In Vivo Treatment of AML Patients with High-Dose Simvastatin Inhibits Geranylgeranylation In AML Cells

Abstract 3280

Statins, like simvastatin, inhibit 3-hydroxy-3-methylglutaryl CoA reductase (HMG-CoAR), the rate-limiting enzyme of the mevalonate pathway. This leads to inhibition of cholesterol synthesis and to decreased expression of the cholesterol efflux transporters ABCA1 and ABCG1 in various cell types. In addition, the production of the isoprenoids farnesyl pyrophosphate (FPP) and geranylgeranyl pyrophosphate (GGPP) is suppressed by statins. Isoprenylation is required for the binding to the plasma membrane of small GTP-ases, like Ras and Rho, and their subsequent participation in signal transduction. The cytotoxic effects of statins in acute myeloid leukemia (AML) have been attributed to their cholesterol-lowering properties. Yet, recent in vitro experiments have demonstrated a key role for isoprenylation in this respect. However, direct effects of in vivo applied statins on AML cells have not been demonstrated. The aim of this study was to verify whether effects of simvastatin on AML cells in vitro also translate to the in vivo situation. AML patients (n=12) were treated for 7 days with high dose simvastatin (7.5-15 mg/kg/day) before initiating chemotherapy. Before and at the end of simvastatin-treatment serum lipid levels were determined and AML mononuclear cells (MNCs) from bone marrow or peripheral blood were collected. Despite a decrease in serum cholesterol (from 4.0 to 2.3 mmol/L, p=0.002), lathosterol (indicator of cholesterol synthesis, from 3.4 to 0.6 μmol/L, p=0.02), and low density lipoprotein (LDL, from 2.5 to 1.1 mmol/L, p=0.004) after simvastatin-treatment, the expected changes in mRNA expression of cholesterol metabolism genes (HMG-CoAR, LDLR, ABCA1 and ABCG1) were not observed in AML MNCs. Gene set enrichment analysis on paired samples of AML patient MNCs before and at the end of simvastatin treatment revealed that mainly pathways involved in cell signaling (e.g., MAPK, GPCR, and RHO pathways) and immune responses were affected by simvastatin, suggesting that indeed inhibition of isoprenylation could play a role. These data prompted us to employ a mouse model that allowed comparison of the effects in bone marrow MNCs versus liver cells, the major target organ of statins. Mice were treated for 7 days with a dose equivalent to 150 mg/kg/day, after which bone marrow MNCs and liver cells were analyzed. Again, no changes in HMG-CoAR and LDLR mRNA expression were found in bone marrow MNCs upon simvastatin treatment, while ABCA1 and ABCG1 were decreased 1.6-fold (p=0.004) and 2.1-fold (p=0.005), respectively. In contrast, in liver cells HMG-CoAR (12-fold, p<0.001) and LDLR (2-fold, p=0.02) were upregulated, whereas ABCA1 and ABCG1 mRNA expression remained unchanged. Simvastatin treatment reduced geranylgeranylation of Rap1, a protein that is exclusively geranylgeranylated, in bone marrow MNCs (1.6-fold, p=0.004) but not in liver cells. No differences in the degree of farnesylation of DnaJ (exclusively farnesylated) before and after statin treatment were observed in either bone marrow MNCs or liver cells. In vitro experiments on primary mouse and human bone marrow MNCs indicated that inhibition of geranylgeranylation occurs at a much lower, physiologically achievable concentration (<1 μM) of simvastatin, than inhibition of farnesylation or changes in cholesterol metabolism gene expression (>25 μM). In conclusion, we demonstrated an inhibition of especially geranylgeranylation in bone marrow MNCs, not liver cells, upon in vivo treatment with a high dose of simvastatin, which may be involved in the cytotoxic effects on AML cells.

Supported by a grant of the Dutch Cancer Society.