Umbilical Cord-Derived Mesenchymal Stem Cells Suppress Autophagy of T Cells in Patients with Systemic Lupus Erythematosus via Transfer of Mitochondria

Aberrant autophagy played an important role in the pathogenesis of autoimmune diseases, especially in systemic lupus erythematosus (SLE). In this study, we showed that T cells from SLE patients had higher autophagic activity than that from healthy controls. A correlation between autophagic activity and apoptotic rate was observed in activated T cells. Moreover, activation of autophagy with rapamycin increased T cell apoptosis, whereas inhibition of autophagy with 3-MA decreased T cell apoptosis. Umbilical cord-derived mesenchymal stem cells (UC-MSCs) could inhibit respiratory mitochondrial biogenesis in activated T cells to downregulate autophagy and consequently decrease T cell apoptosis through mitochondrial transfer and thus may play an important role in SLE treatment.


Introduction
Mesenchymal stem cells, or MSCs, are multipotent stromal cells that can differentiate into a variety of cell types, including osteoblasts, chondrocytes, myocytes, and adipocytes [1]. They can be derived from bone marrow and other nonmarrow tissues, such as the placenta, umbilical cord blood, adipose tissue, adult muscle, corneal stroma, or the dental pulp of deciduous baby teeth. Besides differentiation capacity, numerous studies have demonstrated MSCs have immunomodulatory effects [2]. And transplantation of umbilical cord-derived mesenchymal stem cells (UC-MSCs) has been shown effective in patients with autoimmune diseases, especially systemic lupus erythematosus (SLE) [3][4][5]. UC-MSCs could inhibit T cell proliferation in lupus patients by secreting large amounts of indoleamine 2,3-dioxygenase (IDO), which is enhanced by IFN-produced by CD8+T cells [6]. They also suppress B cell proliferation and differentiation [7] and regulate Treg/Th17 balance in SLE patients [8]. However, their regulation on autophagy has not been assessed. Recent works indicated the important role of autophagy in SLE [9,10], which demonstrated that T cells from SLE patients have overactivated autophagy [11,12].
Autophagy is the natural, destructive mechanism that disassembles, through a regulated process, unnecessary or dysfunctional cellular components. Peripheral T cells from SLE patients have been shown to have defective mitochondria, demonstrated by mitochondrial hyperpolarization and ATP depletion [13]. Prolonged autophagy activation usually leads to a high turnover rate of proteins and organelles, and a high rate above the survival threshold could result in cell death [14] and thus may be related to increased apoptosis of peripheral T cells from SLE patients [15]. In this study, the link between autophagy and apoptosis of activated T cells from SLE patients was addressed and the regulation of UC-MSCs on T cell autophagy was investigated.   (3/9) Clinical characteristics are presented as mean ± SEM or number (%). SLEDAI, SLE disease activity index; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; HCQ, hydroxychloroquine; ISA, immunosuppressive agent. The normal ranges were as follows: ESR (0-20 mm/h) and CRP (0-8 mg/L).

Patients and Healthy Controls.
We included 32 SLE patients (male : female = 7 : 25, mean age = 36, range 16 to 49 years old) who were diagnosed according to the classification criteria of the American College of Rheumatology [16]. Current SLE disease activity was measured using the SLE Disease Activity Index (SLEDAI) [17]. The mean ± standard error of mean (SEM) SLEDAI score was 6.5 ± 0.9. Among these patients, 9 were measured for basal autophagic activity and mitochondrial mass (detailed clinical characteristics and laboratory features were shown in Table 1 Apoptosis was measured after autophagy activation with rapamycin (Cayman, 53123-88-9) or AMPK Activator VI RSVA314 (Merck&Millipore, 171272-10MGCN) and after being treated with autophagy suppressor 3-MA (Cayman, 13242-50 mg) or chloroquine (Sigma-Aldrich, C6628-25G).
Respiratory mitochondrial mass (mitochondria that maintains transmembrane potential) was detected with Mito-Tracker Deep Red (Invitrogen, M22426), and total mitochondrial mass (including respiratory mitochondria and mitochondria that have lost transmembrane potential) was detected with MitoTracker Green (Invitrogen, M7514) staining. Data were collected on FACS Calibur (BD Bioscience) and analyzed using FlowJo software (Tree Star).

RNA Interference Transfection in Human T Lymphocytes.
The silencing of ATG5 was performed with a chemically synthesized siRNA (5 -AACCUUUGGCCUAAGAAGAdTdT-3 ; Bioinshine, China) and a control siRNA (Bioinshine, China) with a random sequence not present in the human genome (5 -UUCUCCGAACGUGUCACGUdTdT-3 ). 10 6 T lymphocytes were resuspended in 250 l of siRNA-lipid complex (Lipofectamine RNAiMAX Reagent, Invitrogen) and then incubated in complete medium for 48 h before harvesting. Apoptosis of T lymphocytes was detected by flow cytometry.

Statistical
Analysis. Data were summarized as mean ± SEM. Statistical significance was performed by Student'stest and the correlation coefficient between autophagy and apoptosis was analyzed by linear regression. A value < 0.05 was considered statistically significant. In order to avoid the influence of different cell counts among data, some flow cytometry results were shown as "normalized to mode" (mode mean "the most frequent value in a data set") which was a ratio of cell numbers. All statistical analyses were performed using GraphPad Prism software (Graph-Pad, San Diego, CA, USA) or SPSS version 13.0 statistical software (IBM SPSS, Chicago, IL, USA).

SLE Patients Had Excessive Mitochondria within T Cells.
As autophagy was essential for mitochondrial turnover, we detected mitochondrial mass in T cells from SLE patients. With the use of MitoTracker Deep Red (MDR) and Mito-Tracker Green (MG) staining, we found T cells from SLE patients had elevated respiratory mitochondrial mass (152.50 ± 1.50 versus 13.52 ± 3.54, < 0.0001) and total mitochondrial mass (3079.0 ± 348.5 versus 2165.0 ± 218.1, = 0.0461) in comparison to that from controls (Figure 4). Respiratory mitochondrial mass of T cells from SLE patients further increased following anti-CD3/28 stimulation (232.70 ± 1.5017.80 versus 152.50 ± 1.50, = 0.0462) (Figure 4(a)). As we know, cell activation would induce increased cellular biogenesis which is ATP-dependent, and mitochondria are the main source of ATP synthesis. However, some studies showed that the mitochondrial function of lymphocytes from SLE patients was defective, characterized by high mitochondrial transmembrane potential (ΔΨm) and low ATP production [13,18]. These results indicated that T cell activation may increase mitochondrial biogenesis in SLE patients, but it did not improve their energy "starvation" state and activated autophagy failed to eliminate the excessive mitochondria in those T cells. As T cells from SLE patients were characterized by increased ΔΨm, we detected the defective mitochondria in them with potential-dependent probe, MDR, in the following experiments.
In these experiments, UC-MSCs were prestained with MDR and then cultured with PBMCs which had been stimulated with anti-CD3/CD28 antibodies. As demonstrated, activated CD3+T cells got much more stained mitochondria than other cells (CD3 negative) in PBMCs (Figure 7(a)) and the effect was cell-to-cell contact-dependent (11.61 ± 1.34 versus 1.23 ± 0.11, = 0.0136 for CD4+T cells) (Figures 7(b) and 7(c)). Further studies using fluorescence microscopy confirmed that UC-MSCs transferred their functional mitochondria to T cells (Figure 7(d)). Interestingly, we could distinguish two populations of CD3+T cells (Figure 7(a)), and CD4+T cells got more mitochondria than CD8+T cells (Figure 7(b)), indicating there was a precedence of mitochondrial transfer for CD4+T cells. In summary, UC-MSCs may improve energy state of T cells via mitochondrial transfer and thus suppressed excessive mitochondrial biogenesis and accumulation, which finally rescued them from activated autophagy and apoptosis.

Functional Mitochondria Were Essential for UC-MSCs to Regulate T Cell Autophagy.
To confirm the essential role of mitochondria in autophagy regulation, we used chloroquine, an autophagy-lysosomal inhibitor, to block late-stage autophagy of UC-MSCs. As expected, we found blockade of autophagic flux caused accumulation of mitochondria in UC-MSCs, indicating impairment to eliminate excessive or dysfunctional mitochondria (145.0 ± 2.65 versus 119.50 ± 8.50, = 0.0383) (Figure 8(a)). Interestingly, it subsequently  abrogated UC-MSCs' regulation on T cell autophagy (Figure 8(b)), which suggested that UC-MSCs could not regulate T cell autophagy without intact mitochondria.

Discussion
In this paper, we showed that activated autophagy increased apoptosis of T cells in SLE patients, and UC-MSCs could regulate autophagy via mitochondrial transfer. Recent studies found that aberrant autophagy played an important role in the development of autoimmune diseases [9,19], especially in SLE. Gros et al. detected more autophagosomes in mice and human lupus T lymphocytes [12]. However, what causes aberrant autophagy in SLE patients is not clear yet. It was reported that autophagy deregulation of SLE T cells was not an immediate consequence of the inflammatory event.
Autophagy of T cells from CBA/J mice could not be activated by LPS [12]. Meanwhile, another study showed that T cells from SLE patients were resistant to induction of autophagy by serum IgG from patients with SLE [11]. Consistent with these studies, we observed higher autophagy level in T cells from SLE patients than that from healthy controls, as demonstrated by LC3-IIB and p62. Additionally, T cells from SLE patients seemed less responsive to anti-CD3/CD28 antibody-induced autophagy activation. Nevertheless, we found a significant increase of LC3-IIB expression in T cells from SLE patients after anti-CD3/CD28 stimulation, following elevated respiratory mitochondrial mass. And LC3-IIB expression markedly decreased after suppression of mitochondrial accumulation. So we supposed that increased autophagy in T cells from SLE patients may be partly due to energy "starvation" [13,20] and excessively accumulated mitochondria. Additionally, inflammatory microenvironment in SLE patients may aggravate the situation. Furthermore, we found overactivated autophagy promoted apoptosis of T cells from SLE patients. There have been lots of studies showing the complex relationship between autophagy and apoptosis [21][22][23][24][25][26]. Generally, autophagy could rescue cells from death when they are under stress such as DNA damage [27], mitochondrial dysfunction [28], or starvation [29]. Prolonged autophagy activation leads to a high turnover rate of proteins and organelles. However, a high rate above the survival threshold may lead to cell death [14]. Previous studies suggested that peripheral T cells from SLE patients had defective mitochondria and were under energy "starvation" [13,20]. In our study, activation of T cells induced mitochondrial biogenesis and autophagy, which seemed to increase apoptosis in contrary. However, inhibition of autophagy not always rescued T cells from apoptosis, as shown by the difference between 3-MA and chloroquine (CQ). 3-MA is a PI3K inhibitor and inhibits the initial process of autophagy, while CQ blocks lysosomal degradation, which is the late-stage of autophagy. So treatment with CQ prevented the contents in autolysosome from recycling and was unable to maintain cellular energy levels. These results further indicated the complicated relationship between autophagy and apoptosis. And additional studies must be done to elucidate the mechanism more clearly. Interestingly, UC-MSCs could rescue T cells from SLE patients by suppressing autophagy, which was mediated by mitochondrial transfer. It has been reported before that mesenchymal stem cells (MSCs) could transfer mitochondria to tissue cells in wound healing [30]. Further studies suggested that the transfer of mitochondria was through tunneling nanotubes (TNT) from MSCs to the injured cells [31] and regulated by Miro1 [32]. It is, however, the first demonstration that they can regulate energy state of T cells from SLE patients in this way, which may reveal a novel mechanism of treating SLE patients with UC-MSCs. As we know, T cells are characteristically activated in patients with SLE, especially in those with high disease activity. Since accumulated defective mitochondria could not match the high energy consumption within T cells, they may thus induce apoptosis, subsequently increasing autoantibody formation [33]. Thus regulation of energy state of T cells could probably be a new way to treat patients with SLE.
As demonstrated, UC-MSCs showed a preference for CD4+T cells in both respiratory mitochondrial mass regulation and mitochondrial transfer. In our study, we found respiratory mitochondrial accumulation was more significant in CD4+T than in CD8+T cells after stimulation. These results indicated that mitochondrial dysfunction may play a more important role in CD4+T cell aberration. However, we need further investigations to confirm this hypothesis. Besides UC-MSCs, recent work identified platelets as potential donors of mitochondria in blood. It is found that activated platelets could release respiratory-competent mitochondria, both within membrane-encapsulated microparticles and as free organelles, which would interact with neutrophils and lead to inflammatory responses [34]. These findings suggested the relationship between mitochondria and inflammation and demonstrated the potential for treating inflammatory diseases through maintaining mitochondrial homeostasis.
Besides T cells, other cells in SLE patients, such as B cells and plasmacytoid dendritic cells [35], are also pathogenic. And it was recently reported that autophagy activation was required for plasmablast development [19]. However, whether they have defective mitochondria as T cells is not clear yet. And if UC-MSCs could regulate their autophagy in the same way still needs to be investigated. Mitochondrial transmembrane potential.