Genomic and expression profiling identifies the B-cell associated tyrosine kinase Syk as a possible therapeutic target in mantle cell lymphoma

Genomic and expression profiling identifies the B-cell associatedtyrosine kinase Syk as a possible therapeutic target in mantlecell lymphoma

Andrea Rinaldi,1 Ivo Kwee,1,2 Monica

Taborelli,1 Cristina Largo,3 Silvia

Uccella,4 Vittoria Martin,4 Giulia

Poretti,1 Gianluca Gaidano,5 Giuseppe

Calabrese,6 Giovanni Martinelli,7 Luca

Baldini,8 Giancarlo Pruneri,9 Carlo

Capella,4 Emanuele Zucca,10 Finbarr E.

Cotter,11 Juan C. Cigudosa,3 Carlo V.

Catapano,1 Maria G. Tibiletti4 and

Francesco Bertoni1,11

1Laboratory of Experimental Oncology, Oncology

Institute of Southern Switzerland, Bellinzona,2Istituto Dalle Molle di Studi sull’Intelligenza

Artificiale, Manno, Switzerland, 3Cytogenetics

Unit, Centro Nacional Investigaciones

Oncologicas (CNIO), Madrid, Spain, 4Anatomic

Pathology Unit, University of Insubria, Ospedale

di Circolo, Varese, 5Division of Haematology,

Department of Medical Sciences and IRCAD,

Amedeo Avogadro University of Eastern

Piedmont, Novara, 6Dipartimento di Scienze

Biomediche, Sezione di Genetica Medica,

Universita’ di Chieti, Chieti, 7Department of

Haematoncology, European Institute of Oncology,

Milan, 8Ematologia 1, Dipartimento di Scienze

Mediche, Universita’ degli Studi di Milano,

Ospedale Maggiore IRCCS, Milano, 9Division of

Pathology and Laboratory Medicine, European

Institute of Oncology, Milan, Italy, 10Lymphoma

Unit, Oncology Institute of Southern Switzerland,

Bellinzona, Switzerland, and 11Department of

Experimental Haematology, Barts and The

London – Queen Mary’s School of Medicine and

Dentistry, London, UK

Received 26 August 2005; accepted for

publication 17 October 2005

Correspondence: Francesco Bertoni, MD,

Laboratory of Experimental Oncology,

Oncology Institute of Southern Switzerland, via

Vincenzo Vela 6, 6500 Bellinzona, Switzerland.

E-mail: [email protected]

Summary

Among B-cell lymphomas mantle cell lymphoma (MCL) has the worst

prognosis. By using a combination of genomic and expression profiling

(Affymetrix GeneChip Mapping 10k Xba131 and U133 set), we analysed 26

MCL samples to identify genes relevant to MCL pathogenesis and that could

represent new therapeutic targets. Recurrent genomic deletions and gains

were detected. Genes were identified as overexpressed in regions of DNA gain

on 3q, 6p, 8q, 9q, 16p and 18q, including the cancer genes BCL2 and MYC.

Among the transcripts with high correlation between DNA and RNA, we

identified SYK, a tyrosine kinase involved in B-cell receptor signalling. SYK

was amplified at DNA level, as validated by fluorescence in situ hybridisation

(FISH) analysis, and overexpressed at both RNA and protein levels in the

JeKo-1 cell line. Low-level amplification, with protein overexpression of Syk

was demonstrated by FISH in a small subset of clinical samples. After

treatment with low doses of the Syk inhibitor piceatannol, cell proliferation

arrest and apoptosis were induced in the cell line overexpressing Syk, while

cells expressing low levels of Syk were much less sensitive. A combination of

genomic and expression profiling suggested Syk inhibition as a new

therapeutic strategy to be explored in lymphomas.

Keywords: piceatannol, Affymetrix, isochromosome, 9q, Syk.

research paper

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 doi:10.1111/j.1365-2141.2005.05883.x

Mantle cell lymphoma (MCL) accounts for approximately 6%

of all non-Hodgkin lymphomas and patients affected by MCL

have very poor median time to progression and overall survival

(The Non-Hodgkin’s Lymphoma Classification Project, 1997).

To date, there are no standard treatments for MCL (Bertoni

et al, 2004a; Fisher, 2005).

A number of distinct genetical and biological alterations are

associated with the disease including the t(11;14)(q13;q32)

translocation, ATM alterations and 11q deletion (Bertoni et al,

2004b; Fernandez et al, 2005). Cell cycle regulation and the

nuclear factor (NF)-jB pathway are critically affected in MCL

(Martinez et al, 2003; Pham et al, 2003; Rosenwald et al, 2003;

Fernandez et al, 2005).

Microarray-based comparative genomic hybridisation (ar-

rayCGH) enables the study of unbalanced chromosomal

abnormalities, i.e. genomic DNA losses and gains at high

resolution (Pollack et al, 1999; Monni et al, 2001; Bignell et al,

2004; Huang et al, 2004). It offers the opportunity to identify

small affected regions, making it more feasible to identify

cancer-associated genes.

We applied a combined approach using arrayCGH to detect

genomic lesions and gene expression profiling to identify the

genes affected by DNA gains and losses. Because of the stability

of DNA, gene amplification is easier to measure than RNA or

protein overexpression. Therefore, determination of gene

amplification would be optimally suited to diagnostic appli-

cations, as in the case the HER2 status in clinical breast cancer

samples (Hicks & Tubbs, 2005). We applied the arrayCGH

technique using the GeneChip Mapping 10K Xba131 (Affyme-

trix Inc., Santa Clara, CA, USA) on 26 MCL samples to detect

the most recurrent gained and lost regions, and we combined

the gene expression and the whole genome profiling to identify

potential therapeutic targets.

Materials and methods

Cell lines and conventional cytogenetics

Four established human MCL cell lines (JeKo-1, Granta-519,

REC and NCEB1) were maintained in culture as previously

described (Lacrima et al, 2005). Cytogenetic analysis was

conventionally performed on chromosome preparations

obtained from cell line cultures. Cells were incubated

overnight with Colcemid (0Æ02 mg/ml) (Celbio, Milan,

Italy), harvested by hypotonic treatment with 1% sodium

citrate and repeated fixation in methanol:acetic acid 3:1.

Karyotype evaluation was performed using the Q-banding

technique. Chromosome abnormalities were described

according to the recommendations of the International

System for Human Cytogenetic Nomenclature (Mitelman,

1995). Only clonal abnormalities were considered in the

description of the tumour karyotype. The same structural

rearrangement or chromosomal gain had to be present in at

least two metaphases whereas the imbalance of a chromo-

some had to be detected in at least three metaphases. Where

different tumour cell populations were identified, the

karyotypes of the more represented cell populations have

been given in the Results.

DNA extraction

DNA from cell lines was isolated using the Puregen DNA

Isolation Kit (Gentra Systems, Minneapolis, MN, USA).

DNA from clinical material was extracted using the Qiagen

DNA Mini kit (Qiagen, Hilden, Germany). All DNA samples

were dissolved in reduced EDTA TE buffer (0Æ1 mmol/l of

EDTA, 10 mmol/l of Tris–HCl, pH 8Æ0).

Fluorescence in situ hybridisation and spectral karyotypingFISH

Fluorescence in situ hybridisation (FISH) and spectral

karyotyping FISH (SKY-FISH) and FISH analysis were

performed as previously described (Tibiletti et al, 1996;

Calabrese et al, 2000). Briefly, chromosomal preparations

were mounted in antifade solution containing 4¢-6-diamidi-

no-2-phenylindol (DAPI; Vysis, Downers Grove, IL, USA)

and observed on a Leica DMRA fluorescence microscope

(Leica, Wetzlar, Germany) with a cooled CCD camera

(Joko-cho, Hamamatsu City, Japan) coupled with FISH

software (Casti Amplimedical, Assago, Italy). Commercially

available probes, directly labelled with fluorescein or orange

fluorochromes were used (Vysis). The following 12 locus-

specific probes were hybridised: dual colour dual fusion

translocation probes LSI IgH/BCL2, LSI IgH/CCND1, LSI

API2/MALT1, LSI BCR/ABL; dual colour probes LSI P53/

ATM, LSI D13S319/13q34; dual colour break-apart rear-

rangement probe LSI BCL6; tri-colour dual fusion trans-

location probe LSI IgH/MYC and Cep8. Each probe was

checked on both normal metaphases and nuclei. The

interphasic results with the different types of probes were

obtained using specific cut-off values obtained with FISH

experiments on normal nuclei. For the interphasic evalua-

tion, we considered the number of copies of chromosome

regions with the conventional and the SKY approach and

the number of spots per cell. When more than one cell

population was identified the copies of regions in each cell

population was reported. FISH analysis was always per-

formed both on a minimum of 20 metaphases and at least

200 intact nuclei for each probe.

Cell lines were first tested by FISH and polymerase chain

reaction (PCR) for the MCL-specific translocations involving

the 14q32 (IgH) and the 11q13 (CCND1) loci. All of them

were positive by FISH. Conversely, PCR for t(11;14)(q13;32),

performed with the BCL1/JH Translocation Assay (InVivo-

Scribe Technologies, San Diego, CA, USA) was positive only in

REC and JeKo-1, but negative in Granta-519 and NCEB-1.

For validation of SYK gain by FISH, probes were made from

bacterial artificial chromosomes (BAC) clone RP11-489I19

(provided by M. Rocchi, Bari, Italy; http://www.biologia.

A. Rinaldi et al

304 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316

uniba.it/rmc/), chosen because it contains the gene. BAC DNA

was labelled with biotinylated 16-dUTP using a nick-transla-

tion kit (Boehring Mannheim-Roche, Mannheim, Germany).

A CEP 9 SpectrumGreen probe (Vysis), targeting the centro-

meric-alpha satellite DNA sequence at 9p11–9q11, was used as

an internal FISH control. Briefly, chromosomal DNA was

denatured in 70% formamide, 2x saline sodium citrate (SSC)

pH 7Æ0 at 72�C for 2 min. Cot-1 and BAC DNA were dissolved

and denatured at 80�C for 10 min and pre-annealed for 1 h at

37�C. BAC DNA was mixed with denatured alpha-satellite

probe. Hybridisation was performed at 42�C overnight in

Hybryte (Vysis). Washings were performed twice at 40�C for

10 min in 50% formamide, 2x SSC and three times for 5 min

in 2x SSC. The slides were then incubated for 30 min in non-

fat dried milk. A FISH performed on normal metaphases

showed the correct localisation of the two probes (data not

shown). FISH analysis on MCL clinical samples was performed

on nuclei obtained after conventional chromosome prepar-

ation of fresh MCL samples using BAC-containing SYK and

centromere of chromosome 9 as probes.

RNA extraction and gene expression microarrays

RNA was extracted from cell lines with the Trizol method

(Trizol; Invitrogen Life Technologies, San Diego, CA, USA),

with an additional RNA clean-up step using the RNAeasy

Purification kit (Qiagen). The quantity and quality of RNA

was assessed by obtaining the ratio of absorbance values at

260 and 280 nm on a Nanodrop spectrophotometer (Na-

noDrop Technologies Inc., Wilmington, DE, USA), and by

visualisation of intact 28S and 18S ribosomal RNA bands on

a Bioanalyser 2100 (Agilent Technologies Inc., Palo Alto,

CA, USA). Labelling and hybridisations were performed

according to the standard protocol from Affymetrix, starting

with 10 lg of total RNA and using Affymetrix U133A and

U133B GeneChip microarrays. Washings and scanning were

performed according to Affymetrix protocols using Fluidics

Station 400 and GeneChip Scanner 3000 (Affymetrix Inc.).

Data acquisition was performed using the Affymetrix

GeneChip GCOS 1.1. Raw data will be deposited at the

Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.-

gov/geo/).

ArrayCGH: cDNA arrays

The cDNA microarray assays were performed using the CNIO

OncoChip TM (v1.1a) (CNIO, Madrid, Spain). The micro-

array contains 7657 cancer- or tissue-specific cDNA clones,

corresponding to 6386 known genes and expressed sequence

tags and 142 non-human clones as negative controls. The full

list of genes on the array is available at http://bioinfo.cnio.es/

data/oncochip. Experiments were performed as previously

described (Pollack et al, 1999; Monni et al, 2001). Seven

micrograms of the test and reference DNA were digested for

15 h with AluI and RsaI (Life Technologies, Inc., Rockville,

MD, USA). The digested DNAs were labelled with fluorescent

Cy5-dUTP (test) and Cy3-dUTP (reference) using the Bio-

prime Labeling Kit (Invitrogen, Basel, Switzerland). Hybrid-

isations and post-hybridisation washings were performed at

50�C as described for expression arrays (Tracey et al, 2003).

Slides were scanned for Cy3 and Cy5 fluorescence in an Agilent

Array Scanner (Agilent Technologies).

ArrayCGH: single nucleotide polymorphism arrays

Affymetrix GeneChip Mapping 10 K Xba131 arrays have been

used according to the GeneChip Mapping Assay Protocol.

Briefly, 250 ng of genomic DNA was digested with XbaI (New

England Biolabs, Beverly, MA, USA), ligated with Xba adapter

and PCR amplified in five replicates on a MJR Thermal Cycler

200 (MJ Research, Cambridge, MA, USA). PCR products were

purified with the MinElute 96UF PCR Purification Plate and

PCR purification kit (Qiagen). Purified PCR products (20 lg)were fragmented and labelled. The microarrays were hybrid-

ised at 48�C for 16–18 h, washed and stained on an Affymetrix

Fluidics Station 400, and scanned using the Affymetrix

GeneChip Scanner 3000. Data acquisition was performed

using the Affymetrix GeneChip GCOS 1.1 and GDAS 1.0. Raw

data will be deposited at the GEO (http://www.ncbi.nlm.nih.-

gov/geo/).

Data analysis: gene expression

Robust Multichip Analysis (RMA) expression values were

calculated for both HG-U133A and HG-U133B chips from the

raw CEL files using the Bioconductor statistical package

(http://www.bioconductor.org) (Irizarry et al, 2003; Gentle-

man et al, 2004). Probes present on both A and B chips were

represented by their mean value.

Data analysis: cDNA arrayCGH

Spot data were quantified using the GenePix Pro 5Æ0 software

(Axon Instruments Inc., Union City, CA, USA). To normalise

the data, the ratio Cy3:Cy5 was adjusted to a normalised factor

equal to the median ratio value of all spots in the array. Only

measurements with fluorescence intensities over 90% of the

630 No-DNA spots included on the array were considered

reliable. Bad spots or areas of the array with obvious defects

were manually flagged. The Cy3:Cy5 ratios of the duplicated

spots of the array were averaged. Upper and lower thresholds

were established at 0Æ85 and 1Æ22 Cy5:Cy3 ratios after the

hybridisation of genomic DNAs from 45XO, 46XX, 47XXX,

48XXXX and 49XXXXX cell lines versus the control pool

(46XX) (Pollack et al, 1999). DNA copy number (CN) profiles

were displayed as a moving average (symmetric three-nearest

neighbours). Map positions for arrayed human cDNA clones

were assigned according to the July 2003 version of the

University of California-Santa Cruz Biotechnology Human

Genome Working Draft (http://genome.ucsc.edu/).

Syk in Mantle Cell Lymphoma

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 305

Data analysis: single nucleotide polymorphism arrayCGH

The GeneChip Chromosome Copy Number Analysis Tool

(CNAT; Affymetrix) v 2.0 was used to calculate CN and loss

of heterozygosity (LOH) probability (Huang et al, 2004)

starting from the GDAS-derived CEL files. The CN was

probe-wise normalised using the median signal of all patient

samples and scaled back globally so as to retain its previous

global median value. We repeated this procedure for the

autosomes and the X chromosome separately. We found the

normalisation to be important, to remove remaining

systematic undulations in the signal. Using this obtained

signal, the underlying CN was inferred using the circular

binary segmentation (CBS) algorithm (Olshen et al, 2004).

The CBS algorithm finds the best possible breakpoints along

the chromosome and regresses the data using a piece-wise

constant function. The CNAT LOH values were not

smoothed. Regions with significant gain were then defined

as those with a CN >2Æ5, and those with significant loss as

those having a CN <1Æ5. We defined regions with significant

LOH as those with a value >4.

Data analysis: correlation between genomic and expressionvalues

To quantify the correlation between gene expression and

genomic alteration, we computed, probe wise across the

samples, the Pearson’s correlation coefficients for each single

nucleotide polymorphism (SNP) probe paired with each

expression probe situated between its neighbouring SNPs,

and repeated this for all SNP probes. In this way, each

expression probe was assigned twice to each neighbouring

SNP, except for the terminal ends of the chromosome.

Expression values were prefiltered for a minimum standard

deviation of 0Æ2 and a positive genomic alteration (CN

higher than three) in at least one of the samples. We

additionally required a maximum deviation from the median

expression value of at least 1 in RMA scale, i.e. a twofold

change in absolute scale. Correlation coefficients more

extreme than 0Æ83 were identified as being statistically

significant for our sample size of four. The obtained set of

genes after filtering, represent those genes with significant

differential expression and positively correlating with the

underlying chromosomal gain.

Immunoblotting

Cells were lysed in 50 mmol/l of Tris–HCl pH 7Æ4, 250 mmol/l

of sodium chloride, 5 mmol/l of EDTA, 2 mmol/l of phenyl-

methylsulphonyl fluoride, 50 mmol/l of sodium fluoride,

10 mmol/l of sodium orthovanadate, 0Æ1% Nonidet P-40

(Sigma, Fluka Chemie GmbH, Buchs, Switzerland), 1%

protease inhibitor cocktail (Sigma). Proteins were separated

on sodium dodecyl sulphate-polyacrylamide gels and trans-

ferred onto polyvinylidene difluoride membranes. Blots were

incubated with antibodies against Syk (LR; Santa Cruz

Biotechnology Inc., Santa Cruz, CA, USA), Phospho-Syk

(Phospho Tyr323; Cell Signaling Technology, Beverly, MA,

USA) and anti-a-tubulin (Calbiochem, La Jolla, CA, USA) and

then developed with peroxidase-conjugated secondary anti-

bodies and the enhanced chemiluminescence system (ECL;

Amersham Biosciences, Buckinghamshire, UK).

Immunohistochemistry

Ten MCL specimens (CD20, CD5, and Cyclin D1 positive and

CD23 negative) were selected from the Pathology archives.

Eight were nodal, two extranodal (rectal and oropharyngeal).

Immunohistochemistry was performed on 3-lm thick sections

obtained from paraffin blocks of each lymphoma. After

deparaffinisation, endogenous peroxidase was quenched with

3% hydrogen peroxide. The sections were incubated overnight

with monoclonal anti-Syk antibody (clone SP147; Spring

Bioscience, Freemont, CA, USA) at 4�C. Biotin-labelled

secondary horse anti-mouse antibody was incubated for 1 h

at room temperature, followed by ABC-peroxidase complex

(1 h at room temperature). Specificity controls consisted of the

omission of the first layer, and the use of control tissues known

to express or not the antigen. An intensity score was used to

evaluate the intensity of Syk immunoreactivity: +, very faint;

1+, faint; 2+, moderate; 3+, intense. In addition, a semi-

quantitative evaluation of the percentage of immunoreactive

neoplastic cells was performed.

Cell proliferation, cell growth inhibition and apoptosisassays

Cell proliferation was assessed by measuring 3-(4,5-dimethyl-

thiazol-2-yl)-2,5-dimethyl tetrazolium bromide (MTT) dye

absorbance. Cells were exposed to increasing doses of picea-

tannol (3,4,3¢,5¢-tetrahydroxy-trans-stilbene; Sigma) for 24, 48

and 72 h. Adsorbance was read at a wavelength of 570 nm on a

Beckman Coulter AD 340D microplate reader (Becton Dickin-

son, Mountain View, CA, USA). Results were given as

percentage of the untreated control samples. The doses

corresponding to the 50% inhibitory concentration (IC50) were

estimated using the R statistical package (Troester et al, 2004).

For cell growth inhibition, cell lines were continuously

exposed to the IC50 doses. Cell number and cell viability were

determined daily using the trypan blue dye exclusion test. A

P < 0Æ05, obtained with a paired two-sample Student’s t-test,

was considered to be statistically significant.

For detection of apoptotic cells by annexin V-propidium

iodide (PI) staining, control and drug-treated cells were

harvested, washed with phosphate-buffered saline, and incu-

bated with annexin V-fluorescein isothiocyanate (Bender

MedSystems, Vienna, Austria) and PI according to the

manufacturer’s instructions before analysis by flow cytometry.

Intact cells (annexin)/PI)) were discriminated from apoptotic

cells and necrotic cells (annexin+/PI+) using the Cell Quest

A. Rinaldi et al

306 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316

Fig

1.HeatplotsofsignificantlyalteredDNAregionsin

26mantlecelllymphoma(M

CL)cases;

x-axis,chromosomelocalisationandphysicalmapping;

y-axis,MCLcases.Thedashed

linerepresentsthe

locationofthecentromeres.(A)Heatplotofgain

(white)

andlosses

(black)as

assessed

bythresholdingtheinferred

copynumber

estimates.T

hethresholdsusedwere1Æ5and2Æ5,fordetermininglossand

gain

respectively.(B)Heatplotofsignificantloss

ofheterozygosity

(LOH)probability(w

hite)

asassessed

bythresholdingtheraw

LOH

estimates.Thethreshold

usedwas

LOH

>4,

fordetermining

significantLOH.

Syk in Mantle Cell Lymphoma

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 307

–40–2002040

Lo

ss f

req

uen

cy (

low

er p

lot)

ver

sus.

LO

H f

req

uen

cy (

up

per

plo

t)

Frequency (%)

12

34

56

78

910

1112

1314

1516

1718

1920

2122

X

010203040

Gai

n f

req

uen

cy

Frequency (%)

12

34

56

78

910

1112

1314

1516

1718

1920

2122

X

Fig

2.Frequency

ofgenomicaberrationsin

22mantlecelllymphoma(M

CL)clinicalsamplesandin

fourMCLcelllines:(A)regionsofDNAlosses,asassessed

byareductionin

copynumber

(lower

plot)

orbythepresence

ofloss

ofheterozygosity

(upper

plot);

x-axis,chromosomelocalisationandphysical

mappingin

Mb;

y-axis,percentage

ofcasesshowingDNAlosses;(B)regionsofDNAgains,as

assessed

byan

increase

incopynumber;

x-axis,chromosomelocalisationandphysical

mappingin

Mb;

y-axis,percentage

ofcasesshowingDNAgain.

A. Rinaldi et al

308 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316

software (Becton Dickinson). Each treatment was performed

in triplicate and experiments were repeated at least three times.

Results

ArrayCGH validation

To assess the impact of the cell ploidy on arrayCGH results, we

compared cytogenetic and FISH results on a panel of 12 loci

with the data obtained using the GeneChip Xba 131 and the

CNIO cDNA OncoChip microarrays. Conventional cytogen-

etics and SKY-FISH analysis (Fig S1) demonstrated homogen-

eous diploid karyotypes in REC and Granta-519 cell lines. On

the contrary, JeKo-1 and NCEB1 showed heterogeneous

chromosome complements, with triploid and tetraploid cell

populations respectively. Cell lines were studied using the

GeneChip Xba 131 arrays, the Affymetrix SNP-based oligonu-

cleotides microarrays, and with the CNIO cDNA OncoChip

(JeKo-1 profile is shown in Fig S2). Assuming that FISH is the

best technique to quantify chromosome CNs, the two array-

CGH techniques gave similar results in terms of their ability to

detect DNA losses or gains (Table S1). The rate of concor-

dance between arrayCGH and FISH was 74% (71–77%) as

evaluated on a panel of 12 loci analysed in the four MCL cell

lines. The results were worse when analysing samples charac-

terised by non-diploid versus diploid cells: 60% (58–62%) vs.

87% (83–92%). Based upon the literature (Bignell et al, 2004;

Huang et al, 2004) and the obtained results, we analysed MCL

clinical samples with the SNP-based platform, commercially

available and with good genome coverage.

Recurrent DNA losses and gains

To detect recurrent DNA losses and gains, 26 MCL samples,

including the four MCL cell lines, were analysed with the

Affymetrix GeneChip Xba 131 microarrays (Fig 1).

Figure 2A shows the regions of losses among 26 MCL

samples as detected by a reduction in CN or by LOH. Table I

shows the genomic regions deleted in more than 18% of the

clinical samples. Reduction in CN and the presence of LOH

behaved in a similar way, with minor differences because of

technical issues, such as percentage of normal infiltrating cells

in the tumour sample or for subtelomeric deletion with no

LOH, the small number of probes available to compute the

LOH score. However, LOH in the absence of any CN change

was detected in 27% of the patients on 11p12.2 (50 096 999–

55 544 415, according to the National Center for Biotechno-

logy Information Build 35, http://www.ncbi.nlm.nih.gov/

genome/guide/human) and in 18% on 2p24.3 (14 756 245–

15 623 846).

Figure 2B shows the regions of recurrent gains in patients and

cell lines. Recurrent gains, present inmore than 20% of theMCL

clinical samples, were detected in 3q25.1–3q29 (150 938 349–

194 075 850) with similar frequency among patients and the cell

lines (27% and 25% respectively). Gains were also observed in

18q21.32–18q22Æ1 (55 411 565–64 670 508) in 18% of the

patients and in one of the four cell lines.

Differences were observed between the frequency of recur-

rent deletions and gains in clinical samples versus cell lines,

especially as expected, regarding lesions affecting TP53 at 17p.

In general, LOH was observed at higher frequency in cell lines,

Table II. Genes showing a positive correlation

between increased DNA copy number and gene

expression in mantle cell lymphoma cell lines.

Gene symbol Localisation

GYG, SSR3, LAMP3, PLR2H, HES1, EST 3q24–3q29

FGD2, FLJ11236, MRPL2, C6ORF89, MTCH1, EST 6p21-ter

SYK, BICD2, PHF2, PTCH, TGFBR1, SEC61B, TMEFF1, CDW92, ENDOG 9q22–9q31.2

ABAT, CARHSP1, USP7, PROO149, MHC2TA, LOC51760, ESTs 16p12.3–16p13.2

NEDD4L, ZNF532, LMAN1, BCL2, FVT1, VPS4B, TXNDC10, RTTN, SOCS6 18q21–18q22.2

Table I. Frequency of DNA losses among 26

mantle cell lymphoma samples, and comparison

between patients and cell lines.

Chromosome Mapping (NCBI Build 35) Patients (%) Cell lines (%)

1p35.1–1p36.12 147 558 99–347 961 67 23–32 25–50

9p21.3 21 971 583–22 026 367 27 50

13q14.2–13q21.1 46 240 774–54 701 519 27 0

1p13.3–21.3 98 641 388–108 570 569 18 18–23

12q24.33 130 375 262–130 375 660 36 0

11q22.1–11q23.1 101 131 785–111 865 891 23 0–25

6p21.31 32 521 295–34 589 070 18 25

6q23.3–6q25.1 135 371 662–150 859 661 18 25

7q31.31–7q32.3 117 126 103–131 824 485 18 25

10p15.3 135 698–1 636 959 18 0

17p12–17p13.1 7 478 517–15 053 487 18 100

9q21.11 118 253 831–133 267 975 18 0

Syk in Mantle Cell Lymphoma

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 309

and this could be due to both selections during in vitro growth,

but also to the presence of normal contaminating cells in the

patient-derived material that can decrease the LOH detection

rate.

In the cell lines, to select candidate cancer genes, we

searched for genes with a correlation between an increased

DNA level (CN > 3) and high expression. The analysis

identified 108 probes corresponding to 37 Unigene clusters,

including well-known lymphoma genes such as BCL2 and

CMYC (Table II).

SYK amplification and expression in cell lines and clinicalsamples

The SYK, involved in the B-cell receptor (BCR) signalling

pathway (Niiro & Clark, 2002), was among the genes with a

Fig 3. Fluorescence in situ hybridisation (FISH) showing gain of SYK in JeKo-1 cell line and in one mantle cell lymphoma (MCL) clinical sample: (A)

spectral karyotyping FISH magnification showing three markers derivatives from chromosome 9; (B) FISH analysis using bacterial artificial chro-

mosomes targeting SYK region (red) and centromeric probe (green) on metaphases of JeKo-1 MCL cell line. The derivative chromosomes showed in

tandem duplication of 9q22.1 region. The white arrow indicates the 9q isochromosome with double 9q22.1 in tandem duplications; (C) immu-

nohistochemistry showing Syk expression in one MCL clinical sample (original magnification ·440), matched with the corresponding FISH on

interphase nuclei (SYK region in red and centromeric probe in green).

A. Rinaldi et al

310 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316

high correlation between DNA and gene expression in the cell

lines (Table II). As SYK is a member of the recently described

‘BCR/proliferation’ gene cluster of diffuse large B-cell lympho-

mas (Monti et al, 2005) and specific Syk inhibitors are

available (Wieder et al, 2001; Wong et al, 2004; Wong &

Leong, 2004), it appeared to be an interesting gene for further

validation and study. ArrayCGH and gene expression data

obtained in JeKo-1 cell line are shown in Fig S3. The raw CN

plot was suggestive of a higher CN of the SYK containing

region in comparison with the rest of the 9q22 region, all

gained in JeKo-1. These data were validated by FISH using

simultaneously BAC and centromeric probes both on inter-

phasic nuclei and on metaphases (Fig 3A). As BAC RP11-

489I19 contained SYK as the only gene, FISH experiments

demonstrated that JeKo-1 had SYK amplification (six, seven,

eight or nine copies per cell) in contrast to the presence of

three or four chromosome 9 centromeric regions. FISH

showed that the long arms of two markers contained in

tandem duplications of the genomic region corresponding to

SYK. In 30% of JeKo-1 cells, FISH also revealed the presence of

an isochromosome that originated from 9q duplication

containing in tandem duplication of 9q22.2 region. Conven-

tional cytogenetic and SKY analysis previously demonstrated

that JeKo-1 karyotype was heterogeneous, showing different

chromosome markers (3, 4, 5 and 6 markers for cell) that

originated from chromosome 9 (Fig 3B). We then looked at

Syk protein level. Immunoblotting (Fig 4) showed that all the

cell lines expressed Syk. JeKo-1 had very high expression of

Syk, in accordance with arrayCGH, gene expression and FISH

data. The protein was phosphorylated in all the four cell lines

on the activation sites TYR 525/526 (Zhang et al, 2000). To

assess the relevance of this observation on clinical samples, we

performed immunohistochemistry for Syk and FISH analysis

on an independent series of MCLs. Immunohistochemistry in

10 MCL clinical samples showed Syk expression in all cases,

with a range of 1+ to 3+ intensity score. Eight of 10 MCL

showed Syk overexpression, with 2+ or 3+ intensity score and

a percentage of immunoreactive cells higher than 40%

(Fig 3C). For comparison, in normal lymph nodes used as

controls we observed Syk immunoreactivity in B-cell zones,

with mantle zone showing an intensity score of 1+/2+ while

follicular centres showed an intensity score of ±; T-cell zone

A REC

120

100

80

60

40

20

00 50

RecJeko1Granta519Nceb

6·25

Piceatannol

% o

f cel

l gro

wth

12·5

Granta-519 NCEB-1 JeKo-1

Syk

α-tubulin

B

Fig 4. Immunoblotting and 3-(4,5-dimethylthiazol-2-yl)-2,5-dimethyl tetrazolium bromide (MTT) assay showing SYK overexpression in JeKo-1 and

the high sensitivity to increasing doses of piceatannol in JeKo-1. (A) Syk protein expression in four mantle cell lymphoma (MCL) cell lines. (B)

Cytotoxic effect on MCL cell lines treated with increasing doses of SYK inhibitor piceatannol for 72 h; x-axis, piceatannol concentrations (lmol/l);

y-axis, percentage of untreated cells.

Table III. Concomitant fluorescence in situ hybridisation (FISH)

analysis and immunohistochemistry on five mantle cell lymphoma

clinical samples.

Cases

FISH

Immunohisto-

chemistry

SYK status Cells (%)

Syk

intensity

Cells

(%)

Case 1 Low level amplification 50 3+ 80

Case 2 Trisomy 9 20 3+ 40

Low level amplification 18

Case 3 Low level amplification 50 2+ 70

Case 4 Low level amplification 15 1+/2+ 80

Case 5 Diploid 100 1+/2+ 70

Low level of amplification at FISH indicates five to eight copies of SYK

in the presence of two chromosome 9 centromeres.

Syk in Mantle Cell Lymphoma

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 311

was completely negative. In five of these 10 MCLs, nuclei

obtained with conventional cytogenetics were available. Inter-

phasic FISH analysis performed on these cases revealed low

level of SYK amplification (five, six, seven or eight copies of

SYK in contrast to two copies of centromeres of chromosome

9) in four samples (Fig 3C). As shown in Table III, only one

clinical sample revealed disomic complement of Syk. Interest-

ingly, immunohistochemical overexpression was found in

cases showing Syk amplification (Table III).

Syk as a therapeutic target

MCL cell lines were treated with increasing doses of piceatan-

nol, a previously reported specific Syk inhibitor (Wieder et al,

2001). Piceatannol had a strong cytotoxic effect in JeKo-1,

which constitutively overexpressed Syk, with more than 80% of

cell growth arrest at 12Æ5 lmol/l (Fig 4). Granta-519, REC and

NCBE-1 cell lines, which expressed lower levels of Syk protein,

were much less sensitive to piceatannol. The IC50 doses,

calculated after 72 h of drug exposure, were 8 lmol/l for JeKo-

1, 30 lmol/l for Granta-519, 50 lmol/l for REC and over

100 lmol/l for NCEB-1. JeKo-1 underwent statistically signi-

ficant growth inhibition before 24 h when treated with the IC50

piceatannol dose, in accordance with MTT data (Fig 5A).

Treatment with piceatannol IC50 dose decreased the level of Syk

phosphorylation (Fig 5B). The ability of piceatannol to induce

apoptosis was evaluated by Annexin V staining. When exposed

to a concentration of 12Æ5 lmol/l, JeKo-1 showed a high

percentage of necrotic and apoptotic cells (39%) (Fig 5C). No

effects were observed in the other cell lines REC and Granta-

519, when exposed to the same piceatannol dose.

Discussion

By using a combination of genomic and expression profiling,

we analysed a series of MCL samples to identify regions

containing genes that might be relevant to the MCL patho-

genesis and that could represent new therapeutic targets.

18

15

12

9

6

3

00 hr

Piceatannol

Piceatannol

6%

4%

0%

21%

9%

9%–

–

p-Syk

α-tubulin

+

+

Cel

l cou

nt (

105 /

ml)

24 hr 48 hrUntreated Treated

72 hr

A

B C

104

104

103

103

102

102

Empty

Em

pty

101

101

100

104

103

102

Em

pty

101

100

100104103102

Empty101100

Fig 5. Growth inhibition and induction of apoptosis in JeKo-1 when treated with SYK inhibitor piceatannol. (A) Growth inhibition of Jeko-1

exposed to 12Æ5 lmol/l piceatannol; x-axis, days; y-axis, viable cell numbers. (B) Decreased level of Syk phosphorylation in JeKo-1 after exposure to

piceatannol IC50 for 24 h. (C) Apoptosis measurement by Annexin V staining in JeKo-1 cell line, untreated and treated with the 12Æ5 lmol/l

piceatannol for 48 h. For each plot: lower left quadrant, viable not apoptotic cells; upper right quadrant, damaged cell, late stage apoptosis; upper left

quadrant, dead cells, end stage apoptotic or necrotic; lower right quadrant, cells in early stage apoptosis; x-axis, Annexin V fluorescence intensity; y-

axis, propidium iodide fluorescence intensity.

A. Rinaldi et al

312 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316

First, we validated arrayCGH, and especially the Affymetrix

GeneChip Mapping 10k Xba131, as an appropriate method to

investigate genomic amplification and loss. Both the GeneChip

microarrays – and the CNIO OncoChip cDNA-microarrays,

compared well with the classic FISH analysis. The concordance

rate was very high for diploid cells, but lower for triploid or

tetraploid cells. The discordances between FISH and microar-

rays were generally due to changes of one CN, such loss or gain

in the context of tetraploidy, whilst all high CN changes were

detected by arrayCGH. No false-positive gains were detected.

Then, we analysed 26 MCL samples, including the four

established cell lines, with the Affymetrix GeneChip Xba131.

Recurrent deletions were detected at 1p, 9p, 13q, 12q, 11q, 6q,

7q, 10p, 17p and 9q, while gains affected chromosome 3q and

18q. As a whole, our data showed recurrently deleted and

gained regions similar to other recently reported series (De

Leeuw et al, 2004; Rubio-Moscardo et al, 2005; Schraders et al,

2005; Tagawa et al, 2005). Some differences can be partially

explained by the relatively small number of cases in our study,

and by the different technologies used and the different

algorithms applied in data analysis. As is well known for gene

expression data derived from different platforms (Marshall,

2004), discrepancies can also be expected for arrayCGH

studies. In our study, the genome profiling obtained with

two different platforms, SNP- and cDNA microarrays were

similar, but not perfectly overlapping, which is also shown by

Zhao et al (2004). The technique we used is based on 25mer

oligonucleotides initially designed for large-scale genotyping

(Kennedy et al, 2003; Matsuzaki et al, 2004), and shown to be

applicable for the detection of cancer alterations (Bignell et al,

2004; Huang et al, 2004; Zhao et al, 2004). The other

published MCL studies all used microarrays that were obtained

with spotting probes derived from BAC genomic clones (De

Leeuw et al, 2004; Rubio-Moscardo et al, 2005; Schraders et al,

2005; Tagawa et al, 2005). The spots of BAC arrays can contain

individual probes more than 1-kb long and each spot can be

composed of a mixture of probes recognising any DNA

fragment contained in up to 120 kb. Thus, they are different to

cDNA-microarrays, where the target DNA of each probe is

usually <1 kb, and obviously, even more diverse than oligo-

arrays. One of the advantages of using a SNP-based platform is

the possibility of detecting LOH with no reduction in CN,

suggestive of partial uniparental disomy, as also recently

reported in acute leukaemias (Raghavan et al, 2005): based

upon our data, in MCL two regions are worthy of further

analysis in 2p and 11p.

We combined gene expression and whole genome profiling

to identify possible therapeutic targets. The discovery of even

individual cases bearing small highly amplified regions

containing one highly expressed gene could highlight pathways

fundamental for the cancer cells. Indeed, both immortalised

cancer cell lines and clinical samples presented small amplified

DNA regions. We looked at genes that map within regions

showing high CNs at DNA level and that were overexpressed.

Single genes were identified, some previously described as

affected in lymphomas such as BCL2 and CMYC. The genes

BCL2 (and its neighbour FVT1), amplified in Granta-519, and

CMYC, amplified in JeKo-1 are often amplified (Rao et al,

1998). They also happened to be among the 12 loci used to

assess the arrayCGH, thus they have been validated by FISH.

Among the transcripts with a high correlation between DNA

and RNA, we identified the gene coding for Syk on 9q22. The

gene was amplified at DNA level, as validated by FISH analysis,

and overexpressed at both RNA and protein levels in the JeKo-

1 MCL cell line. Interestingly, the amplification of the 9q22.2

region containing SYK was due to specific chromosome

rearrangements: in tandem duplication of 9q22.2 region and

subsequently chromosome arm duplication through 9q

isochromosome formation. This same mechanism of amplifi-

cation was also demonstrated for MALT1 and BCL2 in Granta-

519 (data not shown).

Syk is a tyrosine kinase involved in BCR signalling and it

represents the B-cell counterpart of Zap-70 (Cheng et al,

1995; Turner et al, 1995; Cornall et al, 2000; Pogue et al,

2000; Latour & Veillette, 2001; Niiro & Clark, 2002). It is

expressed in all B lymphocytes as well as in B-cell

lymphomas, where it is part of the recently described diffuse

large B-cell lymphoma BCR/proliferation cluster (Pozzobon

et al, 2004; Marafioti et al, 2005; Monti et al, 2005). Our

immunohistochemistry and FISH data showed that MCL

cases can express high levels of Syk and the overexpression

might be given, at least in part, by genomic amplification of

SYK. Based upon these data and because of the availability of

piceatannol, a natural product selective Syk inhibitor (Wieder

et al, 2001), we treated the four MCL cell lines. The JeKo-1

cell line, bearing Syk overexpression, experienced cell

proliferation arrest, growth inhibition and apoptosis at a

dose much lower than the remaining MCL cell lines. In B

cells, Syk leads to intracellular calcium mobilisation, activa-

tion of AKT, mitogen-activated protein kinases and NFjBactivation (Beitz et al, 1999; Cornall et al, 2000; Yokozeki

et al, 2003). The BCR signalling pathway is believed to play a

major role in B-cell lymphoma growth and survival (Kuppers,

2005). A large proportion of lymphomas show constitutive

activation of NFjB, and the cause of the latter phenomenon

is seldom known (Martinez et al, 2003; Perkins, 2004;

Feuerhake et al, 2005). The role played by Syk in the BCR

cascade together with our data of genomic amplification

detected also in clinical samples and of the possibility of

inducing apoptosis with Syk inhibitors in lymphoma cell lines

makes it a very good candidate for further studies in MCL

other lymphoma subtypes. Future studies are needed to

define the piceatannol mechanism of action better, since it is

not yet possible to exclude an effect on other kinases, such as

Zap-70 or Tyk2. Zap-70 is expressed at high levels in a subset

of cases of B-cell chronic lymphocytic leukaemia (B-CLL)

with unmutated immunoglobulin heavy chain genes and it is

associated with poor prognosis (Crespo et al, 2003). Also a

low percentage of B-cell lymphomas, including some MCLs

express Zap-70 (Admirand et al, 2004; Carreras et al, 2005).

Syk in Mantle Cell Lymphoma

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 313

Piceatannol has been previously reported to induce apoptosis

in a Burkitt lymphoma cell line as well as in acute

lymphoblastic leukaemia primary cells at doses of 25–

50 lmol/l (Wieder et al, 2001), higher than the IC50 we

observed in Jeko-1 and similar to doses active in REC and in

Granta-519. As at doses of 50–100 lmol/l piceatannol could

inhibit kinases other than Syk, in particular TYK2, acting as

STAT3 inhibitor (Su & David, 2000; Alas & Bonavida, 2003),

and as in the previous report no mention was made of Syk

status, it is possible that piceatannol might have an anti-

lymphoma effect also through a Syk-independent pathway.

Indeed, our data cannot exclude that the observed effect in

Granta-519 and in REC is not via TYK2 inhibition. However,

the cytotoxic effect in JeKo-1 that has very high Syk levels,

was obtained at doses of piceatannol of <10 lmol/l, reported

to be ineffective on TYK2 (Su & David, 2000).

In conclusion, our MCL data together the very recent

report of overexpression of Syk in splenic marginal zone

B-cell lymphomas (Ruiz-Ballesteros et al, 2005) and the

important role played by Syk-analogue Zap-70 in B-CLL

(Crespo et al, 2003), suggest Syk inhibition as a new

therapeutic strategy to be explored in lymphoid neoplasms.

Syk inhibitors are already under clinical development for

treatment of asthma (Wong & Leong, 2004). Although Syk is

widely expressed, treatment with Syk inhibitors might be

associated with low toxicity because of the existence of Syk-

independent pathways in normal tissues, (Wong et al, 2004),

while lymphoma cells that overexpress Syk might be highly

sensitive to its inhibition.

Acknowledgements

This work was partially supported by the Krebsforschung

Schweiz and the Swiss Group for Clinical Research (SAKK).

G.P. is receiving a fellowship from the San Salvatore Founda-

tion. C.L. is receiving a Fellowship from the Gobierno de

Navarra (Spain). The Authors thank Dr Rita Oldini (Varese,

Italy) for immunohistochemistry (Varese, Italy).

References

Admirand, J.H., Rassidakis, G.Z., Abruzzo, L.V., Valbuena, J.R., Jones,

D. & Medeiros, L.J. (2004) Immunohistochemical detection of ZAP-

70 in 341 cases of non-Hodgkin and Hodgkin lymphoma. Modern

Pathology, 17, 954–961.

Alas, S. & Bonavida, B. (2003) Inhibition of constitutive STAT3 activity

sensitizes resistant non-Hodgkin’s lymphoma and multiple mye-

loma to chemotherapeutic drug-mediated apoptosis. Clinical Cancer

Research, 9, 316–326.

Beitz, L.O., Fruman, D.A., Kurosaki, T., Cantley, L.C. & Scharenberg,

A.M. (1999) SYK is upstream of phosphoinositide 3-kinase in B cell

receptor signaling. Journal of Biological Chemistry, 274, 32662–32666.

Bertoni, F., Zucca, E. & Cavalli, F. (2004a) Mantle cell lymphoma.

Current Opinion in Hematology, 11, 411–418.

Bertoni, F., Zucca, E. & Cotter, F.E. (2004b) Molecular basis of mantle

cell lymphoma. British Journal of Haematology, 124, 130–140.

Bignell, G.R., Huang, J., Greshock, J., Watt, S., Butler, A., West, S.,

Grigorova, M., Jones, K.W., Wei, W., Stratton, M.R., Futreal, P.A.,

Weber, B., Shapero, M.H. & Wooster, R. (2004) High-resolution

analysis of DNA copy number using oligonucleotide microarrays.

Genome Research, 14, 287–295.

Calabrese, G., Fantasia, D., Spadano, A., Morizio, E., Di Bartolomeo, P.

& Palka, G. (2000) Karyotype refinement in five patients with acute

myeloid leukemia using spectral karyotyping. Haematologica, 85,

1219–1221.

Carreras, J., Villamor, N., Colomo, L., Moreno, C., Ramon y Cajal, S.,

Crespo, M., Tort, F., Bosch, F., Lopez-Guillermo, A., Colomer, D.,

Montserrat, E. & Campo, E. (2005) Immunohistochemical analysis

of ZAP-70 expression in B-cell lymphoid neoplasms. Journal of

Pathology, 205, 507–513.

Cheng, A.M., Rowley, B., Pao, W., Hayday, A., Bolen, J.B. & Pawson,

T. (1995) Syk tyrosine kinase required for mouse viability and B-cell

development. Nature, 378, 303–306.

Cornall, R.J., Cheng, A.M., Pawson, T. & Goodnow, C.C. (2000) Role

of Syk in B-cell development and antigen-receptor signaling. Pro-

ceedings of the National Academy of Sciences of the United States of

America, 97, 1713–1718.

Crespo, M., Bosch, F., Villamor, N., Bellosillo, B., Colomer, D., Roz-

man, M., Marce, S., Lopez-Guillermo, A., Campo, E. & Montserrat,

E. (2003) ZAP-70 expression as a surrogate for immunoglobulin-

variable-region mutations in chronic lymphocytic leukemia. New

England Journal of Medicine, 348, 1764–1775.

De Leeuw, R.J., Davies, J.J., Rosenwald, A., Bebb, G., Gascoyne, R.D.,

Dyer, M.J., Staudt, L.M., Martinez-Climent, J.A. & Lam, W.L.

(2004) Comprehensive whole genome array CGH profiling of

mantle cell lymphoma model genomes. Human Molecular Genetics,

13, 1827–1837.

Fernandez, V., Hartmann, E., Ott, G., Campo, E. & Rosenwald, A.

(2005) Pathogenesis of mantle-cell lymphoma: all oncogenic roads

lead to dysregulation of cell cycle and DNA damage response

pathways. Journal of Clinical Oncology, 23, 6364–6369.

Feuerhake, F., Kutok, J.L., Monti, S., Chen, W., Lacasce, A.S.,

Cattoretti, G., Kurtin, P., Pinkus, G.S., de Leval, L., Harris, N.L.,

Savage, K.J., Neuberg, D., Habermann, T.M., Dalla-Favera, R.,

Golub, T.R., Aster, J.C. & Shipp, M.A. (2005) NF{kappa}B activity,

function and target gene signatures in primary mediastinal large

B-cell lymphoma and diffuse large B-cell lymphoma subtypes.

Blood, 106, 1392–1399.

Fisher, R.I. (2005) Mantle cell lymphoma: at last, some hope for suc-

cessful innovative treatment strategies. Journal of Clinical Oncology,

23, 657–658.

Gentleman, R.C., Carey, V.J., Bates, D.M., Bolstad, B., Dettling, M.,

Dudoit, S., Ellis, B., Gautier, L., Ge, Y., Gentry, J., Hornik, K.,

Hothorn, T., Huber, W., Iacus, S., Irizarry, R., Leisch, F., Li, C.,

Maechler, M., Rossini, A.J., Sawitzki, G., Smith, C., Smyth, G.,

Tierney, L., Yang, J.Y. & Zhang, J. (2004) Bioconductor: open

software development for computational biology and bioinfor-

matics. Genome biology, 5, R80.

Hicks, D.G. & Tubbs, R.R. (2005) Assessment of the HER2 status in

breast cancer by fluorescence in situ hybridization: a technical

review with interpretive guidelines. Human Pathology, 36, 250–261.

Huang, J., Wei, W., Zhang, J., Liu, G., Bignell, G.R., Stratton, M.R.,

Futreal, P.A., Wooster, R., Jones, K.W. & Shapero, M.H. (2004)

Whole genome DNA copy number changes identified by high

density oligonucleotide arrays. Human Genomics, 1, 287–299.

A. Rinaldi et al

314 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316

Irizarry, R.A., Hobbs, B., Collin, F., Beazer-Barclay, Y.D., Antonellis,

K.J., Scherf, U. & Speed, T.P. (2003) Exploration, normalization,

and summaries of high density oligonucleotide array probe level

data. Biostatistics, 4, 249–264.

Kennedy, G.C., Matsuzaki, H., Dong, S., Liu, W.M., Huang, J., Liu, G.,

Su, X., Cao, M., Chen, W., Zhang, J., Liu, W., Yang, G., Di, X.,

Ryder, T., He, Z., Surti, U., Phillips, M.S., Boyce-Jacino, M.T.,

Fodor, S.P. & Jones, K.W. (2003) Large-scale genotyping of complex

DNA. Nature Biotechnology, 21, 1233–1237.

Kuppers, R. (2005) Mechanisms of B-cell lymphoma pathogenesis.

Nature Reviews Cancer, 5, 251–262.

Lacrima, K., Valentini, A., Lambertini, C., Taborelli, M., Rinaldi, A.,

Zucca, E., Catapano, C., Cavalli, F., Gianella-Borradori, A., Mac-

callum, D.E. & Bertoni, F. (2005) In vitro activity of cyclin-depen-

dent kinase inhibitor CYC202 (Seliciclib, R-roscovitine) in mantle

cell lymphomas. Annals of Oncology, 16, 1169–1176.

Latour, S. & Veillette, A. (2001) Proximal protein tyrosine kinases in

immunoreceptor signaling. Current Opinion in Immunology, 13,

299–306.

Marafioti, T., Pozzobon, M., Hansmann, M.L., Gaulard, P., Barth, T.F.,

Copie-Bergman, C., Roberton, H., Ventura, R., Martin-Subero, J.I.,

Gascoyne, R.D., Pileri, S.A., Siebert, R., Hsi, E.D., Natkunam, Y.,

Moller, P. & Mason, D.Y. (2005) Expression pattern of intracellular

leukocyte-associated proteins in primary mediastinal B cell lym-

phoma. Leukemia, 19, 856–861.

Marshall, E. (2004) Getting the noise out of gene arrays. Science, 306,

630–631.

Martinez, N., Camacho, F.I., Algara, P., Rodriguez, A., Dopazo, A.,

Ruiz-Ballesteros, E., Martin, P., Martinez-Climent, J.A., Garcia-

Conde, J., Menarguez, J., Solano, F., Mollejo, M. & Piris, M.A.

(2003) The molecular signature of mantle cell lymphoma reveals

multiple signals favoring cell survival. Cancer Research, 63,

8226–8232.

Matsuzaki, H., Loi, H., Dong, S., Tsai, Y.Y., Fang, J., Law, J., Di, X.,

Liu, W.M., Yang, G., Liu, G., Huang, J., Kennedy, G.C., Ryder, T.B.,

Marcus, G.A., Walsh, P.S., Shriver, M.D., Puck, J.M., Jones, K.W. &

Mei, R. (2004) Parallel genotyping of over 10 000 SNPs using a one-

primer assay on a high-density oligonucleotide array. Genome

Research, 14, 414–425.

Mitelman, F. (ed.) (1995) ISCN (1995): An International System for

Human Cytogenetic Nomenclature. S. Karger, Basel.

Monni, O., Barlund, M., Mousses, S., Kononen, J., Sauter, G.,

Heiskanen, M., Paavola, P., Avela, K., Chen, Y., Bittner, M.L. &

Kallioniemi, A. (2001) Comprehensive copy number and gene

expression profiling of the 17q23 amplicon in human breast cancer.

Proceedings of the National Academy of Sciences of the United States of

America, 98, 5711–5716.

Monti, S., Savage, K.J., Kutok, J.L., Feuerhake, F., Kurtin, P., Mihm,

M., Wu, B., Pasqualucci, L., Neuberg, D., Aguiar, R.C., Dal Cin, P.,

Ladd, C., Pinkus, G.S., Salles, G., Harris, N.L., Dalla-Favera, R.,

Habermann, T.M., Aster, J.C., Golub, T.R. & Shipp, M.A. (2005)

Molecular profiling of diffuse large B-cell lymphoma identifies

robust subtypes including one characterized by host inflammatory

response. Blood, 105, 1851–1861.

Niiro, H. & Clark, E.A. (2002) Regulation of B-cell fate by antigen-

receptor signals. Nature Reviews Immunology, 2, 945–956.

Olshen, A.B., Venkatraman, E.S., Lucito, R. & Wigler, M. (2004)

Circular binary segmentation for the analysis of array-based DNA

copy number data. Biostatistics, 5, 557–572.

Perkins, N.D. (2004) NF-kappaB: tumor promoter or suppressor?

Trends Cell Biology, 14, 64–69.

Pham, L.V., Tamayo, A.T., Yoshimura, L.C., Lo, P. & Ford, R.J. (2003)

Inhibition of constitutive NF-kappaB activation in mantle cell

lymphoma B cells leads to induction of cell cycle arrest and apop-

tosis. Journal of Immunology, 171, 88–95.

Pogue, S.L., Kurosaki, T., Bolen, J. & Herbst, R. (2000) B cell

antigen receptor-induced activation of Akt promotes B cell sur-

vival and is dependent on Syk kinase. Journal of Immunology, 165,

1300–1306.

Pollack, J.R., Perou, C.M., Alizadeh, A.A., Eisen, M.B., Per-

gamenschikov, A., Williams, C.F., Jeffrey, S.S., Botstein, D. & Brown,

P.O. (1999) Genome-wide analysis of DNA copy-number changes

using cDNA microarrays. Nature Genetics, 23, 41–46.

Pozzobon, M., Marafioti, T., Hansmann, M.L., Natkunam, Y. &

Mason, D.Y. (2004) Intracellular signalling molecules as im-

munohistochemical markers of normal and neoplastic human leu-

cocytes in routine biopsy samples. British Journal of Haematology,

124, 519–533.

Raghavan, M., Lillington, D.M., Skoulakis, S., Debernardi, S., Chaplin,

T., Foot, N.J., Lister, T.A. & Young, B.D. (2005) Genome-wide single

nucleotide polymorphism analysis reveals frequent partial uni-

parental disomy due to somatic recombination in acute myeloid

leukemias. Cancer Research, 65, 375–378.

Rao, P.H., Houldsworth, J., Dyomina, K., Parsa, N.Z., Cigudosa, J.C.,

Louie, D.C., Popplewell, L., Offit, K., Jhanwar, S.C. & Chaganti, R.S.

(1998) Chromosomal and gene amplification in diffuse large B-cell

lymphoma. Blood, 92, 234–240.

Rosenwald, A., Wright, G., Wiestner, A., Chan, W.C., Connors, J.M.,

Campo, E., Gascoyne, R.D., Grogan, T.M., Muller-Hermelink, H.K.,

Smeland, E.B., Chiorazzi, M., Giltnane, J.M., Hurt, E.M., Zhao, H.,

Averett, L., Henrickson, S., Yang, L., Powell, J., Wilson, W.H., Jaffe,

E.S., Simon, R., Klausner, R.D., Montserrat, E., Bosch, F., Greiner,

T.C., Weisenburger, D.D., Sanger, W.G., Dave, B.J., Lynch, J.C.,

Vose, J., Armitage, J.O., Fisher, R.I., Miller, T.P., LeBlanc, M., Ott,

G., Kvaloy, S., Holte, H., Delabie, J. & Staudt, L.M. (2003) The

proliferation gene expression signature is a quantitative integrator of

oncogenic events that predicts survival in mantle cell lymphoma.

Cancer Cell, 3, 185–197.

Rubio-Moscardo, F., Climent, J., Siebert, R., Piris, M.A.,Martin-Subero,

J.I., Nielander, I., Garcia-Conde, J., Dyer, M.J., Terol, M.J., Pinkel, D.

& Martinez-Climent, J.A. (2005) Mantle-cell lymphoma genotypes

identified with CGH to BACmicroarrays define a leukemic subgroup

of disease and predict patient outcome. Blood, 105, 4445–4454.

Ruiz-Ballesteros, E., Mollejo, M., Rodriguez, A., Camacho, F.I., Algara,

P., Martinez, N., Pollan, M., Sanchez-Aguilera, A., Menarguez, J.,

Campo, E., Martinez, P., Mateo, M. & Piris, M.A. (2005) Splenic

marginal zone lymphoma: proposal of new diagnostic and prog-

nostic markers identified after tissue and cDNA microarray analysis.

Blood, 106, 1831–1838.

Schraders, M., Pfundt, R., Straatman, H.M., Janssen, I.M., van Kessel,

A.G., Schoenmakers, E.F., van Krieken, J.H. & Groenen, P.J. (2005)

Novel chromosomal imbalances in mantle cell lymphoma detected

by genome-wide array-based comparative genomic hybridization.

Blood, 105, 1686–1693.

Su, L. & David, M. (2000) Distinct mechanisms of STAT phosphor-

ylation via the interferon-alpha/beta receptor. Selective inhibition of

STAT3 and STAT5 by piceatannol. Journal of Biological Chemistry,

275, 12661–12666.

Syk in Mantle Cell Lymphoma

ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316 315

Tagawa, H., Karnan, S., Suzuki, R., Matsuo, K., Zhang, X., Ota, A.,

Morishima, Y., Nakamura, S. & Seto, M. (2005) Genome-wide

array-based CGH for mantle cell lymphoma: identification of

homozygous deletions of the proapoptotic gene BIM. Oncogene, 24,

1348–1358.

The Non-Hodgkin’s Lymphoma Classification Project (1997) A clinical

evaluation of the International Lymphoma Study Group classifica-

tion of non-Hodgkin’s lymphoma. Blood, 89, 3909–3918.

Tibiletti, M.G., Bernasconi, B., Furlan, D., Riva, C., Trubia, M.,

Buraggi, G., Franchi, M., Bolis, P., Mariani, A., Frigerio, L., Capella,

C. & Taramelli, R. (1996) Early involvement of 6q in surface epi-

thelial ovarian tumors. Cancer Research, 56, 4493–4498.

Tracey, L., Villuendas, R., Dotor, A.M., Spiteri, I., Ortiz, P., Garcia, J.F.,

Peralto, J.L., Lawler, M. & Piris, M.A. (2003) Mycosis fungoides

shows concurrent deregulation of multiple genes involved in the

TNF signaling pathway: an expression profile study. Blood, 102,

1042–1050.

Troester, M.A., Hoadley, K.A., Sorlie, T., Herbert, B.S., Borresen-Dale,

A.L., Lonning, P.E., Shay, J.W., Kaufmann, W.K. & Perou, C.M.

(2004) Cell-type-specific responses to chemotherapeutics in breast

cancer. Cancer Research, 64, 4218–4226.

Turner, M., Mee, P.J., Costello, P.S., Williams, O., Price, A.A., Duddy,

L.P., Furlong, M.T., Geahlen, R.L. & Tybulewicz, V.L. (1995) Peri-

natal lethality and blocked B-cell development in mice lacking the

tyrosine kinase Syk. Nature, 378, 298–302.

Wieder, T., Prokop, A., Bagci, B., Essmann, F., Bernicke, D., Schulze-

Osthoff, K., Dorken, B., Schmalz, H.G., Daniel, P.T. & Henze, G.

(2001) Piceatannol, a hydroxylated analog of the chemopreventive

agent resveratrol, is a potent inducer of apoptosis in the lymphoma

cell line BJAB and in primary, leukemic lymphoblasts. Leukemia, 15,

1735–1742.

Wong, W.S. & Leong, K.P. (2004) Tyrosine kinase inhibitors: a new

approach for asthma. Biochimica et Biophysica Acta, 1697, 53–69.

Wong, B.R., Grossbard, E.B., Payan, D.G. & Masuda, E.S. (2004)

Targeting Syk as a treatment for allergic and autoimmune disorders.

Expert Opinion on Investigational Drugs, 13, 743–762.

Yokozeki, T., Adler, K., Lankar, D. & Bonnerot, C. (2003) B cell

receptor-mediated Syk-independent activation of phosphatidyl-

inositol 3-kinase, Ras, and mitogen-activated protein kinase path-

ways. Journal of Immunology, 171, 1328–1335.

Zhang, J., Billingsley, M.L., Kincaid, R.L. & Siraganian, R.P. (2000)

Phosphorylation of Syk activation loop tyrosines is essential for Syk

function. An in vivo study using a specific anti-Syk activation loop

phosphotyrosine antibody. Journal of Biological Chemistry, 275,

35442–35447.

Zhao, X., Li, C., Paez, J.G., Chin, K., Janne, P.A., Chen, T.H., Girard,

L., Minna, J., Christiani, D., Leo, C., Gray, J.W., Sellers, W.R. &

Meyerson, M. (2004) An integrated view of copy number and allelic

alterations in the cancer genome using single nucleotide poly-

morphism arrays. Cancer Research, 64, 3060–3071.

Supplementary Material

The following material is available for this article online:

Table S1. Comparison between FISH and arrayCGH.

Figure S1. Karyotype of the four MCL cell lines, as defined

by conventional QFQ banding and SKY-FISH.

Figure S2. Whole genome profiling of JeKo-1 MCL cell line:

(A) raw copy number estimate as obtained with the Affymetrix

Genechip Human Mappping Xba 131; x-axis, physical map-

ping; y-axis, CNAT estimated copy number. (B) LOH as

obtained with the Affymetrix Genechip Human Mappping Xba

131; x-axis, physical mapping; y-axis, CNAT derived LOH

probability. (C) Piece-wise constant estimate of the copy

number profiling as obtained with the Affymetrix Genechip

Human Mappping Xba 131; x-axis, physical mapping; y-axis,

piece-wise constant estimate of the copy number. (D) Whole

genome profiling as obtained with the cDNA CNIO Onco-

Chip; x-axis, physical mapping; y-axis, log 2 ratios between

tumour and normal reference DNA.

Figure S3. SYK involvement in Jeko-1 cell: DNA and

RNA (for both panels: squares, JeKo-1; dots, Granta-519; x,

REC; triangle, NCEB-1. (A) Raw copy number estimate as

obtained with the Affymetrix Genechip Human Mappping

Xba 131 for SNPprobes mapped in the 88-128 Mb segment

of 9q22; x-axis, physical mapping; y-axis, CNAT estimated

copy number. (B) Piece-wise constant estimate of the copy

number for SNPprobes mapped in the 88- to 128-Mb

segment of 9q22; x-axis, physical mapping; y-axis, piece-wise

constant estimate of the copy number. (C) RMA gene

expression values for Affymetrix probes targeting genes

mapped in the 88- to 128-Mb segment of 9q22 and showing

a significant correlation between CN and gene expression

(no such genes were present the 106–128 Mb); SYK probes

are highlighted; y-axis, RMA values.

The material is available as part of the online article from

http://www.blackwell-synergy.com.

A. Rinaldi et al

316 ª 2005 Blackwell Publishing Ltd, British Journal of Haematology, 132, 303–316