Epithelial cells trigger frontline immunoglobulin class switching through an SLPI-regulated pathway (41.7)

Epithelial cells (ECs) recruit B cells expressing class-switched IgG and IgA into mucosal effector sites and transport IgG and IgA antibodies across mucous membranes. Whether ECs initiate IgG and IgA class switching remains unknown. We found that ECs lining tonsillar crypts formed pockets populated by B cells expressing activation-induced cytidine deaminase (AID), a DNA-editing enzyme associated with ongoing class switch DNA recombination (CSR). ECs released AID- and CSR-inducing factors, such as B cell-activating factor of the TNF family (BAFF), after sensing microbial products through Toll-like receptors (TLRs). EC-induced IgG and IgA CSR was further amplified by thymic stromal lymphopoietin (TSLP), an epithelial IL-7-like cytokine that enhanced BAFF production by dendritic cells. Conversely, EC-induced IgG and IgA CSR was restrained by secretory leukocyte protease inhibitor (SLPI), an epithelial homeostatic protein that inhibited BAFF signaling in B cells. Our findings indicate that ECs function as mucosal guardians that orchestrate frontline IgG and IgA class switching through a TLR-inducible SLPI-regulated signaling program.

This work was supported by NIH grants AI057653 (to A.C.) and by NIH T32 training grant AI07621 (to W. X.).

Kaposi sarcoma-associated herpesvirus in non-Hodgkin lymphoma and reactive lymphadenopathy in Uganda

Kaposi sarcoma-associated herpesvirus (KSHV) causes Kaposi sarcoma and is also associated with primary effusion lymphoma, a subset of diffuse large B-cell lymphomas, and multicentric Castleman disease. Because KSHV infection is endemic in sub-Saharan Africa, we sought to identify cases of KSHV-positive non-Hodgkin lymphomas (NHLs) and reactive lymphadenopathy in this region. One hundred forty-four cases (80 NHLs, 64 reactive lymph nodes) from the major pathology laboratory in Uganda were reviewed. One NHL was KSHV-positive, as indicated by staining for the viral latent nuclear antigen. This NHL was a diffuse large B-cell lymphoma in a 5-year-old boy. The tumor was also Epstein-Barr virus-positive. In addition, 2 reactive lymph nodes, both classified histologically as follicular involution, stained KSHV latent nuclear antigen-positive and thus most likely represent multicentric Castleman disease. In all 3 KSHV-positive cases, a minority of cells expressed KSHV viral interleukin 6, a biologically active cytokine homolog. In conclusion, we show that KSHV is rarely associated with lymphoproliferative disorders in sub-Saharan Africa. We describe the first case of a KSHV-positive NHL from this region; this case is also the first reported pediatric lymphoma associated with KSHV infection.

Epithelial cells trigger frontline immunoglobulin class switching through a pathway regulated by the inhibitor SLPI

Epithelial cells (ECs) transport class-switched immunoglobulin G (IgG) and IgA antibodies across mucous membranes. Whether ECs initiate class switching remains unknown. Here we found that ECs lining tonsillar crypts formed pockets populated by B cells expressing activation-induced cytidine deaminase (AID), an enzyme associated with ongoing class switching. ECs released B cell-activating AID-inducing factors after sensing microbial products through Toll-like receptors. The resulting class switching was amplified by thymic stromal lymphopoietin, an epithelial interleukin 7-like cytokine that enhanced the B cell 'licensing' function of dendritic cells, and was restrained by secretory leukocyte protease inhibitor, an epithelial homeostatic protein that inhibited AID induction in B cells. Thus, ECs may function as mucosal 'guardians' orchestrating frontline IgG and IgA class switching through a Toll-like receptor-inducible signaling program regulated by secretory leukocyte protease inhibitor.NOTE: In the version of this article initially published online, the middle label above Figure 6c is incorrect. The correct label should be 'BAFF'. The error has been corrected for all versions of the article.

beta-Glucuronidase is an optimal normalization control gene for molecular monitoring of chronic myelogenous leukemia

Quantitative monitoring of breakpoint cluster region (BCR)-Abelson kinase (ABL) transcripts has become indispensable in the clinical care of patients with chronic myelogenous leukemia. Because quantity and quality of RNA in clinical samples are highly variable, a suitable internal normalization control is required for accurate BCR-ABL quantification. However, few studies have examined suitability of the control genes using criteria relevant to residual disease testing. In this study, we evaluated a number of control genes with the application of several novel criteria, including control gene performance on serial patient sample testing and in a residual disease model. We also examined expression of the control genes in BCR-ABL-positive K562 cells in response to Gleevec treatment. We found that beta-glucuronidase is the best control gene among those studied. Importantly, ABL, a widely used control gene, generates misleading BCR-ABL changes that potentially affect the clinical management of chronic myelogenous leukemia patients.

Hodgkin lymphoma cells express TACI and BCMA receptors and generate survival and proliferation signals in response to BAFF and APRIL

Hodgkin lymphoma (HL) originates from the clonal expansion of malignant Hodgkin and Reed-Sternberg (HRS) cells. These B-cell-derived elements constitute less than 10% of the tumoral mass. The remaining tissue is comprised of an inflammatory infiltrate that includes myeloid cells. Myeloid cells activate B cells by producing BAFF and APRIL, which engage TACI, BCMA, and BAFF-R receptors on the B cells. Here, we studied the role of BAFF and APRIL in HL. Inflammatory and HRS cells from HL tumors expressed BAFF and APRIL. Unlike their putative germinal center B-cell precursors, HRS cells lacked BAFF-R, but expressed TACI and BCMA, a phenotype similar to that of plasmacytoid B cells. BAFF and APRIL enhanced HRS cell survival and proliferation by delivering nonredundant signals via TACI and BCMA receptors through both autocrine and paracrine pathways. These signals caused NF-kappaB activation; Bcl-2, Bcl-xL, and c-Myc up-regulation; and Bax down-regulation, and were amplified by APRIL-binding proteoglycans on HRS cells. Interruption of BAFF and APRIL signaling by TACI-Ig decoy receptor, which binds to and neutralizes BAFF and APRIL, or by small-interfering RNAs targeting BAFF, APRIL, TACI, and BCMA inhibited HRS cell accumulation in vitro and might attenuate HL expansion in vivo.

HIV-1 envelope triggers polyclonal Ig class switch recombination through a CD40-independent mechanism involving BAFF and C-type lectin receptors

Switching from IgM to IgG and IgA is essential for antiviral immunity and requires engagement of CD40 on B cells by CD40L on CD4(+) T cells. HIV-1 is thought to impair CD40-dependent production of protective IgG and IgA by inducing progressive loss of CD4(+) T cells. Paradoxically, this humoral immunodeficiency is associated with B cell hyperactivation and increased production of nonprotective IgG and IgA that are either nonspecific or specific for HIV-1 envelope glycoproteins, including gp120. Nonspecific and gp120-specific IgG and IgA are sensitive to antiretroviral therapy and remain sustained in infected individuals with very few CD4(+) T cells. One interpretation is that some HIV-1 Ags elicit IgG and IgA class switch DNA recombination (CSR) in a CD40-independent fashion. We show that a subset of B cells binds gp120 through mannose C-type lectin receptors (MCLRs). In the presence of gp120, MCLR-expressing B cells up-regulate the CSR-inducing enzyme, activation-induced cytidine deaminase, and undergo CSR from IgM to IgG and IgA. CSR is further enhanced by IL-4 or IL-10, whereas Ab secretion requires a B cell-activating factor of the TNF family. This CD40L-related molecule is produced by monocytes upon CD4, CCR5, and CXCR4 engagement by gp120 and cooperates with IL-4 and IL-10 to up-regulate MCLRs on B cells. Thus, gp120 may elicit polyclonal IgG and IgA responses by linking the innate and adaptive immune systems through the B cell-activating factor of the TNF family. Chronic activation of B cells through this CD40-independent pathway could impair protective T cell-dependent Ab responses by inducing immune exhaustion.

Human immunodeficiency virus 1 Nef suppresses CD40-dependent immunoglobulin class switching in bystander B cells

Immunoglobulin class switching from immunoglobulin M (IgM) to IgG and IgA is central to immunity against viruses and requires the activation of B cells by T cells via CD154 (CD40 ligand) and cytokines. These molecules limit their signaling activity in immune cells by turning on negative feedback proteins, including IkappaB and SOCS. We show here that negative factor (Nef) protein, an immunosuppressive human immunodeficiency virus 1 protein expressed and released by infected cells, penetrates B cells both in vivo and in vitro. Nef suppressed immunoglobulin class-switch DNA recombination by inducing IkappaBalpha and SOCS proteins, which blocked CD154 and cytokine signaling via NF-kappaB and STAT transcription factors. Thus, human immunodeficiency virus 1 may evade protective T cell-dependent IgG and IgA responses by 'hijacking' physiological feedback inhibitors in B cells via Nef.

Mutational analysis of PRDM1 indicates a tumor-suppressor role in diffuse large B-cell lymphomas

The PR (PRDI-BF1-RIZ) domain zinc finger protein 1 (PRDM1) is a transcription repressor with a pivotal role in plasma-cell differentiation. We identified clonal inactivating mutations in PRDM1 in the diffuse large B-cell lymphoma (DLBCL) cell line OCI-Ly3 and in 8 of 35 de novo clinical DLBCL samples. The mutational spectrum consists predominantly (7 cases) of single-nucleotide mutations affecting consensus splice donor sites, some of which are recurrent, that lead to splicing aberrations and premature translation termination. In 2 of these cases, point mutations appear to be caused by RNA editing with G-to-A and U-to-G conversions. Other mutations include frame-shift deletion and chromosomal inversion. Except for one mutant, which may act as a dominant-negative, all mutations are associated with either deletion or silencing of the paired PRDM1 allele. This study identifies PRDM1 inactivation as a recurrent genetic defect in DLBCL cells and establishes PRDM1 as a potential tumor suppressor gene in DLBCL. Moreover, it implies inhibition of terminal differentiation as a pathogenetic pathway in DLBCL, particularly for the activated B-cell-like DLBCL. It also demonstrates for the first time the potential role of RNA editing in lymphomagenesis.

Mutually Exclusive Cyclin-Dependent Kinase 4/Cyclin D1 and Cyclin-Dependent Kinase 6/Cyclin D2 Pairing Inactivates Retinoblastoma Protein and Promotes Cell Cycle Dysregulation in Multiple Myeloma

Multiple myeloma, the second most common hematopoietic cancer, ultimately becomes refractory to treatment when self-renewing multiple myeloma cells begin unrestrained proliferation by unknown mechanisms. Here, we show that one, but not more than one, of the three early G(1) D cyclins is elevated in each case of multiple myeloma. Cyclin D1 or D3 expression does not vary in the clinical course, but that alone is insufficient to promote cell cycle progression unless cyclin-dependent kinase 4 (cdk4) is also elevated, in the absence of cdk6, to phosphorylate the retinoblastoma protein (Rb). By contrast, cyclin D2 and cdk6 are coordinately increased, thereby overriding the inhibition by cdk inhibitors p18(INK4c) and p27(Kip1) and phosphorylating Rb in conjunction with the existing cdk4. Thus, cyclin D1 pairs exclusively with cdk4 and cdk6 pairs only with cyclin D2, although cyclin D2 can also pair with cdk4 in multiple myeloma cells. The basis for this novel and specific cdk/D cyclin pairing lies in differential transcriptional activation. In addition, cyclin D1- or cyclin D3-expressing multiple myeloma cells are uniformly distributed in the bone marrow, whereas cdk6-specific phosphorylation of Rb occurs in discrete foci of bone marrow multiple myeloma cells before proliferation early in the clinical course and is then heightened with proliferation and disease progression. Mutually exclusive cdk4/cyclin D1 and cdk6/cyclin D2 pairing, therefore, is likely to be a critical determinant for cell cycle reentry and progression and may play a pivotal role in the expansion of self-renewing multiple myeloma cells.

Distinct subsets of primary effusion lymphoma can be identified based on their cellular gene expression profile and viral association

Primary effusion lymphomas (PELs) are specifically associated with Kaposi's sarcoma-associated herpesvirus (KSHV) infection and most frequently occur in human immunodeficiency virus (HIV)-positive individuals as lymphomatous effusions in the serous cavities without a detectable solid tumor mass. Most PELs have concomitant Epstein-Barr virus (EBV) infection, suggesting that EBV is an important pathogenetic cofactor, although other as yet unidentified cofactors, such as cellular genetic alterations, are also likely to play a role. Lymphomatous effusions that lack KSHV also occur; these are frequently EBV associated in the setting of HIV infection. Here we used gene expression profile analysis to determine the viral impact on cellular gene expression and the pathogenesis of these lymphomatous effusions. Our results show that many genes, including cell cycle and signal transduction regulators, are differentially expressed between KSHV-positive PELs and KSHV-negative lymphomatous effusions and also between KSHV-positive, EBV-positive and KSHV-positive, EBV-negative PELs. Our results confirm that KSHV plays an important role in the pathogenesis of PELs, as its presence selects for a very distinct cellular gene expression category and a clearly different lymphoma type. Within the KSHV-positive PELs, the effect of EBV is more subtle but nevertheless clear.

KSHV-Positive Solid Lymphomas Represent an Extra-Cavitary Variant of Primary Effusion Lymphoma

Primary effusion lymphoma (PEL) is a unique form of non-Hodgkin lymphoma (NHL) associated with Kaposi sarcoma-associated herpesvirus (KSHV; HHV-8) that displays a distinct constellation of clinical, morphologic, immunologic, and molecular characteristics. Rare KSHV-containing immunoblastic lymphomas occurring in solid tissues have been described. Whether they represent part of the spectrum of PEL has not been determined. The morphologic, immunophenotypic, and molecular features of KSHV-positive solid lymphomas occurring in 8 HIV+/AIDS patients were systematically investigated and compared with those of 29 similarly analyzed PELs. The 8 KSHV-positive solid lymphomas were virtually indistinguishable from the 29 PELs based on morphology (immunoblastic/anaplastic), immunophenotype (CD45 positive; T cell antigen negative; CD30, EMA, CD138 positive; CD10, CD15, BCL6 negative) and genotype (100% immunoglobulin genes rearranged; no identifiable abnormalities in C-MYC, BCL6, BCL1, BCL2; and uniformly EBV positive). The only identifiable phenotypic difference was that the KSHV-positive solid lymphomas appeared to express B cell-associated antigens (25%) and immunoglobulin (25%) slightly more often than the PELs (<5% and 15%, respectively; P = 0.11 and P = 0.08, respectively). The clinical presentation and course of the patients who develop KSHV-positive solid lymphomas were also similar, except for the lack of an effusion and somewhat better survival (median 11 months vs. 3 months). However, the 3 KSHV-positive solid lymphoma patients alive without disease 11, 25, and 44 months following initial presentation were recently diagnosed patients and, unlike the other patients with KSHV-positive solid lymphomas, received anti-retroviral therapy. These findings strongly suggest that these decidedly rare KSHV-positive solid lymphomas belong to the spectrum of PEL. Therefore, we propose that the KSHV-positive solid lymphomas be designated extra-cavitary PELs.

Immunophenotypic and genotypic characterization of progression in follicular lymphomas

Progression of follicular lymphomas (FLs) is often accompanied by a spectrum of histologic changes and an aggressive clinical course. Although molecular alterations have been implicated in this event, the underlying factors are largely unknown. We studied the expression of selected tumor suppressor genes (P53 and retinoblastoma [RB]), oncogenes (MYC and BCL2), and a transferrin-receptor related protein (Trump) in sequential biopsies in 16 patients. Eleven patients progressed from grade I or II FL to aggressive B-cell lymphomas with diffuse morphology, whereas 5 patients presented with diffuse aggressive lymphomas and recurred with indolent lymphomas. Immunoreactivity for P53 correlated with higher histologic grade in lymphomas progressing from indolent to aggressive; however, only 1 patient who presented with aggressive lymphoma demonstrated a P53 gene mutation. Neither P53 immunoreactivity nor genotypic alterations correlated with presentation with an aggressive histology and relapse with FL. Growth fraction, as assessed by Ki-67 staining, and Trump expression correlated with histologic grade. Immunoreactivity for RB, BCL2, and MYC was seldom associated with progression. Eight of 9 cases tested exhibited identical immunoglobulin heavy and light chain rearrangements or identical BCL2 gene rearrangements in the sequential lymphomas. We conclude that P53 and Trump protein expression and proliferation activity correlate with histologic grade, but not with recurrence or progression of FL. Our results further indicate that progression of FL to diffuse aggressive lymphomas and presentation of an aggressive B-cell lymphoma followed by FL are clonally related.

Lymphoma B cells evade apoptosis through the TNF family members BAFF/BLyS and APRIL

The mechanisms underlying the autonomous accumulation of malignant B cells remain elusive. We show in this study that non-Hodgkin's lymphoma (NHL) B cells express B cell-activating factor of the TNF family (BAFF) and a proliferation-inducing ligand (APRIL), two powerful B cell-activating molecules usually expressed by myeloid cells. In addition, NHL B cells express BAFF receptor, which binds BAFF, as well as transmembrane activator and calcium modulator and cyclophilin ligand interactor (TACI) and B cell maturation Ag (BCMA), which bind both BAFF and APRIL. Neutralization of endogenous BAFF and APRIL by soluble TACI and BCMA decoy receptors attenuates the survival of NHL B cells, decreases activation of the prosurvival transcription factor NF-kappaB, down-regulates the antiapoptotic proteins Bcl-2 and Bcl-x(L), and up-regulates the proapoptotic protein Bax. Conversely, exposure of NHL B cells to recombinant or myeloid cell-derived BAFF and APRIL attenuates apoptosis, increases NF-kappaB activation, up-regulates Bcl-2 and Bcl-x(L), and down-regulates Bax. In some NHLs, exogenous BAFF and APRIL up-regulate c-Myc, an inducer of cell proliferation; down-regulate p53, an inhibitor of cell proliferation; and increase Bcl-6, an inhibitor of B cell differentiation. By showing that nonmalignant B cells up-regulate BAFF and APRIL upon stimulation by T cell CD40 ligand, our findings indicate that NHL B cells deregulate an otherwise physiological autocrine survival pathway to evade apoptosis. Thus, neutralization of BAFF and APRIL by soluble TACI and BCMA decoy receptors could be useful to dampen the accumulation of malignant B cells in NHL patients.

Etiology and pathogenesis of AIDS-related non-Hodgkin's lymphoma

The incidence of NHL is greatly increased in HIV-infected individuals; malignant lymphoma is the second most common neoplasm that occurs in association with AIDS. The vast majority of neoplasms are clinically aggressive, monoclonal B-cell neoplasms that exhibit Burkitt's, immunoblastic, large cell, or transitional histopathology. Approximately 80% arise systemically (nodal or extranodal) and 20% arise as primary CNS lymphomas. A small proportion of neoplasms are body cavity-based, primary effusion lymphomas that are uniquely associated with KSHV infection. Recently, HIV-associated polymorphic lymphoproliferative disorders have been described as well. AIDS-related NHLs appear to exhibit distinctive clinical characteristics according to their histopathology and anatomic site of origin. Factors that contribute to lymphoma development include HIV-induced immunosuppression, impaired immune surveillance, cytokine release and deregulation, and chronic antigenic stimulation. This environment is associated with the development of oligoclonal B-cell expansions. The appearance of NHL is characterized by the presence of a monoclonal B-cell population that displays a variety of genetic lesions, including, for example, EBV infection, MYC gene rearrangement, BCL6 gene rearrangement, P53 mutations and deletions, and RAS gene mutations. The number and type of genetic lesions vary somewhat among AIDS-related NHLs according to their histopathologic category and anatomic site of origin. These findings suggest that more than one pathogenetic mechanism is operational in the development and progression of AIDS-related NHLs. Further work is necessary to develop a complete understanding of the etiology and pathogenesis of NHL in the setting of HIV infection. AIDS-related NHL is an important biologic model for investigating the development and progression of high-grade NHLs and NHLs that develop in immunedeficient hosts.

Alterations of mRNA splicing in primary effusion lymphomas

Primary effusion lymphoma (PEL) is a unique form of malignant lymphoma associated with infection by the Kaposi's sarcoma-associated herpesvirus (KSHV)/human herpesvirus-8 (HHV-8). The majority of PELs also contain the EBV genome. Although viral infection is believed to play a critical role in the pathogenesis of PEL, it has been suggested that additional molecular lesions are required for the development of PEL. Alternative splicing of pre-mRNA is an important mechanism in the regulation of cellular and viral gene expression. Deregulation of pre-mRNA splicing may shift the gene expression balance and lead to the development of cancer. In order to investigate mRNA splicing in PELs, we examined mRNA splicing of three genes, DNA polymerase beta (pol beta), Bcl-x and CD45, in eight PEL cell lines. We found that the average variant percentage of pol beta in PEL cell lines is two times higher than in peripheral blood mononuclear cells (PBMC) and that the variant pattern of genes bcl-x and CD45 is quite different in PEL cell lines than in PBMC. In addition, we also found that the percentage of variant pol beta increased two-fold in PBMC following Epstein-Barr virus (EBV) infection. Therefore, viral infection may contribute to mRNA alternative splicing in PEL. In order to explore the mechanism by which viral infection affects mRNA splicing, we also examined the roles of genes KS-SM, SM and EBERs and viral copies in mRNA splicing. Our findings indicate that various factors acting as positive or negative regulators may be involved in mRNA alternative splicing caused by viral infection. In conclusion, mRNA splicing in PEL can be altered by viral infection and this alteration may contribute to the pathogenesis of PEL.

Human immunodeficiency virus (HIV)-associated polymorphic lymphoproliferative disorders

The majority of AIDS-related non-Hodgkin's lymphomas are clinically aggressive monoclonal B-cell Burkitt's lymphomas, large cell lymphomas, or immunoblastic lymphomas. In contrast, the lymphoid proliferations arising in solid organ transplant recipients, collectively referred to as posttransplantation lymphoproliferative disorders (PT-LPDs), represent a clinically and histopathologically heterogeneous group of Epstein-Barr virus (EBV)-driven B-cell proliferations of variable clonal composition. During a retrospective histopathologic review of lymphoid proliferations associated with human immunodeficiency virus (HIV) infection we identified 10 cases that morphologically resemble the polymorphic PT-LPDs. They arose in lymph nodes (five), lungs (two), and the parotid gland, perineum, and skin (one each). They exhibit a diffuse growth pattern and are composed of a polymorphic lymphoid cell population exhibiting a variable degree of plasmacytic differentiation, cytologic atypia, and numbers of atypical immunoblasts. A clonal B-cell population was detected by immunoglobulin heavy and light chain gene rearrangement and/or EBV terminal repeat analysis in 8 of the 10 (80%) cases by Southern blotting. The nongermline hybridizing bands were usually faint, however, suggesting that the clonal B-cell population represented only a subpopulation within the polymorphic lesion. Strong clonal rearrangement bands were present in one case in which there was clear morphologic evidence of transformation to diffuse large cell lymphoma. This case exhibited C-MYC, BCL-6, and p53 gene mutations. One other case exhibited a p53 gene mutation. The remaining eight cases lacked C-MYC, BCL-6, RAS, and p53 gene alterations. Clonal EBV infection was detected in 4 of the 10 (40%) lesions. Like EBV-containing PT-LPDs, all four EBV-positive HIV-associated polymorphic lesions were associated with type A EBV. The Kaposi's sarcoma-associated herpesvirus was detectable in two cases by polymerase chain reaction analysis, but not by Southern blotting. In situ hybridization demonstrated Kaposi's sarcoma-associated herpesvirus in some of the cytologically malignant-appearing cells. In conclusion, polymorphic B-cell lymphoproliferative disorders comparable morphologically and molecularly to those arising after solid organ transplantation also occur in association with HIV infection. As in the case of their polymorphic PT-LPD counterparts, their malignant status, biologic significance, and relationship to monomorphic B-cell lymphomas remain to be elucidated.

DNA polymerase mu gene expression in B-cell non-Hodgkin's lymphomas: an analysis utilizing in situ hybridization

DNA polymerase mu (pol mu) is a novel error-prone DNA repair enzyme bearing significant structural homology with terminal deoxynucleotidyltransferase. Whereas other human error-prone DNA polymerases identified thus far show no preferential lymphoid tissue distribution, the highest levels of pol mu mRNA have been detected in peripheral lymphoid tissues, particularly germinal center B cells. Conceivably, up-regulation of the pol mu gene may be biologically significant in lymphomagenesis, especially in the development of B-cell non-Hodgkin's lymphomas (B-NHLs), because of enhanced error-prone DNA repair activities. To explore this possibility, we generated a digoxigenin-labeled riboprobe to pol mu mRNA and used the probe and in situ hybridization to examine the expression pattern of the pol mu gene in formalin-fixed, paraffin-embedded tissue sections of 37 B-NHLs. This included eight chronic lymphocytic leukemia/small lymphocytic lymphomas, six mantle cell lymphomas, seven follicular lymphomas, nine diffuse large B-cell lymphomas, three splenic marginal zone lymphomas, two Burkitt's lymphomas, and two precursor B-lymphoblastic lymphomas. We also correlated the pol mu mRNA expression levels with the tumor proliferation index, which was assessed in each case by image analysis of Ki-67 immunostained slides. Nineteen of 21 (90%) B-NHLs arising from postgerminal center B cells (follicular lymphomas, diffuse large B-cell lymphomas, splenic marginal zone lymphomas, and Burkitt's lymphomas) exhibited high expression of pol mu mRNA. In contrast, only 2 of 16 (13%) B-NHLs arising from pregerminal center B cells (chronic lymphocytic leukemia/small lymphocytic lymphomas, mantle cell lymphomas, and precursor B-lymphoblastic lymphomas) expressed significant levels of pol mu mRNA. Pol mu gene expression did not seem to correlate with the proliferation index, especially because a significant level of pol mu mRNA was not detected in either case of precursor B-lymphoblastic lymphomas. In conclusion, pol mu gene expression is highly associated with B-NHLs of postgerminal center B-cell derivation. Furthermore, the expression level is independent of the proliferation rate and thus is unrelated to the biological aggressiveness of the tumors. These findings, along with the error-prone nature of the enzyme, suggest that up-regulation of pol mu gene expression may be a contributing factor to the pathogenesis of a subset of B-NHLs through DNA repair-associated genomic instability.

Tumours of histiocytes and accessory dendritic cells: an immunohistochemical approach to classification from the International Lymphoma Study Group based on 61 cases

Neoplasms of histiocytes and dendritic cells are rare, and their phenotypic and biological definition is incomplete. Seeking to identify antigens detectable in paraffin-embedded sections that might allow a more complete, rational immunophenotypic classification of histiocytic/dendritic cell neoplasms, the International Lymphoma Study Group (ILSG) stained 61 tumours of suspected histiocytic/dendritic cell type with a panel of 15 antibodies including those reactive with histiocytes (CD68, lysozyme (LYS)), Langerhans cells (CD1a), follicular dendritic cells (FDC: CD21, CD35) and S100 protein. This analysis revealed that 57 cases (93%) fit into four major immunophenotypic groups (one histiocytic and three dendritic cell types) utilizing six markers: CD68, LYS, CD1a, S100, CD21, and CD35. The four (7%) unclassified cases were further classifiable into the above four groups using additional morphological and ultrastructural features. The four groups then included: (i) histiocytic sarcoma (n=18) with the following phenotype: CD68 (100%), LYS (94%), CD1a (0%), S100 (33%), CD21/35 (0%). The median age was 46 years. Presentation was predominantly extranodal (72%) with high mortality (58% dead of disease (DOD)). Three had systemic involvement consistent with 'malignant histiocytosis'; (ii) Langerhans cell tumour (LCT) (n=26) which expressed: CD68 (96%), LYS (42%), CD1a (100%), S100 (100%), CD21/35 (0%). There were two morphological variants: cytologically typical (n=17) designated LCT; and cytologically malignant (n=9) designated Langerhans cell sarcoma (LCS). The LCS were often not easily recognized morphologically as LC-derived, but were diagnosed based on CD1a staining. LCT and LCS differed in median age (33 versus 41 years), male:female ratio (3.7:1 versus 1:2), and death rate (31% versus 50% DOD). Four LCT patients had systemic involvement typical of Letterer-Siwe disease; (iii) follicular dendritic cell tumour/sarcoma (FDCT) (n=13) which expressed: CD68 (54%), LYS (8%), CD1a (0%), S100 (16%), FDC markers CD21/35 (100%), EMA (40%). These patients were adults (median age 65 years) with predominantly localized nodal disease (75%) and low mortality (9% DOD); (iv) interdigitating dendritic cell tumour/sarcoma (IDCT) (n=4) which expressed: CD68 (50%), LYS (25%), CD1a (0%), S100 (100%), CD21/35 (0%). The patients were adults (median 71 years) with localized nodal disease (75%) without mortality (0% DOD). In conclusion, definitive immunophenotypic classification of histiocytic and accessory cell neoplasms into four categories was possible in 93% of the cases using six antigens detected in paraffin-embedded sections. Exceptional cases (7%) were resolvable when added morphological and ultrastructural features were considered. We propose a classification combining immunophenotype and morphology with five categories, including Langerhans cell sarcoma. This simplified scheme is practical for everyday diagnostic use and should provide a framework for additional investigation of these unusual neoplasms.

Expression of CD56/neural cell adhesion molecule correlates with the presence of lytic bone lesions in multiple myeloma and distinguishes myeloma from monoclonal gammopathy of undetermined significance and lymphomas with plasmacytoid differentiation

Unlike monoclonal gammopathy of undetermined significance (MGUS) or non-Hodgkin's lymphomas (NHLs) with plasmacytoid differentiation, multiple myeloma (MM) is commonly associated with lytic bone lesions. Although the mechanisms of increased osteoclast activity are partially understood, comparatively little is known about the mechanisms that lead to the observed decrease in osteoblast function. Studies have shown neural cell adhesion molecule (NCAM) homophilic binding between MM cell lines and osteosarcoma cell lines, and that binding results in decreased osteoid production in vitro. Thus, we postulated that the expression of NCAM by MM cells contributes to lytic lesion formation by causing decreased osteoid production in vivo. We used immunohistochemistry in bone marrow core biopsies to assess NCAM expression in osteoblasts and plasma cells (PCs) in vitro. We found consistent, strong, uniform NCAM expression by the osteoblasts in all bone marrow core biopsies (352 of 352, 100%). Strong expression of NCAM by PCs correlated with the presence of lytic bone lesions (chi-square, 33.39: P <0.000; odds ratio, 16.9). There was also a strong correlation between NCAM expression and the diagnosis of MM in comparison to reactive PCs, MGUS, or NHLs with plasmacytoid differentiation (all P values <0.000). In conclusion, using immunohistochemistry, we found strong expression of NCAM by osteoblasts and that when equal to the intensity of osteoblast expression, NCAM expression by PCs correlates with the presence of lytic bone lesions and distinguishes MM from reactive plasmacytosis, NHLs with plasmacytoid differentiation, and most cases of MGUS.

Expression of Kaposi's sarcoma-associated herpesvirus G protein-coupled receptor monocistronic and bicistronic transcripts in primary effusion lymphomas

Kaposi's sarcoma-associated herpesvirus (KSHV/HHV-8) encodes a G protein-coupled receptor (vGPCR) in open reading frame (ORF) 74, which is homologous to human chemokine receptors. KSHV vGPCR is constitutively active and induces VEGF-mediated angiogenesis. Previous studies have shown that ORF 74 is transcribed as part of a bicistronic message containing ORF K14 upstream of ORF 74, with an early lytic pattern of expression. We have now extended these studies by analyzing three different KSHV-positive primary effusion lymphoma (PEL) cell lines and three PEL clinical samples. In addition, we have identified another less abundant monocistronic transcript containing only ORF 74. Both transcripts were identified at low but similar levels in two PEL clinical samples. We evaluated the degree of sequence and functional conservation of ORF74 in three additional PELs and two KS clinical specimens, demonstrating complete identity at the amino acid level among all isolates. While it is expressed as an early lytic transcript in PEL cell lines, in primary clinical PEL samples transcription of KSHV vGPCR can be readily detected.

Polymerase chain reaction detection of Kaposi's sarcoma-associated herpesvirus-optimized protocols and their application to myeloma

Since its discovery in 1994, KSHV (also called human herpesvirus-8 or HHV8) has been implicated in a variety of disorders. Although the association of KSHV with Kaposi's sarcoma (KS), primary effusion lymphoma (PEL), and multicentric Castleman's disease has been well established, its presence in some other diseases, such as multiple myeloma, remains controversial. Because most KSHV studies are based on polymerase chain reaction (PCR) analysis, the conflicting data may be attributable to variations in the methods, primer sets, and target sequences selected. To establish an efficient and reliable PCR approach for KSHV detection we designed eight sets of primers to six regions (ORFK1, ORFK2, ORFK9, ORK26, ORF72, and ORF74) of the KSHV genome using appropriate database and software. The detection sensitivity of these primers was carefully assessed and their reliability was strictly validated in a series of positive (15 KS and PEL samples) and negative (16 lymphoid tissues) controls. We found that primer sets to the ORFK9 region showed the highest sensitivity, whereas primer sets to ORFK1 and ORF74 showed the lowest sensitivity. Primer sets to ORFK9, ORF26 and ORF72 regions detected all of the positive cases, whereas other primer sets showed varying detection rates or nonspecific bands. All 16 negative controls were negative with all primer sets. However, six of 16 negative controls became positive when we used nested PCR targeting ORF26. Therefore, multiple target KSHV sequences increase the detection efficiency, while nested PCR protocols are likely to introduce false positivity. Using ORFK9, ORF26 and ORF72 primer sets, we screened bone marrow biopsies from 18 cases of multiple myeloma, and failed to detect any KSHV sequences. This finding supports the conclusion that KSHV is not associated with multiple myeloma. Indeed, our results further confirm that although KSHV is universally present in Kaposi's sarcoma and primary effusion lymphoma, it is not ubiquitious.

BCL-6 protein is expressed in precursor T-cell lymphoblastic lymphoma and in prenatal and postnatal thymus

The organization and expression of the BCL-6 gene in normal and neoplastic thymic T cells has not been fully determined. We examined 8 precursor T-cell lymphoblastic lymphomas (T-LBLs) and found significant BCL-6 expression in 4 cases. Three of the BCL-6(+) cases expressed a common thymocyte phenotype (CD4(+), CD8(+)), and one expressed a precursor thymocyte phenotype (CD4(-), CD8(-)). In 6 cases evaluated, including those expressing BCL-6, molecular analyses demonstrated a germline configuration of the BCL-6 gene and a wild-type BCL-6 gene first exon-intron boundary region. We also evaluated 12 normal prenatal and postnatal thymuses for BCL-6 protein. BCL-6 was expressed by most cortical thymocytes and by scattered medullary thymocytes. BCL-6(+) cortical and medullary thymocytes also expressed CD2, CD3, CD4, CD5, CD7, or CD8. We further analyzed the pattern of BCL-2 and BCL-X(L) expression and their coexpression with BCL-6 in normal thymus and T-LBL and compared it to that of follicle centers of reactive lymph nodes and follicular lymphoma. BCL-6(+) cortical thymocytes coexpressed BCL-X(L) but not BCL-2. All 4 BCL-6(+) T-LBLs and 4 BCL-6(-) T-LBLs coexpressed BCL-2 and BCL-X(L). Conceivably, T-LBLs may arise through clonal expansion of cortical thymocytes normally expressing the BCL-6 protein. The pattern of BCL-6, BCL-2, and BCL-X(L) expression in cortical thymocytes is highly reminiscent of germinal centers, and the abnormal coexpression of BCL-2, BCL-X(L), and BCL-6 in T-LBL is analogous to coexpression in follicle center cell lymphomas, suggesting that coexpression of these anti-apoptotic genes may contribute to the pathogenesis of T-LBL.

Ongoing hypermutation in the Ig V(D)J gene segments and c-myc proto-oncogene of an AIDS lymphoma segregates with neoplastic B cells at different sites: implications for clonal evolution

To investigate the role of somatic Ig hypermutation in the evolution of AIDS-associated B cell lymphomas, we analyzed the Ig V(D)J and c-myc genes expressed by neoplastic B cells in two extranodal sites, testis and orbit, and clonally related cells in the bone marrow. Testis and orbit B cells expressed differentially mutated but collinear V(H)DJ(H), V kappa J kappa and c-myc gene sequences. Shared mutations accounted for 10.2%, 8.4%, and 4.3% of the overall V(H)DJ(H), V kappa J kappa, and c-myc gene sequences. Tumor-site specific V(H)DJ(H), V kappa J kappa, and c-myc mutations were comparable in frequency, and a single point-mutation gave rise to an EcoRI site in the testis c-myc DNA. Both shared and tumor site-specific V(H)DJ(H), V kappa J kappa, and c-myc mutations displayed predominance of transitions over transversions. The "neoplastic" V(H)DJ(H) sequence was expressed by about 10(-5) cells in the bone marrow, and contained two of the three orbital, but none of the testicular V(H)DJ(H) mutations. The nature and distribution of the Ig V(D)J mutations found in the kappa chain suggested a selection by antigen in testis and orbit. Our data suggest that, in AIDS-associated B cell lymphomas, the Ig hypermutation machinery targets V(H)DJ(H), V kappa J kappa, and c-myc genes with comparable efficiency and modalities.