ENDOPLASMIC RETICULUM STRESS DISRUPTS THERAPEUTIC PROTEIN SECRETION AND SIRNA PROCESSING By Ryan Chauncey Splichal A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemical Engineering – Doctor of Philosophy 2025 ABSTRACT Therapeutic proteins and small interfering RNA (siRNA) therapeutics are emerging technologies to increase beneficial protein levels or reduce disease causing protein expression. To develop new therapeutics, critical challenges relating to the cost of production and delivery to the cells of interest will need to be overcome. The cell membrane, as part of the endomembrane system, is a barrier to both secretion of therapeutic proteins from cells and entry of siRNA therapeutics into cells. Disruption of normal cell function due to environmental stresses is known to alter the function of the endomembrane system. The effects of this disruption on the production and function of biotherapeutics remains poorly understood. To inform protein production process and siRNA therapeutic designs, we examined how environmental stresses altered protein secretion and siRNA processing. Industrial scale production of many therapeutic proteins relies on mammalian cells grown in batch bioreactors. To maximize productivity, the bioreactor contents must be mixed. However, excessive mixing can damage cells through shear. As a result, pockets of low nutrients and oxygen, high or low temperature, and cellular waste products can form. The cellular response to these types of environmental stresses is often to increase the production of protective proteins, which when combined with the burden of therapeutic protein production, overwhelms the cellular protein quality control system and activates the Unfolded Protein Response (UPR). Using small molecule inhibitors and activators, this research investigated how the UPR-activated protein degradation pathways of autophagy and proteasomal degradation impacted protein production. Interestingly, we showed that increasing proteasomal degradation improved protein secretion. Obesity, smoking, hypertension, and diabetes activate the UPR in patients’ cells. For patients receiving siRNA-based therapeutics, activation of UPR is likely to alter the intracellular processing of the siRNAs. We compared the changes in function of cationic lipid delivered siRNAs between control cells and cells with the UPR activated. Using chemical inhibitors and activators, we found that endosome/autophagy crosstalk is linked to differences in siRNA accumulation, distribution, and activity. Our results will inform future siRNA therapeutic designs by reinforcing that accumulation of siRNAs in a cell does not necessarily correlate to silencing levels and that the disease state of individual patients may affect the activity of siRNA therapeutics. This thesis is dedicated to Mom and Dad, my family, and everyone I have learned from in life. Thank you for always fostering my curiosity and encouraging me to push for excellence. iii ACKNOWLEDGEMENTS I would like to thank my advisors, Dr. Walton and Dr. Chan, for their dedication to me as a pupil and demonstrating how to balance questioning and curiosity with confidence. I would like to thank the undergraduate researchers who have helped me in this work through assistance with cell culture, nucleic acid preparation, and most importantly comradery including, Hannah Cavagnetto, McKenna Coskie, Ryan Jinn, Emma Smith, and Courtney Blair. I would like to thank Dr. Daniel Vocelle not only for his mentorship and support at the beginning of my thesis but also for assistance gathering data through the MSU Flow Cytometry Core. Likewise, I would like to thank Dr. Melinda Frame for assisting with Confocal Microscopy. I am grateful for the funding provided for this research from the National Institute of General Medical Sciences (NIGMS), National Science Foundation, and most importantly the Integrated Pharmacological Sciences Training Program for funding and providing me the biological science foundation to investigate this dissertation. Finally, I am grateful for the graduate students and friends who have kept me sane through their friendship and time playing sports while also offering scientific and personal advice whenever needed. iv TABLE OF CONTENTS LIST OF ABBREVIATIONS ................................................................................................vi CHAPTER 1: INTRODUCTION ..........................................................................................1 1.1 Significance ...........................................................................................................1 1.2 Endomembrane System .........................................................................................1 1.3 ER Stress ...............................................................................................................3 1.4 Protein Therapeutics Challenges and Limitations ..................................................8 1.5 siRNA Therapeutics Challenges and Limitations ..................................................10 1.6 Specific Aims ........................................................................................................14 CHAPTER 2: SECRETED PROTEIN PRODUCTION IS IMPROVED BY CONTROLLING ER STRESS ASSOCIATED PROTEIN DEGRADATION. ....................15 2.1 Introduction ...........................................................................................................15 2.2 Results ...................................................................................................................17 2.3 Discussion ..............................................................................................................27 CHAPTER 3: ER STRESS INDUCED AUTOPHAGY ALTERS CELLULAR PROCESSING OF CATIONIC LIPID DELIVERED SIRNAS ............................................30 3.1 Introduction ...........................................................................................................30 3.2 Results ...................................................................................................................32 3.3 Discussion ..............................................................................................................37 CHAPTER 4: CONCLUSIONS AND FUTURE DIRECTIONS ..........................................39 4.1 Conclusions ...........................................................................................................39 4.2 Future Directions ...................................................................................................39 BIBLIOGRAPHY ..................................................................................................................42 APPENDIX A: SUPPLEMENTAL MATERIAL FOR CHAPTER 2 ...................................55 APPENDIX B: MODULATING POLYMER-SIRNA BINDING DOES NOT PROMOTE POLYPLEX MEDIATED SILENCING ............................................................57 APPENDIX C: LIPOPOLY SACCHARIDES ALTER ENDOCYTOSIS PATTERNS OF CATIONIC-LIPID DELIVERED SIRNA .......................................................................69 APPENDIX D: NUCLEIC ACID CARGO DOES NOT IMPACT LIPOFECTAMINE 2000 ENDOCYTOSIS PATTERNS ......................................................................................71 APPENDIX E: ROLIPRAM IMPROVES PROTEIN PRODUCTION IN HEK293 CELLS TRANSFECTED WITH PDNA ...............................................................................72 APPENDIX F: METHODS AND MATERIALS FOR CHAPTER 2 ...................................73 APPENDIX G: METHODS AND MATERIALS FOR CHAPTER 3 ...................................75 v LIST OF ABBREVIATIONS 3MA ACC2 ATF4 ATF6 3-Methyladenine Acetyl-CoA Carboxylase 2 Activating Transcription Factor 4 Activating Transcription Factor 6 AMPK AMP-activated Protein Kinase Ago2 Argonaute 2 ASPGR Asialoglycoprotein Receptor ATG Autophagy related protein BAFA1 Bafilomycin A1 BiP Binding Immunoglobuilin Protein CHOP CCAAT/enhancer-binding protein homologous protein CMA CHO PI3K Chaperone Mediated Autophagy Chinese Hamster Ovary class III Phosphoinositide 3-Kinase CuAAC Copper(1)-catalyzed Alkyne-Azide Cycloaddition CD Cytosolic Domain DAPK1 Death Associated Protein Kinase 1 DNA Deoxyribonucleic Acid DGAT2 Diacylglycerol O-acyltransferase 2 ER Endoplasmic Reticulum ERAD ER Associated Degradation ERdj3 ERdj4 ERdj5 eIF2α Gluc GRP78 HSPA5 HSPA8 IRE1α ER DnaJ 3 ER DnaJ 4 ER DnaJ 5 eukaryotic Initiation Factor 2α Gaussia Luciferase Glucose-Regulated Protein 78 Heat Shock Protein A5 Heat Shock Protein A8 Inositol-Requiring transmembrane kinase/endoribonuclease 1α vi IFN-gamma Interferon-gamma IRES LNP LF2K LD Internal Ribosome Entry Site Lipid Nanoparticle Lipofectamine 2000 Luminal Domain mTORC1 mammalian Target of Rapamycin Complex 1 mRNA messenger RNA GalNAc N-Acetylgalactosamine PS pDNA PPGL PEG PEI PERK RIDD RNA RP S1P S2P siRNA SCD1 Phopshorothioate plasmid DNA Poly(Propargyl Glycolide) Polyethyeneglycol Polyethylenimine Protein Kinase R-like ER Kinase Regulated IRE1-dependent decay Ribonucleic Acid Rolipram Site 1 Protease Site 2 Protease Small interfering RNA Stearoyl-CoA desaturase-1 SREBP1 Sterol Regulatory Element Binding Protein 1 SREBP2 Sterol Regulatory Element Binding Protein 12 SEL1L TRBP TGN TM TMD UBR4 UBR5 ULK1 Suppressor Enhancer Lin-12-Like Tar RNA bindign protein Trans Golgi Network Tunicamycin Transmembrane Domain Ubiquitin Protein Ligase E3 Component N-Recognin 4 Ubiquitin Protein Ligase E3 Component N-Recognin 5 Unc-51-like Kinase 1 vii UPR XBP1 Unfolded Protein Response X-box Binding Protein 1 XBP1s X-box Binding Protein 1 - spliced viii CHAPTER 1: INTRODUCTION Note: Parts of this chapter have been modified from previously published work1. 1.1 Significance Biotherapeutics are one of the fastest growing segments of the pharmaceutical industry. In 2023, the global biotherapeutics market was valued at $478.20 billion and is expected to grow to $709.91 billion by 20282. While small molecule drugs are still the most used therapeutics, biotherapeutics allow for treatment of diseases that involve protein dysregulation. This dissertation focuses on two classes of biotherapeutics: protein-based therapeutics, which enhance the activity of an under-expressed, mutated, or missing protein; and small interfering RNA (siRNA) therapeutics that selectively down regulate the expression of disease causing proteins. Despite recent advances, widespread adoption of both biotherapeutics is limited by delivery challenges. Protein-based therapeutics are hindered by the cost of process development and production. siRNA therapeutics are limited by the ability of the nucleic acids to arrive at the correct location to achieve gene silencing without toxic effects. Both of these challenges share a common obstacle, the ability of a biotherapeutic to cross the cell membrane. The cell membrane is part of the endomembrane system that is controlled by the endoplasmic reticulum (ER). As the site of protein and lipid synthesis, the ER and protein quality control system can be disrupted by responses to environmental conditions in bioreactors or in vivo. The Unfolded Protein Response initiates a cascade of signaling events, including autophagy induction, which alter cell membrane characteristics. This dissertation investigates how perturbations to the endomembrane system can impact protein production and siRNA delivery and how this knowledge can further advance the state of these two biotherapeutics. 1.2 Endomembrane System The endomembrane system is a network of cellular components responsible for producing, maintaining, separating, and transporting membrane components and other biomolecules throughout the cell1,3 (Figure 1.1). All components in the system contain lipid bilayers that prevent large or charged molecules from passively transporting across4. Thus, the endomembrane system participates in both therapeutic protein production and nucleic acid uptake and release. 1 Figure 1.1 The endomembrane system linking the cell membrane, Golgi, lysosome, and other vesicles through control of shared membrane components. Beginning at the cell membrane, endosomes are trafficked to the early/sorting endosome. From the sorting endosomes, endosomal vesicles can be recycled or trafficked to the ER, Golgi, or lysosome. COPII coated vesicles transport material from the ER to the Golgi and COPI coated vesicles transport from the Golgi to the ER. Additional vesicles can leave the ER and Golgi for secretion or degradation in lysosomes. Lysosomal membrane components can be transported to the cell membrane to re-enter the endomembrane system. Created with BioRender.com The major components of the endomembrane system include the ER, Golgi apparatus, lysosomes, nuclear envelope, endosomes, and transport vesicles3,5. The ER is a primary site of protein and lipid synthesis; the Golgi apparatus then modifies and packages proteins for delivery to their specific destinations. Lysosomes degrade proteins and vesicles, recycling their resources, and the nuclear envelope separates the nucleus from the cytoplasm. Endosomes mediate the internalization of material from the cell membrane, and transport vesicles move cargo among endomembrane compartments5. The endomembrane system enables intracellular transport, membrane and protein production, and separates the cellular components from each other and extracellular material. 2 Secretion of recombinant proteins begins in the rough ER, where soluble proteins are translated and inserted into the ER lumen (Figure 1.1). In the ER, proteins are post- translationally modified, including disulfide bond formation and glycosylation6. Proteins are subsequently transported in COPII-coated vesicles from the ER to the Golgi apparatus, which is made up of a series of cis and trans compartments (cisternae)6,7. The trans Golgi network (TGN) is responsible for protein secretion, where clathrin-coated vesicles bud off the TGN and traffic the protein to the plasma membrane8. Simultaneously, retrograde transport from the Golgi to the ER occurs in COPI-coated vesicles9. The TGN also directs transport vesicles to early endosomes, where the fate of the vesicle contents is determined by the proteins present on the vesicle membrane (e.g., Rab4, Rab5, Rab7, or Rab11)10. Transport of membrane bound vesicles requires cytoskeletal rearrangement11. While there are additional mechanisms available to the cell for protein secretion (including lysosome secretion, transport through protein transporters, multivesicular body secretion, and membrane blebbing), the canonical secretion pathway through the TGN is the most common mechanism for protein secretion7. As the endomembrane system component that separates the cytoplasm from external material, the cell membrane acts as the main barrier to siRNA entry and its therapeutic effect12 12. To enter cells, siRNAs must be endocytosed13. Endocytosis starts with an initiation event that signals for an invagination of the cell membrane14. After the invagination grows and engulfs the material to be endocytosed, an endosome is cinched from the cell membrane12,14. The newly created vesicle is trafficked to the sorting endosome where the fate of the material is determined for recycling to the cell membrane, degradation in the lysosome, or further processing in the ER or Golgi15. 1.3 ER Stress Stresses from the environment, e.g., chemicals, pathogens, genetic manipulation, oxidative stress, and cytokines, induce ER stress and alter cellular protein demands. Increasing protein production in the cells can overwhelm the quality control machinery in the ER, leading to a buildup of unfolded or misfolded proteins which induces ER stress16. To cope with ER stress the unfolded protein response (UPR) is activated to restore proteostasis. The UPR is initiated by 3 sensor proteins: Inositol-Requiring transmembrane kinase/endoribonuclease 1α (IRE1α), PRK- like endoplasmic reticulum kinase (PERK), and Activating Transcription Factor 6 (ATF6). UPR activation (1) decreases global protein expression, (2) increases chaperone protein expression, 3 (3) increases protein degradation and secretion, and, if these measures do not restore proteostasis, (4) signals apoptosis17 (Figure 1.2). Figure 1.2 Summary of the Unfolded Protein Response (UPR). The UPR consists of three protein sensors of ER stress: IRE1α, PERK, and ATF6. In response to ER stress from protein overproduction or stress cues in bioreactors, IRE1α dimerizes and oligomerizes creating an active site for the splicing of XBP1 mRNA which is then translated into the XBP1s transcription factor. PERK dimerizes to phosphorylate and activate transcription factors eIF2α and ATF4. Activated ATF6 is trafficked to the Golgi where site-1-protease (S1P) and site-2-protease (S2P) cleaves the ER portion and cytosolic portion of the protein to create the nuclear factor of ATF6. Together these branches decrease global protein expression, increase chaperone expression, increase proteolysis, and - if ER stress persists – activate apoptosis. ER stress signaling therefore can alter the number of cells producing recombinant protein and the amount of protein each cell produces. Created with BioRender.com IRE1α, a ubiquitously expressed serine-threonine kinase, is a type I transmembrane protein of the ER with an N-terminal luminal domain (LD) that acts as an ER stress sensor, a C- terminal cytosolic domain (CD) containing both Ser/Thr kinase and endoribonuclease (RNase) activities, and a transmembrane domain (TMD) that senses membrane lipid saturation18,19. During ER stress, misfolded proteins bind binding immunoglobulin protein (BiP) (also known as 78 kDa glucose-regulated protein, GRP78, or heat shock protein A5, HSPA5), releasing IRE1α and allowing the LD of IRE1α to dimerize/oligomerize and initiate signaling20. Dimerization/oligomerization facilitates trans-autophosphorylation and the activation of 4 the RNase domain21. This active form of IRE1α catalyzes the unconventional processing of the mRNA encoding X-Box binding protein-1 (XBP1) by splicing a 26-nucleotide intron from the XBP1 mRNA to generate the coding sequence for an active transcription factor, spliced XBP1 (XBP1s)22. XBP1s upregulates genes involved in enhancing ER protein-folding capacity and degrading unfolded or misfolded proteins23. In addition to catalyzing the splicing of XBP1, IRE1α induces regulated IRE1-dependent decay (RIDD), which results in the degradation of RNAs including mRNAs, microRNAs, and ribosomal RNAs24,25 with XBP1-like endomotifs (consensus sequence CNG|CAGN). However, RIDD can also degrade mRNAs without such motifs through a more promiscuous, endomotif-independent processing that requires phospho- oligomers26. Through RIDD, IRE1α promotes the degradation of mRNAs encoding ER-targeted proteins to reduce the protein load in the ER27,28. As with IRE1α, ER stress promotes dimerization of the LD of PERK. PERK belongs to the eukaryotic translation initiation factor 2α (eIF2α) kinase subfamily, containing a Ser/Thr kinase domain in the cytosol. Upon dimerization of the LD, the cytosolic kinase domain undergoes activation by trans-autophosphorylation29. Activated PERK phosphorylates eIF2α at Ser 51, and phosphorylated eIF2α impedes global translation initiation, decreasing the protein expression load in the ER. However, activated eIF2α also increases the translation of activating transcription factor 4 (ATF4)30. Under prolonged ER stress, ATF4 activates CCAAT/enhancer- binding protein homologous protein (CHOP), which contributes to the upregulation of apoptotic pathways31,32. ER stress reduces protein expression by phosphorylating eIF2α, which prevents eIF2ß guanine nucleotide exchange factor from converting eIF2α back into the active form. This prevents recognition of mRNAs and further translation. Chaperone proteins (e.g., HSP70), essential proteins (e.g., insulin receptor), and viral RNAs (e.g., picornaviral RNAs) avoid decreased expression by using an internal ribosome entry site (IRES) that is recognized by ribosomes in an eIF2α-independent manner33. IRESs are currently used in antibody production to ensure both chains of the antibody are expressed once the mRNA is recognized at the ribosome34. Further use of IRESs to avoid decreased stress-induced translation is worth considering. While IRE1α and PERK are type I transmembrane proteins with single α-helical TMDs and cytosolic kinase domains, ATF6α is a type II transmembrane transcription factor containing 5 several α-helical TMDs and a DNA-binding domain with a basic leucine zipper motif35. Upon ER stress, ATF6 localizes to the Golgi apparatus and is further cleaved by site 1 and 2 proteases (S1P, S2P) allowing translocation of ATF6 to the nucleus to form the active transcription factor pATF6a36. An important role of ATF6 is to upregulate molecular chaperones and folding enzymes to increase the protein folding capacity of the ER. Additionally, if ER stress persists, ATF6 and PERK can work synergistically to induce CHOP and apoptosis37. ER stress also activates ER associated degradation (ERAD) as an additional mechanism for restoring proteostasis and protein quality control in the cell. The accumulation of misfolded proteins activates the UPR and ERAD clears the misfolded proteins through the cytosolic ubiquitin-proteasomal degradation pathway38. Proteins for degradation are identified, exported from the ER, ubiquitinated, and transported to the proteasomes for destruction. 6 Figure 1.3 Recombinant proteins exit the secretory pathway for degradation. Secreted recombinant proteins are synthesized by ribosomes in the ER lumen. Chaperone proteins ensure proper folding of synthesized proteins. From the ER, proteins enter the secretory pathway through transport vesicles bound for the Golgi. During ER stress events when the quality control machinery is overwhelmed, proteins can escape the secretory pathway for degradation to reduce the protein load on the ER and secretion machinery. Proteins can be ubiquitinated, transported across the ER membrane, and trafficked to proteasomes for degradation (ERAD). Autophagy can remove proteins from the secretory pathway by chaperone mediated trafficking to autophagosomes or by vesicle deviation to autophagosomes leading to lysosomal degradation of proteins instead of secretion. Created with BioRender.com Finally, in addition to ubiquitin-proteasome degradation, autophagy traffics material to lysosomes for degradation (Figure 1.3). Autophagy-lysosomal degradation can be divided into three categories: microautophagy, chaperone-mediated autophagy, and macroautophagy39. Microautophagy is the basal level breakdown of bulk materials, such as organelles and proteins by the lysosome through direct encapsulation39. Chaperone-mediated autophagy (CMA) occurs when chaperone proteins like Hsc70/HSPA8 recognize the CMA-targeting motif of misfolded proteins and translocate the protein to the lysosome for degradation40. CMA targets the KFERQ or similar motif and through careful sequence selection can be avoidable for recombinant proteins41. Macroautophagy, commonly known as autophagy, starts with the formation of a phagophore from ER-associated components that matures into a new membrane bound vesicle 7 called an autophagosome39. The autophagosome fuses with lysosomes for degradation of the autophagosome’s contents39. In contrast with proteasomal degradation, autophagy can degrade protein aggregates, organelles, and insoluble material39. For proteins that can be degraded by either proteasomes or autophagy, the half-life of the protein dictates the degradation pathway, with longer half-life proteins being more prone to degradation by autophagy42. 1.4 Protein Therapeutics Challenges and Limitations Recombinant proteins have a wide range of uses in medical clinics from diagnosis and prevention of diseases to cancer and infectious disease therapies. There are currently over 7,000 clinical trials and 894 FDA approved therapeutic proteins43. At present, therapeutic proteins are cost prohibitive with antibody therapeutics pricing well over 500,000 USD per year (Soliris/eculizumab)44 and down to about 3,375 USD per year for Rituximab for multiple sclerosis treatment45. There are many contributing factors to the price of therapeutic proteins. Feed material, technical training, and maintaining bioreactor conditions contribute to the cost, but the most expensive parts of therapeutic protein production are research and development of the therapeutic protein, research and development of an industrial scale production process meeting FDA standards, and purification of the product protein. Therapeutic proteins are produced in batch and semi-batch bioreactors46. Yeast and bacteria can be used to produce therapeutic peptides like insulin at reduced cost, but proteins requiring mammalian specific post-translational modifications use cell lines, such as Chinese Hamster Ovary (CHO), mouse myeloma (NS0), and Human Embryonic Kidney 293 (HEK293) cells16. Most therapeutic proteins are designed to be secreted into culture media to ease the purification of product proteins from the cells; however, CHO, NS0, and HEK293 are not specialized secretory cells and secretion can be a rate limiting step in their use for therapeutic protein production47. Industrial cell culture comes with many challenges that can alter protein synthesis and processing at the cellular level. Cell growth and activity in bioreactors can be impacted by shear stress from mixing48, low nutrient49 and gas50 pockets (imperfect mixing), temperature gradients51, and metabolites (Table 1). In response to stress, cells increase production of protective proteins. This combined with the increased protein production, overwhelms protein quality control machinery in the ER becomes overwhelmed and activates the UPR. Due to the nature of industrial cell processing, ER stress is unavoidable, however steps can be taken to 8 mitigate ER stress in bioreactors52. TABLE 1: ER stress induction by metabolites Metabolites/others ER stress sensor UPR activities Cell line Glucose IRE1α Palmitate, oleate IRE1α, PERK, ATF6 Palmitate, oleate, Stearate IRE1α, PERK 7-ketocholesterol*, 4- hydroxynonenal* IRE1α, PERK, ATF6 Insulin IRE1α Glucagon/epinephrine IRE1α Glucagon/(PA, OA, SA, LA at equimolar) IL-4 IRE1α IRE1α IRE1α phosphorylation, XBP1 splicing Beta cells53-55 IRE1α phosphorylation, PERK, eIF2a, XBP1 splicing Beta cells56 IRE1α phosphorylation, PERK, eIF2a, XBP1 splicing IRE1α phosphorylation, PERK, eIF2a, XBP1 splicing IRE1α phosphorylation, XBP1 splicing IRE1α phosphorylation, no XBP1 splicing IRE1α phosphorylation, XBP1 splicing IRE1α phosphorylation, XBP1 splicing Myeloid cells57 Endothelial cells58 Hepatocytes59 Hepatocytes60 Hepatocytes61 Macrophage62 In addition to the challenges of protein production, purification is also costly63. Separation techniques used for protein products must not damage the proteins, adjust to inconsistent feeds, and must reach FDA approved quality standards64. Protein A chromatography (PAC) is the most common purification method and is performed in batch or semi-batch mode even if the protein production is a continuous process, thereby making scale-up difficult65. In PAC, centrifuged or filtered culture media containing the protein product is passed through a column with immobilized protein A which binds the protein product. Then contaminants are eluted from the column. Next, the product proteins are removed by pH control and finally, the column is regenerated66. The quality of the proteins entering PAC can impact the efficiency, isoforms, dimers, aggregates, fragments, or bound host cell proteins increase the difficulty of separation66. The current method for process development and optimization involves directed selection 9 of cell lines producing the protein67. Cells are transfected, virally transduced, or electroporated to express the protein and selected genes to improve viability, secretion, and stability68. Cells that incorporate the pDNA into stable expression are selected, split into single cell isolates and placed into new culture containers. The cells that express the protein with highest quality, duration, and quantity are tested in pilot scale operations68. This process is inefficient in time and materials and does not guarantee that any successful clones will be generated that meet FDA approval69. In summary, lowering the production and development cost of protein therapeutics will increase the number of protein therapeutics that are economically viable treatment options. Any actions taken to increase the productivity of cell lines, increase cell line product purity, increase cell survivability, and improve process development will make strides toward this goal. 1.5 siRNA Therapeutics Challenges and Limitations Small interfering RNAs (siRNAs) are 15-30 nucleotide long double stranded RNA molecules that downregulate the expression of a protein by targeted degradation of an mRNA70. siRNAs function via the RNA interference (RNAi) pathway. In the cytoplasm, siRNAs associate with the RNA induced silencing complex where the passenger strand is degraded71. The guide strand along with Tar RNA binding protein (TRBP), Dicer, and Argonaute 2 (Ago2) then cleave the mRNA complementary to the guide strand preventing translation into protein72. Originally discovered by Fire and Mello in 199873, exogenous activation of the RNAi pathway has found clinical success with 7 FDA approved therapeutics targeting liver diseases and 156 active or recruiting clinical43,74-80. Clinical use begins when siRNA are injected intravenously or intramuscularly. First, there are limitations on siRNAs getting to the target tissue including serum nuclease activation, immune system activation, renal clearance, and binding to positively charged proteins81,82. Next siRNAs must be internalized by the target cells13,82. Finally, the siRNAs must cross the cell membrane or vesicle membrane to enter the cytoplasm81. To overcome these barriers to function, siRNAs are modified directly and/or complexed with a delivery vehicle81,82. Covalent modifications to siRNA are utilized to increase circulation time, decrease toxicity, and improve target tissue uptake without impacting the ability to cleave target mRNA. Modifications to the phosphate backbone increase circulation time by decreasing the enzymatic activity of serum nucleases83. The phosphate group of siRNAs can be modified by replacing one or more oxygens with sulfur ions or organic groups. Phopshorothioate (PS) modifications are 10 used by six of the seven FDA approved siRNA therapeutics to avoid serum nuclease degradation and increase circulation time83. Modifications to the 2’ oxygen (2’-deoxy-2’-fluoro and 2’-O- methyl) result in decrease immune system activation and toxicity84. Finally, the ends of siRNA can be covalently bound to ligands for receptor mediated endocytosis to increase the quantity entering cells. N-Acetylgalactosamine (GalNAc) is a liver targeting ligand that is covalently bound to all siRNAs actively in use85. Numerous other covalent modifications are made to siRNA to indirectly increase delivery83. In addition to covalent modifications, siRNAs can be packaged with delivery vehicles to achieve better incorporation into target cells. Because siRNAs are negatively charged, delivery vehicles are typically positively charged and can be generally divided into 4 categories: lipid- based, polymer-based, solid-core nanoparticles, and cell derived systems. Lipid-based carriers utilize amphiphilic molecules with charged heads and non-polar tails to encapsulate siRNAs86. In addition to siRNA, lipid nanoparticles (LNPs) are used in mRNA vaccines for COVID-1987,88. Polymeric carriers utilize positively charged polymers like polyethylenimine (PEI) and poly(l- lysine) to bind and encapsulate siRNAs89. Solid-core nanoparticles are made from inorganic materials including metals (gold, iron, silver) and silica in conjunction with either a polymer or lipid coating to charge the particle90. Cell derived delivery systems are lipids and proteins structures that originate in cells and have been isolated after secretion or rupturing of the cell and can be divided into two classes: membrane coated nanoparticles and extracellular vesicles91,92. siRNA can be incorporated into these nanostructures by expression in the parent cells or post- production replacement of natively incorporated RNA 93,94. Delivery vehicle strategies can be combined to create more classes and custom carriers like including polyethyeneglycol (PEG) in lipid-based delivery vehicles to decrease charge density and give structure95. To date only lipid- based delivery has seen clinical success with Patisiran, the first FDA approved therapeutic utilizing this technology74. Once siRNAs arrive at the target cells, they must cross the cell membrane to reach the cytoplasm and decrease protein expression81,4. Unlike small molecule drugs, siRNAs cannot passively diffuse across cell membranes and must be actively transported into cells, a process called endocytosis. In endocytosis, an initiating event, usually a receptor-ligand binding event causes the accumulation of proteins at the cell membrane that begin to curve the membrane into the cell, eventually forming an endosome14. From the membrane, endosomes are trafficked to the 11 early/sorting endosome where, based on the Rab protein content of the endosome membrane, the contents are either recycled to the cell membrane (Rab4, Rab5), mature toward lysosomes (Rab7), or translocate toward the ER (Rab1, Rab6) or Golgi (Rab1, Rab6, Rab9)96. The fate of siRNAs is determined by the destination of the endosomes; ultimately less than 1-2% of siRNAs that enter a cell actually participate in gene silencing97,98. Most are recycled out of the cell or degraded in lysosomes98. siRNA cellular entry alone does not mean it will escape the endosome and enter the cytoplasm. The exact timing and mechanism of endosomal escape for siRNAs is not well understood99. The current opinion of the field is most siRNAs that escape endosomes do so as early endosomes mature into late endosomes and lysosomes rather than during trafficking to the ER or Golgi or during endosome recycling. During maturation, proteins are being replaced in the cell membrane and acidification of the endosome is occurring that likely make the vesicle more permeable to siRNA100,101. There are 3 proposed mechanisms for siRNA release from endosomes. The first is the “Proton Sponge Effect” where hydrogen accepting groups in the delivery vehicle buffer the endosomal pH requiring more hydrogen and water to be transported into the endosome to lower the pH, eventually enough water accumulates in the endosome that it ruptures releasing the cargo into the cytoplasm102. This is the method for endosomal escape for some delivery vehicles, notably PEI. However, it is unlikely to be the mechanism for clinically relevant delivery vehicles because rupturing intracellular vesicles results in significant cytotoxicity102. The second reported mechanism is membrane fusion of the delivery vehicle and the endosome (or plasma) membrane. In this mechanism, the amphiphilic components of the delivery vehicle become increasingly non-polar as the pH of the endosome decreases. Eventually the electrostatic repulsion of the endosome membrane and delivery vehicle reduces and the delivery vehicle merges with or forms a pore in the endosome membrane, releasing the contents into the cytoplasm103. The final mechanism is direct protein transport across the membrane. In this mechanism, the siRNA (or siRNA-vehicle complex) is modified to interact with a membrane transport protein, and upon recognition, the transport protein actively internalizes the siRNA directly to the cytoplasm; however, mammals do not have functioning RNA transport proteins81. Learning how to best design delivery vehicles to administer siRNAs into the cytoplasm is an ongoing challenge. To overcome all of these challenges, currently approved siRNA therapeutics all use a 12 GalNAc-end modification and modified phosphate backbone75-80 (Patisiran, which used a lipid- based delivery system, has been largely replaced by a similar siRNA using the GalNAc modification74). GalNAc is a sugar that is recognized by the asialoglycoprotein receptor (ASPGR) in the liver. The backbone modifications allow for the circulating siRNA to go undetected by the clearance systems, and the GalNAc addition initiates endocytosis into liver cells. While the accumulation of GalNAc-siRNA is sufficient, endosomal escape is still quite inefficient with GalNAc-siRNA with Alnylam scientists reporting that only 0.3% of endocytosed siRNA enters the cytoplasm104. These therapeutics rely on the quantity and efficiency of delivery to the liver, the main blood detoxification organ. Future siRNAs will need to target additional tissues beyond the liver. In addition to using ligand modified siRNAs, ligands can be incorporated into delivery vehicles. Folate-, 2-[3-(1,3-dicarboxy propyl)-ureido pentanedioic acid) DUPA-, and aptamer-modified siRNAs have been used to target breast, prostate, and lung cancer105. None of these treatments have reached FDA approval and further identification of receptor-ligand combinations that are as efficient as the GalNAc/ASPGR combination will be crucial for further applications. Despite the recent success, challenges remain for the widespread use of siRNA therapeutics. Many past clinical trials have failed due to toxic and off-target effects106. As trial population size increases, patients with more and more varied comorbidities are incorporated into the study107. Common comorbidities such as obesity, diabetes, smoking, and hypertension are omitted from early clinical trial phases but are present in the general population of therapy recipients108. Toxic effects can be attributed to off-target effects, immune system activation, unanticipated effects from silencing a gene in another location, RNAi saturation, and delivery vehicle toxicity. There is a necessary careful balance between accumulating sufficient levels of siRNA to treat a disease while not accumulating so much as to cause toxicity. Overall, while siRNA therapeutics have achieved clinical success for 6 liver diseases, many challenges remain for widespread use. Efficient cellular entry and endosomal escape are critical barriers to siRNA clinical use beyond the liver. In addition, strides need to be made in the ability to predict in vivo toxicology from in vitro experiments to improve clinical trial success rates. Further research is needed to develop more effective and targeted siRNA delivery strategies to reach and achieve function in diverse cell types without increasing toxicity. 13 1.6 Specific Aims 1.6.1 Improving therapeutic protein production through degradation control Because ER stress is unavoidable in bioreactors, we investigated the components of ER stress that alter protein production including IRE1α activation, autophagy, and ERAD. Using tunicamycin (TM, ER stress inducer), HeLa cells, and pDNA and mRNA that code for a secreted luciferase, we showed that ER stress leads to lower secretion of proteins and higher intracellular accumulation. Since ER stress can alter the protein degradation balance between autophagy and proteasomal degradation, we used 3-methyladenine (3MA) to inhibit autophagy, VR23 to inhibit proteasomes and Rolipram (RP) proteasome activator to influence degradation in ER stress cells. Our results indicate that autophagy inhibition increases secretion under ER stress and that proteasome activation increases protein secretion under ER stress. Taken together our findings suggest that activating proteasomes to degrade intracellular proteins increases the secretory capacity of cells. In the future, these results should be validated in other cell lines and stress conditions, and the quality of secreted proteins should be assessed under small molecule alterations to degradation. 1.6.2 Investigating ER stress impact on siRNA delivery Using Lipofectamine 2000 (LF2K)-siRNA complexes and HeLa cells expressing EGFP, we examined how ER stress induced by TM influenced siRNA accumulation and silencing. We found that increasing TM concentrations lead to increased siRNA accumulation but reduced silencing efficacy. To further elucidate the differences in siRNA processing under ER stress, we used 3MA to inhibit ER stress induced autophagosome formation. TM and 3MA treatment decreased siRNA accumulation compared to TM treatment alone. Bafilomycin A1 (BAFA1) was used to inhibit endosome maturation and autophagosome acidification and showed that siRNA is being trafficked from early endosomes to autophagosomes. To investigate if the siRNA that is accumulating was still able to achieve silencing we allowed cells to recover in TM containing media for 5 days post siRNA transfection and demonstrated that ER stress prolonged the silencing effect of siRNA. Future studies should validate these results are consistent with other ER stress inducers like palmitate, thapsigargin, or glucose starvation. 14 CHAPTER 2: SECRETED PROTEIN PRODUCTION IS IMPROVED BY CONTROLLING ER STRESS ASSOCIATED DEGRADATION 2.1 Introduction Recombinant proteins have a wide range of uses in medicine from diagnosis and prevention of diseases to treatments for cancer and infectious diseases. Industrial cell culture of mammalian cells requires the use of temperature-controlled bioreactors with mixing of highly viscous cultures. Agitation causes shear stress and nonetheless leaves heterogeneities in temperature, nutrients, and oxygen48-51. In response to these kinds of environmental stressors, cells, especially those overexpressing a product protein, can experience a loss of proteostasis, causing the protein quality control machinery in the endoplasmic reticulum (ER) to become overwhelmed, a condition referred to as ER stress1,17,109. While steps can be taken to mitigate ER stress in bioreactors, due to the requirements of industrial bioprocesses, cells are always experiencing ER stress110-112. Cells respond to ER stress via the unfolded protein response (UPR). The UPR has three pathways mediated by three distinct sensor proteins: Inositol-Requiring Enzyme 1α (IRE1α),109 Protein Kinase R-like Endoplasmic Reticulum Kinase (PERK), and Activating Transcription Factor 6 (ATF6)1, 113. IRE1α dimerizes and autophosphorylates in the presence of misfolded proteins, activating an RNase domain that cleaves X-box Binding Protein-1 (XBP1) mRNA into the XBP1-spliced (XBP1-s) mRNA, which encodes for the transcription factor XBP1-s113. XBP1-s initiates signaling that leads to an increase in the size of the ER, upregulates chaperone protein synthesis, and activates ER-associated degradation (ERAD) of excess/misfolded proteins114. PERK dimerizes in response to ER stress and activates Activating Transcription Factor 4 (ATF4) to increase chaperone production and C/EBP homologous protein (CHOP) pathways115. PERK also phosphorylates eukaryotic translation initiating factor 2α (eIF2α), which results in an overall reduction of protein synthesis by reducing mRNA incorporation into ribosomes115. ER stress causes dissociation of ATF6 from Binding Immunoglobulin Protein (BiP, GRP78, HSPA5). In turn, ATF6 translocates to the Golgi apparatus where it is cleaved by Site-1 protease (S1P) and Site-2 protease (S2P) to form the nuclear factor ATF6116. Nuclear factor ATF6 amplifies the IRE1α and PERK responses to ER stress by upregulating XBP1 mRNA expression, chaperone expression, and CHOP expression. Overall, the UPR upregulates chaperone protein synthesis, decreases global protein production, increases protein degradation of misfolded or unfolded proteins, and, if proteostasis is not restored after these measures, the 15 UPR induces apoptosis via the PERK-eIF2a-ATF4-CHOP pathway1,113. The UPR also alters cellular secretion pathways. The endomembrane system connects the ER, Golgi, plasma membrane, and lysosomes through membrane bound vesicles. XBP1-s activation enlarges the size of the ER and ER-associated vesicles to reduce the concentration of misfolded proteins117,118. The UPR promotes protein degradation through ERAD and autophagy, and both degrade proteins that could otherwise be secreted38,39. ERAD occurs through enhancement of proteasomal degradation in three steps: recognition, extraction, and degradation39. Misfolded proteins are recognized by chaperone proteins such as heat shock proteins ER DnaJ 3 (ERdj3), ERdj4, and ERdj5119. Retrotranslocation proteins recognize and bind misfolded proteins in the ER lumen and undergo a conformational change to move the misfolded protein to the cytoplasm, where they are ubiquitinated by UPR regulated E3 enzymes including Suppressor Enhancer Lin-12-Like (SEL1L), Ubiquitin Protein Ligase E3 Component N-Recognin 4 (UBR4), and Ubiquitin Protein Ligase E3 Component N-Recognin 5 (UBR5)122,123. Ubiquitinated proteins are trafficked to the 26S proteasome and degraded38. Because ERAD only degrades soluble proteins38,122, cells need additional pathways for degradation of insoluble proteins and other cellular components. Autophagy or “self-eating” refers to three different processes: microautophagy, chaperone-mediated autophagy (CMA), or macroautophagy39. Microautophagy is the homeostatic lysosomal degradation of material and uses direct transport of cargo into lysosomes123. In CMA, heat shock protein (HSP)A8 binds misfolded proteins for retrotranslocation to the cytoplasm and subsequent insertion of the misfolded protein into the lysosome for degradation41. Most commonly, as we do here, autophagy refers to macroautophagy. Macroautophagy breaks down insoluble material or large amounts of material to maintain cellular function; macroautophagy is active in degrading protein aggregates, recycling organelles, and generating substrates and intermediates in response to nutrient deprivation124. In macroautophagy, Unc-51-like Kinase 1 (ULK1) initiates a membrane formation event that creates a phagophore to encapsulate material to be degraded125. This vesicle matures into a double membrane vesicle called an autophagosome. Autophagosomes merge with lysosomes to initiate degradation of their contents39. Thus, ERAD and autophagy, which are both activated by ER stress, result in altered trafficking of expressed proteins (i.e., to degradation rather than secretion). 16 Here, we modeled ER stress on protein production in cell culture, with the goal of understanding how ER stress alters protein trafficking and secretion. Our results demonstrated that the UPR in response to ER stress (specifically IRE1α activation) improved protein secretion. Importantly, we show that, in cells under ER stress, inhibiting autophagy and enhancing proteasomal degradation improved secretion of proteins, presenting a potential opportunity for improving therapeutic protein secretion in industrial bioprocesses. 2.2 Results 2.2.1 ER stress reduces secretion relative to intracellular accumulation To investigate the impact of ER stress on the secretion of expressed proteins, we tested the ability of HeLa and MDA-MB-231 cells to synthesize and secrete Gaussia princeps luciferase (Gluc) from transfected mRNA in the presence of increasing concentrations of the N- glycosylation inhibitor tunicamycin (TM) (Figure 2.1). Inhibition of N-glycosylation by TM reduces conformational stability and increases aggregation, leading to accumulation of misfolded proteins and ER stress126. Increasing concentrations of TM reduced cell numbers for both HeLa and MDA-MB-231 cells (Figure 2.1A), likely due to both reduced growth rate and cytotoxicity. Increasing TM concentration also reduced the amount of Gluc secreted into the culture media (Figure 2.1B). Plotting cell number vs. Gluc secreted showed a significant correlation (Figure 2.1C). In contrast, increasing levels of TM resulted in no change (HeLa) but increased (MDA- MB-231) Gluc in cell lysates (Figure 2.1D). Lysate Gluc signal and cell number were not correlated (Figure 2.1E). To capture both these results simultaneously, we established a metric for the efficiency by which cells secrete expressed protein, the secretion ratio, which is the quantity of Gluc in the media divided by the quantity in the cell lysate. Secretion ratio decreased with increasing TM for both cell lines (Figure 2.1F). The changes in secretion ratio were associated with activation of autophagy (see Figure A1A-F for representative micrographs of Gluc retention and the autophagosome marker Microtubule-associated proteins 1A/1B light chain 3B (LC3B)). Taken together these results suggest that ER stress i) reduced overall Gluc production in both cell lines by limiting cell growth, and ii) increased the fraction of the expressed, functional Gluc that was sequestered within the cells rather than secreted. 17 Figure 2.1 The impact of tunicamycin treatment on protein secretion. A) Normalized cell number vs. TM concentration for HeLa (solid) and MDA-MB-231 (striped) cells. B) Normalized 24-hour Gluc accumulation in the media vs TM concentration. C) Normalized cell number vs. normalized 24-hour media Gluc accumulation. D) Normalized Gluc in the cell lysate 24-hours post mRNA transfection vs. TM concentration. E) Normalized cell number vs. normalized Gluc in cell lysate. F) Normalized secretion ratio (24-hour media accumulation of Gluc divided by normalized Gluc in cell lysate) vs. TM concentration. Error bars represent standard deviations. Values indicate Student’s t-test pair-wise p-values compared to 0 µg/mL TM, N = 3. Bold indicates p < 0.05. 18 2.2.2 IRE1α mitigates the impact of ER stress on secretion IRE1α splicing of XBP1 mRNA and subsequent transcriptional changes are known to be important cellular responses to maintain secretory capacity127. To investigate the role of IRE1α activity on secretion under ER stress, we used MDA-MB-231 wild type (WT) and IRE1α-/- (KO) cells. TM treatment decreased the cell number and media accumulation of Gluc in both the WT and KO cells (Figures 2.2A&B). The amount of Gluc that accumulated in WT TM-treated cells was similar to the amount in WT cells (Figure 2.2C, left two columns). In contrast, the amount of Gluc that accumulated in the TM-treated KO cells was greater than the amount that accumulated in untreated KO cells (Figure 2.2C, right two columns). Thus, knocking out IRE1α further reduced the secretion ratio in response to ER stress induced by TM (Figure 2.2D, see Figure A2 for representative micrographs depicting increased accumulation and autophagy maker LC3B in KO cells). Hence, IRE1α function is important for supporting secretion in cells experiencing TM-mediated ER stress. 19 Figure 2.2 IRE1α knockout reduces secretion in response to TM. A) Normalized cell numbers for MDA-MB-231 cells that are wild type (WT), wild type with 5 µg/mL TM (WTTM), IRE1α-/- (KO), and IRE1α-/- with 5 µg/mL TM (KOTM) 24 hours after transfection of Gluc mRNA. B) Normalized 24-hour media Gluc accumulation for WT, WTTM, KO, and KOTM. C) Normalized Gluc in cell lysate for WT, WTTM, KO, and KOTM. D) Secretion ratio for WT, WTTM, KO, and KOTM. Error bars represent standard deviations. The table represents Student’s t-test pair-wise p-values, N = 3. Bold indicates p < 0.05. 2.2.3 Autophagy inhibition improves secretion ratio Under ER stress, in addition to IRE1α activation of XBP1-s transcription, in some cases autophagy is activated as a pro-survival mechanism128. Having determined that was the case here (Figures A1A-F) we investigated how autophagy activation affected Gluc secretion. We used the class III phosphoinositide 3-kinase (PI3K) inhibitor, 3-methyladenine (3MA), to prevent the formation of phagophores and inhibit autophagy. The treatments of TM, 3MA, and both 20 (3MATM) did not significantly alter the number of HeLa cells. However, the amount of Gluc secreted into the media was strongly correlated with cell number (Figure 2.3A), with 3MA increasing cell number and protein secreted both in the absence and presence of TM. The amount of Gluc in the cell lysate was not affected by exposure to TM or cotreatment of 3MATM. A plot of cell number vs lysate showed weak correlation, with the slope not statistically different from zero (Figure 2.3B). 3MA treatment alone caused no change in the secretion ratio of Gluc. In contrast, autophagy inhibition in combination with TM exposure increased secretion over TM alone, albeit lower than control (Figure 2.3C). Therefore, autophagy inhibition increased the likelihood of a synthesized Gluc protein being secreted rather than retained in the cell during ER stress. We confirmed the inhibition of autophagy and the corresponding changes in secretion using confocal microscopy (Figure A1A-L). 21 Figure 2.3 Effects of autophagy inhibition and TM-induced ER stress on secretion ratio. HeLa cells transfected with mRNA coding for Gluc and allowed to recover in control- (filled), 3MA- (squares), TM- (diamonds), or 3MA with TM-media (unfilled). A) Normalized 24-hour media accumulation of Gluc vs. normalized cell number. B) Normalized Gluc in cell lysate vs. normalized cell number. C) Secretion ratio of control, 3MA-treated, TM-treated, and 3MATM- treated cells. The table is pairwise Student’s t-tests. Slope p-values calculated by linear regression and compared to a zero slope. Error bars represent standard deviation, N = 3. Bold indicates p < 0.05. 22 2.2.4 Proteasome activation restores secretion ratio Autophagy and proteasomal degradation cooperate to address dysregulated proteostasis. To investigate how proteasomal degradation affected secretion of Gluc in our system, we used the proteasome inhibitor VR23 (VR) and proteasome activator (through PDE4 inhibition) Rolipram (RP). VR reduced the cell number and total Gluc secreted into the media for both TM- stressed and unstressed cells; cell number and Gluc secreted were correlated (Figure 2.4A). The amount of Gluc in the lysate increased with VR treatment for both the TM-stressed and unstressed cells (Figure 2.4B). Thus, VR decreased the secretion ratio, and co-treatment with TM compounded this effect (Figure 2.4C). Counterintuitively, proteasome inhibition, which should increase the amount of protein available for secretion, increased intracellular accumulation of Gluc and reduced secretion. To support this result, we examined whether proteasome activation would concomitantly increase secretion. Indeed, RP activation of proteasomes increased secretion in both the stressed and unstressed cells (Figure 2.5A), without altering the amount of Gluc in the lysates, regardless of TM treatment (Figure 2.5B). Thus, RP treatment increased the secretion ratio in both the TM- treated and unstressed cells (Figure 2.5C). Proteasomal activation was accompanied by a downregulation in autophagy (Figure A1A-F and A1M-X), suggesting that sequestration of secreted proteins in cells occurs through autophagy. 23 Figure 2.4 Proteasome inhibition reduced secretion. HeLa cells transfected with mRNA coding for Gluc and allowed to recover for 24 hours in control- (filled), VR23- (squares), TM- (diamonds), or VR23 with TM-media (unfilled). A) Normalized media accumulation of Gluc vs. normalized cell number. B) Normalized Gluc in cell lysate vs. normalized cell number. C) Secretion ratio of control cells and cells treated with VR23 alone, TM alone, and VR23 with TM. Slope p-values calculated by linear regression and compared to a zero slope. The table is pairwise Student’s t-tests. Error bars represent standard deviation, N = 3. Bold indicates p < 0.05. 24 Figure 2.5 Proteasome activation improved secretion. HeLa cells transfected with mRNA coding for Gluc and allowed to recover for 24 hours in control- (filled), RP- (squares), TM- (diamonds), or RP with TM-media (unfilled). A) Normalized media accumulation of Gluc vs. normalized cell number. B) Normalized Gluc in cell lysate vs. normalized cell number. C) Secretion ratio of control cells and cells treated with RP alone, TM alone, and RP with TM. Slope p-values calculated by linear regression and compared to a zero slope. The table is pairwise Student’s t-tests. Error bars represent standard deviation, N = 3. Bold indicates p < 0.05. 25 2.2.5 Secretion ratio changes with RP treatment are independent of transfected nucleic acid ER stress alters several translation pathways that could alter the expression of secreted proteins108. To determine if the changes in secretion ratio were due to transcriptional regulation, we transfected cells with mRNA and pDNA complexed with Lipofectamine 2000 (LF2K) (Figure 2.6A). Post-transfection treatment with TM, RP, RPTM, VR, and VRTM all showed the same trend on secretion ratio comparing mRNA- and pDNA-based protein secretion. 3MA and 3MATM treatment reduced pDNA-based protein secretion as compared to control and TM treatment respectively, whereas 3MATM improved mRNA-based protein secretion as compared to TM treatment (Figure 2.3). Overall, the transfections of mRNA hindered cell growth less than pDNA transfection (Figure 2.6B) with 3MA causing similar toxicity as TM in pDNA transfected cells, which could suggest pDNA transfection caused additional stress/cytotoxicity through increased LF2K concentration that autophagy helps to alleviate. These results confirmed that the effects of proteasomal activation/inhibition on secretion are due to alterations in translation and protein processing, not transcription. Figure 2.6 Changes to proteasomal degradation were independent of transfected nucleic acid type. A) Secretion ratio responses to small molecule alteration of degradation pathways with either mRNA (black) based protein expression or pDNA (gray) based protein expression. B) Relative cell number for each small molecule alteration of degradation and mRNA (black) or pDNA (gray) transfection. Error bars represent standard deviations, N=3. 2.2.6 ER stress dependence on autophagy and proteasomal activity is cell type dependent We used MDA-MB-231 and HeLa cells to determine if the influence of 3MA and RP on the secretion patterns under ER stress is cell type dependent. For MDA-MB-231 breast cancer cells, TM reduced secretion ratio (Figures 2.7A-C). Reducing degradation via 3MA and VR23 also reduced secretion ratio (Figures 2.7A and B), while increasing proteasomal degradation with RP did not statistically impact secretion (Figure 2.7C). The combination of TM and degradation 26 inhibition further reduced secretion, relative to TM alone (Figures 2.7A and B). TM and RP treatment in MDA-MB-231 was not statistically different from TM alone (Figure 2.7C). The normalized number of cells remaining after 24 hours of protein production was lower for the MDA-MB-231 cells than HeLa cells, suggesting that apoptosis was more likely to be induced in MDA-MB-231 cells vs. HeLa cells under these conditions (Figure 2.7D). These results indicate that altering degradation processes during ER stress to increase protein production will be cell line dependent. Nonetheless, increasing proteasomal degradation either improves secretion (HeLa) or has no negative impact on secretion (MDA-MB-231), making it a consideration for protein production processes. Figure 2.7 Restoration of secretion patterns altered by ER stress is cell line dependent. HeLa (solid) and MDA-MB-231 (striped) cells transfected with mRNA coding for Gluc and allowed to recover for 24 hours in media containing A) no small molecules, TM, 3MA, or both; B) no small molecules, TM, VR23, or both; and C) no small molecules, TM, RP, or both. D) Normalized cell number for each treatment (untreated, 3MA, RP, and VR23) and the cotreatments with TM. 2.3 Discussion In this work, we demonstrated that ER stress reduced the amount of Gluc secreted into the media not only by reducing cell number but also by reducing the fraction of synthesized proteins that are secreted rather than retained in the cells. Autophagy inhibition improved 27 secretion, while, counterintuitively, increasing proteasomal degradation increased the likelihood of secretion. Thus, the capacity of cells to secrete synthesized proteins is related to the activity of its protein degradation mechanisms. We focused this work on understanding how ER stress and the UPR influence protein secretion because, as stated above, cells used in industrial bioprocesses continually experience ER stress due to a variety of causes, which, in turn, reduces yield of expressed proteins. Our model for ER stress induction, administration of TM, is well-established for studying ER stress and the UPR and causes similar changes to those experienced by cells in bioprocesses, despite the differences in the laboratory and industrial contexts (e.g., chemical vs. physical stress)129. For instance, nutrient and oxygen deprivation, heat stress, and shear stress disrupt N-glycosylation similar to TM129. Reduced protein yields can result from many bottlenecks. It is well-established that ER stress reduces cell growth and protein production130, consistent with our results here. But, the effects of ER stress on intracellular protein processing have not previously been explored112. Disruptions in glycosylation impair the function of many transmembrane glycoproteins that control vesicle trafficking131, making it reasonable to presume that ER stress would alter protein secretion. Our results here provide clear evidence that controlling intracellular processing of expressed proteins can directly improve the secretion of active protein. ER stress upregulates the autophagic degradation pathways and ERAD as a pro-survival mechanism38,39. These two forms of degradation play a cooperative role in decreasing the misfolded protein load that accumulates and leads to ER stress. Previous work indicated that induction of autophagy caused an increase in protein secreted into the media132. However, they did not determine the per cell production or secretion ratio, only that the induction of autophagy is pro-survival and increases the number of cells producing protein. Our results showed that TM- induced autophagy hindered protein production, when normalized for cell number, and that inhibiting TM-induced autophagy improved protein secretion per cell. This suggests that there is an optimal level of autophagy that results in the highest protein production per cell with the optimum survivability of cells under ER stress. Since autophagy inhibition improves secretion and proteasomal degradation could decrease autophagy, we investigated RP to increase proteasomal degradation and ERAD. RP is a cAMP-specific phosphodiesterase 4 (PDE4) inhibitor that is used clinically to treat neurological 28 disorders133. Increased cAMP levels increase proteasome phosphorylation and activity. RP treatment increases cell number and protein secretion without increasing protein accumulation in cells. The increase in protein secretion of cells treated with RP could be due to improved protein trafficking, since reduced ER stress would result from increased degradation of unfolded proteins. Additionally, translation could be generally accelerated due to the resulting increased concentration of free amino acids in the cytosol that result from degraded proteins. Elevated cAMP levels could also activate other pathways, including the AMP-activated Protein Kinase pathway134, that promote protein production and cell viability. We demonstrated that inhibition of proteasome activity reduced Gluc secretion per cell and increased intracellular Gluc accumulation and autophagy, thereby highlighting the importance of both autophagy and proteasomal degradation in maintaining cellular protein secretion capacity. We expect that our results will inform the future development of therapeutic protein bioprocesses, by encouraging investigation of methods to increase misfolded protein degradation without increasing autophagy. 29 CHAPTER 3: ER STRESS INDUCED AUTOPHAGY ALTERS CELLULAR PROCESSING OF CATIONIC LIPID DELIVERED SIRNAS 3.1 Introduction Small interfering RNAs (siRNAs) are an increasingly important class of therapeutics, with seven FDA approved therapeutics and 156 active or recruiting clinical trials43,74-80. siRNAs are large (compared to small molecule drugs), charged molecules and must be modified or complexed with a delivery vehicle to enter the cytoplasm of a target cell and achieve therapeutic effect83. Disruptions to endocytosis and intracellular processing from common comorbidities such as obesity, diabetes, smoking, and hypertension could therefore affect siRNA activity107, principally through disruption of the function of the endoplasmic reticulum (ER). The ER is a primary controller of membrane turnover and endocytosis, which are required for uptake and trafficking of siRNAs. In response to these comorbidities, cells respond by upregulating synthesis of defensive proteins, including chaperones, cytokines, and metabolic enzymes108. In many cases, the increased protein synthesis burden overwhelms the quality control systems of the ER, resulting in an accumulation of unfolded or misfolded proteins, a condition broadly labelled as ER stress17,109. In response to ER stress, the unfolded protein response (UPR), is initiated by three sensor proteins IRE1α, PERK, and ATF6. The UPR downregulates protein expression, increases chaperone protein expression, and increases protein degradation. If these changes fail to restore proteostasis, apoptosis is initiated108. In the presence of unfolded proteins, IRE1α dimerizes and trans-autophosphorylates, activating an RNase domain that splices XBP1 mRNA leading to expression of an active transcription factor (XBP1s)135. XBP1s increases expression of JNK pathway, apoptosis, and ER associated degradation (ERAD) proteins136. PERK is activated by the presence of misfolded proteins. PERK subsequently phosphorylates eIF2α, which inhibits mRNA translation, and activates ATF4, which increases expression of ATF target genes, including autophagy regulating genes137. ATF6 translocates to the Golgi, where proteases S1P and S2P cleave ATF6 into the active transcription factor ATF6f, which upregulates XBP1 mRNA, Death Associated Protein Kinase 1 (DAPK1), and Interferon-gamma (IFN-γ) to further promote autophagy138. Taken together, these pathways increase degradation of unfolded proteins and reduce the protein production burden on the ER. To achieve silencing, siRNAs must enter the cytoplasm by crossing the endosome membrane after endocytosis4,81. The exact mechanism of endosomal escape of cationic lipid- 30 delivered siRNAs is unclear, but there is evidence supporting the role of pore formation and membrane fusion in endosomal escape, as well as direct siRNA transport into the cytoplasm81,102,103. Regardless, delivery to the cytoplasm involves electrostatic interactions between the siRNA-containing complexes and a membrane81. The lipid content of membranous vesicles is determined by lipid synthesis that occurs in the ER and changes due to ER stress139. Sterol Regulatory Element Binding Protein 1 (SREBP1) and SREBP2 translocate along with ATF6 to the Golgi, where they are cleaved by S1P and S2P to become activate transcription factors that increase cholesterol synthesis, decreasing the polarity of cell membranes, thereby increasing electrostatic repulsion with charged molecules in siRNA-containing complexes140. IRE1α increases Diacylglycerol O-acyltransferase 2 (DGAT2), Stearoyl-CoA Desaturase (SCD1), and Acetyl-CoA Carboxylase (ACC2) expression, which are involved in the synthesis of triglycerides, monounsaturated fatty acids, and saturated fatty acids, respectively, all of which decrease membrane fluidity141,142. PERK activation of ATF4 leads to increased expression of SREBP1 and SCD1143. Overall, ER stress increases membrane rigidity and decreases membrane fluidity. In addition to altering membrane content, ER stress can alter the trafficking of endomembrane vesicles by increasing autophagy. Macroautophagy (autophagy) is a protective mechanism initiated by all three branches of the UPR to degrade unfolded proteins, increase energy production, and sequester stress causing material144. There are numerous reported mechanisms for autophagy initiation145. After an initiation event, membrane lipids and proteins from the ER are recruited to form a phagophore that elongates to surround the material to be degraded. When the phagophore has fully engulfed the material, it becomes a bilayer autophagosome, which eventually merges with a lysosome to degrade the cargo39. Autophagy impacts endocytosis through a shared membrane resource pool146. Phagophore elongation requires lipid material from nearby membranes including endosomes and the ER146. Autophagosomes can also attract endosome trafficking Rab protein-coated vesicles and COPII- coated vesicles to merge with autophagosomes and/or insert autophagy related proteins (ATGs) to reprogram the vesicle into an autophagosome124. Generally, autophagy removes lipids, proteins, and vesicles from the rest of the endomembrane system, potentially altering siRNA endocytosis, trafficking, and function. In this study, we investigated the impact of tunicamycin (TM)-induced ER stress on the 31 accumulation and effectiveness of the siRNAs delivered to HeLa cells. ER stress increased siRNA accumulation but at short times reduced silencing. We identified autophagy as a major contributor to the increase in siRNA accumulation. Finally, we demonstrated that the increased siRNA accumulation in ER –stressed cells is still functional and extends the duration of silencing. 3.2 Results 3.2.1 ER stress increased siRNA accumulation and decreased silencing To determine the impact of ER stress on siRNA function, the N-glycosylation inhibitor tunicamycin (TM) was used to induce ER stress in HeLa cells that constitutively express EGFP. Transfection of cy5-tagged siRNA using LF2K in cells treated with increasing levels of TM resulted in increasing accumulation of siRNA 24 hours post transfection (Figure 3.1A). Counterintuitively, the increase in accumulated siRNA corresponded to a decrease in the silencing achieved by the siRNAs (Figure 3.1B). This indicated that siRNAs entered the cell and were sequestered in such a way to prevent initiation of RNAi. Figure 3.1 The impact of TM on siRNA accumulation and EGFP silencing. A) siRNA accumulated vs. TM concentration. B) EGFP silencing vs. TM concentration. Labels for each data point are the p-values of Student’s T-tests comparing the data point to the 0 µg/mL TM data point; bold indicates p < 0.05. Error bars represent standard deviation, N = 3. 3.2.2 Autophagy inhibition reduced ER stress-induced siRNA accumulation We hypothesized that the sequestration of siRNAs could be due to autophagy resulting from ER stress. Autophagy was inhibited using the class III phosphoinositide 3-kinase (PI3K) inhibitor 3-methyladenine (3MA). The presence of 3MA in the media during transfection 32 reduced the amount of siRNA that accumulated in cells during TM exposure (Figure 3.2A). Autophagy inhibition reduced silencing in both TM stressed and control cells, with the combination of TM and 3MA treatment reducing silencing most (Figure 3.2B). Calculating the specific silencing or ratio of silencing achieved to the amount of siRNA accumulated in cells provides a measure of the efficiency of cytoplasmic entry (i.e., siRNAs that cannot escape the vesicles are prevented from initiating silencing). TM treatment altered the silencing/siRNA ratio, while treatment with 3MA did not result in a significant change (Figure 3.2C). This indicated that while autophagy inhibition reduces the accumulation of siRNA, ER stress and TM treatment alters the intracellular processing of siRNA complexes. 33 Figure 3.2 Effect of 3MA and TM on siRNA accumulation, silencing, and silencing per siRNA. A) siRNA accumulation in control, 3MA treated, TM treated, and co-treated with TM and 3MA cells. B) Fractional silencing in control, 3MA treated, TM treated, and co-treated with TM and 3MA cells. C) specific silencing or silencing divided by siRNA accumulated in control, 3MA treated, TM treated, and co-treated with TM and 3MA cells. p-values are the result of Student’s T-tests comparing indicated values. Error bars represent standard deviation, N = 3. 34 3.2.3 Endosome maturation inhibition alters siRNA delivery Bafilomycin A1 (BAFA1) is a V-ATPase inhibitor that prevents the acidification of endosomes and prevents the maturation of early endosomes to late endosomes and lysosomes. High concentrations of BAFA1 induce ER stress and disrupt glycosylation by altering Golgi pH32. Treatment with 50 nM BAFA1 during transfection led to an increasing accumulation of siRNA (Figure 3.3A). Silencing and cell number did not change significantly (Figure 3.3B and C). Specific silencing decreased with BAFA1 exposure. This indicates that siRNAs are being retained before the late endosome, not in the lysosome, and that other ER stress inducers that indirectly impact glycosylation have similar effects to those seen in TM exposure. Figure 3.3 Bafilomycin A1 increased siRNA accumulation. A) Normalized siRNA accumulation vs. BAFA1 concentration. B) Normalized silencing vs. BAFA1 concentration. C) Normalized cell counts vs. BAFA1 concentration. Values are pair-wise p-values calculated from Student’s T-tests compared to 0 nM BAFA1. Unlisted p-values are greater than 0.1 and bold represents p < 0.05. Error bars represent standard deviation, N = 3. 35 3.2.4 ER stress prolongs siRNA accumulation and silencing effect Recognizing that siRNAs were being sequestered in ER-stressed cells, we investigated whether the siRNAs were degraded or remained intact and able to initiate silencing upon reaching the cytoplasm. Cells were transfected in media with TM and then allowed to recover in TM culture media or TM-free culture media. 24 hours post transfection, cells that received TM at any point had a higher accumulation of siRNA than control cells (Figure 3.4A). At 72 hours, siRNA fluorescence in control cells approached background, while the siRNA fluorescence remained elevated in cells treated with TM during this recovery period. At 120 hours, cells that recovered in TM retained more siRNA, while the siRNA fluorescence was reduced in cells that received TM only during transfection which matched the reduction in retention of control cells (Figure 3.4A). Unstressed cells silenced the most at 24 hours followed by cells that were exposed to TM only after transfection. Cells transfected in TM had the lowest silencing at 24 hours (Figure 3.4B). At 72 hours, cells exposed to TM at any point had increased silencing which matched that seen in the control cells at 24 h, while silencing in the control cells decreased. At 120 hours, siRNA silencing was absent in the control cells, but the TM-treated cells retained some degree of silencing while cells that recovered in TM achieved about half of the maximum measured silencing (Figure 3.4B). Taken together, these results indicate that sequestered siRNAs in the TM-treated cells were still functional and that continued ER stress prolonged the sequestration of the siRNA-containing complexes without reducing their efficacy. Figure 3.4 Duration of siRNA accumulation and EGFP silencing to unstressed (circles), TM during transfection (squares), TM during recovery (triangle), and TM during both transfection and recovery (diamonds). A) siRNA accumulation vs. time. B) EGFP fractional silence vs. time. Error bars are standard deviation, N = 3. 36 3.3 Discussion We showed that ER stress can alter the delivery and processing of siRNA-containing complexes in TM-treated HeLa cells. TM-induced ER stress increased the accumulation of siRNA while decreasing silencing. Autophagy inhibition decreased the retention of siRNA. By improving our understanding of how diseases alter siRNA processing, we believe our work will inform the design of future siRNA therapeutics and tailoring their dosing (quantity and frequency) for specific patients. Induction of the UPR is both stimulus- and cell line-dependent; however, there are many available ER stress models for UPR activation in cell culture108. TM initiates ER stress by inhibiting N-glycosylation. which is altered or impaired in many diseases, including rheumatoid diseases, metastatic cancers, and neurological conditions that could benefit from siRNA therapy135. Our results showed that TM-induced ER stress resulted in the accumulation of siRNA with a corresponding reduction in silencing. While this result is counterintuitive, it adds to growing evidence that siRNA accumulation in a cell does not necessarily result in increased silencing13,102. ER stress (and specifically TM-induced ER stress) is linked to increases in autophagy136. Our results indicate that autophagy mediates siRNA sequestration but does not damage the siRNA or prevent its function upon release to the cytoplasm. Rapamycin, a mammalian Target of Rapamycin Complex 1 (mTORC1) inhibitor, induced autophagy and improved silencing in H1299 cells using LF2K while 3MA, which inhibits autophagy, reduced silencing137. Our results support that 3MA reduces silencing but furthermore showed that there is a reduction in the accumulation of siRNA in the autophagy-inhibited cells. TM has been shown to activate and inhibit mTORC1 pathway depending on the condition147. Because the silencing per siRNA is consistent in the unstressed and stressed cells, it is possible that other UPR factors or TM- specific factors are responsible for the discrepancy in silencing efficiency reported in this study (Figure 3.2C) and that reported by Song et al137. Alternatively, it could be attributed to a difference in the cell type or the timing of the silencing measurements as Song et al use 48 hours for recovery compared to our 24 hours. BAFA1 is a V-ATPase inhibitor that inhibits the acidification of membrane bound vesicles and as such is an autophagy inhibitor and endosome maturation inhibitor. Low 37 concentrations of BAFA1 prevent acidification without causing ER stress, while high concentrations of BAFA1 lead to GRP-78, CHOP activation, and mTOR activation148. The fact that high BAFA1 concentrations impact siRNA delivery similarly to TM in our cells suggests that TM is causing a disruption in maturation from the early endosomes to late endosomes/lysosomes and promoting trafficking to a more autophagosome-like vesicle. siRNA release to the cytoplasm occurs during early to late endosome transition, and the results in Figures 3.1, 3.2, and 3.3 suggest that ER stress blocks this transition in favor of autophagy generation. The role TM plays on the cellular pharmacodynamics helps situate our data better in the literature. When analyzing the 24 hour data it appears that ER stress and TM are counterproductive to silencing - higher siRNA accumulation, lower silencing. In contrast, at the 72 hour and 120 hour measurements, autophagy and ER stress improve siRNA accumulation and silencing. These results should inform delivery vehicle designers that the accumulation of siRNA at a target tissue dynamic process and that it does not correlate with silencing efficacy. Another key takeaway is that the ER stress state can influence the processing characteristics of siRNA and that comorbidities that increase ER stress especially those that activate mTOR pathway will delay silencing while siRNA accumulates in the tissues. This could result in increased toxicities if not properly accounted for due to the increased retention of exogenous material. In conclusion, the results of our study show that in the short term, TM increases siRNA accumulation and decreases silencing possibly in an mTOR dependent manner. Autophagy inhibition can restore normal accumulation patterns, but ER stress still reduces the specific silencing of the siRNA. Finally, our results suggest that cells experiencing ER stress retained more siRNA which eventually is released to silenced its target gene. 38 CHAPTER 4: CONCLUSIONS AND FUTURE DIRECTIONS 4.1 Conclusions The objective of this work was to gain insight into how perturbations to the endomembrane system alter therapeutic protein production and effective delivery of siRNA therapeutics. The use of secreted luciferase and TM as a model system for ER stress in cell culture (as proxy a for bioreactors) demonstrated that ER stressed cells were less effective at producing secreted proteins and more prone to higher intracellular accumulation. The activation of proteasomal degradation over autophagy is important in mediating the response to ER stress. Inhibiting autophagy reduced the amount of proteins that accumulated in ER stressed cells. Interestingly, activating proteasomal degradation increased secretion and reduced intracellular accumulation in ER stressed cells. These results showed that proteasome activity can be enhanced with small molecules and should be monitored during cell line development to increase the efficiency of therapeutic protein production. Despite the recent advances and clinical successes of siRNA therapeutics, challenges remain in their widespread adoption by the population at large. One of the challenges is the ability to predict pharmacodynamic properties in diverse populations. Our use of TM, fluorescently labeled siRNAs, and HeLa cells expressing EGFP showed that siRNAs accumulate at higher levels in cells experiencing ER stress. Inhibiting autophagy reduced the accumulation of siRNA in cells while inhibiting endosome maturation determined siRNAs accumulated in early endosomes trafficking to autophagosomes. This increased accumulation did not lead to an increase in early silencing. The accumulated siRNAs remain functional and increased the duration of silencing. These results inform future siRNA therapeutic design that in addition to tissue characteristics, ER stress levels should be considered for delivery system selection and dosing parameters. 4.2 Future Directions 4.2.1 Activation of proteasomal degradation in batch bioreactors While these results will inform design of more efficient protein production processes, further research is needed to assess the commercial viability of proteasomal activation. Using a model of ER stress, TM, and HeLa cells do not correlate exactly with the stresses experienced in bioreactors. As such, the next steps is to replicate the methodology with suspension cultures of cells used in protein production (CHO, HEK293, NS0). While Rolipram is a relatively cheap 39 compound and is an FDA approved antipsychotic drug, it is potentially more cost and developmentally effective to increase proteasomal degradation by developing cell lines that express higher levels of proteasomes through stable transfection and selection. In addition to testing other culture systems, these results may help with downstream purification. Purification of therapeutic proteins from host cells face challenges due to secretion of aggregates, and fractional proteins that have similar binding affinity in PAC. This study did not measure the quality of the secreted and intracellularly accumulated proteins, however, the luciferase protein was functional. ER stress associated proteins, mostly chaperone proteins, are one of the main contaminants that are bound to product proteins during purification. It seems likely that increasing the rate of degrading unfolded and misfolded proteins will increase chaperone proteins transport into ERAD and reduce the number of chaperone proteins bound to product proteins. Proteasome activation should also increase degradation of the aggregates and fractional proteins. In future experiments, measuring protein size disparity with HPLC should be included to assess the impact of proteasome activation on aggregates and fragments. 4.2.2 Impact of additional ER stressors on siRNA delivery TM is one of the most impactful ER stress inducing agents available. Nonetheless, further studies should be conducted with clinically relevant models of ER stress such as, lipopolysaccharides (LPS), free fatty acids, fevers, and cytokine responses. Initial results toward this goal have been obtained (Appendix 3) and it appears that LPS changes the way siRNA enters a cell but the impact on silencing effect and siRNA accumulation is unclear. 4.2.3 Impact of in vitro ER stress on siRNA delivery and toxicity Failures with siRNAs therapeutics in clinical trials are largely due to toxicity. They include over accumulation of siRNA in target cells leading to RNAi machinery saturation and delivery vehicle toxicities. If my hypothesis that ER stress increases siRNA accumulation and delays and extends the silencing effect, the next step would be to test in mice with different degrees of ER stress (e.g., high fat diet) and measuring the biodistribution of siRNA. Over-, under-, and optimal dosing could be impacted by ER stress. Overdosing occurs when a therapeutic is given too often and the therapeutic window is exceeded leading to toxic effects. Underdosing refers to long periods between doses where the bodily concentration of the drug goes below the therapeutic window whereby the drug is not effective. Optimal dosing occurs when the correct quantity and dosing period produces maximum therapeutic 40 concentration at time of dosing which decreases to the minimum therapeutic window prior to the next dosing. siRNA therapeutics that are re-administered too frequently under ER stress conditions leading to delayed siRNA degradation, could result in overdosing and toxic effects. This could be contributing to the toxicity observed in clinical is RNA trials. Comparing the dosing regimen and pharmacokinetics of control mice with mice experiencing systemic ER stress should be investigated. 41 BIBLIOGRAPHY Splichal RC, Chen K, Walton SP, Chan C. The role of endoplasmic reticulum stress on 1 reducing recombinant protein production in mammalian cells. Biochem Eng J 2024;210:109434. https://doi.org/10.1016/j.bej.2024.109434. Mordor Intelligence. Biopharmaceutical Industry Size and Share Analysis - Growth 2 trends and forecasts (2024 - 2029). 2024. Mysior MM, Simpson JC. Cell3: a new vision for study of the endomembrane system in 3 mammalian cells. Biosci Rep 2021;41:1–11. https://doi.org/10.1042/BSR20210850C. Smith SA, Selby LI, Johnston APR, Such GK. The Endosomal Escape of Nanoparticles: 4 Toward More Efficient Cellular Delivery. Bioconjug Chem 2019;30:263–72. https://doi.org/10.1021/acs.bioconjchem.8b00732. Long M, Kranjc T, Mysior MM, Simpson JC. RNA Interference Screening Identifies 5 Novel Roles for RhoBTB1 and RhoBTB3 in Membrane Trafficking Events in Mammalian Cells. Cells 2020;9:. https://doi.org/10.3390/cells9051089. Rabouille C. Pathways of Unconventional Protein Secretion. Trends Cell Biol 6 2017;27:230–40. https://doi.org/10.1016/j.tcb.2016.11.007. Nickel W. Unconventional secretory routes: Direct protein export across the plasma 7 membrane of mammalian cells. Traffic 2005;6:607–14. https://doi.org/10.1111/j.1600- 0854.2005.00302.x. Kienzle C, von Blume J. Secretory cargo sorting at the trans-Golgi network. Trends Cell 8 Biol 2014;24:584–93. https://doi.org/10.1016/j.tcb.2014.04.007. Yu J, Chia J, Canning CA, Jones CM, Bard FA, Virshup DM. WLS Retrograde transport 9 to the endoplasmic reticulum during Wnt secretion. Dev Cell 2014;29:277–91. https://doi.org/10.1016/j.devcel.2014.03.016. Agola JO, Jim PA, Ward HH, Basuray S, Wandinger-Ness A. Rab GTPases as regulators 10 of endocytosis, targets of disease and therapeutic opportunities. Clin Genet 2011;80:305–18. https://doi.org/10.1111/j.1399-0004.2011.01724.x. Porat-Shliom N, Milberg O, Masedunskas A, Weigert R. Multiple roles for the actin 11 cytoskeleton during regulated exocytosis. Cellular and Molecular Life Sciences 2013;70:2099– 121. https://doi.org/10.1007/s00018-012-1156-5. 12 Whitehead KA, Langer R, Anderson DG. Knocking down barriers: Advances in siRNA delivery. Nat Rev Drug Discov 2009. https://doi.org/10.1038/nrd2742. Vocelle D, Chan C, Walton SP. Endocytosis Controls Small Interfering RNA Efficiency: 13 Implications for Small Interfering RNA Delivery Vehicle Design and Cell-Specific Targeting. Nucleic Acid Therapies 2020;30:22–32. https://doi.org/https://doi.org/10.1089/nat.2019.0804. Reifarth M, Hoeppener S, Schubert US. Uptake and Intracellular Fate of Engineered 14 Nanoparticles in Mammalian Cells: Capabilities and Limitations of Transmission Electron Microscopy—Polymer-Based Nanoparticles. Advanced Materials 2018;30:1–28. https://doi.org/10.1002/adma.201703704. 15 Rennick JJ, Johnston APR, Parton RG. Key principles and methods for studying the 42 endocytosis of biological and nanoparticle therapeutics. Nat Nanotechnol 2021;16:266–76. https://doi.org/10.1038/s41565-021-00858-8. Gutierrez JM, Feizi A, Li S, Kallehauge TB, Hefzi H, Grav LM, et al. Genome-scale 16 reconstructions of the mammalian secretory pathway predict metabolic costs and limitations of protein secretion. Nat Commun 2020;11:1–10. https://doi.org/10.1038/s41467-019-13867-y. 17 Hetz C, Zhang K, Kaufman RJ. Mechanisms, regulation and functions of the unfolded protein response. Nat Rev Mol Cell Biol 2020;21:421–38. https://doi.org/10.1038/s41580-020- 0250-z. Volmer R, Ploeg K Van Der, Ron D. Membrane lipid saturation activates endoplasmic 18 reticulum unfolded protein response transducers through their transmembrane domains. Proc Natl Acad Sci U S A 2013;110:4628–33. https://doi.org/10.1073/pnas.1217611110. Cho H, Stanzione F, Oak A, Kim GH, Yerneni S, Qi L, et al. Intrinsic Structural Features 19 of the Human IRE1α Transmembrane Domain Sense Membrane Lipid Saturation. Cell Rep 2019;27:307-320.e5. https://doi.org/10.1016/j.celrep.2019.03.017. Oikawa D, Kimata Y, Kohno K, Iwawaki T. Activation of mammalian IRE1α upon ER 20 stress depends on dissociation of BiP rather than on direct interaction with unfolded proteins. Exp Cell Res 2009;315:2496–504. https://doi.org/10.1016/j.yexcr.2009.06.009. Ali MMU, Bagratuni T, Davenport EL, Nowak PR, Silva-Santisteban MC, Hardcastle A, 21 et al. Structure of the Ire1 autophosphorylation complex and implications for the unfolded protein response. EMBO Journal 2011;30:894–905. https://doi.org/10.1038/emboj.2011.18. Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. XBP1 mRNA Is Induced by 22 ATF6 and Spliced by IRE1 in Response to ER Stress to Produce a Highly Active Transcription Factor phorylation, the activated Ire1p specifically cleaves HAC1 precursor mRNA to remove an intron of 252 nucle-otides. The cleaved 5 and . Cell 2001;107:881–91. Lee A-H, Iwakoshi NN, Glimcher LH. XBP-1 Regulates a Subset of Endoplasmic 23 Reticulum Resident Chaperone Genes in the Unfolded Protein Response. Mol Cell Biol 2003;23:7448–59. https://doi.org/10.1128/mcb.23.21.7448-7459.2003. Hollien J, Weissman JS. Decay of Endoplasmic Reticulum-Localized mRNAs During the 24 Unfolded Protein Response. Science (1979) 2006;313:104–7. https://doi.org/10.1126/science.1126121. 25 Tirasophon W, Lee K, Callaghan B, Welihinda A, Kaufman RJ. The endoribonuclease activity of mammalian IRE1 autoregulates its mRNA and is required for the unfolded protein response. Genes Dev 2000;14:2725–36. https://doi.org/10.1101/gad.839400. 26 Thomas A Le, Ferri E, Marsters S, Harnoss JM, Lawrence DA, Zuazo-Gaztelu I, et al. Decoding non-canonical mRNA decay by the endoplasmic-reticulum stress sensor IRE1α. Nat Commun 2021;12:. https://doi.org/10.1038/s41467-021-27597-7. Hollien J, Lin JH, Li H, Stevens N, Walter P, Weissman JS. Regulated Ire1-dependent 27 decay of messenger RNAs in mammalian cells. Journal of Cell Biology 2009;186:323–31. https://doi.org/10.1083/jcb.200903014. 28 Han D, Lerner AG, Walle L Vande, Upton JP, Xu W, Hagen A, et al. IRE1α Kinase Activation Modes Control Alternate Endoribonuclease Outputs to Determine Divergent Cell 43 Fates. Cell 2009;138:562–75. https://doi.org/10.1016/j.cell.2009.07.017. Liu CY, Schroder M, Kaufman RJ. Ligand-independent dimerization activates the stress 29 response kinases IRE1 and PERK in the lumen of the endoplasmic reticulum. Journal of Biological Chemistry 2000;275:24881–5. https://doi.org/10.1074/jbc.M004454200. 30 B’Chir W, Maurin AC, Carraro V, Averous J, Jousse C, Muranishi Y, et al. The eIF2α/ATF4 pathway is essential for stress-induced autophagy gene expression. Nucleic Acids Res 2013;41:7683–99. https://doi.org/10.1093/nar/gkt563. 31 McCullough KD, Martindale JL, Klotz L-O, Aw T-Y, Holbrook NJ. Gadd153 Sensitizes Cells to Endoplasmic Reticulum Stress by Down-Regulating Bcl2 and Perturbing the Cellular Redox State. Mol Cell Biol 2001;21:1249–59. https://doi.org/10.1128/mcb.21.4.1249-1259.2001. Lu TH, Tseng TJ, Su CC, Tang FC, Yen CC, Liu YY, et al. Arsenic induces reactive 32 oxygen species-caused neuronal cell apoptosis through JNK/ERK-mediated mitochondria- dependent and GRP 78/CHOP-regulated pathways. Toxicol Lett 2014;224:130–40. https://doi.org/10.1016/j.toxlet.2013.10.013. Boye E, Grallert B. eIF2α phosphorylation and the regulation of translation. Curr Genet 33 2020;66:293–7. https://doi.org/10.1007/s00294-019-01026-1. Nevoltris D, Chames P. Antibody Engineering Methods and Protocols Third Edition. 34 Antibody Eng, vol. 1827. 2018. p. 505–40. Haze K, Okada T, Yoshida H, Yanagi H, Yura T, Negishi M, et al. Identification of the 35 G13 (cAMP-response-element-binding protein-related protein) gene product related to activating transcription factor 6 as a transcriptional activator of the mammalian unfolded protein response. Biochem J 2001;355:19–28. Ye J, Rawson RB, Komuro R, Chen X, Davé UP, Prywes R, et al. ER Stress Induces 36 Cleavage of Membrane-Bound ATF6 by the Same Proteases that Process SREBPs. Mol Cell 2000;6:1355–64. Okada T, Yoshida H, Akazawa R, Negishi M, Mori K. Distinct roles of activating 37 transcription factor 6 (ATF6) and double-stranded RNA-activated protein kinase-like endoplasmic reticulum kinase (PERK) in transcription during the mammalian unfolded protein response. Biochem J 2002;366:585–94. 38 Meusser B, Hirsch C, Jarosch E, Sommer T. ERAD: The long road to destruction. Nat Cell Biol 2005;7:766–72. https://doi.org/10.1038/ncb0805-766. Ichimiya T, Yamakawa T, Hirano T, Yokoyama Y, Hayashi Y, Hirayama D, et al. 39 Autophagy and autophagy‐related diseases: A review. Int J Mol Sci 2020;21:1–21. https://doi.org/10.3390/ijms21238974. Bonam SR, Ruff M, Muller S. HSPA8/HSC70 in Immune Disorders: A Molecular 40 Rheostat that Adjusts Chaperone-Mediated Autophagy Substrates. Cells 2019;8:849–75. Dice JF. Chaperone-mediated autophagy. Autophagy 2007;3:295–9. 41 https://doi.org/10.4161/auto.4144. 42 Majeski AE, Fred Dice J. Mechanisms of chaperone-mediated autophagy. International Journal of Biochemistry and Cell Biology 2004;36:2435–44. 44 https://doi.org/10.1016/j.biocel.2004.02.013. 43 ClinicalTrials.gov. U.S. National Library of Medicine. 2024. 44 Lien PW, Joshi M, Tice JA, Agboola F, Nikitin D, Withanawasam V, et al. Cost- effectiveness of eculizumab and efgartigimod for the treatment of anti–acetylcholine receptor antibody–positive generalized myasthenia gravis. J Manag Care Spec Pharm 2024;30:517–27. https://doi.org/10.18553/jmcp.2024.30.6.517. Navarro CE, Betancur JE. Cost-Utility Analysis Comparing Ocrelizumab vs Rituximab in 45 the Treatment of Relapsing-Remitting Multiple Sclerosis: The Colombian Perspective. Value Health Reg Issues 2023;36:83–91. https://doi.org/10.1016/j.vhri.2023.02.008. 46 Mahal H, Branton H, Farid SS. End-to-end continuous bioprocessing: Impact on facility design, cost of goods, and cost of development for monoclonal antibodies. Biotechnol Bioeng 2021;118:3468–85. https://doi.org/10.1002/bit.27774. Torres M, Hussain H, Dickson AJ. The secretory pathway–the key for unlocking the 47 potential of Chinese hamster ovary cell factories for manufacturing therapeutic proteins. Crit Rev Biotechnol 2023;43:628–45. https://doi.org/10.1080/07388551.2022.2047004. Keane JT, Ryan D, Gray PP. Effect of shear stress on expression of a recombinant protein 48 by Chinese hamster ovary cells. Biotechnol Bioeng 2003;81:211–20. https://doi.org/10.1002/bit.10472. Coulet M, Kepp O, Kroemer G, Basmaciogullari S. Metabolic Profiling of CHO Cells 49 during the Production of Biotherapeutics. Cells 2022;11:1–21. https://doi.org/10.3390/cells11121929. Handlogten MW, Zhu M, Ahuja S. Intracellular response of CHO cells to oxidative stress 50 and its influence on metabolism and antibody production. Biochem Eng J 2018;133:12–20. https://doi.org/10.1016/j.bej.2018.01.031. Kaufmann H, Mazur X, Fussenegger M, Bailey JE. Influence of low temperature on 51 productivity, proteome and protein phosphorylation of CHO cells. Biotechnol Bioeng 1999;63:573–82. https://doi.org/10.1002/(SICI)1097-0290(19990605)63:5<573::AID- BIT7>3.0.CO;2-Y. Zhou Y, Raju R, Alves C, Gilbert A. Debottlenecking protein secretion and reducing 52 protein aggregation in the cellular host. Curr Opin Biotechnol 2018;53:151–7. https://doi.org/10.1016/j.copbio.2018.01.007. Qiu Y, Mao T, Zhang Y, Shao M, You J, Ding Q, et al. A Crucial Role for RACK1 in the 53 Regulation of Glucose-Stimulated IRE1a Activation in Pancreatic b Cells. Sci Signal 2010;3:1– 10. Lipson KL, Fonseca SG, Ishigaki S, Nguyen LX, Foss E, Bortell R, et al. Regulation of 54 insulin biosynthesis in pancreatic beta cells by an endoplasmic reticulum-resident protein kinase IRE1. Cell Metab 2006;4:245–54. https://doi.org/10.1016/j.cmet.2006.07.007. Hassler JR, Scheuner DL, Wang S, Han J, Kodali VK, Li P, et al. The IRE1α/XBP1s 55 Pathway Is Essential for the Glucose Response and Protection of β Cells. PLoS Biol 2015;13:1– 22. https://doi.org/10.1371/journal.pbio.1002277. 45 Cunha DA, Hekerman P, Ladrière L, Bazarra-Castro A, Ortis F, Wakeham MC, et al. 56 Initiation and execution of lipotoxic ER stress in pancreatic β-cells. J Cell Sci 2008;121:2308– 18. https://doi.org/10.1242/jcs.026062. Robblee MM, Kim CC, Abate JP, Valdearcos M, Sandlund KLM, Shenoy MK, et al. 57 Saturated Fatty Acids Engage an IRE1α-Dependent Pathway to Activate the NLRP3 Inflammasome in Myeloid Cells. Cell Rep 2016;14:2611–23. https://doi.org/10.1016/j.celrep.2016.02.053. Sanson M, Augé N, Vindis C, Muller C, Bando Y, Thiers JC, et al. Oxidized low-density 58 lipoproteins trigger endoplasmic reticulum stress in vascular cells: Prevention by oxygen- regulated protein 150 expression. Circ Res 2009;104:328–36. https://doi.org/10.1161/CIRCRESAHA.108.183749. Ning J, Hong T, Ward A, Pi J, Liu Z, Liu HY, et al. Constitutive role for IRE1α-XBP1 59 signaling pathway in the insulin-mediated hepatic lipogenic program. Endocrinology 2011;152:2247–55. https://doi.org/10.1210/en.2010-1036. 60 Mao T, Shao M, Qiu Y, Huang J, Zhang Y, Song B, et al. PKA phosphorylation couples hepatic inositol-requiring enzyme 1α to glucagon signaling in glucose metabolism. Proc Natl Acad Sci U S A 2011;108:15852–7. https://doi.org/10.1073/pnas.1107394108. Shao M, Shan B, Liu Y, Deng Y, Yan C, Wu Y, et al. Hepatic IRE1α regulates fasting- 61 induced metabolic adaptive programs through the XBP1s-PPARα axis signalling. Nat Commun 2014;5:3528. https://doi.org/10.1038/ncomms4528. 62 Shan B, Wang X, Wu Y, Xu C, Xia Z, Dai J, et al. The metabolic ER stress sensor IRE1α suppresses alternative activation of macrophages and impairs energy expenditure in obesity. Nat Immunol 2017;18:519–29. https://doi.org/10.1038/ni.3709. 63 Chen C, Garcia Z, Chen D, Liu H, Trelstad P. Cost and supply considerations for antibody therapeutics. MAbs 2025;17:. https://doi.org/10.1080/19420862.2025.2451789. Dipaolo B, Pennetti A, Nugent L, Venkat K, Venkat K. Monitoring impurities in 64 biopharmaceuticals produced by recombinant technology. Pharmaceuitical Science & Technology Today 1999;2:70–82. 65 Walls D, Loughran S. Methods in Molecular Biology: Protein Chromatography. vol. 1485. 2017. Ramos-de-la-Peña AM, González-Valdez J, Aguilar O. Protein A chromatography: 66 Challenges and progress in the purification of monoclonal antibodies. J Sep Sci 2019:1816–27. https://doi.org/10.1002/jssc.201800963. 67 Le H, Vishwanathan N, Jacob NM, Gadgil M, Hu WS. Cell line development for biomanufacturing processes: recent advances and an outlook. Biotechnol Lett 2015;37:1553–64. https://doi.org/10.1007/s10529-015-1843-z. Tihanyi B, Nyitray L. Recent advances in CHO cell line development for recombinant 68 protein production. Drug Discov Today Technol 2020;38:25–34. https://doi.org/10.1016/j.ddtec.2021.02.003. 69 Majumdar S, Desai R, Hans A, Dandekar P, Jain R. From Efficiency to Yield: Exploring Recent Advances in CHO Cell Line Development for Monoclonal Antibodies. Mol Biotechnol 46 2024;67:369–92. https://doi.org/10.1007/s12033-024-01060-6. Narasipura EA, VanKeulen-Miller R, Ma Y, Fenton OS. Ongoing Clinical Trials of 70 Nonviral siRNA Therapeutics. Bioconjug Chem 2023;34:1177–97. https://doi.org/10.1021/acs.bioconjchem.3c00205. Pratt AJ, MacRae IJ. The RNA-induced silencing complex: A versatile gene-silencing 71 machine. Journal of Biological Chemistry 2009;284:17897–901. https://doi.org/10.1074/jbc.R900012200. 72 Chendrimada TP, Gregory RI, Kumaraswamy E, Norman J, Cooch N, Nishikura K, et al. TRBP recruits the Dicer complex to Ago2 for microRNA processing and gene silencing. Nature 2005. https://doi.org/10.1038/nature03868. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific 73 genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998;391:806– 11. https://doi.org/10.1038/35888. Adams D, Gonzalez-Duarte A, O’Riordan WD, Yang CC, Ueda M, Kristen A V., et al. 74 Patisiran, an RNAi therapeutic, for hereditary transthyretin amyloidosis. New England Journal of Medicine 2018;379:11–21. https://doi.org/10.1056/NEJMoa1716153. Scott LJ. Givosiran: First Approval. Drugs 2020;80:335–9. 75 https://doi.org/10.1007/s40265-020-01269-0. 76 Garrelfs SF, Frishberg Y, Hulton SA, Koren MJ, O’Riordan WD, Cochat P, et al. Lumasiran, an RNAi Therapeutic for Primary Hyperoxaluria Type 1. New England Journal of Medicine 2021;384:1216–26. https://doi.org/10.1056/nejmoa2021712. 77 Wołowiec Ł, Osiak J, Wołowiec A, Wijata A, Grześk E, Kozakiewicz M, et al. Inclisiran—Safety and Effectiveness of Small Interfering RNA in Inhibition of PCSK-9. Pharmaceutics 2023;15:. https://doi.org/10.3390/pharmaceutics15020323. 78 Adams D, Tournev IL, Taylor MS, Coelho T, Planté-Bordeneuve V, Berk JL, et al. Efficacy and safety of vutrisiran for patients with hereditary transthyretin-mediated amyloidosis with polyneuropathy: a randomized clinical trial. Amyloid 2023;30:18–26. https://doi.org/10.1080/13506129.2022.2091985. Syed YY. Nedosiran: First Approval. Drugs 2023;83:1729–33. 79 https://doi.org/10.1007/s40265-023-01976-4. Lu D, Dou F, Gao J. Fitusiran : The first approved siRNA therapy for hemophilia via 80 reducing plasma antithrombin levels 2025:1–2. https://doi.org/10.5582/ddt.2025.01031. 81 Dowdy SF, Setten RL, Cui XS, Jadhav SG. Delivery of RNA Therapeutics: The Great Endosomal Escape! Nucleic Acid Ther 2022;32:361–8. https://doi.org/10.1089/nat.2022.0004. Ali Zaidi SS, Fatima F, Ali Zaidi SA, Zhou D, Deng W, Liu S. Engineering siRNA 82 therapeutics: challenges and strategies. J Nanobiotechnology 2023;21:1–15. https://doi.org/10.1186/s12951-023-02147-z. Hu B, Zhong L, Weng Y, Peng L, Huang Y, Zhao Y, et al. Therapeutic siRNA: state of 83 the art. Signal Transduct Target Ther 2020;5:. https://doi.org/10.1038/s41392-020-0207-x. 84 Sioud M, Furset G, Cekaite L. Suppression of immunostimulatory siRNA-driven innate 47 immune activation by 2′-modified RNAs. Biochem Biophys Res Commun 2007;361:122–6. https://doi.org/10.1016/j.bbrc.2007.06.177. Zhang L, Liang Y, Liang G, Tian Z, Zhang Y, Liu Z, et al. The therapeutic prospects of 85 N-acetylgalactosamine-siRNA conjugates. Front Pharmacol 2022;13:1–13. https://doi.org/10.3389/fphar.2022.1090237. Subhan MA, Filipczak N, Torchilin VP. Advances with Lipid-Based Nanosystems for 86 siRNA Delivery to Breast Cancers. Pharmaceuticals 2023;16:. https://doi.org/10.3390/ph16070970. Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, et al. Safety and 87 Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. New England Journal of Medicine 2020;383:2603–15. https://doi.org/10.1056/nejmoa2034577. Baden LR, El Sahly HM, Essink B, Kotloff K, Frey S, Novak R, et al. Efficacy and 88 Safety of the mRNA-1273 SARS-CoV-2 Vaccine. New England Journal of Medicine 2021;384:403–16. https://doi.org/10.1056/nejmoa2035389. 89 Wang H, Zhang S, Lv J, Cheng Y. Design of polymers for siRNA delivery: Recent progress and challenges. View 2021;2:1–24. https://doi.org/10.1002/VIW.20200026. Friedrich M, Aigner A. Therapeutic siRNA: State-of-the-Art and Future Perspectives. 90 BioDrugs 2022;36:549–71. https://doi.org/10.1007/s40259-022-00549-3. Yurkin ST, Wang Z. Cell membrane-derived nanoparticles: Emerging clinical 91 opportunities for targeted drug delivery. Nanomedicine 2017;12:2007–19. https://doi.org/10.2217/nnm-2017-0100. O’Brien K, Breyne K, Ughetto S, Laurent LC, Breakefield XO. RNA delivery by 92 extracellular vesicles in mammalian cells and its applications. Nat Rev Mol Cell Biol 2020;21:585–606. https://doi.org/10.1038/s41580-020-0251-y. 93 Mosbach ML, Pfafenrot C, von Strandmann EP, Bindereif A, Preußer C. Molecular determinants for rna release into extracellular vesicles. Cells 2021;10:. https://doi.org/10.3390/cells10102674. Roerig J, Mitrach F, Schmid M, Hause G, Hacker MC, Wölk C, et al. Synergistic siRNA 94 Loading of Extracellular Vesicles Enables Functional Delivery into Cells. Small Methods 2022;6:. https://doi.org/10.1002/smtd.202201001. El Moukhtari SH, Garbayo E, Amundarain A, Pascual-Gil S, Carrasco-León A, Prosper 95 F, et al. Lipid nanoparticles for siRNA delivery in cancer treatment. Journal of Controlled Release 2023;361:130–46. https://doi.org/10.1016/j.jconrel.2023.07.054. Homma Y, Hiragi S, Fukuda M. Rab family of small GTPases: an updated view on their 96 regulation and functions. FEBS Journal 2021;288:36–55. https://doi.org/10.1111/febs.15453. 97 Lönn P, Kacsinta AD, Cui XS, Hamil AS, Kaulich M, Gogoi K, et al. Enhancing Endosomal Escape for Intracellular Delivery of Macromolecular Biologic Therapeutics. Sci Rep 2016;6:1–9. https://doi.org/10.1038/srep32301. Dowdy SF. Endosomal escape of RNA therapeutics: How do we solve this rate-limiting 98 problem? Rna 2023;29:396–401. https://doi.org/10.1261/rna.079507.122. 48 Vocelle D, Chesniak OM, Smith MR, Chan C, Walton SP. Kinetic analysis of the 99 intracellular processing of sirnas by confocal microscopy. Microscopy 2020:1–7. https://doi.org/10.1093/jmicro/dfaa031. 100 Deng Y, Storrie B. Animal cell lysosomes rapidly exchange membrane proteins. Proc Natl Acad Sci U S A 1988;85:3860–4. https://doi.org/10.1073/pnas.85.11.3860. Scott CC, Vacca F, Gruenberg J. Endosome maturation, transport and functions. Semin 101 Cell Dev Biol 2014;31:2–10. https://doi.org/10.1016/j.semcdb.2014.03.034. 102 Benjaminsen R V., Mattebjerg MA, Henriksen JR, Moghimi SM, Andresen TL. The possible "proton sponge " effect of polyethylenimine (PEI) does not include change in lysosomal pH. Molecular Therapy 2013;21:149–57. https://doi.org/10.1038/mt.2012.185. Takikawa M, Fujisawa M, Yoshino K, Takeoka S. Intracellular distribution of lipids and 103 encapsulated model drugs from cationic liposomes with different uptake pathways. Int J Nanomedicine 2020;15:8401–9. https://doi.org/10.2147/IJN.S267638. 104 Brown CR, Gupta S, Qin J, Racie T, He G, Lentini S, et al. Investigating the pharmacodynamic durability of GalNAc-siRNA conjugates. Nucleic Acids Res 2021;48:11827– 44. https://doi.org/10.1093/nar/gkaa670. 105 Abdelaal AM, Kasinski AL. Ligand-mediated delivery of RNAi-based therapeutics for the treatment of oncological diseases. NAR Cancer 2021;3:1–23. https://doi.org/10.1093/narcan/zcab030. 106 Ranjbar S, Zhong X bo, Manautou J, Lu X. A holistic analysis of the intrinsic and delivery-mediated toxicity of siRNA therapeutics. Adv Drug Deliv Rev 2023;201:. https://doi.org/10.1016/j.addr.2023.115052. 107 Unger JM, Hershman DL, Fleury ME. Association of Patient Comorbid Conditions With Cancer Clinical Trial Participation. JAMA Oncol 2019;5:326–33. https://doi.org/10.1001/jamaoncol.2018.5953. 108 Yoshida H. ER stress and diseases. FEBS Journal 2007;274:630–58. https://doi.org/10.1111/j.1742-4658.2007.05639.x. Schröder M, Kaufman RJ. ER stress and the unfolded protein response. Mutation 109 Research - Fundamental and Molecular Mechanisms of Mutagenesis 2005;569:29–63. https://doi.org/10.1016/j.mrfmmm.2004.06.056. Sinharoy P, Aziz AH, Majewska NI, Ahuja S, Handlogten MW. Perfusion reduces 110 bispecific antibody aggregation via mitigating mitochondrial dysfunction-induced glutathione oxidation and ER stress in CHO cells. Sci Rep 2020;10:1–12. https://doi.org/10.1038/s41598- 020-73573-4. 111 Godoy-Silva R, Chalmers JJ, Casnocha SA, Bass LA, Ma N. Physiological responses of CHO cells to repetitive hydrodynamic stress. Biotechnol Bioeng 2009;103:1103–17. https://doi.org/10.1002/bit.22339. 112 Ha TK, Hansen AH, Kildegaard HF, Lee GM. BiP Inducer X: An ER Stress Inhibitor for Enhancing Recombinant Antibody Production in CHO Cell Culture. Biotechnol J 2019;14:. https://doi.org/10.1002/biot.201900130. 49 113 Hetz C. The unfolded protein response: Controlling cell fate decisions under ER stress and beyond. Nat Rev Mol Cell Biol 2012;13:89–102. https://doi.org/10.1038/nrm3270. 114 Hwang J, Qi L. Quality Control in the Endoplasmic Reticulum: Crosstalk between ERAD and UPR pathways. Trends Biochem Sci 2018;43:593–605. https://doi.org/10.1016/j.tibs.2018.06.005. 115 Diehl JA, McQuiston A. Recent insights into PERK-dependent signaling from the stressed endoplasmic reticulum. F1000Res 2017;6:1–11. https://doi.org/10.12688/f1000research.12138.1. Lei Y, Yu H, Ding S, Liu H, Liu C, Fu R. Molecular mechanism of ATF6 in unfolded 116 protein response and its role in disease. Heliyon 2024;10:e25937. https://doi.org/10.1016/j.heliyon.2024.e25937. Tigges M, Fussenegger M. Xbp1-based engineering of secretory capacity enhances the 117 productivity of Chinese hamster ovary cells. Metab Eng 2006;8:264–72. https://doi.org/10.1016/j.ymben.2006.01.006. Shaffer AL, Shapiro-Shelef M, Iwakoshi NN, Lee AH, Qian SB, Zhao H, et al. XBP1, 118 downstream of Blimp-1, expands the secretory apparatus and other organelles, and increases protein synthesis in plasma cell differentiation. Immunity 2004;21:81–93. https://doi.org/10.1016/j.immuni.2004.06.010. 119 Daverkausen-Fischer L, Draga M, Pröls F. Regulation of Translation, Translocation, and Degradation of Proteins at the Membrane of the Endoplasmic Reticulum. Int J Mol Sci 2022;23:. https://doi.org/10.3390/ijms23105576. Lin LL, Wang HH, Pederson B, Wei X, Torres M, Lu Y, et al. SEL1L-HRD1 interaction 120 is required to form a functional HRD1 ERAD complex. Nat Commun 2024;15:1–20. https://doi.org/10.1038/s41467-024-45633-0. 121 Tang D, Sandoval W, Lam C, Haley B, Liu P, Xue D, et al. UBR E3 ligases and the PDIA3 protease control degradation of unfolded antibody heavy chain by ERAD. Journal of Cell Biology 2020;219:. https://doi.org/10.1083/JCB.201908087. 122 Fujita E, Kouroku Y, Isoai A, Kumagai H, Misutani A, Matsuda C, et al. Two endoplasmic reticulum-associated degradation (ERAD) systems for the novel variant of the mutant dysferlin: Ubiquitin/proteasome ERAD(I) and autophagy/lysosome ERAD(II). Hum Mol Genet 2007;16:618–29. https://doi.org/10.1093/hmg/ddm002. 123 Wang L, Klionsky DJ, Shen HM. The emerging mechanisms and functions of microautophagy. Nat Rev Mol Cell Biol 2023;24:186–203. https://doi.org/10.1038/s41580-022- 00529-z. 124 Yu L, Chen Y, Tooze SA. Autophagy pathway: Cellular and molecular mechanisms. Autophagy 2018;14:207–15. https://doi.org/10.1080/15548627.2017.1378838. Petherick KJ, Conway OJL, Mpamhanga C, Osborne SA, Kamal A, Saxty B, et al. 125 Pharmacological inhibition of ULK1 kinase blocks mammalian target of rapamycin (mTOR)- dependent autophagy. Journal of Biological Chemistry 2015;290:11376–83. https://doi.org/10.1074/jbc.C114.627778. 126 Jayaprakash NG, Surolia A. Role of glycosylation in nucleating protein folding and 50 stability. Biochemical Journal 2017;474:2333–47. https://doi.org/10.1042/BCJ20170111. 127 Ku SCY, Toh PC, Lee YY, Chusainow J, Yap MGS, Chao SH. Regulation of XBP-1 signaling during transient and stable recombinant protein production in CHO cells. Biotechnol Prog 2010;26:517–26. https://doi.org/10.1002/btpr.322. 128 Das CK, Banerjee I, Mandal M. Pro-survival autophagy: An emerging candidate of tumor progression through maintaining hallmarks of cancer. Semin Cancer Biol 2020;66:59–74. https://doi.org/10.1016/j.semcancer.2019.08.020. 129 Gerlach JQ, Sharma S, Leister KJ, Joshi L. Endoplasmic Reticulum Stress in Health and Disease. Dordrecht: Springer Netherlands; 2012. Savage KE, Baur PS. Effect of tunicamycin, an inhibitor of protein glycosylation, on 130 division of tumour cells in vitro. J Cell Sci 1983;64:295–306. https://doi.org/10.1242/jcs.64.1.295. 131 Olden K, Pratt RM, Jaworski C, Yamada KM. Evidence for role of glycoprotein carbohydrates in membrane transport: Specific inhibition by tunicamycin. Proc Natl Acad Sci U S A 1979;76:791–5. https://doi.org/10.1073/pnas.76.2.791. 132 Cavalli G, Cenci S. Autophagy and Protein Secretion. J Mol Biol 2020;432:2525–45. https://doi.org/10.1016/j.jmb.2020.01.015. Lokireddy S, Kukushkin NV, Goldberg AL. cAMP-induced phosphorylation of 26S 133 proteasomes on Rpn6/PSMD11 enhances their activity and the degradation of misfolded proteins. Proc Natl Acad Sci U S A 2015;112:E7176–85. https://doi.org/10.1073/pnas.1522332112. 134 Aslam M, Ladilov Y. Emerging Role of cAMP/AMPK Signaling. Cells 2022;11:308. https://doi.org/10.3390/cells11020308. Esmail S, Manolson MF. Advances in understanding N-glycosylation structure, function, 135 and regulation in health and disease. Eur J Cell Biol 2021;100:151186. https://doi.org/10.1016/j.ejcb.2021.151186. Luhr M, Torgersen ML, Szalai P, Hashim A, Brech A, Staerk J, et al. The kinase PERK 136 and the transcription factor ATF4 play distinct and essential roles in autophagy resulting from tunicamycin-induced ER stress. Journal of Biological Chemistry 2019;294:8197–217. https://doi.org/10.1074/jbc.RA118.002829. 137 Song W, Ma Z, Zhang Y, Yang C. Autophagy plays a dual role during intracellular siRNA delivery by lipoplex and polyplex nanoparticles. Acta Biomater 2017;58:196–204. https://doi.org/10.1016/j.actbio.2017.05.038. 138 Gade P, Ramachandran G, Maachani UB, Rizzo MA, Okada T, Prywes R, et al. An IFN- γ-stimulated ATF6-C/EBP-β-signaling pathway critical for the expression of death associated protein kinase 1 and induction of autophagy. Proc Natl Acad Sci U S A 2012;109:10316–21. https://doi.org/10.1073/pnas.1119273109. Zhou H, Liu R. ER stress and hepatic lipid metabolism. Front Genet 2014;5:1–7. 139 https://doi.org/10.3389/fgene.2014.00112. 140 Maruyama R, Kamoshida Y, Shimizu M, Inoue J, Sato R. ATF6α stimulates 51 cholesterogenic gene expression and de novo cholesterol synthesis. Biosci Biotechnol Biochem 2013;77:1734–8. https://doi.org/10.1271/bbb.130295. 141 Qian H, Zhao Y, Wang Y, Zhao H, Cui J, Wang Z, et al. ATR induces hepatic lipid metabolism disorder in rats by activating IRE1α/XBP1 signaling pathway. Toxicology 2024;501:153696. https://doi.org/10.1016/j.tox.2023.153696. 142 Liu HH, Xu Y, Li CJ, Hsu SJ, Lin XH, Zhang R, et al. An SCD1-dependent mechanoresponsive pathway promotes HCC invasion and metastasis through lipid metabolic reprogramming. Molecular Therapy 2022;30:2554–67. https://doi.org/10.1016/j.ymthe.2022.03.015. 143 Chen H, Yuan R, Zhang Y, Zhang X, Chen L, Zhou X, et al. ATF4 regulates SREBP1c expression to control fatty acids synthesis in 3T3-L1 adipocytes differentiation. Biochim Biophys Acta Gene Regul Mech 2016;1859:1459–69. https://doi.org/10.1016/j.bbagrm.2016.07.010. 144 Rashid HO, Yadav RK, Kim HR, Chae HJ. ER stress: Autophagy induction, inhibition and selection. Autophagy 2015;11:1956–77. https://doi.org/10.1080/15548627.2015.1091141. Tanida I. Autophagosome formation and molecular mechanism of autophagy. Antioxid 145 Redox Signal 2011;14:2201–14. https://doi.org/10.1089/ars.2010.3482. 146 Birgisdottir ÅB, Johansen T. Autophagy and endocytosis – interconnections and interdependencies. J Cell Sci 2020;133:. https://doi.org/10.1242/jcs.228114. 147 Dong G, Liu Y, Zhang L, Huang S, Ding HF, Dong Z. mTOR contributes to ER stress and associated apoptosis in renal tubular cells. Am J Physiol Renal Physiol 2015;308:F267–74. https://doi.org/10.1152/ajprenal.00629.2014. Lee GH, Kim DS, Kim HT, Lee JW, Chung CH, Ahn T, et al. Enhanced lysosomal 148 activity is involved in bax inhibitor-1-induced regulation of the endoplasmic reticulum (ER) stress response and cell death against ER stress: Involvement of vacuolar H +-ATPase (V- ATPase). Journal of Biological Chemistry 2011;286:24743–53. https://doi.org/10.1074/jbc.M110.167734. 149 Splichal RC, Gredell JA, Vogel EB, Malefyt A, Comiskey G, Smith MR, et al. Modulating Polymer-siRNA Binding Does Not Promote Polyplex-Mediated Silencing. Nucleic Acid Ther 2021;31:229–36. 150 Oliveira S, Storm G, Schiffelers R. Targeted delivery of siRNA. J Biomed Biotechnol 2006;2006:1–9. https://doi.org/10.1155/JBB/2006/63675. 151 Angart P, Vocelle D, Chan C, Walton SP. Design of siRNA Therapeutics from the Molecular Scale. Pharmaceuticals 2013;6:440–68. https://doi.org/10.3390/ph6040440. Setten RL, Rossi JJ, Han S ping. The current state and future directions of RNAi-based 152 therapeutics. Nat Rev Drug Discov 2019;18:421–46. https://doi.org/10.1038/s41573-019-0017-4. 153 Kanasty R, Dorkin JR, Vegas A, Anderson D. Delivery materials for siRNA therapeutics. Nat Mater 2013;12:967–77. https://doi.org/10.1038/nmat3765. 154 O’Mahony AM, Ogier J, Desgranges S, Cryan JF, Darcy R, O’Driscoll CM. A click chemistry route to 2-functionalised PEGylated and cationic β-cyclodextrins: Co-formulation 52 opportunities for siRNA delivery. Org Biomol Chem 2012;10:4954–60. https://doi.org/10.1039/c2ob25490e. Jiang X, Vogel EB, Smith MR, Baker GL. ‘Clickable’ polyglycolides: Tunable synthons 155 for thermoresponsive, degradable polymers. Macromolecules 2008;41:1937–44. https://doi.org/10.1021/ma7027962. 156 Howard KA, Rahbek UL, Liu X, Damgaard CK, Glud SZ, Andersen M, et al. RNA Interference in Vitro and in Vivo Using a Novel Chitosan/siRNA Nanoparticle System. Molecular Therapy 2006;14:476–84. https://doi.org/10.1016/j.ymthe.2006.04.010. 157 Nelson CE, Kintzing JR, Hanna A, Shannon JM, Gupta MK, Duvall CL. Balancing cationic and hydrophobic content of PEGylated siRNA polyplexes enhances endosome escape, stability, blood circulation time, and bioactivity in vivo. ACS Nano 2013;7:8870–80. https://doi.org/10.1021/nn403325f. Patil Y, Panyam J. Polymeric nanoparticles for siRNA delivery and gene silencing. Int J 158 Pharm 2009;367:195–203. https://doi.org/10.1016/j.ijpharm.2008.09.039. 159 Tsai LR, Chen MH, Chien C Te, Chen MK, Lin FS, Lin KMC, et al. A single-monomer derived linear-like PEI-co-PEG for siRNA delivery and silencing. Biomaterials 2011;32:3647– 53. https://doi.org/10.1016/j.biomaterials.2011.01.059. Zhu C, Jung S, Luo S, Meng F, Zhu X, Park TG, et al. Co-delivery of siRNA and 160 paclitaxel into cancer cells by biodegradable cationic micelles based on PDMAEMA-PCL- PDMAEMA triblock copolymers. Biomaterials 2010;31:2408–16. https://doi.org/10.1016/j.biomaterials.2009.11.077. 161 Kolb HC, Finn MG, Sharpless KB. Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angewandte Chemie - International Edition 2001;40:2004–21. https://doi.org/10.1002/1521-3773(20010601)40:11<2004::AID-ANIE2004>3.0.CO;2-5. 162 Günay KA, Theato P, Klok HA. Standing on the shoulders of hermann staudinger: Post- polymerization modification from past to present. J Polym Sci A Polym Chem 2013;51:1–28. https://doi.org/10.1002/pola.26333. 163 Rostovtsev V V., Green LG, Fokin V V., Sharpless KB. A stepwise huisgen cycloaddition process: Copper(I)-catalyzed regioselective ‘ligation’ of azides and terminal alkynes. Angewandte Chemie - International Edition 2002;41:2596–9. https://doi.org/10.1002/1521-3773(20020715)41:14<2596::AID-ANIE2596>3.0.CO;2-4. 164 Haldón E, Nicasio MC, Pérez PJ. Copper-catalysed azide-alkyne cycloadditions (CuAAC): An update. Org Biomol Chem 2015;13:9528–50. https://doi.org/10.1039/c5ob01457c. 165 Portis AM, Carballo G, Baker GL, Chan C, Walton SP. Confocal microscopy for the analysis of siRNA delivery by polymeric nanoparticles. Microsc Res Tech 2010;73:878–85. https://doi.org/10.1002/jemt.20861. 166 Hill A. The possible effects of the aggregation of molecules of haemoglobin on its dissociation curves. Journal of Physiology 1910;40:iv–vii. Liu X, Howard KA, Dong M, Andersen M, Rahbek UL, Johnsen MG, et al. The 167 influence of polymeric properties on chitosan/siRNA nanoparticle formulation and gene silencing. Biomaterials 2007;28:1280–8. https://doi.org/10.1016/j.biomaterials.2006.11.004. 53 168 Kakizawa Y, Kataoka K. Block copolymer micelles for delivery of gene and related compounds. Adv Drug Deliv Rev 2002;54:203–22. https://doi.org/10.1016/S0169- 409X(02)00017-0. 169 Kim HJ, Miyata K, Nomoto T, Zheng M, Kim A, Liu X, et al. SiRNA delivery from triblock copolymer micelles with spatially-ordered compartments of PEG shell, siRNA-loaded intermediate layer, and hydrophobic core. Biomaterials 2014;35:4548–56. https://doi.org/10.1016/j.biomaterials.2014.02.016. 170 Zhu J, Tang A, Law LP, Feng M, Ho KM, Lee DKL, et al. Amphiphilic core-shell nanoparticles with poly(ethylenimine) shells as potential gene delivery carriers. Bioconjug Chem 2005;16:139–46. https://doi.org/10.1021/bc049895l. 171 Boussif O, LezoualC’H F, Zanta MA, Mergny MD, Scherman D, Demeneix B, et al. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: Polyethylenimine. Proc Natl Acad Sci U S A 1995;92:7297–301. https://doi.org/10.1073/pnas.92.16.7297. 172 Kandil R, Xie Y, Heermann R, Isert L, Jung K, Mehta A, et al. Coming in and Finding Out: Blending Receptor‐Targeted Delivery and Efficient Endosomal Escape in a Novel Bio‐ Responsive siRNA Delivery System for Gene Knockdown in Pulmonary T Cells. Adv Ther (Weinh) 2019;2:1900047–61. https://doi.org/10.1002/adtp.201900047. Sahay G, Querbes W, Alabi C, Eltoukhy A, Sarkar S, Zurenko C, et al. Efficiency of 173 siRNA delivery by lipid nanoparticles is limited by endocytic recycling. Nat Biotechnol 2013;31:653–8. https://doi.org/10.1038/nbt.2614. 174 Kaufman WL, Kocman I, Agrawal V, Rahn HP, Besser D, Gossen M. Homogeneity and persistence of transgene expression by omitting antibiotic selection in cell line isolation. Nucleic Acids Res 2008;36:1–9. https://doi.org/10.1093/nar/gkn508. 54 APPENDIX A: SUPPLEMENTAL MATERIAL FOR CHAPTER 2 Comparing micrographs of control cells (Figure A1A-C) with TM treated cells (Figure A1D-F) showed that in each cell more LC3B autophagy marker and Gluc are retained. 3MA treatment inhibited autophagy as shown by the lack of LC3B in both TM-stressed and unstressed cells (Figure A1G and J) with a reduction in intracellular Gluc signal (Figure A1H and K). VR23 treatment caused an increase in autophagy (Figure A1M). VR23 and TM cotreatment caused the most significant autophagy increase and Gluc accumulation (Figure A1P-R). RP treatment reduced the LC3B signal (SF1S and V) without impacting Gluc signal (Figure A1T and W). These findings are consistent with data generated by cell lysate luciferase activity assays. Figure A1 Micrographs of HeLa cells taken 48 hours post transfection with pDNA encoding CFP-Gluc and allowed to recover in the presence or absence of TM and small molecules affecting degradation. (A,D,G,J,M,P,S,V) LC3B-yellow; (B,E,H,K,N,Q,T,W) Gluc-blue; and (C,F,I,L,O,R,U,X) the overlay of LC3B and Gluc. A-C) no small molecule drugs, D-F) 5 µg/mL TM, G-I) 5 mM 3MA, J-L) 5 µg/mL TM and 5 mM 3MA, M-O) 6 µM VR23, P-R) 6 µM VR23 and 5 µg/mL TM, S-U) 60 µM RP, and V-X) 60 µM RP and 5 µg/mL TM. TM treatment caused an increase in the LC3B signal and Gluc signal in both the WT and IRE1α-/- cells (compare Figure A2A/B with D/E and G/H with J/I). TM increased retention of both LC3B and Gluc in the IRE1α-/- cells as compared with control (compare Figure A2D/E with J/I). These findings are consistent with cell lysate luciferase activity measurements. 55 Figure A2 Micrographs of MDA-MB-231 WT (A-F) and KO (G-L) cells taken 48 hours post transfection with pDNA encoding CFP-Gluc and allowed to recover in the absence (A-C and G- I) or presence (D-F and J-L) of TM. 56 APPENDIX B: MODULATING POLYMER-SIRNA BINDING DOES NOT PROMOTE POLYPLEX MEDIATED SILENCING Note: this appendix is modified from previously published work149. AB.1 Introduction Widespread use of small interfering RNAs (siRNAs) in clinical practice will require improvement in the approaches and molecules used for siRNA delivery. Viral and non-viral delivery approaches are being developed. While viral vectors are generally more efficient150, safety concerns make further development of non-viral delivery approaches essential. To maximize specific activity of the delivered siRNA, delivery vehicles must be designed to (i) protect siRNAs from degradation by serum nucleases, (ii) avoid non-targeted tissues, (iii) be endocytosed efficiently by the target cells, (iv) allow siRNAs to escape endosomal vesicles, and (v) be cytocompatible151-153. To date, most studies that have explored the structure-function relationships of siRNA delivery vehicles have used cell culture screens to identify delivery systems for further investigation in vivo154. The basic chemistries used in these studies are typically cationic lipids or polymers, as they can readily complex with siRNAs by electrostatic self-assembly. Cationic polymers provide tunability in designing delivery systems with controlled physical and chemical characteristics155,156. The size of cationic polymer-siRNA complexes can be controlled by altering the length of the polymer, the relative amounts of hydrophobic and hydrophilic functionality of the polymer, and the amount of siRNA incorporated in the complex157. Charge and charge density can be controlled by varying the chemistry and quantity of cationic monomer or through incorporation of non-charged segments such as poly(ethylene glycol) (PEG)158. Incorporation of PEG increased the half-life of siRNA-containing complexes in vivo and generally decreased cytotoxicity159. Additionally, monomers can be selected to add biodegradability and pH sensitivity to block co-polymeric complexes that have been proven effective siRNA delivery vehicles 159-161. In this work, we sought to identify structure-function relationships for polymeric siRNA delivery vehicles. To achieve structural diversity, we used a post-polymerization modification strategy to synthesize a diverse set of polymers using copper(1)-catalyzed alkyne-azide cycloaddition (CuAAC), a type of “click” chemistry, to attach various side chains to a degradable poly(propargyl glycolide) (PPGL) backbone162-164. Using this approach allowed us to test polymers with structures/functionalities that mimic polymeric vehicles that have previously 57 proven useful as delivery vehicles and to create and test new structures. We measured the in vitro binding of our polymers to short nucleic acids. Subsequently, we tested their utility in delivering siRNAs for silencing in cell culture. While we were able to vary binding over a wide range from 13 µg/mL to over 10,000 µg/mL and successfully delivered siRNAs to cells, silencing was not achieved using any of the polymers. Possible explanations for the lack of silencing activity include a lack of release of siRNAs from the endosome or quick recycling of siRNA complexes out of the cell or detachment and endocytosis of the dye without siRNA. AB.2 Methods and materials Polymer preparation: A Schlenk flask was used to dissolve 40-60 mg of poly(ethylene glycol)- co-poly(propargyl glycolide) into DMF. The PPGL was alkyne functionalized as described165. The desired mole fractions of amine and alkyl sidechains (azide-functionalized) and 24 mole percent sodium ascorbate were then added. The flask was degassed 3-4 times using a freeze- pump-thaw cycle and backfilled with nitrogen gas. A 0.1 M CuCl2•2H2O solution was dripped in and stirred overnight at room temperature. The resulting solution was filtered to remove solids. Copper ions were removed by adding Amberlite IRC-748 ion exchange resin beads before filtering again. The DMF was removed in vacuo. Remaining polymer was dissolved in a 3:1 water/acetone mixture and dialyzed in a 12-14 kDa MWCO dialysis bag for 2-3 days. The dialysis solvents were removed in vacuo. Structures of sidechains attached to PPGL backbones depicted in Figure S1. Polymer Binding Gels: Polymers at varying concentrations were dissolved in 18 MΩ water containing 200 nM FAM-tagged dsDNA and incubated for 15 minutes. dsDNA is less susceptible to degradation during the gel electrophoresis process than siRNA and we have previously shown the use of dsDNA to test polymer binding to be a suitable, cost effective replacement for siRNA while exhibiting similar binding to the synthesized polymers165. A 0.8% agarose gel containing two rows of 13 lanes, used for two, simultaneous independent experiments, was used to separate bound from free dsDNA. The outermost lanes contained no polymer and were used as control intensities for 100% unbound dsDNA. The fraction of bound dsDNA was calculated by dividing the intensity of the free DNA signal in each experimental lane by the intensity of the free DNA signal in the control lane and subtracting the resulting ratio from 1. Calculated fractional binding values from all independent experiments were plotted together. Data were then fit using a modified Hill Equation166 (Equation 1), where K is the 58 binding coefficient at 15 min, n is a fitted parameter that suggests the presence of cooperative binding, and [P] is the concentration of polymer. 𝐹𝑟𝑎𝑐𝑡𝑖𝑜𝑛𝐵𝑜𝑢𝑛𝑑 = [𝑃]𝑛 𝐾𝑛+[𝑃]𝑛 (1) The term binding coefficient is used in place of the dissociation constant in the Hill equation because dissociation constant indicates the polymer-siRNA complexes reached equilibrium. As we only used an incubation period of 15 minutes before testing binding, we cannot guarantee equilibrium was reached in all cases. Cell Culture: Cells were maintained as described165. Briefly, NCI-H1299 (human lung carcinoma) cells expressing enhanced green fluorescent protein (EGFP) with 2 hour half-life were obtained from Dr. Jørgen Kjems, University of Aarhus, Denmark168. Cells were passaged approximately weekly by trypsinization. Cells were grown in Dulbecco’s Modified Eagle’s Medium High Glucose supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 1% Geneticin and incubated at 37°C, 5% CO2, 100% relative humidity. EGFP Silencing and Cytotoxicity: NCI-H1299-EGFP cells were plated at 20,000 cells per well in 0.1 mL of complete media without antibiotics in a 96-well black side, clear bottom plate and incubated for 24 hours. 24 µL of Opti-Mem was used to complex siRNA and polymer for final concentrations of up to 100 nM siRNA and sufficient polymer to bind the siRNA according to the results of the binding gels. These were mixed and incubated at room temperature for 30 min prior to transfection. LPEI- and LF2K-siRNA complexes were prepared to confirm the activity of the siRNA. Transfection solution was left on the cells for 24 hours at 37°C, 5% CO2, and 100% humidity. Before measuring EGFP fluorescence using a Gemini EM fluorescent plate reader (480 nm excitation/525 emission), cells were washed twice with DPBS. Cytotoxicity was determined by comparing average EGFP signal from 3 wells of a 96-well plate of NCI-H1299- EGFP cells treated with only transfection media with average EGFP signal from 3 wells treated with transfection media containing polymer. Silencing was measured by comparing average EGFP signal from 3 wells of a 96-well plate of NCI-H1299-EGFP cells treated with transfection media containing polymer-siRNA complexes with average EGFP signal from 3 wells treated with transfection media containing only polymer (no siRNA). Microscopy: Transfection media was removed, and cells were rinsed before being placed in 0.5 mL of Leibovitz Medium (L-15) media for imaging. An Olympus FluoView 1000 Inverted IX81 microscope with 40x oil objective was used for confocal laser scanning microscopy. EGFP was 59 excited using a 488 nm multi-line Argon laser and detected through a BA505-52 nm emission filter. Dy547-tagged siRNA was excited with a 543 nm HeNe laser and detected through a BA560-IF nm filter. Imaging was done sequentially using a Kalman average of 2 at the focal plane with the highest intensity EGFP signal. 60 AB.3 Results Figure A3 Synthesis and Polymer Schematic. A.) Schematic showing the shape of the polymers. The amine-containing moieties and alkyl chains extend radially from the block co- polymer backbone in a brush-like fashion. B.) Alkyne-azide reaction used to add functional sidechains to PEG-PPGL block co-polymers. C) The amine-containing moieties and alkyl chains used in this study with associated abbreviation and molecular weight in parentheses where R indicates a conection to the PPGL backbone, R1 indicates a connection to the alkyl group on the PPGL backbone, and R2 indicates a connection to one of the side chains. Additional structures can be found in the supplemental information. 61 By using click chemistry to modify PEG-PPGL backbone polymers, we were able to rapidly synthesize 59 unique cationic biodegradable polymers to be complexed with siRNA for transfection of NCI-H1299 cells. The polymers consisted of four structural elements (Figure A3A). PEG and PGL were linked and served as the polymer backbone. “Click” chemistry (Figure A3B) was used to functionalize the PGL monomers with amine-containing or alkyl sidechains (Figure A3C). PEG length was varied, with each polymer having a PEG segment length of 0, 8, or 115 repeat units (Table A1). PGL length was also varied from 5 to 150 repeat units. The amine functional groups added to the PGL backbone were either primary, secondary, tertiary or a combination of primary and secondary amines (Figure A3C). Alkyl sidechains ranged from 4 to 18 carbons in length. TABLE A1: Polymers tested with name, structure, binding constant, K, and potential cooperativity, n Name Structure Hill Equation K (µg/mL) +/- n +/- P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 PEG8PPGL45(A1280C420) PEG115PPGL45(A1280C420) PEG8PPGL45(A1280C1020) PEG115PPGL45(A1280C1020) PEG8PPGL45(A1280C1620) PEG115PPGL45(A1280C1620) PEG115PPGL45(A180C1020) PEG115PPGL45(A280C1020) PEG115PPGL45(A380C1020) PEG115PPGL45(A140A240C1020) PEG8PPGL45(A1280C8 820) PEG8PPGL45(A12280C1020) PEG8PPGL56[A12100] PEG8PPGL45(A12250C1EO4C630X20) PPGL5SSPPGL5[A12100] PEG8PPGL45[A1250LC50] PEG8PPGL45[A12100] PEG8PPGL45(A1290C1010) PEG8PPGL10(A1280C1020) PEG8PPGL45[A1280C1EO410ChS10] 135.40 365.40 31.50 236.00 13.30 63.90 > 10,000 > 10,000 > 10,000 > 10,000 52.80 13.00 18.44 28.30 46.10 50.97 52.34 52.50 53.20 67.97 62 9.40 35.70 3.05 17.40 0.95 1.36 1.01 0.90 1.43 0.08 0.09 0.16 1.65 0.21 1.94 0.28 2.62 0.17 4.55 0.96 1.25 3.89 2.74 2.95 2.88 5.01 3.73 8.07 1.65 0.24 1.19 0.12 0.74 0.04 0.95 0.11 0.85 0.04 1.60 0.14 1.09 0.07 1.43 0.19 1.38 0.13 1.08 0.12 TABLE A1 (cont’d) P21 P22 P23 P24 P25 P26 P27 P28 P29 P30 P31 P32 P33 P34 P35 P36 P37 P38 P39 P40 P41 P42 P43 P44 P45 P46 P47 P48 P49 P50 P51 P52 P53 P54 P55 PEG8PPGL35(A1270C1030) PPGL45(A1280C1EO3C1220) PEG8PPGL45(A1270C1030) PEG8PPGL45(A1290X10) PEG8PPGL45(A1260C1040) PEG8PPGL45(A1270C1020CL10) PPGL5SSPPGL5[A1280] PEG115PPGL45(A1280C1620) PEG8PPGL45(A1280C1EO4C620) PEG8PPGL45[A1280C1EO410ChR10] PEG8PPGL45(A1270AP10C1020) PEG8PPGL45(A1275C1020G5) PEG8PPGL45[A1290Ch10] PEG115PPGL45(A1280X20) PEG115PPGL45(A1290C1010) 70.50 71.00 82.60 85.90 118.50 141.80 150.97 154.20 173.20 180.41 203.30 242.40 254.63 260.80 275.30 PPGL30PEG115PPGL30[A1260C1EO4C640) 306.53 PEG8PPGL45(A1280X20) PPGL50[A1236C1EO4C664) PPGL150(A1270C1030) PEG8PPGL45[A1280E01C220] PPGL50[A12218C1EO4C650X32) PEG115PPGL45(A1290X10) PEG115PPGL45(A1270C1030) PEG115PPGL45(A1280X20) PPGL30PEG115PPGL30(A1250X50) PEG8PPGL45(A1270X30) PEG115PPGL45(A1270X30) 484.00 710.18 776.00 916.63 1184.20 1198.00 1224.00 1255.00 1452.00 2248.00 2639.00 3.34 4.67 4.99 5.03 15.80 28.10 14.33 14.40 15.10 10.19 14.70 20.00 23.20 21.90 14.70 39.28 45.70 63.45 32.90 89.26 158.86 104.00 44.50 2.42 0.35 2.04 0.30 1.47 0.15 1.27 0.11 1.18 0.20 0.60 0.08 0.93 0.07 1.33 0.18 1.06 0.09 0.82 1.42 0.03 0.16 1.68 0.21 0.95 0.07 0.97 0.09 1.21 0.09 0.86 0.09 0.85 0.08 0.73 0.05 1.71 0.73 1.33 1.27 2.96 0.18 0.06 0.23 0.17 0.38 166.70 0.73 0.09 135.00 0.76 0.06 270.00 0.55 0.04 207.00 0.79 0.07 PEG115PPGL45[A1290Ch10] 3238.44 1048.16 0.40 0.06 539.00 0.65 0.07 542.36 0.86 0.09 PEG115PPGL45(A1250X50) PEG8PPGL45(A1218C1EO4C650X22) PEG115PPGL45(A1260C1040) PEG115PPGL45(A1250C1050) PPGL55(C1EO480C1020) PEG8PPGL45(A280C1020) PEG8PPGL45(A380C1020) 4081.00 4297.30 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000 63 TABLE A1 (cont’d) PEG8PPGL45(A1270X30) PEG115PPGL45(A170C1030) PEG8PPGL67[A1280C2EO2C210] PEG8PPGL45[A1250CA50] P56 P57 P58 P59 PLL LPEI25 LPEI2.5 LF2K BPEI > 10,000 > 10,000 >10,000 >10,000 0.36 3.00 3.17 4.87 37.80 0.06 0.41 0.56 0.73 1.84 0.92 0.16 1.80 0.43 1.07 0.16 0.94 0.12 2.64 0.27 Polymer-nucleic acid complex formation was measured using a gel-shift assay (Figure A4). Binding coefficients varied from 13 µg/mL to greater than 10,000 µg/mL. We tested binding for a number of commercially available siRNA delivery systems that have demonstrated silencing (PLL, LPEI, LF2K and BPEI) and found these to have binding coefficient between 0.358 µg/mL and 37.8 µg/mL. In addition to binding coefficient, K, Table A1 lists n, a fit parameter that may be related to cooperativity of binding. An n value greater than one suggests that binding of siRNA to a polymer influences more binding of siRNA to the polymer169. We observed no trends relating n to structural elements. From our binding results, we showed that more PEG monomers per polymer decreases binding between dsDNA and the polymer (Table A1; compare P1 to P2, P3 to P4, and P5 to P6). Polymers with sidechains containing primary amines (P7), secondary amines (P8), tertiary amines (P9), or a combination of primary and secondary amines on separate sidechains (P10) were unable to achieve binding coefficients less than 10,000 µg/mL; however, sidechains containing both primary and secondary amines on the same branch (P4 and polymers containing A12, A112, or A122) have measurable binding coefficients less than 10,000 µg/mL. Additionally, our results demonstrated that the longer the alkyl sidechains, the lower the binding coefficient (stronger binding interaction) (compare P1 (4 carbon alkyl chain), P10 (8 carbon alkyl chain), P3 (10 carbon alkyl chain), and P5 (16 carbon alkyl chain)). 64 Figure A4 Gel-shift Analyses. A.) DNA binding gel for P41 (top) depicting the change in intensity of free DNA as the amount of polymer increases from left to right. B.) The graph shows results from four independent binding gels of P41 with the modified Hill Equation fit (K = 28.3 µg/mL, n = 0.95). Confocal laser scanning microscopy was used to test the ability of our polymers to deliver siRNAs. A representative example, polymer P3, was able to deliver siRNA to NCI- H1299 cells (Figure A5); however, despite accumulation at levels comparable to those achieved with LF2K (Figure A5, compare C and I), the siRNAs did not reduce the expression of EGFP. Indeed, none of the synthesized polymers led to a statistically significant decrease in EGFP, regardless of the strength of the binding interaction between the siRNA and polymer. 65 Figure A5 siRNA Accumulation and Silencing. Shown are the delivery of Dy547-siRNA and silencing of EGFP in NCI-H1299-EGFP cells using LF2K, all scale bars are 20 μm. (A.-C.) at 1 µg/mL and 100 nM siRNA for 24 hours where A is the green channel, C is the red channel, and B is the overlay of A and C. Panels D.)-F.) show untreated, control NCI-H1299-EGFP cells. Panels G.)-I.) show NCI-H1299-EGFP cells after 24 hours in transfection solution containing 50 µg/mL of P3 and 100 nM Dy547-siRNA. Average EGFP intensity in the cells receiving siRNA via P3 did not change compared to the control cells. The cytotoxicity of the polymers was tested by comparing EGFP expression in cells treated with the polymer in the absence of siRNA and untreated control EGFP cells; no statistically significant toxicity was seen for any polymer at the concentration at which it was tested. 66 AB.4 Discussion For siRNAs to accumulate in cells, delivery vehicles must bind them and protect them from nuclease degradation during delivery. We have shown increasing PEGylation of polymers impairs their binding with dsDNA/siRNA, while combining primary and secondary amines, increasing the number of amines, and increasing the length of alkyl sidechains improves binding. It has been shown previously that PEGylation decreases binding affinity of nucleic acid/polymer complexes via charge shielding158. Interestingly, isolated primary, secondary and tertiary amines all have pKa values above physiological pH; therefore, differences in pKa among the amine classes does not account for the differences in binding. The inclusion of two nearby amines in the A12 sidechain potentially allowed for multiple binding sites between the siRNA and polymer, stabilizing the electrostatic interaction169. Binding of a polymer to siRNAs can be sterically limited170. Incorporation of alkyl chains can produce nanoparticles with a hydrophobic core with cationic arms extending into the hydrophilic solution171. These arms are then more spatially available to conform to the siRNA, improving binding. Amphiphilic polymer nanoparticles can also form micelle-like conformations surrounding siRNAs157. The inclusion of a double bond in P10 impairs binding compared to P5 with the same number of carbons, indicating that hydrophobic packing may be crucial for the enhanced binding observed with longer alkyl sidechains. This further supports our hypothesis that these polymers form a micellar structure. It is notable that many of our polymers had binding interactions of similar strength to those of the commercial vehicles. Moreover, our polymers successfully delivered siRNAs into cells at comparable concentrations. Nonetheless, none of the polymers tested was able to lead to silencing of EGFP. One possible explanation for the lack of efficacy of the polymers is the absence of secondary and tertiary amines, which can buffer solutions as low as pH = 3 and therefore may lead to the proton sponge effect and endosomal rupturing102,172. Amphiphilic siRNA delivery vehicles have also been shown to disrupt endosomal membrane stability and thereby allow siRNAs to enter the cytoplasm174. In order to disrupt the membrane, delivery vehicles must be near neutral charge at endosomal pH to avoid electrostatic exclusion from the membrane173. However, our polymers are likely too positively charged to interact with the membrane in this manner. Another possible explanation for the lack of effective delivery by our polymers is that endosomes containing the polymers are quickly tagged for recycling or 67 degradation, not allowing for sufficient time to release their cargo174. We have previously shown that accumulation of siRNAs in cells does not directly correlate to the degree of silencing achieved13. Thus, it is likely that the siRNAs delivered by our polymers may have been endocytosed but precluded from reaching the cytoplasm and initiating RNAi. Lastly, it is possible that the Dy547 fluorophore detaches from the siRNA prior to cellular internalization and enters either alone or with the PPGL nanoparticle. The development of effective and safe delivery systems remains a limiting step in the clinical application of siRNAs. Our polymers offer a unique insight into the vehicle characteristics that influence downstream siRNA activity because of their facile modification via click chemistry. The benefits of having multiple classes of amines incorporated into one polymer remain unclear but could stem from better binding affinity or an increased buffering capacity. The incorporation of higher molecular weight alkyl sidechains requires future investigation to find the length of alkyl chain that optimizes binding with siRNA without impairing solubility or increasing size. Finally, it will be important to understand how alkyl sidechain length impacts stability, conformation, and, ultimately, delivery effectiveness of polymeric vehicles for siRNAs. 68 APPENDIX C: LIPOPOLYSACCHARIDES ALTER ENDOCYTOSIS PATTERNS OF CATIONIC-LIPID DELIVERED SIRNA Bacterial infections are common in hospitals and in patients with reduced immune function. Lipopolysaccharides are large sugar molecules found in the bacterial membrane of Gram-negative bacteria that initiate an immune response from stimulation of the Toll-like Receptor 4 (TLR4). TLR4 activation causes a signaling cascade that increases the expression of cytokines and pro-inflammatory signaling. LPS exposure has been linked to ER stress. Our lab has previously developed a method of determining which endocytosis mechanisms are utilized by siRNA delivery vehicles. In short, prior to and during transfections, cells are incubated in media containing none or one of six inhibitors of proteins known to impact endocytosis. If you combine the knowledge of which proteins each inhibitor impacts and which proteins are necessary for the endocytic pathway, a logic matrix can be derived to figure out which mechanism of endocytosis is utilized by the siRNA-complexes based off how the small molecule inhibitors impact gene silencing and siRNA accumulation. Our preliminary results (Figure A6) indicate that dynamin inhibition by dynasore and actin polymerization inhibition by cytochalasin D have altered impact on LPS induced cells compared to control cells. Actin polymerization is known to be altered by the UPR and dynamin has been shown to interact with multiple chaperone proteins. These results lead us to believe that more than just TM induced ER stress will impact the delivery of siRNAs. 69 Figure A6 Comparison of endocytosis inhibitors impact on siRNA accumulation and silencing with LPS induction. A) Results of endocytosis inhibitors on EGFP silencing and siRNA accumulation in control NCI-H1299 cells13. MßCD and dynasore inhibit accumulation and silencing13. B) Results of endocytosis inhibitors on EGFP silencing and siRNA accumulation in LPS-induced NCI-H1299. Only MßCD inhibits uptake and silencing. 70 APPENDIX D: NUCLEIC ACID CARGO DOES NOT IMPACT LIPOFECTAMINE 2000 ENDOCYTOSIS PATTERNS Previous results have indicated that charge, charge density, size, shape, and nitrogen/phosphate ratios are important characteristics for understanding how siRNA-lipid complexes will be endocytosed by cells. Following that logic we hypothesized that DNA plasmids ( >2000 kb, circular) and siRNA (RNA, 21 bp, linear) would impact the endocytic mechanism used by LF2K. Using a plasmid encoding for EGFP, LF2K, and HEK293 cells we learned that the dynasore, cytochalasin D, and methyl-ß-cyclodextrin (MßCD) decreased both siRNA delivery and pDNA delivery and amiloride also impacted siRNA delivery (Figure A7). According to the logic matrix developed by Vocelle et al.13, both siRNA delivery and pDNA delivery are utilizing GRAF1-mediated endocytosis. Figure A7 Impact of endocytosis inhibitors on delivery and efficacy of siRNA and pDNA complexed with LF2K. A) Results of endocytosis inhibitors on EGFP silencing and siRNA accumulation in control HEK293 cells13. MßCD, dynasore, and cytochalasin D inhibit accumulation and silencing. Amiloride increases accumulation but reduces silencing. B) Impact of endocytosis inhibitors on EGFP expression from pDNA delivered by LF2K. MßCD, dynasore, and cytochalasin D inhibit expression. 71 APPENDIX E: ROLIPRAM IMPROVES PROTEIN PRODUCTION IN HEK293 CELLS TRANSFECTED WITH PDNA To further study the impact of ER stress and degradation on protein production initial experiments were performed using pDNA encoding Gluc, PEI, HEK293 cells, TM and RP. PEI causes nucleic acid endosomal escape through proton sponge effect, whereas LF2K in prior experiments causes endosomal escape through membrane fusion. HEK293 cells are used to produce proteins industrially; however, they require extra care when performing experiments because they are semi-adherent. When measuring the cell lysate all media and wash solutions removed from the cells must be collected and centrifuged at 300 rcf to collect cells that de- adhere. Our results using this model are consistent with our finding that RP improves protein production in both stressed and unstressed cells (Figure A8). Further optimization of PEI, pDNA and RP concentrations will need to be conducted to confirm the results. Figure A8 RP improves protein production in HEK293 cells. A) HeLa cells transfected with pRNA coding for Gluc using PEI and allowed to recover for 24 hours in control, TM-, RP-, or RP with TM-media (RPTM). A) Media accumulation of Gluc. B) Cell lysate Gluc signal. C) Secretion ratio of control cells and cells treated with TM alone, RP alone, and RP with TM. Slope p-values calculated by linear regression and compared to a zero slope. Error bars represent standard deviation, one biological replicate, three technical replicates. 72 APPENDIX F: METHODS AND MATERIALS FOR CHAPTER 2 Cell Culture: HeLa and MDA-MB-231 cells were purchased from ATCC. All cells were maintained in high glucose DMEM (Gibco 11965092) containing 10% FBS (R&D Systems S11550) and 50 mM HEPES (Fisher BP410-500) at 5% CO2 and 37C. Cells were passaged three times per week. MDA-MB-231 IRE1α knockout cells were generated using CRISPR-Cas9 knockout with plasmids gifted by Dr. Feng Zhang (Addgene plasmid # 48138 and # 62988) and Dr. Andrea Ventura (Addgene plasmid # 64073). Nucleic Acid Preparation: A plasmid for expressing secreted Gaussia princeps luciferase was generously provided by Dr. R. Alexander Wesselhoeft and Dr. Daniel G. Anderson. Plasmid was linearized using Xba1 restriction digest enzyme (NEB R0145S) per manufacturer’s instructions prior to in vitro transcription according to manufacturer’s instruction using T7 high yield in vitro transcription kit (Thermo K0441) and MEGAClear transcription clean-up kit (Thermo AM1908). Plasmid DNA coding for secreted Gaussia luciferase conjugated to cyan fluorescent protein was generously provided by Professor Andrew Dillin at UC Berkeley. Bacteria were grown in LB broth (Affymetrix 75852) containing 200 µg/mL ampicillin (Fisher BP1760-25). Plasmid DNA was isolated using a ThermoFisher Gene Jet Plasmid Miniprep (K0503). Transfections: All cells were plated in 0.5 mL of DMEM containing 80,000 cells/mL in a 24 well plate (Costar 3526). 24 hours later, culture DMEM was replaced with serum free DMEM 1 hour prior to transfection. Nucleic acids were mixed with LF2K (Invitrogen 52887) per manufacturer’s instructions. Final concentrations (once added to the cells) for RNA transfections were 160 ng/mL RNA and 200 ng/mL LF2K and for pDNA transfection final concentrations were 200 ng/mL and 1200 ng/mL LF2K. Transfection solutions were added to cells in serum free DMEM and allowed to incubate at 37C for 4 hours. After cells were washed with 1 mL culture DMEM, fresh culture DMEM containing stressors, inhibitors, or activators was added, and cells recovered for 24 hours before analysis. Concentrations of chemicals used were TM (T7765-5MG) – 5 µg/mL 3MA (SAE0107-10ML) – 5 mM Rolipram (Bio-Techne 0905)– 60 µM VR23 (AdipoGen 1624602-30-7) – 3 µM Luciferase Measurement: Luciferase measurements were performed using the Thermofisher Gaussia Luciferase Kit (Thermo 16161). 20 µL of recovery media was added a well in a Corning 73 3904 black wall 96 well plate. The remaining media was aspirated, and cells were washed with 1 mL of DPBS. Cells were lysed in 150 µL of 2X cell lysis solution mixed with 150 µL of DPBS. Plates containing lysis solution were placed on an orbital shaker at 65 rpm for 1 hour. 50 µL of lysis solution was added to the 96-well plate. For each test, 0.5 µL of 100X luciferase reagent solution was mixed with 49.5 µL of DPBS. The plate was loaded into a BioTek Synergy H1 Microplate Reader and shaken for 25 seconds, then raw luminescence was collected for 3 seconds in each well. Microscopy: Cells were plated Millicell EZ Slides (PEZGS0816) at 90000 cells/mL in 500 µL of culture DMEM. Cells were treated with the same conditions described above. After treatment, cells were fixed by adding 500 µL of 3:1 Methanol (Sigma 34860-4L-R) : Glacial Acetic acid (Sigma AX00073-75). After 5 minutes, the culture media/fixative mixture was removed and replaced with 500 µL of fresh 3:1 Methanol:Glacial Acetic Acid for 10 min. Cells were rinsed for 5 min in DI water twice. LC3B staining was for 8 hours at 1:500 LC3B-DyLight550 antibody (Novus NB100-2220R) : DPBS. Fixed and stained cells were rinsed three times with fresh DPBS before 40 µL of Prolong Antifade Glass Mount (Invitrogen P36984) was added to each well surface and a glass coverslip was applied. The mount was allowed to cure overnight before imaging on the Leica Stellaris 5 Confocal Laser Scanning microscope at 100x. LC3B-DyLight was imaged at Ex:562nm/Em 572-700nm. CFP was imaged at Ex 488nm/Em 500-550nm. Statistics: Statistics were calculated first using two-way ANOVA to determine significance followed by Student’s t-test to calculate pair-wise p-values. Measurements were made using 3 plates of cells with 3 replicates using independent solutions for each plate. 74 APPENDIX G: METHODS AND MATERIALS FOR CHAPTER 3 Cell culture: HeLa-EGFP cells were provided by Dr. Manfred Gossen175. Cells were passaged via trypsinization three times per week and maintained in antibiotic-free Dulbecco's Modified Eagle Medium (DMEM, Gibco 119625), 10mM HEPES, and 10% FBS within an incubator at 37°C and 5% CO2. EGFP Silencing during ER Stress: HeLa-EGFP cells were plated in 24-well plates (Corning 3526) at 90,000 cells/mL and 0.5 mL per well. 24 hours after plating, culture media was replaced with serum-free DMEM containing increasing levels of TM (MP Biomedicals, 11089-65-9). For one hour cells were allowed to incubate in the serum-free TM media while Cy5-siRNA/LF2K complexes were formed per manufacturer’s instruction. Complexes were made to achieve a final concentration of 1.75 µg/mL LF2K (Invitrogen, 11668019) and 150 nM siRNA (Horizon Discovery, Sequence: GGCUACGUCCAGGAGCGCAUU) in 50 µL of Opti-MEM (Gibco, 31985062) and 500 µL of DMEM. Cells were incubated with the transfection solution for 4 hours and then washed with 50 µg/mL Heparin Sulfate (AMSBIO, GAG-HS01) in 1 mL DPBS (Gibco, 14190144) for 10 min. Upon removal of the Heparin Sulfate, 0.5 mL of fresh culture media was added to cells and they were allowed to incubate for 24 hours before being fixed for flow cytometry via the protocol in this methods section. To ensure all siRNAs that accumulate entered the cells via endocytosis and not by delayed membrane fusion from siRNA that was stuck to cell membranes and the well plate, immediately following transfection, cells were washed with the negatively charged heparin sulfate. 24 hours post heparin sulfate washing cells were fixed and taken to flow cytometry. In this thesis we define silencing to mean the fractional reduction of EGFP signal in cells treated with siRNA compared to cells treated with all other media and transfection conditions with 0 nM siRNA (equation 1). EGFP Silencing under Autophagy Inhibition: HeLa-EGFP cells were plated in 24-well plates at 90,000 cells/mL and 0.5 mL per well. 24 hours after plating, culture media was replaced serum free DMEM containing 5 mM 3MA (Sigma, SAE0107). After two hours of 3MA incubation, TM was added to a final concentration of 5 µg/mL. Cells were left to incubate in 3MA/TM serum-free media for 1 hour while Cy5-siRNA/LF2K complexes were formed per manufacturer’s instruction. Complexes were made to achieve a final concentration of 1.75 µg/mL LF2K and 150 nM siRNA in 50 µL of Opti-mem and 500 µL of DMEM. Cells were 75 incubated with the transfection solution for 4 hours and then washed with 50 µg/mL Heparin Sulfate in 1 mL DPBS for 10 min. Upon removal of the Heparin Sulfate, 0.5 mL of fresh culture media with or without 3MA was added to cells, and they were allowed to incubate for 24, hours before being fixed for flow cytometry via the protocol in this methods section. EGFP Silencing under Endosome Maturation Inhibition: 10 µg of Bafilomycin A1 (MSPP- TLRLBAF1, VWR) was dissolved in 500 µL of DMSO. HeLa-EGFP cells were plated in 24- well plates at 90,000 cells/mL and 0.5 mL per well. 24 hours after plating, culture media was replaced serum free DMEM containing increasing levels of the BAFA1 solution. Cells were left to incubate in BAFA1 serum-free media for 1 hour while Cy5-siRNA/LF2K complexes were formed per manufacturer’s instruction. Complexes were made to achieve a final concentration of 1.75 µg/mL LF2K and 150 nM siRNA in 50 µL of Opti-mem and 500 µL of DMEM. Cells were incubated with the transfection solution for 4 hours and then washed with 50 µg/mL Heparin Sulfate in 1 mL DPBS for 10 min. Upon removal of the Heparin Sulfate, 0.5 mL of fresh culture media with or without 3MA was added to cells, and they were allowed to incubate for 24, hours before being fixed for flow cytometry via the protocol in this methods section. Cell Fixation and Flow Cytometry: Cells ready for flow cytometry were washed with 300 µL of Versene solution (Gibco, 15040066) before being removed from the plate with 200 µL of trypsin-EDTA solution (Gibco, 25200072). After 3 min of incubation with trypsin-EDTA at 37°C, 2 mL of culture media was added to inactivate the trypsin. This solution was moved to a centrifuge tube and centrifuged at 220 rcf for 3.5 min. The supernatant was removed. Cells stained with Zombie Violet LIVE/DEAD (BioLegend) were stained for 15 min at room temperature in 100 µL at a ratio of 1:1000 Zombie Violet to DPBS. All cells were then washed with 5 mL of DPBS. Cells were then centrifuged for 3.5 min at 220 rcf before the DPBS was removed and replaced with DPBS containing 2% paraformaldehyde (Sigma, 158127) at 37°C. Cells were then incubated in the fixative solution for 20 minutes before being centrifuged and rinsed twice with fresh DPBS. Finally, the cells were resuspended in 250 µL of sorting solution and ran on an Attune CyPix flow cytometer. The flow cytometer measured geometric means of Zombie Violet (ex. 405 nm / em. 423 nm), EGFP (ex. 489 nm / em. 511 nm) and Cy5 (ex. 651 nm / em. 670 nm) that was tagged to the siRNA. Statistics: Unless otherwise stated, replicates were produced by plating cells into 3 different plates with 3 replicates on each plate. Each plate received individually prepared solutions of TM, 76 siRNA/LF2K and recovery media. Statistical analysis was first done by two-way ANOVA and followed by student t-test to calculate pair-wise p-values. 77