兩個TBP結合蛋白－HMGB1和p53參與聚麸醯胺擴增誘導的神經病變

全文

(1)國立台灣師範大學生命科學系碩士國立台灣師範大學生命科學系碩士論文命科學系碩士論文. 兩個 TBP 結合蛋白結合蛋白－－HMGB1 和 p53 參與聚麸醯胺擴增誘導聚麸醯胺擴增誘導的誘導的神經病變神經病變 Two TBP interacting proteins, HMGB1 and p53, are involved in polyQ induced neuropathies. 研究生: 研究生王孝文 Shiao-Wen Wang 指導教授: 指導教授蘇銘燦博士 Dr. Ming-Tsan Su. 中華民國九十八年六月.

(2) Table of Content. 中文摘要中文摘要.........................................................................................2. Abstract …………………………………………………………..3. Introduction ……………………………………………………...5. Materials and Methods ………………………………………...15. Results…………………………………………………………18. Discussion……………………………………………………….28. References……………………………………………………….31. Figures…………………………………………………………42. Supplementary Data……………………………………………67. - 1.

(3) 中文摘要中文摘要. 脊髓小腦萎縮症第十七型是由於 TATA-box binding protein (TBP) 基因外引子上 CAG 三核苷酸異常擴增所導致。TBP 普遍且廣泛表現在所有型態的細胞，但其 N 端聚麸醯胺的擴增造成的病理徵狀卻只出現在小腦、皮質等中樞神經。前人的研究顯示人類 High mobility group box 1 (HMGB1)和果蠅 Dorsal switch protein 1 (DSP1) 在結構上有高度同源性，並且在 RNA 聚合酶 II 轉錄上扮演抑制子的角色。HMGB1 和 DSP1 能夠直接和 TBP 的 N 端麸醯胺豐富的區域結合，並和結合在 TBP 上的 DNA 形成穩定的三元複合體，分別藉由干擾 TFIIB 和 TFIIA 與 TBP 結合，抑制 RNA 聚合酶 II 轉錄的啟始階段。本研究的主要目的，在 SCA3、SCA17 以及 HD 疾病的果蠅模式下，探討 HMGB1 和 DSP1 和聚麸醯胺擴增的蛋白質間交互作用。我們推測，突變的聚麸醯胺蛋白直接和 HMGB1 結合，並把 HMGB1 隔離在核內包涵體 (intranuclear inclusions)，而 HMGB1 的功能喪失，使得轉錄功能的喪失，使得活化轉錄功能的失調，導致細胞凋亡以及神經退化病徵。在研究中我們發現過度表現果蠅內生性 DSP1 干擾了正常的 TBP 的功能，而過度表現 HMGB1 則對 TBP 沒有影響，顯示 DSP1 和 HMGB1 雖然具有高度同源性，但在轉錄中扮演的角色不完全相同。另一方面，我們發現補充 DSP1 和 HMGB1 可以顯著的減輕全長的 TBP109Q，或是截斷型的聚麸醯胺蛋白如：TBP-109Q NTD、MJD-78Q 和 Htt-97Q 所造成的神經退化。這意味著這些異常擴增的聚麸醯胺蛋白能夠直接與 HMGB1 作用，並將之隔離在蛋白質聚合體 (aggregates)，而 HMGB1 喪失的功能可能是聚麸醯胺疾病的致病的普遍原因之一。同時過度表現 DSP1 和 HMGB1 增加了和變異的聚麸醯胺蛋白結合的機會，也減少了聚麸醯胺蛋白捕捉其他轉錄因子的數量，使得部分失調的轉錄功能可以恢復。. - 2.

(4) Abstract. Spinocerebellar ataxia type 17 (SCA17) is caused by expansion of CAG trinucleotide repeats in the exon TATA box-binding protein (TBP) gene. TBP is commonly and ubiquitously expressed in all cells of multiple cell organsisms due to its essential role in transcription. Nevertheless, the pathological phenotype, which resulted from mutant TBP, are mostly found in central nervous system (CNS), such as cerebellum and cortex. Previous studies showed that RNA polymerase II dependent transcriptional repressor, High mobility group box 1 (HMGB1), binds directly to polyQ rich domain of TBP and was sequested by expanded polyQ containing proteins, including HD and AT1, suggesting HMB box containing proteins play a role in polyQ mediated diseases. Additionally, p53 has also been shown to be a TBP binding protein and implicated in pathogenesis of SCA17 . Herein, we employed an established Drosophila model of SCA3, SCA17 and HD to investigate roles of two TBP interacting proteins HMGB1/DSP1 and p53 in the polyQ induced neurodgenerations. We find that loss of function of HMGB1 results in transcriptional dysfunction and leads to apoptosis and neurodegeneration. We further demostrate that overexpression of DSP1 but not HMGB1 in fly eyes interferes with wild-type and TBP54Q in significant level, suggesting different roles in transcription. On the other hand, complement of DSP1 and HMGB1 markedly ameliorates neurodegeneration caused by full-length TBP109Q and truncated polyQ protein, including TBP-109QNTD, MJD78Q, and Htt97Q. These data suggested that mutant polyQ proteins with expanded Q stretch directly interact with HMGB1 and - 3.

(5) sequestrate them into aggregates, increasing the possibility that HMGB1 loss of function involves in the pathologies of general polyQ protein diseases. In this study we showed that overexprssion of DSP1 and HMGB1 is likely to be a polyQ binding protein, and loss-of-function of DSP1/HMGB1 may play an important role in polyQ mediated diseases.. - 4.

(6) Introduction Autosomal dominant cerebellar ataxia Autosomal dominant cerebellar ataxias (ADCAs) consist of a complex group of neurodegenerative disorders characterized by late onset (although rare infantile and juvenile cases are reported), a progressive deterioration course, and significant dysfunction involving specific neuronal groups in the cerebellum, brain stem and spinal cord1. ADCAs are clinically characterized by some behavior disorders such as ataxia, dysathria, ophthalmoplegia, pyramidal and extrapyramidal signs and peripheral neuraphathy. Dementia occurs only in some forms of spinocerebellar ataxia (SCA), such as SCA12, SCA2, SCA3/Machado-Joseph disease (MJD) 3, SCA124, rising in the late stages of the disease. In case of SCA175 and dentatorubral-pallidolluysian atrophy (DRPLA) 6, dementia is a constant feature of the phenotype, as seen in Huntington disease7. Genetically, more than 25 types of causative mutations in various genes have been identified and eight of these disorders including SCA1, 2, 3, 6, 7, 8, 17 and DRPLA are closely associated with an instable expansion of tandemly repeated CAG trinucleotides that code for polyglutamine tracts in the functional protein of the respective gene8. Increased numbers of CAG repeats are believed to be the main cause for the pathogenesis of these polyglutamine diseases because they involve increasingly unstable chromosomes and malfunction of the encoded proteins 5,9,10,11,12.. Spinocerebellar ataxia type 17 Spinocerebellar ataxia type 17 (SCA17) is caused by the expansion larger - 5.

(7) than 44 CAG/CAA repeats in the coding region of the TATA box binding protein (TBP) gene leading to an abnormal expansion of a polyglutamine stretch in the corresponding protein. The coding region of TBP contains 29-42 CAA/CAG repeats with the most common alleles containing 32-39 units of a CAA/CAG repeat. Alleles with 43-48 represent a zone of partial penetrance or late onset14,15. Disease causing alleles with up to 63 Gln codons were identified16. Clinically, SCA17 may phenocopy Huntington disease (HD) and is alternatively named HD-like 4, or HDL47,17. Prominent features include cerebellar signs associated with incoordination, imbalance, oculomotor abnormalities, and involuntary movement, psychiatric symptoms such as psychosis, depression, and dementia. Epilepsy is also seen in some case18. One of the pathological features of this disease is selective neuronal loss in affected patients. The Purkinje cells in the cerebellum are primarily involved; cells in other areas of the brain may also be affected. Neuropathological findings include atrophy of the cerebellar, cortical, subcortical, striatum, thalamus, inferior olive, and nucleus accumbens5,17,19,20. Histological results have revealed severe loss of Purkinje cells in the molecular layer of the cerebellum accompanied with apparent gliosis as well as in the caudate nucleus19,21. Immunohistochemical examination of postmortem brain tissue from SCA17 patients has detected ubiquitinated neuronal intranuclear inclusions (NIIs) immunopositive for TBP and polyglutamine5,19,21. It is reported that polyglutanmine expansions dower mutant proteins with a gain of toxic properties, then these mutant proteins display aberrant interaction with other partner proteins and accumulated exclusively in the neuronal nuclei of affected brain regions to form NIIs, an - 6.

(8) important histopathological hallmark of polyglutamine diseases18-29.. TBP and polyglutamine diseases TBP involves in not only the pathology of SCA17 but also other polyglutamine diseases. In an immunohistochemical study of postmortem brain tissue from SCA1, SCA2, SCA3 and DRPLA, TBP was found to co-localize with the disease associated protein in nuclear aggregates29,30. It was suggested that the presence of a polyglutamine stretch was necessary for recruitment into these aggregates. In HD it was shown that TBP and Huntingtin are part of the insoluble protein fraction31. Therefore, TBP loss of function due to sequestration of TBP in nuclear inclusion may be one common. molecular. process. implicated. in. polyglutamine. induced. neurodegenerations.. High mobility group box-1 protein High mobility group box-1 protein (HMGB1) is an abundant, ubiquitous and extremely conserved (99% mammal identity) chromosomal protein found particularly in all eukaryotes32,33. In mammalians, most cell types contain up to one million molecules of HMGB134. It contains three domains. N-terminal two-thirds of HMGB1 is composed of two highly conserved HMG domains, referred to as A and B box (amino acids 1-89 and 91-176), each containing about 80 amino acids which riches in lysines and arginines. On the contrary, the C terminus is extremely acidic with the last 30 residues being a stretch of entirely aspartate and glutamate residues32,35-38. The HMG box domain has been recognized for its ability to bind and curve DNA - 7.

(9) structures without sequence specificity. Transcription factors like SRY or LEF-1 contains a HMG box domain structurally similar to those of HMGB139,40. Unlike HMG1 which binds weakly and nonspecifically to common B form DNA with TBP, SRY, LEF-1 and UBF interact with DNA through the minor groove with certain sequence specificity. HMGB1 bends the DNA molecule upon binding to linear DNA. However, it binds to angled DNA, such as four-way junctions, bulged DNA and cisplatin- modified DNA40-47, with rather high affinity. This suggests that HMGB1 may function as an architectural protein by distorted linear DNA into a bent conformation which makes DNA easily to be accessed by other transcriptional factors34. The HMG box domains of HMGB1 are also participated in protein-protein interaction. Many members of the HMG box family are known to be transcription factors. Additionally, HMGB1 has been reported to interact with many activated transcription factors and function as a coactivator by a mechanism that augments the affinity of various regulatory proteins to their DNA recognition sequences. Activators that are capable of recruiting HMGB1 include class I (steroid) hormone receptors48-51, Oct-1, -2, -4 and -652,53, HOX954, p5355,56, p7357, Rel family58,59 , ZERBA60 and ICP461. However, there is usually no ternary complex being detected by electrophoretic mobility shift assays (EMSAs), as if HMGB1 finished its job and then disassociated with the DNA34. HMGB1 is expected to play a significant role in promoting enhanceosome assembly. HMGB1 facilitates binding of ZEBRA and Sp1 to the promoter in vitro, to form a simple enhanceosome, which in turn recruits TFIID and TAIIA.. - 8.

(10) HMGB1 and TBP HGMB1 has been reported to function as a transcriptional repressor in class II gene transcription by interacting or competiting with different basal transcription factors62-65. Binding of HMGB1 to TBP increases the affinity of TBP for the TATA box and forms a stable HMGB1/TBP/TATA complex62,63. The interaction takes place between the highly conserved core domain of TBP and the HMGB1 box A64. It was recognized that a stable HMGB1/TBP/TATA complex formation requires the acidic C-terminus of HMGB1 and the Q-tract in the N-terminus of TBP63. This complex inhibits basal and activated transcription by preventing the recruitment of TFIIB to a TBP/TATA complex and, subsequent formation of preinitiation complex (PIC)62,63. TFIIA can bind preferentially to TBP and inhibits HMGB1 binding. Additionally, TFIIA can also antagonize HMGB1-dependent repression by dissociate HMGB1 from the preformed HMGB1/TBP/TATA complex. The binding site of HMGB1 on TBP may be very close to that of TFIIB on the C-terminus, whereas HMGB1 and TFIIA may also bind to the same or overlapping sites on TBP62,65. This can explain that HMGB1 blocks the interaction of TFIIB with TBP and is competed by TFIIA. More precisely, TFIIA binding to TBP excludes HMGB1 binding, while HMGB1 and TFIIB are not mutually exclusive, but TFIIB is repositioned and loosely bound to TBP. It was inferred that HMGB1 binding induces a conformational change in TBP and less accessible to TFIIB that brake the preinitiation complex assembly to a stop64.. HMGB1 and polyglutamine diseases Previous study suggested that HMGB1/2 colocalized with mutant Htt and - 9.

(11) AT1 in the inclusion bodies of primary cortical neurons and Hela cells. Colocalization of HMGB1/2 and mutant Htt proteins in nuclear inclusion bodies has also been detected in the striatal neurons of Htt transgenic mice. Immunohistochemistry with striatal neurons of Htt transgenic mice and cerebellar neurons of AT1 knock-in mice showed a reduction of HMGB1/2 in nucleus.66 Complementation of HMGB1/2 ameliorated toxicity of mutant Htt or AT1 in primary neurons in vitro. Neurite extension, branching and survival rate of transgenic animal were recovered by reversing repressed basal transcription and abnormal cell death. HMGB1 can also rescue eye phenotype, including pigment cell degeneration, disordered retina morphology and loss of photoreceptor neurons induced by mutant Htt or AT1 in Drosophila models in vivo.66. Dorsal Switch Protein 1 Dorsal Switch Protein 1 (DSP1), the Drosophila protein, is a member of the high mobility group (HMG) family of non-histone chromosomal DNA-binding proteins38. DSP1 primary amino acid sequence can be divided into two major subdomains. The N-terminal domain contains two glutamine-rich regions, and the C-terminal domain contains two HMG boxes (A and B) and an acidic tail (amino acids 174-393). Alignment of the sequences of mammalian HMGB1 and DSP1 suggested that the two proteins are highly homologous67. DSP1 is expressed ubiquitously by the end of germ band retraction (stage 12) during embryogenesis. It is later restricted in the ventral nerve cord and brain (stage 15-16). In adult flies, DSP1 is - 10.

(12) detected only in brains and ovaries68,69. DSP1 was first identified as a co-repressor that converts the Drosophila protein Dorsal, the mammalian NF-κB protein complex, from transcriptional activators to repressors when a negative regulatory element (NRE) is present adjacent to Dorsal-binding sites in the zen promoter and adjacent to the NF-κB -binding site in the mammalian interferon- β (IFN-β) enhancer67. Dorsal and NF-κB both belong to Rel family of transcription factors70,71, which includes the c-rel proto-oncogene72, the insect Dif73, Relish74 and Gambif175. DSP1 is proposed to interact with mammalian protein SP100B, a splice variant of SP100A, and that SP100B, in turn, bind to mammalian homologs of the Drosophila heterochromatin protein 1, HP1, forming a DSP1/SP100B/HP1 transcriptional silencing complex that could be recruited to DNA by interaction with various activators of the Rel family such as Dorsal76. Moreover, it has been reported that two transcription factors, dAP-1 and Stat92E, which are activated by LPS- or PGN-induced signaling due to continuous immune response, form a repressosome complex with DSP1. The complex replaces Relish at the promoters of diverse immune effector. genes,. and. then. recruits. histone. deacetylase. to. inhibit. transcription77. Like human HMGB1, DSP1 has been found to bind directly to TBP and form a stable ternary complex with TBP bound to DNA. Screening for DSP1 deletion mutants demonstrated that TBP binding requires at least one intact HMG box. DSP1 preferentially disrupts TFIIA-TBP-DNA complex by displacing TFIIA from binding to TBP. In contrast, DSP1 reduces but does not abolish TFIIB-TBP binding78. Unlike other repressors that are able to bind TBP79, DSP1 was shown to specifically inhibit activated but not basal - 11.

(13) transcription in vitro78. It was proved that TFIIA, the general transcription factor, has only minor stimulatory effect on basal transcription levels80, but plays. a. distinct. role. in. activator-mediated. transcription81-84.. Activator-regulated transcription requires TFIIA in addition to TAFs and various coactivators85. These results support the finding that DSP1 represses activated but not basal transcription through interfering with the binding of TBP by TFIIA, which implicates in activator- regulated transcription78.. p53 and TBP The tumor suppressor p53 functions as a transcriptional activator when bound to promoters containing p53 response elements through a heterologous DNA-binding domain86-88. On the other hand, p53 has been reported to repress a variety of genes that lack p53 response elements89-94. A number of transcription factors have been shown to directly interact with p53, including the TBP92,95-97 and TBP-associated factors, Drosophila TAFII40 and TAFII60 and the corresponding human factors TAFII31 and TAFII7098,99. An interaction between the p53 activation domain and the DNA binding domain of TBP, leads to transcriptional repression92,100,101. It has further been proved that a p53–TBP interaction is not sufficient for transcriptional repression by p53 and an interaction between p53 and other factors, such as TAFs are required for activated but not basal transcription100.. p53 and polyglutamine diseases It has been revealed that mutant Htt with expanded polyglutamine binds to - 12.

(14) p53 and upregulates levels of nuclear p53 as well as p53 transcriptional activity in primary cortical cultures. p53 levels are also increased in the brains of mutant Htt transgenic mice and HD patients. Genetic deletion of p53 diminished mutant Htt-induced retinal degeneration in Drosophila and normalized neurobehavioral deficits of mutant Htt transgenic mice. Other research manifested that p53 deficiency in mice resulted in a reduction of mutant Htt expression in brain, an increase in proenkephalin mRNA expression, and a significant increase in nuclear aggregate formation, which was suggested to be a protective mechanism, in the striatum103. Taken together, it has been proposed that p53 exerts its effect after the initial pathogenic events in SCA1 to promote the progression of neuronal degeneration in SCA1 mice, but this effect may be independent of neuronal cell death103.. - 13.

(15) Objective. Since HMGB1 can interact with expanded polyQ tract but not normal Htt and AT1, we also have known that HMG box containing proteins, such as HMGB1 and DSP1, can bind to TBP and function as repressors in RNA transcription which may implicated in polyQ induced neurodegenerations. We have utilized the genetic trackable organism, Drosophila melanogaster, to model neuropathies of polyQ diseases including SCA3, SCA17 and HD. Herein, I used Drosophila to examine the genetic interaction of HMGB1/DSP1 and polyQ-expanded TBP, AT3 and Htt. Furthermore, it is essential to learn the role of HMGB1 in pathological mechanism of polyQ disease. Because HMGB1 might exhibit “architectural activity”, such as local deformation of DNA and increasing in accessibility of preinitiation complex, we suspected that HMGB1 plays an important role in transcriptional activation. Expanded polyQ containing proteins sequestrate HMGB1 into nuclear inclusion. It is expected that loss of HMGB1 function may cause transcriptional dysfunction, apoptosis and neurodegeneration. On the other hand, complement of HMGB1/DSP1 may reduce the cytotoxicity caused by mutant polyQ protein because they compete with other proteins such as transcription factors for being captured by mutant polyQ proteins.. - 14.

(16) Materials and Methods. Fly strains and Genetics All. Drosophila. melanogaster. stocks. were. raised. on. standard. cornmeal-yeast-agar medium supplied by Institute of Molecular Biology Academia Sinica and cultured at 25°C. We anaesthetize flies with CO2 then observe and manipulate them under a stereomicroscope. Most Drosophila lines were supplied by Bloomington stock center, and the transgenic lines UAS-TBP-36Q, -54Q, -109Q, -109Q N-terminal domain (NTD) are generated in our laboratory. TBP-109QNTD is a C-terminal truncated fragment retaining with an expanded-polyglutamine stretch. Unlike full-length 109Q TBP, TBP-109QNTD is unable to bind TATA-box DNA due to the lack of the DNA-binding domain at C-terminus. All UAS transgenic and recombination lines were maintained over the attached balancer (Cyo or TM3Sb) or double balancer (STB).. UAS-GAL4 system The UAS-Gal4 system is a method for ectopic gene expression in Drosophila which allows the selective activation of any cloned gene in tissue or cell specific patterns (Brand and Perrion, 1993). For genetic interaction, UAS transgenic lines were placed in trans to recombination lines.. Scanning Electron Microscopy - 15.

(17) 8-10 Drosophila’s were decapitated and immersed in 2% glutaraldehyde. The samples were fixed at 4°C overnight. After fixation, the samples were rinsed with PBS and immerse in 2% osmium acid at room temperature for 5 hours. Dehydration were performed in 30% ethanol, then immerse in 30% ethanol at room temperature for 1.5 hours. Dehydrate through an ethanol series (each twice in 50% and 75% ethanol at room temperature for 20 minutes). Head was immerse in 100% acetone at 4°C overnight before critical point dry (CPD) and sputter-coat (Kimmel et al., 1990).. Immunohistochemstry A rhodamine conjugated phalloidin (Invetrogen) was used to stain adult eyes at 2, 7, 14 days after eclosion. Adult retinas were dissected and fixed for 15 min in 4% formaldehyde then immersed in PBST for 15-20 minutes. For ommatidia staining, PBS containing 1% rhodamine-phalloidin and 0.3% Triton X-100 was conducted at 37°C for 30 minutes. Wash of retina are carried through three 10 min changes of PBST. Leica TCS SP2 laser confocal microscope was used for morphological assessments.. Acridine Orange Staining Acridine orange was used to identify dying cells. Adult retinas are dissected without fixation. Tissues are immersed in acridine orange in PBST (1.6:1000) for 5 minutes. Wash the tissues for 2 minutes and examined immediately using confocal microscope.. Climbing Assay - 16.

(18) A graded climbing assay was applied to access the locomotor activity of flies. The climbing apparatus which consists of a glass tube (30 cm in length and1.5 cm in diameter) with an attached plastic funnel at one end was constructed accordingly. The glass tube is divided into a series of five 2-cm scoring areas (scored 1-5) and a buffer zone. For each trial, 10 male flies were tapped down to the bottom of the climbing apparatus, and allowed 10 seconds to climb up into scoring areas. All the trial sections were repeated 10 times and a total of 50 flies were assayed for each time point. The number of flies in each scoring areas were counted, and the climbing index (CI) was calculated as follows: climbing index=Σ(nm)/10, where n is number of flies in a given scoring area, and m is the score for the given score area (1-5). The results were statistically analyzed and graphed using SigmaPlot software (Systat Software Inc., San Jose, California USA).. Lifespan analysis Flies were raised to adulthood at 25°C and newly eclosed flies were placed in vials at low density (10–25 flies per vial). Males and females were kept in separate vials. Flies were transferred to fresh vials every 5 days. The number of surviving flies was scored. Survival curves were generated by calculating the percentage of surviving flies.. - 17.

(19) Results. DSP1 exacerbates loss of TBP function in fly Given that DSP1, the fly homolog of HMGB1, has been identified to bind directly to human TBP and represses activated transcription78. We first examined whether DSP1 exert its effect on endogenous Drosophila TBP. TBP-RNAi flies were mated with homozygous EP-DSP1 flies using GAL4/UAS binary system. The expression is triggered by pan-retinal GMR-GAL4 driver, which directs expression in all cell types of the eyes during developing and adult stage. Examination of the external eye morphology using light microscopy and scanning electron microscopy (SEM) was performed on male and female flies. Compared with the driver alone, the fly eyes overexpressing DSP1 showed mild pigment loss (Fig. 1A vs. 1B). dsRNAi knockdown of the expression dTBP in eyes cause severe depigmentation (Fig.1C). In the genetic background of dTBP RNAi, DSP1 overexpression led to lethality in pupal stage and a significant decline in eclosion rate (Fig. 1D, I). Scanning Electron Microscope analysis revealed apparent loss of bristles when DSP1was overexpressed alone (Fig. 1E vs. 1F) compared with driver-alone control. Knockdown dTBP expression in eyes by siRNA showed disordered ommatidial arrays and bristles (Fig. 1G), and the rough-eye phenotype was enhanced by DSP1 overexpression, leading to severed fused ommatidia (Fig. 1H).. - 18.

(20) DSP1 interferes with the function of normal hTBP In human, the normal N-terminus of TBP contains 25-42 glutamine residues. Expanded repeats of over 42 glutamines generally result in SCA17. To create a Drosophila model of mutant TBP-inducing toxicity, we generated transgenic flies expressing several forms of hTBP: (1) normal human TBP (hTBP-36Q), (2) mutant TBP with elongated polyglutamine (hTBP-54Q and -109Q), (3) a truncated N-terminal fragment of mutant TBP with 109 glutamine residues (hTBP-109QNTD). We first tested the interaction of DSP1 and normal hTBP. Light Microscopy analysis was performed at 1, 3, 5 and 7 weeks on male flies. Compared with driver-alone control (Fig 2A), over-expression of DSP1 displayed a progressive pigment loss (Fig 2B). Eye of transgenic fly misexpressing hTBP-36Q. is. normal(Fig 2C). Nevertheless, coexpression of DSP1 and hTBP-36Q results in moderately eye degeneration with depigmentation phenotype (Fig 2D). SEM images revealed that the external eye morphology of GMR-GAL4 driver is normal (Fig 3A). Overexpression of DSP1 caused severe loss of bristles (Fig 3B). Transgenic misexpressing hTBP-36Q did not lead to any eye defect as seen in control flies (Fig 3C). Similarly, coexprssion of hTBP-36Q and DSP1 resulted in profound bristle loss s and moderately degree of rough eye phenotype (Fig 3D). In order to compare the internal morphology, confocal imaging of dissected adult retinas stained with rhodamine-phalloidin was performed. The control Gal4 driver shows a characteristic trapezoidal pattern in ommatidia. Each ommatidium contains eight photoreceptor neurons, seven of which are typically visible in a tangential optical plane with light-gathering organelles (rhabdomeres) (Fig - 19.

(21) 3E). Overexpression of DSP1 in eyes produced a relatively normal array of rhabdomeres with small degree of polarity defect (Fig 3F). Transgenic eyes expressing hTBP-36Q appeared normal retinal architecture (Fig 3G). Coexpression of hTBP-36Q and DSP1 generated relatively slightly abnormal trapezoidal arrays and mildly loss of photoreceptor phenotype (Fig 3H). However, quantification analysis did not show significantly difference on the number of photoreceptor among various genetic backgrounds (Fig 3I).. DSP1 suppresses toxicity induced by polyQ-expanded hTBP We then examined whether complement of DSP1 modulate polyglutamine protein toxicity. Light microscopy showed that TBP-54Q driven by GMR-GAL driver exhibits nearly normal eye (Fig 4A), whereas mild rough eye phenotype when DSP1 and TBP-54Q was coexpressed (Fig 4B). Retinal misexpression of TBP-109Q caused progressive loss of pigment cells (Fig 4C), and the phenotype were enhanced by with the expression of DSP1 (Fig 4D). Electron micrographs revealed that the transgenic fly expressing TBP-54Q exhibits normal external eye morphology (Fig 5A). Nevertheless, occasionally loss of photoreceptor was observed (Fig 5E). Although overexpression of DSP1 and TBP-54Q resulted in profoundly loss of bristles (Fig 5B). The loss of photoreceptor phenotype was recovered (Fig. 5F). Transgenic fly misexpressing TBP-109Q produced markedly rough eyes with abnormally fused ommatidia and missing bristles (Fig 5C). Significant retina degeneration with disarranged ommatidial morphology and loss of photoreceptor neurons was also observed (Fig. 5G). In contrast, - 20.

(22) coexpression of DSP1 was able to restore rough eye phenotype (Fig 5D and 5H).. DSP1 alleviates the toxicity caused by N-terminal fragments of polyQ-expanded TBP Given that the binding of HMGB1 on TBP has been mapped to glutamine rich tract63, we further investigated weather DSP1 interacts with 109Q N-terminus of TBP (TBP-109QNTD) and modulates the polyQ toxicity. Our data revealed that transgenic flies misexpressing TBP-109QNTD driven by GMR-GAL4 couses high lethality rate (Fig. 6A, C). Coexprssion of TBP-109QNTD and DSP1 conferred a nearly 95% eclosion rate (Fig. 6B and 6C). DSP1 overexpression in eyes assessed by SEM showed predominant loss of bristles (Fig. 7A, B), but internal eye morphology was greatly improved (Fig. 7C) Transgenics misexpressing TBP-109QNTD resulted in severe retinal degeneration with missing bristles, disordered and fused ommatidia (Fig. 7D and 7E). The internal structure of the retina was completely disrupted, producing misshapen ommatidial array and a striking loss of photoreceptor neurons (Fig. 7F). In contrast, transgenic fly coexpressing TBP-109QNTD and DSP1 displayed more normally organized ommatidial arrays and bristles (Fig. 7G and 7H). The internal retinal architecture was (Fig. 7I). We quantified the number of rhabdomeres per ommatidium in 14-day-old flies expressing DSP1, TBP-109QNTD alone or both DSP1 and TBP-109Q. We demonstrate that TBP-109QNTD induced neurodegeneration was significantly suppressed by DSP1 (Fig. 7J).. - 21.

(23) DSP1 represses mutant Htt- and AT3- induced retinal degeneration To investigate the effects of DSP1 on the toxicity of other polyQ expanded proteins, flies expressing expanded truncated human huntingtin and ataxin-3 were crossed with EP-DSP1 lines. Light microscopy revealed that truncated ataxin-3 with 78Q (MJD-78Q) induced severe loss of pigmentation and retina degeneration (Fig. 8A). When DSP1 was coexpressed with MJD-78Q, a better phenotype with less pigment loss was detected (Fig. 8B). Similarly, overexpression of DSP1 ameliorated the strong depigmentaion elicited by mutant Htt with 97Q (Fig. 8C vs. 7D).. DSP1 modulates neurobehavioral abnormalities in HD and SCA3 transgenic flies We then examined whether DSP1 moduates behavior phenotypes of mutant Htt or AT-3 transgenic flies. We observed that adult flies misexprssing Htt-97Q or MJD-78Q driven by ELAV-GAL4show an earlier onset of locomotor impairment at day 5 assessed by climbing assay. At day 20, both genotypes displayed markedly exacerbated motor deficit as compared with driver alone, and transgenic flies died before day 40, shown as 0 on climbing index. (Fig. 9A, yellow bar and dark green bar). In contrast, overexpreesion of DSP1 rescued the mutant Htt97Q or MJD78Q induced motor dysfunction (red bar and light green bar). Lifespan analysis showed that Htt97Q-expressing flies and MJD78Q-expressing flies both have a strikingly shorter lifespan than diver-alone, whereas the expression of Dsp1 - 22.

(24) apparently extended lifespan of transgenic flies (Fig. 9B). Taken together, these results suggested that DSP1 provides protective affect against neurodegenerations induced by disease-causing polyQ proteins, including TBP, Htt and AT3.. HMGB1 ameliorates internal retinal degeneration caused by polyQ- expanded TBP To investigate how HMGB1 affects the pathology of SCA17, UAS-HMGB1 trangenic flies were crossed with UAS-TBP36Q/54Q and 109Q driven by GMR-GAL4. Unlike Dsp1, expression of HMGB1 did not affect eye morphology of transgenic flies expressing TBP(Fig 10D) and TBP54Q (Fig 10F). In two-week-old adult eyes stained with phalloidine, we examined changes in retinal thickness of retinal neuronal layers caused by interaction of TBP109Q and HMGB1. Compared with driver- alone control (Fig 11A), retinal thickness of flies expressing HMGB1 is normal (Fig 11B). Overexpression of TBP109Q led to substantial reduction in the depth of retina (Fig 11C). In contrast, TBP109Q induced retinal thinning was significantly rescued by expression of HMGB1 (Fig 11D). We then observed the morphology of phalloidine-staining retina in tangential views. Transgnic flies overexpressing HMGB1 displayed a near- normal retinal morphology (Fig 11F), as in control driver line (Fig 11E). Overexpression of TBP109Q produced disordered retinal architecture with remarkable loss of photoreceptors neurons (Fig 11G and 11J). In contrast, the coexpression of TBP109Q and HMGB1 displayed relative normal ommatidial arrays with increase in number of photoreceptors (Fig 11H and 11J). - 23.

(25) HMGB1 and DSP1 inhibit TBP109Q- induced cell death To learn if Dsp1 ameliorates retinal degeneration by repressing mutant polyQ protein induce apoptosis, we used acridine orange to label apoptotic cells in retina. Adult eyes of 1-week-old and 2-week-old transgenic flies were carefully dissected, stained with acridine orange and observed immediately using confocal microscopy. Compared with GMR-GAL4 driver, retinal expression of TBP-Q109 cause widespread cell death and the yellow-green AO positive punctuated staining pattern were increased from day. 7. to. day. 14. ((Fig.. 12A. and. 12E. vs.. 12B. and. 12F).. Polyglutamine-induced cell death in retina was strongly inhibited by the expression of HMGB1 (Fig. 12B vs. 12F) or DSP1 (Fig. 12Dvs. 12H), as indicated by markedly reduction in cell death in day 7 and day 14 compared with eyes expressing TBP109Q alone.. HMGB1 suppresses toxicity of mutant TBP, AT3 and Htt in vivo Overexpression of HMGB1 did not disturb external and internal eye morphology (Fig. 13B and 13F). In contrast, the severe retinal degeneration elicited by TBP-109QNTD with sunken and misshaped ommatidial arrays () was substantially rescued by overexpressing HMGB1 (Fig. 13C and 13G vs. 13D and 13H). We quantify the amount of photoreceptors and found that flies coexpressing HMGB1 and 109QNTD had numbers distribute among 4 to 6, whereas most flies expressing 109QNTD alone had 1~2 photoreceptors in each ommatidia (Fig. 13I). We further examined the effect of HMGB1 on - 24.

(26) mutant AT3/Htt- induced eye degeneration. Complement of HMGB1 moderately ameliorated the mutant AT3/Htt- induced sunken ommatidia (Fig14A and 14B) and the loss of photoreceptor neurons (Fig14E and 14F). A similar but more notable effect was observed that HMGB1 repressed mutant Htt-induced rough eye phenotype with fused ommatidia (Fig. 14C vs. 14D) and loss of photoreceptors (Fig. 14G vs. 14H).. TBP ameliorates p53- promoted retinal cell death It has been proposed that wild-type p53 binds directly to human TBP and represses TATA-dependent transcription92,100,101. If p53 was exerting its repressive effects via interacting with TBP, then it is conceivable that p53-mediated repression could be rescued by overexpression of TBP. To investigate the genetic interaction of p53 and wild-type and polyQ-expanded TBP, Transgenic flies expressing wild-type human p53 were mated with transgenic flies expressing human TBP36Q, 54Q or 109Q. Light microscopy analysis. indicated. that. compared. with. driver-alone. control,. p53. overexpression caused markedly reduction in eye size and severe pigment loss (Fig 15A vs. 15B). Coexpression of TBP-36Q modestly ameliorated the eye degeneration (Fig 15C), whereas the overexpression of both p53 and TBP-54Q worsen the rough eye phenotype (Fig 15D). To our surprise that overexpression of TBP-109Q mildly rescued the rough phenotype induced by p53 (Fig 15E). SEM micrographs revealed a normal external retinal morphology of driver-alone control (Fig 15F). In contrast, transgenic flies overexpressing p53 resulted in apparent reduction in eye size and destruction of retinal architecture (Fig 15G). When TBP-36Q was coexpressed, there - 25.

(27) was a modest increase in eye size, but still disrupted ommatidial arrays (Fig 15H). TBP-54Q did not improve the rough eye phenotype when it is coexpressed with p53 (Fig 15I). To our surprise that the eye size of transgenic flies coexpressing p53 and TBP-109Q was modestly larger than that of fly expressing p53 alone (Fig 15J). Analysis of the eclosion rate demonstrated that TBP-36Q moderately rescued the severe papal lethality (Fig 15K). Although the eclosion rate of fly expressiong p53 and TBP-109Q slightly higher than that of fly expressing p53 alone, but it does not reach a statistically difference (Fig 15K). Pan-neuronal overexpression of p53 causes early lethality and adult flies can not survive beyond 4 days after eclosion. Expression of TBP-36Q slightly lengthened the lifespan of transgenic flies expressing p53 (Fig S1A), but it did not improve the motor function (Fig S1B).. Deactivation of p53 ameliorates polyQ mediated toxicity To investigate the loss-of-function effect of p53 in Drosophila model of polyQ protein diseases, transgenic flies that expressing the dominant negative form of endogenous p53 (DN-Dmp53) were crossed with flies expressing mutant human TBP and Htt. Light microscopy revealed that mutant TBP induced progressive degmentation phenotype was inhibits by overexpression of DN-Dmp53 (Fig. 13A vs. 13B). Likewise, Htt97Qinduced retinal degeneration was markedly alleviated when DN-Dmp53 was coexpressed (Fig. 16C vs. Fig. 16D). SEM photographs clearly demonstrated that the eye degeneration resulted from expression of TBP109Q or Htt97Q was rescued by coexpression of DN-Dmp53 (Fig 17A, 17B and 17C). - 26.

(28) Confocal analysis of phalloidin-stained retina demonstrated that flies coexpressing the DN-Dmp53 with TBP109Q or Htt97Q display better retinal architecture with distinguishable photoreceptor neurons (Fig. 17F and 17H vs. Fig. 17E and 17G). Neuronal overexpression of DN-Dmp53 extended lifespan of transgenic flies expressing Htt97Q or TBP109Q (Fig. 17I). Overexpression of DN-Dmp53 alone showed normal external and internal eye morphology (Fig. 18A and 18E vs. Fig. 18B and 18F). Expression of mutant AT3 disrupts ommatidial architecture and trapezoidal arrays of rhabdomeres (Fig. 18C and 18G), whereas flies coexpressing AT3-78Q and DN-Dmp53 showed better ommatidial architecture with distinguishable photoreceptors (Fig. 18D and 18H).. - 27.

(29) Discussion. Previous in vivo studies have provided strong evidence that the toxicity of mutant polyQ proteins is a gain-of-function phenomenon24. SCA17 is a typical polyQ disease because it is a general transcription factor which is essential for transcriptional initiation mediated through three major RNA polymerases (RNAP I, II, and III) in eukaryotic nuclei. Tws aspects of pathogenic mechanisms might be elicited by polyQ expansion within TBP: (1) Transcriptional dysfunction resulted from loss of TBP function; (2) Gain of toxicity function as a result of the expanded polyQ tract. Normal human TBP generally contains 25-39 units of polygutamine within the amino terminal region. Alleles with expansion of the polyQ tract beyond a threshold of 43 Gln codons prepresent a disease- causing alleles and lead to pathogenic characteristics. In our Drosophila model of SCA17, we found that TBP-54Q has no toxicity to the neurons and all cell types in eyes and does confer any degenerative phenotype as observed in transgenics misexpressing TBP-109Q. This signified a different endurance level against polyQ- induced toxicity among species. DSP1 interacts directly with human TBP and inhibits transcription initiation. Herein, we investigated the in vivo effects of DSP1 with TBP containing different length of polyglutamine. stretch. The. results. demonstrated that DSP1 disturb the transcription of wild-type human TBP. Note that the in vivo effects were markedly observed in retinal epithelial pigment cells (Fig. 2D) and bristle cells (Fig. 3D), but not in photoreceptor neurons (Fig. 3H), implicating a cell type-specific function of DSP1. Similar - 28.

(30) phenotypes were observed when TBP-54Q and DSP1 were coexpressed (Fig. 4B). HMGB1, a multi-function protein, has been shown to participate in the repression of RNA polymerase II- directed transcription. HMGB1 functions as an architectural protein that facilitate enhanceosome formation by introducing bends into the DNA. Furthermore, HMGB1 can modulate key transcriptional factors in nucleosome remodeling. In our studies, we found that complement of HMGB1 did not alter the normal eye phenotype of transgenic flies expressing TBP36Q or TBP54Q. This suggested that DSP1 and HMGB1 may exhibit diverse functions on transcription. DSP1 binds to wild-type TBP and interferes with transcription. In contrast, the effect of HMGB1’s architectural function promotes enhanceosome assembly and chromatin remodeling is more important and profound than its interaction with TBP as a repressor in transcription of class II genes. In the case of the SCA17 of Drosophila model, we demonstrated that DSP1 and HMGB1 counteract the effects of eye degeneration caused by TBP-109Q. In addition, DSP1 and HMGB1 ameliorate the toxicity induced by TBP-109QNTD, which exhibited stronger toxicity than full-length mutant TBP. Furthermore, DSP1 suppressed the eye degeneration and behavior defect in fly models of HD and SCA3. These results confirmed the interaction between DSP1/HMGB1 and polyQ tract. A possible explanation for the genetic effects of DSP1/HMGB1 against polyglutamine-induced toxicity can be attributed to the fact that expanded polyQ containing proteins bind directly to DSP1/HMGB1, which is sequestered. and. inactivated. by. inclusion,. and. augmentation. of. DSP1/HMGB1 titrate mutant polyQ protein and reduces its effect on general - 29.

(31) transcription factors. Thus, DSP1/HMGB1 reduces the chance that other transcription factors includes TBP being captured into polyQ aggregates and dysfunction of TBP dependent transcription, which has been implicated in many neurodegenerative diseases.. - 30.

(32) References 1. Harding AE. Clinical features and classification of inherited ataxias. Advances in Neurology. 1993;61:1-14. 2. B ü rk K, Bösch S, Globas C, et al. Executive dysfunction in spinocerebellar ataxia type 1. Eur. Neurol. 2001;46(1):43-8. 3. Tang B, Liu C, Shen L, et al. Frequency of SCA1, SCA2, SCA3/MJD, SCA6, SCA7, and DRPLA CAG trinucleotide repeat expansion in patients with hereditary spinocerebellar ataxia from Chinese kindreds. Arch. Neurol. 2000;57(4):540-4. 4. O'Hearn E, Holmes SE, Calvert PC, Ross CA, Margolis RL. SCA-12: Tremor with cerebellar and cortical atrophy is associated with a CAG repeat expansion. Neurology. 2001;56(3):299-303. 5. Nakamura K, Jeong S. [SCA17, a novel polyglutamine disease caused by the expansion of polyglutamine tracts in TATA-binding protein]. Rinsho Shinkeigaku. 2001;41(12):1123-5. 6. Smith JK, Gonda VE, Malamud N. Unusual form of cerebellar ataxia; combined dentato-rubral and pallido-Luysian degeneration. Neurology. 1958;8(3):205-9. 7. Bauer P, Laccone F, Rolfs A, et al. Trinucleotide repeat expansion in SCA17/TBP in white patients with Huntington's disease-like phenotype. J. Med. Genet. 2004;41(3):230-2. 8. Margolis RL. The spinocerebellar ataxias: order emerges from chaos. Current Neurology and Neuroscience Reports. 2002;2(5):447-56. 9. Zühlke C, Dalski A, Schwinger E, Finckh U. Spinocerebellar ataxia type 17: report of a family with reduced penetrance of an unstable Gln49 TBP allele, haplotype analysis supporting a founder effect for unstable alleles and comparative analysis of SCA17 genotypes. BMC Med. Genet. 2005;6:27. - 31.

(33) 10. Tomiuk J, Bachmann L, Bauer C, et al. Repeat expansion in spinocerebellar ataxia type 17 alleles of the TATA-box binding protein gene: an evolutionary approach. Eur. J. Hum. Genet. 2007;15(1):81-7. 11. Choudhry S, Mukerji M, Srivastava AK, Jain S, Brahmachari SK. CAG repeat instability at SCA2 locus: anchoring CAA interruptions and linked single nucleotide polymorphisms. Hum. Mol. Genet. 2001;10(21):2437-46. 12. MacDonald ME, Ambrose CM, Duyao MP, et al. A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromosomes. Cell. 1993;72(6):971-983. 13. Zühlke C, Hellenbroich Y, Dalski A, et al. Different types of repeat expansion in the TATA-binding protein gene are associated with a new form of inherited ataxia. Eur. J. Hum. Genet. 2001;9(3):160-4. 14. Silveira I, Miranda C, Guimarães L, et al. Trinucleotide repeats in 202 families with ataxia: a small expanded (CAG)n allele at the SCA17 locus. Arch. Neurol. 2002;59(4):623-9. 15. Koide R, Kobayashi S, Shimohata T, et al. A neurological disease caused by an expanded CAG trinucleotide repeat in the TATA-binding protein gene: a new polyglutamine disease? Hum. Mol. Genet. 1999;8(11):2047-53. 16. Toyoshima Y, Yamada M, Onodera O, et al. SCA17 homozygote showing Huntington's disease-like phenotype. Ann. Neurol. 2004;55(2):281-6. 17. Friedman MJ, Wang C, Li X, Li S. Polyglutamine expansion reduces the association of TATA-binding protein with DNA and induces DNA binding-independent neurotoxicity. J. Biol. Chem. 2008;283(13):8283-90. 18. Bruni AC, Takahashi-Fujigasaki J, Maltecca F, et al. Behavioral disorder, dementia, ataxia, and rigidity in a large family with TATA box-binding protein mutation. Arch. Neurol. 2004;61(8):1314-20.. - 32.

(34) 19. Lasek K, Lencer R, Gaser C, et al. Morphological basis for the spectrum of clinical deficits in spinocerebellar ataxia 17 (SCA17). Brain. 2006;129(Pt 9):2341-52. 20. Rolfs A, Koeppen AH, Bauer I, et al. Clinical features and neuropathology of autosomal dominant spinocerebellar ataxia (SCA17). Ann. Neurol. 2003;54(3):367-75. 21. Perez MK, Paulson HL, Pendse SJ, et al. Recruitment and the role of nuclear localization in polyglutamine-mediated aggregation. J. Cell Biol. 1998;143(6):1457-70. 22. Davies SW, Beardsall K, Turmaine M, et al. Are neuronal intranuclear inclusions the common neuropathology of triplet-repeat disorders with polyglutamine-repeat expansions? Lancet. 1998;351(9096):131-3. 23. Orr AL, Huang S, Roberts MA, et al. Sex-dependent effect of BAG1 in ameliorating motor deficits of Huntington disease transgenic mice. J Biol Chem. 2008;283(23):16027-36. 24. Orr HT, Zoghbi HY. Trinucleotide repeat disorders. Annu. Rev. Neurosci. 2007;30:575-621. 25. Schaffar G, Breuer P, Boteva R, et al. Cellular toxicity of polyglutamine expansion proteins: mechanism of transcription factor deactivation. Mol Cell. 2004;15(1):95-105. 26. DiFiglia M, Sapp E, Chase KO, et al. Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain. Science. 1997;277(5334):1990-3. 27. Gutekunst CA, Li SH, Yi H, et al. Nuclear and neuropil aggregates in Huntington's disease: relationship to neuropathology. J Neurosci. 1999;19(7):2522-34. 28. Bichelmeier U, Schmidt T, Hubener J, et al. Nuclear Localization of Ataxin-3 Is Required for the Manifestation of Symptoms in SCA3: In Vivo Evidence. J. Neurosci. 2007;27(28):7418-7428. - 33.

(35) 29. Perez MK, Paulson HL, Pendse SJ, et al. Recruitment and the role of nuclear localization in polyglutamine-mediated aggregation. J. Cell Biol. 1998;143(6):1457-1470. 30. Uchihara T, Fujigasaki H, Koyano S, et al. Non-expanded polyglutamine proteins in intranuclear inclusions of hereditary ataxias--triple-labeling immunofluorescence study. Acta Neuropathol. 2001;102(2):149-152. 31. Huang CC, Faber PW, Persichetti F, et al. Amyloid formation by mutant huntingtin: threshold, progressivity and recruitment of normal polyglutamine proteins. Somat. Cell Mol. Genet. 1998;24(4):217-233. 32. Bustin M, Lehn DA, Landsman D. Structural features of the HMG chromosomal proteins and their genes. Biochim Biophys Acta. 1990;1049(3):231-43. 33. Hardman CH, Broadhurst RW, Raine AR, et al. Structure of the A-domain of HMG1 and its interaction with DNA as studied by heteronuclear three- and four-dimensional NMR spectroscopy. Biochemistry. 1995;34(51):16596-607. 34. Agresti A, Bianchi ME. HMGB proteins and gene expression. Curr Opin Genet Dev. 2003;13(2):170-8. 35. Landsman D, Bustin M. A signature for the HMG-1 box DNA-binding proteins. Bioessays. 1993;15(8):539-46. 36. Bustin M, Reeves R. High-mobility-group chromosomal proteins: architectural components that facilitate chromatin function. Prog Nucleic Acid Res Mol Biol. 1996;54:35-100. 37. Bustin M. Regulation of DNA-dependent activities by the functional motifs of the high-mobility-group chromosomal proteins. Mol. Cell. Biol. 1999;19(8):5237-46. 38. Bianchi ME, Falciola L, Ferrari S, Lilley DM. The DNA binding site of HMG1 protein is composed of two similar segments (HMG boxes), both of - 34.

(36) which have counterparts in other eukaryotic regulatory proteins. EMBO J. 1992;11(3):1055-63. 39. Love JJ, Li X, Case DA, et al. Structural basis for DNA bending by the architectural transcription factor LEF-1. Nature. 1995;376(6543):791-5. 40. Werner MH, Huth JR, Gronenborn AM, Clore GM. Molecular basis of human 46X,Y sex reversal revealed from the three-dimensional solution structure of the human SRY-DNA complex. Cell. 1995;81(5):705-14. 41. Grosschedl R, Giese K, Pagel J. HMG domain proteins: architectural elements in the assembly of nucleoprotein structures. Trends Genet. 1994;10(3):94-100. 42. Bianchi ME, Beltrame M, Paonessa G. Specific recognition of cruciform DNA by nuclear protein HMG1. Science. 1989;243(4894 Pt 1):1056-9. 43. Locker D, Decoville M, Maurizot JC, Bianchi ME, Leng M. Interaction between cisplatin-modified DNA and the HMG boxes of HMG 1: DNase I footprinting and circular dichroism. J Mol Biol. 1995;246(2):243-7. 44. Ferrari S, Harley VR, Pontiggia A, et al. SRY, like HMG1, recognizes sharp angles in DNA. EMBO J. 1992;11(12):4497-506. 45. P-ohler JR, Norman DG, Bramham J, Bianchi ME, Lilley DM. HMG box proteins bind to four-way DNA junctions in their open conformation. EMBO J. 1998;17(3):817-26. 46. Giese K, Cox J, Grosschedl R. The HMG domain of lymphoid enhancer factor 1 bends DNA and facilitates assembly of functional nucleoprotein structures. Cell. 1992;69(1):185-95. 47. Pontiggia A, Rimini R, Harley VR, et al. Sex-reversing mutations affect the architecture of SRY-DNA complexes. EMBO J. 1994;13(24):6115-24. 48. Boonyaratanakornkit V, Melvin V, Prendergast P, et al. High-mobility group chromatin proteins 1 and 2 functionally interact with steroid hormone receptors to enhance their DNA binding in vitro and transcriptional activity - 35.

(37) in mammalian cells. Mol Cell Biol. 1998;18(8):4471-87. 49. Melvin VS, Roemer SC, Churchill MEA, Edwards DP. The C-terminal extension (CTE) of the nuclear hormone receptor DNA binding domain determines interactions and functional response to the HMGB-1/-2 co-regulatory proteins. J Biol Chem. 2002;277(28):25115-24. 50. Melvin VS, Edwards DP. Coregulatory proteins in steroid hormone receptor action: the role of chromatin high mobility group proteins HMG-1 and -2. Steroids. 1999;64(9):576-86. 51. Verrijdt G, Haelens A, Schoenmakers E, Rombauts W, Claessens F. Comparative analysis of the influence of the high-mobility group box 1 protein on DNA binding and transcriptional activation by the androgen, glucocorticoid, progesterone and mineralocorticoid receptors. Biochem J. 2002;361(Pt 1):97-103. 52. Butteroni C, De Felici M, Schöler HR, Pesce M. Phage display screening reveals an association between germline-specific transcription factor Oct-4 and multiple cellular proteins. J. Mol. Biol. 2000;304(4):529-40. 53. Zwilling S, König H, Wirth T. High mobility group protein 2 functionally interacts with the POU domains of octamer transcription factors. EMBO J. 1995;14(6):1198-208. 54. Zappavigna V, Falciola L, Helmer-Citterich M, Mavilio F, Bianchi ME. HMG1 interacts with HOX proteins and enhances their DNA binding and transcriptional activation. EMBO J. 1996;15(18):4981-91. 55. Jayaraman L, Moorthy NC, Murthy KG, et al. High mobility group protein-1 (HMG-1) is a unique activator of p53. Genes Dev. 1998;12(4):462-72. 56. McKinney K, Prives C. Efficient specific DNA binding by p53 requires both its central and C-terminal domains as revealed by studies with high-mobility group 1 protein. Mol Cell Biol. 2002;22(19):6797-808. 57. Stros M, Ozaki T, Bacikova A, Kageyama H, Nakagawara A. HMGB1 - 36.

(38) and HMGB2 cell-specifically down-regulate the p53- and p73-dependent sequence-specific transactivation from the human Bax gene promoter. J. Biol. Chem. 2002;277(9):7157-64. 58. Brickman JM, Adam M, Ptashne M. Interactions between an HMG-1 protein and members of the Rel family. Proc. Natl. Acad. Sci. U.S.A. 1999;96(19):10679-83. 59. Decoville M, Giraud-Panis MJ, Mosrin-Huaman C, Leng M, Locker D. HMG boxes of DSP1 protein interact with the rel homology domain of transcription factors. Nucleic Acids Res. 2000;28(2):454-62. 60. Ellwood KB, Yen YM, Johnson RC, Carey M. Mechanism for specificity by HMG-1 in enhanceosome assembly. Mol Cell Biol. 2000;20(12):4359-70. 61. Carrozza MJ, DeLuca N. The high mobility group protein 1 is a coactivator of herpes simplex virus ICP4 in vitro. J. Virol. 1998;72(8):6752-7. 62. Ge H, Roeder RG. The high mobility group protein HMG1 can reversibly inhibit class II gene transcription by interaction with the TATA-binding protein. J. Biol. Chem. 1994;269(25):17136-40. 63. Das D, Scovell WM. The binding interaction of HMG-1 with the TATA-binding protein/TATA complex. J. Biol. Chem. 2001;276(35):32597-605. 64. Sutrias-Grau M, Bianchi ME, Bernués J. High mobility group protein 1 interacts specifically with the core domain of human TATA box-binding protein and interferes with transcription factor IIB within the pre-initiation complex. J. Biol. Chem. 1999;274(3):1628-34. 65. Dasgupta A, Scovell WM. TFIIA abrogates the effects of inhibition by HMGB1 but not E1A during the early stages of assembly of the transcriptional preinitiation complex. Biochim. Biophys. Acta. 2003;1627(2-3):101-10. 66. Qi M, Tagawa K, Enokido Y, et al. Proteome analysis of soluble nuclear - 37.

(39) proteins reveals that HMGB1/2 suppress genotoxic stress in polyglutamine diseases. Nat. Cell Biol. 2007;9(4):402-14. 67. Lehming N, Thanos D, Brickman JM, et al. An HMG-like protein that can switch a transcriptional activator to a repressor. Nature. 1994;371(6493):175-9. 68. Canaple L, Decoville M, Leng M, Locker D. The Drosophila DSP1 gene encoding an HMG 1-like protein: genomic organization, evolutionary conservation and expression. Gene. 1997;184(2):285-90. 69. Mosrin-Huaman C, Canaple L, Locker D, Decoville M. DSP1 gene of Drosophila melanogaster encodes an HMG-domain protein that plays multiple roles in development. Dev Genet. 1998;23(4):324-34. 70. Siebenlist U, Franzoso G, Brown K. Structure, regulation and function of NF-kappa B. Annu Rev Cell Biol. 1994;10:405-55. 71. Steward R. Dorsal, an embryonic polarity gene in Drosophila, is homologous to the vertebrate proto-oncogene, c-rel. Science. 1987;238(4827):692-4. 72. Gilmore TD, Koedood M, Piffat KA, White DW. Rel/NF-kappaB/IkappaB proteins and cancer. Oncogene. 1996;13(7):1367-78. 73. Ip YT, Reach M, Engstrom Y, et al. Dif, a dorsal-related gene that mediates an immune response in Drosophila. Cell. 1993;75(4):753-63. 74. Dushay MS, Asling B, Hultmark D. Origins of immunity: Relish, a compound Rel-like gene in the antibacterial defense of Drosophila. Proc Natl Acad Sci U S A. 1996;93(19):10343-7. 75. Barillas-Mury C, Charlesworth A, Gross I, et al. Immune factor Gambif1, a new rel family member from the human malaria vector, Anopheles gambiae. EMBO J. 1996;15(17):4691-701. 76. Lehming N, Le Saux A, Schüller J, Ptashne M. Chromatin components - 38.

(40) as part of a putative transcriptional repressing complex. Proc Natl Acad Sci U S A. 1998;95(13):7322-6. 77. Kim LK, Choi UY, Cho HS, et al. Down-regulation of NF-kappaB target genes by the AP-1 and STAT complex during the innate immune response in Drosophila. PLoS Biol. 2007;5(9):e238. 78. Kirov NC, Lieberman PM, Rushlow C. The transcriptional corepressor DSP1 inhibits activated transcription by disrupting TFIIA-TBP complex formation. EMBO J. 1996;15(24):7079-87. 79. Johnson AD. The price of repression. Cell. 1995;81(5):655-8. 80. Tyree CM, George CP, Lira-DeVito LM, et al. Identification of a minimal set of proteins that is sufficient for accurate initiation of transcription by RNA polymerase II. Genes Dev. 1993;7(7A):1254-65. 81. Ozer J, Moore PA, Bolden AH, et al. Molecular cloning of the small (gamma) subunit of human TFIIA reveals functions critical for activated transcription. Genes Dev. 1994;8(19):2324-35. 82. Stargell LA, Struhl K. The TBP-TFIIA interaction in the response to acidic activators in vivo. Science. 1995;269(5220):75-8. 83. Sun X, Ma D, Sheldon M, Yeung K, Reinberg D. Reconstitution of human TFIIA activity from recombinant polypeptides: a role in TFIID-mediated transcription. Genes Dev. 1994;8(19):2336-48. 84. Yokomori K, Zeidler MP, Chen JL, et al. Drosophila TFIIA directs cooperative DNA binding with TBP and mediates transcriptional activation. Genes Dev. 1994;8(19):2313-23. 85. Goodrich JA, Tjian R. TBP-TAF complexes: selectivity factors for eukaryotic transcription. Curr Opin Cell Biol. 1994;6(3):403-9. 86. Fields S, Jang SK. Presence of a potent transcription activating sequence in the p53 protein. Science. 1990;249(4972):1046-1049.. - 39.

(41) 87. Raycroft L, Wu HY, Lozano G. Transcriptional activation by wild-type but not transforming mutants of the p53 anti-oncogene. Science. 1990;249(4972):1049-1051. 88. Vogelstein B, Kinzler KW. p53 function and dysfunction. Cell. 1992;70(4):523-526. 89. Ginsberg D, Mechta F, Yaniv M, Oren M. Wild-type p53 can down-modulate the activity of various promoters. Proc. Natl. Acad. Sci. U.S.A. 1991;88(22):9979-9983. 90. Mercer WE, Shields MT, Lin D, Appella E, Ullrich SJ. Growth suppression induced by wild-type p53 protein is accompanied by selective down-regulation of proliferating-cell nuclear antigen expression. Proc. Natl. Acad. Sci. U.S.A. 1991;88(5):1958-1962. 91. Santhanam U, Ray A, Sehgal PB. Repression of the interleukin 6 gene promoter by p53 and the retinoblastoma susceptibility gene product. Proc. Natl. Acad. Sci. U.S.A. 1991;88(17):7605-7609. 92. Seto E, Usheva A, Zambetti GP, et al. Wild-type p53 binds to the TATA-binding protein and represses transcription. Proc. Natl. Acad. Sci. U.S.A. 1992;89(24):12028-12032. 93. Chin KV, Ueda K, Pastan I, Gottesman MM. Modulation of activity of the promoter of the human MDR1 gene by Ras and p53. Science. 1992;255(5043):459-462. 94. Shiio Y, Yamamoto T, Yamaguchi N. Negative regulation of Rb expression by the p53 gene product. Proc. Natl. Acad. Sci. U.S.A. 1992;89(12):5206-5210. 95. Martin DW, Subler MA, Muñoz RM, et al. p53 and SV40 T antigen bind to the same region overlapping the conserved domain of the TATA-binding protein. Biochem Biophys Res Commun. 1993;195(1):428-34. 96. Liu X, Miller CW, Koeffler PH, Berk AJ. The p53 activation domain binds the TATA box-binding polypeptide in Holo-TFIID, and a neighboring - 40.

(42) p53 domain inhibits transcription. Mol. Cell. Biol. 1993;13(6):3291-3300. 97. Truant R, Xiao H, Ingles CJ, Greenblatt J. Direct interaction between the transcriptional activation domain of human p53 and the TATA box-binding protein. J. Biol. Chem. 1993;268(4):2284-2287. 98. Thut CJ, Chen JL, Klemm R, Tjian R. p53 transcriptional activation mediated by coactivators TAFII40 and TAFII60. Science. 1995;267(5194):100-104. 99. Lu H, Levine AJ. Human TAFII31 protein is a transcriptional coactivator of the p53 protein. Proc. Natl. Acad. Sci. U.S.A. 1995;92(11):5154-5158. 100. Farmer G, Friedlander P, Colgan J, Manley JL, Prives C. Transcriptional repression by p53 involves molecular interactions distinct from those with the TATA box binding protein. Nucleic Acids Res. 1996;24(21):4281-4288. 101. Crighton D, Woiwode A, Zhang C, et al. p53 represses RNA polymerase III transcription by targeting TBP and inhibiting promoter occupancy by TFIIIB. EMBO J. 2003;22(11):2810-2820. 102. Ryan AB, Zeitlin SO, Scrable H. Genetic interaction between expanded murine Hdh alleles and p53 reveal deleterious effects of p53 on Huntington's disease pathogenesis. Neurobiol. Dis. 2006;24(2):419-427. 103. Shahbazian MD, Orr HT, Zoghbi HY. Reduction of Purkinje cell pathology in SCA1 transgenic mice by p53 deletion. Neurobiol. Dis. 2001;8(6):974-981.. - 41.

(43) Figure 1. DSP1 enhances toxicity caused by TBP loss of function. (A-D): light microscopic images. The normal-eye phenotype observed in the control GMR-GAL4 driver (A). DSP1 overexpression results in mild pigment loss (B). dTBP RNAi induces severe depigmentation and cell death in retina (C), and is critically enhanced when its coexpressed with DSP1, leading to lethality at late pupae stage (D). (E-H): SEM images. Compared - 42.

(44) with the external eye morphology of driver-alone control (E), DSP1 overexpression causes an apparent loss of bristles (F). TBP RNAi causes mild rough eye (G), whereas the phenotype is severely enhanced by DSP1 overexpression (H). Quantitative analysis of eclosion rate (I). Newly eclosed flies produced by each genotype crossing with GMR-GAL4 driver are calculated. For each genotype, at least five cohorts consisted 8-10 virgin flies were tested. Values shown represent mean± SD. * P<0.05, * P<0.001.. - 43.

(45) Figure 2. DSP1 interferes with the function of wild-type hTBP (A-D) Light microscopic images. The normal-eye phenotype observed using wild. type. crossing. homozygous. GMR-GAL4. driver. (A).. DSP1. overexpression causes progressive pigment loss in Drosophila eyes (B). A nearly normal phenotype produced by eyes misexpressing hTBP-36Q (C), which is disturbed when DSP1 was coexprssed (D).. - 44.

(46) Figure 3. Interaction of DSP1 and TBP is cell specific in Drosophila (A-D): SEM images. Driver-alone control shows normal-eye (A). Overexprssion of DSP1 causes approximately total bristle loss at 3 week after eclosion (B). Transgenic misexpressing hTBP-36Q did not affect eye development (C), which is disrupted by DSP1 overexpression (D). (E-H): whole mounts retina staining with rhodamine-phalloidin conjugate to - 45.

(47) rhabdomeres of photoreceptor neurons (single tangential confocal section). At this apical section, normally, clusters of seven photoreceptors within each ommatidium form a characteristic chevron-shaped structure as driver-alone control (E). Moderate abnormality in ommatidial polarity was detected with the up-regulation of DSP1 (F). Transgenic flies misexpressing hTBP36Q producea normal retinal morpgology (G). Flies coexpressing DSP1 and hTBP36Q display a relatively normal ommatidial arrays with mild polarity defect (H). (I) Quantitative analysis of the rhabdomere numbers per ommatidium in flies corresponding to the colored boxed panels. In each group, more than 100 ommatidia were analyzed by directly counting non-atrophic rhabdomeres in magnified pictures taken by confocal microscopy.. - 46.

(48) Figure 4. Interaction of DSP1 and polyglutamine- expanded TBP. (A-D) Light microscopic images. Transgenic flies misexpressing TBP-54Q display a normal eye phenotype (A). Coexpression of TBP-54Q and DSP1 results in depigmentation (B). Progressive loss of pigmentation induced by TBP109Q is mild enhanced when DSP1 was coexpressed(C).. - 47.

(49) - 48.

(50) Figure 5. DSP1 suppresses retinal degeneration induced by mutant TBP A-D: SEM images. Mutant TBP-54Q exhibits no toxicity on external structure of eyes (A). The normal phenotype of TBP-54Q is interrupted by DSP1 overexpression, leading to a severe deficit in bristle numbers (B). Transgenic misexpressing TBP-109Q produces a severe rough-eye phenotype (C), which is ameliorated when coexpressed with DSP1 (D). G-H: Confocal images of adult retina stained with rhodamine-phalloidin (red). The eyes expressing hTBP-54Q display a largely normal trapezoidal array of rhabdomeres (G), whereas coexpression of hTBP-54Q and DSP1 gives rise to mild polarity defects (H). Ommatidial disorganization and cell loss are apparent in the eyes misexpressing TBP-109Q (I) are ameliorated by DSP1 overexpression (J). Quantitative analysis of fusion number. For each group, more than 10 adult flies are analysed by directly counting fused ommatidia in magnified pictures taken by SEM. Quantitative analysis of the photoreceptor numbers per ommatidium.. - 49.

(51) Figure 6. DSP1 rescues polyglutamine-induced pupae lethality. Overexpression of TBP-109QNTDdriven by GMR-GAL4 causes severe pupae lethality. (none of males survives to adulthood) (A). The dual TBP-109QNTD/DSP1 transgenic flies demonstrates ~ 95% eclosion rate (B). Quantitative analysis of eclosion rate exhibits significant increase in flies coexpressing TBP-109QNTD and DSP1 compared with flies expressing TBP-109QNTD alone (C).. - 50.

(52) Figure 7. DSP1 represses TBP-109QNTD toxicity in eyes. (A ,D and G) SEM images. (B,E and H) 1000X magnified pictures taken by SEM. (C,F and I) Immunohistologic staining of whole-mount retina, DSP1 overexpression results in severe bristle loss (A) but ordered ommatidia array and external eye morphology were not affected (B) with relatively normal retinal architecture(C). Transgenic eyes overexpressing TBP-109QNTD generates severe bristle abnormality (D), fused ommatidia (E), and ommatidial disorganization accompanied with cell death in the internal retina (F). Coexpression of TBP-109QNTD and DSP1 suppresses eye degeneration, restoring bristle numbers (G), ommatidial array (H) and retina morphology (I). Histological analysis revealed the protective effect of DSP1 on TBP-109QNTD -induced eye degeneration (J). - 51.

(53) Figure 8. DSP1 rescues the rough-eye phenotype of transgenic HD and SCA3 fly models Overexpression of MJD-78Q driven by GMR-GAL4 promotes severe depigmentation and cell death (A). Up-regulation of DSP1 mitigates the loss of pigmentation induced by mutant AT3 (B). Eyes of transgenic flies misexpressing Htt-97Q show severe depigmentation (C). DSP1 ameliorates the pigmentation loss caused by mutant Htt (D).. - 52.

(54) Figure 9. Neuronal overexpression of DSP1 alleviates mutant polyQ protein induced motor dysfunction and pre-matured death. Climbing assays demonstrates that neuronal expression of Htt97Q (yellow bar) and MJD78Q (dark green bar) causes progressive locomotor deficits. Expression of DSP1 improved motor function of flies expressing Htt97Q (red bar) MJD78Q (light green bar) flies at each time point (A). Lifespan analysis: The lifespan of transgenic flies overexpressing Htt97Q or MJD78Q were dramatically reduced as compared with that of control driver line. Dsp1 - 53.

(55) also extended the lifespan of flies expressing Htt97Q or MJD78Q (B).. - 54.

(56) Figure 10. Genetic interaction of HMGB1 and TBP36Q/54Q. Compared with the control driver (A), flies expressing HMGB1 have normal phenotype (B). Flies coexpressing HMGB1 and TBP36Q (D) or TBP54Q (F) showed normal phenotype compared with that expressing TBP36Q (C) or TBP54Q alone (E).. - 55.

(57) - 56.

(58) Figure 11. HMGB1 suppresses internal retinal degeneration induced by polyQ- expanded TBP. (A-D). Longitudinal views of adult retina stained with phalloidin. Internal degeneration can be reflected in the thickness of the retina, indicated by the white double-headed arrows. Compared with driver-alone control (A), flies misexpressing HMGB1 have a normal phenotype (B). Misexpression of TBP-109Q results in severe reduction in retinal thickness (C). Flies coexpressing HMGB1 with TBP-109Q dramatically increase retinal thickness (D). (E-H). Tangential views of adult retina stained with phalloidin. Compared with diver-alone control (E), flies expressing HMGB1 show a normal trapezoidal array of rhabdomeres (F). Misexpression of TBP-109Q induces severe degeneration, with loss of photoreceptor cells and disorganized retinal morphology (G). Coexpression of HMGB1 and TBP-109Q markedly restores the number of photoreceptors and ommatidial arrays (H). (J) Distribution of ommatidia in 14 day flies. Transgenic flies coexpressing HMGB1 and TBP109Q display ommatidia with mostly 5, 6 and 7 photoreceptors, whereas flies expressing TBP109Q alone have ommatidia with 2~4 photoreceptors.. - 57.

(59) Figure 12. HMGB1 and DSP1 inhibit TBP109Q-induced cell death in adult retina. Acridine orange staining of adult eyes at 7 day (A-D) and 14 day (E-H) after eclosion. GMR-GAL4>UAS-TBP109Q eyes show widespread cell death and progressive increase in yellow-green punctuate staining during adult stage (B and F). The eyes of transgenic flies expressing HMGB1 and TBP109Q (C and G), as well as DSP1 and TBP109Q (D and H) exhibit small amounts of dead cells at day 14, and are indistinguishable from the GMR-GAL4 control (A and E). - 58.

(60) Figure 13. HMGB1 suppresses toxicity of TBP109QNTD. Flies expressing HMGB1 exhibit external (B) and internal (F) morphologic defects as compared with driver alone (A and E). The severe phenotypes with sunken ommatidia (C) and disordered hexagonal arrays of ommatidia (G) caused by 109QNTD are markedly rescued when HMGB1 was coexpressed (D and H).. - 59.

(61) Figure 14. HMGB1 rescues retinal degeneration induced by mutant AT3 and Htt. Mutant AT3 induced severe phenotypes with sunken ommatidia (A) and loss of photoreceptor neurons (E) were moderately suppressed by HMGB1 (B and F). Mutant Htt induced retinal degeneration (C and G) was ameliorated in the same way by HMGB1 (D and H).. - 60.

(62) - 61.

(63) Figure 15. Genetic interaction of p53 and wild-type or mutant TBP. (A-E): light microscopic images. Control flies expressing gmr-GAL4 alone display normal eye morphology (A). Flies overexpressing p53 leads to severe degeneration with reduced eye size, loss of pigmentation and collapse of the necrotic retina (B). Compared to flies expressing p53 alone, flies coexpressing TBP36Q moderately rescued eye phenotype (C). Flies coexpressing TBP54Q show similar level of eye degeneration (D) compared with flies expressing p53 alone. Flies coexpressing TBP109Q moderately ameliorate p53-indiced eye degeneration. (F-J): SEM images. The normal phenotype of driver-alone control (F) is disrupted by overexpression of p53, leading to a striking reduction in eye size and destruction of entire ommatidial arrays (G). Coexpression of TBP36Q moderately rescued the eye size but not ommatidal architecture (H). Coexpression of TBP54Q does not evidently affect the severe eye phenotype caused by p53 overexpression (I). Coexpression of TBP109Q mildly increased the eye size (J). Flies coexprssing TBP36Q and p53 significantly increase the eclosion rate as compared with flies expressing p53 alone (K).. - 62.

(64) Figure 16. Deactivation of p53 ameliorates toxicity of mutant polyQ proteins. Compared with progressively depigmentation of fly expressing TBP109Q (A), flies displayed a normal eye phenotype when dominant negative Dmp53 was expressed (B). Flies expressing Htt97Q lead to severe depigmentation (C), whereas coexpreesion of DN-Dmp53 markedly regains the eye pigmentation (D).. - 63.

(65) Figure 17. DN-Dmp53 alleviates mutant TBP and Htt- induced neurodegeneration. The severe rough phenotype with fused ommtidia induced by TBP109Q (A) and disorganized ommatidial arrays (E) were restored by coexpression of dominant negative forms of Dmp53 (B and F). Mutant Htt display stronger - 64.

(66) toxic than that TBP109Q, and produced more fused ommatidia and a greater loss of photoreceptors (C and G) than TBP109Q does. Coexpression of DN-Dmp53 markedly rescued the eye degeneration in external (D) and internal (H) morphology. Neuronal overexpression of HMGB1 and Htt97Q or TBP109Q extended lifespan as compared with fly expressing TBP109Q or Htt97Q.. - 65.

(67) Figure 18. DN-Dmp53 represses toxicity of mutant AT3. Compared with driver-alone control (A and E), flies expressing DN-Dmp53 exhibited normal phenotype (B and F). Mutant AT3 induced loss of ommatidial boundary (C) and misshaped trapezoidal arrays (G) were rescued when DN-Dmp53 was coexpressed (D and H).. - 66.

(68) Figure. S1.. Locomotor. ability. and. lifespan. of. pan-neuronally. coexpression of human p53 and TBP. The transgenic flies expressing p53 and TBP36Q survives slightly longer as compared with flies expressing p53 alone (A). Flies expressing p53 alone or coexpressing TBP with different lengths of polyQ showed severe locomotor defects right after eclosed (B).. - 67.

(69) Figure S2. Genetic interaction of dominant negative p53 and TBP (A-D) light microscopic images. Compared with driver-alone control (A), overexpression of Dmp53 showed a normal eye phenotype (B). Coexpression of Dmp53 with TBP36Q (C) or TBP54Q (D) also showed - 68.

(70) normal eye phenotype. (E-F) SEM images The normal eye morphology of driver-alone control (E) does not effect by overexpression of DN-Dmp53 (F). Coexpression of DN-Dmp53 and TBP36Q (G) or TBP54Q (H) both showed normal eye phenotype.. - 69.

(71)