|Home | About | Journals | Submit | Contact Us | Français|
Lymphoid enhancer factor-1 (lef-1) is overexpressed in B-cell chronic lymphocytic leukemia (CLL) when compared with normal B cells and transcribes several genes implicated in the pathogenesis of CLL. We therefore hypothesize that antagonism of lef-1 might lead to killing of CLL cells. We used two small molecule inhibitors of Wnt/β-catenin/lef-1 signaling (CGP049090 and PKF115-584) to test our hypothesis.
Enriched CLL cells and healthy B cells were used in this study. Small interfering RNA (siRNA)-mediated knockdown of lef-1 in primary CLL cells was done using nucleofection, and 50% lethal concentration (LC50) of two small molecules was assessed using ATP-based cell viability assay. Apoptotic response was investigated in time course experiments with different apoptotic markers. Specificity of the small molecules was demonstrated by coimmunoprecipitation experiments for the lef-1/β-catenin interaction. In vivo studies were done in JVM-3 subcutaneous xenograft model.
Inhibition of lef-1 by siRNA leads to increased apoptosis of CLL cells and inhibited proliferation of JVM-3 cell lines. The two small molecule inhibitors (CGP049090 and PKF115-584) efficiently kill CLL cells (LC50<1 µM), whereas normal B cells were not significantly affected. Coimmunoprecipitation showed a selective disruption of β-catenin/lef-1 interaction. In vivo studies exhibited tumor inhibition of 69% with CGP049090 and 57% with PKF115-584 when compared with vehicle-treated controls, and the intervention was well tolerated.
We have demonstrated that targeting lef-1 is a new and selective therapeutic approach in CLL. CGP049090 or PKF115-584 may be attractive compounds for CLL and other malignancies that deserve further (pre)clinical evaluation.
The progressive accumulation of mature dysfunctional CD5+, CD19+, and CD23+ B cells due to failed apoptosis is the major pathophysiological feature of B-cell chronic lymphocytic leukemia (CLL) . Whereas previous studies on the apoptotic block in CLL have mainly focused on the B-cell lymphoma-2 (bcl-2) gene family , aberrant wnt signaling has been implicated in the pathogenesis of CLL as well .
The wnt/β-catenin/lef-1 signal transduction pathway has been found to be activated in many types of cancer . Binding of secreted wnt protein to its membrane-bound receptor complex, composed of a member of frizzled receptor (fzd) family and the coreceptor LRP 5/6, leads to inhibition of phosphorylation of β-catenin by glycogen synthase kinase (gsk-3β). Unphosphorylated β-catenin remains stable, accumulates in the cytoplasm, and translocates into the nucleus, where it activates target gene expression through interaction with the transcription factors T-cell factor (tcf) and lymphoid enhancer factor-1 (lef-1) [4,5].
This tcf/lef-1/β-catenin signaling regulates the expression of c-myc, cyclin D1 [6–8], and many other target genes involved in the regulation of leukemic cell adhesion, B-cell proliferation, and survival [9–11]. In CLL, the messenger RNA (mRNA) levels of 6 of 18 wnts (wnt-3, wnt 5b, wnt-6, wnt-14, and wnt-16) and fzd-3 were clearly elevated in CLL compared with those in peripheral blood lymphocytes and normal B cells . Lef-1 is a nuclear protein preferentially expressed in pre-B cells but not in mature B cells . Most interestingly, three independent studies showed that lef-1 was overexpressed (~3000-fold) in CLL cells when compared with that in normal B lymphocytes [13–15]. A previous study on β-catenin expression in hematopoietic cells indicates that normal T lymphocytes do not express β-catenin, whereas several leukemia/lymphoma cell lines and primary leukemia cells have an increased expression . Altogether, β-catenin/lef-1 signaling is active in CLL, which might prevent apoptosis and thus contribute to the extended life span of CLL cells.
Recently, 7000 purified natural compounds from proprietary and public collections were screened, and two small molecules, PKF115-584 and CGP049090, were identified, which specifically inhibit β-catenin/tcf/lef-1 interaction in the wnt signaling pathway. The potency of these compounds was demonstrated by several in vitro assays and reporter gene activation . The aim of this study was to determine the role of lef-1 in the survival of CLL cells and whether knocking down of this potent transcription factor would be of any functional relevance.
MEC-1 and JVM-3 cell lines were obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany) and maintained in RPMI-1640 with 20% fetal bovine serum and penicillin/streptomycin (Biochrom, Berlin, Germany). With informed consent, peripheral blood was taken from patients with CLL during routine diagnostic phlebotomy. Samples were collected into heparinized tubes from patients treated at our institution. B cells included in the current study were enriched using RosetteSep (Stem Cells, Vancouver, Canada). The study was conducted according to the Declaration of Helsinki, 6th Revision (2008; World Medical Association, Seoul/South Korea) and approved by the local ethics committee at the University of Cologne (approval no. 04-231).
CGP049090 and PKF115-584 were obtained from Novartis Pharmaceuticals, Inc (Basel, Switzerland). They were dissolved completely in 70% DMSO in a stock concentration of 3.3 mM and stored in aliquots at -20°C. For in vivo studies, the inhibitors were first dissolved in one part of ethanol and then mixed in one part Cremophor EL and nine parts of sterile water.
CLL cells (8 x 106), resuspended in 100 µl of Cell line Solution Kit V (Amaxa, Cologne, Germany) with 0.5 µM of ON-TARGETplus SMARTpool lef-1 small interfering RNA (siRNA) or ON-TARGETplus siCONTROL nontargeting pool as negative control (Dharmacon, Lafayette, CO), were transfected with the Amaxa Nucleofector I device (program U-013), cultured in six-well plates in complete medium for 16 hours and then examined for cell viability and lef-1 protein expression by Western blot analysis.
Cell lines, primary CLL cells, and rosetted B cells from healthy volunteers were seeded in 96-well plates at 2 x 105 cells per well and treated with various concentrations (0.01, 0.1, 1, and 10 µM) of the inhibitors for 24 hours including vehicle (DMSO) and untreated control. Cell viability was measured using CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI) according to the manufacturer's instructions. Results were generated from three independent experiments.
Detection of apoptosis by flow cytometry (BD FACS Canto, San Jose, CA) was determined using Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit (BD Pharmingen, San Diego, CA). Annexin/PI-negative cells were used to denote the percent surviving populations of cells.
Total RNA from 5 x 106 freshly isolated CLL cells was extracted using the QIAmp RNA Blood mini kit (Qiagen GmbH, Hilden, Germany) and converted into first-strand complementary DNA using SuperScript III First-Strand System (Invitrogen, Carlsbad, CA) according to manufacturer's instructions. Transcript levels were quantified in duplicate by real-time reverse transcription-polymerase chain reaction (RT-PCR) on the Roche LightCycler system using the primer sets and reaction were conditions described in Supporting Information. The relative mRNA levels were assessed by normalizing the obtained fluorescence data of the target gene by those of the housekeeping gene c-abl (ΔΔCt method).
CLL samples (n = 3) were seeded into 96-well plates at 2 x 105 cells per well and treated with 5 µM of the inhibitors at various time points (0, 2, 4, 6, and 8 hours) including vehicle (0.2% DMSO) and negative control. Caspase activity was determined using Caspase-Glo 3/7 Assay (Promega) according to the manufacturer's instructions.
Primary CLL samples were incubated with 5 µM of the inhibitors, and the JVM-3 cell lines were incubated with 10 µM of inhibitors for immunoblot analysis. Antibodies targeting human β-catenin, cyclin D1, β-actin, caspase 8, and PARP-cleaved form were obtained from BD Biosciences (San Jose, CA). Anti-human c-myc antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against human bid; caspases 3, 7, and 9; and human lef-1 were obtained from Cell Signaling Technology (Danvers, MA). Anti-human dephospho-β-catenin was obtained from Axxora Life Sciences (San Diego, CA). Mouse IgG1 κ isotype control antibody was obtained from eBioscience (San Diego, CA). BD ApoBlock Caspase Inhibitor was obtained from BD Biosciences. Cytochrome c Apoptosis Detection Kit was obtained from Promocell GmbH (Heidelberg, Germany).
Total cell protein extracts were prepared in mammalian protein extraction reagent (M-PER; Thermo Scientific, Waltham, MA) with complete protease inhibitor cocktail tablets (Roche Diagnostics, Mannheim, Germany) by ultrasonication performed on ice. Nuclear and cytoplasmic fractionation made using nuclear extract kit (Activemotif, Carlsbad, CA) according to the manufacturer's instructions. Equivalent amounts of protein were resolved by NuPAGE 4% to 12% bis-tris gel (Invitrogen) and transferred to nitrocellulose membranes and immunoblotted as previously described .
Coimmunoprecipitation was carried out as previously described  with minor modifications. The eluate was subjected to SDS-PAGE gel electrophoresis and probed for β-catenin and lef-1 by immunoblot analysis.
A total of 5 x 106 primary CLL cells were incubated with 5 µM PKF115-584 for 3 hours. The cells were washed twice in PBS, fixed in 4% paraformaldehyde, permeabilized using Intraprep solution (Beckman Coulter, Krefeld, Germany) and incubated with primary antibody (lef-1 or isotype control) for 60 minutes. Cells were later incubated with secondary FITC-labeled antibody for 30 minutes. Cells were applied to ethanol-rinsed microscopic glass slides, air-dried, and mounted in VectaShield 4′,6-diamidino-2-phenylindole (DAPI) mounting medium (Vector Laboratories, Burlingame, CA). Images were acquired by fluorescence microscopy (Nikon Eclipse E800; Nikon, Duesseldorf, Germany) and analyzed using Lucia software (Nikon).
A CLL-like xenograft model in nude mice was established as previously described . The mice were monitored for a week for visible tumor growth, and tumor size was measured using standard vernier calipers. When the tumor volume (TV) reached a minimum of ~100 mm3, the mice were divided into three groups (eight mice per group) and treated intraperitoneally with 25 mg/kg of CGP049090, PKF115-584, or the vehicle control every day for 12 doses. The TV was measured every third day, and mice that reached a maximum TV of 1000 mm3 were removed from the study and euthanized owing to ethical reasons. The day of euthanasia owing to the enlarged TV was considered as a death event for Kaplan-Meier survival analysis. TV was measured before therapy and every third day after initiation of therapy. TV was calculated according to the formula: TV = L (mm) x W (mm) x H (mm) / 2, where L is the tumor length and W is the tumor width. The inhibition rate (IR) of tumor growth on the basis of TV was calculated according to the formula: IR = (l - RVt / RVc) x 100%, where RVt represents the mean ratio of TV on day n to TV on day 0 of a treated group and RVc indicates the ration of the control group. The largest value for IR was designated as IRmax, which indicates the greatest effect of each test inhibitor. The animal experiments were carried out with prior approval by the local ethics committee at University of Cologne (approval no. 50.203.2-K17, 17/05).
GraphPad Prism (GraphPad Software, San Diego, CA) was used for all statistical analysis. Statistical analyses of data were performed using unpaired Student's t-test with 2-tailed comparison unless otherwise indicated. Differences of P < .05 were considered significantly different from the control.
To confirm the overexpression of lef-1 in CLL, we performed quantitative real-time PCR in primary CLL samples (n = 66) and healthy B cells (n = 7). The relative mRNA expression of lef-1 was significantly higher in the CLL samples when compared with the healthy B cells (P = .0041; Mann-Whitney test; Figure 1A). We could also show significant overexpression of lef-1 and the presence of dephospho-β-catenin expression in CLL in comparison with healthy B cells at the protein level. Figure 1B shows the protein expression of lef-1 and β-catenin in CLL samples (n = 7), healthy B cells (n = 2), and CLL cell lines JVM-3 and MEC-1. The CLL cell lines MEC-1 and JVM-3 express moderate levels of lef-1 and active β-catenin. Finally, lef-1 is predominantly localized in the nucleus along with β-catenin in primary CLL, indicating constitutive activation of β-catenin/lef-1 signaling in CLL (Figure 1C).
Lef-1 knockdown experiments were performed in six primary CLL samples by nucleofection using SMARTpool siRNA against lef-1 with appropriate controls, and cell viability was monitored after 16 hours by flow cytometry using Annexin V/PI staining. Cells nucleofected with lef-1 siRNA had a significantly increased rate of apoptosis when compared with the cells nucleofected with nontargeting siRNA. The percent Annexin V-/PI- cells (mean ± SE) was 45.0 ± 3.7% in the control siRNA transfected cells and 32.4 ± 3.6% in lef-1-transfected cells (P = .031) as shown in Figure 2A. This validates that knockdown lef-1 promotes apoptosis of CLL cells. Complete silencing of lef-1 was not achievable because of experimental difficulties of CLL viability after transfection coupled with massive overexpression of lef-1. However, lef-1 knockdown was determined at the protein level in two representative samples by using anti-lef-1 antibody and anti-β-actin antibody as loading control (Figure W1, A and B).
Lef-1 knockdown experiments were performed in JVM-3 cell lines by nucleofection using SMARTpool siRNA against lef-1 with appropriate controls, and changes in protein level were monitored after 48 hours by immunoblot analysis. Knockdown of lef-1 resulted in growth inhibition in JVM-3 cell lines (Figure 2B). There was downregulation not only of lef-1 target genes c-myc and cyclin D1 but also of poly(ADP-ribose) polymerase (PARP), pro-caspase 3, and proliferating cell nuclear antigen (PCNA). Marginal reduction of PARP and procaspase 3 was seen, but their cleaved fragments were not detectable, indicating lack of apoptotic induction. However, the levels of β-catenin remained stable, indicating that β-catenin expression is independent of lef-1 expression.
Because CLL is a highly complex and heterogeneous disease, we performed ATP-dependent cell survival assays as described in Materials and methods in 24 different CLL patient samples to illustrate the distinct in vitro cytotoxic response that these compounds have with respect to different subsets of CLL. Table W1 shows the patient's individual response (normalized to the vehicle control) to the inhibitors with their corresponding percentage cell kill at 1 µM after 24 hours. No correlation was seen between the patient's responses and ZAP70/CD38 status, indicating that both inhibitors exert their effect independent of these clinical subtypes. The 50% lethal concentration (LC50) calculated from 24 CLL patients was 0.89 µM for CGP049090 and 0.77 µM for PKF115-584 (Figure 3A). The small molecules did not show significant cytotoxicity in healthy B cells as seen in the CLL samples. LC50 values in healthy B cells could not be obtained owing to lack of complete cell kill (Figure 3B). To date, MEC-1 and JVM-3 are among the few cell lines that can be considered CLL-like cells. They were originally derived from a patient with prolymphocytic transformation of CLL. Both cell lines express κ light chains, surface CD19, and CD23, but both are negative for CD5 and have a complex karyotype. Both compounds inhibited the survival of these cells in a dose-dependent manner. The LC50 values were 0.12 µM for CGP049090 and 0.49 µM for PKF115-584 in MEC-1 cells (Figure 3C) and 0.42 µM for CGP049090 and 0.93 µM for PKF115-584 in JVM-3 cells (Figure 3D).
To assess whether the observed cell death was mediated by apoptosis, we investigated caspase activation using a luminescent caspase 3/7 assay and the accumulation of cleaved PARP by immunoblot analysis on inhibitor treatment in primary CLL cells. Measuring of caspase 3/7 activity at various intervals allowed us to determine the progression of apoptosis induction by the inhibitors. CLL cells (n = 3) were incubated in the presence or absence of the inhibitors (5 µM) or the vehicle control (0.1% DMSO) for 0, 2, 4, 6, and 8 hours. Both inhibitors showed a significant induction of caspase 3 and caspase 7 activities over time as early as 2 hours after treatment compared with the controls (Figure 4A). These data indicate that cell death induced by the inhibitors is mediated by apoptosis.
To avoid the heterogeneity seen in primary CLL cells derived from different donors, the JVM-3 cell line was used as a model system to enumerate the apoptotic mechanism involved. Time course experiments on JVM-3 cells treated with the inhibitors for 6, 12, and 24 hours and subsequent immunoblot analysis of whole-cell extracts revealed cleavage of caspases 9, 3, 7, and 8. In addition, there was also a parallel decrease of antiapoptotic proteins such as Mcl-1 and XIAP on treatment (Figure 4B).
To distinguish from extrinsic and intrinsic apoptotic pathway, cytochrome C exit and bid cleavage were analyzed. Additional time course experiments between 0 and 6 hours performed in JVM-3 cell line revealed accumulation of cytochrome C in the cytoplasmic fraction, which leads to activation of caspase 3 and caspase 8 at 4-hour interval in cells treated with PKF115-584 and at 2-hour interval in cells treated with CGP049090. However, complete down-regulation of mcl-1 and XIAP did not precede cytochrome C release in the cells treated with inhibitors (Figure W2, A and B).
We used pan caspase inhibitors to determine 1) whether the apoptosis induced by CGP049090 and PKF115-584 is inhibited by pretreatment with ZVAD-FMK and 2) whether the inhibitors of apoptosis and lef-1 target proteins are downregulated preceding or succeeding caspase activation. We treated primary CLL cells with 3 µM of inhibitors for 16 hours in the presence or absence of 50 µM pan caspase inhibitor (ZVAD-FMK) with respective vehicle control for 16 hours. Pretreatment of CLL cells with pan caspase inhibitors leads to inhibition of lef-1, caspase 3, and PARP. Results also indicate that downregulation of antiapoptotic proteins (XIAP and Mcl-1) and lef-1 target genes (c-myc and cyclin D1) were due to initiation of apoptotic events (Figure 4C).
Previous studies have pointed out that the above-mentioned inhibitors specifically inhibit interaction between β-catenin and members of the tcf/lef-1 transcription factor family. Because β-catenin is a direct target of caspase 3 activity, there is a rapid apoptotic loss of β-catenin on in vitro treatment; hence, we decided to investigate the ability of these compounds to inhibit β-catenin/lef-1 interaction in whole-cell protein extracts of primary CLL samples treated with the inhibitors by coimmunoprecipitation. The immunoprecipitate was immunoblotted and probed for β-catenin and lef-1. The results indicate that β-catenin/lef-1 interaction was almost completely inhibited on 16 hours of incubation of whole-cell extracts with 10 µM CGP049090 or PKF115-584 before coimmunoprecipitation. This underlines the efficacy of the compounds to β-catenin/lef-1 inhibition (Figure 5A).
Because CGP049090 and PKF115-584 inhibit β-catenin/lef-1 signaling, we assessed the expression of lef-1 target proteins, namely lef-1, c-myc, and cyclin D1, in three different primary CLL cells by immunoblot analysis. The cells were treated with 5 µM of the inhibitors for 48 hours, and whole-cell extract was prepared. The levels of β-catenin, c-myc, cyclin D1, and lef-1 were found to be significantly reduced, demonstrating the effect of the inhibitors on inducing apoptosis in different CLL cells (Figure 5B).
Of the two inhibitors tested PKF115-584 had substantial autofluorescence properties (Figure W3). Results of indirect immunofluorescence studies revealed a high expression of lef-1 in both the cytoplasm and the nucleus in primary CLL cells. PKF115-584 had a strong cytoplasmic localization during apoptosis, implying that the inhibition occurs in the cytoplasmic level at lef-1 or β-catenin. Figure 5C shows the independent and merged channels of a CLL cell treated with PKF115-584 for 3 hours and labeled with FITC lef-1 antibody and nucleus stained in DAPI. The yellow region in the cytoplasm of the merged picture represents colocalization of lef-1 and PKF115-584, suggesting possible inhibition of lef-1 transport into the nucleus by the inhibitors.
We determined the in vivo efficacy and systemic tolerance of these inhibitors in a JVM-3 xenograft model. The in vivo efficacy and systemic tolerance were first determined in a small cohort (n = 4 per group) of animals that showed a very good efficacy (Figure 6A). Table W2 shows this observation in initial and final TV (mm3) and the tumor inhibitory rate (IRmax) of each of the inhibitors calculated as described in Materials and methods. There was a significant cessation of tumor growth in the mice treated with CGP049090 and PKF115-584. The experiment was then repeated in a larger cohort (n = 8 per group), and the tumor growth response was monitored after administration of 12 doses at 25 mg/kg during a period of 30 days after initiation of therapy. Both the inhibitors were well tolerated in vivo because there was no toxic lethality in mice of the tested groups. Kaplan-Meier survival curves reveal that treatment significantly improves median survival by 12.5 days (P = .003) with CGP049090 (Figure 6B) and 15.5 days (P = .0023) with PKF115-584 (Figure 6C). Immunohistochemical analysis of the tumor sections reveal down-regulation of lef-1 in mice treated with CGP049090 and PKF115-584 when compared with the controls (Figure W4). Furthermore, we observed down-regulation of PCNA and increased PARP cleavage with CGP049090- and PKF115-584-treated mice when compared with the controls (Figure W5).
Our results consistently confirm overexpression and nuclear localization of lef-1 in CLL cells, and targeting lef-1/β-catenin complex using small molecule inhibitors CGP049090 and PKF115-584 resulted in induction of apoptosis in CLL cells.
Overexpression of bcl-2 and other inhibitors of apoptosis is known to be hallmark feature in CLL, which leads to the inhibition of apoptosis . Here, we report that the overexpression of lef-1 in CLL is another characteristic feature leading to its extended survival. However, whereas we could not observe a massive accumulation of β-catenin in CLL, there is a steady detectable pool of both phosphorylated and dephosphorylated forms of β-catenin. The maintenance of dephosphorylated β-catenin is attributed to autocrine and paracrine action (microenvironment) of wnt ligands and BCR ligation on the CLL cells [3,19–21]. Hence, the aberrant activation of wnt signaling in CLL is in cooperation with massive expression of lef-1 and a steady pool of dephosphorylated β-catenin.
Lef-1 can activate transcription of several target genes in a β-catenin-dependent or -independent manner. c-myc and cyclin D1 are target genes of lef-1, which act in a β-catenin-dependent pathway [6,7], whereas E2F1 is expressed in a β-catenin-independent pathway . Hence, even in the absence of wnt ligand stimulation, it can be speculated that lef-1 can transcribe context-dependent genes in CLL. SiRNA-mediated knockdown experiments of lef-1 resulted in an increased apoptosis in nonproliferative CLL cells. Complete lef-1 knockdown in primary CLL cells was not achievable owing to spontaneous apoptosis exhibited by CLL cells in vitro, poor transfection efficiency and viability after transfection, and substantial expression of lef-1 in CLL cells. We performed siRNA-mediated lef-1 knockdown experiments in JVM-3 cell line, which resulted in expected down-regulation of lef-1 target genes (c-myc and cyclin D1) but led to inhibition of cell proliferation instead of apoptosis. Cleavage of PARP and procaspase 3 are widely accepted markers of apoptosis, surprisingly, we observed down-regulation of procaspase 3 and PARP on lef-1 knockdown in JVM-3 cell lines. Hence, we investigated the levels of PCNA that is expressed in the nuclei of cells during the DNA synthesis phase of the cell cycle and considered as a marker for proliferation . There was reduction of PCNA levels on lef-1 inhibition, leading to proliferative arrest of the JVM-3 cell line.
Levels of β-catenin remained stable, indicating β-catenin expression to be independent of lef-1 expression. c-myc, which is a potent transcription factor by itself, governs the expression of 15% genes of human genome involved in cell cycle progression, apoptosis, and cellular transformation [24,25], which might lead to significantly shutting down the transcriptional machinery in the CLL cells, leading to growth arrest. However, cyclin D1 down-regulation must be regarded as a surrogate for lef-1 inhibition in the context of our investigation, but no statement can be made about its functional relevance with regard to CLL. Lef-1 down-regulation has the potential to inhibit the expression of several target genes in CLL including lef-1 itself because its lef-1/β-catenin complexes can activate the promoter for full-length lef-1 under a positive feedback loop . Taken together, down-regulation of the above crucial proteins underlines the hierarchical importance of β-catenin/lef-1 signaling in CLL cells, and inhibition of this activity could be a viable therapeutic strategy.
We used the known lef-1/β-catenin inhibitors CGP049090 and PKF115-584 to test the therapeutic benefit of lef-1 inhibition in CLL. Interestingly, both substances induced cell death in CLL cells and cell lines at the submicromolar levels (LC50 < 1 µM). However, we observed that both inhibitors show a wide range of % cell kill (10%–98%) in 24 different patient samples, with a mean cell kill of approximately 55% at 1 µM (Table W1). Hence, we made our mechanistic studies in JVM-3 cell line and used concentrations higher than the LC50 values in primary CLL cells. It has to be noted that a previous study attempted the same in CLL using the β-catenin inhibitor R-etodolac, which was unsuccessful owing to its high LC50 greater than 250 µM . Remarkably, despite its efficient induction of cell death in CLL, the healthy B-cell counterparts were significantly less affected by the inhibitors. Furthermore, these inhibitors do not affect the healthy peripheral blood mononuclear cells as seen from the estimated LC50 values of 37.83 and 113.31 µM for PKF115-584 and CGP049090, respectively, shown previously by our group . Hence, both these small molecules have the potency to induce cell death in leukemic cells at the same time sparing the healthy cells.
Because both CGP049090 and PKF115-584 are still lead compounds from a high-throughput screen, there is little evidence on their mechanism of apoptotic induction. A short time course for caspase 3 and 7 activation in CLL cells incubated with 5 µM of the inhibitors revealed a rapid onset of caspase 3 and 7 activation by luminescent assay. Caspase 3 and 7 are executioner caspases that cleave several intracellular protein targets promoting apoptosis [27,28] as evident from the detection of cleaved PARP . Mechanistic studies carried out in JVM-3 cell lines indicate activation of intrinsic pathway by cytochrome C release. Previous studies have indicated that Fas (CD95/APO-1) signaling is defective in CLL  and that this system does not play a crucial role in inducing apoptosis by cytotoxic drugs or radiation and observed that procaspase 8 processing accompanied the activation of procaspase 3 . We also observed a similar effect where cleavage of caspase 3 accompanied caspase 8 cleavage on incubation with the inhibitors as previously described. However, down-regulation of mcl-1 and XIAP did not precede cytochrome C release as evident from cytochrome C release assay (Figure W2, A and B) and pan caspase inhibition (Figure 4C). Hence, it is evident that the apoptotic response was mediated by intrinsic pathway and activation of caspase 3 and 7 leading to cleavage of mcl-1 and XIAP, which are known inhibitors of apoptosis in CLL [32–34].
Coimmunoprecipitation experiments concord to the previous findings that both CGP049090 and PKF115-584 efficiently inhibit lef-1/β-catenin interaction , However, our caspase inhibition experiments reveal that there is down-regulation of lef-1 in CLL cells and parallel induction of apoptosis, indicating that these small molecules possess both inhibiting lef-1/β-catenin interaction and antileukemic cytotoxicity.
CGP049090 and PKF115-584 exert their effect within the micromolar range and do not significantly affect normal B cells, but it must be questioned, however, whether the inhibitors possess a toxic potential on less differentiated and proliferating hematopoietic cells or other tissues in vivo. We treated a small cohort of nontransplanted mice with 50 mg/kg bodyweight daily of CGP049090 or PKF115-584 for 14 days, and they did not develop any gross systemic toxicity (data not shown). This is consistent to a previous study that investigated pharmacokinetics and metabolism of another compound, calphostin C, which structurally resembles PKF115-584 . The authors reported that intraperitoneal administration of 40 mg/kg bodyweight in CD1 mice resulted in therapeutically relevant plasma levels (Cmax = 2.9 µM). Moreover, no toxic signs or fatalities could be observed in the animals during a 30-day observation period after administration.
Initial attempts made to test these inhibitors in a transgenic Eµ-TCL-1 model were highly unsuccessful because of the unpredictable deaths in both treated and control groups and the extended period before the TCL-1 transgenic mice developed leukemia. Hence, owing to a lack of convincing in vivo tumor model for CLL, we carried out in vivo studies in JVM-3 subcutaneous xenografts. The results of in vivo experiments indicate that all of the xenograft mice in the treated groups survived the entire regimen, with a significant tumor inhibition and extended survival mediated by lef-1 knockdown when compared with the controls, regardless of the high proliferative rate exhibited by the EBV-transformed JVM-3 cells. Hence, it can be stated that both inhibitors are well tolerated at doses that are effective for CLL cell killing in vivo.
To conclude, we report that lef-1 is an attractive therapeutic target for CLL therapy. Both CGP049090 and PKF115-584 are potential anticancer agents in CLL and other neoplastic malignancies. Further investigations are warranted to determine the lef-1 transcriptional activity in different subsets of CLL and the feasibility of these small molecules for therapeutic approach in humans.
The tumor sections embedded in paraffin were incubated in three washes of xylene for 5 minutes each to deparaffinize the section. Later, the sections were rehydrated with decreasing concentration of ethanol (100%, 95%, 75%) 10 minutes each. After rinsing the slides in ddH2O, the slides are placed in antigen unmasking buffer (TE buffer) and boiled for 20 minutes. After cooling them down for 5 minutes, the slides are washed in ddH2O and incubated in 3% hydrogen peroxide for 10 minutes. The sections were then washed in ddH2O and blocked in 500 µl of blocking solution for 1 hour. After removing the blocking solution, 500 µl of primary antibody diluted in the antibody diluent to each is added to the sections and incubated overnight at 4°C in a humidified chamber.
The next day, the sections are rinsed in wash buffer three times for 5 minutes followed by incubation with 500 µl of biotinylated secondary antibody, diluted in Tris-buffered Saline Tween-20, and incubated in room temperature for 1 hour. The slides are washed again three times for 5 minutes, and 500 µl of avidin-biotin complex (ABC) reagent is added to the sections and incubated for 1 hour. The sections were washed in wash buffer thrice, and 200 µl of 3,3′-diaminobenzidine (DAB) substrate is added to each section and incubated for 20 minutes until the appropriate color reaction develops. The sections were washed in ddH2O, mounted with coverslips, and observed under documented with the light microscope.
R.K.G. designed, performed, analyzed research, and wrote the manuscript; K.M., I.G., and G.P. analyzed data; A.S., E.K.S., and P.S. contributed tools and reagents. M.H. analyzed research and reviewed the manuscript. K.-A.K. designed and performed research; collected, analyzed, and interpreted data; and wrote the manuscript. The authors declare no conflict of interests.
The authors thank A. Behle, R. Semrau, and S. Tawadros for technical assistance. Parts of the present work will be included into the academic theses of R.K.G.