A wide range of hematologic malignancies arises from numerous cell types. In an attempt to offer a new target for treating B-cell malignancies, in this study, the authors tested the possibility of using the CD40/CD40L system as a common targeting system for the various malignancies in this group.
Two chimeric proteins, soluble CD40 ligand (sCD40L)-caspase 3 (sCD40L-l-Caspase3) and sCD40L-pseudomonas exotoxin 38 (PE38) (sCD40L-l-PE38), were constructed, expressed, and partially purified. The ability of the chimeric proteins to kill tumor cells that expressed CD40 was tested by using proliferation assays. In addition, the induction of apoptosis in treated cells was followed by measuring expression levels of apoptotic proteins using real-time polymerases chain reaction analysis, caspase 3 enzymatic activity, and tracking changes in the cell cycle with fluorescence-activated cell-sorting analysis.
The chimeric proteins exhibited concentration-dependent and time-dependent killing ability. The new chimeric proteins had no effect in several carcinoma cell lines that did not express the CD40 receptor. Treating tumor cells with sCD40L-based chimeric proteins led to internalization of the fusion proteins into the cell cytoplasm of B cells. Shortly after treatment, a sharp rise in B-cell chronic lymphocytic leukemia/lymphoma 2 (Bcl2) expression levels occurred. Approximately 36 hours after the initiation of treatment, Bcl2 levels dropped, whereas Bcl2-associated X protein (Bax) expression levels rose, pushing the cells toward apoptosis. Concomitantly, caspase 3 RNA levels rose.
Approximately 55% of hematologic malignancies are lymphomas, and most are non-Hodgkin lymphoma (NHL). NHL is the fifth most frequently diagnosed cancer in the United States, and approximately 85% to 90% of all NHL cases in the United States and Western Europe are of B-cell origin. These diseases comprise a heterogeneous group of malignancies with differing patterns of behavior, responses to treatment, and prognosis.
Traditionally, these diseases were treated mainly with chemotherapy and radiotherapy, which are nonspecific treatments that damage both malignant and healthy cells and cause severe side effects. Recently new, targeted therapeutic agents, namely, monoclonal antibodies, have become increasingly significant in treating hematologic malignancies. The most prominent of these antibodies are rituximab, which targets the B-cell–specific CD20 antigen and, thus, is used for treating a wide range of B-cell malignancies,1 and alemtuzumab, which targets the CD52 antigen expressed on B and T lymphocytes, which is used to treat chronic lymphocytic leukemia.2 Monoclonal antibodies may induce cell death through several mechanisms.3 Several of these antibodies also have been conjugated with a radioactive isotope, thus producing high levels of radiation locally in the tumor environment.1 Concomitantly, several new classes of drugs with different mechanisms of action have evolved, including tyrosine kinase inhibitors (eg, imatinib, dasatinib), histone deacetylase inhibitors (vorinostat), demethylating agents (azacitidine), immunomodulators (lenalidomide), and proteasome inhibitors (bortezomib). These drugs can be used alone or in combination with chemotherapy.4
In this study, we offer another treatment approach for B-cell malignancies: targeted chimeric proteins. Chimeric proteins are composed of 2 distinct protein moieties that are fused at the cDNA level. The first moiety is the targeting moiety, which is designed to target the chimeric protein to the tumor cell by binding a specific antigen presented on the cell's surface. Once bound, the chimeric protein is internalized into the cell's cytoplasm, thus enabling the second moiety—the killing moiety—to exert its action. Classically, the killing moiety was based on plant or bacterial toxins. Because these toxins are nonhuman proteins, they evoke the production of neutralizing antibodies in the human body. Furthermore, these chimeric proteins have a nonspecific, concentration-dependent activity. These problems limit the dosage levels and prevent repeat administration. In the last few years, to overcome these problems, we have been using human proapoptotic proteins as the killing moiety, which cause the target cells to die by apoptosis. The proapoptotic proteins are from the B-cell chronic lymphocytic leukemia/lymphoma 2 (Bcl2) family (Bcl2-associated X protein [Bax], Bcl2 homologous antagonist killer [Bak], and Bcl2-interacting killer [Bik])5-8 as well as the caspase-activated DNase DNA fragmentation factor, 40 kDa, beta polypeptide (DFF40),9 and caspases like as Caspase 3.10-12 Caspase 3 is a key protease in apoptosis, acting on multiple substrates within the cell. Its activity is especially significant in lymphoid cells. In addition, because caspase 3 acts relatively late in the apoptosis process, its use can partially circumvent the cellular apoptotic process, which is defective in many cancers.
In this study we chose to target malignant B lymphocytes. To do so, we used a characteristic antigen expressed by B cells—CD40. CD40 is a 48-kDa type 1 transmembrane cell surface receptor from the tumor necrosis factor (TNF)-receptor family. CD40 initially was identified in B lymphocytes, in which it is expressed during all stages of development and maturation. Its activation triggers numerous key processes.13 Further research demonstrated CD40 expression in a much broader pattern, both within the hematopoietic system, in monocytes, dendritic cells, platelets; and outside of it, in endothelial cells and fibroblasts. CD40 also is expressed on the cell surface of other malignant cells.14, 15
The natural ligand of the CD40 receptor is CD40L (CD154), a type 2 transmembrane protein from the TNF superfamily. It is expressed mainly on activated CD4+ T cells, but it also can be identified on other cell types, such as basophils, eosinophils, macrophages, natural killer (NK) cells, platelets, epithelial cells, and endothelial cells. Full-length CD40L is a 33-kDa transmembrane protein constructed from 261 amino acids. CD40L forms symmetric homotrimers on the cell membrane.16
Apart from the full-length transmembrane protein, CD40L is also expressed in several shorter, stable, soluble forms (sCD40L). These shorter forms, 31 kDa, 18 kDa, and 14 kDa in length, retain both the trimeric structure of the full-length protein and its biologic functions,14 enabling CD40L to function as a soluble cytokine. The existence of stable and biologically active, soluble forms allowed us to use sCD40L as a targeting moiety for our chimeric protein. Normally, ligation of membrane-bound CD40L to CD40 induces internalization of the entire complex to the T cell through CD40L.17 However, for our chimeric protein to function, it must internalize into the B cell through CD40. There are some data supporting the ability of CD40 to internalize into B cells post-CD40L binding,18-20 leading us to believe that a mechanism for CD40 internalization does exist in lymphocytes, which could be exploited for our purposes.
Two chimeric proteins, sCD40L-l-Caspase3 and sCD40L-pseudomonas exotoxin 38 (PE38) (sCD40L-l-PE38), were constructed, expressed, and partially purified. We tested the proteins' ability to kill tumor cells that expressed CD40. The chimeric proteins exhibited specific concentration-dependent and time-dependent killing ability. Death induced by the sCD40L-based chimeric protein was by apoptosis. Our results also validated, for the first time, the ability of sCD40L to serve as a direct delivery system for targeted molecules.
MATERIALS AND METHODS
Construction of the Coding Plasmids
Construction of the pRK02 and pRK03 plasmids encoding sCD40L-l-PE38 and sCD40L-l-Caspase3, respectively
To obtain the sCD40L (amino acids [aa] 108-261) sequence, total RNA was isolated from phytohemagglutinin (PHA)-activated, healthy human T lymphocytes using the TriPure Isolation Reagent (Roche, Manheim, Germany) then reverse transcribed into cDNA using a reverse transcription system (Reverse-iT First Strand Synthesis Kit; ABgene, Epsom, United Kingdom). The sCD40L sequence (aa 108-261), flanked by the NdeI and KpnI restriction sites, was generated by polymerase chain reaction (PCR) using total T-lymphocyte cDNA and the following synthetic oligonucleotide primers: 5′-G GAA TTC CAT ATG GAA AAC AGC TTT GAA-3′ (sense), and 5′-GG GGT ACC GAG TTT GAG TAA GCC-3′ (antisense). The PCR product was digested with NdeI and KpnI restriction enzymes.
A plasmid encoding the gonadotropin-releasing hormone (GnRH)-l-PE38 sequence that pre-existed in the laboratory was digested using NdeI and KpnI restriction enzymes, thus removing the GnRH coding sequence. This vector and the digested sCD40L fragment (described above) were ligated to produce the pRK02 plasmid encoding the sCD40L-l-PE38 chimeric protein.
A plasmid encoding GnRH-l-Caspase310 was digested using NdeI and KpnI restriction enzymes, thus removing the GnRH coding sequence. The sCD40L-digested fragment (see above) was ligated into the vector 5′ to the l-Caspase3 coding sequence (lacking the prodomain; aa 29-277), producing the pRK03 plasmid that encodes sCD40L-l-Caspase3. The plasmid encoding the anti-CD22 recombinant immunotoxin (RFB4[Fv])-PE38 chimeric protein was obtained from D.J. Fitzgerald (National Institutes of Health, Bethesda, Md).
Expression and Partial Purification of the Chimeric Proteins
After the transformation of plasmids into the Escherichia coli (E. coli) strain BL21 (λDE3), cells were grown in LB medium supplemented with 50 μg/mL ampicillin at 37°C. At an optical density at a 600-nm wavelength of 0.6 to 0.9, protein expression was induced using 1 mM isopropyl-1-thio-D-galactopyranoside (IPTG). After a 2-hour incubation period at 37°C, the cells were centrifuged for 15 minutes at ×3150 g, and the pellet was stored at −70°C overnight. The frozen cells were thawed and suspended in lysis buffer (20 mM Tris-HCl, pH 8.0; 1 mM ethylene diamine tetra acetic acid [EDTA]; 0.2 mg lysosyme per 1 mL; 1 mM phenylsullfonyl fluoride [PMSF]), then further lysed using sonication. The lysed cells were then centrifuged at ×35,000 g for 30 minutes. The supernatant (soluble fraction) was removed, and the pellet (containing inclusion bodies) was suspended in denaturation buffer (8 M urea; 20 mM Tris-HCl, pH 8.6; 1 mM EDTA; 0.05 M NaCl; 10 mM dithiothreitol [DTT]) and continuously stirred for at least 1 hour. The solution was cleared by centrifugation at ×35,000 g for 15 minutes. The denaturated proteins were diluted 1:100 in refolding buffer (20 mM Tris-HCl, pH 8.0; 1 mM EDTA; 0.25 M NaCl; 0.25 M L-arginine; 1 mM glutathione disulfide; 1 mM glutathione). Finally, the refolded protein solutions were dialyzed against phosphate-buffered saline (PBS), aliquoted, and stored at −20°C.
Western Blot Analysis
Samples were separated on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels. The proteins were then electrotransferred onto Immobilon-P transfer membrane (Millipore, Millipore, Bradford, Mass) and blotted with anti-caspase 3 (Dako, Glostrup, Denmark; 1:10,000 dilution), anti-CD40L (Santa Cruz Biotechnology Inc., Santa Cruz, Calif; 1:15,000 dilution), anti-pseudomonas exotoxin (anti-PE) (produced by the laboratory; 1:10,000 dilution), anti-Bcl2 (Dako, Glostrup, Denmark; 1:250 dilution), anti-Bax (Santa Cruz Biotechnology Inc.; 1:5000 dilution), anti-Bik (Santa Cruz Biotechnology Inc.; 1:5000 dilution), or anti-α-tubulin (Serotec, Oxford, United Kingdom; 1:20,000 dilution). Band visualization was achieved using an enhanced chemiluminescence kit (ECL; Biological Industries, Beit Ha'emek, Israel).
Blue Native Polyacrylamide Gel Electrophoresis
To evaluate the native form of sCD40L-l-Caspase3 chimeric protein, we denatured the chimeric protein by boiling it in 1% SDS and 50 mM DTT for 5 minutes followed by dialysis against PBS. Protein samples were mixed with the sample buffer containing Coomassie blue G-250. The native and denatured chimeric protein preparations were run on blue native-PAGE gels (4%-15% volume/volume), as described by Schagger and Von Jagow,21 at natural pH, 90V, and 4°C for 4 hours.
Human B-lineage cells (kindly provided by Hanna Ben-Bassat; Hadassah Medical Center, Jerusalem, Israel) were grown in RPMI medium supplemented with 20% heat-inactivated fetal calf serum (FCS), 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin. MCF-7 and LNCaP cells were grown in the same medium supplemented with 10% FCS. The cell lines 293 and A204 were grown in Dulbecco modified Eagle medium supplemented with 10% FCS. The cells were grown in a highly humidified atmosphere of 5% CO2/95% air at 37°C. All cell cultures were tested for Mycoplasma contamination and were negative (EZ-PCR Mycoplasma Test Kit; Biological Industries).
Receptor Expression Level Analysis
Cells (0.5 × 106) from each cell line were suspended in 50 μL fluorescence-activated cell-sorting (FACS) analysis medium (3% FCS and 0.02% sodium azide in PBS). The cells were stained with fluorescein isothiocyanate (FITC)-conjugated anti-CD40 antibody (Serotec; 1:100 dilution) by incubation for 1 hour on ice without prior fixation or permeabilization. The cells were then washed 3 times with PBS and analyzed using a FACScan immunocytometry system (Becton Dickinson, San Jose, Calif) and the CellQuest software program (Becton Dickinson).
Cells growing in suspension (104/100 μL per well) or adherent cells (5 × 103/100 μL per well) were seeded and treated with increasing concentrations of the chimeric proteins for 72 hours, after which, cellTiter-Blue reagent (Promega, Madison, Wis) was added according to the manufacturer's instructions to determine cell survival. All treatments were performed in triplicate.
Internalization of Chimeric Proteins Into Receptor-Expressing Cells
Internalization of chimeric proteins into receptor-expressing cells analyzed by Western blot analysis
Bjab or Daudi cells (3 × 106/1.7 mL) were treated with 35 μg/mL sCD40L-l-Caspase3, 24 μg/mL RFB4(Fv)-PE38 or 24 μg/mL sCD40L-l-PE38 for 4.5 hours, then collected and washed 3 times with PBS. Next, the cells were treated with acid solution (0.01 M citric acid, pH 2.0; 0.14 M NaCl) for 2 minutes at room temperature; then, 100 μL of 1.5M Tris-HCl buffer, pH 8.8, were added. Finally, the cells were washed 3 more times with PBS. To extract intracellular proteins, cells were treated with lysis buffer (20 mM NaH2PO4, 250 mM NaCl, 30 mM Na4P2O710H2O, 0.1% NP-40, 5 mM EDTA, 1 mM DTT, and 1 mM PMSF) for 10 minutes on ice, then centrifuged at ×15,300 g for 10 minutes at 4°C. The supernatant was separated from the pellet, and total protein content was determined using a Bradford assay. The gels were blotted with rabbit anti-caspase 3 antibodies (Dako; 1:3000 dilution) or with rabbit anti-PE antibodies (1:7500 dilution). The result was observed by ECL, as described above.
Kinetics of CD40L-l-PE38 internalization demonstrated by confocal microscopy
Bjab cells (2 × 105/2 mL) were treated with 24 μg/mL sCD40L-l-PE38 for 2 hours, 4 hours, and 6 hours; washed twice with cold PBS; fixated with 3.7% formaldehyde for 10 minutes; and washed twice again. Next, the cells were treated with permeabilization solution (0.2% Triton-X 100 and 1% bovine serum albumin [BSA] in PBS) for 5 minutes and washed twice. To block unwanted, nonspecific interactions, the cells were treated with 1% BSA in PBS for 20 minutes. sCD40L-l-PE38 protein was stained with rabbit anti-PE antibody (1:50 dilution) and observed with secondary indocarbocyanine (Cy5)-conjugated antirabbit antibody (Jackson ImmunoResearch Laboratories Inc., West Grove, Pa; 1:200 dilution). Finally, the cells were cytospun onto slides and covered with mounting buffer containing 4′,6-diamidino-2-phenylindole (DAPI). The resulting slides were analyzed with a Zeiss LSM410 confocal laser-scanning microscope (Zeiss, Oberkochen, Germany) using an HeNe laser (excitation 633 nm/emission 660 nm), with a planapochromal ×63 oil lens at ×2 magnification.
Polymerase Chain Reaction Analysis of Apoptosis-Related mRNAs
Semiquantitative polymerase chain reaction analysis of apoptosis-related mRNAs in untreated cells
Total RNA was extracted from untreated cells and reversed-transcribed into cDNA as described above. Specific mRNAs were amplified by PCR using the ReddyMix Master Mix (Thermo Scientific, Waltham, Mass) and a T-Gradient Thermocycler (Whatman Biometra, Gottingen, Germany) with the following primers: Bcl2, 5′-GTG CCA CCT GTG GTC CAC CTG-3′ (sense) and 5′-ctt cac ttg tgg ccc aga tag g-3′ (421 base pairs [bp]; antisense); Bax, 5′-CGG AAT TCA AGC TTT GGA CGG GTC CGG GGA-3′ (sense) and 5′-CGG AAT TCA GGT CCT TCA GCC CAT CTT CTT C-3′ (576 bp; antisense); glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 5′-CCA CCC ATG GCA AAT TCC ATG GCA-3′ (sense) and 5′-TCT AGA CGG CAG GTC AGG TCC ACC-3′ (596 bp; antisense); and β-actin, 5′-act ctt cca gcc ttc cttc-3′ (sense) and 5′-agt gat ctc ctt ctg cat cc-3′ (200 bp; antisense).
Real-time polymerase chain reaction analysis of apoptosis-related mRNAs in sCD40L-l-Caspase3–treated cells
Bjab cells (106/4 mL) were treated with 20 μg/mL sCD40L-l-Caspase3 for 3 to 48 hours. Total mRNA was extracted and reverse-transcribed into cDNA as described above. Individual mRNA levels were quantified using Real-Time PCR (Applied Biosystems, Foster City, Calif). Each 2-μL sample contained 1 μL primers (10 ng), 10 μL SYBR Green (Applied Biosystems), and 7 μL H2O in a total volume of 20 μL per sample. The following primers were used: Bax (exons 4-5, 116 bp), TCT GAC GGC AAC TTC AA CTG (sense) and CAG CCC ATG ATG GTT CTGA (antisense); Bcl2 (exons 2-3, 134 bp), CCC CTG GTG GAC AAC ATC (sense) and CAG CCA GGA GAA ATC AAA CAG (antisense); caspase 3 (exons 7-8, 133 bp), GAA CTG GAC TGT GGC ATT GA (sense) and CCT TTG AAT TTC GCC AAG AA (antisense); and glucose-6-phosphate dehydrogenase (G6PD) (exons 6-7, 283 bp), TCT ACC GCA TCG ACC ACT ACC (sense) and GCG ATG TTG TCC CGG TTC (antisense). The data were analyzed by primer express program (Applied Biosystems).
In Vitro Caspase 3 Activity Assay
Bjab cells (105/100 μL per well) were treated with the chimeric protein for 12 to 48 hours. Caspase 3 activity within the cells was assessed by using the Apo-ONE Homogeneous Caspase 3/7 Assay Kit (Promega). Experiments were carried in parallel with cell viability assays.
Cell Cycle Analysis by Propidium Iodide Staining
Bjab and Daudi cells were treated with 35 μg/mL sCD40L-l-caspase3, for 12 to 48 hours. Samples were removed for assaying cell viability, and the remaining cells were centrifuged at ×500 g for 6 minutes at 4°C, washed with cold PBS, and resuspended in 0.5 mL of propidium iodide (PI) hypotonic solution (50 μg/mL PI, 0.1% sodium citrate, 0.1% Triton-X 100).22 After overnight incubation at 4°C, cell cycle analysis of the cells was performed using the FACScan system and the CellQuest software program (Becton Dickinson).
Construction, Expression, Partial Purification, and Characterization of the sCD40L-Based Chimeric Proteins
Our objective was to design chimeric proteins capable of specifically targeting and eliminating B lymphocytes that expressed the CD40 receptor. Therefore, as a killing moiety, we chose to use the human proapoptotic protein caspase 3. For the CD40L component of the chimeric protein, we chose 1 of the soluble forms of the protein, the sCD40L (aa 108-261) 18-kDa form, known as the most active,14 as the targeting moiety for our chimeric protein. To our knowledge, sCD40L has never been used before as part of a direct targeting system for chimeric proteins. Thus, to validate the sCD40L as a new delivery system, we first constructed the sCD40L-l-PE38 plasmid. The bacterial toxin PE is established as a very efficient killing moiety. It is used in numerous chimeric proteins, fused to various targeting moieties.23, 24 We also constructed the pRK03 plasmid encoding sCD40L-l-Caspase3 by using human caspase 3 (which lacks the prodomain; aa 29-277) as a killing moiety. Figure 1A illustrates the schematic structure of the new chimeric proteins.
The chimeric proteins were expressed in the E. coli BL21 (λDE3) expression system, and various subcellular fractions were prepared, separated by SDS-PAGE, and characterized both by Coumassie blue staining and Western blot analysis. Figure 1B,E indicates that Coumassie blue staining revealed major bands with estimated masses corresponding to those of the overexpressed proteins: 47.1 kDa for sCD40L-l-Caspase3 and 54.8 kDa for sCD40L-l-PE38. The newly designed chimeric proteins were identified mainly in the inclusion bodies fraction and constituted the majority of the inclusion body proteins (Fig. 1B,E).
When tested by Western blot analysis, sCD40L-based proteins reacted both with the anti-CD40L antibody (Fig. 1C,F) and with an antibody against the killing moiety (anti-caspase 3 for sCD40L-l-Caspase3 and anti-PE for sCD40L-l-PE38) (Fig. 1D,G, respectively), yielding major bands at the expected sizes. These analyses confirmed the cloning and production of the in-frame, full-length chimeric proteins.
It should be noted that the natural active form of CD40L is a homotrimer of 3 subunits. However, in our construct, we used a monomer. We assumed that, during the protein's refolding process, separate molecules would fold together to form the active trimeric protein. Because the fused killing moiety may cause a steric disturbance to the refolding process, thus preventing trimer formation, we separated the sCD40L moiety from the PE moiety by adding a 15-aa flexible linker, similar to that separating the 2 parts of recombinant antibodies produced as single-chain antibodies in a bacterial system.25
To evaluate the tertiary structure of sCD40L-l-Caspase3, we ran a blue native-polyacrylamide gel electrophoresis (BN-PAGE) and blotted it with anti-caspase 3 antibodies. We observed that the native form of sCD40L-l-Caspase3 is a high-molecular-weight protein, most probably a homotrimer. Upon denaturing the chimeric protein preparation (boiling the chimeric protein in 1% SDS and 50 mM DTT for 5 minutes) to dissociate higher protein forms, the chimeric protein runs as a low-molecular-weight protein (results not shown). However, additional experiments should be performed once a highly purified sCD40L-l-Caspase3 protein preparation is obtained to evaluate the exact tertiary structure of the chimeric protein.
Internalization of the Chimeric Proteins Into Receptor-Expressing Cells
To kill targeted cells, both caspase 3 and PE38 must find their way into the cells' cytoplasm. When using these constructs, it is up to the targeting moiety both to bind the specific receptors on the cells' surface and to activate the internalization mechanisms that eventually will lead the chimeric proteins into the cytoplasm. There is only indirect evidence supporting the ability of sCD40L to perform this task.
To test our hypothesis regarding internalization of sCD40L into CD40-expressing B cells, Bjab and Daudi cells, known to express the CD40 receptor (see below), were incubated with each of the 2 chimeric proteins separately for 4.5 hours at 37°C, allowing the chimeric proteins to bind the appropriate receptors and to enter the cells. Cells were then thoroughly washed and treated with acid to remove any nonspecifically bound proteins from the cell surface and the surrounding solution. Figure 2A-C illustrates Western blot analyses of the treated cell extracts: sCD40L-based chimeric proteins are detectable in the cytoplasm of treated cells after 4.5 hours of incubation at the expected molecular weights.
Next, we incubated Bjab cells with sCD40L-l-PE38 for 2 hours, 4 hours, and 6 hours. After this treatment, we were able to demonstrate sCD40L-l-PE38 internalization into target cells using 2 approaches: 1) Western blot analysis, as demonstrated in Figure 2D, in which sCD40L-l-PE38 is detectable within the treated cells as early as 2 hours after treatment initiation; and 2) confocal microscopy. Figure 2E indicates that sCD40L-l-PE38 is detected within the cells as early as 2 hours after initiation of treatment, similar to the results from Western blot analysis. Because human cells do not express PE, any PE detected within the cells by this method must originate from the internalized sCD40L-l-PE38 chimeric protein. These experiments directly demonstrate for the first time the ability of soluble CD40L and sCD40L-based chimeric proteins to induce CD40 internalization into B cells upon its binding, thus establish the feasibility of using the CD40-CD40L system as a new and efficient targeting system for B-cell malignancies.
Effect of the Chimeric Proteins on Cell Viability
Killing is specific, dose-dependent, and time-dependent
The chimeric proteins were tested for their ability to kill target cells. Bjab cells were treated with PBS (control), sCD40L-l-Caspase3 (Fig. 3A), or sCD40L-l-PE38 (Fig. 3B) in increasing doses for 72 hours. The chimeric proteins induced a linear, dose-dependent decrease in the cells' viability. In this experiment, as a positive control, we used the well tested chimeric protein RFB4(Fv)-PE38.26-28 Because Bjab cells do not express the GnRH receptor, we used an unrelated protein, GnRH-PE66,29 as a negative control. GnRH-PE66 had no effect on the cells' viability, as expected (Fig. 3A,B).
Next, Bjab cells were treated with 1 of the chimeric proteins at 1 selected concentration, and viability was assessed every 12 hours for up to 72 hours. Figure 3C,D demonstrates the kinetics of killing of the 2 chimeric proteins. sCD40L-l-Caspase3 (Fig. 3C) or sCD40L-l-PE38 (Fig. 3D) caused very rapid cell death. The maximal effect was achieved at 36 hours, when the percentage of live cells roughly stabilized at a low of 14% and 19%, with a 50% inhibitory concentration (IC50) of 23.5 μg/mL and 3.4 μg/mL, respectively. Both chimeric proteins, sCD40L-l-Caspase3 and sCD40L-l-PE38, induced a similar linear, dose-dependent and time-dependent death when added to Daudi CD40-receptor expressing cells (see Fig. 4B), representing the blast I maturation stage of B-cell differentiation, with IC50 values of 25 μg/mL and 4.2 μg/mL, respectively (results not shown). In addition, to assess the specificity of the proteins' activity, we tested their ability to induce death in cells that lacked expression of CD40, namely, 293 renal cell carcinoma cells, A204 rhabdomyosarcoma cells, MCF7 breast adenocarcinoma cells, and LNCaP prostate adenocarcinoma cells (Fig. 3F). The tested cell lines did not respond to any of the newly designed chimeric proteins (Fig. 3E). GnRH-PE6629 was used as a positive control for 293, MCF7, and LNCaP cells (Fig. 3E).
Effect of the Chimeric Proteins on B Lymphocytes at Different Stages of Development
sCD40L-based chimeric proteins specifically killed the Burkitt lymphoma Bjab cell line, representing the B-blast maturation stage, as demonstrated previously. However, B-cell–derived hematologic malignancies arise from B lymphocytes at all maturation stages. To test the chimeric proteins' ability to target and treat all such malignancies, we tested their activity on several cells lines, representing malignancies arising from different B-cell maturation stages: Km-3 non-T/non-B ALL cells, Nalm-6 pre-B ALL cells, Bjab Burkitt lymphoma cells (B blast), ARH-77 myeloma cells, and LAM B-cell follicular lymphoma t(14;18) cells.
The various cells were treated with increasing doses of sCD40L-l-PE38 for 72 hours, and the cells' viability was measured. The results in Figure 4A illustrate the different responses of the various cell lines to treatment with the chimeric protein. The young non-B/non-T Km-3 cells as well as pre-B Nalm-6 cells had no response even to relatively high levels of sCD40L-l-PE38. Even relatively low amounts of sCD40L-l-PE38 induced death in 80% to 90% of B-blast Bjab cells, as observed previously. Plasma ARH-77 cells had a mild response to treatment with sCD40L-l-PE38; with an IC50 of 22 μg/mL. LAM follicular lymphoma t(14;18) cells had virtually no response to treatment with sCD40L-l-PE38, even at relatively high doses.
Characterization of B Lymphocytes at Different Stages of Development
The response of cells to treatment with sCD40L-based chimeric proteins is likely to be influenced by the level of receptor expression on the cell surface. In addition, because we assumed that our chimeric proteins would induce apoptosis in treated cells (see below), the ratio of proapoptotic and antiapoptotic protein expression within the cells is important. Thus, these 2 factors were studied next.
We used FACS analysis to verify and quantify CD40 expression on the cell surface of cell lines that represented various differentiation stages using FITC-conjugated, anti-CD40 antibodies. It is demonstrated in Figure 4B that that non-T/non-B Km-3 cells express very low levels of CD40. Pre-B Nalm-6 cells do not express detected levels of CD40. B-blast Bjab cells express high levels of the receptor (97%). B-blast Daudi cells have a similar expression pattern. The plasma cell line ARH-77 also expresses relatively high levels of the receptor (97%), whereas LAM cells express relatively low levels of the receptor.
Characteristics of B cells at different stages of maturation
B cells, as they mature, undergo substantial changes, which influence and adjust their response to internal and external death stimuli. One of the mechanisms involved is changes in expression levels of proapoptotic and antiapoptotic proteins. Thus, these changes may be significant to the response of cells to treatment with our chimeric proteins.
To assess these changes we examined the level of expression, at the mRNA and protein levels, of both proapoptotic proteins (Bax and Bik) and of an antiapoptotic protein (Bcl2), in cell lines representing sequential stages of B-cell maturation. Total mRNA and proteins were extracted from untreated B cell lines at sequential stages of development. Reverse transcriptase-PCR and Western blot analyses were used to assess the levels of mRNA and proteins.
It is clear from Figure 5A,B that immature cells (the non-B/non-T Km3 or Reh cells and the pro-B Nalm-6 cells) express high levels of Bcl2; and, as the cells mature into blast cells (Daudi and Bjab cells), Bcl2 expression drops dramatically to almost undetectable levels. Further maturation into plasma cells (ARH-77 myeloma cells) is followed by an increase in Bcl2 expression levels. LAM cells that carry the t(14;18) translocation express high levels of Bcl2, as expected. These changes are observed both at the mRNA (Fig. 5A) and protein (Fig. 5B) levels.
Figure 5C,D indicates that, in contrast to Bcl2, Bax expression levels in young cells (Km-3/Reh and Nalm-6 cells) is low. Bax levels rise as cells mature into blast cells and remain at the same level throughout maturation into plasma cells. LAM cells also express high levels of Bax. Like Bax levels, the levels of the proapoptotic Bik protein are low in very young cells and rise gradually through maturation into the blast stage. However, unlike Bax, as cells mature into plasma cells (ARH-77 cells), Bik expression levels are decreased again. LAM cells express high levels of Bik. (Fig. 5D).
Effect of Chemotherapy on Cells at Various Stages of Maturation
To further explore the relation between the expression of proapoptotic and antiapoptotic proteins and the tendency of cells to undergo apoptosis, we treated the various cell lines with 2 well tested apoptosis-inducing chemotherapeutic drugs—doxorubicin30 and vinblastine.31 The cell lines were treated with either doxorubicin (3 μg/mL for 6 hours) or vinblastine (1 μg/mL for 24 hours); then, their viability was assessed using a cell viability assay (Fig. 5F,G). Table 1 summarizes the link between the expression of proapoptotic and antiapoptotic proteins and the response to treatment with doxorubicin for each cell line.
Table 1. The Relation Between Antiapoptotic and Proapoptotic Proteins Levels and Response to Chemotherapya
aThis table summarizes the expression levels of proapoptotic and antiapoptotic proteins in the various cell lines, representing sequential stages of B-lymphocyte maturation, and the correlation between the antiapoptotic:proapoptotic ratio and the response to treatment with doxorubicin.
The results indicate that, as the antiapoptotic to proapoptotic protein ratio increases, cells become less sensitive to treatment with doxorubicin. A similar trend was observed with vinblastine. Because we assumed that our chimeric proteins would induce apoptosis in treated cells (see below), we hypothesized that the ratio of antiapoptotic to proapoptotic proteins also would influence the response of cells to treatment.
Summarizing the cells' response to treatment with sCD40L-l-PE38 (Fig. 4) in light of the parameters of receptor expression and the antiapoptotic/proapoptotic ratio (Table 1), we concluded that, as the receptor density increased and the antiapoptotic/proapoptotic ratio decreased, sCD40L-l-PE38 treatment had a greater effect on the cells, which was the case for B-blast Bjab and Daudi cells. In contrast, when receptor density was low and the antiapoptotic/proapoptotic ratio was high, as in non-T/non-B KM-3 cells, there was a low response to treatment with our chimeric proteins.
Effect of sCD40L-l-Caspase3 Chimeric Proteins on Cellular Mechanisms: Death by Apoptosis
To gain some insight regarding the mechanism of action of sCD40L-l-Caspase3 within target cells, we explored the changes that occurred in the cellular apoptotic proteins both at the mRNA level by real-time PCR and at the protein activity level by measuring caspase 3 enzymatic activity.
Effect of CD40L-l-Caspase3 on the expression levels of apoptotic proteins
We examined the effect of the sCD40L-l-Caspase3 chimeric protein on the expression levels of apoptotic proteins using real-time PCR. Cellular mRNA levels of Bcl2, Bax, and caspase 3 increased in response to treatment, as indicated in Figure 6. This increase was dramatic for Bcl2-mRNA levels (>90-fold), which peaked at 36 hours and then declined. The increases in both Bax and caspase 3 mRNA levels were much less (from 2.0-fold to 2.5-fold greater than the increases in control, untreated cells) (Fig. 6A-C).
Effect of the chimeric proteins on cellular caspase 3 enzymatic activity
First, we tested sCD40L-l-Caspase3 for in vitro enzymatic activity using a synthetic substrate. These experiments demonstrated that sCD40L-l-Caspase3 has no caspase 3 activity in a cell-free system (data not shown). However, it should be pointed out that previous experience with another chimeric protein using interleukin 2 (IL-2) produced in the laboratory, namely, IL-2-Caspase3, demonstrated that, although this chimeric protein also had no in vitro caspase 3 enzymatic activity in a cell-free system, it has significant biologic activity on cell lines and in vivo.11, 12 Thus, we assayed the enzymatic activity of caspase 3 within target cells after treatment with our chimeric proteins.
Bjab and Daudi cells were treated with sCD40L-l-Caspase3 for 12 to 48 hours; then, caspase 3 activity within the cells was measured. Simultaneously, identical samples were analyzed for cell viability. Treatment of Bjab cells with sCD40L-l-Caspase3 increased cellular caspase 3 activity in Bjab cells by 12-fold after 12 hours of treatment, the activity plateaued at 24 hours, then it declined after longer exposure to the chimeric protein, as illustrated in Figure 6D. The increase in cellular caspase 3 activity coincided with a decrease in cell viability. Similar results were obtained when treating Daudi cells (data not shown).
Effect of the chimeric proteins on the cell cycle
To demonstrate the effect of our chimeric proteins on the cell cycle, Bjab and Daudi cells were treated with the chimeric proteins for 12 to 48 hours. Bjab cells that were treated with sCD40L-l-Caspase3 had already started forming apoptotic bodies after 12 hours of incubation (Fig. 7A). The percentage of cells in sub-G1 remained constant (15%-18%) throughout the longer incubation periods up to 36 hours (Fig. 7A-E). A similar response was obtained with Daudi cells (Fig. 7F), thus confirming that death induced by CD40L-l-Caspase3 is by apoptosis.
Considerable efforts have been invested in recent years in developing new, effective, yet less toxic drugs for targeted cancer therapy that may either replace current treatment or be added to such treatment to enhance efficacy. Currently, this diverse group of drugs includes monoclonal antibodies, immunomodulators, proteasome inhibitors, tyrosine kinase inhibitors, and deacetylation agents.
Another approach to treating malignancies is the use of chimeric proteins. In the current work, our objective was to develop a chimeric protein to treat B-cell malignancies. We used a new targeting system to direct the human proapoptotic protein caspase 3 to malignant B lymphocytes—the CD40/CD40L system. The resulting chimeric protein, sCD40L-l-Caspase3, was capable of targeting, entering, and efficiently killing cells that expressed CD40 in a dose-dependent and time-dependent manner and caused cell death by apoptosis.
A wide range of hematologic malignancies arises from numerous cell types. The different phenotypes include a unique panel of cell surface molecules characteristic to each disease. On 1 hand, this variety offers a diversity of potential targets for treatment with chimeric cytotoxic proteins. Conversely, it means that numerous chimeric proteins must be developed to match the specific requirement of each disease. In an attempt to provide a new target for treating B-cell malignancies, we tested the possibility of using the CD40/CD40L system as a common targeting system for the different types of malignancies in this group.
One of the major requirements for a cell surface protein to be used as a target for a chimeric cytotoxic protein is the ability to recruit an internalizing mechanism, which will carry the chimeric protein into the cells' cytoplasm. It is well known that binding of CD40, which is located on the membrane of B cells, to membrane-bound CD40L, which is located on the membrane of T cells, leads to internalization by CD40L of the entire complex into the T cell13; however, it was unclear whether a mechanism to internalize CD40 into the B cell also exists. Some findings, including the incorporation of CD40 into glycolipid-rich domains in the cell membrane upon binding to CD40L19 and the induction of internalization when binding an anti-CD40 antibody,20 suggested that such a mechanism may exist and, thus, could be exploited for our purposes. In addition, several targeted molecules have been designed to target CD40, such as dacetuzumab (SGN-40), a humanized monoclonal antibody. Preliminary trials treating NHL, chronic lymphocytic leukemia, and multiple myeloma with dacetuzumab have produced promising results.32, 33 An immunotoxin targeting CD40 (G28-5 sFv-PE40) has demonstrated in vitro cytotoxic activity on CD40-expressing lymphoma cells.34-36
However, the natural ligand of CD40 or 1 of its soluble forms has not previously been used as a direct delivery system to target CD40. The binding of the sCD40L-based chimeric proteins, sCD40L-l-PE38 and sCD40L-l-Caspase3, to CD40 successfully induced internalization and translocation of the chimeric protein into the cells' cytoplasm, as illustrated in Figure 2. This proof-of-principal experiment establishes for the first time the feasibility of using the CD40/CD40L system as a targeting system for malignant B cells. This system can now be used to construct various chimeric cytotoxic proteins, but it is not limited to this modality. It also can be used as a targeting system to deliver radioactive isotopes, liposomal structures carrying chemotherapy agents, adenoviruses carrying therapeutic genes, etc. Moreover, because CD40 is also expressed on various carcinomas, this delivery system is not limited to treating B-cell hematologic malignancies.
CD40 is widely expressed on various normal cells; thus, 1 possible concern is that systemic administration of our newly designed molecules could cause serious side effects. However, its expression on malignant cells is highly up-regulated. Moreover, a few phase 1 clinical trials with the humanized monoclonal antibody dacetuzumab (see above) in patients with multiple myeloma, chronic lymphocytic leukemia, and refractory or recurrent NHL demonstrated that dacetuzumab was well tolerated.37-39 Thus, when treating with sCD40L-based chimeric proteins, serious side effects are not anticipated.
The use of human proapoptotic proteins like caspase 3 as the killing moieties of chimeric proteins has several advantages. First, apoptotic cell death prevents leakage of the cellular content into the surrounding matrix, thus minimizing the inflammatory response caused by the cells' death. Second, although caspase 3 is an intracellular protein and is not usually present in the bloodstream, it is of human origin and, as such, is much less immunogenic than bacterial toxins. The third advantage is that caspase 3 is an intracellular protein without any cell surface receptor that would allow its binding and translocation into the cytoplasm; thus, it is expected to have much less nonspecific activity. Fourth, some hematologic malignancies respond poorly to chemotherapy and radiotherapy because of defective apoptotic mechanisms or because of extremely high levels of antiapoptotic protein expression. In these cases, using chimeric proteins based on the downstream effector caspase 3 may circumvent these obstacles and render the cells more susceptible to chemotherapy and radiotherapy.
These results indicate that, as more cell surface molecules are recognized and used as targets for chimeric proteins therapy, combinations of chimeric proteins or chimeric proteins with monoclonal antibodies as well as with chemotherapy agents can be used. The “multiple weapons” approach using combinations of several chemotherapy agents and monoclonal antibodies is vastly used in oncology and hemato-oncology.
In the current report, we present a new chimeric protein aimed at treating B-cell hematologic malignancies. This chimeric protein, sCD40L-l-Caspase3, is based on the CD40/CD40L targeting system, that has been validated for the first time as an efficient delivery targeting system, and the human proapoptotic protein Caspase 3 as a killing moiety. sCD40L-l-Caspase3 is able to efficiently and specifically target and kill cells that express CD40, thus offering another weapon in the everlasting war against cancer.