Cancer Genetics and Cytogenetics 125 (2001) 156–160 Modified order of allelic replication in lymphoma patients at different disease stages Aliza Amiel,* Avishay Elis, Danith Blumenthal, Elena Gaber, Moshe D. Fejgin, Ron Dubinsky, Michael Lishner Genetic Institute and Department of Medicine and Hematology, Meir Hospital, Sapir Medical Center, Kfar-Saba and Department of Medicine, Wolfson Medical Center, Holon and Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel Received 12 July 2000; accepted 11 September 2000 Abstract Asynchronous replication of homologous loci was reported in lymphocytes of patients with lymphoma, ovarian and renal cancer as well as in lymphocytes of patients with premalignant conditions, for example, essential mixed cryoglobulinemia associated with hepatitis C virus and in monoclonal gammopathy of unknown significance. In the present study we evaluated the replication pattern in lymphocytes of four groups of patients with intermediate grade of non-Hodgkin lymphoma at various stages of their disease: 1) at diagnosis; 2) during cytotoxic treatment; 3) in remission; and 4) in relapse. A significantly higher proportion of the asynchronous pattern of replication at diagnosis, during cytotoxic treatment, and in relapse was noted as compared to healthy controls and to patients who achieved remission of their lymphoma. Also, the frequency of the two doublets (DD) pattern in every group studied was significantly lower than in the controls. If our findings can be confirmed in larger, long-term prospective studies, it may allow the use of a simple and inexpensive tool to closely observe patients with lymphoma who are at high risk for relapse. © 2001 Elsevier Science Inc. All rights reserved. 1. Introduction In the past, in order to follow the replication pattern of a given DNA sequence, prelabelling of cells with BrdU was the method of choice, usually accompanied by cell synchronization or cell sorting [1–3]. The more recent method of fluorescence in situ hybridization (FISH) enables us to determine replication timing of allelic DNA sequences in unsynchronized cell populations [4–6]. Accordingly, an unreplicated DNA sequence at interphase is manifested by a single fluorescent signal (singlet[S]) whereas a replicated sequence gives rise to a double signal (doublet [D]). Thus, in a population of replicating cells, a high frequency of nuclei with two similar hybridization signals—either two singlets (SS) or two doublets (DD)—would indicate a pair of allelic loci that replicate synchronously. On the other hand, allelic loci that replicate asynchronously are revealed by a high frequency of nuclei containing two different hybridization signals, a singlet and a doublet (SD) [4–6]. * Corresponding author. Tel.: 1972-9-7472220; fax: 1972-9-7419851. E-mail address: alizaamiel@hotmail.com (A. Amiel). A close association usually exists between the specific time interval during S-phase at which a particular DNA sequence replicates in a given tissue and its transcriptional status: expressed loci are usually early replicating, whereas unexpressed ones replicate late. Hence, housekeeping genes, encoding products which are essential for cell maintenance, replicate early in most cell types, whereas tissuespecific genes reveal a differentiation-dependent pattern of replication, undergoing early replication in cells where they are expressed, and late in cells where they are silent [1–3]. It has been shown that homologous regions of the TP53, C-MYC, HER-2/neu, and 21q22 loci, each known to accommodate genes associated with various aspects of malignancy, replicate synchronously in different types of normal diploid cells, such as peripheral leukocytes [7–9] and bone-marrow cells [8]. On the other hand, these same loci, when present in malignant cells of patients suffering from hematological neoplasms, for example, chronic lymphocytic leukemia, chronic myeloid leukemia and lymphoma, show an asynchronic replication, giving rise to an early and late replicating allele for each locus [7,8]. The replication pattern of lymphocytes derived from patients with premalignant conditions is different from that of normal controls or patients 0165-4608/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S0165-4608(00)00 3 8 1 - 2 157 A. Amiel et al. / Cancer Genetics and Cytogenetics 125 (2001) 156–160 with established malignancies. For example, we have reported that a significantly higher rate of asynchronized replication of different loci is present in patients with monoclonal gammopathy of unknown significance than in controls, but this rate is significantly lower than in patients with myeloma [10]. Similar results were obtained in patients with cancer of the cervix when evaluated by the degree of malignancy [11]. Asynchronous replication of homologous alpha-satellite sequences (repetitive DNA arrays lacking transcriptional capability), which are engaged in chromosome segregation, was reported in ovarian tumor cells as well as in leukocytes of patients with ovarian cancer [12]. Also, lymphocytes of patients with renal cancer exhibit a modified order of allelic replication [9]. This represents a non-genetic alteration associated with malignancy and offers a potential test for cancer identification. In the current study we evaluated the replication pattern in peripheral lymphocytes of four groups of patients with non-Hodgkin lymphoma at various stages of their disease: 1) at diagnosis; 2) during cytotoxic treatment; 3) in remission and 4) in relapse. We used the FISH method, with probes for RB-1 and HER-2/neu, for analysis of synchronization of replication in each of these phases of the disease. 2.3. In-situ-hybridization Fresh slide spreads were denatured for 2 min in 70% formamide 2 3 SSC at 708C and dehydrated in a graded ethanol series. The probe mix was then applied to air-warmed slides (30m mix sealed under a 24 3 50 mm glass coverslip) and hybridized for 18 hrs at 378C in a moist chamber. Following hybridization, the slides were washed in 50% formamide 2 3 SSC for 20 min at 438C, rinsed in two changes of 2 3 SSC at 378C for 4 min each, and placed in 0.05% Tween 20 (Sigma Chemicals, St. Louis, MO). The slides were counterstained in DAPI (Sigma) antifade solution and analyzed for simultaneous viewing of FITC, Texas red, and DAPI. An Applied Imaging system was used for the FISH analysis. The replication pattern was assessed only in cells with two fluorescent signals. FISH efficacy was 98%. In each sample we analyzed the two chromosome regions, 17q11.2–q12 and 13q14.3. Following monocolor FISH, Table 1 Clinical characteristic of therapeutic approaches Bone-marrow Patient no. Sex/Age Lymphoma sub-types involvement Treatment 2. Material and methods Seventeen patients with intermediate grade lymphoma (study group) and six healthy controls, matched for age, were analyzed. The diagnosis and classification of lymphoma was according to The Working Formulation. The patients were randomly selected from the patients attending the clinic, by the phase of their clinical course. Each patient consented to participate in the study. Lymphocytes were incubated for short-term culture in an F10-supplemented medium in a 378C moist chamber for 72 hours. The supplemented medium contained 20% FCS, 3% PHA (phytohemagglutinin), 0.2 heparin, and 1% antibiotics. After incubation, colchicine (final concentration 0.1 mg/ ml) was added to the cultures for 1 hour, followed by hypotonic treatment (0.075-M KCl at 378C for 15 min) and four washes each with a fresh cold 3:1 methanol: acetic acid solution. The lymphocyte suspensions of the three samples were stored at 248C. 2.1. Slide preparation Glass slides were precleaned for FISH by incubation in concentrated sulfochromic solutions, rinsed with distilled water followed by two series of absolute ethanol, and then dried with a clean cloth. The stored cell suspensions were washed with a fresh cold 3:1 methanol: acetic acid solution and then dropped onto the precleaned slides and air-dried. 2.2. Probes Two direct labeled commercial probes were used: one for the HER-2/neu locus (17q11.2–q12, Vysis 32-1900003), and one for the RB-1 locus (13q14.3, Vysis 32-190045). 1a 2a 3a 4a M/68 F/60 F/53 F/55 5b M/73 6b 7b M/71 F/76 8b M/72 9b 10c 11c F/70 M/49 F/71 12c 13c M/72 F/73 14d M/49 15d 16d 17d M/77 F/68 M/61 Follicular large cell DLCL DLCL Diffuse mixed small & large cells DLCL 1 2 2 2 DLCL Diffuse mixed small cleaved & large cell DLCL 2 1 DLCL DLCL Diffuse mixed small & large cell DLCL Diffuse mixed small & large cell Follicular mixed small & large cell 1 2 2 CHOP 3 8 DVIP 3 4 CHOP 3 1 CHOP 3 2 CHOP 3 8 CHOP 3 4 2 2 CHOP 3 6 CHOP 3 6 2 Prednison Lukeron CHOP 3 6 CHOP 3 6 CHOP 3 6 CHOP 3 8 CHOP 3 8 DLCL DLCL Intermediate grade 2 1 2 2 2 CHPOP 3 4 DVIP 3 3 CAMP 3 1 CHOP 3 4 CHOP 3 4 CHOP 3 5 Abbreviations: DLCL, diffuse large cell lymphoma; CHOP, cyclophosphamide, adriamycin, oncovin, prednisone; DVIP, dexacort, etoposide, ifosphamide, cisplatinum; CAMP–CCNU, cytosine arabinoside, mitoxantrone, prednisone. a At diagnosis. b During cytotoxic treatment. c In remission. d In relapse. 158 A. Amiel et al. / Cancer Genetics and Cytogenetics 125 (2001) 156–160 we recorded for each probe 92–125 interphase cells that showed two hybridization signals (Tables 1 and 2). The examined cells were classified into three categories, SS, DD, and SD, according to the replication status of the two homologous loci. The slides were blindly scored by two different readers. groups. Pearson Chi-square and Fisher’s Exact tests were applied in order to examine differences between the study groups for the categorical parameters. All tests applied were two-tailed, and p value of 5% or less was considered statistically significant. The data were analyzed using the SAS software (SAS Institute, Cary North Carolina). 2.4. Statistical methods The following statistical tests were used in the analysis of the data presented in this paper: The two-sample t-test and non parametric test were applied for testing differences between the study groups for quantitative parameters. The Multiple Comparisons Tests (Duncan’s method) were applied for testing quantitative parameters between the study Table 2 Replication pattern of RB-1 locus in peripheral blood mononuclear cells Cells in synchronization Patient Control 1 2 3 4 5 6 Mean Patients at diagnosis 1 2 3 4 Mean Patients during chemotherapy administration 1 2 3 4 5 Mean Patients in remission 1 2 3 4 Mean Patients in relapse 1 2 3 4 Mean SS pattern DD pattern Cells in asynchronization SD pattern Total number of cells 68 85 80 57 65 68 62.6 6 1.5 32 29 33 32 25 28 26.7 6 1.3 11 (10) 11 (9) 12 (10) 11 (11) 13 (13) 13 (12.5) 10.6 6 0.6 111 125 125 100 103 104 60 60 80 70 67.5 6 4.7 10 16 5 10 10.3 6 2.2 30 (30) 24 (24) 15 (15) 20 (20) 22.2 6 3.2 100 100 100 100 76 68 80 75 75 73.9 6 2.5 6 13 2 10 6 7.3 6 1.8 20 (20) 24 (23) 18 (18) 15 (15) 19 (19) 18.9 6 1.3 102 105 100 100 100 67 85 79 75 74 6 6.0 18 5 7 10 10 6 2.9 25 (23) 10 (10) 14 (14) 15 (15) 15 6 3.2 110 100 100 100 84 70 80 65 76.6 6 5.8 1 10 4 10 6.3 6 2.2 7 (8) 20 (20) 16 (16) 25 (25) 17.2 3 3.7 92 100 100 100 SS cells with two single signals; DD cells with two double signals; SD cells with one single and one double signal. Number in parentheses 5 the proportion of cells from the total cell population. 3. Results There were eight women and nine men with a mean age of 62 years. The lymphoma subtypes, clinical characteristics, and treatment approaches are presented in Table 1. The number of cells with synchronous (SS and DD) and asynchronous (SD) pattern of replication of the RB-1 and HER-2/neu Table 3 Replication pattern of HER-2/neu locus in peripheral blood mononuclear cells SS pattern DD pattern Cells in asynchronization SD pattern 50 52 59 50 50 56 52.6 6 1.2 40 38 32 35 40 35 36.4 6 1.4 12 (12)a 10 (10) 13 (12.5) 12 (12) 10 (10) 11 (11) 11.2 6 0.5 102 100 104 97 100 102 60 75 65 60 65.0 6 35 15 8 10 12 11.2 6 1.5 25 (25) 17 (17) 25 (25) 28 (28) 23.7 6 2.3 100 100 100 100 85 57 75 62 75 69.5 6 4.2 6 14 10 8 7 8.9 6 1.5 17 (16) 29 (29) 15 (15) 30 (30) 18 (18) 21.6 6 3.3 108 100 100 100 100 56 81 83 75 73.8 6 6.2 24 7 5 10 11.5 6 4.3 20 (20) 12 (12) 12 (12) 15 (15) 14.8 6 1.9 100 100 100 100 82 66 65 55 66.6 6 5.2 5 14 10 13 10.5 6 3.7 15 (15) 20 (20) 25 (25) 32 (32) 22.9 3 3.7 102 100 100 100 Cells in synchronization Patient Control 1 2 3 4 5 6 Mean Patients at diagnosis 1 2 3 4 Mean Patients during chemotherapy administration 1 2 3 4 5 Mean Patients in remission 1 2 3 4 Mean Patients in relapse 1 2 3 4 Mean Total number of cells SS cells with two single signals; DD cells with two double signals; SD cells with one single and one double signal. Number in parentheses 5 the proportion of cells from the total cell population. A. Amiel et al. / Cancer Genetics and Cytogenetics 125 (2001) 156–160 159 Fig. 1. Replication timing of different lymphoma patients and control group for RB-1 locus. (1) Patients at diagnosis. (2) Patients during cytotoxic treatment. (3) Patients in relapse. (4) Patients in remission. (5) Control group. probes is presented in Tables 2 and 3 and Figs. 1 and 2. A significantly higher proportion of peripheral lymphocytes exhibited an asynchronous pattern of replication at diagnosis, during cytotoxic treatment, and in relapse, as compared to healthy controls (P , 0.01 and 0.01, respectively) and to patients who achieved remission of their lymphoma (P , 0.01 and 0.01 and 0.05, respectively, Tables 2 and 3). Also, a significantly higher degree of asynchronous replication was demonstrated in NHL patients in remission than in healthy controls (P , 0.01). The frequency of DD pattern in each stage of NHL was lower than in the controls with both probes (P , 0.01). Thus, it seems that one allele of these genes replicated later than in the normal controls. 4. Discussion In the current study, we evaluated the replication pattern in lymphocytes of four groups of patients with NHL during different stages of their disease. We found a significantly higher rate of asynchronous pattern of replication at diagnosis, during cytotoxic treatment, and at relapse with both probes RB-1 and HER-2/neu loci than in healthy controls. The rates of asynchronous replication were not different between the study groups with known disease or actively treated patients. However, there was a marked variability in pattern of replication within the groups. The degree of asynchronous replication during remission was lower than during active disease or anticancer treatment, but it did not reach the level of healthy controls. Finally, in patients with various stages of lymphoma, a consistent pattern was found in which one allele replicated later than in healthy controls. The presence of asynchronous pattern of replication at diagnosis of NHL and during relapse is not surprising. We and others have shown that this pattern of replication is found in the peripheral blood lymphocytes of patients with various types of malignancies and even in premalignant conditions like monoclonal gammopathy and in hepatitis C patients with predisposition to lymphoma [7,9,10,13]. Thus, Fig. 2. Replication timing of different lymphoma patients and normal controls for HER-2/neu locus. (1) Patients at diagnosis. (2) Patients during cytotoxic treatment. (3) Patients in relapse. (4) Patients in remission. (5) Control group. 160 A. Amiel et al. / Cancer Genetics and Cytogenetics 125 (2001) 156–160 our findings lend further support to the concept that asynchronous replication of homologous loci is not disease-specific but, rather, related to the abnormal control of replication associated with the malignant phenotype, and that this can be demonstrated also in unaffected tissues [9]. Although the rate of asynchronous replication during treatment was significantly higher than in controls, it was similar to the rate at diagnosis and in relapse. This may be an indication that the abnormal replication pattern during treatment represents an ongoing disease state or the effect of treatment. The effect of chemotherapy on replication pattern may be significant in the acute phases of treatment due to bone marrow depression and proliferate responses that are involved. Indeed, an asynchronous pattern of replication as recently demonstrated after the administration of G-CSF to normal bone marrow donors [14]. In light of this, the replication status should also be examined after discontinuation of cytotoxic treatment to evaluate its long-term effects and relation to the development of secondary leukemias. An interesting observation is the inter-patient variability of the replication pattern in the different stages of lymphoma. It is possible that patients with higher frequency of SD pattern during remission are also at higher risk of relapse. If this finding is confirmed in larger, long-term, prospective studies, it may allow the use of this simple and inexpensive tool (FISH) to closely observe patients who are at high risk for relapse. It was previously demonstrated that homologous alphasatellite sequences in women with predisposition to ovarian cancer and in patients with other premalignant conditions show the abnormal replication pattern of certain genes [10– 12]. Thus, we and others suggested that the delay in the replication of one allele is equivalent to loss of heterozygosity (LOH) caused by allelic deletion or mutation [7–9]. These abnormalities are common in cancer and may be associated with the second hit in the Knudson two-hit model of inactivation of tumor suppressor genes [15]. Whether this phenomenon is related to the well-known increase in the rate of second malignancies in chemotherapy-treated patients cannot be concluded from our findings. A consistent finding in every study group was the lower frequency of the DD pattern when compared to healthy controls. Thus, it seems that one allele replicated later in the malignant situation. It is still unclear whether this represents a mechanism of carcinogenesis such as LOH or if it is a reflection of the loss of control of replication and cell-cycle progression. The current study, although small, demonstrates that an asynchronous pattern of replication can be found in patients with various stages of NHL. The findings should be confirmed in larger studies. For now, our observations open new avenues in the research of the short- and long-term effects of chemotherapy on normal and malignant tissues. They also call for large clinical studies of the appli- cation of this simple method for early detection of relapse or second cancer in high-risk patients. Acknowledgment This work was performed by Danith Blumenthal in partial fulfillment of the M.D. thesis requirements of the Sackler Faculty of Medicine, Tel-Aviv University. References [1] Holmquist GP. Role of replication time in the control of tissue specific gene expression. Am J Hum Genet 1987;40:151–73. [2] Haton KS, Dhar VH, Brown EH. Replication program of active and inactive multigene families in mammalian cells. Mol Cell Biol 1988; 8:2149–58. [3] Hansen RS, Canfield TK, Lamb MM, Gartler SM, Laird CD. Association of fragile X syndrome with delayed replication of the FMRI gene. Cell 1993;73:1403–9. [4] Selig S, Okumura K, Ward DC, Cedar H. Delineation of DNA replication time zones by fluorescence in situ hybridization. EMBO J 1992; 11:1217–25. [5] Boggs BA, Chinault AC. Analysis of DNA replication by fluorescence in situ hybridization. Methods 1997;13:259–70. [6] Kitsberg D, Selig S, Brandoir M, Simon I, Keshet I, Driscoll DJ, Nicholls RD, Cedar H. Allele-specific replication timing of imprinted gene regions. Nature 1993;364:459–63. [7] Amiel A, Litmanovitch T, Gaber E, Lishner M, Avivi L, Fejgin M. Asynchronous replication of p53 and 21q22 loci in chronic lymphocytic leukemia. Hum Genet 1997;101:219–22. [8] Amiel A, Litmanovitch T, Lishner M, Mor A, Gaber E, Tangi I, Fejgin M, Avivi L. Temporal differences in replication timing of homologous loci in malignant cells derived from CML and lymphoma patients. Genes Chromosom Cancer 1998;22:225–31. [9] Dotan ZA, Dotan A, Litmanovitch T, Ravia Y, Oniashvili N, Leibovitch I, Ramon J, Avivi L. Modification in the inherent mode of allelic replication in lymphocytes of patients suffering from renal cell carcinoma: a novel genetic alteration associated with malignancy. Genes Chromosom Cancer 2000;27:270–7. [10] Amiel A, Kigner I, Gaber E, Manor Y, Fejgin M, Lishner M. Replication pattern in cancer: asynchronous replication in multiple myeloma and in monoclonal gammopathy. Cancer Genet Cytogenet 1999;108: 32–7. [11] Amiel A, Kolodizner T, Fishman A, Gaber E, Klein Z, Beyth Y, Fejgin MD. Replication pattern of the p53 and 21q22 loci in the premalignant and malignant stages of carcinoma of the cervix. Cancer 1998;83:1966–71. [12] Litmanovitch T, Altaras MM, Dotan A, Avivi L. Asynchronous replication of homologous alpha-satellite DNA loci in man is associated with non-disjunction. Cytogenet Cell Genet 1998;81:26–35. [13] Amiel A, Kitay-Cohen Y, Fejgin MD, Lishner M. Replication status as a marker for predisposition for lymphoma in patients with chronic hepatitis C with and without cryoglobulinemia. Exp Hematol 2000; 28:156–60. [14] Nagler A, Toren A, Samuel S, Avivi L, Korenstein A. Effect of granulocyte colony stimulating factor (GCSF) administration for stem cell mobilization on temporal order of allelic replication. Blood 1999;94:605. [15] Knudson AG. Antioncogenes and human cancer. Proc Natl Acad Sci USA 1993;90:10914–21.