Conceived and designed the experiments: LS DW AB. Performed the experiments: LS AH AB. Analyzed the data: LS AB. Contributed reagents/materials/analysis tools: DW RD EH AB. Wrote the paper: LS AB. Manuscript review: RD EH DW.
BioResponse Nutrients, LLC donated the DIM formulation used in this study, BioResponse-DIM. The monetary value of this donation is less than $500. BioResponse Nutrients, LLC did not have any involvement in the experiment design, execution, analysis or composition of this manuscript. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials. USANA Health Sciences provided financial support for this research, via its support of the Linus Pauling Institute and associated laboratories. USANA did not have any involvement in the experiment design, execution, analysis or composition of this manuscript. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
Certain bioactive food components, including indole-3-carbinol (I3C) and 3,3′-diindolylmethane (DIM) from cruciferous vegetables, have been shown to target cellular pathways regulating carcinogenesis. Previously, our laboratory showed that dietary I3C is an effective transplacental chemopreventive agent in a dibenzo[
Acute lymphoblastic leukemia (ALL), the most frequently diagnosed cancer in children ages 0 to 19 years
New strategies in cancer therapy utilize drugs that specifically target aberrant signaling pathways in order to reduce toxic side effects, yet such specific therapies are only effective in a small percentage of this highly heterogeneous disease population. For example, more than half of T-ALL cases are characterized by a gain-of-function mutation in the Notch1 receptor, which leads to constitutive activation of Notch-mediated cell proliferation and survival
Evidence from epidemiological and animal studies shows that modification of the diet to increase consumption of cruciferous vegetables is sufficient to reduce cancer risk (reviewed in
Herein, we report for the first time that DIM markedly reduces the proliferation and survival of four different human T-ALL cell lines, which were selected to represent the heterogeneity of the disease. The anticancer effects of DIM were exerted by modification of critical regulators in the cell cycle pathway leading to induction of G1 arrest and apoptosis. Subtle differences in sensitivity within this group of cell lines were observed, although DIM was more potent than its parent compound I3C in all cases. Of particular importance was the observation that DIM reduced growth of human CEM cells in a xenograft model when supplemented through the diet. Collectively, the data presented below suggest that DIM could be an effective anticancer agent in T-ALL cases originating from T cells at different stages of differentiation and at concentrations that can be reasonably achieved
The following chemicals and reagents were purchased from the indicated suppliers: I3C from Sigma-Aldrich Co. (St. Louis, MO), Matrigel Matrix from BD Biosciences (Franklin Lakes, NJ) and ViaCount Flex Reagent from Millipore (Billerica, MA). DNase I and the NuPAGE system for SDS-PAGE, including 10% and 4–12% Bis-Tris gels and appropriate electrophoresis and transfer buffers, were purchased from Invitrogen (Carlsbad, CA). Antibodies for immunoblotting were obtained from Cell Signaling Technology (Danvers, MA), including β-actin and α-tubulin primary antibodies and the Cell-Cycle Regulation Antibody Sampler Kit (contains primary antibodies for CCND3, CDK4, and CDK6 as well as HRP-linked anti-mouse and anti-rabbit IgG secondary antibodies). DIM was kindly provided in a bioavailable formula (BioResponse-DIM, herein referred to as DIM) by BioResponse, LCC (Boulder, CO), which was certified to contain 30% DIM (wt/wt) by Eurofins-Alpha Laboratories (Petaluma, CA). This bioavailable form of DIM, rather than the pure crystalline DIM, has been utilized for many of the preclinical and clinical studies in the published literature and is the common form provided in commercial dietary supplements. For these reasons, we selected the BioResponse formula for the experiments outlined below. Experimental concentrations reported in this study were adjusted accordingly (
T-ALL is a heterogeneous disease resulting from the developmental arrest and abnormal proliferation of T-cells at different stages of maturation
Cell line (abbreviation) | CCRF-HSB2 (HSB2) | CCRF-CEM (CEM) | SUP-T1 | Jurkat (JM) | Reference |
|
NIH-AIDS (497) | ATCC (CCL-119) | NIH-AIDS (100) | NIH-AIDS (4668) | |
|
11/m | 3/f | 8/m | 14/m |
|
|
Pre-T | Pre-T | Cortical T | Mature T |
|
CD3 (cytoplasmic) | − | + | + | + |
|
CD3 (surface) | − | − | − | + |
|
CD4 | − | + | + | + |
|
CD8 | − | − | + | − |
|
CD1a | − | − | + | − |
|
|
|||||
CDKN2A(p16) | + | + | + | + |
|
RB1 | − | − | − | − |
|
TP53 | − | + | + | + |
|
PTEN | − | + | − | + |
|
|
|||||
NOTCH1 activating mutation (50–60%) |
+ | + | + | + | |
TAL1 expression (25%) |
+++ | ++ | − | ++ |
|
STIL-TAL1 fusion |
+ | + | − | − |
|
LYH1 expression |
+ | ++ | − | − |
|
ATCC, American Type Tissue Collection; NIH AIDS Research and Reference Reagent Program.
Indicates presence (+) or absence (−).
Indicates relative expression (−, +,++, or +++).
T-ALL cells were treated with 0 up to 60 µM DIM or 0 up to 500 µM I3C for up to 48 hr. The concentration of viable cells was determined at each indicated time point by the ViaCount Assay (Millipore, Billerica, MA) as recommended by the manufacturer using either the Guava Personal Cell Analyzer (Guava Technologies, Inc., Hayward, CA) or the Accuri C6 flow cytometer (BD Accuri Cytometers, Inc., Ann Arbor, MI); assay performance was comparable on both instruments. Raw data were compared to the time-zero control for cell proliferation and the time-matched control for viability. Concentration values for 50% inhibition (IC50) of T-ALL cell proliferation and viability by I3C and DIM were calculated by non-linear regression using a sigmoidal dose-response with variable slope (Prism 5, GraphPad Software, La Jolla, CA).
T-ALL cells were treated with 0 to 15 µM DIM for up to 48 hr, rinsed in cold PBS, fixed in ice cold 70% EtOH, and stored at least overnight at −20°C. On the day of analysis, cells were washed with PBS and incubated for 30 min in the dark in staining solution (25 µg/ml propidium iodide, 0.1% (v/v) Trition X-100 and 0.2 mg/ml RNase in PBS). Flow cytometry was used to determine cellular DNA distribution using the Guava PCA or Accuri C6 instruments and the number of cells in each cycle were analyzed using MultiCycle software (Phoenix Flow System, San Diego, CA) or FlowJo Cytometry Analysis Software (Ashland, OR).
Cells were treated with 0 to 15 µM DIM for 12 or 24 hr or with 0 to 500 µM I3C for 24 hr, then lysed in IP lysis buffer (20 mM Tris, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% (v/v) Triton X-100, 2.5 mM Na4P2O7·10H2O, 1 mM C3H9O6P, 1 mM Na3VO4, 1 µg/ml leupeptin and 0.5% protease inhibitor cocktail III (EMD Chemicals, Gibbstown, NJ)). Protein concentration was determined using the Coomasie Plus Assay (Thermo Scientific, Rockford, IL) and an equal amount of protein for each sample was separated by SDS-page electrophoresis and transferred to nitrocellulose membranes. Membranes were blocked for 1 hr in 5% non-fat milk or BSA prior to overnight incubation at 4°C with primary antibodies for CCND3, CDK4, or CDK6 (all 1∶1000 dilution). Membranes were subsequently incubated with the appropriate HRP-conjugated secondary antibody for 1 hr. Immunoreactive proteins were visualized using an Alpha Innotech Imaging Station (Cell Biosciences, Santa Clara, CA) and the Western Lightning ECL reagent (Perkin Elmer, Waltham, MA). Protein bands of interest were measured by densitometry using FluorChem 8800 software (Cell Biosciences, Santa Clara, CA). Membranes were stripped using Restore Western Blot Stripping Buffer (Thermo Scientific) and tested for removal of antibodies before re-probing with β-actin or α-tubulin. Changes in protein expression, normalized to β-actin or α-tubulin, were calculated as the mean difference in percentage compared to time-matched vehicle controls (0.1% DMSO), which were assigned a value of 100%.
The terminal deoxynucleotidyl transferase dUTP nick end labeling method (TUNEL) was applied to CEM treated with 0–15 µM DIM for 48 hr. The
Total RNA was extracted using TRIZOL reagent (Sigma-Aldrich) as recommended by the manufacturer from triplicate samples of CEM cells treated with 7.5 µM DIM for 4 or 24 hr. cDNA synthesis was performed using 2 µg RNA per sample with the RT2 First Strand Synthesis Kit (SABiosciences, Frederick, MD); quantitative PCR analysis for 84 genes related to human apoptosis was performed using the RT2 Profiler PCR Array System (SABiosciences) with the iCYCLER iQ5 Real-Time PCR System (Bio-Rad, Hercules, CA). Relative gene expression was calculated using the ΔΔCt method
All protocols for the handling and treatment of mice were reviewed and approved by the Oregon State University Institutional Animal Care and Use Committee (Animal Care and Use Protocol #3837). Male NOD.CB17-
Detailed methods for the xenograft study are provided in
Detailed methods for staining and analysis of xenograft tissues by TUNEL for detection of apoptosis are provided in
GraphPad Prism 5 software (La Jolla, CA) was used for all statistical analyses. One or two-way ANOVA were performed as appropriate for the number of experimental factors being examined. Statistical significance was inferred when
To determine the impact of DIM and I3C on growth of a representative human T-ALL cell line, a time-course study was performed over a range of concentrations using CEM cells. DIM and I3C blocked the proliferation of CEM cells in a time- and concentration-dependent manner (
Cells were treated with 0 (□), 1.9 (▴), 3.8 (▿), 7.5 (⧫), 15 (○), or 30 (▪) µM DIM (panels A,C) or 0 (□), 15.6 (▴), 31.3 (▿), 62.5 (⧫), 125 (○), 250 (▪), or 500 (⋄) µM I3C (panels B,D) for 24 or 48 hr, then stained with ViaCount reagent for analysis of viable cell concentration and percent viability. Values are the mean fold change in cell proliferation (panels A, B) or percent viability (panels C, D) ± SEM (n = 3 independent experiments) normalized to control cells at 0 hr. **,
Concentration-response experiments were performed in four different T-ALL cell lines to determine whether DIM and I3C are similarly effective in reducing growth of T-ALL cells derived from T-cells at different stages of differentiation. In all cell types,
Human CEM, HSB2, SUP-T1 and Jurkat cells were treated for 48 hr with I3C (15.6 up to 500 µM) or DIM (1.9 up to 60 µM), then stained with ViaCount reagent for analysis of cell concentration and percent viability. Values are the mean level of cell proliferation (panels A–D) or the mean percent viable cells (panels E–F) ± SEM (n = 3 independent experiments), normalized to the time-matched vehicle control (0.1% DMSO). Non-linear regression analysis (four parameter, variable slope) was performed (GraphPad Prism) to generate the concentration-response curve for each chemical in each cell line, from which IC50 values were obtained (see
IC50 (µM) for DIM | IC50 (µM) for I3C | |||
Cell line | Proliferation | Viability | Proliferation | Viability |
CEM | 15 | 27 | 122 | 223 |
HSB2 | 8 | 7 | 86 | 83 |
SUP-T1 | 13 | 14 | 262 | 284 |
Jurkat | 9 | 15 | 228 | 222 |
The marked suppression of proliferation by DIM prompted us to evaluate cellular DNA content by flow cytometry in each of the four T-ALL cell lines. Treatment of CEM or HSB2 cells with 7.5 or 15 µM DIM for 48 hr resulted in a significant G1 cell cycle arrest, with substantially fewer cells progressing to the G2/M phase (
Cells were treated with 0, 3.8, 7.5, or 15 µM DIM for 48 hr, then fixed in ice-cold 70% EtOH and stained with propidium iodide. DNA content distribution was analyzed by Guava PCA or Accuri C6 flow cytometry. (A–B) Representative histograms are shown for control and 15 µM DIM treatments at 48 hr in human CEM cells. (C–D) Distributions of CEM, HSB2, SUP-T1 or Jurkat cells in G1 (black), S (white), and G2 (grey) phases of cell-cycle progression at 48 hr (n = 3 to 5 independent experiments). *,
Next, we measured the expression of key regulatory proteins of cell cycle progression by immunoassay in CEM cells. DIM suppressed expression of key cell cycle regulatory proteins
Following either 12 hr (gray bars) or 24 hr (black bars) treatment with increasing concentrations of DIM, CEM cells were harvested and protein immunoassays were performed for detection of CCND3, CDK4 and CDK6 proteins (three replicate experiments performed). (A) A representative immunoblot is shown for each protein assay. (B) Values shown are average protein expression ± SEM normalized to β-actin, expressed as a percentage difference from time-matched vehicle controls (0.1% DMSO), which were assigned a value of 100%. *,
Two methods for assessing the impact of DIM on apoptosis were used in this study. First, the portion of apoptotic cells following treatment with DIM for 48 hr was determined by the ViaCount assay. In all four T-ALL cell types, treatment with 15 µM DIM caused a significant increase in the percentage of apoptotic cells (
CEM, HSB2, SUP-T1 and Jurkat cells were treated with 3.8 to 15 µM DIM for 48 hr. Values are the proportion of apoptotic cells as determined using the ViaCount assay + SEM (n = 3 to 4 independent experiments). *,
Next, the extent of DNA strand breaks
The
We determined the effects of DIM on expression of gene targets relevant for regulation of apoptosis in human cells. Fold change values and results of the statistical analyses for all gene targets on the apoptosis PCR pathway array are provided in
Log2 R ( |
||||
Unigene | Symbol | Description | 4 hours | 24 hours |
Hs.283672 | BCL2L10 | BCL2-like 10 (apoptosis facilitator) | ||
Hs.592244 | CD40LG | CD40 ligand | ||
Hs.87247 | HRK | Harakiri, BCL2 interacting protein (contains only BH3 domain) | ||
Hs.241570 | TNF | Tumor necrosis factor (TNF superfamily, member 2) | −0.39 (0.026) | |
Hs.279594 | TNFRSF1A | Tumor necrosis factor receptor superfamily, member 1A | −0.46 (0.331) | |
Hs.462529 | TNFRSF25 | Tumor necrosis factor receptor superfamily, member 25 | nd |
|
Hs.654445 | TNFSF8 | Tumor necrosis factor (ligand) superfamily, member 8 | 0.34 (0.009) | |
Hs.8375 | TRAF4 | TNF receptor-associated factor 4 | nd |
A complete list of DIM-induced changes in gene expression, including all genes on the RT2 Profiler Apoptosis array, is provided in
Log2 fold change (R) values are highlighted in bold if level of change is >1.5-fold (Log2 R<−0.58 or >0.58) compared to vehicle (0.1% DMSO) control.
nd, not detected by RT2 PCR profiler array at this time point (Ct>35).
Another key objective of this study was to determine whether dietary DIM or I3C reduced the growth of human CEM cells
Male NOD.CB17-
Treatment | Final tumor volumemm3 ± SEM | Tumor doubling time |
Control | 1360±222 | 6.43 (5.22–8.35) |
100 ppm DIM (350 ppm BR-DIM) | 761±153 |
10.2 (7.51–15.9) |
500 ppm I3C | 1030±318 | 7.57 (5.43–12.5) |
2000 ppm I3C | 994±191 | 8.45 (6.47–12.2) |
,
,
,
Tumor growth rates were modeled by non-linear regression analyses using the exponential growth equation with least-squares fit (Prism 5). Average doubling time (DT) values are shown and were calculated as follows: DT = [(To−Ti)×ln2]/ln(Vo/Vi) where
Because high rates of apoptosis were detected in CEM cells exposed to DIM
(A) The
We provide evidence for the first time that DIM significantly impairs the growth of human T-ALL cells
The SCID mouse model supports the solid growth of subcutaneously injected human acute leukemia blast cells in a manner that is easily measurable and exhibits a dissemination pattern analogous to the human disease
We selected a dietary concentration of 2000 ppm I3C level based on the apparent anticancer effects at this level observed in our previous studies
DIM was also significantly more potent than I3C
Other plausible explanations exist for the distinctive responses to DIM observed in the four T-ALL cell lines studied, which are characterized by different lineages of T-cell differentiation (pre-T, cortical-T and mature-T), as well as different ages and genders of the source patients (
Although beyond the scope of this study, the different combinations of these aberrations across the cell lines tested are likely to play a role in the therapeutic effect of DIM. The variable responses to both targeted and conventional chemotherapeutic drugs that have been previously observed in T-ALL cells are likely a consequence of the respective mutations harbored by the cell lines (e.g.,
Treatment of CEM and HSB cells with DIM caused a blockade of cell-cycle progression at the G1 phase checkpoint, although this effect was not observed in more differentiated T-ALL cell lines (SUP-T1 or Jurkat); DIM (and I3C) also suppressed expression of CCND3, CDK4 and CDK6 cell cycle regulatory proteins in CEM cells. Early progression of the eukaryotic cell cycle is positively regulated by the coupling of D-type cyclins with the highly homologous CDK4 or CDK6 proteins and negatively regulated by cyclin dependent kinase inhibitors and phosphatases
DIM treatment effectively induced apoptosis in human T-ALL cells
Bioactive dietary components may be utilized as part of a healthy lifestyle aimed at disease prevention or therapy. The ability of I3C/DIM to target multiple pro-survival pathways in cancer cells, while causing few adverse effects on normal cells, has been explored in a number of cancer models with substantial success. Collectively, our work points to the potential benefit of exposure to these agents at early life stages for chemoprotection, from gestation through adolescence, when leukemia is most prevalent
(TIF)
(PDF)
(PDF)
The authors would like to thank the staff of the Laboratory Animal Resource Center and the Cancer Chemoprevention Core Labs at Oregon State University. Finally, we greatly appreciate the technical assistance provided by Mohaiza Dashwood, Dr. Carmen Wong, Dr. Praveen Rajendran, Marilyn Henderson, Lisbeth Siddens and Deanna Larson.