SPRED1, a RAS MAPK pathway inhibitor that causes Legius syndrome, is a tumour suppressor downregulated in paediatric acute myeloblastic leukaemia

Abstract

Constitutional dominant loss-of-function mutations in the SPRED1 gene cause a rare phenotype referred as neurofibromatosis type 1 (NF1)-like syndrome or Legius syndrome, consisted of multiple café-au-lait macules, axillary freckling, learning disabilities and macrocephaly. SPRED1 is a negative regulator of the RAS MAPK pathway and can interact with neurofibromin, the NF1 gene product. Individuals with NF1 have a higher risk of haematological malignancies. SPRED1 is highly expressed in haematopoietic cells and negatively regulates haematopoiesis. SPRED1 seemed to be a good candidate for leukaemia predisposition or transformation. We performed SPRED1 mutation screening and expression status in 230 paediatric lymphoblastic and acute myeloblastic leukaemias (AMLs). We found a loss-of-function frameshift SPRED1 mutation in a patient with Legius syndrome. In this patient, the leukaemia blasts karyotype showed a SPRED1 loss of heterozygosity, confirming SPRED1 as a tumour suppressor. Our observation confirmed that acute leukaemias are rare complications of the Legius syndrome. Moreover, SPRED1 was significantly decreased at RNA and protein levels in the majority of AMLs at diagnosis compared with normal or paired complete remission bone marrows. SPRED1 decreased expression correlated with genetic features of AML. Our study reveals a new mechanism which contributes to deregulate RAS MAPK pathway in the vast majority of paediatric AMLs.

Introduction

In 2007, germline dominant loss-of-function mutations in the SPRED1 (sprouty-related, EVH1 domain containing 11; NM_152594) gene were identified in patients fulfilling the NIH clinical diagnostic criteria for neurofibromatosis type 1 (NF1) with no NF1 mutation found.1, 2, 3, 4 Their phenotype referred as NF1-like syndrome or Legius syndrome (OMIM 611431), consisted of multiple café-au-lait macules, axillary freckling, mild neurocognitive impairment and macrocephaly but without features common in NF1 such as neurofibromas and iris Lisch nodules. Current knowledge of natural history of Legius syndrome is based on clinical manifestations of fewer than 150 individuals with a molecularly confirmed diagnosis.1, 2, 3, 4

NF1 and Legius syndrome belong to the neuro-cardio-facio-cutaneous syndromes that are caused by deregulating constitutional mutations of the RAS MAPKs (mitogen-activated protein kinases) signalling pathway.5 These RASopathies that include Noonan syndrome, LEOPARD syndrome, cardio-facio-cutaneous syndrome, Costello syndrome, NF1 and the Legius syndrome, share characteristic overlapping features, including predisposition to develop multiple types of cancer. Individuals with NF1 and Noonan syndrome have a higher risk of haematological malignancies, including acute leukaemia (AL) and the rare disorder juvenile myelomonocytic leukaemia.

We previously reported the observation of an 11-month-old boy with a SPRED1 constitutional mutation, who developed a paediatric acute myeloblastic leukaemia (AML).6 Although rare, inherited predispositions to myeloid leukaemia have uncovered a critical role of hyperactive RAS MAPKs signalling in normal myeloid growth and leukemogenesis.7 SPRED1 and SPRED2 are members of the evolutionarily conserved SPROUTY/SPRED family of membrane-associated negative regulators of growth factor-induced ERK (extracellular-regulated kinase) activation.8 In mice, SPRED1 inhibits ERK activation by suppressing RAF1 kinase phosphorylation. SPRED1 is highly expressed in haematopoietic cells and negatively regulates haematopoiesis by suppressing stem cell factor and interleukin-3 (IL-3)-induced ERK activation.9 Bone marrow (BM)-derived mast cells and eosinophils from Spred1–/– mice were more sensitive to IL-3 and IL-5, respectively, than those from wild-type mice.9, 10 Moreover, recent evidences indicate that SPRED1 regulates neurofibromin RAS-GTPase activity by interacting with the NF1 gene product.11 With regards to these observations, SPRED1 seemed to be a good candidate gene for leukaemia predisposition or transformation.

In this study, we performed SPRED1 mutation screening and expression status in a large series of 230 paediatric acute lymphoblastic and myeloblastic leukaemias. We found four new SPRED1 variations that were constitutional events. Among them, one patient had Legius syndrome clinical features, which were not diagnosed at the time of leukaemia onset. In this patient, the leukaemia blasts karyotype showed a loss of heterozygosity (LOH) of the SPRED1 locus, confirming SPRED1 as a tumour suppressor. Our observation confirmed that AL are complication of the Legius syndrome stressing the risk of developing leukaemia in constitutional carriers of SPRED1 mutations. Moreover, SPRED1 was significantly decreased at RNA and protein levels in the majority of AML at diagnosis compared with normal BM or paired complete remission (CR) BM samples. Interestingly, SPRED1 decreased expression correlated with genetic features of AML. Our study reveals a new mechanism which contributes to deregulate RAS-activating MAPK pathway in the vast majority of paediatric myeloblastic leukaemias.

Results

SPRED1 somatic mutation is not a frequent event in paediatric AL

Two hundred and thirty paediatric AL DNAs (83 AML, 110 B-ALL (B-cell acute lymphoblastic leukaemia) and 37 T-ALL (T-cell acute lymphoblastic leukaemia)) were screened for mutation in the SPRED1 gene. Four SPRED1 nucleotide variations (2%) were identified in three B-ALL and one AML at the time of diagnosis (Table 1): one isosemantic transition c.674C>T (B-ALL/2), two missense variations: c.124G>A leading to p.Val42Ile (AML/3; AML) and c.1089A>G leading to p.Ile363Met (B-ALL/1), and one frameshift mutation c.395dupA leading to p.Asn133fs (B-ALL/3). The isosemantic c.674C>T and the two missense variations (Val42Ile and p.Ile363Met) were shown to have no deleterious effect on RNA splicing as SPRED1 cDNA (complementary DNA) sequencing showed no splicing aberration, ruling out a putative RNA decay or splice modification. The two missense variations affected two SPRED1 functional domains: N-terminus Ena/vasodilator-stimulated phosphoprotein homology (EVH1), and C-terminus SPROUTY-like (SPR) domains, respectively. EVH1 domain binds proline-rich sequences, and is necessary for interacting with neurofibromin, the NF1 gene product. The SPR domain is important for plasma membrane localization11 and for the interaction with RAF proteins.8 Most previously described SPRED1 mutations affected these two evolutionary conserved domains.4 Moreover, the two missense mutations both affect two highly conserved amino acids (Supplementary Figure 1).

Table 1 Clinical characteristics of the four patients with SPRED1 constitutional variants

The four heterozygous nucleotide variations were also found at the time of remission. This finding demonstrated that these variations were constitutional rather than somatic. Recently the p.Val42Ile missense variation was described in two unrelated individuals affected by Legius syndrome.4 However, whether the two constitutional missense variations are disease causing mutations, or only rare neutral variants, remain to be established and their precise consequences should be determined at the functional protein level. Patient B-ALL/3 with the frameshift SPRED1 loss-of-function mutation (c.395dupA; p.Asn133fs), showed typical Legius syndrome clinical features: café-au-lait spots and axillary freckling. Clinical information of the four patients with SPRED1 constitutional variations is summarized in Table 1.

SPRED1 somatic LOH was found in leukaemia blasts in one patient with a constitutional SPRED1 mutation

At the time of diagnosis, a somatic mosaicism of a heterozygous deletion on chromosome 15q -including the SPRED1 locus- was found in medullar leukaemia blasts of patient B-ALL/3 with the frameshift SPRED1 loss-of-function mutation (Table 1). At the time of remission, a normal karyotype was observed. The three other carriers of SPRED1 variations (patients B-ALL/2, AML/3, and B-ALL/1) had no typical cutaneous symptoms of Legius syndrome (including café-au-lait macules, and axillary freckling) and no NF1 characteristic complications. No SPRED1 LOH was found in these three patients’ leukaemia blasts at the time of diagnosis.

SPRED1 polymorphisms are not associated with inherited susceptibility to paediatric AL

Previous reports of genome-wide association studies have identified constitutional single nucleotide polymorphisms (SNPs) at different loci that were significantly enriched in leukaemia cases compared with controls, with relevance to the molecular pathogenesis of leukaemia.12, 13, 14 In order to assess the potential role of SPRED1 as a susceptibility gene for paediatric leukaemia, we compared the distribution of two known SPRED1 synonymous exonic SNPs between the AL cohort herein studied and the HapMap samples with the same European origin. Comparison of observed and expected (HapMap samples) distributions for these two SPRED1 SNP genotypes showed no evidence of association (P>0.05) between germline genomic SPRED1 variants and childhood AL as a whole or as three distinct subgroups (T-ALL, B-ALL and AML) making unlike that SPRED1 inherited common alleles predispose to leukaemia. The frequencies in individuals with European ethnic origin in the study cohort are summarized in Supplementary Table 1.

RAS MAPK pathway genes alterations were identified in 50% of the AML samples, including high frequency of NF1 mutations

Figure 1 summarizes the RAS MAPK pathway genes alterations identified in the 73 paediatric AML samples screened for SPRED1, NF1, NRAS, FLT3-ITD, c-KIT, BRAF and PTPN11 mutations. Interestingly, NF1 loss-of-function mutations were found in 5/73 (6.8%) of AML samples: AML/7: c.4718A>G (Tyr1573Cys), AML/31: c.2446C>T, (Arg816*), AML/36: c.8041_8042insA (frameshift), AML/54: IVS731-1G>C (splicing) and AML/67: IVS1642-5T>A (splicing). In samples AML/36 and AML/54, the loss-of-function point mutations (c.8041_8042insA, and IVS731-1G>C, respectively) were found to be associated with a complete NF1 wild-type allele deletion. Overall, a mutation in NF1, NRAS, SPRED1 or c-KIT was found in 37/73 samples (50.7%). NRAS, FLT3-ITD and c-KIT mutations were found in 10/73 (13.7%), 11/73 (15.1%) and 9/73 AML samples (12.3%), respectively. Mutations in NF1, NRAS and SPRED1 were found to be mutually exclusive (Figure 1).

Figure 1
figure1

Distribution of different RAS MAPKs pathway activating mutations in the AML panel (N=73), according to FAB (French–American–British) and cytogenetic risk groups (Medical Research Council, MRC criteria). In samples AML/36 and AML/54, the NF1 loss-of-function point mutations were found to be associated with a complete NF1 wild-type allele deletion.

SPRED1 expression is significantly decreased in paediatric AML at diagnosis compared with normal BM

We determined the median values and range of SPRED1 mRNA levels at the time of diagnosis in 120 leukaemia samples (58 AML, 40 B-ALL and 22 T-ALL) and normal controls (Figure 2). SPRED1 was expressed in normal haematopoietic cells and, compared with total BM, levels were higher in the CD34+ enriched fraction and lower in peripheral blood mononuclear cells or thymocytes. We could quantify the expression of SPRED1 in all the leukaemic samples although values varied greatly. SPRED1 median normalized expression value was 1.2 (0.003–4.33) in B-ALL, 0.26 (0.01–2.14) in T-ALL and 0.09 (0.001–5.46) in AML. Thus, SPRED1 appeared to be significantly decreased in AML (P=0.006) and in T-ALL (P=0.056), when compared with normal BM (Figure 2a). We also assessed the expression levels of the SPRED2, and NF1 genes in the 58 AML samples. Three samples were removed from the analysis because of unqualified reference genes CT. No significant SPRED2, and NF1 downregulation at the RNA level was found in the AML samples (Supplementary Figure 2). We further evaluated SPRED1 expression in the 47 paired diagnosis and CR AML samples, and three paired B-ALL which had a very low expression at diagnosis. In all cases, SPRED1 expression was normal or higher than normal in the remission BM samples (Figure 2b), confirming the downregulation was specific of leukaemic cells. We assessed the expression levels of miR-126 in 55 AML samples and in normal haematopoietic cells (Supplementary Figure 3). AML samples that strongly expressed miR-126 showed no SPRED1 expression. However, no miR-126 upregulation was found in a significant proportion of samples with a SPRED1 downregulation.

Figure 2
figure2

SPRED1 mRNA expression in leukaemia and normal cells at diagnosis and remission time. (a) SPRED1 expression was assessed in 120 leukaemia samples (58 AML, 40 B-ALL and 22 T-ALL) and normal controls. SPRED1 was expressed in normal BM cells. When compared with the total BM, SPRED1 expression levels were higher in the CD34+ enriched BM fraction, and lower in peripheral blood mononuclear cells. When compared with normal BM, SPRED1 was significantly decreased in AML (**P=0.006) and, at a lesser extent in T-ALL (P=0.056). MRNA levels were normalized such that the median value of normal BM was 1. (b) SPRED1 expression was assessed in 47 paired diagnostic and CR AML samples. SPRED1 expression increased to normal or higher than normal level control in AML CR samples (***P=0.0001).

SPRED1 expression is downregulated at protein level and correlates with high phosphoERK1/2 levels

We evaluated SPRED1 expression at the protein level. Using immunofluorescence microscopy, we showed that SPRED1 protein was expressed in human haematopoietic cells, localized in the cytoplasm of precursor CD34 positive cells and BM cells (Figure 3). The amount of protein observed in pathological and normal cells was coherent with mRNA expression data (Figure 3). Western blot analysis confirmed a decrease of functional SPRED1 protein in AML samples (Supplementary Figure 4). Because SPRED1 is considered to act as a negative regulator of the RAS MAPKs signalling, we checked for phosphoERK1/2 (pERK1/2) levels in primary AML in patients with low SPRED1 protein and B-ALL/9 patient sample in which SPRED1 transcript was bearable detected. All tested patients (n=11) had high pERK1/2 levels at diagnosis (Figure 4) and in 7 out of 11 patients pERK1/2 levels decreased (30–90% of reduction compared with the diagnostic samples) in remission samples, concomitantly with SPRED1 recovery.

Figure 3
figure3

SPRED1 immunofluorescent microscopy. SPRED1 is expressed in the cytoplasm of (a) hepatocyte cell line HEPG3 is a positive control; (b) BM precursor CD34 positive cells; (c and e) two AML patients at diagnosis and (d and f) once reached the CR. The amount of protein observed in pathological and normal cells was coherent with mRNA expression: SPRED1 relative expression values (RQ) at diagnostic (DG) and complete remission (CR) are indicated.

Figure 4
figure4

ERK1/2 activation in AML samples with dowregulated SPRED1 expression. Relative pERK/ERK and SPRED mRNA levels are indicated at diagnosis (DG) and at the time of cytological remission (CR). SPRED1 values are calculated relative to BM calibrator (value=1). All tested patients (N=11) had high phosphoERK1/2 at diagnosis.

Correlation of SPRED1 expression levels with NF1, NRAS, FLT3-ITD and c-KIT mutations in AML patients

We checked for an association between SPRED1 expression levels and the mutations found in the RAS MAPKs pathway. AML samples were dichotomized according to SPRED1 median expression (0.09) into low level (N=29; (0.001–0.09)) or high level (N=29; (0.1–5.46)). The lowest levels of SPRED1 expression were more often associated with FLT3-ITD mutations (8/10 cases), whereas relatively higher or near to the normal SPRED1 levels were more often associated with NRAS mutations (8/9 cases) and NF1 mutations (3/4 cases). Among the seven tested c-KIT mutated samples, three samples showed a weak SPRED1 expression, and four samples showed a high SPRED1 expression (Table 2). Samples with no mutations identified were equally distributed between high and low SPRED1 expression levels. By comparing the median SPRED1 expression values in NF1-, NRAS-, FLT3-ITD- or c-KIT-mutated AML BM samples at diagnosis, we showed that SPRED1 expression was significantly lower in FLT3-ITD-mutated AML (P=0.01) (Figure 5).

Table 2 Correlation between SPRED1 expression and known activating RAS MAPKs pathway mutations in the AML panel
Figure 5
figure5

SPRED1 normalized expression (mean) in NF1 (N=4), NRAS (N=9), FLT3-ITD (N=10) or c-KIT (N=7)-mutated AML BM cells at diagnosis. SPRED1 was significantly decreased in FLT3-ITD-mutated versus NF1-, NRAS- and c-KIT-mutated AML BM cells (P=0.01).

Discussion

In this study, we assessed the potential role of SPRED1, an inhibitor of the RAS MAPKs pathway, in leukemogenesis. The RAS MAPKs pathway has been estimated to be mutated in 30% of all cancers, with a peculiar role in haematopoietic malignancies.7 NRAS or KRAS somatic mutations occur in 20% of AML, and RAS MAPKs signalling is upregulated by FLT3-ITD mutations and C-KIT mutations in 25–40% of cases.15, 16 SPRED1 downregulates the RAS MAPKs signalling pathway through an interaction with the NF1 protein, neurofibromin.8, 11, 17

We report the first comprehensive molecular screening of SPRED1 in a large cohort of 230 paediatric patients with de novo AL. Comparison of the distribution of two common SPRED1 SNP genotypes showed that SPRED1-inherited common alleles did not predispose to leukaemia. Using SPRED1 sequencing, we reported four constitutional SPRED1 heterozygous nucleotide variations. Among them, one loss-of-function frameshift SPRED1 mutation was found in a girl who had multiple café-au-lait spots since the age of 2 years, confirming the diagnosis of Legius syndrome. At the time of diagnosis, a somatic interstitial deletion on chromosome 15q including the SPRED1 locus was found in medullar leukaemia blasts (Table 1). This is the first case where a LOH of SPRED1 is shown in leukaemia, supporting the tumour suppressor role of SPRED1 in the development of the B-ALL found in this young patient with Legius syndrome. Our study confirms that AL should be regarded as a complication of the rare Legius syndrome. Constitutional SPRED1 mutation should be tested in case of an association of multiple café-au-lait spots and leukaemia, as well as biallelic mutations in mismatch repair genes, or heterozygous NF1 mutations.

SPRED1, SPRED2 and NF1 expression was then studied in leukaemia RNA samples. Interestingly, SPRED1 transcript levels were significantly decreased in AML samples at the time of diagnosis compared with normal BMs (P=0.006; Figure 2a; Supplementary Figure 2). No significant SPRED2, and NF1 downregulation was found in the AML samples, emphasizing the specificity of SPRED1 downregulation (Supplementary Figure 2). SPRED1 lower expression in AML was confirmed at the protein level (Figure 3; Supplementary Figure 4). SPRED1 downregulated expression correlated with RAS MAPKs pathway activation visualized by increased phosphoERK1/2 levels (Figure 4), suggesting that SPRED1 dosage effects may be a pathogenetically relevant event in AML. Analysis of polymorphic microsatellites located at SPRED1 locus showed that SPRED1 dowregulation was not caused by LOH, suggesting other mechanisms implying a transcriptional or post-transcriptional regulation. Overexpression of miR-126 in haematopoietic cell lines was previously shown to increase proliferation rate and ERK activation in a SPRED1-dependent manner.18 We therefore assessed the expression levels of miR-126 in 55 AML samples (Supplementary Figure 3). Our study suggests that miR-126 overexpression may cause SPRED1 transcript downregulation in a subset of the AML samples.

Here, we demonstrate that reduction of SPRED1 expression which is correlated to the decrease of the protein, likely contributes to the transformation process through the activation of RAS MAPKs pathway in AML. Implication of SPRED1 downregulation in tumorigenesis was demonstrated in previous studies. Overexpression of Spred1 can efficiently suppress tumorigenesis in metastatic osteosarcoma cell lines injected in nude mice.19 SPRED1 expression was reported to be reduced in human hepatocellular carcinoma tissue and its level was inversely correlated with the capacity of hepatocellular carcinoma to invasion and metastasis.20 SPRED1 has therefore to be considered as a tumour suppressor gene, as is NF1. Recently Stowe et al. showed that SPRED1 is necessary for NF1 gene product inhibitory activity. In this work, they demonstrated that the N-terminal EVH1 domain of SPRED1 interacts with neurofibromin and mediates its translocation to the plasma membrane where neurofibromin subsequently downregulates RAS-GTP levels.

SPRED1 was originally identified as c-KIT interacting protein via its central domain. C-KIT has an essential role in haematopoietic stem cell and progenitors proliferation. It is thus tempting to speculate that SPRED1 takes part in dampening c-KIT-mediated RAS activation via its interaction with neurofibromin. Taken together these observations may also explain that ALs were observed in young children with Legius syndrome, rather than other tumour types.

We also screened the AML samples for NF1 mutations using next generation sequencing. A NF1 mutation was found in 7% of AMLs (Figure 1). Recently, pan cancer studies confirmed the role of NF1 as a major tumour suppressor in leukaemias (http://www.nature.com/tcga/).21 Molecular characterization of additional genetic alterations in AML pointed a significant association between SPRED1 low expression rates and FLT3-ITD mutation (Figure 5). This observation reinforces the functional implication of SPRED1 expression decrease in AML pathological process. FLT3 belongs to the class III receptor tyrosine kinase family.22 Somatic gain of function FLT3-ITD mutations have powerful transforming potential in myeloid cells mediated by RAS activation.23 FLT3-ITD and RAS mutations have been shown to represent cooperating events in leukemogenesis. Previous studies in mice model have shown that Nf1 deficiency can cooperate with oncogenic Kras to induce AML in mice.24 Moreover, next-generation sequencing of a primary lung tumour found that at least eight genes in the RAS MAPK pathway were either mutated or amplified in this tumour.25 SPRED1 downregulation and FLT3-ITD mutations might then cooperate to activate RAS MAPK pathway during leukemogenesis. Full confirmation of the herein suggested critical cooperation between SPRED1 downregulation and FLT3-ITD mutation needs further functional studies. Given the importance of cooperating gene mutations in signalling pathways in the generation of the leukaemia phenotype, it might be helpful to determine how specific leukaemia-associated genetic alterations affect the behavior of signalling networks both alone and in combination.16, 26

Our observations showed that SPRED1 is a tumour suppressor of the RAS MAPK pathway, altered in AML pathologic process. We observed for the first time a LOH in a Legius syndrome-associated leukaemia. Moreover, we showed SPRED1 downregulation in AML, which was associated with abnormal RAS MAPKs pathway activation. We concluded that SPRED1 functions as a tumour suppressor in human AML. Elucidation of the role of cellular signalling pathways in leukemogenesis will enable to establish molecular targeted therapeutics. Hopefully, these insights will contribute to the development of therapeutic strategies capable to exploit inherent differences between normal and malignant signalling pathways and thereby improve the therapeutic index of current cancer treatments.27

Materials and Methods

Patient samples

BM or blood samples from 230 children (age: 0–16 years) with AL (83 AML, 110 B-ALL, and 37 T-ALL) were collected at diagnosis, separated by ficoll-hypaque gradient, and mononuclear cells frozen in liquid nitrogen until used in experiments. Blasts percentage varied from 60 to 100% (median 85%) among AML samples, and from 85 to 99% (median 93%) among ALL samples. DNA was extracted from each sample according to standard procedures for SPRED1 mutations screening. High-quality RNA was obtained at diagnosis from 120 samples (58 AML, 40 B-ALL and 22 T-ALL) and from 50 paired BM samples, at the time of the CR, (47 AML, and three B-ALL). The physiological SPRED1 expression in non pathological haematopoietic tissues was further evaluated at both RNA and protein levels in ten BM, eight peripheral blood mononuclear cells, one thymus and one CD34-positive enriched BM samples obtained from healthy volunteers donors. Patients were treated between 2002 and 2010 in Trousseau Hospital, AP-HP, Paris; one case was referred for molecular investigation of clinical features evoking Legius syndrome. Diagnosis was established according to World Health Organization criteria. Written institutional review-board-approved informed consent was obtained from parents or legal tutors, according to the declaration of Helsinki.

Isolation and quality control of RNA and cDNA preparation

Total RNAs were extracted with RNAble reagent (Eurobio, Les Ulis, France) according to the manufacturer’s instruction. DNA was extracted using standard methods. RNA quantity and quality were assessed by Nanodrop ND-1000 spectrophotometer, and by electrophoresis of 200 ng of each RNA on a denaturing agarose gel. Only samples with coherent amount of RNA measured with both techniques and a 28S/18S RNA ratio between 1.8 and 2 were subsequently analysed. cDNA was then reverse transcribed from 1.5 μg of total RNA in a final volume of 20 μl by using the high-capacity cDNA reverse transcription kit (Applied Biosystems, Les Ulis, France) following the manufacturer’s instructions.

SPRED1 mutation screening

We performed SPRED1 mutation screening at genomic DNA and RNA levels as previously described.2 Sequences were aligned with Seqscape analysis software (Applied Biosystems). The primer oligonucleotide sequences and PCR conditions are available on demand. Genomic DNA was amplified with primers specific for SPRED1 coding exons and their IVS boundaries. PCR was performed with the Taqman PCR Core Reagent Kit (Applied Biosystems). Mutation screening was performed using bidirectional DNA sequencing of purified PCR products. SPRED1 nucleotide variations found at DNA level were checked on RNA through the following procedure. After reverse transcription, two overlapping PCR fragments were generated by Taqman PCR Core Reagent Kit (Applied Biosystems). Mutation screening was performed with the ABI BigDye terminator sequencing kit (Applied Biosystems) on an ABI Prism 3130 automatic.

RAS MAPKs pathway activating mutations

NRAS, KRAS (codons 12, 13, and 61), FLT3 (ITD and D835V), BRAF (V600) and PTPN11 (exons 3 and 13) hotspot mutations were further searched in 73/83 of AML diagnostic samples, as previously described.28, 29, 30 NF1 sequencing was performed using a next generation sequencing approach specifically targeting the NF1 gene (Eric Pasmant, manuscript in preparation) in 73/83 of AML samples. Briefly, the custom primers panel targeting the NF1 gene (coding exons and IVS boundaries) was designed using the AmpliSeq Designer. The targeted region was amplified by 197 amplicons (125–225 bp length); 20 ng of genomic DNA were amplified to generate the library using the Ion AmpliSeq Library Kit 2.0 (Life technologies, Saint-Aubin, France). Experiments were performed in the next generation sequencing platform of the Cochin hospital, Paris (Assistance Publique—Hôpitaux de Paris, France). Next generation sequencing libraries preparation was performed using the Ion AmpliSeq Library Kit 2.0 (Life technologies), according to the manufacturer’s instructions (Ion AmpliSeq Library Preparation, Publication Part Number MAN0006735, Revision 5.0, July 2013, Life technologies). The template-positive ion sphere particles were loaded on Ion 316 chips and sequenced with an Ion Personal Genome Machine System (Life technologies). Sequence alignment and extraction of SNPs and short insertions/deletions were performed using the Variant Caller plugin on the Ion Torrent Browser and DNA sequences visualized using the Integrated Genomics Viewer (IGV, version 2.2) from Broad Institute (Cambridge, MA, USA). The NextGENe software (Softgenetics, State College, PA, USA) was also used for sequence alignment, extraction of SNPs and short indels and their visualization.

SPRED1 SNPs genotyping

Genotype of two SPRED1 SNPs: rs7182445 (NM_152594.2: c.291G>A) and rs3751526 (NM_152594.2: c.1044T>C) was assessed by sequencing.

SPRED1 microsatellite typing

A series of three 15q14-linked microsatellite markers were characterized in AML with SPRED1 dowregulation. Two intragenic (IVS1-AT19 and IVS2-TG19) and one extragenic (AC23 at 5′ extremity) SPRED1 polymorphic microsatellites were studied. The primer oligonucleotide sequences are available on demand. After amplification with a 5′-end-labelled primer, followed by the addition of an internal size standard (Genescan 2500-ROX: Applied Biosystems), the PCR products were separated on 6% polyacrylamide/7 μM urea gel, using an ABI PRISM 310 DNA sequencer (Applied Biosystems). The results were analysed with the Genescan 672 (version 1.2) software package.

Real-time RT-PCR for protein-coding genes (SPRED1, SPRED2 and NF1)

The theoretical and practical aspects of real-time quantitative RT-PCR have been described in detail elsewhere.31 Quantitative reverse transcription RT-PCR was performed using specific primers for SPRED1 (Hs00544790_m1 expression assays, Applied Biosystems), SPRED2 (Hs_SPRED2_1_SG QuantiTect primer assay, Qiagen, Courtaboeuf, France) and NF1 (upper primer, 5′-ACA GAG CGT GGC CTA CTT AGC A-3′, and lower primer, 5′-GGA CCA TGG CTG AGT CTC CTT T-3′). One hundred nanogram of RT-equivalent RNA from each sample were amplified according to standard procedure in a TaqMan 7900 HT devise (Applied Biosystems). The ABL1 gene was used as ubiquitous control gene and quantified using the EAC (Europe Against Cancer program) Taqman set.31 Samples were considered eligible for testing only when the Ct of the internal reference ABL1 was lower than 28. We calculated the expression values with the comparative CT method, which uses the formula 2−ΔΔCT to calculate the expression of a target gene normalized to a calibrator. The calibrator was determined as follows: first we calculated the ΔCT values (ΔCT=CT (target gene) –CT (internal reference, ABL1)) for each normal BM (n=10), and then for the pooled normal BM samples. The value of the pool of BM (ΔCt=2.29) was almost identical to the mean value of normal BM values (ΔCt=2.30; (0.12–2.78)). Then, we gave a relative value of 1 to the target gene expression level quantified in the pool of BM. In this way, we determined gene expression levels comparatively to the calibrator such that ΔΔCTCT (AML sample)–ΔCT (pooled BM sample).

Real-time RT-PCR for microRNA-126 (miR-126)

The specific expression of hsa-mir-126 and of an internal control was quantified with real-time PCR using human hsa-miR-126-5pLNA and SNORD44 PCR microRNA assay kits, respectively, according to the manufacturer’s protocol (Exiqon A/S Skelstedet 16, DK-2950 Vedbaek, Denmark). U44 small nucleolar RNA was used as an internal control to normalize RNA input in the real-time RT-PCR assay.

Western blot analysis

Paired AML samples taken at diagnosis (DG) and CR were further analysed for SPRED1, MEK and ERK1/2 protein expression and phosphorylation by western blot. Western blots were performed as previously described.32 The anti-phosphoERK1/2 (Thr202/Tyr204) antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA), the anti-ERK1/2 (Sc-94, K23) antibodies obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA), and anti-SPRED1 (M23 P2G3) from ABCAM.8 Total proteins from 2–5 × 106 cells were extracted with 1 × Laemmli buffer. Proteins were separated by 12% SDS–PAGE and transferred onto nitrocellulose membrane (Schleicher & Schuell, Versailles, France). Membranes were incubated in Tris-buffered saline with 5% non-fat milk containing antibodies against SPRED1. After staining with secondary antibodies, the proteins were detected with Supersignal West Pico chemiluminescent substrate (Thermo Fisher Scientific, Villebon-sur-Yvette, France). Hybridization signals were quantified by using the GS-800 calibrated densitometer with Quantity One 1D analysis software (Bio-Rad Laboratories, Marnes-la-Coquette, France).

SPRED1 immunofluorescence microscopy

Cells were centrifuged onto glass coverslips for 3 min at 300 rpm in a cytospin centrifuge. Cells were fixed with 4% paraformaldehyde for 20 min at room temperature, washed with PBS1X, and then permeabilized with 0.01% Triton X100 for 30 min at room temperature. After extensive washing with PBS1X, cells were incubated with PBS1X containing 5% BSA for 1 h at room temperature. Rabbit anti-SPRED1 (1/100, Santa-Cruz, Y-24) or normal rabbit control IgG were added for 3 h at room temperature. After washing with PBS1X, cells were incubated with Alexa 555-labelled anti-rabbit (1/500, Invitrogen, Cergy Pontoise, France) and DAPI (1/100, Sigma Aldrich, Saint-Quentin Fallavier, France) for 1 h at room temperature. Fluorescence was analysed with Leica DMI 6000 microscope (Leica, Wetzlar, Germany) equipped with a 63 × 1.6 oil-immersion objective and a coupled device camera MicroMAX (Princeton Instruments, Trenton, NJ, USA). Pictures were analysed using ImageJ software.

Statistical analysis

SPRED1 genotype frequencies for SNPs rs7182445 and rs3751526 were calculated by gene counting and were compared with HapMap values by using the Chi-square test (Fisher exact test was used when the expected cell frequency was <5). Statistical testing was done with a two-tailed α level of 0.05. Because almost all patients with AL were of European origin, we used HapMap SNP data (the International HapMap Project/Public Release #21) from the CEU (CEPH collection of 113 Utah residents of northern and western European ancestry) and TSI (Toscani in Italia: 87 individuals) HapMap samples to estimate the expected genotype frequencies. Quantitative SPRED1 expression RT-PCR analysis statistical analysis was done using the Student’s t-test with α level of 0.05.

References

  1. 1

    Brems H, Chmara M, Sahbatou M, Denayer E, Taniguchi K, Kato R et al. Germline loss-of-function mutations in SPRED1 cause a neurofibromatosis 1-like phenotype. Nat Genet 2007; 39: 1120–1126.

    CAS  Article  Google Scholar 

  2. 2

    Pasmant E, Sabbagh A, Hanna N, Masliah-Planchon J, Jolly E, Goussard P et al. SPRED1 germline mutations caused a neurofibromatosis type 1 overlapping phenotype. J Med Genet 2009; 46: 425–430.

    CAS  Article  Google Scholar 

  3. 3

    Messiaen L, Yao S, Brems H, Callens T, Sathienkijkanchai A, Denayer E et al. Clinical and mutational spectrum of neurofibromatosis type 1-like syndrome. JAMA 2009; 302: 2111–2118.

    CAS  Article  Google Scholar 

  4. 4

    Brems H, Pasmant E, Van Minkelen R, Wimmer K, Upadhyaya M, Legius E et al. Review and update of SPRED1 mutations causing Legius syndrome. Hum Mutat 2012; 33: 1538–1546.

    CAS  Article  Google Scholar 

  5. 5

    Bentires-Alj M, Kontaridis MI, Neel BG . Stops along the RAS pathway in human genetic disease. Nat Med 2006; 12: 283–285.

    CAS  Article  Google Scholar 

  6. 6

    Pasmant E, Ballerini P, Lapillonne H, Perot C, Vidaud D, Leverger G et al. SPRED1 disorder and predisposition to leukemia in children. Blood 2009; 114: 1131.

    CAS  Article  Google Scholar 

  7. 7

    Schubbert S, Shannon K, Bollag G . Hyperactive Ras in developmental disorders and cancer. Nat Rev Cancer 2007; 7: 295–308.

    CAS  Article  Google Scholar 

  8. 8

    Wakioka T, Sasaki A, Kato R, Shouda T, Matsumoto A, Miyoshi K et al. Spred is a Sprouty-related suppressor of Ras signalling. Nature 2001; 412: 647–651.

    CAS  Article  Google Scholar 

  9. 9

    Nonami A, Kato R, Taniguchi K, Yoshiga D, Taketomi T, Fukuyama S et al. Spred-1 negatively regulates interleukin-3-mediated ERK/mitogen-activated protein (MAP) kinase activation in hematopoietic cells. J Biol Chem 2004; 279: 52543–52551.

    CAS  Article  Google Scholar 

  10. 10

    Inoue H, Kato R, Fukuyama S, Nonami A, Taniguchi K, Matsumoto K et al. Spred-1 negatively regulates allergen-induced airway eosinophilia and hyperresponsiveness. J Exp Med 2005; 201: 73–82.

    CAS  Article  Google Scholar 

  11. 11

    Stowe IB, Mercado EL, Stowe TR, Bell EL, Oses-Prieto JA, Hernandez H et al. A shared molecular mechanism underlies the human rasopathiesLegius syndrome and Neurofibromatosis-1. Genes Dev 2012; 26: 1421–1426.

    CAS  Article  Google Scholar 

  12. 12

    Papaemmanuil E, Hosking FJ, Vijayakrishnan J, Price A, Olver B, Sheridan E et al. Loci on 7p12.2, 10q21.2 and 14q11.2 are associated with risk of childhood acute lymphoblastic leukemia. Nat Genet 2009; 41: 1006–1010.

    CAS  Article  Google Scholar 

  13. 13

    Treviño LR, Yang W, French D, Hunger SP, Carroll WL, Devidas M et al. Germline genomic variants associated with childhood acute lymphoblastic leukemia. Nat Genet 2009; 41: 1001–1005.

    Article  Google Scholar 

  14. 14

    Levine RL . Inherited susceptibility to pediatric acute lymphoblastic leukemia. Nat Genet 2009; 41: 957–958.

    CAS  Article  Google Scholar 

  15. 15

    Case M, Matheson E, Minto L, Hassan R, Harrison CJ, Bown N et al. Mutation of genes affecting the RAS pathway is common in childhood acute lymphoblastic leukemia. Cancer Res 2008; 68: 6803–6809.

    CAS  Article  Google Scholar 

  16. 16

    Braun BS, Shannon K . Targeting Ras in myeloid leukemias. Clin Cancer Res 2008; 14: 2249–2252.

    CAS  Article  Google Scholar 

  17. 17

    Bundschu K, Walter U, Schuh K . The VASP-Spred-Sprouty domain puzzle. J Biol Chem 2006; 281: 36477–36481.

    CAS  Article  Google Scholar 

  18. 18

    Ishizaki T, Tamiya T, Taniguchi K, Morita R, Kato R, Okamoto F et al. miR126 positively regulates mast cell proliferation and cytokine production through suppressing Spred1. Genes Cells 2011; 16: 803–814.

    CAS  Article  Google Scholar 

  19. 19

    Miyoshi K, Wakioka T, Nishinakamura H, Kamio M, Yang L, Inoue M et al. The Sprouty-related protein, Spred, inhibits cell motility, metastasis, and Rho-mediated actin reorganization. Oncogene 2004; 23: 5567–5576.

    CAS  Article  Google Scholar 

  20. 20

    Yoshida T, Hisamoto T, Akiba J, Koga H, Nakamura K, Tokunaga Y et al. Spreds, inhibitors of the Ras/ERK signal transduction, are disregulated in human hepatocellular carcinoma and linked to the malignant phenotype of tumors. Oncogene 2006; 25: 6056–6066.

    CAS  Article  Google Scholar 

  21. 21

    Kandoth C, McLellan MD, Vandin F, Ye K, Niu B, Lu C et al. Mutational landscape and significance across 12 major cancer types. Nature 2013; 502: 333–339.

    CAS  Article  Google Scholar 

  22. 22

    Kindler T, Lipka DB, Fischer T . FLT3 as a therapeutic target in AML: still challenging after all these years. Blood 2010; 116: 5089–5102.

    CAS  Article  Google Scholar 

  23. 23

    Mizuki M, Fenski R, Halfter H, Matsumura I, Schmidt R, Müller C et al. Flt3 mutations from patients with acute myeloid leukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways. Blood 2000; 96: 3907–3914.

    CAS  Google Scholar 

  24. 24

    Cutts BA, Sjogren AK, Andersson KM, Wahlstrom AM, Karlsson C, Swolin B et al. Nf1 deficiency cooperates with oncogenic K-RAS to induce acute myeloid leukemia in mice. Blood 2009; 114: 3629–3632.

    CAS  Article  Google Scholar 

  25. 25

    Lee W, Jiang Z, Liu J, Haverty PM, Guan Y, Stinson J et al. The mutation spectrum revealed by paired genome sequences from a lung cancer patient. Nature 2010; 465: 473–477.

    CAS  Article  Google Scholar 

  26. 26

    Renneville A, Roumier C, Biggio V, Nibourel O, Boissel N, Fenaux P et al. Cooperating gene mutations in acute myeloid leukemia: a review of the literature. Leukemia 2008; 22: 915–931.

    CAS  Article  Google Scholar 

  27. 27

    Walter RB, Appelbaum FR, Tallman MS, Weiss NS, Larson RA, Estey EH . Shortcomings in the clinical evaluation of new drugs: acute myeloid leukemia as paradigm. Blood 2010; 116: 2420–2428.

    CAS  Article  Google Scholar 

  28. 28

    Lapillonne H, Llopis L, Auvrignon A, Renneville A, Labopin M, Mazingue F et al. Extensive mutational status of genes and clinical outcome in pediatric acute myeloid leukemia. Leukemia 2010; 24: 205–209.

    CAS  Article  Google Scholar 

  29. 29

    Christiansen DH, Andersen MK, Desta F, Pedersen-Bjergaard J . Mutations of genes in the receptor tyrosine kinase (RTK)/RAS-BRAF signal transduction pathway in therapy related myelodysplasia and acute myeloid leukemia. Leukemia 2005; 19: 2232–2240.

    CAS  Article  Google Scholar 

  30. 30

    Tartaglia M, Martinelli S, Cazzaniga G, Cordeddu V, Iavaroni I, Spinelli M et al. Genetic evidence for lineage-related and differentiation stage-related contribution of somatic PTPN11 mutations to leukemogenesis in childhood acute leukemia. Blood 2004; 14: 307–313.

    Article  Google Scholar 

  31. 31

    Gabert J, Beillard E, van der Velden VH, Bi W, Grimwade D, Pallisgaard N et al. Standardization and quality control studies of 'real-time' quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia Europe Against Cancer program. Leukemia 2003; 17: 2318–2357.

    CAS  Article  Google Scholar 

  32. 32

    Letourneux C, Rocher G, Porteu F . B56-containing PP2A dephosphorylate ERK and their activity is controlled by the early gene IEX-1 and ERK. EMBO J 2006; 25: 727–738.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This work was supported in part by grant from l'Association pour la Recherche sur les Maladies Hématologiques de l'Enfant (A.R.M.H.E.). We thank the patients and their parents for their participation. We are grateful to Pr. Akihiko Yoshimura for providing anti-Spred-1 antibody, Mrs. Frédérique Siotto for expert technical assistance, and Dr. Audrey Sabbagh for her statistical expertise.

Author information

Affiliations

Authors

Corresponding authors

Correspondence to E Pasmant or P Ballerini.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Pasmant, E., Gilbert-Dussardier, B., Petit, A. et al. SPRED1, a RAS MAPK pathway inhibitor that causes Legius syndrome, is a tumour suppressor downregulated in paediatric acute myeloblastic leukaemia. Oncogene 34, 631–638 (2015). https://doi.org/10.1038/onc.2013.587

Download citation

Keywords

  • SPRED1
  • tumour suppressor gene
  • childhood leukaemia
  • Legius syndrome
  • café-au-lait spots
  • neurofibromatosis type 1

Further reading

Search

Quick links