Introduction

The CD20 antigen, a 33- to 37-kDa membrane protein of unknown function, is an excellent immunotherapy target as it is expressed only on mature B cells and not on B-cell precursors. This antigen is not shed, internalized or modulated to a significant degree once antibody–antigen binding has occurred,¹ being able to activate effector mechanisms, such as complement-mediated lysis, antibody-dependent cellular cytotoxicity (ADCC) and apoptosis.² The anti-CD20 chimeric monoclonal antibody (mAb) Rituximab has emerged as an effective single agent for treatment of patients with CD20-positive non-Hodgkin's lymphomas (NHL) and chronic lymphocytic leukemia. In 1997, Rituximab was approved by US FDA for treatment of recurrent/refractory follicular NHL and, recently, of untreated aggressive NHL in combination with CHOP (cyclophosphamide, adriamycin, oncovin, prednisone) regimen. Rituximab treatment showed a response rate of about 50% in relapsed low-grade NHL.³ In combination with CHOP, it showed more than 70% of complete response in elderly patients with untreated aggressive NHL.⁴ In an effort to improve response rates and remission duration, several groups investigated the possibility of radiolabelling anti-CD20 antibodies with iodine-131 or yttrium-90.^{5, 6, 7, 8, 9} Despite the encouraging results reported in the latter studies, the majority of patients treated with anti-CD20 radioimmunoconjugates relapsed. Moreover, some side effects have been described, such as prolonged myelosuppression, a potential risk of treatment-associated myelodysplastic syndrome and acute myelogenous leukemia.⁹ In addition, a pilot study introducing a novel 'pretargeting' strategy of two-step radioimmunotherapy using streptavidin-labeled anti-CD20 antibodies was recently published.¹⁰

A different way to improve mAbs effects is to conjugate them to a toxic moiety, in order to obtain immunotoxins, chimeric proteins with specific cytotoxic effect. To this purpose, a variety of bacterial and plant toxins have been utilized.¹¹ Among plant toxins, ribosome-inactivating proteins (RIPs) are most frequently employed for the preparation of conjugates. Most commonly, RIPs are divided into type 1, consisting of a single-chain protein, and type 2, consisting of an enzymatic A chain linked to a B chain with lectin properties.^{12, 13} Type 1 RIPs and the A chain of type 2 RIPs cleave one or more adenine molecules from ribosomal RNA,¹⁴ thus damaging ribosomes in an irreversible manner. They also remove adenine residues from DNA, hence the denomination of adenine polynucleotide glycosylase proposed for RIPs.¹⁵ To date, several clinical trials have been conducted with immunotoxins, and some of them are still ongoing, to evaluate their antitumor efficacy. The response rates observed in phase I/II trials have often been higher than those reported for some of the conventional antiblastic drugs.^{11, 16, 17} In this research, the cytotoxic effect of Rituximab conjugated to the RIP saporin-S6 has been studied for the first time. Moreover, we explored the possibility of combining immunotoxin and the chemotherapic drug Fludarabine, to augment the efficiency of killing target cells.

Materials and methods

Monoclonal antibodies

MAb Rituximab (anti-human CD20) was obtained from Roche (Milano, Italy). A fluorescein isothiocyanate conjugated goat anti-human IgG (FITC-GAH) (Beckman-Coulter, Hialeah, FL, USA) was used as second step reagent for immunofluorescence staining.

Ribosome-inactivating protein

The type 1 RIP saporin (saporin-S6), from the seeds of Saponaria officinalis, was purified as previously described.¹⁸ Saporin-S6 was labeled with ¹²⁵I with the Iodogen reagent (Pierce Chemical Co., Catex, USA) according to the manufacturer's instructions.

Immunotoxins

The RIP and the mAb were conjugated via a disulfide bond between chemically inserted sulfhydryl (SH) groups. Rituximab and saporin-S6, the latter containing a trace of ¹²⁵I-RIP, were dissolved in 50 mM sodium borate buffer, pH 9.0, at a concentration of 2 mg/ml (mAb) and 8 mg/ml (RIP), and were modified by adding 2-iminothiolane (Sigma, St Louis, MO, USA) to a final concentration of 0.4 mM (antibody) or 1.0 mM (RIP), as already described.¹⁹ The antibody and the reduced RIP, in a 10:1 molar ratio, were allowed to react for 16 h at room temperature. The resulting conjugate was separated from the unreacted reagents and from RIP homopolymers by gel filtration on a Sephacryl S200 high-resolution column (100 cm 2.5 cm) (Pharmacia), equilibrated and eluted with phosphate-buffered saline (PBS, 0.14 M sodium chloride in 5 mM sodium phosphate buffer, pH 7.4).

The conjugate was analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) under non-reducing conditions. Proteins were incubated in sample buffer (40 mM Tris-HCl pH 6.8, 2% SDS, 0.005% bromophenol blue) containing 1 mg/ml iodoacetamide, for 30 min at room temperature, analyzed on a 4–15% PhastGel gradient, and then stained with Coomassie brilliant blue, following the recommended protocol from the manufacturer (Pharmacia Biotech). Molecular weight markers were from Sigma: myosin (205 kDa), -galactosidase (116 kDa), phosphorylase B (97 kDa), bovine serum albumin (66 kDa). The RIP/antibody ratio of the immunotoxin was estimated by densitometric analysis, performed with a Kodak DC 290 apparatus, using Kodak 1D, 3.6 software version.

Another gel was run at the same time and blotted on nitrocellulose paper (BioRad) with the Phast Transfer Kit (Pharmacia Biotech), as described in the manufacturer's protocol. Saporin-S6 was revealed by incubation overnight with a rabbit anti-saporin-S6 serum.¹⁵ Detection was carried out with a horseradish peroxidase-conjugated goat anti-(rabbit IgG)Ig and with 4-chloro-naphtol (Sigma).

Protein synthesis inhibitory activity of the immunotoxin was assayed on a rabbit reticulocyte lysate as described below. The conjugate was sterilized by filtering through a 0.22 m filter, divided into aliquots and stored in liquid nitrogen at M concentrations, as RIP.

A conjugate anti-IgG murine/saporin-S6 was used in cytotoxicity tests as irrelevant immunotoxin.

Cells

The activity of the conjugate was assayed on the CD20-positive cell lines Raji, derived from a Burkitt's lymphoma, and D430B, an Epstein–Barr virus (EBV)-infected B cell line.²⁰ The CD20-negative cell line Jurkat, derived from T-cell leukemia, was used to detect the aspecific cytotoxicity.

Cells were maintained in RPMI 1640 medium (Gibco, Life Technologies Inc., Rockville, MD, USA), supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Biospa, Milano, Italy), 2 mM L-glutamine, 100 U/ml penicillin and 100 g/ml streptomycin (Bio Whittaker, Verviers, Belgium) (hereafter named complete medium), in humidified air with 5% CO₂ at 37°C. Viability was checked before each experiment by Trypan blue dye exclusion.

NHL neoplastic cells were obtained from four patients, after informed consent. B cells were more than 90% of the total population, as evaluated by flow cytometry and FITC-CD19 mAb staining.

Rituximab affinity for CD20-expressing cell lines

The antigen expression on cell lines and the reactivity with immunotoxin were ascertained by an indirect immunofluorescence method. Briefly, cells were harvested and adjusted to a concentration of 10⁶ cells/ml in complete medium. To 200 l of cell suspension, 200 l of Rituximab, free or conjugated to saporin-S6, was added to final concentrations ranging from 0.01 to 10 nM as mAb, and control samples were run with medium alone. Cells were incubated for 15 min at room temperature (21°C), washed twice in PBS, containing 1% FBS, and incubated again for 10 min at room temperature in a volume of 50 l with 2 l of FITC-GAH. After three washes as above, the samples were fixed with PBS containing 1% formalin. The binding of mAb and immunotoxin was assessed by flow cytometry with an EPICS XL equipment (Beckman-Coulter). Histograms and statistics were generated with the software of the EPICS-dedicated computer. The intensity of fluorescence, evaluated as mean channel value (MCV), was used as a measure of the binding capacity of the antibody alone or of the corresponding immunotoxin.

Protein synthesis inhibition assays

The inhibitory activity of immunotoxin and free RIP on cell-free protein synthesis was evaluated using a rabbit reticulocyte lysate system, prepared as described by Allen and Schweet.²¹ After reduction with 20 mM 2-mercaptoethanol for 30 min at 37°C, immunotoxin and RIP were appropriately diluted and then added to a reaction mixture, as described by Bolognesi et al.¹⁹ Each experiment was carried out in duplicate. The concentration of immunotoxin, expressed as RIP content, causing 50% inhibition of [¹⁴C]leucine incorporation (IC₅₀) was calculated by linear regression analysis.

The cytotoxicity of the immunotoxin was evaluated from the inhibition of [³H]leucine incorporation. Cells (5 10³/well) were seeded in 96-well microtiter plates (Falcon, Becton Dickinson, Franklin Lakes, NJ, USA) in 100 l of complete medium, and 100 l of immunotoxin was added to final concentrations ranging from 0.001 to 100 nM for Raji and D430B, and from 0.1 to 100 nM for Jurkat. Concentrations are expressed as RIP content. Control samples were run with RIP alone, mAb alone, a mixture of unconjugated anti-CD20 mAb and RIP, or with an irrelevant anti-mouse immunotoxin. After 96 h cells were washed and 5 kBq of L-[4,5-³H]leucine (Amersham) was added, and after further 6 h cells were harvested with an automatic cell harvester (Skatron Instruments, Lier, Norway) onto glass-fiber diskettes. The radioactivity incorporated was determined as previously described.¹⁹

To verify a possible additive effect, Raji cells were exposed to both Rituximab/saporin-S6 immunotoxin and the chemotherapic drug Fludarabine monophosphate (FLU) (Schering AG, Milano, Italy), an adenine nucleoside analog.²² Cells (5 10³/well) were seeded, as described above, in 100 l of complete medium, and 50 l of immunotoxin was added to final concentrations ranging from 0.01 to 10 nM, as RIP, and 50 l of FLU to final concentrations of 0.5, 1 and 1.5 M. After 96 h, protein synthesis was measured as above.

Evaluation of apoptotic membrane changes

On Raji and D430B cell lines, viable and dead cells were evaluated by a double staining via FITC-AnnexinV and propidium iodide with a kit purchased from Bender Medsystem (Wien, Austria), following the manufacturer's instructions. Cells were incubated for 72 h with complete medium containing Rituximab/saporin-S6 at concentrations ranging from 0.001 to 10 nM or containing the single immunotoxin components or a mixture of unconjugated Rituximab and saporin-S6. Analysis was performed by flow cytometry as described by Tazzari et al.²³

Cells from four NHL patients were incubated for 72 h with immunotoxin or with a mixture of unconjugated mAb and RIP. Cells were then collected and incubated with AnnexinV, as described above. Positivity was evaluated by flow cytometry, by gating B cells with phycoerythrin (PE)-conjugated CD19 mAb (PE-CD19).

Colony assays

The effect of short-term exposure to immunotoxin was also evaluated on the Raji cell line by a clonogenic assay. Briefly, 2 10⁴ cells were incubated for 3 h with immunotoxin or a mixture of unconjugated Rituximab and saporin-S6 at concentrations ranging from 0.1 to 100 nM in a total volume of 200 l. After one wash with 2 ml of PBS, 2 10³ tumor cells were plated in 50 l of complete medium. Methylcellulose was added in a volume of 0.5 ml to 1.1% final concentration. Aggregates >50 cells were scored with an inverted microscope after 10 days of culture.

In an additional experiment, performed as above, Raji cells were incubated with 1 nM immunotoxin, 1 M FLU or with a mixture of both.

Reactivity of Rituximab with CD34+ cells

To exclude crossreactivity of Rituximab with CD34+ cells, we performed a double staining as follows. Cells obtained from three normal donors and three NHL patients (after informed consent), all submitted to G-CSF treatment, were stained with a PE-conjugated CD34 mAb (PE-CD34, Beckman Coulter) for 20 min. After one wash with PBS at room temperature, staining with Rituximab and Rituximab/saporin-S6 was performed as described above. Control samples were run with PE-CD34 alone and with PE-CD34 and FITC-GAH. Analysis was performed by flow cytometry, as described above.

Statistical analyses

Results are given as meanss.d. Data were analyzed by ANCOVA and ANOVA test with Bonferroni correction for multiple comparison. Immunotoxin and Fludarabine combined effects were analyzed by ANOVA, with Bonferroni correction, followed by a comparison of single vs mixed drugs with Dunnett's test. Statistical analyses were conducted using the XLSTAT-Pro software, version 6.1.9 (Addinsoft 2003).

Results

Immunotoxin properties

Saporin-S6 was conjugated to Rituximab through an artificial disulfide bond. SH groups were inserted in each type of molecule by an imidoester reaction between 2-iminothiolane and the primary amino groups of the proteins. Rituximab showed a marked reactivity with 2-iminothiolane, with an average of 3.7 SH groups inserted per molecule, using a standard concentration of the linking reagent (Table 1). The composition of purified conjugate was analyzed by SDS–PAGE in non-reducing conditions, using a 4–15% gradient gel. The ratio of the different reaction products was determined by densitometric scanning of Coomassie blue-stained gel (Figure 1). This analysis demonstrates a toxin/mAb molar ratio of about 1.89. The conjugate is constituted of four different molecular products containing 1–4 saporin-S6 molecules per Rituximab molecule. The most represented products are the conjugates 2/1 RIP/mAb (31.3%) and 1/1 RIP/mAb (30.4%). The percentage of unconjugated Rituximab is less than 10%. The presence of saporin-S6 in the products stained by Coomassie brilliant blue was confirmed by Western blot results (data not shown).

Figure 1.

Analysis of Rituximab (lane 1) and Rituximab/saporin-S6 (lane 2) by SDS–PAGE under non-reducing conditions on a 4–15% PhastGel. The RIP/mAb ratio was determined by densitometric scanning of Coomassie blue-stained gel. Each band is expressed as percentage of the total intensity.

Full figure and legend (60K)

Table 1 - Characteristics of the Rituximab/Saporin-S6 immunotoxin.

Full table

After conjugation, we investigated whether saporin-S6 and Rituximab maintained their characteristics. First, the enzymatic activity of saporin-S6 was evaluated on a cell-free protein synthesis system (rabbit reticulocytes lysate). This test showed that protein synthesis was efficiently inhibited by immunotoxin with an IC₅₀ of 70 pM, which was comparable to that of free saporin-S6 (58 pM), tested in the same condition (Table 1). The binding kinetics of free and conjugated mAb were obtained by incubating Raji and D430B cells with different amounts of Rituximab and Rituximab/saporin-S6. The results are shown in Figure 2, where the mean fluorescence intensity is plotted against the concentration of the mAb. The comparison of Rituximab and immunotoxin binding to target cells indicates that both reagents have similar affinity for the CD20 molecule. Thus it could be concluded that the chemical link between saporin-S6 and Rituximab does not significantly modify either antibody or RIP properties.

Figure 2.

Cytofluorimetric analysis of the antigen CD20 expressed by Raji and D430B cell lines. Rituximab, free () or conjugated to saporin-S6 (), was added to cells at concentrations ranging from 0.01 to 10 nM for 15 min at room temperature. After 10 min incubation with FITC-GAH, the intensity of fluorescence, evaluated as MCV, was used as a measure of the binding capacity. Results are means of three different experiments. s.d. never exceeds 15%. ANCOVA test shows no significant differences between conjugated and free Rituximab antigen binding.

Full figure and legend (32K)

To estimate the ability of the immunotoxin to eliminate CD20+ cells, we used three different tests, that is, inhibition of protein synthesis, evaluation of apoptosis and clonogenic assay.

Protein synthesis inhibition assays

Protein synthesis inhibition assays were performed on two target cell lines, namely Raji and D430B, and on a non-target cell line, Jurkat. Conjugation with Rituximab enhanced saporin-S6 protein synthesis inhibitory activity on target cells by at least 2 logs, with IC₅₀ values of 0.123 nM (Raji) and 0.313 nM (D430B) (Figure 3, Table 2). Moreover, in both cell lines, protein synthesis was completely inhibited by immunotoxin at 10 nM concentration, as RIP content. No inhibition was induced by free mAb. The irrelevant immunotoxin anti-mouse/saporin-S6 exerted similar toxicity as native RIP. Protein synthesis of CD20-negative Jurkat cells was not affected by the immunotoxin at concentrations up to 10 nM.

Figure 3.

Protein synthesis inhibition assay on Raji, D430B and Jurkat cell lines treated with Rituximab/saporin-S6 immunotoxin (), Rituximab (), saporin-S6 (), a mixture of unconjugated Rituximab and saporin-S6 () or an irrelevant immunotoxin (). A total of 5 10³ cells were seeded in 96-well plates in a total volume of 200 l of complete medium containing various concentrations of substances. After 96 h of incubation and further 6 h with [³H]leucine, the radioactivity incorporated was determined. Results are means of three different experiments each performed in triplicate. s.d. never exceeds 10%. Data were analyzed by ANCOVA/Bonferroni test. Raji and D430B protein synthesis inhibitions by immunotoxin are significantly different (P<0.0001) from that by Rituximab, saporin-S6 and the mixture of unconjugated RIP and mAb. No significant differences are observed on Jurkat non-target cells, between immunotoxin and the other tested substances.

Full figure and legend (140K)

Table 2 - Effect of Rituximab/Saporin-S6 immunotoxin on protein synthesis and viability by Raji and D430B cell lines.

Full table

Evaluation of apoptotic membrane changes

Since in previous works we demonstrated that RIP and RIP-containing immunotoxins induce programmed cell death (ie apoptosis) in target cells,^{23, 24} we evaluated apoptotic changes in CD20+ cell lines. In fact the exposure of Raji and D430B cells to Rituximab/saporin-S6 immunotoxin induced dose-dependent apoptotic membrane changes, evaluated after 72 h by a double staining via FITC-AnnexinV and propidium iodide. The percentage of AnnexinV-positive cells was over 95% in both cell lines treated with 10 nM immunotoxin. RIP alone or a mixture of unconjugated RIP and mAb in the same conditions caused less than 30% of apoptosis. On exposing cells to Rituximab alone, no significant increase of apoptosis over controls was observed (Figure 4).

Figure 4.

Flow cytometry analysis of Raji and D430B cell lines stained via FITC-AnnexinV and propidium iodide. Cells were treated for 72 h with 0.001–10 nM concentrations of Rituximab/saporin-S6 immunotoxin (), Rituximab (), saporin-S6 () or a mixture of Rituximab and saporin-S6 (). Results are means of two different experiments. s.d. never exceeded 10%. Apoptosis induced by immunotoxin results very significant vs Rituximab, saporin-S6 and their mixture, when compared by ANCOVA/Bonferroni (P<0.0001).

Full figure and legend (43K)

Colony assays

To evaluate the potential antitumor activity of anti-CD20 Rituximab/saporin-S6, the clonogenicity of Raji cell line was determined after a short-term exposure (3 h) to the immunotoxin. A complete elimination of Raji clones was reached with 10 nM immunotoxin (Figure 5). At the same concentration, a mixture of free RIP and mAb gave only a 10% inhibition of clonogenic growth.

Figure 5.

Influence of Rituximab/saporin-S6 immunotoxin on the clonogenic growth of Raji cell line. Cells were exposed for 3 h to concentrations ranging from 0.1 to 100 nM of immunotoxin () or a mixture of Rituximab and saporin-S6 (). In all, 2 10³ cells were plated in semisolid medium. Aggregates of >50 cells were scored after 10 days of culture. Results are means of two different experiments each performed in triplicate. s.d. never exceeds 15%. Data are significant by ANCOVA/Bonferroni tests (P<0.0001, immunotoxin vs Rituximab, saporin-S6 and mixture).

Full figure and legend (20K)

Immunotoxin and Fludarabine combined cytotoxicity

Further series of experiments were designed to enhance the killing efficiency of immunotoxin. Since Fludarabine (FLU) is one of the most widely used drugs for the treatment of indolent NHL,²² we tested a mixture of the drug and the immunotoxin. As expected, sensitivity of Raji to Rituximab/saporin-S6 immunotoxin was augmented when cells were coincubated with FLU (Figure 6). The ANOVA test was utilized to compare toxicity by immunotoxin (IT) and Fludarabine alone or mixed. Bonferroni and Dunnett tests demonstrate that three combinations of FLU and Rituximab/saporin-S6 reduce protein synthesis significantly compared to single compounds (FLU 1 M+IT 1 nM, P<0.0001; FLU 1 M+IT 0.1 nM, P=0.0002; FLU 0.5 M+IT 0.1 nM, P<0.0001).

Figure 6.

Combined cytotoxic effect of FLU and Rituximab/saporin-S6 immunotoxin on Raji cells protein synthesis. Cells were incubated with different immunotoxin concentrations, as reported on the X-axis. Symbols represent different FLU concentrations: 1.5 M (), 1 M () and 0.5 M (). Asterisk (^*) represents immunotoxin without FLU. Symbols on the Y-axis indicate protein synthesis inhibition caused by FLU alone. Results are means of two different experiments. s.d. never exceeded 10%. Protein synthesis inhibition by immunotoxin is significantly augmented after combination with FLU 0.5, 1 and 1.5 M (ANCOVA/Bonferroni tests, P<0.0001). ANOVA/Bonferroni followed by Dunnett's test demonstrates only three immunotoxin and Fludarabine combinations (dark symbols) results as significant vs single drugs at the same concentrations (FLU 1 M+IT 1 nM, P<0.0001; FLU 1 M+IT 0.1 nM, P=0.0002; FLU 0.5 M+IT 0.1 nM, P<0.0001).

Full figure and legend (27K)

In addition, the antitumor activity of the combination of 1 M FLU and 1 nM Rituximab/saporin-S6 was assayed in a clonogenic assay. As shown in Figure 7, both immunotoxin and FLU reduced Raji clonogenic growth by about 80%, but did not eliminate it completely, while cell growth was nearly abolished by applying the reagents together, confirming the protein synthesis results.

Figure 7.

Combined cytotoxic effect of 1 M FLU and 1 nM Rituximab/saporin-S6 immunotoxin on Raji cells clonogenic assay. Aggregates of >50 cells were scored after 10 days of culture. Results are means of two different experiments each performed in triplicate. s.d. never exceeded 15%. ANOVA/Bonferroni followed by Dunnett's test demonstrates the combination of immunotoxin and Fludarabine results as significant vs single drugs (P=0.001).

Full figure and legend (49K)

Apoptosis of NHL patients' treated cells

The exposure of lymphoma cells from four NHL patients to Rituximab/saporin-S6 immunotoxin induced dose-dependent apoptotic membrane changes, evaluated after 72 h by FITC-AnnexinV staining. Positivity was evaluated by flow cytometry, by gating B cells with PE-CD19 mAb. The percentage of AnnexinV-positive cells was over 80% in cells treated with 10 nM immunotoxin. A mixture of unconjugated RIP and mAb at 10 nM concentration gave less than 40% of AnnexinV+ cells (Figure 8).

Figure 8.

Flow cytometry analysis of cells from four NHL patients, stained via FITC-AnnexinV and propidium iodide. Cells were treated for 72 h with 0.1–10 nM concentrations of immunotoxin or 10 nM Rituximab+Saporin-S6. s.d. never exceeded 15%. ANOVA/Bonferroni followed by Dunnett's test demonstrates immunotoxin (0.1, 1 and 10 nM) results as significant vs control (P<0.0001) and vs the mixture of RIP and mAb (P<0.0001).

Full figure and legend (52K)

Reactivity of Rituximab with CD34+ cells

To exclude a possible toxicity on stem cells, we performed a double staining on peripheral blood obtained from normal donors and NHL patients primed with G-CSF. CD34+ cells were negative when stained with Rituximab. Cells from both donors and patients showed the same pattern (data not shown).

Discussion

Monoclonal antibodies are currently used in cancer therapy, both alone and conjugated to radioactive compounds or toxic moieties.²⁵ This approach might help in the management of cancer patients, since toxicity due to chemotherapy might be reduced. In the past years, many protocols have been developed for CD20+ neoplasias utilizing the mAb Rituximab. This is a chimeric molecule consisting of human Fc and constant region and mouse variable region, recognizing the CD20-related antigen.²⁶ Rituximab is actually accepted as an effective treatment, with few side effects, for NHL and other B-cell neoplasias. The basic effects on target cells seem to be due to apoptotic triggering and complement- and cell-mediated cytotoxicity. An improvement of Rituximab-based therapy could be obtained by linking the mAb to a toxic moiety able to kill target cells without needing any cofactor or cell-mediated immunity.

Plant toxins linked to mAbs generate hybrid molecules, named immunotoxins, which are potent reagents useful in killing cell population bearing a selected antigen. These molecules have been mainly devised for cancer immunotherapy and would be particularly suitable to eliminate residual neoplastic disease after chemo- or radiotherapy. Immunotoxins could also be useful for treatment of graft-versus-host disease rejection and immune disorders.²⁷ To date, several clinical trials have been conducted, and some are still ongoing, to evaluate the antitumor efficacy of immunotoxins. The response rates observed in phase I/II trials have often been higher than those reported for some of the conventional drugs.^{28, 29}

In our study, we constructed and tested in vitro the anticancer properties of a chemical anti-CD20 immunotoxin composed of the anti-CD20 mAb Rituximab and the type 1 RIP saporin-S6, which inhibits protein synthesis by means of a peculiar adenine polynucleotide glycosylase activity.

This immunotoxin is specifically cytotoxic for the CD20+ cell lines Raji and D430B. Rituximab/saporin-S6 is able to inhibit protein synthesis completely, induce apoptosis and abolish clonogenic growth.

It should be highlighted that in a previous study³⁰ unconjugated Rituximab, at 30 nM concentration, induced apoptotic changes in 8–15% of target cells, and only when crosslinking was induced, by using a secondary anti-human IgG antibody, the percentage of apoptotic cells arose to 20–38%. In our study, Rituximab/saporin-S6 caused apoptosis in almost 95% of cells at concentrations at least three-fold lower, without any crosslinking. Moreover, crosslinking is very difficult to obtain in vivo.

As already demonstrated in previous papers,^{31, 32} Rituximab effects rely on complement- and cell-mediated cytotoxicity, which depend on CD55/CD59 levels³³ and on phagocytic activity, respectively. These activities are variable among individuals, and final efficiency on target cells may be different in each patient. Thus the use of Rituximab/saporin-S6 immunotoxin could reduce the variability mentioned above and give higher reproducibility of the clinical response.

In previous studies it was reported that CD20 antigen is not internalized after the antibody binding.^{1, 34} Nevertheless, the strongly increased selective cytotoxicity to CD20+ cells of saporin-S6 conjugated with Rituximab indicates an effective RIP translocation into the cytoplasm. This is in agreement with previous observations made with the mAb Campath-1, which was efficiently internalized only after conjugation with saporin-S6.³⁵ Moreover, in previous papers, we have demonstrated that immunotoxin uptake, routing and intracellular fate not only depends on the carrier properties but is also influenced by the toxic moiety.³⁶

The results described above confirm that a conjugate containing an mAb and an RIP is able to kill target cells in a more efficient fashion than the mAb alone. Rituximab/saporin-S6 could have many other advantages. In addition to the specific toxicity without the variability due to CD55/CD59 levels or to the phagocytic activity, immunotoxin can induce apoptosis through different mechanisms. Moreover, it should be considered that immunotoxins containing RIP are also active against non-dividing cells, do not induce drug resistance and are active even on multidrug-resistant cells.³⁷

In an attempt to improve the antitumor efficacy of the treatment, we also tried a simultaneous incubation of target cells with Rituximab/saporin-S6 immunotoxin and FLU. FLU is an adenine nucleoside analog that shows therapeutic activity in clinical treatment of many B-cell neoplasias and is one of the most used drugs for low-grade NHL therapy.^{38, 39} Previous studies demonstrated that neoplastic B cells treated with FLU showed the characteristic fragmentation pattern of apoptosis associated with an increased internucleosomal DNA breakdown.⁴⁰ The combination of FLU and Rituximab was effective, with modest side effects, in the clinical treatment of indolent lymphoma (reviewed by Boye et al²⁶).

In our experiments, we found that simultaneous incubation of target cells with Rituximab/saporin-S6 and FLU produced a synergistic toxic effect that led to a total elimination of the neoplastic population. The low drug concentrations used in this assay, which are approximately 10-fold lower than those achievable in patients, strongly suggest that a combined immunotoxin/FLU therapy should also give synergistic cytotoxic effects in vivo.

Rituximab was found negative on CD34+ cells, thus making safe a possible ex vivo bone marrow purging with Rituximab-containing immunotoxin, while further in vivo studies (biodistribution, pharmacokinetic and systemic toxicity) will be required prior to its use in clinical trials.

In conclusion, our study points out that it is possible to augment specifically the Rituximab toxicity on target cells by linking it to saporin-S6 and that a further increase of toxicity is achievable by combination of the immunotoxin with the chemotherapic drug FLU. Thus in a near future it could be possible to devise a therapy schedule of B-cell NHLs including these two cytotoxic agents.

References

1. Anderson KC, Bates MP, Slaughenhoupt BL, Pinkus GS, Schlossman SF, Nadler LM. Expression of human B cell-associated antigens on leukemias and lymphomas: a model of human B cell differentiation. Blood 1984; 63: 1424–1433. | PubMed | ISI | ChemPort |
2. Maloney DG. Mechanism of action of rituximab. Anticancer Drugs 2001; 12 (Suppl. 2): S1–S4. | PubMed |
3. McLaughlin P, Grillo-Lopez AJ, Link BK, Levy R, Czuczman MS, Williams ME et al. Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol 1998; 16: 2825–2833. | PubMed | ISI | ChemPort |
4. Coiffier B, Lepage E, Briere J, Herbrecht R, Tilly H, Bouabdallah R et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346: 235–242. | Article | PubMed | ISI | ChemPort |
5. Ma D, McDevitt MR, Barendswaard E, Lai L, Curcio MJ, Pellegrini V et al. Radioimmunotherapy for model B cell malignancies using ⁹⁰Y-labeled anti-CD19 and anti-CD20 monoclonal antibodies. Leukemia 2002; 16: 60–66. | Article |
6. Goldenberg DM. The role of radiolabeled antibodies in the treatment of non-Hodgkin's lymphoma: the coming of age of radioimmunotherapy. Crit Rev Oncol Hematol 2001; 39: 195–201. | Article | PubMed | ISI | ChemPort |
7. Kaminski MS, Estes J, Zasadny KR, Francis IR, Ross CW, Tuck M et al. Radioimmunotherapy with Iodine 131 Itositumomab for relapsed or refractory B-cell non-Hodgkin's lymphoma: updated results and long-term follow-up of the University of Michigan experience. Blood 2000; 96: 1259–1266. | PubMed | ISI | ChemPort |
8. Witzig TE, Gordon LI, Cabanillas F, Czuczman MS, Emmanouilides C, Joyce R et al. Randomized controlled trial of yttrium-90-labeled ibritumomab tiuxetan radioimmunotherapy versus rituximab immunotherapy for patients with relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin's lymphoma. J Clin Oncol 2002; 20: 2453–2462. | Article | PubMed | ISI | ChemPort |
9. Cheson BD. Radioimmunotherapy of non-Hodgkin lymphomas. Blood 2003; 101: 391–398. | Article | PubMed | ISI | ChemPort |
10. Weiden PL, Breitz HB, Press O, Appelbaum JW, Bryan JK, Gaffigan S et al. Pretargeted radioimmunotherapy (PRIT) for treatment of non-Hodgkin's lymphoma (NHL): initial phase I/II study results. Cancer Biother Radiopharm 2000; 15: 15–29. | PubMed | ISI | ChemPort |
11. Pastan I, Kreitman RJ. Immunotoxins for targeted cancer therapy. Adv Drug Deliv Rev 1998; 31: 53–88. | Article | PubMed |
12. Barbieri L, Battelli MG, Stirpe F. Ribosome-inactivating proteins from plants. Biochim Biophys Acta 1993; 1154: 237–282. | Article | PubMed | ISI | ChemPort |
13. Peumans WJ, Hao Q, Van Damme EJ. Ribosome-inactivating proteins from plants: more than RNA N-glycosidases? FASEB J 2001; 15: 1493–1506. | Article | PubMed | ChemPort |
14. Barbieri L, Valbonesi P, Bonora E, Gorini P, Bolognesi A, Stirpe F. Polynucleotide:adenosine glycosidase activity of ribosome-inactivating proteins: effect on DNA, RNA and poly(A). Nucleic Acids Res 1997; 25: 518–522. | Article | PubMed | ISI | ChemPort |
15. Bolognesi A, Polito L, Lubelli C, Barbieri L, Parente A, Stirpe F. Ribosome-inactivating and adenine polynucleotide glycosylase activity in Mirabilis jalapa L. tissues. J Biol Chem 2002; 277: 13709–13716. | Article | PubMed | ChemPort |
16. Frankel AE, Kreitman RJ, Sausville EA. Targeted toxins. Clin Cancer Res 2000; 6: 326–334. | PubMed | ISI | ChemPort |
17. Thrush GR, Lark LR, Clinchy BC, Vitetta ES. Immunotoxins: an update. Annu Rev Immunol 1996; 14: 49–71. | Article | PubMed | ISI | ChemPort |
18. Barbieri L, Stoppa C, Bolognesi A. Large-scale chromatographic purification of ribosome-inactivating proteins. J Chromatogr 1987; 408: 235–243. | Article |
19. Bolognesi A, Polito L, Tazzari PL, Lemoli RM, Lubelli C, Fogli M et al. In vitro anti-tumor activity of anti-CD80 and anti-CD86 immunotoxins containing type 1 ribosome-inactivating proteins. Br J Haematol 2000; 110: 351–361. | Article | PubMed |
20. Tazzari PL, de Totero D, Bolognesi A, Testoni N, Pileri S, Roncella S et al. An Epstein–Barr virus-infected lymphoblastoid cell line (D430B) that grows in SCID-mice with the morphologic features of a CD30+ anaplastic large cell lymphoma, and is sensitive to anti-CD30 immunotoxins. Haematologica 1999; 84: 988–995. | PubMed |
21. Allen EH, Schweet RS. Synthesis of hemoglobin in a cell free system. Properties of the complete system. J Biol Chem 1962; 237: 760–767. | PubMed | ISI | ChemPort |
22. Johnson SA. Nucleoside analogues in the treatment of haematological malignancies. Expert Opin Pharmacother 2001; 2: 929–943. | Article | PubMed | ChemPort |
23. Tazzari PL, Polito L, Bolognesi A, Pistillo MP, Capanni P, Palmisano GL et al. Immunotoxins containing recombinant anti-CTLA-4 single-chain fragment variable antibodies and saporin: in vitro results and in vivo effects in an acute rejection model. J Immunol 2001; 167: 4222–4229. | PubMed |
24. Bolognesi A, Tazzari PL, Olivieri F, Polito L, Falini B, Stirpe F. Induction of apoptosis by ribosome-inactivating proteins and related immuntoxins. Int J Cancer 1996; 68: 349–355. | Article | PubMed | ChemPort |
25. Carter P. Improving the efficacy of antibody-based cancer therapies. Nat Rev Cancer 2001; 1: 118–129. | Article | PubMed | ChemPort |
26. Boye J, Elter T, Engert A. An overview of the current clinical use of the anti-CD20 monoclonal antibody rituximab. Ann Oncol 2003; 14: 520–535. | Article | PubMed | ChemPort |
27. Wu M. Are immunoconjugates useful for therapy with autoimmune diseases? Int J Immunopharmacol 1997; 19: 83–93. | Article | PubMed |
28. Pastan I I, Kreitman RJ. Immunotoxins for targeted cancer therapy. Adv Drug Deliv Rev 1998; 31: 53–88. | Article | PubMed |
29. Frankel AE, Kreitman RJ, Sausville EA. Targeted toxins. Clin Cancer Res 2000; 6: 326–334. | PubMed | ISI | ChemPort |
30. Cardarelli PM, Quinn M, Buckman D, Fang Y, Colcher D, King DJ et al. Binding to CD20 by anti-B1 antibody or F(ab')₂ is sufficient for induction of apoptosis in B cell lines. Cancer Immunol Immunother 2002; 51: 15–24. | Article | PubMed | ISI | ChemPort |
31. Van der Kolk LE, Grillo-Lopez AJ, Baars JW, Hack CE, van Oers MHJ. Complement activation plays a key role in the side-effects of rituximab treatment. Br J Haematol 2001; 115: 807–811. | Article | PubMed | ISI | ChemPort |
32. Bellosillo B, Villamor N, Lopez-Guillermo A, Marce S, Esteve J, Campo E et al. Complement-mediated cell death induced by rituximab in B-cell lymphoproliferative disorders is mediated in vitro by a caspase-independent mechanism involving the generation of reactive oxygen species. Blood 2001; 98: 2771–2777. | Article | PubMed | ChemPort |
33. Treon SP, Mitsiades C, Mitsiades N, Young G, Doss D, Schlossman R et al. Tumor cell expression of CD59 is associated with resistance to CD20 serotherapy in patients with B-cell malignancies. J Immunother 2001; 24: 263–271. | Article | ChemPort |
34. Countouriotis A, Moore TB, Sakamoto KM. Cell surface antigen and molecular targeting in the treatment of hematologic malignancies. Stem Cells 2002; 20: 215–229. | Article | PubMed | ISI | ChemPort |
35. Tazzari PL, Barbieri L, Gobbi M, Dinota A, Rizzi S, Bontadini A et al. An immunotoxin containing a rat IgM monoclonal antibody (Campath 1) and saporin 6: effect on T lymphocytes and hemopoietic cells. Cancer Immunol Immunother 1988; 26: 231–236. | PubMed |
36. Bolognesi A, Tazzari PL, Tassi C, Gromo G, Gobbi M, Stirpe F. A comparison of anti-lymphocyte immunotoxins containing different ribosome-inactivating proteins and antibodies. Clin Exp Immunol 1992; 89: 341–346. | PubMed |
37. Dinota A, Tazzari PL, Michieli M, Visani G, Gobbi M, Bontadini A et al. In vitro bone marrow purging of multidrug-resistant cells with a mouse monoclonal antibody directed against Mr 170,000 glycoprotein and a saporin-conjugated anti-mouse antibody. Cancer Res 1990; 50: 4291–4294. | PubMed |
38. Lazzarino M, Orlandi E, Montillo M, Tedeschi A, Pagnucco G, Astori C et al. Fludarabine, cyclophosphamide, and dexamethasone (FluCyD) combination is effective in pretreated low-grade non-Hodgkin's lymphoma. Ann Oncol 1999; 10: 59–64. | Article | PubMed |
39. Zinzani PL, Magagnoli M, Moretti L, De Renzo A, Battista R, Zaccaria A et al. Randomized trial of fludarabine versus fludarabine and idarubicin as frontline treatment in patients with indolent or mantle-cell lymphoma. J Clin Oncol 2000; 18: 773–779. | PubMed |
40. Zinzani PL, Buzzi M, Farabegoli P, Tosi P, Fortuna A, Visani G et al. Induction of 'in vitro' apoptosis by Fludarabine in freshly isolated B-chronic lymphocytic leukemia cells. Leuk Lymphoma 1994; 13: 95–97. | PubMed |

Acknowledgements

This study was supported by the University of Bologna, funds for selected research topics, from the Ministero della Salute, by the Pallotti's Legacy for Cancer Research, by the Fondazione Cassa di Risparmio in Bologna and by Progetto Strategico Oncologia n.74 (DD 19Ric, 09/01/02) from Ministero Istruzione Università e Ricerca.