| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Respirology (B2) and
Department of Pathology, School of Medicine, Chiba University, Chiba, Japan;
Division of Pulmonary and Critical Care Medicine and Bellevue Chest Service and
Department of Pathology, New York University Medical Center, New York, NY 10016
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ig gene somatic hypermutations occur within the germinal center (GC) at a stage in B cell development that follows VH-D-JH rearrangement, and provide the structural correlation for Ag-driven intraclonal selection and affinity maturation. Somatic hypermutations in Ig VH genes occur in the GC and post-C cells. It has been demonstrated that the process of hypermutation remains ongoing in MALT lymphomas because of the presence of intraclonal variations in the Ig VH genes (4). Somatic hypermutations have also been reported to target non-Ig loci, characterized in the 5'-intronic region of the BCL-6 gene locating 
2 kb 3' from the transcription initiation site (5). In a study at the single cell level, it was found that BCL-6 mutations tend to occur in normal B cells containing mutations in their Ig VH genes (6). Mutations in the 5' regulatory region are frequently found in normal GC and memory B cells, as well as in lymphomas displaying GC/post-C phenotypes and harboring mutated Ig VH genes, but not in naive B cells or in other tumor types (7). On the basis of these findings, mutations in BCL-6 are regarded as a marker of B cell transit through the GC.
In this study, we used a PCR and sequence procedure to analyze Ig VH region genes and mutations in the 5' regulatory region of BCL-6 genes in pulmonary lymphoproliferative disorders (MALT lymphomas, HIV-related, EBV-related, and virus-negative LIP). Our data support the view that pulmonary MALT lymphoma and HIV-negative LIP are under the influence of GC hypermutation mechanisms and that BCL-6 mutations in HIV-related LIP result from a different process than that found in other pulmonary lymphoproliferative disorders.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tissue specimens from 40 patients with pulmonary lymphoproliferative disorders were collected from our own files for the years 1982–2002. Cases included 20 patients with pulmonary extranodal marginal zone B cell lymphoma of the MALT type and 20 patients with LIPs (5 HIV-related, 5 EBV-related, and 10 virus-negative) (Table I). The criteria used to identify primary pulmonary lymphoma of the lung were a lack of evidence of disseminated lymphoma at the time of diagnosis, as well as no such evidence for 3 mo after diagnosis (8). When a diagnosis of pulmonary MALT lymphoma was reached, the lesion was subclassified according to the new World Health Organization classification (9). In two cases (cases 5 and 8), mass shadows expanded very slowly and subsequent percutaneous needle biopsy (PCNB) was performed 8 years after the open-chest biopsy (OC) procedure. All HIV-related LIP cases had normal absolute CD4+ T cell numbers classified as group II or III according to the Centers for Disease Control classification (10). Two cases (cases 9 and 10) had a history of HIV-positive parents and another two cases (cases 11 and 13) had a history of transfusion. The risk factors for HIV infection were unknown for case 12 in this study. In HIV-negative patients, two patients had collagen disease (case 17: Sjogren syndrome and case 18: rheumatoid arthritis). Because symptoms (dry cough and dyspnea) had gradually became more severe and the reticulonodular shadows on the chest x-ray had spread to a wider area, three patients (cases 12, 13, and 16) with LIP (2 HIV-related and 1 virus-negative LIP) underwent transbronchial lung biopsy (TBLB) 1–4 years after OC.
|
Materials were fixed in formalin and embedded in paraffin wax. Sections were subjected to H&E and immunohistochemical stainings. Immunohistochemical studies were performed with the following markers: leukocyte common Ag (Dakopatts, Glostrup, Denmark), CD20 (L26; Dakopatts), CD45RA (MB1; Euro-Diagnostica, Malmo, Sweden), CD43 (MT1; Euro-Diagnostica), CD45RO (UCHL1; Dakopatts), and anti-
and 
L chains (Dakopatts).
Microdissection
To increase the concentration of B lymphocyte DNA relative to total tissue DNA, materials were microdissected as follows: a CD20-stained section was visualized under a 5x objective and the nonlymphoid tissue and GCs were carefully scraped off, leaving a zone as small as 1 mm2, rich in the infiltrate of B lymphocytes. Microdissected areas from four sections were then scraped into one tube (11).
DNA extraction and PCR
DNA was obtained from microdissected materials and peripheral lymphocytes using TAKARA DEXPAT (Takara, Kyoto, Japan) and COLLECTAGENE (Takara).
PCR amplification of BCL-6 5' noncoding regions was performed on three partially overlapping fragments (E1.10, E1.11, and E1.12) spanning 742 bp located downstream of the first BCL-6 noncoding region (12). Primers used for PCR analysis of Ig VH genes and the EBV genome have been reported previously (11, 13).
Cloning and sequencing of PCR products
PCR products encompassing fragments of BCL-6 (E1.10, E1.11, and E1.12) and fragments of the Ig VH gene were cloned and ligated to the PCR vector, and the ligation mixture was transformed into One Shot competent cells using a Zero Blunt TOPO PCR Cloning kit (Invitrogen, San Diego, CA). Ligated clones were chosen at random and phage DNA was purified. The inserts in the PCR vector were sequenced by the Dye Primer method using a Taq Dye Primer Cycle Sequencing Core kit (Applied Biosystems, Norwalk, CT). In each sample subjected to DNA sequencing, 20 independent subclones of the Ig VH gene and 40 independent subclones of the BCL-6 gene were analyzed. Sequences were aligned with the BCL-6 hypermutation region germline sequence previously established (GenBank accession number AF 191831) (14). The first nucleotide of the amplified BCL-6 gene region was arbitrarily defined as position +1. A mutation observed more than once among the BCL-6 gene clones from the same tumor specimen was defined as a confirmed mutation. The Ig VH, D, and JH elements were analyzed using the international ImmunoGeneTics (IMGT) database (http://imgt.cines.fr) (15).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Twenty cases of pulmonary MALT lymphoma were classified as extranodal marginal zone B cell lymphoma (B cell lymphoma of MALT type). In all cases, reactive follicles were sparsely present and surrounded by tumor cell infiltrations consisting of small- to medium-sized lymphoid cells. Every case demonstrated lymphoepithelial lesions, which are a characteristic feature of MALT lymphomas. All cases were classified as B cell lymphoma on the basis of the expression of at least two B cell markers.
All 20 LIP samples demonstrated dense and mixed interstitial infiltrates of lymphocytes, plasma cells, and histiocytes with prominent GCs. No cytologic atypia of the lymphoid infiltrates was seen. Immunohistochemical staining of the specimens revealed polyclonal B cell populations with numerous T cells.
Ig VH region sequences of lymphoproliferative disorders with BCL-6 mutations
Five LIP case were defined as EBV-positive because of the amplification of the EBV-genome demonstrated by PCR analysis. Ig VH gene sequences of all eight cases of pulmonary MALT lymphoma with BCL-6 mutations showed closely related sequences, indicating amplification from dominant lymphoma cell clones. The closest germline Ig VH genes and their degree of similarity to each case are shown in Table II. For all eight cases of primary pulmonary lymphoma, a single major Ig VH gene sequence was identified with rates of 12/20 to 17/20. Six lymphoma clones expressed Ig VH genes derived from the VH3 family, whereas the other two derived from the VH4 family. Five lymphoma Ig VH genes (patients 1–5) showed evidence of intraclonal heterogeneity. In contrast, the other three lymphoma Ig VH genes (patients 6–8) demonstrated a lack of intraclonal heterogeneity.
|
In pulmonary MALT lymphoma and LIP (HIV-related, EBV-related, and HIV-negative) cases with BCL-6 mutations, the respective lengths of complementarity determining region 3 averaged 42 bases (range: 30–60 bases), 52 bases (range: 42–60 bases), 49 bases (range: 45–54 bases), and 38 bases (range: 30–51 bases) without a major contribution of the D region, respectively. JH4 was frequently found in pulmonary MALT lymphoma (50%), LIP (HIV-related (33%), EBV-related (50%), and HIV-negative (66%)) cases. Another frequently found JH region was JH6 (40%) in HIV-related LIP.
Sequencing analysis of the BCL-6 5' noncoding regions of pulmonary lymphoproliferative disorders
Eight of 20 (40%) pulmonary MALT lymphoma and 10 of 20 LIP (5 of 5 (100%) HIV-related, 2 of 5 (40%) EBV-related, and 3 of 10 (30%) virus-negative LIP) cases demonstrated BCL-6 mutations. The results of sequencing analysis of the BCL-6 mutations are presented as the rates of incidence of the predominant sequences of the total number of vector clones analyzed (Table III). Five (cases 1–5) of eight pulmonary MALT lymphoma cases with BCL-6 mutations showed evidence of intraclonal heterogeneity, i.e., clones of a given sample differed from the others in one or more nucleotides. Another three cases (cases 6–8) demonstrated BCL-6 mutations without intraclonal heterogeneity, assuming that these mutations represent single events and not independent mutations.
|
|
In two MALT lymphoma cases (patients 5 and 8), subsequent PCNB specimens (8 years later) demonstrated monomorphous lymphocytic infiltration suggestive of MALT lymphoma. These lymphoma cells were similar to those observed in the previous OC specimens. In one case (patient 5), some nucleotide substitutions of major Ig VH genes derived from lymphoma clones were found between the surgical resection (SR) and PCNB specimens. This case showed continuous somatic mutations of the BCL-6 gene in the first and second biopsy materials (Table V). In patient 8, Ig VH genes of the lymphoma cell clones derived from the OC and PCNB were identical. No continuous somatic mutations of the BCL-6 gene were found in SR and PCNB specimens.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because single strand conformational polymorphism is less sensitive for the detection of mutations than sequencing analysis and may underestimate the extent of the mutational process, we applied a sequencing approach to clarify the suggestive link between the occurrence of mutations in the Ig VH and BCL-6 genes in lymphoproliferative disorders. As the use of high-fidelity DNA polymerases in PCR is essential for reducing the introduction of amplification errors in the PCR products, we used a high-fidelity polymerase, Pfu. Unlike TaqDNA polymerase, Pfu possesses a 3' to 5' exonuclease proofreading activity that enables the polymerase-generated PCR fragments to have fewer errors than Taq-generated PCR inserts. The error rate for Pfu has been reported to be one mutation per 7.8 x 105 bp per cycle (17).
The percentage of MALT lymphomas carrying BCL-6 mutations reported here is not significantly different from that reported in a previous study (18). When examining the relationship between Ig VH genes and BCL-6 mutations, the analysis of this relationship is particularly informative because these somatic hypermutations are useful in pulmonary MALT lymphoma for defining the molecular heterogeneity of the entity in cases of pulmonary MALT lymphoma. Intraclonal heterogeneity of BCL-6 mutations was observed only in pulmonary MALT lymphoma cases (cases 1–5) whose Ig VH genes also showed intraclonal heterogeneity. To determine whether BCL-6 mutations persist over time at the site of MALT lymphoma lesions, we had the opportunity to study PCNB specimens (case 5) obtained 8 years after OC, and were able to show continuous somatic mutations of the BCL-6 gene in the first and second biopsy materials. Considering that the BCL-6 gene can be induced to hypermutate in vitro in human B cells by the same stimuli required for the induction of Ig VH gene hypermutation (19) and that Ig VH gene hypermutation requires the microenvironment of the GC and will not occur in marginal zones (20), the ongoing BCL-6 mutations might reflect re-entry into a GC pathway to further mutations. Previous studies have demonstrated that the BCL-6 gene can act as a histogenetic marker of B cell transition through the GC in analyses of lymphomas. These ongoing BCL-6 mutations suggest that pulmonary MALT lymphoma cells are influenced by GC-hypermutation mechanisms over a long time period, and that these GC reactions may play a role in the pathogenesis of pulmonary MALT lymphomas. In contrast, three lymphoma cases (cases 6–8) without intraclonal heterogeneity of the Ig VH gene also demonstrated no intraclonal heterogeneity of the BCL-6 mutations. These lymphoma cells might not undergo re-entry into the GC environment and might no longer be susceptible to the hypermutation mechanism. Sequencing of the PCR products from the OC materials and the subsequent PCNB specimen obtained from case 8 confirmed this possibility. Because all pulmonary MALT lymphoma cases except two (cases 5 and 8) have remained asymptomatic without recurrence since SR (3–6 years), long-term follow-up is especially important for determining whether outcome variables will differ in patients with pulmonary MALT lymphomas with ongoing and nonongoing mutations in the BCL-6 gene.
In pulmonary MALT lymphomas, the pattern of BCL-6 mutation is similar to that of the Ig VH gene (21): 1) a preference for single base pair substitutions with a small number of deletions; 2) transition mutations more frequent than transversions, as is observed in somatic hypermutation of Ig VH genes (16, 22); and 3) preferential targeting of RGYW/WRCY motifs by the mutational machinery of BCL-6 genes. Randomly occurring point mutations would be expected to be one-third transitions and two-thirds transversions, leading to a value of 0.5 for randomly occurring mutations. BCL-6 point mutations derived from pulmonary MALT lymphoma showed a bias for transitions over transversions and standard polarity. It has been suggested that the quadruplet motif RGYW and the inverse repeat WRCY are targets for increased mutational activity in the hypermutation of Ig VH genes. This motif targeting is a striking feature of somatic hypermutations of the Ig VH gene, especially within the CDR regions of the Ig VH genes (22). BCL-6 mutations of pulmonary MALT lymphomas demonstrate preferential targeting in the consensus sequence RGYW and the inverse repeat WRCY (16, 23). These consensus sequences have been identified as mutational hotspots in human Ig VH genes and in the BCL-6 gene of human GC B cells or in diffuse large cell lymphoma (6).
After establishing sequencing techniques for the BCL-6 gene, we analyzed the sequences in HIV-related and virus-negative LIP. The findings of this study are consistent with the different immunological situations of HIV-related and virus-negative LIP. We detected BCL-6 mutations in 5 of 5 (100%) patients with HIV-related LIP and 3 of 10 (30%) virus-negative LIP patients. In GC B cells, the frequency of BCL-6 mutations was found to be 20- to 100-fold lower than that of mutations in the Ig VH-D-JH gene segment, suggesting a relative inefficiency in the somatic hypermutation process of BCL-6. However, the frequency of BCL-6 gene mutations is higher in HIV-related LIP than in virus-negative LIP. Not only pulmonary MALT lymphomas but also virus-negative LIP cases have some features of Ig VH gene mutations. In virus-negative pulmonary lymphoproliferative disorders (pulmonary MALT lymphomas and HIV-negative LIP), BCL-6 genes are likely to be mutated by B cell hypermutational mechanisms. Consistent with this, a recent study demonstrated that mutations in Ig VH genes and BCL-6 occurred in parallel in a B cell lymphoma line stimulated in vivo (19). In contrast, BCL-6 mutations in HIV-related LIP lack the predominance of transitions over transversions, lose preferential motif (RGYW/WRCY) targeting, and may not be governed by the same mechanisms that account for hypermutations in Ig VH genes. The loss of the characteristics of Ig VH gene hypermutation in BCL-6 genes in HIV-related LIP may be the very important evidence that BCL-6 mutations in HIV-related LIP result from a process different from that found in virus-negative pulmonary lymphoproliferative disorders. Our results indicate that mutations in the 5' noncoding regions of BCL-6 may represent genetic instability in HIV-related LIP and that the immunological reaction in HIV-related LIP is different from that in virus-negative pulmonary lymphoproliferative disorders. To determine whether this finding is unique to HIV or whether other viral infections induce similar effects, we analyzed BCL-6 mutations in five patients with EBV-related LIP. Two of the five patients with EBV-positive LIP showed BCL-6 mutations with a predominance of transitions over transversions and with preferential targeting for RGYW/WRCY motifs. A high frequency of BCL-6 mutations may be unique to HIV-related LIP. More precise molecular analyses are required to determine whether this is the case.
TBLB specimens were obtained from three patients with LIP (two HIV-related and one virus-negative) 1–4 years after OC. By analyzing these TBLB specimens, BCL-6 mutations were found to be ongoing in all three LIP cases. Loss of the characteristics of Ig VH gene hypermutation in the BCL-6 genes was demonstrated in the HIV-related LIP cases but not in the virus-negative LIP case. This finding lends further support to the concept that aberrant immunological reactions remain unchanged over the years, although a much longer follow-up period is required to confirm this conclusion.
This study demonstrates that the intraclonal heterogeneity of BCL-6 mutations is observed only in pulmonary MALT lymphoma cases whose Ig VH genes also show intraclonal heterogeneity, and that BCL-6 mutations in HIV-related LIP result from a process different from that found for HIV-negative pulmonary lymphoproliferative disorders.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Katsushi Kurosu, Department of Respirology (B2), Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan. E-mail address: kurosu{at}faculty.chiba-u.jp
3 Abbreviations used in this paper: LIP, lymphocytic interstitial pneumonia; MALT, mucosa-associated lymphoid tissue; GC, germinal center; PCNB, percutaneous needle biopsy; OC, open-chest biopsy; TBLB, transbronchial lung biopsy; SR, surgical resection.
Received for publication August 27, 2003. Accepted for publication March 3, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
