The Class II Tumor-Suppressor Gene RARRES3 Is Expressed in B Cell

Leukemia (2001) 15, 1521–1526  2001 Nature Publishing Group All rights reserved 0887-6924/01 $15.00 www.nature.com/leu The class II tumor-suppressor gene RARRES3 is expressed in B cell lymphocytic leukemias and down-regulated with disease progression B Casanova1, MT de la Fuente1, M Garcia-Gila1, L Sanz1, A Silva1, JA Garcia-Marco2 and A Garcia-Pardo1

1Departamento de Inmunologıá, Centro de Investigaciones Biolo´gicas, CSIC, Madrid; and 2Servicio de Hematologıá, Hospital Universitario Clıńica Puerta de Hierro, Madrid, Spain

The molecular pathogenesis of B cell chronic lymphocytic leu- treatment and poor prognosis9,15 and alterations of the retinoic kemia (B-CLL), the most common form of leukemia, remains acid receptor alpha (RARa) gene at chromosome 17 (6% of unknown. We have used the mRNA differential display tech- 16,17 nique to analyze genes that may be involved in the cases). development/progression of B-CLL. We have identified the To gain some insight on the involvement of these gene alter- tumor suppressor retinoic acid receptor responder 3 ations in the development and progression of B-CLL, we have (RARRES3) as a B-CLL-related gene. RARRES3 maps to chro- used the mRNA differential display technique (DD)18 to ana- mosome band 11q23, a region frequently deleted in lymphopro- lyze gene expression differences in B-CLL and normal B cells. liferative disorders. To assess the potential involvement of Among the differences detected we have identified the reti- RARRES3 in leukemogenesis, we examined 24 cases of B-CLL, 10 of acute lymphocytic leukemia (ALL) and five related cell noic acid receptor responder 3 gene (RARRES3), a recently lines by RT-PCR and sequence analyses. We report a corre- described growth regulatory gene which functions as a class lation between RARRES3 down-regulation and B-CLL pro- II tumor suppressor and mediates some of the growth sup- gression. We also found decreased RARRES3 gene levels in pressive effects of retinoids.19 We have analyzed the ALL cases and in the five cell lines studied. We did not find expression and sequence of this gene in 24 cases of B-CLL mutations in any of the leukemia samples assayed, including and have found a clear correlation between down-regulation those with 11q23 deletion. These results indicate that RARRES3 may play a role in B-CLL progression. Leukemia of RARRES3 expression and B-CLL progression. Furthermore, (2001) 15, 1521–1526. we have extended these analyses to 10 cases of acute lym- Keywords: RARRES3; B-chronic lymphocytic leukemia; mRNA phoblastic leukemia (ALL), another lymphoproliferative dis- differential display; tumor suppressor gene; lymphoproliferative order, and to five Epstein–Barr virus (EBV)-transformed cell disorders lines established from normal and malignant B lymphocytes.

Introduction Materials and methods

B cell chronic lymphocytic leukemia (B-CLL) is the most com- Patients, cell purification and cell lines mon form of leukemia accounting for 0.8% of all cancers and nearly 30% of adult leukemia.1,2 However, the molecular Peripheral blood samples were obtained with informed con- pathogenesis of B-CLL remains largely unknown and no spe- sent from24 B-CLL and 10 B-cell ALL patients, diagnosed cific gene has been shown to play a major role in this dis- according to established clinical and laboratory criteria (Table ease.1–4 B-CLL progression is usually associated with clonal 1). The B-CLL diagnosis was staged according to Binet and karyotypic alterations described in up to 79% of B-CLL Rai’s classifications.20,21 CD5+ malignant B-lymphocytes were patients.5,6 The most frequent abnormalities are deletions in purified fromthe peripheral blood of B-CLL samplesas pre- chromosome bands 13q12–q14 (30–80% of cases), followed viously described.22 B-lymphocytes from healthy donors were by trisomy 12 (15–35%) and deletions at 11q (20%).5,7–10 11q purified frombuffy coat cells obtained fromthe Centro de deletions correlate with extensive nodal involvement, younger Transfusiones de la Comunidad de Madrid (Madrid, Spain) by age incidence and more aggressive disease progression.5,7,11 Ficoll–Hypaque centrifugation followed by incubation with Some of the reported genetic aberrations in B-CLL affect inter- anti-CD19-coated Dynabeads (Dynal, Oslo, Norway) accord- esting genes, such as LEU2B (located at 13q14.3) which is ing to the manufacturer’s instructions. Purified B cells were frequently deleted in B-CLL,12 although its involvement in the Ͼ95% CD19+ and Ͼ98% viable as determined by flow cyto- disease has not been yet elucidated. Likewise, ATM (ataxia metry and trypan blue dye exclusion, respectively. Although telangiectasia mutated) and RDX (radixin), two potential normal B lymphocytes are not phenotypically identical to B- tumor-suppressor genes, are located at 11q22.3–q23.1, CLL since most of them do not express CD5+ or CD23, they another frequently deleted region. Poor outcome in B-CLL has are closely related cells and generally accepted as control for been associated with absence of ATM protein or loss of the B-CLL cells. ATM gene,11,13 suggesting that mutations of ATM at the germ- The EHEB cell line, established froma patient with B-CLL line could be a risk for development of B-CLL.14 Other fre- by in vitro transformation with EBV, was obtained from the quent abnormalities in B-CLL are deletions of 6q21–q23;9 German Collection of Microorganisms and Cell Cultures deletions/mutations of the P53 tumor suppressor gene at (Braunschweig, Germany). The HUT112, GER112, CO31 and 17p13.1 (16% of cases) which associate with resistance to CO43 cell lines, established from normal B-lymphocytes by in vitro transformation with EBV, were obtained from Dr S Rodrı´guez de Co´rdoba (Centro de Investigaciones Biolo´gicas, Madrid, Spain) and Dr R de Pablo (Hospital Universitario Correspondence: A Garcia-Pardo, Centro de Investigaciones Bio- lo´gicas, CSIC, Vela´zquez 144, 28006 Madrid, Spain; Fax: 34 91 Clıńica Puerta de Hierro, Madrid, Spain). JM and CEM cells 562 7518 (T cell leukemia) were obtained from Dr C Bernabeu (Centro Received 8 March 2001; accepted 13 June 2001 de Investigaciones Biolo´gicas); RPMI 8866 (B lymphoblastic) RARRES3 expression in lymphoproliferative disorders B Casanova et al 1522 Table 1 Clinical characteristics of patients

Patient Sex/Age Morphol Stagea Therapy Karyotype

B-CLL patients 1 M/72 Typical C/III No Normal 2 M/NANANANANA 3 M/59 NA B No ND 4 F/71 Atypical C/IV FDR 46,XX, del(11)(q23)[5] 5 F/78 Typical A/I LK 46, XX, del(6)(q15)[10], 46,XX[10] 6 M/49 Atypical C/IV FDR 44–46, XY, −2, der(5)t(5;12) (q35;q12), der(12) t(12;?)(q12;?), del(14)(q12), i(17)(q10),der(18) t(18;21;?)(p11;q11;?), −21, +mar[cp35], 88–115 hiperploidy[10] 7 M/67 Typical C/IV LK 46,XY,del(11)(q23)[12], 46,XY[8] 8 M/77 Typical A/0 No 46,XY[20] 9 F/88 Typical A/1 No No metaphase 10 F/72 Typical A/0 No 46,XX, t(4;13)(q34;q13)[5], t(2;10) (p11;p15), del(11)(q21),− 13,+mar[5], 46,XX[10] 11 M/45 Typical A/0 No 47,XY,+12[13],46,XY,+12,−18[3],46,XY del(13)(q12-q14)[3] 12 F/68 Atypical B/II FDR 45,X,t(X;3)(p22;p23),−7,add(15)(p13),−17, der(19) t(7;19) (p11;p13) ins(19;1)(p13;q11;q44)[16] 13 M/69 Typical C/III FMC 46,XY[20] 14 M/70 Typical A/I HYDREA 46,XY[20] 15 M/67 Typical B/II No 46,XY[20] 16 M/74 NA A/I-II NA NA 17 M/NA NA A/I No NA 18 M/65 Atypical A/0 No 46,XY, del(6)(q16),add(7)(p22)[19], 46, XY, add(7)(p22)[6] 19 F/36 Atypical B/II FDR 46,XX[20] 20 F/73 Atypical A/0 No 46,XX, del(8)(p11)[15], 46,XX[5] 21 M/46 Typical C/IV CHOP 46,XY, del(11)(q13)[20] 22 M/73 Atypical A/0 No 46,XY, del(11)(q23)[16]; 46, XY[4] 23 M/44 Atypical C/IV ESHAP 46,XY, del(11)(q21)[10]; 46,XY, del(11)(q21), der(21)t(11;21) (q12q23;p11)[9]; 47,XY, der(1)del(1)(q10),+der(1)del(1)(p11), del(11) (q21),der(21)t(11;21)(q12q23;p11)[3]; 46,XY[2] 24 M/58 Typical B/II No 46,XY,t(7;11;?)(q36;q12q25;?),del(11)(q22)[15]

ALL patients 1 M/16 46,XY[16], 46, XY, t(2;14)(p21;q22)[4] 2 M/20 47, XX, t(9;22)(q34;q11), +der(22) t(q;22)(q34;q11)[16], 46, XX, der(9) t(9;22)(q34;q11)[2], 46, XX, t(9;22)(q34;q11)[2] 3 M/66 35, XY, −3,−4,−5,−7,−9,−13,−14,−15,−16,−17,−19, del(2)(q11)[13], 46,XY, del(11)(q23)[3], 46,XY[4] 4 M/53 46, XY[16], 47, XY, add(3)(q28), +mar[4] 5 F/18 46, XX[20] 6 M/14 46, XY, i(9)(q10)[9] 7 M/12 ND 8 M/20 60, XY, add(1)(q44), +4, +5, +6, +del(6)(q16), +8, ?t(8;14)(q22;q32)?, +11, +12, +14, +15, +i(17)(q10), +18, +21, +2mer[20], 46, XY,[15] 9 M/26 47, XY, +4, t(11;12)(q22;p13)[22] 10 M/NA 49, XY, +X, +der(1) t(1;22)(p11;q11), +8, +8, +13, −21, −22[18]

aAccording to Binet et al20 and Rai et al.21 NA, not available; ND, not determined; FDR, ﬂudarabine; FMC, combination of ﬂudarabine, mitoxantrone and chlorambucil; LK, leukeran; CHOP, combination of cyclophosphamide, adriamycin, vincristine and prednisone; ESHAP, combination of etoposide, cisplatin, ara-C and prednisone.

and RPMI 8226 (myeloma) cells were obtained from Dr F mRNA differential display Sa´nchez-Madrid (Hospital de la Princesa, Madrid, Spain). Total RNA was isolated with TRIzol reagent (GIBCO-BRL, Life Technologies, Grand Island, NY, USA) following the manufac- Chromosome analysis turer’s protocol. DNA contamination was removed by treatment with DNAsel (Boehringer Mannheim, Indianapolis, IN, Mononuclear cells frombone marrowand/or peripheral blood USA) according to the manufacturer’s instructions. mRNA dif- for CLL samples and from bone marrow for ALL samples were ferential display between B-lymphocytes from a normal sub- cultured at 37°C for 3–5 days in the presence of tetrade- ject (control cells) and froma B-CLL patient (patient 1 on canoylphorbol-13-acetate (0.05 mg/ml) for CLL samples and Table 1) was performed in duplicate as described.18 In brief, 2–3 days without addition of mitogens for ALL samples. At total RNA was divided in aliquots of 0.3 ␮g and used to dupli- this time standard cytogenetic preparations were made, G- cate all subsequent reactions. First-strand cDNA was synthe- banded and karyotyped according to the International System sized using 630 U/␮g RNA of Moloney murine leukemia virus for Cytogenetic Nomenclature.23 If available, 20 metaphases reverse transcriptase (M-MLV) (Amersham Life Science, Cleve- were evaluated for each case. land, OH, USA) in three different reverse transcription (RT)

Leukemia RARRES3 expression in lymphoproliferative disorders B Casanova et al 1523 reactions primed by three downstream one-base anchored Statistical analysis Ј oligo-dT primers (oligo-dT11V). Their sequence were 5 - Ј ␮ AAGCT11V-3 (where V can be A, C, or G). 2.0 l of the RT For statistical comparison between groups the Student’s non- reaction were amplified as described.18 1 ␮Ci of [␣-33P]dATP paired t-test was used. Two-tailed statistical significance was (2000 Ci/mmol; NEN Life Science Products Inc, Boston, MA, determined. A P-value Ͻ0.05 was considered significant. USA) was used per reaction tube. We employed the three Analyses were performed using GraphPad InStat V2.04a oligo-dT11V primers described before in combinations with software (GraphPad Software, San Diego, CA, USA). five upstreamarbitrary primers(13 bases long). The arbitrary primers were: AP1 (5Ј-AAGCTTGATTGCC-3Ј), AP2 (5Ј- AAGCTTGACTGT-3Ј), AP3 (5Ј-AAGCTTTGGTCAG-3Ј), AP14 Results and discussion (5Ј-AAGCTTCAGCGAA-3Ј) and AP45 (5Ј-AAGCTTACTC CAC-3Ј). The duplicate PCR products were separated on To identify genes which may be involved in the development sequencing gels under denaturing conditions. Dried gels were and/or progression of B-CLL, we compared the pattern of gene exposed to Kodak X-OMAT AR film(Eastman-Kodak,Roches- expression in malignant B lymphocytes from a B-CLL patient ter, NY, USA) for 24–48 h. Bands that were unique to the con- (patient 1 in Table 1) and froma normalsubject using the trol or B-CLL cells and were present in both duplicate reac- mRNA DD method. A 197-bp fragment of cDNA was found tions were isolated and re-amplified as described.18 Bands to be expressed in normal B cells, but absent in the B-CLL successfully amplified were ligated in pGEM-T Easy Vector sample (Figure 1a). We initially called this cDNA ‘leukemia (Promega Corporation, Madison, WI, USA) and cloned in related gene-1’ (LRG-1). DNA sequencing of this cDNA and DH5␣ competent cells. subsequent search for homology in the public GenBank data- base revealed that LRG-1 showed 99% identity to nucleotides 328–512 of the retinoic acid receptor responder 3 gene PCR analysis of RARRES3 expression (RARRES3, GenBank accession no. AF060228) resulting in highly homologous protein products (Figure 1b). We will Total RNA (1 ␮g) was reverse transcribed with 50 mM therefore refer to LRG-1 as RARRES3. This gene is a well- oligo(dT) primer and 50 U/␮g RNA of M-MLV in a volume of known growth regulator whose expression correlates inversely 20 ␮lat42°C for 45 min. 0.2 ␮l of this reaction were PCR with cell proliferation and is decreased in several cancer cell amplified by using the primer RARRES3-1F 5Ј-TCAAGC lines and in some primary tumors such as lymphoma, ureter, TTGGAGCACCAGACCTCTC-3Ј or the primers described by kidney, rectal and uterine.19 DiSepio et al19 F: 5Ј-TCAAGCTTCCACCATGGCTTCGCCA- RARRES3 maps to the 11q23 chromosomal region and thus CACCAAGAGCCCA-3Ј and R: 5Ј-TTGGATCCTGTGGCT below the ATM gene.19 Aberrations at 11q regions are recur- GCTTCAGGCGTTGC-3Ј. A control PCR product, correspond- ring abnormalities in various types of neoplasias including ing to human glyceraldehyde-3-phosphate dehydrogenase lymphoproliferative disorders.24–28 In addition, previous stud- (HGAPDH), was used under the same conditions for nor- ies have shown that loss of heterozygosity (LOH) at the gen- malizing measurement differences (HGAPDH-203F: 5Ј- omic region 11q22–q23 is one of the most common structural ATGGCACCGTCAAGGCTGAG-3Ј and HGAPDH-880R: 5Ј- chromosome aberrations in B-CLL,5,7–10 suggesting that a AGACCACCTGGTGCCAGTG-3Ј). 1 ␮Ci of [␣-32P]dCTP novel tumor suppressor gene may be located in this region. (3000 Ci/mmol. Amersham International) per reaction tube Because RARRES3 is mapped to this hot spot chromosome was used. Aliquots of 4 ␮l were taken fromthe reaction mix- region and this gene has not been analyzed thus far in hema- ture at cycles 10, 15, 18, 21, 25 and 28 for HGAPDH product, tologic malignancies, we performed expression and and 21, 25, 28, 30, 35, and 40 for RARRES3 product. The mutational analyses to determine whether RARRES3 could amplified products were separated on a 4% agarose gel, trans- play a role in B-CLL pathogenesis. ferred to nylon membranes and radioactivity quantified by For these studies we used RT-PCR to analyze samples from Phosphorlmager scanning (Molecular Dynamics, Sunnyvale, the 24 B-CLL patients listed in Table 1. The overall incidence CA, USA). For subsequent analyses, we used 18 cycles of of clonal karyotype abnormalities in the B-CLL cases studied amplification for HGAPDH products and 25 cycles for was 65% and seven out of 20 patients (35%) had LOH at RARRES3 products and proceeded as described above. To 11q23 based on chromosome analysis (Table 1). Samples quantify the expression of RARRES3 gene we calculated the fromfour normalsubjects were used as reference. We also ratio of the ‘signal of RARRES3 product’ vs ‘signal of HGAPDH carried out these analyses in four normal EBV-B cell lines and product’. Each result was validated by two sets of experiments in EHEB cells. As shown in Figure 2a and Table 2, RARRES3 (PCR and quantification). was expressed in normal B-lymphocytes (1.069 ± 0.076, arbitrary units) and this expression was significantly decreased in the normal EBV-lymphocytes and EHEB cell line (Figure 2a, Sequence analysis Table 2). These results are in agreement with the role of RARRES3 as a class II tumor suppressor gene, known to be 0.2 ␮l of the RT reaction described above were PCR amplified functionally active, but expressed at low levels in cell lines by using primers RARRES3-1F 5Ј-TCAAGCTTGGAGCACA- and/or tumors.19 GACTCCT C-3Ј and RARRES3-678R 5Ј-TTGGATCCTCC As shown in Table 2, RARRES3 expression values at early TTCAGTCTTGTTTCAATTAG-3Ј. Amplified products were run stages (A/0–1) of B-CLL did not differ significantly fromcontrol on an 2% of agarose gel, purified with the Qiaquick PCR puri- levels. At later stages (B and C) however, there was a progress- fication kit (Qiagen, Valencia, CA, USA) and sequenced with ive significant decrease in the expression of this gene. These the corresponding forward and reverse primers using the dide- data indicate that down-regulation of RARRES3 gene oxy terminator reaction chemistry in an automated ABI-Prism expression was clearly related to B-CLL disease progression. 377 sequencing system (Applied Biosystems-Prism-Perkin Since one of the characteristics of this progression is the tran- Elmer, Foster City, CA, USA). sition froma low mitoticactivity phase (stage A) to a higher

Leukemia RARRES3 expression in lymphoproliferative disorders B Casanova et al 1524

Figure 2 Expression analysis of RARRES3 gene by RT-PCR. 4 ␮lof each ampliﬁed product were run on a 4% agarose gel, transferred to nylon membrane and exposed to PhosPhorlmager scanner to quantify radiactivity. (a) RT-PCR from: P, B-CLL patients; CTR, control subjects; EHEB, B-CLL cell line; HUT112, GER112, CO43, and CO31, EBV- cell lines fromnormalsubjects. P6–2 and P8–2, are repeated extrac- tions frompatients P6 and P8 obtained at a later stage of the disease. The size of RARRES3 product (primers RARRES3-1F and R) is 550 bp. The size of RARRES3 product at cell line CO31 (ampliﬁed with primers F and R) is 526 bp. The size of control product HGAPDH (primers HGAPDH-203F and HGAPDH-880R) is 678 bp. (b) RT-PCR fromALL patients.

otype abnormalities in the nine cases studied was 88.9% and one patient or 11.1% had deletion at 11q23 (Table 1). As shown in Figure 2b and Table 2, the expression of RARRES3 was also decreased in ALL samples (0.239 ± 0.045) when compared with normal B-cells (1.069 ± 0.076). Therefore down-regulation of this gene could also contribute to the typical lymphocyte high proliferation rate observed in this malignancy.29 Figure 1 (a) Section of differential display gel. Total RNA fromB- Characterization of RARRES3 as a true tumor-suppressor CLL patient 1 (P1) and froma normalsubject (CTR) were subjected gene would require the identification of loss of heterozygosity to differential display (primer combination oligo-dT11G-AP3). P1-R1 and P1-R2, cDNAs amplified from RT reactions 1 and 2 from patient or/and inactivating mutations in cancer patients. Thus, we car- 1. CTR-R1 and CTR-R2, cDNAs amplified from RT reactions 1 and 2 ried out sequence analyses on cDNA samples obtained by RT- froma normalsubject. The position of the band (LRG-1) that was PCR amplification from EHEB and EBV-transformed cell lines, further analyzed is indicated. (b) Multiple sequence alignment show- normal B cells and all B-CLL and ALL samples listed in Table ing homology between LRG-1, RARRES3, human H-rev 107 (H-rev), 1. Our analyses revealed three nucleotide differences between and rat H-rev 107 (R-rev). Consens, consensus sequence. Differences in residues 63 and 118 between LRG-1 and RARRES3 are enclosed the RARRES3 cDNA amplified by us (LRG-1) and the pre- 19 in black squares. viously reported RARRES3 sequence, which resulted in two amino-acid changes (Figure 1b, Table 3). These differences were found in all leukemia samples assayed as well as in T proliferation stage reflected by a lymphocyte doubling time of (JM, CEM) and B (RPMI 8866, RPMI 8226) cell lines, suggest- less than 12 months2 and RARRES3 appears to be an inhibitor ing that these nucleotide changes may represent the wild of cell proliferation,19 the reduced expression of this gene sequence, rather than polymorphism. In this context, we did observed in our study could facilitate the increase of CD5+ not find mutations in any of the 36 leukemia samples assayed, malignant B-lymphocytes that accompany B-CLL progression. including the seven B-CLL and one ALL cases with 11q23 RARRES3 was also significantly decreased in 11q-deleted B- deletion (see Table 1). CLL samples compared to control values (Figure 2a, Table 2). Failure to detect mutations in these cases might suggest that We have also examined the expression of RARRES3 in 10 RARRES3 is not the targeted gene in 11q23 deletions or that B lineage ALL samples. The overall incidence of clonal kary- the remaining allele, which may account for the observed

Leukemia RARRES3 expression in lymphoproliferative disorders B Casanova et al 1525 Table 2 Summary of RARRES3 expression

Study group Study group Mean ± s.e.m. Mean ± s.e.m. P value (relative units) (relative units)

Control (n = 4) vs A/0-I (n = 10) 1.069 ± 0.076 vs 1.226 ± 0.208 NSa Control (n = 4) vs B/II (n = 5) 1.069 ± 0.076 vs 0.508 ± 0.124 0.0086 Control (n = 4) vs C/III–IV (n = 8) 1.069 ± 0.076 vs 0.470 ± 0.107 0.0042 Control (n = 4) vs 11q del B-CLL (n = 7) 1.069 ± 0.076 vs 0.619 ± 0.116 0.0245 A/0–1(n = 10) vs C/III–IV (n = 8) 1.226 ± 0.208 vs 0.470 ± 0.107 0.0086 Control (n = 4) vs ALL (n = 10) 1.069 ± 0.076 vs 0.239 ± 0.045 0.0001 Control (n = 4) vs EBV-lymphocytes (n = 4) 1.069 ± 0.076 vs 0.525 ± 0.183 0.0336 Control (n = 4) vs EBV-lymphocytes + 1.069 ± 0.076 vs 0.498 ± 0.144 0.0145 EHEB (n = 5) aNS, not significant; P Ͻ 0.05, significant; P Ͻ 0.01, very significant, P Ͻ 0.001, extremely significant.

Table 3 Nucleotide and amino acid differences between LRG-1 Interministerial de Ciencia y Tecnologıá (CICYT); and RARRES3 sequences 08.1/0028/99 (to AGP) and 08.1/012/97 (to AS and AGP) from the Comunidad Autońoma de Madrid (CAM). B Casanova, MT Nucleotide Amino acid de la Fuente and L Sanz were supported by fellowships from LRG-1/RARRES3 LRG-1/RARRES3 CAM; M Garcıá-Gila was supported by a fellowship from CICYT. We thank Drs Felipe Prosper and M Jose´ Terol Base pair 217 A/G Glu/Gly (position 63) Base pair 381 A/G Thr/Ala (position 118) (Hospital Clinico Universitario, Valencia, Spain) and Dr Erne- Base pair 515 G/A Ala/Ala (position 162) sto Roldań (Hospital Ramoń y Cajal, Madrid, Spain) for some of the samples and clinical data of B-CLL patients; Dr Santiago Rodrı´guez de Co´rdoba for critical reading of the manuscript and Ms Mercedes Hemańdez del Cerro for excellent techni- RARRES3 low expression in 11q23 deleted samples, is inacti- cal assistance. vated by mechanisms different from mutation or deletion. These alternative mechanisms could affect the expression or inactivate RARRES3 in non 11q23-deleted cases. In this References regard, increasing evidence has suggested that silencing the tumor suppressor genes BRCA1, p27Kip1, p53 and p73, and 1 Gale RP, Foon KA. Biology of chronic lymphocytic leukemia. cyclin-dependent kinase inhibitors genes p15 and p16, by Semin Hematol 1987; 24: 209–229. promoter methylation, may play a role in the development 2 Kipps TJ. Chronic lymphocytic leukemia. Curr Opin Hematol and/or poor prognosis of some neoplasias.30–34 On the other 2000; 7: 223–234. hand, it cannot be excluded that some alterations (ie 3 Moore JS, Friedman DF, Silberstein LE, Besa EC, Nowell PC. Clini- mutations, deletions) do not involve the RARRES3 gene cal heterogeneity reflects biologic diversity in chronic lymphocytic leukemia. Crit Rev Oncol Hematol 1995; 20: 141–164. coding region itself, but affect its expression by modifying 4 Rozman C, Montserrat E. Chronic lymphocytic leukemia. N Engl regulatory sequences. Our results do not rule out these alter- JMed1995; 333: 1052–1057. native mechanisms of tumor suppressor inactivation, and 5 Fegan C, Robinson H, Thompson P, Whittaker JA, White D. Kary- possibly the combination of mono-allelic deletion and/or pro- otypic evolution in CLL: identification of a new sub-group of moter methylation or mutation of noncoding regulatory patients with deletions of 11q and advanced progessive disease. elements could take place in the leukemic cases studied. Leukemia 1995; 9: 2003–2008. 6 Larramendy ML, Siitonen SM, Zhu Y, Hurme M, Vilpo L, Vilpo JA, To our knowledge RARRES3 has not been linked so far to Knuutila S. Optimized mitogen stimulation induces proliferation of the pathogenesis of malignant disease, nor has it been neoplastic B cells in chronic lymphocytic leukemia: significance reported in hematologic malignancies. Our study does not for cytogenetic analysis. The Tampere CLL group. Cytogenet Cell suggest a primary role of RARRES3 gene in B-CLL and ALL Genet 1998; 82: 215–221. leukemogenesis and we did not observe the classical tumor 7 Stilgenbauer S, Liebisch P, James MR, Schro¨der M, Schlegelberger suppressor inactivation mechanism consisting in loss of one B, Fischer K, Bentz M, Lichter P, Do¨hner H. Molecular cytogenetic delineation of a novel critical genomic region in chromosome allele and mutation of the remaining one. However, our bands 11q22.3-q23.1 in lymphoproliferative disorders. Proc Natl results argue in favor of the involvement of RARRES3 function Acad Sci USA 1996; 93: 11837–11841. in B-CLL (and possibly ALL) progression. Down-regulation or 8 Karhu R, Knuutila S, Kallioniemi OP, Siltonen S, Aine R, Vilpo L, deletion of RARRES3, a cell proliferation inhibitor,19 may pro- Vilpo J. Frequent loss of the 11q14-24 region in chronic lympho- vide a dramatic growth advantage for malignant cells and thus cytic leukemia: a study by comparative genomic hybridization. accelerate the leukemic process. Further functional studies Tampere CLL group. Genes Chromosom Cancer 1997; 19: 286– 290. will be necessary to determine the exact role of this gene 9Do¨hner H, Stilgenbauer S, Do¨hner K, Bentz M, Lichter P. Chromo- in the development and progression of hematopoietic some aberrations in B-cell chronic lymphocytic leukemia: reas- neoplasms. sessment based on molecular cytogenetic analysis. J Mol Med 1999; 77: 266–281. 10 Zhu Y, Monni O, El-Rifai W, Siitonen SM, Vilpo L, Vilpo J, Knuut- Acknowledgements ila S. Discontinuous deletions at 11q23 in B cell chronic lymphocytic leukemia. Leukemia 1999; 13: 708–712. 11 Do¨hner H, Stilgenbauer S, James MR, Benner A, Weilguni T, Bentz This work was supported by grants SAF97-0064-CO3-02 (to M, Fischer K, Hunstein W, Lichter P. 11q deletions identify a new AGP) and SAF97-0064-CO3-03 (to AS) fromthe Comisio ń subset of B cell chronic lymphocytic leukemia characterized by

Leukemia RARRES3 expression in lymphoproliferative disorders B Casanova et al 1526 extensive nodal involvement and inferior prognosis. Blood 1997; 22 De la Fuente MT, Casanova B, Garcia-Gila M, Silva A, Garcia- 89: 2516–2522. Pardo A. Fibronectin interaction with ␣4␤1 integrin prevents 12 Corcoran MM, Sangfelt O, Kapanadze B, Makeevaa N, Ibbotson apoptosis in B-cell chronic lymphocytic leukemia: correlation RE, Baranova A, Zabarovsky E, Merup M, Jansson O, Gahrton G, with Bcl-2 and Bax. Leukemia 1999; 13: 266–274. Einhorn S, Grander D, Oscier DG. Characterisation of a candidate 23 Mitelman F. ISCN 1995: an International System for Cytogenetic tumor suppressor gene LEU2B located at 13q14.3 in chronic lym- Nomenclature. Karger: Basel, 1995. phocytic leukemia. Blood 2000; 96 (Suppl. 1): 702a. 24 Rasio D, Negrini M, Maneti G, Dragani TA, Croce C. Loss of het- 13 Starostik P, Manshouri T, O’Brien S, Freireich E, Kantarjian H, Hai- erozygosity at chromosome 11q in lung adenocarcinoma: identi- dar M, Lerner S, Keating M, Albitar M. Deficiency of the ATM fication of three independent regions. Cancer Res 1995; 55: protein expression defines an aggressive subgroup of B-cell 3988–3991. chronic lymphocytic leukemia. Cancer Res 1998; 58: 4552–4557. 25 Mitelman F. Catalog of Chromosome Aberrations in Cancer, 6th 14 Bullrich F, Rasio D, Kitada S, Starostik P, Kipps T, Keating M, Albi- edn. Alan R Liss: New York, 1996. tar M, Reed JC, Croce CM. ATM mutations in B-cell chronic lym- 26 Monni O, Oinonen R, Elonen E, Franssila K, Teerenhovi L, Joensuu phocytic leukemia. Cancer Res 1999; 59: 24–27. H, Knuutila S. Gain of 3q and deletion of 11q22 are frequent aber- 15 Cordone I, Masi S, Mauro FR, Soddu S, Morsilli O, Valentini T, rations in mantle cell lymphoma. Genes Chromosom Cancer Vegna ML, Guglielmi C, Mancini F, Giuliacci S, Sacchi A, Mand- 1998; 21: 298–307. elli F, Foa R. p53 expression in B-cell chronic lymphocytic leuke- 27 Monni O, Zhu Y, Franssila K, Oinonen R, Hoglund P, Elonen E, mia: a marker of disease progression and poor prognosis. Blood Joensuu H, Knuutila S. Molecular characterization of deletion at 1998; 91: 4342–4349. 11q22.1–23.3 in mantle cell lymphoma. Br J Haematol 1999; 104: 16 Padua RA, McKenna S, Butler PC, Williams-Gregory D, Jordan D, 665–671. Smith M, White D, Whittaker J, Bentley P, Burnett AK, Jacobs A. 28 Zhu Y, Monni O, Franssila K, Elonen E, Vilpo J, Joensuu H, Knuut- Identification of retinoic acid receptor alpha (RAR␣) gene ila S. Deletions at 11q23 in different lymphoma subtypes. Haema- rearrangements in leukaemias: cloning of a novel RAR␣-PLZF tologica 2000; 85: 908–912. (promyelocytic leukaemia zinc finger) fusion in a patient with 29 Laport GF, Larson RA. Treatment of adult acute lymphoblastic leu- chronic lymphocytic leukaemia. Blood 1995; 86 (Suppl. 1): 34a. kemia. Semin Oncol 1997; 24: 70–82. 17 Parrado A, West R, Jordan D, Bastard C, McKenna S, Whittaker 30 Rice JC, Ozcelik H, Maxeiner P, Andrulis I, Futscher BW. Methyl- JA, Bentley P, White D, Chomienne C, Padua RA. Alterations of ation of the BRCA1 promoter is associated with decreased BRCA1 the retinoic acid receptor alpha (RAR alpha) gene in myeloid and mRNA levels in clinical breast cancer specimens. Carcinogenesis lymphoid malignancies. Br J Haematol 1999; 104: 738–741. 2000; 21: 1761–1765. 18 Liang P, Pardee AB. Differential display of eukaryotic messenger 31 Corn PG, Kuerbitz SJ, van Noesel MM, Esteller M, Compitello N, RNA by means of the polymerase chain reaction. Science 1992; Baylin SB, Herman JG. Transcriptional silencing of the p73 gene 257: 967–971. in acute lymphoblastic leukemia and Burkitt’s lymphoma is asso- 19 DiSepio D, Ghosn C, Eckert RL, Deucher A, Robinson N, Duvic ciated with 5Ј CpG island methylation. Cancer Res 1999; 59: M, Chandraratna RAS, Nagpal S. Identification and characteriz- 3352–3356. ation of a retinoid-induced class II tumor suppressor/growth regu- 32 Baur AS, Shaw P, Burri N, Delacretaz F, Bosman FT, Chaubert P. latory gene. Proc Natl Acad Sci USA 1998; 95: 14811–14815. Frequent methylation silencing of p15(INK4b) (MTS2) and 20 Binet JL, Auquier A, Dighiero G, Chastang C, Piguet H, Goasguen p16(INK4a) (MTS1) in B cell and T cell lymphomas. Blood 1999; J, Vaugier G, Potron G, Colona P, Oberling F, Thomas M, Tchernia 94: 1773–17781. G, Jacquillat C, Boivin P, Lesty C, Duault MT, Monconduit M, 33 Drexler HG. Review of alterations of the cyclin-dependent kinase Belabbes S, Gremy F. A new prognostic clasification of chronic inhibitor INK4 family genes p15, p16, p18 and p19 in human lymphocytic leukemia derived from a multivariate analysis. Can- leukemia–lymphoma cells. Leukemia 1998; 12: 845–859. cer 1981; 48: 198–206. 34 Warnecke PM, Bestor TH. Cytosine methylation and human can- 21 Rai KK, Sawitsky A, Cronkite EP, Chanana AD, Levy RN, Pasternak cer. Curr Opin Oncol 2000; 12: 68–73. BS. Clinical staging of chronic lymphocytic leukemia. Blood 1975; 46: 219–234.

Leukemia