Glycogen is an important reserve polysaccharide from bacteria to human. It is organized in glycogen granules that also contain several proteins involved in their metabolism. Glycogen granules can be mobilized in mammalian lysosomes and yeast vacuoles. They are delivered to these organelles by macroautophagy (hereafter autophagy). However, whether this is a selective or a non-selective process remains a matter of debate. It was proposed to be selective and called \"glycophagy\" (for selective autophagy of glycogen) in the mouse liver. However, the evidence of this selectivity is lacking in other glycogen-rich organs, such as the heart and skeletal muscle, which both are heavily impacted by the aberrant lysosomal accumulation of glycogen in Pompe disease. We recently developed the Komagataella phaffii yeast as a simple model to study the relationship of glycogen and autophagy. Using this model, we showed that cytosolic glycogen granules are delivered to the vacuole by non-selective autophagy, at least during nitrogen starvation. We speculate that this type of autophagy might be responsible for the lysosomal glycogen turnover in non-hepatic mammalian tissues.

Autophagy was initially recognized as a bulk degradation process that randomly sequesters and degrades cytoplasmic material in lysosomes (vacuoles in yeast). In recent years, various types of selective autophagy have been discovered. Glycophagy, the selective autophagy of glycogen granules, is one of them. While autophagy of glycogen is an important contributor to Pompe disease, which is characterized by the lysosomal accumulation of glycogen, its selectivity is still a matter of debate. Here, we developed the Komagataella phaffii yeast as a simple model of glycogen autophagy under nitrogen starvation conditions to address the question of its selectivity. For this, we turned the self-glucosylating initiator of glycogen synthesis, Glg1, which is covalently bound to glycogen, into the Glg1-GFP autophagic reporter. Our results revealed that vacuolar delivery of Glg1-GFP and its processing to free GFP were strictly dependent on autophagic machinery and vacuolar proteolysis. Notably, this process was independent of Atg11, the scaffold protein common for many selective autophagy pathways. Importantly, the non-mutated Glg1-GFP (which synthesizes and marks glycogen) and mutated Glg1Y212F-GFP (which does not synthesize glycogen and is degraded by non-selective autophagy as cytosolic Pgk1-GFP) were equally well delivered to the vacuole and had similar levels of released GFP. Therefore, we concluded that glycogen autophagy is a non-selective process in K. phaffii yeast under nitrogen starvation conditions.

MicroRNA-7a2 (miR-7a2) plays fundamental roles in the female reproductive axis, and estrogen is indispensable for maintaining ovary function. However, the interaction between miR-7a2 and ovarian function is unclear. The present study aimed to determine whether and how miR-7a2 functions in estrogen synthesis. Firstly, the results verified that miR-7a was highly expressed in ovarian granulosa cells. The knockout (KO) of miR-7a2 caused infertility and abnormal ovarian function in mice. Concomitantly, the Cyp19a1 expression and estrogen synthesis were significantly inhibited, which was validated in primary granulosa cells. The mice transplanted with miR-7a2 KO ovaries showed similar results; however, estrogen supplementation reversed infertility. In the in vitro experiment, follicle-stimulating hormone (FSH) significantly improved the expression of miR-7a and Cyp19a1 and the synthesis of estrogen. However, the miR-7a2 KO markedly reversed the function of FSH. Also, FSH upregulated miR-7a by activating the (c-Jun N-terminal kinase) JNK signaling pathway. In addition, Golgi apparatus protein 1 (Glg1) was shown to be the target gene of miR-7a2. These findings indicated that miR-7a2 is essential for ovarian functions with respect to estrogen synthesis through the targeted inhibition of the expression of Glg1 and then promoting Cyp19a1 expression; the physiological process was positively regulated by FSH via the JNK signaling pathway in granulosa cells.

Ewing sarcoma (EwS) is an aggressive cancer displaying an undifferentiated small-round-cell histomorphology that can be easily confused with a broad spectrum of differential diagnoses. Using comparative transcriptomics and immunohistochemistry (IHC), we previously identified BCL11B and GLG1 as potential specific auxiliary IHC markers for EWSR1-FLI1-positive EwS. Herein, we aimed at validating the specificity of both markers in a far larger and independent cohort of EwS (including EWSR1-ERG-positive cases) and differential diagnoses. Furthermore, we evaluated their intra-tumoral expression heterogeneity. Thus, we stained tissue microarrays from 133 molecularly confirmed EwS cases and 320 samples from morphological mimics, as well as a series of patient-derived xenograft (PDX) models for BCL11B, GLG1, and CD99, and systematically assessed the immunoreactivity and optimal cut-offs for each marker. These analyses demonstrated that high BCL11B and/or GLG1 immunoreactivity in CD99-positive cases had a specificity of 97.5% and an accuracy of 87.4% for diagnosing EwS solely by IHC, and that the markers were expressed by EWSR1-ERG-positive EwS. Only little intra-tumoral heterogeneity in immunoreactivity was observed for differential diagnoses. These results indicate that BCL11B and GLG1 may help as specific auxiliary IHC markers in diagnosing EwS in conjunction with CD99, especially if confirmatory molecular diagnostics are not available.

Ewing sarcoma is an undifferentiated small-round-cell sarcoma. Although molecular detection of pathognomonic EWSR1-ETS fusions such as EWSR1-FLI1 enables definitive diagnosis, substantial confusion can arise if molecular diagnostics are unavailable. Diagnosis based on the conventional immunohistochemical marker CD99 is unreliable due to its abundant expression in morphological mimics. To identify novel diagnostic immunohistochemical markers for Ewing sarcoma, we performed comparative expression analyses in 768 tumors representing 21 entities including Ewing-like sarcomas, which confirmed that CIC-DUX4-, BCOR-CCNB3-, EWSR1-NFATc2-, and EWSR1-ETS-translocated sarcomas are distinct entities, and revealed that ATP1A1, BCL11B, and GLG1 constitute specific markers for Ewing sarcoma. Their high expression was validated by immunohistochemistry and proved to depend on EWSR1-FLI1-binding to highly active proximal super-enhancers. Automated cut-off-finding and combination-testing in a tissue-microarray comprising 174 samples demonstrated that detection of high BCL11B and/or GLG1 expression is sufficient to reach 96% specificity for Ewing sarcoma. While 88% of tested Ewing-like sarcomas displayed strong CD99-immunoreactivity, none displayed combined strong BCL11B- and GLG1-immunoreactivity. Collectively, we show that ATP1A1, BCL11B, and GLG1 are EWSR1-FLI1 targets, of which BCL11B and GLG1 offer a fast, simple, and cost-efficient way to diagnose Ewing sarcoma by immunohistochemistry. These markers may significantly reduce the number of misdiagnosed patients, and thus improve patient care.

Microglia are tissue macrophages and mediators of innate immune responses in the brain. The protein-modifying glycan polysialic acid (polySia) is implicated in modulating microglia activity. Cultured murine microglia maintain a pool of Golgi-confined polySia, which is depleted in response to lipopolysaccharide (LPS)-induced activation. Polysialylated neuropilin-2 (polySia-NRP2) contributes to this pool but further polySia protein carriers have remained elusive. Here, we use organotypic brain slice cultures to demonstrate that injury-induced activation of microglia initiates Golgi-confined polySia expression in situ. An unbiased glycoproteomic approach with stem cell-derived microglia identifies E-selectin ligand-1 (ESL-1) as a novel polySia acceptor. Together with polySia-NRP2, polySia-ESL-1 is also detected in primary cultured microglia, in brain slice cultures and in phorbol ester-induced THP-1 macrophages. Induction of stem cell-derived microglia, activated microglia in brain slice cultures and THP-1 macrophages by LPS, but not interleukin-4, causes polySia depletion and, as shown for stem cell-derived microglia, a metalloproteinase-dependent release of polySia-ESL-1 and polySia-NRP2. Moreover, soluble polySia attenuates LPS-induced production of nitric oxide and proinflammatory cytokines. Thus, shedding of polySia-ESL-1 and polySia-NRP2 after LPS-induced activation of microglia and THP-1 macrophages may constitute a mechanism for negative feedback regulation. GLIA 2016 GLIA 2016;64:1314-1330.