in diagnosis and therapy.
Science (Washington DC)
,
252
:
1657
-1662,  
1991
.
8
Colcher D., Goel A., Pavlinkova G., Beresford G., Booth B., Batra S. K. Effects of genetic engineering on the pharmacokinetics of antibodies.
Q. J. Nucl. Med.
,
43
:
132
-139,  
1999
.
9
Jain R. K. Transport of molecules across tumor vasculature.
Cancer Metastasis Rev.
,
6
:
559
-593,  
1987
.
10
Pluckthun A., Pack P. New protein engineering approaches to multivalent and bispecific antibody fragments.
Immunotechnology (Amsterdam)
,
3
:
83
-105,  
1997
.
11
Hudson P. J. Recombinant antibody constructs in cancer therapy.
Curr. Opin. Immunol.
,
11
:
548
-557,  
1999
.
12
Bird R. E., Hardman K. D., Jacobson J. W., Johnson S., Kaufman B. M., Lee S. M., Lee T., Pope S. H., Riordan G. S., Whitlow M. Single-chain antigen-binding proteins.
Science (Washington DC)
,
242
:
423
-426,  
1988
.
13
Brinkmann U., Reiter Y., Jung S. H., Lee B., Pastan I. A recombinant immunotoxin containing a disulfide-stabilized Fv fragment.
Proc. Natl. Acad. Sci. USA
,
90
:
7538
-7542,  
1993
.
14
Colcher D., Bird R., Roselli M., Hardman K. D., Johnson S., Pope S., Dodd S. W., Pantoliano M. W., Milenic D. E., Schlom J. In vivo tumor targeting of a recombinant single-chain antigen-binding protein.
J. Natl. Cancer Inst. (Bethesda)
,
82
:
1191
-1197,  
1990
.
15
Milenic D. E., Yokota T., Filpula D. R., Finkelman M. A., Dodd S. W., Wood J. F., Whitlow M., Snoy P., Schlom J. Construction, binding properties, metabolism, and tumor targeting of a single-chain Fv derived from the pancarcinoma monoclonal antibody CC49.
Cancer Res.
,
51
:
6363
-6371,  
1991
.
16
Yokota T., Milenic D. E., Whitlow M., Schlom J. Rapid tumor penetration of a single-chain Fv and comparison with other immunoglobulin forms.
Cancer Res.
,
52
:
3402
-3408,  
1992
.
17
Adams G. P., McCartney J. E., Tai M. S., Oppermann H., Huston J. S., Stafford W. F., III, Bookman M. A., Fand I., Houston L. L., Weiner L. M. Highly specific in vivo tumor targeting by monovalent and divalent forms of 741F8 anti-c-erbB-2 single-chain Fv.
Cancer Res.
,
53
:
4026
-4034,  
1993
.
18
Pavlinkova G., Beresford G. W., Booth B. J., Batra S. K., Colcher D. Pharmacokinetics and biodistribution of engineered single-chain antibody constructs of MAb CC49 in colon carcinoma xenografts.
J. Nucl. Med.
,
40
:
1536
-1546,  
1999
.
19
Begent R. H., Verhaar M. J., Chester K. A., Casey J. L., Green A. J., Napier M. P., Hope-Stone L. D., Cushen N., Keep P. A., Johnson C. J., Hawkins R. E., Hilson A. J., Robson L. Clinical evidence of efficient tumor targeting based on single-chain Fv antibody selected from a combinatorial library.
Nat. Med.
,
2
:
979
-984,  
1996
.
20
Larson S. M., El-Shirbiny A. M., Divgi C. R., Sgouros G., Finn R. D., Tschmelitsch J., Picon A., Whitlow M., Schlom J., Zhang J., Cohen A. M. Single chain antigen binding protein (sFv CC49): first human studies in colorectal carcinoma metastatic to liver.
Cancer (Phila.)
,
80
:
2458
-2468,  
1997
.
21
Pavlinkova G., Booth B. J., Batra S. K., Colcher D. Radioimmunotherapy of human colon cancer xenografts using a dimeric single-chain Fv antibody construct.
Clin. Cancer Res.
,
5
:
2613
-2619,  
1999
.
22
Johnson V. G., Schlom J., Paterson A. J., Bennett J., Magnani J. L., Colcher D. Analysis of a human tumor-associated glycoprotein (TAG-72) identified by monoclonal antibody B72.
3. Cancer Res.
,
46
:
850
-857,  
1986
.
23
Wu A. M., Chen W., Raubitschek A., Williams L. E., Neumaier M., Fischer R., Hu S. Z., Odom-Maryon T., Wong J. Y., Shively J. E. Tumor localization of anti-CEA single-chain Fvs: improved targeting by non-covalent dimers.
Immunotechnology (Amsterdam)
,
2
:
21
-36,  
1996
.
24
Adams G. P., Schier R., McCall A. M., Crawford R. S., Wolf E. J., Weiner L. M., Marks J. D. Prolonged in vivo tumour retention of a human diabody targeting the extracellular domain of human HER2/neu.
Br. J. Cancer
,
77
:
1405
-1412,  
1998
.
25
Beresford G. W., Pavlinkova G., Booth B. J., Batra S. K., Colcher D. Binding characteristics and tumor targeting of a covalently linked divalent CC49 single-chain antibody.
Int. J. Cancer
,
81
:
911
-917,  
1999
.
26
Kipriyanov S. M., Breitling F., Little M., Dubel S. Single-chain antibody streptavidin fusions: tetrameric bifunctional scFv-complexes with biotin binding activity and enhanced affinity to antigen.
Hum. Antib. Hybrid.
,
6
:
93
-101,  
1995
.
27
Pack P., Muller K., Zahn R., Pluckthun A. Tetravalent miniantibodies with high avidity assembling in Escherichia coli.
J. Mol. Biol.
,
246
:
28
-34,  
1995
.
28
Rheinnecker M., Hardt C., Ilag L. L., Kufer P., Gruber R., Hoess A., Lupas A., Rottenberger C., Pluckthun A., Pack P. Multivalent antibody fragments with high functional affinity for a tumor-associated carbohydrate antigen.
J. Immunol.
,
157
:
2989
-2997,  
1996
.
29
Iliades P., Kortt A. A., Hudson P. J. Triabodies: single chain Fv fragments without a linker form trivalent trimers.
FEBS Lett.
,
409
:
437
-441,  
1997
.
30
Kortt A. A., Lah M., Oddie G. W., Gruen C. L., Burns J. E., Pearce L. A., Atwell J. L., McCoy A. J., Howlett G. J., Metzger D. W., Webster R. G., Hudson P. J. Single-chain Fv fragments of anti-neuraminidase antibody NC10 containing five- and ten-residue linkers form dimers and with zero-residue linker a trimer.
Protein Eng.
,
10
:
423
-433,  
1997
.
31
Libyh M. T., Goossens D., Oudin S., Gupta N., Dervillez X., Juszczak G., Cornillet P., Bougy F., Reveil B., Philbert F., Tabary T., Klatzmann D., Rouger P., Cohen J. H. A recombinant human scFv anti-Rh(D) antibody with multiple valences using a C-terminal fragment of C4-binding protein.
Blood
,
90
:
3978
-3983,  
1997
.
32
Coloma M. J., Morrison S. L. Design and production of novel tetravalent bispecific antibodies.
Nat. Biotechnol.
,
15
:
159
-163,  
1997
.
33
Santos A. D., Kashmiri S. V., Hand P. H., Schlom J., Padlan E. A. Generation and characterization of a single gene-encoded single-chain-tetravalent antitumor antibody.
Clin. Cancer Res.
,
5(Suppl.)
:
3118s
-3123s,  
1999
.
34
Le Gall F., Kipriyanov S. M., Moldenhauer G., Little M. Di-, tri- and tetrameric single chain Fv antibody fragments against human CD19: effect of valency on cell binding.
FEBS Lett.
,
453
:
164
-168,  
1999
.
35
Alt M., Muller R., Kontermann R. E. Novel tetravalent and bispecific IgG-like antibody molecules combining single-chain diabodies with the immunoglobulin γ1 Fc or CH3 region.
FEBS Lett.
,
454
:
90
-94,  
1999
.
36
Schott M. E., Frazier K. A., Pollock D. K., Verbanac K. M. Preparation, characterization, and in vivo biodistribution properties of synthetically cross-linked multivalent antitumor antibody fragments.
Bioconjug. Chem.
,
4
:
153
-165,  
1993
.
37
King D. J., Turner A., Farnsworth A. P., Adair J. R., Owens R. J., Pedley R. B., Baldock D., Proudfoot K. A., Lawson A. D., Beeley N. R., et al .
Improved tumor targeting with chemically cross-linked recombinant antibody fragments. Cancer Res.
,
54
:
6176
-6185,  
1994
.
38
Werlen R. C., Lankinen M., Offord R. E., Schubiger P. A., Smith A., Rose K. Preparation of a trivalent antigen-binding construct using polyoxime chemistry: improved biodistribution and potential for therapeutic application.
Cancer Res.
,
56
:
809
-815,  
1996
.
39
Pantoliano M. W., Bird R. E., Johnson S., Asel E. D., Dodd S. W., Wood J. F., Hardman K. D. Conformational stability, folding, and ligand-binding affinity of single-chain Fv immunoglobulin fragments expressed in Escherichia coli.
Biochemistry
,
30
:
10117
-10125,  
1991
.
40
Goel A., Beresford G. W., Colcher D., Pavlinkova G., Booth B. J. M., Baranowska-Kortylewicz J., Batra S. K. Divalent form of CC49 single-chain antibody constructs in Pichia pastoris: expression, purification and characterization.
J. Biochem. (Tokyo)
,
127
:
829
-836,  
2000
.
41
Lowry O., Rosenbrough N. J., Farr A. L., Randall R. J. Protein measurement by Folin phenol.
J. Biol. Chem.
,
193
:
265
-275,  
1951
.
42
Laemmli U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature (Lond.)
,
227
:
680
-685,  
1970
.
43
Colcher D., Zalutsky M., Kaplan W., Kufe D., Austin F., Schlom J. Radiolocalization of human mammary tumors in athymic mice by a monoclonal antibody.
Cancer Res.
,
43
:
736
-742,  
1983
.
44
Colcher D., Pavlinkova G., Beresford G., Booth B. J., Batra S. K. Single-chain antibodies in pancreatic cancer.
Ann. NY Acad. Sci.
,
880
:
263
-280,  
1999
.
45
Karlsson R., Michaelsson A., Mattsson L. Kinetic analysis of monoclonal antibody-antigen interactions with a new biosensor based analytical system.
J. Immunol. Methods
,
145
:
229
-240,  
1991
.
46
Tom B. H., Rutzky L. P., Jakstys M. M., Oyasu R., Kaye C. I., Kahan B. D. Human colonic adenocarcinoma cells.
I. Establishment and description of a new line. In Vitro (Rockv.)
,
12
:
180
-191,  
1976
.
47
Buchsbaum D. J., Rogers B. E., Khazaeli M. B., Mayo M. S., Milenic D. E., Kashmiri S. V., Anderson C. J., Chappell L. L., Brechbiel M. W., Curiel D. T. Targeting strategies for cancer radiotherapy.
Clin. Cancer Res.
,
5(Suppl.)
:
3048s
-3055s,  
1999
.
48
Divgi C. R., Scott A. M., McDermott K., Fallone P. S., Hilton S., Siler K., Carmichael N., Daghighian F., Finn R. D., Cohen A. M., et al .
Clinical comparison of radiolocalization of two monoclonal antibodies (mAbs) against the TAG-72 antigen. Nucl. Med. Biol.
,
21
:
9
-15,  
1994
.
49
Mulligan T., Carrasquillo J. A., Chung Y., Milenic D. E., Schlom J., Feuerstein I., Paik C., Perentesis P., Reynolds J., Curt G., et al .
Phase I study of intravenous Lu-labeled CC49 murine monoclonal antibody in patients with advanced adenocarcinoma. Clin. Cancer Res.
,
1
:
1447
-1454,  
1995
.
50
Safavy A., Khazaeli M. B., Safavy K., Mayo M. S., Buchsbaum D. J. Biodistribution study of 188Re-labeled trisuccin-HuCC49 and trisuccin- HuCC49deltaCh2 conjugates in athymic nude mice bearing intraperitoneal colon cancer xenografts.
Clin. Cancer Res.
,
5(Suppl.)
:
2994s
-3000s,  
1999
.
51
Buchsbaum D. J., Khazaeli M. B., Mayo M. S., Roberson P. L. Comparison of multiple bolus and continuous injections of 131I-labeled CC49 for therapy in a colon cancer xenograft model.
Clin. Cancer Res.
,
5(Suppl.)
:
3153s
-3159s,  
1999
.
52
Pavlinkova G., Beresford G., Booth B. J., Batra S. K., Colcher D. Charge-modified single chain antibody constructs of monoclonal antibody CC49: generation, characterization, pharmacokinetics, and biodistribution analysis.
Nucl. Med. Biol.
,
26
:
27
-34,  
1999
.
53
McGregor D. P., Molloy P. E., Cunningham C., Harris W. J. Spontaneous assembly of bivalent single chain antibody fragments in Escherichia coli.
Mol. Immunol.
,
31
:
219
-226,  
1994
.
54
Whitlow M., Filpula D., Rollence M. L., Feng S. L., Wood J. F. Multivalent Fvs: characterization of single-chain Fv oligomers and preparation of a bispecific Fv.
Protein Eng.
,
7
:
1017
-1026,  
1994
.
55
Wolff E. A., Schreiber G. J., Cosand W. L., Raff H. V. Monoclonal antibody homodimers: enhanced antitumor activity in nude mice.
Cancer Res.
,
53
:
2560
-2565,  
1993
.
56
Kipriyanov S. M., Little M., Kropshofer H., Breitling F., Gotter S., Dubel S. Affinity enhancement of a recombinant antibody: formation of complexes with multiple valency by a single-chain Fv fragment-core streptavidin fusion.
Protein Eng.
,
9
:
203
-211,  
1996
.
57
Muller K. M., Arndt K. M., Pluckthun A. Model and simulation of multivalent binding to fixed ligands.
Anal. Biochem.
,
261
:
149
-158,  
1998
.
58
Mattes M. J. Binding parameters of antibodies reacting with multivalent antigens: functional affinity or pseudo-affinity.
J. Immunol. Methods
,
202
:
97
-101,  
1997
.
59
Pavlinkova G., Colcher D., Booth B. J. M., Goel A., Batra S. K. Pharmacokinetics and biodistribution of a light chain shuffled CC49 single-chain Fv antibody construct.
Cancer Immunol. Immunother.
,
49
:
267
-275,  
2000
.
60
Crothers D. M., Metzger H. The influence of polyvalency on the binding properties of antibodies.
Immunochemistry
,
9
:
341
-357,  
1972
.
61
Kaufman E. N., Jain R. K. Effect of bivalent interaction upon apparent antibody affinity: experimental confirmation of theory using fluorescence photobleaching and implications for antibody binding assays.
Cancer Res.
,
52
:
4157
-4167,  
1992
.
62
Hu S., Shively L., Raubitschek A., Sherman M., Williams L. E., Wong J. Y., Shively J. E., Wu A. M. Minibody: a novel engineered anti-carcinoembryonic antigen antibody fragment (single-chain Fv-CH3) which exhibits rapid, high-level targeting of xenografts.
Cancer Res.
,
56
:
3055
-3061,  
1996
.
63
Olafsen T., Rasmussen I. B., Norderhaug L., Bruland O. S., Sandlie I. IgM secretory tailpiece drives multimerization of bivalent scFv fragments in eukaryotic cells.
Immunotechnology (Amsterdam)
,
4
:
141
-153,  
1998
.
64
Viti F., Tarli L., Giovannoni L., Zardi L., Neri D. Increased binding affinity and valence of recombinant antibody fragments lead to improved targeting of tumoral angiogenesis.
Cancer Res.
,
59
:
347
-352,  
1999
.
65
Wolff E. A., Esselstyn J., Maloney G., Raff H. V. Human monoclonal antibody homodimers.
Effect of valency on in vitro and in vivo antibacterial activity. J. Immunol.
,
148
:
2469
-2474,  
1992
.
66
Ghetie M. A., Podar E. M., Ilgen A., Gordon B. E., Uhr J. W., Vitetta E. S. Homodimerization of tumor-reactive monoclonal antibodies markedly increases their ability to induce growth arrest or apoptosis of tumor cells.
Proc. Natl. Acad. Sci. USA
,
94
:
7509
-7514,  
1997
.
67
Casey J. L., Pedley R. B., King D. J., Green A. J., Yarranton G. T., Begent R. H. Dosimetric evaluation and radioimmunotherapy of anti-tumour multivalent Fab’ fragments.
Br. J. Cancer
,
81
:
972
-980,  
1999
.
68
Colcher D., Minelli M. F., Roselli M., Muraro R., Simpson-Milenic D., Schlom J. Radioimmunolocalization of human carcinoma xenografts with B72.
3 second generation monoclonal antibodies. Cancer Res.
,
48
:
4597
-4603,  
1988
.
69
Schlom J., Eggensperger D., Colcher D., Molinolo A., Houchens D., Miller L. S., Hinkle G., Siler K. Therapeutic advantage of high-affinity anticarcinoma radioimmunoconjugates.
Cancer Res.
,
52
:
1067
-1072,  
1992
.
70
Weinstein J. N., Eger R. R., Covell D. G., Black C. D., Mulshine J., Carrasquillo J. A., Larson S. M., Keenan A. M. The pharmacology of monoclonal antibodies.
Ann. NY Acad. Sci.
,
507
:
199
-210,  
1987
.
71
Baxter L. T., Jain R. K. Transport of fluid and macromolecules in tumors.
I. Role of interstitial pressure and convection. Microvasc. Res.
,
37
:
77
-104,  
1989
.