Clonal analysis of NRAS activating mutations in KIT-D816V systemic mastocytosis

Abstract

Cooperating genetic events are likely to contribute to the phenotypic diversity of KIT-D816V systemic mastocytosis. In this study, 44 patients with KIT-D816V systemic mastocytosis were evaluated for coexisting NRAS, KRAS, HRAS or MRAS mutations. Activating NRAS mutations were identified in 2 of 8 patients with advanced disease. NRAS mutations were not found in patients with indolent systemic mastocytosis. To better understand the clonal evolution of mastocytosis, we evaluated the cell compartments impacted by the NRAS and KIT mutations. Clonal mast cells harbored both mutations. KIT-D816V was not detected in bone marrow CD34⁺ progenitors, whereas the NRAS mutation was present. These findings suggest that NRAS mutations may have the potential to precede KIT-D816V in clonal development. Unlike other mature lineages, mast cell survival is dependent on KIT and the presence of these two activating mutations may have a greater impact on the expansion of this cell compartment and in resultant disease severity. (Clinicaltrials.gov identifier: NCT00044122, NCT00001756)

Introduction

Systemic mastocytosis (SM) is a heterogeneous disorder characterized by the pathological accumulation of mast cells within tissues. The majority of adult patients are classified with indolent systemic mastocytosis (ISM) which generally carries a low risk of transformation to an aggressive state and is not thought to affect lifespan. However, a subset of patients with aggressive systemic mastocytosis (ASM) have a poor prognosis and will require cytoreductive therapy. Somatic activating mutations in c-kit, most notably KIT-D816V, are detected in the majority of adult patients.1 Although multilineage involvement by KIT-D816V clearly has an impact on disease severity and progression,2 additional unidentified genetic abnormalities are likely to contribute to more advanced forms of the disease.

RAS proteins are small membrane associated GTPases that play a pivotal role in signal transduction events regulating cell proliferation, differentiation and survival. Somatic mutations which disrupt this intrinsic GTPase activity and lock RAS in an active GTP-bound state are frequent among myeloid malignancies, predominantly involving KRAS and NRAS. In murine models, oncogenic NRAS has not only produced chronic myelogenous leukemia and acute myelogenous leukemia-like diseases, but also increased mast cells in the blood, bone marrow, liver and spleen, a phenotype consistent with aggressive systemic mastocytosis.3–4

In this study, we demonstrate that RAS gene expression increases with mast cell maturation and that activating mutations, specifically in NRAS, are found exclusively in advanced forms of systemic mastocytosis and may precede the KIT-D816V mutation in clonal development.

Design and Methods

Patients

Forty-four patients with systemic mastocytosis were evaluated at the National Institutes of Health (NIH, Bethesda, MD, USA) between 2006 and 2009 as part of an Institutional Review Board-approved research protocol designed to study the pathogenesis and natural history of systemic mastocytosis (NCT00044122). This included 27 patients with indolent systemic mastocytosis (ISM), 9 patients with smoldering systemic mastocytosis (SSM), 4 patients with systemic mastocytosis with an associated clonal hematologic non-mast cell lineage disease (SM-AHNMD) and 4 patients with aggressive systemic mastocytosis (ASM). All patients were diagnosed according to the World Health Organization (WHO) criteria5 and carried the KIT-D816V mutation.

Sample processing

RNA/cDNA was prepared from bone marrow mononuclear cells and cell lines as described.6 Buccal gDNA was isolated using the Gentra Puregene DNA Purification Kit (Qiagen) followed by amplification using a Qiagen REPLI-g Mini kit. HMC1, LAD2 and CD34 derived human mast cells (NCT00001756) were cultured as described.7

Immunophenotypic analysis of mast cells and flow cytometry cell sorting

Bone marrow mast cells were analyzed as described6 using CD45 PerCP, CD117 APC and CD25 FITC (BD Biosciences) antibodies and FACSCanto II flow cytometer (BD Biosciences).

To obtain mast cells, CD34 cells, monocytes, granulocytes, eosinophils, B- and T-cell fractions, a CD45 enriched population (Whole Blood CD45 MicroBeads; Miltenyi Biotec) were stained using CD45 Tri Color, CD3 PE-TR, CD19 PE-TR (Invitrogen), CD14 FITC, CD49d PE, CD34 FITC (BD Biosciences), CD117 PE (Dako), DAPI and sorted using a FACSVantage SE flow cytometer (BD Biosciences). Sort purity routinely exceeded 98%.

Mutational analysis

The KIT-D816V mutation was detected by RT-PCR/RFLP as described.6 Two round PCR followed by RFLP was used for flow sorted cells. NRAS, HRAS, KRAS and MRAS open reading frames were amplified from cDNA either directly or by nested PCR (flow-sorted cells). PCR products were gel purified and directly sequenced in both sense and antisense directions using BigDye terminator v3.1 chemistry and an ABI-3100 genetic analyzer according to standard protocols. Sequencing data were analyzed by Sequencher (Version 4.5, Softgenetics). Primers and conditions used for all PCR reactions are found in the Online Supplementary Table S1.

RAS real-time PCR

Real-time PCR was performed using RT2 SYBR Green qPCR Master Mixes (SABiosciences) and the ABI7500 real-time PCR system (Applied Biosystems). The 2 method was used to calculate the relative expression level of each gene to GAPDH.

Results and Discussion

The phenotypic diversity displayed among myeloproliferative disorders is thought to be the result of multiple and complex molecular events. KIT-D816V SM shares this phenotypic heterogeneity and coexisting mutations are increasingly being identified. The JAK2 V617F mutation was detected in a rare subset of patients with KIT D816V systemic mastocytosis associated with chronic idiopathic myelofibrosis.8 More recently, loss of function mutations in the putative tumor suppressor gene, TET2, were frequently found in patients with systemic mastocytosis although this did not appear to alter prognosis.9 We now report coexisting NRAS activating mutations which potentially collaborate with KIT-D816V in disease pathogenesis.

Table 1.Characteristics of the study population with advanced forms of KIT-D816V systemic mastocytosis. Shading indicates patients with additional NRAS mutations.

Two of 44 patients (4.5%) harbored an NRAS activating mutation. NRAS-G12D and NRAS-G13D mutations were identified in one patient with SM-CMML and one patient with aggressive systemic mastocytosis, respectively (Figure 1A). Bone marrow histology supported these classifications and although a hypercellular marrow was observed in the patient with aggressive systemic mastocytosis, the overall findings did not meet 2008 WHO criteria for any myeloproliferative or myelodysplastic disorder (Figure 2). Together, 25% (2/8) of patients with advanced forms of systemic mastocytosis harbored activating NRAS mutations, although no associated phenotype was observed within this subset (Table 1). These findings parallel observations made in other myeloproliferative disorders such as acute myelogenous leukemia, where RAS mutation frequency does not vary with gender, age, leukocytosis, or WHO performance status.10 Of similar importance is the absence of NRAS mutations in 36 patients with indolent systemic mastocytosis (n=27) or smoldering systemic mastocytosis (n=9). This observation supports the current premise that more benign forms of mastocytosis are mainly KIT-D816V driven and additional mutations may be required for more severe forms of the disease. Indeed, NRAS mutations associated with progression from myelodysplastic syndromes to acute myelogenous leukemia are described.11–12

Figure 1.(A) Activating NRAS mutations in 2 patients with severe forms of KIT-D816V systemic mastocytosis. KIT-D816V RT-PCR/RFLP: Detection of KIT-D816V (arrow) in the bone marrow mononuclear cells (BMMC) of the 2 patients. HMC1.2 cells were used as positive control. NRAS sequencing: Patient 1: Compared to control, a heterozygous missense mutation GGT→GAT at codon 13 was identified resulting in a Glycine to Aspartic Acid substitution (NRAS-G13D). Patient 2: A similar heterozygous missense mutation GGT→GAT at codon 12 was identified resulting in a Glycine to Aspartic Acid substitution (NRAS G12D). A wild-type sequence in the buccal germline DNA confirmed a somatic event. (B) Patient 2: Segregation of KIT-D816V and NRAS-G12D in flow sorted bone marrow cell populations. KIT-D816V is present in mast cells (arrow), but absent in the CD34⁺ cells. NRAS-G12D is present in both populations (asterisks). Results are representative of three separate flow sorting experiments and summarized in Table 1. (C) Mast Cell Immunophenotyping. Bone marrow mast cells were initially identified as CD117 bright positive, CD45 positive cells with characteristic forward and side scatter properties (circled). Compared to the isotype control, gated mast cells uniformly expressed the aberrant CD25 marker. (D) RAS gene expression in cultured human mast cells. HRAS, NRAS, KRAS and MRAS gene expression standardized to GAPDH is plotted on the Y axis. Mast cells on the X axis include CD34⁺ derived human mast cells (HuMC) harvested at weeks one, 2, 4 and 8; HMC1.1, HMC1.2 and LAD2 mast cell lines. Error bars are the Standard Error Mean (SEM) of three separate experiments each performed in triplicate. Different CD34⁺ donors were used for each HuMC experiment. *Note: KIT-D816V RT-PCR/RFLP gel lanes are cropped from the same gel without enhancement.

Figure 2.Histopathological changes in bone marrow biopsies from 2 NRAS positive patients. Both patients have hypercellular marrow biopsies, but with markedly different mast cell burden, as assessed by % mast cell involvement of the biopsy sections. Representative H&E (A) and CD117 (B) immunostained sections show that Patient 1 has minimal involvement by mast cell aggregates, while Patient 2 has extensive marrow replacement by mast cells [(C) H&E; (D) CD117]. In addition, Patient 1 has myeloid hyperplasia with increase in monocytic precursors and blasts, while Patient 2 has normal trilineage maturation. (Magnification 40x for all photographs)

Efforts to molecularly dissect the cell compartments impacted by these two mutations in aggressive systemic mastocytosis (Patient 2) revealed that bone marrow mast cells harbored both mutations and uniformly expressed the aberrant CD25 marker, indicating a clonal population (Figure 1B and C). KIT-D816V and NRAS-G12D were also detected in both myeloid and lymphoid lineages (Figure 1B). This is consistent with previous observations that aggressive forms of mastocytosis display multilineage involvement, likely the result of a common progenitor.2, 13 However, detection of KIT-D816V in CD34 bone marrow progenitors has varied between studies. Akin et al. did not detect KIT-D816V in the CD34 cells of 3 of 3 patients with systemic mastocytosis displaying multilineage involvement.13 In contrast, Garcia-Montero et al. observed that 3 of 4 of patients with aggressive systemic mastocytosis harbored KIT-D816V in CD34 progenitors.2 In our study we detected only NRAS-G12D in the CD34 progenitors, despite the KIT-D816V RT-PCR/RFLP assay having greater sensitivity than cDNA sequencing. This observation suggests that NRAS-G12D may have preceded KIT-D816V in clonal development. According to the clonal expansion model, early mutations should be more prevalent in the clonal population than late mutations.14 Consistent with this model, NRAS-G12D penetrated more cell populations than KIT-D816V (Figure 1B).

Cooperating Class I (enhanced proliferation and/or survival) and Class II (impaired differentiation) mutations are thought necessary for leukemogenesis.15 NRAS and KIT mutations both represent Class I mutations. Although it is likely that an unidentified Class II mutation may exist, this “exception” has been observed in other studies.16–17 A model utilizing information from the Cancer Genome Atlas recently predicted that NRAS activating mutations would coexist with KIT mutations in hematopoietic malignancies and were, therefore, strong candidates for cosequencing.14 Mutations in other RAS genes were not predicted to coexist with KIT mutations. In support of this prediction, we did not detect KRAS, HRAS or MRAS mutations in our KIT-D816V population. Arguably, this may be a reflection of the sampling size or alternatively NRAS may have a significant role in mast cell homeostasis.

The RAS gene family appears to play a role in mast cell development, as their relative expression levels uniformly increased as mast cells matured in vitro (Figure 1D). Peak RAS expression was observed at eight weeks, at which time mature mast cells were the only cell type present. NRAS and KRAS, but not HRAS are reported to be the dominant isoforms in LAD2 and HMC1 cell lines.18 We observed that the message for all isoforms was detectable, including MRAS, albeit expression was relatively low and comparable to that of an immature mast cell. This may reflect the maturity state of the cell lines and/or the cell division rate. Cell lines and cultured CD34 cells at 1–2 weeks are rapidly dividing and may not require significant RAS expression, whereas the relatively quiescent mature mast cell may be more dependent. RAS mutations were not observed in the cell lines.

Clinical trials targeting KIT-D816V have demonstrated only modest efficacy in systemic mastocytosis.19–20 This study contributes to growing evidence that additional genetic alterations are present in KIT-D816V SM and effective treatment will likely require a multi-targeted approach.21–22 Specifically identifying RAS mutations may influence the choice and dosing of cytoreductive therapy, as acute myelogenous leukemia patients carrying activated forms of RAS appear to benefit from higher cytosine arabinoside doses in response rate and overall survival.23–24 As advanced systemic mastocytosis is relatively rare, large multicenter studies will be required to support not only our findings, but future studies investigating the molecular pathogenesis of systemic mastocytosis.

Acknowledgments

We wish to thank the patients for their participation, as well as Alasdair Gilfillan and Kimberly Dyer and the clinical research staff for their scientific and medical assistance.

Footnotes

• Funding: this study was funded by the Intramural Research Program of the NIAID, NIH.
• The online version of this article has a Supplementary Appendix.
• Authorship and Disclosures The information provided by the authors about contributions from persons listed as authors and in acknowledgments is available with the full text of this paper at www.haematologica.org.
• Financial and other disclosures provided by the authors using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are also available at www.haematologica.org.

• Received August 9, 2010.
• Revision received November 3, 2010.
• Accepted November 30, 2010.