The Autophagy Protein Pacer Positively Regulates the Therapeutic Potential of Mesenchymal Stem Cells in a Mouse Model of DSS-Induced Colitis

Mesenchymal stem cells (MSC) have emerged as a promising tool to treat inflammatory diseases, such as inflammatory bowel disease (IBD), due to their immunoregulatory properties. Frequently, IBD is modeled in mice by using dextran sulfate sodium (DSS)-induced colitis. Recently, the modulation of autophagy in MSC has been suggested as a novel strategy to improve MSC-based immunotherapy. Hence, we investigated a possible role of Pacer, a novel autophagy enhancer, in regulating the immunosuppressive function of MSC in the context of DSS-induced colitis. We found that Pacer is upregulated upon stimulation with the pro-inflammatory cytokine TNFα, the main cytokine released in the inflammatory environment of IBD. By modulating Pacer expression in MSC, we found that Pacer plays an important role in regulating the autophagy pathway in this cell type in response to TNFα stimulation, as well as in regulating the immunosuppressive ability of MSC toward T-cell proliferation. Furthermore, increased expression of Pacer in MSC enhanced their ability to ameliorate the symptoms of DSS-induced colitis in mice. Our results support previous findings that autophagy regulates the therapeutic potential of MSC and suggest that the augmentation of autophagic capacity in MSC by increasing Pacer levels may have therapeutic implications for IBD.


Introduction
Mesenchymal stem cells (MSC) display high self-renewing properties and can differentiate into a variety of cell types, including osteoblasts, chondrocytes and adipocytes, maintaining adult mesenchymal tissues [1][2][3]. MSC are adult progenitor cells and are found in almost all postnatal tissues and organs with similar morphological and functional features, which include the capacity for multi-lineage differentiation as well as immunosuppressive and pro-angiogenic characteristics [4][5][6][7]. Stem cell therapy with MSC shows promise for an increasing number of autoimmune, degenerative and inflammatory diseases, including inflammatory bowel disease (IBDs), graft versus host disease (GvHD), systemic lupus erythematosus (SLE) and rheumatoid arthritis [8,9]. IBD mouse models using dextran sulfate sodium (DSS) to induce colitis are extensively used to study the immunomodulatory properties of MSC [10,11]. DSS-induced colitis mouse models mimic some key immunological and histopathological features of IBDs in humans, hence these models can thus be valuable tools to test the evolving therapeutic strategies in a preclinical setting [12].
Tumor necrosis factor alpha (TNFα) is a pleiotropic cytokine involved in a wide range of pathological processes, including IBD. Increased levels of TNFα have been reported in serum, stool or mucosal biopsy specimens of IBD patients [13][14][15][16]. Furthermore, the successful treatment of IBD in patients with TNFα chimeric monoclonal antibodies (cA2 or infliximab) established a clear association of the role of this cytokine in the pathogenesis of IBD [17]. However, many patients with IBD show primary non-response to this therapy or demonstrate loss of response over time (secondary non-response) [18]. Hence, stem cells have emerged as an attractive novel approach for clinical therapy for patients with IBD [19]. The administration of MSC in both mice and humans has been shown to modulate the function of all immune cells affecting both innate and adaptive immune responses [20]. MSC can secrete several anti-inflammatory molecules, such as prostaglandin E2 (PGE2), nitric oxide (NO), transforming growth factor 1 beta (TGF1β), hepatocyte growth factor (HGF), interleukin-6 (IL6) and interleukin-10 (IL10), to inhibit the proliferation and function of immune cells, including dendritic cells, T and B lymphocytes, thereby dampening the severity of inflammation. The exposure to an inflammatory tissue environment can modulate the immunosuppressive function of MSC [21]. Thus, TNFα and other pro-inflammatory cytokines, such as interferon

Introduction
Mesenchymal stem cells (MSC) display high self-renewing properties and can differentiate into a variety of cell types, including osteoblasts, chondrocytes and adipocytes, maintaining adult mesenchymal tissues [1][2][3]. MSC are adult progenitor cells and are found in almost all postnatal tissues and organs with similar morphological and functional features, which include the capacity for multi-lineage differentiation as well as immunosuppressive and pro-angiogenic characteristics [4][5][6][7]. Stem cell therapy with MSC shows promise for an increasing number of autoimmune, degenerative and inflammatory diseases, including inflammatory bowel disease (IBDs), graft versus host disease (GvHD), systemic lupus erythematosus (SLE) and rheumatoid arthritis [8,9]. IBD mouse models using dextran sulfate sodium (DSS) to induce colitis are extensively used to study the immunomodulatory properties of MSC [10,11]. DSS-induced colitis mouse models mimic some key immunological and histopathological features of IBDs in humans, hence these models can thus be valuable tools to test the evolving therapeutic strategies in a preclinical setting [12].
Tumor necrosis factor alpha (TNFα) is a pleiotropic cytokine involved in a wide range of pathological processes, including IBD. Increased levels of TNFα have been reported in serum, stool or mucosal biopsy specimens of IBD patients [13][14][15][16]. Furthermore, the successful treatment of IBD in patients with TNFα chimeric monoclonal antibodies (cA2 or infliximab) established a clear association of the role of this cytokine in the pathogenesis of IBD [17]. However, many patients with IBD show primary non-response to this therapy or demonstrate loss of response over time (secondary non-response) [18]. Hence, stem cells have emerged as an attractive novel approach for clinical therapy for patients with IBD [19]. The administration of MSC in both mice and humans has been shown to modulate the function of all immune cells affecting both innate and adaptive immune responses [20]. MSC can secrete several anti-inflammatory molecules, such as prostaglandin E2 (PGE2), nitric oxide (NO), transforming growth factor 1 beta (TGF1β), hepatocyte growth factor (HGF), interleukin-6 (IL6) and interleukin-10 (IL10), to inhibit the proliferation and function of immune cells, including dendritic cells, T and B lymphocytes, thereby dampening the severity of inflammation. The exposure to an inflammatory tissue environment can modulate the immunosuppressive function of MSC [21]. Thus, TNFα and other pro-inflammatory cytokines, such as interferon ɣ (INFɣ) and interleukin 1 (IL1), are present in the inflammatory tissues and can potentiate the immunosuppressive function of MSC [22][23][24].
Although a consensus exists about the regulatory function of autophagy in MSC toward their stemness, differentiation capacity and immunosuppressive function, the specific effects of modulating autophagy in MSC have been reported in several studies with opposing findings (reviewed in Ref [25]). Autophagy is a highly conserved cellular process that maintains cellular homeostasis by recycling cytoplasmic materials, such as damaged organelles or misfolded proteins, by delivering them to the lysosome for degradation [26]. Through its recycling function, autophagy also provides energy for cellular renovation and homeostasis [27]. Cellular autophagic capacity can be rapidly increased in response to stress signals, such as starvation, hypoxia, infection and inflammation, primarily as a survival mechanism [28]. In MSC, most studies on autophagy have focused on their roles in differentiation [29,30]; whether autophagy plays a role in the immunosuppressive function of MSC remains unknown. Gao et al. demonstrated that the autophagy inhibitor 3-methyladenine (3-MA) weakens the immunosuppressive function of MSC, whereas autophagy inducer rapamycin enhances this function mediated by TGF1β secretion [31]. In this line, MSC treated with rapamycin aggregate facilitating full-layer cutaneous wound healing and regeneration mediated by vascular endothelial growth factor (VEGF) secretion [32]. On the other hand, MSC depleted of Beclin1, a core subunit of distinct phosphatidylinositol 3 kinases (PI3K) complexes, which mediate multiple steps of the autophagy process, lose their regenerative capacity [32]. Hence, the activation of autophagy may be directly involved in the regulation of the immunosuppressive function

Introduction
Mesenchymal stem cells (MSC) display high self-renewing properties and can differentiate into a variety of cell types, including osteoblasts, chondrocytes and adipocytes, maintaining adult mesenchymal tissues [1][2][3]. MSC are adult progenitor cells and are found in almost all postnatal tissues and organs with similar morphological and functional features, which include the capacity for multi-lineage differentiation as well as immunosuppressive and pro-angiogenic characteristics [4][5][6][7]. Stem cell therapy with MSC shows promise for an increasing number of autoimmune, degenerative and inflammatory diseases, including inflammatory bowel disease (IBDs), graft versus host disease (GvHD), systemic lupus erythematosus (SLE) and rheumatoid arthritis [8,9]. IBD mouse models using dextran sulfate sodium (DSS) to induce colitis are extensively used to study the immunomodulatory properties of MSC [10,11]. DSS-induced colitis mouse models mimic some key immunological and histopathological features of IBDs in humans, hence these models can thus be valuable tools to test the evolving therapeutic strategies in a preclinical setting [12].
Tumor necrosis factor alpha (TNFα) is a pleiotropic cytokine involved in a wide range of pathological processes, including IBD. Increased levels of TNFα have been reported in serum, stool or mucosal biopsy specimens of IBD patients [13][14][15][16]. Furthermore, the successful treatment of IBD in patients with TNFα chimeric monoclonal antibodies (cA2 or infliximab) established a clear association of the role of this cytokine in the pathogenesis of IBD [17]. However, many patients with IBD show primary non-response to this therapy or demonstrate loss of response over time (secondary non-response) [18]. Hence, stem cells have emerged as an attractive novel approach for clinical therapy for patients with IBD [19]. The administration of MSC in both mice and humans has been shown to modulate the function of all immune cells affecting both innate and adaptive immune responses [20]. MSC can secrete several anti-inflammatory molecules, such as prostaglandin E2 (PGE2), nitric oxide (NO), transforming growth factor 1 beta (TGF1β), hepatocyte growth factor (HGF), interleukin-6 (IL6) and interleukin-10 (IL10), to inhibit the proliferation and function of immune cells, including dendritic cells, T and B lymphocytes, thereby dampening the severity of inflammation. The exposure to an inflammatory tissue environment can modulate the immunosuppressive function of MSC [21]. Thus, TNFα and other pro-inflammatory cytokines, such as interferon ɣ (INFɣ) and interleukin 1 (IL1), are present in the inflammatory tissues and can potentiate the immunosuppressive function of MSC [22][23][24].
Although a consensus exists about the regulatory function of autophagy in MSC toward their stemness, differentiation capacity and immunosuppressive function, the specific effects of modulating autophagy in MSC have been reported in several studies with opposing findings (reviewed in Ref [25]). Autophagy is a highly conserved cellular process that maintains cellular homeostasis by recycling cytoplasmic materials, such as damaged organelles or misfolded proteins, by delivering them to the lysosome for degradation [26]. Through its recycling function, autophagy also provides energy for cellular renovation and homeostasis [27]. Cellular autophagic capacity can be rapidly increased in response to stress signals, such as starvation, hypoxia, infection and inflammation, primarily as a survival mechanism [28]. In MSC, most studies on autophagy have focused on their roles in differentiation [29,30]; whether autophagy plays a role in the immunosuppressive function of MSC remains unknown. Gao et al. demonstrated that the autophagy inhibitor 3-methyladenine (3-MA) weakens the immunosuppressive function of MSC, whereas autophagy inducer rapamycin enhances this function mediated by TGF1β secretion [31]. In this line, MSC treated with rapamycin aggregate facilitating full-layer cutaneous wound healing and regeneration mediated by vascular endothelial growth factor (VEGF) secretion [32]. On the other hand, MSC depleted of Beclin1, a core subunit of distinct phosphatidylinositol 3 kinases (PI3K) complexes, which mediate multiple steps of the autophagy process, lose their regenerative capacity [32]. Hence, the activation of autophagy may be directly involved in the regulation of the immunosuppressive function ) and interleukin 1 (IL1), are present in the inflammatory tissues and can potentiate the immunosuppressive function of MSC [22][23][24].
Although a consensus exists about the regulatory function of autophagy in MSC toward their stemness, differentiation capacity and immunosuppressive function, the specific effects of modulating autophagy in MSC have been reported in several studies with opposing findings (reviewed in Ref [25]). Autophagy is a highly conserved cellular process that maintains cellular homeostasis by recycling cytoplasmic materials, such as damaged organelles or misfolded proteins, by delivering them to the lysosome for degradation [26]. Through its recycling function, autophagy also provides energy for cellular renovation and homeostasis [27]. Cellular autophagic capacity can be rapidly increased in response to stress signals, such as starvation, hypoxia, infection and inflammation, primarily as a survival mechanism [28]. In MSC, most studies on autophagy have focused on their roles in differentiation [29,30]; whether autophagy plays a role in the immunosuppressive function of MSC remains unknown. Gao et al. demonstrated that the autophagy inhibitor 3-methyladenine (3-MA) weakens the immunosuppressive function of MSC, whereas autophagy inducer rapamycin enhances this function mediated by TGF1β secretion [31]. In this line, MSC treated with rapamycin aggregate facilitating full-layer cutaneous wound healing and regeneration mediated by vascular endothelial growth factor (VEGF) secretion [32]. On the other hand, MSC depleted of Beclin1, a core subunit of distinct phosphatidylinositol 3 kinases (PI3K) complexes, which mediate multiple steps of the autophagy process, lose their regenerative capacity [32]. Hence, the activation of autophagy may be directly involved in the regulation of the immunosuppressive function of MSC. Recently, we and others described a novel autophagy-related protein called Pacer (protein associated with UVRAG as autophagy enhancer) [33][34][35]. Pacer associates with Beclin1 and positively regulates autophagosome maturation by complex association with UV radiation resistance-associated gene protein (UVRAG) and stimulation of Vps34 kinase activity [33,34,36]. Pacer was shown to be modulated through phosphorylation by the mammalian target of rapamycin complex 1 (mTORC1), which prevents the interaction of Pacer with STX17 and the HOPS complex, resulting in disrupted autophagosome maturation [35]. Dephosphorylation of Pacer promotes its acetylation in the RH domain, which was shown to enhance the interaction with the above-mentioned proteins and promote autophagosome maturation [35]. It has been suggested that Pacer has an important role in autophagy-mediated hepatic lipolysis to alleviate liver inflammation and injury [35]. Furthermore, Pacer has been shown to be involved in amyotrophic lateral sclerosis (ALS) pathogenesis where its loss of function led to increased ALS-associated protein aggregation and neuronal cell death [33]. A role for Pacer in MSC autophagic capacity or immunosuppressive function has not been investigated previously.
Here, we investigated a possible role of Pacer in regulating the immunosuppressive capacity of MSC through its function in the autophagy pathway. We found Pacer to be upregulated upon stimulation with the pro-inflammatory cytokine TNFα. By modulating Pacer expression in MSC, we found that Pacer plays an important role in regulating the autophagy pathway in MSC in response to TNFα stimulation, as well as in regulating the immunosuppressive capacity of MSC toward T-cell proliferation. Furthermore, increased expression of Pacer in MSC enhanced the ability of MSC to ameliorate the symptoms of DSS-induced colitis in mice. Our results suggest that the augmentation of autophagic capacity in MSC by increasing Pacer levels may have therapeutic implications for IBD. Finally, our findings also provide one of the first insights as to how autophagy could modulate the immunoregulatory function of MSC.

Autophagy Assays
To induce autophagy in MSC, cells were treated with rapamycin (200 nM, Enzo Life Sciences, Farmingdale, NY, USA, BML-A275), EBSS (Sigma, E2888) for 4 h and human TNFα (hTNFα) (10 µg/mL) for 2 and 4 h. For autophagy flux experiments, MSC cells were transfected with vectors encoding human Pacer-Flag or empty vector (OriGene) using TransIT-x2 dynamic delivery system (Mirus Bio, Madison, WI, USA, MR.MIR6000) according to manufacturer s instructions or siRNA oligos targeting mouse Pacer or scrambled siRNAs as a control (Dharmacon) transfected with Dharmafect Transfection Reagents (Dharmacon, T-2001-01) using the manufacturer s protocol. To induce autophagy, cells were treated with hTNFα (10 µg/mL) (Sigma, SRP3177) for 30 min, 2 h or 4 h. To inhibit autophagosome-lysosome fusion, cells were pretreated with Bafilomycin A1 for 30 min (0.5 µM, Sigma, B1793-10UG) and then stimulated with hTNFα at the same times described above. Autophagy flux was calculated as described in Ref. [37]: LC3II flux per sample equals LC3 II densitometric values (after normalization to β-Actin) of lysosomal inhibitortreated samples minus lysosomal inhibitors untreated controls. This was performed for each independent experiment, and the 3 N were graphically represented. The same was performed for p62 flux.

DSS-induced Colitis Mouse Model
To model acute colitis, 2.5% dextran sulfate sodium (DSS) (Millipore, Burlington, MA, USA, MP-0216011080) was dissolved in drinking water and was orally administered to female C57BL/6 mice (14-16-weeks old) for 7 days. Murine MSC were intraperitoneally (i.p.) administered into mice on day 3. The mice were divided into four groups: (i) healthy controls not treated with DSS, (ii) mice treated with DSS, (iii) mice treated with DSS and i.p. injected with EV-MSC, and (iv) mice treated with DSS and i.p. injected with hPacer-Cells 2022, 11, 1503 6 of 16 MSC. A total of 1 × 10 6 MSC in a volume of 100 µL phosphate buffer saline (PBS) (Gibco, 10010049) was used for i.p. injection. Weight loss, stool consistency/diarrhea and the presence of rectal bleeding were daily assessed. The Disease Activity Index (DAI) was calculated as previously described [39]. Briefly, the sum of scores for (i) weight loss (0-4), (ii) stool consistency/diarrhea (0-4) and (iii) rectal bleeding (0-4) was calculated. The DAI is classified as 0 (healthy mice) to 12 (severe colitis). Mice were sacrificed at day 14, and the colon length was measured previous to processing colon tissue for biochemical and histological analyses.

Histological Evaluation and Immunohistochemistry (IHC)
The Swiss-roll technique was used to prepare tissues and perform the histological analyses of the mouse intestine. Colon samples were fixed in 4% paraformaldehyde; 5 µmthick sections were stained with hematoxylin and eosin (H&E). Colonic inflammation was graded by a histopathological scoring system. The scores were determined as follows: No evidence of inflammation-Score 2; Low leukocyte infiltration (<10% of section), no structural damage-Score 4; moderate leukocyte infiltration restricted to the mucosal layer (10-25% of section) crypt, elongation, partial loss of goblet cells, bowel wall thickening, no ulcerations-Score 6; severe leukocyte infiltration beyond the mucosal layer (25-50% of section), crypt elongation, bowel wall thickening, superficial ulcerations-Score 8; transmural leukocyte infiltration seen in >50% of section, distorted crypts, marked loss of goblet cells, bowel wall thickening, extensive ulcerations-Score 10 [40]. The histopathological score was normalized by total colon area.

Statistical Analysis
Statistical analysis was performed with GraphPad Prism V7 software. Statistically significant values were determined using one-way ANOVA, two-way ANOVA or Student's t-test.

The Autophagy Enhancer Pacer Is Upregulated in MSC in Response to the Pro-Inflammatory Cytokine TNFα
We and others have recently described the novel autophagy protein Pacer to be part of the autophagic machinery required for responding to a higher demand for autophagic capacity in the context of neurodegenerative and liver disease [33,34]. Autophagy previously has been shown to be involved in the response of MSC to the inflammatory environment they face in various disease contexts, including experimental autoimmune encephalomyelitis (EAE), Alzheimer, SLE, among others [41][42][43]. To determine whether Pacer could be involved in the response of MSC to an inflammatory environment during colitis, we first assessed the levels of Pacer and other autophagy markers by Western blot in the presence of TNFα, since this cytokine plays an important role in the pathogenesis of colitis [44]. We found that the treatment with TNFα led to a significant upregulation of the protein levels of Pacer, the autophagy core component Beclin1 and the autophagy marker LC3II, as well as a significant downregulation of the autophagy substrate p62, similar to treatments with EBSS (starvation) or rapamycin, two conditions generally used to induce autophagy ( Figure 1A,B). These results suggest an activation of the autophagy pathway in MSC while encountering a pro-inflammatory environment mainly defined by TNFα, such as the one found during IBD. Since Pacer has been reported to promote autophagic activity [33,34], we determined whether the presence of Pacer is related to the autophagic response to TNFα by performing autophagic flux assays under TNFα treatment for 0.5, 2 or 4 h, while also inhibiting lysosomal degradation with lysosomal inhibitors ( Figure 1C). To deplete Pacer levels in MSC, we employed siRNA targeting Pacer mRNA (siPacer). As a control, scrambled siRNA (siCtrl) was used. Interestingly, we found that under TNFα stimulation, the levels of endogenous Pacer were notably increased under lysosomal inhibition, indicating its degradation by the lysosome (Figure 1C). On the other hand, we found that a decrease in Pacer expression by a knockdown in TNFα-stimulated MSC resulted in a Cells 2022, 11, 1503 7 of 16 decrease in LC3II levels under lysosomal inhibition, however, with no significant changes in p62 levels ( Figure 1C,D), indicating impaired autophagy flux under these conditions. These results are in line with the reported role of Pacer as an autophagy enhancer [33,34]. Conversely, to investigate whether an increase in Pacer levels could improve autophagic capacity under TNFα stimulation, we overexpressed human Flag-tagged Pacer (hPacer) in MSC and determined the autophagic flux compared to an empty vector as a control (EV) ( Figure 1E). We found that increased Pacer expression enhanced the autophagy activity of MSC, as indicated by an increase in LC3II under lysosomal inhibition, however, with no significant changes in p62 levels ( Figure 1E,F). Together these results suggest that a pro-inflammatory environment results in the activation of autophagy in MSC and that this increase in autophagic capacity can be modulated by alterations in the levels of Pacer.

Pacer Is Required for the Immunosuppressive Function of MSC
To investigate the response of MSC to TNFα, we determined the mRNA levels of several pro-or anti-inflammatory molecules, including, IL6, TGF1β and prostaglandin endoperoxide synthase 2 (PTGS2, also referred to as COX2) (Figure 2A). We found a significant increase in the mRNA levels of IL6, TGF1β and PTGS2 in MSC in response to TNFα treatment ( Figure 2A). PTGS2 is responsible for the synthesis of to the potent immunosuppressor Prostaglandin E2 [45], and a link between Beclin1 and PTGS2 has been reported previously [41]. Since PTGS2 is an important immune modulator in MSC, we confirmed its expression by determining its protein levels by Western blot ( Figure 2B). We found increased levels of this enzyme at 2 h and 4 h of TNFα treatment ( Figure 2B). Interestingly, rapamycin treatment but not starvation conditions (EBSS) also induced increased mRNA levels of IL6, TGF1β and PTGS2, as well as PTGS2 protein levels (Figure 2A,B). These results suggest that in MSC, the autophagy pathway is in part regulated by the levels of TNFα found in the tissue environment and that the autophagic capacity of MSC may influence their secretion of cytokines, hence their immunomodulatory potential.
We hypothesized that the modulation of Pacer may play a role in the immunomodulatory functions of MSC, hence we determined the protein levels of PTGS2 under Pacer loss and gain of function ( Figure 2C,D). We found that the depletion of Pacer in MSC resulted in diminished PTGS2 levels and a lack of PTGS2 upregulation upon TNFα treatment ( Figure 2C), while the overexpression of Pacer resulted in significantly increased PTGS2 levels under non-treated conditions, which were maintained under TNFα treatment ( Figure 2D). Hence, these results, together with our previous results, suggest that the immunomodulatory response of MSC to a TNFα pro-inflammatory environment may be dependent on Pacer expression levels. It has been previously reported that TNFα can enhance the immunosuppressive function of MSC [22]. Thus, we confirmed this observation by performing a splenocyte proliferation assay, where we co-cultured MSC stimulated with or without TNFα together with mouse splenocytes ( Figure S1). To study the potential role of Pacer in regulating the immunosuppressive capacity of MSC, we depleted MSC of Pacer expression and then examined their effects on T-cell proliferation. By using RNA interference targeting Pacer mRNA (siPacer) and the appropriate nonsense control (siCtrl), we achieved a knockdown of Pacer mRNA and protein levels of approximately 50% ( Figure 3A,B). Next, we assessed the immunosuppressive function of these cells. MSC with diminished Pacer levels suppressed the proliferation of activated T cells less efficiently than the MSC treated with siCtrl or untreated MSC ( Figure 3C). To determine whether increased expression of Pacer in MSC would have the inverse effect, we generated MSC that overexpress human Flag-tagged Pacer (hPacer) using lentiviral transduction ( Figure 3D,E) and then examined their effects on T-cell proliferation. MSC transduced with lentivirus carrying a construct for hPacer expression (hPacer-MSC) suppressed the proliferation of activated T cells more efficiently than MSC transduced with lentivirus carrying an empty vector (EV-MSC) ( Figure 3F). Together, these results suggest that Pacer is required for the immunosuppressive capacity of MSC and that increasing Pacer levels may enhance this capacity. expression enhanced the autophagy activity of MSC, as indicated by an increase in LC3II under lysosomal inhibition, however, with no significant changes in p62 levels ( Figure  1E,F). Together these results suggest that a pro-inflammatory environment results in the activation of autophagy in MSC and that this increase in autophagic capacity can be modulated by alterations in the levels of Pacer.  mRNA levels of IL6, TGF1β and PTGS2, as well as PTGS2 protein levels (Figure 2A,B). These results suggest that in MSC, the autophagy pathway is in part regulated by the levels of TNFα found in the tissue environment and that the autophagic capacity of MSC may influence their secretion of cytokines, hence their immunomodulatory potential.  . β -Actin mRNA levels were used as a reference. Student's t-test was performed. Mean and SEM are shown: *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001; n.s., non-significant. (B) PTGS2 protein levels were determined by Western blot analysis (n = 3). β-Actin serves as a loading control. Densitometric quantifications were performed of three independent experiments. Student's t-test was performed. Mean and SEM are shown: **, p ≤ 0.01; ***, p ≤ 0.001; n.s., non-significant. (C) PTGS2 protein levels under Pacer knockdown (siPacer) were determined by Western blot. Cells were treated with hTNFα (10 µg/mL) for 4 h. As a mock control, a scrambled siRNA (siCtrl) oligo was used. Mouse Pacer and PTGS2 levels of three independent experiments were detected (n = 3). β-Actin serves as a loading control. Densitometric quantification of PTGS2 levels normalized to β-Actin levels is shown (n = 3). Student's t-test was performed. Mean and SEM are shown: **, p ≤ 0.01; and ***, p ≤ 0.001; n.s., non-significant. (D) PTGS2 protein levels were determined by Western blot under human Pacer (hPacer) overexpression. Cells were treated with hTNFα (10 µg/mL) for 4 h. As a mock control, an empty vector (EV) construct was used. hPacer and PTGS2 were detected in five independent experiments (n = 5). β-Actin serves as a loading control. Densitometric quantifications of PTGS2 normalized to β-Actin levels is shown (n = 5). Statistical analyses were performed using Student's t-test. Mean and SEM are shown: **, p ≤ 0.01; and ***, p ≤ 0.001; n.s., non-significant.

Increasing Pacer Levels in MSC Improves Their Therapeutic Effect in a Mouse Model of DSS-Induced Colitis
Since increased Pacer levels enhanced the immunosuppressive capacity of MSC in vitro, we tested whether hPacer-MSC have a comparable effect in vivo. To explore this potential effect, we used the DSS-induced acute inflammatory colitis mouse model. Feeding mice for several days with DSS polymers in the drinking water induces an acute colitis characterized by weight loss, bloody diarrhea, rectal bleeding, ulcerations and infiltrations with granulocytes. In our protocol, we used DSS at a concentration of 2.5% (w/v) in the drinking water for 7 days, which induces strong colitis with very low mortality rates [12]. EV-MSC or hPacer-MSC were intraperitoneally injected into the mice on day 3 of the DSS treatment ( Figure S2A). The overexpression of hPacer significantly improved the therapeutic effects of MSC toward intestinal injury, which was assessed by determining the disease activity index ( Figure 4A), which included weight recovery ( Figure S2B). At day 14, the large intestine and cecum were extracted from the peritoneal cavity. The colon length was measured to assess general organ damage. The colon of DSS treated mice was significantly shorter than the colon of healthy control mice, whereas mice treated with DSS and injected with MSC overexpressing hPacer displayed a similar colon length as healthy control mice, which was also improved compared to mice treated with DSS and EV-MSC ( Figure 4B,C). Histological analysis of hematoxylin/eosin-stained colon sections showed that the administration of hPacer-MSC was able to regenerate the damaged intestinal epithelium more efficiently than EV-MSC ( Figure 4D). Furthermore, a lower level of inflammation with scattered infiltrating mononuclear cells (1-2 foci) was observed with hPacer-MSC treatment compared to EV-MSC treatment ( Figure 4D,E). These results show that increased levels of Pacer improve the therapeutic capability of MSC to alleviate DSS-induced inflammatory colon injury.
Cells 2022, 11, x FOR PEER REVIEW 11 of 16 length was measured to assess general organ damage. The colon of DSS treated mice was significantly shorter than the colon of healthy control mice, whereas mice treated with DSS and injected with MSC overexpressing hPacer displayed a similar colon length as healthy control mice, which was also improved compared to mice treated with DSS and EV-MSC ( Figure 4B,C). Histological analysis of hematoxylin/eosin-stained colon sections showed that the administration of hPacer-MSC was able to regenerate the damaged intestinal epithelium more efficiently than EV-MSC ( Figure 4D). Furthermore, a lower level of inflammation with scattered infiltrating mononuclear cells (1-2 foci) was observed with hPacer-MSC treatment compared to EV-MSC treatment ( Figure 4D,E). These results show that increased levels of Pacer improve the therapeutic capability of MSC to alleviate DSSinduced inflammatory colon injury. length was measured to assess general organ damage. The colon of DSS treated mice was significantly shorter than the colon of healthy control mice, whereas mice treated with DSS and injected with MSC overexpressing hPacer displayed a similar colon length as healthy control mice, which was also improved compared to mice treated with DSS and EV-MSC ( Figure 4B,C). Histological analysis of hematoxylin/eosin-stained colon sections showed that the administration of hPacer-MSC was able to regenerate the damaged intestinal epithelium more efficiently than EV-MSC ( Figure 4D). Furthermore, a lower level of inflammation with scattered infiltrating mononuclear cells (1-2 foci) was observed with hPacer-MSC treatment compared to EV-MSC treatment ( Figure 4D,E). These results show that increased levels of Pacer improve the therapeutic capability of MSC to alleviate DSSinduced inflammatory colon injury.

Introduction
Mesenchymal stem cells (MSC) display high self-renewing properties and can differentiate into a variety of cell types, including osteoblasts, chondrocytes and adipocytes, maintaining adult mesenchymal tissues [1][2][3]. MSC are adult progenitor cells and are found in almost all postnatal tissues and organs with similar morphological and functional features, which include the capacity for multi-lineage differentiation as well as immunosuppressive and pro-angiogenic characteristics [4][5][6][7]. Stem cell therapy with MSC shows promise for an increasing number of autoimmune, degenerative and inflammatory diseases, including inflammatory bowel disease (IBDs), graft versus host disease (GvHD), systemic lupus erythematosus (SLE) and rheumatoid arthritis [8,9]. IBD mouse models using dextran sulfate sodium (DSS) to induce colitis are extensively used to study the immunomodulatory properties of MSC [10,11]. DSS-induced colitis mouse models mimic some key immunological and histopathological features of IBDs in humans, hence these models can thus be valuable tools to test the evolving therapeutic strategies in a preclinical setting [12].
Tumor necrosis factor alpha (TNFα) is a pleiotropic cytokine involved in a wide range of pathological processes, including IBD. Increased levels of TNFα have been reported in serum, stool or mucosal biopsy specimens of IBD patients [13][14][15][16]. Furthermore, the successful treatment of IBD in patients with TNFα chimeric monoclonal antibodies (cA2 or infliximab) established a clear association of the role of this cytokine in the pathogenesis of IBD [17]. However, many patients with IBD show primary non-response to this therapy or demonstrate loss of response over time (secondary non-response) [18]. Hence, stem cells have emerged as an attractive novel approach for clinical therapy for patients with IBD [19]. The administration of MSC in both mice and humans has been shown to modulate the function of all immune cells affecting both innate and adaptive immune responses [20]. MSC can secrete several anti-inflammatory molecules, such as prostaglandin E2 (PGE2), nitric oxide (NO), transforming growth factor 1 beta (TGF1β), hepatocyte growth factor (HGF), interleukin-6 (IL6) and interleukin-10 (IL10), to inhibit the proliferation and function of immune cells, including dendritic cells, T and B lymphocytes, thereby dampening the severity of inflammation. The exposure to an inflammatory tissue environment can modulate the immunosuppressive function of MSC [21]. Thus, TNFα and other pro-inflammatory cytokines, such as interferon ɣ (INFɣ) and interleukin 1 (IL1), are present in the inflammatory tissues and can potentiate the immunosuppressive function of MSC [22][23][24].
Although a consensus exists about the regulatory function of autophagy in MSC toward their stemness, differentiation capacity and immunosuppressive function, the specific effects of modulating autophagy in MSC have been reported in several studies with opposing findings (reviewed in Ref [25]). Autophagy is a highly conserved cellular process that maintains cellular homeostasis by recycling cytoplasmic materials, such as damaged organelles or misfolded proteins, by delivering them to the lysosome for degradation [26]. Through its recycling function, autophagy also provides energy for cellular renovation and homeostasis [27]. Cellular autophagic capacity can be rapidly increased in response to stress signals, such as starvation, hypoxia, infection and inflammation, primarily as a survival mechanism [28]. In MSC, most studies on autophagy have focused on their roles in differentiation [29,30]; whether autophagy plays a role in the immunosuppressive function of MSC remains unknown. Gao et al. demonstrated that the autophagy inhibitor 3-methyladenine (3-MA) weakens the immunosuppressive function of MSC, whereas autophagy inducer rapamycin enhances this function mediated by TGF1β secretion [31]. In this line, MSC treated with rapamycin aggregate facilitating full-layer cutaneous wound healing and regeneration mediated by vascular endothelial growth factor (VEGF) secretion [32]. On the other hand, MSC depleted of Beclin1, a core subunit of distinct phosphatidylinositol 3 kinases (PI3K) complexes, which mediate multiple steps of the autophagy process, lose their regenerative capacity [32]. Hence, the activation of autophagy may be directly involved in the regulation of the immunosuppressive function ) and (v) promoting the release of anti-inflammatory factors (IL10, TGF1β) (reviewed in Ref [51]). However, the molecular mechanism associated with the immunomodulatory function of MSC remained poorly understood. Recent mounting evidence suggests that autophagy could play a central role in the ability of MSC to modulate the host immune response. For instance, autophagy is activated in MSC in an inflammatory environment [41,52], and this increase in autophagic capacity was found to improve their immunosuppressive and therapeutic functions [53,54]. Furthermore, Beclin1, a core component of the autophagy machinery, has been shown to be involved in the therapeutic properties of MSC in a murine encephalomyelitis model [41]. In this study, we investigated the potential role of the novel autophagy enhancer Pacer in the immunomodulatory properties of MSC. We found Pacer, Beclin1 and LC3II levels to be upregulated, while p62 levels were downregulated in MSC upon stimulation with the pro-inflammatory cytokine TNFα, one of the main cytokines present in the inflammatory environment of IBD, indicating an increase in the autophagy capacity of MSC under these conditions. Furthermore, Pacer loss-and gain-of-function experiments in combination with autophagy flux assays suggested that Pacer is a possible regulator/modulator of the autophagy machinery under TNFα pro-inflammatory condition, since its depletion in MSC resulted in impaired autophagy flux, while the augmentation of its levels resulted in enhanced autophagy flux. Additionally, a TNFα-mediated pro-inflammatory environment also upregulated the expression levels of immunomodulatory factors in MSC, such as IL6, TGF1β and PTGS2. Interestingly, we found PTGS2 levels to be highly dependent on the expression levels of Pacer, which suggests the Pacer may not only act as a regulator of autophagic capacity in MSC, but it could also be a modulator of the secretion of immunomodulatory factors of MSC, such as prostaglandins. The finding that alterations in Pacer levels can modulate the immunosuppressive capability of MSC further supports this conclusion. Our results show that Pacer loss of function led to impaired MSC control over splenocyte proliferation, while Pacer gain of function enhanced the expected immunosup-pressive function of MSC to control splenocyte proliferation. Strikingly, these enhancing effects of increased Pacer expression in MSC were directly translatable to an acute in vivo mouse model of DSS-induced colitis, where MSC were used as a therapy. The increased expression of Pacer in MSC enhanced the ability of MSC to ameliorate the symptoms of DSSinduced colitis in mice. Our results are further supported by previous findings [31] that autophagy regulates the therapeutic potential of MSC and suggest that the augmentation of autophagic capacity in MSC by increasing Pacer levels may have therapeutic implications for IBD. Moreover, the modulation of MSC autophagy has been proposed as a possible strategy to favor MSC-induced T-cell polarization toward regulatory cells [55]. Thus, MSCs derived from adipose tissue pre-treated with rapamycin were more effective than untreated MSC in suppressing the in vitro expansion of T helper 17 cells and in reducing the mortality and clinical severity of acute graft versus host disease induced in mice [55]. In vivo benefits are associated with a reduction in T helper 17 cells and an increase in regulatory T cells. The potentiation of immunoregulatory MSC function was correlated with the activation of the autophagic machinery, since increased mRNA expression of some autophagy genes, such as ATG5 and LC3, increased protein expression of Beclin1, ATG5, ATG7 and LC3-II, and concomitant suppression of the expression of MTOR and MTOR components (RICTOR and RPTOR) was found [55]. We demonstrate that the enhancement of autophagy in MSC by increasing Pacer levels improves the immunosuppressive effects of MSC toward T-cell proliferation, resulting in augmented therapeutic efficacy in vivo. These findings highlight the importance of autophagy in regulating the immunomodulatory function of MSC in the inflammatory microenvironment and suggest Pacer as a possible mechanistic link. In summary, our findings demonstrate a critical role of Pacer in the dual regulation of the inflammatory environment on the immunosuppressive function of MSC; while proinflammatory cytokines empower MSC to suppress immune responses, these cytokines increase Pacer expression, which in turn enhances autophagic capacity, improving the immunosuppressive function of MSC. Therefore, the modulation of Pacer in MSC may provide a novel strategy to improve MSC-based immunotherapy. Given its potent effects on reducing T-cell responses, it will be valuable to investigate whether the overexpression of Pacer in MSC may augment their therapeutic efficacy in other inflammatory-related diseases.

Patents
Parts of the results presented in this study have been submitted for patent application.