Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Dec 1.
Published in final edited form as: Bioconjug Chem. 2008 Dec;19(12):2527–2534. doi: 10.1021/bc800113v

Site-specific, Thiol-mediated Conjugation of Fluorescent Probes to Cysteine-modified Diabodies Targeting CD20 or HER2

Shannon J Sirk 1,*, Tove Olafsen 1, Bhaswati Barat 1, Karl B Bauer 1, Anna M Wu 1
PMCID: PMC2668938  NIHMSID: NIHMS91453  PMID: 19053310

Abstract

Small, engineered antibody fragments such as diabodies (50 kDa non-covalent dimers of single-chain Fv fragments) are useful alternatives to their larger antibody counterparts. However, due to their size, they are more susceptible to disruption of their antigen binding sites when modified using random conjugation techniques. Previous work has demonstrated the utility of a C-terminal cysteine modification for site-specific radiolabeling of an anti-CEA diabody, resulting in the creation of a cys-diabody (CysDb). In the present work, the adaptability of the CysDb system was explored by creating two additional CysDbs: one specific for CD20 and one for HER2. Purified CysDbs of both specificities demonstrated behavior consistent with stable, covalent dimers harboring a readily reducible disulfide bond. Each CysDb was site-specifically conjugated to three different fluorophores for optical detection: the large fluorescent proteins phycoerythrin (PE) and allophycocyanin (APC), and the small fluorescent molecule Alexa Fluor 488. Fluorophore-conjugated CysDbs bound specifically to their targets in both antigen systems and with each different fluorescent tag as determined by flow cytometry. In vitro specific antigen binding was observed in the presence of a mixture of specific and non-specifically conjugated CysDbs. Conjugates retained both specificity and fluorescence, demonstrating the successful expansion of the CysDb repertoire to new targets and to new site-specific conjugation possibilities.

INTRODUCTION

Monoclonal antibodies (mAbs) comprise a unique class of tumor-targeting agents capable of specific recognition of cell surface antigens, including cancer biomarkers. Many therapeutic mAbs have been used to target and treat a wide range of diseases, including breast cancer, lymphoma, colorectal cancer, and leukemia (1). When conjugated to radionuclides, antibody derivatives specific for cancer biomarkers can be employed for tumor detection using imaging methods such as positron emission tomography (PET) (2–4). MAbs armed with therapeutic radionuclides or other cytotoxic agents have been used to deliver targeted toxicity to a tumor while minimizing damage to healthy tissue (5, 6).

Intact mAbs are highly specific targeting agents but are not suited for every application due to their large size and full Fc domain. Engineered antibody fragments harness the specificity of the variable domains while modifying or deleting the constant domains to suit the intended targeting need (7, 8). Single-chain Fvs (scFvs), consisting of the variable light (VL) and variable heavy (VH) domains joined by an amino acid linker, are the smallest antibody fragments that still retain specificity and are commonly used antigen-specific probes (9–11). ScFv linker length influences fragment multimerization, with shorter linkers (3–12 residues) supporting the formation of bivalent diabodies (Dbs) (12). Compared to scFvs, Dbs demonstrate superior binding in vitro and improved tumor accumulation in vivo due to increased avidity and therefore increased functional affinity (13).

Targeting applications regularly involve labeling antibodies and antibody fragments, often with a radionuclide. Common techniques involve conjugation via random tyrosine or lysine residues (14, 15). Intact antibodies (150 kDa) are large enough to offer several sites that will not disrupt antigen binding when conjugated to a labeling agent. At a fraction of the size of intact antibodies, Dbs (50 kDa) have fewer available conjugation sites and it is therefore more likely that conjugation will disrupt antigen binding or overall conformation and efficacy of the molecule as a tumor targeting agent.

To address this problem, methods for site-specifically labeling Dbs and bypassing potential adverse effects of random conjugation have been pursued. The addition of C-terminal cysteinyl peptides was originally explored as a method to induce dimerization of scFv molecules (16, 17) and provide a defined site for radiometal chelation (18). Site-specific PEGylation has been used to adjust the circulating half life of cysteine-modified scFvs (19) and liposomes have been coupled via thiol chemistry to scFvs targeting tumor biomarkers and tumor stromal cells (20, 21).

Crystallographic analysis of an anti-carcinoembryonic antigen (CEA) Db revealed that the antigen binding sites are located on opposite sides from the C-termini of the VH domains (22). An anti-CEA Db with a cysteine residue incorporated at the C-terminus of the VH domain of each scFv monomer was created (CysDb) and expressed in mammalian cell culture, spontaneously yielding stable covalent dimers (23). Anti-CEA CysDbs demonstrated excellent binding in vitro and, following site-specific chelate conjugation and radiometal labeling, specifically targeted CEA-expressing tumor xenografts in a microPET mouse model (23).

Based on this initial success, the CysDb format was extended to two additional cancer targets – CD20 (lymphoma) and HER2 (breast cancer). CD20 and HER2 were chosen as candidate antigen systems because there are pre-existing, well-characterized therapeutic antibodies (rituximab and trasuzumab) specific for these targets. Functional CysDbs of both specificities were produced and characterized. Both CysDbs retained specific antigen binding following site-specific conjugation to three different fluorophores. Bifunctionality of each conjugate is demonstrated through the observed antigen specificity of each CysDb as well as the unique fluorescence profile of each attached fluorophore. Successful expansion of the C-terminal cysteine modification to two new specificities supports the utility of the CysDb format as a generalizable platform for site-specific conjugation of tumor-targeting molecules with functional payloads.

EXPERIMENTAL PROCEDURES

Design of cysteine-modified diabodies

Diabodies were constructed from existing gene constructs containing variable regions recognizing HER2 (4) or CD20 (Olafsen et al. [abstr.], J. Nucl. Med. 47:48P, 2006). Overlap-extension PCR and site-directed mutagenesis (QuikChange XL, Stratagene) were used to create CysDb genes with a 5-amino acid linker (SGGGG) between the variable light (VL) and variable heavy (VH) domains and a C-terminal cysteine modification (GGC) to give the final CysDb construct: (VL – SGGGG – VH – GGC) in the pEE12 mammalian expression vector (Lonza Biologics). All constructs were confirmed by DNA sequencing.

Mammalian expression, selection, and purification

2.5 × 106 NS0 mouse myeloma cells (ECACC #85110503) were transfected by electroporation with 10μg linearized plasmid DNA and selected in glutamine-deficient DMEM (Mediatech Inc.) as described (24). Transfectants were screened for CysDb expression by assaying cell culture supernatant by ELISA and/or SDS-PAGE using pre-cast 4–20% gels (Bio-Rad).

ELISA

For anti-CD20 CysDb ELISA, microtiter plate wells were coated with 100μL (2μg/mL) goat-anti-mouse Fab (Sigma), then blocked with 1%BSA. 100μL cell culture supernatant was assayed and anti-CD20 CysDb was detected with 100μL (0.4μg/mL) alkaline phosphatase-conjugated goat-anti-mouse Fab (Jackson Immunolabs). Signal was developed using phosphatase substrate tablets (Sigma) and absorbance was detected at 405nm using a microplate reader (Tecan). The same procedure was performed for anti-HER2 CysDb ELISA, except using Protein L (Sigma) for coating wells and Protein L-horse radish peroxidase (HRP) (Sigma) for detection. Signal was developed with CN/DAB Substrate Kit (Pierce) and analyzed visually for black precipitate.

Gel electrophoresis and immunoblotting

SDS-PAGE gels were run with 30μL cell culture supernatant and either stained with Microwave Blue (Protiga), or transferred to nitrocellulose membrane for Western analysis. Samples were run under both reducing (100mM DTT) and non-reducing conditions. Anti-CD20 CysDb was detected on Western blot with 0.4μg/mL goat anti-mouse Fab-AP (Jackson Immunolabs) and color was developed with NBT/BCIP substrate (Promega). Anti-HER2 CysDb was detected on Western blot by Protein L-HRP (Sigma) and color was developed with CN/DAB Substrate Kit (Pierce).

Purification

Clones showing the highest expression were expanded into triple flasks (Corning), and approximately 700mL cell culture supernatant was collected and filtered on a polyethersulfone (PES) membrane (Corning). Supernatant was loaded onto a Protein L column (Pierce) with a bed volume of 1mL at a flow rate of 0.4mL/min in PBS using an ÄKTA Purifier (GE Healthcare Life Sciences). Bound protein was eluted with 0.1M glycine, pH 2.5 over 20 column volumes. Eluted fractions were neutralized upon collection in 5% final volume 2M Tris, pH 8.0. Fractions containing protein were identified by SDS-PAGE and combined. Buffer was exchanged to PBS and protein was concentrated using a spin column (Vivaspin 20, 10 kDa cut-off), and protein was stored at 4°C.

Characterization of purified CysDbs

Purified, concentrated protein was analyzed by SDS-PAGE followed by Microwave Blue gel staining or Western blot analysis as described above. All molecular weights were approximated by comparison to Kaleidoscope pre-stained molecular weight standard (Bio-Rad) and a previously characterized anti-CEA CysDb with a known molecular weight of 55 kDa. CysDbs were also subjected to size-exclusion chromatography on a Superdex 75 column (GE Healthcare Life Sciences) in PBS at a flow rate of 0.5mL/min over one column volume (24mL) on an ÄKTA Purifier. Protein purity was estimated from size exclusion chromatography data. Anti-CD20 CysDb (only) was analyzed for specific binding by indirect flow cytometry, described below.

Site-specific fluorescent labeling

Cys-diabodies were site-specifically conjugated using thiol chemistry to three different fluorophores: phycoerythrin activated with succinimidyl 4-[N-maleimidomethyl]-cyclohexane-1-carboxylate (PE-SMCC, 240 kDa, Prozyme), allophycocyanin (APC)-SMCC (104 kDa, Prozyme), or Alexa Fluor 488 C5-maleimide (A488, 720 Da, Invitrogen). Each CysDb was conjugated to each fluorophore individually, resulting in six conjugates (3 fluorescent tags × 2 CysDb specificities). Mock conjugation controls were treated under the same conditions, except with no CysDb added. Following the Phycolink conjugation kit guidelines (Prozyme), 150μg of CysDb in 60μL PBS was incubated at room temperature with 16mM DTT for 30 minutes to reduce disulfide bonds. DTT was removed from reduction reactions using a Micro Bio-Spin 6 chromatography column (Bio-Rad), equilibrated with Exchange Buffer (50mM MES, 2mM EDTA, pH 6.0). Reduced CysDb was then incubated with 240μg activated PE or APC (10mg/mL in 50mM MES, 2mM EDTA, pH 6.0, with 1μg/mL pentachlorophenol [preservative]), or 12μg A488 (1μg/mL, 1.4mM in water). Molar ratios in the conjugation reactions were approximately 1:3, 1:1.25, and 5.6:1 for PE:CysDb, APC:CysDb, and A488:CysDb, respectively. Addition of fluorophore was followed by incubation in the dark at room temperature for one hour with occasional mixing. Remaining free sulfhydryls were then blocked with 5.1μg N-ethylmaleimide (NEM, 10mg/mL in DMSO, Sigma). Reactions were wrapped in foil, incubated at room temperature with occasional mixing for 30 minutes, and then applied to new spin columns equilibrated with Storage Buffer (10mM Tris-HCl, 150mM NaCl, pH 8.2). Fluorophore-conjugated CysDbs were stored in 1.5mL Eppendorf tubes at 4°C, wrapped in aluminum foil. Protein yield and conjugation ratios were determined using a UV/Visible spectrophotomer (Ultrospec 1100 Pro, GE Healthcare Life Sciences) to measure absorption at 280 and at the peak absorption wavelength of each flurophore.

Flow cytometry

Flow cytometry was used to evaluate binding of CysDbs to cell-surface antigen before (anti-CD20 CysDb only) and after site-specific fluorescent labeling. Unconjugated anti-CD20 CysDb was tested on 38C13 mouse myeloma cells (ATCC #HB-132) transfected with human CD20 (25). Conjugated CysDbs were tested on SKW 6.4 (human B cell lymphoma, CD20-positive, ATCC #TIB-215), Daudi (human B cell lymphoma, CD20-positive, ATCC #CCL-213), and MCF-7 HER2+ (human breast cancer, HER2-transfected, gift of Dr. Dennis Slamon, UCLA School of Medicine) (26). Cells were harvested and resuspended in PBS/1%FBS to a final concentration of 1×106 cells/mL. For each reaction, 5×105 cells (500 μL) were incubated with fluorophore-conjugated CysDb (0.5 μL or 5 μL) for one hour on ice with occasional mixing.

38C13-CD20+ cells were incubated first with unconjugated anti-CD20 CysDb, followed by PE-conjugated anti-mouse Fab secondary antibody (Jackson Immunolabs). SKW 6.4 cells were incubated with PE-conjugated anti-CD20 CysDb only.

Each fluorescent CysDb and each mock conjugate was incubated in separate reactions with Daudi cells and MCF7-HER2+ cells. Experiments were performed with 0.5μL (low concentration sample) or 5μL (high concentration sample) of conjugated CysDb or mock-conjugated dye. Each fluorescent CysDb was also assayed for specific binding to either Daudi (anti-CD20 CysDbs) or MCF7-HER2+ (anti-HER2 CysDbs) in the presence of other CysDbs. Mixtures were composed of one antigen-specific and two antigen-non-specific fluorescent conjugates, with every combination represented. Each fluorescent CysDb (3μL) was incubated with its target cell type in the presence of the other two (non-specific) CysDb (3μL each conjugate). Positive control samples were incubated with either anti-CD20 scFvFc (27) for Daudi, or anti-HER2 scFvFc (4) for MCF7-HER2+ cells, and then washed and incubated with each of three fluorescently-labeled anti-human-IgG-Fc-specific secondary antibodies (PE- and APC-conjugated, Jackson Immunolabs; Alexafluor 488-conjugated, Molecular Probes). All samples were then washed with PBS/1%FBS, resuspended in 500μL PBS/1%FBS, filtered, and analyzed on a modified FACScan flow cytometer using CellQuest software (BD Biosciences). Fluorescence of 10,000 gated viable cells was collected simultaneously in 3 channels for each sample. A488 signal (519nm emission wavelength) was detected in FL1, PE signal (578nm emission wavelength) was detected in FL2, and APC signal (661nm emission wavelength) was detected in FL4.

RESULTS

Design and construction of CysDbs

C-terminal GGC modifications were added to anti-HER2 and anti-CD20 scFv gene constructs as shown (Fig 1). Anti-CD20 CysDb was constructed from an existing anti-CD20 single chain variable fragment (scFv, VL-SGGGG-VH), composed of rituximab (Rituxan™, Genentech) mouse variable regions (27). Anti-HER2 CysDb was constructed from an existing minibody, composed of two trastuzumab (Herceptin™, Genentech) humanized scFvs (28) linked to the CH3 domain of human IgG1 (4). The scFv orientation and linker of the anti-HER2 minibody were as follows: VL-GSTSGGGSGGGSGGGGSS-VH. Overlap extension PCR was used to shorten the 18 amino-acid-linker in the anti-HER2 scFv gene with a 5 amino-acid-linker (SGGGG).

Figure 1.

Figure 1

Open in a new tab

Schematic representation of CysDb. (A) Gene assembly for anti-CD20 and anti-HER2 CysDb (scFv monomer) with 5-aa linker and cysteine modification shown. (B) CysDb with antigen binding sites indicated by starbursts. CysDb is shown in covalent dimer form, with disulfide bond in the predicted location. (C) Conjugation reaction linking cys-diabody to fluorescent probe.

Mammalian expression

Following transfection of NS0 cells with either anti-CD20 or anti-HER2 CysDb gene construct, positive clones were identified by growth of single colonies in microtiter plates. CysDb-expressing transfectants were screened by ELISA and Western blot (data not shown). Color development verified positive clones by ELISA and allowed ranking of relative expression levels within this group. Bands on Western blots were detected at approximately 50 kDa (non-reduced) and 25 kDa (reduced), providing confirmation of the correct size of both dimer and monomeric subunit forms of the protein.

Purification

Clones expressing anti-CD20 CysDb and clones expressing anti-HER2 CysDb were expanded and CysDb was purified from the media by affinity chromatography on a Protein L column. Final protein yields were 1.6 mg/L of culture supernatant for anti-CD20 CysDb, and 3.0 mg/L of culture supernatant for anti-HER2 CysDb. Purities of >99% for anti-CD20 CysDb and >90% for anti-HER2 CysDb were determined from size exclusion chromatography data.

Characterization of purified CysDbs

Gel electrophoresis and immunoblotting

SDS-PAGE analysis demonstrated correct size and covalent dimerization of purified CysDbs. Under non-reducing conditions, both proteins migrated near the expected molecular weights of approximately 50 kDa, verifying a population of covalently-dimerized molecules in the final purified product. Under reducing conditions, CysDbs migrated at approximately 25 kDa, consistent with the expected scFv monomer (Fig 2). Purified CysDbs were specifically detected by Western blot at approximately 50 kDa (non-reduced) and approximately 25 kDa (reduced) (data not shown).

Figure 2.

Figure 2

Open in a new tab

SDS-PAGE of three CysDbs under non-reducing (left) and reducing (right) conditions. Previously characterized anti-CEA CysDb shown for comparison purposes. Non-reduced samples migrate at expected size for covalent dimer (50 kDa) and reduced samples migrate at expected size for monomeric scFv (25 kDa).

Size exclusion chromatography

Characterization of purified CysDbs by size-exclusion chromatography resulted in elution profiles similar to each other and to previously characterized diabodies (Fig 3). Anti-CD20 CysDb (25.2 min) and anti-HER2 CysDb (22.5 min) both eluted at times similar to the previously characterized anti-CEA CysDb (21.5 min) (23), and anti-CD20 Db (23.7 min) (T. Olafsen, unpublished). Anti-CD20 CysDb eluted slightly later than its parental, non-cysteine-modified diabody, suggesting a more compact form of the protein, as expected with covalent dimerization (23). Anti-CD20 CysDb eluted later than either of the other CysDbs, consistent with its apparent smaller size as seen by SDS-PAGE (Fig 2).

Figure 3.

Figure 3

Open in a new tab

Size exclusion chromatography of cys-diabodies on Superdex 75 column. Previously characterized anti-CEA CysDb shown for comparison purposes.

Flow cytometry

Specific antigen binding was demonstrated for anti-CD20 CysDb by indirect flow cytometry. Fluorescence of gated viable CD20+ cells showed binding comparable to that of parental, non-cysteine-modified anti-CD20 diabody (data not shown).

Site-specific fluorescent labeling and specific binding of conjugated CysDbs

Single-channel flow cytometry

Anti-CD20 CysDbs site-specifically conjugated to the fluorescent dyes PE, APC, or A488 demonstrated specific binding to CD20-positive Daudi cells and not to CD20-negative MCF7-HER2+ cells (Fig 4). Quantities of APC and PE used were half that suggested in the manufacturer’s guidelines, as this modification yielded comparable conjugation efficiency in our hands and conserved reagent. Product guidelines for A488 recommend 10–20 moles of dye per mole of protein when conjugating thiol-reactive probes (Invitrogen). A final value of 5.6 moles of A488 C5-maleimide per mole of CysDb was used in an attempt to reduce excessive background from unconjugated dye. Dye:protein ratios were calculated for A488 batches only, as this was the only non-protein fluorophore and therefore gave more consistent quantitative results. The molar ratio of A488:CysDb was calculated to be 2:1.

Figure 4.

Figure 4

Open in a new tab

Flow cytometry of single-channel labeling of Daudi cells. Unlabeled cells (UL) shown in solid grey. (A) Three channels of dye-conjugated anti-CD20 CysDb (Conj., thick line) and mock conjugated dye (Mock, dotted line) demonstrating minimal background from non-specific binding of free dye molecules. (B) Three channels of low (0.5μl, dotted line) and high (5μl, thick line) concentrations of dye-conjugated anti-CD20 CysDb demonstrating concentration-dependent shift in mean fluorescence with increasing quantities of conjugated CysDb. (C) Three channels of dye-conjugated anti-CD20 CysDb (Conj., thick line) with minimal binding to CD20-negative MCF7-HER2 cells, demonstrating antigen-specificity of binding.

Single-channel experiments showed binding to Daudi cells of each of the three fluorescent conjugates of anti-CD20 CysDb (Fig 4A). Mock-conjugated fluorophore background showed minimal shift in FL1, FL2 or FL4 compared to the antigen-specific binding shift of the CysDb. PE-labeled anti-CD20 CysDb also demonstrated specific binding to an additional CD20-positive cell line, SKW 6.4, by flow cytometry (data not shown). All three anti-CD20 CysDb conjugates showed a concentration-dependent shift in fluorescence from the low to the high concentration samples, with the APC version exhibiting a higher mean fluorescence in general, across all experiments performed (Fig 4B). All three anti-CD20 CysDb conjugates showed minimal non-specific background binding to CD20-negative MCF7-HER2+ cells (Fig 4C).

Analogous results were seen with PE-, APC- and A488-conjugated anti-HER2 CysDbs. All three conjugates showed specific binding to MCF7-HER2+ cells. Fluorescence shifts were seen in the expected channels for each of the conjugated CysDbs, but not for the mock-conjugated dye (Fig 5A). Cells incubated with a low concentration of PE- and APC-conjugated anti-HER2 CysDb showed very little difference in mean fluorescence intensity compared with cells incubated with a high concentration of these CysDbs (Fig 5B). When incubated with HER2-negative Daudi cells, fluorescently-conjugated anti-HER2 CysDbs showed minimal background binding (Fig 5C). PE-labeled anti-HER2 CysDb also showed no binding to HER2-negative, CD20-positive SKW 6.4 cells by flow cytometry (data not shown).

Figure 5.

Figure 5

Open in a new tab

Flow cytometry of single-channel labeling of MCF7-HER2+ cells. Unlabeled cells (UL) shown in solid grey. (A) Three colors of dye-conjugated anti-HER2 CysDb (Conj., thick line) and mock conjugated dye (Mock, dotted line) demonstrating minimal background from non-specific binding of free dye molecules. (B) Three channels of low (0.5μl, dotted line) and high (5μl, thick line) concentrations of dye-conjugated anti-HER2 CysDb demonstrating concentration-dependent shift in mean fluorescence with increasing quantities of conjugated CysDb. (C) Three channels of dye-conjugated anti-HER2 CysDb (Conj., solid line) with minimal binding to HER2-negative Daudi cells, demonstrating antigen-specificity of binding.

Multipl- channel flow cytometry

Figure 6A shows representative plots of three-channel flow cytometry experiments in which Daudi cells were incubated with a cocktail of three different CysDb conjugates: one anti-CD20 CysDb and two anti-HER2 CysDbs, each conjugated to a different fluorophore. Each of the three anti-CD20 CysDb fluorescent conjugates was tested in combination with two anti-HER2 CysDb fluorescent conjugates. For anti-CD20 CysDb-PE, specific binding to Daudi cells was demonstrated by signal in FL2 (upward shift). Minimal background signal was detected in FL4 or FL1 from either APC- or A488-conjugated anti-HER2 CysDb, respectively, as seen by a lack of rightward shift in fluorescence (Fig 6A). Similar results were seen when APC- or A488-conjugated anti-CD20 CysDb was incubated with Daudi cells in a three-conjugate mix with anti-HER2 CysDb and subjected to flow cytometric analysis in all three fluorescence channels (Fig 6B).

Figure 6.

Figure 6

Open in a new tab

Flow cytometry of 3-channel labeling of Daudi (A and B) and MCF7-HER2 (C and D) cells with antigen-specific and isotype control dye-conjugated CysDbs. (A) Two representative plots from 3-channel experiments demonstrating detection in FL2 of specific binding of PE-conjugated anti-CD20 CysDb with minimal signal in FL1 or FL4 from co-incubated APC- or A488-conjugated anti-HER2 CysDbs, respectively. (B) Bar graph presenting compiled results of all 3-channel labeling experiments on Daudi cells, demonstrating consistent, specific detection of all three dye-conjugated anti-CD20 CysDbs and minimal background from non-specific binding of dye-conjugated anti-HER2 CysDbs. Analogous experiments were performed with dye-conjugated anti-HER2 CysDbs on MCF7-HER2 cells, using dye-conjugated anti-CD20 CysDbs as isotype controls (C, D).

Parallel experiments were run with MCF7-HER2+ cells which showed specific binding of conjugated anti-HER2 CysDb when mixed with anti-CD20 CysDb conjugates in three-conjugate cocktail experiments as described above. Figure 6C shows a representative plot demonstrating specific detection in FL2 of PE-conjugated anti-HER2 CysDb bound to MCF7-HER2+ cells, while minimal background signal of APC- or A488-conjugated anti-CD20 CysDb was seen in FL4 or FL1, respectively. Similar results were seen for the other combinations of fluorescent conjugates. For each sample, nearly 100% of the signal was in the channel specific to the anti-HER2 conjugate being detected, while fluorescence from the non-specific anti-CD20 CysDb was not seen, with the exception of a slightly higher background from the APC-conjugated anti-CD20 CysDb (Fig 6D).

DISCUSSION

Cysteine-modified diabodies were engineered for use as site-specifically modifiable, cancer biomarker-specific antibody fragments. The decision to place the cysteine modification at the C-terminus of the VL-linker-VH construct was based on previous success with a similar modification which generated the T84.66 anti-CEA CysDb (23). The current work extends the CysDb format from CEA to two additional antibodies recoginizing CD20 and HER2, demonstrating that this format is general. Since the C-termini of the diabody subunits are physically opposite from the antigen-binding sites on the CysDb, following reduction, any further modification to the free sulfhydryls should not interfere with binding activity (23). Purified CysDb exists as a stable covalent dimer until subjected to reducing conditions, demonstrating a system with simple and reliable control over the availability of free sulfhydryls for thiol-specific conjugation.

The presence of C-terminal cysteine residues clearly induces efficient disulfide bridge formation and a compact overall conformation. Olafsen et. al. first observed that the cysteine modification and subsequent disulfide bridge formation caused an anti-CEA CysDb to appear more condensed than the parental diabody (23). We also found that for the two diabody/CysDb pairs examined (anti-CD20, anti-HER2), the cysteine-modified version consistently eluted later on size exclusion chromatography compared to the native diabody (data not shown), consistent with a more closed conformation of the covalent dimer. Elution time by size exclusion appears to be an innate property of each CysDb (Fig. 3), and was reproducible among purified batches of the same CysDb. The anti-HER2 CysDb and the anti-CD20 CysDb both have a predicted molecular weight of 50 kDa, although the anti-CD20 CysDb consistently eluted later by size exclusion chromatography (Fig 3). Observed differences in elution times may reflect the sequence difference in the binding sites, as well as the fact that the rituximab framework is murine whereas trastuzumab is humanized. Minor differences in migration were also observed on SDS-PAGE (Fig. 2), indicating that mobility was influenced by primary sequence as well as molecular weight.

Reduction of the CysDb disulfide bridges provided a flexible and reproducible approach for site-specific conjugation. Fluorescent dyes spanning a molecular weight range from sub-kilodalton (A488) to several hundred kilodaltons (PE, APC) were readily coupled to the CD20- and HER2-specific CysDbs to yield six distinct bifunctional conjugates. At the same time, these conjugates retained the ability to bind to antigen, indicating that the internal disulfide bridges in the V regions, characteristic of the conserved immunoglobulin fold, were not disrupted. In addition, spectrophotometric quantitation of the molar conjugation ratio indicated ~ 2 dye molecules per diabody, consistent with conjugation of fluorophores only onto the two most easily accessible sulfhydryls, at the C-termini.

Fluorescent antibody fragments are regularly used for in vitro detection applications and in vivo studies in small animals for non-invasive monitoring and radiation-free imaging (29). Multi-channel flow cytometry experiments in the current work demonstrated antigen-specific binding with the use of a cocktail of fluorescent CysDbs with a spectrum of labels, supporting this format as a robust and easily interchangeable targeting molecule. Increasingly, optically-labeled antibody fragments are being used for fluorescence-guided surgery and intraoperative visualization of metastastes (30). CysDbs are promising for this application, as their small size should allow them to penetrate a tumor more deeply than a larger, intact antibody. Optical detection and diagnosis using a cocktail of CysDbs recognizing a set different tumor biomarkers could provide a means for simultaneous evaluation of multiple proteins, subject to the known limitations of detection of visible light in tissues and living organisms.

The CysDb format provides a general approach for combining minimal antibody binding domains with a variety of functional moieties. Previously, the cysteine modification was used for site-specific conjugation of the macrocyclic metal chelator DOTA(1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid), enabling radiolabelling of the anti-CEA CysDb with 64Cu for microPET detection of CEA-expressing LS174T xenografts in mice (23). In addition to radiolabeling, site-specific PEGylation of cysteine-modified scFvs or Dbs has been used to modify the pharmacokinetics of these small antibody formats. (31). The incremental addition of several kilodaltons of PEG increases serum residence time in a predictable, size-dependent manner (32). The ability of the diabodies to accept large payloads is not limited to inert substances such as PEG. Venisnik et. al. fused bioluminescent enzymes to an anti-CEA diabody, demonstrating the ability of the diabody to deliver the Renilla luciferase and Gaussia luciferase to a CEA-expressing tumor in vivo (33, 34). The success of this fusion protein approach confirms that relatively large functional groups can be linked to the C-termini of diabodies while maintaining binding and targeting capabilities. However, the fusion protein strategy lacks the flexibility of the CysDb approach, as each individual combination requires engineering and expression of a distinct construct.

CysDbs are especially attractive for nanotechnology applications, due to their overall size (diabody dimensions are approximately 5 × 7 nm) (22), bivalency, and capability for oriented coupling. Nanoparticles, including quantum dots, iron oxide particles, dendrimers, liposomes, and the like, are emerging as a distinct class of reagents with significant potential for in vivo applications (21, 35, 36). Several nanoparticles have been developed for use in imaging and therapy, but a general challenge in the field remains the need for targeted in vivo delivery (37). Oriented conjugation using antigen-specific CysDbs may provide one solution. For example, CysDbs have recently been conjugated to quantum dots, demonstrating specific binding by flow cytomety and confocal microscopy (Barat, B., Sirk, S., et. al, submitted) Site-specific thiol coupling of CysDbs to nanoparticles or nanosensor substrates would automatically orient the antigen-binding domains outward to enable interaction with molecules of interest. Furthermore, attachment of CysDbs to nanodevices or nanowires should also shorten the overall interaction distances compared to coupling of intact antibodies (38, 39).

These studies show the utility of the CysDb platform as a simple C-terminal modification with the potential to be applied to any existing scFv or or diabody. Six fluorescent antibody fragments were produced, demonstrating the versatility and reproducibility of the system. CysDbs are promising as targeting agents for in vivo delivery due to their small size and retention of bivalent binding. CysDbs represent a robust and simple method for reliable, site-specific, and oriented conjugation of tumor-targeting proteins to cargo of a range of size and function.

Acknowledgments

This work was supported by the Research Training in Pharmacological Sciences Training Grant (NIH T32 GM008652), and National Cancer Institute grants P50 CA107399, U54 CA119367, R01 EB000312, and P50086306. Flow cytometry was performed in the UCLA Jonsson Comprehensive Cancer Center and Center for AIDS Research Flow Cytometry Core Facility (NIH CA16042 and AI28697). A. W. is a member of the Jonsson Comprehensive Cancer Center (CA16042). The authors thank Roland Remenyi for experimental contributions and Katelyn McCabe for careful review of the manuscript.

LITERATURE CITED

  • 1.Adams GP, Weiner LM. Monoclonal antibody therapy of cancer. Nat Biotechnol. 2005;23:1147–57. doi: 10.1038/nbt1137. [DOI] [PubMed] [Google Scholar]
  • 2.Robinson MK, Doss M, Shaller C, Narayanan D, Marks JD, Adler LP, Gonzalez Trotter DE, Adams GP. Quantitative immuno-positron emission tomography imaging of HER2-positive tumor xenografts with an iodine-124 labeled anti-HER2 diabody. Cancer Res. 2005;65:1471–8. doi: 10.1158/0008-5472.CAN-04-2008. [DOI] [PubMed] [Google Scholar]
  • 3.Cai W, Olafsen T, Zhang X, Cao Q, Gambhir SS, Williams LE, Wu AM, Chen X. PET imaging of colorectal cancer in xenograft-bearing mice by use of an 18F-labeled T84.66 anti-carcinoembryonic antigen diabody. J Nucl Med. 2007;48:304–10. [PubMed] [Google Scholar]
  • 4.Olafsen T, Kenanova VE, Sundaresan G, Anderson AL, Crow D, Yazaki PJ, Li L, Press MF, Gambhir SS, Williams LE, Wong JY, Raubitschek AA, Shively JE, Wu AM. Optimizing radiolabeled engineered anti-p185HER2 antibody fragments for in vivo imaging. Cancer Res. 2005;65:5907–16. doi: 10.1158/0008-5472.CAN-04-4472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Sharkey RM, Goldenberg DM. Targeted therapy of cancer: new prospects for antibodies and immunoconjugates. CA Cancer J Clin. 2006;56:226–43. doi: 10.3322/canjclin.56.4.226. [DOI] [PubMed] [Google Scholar]
  • 6.Wu AM, Senter PD. Arming antibodies: prospects and challenges for immunoconjugates. Nat Biotechnol. 2005;23:1137–46. doi: 10.1038/nbt1141. [DOI] [PubMed] [Google Scholar]
  • 7.Wu AM, Yazaki PJ. Designer genes: recombinant antibody fragments for biological imaging. Q J Nucl Med. 2000;44:268–83. [PubMed] [Google Scholar]
  • 8.Olafsen T, Kenanova VE, Wu AM. Tunable pharmacokinetics: modifying the in vivo half-life of antibodies by directed mutagenesis of the Fc fragment. Nat Protoc. 2006;1:2048–60. doi: 10.1038/nprot.2006.322. [DOI] [PubMed] [Google Scholar]
  • 9.Albrecht H, Denardo GL, Denardo SJ. Development of anti-MUC1 di-scFvs for molecular targeting of epithelial cancers, such as breast and prostate cancers. Q J Nucl Med Mol Imaging. 2007;51:304–13. [PubMed] [Google Scholar]
  • 10.Huston JS, Levinson D, Mudgett-Hunter M, Tai MS, Novotny J, Margolies MN, Ridge RJ, Bruccoleri RE, Haber E, Crea R, et al. Protein engineering of antibody binding sites: recovery of specific activity in an anti-digoxin single-chain Fv analogue produced in Escherichia coli. Proc Natl Acad Sci U S A. 1988;85:5879–83. doi: 10.1073/pnas.85.16.5879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Uchino J, Takayama K, Harada A, Sone T, Harada T, Curiel DT, Kuroki M, Nakanishi Y. Tumor targeting carboxylesterase fused with anti-CEA scFv improve the anticancer effect with a less toxic dose of irinotecan. Cancer Gene Ther. 2008;15:94–100. doi: 10.1038/sj.cgt.7701100. [DOI] [PubMed] [Google Scholar]
  • 12.Kortt AA, Dolezal O, Power BE, Hudson PJ. Dimeric and trimeric antibodies: high avidity scFvs for cancer targeting. Biomol Eng. 2001;18:95–108. doi: 10.1016/s1389-0344(01)00090-9. [DOI] [PubMed] [Google Scholar]
  • 13.Adams GP, Tai MS, McCartney JE, Marks JD, Stafford WF, 3rd, Houston LL, Huston JS, Weiner LM. Avidity-mediated enhancement of in vivo tumor targeting by single-chain Fv dimers. Clin Cancer Res. 2006;12:1599–605. doi: 10.1158/1078-0432.CCR-05-2217. [DOI] [PubMed] [Google Scholar]
  • 14.Behr TM, Gotthardt M, Becker W, Behe M. Radioiodination of monoclonal antibodies, proteins and peptides for diagnosis and therapy. A review of standardized, reliable and safe procedures for clinical grade levels kBq to GBq in the Gottingen/Marburg experience. Nuklearmedizin. 2002;41:71–9. [PubMed] [Google Scholar]
  • 15.Nikula TK, Bocchia M, Curcio MJ, Sgouros G, Ma Y, Finn RD, Scheinberg DA. Impact of the high tyrosine fraction in complementarity determining regions: measured and predicted effects of radioiodination on IgG immunoreactivity. Mol Immunol. 1995;32:865–72. doi: 10.1016/0161-5890(95)00052-g. [DOI] [PubMed] [Google Scholar]
  • 16.McCartney JE, Tai MS, Hudziak RM, Adams GP, Weiner LM, Jin D, Stafford WF, 3rd, Liu S, Bookman MA, Laminet AA, et al. Engineering disulfide-linked single-chain Fv dimers [(sFv′)2] with improved solution and targeting properties: anti-digoxin 26–10 (sFv′)2 and anti-c-erbB-2 741F8 (sFv′)2 made by protein folding and bonded through C-terminal cysteinyl peptides. Protein Eng. 1995;8:301–14. doi: 10.1093/protein/8.3.301. [DOI] [PubMed] [Google Scholar]
  • 17.Adams GP, McCartney JE, Tai MS, Oppermann H, Huston JS, Stafford WF, 3rd, Bookman MA, Fand I, Houston LL, Weiner LM. Highly specific in vivo tumor targeting by monovalent and divalent forms of 741F8 anti-c-erbB-2 single-chain Fv. Cancer Res. 1993;53:4026–34. [PubMed] [Google Scholar]
  • 18.George AJ, Jamar F, Tai MS, Heelan BT, Adams GP, McCartney JE, Houston LL, Weiner LM, Oppermann H, Peters AM. Radiometal labeling of recombinant proteins by a genetically engineered minimal chelation site: technetium-99m coordination by single-chain Fv antibody fusion proteins through a C-terminal cysteinyl peptide. Proc Natl Acad Sci U S A. 1995;92:8358–62. doi: 10.1073/pnas.92.18.8358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Yang K, Basu A, Wang M, Chintala R, Hsieh MC, Liu S, Hua J, Zhang Z, Zhou J, Li M, Phyu H, Petti G, Mendez M, Janjua H, Peng P, Longley C, Borowski V, Mehlig M, Filpula D. Tailoring structure-function and pharmacokinetic properties of single-chain Fv proteins by site-specific PEGylation. Protein Eng. 2003;16:761–70. doi: 10.1093/protein/gzg093. [DOI] [PubMed] [Google Scholar]
  • 20.Weng KC, Noble CO, Papahadjopoulos-Sternberg B, Chen FF, Drummond DC, Kirpotin DB, Wang D, Hom YK, Hann B, Park JW. Targeted tumor cell internalization and imaging of multifunctional quantum dot-conjugated immunoliposomes in vitro and in vivo. Nano Lett. 2008;8:2851–7. doi: 10.1021/nl801488u. [DOI] [PubMed] [Google Scholar]
  • 21.Messerschmidt SK, Kolbe A, Muller D, Knoll M, Pleiss J, Kontermann RE. Novel single-chain Fv′ formats for the generation of immunoliposomes by site-directed coupling. Bioconjug Chem. 2008;19:362–9. doi: 10.1021/bc700349k. [DOI] [PubMed] [Google Scholar]
  • 22.Carmichael JA, Power BE, Garrett TP, Yazaki PJ, Shively JE, Raubischek AA, Wu AM, Hudson PJ. The crystal structure of an anti-CEA scFv diabody assembled from T84.66 scFvs in V(L)-to-V(H) orientation: implications for diabody flexibility. J Mol Biol. 2003;326:341–51. doi: 10.1016/s0022-2836(02)01428-6. [DOI] [PubMed] [Google Scholar]
  • 23.Olafsen T, Cheung CW, Yazaki PJ, Li L, Sundaresan G, Gambhir SS, Sherman MA, Williams LE, Shively JE, Raubitschek AA, Wu AM. Covalent disulfide-linked anti-CEA diabody allows site-specific conjugation and radiolabeling for tumor targeting applications. Protein Eng Des Sel. 2004;17:21–7. doi: 10.1093/protein/gzh009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Yazaki PJ, Shively L, Clark C, Cheung CW, Le W, Szpikowska B, Shively JE, Raubitschek AA, Wu AM. Mammalian expression and hollow fiber bioreactor production of recombinant anti-CEA diabody and minibody for clinical applications. J Immunol Methods. 2001;253:195–208. doi: 10.1016/s0022-1759(01)00388-x. [DOI] [PubMed] [Google Scholar]
  • 25.Golay J, Cittera E, Di Gaetano N, Manganini M, Mosca M, Nebuloni M, van Rooijen N, Vago L, Introna M. The role of complement in the therapeutic activity of rituximab in a murine B lymphoma model homing in lymph nodes. Haematologica. 2006;91:176–83. [PubMed] [Google Scholar]
  • 26.Pietras RJ, Arboleda J, Reese DM, Wongvipat N, Pegram MD, Ramos L, Gorman CM, Parker MG, Sliwkowski MX, Slamon DJ. HER-2 tyrosine kinase pathway targets estrogen receptor and promotes hormone-independent growth in human breast cancer cells. Oncogene. 1995;10:2435–46. [PubMed] [Google Scholar]
  • 27.Wu AM, Tan GJ, Sherman MA, Clarke P, Olafsen T, Forman SJ, Raubitschek AA. Multimerization of a chimeric anti-CD20 single-chain Fv-Fc fusion protein is mediated through variable domain exchange. Protein Eng. 2001;14:1025–33. doi: 10.1093/protein/14.12.1025. [DOI] [PubMed] [Google Scholar]
  • 28.Carter P, Presta L, Gorman CM, Ridgway JB, Henner D, Wong WL, Rowland AM, Kotts C, Carver ME, Shepard HM. Humanization of an anti-p185HER2 antibody for human cancer therapy. Proc Natl Acad Sci U S A. 1992;89:4285–9. doi: 10.1073/pnas.89.10.4285. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Barrett T, Koyama Y, Hama Y, Ravizzini G, Shin IS, Jang BS, Paik CH, Urano Y, Choyke PL, Kobayashi H. In vivo diagnosis of epidermal growth factor receptor expression using molecular imaging with a cocktail of optically labeled monoclonal antibodies. Clin Cancer Res. 2007;13:6639–48. doi: 10.1158/1078-0432.CCR-07-1119. [DOI] [PubMed] [Google Scholar]
  • 30.Koyama Y, Hama Y, Urano Y, Nguyen DM, Choyke PL, Kobayashi H. Spectral fluorescence molecular imaging of lung metastases targeting HER2/neu. Clin Cancer Res. 2007;13:2936–45. doi: 10.1158/1078-0432.CCR-06-2240. [DOI] [PubMed] [Google Scholar]
  • 31.Natarajan A, Xiong CY, Albrecht H, DeNardo GL, DeNardo SJ. Characterization of site-specific ScFv PEGylation for tumor-targeting pharmaceuticals. Bioconjug Chem. 2005;16:113–21. doi: 10.1021/bc0498121. [DOI] [PubMed] [Google Scholar]
  • 32.Lee LS, Conover C, Shi C, Whitlow M, Filpula D. Prolonged circulating lives of single-chain Fv proteins conjugated with polyethylene glycol: a comparison of conjugation chemistries and compounds. Bioconjug Chem. 1999;10:973–81. doi: 10.1021/bc990076o. [DOI] [PubMed] [Google Scholar]
  • 33.Venisnik KM, Olafsen T, Loening AM, Iyer M, Gambhir SS, Wu AM. Bifunctional antibody-Renilla luciferase fusion protein for in vivo optical detection of tumors. Protein Eng Des Sel. 2006;19:453–60. doi: 10.1093/protein/gzl030. [DOI] [PubMed] [Google Scholar]
  • 34.Venisnik KM, Olafsen T, Gambhir SS, Wu AM. Fusion of Gaussia luciferase to an engineered anti-carcinoembryonic antigen (CEA) antibody for in vivo optical imaging. Mol Imaging Biol. 2007;9:267–77. doi: 10.1007/s11307-007-0101-8. [DOI] [PubMed] [Google Scholar]
  • 35.Michalet X, Pinaud FF, Bentolila LA, Tsay JM, Doose S, Li JJ, Sundaresan G, Wu AM, Gambhir SS, Weiss S. Quantum dots for live cells, in vivo imaging, and diagnostics. Science. 2005;307:538–44. doi: 10.1126/science.1104274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Alexis F, Rhee JW, Richie JP, Radovic-Moreno AF, Langer R, Farokhzad OC. New frontiers in nanotechnology for cancer treatment. Urol Oncol. 2008;26:74–85. doi: 10.1016/j.urolonc.2007.03.017. [DOI] [PubMed] [Google Scholar]
  • 37.Kam NW, O’Connell M, Wisdom JA, Dai H. Carbon nanotubes as multifunctional biological transporters and near-infrared agents for selective cancer cell destruction. Proc Natl Acad Sci U S A. 2005;102:11600–5. doi: 10.1073/pnas.0502680102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zheng G, Patolsky F, Cui Y, Wang WU, Lieber CM. Multiplexed electrical detection of cancer markers with nanowire sensor arrays. Nat Biotechnol. 2005;23:1294–301. doi: 10.1038/nbt1138. [DOI] [PubMed] [Google Scholar]
  • 39.Gruner G. Carbon nanotube transistors for biosensing applications. Anal Bioanal Chem. 2006;384:322–35. doi: 10.1007/s00216-005-3400-4. [DOI] [PubMed] [Google Scholar]

RESOURCES