In normal epithelia, the epithelial cell adhesion molecule (EpCAM) expression is relatively low and only present at the basolateral cell surface. In contrast, EpCAM is aberrantly overexpressed in various human carcinomas. Therefore, EpCAM is considered to be a highly promising target for antibody-based cancer immunotherapy. Here, we present a new and fully human cytolytic fusion protein (CFP), designated “anti–EpCAM(scFv)-MAP,” that is comprised of an EpCAM-specific antibody fragment (scFv) genetically fused to the microtubule-associated protein tau (MAP). Anti–EpCAM(scFv)-MAP shows potent EpCAM-restricted proapoptotic activity toward rapidly proliferating carcinoma cells. In vitro assays confirmed that treatment with anti–EpCAM(scFv)-MAP resulted in the colocalization and stabilization of microtubules, suggesting that this could be the potential mode of action. Dose-finding experiments indicated that anti–EpCAM(scFv)-MAP is well tolerated in mice. Using noninvasive far-red in vivo imaging in a tumor xenograft mouse model, we further demonstrated that anti–EpCAM(scFv)-MAP inhibited tumor growth in vivo. In conclusion, our data suggest that anti–EpCAM(scFv)-MAP may be of therapeutic value for the targeted elimination of EpCAM+ carcinomas. Mol Cancer Ther; 13(9); 2194–202. ©2014 AACR.

Standard chemo- and radiotherapy remain routine cancer treatment regimens in daily clinical practice. Despite encouraging initial responses, most solid tumors eventually develop resistance, metastasize, and relapse as incurable disease in response to these treatments (1, 2). In addition, unacceptable side effects strongly limit the therapeutic applicability of conventional therapies. Thus, there is an urgent demand for novel therapeutic avenues that are both selective and have high cytotoxic efficacy against malignant tumor cells.

Among all cancers, carcinoma is the most diagnosed cancer type, originating in the skin, lungs, breasts, pancreas, glands, or various other organs. Although each type of carcinoma has its own specific properties, many types often aberrantly overexpress the epithelial cell adhesion molecule (EpCAM; CD326; ref. 3). In fact, the upregulation of this transmembrane glycoprotein has been associated with the development and progression of various carcinomas, including gastrointestinal carcinoma (gall bladder, pancreas; ref. 4, 5), lung carcinoma (squamous cell carcinoma, adenocarcinoma; refs. 6, 7), breast carcinoma (8), gynecological carcinoma (cervix; ref. 9), and urological carcinoma (urothelial bladder, prostate; refs. 10, 11). The tumor-promoting nature of EpCAM has been accredited to its ability to modulate cell proliferation and cell–cell interactions, which consequently results in tissue invasion and metastasis (9, 12).

There are several characteristics of EpCAM+ carcinomas that make this protein an attractive target in cancer diagnosis and therapy. Although it is present in healthy epithelial tissue, the expression of EpCAM is low and the protein localizes solely to the basolateral membrane. In contrast, in carcinomas, EpCAM expression pattern is significantly altered and high levels of the protein can be found uniformly across the entire cell membrane (13). Moreover, EpCAM was shown to be hyperglycosylated in carcinomas versus in healthy epithelia (14). Overexpression of EpCAM has further been associated with poor prognosis and overall survival (15). Together, these differences would allow for the development of unique monoclonal antibodies (mAb) against EpCAM in carcinoma cells that are of high therapeutic value.

Several anti-EpCAM mAbs have been developed showing promising activity in preclinical studies. However, early-phase clinical evaluation of naked anti-EpCAM mAbs in cancer treatment has been disappointing (3, 16–18). The therapeutic effects of unmodified or “naked” mAbs are predominantly dependent on natural effector functions of a fully active human immune response and include complement-dependent cytotoxicity (CDC), antibody-dependent cellular cytotoxicity (ADCC), and antibody-dependent cellular phagocytosis (ADCP). However, in patients with cancer, the immune system may be severely compromised. Moreover, cancer cells frequently develop resistance mechanisms that allow them to evade current mAb-based approaches.

Therefore, advanced antibody-based cancer therapeutics, such as antibody–drug conjugates (ADC) or immunotoxins, which deliver cytotoxic molecules to their specific target, were developed (19). Classic ADCs are conjugates in which a conventional antitumor mAb is chemically linked to a small synthetic cytotoxic drug molecule. The main goal of ADCs is not primarily the recruitment of immune effector cells via their Fc-domain, but rather the induction of cancer cell death after their internalization and the intracellular release of the coupled cytotoxic drug. However, because of their relative bulky size, ADCs have a reduced ability to efficiently penetrate solid tumors, limiting their application to early-stage disease and to nonsolid cancer types (20). In addition, a new generation of chimeric immunotoxins, which is composed of a small tumor-targeting ligand (e.g., growth factors and cytokines) or an antibody fragment [e.g., single-chain fragment variable (scFv)] and a cytotoxic protein, was developed (21). Although there have been encouraging results obtained in preclinical and clinical settings, the use of bacteria- or plant-derived protein toxins, such as Pseudomonas aeruginosa exotoxin A (ETA) or ricin (22, 23), can result in the induction of neutralizing antibodies and dose-limiting side effects, including vascular leak syndrome and liver toxicity (24, 25).

During the last decade, various human proteins have been identified that seemed suitable to be exploited as antitumor effector proteins in immunotherapeutic approaches. Several human enzymes, including the protease granzyme B, RNase angiogenin, or death-associated protein kinase 2, have been demonstrated to induce apoptosis in target cells, when delivered by antigen-specific ligands or antibody fragments (26–29). Similarly, death ligands of the tumor necrosis factor (TNF) family have been genetically fused to anticancer scFvs to selectively induce apoptosis in various forms of cancer (30, 31).

Recently, we reported on the ability of the microtubule-binding protein tau (MAP) to specifically kill epidermal growth factor (EGF) receptor–overexpressing cancer cells, when fused to EGF (32). In contrast with enzymes, MAP activity is not blocked by endogenous inhibitors. Its cytotoxic mechanism of action involves the binding to and stabilizing of spindle microtubules. The dependence of this process on cell proliferation makes frequently and rapidly diving cancer cells the target of choice for MAP-based cytolytic fusion proteins (CFP).

In this study, we present a new human CFP, designated EpCAM(scFv)-MAP, consisting of a fully human anti-EpCAM scFv antibody fragment genetically fused to human MAP. Anti–EpCAM(scFv)-MAP was expressed in Escherichia coli (E. coli) and purified by standard chromatographic methods. Anti–EpCAM(scFv)-MAP potently induced apoptosis in EpCAM+ carcinoma cell lines, including L3.6pl (IC50, 43 nmol/L) and A431 (IC50, 67 nmol/L). The chimeric anti–EpCAM(scFv)-ETA' immunotoxin was used as a positive control. Colocalization and tubulin polymerization assays indicated that stabilization of microtubules could be a possible mode of action of anti–EpCAM(scFv)-MAP. Furthermore, anti–EpCAM(scFv)-MAP was well tolerated when injected in mice. L3.6pl cells transfected with the far-red fluorescent protein Katushka-2 were used in a mouse xenograft model in vivo to confirm that anti–EpCAM(scFv)-MAP significantly inhibited tumor growth.

Here, we present a novel approach for the treatment of EpCAM+ carcinomas combining the binding specificity of a human high-affinity anti-EpCAM scFv antibody fragment with the selective cytotoxicity of human MAP toward proliferating carcinoma cells.

Generation of anti–EpCAM(scFv)-MAP

The open reading frame (ORF) of MAP (Gene ID, 4137; modified as described in ref. 32) was modified with 5′-NotI and 3′-BlpI restriction sites by PCR followed by ligation into a NotI/BlpI-linearized pMT vector. The anti-EpCAM(scFv) sequence (33) was cloned 5′ of MAP via XbaI/NotI. The sequence for anti–EpCAM(scFv)-ETA' was constructed using the same strategy. Expression, purification, and protein analysis was carried out as described previously (32).

Cell culture

EpCAM+ human carcinoma cancer cell lines L3.6pl (34), A431 (ATCC: CRL-1555), C4-2 (kindly provided by Prof. Dr. Elsässer-Beile, University Hospital, Freiburg, Germany), SU86.86 (ATCC: CRL-1837), 22Rv1 (ATCC: CRL-2505), and the EpCAM human embryonic kidney cell line HEK 293T (ATCC: CRL-11268) were grown in RPMI-1640 medium supplemented with 10% (v/v) fetal calf serum, 2 mmol/L l-glutamine, 50 μg/mL penicillin, and 100 μg/mL streptomycin at 37°C, 100% humidity, and 5% CO2. Cell line authentication was carried out by short tandem repeat (STR) analysis. L3.6pl (6/2008), A431 (3/2005), HEK293 (12/2004), 22Rv1 (1/2013), and SU86.86 (10/2001) cells were authenticated at the Fraunhofer Institute IME in Aachen. The cell line C4-2 was not authenticated in-house.

Transfection of L3.6pl cells with Katushka-2

Transfection was carried out as described previously (35). Briefly, we seeded 1 × 105 L3.6pl cells in a 12-well plate and transfected them with 1 μg DNA encoding the far-red fluorescent protein Katushka-2 (pTag-Katushka2-N; Evrogen) using Roti-Fect (Roth) according to the manufacturer's instructions. Transfected cells were selected with 500 ng/mL G418 (Nalgene) and sorted using the FACSVantage Cell Sorter (Becton Dickinson).

Binding analysis

The cell-binding activity of purified anti–EpCAM(scFv)-MAP was analyzed by flow cytometry. We incubated 4 × 105 cells with 1 μmol/L anti–EpCAM(scFv)-MAP in PBS (pH, 7.4) containing 2 mmol/L EDTA and 0.5% (w/v) BSA for 30 minutes on ice followed by washing with PBS. Fluorescence staining was performed using an anti–penta-His-Alexa Fluor 488 antibody (1:100; Qiagen) for 30 minutes on ice in the dark. Finally, the cells were washed twice with PBS and analyzed on a FACSCalibur flow cytometer (Becton Dickinson).

In vitro cytotoxicity assay

The cytotoxic effect of anti–EpCAM(scFv)-MAP was assessed by measuring the conversion of XTT to a water-soluble orange formazan dye. In short, 1 × 105 cells were seeded per well into a 96-well microtiter plate and were incubated with various dilutions of the recombinant protein for 72 hours at 37°C, 5% CO2, and 100% humidity. The EpCAM cell line HEK293 and the chimeric immunotoxin anti–EpCAM(scFv)-ETA', and the SNAP-tagged anti-EpCAM(scFv) were used as a controls. Next, 50 μL XTT/phenanzine methosulfate (100:1; SERVA and Sigma) were added to each well and incubated the plates as above for 3 to 4 hours before measuring the absorbance at 450 and 630 nm using an Epoch Microplate Spectrophotometer (BioTek). The concentration required to achieve 50% reduction of protein synthesis (IC50) relative to untreated control cells was calculated using GraphPad Prism 5 (GraphPad Software). All experiments were carried out in triplicate.

Apoptosis assay

An AnnexinV/propidium iodide assay was used to determine the proapoptotic impact of anti–EpCAM(scFv)-MAP. We incubated 2.5 × 105 cells/mL with 100 nmol/L anti–EpCAM(scFv)-MAP in a 12-well plate (Greiner Bio-One) for 48 hours at 37°C, 5% CO2, and 100% humidity. EGFR HEK293 cells were used as a control. After incubation, the cells were washed twice with PBS (pH 7.4) and stained with AnnexinV-FITC (eBioscience) in AnnexinV binding buffer [10 mmol/L HEPES/NaOH (pH 7.4), 140 mmol/L NaCl, 2.5 mmol/L CaCl2] for 30 minutes at room temperature. Finally, the cells were washed, resuspended in AnnexinV buffer containing 10 μg/mL propidium iodide, and analyzed by flow cytometry using a FACSCalibur flow cytometer (Becton Dickinson).

Tubulin polymerization assay

The stabilizing effect of anti–EpCAM(scFv)-MAP on tubulin polymerization was tested using a Tubulin Polymerization Assay Kit (tebu-bio) according to the manufacturer's instructions. Paclitaxel (10 μmol/L) and general tubulin buffer were used as controls. All measurements were carried out in duplicate. Vmax represents the slope of the linear phase in milli-extinction per minute.

Colocalization assay

To visualize colocalization of MAP with microtubules, EGFR+ A549 cells were transfected with a SNAP-TubB3 DNA construct [amplified by PCR from pCMV6-XL5-βIII-tubulin template (OriGene)] facilitating directed fluorescence labeling of tubulin via the SNAP-tag. Transfected A549SNAP-TubB3 cells were stained with 10 nmol/L SNAP-Cell TMR-Star (New England BioLabs) overnight. Afterward, cells were fixed by 4% (v/v) PFA/PBS (Polysciences Inc.) for 15 minutes at room temperature, permeabilized with 0.1% (v/v) Triton/PBS (Roth), and blocked with 3% (w/v) BSA/PBS. Of note, 1 μmol/L of EGF–MAP (32) or the truncated version EGF–MAPf151 diluted in 1% (w/v) BSA/PBS was added and incubated for 2 hours at room temperature. The primary antibody detecting MAP/MAPf151 was anti-human Tau [1:5,000 in 1% (w/v) BSA/PBS for 2 hours at room temperature (Thermo Fisher Scientific)]. The secondary antibody was goat-anti-mouse Alexa Fluor 647 [1:1,500 in 1% (w/v) BSA/PBS for 1 hour at room temperature (Life Technologies)]. DAPI (1:10,000 in PBS; Invitrogen) was used to counterstain the nucleus. Images were taken using an Opera QEHS 2.0 confocal fluorescence microscope (PerkinElmer) at ×40 magnification.

In vivo efficacy

The local Animal Care and Use Review Committee officially approved the animal experiments. All animals received humane care in accordance with the requirements of the German Tierschutzgesetz, §8 Abs. 1 and with the guide for the care and use of laboratory animals published by the NIH in 2011. Animal experiments were carried out as described previously (35). Briefly, 1 × 106 L3.6pl cells expressing Katushka-2 were suspended in approximately 20 μL of PBS (pH 7.4) and subcutaneously injected into the right hind limb of 6- to 8-week-old female Balb/C nu/nu mice (Charles River Laboratories). For imaging experiments, mice were placed on a purified, chlorophyll-free diet (AIN93G; SSNIFF GmbH) 11 days before the imaging experiments began. The animals were injected intravenously with anti–EpCAM(scFv)-MAP (4 mg/kg) or PBS (pH 7.4) on days 0, 1, 3, and 5. Readouts were taken on the same days as the treatment, using the Maestro CRi optical imaging system (Maestro CRi Inc.). Images were analyzed using the Maestro spectral imaging software as previously described (36). The Katushka-2 signal was detected using the yellow filter set (630–850 nm).

Statistical analysis

Statistical analysis was performed using GraphPad Prism 5 (GraphPad Software). Data were expressed as mean ± SD or SEM as indicated. Statistical comparisons were made using a two-tailed unpaired Student t test or one-way ANOVA, where appropriate; *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

Generation of anti–EpCAM(scFv)-MAP and characterization of binding activity

We previously reported the identification of a new cytolytic effector protein, called MAP, which proved effective in specifically killing proliferating cancer cells, when fused to EGF (32). Here, the complete ORF of MAP was cloned in frame with a 5′-ORF encoding the fully-human anti-EpCAM(scFv) into a bacterial expression vector. Anti–EpCAM(scFv)-MAP was successfully expressed under osmotic stress conditions in the presence of compatible solutes (37), yielding up to 1 mg of purified protein per liter of bacterial culture. After purification by affinity chromatography, specific binding activity to L3.6pl and A431 cells was confirmed by flow cytometry, which also revealed the absence of nonspecific binding to EpCAM HEK293 cells (Fig. 1).

Figure 1.

Cell-binding activity of anti–EpCAM(scFv)-MAP to L3.6pl and A431 cells. Purified anti–EpCAM(scFv)-MAP was tested for its ability to bind EpCAM+ target cells. EpCAM HEK293 cells were used as a negative control. Bound anti–EpCAM(scFv)-MAP was detected with anti–penta-His-Alexa Fluor 488 antibody. relMFI, relative mean fluorescence intensity = MFI stained cells/MFI background signal.

Figure 1.

Cell-binding activity of anti–EpCAM(scFv)-MAP to L3.6pl and A431 cells. Purified anti–EpCAM(scFv)-MAP was tested for its ability to bind EpCAM+ target cells. EpCAM HEK293 cells were used as a negative control. Bound anti–EpCAM(scFv)-MAP was detected with anti–penta-His-Alexa Fluor 488 antibody. relMFI, relative mean fluorescence intensity = MFI stained cells/MFI background signal.

Close modal

Cytotoxicity of anti–EpCAM(scFv)-MAP in vitro

The EpCAM-specific scFv has been shown to be an attractive target for immunotherapy as it induces internalization of the receptor–ligand complex upon binding. Having taken this into consideration (33, 38), the in vitro cytotoxicity of anti–EpCAM(scFv)-MAP was determined using an XTT cell viability assay. Anti–EpCAM(scFv)-MAP showed specific toxicity toward L3.6pl and A431 cells reaching IC50 values of 43 nmol/L and 67 nmol/L, respectively (Fig. 2, top). Although the bacterial ETA is not comparable with MAP with respect to the mode of action, the cytotoxic efficacy of anti–EpCAM(scFv)-MAP was comparable with that obtained for anti–EpCAM(scFv)-ETA' (409 nmol/L on L3.6pl and 91 nmol/L on A431 cells; Fig. 2, middle) that was used as a positive control. The diagnostic tool molecule anti–EpCAM(scFv)-SNAP, which does not contain a cytotoxic load, showed no adverse effects on cell viability (Fig. 2, bottom). In addition, anti–EpCAM(scFv)-MAP bound and was cytotoxic to other cancer cell lines expressing different levels of EpCAM (Supplementary Fig. S1 and Table 1, respectively).

Figure 2.

Dose-dependent cytotoxicity of anti–EpCAM(scFv)-MAP. XTT cell viability assays were carried out using anti–EpCAM(scFv)-MAP against L3.6pl, A431, and HEK293 cells. Anti–EpCAM(scFv)-ETA' and anti–EpCAM(scFv)-SNAP were used as controls. Nonlinear regression and IC50 calculations were performed using GraphPad Prism 5 software (GraphPad Software). Values are presented as mean ± SD of three independent measurements.

Figure 2.

Dose-dependent cytotoxicity of anti–EpCAM(scFv)-MAP. XTT cell viability assays were carried out using anti–EpCAM(scFv)-MAP against L3.6pl, A431, and HEK293 cells. Anti–EpCAM(scFv)-ETA' and anti–EpCAM(scFv)-SNAP were used as controls. Nonlinear regression and IC50 calculations were performed using GraphPad Prism 5 software (GraphPad Software). Values are presented as mean ± SD of three independent measurements.

Close modal
Table 1.

Cytotoxicity of anti–EpCAM(scFv)-MAP to different EpCAM+ cell lines

Cytotoxicity of anti–EpCAM(scFv)-MAP to different EpCAM+ cell lines
Cytotoxicity of anti–EpCAM(scFv)-MAP to different EpCAM+ cell lines

Anti–EpCAM(scFv)-MAP promotes apoptosis

The mode of action of anti–EpCAM(scFv)-MAP was investigated by incubating L3.6pl cells with anti–EpCAM(scFv)-MAP followed by staining with AnnexinV-FITC and propidium iodide. As this assay was performed to demonstrate qualitatively the proapoptotic effect of the anti–EpCAM(scFv)-MAP, the concentration (100 nmol/L ∼ 2 × IC50) and the incubation time (48 hours) were appropriately chosen to capture the cells within the early and late apoptotic stage. Cells treated with equal volume of PBS were used as a control. The population of late apoptotic cells (upper right quadrant) increased strongly, demonstrating the cytotoxicity of anti–EpCAM(scFv)-MAP toward EpCAM+ target cells. In addition to apoptotic cells, the population in the upper left quadrant, representing necrotic cells, increased significantly. There was no induction of apoptosis in HEK293 cells (Fig. 3).

Figure 3.

Cytotoxicity of anti–EpCAM(scFv)-MAP is based on induction of apoptosis. L3.6pl and HEK293 cells were incubated with 100 nmol/L of anti–EpCAM(scFv)-MAP for 48 hours. Detection of late apoptotic cells was carried out using recombinant AnnexinV-FITC and propidium iodide staining. Values are presented as mean ± SD of three independent measurements. Statistical analysis was carried out using one-way ANOVA; ***, P ≤ 0.001.

Figure 3.

Cytotoxicity of anti–EpCAM(scFv)-MAP is based on induction of apoptosis. L3.6pl and HEK293 cells were incubated with 100 nmol/L of anti–EpCAM(scFv)-MAP for 48 hours. Detection of late apoptotic cells was carried out using recombinant AnnexinV-FITC and propidium iodide staining. Values are presented as mean ± SD of three independent measurements. Statistical analysis was carried out using one-way ANOVA; ***, P ≤ 0.001.

Close modal

Anti–EpCAM(scFv)-MAP colocalizes with and stabilizes microtubules

To provide first indications for the mode of action of MAP-based CFPs, we established a cell-based in vitro fluorescence assay, in which colocalization of MAP can be visualized. Therefore, target cells had to be transfected with SNAP-tagged tubulin DNA constructs allowing the specific labeling of tubulin molecules with fluorescent dyes. Unfortunately, EpCAM+ L3.6 and A431 cells are difficult to transfect. Therefore, we used EGFR+ A549 cells in combination with previously described EGF–MAP and a truncated version EGF–MAPf151 to show the representative mode of action for MAP-based CFPs. Figure 4A exemplifies the successful demonstration of MAP/MAPf151-tubulin colocalization (hotspots indicated by arrows). Binding of MAP to tubulin was anticipated to interfere with mitosis by stabilizing microtubules. To determine whether MAP could, in fact, stabilize microtubules and promote polymerization, we carried out a tubulin polymerization assay. Of note, 50 nmol/L (∼IC50) of anti–EpCAM(scFv)-MAP was able to stabilize microtubules effectively by promoting polymerization. Paclitaxel (10 μmol/L; concentration recommended by the manufacturer) was used as a control (Fig. 4B). Combining these two experimental outcomes indicates that MAP-based CFPs are able to bind to and to stabilize microtubules, thus affecting microtubule dynamics within the cell.

Figure 4.

Anti–EpCAM(scFv)-MAP colocalizes with and stabilizes microtubules. A, colocalization assays were performed using EGFR+ A549 cells transfected with a SNAP-tagged tubulin DNA construct. Expressed SNAP-tagged tubulin molecules were labeled with SNAP-Cell TMR-Star (green). EGF–MAP and the truncated version EGF–MAPf151 were used as representatives for MAP-based CFPs. MAP and MAPf151 were detected using mouse anti-human Tau and goat anti-mouse Alexa Fluor 647 antibodies (red). DAPI was used to counterstain the nucleus (blue). Images were taken using an Opera QEHS 2.0 confocal fluorescence microscope at ×40 magnification. Selected colocalization events of MAP/MAPf151 with tubulin are indicated by arrows. B, tubulin polymerization assay using 50 nmol/L (IC50) anti–EpCAM(scFv)-MAP (Ep-MAP), with general tubulin buffer and paclitaxel (10 μmol/L) as controls. Vmax values were calculated using the slope of the linear phase (mE, milli-extinction). Statistical analysis was carried out using a two-tailed unpaired Student t test; ***, P ≤ 0.001 compared with the buffer control.

Figure 4.

Anti–EpCAM(scFv)-MAP colocalizes with and stabilizes microtubules. A, colocalization assays were performed using EGFR+ A549 cells transfected with a SNAP-tagged tubulin DNA construct. Expressed SNAP-tagged tubulin molecules were labeled with SNAP-Cell TMR-Star (green). EGF–MAP and the truncated version EGF–MAPf151 were used as representatives for MAP-based CFPs. MAP and MAPf151 were detected using mouse anti-human Tau and goat anti-mouse Alexa Fluor 647 antibodies (red). DAPI was used to counterstain the nucleus (blue). Images were taken using an Opera QEHS 2.0 confocal fluorescence microscope at ×40 magnification. Selected colocalization events of MAP/MAPf151 with tubulin are indicated by arrows. B, tubulin polymerization assay using 50 nmol/L (IC50) anti–EpCAM(scFv)-MAP (Ep-MAP), with general tubulin buffer and paclitaxel (10 μmol/L) as controls. Vmax values were calculated using the slope of the linear phase (mE, milli-extinction). Statistical analysis was carried out using a two-tailed unpaired Student t test; ***, P ≤ 0.001 compared with the buffer control.

Close modal

Antitumor activity of anti–EpCAM(scFv)-MAP in vivo

Before treatment experiments, anti–EpCAM(scFv)-MAP was tested in tumor-free mice to investigate its off-target toxicity. Three mice per treatment group received 0.5, 1.0, 2.0, or 4.0 mg/kg of anti–EpCAM(scFv)-MAP intravenously on days 0, 1, 3, and 5. The overall health status and body weight of the mice were monitored daily. All mice survived and showed normal behavior. No reduction of body weight was observed (Fig. 5A). Although a maximum tolerated dose could not be determined, our data suggest that anti–EpCAM(scFv)-MAP is well tolerated.

Figure 5.

In vivo performance of anti–EpCAM(scFv)-MAP. A, to determine the maximum tolerated dose of anti–EpCAM(scFv)-MAP in vivo, four different doses (0.5, 1.0, 2.0, and 4.0 mg/kg) were administered to mice (n = 3) on days 0, 1, 3, and 5. Values are presented as means ± SEM. B, comparison of tumor growth between untreated (PBS) and treated (anti–EpCAM(scFv)-MAP) groups. The mean tumor volumes of both groups were normalized to 100 on day 0. Values are presented as means ± SEM (n = 5). Statistical analysis was carried out using a two-tailed unpaired Student t test; *, P ≤ 0.05; **, P ≤ 0.01 compared with the respective PBS control.

Figure 5.

In vivo performance of anti–EpCAM(scFv)-MAP. A, to determine the maximum tolerated dose of anti–EpCAM(scFv)-MAP in vivo, four different doses (0.5, 1.0, 2.0, and 4.0 mg/kg) were administered to mice (n = 3) on days 0, 1, 3, and 5. Values are presented as means ± SEM. B, comparison of tumor growth between untreated (PBS) and treated (anti–EpCAM(scFv)-MAP) groups. The mean tumor volumes of both groups were normalized to 100 on day 0. Values are presented as means ± SEM (n = 5). Statistical analysis was carried out using a two-tailed unpaired Student t test; *, P ≤ 0.05; **, P ≤ 0.01 compared with the respective PBS control.

Close modal

To test the efficacy of anti–EpCAM(scFv)-MAP in vivo, we transfected L3.6pl cells with the far-red fluorescent protein, Katushka-2, and injected 1 × 106 transfected cells subcutaneously into the right hind leg of 10 Balb/C nu/nu mice to induce tumors. We monitored tumor growth using the in vivo imaging system Maestro Cri, based on the premise that the total Katushka-2 signal correlates with cell number. Eleven days after inoculation, all mice showed palpable tumors (Fig. 5B). Mice were then randomly divided into two groups (n = 5) for one treatment cycle with anti–EpCAM(scFv)-MAP (4 mg/kg) or PBS (pH 7.4). Although the tumor volume of PBS-treated mice increased 2.9 fold on day 5 and 3.6 fold on day 7 (both compared with day 0), only 1.3 fold and 1.9-fold volume increase could be determined for mice treated with anti–EpCAM(scFv)-MAP, respectively. Hence, our results clearly demonstrate that anti–EpCAM(scFv)-MAP causes a significant reduction in tumor growth compared with the PBS control group.

In this study, we present a novel MAP-based CFP anti–EpCAM(scFv)-MAP that has potent EpCAM-restricted proapoptotic activity toward carcinoma cells. Its cytotoxic mode of action is provided by the proapoptotic effector protein MAP that selectively affects proliferating cells (32). The effector protein MAP in anti–EpCAM(scFv)-MAP was cytotoxic toward cancer cells in a manner that mechanistically seems to be comparable with that of conventional microtubules modifying cytostatic agents. Binding to and stabilizing of polymerized microtubule filaments by the MAP domain may result in a change in microtubule dynamics, which then leads to cell-cycle arrest and apoptosis (39). Because most tumor markers, including EpCAM, are also expressed on physiologically normal cells, superior selectivity for proliferating cells may reduce undesired off-target toxic effects.

In contrast with synthetic cytostatic drugs, which require chemical conjugation to the targeting component and often lead to a heterogenic protein preparation, recombinant CFPs, such as the new anti–EpCAM(scFv)-MAP, are considered inherently safer. They are produced as an integral whole in a one-step fermentation omitting additional postpurification conjugation procedures. This feature allows for the production of a homogenous therapeutic product. Furthermore, we could successfully express anti–EpCAM(scFv)-MAP in E. coli, pointing to the ability to exploit prokaryotic expression systems, which provides excellent prospects for scaling up production (40).

The chimeric immunotoxin 4D5MOCB-ETA, consisting of a full-length anti-EpCAM antibody chemically coupled to ETA', has been reported to inhibit the growth of squamous cell carcinoma and adenocarcinoma cell lines in the femtomolar range (41). Although systemic administration of 250 μg/kg dose of 4D5MOCB-ETA did not cause any hepatotoxicity in immunocompetent mice, hepatocyte necrosis was induced at a dose of 1 mg/kg, highlighting one of the common disadvantages associated with chimeric immunotoxins, namely dose limitation due to vascular leak syndrome (42). Similar problems have been described for ADCs, which are bifunctional molecules commonly composed of a tumor-specific mAb chemically linked to a synthetic small-molecule drug. In fact, linker chemistry is a key determinant of ADC activity, pharmacokinetics, and therapeutic index; although the linker must be stable in circulation to ensure delivery of the intact ADC to the target antigen, it must become labile upon internalization to allow release of the cytotoxic drug into the cytosol. Although various linker designs are available, most of them suffer from instability in plasma with half-lives of 48 to 72 hours (43). On the other hand, dissociated free cytotoxic agents result in undesirable off-target effects and limit the clinical application of many ADCs (44). To reduce immunogenic potential and off-target effects, proapoptotic human enzymes have been exploited in the context of fully human CFPs. In 2008, an entirely human granzyme B–based CFP directed against CD64 was found to be toxic for an acute myeloid leukemia (AML)–related cell line and primary AML cells with IC50 values ranging between 1.7 and 17 nmol/L (27). The human RNase angiogenin was shown to be specifically cytotoxic toward CD30-overexpressing Hodgkin lymphoma–derived cell lines, when fused to the CD30 ligand (26). Another CD30-targeted CFP was shown to kill CD30-overexpressing tumor cells efficiently when combined with human death-associated protein kinase 2 (28). The cytotoxic efficacy of anti–EpCAM(scFv)-MAP showed promising results at approximately 50 nmol/L, which was comparable or even slightly better than that of fusion protein anti–EpCAM(scFv)-ETA'. In contrast with the bivalent full-length construct 4D5MOCB-ETA, monovalent anti–EpCAM(scFv)-ETA', which we used in our study, by far did not reach femtomolar efficacy toward A431 and L3.6pl cells (IC50 values ranged from 90 to 400 nmol/L). This reflects the cytotoxic potential of MAP, which, in contrast with ETA', lacks cytotoxic enzymatic activity, but rather induces cell-growth arrest by stoichiometric binding to microtubules. Interestingly, the escape mechanism for human effector proteins from the endosome to the cytosol is somewhat unclear. One tentative hypothesis discussed in the literature is that upon the pH shift in the endosome, the conformation of the protein is changed, leading to the exposure of potential processing and/or signal sequences (45). However, the cytotoxic activity of human CFPs can be improved further by the insertion of adapter sequences that facilitate the translocation of the effector molecule from the endosome to the cytosol (29). Therefore, our future research will focus on improving the translocation efficiency of anti–EpCAM(scFv)-MAP, to further enhance its therapeutic potential. In regard to the cytotoxic mode of action, we have clearly demonstrated the proapoptotic effect of MAP, previously (32). Interestingly, but not totally unexpected, the surface expression levels of EpCAM did not correlate quantitatively (IC50 values) with the sensitivity of different cell lines. EpCAM is known to be a highly glycosylated protein. Differences in glycosylation patterns of EpCAM have been attributed to differences in protein stability and thus functionality (46). Furthermore, distinct EpCAM glycosylation has been suggested to be one possible discriminating criterion between healthy and malignant cells (14, 47). Incubation with anti–EpCAM(scFv)-MAP led to a significant increase of the necrotic cell population, which seems to be the result of prolonged treatment. In regard to immunogenicity, fully human CFPs seem superior to ADCs or chimeric immunotoxins due to the fact that they show comparable efficacy but are expected to have significantly lower immunogenicity. In this study, intravenous injection of 4.0 mg/kg of anti–EpCAM(scFv)-MAP in vivo still did not reach the maximum tolerated dose, highlighting favorable tolerability of the new human CFP.

Given the fact that EpCAM is also expressed on healthy epithelia, a diagnostic preselection of patients with overexpression of EpCAM in cancerous lesions cells may provide a promising strategy to enhance the efficacy of EpCAM-directed cancer immunotherapy. Previously, it was demonstrated that postoperative intraperitoneal application of the anti-CD3/anti-EpCAM bispecific mAb catumaxomab proved to be beneficial in the treatment of patients with gastric cancer (48). It remains to be studied if perhaps anti–EpCAM(scFv)-MAP also shows enhanced efficacy when topically applied in a similar fashion.

In conclusion, the fully human fusion protein anti–EpCAM(scFv)-MAP shows promising carcinoma-selective cytotoxicity both in vitro and in an in vivo animal tumor model and may be of potential in the treatment of EpCAM-overexpressing malignancies.

D. Hristodorov, R. Mladenov, T. Thepen, and S. Barth have ownership interest (including patents) in MAP-based cytolytic fusion proteins. No potential conflicts of interest were disclosed by the other authors.

Conception and design: D. Hristodorov, W. Helfrich, R. Fischer, T. Thepen, S. Barth

Development of methodology: D. Hristodorov, M. Amoury, N. Berges, S. Di Fiore, T. Thepen, S. Barth

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D. Hristodorov, M. Amoury, R. Mladenov, J. Niesen, K. Arens, L. Hein, A.-T. Pham, W. Helfrich

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): D. Hristodorov, R. Mladenov, N. Berges, S. Di Fiore, M. Huhn, T. Thepen

Writing, review, and/or revision of the manuscript: D. Hristodorov, W. Helfrich, R. Fischer, T. Thepen, S. Barth

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A.-T. Pham, M. Huhn

Study supervision: N. Berges, S. Di Fiore, T. Thepen

The authors thank Reinhard Rosinke for the purification of recombinant proteins.

This work was funded in part by the ForSaTum project, which is sponsored within the NRW-EU Ziel 2-Programm “Regionale Wettbewerbsfähigkeit und Beschäftigung 2007-2013” (ERFE).

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.
Ma
H
,
Miao
X
,
Ma
Q
,
Zheng
W
,
Zhou
H
,
Jia
L
. 
Functional roles of glycogene and N-glycan in multidrug resistance of human breast cancer cells
.
IUBMB Life
2013
;
65
:
409
22
.
2.
Rebucci
M
,
Michiels
C
. 
Molecular aspects of cancer cell resistance to chemotherapy
.
Biochem Pharmacol
2013
;
19
:
00121
4
.
3.
Baeuerle
PA
,
Gires
O
. 
EpCAM (CD326) finding its role in cancer
.
Br J Cancer
2007
;
96
:
417
23
.
4.
Fong
D
,
Steurer
M
,
Obrist
P
,
Barbieri
V
,
Margreiter
R
,
Amberger
A
, et al
Ep-CAM expression in pancreatic and ampullary carcinomas: frequency and prognostic relevance
.
J Clin Pathol
2008
;
61
:
31
5
.
5.
Varga
M
,
Obrist
P
,
Schneeberger
S
,
Muhlmann
G
,
Felgel-Farnholz
C
,
Fong
D
, et al
Overexpression of epithelial cell adhesion molecule antigen in gallbladder carcinoma is an independent marker for poor survival
.
Clin Cancer Res
2004
;
10
:
3131
6
.
6.
Piyathilake
CJ
,
Frost
AR
,
Weiss
H
,
Manne
U
,
Heimburger
DC
,
Grizzle
WE
. 
The expression of Ep-CAM (17-1A) in squamous cell cancers of the lung
.
Hum Pathol
2000
;
31
:
482
7
.
7.
Kim
Y
,
Kim
HS
,
Cui
ZY
,
Lee
HS
,
Ahn
JS
,
Park
CK
, et al
Clinicopathological implications of EpCAM expression in adenocarcinoma of the lung
.
Anticancer Res
2009
;
29
:
1817
22
.
8.
Osta
WA
,
Chen
Y
,
Mikhitarian
K
,
Mitas
M
,
Salem
M
,
Hannun
YA
, et al
EpCAM is overexpressed in breast cancer and is a potential target for breast cancer gene therapy
.
Cancer Res
2004
;
64
:
5818
24
.
9.
Litvinov
SV
,
van Driel
W
,
van Rhijn
CM
,
Bakker
HA
,
van Krieken
H
,
Fleuren
GJ
, et al
Expression of Ep-CAM in cervical squamous epithelia correlates with an increased proliferation and the disappearance of markers for terminal differentiation
.
Am J Pathol
1996
;
148
:
865
75
.
10.
Brunner
A
,
Prelog
M
,
Verdorfer
I
,
Tzankov
A
,
Mikuz
G
,
Ensinger
C
. 
EpCAM is predominantly expressed in high grade and advanced stage urothelial carcinoma of the bladder
.
J Clin Pathol
2008
;
61
:
307
10
.
11.
Poczatek
RB
,
Myers
RB
,
Manne
U
,
Oelschlager
DK
,
Weiss
HL
,
Bostwick
DG
, et al
Ep-Cam levels in prostatic adenocarcinoma and prostatic intraepithelial neoplasia
.
J Urol
1999
;
162
:
1462
6
.
12.
Litvinov
SV
,
Bakker
HA
,
Gourevitch
MM
,
Velders
MP
,
Warnaar
SO
. 
Evidence for a role of the epithelial glycoprotein 40 (Ep-CAM) in epithelial cell-cell adhesion
.
Cell Adhes Commun
1994
;
2
:
417
28
.
13.
Xie
X
,
Wang
CY
,
Cao
YX
,
Wang
W
,
Zhuang
R
,
Chen
LH
, et al
Expression pattern of epithelial cell adhesion molecule on normal and malignant colon tissues
.
World J Gastroenterol
2005
;
11
:
344
7
.
14.
Pauli
C
,
Munz
M
,
Kieu
C
,
Mack
B
,
Breinl
P
,
Wollenberg
B
, et al
Tumor-specific glycosylation of the carcinoma-associated epithelial cell adhesion molecule EpCAM in head and neck carcinomas
.
Cancer Lett
2003
;
193
:
25
32
.
15.
Gastl
G
,
Spizzo
G
,
Obrist
P
,
Dunser
M
,
Mikuz
G
. 
Ep-CAM overexpression in breast cancer as a predictor of survival
.
Lancet
2000
;
356
:
1981
2
.
16.
Punt
CJ
,
Nagy
A
,
Douillard
JY
,
Figer
A
,
Skovsgaard
T
,
Monson
J
, et al
Edrecolomab alone or in combination with fluorouracil and folinic acid in the adjuvant treatment of stage III colon cancer: a randomised study
.
Lancet
2002
;
360
:
671
7
.
17.
Naundorf
S
,
Preithner
S
,
Mayer
P
,
Lippold
S
,
Wolf
A
,
Hanakam
F
, et al
In vitro and in vivo activity of MT201, a fully human monoclonal antibody for pancarcinoma treatment
.
Int J Cancer
2002
;
100
:
101
10
.
18.
Brischwein
K
,
Schlereth
B
,
Guller
B
,
Steiger
C
,
Wolf
A
,
Lutterbuese
R
, et al
MT110: a novel bispecific single-chain antibody construct with high efficacy in eradicating established tumors
.
Mol Immunol
2006
;
43
:
1129
43
.
19.
Sievers
EL
,
Senter
PD
. 
Antibody-drug conjugates in cancer therapy
.
Annu Rev Med
2013
;
64
:
15
29
.
20.
Firer
MA
,
Gellerman
G
. 
Targeted drug delivery for cancer therapy: the other side of antibodies
.
J Hematol Oncol
2012
;
5
:
70
.
21.
Pastan
I
,
Hassan
R
,
FitzGerald
DJ
,
Kreitman
RJ
. 
Immunotoxin treatment of cancer
.
Annu Rev Med
2007
;
58
:
221
37
.
22.
Cao
Y
,
Marks
JD
,
Huang
Q
,
Rudnick
SI
,
Xiong
C
,
Hittelman
WN
, et al
Single-chain antibody-based immunotoxins targeting Her2/neu: design optimization and impact of affinity on antitumor efficacy and off-target toxicity
.
Mol Cancer Ther
2012
;
11
:
143
53
.
23.
Janthur
WD
,
Cantoni
N
,
Mamot
C
. 
Drug conjugates such as antibody drug conjugates (ADCs), immunotoxins and immunoliposomes challenge daily clinical practice
.
Int J Mol Sci
2012
;
13
:
16020
45
.
24.
Kuan
CT
,
Pai
LH
,
Pastan
I
. 
Immunotoxins containing Pseudomonas exotoxin that target LeY damage human endothelial cells in an antibody-specific mode: relevance to vascular leak syndrome
.
Clin Cancer Res
1995
;
1
:
1589
94
.
25.
Siegall
CB
,
Liggitt
D
,
Chace
D
,
Mixan
B
,
Sugai
J
,
Davidson
T
, et al
Characterization of vascular leak syndrome induced by the toxin component of Pseudomonas exotoxin-based immunotoxins and its potential inhibition with nonsteroidal anti-inflammatory drugs
.
Clin Cancer Res
1997
;
3
:
339
45
.
26.
Huhn
M
,
Sasse
S
,
Tur
MK
,
Matthey
B
,
Schinkothe
T
,
Rybak
SM
, et al
Human angiogenin fused to human CD30 ligand (Ang-CD30L) exhibits specific cytotoxicity against CD30-positive lymphoma
.
Cancer Res
2001
;
61
:
8737
42
.
27.
Stahnke
B
,
Thepen
T
,
Stocker
M
,
Rosinke
R
,
Jost
E
,
Fischer
R
, et al
Granzyme B-H22(scFv), a human immunotoxin targeting CD64 in acute myeloid leukemia of monocytic subtypes
.
Mol Cancer Ther
2008
;
7
:
2924
32
.
28.
Tur
MK
,
Neef
I
,
Jost
E
,
Galm
O
,
Jager
G
,
Stocker
M
, et al
Targeted restoration of down-regulated DAPK2 tumor suppressor activity induces apoptosis in Hodgkin lymphoma cells
.
J Immunother
2009
;
32
:
431
41
.
29.
Hetzel
C
,
Bachran
C
,
Fischer
R
,
Fuchs
H
,
Barth
S
,
Stocker
M
. 
Small cleavable adapters enhance the specific cytotoxicity of a humanized immunotoxin directed against CD64-positive cells
.
J Immunother
2008
;
31
:
370
6
.
30.
de Bruyn
M
,
Bremer
E
,
Helfrich
W
. 
Antibody-based fusion proteins to target death receptors in cancer
.
Cancer Lett
2013
;
332
:
175
83
.
31.
Bremer
E
,
Helfrich
W
. 
A better TRAIL variant for tumor cell-specific targeting? - letter
.
Mol Cancer Ther
2010
;
9
:
2853
.
32.
Hristodorov
D
,
Mladenov
R
,
Pardo
A
,
Pham
AT
,
Huhn
M
,
Fischer
R
, et al
Microtubule-associated protein tau facilitates the targeted killing of proliferating cancer cells in vitro and in a xenograft mouse tumour model in vivo
.
Br J Cancer
2013
;
109
:
1570
8
.
33.
Huls
GA
,
Heijnen
IA
,
Cuomo
ME
,
Koningsberger
JC
,
Wiegman
L
,
Boel
E
, et al
A recombinant, fully human monoclonal antibody with antitumor activity constructed from phage-displayed antibody fragments
.
Nat Biotechnol
1999
;
17
:
276
81
.
34.
Bruns
CJ
,
Harbison
MT
,
Kuniyasu
H
,
Eue
I
,
Fidler
IJ
. 
In vivo selection and characterization of metastatic variants from human pancreatic adenocarcinoma by using orthotopic implantation in nude mice
.
Neoplasia
1999
;
1
:
50
62
.
35.
Pardo
A
,
Stocker
M
,
Kampmeier
F
,
Melmer
G
,
Fischer
R
,
Thepen
T
, et al
In vivo imaging of immunotoxin treatment using Katushka-transfected A-431 cells in a murine xenograft tumour model
.
Cancer Immunol Immunother
2012
;
61
:
1617
26
.
36.
Kampmeier
F
,
Niesen
J
,
Koers
A
,
Ribbert
M
,
Brecht
A
,
Fischer
R
, et al
Rapid optical imaging of EGF receptor expression with a single-chain antibody SNAP-tag fusion protein
.
Eur J Nucl Med Mol Imaging
2010
;
37
:
1926
34
.
37.
Barth
S
,
Huhn
M
,
Matthey
B
,
Klimka
A
,
Galinski
EA
,
Engert
A
. 
Compatible-solute-supported periplasmic expression of functional recombinant proteins under stress conditions
.
Appl Environ Microbiol
2000
;
66
:
1572
9
.
38.
Amoury
M
,
Blume
T
,
Brehm
H
,
Niesen
J
,
Tenhaef
N
,
Barth
S
, et al
SNAP-tag based agents for preclinical in vitro imaging in malignant diseases
.
Curr Pharm Des
2013
;
19
:
5429
36
.
39.
Jordan
MA
,
Kamath
K
. 
How do microtubule-targeted drugs work? An overview
.
Curr Cancer Drug Targets
2007
;
7
:
730
42
.
40.
Horn
U
,
Strittmatter
W
,
Krebber
A
,
Knupfer
U
,
Kujau
M
,
Wenderoth
R
, et al
High volumetric yields of functional dimeric miniantibodies in Escherichia coli, using an optimized expression vector and high-cell-density fermentation under non-limited growth conditions
.
Appl Microbiol Biotechnol
1996
;
46
:
524
32
.
41.
Zimmermann
S
,
Wels
W
,
Froesch
BA
,
Gerstmayer
B
,
Stahel
RA
,
Zangemeister-Wittke
U
. 
A novel immunotoxin recognising the epithelial glycoprotein-2 has potent antitumoural activity on chemotherapy-resistant lung cancer
.
Cancer Immunol Immunother
1997
;
44
:
1
9
.
42.
Kreitman
RJ
. 
Immunotoxins for targeted cancer therapy
.
AAPS J
2006
;
8
:
E532
51
.
43.
Senter
PD
. 
Potent antibody drug conjugates for cancer therapy
.
Curr Opin Chem Biol
2009
;
13
:
235
44
.
44.
Shen
BQ
,
Xu
K
,
Liu
L
,
Raab
H
,
Bhakta
S
,
Kenrick
M
, et al
Conjugation site modulates the in vivo stability and therapeutic activity of antibody-drug conjugates
.
Nat Biotechnol
2012
;
30
:
184
9
.
45.
Weidle
UH
,
Georges
G
,
Brinkmann
U
. 
Fully human targeted cytotoxic fusion proteins: new anticancer agents on the horizon
.
Cancer Genomics Proteomics
2012
;
9
:
119
33
.
46.
Munz
M
,
Fellinger
K
,
Hofmann
T
,
Schmitt
B
,
Gires
O
. 
Glycosylation is crucial for stability of tumour and cancer stem cell antigen EpCAM
.
Front Biosci
2008
;
13
:
5195
201
.
47.
Misonou
Y
,
Shida
K
,
Korekane
H
,
Seki
Y
,
Noura
S
,
Ohue
M
, et al
Comprehensive clinico-glycomic study of 16 colorectal cancer specimens: elucidation of aberrant glycosylation and its mechanistic causes in colorectal cancer cells
.
J Proteome Res
2009
;
8
:
2990
3005
.
48.
Atanackovic
D
,
Reinhard
H
,
Meyer
S
,
Spock
S
,
Grob
T
,
Luetkens
T
, et al
The trifunctional antibody catumaxomab amplifies and shapes tumor-specific immunity when applied to gastric cancer patients in the adjuvant setting
.
Hum Vaccin Immunother
2013
;
9
:
12
.