抑制調控果蠅5-磷酸核醣異構酶能減輕Tau引起的毒性

全文

(1)國立臺灣師範大學生命科學系博士論文. 抑制調控果蠅5-磷酸核醣異構酶能減輕 Tau引起的毒性 Down-regulation of ribose-5-phosphate isomerase attenuates Tau mediated toxicity in Drosophila. 研究生：蕭欣杰 Hsin-Chieh Shiau 指導教授：蘇銘燦博士 Ming-Tsan Su 中華民國 104 年 2 月.

(2) Acknowledgement 此博士論文的完成，真心地謝謝所有給予我幫助及支持的貴人，讓我得以邁入人生的另一個階段。漫長的研究生生活，真的非常感謝蘇銘燦老師的包容及學業上的指導。在低潮時，李桂楨老師及謝秀梅老師的鼓勵及幫助給了我力量。謝謝汪宏達老師及廖國楨老師在論文口試的指導及寶貴的意見。我也要感謝協助、陪伴及鼓勵我的人，包含麗卿、芳足、國華、柏安、政光、瑞宏、玄原、怡辰、承岳、永祥、聖文、煥雯、俊彥、敦傑、俞鈞、勝安、郁芬、進勝、建文、昱嘉及科辦的漢英、晉怡、淑美、友信、春美等人。最後，我要感謝的父母和家人，謝謝你們的關心、支持、體諒及包容，讓我能專心於研究，順利的完成此論文。.

(3) Index Abstract (Chinese)…………………...………………………………....01 Abstract (English).………………………...……………………………02 Introduction…...………….……………………………………………..03 Materials and methods...……………………………………….………..08 Fly strains and culture conditions……………………..……...........08 Bristle quantification and eye morphology assessment……………09 Lifespan assays…………………………………………………….10 Climbing ability assay……………………………………………..10 Immunoblotting……………………………………………………11 NADPH quantitation Assay..………………………………...........11 Reduced glutathione Assay…………...……………………….…..12 Results…………………………..…………………………………..…..13 A targeted screen for modifiers of tauopathy in Drosophila…..…..13 Down regulation of rpi extends the lifespan and improves the motor function of tauopathy flies………………………………..……………..16 rpi downregulation does not decrease phosphor-Tau species……...17 Reduced rpi increases cellular NADPH and reduced form of glutathione in tauopathy flies………………………….………………..18 Overexpression of hTTLL1 attenuates Tau toxicity and extends the lifespan of tauopathy flies………………………………………………19 Discussion……………………………..………………………….…….22 Appendix……………………………………………..…………………26 Abstract (Chinese)...…………………………………….………………28 Abstract (English)…………………………………….…………………29 I.

(4) Introduction……………………………………………………………..30 Materials and Methods………………….………………………………35 Dual luciferase reporter constructs………………………………..35 Cell culture, transfections, and luciferase assay…………………..36 Results………………………………………………………………….37 IRES activity of ATXN8OS RNA..………………………………. 37 Discussion………………………………………………………………39 References..……………………………………………………..………40 Figures and Tables………………………………………………..……..46 Figure 1. Genetic modifier screening of Tau induced toxicity in Drosophila eye and notal bristle of 1-day-old adult flies...…..46 Figure 2. SEM of eye micrography and notum used in the genetic modifier screens……………………………………..………..49 Figure 3. Downregulation of rpi extends the lifespan and improves the motor function of tauopathy flies…………….........................50 Figure 4. The knockdown of rpi does not alter Tau phosphorylation ….52 Figure 5. Neuronal downregulation of rpi increases NADPH and the reduced form of glutathione levels in the tauopathy flies……54 Figure 6. Reduced tubulin tyrosine ligase-like (TTLL) genes expressions enhance Tau induced toxicity………….……………………..…………55 Figure 7. Overexpression of hTTLL1 attenuates Tau toxicity and extends the lifespan of tauopathy flies…………….…………………..…………57 Figure 8. Schematic diagrams of the ATXN8OS cDNA and pRF dual luciferase reporter constructs………………………..……......59 Figure 9. IRES activity of the ATXN8OS transcript………….................60 II.

(5) Table 1. Genetic modifiers of Tau toxicity…..………………..………..62. III.

(6) 摘要 Tauopathy是一種多因子疾病，透過遺傳及分子方法已確認出許多致病途徑。為探究Tau引起的毒性及找出新的治療標的，我們利用眼睛及背毛模式去篩選與其他神經退化疾病有關的致病因子。在篩選研究中，靜默5-磷酸核醣異構酶(rpi)基因呈現出最佳的性狀改善結果。本研究指出在神經中抑制調控rpi基因表現可以延長tauopathy果蠅的壽命及改善運動功能。而在西方墨點實驗結果，顯示出tauopathy果蠅性狀改善不是經由改變Tau蛋白質的磷酸化。此外，進一步發現 tauopathy果蠅細胞中煙草醯胺腺嘌呤二核苷酸磷酸鹽(NADPH)及還原態穀胱甘肽(GSH)濃度會上升。我們也得到過量表現TTLL1可減緩 Tau的毒性及延長tauopathy果蠅的壽命。因此，在果蠅模式中，減少 rpi的表現及大量表現TTLL1可以抑制Tau的毒性。. 關鍵字：5-磷酸核醣異構酶 1.

(7) Abstract Tauopathy is a multifactorial disease in which many pathogenic pathways have been identified through genetics and molecular approaches. To better evaluate Tau induced toxicity and to find novel therapeutic targets, we screen components of pathogenic pathways that might implicated in various neurodegenerative diseases using both eye and notal bristle as model systems. We have chosen to study ribose-5-phosphate isomerase (rpi), because silencing of rpi exhibited the greatest phenotypic improvement among all other modifiers in our screening. Our data indicated that reduced neuronal expression of rpi extends the lifespan and improves the motor function of tauopathy flies. Results of immunoblotting experiments reveal that the phosphorylated Tau species were not significantly decreased when rpi were downregulated, suggesting that the beneficial effect of reduced rpi on tauopathy flies is not mediated through altering the phosphorylation of Tau. Additionally, reduced rpi increases cellular nicotinamide adenine dinucleotide phosphate (NADPH) and reduced form of glutathione (GSH) in tauopathy flies. We also found that overexpression of Tubulin tyrosine ligase-like 1 (TTLL1) attenuates Tau toxicity and extend the lifespan of tauopathy flies. Taken together our findings demonstrate that reduced rpi expression and TTLL1 overexpression suppress Tau toxicity in Drosophila.. Keywords: ribose-5-phosphate isomerase 2.

(8) Introduction Tauopathies are a class of at least 15 different neurodegenerative disorders of which share a common pathological hallmark, aberrant deposit of Tau aggregates. The causative role of Tau in neurodegeneration has been established because of the demonstration that dominant mutation in the Tau gene causes frontotemporal dementia and Parkinsonism linked to chromosome 17 (FTDP-17) [1-3]. Currently, more than 30 different mutations have been associated with Tau [4]. These mutations affect Tau in many aspects, including reduction in microtubule binding, increasing in the propensity to form abnormal aggregates, and alteration in the ratio of Tau isoforms [4]. Interestingly, in Alzheimer’s disease (AD), the most prevail tauopathy, Tau was found to be normal. Nevertheless, the appearance and anatomic distribution of pathological Tau aggregates correlates well with neuronal loss, cognitive dysfunction, duration and severity in AD [5,6], suggesting that wild-type Tau may directly contribute to the pathogenesis of tauopathies. Tau protein is a small microtubule-associated protein (MAP) which interacts with tubulin to stabilize microtubules through two controlling ways (isoforms and phosphorylation) in neurons of the central nervous system. There are six molecular isoforms in human brain through alternative splicing of microtubule-associated protein Tau (MAPT) gene located on chromosome 17q21. According to the inserts of amino acid (exon 2 and exon 3) at the N-terminal part (N) and three or four microtubule binging repeats (exon 10) at C-terminal part (R), the six isoform of Tau are 0N3R (352 amino acid, Tau352), 1N3R (381 amino 3.

(9) acid, Tau381), 2N3R (410 amino acid, Tau410), 0N4R (383 amino acid, Tau383), 1N4R (412 amino acid, Tau412), and 2N4R (441 amino acid, Tau441) [7]. A group of serine and threonine phosphorylation sites on Tau protein that are highly associated with neurodegeneration are designated as SP/TP sites [8]. Hyperphosphorylation of Tau protein can result in self-assembly of Tau into tangles of paired helical filaments (PHFs) and straight filaments [9]. In AD brains, all of the six Tau isoforms were hyperphosphorylated and aggregated into paired helical filaments. It was demonstrated that phosphorylation of Tau is regulated by a group of kinases and phosphatases [10]. The imbalance of kinases and phosphatases has shown to contribute the hyperphosophorylation of Tau, thereby enhanced Tau toxicity. Polyglutamylation of tubulin can cause conformational shifts to regulate the binding affinity of Tau protein through the growth of polyglutamyl chain from one to six glutamyl units [11]. Tubulin post-translational modifications include detyrosination, delta2-tubulin generation, acetylation, glutamylation, and glycylation which occur on the C-terminal domains of tubulin. The C-terminus of the tubulin is the major binding sites of microtubule-based molecular motors (kinesins and dynesins) and microtubule-associated proteins (MAPs). The glutamylation and glycylation enzymes are known to be members of the tubulin tyrosine ligase-like (TTLL) family with a tubulin-tyrosine ligase (TTL) homology domain which catalyze ligations of different amino acids to tubulins. The TTL enzyme can catalyze the ATP-dependent 4.

(10) post-translational addition of a tyrosine to the C-terminal end of detryosinated α-tubulin [12]. The mammalian TTLL family includes 13 TTLL proteins which have identified 9 glutamylases (TTLL1, TTLL2, TTLL4, TTLL5, TTLL6, TTLL7, TTLL9, TTLL11, and TTLL13) and 3 glycylases (TTLL3, TTLL8, and TTLL10) by sequence analysis and enzymatic characterization [13,14]. Neuronal tubulin polyglutamylase in mouse brain is a multimeric protein complex (360 kD) composed of five polyglutamylase subunits (PGs) which are PGs1, PGs2, PGs3, PGs4, and PGs5 [15-17]. PGs3 is the ortholog of the human tubulin tyrosine ligase-like 1 (TTLL1) protein and maybe the catalytic subunit of neural tubulin polyglutamylase. The TTLL1 preferentially polyglutamylates α-tubulin and has a side chain elongating activity. Since oxidative insults is a critical pathogenic factor of tauopathy and GSH is the major brain antioxidant [18,19], the increasing in nicotinamide adenine dinucleotide phosphate (NADPH) and reduced glutathione (GSH) contents are likely to protect neurons from Tau toxicity. The reduced neuronal expression of ribose-5-phosphate isomerase (rpi) exhibits increased resistance to oxidative stress, extends lifespan, and attenuates polyglutamine neurotoxicity by means of increasing NADPH level through elevated G6PD activity [20]. Ribose-5-phosphate isomerase isomerizes ribulose 5-phosphate (Ru5P) to ribose 5-phosphate (R5P) in non-oxidative phase of pentose phosphate pathway (PPP). PPP is an important metabolic pathway that includes irreversible oxidative phase (generaion of NADPH) and reversible non-oxidative phase (synthesis of nucleotides and nucleic acids). The primary result of PPP is the 5.

(11) generation of NADPH, ribose-5-phosphate (R5P), and erythrose-4-phosphate (E4P). NADP+ is reduced to NADPH by using the energy from the conversion of glucose-6-phosphate to ribulose 5-phosphate through glucose 6-phosphate dehydrogenase (G6PD), 6-phosphogluconolactonase, and 6-phosphogluconate dehydrogenase in oxidative phase. G6PD is a rate-limiting enzyme to control NADPH levels by allosteric regulation of NADP+. In humans, the PPP occurs in the cytosol and generates approximately 60% of NADPH production. The major biological function of NADPH in the cell is to provide the redox (reduction and oxidation) reaction involved in protecting against reactive oxygen species (ROS) from mitochondrial oxidative phosphorylation by producing antioxidant molecules such as GSH and thioredoxin (TRX). NADPH is also involved in lipid synthesis, cholesterol synthesis, and fatty acid chain elongation. Models for human tauopathies in simple and genetically tractable organisms, such as Drosophila melanogaster, have provided valuable tools to dissect the molecular pathomechanisms underlying tauopathies [21-26]. A number of genetic modifier screens have been conducted to identify key molecular players, including kinases, phosphatases, cytoskeleton proteins, cytoskeleton motors, autophagic components, RNA associated proteins, and cell adhesion molecules, involved in the pathogenesis of tauopathies [24,27-32]. Importantly, some of these modifiers were also identified as AD risk factors in genome-wide association studies (GWAS), indicating that the Drosophila models. 6.

(12) faithfully recapitulated the pathogenesis of tauopathies, and findings from disease fly models may have relevant medical implications in human. Previously, we found that overexpression of human Tau in the Drosophila nota leaded to bristle loss [33]. The bristle loss phenotype has provided us a quantitative assay tool for assessing Tau toxicity. To elucidate the mechanisms of Tau neurotoxicity, we conducted modifier screen of tauopathy using the notal bristle system. My results suggest that constituents of the protein quality control system are major modulators of Tau toxicity. Moreover, novel modifiers, including the yu (dual-specificity A-kinase anchor protein) and ribose-5-phosphate isomerase (rpi), were also identified. Further validation demonstrated that rpi inhibition was neuroprotective. In sum, our studies underscore the important role of protein quality control in tauopathies and provide potential new targets toward therapeutics.. 7.

(13) Materials and methods Fly strains and culture conditions The following Gal4 driver lines and mutant alleles were obtained from the Bloomington Stock Center: Gmr-gal4 (retinal), Elav-gal4 (panneural), torΔP (Stock No. 7014), tork17004(Stock No. 11218), vps4KG1292(Stock No. 13421), vps4EP970(Stock No. 10141), atg5d04577(Stock No. 19206), UAS-HSPA1L(Stock No. 7454), UAS-Hsc70-4K71S(Stock No. 7453), tpr2KG08262(Stock No. 14711), tpr2EY21644(Stock No. 22491), mrjEY04743(Stock No. 15938), UAS-Hdj1, pros261(Stock No. 6182), smt304439(Stock No. 11378), yuKG02745(Stock No. 12879), yuEP1400(Stock No. 11236), and dbrEP9(Stock No. 8551). The following RNAi lines were obtained from VDRC：CG32234-Ri109628, CG11323-Ri108627, CG11201-Ri104449, CG1550-Ri106694, CG5987-Ri103773, and CG6931-Ri108070. Eq-gal4 was provided by Yi-Han Sun [34]. UAS-tauWT was kindly provided by Mel Feany [26]. UAS-Hsp27, UAS-rpi and UAS-rpiRNAi were provided by Horng-Dar Wang [20]. UAS-hTTLL1 was engineered by Yi-Chun Wang. All fly stocks and genetic crosses were maintained on standard cornmeal-yeast-agar medium at 25°C. Flies were age and sex-matched in assessing modification of brain, eye, and notum toxicity.. 8.

(14) Bristle quantification and eye morphology assessment For quantification of notal bristle, similar protocol was performed as described [33]. One-day-old flies were collected, frozen, and immobilized on glass slides using polyvinyl acetate resin adhesive (Nanpao Co. Taiwan) with their dorsal sides face upward. Serial focal plane images of notum bristles were acquired by a Zeiss Axiovert 35 inverted fluorescent microscope (Carl Zeiss) equipped with a 10X-Ph1 objective and a digital CoolSNAP Charge-Coupled Device Camera (CoolSnap 5.0, Photometrics). One completely focused image were created by Helicon Focus software (Helicon Soft Ltd.) and analyzed with Photoshop software (Adobe Systems Incorporated). Bristles were quantified by Northern Eclipse software (Empix Imaging Inc.). Three independent experiments with 10 flies (5 males and 5 females) in each experiment were performed. Student’s t-test was performed for statistical analysis. For assessment of external eye morphology, one-day-old flies were collected, frozen, and immobilized on glass slides using polyvinyl acetate resin adhesive (Nanpao Co. Taiwan). Serial focal plane images of eyes were acquired by a Zeiss Axiovert 35 inverted fluorescent microscope (Carl Zeiss) equipped with a 10X-Ph1 objective and a digital CoolSNAP Charge-Coupled Device Camera (CoolSnap 5.0, Photometrics). One completely focused image was created by Helicon Focus software (Helicon Soft Ltd.).. 9.

(15) Lifespan assays For lifespan assay, a group of 10 adult flies were cultured in a vial and maintained at 25°C. At least 100 flies were analyzed for each sex. Dead flies were scored everyday and transferred to a new vial in 3-day interval until all were dead. At least 200 flies were analyzed for each genotype. The log-rank test was used for survival analysis.. Climbing ability assay The novel climbing ability assay was used to assess the locomotor activity of flies due to the natural negative geotaxis [35].The climbing apparatus consist of a 30 cm long glass tube with a diameter of 1.5 cm and plugged with cotton. For each trial, 10 male flies were allowed to climb up in a graded glass tube which was divided into a series of five 2-cm scoring areas (scored 1-5) from the base in 10 seconds. A total of 10 trials per climbing session were scored and calculated the climbing index. The climbing index (CI) was computed as follows: CI =Σ(nm)/10, where n is the number of flies in a given scoring area, and m is the score for the given scoring area (1-5). A total of 100 male flies were tested for each genotype and assayed every five days. Dead flies were scored everyday and transferred to a new vial in 3-day interval until all were dead.. 10.

(16) Immunoblotting For immunoblotting assay, 30 heads (15 males and 15 females) from adult flies at one-day posteclosion were homogenized in 100 μl Laemmli buffer and denatured before resolution in a 10% SDS polyacrylamide gel. Proteins were blotted onto PVDF membrane (BIO-RAD) and the protein blot was blocked in Tris-buffered saline with 0.1% Tween-20 and 5% BSA. The primary antibodies used in the immunoblotting analysis were polyclonal anti-Tau (1: 50,000, DAKO), pS202 (1:1,000, AnaSpec Co.), pT212 (1:1,000, Invitrogen), pT231 (1:3000, Abcam), pS262 (1:3000, Abcam), pS396 (1:3,000, Invitrogen), and anti-β actin (GeneTex, 1:5000). The appropriate secondary antibody was used at a dilution of 1:20,000 (Jackson ImmunoResearch). Signal was detected by Immobilon Western Chemiluminescent HRP Substrate (Millipore) and captured using an imaging system (Fujifilm LAS-4000). Signal was quantitation by ImageJ (NIH). Three independent experiments were performed.. NADPH quantitation assay 50 heads (25 males and 25 females) from one-day-old flies were homogenized in 150 μl NADP/NADPH Extraction Buffer. Following centrifugation at 14,000 rpm for 5 minutes, the supernatant was quantified by the Bio-Rad protein assay. According to the protocol in NADP/NADPH quantitation kit (Cat. # K347-100, Bio Vision), the values of the samples and NADPH standard at O. D.450nm were measured in duplicate. Standard curve were used to calculate NADPH 11.

(17) concentration of each sample. The amount of proteins in the tissue homogenates was used to normalize the NADPH value in each Three independent experiments were performed in duplicate. Student’s t-test was used to calculate P value to determine statistical significance.. Reduced glutathione assay Fifty heads (25 males and 25 females) from one-day-old flies were homogenized in 150 µl 1× PBS containing 2mM EDTA. Following centrifugation at 14,000 rpm for 5 minutes, the supernatant was quantified by the Bio-Rad protein assay kit. According to the protocol in GSH-GloTM Glutathione Assay kit (Cat.#V6911, Promega), the luminescence of the samples and GSH standard were measured using a SpectraMaxL microplate reader (Molecular Devices). Standard curve used to calculate GSH concentration of each sample. The amount of proteins in the tissue homogenates was used to normalize the reduced GSH concentration in each sample. Three independent experiments were performed in duplicate. Student’s t-test was used to calculate P value to determine statistical significance.. 12.

(18) Results A targeted screen for modifiers of tauopathy in Drosophila We screened components of pathogenic pathways that might implicated in various neurodegenerative diseases using both eye and notal bristle as model systems to evaluate Tau induced toxicity. Both eye and notal bristle models have shown to be excellent models for assessing Tau induced toxicity and applicable for finding genetic modifiers for tauopathies [32,33]. As shown in Fig. 1, retinal expression of human Tau driven by GMR-Gal4 driver (GMR>Tau) causes moderately rough eye phenotype (Fig. 1A vs. 1B, and [33]). Similarly, notal bristles appeared normal in the control flies expressing GFP driven by Eq-gal4 driver (Eq>GFP, Fig. 1A’). Nevertheless, notal bristles were significantly reduced when human Tau was overexpressed (Eq>Tau, Fig. 1B’ and Fig. 1V). Before the genetic screen, all the candidate modifiers have been tested not able to induce rough eye or bristle loss phenotype by their own. The modifier screen was conducted by crossing the GMR>Tau or Eq>Tau flies to each individual lines and examining the progeny for enhancement or suppression of Tau induced rough eye or bristle loss phenotype (Table 1). Suppressor of Tau toxicity either restored rough eye to a more regular ommatidial organization or increased notal bristle. In contrast, enhancer of Tau toxicity either reduced the eye size and increased ommatidial fusion or reduced notal bristle. We first evaluated components of the autophagy pathway in our modifier screen, because of their neuroprotective roles in tauopathy (for recent reviews [36,37]). We found that both eye size and notal bristle 13.

(19) were significantly reduced in Tau/torΔP progenies (Fig. 1C and 1C’). Despite the rough eye phenotype was not enhanced in Tau/tork17004 flies, the bristle was decreased, we suggests that loss of target of rapamycin (tor) was an enhancer of tauopathy (Fig. 1D and 1D’). The role of vacuolar protein sorting 4 (vps4), whose function in autophagic vacuole fusion, was paradoxical because both gain- and loss-of-function alleles suppressed either rough eye or bristle loss phenotype (Fig. 1E, 1E’, 1F, 1F’ and 1V). atg5d04577 may function as a suppressor because the ommatidial structure was restored with better organization and notal bristle was increased significantly (Fig. 1G, 1G’ and 1V). One of the pathomechanisms of tauopathy is the pathological accumulation of the misfolded Tau aggregates, including paired helical filaments (PHFs), neurofibrillary tangles (NFTs) and other abnormal oligomers. Since molecular chaperones are the major components of the protein quality control system whose primary functions are to assist the correct folding of proteins and prevent the formation of misfolded protein aggregates, it was believed that molecular chaperones would modulate the pathogenesis of tauopathy [38-40]. To test this we have included an array of chaperones and co-chaperones, including Hsp70, Hsc70, tpr2, mrj, Hdj1 and Hsp27, in our modifier screen (Fig. 1H-1N and 1H’-1N’). Hsp70 exhibited no effect on modulating Tau toxicity in flies, because overexpression of either human Hspa1L, a homolog of Drosophila Hsp70, or a dominant negative Hsp70, Hsc70-4K71S, did not alter the rough eye and bristle loss phenotypes in flies (Fig. 1H-1I and 1H’-1I’). Two co-chaperone encoded genes, tpr2 and mrj, were enhancers, because 14.

(20) bristle loss phenotype was enhanced under either loss-of tpr2 or loss-of mrj function background despite the rough eye phenotype was not significant different (Fig. 1J-1L and 1J’-1L’). Overexpression of Hdj1, a human DnaJ homolog, suppressed rough eye and bristle loss phenotypes in tauopathy fly models (Fig. 1M and 1M’). Nevertheless, Hsp27, a small pro-survival heat shock protein, has no effect on suppressing Tau induced toxicity (Fig. 1N and 1N’). The ubiquitin–proteasome pathway is one of the major protein clearance routes in cells. It was reported that Tau protein containing PHFs sequestered and deactivated proteasome, suggesting that inhibition of proteasome activity was sufficient to induce neuronal degeneration in AD brain [41]. In our screen, we found that a loss of function allele of proteasome β6 subunit, pros261, has no effect in both eye and bristle models (Fig. 1O and 1O’). In the case of smt304493, a gene encoded for an ubiquitin-related domain containing protein, was identified as a suppressor (Fig. 1P and 1P’). Since many neurodegenerative diseases show some common pathological pathway, it is possible that genetic modifiers of other neurodegenerations may also modify tauopathies. Therefore, we have also included modifiers of other neurodegenerative diseases in our screen. We uncovered gain-of-function yu gene, a suppressor of spinocerebellar ataxia type 8 (SCA8), as a suppressor of Tau induced toxicity (Fig. 1Q and 1Q’). The interaction seems to be specific because the gain-of-function yuEP1400 allele suppressed Tau induced phenotype, whereas the loss-of-function yuKG02745 allele enhanced Tau toxicity (Fig. 15.

(21) 1Q-1R, and 1Q’-1R’). The dbrEP9 allele, which was a suppressor of the expanded polyglutamine (polyQ) mediated neurodegeneration, was recovered as a weak enhancer of Tau induced degeneration in our screen (Fig. 1S and 1S’). The pentose phosphate pathway (PPP) ribose-5-phosphate isomerase (rpi) encoding gene reports to be involved in oxidative stress resistance and longevity in flies [20]. Downregulation of rpi has found to attenuate polyQ induced neurodegeneration [20]. Interestingly, silencing of rpi also significantly suppressed Tau induced eye degeneration and bristle loss (Fig. 1T and 1T’). Nevertheless, overexpression of rpi has no effect on Tau toxicity (Fig. 1U and 1U’). The SEM of rough eye from interesting modifiers were consistent with our notal bristle screening system (Fig. 2).. Down regulation of rpi extends the lifespan and improves the motor function of tauopathy flies To validate the genetic interactions uncovered in our screen, we have chosen to study rpi, because silencing of rpi exhibited the greatest phenotypic improvement among all other modifiers in our screening (Fig. 1, 2 and Table 1). We first tested that knocking down the expression of rpi has no effect itself when expressed by Eq-gal4 (Fig. 3A-C), but it did potentiate Tau induced toxicity (Fig. 1T and 1T’). To further validate if the beneficial effect of reduced rpi would act in neuronal tissues, we conducted negative geotaxis and survivorship assays. The mobility performance of the control elav-gal4 lines did not decline significantly 16.

(22) over the course of 30 days (Fig. 3D). Neuronal overexpression of Tau driven by elav-gal4 (elav>Tau) impaired motor function of flies in an age dependent manner (Fig. 3D). Although the mobility defect of elav>Tau flies was not evident at age of 1 day, the climbing index was significantly lower that of the elav-gal4 control flies at aged of 15 and 30 days. Knocking down the expression of endogenous rpi in the neuronal tissues using RNAi (elav>rpiRNAi ) did not impair the motor function of flies (Fig. 3D). Nevertheless, neuronal downregulation of rpi significantly improved the climbing ability of tauopathy flies at ages of 15 and 30 days (Fig. 3D). In the survivorship assay, we found that the lifespan of elav>Tau flies was significantly reduced when compared with the elav-gal4 control flies (Fig. 3E). Consistent with a previous study, neuronal silencing the expression of rpi extended the lifespan of flies (Fig. 3E) [20]. Additionally, we found that reduction of rpi in the neuronal tissues significantly increased the survivorship of tauopathy flies (Fig. 3E). The average lifespans of the elav>Tau and elav>Tau/rpiRNAi flies were 48.7 and 66.3 days respectively. Overall, our data suggested that neuronal reduced rpi exhibited beneficial effect on tauopathy flies.. rpi downregulation does not decrease phosphor-Tau species The pathologically accumulation of phosphor-Tau species in the brains of affected individuals is considered to be one of the major features of tauopathies [42]. A group of phosphorylation sites on Tau that are highly associated with neurodegeneration are designated as SP/TP sites [43]. Since downregulation of rpi exhibited neuroprotective effect, we 17.

(23) would like to test if the phosphorylation status of these SP/TP sites were altered when rpi were silenced in tauopathy flies. Immunoblotting experiments were conducted to address the question using antibodies, including pS202, pT212, pT231, pS262 and pS396, which recognized these SP/TP sites specifically. As shown in Fig. 4A, the above antibodies reacted with these SP/TP sites in the brains of Tau-expressing flies, but not with those of elav-gal4 control flies. Knocking down the expression of rpi did not alter the levels of phosphor-Tau in the brains of tauopathy flies (Fig. 4B). Additionally, the amounts of total Tau which were revealed using an antibody against phosphorylated and unphosphorylated form of Tau, were not significantly different between elav>Tau, and elav>Tau/rpiRNAi flies (Fig. 4B). Taken together, we concluded that the beneficial effect of reduced rpi on tauopathy is not mediated through decreasing the amounts of Tau or phosphor-Tau species in flies.. Reduced rpi increases cellular NADPH and reduced form of glutathione in tauopathy flies The rpi gene encoding ribose-5-phosphate isomerase is a component of the PPP whose main functions are to generate pentose for the synthesis of nucleotides, and NADPH for anabolism and the production of reduced GSH. It was reported that reduction in rpi expression causes a shunt of ribulose-5-phosphate back to glucose-6-phosphate to generate more NADPH [20]. Since NADPH is required for generation of reduced GSH, a potent free radical scavenger, and oxidative stress is a critical 18.

(24) pathogenic factor of tauopathy [19], we reasoned that the neuroprotective effect of reduced rpi is likely to mediate through the increasing in the production of NADPH and reduced form of GSH. To address this, we determined the levels of NADPH in heads of flies. Equivalent amounts of NADPH were detected in control elav-gal4 and elav>Tau flies (Fig. 5A). Knocking down the expression of rpi increased NADPH levels by 40% when compared with the control elav-gal4 or elav>Tau flies (Fig. 5A). Additionally, a 30% increase of NADPH were detected in the elav>Tau; rpiRNAi when compared with the elav>Tau flies (Fig. 5A). We also determined the levels of reduced GSH in tested animals, and found that the reduced GSH levels were significantly decreased in the heads of elav>Tau flies when compared with elav-gal4 control flies (Fig. 5B). The levels of reduced GSH were not significant different between control elav-gal4 and elav>rpiRNAi flies (Fig. 5B). However, rpi knockdown resulted in a 1.67-fold increase in the levels of reduced GSH in tauopathy flies (Fig. 5B). Taken together, we concluded that the neuroprotective effect of reduced rpi probably acts through an increase in the contents of NADPH and reduced form of GSH.. Overexpression of hTTLL1 attenuates Tau toxicity and extends the lifespan of tauopathy flies Tubulin post-translational modifications include detyrosination, delta2-tubulin generation, acetylation, glutamylation, and glycylation which occur on the C-terminal domains of tubulin. The C-terminus of the 19.

(25) tubulin is the major binding sites of microtubule-based molecular motors (kinesins and dynesins) and microtubule-associated proteins (MAPs). The glutamylation and glycylation enzymes are known to be members of the tubulin tyrosine ligase-like (TTLL) family with a tubulin-tyrosine ligase (TTL) homology domain which catalyze ligations of different amino acids to tubulins. Downregulation of dTTLL1, dTTLL3A, dTTLL3B, dTTLL6, dTTLL12, and dTPGS2 enhanced Tau induced toxicity (Fig. 6A-6I). Protein sequence alignment between dTTLL1 and hTTLL1 shows high identity (53.9%) and similarity (65.3%) (Fig. 6J). To study the interaction between hTTLL1 and Tau, we first tested that overexpression of hTTLL1 has no effect itself when expressed by Eq-gal4 (Fig. 7A, 7B, and 7E), but it did potentiate Tau induced toxicity (Fig. 7C, 7D, and 7E). To further validate if the beneficial effect of hTTLL1 would act in neuronal tissues, we conducted survivorship assays. The lifespan of tauopathy flies (average lifespan: 44.83 days, n=120) was significantly reduced when compared with the control flies (average lifespan: 67.35 days, n=117). Knocking down the expression of dTTLL1 does not shorten the lifespan of flies (average lifespan: 70.71 days, n=115), but it significantly decreased the survivorship of tauopathy flies (average lifespan: 39.71 days, n=120). Overexpression of hTTLL1 does not shorten the lifespan of flies (average lifespan: 70.44 days, n=113) and significantly increased the survivorship of tauopathy flies (average lifespan：56.26 days, n=119). 20.

(26) (Fig. 7F). Overexpression of hTTLL1 attenuates Tau mediated toxicity in Drosophila.. 21.

(27) Discussion To access the tauopathy, many disease fly models have been generated of which assisted greatly in dissecting the pathomechanisms of tauopathy [22,25,26,33]. These tauopathy fly models have also been used in genetic modifier screens to reveal the genetic pathways controlling Tau toxicity [27,28,30-32]. In most screenings, Drosophila eyes were used as an assessment tool, and the modifiers were identified based on the change in eyes’ roughness, size or volume [24,28,31,32]. In our previous study, we found that Drosophila notal bristle is a versatile system for quantitatively assessing Tau toxicity [33]. To further test the versatility of our system, we conducted a genetic modifier screen using the notal bristle as an assessment tool. For comparison, the eyes of flies were also used in parallel in the screen. We found that both systems gave us similar results (Fig. 1, 2 and Table 1). However, the notal bristle system seemed more sensitive because many modifiers were recovered by using notal bristle system but not by the eye system (Table 1). In most cases, the phenotypic changes in notal bristles were stronger and easier to be identified (Fig. 1 and Fig. 2). Abnormal accumulation of misfolded Tau in neurons is considered to be the primary cause of tauopathies [44]. Since molecular chaperons assist correct folding of proteins and target misfolded proteins for degradation, an array of chaperones were tested in the tauopathy modifier screen. We found that different chaperones exhibited differential effects in tauopathy flies. For instance, overexpression of human Hsp70 (Hspa1L), had no effect in modulating Tau toxicity (Fig. 1H, 1H’ and 22.

(28) Table 1). Consistently, expression of a dominant-negative mutant form of the constitutively expressed Hsp70 (Hsc70-4K71S) also did not alter the Tau induced bristle loss and rough eye phenotypes in flies (Fig. 1I, 1I’ and Table 1). It has been shown that Hsp70 more effective at inhibiting aggregation formed by 3-repeat Tau isoform [45]. Because the tauopathy fly model was generated by overexpression of the 0N4R Tau isoform, this probably explains why Hsp70 can not attenuate Tau toxicity in our screen [26,46]. Of all chaperones, we found that Hdj1 exhibited the strongest suppressive effect on Tau toxicity in our screen (Fig. 1M, 1M’, 1V and Table 1). This finding is in direct contrast with a previously study in which overexpression of Hdj1 enhanced toxicity of TauV337M, a mutant Tau associated with FTDP-17 [30]. The discrepancy could be due to the differences in Tau variants used in the modifier screens. Indeed, it was reported that overexpression of Hdj1 had no effect on eye phenotype induced by wild type Tau in the same study [30]. In our study, we observed that the size and structure of the ommatidia in flies co-expressing Hdj1 and Tau were consistently larger and better organized than those of flies expressing Tau alone (Fig. 1B vs. 1M). As discussed above the eye phenotype seemed less sensitive for identifying genetic modifier of tauopathy. To prevent the experimental bias, we further showed that Hdj1 increased notal bristles significantly in Eq>Tau flies (Fig. 1M’, 1V and Table 1). Since both systems gave us consistent results, we concluded that Hdj1 is a suppressor of tauopathy in flies.. 23.

(29) Although the scale of our modifier screen was small and only focused on components of certain pathways, we were able to identify novel modifiers of tauopathy. Notably, we found that co-expression of yu strikingly suppressed toxicity of Tau, whereas, down-regulation of yu expression enhanced Tau toxicity (Fig. 1Q’, 1R’ and Table 1). yu had been identified as a suppressor of SCA8 [47], indicating that common pathomechanism may operate in both neurodegenerations. The yu gene encodes a dual-specificity A-kinase anchor protein (AKAP) and plays roles in oogenesis and long-term memory formation in Drosophila [48,49]. It is interesting to find that the mammalian AKAP also plays a role in learning memory through regulating synaptic plasticity [50]. Since one of the major pathological manifestations of tauopathies is deficits in learning and memory, it will be interested to test whether AKAP would suppress Tau induced cognitive dysfunction. Reducing the expression of rpi exhibited the strongest suppressive effect against Tau induced eye degeneration and notal bristle loss (Fig. 1T and 1T’), suggesting that rpi is a good therapeutic target for tauopathy. Indeed, neuronal downregulation of rpi improved motor function and extended lifespan of the tauopathy flies (Fig. 3). To investigate the molecular mechanisms underlying the neuroprotective effect of reduced rpi, we found that the NADPH and reduced GSH levels were increased in tauopathy flies when rpi was downregulated (Fig. 5). Since oxidative insults is a critical pathogenic factor of tauopathy and GSH is the major brain antioxidant [18,19], the increasing in NADPH and GSH contents is likely to protect neurons from Tau toxicity. Consistent with our findings, 24.

(30) upregulation of antioxidant defenses genes, such as thioredoxin peroxidase (Tpx) and superoxide dismutase 2 (Sod2), or antioxidant treatment, significantly suppressed Tau toxicity in flies [19]. Moreover, cellular oxidative status did not affect Tau phosphorylation because phosphor Tau was not altered when antioxidant defenses genes were downregulated [19]. This may also explain why reduced rpi did not change the phosphorylation of Tau in tauopathy flies (Fig. 4). Downregulation of tubulin tyrosine ligase-like (TTLL) genes expression enhance Tau induced toxicity. dTTLL1 and hTTLL1 show high identity and similarity (Fig. 6J). The hTTLL1 protein maybe the catalytic subunit of neural tubulin polyglutamylase and polyglutamylate α-tubulin. The interaction between hTTLL1 and Tau, we observed initial result that overexpression of hTTLL1 attenuates Tau toxicity and extends the lifespan of tauopathy flies (Fig. 7). But the relationship between hTTLL1 and Tau are worthy of further investigation.. 25.

(31) Appendix. 26.

(32) Studies of Cap Independent Internal Ribosome Entry Segment Activity of ATXN8OS RNA. 27.

(33) 摘要第八型脊髓小腦運動失調症與ATXN8OS CUG重複擴增RNA 及ATXN8多麩醯胺擴增的蛋白質相關。雖然在CTG方向被轉錄、選擇性裁接及加上polyA尾，ATXN8OS RNA上僅存在小的開放解讀架構(ORF)。為探究ATXN8OS RNA上一包含102胺基酸之ORF1 及一包含41胺基酸與多白胺酸之ORF2是否被轉譯，我們選殖了人的ATXN8OS cDNA 5端片段，置入水母及螢火蟲雙冷光酶基因間。短暫轉染人類胚胎腎細胞及冷光報告測驗顯示，ATXN8OS 上帶有兩個不依賴Cap的內部核醣進入位置(IRESs)。因此， ATXN8OS具有IRESs的活性，可進一步探討第八型脊髓小腦運動失調症的致病機制。. 關鍵字：第八型脊髓小腦運動失調症, 內部核醣進入位置 28.

(34) Abstract Spinocerebellar ataxia type 8 (SCA8) involves bidirectional expression of CUG (ATXN8OS) and CAG (ATXN8) expansion transcripts. Although being transcribed, alternatively spliced and polyadenylated in the CTG orientation, only small open reading frames (ORFs) were noted in ATXN8OS RNA. To investigate if a 102 amino acids containing-ORF (ORF1) and a 41 amino acids plus a polyleucine tract containing-ORF (ORF2) in the ATXN8OS RNA could be translated, we cloned 5’ region of the human ATXN8OS cDNA into dicistronic Renilla and firefly luciferase reporter. Transient transfection in human embryonic kidney (HEK)-293 cells and reporter assay revealed the presence of bipartite cap independent internal ribosome entry segments (IRESs) in the ATXN8OS RNA. Therefore, we could possibly define the IRES activity of ATXN8OS to study the plausible pathogenesis of SCA8.. Keywords: Spinocerebellar ataxia type 8, internal ribosome entry site 29.

(35) Introduction The hereditary ataxias are a group of genetic disorders categorized by inheritance pattern and pathogenic gene. Spinocerebellar ataxia type 8 (SCA8) is one of the late onset autosomal dominant cerebellar ataxia type I (ADACI) which characterized by cerebellar ataxia and neurologic features (dementia, pyramidal and extrapyramidal signs, peripheral neuropathy, and opthalmoplegia) [51]. The clinical feature of SCA8 is a slowly progressive cerebellar ataxia with cerebellar atrophy. The common symptoms are gait, limb, speech and oculomotor incoordination, spasticity, and sensory loss. The phenomena of anticipation which present at an earlier age and with more severe manifestations from one generation to the next often observes in SCA8. The genomic DNA of a SCA8 patient with a non-coding CTG expansion, located on chromosome 13q21, was first reported using the repeat analysis pooled isolation and detection (RAPID) cloning in 1999 [52]. The polymorphic CTG repeat size of ataxin 8 opposite strand (ATXN8OS, formerly SCA8 or KLHL1AS) allele ranges from 15 to 50 in the general population and 80 to 250 in the SCA8. Linkage analysis between SCA8 and the CTG expansion in a seven-generation kindred with 84 members (MN-A family) gets a maximum lod score of 6.8 at θ=0.00. The CTG repeat expansion is tightly linked to ATXN8OS. The ATXN8OS alternative splice variants is identified by means of rapid amplification of cDNA ends (RACE) analysis. The 30.

(36) ATXN8OS transcript consists of up to six exons (D, C3, C2, C1, B, and A). The trinucleotide repeat expansion is located in exon A. Interruptions within the expanded alleles by one or more CCG, CTA,CTC,CCA, or CTT trinucleotides are introduced and duplicated [53]. The intergenerational variation in CTG repeat length almost always expands on maternal transmission and contracts on paternal transmission. Trinucleotide repeat instability and reduced penetrance observe in ATXN8OS expansions. SCA8 transgenic Drosophila and mouse models have been developed. In Drosophila model, RNA binding protein mutants screen for interaction with SCA8 by using retinal neurodegenerative phenotype. yu (a putative protein kinase A anchor protein), muscleblind, split ends, and staufen are genetic modifiers of SCA8 [47]. According to the BAC transgenic mice model, SCA8 has two genes spanning the repeat in opposite direction. One is CTG repeat expansion in ATXN8OS gene. The other is a CAG repeat expansion in ataxin 8 (ATXN8) gene. ATXN8 encods polyglutamine tracts (polyQ) that caused toxic protein aggregation and formed intranuclear inclusions. ATXN8OS transcribed a noncoding CUG expansion RNA which includes a RNA gain-of-function mechanism (nuclear RNA foci) and an anti-sense RNA interference mechanism (antisense regulator of KLHL1) [54-56]. Exon D of the ATXN8OS transcript overlaps the 5’ end of Kelch-like 1 (KLHL1) transcript and maybe act as an antisense regulator of KLHL1 [57-59]. The KLHL1 is an actin-binding cytoplasmic protein which expressed mainly in brain tissues, including the cerebellum, substantia 31.

(37) nigra, frontal lobe, and medulla. The 748 amino acid KLHL1 protein is homologous to Drosophila kelch which maintains ring canal organization during oogenesis. The ring canal organization activity of kelch protein includes ring canal localization, dimerization, and ring canal organization. The kelch protein consists of four protein domains including amino-terminal region (NTR), BR-C, ttk, and bab/Pox virus and Zinc finger domain (BTB/POZ domain), intervening region (IVR), and kelch repeat domain (KREP). The NTR regulates the timing of ring canal localization. The BTB and IVR mediate dimerization or oligomerization. The KREP, which constitutes the actin-binding domain, provides the ring canal localization and organizational function [60]. Mayven (KLHL2), Ectodermal-Neural Cortex 1 (ENC1) and Mayven-related protein 2 (MRP2) which are brain-specific, actin-binding kelch-related protein may perform similar cellular function of KLHL1 [61-63]. SCA8 was first proposed to be cause by an RNA gain-of-function mechanism and analysis of ATXN8OS sequence did not reveal any possible spliced isoform possessing an ORF to extend through the expansion [64]. In this study, sequence analysis reveals the existence of a 102 amino-acid ORF1 and a 41 amino acids plus a polyleucine tract ORF2 in ATXN8OS RNA (Fig. 8A). The two ORFs could be translated if ATXN8OS RNA possesses a cap-independent internal ribosome entry site (IRES) activity. Two mechanisms of initiation of translation in eukaryotic cells are cap-dependent scanning manner and cap-independent internal 32.

(38) ribosome entry manner. The cap-dependent translation initiation mechanism depends on the 7-methyl-guanosine cap structure, m7GpppN m7GpppN (where N is any nucleotide), at 5’ end of mRNAs and eukaryotic initiation factor (eIF). eIF4E recognizes the 5’ cap structure of mRNA and recruits the 43S preinitiation complex (a complex with the 40S ribosomal subunit and the ternary complex consisting of the initiator methionyl-tRNAi, eIF2, and GTP) to mRNA. Ribosomal scanning of mRNA commences until the first AUG initiation codon is recognized and the 48S initiation complex is formed. In turn, joining of the 60S ribosomal subunit to create the 80S ribosome gives rise to start polypeptide synthesis. The alternative mechanism of initiation of translation in eukaryotic cells is cap-independent internal ribosome entry manner [65,66]. Internal ribosome entry site (IRES) elements were first reported in poliovirus (PV) RNA and encephalomyocarditis virus (ECMV) RNA in 1988. Eukaryotic mRNA containing IRES also have been found in yeasts, Drosophila, and mammals later. The IRES elements are specialized RNA regulatory sequence that located in the 5’ untranslated region (5’ UTR) of eukaryotic mRNAs and allow cap-independent translation initiation during cellular stress [67]. The conditions of cell stress include DNA damage, nutrient deprivation, temperature shock, and oxygen shock. A common consequence of cellular stress is the inhibition of target of rapamycin (TOR) pathway leading to inhibit eIF4E-binding proteins phosphorylation or to reduce the activities of eIF4A and eIF4B. On the other hand, GCN2, PERK, HRI, and PKR stress-related kinases 33.

(39) phosphorylate eIF2α to prevent eIF2 recycling and reduce translation initiation rates. When cap-dependent translation is compromised during cellular stress, the cap-independent internal ribosome entry mechanism operates to respond rapidly to changes in cellular environment. In general, IRES mediated translation can be controlled by a group of cofactors known as IRES trans-acting factors (ITAFs) involving polypyrimidine tract-binding protein (PTB), double-stranded RNA-binding protein 76 (DRBP76), erbB-3-binding protein 1 (Ebp 1), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), heterogeneous nuclear ribonucleoprotein (hnRNP), lupus La antigen (La), Gemin5, poly-r(C) binding protein (PCBP2), upstream of N-ras (Unr), SR splicing factor (SRp20), far upstream element binding protein 2 (FBP2), and Nucleolin [68]. These ITAFs may assist to recruit the IRES to the 40S ribosomal subunit through a positive or negative influence (stimulation, repressor, or downregulation). All published IRES elements are available in the IRESite database at http://www.iresite.org [69]. To investigate if these two ORFs can be translated via a cap-independent IRES activity, we first examine the cap-independent IRES activity in the ATXN8OS RNA using a dual luciferase reporter assay.. 34.

(40) Materials and methods Dual luciferase reporter constructs ATXN8OS cDNA containing exon D, C2, C1, B, and A [57] (Fig. 8A) was cloned as described [55]. The ATXN8OS cDNA (NR_002717) were then cloned into the EcoRI site of pEGFP-N1 (Clontech). To construct a dual luciferase reporter, a 76-bp XbaI-BamHI polylinker region of pcDNA3 was first added between the XbaI and BamHI sites of phRL-TK vector (Promega) to introduce an XhoI site as well as remove the SV40 late poly(A) region. Then a 1972-bp XhoI-BamHI fragment containing the firefly luciferase gene and the SV40 late poly(A) signal from pGL3-Basic vector (Promega) was placed between the XhoI and BamHI sites of the modified phRL-TK vector. The resulting pRF plasmid has Renilla luciferase and firefly luciferase gene between the TK promoter and polyadenylation signal (Fig. 8B). To engineer dual luciferase reporter constructs containing different 5’ regions of human ATXN8OS, the ATXN8OS cDNA in pEGFP-N1 was restricted with XhoI, StyI, and/or HaeIII to generate various 5’portion of ATXN8OS cDNA (NR_002717; 1~1032, 1~395, and 396~1032) (Fig. 9A). The blunted XhoI-StyI, XhoI-HaeIII, and HaeIII-StyI fragments were placed in the blunted XhoI site between the two luciferase genes. The 632-bp blunted XhoI-MscI IRES fragment from pIRES2-EGFP (Clontech) was inserted between the two luciferase genes as a positive control.. 35.

(41) Cell culture, transfections, and luciferase assay HEK-293 cells cultivated in Dulbecco’s modified Eagle’s medium containing 10% FCS were plated into 12-well dishes (2 x 105/well), grown for 20 hours and transfected by the lipofection method (GibcoBRL) with the test dual luciferase reporter plasmid (1.5 μg). The cells were grown for 48 hours. Then cell lysates were prepared and luciferase activity was measured by a luminometer using a Dual-Luciferase Reporter Assay System (Promega). The IRES activity of each ATXN8OS cDNA fragment was directly measured by the ratio of the firefly luciferase level to the Renilla luciferase level. For each construct, three independent transfection experiments were performed.. 36.

(42) Results IRES activity of ATXN8OS RNA To study the plausible pathogenesis of SCA8, we cloned the ATXN8OS cDNA containing spliced exons D, C2, C1, B, and A [57]. Sequence analysis revealed the existence of a 102 amino-acid ORF1 and a 41 amino acids plus a polyleucine tract ORF2 in ATXN8OS RNA (Fig. 8A). To investigate if these two ORFs can be translated via a cap-independent IRES activity, we constructed a dicistronic vector pRF in which firefly luciferase was placed after the Renilla luciferase (Fig. 8B). The expression construct was under the control of the HSV-TK promoter. Regions upstream of ATXN8OS ORF1 and ORF2 (Fig. 9A) were inserted into the intercistronic region of the pRF. The IRES from the encephalomyocarditis virus (ECMV) was inserted as a positive control. By comparing the expression levels of firefly luciferase and Renilla luciferase among these constructs after transfection, we could possibly define the IRES activity of ATXN8OS. Transient transfection in HEK-293 cells and reporter assay revealed the presence of bipartite cap-independent IRESs in the ATXN8OS RNA. When the expressed luciferase level of the ECMV IRES was set as 100%, the ATXN8OS 5’ fragments 1~1032, 1~395, and 396~1032 directed firefly luciferase synthesis to a level of 23.5 %, 33.7 %, and 22.5 %, respectively, as compared to the ECMV IRES sequences (Fig. 9B). Both non-overlapping 1~395 and 396~1032 fragments displayed IRES activity; the presence of multiple internal ribosome entry sites was indicated. As the ATXN8OS 5’ fragment 1~395 expressed significant higher level of relative luciferase 37.

(43) activity as compared to fragment 1~1032 (p = 0.024) and fragment 396~1032 (p = 0.014). The results suggest the possible IRES activity existing in the 5’ regions upstream of ATXN8OS ORF1 and ORF2.. 38.

(44) Discussion SCA8 was first proposed to be caused by an RNA gain-of-function mechanism and analysis of ATXN8OS sequence did not reveal any possible spliced isoform possessing an ORF to extend through the expansion [64]. In this study, we revealed the presence of bipartite cap-independent IRESs in the ATXN8OS RNA. The lower luciferase activity of fragment 1~1032 was observed when compared with fragment 1~395 in ATXN8OS RNA. This result may be due to the specialized structure of different RNA fragment which could recruit ribosome or protein factors. In the present study, expression of chimeric construct with an in-frame ORF-EGFP gene demonstrated that ATXN8OS RNA is translatable in various human cells [70]. It indicated that the ATXN8OS putative ORF protein could be translatable and may be expressed via cap-independent IRES. We could possibly define the IRES activity of ATXN8OS to study the plausible pathogenesis of SCA8.. 39.

(45) References 1. Hong M, Zhukareva V, Vogelsberg-Ragaglia V, Wszolek Z, Reed L, et al. (1998) Mutation-specific functional impairments in distinct tau isoforms of hereditary FTDP-17. Science 282: 1914-1917. 2. Hutton M, Lendon CL, Rizzu P, Baker M, Froelich S, et al. (1998) Association of missense and 5'-splice-site mutations in tau with the inherited dementia FTDP-17. Nature 393: 702-705. 3. Spillantini MG, Murrell JR, Goedert M, Farlow MR, Klug A, et al. (1998) Mutation in the tau gene in familial multiple system tauopathy with presenile dementia. Proc Natl Acad Sci U S A 95: 7737-7741. 4. Goedert M, Jakes R (2005) Mutations causing neurodegenerative tauopathies. Biochim Biophys Acta 1739: 240-250. 5. Braak H, Braak E (1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82: 239-259. 6. Arriagada PV, Growdon JH, Hedley-Whyte ET, Hyman BT (1992) Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer's disease. Neurology 42: 631-639. 7. Ingram EM, Spillantini MG (2002) Tau gene mutations: dissecting the pathogenesis of FTDP-17. Trends Mol Med 8: 555-562. 8. Steinhilb ML, Dias-Santagata D, Mulkearns EE, Shulman JM, Biernat J, et al. (2007) S/P and T/P phosphorylation is critical for tau neurotoxicity in Drosophila. J Neurosci Res 85: 1271-1278. 9. Alonso A, Zaidi T, Novak M, Grundke-Iqbal I, Iqbal K (2001) Hyperphosphorylation induces self-assembly of tau into tangles of paired helical filaments/straight filaments. Proc Natl Acad Sci U S A 98: 6923-6928. 10. Taniguchi T, Kawamata T, Mukai H, Hasegawa H, Isagawa T, et al. (2001) Phosphorylation of tau is regulated by PKN. J Biol Chem 276: 10025-10031. 11. Boucher D, Larcher JC, Gros F, Denoulet P (1994) Polyglutamylation of tubulin as a progressive regulator of in vitro interactions between the microtubule-associated protein Tau and tubulin. Biochemistry 33: 12471-12477. 12. Ersfeld K, Wehland J, Plessmann U, Dodemont H, Gerke V, et al. (1993) Characterization of the tubulin-tyrosine ligase. J Cell Biol 120: 725-732. 40.

(46) 13. Rogowski K, Juge F, van Dijk J, Wloga D, Strub JM, et al. (2009) Evolutionary divergence of enzymatic mechanisms for posttranslational polyglycylation. Cell 137: 1076-1087. 14. van Dijk J, Rogowski K, Miro J, Lacroix B, Edde B, et al. (2007) A targeted multienzyme mechanism for selective microtubule polyglutamylation. Mol Cell 26: 437-448. 15. Trichet V, Ruault M, Roizes G, De Sario A (2000) Characterization of the human tubulin tyrosine ligase-like 1 gene (TTLL1) mapping to 22q13.1. Gene 257: 109-117. 16. Regnard C, Fesquet D, Janke C, Boucher D, Desbruyeres E, et al. (2003) Characterisation of PGs1, a subunit of a protein complex co-purifying with tubulin polyglutamylase. J Cell Sci 116: 4181-4190. 17. Janke C, Rogowski K, Wloga D, Regnard C, Kajava AV, et al. (2005) Tubulin polyglutamylase enzymes are members of the TTL domain protein family. Science 308: 1758-1762. 18. Wei H, Kim SJ, Zhang Z, Tsai PC, Wisniewski KE, et al. (2008) ER and oxidative stresses are common mediators of apoptosis in both neurodegenerative and non-neurodegenerative lysosomal storage disorders and are alleviated by chemical chaperones. Hum Mol Genet 17: 469-477. 19. Dias-Santagata D, Fulga TA, Duttaroy A, Feany MB (2007) Oxidative stress mediates tau-induced neurodegeneration in Drosophila. J Clin Invest 117: 236-245. 20. Wang CT, Chen YC, Wang YY, Huang MH, Yen TL, et al. (2012) Reduced neuronal expression of ribose-5-phosphate isomerase enhances tolerance to oxidative stress, extends lifespan, and attenuates polyglutamine toxicity in Drosophila. Aging Cell 11: 93-103. 21. Wentzell J, Kretzschmar D (2010) Alzheimer's Disease and tauopathy studies in flies and worms. Neurobiol Dis 40: 21-28. 22. Iijima-Ando K, Iijima K (2010) Transgenic Drosophila models of Alzheimer's disease and tauopathies. Brain Struct Funct 214: 245-262. 23. Feuillette S, Miguel L, Frebourg T, Campion D, Lecourtois M (2010) Drosophila models of human tauopathies indicate that Tau protein toxicity in vivo is mediated by soluble cytosolic phosphorylated forms of the protein. J Neurochem 113: 895-903. 41.

(47) 24. Chen X, Li Y, Huang J, Cao D, Yang G, et al. (2007) Study of tauopathies by comparing Drosophila and human tau in Drosophila. Cell Tissue Res 329: 169-178. 25. Jackson GR, Wiedau-Pazos M, Sang TK, Wagle N, Brown CA, et al. (2002) Human wild-type tau interacts with wingless pathway components and produces neurofibrillary pathology in Drosophila. Neuron 34: 509-519. 26. Wittmann CW, Wszolek MF, Shulman JM, Salvaterra PM, Lewis J, et al. (2001) Tauopathy in Drosophila: neurodegeneration without neurofibrillary tangles. Science 293: 711-714. 27. Shulman JM, Imboywa S, Giagtzoglou N, Powers MP, Hu Y, et al. (2014) Functional screening in Drosophila identifies Alzheimer's disease susceptibility genes and implicates Tau-mediated mechanisms. Hum Mol Genet 23: 870-877. 28. Ambegaokar SS, Jackson GR (2011) Functional genomic screen and network analysis reveal novel modifiers of tauopathy dissociated from tau phosphorylation. Hum Mol Genet 20: 4947-4977. 29. Wheeler JM, Guthrie CR, Kraemer BC (2010) The role of MSUT-2 in tau neurotoxicity: a target for neuroprotection in tauopathy? Biochem Soc Trans 38: 973-976. 30. Blard O, Feuillette S, Bou J, Chaumette B, Frebourg T, et al. (2007) Cytoskeleton proteins are modulators of mutant tau-induced neurodegeneration in Drosophila. Hum Mol Genet 16: 555-566. 31. Karsten SL, Sang TK, Gehman LT, Chatterjee S, Liu J, et al. (2006) A genomic screen for modifiers of tauopathy identifies puromycin-sensitive aminopeptidase as an inhibitor of tau-induced neurodegeneration. Neuron 51: 549-560. 32. Shulman JM, Feany MB (2003) Genetic modifiers of tauopathy in Drosophila. Genetics 165: 1233-1242. 33. Yeh PA, Chien JY, Chou CC, Huang YF, Tang CY, et al. (2010) Drosophila notal bristle as a novel assessment tool for pathogenic study of Tau toxicity and screening of therapeutic compounds. Biochem Biophys Res Commun 391: 510-516. 34. Tang CY, Sun YH (2002) Use of mini-white as a reporter gene to screen for GAL4 insertions with spatially restricted expression pattern in the developing eye in drosophila. Genesis 34: 39-45.. 42.

(48) 35. Todd AM, Staveley BE (2008) Pink1 suppresses alpha-synuclein-induced phenotypes in a Drosophila model of Parkinson's disease. Genome 51: 1040-1046. 36. Zhu XC, Yu JT, Jiang T, Tan L (2013) Autophagy modulation for Alzheimer's disease therapy. Mol Neurobiol 48: 702-714. 37. Lee MJ, Lee JH, Rubinsztein DC (2013) Tau degradation: the ubiquitin-proteasome system versus the autophagy-lysosome system. Prog Neurobiol 105: 49-59. 38. Petrucelli L, Dickson D, Kehoe K, Taylor J, Snyder H, et al. (2004) CHIP and Hsp70 regulate tau ubiquitination, degradation and aggregation. Hum Mol Genet 13: 703-714. 39. Sahara N, Murayama M, Mizoroki T, Urushitani M, Imai Y, et al. (2005) In vivo evidence of CHIP up-regulation attenuating tau aggregation. J Neurochem 94: 1254-1263. 40. Wang Y, Martinez-Vicente M, Kruger U, Kaushik S, Wong E, et al. (2009) Tau fragmentation, aggregation and clearance: the dual role of lysosomal processing. Hum Mol Genet 18: 4153-4170. 41. Keck S, Nitsch R, Grune T, Ullrich O (2003) Proteasome inhibition by paired helical filament-tau in brains of patients with Alzheimer's disease. J Neurochem 85: 115-122. 42. Lee VM, Goedert M, Trojanowski JQ (2001) Neurodegenerative tauopathies. Annu Rev Neurosci 24: 1121-1159. 43. Yen SH, Liu WK, Hall FL, Yan SD, Stern D, et al. (1995) Alzheimer neurofibrillary lesions: molecular nature and potential roles of different components. Neurobiol Aging 16: 381-387. 44. Spires-Jones TL, Stoothoff WH, de Calignon A, Jones PB, Hyman BT (2009) Tau pathophysiology in neurodegeneration: a tangled issue. Trends Neurosci 32: 150-159. 45. Voss K, Combs B, Patterson KR, Binder LI, Gamblin TC (2012) Hsp70 alters tau function and aggregation in an isoform specific manner. Biochemistry 51: 888-898. 46. Gistelinck M, Lambert JC, Callaerts P, Dermaut B, Dourlen P (2012) Drosophila models of tauopathies: what have we learned? Int J Alzheimers Dis 2012: 970980. 47. Mutsuddi M, Marshall CM, Benzow KA, Koob MD, Rebay I (2004) The spinocerebellar ataxia 8 noncoding RNA causes neurodegeneration and associates with staufen in Drosophila. Curr Biol 14: 302-308. 43.

(49) 48. Hadad M, Bresler-Musikant T, Neuman-Silberberg FS (2011) Drosophila spoonbill encodes a dual-specificity A-kinase anchor protein essential for oogenesis. Mech Dev 128: 471-482. 49. Lu Y, Lu YS, Shuai Y, Feng C, Tully T, et al. (2007) The AKAP Yu is required for olfactory long-term memory formation in Drosophila. Proc Natl Acad Sci U S A 104: 13792-13797. 50. Carnegie GK, Means CK, Scott JD (2009) A-kinase anchoring proteins: from protein complexes to physiology and disease. IUBMB Life 61: 394-406. 51. Harding AE (1982) The clinical features and classification of the late onset autosomal dominant cerebellar ataxias. A study of 11 families, including descendants of the 'the Drew family of Walworth'. Brain 105: 1-28. 52. Koob MD, Moseley ML, Schut LJ, Benzow KA, Bird TD, et al. (1999) An untranslated CTG expansion causes a novel form of spinocerebellar ataxia (SCA8). Nat Genet 21: 379-384. 53. Day JW, Schut LJ, Moseley ML, Durand AC, Ranum LP (2000) Spinocerebellar ataxia type 8: clinical features in a large family. Neurology 55: 649-657. 54. He Y, Zu T, Benzow KA, Orr HT, Clark HB, et al. (2006) Targeted deletion of a single Sca8 ataxia locus allele in mice causes abnormal gait, progressive loss of motor coordination, and Purkinje cell dendritic deficits. J Neurosci 26: 9975-9982. 55. Chen IC, Lin HY, Lee GC, Kao SH, Chen CM, et al. (2009) Spinocerebellar ataxia type 8 larger triplet expansion alters histone modification and induces RNA foci. BMC Mol Biol 10: 9. 56. Moseley ML, Zu T, Ikeda Y, Gao W, Mosemiller AK, et al. (2006) Bidirectional expression of CUG and CAG expansion transcripts and intranuclear polyglutamine inclusions in spinocerebellar ataxia type 8. Nat Genet 38: 758-769. 57. Nemes JP, Benzow KA, Moseley ML, Ranum LP, Koob MD (2000) The SCA8 transcript is an antisense RNA to a brain-specific transcript encoding a novel actin-binding protein (KLHL1). Hum Mol Genet 9: 1543-1551. 58. Benzow KA, Koob MD (2002) The KLHL1-antisense transcript ( KLHL1AS) is evolutionarily conserved. Mamm Genome 13: 134-141. 44.

(50) 59. Chen WL, Lin JW, Huang HJ, Wang SM, Su MT, et al. (2008) SCA8 mRNA expression suggests an antisense regulation of KLHL1 and correlates to SCA8 pathology. Brain Res 1233: 176-184. 60. Robinson DN, Cooley L (1997) Drosophila kelch is an oligomeric ring canal actin organizer. J Cell Biol 138: 799-810. 61. Jiang S, Seng S, Avraham HK, Fu Y, Avraham S (2007) Process elongation of oligodendrocytes is promoted by the Kelch-related protein MRP2/KLHL1. J Biol Chem 282: 12319-12329. 62. Soltysik-Espanola M, Rogers RA, Jiang S, Kim TA, Gaedigk R, et al. (1999) Characterization of Mayven, a novel actin-binding protein predominantly expressed in brain. Mol Biol Cell 10: 2361-2375. 63. Hernandez MC, Andres-Barquin PJ, Martinez S, Bulfone A, Rubenstein JL, et al. (1997) ENC-1: a novel mammalian kelch-related gene specifically expressed in the nervous system encodes an actin-binding protein. J Neurosci 17: 3038-3051. 64. Ranum LP, Day JW (2004) Pathogenic RNA repeats: an expanding role in genetic disease. Trends Genet 20: 506-512. 65. Martinez-Salas E, Pineiro D, Fernandez N (2012) Alternative Mechanisms to Initiate Translation in Eukaryotic mRNAs. Comp Funct Genomics 2012: 391546. 66. Stoneley M, Willis AE (2004) Cellular internal ribosome entry segments: structures, trans-acting factors and regulation of gene expression. Oncogene 23: 3200-3207. 67. Spriggs KA, Bushell M, Willis AE (2010) Translational regulation of gene expression during conditions of cell stress. Mol Cell 40: 228-237. 68. Pacheco A, Martinez-Salas E (2010) Insights into the biology of IRES elements through riboproteomic approaches. J Biomed Biotechnol 2010: 458927. 69. Mokrejs M, Vopalensky V, Kolenaty O, Masek T, Feketova Z, et al. (2006) IRESite: the database of experimentally verified IRES structures (www.iresite.org). Nucleic Acids Res 34: D125-130. 70. Chen IC, Lin HY, Hsiao YC, Chen CM, Wu YR, et al. (2013) Internal ribosome entry segment activity of ATXN8 opposite strand RNA. PLoS One 8: e73885.. 45.

(51) Figure 1. Genetic modifier screening of Tau induced toxicity in Drosophila eye and notal bristle of 1-day-old adult flies. (A-U) Eye; 46.

(52) (A’-U’) Notal bristle. (A) Control flies overexpressing GFP showing normal eye morphology with regular arrangement of ommatidia. Genotype: GMR-Gal4/UAS-GFP;+/+. (A’) Normal bristle pattern in flies expressing GFP. Genotype: UAS-GFP/+; Eq-GAL4/+. (B) Expression of human Tau caused moderate rough eye phenotype. Genotype: GMR-Gal4/+; UAS-Tau/+. (B’) Bristle loss induced by expression of Tau. Genotype: +/+; Eq-Gal4/UAS-Tau. (C-G) Selected genes involved in autophagy pathway modulate ectopic Tau induced toxicity in eye. Genotype: GMR-Gal4/CyO; UAS-Tau/TM3 in trans to alleles indicated. (C’-G’) Selected autophagy genes modulated Tau toxicity in the nota. Genotype: +/+; Eq-Gal4, UAS-Tau/TM6B in trans to alleles indicated. (H-N) Chaperones exhibited less effect on Tau induced toxicity in retina of flies. Genotype: GMR-Gal4/CyO; UAS-Tau/TM3 in trans to alleles indicated. (H’-N’) Chaperones modulated Tau toxicity in nota of flies. Genotype: +/+; Eq-Gal4, UAS-Tau/TM6B in trans to alleles indicated. (O and O’) Functional deficits in proteasome did not affect ectopic Tau induced toxicity in either eyes or nota. Genotype: (O) GMR-Gal4/+; UAS-Tau/pros261. (O’) +/+; Eq-Gal4, UAS-Tau/pros261. (P and P’) Downregulation of smt3 suppressed Tau toxicity in moderate degree in both retina and nota of flies. Genotype: (P) GMR-Gal4/+; UAS-Tau/smt304493. (P’) +/+; Eq-Gal4, UAS-Tau/ smt304493. (Q and Q’) Expression of the yu gene suppressed Tau toxicity in both eyes and nota. Genotype: (Q) yuEP1400/+; GMR-Gal4/+; UAS-Tau/+. (Q’) yuEP1400/+; +/+; Eq-Gal4, UAS-Tau/ +. (R and R’) Loss-of-function mutations in yu enhanced Tau toxicity in both eyes and nota. (R) yuKG02745/+; 47.

(53) GMR-Gal4/+; UAS-Tau/+. (R’) yuKG02745/+; +/+; Eq-Gal4, UAS-Tau/ +. (S and S’) Loss-of-function mutations in dbr enhanced Tau toxicity in both eyes and nota. (S) GMR-Gal4/+; UAS-Tau/dbrEP9. (S’) +/+; Eq-Gal4, UAS-Tau/dbrEP9. (T and T’) Silencing of rpi suppressed Tau toxicity in both eyes and nota. (T) GMR-Gal4/+; UAS-Tau/UAS-rpiRNAi. (T’) +/+; Eq-Gal4, UAS-Tau/UAS-rpiRNAi. (U and U’) Overexpression of rpi has no effect on Tau induced toxicity in both eyes and nota. Genotype: (U) GMR-Gal4/+; UAS-Tau/UAS-rpi. (U’) +/+; Eq-Gal4, UAS-Tau/UAS-rpi. (V) Quantification of the notal bristle number of 1-day-old flies (A’-U’) assessed by student’s t-test. Each bar represents the mean ± SD of three independent experiments. ** P < 0.01; *** P < 0.001. Quantification of the bristle number as follows: (A’) 225.16±19.21, n=50; (B’) 98.67±7.91, n=55; (C’) 82.35±8.41, n=20; (D’) 89.50±6.46, n=20; (E’) 109.00±8.70, n=20; (F’) 112.80±8.35, n=20; (G’) 140.80±8.53, n=20; (H’)103.27±11.88,n=30; (I’) 96.00±8.96, n=30; (J’) 86.80±14.09, n=20; (K’) 79.80±7.19, n=20; (L’) 84.41±7.12, n=22; (M’) 121.23± 6.70,n=22; (N’) 100.81±8.66, n=21; (O’) 101.13±12.41, n=30; (P’) 124.87±13.50, n=30; (Q’) 127.35±11.05, n=20; (R’) 90.25±9.41, n=20; (S’) 90.44±9.07, n=16; (T’) 144.70±8.24, n=20; (U’) 96.93±11.29, n=30.. 48.

(54) Figure 2. SEM of eye micrography and notum used in the genetic modifier screens. (A-J) Eye; (A’-J’) Nortal bristle. Both systems gave the similar results. The notal bristle system seemed more sensitive because many modifiers were recovered by using notal bristle system but not by the eye system (Fig. 3 and Table 1). In most cases, the phenotypic changes in notal bristles were stronger and easier to be identified.. 49.

(55) Figure 3. Downregulation of rpi extends the lifespan and improves the motor function of tauopathy flies. (A and B) Representative notum images of 1-day-old control (A) and rpiRNAi (B) flies driven by Eq-Gal4. Both control and rpi silencing flies shown normal bristle pattern. Genotype: (A) Eq-Gal4/+. (B) Eq-Gal4/UAS-rpiRNAi. (C) Quantification of the bristle number of control (243.17±24.85) and rpiRNAi (240.87±22.67) driven by Eq-Gal4. The RNA interference knockdown expression of rpi has no effect itself when compared with the control flies assessed by Student’s t-test (P = 0.709). (D) In the survivorship assay, we found that the lifespan of tauopathy flies (average lifespan: 50.3 days, n=225) was significantly reduced when compared with the control flies (average lifespan: 70.8 days, n=234). Knocking down the expression of rpi 50.

(56) extended the lifespan of flies (average lifespan: 78.5 days, n=212) and significantly increased the survivorship of tauopathy flies (average lifespan: 61.4 days, n=233). The log-rank test was used for survival analysis. (E) The novel climbing ability assay was used to assess the locomotor activity of flies due to the natural negative geotaxis. The climbing index (CI) was computed as follows: CI =Σ(nm)/10, where n is the number of flies in a given scoring area, and m is the score for the given scoring area (1-5). The mobility performance of the control and downregulation rpi flies did not decline significantly over the course of 30 days. Tauopathy flies impaired motor function and CI was significantly lower than control flies at age 15 and 30 days. Nevertheless, neuronal downregulation of rpi significantly improve the climbing ability of tauopathy flies at age of 15 and 30 days. Statistical analysis were performed using Student’s t-test (n = 100, ** P < 0.01, *** P < 0.001). (D and E) Genotype: control: elav-Gal4/+. Tau: elav-Gal4/+; UAS-Tau/+. rpiRNAi: elav-Gal4/+; UAS-rpiRNAi/+. Tau + rpiRNAi: elav-Gal4/+;UAS-Tau/UAS-rpiRNAi.. 51.