Abstract
Background: The mechanisms promoting malignant transformation from chronic Helicobacter pylori-gastritis to gastric extranodal marginal zone lymphoma (MALT lymphoma) are insufficiently characterized. This follow-up study aimed to validate candidate microRNAs (miRs) in the process of neoplastic transformation. Materials and Methods: MicroRNA expression signatures (n=20) were generated for a total of 60 cases of gastric lesions ranging from Wotherspoon 0-5 employing a quantitative real-time polymerase chain reaction (PCR) approach. Morphological and immunohistochemical characterization of the cohort was supplemented by PCR-based immunoglobulin heavy chain recombination studies. Results: Quantitative expression of miR-150, miR-142.3p, miR-375 and miR-494 was significantly de-regulated in samples from MALT lymphoma compared to those from gastritis. Conclusion: The previously reported up-regulation of miR-150 in marginal zone lymphoma of MALT type was verified in an independent cohort of lymphoma samples employing a modified methodology. This further substantiates the role of miR-150 as a potential oncomiR in MALT lymphoma.
Extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT) type constitutes a rare form of low-grade non-Hodgkin's lymphoma of B-cell lineage, with an annual incidence of approximately 0.5 to 1 per 100,000 (1, 2). Despite the gastrointestinal tract (stomach) being the most common site of manifestation (>50%) other localizations including skin, ocular adnexa and lung have been reported (3). It has been shown that Helicobacter pylori is an important stimulus in the pathogenesis of gastric MALT lymphoma (4, 5) and similar pathophysiological mechanisms have been proposed for ocular adnexal MALT lymphoma and Chlamydia psitaci, immunoproliferative small intestinal disease (IPSID) and Campylobacter jejuni, MALT lymphoma of the fallopian tube and Acinetobacter spp. as well as cutaneous MALT lymphoma of the skin and Borrelia burgdorferi (6, 7). Histopathological diagnosis of gastric MALT lymphoma in the context of morphological overlap between reactive or inflammatory changes during progressive malignant transformation is considered difficult. Most commonly, the score-based classification system as proposed by Wotherspoon et al. supplemented by molecular clonality assessment of the B-cell population is employed (8, 9). In cases of Helicobacter pylori-associated gastric MALT lymphoma, H. pylori eradication therapy can induce sustained remission (10, 11). The most frequently detectable chromosomal aberration in gastric MALT lymphoma [t(11;18)(q21;q21)] affecting the MALT1 gene locus, however, coincides with resistance to H. pylori eradication therapy (11). Both radiation and local surgical treatment are viable options often followed by complete remission or prolonged disease-free intervals in these cases (3). Despite the well-established link between H. pylori-associated gastritis and MALT lymphoma, the mechanisms driving the progressive transformation of normal mucosa to lymphoma are insufficiently characterized at the molecular level. Over the past decade, microRNAs have emerged as a key element in the epigenetic regulation of gene expression (12). An involvement of these small non-coding RNAs in numerous biological processes, including cell growth, differentiation, apoptosis and pathogenesis of malignant neoplasia, by down-regulating one or several genes by means of translational repression and target degradation has been identified (13, 14). Quantitative expression of microRNAs appears to differ between normal and neoplastic tissues and it is believed that specific microRNA profiles exist for all types of tissues and tumors. Moreover, distinct microRNA signatures have been characterized for certain types of inflammation, including H. pylori-associated gastritis (15). Additionally microRNAs are valuable as prognostic factors in various types of malignancies (16).
Histological scoring of lymphoid infiltrations in the gastric mucosa according to Wotherspoon et al. (9).
In a recent study, we identified microRNA expression signatures of gastric biopsies with lesions corresponding to Wotherspoon scores 0 to 5 in order to identify microRNAs which might be involved in the process of neoplastic transformation. Quantitative aberrations of several microRNAs were associated with the presence of a predominantly lymphocytic inflammatory infiltrate (e.g. miR-566 and miR-212) or an underlying H. pylori infection (e.g. LET7f) (15). A set of five microRNAs, namely miR-150, miR-550, miR-124a, miR-518b and miR-539, however, was revealed to be differentially expressed in gastritis as opposed to MALT lymphoma (17). An elevated expression of miR-155 was detectable in a subset of patients, which at the time we attributed mainly to the presence of an underlying H. pylori infection in accordance with the literature (15). Recent findings by Saito et al. however suggest that these expression levels may also indicate an elevated degree of resistance towards eradication therapy (18). We reasoned that these aberrantly expressed microRNAs are a potential key element in the process of malignant transformation from chronic inflammation to lymphomagenesis.
Emphasizing on the need for validation of our initial findings, in the present study, we generated microRNA signatures in a comprehensive and independent cohort of gastric MALT lymphoma samples employing an alternate methodological approach alongside a modified statistical normalization.
Materials and Methods
Patients. Formalin-fixed and paraffin-embedded (FFPE) tissue biopsy samples from a total of 60 cases were retrieved from the registry of the Reference Center for Lymph Node Pathology and Hematopathology, University Hospital of Schleswig-Holstein, Campus Luebeck. All samples were collected as part of standard clinical care and the local ethics committee has approved the study.
MicroRNAs investigated in the current study.
Two experienced hematopathologists without knowledge of microRNA expression data reassessed all cases for independent pathology review. As described previously, all biopsies were re-evaluated and scored according to the criteria shown in Table I (17).
Immunohistochemistry. Immunohistochemical studies were performed on FFPE sections according to a standard, three-step immunoperoxidase technique using an automated TechMate system (DAKO, Glostrup, Denmark) and the BrightVision Kit (ImmunoLogic, Duiven, the Netherlands). Conventional stainings (hematoxylin and eosin, and Giemsa) were supplemented by immunohistochemical reactions for CD20, CD3, CD23, CD10, cyclin D1, Ki67 and light chains (kappa and lambda).
Immunoglobulin heavy chain recombination studies. In cases of non-clonal light chain expression by means of immunohistochemical reactions, immunoglobulin heavy chain gene rearrangement was assessed for clonality in two DNA samples separately obtained from FFPE specimen by Polymerase Chain Reaction (PCR) and subsequent capillary gel electrophoresis according to the BIOMED2 protocol as described for cases classified as Wotherspoon score 3 or 4 (19).
Definition of pooled tumor groups.
Nucleic acid isolation and cDNA synthesis. Total RNA was isolated from four 20-μm sections of FFPE tissues using the Ambion Recover All kit (Ambion, Austin, TX, USA) according to the manufacturer's protocol. In order to assess microRNA expression, a total of 10 ng of RNA was reverse-transcribed using the TaqMan® universal PCR master mix, No AmpErase® UNG-kit and the TaqMan® microRNA reverse transcription kit from Applied Biosystems (Foster City, CA, USA). All reactions were diluted 1:60 and stored at -20°C.
Quantitative real time-PCR and data analysis. Primers and probes for quantification of microRNA expression were obtained from Applied Biosystems. Expression signatures of 20 microRNAs (Table II) and RNU6b were generated in triplicate employing a quantitative real-time PCR (RT-PCR) approach on a LightCycler® 480 System (Roche, Mannheim, Germany) according to manufacturer's instructions.
MicroRNA expression data were normalized using expression of RNU6b in triplicate. Cases with more than 15% missing values (n=1) or equivocal data on clonality for Wotherspoon score 3 and 4 (n=2) were excluded.
To identify microRNAs with changed expression values in biopsies with increasing Wotherspoon scores, Jonckheere-Terpstra trend tests (R-package “clinfun” R version 2.11.1) were performed. These analyses were also performed for samples of Wotherspoon scores 2 to 5. To compare expression values of microRNAs in gastritis (Wotherspoon scores 1, 2, and polyclonal 3 and 4) and lymphoma (monoclonal scores 3, 4, and 5), Wilcoxon-Mann-Whitney tests were used.
Results
Morphological, immunophenotypic and molecular features of the study group. By means of morphological (hematoxylin and eosin, and Giemsa) and immunohistochemical investigations (CD20, CD3, CD23, CD10, cyclin D1 and MIB1) the cases enrolled in the present study were classified in accordance with the scoring criteria proposed by Wotherspoon et al. (9). Employed criteria and case distribution for the current study according to the Wotherspoon system are briefly summarized in Table I.
Clonality analyses of tissues of Wotherspoon scores 3 and 4 by means of immunohistochemistry for light chains kappa and lambda and by PCR for the immunoglobulin heavy chain gene locus according to BIOMED-2 protocols revealed clonality in 2/10 and 6/10 cases, respectively.
Expression of microRNAs in chronic gastritis and MALT lymphoma samples. All microRNAs investigated in the current study were shown to be constitutively expressed in gastric biopsy specimen biopsies in all cases included in the study (n=59).
Changes in microRNA expression signatures concur with malignant transformation from chronic H. pylori-associated gastritis to MALT lymphoma. MicroRNA signatures differ significantly between gastric biopsies with different Wotherspoon scores. In summary, four of the investigated microRNAs showed altered expression levels from score 0 to 5. Out of these, expression levels of two microRNAs decreased (miR-150 and miR142-3p), whereas expression levels of two microRNAs increased (miR-375 and miR-474) (see Table IV and Figure 1).
Comparative analysis of gastritis (Wotherspoon scores 1, 2, and polyclonal 3, 4) and MALT lymphoma (Wotherspoon score 5 and monoclonal scores 3 and 4) showed four microRNAs, namely miR-150, miR-494, miR-124a and miR-142-3p, to be differentially expressed after adjusting for multiple testing (see Table V and Figure 2).
Discussion
Extranodal marginal zone lymphoma of MALT type is recognized as a distinct entity by the WHO classification and is in many instances associated with chronic antigenic stimulus and chronic inflammation (2). Pathogenetically, both infection (H. pylori and Chlamydia psitaci) and autoimmune disorders, such as Sjögren syndrome and Hashimoto thyroiditis, has been shown to be of relevance. The causal link between an inflammatory stimulus and B-cell lymphomagenesis is best established for gastric MALT lymphomas and H. pylori (5). Seeking to validate and verify our initial findings of miR-150, miR-550, miR-124a, miR-518b and miR-539 being differentially expressed in gastritis as opposed to MALT lymphoma, we performed RT-PCR-based expression profiling of 20 microRNAs selected due to their previously published expression levels in an independent cohort of 60 gastric MALT lymphomas (17). Deviating from the initial setting normalization was performed by applying the deltaCT method using the quantitative expression of RNU6 as a housekeeping microRNA. In the current study, we were able to reproduce our initial findings of an increased expression of miR-150. The aberrantly-elevated expression of miR-150 remains remarkable in the light of this microRNAs previously postulated role as a tumor suppressor in the hematopoietic system. We were able to identify significant down-regulation of miR-150 in diffuse large B-cell lymphoma but not follicular lymphoma (20-22). Thus, miR-150 was expected to be down-regulated in lymphoma (23-26). Recently, Monsalvez et al. identified a similar phenomenon in primary cutaneous marginal zone lymphoma, however, the authors found low levels of miR-150 to be associated with shortened overall survival (27). Moreover, Cai et al. made corresponding observations for primary conjuctival MALT lymphoma (28). This leaves room for speculation as to the functional consequences of increased miR-150 expression as it appears to be a molecular hallmark of marginal zone lymphomas regardless of their site of primary manifestation, while at the same time it positively affects patient outcome. The role of miR-150 as a key regulatory element in the differentiation of B- and T-cells has been established in the past. In light of our present findings, alongside previously published results from Wu et al. describing high miR-150 expression in gastric carcinoma, we reason that the localization of a malignant tumor may vitally influence the oncogenic or tumor-suppressive behavior of miR-150 (25). Our initial findings regarding aberrant expression of miR-550, miR-124a, miR-518b and miR-539 failed to reach statistical significance in the current study, indicating a variable degree of contribution to MALT lymphoma pathogenesis. Unlike the results obtained in our initial study, we here demonstrate miR142-3p as another microRNA significantly increased in expression, whereas miR-375 and miR-474 decreased in expression. These findings had merely bordered on statistical significance in our previous investigations. Integrating results from both cohorts, a role of these microRNAs can be presumed. Further studies, integrating functional analyses, are needed in order to validate this assumption. In summary, we were able to underline the pathogenetic implications of miR-150 in gastric extranodal marginal zone lymphomas and its pre-malignant precursor lesions, as elevated expression was detectable in both our initial study as well as the present validation study employing modified methodology and statistical normalization approaches. Although these findings emphasize on the role of miR-150 as an oncomiR in gastric MALT lymphoma, they also necessitate the elucidation of molecular mechanisms by which miR-150 drives malignant transformation of pre-neoplastic lymphocytic lesions. It is tempting to speculate about a potential implication for novel targeted therapy approaches.
Trend in CT values for different groups of Wotherspoon score for miR-150. Boxes range from 25th to 75th percentile and include median values. Whiskers indicate range and circles display outliers.
Significant microRNAs (Wotherspoon groups), median difference (ΔCT values) for groups 0 and 5 with corresponding confidence interval (CI) and unadjusted p-value.
Significant microRNAs (pooled groups), median difference (ΔCT-values) for groups 2 and 3 (see Table III) with corresponding confidence interval (CI) and unadjusted p-value.
Boxplots for pooled groups (1 - normal, 2 - gastritis and 3 - gastric mucosa associated lymphoid tissue lymphoma) and miR-150. Boxes range from 25th to 75th percentile and include median values. Whiskers indicate range and circles display outliers.
Acknowledgments
This work was supported by Deutsche Forschungsgemeinschaft“ Grant-No. TH 813/4-1 to C.T. We thank Annette Aufseß for excellent technical assistance.
Footnotes
-
Declaration of Interests
-
The Authors declare that there are no financial or other potential conflicts relevant to the manuscript.
- Received January 29, 2014.
- Revision received February 17, 2014.
- Accepted February 18, 2014.
- Copyright © 2014 The Author(s). Published by the International Institute of Anticancer Research.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.