Treatment with high-dose simvastatin inhibits geranylgeranylation in AML blast cells in a subset of AML patients

Experimental Hematology 2012;40:177–186

Treatment with high-dose simvastatin inhibits geranylgeranylation in AML blast cells in a subset of AML patients Karen van der Weidea,b, Susan de Jonge-Peetersa,b, Gerwin Hulsa, Rudolf S.N. Fehrmannb,c, Jan Jacob Schuringaa, Folkert Kuipersd, Elisabeth G.E. de Vriesb, and Edo Vellengaa a Department of Hematology; bDepartment of Medical Oncology; cDepartment of Genetics, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands; dDepartment of Pediatrics, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands

(Received 3 August 2011; revised 26 October 2011; accepted 22 November 2011)

It is currently unknown whether the in vitro effects observed with statins in acute myeloid leukemia (AML) cells, including lowering of cholesterol, inhibition of isoprenylation, and sensitization to chemotherapy, also occur in vivo. Therefore, AML mononuclear cells (MNCs) were isolated from 12 patients before and after 7 days of high-dose (7.5–15 mg/kg/ day) simvastatin treatment. Parallel mouse studies were performed to have, in addition to AML cells, access to liver tissue, a major target of statins. Serum cholesterol levels were lowered by simvastatin in all patients, however, only limited changes in the messenger RNA expression of cholesterol metabolism genes were seen in patient and mouse MNCs compared to murine liver cells. Still, two out of seven patients displayed an increased in vitro chemosensitivity of their AML cells upon simvastatin treatment. Gene set enrichment analysis on microarray data of AML patient cells and Western blot analysis for the isoprenylated proteins DnaJ and Rap1 on murine and AML patient MNCs demonstrated that in vivo simvastatin treatment resulted in inhibition of geranylgeranylation in murine MNCs and in a subset of patient AML MNCs. In summary, our data demonstrate that simvastatin treatment results in chemosensitization and inhibition of geranylgeranylation in AML cells of a subset of patients. Ó 2012 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc.

Different mechanisms have been identified that protect acute myeloid leukemia (AML) cells against the cytotoxic effects of chemotherapy, including adaptation of cellular cholesterol homeostasis [1,2]. Cellular cholesterol levels are maintained by balanced de novo synthesis, influx, and efflux. Cholesterol synthesis is initiated by 3-hydroxy-3methylglutaryl CoA reductase (HMG-CoAR), which is the rate-controlling enzyme of the mevalonate pathway that yields cholesterol as well as isoprenoids. The low-density lipoprotein receptor (LDLR) is responsible for cholesterol influx into cells [3]. The cholesterol efflux pumps ABCA1 and ABCG1 promote the transfer of cellular cholesterol to apolipoprotein A-I and high-density lipoprotein, which is regulated by liver X receptor, a nuclear receptor that is activated by oxidized cholesterol derivatives (oxysterols) [4–6]. Offprint requests to: Edo Vellenga, M.D., Ph.D., Department of Hematology, University Medical Center Groningen, Hanzeplein 1, 9713 GZ Groningen, The Netherlands; E-mail: [email protected] Supplementary data related to this article can be found online at doi: 10.1016/j.exphem.2011.11.008

Several studies have demonstrated that cholesterol metabolism is increased in AML cells, as reflected by increased expression of HMG-CoAR and LDLR [7–9]. In addition, AML cells increase their cholesterol levels upon in vitro exposure to chemotherapeutic drugs, which renders these cells less susceptible to these drugs [10,11]. Consequently, a role for statins to improve standard antileukemic treatment has been suggested. Statins inhibit HMG-CoAR activity, resulting in a blockade of cholesterol synthesis as well as inhibition of the production of isoprenoids, such as farnesyl and geranylgeranyl. Farnesylation or geranylgeranylation, collectively referred to as isoprenylation, is the transfer of a farnesyl or geranylgeranyl moiety to proteins. Isoprenylation is required for the attachment of small GTPases (e.g., Ras and Rho) to the plasma membrane and their subsequent participation in signal transduction pathways, such as the phosphoinositide 3-kinase/Akt and Ras/MEK/ERK pathways [12]. The cytotoxic effects of statins were originally attributed to their serum cholesterol-lowering capacity [13]. However, in vitro studies with AML cell lines and primary AML cells

0301-472X/$ - see front matter. Copyright Ó 2012 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. doi: 10.1016/j.exphem.2011.11.008

178

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

have suggested that the cytotoxic effects of simvastatin are caused by a blockade of the isoprenylation route rather than a blockade of cholesterol synthesis [14]. We aimed to verify whether the in vitro observed effects by statins are actually present in blast cells of AML patients treated with simvastatin. Additionally, we used a mouse model to validate our findings in more detail. The results show that, in a subset of patients, treatment with simvastatin sensitizes AML cells to in vitro chemotherapy. In addition, geranylgeranylation is inhibited in AML cells from simvastatin-treated patients, with only minor effects on the expression of cholesterol metabolism genes.

Materials and methods Patients and patient material Patient samples (n 5 12) were obtained from a phase II feasibility study in which AML patients were treated with high-dose simvastatin. Median age of the patients was 67 years (range, 23–82 years) and the male-to-female ratio was 9:3. For this study, eligible patients were those with AML at diagnosis or relapsed AML who would receive intensive chemotherapy (n 5 4) or palliative chemotherapy (n 5 8) according to the ongoing HOVON studies of the Dutch-Belgian Hemato-Oncology Cooperative Group (HOVON) [15,16]. Simvastatin was administered orally at a dose of 7.5 to 15 mg/kg/day during 7 days. Eight patients were treated with 7.5 mg/kg/day simvastatin and four patients received 15 mg/kg/day. On day 7 of the simvastatin treatment, chemotherapy was initiated. The study was approved by the Medical Ethical Committee of the University Medical Center Groningen. All patients provided written informed consent. Bone marrow (BM) or peripheral blood (PB) samples were collected before and after 7 days of simvastatin treatment, but before chemotherapy application. The mononuclear cell (MNC) fraction was stored and thawed as described earlier [17]. Serum lipid levels were monitored using standard laboratory techniques. For additional experiments BM or PB MNCs were used that were collected from AML patients at diagnosis after informed consent, as well as granulocyte colony-stimulating factor
mobilized peripheral blood stem cells from patients eligible for autologous hematopoietic stem cell transplantation in accordance with institutional guidelines. Animals As the liver is the primary target of statins, it is of interest to compare our findings in BM cells with those in liver cells. To have access to liver cells, a mouse study was performed. Eleven- to 13-week-old male C57Bl/6OlaHsd mice were purchased from Harlan (Horst, the Netherlands) and housed under clean conventional conditions. All experimental protocols were approved by the institutional ethical committee on animal experiments. The mice were fed normal chow (RMH-B; Arie Blok, Woerden, the Netherlands), chow containing 0.1% w/w simvastatin (Merck Chemical Ltd., Nottingham, UK) (7 days), chow containing 2% w/w WelChol (Colesevelam HCL; Daiichi Sankyo Inc., Munich, Germany) (14 days), or the combination of the two diets (7 days WelChol, followed by 7 days WelChol þ simvastatin). After this treatment, a PB sample was taken. The animal was thereafter sacrificed and liver was harvested and snap-frozen. BM cells were flushed from the femurs

and tibias. After standard erythrocyte lysis, nucleated cells were treated as described here later. Primary rat liver cells were obtained and cultured as described by Vrenken et al. [18]. Cell sorting and culture of human AML cells and murine BM cells The patient AML MNCs were incubated with a fluorescein isothiocyanate
conjugated antibody against CD34 (Becton Dickinson, San Jose, CA, USA). Sorting of CD34þ cells was performed with the use of a MoFlo Cell Sorter (DakoCytomation, Carpinteria, CA, USA). CD34þ cells and total MNCs were cultured in long-term culture medium, as described by van Gosliga et al. [19]. All cultures were kept at 37 C and 5% CO2. Mouse BM cells were cultured and analyzed as described in the Supplementary Material and Methods (online only, available at www.exphem.org). Cell viability Viability of human (CD34þ) AML cells and mouse BM cells was determined by the Cell Titer-Glo Luminescent cell viability assay (Promega, Madison, WI, USA) according to manufacturer’s instructions. Ninety-six
well plates were prepared with 100 mL medium supplemented as described and 10,000 cells per well. Cells were incubated with cytarabine (Mayne Pharma Benelux, Brussels, Belgium) and after 24 hours viability was assessed in duplicate. Microarray experiments, gene enrichment analysis, and quantitative polymerase chain reaction Microarray experiments were performed as described recently using human AML mononuclear cells [20]. Gene set enrichment analyis (GSEA) was performed with the software package GSEA 2.0, which was developed by the Broad Institute of the Massachusetts Institute of Technology and Harvard [21,22] as described elsewhere [23], with 37,804 genes. A total of 156 functional genes sets as reported in the Kyoto Encyclopedia of Genes and Genomes database, 125 from the Biocarta database (http://www.biocarta.com), and 83 as reported in the GenMAPP database were analyzed. The statistical significance of enrichment was determined using a randomization test based on 1000 gene permutations. For each functional set, the false discovery rate was calculated, which represents the estimated probability that a given enrichment score represents a false-positive finding. Only gene sets with an enrichment p value of !0.05 and a false discovery rate of !0.25 are reported. The leading-edge subset is defined as the subset of genes in a functional gene set that appears in the ranked list of 37,804 genes at, or before, the point in which the running enrichment score reaches its maximum deviation from zero. The genes within this subset can be interpreted as the most important in the enrichment of the functional gene set. Leading-edge subset analysis was done as described before [23]. Quantitative polymerase chain reaction on human AML MNCs and mouse BM and peripheral blood MNCs was performed as described earlier [17] using primers and probes for the human and murine ABC transporters and HMG-CoAR and LDLR [24–28]. The housekeeping gene glyceraldehyde 3-phosphate dehydrogenase served as endogenous control for patient material and 18S for mouse material. Western blotting To determine DnaJ and Rap1 protein expression cells (0.5  106 per mL) were cultured in the presence or absence of simvastatin

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

179

Table 1. Clinical and cellular characteristics of AML patients

AML

Age (y)

Blasts*

WBC (109/L)

FAB Classification

Cytogenetics

Risk group stratificationy

Treatment

Simvastatin (mg/kg/d)

CD34in AML MNCs (%)

65 60 72 23 82 73 33 57 68 66 76 76

50 56 24 38 22 38 80 O20 20 25 23 80

26.1 101.0 17.2 3.8 1.1 5.6 2.4 1.2 4.8 13.2 2.2 16.8

M5 M2 M2 M1 M1 M2 M5 M6 M2 M6 M2 M5

NK NK 45,X,-X,1-13dmin; 46,XX,1-23dmin 46,XY,5q-,9qþ & 46,XY, t(5,9)? NK NK t6;11 NK 7qNK NK ND

Intermediate Intermediate Poor Poor Intermediate Intermediate Poor Intermediate Poor Intermediate Intermediate ND

it pt pt pt pt pt pt it it it pt pt

7.5 7.5 15 7.5 15 15 7.5 7.5 7.5 7.5 15 7.5

8 36 !1 23 33 29 20 24 45 6 26 !1

1 2 3 4 5 6 7 8 9 10 11 12

FAB 5 French-American-British; it 5 intensive treatment; pt 5 palliative treatment; ND 5 not determined; NK 5 normal karyotype; WBC 5 white blood cell counts. *Bone marrow blast percentage at presentation. y Risk group stratification is based on (un)favorable cytogenetics combined with peripheral blood blast cell counts ($20  100/L according to HOVON criteria of the Dutch-Belgian Hemato-Oncology Cooperative Group).

(Merck Chemical Ltd.) for 24 hours. Experimental procedures are described in the Supplementary Material and Methods (online only, available at www.exphem.org). Densitometry was carried out using ImageJ [29]. Statistics The nonparametric Wilcoxon signed rank test was used to test whether there were differences in serum lipid concentrations, blood parameters, and gene expression between before and after treatment of AML patients with simvastatin. Student’s t test was used in the other analyses.

Results Treatment of AML patients with simvastatin decreases in vivo cholesterol synthesis Main characteristics of the treated patients (n 5 12) are shown in Table 1. To monitor the in vivo effectiveness of simvastatin, serum lipid levels were analyzed before and at the end of simvastatin treatment. The lathosterol-tocholesterol ratio, which is generally used as a measure for total body cholesterol synthesis [30], was decreased after 7 days of statin treatment (0.71 6 0.57 vs 0.23 6 0.21; p 5 0.02; Fig. 1A). No distinct changes were observed in the PB cell counts or the blast percentages of the PB (data not shown). Treatment with simvastatin increases in vitro sensitivity of sorted AML CD34þ cells to chemotherapy in a subset of patients In vitro simvastatin sensitizes primary AML cells to chemotherapy [2]. We investigated whether the AML cells exposed to simvastatin in vivo are more sensitive to in vitro chemotherapy than their untreated counterparts. Previous studies have shown that the stem cell enriched CD34þ

(AML) cell fraction displays higher levels of cholesterol metabolism genes [25,29], and is therefore likely more dependent on high cholesterol turnover. Therefore, we sorted CD34þ AML cells from samples that were collected before and at the end of simvastatin treatment. In 7 of 12 patients, sufficient (O500,000) CD34þ cells could be collected. In vitro culture with cytarabine increased cytotoxicity in 2 of 7 AML CD34þ cells exposed in vivo to simvastatin (Fig. 1B). Viabilities of these two responsive samples were, upon cytarabine treatment, 58% and 62% before and 26% and 47% after simvastatin pretreatment, respectively, while cells of the remaining five patients showed 95% 6 9% viability before and 95% 6 5% after in vivo simvastatin. Mononuclear cells of AML patients treated with simvastatin show limited changes in cholesterol metabolism genes To delineate the changes that occur in AML cells upon treatment of patients with simvastatin, microarray experiments were performed on paired AML MNCs (n 5 10) obtained before and at the end of simvastatin treatment. In the majority of patients simvastatin treatment resulted in slight increases in HMG-CoAR (6 of 10) and LDLR (8 of 10) messenger RNA (mRNA) levels in AML MNCs (1.3-fold 6 0.2-fold; range, 1.1–1.6-fold; and 1.8-fold 6 0.9-fold; range, 1.1–3.9-fold, respectively), whereas ABCA1 (7 of 10) and ABCG1 (6 of 10) were decreased (2.3-fold 6 1.0-fold; range, 1.2–4.1-fold; and 3.4-fold 6 1.6-fold; range, 1.7–6.1-fold, respectively) (Fig. 1C). However, these changes were not significant and not evident in all samples, despite the consistent decline in serum cholesterol levels. In addition, the degree of gene expression changes in AML cells did not relate to the degree of decrease in serum cholesterol levels. The expression levels of the tested genes were

180

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Figure 1. In vivo effects of simvastatin on serum lipid levels and cholesterol metabolism gene expression in cells of AML patients. (A) The lathosterol-tocholesterol ratio of patients before (d0) and after 7 days (d7) of treatment with simvastatin. (B) Viability of sorted CD34þ cells obtained from patients before (d0) and after (d7) treatment simvastatin upon in vitro treatment with 0.01 mg/mL cytarabine vs untreated control cells (100%), as shown for individual patients. (C) Gene expression, as determined by microarray analysis, of HMG-CoAR, LDLR, ABCA1, and ABCG1 in the total fraction of AML MNCs (n 5 10) before (d0) and after 7 days of simvastatin treatment. Closed circles: 7.5 mg/kg/day; open squares: 15 mg/kg/day. *p ! 0.05.

confirmed by quantitative polymerase chain reaction (Supplementary Figure E1; online only, available at www. exphem.org). Biological pathway analysis To identify biological pathways affected by simvastatin treatment, GSEA was performed (Supplementary Figure E2; online only, available at www.exphem.org). Using Kyoto Encyclopedia of Genes and Genomes pathway definitions, GSEA identified 41 pathways enriched in simvastatin-exposed samples (Supplementary Table E1; online only, available at www.exphem.org), of which the most significantly enriched pathways are shown in Table 2A. No pathways were enriched in untreated samples vs simvastatin-exposed samples, when p ! 0.05 and false discovery rate !0.25 were used. GSEA revealed 18 and 12 pathways enriched in simvastatin-exposed AML samples (Table 2B, C and Supplementary Tables E2, E3;

online only, available at www.exphem.org) according to Biocarta and GenMAPP, respectively. Based on the mechanism of action of simvastatin, we expected to find pathways involved in cholesterol synthesis and cell signaling due to inhibition of isoprenylation of small GTPases. These pathways were indeed affected, e.g., the Rho pathway requires isoprenylated Rho and the vascular endothelial growth factor pathway may involve isoprenylation of Ras. However, the genes that contributed to enrichment of the cholesterol (or steroid) synthesis pathway did not include HMG-CoAR or LDLR, which are known to be increased in vitro. Leading edge analysis revealed key regulatory genes common to the identified pathways, such as AKT1 and MAPK3 (ERK1) for Kyoto Encyclopedia of Genes and Genomes pathways and PRKCB1 and PRKCA for Biocarta (Supplementary Table E4; online only, available at www.exphem.org), suggesting that isoprenylationdependent signaling may have been affected.

K. van der Weide et al./ Experimental Hematology 2012;40:177–186 Table 2. Gene sets enriched in MNCs using pathway definitions from Kyoto Encyclopedia of Genes and Genomes (A), Biocarta (B), and GenMAPP (C) with p ! 0.001

Pathway

False discovery Enriched p Value rate in

A Hematopoietic cell lineage Cell adhesion molecules Natural killer cell
mediated cytotoxicity Type I diabetes mellitus T cell receptor signaling pathway Biosynthesis of steroids VEGF signaling pathway Antigen processing and presentation Cytokine cytokine receptor interaction Wnt signaling pathway MAPK signaling pathway Calcium signaling pathway Regulation of actin cytoskeleton

!0.001 !0.001 !0.001

0.00 0.00 0.00

Post Post Post

!0.001 !0.001 !0.001 !0.001 !0.001 !0.001 !0.001 !0.001 !0.001 !0.001

0.00 0.00 0.01 0.01 0.01 0.02 0.02 0.05 0.05 0.11

Post Post Post Post Post Post Post Post Post Post

Caspase pathway Chemical pathway TCR pathway GPCR pathway Rho pathway HIVNef pathway

!0.001 !0.001 !0.001 !0.001 !0.001 !0.001

0.01 0.01 0.01 0.02 0.02 0.05

Post Post Post Post Post Post

Apoptosis GenMAPP Apoptosis Apoptosis Kyoto Encyclopedia of Genes and Genomes Pyruvate metabolism Propanoate metabolism Calcium regulation in cardiac cells

!0.001 !0.001 !0.001

0.00 0.00 0.00

Post Post Post

!0.001 !0.001 !0.001

0.00 0.01 0.06

Post Post Post

B

C

GPCR 5 G-protein-coupled receptor; MAPK 5 mitogen-activated protein kinase; TCR 5 T-cell receptor; VEGF 5 vascular endothelial growth factor.

Differential effects of simvastatin on cholesterol metabolism genes expression in murine bone marrow and liver cells Next, we questioned whether the absence of changes in cholesterol metabolism genes in MNCs, despite the physiological decrease of serum cholesterol levels, reflects celltype
dependent effects. Therefore, a mouse model was used, which enabled to make a distinction between the effects of simvastatin on hematopoietic cells and liver cells, the latter being the major source of cholesterol production. Mice treated with simvastatin or simvastatin in combination with WelChol, to mimic the increased cholesterol metabolism in AML cells, displayed no different cell counts in BM and PB composition compared to their untreated counterparts (Supplementary Table E5; online only, available at www.exphem.org). BM progenitor frequencies, as determined by fluorescence-activated cell sorting analysis and colony-forming cell assay remained unchanged, although a slight increase in Lin Sca Kit (LSK) number (6.3% 6

181

0.5% to 8.3% 6 1.4%; p 5 0.03) upon simvastatin treatment was observed (Supplementary Figure E3; online only, available at www.exphem.org). In addition, we compared changes in expression of cholesterol metabolism genes in murine BM MNCs with liver cells upon in vivo treatment with simvastatin. In liver cells, simvastatin induced LDLR and HMG-CoAR mRNA expression by 2- (p 5 0.02) and 12-fold (p ! 0.001), respectively, whereas no changes in ABCA1 and ABCG1 expression were noticed (Fig. 2). BM MNCs demonstrated no changes in LDLR and HMG-CoAR expression upon simvastatin-treatment. In contrast to liver cells, ABCA1 and ABCG1 mRNA expression was reduced in simvastatin-treated MNCs (1.6-fold; p 5 0.004 and 2.1fold; p 5 0.005), respectively). Pretreatment with WelChol to induce cholesterol synthesis increased the expression of LDLR 2-fold (p 5 0.009) and HMG-CoAR was 8-fold increased (p ! 0.001) in liver cells. HMG-CoAR increased even 47-fold in these cells when WelChol treatment was combined with simvastatin (p 5 0.003). In contrast, treatment with both WelChol and simvastatin did not affect expression of cholesterol metabolism genes in the BM MNC. In addition, the expression in circulating MNCs was not affected by simvastatin treatment (Fig. 2). Simvastatin treatment of patients and mice inhibits geranylgeranylation in BM cells An alternative way in which statins might affect BM MNCs is by inhibition of farnesylation and/or geranylgeranylation. This would be in line with our GSEA data, demonstrating that simvastatin affected pathways involved in signal transduction (Supplementary Tables E1–E3; online only, available at www.exphem.org) that often require isoprenylation of upstream regulators. Therefore, we determined whether in vivo simvastatin treatment of mice also resulted in inhibition of isoprenylation. No inhibition of farnesylation was observed in mouse BM MNCs and liver cells upon in vivo treatment with simvastatin. However, in vivo treatment of mice with simvastatin resulted in differential effects on geranylgeranylation in BM and liver cells: simvastatin increased unprocessed Rap1 levels, a protein that is exclusively geranylgeranylated, in BM MNCs (Fig. 3A), which could not be demonstrated in liver cells (Fig. 3B). Also, in one of the four AML patients tested, inhibition of geranylgeranylation was detected upon in vivo treatment with simvastatin (Fig. 3C), whereas no inhibition of farnesylation was observed. Inhibition of geranylgeranylation occurs at a low concentration of simvastatin Next, we analyzed whether the limited effects of treatment with simvastatin on cholesterol metabolism gene expression in BM MNCs may be related to the relative low concentration achieved in the BM compartment. In vitro treatment of primary human and mouse BM MNCs demonstrated that effects on cell survival, cholesterol metabolism

182

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Figure 2. mRNA expression of cholesterol metabolism genes upon in vivo treatment with simvastatin. mRNA expression of HMG-CoA reductase, LDL receptor, ABCA1, and ABCG1 was determined in BM, PB, and liver cells of mice that were fed with a diet containing 0.1% w/w simvastatin (S), 2% w/w welchol (W), or both (WþS). C: control chow fed mice. Mean 6 standard deviation is shown for 11 (control and simvastatin), or 3 (welchol and simvastatin þ welchol) mice per group. 18S served as a housekeeping gene. Expression is shown as the relative expression vs 18S, with the first control set at one. ND 5 not determined. *p ! 0.05.

gene expression and the degree of farnesylation were pronounced only at high concentrations that are not likely to be achieved in vivo (O5 mM) [31] (Fig. 4A, Supplementary Figure E4; online only, available at www.

exphem.org) when incubated for 24 hours. Inhibition of Rap1 geranylgeranylation, however, occurred at concentrations as low as 0.2 mM simvastatin in the culture medium and was maximal at 1 to 5 mM in both mouse and human

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Figure 3. Inhibition of isoprenylation upon in vivo treatment with simvastatin. Inhibition of geranylgeranylation of Rap1 and of farnesylation of DnaJ in control mice (chow) and mice treated for 7 days with 0.1% w/w simvastatin in BM MNCs (A) and liver (B) cells. Unprenylated (unpren) and total levels of Rap1 are shown, as well as unprenylated and prenylated (pren) levels of DnaJ for seven mice in each group. (C) Prenylation status of MNCs from four AML patients before (c) and after 7 days of simvastatin treatment (s).

normal BM MNCs (Fig. 4B). In contrast, in primary hepatocytes, inhibition of geranylgeranylation was observed only at concentrations of 25 mM or higher (Fig. 4B).

Discussion The present study is the first to demonstrate that treatment of AML patients with high-dose simvastatin inhibits isoprenylation in their MNCs, but does not affect the expression of cholesterol metabolism genes in these cells, despite a reduction in serum cholesterol levels. This striking finding provides a novel insight in the mechanism behind statininduced effects in normal and leukemic hematopoietic cells. Statin-treatment resulted in an increased in vitro sensitivity of AML CD34þ cells to chemotherapy in 2 of

183

7 patients. These findings extend in vitro data demonstrating that AML cells display an increased sensitivity to chemotherapy upon statin treatment [10,11]. Differences in response to chemotherapy combined with statins were reported before in a clinical study with pravastatin combined with Idarubicin and high-dose Ara-C [13]. These differences were associated with the occurrence of cholesterol rebounds after the initial decrease of cholesterol levels, but can also be due to other interpatient differences in cholesterol metabolism [13,32]. However, in our study chemosensitivity was not related to basal or simvastatinaffected expression levels of cholesterol metabolism genes. Only AML samples that are in vitro sensitive to chemotherapy become more sensitive when in vivo pretreated with simvastatin (Fig. 1B). This is in line with an earlier report showing that in vitro combination treatment with lovastatin and daunorubicin or cytarabine was only more effective in AML samples that displayed sensitivity to chemotherapy alone [33]. Simvastatin-treatment of the AML patients resulted in a significant decline in serum cholesterol levels, reflecting a decreased cholesterol synthesis by the liver. We did see that cholesterol metabolism gene expression was affected in mouse liver cells by statin treatment. However, only limited effects were found on the mRNA expression of single cholesterol metabolism genes in the total fraction from AML patients who underwent simvastatin treatment. This challenges previous in vitro reports describing that statins predominantly affect cholesterol metabolism in AML cells [2,10]. Apparently, simvastatin-mediated effects are dependent on the cell type investigated [34]. Alternatively, the plasma concentration of simvastatin might not be sufficiently high due to low bioavailability [35,36]. Peak plasma levels upon treatment with 4 mg/kg/day lovastatin, a statin that is pharmacokinetically comparable to simvastatin [36,37], ranged from 0.1 to 3.9 mM [38]. With in vitro experiments, we demonstrated that this concentration has limited effects on expression of cholesterol metabolism genes in AML cells. Studying the expression of single genes did not reveal clear changes upon simvastatin treatment. However, subtle but coordinated changes in gene expression within a certain pathway can have significant biological effects. With GSEA, we showed that there were signaling pathways affected by simvastatin treatment, of which some participants require isoprenylation. This observation elaborates on the previous finding that the synthesis of isoprenoids, products of the mevalonate pathway, is inhibited by statins in vitro [14,39]. Prenylated proteins belong mainly to the superfamily of Ras-GTPases [40]. In many AML patients, Ras (predominantly N-Ras) is constitutively active due to mutations or autocrine production of growth factors [41,42]. Interestingly, in vitro data indicate that, e.g., Ras signaling can be affected by statins, which results in inhibition of a number of downstream signaling pathways, including the MEK/ ERK pathway [43–45].

184

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Figure 4. In vitro effects of simvastatin on prenylation status. Inhibition of farnesylation of DnaJ (A) and geranylgeranylation of Rap1 (B) after 24 hours of treatment with simvastatin in primary mouse BM MNCs, rat hepatocytes, human AML MNCs, and normal human MNCs. unpren: unprenylated protein; pren: prenylated protein. Representative blots of at least three experiments are shown.

It appeared that simvastatin inhibited mainly geranylgeranylation in AML cells of simvastatin treated patients. Our in vitro data reflect these findings, as we observed inhibition of geranylgeranylation in AML MNCs and mouse BM cells at a low simvastatin concentration, whereas other effects, e.g., inhibition of farnesylation or changes in cholesterol metabolism genes, could only be observed at concentrations that were 25- to 100-fold higher. These high concentrations are also required to inhibit ERK phosphorylation and to induce cytotoxic effects [14]. As such high simvastatin levels cannot be achieved in vivo [38], it is unlikely that in vivo treatment with statins will be effective in totally blocking Ras/MEK/ERK signaling. For future research, it would be attractive to combine statins with tipifarnib, a farnesyltransferase inhibitor, since the combined use of these agents additively blocked ERK-signaling in AML cells in vitro [17] and might affect both farnesylation and geranylgeranylation at clinical achievable concentrations. Inhibition of geranylgeranylation and chemosensitization was not found in all patients. Our previous in vitro data showed that, within different patient samples, a heterogeneity in response to simvastatin exists [14,17]. Not surprisingly, this also holds true for the clinical setting, and may be even more affected by interpatient differences that do not read out in vitro, such as varying simvastatin plasma levels due to variation in, e.g., CYP3A4 activity [32]. GSEA on the total AML MNC fraction revealed also pathways involved in immune signaling to be affected by simvas-

tatin. These findings challenge to interpret the studies in AML patients that underwent allogeneic stem cell transplantation differently. Here, a reduced incidence of graft-vs-host disease was observed when the donors had been using low-dose simvastatin [46]. These effects of simvastatin may be attributed to their effect on prenylated proteins (e.g., Ras, Rac, Rho, Rap1, and CDC42), which are involved in immune function [47]. In summary, the present study demonstrates that treatment of AML patients with simvastatin inhibits primarily geranylgeranylation in a subset of patients in the absence of marked changes in cholesterol metabolism gene expression in AML MNCs. This finding should be confirmed in more patients, as we have shown this in one out of four patient samples so far. Nevertheless, this finding breaks with the conventional idea that inhibition of cholesterol synthesis does the trick in the treatment of AML patients with simvastatin. Our findings encourage further research to exploit statin-induced inhibition of geranylgeranylation in AML patients.

Funding disclosure This work was supported by a grant of the Dutch Cancer Society (RUG 2006-3580).

Acknowledgments We would like to acknowledge Bertien Dethmers-Ausema for her skillful assistance with the mouse experiments, Annet Vos and

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Ingrid Leegte for their technical assistance with the microarray experiments, and Henk Moes and Geert Mesander for their assistance with the MoFlo Cell Sorter.

18.

Conflict of interest disclosure No financial interest/relationships with financial interest relating to the topic of this article have been declared.

19.

20.

References 1. Jonge-Peeters SD, Kuipers F, de Vries EG, Vellenga E. ABC transporter expression in hematopoietic stem cells and the role in AML drug resistance. Crit Rev Oncol Hematol. 2007;62:214–226. 2. Li HY, Appelbaum FR, Willman CL, Zager RA, Banker DE. Cholesterol-modulating agents kill acute myeloid leukemia cells and sensitize them to therapeutics by blocking adaptive cholesterol responses. Blood. 2003;101:3628–3634. 3. Goldstein JL, Brown MS. The LDL receptor. Arterioscler Thromb Vasc Biol. 2009;29:431–438. 4. Wang N, Lan D, Chen W, Matsuura F, Tall AR. ATP-binding cassette transporters G1 and G4 mediate cellular cholesterol efflux to highdensity lipoproteins. Proc Natl Acad Sci U S A. 2004;101:9774–9779. 5. Wang N, Silver DL, Costet P, Tall AR. Specific binding of ApoA-I, enhanced cholesterol efflux, and altered plasma membrane morphology in cells expressing ABC1. J Biol Chem. 2000;275:33053–33058. 6. Kaneko E, Matsuda M, Yamada Y, et al. Induction of intestinal ATPbinding cassette transporters by a phytosterol-derived liver X receptor agonist. J Biol Chem. 2003;278:36091–36098. 7. Vitols S, Gahrton G, Ost A, Peterson C. Elevated low density lipoprotein receptor activity in leukemic cells with monocytic differentiation. Blood. 1984;63:1186–1193. 8. Vitols S, Norgren S, Juliusson G, Tatidis L, Luthman H. Multilevel regulation of low-density lipoprotein receptor and 3-hydroxy-3methylglutaryl coenzyme A reductase gene expression in normal and leukemic cells. Blood. 1994;84:2689–2698. 9. Ho YK, Smith RG, Brown MS, Goldstein JL. Low-density lipoprotein (LDL) receptor activity in human acute myelogenous leukemia cells. Blood. 1978;52:1099–1114. 10. Banker DE, Mayer SJ, Li HY, et al. Cholesterol synthesis and import contribute to protective cholesterol increments in acute myeloid leukemia cells. Blood. 2004;104:1816–1824. 11. Stirewalt DL, Appelbaum FR, Willman CL, Zager RA, Banker DE. Mevastatin can increase toxicity in primary AMLs exposed to standard therapeutic agents, but statin efficacy is not simply associated with ras hotspot mutations or overexpression. Leuk Res. 2003;27:133–145. 12. Steelman LS, Abrams SL, Whelan J, et al. Contributions of the Raf/ MEK/ERK, PI3K/PTEN/Akt/mTOR and Jak/STAT pathways to leukemia. Leukemia. 2008;22:686–707. 13. Kornblau SM, Banker DE, Stirewalt D, et al. Blockade of adaptive defensive changes in cholesterol uptake and synthesis in AML by the addition of pravastatin to idarubicin þ high-dose Ara-C: a phase 1 study. Blood. 2007;109:2999–3006. 14. van der Weide K, Korthuis PM, Kuipers F, de Vries EG, Vellenga E. Heterogeneity in simvastatin-induced cytotoxicity in AML is caused by differences in Ras-isoprenylation. Leukemia; 2011;doi:10.1038/leu.2011.259. 15. Lowenberg B, Ossenkoppele GJ, van PW, et al. High-dose daunorubicin in older patients with acute myeloid leukemia. N Engl J Med. 2009;361:1235–1248. 16. Vellenga E, van PW, Ossenkoppele GJ, et al. Autologous peripheral blood stem cell transplantation for acute myeloid leukemia. Blood. 2011;118:6037–6042. 17. van der Weide K, Jonge-Peeters SD, Kuipers F, de Vries EG, Vellenga E. Combining simvastatin with the farnesyltransferase inhibitor tipifarnib

21.

22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

32.

33.

34.

35. 36.

185

results in an enhanced cytotoxic effect in a subset of primary CD34þ acute myeloid leukemia samples. Clin Cancer Res. 2009;15:3076–3083. Vrenken TE, Buist-Homan M, Kalsbeek AJ, Faber KN, Moshage H. The active metabolite of leflunomide, A77 1726, protects rat hepatocytes against bile acid-induced apoptosis. J Hepatol. 2008; 49:799–809. van Gosliga D, Schepers H, Rizo A, et al. Establishing long-term cultures with self-renewing acute myeloid leukemia stem/progenitor cells. Exp Hematol. 2007;35:1538–1549. de Jonge HJ, Woolthuis CM, Vos AZ, et al. Gene expression profiling in the leukemic stem cell-enriched CD34(þ) fraction identifies target genes that predict prognosis in normal karyotype AML. Leukemia; 2011 Jul 15;doi:10.1038/leu.2011.172. [Epub ahead of print]. Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102:15545–15550. Mootha VK, Lindgren CM, Eriksson KF, et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet. 2003;34:267–273. Leffers N, Fehrmann RS, Gooden MJ, et al. Identification of genes and pathways associated with cytotoxic T lymphocyte infiltration of serous ovarian cancer. Br J Cancer. 2010;103:685–692. Grefhorst A, Elzinga BM, Voshol PJ, et al. Stimulation of lipogenesis by pharmacological activation of the liver X receptor leads to production of large, triglyceride-rich very low density lipoprotein particles. J Biol Chem. 2002;277:34182–34190. Kruit JK, Plosch T, Havinga R, et al. Increased fecal neutral sterol loss upon liver X receptor activation is independent of biliary sterol secretion in mice. Gastroenterology. 2005;128:147–156. Plosch T, Kok T, Bloks VW, et al. Increased hepatobiliary and fecal cholesterol excretion upon activation of the liver X receptor is independent of ABCA1. J Biol Chem. 2002;277:33870–33877. Peeters SD, van der Kolk DM, de Haan G, et al. Selective expression of cholesterol metabolism genes in normal CD34þCD38- cells with a heterogeneous expression pattern in AML cells. Exp Hematol. 2006; 34:622–630. Lowenberg B, Boogaerts MA, Daenen SM, et al. Value of different modalities of granulocyte-macrophage colony-stimulating factor applied during or after induction therapy of acute myeloid leukemia. J Clin Oncol. 1997;15:3496–3506. Abramoff MD, Magelhaes PJ, Ram SJ. Image processing with ImageJ. Biophotonics Int. 2004;11:36–42. Kempen HJ, Glatz JF, Gevers Leuven JA, van der Voort HA, Katan MB. Serum lathosterol concentration is an indicator of whole-body cholesterol synthesis in humans. J Lipid Res. 1988;29:1149–1155. de Grouw EP, Raaijmakers MH, Boezeman JB, et al. Preferential expression of a high number of ATP binding cassette transporters in both normal and leukemic CD34þ. Leukemia. 2006;20:750–754. Kajinami K, Takekoshi N, Brousseau ME, Schaefer EJ. Pharmacogenetics of HMG-CoA reductase inhibitors: exploring the potential for genotype-based individualization of coronary heart disease management. Atherosclerosis. 2004;177:219–234. Jonge-Peeters SD, van der Weide K, Kuipers F, et al. Variability in responsiveness to lovastatin of the primitive CD34þ AML subfraction compared to normal CD34þ cells. Ann Hematol. 2009;88: 573–580. Dimitroulakos J, Ye LY, Benzaquen M, et al. Differential sensitivity of various pediatric cancers and squamous cell carcinomas to lovastatininduced apoptosis: therapeutic implications. Clin Cancer Res. 2001;7: 158–167. Chan KK, Oza AM, Siu LL. The statins as anticancer agents. Clin Cancer Res. 2003;9:10–19. Corsini A, Bellosta S, Baetta R, et al. New insights into the pharmacodynamic and pharmacokinetic properties of statins. Pharmacol Ther. 1999;84:413–428.

186

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

37. Shitara Y, Sugiyama Y. Pharmacokinetic and pharmacodynamic alterations of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors: drug-drug interactions and interindividual differences in transporter and metabolic enzyme functions. Pharmacol Ther. 2006;112:71–105. 38. Thibault A, Samid D, Tompkins AC, et al. Phase I study of lovastatin, an inhibitor of the mevalonate pathway, in patients with cancer. Clin Cancer Res. 1996;2:483–491. 39. van de Donk NW, Kamphuis MM, van KB, Lokhorst HM, Bloem AC. Inhibition of protein geranylgeranylation induces apoptosis in myeloma plasma cells by reducing Mcl-1 protein levels. Blood. 2003; 102:3354–3362. 40. Konstantinopoulos PA, Karamouzis MV, Papavassiliou AG. Posttranslational modifications and regulation of the RAS superfamily of GTPases as anticancer targets. Nat Rev Drug Discov. 2007;6: 541–555. 41. Reuter CW, Morgan MA, Bergmann L. Targeting the Ras signaling pathway: a rational, mechanism-based treatment for hematologic malignancies? Blood. 2000;96:1655–1669.

42. Birkenkamp KU, Geugien M, Schepers H, et al. Constitutive NFkappaB DNA-binding activity in AML is frequently mediated by a Ras/PI3-K/PKB-dependent pathway. Leukemia. 2004;18:103–112. 43. van de Donk NW, Kamphuis MM, Lokhorst HM, Bloem AC. The cholesterol lowering drug lovastatin induces cell death in myeloma plasma cells. Leukemia. 2002;16:1362–1371. 44. Dai Y, Khanna P, Chen S, et al. Statins synergistically potentiate 7hydroxystaurosporine (UCN-01) lethality in human leukemia and myeloma cells by disrupting Ras farnesylation and activation. Blood. 2007;109:4415–4423. 45. Wong J, Quinn CM, Brown AJ. Statins inhibit synthesis of an oxysterol ligand for the liver x receptor in human macrophages with consequences for cholesterol flux. Arterioscler Thromb Vasc Biol. 2004;24:2365–2371. 46. Rotta M, Storer BE, Storb RF, et al. Donor statin treatment protects against severe acute graft-versus-host disease after related allogeneic hematopoietic cell transplantation. Blood. 2010;115:1288–1295. 47. Greenwood J, Steinman L, Zamvil SS. Statin therapy and autoimmune disease: from protein prenylation to immunomodulation. Nat Rev Immunol. 2006;6:358–370.

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

186.e1

Supplementary materials and methods Mouse stem cell and progenitor assays Mouse BM cells were cultured in StemSpan (StemCell Technologies, Grenoble, France) supplemented with 10% fetal calf serum, 20 ng/mL recombinant mouse interleukin-11 (R&D Systems, Minneapolis, MN, USA), 300 ng/mL polyethylene glycol-complexed recombinant rat stem cell factor (Amgen, Thousand Oaks, CA, USA), and 10 ng/mL Flt3-ligand (Amgen). To quantify the number of mouse progenitors after diet, the number of colony-forming units
granulocyte/macrophage was determined using standard methylcellulose cultures (0.8% methylcellulose; Sigma-Aldrich), 30% fetal calf serum in a
minimum essential medium (StemCell Technologies) supplemented with 20 ng/mL recombinant murine granulocytemacrophage colony-stimulating factor (R&D Systems) and 100 ng/mL polyethylene glycol-complexed recombinant rat stem cell factor (Amgen). Cells were cultured at 37 C and 5% CO2, and counted on day 7 using an inverted microscope. Stem cell and progenitor frequencies were determined by fluorescence-activated cell sorting analysis on the basis of the combinatorial expression of cell surface antigens. Cells were stained with a panel of biotin-conjugated lineage-specific antibodies (containing antibodies to CD3e, CD11b [Mac1], CD45R/B220, Gr-1 [Ly-6G and Ly-6C], and TER-119 [Ly-76]), phycoerythrin-conjugated Sca-1, allophycocyanin-conjugated antibody to c-Kit, PacificBlue-conjugated antibody to CD34, and phycoerythrin-Cy7-conjugated antibody to CD16/32. After being washed, cells were incubated with streptavidinallophycocyanin-Cy7. All antibodies were purchased from Pharmingen. The fluorescence-activated cell-sorting analyses were performed on an LSR-II (BD Biosciences). Lineage-depleted (Lin
) BM cells were defined as the 5% of cells showing the least allophycocyanin-Cy7 intensity. Hematopoietic stem cells were defined as Lin
Sca-1þc-kitþ and progenitors as Lin
Sca-1
c-kitþ. In addition, the progenitor cell fraction was separated into common myeloid progenitors (Lin
Sca-1þ c-kitþCD34þCD16/32lo), granulocyte-macrophage progenitors (Lin
Sca-1þc-kitþCD34þCD16/32þ), and megakaryocyteerythroid progenitors (Lin
Sca-1þc-kitþCD34
CD16/32
). Data were analyzed using FlowJo (Tree Star, Ashland, OR, USA) software. Western blotting Whole-cell extracts were obtained by lysing 5  105 cells in boiling Laemmli sample buffer for 5 minutes. Proteins were separated by 10% or 12.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane (Millipore, Bedford, MA, USA) in Tris-buffer using a semidry electroblotter from Bio-Rad Laboratories (Veenendaal, The Netherlands). After blocking in 0.1% Tween-20 containing 5% skim powdered milk and 2% bovine serum albumin in Tris-buffered saline, membranes were probed with antibodies against DnaJ (HDJ-2; Labvision, Fremont, CA, USA), Rap1 (total) and Rap1a (unprenylated) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) according to manufacturer’s protocols. Antibody binding was visualized with enhanced chemiluminescence detection or using an Odyssey infrared scanner (Li-Cor Biosciences, Lincoln, NE, USA) after incubation with a horseradish peroxidase
conjugated (Dako, Glostrup, Denmark) or an Alexa680- or IRDye800-labeled secondary antibody (Invitrogen).

186.e2

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Supplementary Figure E1. Verification of microarray expression data of cholesterol metabolism genes by quantitative reverse transcriptase polymerase chain reaction (PCR). Expression of HMG-CoAR, LDLR, ABCA1, and ABCG1 in three total MNC fractions and one CD34þ cell fraction, as determined by microarray analysis, was compared with quantitative PCR expression levels by linear regression.

Supplementary Figure E2. Examples of GSEA enrichment plots for the most enriched pathways in samples from simvastatin-treated AML patients based on Kyoto Encyclopedia of Genes and Genomes (A), Biocarta (B), and Genmap (C) pathways.

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

186.e3

Supplementary Figure E3. In vivo effects of simvastatin on progenitor frequencies of mouse BM cells. Fluorescence-activated cell sorted staining for stem cells and progenitors (A) and a colony-forming assay (B) using BM MNCs from control mice (chow) and from mice treated with 0.1% w/w simvastatin (simva). Lin Sca Kit (LSK) and progenitors are shown as a percentage within the 5% lineage-negative cells; common myeloid progenitors (CMP), granulocyte-macrophage progenitors (GMP), and megakaryocyte-erythroid progenitors (MEP) are shown as a percentage of the total progenitor fraction. Mean 6 standard deviation is shown for four control mice and five simvastatin-fed mice. *p ! 0.05.

Supplementary Figure E4. In vitro effects of simvastatin cell viability and gene expression. Cell viability of mouse BM (A) and human AML MNCs (B) upon treatment with different simvastatin concentrations for 24 hours. mRNA expression of cholesterol metabolism genes in mouse BM (C) and human AML MNCs (D) upon simvastatin treatment. Control was set at 18S (mouse) or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (human) served as a housekeeping gene. Data are shown as mean 6 standard deviation of n 5 7 (A), n 5 12 (B), n 5 8 (C), and n 5 6 (D). *p ! 0.05 vs control.

186.e4

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Supplementary Table E1. Results of gene set enrichment analysis of MNCs using pathway definitions from Kyoto Encyclopedia of Genes and Genomes

Pathway

p Value

False discovery rate

Hematopoietic cell lineage Cell adhesion molecules Natural killer cell
mediated cytotoxicity Type 1 diabetes mellitus T-cell receptor signaling pathway Biosynthesis of steroids VEGF signaling pathway Antigen processing and presentation Cytokine cytokine receptor interaction Wnt signaling pathway MAPK signaling pathway Calcium signaling pathway Regulation of actin cytoskeleton Focal adhesion Leukocyte transendothelial migration JAK STAT signaling pathway Tight junction Axon guidance Small cell lung cancer Pathogenic Escherichia coli infection EPEC B-cell receptor signaling pathway Toll-like receptor signaling pathway Pancreatic cancer Long-term potentiation Fc epsilon RI signaling pathway Pyruvate metabolism Pathogenic E. coli infection EHEC Fatty acid metabolism Adipocytokine signaling pathway Glioma Benzoate degradation via CoA ligation Apoptosis Arachidonic acid metabolism Neuroactive ligand receptor interaction Renal cell carcinoma TGF-b signaling pathway PPAR signaling pathway Glycolysis and gluconeogenesis Propanoate metabolism Adherens junction ECM receptor interaction Pentose and glucuronate interconversions

!0.001 !0.001 !0.001

0.00 0.00 0.00

Post Post Post

!0.001 !0.001 !0.001 !0.001 !0.001 !0.001

0.00 0.00 0.01 0.01 0.01 0.02

Post Post Post Post Post Post

!0.001 !0.001 !0.001 !0.001 0.001 0.001 0.001 0.001 0.002 0.002 0.002

0.02 0.05 0.05 0.11 0.06 0.05 0.07 0.06 0.01 0.05 0.04

Post Post Post Post Post Post Post Post Post Post Post

0.002 0.003 0.003 0.005 0.005 0.005 0.006 0.010 0.011 0.012 0.012

0.04 0.06 0.06 0.05 0.05 0.04 0.04 0.06 0.11 0.12 0.05

Post Post Post Post Post Post Post Post Post Post Post

0.013 0.018 0.022

0.11 0.11 0.26

Post Post Post

0.027 0.028 0.031 0.034 0.036 0.037 0.044 0.046

0.16 0.18 0.14 0.16 0.12 0.18 0.18 0.13

Post Post Post Post Post Post Post Post

Enriched in

ECM 5 extra cellular matrix; EHEC 5 enterohemorrhagic E. coli; EPEC 5 enteropathogenic E. coli; JAK 5 Janus activating kinase; MAPK 5 mitogenactivated protein kinase; PPAR 5 peroxisome proliferator-activated receptor; STAT 5 signal transducers and activators of transcription; TGF 5 transforming growth factor; VEGF 5 vascular endothelial growth factor.

Supplementary Table E2. Results of gene set enrichment analysis of MNCs using pathway definitions from Biocarta

Pathway

p Value

False discovery rate

Enriched in

Caspase pathway Chemical pathway TCR pathway GPCR pathway Rho pathway HIVNef pathway BCR pathway Calcineurin pathway P53hypoxia pathway CREB pathway Biopeptides pathway Fas pathway AMI pathway CTLA4 pathway NOS1 pathway Csk pathway Mitochondria pathway Death pathway VIP pathway FceR1 pathway PAR1 pathway NKcells pathway GATA3 pathway TEL pathway Stem pathway Ceramide pathway Erk pathway IL12 pathway ECM pathway PGC1a pathway PDGF pathway PPARa pathway NO2IL12 pathway MEF2D pathway SppA pathway ATM pathway NDK dynamin pathway

!0.001 !0.001 !0.001 !0.001 !0.001 !0.001 0.002 0.002 0.002 0.003 0.003 0.003 0.003 0.003 0.005 0.007 0.007 0.008 0.009 0.010 0.012 0.012 0.020 0.021 0.022 0.023 0.025 0.025 0.025 0.027 0.029 0.030 0.031 0.033 0.038 0.046 0.047

0.01 0.01 0.01 0.02 0.02 0.05 0.02 0.01 0.04 0.03 0.05 0.05 0.05 0.03 0.04 0.04 0.04 0.05 0.06 0.08 0.07 0.05 0.08 0.08 0.07 0.08 0.09 0.08 0.11 0.11 0.14 0.13 0.10 0.13 0.15 0.15 0.18

Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post

BCR 5 B-cell receptor; ECM 5 extra cellular matrix; GPCR 5 G-proteincoupled receptor; IL 5 interleukin; NK 5 natural killer; PDGF 5 plateletderived growth factor; PPARa 5 peroxisome proliferator-activated receptor alpha; TCR 5 T-cell receptor.

K. van der Weide et al./ Experimental Hematology 2012;40:177–186 Supplementary Table E3. Results of gene set enrichment analysis of MNCs using pathway definitions from GenMAPP Pathway Apoptosis GenMAPP Apoptosis Apoptosis KEGG Pyruvate metabolism Propanoate metabolism Calcium regulation in cardiac cells GPCRDB class A rhodopsin like Smooth muscle contraction Peptide GPCRs Mitochondrial fatty acid betaoxidation Cholesterol biosynthesis Glycolysis Glycolysis and gluconeogenesis Gluconeogenesis Valine leucine and isoleucine degradation Integrin-mediated cell adhesion KEGG Lysine degradation Citrate cycle TCA cycle

p Value !0.001 !0.001 !0.001 !0.001 !0.001 !0.001 0.001 0.002 0.003 0.004 0.005 0.011 0.012 0.013 0.019 0.030 0.035 0.049

FDR 0.00 0.00 0.00 0.00 0.01 0.06 0.12 0.09 0.06 0.01 0.03 0.11 0.11 0.10 0.10 0.17 0.15 0.13

186.e5

Supplementary Table E4. Results of leading edge analysis using p ! 0.05 and false discover rate !0.25

Enriched in Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post Post

GPCR 5 G-protein-coupled receptor; KEGG 5 Kyoto Encyclopedia of Genes and Genomes; TCA 5 the citrate cycle.

Gene symbol

Protein

MNCs Kyoto Encyclopedia of Genes and Genomes AKT1 PKB/Akt MAPK3 ERK1 AKT2 RAC-b PRKCA PKCa PRKCB1 PKCb RAC2 Rac2 PPP3CC Calcineurin A-g ITGB1 Integrin b-1; CD29 PIK3R1 PI3K regulatory subunit 1 EGFR EGFR FYN Proto-oncogene tyrosine kinase Fyn ROCK1 ROCK IKBKG IKKg NFKB1 NF-kB Biocarta PRKCB1 PKCb PRKCA PKCa MAPK3 ERK1 CALM3 calmodulin 3 CALM2 calmodulin 2 MAPK8 JNK PPP3CC Calcineurin A-g MAPK1 ERK2 NFATC2 nuclear factor of activated T cells NFATC1 nuclear factor of activated T cells PRKAR1A PKA type I-a PIK3R1 PI3K regulatory subunit 1 GenMAPP ALDH1B1 Aldehyde dehydrogenase ALDH9A1 Aldehyde dehydrogenase ALDH1A1 Aldehyde dehydrogenase ALDH1A2 Aldehyde dehydrogenase ALDH1A3 Aldehyde dehydrogenase LDHC Lactate dehydrogenase LDHB Lactate dehydrogenase EHHADH Enoyl-CoA hydratase ACAT2 Acetyl-CoA acetyltransferase NFKB1 NF-kB PDHB Pyruvate dehydrogenase PKM2 Pyruvate kinase isoenzymes M1/M2 PKLR Pyruvate kinase isozymes R/L

No. of gene sets

15 14 13 13 12 12 10 9 9 8 8 8 8 8 14 14 13 11 11 10 10 8 8 8 8 7 6 6 6 6 6 5 5 4 4 4 4 4 4

186.e6

K. van der Weide et al./ Experimental Hematology 2012;40:177–186

Supplementary Table E5. Cell counts and blood parameters of mice treated with simvastatin and/or welchol

BM % LYMPH % GRAN % MID PB WBC (10E6/mL) % LYMPH % GRAN % MID HCT (L/L) MCV (fl) RBC (10E9/mL) HGB (mmol/L) MCH (fmol) MCHC (mmol/L) RDW (%) MPV (fl) PLT (10E6/mL)

Control

Simvastatin

Welchol

Simvastatin þ welchol

58.3 6 4.5 13.4 6 3.2 28.3 6 2.1

60.3 6 5.0 11.6 6 3.3 28.1 6 2.1

66.7 6 0.8* 9.2 6 0.4 24.1 6 0.5*

63.2 6 1.2y 10.9 6 0.6y 24.5 6 2.8*,z

6.1 86.8 2.9 10.3 0.37 40.9 9.0 9.0 1.0 24.5 14.9 6.8 540

6 6 6 6 6 6 6 6 6 6 6 6 6

2.1 3.5 1.5 2.6 0.03 1.7 0.7 0.7 0.04 0.5 0.6 0.2 116

7.0 81.7 3.2 15.1 0.37 41.0 9.0 9.1 1.0 24.6 14.8 6.9 619

6 6 6 6 6 6 6 6 6 6 6 6 6

2.1 7.2* 1.6 6.2* 0.03 1.7 0.6 0.6 0.04 0.5 0.6 0.2 109

7.6 90.8 1.6 7.6 0.40 39.8 10.0 9.6 1.0 24.1 15.6 7.0 605

6 6 6 6 6 6 6 6 6 6 6 6 6

1.5 0.9 0.6 0.5 0.01 0.1 0.2* 0.2 0.01 0.2 0.3 0.1 45

7.3 85.3 2.7 12.0 0.39 39.4 9.9 9.3 0.9 24.1 15.7 7.2 613

6 6 6 6 6 6 6 6 6 6 6 6 6

1.1 2.5 0.8 1.7 0.00 0.4 0.2 0.0 0.02 0.2 0.1 0.4* 279

BM 5 bone marrow; GRAN 5 granulocytes; HCT 5 hematocrit; HGB 5 hemoglobin; LYMPH 5 lymphocytes; MCH 5 mean corpuscular hemoglobin; MCHC 5 mean corpuscular hemoglobin concentration; MCV 5 mean corpuscular volume; MID 5 other cells including monocytes, eosinophils, basophils, and precursor white cells; MPV 5 mean platelet volume; PB = peripheral blood; PLT 5 platelets; RBC 5 red blood cells; RDW 5 red blood cell distribution width; WBC 5 white blood cells. *p ! 0.05 vs control. y p ! 0.05 vs welchol. z p ! 0.05 vs simvastatin.