Osteoarthritic chondrocytes undergo a glycolysis-related metabolic switch upon exposure to IL-1b or TNF.

BACKGROUND: Osteoarthritis is an age-related disease that currently faces a lack of symptomatic treatment. Inflammation, which is mainly sustained by pro-inflammatory cytokines such as IL-1b, TNF, and IL-6, plays an important role in osteoarthritis progression. In this context, pro-inflammatory cytokines are widely used to mimic the inflammatory component of osteoarthritis in vitro. However, the therapeutic failures of clinical trials evaluating anti-cytokines drugs highlight the lack of overall understanding of the effects of these cytokines on chondrocytes. METHODS: Here, we generated a comprehensive transcriptomic and proteomic dataset of osteoarthritic chondrocytes treated with these cytokines to describe their pro-inflammatory signature and compare it to the transcriptome of non-osteoarthritic chondrocytes. Then, the dysregulations highlighted at the molecular level were functionally confirmed by real-time cellular metabolic assays. RESULTS: We identified dysregulation of metabolic-related genes in osteoarthritic chondrocytes but not in non-osteoarthritic chondrocytes. A metabolic shift, toward increased glycolysis at the expense of mitochondrial respiration, was specifically confirmed in osteoarthritic chondrocytes treated with IL-1b or TNF. CONCLUSION: These data show a strong and specific association between inflammation and metabolism in osteoarthritic chondrocytes, which was not found in non-osteoarthritic chondrocytes. This indicates that the link between inflammation and metabolic dysregulation may be exacerbated during chondrocyte damage in osteoarthritis. Video Abstract.

Oncostatin M overexpression induces skin inflammation but is not required in the mouse model of imiquimod-induced psoriasis-like inflammation.

Oncostatin M (OSM) has been reported to be overexpressed in psoriasis skin lesions and to exert proinflammatory effects in vitro on human keratinocytes. Here, we report the proinflammatory role of OSM in vivo in a mouse model of skin inflammation induced by intradermal injection of murine OSM-encoding adenovirus (AdOSM) and compare with that induced by IL-6 injection. Here, we show that OSM potently regulates the expression of genes involved in skin inflammation and epidermal differentiation in murine primary keratinocytes. In vivo, intradermal injection of AdOSM in mouse ears provoked robust skin inflammation with epidermal thickening and keratinocyte proliferation, while minimal effect was observed after AdIL-6 injection. OSM overexpression in the skin increased the expression of the S100A8/9 antimicrobial peptides, CXCL3, CCL2, CCL5, CCL20, and Th1/Th2 cytokines, in correlation with neutrophil and macrophage infiltration. In contrast, OSM downregulated the expression of epidermal differentiation genes, such as cytokeratin-10 or filaggrin. Collectively, these results support the proinflammatory role of OSM when it is overexpressed in the skin. However, OSM expression was not required in the murine model of psoriasis induced by topical application of imiquimod, as demonstrated by the inflammatory phenotype of OSM-deficient mice or wild-type mice treated with anti-OSM antibodies.

IL-38 overexpression induces anti-inflammatory effects in mice arthritis models and in human macrophages in vitro.

OBJECTIVES: Interleukin (IL)-38 is a newly characterised cytokine that belongs to the IL-1 family. This cytokine is expressed in the rheumatoid arthritis (RA) synovial tissue and IL-38 deficient mice have exacerbated arthritis. Here, we analysed the effect of IL-38 overexpression in the joints of arthritic mice, in human macrophages and synovial fibroblasts in vitro. METHODS: Articular injections of an adeno-associated virus (AAV) 2/8 encoding IL-38 were performed in collagen-induced arthritis (CIA), K/BxN serum transfer-induced arthritis (STIA) and antigen-induced arthritis (AIA) in mice. The effect of IL-38 overexpression was evaluated through clinical scores, immunohistochemistry, microCT, Luminex and RT-qPCR analysis. THP-1 macrophages were transduced with a lentiviral vector to overexpress IL-38. RESULTS: Clinical inflammatory scores were significantly decreased after AAV IL-38 injection in joints of mice with CIA and STIA, but not AIA. This decrease was accompanied by reduced macrophage infiltration and a decreased expression of Th17 cytokines (IL-17, IL-23, IL-22) and TNFα. However, IL-38 overexpression had no effect on cartilage or bone destruction. In vitro, the THP-1 monocytic cell line expressed less IL-6, TNFα and IL-23 after IL-38 overexpression. Conditioned media from these cells, containing released IL-38, also exert an anti-inflammatory effect on human primary macrophages and synovial fibroblasts from patients with RA. CONCLUSIONS: This study shows for the first time that IL-38 overexpression attenuates the severity of experimental arthritis. IL-38 may exert its anti-inflammatory effects by decreasing the production of proinflammatory cytokines by macrophages and synovial fibroblasts. This effect can lead to the development of novel treatment strategies in arthritis.

IMQ-induced skin inflammation in mice is dependent on IL-1R1 and MyD88 signaling but independent of the NLRP3 inflammasome.

The pathogenesis of inflammatory skin diseases such as psoriasis involves the release of numerous proinflammatory cytokines, including members of the IL-1 family. Here we report overexpression of IL-1α, IL-1ß, and IL-1 receptor antagonist mRNA, associated to expression of IL-23p19, IL-17A, and IL-22 in skin cells, upon topical application of the TLR7 agonist imiquimod (IMQ) in C57BL/6J mice. IMQ-induced skin inflammation was partially reduced in mice deficient for both IL-1α/IL-1ß or for IL-1 receptor type 1 (IL-1R1), but not in IL-1α- or IL-1ß-deficient mice, demonstrating the redundant activity of IL-1α and IL-1ß for skin inflammation. NLRP3 or apoptosis-associated Speck-like protein containing a Caspase recruitment domain-deficient mice had no significant reduction of skin inflammation in response to IMQ treatment, mainly due to the redundancy of IL-1α. However, IMQ-induced skin inflammation was abolished in the absence of MyD88, the adaptor protein shared by IL-1R and TLR signaling pathways. These results are consistent with the TLR7 dependence of IMQ-induced skin inflammation. Thus, IL-1R1 contributes to the IMQ-induced skin inflammation, and disruption of MyD88 signaling completely abrogates this response.

Oncostatin m, an inflammatory cytokine produced by macrophages, supports intramembranous bone healing in a mouse model of tibia injury.

Different macrophage depletion strategies have demonstrated a vital role of macrophages in bone healing, but the underlying molecular mechanisms are poorly understood. Here, with the use of a mouse model of tibia injury, we found that the cytokine oncostatin M [OSM or murine (m)OSM] was overexpressed during the initial inflammatory phase and that depletion of macrophages repressed mOSM expression. In Osm(-/-) mice, by micro-computed tomography and histology we observed a significant reduction in the amount of new intramedullar woven bone formed at the injured site, reduced number of Osterix(+) osteoblastic cells, and reduced expression of the osteoblast markers runt-related transcription factor 2 and alkaline phosphatase. In contrast, osteoclasts were normal throughout the healing period. One day after bone injury, Stat3, the main transcription factor activated by mOSM, was found phosphorylated/activated in endosteal osteoblastic cells located at the hedge of the hematoma. Interestingly, we observed reduced activation of Stat3 in Osm(-/-) mice. In addition, mice deficient in the mOSM receptor (Osmr(-/-)) also had reduced bone formation and osteoblast number within the injury site. These results suggest that mOSM, a product of macrophages, sustains intramembranous bone formation by signaling through Osmr and Stat3, acting on the recruitment, proliferation, and/or osteoblast differentiation of endosteal mesenchymal progenitor cells. Because bone resorption is largely unaltered, OSM could represent a new anabolic treatment for unconsolidated bone fractures.

BYL719, a new α-specific PI3K inhibitor: single administration and in combination with conventional chemotherapy for the treatment of osteosarcoma.

It has been established that disturbances in intracellular signaling pathways play a considerable part in the oncologic process. Phosphatidylinositol-3-kinase (PI3K) has become of key interest in cancer therapy because of its high mutation frequency and/or gain in function of its catalytic subunits in cancer cells. We investigated the therapeutic value of BYL719, a new specific PI3Kα inhibitor that blocks the ATP site, on osteosarcoma and bone cells. The in vitro effects of BYL719 on proliferation, apoptosis, and cell cycle were assessed in human and murine osteosarcoma cell. Its impact on bone cells was determined using human mesenchymal stem cells (hMSC) and human CD14+ osteoclast precursors. Two different murine preclinical models of osteosarcoma were used to analyze the in vivo biological activities of BYL719. BYL719 decreased cell proliferation by blocking cell cycle in G0/G1 phase with no outstanding effects on apoptosis cell death in HOS and MOS-J tumor cells. BYL719 inhibited cell migration and can thus be considered as a cytostatic drug for osteosarcoma. In murine preclinical models of osteosarcoma, BYL719 significantly decreased tumor progression and tumor ectopic bone formation as shown by a decrease of Ki67+ cells and tumor vascularization. To explore the maximum therapeutic potential of BYL719, the drug was studied in combination with conventional chemotherapeutic drugs, revealing promising efficacy with ifosfamide. BYL719 also exhibited dual activities on osteoblast and osteoclast differentiation. Overall, the present work shows that BYL719 is a promising drug in either a single or multidrug approach to curing bone sarcoma.

Oncostatin M induces IL-33 expression in liver endothelial cells in mice and expands ST2+CD4+ lymphocytes.

Interleukin (IL)-33 is crucially involved in liver pathology and drives hepatoprotective functions. However, the regulation of IL-33 by cytokines of the IL-6 family, including oncostatin M (OSM) and IL-6, is not well studied. The aim of the present study was to determine whether OSM mediates regulation of IL-33 expression in liver cells. Intramuscular administration in mice of an adenovirus encoding OSM (AdOSM) leads to increase in expression of OSM in muscles, liver, and serum of AdOSM-infected mice compared with control mice. The increase of circulating OSM markedly regulated mRNA of genes associated with blood vessel biology, chemotaxis, cellular death, induction of cell adhesion molecules, and the alarmin cytokine IL-33 in liver. Steady-state IL-33 mRNA was upregulated by OSM at an early phase (8 h) following AdOSM infection. At the protein level, the expression of IL-33 was significantly induced in liver endothelial cells [liver sinusoidal endothelial cells (LSEC) and vascular endothelial cells] with a peak at 8 days post-AdOSM infection in mice. In addition, we found OSM-stimulated human microvascular endothelial HMEC-1 cells and human LSEC/TRP3 cells showed a significant increase in expression of IL-33 mRNA in a dose-dependent manner in cell culture. The OSM-mediated overexpression of IL-33 was associated with the activation/enrichment of CD4(+)ST2(+) cells in liver of AdOSM-infected mice compared with adenovirus encoding green fluorescent protein-treated control mice. In summary, these data suggest that the cytokine OSM regulates the IL-33 expression in liver endothelial cells in vivo and in HMEC-1/TRP3 cells in vitro and may specifically expand the target CD4(+)ST2(+) cells in liver.

IL-1ß and TNFα promote monocyte viability through the induction of GM-CSF expression by rheumatoid arthritis synovial fibroblasts.

BACKGROUND: Macrophages and synovial fibroblasts (SF) are two major cells implicated in the pathogenesis of rheumatoid arthritis (RA). SF could be a source of cytokines and growth factors driving macrophages survival and activation. Here, we studied the effect of SF on monocyte viability and phenotype. METHODS: SF were isolated from synovial tissue of RA patients and CD14+ cells were isolated from peripheral blood of healthy donors. SF conditioned media were collected after 24 hours of culture with or without stimulation with TNFα or IL-1ß. Macrophages polarisation was studied by flow cytometry. RESULTS: Conditioned medium from SF significantly increased monocytes viability by 60% compared to CD14+ cells cultured in medium alone (P < 0.001). This effect was enhanced using conditioned media from IL-1ß and TNFα stimulated SF. GM-CSF but not M-CSF nor IL34 blocking antibodies was able to significantly decrease monocyte viability by 30% when added to the conditioned media from IL-1ß and TNFα stimulated SF (P < 0.001). Finally, monocyte cultured in presence of SF conditioned media did not exhibit a specific M1 or M2 phenotype. CONCLUSION: Overall, rheumatoid arthritis synovial fibroblasts stimulated with proinflammatory cytokines (IL-1ß and TNFα) promote monocyte viability via GM-CSF but do not induce a specific macrophage polarization.

Biology and physiology of tendon healing.

Tendon disorders affect people of all ages, from elite and recreational athletes and workers to elderly patients. After an acute injury, 3 successive phases are described to achieve healing: an inflammatory phase followed by a proliferative phase, and finally by a remodeling phase. Despite this process, healed tendon fails to recover its original mechanical properties. In this review, we proposed to describe the key factors involved in the process such as cells, transcription factors, extracellular matrix components, cytokines and growth factors and vascularization among others. A better understanding of this healing process could help provide new therapeutic approaches to improve patients' recovery while tendon disorders management remains a medical challenge.

Phase composition of calcium phosphate materials affects bone formation by modulating osteoclastogenesis.

Human mesenchymal stromal cells (hMSCs) seeded on calcium phosphate (CaP) bioceramics are extensively explored in bone tissue engineering and have recently shown effective clinical outcomes. In previous pre-clinical studies, hMSCs-CaP-mediated bone formation was preceded by osteoclastogenesis at the implantation site. The current study evaluates to what extent phase composition of CaPs affects the osteoclast response and ultimately influence bone formation. To this end, four different CaP bioceramics were used, hydroxyapatite (HA), ß-tricalcium phosphate (ß-TCP) and two biphasic composites of HA/ß-TCP ratios of 60/40 and 20/80 respectively, for in vitro osteoclast differentiation and correlation with in vivo osteoclastogenesis and bone formation. All ceramics allowed osteoclast formation in vitro from mouse and human precursors, except for pure HA, which significantly impaired their maturation. Ectopic implantation alongside hMSCs in subcutis sites of nude mice revealed new bone formation at 8 weeks in all conditions with relative amounts for ß-TCP > biphasic CaPs > HA. Surprisingly, while hMSCs were essential for osteoinduction, their survival did not correlate with bone formation. By contrast, the degree of early osteoclastogenesis (2 weeks) seemed to define the extent of subsequent bone formation. Together, our findings suggest that the osteoclastic response could be used as a predictive marker in hMSC-CaP-based bone regeneration and strengthens the need to understand the underlying mechanisms for future biomaterial development. STATEMENT OF SIGNIFICANCE: The combination of mesenchymal stromal cells (MSCs) and calcium phosphate (CaP) materials has demonstrated its safety and efficacy for bone regeneration in clinical trials, despite our insufficient understanding of the underlying biological mechanisms. Osteoclasts were previously suggested as key mediators between the early inflammatory phase following biomaterial implantation and the subsequent bone formation. Here we compared the affinity of osteoclasts for various CaP materials with different ratios of hydroxyapatite to ß-tricalcium phosphate. We found that osteoclast formation, both in vitro and at early stages in vivo, correlates with bone formation when the materials were implanted alongside MSCs in mice. Surprisingly, MSC survival did not correlate with bone formation, suggesting that the number or phenotype of osteoclasts formed was more important.

New chondrosarcoma cell lines and mouse models to study the link between chondrogenesis and chemoresistance.

Chondrosarcomas are cartilage-forming, poorly vascularized tumors. They represent the second malignant primary bone tumor of adults after osteosarcoma, but in contrast to osteosarcoma they are resistant to chemotherapy and radiotherapy, surgical excision remaining the only therapeutic option. Few cell lines and animal models are available, and the mechanisms behind their chemoresistance remain largely unknown. Our goal was to establish new cell lines and animal cancer models from human chondrosarcoma biopsies to study their chemoresistance. Between 2007 and 2012, 10 chondrosarcoma biopsies were collected and used for cell culture and transplantation into nude mice. Only one transplanted biopsy and one injected cell line has engrafted successfully leading to conventional central high-grade chondrosarcoma similar to the original biopsies. In culture, two new stable cell lines were obtained, one from a dedifferentiated and one from a grade III conventional central chondrosarcoma biopsy. Their genetic characterization revealed triploid karyotypes, mutations in IDH1, IDH2, and TP53, deletion in CDKN2A and/or MDM2 amplification. These cell lines expressed mesenchymal membrane markers (CD44, 73, 90, 105) and were able to produce a hyaline cartilaginous matrix when cultured in chondrogenic three-dimensional (3D) pellets. Using a high-throughput quantitative RT-PCR approach, we observed that cell lines cultured in monolayer had lost expression of several genes implicated in cartilage development (COL2A1, COMP, ACAN) but restored their expression in 3D cultures. Chondrosarcoma cells in monolayer were sensitive to several conventional chemotherapeutic agents but became resistant to low doses of mafosfamide or doxorubicin when cultured in 3D pellets, in parallel with an altered nucleic accumulation of the drug. Our results indicate that the cartilaginous matrix produced by chondrosarcoma cells may impair diffusion of several drugs and thus contribute to chemoresistance. Therefore, 3D chondrogenic cell pellets constitute a more relevant model to study chondrosarcoma chemoresistance and may be a valuable alternative to animal experimentations.

Oncostatin M is a growth factor for Ewing sarcoma.

Primary bone tumors, osteosarcomas and chondrosarcomas, derive from mesenchymal stem cells committed into osteoblasts and chondrocytes; in Ewing sarcomas (ESs), the oncogenic fusion protein EWS-FLI1 prevents mesenchymal differentiation and induces neuroectodermic features. Oncostatin M (OSM) is a cytokine from the IL-6 family that modulates proliferation and differentiation in numerous cells. The basis for inhibition versus induction of proliferation by this cytokine is obscure, although MYC was described as a potent molecular switch in OSM signaling. We show herein that, in contrast to osteosarcomas and chondrosarcomas, for which OSM was cytostatic, OSM induced proliferation of ES cell lines. Knockdown experiments demonstrated that growth induction by OSM depends on both types I [leukemia inhibitory factor receptor (LIFR)] and II [OSM receptor (OSMR)] receptors, high STAT3 activation, and induction of MYC to a high expression level. Indeed, ES cell lines, mice xenografts, and patient biopsy specimens poorly expressed LIF, precluding LIFR lysosomal degradation and OSMR transcriptional induction, thus leading to a high LIFR/OSMR ratio. Because other neuroectodermic tumors (ie, glioma, medulloblastoma, and neuroblastoma) had a similar expression profile, the main role of EWS-FLI1 could be through maintenance of stemness and neuroectodermic features, characterized by a low LIF, a high LIFR/OSMR ratio, and high MYC expression. Thus, this study on rare bone malignancies gives valuable insights on more common cancer regulatory mechanisms and could provide new therapeutic opportunities.

Induction of osteogenesis in mesenchymal stem cells by activated monocytes/macrophages depends on oncostatin M signaling.

Bone resorption by osteoclasts and bone formation by osteoblasts are tightly coupled processes implicating factors in TNF, bone morphogenetic protein, and Wnt families. In osteoimmunology, macrophages were described as another critical cell population regulating bone formation by osteoblasts but the coupling factors were not identified. Using a high-throughput approach, we identified here Oncostatin M (OSM), a cytokine of the IL-6 family, as a major coupling factor produced by activated circulating CD14+ or bone marrow CD11b+ monocytes/macrophages that induce osteoblast differentiation and matrix mineralization from human mesenchymal stem cells while inhibiting adipogenesis. Upon activation of toll-like receptors (TLRs) by lipopolysaccharide or endogenous ligands, OSM was produced in classically activated inflammatory M1 and not M2 macrophages, through a cyclooxygenase-2 and prostaglandin-E2 regulatory loop. Stimulation of osteogenesis by activated monocytes/macrophages was prevented using neutralizing antibodies or siRNA to OSM, OSM receptor subunits gp130 and OSMR, or to the downstream transcription factor STAT3. The induced osteoblast differentiation program culminated with enhanced expression of CCAAT-enhancer-binding protein δ, Cbfa1, and alkaline phosphatase. Overexpression of OSM in the tibia of mice has led to new bone apposition with no sign of bone resorption. Two other cytokines have also a potent role in bone formation induced by monocytes/macrophages and activation of TLRs: IL-6 and leukemia inhibitory factor. We propose that during bone inflammation, infection, or injury, the IL-6 family signaling network activated by macrophages and TLR ligands stimulates bone formation that is largely uncoupled from bone resorption and is thus an important target for anabolic bone therapies.

New insights into inflammatory osteoclast precursors as therapeutic targets for rheumatoid arthritis and periodontitis.

Rheumatoid arthritis (RA) and periodontitis are chronic inflammatory diseases leading to increased bone resorption. Preventing this inflammatory bone resorption is a major health challenge. Both diseases share immunopathogenic similarities and a common inflammatory environment. The autoimmune response or periodontal infection stimulates certain immune actors, leading in both cases to chronic inflammation that perpetuates bone resorption. Moreover, RA and periodontitis have a strong epidemiological association that could be explained by periodontal microbial dysbiosis. This dysbiosis is believed to be involved in the initiation of RA via three mechanisms. (i) The dissemination of periodontal pathogens triggers systemic inflammation. (ii) Periodontal pathogens can induce the generation of citrullinated neoepitopes, leading to the generation of anti-citrullinated peptide autoantibodies. (iii) Intracellular danger-associated molecular patterns accelerate local and systemic inflammation. Therefore, periodontal dysbiosis could promote or sustain bone resorption in distant inflamed joints. Interestingly, in inflammatory conditions, the existence of osteoclasts distinct from "classical osteoclasts" has recently been reported. They have proinflammatory origins and functions. Several populations of osteoclast precursors have been described in RA, such as classical monocytes, a dendritic cell subtype, and arthritis-associated osteoclastogenic macrophages. The aim of this review is to synthesize knowledge on osteoclasts and their precursors in inflammatory conditions, especially in RA and periodontitis. Special attention will be given to recent data related to RA that could be of potential value in periodontitis due to the immunopathogenic similarities between the two diseases. Improving our understanding of these pathogenic mechanisms should lead to the identification of new therapeutic targets involved in the pathological inflammatory bone resorption associated with these diseases.

Micromolding-based encapsulation of mesenchymal stromal cells in alginate for intraarticular injection in osteoarthritis.

Osteoarthritis (OA) is an inflammatory joint disease that affects cartilage, subchondral bone, and joint tissues. Undifferentiated Mesenchymal Stromal Cells are a promising therapeutic option for OA due to their ability to release anti-inflammatory, immuno-modulatory, and pro-regenerative factors. They can be embedded in hydrogels to prevent their tissue engraftment and subsequent differentiation. In this study, human adipose stromal cells are successfully encapsulated in alginate microgels via a micromolding method. Microencapsulated cells retain their in vitro metabolic activity and bioactivity and can sense and respond to inflammatory stimuli, including synovial fluids from OA patients. After intra-articular injection in a rabbit model of post-traumatic OA, a single dose of microencapsulated human cells exhibit properties matching those of non-encapsulated cells. At 6 and 12 weeks post-injection, we evidenced a tendency toward a decreased OA severity, an increased expression of aggrecan, and a reduced expression of aggrecanase-generated catabolic neoepitope. Thus, these findings establish the feasibility, safety, and efficacy of injecting cells encapsulated in microgels, opening the door to a long-term follow-up in canine OA patients.

Extensive mannose phosphorylation on leukemia inhibitory factor (LIF) controls its extracellular levels by multiple mechanisms.

In addition to soluble acid hydrolases, many nonlysosomal proteins have been shown to bear mannose 6-phosphate (Man-6-P) residues. Quantification of the extent of mannose phosphorylation and the relevance to physiological function, however, remain poorly defined. In this study, we investigated the mannose phosphorylation status of leukemia inhibitory factor (LIF), a previously identified high affinity ligand for the cation-independent mannose 6-phosphate receptor (CI-MPR), and we analyzed the effects of this modification on its secretion and uptake in cultured cells. When media from LIF-overexpressing cells were fractionated using a CI-MPR affinity column, 35-45% of the total LIF molecules were bound and specifically eluted with free Man-6-P thus confirming LIF as a bona fide Man-6-P-modified protein. Surprisingly, mass spectrometric analysis of LIF glycopeptides enriched on the CI-MPR column revealed that all six N-glycan sites could be Man-6-P-modified. The relative utilization of these sites, however, was not uniform. Analysis of glycan-deleted LIF mutants demonstrated that loss of glycans bearing the majority of Man-6-P residues leads to higher steady-state levels of secreted LIF. Using mouse embryonic stem cells, we showed that the mannose phosphorylation of LIF mediates its internalization thereby reducing extracellular levels and stimulating embryonic stem cell differentiation. Finally, immunofluorescence experiments indicate that LIF is targeted directly to lysosomes following its biosynthesis, providing another mechanism whereby mannose phosphorylation serves to control extracellular levels of LIF. Failure to modify LIF in the context of mucolipidosis II and its subsequent accumulation in the extracellular space may have important implications for disease pathogenesis.

Temporal deep learning framework for retinopathy prediction in patients with type 1 diabetes.

The adoption of electronic health records in hospitals has ensured the availability of large datasets that can be used to predict medical complications. The trajectories of patients in real-world settings are highly variable, making longitudinal data modeling challenging. In recent years, significant progress has been made in the study of deep learning models applied to time series; however, the application of these models to irregular medical time series (IMTS) remains limited. To address this issue, we developed a generic deep-learning-based framework for modeling IMTS that facilitates the comparative studies of sequential neural networks (transformers and long short-term memory) and irregular time representation techniques. A validation study to predict retinopathy complications was conducted on 1207 patients with type 1 diabetes in a French database using their historical glycosylated hemoglobin measurements, without any data aggregation or imputation. The transformer-based model combined with the soft one-hot representation of time gaps achieved the highest score: an area under the receiver operating characteristic curve of 88.65%, specificity of 85.56%, sensitivity of 83.33% and an improvement of 11.7% over the same architecture without time information. This is the first attempt to predict retinopathy complications in patients with type 1 diabetes using deep learning and longitudinal data collected from patient visits. This study highlighted the significance of modeling time gaps between medical records to improve prediction performance and the utility of a generic framework for conducting extensive comparative studies.

IL-34 deficiency impairs FOXP3+ Treg function in a model of autoimmune colitis and decreases immune tolerance homeostasis.

BACKGROUND: Immune homeostasis requires fully functional Tregs with a stable phenotype to control autoimmunity. Although IL-34 is a cytokine first described as mainly involved in monocyte cell survival and differentiation, we recently described its expression by CD8+ Tregs in a rat model of transplantation tolerance and by activated FOXP3+ CD4+ and CD8+ Tregs in human healthy individuals. However, its role in autoimmunity and potential in human diseases remains to be determined. METHODS: We generated Il34-/- rats and using both Il34-/- rats and mice, we investigated their phenotype under inflammatory conditions. Using Il34-/- rats, we further analyzed the impact of the absence of expression of IL-34 for CD4+ Tregs suppressive function. We investigated the potential of IL-34 in human disease to prevent xenogeneic GVHD and human skin allograft rejection in immune humanized immunodeficient NSG mice. Finally, taking advantage of a biocollection, we investigated the correlation between presence of IL-34 in the serum and kidney transplant rejection. RESULTS: Here we report that the absence of expression of IL-34 in Il34-/- rats and mice leads to an unstable immune phenotype, with production of multiple auto-antibodies, exacerbated under inflammatory conditions with increased susceptibility to DSS- and TNBS-colitis in Il34-/- animals. Moreover, we revealed the striking inability of Il34-/- CD4+ Tregs to protect Il2rg-/- rats from a wasting disease induced by transfer of pathogenic cells, in contrast to Il34+/+ CD4+ Tregs. We also showed that IL-34 treatment delayed EAE in mice as well as GVHD and human skin allograft rejection in immune humanized immunodeficient NSG mice. Finally, we show that presence of IL-34 in the serum is associated with a longer rejection-free period in kidney transplanted patients. CONCLUSION: Altogether, our data emphasize on the crucial necessity of IL-34 for immune homeostasis and for CD4+ Tregs suppressive function. Our data also shows the therapeutic potential of IL-34 in human transplantation and auto-immunity. HIGHLIGHTS: -Absence of expression of IL-34 in Il34-/- rats and mice leads to an unstable immune phenotype, with a production of multiple auto-antibodies and exacerbated immune pathology under inflammatory conditions. -Il34-/- CD4+ Tregs are unable to protect Il2rg-/- rats from colitis induced by transfer of pathogenic cells. -IL-34 treatment delayed EAE in mice, as well as acute GVHD and human skin allograft rejection in immune-humanized immunodeficient NSG mice.

Direct anti-cancer effect of oncostatin M on chondrosarcoma.

The cytokine Oncostatin M (OSM) is cytostatic, pro-apoptotic and induces differentiation of osteosarcoma cells into osteocytes, suggesting new adjuvant treatment for these bone-forming sarcomas. However, OSM systemic over-expression could lead to adverse side effects such as generalized inflammation, neoangiogenesis and osteolysis. We determine here the effect of OSM on chondrosarcoma, another primary bone sarcoma characterized by the production of cartilage matrix and altered bone remodelling. Chondrosarcomas are resistant to conventional chemotherapy and radiotherapy, and wide surgical excision remains the only available treatment. We found that OSM blocked the cell cycle in four of five chondrosarcoma cell lines, independently of p53 and presumably through the JAK3/STAT1 pathway. In two tested cell lines, OSM induced a hypertrophic chondrocyte differentiation, with an induced Cbfa1/SOX9 ratio and induced Coll10, matrix metalloproteinase 13 (MMP13) and RANKL expression. Adenoviral gene transfer of OSM (AdOSM) in the Swarm rat chondrosarcoma (SRC) model indicated that local intra-tumoral OSM over-expression reduces chondrosarcoma development not only with reduced tumor proliferation and enhanced apoptosis but also with enhanced RANKL expression, osteoclast formation and reduced bone volumes. Flu-like symptoms were induced by the AdOSM, but there was no effect on tumor angiogenesis. Therefore, OSM could be considered as a new adjuvant anti-cancer agent for chondrosarcomas. A local application of this cytokine is presumably needed to overcome the poor vascularization of these tumors and to limit the deleterious effect on other tissues. Its side effect on bone remodeling could be managed with anti-resorption agents, thus offering potential new lines of therapeutic interventions.