Metabolic clogging of mannose triggers dNTP loss and genomic instability in human cancer cells.

Mannose has anticancer activity that inhibits cell proliferation and enhances the efficacy of chemotherapy. How mannose exerts its anticancer activity, however, remains poorly understood. Here, using genetically engineered human cancer cells that permit the precise control of mannose metabolic flux, we demonstrate that the large influx of mannose exceeding its metabolic capacity induced metabolic remodeling, leading to the generation of slow-cycling cells with limited deoxyribonucleoside triphosphates (dNTPs). This metabolic remodeling impaired dormant origin firing required to rescue stalled forks by cisplatin, thus exacerbating replication stress. Importantly, pharmacological inhibition of de novo dNTP biosynthesis was sufficient to retard cell cycle progression, sensitize cells to cisplatin, and inhibit dormant origin firing, suggesting dNTP loss-induced genomic instability as a central mechanism for the anticancer activity of mannose.

In order to grow and divide, cells require a variety of sugars. Breaking down sugars provides energy for cells to proliferate and allows them to make more complex molecules, such as DNA. Although this principle also applies to cancer cells, a specific sugar called mannose not only inhibits cancer cell division but also makes them more sensitive to chemotherapy. These anticancer effects of mannose are particularly strong in cells lacking a protein known as MPI, which breaks down mannose. Evidence from honeybees suggests that a combination of mannose and low levels of MPI leads to a build-up of a modified form of mannose, called mannose-6-phosphate, within cells. As a result, pathways required to release energy from glucose become disrupted, proving lethal to these insects. However, it was not clear whether the same processes were responsible for the anticancer effects of mannose. To investigate, Harada et al. removed the gene that encodes the MPI protein in two types of human cancer cells. The experiments showed that mannose treatment was not lethal to these cells but overall slowed the cell cycle ­ a fundamental process for cell growth and division. More detailed biochemical experiments showed that cancer cells with excess mannose-6-phosphate could not produce the molecules required to make DNA. This prevented them from doubling their DNA ­ a necessary step for cell division ­ and responding to stress caused by chemotherapy. Harada et al. also noticed that cancer cells lacking MPI did not all react to mannose treatment in exactly the same way. Therefore, future work will address these diverse reactions, potentially providing an opportunity to use the mannose pathway to search for new cancer treatments.

IRE1-XBP1 pathway regulates oxidative proinsulin folding in pancreatic ß cells.

In mammalian pancreatic ß cells, the IRE1α-XBP1 pathway is constitutively and highly activated under physiological conditions. To elucidate the precise role of this pathway, we constructed ß cell-specific Ire1α conditional knockout (CKO) mice and established insulinoma cell lines in which Ire1α was deleted using the Cre-loxP system. Ire1α CKO mice showed the typical diabetic phenotype including impaired glycemic control and defects in insulin biosynthesis postnatally at 4-20 weeks. Ire1α deletion in pancreatic ß cells in mice and insulinoma cells resulted in decreased insulin secretion, decreased insulin and proinsulin contents in cells, and decreased oxidative folding of proinsulin along with decreased expression of five protein disulfide isomerases (PDIs): PDI, PDIR, P5, ERp44, and ERp46. Reconstitution of the IRE1α-XBP1 pathway restored the proinsulin and insulin contents, insulin secretion, and expression of the five PDIs, indicating that IRE1α functions as a key regulator of the induction of catalysts for the oxidative folding of proinsulin in pancreatic ß cells.

MPP+ induces necrostatin-1- and ferrostatin-1-sensitive necrotic death of neuronal SH-SY5Y cells.

Regulation of cell death is potentially a powerful treatment modality for intractable diseases such as neurodegenerative diseases. Although there have been many reports about the possible involvement of various types of cell death in neurodegenerative diseases, it is still unclear exactly how neurons die in patients with these diseases, thus treatment strategies based on cell death regulation have not been established yet. To obtain some insight into the mechanisms of cell death involved in neurodegenerative diseases, we studied the effect of 1-methyl-4-phenylpyridinium (MPP+) on the human neuroblastoma cell line SH-SY5Y (a widely used model of Parkinson's disease). We found that MPP+ predominantly induced non-apoptotic death of neuronally differentiated SH-SY5Y cells. This cell death was strongly inhibited by necrostatin-1 (Nec-1), a necroptosis inhibitor, and by an indole-containing compound (3,3'-diindolylmethane: DIM). However, it occurred independently of receptor-interacting serine/threonine-protein kinase 1/3 (RIP1/RIP3), indicating that this form of cell death was not necroptosis. MPP+-induced cell death was also inhibited by several inhibitors of ferroptosis, including ferrostatin-1 (Fer-1). Although MPP+-induced death and ferroptosis shared some features, such as occurrence of lipid peroxidation and inhibition by Fer-1, MPP+-induced death seemed to be distinct from ferroptosis because MPP+-induced death (but not ferroptosis) was inhibited by Nec-1, was independent of p53, and was accompanied by ATP depletion and mitochondrial swelling. Further investigation of MPP+-induced non-apoptotic cell death may be useful for understanding the mechanisms of neuronal loss and for treatment of neurodegenerative diseases such as Parkinson's disease.

Vital staining for cell death identifies Atg9a-dependent necrosis in developmental bone formation in mouse.

Programmed cell death has a crucial role in various biological events, including developmental morphogenesis. Recent evidence indicates that necrosis contributes to programmed cell death in addition to apoptosis, but it is unclear whether necrosis acts as a compensatory mechanism for failure of apoptosis or has an intrinsic role during development. In contrast to apoptosis, there have been no techniques for imaging physiological necrosis in vivo. Here we employ vital staining using propidium iodide to identify cells with plasma membrane disruption (necrotic cells) in mouse embryos. We discover a form of necrosis at the bone surface, which does not occur in embryos with deficiency of the autophagy-related gene Atg9a, although it is unaffected by Atg5 knockout. We also find abnormalities of the bone surface in Atg9a knockout mice, suggesting an important role of Atg9a-dependent necrosis in bone surface formation. These findings suggest that necrosis has an active role in developmental morphogenesis.

Reconstitution and characterization of the unconventional splicing of XBP1u mRNA in vitro.

Upon endoplasmic reticulum (ER) stress, mammalian cells induce the synthesis of a transcriptional activator XBP1s to alleviate the stress. Under unstressed conditions, the messenger RNA (mRNA) for XBP1s exists in the cytosol as an unspliced precursor form, XBP1u mRNA. Thus, its intron must be removed for the synthesis of XBP1s. Upon ER stress, a stress sensor IRE1α cleaves XBP1u mRNA to initiate the unconventional splicing of XBP1u mRNA on the ER membrane. The liberated two exons are ligated to form the mature XBP1s mRNA. However, the mechanism of this splicing is still obscure mainly because the enzyme that joins XBP1s mRNA halves is unknown. Here, we reconstituted the whole splicing reaction of XBP1u mRNA in vitro. Using this assay, we showed that, consistent with the in vivo studies, mammalian cytosol indeed had RNA ligase that could join XBP1s mRNA halves. Further, the cleavage of XBP1u mRNA with IRE1α generated 2', 3'-cyclic phosphate structure at the cleavage site. Interestingly, this phosphate was incorporated into XBP1s mRNA by the enzyme in the cytosol to ligate the two exons. These studies reveal the utility of the assay system and the unique properties of the mammalian cytosolic enzyme that can promote the splicing of XBP1u mRNA.

Mammalian ER stress sensor IRE1ß specifically down-regulates the synthesis of secretory pathway proteins.

Accumulation of unfolded proteins in the endoplasmic reticulum (ER) causes ER stress. The ER stress sensor inositol requiring enzyme-1beta (IRE1ß), which is specifically expressed in intestinal epithelial cells, is thought to be involved in translational repression. However, its mechanism of action is not fully understood. Using a reporter that can evaluate and distinguish between translation efficiency in the cytosol and on the ER membrane, we show here that IRE1ß represses translation on the ER membrane but not in the cytosol, and that this selective repression depends on the RNase activity of IRE1ß.

Cotranslational targeting of XBP1 protein to the membrane promotes cytoplasmic splicing of its own mRNA.

Endoplasmic reticulum (ER) stress triggers the cytoplasmic splicing of XBP1 mRNA by the transmembrane endoribonuclease IRE1alpha, resulting in activation of the unfolded protein response, which maintains ER homeostasis. We show that the unspliced XBP1 (XBP1u) mRNA is localized to the membrane, although its product is neither a secretory nor a membrane protein and is released to the cytosol after splicing. Biochemical and mutagenic analyses demonstrated that membrane localization of XBP1u mRNA required its in-frame translation. An insertional frame-shift mutation greatly diminished both membrane localization and splicing of the XBP1u mRNA. Furthermore, membrane localization was compromised by puromycin treatment and required a hydrophobic region within XBP1u. These data demonstrate that the nascent XBP1u polypeptide recruits its own mRNA to the membrane. This system serves to enhance cytoplasmic splicing and could facilitate a more rapid response to ER stress, and represents a unique way of cotranslational protein targeting coupled to mRNA maturation.

RNase domains determine the functional difference between IRE1alpha and IRE1beta.

Endoplasmic reticulum (ER) stress is associated with the functional disorder of the ER. During conditions of ER stress, cells induce at least two responses to maintain ER function: transcriptional upregulation of ER quality control genes, and translational attenuation of protein synthesis. Induction of ER quality control proteins is mediated by IRE1alpha, which activates the transcription factor XBP1 via an unconventional splicing event, while a partial translational attenuation is mediated by IRE1beta. Here, we show by both in vivo and in vitro analyses that the RNase domain of IRE1 determines the functional specificities of each of these isoforms.