Inhibiting the palmitoylation/depalmitoylation cycle selectively reduces the growth of hematopoietic cells expressing oncogenic Nras

Abstract

The palmitoylation/depalmitoylation cycle of posttranslational processing is a potential therapeutic target for selectively inhibiting the growth of hematologic cancers with somatic NRAS mutations. To investigate this question at the single-cell level, we constructed murine stem cell virus vectors and assayed the growth of myeloid progenitors. Whereas cells expressing oncogenic N-Ras(G12D) formed cytokine-independent colonies and were hypersensitive to GM-CSF, mutations within the N-Ras hypervariable region induced N-Ras mislocalization and attenuated aberrant progenitor growth. Exposing transduced hematopoietic cells and bone marrow from Nras and Kras mutant mice to the acyl protein thioesterase inhibitor palmostatin B had similar effects on protein localization and colony growth. Importantly, palmostatin B-mediated inhibition was selective for Nras mutant cells, and we mapped this activity to the hypervariable region. These data support the clinical development of depalmitoylation inhibitors as a novel class of rational therapeutics in hematologic malignancies with NRAS mutations.

Figure 1

Functional analysis of N-Ras^G12D mutant proteins. (A) CFU-GM growth of GFP⁺ fetal liver cells expressing WT N-Ras, N-Ras^G12D, and N-Ras^G12D HVR mutant proteins over a range of GM-CSF concentrations. The data are shown as percentage of maximal growth (left panel) and the absolute number of colonies (right panel) for each construct. The data presented are from 3 independent experiments. Asterisks on the right panel indicate statistically significant differences in colony growth: *P < .05; **P < .005; ***P < .0005. Cytokine-independent CFU-GM growth was only observed in cells expressing N-Ras^{G12D, SSDD} or N-Ras^G12D, and was significantly lower for the SSDD mutant. For statistical analyses, the number of CFU-GM colonies that formed in cells expressing WT N-Ras in the presence of a saturating concentration of GM-CSF (10 ng/mL) was compared with all 3 mutants. Cells expressing N-Ras^{G12D, SSDD} or N-Ras^G12D formed significantly more colonies, whereas cells expressing N-Ras^{G12D, C181S} formed significantly fewer. (B) Confocal imaging of macrophages differentiated from GFP⁺ fetal liver cells. Note that the SSDD mutant protein accumulates in the Golgi and that the C181S mutant is absent from the plasma membrane. The confocal images were acquired on the Zeiss LSM 510 NLO Meta using the Plan-APOCHROMAT 63×/1.4 aperture oil objective. Images were taken on live cells grown on Lab-Tek chambered coverglass w/cvr at 25°C. We used GFP as the fluorochrome, and fluorescent signals were detected using photomultiplier tubes. We used the acquisition software LSM 510 and no further manipulation of the images was performed. (C) Biochemical analysis of cultured GFP⁺ fetal liver cells differentiated into macrophages in vitro. The cells were deprived of serum overnight and stimulated with 10 ng/mL GM-CSF for 20 minutes. The 3 G12D mutant proteins accumulate in the GTP-bound conformation, and both total Ras expression and ERK activation are severely attenuated by the C181S substitution. (D) CFU-GM growth of fetal liver cells expressing WT N-Ras and WT N-Ras with the C181S mutation over a range of GM-CSF concentrations. The data presented are from 3 independent experiments.

Figure 2

Effects of palmostatin B on CFU-GM and blast colony growth. (A-B) CFU-GM were grown from fetal liver cells expressing N-Ras^G12D, K-Ras^G12D, N-Ras^{G12D, KHVR}, and K-Ras^{G12D, NHVR} at 0 and 0.1 ng/mL GM-CSF in the presence palmostatin B. As in Figure 1, the data presented are from 3 independent experiments. Asterisks indicate statistically significant reductions in colony growth compared with untreated cells that were transduced with the same vector and plated in parallel: *P < .05; ***P < .0005. Only cells infected with MSCV vectors encoding proteins containing the N-Ras HVR are sensitive to treatment. (C) Confocal imaging of differentiated macrophages from GFP⁺ fetal liver cells after treatment with 10μM palmostatin B for 15 minutes. Proteins containing the N-Ras HVR show reduced localization at the plasma membrane by palmostatin B treatment. The confocal images were acquired on the Zeiss LSM 510 NLO Meta using the Plan-APOCHROMAT 63×/1.4 aperture oil objective. Images were taken on live cells grown on Lab-Tek chambered coverglass w/cvr at 25°C. We used GFP as the fluorochrome, and fluorescent signals were detected using photomultiplier tubes. We used the acquisition software LSM 510 and no further manipulation of the images was performed. (D) Effects of palmostatin B on cytokine-independent CFU-GM growth from the bone marrows of 3-month-old Mx1-Cre; LSL-Nras^G12D/+, Mx1-Cre; LSL-Kras^G12D/+, Mx1-Cre; LSL-Nras^G12D/G12D mice as well as from 6-month-old Mx1-Cre, LSL-Nras^G12D/G12D mice. One mouse of each genotype was analyzed in 2 independent experiments. Asterisks indicate statistically significant reductions in CFU-GM growth compared with untreated cells of the same genotype: *P < .05; **P < .005; ***P < .0005. (E) Recipient mice that were transplanted with Mx1-Cre, LSL-Nras^G12D and Mx1-Cre, LSL-Kras^G12D AML cells died of aggressive leukemia (leukocyte counts > 100 000/mm³). Blast colony growth was assessed from bone marrow cells plated in 10 ng/mL GM-CSF with or without palmostatin B. The growth of CFU-GM colonies from a WT mouse was compared with blast colony growth from the same Kras mutant AML and from 2 Nras AMLs in 2 independent experiments. Asterisks indicate statistically significant reductions in CFU-GM growth compared with untreated cells from the same mice: *P < .05; **P < .005.