Rearrangements of Chromosome Band 1p36 in Non-Hodgkin’s Lymphoma¹

NHLs³ are a heterogeneous group of lymphoid neoplasms that display distinct morphological, immunological, cytogenetic, and molecular genetic features. Consistent structural chromosome aberrations in NHL have provided substantial insight into the genetic mechanisms of lymphomagenesis (1, 2, 3, 4). The majority of NHL cases have clonal chromosome aberrations; however, tumors with sole karyotypic changes are relatively rare, constituting only 10–30% of the cases reported in large studies (5, 6, 7). Despite the complexity of the karyotypes, the recurrence of some karyotypic abnormalities in particular histopathological subtypes of NHL has been established. These include t(14;18) in follicular lymphomas, t(11;14) in mantle cell lymphomas, t(3;14) in diffuse large B-cell lymphomas, and t(8;14) in Burkitt’s lymphoma (8, 9, 10). All of these rearrangements are balanced translocations, which involve the immunoglobulin heavy chain gene mapped to 14q32. The juxtaposition of different chromosomal loci, which bear oncogenes, to the actively transcribing immunoglobulin locus results in the activation of the proto-oncogenes and contributes to malignant transformation (11, 12, 13). The chromosomal translocations in NHL have, thus, been helpful in understanding oncogene activation via chromosomal rearrangement in neoplasia. In recent years, much attention has also been attracted to the role of tumor suppressor genes in malignant transformation. Deletion or loss of an entire chromosome or unbalanced structural rearrangements may lead to the inactivation or loss of a tumor suppressor gene(s). Independent lines of evidence suggest a role of the short arm of chromosome 1 in the suppression of biological parameters related to malignancy (14).

Rearrangements of the distal segments of chromosome 1p are found in >10% of NHL cases (15, 16). However, because they are frequently reported to be in concert with other chromosomal changes, they are considered to be secondary abnormalities in NHL (17). Investigators have referred to a wide range of breakpoints between 1p32–36 (18, 19, 20). The purpose of this report was to focus on the involvement of chromosome band 1p36 in NHL and to determine the translocation partner chromosomes involved in the rearrangements with this chromosomal region. The data presented here are from a single institution-based study performed on pretreatment NHL cases. This is the first report delineating the cases with pretreatment karyotypes exhibiting abnormalities of chromosome band 1p36, a region that includes many candidate tumor suppressor gene(s) (21, 22, 23, 24, 25, 26, 27). Because 1p36 alterations seldom occur as a sole abnormality and most often occur with other NHL-specific clonal abnormalities, these alterations are presumably a secondary change due to or leading to the disease. In an ongoing effort, we have used FISH to identify the origin of the translocation partner(s) on the derivative chromosome 1, and these data are included here.

Between January 1982 and December 1996, >1000 consecutive specimens (of which 850 contained abnormal karyotypes) were histologically confirmed as NHL at University of Nebraska Medical Center or other hospitals in Nebraska and western Iowa. The lymphomas have been classified according to the International Working Formulation (28). Biopsy materials were divided for histopathological, cytogenetic, and immunological analysis as described previously (29). B- or T-cell lineage has been assigned based on the results of immunophenotypic and/or immunogenotypic analysis. Cytogenetic studies in all these previously untreated cases have been performed following the protocol described previously (29).

Briefly, chromosome preparations were obtained from lymph node biopsies according to conventional methods following 24-h short-term culture of mechanically disaggregated cells at 37° in RPMI 1640 (Irvine Scientific, Santa Ana, CA) including 20% fetal bovine serum and antibiotics. The cultures were performed without the use of mitogens. One h before the initiation of harvest, the cells were exposed to colcemid (0.05 μg/ml; Irvine Scientific). Following hypotonic treatment (0.074 m KCl solution for 20 min at 37°), the cells were fixed in freshly prepared fixative, methanol:glacial acetic acid (3:1). After three washes with the fixative, air-dried slides were prepared and aged at 60° overnight. Giemsa banding using Wright’s stain (GW banding) was performed. When available, at least 20 metaphases were analyzed. Otherwise, as many metaphases as were available were characterized. Karyotypes were described according to International System for Human Cytogenetic Nomenclature, and those completed before 1995 have been revised according to the most recent nomenclature system (30). Chromosome abnormalities were defined as clonal if at least two cells had the same structural abnormality or gain of a chromosome or if three cells had loss of a specific chromosome. When an abnormal karyotype contained a known NHL-associated translocation [e.g., t(14;18)(q32;q21)] and was observed in only one cell, we included it in the nomenclature. Also, even when a single normal karyotype was observed, it was included in the nomenclature.

FISH was performed on fresh slides that were cytogenetically analyzed and/or in some cases slides prepared from frozen fixed cell pellets, which have been maintained in our cell bank. Slides that were previously stained with GW staining procedures for cytogenetic analysis were destained by sequential washes in 70, 80, and 90% cold ethanol series for 10 min each. In cases for which fixed cell pellets were available, fresh slides were prepared using the cell suspension and aged following routine procedures for FISH (30 min in 2× SSC, pH 7.0, at 37°). For each specimen, the selection of the probes to be used was dependent upon the most likely source of chromosome material as determined by cytogenetics. Commercially available chromosome painting probes, centromeric probes, cosmid probes, and satellite probes (Oncor, Inc., Gaithersburg, MD; and Vysis, Inc., Downers Grove, IL) were used for hybridization. Specific probes for centromeric regions of chromosome 1 were used to assist with target chromosome identification and chromosome-painting probes were used to identify the chromosome translocated to 1p36 region. Hybridization was performed following general protocols described for each specific probe used. Briefly, the slides were denatured in 70% formamide-2× SSC for 2 min at 72° and then sequentially dehydrated. The probe mixture was denatured according to the manufacturer’s protocol (Oncor or Vysis) before it was added to the target DNA and hybridized overnight. To circumvent the separate denaturation processes for the target DNA and the probes, we have also successfully used the Hybrite system (Vysis) for codenaturation using the program specified by the manufacturer for whole chromosome painting probes (WCP). Depending on the probes used (direct- or indirect-labeled), the postwashes and detection (in the case of indirect-labeled probes) were also carried out following the manufacturer’s protocol. Following the hybridization and post wash process, the slides were counterstained with 4,6-diamino-2-phenylindol hydrochloride (DAPI; 0.1 μg/μl). Analysis of the hybridization signals was performed on a Zeiss Axioscope equipped with appropriate filters and imaged with the Cytovision System for FISH and comparative genomic hybridization (Applied Imaging, Pittsburgh, PA). In the event of failure to determine the translocation partner chromosome with the specific probes used and when the available sample was small, slides were subjected to subsequent denaturation and hybridization with probes for other chromosomes.

Of the 850 pretreatment NHL, cytogenetically abnormal cases analyzed, 25% contained structural abnormalities of chromosome 1. Fifty % of these exhibited rearrangements involving band 1p36, including complete or partial deletions and duplications of 1p. Numerical anomalies of chromosome 1 were not considered in this analysis because no specific breakpoints are involved. Deletions and duplications result in the loss or amplification of a relatively large amount of genetic material encompassing a wide array of breakpoints. We have particularly focused only on those specimens with the karyotypic nomenclature precisely describing the rearrangement of 1p36. We observed that 1p36 rearrangements were significantly higher compared to other breakpoints on chromosome 1. Fig. 1 displays the relative frequency of the breakpoints involved in structural rearrangements of chromosome 1 in 213 cases. Duplications were very rare in the p arm of chromosome 1 and never involved the p36 region. However, deletions of 1p almost always involved the loss of 1p36 (Fig. 2). The karyotypes of the patients included in Fig. 2 are not described here, except for one case, with a deletion of only the p36 region. (Table 1, serial no. 13). In both figures, we have described the number of specimens containing the abnormality; hence, any abnormality, even if it occurred more than once in the same karyotype, has been depicted only once in the figure. When two or more breakpoints were involved in one specimen, all of the breakpoints were noted (39 of 213 cases had more than one abnormality involving chromosome 1).

The clinical characteristics, including age, sex, histological type of NHL, the source of specimens, and the complete karyotypes of the 53 NHL cases with 1p36 rearrangements, are described in Table 1. All cases were B-cell NHLs (except for two cases, indicated in Table 1). With an age range of 19–86 years and a median age of 58 years, this group of patients included 35 males and 18 females. Specimens for cytogenetic analysis were collected from lymph node, bone marrow, spleen, tonsil, gastrointestinal mass, and soft tissue. Abnormalities of band 1p36 were found in low, intermediate, and high-grade NHLs. In 35 of 53 cases (66%), a 1p36 abnormality was present, together with t(14;18)(q32;q21). One specimen had a t(1;3;14)(q32;q23;q32), and in two cases, there was a t(1;14)(p36;q32), involving the 14q32 region. Also, in four other cases (serial nos. 35, 36, 38, and 47), we found either a duplication or translocation of an unidentified chromosome involving band 14q32. Sixty % of the cases (32 of 53) containing 1p36 rearrangements were follicular lymphomas, 6% (3 of 53) were diffuse small cleaved or diffuse mixed lymphomas, and 23% (12 of 53) were diffuse lymphomas with large cell component or immunoblastic lymphomas. Fig. 3 depicts representative karyotypes of the abnormal clone containing a 1p36 abnormality from two cases.

Table 2 describes the reciprocal breakpoints involved in translocation with 1p36. In 31 of 53 specimens, the translocation partners have been characterized, 11 of these were resolved by FISH. With clustering of breakpoints within 1q11–q25, chromosome 1 was the most frequent translocation partner (11 of 31 cases). In one case, a deletion of band 1p36 was observed. Also, a recurrent translocation involving chromosome band 9q13 was observed in four cases. In two cases, a t(1;14)(p36;q32) was observed. Table 3 presents the revised cytogenetic nomenclature for those cases in which FISH technique helped determine the translocation partners. Representative metaphases from the cases in which the FISH technique was used to determine the origin of unidentified segments translocated to 1p36 are shown in Fig. 4.

The incidence of NHL has been increasing during the past 15 years (31, 32). Much is known about the genetic changes occurring in different types of NHL, and yet much remains to be explored. There are many recurring cytogenetic abnormalities that have been associated with particular subtypes of lymphomas; however, the association has not been absolute, and new, cytogenetic, histological, and clinical correlations continue to be investigated. Large single-institution ascertainment of karyotypically abnormal cases continues to reveal novel clinical and histological subsets of lymphoid neoplasia, including NHL (15, 16, 18, 19, 20). These studies have also facilitated new molecular approaches to understand the etiology and clinical behavior of NHL. We have studied the diagnostic karyotypes of 850 NHL cases in which 12% of cases include a rearrangement of band 1p36. These chromosomal alterations include deletions, inversions, and balanced and unbalanced translocations.

Recurrent chromosomal translocations have always served to facilitate the investigation of molecular changes and genes affected as a result of these translocations. Here, we describe three recurrent translocations involving chromosome band 1p36 (Tables 1,2,3). NHLs have shown one of the widest varieties of recurrent chromosomal changes, and these are not frequently seen as sole chromosomal alterations (33). In compiling recurrent abnormalities among hematological malignancies (including lymphomas), Mitelman et al. (33) described only those rearrangements that have been reported in three cases or more. Thirty cytogenetically different balanced recurrent abnormalities have been reported in NHL, and only one of these involved chromosome band 1p36, i.e.t(1;17)(p36;q21). This was reported in three cases, once as a sole abnormality. Also, 180 different recurrent unbalanced abnormal clones have been reported in NHL. These include defined unbalanced translocations resulting in duplications and/or deletions and derivative chromosomes. Only two unbalanced abnormalities were related to band 1p36, i.e., del(1)(p36) and der(1)t(1;1)(p36;q21). These were described in five cases each and never as a sole abnormality (33). Our data suggest that the translocation partners represent various chromosomes, with chromosome 1q11-q25 being the most frequent partner. We described three cases with der(1)t(1;1)(p36;q21) (Table 1, serial nos. 12, 23, and 46). Two other cases showed translocations of 1p36 with bands adjacent to q21, i.e., at q22 and q23, respectively (Table 1, serial nos. 42 and 51; Table 2). We also report another recurrent rearrangement of 1q25 with 1p36 (Table 1, serial nos. 18, 43, and 48). This occurred as a balanced translocation t(1;1)(p36;q25) in two cases and as a der(1)t(1;1)(p36;q25) in one case. All these cases also included a t(14;18)(q32;q21). This emphasizes that 1p36 rearrangements may be secondary alteration, due to or leading to NHL.

The potential of FISH techniques to determine the origin of derivative and marker chromosomes and to improve the accuracy of cytogenetic interpretations is well accepted and is being applied as an adjunct to conventional cytogenetic analysis of many cancers, including lymphomas (34, 35, 36, 37). There are two reports from a group of investigators (34, 35) describing the translocation partners of chromosome band 1p36 determined by FISH. The authors have resolved two and nine NHL cases, respectively; however, no recurrent translocation was reported. From our preliminary data, we are convinced that FISH analysis has helped in revealing the unknown material added to 1p36 and has helped define another recurrent translocation [der(1)t(1;9)(p36;q13)] in NHL (Tables 1,2,3). Offit et al. (38) have shown that clusters of chromosome 9 aberrations involving 9q11–q13 were associated with diffuse lymphomas with a large cell component. However, the karyotypes described in their report did not have a reciprocal breakpoint on 1p36. The four cases showing the recurrent translocation t(1;9)(p36;q13) in our group of patients did not share a common histological subtype.

We observed that duplications of chromosome 1p were very rare and never involved the p36 region, whereas deletions were frequent and almost always included p36 region as part of the deletion (Fig. 2). Chromosome rearrangements, like deletions, duplications, and translocations, may lead to amplification, loss, or disruption of the functional ability of oncogenes and/or tumor suppressor genes located on that chromosomal region. In recent years, much attention has been directed to the role of tumor suppressor genes in malignant transformation (39, 40). Many candidate tumor suppressor genes have been mapped to chromosome 1p36. These include CDC2L1 (p58 or PITSLRE), a cell cycle-regulated kinase gene with homology to CDC2 (21, 22); TNFR2, one of the two tumor necrosis factor receptor genes (23); ID3, a member of the Id family of developmental negative regulatory genes (24); PAX7, the developmental regulator gene (25); and DAN, a transcription factor gene homologous to a mouse tumor suppressor gene (26). A very recent addition to this set of putative tumor suppressor genes on 1p36 is p73, a strong candidate suppressor gene that bears homology to p53 (27).

Besides the cytogenetically visible deletions of chromosome 1p involving the p36 region, we hypothesize that as a molecular consequence of other structural rearrangements involving 1p36, one or more of the putative tumor suppressor gene(s) might be deleted or functionally disrupted. Loss of one or more of these candidate tumor suppressor genes on 1p36 have been extensively studied in neuroblastoma, in which deletion of 1p36 is considered a primary chromosomal change (41, 42). Deletions of tumor suppressor genes on 1p36 have also been investigated in many other cancers, including melanoma (43), intestinal cancer (44), and colon cancer (45), and more recently in ovarian cancer (46); however, there are no such reports on NHL. CDC2L1 is the most distally located candidate tumor suppressor gene on 1p36. Investigations to detect the possible deletion of this region are in progress in our laboratory, and we have observed the deletion of CDC2L1 gene locus among 23 of the 26 cases studied thus far (47). Whether the loss of this region containing putative tumor suppressor gene(s) significantly contributes to the disease progression and survival remains to be determined. It has been reported that the overall survival of the patients bearing breaks at 1p32–36 is shorter than that of the patients without these abnormalities (18). A correlation of breakpoints in this region suggested decreased disease-free survival (48). Breaks in this region, detected at the time of presentation or relapse, have also been correlated with bone marrow involvement in NHL (49). Clinical correlations of patients included in this study are in progress, and the data generated will provide insight into the probable role of 1p36 abnormalities in the survival of these patients. However, from the data compiled in this report, it is clearly evident that chromosome band 1p36 is implicated in NHL.