Speckled-like Pattern in the Germinal Center (SLIP-GC), a Nuclear GTPase Expressed in Activation-induced Deaminase-expressing Lymphomas and Germinal Center B Cells

We identified a novel GTPase, SLIP-GC, with expression limited to a few tissues, in particular germinal center B cells. It lacks homology to any known proteins, indicating that it may belong to a novel family of GTPases. SLIP-GC is expressed in germinal center B cells and in lymphomas derived from germinal center B cells such as large diffuse B cell lymphomas. In cell lines, SLIP-GC is expressed in lymphomas that express activation-induced deaminase (AID) and that likely undergo somatic hypermutation. SLIP-GC is a nuclear protein, and it localizes to replication factories. Reduction of SLIP-GC levels in the Burkitt lymphoma cell line Raji and in non-Hodgkin lymphoma cell lines resulted in an increase in DNA breaks and apoptosis that was AID-dependent, as simultaneous reduction of AID abrogated the deleterious effects of SLIP-GC reduction. These results strongly suggest that SLIP-GC is a replication-related protein in germinal center B cells whose reduction is toxic to cells through an AID-dependent mechanism.The germinal center (GC)³ is a transient structure formed during T-dependent B cell responses wherein B cell affinity maturation to a specific antigen occurs, leading to the formation of high affinity memory B cells (1–3). Many features of this reaction are unique in biology such as the somatic hypermutation (SHM) of immunoglobulin (Ig) genes, the genetic rearrangement of the constant domains in class switch recombination to generate B cells bearing receptors of downstream isotypes such as IgG, IgE, and IgA, and the cellular selection process that recruits high affinity variants generated via SHM. In SHM the variable (V) regions of the heavy and light chain loci of Ig genes undergo a directed process of hypermutation where base substitutions accumulate, particularly in regions encoding the antigen binding pockets of the B cell receptor. The molecular basis for SHM is not fully understood, but it is known to be triggered by a cytosine deaminase, AID (4, 5). However, it is clear that novel factors are yet to be discovered in SHM. For example, AID alone is not sufficient for proper targeting to the Ig locus, and it is likely that a novel factor targets AID to the Ig locus (6). In addition, AID-mediated deamination of cytosines explains only mutations at G:C base pairs, yet mutations at A:T base pairs occur at approximately the same rate as G:C mutations. Although A:T mutations have been linked to the activities of the mismatch repair (MMR) proteins MSH/MSH6 and the error-prone DNA polymerase η, hypermutating Burkitt lymphoma cell lines have intact MMR and polymerase η, yet mutations at A:T base pairs are markedly reduced (7). The class switch recombination reaction is also only partly understood. Targeting of AID, the DNA substrate subjected to AID deamination, and the subsequent DNA breaks and their repair also remain only partially defined for class switch recombination. Finally, it remains unclear how these reactions are coordinated in the GC environment with both cellular selection for increased affinity to foreign antigen and tolerance mechanisms to prevent or minimize autoreactivity acquired during hypermutation that can lead to high affinity pathogenic IgG antibodies (8, 9). Clearly, efforts to understand these mechanisms and to identify novel proteins that contribute to this unique environment are needed.To identify proteins that may contribute to SHM or other aspects of the GC reaction, we mined expression libraries generated by the I.M.A.G.E. Consortium (10) through informatics tools in the Cancer Genome Anatomy Group website (11). Given that BCL6 is a critical protein for the GC reaction (12, 13), we pooled libraries derived from GC B cells with BCL6 expression and compared them to all other libraries (see Fig. 1A for the scheme). This strategy led us to the discovery of a novel protein, SLIP-GC (speckled-like pattern in the germinal center), expressed in GC B cells, and its expression profile was similar to that of AID. Subsequent experiments showed that this protein is expressed in GC B cells and localizes to replication factories in the nucleus and when reduced in AID⁺ lymphoma cell lines results in an increase in DNA breaks and in cell death. These studies reveal SLIP-GC to be a novel factor that likely contributes to the unique reactions in GCs. The data also suggest that SLIP-GC reduction is toxic to B cells through an AID-mediated mechanism.Open in a separate windowFIGURE 1.Identification of a novel GTPase expressed in germinal center B cells. A, Shown is a schematic representation of the method used to identify novel proteins primarily expressed in germinal center B cells by mining EST libraries. B, shown is the amino acid sequence of SLIP-GC. Italic boldface motifs are potential nuclear localization signals. The underlined boldfaced motif is a GTPase motif (P-loop), whereas the underlined italic sequence near the C terminus is the coiled-coil region. C, shown is a representative graph of GTPase assay of immunoprecipitated SLIP-GC from lipopolysaccharide-activated and unstimulated B cells. CRL-2289 is a cell line with endogenous SLIP-GC. As a positive control, RhoA, a ubiquitous GTPase, was immunoprecipitated with specific antibodies and tested the same way as SLIP-GC. We also performed GTPase assays on CRL-2631, which does not express SLIP-GC. Accordingly, a Coomassie Blue gel of immunoprecipitated SLIP-GC only shows a SLIP-GC band in the CRL-2289 extracts (data not shown), and SLIP-GC GTPase activity in lipopolysaccharide (LPS)-activated CRL-2631 was negligible, whereas RhoA GTPase activity was high (SLIP-GC, 0.36 nmol GDP/min; RhoA, 28.48 nmol GDP/min). The assay was done at least two times.