BAFF attenuates oxidative stress-induced cell death by the regulation of mitochondria membrane potential via Syk activation in WiL2-NS B lymphoblasts

Cell survival is facilitated by the maintenance of mitochondrial membrane potential (MMP). B cell activating factor (BAFF) plays a role in survival, differentiation, and maturation of B cells. In the present study, we examined whether BAFF could attenuate oxidative stress-induced B cell death by the regulation of MMP collapse via spleen tyrosine kinase (Syk) activation using WiL2-NS human B lymphoblast cells. BAFF binds to receptors on WiL2-NS cells. When the cells were incubated in serum-deprived conditions with 1% fetal bovine serum (FBS), BAFF reduced the percentage of dead cells as determined through trypan blue staining and caspase 3 activity. BAFF also inhibited MMP collapse with 1% FBS, as indicated by a decrease in the number of cells with high-red fluorescence of MitoProbe™ JC-1 reagent or a decrease in the percentage of DiOC6-stained cells. Reactive oxygen species (ROS) production was reduced by incubation with BAFF in the presence of 10% or 1% FBS. BAFF inhibited MMP collapse, cell growth retardation, dead cell formation, and caspase 3 activation caused by treatment with H2O2. Syk phosphorylation on tyrosine (Y) 525/526 was increased in cells incubated with 1% FBS in the presence of BAFF than cells incubated with 1% FBS or BAFF alone. BAY61-3606, a Syk inhibitor reduced the effect of BAFF on MMP collapse, caspase 3 activation, cell growth retardation, and dead cell formation. Together, these data demonstrate that BAFF might attenuate oxidative stress-induced B cell death and growth retardation by the maintenance of MMP through Syk activation by Y525/526 phosphorylation. Therefore, BAFF and Syk might be therapeutic targets in the pathogenesis of B cell-associated diseases such as autoimmune disease.


Measurement of mitochondrial membrane potential.
To measure mitochondrial membrane potential (MMP), cells were stained with 2.5 μg/ml JC-1 or DiOC 6 for 10 min at 37 °C. Then, cells stained with JC-1 were observed by 485 nm filter of fluorescence microscope and five areas were pictured to count number of high red fluorescent (J-aggregate) live cells. Dead or dying cells exhibit yellow or green fluorescence (J-monomer) with collapsed mitochondrial potential, respectively. Data were represented by the percentage of high red-fluorescence cells 37 . In addition, cells stained with DiOC 6 were analyzed by FACSCalibur™ (Becton Dickinson, San Joes, CA, USA). Cells with MMP collapse showed a decrease in DiOC 6  trypan blue exclusion assay. Diluted cell suspension was mixed with equal volume of 0.4% trypan blue in PBS. Dying or dead cells were stained with blue color and viable cells were unstained. Each cell was counted by using hemocytometer under light microscope (Olympus Korea Co., Ltd, Seoul, Rep. of Korea) 35 .
Statistical analyses. Experimental differences were examined separately for statistical significance using ANOVA and Students' t-distribution. The p value of < 0.05 or < 0.01 was considered to be significant 35,37 .

BAff rescued cells from oxidative stress-induced cell death.
To examine the effects of BAFF on a human B cell line, we determined whether BAFF could bind to the cell surface of WiL2-NS human B cells that expressed BAFF receptors. To do this, cells were incubated with biotin-labelled BAFF or antibodies to BAFF receptors. The results revealed a significant increase in BAFF binding on the surface of WiL2-NS cells (Fig. 1A). The expression level of BAFF-R was higher than that of TACI (Fig. 1B). However, no BCMA expression was detected (data not shown), which suggests that BAFF might affect WiL2-NS cells via BAFF binding on the cell surface. Next, we used WiL2-NS cells to assess the effect of BAFF on molecular and cellular changes that control B cell death or survival. BAFF increases B cell survival 1,9,10 and is associated with an increase in the incidence of autoimmune disease. Additionally, serum deprivation (SD) leads to apoptotic cell death 38,39 ; therefore, we examined the effects of BAFF on oxidative stress-induced cell death by using 1% rather than 10% FBS. As shown in Fig. 1C, the cell number was reduced to about half by incubation with 1% FBS compared to 10% FBS. When cells were incubated with 10% or 1% FBS-containing medium in the presence or absence of BAFF, no changes were detected in total cell number. The number of dead cells was determined using a trypan blue staining assay. The percentage of cell survival with 1% FBS decreased to about 50% as compared to about 80% with 10% FBS. When cells were incubated with 1% FBS-containing medium in the presence of BAFF, the percentage of cell survival was rescued to about 70% (Fig. 1D). In 1% FBS, a caspase 3 assay revealed that cell death increased by about 1.6 times compared to that with 10% FBS. Caspase 3 activity induced by 1% FBS decreased about 1.2 times by treatment with BAFF (Fig. 1E). This suggests that BAFF binding might activate or change molecules or intracellular events to control B cell death.
BAFF attenuated MMP collapse in serum-deprived (SD) conditions. MMP plays a role in the regulation of various cellular functions 37,40 , and SD induces apoptotic cell death of rat retinal ganglion cells via mitochondrial changes 38 . Therefore, we determined the effects of BAFF on MMP changes using MitoProbe™ JC-1 reagent in WiL2-NS cells that were incubated with 1% FBS. The number of live cells with high red fluorescence was determined to assess the MMP of cells incubated with 1% FBS-containing medium. The MMP decreased to about 40% as compared to 83% with 10% FBS. When cells were incubated with 1% FBS-containing medium in the presence of BAFF, the MMP was significantly rescued to about 60% ( Fig. 2A). MMP changes by 1% FBS were also determined by incubation with DiOC 6 . As shown in Fig. 2B, the fluorescence intensity of DiOC 6 was decreased by incubation with 1% FBS. The percentage of cells with low fluorescence intensity of DiOC 6 increased . MFI was represented as bar graph with the means ± SD (B, right). All experiments were performed four times. **p < 0.01; significant difference as compared to BAFF-untreated control with 10% FBS. # p < 0.05; significant difference as compared to BAFF-untreated control with 1% FBS.    (Fig. 4B). Incubation with H 2 O 2 reduced the total cell number by about 40% compared to the control group, which was increased slightly in the presence of BAFF (Fig. 4C). Using trypan blue staining to assess the number of dead cells, the percentage of cell survival was reduced to about 89% as compared to about 96% in 10% FBS. When cells were incubated with H 2 O 2 in the presence of BAFF, the percentage of cell survival was rescued to about 95% (Fig. 4D). This suggests that BAFF binding might activate or change molecules, or alter intracellular events to control H 2 O 2 -induced B cell death.
Syk phosphorylation at Y525/526 was activated by BAFF or serum deprivation. Because Syk can be activated by BAFF 42 and plays an important role in B cell survival 42 , we sought to identify the tyrosine site of Syk that might be activated by BAFF binding to WiL2-NS lymphoblast B cells. When cells were incubated with 20 ng/ml of BAFF, we observed a time-dependent increase in Syk phosphorylation at Y525/526 (Fig. 5A). However, few changes were detected in Syk phosphorylation at Y323 or Y352. Syk phosphorylation at Y525/526 was enhanced by incubation with 1% FBS (Fig. 5B). When cells were incubated with 1% FBS in the presence of BAY61-3606, a Syk inhibitor, we observed a decrease in Syk phosphorylation at Y525/526 (Fig. 5C). In contrast, when cells were incubated with 1% FBS-containing medium in the presence of BAFF, Syk phosphorylation at Y525/526 was greatly increased compared to that in the control group with 1% FBS (Fig. 5D). These results indicate that BAFF provided a much stronger signal for B cell survival via Syk phosphorylation. Therefore, Syk activation might be required for B cells to survive in an environment that induces cell death.

Syk inhibitor enhanced MMp collapse in serum-deprived conditions. Because Syk is localized
in the mitochondrial inter-membrane space where it can regulate respiratory activity 42 , we tested the role of  (Fig. 6A, top). The results revealed that the MMP with 1% FBS decreased to about 18% as compared to 93% with 10% FBS. Treatment with BAFF significantly rescued the MMP to about 32%, which was inhibited by BAY61-3606. However, few changes were observed by treatment with BAY61-3606 alone (Fig. 6A, bottom). DiOC 6 was also used to confirm the effect of the Syk inhibitor on MMP collapse. As shown in Fig. 6B, the fluorescence intensity of DiOC 6 was decreased by incubation with 1% FBS. The percentage of cells with low fluorescence intensity of DiOC 6 was increased to about 28.3%, which was reduced to about 20.6% by incubation with BAFF as compared to 3.2% with 10% FBS (Fig. 6B, top). The MFI of the cell population with 1% FBS was 78.5, which was increased to about 88.7 by incubation with BAFF as compared to about 106.6 with 10% FBS. Additionally, when cells were incubated with 1% FBS in the presence of BAY61-3606, the number of cells with low fluorescence intensity of DiOC 6 significantly increased to about 60.7%, which was recovered to 53.4% by incubation with BAFF. The MFI of the cell population in the presence of BAY61-3606 was decreased to about 61.1, which was also enhanced to 68.5 by incubation with BAFF (Fig. 6B, bottom). This suggests that MMP collapse could be maintained by BAFF via Syk activation.

BAff-mediated inhibition of B cell death was attenuated by Syk inhibition.
To confirm the role of Syk activation by BAFF in the inhibition of B cell death, WiL2-NS cells were incubated with 1% FBScontaining medium in the presence of BAY61-3606 and/or BAFF. Caspase 3 activity was enhanced by 1% FBS about 1.8 times as compared to that with 10% FBS. Caspase 3 activity induced by 1% FBS was reduced to about 1.2 times by treatment with BAFF, which was inhibited by BAY61-3606. Additionally, almost the same changes in caspase 3 activity were observed by treatment with BAY61-3606 alone (Fig. 7A). The total cell number was reduced to about 70% by incubation with 1% FBS as compared to that with 10% FBS. When cells were incubated with 1% FBS-containing medium in the presence or absence of BAFF, the total cell number decreased by incubation with 1% FBS was increased to about 90% by treatment with BAFF. However, the total cell number was highly reduced by treatment with BAY61-3606 in the presence or absence of BAFF (Fig. 7B). Trypan blue staining was used to determine the number of dead cells; the percentage of cell survival was calculated and was found to be reduced about 92% with 1% FBS as compared to about 98% with 10% FBS, which was inhibited by BAY61-3606. When cells were incubated with 1% FBS-containing medium in the presence of BAFF, the percentage of cell survival was rescued to about 99%, which was also inhibited by BAY61-3606 (Fig. 7C). These results  www.nature.com/scientificreports/ were consistent with changes in Syk phosphorylation at Y525/526 (Fig. 7D, left). When cells were incubated with 1% FBS-containing medium, Syk phosphorylation at Y525/526 increased about 2.5-fold as compared to control with 10% FBS, which was inhibited by BAY61-3606. Syk phosphorylation at Y525/526 increased about 4.7-fold by incubation with 1% FBS in the presence of BAFF, which was also inhibited by BAY61-3606 (Fig. 7D,  right). This suggests that BAFF might regulate B cell survival by controlling MMP via oxidative stress-induced Syk activation (Fig. 7E).

Discussion
B cells play a role in the progression of autoimmune disease by releasing inflammatory cytokines, presenting autoantigens, producing antibodies, and ensuring interactions with T cells 43 . Therefore, many researchers are investigating how B cells can survive in the aggravation of autoimmune disease. ROS are increased in most inflammatory disease sites and are highly reactive due to their unpaired electron [11][12][13][14] . ROS that are excessively increased in inflammatory responses could affect cell fate by damaging various cellular components such as DNA, lipids, proteins 13,14 , and organelles 15 . For example, the electron transport chain in mitochondria can be disrupted in the middle of cell death 15 . BAFF enhances B cell survival, which is indispensable for B cell maturation and the enhancement of immune responses. Excessive responses by elevated BAFF levels lead to the accumulation of antigen-antibody immune complexes, which aggravate autoimmune diseases such as RA, SLE, and type 1 diabetes 1,9,10 . ROS can induce B cell activation, which is similar to the response mediated via B cell antigen receptors. Syk is a molecule that is activated by ROS 28,29 . However, little is known about the regulation of B cell survival by Syk activation-associated MMP control via ROS induction. BAFF mediates ROS control: here, we investigated whether B cell survival could be regulated via BAFF-mediated control of ROS and MMP collapse    www.nature.com/scientificreports/ BAFF can bind three receptors: BAFF-R, TACI, and BCMA. Among these, BAFF-R has binding affinity for only BAFF, and BCMA has the weakest binding affinity. TACI and BCMA also have affinity to APRIL, which has a similar structure to BAFF 1 . Our results revealed that BAFF binding was significant on the cell surface of WiL2-NS human B lymphocytes. While WiL2-NS cells expressed high levels of BAFF-R and low levels of TACI, no BCMA expression was detected. BAFF has a different binding affinity to each BAFF receptor, so it is necessary to determine how much each receptor contributes to BAFF-induced survival.
Although the exact cause of autoimmune disease is not well-understood, auto-antibodies from auto-reactive B cells that attack host cells are cited as one of the main causes 44,45 . To produce autoantibodies, B cells must overcome many harmful kinds of environments such as those with a high amount of ROS or MMP collapse, which threaten cell death in disease sites. BAFF enhances B cell survival, which is indispensable for B cell immune responses that aggravate autoimmune diseases 1,9,10 . Our research, conducted in serum-deprived conditions, provides the groundwork for an experimental model to test the effect of therapeutic candidates on BAFF-mediated autoimmune disease-like states with ROS production and MMP collapse.
BAFF up-regulated MMP through AKT and PKC-β in mouse B cells 19 . BAFF also activated Syk 31 in B cells of different species. BAFF expression was also increased by ROS in mouse macrophages 20 . Therefore, it is possible for BAFF expression to be regulated by Syk activation.
Syk belongs to the Syk family of tyrosine kinases and transmit signals from B-cell receptor (BCR), T-cell receptor (TCR), and BAFF receptors for various cellular responses 26 . Syk consists of many tyrosine residues that can potentially be phosphorylated. In human Syk, Y323, Y352, and Y525/526 are the most reported residues. Each of the tyrosine phosphorylation sites leads to a different downstream pathway. Phosphorylated Y323 becomes the binding site for the E3 ubiquitin ligase CBL, which mediates Syk ubiquitylation and degradation 9,23,25 . Phosphorylated Y352 is the binding site for the PI3K regulatory subunit p85α N-terminal SH2 domain. It is also the binding site for the C-terminal SH2 domain of PLC-γ 24 . Y525/526 is located in the kinase domain, and its phosphorylation plays a crucial role in Syk activation 27 . Our results indicate that SD or BAFF treatment increased Syk phosphorylation at Y525/526. This suggests that BAFF-mediated cell survival could be dependent on Syk activation to control MMP collapse by SD.
Syk also contributes to B cell survival and differentiation 21,22 . Therefore, Syk activation in our study could be associated with different molecules to maintain cell survival. Syk might inhibit the activation of caspase-9 through AKT 13 . AKT plays an important role in B cell survival and is activated in a Syk-dependent pathway 46 . Syk also protects cells from apoptosis induced by oxidative or genotoxic stress by stabilizing the mRNA for Bclx(L), an antiapoptotic protein 32 . Syk also transduces BAFF survival signals via extracellular signal-regulated kinase (ERK) and PI3 kinase 31 .
At the same time, the possibility that Syk might activate cell death pathways cannot be ruled out. Sykdependent PLCγ2 activation was required for acceleration towards apoptosis following oxidative stress 13 . Mule (also known as ARF-BP1, HUWE1, Ureb1, LASU1, and HECTH9) belongs to the E5-AP C terminus (HECT) domain-containing ubiquitin ligase (E3). Mule phosphorylation by Syk induces TNF-induced c-Jun N-terminal kinase (JNK) activation and cell death 47 . As such, it is necessary to define Syk-associated molecules by BAFF binding on cell surfaces in the inhibition of SD-induced MMP collapse.
In conclusion, our findings demonstrate that SD-induced ROS production causes MMP collapse through oxidative stress, which leads to cell death. Furthermore, this was attenuated by BAFF binding on BAFF receptors via Syk activation. Together, our results suggest that Syk activation by BAFF-BAFF receptor binding could regulate ROS production and inhibit MMP collapse and B cell death. These results provide additional evidence that BAFF or Syk may be valuable targets for B cell-mediated autoimmune disease therapies, such as the chemical KR33426 36 .