Mol Diagn 2003; 7 (3): 163-167ORIGINAL RESEARCH ARTICLE 1084-8592/03/0003-0163/$30.00/0

© 2003 Adis Data Information BV. All rights reserved.

Epstein-Barr Virus Transformation of HumanLymphoblastoid Cells from Patients with FragileX Syndrome Induces Variable Changes on CGGRepeats Size and Promoter MethylationVictoria Bonilla, Francisco Sobrino, Miguel Lucas and Elizabeth Pintado

Department of Biochemical Medicine and Molecular Biology, Faculty of Medicine and University Hospital of the VirginMacarena, University of Seville, Seville, Spain

Background: Our understanding of fragile X syndrome can be improved by reversing the expression of theAbstractsilenced fragile X mental retardation 1 (FMR1) gene in immortalized cells from these patients. Epstein-Barrvirus (EBV) infection has been extensively used to transform B cells into a permanent lymphoblastoid cell line.Methods: We immortalized B lymphocytes from three different fragile X patients and one normal male. Weanalyzed the CGG triplet repeats and methylation status of the FMR1 and interferon (IFN)-γ promoter. We alsoassayed FMR1 mRNA levels by real-time PCR and FMR1 protein (FMRP) by Western blot.Results: We observed that EBV transformation may induce the instability of CGG repeats and DNA demethyla-tion that can lead to the modification of mRNA expression.Conclusions: EBV transformation may induce variable changes in the genome that can lead to the misinterpreta-tions of experimental data obtained from these cells. Thus, periodic testing of DNA from immortalized cellsshould be routinely undertaken to detect undesired effects.

Fragile X syndrome is one of the most common forms of extensively used in both human populations and medical geneticsinherited mental retardation. The molecular basis of fragile X to dispose of a permanent source of cells and DNA.[12-15] For thesesyndrome is an expansion of the CGG triplet repeats located purposes, we immortalized B cells from three fragile X patientswithin the 5′ untranslated region (UTR) region of the fragile X and a normal individual by using the EBV infection. We observedmental retardation 1 (FMR1) gene, resulting in the absence of the that the transformation of fragile X cells with EBV may induceencoded protein FMR1 protein (FMRP), which is a ribosome-as- CGG changes and demethylation of the FMR1 promoter increas-sociated RNA-binding protein.[1-5] The presence of large expan- ing mRNA expression. Therefore, periodic DNA testing fromsion (n > 200) is generally associated with abnormal methylation immortalized cells helps in our understanding of the effects ofof the surrounding DNA and the suppression of FMR1 expression EBV on the genome. In addition, the study of the pathway byand translation.[6,7] Rare individuals of normal intelligence have which EBV induces demethylation may give some clue to thebeen shown to carry a completely or partially unmethylated full reactivation of FMR1 in fragile X patients.mutation and to express FMRP. The variable manifestation of the

Material and Methodsdisease in female carriers of a full mutation is explained by thedifferential X inactivation.[8,9]

The Epstein-Barr virus (EBV) is a human herpes virus that Cell Cultureinteracts with various immunocompetent cells that carry the EBVreceptor and can transform, in vitro, human B cells into continu- Lymphoblastoid cell lines (LCL) were established by EBVously growing lymphoid cell lines.[10,11] This method has been transformation of peripheral blood lymphocytes from three male

164 Bonilla et al.

fragile X patients and one normal male using a standard procedure. transcribed using random hexamers, and the cDNA was amplifiedThis research was performed with protocols approved by the local using specific primers as described by Pietrobono et al.[15] Theinstitutional review board and with the informed consent of the primers used for FMR1 expression were: forward, 5′-GGA ACAparticipants and/or their guardians. AAG GAC AGC ATC GC-3′; reverse, 5′-CTC TCC AAA CGC

AAC TGG TCT-3′, given an 89 bp band (Roche DiagnosticsBriefly, lymphoid cells were obtained from 20mL of freshCorporation Biolabs, Barcelona, Spain). The expression of theblood by Ficoll-Paque density gradient (Amersham Biosciences,guanine hypoxantine phosphorribosyl transferase (HPRT) wasUppsala, Sweden). Cells were washed twice in 10mL of standardused as an internal control. PCR products were visualized onmedium RPMI (BioWhittaker Europe, Verviers, Belgium), 10%agarose gel stained with ethidium bromide.inactivated calf fetal serum (Gibco Brl, Life Technologies BV,

Barcelona, Spain), 2mM glutamine plus antibiotics (Sigma, Ma-Real-Time PCRdrid, Spain), suspended in 4mL EBV, and placed in inclined tubes

for 2 hours in a 5% CO2 incubator. Following this, the cells wereReal-time PCR was performed in an ABI Prism® 7000 Se-centrifugated and incubated in tissue culture flasks in a vertical

quence Detection System (Applied Biosystems, Foster City, CA,position with 10mL RPMI plus 20% calf fetal serum, 2mMUSA) using SYBR® Green PCR Master mix (Applied Biosys-glutamine, antibiotics and 1 mg/L ciclosporine (Sandoz Ltd,tems, Foster City, CA, USA) and the thermocycler conditionsSchoenenwerd, Switzerland). Every 7 days the medium was re-recommended by the manufacturer. PCRs were performed inmoved and replaced with fresh medium without centrifugation.duplicates in a total volume of 25μL containing 2–3μL of theAfter 21–25 days the cells showed conglomerations and couldreverse transcriptase reaction. Each sample was analyzed foreither be frozen or used in the experiments. For these experiments,cyclophylin to normalize for RNA input amounts and to performthe standard medium was changed every 48 hours, cells wererelative quantifications. Primers were designed using the computercounted, split, and seeded at the initial concentration of 1–2 × 106program Primer Express® (Applied Biosystems, Foster City, CA,cells/mL and the cells were harvested and used for DNA, RNA, orUSA) and for FMR1 were the same as those used in conventionalprotein extraction.reverse transcription (RT)-PCR. For cyclophylin the amplicongives a 74 bp and the primers (Roche Diagnostics Corporation,Southern BlottingBarcelona, Spain) were: forward 5′-GCA CTG GTG GCA AGT

DNA samples were prepared from fresh peripheral lympho- CCA T-3′ and reverse 5′-GCC AGG ACC TGT ATG CTTcytes and LCL by standard procedures. 10μg of genomic DNA ACG-3′. Melting curve analysis showed a single sharp peak withwas digested overnight with the restriction enzymes EcoRI-EagI the expected melting temperature (Tm) for all samples.[17]

(New England Biolabs, Hitchin, England), this last one beingsensitive to methylation. Restriction fragments were separated by Western Blottingelectrophoresis on 0.8% agarose gel, Southern blotted, and hybrid-

Cells were homogenized in cold lysis hypotonic buffer contain-ized with radiolabeled StB12.3 probe as previously described.[16]

ing 20mM Hepes pH 7.9, 10mM KCl, 10% glycerol, 1mM EDTA,Methylation Rate of Interferon-γ Promoter by PCR 1mM DTT, 0.2% NP40, protease inhibitors (0.01% PMSF, 1mM

Na3VO4, 10 mg/L aprotinin, 10 mg/L leupeptin), and 1.5 mg/LGenomic DNA (2.5μg) from fresh peripheral lymphocytes and

RNase-DNase free for 10 minutes at 4°C. The homogenate wasLCL from patient P3 was digested by the methylation sensitive

centrifuged 5 minutes at 13 000g, the supernatant recollected andrestriction enzyme SnaBI at 37°C overnight. The primers were

centrifuged 60 minutes at 100 000g. The pellet was suspended indesigned to contain the conserved CpG target for methylation

hypertonic buffer containing 20mM Hepes pH 7.9, 10mM KCl,(recognition site TACGTA) at position -54. 5′-GAC CCA AGG

420mM NaCl, 20% glycerol, 1mM EDTA, 1mM DTT, 0.2%AGT CTA AAG GAA ACT CTA ACT-3′ and 5′-CTG ATC TTC

NP40 and the same protease inhibitors as described with minorAGA TGA TCA GAA CAA TGT GCT-3′. The 20μL of PCR

modifications.[18] The protein concentration was determined withproducts were loaded on 1% agarose gel, and the results analyzed

Bradford protein assay reagent. 20–40μg of the lysate was loadedby densitometry.

and resolved on 7.5% SDS-PAGE transferred ontopoly(vinylidene difluoride) [PVDF] membrane (Amersham Bios-

Reverse Transcription PCRciences, Uppsala, Sweden) and subjected to immunodetection

Total RNA was extracted from LCL and fresh peripheral lym- using a 1 : 10 000 dilution of primary antibody (Chemicon Inter-phocytes by the guanidine phenol method. RNA was reverse national, Temecula, CA, USA) and enhanced chemiluminescence

© 2003 Adis Data Information BV. All rights reserved. Mol Diagn 2003; 7 (3)

DNA Changes in EBV Immortalized Fragile X Cells 165

detection system (Amersham Biosciences, Uppsala, Sweden). Allother cited reagents were of the best quality commercially avail-able.

Results

Patients P1, P2 and P3 were mentally retarded boys with IQs of35, 40, and 51, respectively. They were diagnosed with the fragileX syndrome by Southern blotting in our laboratory.[19] To have apermanent source of cells and DNA we immortalized B cells byEBV infection from these three patients and a male without fragileX sydrome. Figure 1 shows the CGG size and the FMR1 CpGisland methylation status from fresh lymphocytes and EBV trans-formed lymphoblastoid cells obtained from the three patients.Lines 1, 3, and 5 are from patients P1, P2 and P3, respectively,before immortalization and lines 2, 4, and 6 after EBV transforma-tion. After immortalization patient P1 shows an increase of about800 bp in the CGG repeats and patient P2 experienced a smalldecrease of about 100 bp. Patient P3 had an expanded and methy-lated band of 5.8kb (600 bp expansion) and a non-methylated bandof about 2.9kb. In this case, after immortalization the methylatedallele almost disappeared and a non-methylated band with thesame expansion (600 bp) came into sight.

To further demonstrate that EBV transformation may inducegenomic demethylation we analyzed the methylation status of theinterferon (IFN)-γ promoter which has a highly conserved CpGtarget for methylation at a position critical for promoter func-tion.[20] This analysis was performed through the use of the meth-ylation-sensitive restriction endonuclease SnaBI. The digested

1.2

1.0

0.8

0.6

0.4

0.2pI

FN

γ m

RN

A/G

AP

DH

mR

NA

fL iL

P3

fL iL

P3

pIFNγ

GAPDH

204

600

bp

1 2 3 4

b

a

Fig. 2. PCR analysis of the degree of CpG methylation in the interferon(IFN)-γ promoter region. Genomic DNA of fresh (fL) and immortalized (iL)cells from patient (P) 3 was digested with excess of SnaBI at 37°C over-night. 5μL of digestion mix was used for detection of CpG methylationstatus by PCR. The levels of methylation were determine on 1% agarosegel and analyzed by densitometry. (a) Lines 1 and 2 represent digestedDNA from patient P3 before immortalization and lines 3 and 4 after Ep-stein-Barr virus (EBV) transformation. Glyceraldehyde-3-phosphate dehy-drogenase (GAPDH) from nondigested DNA was used as a control. (b)The relative mRNA expression of plasmid IFNγ (pIFNγ) and GAPDH quan-tified by integrating densitometry.

DNA, before and after immortalization, was used to measure the

degree of methylation. As shown in figure 2 (a and b), the SnaBI

site in fresh lymphocytes from patients P3 was highly methylated

compared with LCL from this patient. PCR from nondigested

DNA was used as control.

FMR1 mRNA expression studied by conventional and real-

time RT-PCR shows that only patient P3 expressed a detectable

amount of FMR1 mRNA either in fresh or immortalized cells

(figure 3a). Real-time PCR showed that immortalized lympho-

cytes from patient P3 express four times more FMR1 mRNA than

fresh lymphocytes from this patient (figure 3b), although Western

blot from immortalized cells shows that patient P3 expressed the

80kD FMRP at lower levels than the control sample (figure 4).

5.2

2.8

1 2 3 4 5 6

KbfL iL fL iL fL iL

P1 P2 P3

Fig. 1. Restriction pattern of fresh and Epstein-Barr virus (EBV) immortal-ized cells from patients with fragile X syndrome. Southern blot of genomicDNA digested to completion with EcoRI-EagI (New England Biolabs, Hitch-in, England). Non-fragile X individuals will show a 2.8kb fragments (males)or a 2.8kb and 5.2kb fragments (females). Lines 1, 3, and 5 correspondedto fresh lymphocytes from patients (P) 1, 2, and 3, respectively. Lines 2, 4and 6 are EBV immortalized cells from the same patients. fL = fresh cells;iL = immortalized cells.

© 2003 Adis Data Information BV. All rights reserved. Mol Diagn 2003; 7 (3)

166 Bonilla et al.

any of the several achieved blotting. It has recently been observedthat cell lines from transgenic mice generated by Simian virus-40(SV40) infection show dramatic CGG repeats instability. This is apostzygotic mechanism induced by the inclusion of the origin ofreplication of SV40 in the plasmid constructed for injection.[23]

Therefore, the possibility that some viruses may have a directaction on the genome should not be underestimated.

It has been reported that EBV-transformed human lymphocyteskept in culture underwent early, strong overall demethylation ofthe genome.[24] In our immortalized cells we observed demethyla-tion of the FMR1 promoter only in patient P3, the sample with thesmallest CGG expansion. The unmethylated band observed inimmortalized cells seems to correspond to the methylated allele infresh lymphocytes since the size of expansion was exactly thesame in both cases. It is very unlikely that only selection eventswere responsible for the observed band. It is more feasible that thedemethylation occurred after selection of a population of cellswith the full mutation. Higher expansions at the FMR1 promoter,as seen in patients P1 and P2, may be more resistant to demethyla-tion than the smallest ones. Since the reported demethylationinduced by EBV infection is spread on the genome[24] we tested, asan example, the methylation status of the human IFNγ promoter.The marked decrease in methylation observed after immortaliza-tion supports the view of a general and, probably random, processof demethylation induced by EBV transformation.

1.0

0.8

0.6

0.4

0.2

0

Fol

d ch

ange

fL iL fL iL

N P3

FMR1

HPRT

1 2 3 4 5 6 7

fL iL fL iL fL iL iL

89

700

bp

N P1 P3 P2a

b

Fig. 3. Fragile X mental retardation 1 (FMR1) expression in patients withfragile X syndrome assessed by reverse transcription (RT)-PCR. (a)Agarose gel separation of RT-PCR products obtained from total RNA fromfresh (fL) and immortalized lymphocytes (iL). FMR1 gives an 89 bp bandand hypoxantine phosphorribosyl transferase (HPRT) a 700 bp. Lines 1, 3,and 5 correspond to fL from a healthy individual (N), patients (P) 1 and 3,respectively. Lines 2, 4, 6, and 7 are from iL cells from a healthy individual,P1, P3, and P2, respectively. (b) Real-time PCR from a healthy individualand P3 either from fL or iL. Average-fold change values ± SE from ahealthy individual and P3 for 2 separate RT-PCR are presented.

FMR1 mRNA analyzed by real-time PCR indicates that theimmortalization process increased FMR1 expression only whenDiscussionthe EBV induced demethylation. It has been reported that premu-tated carriers show special phenotype and express more FMR1EBV transformation is a common procedure used to maintain amRNA than control individuals.[25-28] However, fresh cells frompermanent source of cells and DNA, thus it is important to knowpatient P3 expressed 10 times less FMR1 mRNA than cells from ahow the virus itself may affect the genome of the immortalizednormal individual, which is in accordance with the IQ (51) observ-cells. We have observed that Southern blot analysis of the FMR1ed in this patient. After immortalization, cells from patient P3gene in human lymphocytes gives different patterns before andacted as a carrier by showing reduced FMRP levels despite signifi-after immortalization. It is possible that the observed changes incantly elevated FMR1 mRNA levels.FMR1 occurred between the time of diagnosis (several years

earlier) and the transformation of samples used in this study. ThisConclusionpossibility was, however, easily rejected because no alterations

were seen after comparisons with fresh cells from the same sampleOur study shows that EBV transformation has a variable effect

used for immortalization. Thus, EBV transformation induced theon the FMR1 gene in patients with fragile X syndrome. Since

changes in the FMR1 gene described here. Our observations are inagreement with the generally accepted opinion that in differentiat-ed cells, after the period of expansion in early embryogenesis,expanded CGG repeats of the FMR1 gene show strong somaticstability.[21,22]

Although the apparent instability detected by Southern blotmay arise from subcloning events which are known to occur incontinuous culturing of a cell population, it is surprising that thealleles of selected cells were not visible before immortalization in

FMRP

P1 P2 P3 N kD

80

Fig. 4. FMRP expression in Epstein-Barr virus (EBV) immortalized cellsfrom patients with fragile X syndrome. Western blot of FMRP in EBVimmortalized cells from patients (P) 1, 2, 3, and a normal male (N).

© 2003 Adis Data Information BV. All rights reserved. Mol Diagn 2003; 7 (3)

DNA Changes in EBV Immortalized Fragile X Cells 167

13. Chiurazzi P, Pomponi MG, Pietrobono R, et al. Synergistic effect of histoneimmortalization of B cells by EBV infection is a common proce-hyperacetylation and DNA demethylation in the reactivation of the FMR1 gene.

dure in maintaining a permanent source of cells and is extensively Hum Mol Genet 1999; 8: 2317-23

used to reactivate FMR1 expression these observations should be 14. Coffee B, Zhang F, Warren ST, et al. Acetylated histones are associated withFMR1 in normal but not fragile X-syndrome cells. Nat Genet 1999; 22: 98-101taken into consideration when working with EBV transformed

cells. 15. Pietrobono R, Pomponi MG, Tabolacci E, et al. Quantitative analysis of DNAdemethylation and transcriptional reactication of the FMR1 gene in fragile Xcells treated with 5-azadeoxyxytidine. Nucleic Acids Res 2002; 30: 3278-85

16. Hallmayer J, Pintado E, Lotspeich L, et al. Molecular analysis and test of linkageAcknowledgementsbetween the FMR-1 gene and infantile autism in multiplex families. Am J HumGenet 1994; 55: 951-9

We are very grateful to Dr R. Fernandez-Munoz (Laboratorio de Viro- 17. Del Toro R, Levitsky KL, Lopez-Barneo J, et al. Induction of T-Type calciumlogia, Hospital Ramon y Cajal, Madrid) for the EBV aliquots used in this study channel gene expression by chronic hypoxia. J Biol Chem 2003; 278: 22316-24

and for his useful advice on how to manipulate them. We thank Dr J.L. Mandel 18. Hmadcha A, Bedoya FJ, Sobrino F, et al. Methylation-dependent gene silencing(INSERM, Strasburg, France) for the StB12.3 probe. We also wish to thank to induced by interleukin 1β via nitric oxide production. J Exp Med 1999; 11:Dr M. Carrasco (Ministry of Education of the Andelusian Government, 1595-603

Huelva, Spain) for the IQ data. We would like to express our gratitude to Dr J. 19. De Diego Y, Hmadcha A, Moron F, et al. Fragile X founder effect and distributionLopez-Barneo (LIB, Seville) for his advise in the writing of the manuscript of CGG repeats among the mentally retarded population of Andalusia, Southand for his generosity in making the ABI Prism 7000 available. Spain. Genet Mol Biol 2002; 25: 1-6

This work was supported by grants 01/1132 from the Ministerio de 20. Xiaofang CH, Lopez DM. CD4+, but not CD8+, T cells from mammary tumor-Sanidad y Consumo and 80/01 from the Servicio Andaluz de Salud, Spain. bearing mice have a down-regulated production of IFN-γ: role of phosphatidyl

serine. J Immunol 1998; 160: 2735-41

21. Tassone F, Hagerman RJ, Gane LW, et al. Strong similarities of the FMR1mutation in multiple tissues: postmortem studies of male with a full mutation

References and a male carrier of a premutation. Am J Med Genet 1999; 84: 240-41. Fu YH, Kuhl DP, Pizzuti A, et al. Variation of the CGG repeat at the fragile site

22. Burman RW, Popovich BW, Jacky PB, et al. Fully expanded FMR1 CGG repeatsresults in genetic instability: resolution of the Sherman paradox. Cell 1991; 67:exhibit a length-and differentiation-dependent instability in cell hybrids that is1047-58independent of DNA methylation. Hum Mol Genet 1999; 8: 2293-302

2. Oberle I, Rousseau F, Heitz D, et al. Instability of a 550-base pair DNA segmentand abnormal methylation in fragile X syndrome. Science 1991; 252: 1097-102 23. Baskaran S, Datta S, Mandal A, et al. Instability of CGG repeats in transgenic

mice. Genomics 2002; 80: 151-73. Siomi H, Siomi MC, Nussbaum RL, et al. The product of the fragile X gene,FMR1, has characteristics of an RNA-binding protein. Cell 1993; 74: 291-8 24. Vilain A, Bernardino J, Gerbault-Seureau M, et al. DNA methylation and chromo-

4. O’Donnell WT, Warren ST. A decade of molecular studies of fragile X syndrome. some instability in lymphoblastoid cell lines. Cytogenet Cell Genet 2000; 90:Annu Rev Neurosci 2002; 25: 315-38 93-101

5. Oostra BA, Willemsen R. A fragile balance: FMR1 expression levels. Hum Mol25. Pintado E, De Diego Y, Hmadcha A, et al. Instability of the CGG repeat at the

Genet 2003 Oct 15; 12 Spec No. 2: R249-57FRAXA locus and variable phenotypic expression in a large fragile X pedigree.

6. Pieretti M, Zhang F, Fu YH, et al. Absence of expression of the FMR-1 gene in J Med Genet 1995; 32: 907-9fragile X syndrome. Cell 1991; 66: 817-22

26. Tassone F, Hagerman RJ, Taylor AK, et al. Elevated levels of FMR1 mRNA in7. Hmadcha A, De Diego Y, Pintado E. Assessment of FMR1 expression by reversecarrier males: a new mechanism of involvement in the fragile-x syndrome. Amtranscriptase-polymerase chain reaction of KH domains. J Lab Clin Med 1998;J Hum Genet 2000; 66: 6-15131: 170-3

8. Hagerman RJ, Hull CE, Safanda JF, et al. High functioning fragile X males: 27. Tassone F, Hagerman RJ, Chamberlain WD, et al. Transcription of the FMR1 genedemonstration of an unmethylated fully expanded FMR1 mutation associated in individuals with fragile X syndrome. Am J Med Genet 2000; 97: 195-203with protein expression. Am J Med Genet 1994; 51: 298-308

28. Kenneson A, Zhang F, Hagedorn CH, et al. Reduced FMRP and increased FMR19. Smeets HJ, Smits AP, Verheij CE, et al. Normal phenotype in two brothers with a

transcription is proportionally associated with CGG repeat number in inter-full FMR1 mutation. Hum Mol Genet 1995; 4: 2103-8

mediate-length and premutation carriers. Hum Mol Genet 2001; 10: 1449-5410. Cambier JC, Pleiman CM, Clash MR. Signal transduction by B-cell antigen

receptor and co-receptor. Annu Rev Immunol 1994; 12: 457-86

11. D′Addario M, Ahmad A, Morgan A, et al. Binding of the Epstein-Barr virus major Correspondence and offprints: Dr Elizabeth Pintado, Departemento de Bio-envelope glycoprotein gp350 results in the upregulation of the TNF-α gene

quimica Medica y Biologia Molecular, Facultad de Medicina y Hospitalexpression in monocytic cells via NF-κB involving PKC, PI3-K and tyrosineUniversitario Virgen Macarena, Avenida Sanchez Pizjuan, 4, Sevilla,Kinases. J Mol Biol 2000; 298: 765-78E-41009, Spain.12. Chiurazzi P, Pomponi MG, Willemsen R, et al. In vitro reactivation of the FMR1

gene involved in fragile X syndrome. Hum Mol Genet 1998; 7: 109-13 E-mail: elizabet@us.es

© 2003 Adis Data Information BV. All rights reserved. Mol Diagn 2003; 7 (3)