BRAF mutations in chronic lymphocytic leukemia
Billy Michael Chelliah Jebaraj1, Dirk Kienle2, Andreas Bühler1, Dirk Winkler1, Hartmut Döhner1, Stephan Stilgenbauer1 & Thorsten Zenz3,4

1Department of Internal Medicine III, University of Ulm, Ulm, Germany, 2Kantonspital Chur, Chur, Switzerland,
3Department of Translational Oncology, National Center for Tumor Diseases (NCT) and German Cancer Research Center (DKFZ), Heidelberg, Germany and 4Department of Medicine V, University of Heidelberg, Heidelberg, Germany

Abstract
BRAF mutations have been shown to occur at a high frequency in melanoma and thyroid cancer, but also at lower frequencies in hematological malignancies. To assess the potential role of BRAF, we have sequenced exons 11 and 15 of BRAF in 138 cases with chronic lymphocytic leukemia (CLL) and 32 cases of B-cell prolymphocytic leukemia (B-PLL). We found an incidence of BRAF mutations of 2.8% in CLL (4/138), while no cases with B-PLL showed BRAF mutations. The analysis of a cohort of patients with fludarabine-refractory disease (n = 87) showed no increase in the mutation incidence, suggesting that this mutation is not selected for during the disease progression. A limited analysis of the effect of BRAF inhibition in primary CLL cells showed no cell death induction in CLL samples with and without BRAF mutations. Our analysis suggests that BRAF mutations occur at a low frequency in CLL. The pharmacological inhibition of MEK/ ERK signaling using the mutant BRAF inhibitor PLX4720 showed no effect on viability in vitro in CLL cases.

Raf-1 and phosphorylates MEK leading to activation of the downstream growth and survival pathways [8]. Activating BRAF mutations have been shown to occur across tumor entities, with almost universal presence in hairy cell leuke- mia [9], a high frequency in melanoma and papillary thyroid carcinoma [10–12], and at lower frequencies in hematologi- cal malignancies [13,14].
Over 90% of the BRAF mutations in cancer correspond to the hotspot transversion T1799A (V600E) (http://www. sanger.ac.uk/genetics/CGP/cosmic/). Additional missense mutations reside in the glycines of the P-loop in exon 11 or in the activation segment in exon 15, mostly at codon 600. These mutations are oncogenic by disrupting the interaction of the P-loop and the activation segment, resulting in the loss of the inactive conformation [15], which leads to activation of the mitogen activated protein kinase (MAPK) pathway. The RAS/RAF/MEK/ERK signaling pathway regulates many key cellular processes and is often deregulated in human can-

cer. Active RAS-guanosine triphosphate (GTP) binds in the

Keywords: Lymphoid leukemia, genetic and other predisposing conditions, pharmacotherapeutics

Introduction
Mutations of ATM, NOTCH1, SF3B1 or TP53 have been described in chronic lymphocytic leukemia (CLL) [1–3]. Efforts to identify mutations by whole genome and exome sequencing have led to the identification of low frequency mutations in MYD88, FBXW7, ZMYM3, DDX3X, MAPK1,
XPO1, KLHL6, POT1 and CHD2 in CLL [4–7]. In the pres- ent study we assessed the role of BRAF mutations in CLL and B-cell prolymphocytic leukemia (B-PLL). v-Raf murine sarcoma viral oncogene homolog B1 (BRAF) is a serine/ threonine kinase involved in RAS/RAF/MEK/ERK/MAPK pathway signaling. As a response to extracellular growth factors, BRAF undergoes RAS-dependent dimerization with

membrane to multiple effector proteins, including the three members of the RAF kinase family (ARAF, BRAF and CRAF). Binding of these kinases to RAS-GTP leads to the activation of a cascade of kinases including RAF, MEK1/MEK2 and ERK1/ERK2. Constitutive activation of MAPK signaling has been reported in about half of CLL cases [16,17], which hints at a critical role of MAPK signaling in CLL cell survival. Iden- tifying activating mutations of BRAF in CLL may not only increase our understanding of the disease but also may have a direct impact on treatment, even if the size of the patient subgroup is small.

Materials and methods
Patient cohort
We analyzed cohorts of patients with CLL and B-PLL for
BRAF mutations in exon 11 and exon 15 by direct Sanger

Correspondence: Thorsten Zenz, MD, Department of Translational Oncology, National Center for Tumor Diseases (NCT) and German Cancer Research Center (DKFZ), Im Neuenheimer Feld 460, 69120 Heidelberg, Germany. Tel: + 49-6221-56-36039. Fax: + 49-6221-56-5388. E-mail: thorsten.zenz@ nct-heidelberg.de; Stephan Stilgenbauer, MD, Department of Internal Medicine III, University of Ulm, Albert-Einstein-Allee 23, 89081 Ulm, Germany. Tel: + 49-731-500-45548. Fax: + 49-731-500-24405. E-mail: [email protected]
Received 14 August 2012; revised 25 September 2012; accepted 17 October 2012
1177

sequencing. To assess the selectability of BRAF mutations as a potential indication of prognostic impact, we compared the incidence of the mutation in an unselected cohort treated at Ulm University (pretreated/untreated) and a cohort of patients with fludarabine-refractory CLL. Fluorescence in situ hybridization (FISH), immunoglobulin heavy chain variable (IGHV) gene mutation status and TP53 mutations were analyzed as previously described [18,19]. Analyses were performed with approval from the local ethics committee. The patient characteristics are summarized in Table I.

BRAF mutation analysis
DNA was isolated from fresh/frozen peripheral blood mono- nuclear cells (PBMCs) using TRIzol reagent (Invitrogen) or a Qiagen Allprep DNA/RNA Mini Kit and analyzed for the presence of mutations in exons 15 and 11 of BRAF by direct Sanger sequencing. The forward and reverse primers used for exon 15 of BRAF were E15For: 5 TAC CTA AAG TCT TCA TAA TGC TTG C 3 and E15Rev: 5 GTA ACT CAG CAG CAT
CTC AGG G 3 respectively. The forward and reverse primers
used for the analysis of exon 11 of BRAF were E11For: 5 CTG TTT GGC TTG ACT TGA CTT 3 and E11Rev: 5 CTT GTC
ACA ATG TCA CCA CA 3, respectively. Polymerase chain reactions (PCRs) were carried out in a 30 L volume using Discoverase denaturing high pressure liquid chromatogra- phy (dHPLC) polymerase from Invitrogen. Sequencing was done with a 3130xl genetic analyzer (Applied Biosystems).

In vitro experiments
In order to assess the effect of the BRAF inhibitor, PLX4720, and sorafenib (all inhibitors were purchased from Axon Medchem) on primary CLL cell viability, viably frozen PBMCs were cultured in RPMI 1640 medium with 10% fetal calf serum and exposed to increasing concentrations of the inhibitors or with 0.1–0.5% dimethylsulfoxide (DMSO) as control. The treatments with the inhibitors and DMSO were done in triplicate. Due to limitations of BRAF mutant sample availability, a sample with p.G469A from an independent analysis was included for the in vitro experiments, and the PLL cell line JVM3 harboring the BRAF mutation p.K601N was used for obtaining the optimal dose and to demonstrate

inhibitor activity (Supplementary Figures 1 and 2 to be found online at http://informahealthcare.com/doi/abs/10.3109/ 10428194.2012.742525). The acute promyelocytic leukemia cell line HL60, which is sensitive to inhibition of MEK (using PD0325901 from Axon Medchem), was used as a control for the activity of phospho-ERK antibody (Supplementary Figures 3 to be found online at http://informahealthcare. com/doi/abs/10.3109/10428194.2012.742525).

Cell viability and apoptosis assays
Cell viability was assessed at 24 h following treatment, using CellTitre-Glo reagent (purchased from Promega) and ana- lyzed using a GloMax microplate luminometer (Promega). Apoptosis was measured at 24 and 48 h after treatment using annexin V/7-aminoactinomycin D (7-AAD) staining (annexin V–fluorescein isothiocyanate [FITC] and 7-AAD purchased from BD Biosciences) and analyzed using fluo- rescence activated cell sorting (FACS). Percentage viability and percentage increase in apoptosis were calculated in each case by normalizing to their respective DMSO treated controls.

Immunoblotting
Protein lysates were prepared after 3 and 12 h of treatment with PLX4720 and sorafenib. Protein concentration was quantified using Bradford’s reagent, and an equal amount of protein (5–20 g) was separated using 4–12% NuPAGE gels (Invitrogen), transferred to a polyvinylidene fluoride (PVDF) membrane and hybridized with antibodies for phospho- ERK (pERK), total ERK 1/2, lamin B (antibodies purchased from Santa Cruz Biotechnology) and phospho-MEK1/2 (pMEK1/2) (antibody purchased from Cell Signaling). The blots were developed and quantified using ImageJ and in each case normalized to lamin.

Results
BRAF mutations occur at a low frequency in CLL
In the cohort of 170 patients with CLL/B-PLL, four patients were found to have BRAF mutations. More specifically, four of 138 patients with CLL (2.8%) had BRAF exon 15 mutations,

Table I. Patient characteristics of cohorts investigated for BRAF mutations in CLL/B-PLL.

BRAF exon 11/15 wild-type BRAF exon 15 mutation
B-PLL 32
CLL (unselected) 51 2
F-refractory CLL 87 2
FISH (CLL) Unselected F-refractory Unselected F-refractory
del(17p13) 9 28
del(11q23) 9 18 1
+ 12q 3 7 1
del(13q14) 18 18
Other 2 1
Normal 10 14 1 1
Total 51 86 2 2
IGHV unmutated 31 67 2 1
IGHV mutated 20 20 1
TP53 wild-type 36 53 2 2
TP53 mutation 11 31
CLL, chronic lymphocytic leukemia; B-PLL, B-cell prolymphocytic leukemia; F, fludarabine; FISH, fluorescence in situ
hybridization; IGHV, immunoglobulin heavy chain variable.
Association of BRAF mutations with diagnosis, disease stage and genetic categories in unselected and refractory
CLL cohorts.

while no patient with B-PLL (n = 32) was found to have a mutation. The mutations were heterozygous and targeted the activation segment at or around codon 600 (Supplemen- tary Table 1 to be found online at http://informahealthcare. com/doi/abs/10.3109/10428194.2012.742525). Nomutations were found in exon 11. The low incidence of BRAF mutations in CLL makes it difficult for assessment of associations with clinical and biological characteristics. The incidence was not different in the unselected cohort, which included pretreated and untreated patients (2/51), and the cohort with refractory CLL (2/87), suggesting that the mutation was not selected with advanced disease as is the case with TP53 mutations [1]. As shown in Table I there was no evident association of BRAF mutations with particular genetic subgroups, but incidences were low. Mutations of TP53 and BRAF did not coexist.

MAPK/ERK pathway activation in BRAF
mutant/wild-type CLL subgroups
We were interested to assess the effect of the BRAF inhibi- tor, PLX4720, and sorafenib in primary CLL cells. Thus we analyzed seven primary CLL samples with wild-type BRAF and three samples with BRAF mutation (p.V600T, p.V600E and p.G469A) for the basal expression levels of pERK and total ERK1/2. The HL60 cell line treated with MEK inhibitor (PD0325901 at 1 M) was used as a control for the activity of pERK antibody (Supplementary Figures 3 to be found online at http://informahealthcare.com/doi/abs/10.3109/1042819 4.2012.742525). The pERK and ERK1/2 expression levels as analyzed by Western blotting were variable among the dif- ferent CLL samples, but surprisingly, the cases harboring a mutant BRAF did not show an increase in phosphorylation status of ERK compared to BRAF wild-type cells (Figure 1). This was contrary to what has been reported for other can- cers with activating BRAF mutations.

Figure 1. Basal levels of phospho-ERK and ERK1/2 were analyzed in samples (n = 10) using Western blotting with lamin B as loading control. Blots were quantified using ImageJ. pERK and total ERK levels were normalized to lamin B. The level of pERK was normalized to total ERK1/2 for comparison of basal expression of pERK among samples.

BRAF mutant/wild-type CLLs respond similarly to treatment with PLX4720 and sorafenib
Treatment of CLL cells in vitro with increasing concentra- tions of the mutant BRAF inhibitor PLX4720 (1 M, 10 M and 20 M) marginally increased apoptosis in the BRAF wild-type and mutant cells (11.9% vs. 11.8%, respectively, after 20 M at 48 h) as shown in Figure 2(a). No specific sensitivity of BRAF mutants to PLX4720 was observed in the samples analyzed. Similar responses to treatment with sorafenib (1 M, 10 M and 20 M) were obtained in the BRAF mutant and wild-type cells. There was apoptosis of 45.4% of cells in BRAF mutant versus 48.1% in BRAF wild- type samples (20 M sorafenib after 48 h) [Figure 2(b)]. These inhibitors were effective only at high concentrations (Figure 2) and at the later time point (data not shown). In the BRAF mutant PLL cell line JVM3, inhibition after treat- ment with PLX4270 and sorafenib was detectable only at higher drug concentrations, as analyzed by Western blotting for phosphorylation of MEK (Supplementary Figures 2 to be found online at http://informahealthcare. com/doi/abs/10.3109/10428194.2012.742525).

Figure 2. BRAF wild-type (n = 7) and mutant (n = 3) samples were treated with 1 M, 10 M and 20 M concentrations of PLX4720 (a) and sorafenib (b) in triplicate and apoptosis was detected at 24 h (data not shown) and 48 h using annexin V/7-AAD FACS. Percent increase in annexin V positive cells was calculated by normalizing with DMSO control. Error bars represent standard error of mean of the samples.

Figure 3. Representative blots of samples which showed a low (a)/no (b) decrease in pERK levels on treatment with PLX4720 and sorafenib (20 M at 12 h) compared to DMSO treated. Blots were quantified using ImageJ and pERK levels were normalized to lamin B.

No correlation between phosphorylation status of pERK and apoptosis in CLL cells
The effect of PLX4720 and sorafenib on MAPK signaling in CLL was analyzed by investigating phosphorylation levels of ERK after 12 h of treatment with PLX4720 and sorafenib. A decrease in phosphorylation of > 40% (compared to DMSO control) was observed in five out of the 10 CLL samples (Figure 3). Surprisingly, the samples with BRAF mutation did not show a decrease in pERK levels after treatment with the inhibitors. Also, the response to treatment with PLX4720 and sorafenib did not correlate with the basal levels of pERK or the decrease in ERK phosphorylation post-treatment.

Discussion
Our data show that BRAF mutations in exon 15 occur at a low frequency in CLL and these mutations do not seem to be selected for during progression to refractory CLL. Treatment with the multikinase inhibitor, sorafenib, induced cell death in mutant BRAF and the wild-type BRAF subgroups, while treatment with the mutant BRAF inhibitor, PLX4720, failed to produce a significant effect on the BRAF mutant CLL cells in vitro.
BRAF mutations detected in CLL and the cell line JVM3 were found to be heterozygous, as observed by the presence of a wild-type peak in the sequences from the BRAF mutant

samples (Supplementary Table 1 to be found online at http://informahealthcare.com/doi/abs/10.3109/10428194. 2012.742525). Heterozygosity in BRAF mutations is preva- lent across different cancers including hairy cell leukemia [9], melanomas [20] and papillary thyroid carcinomas [12]. Melanoma cell lines with heterozygous BRAF mutations were found to be less responsive to treatment with BRAF inhibitors compared to the homozygotes [21], but the impact on the response in primary samples is not estab- lished. Cancers with heterozygous mutations in BRAF have been shown to be sensitive to inhibition in vitro [9,22] and in vivo [23].
The sensitivity of cells with BRAF mutation to inhibi- tors appears tissue context specific: data from colorectal cancer suggest that colon cancer cells with a V600E BRAF mutation are non-responsive to the inhibition of mutant BRAF by up-regulated EGFR signaling [24]. Our data do not show a clear activation of the MAPK pathway in the BRAF mutants compared to wild-types. This is contrary to what is observed in cases with activating BRAF muta- tions in other cancers, where a robust activation of the MAPK signaling, independent of external growth factors, is observed [9,22].
The multikinase inhibitor sorafenib, which inhibits vascular endothelial growth factor receptor 2 (VEGFR2), platelet derived growth factor receptor  (PDGFR), PDGFR and c-KIT in addition to CRAF and BRAF, was shown to induce apoptosis in CLL cells. However, the absence of a correlation between response to sorafenib and decrease in pERK suggests a MEK/ERK independent mechanism in the induction of apoptosis in these cases. Raf1, which is inhibited by sorafenib, has been shown to be involved in antagonizing apoptosis in a MEK/ERK
independent manner [25].
Two other studies have shown the occurrence of BRAF mutations in a similar proportion in CLL [6,7], but the response of CLL cells to inhibition of mutant BRAF has not been reported. Surprisingly, the V600E mutation that was found to constitute 90% of all BRAF mutations [10] is very rare in CLL [6,7,26,27] and other hematological malignancies, excluding hairy cell leukemia [28]. This mutation is of particular interest, as cancer cells carrying this mutation have been shown to be sensitive to treat- ment with PLX4720 and PLX4032. The sensitivity of other BRAF mutations to the available BRAF inhibitors has been shown to be variable [29]. Our analysis identified one case of CLL with a V600E mutation, but at least in vitro treatment showed no sensitivity to PLX4720 compared to the other BRAF mutants/BRAF wild-type cases. Treatment of cells harboring mutations other than V600E or wild- type BRAF with PLX4720 might prove to be detrimental, as the inhibitor has also been shown to confer pro-survival signals by trans-activating ERK signaling in these cells [22,30].

Acknowledgements
We thank Doris Winter for technical assistance.

Potential conflict of interest: Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.
This study was supported by the CLL Global Research Foundation, the “Helmholtz Gesellschaft” and “Dt. Krebshilfe” (109674).

References
[1] Zenz T, Mertens D, Kuppers R, et al. From pathogenesis to treatment of chronic lymphocytic leukaemia. Nat Rev Cancer 2010;10:37–50.
[2] Chiorazzi N, Rai KR, Ferrarini M. Chronic lymphocytic leukemia.
N Engl J Med 2005;352:804–815.
[3] Fabbri G, Rasi S, Rossi D, et al. Analysis of the chronic lymphocytic leukemia coding genome: role of NOTCH1 mutational activation. J Exp Med 2011;208:1389–1401.
[4] Puente XS, Pinyol M, Quesada V, et al. Whole-genome sequencing
identifies recurrent mutations in chronic lymphocytic leukaemia. Nature 2011;475:101–105.
[5] Wang L, Lawrence MS, Wan Y, et al. SF3B1 and other novel
cancer genes in chronic lymphocytic leukemia. N Engl J Med 2011; 365:2497–2506.
[6] Quesada V, Conde L, Villamor N, et al. Exome sequencing identifies
recurrent mutations of the splicing factor SF3B1 gene in chronic lymphocytic leukemia Nat Genet 2011;44:47–52.
[7] Zhang X, Reis M, Khoriaty R, et al. Sequence analysis of 515 kinase
genes in chronic lymphocytic leukemia. Leukemia 2011;25:1908–1910.
[8] Weber CK, Slupsky JR, Kalmes HA, et al. Active Ras induces heterodimerization of cRaf and BRaf. Cancer Res 2001;61:3595–3598.
[9] Tiacci E, Trifonov V, Schiavoni G, et al. BRAF mutations in
hairy-cell leukemia. N Engl J Med 2011;364:2305–2315.
[10] Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417:949–954.
[11] Cohen Y, Xing M, Mambo E, et al. BRAF mutation in papillary
thyroid carcinoma. J Natl Cancer Inst 2003;95:625–627.
[12] Nikiforova MN, Kimura ET, Gandhi M, et al. BRAF mutations in thyroid tumors are restricted to papillary carcinomas and anaplastic or poorly differentiated carcinomas arising from papillary carcinomas. J Clin Endocrinol Metab 2003;88:5399–5404.
[13] Smith ML, Snaddon J, Neat M, et al. Mutation of BRAF is
uncommon in AML FAB type M1 and M2. Leukemia 2003;17:274–275.
[14] Hou P, Liu D, Xing M. The T1790A BRAF mutation (L597Q) in childhood acute lymphoblastic leukemia is a functional oncogene. Leukemia 2007;21:2216–2218.
[15] Wan PT, Garnett MJ, Roe SM, et al. Mechanism of activation
of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell 2004;116:855–867.
[16] Muzio M, Apollonio B, Scielzo, et al. Constitutive activation of
distinct BCR-signaling pathways in a subset of CLL patients: a molecular signature of anergy. Blood 2008;112:188–195.
[17] Messmer D, Fecteau JF, O’Hayre M, et al. Chronic lymphocytic
leukemia cells receive RAF-dependent survival signals in response to CXCL12 that are sensitive to inhibition by sorafenib. Blood 2011;117:882–889.
[18] Kröber A, Seiler T, Benner A, et al. V(H) mutation status, CD38
expression level, genomic aberrations, and survival in chronic lymphocytic leukemia. Blood 2002;100:1410–1416.
[19] Döhner H, Stilgenbauer S, Benner A, et al. Genomic aberrations
and survival in chronic lymphocytic leukemia. N Engl J Med 2000;343:1910–1916.
[20] Willmore-Payne C, Holden JA, Hirschowitz S, et al. BRAF and
c-kit gene copy number in mutation-positive malignant melanoma. Hum Pathol 2006;37:520–527.
[21] Søndergaard JN, Nazarian R, Wang Q, et al. Differential sensitivity
of melanoma cell lines with BRAFV600E mutation to the specific Raf inhibitor PLX4032. J Transl Med 2010;8:39.
[22] Halaban R, Zhang W, Bacchiocchi A, et al. PLX4032, a selective
BRAF(V600E) kinase inhibitor, activates the ERK pathway and enhances cell migration and proliferation of BRAF melanoma cells. Pigment Cell Melanoma Res 2010;23:190–200.
[23] Dietrich S, Glimm H, Andrulis M, et al. BRAF inhibition in
refractory hairy-cell leukemia. N Engl J Med 2012;366:2038–2040.
[24] Prahallad A, Sun C, Huang S, et al. Unresponsiveness of colon cancer to BRAF (V600E) inhibition through feedback activation of EGFR. Nature 2012;483:100–103.

[25] Chen J, Fujii K, Zhang L, et al. Raf-1 promotes cell survival by antagonizing apoptosis signal-regulating kinase 1 through a MEK-ERK independent mechanism. Proc Natl Acad Sci USA 2001; 98:7783–7788.
[26] Ping N, Wang Q, Wang Q, et al. Absence of BRAF V600E mutation
in hematologic malignancies excluding hairy-cell leukemia. Leuk Lymphoma 2012 Jul 9. [Epub ahead of print]
[27] Langabeer SE, Quinn F, O’Brien D, et al. Incidence of the BRAF
V600E mutation in chronic lymphocytic leukaemia and prolymphocytic leukaemia. Leuk Res 2012;36:483–484.

Supplementary material available online
Supplementary Figures 1–3 and Table I showing additional results.

[28] Trifa AP, Popp RA, Cucuianu A, et al. Absence of BRAF V600E mutation in a cohort of 402 patients with various chronic and acute myeloid neoplasms. Leuk Lymphoma 2012 Apr 18. [Epub ahead of print]
[29] Yang H, Higgins B, Kolinsky K, et al. RG7204 (PLX4032), a selective
BRAFV600E inhibitor, displays potent antitumor activity in preclinical melanoma models. Cancer Res 2010;70:5518–5527.
[30] Poulikakos PI, Zhang C, Bollag G, et al. RAF inhibitors transactivate
RAF dimers and ERK signalling in cells with wild-type BRAF. Nature 2010;464:427–430.PLX-4720