Annexe 2 : complément à la discussion générale. Persistent Polyclonal B Cell lymphocytosis: The Making of a Lymphoma? An immunological perspective. (Version intégrale)

Table des matières

Marguerite Massinga Loembé1, Robert Delage2 and André Darveau1 .

1CREFSIP, Département de Biochimie et Microbiologie, Université Laval, Québec, Canada.

2 Centre d'Hématologie et d'Immunologie Clinique, CHA, Pavillon St-Sacrement, Québec, Canada.

Address correspondence to:

André Darveau, CREFSIP, Pavillon Marchand, Université Laval, Québec, G1K 7P4, Qc, Canada. Phone: (418) 656-2131, ext.3214. Fax: 656-7176. E-mail: adarveau@rsvs.ulaval.ca.

First described twenty years ago, persistent polyclonal B cell lymphocytosis (PPBL) has since stirred much perplexity among researchers and clinicians. Diagnosed principally in adult female smokers, this unusual haematological disorder of B lymphocytes shares both features of benignity (polyclonal expansion, polyconal IgM secretion, lack of significant clinical symptoms, stable and mostly uneventful course) and features of malignancy: (morphologically atypical binucleated cells, bcl-2/Ig genes translocations, chromosome 3 anomalies, bone marrow involvement and, in rare patients, concurrent occurrence of neoplasia). As explained in the present review, the polyclonal nature of PPBL and the apparent heterogeneity of the involved cellular population has long impeded phenotypic and functional characterization of the disorder. The last years however have witnessed tremendous progresses in the field of B lymphocyte immunobiology. Technical inputs from the molecular field have lead to a better discrimination of developmentally distinct B cells subsets and, by extension, of B cell lymphoid disorders. Numerous molecular mechanisms and enzymatic factors involved in peripheral B cell maturation have been uncovered. These achievements have yielded a better understanding of the lymphomagenesis process. As far as PPBL is concerned, formal characterization of the expanding IgM+IgD+ memory B lymphocyte subset has ensued and a better delineation of its clinical definition has emerged. Indeed, deceitful cytogenetic and histologic features in PPBL have formerly exposed some patients to unneeded intrusive therapy, emphasizing the requirement for proper diagnosis of this syndrome. Moreover, it is still unclear at this point whether progression to malignancy represent the long term issue in PPBL, making attentive monitoring of patients mandatory. Nevertheless, as testified by is virtual absence from recent haematology manuals, or by the lack of recent comprehensive review on the subject, PPBL has not yet gained mainstream recognition as a distinct physiopathological entity. In this paper, our aim is to provide a retrospective view of the research realizations during these last twenty years and an up-to-date picture of PPBL.

Nearly two decades ago, Gordon and collaborators were the first to report an unusually stable proliferative disorder of B lymphocytes, which they termed persistent polyclonal lymphocytosis of B lymphocytes, or PPBL (Gordon , et al 1982). The syndrome, observed in three adult female smokers, consisted in a marked elevation of peripheral lymphocytes numbers. PPBL involved mainly atypical B lymphocytes, medium sized, with an abundant cytoplasm and indented to fully bilobulated nuclei, and it was accompanied by a rise in serum IgM, with low IgG and IgA. Lymphocytosis is typically a transient event in healthy adults, often related to benign causes such as infectious reactions (see Granados , et al 1998 for a review). Persistent lymphocytosis, on the other hand, is frequently the sign of a neoplastic condition, with a clonal origin and a malignant evolution (i.e.: B-chronic lymphocytic leukaemia or B-CLL, hairy cell leukaemia or HCL, prolymphocytic leukemia or PLL, Waldenstrom’s macroglobinaemia, the leukemic phase of some non Hodgkin lymphoma) (Hoffbrand , et al 2001). Yet, unlike most persistent lymphoproliferations of B cells, PPBL was remarkable by its polyclonal nature, evidenced by the expression of both λ and κ immunoglobulin light chains in the B cell population, and its extreme stability, with up to 25 years follow-up. Few clinical signs were associated with the syndrome, except for occasional lymphoid hyperplasia (mostly splenomegaly) and tobacco related respiratory track irritation or minor infections. The fact that all three patients in the study expressed the HLA-DR7 antigen, only observed in ~20% of the Caucasian population (Linet , et al 1988), was especially noteworthy, and hinted at a genetic predisposition to the syndrome. Moreover, the atypical lymphocyte morphology suggested the involvement of an infectious agent in the aetiology of PPBL. No virus could be detected by means of electron microscopy, nevertheless there were serological evidences of past infection with the Esptein-Barr virus (EBV), cytomegalovirus (CMV) and herpes simplex virus (HSV) in all patients. According to the authors, this lymphocytosis constituted a new disorder, most unusual among typically malignant B cell lymphoproliferations, probably benign, but which cause still needed to be fully elucidated.

Owing mostly to the lack of physical symptoms, PPBL was at first seldom diagnosed and considered a rare entity. The presence of binucleated B lymphocytes, a regular observation in patients, exceptional in normal individuals, soon became a hallmark of the lymphocytosis, and enabled the identification of several new cases by way of blood smear examination (see figure 19). The years have subsequently proven PPBL to be more prevalent than initially thought: as of 2004, approximately 118 cases have been described in the literature worldwide (see table 13 for an overview). These additional reports have confirmed the marked predominance of the feminine gender (104/118), positive EBV serology (52/82 tested), HLA-DR7 (81/118) phenotype and smoking in PPBL (113/118), making them all prospective susceptibility factors. Nevertheless, the syndrome was also diagnosed in a few men (13/118) and one infant (Gomez , et al 2000), in non-smoking patients (4/118) and in individuals negative for the HLA-DR7 haplotype (18/118) (see table 13, column 2 and 5). The extent to which each of the aforementioned features contributes to the pathogenesis of PPBL has been the subject of investigations and debates. However, as we discuss in the first part of this review, evidences that have accumulated since the first PPBL case reports only allow a partial resolution of the enigma.

Despite the fact that B lymphocytes account for the origin of 80% of all lymphoma types (Isaacson 2000), study of the different B cell subsets, particularly the mature pool, has long remained an uncharted territory. Accordingly, early lymphoma classification systems, such as the Rappaport classification (1970) or the Working Formulation (1982), relied mainly on the histological properties of malignant cells and clinical progression for the identification of B cell haematological disorders. However, owing to an inherent lack of flexibility, these methods did not always allow proper discrimination between distinct neoplastic entities (Isaacson 2000). The incorporation of basic phenotypic data (cell surface receptors), as illustrated in the Kiel Classification (1988), and of disease specific genetic features (chromosomal anomalies) later offered a supplementary level of analysis. Contrastingly, the recent years have witnessed a tremendous influx of scientific data regarding B cell developmental biology. Technical advances in the field of molecular genetics (i.e.: exhaustive Ig heavy chain variable region [VH] analysis at the single cell level, microarrays technology), along with the discovery of new cell surface markers, have had a considerable impact on the categorisation of the different human B cell compartments. Uncovering of enzymatic actors directing B cell maturation (exemplified lately by the exciting discovery of the AID cytidine deaminase) and clarification of implicated molecular processes have yielded tremendous insights into B cell immunobiology. Establishing a complete phenotypic and molecular fingerprint for a monoclonal lymphoid disorder is now feasible. At the same time, the notion that lymphoid malignancies are in a large part genetic disorders, resulting from the stepwise accumulation of anomalies in developmentally crucial genes, has gained wider acceptance among the scientific and medical communities (Dolcetti and Boiocchi 1996). As a result, elaborate immunophenotyping and molecular analysis are steadily becoming mainstream tools for the scientific investigation of lymphoid disorders, a reality clearly attested by a rapid overview of recent publications in this domain. With the lineage of a particular malignant clone, and provided with detailed molecular analysis of the B cell antigen receptor and/or molecular profiling, further attempts can be made to retrace the natural history of the neoplasm and even, to a certain level, to predict its clinical evolution according to the functional properties of the proposed normal cellular progenitor. This has been well illustrated with B-CLL, were distinct diagnostic categories were shown to be directly correlated with the molecular status (mutated versus unmutated) of VH genes. The clinical relevance of this analytical approach has been further demonstrated by the elaboration of new systems, namely the REAL (1994) and WHO (2001) lymphoma classifications, which rely not only on morphologic and clinical features, but further encompass immunophenotypic and molecular properties in the definition of a lymphoid neoplasm (Isaacson 2000). Contrarily to their predecessors, these methods have achieved an international consensus and demonstrated an appreciable efficiency.

The second part of this review will focus on the immunological basis for the phenotypic and molecular fingerprinting of a given B cell clone, and illustrate how this approach led to the discovery that the expanding B cell population in PPBL consisted in polyclonal, somatically mutated IgM+IgD+CD27+ memory B cells. The hypothesis that can be put forward regarding the lineage of these cells and the possible cause(s) behind the lymphocytosis will be explored. Finally a survey will be made of the avenues that can tentatively be offered to answer the question which probably is of most concern to patients, clinicians and scientists alike: despite an apparent indolent course, is there a risk for long term malignant evolution in PPBL patients?

The main challenge facing researchers and clinicians was the identity of the expanding B cell population in PPBL: was the lymphocytosis the result of a global expansion of B lymphocytes or only that of a specific subset? Evaluation of cellular morphology and immunophenotyping, a classical approach for the characterisation of malignant clones, was undertaken (table 13, column 9). Steady expression of the pan B cell markers CD19, CD20 CD22, CD24 and FMC7 corroborated the B lineage of the lymphocytosis. Expression of CD5 (B1 B cell marker), CD23 (naïve B cell activation marker) and CD38 (immature and germinal center [GC]B cell marker), often used for the differential diagnosis of CLL (Chiorazzi and Ferrarini 2003), was negative, as did expression of GC marker CD10. When tested, the CD103 marker, specific for both HCL and a polyclonal counterpart (hairy B cell lymphoproliferative disorder or HBLD) (Machii , et al 1997), was negative. CD11c, a second HCL and HBLD specific marker, was variably expressed. Most B lymphocytes displayed surface IgM and IgD which, according to the Bm1-5 classification system of B lymphocytes subsets (Liu and Arpin 1997, Pascual , et al 1994), pointed to a Bm1 or naïve phenotype. Contrastingly, features consistent with cellular activation were also reported: a high percentage of cells (up to 50%) presented with an abundant, basophilic cytoplasm, enlarged, often indented nucleus or apparently two completely separated nuclei (figure 19). Expression of activation marker CD25 (IL-2 receptor α) and of CD21 (EBV and complement receptor) was frequently observed. In addition, cytoplasmic immunoglobulin positivity, suggesting an early plasma cell phenotype, was documented (Feugier , et al 2004, Gordon , et al 1982, Ide , et al 2002). The lack of consistency of those observations illustrates the limitations resulting from the polyclonal nature of the disorder, and the probable involvement of a heterogeneous B cell population. Karyotyping and molecular analysis have further disclosed genetic anomalies within the B cell compartment in PPBL (table 13, column 8): premature chromosome condensation (PCC), isochromosome +i(3q) and multiple bcl-2/Ig genes rearrangements, were recurring findings in most cases tested. Yet again, these two last chromosomal anomalies were distributed randomly in the B lymphocyte population, regardless of the morphological aspect of the cell, whether binucleated (Callet-Bauchu , et al 1999, Callet-Bauchu , et al 1997, Espinet , et al 2000, Lancry , et al 2001, Mossafa , et al 1999). Thus, at that point, the exact identity of the expanding subset in PPBL remained elusive.

The high predominance of smokers among PPBL patients (114/118, table 13, column 2) has raised questions about the possible involvement of tobacco in the aetiology of the syndrome. These suspicions appeared to be supported early on, after the lymphocytosis had apparently resolved in one patient when she ceased smoking, but subsequently returned upon tobacco reintroduction (Carstairs , et al 1985). As additional PPBL cases were identified, normalization of the lymphocytosis, as well as diminution of serum IgM levels, was correspondingly observed in other patients after they quit smoking (Bain , et al 1998, Rodriguez , et al 1996, Tonelli , et al 2000). As a result, tobacco has been designated a causal factor for the disorder. However, tobacco cessation has not always yielded decreased B cell counts, and cytogenetic abnormalities have persisted (our own unpublished informations in three patients and Mossafa , et al 1999). Furthermore, atypical bilobulated B lymphocytes have remained present in ex-smoking patients, even in the absence of measurable absolute lymphocytosis (Bain , et al 1998, Himmelmann , et al 2001b, Rodriguez , et al 1996). This is especially noteworthy, as it indicates that the syndrome may well be present but goes unnoticed in many individuals. Meanwhile, control screenings of asymptomatic heavy smokers could not reveal the presence of characteristic bilobulated cells (Troussard and Flandrin 1996) or multiple bcl-2/Ig gene rearrangements (Delage , et al 2001). Finally, heavy smoking in healthy individual is significantly associated with lowered IgM levels (Mili , et al 1991, Moszczynski , et al 2001), making tobacco an unlikely suspect to explain the IgM elevation constantly observed in PPBL (Vignes , et al 2000). As the disorder has also been observed in non smokers, it therefore seems that, though tobacco undoubtedly exacerbates the distinctive PPBL associated polyclonal B lymphocyte expansion, it is probably not a key factor in the aetiology of the disease.

Detection of the HLA-DR7 haplotype, present homozygously among the first three identified PPBL cases (Gordon , et al 1982), and either homozygously or heterozygously in most of the later cases (99/118, table 13, column 5), was intriguing given the low predominance of this antigen in the Caucasian population (~20%). It suggested a genetic predisposition to the disease. Interestingly, the HLA-DR7 has been associated with persistent infection with hepatitis B virus (Almarri and Batchelor 1994), possibly due to inadequate recognition by cytotoxic T cells, a phenomenon which, according to some authors, could likewise contribute to the apparent implication of EBV in PPBL (Chow , et al 1992, Mitterer , et al 1995). The finding of bilobulated lymphocytes in two of one patient’s relatives (Troussard , et al 1994) and, more notably, the occurrence of PPBL in identical twins (Carr , et al 1997), has reinforced the notion of a genetic component in the disease. Ensuing investigations have provided decisive arguments in favour of a familial inheritance pattern in PPBL. The study by Delage et al (Delage , et al 2001), that took advantage of a patients’ large family of nine first degree relatives, fourteen second degree relatives, and span three generations overall, has established that PPBL-associated criteria such as elevated serum IgM and B cells numbers, but particularly multiple bcl-2/Ig genes rearrangements, are more frequent among family members. As a matter of fact, it has led to the identification of two more PPBL cases among the patient’s siblings. Singularly, the HLA-DR7 haplotype was absent in this family. A second study was conducted in a family of four siblings, which all presented PPBL associated risk factors (EBV positive serology, HLA-DR7 antigen expression, tobacco usage) (Himmelmann , et al 2001b). A second study was conducted in a family of four siblings, which all displayed PPBL associated risk factors (EBV positive serology, HLA-DR7 antigen expression, tobacco usage) (Himmelmann , et al 2001b). Again, two of the siblings presented with the disorder, emphasizing the plausibility of a familial inheritance pattern. Nevertheless, the fact that two individuals in this family have remained unaffected raises the possibility that different or additional factors could be involved in the pathogenesis of PPBL. In both studies, relatives affected by PPBL either had a normal total lymphocytes count or a mild lymphocytosis. This again stresses the fact that the number of officially reported PPBL cases, which were referred and diagnosed as a result of elevated cell counts, only give a conservative estimate of the actual frequency of this disorder in the general population.

As the peculiar B cell morphology in PPBL suggested a viral infection, attempts were rapidly made to ascertain which infectious factors could play a role in its pathogenesis. Although, as mentioned, no viral agent had been detected by electron microscopy, serologic assays indicated past EBV infection in a majority of reported cases (52/82 tested, see table 13, column 7). Nevertheless, in the absence of appropriate controls, individual measurements of positive EBV serology did not warrant the implication of the virus in the disorder (Wyatt and Coyle 1991). Moreover, this observation is in line with the fact that the virus is carried asymptomatically by more than 90% of the adult population as a lifelong persistent infection (Crawford 2001) . Meanwhile, detection of viral genomes by in situ hybridation and PCR amplification in PPBL patients has been inconsistent (see table 13, column 7). Still, EBV was considered the most plausible candidate, owing to its capacity to drive B cell polyclonal proliferation and its frequent aetiological association with lymphoproliferative diseases, notably in immunosuppressed individuals (post-transplant and AIDS patients) (Crawford 2001, Thorley-Lawson and Gross 2004). The apparent association between PPBL, tobacco, a suspected inducer of EBV active infection, and the HLA-DR7 phenotype, involved in the immune evasion of virally infected cells, has reinforced the notion of a causal chain leading to the emergence of the lymphocytosis (Mitterer , et al 1995). This hypothesis is nonetheless challenged by the fact that, as we previously pointed out, tobacco usage and the HLA-DR7 haplotype are not shared by all PPBL patients. Moreover, when assayed in vitro , tobacco constituents have had no detectable effect upon active EBV replication (Jenson , et al 1999), whereas immune evasion of EBV has only been reported in relation with the HLA-A11 haplotype (de Campos-Lima , et al 1993).

Actual chronic active EBV infection was reported in one patient (Mitterer , et al 1995), from which a stable lymphoblastoid cell line was derived (Larcher , et al 1995) and functional studies conducted. Whether this cell line is representative of the disorder can be questioned, as it does not express surface IgD, contrarily to the expanding B cell subset in PPBL patients. Potentially more significant results arose from the molecular analysis of the carboxy-terminal region of the virus latent LMP1 gene. It indeed led to the identification of a unique 69 bp deletion that was not observed in four of the patient’s siblings, despite evidence of past EBV infection. In these healthy relatives, distinct 30 bp deletions and point mutations were observed, reflecting the natural polymorphism of the LMP1 gene in the general population (Khanim , et al 1996). The 69 bp deletion, located within LMP1 NF-kB activation domain, was nonetheless susceptible to profoundly alter the signalling properties of this CD40 analog (Kilger , et al 1998). The authors have accordingly proposed that the variant form of the LMP1 oncoprotein, through its effect on B cell physiology in infected patients, could lead to PPBL (Larcher , et al 1997). In our experience, amplification and sequencing of LMP1 carboxy-terminal region in nine PPBL patients has disclosed the presence of independent point mutations in every one. Moreover, in four of them, the 30 bp deletion detected in relatives of the index case was also identified. In opposition, the 69 bp mutation was never found (Carle Ryckmann, unpublished results). Altogether, these results challenge the hypothesis of an infection with a specific variant EBV strain that could be implicated in the pathogenesis PPBL by way of impaired LMP1 signalling.

Following primary infection, EBV persists in the host through latent infection of peripheral blood memory B cells. The virus is narrowly detected in the IgD-CD27+ memory B lymphocytes pool (Joseph , et al 2000) and latent gene expression is null, excepted for the EBNA1 protein in dividing cells. Conversely, LMP1 latent protein expression is reported in naïve and activated GC B cells in the lymph nodes (Thorley-Lawson and Gross 2004). When tested, amplification of the latent viral genes LMP1 and EBNA1, using a sensitive nested RT-PCR technique, invariably led to their amplification from PBMC in PPBL patients (LMP1:,7/7, EBNA1:7/7 ), whereas they were seldom detected in healthy controls (LMP1: 0/7, EBNA1:1/7) even when using DNA from purified B cells (Carle Ryckmann, unpublished results). It can be argued that expansion of the CD27+ memory B cell subset in PPBL patients could explain the preferential detection of the virus in PPBL patients. However some questions remain unanswered. It is somewhat puzzling that patients, who display an expansion of the IgD+ memory fraction, which allegedly excludes latent infection, show a parallel increased expression of latency associated viral genes. Moreover, it is not clear at this point what the significance of LMP1 expression in this specific memory compartment is. Subsetting memory B cells based on surface Ig expression, and further molecular investigations will undoubtedly help clarify the meaning of these observations.

Overall, the various studies conducted over the last twenty years have led to the accumulation of evidences for a genetic basis in PPBL. However, the precise mode of transmission, as well as the exact factor or combination of factors required for full expression of the disease, are still the subject of speculations and, for the most part, remained to be elucidated.

B cell maturation is punctuated by developmental checkpoints aiming at the expression of a functional receptor on the cell surface and its subsequent remodelling in order to provide the organism with the means to respond efficiently to antigenic challenge. At the molecular level, B cell maturation is reflected by sequential molecular modifications of the Ig gene locus. At the immunophenotypic level, differential expression of CD cell surface markers provides supplementary means to characterize developmentally distinct B cells.

B cell ontogeny in humans is classically divided into the antigen-independent and the antigen-dependent developmental stages (overview provided in Duchosal 1997). The antigen-dependent stage takes place in the foetal liver and the bone marrow. It allows the successful assembly of the structural elements composing the Ig molecule, and its ensuing expression at the cell surface as part of the B cell receptor (BCR). In particular, elements of the Ig heavy (VH) and light (VL) chains variable domain are first brought together before being joined to a constant domain. These elements, namely the variable (V), diversity (D, VH only) and joining (J) gene segments, are scattered on non adjacent parts of the chromosome. Accordingly, expression of a functional Ig receptor is contingent upon germline DNA rearrangement, or V(D)J gene segments recombination. Introduction of DNA double strand breaks, by the recombination activating enzymes RAG1 and RAG2, initiates the V(D)J recombination process which is next resolved by enzymes from the ubiquitous non-homologue ends-joining repair apparatus. The key to antibody diversity lays in the fact that V, D and J segments are encoded by multiple distinct gene copies which are assembled in a purely random fashion. Further variation is introduced by nucleotides additions or deletions at junction sites between segments as a result of the terminal deoxynucleotidyl transferase (TdT) enzyme activity. Hence, each B cell clone is provided with a unique Ig gene combination and the capacity to specifically recognise a singular antigenic determinant. The immune repertoire thus generated is referred to as the primary immune repertoire.

Mature naïve B cells, the outcome of the antigen-dependent differentiation stage, represent ~60% of circulating B lymphocytes, express surface IgM and IgD with (~45%) or without (~15%) CD5 (see figure 20 and Klein , et al 1998). The IgM+IgD+CD5+ subset (or B-1 cells) represent a presumably distinct lineage that is associated with the natural, T-independent (TI), immune response (Youinou , et al 1999). IgM+IgD+CD5-, on the other hand, recirculate actively between the follicules of secondary lymphoid organs until antigen encounter, interaction with an antigen-specific T cell, and initiation of the antigen-dependent development stage. Once activated, B cells either migrate to the T-cell zone of lymphoid organs (the extra-follicular region) and differentiate to low affinity short-lived plasmocytes, or they seed a follicule and give rise to a germinal centre (GC) (Jacob and Kelsoe 1992, Liu , et al 1991). Surrounding unactivated naïve B lymphocytes within the follicule are displaced to the periphery and form the follicular mantle zone (Liu , et al 1991). Inside the GC specialized microenvironment, B cells will be submitted to affinity maturation of their Ig receptor, and the secondary immune repertoire will be generated. Activated naïve B lymphocytes first undergo massive clonal expansion and differentiate to centroblasts, forming the GC dark zone. During this stage, the Ig receptor is diversified by somatic hypermutation (SH), which introduces random point mutations within the variable domains of the Ig H and, to a lesser extent, L chains (Jacob , et al 1991, Kuppers , et al 1993). Thereafter, centroblasts differentiate to centrocytes and migrate to the light zone of GC. SH operates at random, and yields B cell mutants with either improved, lessen or lost affinity for the Ag. Correspondingly, an antigen-driven selection mechanism, whose specifics will be discussed in more details in a latter section, ensures that further centrocyte differentiation and entry into the effector compartment (plasma or memory B cells) is restricted to mutants with increased Ag-binding capacity. Imprint of such a selection process is visible at the molecular level as replacement mutations within the framework regions (FR) of V segments that are essential to antibody structure will be counter-selected. Silent mutations on the other hand will be favoured, yielding a lower R/S ratio than that expected by chance only (statistical methods for analysis of mutations distribution within V domains of Ig genes are presented in Lossos , et al 2000). Some mutants B cells will additionally switch to the expression of alternative (downstream) constants domains, conferring the Ig receptor new functional properties while maintaining its Ag-binding specificity (Liu , et al 1996). Effector B cells lastly quit the GC and migrate either to the marginal zone of the secondary lymphoid organs, peripheral blood or the bone marrow (Duchosal 1997, Klein , et al 1998, Liu , et al 1988). In the peripheral blood, the secondary immune repertoire will hence be comprised of somatically mutated memory B cells with switched (IgG and IgA principally, ~15%), IgD+IgM+ (~15%), IgM only (~10%), and a minority of IgD only (<1%) isotype (Klein , et al 1998). Contrarily to the antigen-dependent phase, enzymatic effectors and mechanisms involved in the affinity maturation of the Ig receptor were very only poorly defined. The recent discovery of the activation induced cytidine deaminase AID, a key molecule in the class-switch (CS) and SH processes, has however lifted a corner the veil and allowed for a better comprehension of the GC reaction (Muramatsu , et al 1999, Okazaki , et al 2003).

Various systems have been elaborated over the years to categorize developmentally distinct B lymphocytes. One, mentioned previously, segregates lymphocytes according to surface expression of CD5, with CD5+ or B-1 B cells mediating the natural or TI response and CD5- or B-2 cells mediating the conventional T-dependent (TD) response. However the pertinence of using CD5 as an indicator for cell lineage, although common in the mouse, has been debated in humans (Sagaert and De Wolf-Peeters 2003, Youinou , et al 1999). Another approach, the Bm1-Bm5 classification, rather focuses on the differentiation status of B lymphocytes (naïve, centroblast, centrocyte, memory) as indicated by morphology and differential expression of phenotypic markers (sIg, CD38, CD10, CD44 and CD77) and their corresponding location within discrete micro-anatomical structures of the GC (follicular mantle, dark zone, light zone, marginal zone) (Pascual , et al 1994). This system has proven very useful for the characterization of B cells subsets isolated from human secondary lymphoid clinical specimens (tonsils, lymph nodes, spleen sections). Nevertheless, it has had serious limitations as far as categorization of the heterogeneous peripheral blood B cell population has been concerned. Contrarily to follicular B cells, a significant proportion of sIgD-expressing circulating B lymphocytes carries hypermutated Ig genes (Klein , et al 1998), making IgD expression an improbable indicator for naïve B cells. As a consequence, discrimination of memory and naïve subsets has not been possible based on morphological features and phenotypic markers alone. These difficulties have been resolved with the identification of CD27 as a marker for somatically mutated memory B lymphocytes (Agematsu 2000, Klein , et al 1998). Consequently, Ig configuration, along with differential expression of phenotypic markers, now provides any isolated B cell clone with a unique fingerprint that makes it possible to establish its developmental status. Using above criteria, clonal lymphoid disorders can thus be separated in three broad categories (Hummel and Stein 2000):

-Pre-GC origin B cells, with rearranged but unmutated Ig genes (ie: some CLL, mantle cell lymphoma or MCL).

-GC origin B cells, with rearranged, mutated Ig genes and intra-clonal diversity indicative of ongoing SH (ie: follicular lymphoma or FL, some diffuse large B cell lymphomas or GC-like DLBCL).

-Post-GC origin B cells, with rearranged, mutated Ig genes and no intra-clonal diversity (ie: some CLL, marginal zone lymphoma or MZL, HCL, Burkitt lymphoma or BL).

An ultimate level of refinement has been brought to this classification system with the advent of the DNA micrroarrays technology which, by allowing the simultaneous analysis of thousand of genes, yields an extremely precise gene-expression signature for each distinct stage in B cell development, particularly those involved in the GC reaction (Klein , et al 2003).

In recent years, evolution of molecular biology techniques has made analysis of chromosomal anomalies and Ig genes configuration accessible procedures and, in turn, has had a considerable impact on the clinical management of lymphoid disorders. The biologic properties of a malignant cell are partly dependent on the differentiation status of its normal cellular progenitor. Thus, molecular tools, which make it possible to retrace the non-transformed counterpart for a given clonal lymphoid proliferation, allow inferences into the natural history of the disease, but also predictions about its clinical behaviour and prognosis. Additionally, classification of lymphoma based on the molecular structure of Ig genes has made it evident that most lymphoid disorders had a GC or post-GC origin, suggesting an active role for the GC reaction in the generation of malignant cells.

This notion has recently been backed-up by a consistent body of experimental evidences which go far beyond the scope of this review, but has been the object of excellent analysis in recent publications (Davila , et al 2001, Shaffer , et al 2002). Briefly, though those molecular mechanisms involved in the maturation of the Ig receptor are by essence highly mutagenic, strict regulatory mechanisms had been identified that were thought to circumscribe danger for the organism. Among them, the specific targeting to the Ig locus directed by the presence of specific genes sequences such as recombination signal sequences (RSS) and switch region (SR), or the restricted expression of enzymatic effectors such as the RAG proteins, to a narrowly defined stage of B cell development (B cell precursors). But recently, analysis of GC derived cells in healthy individuals have made it clear that the SH machinery was acting outside the Ig gene locus and targeting potential proto-oncogenes such as the BCL-6 or CD95 proteins (Muschen , et al 2000, Pasqualucci , et al 1998). Similarly, RAG proteins have been found to be re-expressed in GC (Girschick , et al 2001) and are believed to play an active role in the generation of presumably transforming chromosomal translocations such as the bcl2/Ig or c-myc/Ig translocations (Davila , et al 2001). Thus there are many reasons to believe that malignant cells are in fact generated as by-products of the GC reaction.

Supplementary immunophenotyping of peripheral B lymphocytes in PPBL, using the newly defined memory marker CD27, and molecular analysis of Ig variable regions were undertaken in order to gain some insight in the developmental status of circulating B lymphocytes in patients. The memory subset, which represents 40% of peripheral B lymphocytes in healthy individuals, has thus been found to be increased up to 80% in PPBL patients. Noticeably, the memory compartment almost exclusively consists in IgM++IgD+ coexpressing B lymphocytes (Himmelmann , et al 2001a, Loembe , et al 2002, Salcedo , et al 2002), a population that only accounts for about one third of CD27+ memory B cells in healthy individuals, alongside class-switched, IgM-only, and a few IgD-only B lymphocytes (Klein , et al 1998). This raise in CD27 expression is corroborated by a parallel elevation in CD148 expression, a second memory B cell marker, originally identified on somatically mutated splenic marginal zone (MZ) B cells (Himmelmann , et al 2001a, Tangye , et al 1998). Taking into account the observed lymphocytosis, this amounts to a twenty fold increase of the specific IgM++IgD+CD27+ population in patients. Those observations have simultaneously led us, and another team, to proceed with the in depth molecular analysis of immunoglobulin genes in patients, since this strategy had proven so useful for investigation and classification of neoplasic lymphoid disorders (Loembe , et al 2002, Salcedo , et al 2002). Both studies have generated concordant results: in an average proportion of 73%, VH genes have been found to be somatically mutated. No bias in the Ig repertoire has been detected: VH genes usage among PPBL patients was heterogeneous and reflected the relative abundance of each VH gene family. Among both studies, the mutation frequency varied between 0.52% and 3%, with a mean value of 1.85%. Statistical analysis of mutations distribution in VH genes was also undertaken. Evidence for antigen-driven negative selection, specifically suppression of replacement mutations in FR, has not been found, except in a few mutated sequences (Loembe et al : 4/21, Salcedo et al : 12/51). It therefore seems PPBL results from the preferential expansion of the memory IgM++IgD+CD27+ B lymphocyte population. Intriguingly, the majority of these cells do not display the imprint of antigen selection that is considered to be the hallmark of the affinity maturation process in germinal centres.

Presence of somatic hypermutations in VH genes, CD27 expression, and absence of GC associated markers CD38 and CD10, all suggests a post-germinal centre origin for the expanding IgM++IgD+ B cell population in PPBL. In human, hypermutated memory IgM+IgD+ populations have been identified in the bone marrow (Paramithiotis and Cooper 1997), peripheral blood (Agematsu , et al 1997, Klein , et al 1998), and among the MZ B cells (Dono , et al 2000, Tangye , et al 1998) (table 14). The latter two subsets especially have multiple phenotypical features in common with PPBL including: CD27 expression, high levels of sIgM, large abundant cytoplasm and, specifically for MZ B lymphocytes, irregular nuclear morphology with frequent expression of CD21 and CD25. Mutation frequencies observed in VH genes in PPBL (~1.85%) are closest to that seen in marginal zone and bone marrow IgM+IgD+ memory B subsets (see table 14), and are definitely lower than that reported in class-switched memory B cells (Pascual , et al 1994).

The marginal zone is a micro-anatomically defined structure of the spleen, defined as the outermost margin of the follicular mantle zone. It is mainly populated by B cells displaying a phenotype intermediate between small resting lymphocytes and activated plasmablasts: the MZ B cells (Dono , et al 1996, Hsu 1985). This lymphoid compartment’s function as a reservoir for memory B cells involved in the response to TD and T-independent type 2 (TI-2) antigens was first extensively described in the rat (MacLennan and Liu 1991). In humans, the term marginal zone is now commonly used in a larger sense not only to refer to the extrafollicular areas of the spleen but also to similar micro-anatomical regions in other lymphoid tissues, namely: the subepithelial zone of the tonsils, the dome of the Peyer’s patches, the subcapsular areas of the lymph nodes, and extranodal reactive mucosa-associated lymphoid tissues (MALT) (Dono , et al 2003). Typically, MZ B cells are positioned at the front line of potential antigenic assault. B cells in this subset share common morphological attributes (see table 14), are usually non cycling and express high level of IgM with no or low IgD (Dono , et al 1996, Hsu 1985). Molecular studies, on the other hand, have revealed quite a level of heterogeneity among MZ B cells within a single anatomical site, with some expressing unmutated and others mutated VH genes. Those cells with mutations may or not present evidence of antigen-driven selection (Dono , et al 2000, Dunn-Walters , et al 1995, Tierens , et al 1999). It is conventionally believed that the non cycling hypermutated MZ B cells are the progeny of adjacent GC (Liu , et al 1988). Functional diversity in MZ B lymphocytes is further attested by the differential expression of cell surface markers according to their anatomic location: CD21 expression is high in splenic MZ B cells but low in their tonsillar counterparts (Dono , et al 2003), whereas the newly discovered IRTA-1 receptor is selectively expressed by tonsillar, MALT, and Peyer’s patches MZ B lymphocytes (Falini , et al 2003). Moreover, IgA and IgG expression is mostly restricted to splenic MZ B lymphocytes (Dono , et al 1996, Tangye , et al 1998). Differences in the type of antigenic challenge (TD versus TI-2) and in the level of antigen exposure could largely account for this heterogeneity in MZ B lymphocytes. The particular IgM++IgD+ MZ B cells subset has been investigated in the tonsils (Dono , et al 1996, Dono , et al 2003) and in the spleen (Tangye , et al 1998). This population is chiefly comprised of memory B cells as attested by CD27 and CD148 expression, presence of somatic mutations in VH genes and, in the tonsil subset, evidence for antigen-driven selection (table 14). Functionally, in analogy to animal models, these cells take part in both TD and TI-2 immune responses (Dono , et al 2003). Furthermore, following CD40 engagement, splenic IgM++IgD+ MZ B cells secrete both IgM and switched isotypes. In the rat, MZ B cells do not recirculate, except for their recruitment to the follicules upon recall immune reponses (Liu , et al 1988). Contrastingly in humans, this concept is challenged by the presence of B cells with characteristics similar to MZ B lymphocytes in the blood and to a lesser degree in the bone marrow (table 14 and Weller , et al 2004). Some authors have consequently proposed that, following antigenic stimulation and somatic diversification, either in a TD or TI fashion, MZ B cells could transit trough peripheral blood before homing to the marginal zone of neighbouring lymphoid organs. They could also migrate to the bone marrow where local environmental signals would promote terminal differentiation to high affinity plasma cells (Paramithiotis and Cooper 1997, Tierens , et al 1999).

Ensuing the extensive cytofluorometric assessment of cell surface molecules expression in PPBL B lymphocytes, complemented by the molecular analysis of Ig genes, Salcedo and collaborators have designated memory cells from the marginal zone compartment as normal counterparts for circulating IgM++IgD+CD27+ B cell in patients (Salcedo , et al 2002). Although this hypothesis seems highly plausible, supplementary immunophenotyping and functional studies are needed to warrant definitive validation of the developmental lineage in PPBL B lymphocytes.

Recently, somatically mutated IgM+IgD+CD27+ memory B cells were identified in the peripheral blood of hyper IgM (HIGM1) patients who cannot form GC because of an invalidating mutation in CD154, the CD40 ligand. Conversely, and in accordance with their GC origin, class-switched and IgM only CD27+ memory B cells were not detected (Weller , et al 2001). This observation has raised a serious debate about the notion of a GC-dependent origin in IgM+IgD+CD27+ memory B lymphocytes. Weller and collaborators, the team at the origin of this significant discovery, suggested that peripheral blood IgM+IgD+CD27+ memory B cells are the circulating counterparts of MZ IgM++IgD+CD27+ B cells. Furthermore, these cells would be involved in TI responses, particularly TI-2 responses (Kruetzmann , et al 2003, Weller , et al 2004), implying that somatic diversification of the Ig receptor in HIGM1 patients would arise from a GC-independent developmental pathway (Weller et al 2003). In line with this view, cycling (KI-67+) MZ B cells with a memory phenotype were identified in nodal tissues sections and they displayed no clonal relationship to proximate GC, suggesting that these lymphocytes had possibly mutated in situ independently of the GC environment (Tierens , et al 1999). In similitude to mice B1 B cells, antigen exposure would be a prerequisite to specific antibody production by IgM++IgD+CD27+ B cells (Fagarasan and Honjo 2000, Kruetzmann , et al 2003). In this sense, this subset would not represent the "true", GC-derived, memory pool, but rather mediate a natural yet specific immunity, at the junction between the innate and the T-dependent, adaptive immune response, insuring the first line of defense against encapsulated bacteria (Kruetzmann , et al 2003, Weller , et al 2004). The fact that deregulation of homeostasis observed in PPBL patients solely affects the IgM++IgD+CD27+ lymphoid compartment makes a distinct origin for this particular memory subset a conceivable hypothesis. This disorder could thus add to the emerging picture of a GC-independent developmental pathway for IgM+IgD+CD27+ B cells.

On a practical level, addressing the phenotype and lineage of the expanding B cell population in PPBL patients has had appreciable repercussions for clinicians as it provided them with additional and valuable diagnostic tools. But the investigative objectives pursued by researchers were primarily to uncover the cause(s) behind the disorder and to predict its evolution, if any was to be expected. And, in this regard, no clear-cut answer has been offered as yet. Two scenarios have nevertheless been advanced to explain polyclonal amplification of the IgM++IgD+CD27+ B cell lymphoid compartment in patients: either 1) an increase in memory B cells production (as a result of chronic antigenic stimulation) or 2) a decrease in their developmentally programmed elimination (due to impairment of apoptosis).

The first supposition is based on several lines of evidence demonstrating a connection between infectious agents and the emergence of various B cell lymphoid disorders (Dolcetti and Boiocchi 1996). The cause-effect relationship between B cell-tropic viruses with transforming capacities, such as EBV, and lymphoma in immunosupressed individuals is a well documented example (Crawford 2001). More recently, evidences have also accumulated for the indirect implication of pathogens in B cell lymphoproliferations, specifically those originating from the marginal zone, as a result of sustained antigenic stimulation. Such is the case for Helicobacter pylori infection and gastric mucosa associated lymphoid tissue (MALT) lymphoma where pathogen eradication by antibiotherapy frequently leads to disease regression (Wotherspoon , et al 1993). Epidemiological studies suggest a similar cause-effect relationship between splenic marginal zone lymphoma with villous lymphocytes (SLVL) and the hepatitis C virus (Hermine , et al 2002). As far as PPBL is concerned, a role for EBV in the natural history of the disease has long been postulated as formerly discussed. Although direct B cell infection and transformation by a virus variant could not be formally evidenced, the possibility remains that an as yet unidentified (environmental ?) factor could drive chronic reactivation of lytic infection in patients, generating a persistent memory IgM++IgD+ B cell immune response, in either a GC-dependent or GC-independent pathway. Coincidentally, one of the anticipated anatomical location for IgM++IgD+CD27+ memory cells, namely the subepithelial zone of the tonsils, was also proposed as the primary site for EBV infection in healthy individuals (Faulkner , et al 2000). Repeated virus production in this site could evoke a sustained immune response in resident IgM++IgD+CD27+ memory cells, in similitude to what is observed in MALT lymphoma. On the other hand, no restriction of the immunoglobulin VH gene repertoire was evidenced in PPBL patients. Contrastingly, this phenomenon has been reported in lymphoma derived from antigen-experienced memory B cells (Fais , et al 1998, Weng and Levy 2003), or even after Haemophilus influenzae immunization in healthy individuals (Adderson , et al 1991). In addition, the lack of evidence for antigenic selection pressure in expressed VH gene regions further challenges the model of chronic antigenic selection in PPBL.

As we already pointed out, maintenance of homeostasis in the B cell lymphoid compartment is dependent upon stringent regulation mechanisms which insure control of the molecular processes involved in somatic diversification of the antibody repertoire. In the GC, balanced survival and elimination of mutated clones, through the process of antigen-driven selection, is necessary to delete potentially harmful autoreactive clones. It also prevents the feeding of superfluous low-affinity mutants in the memory pool. Thereby, only the best-fitted antigen-specific mutants are allowed to differentiate to memory B cells. The fact that IgM++IgD+CD27+ B cells in PPBL patients showed no evidence for antigen-driven selection, as indicated by the distribution of replacement versus silent mutations in Ig VH genes, was intriguing given that both their immunophenotype and the molecular configuration of their Ig genes were otherwise indicative of a post-GC origin. Lack of evidence for antigen-driven selection can signify that the process of affinity maturation, the hallmark of a TD immune response, is altered in PPBL patients. In light of those observations, we have put forward a second scenario whereby impairment of the antigen-driven selection mechanism would enable the survival of low affinity mutants B cells within GC in PPBL, allow their recruitment into the memory B cell pool, and subsequently cause expansion of this compartment in the periphery. This hypothesis appears to be corroborated by several experimental facts. First, expression the Bcl-2 and the Bcl-xL anti-apoptotic proteins have been reported to be up-regulated in PPBL. The interplay between differential expression of key pro and anti-apoptotic genes in GC B cells is crucial to the antigen-driven selection process. Tonsillar GC B cells accordingly express an apoptosis-sensitive phenotype, namely high expression of the pro-apoptotic CD95, Bax, Bak, Bim, and c-Myc proteins combined with low expression of the survival protein Bcl-2 (Liu and Arpin 1997, Yokoyama , et al 2002). This expression pattern apparently predisposes them to deletion through CD95/Fas-mediated killing, unless they can be rescued by efficient binding of immunizing Ag on the surface of follicular dendritic cells (FDC) and interaction with antigen specific CD40-L expressing T cells in the light zone of GC (Liu , et al 1989). CD95-induced cellular apoptosis can utilize two different signalling pathways (Mizuno , et al 2003). Type I apoptosis proceeds through association of a death inducing complex (DISC) where CD95 death domain (DD) recruits the CD95/Fas containing associated death-domain containing adapter protein (FADD) and procapase 8, later leading to the activation of caspase 8 and the ensuing apoptotic signalling cascade. In type II or mitochondrion-dependent apoptosis, an amplification step is necessary which involves mitochondrial release of cytochrome C, apoptosome assembly and activation of capase 9. This latter type can specifically be blocked by members of the Bcl-2 protein family. Which type of Fas-induced apoptosis is predominant in GC B cells is still subjected to debate. Some experimental data indicates that both types I and II apoptosis could contribute to the process of antigen-driven selection. In point of fact, animals with decreased CD95 expression ( lpr/lpr phenotype) (Takahashi , et al 2001), as well as those possessing a transgene for either Bcl-2 (Smith , et al 2000) or Bcl-xL (Takahashi , et al 1999) constitutive expression, all display a distortion of affinity maturation which leads to the accumulation of low affinity memory B cells. Contrarily to Bcl-2, Bcl-xL expression is high in GC B cells (Tuscano , et al 1996). Moreover, rescue from CD95-mediated apoptosis, mimicked in vitro by CD40 stimulation and Ig cross-linking, correlates with up-regulation of Bcl-xL (Zhang , et al 1996), suggesting an active regulatory role for this anti-apoptotic protein. In the same way, increased Bcl-xL expression in PPBL patients could confer resistance to CD95-mediated apoptosis. However, one cautionary note regarding this conclusion resides in the fact that Bcl-2 and the Bcl-xL expression levels in patients were compared to protein expression in total B cells, which comprised a majority (~60%) of naïve B cells. Whereas Bcl2 protein expression is similar in naïve and memory B cells, Bcl-xL mRNA is about 8-fold higher in the memory compartment (Bovia , et al 1998). The apparent up-regulation of this protein in patients could thus be a mere reflect of an increased proportion in memory B cells. Nevertheless, in vitro resistance to Fas-mediated apoptosis has concomitantly been reported in PPBL B lymphocytes (Roussel , et al 2003). Again, normal controls consisted not in memory B cells, but rather in the Ramos cell line, with a GC phenotype and a conceivably differential expression of apoptosis-related genes. Nevertheless, since PPBL B lymphocytes displayed apoptotic features when treated with the anti-neoplastic etoposide reagent, the observed resistance appeared to be specific to the CD95-induced apoptotic pathway. CD95 has been described as a tumour suppressor gene, as loss of susceptibility to CD95-mediated killing often correlates with tumour progression. Notably, impairment of CD95 function would allow the survival of GC generated autoreactive or pre-malignant B clones that would otherwise be eliminated by antigen-driven selection. Support for this last role is provided by studies in animal models with a lpr/lpr phenotype who display increased susceptibility for B cell malignancies (Davidson , et al 1998). The same holds true for CD95 deficient human patients with the autoimmune lymphoproliferative syndrome (APLS) who lack a functional CD95 protein (Straus , et al 2001). Down-regulation of CD95 protein expression, disruption of the CD95 signalling cascade, or somatic mutations of the CD95 encoding gene, particularly those affecting the death domain (DD) region, are all mechanisms that have been identified in lymphoid malignancies and which could promote lymphomagenesis (for a detailed review of CD95 resistance mechanisms and their contribution to lymphomagenesis please see Mizuno , et al 2003 and Muschen , et al 2002) With regards to PPBL, CD95 expression in IgM++gD+CD27+ memory B lymphocytes is similar or superior to that detected in healthy controls (Salcedo , et al 2002 and our own observations), therefore the observed resistance cannot be related to negative modulation of the receptor in patients. Still, mutations in CD95 DD could be present that would allow cell surface expression of the protein but interfere with signalling. Presence of the DISC components FADD and caspase 8 has been detected (Roussel , et al 2003), but it is not known at this point if they assemble properly. Moreover, expression of FLIP or FAIM, known mediators of CD95 resistance (Schneider , et al 1999, Wang , et al 2000), has not yet been investigated. The postulate of a deficient antigen-driven selection mechanism in PPBL has recently been complemented by the uncovering of IgM anti-phospolipid antibodies (apA/cofactor) in patients despite the absence of noticeable auto-immune disease (Granel , et al 2002). Hence, propensity for increased survival of auto-reactive and low affinity B cells clones seems to exist in PPBL patients. We ergo propose that a deregulation of the physiological processes implicated in the naïve to memory B cell transition could contribute significantly to the disruption of homeostasis in PPBL patients.

Although the majority of patients have had uneventful follow-ups amounting to more than 25 years in some cases, there is a persistent concern that the disruption of homeostasis observed in PPBL could in fact represent the first stage in a multi-step progression toward a more aggressive proliferation. Accordingly, in two patients, PPBL diagnosis has been associated with the occurrence of non Hodgkin lymphomas: DLBCL (19 years after diagnosis) (Roy , et al 1998), and MALT lymphoma (concomitantly with diagnosis) (Callet-Bauchu , et al 1999). Clonal or oligoclonal VH genes rearrangements have been detected in four patients, yet no further evolution was reported for these cases (Chan , et al 1990, Delage , et al 1997, Feugier , et al 2004). These observations nevertheless emphasize the notion that emergence of a predominant, and potentially malignant, clone is a likely outcome in PPBL. Bone marrow B cell intra-vascular infiltration was recently reported as recurrent finding in PPBL patients (Feugier , et al 2004). Moreover, careful microscopic observation has also revealed the presence of nuclear pockets in atypical B lymphocytes, a feature which is usually witnessed in pre-leukemic or leukemic leucocytes (Casassus , et al 1987, Espinet , et al 2000, Woessner , et al 1999). The discovery of clonal genetic abnormalities in PPBL B cells, contrasting with the polyclonal nature of the proliferation, illustrates the risk of subsequent transformation in this disorder. As mentioned previously, isochromosome (+i3q), first documented as an isolated event in 1989 (Perreault , et al 1989), has since been demonstrated to be a recurrent finding among PPBL patients (Callet-Bauchu , et al 1997, Mossafa , et al 1996): it was observed in 29/41 cases tested (see table 13, column 8). Additional numerical aberrations involving chromosome 3, namely trisomy 3, partial duplication 3q, and derivative 3, were also sporadically reported (table 13, column 8). Unlike PCC, whose occurrence has been linked to the presence of multinucleated cells (Mossafa , et al 1999), the (+i3q) aberration is apparently not restricted to a morphologically distinct B cell population and both events are independent of the light chain isotype (Callet-Bauchu , et al 1997, Espinet , et al 2000). Interestingly, observations made at referral and then at 2 years follow-up seem to indicate an accumulation of these chromosomal anomalies in at least one patient (Callet-Bauchu , et al 1999). Among non Hodgkin lymphoid disorders, chromosome 3 aberrations are more frequently identified in MZL, more precisely in the extranodal (MALT), nodal (monocytoid B cell lymphoma or MCBL) and splenic marginal zone B cell lymphoma (or SMZL, including splenic lymphoma with villous lymphocytes [or SLVL]) types (Dierlamm , et al 1996). Consistent with the postulated origin for IgM++IgD+CD27+ B lymphocytes in PPBL, SMZL shares many additional features with this disorder:

-Generally slow clinical progression.

-A heterogeneous morphological composition involving a mixture of plasma, blast-like and small cells, occasionally presenting a cleaved nucleus.

-Peripheral blood and bone marrow involvement (Thieblemont , et al 2003).

-Lack of expression of CD5, CD10 and CD23.

-Heterogeneous utilisation of VH genes among cases, and presence of somatic hypermutations without evidence of antigenic selection in most cases (Dierlamm , et al 1996).

Unlike PPBL however, SMZL are clearly monoclonal with detectable IgH rearrangements. In addition, no bcl2 gene rearrangement can be observed, which is in sharp contrast to what is observed in PPBL.

In eleven patients of the cohort followed by our team in Quebec City, nested-PCR amplification has allowed the detection of multiple distinct bcl-2/Ig genes rearrangements (up to seven), involving both the mcr (minor cluster region) and MBR (major breakpoint region), in all but one case (only one rearrangement) (Delage , et al 1997, Delage , et al 1998). These observations were later reproduced in additional patients, and the frequency of the t(14;18) translocation was estimated between 1/102 and 1/104 cells (see table 13, column 8). In analogy to the +i(3q) chromosomal anomaly, there appears to be an accumulation of the bcl-2/Ig rearrangements in some patients (Delage , et al 1998). bcl-2/Ig genes rearrangements are also reported in approximately 50% of healthy individuals, however their frequency is lower (1/105 to 1/106) and multiple rearrangements are only revealed with very sensitive detection methods (Ji , et al 1995). Bcl-2 oncogenic potential is illustrated by the occurrence of the t(14;18) translocation in 80% of cases of follicular lymphoma (FL), the most frequent human B cell malignancy. A positive correlation was observed between age or tobacco usage and the presence of bcl-2/Ig rearrangements in healthy individuals (Bell , et al 1995, Liu , et al 1994). As the incidence of non Hodgkin lymphoma is also increased in aged people and smokers, authors have accordingly proposed that the presence of bcl-2/Ig rearrangements reflects an individual’s risk for developing a subsequent lymphoid malignancy.

At the cellular level, the t(14;18) chromosomal translocation produces a bcl-2 -immunoglobulin fusion gene and, owing to the resulting proximity with the Ig transcriptional enhancer, leads to an overexpression of the Bcl-2 protein (Graninger , et al 1987). Furthermore, this deregulated expression interferes with apoptosis, though it does not promote cellular proliferation (Hockenbery , et al 1990). Interestingly, mice models bearing a bcl-2/Ig minigene initially display an indolent lymphoid hyperplasia consisting of polyclonal IgM+IgD+ resting B cells. Slow progression to clonal DLCL, apparently as the result of secondary genetic alterations ( c-myc translocation), ensues in these mice (McDonnell and Korsmeyer 1991). Thus bcl-2 acts as a proto-oncogene who, once translocated, promotes B cell survival and increases the risk for subsequent tumorigenic genome alterations leading to the emergence of neoplasia. In humans, the similitude with PPBL is remarkable, and that a similar outcome could occur in patients is indeed a troubling probability. Still, at the physiological level, presence of bcl2 gene translocations doesn’t always translate into increased protein expression, especially at the relatively low translocation frequency observed in PPBL patients. Accordingly, reports of Bcl2 protein upregulation in PPBL patients have been negative in most cases (Delage , et al 1998, Himmelmann , et al 2001a, Lancry , et al 2001). Nevertheless, cytoplasmic expression of the protein has been detected by immunocytochemistry in both binucleated and non binucleated B cells (Feugier , et al 2004, Lancry , et al 2001). Again, utilisation of the corresponding normal memory cellular subset as basis for comparison could provide a definitive answer regarding Bcl-2 expression levels in patients.

Similarly to animal models, progression from low (ie: FL) to high (ie: DLBCL) grade disease apparently requires additional genetic mutations secondary to the FL-characteristic blc-2/Ig translocation. Oncogenic lesions affecting genes with growth promoting ( c-myc ), differentiation blocking ( bcl-6, pax5 ) or apoptosis blocking ( bcl-xL , NF-κB activators) properties can presumably synergize with Bcl-2 to promote progression from low to high grade lymphoma (Shaffer , et al 2002). Mutations in the open reading frame of the blc-2 gene itself have been correlated with morphologic transformation form FL to DLBCL (Matolcsy , et al 1996). Since they are restricted to the GC and post GC compartments, these genetic alterations apparently accumulate as by-products of affinity maturation molecular machinery (Shaffer , et al 2002). Which bring us back to PPBL: although they preferentially differentiate to plasma cell upon recall responses, memory B cells can also participate to further rounds of GC-dependent affinity maturation (Liu 1997). Higher frequency in memory B cells could thus increase the risk for genomic instability in patients. A presumably deficient antigen-driven selection in PPBL patients could substantially contribute to the emergence of clonal proliferations. To this day, no molecular analysis of those genes related with low to high grade disease progression was conducted in PPBL patients, with the exception of studies regarding bcl-2 translocations. bcl-2 mutational status per se has not been ascertained however. A special focus on lymphoma-associated oncogenes, aiming at the estimation of both gene expression and gene structure, is an avenue that was only scarcely explored in PPBL and which need to be further investigated in the near future as it could prove extremely useful to predict the outcome of this disorder.

Throughout this review, we have purposely left out the atypical PPBL case diagnosed in a newborn child by Gomez et al in 2000. That case undeniably stands apart among classical PPBL reports as far as the age of the patient is concerned: this is the only diagnosis that has been described in a child during the last twenty years. More importantly, atypical B lymphocytes display an unusual morphology with no binucleated cells, and the occasional presence of cytoplasmic villi or a lymphoplasmacytoid appearance. None of the prospective PPBL predisposition factors (positive EBV serology, HLA-DR7 haplotype), or characteristic genetic instability ( bcl-2/Ig genes rearrangements, +i(3q)) have been identified. IgM levels are increased, but only slightly. Finally, B cells display a distinctive surface immunophenotype with expression of CD23, CD25, CD38, CD103, CD5 and no expression of CD11c (Gomez , et al 2000 and personal communication). These clinical features are rather reminiscent of those seen in hairy B cell lymphoproliferative disorder (HBLD), the alleged polyclonal counterpart of the Japanese variant of HCL, which however is negative for CD25 and positive for CD11c (Machii , et al 1997). To our knowledge, CD5+ persistent polyclonal B cell lymphocytosis has only been reported once in a male adult (Reeder and Conley 1999). In neither of those two cases was the cell surface expression of IgM, IgD and CD27 assessed. Configuration of Ig VH genes in these patients has not been determined either. Until these elements are evaluated, it will not be possible to definitely resolve whether this atypical CD5+ PPBL is a variant form of classic PPBL or whether it is a separate entity as its distinct immunophenotype seems to indicate.

In order to better understand the aetiology of PPBL and estimate the risks of malignant progression in patients, it will be mandatory to refine the characterisation of the expanding IgM+IgD+CD27+ subset. Supplementary immunophenotyping, particularly exploring those markers such as IRTA-1, or the newly described CD1c, which are specifically expressed among MZ B cell subsets (Falini , et al 2003, Weller , et al 2004), could be useful to confirm whether PPBL B lymphocytes actually are a MZ derived population. If access to secondary lymphoid organ specimens should ever become possible, it would be very interesting to conduct in situ studies in order to retrace the exact anatomical origin for IgM+IgD+CD27+ B lymphocytes in patients. Molecular analysis of individually picked B cells, isolated from the distinct areas of secondary follicules (dark zone, light zone, mantle and marginal zone), as was elegantly presented by Küppers et al (Kuppers , et al 1993), may allow researchers to establish the genealogy between GC founders and hypermutated memory B cells. This could well represent an authoritative answer in the debate regarding the GC origin of IgM+IgD+CD27+ B lymphocytes not only in patients but also in healthy individuals. If indeed they originate from a distinct lineage, molecular profiling and the establishment of the gene signature of IgM+IgD+CD27+ B lymphocytes both in patients and healthy individuals, might provide an effective way to uncover the possible physiological deregulation present in PPBL.

Very few functional studies have been conducted on PPBL B lymphocytes (see table 15). The polyclonal nature of the population under scrutiny, and the lack of an accurately identified normal counterpart, long impeded such experiments. The results, when obtained, were difficult to interpret. Nonetheless, early functional studies gave clear indication of the functional distinctiveness of PPBL B cells (Loembe , et al 2001, Reimer , et al 2000). This already suggested that the majority of peripheral blood B lymphocytes in patients presented a different developmental status relative to healthy individuals. The subsequent characterisation of the expanding IgM+IgD+CD27+ population, corroborated those preliminary observations and, at least in theory, greatly reduced the obstacles to further functional studies. However, one was still confronted with the scarcity of comprehensible information regarding peripheral blood memory B lymphocytes, notably unswitched, hypermutated IgM+IgD+ cells, long considered a naïve subset. As this field is now rapidly being uncovered, it should be expected that additional investigations will be undertaken in the coming years. It will then be possible to gain a better understanding of the physiological function of the IgM+IgD+CD27+ B cell population in patients, notably the type of immunity it mediates (TD versus TI), and its capacity to participate to recall immune responses. In this regard, it will be interesting to determine the role played by those cells as far as antibodies production is concerned. Assessment of the differentiation capacity in PPBL IgM+IgD+CD27+ B lymphocytes might ascertain the hypothesis, advanced by some authors, that impaired maturation would be accountable for the selective expansion of this subset, yielding the high serum IgM levels observed in patients.

Definitive confirmation of the preliminary results hinting at a defect in the antigen-driven selection process will require in vitro studies. The pattern of pro-and anti-apoptotic genes expression in PPBL patients (especially regarding Bcl-xL) as well as the dynamics of CD95 DISC assembly and its ensuing signalling cascade will need to be evaluated. It will be however mandatory that the normal counterpart, ie: IgM+IgD+CD27+ memory B cell population from healthy donors, be used for comparison. Moreover, to more closely recreate the GC microenvironment, stimulation systems should include FDC, a cellular type which has been shown to play a preponderant part in the antigen-driven selection process, along with CD40-L and CD95-L expressing T lymphocytes (Li and Choi 2002).

The frequent involvement of chromosome 3 numerical anomalies could equally prove of clinical significance. Thus, in depth genetic investigation are required to 1) locate regions bearing genes of prospective interest on this chromosome 2) determine if the function of those genes could be significant to the pathogenesis of PPBL.

Finally, a freshly published paper has reported defective expression of adhesion molecules in PPBL (Feugier , et al 2004), highlighting a possible role for improper homing in this disorder that could cause accumulation of IgM+IgD+CD27+ memory B cell in the peripheral blood. Surely, studies in additional patients that would combine analysis of cellular adhesion and chemotaxis could be yet another interesting avenue that would benefit from deeper investigation and might provide considerable insight into the genesis of this disorder.

PPBL is an unusual haematological disorder, sharing attributes of both indolent proliferation (polyclonality, stability, lack of significant symptoms) and malignant proliferation (atypical cellular morphology, chromosomal anomalies, bone marrow infiltration). As such, it has long elicited puzzlement among the medical and scientific communities. Paralleling the significant achievements made with regards to the understanding of B cell immunobiology, a clearer clinical definition of the disorder has notwithstanding been emerging in the recent years. The proliferating subset has been clearly delineated, specific genetic anomalies have been uncovered, and a familial predisposition has been evidenced. Nevertheless, general awareness about this disorder is still lagging behind, inasmuch as it is not yet listed as a distinct pathological entity in recent haematology manuals.

In our opinion, two principal reasons justify bringing this disorder to the forefront. Firstly, despite its apparent rarity, PBBL could be relatively common, especially among family members of established cases. The repeated detection of patients lacking any clinical evidence of the disorder, other than occasional presence of circulating binucleated B lymphocytes, emphasizes the fact that PPBL does not always manifest itself as an overt leucocytosis. More cases could thus go unrecognized in the general population. It is necessary that the clinical picture of PPBL, especially its indolent progression, be widely acknowledged so these prospective cases be not submitted to unnecessary aggressive therapies, as were some former patients (Perreault , et al 1989). Secondly, as the risk for malignant evolution in PPBL cannot be dismissed altogether at this point, vigilant long-term monitoring should be advised and special efforts should be invested so that patients do not become lost to follow-up.

On a more fundamental level, advances regarding the fundamentals of B cell developmental biology and technical inputs from the molecular field have made it possible to gain remarkable insight into the genesis of lymphoid disorders in general, and PPBL in particular. This clinical model clearly illustrates the multi-step theory of lymphomagenesis. Disruption of peripheral lymphoid homeostasis and genetic instability in patients are presumably preliminary steps towards malignant progression (see figure 21). Apparently however, they are insufficient to drive definitive clonal transformation, and/or specific safeguard mechanisms operate in PPBL. Surely future investigations will grant a better understanding of the lymphoid transformation process, or lack of thereof, in patients, and by extension in the general population.

Figure 19 : Morphology of atypical lymphocytes in PPBL.

Pictures illustrating the heterogeneous aspect of atypical B lymphocytes in PPBL which can present with an enlarged (A), either slightly (B) to deeply indented (C), or fully binucleated (D) nucleus.

Figure 20 : Peripheral B cell development as indicated by Ig genes configuration and cell surface immunophenotype.* see (Klein, et al 1998).

Figure 21 : PPBL, past and future: the aetiopathology of PPBL and its prospective clinical evolution.

Adderson, E.E., Shackelford, P.G., Quinn, A. & Carroll, W.L. (1991) Restricted Ig H chain V gene usage in the human antibody response to Haemophilus influenzae type b capsular polysaccharide. J Immunol, 147, 1667-1674.

Agematsu, K. (2000) Memory B cells and CD27. Histol Histopathol, 15, 573-576.

Agematsu, K., Nagumo, H., Yang, F.C., Nakazawa, T., Fukushima, K., Ito, S., Sugita, K., Mori, T., Kobata, T., Morimoto, C. & Komiyama, A. (1997) B cell subpopulations separated by CD27 and crucial collaboration of CD27+ B cells and helper T cells in immunoglobulin production. Eur J Immunol, 27, 2073-2079.

Agrawal, S., Matutes, E., Voke, J., Dyer, M.J., Khokhar, T. & Catovsky, D. (1994) Persistent polyclonal B-cell lymphocytosis. Leuk Res, 18, 791-795.

Almarri, A. & Batchelor, J.R. (1994) HLA and hepatitis B infection. Lancet, 344, 1194-1195.

Bain, B., Matutes, E. & Catovsky, D. (1998) Teaching cases from the Royal Marsden and St Mary's Hospitals. Case 14: persistent lymphocytosis in a middle aged smoker. Leuk Lymphoma, 28, 623-625.

Bell, D.A., Liu, Y. & Cortopassi, G.A. (1995) Occurrence of bcl-2 oncogene translocation with increased frequency in the peripheral blood of heavy smokers. J Natl Cancer Inst, 87, 223-224.

Bovia, F., Nabili-Tehrani, A.C., Werner-Favre, C., Barnet, M., Kindler, V. & Zubler, R.H. (1998) Quiescent memory B cells in human peripheral blood co-express bcl-2 and bcl-x(L) anti-apoptotic proteins at high levels. Eur J Immunol, 28, 4418-4423.

Callet-Bauchu, E., Gazzo, S., Poncet, C., Pages, J., Morel, D., Alliot, C., Coiffier, B., Coeur, P., Salles, G. & Felman, P. (1999) Distinct chromosome 3 abnormalities in persistent polyclonal B-cell lymphocytosis. Genes Chromosomes Cancer, 26, 221-228.

Callet-Bauchu, E., Renard, N., Gazzo, S., Poncet, C., Morel, D., Pages, J., Salles, G., Coeur, P. & Felman, P. (1997) Distribution of the cytogenetic abnormality +i(3)(q10) in persistent polyclonal B-cell lymphocytosis: a FICTION study in three cases. Br J Haematol, 99, 531-536.

Carr, R., Fishlock, K. & Matutes, E. (1997) Persistent polyclonal B-cell lymphocytosis in identical twins. Br J Haematol, 96, 272-274.

Carstairs, K.C., Francombe, W.H., Scott, J.G. & Gelfand, E.W. (1985) Persistent polyclonal lymphocytosis of B lymphocytes, induced by cigarette smoking? Lancet, 1, 1094.

Casassus, P., Lortholary, P., Komarover, H., Lejeune, F. & Hors, J. (1987) Cigarette smoking-related persistent polyclonal B lymphocytosis. A premalignant state. Arch Pathol Lab Med, 111, 1081.

Chan, M.A., Benedict, S.H., Carstairs, K.C., Francombe, W.H. & Gelfand, E.W. (1990) Expansion of B lymphocytes with an unusual immunoglobulin rearrangement associated with atypical lymphocytosis and cigarette smoking. Am J Respir Cell Mol Biol, 2, 549-552.

Chiorazzi, N. & Ferrarini, M. (2003) B cell chronic lymphocytic leukemia: lessons learned from studies of the B cell antigen receptor. Annu Rev Immunol, 21, 841-894.

Chow, K.C., Nacilla, J.Q., Witzig, T.E. & Li, C.Y. (1992) Is persistent polyclonal B lymphocytosis caused by Epstein-Barr virus? A study with polymerase chain reaction and in situ hybridization. Am J Hematol, 41, 270-275.

Crawford, D.H. (2001) Biology and disease associations of Epstein-Barr virus. Philos Trans R Soc Lond B Biol Sci, 356, 461-473.

Davidson, W.F., Giese, T. & Fredrickson, T.N. (1998) Spontaneous development of plasmacytoid tumors in mice with defective Fas-Fas ligand interactions. J Exp Med, 187, 1825-1838.

Davila, M., Foster, S., Kelsoe, G. & Yang, K. (2001) A role for secondary V(D)J recombination in oncogenic chromosomal translocations? Adv Cancer Res, 81, 61-92.

de Campos-Lima, P.O., Gavioli, R., Zhang, Q.J., Wallace, L.E., Dolcetti, R., Rowe, M., Rickinson, A.B. & Masucci, M.G. (1993) HLA-A11 epitope loss isolates of Epstein-Barr virus from a highly A11+ population. Science, 260, 98-100.

de Jaureguiberry, J.P., Pignon, D., Galzin, M., Carli, P., Jaubert, D. & Chagnon, A. (1997) [Persistent polyclonal B-cell lymphocytosis with binucleated lymphocytes in a male patient]. Rev Med Interne, 18, 258.

Delage, R., Jacques, L., Massinga-Loembe, M., Poulin, J., Bilodeau, D., Mignault, C., Leblond, P.F. & Darveau, A. (2001) Persistent polyclonal B-cell lymphocytosis: further evidence for a genetic disorder associated with B-cell abnormalities. Br J Haematol, 114, 666-670.

Delage, R., Roy, J., Jacques, L., Bernier, V., Delage, J.M. & Darveau, A. (1997) Multiple bcl-2/Ig gene rearrangements in persistent polyclonal B-cell lymphocytosis. Br J Haematol, 97, 589-595.

Delage, R., Roy, J., Jacques, L. & Darveau, A. (1998) All patients with persistent polyclonal B cell lymphocytosis present Bcl-2/Ig gene rearrangements. Leuk Lymphoma, 31, 567-574.

Delannoy, A., Djian, D., Wallef, G., Deneys, V., Fally, P., Martiat, P. & Michaux, J.L. (1993) Cigarette smoking and chronic polyclonal B-cell lymphocytosis. Nouv Rev Fr Hematol, 35, 141-144.

Dierlamm, J., Pittaluga, S., Wlodarska, I., Stul, M., Thomas, J., Boogaerts, M., Michaux, L., Driessen, A., Mecucci, C., Cassiman, J.J. & et al. (1996) Marginal zone B-cell lymphomas of different sites share similar cytogenetic and morphologic features. Blood, 87, 299-307.

Dolcetti, R. & Boiocchi, M. (1996) Cellular and molecular bases of B-cell clonal expansions. Clin Exp Rheumatol, 14 Suppl 14, S3-13.

Dono, M., Burgio, V.L., Tacchetti, C., Favre, A., Augliera, A., Zupo, S., Taborelli, G., Chiorazzi, N., Grossi, C.E. & Ferrarini, M. (1996) Subepithelial B cells in the human palatine tonsil. I. Morphologic, cytochemical and phenotypic characterization. Eur J Immunol, 26, 2035-2042.

Dono, M., Zupo, S., Colombo, M., Massara, R., Gaidano, G., Taborelli, G., Ceppa, P., Burgio, V.L., Chiorazzi, N. & Ferrarini, M. (2003) The human marginal zone B cell. Ann N Y Acad Sci, 987, 117-124.

Dono, M., Zupo, S., Leanza, N., Melioli, G., Fogli, M., Melagrana, A., Chiorazzi, N. & Ferrarini, M. (2000) Heterogeneity of tonsillar subepithelial B lymphocytes, the splenic marginal zone equivalents. J Immunol, 164, 5596-5604.

Duchosal, M.A. (1997) B-cell development and differentiation. Semin Hematol, 34, 2-12.

Dunn-Walters, D.K., Isaacson, P.G. & Spencer, J. (1995) Analysis of mutations in immunoglobulin heavy chain variable region genes of microdissected marginal zone (MGZ) B cells suggests that the MGZ of human spleen is a reservoir of memory B cells. J Exp Med, 182, 559-566.

Espinet, B., Florensa, L., Sole, F., Lloveras, E., Abella, E., Besses, C., Sans-Sabrafen, J. & Woessner, S. (2000) Isochromosome +i(3)(q10) in a new case of persistent polyclonal B-cell lymphocytosis (PPBL). Eur J Haematol, 64, 344-346.

Fagarasan, S. & Honjo, T. (2000) T-Independent immune response: new aspects of B cell biology. Science, 290, 89-92.

Fais, F., Ghiotto, F., Hashimoto, S., Sellars, B., Valetto, A., Allen, S.L., Schulman, P., Vinciguerra, V.P., Rai, K., Rassenti, L.Z., Kipps, T.J., Dighiero, G., Schroeder, H.W., Jr., Ferrarini, M. & Chiorazzi, N. (1998) Chronic lymphocytic leukemia B cells express restricted sets of mutated and unmutated antigen receptors. J Clin Invest, 102, 1515-1525.

Falini, B., Tiacci, E., Pucciarini, A., Bigerna, B., Kurth, J., Hatzivassiliou, G., Droetto, S., Galletti, B.V., Gambacorta, M., Orazi, A., Pasqualucci, L., Miller, I., Kuppers, R., Dalla-Favera, R. & Cattoretti, G. (2003) Expression of the IRTA1 receptor identifies intraepithelial and subepithelial marginal zone B cells of the mucosa-associated lymphoid tissue (MALT). Blood, 102, 3684-3692.

Faulkner, G.C., Krajewski, A.S. & Crawford, D.H. (2000) The ins and outs of EBV infection. Trends Microbiol, 8, 185-189.

Feugier, P., De March, A.K., Lesesve, J.F., Monhoven, N., Dorvaux, V., Braun, F., Gregoire, M.J., Jonveaux, P., Lederlin, P., Bene, M.C. & Labouyrie, E. (2004) Intravascular bone marrow accumulation in persistent polyclonal lymphocytosis: a misleading feature for B-cell neoplasm. Mod Pathol .

Gil-Fernandez, J.J., Blas, C. & Fernandez-Ranada, J.M. (2001) Persistent polyclonal B-cell lymphocytosis in a middle-aged, smoking woman with typical morphologic and genetic hallmarks. Haematologica, 86, 1007.

Girschick, H.J., Grammer, A.C., Nanki, T., Mayo, M. & Lipsky, P.E. (2001) RAG1 and RAG2 expression by B cell subsets from human tonsil and peripheral blood. J Immunol, 166, 377-386.

Gomez, P., Matutes, E., Sanchez, J., Garcia, J.M., Roman, J., Gruszka-Westwood, A. & Torres, A. (2000) An unusual form of persistent polyclonal B lymphocytosis in an infant. Br J Haematol, 110, 430-433.

Gordon, D.S., Jones, B.M., Browning, S.W., Spira, T.J. & Lawrence, D.N. (1982) Persistent polyclonal lymphocytosis of B lymphocytes. N Engl J Med, 307, 232-236.

Granados, E., Llamas, P., Pinilla, I., Tomas, J.F., Font, P., Camara, R., Olmeda, F., Arranz, R. & Fernandez-Ranada, J.M. (1998) Persistent polyclonal B lymphocytosis with multiple bcl-2/IgH rearrangements: a benign disorder. Haematologica, 83, 369-375.

Granel, B., Serratrice, J., Disdier, P., SanMarco, M., Hubert, M.M., Alessi, M.C., Cannoni, H., Camoin, L., Bouabdallah, R., Juhan-Vague, I. & Weiller, P.J. (2002) Anti-phospholipid/cofactor antibodies in three cases of persistent polyclonal B lymphocytosis. Br J Haematol, 119, 875-876.

Graninger, W.B., Seto, M., Boutain, B., Goldman, P. & Korsmeyer, S.J. (1987) Expression of Bcl-2 and Bcl-2-Ig fusion transcripts in normal and neoplastic cells. J Clin Invest, 80, 1512-1515.

Hermine, O., Lefrere, F., Bronowicki, J.P., Mariette, X., Jondeau, K., Eclache-Saudreau, V., Delmas, B., Valensi, F., Cacoub, P., Brechot, C., Varet, B. & Troussard, X. (2002) Regression of splenic lymphoma with villous lymphocytes after treatment of hepatitis C virus infection. N Engl J Med, 347, 89-94.

Himmelmann, A., Gautschi, O., Nawrath, M., Bolliger, U., Fehr, J. & Stahel, R.A. (2001a) Persistent polyclonal B-cell lymphocytosis is an expansion of functional IgD(+)CD27(+) memory B cells. Br J Haematol, 114, 400-405.

Himmelmann, A., Ruegg, R. & Fehr, J. (2001b) Familial persistent polyclonal B-cell lymphocytosis. Leuk Lymphoma, 41, 157-160.

Hockenbery, D., Nunez, G., Milliman, C., Schreiber, R.D. & Korsmeyer, S.J. (1990) Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature, 348, 334-336.

Hoffbrand, A.V., Pettit, J.E. & Moss, P.H.A. (2001) Essential Haematology. Blackwell Science, Oxford.

Hsu, S.M. (1985) Phenotypic expression of B lymphocytes. III. Marginal zone B cells in the spleen are characterized by the expression of Tac and alkaline phosphatase. J Immunol, 135, 123-130.

Hummel, M. & Stein, H. (2000) Clinical relevance of immunoglobulin mutation analysis. Curr Opin Oncol, 12, 395-402.

Ide, L., Dekoninck, A., Verburgh, E., Goossens, W., Brusselmans, C., Boeckx, N., Emonds, M.P. & Vandekerckhove, P. (2002) Persistent polyclonal B-cell lymphocytosis. Acta Clin Belg, 57, 31-33.

Isaacson, P.G. (2000) The current status of lymphoma classification. Br J Haematol, 109, 258-266.

Jacob, J. & Kelsoe, G. (1992) In situ studies of the primary immune response to (4-hydroxy-3-nitrophenyl)acetyl. II. A common clonal origin for periarteriolar lymphoid sheath-associated foci and germinal centers. J Exp Med, 176, 679-687.

Jacob, J., Kelsoe, G., Rajewsky, K. & Weiss, U. (1991) Intraclonal generation of antibody mutants in germinal centres. Nature, 354, 389-392.

Jenson, H.B., Heard, P. & Moyer, M.P. (1999) Evaluation of the effect of smokeless tobacco purified products and extracts on latent Epstein-Barr virus. Toxicology, 133, 35-42.

Ji, W., Qu, G.Z., Ye, P., Zhang, X.Y., Halabi, S. & Ehrlich, M. (1995) Frequent detection of bcl-2/JH translocations in human blood and organ samples by a quantitative polymerase chain reaction assay. Cancer Res, 55, 2876-2882.

Joseph, A.M., Babcock, G.J. & Thorley-Lawson, D.A. (2000) EBV persistence involves strict selection of latently infected B cells. J Immunol, 165, 2975-2981.

Khanim, F., Yao, Q.Y., Niedobitek, G., Sihota, S., Rickinson, A.B. & Young, L.S. (1996) Analysis of Epstein-Barr virus gene polymorphisms in normal donors and in virus-associated tumors from different geographic locations. Blood, 88, 3491-3501.

Kilger, E., Kieser, A., Baumann, M. & Hammerschmidt, W. (1998) Epstein-Barr virus-mediated B-cell proliferation is dependent upon latent membrane protein 1, which simulates an activated CD40 receptor. Embo J, 17, 1700-1709.

Klein, U., Rajewsky, K. & Kuppers, R. (1998) Human immunoglobulin (Ig)M+IgD+ peripheral blood B cells expressing the CD27 cell surface antigen carry somatically mutated variable region genes: CD27 as a general marker for somatically mutated (memory) B cells. J Exp Med, 188, 1679-1689.

Klein, U., Tu, Y., Stolovitzky, G.A., Keller, J.L., Haddad, J., Jr., Miljkovic, V., Cattoretti, G., Califano, A. & Dalla-Favera, R. (2003) Transcriptional analysis of the B cell germinal center reaction. Proc Natl Acad Sci U S A, 100, 2639-2644.

Kruetzmann, S., Rosado, M.M., Weber, H., Germing, U., Tournilhac, O., Peter, H.H., Berner, R., Peters, A., Boehm, T., Plebani, A., Quinti, I. & Carsetti, R. (2003) Human immunoglobulin M memory B cells controlling Streptococcus pneumoniae infections are generated in the spleen. J Exp Med, 197, 939-945.

Kuppers, R., Zhao, M., Hansmann, M.L. & Rajewsky, K. (1993) Tracing B cell development in human germinal centres by molecular analysis of single cells picked from histological sections. Embo J, 12, 4955-4967.

Lancry, L., Roulland, S., Roue, G., Mossafa, H., Salaun, V., Cheze, S., Gauduchon, P., Hardouin, A., Sola, B. & Troussard, X. (2001) No BCL-2 protein over expression but BCL-2/IgH rearrangements in B cells of patients with persistent polyclonal B-cell lymphocytosis. Hematol J, 2, 228-233.

Larcher, C., Fend, F., Mitterer, M., Prang, N., Schwarzmann, F. & Huemer, H.P. (1995) Role of Epstein-Barr virus and soluble CD21 in persistent polyclonal B-cell lymphocytosis. Br J Haematol, 90, 532-540.

Larcher, C., McQuain, C., Berger, C., Mitterer, M., Quesenberry, P.J., Huemer, H.P. & Knecht, H. (1997) Epstein-Barr virus-associated persistent polyclonal B-cell lymphocytosis with a distinct 69-base pair deletion in the LMP1 oncogene. Ann Hematol, 74, 23-28.

Lawlor, E., Murray, M., O'Briain, D.S., Blaney, C., Foroni, L., Sarsfield, P., Condell, D., Sullivan, F. & McCann, S.R. (1991) Persistent polyclonal B lymphocytosis with Epstein-Barr virus antibodies and subsequent malignant pulmonary blastoma. J Clin Pathol, 44, 341-342.

Li, L. & Choi, Y.S. (2002) Follicular dendritic cell-signaling molecules required for proliferation and differentiation of GC-B cells. Semin Immunol, 14, 259-266.

Linet, M.S., Bias, W.B., Dorgan, J.F., McCaffrey, L.D. & Humphrey, R.L. (1988) HLA antigens in chronic lymphocytic leukemia. Tissue Antigens, 31, 71-78.

Liu, Y., Hernandez, A.M., Shibata, D. & Cortopassi, G.A. (1994) BCL2 translocation frequency rises with age in humans. Proc Natl Acad Sci U S A, 91, 8910-8914.

Liu, Y.J. (1997) Reuse of B lymphocytes in germinal centers. Science, 278, 238-239.

Liu, Y.J. & Arpin, C. (1997) Germinal center development. Immunol Rev, 156, 111-126.

Liu, Y.J., Joshua, D.E., Williams, G.T., Smith, C.A., Gordon, J. & MacLennan, I.C. (1989) Mechanism of antigen-driven selection in germinal centres. Nature, 342, 929-931.

Liu, Y.J., Malisan, F., de Bouteiller, O., Guret, C., Lebecque, S., Banchereau, J., Mills, F.C., Max, E.E. & Martinez-Valdez, H. (1996) Within germinal centers, isotype switching of immunoglobulin genes occurs after the onset of somatic mutation. Immunity, 4, 241-250.

Liu, Y.J., Oldfield, S. & MacLennan, I.C. (1988) Memory B cells in T cell-dependent antibody responses colonize the splenic marginal zones. Eur J Immunol, 18, 355-362.

Liu, Y.J., Zhang, J., Lane, P.J., Chan, E.Y. & MacLennan, I.C. (1991) Sites of specific B cell activation in primary and secondary responses to T cell-dependent and T cell-independent antigens. Eur J Immunol, 21, 2951-2962.

Loembe, M.M., Lamoureux, J., Deslauriers, N., Darveau, A. & Delage, R. (2001) Lack of CD40-dependent B-cell proliferation in B lymphocytes isolated from patients with persistent polyclonal B-cell lymphocytosis. Br J Haematol, 113, 699-705.

Loembe, M.M., Neron, S., Delage, R. & Darveau, A. (2002) Analysis of expressed V(H) genes in persistent polyclonal B cell lymphocytosis reveals absence of selection in CD27+IgM+IgD+ memory B cells. Eur J Immunol, 32, 3678-3688.

Lossos, I.S., Tibshirani, R., Narasimhan, B. & Levy, R. (2000) The inference of antigen selection on Ig genes. J Immunol, 165, 5122-5126.

Machii, T., Yamaguchi, M., Inoue, R., Tokumine, Y., Kuratsune, H., Nagai, H., Fukuda, S., Furuyama, K., Yamada, O., Yahata, Y. & Kitani, T. (1997) Polyclonal B-cell lymphocytosis with features resembling hairy cell leukemia-Japanese variant. Blood, 89, 2008-2014.

MacLennan, I.C. & Liu, Y.J. (1991) Marginal zone B cells respond both to polysaccharide antigens and protein antigens. Res Immunol, 142, 346-351.

Matolcsy, A., Casali, P., Warnke, R.A. & Knowles, D.M. (1996) Morphologic transformation of follicular lymphoma is associated with somatic mutation of the translocated Bcl-2 gene. Blood, 88, 3937-3944.

McDonnell, T.J. & Korsmeyer, S.J. (1991) Progression from lymphoid hyperplasia to high-grade malignant lymphoma in mice transgenic for the t(14; 18). Nature, 349, 254-256.

Mili, F., Flanders, W.D., Boring, J.R., Annest, J.L. & Destefano, F. (1991) The associations of race, cigarette smoking, and smoking cessation to measures of the immune system in middle-aged men. Clin Immunol Immunopathol, 59, 187-200.

Mitterer, M., Pescosta, N., Fend, F., Larcher, C., Prang, N., Schwarzmann, F., Coser, P. & Huemer, H.P. (1995) Chronic active Epstein-Barr virus disease in a case of persistent polyclonal B-cell lymphocytosis. Br J Haematol, 90, 526-531.

Mizuno, T., Zhong, X. & Rothstein, T.L. (2003) Fas-induced apoptosis in B cells. Apoptosis, 8, 451-460.

Mossafa, H., Malaure, H., Maynadie, M., Valensi, F., Schillinger, F., Garand, R., Jung, G., Flandrin, G. & Troussard, X. (1999) Persistent polyclonal B lymphocytosis with binucleated lymphocytes: a study of 25 cases. Groupe Francais d'Hematologie Cellulaire. Br J Haematol, 104, 486-493.

Mossafa, H., Troussard, X., Valensi, F., Schillinger, F., Maynadie, M., Bulliard, G., Macintyre, E. & Flandrin, G. (1996) Isochromosome i(3q) and premature chromosome condensation are recurrent findings in chronic B-cell lymphocytosis with binucleated lymphocytes. Leuk Lymphoma, 20, 267-273.

Moszczynski, P., Zabinski, Z., Moszczynski, P., Jr., Rutowski, J., Slowinski, S. & Tabarowski, Z. (2001) Immunological findings in cigarette smokers. Toxicol Lett, 118, 121-127.

Muramatsu, M., Sankaranand, V.S., Anant, S., Sugai, M., Kinoshita, K., Davidson, N.O. & Honjo, T. (1999) Specific expression of activation-induced cytidine deaminase (AID), a novel member of the RNA-editing deaminase family in germinal center B cells. J Biol Chem, 274, 18470-18476.

Muschen, M., Rajewsky, K., Kronke, M. & Kuppers, R. (2002) The origin of CD95-gene mutations in B-cell lymphoma. Trends Immunol, 23, 75-80.

Muschen, M., Re, D., Jungnickel, B., Diehl, V., Rajewsky, K. & Kuppers, R. (2000) Somatic mutation of the CD95 gene in human B cells as a side-effect of the germinal center reaction. J Exp Med, 192, 1833-1840.

Okazaki, I., Yoshikawa, K., Kinoshita, K., Muramatsu, M., Nagaoka, H. & Honjo, T. (2003) Activation-induced cytidine deaminase links class switch recombination and somatic hypermutation. Ann N Y Acad Sci, 987, 1-8.

Paramithiotis, E. & Cooper, M.D. (1997) Memory B lymphocytes migrate to bone marrow in humans. Proc Natl Acad Sci U S A, 94, 208-212.

Pascual, V., Liu, Y.J., Magalski, A., de Bouteiller, O., Banchereau, J. & Capra, J.D. (1994) Analysis of somatic mutation in five B cell subsets of human tonsil. J Exp Med, 180, 329-339.

Pasqualucci, L., Migliazza, A., Fracchiolla, N., William, C., Neri, A., Baldini, L., Chaganti, R.S., Klein, U., Kuppers, R., Rajewsky, K. & Dalla-Favera, R. (1998) BCL-6 mutations in normal germinal center B cells: evidence of somatic hypermutation acting outside Ig loci. Proc Natl Acad Sci U S A, 95, 11816-11821.

Perreault, C., Boileau, J., Gyger, M., de Bellefeuille, C., D'Angelo, G., Belanger, R., Lacombe, M., Lavallee, R., Bonny, Y., Paquin, M. & et al. (1989) Chronic B-cell lymphocytosis. Eur J Haematol, 42, 361-367.

Reeder, C.B. & Conley, C.R. (1999) CD5+ persistent polyclonal B-cell lymphocytosis in a male. Leuk Lymphoma, 33, 593-596.

Reimer, P., Weissinger, F., Tony, H.P., Koniczek, K.H. & Wilhelm, M. (2000) Persistent polyclonal B-cell lymphocytosis--an important differential diagnosis of B-cell chronic lymphocytic leukemia. Ann Hematol, 79, 327-331.

Rodriguez, J.N., Dieguez, J.C., Martino, M.L. & Prados, D. (1996) Persistent polyclonal B lymphocytosis. Am J Hematol, 51, 246-247.

Roussel, M., Roue, G., Sola, B., Mossafa, H. & Troussard, X. (2003) Dysfunction of the Fas apoptotic signaling pathway in persistent polyclonal B-cell lymphocytosis. Haematologica, 88, 239-240.

Roy, J., Ryckman, C., Bernier, V., Whittom, R. & Delage, R. (1998) Large cell lymphoma complicating persistent polyclonal B cell lymphocytosis. Leukemia, 12, 1026-1030.

Sagaert, X. & De Wolf-Peeters, C. (2003) Classification of B-cells according to their differentiation status, their micro-anatomical localisation and their developmental lineage. Immunol Lett, 90, 179-186.

Salcedo, I., Campos-Caro, A., Sampalo, A., Reales, E. & Brieva, J.A. (2002) Persistent polyclonal B lymphocytosis: an expansion of cells showing IgVH gene mutations and phenotypic features of normal lymphocytes from the CD27+ marginal zone B-cell compartment. Br J Haematol, 116, 662-666.

Schneider, T.J., Fischer, G.M., Donohoe, T.J., Colarusso, T.P. & Rothstein, T.L. (1999) A novel gene coding for a Fas apoptosis inhibitory molecule (FAIM) isolated from inducibly Fas-resistant B lymphocytes. J Exp Med, 189, 949-956.

Schonermarck, U., Diem, H., Lohse, P. & Samtleben, W. (2003) [Persistent polyclonal B-cell lymphocytosis]. Dtsch Med Wochenschr, 128, 1115-1118.

Shaffer, A.L., Rosenwald, A. & Staudt, L.M. (2002) Lymphoid malignancies: the dark side of B-cell differentiation. Nat Rev Immunol, 2, 920-932.

Smith, K.G., Light, A., O'Reilly, L.A., Ang, S.M., Strasser, A. & Tarlinton, D. (2000) bcl-2 transgene expression inhibits apoptosis in the germinal center and reveals differences in the selection of memory B cells and bone marrow antibody-forming cells. J Exp Med, 191, 475-484.

Straus, S.E., Jaffe, E.S., Puck, J.M., Dale, J.K., Elkon, K.B., Rosen-Wolff, A., Peters, A.M., Sneller, M.C., Hallahan, C.W., Wang, J., Fischer, R.E., Jackson, C.M., Lin, A.Y., Baumler, C., Siegert, E., Marx, A., Vaishnaw, A.K., Grodzicky, T., Fleisher, T.A. & Lenardo, M.J. (2001) The development of lymphomas in families with autoimmune lymphoproliferative syndrome with germline Fas mutations and defective lymphocyte apoptosis. Blood, 98, 194-200.

Takahashi, Y., Cerasoli, D.M., Dal Porto, J.M., Shimoda, M., Freund, R., Fang, W., Telander, D.G., Malvey, E.N., Mueller, D.L., Behrens, T.W. & Kelsoe, G. (1999) Relaxed negative selection in germinal centers and impaired affinity maturation in bcl-xL transgenic mice. J Exp Med, 190, 399-410.

Takahashi, Y., Ohta, H. & Takemori, T. (2001) Fas is required for clonal selection in germinal centers and the subsequent establishment of the memory B cell repertoire. Immunity, 14, 181-192.

Tangye, S.G., Liu, Y.J., Aversa, G., Phillips, J.H. & de Vries, J.E. (1998) Identification of functional human splenic memory B cells by expression of CD148 and CD27. J Exp Med, 188, 1691-1703.

Thieblemont, C., Felman, P., Callet-Bauchu, E., Traverse-Glehen, A., Salles, G., Berger, F. & Coiffier, B. (2003) Splenic marginal-zone lymphoma: a distinct clinical and pathological entity. Lancet Oncol, 4, 95-103.

Thorley-Lawson, D.A. & Gross, A. (2004) Persistence of the Epstein-Barr virus and the origins of associated lymphomas. N Engl J Med, 350, 1328-1337.

Tierens, A., Delabie, J., Michiels, L., Vandenberghe, P. & De Wolf-Peeters, C. (1999) Marginal-zone B cells in the human lymph node and spleen show somatic hypermutations and display clonal expansion. Blood, 93, 226-234.

Tissot, J.D., Schmidt, P.M., von Fliedner, V. & Knecht, H. (1991) [Persistent B-cell polyclonal lymphocytosis: a benign lymphoproliferative syndrome]. Schweiz Med Wochenschr, 121, 1582-1584.

Tonelli, S., Vanzanelli, P., Sacchi, S., Fiorani, C., Castelli, I., Temperani, P. & Bonacorsi, G. (2000) Persistent polyclonal B lymphocytosis: morphological, immunological, cytogenetic and molecular analysis of an Italian case. Leuk Res, 24, 877-879.

Troussard, X. & Flandrin, G. (1996) Chronic B-cell lymphocytosis with binucleated lymphocytes (LWBL): a review of 38 cases. Leuk Lymphoma, 20, 275-279.

Troussard, X., Mossafa, H. & Flandrin, G. (1997a) Identity between hairy B-cell lymphoproliferative disorder and persistent polyclonal B lymphocytosis? Blood, 90, 2110-2113.

Troussard, X., Mossafa, H., Valensi, F., Maynadie, M., Schillinger, F., Bulliard, G., Malaure, H. & Flandrin, G. (1997b) [Polyclonal lymphocytosis with binucleated lymphocytes. Morphological, immunological, cytogenetic and molecular analysis in 15 cases]. Presse Med, 26, 895-899.

Troussard, X., Valensi, F., Debert, C., Maynadie, M., Schillinger, F., Bonnet, P., Macintyre, E.A. & Flandrin, G. (1994) Persistent polyclonal lymphocytosis with binucleated B lymphocytes: a genetic predisposition. Br J Haematol, 88, 275-280.

Tuscano, J.M., Druey, K.M., Riva, A., Pena, J., Thompson, C.B. & Kehrl, J.H. (1996) Bcl-x rather than Bcl-2 mediates CD40-dependent centrocyte survival in the germinal center. Blood, 88, 1359-1364.

Vignes, S., Oksenhendler, E., Quint, L., Daniel, M.T., Mariette, X. & Clauvel, J.P. (2000) [Polyclonal B lymphocytosis and hyper-IgM: immunodeficiency and/or benign lymphoid proliferation associated with tobacco?]. Rev Med Interne, 21, 236-241.

Vincenot-Blouin, A., Timbely, O., Abarah-Atassi, W., Mossafa, H., Allard, C., Michel, F. & Andre-Kerneis, E. (2003) [Binucleated lymphocyte lymphocytosis]. Ann Biol Clin (Paris), 61, 454-457.

Wang, J., Lobito, A.A., Shen, F., Hornung, F., Winoto, A. & Lenardo, M.J. (2000) Inhibition of Fas-mediated apoptosis by the B cell antigen receptor through c-FLIP. Eur J Immunol, 30, 155-163.

Weller, S., Braun, M.C., Tan, B.K., Rosenwald, A., Cordier, C., Conley, M.E., Plebani, A., Kumararatne, D.S., Bonnet, D., Tournilhac, O., Tchernia, G., Steiniger, B., Staudt, L.M., Casanova, J.L., Reynaud, C.A. & Weill, J.C. (2004) Human blood IgM "memory" B cells are circulating splenic marginal zone B cells harboring a pre-diversified immunoglobulin repertoire. Blood .

Weller, S., Faili, A., Garcia, C., Braun, M.C., Le Deist, F.F., de Saint Basile, G.G., Hermine, O., Fischer, A., Reynaud, C.A. & Weill, J.C. (2001) CD40-CD40L independent Ig gene hypermutation suggests a second B cell diversification pathway in humans. Proc Natl Acad Sci U S A, 98, 1166-1170.

Weng, W.K. & Levy, S. (2003) Hepatitis C virus (HCV) and lymphomagenesis. Leuk Lymphoma, 44, 1113-1120.

Woessner, S., Florensa, L. & Espinet, B. (1999) Bilobulated circulating lymphocytes in persistent polyclonal B-cell lymphocytosis. Haematologica, 84, 749.

Wotherspoon, A.C., Doglioni, C., Diss, T.C., Pan, L., Moschini, A., de Boni, M. & Isaacson, P.G. (1993) Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori. Lancet, 342, 575-577.

Wyatt, D.E. & Coyle, P.V. (1991) Persistent polyclonal B lymphocytosis with Epstein-Barr virus antibodies and subsequent malignant pulmonary blastoma. J Clin Pathol, 44, 966.

Yokoyama, T., Tanahashi, M., Kobayashi, Y., Yamakawa, Y., Maeda, M., Inaba, T., Kiriyama, M., Fukai, I. & Fujii, Y. (2002) The expression of Bcl-2 family proteins (Bcl-2, Bcl-x, Bax, Bak and Bim) in human lymphocytes. Immunol Lett, 81, 107-113.

Youinou, P., Jamin, C. & Lydyard, P.M. (1999) CD5 expression in human B-cell populations. Immunol Today, 20, 312-316.

Zhang, X., Li, L., Choe, J., Krajewski, S., Reed, J.C., Thompson, C. & Choi, Y.S. (1996) Up-regulation of Bcl-xL expression protects CD40-activated human B cells from Fas-mediated apoptosis. Cell Immunol, 173, 149-154.