Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Lymphoma

Differential expression of NF-κB target genes in MALT lymphoma with and without chromosome translocation: insights into molecular mechanism

Abstract

Mucosa-associated lymphoid tissue (MALT) lymphoma is characterized by t(11;18)(q21;q21)/API2-MALT1, t(1;14)(p22;q32)/BCL10-IGH and t(14;18)(q32;q21)/IGH-MALT1, which commonly activate the nuclear factor (NF)-κB pathway. Gastric MALT lymphomas harboring such translocations usually do not respond to Helicobacter pylori eradication, while most of those without translocation can be cured by antibiotics. To understand the molecular mechanism of these different MALT lymphoma subgroups, we performed gene expression profiling analysis of 21 MALT lymphomas (13 translocation-positive, 8 translocation-negative). Gene set enrichment analysis (GSEA) of the NF-κB target genes and 4394 additional gene sets covering various cellular pathways, biological processes and molecular functions have shown that translocation-positive MALT lymphomas are characterized by an enhanced expression of NF-κB target genes, particularly toll like receptor (TLR)6, chemokine, CC motif, receptor (CCR)2, cluster of differentiation (CD)69 and B-cell CLL/lymphoma (BCL)2, while translocation-negative cases were featured by active inflammatory and immune responses, such as interleukin-8, CD86, CD28 and inducible T-cell costimulator (ICOS). Separate analyses of the genes differentially expressed between translocation-positive and -negative cases and measurement of gene ontology term in these differentially expressed genes by hypergeometric test reinforced the above findings by GSEA. Finally, expression of TLR6, in the presence of TLR2, enhanced both API2-MALT1 and BCL10-mediated NF-κB activation in vitro. Our findings provide novel insights into the molecular mechanism of MALT lymphomas with and without translocation, potentially explaining their different clinical behaviors.

Introduction

Extranodal marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue (MALT) originates from the MALT acquired as a result of chronic inflammatory or autoimmune disorders.1 The etiological factors underlying these chronic inflammatory disorders have a pivotal role in MALT lymphomagenesis. This is best exemplified by the causative role of Helicobacter pylori infection in gastric MALT lymphoma as shown by the compelling evidence from the epidemiological, laboratory and particularly clinical studies, which show long-term complete remission of the lymphoma following H. pylori eradication in 70% of cases.1 In spite of this, the molecular mechanisms underlying the lymphoma development are not fully understood. Stimulations of antigen receptor by autoantigen, and co-stimulatory molecule CD40 by H. pylori-specific T cells are believed to have an important role. Recent studies on MALT lymphoma-associated chromosome translocations provide further insights into its molecular pathogenesis.

t(11;18)(q21;q21)/API2-MALT1, t(1;14)(p22;q32)/BCL10-IGH and t(14;18)(q32;q21)/IGH-MALT1 are specifically associated with MALT lymphoma albeit occurring at considerately variable frequencies in different anatomic sites.2, 3, 4, 5 Although these translocations involve different oncogenes, molecularly their encoded products commonly activate the canonical nuclear factor (NF)-κB pathway,6, 7, 8 accounting for their critical role in lymphomagenesis. Nonetheless, overexpression of these oncogenes alone is insufficient for malignant transformation as both Eμ-API2-MALT1 and Eμ-BCL10 transgenic mice developed splenic marginal zone hyperplasia, but not lymphoma.9, 10 Thus, other molecular events are required to cooperate with these chromosome translocations in MALT lymphoma development.

The above chromosomal translocations are always mutually exclusive and t(11;18), the most frequent translocation in MALT lymphoma, occurs often as the sole cytogenetic abnormality.11 Several studies suggest that there is a potential cooperation between MALT lymphoma-associated oncogenic products and immunological stimuli in lymphomagenesis. In the Eμ-API2-MALT1 transgenic mice, immunization with the Freund′s complete adjuvant led to development of a splenic marginal zone lymphoma-like hyperplasia.12 In vitro assay showed that CD40 stimulation enhanced both API2-MALT1 and MALT1-induced NF-κB activation.13 However, the extent of potential cooperation between MALT lymphoma-associated oncogenic products and immune surface receptor signaling is unknown.

In spite of the presence of a potential overlap in the molecular mechanism of MALT lymphoma with and without translocation as discussed above, there are important differences in the clinical and histological presentations between these different subgroups. Clinically, gastric MALT lymphomas with t(11;18) or t(1;14) are significantly associated with advanced stages and resistance to H. pylori eradication.14, 15 Histologically, t(11;18)-positive MALT lymphomas seem to be more monotonous, lacking apparent transformed blasts.16 These distinct clinico-pathological characteristics may indicate important differences in molecular mechanisms between MALT lymphomas with and without translocation. To analyze this and understand further the molecular mechanism of MALT lymphoma, we studied the transcriptional profiles of a well-characterized series of cases with different translocation status and further validated the genes identified and the hypothesis generated.

Materials and methods

Patient materials

Fresh frozen tissues from 24 well-characterized MALT lymphomas (Supplementary Table S1), 7 follicular lymphoma (FL) and 7 mantle cell lymphoma (MCL) were used for gene expression microarray analysis. The MALT lymphomas were nine positive for t(11;18)/API2-MALT1 (eight gastric and one pulmonary), four positive for t(1;14)/BCL10-IGH or t(1;2)/BCL10-IGκ (three gastric and one pulmonary), two positive for t(14;18)/IGH-MALT1 (one hepatic and one ocular adnexal) and nine gastric cases negative for all known MALT lymphoma-associated translocations. The percentage of tumor cells was estimated on hematoxylin and eosin stained slides and crude microdissection was performed to ensure that at least 70% tumor cells was used for expression microarray analysis.

In addition, 73 cases of MALT lymphoma, including 18 positive for t(11;18), 8 positive for t(1;14), 9 positive for t(14;18) and 38 negative for these translocations, were used for validation of the expression microarray findings. The use of archival tissues for research was approved by the local research ethics committees of the authors’ institutions.

Gene expression microarray

RNA extraction, synthesis of labeled complementary RNA by in vitro transcription and hybridization to Affymetrix (Affymetrix UK Ltd., High Wycombe, UK) GeneChip HG-U133A (MALT lymphoma) or Affymetrix H133 plus 2 (FL and MCL), quality control analysis, microarray data normalization and nonspecific filtering are detailed in Supplementary Methods. All microarray data have been deposited with Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/, GSE18736).

Unsupervised clustering

This was performed using Pearson's correlation coefficient and average linkage as the similarity measure and clustering algorithm respectively within Genespring GX 7.3.1. Separate clustering was performed among all MALT lymphoma, FL and MCL cases and also within the MALT lymphoma group.

Gene set enrichment analysis (GSEA)

GSEA was used to identify gene sets differentially regulated between MALT lymphoma with (13 cases) and without (8 cases) chromosome translocation.17 As the original GSEA only identifies the gene set showing either uniformly up- or downregulation, for the gene sets showing both up- and downregulated genes, absolute GSEA was additionally performed.18 A total of 4395 gene sets were analyzed (Supplementary Table S2) and they included (1) NF-κB target genes, collated from online database, published works and careful bioinformatic search (Supplementary Table S3); (2) biological pathways involved in inflammatory and immune responses from human immunome database,19 gene ontology (GO) and ingenuity and (3) gene sets from Molecular Signature database. The GSEA results were ranked according to the nominal P-value (<0.05) and false discovery rate (0.25) as described previously.17

For the gene sets differentially regulated between MALT lymphoma with and without translocation, leading edge analysis was performed to identify the biologically important gene subset.17 When generating gene sets, for each sample, only the maximum expression value of the multi-probes for a given gene was used for GSEA as described previously.17

Analysis of differential gene expression in MALT lymphomas with and without translocation

For identification of differentially expressed genes between MATL lymphoma with and without translocation, the MAS5 normalized and filtered data set was used as suggested previously.20 The genes differentially expressed (one-way analysis of variance test, P<0.05) between translocation-positive and -negative MALT lymphomas were identified, and those showing >2.5-fold differences were selected for functional annotation using GO.

Functional annotation using GO

To assess the biological implication of differential gene expression in MALT lymphomas with and without translocation, we measured the representation of GO terms (association of gene products with their related biological processes and molecular functions) in the above differentially expressed genes using Genespring and hypergeometric tests provided in the R package (GOstats, version 2.8.0, http://www.bioconductor.org). This allowed us to examine whether any GO term was over- or under-represented as compared with what can occur by chance. Independent analyses of GO categories were performed for overexpressed genes in both translocation-positive and -negative MALT lymphoma.

Quantitative reverse-transcription PCR (qRT-PCR), immunohistochemistry and western blot analysis

Please see description in the Supplementary Methods (Supplementary Table S4S5).

NF-κB reporter assay

The potential cooperation between BCL10 (or API2-MALT1) and TLR6 expression in NF-κB activation was analyzed in vitro using a Luciferase Reporter Assay and the experimental details are given in the Supplementary Methods.

Results

Transcriptional profiling defines MALT lymphoma as a distinct entity

The microarray data from 21 MALT lymphomas (13 translocation-positive and 8 translocation-negative), 5 FL and 7 MCL passed the microarray hybridization quality control and were further analyzed and presented below (Supplementary Table S6). The standard normalization and filtering across all these cases yielded a set of 2629 probes. As expected, CD10 and BCL6 were found most highly expressed in FL, CCND1 most highly expressed in MCL, and MALT1 most highly expressed in MALT lymphoma with t(14;18) or t(11;18) (Supplementary Figure S1). Unsupervised hierarchical clustering showed that MALT lymphomas were clustered as a single branch, irrespective of their origin from different anatomic sites. Within the MALT lymphoma group, translocation-positive cases were intermingled with translocation-negative cases (Figure 1), indicating that the translocation status did not have major effect on the hierarchical clustering. We also repeated the unsupervised hierarchical clustering analysis exclusively on MALT lymphoma cases using a set of 6893 variant probes derived from U133A and very similar results were found (Supplementary Figure S2).

Figure 1
figure1

MALT lymphoma shows distinct gene expression profiles from FL and MCL. After standard normalization and filtering, a set of 2629 probes were obtained and used for unsupervised hierarchical clustering analysis. All MALT lymphomas are clustered as single branch, clearly separated from both FL and MCL. Nonetheless, within MALT lymphoma group, cases with and without chromosome translocation are intermingled together. The gene tree was ordered according to biological processes defined by GO, and the gene clusters enriched for a particular biological process in a lymphoma subtype as shown by hypergeometric testing are indicated. The gene sets for cell cycle (GO:7049) and regulation of progression through cell cycle (GO:74) are highly enriched in MCL, whereas those for immune cell activation (GO:45321, GO:46649) are enriched in FL. The gene sets for immune response to biotic stimulus (GO:9607), external biotic stimulus such as pest, pathogen or parasite (GO:9613) and cellular defense response (GO:6952) are highly enriched in MALT lymphoma. The chromosome translocation status of MALT lymphoma, FL and MCL are indicated by different color scheme. On the heatmap, red represents upregulated genes and green downregulated genes, with the scale showed at the bottom of the figure. FL, follicular lymphoma; MCL, mantle cell lymphoma; NEG, translocation-negative MALT lymphoma.

Differential expression of NF-κB target genes in MALT lymphoma with and without chromosome translocation

As expected, the absolute GSEA revealed that a subset of the NF-κB target genes was over-represented in the translocation-positive MALT lymphomas, whereas another subset was enriched in translocation-negative cases (P=0.011, false discovery rate=0.005, Figure 2a, Supplementary Table S7). Leading edge analysis showed that 19 core genes accounted for the significant enrichment in translocation-positive cases and the top 10 genes included CCR2A, BCL2, CD69, TLR6, TFEC, IRF4, PRDM2, REL, CCR7 and CCR5. Whereas, 34 core genes underscored the significant enrichment in translocation-negative cases and the top 10 genes were PTGS2, PLAU, NR4A3, PTGIS, IL8, CD86, CCL2, CCL11, CXCL5 and CXCL1.

Figure 2
figure2

GSEA of NF-κB target genes (a) and immune response genes (GO:6955) (b) in MALT lymphomas with and without chromosome translocation. Left panel shows the distribution of NF-κB target genes or immune response genes according to their rank position. Right panel shows heatmap illustration of their expression between MALT lymphoma with and without chromosome translocation. The top 15 leading edge core genes are shown. Trans –ve: translocation-negative MALT lymphoma; trans +ve: translocation-positive MALT lymphoma.

NF-κB target genes potentially underpins the differential representation of significant gene sets between MALT lymphomas with and without chromosome translocation

To gain further insights into the potential difference in molecular mechanisms between translocation-positive and -negative MALT lymphomas, we performed GSEA, in which indicated absolute GSEA, on 4394 gene sets covering various cellular pathways, biological processes and molecular functions. A total of 33 gene sets (not including those with very general term or those containing <20 genes) were differentially over-represented between MALT lymphomas with and without translocation (P<0.05, false discovery rate<0.20, Table 1).17 As there was a considerable overlap among the gene sets associated with the related cellular pathways or biological processes, they were grouped according to their involvement in the NF-κB activation pathway, inflammation/immune responses, chemokine and cell migration, G protein-coupled receptor (GPR) signaling and cell proliferation/apoptosis (Table 1). Leading edge analysis was performed to identify the core subset genes that underscored the significant enrichment and were thus most likely biologically important. Interestingly, the NF-κB target genes were frequently presented in each of these core subset genes, often on top of the list (Supplementary Tables S8S16, Supplementary Figure S3). Figure 2b shows the results of GSEA of immune response genes (GO:6955) with the top 15 leading edge core genes indicated in the heatmap illustration. Several NF-κB targets such as CCR2, BCL2, TLR6 and IRF4 were enriched in translocation-positive MALT lymphoma, whereas IL8, CD86 CCL2 and ICOS were over-represented in translocation-negative cases (Supplementary Table S12).

Table 1 Gene sets differentially over-represented between MALT lymphoma with and without chromosome translocation

Differential gene expression between MALT lymphomas with and without chromosome translocation

Using one-way analysis of variance test (P<0.05) and a 2.5-folds change as the threshold, we identified 26 and 62 genes significantly overexpressed in translocation-positive and translocation-negative MALT lymphoma respectively (Supplementary Table S17). To assess the biological implication of this differential gene expression in MALT lymphoma with and without chromosome translocation, we measured the representation of GO terms in the above gene sets using hypergeometric tests. Among the genes overexpressed in translocation-positive MALT lymphoma, the GO terms assocaited with NF-κB pathway activation, chemokine/GPR signaling, and antigen presentation were significantly over-represented (Supplementary Table S18). Although among the genes overexpressed in translocation-negative MALT lymphoma, the GO terms related to immune/defense response were significantly over-represented (Supplementary Table S18). These findings from analysis of differentially expressed genes between MALT lymphomas with and without translocation reinforce the above observations by GSEA.

Validation of gene expression by qRT-PCR and immunohistochemistry

To confirm the differential expression of the key candidate genes between MALT lymphoma with and without translocation identified by the above microaray analyses, we analyzed the expression of MALT1, BCL10, TLR6, CD69, CCR2A, CCR5, CD86 and NR4A3 by qRT-PCR of the microdissected tumor cells from formalin-fixed paraffin-embedded tissue and wherein possible by immunohistochemistry in a total of 58 cases including 16 used in gene expression profiling. As expected, MALT1 was most highly expressed in cases with t(14;18)/IGH-MALT1 and BCL10 was highest expressed in those with t(1;14) (Supplementary Figure S4). In keeping with the expression microarray data, TLR6, CD69 and CCR2A were highly expressed in t(1;14) or t(11;18)-positive MALT lymphomas and CCR5 was highly expressed in t(1;14)-positive cases in comparison with translocation-negative cases (Supplementary Figure S4). Conversely, CD86 and NR4A3 were significantly highly expressed in translocation-negative MALT lymphomas (Supplementary Figure S4).

In keeping with the above qRT-PCR data, immunohistochemistry showed that most translocation-positive MALT lymphomas showed strong to moderate homogeneous BCL2 and CD69 staining in >70% tumor cells, often in most tumor cells, whereas majority of translocation-negative cases showed heterogeneous staining in 30–70% cells or a negative result (Figure 3). Similarly, western blot analyses showed that TLR6 was highly expressed in translocation-positive MALT lymphoma in comparison with the translocation-negative cases (Figure 3). In contrast, most translocation-positive MALT lymphomas showed no CD86 staining, whereas the majority of translocation-negative cases showed heterogeneous CD86 staining albeit variable in both positivity and intensity (Figure 3).

Figure 3
figure3

BCL2, CD69, CD86 and TLR6 protein expression in MALT lymphomas with and without chromosomal translocation. (a) Top panel: examples of BCL2, CD69 and CD86 immunohistochemistry in MALT lymphomas with and without chromosome translocation. Original magnification × 400 for all panels. Lower panel summaries BCL2, CD69 and CD86 immunohistochemical results in MALT lymphoma with and without chromosome translocation. BCL2 and CD69 are more strongly and homogeneously expressed in translocation-positive than translocation-negative MALT lymphoma (P=6.9 × 10−5, P=2.2 × 10−4 respectively by Fisher's exact test), whereas CD86 is more strongly expressed in translocation-negative than translocation-positive MALT lymphoma (P=6.4 × 10−7 by Fisher's exact test). FC, follicle center; Trans+ve, translocation-positive; Trans −ve, translocation-negative. (b) Western blot analysis shows that TLR6 is highly expressed in translocation-positive MALT lymphoma, but at low levels in translocation-negative cases. M, MALT lymphoma; FL, follicular lymphoma.

TLR6 expression enhances NF-κB activation by BCL10 and API2-MALT1 in vitro

Among the genes highly expressed in translocation-positive MALT lympoma, TLR6, CCR2A, CD69 and BCL2 were particularly interesting and we selected TLR6 for further functional investigation because overexpression of this pattern recognition receptor may sensitize the response of tumor cells, particularly those with translocation, to stimulation by microbial antigens. To attest this, we performed a series of NF-κB reporter assays in Jurkat T cells, which are known not responding to lipopolysaccharides (LPS) stimulation, thus ideal for analyzing TLR signaling. Expression of TLR6 alone did not enhance BCL10 or API2-MALT1-induced NF-κB activation in Jurkat cells even in the presence of LPS stimulation (Supplementary Figure S5). TLR6 functions through the formation of heterodimer with its family member and typically forms heterodimer with TLR2 in responding to stimulation by bacterial antigen.21 We next analyzed whether co-expression of TLR6 and TLR2 could enhance BCL10 or API2-MALT1-mediated NF-κB activation in the presence of LPS stimulation. As expected, TLR6/2 co-expression, in the presence of LPS stimulation, were synergistic with BCL10 and API2-MALT1 in activating the NF-κB pathway (Figure 4). In contrast, there was no cooperation between co-expression of TLR6/1 and BCL10 or API2-MALT1 in NF-κB activation (Figure 4).

Figure 4
figure4

TLR6 enhances BCL10 and API2-MALT1-mediated NF-κB activation, in presence of TLR2 but not TLR1, in Jurkat T cells. Jurkat T cells were co-transfected with vector (pIRESpuro2) or plasmids containing BCL10, API2-MALT1 (AM), TLR6, TLR2 and TLR1 as indicated, together with NF-κB luciferase reporter gene. The transfected cells were seeded in multiple cell plates, cultured for 20 h and then treated with LPS or vehicle alone for 6 h. NF-κB activities were measured in quadruplet experiments and recorded as fold increase in the vector control. Western blot in the lower panel shows appropriate expression of various expression constructs. *P<0.01 by Student's t-test.

Discussion

This study showed that MALT lymphoma was characterized by distinct expression profile in comparison with FL and MCL, in line with the recent finding by Chng et al.22 Although unsupervised clustering analyses showed considerable overlap in the gene expression profiles between MALT lymphomas with and without chromosome translocation, there was important difference in the expression of NF-κB target genes between the two subgroups. By exhaustive GSEA of various molecular pathways and biological processes, we also showed that the gene sets related to inflammation, immune responses, chemokine and GPR signaling are differentially over-represented between these different subgroups. Importantly, several of these molecular pathways or biological processes also lead to NF-κB activation. These findings were reinforced by independent analyses of differentially expressed genes between MALT lymphomas with and without translocation using hypergeometric tests. Our observations provide several novel insights into the molecular mechanisms of both translocation-positive and -negative MALT lymphomas and potentially explain their different clinical and histological presentations.

Molecular mechanism of translocation-positive MALT lymphoma

In comparison with translocation-negative MALT lymphoma, GSEA and leading edge analyses revealed a common core subset of genes that were overexpressed in translocation-positive cases and a high proportion of them are NF-κB target genes involving multiple related biological processes or molecular pathways. The top examples included immune receptors such as TLR6, TLR7, CD69 and CD1D, and chemokine receptor such as CCR2, CXCR4, CCR6 and CCR7, the apoptosis inhibitor BCL2, and positive regulators of the NF-κB pathway such as REL and molecules involved in GPR signaling (Figure 2, Supplementary Tables S7S16). The overexpression of CCR2, BCL2, CD69 and TLR6 in translocation-positive cases was further confirmed in a large cohort of MALT lymphomas by qRT-PCR and/or immunohistochemistry/western blot analysis. All these molecules are expected to promote tumor cell survival and proliferation either directly or indirectly. Among these, the overexpression of the above immune surface receptors and chemokine receptors is particularly interesting.

TLR are critical in surveillance of microbial infection by recognizing pathogen-associated molecular patterns such as LPS and bacterial lipopeptides. In mouse model, it has been shown that TLR signaling promotes marginal zone B-cell activation and migration.23 TLR6 typically forms heterodimers with TLR2 on the cell surface to recognize bacterial antigens.21 TLR2/TLR6 signaling activates not only IkappaB kinase (IKK) complex that leads to activation of the NF-κB transcriptional factor, but also the mitogen-activated protein kinase (MAPK) p38 and Jun amino-terminal kinase that lead to activation of the activator protein 1 (AP-1) transcriptional factor.24 Hence, overexpression of TLR6 in translocation-positive MALT lymphoma could potentially augment the NF-κB activity mediated by MALT lymphoma-associated oncogenic products and also activate the MAPK pathways. In this study, we tested the former hypothesis and showed indeed that expression of TLR6, in presence of TLR2, could enhance both BCL10 and API2-MALT1-mediated NF-κB activation in vitro and this effect was particularly significant on LPS stimulation. A role of TLR signaling in the pathogenesis of translocation-positive MALT lymphoma is also suggested by the followings: (1) H. pylori infection is invariably associated with translocation-positive gastric MALT lymphoma; (2) H. pylori activates NF-κB through both the classical and alternative pathway in B lymphocytes and this effect is dependent on LPS but not cag pathogenecity island;25 (3) H. pylori-associated LPS-induced NF-κB activation requires TLR2/TLR6 or TLR2/TLR1 complex.26 Taken together, these findings suggest that there is a potential biological cooperation between MALT lymphoma translocation and TLR signaling in the lymphomagenesis.

CD69, a type II transmembrane glycoprotein, is a potential co-stimulatory receptor and may also have an immunoregulatory role.27 Although the precise function of CD69 in B cells is largely unknown, it is a well-described activation marker in several cell types, and its expression is upregulated in marginal zone B cells on TLR stimulation.23 CD69 is frequently expressed in low-grade B-cell lymphomas, and in FL, its expression is associated with poor treatment outcome.28, 29 Our finding of enriched expression of CD69 in translocation-positive MALT lymphoma further implicates its role in lymphoma pathogenesis.

CCR are GPRs and mediate immune cells migration and their retention in the inflammatory site. As B-cell homeostatic chemokine receptor, CCR7, CCR6 and CXCR4 are crucial for this homing process. For example, CCR7 has a central role in the regulation of normal mucosal lymphocyte re-circulation and homeostasis, particularly in the stomach,30 and CXCR4 is critical for B-cell homing to the Peyer's patches and splenic marginal zone.31 Although the specific role of CCR2 in B-cell trafficking and homing is unclear, it forms heterodimer with CXCR4,32 thus potentially having a role in mature B-cell homing process. In both low-grade B-cell lymphomas and classic Hodgkin lymphomas, CCR7 and CXCR4 overexpression were associated with a wide lymph node spread, supporting their role in lymphoma pathogenesis.33, 34, 35 In addition to homing process, CCR signaling may also promote cell survival and proliferation through its activation of MAPK pathways. In this context, it is noteworthy that GPR is also targeted by chromosomal translocation in MALT lymphoma. A recent study reported deregulation of GPR34 expression by t(X;14)(p11;q32) in a salivary gland MALT lymphoma.36 Importantly, expression of GPR34-induced activation of both the NF-κB and MAPK pathways in vitro.36 In keeping with these findings, our GSEA also showed that several gene sets related to GPR signaling and MAPK pathways were enriched in translocation-positive MALT lymphoma.

As discussed above, several molecular pathways including signaling through TLR, and chemokine receptor may be operational in translocation-positive MALT lymphomas and contribute to the activation of the NF-κB pathway (Figure 5). Together with MALT lymphoma-associated oncogenic products, they cause relentless NF-κB activation, leading to the prolonged survival of tumor cells even in the case of obliteration of microbe-mediated immune responses, such as H. pylori eradication in gastric MALT lymphoma. In this regard, it is to be noted that the apoptosis inhibitor BCL2 was remarkably uniformly overexpressed virtually in all tumor cells in nearly all translocation-positive cases. In contrast, the protein was heterogeneously expressed, at a much lower level, in tumor cells of translocation-negative cases.

Figure 5
figure5

Summary and hypothesis on molecular mechanism of MALT lymphoma with and without chromosomal translocation. In translocation-positive MALT lymphoma, overexpression of API2-MALT1, BCL10 and MALT1 activates the canonical NF-κB pathway, leading to enhanced expression of the NF-κB target genes, particularly TLR6, CCR2, CD69 and BCL2. Overexpression of TLR6 may provide a further positive feedback to the activation of the NF-κB pathway. Similar positive feedback may also be expected from the CCR2 signaling, and in addition both TLR6 and CCR2 may trigger activation of the MAPK pathway. The pathogenic implication of enhanced CD69 expression is currently unknown. Overexpression of BCL2 is expected to promote the tumor cell survival. In essence, the above chromosome translocations cause constitutive NF-κB activation with expression of its target genes forming a potential positive feedback loop, and the relentless NF-κB activation, in the case of gastric MALT lymphoma, confers its resistance to H. pylori eradication. In translocation-negative MALT lymphoma, the ongoing inflammatory and immune responses maintain active cognate B and T cell interaction through co-stimulating molecules CD86/CD28, B7RP1/ICOS, which are the major determinants of tumor cell survival and thus explain, in the cases of gastric MALT lymphoma, their responses to H. pylori eradication.

In spite of the above overwhelming evidence of NF-κB activation in translocation-positive MALT lymphoma, there was considerable heterogeneity in the expression of NF-κB target genes among these lymphomas. Not all translocation-positive MALT lymphomas showed uniform overexpression of the leading edge core set of the NF-κB target genes described above, nor each of the translocation-negative cases showed a complete lack of expression of these NF-κB target genes (Figure 2, Supplementary Figure S4). This is also consistent with the clinical response of gastric MALT lymphoma to H. pylori eradication therapy. Although most of t(11;18)-positive gastric MALT lymphomas do not respond to H. pylori eradication, there are occasional cases responsive to the antibiotic treatment,37 suggesting that not all translocations have the same biological effect. Equally, the majority of translocation-negative gastric MALT lymphomas can be cured by H. pylori eradication, but there are 10–20% cases that are negative for MALT1, BCL10 and FOXP1 involved translocations, and do not respond to H. pylori eradication,37 suggesting presence of other unknown genetic abnormalities that may also target the NF-κB pathway.

Molecular mechanism of translocation-negative MALT lymphoma

In contrast to translocation-positive MALT lymphoma, translocation-negative cases were characterized by expression of a strong inflammatory gene signature. GSEA and leading edge analysis also revealed common core subset genes involving several related biological processes or molecular pathways, which were enriched in translocation-negative MALT lymphoma. The top examples included proinflammatory cytokines IL8 and IL1β, molecules involved in B- and T-cell interaction such as CD86, CD28 and ICOS, several chemokine and chemokine receptors and NR4A3 (also known as MINOR) (Figure 2, Supplementary Tables S7, S9, S10, S12, S14).

IL8 and IL1β are the hallmark of proinflammatory cytokine profile in response to H. pylori infection. IL8 is critical for neutrophil infiltration and activation, whereas IL1β induces gastrin release, inhibits acid secretion and promote apoptosis of epithelial cells.38 The finding of overexpression of these proinflammatory cytokines in translocation-negative gastric MALT lymphomas, indicates the presence of active H. pylori infection. In keeping with this, translocation-negative gastric MALT lymphomas show a higher number of blast cells than translocation-positive cases.16 In addittion, a number of chemokines and chemokine receptors were highly expressed in the translocation-negative cases. This may reflect the trafficking and retention of various immune cells in response to an active H. pylori infection.

Most importantly, GSEA showed enriched expression of the surface molecules involved in B- and T-cell interaction namely CD86, CD28 and ICOS in translocation-negative gastric MALT lymphoma. Although residual reactive follicles may be present and contribute to the high CD86, CD28 and ICOS expression in translocation-negative cases, the germinal center markers CD10 and BCL6 were expressed in much low levels in MALT lymphoma (Supplementary Figure S1), and more importantly overexpression of CD86 in tumor cells was clearly shown by qRT-PCR and immunohistochemistry. In line with our finding, a previous study showed significantly higher CD86 expression in gastric MALT lymphomas that responded to H. pylori eradication than those resisted to the therapy (66% vs 10%).39 Although the chromosome translocation status in these cases is not available, it is most likely that the cases responded to H. pylori were translocation-negative.14 Taken together, these findings suggest that there is an active immune response to H. pylori infection in translocation-negative gastric MALT lymphoma, and this most likely underscores the tumor cell survival and expansion, and thus determines their response to H. pylori eradication (Figure 5).

In summary, this study shows that (1) translocation-positive MALT lymphoma is in general characterized by an enhanced expression of NF-κB target genes, particularly CCR2, TLR6 and BCL2; (2) the oncogenic products of MALT lymphoma-associated translocation may cooperate with signaling from several surface receptors including TLR6 and chemokine receptors in activation of the NF-κB pathways; and (3) translocation-negative MALT lymphoma is featured by active inflammatory and immune responses to H. pylori infection, and tumor cell interaction with infiltrating T cells through co-stimulating molecules (especially CD86/CD28) may have an important role in their survival and clonal expansion.

References

  1. 1

    Isaacson PG, Du MQ . MALT lymphoma: from morphology to molecules. Nat Rev Cancer 2004; 4: 644–653.

    CAS  Article  PubMed  Google Scholar 

  2. 2

    Ye H, Liu H, Attygalle A, Wotherspoon AC, Nicholson AG, Charlotte F et al. Variable frequencies of t(11;18)(q21;q21) in MALT lymphomas of different sites: significant association with CagA strains of H pylori in gastric MALT lymphoma. Blood 2003; 102: 1012–1018.

    CAS  Article  PubMed  Google Scholar 

  3. 3

    Ye H, Gong L, Liu H, Hamoudi RA, Shirali S, Ho L et al. MALT lymphoma with t(14;18)(q32;q21)/IGH-MALT1 is characterized by strong cytoplasmic MALT1 and BCL10 expression. J Pathol 2005; 205: 293–301.

    CAS  Article  PubMed  Google Scholar 

  4. 4

    Streubel B, Simonitsch-Klupp I, Mullauer L, Lamprecht A, Huber D, Siebert R et al. Variable frequencies of MALT lymphoma-associated genetic aberrations in MALT lymphomas of different sites. Leukemia 2004; 18: 1722–1726.

    CAS  Article  PubMed  Google Scholar 

  5. 5

    Goatly A, Bacon CM, Nakamura S, Ye H, Kim I, Brown PJ et al. FOXP1 abnormalities in lymphoma: translocation breakpoint mapping reveals insights into deregulated transcriptional control. Mod Pathol 2008; 21: 902–911.

    CAS  Article  PubMed  Google Scholar 

  6. 6

    Lucas PC, Yonezumi M, Inohara N, McAllister-Lucas LM, Abazeed ME, Chen FF et al. Bcl10 and MALT1, independent targets of chromosomal translocation in MALT lymphoma, cooperate in a novel NF-kappa B signaling pathway. J Biol Chem 2001; 276: 19012–19019.

    CAS  Article  PubMed  Google Scholar 

  7. 7

    Zhou H, Wertz I, O’Rourke K, Ultsch M, Seshagiri S, Eby M et al. Bcl10 activates the NF-kappaB pathway through ubiquitination of NEMO. Nature 2004; 427: 167–171.

    CAS  Article  PubMed  Google Scholar 

  8. 8

    Lucas PC, Kuffa P, Gu S, Kohrt D, Kim DS, Siu K et al. A dual role for the API2 moiety in API2-MALT1-dependent NF-kappaB activation: heterotypic oligomerization and TRAF2 recruitment. Oncogene 2007; 26: 5643–5654.

    CAS  Article  PubMed  Google Scholar 

  9. 9

    Baens M, Fevery S, Sagaert X, Noels H, Hagens S, Broeckx V et al. Selective expansion of marginal zone B cells in Emicro-API2-MALT1 mice is linked to enhanced IkappaB kinase gamma polyubiquitination. Cancer Res 2006; 66: 5270–5277.

    CAS  Article  PubMed  Google Scholar 

  10. 10

    Li Z, Wang H, Xue L, Shin DM, Roopenian D, Xu W et al. Emu-BCL10 mice exhibit constitutive activation of both canonical and noncanonical NF-kappaB pathways generating marginal zone (MZ) B-cell expansion as a precursor to splenic MZ lymphoma. Blood 2009; 114: 4158–4168.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    Zhou Y, Ye H, Martin-Subero JI, Hamoudi R, Lu YJ, Wang R et al. Distinct comparative genomic hybridisation profiles in gastric mucosa-associated lymphoid tissue lymphomas with and without t(11;18)(q21;q21). Br J Haematol 2006; 133: 35–42.

    CAS  Article  PubMed  Google Scholar 

  12. 12

    Sagaert X, Theys T, Wolf-Peeters C, Marynen P, Baens M . Splenic marginal zone lymphoma-like features in API2-MALT1 transgenic mice that are exposed to antigenic stimulation. Haematologica 2006; 91: 1693–1696.

    CAS  PubMed  Google Scholar 

  13. 13

    Ho L, Davis RE, Conne B, Chappuis R, Berczy M, Mhawech P et al. MALT1 and the API2-MALT1 fusion act between CD40 and IKK and confer NF-kappa B-dependent proliferative advantage and resistance against FAS-induced cell death in B cells. Blood 2005; 105: 2891–2899.

    CAS  Article  PubMed  Google Scholar 

  14. 14

    Liu H, Ye H, Ruskone-Fourmestraux A, de Jong D, Pileri S, Thiede C et al. T(11;18) is a marker for all stage gastric MALT lymphomas that will not respond to H pylori eradication. Gastroenterology 2002; 122: 1286–1294.

    CAS  Article  PubMed  Google Scholar 

  15. 15

    Ye H, Gong L, Liu H, Ruskone-Fourmestraux A, de Jong D, Pileri S et al. Strong BCL10 nuclear expression identifies gastric MALT lymphomas that do not respond to H pylori eradication. Gut 2006; 55: 137–138.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  16. 16

    Okabe M, Inagaki H, Ohshima K, Yoshino T, Li C, Eimoto T et al. API2-MALT1 fusion defines a distinctive clinicopathologic subtype in pulmonary extranodal marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue. Am J Pathol 2003; 162: 1113–1122.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. 17

    Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 2005; 102: 15545–15550.

    CAS  Article  Google Scholar 

  18. 18

    Saxena V, Orgill D, Kohane I . Absolute enrichment: gene set enrichment analysis for homeostatic systems. Nucleic Acids Res 2006; 34: e151.

    Article  PubMed  PubMed Central  Google Scholar 

  19. 19

    Ortutay C, Vihinen M . Immunome: a reference set of genes and proteins for systems biology of the human immune system. Cell Immunol 2006; 244: 87–89.

    CAS  Article  PubMed  Google Scholar 

  20. 20

    Pepper SD, Saunders EK, Edwards LE, Wilson CL, Miller CJ . The utility of MAS5 expression summary and detection call algorithms. BMC Bioinformatics 2007; 8: 273.

    Article  PubMed  PubMed Central  Google Scholar 

  21. 21

    Gomariz RP, Arranz A, Juarranz Y, Gutierrez-Canas I, Garcia-Gomez M, Leceta J et al. Regulation of TLR expression, a new perspective for the role of VIP in immunity. Peptides 2007; 28: 1825–1832.

    CAS  Article  PubMed  Google Scholar 

  22. 22

    Chng WJ, Remstein ED, Fonseca R, Bergsagel PL, Vrana JA, Kurtin PJ et al. Gene expression profiling of pulmonary mucosa-associated lymphoid tissue lymphoma identifies new biologic insights with potential diagnostic and therapeutic applications. Blood 2009; 113: 635–645.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. 23

    Rubtsov AV, Swanson CL, Troy S, Strauch P, Pelanda R, Torres RM . TLR agonists promote marginal zone B cell activation and facilitate T-dependent IgM responses. J Immunol 2008; 180: 3882–3888.

    CAS  Article  PubMed  Google Scholar 

  24. 24

    Akira S, Takeda K . Toll-like receptor signalling. Nat Rev Immunol 2004; 4: 499–511.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  25. 25

    Ohmae T, Hirata Y, Maeda S, Shibata W, Yanai A, Ogura K et al. Helicobacter pylori activates NF-kappaB via the alternative pathway in B lymphocytes. J Immunol 2005; 175: 7162–7169.

    CAS  Article  PubMed  Google Scholar 

  26. 26

    Yokota S, Ohnishi T, Muroi M, Tanamoto K, Fujii N, Amano K . Highly-purified Helicobacter pylori LPS preparations induce weak inflammatory reactions and utilize Toll-like receptor 2 complex but not Toll-like receptor 4 complex. FEMS Immunol Med Microbiol 2007; 51: 140–148.

    CAS  Article  PubMed  Google Scholar 

  27. 27

    Sancho D, Gomez M, Sanchez-Madrid F . CD69 is an immunoregulatory molecule induced following activation. Trends Immunol 2005; 26: 136–140.

    CAS  Article  PubMed  Google Scholar 

  28. 28

    Erlanson M, Gronlund E, Lofvenberg E, Roos G, Lindh J . Expression of activation markers CD23 and CD69 in B-cell non-Hodgkin's lymphoma. Eur J Haematol 1998; 60: 125–132.

    CAS  Article  PubMed  Google Scholar 

  29. 29

    de Jong D, Koster A, Hagenbeek A, Raemaekers J, Veldhuizen D, Heisterkamp S et al. Impact of the tumor microenvironment on prognosis in follicular lymphoma is dependent on specific treatment protocols. Haematologica 2009; 94: 70–77.

    Article  PubMed  Google Scholar 

  30. 30

    Hopken UE, Wengner AM, Loddenkemper C, Stein H, Heimesaat MM, Rehm A et al. CCR7 deficiency causes ectopic lymphoid neogenesis and disturbed mucosal tissue integrity. Blood 2007; 109: 886–895.

    Article  PubMed  Google Scholar 

  31. 31

    Nie Y, Waite J, Brewer F, Sunshine MJ, Littman DR, Zou YR . The role of CXCR4 in maintaining peripheral B cell compartments and humoral immunity. J Exp Med 2004; 200: 1145–1156.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  32. 32

    Springael JY, Urizar E, Parmentier M . Dimerization of chemokine receptors and its functional consequences. Cytokine Growth Factor Rev 2005; 16: 611–623.

    CAS  Article  PubMed  Google Scholar 

  33. 33

    Lopez-Giral S, Quintana NE, Cabrerizo M, Alfonso-Perez M, Sala-Valdes M, De Soria VG et al. Chemokine receptors that mediate B cell homing to secondary lymphoid tissues are highly expressed in B cell chronic lymphocytic leukemia and non-Hodgkin lymphomas with widespread nodular dissemination. J Leukoc Biol 2004; 76: 462–471.

    CAS  Article  PubMed  Google Scholar 

  34. 34

    Hopken UE, Foss HD, Meyer D, Hinz M, Leder K, Stein H et al. Up-regulation of the chemokine receptor CCR7 in classical but not in lymphocyte-predominant Hodgkin disease correlates with distinct dissemination of neoplastic cells in lymphoid organs. Blood 2002; 99: 1109–1116.

    CAS  Article  PubMed  Google Scholar 

  35. 35

    Trentin L, Cabrelle A, Facco M, Carollo D, Miorin M, Tosoni A et al. Homeostatic chemokines drive migration of malignant B cells in patients with non-Hodgkin lymphomas. Blood 2004; 104: 502–508.

    CAS  Article  PubMed  Google Scholar 

  36. 36

    Novak A, Akasaka T, Manske M, Gupta M, Witzig T, Dyer MJS et al. Elevated expression of GPR34 and its association with a novel translocation t(X;14)(p11;q32) involving IGHS and GPR34 in MALT lymphoma. Blood 2008; 112: 785, Abstract 2251.

    Google Scholar 

  37. 37

    Du MQ, Atherton JC . Molecular subtyping of gastric MALT lymphomas: implications for prognosis and management. Gut 2006; 55: 886–893.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  38. 38

    McNamara D, El Omar E . Helicobacter pylori infection and the pathogenesis of gastric cancer: a paradigm for host-bacterial interactions. Dig Liver Dis 2008; 40: 504–509.

    CAS  Article  PubMed  Google Scholar 

  39. 39

    de Jong D, Vyth-Dreese F, Dellemijn T, Verra N, Ruskone-Fourmestraux A, Lavergne-Slove A et al. Histological and immunological parameters to predict treatment outcome of Helicobacter pylori eradication in low-grade gastric MALT lymphoma. J Pathol 2001; 193: 318–324.

    CAS  Article  PubMed  Google Scholar 

Download references

Acknowledgements

We would like to thank Professor Ahmet Dogan and Dr Ellen Remstein, Department of Pathology, Mayo Clinic for sharing their pulmonary MALT lymphoma gene expression microarray data; Dr Ian McFarlane, the Microarray CoreLab, National Institute of Health Research, Cambridge Comprehensive Biomedical Research Centre for his technical assistance, Dr Koichi Kuwano, Department of Infectious Medicine, Kurume University School of Medicine, Japan for providing TLR1, TLR2 and TLR6 expression constructs and both previous and present members of Du lab for helpful discussion and assistance. The research in Du lab was supported by research grants from Leukemia Research, UK and the National Institute for Health Research Cambridge Biomedical Research Center. BS was supported by Grant FWF (P19346-B12). LdL is a senior research associate of the FRS-FNRS.

Author information

Affiliations

Authors

Corresponding author

Correspondence to M-Q Du.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Authors contribution

RAH designed the experiment, collected and analyzed the data. AA, HY and LG contributed to the design and experimental data collection and analysis; ARF, BS, AC, MR, IW, CDWP, KAM, LdL and PGI provided lymphoma cases; MQD designed, analyzed the data and wrote the paper.

Supplementary Information accompanies the paper on the Leukemia website

Supplementary information

Supplementary Figure S1 (PPT 278 kb)

Supplementary Figure S2 (PPT 287 kb)

Supplementary Figure S3 (PPT 138 kb)

Supplementary Figure S4 (PPT 226 kb)

Supplementary Figure S5 (PPT 145 kb)

Supplementary Figure legends (DOC 28 kb)

Supplementary Table S1 (XLS 30 kb)

Supplementary Table S2 (XLS 37 kb)

Supplementary Table S3 (XLS 49 kb)

Supplementary Table S4 (XLS 22 kb)

Supplementary Table S5 (XLS 27 kb)

Supplementary Table S6 (XLS 43 kb)

Supplementary Table S7 (XLS 34 kb)

Supplementary Table S8 (XLS 27 kb)

Supplementary Table S9 (XLS 33 kb)

Supplementary Table S10 (XLS 33 kb)

Supplementary Table S11 (XLS 29 kb)

Supplementary Table S12 (XLS 31 kb)

Supplementary Table S13 (XLS 27 kb)

Supplementary Table S14 (XLS 119 kb)

Supplementary Table S15 (XLS 26 kb)

Supplementary Table S16 (XLS 28 kb)

Supplementary Table S17 (XLS 76 kb)

Supplementary Table S18 (XLS 31 kb)

Supplementary Methods (DOC 50 kb)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hamoudi, R., Appert, A., Ye, H. et al. Differential expression of NF-κB target genes in MALT lymphoma with and without chromosome translocation: insights into molecular mechanism. Leukemia 24, 1487–1497 (2010). https://doi.org/10.1038/leu.2010.118

Download citation

Keywords

  • MALT lymphoma
  • gene expression profiling
  • NF-κB activation
  • chromosome translocation

Further reading

Search

Quick links