The Interactive Fly
Zygotically transcribed genes

Unfolding Protein Response - a homeostatic signaling pathway that is activated by excessive protein misfolding in the endoplasmic reticulum


  • Review: Ryoo, H. D. (2015). Drosophila as a model for unfolded protein response research. BMB Rep 48: 445-453. PubMed ID: 25999177
  • Drosophila as a Model for Human Diseases: ER stress and Unfolded protein response
  • Drosophila melanogaster Activating Transcription Factor 4 regulates glycolysis during endoplasmic reticulum stress
  • A modified UPR stress sensing system reveals a novel tissue distribution of IRE1/XBP1 activity during normal Drosophila development
  • Changes in Drosophila mitochondrial proteins following chaperone-mediated lifespan extension confirm a role of Hsp22 in mitochondrial UPR and reveal a mitochondrial localization for cathepsin D
  • Exploring the Conserved role of MANF in the unfolded protein response in Drosophila melanogaster
  • Reduced sleep during social isolation leads to cellular stress and induction of the Unfolded Protein Response (UPR)
  • EDEM function in ERAD protects against chronic ER Proteinopathy and age-related physiological decline in Drosophila

    Cryptocephal/ATF4 - mediates a shift from a metabolism based on oxidative phosphorylation to one more heavily reliant on glycolysis,
    reminiscent of aerobic glycolysis or the Warburg effect observed in cancer and other proliferative cells

    cap'n'collar
    transcription factor - basic leucine zipper - involved in the segmentation of the head effecting both labral and mandibular structures -
    vertebrate Nrf1, Nrf2, and Nrf3, the Caenorhabditis elegans SKN-1, and Drosophila CncC mediate adaptive responses to cellular stress

    Inositol-requiring enzyme-1
    protein kinase, endonuclease, a regulator of the unfolded protein response, leading to the activation
    of the transcription factor Xbp1 - regulated Ire1-dependent decay - regulation of RNA splicing - photoreceptor differentiation and rhabdomere morphogenesis

    Mesencephalic astrocyte-derived neurotrophic factor
    a secreted protein - unfolded protein response - immune modulation - expressed
    in glial cells - supports the dopaminergic system in non-cell-autonomous manner

    pancreatic eIF-2α kinase (common alternative name: Perk)
    phosphorylates and inhibits the translation initiation factor 2 α, control of intestinal stem cell proliferation,
    homeostatic regeneration, unfolded protein response of the ER, endoplasmic reticulum stress

    X-box-binding protein 1
    enhances the expression of genes encoding ER chaperones, enzymes, and the ER protein degradation machinery



    Drosophila melanogaster Activating Transcription Factor 4 regulates glycolysis during endoplasmic reticulum stress

    Endoplasmic reticulum (ER) stress results from an imbalance between the load of proteins entering the secretory pathway and the ability of the ER to fold and process them. The response to ER stress is mediated by a collection of signaling pathways termed the unfolded protein response (UPR), which plays important roles in development and disease. This study shows that in Drosophila melanogaster S2 cells, ER stress induces a coordinated change in the expression of genes involved in carbon metabolism. Genes encoding enzymes that carry out glycolysis were up-regulated, whereas genes encoding proteins in the TCA cycle and respiratory chain complexes were down-regulated. The UPR transcription factor Atf4/Cryptocephal was necessary for the up-regulation of glycolytic enzymes and Lactate dehydrogenase (Ldh). Furthermore, Atf4 binding motifs in promoters for these genes could partially account for their regulation during ER stress. Finally, flies up-regulated Ldh and produced more lactate when subjected to ER stress. Together these results suggest that Atf4 mediates a shift from a metabolism based on oxidative phosphorylation to one more heavily reliant on glycolysis, reminiscent of aerobic glycolysis or the Warburg effect observed in cancer and other proliferative cells (Lee, 2015).

    As the flux of proteins through the ER varies considerably among cell types and in different conditions, cells maintain a balance between the load on the ER and its protein folding capacity. However, a number of biochemical, physiological, and pathological stimuli can disrupt this balance, resulting in ER stress. To re-establish ER homeostasis, the unfolded protein response is activated. This network of pathways up-regulates genes encoding ER-specific chaperones and other proteins involved in protein secretion, while also attenuating protein translation and degrading certain ER-associated mRNAs. The UPR is broadly conserved across eukaryotes and is essential for normal development in several model organisms, particularly for professional secretory cells, where it is thought to be important for the establishment and maintenance of high levels of protein secretion . It is also induced during many metabolic conditions including diabetes, hyperlipidemia, and inflammation, and has been implicated in various cancers, especially in the growth of large tumors that rely on an effective response to hypoxia (Lee, 2015).

    The UPR is carried out by three main signaling branches. One of these is initiated by the ER transmembrane protein Inositol-requiring enzyme 1 (Ire1). When activated by ER stress, the cytosolic endoribonuclease domain of Ire1 cleaves the mRNA encoding the transcription factor Xbp1, thereby initiating an unconventional splicing event that produces the mRNA template encoding a highly active form of Xbp1. Ire1 also cleaves other mRNAs associated with the ER membrane, through a pathway that is particularly active in Drosophila cells and that may reduce the load on the ER. A second sensor of ER stress, Activating transcription factor 6 (Atf6), is activated by proteolysis, which releases it from the ER membrane and allows it to travel to the nucleus and regulate gene expression. Finally, Protein kinase RNA (PKR)- like Pancreatic ER kinase (Perk) phosphorylates eukaryotic initiation factor 2 alpha, leading to a general attenuation of protein synthesis as well as the translational up-regulation of certain mRNAs that contain upstream open reading frames (ORFs) in their 5' untranslated regions. Activating transcription factor 4 (Atf4) is among those proteins that are up-regulated translationally during ER stress, and regulates genes involved in protein secretion as well as amino acid import and resistance to oxidative stress (Lee, 2015).

    In addition to its direct effects on the protein secretory pathway, the UPR influences several other cellular pathways including apoptosis, inflammation, and lipid synthesis. Furthermore, the UPR (particularly the Perk/Atf4 branch) appears to have close ties to mitochondrial function. For example, knockout of Mitofusin 2, a key mitochondrial fusion protein, activates Perk, leading to enhanced reactive oxygen species (ROS) production and reduced respiration. Atf4 also increases expression of Parkin, which mediates degradation of damaged mitochondria, protecting cells from ER stress-induced mitochondrial damage. Despite clear links between ER stress and mitochondria, the mechanistic relationship between the UPR and mitochondrial metabolism is not well-understood (Lee, 2015).

    This study reports that the UPR in Drosophila S2 cells triggers a coordinated change in the expression of genes involved in carbon metabolism. The metabolism of glucose as an energy source produces pyruvate, which can then enter the mitochondria and the tricarboxylic acid (TCA) cycle to produce reducing equivalents for oxidative phosphorylation (OXPHOS). For most cells in normal conditions, the majority of ATP is produced through OXPHOS. However, in hypoxic conditions when OXPHOS is limited, cells rely heavily on glycolysis to compensate for the decrease in ATP production, and convert the excess pyruvate to lactate, which then leaves the cel. This shift from OXPHOS to glycolysis is seen in a variety of cancers even when cells have access to oxygen, an effect known as aerobic glycolysis or the Warburg effect, and is thought to be a hallmark of cancer cells. Aerobic glycolysis is also becoming increasingly recognized as a metabolic signature of other cell types as well, including stem cells and activated immune cells (Lee, 2015).

    In Drosophila, the Estrogen-related receptor (dERR) is the only transcription factor known to regulate glycolytic genes (Li et al. 2013). Its activity is temporally regulated during mid-embryogenesis to support aerobic glycolysis during larval growth (Tennessen, 2011). Moreover, a recent study found that glycolytic gene expression under hypoxic conditions in larvae is partially dependent on dERR (Li, 2013). This study shows that the UPR transcription factor Atf4 also regulates glycolytic genes, contributing to a broad regulation of metabolic gene expression during ER stress that is reminiscent of the Warburg effect (Lee, 2015).

    This study has shown that Drosophila S2 cells subjected to ER stress up-regulate glycolytic genes and Ldh and down-regulate genes involved in the TCA cycle and respiratory chain complex. Furthermore, Atf4 is responsible for the up-regulation of glycolytic genes and Ldh. How TCA cycle and respiratory chain complex genes are down-regulated during ER stress requires further investigation, although the lack of effect of Atf4 depletion suggests that these are not regulated as an indirect consequence of glycolysis up-regulation (Lee, 2015).

    Despite a highly coordinated change in gene expression for metabolic genes during ER stress, this study did not detect any changes in actual metabolism in S2 cells. Because these cells have been in culture for decades and have likely been selected for rapid proliferation, it is possible that they are already undergoing some version of aerobic glycolysis, such that the underlying gene regulation during ER stress is preserved but any metabolic changes are masked. Others have also noted that S2 cells are resistant to hypoxia, and do not produce more lactate except in extreme conditions. The increase in lactate observed through in vivo studies in flies subjected to ER stress, however, suggests that in a more physiological setting, the gene expression changes shown here do mediate a metabolic shift toward aerobic glycolysis (Lee, 2015).

    Up-regulation of glycolytic genes during ER stress has not been observed in genome-wide studies of mammalian cells. However, several lines of evidence suggest that mammalian cells subjected to ER stress may undergo a glycolytic shift. For example, a recent study examining human gliomas found coordinated up-regulation of UPR targets and glycolysis, which correlated with poor patient prognosis; and both ER stress and overexpresson of Perk have been shown to reduce mitochondrial respiration in cultured mammalian cells (Lee, 2015).

    The link between ER stress and metabolism can be rationalized by the need to generate building blocks for biosynthesis of glycoproteins and lipids. Early intermediates of glycolysis are necessary for production of uridine diphosphate-N-acetylglucosamine (UDP-GlcNAc), an important donor molecule for N-glycosylation of proteins in the ER. Both fructose-6-phosphate and dihydroxyacetone phosphate are also required for synthesis of glycolipids. An increased flux through glycolysis may therefore be important to support the increased production of glycerophospholipids and glycoproteins that are associated with the UPR. In support of this view, glucose deprivation or inhibition of glycolysis with 2-deoxy-D-glucose induces the UPR, which contributes to cell death, especially in cancer cells, and this effect can be rescued by UDP-GlcNAc. The hexosamine biosynthetic pathway generating UDP-GlcNAc is also directly activated by Xbp1, and stimulates cardioprotection during ischemia/reperfusion injury and increases longevity in worms (Lee, 2015).

    A second, non-mutually exclusive explanation for a shift to glycolysis during ER stress is the need to limit production of ROS. Along with mitochondrial respiration, protein folding in the ER is one of the main sources of ROS, which are produced by the normal process of disulfide bond-coupled folding. If allowed to accumulate, these ROS can cause oxidative stress and damage to cells, eventually leading to apoptosis. Several studies have confirmed that ROS are produced during ER stress, when protein folding is inefficient and more rounds of oxidation and reduction are required to fold proteins. Limiting other sources of oxidative stress, such as by down-regulating the TCA cycle and thereby restricting the flux through OXPHOS (the main source of ROS in the mitochondria), may be a way to mitigate the damage and allow cells to recover more effectively (Lee, 2015).

    Finally, the advantage of the Warburg effect for tumor growth may arise from the increased rate of ATP production by glycolysis compared to OXPHOS, despite its lower efficiency of conversion. By analogy, a metabolic shift during ER stress could rapidly supply ATP necessary for protein folding and processing. Indeed, cancer cells showing elevated levels of ENTPD5, an ER UDPase, promotes aerobic glycolysis to increase ATP for protein N-glycosylation and refolding (Lee, 2015).

    Overall, these results identify Atf4 as a transcriptional regulator of glycolysis during ER stress. As Atf4 is expressed throughout fly development (Hewes, 2000), it may regulate glycolysis in other situations as well: notably, Atf4 mutant flies are lean and have reduced circulating carbohydrates, suggesting a role in metabolism. Furthermore, because the Perk-Atf4 branch of UPR is activated during hypoxia, it will be interesting to see whether Atf4 contributes to regulation of glycolysis in other developmental, physiological (hypoxia), or pathological process during which glycolysis regulated. More broadly, since the UPR is activated in many types of cancer, its ability to regulate glucose metabolism may play a contributing role in the Warburg effect (Lee, 2015).

    A modified UPR stress sensing system reveals a novel tissue distribution of IRE1/XBP1 activity during normal Drosophila development

    Eukaryotic cells respond to stress caused by the accumulation of unfolded/misfolded proteins in the endoplasmic reticulum by activating the intracellular signaling pathways referred to as the unfolded protein response (UPR). In metazoans, UPR consists of three parallel branches, each characterized by its stress sensor protein, IRE1, ATF6, and PERK, respectively. In Drosophila, IRE1/XBP1 pathway is considered to function as a major branch of UPR; however, its physiological roles during the normal development and homeostasis remain poorly understood. To visualize IRE1/XBP1 activity in fly tissues under normal physiological conditions, previously reported XBP1 stress sensing systems, were modified based on the recent reports regarding the unconventional splicing of XBP1/HAC1 mRNA. The improved XBP1 stress sensing system allowed detection of new IRE1/XBP1 activities in the brain, gut, Malpighian tubules, and trachea of third instar larvae and in the adult male reproductive organ. Specifically, in the larval brain, IRE1/XBP1 activity was detected exclusively in glia, although previous reports have largely focused on IRE1/XBP1 activity in neurons. Unexpected glial IRE1/XBP1 activity may provide novel insights into the brain homeostasis regulated by the UPR (Sone, 2013).

    Inadequate sensitivity of existing XBP1 stress sensing systems can be overcome by improving the efficiency of unconventional splicing of xbp1 mRNA. Recent reports regarding the cellular localization of XBP1/HAC1 mRNA during its splicing allowed construction of a highly sensitive HG stress indicator that can visualize the activation of IRE1/XBP1 pathway at the third instar larval stage during normal Drosophila development. Several types of cells in the organs where IRE1/XBP1 activity was detected are known for having high secretory capacity (Sone, 2013).

    In the larval brain, significant IRE1/XBP1 activity was found in glial cells. While glia had been originally thought to function as the structural support cells in the nervous system, it has been revealed that they play several important roles in the development and homeostasis of the nervous system. In Drosophila, glial cells are classified into three classes (surface-, cortex-, and neuropil-associated glia), each of which is subdivided further morphologically. Whether IRE1/XBP1 active glia is restricted to only a subtype of those glia, or more broadly, is currently under investigation (Sone, 2013).

    In mammals, oligodendrocytes in the central nerve system and Schwann cells in the peripheral nerve system myelinate axons by producing a large amount of myelin membrane proteins, cholesterol, and membrane lipids through the secretory pathway. Recent reports suggested that ER stress in myelinating cells is important in the pathogenesis of various disorders of myelin. Neuropil glia and peripheral glia in Drosophila are the counterparts of oligodendrocytes and Schwann cells, respectively. Therefore, these cells are the candidates that show constitutive IRE1/XBP1 activity. Although Drosophila glia do not generate myelin sheaths, they form multi-layered membrane sheaths around neurons that are morphologically similar to the myelin sheaths in mammals. Thus, it is possible that the IRE1/XBP1 active glia protect neurons from their deterioration through this ensheathment, thereby contributing to brain homeostasis. Further studies are expected to be informative as to the pathological significance of IRE1/XBP1 functions in human glia (Sone, 2013).

    IRE1/XBP1 pathway does not appear to be active in neuron. However, the possibility of neuronal IRE1/XBP1 activation in the brain was not excluded. In fact, slight neuronal IRE1/XBP1 activity was occasionally observed in the ventral nerve cord during repeated experiments. In this study, it is concluded that in the third instar larval brain, the IRE1/XBP1 pathway is predominantly activated in glia while the activation is not detectable in neurons (Sone, 2013).

    The importance of IRE1/XBP1 activity in the gut has already been studied in Caenorhabditis elegans and mammals. Intra-tissue distribution of IRE1/XBP1 activity was detected in the proventriculus region of the gut. In the larval midgut and hindgut, an irregular distribution of IRE1/XBP1 active cells was observed. These were not entero-endocrine cells, as they did not colocalize with anti-Prospero antibody that marks those cells. Secretory intestinal cells in the midgut other than entero-endocrine cells including the intestinal stem cells are possible candidates for these IRE1/XBP1 active cells (Sone, 2013).

    IRE1/XBP1 activity in the fly Malpighian tubules (analogous to the kidney in mammals) was also unexpected. The activity was detected throughout the organ, but not all of the cells were IRE1/XBP1 active. Although the Malpighian tubules are attached at the junction of the midgut and the hindgut, they are morphologically and functionally independent from both of them. Identification of the IRE1/XBP1 active cells in the gut and the Malpighian tubules might reflect a shared physiological function of both organs. One possible shared function may be the selective uptake of the essential molecules, including several metal ions, from the contents passing through those organs. IRE1/XBP1 pathway might regulate the function of some transporter channels in these organs. Drosophila Malpighian tubules are expected to be one of the models for the mammalian diabetic kidney diseases that are associated with UPR activation (Sone, 2013).

    In this study, IRE1/XBP1 activity was also detected in the trachea. Previous reports suggest its relevance to glial IRE1/XBP1 activity. One of them showed that tracheal development in Drosophila brain was controlled by signals from glia. According to the report, the branches of cerebral trachea grow around the neuropile. If IRE1/XBP1 active glia were neuropile-associated glia, assessing IRE1/XBP1 activity at neuropile-associated glia is likely reveal the shared physiological function of IRE1/XBP1 pathway between brain and trachea. The other report, using embryonic trachea, indicated that the proper combination of secretory activity and endocytotic activity was importaAnt for the maturation of trachea as an airway. In tracheal maturation, Sar1, one of the core COPII proteins, was required for the secretion of protein, the luminal matrix assembly, and the following expansion of tube diameter to avoid the clogging of protein, while Rab5, the small GTPase that regulates the early stage of endocytosis, was required for the clearance of deposited materials in the lumen. It can be predicted that, even in larval trachea, IRE1/XBP1 pathway plays a crucial role in tracheal maturation by supplying the properly folded proteins to the transport machinery. In that case, in view of second instar larval lethality of xbp1-/- hypomorph mutant, it could also be hypothesized that the tracheal maturation/maintenance is still important for larval lethality, in addition to its importance for the embryonic development (Sone, 2013).

    IRE1/XBP1 activity in the salivary gland has already been reported in a previous study. The salivary gland is commonly used for the determination of the subcellular localization of the protein in Drosophila cells due to its morphological features. The nuclear localization of HG indicator, XBP1-EGFP molecule, was clearly indicated. In addition, weak IRE1/XBP1 activity was detected in the fat body; it was attached to the salivary gland. Generally, the Drosophila fat body, which is equivalent to mammalian adipose tissue, functions as the organ for energy/lipid storage and is distributed throughout the larval body (Sone, 2013).

    In addition to the larval tissues, IRE1/XBP1 activity was analyzed in the adult male reproductive organs. Though a previous RT-PCR suggested the activity in the testis, the areas where IRE1/XBP1 activity was detected were the accessory glands and a limited area of the testis close to the testicular duct. In the accessory gland, seminal fluid containing several hormones, which facilitate reproductive traits such as sperm transfer, sperm storage, female receptivity, ovulation, and oogenesis, are produced and secreted. There are two morphologically distinct secretory cell types in Drosophila accessory gland. Ninety-six percent of the secretory cells are categorized as main cells and the others are secondary cells. Based on the intra-tissue distribution of IRE1/XBP1 active cells in the accessory gland, the active cells are likely to be main cells. Since each of these cell types expresses a unique set of genes, the confirmation of IRE1/XBP1 active cell type is expected to allow ingnarrow down the proteins related to IRE1/XBP1 activity. IRE1/XBP1 pathway is likely to function, to some extent, in maintaining proper fertility (Sone, 2013).

    On the other hand, a possibility is considered that the EGFP signal detected in each organ might not necessarily reflect the unconventional splicing of xbp1-EGFP mRNA. Higher concentrations of the spliced xbp1-EGFP mRNA and resulting XBP1-EGFP in the cells induced by the Gal4/UAS system might cause the artifactual EGFP signal. The possibility was excluded that the EGFP signal in this study was detected independently of the unconventional splicing, based on the results in this study and the following reasoning (Sone, 2013).

    There are two possible molecular mechanisms that cause the artifactual EGFP signal which is not derived from the unconventional splicing of xbp1-EGFP mRNA. One is the generation of EGFP or abnormal EGFP fusion proteins, resulting from translation initiation at the start codon of the EGFP coding sequence or at ATG codons coding Met residues in XBP1(s), respectively. The other is the proteolytic digestion of XBP1-EGFP fusion protein at the junction of XBP1 and EGFP portions. Both of these are prone to happen upon overexpression of fusion proteins in cells. In particular, the proteolytic digestion is often observed in the overexpression of GST fusion protein in Escherichia coli (Sone, 2013).

    In this system, there is no nuclear localization signal (NLS) on the EGFP molecule. In contrast, xbp1 gene carries a NLS coding sequence located upstream of the unconventional splice site. There are a total of 11 ATG codons that code the Met residues of XBP1(s) molecule. Eight of the ATG codons are located downstream of NLS coding sequence. Therefore, due to the lack of NLS, both EGFP and the EGFP fusion proteins using these eight ATG codons as start codons should diffuse all over the cell upon their synthesis, if they were generated. EGFP signal was detected exclusively in the nucleus in the salivary gland, which is often used for the analysis of the cellular localization of the proteins in Drosophila. Hence, it is not reasonable to conclude that either EGFP or the possible eight EGFP fusion proteins above were expressed in the cells. Only the EGFP fusion proteins that use the other three ATG codons that are upstream of the NLS as start codons should be synthesized upon unconventional splicing and localized in the nucleus. The estimated molecular weights of those fusion proteins are 73.7, 74.3, and 80.0 kDa, respectively. Only a significant single band that represented the intact XBP1-EGFP (80.0 kDa) was detected in the S2 cell extract, in which XBP1-EGFP was overexpressed through the Gal4/UAS system. Therefore, it is concluded that the EGFP fusion protein synthesized in this study was the intact XBP1-EGFP (Sone, 2013).

    Additionally, there are several ATG codons that are located upstream of the unconventional splice site and are in frame with EGFP coding sequence on unspliced xbp1 mRNA. However, inside the 23 bp of unconventional-spliced fragment, there is a TGA stop codon that is also in frame with EGFP coding sequence. Even if the translation initiated from these start codons, the synthesis of these products should be terminated at this TGA stop codon before the ribosome would reach the EGFP coding region (Sone, 2013).

    Regarding the proteolytic digestion of XBP1-EGFP fusion protein, the resulting EGFP should diffuse all over the cell upon its synthesis due to the lack of NLS. Therefore, the possibility of the proteolytic digestion is also excluded based on the same reasoning as above. Taken together, it is concluded that the detected EGFP signal in this study exclusively reflected the occurrence of unconventional splicing of xbp1-EGFP mRNA (Sone, 2013).

    In summary, this study improved the sensitivity of the XBP1 stress sensing system and newly identified several organs where IRE1/XBP1 pathway is constitutively activated under normal physiological conditions. In particular, in the larval brain, significant glial specific activation was detected. The improved system is expected to provide with a number of clues to reveal the molecular mechanisms underlying the normal development and homeostasis controlled by IRE1/XBP1 pathway (Sone, 2013).

    Changes in Drosophila mitochondrial proteins following chaperone-mediated lifespan extension confirm a role of Hsp22 in mitochondrial UPR and reveal a mitochondrial localization for cathepsin D

    Hsp22 is a small mitochondrial heat shock protein (sHSP) preferentially up-regulated during aging in Drosophila. Its developmental expression is strictly regulated and it is rapidly induced in conditions of stress. Hsp22 is one of the few sHSP to be localized inside mitochondria, and is the first sHSP to be involved in the mitochondrial unfolding protein response (UPRMT) together with Hsp60, mitochondrial Hsp70 and TRAP1. The UPRMT is a pro-longevity mechanism, and interestingly Hsp22 over-expression by-itself increases lifespan and resistance to stress. Among the proteins influenced by Hsp22 expression were proteins from the electron transport chain (ETC), the TCA cycle and mitochondrial Hsp70. Hsp22 co-migrates with ETC components and its over-expression is associated with an increase in mitochondrial protease activity. Interestingly, the only protease that showed significant changes upon Hsp22 over-expression was cathepsin D, which is localized in mitochondria in addition to lysosome in D. melanogaster as evidenced by cellular fractionation. Together the results are consistent with a role of Hsp22 in the UPRMT and in mitochondrial proteostasis (Morrow, 2016).

    Exploring the Conserved role of MANF in the unfolded protein response in Drosophila melanogaster

    Disturbances in the homeostasis of endoplasmic reticulum (ER) referred to as ER stress is involved in a variety of human diseases. ER stress activates unfolded protein response (UPR), a cellular mechanism the purpose of which is to restore ER homeostasis. Previous studies show that Mesencephalic Astrocyte-derived Neurotrophic Factor (MANF) is an important novel component in the regulation of UPR. In vertebrates, MANF is upregulated by ER stress and protects cells against ER stress-induced cell death. Biochemical studies have revealed an interaction between mammalian MANF and GRP78, the major ER chaperone promoting protein folding. This study discovered that the upregulation of MANF expression in response to drug-induced ER stress is conserved between Drosophila and mammals. Additionally, by using a genetic in vivo approach genetic interactions was found between Drosophila Manf and genes encoding for Drosophila homologues of GRP78, PERK and XBP1, the key components of UPR. These data suggest a role for Manf in the regulation of Drosophila UPR (Lindstrom, 2016).

    The accumulation of unfolded or misfolded proteins causes disturbances in endoplasmic reticulum (ER) homeostasis, a phenomenon referred to as ER stress. ER stress in turn activates the unfolded protein response (UPR). In order to overcome ER stress, UPR leads to attenuation of protein synthesis, enhancement of degradation of unfolded proteins, and activation of specific signalling cascades. These events aim to reduce the overall protein load in the ER and to enhance the protein folding capacity by selective transcription of chaperones. UPR is activated through three signalling cascades by ER transmembrane sensor proteins PERK (PRKR-like endoplasmic reticulum kinase), IRE1 (inositol requiring enzyme 1), and ATF6 (activating transcription factor 6). All of these three proteins are maintained inactive in normal cellular status by binding to the major ER chaperone GRP78/BiP (Glucose-regulated protein 78/Binding immunoglobulin protein). Upon ER stress, GRP78 is dissociated from the sensor proteins which are subsequently activated. The most ancient of these signalling cascades is mediated by IRE1, the sole branch of UPR identified in Saccharomyces cerevisiae. IRE1 has kinase activity and endoribonuclease activity needed for degradation of mRNAs in order to relieve the protein synthesis load. IRE1 is also responsible for the unconventional splicing and thus activation of transcription factor XBP1 (X-box Binding Protein-1), a positive regulator of ER chaperone and other UPR related gene expression. Activated PERK attenuates overall protein synthesis through phosphorylating and thus inhibiting EIF2α (eukaryotic translation initiation factor 2, subunit 1 α). However, the decreased activation of EIF2α results in an upregulated translation of specific target mRNAs including ATF4 (activating transcription factor 4). The third signalling pathway is mediated through ATF6, a transcription factor activated by its cleavage and translocation to the nucleus (Lindstrom, 2016).

    In Drosophila, both IRE1- and PERK-mediated UPR signalling cascades are conserved. The amino acid sequence of the Drosophila homologue of ATF6 is highly similar to its mammalian counterpart, but experimental evidence for its involvement in Drosophila UPR is lacking. Similar to mammals, the expression of Drosophila homologue of GRP78, Hsc3 (Heat shock protein cognate 3), is upregulated upon induced ER stress in Xbp1-dependent manner but no biochemical data are available to show its association with ER stress sensor proteins (Lindstrom, 2016).

    The MANF/CDNF family of neurotrophic factors was first characterized based on its trophic function on dopaminergic neurons in vitro and in vivo. When injected into the brain, recombinant mammalian MANF (Mesencephalic Astrocyte-derived Neurotrophic Factor) and CDNF (Cerebral Dopamine Neurotrophic Factor) protect and repair dopaminergic neurons in toxin-induced rodent models of Parkinson's disease (PD) in vivo. The sole Drosophila homologue, DmManf, is expressed in and secreted from glial cells and supports the dopaminergic system in non-cell-autonomous manner (Palgi, 2009). The role of MANF as an extracellular trophic factor is further supported by the evidence that mammalian MANF is protective against ischemic injury in both neurons and cardiomyocytes (Airavaara, 2010; Glembotski, 2012). However, the biology of MANF is not thoroughly understood. Intriguingly, MANF localizes to the ER and has a protective role against ER stress in vitro and in vivo. Additionally, mammalian MANF binds GRP78 in Ca2+-dependent manner in vitro and this binding may regulate MANF secretion. MANF can be retained in the ER by its C-terminal signal sequence, RTDL in human and RSEL in Drosophila (Glembotski, 2012; Lindström, 2013). Experimental evidence suggests that mammalian MANF interacts with KDEL-R [KDEL (Lys-Asp-Glu-Leu) endoplasmic reticulum protein retention receptor] and that the C-terminal RTDL sequence of MANF is responsible for this interaction. The relevance of KDEL-R as a mediator of the functions of MANF has not been explored in vivo, yet. Recently, MANF was also shown to regulate the expression of ER-resident protein CRELD2 (Lindstrom, 2016).

    Both in vivo and in vitro studies have shown that MANF is upregulated after chemically induced ER stress and by misfolded mutant proteins accumulating in the ER. Mammalian MANF expression is activated upon ER stress by ATF6 and XBP1 through an ER stress response element II found in the promoter region of MANF. Based on a knockout mouse model, MANF was found to be essential for the survival of pancreatic β-cells and its loss resulted in severe diabetes due to reduction of beta cell mass and activation of UPR in the pancreatic islets. The protective role against 6-OHDA induced and ischemic neuronal damage has been suggested to rise from the ER-related functions of MANF as these processes have been shown to induce ER stress (Lindstrom, 2016).

    In Drosophila, the loss of DmManf is associated with upregulated expression of genes involved in UPR (Palgi, 2012). Additionally, the overexpression of DmManf resulted in downregulation of several UPR-related genes (Palgi, 2012). This study shows that, similar to mammalian MANF, the expression of DmManf is induced in response to ER stress in vitro. Further, transgenic approaches for gene silencing in vivo were applied to reveal genetic interactions between DmManf and genes with known functions in the maintenance of ER homeostasis and in UPR (Lindstrom, 2016).

    Increasing evidence indicates that ER stress and UPR play a major role in variety of human diseases including diabetes mellitus and neurodegenerative disorders. MANF is a secreted protein (Palgi, 2009; Lindholm, 2008), but also localizes to the ER and has a role in mammalian UPR (Mizobuchi, 2007; Palgi, 2012). This study examined the role of DmManf in UPR in the Drosophila model. Upregulation of MANF mRNA expression by ER stress-inducing agents was shown to be conserved in Drosophila S2 cells. Additionally, genetic interaction between DmManf and genes known to function in the ER and UPR were shown (see A simplified presentation of UPR and genetic interactions discovered for Drosophila Manf; Lindstrom, 2016).

    One of the interacting partners was Hsc3, the Drosophila homologue of mammalian chaperone GRP78. The silencing of Hsc3 in the wing resulted in an abnormal wing phenotype in wild type background. This wing phenotype was stronger in DmManf-overexpressing background. In cultured mammalian cells MANF has been shown to bind GRP78 in Ca2+-dependent manner and the loss of interaction between mammalian MANF and GRP78 was associated with increased secretion of MANF (Glembotski, 2012). In line, the knockdown of Hsc3 could lead to increased secretion of DmManf and lead to deprivation of intracellular DmManf. In a previous study, it was noticed that the deletion of ER retention signal RSEL increased the secretion of DmManf in S2 cells and decreased its functionality in rescue experiments in vivo (Lindström, 2013). Based on the physical interaction found between mammalian MANF and GRP78, the simultaneous overexpression of DmManf and knockdown of Hsc3 could also result in the abundant DmManf binding the residual Hsc3 and preventing other important cellular functions of Hsc3. Alternatively, the loss of Hsc3 could lead to decreased protein folding capacity in the ER and activation of UPR. The vast amount of DmManf protein could exhaust this already disturbed cellular state (Lindstrom, 2016).

    In previous studies, mammalian MANF has been suggested to have chaperone-like functions, e.g. by binding unfolded proteins in vitro but the putative chaperone activity remains unconfirmed. The major ER chaperone Hsc3 and DmManf clearly have distinct roles as either the overexpression or the loss of one could not complement for the loss of the other. However, the current study indicates that the interaction between MANF and GRP78 (Glembotski, 2012) is conserved. In future, the functional significance of this intriguing interaction deserves to be addressed in detail (Lindstrom, 2016).

    DmManf genetically interacted with PEK/PERK, an ER stress sensor protein. Similar to the silencing of Hsc3, simultaneous overexpression of DmManf worsened the phenotypes observed in PEK knockdown flies. Previous studies have indicated functional conservation of PERK in Drosophila and mammals. The Drosophila homologue to ATF4, the downstream target of activated PERK and selectively upregulated by UPR, showed no genetic interaction with DmManf in this study. It was previously shown that the abolishment of both zygotic and maternal DmManf resulted in increased phosphorylation of eIF2α, another molecular marker used for detecting ER stress (Palgi, 2012). This study abolished only the zygotic DmManf while maternal DmManf was still present. The loss of zygotic DmManf alone did not induce UPR when evaluated by other readouts, i.e. increased Hsc3 mRNA level and splicing of Xbp1. Although the zygotic DmManfΔ96 mutant larvae show only low amount of persisting maternal DmManf mRNA and protein, it could be sufficient to prevent the induction of UPRl (Lindstrom, 2016).

    Additionally, a genetic interaction was discovered between DmManf and Xbp1, a transcription factor mainly responsible for the regulation of UPR-induced genes. Upon UPR, the mRNA of Xbp1 is spliced by IRE1 and translated into a transcriptional activator of chaperone expression in response to the increased protein folding demand. According to previous studies, the spliced form of Xbp1 could mediate the UPR-induced upregulation of MANF in mammals. MANF has been suggested to have protective role against ER stress. During normal development, ER stress is detected in the secretory cells and the silencing of Xbp1 disturbs this developmental ER stress. Both mammalian and Drosophila MANF has been shown to have especially high expression levels in secretory tissues. Overexpression of DmManf increased Xbp1 mRNA level but the knockdown of Xbp1 did not affect DmManf expression levels. Also, the mRNA levels of Hsc3 were not upregulated in Xbp1 knockdown larvae. This could indicate the lack of transcriptional activation of DmManf and Hsc3 expression by Xbp1s in Xbp1-knockdown larvae. Therefore, knockdown of Xbp1 could compromise the regulation of DmManf expression in the developmental ER stress and deteriorate its function in the secretory cells (Lindstrom, 2016).

    ERAD is a cellular process aiming to clear out the unfolded and misfolded proteins from the ER. According to previous transcriptome analysis, sip3 was downregulated in DmManfΔ96 mutant larvae (Palgi, 2012). This study also found a genetic interaction between DmManf and sip3. Sip3 encodes a homologue to mammalian ER resident E3 ubiquitin ligase synoviolin/HRD1 with specific function in ERAD. Mammalian MANF is upregulated by ERSE-II (ER stress response element II) found in its promoter region. Interestingly, ERSE-II is also found in ERAD-related components HERPUD1 (homocysteine-inducible, ER stress-inducible, ubiquitin-like domain member 1, also known as HERP) and VIMP (VCP-interacting membrane protein, also known as selenoprotein S). ERSE-II has been hypothesized to regulate the protein quality control and degradation of misfolded proteins during ER stress suggesting that MANF could also have a role in these functions (Lindstrom, 2016).

    Surprisingly, overexpression of DmManf led to enhanced phenotypes in flies of which a UPR-related gene was knocked down. Thus far, overexpression of DmManf with any GAL4 driver tested has never resulted in a detectable phenotype or altered viability. According to a previous microarray analysis, DmManf overexpression led to downregulation of UPR-related genes (Palgi, 2012). This suggests that the overexpression of DmManf would disturb UPR signalling. Hypothetically, wild type background cells would be able to deal with the increased DmManf expression and the subsequent downregulation of UPR-related genes whereas the additional knockdown of an important component of UPR, e.g. Hsc3, PEK or Xbp1, could compromise the cell homeostasis (Lindstrom, 2016).

    An alternative explanation for these observations in interaction studies between UPR genes and DmManf would be that DmManf is actually a substrate for UPR. Then, the abundant expression of DmManf by UAS/GAL4 would rather model the effects of increased overall protein synthesis in ER than indicate specific ER-related functions for DmManf. DmManf enters the secretory pathway (Palgi, 2009) and its ectopic expression may cause stress to the protein folding machinery in the ER. Although the Xbp1 mRNA level was increased, the expression of Hsc3 was not altered indicating that overexpression of DmManf induces mild UPR. However, no similar effects were seen with overexpression of membrane-directed GFP suggesting that the observed phenomena were specific for DmManf (Lindstrom, 2016).

    The previous microarray study found that the total loss of DmManf is associated with upregulated expression of genes involved in UPR (Palgi, 2012). However, in the current study the mRNA levels of Hsc3 and Xbp1 were mildly decreased in DmManf mutant larvae. In the previous study, transcriptome analysis was done from the embryonic DmManf mutants lacking both maternal and zygotic DmManf. In the current study, RNA was collected from zygotic DmManf mutants with the persisting maternal DmManf mRNA and protein. The maternal DmManf is apparently sufficient to prevent induction of UPR and upregulation of UPR related genes (Lindstrom, 2016).

    This work provides evidence for the contribution of DmManf in Drosophila UPR. Further biochemical studies on the interaction between DmManf and UPR genes in Drosophila are needed to elucidate the details of this process (Lindstrom, 2016).

    Reduced sleep during social isolation leads to cellular stress and induction of the Unfolded Protein Response (UPR)

    Social isolation has a multitude of negative consequences on human health including the ability to endure challenges to the immune system, sleep amount and efficiency, and general morbidity and mortality. These adverse health outcomes are conserved in other social species. In the fruit fly Drosophila melanogaster, social isolation leads to increased aggression, impaired memory and reduced amounts of daytime sleep. There is a correlation between molecules affected by social isolation and those implicated in sleep in Drosophila. Previous work has demonstrated that acute sleep loss in flies and mice induced the unfolded protein response (UPR), an adaptive signaling pathway. One mechanism indicating UPR upregulation is elevated levels of the endoplasmic reticular chaperone BiP/GRP78. BiP overexpression in Drosophila has been shown to led to increased sleep rebound. Increased rebound sleep has also been demonstrated in socially isolated flies. Flies were used to study the effect of social isolation on cellular stress. Socially isolated flies displayed an increase in UPR markers; there were higher BiP levels, increased phosphorylation of the translation initiation factor eIF2alpha and increased splicing of xbp1. These are all indicators of UPR activation. In addition, the effects of isolation on the UPR were reversible; pharmacologically and genetically altering sleep in the flies modulated the UPR. It is concluded that the reduction in sleep observed in socially isolated flies is a cellular stressor that results in UPR induction (Brown, 2017).

    EDEM function in ERAD protects against chronic ER Proteinopathy and age-related physiological decline in Drosophila

    The unfolded protein response (UPR), which protects cells against accumulation of misfolded proteins in the ER, is induced in several age-associated degenerative diseases. However, sustained UPR activation has negative effects on cellular functions and may worsen disease symptoms. It remains unknown whether and how UPR components can be utilized to counteract chronic ER proteinopathies. This study found that promotion of ER-associated degradation (ERAD) through upregulation of ERAD-enhancing alpha-mannosidase-like proteins [EDEMs; EDEM1 (CG3810) and EDEM2 (CG5682)] protected against chronic ER proteinopathy without inducing toxicity in a Drosophila model. ERAD activity in the brain decreased with aging, and upregulation of EDEMs suppressed age-dependent behavioral decline and extended the lifespan without affecting the UPR gene expression network. Intriguingly, EDEM mannosidase activity was dispensable for these protective effects. Therefore, upregulation of EDEM function in the ERAD protects against ER proteinopathy in vivo and thus represents a potential therapeutic target for chronic diseases (Sekiya, 2017).


    References

    Airavaara M, Chiocco, M. J., Howard, D. B., Zuchowski, K. L., Peränen J., Liu, C., et al. (2010) Widespread cortical expression of MANF by AAV serotype 7: Localization and protection against a brain injury. Exp Neurol. 225: 104-113. PubMed ID: 20685313

    Brown, M. K., Strus, E. and Naidoo, N. (2017). Reduced sleep during social isolation leads to cellular stress and induction of the Unfolded Protein Response (UPR). Sleep [Epub ahead of print]. PubMed ID: 28541519

    Glembotski, C. C., Thuerauf, D. J., Huang, C., Vekich, J. A., Gottlieb, R. A., Doroudgar, S. (2012). Mesencephalic astrocyte-derived neurotrophic factor protects the heart from ischemic damage and is selectively secreted upon sarco/endoplasmic reticulum calcium depletion. J Biol Chem. 287: 25893-25904. PubMed ID: 22637475

    Lee, J.E., Oney, M., Frizzell, K., Phadnis, N. and Hollien, J. (2015). Drosophila melanogaster activating transcription factor 4 regulates glycolysis during endoplasmic reticulum stress. G3 (Bethesda) 5: 667-675.  PubMed ID: 25681259

    Lindholm, P., Peränen, J., Andressoo, J-O., Kalkkinen, N., Kokaia, Z., Lindvall, O., et al. (2008). MANF is widely expressed in mammalian tissues and differently regulated after ischemic and epileptic insults in rodent brain. Mol Cell Neurosci. 39: 356-371. PubMed ID: 18718866

    Lindström, R., Lindholm, P., Kallijärvi, J., Yu, L-Y., Piepponen, T. P., Arumäe, U., et al. (2013). Characterization of the Structural and Functional Determinants of MANF/CDNF in Drosophila In Vivo Model. PLoS ONE 8(9):e73928. PubMed ID: 24019940

    Lindstrom, R., Lindholm, P., Kallijarvi, J., Palgi, M., Saarma, M. and Heino, T. I. (2016). Exploring the Conserved role of MANF in the unfolded protein response in Drosophila melanogaster. PLoS One 11: e0151550. PubMed ID: 26975047

    Mizobuchi N, Hoseki J, Kubota H, Toyokuni S, Nozaki J-I, Naitoh M, et al. (2007). ARMET is a soluble ER protein induced by the unfolded protein response via ERSE-II element. Cell Struct Funct. 32: 41-50. PubMed ID: 17507765

    Morrow, G., et al. (2016). Changes in Drosophila mitochondrial proteins following chaperone-mediated lifespan extension confirm a role of Hsp22 in mitochondrial UPR and reveal a mitochondrial localization for cathepsin D. Mech Ageing Dev 155: 36-47. PubMed ID: 26930296

    Palgi, M., Lindström, R., Peränen, J., Piepponen, T. P., Saarma, M., Heino, T. I. (2009). Evidence that DmMANF is an invertebrate neurotrophic factor supporting dopaminergic neurons. Proc Natl Acad Sci U S A. 2009;106: 2429-2434. PubMed ID: 19164766

    Palgi, M., Greco, D., Lindström, R., Auvinen, P., Heino, T. I. (2012). Gene expression analysis of Drosophila Manf mutants reveals perturbations in membrane traffic and major metabolic changes. BMC Genomics. 13:134. PubMed ID: 22494833

    Sekiya, M., Maruko-Otake, A., Hearn, S., Sakakibara, Y., Fujisaki, N., Suzuki, E., Ando, K. and Iijima, K. M. (2017). EDEM function in ERAD protects against chronic ER Proteinopathy and age-related physiological decline in Drosophila. Dev Cell 41(6): 652-664.e655. PubMed ID: 28633019

    Sone, M., Zeng, X., Larese, J. and Ryoo, H. D. (2013). A modified UPR stress sensing system reveals a novel tissue distribution of IRE1/XBP1 activity during normal Drosophila development. Cell Stress Chaperones 18: 307-319. PubMed ID: 23160805


    Zygotically transcribed genes

    Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

    The Interactive Fly resides on the
    Society for Developmental Biology's Web server.