Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Histone acetylation and deacetylation wikipedia , lookup
Hedgehog signaling pathway wikipedia , lookup
Cellular differentiation wikipedia , lookup
Protein moonlighting wikipedia , lookup
Signal transduction wikipedia , lookup
List of types of proteins wikipedia , lookup
Identification of a novel Endoplasmic Reticulum Stress Response Element regulated by XBP1 Michael Misiewicz 260182952 Department of Anatomy and Cell Biology McGill University, Montreal August 2012 A thesis submitted to the Graduate Studies and Research of McGill University in partial fulfillment of the requirements of the degree of Master of Science. © Michael Misiewicz 2012 Abstract The Prion protein (PrP), which is the causative agent of scrapie diseases, has a still unclear physiological function, despite 30 years of research on its nature. However, a preponderance of evidence is beginning to support the idea that cellular prion protein (PrPC) has a pro-survival function. Here, we study the regulation of the prion protein gene (PRNP) in this context, as the regulation of the PRNP gene is not well understood. By homology, we identified in the PRNP a novel promoter element which bears similarity to the Endoplasmic Reticulum Stress Response Element (ERSE). This novel ERSE (“ERSE-26”) is able to regulate PRNP endogenously in response to endoplasmic reticulum (ER) stress. In order to determine whether or not the ERSE-26 exists elsewhere in the genome and what is co-regulated with PRNP, we conducted a bioinformatic search and identified 38 other genes with an ERSE-26. Their expression was confirmed by treating cultured primary human neurons and MCF-7 cells with the ER stressors Brefeldin A, Tunicamycin and Thapsigargin and conducting Reverse Transcriptase PCR or Quantitative PCR. We found that the genes SESN2, GADD45B and PRNP were significantly upregulated, and others showed an upward trend. Finally, a luciferase reporter construct containing the ERSE-26 only was used to identify that the ER stress transcription factor XBP1 is a transcription factor that induces ERSE-26 activity. Finally, we conducted a literature search to determine what functions of the cell are co-regulated with PRNP with the ERSE-26. Oxidative stress response and pro-survival genes were found in the ERSE-26 genes and found to be the most upregulated by the ERSE-26, strengthening the case for PrPC as a pro-survival gene. Misiewicz 2 Résumé La protéine Prion (PrP), qui est l'agent infectieux causant les encéphalopathies transmissibles, n'a pas toujours un rôle bien identifié dans la cellule, malgré 30 ans de recherche sur sa fonction physiologique. Cependant, de plus en plus de preuves commencent à impliquer PrP dans des fonctions de protection dans la cellule. Dans cette étude, nous avons étudié la régulation peu connue du promoteur du gène qui encode PrP (PRNP). Par homologie de séquence, nous avons identifié un nouvel élément dans le promoteur de PRNP qui ressemble à l'Endoplasmic Reticulum Stress Response Element (ERSE). Ce nouvel ERSE (appelé ERSE-26) est capable de réguler l'expression du PRNP de manière endogène en réponse au stress dans le réticulum endoplasmique (RE). Pour savoir si l'ERSE-26 existe ailleurs dans le génome et afin de trouver d'autres gènes régulé avec PRNP, nous avons fait une recherche bioinformatique dans le génome entier. Nous avons identifié 38 gènes contenant aussi un ERSE-26 dans leur promoteur. Afin de confirmer l'expression de ces gènes en réponse au stress ER, nous avons traité des cultures de neurones primaires humains et des cellules MCF-7 avec les activateurs du stress RE Brefeldin A, Tunicamycin et Thapsigargin, puis vérifié l'expression par Transcriptase Inverse PCR (RT-PCR) ou RT-PCR quantitative. Nous avons montré l'induction des gènes GADD45B, SESN2 et PRNP, et d'autres ont montré une tendance positive. Ensuite, un plasmide rapporteur luciferase contenant l'ERSE-26 seulement a été utilisé pour montrer que le facteur de transcription du stress ER XBP1 est un facteur de transcription responsable pour l'activité de l'ERSE-26. Finalement, nous avons fait une recherche dans la littérature afin de déterminer la fonction des gènes contenant ERSE-26. Les gènes répondant au stress oxydant et les gènes pro-survie étaient parmi les gènes ERSE-26, et aussi ont été le plus induits, soutenant le rôle protecteur du PrP dans la cellule. Misiewicz 3 Preface This thesis focuses on the regulation of the prion protein encoding gene and the discovery and characterization of a novel DNA motif that regulates prion and other gene expression in response to endoplasmic reticulum stress. It contains three chapters: a literature review, the novel research data in a manuscript submitted for publication and a conclusion. Manuscript contained in thesis: Identification of a novel Endoplasmic Reticulum Stress Response Element regulated by XBP1. Misiewicz 4 Acknowledgments I'd like to thank my supervisors, Dr. LeBlanc and Dr. Ruths for their great assistance. Under their scholarship I have learned a great deal during my Master's, and I am very grateful for their time and financial support. I'd also like to thank my lab members for their input, assistance and opinions. In particular, I'd like to thank Dr. Julie Jodoin for her help during my undergraduate degree and the beginning of my Master’s. I'd also like to thank Dr. Vikas Kushal for his great assistance in culturing primary human fetal neurons obtained from the University of Washington Tissue Bank. Misiewicz 5 Abbreviations AD Alzheimer Disease ADP Adenosine diphosphate AMV-RT Avian myeloblastosis virus reverse transcriptase ATF4 Activating Transcription Factor 4 ATF6 Activating Transcription Factor 6 ATP Adenosine triphosphate Bax BCL-2 Associated protein X BCL2 B-Cell Lymphoma 2 BFA Brefeldin A BH2 Bcl Homology Domain 2 BiP Binding Immunoglobin Protein BLAST Basic Linear Alignment and Search Tool bp Basepairs BSE Bovine spongiform encephalopathy cDNA Complimentary DNA ChIP Chromatin Immunoprecipitation CHOP C/EBP homologous protein CJD Creutzfeldt–Jakob disease CNS Central Nervous System CyPrP Cytosolic PrP DMSO Dimethyl sulfoxide eIF2α Eukaryotic initiation factor 2 alpha ER Endoplasmic Reticulum ERAD ER associated decay ERSE Endoplasmic Reticulum Stress FFI Fatal Familial Insomnia FSA Finite state automaton/Finite state automata GABA γ-Aminobutyric acid GADD45B Growth Arrest and DNA Damage Inducible 45 kDa Beta GPI Glycophosphatidylinositol Misiewicz 6 GRC Genome Reference Commission GSS Gerstmann–Sträussler–Scheinker syndrome HERP Homocysteine-inducible, endoplasmic reticulum stress-inducible HMM Hidden Markov Model IRE1α Inositol requiring enzyme 1 alpha mRNA Messenger RNA NF-Y Nuclear Transcription Factor Y ORF Open reading frame PCR Polymerase chain reaction PDI Protein Disulfide Isomerase PEI Polyethylenimine PERK PKR-like endoplasmic reticulum kinase PRNP Human Prion Protein Gene Prnp Mouse Prion Protein Gene PrP Prion Protein PrPC Prion protein, normal cellular form PrPSc Prion protein, scrapie form qPCR Quantitative Polymerase Chain Reaction ROS Reactive Oxygen Species RT-PCR Reverse Transcriptase Polymerase Chain Reaction S1P Site 1 Protease S2P Site 2 Protease SDS-PAGE Sodium dodecyl sulfate polyacrylimide gel electrophoresis sXBP1 Spliced XBP1 TAE Tris, Acetic Acid, EDTA Thps Thapsigargin TM Tunicamycin TNFα Tumor Necrosis Factor UPRE Unfolded Protein Response element XBP1 X-box binding protein 1 Misiewicz 7 List of figures Figure 1 – Known regulatory elements in the PRNP promoter Pg. 17 Figure 2 – Topology and transmembrane forms of PrP. Pg. 24 Figure 3 – ERSE-26 activity is XBP1 dependent. Pg. 64. Figure 4 – Induction of ER stress increases ERSE-26 gene expression Pg. 71. in cultured primary human neurons. Figure 5 – MCF-7 cells show induced ERSE-26 gene expression Pg 74. following ER stress treatment or sXBP1 transfection Table 1 – Genes containing an ERSE-26 in their promoter. Pg. 67. Supplemental Figure 1 – Activity of secreted luciferase in N2a cells Pg. 65. transfected with pML2-EL26 and sXBP1. Supplemental Table I – Primers used in RT-PCR Pg. 57. Supplemental Table II – Primers used in qRT-PCR. Pg. 58. Supplemental Table III – Genes containing and ERSE in their Pg. 68. promoter Misiewicz 8 Table of Contents Abstract ...................................................................................................................................... 2 Résumé ....................................................................................................................................... 3 Preface ........................................................................................................................................ 4 Acknowledgments ................................................................................................................... 5 Abbreviations ............................................................................................................................ 6 List of figures ............................................................................................................................. 8 Table of Contents ....................................................................................................................... 9 Chapter 1 – Literature Review ............................................................................................... 12 Introduction ............................................................................................................................. 12 1 Introduction to prion disease and the prion gene ........................................................ 13 1.1 The human PRNP gene and Prn locus .................................................................. 15 1.1.1 Structure of the PRNP gene .......................................................................................... 15 1.1.1.1 PRNP promoter structure ....................................................................... 16 1.1.1.2 Intron and exons in PRNP and Prn locus members .............................18 1.1.2 Shadoo and Doppel .......................................................................................................18 1.1.3 Patterns of constitutive PNRP expression .................................................................... 19 1.1.3.1 Signal dependent PRNP expression ...................................................... 20 1.1.3.2 Expression of PRNP in cancer and disease ........................................... 20 1.2 Functions and topology of prion protein .............................................................. 21 1.2.1 Topology of PrP ............................................................................................................... 21 1.2.2 Processing and localization ............................................................................................ 22 1.2.2.1 Signal peptides ......................................................................................... 22 1.2.2.2 Secretion .................................................................................................... 23 1.2.2.3 Post-translational modifications .............................................................23 1.2.3 Protective functions ........................................................................................................ 25 1.2.3.1 Anti-Bax function ..................................................................................... 25 1.2.3.2 Copper metabolism and redox activity ................................................. 26 2 The endoplasmic reticulum: functions and properties ................................................27 2.1 Protein folding .......................................................................................................... 28 Misiewicz 9 2.1.1 Mechanisms of enzymatic activity for protein folding in the ER .............................. 28 2.1.2 Post-translational modifications and sorting ............................................................... 29 2.2 ER Stress .................................................................................................................... 30 2.2.1 Discovery and definition ................................................................................................ 30 2.2.2 ER Stress signal transduction .........................................................................................31 2.2.2.1 IRE1α and XBP1 ....................................................................................... 31 2.2.2.2 Site 1 and Site 2 proteases and ATF6 .....................................................32 2.2.2.3 PKR-like endoplasmic reticulum kinase and ATF4 .............................33 2.2.3 Description of the ER stress response program ........................................................... 33 2.2.3.1 Translational ............................................................................................. 34 2.2.3.2 Transcriptional ..........................................................................................35 2.2.4 Regulation of ER stress mediated transcription ......................................................... 35 2.2.4.1 ERSE .......................................................................................................... 36 2.2.4.2 ERSE II ....................................................................................................... 36 2.2.4.3 ERSE-like ................................................................................................... 37 2.2.4.4 UPRE ......................................................................................................... 37 2.2.5 Exit from ER stress .......................................................................................................... 37 2.2.5.1 Restoration of ER homeostasis ............................................................... 38 2.2.5.2 Activation of apoptosis ............................................................................38 2.2.6 ER Stress and Disease ..................................................................................................... 39 3 Study of the Human Genome ......................................................................................... 40 3.1 Sequencing ................................................................................................................ 40 3.1.1 The current state of the art of genome sequencing ..................................................... 41 3.1.2 The Human Reference Genome .....................................................................................41 3.2 Annotations and the definition of loci and their promoters ...............................42 4 Computational pattern matching algorithms in biological sequences ......................43 4.1 Regular expressions ................................................................................................. 44 5 Methods of detecting gene expression levels ............................................................... 45 5.1 Quantitative Real Time PCR ................................................................................... 45 6 Working hypothesis and research objectives ................................................................ 46 Chapter 2 – Manuscript .......................................................................................................... 49 Preface ...................................................................................................................................... 49 Misiewicz 10 Author Contributions ............................................................................................................. 49 Identification of a novel Endoplasmic Reticulum Stress Response Element regulated by XBP1 ............................................................................................................................................50 Chapter 3 – General Discussion ............................................................................................ 81 Conclusion ............................................................................................................................... 84 References ................................................................................................................................ 85 Ethics approval ........................................................................................................................ 99 Misiewicz 11 Chapter 1 – Literature Review Introduction The prion protein is the causative agent of several fatal and incurable neurodegenerative diseases. It was originally discovered to be the first transmissible disease that does not contain nucleic acid, a controversial finding. Although it may also be the cause of disease, the prion protein also has a normal, non-disease state. Given its conservation across the animal kingdom, prion protein must have an important role in the cellular context. This role has been difficult to precisely define, despite 30 years of research on physiological as well as disease prion. Nevertheless, a plurality of evidence now depicts prion as a protective protein in its normal state. Prion protein inhibits activators of apoptosis, and mice lacking prion show subtle but noticeable defects over time. Prion has been identified as a binder of copper ions in the cell, and exhibits anti-superoxide activity. Finally, prion has been identified as a gene upregulated in response to oxidative stress. However, the regulation of prion expression is not well understood. The promoter of the gene has been poorly studied, and a better understanding of the promoter could provide insight into the activation and function of prion protein. In order to address this question, our lab has investigated the prion protein promoter. We identified a novel Endoplasmic Reticulum Stress Response element which is capable of transactivating prion expression in response to Endoplasmic Reticulum Stress. In this thesis, I will describe the search for this element in the entire human genome as well as a characterization of the transcription factors that activate it. Finally, by examining what genes contain this novel element, insight into the function of prion is gained. Misiewicz 12 1 Introduction to prion disease and the prion gene Prion diseases are progressive fatal neurodegenerative diseases with no known cure. Prion diseases can be inherited, arise sporadically, and even be transmitted between hosts due to ingestion of contaminated tissues or iatrogenically through tissue transfers. Many different prion diseases have been described, each with distinct symptoms and presentations (Norrby, 2011). Prion diseases have been well known in livestock for some time, since the mid 18th century (Parry, 1962). Scrapie in sheep and Bovine Spongiform Encephalopathy (BSE) have been widely reported in the media and are a concern for many public health officials due the apparent transmissible nature of these diseases. In humans, the most common is Creutzfeldt–Jakob disease (CJD), which can be contracted genetically, sporadically, or through ingestion of contaminated food. However other known scrapies in humans include Fatal Familial Insomnia (FFI), Kuru, and Gertsman Strankler Strausser syndrome (GSS). In cervids, a current area of concern is Chronic Wasting Disease, which affects deer and moose in the western part of North America. While each prion disease presents different symptoms, all are neurodegenerative diseases which result in death (Norrby, 2011). One of the most surprising aspects of scrapies is the apparent lack of traditional infectious agents. It was not until 1982 that the causes of scrapies were demonstrated to be an infectious protein, which when mis-folded became protease resistant and able to catalyze the conversion of normal cellular prion protein (PrP C) into infectious scrapie prion (PrPSc) (Prusiner, 1982). This proteinaceous infectious particle (hence “prion”) was proven to be devoid of any nucleic acids or any other hallmarks of bacteria or viruses. Because it seemed to contradict the germ theory of transmissible disease, this discovery was quite controversial initially but it has been subsequently accepted and Dr. Stanley Prusiner has won a Nobel prize for his work. Unusually, many of these diseases can be transmitted by eating or otherwise ingesting contaminated tissue. The most well known example of this phenomenon is the Misiewicz 13 consumption of scrapie meat, which causes a variant of CJD to occur. Additionally, iatrogenic transmission of prions is possible through hormone transfusions and improper medial instrument sterilization, a mode of transmission that was more common before there was a broad understanding of the causes of scrapies (Norrby, 2011). In laboratories, experimentally, prion diseases can be transmitted by injecting animals with brain homogenates from afficted animals. The catalytic activity of scrapie can even be reproduced in vitro through an assay called protein misfolding cyclic amplification. Although the effects and modes of transmission are well identified, the exact mechanisms by which PrPSc enters the body, crosses the blood brain barrier and induces misfolding of normal prion remains unclear. In addition to its role as the infection particle in scrapies, prion has a normal, physiological function. Prion's function has remained difficult to define despite 30 years of research. However, it has come to be implicated in the protection of cells from apoptosis induced by BCL-2 associated protein X (Bax). It has been associated with the response to oxidative stress, protection against serum deprivation, anti-apoptotic responses and a responder to endoplasmic reticulum stress. Prion's role in cancer due to its pro-survival properties is currently an area of active research (Westergard et al., 2007; Linden et al., 2008). PrP has been identified to be highly expressed in central nervous system (CNS) tissues, where it has been additionally implicated in the sleep-wakefulness cycle, synapse function, and memory formation. Finally, PrP is thought to be important in copper ion metabolism, and there is a line of evidence to suggest that PrP evolved from a family of metal ion transporters. Although PrP has been associated with many different cellular functions, one area of prion research that remains poorly defined is its regulation and response to various cellular states. There has been some research in this area, but it remains largely unknown how and why physiological prion expression is controlled in the cell. Understanding the processes that regulate the prion gene (PRNP) expression could lead to treatments for cancers and scrapie diseases (the severity of these two disease appears to be exacerbated by high PrP levels) and provide a better understanding of the ways in which cells cope with various stresses throughout their lifetimes. In this thesis, I aim to lay out the known mechanisms of PrP regulation, general mechanisms of pro-survival Misiewicz 14 responses in the cell, and how PrP is regulated by these responses. Our results show that the PRNP promoter contains a novel Endoplasmic Stress Response Element (ERSE) which enables it and other genes in the human genome to respond to perturbations in the homeostasis of the endoplasmic reticulum. This new element is a new mechanism for the regulation of PRNP expression, helps better explain the functionality of the PRNP promoter, and provides a better understanding of the broad cellular response to stress. 1.1 The human PRNP gene and Prn locus Human PrP is encoded by the PRNP gene. Originally mapped in 1986, it is located on chromosome 20, at 20p13 (Basler et al., 1986). In addition to PRNP, two other prion-like genes have been discovered, Doppel (PRND) and Shadoo (SPRN). PRNP and PRND are adjacent, and comprise what is sometimes referred to as the prion locus. PRNP is well conserved in vertebrates, showing conservation in mammals and existence in fish and birds (Rivera-Milla et al., 2006). The strong conservation and deeply branching nature of PRNP suggests that it has a very important role in animals. PRNP encodes a 2.7 kb mRNA (Sayers et al., 2012), with a 761 bp open reading frame (ORF). There are several variants of the PRNP mRNA (5 have been confirmed), but they all encode the same protein. The protein contains two signal peptides, and is almost entirely processed in the ER (Caughey et al., 1989; Yedidia et al., 2001). 1.1.1 Structure of the PRNP gene Human PRNP contains two exons and a 13kb intron. The open reading frame is located entirely in exon two. The large intron is unique to humans. Although the structure is well conserved in primates, other mammals have different numbers of introns and Misiewicz 15 exons. For example, the mouse Prnp gene contains three exons and two introns, although like PRNP the coding sequence is contained entirely in the last exon. 1.1.1.1 PRNP promoter structure The PRNP promoter contains several confirmed transcription factor binding sites, although generally not much is known about the function of the PRNP promoter (Figure 1). PRNP is expressed as a housekeeping gene: its promoter contains a SP1 site, no TATA box, high G/C content and a CCAAT box leading to constitutive PRNP expression and its classification as a housekeeping gene (Basler et al., 1986). In addition to its constitutive expression, the PRNP promoter also contains conditionally activated elements. Putative binding sites for heat shock elements, p53 binding sites, AP-2, Nkx2-5 and Myo-D, IL-6, NF-AT, Ets-1, metal responsive element binding sites, and cell membrane dependent AP-1 have been identified (Bellingham et al., 2009; Funke-Kaiser et al., 2001; Mahal et al., 2001). Finally, several elements of unknown function are well conserved in mammalian PRNP promoters (Mahal et al., 2001). However, of these putative sites, only Sp1, p53, metal transcription factor 1 and heat shock elements have been confirmed to regulate PRNP expression (Qin et al., 2009; Liang et al., 2007; Bellingham et al., 2009; Vincent et al., 2009; Shyu et al., 2002). There is evidence that PRNP might be regulated by p53, but also that PRNP might itself regulate p53; the relationship is unclear (Vincent et al., 2009; Paitel et al., 2003) The elements are located in two clusters around the transcription start site (TSS) of PRNP: from up to -800 bp to 0bp and within exon I (Mahal et al., 2001). Although many putative elements have been identified, a full understanding of the function of PrP will require a complete understanding of the PRNP promoter, and much remains to be discovered. Misiewicz 16 TSS -1000 p53 HSE -755 -737 -700 -680 EL26 -231 -196 MRE -90 -84 SP1 -62 -57 MRE +200 SP1 -39 -33 0bp 6 Figure 1 – Known regulatory elements in the PRNP promoter. Binding sites for p53, Heat Shock Element (HSE), the ERSE-26 (EL26), Metal Responsive Element (MRE) and SP1 binding site (SP1) have been shown to be functional in the PRNP promoter. Misiewicz 17 1.1.1.2 Intron and exons in PRNP and Prn locus members PRNP contains two exons and one intron of 13 kb. Exon I does not encode any part of the open reading frame, the protein is entirely encoded by exon II. PRND, which is located downstream of PRNP, contains a similar structure: two exons, separated by a large intron. The entire mRNA is 4 kbp, and the ORF is contained entirely in the first part of the second exon, like in PRNP. Next in the Prn locus following PRND is PRNT, a testis-specific version of PRNP. It does not appear to be protein coding however, although it also retains a similar structure to PRNP and PRNP. On chromosome 10, outside of the Prn locus is SPRN. It encodes a 3.2 kb intron, which like PRNP, PRND, and PRNT is comprised of two exons separated by a large intron. The open reading frame is located in the beginning part of the first exon. 1.1.2 Shadoo and Doppel SPRN and PRND are two genes that are homologous to PRNP and form the prion family. Doppel was originally discovered accidentally in Prnp knockout mice, when ataxia and Pukinje cell death was observed in aged mice (Sakaguchi et al., 1996). These results were surprising, because Prnp knockout mice normally do not show symptoms, however upon sequencing and further investigation, it was determined that the deletion used to produce the PRNP knockout mouse actually put the PRND gene under the PRNP promoter (Moore et al., 1999). In mice, Doppel is normally expressed in the testis only after birth, and the Prnp promoter led to aberrant expression in the CNS. Doppel was subsequently found to be very similar to the C-terminal end of PRNP, and its expression has been confirmed in many organisms and in various tissue types (Westaway et al., 2011). In addition, Doppel has an important role in spermatogenesis, being primarily Misiewicz 18 expressed in the male reproductive tract (Westaway et al., 2011). But its exact function remains mysterious. Shadoo was discovered by searching nucleotide databases for sequences with homology to PRNP. It is a short protein at 98 residues, and contains a great deal of homology to the N-terminal half of PrP, and the gene is known to be protein-coding in the central nervous system. Shadoo's function in the CNS is unclear, however (Westaway et al., 2011). Although the exact function of these prion-like genes is not precisely known, they might provide new insight into prion function and indicate the presence of a prionlike family of proteins, a relatively new and exciting development. 1.1.3 Patterns of constitutive PNRP expression Human PRNP is expressed in most tissues of the body. Its highest levels of expression are in the CNS, where in mice, it can be detected at embryonic day 8.5, when the transition to oxidative respiration occurs (Miele et al., 2003). In addition to its expression in CNS tissue, PrP has been found in most tissues of the human body (Westergard et al., 2007; Linden et al., 2008). PrP expression appears to be higher in immune system tissues, particularly lymphocytes, which typically have higher expression of binding of immunoglobulin protein (BiP) and other chaperones due to their large secretory load (Haas and Wabl, 1983; Li et al., 2001). Additionally, there is some evidence to suggest that PRNP may play a role in immune system signaling, and is implicated in lymphocyte development (Li et al., 2001). PRNP knockout mice do not show major phenotypes, however. There are defects in cell survival (Prnp -/- neurons die faster when deprived of serum in culture), memory formation, and vulnerability to chemically induced seizure. PRNP knockout mice are immune to prion infection (Linden et al., 2008). Misiewicz 19 1.1.3.1 Signal dependent PRNP expression Despite the putative and confirmed regulatory elements of PRNP, the regulation of PRNP remains unclear. There is evidence that PrP expression is increased in breast cancer carcinoma cells subjected to the ER stressors BFA and Bisphenol A in MCF-7 and TTE3 cells respectively (Tabuchi et al., 2006; Jodoin et al., 2007). Furthermore, during development of the hamster, injection of neuron growth factors induces expression PRNP mRNA expression (Mobley et al., 1988). PRNP expression has been observed in ischemia and in response to oxidative stress. Hydrogen peroxide and hyperbaric oxygen both upregulate PRNP expression. Prnp knockout mice show increased markers for oxidative stress, suggesting that PRNP has an important role in oxidative stress (Shyu et al., 2004; Liang et al., 2007). However, the functionality of the PRNP promoter and its properties remains an area of active research. 1.1.3.2 Expression of PRNP in cancer and disease Another disease role for PrP is in cancer. In the breast cancer carcinoma line MCF-7, cells which gain resistance to cell death by Tumor Necrosis Factor alpha (TNFα) also show large increases in both PRNP mRNA and PrP levels. Furthermore, overexpression of PrP in TNFα-sensitive MCF-7 cells results in a reduction in their sensitivity to TNFα treatment. Additionally, silencing PRNP in MCF-7 cells had the effect of re-sensitizing them to TNF-related apoptosis-inducing ligand mediated apoptosis (Diarra-Mehrpour et al., 2004). Given that cancer is fundamentally a disease of inappropriate cell survival and growth, there is currently great interest in better understanding PrP's pro-survival role in these contexts as it may lead to more effective treatments for cancers. Misiewicz 20 1.2 Functions and topology of prion protein While the exact details are still up for debate, a growing ensemble of evidence collected so far tends to indicate that PrP plays a protective role in the cell (Diarra-Mehrpour et al., 2004; Bounhar et al., 2001; Roucou et al., 2005; Jodoin et al., 2007). Due to its localization at the cell surface and synapse, PrP is thought to play a role in synaptic function as well. PrP knockout mice do not show very noticeable symptoms, however they are prone to seizures and have a shortened lifespan (Walz et al., 1999). Furthermore, PrP knockout mice are immune to infection with PrPSc, consistent with PrP's infectious protein property (Büeler et al., 1993). In addition, PrP has been associated with memory formation, the sleep-wakefulness cycle, synaptic function, immune system function and differentiation, and as a possible receptor or scaffold for intercellular signaling (Westergard et al., 2007; Linden et al., 2008). 1.2.1 Topology of PrP PrP is a 253 aa, 26 kDa protein that is primarily expressed as a glycosyl phosphatidylinositol (GPI)-anchored glycoprotein at the cell surface. It contains a disulfide bridge and two glycosylation sites (Figure 2A). There is a highly conserved octapeptide repeat region, a signal peptide and a GPI anchor sequence. Additionally, there is a transmembrane domain, which has lead to the discovery of two different forms of transmembrane PrP. One isoform, termed PrP Ctm, is found with the C-terminus in the ER lumen, and the N-terminus in the cytosol. PrP Ntm has also been identified, which is the inverse: the N-terminus is found in the lumen, and the C-terminus is in the cytosol (Figure 2B) (Hegde et al., 1998). In the C-terminal globular region of the protein, there are three alpha helices and two beta sheets. Although it was originally thought that PrP's Misiewicz 21 demonstrated ability to inhibit Bax activity (see below) would be conferred by the conserved octapeptide repeat region, it was determined that helix three is necessary and sufficient for this purpose (Laroche-Pierre et al., 2009). 1.2.2 Processing and localization The majority of PrP is present at the cell surface in lipid rafts, although there are several other isoforms that exists. A cytosolic, retrotranslocated form the protein, without any signal peptides exists(“CyPrP”, Figure 2) (Roucou et al., 2003). Since signal peptide cleavage is required for entry into the ER, the lack of signal peptide in CyPrP indicates that retrotranslocation must have occurred. Additionally, full-length PrP with a signal peptide can be found in the cytosol, indicating that the peptide failed to enter the ER. Retrotranslocation of PrP has been shown to be important for protection of cells against apoptotic insults (Lin et al., 2008), and disease causing mutations in PrP affect the ability of PrP to retrotranslocate and exert its protective effect (Jodoin et al., 2007, 2009). PrP has also been detected in nuclei, but the reasons and purpose of PrP in the nucleus remain unclear (Hosokawa et al., 2008). 1.2.2.1 Signal peptides PrP contains two signal peptides. The first N-terminal signal peptide is an ER signal, which recruits PrP to the translocon for translocation into the secretory pathway. This peptide is not considered 'strong', explaining why sometimes PrP fails to enter the secretory pathway (Rane et al., 2004). The C-terminal signal peptide is a GPI anchor signal, targeting PrP to GPI rafts for secretion (Caughey et al., 1989). These signal peptides are cleaved during maturation of the prion protein (Roucou et al., 2003). Misiewicz 22 1.2.2.2 Secretion PrP's C-terminal GPI signal sequence is cleaved by the enzyme GPI protein transamidase, and a glycolipid complex is added to the new C-terminus of the protein. This glycolipid anchor remains embedded in the phospholipid bilayer, and when attached to a protein may indicate protein maturity and readiness for ER exit. The GPIanchored protein is trafficked to the Golgi in a mechanism which is independent of COP-I and COP-II vesicles normally used for transmembrane protein exocytosis. The exact details remain unclear, but once PrP reaches the Golgi, it is secreted like other GPIanchored proteins (Harris, 2003). Once PrP reaches the cell surface, it tends to accumulate in lipid rafts with other GPI proteins (Caughey et al., 1989; Campana et al., 2005). 1.2.2.3 Post-translational modifications Human PrP exists in several post-translationally modified states. A disulfide bridge is present between residues 179 and 214. In addition, PrP is detectable in various stages of the glycoprotein maturation process (i.e. immature glycosylated, and mature) (Haraguchi et al., 1989). These modifications can occur at residues 181 and 197. It appears that glycosylation may affect PrP's ability to form aggregates, although this remains unclear (Lawson et al., 2005). Misiewicz 23 A PrP, full length Disulfide bridge TMD Signal peptide GPI anchor 179 N 22 112 181 197 214 136 C 231 253 Polysaccharide Side chains CyPrP TMD 179 N 181 197 214 C 23 112 136 231 23 B PrP Lumen PrPNtm PrPCtm 231 GPI 231 23 GPI 23 Cytosol 231 N CyPrP C Figure 2 – Topology and transmembrane forms of PrP. A. Schematic diagram of PrP isoforms, both cytosolic and secreted forms. B. Localization of PrPCtm, PrPNtm, normal PrP and CyPrP. Misiewicz 24 1.2.3 Protective functions Prion protein appears to be important in protecting cells against apoptotic insults (Brown et al., 1999; Kuwahara et al., 1999; Bounhar et al., 2001; Roucou et al., 2003; Diarra-Mehrpour et al., 2004). There appear to be two ways in which this occurs. The first is by reduction of reactive oxygen species (ROS). PrP has been observed to possess superoxide dismutase activity, which may be important in neutralizing reactive oxygen species in cells (Brown et al., 1999). PrP also protects against oxidative stress. When PrP deficient cells are subjected to various oxidative stresses, their survivability is reduced (Shyu et al., 2004; Brown et al., 1997b; Wong et al., 2001). PrP's other protective effect is to exert a negative effect on the pro-apoptotic protein Bax. This activity is impaired when PrP contains disease causing familial prion disease mutations, possibly explaining the mechanism for neurodegeneration in familial scrapie diseases (Bounhar et al., 2001; Jodoin et al., 2007, 2009). Finally, PrP has been shown to protect cells against TNFα mediated apoptosis, which possibly implicates PrP in cancer progression (DiarraMehrpour et al., 2004). 1.2.3.1 Anti-Bax function Bax is a 218 aa protein that oligomerizes at the surface of mitochondria and ultimately allows membrane permeabilization leading to caspase dependent cytochrome-c dependent apoptosis (Oltvai et al., 1993). In order to cause cytochrome c release, Bax must oligomerize at the mitochondria surface to solubilize the membrane (Antonsson et al., 2000). Control of Bax is therefore very tight, normally regulated by ubiquitination and proteolysis. PrP's role in Bax activation first came to light when it was noted that the octapeptide repeat region of PrP contains a Bcl2 homology (BH2) domain (Yin et al., Misiewicz 25 1994; Wang and Snyder, 1998). Additionally, in S. cerevisiae, which has no endogenous PRNP or Bax equivalent, PrP rescued against Bax-induced cell cycle arrest and toxicity (Bounhar et al., 2006). Through an unknown partner, PrP appears to indirectly regulate Bax activity. Only mature CyPrP has this effect (Lin et al., 2008). In the presence of CyPrP only (i.e. not other forms), Bax shows reduced oligomerization activity (Roucou et al., 2005). Importantly, PrP that contains GSS, FFI or CJD mutations shows reduced or eliminated Bax inhibition, suggesting one possible way in which these mutations can lead to disease (Jodoin et al., 2007, 2009). The unknown factor that binds with CyPrP to inhibit Bax oligomerization remains unknown. However, by using expoxymicin to inhibit protein retrotranslocation or by introducing a known familial disease-causing mutation to PrP, anti-Bax activity can be eliminated, implicating CyPrP specifically in the protective process (Jodoin et al., 2007). 1.2.3.2 Copper metabolism and redox activity ROS are thought to be one of the major causes of aging and damage in cells. ROS have been implicated in oxidization of organic molecules, DNA damage, and cellular dysfunction. Therefore it is important that cells have mechanisms in place to reduce or control ROS, particularly due to their production during aerobic respiration (Lenaz, 2012). Many metal ions produce ROS if not properly chaperoned. Cu2+ is one such ion, and having a system to chaperone and sequester these ions is important (Lenaz, 2012). In vitro and in vivo, PrP has been shown to be able to reduce Cu2+ ions. The octapeptide repeat region of the PrP has been shown to bind copper, and this region of the protein is one of the best conserved regions of the protein (Brown et al., 1997a; Stöckel et al., 1998; Brown et al., 1999). Based on homology and alignment studies, PrP and its related family members Shadoo and Doppel are thought to have descended from ion chaperones (Schmitt-Ulms et al., 2009). Due to its reductase activity and homology, it is thought this is one of PrP's major functions. Misiewicz 26 2 The endoplasmic reticulum: functions and properties The ER is the organelle of the cell where secreted proteins are folded and processed into their final functional confirmations. In addition to its function in protein folding, the ER also has the following critical functions: providing appropriate enzymes for protein folding, being a sensitive sensor for errors in this process and mediating disposal of proteins which have failed to properly assemble. Three major classes of enzymes enable protein folding: chaperones, foldases and lectins (Schröder and Kaufman, 2005). The most well known protein folding enzymes are chaperones, which provide a hydrophobic folding environment for client proteins. They allow the protein to more rapidly cycle through all possible configurations in space. Foldases are ATP-dependent enzymes that accelerate protein folding by accelerating covalent modifications that must take place during protein maturation (such as disulfide bridge formation). Lectins “read” glycoproteins' carbohydrate side chains and direct them to the Golgi or retain them in the ER if they need more time to fold. If a protein remains attached to quality control enzymes for too long, this will signal the activation of the protein disposal mechanisms. A pathway called Endoplasmic Reticulum Associated Decay (ERAD) is able to recognize proteins that have remained associated with folding enzymes for too long, and removes them to the cytosol where they are subject to proteasomal degradation. In addition to folding enzymes, the ER contains transferases which induce the binding and processing of complex carbohydrates as proteins mature, allowing secreted glycoproteins to attain their correct forms. Glycoproteins receive their immature carbohydrate side chains in the ER, and are then released to the Golgi for cleavage and maturation. Failure to achieve proper glycosylation results in retention in the ER and ultimately activation of ERAD. Finally, the solute content of the ER is carefully controlled by the organelle, in order to assure proper ion concentrations and the efficient production of secreted proteins. In addition to its role in protein synthesis and ion concentration, the ER is the site of sterol and lipid synthesis, about which very little is known, particular in regards Misiewicz 27 to ER stress. Finally, the ER is an intracellular store of calcium ions, which are needed in many cellular functions in or out of the ER. 2.1 Protein folding Protein function is determined by its shape, thus in order to assure function, structure must be assured. In order to reach these structures, cells employ enzymes which provide the proper conditions to catalyze the needed bonds. There are three properties of a protein that give rise to its final conformation. First, is its primary structure, which is determined by its amino acid sequence. Second, a protein's secondary structure consists of beta sheets, alpha helices and coiled coils, structures which are formed due to hydrogen bonding between the various amino acids. Third, the tertiary structure which is more complex and represents how these structures all interact together to form a larger globular protein. 2.1.1 Mechanisms of enzymatic activity for protein folding in the ER The three major classes of protein folding enzymes in the ER. Each kind of enzyme has a different mode of action to assist client proteins assume their correct confirmations. The best known class of protein folding enzymes are the chaperones, and best known of these is BiP (GRP78), although many have since been discovered. Next there are the foldases, and the best known foldase in the ER is Protein Disulfide Isomerase (PDI) which helps establish disulfide bridges between cysteine amino acids. Finally, there are the lectins, and the best known lectins are Calnexin and Calreticulin. The chaperone BiP is an ATP dependent enzyme. ADP-BiP has a high affinity for unfolded client proteins and binds them. In this confirmation, the ADP is exchanged for Misiewicz 28 an ATP, and BiP becomes tightly attached to the client protein which is now able to undergo a folding step. Finally, the ATP is hydrolyzed to ADP, and BiP returns to a lower affinity state for the client protein. This cycle causes protein folding to become an ATP dependent process, and as the protein folds, ATP is consumed. Depleting ATP can inhibit protein folding. The purpose of BiP activity is to provide for greatly accelerated protein folding, because protein folding steps proceed faster when BiP is present (Flynn et al., 1989). Lectins are responsible for ensuring the quality of glycoproteins. The best known lectin proteins are Calnexin and Calreticulin, which bind to newly-glycosylated proteins. While the carbohydrate remains immature (not cleaved to a smaller, more mature form), it is deglycosylated and re-glycosylated and retained by calnexin and calreticulin. Several cycles of this occur, which allows for sufficient folding to take place before cleavage to the mature form occurs. Furthermore, calnexin and calreticulin have the important function of keeping unfolded proteins out of the Golgi, since immature glycoproteins remained retained until proper cleavage occurs (Schröder and Kaufman, 2005; Wang et al., 2012a). Finally, there are foldases. The best known foldase is protein disulfide isomerase (PDI), which catalyzes the formation of a disulfide bridge between two cysteine residues. This requires the formation of a covalent bond between the two cysteines to form cystine. Bonds must be formed between the correct cysteines, and reduction of -SH groups on the cysteine amino acid must take places. PDI ensures that all of these activities are carried out correctly and efficiently. Since the discovery of PDI, other additional foldases have been discovered with various functions (Wilkinson and Gilbert, 2004). 2.1.2 Post-translational modifications and sorting The ER and Golgi are the sites of post-translational protein modification and sorting to their destinations. GPI anchor attachment occurs in the ER, which leads to anchored Misiewicz 29 proteins being targeted to the cell surface. Immature glycosylation occurs in the ER as well, where nitrogen (typically on an asparagine residue) and oxygen (typically on serine or threonine residues) are covalently linked to large carbohydrates. In the Golgi apparatus, these glycans are trimmed, indicating protein maturity. Other modifications such as protein sialylation can occur in the Golgi as well. Following the appropriate glycosylation, proteins exit the secretory pathway for their final destinations (Alberts et al., 2008). 2.2 ER Stress Fundamentally, stress is the response to a perturbation in the environment. The response to stress occurs as the cell responds to that pertuburation to restore optimal functionality: either by adjusting the environment or altering its own functional properties to function most efficiently. Restoration of optimal functionality constitutes the exit from stress. In the context of the ER, stress impairs the ability of the organelle to complete its function. In order to respond to this change, a complex program has evolved in mammalian cells that enables the restoration of homeostasis. 2.2.1 Discovery and definition ER stress is vaguely defined. It was originally noticed that cells upregulated certain proteins when deprived of oxygen or glucose, or in normal tissue that secreted large amounts of protein. This led to the discovery of chaperones, and the idea that high expression of chaperones constitutes a stress (Haas and Wabl, 1983; Munro and Pelham, 1986; Hetz, 2012). Generally, ER stress is defined as an excess of unfolded proteins in the ER, which results in higher than normal chaperone activity. Misiewicz 30 2.2.2 ER Stress signal transduction The cellular response to ER stress entails increasing folding enzyme (e.g. chaperone) activity and altering protein expression levels, both transcriptionally and translationally (Hetz, 2012). However, the status of the contents of the ER cannot be easily signaled to the rest of the cell, since the ER is a membrane bound, closed structure. As a result, unique mechanisms are needed to inform the rest of the cell of the current status of the ER. Three pathways have evolved to transmit the ER stress signal across the ER membrane. All of them are dependent on the level of BiP associated to transmembrane sensors, which is lower when BiP is bound to an excess of unfolded client proteins. Therefore, BiP acts as both a measurement of the amount of unfolded proteins, as well as the tool to reduce the pool of unfolded clients. 2.2.2.1 IRE1α and XBP1 The first transducer of ER stress signals discovered was Inositol Requiring Enzyme 1 (IRE1α) and the transcription factor X Box Binding Protein 1 (XBP1). XBP1 was initially discovered in yeast as Hac1, where it is the only transducer of ER stress mediated gene transactivation; the mammalian version was discovered by homology (Mai and Breeden, 1997). When high levels of unfolded client proteins are present in the ER, BiP dissociates from a transmembrane splicing enzyme Inositol Requiring Enzyme 1 alpha (IRE1α). When IRE1α is no longer associated with BiP, it homodimerizes and autophosphorylates, which results in the activation of endonuclease activity (Li et al., 2010). IRE1α excises a 26bp intron from the XBP1 mRNA, which removes a stop codon. The shortened message encodes the highly active transcription factor sXBP1, which translocates to the nucleus where it acts as a transcription factor for ER stress mediated transactivation (Yoshida et al., 2001). In addition to its endonuclease activity, IRE1α also has the ability to activate JNK signaling through the adapter protein TRAF2. Through this mechanism, IRE1α appears to be able to regulate signaling for apoptosis and Misiewicz 31 autophagy (Hetz and Glimcher, 2009). 2.2.2.2 Site 1 and Site 2 proteases and ATF6 ATF6 was the second ER stress transcription factor identified after yeast Hac1p. It exists in two different isoforms, a longer 90 kDa isoform that is retained in the ER, and a shorter 55 kDa isoform which can be found in the nucleus as a highly active transcription factor. In conjunction with the general transcription factor NF-Y, it binds to DNA and is able to recruit the needed transcription machinery, resulting in high levels of transactivation during ER stress (Haze et al., 1999). In order to be converted from the longer ER resident form to the soluble 55 kDa form, ATF6 is cleaved by two transmembrane Golgi-resident proteases Site 1 Protease and Site 2 Protease (Ye et al., 2000). Normally, ATF6 is retained in the ER, bound to BiP. However, as with XBP1, when BiP dissociates to fold client proteins, ATF6 is able to advance in the secretory pathway to the Golgi. Upon reaching the Golgi, ATF6 is cleaved by Site 1 Protease (S1P) and Site 2 Protease (S2P), resulting in a 55 kDa fragment and a 36 kDa fragment which are able to activate transcription in the nucleus (Yoshida et al., 2000). In non stress conditions, ATF6 is normally fully glycosylated, due to three wellconserved glycosylation sites in its lumenal half. However, during disruptions to ER homeostasis, ATF6 becomes under glycosylated. The less glycosylated form of ATF6 loses affinity for calreticulin, which normally retains ATF6 in the ER during non-stress conditions. No longer retained by calnexin/calreticulin, ATF6 is transported more quickly to the Golgi, where it can be cleaved by S1P and S2P. Thus, ATF6 is also able to act as a sensor for protein glycosylation: improper or impaired protein glycosylation in the ER increases the transcriptional activity of ATF6 by making more of it available for cleavage activation (Hong et al., 2004). Misiewicz 32 2.2.2.3 PKR-like endoplasmic reticulum kinase and ATF4 The final ER stress mediated transcription factor is ATF4. Similar to XBP1, ATF4 is regulated on a translational level. Its mRNA is not transcribed under normal conditions (Vattem and Wek, 2004). Upon activation of the ER stress response however, ATF4 translation is initiated and the active transcription factor is able to transactivate gene expression in the nucleus. During ER stress, cells attempt to reduce the ER load by attenuating the rate of new protein translation, which allows the chaperones, foldases and lectins to clear the unfolded proteins from the ER. In order to achieve this, a transmembrane serine threonine kinase known as PKR-like endoplasmic reticulum kinase (PERK) phosphorylates the translation initiation factor eIF2α, which arrests nearly all translation in the cell (Harding et al., 2000). Similar to the other ER stress sensors, PERK is normally bound to BiP when the ER does not contain an excess of unfolded proteins. During ER stress, BiP dissociates from PERK in favor of client proteins. PERK dimerizes and autophosphorlates, which causes PERK to gain kinase activity (Bertolotti et al., 2000). When eIF2α is phosphorylated, the majority of protein translation is inhibited due to a failure of the ribosome to find the initiation site. However, messages containing a second open reading frame after the first are efficiently translated when eIF2α is phosphorylated. ATF4 mRNA contains one of these sites, resulting in high translation of ATF4 mRNA, the active transcription factor, during ER stress (Vattem and Wek, 2004). 2.2.3 Description of the ER stress response program The cell's response to ER stress is both transcriptional and translation. Fundamentally, Misiewicz 33 the goal of the cell's response to ER stress is to reduce the load of unfolded proteins by attenuating translation, and by properly processing client proteins in the ER. Thus, the genes that are regulated transcriptionally and translationally in the ER stress response help achieve these functions. If the response to ER stress is not sufficient to clear the blockage, cells activate an apoptotic program to initiate caspase-dependent cell death (Hetz, 2012). In addition to its role in protein folding, ER stress can also affect the output and processing of other ER resident molecules, such as lipids and ions, which are produced or processed in the ER. During ER stress, translation of most new proteins is inhibited by the cell due to phosphorylation of eIF2α. However, there are several mRNAs which avoid this translational arrest by containing an additional special ORF. The best example of this is the ATF4 mRNA, which contains two ORFs. The first contains a sequence which prevents appropriate attachment of translation initiation factors when eIF2α is not phosphorylated. However, when eIF2α is phosphorylated, the first inhibitory ORF is not recognized during ribosomal scanning of the mRNA, and instead a second, downstream ORF is recognized and translated. This mechanism allows the cell to impair translation of most new client proteins, while increasing the level of chaperones and transcription factors needed to restore ER homeostasis (Vattem and Wek, 2004). 2.2.3.1 Translational Protein chaperones are amongst the molecules that are most induced by ER stress. Originally, protein chaperones were identified as proteins which appeared on gels when cells were either deprived of glucose or subjected to heat shock (Munro and Pelham, 1986). It was also discovered that certain toxic chemicals (such as a tunicamycin) would also cause Grp78 induction, leading to the idea that this protein did more than just respond to heat and glucose deprivation. Independently, BiP was identified as a protein in the ER of B-cells, and it was identified as a binding partner of secreted proteins (Haas and Wabl, 1983). Eventually, cloning and mapping of BiP and Grp78 lead researchers to Misiewicz 34 the realization that they were actually the same protein (Munro and Pelham, 1986). In addition to BiP, other proteins increased during ER stress have been identified. These include ER enzymes such as HERP, PDI6, Calnexin/Calreticululin, and Grp94. Additionally, cells upregulate the expression of nutrient transporters, such as amino acid transporters and sugar transporters in order to help synthesize proteins and the appropriate glycans for their correct modification (Oyadomari and Mori, 2004; Hetz, 2012). 2.2.3.2 Transcriptional By way of the above mentioned transcription factors, cells are able to activate a diverse transcriptional program. The mRNA levels of chaperones are strongly increased during ER stress due to an increase in chaperone gene transcription levels (Munro and Pelham, 1986). Following excessive ER stress signaling, and especially if IRE1α signaling becomes diminished and PERK signaling increases, pro-apoptotic genes start to be transcribed, activating a cell death program (Oyadomari and Mori, 2004; Hetz, 2012). 2.2.4 Regulation of ER stress mediated transcription ER stress-mediated transcription in higher eukaryotes is regulated by a series of DNA promoter motifs that transactivate gene expression. Originally identified by examination of the promoters of ER stress responsive genes, these ER Stress Response Elements (ERSE) have been identified in many locations throughout the genome and have several different mechanisms of activation. Misiewicz 35 2.2.4.1 ERSE The first ERSE was identified in 1998 by examining the promoters of several ER stress regulated genes (HSPA5, HERP and PDI6). It contains the sequence CCAAT-N9CCACG, where the middle nucleotides are non-specific (Yoshida et al., 1998). It can function in either a forward or a reverse orientation. The element was identified in a one hybrid screen, which identified the transcription factors ATF6 and XBP1 as being the transcription factors able to activate this element (the ATF6-NF-Y complex recruits the general transcription factors needed to initiate transcription). The spacing between CCAAT and CCACG appears to be important, mutation experiments showed that changing the variable region to either 8 or 10 nucleotides resulted in attenuation of transcriptional activity (Yoshida et al., 1998). The general transcription factor NF-Y binds the CCAAT part of the ERSE, while the transcription factor ATF6 binds the CCACG part. The cleavage of ATF6 and its translocation to the nucleus is dependent on ER stress, meaning that the CCACG section provides transcriptional specificity for ER stress (Yoshida et al., 2000). XBP1 is able to bind to the ERSE, and was identified in the original yeast one hybrid screen, however in vivo it does not appear to be responsible for the majority of ERSE mediated transcription (Yoshida et al., 1998; Yamamoto et al., 2004). 2.2.4.2 ERSE II Following the discovery of the ERSE, another ERSE element was discovered. The sequence is ATTGG-N-CCACG, and was found by investigating the HERP promoter. The non-conserved region in the middle is shorter, with only one nucleotide, and the ATF6 binding site is inverted to the opposite strand. Like the classical ERSE, the ERSE-II is bound by the transcription factors ATF6 and NF-Y. It initiates transcription using a similar mechanism (Kokame et al., 2001). XBP1 is also able to bind the ERSE-II element. Additionally, and unlike the classical ERSE, removing XBP1 attenuates, but does not entirely eliminate, the Misiewicz 36 transcriptional activity of the ERSE-II (Yamamoto et al., 2004). 2.2.4.3 ERSE-like A third ERSE-type element identified is the so-called ERSE-like element. This element contains the sequence CCACG-N9-ATTGG. This element is similar to the classical ERSE, except in this case, the NF-Y binding site is on the opposite strand, and the orientation of the element is reversed. Furthermore, this element is dependent on XBP1 binding, but not ATF6. This element was confirmed to regulate the ER-related gene WFS1, though it is thought to work for other genes as well (Kakiuchi et al., 2006). 2.2.4.4 UPRE The Unfolded Protein Response Element (UPRE) is different than the ERSE. Its sequence is TGACGTGG/A, and is bound by XBP1 (Yoshida et al., 2001; Yamamoto et al., 2004). It is known to regulate the expression of the gene HERP. Initially, it was thought that ATF6 would have a high affinity for this element due to its similarity to the ERSE (on the opposite strand), but that has not proven to be the case; XBP1 activates this element. 2.2.5 Exit from ER stress Exiting from ER stress occurs in one of two ways. If ER homeostasis is restored, the Misiewicz 37 transcription and translation of ER stress will subside, and BiP will re-associate with the sensors ATF6, IRE1α and PERK. The alternative is apoptosis, which induced by the ER stress-regulated genes CHOP (GADD153). GADD34 is an eIF2α phosphotase which helps restore translation when the cell exits ER stress. 2.2.5.1 Restoration of ER homeostasis Attenuation of ER stress signaling appears to indicate a return to ER homeostasis. With proper BiP binding, ER stress sensors become inactivated. Importantly, it has been noted that when IRE1α remains active during ER stress, cells avoid apoptosis and restore homeostasis better. In contrast, prolonged activation of PERK due to ER stress appears to have the opposite effect, preventing a return to normalcy, and inducing apoptosis. However, the cell can avoid this outcome if IRE1α signaling is maintained rather than PERK signal transduction (Lin et al., 2009). 2.2.5.2 Activation of apoptosis Prolonged ER stress signaling eventually leads to apoptosis (Hetz, 2012). After long periods of ER stress, the transmembrane sensor transmitting the ER stress signal shifts from IRE1α to PERK. PERK causes transcription of ATF4, and ATF4 activates expression of GADD153/CHOP (Oyadomari and Mori, 2004). Some evidence suggests CHOP protein downregulates B-cell lymphoma 2 protein (Bcl-2) activity, which is a major inhibitor of Bax oligomerization (McCullough et al., 2001). Following a CHOP-mediated removal of Bax inhibition, cytochrome-c is released and caspase-dependent apoptosis is activated. PERK appears to be activated later in ER stress, consistent with a mechanism for to protect the organism from unhealthy cells if earlier protective responses fail to restore cell homeostasis. Misiewicz 38 2.2.6 ER Stress and Disease Accumulating evidence is beginning to show that ER stress plays a role in many diseases. ER stress is implicated in scrapie diseases, due to the high load of unfolded scrapie proteins in cells. Accumulation of PrPSc has been observed in the ER, as well as an increase in BiP levels in CJD patients (Hetz and Glimcher, 2009). Furthermore, PrPSc from mouse brains has been shown to induce ER stress when injected in cell cultures (Torres et al., 2010). Given that scrapie diseases are characterized by plaques of insoluble proteins, the fact that these accumulate in the ER and impede function is not surprising. However, in addition to scrapie diseases, ER stress has also been observed in familial Alzheimer disease (AD) caused by mutations in the gene Presenilin 1 (Doyle et al., 2011). An excess of Aβ protein and formation of plaques in familial AD brains has been observed. Additionally, there is evidence that BiP levels are increased in sporadic AD cases, suggesting distress in AD affected brains (Doyle et al., 2011). In addition to AD, ER stress has been identified in Parkinson's Disease, which also is characterized by protein aggregates in the brain, as well as amyotrophic lateral sclerosis (Doyle et al., 2011). Outside of neural tissue, ER stress has been implicated in Type 2 Diabetes, where it has been observed that ER stress markers (BiP, XBP1, phospho-eIF2) are upregulated in liver and adipose tissues, in conjunction with a decreased sensitivity to insulin signaling (Ozcan and Tabas, 2012). ER stress signaling has also been observed in cancer, where cells undergoing rapid division have a high demand for protein synthesis. ER stress markers BiP and XBP1 have shown upregulation in cancer, and there is a positive correlation between cancer severity and the levels of ER stress markers (Ozcan and Tabas, 2012; Carrasco et al., 2007; Fernandez et al., 2000). Due to the widespread nature of ER stress in disease, an understanding of its mechanisms and systems of control can lead to better treatments for these diseases. Misiewicz 39 3 Study of the Human Genome The human genome is nearly 3 billion basepairs long and consists of 23 chromosomes. The recent complete sequencing of the human genome in 2001 was a milestone of human achievement, although now the process of interpreting these findings is the major challenge facing scientists. Since the completion of this project, it has become clear that the large degree of variability and phenotypes cannot be simply explained by sequence variation alone, so the study of regulation of expression has become increasingly important, and many new technologies to facilitate this has come into being in recent years. 3.1 Sequencing DNA sequencing has made truly incredible advances in the last 10 years. The cost of sequencing (per nucleotide) has been decreasing as a super exponential function (Lee and Tang, 2012). Fundamentally, the process of sequencing involves extracting DNA from an organism and determining the sequences of bases by a chemical reaction that provides a unique signal for each different nucleotide (Sanger et al., 1977). Originally, this “Sanger Sequencing” was conducted using gel electrophoresis, and the sequence of nucleotides could be determined by the position of bands on a very long gel. Subsequently, improvements to classical Sanger sequencing (such as heavy use of robotics and automation and sophisticated alignment algorithms) allowed publication of the first draft of the human genome in 2001 (Lander et al., 2001). While improved Sanger sequencing offered major improvements over older methods, they were still somewhat slow due to the requirement that DNA be cloned in bacterial artificial chromosomes. Recent advances in microarray technologies have allowed for sequencing of entire genomes in very short times, by conducting millions of very short reads over the entire sequence of the organism and aligning the fragments using sophisticated alignment Misiewicz 40 algorithms (Lee and Tang, 2012). 3.1.1 The current state of the art of genome sequencing Current next generation sequencing technologies are highly sophisticated and allow many sequences to be determined in parallel (massively parallel sequencing). In modern next generation sequencing methods, DNA is sheared into various small pieces. The DNA is spotted onto a microwell plate, where it is immobilized, and reacted with Taq DNA polymerase and DNA synthesis reagents in a microwell. Alternatively, the Taq may be immobilized in the microwell, and the DNA is added. In either case, the end result is the same: on a given plate, many thousands of sequences can be read simultaneously, each spot on the plate represents a short DNA fragment. The actual sequence is read using a pyrosequencing method: Each individual nucleotide is added sequentially by an injection system to the entire plate, and a microscope with a digital camera reads the plate to determine the sequence based on following the addition of a reactive nucleotide, which produces detectable light. Once a complete series of short reads has been recorded, they are aligned to a reference genome and eventually a complete sequence can be obtained. Using these techniques a genome can be sequenced in a short matter of time, however computing complexity has dramatically increased. Another advantage to the massive parallel approach is the ability to obtain several reads over several different regions of the genome, which can help improve confidence values if a particular read did not work well for some reason. 3.1.2 The Human Reference Genome Misiewicz 41 These advances in sequencing has resulted in a concerted effort to increase coverage of the human genome. Initially, the first draft sequence was only from a few individuals, but this is problematic due to the number of polymorphisms that are present in a small group (Church et al., 2011). As a result, teams around the world have sought to sequence people from many different backgrounds and populations to gain an understanding of the “reference” genome for humans, notwithstanding an individual's variation. This project has been embodied as the Genome Reference Commission, which contains the “correct” version of the human genome (Church et al., 2011). For the work presented in this thesis, such curated reference versions were used for bioinformatic analysis. 3.2 Annotations and the definition of loci and their promoters Being able to read the human genome is one matter. Interpreting it is another. The process of understanding human genetics has often been compared to a book: first one must determine the letters, then the punctuation, then the meaning of the words. While large scale sequencing projects have solved the question of reading the letters, currently deriving knowledge from this information (i.e. understanding the syntax and structure of this language) is the challenge of our time (Lander et al., 2001). One of the ways this must be done is genome annotation, the process of assigning features to locations in the genome. A great number of cDNAs have already been deposited into nucleotide databases such as Genbank (Sayers et al., 2012). These can simply be aligned to the genome directly. However, there are a great number of genes which are not known, and two methods are being used to locate novel genes. One method is exome sequencing, where the entire RNA content of a cell is sequenced and aligned to the reference genome (a complete reference genome sequence is required for these kinds of studies, which has only recently been the case). This large-scale method is better than simply aligning previously deposited mRNA sequences to the genome because during the process of exome sequencing, new mRNAs can be discovered. However, the cell types being Misiewicz 42 studied in exome sequencing may not express all genes or in levels that allow them to be detected. Thus, the method has some bias. The other method of gene finding is predictive: a statistical model based on formal computer science theory allows for predictions of gene probability based on the emission of certain letters (i.e. TATA often indicates a gene, followed by possible splicing sites, etc.). These methods allow for computational prediction of genes which can be subsequently validated in exome sequencing projects (Ng et al., 2009). 4 Computational pattern matching algorithms in biological sequences Automated pattern matching in large data sets is one of the most useful applications in modern computing. In the 1960s, the idea of using computers to search biological data was first implemented, as previously tasks such as sequence alignments were done manually. While fundamentally the problem is quite simple, searching for large and complex patterns in complex datasets can quickly become computationally taxing. The simplest algorithm is a naïve approach where each letter in the query sequence is compared to the target sequence. Depending on the length of the query and the target sequencings, the computation time and storage requirements can become impossible. In order to overcome this limitation, algorithms such as Smith-Waterman (Smith and Waterman, 1981) and the Basic Local Alignment and Search Tool (BLAST) (Altschul et al., 1990) were developed. They each have optimizations in place to compute optimal or near-optimal alignments between query and target sequences. Although Smith-Waterman and BLAST work well for exact sequences, allowing substitutions and deviations of letters in the sequence can increase complexity and computational requirements. In order to conduct searches in targets allowing complex patterns (for example, ACT[g or c]ATCGTA), other methods are needed. Misiewicz 43 4.1 Regular expressions Regular Expressions were first proposed by a pioneer of computer science, S. C. Kleene (Kleene, 1951). Although at first his paper did not make a very big impact, in it he defined a framework for representing a string of text as a series of “emission” events froma finite state automaton (FSA) (Russell and Norvig, 2010). A FSA is an abstract machine: given a set of arbitrary inputs (which could be a text string, sequence, pattern or any sort of information) the FSA produces an output according to its structure and the probability of certain outputs. Finite state automata can be useful if we want to determine the parameters of a process (for example, the frequency of a certain nucleotide being emitted in a sequence of nucleotides). To achieve this, we make the assumption that a given sequence is output from a FSA, and then use this output to estimate of the parameters of the FSA that must have produced it. Alternatively, one can define the machine and its parameters ahead of time (for example: the FSA emits [A or B] then [C or D]), and determine if a given input string could have been produced by this FSA. The use of FSA to model emission probabilities allows the application of several useful mathematical theorems which can dramatically optimize the speed of text searching. More specifically, when using a FSA for pattern matching, the output to test (i.e. the text to search) can be considered in an arbitrary order. Inputting a user-specified FSA specification allows for implementation of efficient search algorithms (which perform much more quickly than a naïve search of all possible strings that satisfy the pattern). An algorithm implementing Regular Expression based search and pattern matching was first written by the pioneering computer scientist Ken Thompson (Thompson, 1968). His algorithm or derivatives of it have subsequently been implemented in a great number of computer programs that contain abilities for advanced text processing. The approach of Thompson's regular expression evaluation algorithm provides dramatic improvements over the naïve approach due to optimizations. It first generates the FSA corresponding with the user inputted pattern. Next, this FSA is used to produce a table of all possible characters at a given position in the query string, an operation Misiewicz 44 which can be done very quickly. Then, these tables are used to search the target string, by simply comparing each character of the target text to the tables. The naïve approach of searching, comparing each position of the target and query texts of length n and m, requires mn steps in the worst case. Using regular expressions, only n operations are needed to search for a pattern of length m in n. Thus Regular Expression search is an order of magnitude faster than naïve search and allows for great fexibility of patterns, which would not be possible with naïve search due to the enormous cost in computation time and storage space. 5 Methods of detecting gene expression levels In order to study the transcriptional state of the cell, several methods have been developed to assess mRNA identity and abundance. One is Northern Blotting (Alberts et al., 2008), where an oligonucelotide probe is hybridized directly to the mRNAs immobilized on a membrane (typically after transfer from a gel). This provides a very sensitive method for determining RNA concentration and identity (since the probe is made by the researcher). Alternatiely, reverse transcriptase (which makes cDNA from RNA) be used to produce template DNA for polymerase chain reaction (PCR). DNA from PCR can be run on a gel and used for semi-quantitative analysis. Once the image is stored in a computer, pixel intensity values can be computed for the image and used to compute the intensity of the band. The resultant measurement can be normalized against controls to establish semi-quantitative fold induction values. While both of these approaches are functional, there is a better more precise method. 5.1 Quantitative Real Time PCR Although PCR on DNA samples can be quantitative, the method is only relative. Misiewicz 45 Furthermore, agarose gel electrophoresis depends on a DNA staining agent to enable detection and identification of products from the PCR reaction. However, the sensitivity and linear response of the dye used (commonly Ethidium bromide) must be taken into account. Furthermore, ethidium bromide has a comparatively low sensitivity compared to newer dyes such as SYBR Green. This means that in order to identify the amplicon on a gel, many cycles of amplification must take place. This leads to problems however, since PCR is an exponential reaction, after a sufficient number of cycles, reactants can be come depleted and amplification can plateau. Thus, with sufficient cycling, differences in sample concentration can be masked because all of the reactions will have “caught up” to each other. The solution to this problem was proposed by Morris and Morrion (Morrison et al., 1998; Morris et al., 1996). Instead of detecting the double stranded output of the PCR reaction after all thermal cycles have completed, a dye is placed in the PCR reaction itself, and a fuorometer detects fuorescence above background levels (meaning double stranded product has been produced) in the cycler. As soon as fuorescence exceeds background levels, the cycle number is noted, and the fold abundance can be determined by extrapolating from a logarithmic curve (Pfaff, 2001). This method allows for very precise quantitation of PCR amplification, although primers must be designed very carefully and well-characterized to prevent multiple products. Unlike conventional RT-PCR, qPCR cannot be used to quantify multiple size amplicons in the same reaction. Furthermore, complex equipment, expensive dyes and analysis are required for interpretation of qPCR results. 6 Working hypothesis and research objectives The culmination of previous evidence has lead to the theory that PrP is protective against stresses and insults. This is supported by PrP's demonstrated anti-apoptotic function (by inhibiting Bax), the vulnerability of Prnp knockout cells to insults, and PrP's demonstrated redox activity. Furthermore, mutations in prion that cause neurodegenerative disease inhibit this activity. Misiewicz 46 In concert with PrP's protective role, preliminary evidence in our lab indicates that PRNP and PrP are upregulated in response to ER stress in neurons and MCF-7 cells (Jodoin et al. Unpublished). Jodoin et al. also identified two ERSEs in the PRNP promoter as well as a novel ERSE which has similarity to the previously identified ERSEs. This novel element, called ERSE-26, appears to be able to regulate PRNP expression, based on a luciferase reporter construct and mutagenesis experiments. In the same paper, Jodoin et al. showed with Chromatin Immunopreciption (ChIP) that the PRNP promoter can bind both ATF6 and XBP1, but which of the three ERSEs binds which transcription factor is not known. Although the ERSE-26 is present and functional in the PRNP promoter, its existence and functionality elsewhere in the genome remain to be seen. Additionally, the activating transcription factor is not known. In order to address these questions, using bioinformatic tools and the human reference genome (Church et al., 2011), I conducted a genome wide scan of all promoters to identify other ERSE-26 elements. After finding other gene promoters which contain the ERSE-26, I conducted RT-PCR and qPCR on neurons and MCF-7 cells treated with the pharmacological ER stressors Brefeldin A (BFA), Tunicamycin (TM) and Thapsigargin (Thps). These three stressors induced expression of some ERSE-26 genes, confirming its existence and function elsewhere in the genome. Next, I produced a luciferase reporter construct containing the ERSE-26, which when co-transfected with either sXBP1 or ATF6 allowed for the identification of the activating transcription factor. The factor sXBP1 was identified as the transcription factor activating the ERSE-26. In order to confirm this, we transfected MCF7 cells with sXBP1 to see whether or not ERSE-26 containing genes were induced or repressed. The promoter of PRNP is poorly understood, and understanding the mechanisms of PRNP expression and control will lead to a better understanding of its broader function in the context of the cell. Furthermore, by understanding what is co-regulated with PRNP, PrP's role in the cell can be better elucidated. Finally, the role of ER stress in disease is currently being recognized in a wide array of diseases; having a better understanding for these pro-survival responses can help improve treatments for these diseases. Misiewicz 47 Misiewicz 48 Chapter 2 – Manuscript Preface This chapter is the manuscript to be submitted to PloS Genetics: Michael Misiewicz, Julie Jodoin, Derek Ruths, Andréa C. LeBlanc (2012). Identification of a Novel Endoplasmic Reticulum Stress Response Element regulated by XBP1. Author Contributions Michael Misiewicz conduced all experiments and writing, with the exception of the western blots in Figure 4D which were done by Julie Jodoin. Derek Ruths and Andrea LeBlanc were supervisors. Human primary cultures were obtained from The University of Washington, and I wish to thank Vikas Kushal for his assistance in culturing them. Misiewicz 49 Identification of a novel Endoplasmic Reticulum Stress Response Element regulated by XBP1 Misiewicz 50 Introduction The endoplasmic reticulum (ER) is the organelle where all proteins secreted by the cell are folded. In addition to helping route proteins to their proper destinations, the ER is also critical in ensuring that proteins assume their proper confirmations. In order to assure this, eukaryotes have evolved complex mechanisms for quality control in the ER (Hetz, 2012). PrP is a secreted 26 kDa glycosyl phosphatidylinositol (GPI) anchored glycoprotein encoded by the PRNP gene (Westergard et al., 2007; Linden et al., 2008). PrP has been linked to many different important cellular processes, and it is the causative agent of several transmissible, familial and sporadic spongiform encephalopathies, for which it was originally discovered and named (Prusiner, 1982). In addition to its original discovery as the causative agent of scrapie disease, normal cellular prion (PrP C) is known to be involved in copper metabolism, synapse function, cancer and oxidative stress (Brown et al., 1997a; b; Wong et al., 2001; Stöckel et al., 1998; Haeberlé et al., 2000). Additionally, PrP plays a cytoprotective role in neuronal cells (Kuwahara et al., 1999), and can be retrotranslocated from the secretory pathway to the cytosol to exert prosurvival activity by inhibiting the pro-apoptotic protein Bax (Jodoin et al., 2007; Roucou et al., 2003, 2005; Jodoin et al., 2009). Prion diseases are characterized by aggregation of aberrantly folded PrP, which is able to catalyze the conversion of properly folded PrP into mis-folded isoforms (Prusiner, 1982). As a result of this process, the cell experiences ER stress and activates the Unfolded Protein Response (UPR) (Hetz, 2012). The UPR is a well conserved process that cells use to restore ER homeostasis or enter apoptosis. In higher eukaryotes, three sensors of ER status exist. The Inositol Requiring Enzyme 1 (IRE1α) is a transmembrane ribonuclease that splices and activates X-Box Binding protein (XBP1) mRNA when the foldase BiP (Grp78, encoded by HSPA5) dissociates from IRE1α and preferentially attaches to misfolded client proteins. (Yoshida et al., 2001). Additionally, IRE1α is able to activate JNK signalling through the adapter protein TRAF2 (Hetz and Glimcher, 2009). Concurrently, the transmembrane protein Activating Transcription Factor 6 (ATF6α), which is Misiewicz 51 normally bound by BiP, is released upon BiP’s preferential attachment to client proteins. ATF6α translocates to the Golgi complex, where it is cleaved by transmembrane Site 1 and Site 2 proteases to form a highly active transcription factor (Haze et al., 1999). The third branch of ER stress signal transduction is mediated by the transmembrane PKRlike ER Kinase (PERK), which upon BiP dissociation is able to phosphorylate eukaryotic translation initiation factor 2 (eIF2α). Phosphorylation of eIF2 arrests translation of most messages, except for several with upstream activating open reading frames, such as the Activating Transcription Factor 4 (ATF4) mRNA (Hetz, 2012; Vattem and Wek, 2004). Following ER signal transduction, spliced XBP1 (sXBP1), cleaved ATF6 (ΔATF6α) and ATF4 translocate to the nucleus where they are able to activate the expression of ER stress responsive genes. ΔATF6α, sXBP1 or ATF4 interact with several promoter DNA motifs called the ER Stress responsive element (ERSE), resulting in elevated expression of genes containing this element (Yoshida et al., 2000, 1998; Kakiuchi et al., 2006). Originally identified upstream of HSPA5 and the ER-related genes HERP and PDI6, the classical ERSE contains the consensus sequence CCAAT-n9-CCACG. It was shown in a yeast one hybrid screen that this element is able to bind the transcription factors ΔATF6α and XBP1 (Yoshida et al., 2000). Yoshida et al. showed that ΔATF6α binds to the CCACG portion of the ERSE, while the general transcription factor NF-Y binds the CCAAT part (Yoshida et al., 1998). sXBP1 is also able to bind CCACG (Yoshida et al., 1998; Lee et al., 2002). Removing the CCACG portion or altering the variable region to 8 or 10 nucleotides reduces or eliminates the ER stress-mediated transcriptional response. The ERSE is able to function in either a forward or reverse orientation. Subsequently, several other variations on the ERSE have been identified, such as the ERSE-II, which contains the sequence ATTGG-N-CCACG and binds either ΔATF6α or sXBP1, and the ERSE-like, with an inverted NF-Y binding site (Yamamoto et al., 2004; Kakiuchi et al., 2006). Previously, it has been shown that the PRNP promoter contains several ERSEs as well as a novel ER stress response element, called the ERSE-26 (Jodoin et al., unpublished). This element contains the same sequence as the classical ERSE, except the 9-nucleotide variable region is replaced with a 26-nucleotide variable region (giving CCAAT-n26CCACG). Expression of this element in a luciferase reporter system showed induction Misiewicz 52 when cells were subjected to ER stress, and mutagenesis removed this. Furthermore, Jodoin et al. demonstrated that PRNP expression is correlated with cancer severity and ER stress (as determined by BiP), suggesting that ERSE-26 may play a pro-survival role in ER stressed cells. Jodoin et al. demonstrated XBP1 and ATF6 binding to the PRNP promoter, but they did not show which transcription factor activates the ERSE-26, and the existence and functionality of the ERSE-26 elsewhere in the genome remains unconfirmed. In order to gain insight into the broader cellular context of this novel element and to determine whether it is specific to PRNP or not, we conducted a bioinformatic analysis of the complete human reference genome (Church et al., 2011) using a Regular Expression model of the ERSE-26. Following the identification of genes whose promoters contained this element within -2,000 bp or +500 bp from the transcription start site, reverse transcriptase PCR (RT-PCR) and quantitative RT-PCR (qPCR) were conducted to assess the fold changes in ERSE-26 genes in cultured primary human fetal neurons and MCF-7 cells treated with pharmacological ER stress inducers. Additionally, we produced a luciferase reporter construct to identify which transcription factors are responsible for ERSE-26 activity. We identified that sXBP1 transactivates the ERSE-26 in a luciferase reporter assay. Finally, an extensive literature search was conducted to identify which pathways and processes are affected by ERSE-26 regulation. We identified ERSE-26 genes involved in the oxidative stress response, neurotransmitter metabolism, ER homeostasis and mitochondrial function. These co-regulated genes could provide useful future prospects for better elucidating PrP’s cellular function. Misiewicz 53 Materials and Methods Genome-wide search for genes bearing an ERSE-26 element In order to find genes in the human genome with an ERSE-26 in their promoter, a program was written in Python. The sequence of the human genome, version GRC37.1 from the Genome Reference Commission (Church et al., 2011) was scanned using regular expressions representing the plus and minus strand versions of the ERSE-26. The region from -2,000 bp upstream of the transcription start site (TSS) to +500 bp downstream of the TSS was considered to be the promoter region of the gene. The Regular Expression pattern used was “ccaat.{26}ccacg”, matching precisely (no deviations were permitted). Both ERSE-26 elements in forward and reverse orientations were searched. Additionally, promoters in the positive and negative strands of the genome were examined. Source code is provided in a Git repository located at: (https://www.github.com/networkdynamics/erse-like-promoters-analysis). Cell culture conditions Human primary neuron cultures, obtained from human fetal brains subject to McGill University ethical approval, were cultured as previous described (LeBlanc, 1995). Briefy, brains were minced and subject to trypsinization. Following trypsinization, cells were plated on poly-lysine coated fasks and grown in the presence of 5-fuorodeoxyuridine, 1X Penicillin Streptomycin and minimum essential media (MEM) media supplemented with 10% bovine calf serum in a 5% CO 2 environment at 37° C. MCF-7 cells were grown in Roswell Park Memorial Institute (RPMI) media supplemented with 10% Fetal Bovine Serum (FBS) in a 5% CO 2 environment at 37° C. HEK293 cells were grown in Dulbecco’s Minimum Essential Media (DMEM) supplemented with 1.7 g/L NaHCO 3 and 10% fetal bovine serum (FBS) in a 5% CO2 incubator at 37° C. N2a cells were grown in DMEM supplemented with 1.7 g/L NaHCO3 and 10% FBS in a 5% CO2 environment at 37° C. Pharmacological Induction of ER Stress and other cell treatments Neurons were subjected to ER stress, under conditions experimentally optimized to minimize cell death and maximize ER stress induction (Jodoin et al. unpublished). In Misiewicz 54 neurons, ER stress was induced by the drugs Brefeldin A (5 µg/mL), Tunicamycin 0.5 µg/mL and Thapsigargin (3.25 µg/mL), all from Biomol. Drugs were dissolved in dimethyl sulfoxide (DMSO), and cells were treated for 6h, followed by immediate total RNA extraction. One neuron preparation was treated with ER stressors in the additional presence of 1 µg/mL actinomycin D. MCF-7 cells were also treated under conditions optimized to minimize cell death and maximize ER stress induction, as previously determined by Jodoin et al (Jodoin et al. unpublished). Cells were treated for 18h with all drugs at a concentration of 5 µg/mL, and then subject to total RNA extraction. Total RNA extraction and Reverse-Transcriptase Polymerase Chain Reaction Total RNA was extracted from neurons using Trizol (Invitrogen) as described by the manufacturer’s protocol. For MCF-7 cells, the Chomczynski Guanidine Thiocyanate method was used (Chomczynski and Sacchi, 2006). RNA was quantitated on an H4 plate reader (Biotek, Winooski, VT) for neurons, or a Nanodrop ND-1000 spectrophotometer (Thermo Scientific) otherwise. cDNA was prepared using avian myeloblastosis reverse transcriptase (AMV-RT), following the manufacturer’s protocol (Roche). Briefy, 1µg total RNA was used for cDNA synthesis by AMV-RT with poly-A primers, subjected to the manufacturer’s recommended thermal cycle, and used as a template for subsequent PCRs. Primer Design and PCR of ERSE-26 genes For PCRs, primers were designed using the NCBI PrimerBlast software available online at http://www.ncbi.nlm.nih.gov/tools/primer-blast/ (Sayers et al., 2012). Primers were designed to have an annealing temperature of 60° C. All oligonucleotides were ordered from a commercial supplier (Invitrogen), and optimized with a gradient of annealing temperatures. See Supplemental Table I for all primer sequences. During all subsequent PCRs, the optimal annealing temperature of a gene’s primer pair was always used. cDNA was diluted to 1:200 concentration and PCR was performed with Taq DNA polymerase (NEB), following the manufacturer’s protocol. Three biologically independent neuron preparations were used, and two technical replicates per gene per preparation were tested. Owing to differences in genetic backgrounds of the primary cultures, expression levels of each gene were analyzed separately in each preparation. Misiewicz 55 For MCF-7 cells, three independent experiments were conducted for all ER stress drug treatments and XBP1 transfections. Misiewicz 56 Supplemental Table I – Primers used in RT-PCR Note: a blank value in “Optimal TM” indicates no amplification Misiewicz 57 Quantitative PCR of ERSE-26 genes Next, qPCR was conducted on cDNA from neurons and MCF-7 cells using SYBR Green Taq Mastermix (Quanta Biosciences). For several ERSE-26 genes, new primers had to be designed, and the additional qPCR validated primers were obtained from PrimerBank (Wang et al., 2012b), a repository of primer sequences and synthesized by a commercial supplier (Invitrogen). These additional primers can be found in the Supplemental Table II. For neurons, two technical replicates for each of the three biologically distinct preparations were conducted. For MCF-7 cells, two technical replicates per biological replicate were conducted. Using ERSE-26 primers that produced a single unique amplicon, standard curves and amplification constants were determined. An Applied Biosystems 7500Fast qPCR apparatus (Invitrogen) was used, and the manufacturer’s default thermal cycle was used for all experiments. Primers were verified for specificity and performance for qPCR. All output was converted to fold-induction values compared to control (DMSO) treatments normalized with loading controls (HPRT1), using Pfaff’s method (Pfaff, 2001). Supplemental Table II – Primers used in qRT-PCR Note: these primers were designed specifically for qPCR. Misiewicz 58 Agarose gel electrophoresis and quantification TAE agarose gels were run under identical conditions (run for 1hr at 10 V/cm), using ethidium bromide staining. After gels completed, images were digitized using the Syngene GeneSnap software and gel documentation apparatus. After files were saved as uncompressed TIFFs, the ImageJ program from the NIH (Rasband, 2011) was used to quantitate gel bands. Band intensity was converted to relative fold change compared to control lanes (DMSO). Fold induction values determined with ImageJ for each treatment were averaged across all replications of PCRs. RT-PCR of HPRT1 was used as a loading control, and relative values were normalized against the fold change in HPRT1 levels. HPRT1 was chosen as the most stable amongst several other possible loading controls tested (ACTB, GAPDH). Western Blotting of PrP in ER stress treated neurons Cells were lysed for 20 min on ice in 150 mM NaCl, 2 mM EDTA, 0.5% Triton X-100 (v/v), 0.5% sodium deoxycholate (NaDOC) (w/v), 50 mM Tris-HCl, pH 7.5 lysis buffer containing fresh protease and phosphatase inhibitors (38µg/mL AEBSF, 0.5µg/mL leupeptin, 0.1µg/mL pepstatin, 0.1µg/mL N-α-p-tosyl-L-lysine chloromethyl ketone hydrochloride, 4 mM sodium orthovanadate, 20 mM sodium fuoride). After centrifugation of the lysate at 11 500 x g for 10 min at 4˚C, the supernatant was collected as the detergent soluble fraction and the detergent insoluble pellet was resuspended in 2 % SDS and sonicated for 30 sec. The protein concentration was determined with the BCA Protein Assay Reagents (Fisher, Nepean, ON). One hundred µg of protein was precipitated with 4 volumes of ice-cold methanol overnight at -20˚C, centrifuged at 11 500 x g for 15 min at 4˚C and the dried pellet solubilized in Laemmli sample buffer (2% SDS (w/v), 5% β-mercaptoethanol, (v/v), 10% glycerol (v/v), 0.01% bromophenol blue (w/v), 62.5 mM Tris-HCl, pH 6.8). The proteins were boiled for 3 min, separated in a 10% or 15% SDS-PAGE gel and transferred to PVDF membranes. PrP was detected with the 3F4 monoclonal antibody recognizing human PrP (1/2000, anti-PrP109-112) or the polyclonal R155 antiserum (1/500, anti-PrP36-56), which recognizes mouse and human PrP and was produced in our laboratory. Antibodies specific for β-actin (1/5000, clone AC-15, Sigma, Oakville, ON), mitochondrial Hsp70 (1/1000, clone JG1, Affinity Misiewicz 59 BioReagents, Golden, CO), and BiP (GRP 78, 1/250, clone H-129, Santa Cruz Biotechnology Inc), were used for western blot analyses. Immunoreactivity was revealed with 1/5000 anti-mouse or rabbit IgG conjugated to horseradish peroxidase secondary antibodies (GE Healthcare, Baie d’Urfe, QC) and chemiluminescence reagents from GE Healthcare or Millipore (Billerica, MA, USA) and detected with the Molecular Dynamics Storm 840 (GE Healthcare) or Kodak Biomax MR film (Carestream Health, Toronto, ON). The level of PrP was calculated from the intensity of the immunoreactive bands analyzed with the Image Quant TL software (GE Healthcare). The ratio of PrP over βactin was calculated. Cloning of ERSE-26 fragment into pMetLuc2 vector The pMetLuc2-Reporter construct (pML2, obtained from Clontech) is plasmid that encodes a luciferase gene from the copepod Metrida Longa. This gene contains a powerful endogenous signal peptide, causing the protein to be secreted into the growth media. This allows for a no-lysis protocol for assaying luciferase activity in transfected cells by simply reading the luciferase activity (when a substrate is added) of the media. To clone the PRNP promoter fragment into pML2, primers were designed to amplify the region from -391bp to -274bp, around the ERSE-26 (Mahal et al., 2001). The following primers were used (restriction sites in bold): Forward 5’-GAG CTC TCT CCA TTA TGT AAC GGG GA-3’, Reverse 3’-GCG AAT TCT CAG TTG ATA CCG CCT GCG G-5’. Primers contained the restriction endonuclease sites for EcoRI and SacI. The PCR product and pMetLuc2-reporter vector was digested using EcoRI and SacI, followed by ligation with T4 DNA Ligase (Fermantas). DH5α E. coli were transformed using standard protocols and DNA was prepared for transfection using the alkaline lysis method (Birnboim and Doly, 1979). The resulting pML2-EL26 construct was sequenced. Transfection of HEK293 cells and Luciferase assay HEK293 cells were plated in 6 well plates at approximately 500,000 cells/well. Twenty four hours after plating, using the polyethyleneimine (PEI) method (Boussif et al., 1995), cells were transfected with pMetLuc2-Reporter-ERSE-26 and plasmids encoding ΔATF6α (amino acids 1-373) or sXBP1 (Lee et al., 2002). Following 6h of transfection, Misiewicz 60 media was changed, and cells were allowed to grow for 20h. After 20h, media was collected. Luciferase assays were conducted on 50 µl of HEK293 growth media. Substrate (Ready-to-Glow Secreted Luciferase substrate, Clontech) for the luciferase protein was prepared according to the manufacturer’s protocols. Assays were conducted in a 96-well white, opaque plate (Costar) and total luminescence was determined using an H4 plate reader (Biotek, Winooski, VT). Three reads were conducted per well, and two separate pMetLuc2-Reporter-ERSE-26 constructs were tested in three biologically independent experiments each. As a control, total RNA was extracted as before to verify sXBP1 and ATF6α transcripts in transfected cells. Western Blotting of XBP1 and ATF6α in HEK293 cells HEK293 cells were extracted with NP-40 lysis buffer (50 mM Tris pH 8.0, 150 mM NaCl, 1% NP-40, 5 mM EDTA pH 8.0). Extracts were separated by 10% SDS-PAGE, transferred using a Biorad Transblot Turbo apparatus to PVDF membrane, blocked 1h in 5% milk, then incubated with 1:100 anti-XBP1 (SC7160, Santa Cruz) or 1:100 anti-ATF6 (IMG-273, Imgenex) to confirm expression of transgenes. Membranes were detected using antimouse IgG horseradish peroxidase at 1:5000 in 5% milk (Jackson) and visualized with ECL Prime (GE Healthcare) on Kodak BioMax MR Film. Mutagenesis of pMetLuc2-ERSE-26 constructs Following cloning of the PRNP promoter fragment into the pMetLuc2 vector, the ERSE26 was mutated by PCR with mutagenic oligos. Primers were designed to target the second part of the ERSE-26, the putative XBP1/ATF6 binding site: CCACG was changed to ATCTA. Two additional basepairs after the XBP1/ATF6 site were also changed from TC to GA, yielding a final mutation of ATCTAGA. DNA was synthesized using Pfu Turbo (Agilent Technologies) and subject to DpnI digestion to remove non-mutated plasmid DNA. DNA was then transformed into DH5α E. coli using standard techniques and prepared using the alkaline lysis method. Transfection in HEK293 cells was done as before. The following primers were used for mutagenesis, with the mutation in bold, and the former location of the CCACG site underlined: 5’-GAT TTT TAC AGT CAA TGA GAT CTA GAA GGG AGC GAT GGC ACC C-3’ and 5’-GGG TGC CAT CGC TCC CTT CTA GAT CTC ATT GAC TGT AAA AAT C-3’. Misiewicz 61 Transfection of MCF-7 cells MCF-7 cells from ATCC (Manasas, VA) were cultured as described above and transfected with Lipofectamine 2000 following the manufacturer’s protocol (Invitrogen). Cells were transfected with either pCGN-sXBP1 (Bommiasamy et al., 2009) or vehicle (pCGN-EGFP). Total RNA was extracted using Chomczynski’s method, and cDNA was synthesized using AMV-RT as before. Next, qPCR was conducted like in previous MCF7 experiments. Fold induction values were computed using Pfaff’s method as before. Statistical analysis For RT-PCR conducted on neuron preparations, statistics were conducted on the average fold-induction values computed by densitometry. For each biologically independent neuron preparation, a one way analysis of variance was conducted, followed by Dunnet’s post hoc test. For qPCR on MCF-7 cells, an analysis of variance was conducted on each of the genes tested in qPCR, followed by Dunnet’s post-hoc test. For the luciferase assays, a one-way analysis of variance followed by Tukey’s Honest Simple Difference post-hoc test was used to determine whether or not luminescence values were statistically significant. For all statistics, a p-value less than 0.05 was taken to be significant and all were computed using InStat version 3.1a (Graphpad Software). Misiewicz 62 Results XBP1 activates ERSE-26 in HEK293 and N2a cells To determine if the ERSE-26 responds to ER stress-regulated transcription factors, HEK293 cells were co-transfected with the pML2-ERSE-26 luciferase reporter construct and pCGN-sXBP1 or pCGN-ATF6(1-373). Co-transfection of pML2-ERSE-26 and sXBP1 resulted in high luciferase activity (Fig. 3A), suggesting that sXBP1 is a transcription factor that binds the ERSE-26. ΔATF6α did not activate ERSE-26-mediated transcription, unlike its ability to transactivate the canonical ERSE. In cells transfected with pML2empty and sXBP1, some non-significant luciferase activity was detected. This may be due to a low-affinity XBP1 site in the pML2 plasmid as identified by a computation search of the Transfac database of promoter binding sites (results not shown). Mutagenesis of the putative ATF6α/XBP1 binding site (CCACG) of the pML2-ERSE-26 abrogated transactivation by sXBP1 (Fig. 3B). As expected, pML2-luciferase under the control of a CMV promoter showed very high luciferase activity, while mock, nontransfected or pML2-transfected cells showed no luciferase activity (Fig. 3A&B). RT-PCR (Fig. 3C) and western blots (Fig. 3D) confirmed expression of constructs. Co-transfection of pML2-ERSE-26 and sXBP1 in N2a cells gave identical results (Suppl. Fig. 1). Together, these results indicate that sXBP1 is the transactivating factor for the ERSE-26. Misiewicz 63 Misiewicz 64 Figure 3 – ERSE-26 activity is XBP1 dependent. A. Luciferase assay shows XBP1 induces luciferase activity in reporter construct. *: p<0.001. B. Mutation of ATF6/XBP1 binding site removes this activity. *: p< 0.01 C. Ethidium bromide agarose gel of DNA obtained by RT-PCR of transfected cells, indicating sXBP1 and ATF6 expression in transfected cells. D. Western Blot of transfected cells showing sXBP1 and ATF6 expression, confirming transfection. In panels A and B, Error bars indicate SEM. Supplemental Figure 1 – Activity of secreted luciferase in N2a cells transfected with pML2-EL26 and sXBP1. Human sXBP1 activates pML2-EL26 secreted luciferase activity in mouse cells. Misiewicz 65 ERSE-26 was found in the promoters of 38 genes To assess if ERSE-26 regulates other genes, a Python program was written to search the Genome Reference Consortium human genome (Build GRC37.1). In total, 38 genes contained an ERSE-26 within -2000bp and +500bp from the transcription start site of the gene (Table I). None of these genes also contained canonical ERSEs except C6orf20, which contained an ERSE-II, and GADD45B and LRRC55, which contained XBP1 binding sites in their promoters. Twenty-six ERSE-26 matches were found in the reverse (3’ to 5’) direction, and 12 were found in the forward direction. As a control for the search program, the search was repeated with the canonical ERSE (CCAAT-.9-CCACG) as a target pattern. Forty-five genes, including the previously known ERSE-containing genes HSPA5 (Bip, Grp78), XBP1 and PDIA6 contained an ERSE within -2000bp and +500bp from the transcription start site of the gene (Suppl. Table II, page 58). These results show that the ERSE-26 is not unique to the PRNP promoter and suggest that ER stress may regulate a wide variety of genes previously unsuspected to be involved in the ER stress response. Misiewicz 66 Table I Misiewicz 67 Supplemental Table III Misiewicz 68 ERSE-26 genes are upregulated in primary human neuron cultures treated with BFA, TM or Thps To confirm that the ERSE-26 containing promoters are trans-activated during ER stress, primary human neurons were treated with BFA, TM, or Thps (Fig. 4). RT-PCR identified that 20 of the 38 ERSE-26 genes were transcribed in primary human neuron cultures (Fig. 4A). ER stress induction was confirmed in TM and Thps-treated neurons by the presence of the shorter 257 bp sXBP1 amplicon. The lack of sXBP1 amplicon in the BFAtreated human neurons indicates differential regulation of the ER stress transduction pathways. Relative to the DMSO control, each of the pharmacological inducers of ER stress (BFA, TM, or Thps) induced expression of some of the ERSE-26 genes in at least one neuron culture (Fig. 4A). Genetic variability may explain differential ER stress responses in the primary human neuron preparations. Following densitometric analysis (Fig. 4B), GADD45B showed a significant increase for all three stressors in preparation 1, and a significant increase under TM and Thps treatment for preparation 2, but no increases in preparation 3. Despite the variability in responses across preparations, SESN2, NUDT9-T1 (transcript variant 1), GAD2, and ERLEC1 showed an upward trend. In contrast, the genes USP4, YPEL3, LRRC55 and SMPD1 showed downward trends in all cases. The genes GNAQ, KBTBD2 and NOL10 were mostly unchanged across all preparations. Further analyses by qPCR showed that BFA, TM, and Thps all significantly increased HSPA5 (BiP) and sXBP1 mRNA levels, thus confirming the activation of the ER stress response under these conditions (Fig. 4C). Furthermore, at least one of the ER stressors increased PRNP, ERLEC1, GADD45B, SESN2, and SLC38A5 mRNA levels in primary human neurons. Similar to previous observation, the response of some genes to ER stress varied with the type of ER stress despite the fact that each stressor clearly induced ER stress. Increased PrP and BiP accompanied ER stress increased mRNA levels (Fig. 4D). Transcriptional inhibition with actinomycin D stunted the ER stress-mediated upregulation of PrP. Together, these results indicate that ERSE-26-containing genes can be differentially Misiewicz 69 transactivated depending on the ER stress and on the genetic background of the individual. Misiewicz 70 Misiewicz 71 Figure 4 – Induction of ER stress increases ERSE-26 gene expression in cultured primary human neurons. Figure represents three independent preparations, in duplicate. A. Representative Gel electrophoresis of ERSE-26 genes detected by RT-PCR in cultured primary human neurons, indicating which ERSE-26 genes are detectable in neurons. XBP1 and HPRT1 shown as controls. B. Panel B shows each pharmacological treatment as a bar graph, and the expression of each gene in each neuron preparation is indicated by a different color bar within each graph. Upper panel: Fold change of ERSE-26 genes following treatment with BFA compared to DMSO (1 fold change). Middle panel: Fold change of ERSE-26 genes following treatment with TM. Bottom panel: Fold change of ERSE-26 genes following treatment with Thps. *: p<0.05, **: p<0.005. Error bars indicate SEM. Within each treatment, significance was assessed by a one-way analysis of variance followed by Dunnett’s post-hoc test. C. By qPCR, average fold induction of several ERSE-26 genes and ER stress controls averaged across two biological replicates and two technical replicates. Error bars represent SEM. A one-way analysis of variance followed by Dunnet’s post-hoc test were used to assess significance. *: p < 0.05, **: p < 0.005. NT indicates no treatment and -S indicates serum deprivation. D. In cultured primary neurons, PrP increase following ER stress treatment is transcriptionally dependent. The top two panels show induction of PrP by western blot during ER stress. However, when the transcription blocker Actinomycin is applied to the neurons, PrP levels do not increase in ER stress. Misiewicz 72 ERSE-26-containing genes are upregulated in ER stressed breast carcinoma MCF-7 cells. To determine if the variability observed with different ER stressors is dependent on the genetic background of the cells, we stressed MCF-7 cells with BFA, TM or Thps and subjected the mRNA to qPCR. The results showed increased HSPA5 when cells were treated with all three stressors and increased sXBP1 mRNAs when cells were treated with BFA and Thps, confirming the ER stress response (Fig. 5A). PRNP, ERLEC1, GADD45B, SESN2, and SLC38A5 showed a significant increase in mRNA when MCF-7 cells were treated with BFA. However, TM treatment caused increased expression only for ERLEC1, and Thps treatment induced expression of ERLEC1 and GADD45B. To assess if XBP1 alone can transactivate ERSE-26 containing genes in a cellular system, MCF-7 cells were transfected with a vector expressing sXBP1 and expression levels was monitored by qPCR. As expected, sXBP1 mRNA levels were increased 155 fold over mock-transfected cells (Fig. 5B). The control EGFP vector induced a slightly higher level as well but this was not significant. HSPA5 mRNA levels were induced approximately 3 fold with sXBP1, as expected for an ER stress regulated gene. However, only PRNP showed increased mRNA levels with sXBP1 expression. These results suggest that regulation of ERSE-26 gene expression required additional transactivating factors. Misiewicz 73 Misiewicz 74 Figure 5 – MCF-7 cells show induced ERSE-26 gene expression following ER stress treatment or sXBP1 transfection A. Fold induction of ERSE-26 genes in MCF-7 cells treated with BFA, TM or Thps for 18 h as determined by qPCR. Induction of HSPA5 and spliced XBP1 were used as ER stress controls. A one way analysis of variance followed by Dunnet’s post hoc were used to assess significance, *: p<0.05, **: p< 0.005, ***: p< 0.0001. Data represent three biologically independent experiments performed in duplicate. B. ERSE-26 gene expression levels in sXBP1 transfected cells compared to mock transfected cells. **: p < 0.005. Data represent two biologically independent experiments performed in duplicate. Misiewicz 75 Discussion Following the previous identification of the ERSE-26 (Jodoin et al. Manuscript submitted), we set out to determine whether or not this element is unique to PRNP or represents a new, novel mechanism for cells to respond to ER stress. Interestingly, during the discovery and characterization of the ’classical’ ERSE in the late nineties and early 2000s, it was demonstrated that the ERSE is dependent on a precise spacing of nine nucleotides between ATF6 and NF-Y sites (Yoshida et al., 2000). The authors demonstrated this with luciferase reporter constructs with either eight or ten basepairs added between binding sites, but larger numbers were not tested. Our results here suggest that other ERSE configurations are possible, and that the ERSE-26 operates through previously identified ER stress transcription factors. Previously known functions of PrP appeared to be co-regulated with PRNP by the ERSE-26, and we also identified putative new processes that are co-regulated with PRNP. Our search program identified 38 genes in the human genome with the ERSE-26 in their promoter. Importantly, this indicates that the element exists elsewhere in the genome and is not PRNP specific. In addition, most of the genes found in the search did not contain other ERSEs and were not previously known to be ER stress regulated. These results are a new set of ER stress regulated genes, previously unknown. We restricted our search to exact matches only. Allowing for more sophisticated searches, using position weight matrices or Hidden Markov Models might produce for hits. Additionally, our definition of 'promoter' was rather arbitrary. A more nuanced definition of the promoter region to search might expand hits. However, since we found abundant genes using the aforementioned parameters, we did not adjust these values, but it seems like a logical first step for followup experiments. Next, we showed ERSE-26 gene induction in cells that are pharmacologically ER stressed. The behavior of ERSE-26 genes in neuron cultures was inconsistent across genetic backgrounds. All genes showed upregulation in at least one preparation, but it was not repeated across all three for most genes. However, the genes GADD45B showed consistent behavior across at least two treatments in two preparations, and SESN2 showed an upward trend in two treatments in two preparations. In qPCR experiments, Misiewicz 76 which utilize a far more accurate detection method, GADD45B, and SESN2 showed upregulation. Both are previously known as stress responsive genes: their status as ERSE-26 genes might explain one way they are activated. In contrast to neurons, more ERSE-26 gene tested in MCF-7 cells was upregulated, indicating that perhaps the ERSE26 is more or less functional in different cell types. PRNP, which is ERSE-26 regulated, has been shown to be protective against apoptotic insults in breast cancers; perhaps ERSE-26 regulation is the mechanism causing this (Diarra-Mehrpour et al., 2004). Co-transfection of a luciferase reporter construct and the activated ER stress transcription factors ATF6 and XBP1. We elected for co-transfection because the use of the drugs BFA, TM or Thps would interfere with the secretory pathway in a variety of ways, confusing the results. XBP1 is the transcription factor primarily responsible for ERSE-26 activity, while the classical ERSE was found by Yoshida et al. to be mostly only ATF6 activated. However, XBP1 has been shown to bind the ERSE in a yeast one-hybrid screen, and yeast Hac1p (XBP1 homolog) binds the ATF6 promoter (Lee et al., 2002; Yoshida et al., 1998). Subsequent to the discovery of the ERSE, additional ERSE-like motifs have been found, and have been shown to be regulated by XBP1 and/or ATF6 to differing degrees (Yamamoto et al., 2004). Thus, the ERSE-26 behaves similarly to other ERSEs. Despite the evidence for XBP1, we cannot at this time rule out that ATF6 plays a role in ERSE-26 activity. Possibly transfection with ATF6 requires co-transfection with NF-Y, the obligate partner of ATF6 in the classical ERSE. As further evidence for XBP1’s ability to induce ERSE-26 gene expression, transfection of MCF-7 cells showed increased ERSE-26 genes. Thus, experimentally, the ERSE-26 can be activated in two ways: either pharmacologically or genetically. ER stress is implicated in a variety of diseases such as Alzheimer and cancer, understanding how cells respond to ER stress is important to properly treat these diseases. To gain insight into the transcriptional program regulating PRNP during ER stress, we conducted an extensive literature search to identify functions of ERSE-26 genes (automated clustering did not give any significant results). Some potential novel roles for PRNP in the cell were identified as well as confirmation for previously published functions of PrP. One notable functional group identified in this search was genes related to cell adhesion and synapse function. GAD2, LRFN4, LRRC55, and SLC38A5 all play a role in Misiewicz 77 this process. GAD2 is one of two terminal rate limiting enzymes that cleave glutamic acid into γ-Aminobutyric acid (GABA) (Erlander et al., 1991). LRFN4, also known as SALM3 is known to bind Post Synaptic Density 95 protein, and induces neurite outgrowth (Seabold et al., 2008). LRRC55 was recently identified as a protein known to enhance activity of calcium activated potassium channels (Yan and Aldrich, 2012). Finally, SLC38A5, also known as SN2 or SNAT5, is an astrocyte expressed amino acid transporter that permits astrocytes to release glutamic acid for neurons to cleave to GABA (Cubelos et al., 2005). Both drugs and XBP1 transfection showed that LRFN4 and KBTBD2 were very upregulated, as well as SLC38A5. GAD2 showed expression during treatment with BFA. These findings give support to previously published evidence that PrP plays a role in the synapse (Linden et al., 2008). Additionally, we have identified biological processes and co-regulated genes that future studies can use to pinpoint the role of PrP at the synapse. Another notable genes in the results is GADD45B. It was originally discovered as part of a family containing GADD153 (CHOP) (Zhan et al., 1994) and the prionassociated pro-survival EIF2a phosphatase GADD34 (Moreno et al., 2012). This protein is a known activator of p38 MAPK signaling that is known to promote survival (Takekawa and Saito, 1998; Takekawa et al., 2002; Engelmann et al., 2007). GADD45B’s upregulation is consistent with PrP’s previously known pro-survival function (Kuwahara et al., 1999; Roucou et al., 2005; Jodoin et al., 2007). Notably as well, p38 MAPK signaling has been shown to promote changes to the cytoskeleton, which is consistent with our observation that beta-actin levels changed dramatically during ER stress activation (Guay et al., 1997). This gene was consistently upregulated in neurons treated with BFA, TM and Thps, sometimes as much as 8 fold. It seems clear this gene is co-regulated with PrP, and this further strengthens the case for normal PrP’s pro-survival role. In addition to the known stress responder GADD45B, MYEOV2 (myeloma over-expressed 2) was highly upregulated in neurons, although not in MCF-7. This protein, about which nearly nothing is known, was identified as highly expressed gene in myelomas. Thus is seems logical that this gene would be activated in a stress response, especially since it is known that ER stress is a hallmark of cancer. The final group of function that appeared was related to oxidative stress. The genes NUDT9, TIMM44, SESN2 and GADD45B are known to be localized to the Misiewicz 78 mitochondria or as responders to oxidative stress (Takekawa and Saito, 1998; Budanov et al., 2010; Perraud et al., 2003). Although TIMM44 did not show induction following XBP1 transfection or drugs in neurons, it showed activation in MCF-7 cells treated with drugs. NUDT9, a gene whose function remains largely unknown, was strongly induced by XBP1 transfection. Interestingly, both of these proteins are known to localize to the mitochondria, and TIMM44 is known to interact with chaperones (Moro et al., 1999). SESN2 and GADD45B (the latter is even used as a marker for p53 activity) are both regulated by p53 in response to environmental stresses (Budanov, 2011; Takekawa and Saito, 1998). Even more interesting, our results included a transcription factor FOXO4; the FoxO family of transcription factors are known to regulate Sestrin family members (Budanov et al., 2010). However, we were not able to detect FOXO4 expression in MCF-7 cells or neurons. Most importantly though, is that PRNP contains a p53 binding site (Vincent et al., 2009). Thus all of these genes share a common regulator in addition to their ERSE-26. This further bolsters the evidence for the cytoprotective role of PrP. PRNP has been previously identified as a oxidative stress responsive gene (Shyu et al., 2004; Wong et al., 2001; Qin et al., 2009); the evidence here bolsters its anti-oxidative stress role. There appear to be two functional groups of genes identified in this bioinformatic survey. Genes that respond to oxidative stress and promote cell survival, and genes that appear to play a role at the synapse and its function. It has been shown that cytosolic PrP, retrotranslocated from the membrane is needed for PrP to exert its protective functions. Furthermore, it is a well known fact that synapses are required to ensure neuronal health. The genes identified and validated here seem to refect this two-sided nature of physiological PrP, and its easy to imagine how scrapie diseases can disrupt these processes. Although we have not shown that PrP interacts directly with anything identified here, that seems to be a promising area of future study, given these genes’ implications in processes that PrP is involved in. In conclusion, we have identified many new genes that contain a previously described novel promoter element similar to the ERSE. We have showed that this element appears to be responsible for transcriptional activity in a variety of environments, and that genes regulated by it (and co-regulated with PrP) are involved in previously identified cytoprotective roles in the cell (i.e. these genes were already shown Misiewicz 79 to be protective). The results presented here can be used to help identify the exact function of PrP, whose role in all these processes is known but remains unexplained. Misiewicz 80 Chapter 3 – General Discussion The function of PrP in the cell remains somewhat unclear to this day. PrP is implicated in a variety of cellular functions, including metal ion binding, synapse localization, antioxidative stress response, and anti-Bax function (Westergard et al., 2007; Linden et al., 2008). In particular, PrP's role as an anti-Bax agent is convincing; the particular isoform of PrP has been identified, as well as the exact region of the protein that is needed, and the role of disease mutations in this anti-Bax activity (Bounhar et al., 2001; Roucou et al., 2003, 2005; Jodoin et al., 2007; Lin et al., 2008; Jodoin et al., 2009) . However, this remains somewhat controversial in the field of prion research, as there are some investigators who claim to have found a cytotoxic effect for cytosolic PrP. (Ma et al., 2002; Anantharam et al., 2008). Despite these conficting reports however, a clear picture of PrP C as a protective gene is beginning to emerge. One area that, until now, had been poorly studied is the regulation of PRNP in this context. Our lab recently characterized the expression of PRNP as a responder to ER stress. Furthermore, in the same work, the association of PrP with longevity in breast cancer tissue was established and correlated well with BiP expression (Jodoin et al., unpublished). Importantly, it was shown that this ER stress regulation of PRNP was due in part to a novel ERSE. Jodoin et al. demonstrated the existence and function of this element in the PRNP promoter, but the current work expands this into the rest of the genome. Furthermore, I've shown that genes regulated by oxidative stress appear to be co-regulated with PRNP. While ER stress regulation is a new role for PrP, regulation by oxidative stress is congruent with PrP's previously identified function as an oxidative stress responder. The evidence presented here shows that these two responses can be regulated by the same transcriptional element, suggesting overlap between these two responses. There are two important results of the genome-wide search and subsequent search. One, the element is not unique to PRNP, and represents a novel mechanisms for mediating ER stress transcription. This puts the ERSE-26 in the same category as the ERSE, ERSE-II, ERSE-like, and UPRE. Furthermore, it is important to note that in this Misiewicz 81 work, only exact matches were used for the search. More sophisticated models allowing substitutions in the element can be produced, and the number of ERSE-26 genes will be larger than what is reported here. Two, the ERSE-26 is a marker for ER stress mediated genes, like the ERSE. A great deal of the genome remains very poorly understood; many genes have been identified by exome sequencing or computational searches, however nothing is known about them. By examining their promoters for certain regulatory elements, much can be inferred about the function of these newly discovered genes. Indeed, there are several genes like this in the ERSE-26, and given that I have shown this element is a marker for a certain pattern of expression, already we know more about them than we did before. Understanding the nature of what genes are regulated alongside PRNP by the ERSE-26 is also very relevant. The genes that contained an ERSE-26 were genes that have been previously shown to be protective. The response to ER stress in cells is a deeply conserved transcriptional program, going back as far as yeast. Studying these transcriptional programs as a whole is informative because we can gain a much more holistic understanding of how the cell functions. Thus, the genes found in this survey would seem to corroborate the growing protective role for PrP, since the most upregulated genes (GADD45B and SESN2) were previously known stress responders and pro-survival genes. It seems unlikely that the literature suggesting PrP C can be toxic is not correct, given all the evidence now assembled that indicates PrP is protective against a variety of insults. Additionally, PrP C is highly and continuously expressed in many different cell types, which does not make sense for a cytotoxic gene. Furthermore, PrP's protective nature also suggests a mechanism by which scrapie can be exacerbated: mis-folded PrP produces stress in cells, which upregulate PrP in response, which leads to more substrate being available. Having a more complete understanding of ER stress is very relevant to disease. ER stress has been implicated in diseases that are of enormous challenge and burden to healthcare systems around the world, namely Alzheimer Disease and cancer (Doyle et al., 2011; Ozcan and Tabas, 2012). A more complete understanding of ER stress could lead to invaluable treatments or remedies for these diseases which are the cause of untold human suffering and burden. Thus, any better understanding of the regulation of Misiewicz 82 ER stress and how cells cope with stresses is of high importance. In a much broader sense, the methods presented in this paper are of great importance. Biology is now entering the age of the large-scale experiment. Microarrays, protein interaction assays, and genome sequencing projects are producing massive amounts of data, more than can be analyzed by hand. This is a sharp break from the past, where the methods available only allowed the study of a small number of questions simultaneously. As a testament to the amount of data biologists are now producing, one can simply look at the size of Genbank, which grew by nearly 20% last year alone (containing 339 billion basepairs as of August 2011), and doubles every 18 months (Benson et al., 2012). And it is not simply biological data that is being produced: all of humanity produced 1.8 zettabytes of information in 2010, or 1.8 trillion gigabytes, representing 9-fold growth in five years. By 2015 nearly 8 zettabtyes are expected, which will be a major challenge to computer scientists to understand, interpret and extract knowledge from this information (Gantz and Reinsel, 2011) However, there is a difference between information and knowledge, and one of the major challenges of these large scale projects is interpreting the data. This is one of the areas where bioinformatics can be of great use to biologists. As mentioned earlier, “wet bench” methods have a restricted throughput, meaning that in order to gain knowledge about how parts of a cell interact, specific topics of interest must be identified for study. Using bioinformatics will allow scientists to identify research questions and topics for time consuming wet bench research. The sequencing of the human genome is one of the proudest moments for science, on par with the moon landing or the construction of computers and the internet. Recently-released commercial products and advances in sequencing technologies have put us on the cusp of the era of the $1,000 genome, compared to billions in 2000. In a short matter of time, it seems that every individual will have his genome sequenced as a routine matter during his life. The need for scientists who have the tools to deal with this massive deluge of information will only be increased in years to come – but having the ability to confirm findings in the lab will be equally important. The work presented in this thesis represents a combination of the old world and the new, which will be critically important in years to come. Misiewicz 83 Conclusion The evidence presented here shows that PRNP can be expressed in response to ER stress. This activity is regulated by the ERSE-26, a novel ERSE. It is functional and extant elsewhere in the genome, and transactivated by XBP1. The genes upregulated by the ERSE-26 are implicated in the response to Oxidative Stress. This corroborates the previous evidence for PrP function. PRNP helps contribute to the longevity of cells that express it, due to its protective activity, which is why it is upregulated by the ERSE like 26 when cells are stressed. Furthermore, another tissue that experiences large amounts of non-pathological ER stress is the immune system, where BiP was first discovered. In my view, it is not a coincidence that these tissues also express high levels of PrP. Taken together, the ERSE-26 mediated regulation of PRNP corroborates growing evidence that PRNP is a cytoprotective gene, responds to oxidative stress, and helps cells survive perturbations. Misiewicz 84 References Alberts, B., A. Johnson, J. Lewis, M. Raff, K. Roberts, and P. Walter. 2008. Molecular Biology of the Cell. Fifth Edit. Garland Science, New York. Altschul, S.F., W. Gish, W. Miller, E.W. Myers, and D.J. Lipman. 1990. Basic local alignment search tool. Journal of molecular biology. 215:403–10. Anantharam, V., A. Kanthasamy, C.J. Choi, D.P. Martin, C. Latchoumycandane, J.A. Richt, and A.G. Kanthasamy. 2008. Opposing roles of prion protein in oxidative stressand ER stress-induced apoptotic signaling. Free radical biology & medicine. 45:1530–41. Antonsson, B., S. Montessuit, S. Lauper, R. Eskes, and J.C. Martinou. 2000. Bax oligomerization is required for channel-forming activity in liposomes and to trigger cytochrome c release from mitochondria. The Biochemical journal. 345 Pt 2:271–8. Basler, K., B. Oesch, M. Scott, D. Westaway, M. Wälchli, D.F. Groth, M.P. McKinley, S.B. Prusiner, and C. Weissmann. 1986. Scrapie and cellular PrP isoforms are encoded by the same chromosomal gene. Cell. 46:417–28. Bellingham, S. a, L. a Coleman, C.L. Masters, J. Camakaris, and A.F. Hill. 2009. Regulation of prion gene expression by transcription factors SP1 and metal transcription factor-1. The Journal of biological chemistry. 284:1291–301. Benson, D.A., I. Karsch-Mizrachi, K. Clark, D.J. Lipman, J. Ostell, and E.W. Sayers. 2012. GenBank. Nucleic acids research. 40:D48–53. Bertolotti, a, Y. Zhang, L.M. Hendershot, H.P. Harding, and D. Ron. 2000. Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nature cell biology. 2:326–32. Birnboim, H.C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic acids research. 7:1513–23. Bommiasamy, H., S.H. Back, P. Fagone, K. Lee, S. Meshinchi, E. Vink, R. Sriburi, M. Frank, S. Jackowski, R.J. Kaufman, and J.W. Brewer. 2009. ATF6alpha induces XBP1independent expansion of the endoplasmic reticulum. Journal of cell science. 122:1626–36. Misiewicz 85 Bounhar, Y., K.K. Mann, X. Roucou, and A.C. LeBlanc. 2006. Prion protein prevents Baxmediated cell death in the absence of other Bcl-2 family members in Saccharomyces cerevisiae. FEMS yeast research. 6:1204–12. Bounhar, Y., Y. Zhang, C.G. Goodyer, and A. LeBlanc. 2001. Prion protein protects human neurons against Bax-mediated apoptosis. The Journal of biological chemistry. 276:39145–9. Boussif, O., F. Lezoualc’h, M.A. Zanta, M.D. Mergny, D. Scherman, B. Demeneix, and J.P. Behr. 1995. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proceedings of the National Academy of Sciences of the United States of America. 92:7297–301. Brown, D.R., K. Qin, J.W. Herms, A. Madlung, J. Manson, R. Strome, P.E. Fraser, T. Kruck, A. von Bohlen, W. Schulz-Schaeffer, A. Giese, D. Westaway, H. Kretzschmar, and A. Von Bohlen. 1997a. The cellular prion protein binds copper in vivo. Nature. 390:23–26. Brown, D.R., W.J. Schulz-Schaeffer, B. Schmidt, and H.A. Kretzschmar. 1997b. Prion protein-deficient cells show altered response to oxidative stress due to decreased SOD-1 activity. Experimental neurology. 146:104–12. Brown, D.R., B.S. Wong, F. Hafiz, C. Clive, S.J. Haswell, and I.M. Jones. 1999. Normal prion protein has an activity like that of superoxide dismutase. The Biochemical journal. 344 Pt 1:1–5. Budanov, A. V. 2011. Stress-responsive sestrins link p53 with redox regulation and mammalian target of rapamycin signaling. Antioxidants & redox signaling. 15:1679–90. Budanov, A. V, J.H. Lee, and M. Karin. 2010. Stressin’ Sestrins take an aging fight. EMBO molecular medicine. 2:388–400. Büeler, H., a Aguzzi, a Sailer, R. a Greiner, P. Autenried, M. Aguet, and C. Weissmann. 1993. Mice devoid of PrP are resistant to scrapie. Cell. 73:1339–47. Campana, V., D. Sarnataro, and C. Zurzolo. 2005. The highways and byways of prion protein trafficking. Trends in cell biology. 15:102–11. Carrasco, D.R., K. Sukhdeo, M. Protopopova, R. Sinha, M. Enos, D.E. Carrasco, M. Zheng, M. Mani, J. Henderson, G.S. Pinkus, N. Munshi, J. Horner, E. V Ivanova, A. Protopopov, K.C. Anderson, G. Tonon, and R. a DePinho. 2007. The differentiation and Misiewicz 86 stress response factor XBP-1 drives multiple myeloma pathogenesis. Cancer cell. 11:349– 60. Caughey, B., R.E. Race, D. Ernst, M.J. Buchmeier, and B. Chesebro. 1989. Prion protein biosynthesis in scrapie-infected and uninfected neuroblastoma cells. Journal of virology. 63:175–181. Chomczynski, P., and N. Sacchi. 2006. The single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction: twenty-something years on. Nature protocols. 1:581–5. Church, D.M., V.A. Schneider, T. Graves, K. Auger, F. Cunningham, N. Bouk, H.-C. Chen, R. Agarwala, W.M. McLaren, G.R.S. Ritchie, D. Albracht, M. Kremitzki, S. Rock, H. Kotkiewicz, C. Kremitzki, A. Wollam, L. Trani, L. Fulton, R. Fulton, L. Matthews, S. Whitehead, W. Chow, J. Torrance, M. Dunn, G. Harden, G. Threadgold, J. Wood, J. Collins, P. Heath, G. Griffiths, S. Pelan, D. Grafham, E.E. Eichler, G. Weinstock, E.R. Mardis, R.K. Wilson, K. Howe, P. Flicek, and T. Hubbard. 2011. Modernizing reference genome assemblies. PLoS biology. 9:e1001091. Cubelos, B., I.M. González-González, C. Giménez, and F. Zafra. 2005. Amino acid transporter SNAT5 localizes to glial cells in the rat brain. Glia. 49:230–44. Diarra-Mehrpour, M., S. Arrabal, A. Jalil, X. Pinson, C. Gaudin, G. Piétu, A. Pitaval, H. Ripoche, M. Eloit, D. Dormont, and S. Chouaib. 2004. Prion protein prevents human breast carcinoma cell line from tumor necrosis factor alpha-induced cell death. Cancer research. 64:719–27. Doyle, K.M., D. Kennedy, A.M. Gorman, S. Gupta, S.J.M. Healy, and A. Samali. 2011. Unfolded proteins and endoplasmic reticulum stress in neurodegenerative disorders. Journal of cellular and molecular medicine. 15:2025–39. Engelmann, a, D. Speidel, G.W. Bornkamm, W. Deppert, and C. Stocking. 2007. Gadd45β is a pro-survival factor associated with stress-resistant tumors. Oncogene. 27:1429–1438. Erlander, M.G., N.J. Tillakaratne, S. Feldblum, N. Patel, and A.J. Tobin. 1991. Two genes encode distinct glutamate decarboxylases. Neuron. 7:91–100. Fernandez, P.M., S.O. Tabbara, L.K. Jacobs, F.C. Manning, T.N. Tsangaris, a M. Schwartz, Misiewicz 87 K. a Kennedy, and S.R. Patierno. 2000. Overexpression of the glucose-regulated stress gene GRP78 in malignant but not benign human breast lesions. Breast cancer research and treatment. 59:15–26. Flynn, G.C., T.G. Chappell, and J.E. Rothman. 1989. Peptide binding and release by proteins implicated as catalysts of protein assembly. Science. 245:385–90. Funke-Kaiser, H., S. Theis, T. Behrouzi, A. Thomas, K. Scheuch, F.S. Zollmann, M. Paterka, M. Paul, and H.D. Orzechowski. 2001. Functional characterization of the human prion protein promoter in neuronal and endothelial cells. Journal of molecular medicine. 79:529–35. Gantz, J., and D. Reinsel. 2011. Extracting Value from Chaos State of the Universe : An Executive Summary. IDC Whitepaper. 1–12. Guay, J., H. Lambert, G. Gingras-Breton, J.N. Lavoie, J. Huot, and J. Landry. 1997. Regulation of actin filament dynamics by p38 map kinase-mediated phosphorylation of heat shock protein 27. Journal of cell science. 110 ( Pt 3:357–68. Haas, I.G., and M. Wabl. 1983. Immunoglobulin heavy chain binding protein. Nature. 306:387–9. Haeberlé, A.M., C. Ribaut-Barassin, G. Bombarde, J. Mariani, G. Hunsmann, J. Grassi, and Y. Bailly. 2000. Synaptic prion protein immuno-reactivity in the rodent cerebellum. Microscopy research and technique. 50:66–75. Haraguchi, T., S. Fisher, S. Olofsson, T. Endo, D. Groth, A. Tarentino, D.R. Borchelt, D. Teplow, L. Hood, and A. Burlingame. 1989. Asparagine-linked glycosylation of the scrapie and cellular prion proteins. Archives of biochemistry and biophysics. 274:1–13. Harding, H.P., I. Novoa, Y. Zhang, H. Zeng, R. Wek, M. Schapira, and D. Ron. 2000. Regulated translation initiation controls stress-induced gene expression in mammalian cells. Molecular cell. 6:1099–108. Harris, D.A. 2003. Trafficking, turnover and membrane topology of PrP. British medical bulletin. 66:71–85. Haze, K., H. Yoshida, H. Yanagi, T. Yura, and K. Mori. 1999. Mammalian transcription factor ATF6 is synthesized as a transmembrane protein and activated by proteolysis in Misiewicz 88 response to endoplasmic reticulum stress. Molecular biology of the cell. 10:3787–99. Hegde, R.S., J.A. Mastrianni, M.R. Scott, K.A. DeFea, P. Tremblay, M. Torchia, S.J. DeArmond, S.B. Prusiner, and V.R. Lingappa. 1998. A transmembrane form of the prion protein in neurodegenerative disease. Science. 279:827–34. Hetz, C. 2012. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nature reviews. Molecular cell biology. 13:89–102. Hetz, C., and L.H. Glimcher. 2009. Fine-tuning of the unfolded protein response: Assembling the IRE1alpha interactome. Molecular cell. 35:551–61. Hong, M., S. Luo, P. Baumeister, J.-M. Huang, R.K. Gogia, M. Li, and A.S. Lee. 2004. Underglycosylation of ATF6 as a novel sensing mechanism for activation of the unfolded protein response. The Journal of biological chemistry. 279:11354–63. Hosokawa, T., K. Tsuchiya, I. Sato, N. Takeyama, S. Ueda, Y. Tagawa, K.M. Kimura, I. Nakamura, G. Wu, A. Sakudo, C. Casalone, M. Mazza, M. Caramelli, H. Takahashi, T. Sata, K. Sugiura, A. Baj, A. Toniolo, and T. Onodera. 2008. A monoclonal antibody (1D12) defines novel distribution patterns of prion protein (PrP) as granules in nucleus. Biochemical and biophysical research communications. 366:657–63. Jodoin, J., S. Laroche-Pierre, C.G. Goodyer, and A.C. Leblanc. 2007. Defective retrotranslocation causes loss of anti-Bax function in human familial prion protein mutants. Journal of Neuroscience. 27:5081–5091. Jodoin, J., M. Misiewicz, P. Makhijani, P.N. Giannopoulos, J. Hammond, C.G. Goodyer, and A.C. Leblanc. 2009. Loss of anti-Bax function in Gerstmann-Str{ä}ussler-Scheinker syndrome-associated prion protein mutants. PLoS ONE. 4:e6647. Kakiuchi, C., M. Ishiwata, A. Hayashi, and T. Kato. 2006. XBP1 induces WFS1 through an endoplasmic reticulum stress response element-like motif in SH-SY5Y cells. Journal of neurochemistry. 97:545–55. Kleene, S.C. 1951. Representation of events in nerve nets and finite automata. Rand Corp., Santa Monica, Calif. Kokame, K., H. Kato, and T. Miyata. 2001. Identification of ERSE-II, a new cis-acting element responsible for the ATF6-dependent mammalian unfolded protein response. The Misiewicz 89 Journal of biological chemistry. 276:9199–205. Kuwahara, C., A.M. Takeuchi, T. Nishimura, K. Haraguchi, A. Kubosaki, Y. Matsumoto, K. Saeki, T. Yokoyama, S. Itohara, and T. Onodera. 1999. Prions prevent neuronal cellline death. Nature. 400:225–226. Lander, E.S., L.M. Linton, B. Birren, C. Nusbaum, M.C. Zody, J. Baldwin, K. Devon, K. Dewar, M. Doyle, W. FitzHugh, R. Funke, D. Gage, K. Harris, a Heaford, J. Howland, L. Kann, J. Lehoczky, R. LeVine, P. McEwan, K. McKernan, J. Meldrim, J.P. Mesirov, C. Miranda, W. Morris, J. Naylor, C. Raymond, M. Rosetti, R. Santos, a Sheridan, C. Sougnez, N. Stange-Thomann, N. Stojanovic, a Subramanian, D. Wyman, J. Rogers, J. Sulston, R. Ainscough, S. Beck, D. Bentley, J. Burton, C. Clee, N. Carter, a Coulson, R. Deadman, P. Deloukas, a Dunham, I. Dunham, R. Durbin, L. French, D. Grafham, S. Gregory, T. Hubbard, S. Humphray, a Hunt, M. Jones, C. Lloyd, a McMurray, L. Matthews, S. Mercer, S. Milne, J.C. Mullikin, a Mungall, R. Plumb, M. Ross, R. Shownkeen, S. Sims, R.H. Waterston, R.K. Wilson, L.W. Hillier, J.D. McPherson, M. a Marra, E.R. Mardis, L. a Fulton, a T. Chinwalla, K.H. Pepin, W.R. Gish, S.L. Chissoe, M.C. Wendl, K.D. Delehaunty, T.L. Miner, a Delehaunty, J.B. Kramer, L.L. Cook, R.S. Fulton, D.L. Johnson, P.J. Minx, S.W. Clifton, T. Hawkins, E. Branscomb, P. Predki, P. Richardson, S. Wenning, T. Slezak, N. Doggett, J.F. Cheng, a Olsen, S. Lucas, C. Elkin, E. Uberbacher, et al. 2001. Initial sequencing and analysis of the human genome. Nature. 409:860–921. Laroche-Pierre, S., J. Jodoin, and A.C. LeBlanc. 2009. Helix 3 is necessary and sufficient for prion protein’s anti-Bax function. Journal of neurochemistry. 108:1019–31. Lawson, V. a, S.J. Collins, C.L. Masters, and A.F. Hill. 2005. Prion protein glycosylation. Journal of neurochemistry. 93:793–801. LeBlanc, A. 1995. Increased production of 4 kDa amyloid beta peptide in serum deprived human primary neuron cultures: possible involvement of apoptosis. The Journal of neuroscience. 15:7837–7846. Lee, H., and H. Tang. 2012. Next-generation sequencing technologies and fragment assembly algorithms. Methods in molecular biology. 855:155–74. Lee, K., W. Tirasophon, X. Shen, M. Michalak, R. Prywes, T. Okada, H. Yoshida, K. Mori, Misiewicz 90 and R.J. Kaufman. 2002. IRE1-mediated unconventional mRNA splicing and S2Pmediated ATF6 cleavage merge to regulate XBP1 in signaling the unfolded protein response. Genes & development. 16:452–66. Lenaz, G. 2012. Advances in Mitochondrial Medicine. R. Scatena, P. Bottoni, and B. Giardina, editors. Springer Netherlands, Dordrecht. Li, H., A. V Korennykh, S.L. Behrman, and P. Walter. 2010. Mammalian endoplasmic reticulum stress sensor IRE1 signals by dynamic clustering. Proceedings of the National Academy of Sciences of the United States of America. 107:16113–8. Li, R., D. Liu, G. Zanusso, T. Liu, J.D. Fayen, J.H. Huang, R.B. Petersen, P. Gambetti, and M.S. Sy. 2001. The expression and potential function of cellular prion protein in human lymphocytes. Cellular immunology. 207:49–58. Liang, J., F. Bai, G. Luo, and J. Wang. 2007. Hypoxia induced overexpression of PrP(C) in gastric cancer cell lines. Cancer Biology. 769–774. Lin, D.T.S., J. Jodoin, M. Baril, C.G. Goodyer, and A.C. Leblanc. 2008. Cytosolic prion protein is the predominant anti-Bax prion protein form: exclusion of transmembrane and secreted prion protein forms in the anti-Bax function. Biochimica et biophysica acta. 1783:2001–12. Lin, J.H., H. Li, Y. Zhang, D. Ron, and P. Walter. 2009. Divergent effects of PERK and IRE1 signaling on cell viability. PLoS One. 4:e4170. Linden, R., V.R. Martins, M.A.M. Prado, M. Cammarota, I. Izquierdo, and R.R. Brentani. 2008. Physiology of the prion protein. Physiological Reviews. 88:673–728. Ma, J., R. Wollmann, and S. Lindquist. 2002. Neurotoxicity and neurodegeneration when PrP accumulates in the cytosol. Science. 298:1781–5. Mahal, S.P., E.A. Asante, M. Antoniou, and J. Collinge. 2001. Isolation and functional characterisation of the promoter region of the human prion protein gene. Gene. 268:105– 114. Mai, B., and L. Breeden. 1997. Xbp1, a stress-induced transcriptional repressor of the Saccharomyces cerevisiae Swi4/Mbp1 family. Molecular and cellular biology. 17:6491–501. McCullough, K.D., J.L. Martindale, L.O. Klotz, T.Y. Aw, and N.J. Holbrook. 2001. Misiewicz 91 Gadd153 sensitizes cells to endoplasmic reticulum stress by down-regulating Bcl2 and perturbing the cellular redox state. Molecular and cellular biology. 21:1249–59. Miele, G., A.R. Alejo Blanco, H. Baybutt, S. Horvat, J. Manson, and M. Clinton. 2003. Embryonic activation and developmental expression of the murine prion protein gene. Gene expression. 11:1–12. Mobley, W.C., R.L. Neve, S.B. Prusiner, and M.P. McKinley. 1988. Nerve growth factor increases mRNA levels for the prion protein and the beta-amyloid protein precursor in developing hamster brain. Proceedings of the National Academy of Sciences of the United States of America. 85:9811–5. Moore, R.C., I.Y. Lee, G.L. Silverman, P.M. Harrison, R. Strome, C. Heinrich, a Karunaratne, S.H. Pasternak, M. a Chishti, Y. Liang, P. Mastrangelo, K. Wang, a F. Smit, S. Katamine, G. a Carlson, F.E. Cohen, S.B. Prusiner, D.W. Melton, P. Tremblay, L.E. Hood, and D. Westaway. 1999. Ataxia in prion protein (PrP)-deficient mice is associated with upregulation of the novel PrP-like protein doppel. Journal of molecular biology. 292:797–817. Moreno, J. a., H. Radford, D. Peretti, J.R. Steinert, N. Verity, M.G. Martin, M. Halliday, J. Morgan, D. Dinsdale, C. a. Ortori, D. a. Barrett, P. Tsaytler, A. Bertolotti, A.E. Willis, M. Bushell, and G.R. Mallucci. 2012. Sustained translational repression by eIF2α-P mediates prion neurodegeneration. Nature. 485:507–11. Moro, F., C. Sirrenberg, H.C. Schneider, W. Neupert, and M. Brunner. 1999. The TIM17.23 preprotein translocase of mitochondria: composition and function in protein transport into the matrix. The EMBO journal. 18:3667–75. Morris, T., B. Robertson, and M. Gallagher. 1996. Rapid reverse transcription-PCR detection of hepatitis C virus RNA in serum by using the TaqMan fuorogenic detection system. Journal of clinical microbiology. 34:2933–6. Morrison, T.B., J.J. Weis, and C.T. Wittwer. 1998. Quantification of low-copy transcripts by continuous SYBR Green I monitoring during amplification. BioTechniques. 24:954–8, 960, 962. Munro, S., and H.R.B. Pelham. 1986. An Hsp70-like protein in the ER: identity with the 78 kd glucose-regulated protein and immunoglobulin heavy chain binding protein. Cell. Misiewicz 92 46:291–300. Ng, S.B., E.H. Turner, P.D. Robertson, S.D. Flygare, A.W. Bigham, C. Lee, T. Shaffer, M. Wong, A. Bhattacharjee, E.E. Eichler, M. Bamshad, D. a Nickerson, and J. Shendure. 2009. Targeted capture and massively parallel sequencing of 12 human exomes. Nature. 461:272–6. Norrby, E. 2011. Prions and protein-folding diseases. Journal of internal medicine. 270:1–14. Oltvai, Z.N., C.L. Milliman, and S.J. Korsmeyer. 1993. Bcl-2 heterodimerizes in vivo with a conserved homolog, Bax, that accelerates programmed cell death. Cell. 74:609–19. Oyadomari, S., and M. Mori. 2004. Roles of CHOP/GADD153 in endoplasmic reticulum stress. Cell death and differentiation. 11:381–9. Ozcan, L., and I. Tabas. 2012. Role of endoplasmic reticulum stress in metabolic disease and other disorders. Annual review of medicine. 63:317–28. Paitel, E., R. Fahraeus, and F. Checler. 2003. Cellular prion protein sensitizes neurons to apoptotic stimuli through Mdm2-regulated and p53-dependent caspase 3-like activation. The Journal of biological chemistry. 278:10061–6. Parry, H.B. 1962. Scrapie: A transmissible and hereditary disease of sheep. Heredity. 17:75–105. Perraud, A.-L., B. Shen, C.A. Dunn, K. Rippe, M.K. Smith, M.J. Bessman, B.L. Stoddard, and A.M. Scharenberg. 2003. NUDT9, a member of the Nudix hydrolase family, is an evolutionarily conserved mitochondrial ADP-ribose pyrophosphatase. The Journal of biological chemistry. 278:1794–801. Pfaff, M.W. 2001. A new mathematical model for relative quantification in real-time RTPCR. Nucleic acids research. 29:e45. Prusiner, S.B. 1982. Novel proteinaceous infectious particles cause scrapie. Science. 216:136–144. Qin, K., L. Zhao, R.D. Ash, W.F. McDonough, and R.Y. Zhao. 2009. ATM-mediated transcriptional elevation of prion in response to copper-induced oxidative stress. The Journal of biological chemistry. 284:4582–93. Misiewicz 93 Rane, N.S., J.L. Yonkovich, and R.S. Hegde. 2004. Protection from cytosolic prion protein toxicity by modulation of protein translocation. The EMBO journal. 23:4550–9. Rasband, W.S. 2011. ImageJ. Rivera-Milla, E., B. Oidtmann, C.H. Panagiotidis, M. Baier, T. Sklaviadis, R. Hoffmann, Y. Zhou, G.P. Solis, C. a O. Stuermer, and E. Málaga-Trillo. 2006. Disparate evolution of prion protein domains and the distinct origin of Doppel- and prion-related loci revealed by fish-to-mammal comparisons. FASEB journal. 20:317–9. Roucou, X., P.N. Giannopoulos, Y. Zhang, J. Jodoin, C.G. Goodyer, and A. LeBlanc. 2005. Cellular prion protein inhibits proapoptotic Bax conformational change in human neurons and in breast carcinoma MCF-7 cells. Cell death and differentiation. 12:783–95. Roucou, X., Q. Guo, Y. Zhang, C.G. Goodyer, and A.C. LeBlanc. 2003. Cytosolic prion protein is not toxic and protects against Bax-mediated cell death in human primary neurons. J Biol Chem. 278:40877–40881. Russell, S., and P. Norvig. 2010. Artificial intelligence : a modern approach. 3rd ed. Prentice Hall, Upper Saddle River. Sakaguchi, S., S. Katamine, N. Nishida, and R. Moriuchi. 1996. Loss of cerebellar Purkinje cells in aged mice homozygous for a disrupted PrP gene. Nature. Sanger, F., S. Nicklen, and A.R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proceedings of the National Academy of Sciences of the United States of America. 74:5463–7. Sayers, E.W., T. Barrett, D.A. Benson, E. Bolton, S.H. Bryant, K. Canese, V. Chetvernin, D.M. Church, M. Dicuccio, S. Federhen, M. Feolo, I.M. Fingerman, L.Y. Geer, W. Helmberg, Y. Kapustin, S. Krasnov, D. Landsman, D.J. Lipman, Z. Lu, T.L. Madden, T. Madej, D.R. Maglott, A. Marchler-Bauer, V. Miller, I. Karsch-Mizrachi, J. Ostell, A. Panchenko, L. Phan, K.D. Pruitt, G.D. Schuler, E. Sequeira, S.T. Sherry, M. Shumway, K. Sirotkin, D. Slotta, A. Souvorov, G. Starchenko, T.A. Tatusova, L. Wagner, Y. Wang, W.J. Wilbur, E. Yaschenko, and J. Ye. 2012. Database resources of the National Center for Biotechnology Information. Nucleic acids research. 40:D13–25. Schmitt-Ulms, G., S. Ehsani, J.C. Watts, D. Westaway, and H. Wille. 2009. Evolutionary Misiewicz 94 descent of prion genes from the ZIP family of metal ion transporters. PLoS ONE. 4:e7208. Schröder, M., and R.J. Kaufman. 2005. ER stress and the unfolded protein response. Mutation research. 569:29–63. Seabold, G.K., P.Y. Wang, K. Chang, C.-Y. Wang, Y.-X. Wang, R.S. Petralia, and R.J. Wenthold. 2008. The SALM family of adhesion-like molecules forms heteromeric and homomeric complexes. The Journal of biological chemistry. 283:8395–405. Shyu, W.-C., H.-J. Harn, K. Saeki, A. Kubosaki, Y. Matsumoto, T. Onodera, C.-J. Chen, Y.D. Hsu, and Y.-H. Chiang. 2002. Molecular modulation of expression of prion protein by heat shock. Molecular neurobiology. 26:1–12. Shyu, W.-C., S.-Z. Lin, K. Saeki, A. Kubosaki, Y. Matsumoto, T. Onodera, M.-F. Chiang, P. Thajeb, and H. Li. 2004. Hyperbaric oxygen enhances the expression of prion protein and heat shock protein 70 in a mouse neuroblastoma cell line. Cellular and molecular neurobiology. 24:257–68. Smith, T.F., and M.S. Waterman. 1981. Identification of common molecular subsequences. Journal of molecular biology. 147:195–7. Stöckel, J., J. Safar, A.C. Wallace, F.E. Cohen, and S.B. Prusiner. 1998. Prion protein selectively binds copper(II) ions. Biochemistry. 37:7185–93. Tabuchi, Y., I. Takasaki, and T. Kondo. 2006. Identification of genetic networks involved in the cell injury accompanying endoplasmic reticulum stress induced by bisphenol A in testicular Sertoli cells. Biochemical and biophysical research communications. 345:1044–50. Takekawa, M., and H. Saito. 1998. A family of stress-inducible GADD45-like proteins mediate activation of the stress-responsive MTK1/MEKK4 MAPKKK. Cell. 95:521–530. Takekawa, M., K. Tatebayashi, F. Itoh, M. Adachi, K. Imai, and H. Saito. 2002. Smaddependent GADD45beta expression mediates delayed activation of p38 MAP kinase by TGF-beta. The EMBO journal. 21:6473–82. Thompson, K. 1968. Programming Techniques: Regular expression search algorithm. Communications of the ACM. 11:419–422. Torres, M., K. Castillo, R. Armisén, A. Stutzin, C. Soto, and C. Hetz. 2010. Prion Protein Misfolding Affects Calcium Homeostasis and Sensitizes Cells to Endoplasmic Reticulum Misiewicz 95 Stress. PLoS ONE. 5:e15658. Vattem, K.M., and R.C. Wek. 2004. Reinitiation involving upstream ORFs regulates ATF4 mRNA translation in mammalian cells. Proceedings of the National Academy of Sciences of the United States of America. 101:11269–74. Vincent, B., C. Sunyach, H.-D. Orzechowski, P. St George-Hyslop, and F. Checler. 2009. p53-Dependent transcriptional control of cellular prion by presenilins. The Journal of neuroscience. 29:6752–60. Walz, R., O.B. Amaral, I.C. Rockenbach, R. Roesler, I. Izquierdo, E.A. Cavalheiro, V.R. Martins, and R.R. Brentani. 1999. Increased sensitivity to seizures in mice lacking cellular prion protein. Epilepsia. 40:1679–82. Wang, E., and D.S. Snyder. 1998. Handbook of the aging brain. Academic Press, San Diego. Wang, W.-A., J. Groenendyk, and M. Michalak. 2012a. Calreticulin signaling in health and disease. The international journal of biochemistry & cell biology. 44:842–6. Wang, X., A. Spandidos, H. Wang, and B. Seed. 2012b. PrimerBank: a PCR primer database for quantitative gene expression analysis, 2012 update. Nucleic acids research. 40:D1144–9. Westaway, D., N. Daude, S. Wohlgemuth, and P. Harrison. 2011. The PrP-like proteins Shadoo and Doppel. Topics in current chemistry. 305:225–56. Westergard, L., H.M. Christensen, and D.A. Harris. 2007. The cellular prion protein (PrP(C)): its physiological function and role in disease. Biochimica et biophysica acta. 1772:629–44. Wilkinson, B., and H.F. Gilbert. 2004. Protein disulfide isomerase. Biochimica et biophysica acta. 1699:35–44. Wong, B.S., T. Liu, R. Li, T. Pan, R.B. Petersen, M.A. Smith, P. Gambetti, G. Perry, J.C. Manson, D.R. Brown, and M.S. Sy. 2001. Increased levels of oxidative stress markers detected in the brains of mice devoid of prion protein. Journal of neurochemistry. 76:565– 72. Yamamoto, K., H. Yoshida, K. Kokame, R.J. Kaufman, and K. Mori. 2004. Differential Misiewicz 96 contributions of ATF6 and XBP1 to the activation of endoplasmic reticulum stressresponsive cis-acting elements ERSE, UPRE and ERSE-II. Journal of biochemistry. 136:343– 50. Yan, J., and R.W. Aldrich. 2012. BK potassium channel modulation by leucine-rich repeat-containing proteins. Proceedings of the National Academy of Sciences of the United States of America. Ye, J., R.B. Rawson, R. Komuro, X. Chen, U.P. Davé, R. Prywes, M.S. Brown, and J.L. Goldstein. 2000. ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs. Molecular cell. 6:1355–64. Yedidia, Y., L. Horonchik, S. Tzaban, A. Yanai, and A. Taraboulos. 2001. Proteasomes and ubiquitin are involved in the turnover of the wild-type prion protein. The EMBO journal. 20:5383–91. Yin, X.M., Z.N. Oltvai, and S.J. Korsmeyer. 1994. BH1 and BH2 domains of Bcl-2 are required for inhibition of apoptosis and heterodimerization with Bax. Nature. 369:321–3. Yoshida, H., K. Haze, H. Yanagi, T. Yura, and K. Mori. 1998. Identification of the cisacting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins. Involvement of basic leucine zipper transcription factors. The Journal of biological chemistry. 273:33741–9. Yoshida, H., T. Matsui, A. Yamamoto, T. Okada, and K. Mori. 2001. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell. 107:881–91. Yoshida, H., T. Okada, K. Haze, H. Yanagi, T. Yura, M. Negishi, and K. Mori. 2000. ATF6 activated by proteolysis binds in the presence of NF-Y (CBF) directly to the cis-acting element responsible for the mammalian unfolded protein response. Molecular and cellular biology. 20:6755–67. Zhan, Q., K.A. Lord, I. Alamo, M.C. Hollander, F. Carrier, D. Ron, K.W. Kohn, B. Hoffman, D.A. Liebermann, and A.J. Fornace. 1994. The gadd and MyD genes define a novel set of mammalian genes encoding acidic proteins that synergistically suppress cell growth. Molecular and cellular biology. 14:2361–71. Misiewicz 97 Misiewicz 98 Ethics approval Misiewicz 99 Misiewicz 100