Selective constraints and expression pattern of the ataxin 3 like 1 retrogene

Selective constraints and expression pattern of the ataxin-3 like 1 retrogene Andreia Patrícia de Sousa Pinto Mestrado em Genética Forense Departamento de Biologia 2013 Orientador Doutora Sandra Martins, Investigadora Post-doc no Instituto de Patologia e Imunologia Molecular da Universidade do Porto (IPATIMUP) Todas as correções determinadas pelo júri, e só essas, foram efetuadas. O Presidente do Júri, Porto, ______/______/_________ FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 1 Agradecimentos Quero começar por prestar os meus agradecimentos à primeira pessoa responsável pela minha jornada no IPATIMUP, o Professor Amorim, por quem eu nutro uma grande admiração, não só pelo seu incontestável mérito profissional como pela sua personalidade única e de moderada irreverência. O maior agradecimento deve ser entregue à minha orientadora, Sandra Martins, entusiasta, dedicada, persistente, apaixonada, prestável, paciente… Por todas estas características fantásticas e pelo elevado peso que tem tido na minha educação, ensinando-me mais sobre Ciência e sobre o que é ser cientista. A todos os elementos do grupo de Genética Populacional do IPATIMUP que atenderam a diversas dúvidas e necessidades de ajuda por que passei e também pelas chamadas de atenção nos momentos em que errava, sempre de forma informal e a transparecer companheirismo. Um especial obrigado à Alexandra Lopes pela sua colaboração neste trabalho. À Dra. Susana Seixas pela sua colaboração e ajuda dispensada em benefício deste trabalho. Ao IPATIMUP pelo financiamento e apoio científico sem o qual o sucesso do trabalho não teria sido possível. Aos colegas de Genética Forense 2011/2012 pelo companheirismo e aos alunos de mestrado desse mesmo ano pela receção acolhedora e por todo o acompanhamento para mim indispensável e irretribuível, ficando inevitavelmente em dívida para com vocês. Um especial obrigado à Inês Martins por me ter orientado nos primeiros passos do meu trabalho. Às minhas grandes amigas da Generation Y Dance Crew que fazem questão de me acompanhar em todos os meus passos que não sejam só os de dança, mostrando-se sempre orgulhosas de mim em todas as minhas conquistas e obrigando-me a ser sempre melhor em tudo para terem um bom exemplo a seguir. Aos meus pais que me permitem estudar, um “luxo” que lhes foi retirado aos 10 anos de idade, e engrandecer na vida sempre com humildade. 2 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene E por fim, mas não menos importante, à minha avó que no passado ano, já no seu leito de morte, me confessou que levaria consigo um desgosto: ser analfabeta. Talvez por isso, sabia melhor que ninguém a importância da educação e foi, outrora, um dos grandes incentivos a que eu, ainda na escola primária, definisse o meu grande objetivo de vida: obter formação superior na área das Ciências Biológicas, que em tão inocente idade eu entendia como - “quero ir para a universidade para cuidar dos animais”. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 3 Abstract The ataxin-3 gene (ATXN3; 14q32.1) encodes a ubiquitously expressed deubiquitinating enzyme, conserved throughout eukaryotes, with homologues among metazoans and found also in plants and protozoans. In humans, ATXN3 may expand and encode an abnormally long polyglutamine stretch, polyQ, responsible for the most common dominant ataxia worldwide: Machado-Joseph disease (MJD/SCA3). The expanded protein gains toxic properties but the normal ataxin-3 seems also to influence the disease pathogenesis. ATXN3 has two poorly studied paralogues, originated by transposition events: ataxin-3 like (ATXN3L; Xp22.2) and LOC100132280 (8q23.2), here named ATXN3L1 and ATXN3L2, respectively. Interestingly, a recent in vitro study has shown the ability of ATXN3L1 to cleave ubiquitin substrates, with its protease domain significantly more efficient than the ATXN3 domain. Regarding ATXN3L2, in silico analyses previously performed in our lab predicted only short reading frames for its translation. Still, if transcribed, these short mRNA molecules may have functional relevance by regulating the expression of the parental gene, as demonstrated for other transcripts originated from retrocopies. Our aims were (1) to study the mechanisms that led to the human-specific (CAG)n expansion in the parental ATXN3, but not in other lineages or paralogues; (2) to estimate the onset of the retrotransposition events; (3) to compare the rates of evolution and selective constraints among all three genes in the primate lineage; and (4) to assess the mRNA expression pattern of the retrocopy that conserved an intact ORF (ATXN3L1) in human tissues. We have found a ATXN3L1 CAG repeat poorly polymorphic and highly interrupted across the primate lineage, whereas a pure (CAG)n was observed in ATXN3L2 of most species, with humans and chimpanzees showing a polymorphic (CGGCAG)n. Instability encompassing this hexanucleotide may have resulted from a CAG to CGG mutation occurred after the Gorilla-Pan split, followed by unequal crossover. We have also found evidence of unequal crossover events in ATXN3 of orangutan. Phylogenetic results suggested that ATXN3L1 and ATXN3L2 arose by two independent retrotransposition events, in Haplorrhini (~63 MYA), and before the PlatyrrhiniCatarrhini split (~43 MYA), respectively. Branch models performed indicated that ATXN3L1 has been under selective constraints throughout primate evolution, reinforcing its functional relevance, whereas ATXN3L2 gained premature stop codons that likely turned it into a processed pseudogene. In fact, we confirmed by Reverse Transcriptase PCR, that ATXN3L1 is transcriptionally-active in humans, with mRNA expression in testis, placenta, brain, spleen, and 4 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene thymus. Further studies will be done to confirm the presence of endogenous protein in vivo and to better understand the functional diversification of ATXN3L1. Keywords: Ataxin-3; Retrotransposition; Paralogue genes; Machado-Joseph disease; Evolution FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 5 Resumo O gene ataxin-3 (ATXN3; 14q32.1) codifica uma enzima de expressão ubíqua pertencente ao grupo das desubiquitinases. A ataxina-3 é conservada nos eucariotas, com ortólogos descritos nos metazoários, em plantas e, também, em protozoários. Em humanos, o gene ATXN3 pode apresentar uma região repetitiva expandida que codifica uma proteína poliglutamínica mais longa do que o normal. A doença de Machado-Joseph (DMJ/SCA3), considerada a ataxia dominante mais comum em todo o mundo, é causada por ganho de função da proteína expandida, mas a ataxina-3 normal parece ser um modificador da apresentação clínica da doença. O gene ATXN3 tem dois genes parálogos ainda pouco estudados, originados a partir de eventos de retrotransposição: ataxin-3 like (ATXN3L; Xp22.2) e LOC100132280 (8q23.2), denominados por nós como ATXN3L1 e ATXN3L2, respetivamente. Relativamente a ATXN3L1, é interessante referenciar um estudo in vitro realizado recentemente que demonstra a capacidade da ATXN3L, através do seu domínio proteolítico, quebrar ligações entre ubiquitinas mais eficazmente do que a ataxina-3 parental. Quanto a ATXN3L2, análises in silico já realizadas por nós, demonstraram a aquisição de vários codões stop prematuros, o que se reflete na existência de open reading frames (ORFs) muito interrompidas. Ainda assim, não se pode negligenciar a importância deste parálogo pois, tal como demonstrado para outros transcritos orginados a partir de retrocópias, as pequenas moléculas de mRNA podem desempenhar funções ao nível da regulação da expressão do gene parental. Os nossos objetivos foram (1) o estudo dos mecanismos que levaram à expansão do (CAG)n no gene ATXN3 humano; (2) a determinação da origem dos eventos de retrotransposição que levaram ao surgimento dos dois parálogos em estudo; (3) a comparação das taxas de evolução e das pressões seletivas dos 3 genes em estudo ao longo das diversas linhagens de primatas; e (4) a determinação do padrão de expressão do gene ATXN3L1 em tecidos humanos, uma vez que se apresenta como o gene mais conservado. De acordo com os nossos resultados, o motivo repetitivo CAG do gene ATXN3L1 apresenta-se pouco polimórfico e altamente interrompido ao longo das linhagens de primatas, enquanto que o gene ATXN3L2 possui um motivo repetitivo puro em quase todas as espécies estudadas, excetuando, por exemplo, os humanos e os chimpanzés que apresentam o motivo (CGGCAG)n. A instabilidade inerente a este hexanucleotídio poderá ter resultado a partir de uma mutação de CAG para CGG logo após a divergência gorila-chimpanzé, seguida de recombinação desigual. Análises filogenéticas sugerem dois eventos de retrotransposição 12 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene Figure 19: Schematic representation of the unequal crossing over events that may have led to the present (CAG)n tract in A. ATXN3L2 of Pan troglodytes; and B. ATXN3 of Pongo abelii. ..... 66 Figure 20: Phylogenetic tree showing the species where duplications of ATXN3L1 seem to have occurred. .......................................................................................................................................... 68 Figure 21: ATXN3L1 (L1) and ATXN3L1 duplicate (dL1) haplotypes for A. positions c.561T>G, c.743A>G, c.923A>C and c.995C>T in M. fuscata individuals; and B. c.801A>C and c.995C>T in M. fascicularis males. .......................................................................................................................... 69 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 13 Abbreviations A – Adenine APS - ammonium persulfate ATXN1 – Ataxin-1 ATXN1L – Ataxin-1 like ATXN3 – Ataxin-3 ATXN3L1 – Ataxin-3 like 1 ATXN3L2 – Ataxin-3 like 2 bp – base pairs C – Citosine CAMP - Cyclic adenosine monophosphate CBP - CREB-binding protein cDNA – complementary DeoxyriboNucleic Acid CREB - cAMP response element-binding protein dL1 - Ataxin-3 like 1 duplicate DNA – DeoxyriboNucleic Acid dNTP – deoxyNucleotide TriPhosphate DRPLA – Dentatorubral-pallidoluysian atrophy DUB – DeUbiquitinating enzyme ExoFastAP – Exonuclease I and Thermosensitive Alkaline Phosphatase G – Guanine gDNA – genomic DeoxyriboNucleic Acid HAT – Histone Acetiltransferase HD – Huntington’s Disease HDAC - Histone deacetilases Herv – Human Endogenous Retroviruses JD – Josephin Domain JOSD1 – Josephin domain containing protein 1 JOSD2 – Josephin domain containing protein 2 L1 – Long interspersed nuclear element 1 LINE – Long Interspersed Nucleotide Element LTR – Long Terminal Repeat min - minutes 14 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene MJD – Machado-Joseph Disease MMP – Matrix MetalloProteinase mRNA – messenger Ribonucleic Acid MYA – Million Years Ago NCBI – National Center for Biotechnology Information NcoR - Nuclear receptor co-repressor 1 NII – Nuclear inclusions ORF – Open Reading Frame PCAF - P300/CBP-associated factor PCR – Polymerase Chain Reaction PolyQ – poly glutamine RNA – Ribonucleic Acid rpm – rotations per minute RSP - restriction site polymorphisms RT – Reverse Transcriptase SBMA - Spinal and Bulbar Muscular Atrophy SCA – SpinoCerebellar Ataxia sec - seconds SINE – Short Interspersed Nucleotide Element SNP – Single Nucleotide Polymorphism STR – Short Tandem Repeat T – Timine TEMED – Tetramethylethylenediamine TSD – Target Site Duplication Ub – Ubiquitin UCSC –University of California, Santa Cruz UIM – Ubiquitin Interaction Motif UPP – Ubiquitin-proteasome pathway UTR - Untranslated Region Xp – short arm chromosome X Yq – long arm chromosome Y FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 15 Introduction FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 17 1. Origin of new genes The emergence of new genes is fundamental to the evolution of species-specific traits [1]. The major mechanism providing raw material for the origin of new genes is gene duplication. There are two types of gene duplication: direct duplication of genomic DNA and retropositional events [2]. In the first case, gene copies present often the same exon-intron organization and similar expression patterns when compared to parental genes: duplication of chromosomal segments [1]. Gene duplicates may, however, lose their core promoters and their protein-coding potential if tandem duplication or uneven crossing-over is in their origin [3]. In the case of retrotransposition, re-integration of reverse transcribed mRNA molecules occurs in the genome [4]. This is the mechanism by which our genes of interest have arisen, and for this reason, in next topics, we will explain it in more detail. 1.1 Retrotransposition events Transposable elements have contributed greatly to what we now realize to be a highly dynamic genome. An example are the retrotransposons which encode a reverse transcriptase (RT) activity and move by a “copy and paste” process involving an RNA intermediate [5]. L1 elements are autonomous non-LTR (Long Terminal Repeat) retrotransposons classified as LINEs (Long Interspersed Nucleotide Elements) and are widely present in mammalian genomes (about 25% of their genome), being responsible for a burst on the number of retrocopies in Retrotransposons Autonomous LTR (Long Terminal Repeat) LINES (Long Interspersed Nucleotide Elements) Non-autonomous SINES (Short Interspersed Nucleotide Elements) Figure 1: Classification of the different types of retrotransposons. 18 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene mammalian lineages [4]. Besides duplicating themselves, L1 elements likely have been responsible for the genomic expansion of the non-autonomous retrotransposons SINES (Short Interspersed Nucleotide Elements) since they do not encode any proteins and nor possess an independent mobility. Among SINEs, we can find Alu elements, processed pseudogenes, and other elements in the human genome [5]. A. B. Figure 2: Structure of mammalian transposable elements. A. L1 consists of a 5’ untranslated region (5’UTR), two ORFs (Open Reading Frames) separated by a short intergenic region, a 3’UTR, a polyA signal (AATAAA), and a polyA tail (A(n)). L1 elements are often flanked by 7–20 bp target site duplications (TSD). In ORF2, both RT (reverse transcriptase) and EN (endonuclease) domains, as well as a conserved cysteine-rich motif (C) are annotated; B. Alu elements contain two similar sequences, the left (L) and right monomers (R) and end in a polyA tail. (Adapted from Ostertag, 2001) [5]. Retrocopies lack many of the genetic features of their parental genes, such as introns and regulatory elements [6]; for a long time, this has contributed to the idea that only non-functional duplicate gene copies could result from gene retrotransposition. Subsequent analyses suggested, however, that retroposition had efficiently sown the seeds of evolution in genomes [7]; and more recent genome-wide searches of retroposed genes have confirmed the importance of gene retrotransposition in the origin of new functional genes in fly and primates [8-14]. In humans, although selection generally acted against the insertion and high transcription of retrocopies located inside other genes, it favored the emergence of a substantial number of new genes with diverse gene structures and functions [1]. Retrogenes have for long time been described as pseudogenes, however, since raising evidence of retrocopies functionality, the term retrogene has now been widely used to describe FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 19 functional retrocopied genes. On the other hand, pseudogenes consist in genomic DNA sequences lacking the protein-coding capability of their paralogous counterpart because of frameshifts or premature stop codons [15]. Thus, the term “pseudogene” implies nonfunctionality, frequently viewed as a molecular fossil and used to measure background genomic substitution rates [16, 17]. Retrogenes may directly inherit promoters from their parental transcripts as has been observed for individual cases [18], but they often use the regulatory elements of host genes that surround the insertion site (suggesting that retrocopy transcription is frequently driven by the machinery of nearby genes). In fact, regions surrounding retrogenes are transcriptionally more active than regions flanking silent retrocopies. New retrogenes may also emerge from truncated regions coding small RNAs that may regulate other genes expression [1]. The transcription of a retrogene is not, however, sufficient evidence of biological function. Additional confirmation of the functionality of these retrocopies is their ancient origin and strong conservation throughout evolution [15]. The analysis of conserved retrogenes has shown that 50% are indeed conserved across millions of years of primate evolution (far fewer are though to be conserved between species more distant to human, such as rodents) [3]. 1.1.1 Retrotransposons as genetic markers As revealed by pioneering studies on humans, primates and other non-primate groups, retrotransposons afford several advantages that make them very powerful tools as genetic markers for studying human and non-human primate evolutionary histories [19-22]. When compared to retrotransposons, the most often used genetic markers (such as single nucleotide polymorphisms (SNPs), restriction site polymorphisms (RSPs) and short tandem repeats (STRs)) suffer from higher probabilities of homoplasy and greater difficulty of confidently establishing the ancestral marker state. On the other hand, LINE and SINE insertions have two uniquely valuable properties as markers for phylogenetic and population genetic analyses: first, they are virtually free of homoplasy since every observed insertion of the same element sequence is identical by descent; second, the ancestral state of the locus is known to be the absence of the insertion [23]. Retrotransposon insertion polymorphisms have also being used as forensic tools, for example, in species specific DNA detection and quantitation, analysis of complex biomaterials, human gender determination and inference of geographic origin of human samples [24]. 20 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 2. ATXN3 paralogues Ataxin-3 (ATXN3; 14q32.2) was described as containing 11 exons, resulting in at least four different mRNAs due to alternative splicing. More recently, two additional exons (6a and 9a) and 56 alternative transcripts have been described [25, 26]. The (CAG)n tract, located in exon 10, displays usually the (CAG)2 CAA AAG CAG CAA (CAG)n configuration [27]. In most transcripts, including the widely frequent ATXN3-001, this repeat is in the 5’ part of the exon, with a single adenine nucleotide before the reiteration site, at the beginning of the exonic region [28]. Orthologues and paralogues are types of homologous genes that are related by speciation or duplication, respectively [29, 30]. The ATXN3 gene has two paralogues in the genome of several primates. Both ataxin-3 like (Xp22.2) and LOC100132280 (8q23.2) (here named ataxin3 like 1, ATXN3L1; and ataxin-3 like 2, ATXN3L2, respectively) are intronless copies of ATXN3, which hint retrotransposition for their origin. In humans, ATXN3 and ATXN3L1 share 85% sequence identity in their catalytic Josephin domains and high similar crystal structures [31]. A. B. Figure 3: Comparison of the ATXN3 and ATXN3L1 structures. A. Schematic representation of ATXN3 and ATXN3L1 proteins. Josephin domains are shown in light gray, the UIMs are shown in dark gray, and the polyglutamine (polyQ) repeat regions are shown as cross-hatched. The sizes of the UIMs and polyQ regions are not shown to scale. The scale at top shows length in amino acids; B. Superposition of the ATXN3L1 crystal structure (blue) and free ATXN3 solution structure. (Adapted from Weeks et al., 2011) [31]. The other ATXN3 paralogue, ATXN3L2 shares about 70% homology with the parental gene, in humans. This homologous region presents an interrupted ORF, suggesting that this is a processed pseudogene of ATXN3 [32]. Regarding that the two human copies of ATXN3 described above remain poorly studied, it is of great importance to better characterize these retrocopies and to further explore, in the case of ATXN3L1, its potential role in the pathogenesis of Machado-Joseph Disease. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 21 3. J D proteins Josephin Domain (JD) containing proteins are the smallest family of deubiquitinating enzymes (DUBs), all sharing a highly conserved JD catalytic domain. They are found in a wide range of phylogenetic groups, including protists, plants, and metazoans. The term MJD class has also been used after the identification of the neurodegenerative Machado-Joseph disease (MJD), caused by the best-studied protein of this class, ATXN3. This protein family comprises two defined subgroups named ataxins and Josephins, based on the sequence homology of their JD. In almost all species, Josephins (JOSD1 and JOSD2) consist of JD alone, while ataxins (ATXN3 and ATXN3L1) contain an additional C-terminus with two or three ubiquitin interaction motifs (UIMs) and a polyQ tract [33]. Little is known about ATXN3L, JOSD1 and JOSD2, and only very recent studies have provided some insights into their characterization [31, 34]. 3.1 Ataxin-3 Ataxin-3 is the best-characterized JD protein [26]. It presents an approximate molecular weight of 40-43 kDa in normal individuals and structurally is composed by a globular N-terminal Josephin domain with a papain-like fold, combined with an unstructured C-terminus that contains two or three UIMs and the polymorphic polyQ tract [33]. The catalytic pocket consists of a glutamine (Q9) and a cysteine (C14) residue located in the N-terminal part of the JD, and a histidine (H119) and an asparagine (N134) in the JD C-terminal part [26]. When ATXN3 is not bound to other cellular components, its structure is much less compact than when interacting with other cellular partners, which stabilizes the fold of the C-terminus [35]. The N-terminal domain and the catalytic cysteine residue consist mainly of alpha helices, while the catalytic histidine and the orienting glutamine residue are found in a beta-strand context [36]. 28 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene function result in increased histone acetylation of specific target gene promoters, probably leading to altered gene expression in MJD [99, 100]. This repressor activity is dependent on its ubiquitin interaction motifs, suggesting a link between ATXN3 function in the ubiquitin/proteasome pathway and its role in transcriptional regulation [61]. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 29 Objectives FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 31 Given the ongoing evidence of retrogenes functionality, we considered of great importance to characterize ATXN3 retrocopies, since in addition to the interest in the biological processes on their origin, they may play a role in the pathogenesis of MJD. Therefore, our main objectives were:  to address the mechanisms that have led to the human-specific ATXN3 (CAG)n expansion (not observed in other lineages or paralogues);  to determine at which point in primate evolution have the retrotransposition events occurred;  to compare the rates of evolution and selective constraints among ATXN3 and ATXN3L1 in the primate lineage;  to assess the mRNA expression pattern of the retrocopy that conserves an intact ORF (ATXN3L1) in human tissues. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 33 Materials and Methods FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 35 1. DNA samples We amplified and sequenced DNA samples from 10 non-human primate species: Pan troglodytes (n=1), Gorilla gorilla (n=1), Pygmaeus pongus (n=1), Hylobates (n=1), Macaca mulatta (n=3), Macaca fuscata (n=15), Macaca fascicularis (n=1), Cercopithecus aethiops (n=1), Aotus trivirgatus (n=1), and Saguinus oedipus (n=1). Sequences obtained from these samples were analyzed together with others previously assessed by us [32] and with additional ones retrieved from public databases as summarized in tables 1 and 2. Table 2: Sequences retrieved from public databases. Species ATXN3 ATXN3L1 ATXN3L2 Database Accession/Gene ID Database Accession Database Accession H. sapiens Ensembl ENSG00000066427 Ensembl ENSG00000123594 NCBI 100132280 P. troglodytes Ensembl ENSPTRG00000006648 Ensembl ENSPTRG00000028316 UCSC chr8: 109011629-109012743 P. paniscus NCBI 100976906 NCBI 100987250 NA NA G. gorilla Ensembl ENSGGOG00000014166 NCBI 101124367 UCSC chr8: 109925669-109926225 P. abelii Ensembl ENSPPYG00000006075 Ensembl ENSPPYG00000020129 UCSC chr8: 117579303-117579866 N. leucogenys Ensembl ENSNLEG00000017126 Ensembl ENSNLEG00000018731 UCSC GL397267: 24598565-24599123 P. anubis UCSC chr7: 3062 UCSC chrX: 10617308-10618372 UCSC chr8: 105521214-105522300 P. hamadryas NA NA UCSC scaffold3734:61188-61980 NA NA M. mulatta Ensembl ENSMMUG00000020751 Ensembl ENSMMUG00000009370 UCSC chr8: 112963801-112964361 S. boliviensis NCBI 101043060 NCBI 101036136 UCSC Genomic JH378207 C. jacchus Ensembl ENSCJAG00000018025 Ensembl ENSCJAG00000023329 UCSC chr19: 6218140-6218664 T. syrichta Ensembl ENSTSYG00000009902 Ensembl ENSTSYG00000003386 NA NA O. garnettii Ensembl ENSOGAG00000008601 NA NA NA NA NA: not available Species N Pan troglodytes 3 Gorilla gorilla 4 Pongo abelii 2 Pygmaeus pongus 1 Hylobates 1 Papio anubis 3 Macaca mulatta 7 Macaca fuscata 15 Macaca fascicularis 4 Cercopithecus aethiops 1 Aotus trivirgatus 1 Callithrix jacchus 3 Saguinus oedipus 1 Table 1: Non-human primate species sequenced in our lab. 36 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 2. Whole genome DNA amplification In order to have enough amount of DNA to complete our study, all species sequenced in this work, excluding Macaca mulatta, were submitted to a random amplification with GenomiPhi V2 amplification kit (GE Healthcare). Following the instructions provided by the manufacturer, 1 μL of DNA was mixed with 9 μL of sample buffer and denatured for 3 min at 95ºC; then, 1 μL of enzyme mix was added to 9 μL of the denatured product and incubated at 30ºC for 90 min; and, finally, to inactivate the enzyme, the amplified product was heated at 65ºC for 10 min. 3. Amplification of ATXN3L1 and ATXN3L2 We started by amplifying the genes of interest, ATXN3L1 and ATXN3L2, with primers and conditions previously optimized in our lab. Based on the alignment of gene orthologues, primers were designed to specifically amplify ATXN3L1 and ATXN3L2 in most primate species [32]. In cases when the available protocol did not work properly, we optimized new conditions. Both previous and newly optimized protocols were carried out in singleplex PCR reactions using 1 μL of genomic DNA, 5 μL of My Taq™ HS MIX (Bioline, London, United Kingdom), 1 μL of each primer (with a final concentration of 2.5 μM), and 2 μL of H2O in a final volume of 10 μL. The reactions were done in 2720 Thermal Cycler (Applied Biosystems) or Thermal Cycler (BioRad). The summary of amplification parameters is described in figure 7 and table 4. Table 3: Primers designed to specifically amplify ATXN3L1 and ATXN3L2 genes. ATXN3L1 ATXN3L2 Primer Sequence Primer Sequence Forward L1F1 CTCTAACTAGGATACCAGCAAAG L2F1 CATTAACCAAAGAAAGTGGGATAC L1F2 GTTTCCTGTGTGCTCAGCACTG L2F1s GTAGGAAAATAAACACAGTGAAAC L1F3 GAAACCAATAGAGAAGATGAA L2F2 CCAGAGTATCAAAGGCTCAGGATC L1F4 AAGATGAGGAGGATTTTCAGAGG L2F3 AGGCTCGCTTTGTGCTCAGCATTG L2F4P7b GAAGAAATCTCTGGAGGGCAG L2P4A GATAAGATAATTTATATGCGA L2P4G GATAAGATAATTTATATGCGG Reverse L1R1 GGAAAAAGTTCTATGGCAAGAGC L2R1 GGAATCCTATGCTGTAATCACAC L1R2 ACTGGTGACTCCTCCTTCTGCCA L2R1s TCACAGCTGCCTGAAAAGTGG L1R3 GTCTAAAAGCCTTATTTCCTCATCT L2R2 GTGGTCAGCCTTTCACATGGATA L1R4P7 CCTKGCATGCTTAACTCAATAG L2R3 TCTATTACCCAGAGTGGAGTGCAG L2R4 TGGTCTCGAACTCCTGGCCTCA L2R5 CTAAAACTCCTCCTTCTGCC FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 37 Primers L2F1s and L2R1s were specifically designed taking into account Saguinus oedipus sequence obtained in the lab. Similarly, L1R4P7 and L2F4P7b are specific primers to Aotus trivirgatus and L2P4A and L2P4G are allele-specific primers to assess the phase of Gorilla gorilla ATXN3L2 sequences. The following figures help to visualize the annealing site of the primers within our genes. A. B. Figure 6: Schematic representation of A. ATXN3L1; and B. ATXN3L2 genes, with distances (bp) between selected primers. Start and stop codons for ATXN3L1 and the aligned homologous codons for ATXN3L2 are also represented. Hold 1 Hold 3 Hold 2 zx 15’ 30’’ 1’30’’ 95ºC 94ºC wºC y’ 10’ 70ºC 72ºC Figure 7: General PCR protocol with length of time (‘- minutes; ‘’ – seconds) and temperatures (°C) for the amplification ATXN3L1 and ATXN3L2 loci. x (annealing temperature) – 59 for ATXN3L1; and 57 for ATXN3L2; y (extension time) – 2 for ATXN3L1, and 1 for ATXN3L2; z (number of cycles) - 40 for ATXN3L1; and 35 for ATXN3L2. 44 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene Figure 9: ATXN3 gene structure and primers insertion location. Transcribed regions are represented in blue, and not-transcribed and UTR regions in grey. (Adapted from Bettencourt et al., 2011) [26]. We performed a duplex PCR amplification by using 5 µL of My Taq™ HS MIX, 0.5 µL of each primer (2.5 µM), 1 µL of our cDNA and 2 µL of H2O to complete the final volume of 10 µL. The PCR conditions included an initial denatured process at 95ºC for 15 min, a final extension at 60ºC for 30 min and 35 intermediate cycles at 94ºC for 30 sec, 59ºC for 90 sec and 72ºC for 30 sec. The GAPDH (glyceraldehyde-3-phosphate dehydrogenase) gene was used as positive control since it is ubiquitously expressed and one of the most commonly used housekeeping genes in comparisons of gene expression data [108]. Finally, we confirmed specific amplifications by polyacrylamide gel electrophoresis followed by silver staining. E7-8F E7-8R FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 45 Results FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 47 1. Evolution of the (CAG)n tract in ATXN3, ATXN3L1 and ATXN3L2 To gain insight into the processes by which the (CAG)n tracts have been accumulating variation throughout the evolution of each paralogue, we analyzed sequences from all primates obtained from databases and in our lab (Table 9). While analyzing the compiled sequences, we observed the (CAG)n tract conserved in the parental gene along the primate lineage, except in the two orangutans (Pongo abelii and Pygmaeus pongus), which presented several interruptions of triplets coding lysine or phenylalanine. In humans (Homo sapiens), chimpanzees (Pan troglodytes and Pan paniscus), gorilla (Gorilla gorilla), baboon (Papio anubis), rhesus monkey (Macaca mulatta), and African green monkey (Cercopithecus aetiops), a (CAG)2 CAA AAG CAG CAA tract is followed by AAG and/or (CAG)n; in cynomolgous monkey (Macaca fascicularis) some point mutations originated the tract (CAG)2 CAA (CAG)2 AAG (CAG)7. Both configurations give rise to an almost pure polyQ tract, with a single lysine (K) interruption (two in gorilla, baboon and African green monkey). The (CAG)n tract in bushbaby has a (CAG)n CAA (CAG)n, followed by a AAG. The protein tract is almost pure in this species presenting only a terminal lysine. Gibbon (Nomascus leucogenys) and marmoset (Callithrix jacchus) present an almost pure polyQ stretch with one and two glutamic acids (E) interruptions, respectively. Squirrel monkey (Saimiri boliviensis) and tarsier (Tarsius syrichta) are the only species analyzed with a pure polyQ stretch. For ATXN3L1, the repetitive region is also conserved among primates, with a (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG tract observed in several species. The resulting putative protein sequence is interrupted, generally with the Q2 E Q K Q5 configuration. Gorilla and orangutan have a slightly different configuration which gives rise to an additional leucine (L) before the last group of four glutamines. The owl monkey (Aotus trivirgatus), squirrel monkey, marmoset, and cotton-top tamarin (Saguinus oedipus) present a highly interrupted tract: GAG CAA GAA CAA (CAG)2 AGG AAA CAA AAG, giving rise to the interrupted polyQ stretch E Q E Q3 R K Q K. Tarsier is the only species presenting a small pure polyQ tract for ATXN3L1 of just four glutamines. The chimpanzee sequence obtained from Ensembl present a different (CAG)n tract from the other 3 individuals sequenced by us, with a CAC instead of a CAA in the 3’-end of the repeat. As for ATXN3L2, translated products have not been considered since it presents a disrupted ORF. Thus, we observed a pure (CAG)n tract in gorilla, orangutan and gibbon, whereas in humans and chimps it is highly interrupted by CGG triplets. For the first time, a hexanucleotide repeat is observed in ataxin-3 paralogues, instead of a trinucleotide pattern. By comparing both 48 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene retrocopies, ATXN3L1 acquired more interruptions in its CAG repeat region than ATXN3L2, which turned it more stable. Accordingly, the almost pure ATXN3L2 tract in species which diverged earlier in primate evolution, and the polymorphic (CGG CAG)n in humans and chimps show the higher instability associated to this locus in comparison to both parental ATXN3 and ATXN3L1 retrogene. *ATXN3L2 presents a disrupted ORF Table 9: ATXN3, ATXN3L1 and ATXN3L2 (CAG)n tracts obtained both from public databases and our sequencing of several primates. N represents the number of chromosomes analyzed. N Sequence N Sequence N Sequence (CAG)2 CAA AAG CAG CAA (CAG)n (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q3 K Qn Q2 E Q K Q5 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG (CAG)2 CAA AAG CAG CAA (CAG)12 Q2 E Q K Q5 Q3 K Q14 (CAG)2 GAA CAG AAG (CAG)2 CAA CAC CAG Q2 E Q K Q3 H Q (CAG)2 CAA AAG CAG CAA (CAG)11 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q3 K Q13 Q2 E Q K Q5 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG (CAG)2 CAA AAG CAA AAG (CAG)4 Q2 E Q K Q5 Q3 K Q K Q4 (CAG)2 GAA CAG AAG CTG CAG (CAA)2 CAG Q2 E Q K L Q4 (CAG)2 CAA AAG CAG CAA (CAG)2 CCG CAA AAG CAG CAA (CAG)2 CCG CAA (CAG)9 (CAG)2 GAA CAG AAG CTG CAG (CAA)2 CAG 4 (CAG)6 Q3 K Q4 P Q K Q4 P Q10 Q2 E Q K L Q4 1 (CAG)7 (CAG)2 CAA AAG CAG CAA (CAG)2 CCG CAA (CAG)2 CAA (CAG)2 CCG CAA (CAG)7 (CAG)2 GAA CAG AAG CTG CAG (CAA)2 CAG Q3 K Q4 P Q6 P Q8 Q2 E Q K L Q4 (CAG)3 CAA (CAG)5 GAG (CAG)3 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q9 E Q3 Q2 E Q K Q5 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q2 E Q K Q5 (CAG)2 CAA AAG CAG CAA (CAG)2 AAG (CAG)7 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG 3 (CAG)7 CAC CAG Q3 K Q4 K Q7 Q2 E Q K Q5 3 (CAG)8 CAC CAG (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q2 E Q K Q5 (CAG)2 CAA (CAG)2 AAG (CAG)7 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG 13 (CAG)6 CAC CAG Q5 K Q7 Q2 E Q K Q5 2 (CAG)7 CAC CAG (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q2 E Q K Q5 (CAG)2 CAA (CAG)2 AAG (CAG)7 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG 5 (CAG)6 CAC CAG Q5 K Q7 Q2 E Q K Q5 3 (CAG)8 CAC CAG (CAG)2 CAA AAG CAG CAA (CAG)13 AAG (CAG)6 (CAG)2 GAA CAG AAG (CAG)2 (CAA)2 CAG Q3 K Q15 K Q6 Q2 E Q K Q5 GAG CAA GAA CAA (CAG)2 AGG AAA CAA AAG E Q E Q3 R K Q K CAG CAA (CAG)5 (CAA)5 CAG CAA (CAG)12 GAG CAA GAA CAA (CAG)2 AGG AAA CAA AAG Q26 E Q E Q3 R K Q K CAG CAA (CAG)5 CAA (CAG)3 GAG (CAG)3 GAG (CAG)3 GAG CAA GAA CAA (CAG)2 AGG AAA CAA AAG Q11 E Q3 E Q3 E Q E Q3 R K Q K GAG CAA GAA CAA (CAG)2 AGG AAA CAA AAG E Q E Q3 R K Q K (CAA)2 (CAG)2 CAG CAA (CAG)2 Q4 Q4 (CAG)5 CAA (CAG)5 AAG Q11 K Tarsius syrichta Otolemur garnettii 1 1 1 Macaca fascicularis Saimiri boliviensis Callithrix jacchus Saguinus oedipus 1 1 6 1 1 11 Pongo abelii Papio anubis 2 1 2 ATXN3L2* ATXN3 36 1 1 2 9 (CAG)6-17 1 1 1 1 3 2 ATXN3L1 Gorilla gorilla Pygmaeus pongus Nomascus leucogenys Hylobates Homo sapiens Pan troglodytes Pan paniscus 98 1 1 4 1 2 (CAG)6 CAC CAG 2 30 1 2 1 Cercopithecus aethiops Aotus trivirgatus Papio hamadryas 2 1 2 Macaca fuscata 2 6 Macaca mulatta CAG (CGGCAG)5 GGG CAG (CGGCAG)3 GGG CAG 6 CAG (CGGCAG)3-9 GGG CAG 58 (CAG)3 (CGG CAG)4-11 (CAG)6 CAC CAG (CAG)6 (CAG)8 (CAG)3 CAA (CAG)2 30 1 1 1 4 GAG (CAG)4 CAA 1 2 GAG (CAG)3 CAA GAG (CAG)3 CAA FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 49 AMELX AMELY AMELY AMELX 2. Sex determination We determined the gender of primates for which this information was lacking. First, we genotyped the androgen receptor gene based on the theoretical basis of observing homozygozity in males and heterozygozity in females, since AR is located in the X chromosome. The successful genotyping of four AR polymorphic markers by capillary electrophoresis was not enough to determine the sex of all the primates (Table 10). Then, we tried to differentiate the allelic states of the X/Y polymorphism in the amelogenin gene by electrophoresis; the presence of one or two fragments 6 bp distance of each other would identify a female or a male, respectively. All 5 samples analyzed by us appeared to present a single length fragment which led us to suspect that our electrophoresis conditions did not allow the correct visualization of the two allelic states of this Indel. Thus, we proceeded with the sequencing of the AMELX/Y amplified products and aligned with reference sequences (Figure 10). The identification of a 6 bp insertion in AMELY, together with additional AMELX/Y specific point variants across the sequence, allowed us to determine the gender of our primate samples as shown in the appendix section (Table A1). CAG STR1 STR3 DXS1194 Cercopithecus aetiops h H h H Hylobates h ND ND H Pygmaeus pongus h h ND h Saguinus oedipus h ND ND ND Macaca fuscata 2 h H H H H – heterozygous sequence; h – homozygous sequence; ND – not determined. A. B. 3. Intraspecific ATXN3L1 duplication events Table 10: Capillary electrophoresis results for four polymorphic markers within or flanking the androgen receptor gene. Figure 10: Alignment of AMELX and AMELY sequences with A. a male chimpanzee control sample; and B. a female owl monkey control sample. 50 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 3. Intraspecific ATXN3L1 duplication events After confirming the gender of the 15 Macaca fuscata samples, we noticed that all individuals showed to be heterozygous for at least one ATXN3L1 position, including all 5 males. In total, we identified 7 different patterns with 4 SNPs as represented in table 11 and figure 11. Results have been confirmed in all specimens by sequencing in both forward and reverse directions. We therefore, hypothesized that a duplication of the ATXN3L1 locus occurred in this species or in a common recent ancestor of Old World monkeys. Interestingly, we noticed that in some sequences there was a different proportion of each nucleotide (Figure 11B). This could be due to preferential annealing of specific primers designed for ATXN3L1 in the amplification or in the sequencing reaction. Yet, the patterns remained constant through the several PCR and sequencing replications. Table 11: Heterozygozity pattern and polymorphic positions of M. fuscata ATXN3L1. Variant Specimen Sex c.561T>G c.743A>G c.923A>C c.995C>T Pattern M. fuscata 1 Female T G A/C C/T 1 M. fuscata 2 Female T G A/C C/T 1 M. fuscata 3 Male T G A C/T 2 M. fuscata 4 Male T A/G A/C C/T 3 M. fuscata 5 Female T G A C/T 2 M. fuscata 6 Male T A/G A/C C/T 3 M. fuscata 7 Female T G A/C C/T 1 M. fuscata 8 Female T G A C/T 2 M. fuscata 9 Female T/G G A C/T 7 M. fuscata 10 Female T G A/C C/T 1 M. fuscata 11 Female T G A C/T 4 M. fuscata 12 Male T G A C/T 2 M. fuscata 13 Female T A/G A/C C/T 6 M. fuscata 14 Female T G A C/T 5 M. fuscata 15 Male T A/G A/C C/T 3 *in bolt are represented alleles with the highest intensity in sequencing. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 51 A. B. Pattern 1 Pattern 2 Pattern 3 Pattern 4 Pattern 5 Pattern 6 Pattern 7 c.561T>G c.743A>G c.923A>C c.995C>T Figure 11: Example of two ATXN3L1 variant positions in M. fuscata. The two sequences represented in each image (A and B) were sequenced at different time courses and by using alternative primers (forward and reverse) to assure the reproducibility of results; A. Heterozygous position c.995C>T in M. fuscata 8; B. Heterozygous position c.923A>C in M. fuscata 10 showing the two nucleotides unequally represented. Figure 12: The different patterns of heterozygozity found in M. fuscata samples while sequencing ATXN3L1. The four nucleotides are distinguished by four different colours (red – adenine; blue – cytosine; green – timine; and yellow – guanine). 52 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene As for the other Macaca we found two heterozygous position in two M. fascicularis males sequenced by us (with c.995C>T corresponding to M. fuscata c.995C>T), although in this case a pattern could not be established due to the low number of analyzed chromosomes (6). On the other hand, our data did not allow us to confirm if this duplication is present or not in M. mulatta since no biallelic states have been found in the four males sequenced. For this species, a single heterozygous position has been observed in a female individual (c.174C>T). We also found a duplication of the ATXN3L1 locus in Papio anubis. Using the Blat tool in UCSC Genome Browser, we found two sequences located in the X chromosome with 92% similarity to human ATXN3L1. By analyzing our sequencing results of three baboon samples, we observed a male with 4 positions with 2 allelic states, and noticed that these heterozygous positions corresponded to the bases varying between UCSC sequences ChrX: 10617308-10618372 and ChrX: 10636574-10637639 (Table 12). We aligned the two P. anubis sequences retrieved from UCSC with all other ATXN3L1 orthologues and confirmed that both clustered with other ATXN3L1 sequences and shared a common node, suggesting a recent duplication event of this locus. Table 12: ATXN3L1 polymorphic positions found in Papio anubis. Origin Sequences Sex Variant c.280C>T c.319A>G c.987A>T c.1006A>G c.1007C>T c.1011C>T UCSC P. anubis (ChrX: 10617308-10618372) Unknown T G T A T C P. anubis (ChrX: 10636574-10637639) Unknown C A A G C T Our samples Baboon 1 Female C/T A/G A/T A/G C/T C/T Baboon 2 Female C/T A/G A/T A/G C/T C/T Baboon 3 Male ND ND A/T A/G C/T C/T ND: not determined FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 53 4. ATXN3L2 in Aotus trivirgatus Several sequences highly homologous to the human ATXN3 CDS available in public databases have been identified by us as independent-origin retrocopies of ATXN3 [32]. When using different sets of primers in the optimization of ATXN3L2 amplification in Aotus trivirgatus, we found two different sequences with high similarity to human ATXN3L2. The multiple sequence alignment followed by neighbor-joining phylogenetic tree allowed us to identify A. trivirgatus ATXN3L2 whereas the second sequence (here named ATXN3L?) showed a closer molecular distance with other recent retrocopies of ATXN3. Indeed, results showed ATXN3L? sharing a more common recent ancestor with parental genes than with ATXN3 paralogues, as we can see in figure 14. Figure 13: Neighbor-Joining phylogenetic tree for ATXN3 and ATXN3L1 showing the two sequences retrieved from UCSC to Papio anubis. 60 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 6.3 Branch-site model To test if some codons of ATXN3L1 have been evolving differently in different lineages, we applied the branch-site model, in which branches on the phylogeny are divided a priori into foreground (ATXN3L1 branch) and background (ancestral ATXN3 branch) and selective pressures are allowed to vary over sites and branches. All sites seem to be constrained or neutrally evolving with ω values always below or equal 1 (Table 15). Given these results, we decided to define the duplication branch as our foreground and all the other branches as background to test whether a faster evolution has happened right after de retrotransposition event. Once more, no significant values were found. Table 15: Parameter estimates and likelihood scores under different branch-site models. Model comparisons of variable ω ratios among sites and branches. Model Initial ω Likelihood (ln) Parameters 1 (fixed) -3614.52 p= 0.69 0.00 0.31 0.00 H0 ωATXN3= 0.15 1.00 0.15 1.00 ωATXN3L1= 0.15 1.00 1.00 1.00 0 -3614.52 p= 0.69 0.00 0.31 0.00 ωATXN3= 0.15 1.00 0.15 1.00 ωATXN3L1= 0.15 1.00 1.00 1.00 1 -3614.44 p= 0.68 0.01 0.30 0.01 ωATXN3= 0.15 1.00 0.15 1.00 ωATXN3L1= 0.15 1.00 1.00 1.00 H1 2 -3614.44 p= 0.68 0.01 0.30 0.01 ωATXN3= 0.15 1.00 0.15 1.00 ωATXN3L1= 0.15 1.00 1.00 1.00 4 -3614.52 p= 0.69 0.00 0.31 0.00 ωATXN3= 0.15 1.00 0.15 1.00 ωATXN3L1= 0.15 1.00 1.00 1.00 5 -3614.44 p= 0.68 0.01 0.30 0.01 ωATXN3= 0.15 1.00 0.15 1.00 ωATXN3L1= 0.15 1.00 1.00 1.00 Models compared df -2Δl Critical 2 values (P=0.05) H0 vs H1 (0) 1 0.00 3.84 H0 vs H1 (1) 0.14 H0 vs H1 (2) 0.14 H0 vs H1 (4) 0.00 H0 vs H1 (5) 0.14 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 61 7. Nucleotide diversity of ATXN3L1 and ATXN3L2 By analyzing intraspecific nucleotide diversity, we observed that for ATXN3L1, gorilla (G. gorilla) presented the highest number of polymorphisms (26) followed by chimpanzee (P. troglodytes) (11), baboon (P. anubis) (6), japanese macaque (M. fuscata) (4), orangutan (P. abelii) (3), and cynomolgous monkey (M. fascicularis) (2). In gorilla, 11 non-synonymous substitutions were observed, one of them leading to a premature stop codon 21 amino acids before the end of the putative protein. From the 11 polymorphisms found in chimpanzee, 10 were registered only due to the reference sequence retrieved from Ensembl Genome Browse (with 6 of them leading to non-synonymous substitutions, including a premature stop codon). Although the high similarity between our three chimpanzee samples, they are not related to the best of our knowledge. Orangutan showed one non-synonymous substitution over the 3 polymorphisms registered. In baboon, japanese macaque and cynomolgous monkey, we were not able disentangle between the condition of ATXN3L1 polymorphisms and sequence difference resulted from the ATXN3L1 duplicate since all species have a highly homologous duplicate of this retrogene. Rhesus monkey (M. mulatta) and marmoset (C. jacchus) presented only one silent and one nonsynonymous polymorphism each, respectively. Considering ATXN3L2, gorilla was the species showing also the highest number of polymorphisms (41 in total), 4 of them indels (3 insertions and 1 deletion). Actually, deletions have also been found in rhesus monkey (which presented 12 polymorphisms in total, the same number as cynomolgous monkeys) and in orangutans (which had 3 additional SNPs). Baboon showed the longest sequence length variant: a 52 bp insertion in one of the 7 chromosomes analyzed (in addition to 5 SNPs). Neither chimpanzee nor japanese macaque presented length polymorphisms, but only 3 SNPs in 7 and 30 chromosomes analyzed, respectively. More details about this analysis can be retrieved from tables A3 and A4 in appendix. 62 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 100 bp 200 bp Lung GAPDH ATXN3L1 Kidney Esophagus Cervix Adipose tissue Liver Heart Colon Bladder Placenta Thyroid Testis Small intestine Prostate Ovary PCR (-) Chimpanzee Brain Human Brain Trachea Thymus Spleen Skeletal muscle RT-PCR (-) Ladder 8. Transcriptional pattern of ATXN3L1 We tested the presence of ATXN3L1 transcripts in cDNA of different human body tissues as well as in chimpanzee brain by using primers designed to specifically amplify ATXN3L1 cDNA. Sequencing of the RT-PCR product obtained from human brain confirmed that amplified products were specific ATXN3L1 sequences, and not the parental ATXN3. Therefore, we observed that, in humans, ATXN3L1 is transcribed in testis, placenta, brain, spleen, and thymus. Testis and placenta were the tissues for which we observed more amplified product (a similar amplification pattern for GAPDH in all tissues discarded the possible heterogeneity in the cDNA quantity among different tissue samples). Although in a very lower intensity, we can still notice the presence of amplified product in trachea, skeletal muscle, prostate, lung, cervix, heart, bladder, and small intestine. Interestingly, we were also able to test transcription of ATXN3L1 in chimpanzee brain. Regarding the sequence conservation in the annealing region of E7-8 primers between human and chimpanzee ATXN3L1, we tested whether a functional gene was likely to be present in this non-human primate. Indeed, we have shown that ATXN3L1 is expressed in chimpanzee brain, although not so intensively as in humans. Figure 18: ATXN3L1 transcriptional status of 20 human tissues and chimpanzee brain using E7-8 primers. GAPDH cDNA was amplified as a control. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 63 Discussion FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 65 Retrotransposition and retrogenes are gaining increasing attention as recent studies have shown that they may play an important role in genome evolution and in the formation of new genes [10]. Studies focusing on the modulation of parental genes by the respective expressed paralogues may be important to gain insight into the mechanisms by which ATXN3 retrocopies (especially ATXN3L1) may acquire functional relevance. In this study, we benefited from the existence of two paralogues of the ataxin-3 gene in the primate lineage, originated through retrotransposition, to analyze the mechanisms of (CAG)n evolution. This repetitive region is shared by all three paralogues, but expanded exclusively in the parental gene (ATXN3) of humans. Thus, by comparing variability patterns in the normal range of three highly similar genes (regarding the CDS or homologous sequence in ATXN3L2), which have been under different selective pressures can allow us the testing of hypotheses on the causes of expansion and disease. Andres et al. (2004) suggested that human capacity for expansion at repetitive loci could be explained by a higher variance and coefficient of variation observed in our species; thus, loci with increased variance may be more likely to expand and give rise to pathogenic alleles [111]. Still, one question remained: why do these disease associate repeats have higher variance than non-expanding loci? While analyzing the (CAG)n tract of our three paralogues in several primates, we noticed that a cassete-like structure is part of the ATXN3 orangutan and ATXN3L2 chimpanzee sequences (Figure 19). These patterns have, more likely, resulted from unequal crossover events than from successive point mutations in a pure (CAG)n. Other less clear structures are found in the ATXN3 repeat of gorilla and marmoset and additional unequal recombination may have happened in other species, but no evidence can be noticed due to a pure nature of the tract. Similarly, unequal exchange of DNA during recombination may have underlain the evolution of other disease-associated repeats and still remain unnoticed due to the lack of interruptions that allow us to detect them. Actually, in polyalanine disorders, in which repetitive tracts are highly interrupted, unequal crossing-over is often mentioned as the main mechanism for the origin of newly sized alleles [112, 113]. A. A1 A2 CAG (CGGCAG)3 GGG CAG CGG CAG (CGGCAG)3 GGG CAG CGG CAG (CGGCAG)3 GGG CAG CGG CAG (CGGCAG)3 GGG CAG CGG 66 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene A3 A4 A5 B. B1 B2 B3 B4 B5 Figure 19: Schematic representation of the unequal crossing over events that may have led to the present (CAG)n tract in A. ATXN3L2 of Pan troglodytes; and B. ATXN3 of Pongo abelii. The red lines represent the chiasmata. CAG (CGGCAG)3 GGG CAG CAG CGGCAG CGGCAG CGG (CGGCAG)3 GGG CAG CAG (CGGCAG)3 GGG CAG (CGGCAG)3 GGG CAG CAG CGG (CAG)2 CAA AAG CAG CAA (CAG)2 CCG (CAG)9 (CAG)2 CAA AAG CAG CAA (CAG)2 CCG (CAG)9 (CAG)2 CAA AAG CAG CAAA (CAG)2 CCG (CAG)9 (CAG)2 CAA AAG CAG CAA (CAG)2 CCG (CAG)9 (CAG)2 CAA AAG CAG CAAA (CAG)2 CCG (CAG)n (CAG)2 CAA AAG CAG CAA (CAG)2 CCG (CAG)9 (CAG)9 CAG (CGGCAG)3 GGG CAG CGG CAG (CGGCAG)3 GGG CAG CGG CAA AAG CAG CAA (CAG)2 CCG (CAG)9 (CAG)2 CAA AAG CAG CAAA (CAG)2 CCG (CAG)n (CAG)2 CAA AAG CAG CAA (CAG)2 CCG (CAG)9 (CAG)9 (CAG)2 CAA AAG CAG CAA (CAG)2 CCG CAA AAG CAG CAA (CAG)2 CCG CAA (CAG)9 (CAG)11 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 67 Our data suggested that the two retrocopies have independent origins: the ATXN3L1 about 63 MYA, in the Haplorrhini clade, and ATXN3L2 about 43 MYA, before the Platyrrhine-Catarrhine split. Currently, a short CAA2 CAG2 is observed in Tarsier parental gene. Similarly, the ATXN3L1 orthologue has a CAG CAA CAG2 motif, but the acquisition of several interruptions seems to have occurred across the primate lineage. In fact, no more than two consecutive CAGs are observed, which turns this repeat less prone to instability. Although the function of the parental ATXN3 polyQ is poorly understood, it is interesting to notice that despite the repeat interruptions, ATXN3L1 of most primates may encode 5 consecutive glutamines in the 3'-end of its polyQ stretch. The ATXN3L2 pseudogene presents currently an almost pure (CAG)n in several species. Uninterrupted CAGs are also common to parental rodents and porcines homologous sequences, mammalian species prior to the second retrotransposition event on the origin of ATXN3L2 [114, 115]. This way, the repetitive tract has probably a pure nature on its origin that has been maintained throughout evolution in this non-functional gene. Some functions have been suggested to ATXN3 polyQ protein as transcriptional regulation, nuclear localization and protein-protein interactions [37]. A recent study has shown that species with high polyQ protein content have a higher number of proteins bearing domains with functions related to phosphatidylinositol (PI) signaling and ubiquitin-directed protein degradation [116]. As we know by then, aggregates containing proteins with an expanded polyQ stretch are ubiquitinated [95]. These facts suggest that (CAG)n tracts can actually be subjected to selective pressures, in order to maintain a possible functional role in the protein context. Interestingly, while completing data on ataxin-3 paralogous with our own sequencing, we were able to sequence a more recent ATXN3 retrocopy in owl monkey (Aotus trivirgatus). In fact, ATXN3 gene has been on the origin of many different retrocopies along the mammalian lineage [32] maybe due to the abundance of L1 retrotransposons in mammals [4, 5]. In addition, we were able to find evidence of duplication events, occurred after the retrotransposition, in M. fuscata, M. fascicularis and P. anubis species. Given the high similarity between ATXN3L1 and the respective duplicate of each species shown in the obtained phylogenetic tree, we suggest that two different duplications have occurred after speciation (Figure 20). Actually, the ATXN3L1 genes of Papio anubis and Macaca species presented more nucleotide differences between them than between ATXN3L1 and the respective duplicate. Thus, it is more parsimonious that two different duplications have occurred than a single event followed by subsequent selective pressures acting similarly on both copies (ATXN3L1 and ATXN3L1 duplicate) of each species. 68 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene In M. fuscata, we tried to distinguish between ATXN3L1 and ATXN3L1 duplicate based on the seven different patterns found through the sequencing of 15 individuals. We started by comparing the heterozygous with homozygous non-variable positions. Then, we compared also ATXN3L1 sequences of other primates to infer the ancestral allele, i.e. the ATXN3L1 allele. In cases where we found different nucleotide proportions in sequencing, the ancestral allele was considered as being in ATXN3L1 sequence and the derived allele in the duplicate sequence. This way, we were able to discern that two heterozygous positions (c.743A>G and c.923A>C) were indeed polymorphic in the duplicate, and one position (c.995C>T) was polymorphic in both ATXN3L1 and ATXN3L1 duplicate (Figure 21 - A). The fluorescence imbalance in the sequencing suggests that the ATXN3L1 duplicate is, as well, located in the X chromosome. We did the same exercise to M. fascicularis and found alleles A and T to be the ancestral alleles of c.801A>C and c.995C>T, respectively (Figure 21 - B). Figure 20: Phylogenetic tree showing the species where duplications of ATXN3L1 seem to have occurred. The red stars represent the duplication events. Hylobates was used as an outgroup. This tree was obtained from Geneious sotware using neighbor-joining method. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 69 A. Pattern 1 (4 females) Pattern 2 (2 males and 2 females) Pattern 3 (3 males) Pattern 4 (1 female) Pattern 5 (1 female) Pattern 6 (1 female) Pattern 7 (1 female) B. Cynomolgous monkey 1 (male) Cynomolgous monkey 2 (male) Regarding the divergent positions in P. anubis, we could not infer haplotype phases for ATXN3L1 and ATXN3L1 duplicate since both sequences retrieved from the UCSC database seem to contain ancestral alleles at different positions (Table 16). T-G-A-T T-G-A-C T-G-A-T T-G-C-C T-G-A-C T-G-A-T T-G-A-C T-G-A-C T-G-A-T T-G-A-T T-G-A-T T-A-C-C T-G-A-T T-G-A-C T-G-A-C T-G-A-C T-G-A-T T-G-A-C T-G-A-T T-G-A-T T-G-A-C T-A-C-C T-G-A-T T-G-A-T G-G-A-T G-G-A-C T-G-A-T T-G-A-T C-T A-T A-C A-T Figure 21: ATXN3L1 (L1) and ATXN3L1 duplicate (dL1) haplotypes for A. positions c.561T>G, c.743A>G, c.923A>C and c.995C>T in M. fuscata individuals; and B. c.801A>C and c.995C>T in M. fascicularis males. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 77 References FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 79 1. Vinckenbosch, N., I. Dupanloup, and H. Kaessmann, Evolutionary fate of retroposed gene copies in the human genome. Proc Natl Acad Sci U S A, 2006. 103(9): p. 3220-5. 2. Ohshima, K., RNA-Mediated Gene Duplication and Retroposons: Retrogenes, LINEs, SINEs, and Sequence Specificity. International journal of evolutionary biology, 2013. 2013. 3. Pink, R.C., et al., Pseudogenes: Pseudo-functional or key regulators in health and disease? Rna, 2011. 17(5): p. 792-798. 4. Pan, D. and L. Zhang, Burst of young retrogenes and independent retrogene formation in mammals. PloS one, 2009. 4(3): p. e5040. 5. Ostertag, E.M. and H.H. Kazazian Jr, Biology of mammalian L1 retrotransposons. Annual review of genetics, 2001. 35(1): p. 501-538. 6. Cordaux, R. and M.A. Batzer, The impact of retrotransposons on human genome evolution. Nature Reviews Genetics, 2009. 10(10): p. 691-703. 7. Brosius, J. and Retroposons--seeds of evolution Science, 1991. 251(4995): p. 753. 8. Sayah, D.M., et al., Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature, 2004. 430(6999): p. 569-573. 9. Babushok, D.V., et al., A novel testis ubiquitin-binding protein gene arose by exon shuffling in hominoids. Genome research, 2007. 17(8): p. 1129-1138. 10. Kaessmann, H., N. Vinckenbosch, and M. Long, RNA-based gene duplication: mechanistic and evolutionary insights. Nature Reviews Genetics, 2009. 10(1): p. 19-31. 11. Betrán, E., et al., Evolution of the phosphoglycerate mutase processed gene in human and chimpanzee revealing the origin of a new primate gene. Molecular biology and evolution, 2002. 19(5): p. 654-663. 12. Burki, F. and H. Kaessmann, Birth and adaptive evolution of a hominoid gene that supports high neurotransmitter flux. Nature genetics, 2004. 36(10): p. 1061-1063. 13. Betrán, E., K. Thornton, and M. Long, Retroposed new genes out of the X in Drosophila. Genome research, 2002. 12(12): p. 1854-1859. 14. Emerson, J., et al., Extensive gene traffic on the mammalian X chromosome. Science, 2004. 303(5657): p. 537-540. 15. Svensson, Ö., L. Arvestad, and J. Lagergren, Genome-wide survey for biologically functional pseudogenes. PLoS computational biology, 2006. 2(5): p. e46. 16. Graur, D., Y. Shuali, and W.-H. Li, Deletions in processed pseudogenes accumulate faster in rodents than in humans. Journal of molecular evolution, 1989. 28(4): p. 279-285. 17. Zhang, Z. and M. Gerstein, Patterns of nucleotide substitution, insertion and deletion in the human genome inferred from pseudogenes. Nucleic acids research, 2003. 31(18): p. 5338-5348. 18. Féral, C., G. Guellaën, and A. Pawlak, Human testis expresses a specific poly (A)-binding protein. Nucleic acids research, 2001. 29(9): p. 1872-1883. 19. PERNA, N.T., et al., Alu insertion polymorphism: a new type of marker for human population studies. Human biology, 1992: p. 641-648. 20. Ryan, S.C. and A. Dugaiczyk, Newly arisen DNA repeats in primate phylogeny. Proceedings of the National Academy of Sciences, 1989. 86(23): p. 9360-9364. 21. Shimamura, M., et al., Molecular evidence from retroposons that whales form a clade within even-toed ungulates. Nature, 1997. 388(6643): p. 666-670. 22. Shedlock, A.M. and N. Okada, SINE insertions: powerful tools for molecular systematics. Bioessays, 2000. 22(2): p. 148-160. 23. Witherspoon, D., et al., Human population genetic structure and diversity inferred from polymorphic L1 (LINE-1) and Alu insertions. Human heredity, 2006. 62(1): p. 30-46. 80 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 24. Ray, D.A., J.A. Walker, and M.A. Batzer, Mobile element-based forensic genomics. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis, 2007. 616(1): p. 24-33. 25. Bettencourt, C., et al., Increased transcript diversity: novel splicing variants of Machado–Joseph Disease gene (ATXN3). neurogenetics, 2010. 11(2): p. 193-202. 26. Bettencourt, C. and M. Lima, Machado-Joseph Disease: from first descriptions to new perspectives. Orphanet J Rare Dis, 2011. 6: p. 35. 27. Kawaguchi, Y., et al., CAG expansions in a novel gene for Machado-Joseph disease at chromosome 14q32. 1. Nature genetics, 1994. 8(3): p. 221-228. 28. Djian, P., J.M. Hancock, and H.S. Chana, Codon repeats in genes associated with human diseases: fewer repeats in the genes of nonhuman primates and nucleotide substitutions concentrated at the sites of reiteration. Proceedings of the National Academy of Sciences, 1996. 93(1): p. 417421. 29. Fitch, W.M., Distinguishing homologous from analogous proteins. Systematic Biology, 1970. 19(2): p. 99-113. 30. Fitch, W.M., Homology: a personal view on some of the problems. Trends in genetics, 2000. 16(5): p. 227-231. 31. Weeks, S.D., et al., Crystal Structure of a Josephin-Ubiquitin Complex EVOLUTIONARY RESTRAINTS ON ATAXIN-3 DEUBIQUITINATING ACTIVITY. Journal of Biological Chemistry, 2011. 286(6): p. 4555-4565. 32. Martins, M., Evolution and Functional Relevance of Ataxin-3 Paralogues, in Secção Autónoma de Ciências da Saúde2012, Aveiro. 33. Tzvetkov, N. and P. Breuer, Josephin domain-containing proteins from a variety of species are active de-ubiquitination enzymes. Biological chemistry, 2007. 388(9): p. 973-978. 34. Seki, T., et al., JosD1, a Membrane-targeted Deubiquitinating Enzyme, Is Activated by Ubiquitination and Regulates Membrane Dynamics, Cell Motility, and Endocytosis. Journal of Biological Chemistry, 2013. 288(24): p. 17145-17155. 35. Masino, L., et al., Domain architecture of the polyglutamine protein ataxin-3: a globular domain followed by a flexible tail. FEBS letters, 2003. 549(1): p. 21-25. 36. Scheel, H., S. Tomiuk, and K. Hofmann, Elucidation of ataxin-3 and ataxin-7 function by integrative bioinformatics. Human molecular genetics, 2003. 12(21): p. 2845-2852. 37. Matos, C.A., S. de Macedo-Ribeiro, and A.L. Carvalho, Polyglutamine diseases: the special case of ataxin-3 and Machado–Joseph disease. Progress in Neurobiology, 2011. 95(1): p. 26-48. 38. Riess, O., et al., SCA3: neurological features, pathogenesis and animal models. The Cerebellum, 2008. 7(2): p. 125-137. 39. Donaldson, K.M., et al., Ubiquitin-mediated sequestration of normal cellular proteins into polyglutamine aggregates. Proceedings of the National Academy of Sciences, 2003. 100(15): p. 8892-8897. 40. Nicastro, G., et al., Understanding the role of the Josephin domain in the PolyUb binding and cleavage properties of ataxin-3. PloS one, 2010. 5(8): p. e12430. 41. Macedo-Ribeiro, S., et al., Nucleocytoplasmic shuttling activity of ataxin-3. PloS one, 2009. 4(6): p. e5834. 42. Chai, Y., et al., Evidence for proteasome involvement in polyglutamine disease: localization to nuclear inclusions in SCA3/MJD and suppression of polyglutamine aggregation in vitro. Human molecular genetics, 1999. 8(4): p. 673-682. 43. Li, F., et al., Ataxin-3 is a histone-binding protein with two independent transcriptional corepressor activities. Journal of Biological Chemistry, 2002. 277(47): p. 45004-45012. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 81 44. Evert, B.O., et al., Ataxin-3 represses transcription via chromatin binding, interaction with histone deacetylase 3, and histone deacetylation. J Neurosci, 2006. 26(44): p. 11474-86. 45. Zhou, L., et al., Ataxin-3 Protects Cells Against H< sub> 2</sub> o< sub> 2</sub>-Induced Oxidative Stress By Enhancing The Interaction Between Bcl-X< sub> l</sub> And Bax. Neuroscience, 2013. 46. Crespo-Barreto, J., et al., Partial loss of ataxin-1 function contributes to transcriptional dysregulation in spinocerebellar ataxia type 1 pathogenesis. PLoS genetics, 2010. 6(7): p. e1001021. 47. Bowman, A.B., et al., Duplication of Atxn1l suppresses SCA1 neuropathology by decreasing incorporation of polyglutamine-expanded ataxin-1 into native complexes. Nature genetics, 2007. 39(3): p. 373-379. 48. Thomas, P.S., et al., Loss of endogenous androgen receptor protein accelerates motor neuron degeneration and accentuates androgen insensitivity in a mouse model of X-linked spinal and bulbar muscular atrophy. Human molecular genetics, 2006. 15(14): p. 2225-2238. 49. Van Raamsdonk, J.M., et al., Loss of wild-type huntingtin influences motor dysfunction and survival in the YAC128 mouse model of Huntington disease. Human molecular genetics, 2005. 14(10): p. 1379-1392. 50. Van Raamsdonk, J.M., et al., Wild-type huntingtin ameliorates striatal neuronal atrophy but does not prevent other abnormalities in the YAC128 mouse model of Huntington disease. BMC neuroscience, 2006. 7(1): p. 80. 51. Dragatsis, I., M.S. Levine, and S. Zeitlin, Inactivation of Hdh in the brain and testis results in progressive neurodegeneration and sterility in mice. Nature genetics, 2000. 26(3): p. 300-306. 52. Auerbach, W., et al., The HD mutation causes progressive lethal neurological disease in mice expressing reduced levels of huntingtin. Human molecular genetics, 2001. 10(22): p. 2515-2523. 53. Martins, S., et al., A multistep mutation mechanism drives the evolution of the CAG repeat at MJD/SCA3 locus. Eur J Hum Genet, 2006. 14(8): p. 932-40. 54. Orr, H.T. and H.Y. Zoghbi, Trinucleotide repeat disorders. Annu. Rev. Neurosci., 2007. 30: p. 575621. 55. Gatchel, J.R. and H.Y. Zoghbi, Diseases of unstable repeat expansion: mechanisms and common principles. Nature Reviews Genetics, 2005. 6(10): p. 743-755. 56. Zühlke, C., et al., Mitotic stability and meiotic variability of the (CAG) n repeat in the Huntington disease gene. Human molecular genetics, 1993. 2(12): p. 2063-2067. 57. Keckarevic, D., et al., The status of SCA1, MJD/SCA3, FRDA, DRPLA and MD triplet containing genes in patients with Huntington disease and healthy controls. Journal of neurogenetics, 2000. 14(4): p. 257-263. 58. Lorenzetti, D., S. Bohlega, and H.Y. Zoghbi, The expansion of the CAG repeat in ataxin-2 is a frequent cause of autosomal dominant spinocerebellar ataxia. Neurology, 1997. 49(4): p. 10091013. 59. Sobue, G., X-linked recessive bulbospinal neuronopathy (SBMA). Nagoya Journal of Medical Science, 1995. 58: p. 95-106. 60. Koide, R., et al., Unstable expansion of CAG repeat in hereditary dentatorubral–pallidoluysian atrophy (DRPLA). Nature genetics, 1994. 6(1): p. 9-13. 61. Carlson, K.M., J.M. Andresen, and H.T. Orr, Emerging pathogenic pathways in the spinocerebellar ataxias. Current opinion in genetics & development, 2009. 19(3): p. 247-253. 62. Gunawardena, S., et al., Disruption of Axonal Transport by Loss of Huntingtin or Expression of Pathogenic PolyQ Proteins in< i> Drosophila</i>. Neuron, 2003. 40(1): p. 25-40. 82 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 63. Muchowski, P.J., et al., Hsp70 and hsp40 chaperones can inhibit self-assembly of polyglutamine proteins into amyloid-like fibrils. Proceedings of the National Academy of Sciences, 2000. 97(14): p. 7841-7846. 64. Yoo, S.-Y., et al., SCA7 knockin mice model human SCA7 and reveal gradual accumulation of mutant ataxin-7 in neurons and abnormalities in short-term plasticity. Neuron, 2003. 37(3): p. 383-401. 65. Watase, K., et al., A Long CAG Repeat in the Mouse< i> Sca1</i> Locus Replicates SCA1 Features and Reveals the Impact of Protein Solubility on Selective Neurodegeneration. Neuron, 2002. 34(6): p. 905-919. 66. Watase, K., et al., Spinocerebellar ataxia type 6 knockin mice develop a progressive neuronal dysfunction with age-dependent accumulation of mutant CaV2. 1 channels. Proceedings of the National Academy of Sciences, 2008. 105(33): p. 11987-11992. 67. Yeh, S., et al., Generation and characterization of androgen receptor knockout (ARKO) mice: an in vivo model for the study of androgen functions in selective tissues. Proceedings of the National Academy of Sciences, 2002. 99(21): p. 13498-13503. 68. Zeitlin, S., et al., Increased apoptosis and early embryonic lethality in mice nullizygous for the Huntington's disease gene homologue. Nature genetics, 1995. 11(2): p. 155-163. 69. Matilla, A., et al., Mice lacking ataxin-1 display learning deficits and decreased hippocampal paired-pulse facilitation. The Journal of neuroscience, 1998. 18(14): p. 5508-5516. 70. Kiehl, T.-R., et al., Generation and characterization of Sca2 (ataxin-2) knockout mice. Biochemical and biophysical research communications, 2006. 339(1): p. 17-24. 71. Crespo-Barreto, J., et al., Partial loss of ataxin-1 function contributes to transcriptional dysregulation in spinocerebellar ataxia type 1 pathogenesis. PLoS Genet, 2010. 6(7): p. e1001021. 72. Li, F., et al., Ataxin-3 is a histone-binding protein with two independent transcriptional corepressor activities. J Biol Chem, 2002. 277(47): p. 45004-12. 73. Limprasert, P., et al., Analysis of CAG repeat of the Machado-Joseph gene in human, chimpanzee and monkey populations: a variant nucleotide is associated with the number of CAG repeats. Hum Mol Genet, 1996. 5(2): p. 207-13. 74. Takiyama, Y., et al., The gene for Machado–Joseph disease maps to human chromosome 14q. Nature genetics, 1993. 4(3): p. 300-304. 75. Jardim, L.B., et al., A survey of spinocerebellar ataxia in South Brazil–66 new cases with Machado-Joseph disease, SCA7, SCA8, or unidentified disease–causing mutations. Journal of neurology, 2001. 248(10): p. 870-876. 76. Vale, J., et al., Autosomal dominant cerebellar ataxia: frequency analysis and clinical characterization of 45 families from Portugal. European Journal of Neurology, 2010. 17(1): p. 124-128. 77. Zhao, Y., et al., Prevalence and ethnic differences of autosomal‐dominant cerebellar ataxia in Singapore. Clinical genetics, 2002. 62(6): p. 478-481. 78. Tang, B., 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. Archives of neurology, 2000. 57(4): p. 540. 79. Van de Warrenburg, B., et al., Spinocerebellar ataxias in the Netherlands Prevalence and age at onset variance analysis. Neurology, 2002. 58(5): p. 702-708. 80. Schöls, L., et al., Autosomal dominant cerebellar ataxia: phenotypic differences in genetically defined subtypes? Annals of neurology, 1997. 42(6): p. 924-932. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 83 81. Maruyama, H., et al., Difference in disease‐free survival curve and regional distribution according to subtype of spinocerebellar ataxia: A study of 1,286 Japanese patients. American journal of medical genetics, 2002. 114(5): p. 578-583. 82. Kraft, S., et al., Adult onset spinocerebellar ataxia in a Canadian movement disorders clinic. The Canadian Journal of Neurological Sciences, 2005. 32(4): p. 450-458. 83. Moseley, M.L., et al., Incidence of dominant spinocerebellar and Friedreich triplet repeats among 361 ataxia families. Neurology, 1998. 51(6): p. 1666-1671. 84. Alonso, E., et al., Distinct distribution of autosomal dominant spinocerebellar ataxia in the Mexican population. Movement disorders, 2007. 22(7): p. 1050-1053. 85. Storey, E., et al., Frequency of spinocerebellar ataxia types 1, 2, 3, 6, and 7 in Australian patients with spinocerebellar ataxia. American journal of medical genetics, 2000. 95(4): p. 351-358. 86. Saleem, Q., et al., Molecular analysis of autosomal dominant hereditary ataxias in the Indian population: high frequency of SCA2 and evidence for a common founder mutation. Human genetics, 2000. 106(2): p. 179-187. 87. Bryer, A., et al., The hereditary adult-onset ataxias in South Africa. Journal of the neurological sciences, 2003. 216(1): p. 47-54. 88. Brusco, A., et al., Molecular genetics of hereditary spinocerebellar ataxia: mutation analysis of spinocerebellar ataxia genes and CAG/CTG repeat expansion detection in 225 Italian families. Archives of neurology, 2004. 61(5): p. 727. 89. Coutinho, P. and C. Andrade, Autosomal dominant system degeneration in Portuguese families of the Azores Islands A new genetic disorder involving cerebellar, pyramidal, extrapyramidal and spinal cord motor functions. Neurology, 1978. 28(7): p. 703-703. 90. Carvalho, D.R., et al., Homozygosity enhances severity in spinocerebellar ataxia type 3. Pediatric neurology, 2008. 38(4): p. 296-299. 91. Sobczak, K. and W.J. Krzyzosiak, Patterns of CAG repeat interruptions in SCA1 and SCA2 genes in relation to repeat instability. Hum Mutat, 2004. 24(3): p. 236-47. 92. Maciel, P., et al., Correlation between CAG repeat length and clinical features in MachadoJoseph disease. American journal of human genetics, 1995. 57(1): p. 54. 93. Maruyama, H., et al., Molecular features of the CAG repeats and clinical manifestation of Machado-Joseph disease. Human molecular genetics, 1995. 4(5): p. 807-812. 94. Takiyama, Y., et al., Evidence for inter-generational instability in the CAG repeat in the MJD1 gene and for conserved haplotypes at flanking markers amongst Japanese and Caucasian subjects with Machado-Joseph disease. Human molecular genetics, 1995. 4(7): p. 1137-1146. 95. Todi, S.V., et al., Cellular turnover of the polyglutamine disease protein ataxin-3 is regulated by its catalytic activity. Journal of Biological Chemistry, 2007. 282(40): p. 29348-29358. 96. Mueller, T., et al., CK2-dependent phosphorylation determines cellular localization and stability of ataxin-3. Human molecular genetics, 2009. 18(17): p. 3334-3343. 97. Tzvetkov, N. and P. Breuer, Josephin domain-containing proteins from a variety of species are active de-ubiquitination enzymes. Biol Chem, 2007. 388(9): p. 973-8. 98. Burnett, B., F. Li, and R.N. Pittman, The polyglutamine neurodegenerative protein ataxin-3 binds polyubiquitylated proteins and has ubiquitin protease activity. Hum Mol Genet, 2003. 12(23): p. 3195-205. 99. Evert, B.O., et al., Inflammatory genes are upregulated in expanded ataxin-3-expressing cell lines and spinocerebellar ataxia type 3 brains. The Journal of Neuroscience, 2001. 21(15): p. 53895396. 84 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 100. Evert, B.O., et al., Gene expression profiling in ataxin-3 expressing cell lines reveals distinct effects of normal and mutant ataxin-3. Journal of Neuropathology & Experimental Neurology, 2003. 62(10): p. 1006-1018. 101. Santos, D., Mecanismos mutacionais associados aos alelos (CAG)n normais do gene receptor de androgénio, in Biology Department2010, Porto. 102. Morrill, B.H., L.F. Rickords, and H.J. Schafstall, Sequence length polymorphisms within primate amelogenin and amelogenin‐like genes: usefulness in sex determination. American journal of primatology, 2008. 70(10): p. 976-985. 103. Yang, Z., Likelihood ratio tests for detecting positive selection and application to primate lysozyme evolution. Molecular biology and evolution, 1998. 15(5): p. 568-573. 104. Bielawski, J.P. and Z. Yang, Maximum likelihood methods for detecting adaptive evolution after gene duplication, in Genome Evolution. 2003, Springer. p. 201-212. 105. Nielsen, R. and Z. Yang, Likelihood models for detecting positively selected amino acid sites and applications to the HIV-1 envelope gene. Genetics, 1998. 148(3): p. 929-936. 106. Yang, Z., et al., Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics, 2000. 155(1): p. 431-449. 107. Yang, Z. and R. Nielsen, Codon-substitution models for detecting molecular adaptation at individual sites along specific lineages. Molecular biology and evolution, 2002. 19(6): p. 908-917. 108. Barber, R.D., et al., GAPDH as a housekeeping gene: analysis of GAPDH mRNA expression in a panel of 72 human tissues. Physiological genomics, 2005. 21(3): p. 389-395. 109. Jobling, M.A., M. Hurles, and C. Tyler-Smith, Human evolutionary genetics: origins, peoples & disease. 2004. p. 204. 110. Goldman, N. and Z. Yang, A codon-based model of nucleotide substitution for protein-coding DNA sequences. Molecular biology and evolution, 1994. 11(5): p. 725-736. 111. Andres, A.M., et al., Comparative genetics of functional trinucleotide tandem repeats in humans and apes. J Mol Evol, 2004. 59(3): p. 329-39. 112. Arai, H., et al., De novo polyalanine expansion of PHOX2B in congenital central hypoventilation syndrome: unequal sister chromatid exchange during paternal gametogenesis. Journal of human genetics, 2007. 52(11): p. 921-925. 113. Brown, L.Y. and S.A. Brown, Alanine tracts: the expanding story of human illness and trinucleotide repeats. TRENDS in Genetics, 2004. 20(1): p. 51-58. 114. Madsen, L.B., et al., Identification of the porcine homologous of human disease causing trinucleotide repeat sequences. Neurogenetics, 2007. 8(3): p. 207-218. 115. Bakalian, A., N. Delhaye-Bouchaud, and J. Mariani, Quantitative analysis of the Purkinje cell and the granule cell populations in the cerebellum of nude mice. Journal of neurogenetics, 1995. 9(4): p. 207-218. 116. Schaefer, M.H., E.E. Wanker, and M.A. Andrade-Navarro, Evolution and function of CAG/polyglutamine repeats in protein–protein interaction networks. Nucleic acids research, 2012. 40(10): p. 4273-4287. 117. McCole, R.B., et al., A CASE‐BY‐CASE EVOLUTIONARY ANALYSIS OF FOUR IMPRINTED RETROGENES. Evolution, 2011. 65(5): p. 1413-1427. FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene 85 Appendix 92 FCUP Selective constraints and expression pattern of the ataxin-3 like 1 retrogene A – adenine; C – cytosine; G – guanine; T – timine; MAF - maximum allele frequency. Table A5 (continuation): Nucleotide diversity found in ATXN3L2. c.49_101indel 49_101 indel 1/6 0.86 c.102T>G 102 T/G 4/3 0.57 c.373T>G 373 T/G 6/1 0.86 c.556T>G 556 T/G 6/1 0.86 c.987A>G 987 A/G 1/6 0.86 c.1049T>G 1049 T/G 2/5 0.71 c.29_31AAG-del 28_31 delAAG 1/14 0.93 c.36 A>G 36 A/G 1/14 0.93 c.73A>G 73 A/G 7/8 0.53 c.261C>T 261 C/T 1/14 0.93 c.267A>G 267 A/G 14/1 0.93 c.341A>C 341 A/C 10/5 0.67 c.420A>G 420 A/G 1/14 0.93 c.680A>G 680 A/G 1/14 0.93 c.740C>G 740 C/G 14/1 0.93 c.795A>G 795 A/G 14/1 0.93 c.864A>G 864 A/G 10/5 0.67 c.1015C>T 1015 C/T 6/9 0.6 c.575A>G 575 A/G 25/5 0.83 c.579A>G 579 A/G 4/26 0.87 c.581C>T 581 C/T 29/1 0.97 c.73A>G 73 A/G 4/4 0.63 c.134A>G 134 A/G 3/5 0.63 c.180A>C 180 A/C 1/7 0.88 c.341A>C 341 A/C 4/4 0.63 c.471C>T 471 C/T 3/5 0.63 c.495A>G 495 A/G 5/3 0.63 c.507A>G 507 A/G 4/4 0.63 c.549A>G 549 A/G 5/3 0.63 c.627T>G 627 T/G 3/5 0.63 c.821C>T 821 C/T 4/4 0.63 c.864A>G 864 A/G 4/4 0.63 c.1048A>G 1048 A/G 1/7 0.88 M. fuscata M. fascicularis 15 30 8 M. mulatta P. anubis 7