SEPTIN2 and STATHMIN Regulate CD99-Mediated Cellular Differentiation in Hodgkin's Lymphoma

Hodgkin’s lymphoma (HL) is a lymphoid neoplasm characterized by Hodgkin’s and Reed-Sternberg (H/RS) cells, which is regulated by CD99. We previously reported that CD99 downregulation led to the transformation of murine B lymphoma cells (A20) into cells with an H/RS phenotype, while CD99 upregulation induced differentiation of classical Hodgkin’s lymphoma (cHL) cells (L428) into terminal B-cells. However, the molecular mechanism remains unclear. In this study, using fluorescence two-dimensional differential in-gel electrophoresis and matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS), we have analyzed the alteration of protein expression following CD99 upregulation in L428 cells as well as downregulation of mouse CD99 antigen-like 2 (mCD99L2) in A20 cells. Bioinformatics analysis showed that SEPTIN2 and STATHMIN, which are cytoskeleton proteins, were significantly differentially expressed, and chosen for further validation and functional analysis. Differential expression of SEPTIN2 was found in both models and was inversely correlated with CD99 expression. STATHMIN was identified in the A20 cell line model and its expression was positively correlated with that of CD99. Importantly, silencing of SEPTIN2 with siRNA substantially altered the cellular cytoskeleton in L428 cells. The downregulation of STATHMIN by siRNA promoted the differentiation of H/RS cells toward terminal B-cells. These results suggest that SEPTIN2-mediated cytoskeletal rearrangement and STATHMIN-mediated differentiation may contribute to changes in cell morphology and differentiation of H/RS cells with CD99 upregulation in HL.


Introduction
Hodgkin's lymphoma (HL) is one of the most common malignant neoplasms affecting the lymphoid and hematological systems. Classical Hodgkin's lymphoma (cHL) is characterized by Hodgkin cells and multinucleated Reed-Sternberg cells (H/RS) [1]. Accumulating evidence suggests that H/RS cells are derived from clonal B-cells with loss of their B-cell phenotype [2]. Mature B-cells lacking B-cell receptors (BCR) normally die via apoptosis, suggesting that H/RS cells must have developed mechanisms to maintain survival. H/RS cells present a complex immunophenotype. For example, H/RS cells usually express markers associated with the myeloid lineage (CD15) and markers associated with plasma cells (CD138, MUM-1) [3,4], but rarely B-cell markers, such as CD20, Oct-2, Ig, or components of the BCR (CD79a and CD79b) [5]. The cause of the aberrant expression of a large number of B-cell genes is currently unclear. It is proposed that B-cell markers are lost due to aberrant gene regulation and cytoskeletal rearrangements [6]. H/RS cells are also characterized by abortive plasma cell differentiation [7], although, the underlying molecular mechanism is largely unknown.
CD99, a transmembrane glycoprotein encoded by the MIC2 gene, is broadly expressed in hematopoietic cells, such as B-cells, T-cells, mononuclear cells, and neutrophils [8]. CD99 is highly expressed in non-Hodgkin lymphoma, including acute lymphoblastic lymphoma [9], but rarely expressed in H/RS cells in cHL, with the mechanism still elusive. Several studies indicate that the generation of H/RS-like cells might be related to the downregulation of CD99 [10,11]. Kim et al [12] transfected IM9 (Ig-secreting lymphoblast) and BJAB (Burkitt's lymphoma) cell lines with antisense CD99 and found that downregulation of CD99 led to the generation of cells with an H/RS phenotype. We previously reported that upregulation of CD99 in L428 cell line (L428-CD99) resulted in loss of H/RS cells morphology [13]. In addition, downregulation of mouse CD99 antigen-like 2 (mCD99L2) in murine A20 cells led to induction of some H/RScell like morphologies [14]. The mCD99L2 antigen shows sequence homology to human CD99 [15]. A20 is a murine cell line derived from a spontaneously arising tumor in an aged BALB/c mouse with the characteristic pathology of human diffuse large B-cell lymphoma (DLBCL) [16,17]. Taken together, these findings suggest that CD99 plays a critical role in H/RS cellular differentiation.
To investigate the underlying mechanism by which CD99 regulates H/RS cell differentiation, we used two-dimensional differential in-gel electrophoresis (2D-DIGE) combined with matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) to identify the changes in protein expression following CD99 upregulation of L428 cells, and downregulation of mCD99L2 in A20 cells. We found that 21 proteins were upregulated and 17 downregulated in L428 cells with ectopic overexpression of CD99. Meanwhile, 21 upregulated proteins and 20 downregulated proteins were identified in A20 cells following silencing of mCD99L2 with siRNA. Among these identified proteins, SEPTIN2 and STATHMIN, which mediate cytoskeletal reorganization and play an important role in cellular differentiation, were selected for further validation and functional analysis.

Fluorescent dye labeling, 2D-DIGE, and MALDI-TOF MS
The cells were lysed on ice in lysis buffer (Cell Signaling Technology, Boston, MA, USA) supplemented with protease and phosphatase inhibitor (Roche), and 1 mM PMSF (Sigma), and centrifuged at 15,000 × g for 30 min at 4°C. A total of 50 μg of protein was labeled with one of three CyDye DIGE Fluors (GE Healthcare). Protein samples from four different groups (L428-CD99, L428-CTR, A20-mCD99L2-, and A20-CTR) were labeled with Cy3 and Cy5, respectively. The internal standard contained equal amounts of each sample labeled with Cy2. CyDyes were combined with samples at a ratio of 400 pmol of CyDye to 50 μg of sample. The labeling reaction was performed on ice and in the dark for 30 min. The reaction was then quenched by incubating with 1.5 μL of 10 mM lysine on ice and in the dark for 10 min. Both groups of proteins (L428-CD99 vs L428-CTR and A20-mCD99L2-vs A20-CTR) were pooled for 2-D gel electrophoresis for protein identification. 50 μg of each pooled protein sample was diluted in 450 mL of rehydration buffer for isoelectric focusing on Ettan IPGphor IEF System (GE Healthcare) following the manufacturer's instructions with IPG strips (24 cm, pH 3-10 NL, GE Healthcare). The IPG strips were rehydrated prior to use at 50 V for 10 h, followed by focusing at 100 V for 2 h, 300 V for 2 h, 600 V for 2 h, 1000 V for 2 h, 3000 V for 2 h, 9000 V for 2 h, and maintained at 500 V for a total of 90,000 V h. The two-dimensional separation was transferred onto 12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) on Ettan-DALT twelve perpendicular gel electrophoresis system (GE Healthcare). All CyDye-labeled gels were stained with Coomassie Blue R350 (GE Healthcare). The Cy2, Cy3, and Cy5-labeled images were scanned by a FLA5100 scanner system (GE Health) at the excitation/emission of 480 nm/530 nm (Cy2), 540 nm/590 nm (Cy3), and 620 nm/680 nm (Cy5).
Labeled image files were analyzed by DeCyder 2D 6.0 difference analysis software (GE Healthcare). All the protein spots from each gel were analyzed by Differential In-gel Analysis. The same protein spots on different gels were matched using an internal standard and analyzed using the DeCyder software (also known as BVA, Biological Variance Analysis). Differentially expressed protein spots of interest were excised from the stained gels and pooled together. The gel pieces were washed in 25 mM NH 4 HCO 3 , 50% acetonitrile (ACN) for 30 min, and then dehydrated in 100% ACN for 10 min. The proteins were digested using sequencing-grade trypsin (Promega Corporation) overnight at 37°C. Peptides were extracted in 5% TFA and 50% ACN, and dried using a Speedvac. The peptides were resuspended in 0.3% TFA, and co-crystallized by α-cyano-4-hydroxycinnamic acid (CHCA) matrix on a MALDI target. The proteins were identified using an ABI 4800 Proteomic Analyzer MALDI-TOF MS mass spectrometer (Applied Biosystems). Mass spectrometry spectra were identified in the Swiss Prot database using Global Proteome Server Explorer software (Applied Biosystems).

SiRNA Transfection
Transfection of siRNA was carried out using Lipofectamine2000 Transfection Reagent (Invitrogen, CA) following the manufacturer's protocols. For each transfection, 10 μg of siRNA oligos were used for 2 ×10 6 cells. The siRNA sequences are listed in S1 Table. The transfection efficiency was determined by quantitative real-time PCR (qRT-PCR) in triplicate.

RNA isolation, reverse transcription, and qRT-PCR analysis
Total RNA extraction, reverse transcription, and qRT-PCR were carried out as previously described [13]. Primers specific for human SEPTIN2 and STATHMIN are indicated in S2 Table. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control. The reaction conditions were 95°C for 30 sec, followed by 40 cycles of 95°C for 30 sec and 54°C for 34 sec. The relative mRNA levels were calculated using the 2 -44Ct method. The qRT-PCR experiments were repeated independently three times.

Western blot
Cells were harvested and washed twice with cold PBS. Cell lysates were prepared, and equal amounts of protein (50 μg) were separated on 8% SDS-PAGE, and transferred onto polyvinylidene difluoride (PVDF) membranes (Hercules, CA, USA). Membranes were incubated with 5% skim milk in TBS-0.1% Tween-20 for 2 h to block the residual binding sites followed by immunoblotting overnight at 4°C with appropriately diluted antibody. The antibodies used in this study are listed in S3 Table. Specific binding was revealed by mouse HRP-conjugated antirabbit IgG (Santa Cruz) and an enhanced chemiluminescence system (ECL-Plus; Amersham Biosciences, Piscataway, NJ, USA).

Immunohistochemistry and immunocytochemistry analyses
Immunohistochemistry (IHC) and immunocytochemistry (ICC) analyses were performed as previously described [20]. The antibodies used are listed in S3 Table. Evaluation of the immunohistochemical staining results was conducted independently by two pathologists (T.Z. and XH.Z.) who were blinded to the clinical data. Staining was scored as positive if at least 10% of the tumor cells were immunoreactive, and then scored as weak (1+), moderate (2+), or strong (3+) according to staining intensity.

Preparation of paraffin-embedded cell blocks
L428 cells (>5×10 7 /mL) were collected, washed twice with cold PBS, and then fixed in 10% formaldehyde overnight at room temperature without suspension. Next day, the cell block was packaged with a lens paper and placed in the paraffin-embedded box, followed by IHC.
Immunofluorescence Cells (2.0×10 5 /ml) were inoculated into each well of 6-well plates (Corning, NY, USA) and cultured in complete medium for 48 h followed by in serum-free medium for another 24 h. After deposition, fixation and permeabilization, the cells were labeled with appropriate antibodies. The antibodies used are listed in S3 Table. Negative controls were performed by replacing the primary antibodies with PBS. The cells were observed under a fluorescence microscope (Tokyo, Japan).

Flow cytometry
Control L428 cells and L428 cells transfected with STATHMIN-siRNA for 96h were harvested, washed twice with PBS and suspended at a concentration of 1×10 7 /mL. Cell surface expression of CD15, CD45, CD38 and CD138 were analyzed by FC 500 MPL flow cytometer (CA). The antibodies applied are listed in S4 Table. Data were analyzed in FCS Express V3 software (DeNovo Software, Canada).

Statistical analysis
All images, including Western blots and flow cytometry, were representative of at least three independent experiments. Quantitative RT-PCR assays were carried out in triplicate for each experiment. The data shown are expressed as the mean ± SD. Statistical analysis was performed using SPSS version 13.0 software. All comparisons were paired two-sample Student's t-tests, except that qRT-PCR experiments with STATHMIN-siRNAs or SEPTIN2-siRNAs were Oneway ANOVA. P-value of <0.05 was considered statistically significant. Proteomic database analysis to identify the proteins was based on peptide mass fingerprinting (PMF) data by searching MASCOT (http://www.matrixscience.com) software using a MASCOT Distiller, which detects peaks by fitting an ideal isotopic distribution to the experimental data. The parameters were set as follows: mass tolerance 250 ppm, number of missed cleavage sites allowed 1, cysteine residue modified as carbamidomethylcys, variable oxidative modifications (M), minimum number of matched-peptides 5, species selected as Homo sapiens (human) or Mus musculus (mice), with monoisotopic mass values. The mass and protein scores were obtained by searching MASCOT.

Proteomic analysis: differential expression of proteins
We previously found that upregulation of CD99 in H/RS cells induced terminal B-cell differentiation [13], while downregulation of mCD99L2 in BALB/c mice led to the transformation of A20 cells into H/RS-like cells [14]. To explore the mechanism underlying the CD99-mediated differentiation of H/RS cells, we conducted differential proteome analyses of the L428-CD99 cells vs L428-CTR cells using 2D-DIGE with MALDI-TOF MS. A number of spots were found to be significantly altered (P < 0.05) after upregulation of CD99 (Fig 1 and S1 Fig). Thirtyeight single spots were successfully identified using PMF (S5 Table). Of the 38 spots, 21 were upregulated and 17 were downregulated in L428-CD99 cells compared with L428-CTR cells.
To determine the potential functions of the differentially expressed proteins, we performed protein congregation analysis based on the Gene Ontology (GO) software suite (http://61.50. 138.115/GOfact/) in the terms of cellular component, biological process and molecular function (S2 Fig), and identified several closely related proteins that were potentially involved in CD99-mediated cell differentiation (S6 Table).
We also compared the protein expression profile of A20-mCD99L2-cells vs A20-CTR cells using the same methods. Results of 2D-DIGE, PMF, and Mascot-score maps are illustrated in Fig 1 and S1 Fig. Forty-one differentially expressed proteins were identified by PMF. Of these, 21 were upregulated and 20 were downregulated in the A20-mCD99L2-cells compared to the A20-CTR cells (S7 Table). Protein congregation analysis with Gene Ontology identified several proteins closely related to mCD99L2-mediated cell differentiation (S8 Table).
SEPTIN2, associated with cytoskeletal organization, was identified to be significantly downregulated in L428-CD99 cells (Fig 1B and S5 Table) but upregulated in A20-mCD99L2-cells (Fig 1C and S7 Table). In addition, STATHMIN was also overexpressed in A20-mCD99L2cells (Fig 1D and S7 Table). It is well documented that both SEPTIN2 and STATHMIN regulate cytoskeletal organization [21,22], which is associated with cellular differentiation [6]. Accordingly, we selected these two proteins for further investigation.
Differential expression of SEPTIN2 and STATHMIN in L428-CD99 and L428-CTR cells We next validated the expression status of SEPTIN2 and STATHMIN in L428-CD99 and L428-CTR cells by qRT-PCR, Western blot, ICC, and immunofluorescence. Compared with L428-CTR cells, both the mRNA (Fig 2A) and protein (Fig 2B) levels of SEPTIN2 were significantly decreased, while those of STATHMIN were increased in L428-CD99 cells. Similar downregulation of SEPTIN2 and upregulaton of STATHMIN in L428-CD99 cells were revealed by ICC ( Fig 2C) and immunofluorescence (Fig 2D). In addition, ICC and immunofluorescence images showed that SEPTIN2 and STATHMIN were primarily found in the cytoplasm ( Fig  2E). These results showed that SEPTIN2 expression was negatively correlated with CD99, whereas a synergistic expression between CD99 and STATHMIN was observed, which was further validated in the Burkitt's lymphoma cell line BJAB following silencing of CD99 by siRNA (S1 Text).  Table). In RH tissues, strong SEPTIN2 staining was not detected in any of the 5 cases; however, moderately positive staining was detected in 4/5 (80%) and weakly positive was found in 1/5 (20%). Strong  STATHMIN staining was detected in 3/5 cases (60%), with the remainder (40%) showing moderate to intermediate staining.
Moreover, in RH tissues, there was strong staining for STATHMIN in the GC, and less dense, more scattered staining in the mantle zones. Unlike STATHMIN, SEPTIN2 showed weak staining in the GC, but stronger expression in the mantle zones. In capp:ds:hodgkin lymphoma(hl)HL tissues, histologically-confirmed H/RS cells showed strong to intermediate STATHMIN expression, and variable background STATHMIN expression. In contrast, SEP-TIN2 showed intermediate to strong expression in the cytoplasm of H/RS cells, but slightly stronger expression in background Non-H/RS cells. We also observed stronger staining for STATHMIN in the RH cases than in the cHapp:ds:hodgkin lymphoma(hl)L, but weaker expression of SEPTIN2.

SiRNA silencing of SEPTIN2 results in cytoskeletal reorganization
Bioinformatics analysis indicated that SEPTIN2 was associated with cytoskeletal organization (S6 and S8 Tables). The differential expression of SEPTIN2 in L428-CTR and L428-CD99 cells suggests that SEPTIN2 might play a role in the regulation of the cytoskeletal dynamics in cHL. To test this hypothesis, we silenced SEPTIN2 with siRNA and visualized SEPTIN2 and F-actin in L428 cells by immunofluorescence. As shown in Fig 4, SEPTIN2 colocalized with F-actin in the cytoplasm of the control cells with a punctate distribution at the cell-substratum interface. SiRNA-mediated downregulation of SEPTIN2 reduced the expression of SEPTIN2 and F-actin in L428 cells. Moreover, SEPTIN2 downregulation led to fewer filopodia and thinner cortical filamentous actin at the cell surface. Thus, downregulation of SEPTIN2 may induce cytoskeletal rearrangement in L428 cells. The findings were consistent with our previous study showing that CD99-upregulated L428 cells underwent terminal B-cell differentiation and displayed cytoskeleton reorganization, disappearance of filopodia and thinning of cortical filamentous actin [13]. The CD99-upregulated L428 cells showed reduced expression of SEPTIN2. Taken together, these results suggest that SEPTIN2 plays a role in maintaining H/RS cell morphology and SEPTIN2-induced cytoskeletal reorganization may contribute to CD99-mediated differentiation of H/RS cells.

Dynamic expression of STATHMIN is correlated with B-cell lymphoma differentiation
STATHMIN plays an important role in cellular differentiation [22][23][24]. We previously showed that overexpression of CD99 promoted the differentiation of lymphoma cells into terminal Bcells [13]. The differential expression of STATHMIN in L428-CTR and L428-CD99 cells suggests that STATHMIN may contribute to the differentiation of HL. To test this, we evaluated STATHMIN expression by IHC in 65 lymphoid neoplasms, including 20 cHL (S10 Table). In RH, the majority of the STATHMIN-positive cells were located in the GC, and only occasionally found in the interfollicular areas. In cHL, strong to intermediate staining of STATHMIN was detected in H/RS cells. Increasingly variable expression levels were found in other lymphoid cells. DLBCL were also stained positive for STATHMIN with varying degrees of intensity. The staining intensity of STATHMIN in DLBCL derived from GCB cells and plasmacytomas (PCM) was stronger than that in the non-GCB cells (Fig 5A). These data demonstrate that STATHMIN expression decreases from the GC to post-GC stages, and increases from the post-GC to plasma stages. Our results were consistent with the previous finding that STATHMIN was involved in B-cell lymphoma differentiation [25].

Downregulation of STATHMIN induces H/RS cell differentiation toward plasma cells
To further confirm the role of STATHMIN in HL differentiation, we downregulated STATH-MIN by siRNA in L428 cells (Fig 5B and 5C) and assessed the expression of PRDM1, which plays an important role in terminal differentiation of B-cells into immunoglobulin (Ig)- secreting plasmablasts and plasma cells [26]. SiRNA-mediated STATHMIN downregulation led to increase in PDRM1 protein level (Fig 5D). We next evaluated the expression of B-cell differentiation markers using flow cytometry. As shown in Fig 5E, the percentage of cells expressing CD15, the diagnostic marker for cHL, was obviously reduced in STATHMIN-L428-siRNA cells compared with the control cells. Consistent with the role for STATHMIN in B-cell differentiation, plasma cell phenotypic markers CD38 and CD138 were increased following STATH-MIN silencing. We also silenced STATHMIN in L428-CD99 cells (Fig 5F and 5G). In sharp contrast to L428-CD99 cells, PRDM1 protein level was reduced after silencing of STATHMIN in L428-CD99 cells (Fig 5H). Taken together, these results suggest STATHMIN induced differentiation of H/RS cells to plasma cells.

SEPTIN2 silencing leads to downregulation of STATHMIN
Our results clearly showed that SEPTIN2 and STATHMIN were closely associated with CD99 expression, suggesting a potential relationship between SEPTIN2 and STATHMIN. Therefore, we examined the expression of both SEPTIN2 and STATHMIN in L428 cells after downregulation of SEPTIN2 with siRNA and in L428-CD99 cells following downregulation of STATHMIN with siRNA, respectively. Surprisingly, STATHMIN expression was reduced when the expression of SEPTIN2 was decreased in L428 cells (Fig 6A). In contrast, almost no change in SEP-TIN2 levels was observed in the L428-CD99 cells following STATHMIN downregulation ( Fig  6B). These results indicate that SEPTIN2 regulates STATHMIN. The mechanism remains to be determined.

Discussion
In this study, we elucidated the molecular mechanisms of CD99-mediated H/RS cellular differentiation, using 2D-DIGE and MALDI-TOF MS analyses and identified proteins that were differentially expressed in L428-CD99 vs L428-CTR cells and A20-mCD99-L2-vs A20-CTR cells. SEPTIN2 and STATHMIN, which are involved in cytoskeleton organization, were found to be significantly expressed (S6 and S8 Tables), and were selected for further validation and functional analysis.
SEPTIN2 is a member of the septin family that is involved in many cellular functions, such as cell polarity, cell cortex compartmentalization, vesicle transport, and regulation of the actin and tubulin cytoskeleton [27]. Septins are highly expressed in many tumors [28]. Moreover, septins are also associated with a filament-forming cytoskeletal GTPase, which is required for normal organization of the actin cytoskeleton [29]. Actin cytoskeleton not only plays a pivotal role in the regulation of the morphology and apoptosis of B-cells [30], but also mediates the BCR signal transmission during B-cell differentiation [31]. Depolymerization of the actin cytoskeleton inhibits surface BCR clustering [32] and induces a signaling cascade in the absence of antigen [33]. In addition, evidence suggests that SEPTINS play an important role in cytoskeletal dynamics. SEPTIN-F-actin linkage may contribute to higher-order organization of the cortical cytoskeleton [34,35]. The relationship between SEPTIN filaments and actin has been well established, however, very little is known about its role in HL and H/RS cell differentiation. In this study, we found that SEPTIN2 expression was negatively correlated with CD99 and SEP-TIN2 downregulation triggered the reduction of F-actin and pseudopodia apophysis in L428 cells. These results were consistent with the previous finding that upregulation of CD99 altered cytoskeleton in L428 cells [13]. These data indicate that SEPTIN2 plays an important role in maintaining H/RS cell shape. Thus, SEPTIN2-mediated cytoskeleton reorganization may be one of the induction mechanisms of H/RS cellular differentiation by CD99 overexpression. Nevertheless, the relationship between SEPTIN2 and BCR is unknown and needs further investigation.
STATHMIN is another cytoskeletal protein, which is overexpressed in several malignancies with a significant role in cell differentiation [23]. Reducing the level of STATHMIN reverses many phenotypes associated with transformation [36]. Moreover, STATHMIN is a highly conserved cytosolic phosphoprotein implicated in regulating microtubule dynamics [37,38], cell proliferation and differentiation [39][40][41][42]. Nylander et al. reported that STATHMIN exhibited variable expression in malignant lymphomas and proposed that it may be involved in B-cell differentiation [25]. Our results obtained from IHC support this idea. However, whether STATHMIN participates in CD99-mediated H/RS cell differentiation toward terminal B-cells is unclear.
Interestingly, the present study demonstrated that downregulation of STATHMIN in L428 cells resulted in reduction of CD15, a characteristic marker of H/RS cells [43], and expression of plasma-cell markers CD38 and CD138. CD38 expression is tightly regulated during B-cell ontogenesis and is present at high levels in terminally differentiated plasma cells [44]. Furthermore, PRDM1 is a necessary and crucial factor in the regulation of B-cell differentiation toward plasma cells [26]. The results showed that silencing of STATHMIN increased the expression of PRDM1 in L428 cells. Our results suggested that downregulation of STATHMIN may stimulate L428 cellular differentiation toward plasmablasts or plasma cells (terminal B-cells). In addition, we investigated the role of STATHMIN-mediated differentiation in CD99 cells, by silencing STATHMIN in L428-CD99 cells with siRNA. We found that silencing of STATHMIN led to decrease in PRDM1, which is in contrast to that in L428 cells. Therefore, it is possible that STATHMIN was involved in the regulation of CD99-mediated differentiation of H/RS cells toward terminal B-cells.
In addition, our previous study demonstrated that ectopic overexpression of CD99 resulted in cell growth inhibition in L428 cells, but the underling mechanism was unclear [13]. It was reported that STATHMIN regulates dynamics of the mitotic spindle by promoting microtubule depolymerization [45]. When cells enter mitosis, phosphorylation-dependent inactivation of STATHMIN allows microtubules to polymerize and assemble into a mitotic spindle. The microtubule-depolymerizing activity of STATHMIN must be restored in the later phases of mitosis to allow mitotic spindle disassembly and proper exit from mitosis [37]. Both overexpression and downregulation of STATHMIN results in mitotic spindle abnormalities and accumulation of cells in G2/M phases of the cell cycle [46]. Therefore, we speculate that downregulation of STATHMIN in L428-CD99 cells may contribute to its growth inhibition.
To the best of our knowledge, we are the first to report the expression of SEPTIN2 in cHL tissues. The staining intensity of SEPTIN2 in the cytoplasm of H/RS cells was weaker than in other inflammatory cells; whereas in RH tissues, SEPTIN2 was expressed at higher levels in the mantle zone than in the GC. These results suggest that the patterns of SEPTIN2 expression in RH and cHL were contrary to STATHMIN expression, for unknown reasons.
Furthermore, several proteins known to be involved in cHL like ARHGDIB, MSH2 or PRDX2 were found differentially expressed in L428 cells as well as in mCD99L2 downregulated A20 cells (S5 and S7 Tables). Evidence shows that the guanosine triphosphatase (GTPase) inhibitor ARHGDIB is downregulated in H/RS cells and the absence of ARHGDIB might contribute to the apoptotic resistance of H/RS [47]. PRDX2 is also downregulated in H/RS cells and epigenetic silencing of this gene may contribute to the loss of B-cell identity and survival of H/ RS cells [48]. MSH2 transcript is present in most B-cell lymphoma with the exception of plasma cell lymphoma and deregulation of MSH2 in B-cell lymphoma types is characterized by aggressive biologic behavior [49]. In this study, proteomic analysis showed upregulation of CD99 in L428 cells led to high expression of ARHGDIB and PRDX2, while low expression of MSH2. These results suggest that ARHGDIB, PRDX2 and MSH2 may play a role in CD99-induced transformation of H/RS cells toward B-cell.
In summary, we characterized the expression pattern of SEPTIN2 in cHL and provided evidence that SEPTIN2-mediated cytoskeleton reorganization plays an important role in H/RS cell differentiation. Furthermore, we found that CD99 induced the transformation of H/RS cells toward B-cell by regulating the expression of SEPTIN2 and STATHMIN (Fig 7). The present study provides novel insights into the mechanisms underlying CD99-mediated H/RS cell differentiation toward B-cells.   Table. MALDI-TOF MS identification of 38 characteristic proteins with differential expression in L428-CD99 vs L428-CTR cells.

Supporting Information
(XLSX) S6 Table. Closely related proteins involved in the CD99 regulation of cell transformation. (XLSX) S7 Table. MALDI-TOF MS identification of 41 characteristic proteins with differential expression levels in A20-mCD99L2-vs A20-CTR cells.