Background
In 2018, almost 21,000 new cases of acute myeloid leukemia (AML) will be diagnosed with an estimated 10,500 deaths [
1]. In the absence of intensive chemotherapy, median survival of patients with AML is a mere 5–10 months [
2]. The general treatment guidelines for AML have not changed in the last 40 years [
3] and consists of high dose infusion of cytarabine in combination with an anthracycline (doxorubicin or daunorubicin). As such, this aggressive chemotherapy regimen is poorly tolerated and comes with significant side effects [
4]. Newer targeted therapies have shown promise but act only on a small subset of patients [
5,
6]. Furthermore AML typically affects the elderly [
7] which tolerate the aggressive chemotherapy regimen poorly. Also, even with this aggressive toxic chemotherapy regimen, only 15% of patients older than 65 years of age will achieve complete remission [
3]. As such, novel therapy that is both well tolerated and effective is the need of the hour.
Previous studies from our laboratory show that triptolide, a diterpene tri-epoxide, and its water soluble pro-drug, Minnelide, are effective against pre-clinical models of multiple cancers [
8‐
11]. Subsequently a Phase I clinical trial for Minnelide showed promising efficacy and tolerability at a dose of 0.67 mg/m
2/day in patients with advanced gastro-intestinal (GI) malignancies [
12]. This translates to an equivalent mouse dose of ~ 0.21 mg/kg/day [
13]. In the current study we show that Minnelide is effective against AML at a very low dose (well within the dose range tolerated by patients) both in vivo and in vitro. At mechanistic level we demonstrate that triptolide decreases stemness in AML cancer cells and mediates apoptotic cell death via the Myc pathway.
Discussion
Treatment for AML that is effective and also well tolerated has been a focus of intense research for the last six decades [
20]. But despite major breakthroughs in the molecular and phenotypic characterization of AML, standard therapy has remained largely unchanged for the last four decades [
21,
22]. The prognosis with chemo-intensive regimen of cytarabine and anthracyclines is modest at best in young patients. However, a greater proportion of the patients diagnosed with AML are elderly and present with significant co-morbidity, precluding any hope for long term remission [
3]. In this backdrop, there is an unmet need for newer therapeutic modalities to treat AML.
Through our studies, we show that Minnelide is effective against AML in multiple pre-clinical models. Treatment with Minnelide resulted in complete clearance of leukemic burden in AML models of both cell line and patient derived AML blasts. The doses that are able to clear leukemia in these models ranges from 0.1 to 0.15 mg/kg/day. We found that Minnelide, when used for AML, is effective at low doses. These doses are much lower than equivalent human doses that were safely tolerated by patients in the Phase I Clinical Trial of Minnelide [
12].
This is confirmed by our findings in vitro, where triptolide, the active metabolite of Minnelide, leads to apoptosis in cancer cells at doses as low as 5 nM. This efficacy is seen over a wide range of AML cancer cell lines and patient blast samples.
Consistent with its cytotoxic effect, treatment with triptolide resulted in a decrease in clonogenicity, a marker of stemness, as well as reduced expression of stem cell markers. More interestingly though, Minnelide when administered at a higher dose of 0.21 mg/kg/day had no effect on normal hematopoietic myeloid precursors, highlighting its selectivity against leukemic cells.
The broad effect of triptolide against leukemic cell lines and patient samples pointed towards triptolide’s (Minnelide) effect against a pathway that is commonly dysregulated in leukemias. One such transcription factor regulating multiple processes in AML is c-Myc. We found that treating AML cells with triptolide resulted in a significant decrease in transcription and ultimately translation of c-Myc. In AML, c-Myc is overexpressed via multiple mechanisms. Myc amplification is described to occur via double minute chromosomes [
23,
24] in AML and c-Myc overexpression is a common end result of fusion oncoproteins such as AML/ETO, PML-RARA and PLZF-RARA [
25]. Furthermore, activating mutations in the pathways of receptor and non-receptor tyrosine kinase is also shown to dysregulate c-Myc [
26]. These mutations can be found in almost 50% of patients with AML [
27] and other studies have highlighted its role as a central downstream transcription factor mediating AML progression [
28]. Consistent with its role in cell cycle regulation, we saw that inhibition of c-Myc with triptolide caused an arrest in G1–S phase, a progressive decrease in cells in the G2/M phase and an upregulation of the cell cycle checkpoint protein p21. C-Myc is also known to be an important mediator of cell survival in leukemic stem cells [
16,
17] and may explain the reduction in colony forming ability as well as the decrease in stemness markers seen on treatment with triptolide. Subsequently, inhibition of c-Myc with a specific chemical inhibitor resulted in cell death of AML cancer cell lines recapitulating the effects seen with triptolide.
In summary, we show that Minnelide results in a dramatic reduction in leukemic burden in various pre-clinical models of AML at doses that can be easily achieved clinically in patients. In vitro, we see that triptolide, the active component of Minnelide, leads to inhibition of c-Myc transcription in AML cells which subsequently causes cell cycle arrest and cell death via apoptosis. Triptolide is also effective in decreasing the “stemness” of AML cells contributing to the long-lasting remission seen on stopping drug treatment.
Conclusion
Here we show that Minnelide is effective in decreasing leukemic burden in AML through multiple pre-clinical models and at doses that are clinically applicable. In vitro, we show that triptolide, the active metabolite of Minnelide, results in cell cycle arrest, cell death and reduced clonogenicity of leukemic blasts. Evaluation of Minnelide in patients with AML holds great promise in this difficult disease.
Materials and methods
Cell culture and drugs
THP-1, HL-60, Kasumi-1 and KG1a cell lines were obtained from the ATCC and grown in RPMI-1640 with varying concentrations of FBS (10%, 10%, 20% and 20% respectively) with 1% penicillin streptomycin. Luciferase expressing THP-1 were a gift by Dr. Largeaspada from the University of Minnesota. These cell lines were grown under standard conditions at 37 °C in a humidified atmosphere with 5% CO2. All cell culture media were obtained from Invitrogen Life Technologies (MA). To make a stock concentration, Triptolide (EMD Millipore, MA) was first diluted in dimethyl sulfoxide (DMSO) to obtain a 12.4 mM stock, stored at − 20 °C and further diluted to indicated concentrations. 10058-F4 (Catalog No: F3680 Millipore Sigma, MA, US) was diluted at a concentration of 1000 µM in DMSO and used at indicated doses.
Determination of cell viability and apoptosis
Cell Titer Glo (Catalog Number: G755A, Promega, WI) was used as per manufacturer’s directions to evaluate cell viability after treatment of AML cells in 96 well white opaque plates for the indicated time duration and drug concentrations. Apoptosis was measured using caspase 3 Glo assay (Catalog No: G8090, Promega, WI) as described by the manufacturer’s protocol. Briefly, leukemia cells were simultaneously assessed for cell viability as well as caspase 3/7 activity using a pro-luminescent caspase − 3/7 substrate which is cleaved by the caspase 3/7 enzyme and generates a glow type luminescent signal proportional to the caspase 3/7 activity. All experiments were done in triplicates and repeated three independent times.
Flow cytometric analysis of cell cycle
THP-1 and KG1a cells were grown at a concentration of 5 × 106 cells/ml in 5% RPMI for 24 h followed by changing to serum free media overnight. These cells were then treated with triptolide and harvested after 24 h of treatment. After fixing the cells with 70% ethanol, the leukemic cells were subsequently stained with Guava Cell Cycle Reagent for Flow Cytometry (Cat. No. 4500-0220, EMD Millipore, Billerica, MA) and analysed in BD FACSCanto II (BD Biosciences, CA) BD FACS Canto II (BD Biosciences, CA) using FACS Diva and FlowJo (Tree Star) software.
THP-1 and KG1a cells were treated overnight with indicated concentrations of triptolide and were subsequently washed and re-suspended in Human Methylcellulose complete media (Catalog Number: HSC003, R&D Systems, MN, US) in 24 well plates. Cluster of cells consisting of more than 40 cells were counted as colonies and were scored for number of colonies.
Quantitative real-time polymerase chain reaction and western blot
THP-1 and KG1a cells (5 × 10
5 cells/ml) were seeded in 6 well plates and treated with triptolide. These cells were then harvested after washing with PBS (phosphate buffered saline) and RNA was extracted using TRIzol (Life Technologies, MA) reagent according to the manufacturer’s instructions. To make cDNA, total RNA of 2 μg was used with the high capacity cDNA reverse transcription kit (Applied Biosystems, NY). The QuantiTect SYBR Green PCR kit (Qiagen, CA) was used according to the manufacturer’s instructions to run quantitative real-time PCR. Individual values of the genes were expressed relatively by normalizing to the housekeeping gene 18S. Primers were obtained from Integrated DNA Technologies Inc (IA) and are detailed in Additional file
3: Table S1.
For western blotting, AML cells after treatment with indicated drugs and concentrations were washed with PBS once and cell lysates were prepared by adding Rapid Immuno Precipitation Assay Buffer (Boston Bioproducts, MA) with protease (Catalog Number: A32963, Thermo Scientific, MA) (20 μl/ml of lysis buffer) and phosphatase inhibitor (Catalog Number: 78420, Thermo Scientific, MA) in ice. The lysates thus obtained were quantified using Pierce bicinchoninic acid protein estimation kit (Catalog Number: 23250, Thermo Scientific, MA) and electrophoresis was performed on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels by loading 30–50 μg of protein. The proteins were then transferred to a polyvinylidene difluoride (PVDF) membrane (Catalog Number: 1620177 BioRad, CA) and probed using the following antibodies c-Myc (Catalog Number: 13987, Cell Signaling Inc, MA) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Catalog Number: AB2302, EMD Millipore Inc., Germany) was used as the loading control.
In vivo experiments
NRGS Mice (NOD.Cg-Rag1tm1Mom Il2rgtm1Wjl Tg(CMV-IL3, CSF2, KITLG)1Eav/J) are NOD Rag 1 γc null mice that over express human IL-3 and GM-CSF. These mice were sub-lethally irradiated using 250 cGy of irradiation using an X ray irradiator and were injected with 2 × 106 leukemic cells via the tail vein on the same day. In another experiment, mice injected with Luciferase expressing THP-1 cells were serially followed for engraftment using In Vivo Imaging System. d-Luciferin (LUCK-1G, Gold Biotechnology Inc., MO) was prepared at a dose of 15 mg/ml and injected intraperitoneally at 10 μl/g of mouse body weight for measuring bioluminescence. IVIS Spectrum (PerkinElmer, MA, US) was used to measure luminescence followed by image quantitation and analysis using the Living Image Software (PerkinElmer, MA, US). For primary human AML blasts, NRGS mice were followed at various intervals for engraftment of leukemic cells by evaluating the level of human CD45+ (Catalog Number 347723, BD Biosciences) leucocytes in the peripheral blood. Briefly, peripheral blood was evaluated after red blood cell (RBC) lysis using ammonium–chloride–potassium (ACK) buffer. Live cells were stained with Anti hCD45 fluorescein isothiocyanate (FITC) and analyzed by flow cytometry. Animals expressing more than 2% of human CD45 were considered to have positive engraftment. Minnelide, dissolved in saline, was administered intra-peritoneally (i.p) at a dose of 0.05 mg/kg/day to 0.1 mg/kg/day.
Statistical analysis
In the corresponding figures, values are expressed as mean ± SD for in vitro experiments and mean ± SEM for in vivo experiments. For individual experiments, n and p values are mentioned in the corresponding figure legends. Survival was measured using Kaplan–Meier survival curve using Graph pad Prism software (GraphPad Software, inc, California). Data was analyzed using the Mann–Whitney unpaired test for in vivo experiments and in vitro experiments were analyzed with student t-test with two-sided p < 0.05 being considered significant.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.