GALNT11 as a new molecular marker in chronic lymphocytic leukemia

Aberrant mucin O-glycosylation often occurs in different cancers and is characterized by immature expression of simple mucin-type carbohydrates. At present, there are some controversial reports about the Tn antigen (GalNAcα-O-Ser/Thr) expression and there is a great lack of information about the [UDP-N-acetyl-α-d-galactosamine:polypeptide N-acetylgalactosaminyltransferase (GalNAc-Ts)] expression in chronic lymphocytic leukemia (CLL). To gain insight in these issues we evaluated the Tn antigen expression in CLL patient samples using two Tn binding proteins with different fine specificity. We also studied the expression from 14 GalNAc-Ts genes in CLL patients by RT-PCR. Our results have provided additional information about the expression level of the Tn antigen, suggesting that a low density of Tn residues is expressed in CLL cells. We also found that GALNT11 was expressed in CLL cells and normal T cell whereas little or no expression was found in normal B cells. Based on these results, GALNT11 expression was assessed by qPCR in a cohort of 50 CLL patients. We found significant over-expression of GALNT11 in 96% of B-CLL cells when compared to normal B cells. Moreover, we confirmed the expression of this enzyme at the protein level. Finally we found that GALNT11 expression was significantly associated with the mutational status of the immunoglobulin heavy chain variable region (IGHV), [א(2)(1)=18.26; P<0.0001], lipoprotein lipase expression [א(2)(1)=13.72; P=0.0002] and disease prognosis [א(2)(1)=15.49; P<0.0001]. Our evidence suggests that CLL patient samples harbor aberrant O-glycosylation highlighted by Tn antigen expression and that the over-expression of GALNT11 constitutes a new molecular marker for CLL.

"Role of the B-cell receptor and the microenvironment in chronic lymphocytic leukemia''

Despite significant progress in treatment, chronic lymphocytic leukemia (CLL) remains an incurable disease. Advances have been made to understand the molecular pathogenesis underlying CLL progression and treatment resistance. We here review the available evidences concerning the role of the B-cell receptor (BCR) and the tumor microenvironment interactions in CLL pathogenesis. Antigen likely has a key role in the selection of the tumoral clone, the mutational status of immunoglobulin genes is a strong prognostic predictor and BCR signaling has been postulated to have a role for CLL trafficking and interaction with the stromal microenvironment. There is also important evidence, favoring a role for the microenvironment in CLL pathogenesis. Most, if not all, proliferative events occur in the lymph nodes and bone marrow, where leukemic cells receive through microenvironment interactions survival signals aiming to avoid apoptosis and acquire favorable tumoral growing conditions. In addition, the tumoral microenvironment appears to be the site where the acquisition of additional genetic lesions in the clone occur, which should greatly influence clinical outcome. The advent of new tyrosine kinase inhibitors which seem to be able to modulate microenvironment interactions and circumvent the p53 deletion have generated significant promise by raising the possibility that they could provide significant progress in disease treatment.

What do somatic hypermutation and class switch recombination teach us about chronic lymphocytic leukaemia pathogenesis?

B-CLL cells express CD5 and IgM/IgD and thus have a mantle zone-like phenotype of naive cells, which, in normal conditions express unmutated Ig genes. However, recent studies have shown that 50%-70% of CLL harbour somatic mutations of VH genes, as if they had matured in a lymphoid follicle. Interestingly, the presence or absence of somatic hypermutation (SHM) process is associated with the use of particular VH genes. Particular alleles of the VH1-69 gene and the VH4-39 gene are preferentially expressed in an unmutated form, while VH4-34 or the majority of VH3 family genes frequently contain somatic mutations. The fact that some genes like VH1-69 and VH3-07 recombine this VH segment to particular JH segments and the restricted use of CDR3 sequences by CLLs expressing the VH4-39 gene suggest that the observed differences in BCR structure in B-CLL could result from selection by distinct antigenic epitopes. It is currently unclear whether this putative antigen-driven process could occur prior to leukaemic transformation and/or that the precursors were transformed into leukaemic cells at distinct maturational stages. The mutational profile of Ig genes has been shown to be associated with disease prognosis. These results could favour the idea that CLL could correspond to two different diseases that look alike in morphologic and phenotypic terms. In CLL with mutated Ig genes, the proliferating B cell may have transited through germinal centres, the physiologic site of hypermutation, whereas in CLL with unmutated Ig genes the malignant B cell may derive from a pre-germinal centre naïve B cell. Despite these clinical and molecular differences, recent studies on gene expression profiling of B-CLL cells showed that CLL is characterized by a common gene expression signature that is irrespective of Ig mutational status and differs from other lymphoid cancers and normal lymphoid subpopulations, suggesting that CLL cases share a common mechanism of transformation and/or cell of origin. Activation induced cytidine deaminase (AID) plays a key role in SHM and class switch recombination (CSR). However, the mechanisms accounting for AID action and control of its expression remain unclear. In a recent work we have shown that in contrast to normal circulating B-cells, AID transcripts are expressed constitutively in CLL patients undergoing active CSR, but interestingly this expression occurs predominately in unmutated CLL B-cells. These data favour the view that AID protein may act differentially on CSR and SHM pathways, but the role-played by AID in both processes remains to be elucidated. Recent work indicates that AID is expressed in a small fraction of tumoral cells, which could suggest that this small fraction of cells may correspond to B-CLL cells that would have recently experienced an AID-inducing stimulus occurring in a specific microenvironment.

Post-transcriptional regulation of inducible nitric oxide synthase in chronic lymphocytic leukemia B cells in pro- and antiapoptotic culture conditions

Functional inducible NOS (iNOS) may be involved in the prolonged lifespan of chronic lymphocytic leukemia cells (B-CLL), although the exact mechanisms implicated remain elusive as yet. In this work, we have examined iNOS expression in normal B lymphocytes and B-CLL cells in pro- and antiapoptotic conditions. Our results demonstrate: (1) The existence of a new splice variant characterized by a complete deletion of exon 14 (iNOS 13-16(14del)), which was preferentially detected in normal B lymphocytes and may represent an isoform that could play a role in the regulation of enzyme activity. (2) The existence of another alternatively spliced iNOS mRNA transcript involving a partial deletion of the flavodoxin region (iNOS 13-16(neg)) was correlated to a decreased B-CLL cell viability. The 9-beta-D-arabinofuranosyl-2-fluoradenine or fludarabine (F-ara) treatment induced iNOS 13-16(neg) transcript variants, whereas IL-4 enhanced both the transcription of variants, including these exons (iNOS 13-16(pos)), and the expression of a 122 kDa iNOS protein. These results suggest that in B-CLL, a regulation process involving nitric oxide (.- NO) levels could occur by a post-transcriptional mechanism mediated by soluble factors. Our results also provide an insight into a new complementary proapoptotic action of F-ara in B-CLL by the induction of particular iNOS splice variants, leading to the activation of a caspase-3-dependent apoptotic pathway.

Predictive value of serum thymidine kinase level for Ig-V mutational status in B-CLL

In B-CLL IgV(H) genes mutational status is a major prognostic factor. Since sequencing of IgV(H) genes is not available in most laboratories, an easily performed surrogate assay is desirable. To identify the best surrogate assay, and to better discriminate prognostic subgroups we analyzed clinical and biological data from 58 typical CLL cases. A higher serum thymidine kinase level (>15 U/l) proved to be a strong predictor of mutational status, and the only independent one among the studied parameters. To further identify prognostic subgroups, cluster analysis was employed on 38 cases on which all data were available, which segregated two groups including 25 and 13 patients, respectively. These two clusters differed by their proliferative potential and appeared to discriminate patients with very different clinical course and outcome. s-TK was strikingly different among these two clusters, suggesting that s-TK level could be used routinely to identify patients at risk of progression.

Do CLL B cells correspond to naive or memory B-lymphocytes? Evidence for an active Ig switch unrelated to phenotype expression and Ig mutational pattern in B-CLL cells

Recent work suggests that chronic lymphocytic leukemia (B-CLL) expressing unmutated immunoglobulin V genes could correspond to the proliferation of naive B cells whereas those expressing mutated genes, may correspond to the proliferation of post-germinal center B cells. Current data from gene profiling expression have failed to demonstrate a clear-cut distinction between these two forms of B-CLL disease. In the present study, we have investigated the complete V(H) nucleotide sequence and the presence of RNA transcripts from different C(H) domains in 25 B-CLL patients. Our results demonstrate that: (1) expression of IgD is not related to the mutational frequency and activation of the isotype switch pathway; (2) isotype switch, leading to simultaneous expression at the transcriptional and protein level of IgM, IgD, IgG and IgA, occurs in a small percentage of patients, and (3) different mechanisms such as VDJ duplication and trans-splicing or RNA splicing of long nuclear transcript, could be involved in isotype switch. Our results highlight the difficulty in assigning a normal counterpart to B-CLL cells and raise the possibility that a different B cell development pathway, independent from classical germinal centers, might exist in B-CLL.

Tn antigen is a pre-cancerous biomarker in breast tissue and serum in n-nitrosomethylurea-induced rat mammary carcinogenesis

The Tn determinant (GalNAcalpha-O-Ser/Thr), normally a cryptic structure in mucin-type O-glycans, is a tumor-associated marker which has attracted particular interest in cancer biology. We herein report the characterization of N-nitrosomethylurea (NMU)-induced breast cancer in rats as a new model for the study of aberrant O-glycosylation products. Tn-antigen expression is detectable not only in mammary carcinoma induced by NMU but also in carcinogen-initiated mammary epithelium, indicating that Tn could be a pre-cancerous biomarker in rats treated with NMU. Serum Tn levels were followed up longitudinally in 30 rats from the time of the first injection of NMU to the development of advanced breast cancer. Tn antigen increased in serum several weeks before tumor development, and became highly positive after 56 days of carcinogenesis (prior to breast-cancer occurrence), and the levels correlated with Tn expression in mammary tissues. However, during the follow-up after detection of mammary cancer, all animals displayed a significant decrease of serum Tn antigen, and low levels were observed in animals with advanced breast cancer. We have shown that the humoral immune response to cancer, with the production of anti-Tn antibodies, could hamper the detection of Tn antigen in animals with advanced breast cancer. These results suggest that NMU-induced rat mammary carcinogenesis is a useful experimental model to study the regulation of O-glycosylation at the cellular level during malignant transformation.

Production and functional characterization of two mouse/human chimeric antibodies with specificity for the tumor-associated Tn-antigen

In this work, we have constructed two functional mouse/human chimeric antibodies (IgMkappa and IgG1kappa isotypes) by inserting genomic DNA fragments encoding VH and Vkappa variable regions of the murine monoclonal antibody IgMK-83D4 into mammalian expression vectors containing human mu, gamma1, and kappa constant exons, and by transfecting them into the nonsecreting mouse myeloma X-63 cell line. In previous works, we have demonstrated that 83D4 murine mAb reacts with Tn determinant (GalNAcalpha-O-Ser/Thr) expressed in 90% of breast, ovary, and colon carcinomas. Both expressed chimeric antibodies were purified from the transfected cell line supernatant by affinity chromatography, and their reactivities against Tn antigen were confirmed by ELISA on asialo ovine submaxilar mucin and immunofluorescence studies on MCF-7 breast carcinoma cell line. We have demonstrated by gel filtration chromatography, that the principal secreted forms were monomers for IgG1kappa and pentamers for IgMkappa. The binding affinities of these chimeric antibodies against synthetic Tn glycopeptides, were evaluated by surface plasmon resonance showing an affinity constant similar to that of 83D4 native antibody for IgMkappa and a lower affinity constant for IgG1kappa chimeric antibody. On the other hand, the replacement of mouse C regions with human C regions confers both chimeric antibodies the ability to activate human complement. These mouse/human chimeric antibodies should be much less immunogenic and could play an important role in the lysis of tumor cell expressing Tn-antigen. Therefore, these anti-Tn chimeric antibodies could be considered as potential tools for human in vivo studies.

Molecular cloning of a monoclonal anti-tumor antibody specific for the Tn antigen and expression of an active single-chain Fv fragment

We report here the first amino acid sequence of an anti-Tn monoclonal antibody raised against human breast cancer cells and show that a single chain Fv fragment of this IgM retains the Tn-binding specificity as defined by functional assays with asialo-OSM and membrane extracts from MCF-7 cells. Sequence comparisons and molecular modeling of 83D4 indicate that the antibody combining site displays a cavity-like feature primarily defined by the CDR H1 and H2 loops. This pocket could accommodate a single Tn molecule, thus, suggesting a structural explanation for the predominant expression of a particular VH gene segment in a group of antibodies that recognize tumor-associated antigens arising from an aberrant O-glycosylation.