A soluble endoplasmic reticulum factor as regenerative therapy for Wolfram syndrome

Endoplasmic reticulum (ER) stress-mediated cell death is an emerging target for human chronic disorders, including neurodegeneration and diabetes. However, there is currently no treatment for preventing ER stress-mediated cell death. Here, we show that mesencephalic astrocyte-derived neurotrophic factor (MANF), a neurotrophic factor secreted from ER stressed cells, prevents ER stress-mediated β cell death and enhances β cell proliferation in cell and mouse models of Wolfram syndrome, a prototype of ER disorders. Our results indicate that molecular pathways regulated by MANF are promising therapeutic targets for regenerative therapy of ER stress-related disorders, including diabetes, retinal degeneration, neurodegeneration, and Wolfram syndrome.


Introduction
Growing evidence indicates that endoplasmic reticulum (ER) stress plays a critical role in β cell death in type 1 and type 2 diabetes, as well as in neurodegenerative disorders, including Parkinson's disease and amyotrophic lateral sclerosis [1][2][3][4][5]. Despite the underlying importance of ER stress in β cell death, there is currently no diabetes treatment targeting the ER due to the complex nature of type 1 and type 2 diabetes. Our strategy for overcoming this challenge is to focus on a monogenic form of diabetes, Wolfram syndrome. Wolfram syndrome is a rare disease characterized by juvenile-onset diabetes mellitus, optic nerve atrophy, and neurodegeneration [6,7]. As this syndrome is caused by mutations in the WFS1 gene which is involved in ER calcium homeostasis and ER stress-mediated cell death, it is ideal for testing potential new treatments targeting the ER [8][9][10][11][12][13][14].
Mesencephalic astrocyte-derived neurotrophic factor (MANF) is a trophic factor whose expression and secretion is enhanced by ER stress and ER calcium depletion [15][16][17][18]. It has been demonstrated that MANF plays a critical role in the survival of ER stressed β cells and neurons [19,20], raising the possibility that MANF-based treatment can be beneficial for patients suffering from ER stressrelated disorders, including Wolfram syndrome. Here we show that MANF-based treatment prevents β cell death and enhances β cell proliferation in cell and mouse models of Wolfram syndrome. Our results indicate that molecular pathways regulated by MANF are promising drug targets for ER stress-related disorders, including β cell death in diabetes and Wolfram syndrome.

Animal experiments
Wfs1 β cell-specific knockout (βWfs1 (−/−) ) mice were generated by breeding the Cre recombinase driven by rat insulin promoter (Rip2-Cre) transgenic mice (originally from Dr Pedro Herrera) with Wfs1 floxed mice [21]. All animal experiments were performed according to procedures approved by the Institutional Animal Care and Use Committee at the Washington University School of Medicine (A-3381-01).

Immunoblot analysis
INS-1 832/13 cells were washed in cold PBS and lysed with M-PER reagent (Thermo Fisher Scientific) containing Complete™ protease inhibitor cocktail (MilliporeSigma). The equivalent amounts of cell lysates were resolved by SDS-PAGE using 4-20% Mini-PROTEAN® TGX™ Precast Protein Gels (Bio-Rad Laboratories, Hercules, CA) and blotted onto Immobilon-P PVDF membrane (0.45 µm) (MilliporeSigma). The following primary antibodies were used for detecting the protein of interest; WFS1 antibody (Proteintech, Rosemont, IL), cleaved caspase-3, GAPDH, alpha-tubulin and beta-actin antibody (Cell Signaling Technology, Danvers, MA), and anti-MANF antibody (Abnova, Taipei City, Taiwan) at 1:1000 dilution. The secondary antibodies conjugated to horseradish peroxidase were obtained from Cell Signaling Technology. The detection was performed by enhanced chemiluminescenceselect (GE Healthcare Bio-Sciences, Pittsburgh, PA). Fiji/ ImageJ was used for the quantification of immunoblot.

Primary islet culture
Mouse primary islets were taken from βWfs1 (−/−) mice. The mice were anesthetized, and pancreata were infused with 5 ml of 0.45 mg/ml collagenase type V (MilliporeSigma) in Hank's balanced salt solution without Ca 2+ (Thermo Fisher Scientific). After surgical removal, pancreata were incubated for 12 min at 37°C, and then hand-shaken for 2 min. Undigested acinar tissue was removed by using a 70-μm cell strainer and recovered tissues were washed twice with ice-cold Hanks' balanced salt solution followed by centrifugation at 1100 rpm for 1 min. Islets were handpicked and preincubated in RPMI 1640 medium containing 10% FBS and antibiotics before experimentation. Islets of equal size were handpicked to generate 3-5 technical replicates for all experiments. Very large and very small islets were excluded. The results were obtained from at least three independent experiments.

Insulin secretion assay
Primary mouse islets or INS-1 832/13 were cultured for 24 h and batches of ten islets were handpicked on the day of the experiment. Mouse islets or INS-1 832/13 were starved for 1 h in Krebs-Ringer bicarbonate-HEPES buffer (129 mM NaCl, 5 mM NaHCO 3 , 4.8 mM KCl, 1.2 mM KH 2 PO 4 , 1.2 mM MgSO 4 , 10 mM HEPES, and 1 mM CaCl 2 at pH 7.4) containing 0.1% bovine serum albumin (KRBH/ BSA). KRBH/BSA was supplemented with 2.8 mM glucose and then stimulated for 1 h at 37°C in KRBH/BSA containing basal 5.5 mM or stimulatory 16.7 mM glucose. At the end of each incubation, supernatants were collected to measure insulin release, and cellular insulin contents were determined by acid-ethanol extraction followed by ELISA Rat/Mouse Insulin kit (MilliporeSigma).

Cell proliferation
The islets isolated from humans donor or βWfs1 (−/−) mice were dissociated by incubation with 0.25% trypsin-EDTA (Thermo Fisher Scientific) at 37°C for 5 min and treated with MANF peptide (R&D Systems, Minneapolis, MN) 5 µg/ml for 5 days. Two-thirds of the medium were changed daily to fresh medium with MANF peptide. To monitor the cell proliferation rate, the BrdU cell proliferation assay kit (Cell Signaling Technology) was used following the manufacturer's instruction.

Immunostaining
Pancreatic tissue sections were fixed, rehydrated and permeabilized with 0.1% Triton X-100 for 2 min. The sections were washed with 0.1% Tween-20 PBS (PBS-T) containing Image-It FX signal enhancer (Thermo Fisher Scientific) for 1 h and incubated with primary antibodies overnight at 4°C [guinea pig anti-insulin antibody (1:100, Thermo Fisher Scientific), MANF (1:100, Abnova), and Ki67 (1:100, Cell Signaling Technology)]. The tissue sections were washed three times in PBS-T and incubated with secondary antibodies for 1 h at room temperature. Images were obtained with a Zeiss LSM 5 PASCAL confocal microscope with LSM Image software.

Measurement of β-cell mass
For measurement of β-cell mass, every 40th pancreatic section was immunostained with guinea pig anti-insulin antibody (1:100, Thermo Fisher Scientific) and counterstained with hematoxylin. The β-cell mass for each mouse was quantified using Image Pro Plus software (Media Cybernetics, Rockville, MD) by obtaining the fraction of the cross-sectional area of pancreatic tissue (exocrine and endocrine) positive for insulin staining, and then multiplying this by the pancreatic weight.

Measurement of apoptosis through TUNEL assay
Apoptotic cells were detected using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) method as per the manufacturer's protocol (Milli-poreSigma). For the determination of apoptosis, all β-cells per pancreatic sections (five sections per animal) were analyzed to count the total number of TUNEL-positive βcells. An average of 150 islets was counted per animal and the percentage of TUNEL-positive cells was quantitated.

In vivo administration of AAV vectors
The methods for AAV production are described in Supplemental Information. AAV was produced in collaboration with the Hope Center Viral Vectors Core at Washington University. Male (n = 3, 2-3 months of age) and female (n = 4, 3-4 months of age) βWfs1 (−/−) mice received intraperitoneal injections of AAV9-CBA-IRES-GFP or AAV9-CBA-MANF-IRES-GFP at a final dose of 1 × 10 13 viral genome particles diluted in saline per mouse. After 4 weeks of AAV administration, the pancreata were harvested. Dissected pancreas pieces were fixed in 4% formalin. Formalin-fixed paraffin-embedded sections were deparaffinized and rehydrated. To estimate the β-cell replication rate, pancreatic sections were immunostained with anti-insulin and anti-Ki-67 antibody, a marker for cellular proliferation. Overall, 1500-3000 β cells were counted in each animal.

Data analysis
The values are expressed as mean ± SEM. All the statistical analysis was carried out with Prism 8 (ver 8.0.2). Comparisons among the group were done by Student's t test. Multiple comparisons were performed by ANOVA followed by Tukey's test. P < 0.05 was considered statistically significant.

MANF confers protection against cell death induced by ER calcium depletion
We have recently shown that various β cell perturbants, including the loss of function of Wolfram syndrome 1 (WFS1) gene, induce ER calcium depletion and ER stress, leading to β cell death [10,22]. It has been recently reported that loss of MANF in vivo leads to β cell death with ER stress elevation [20]. These considerations prompted us to monitor MANF expression levels in β cells under stressed conditions. Although Manf mRNA expression was not changed by Wfs1 deficiency (Fig. S1), thapsigargin, which is a well-established ER calcium depletion inducer, increased Manf mRNA expression and MANF protein Fig. 2 Effect of MANF on glucose-stimulated insulin secretion. a Doxycycline-inducible shRNA directed against Wfs1 (INS-1 DOX-shWfs1) cells were treated with or without MANF peptide (5 µg/ml) for 24 h, and then treated with doxycycline (DOX). Insulin release was measured at basal (5.5 mM) glucose and stimulatory (16.7 mM) glucose conditions (n = 3, not significant). b Cellular insulin contents were measured after the 24 h pretreatment with MANF peptide (5 µg/ml) followed by DOX treatment (n = 3, not significant). c Glucose-stimulated insulin secretion on control (Ctrl) and MANF overexpressed INS-1 832/ 13 cells (MANF-OE). Insulin release was measured at 5.5 and 16.7 mM glucose conditions (n = 3, not significant). d Primary islets isolated from wild type (WT) and β cell-specific Wfs1 knockout mice (βWfs1 (−/−) ) were pretreated with MANF peptide (5 µg/ml) for 24 h. Insulin release was measured at 5.5 mM and 16.7 mM glucose (n = 3, not significant). secretion in INS-1 832/13 cells (Fig. 1a, b). A smaller band of extracellular MANF corresponds to an isoform lacking RTDL domain which is prone to be secreted, and a larger band corresponds to an isoform containing RTDL domain which is glycosylated [15] (https://www.ncbi.nlm.nih.gov/ protein/NP_001101653.1,XP_006243837.1). Intracellular fraction only contains an isoform with the C-terminal RTDL domain [23]. While Manf knockout INS-1 832/13 cells were more sensitive to ER stress-induced cell death (Fig. 1c), recombinant MANF peptide pretreatment reduced cell death in INS-1 832/13 cells treated with thapsigargin (Fig. 1d, e). Furthermore, mRNA expression level of tribbles pseudokinase 3 (Trb3), which is an ER stress-inducible gene, was significantly suppressed in INS-1 832/13 cells stably overexpressing MANF (MANF-OE) (Fig. 1f). Trb3 is a proapoptotic component of ER stress signaling [24][25][26], suggesting that MANF might suppress the proapoptotic arm of ER stress signaling in those models.

Effect of MANF on insulin secretion
Since the loss of MANF in vivo can lead to β cell dysfunction, we studied the relationship between MANF and insulin secretion. We created INS-1 832/13 cells in which Wfs1 expression can be suppressed by doxycyclineinducible shRNA directed against Wfs1 (INS-1 DOX-shWfs1) [9]. Glucose-stimulated insulin secretion (GSIS) assays were performed in INS-1 DOX-shWfs1 cells, MANF-OE INS-1 832/13, and primary mouse islets isolated from β cell-specific Wfs1 knockout (βWfs1 (−/−) ) mice treated with recombinant MANF peptide. As a consequence, MANF treatment or overexpression did not affect GSIS in those models (Fig. 2a-d).

MANF activates proliferation of human primary islets
The fact that the suppression of ER stress can lead to β cell proliferation raised the possibility that MANF treatment might activate β cell proliferation [22,23]. To test this idea, human primary islets were treated with recombinant MANF peptide and then their proliferation rates were assessed by the BrdU assay. Consequently, MANF treatment significantly induced the proliferation of human primary islets derived from two out of six donors (Fig. 3 and Supplementary Table).

MANF-based treatment for Wolfram syndrome
We have previously shown that ER calcium depletion, followed by ER stress-mediated cell death, plays a role in the pathogenesis of Wolfram syndrome [10,22,27], which prompted us to consider the possibility that MANF-based treatment could prevent β cell death and activate β cell proliferation in Wolfram syndrome. Cell death induced by Wfs1 knockdown in INS-1 DOX-shWfs1 cells was prevented by recombinant MANF peptide treatment shown as cleaved caspase-3 protein and caspase-3/7 activity reduction (Fig. 4a, b). The proliferation of primary islets from βWfs1 (−/−) mice, which is a mouse model of Wolfram syndrome [21], was also enhanced by MANF treatment (Fig. 4c). Moreover, MANF treatment suppressed the expression of proapoptotic ER stress markers (Chop and Trb3) in INS-1 DOX-shWfs1 cells (Fig. 4d) and MANF overexpression improved the viability of Wfs1 knockout INS-1 832/13 cells (Fig. S2).

Discussion
Wolfram syndrome is characterized by juvenile-onset diabetes, optic nerve atrophy and, neurodegeneration due to ER stress-mediated cell death [6,28], and has been established as a prototype of ER stress disease [8, 9, 11-14, 21, 29]. Since there is no treatment that can stop or even slow the progression of this syndrome currently, developing the novel treatment has been an urgent task.
Increasing evidence indicates that MANF possesses regenerative and cytoprotective effects. In the mouse pancreas, MANF overexpression was found to induce the proliferation of pancreatic β cells [20]. Systematic MANF overexpression or recombinant MANF peptide delivery protects the liver of old mice from inflammation and hepatocyte apoptosis [30]. Notably, recombinant human MANF peptide protects human β cells from cytokineinduced ER stress and cell death, and induces β cells proliferation [31]. In this study, we show that MANF treatment activates the proliferation of β cells in human islets and prevents ER stress-mediated β cell death and enhances β cell proliferation in cell and mouse models of Wolfram syndrome. These results broaden the possibility of developing the new treatments for Wolfram syndrome using adeno-associated virus expressing MANF or recombinant MANF peptide. To elucidate the efficacy of MANF treatment, further experiments using the other Wolfram syndrome model mice, or β cells which are differentiated from Wolfram syndrome patient-derived iPSCs would be required [12]. On the other hand, MANF treatment did not change insulin secretion and insulin content in INS-1 832/ 13 cells. These results are in line with the previous report using EndoC-βH1 cells [31]. Moreover, even though MANF overexpression activated the βWfs1 (−/−) mice β cell proliferation, the β cell mass of these mice was not changed. A longer overexpression might be needed to study the effect of MANF on the β cell mass.
MANF was originally isolated from astrocytes as a novel neurotrophic factor [15]. It has been reported that MANF regulates the NF-kB signaling pathway, which is considered to be activated through their receptors [31,32]. However, receptors for MANF have not been identified. Further studies are required to identify these receptors and their signaling pathway in order to develop treatments based on small molecules that act as MANF receptor agonists.
Our results are also relevant to other diseases related to ER stress. Genetic, clinical, and experimental evidence indicates that ER stress-mediated cell death is an important pathogenic component in human chronic disorders, including type 1 and type 2 diabetes, retinal degeneration, Parkinson's disease, amyotrophic lateral sclerosis, inflammatory bowel disease, and multiple sclerosis [3,[33][34][35][36][37][38][39]. It has been reported that plasma MANF protein levels decline with age in flies, mice, and human [30]. In contrast, circulating MANF levels are known to increase in children with type 1 diabetes as compared with control subjects [40]. ER stress in β cells has been linked to autoimmunity and cytokine-mediated β cell death during the onset and progression of type 1 diabetes [41][42][43][44][45][46][47]. Thus, increased MANF levels in patients with type 1 diabetes may be an adaptive response to ER stress in β cells. MANF mutations have been reported in a patient with type 2 diabetes [48]. In such disorders, MANF-based therapy may suppress ER stressmediated cell death and delay the progression of the disease.
International Registry and Clinical Study, Research Clinic, and Clinical Trials for their time and efforts.
Author contributions JM, SM, and FU participated in study conception and design. JM, SM, TY, SL, KK, DA, and CMB participated in data acquisition. JM, SM, SL, KK, DA, and FU participated in data analysis and interpretation. JM, SM, CMB, and FU participated in manuscript writing.

Compliance with ethical standards
Conflict of interest FU and KK are inventors of US Patent 9,891,231 B2 entitled "SOLUBLE MANF IN PANCREATIC BETA CELL DISORDERS." FU and SL are inventors of US 10,441,574, B2 entitled "TREATMENT FOR WOLFRAM SYNDROME AND OTHER ER STRESS DISORDERS." FU received research funding from Eli Lilly, Ono Pharmaceuticals, and Amarantus BioScience for the development of MANF-based regenerative therapy for Wolfram syndrome, optic nerve atrophy, and diabetes. FU received chemical compounds from Amylyx Pharmaceuticals, Mitochon Pharmaceuticals, Aetas Pharma, and National Center for Advancing Translational Sciences for the development of small molecule-based therapies for ER stress-related disorders, including Wolfram syndrome. The other authors declare no conflict of interest.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.