Sarcoplasmic reticulum Ca2+ dynamics in aging Drosophila and correlation with sarcopenia

PROGRAMA DE DOCTORADO EN INVESTIGACIÓN BIOMÉDICA TESIS DOCTORAL Sarcoplasmic reticulum Ca2+ dynamics in aging Drosophila and correlation with sarcopenia Presentada por Alba Del Río Lorenzo para optar al grado de Doctor/a por la Universidad de Valladolid Dirigida por: Dr. Javier García-Sancho Martín Dra. María Teresa Alonso Alonso Valladolid, 2020 A mis padres Javier y Milagros “Educar la mente sin educar el corazón no es educar en absoluto” Aristóteles Para empezar una tesis hace falta valentía, vocación y motivación, pero para acabarla hace falta perseverancia, constancia, disciplina, fuerza y paciencia. Por tanto, en primer lugar quiero agradecer a mis directores de tesis, a Maite y a Javier, por haberme dado esta oportunidad cuando llegué a Valladolid llena de esa vocación y esa motivación. Además agradecerles que me han mantenido llena de ilusión y ganas de aprender ciencia a lo largo de estos años. Siempre serán mis “padres científicos”. Por otro lado, para afrontar el día a día, hace falta todo un equipo que encaje a nivel científico y personal, y hay que decir no podría haber tenido uno mejor. Miriam, que me abrió las puertas el día que llegué, que me ha enseñado a manejarme en el laboratorio y que me ha ayudado como una madre a nivel personal. A Jonathan, que es otro padre científico para mí y que siempre ha dejado de hacer sus cosas por prestarme su incalculable ayuda y atención. A Paloma, que siempre ha sido un referente para mí, ya que, desde que llegué quería alcanzar el lugar al que ella había llegado. A Macarena, que ha sabido entenderme a la perfección a nivel personal y científico y con la que he pasado momentos inolvidables. A Raquel, con la que anduve los primeros pasos de la mano y que resultó ser un ejemplo de trabajo y bondad. A Jesús, por estar siempre dispuesto a ayudar y por los buenos momentos de las tardes de prácticas. A Karla y a Patry con las que he aprendido a enseñar, y que me han sabido ayudar, animar y llevar en la etapa más difícil del camino, el final. Gracias a las dos, por vuestro cariño y por todo lo que me habéis hecho crecer a nivel emocional. Otro ingrediente importante para poder afrontar el día a día son los amigos, y yo no puedo haberme rodeado de mejores personas en esta etapa. Todos juntos, hemos sido, sin duda, todo un equipazo que ha sabido apoyarse en los momentos duros y divertirse como los que más siempre. Sin duda, sois lo mejor de esta etapa y se me queda corto este manuscrito para agradeceros todo lo que me habéis hecho crecer y sentir. Nagore, eres la mejor amiga que se puede desear. Eres, sin duda, todo un referente personal para mí y nuestra conexión mental es tan increíble que hablamos sin palabras. Alberto, eres un esencial en este camino, una de las personas que más me conoce y cuyo apoyo ha sido siempre infinito. Patri, eres de esas personas únicas, capaz de levantar el ánimo a cualquiera y que ven el interior de los demás. Lucía, capaz de dar hasta que se queda sin nada para ella misma, inagotable es tu fuente de amor por los demás. Eu, eres la persona más fuerte y leal que conozco y precisamente por eso la que más he echado de menos en esta última etapa aunque nuestra amistad es un para siempre. Iván, la persona con la que más conversaciones científicas transcendentales he tenido y cuyos consejos tengo siempre presentes. Miguel, que siempre te recuerda qué es lo más importante en la vida. Laura, mi descubrimiento más tardío pero que me ha ayudado a alcanzar la meta como no lo habría hecho nadie, con sus palabras, sus risas, su amor y su increíble forma de ser. Sergio, que siempre lo ve todo con perspectiva y le pone una nota de humor. Y Nuria, que me recuerda a mis comienzos y cuya risa siempre te alegra el día. Gracias a toda la gente del IBGM, a Lola y Diego por responder a todas mis preguntas de moscas y a Miriam por nuestras charlas en el pasillo y en cultivos. A todo nuestro grupo de comer que siempre me hacen olvidarme de los problemas cotidianos: Sara, Jorge, Elisa, Aida, Sendoa y Vero. Una de las experiencias clave en este periodo es la estancia. Yo tuve la suerte de realizarla en Múnich en el laboratorio de la Dr. Fabiana Perocchi, a la que agradezco que me acogiera y me dedicara parte de su tiempo. Pero lo más importante de esta parte imprescindible del doctorado es el crecimiento tanto científico como personal. El ver cómo hacen ciencia en otro país te abre la mente y te prepara para el futuro. Por otro lado conocer a tanta gente con culturas y educaciones diferentes supone un gran enriquecimiento personal. Fueron unos meses increíbles en los que estuve muy bien acompañada y en los que hice amistades muy especiales. Anja, siempre dispuesta a ayudar con cualquier tipo de problema fue mi guía en la cuidad y en el Helmholtz. Sandra y Val, con las que conocí todas las ciudades cercanas a Múnich viviendo momentos increíbles. Elena, mi compañera de confesiones y cervezas. Irem, la que más me ha enseñado y apoyado en todo momento. Kontxe, la persona más dulce que jamás conoceré en mi vida y el resto de compañeros, cada uno ha aportado su granito de arena en mi crecimiento personal: Fran, Ruth, Theresa, René, Guisseppe, Daniela y Tina. Gracias a mis amigas, las de toda la vida, mi grupo de “miss you” que me acompañan en cada paso de mi vida desde que tengo uso de razón. A mis “supernenas”, que aunque lejos, están siempre en mí día a día, y que me sacan sonrisas aunque haya tenido el peor de los días. A Lucía Montero, que con su fuerza y su cariño siempre es capaz de levantarme cuando me caigo y que siempre es un aliento de vida. Gracias por haberme incluido en “tu mundo” cuando más lo necesitaba y presentarme a todas tus personas increíbles que tan buenos momentos me han dado, gracias chicas. A mis “biogirls”, con las que entré en este mundo de la bioquímica y en especial a Bea Ranz, que siempre ha sido como una hermana para mí. Por último, gracias a toda mi familia. A mi tío Gustavo que siempre me anima a alcanzar mis metas, a mi tía Rous que siempre será mi hada madrina y a mi abuela Rosa, que siempre me recuerda que el pasado nunca vuelve y hay que vivir el presente. Gracias a mis abuelos, Paula y Luis, que estarían muy orgullosos de mí. A mi hermana, Coral, que es una inyección de amor puro. Pero sobretodo, gracias a mis padres, a los que va dedicada esta tesis. Gracias, por enseñarme todo lo que sé. Por brindarme una magnífica educación, de la que estaré eternamente agradecida. Por inculcarme los valores más importantes y por vivir siempre en un hogar de amor. Gracias, porque me habéis hecho la persona que soy. ix Some of the results exposed in this memory have been published in the following journals: • Delrio-Lorenzo, A., Rojo-Ruiz, J., Alonso, M. T., & García-Sancho, J. (2020). Sarcoplasmic reticulum Ca 2+ decreases with age and correlates with the decline in muscle function in Drosophila. Journal of cell science, 133(6), jcs240879. Communications have also been presented at the following conferences: • Delrio-Lorenzo, A., Rojo-Ruiz, J., Alonso, M. T., & García-Sancho, J. In vivo Ca 2+ imaging in Drosophila melanogaster reveals correlation between decreased sarcoplasmic reticulum content and sarcopenia in aging. The 8 th workshop of the European Calcium Society on Calcium Signaling in Aging and Neurodegenerative Diseases. Coimbra 2019. Oral communication. • Delrio-Lorenzo, A., Rojo-Ruiz, J., Alonso, M. T., & García-Sancho, J. GAP, flies, sarcoplasmic reticulum and aging. The 8 th workshop of the European Calcium Society on Calcium Signaling in Aging and Neurodegenerative Diseases. Coimbra 2019. Symposium. xvi Figure 43. The resting endoplasmic reticulum Ca 2+ concentration decreases with age in the wing sensory neurons.......................................... 122 Figure 44. Protein expression levels of SERCA, RyR and BiP in Drosophila skeletal muscle with age ................................................................ 124 Figure 45. Protein expression levels of SERCA, RyR and BiP in Drosophila brain with age ................................................................................. 126 Figure 46. Mechanism for the heating effect ................................................ 135 xvii List of tables Table 1. The voltage operated Ca 2+ channels. ................................................ 22 Table 2. Fly lines used in this thesis. .............................................................. 69 Table 3. Primary antibodies for Western blotting. ......................................... 76 Table 4. Secondary antibodies for Western blotting. .................................... 76 Table 5. Reagents and resources used in this thesis. ................................... 88 Table 6. Mean and maximal longevity values obtained from the survival curves shown in Fig. 26. .................................................................. 96 Table 7. Calibration of erGAP3 fluorescent signal into [Ca 2+ ] SR/ER in different cell types. ........................................................................................ 105 Abbreviations xxi Abbreviations [Ca 2+ ]: Ca 2+ concentration [Ca 2+ ] C : cytosolic Ca 2+ concentration [Ca 2+ ] ER : endoplasmic reticulum Ca 2+ concentration [Ca 2+ ] SR : sarcoplasmic reticulum Ca 2+ concentration A-IFM: asynchronous indirect flight muscle ATP: adenosine triphosphate BiP: immunoglobulin protein CaBP: Calcium-Binding-Protein CCh: carbachol CNS: central nervous system CEPIA: Calcium-measuring organelle-Entrapped Protein IndicAtors CR: calreticulin DHPR: dihydropyridine receptors DLM: dorsal longitudinal muscle DNA: deoxyribonucleic acid DR: dynamic range EGTA: Ethylene Glycol Tetraacetic Acid Elav: embryonic lethal abnormal visual system ER: endoplasmic reticulum Fig: Figure FKBP: FK506 binding proteins GAL4: galactose-induced gene 4 GAP: GFP-Aequorin-Protein GCaMP: (Green fluorescent protein-CalModulin Protein) GECI: genetically encoded Ca 2+ indicator GF: giant fiber GFP: Green Fluorescent Protein HRP: horseradish peroxidase IP 3 : Inositol 1,4,5-triphosphate IP 3 R: Inositol 1,4,5-triphosphate receptors Mhc: myosin heavy chain NCX: Na + /Ca 2+ exchanger xxii Abbreviations NCKX: Na + /Ca 2+ /K + exchanger NMJ: neuromuscular junctions PBS: phosphate-buffered saline PMCA: Plasma Membrane Ca 2+ -ATPase R: fluorescence ratio of two individual wavelengths R min: minimal fluorescence ratio R max: maximal fluorescence ratio R 0: basal fluorescence ratio ROI: region of interest ROS: Radical oxygen species RT: room temperature RyR: Ryanodine receptor SERCA: the sarco/endoplasmic reticulum Ca 2+ -ATPase SOCE: Store-Operated Ca 2+ Entry SR: sarcoplasmic reticulum SEM: standard error of the mean SDS: sodium dodecyl sulphate TBH: 2,5-di(tert-butyl)-1,4-benzohydroquinone TC: terminal cisternae UAS: upstream activating sequence VOCC: Voltage-Operated Calcium Channel WT: wild type Summary xxv Summary Aging still remains a mystery of biology and one of the most affected tissues in aging is skeletal muscle, whose loss of muscle mass and strength is called sarcopenia. Age-dependent sarcopenia is not restricted to mammals, as it affects other animal species including nematodes or flies. Cytosolic Ca 2+ ion is the intracellular second messenger that triggers muscle contraction. The sarcoplasmic reticulum is the store of Ca 2+ in the muscle cell, and it releases Ca 2+ to the cytosol when muscle contracts. Sarcopenia has been linked to the loss of Ca 2+ homeostasis that trigger muscle contraction, but mechanistic details remain unsolved. Here we explore the hypothesis that an alteration of the Ca 2+ content within the sarcoplasmic reticulum (SR) is at the origin of this loss of Ca 2+ homeostasis observed in sarcopenia. For investigating this hypothesis, we generated transgenic flies that express the ratiometric low affinity Ca 2+ indicator GAP3 targeted to the muscle sarcoplasmic reticulum (erGAP3), and we developed a new method to calibrate erGAP3 fluorescent signals into SR/ER Ca 2+ concentrations ([Ca 2+ ] SR/ER ). With these tools we measured resting [Ca 2+ ] SR in vivo along the fly life, and found a progressive decrease with aging that results in a tenfold reduction in the [Ca 2+ ] SR in the oldest flies. Then, to explore the molecular mechanisms involved in this decrease of [Ca 2+ ] SR we studied the expression levels of the main proteins involved in [Ca 2+ ] SR resting levels. In old muscle, we found a slight non-significant increase in the ryanodine receptors (RyR) and in the immunoglobulin protein (BiP) expression whereas the expression of the sarco/endoplasmic reticulum Ca 2+ - ATPase (SERCA) decreased by 35%. Moreover, the loss of function of the skeletal muscle was monitored by the well-characterized climbing assay, and found a strong correlation between the Ca 2+ content of the sarcoplasmic reticulum and fly climbing ability with aging. Furthermore, to assess whether the reduction of [Ca 2+ ] SR content in the aged flies also affected the [Ca 2+ ] C transients, we studied the cytosolic Ca 2+ dynamics during muscle contraction in transgenic flies expressing the cytosolic Ca 2+ sensor GCaMP in the muscle tissue. This experiments showed that old flies released less Ca 2+ to the cytosol in comparison to young flies and, thus, these results validated those obtained in the SR. In order to investigate whether the reduction of SR Ca 2+ content observed in muscle was a universal phenomenon of aging that occurred also in other tissues 1 Introduction 1. Physiological aging The inevitability of aging and death has preoccupied humanity for more than 5.000 years; however aging still remains a mystery of biology. Aging began to be viewed as capable of being extended or shaped in the 1980s, with the discovery of single gene mutations that extended lifespan in the nematode Caenorhabditis elegans. Nowadays, hundreds of mutant genes can increase longevity in model organisms, including nematodes, yeast (Saccharomyces cerevisiae), fruit flies (Drosophila melanogaster) and mice (Mus musculus). Most of these genes intervene in the evolutionarily conserved pathways that regulate growth, energy metabolism, nutrient sensing or reproduction. These recent discoveries in the biology of aging indicate that lifespan can be manipulated by genetic, nutritional or pharmacological intervention (Vijg and Campisi, 2008). Aging is defined as the gradual loss of function of tissues and organs that ultimately results in death. This progressive deterioration is associated with many human pathologies including cancer, diabetes, cardiovascular disorders or neurodegenerative diseases (Kennedy et al., 2014). Understanding these agerelated disorders is complicated but necessary to define the complexity of the aging phenotype, in contrast to the ease of defining single gene mutations. However, although some aging phenotypes are visible in laboratory models, they vary among mice strains or individuals and, more importantly some of them are not as obvious in invertebrates. This is the main reason why aging research should focus on searching for biomarkers to better define and characterize the degenerative processes underlying mortality (Vijg and Campisi, 2008). 1.1. The hallmarks of aging Aging is a multifactorial process involving the interaction of genetic and environmental factors. Experts in the field have attempted to identify and categorize the molecular hallmarks of aging in mammals (López-Otín et al., 2013). They propose nine aging hallmarks (Fig. 1) that occur in normal aging and its experimental aggravation or deterioration accelerate or ameliorate aging, 2 Introduction respectively. Moreover, they can be classified in three categories: primary, antagonistic and integrative. Primary hallmarks include: genomic instability that involves random deoxyribonucleic acid (DNA) damage and failure in the DNA repair mechanisms; telomere attrition or shortening; epigenetic alterations, such as changes in the DNA methylation patterns; and, impaired protein homeostasis (termed proteostasis); and they produce negative effects. Antagonistic hallmarks refer to the mechanisms that are initially beneficial but become deleterious at chronic high levels. This class includes mitochondrial dysfunction, which results in increased generation of radical oxygen species (ROS); cellular senescence, defined as a stable arrest from the cell cycle; and deregulated nutrient sensing. By interconnecting the two groups, a new category of hallmarks arises, the integrative that are responsible for the progressive loss of tissue function observed in aging. Among these, stem cell exhaustion as a decline in the regenerative property of tissues, and altered intercellular communication, including endocrine, neuronal or neuroendocrine. Figure 1. The Hallmarks of Aging. The diagram enumerates the proposed nine hallmarks of aging. In clockwise direction: genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, deregulated nutrient sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, and altered intercellular communication. Figure modified from (López-Otín et al., 2013). Deregulated nutrient sensing Mitochondrial dysfunction Loss of proteostasis Genomic instability Telomere attrition Epigenetic alterations Cellular senescence Stem cell exhaustion Altered intercellular communication 3 Introduction Defining the hallmarks of aging may contribute to build a start point for future research on the molecular mechanisms of getting older and to design interventions for the new concept of “healthy aging” that has been defined by the World Health Organization (WHO) as “the process of developing and maintaining the functional ability that enables wellbeing in older age”. 1.2. The history of aging theories Despite the efforts of a number of researchers and the huge advances in molecular biology and genetics, the mechanisms underlying aging are still unknown (Davidovic et al., 2010). In order to increase human lifespan substantially, we first need to understand the primary cause of aging, which leads to the key questions of why and how we age. There are many theories trying to explain why we age; however each of them is focused on one particular aspect of the aging process and, for this reason, each theory is incomplete (Gladyshev, 2016). Aging theories mainly fall into three categories: programmed theories, that imply that aging follows a biological timetable; evolutionary theories, that try to explain the course of aging as an evolutionary mechanism; and, damage or error theories, that emphasize that aging could be a result of environmental action, that causes cumulative cellular damage at various levels (Jin, 2010; Park and Yeo, 2013). The programmed and altruistic theory of aging considers that there is a genetic program that drives senescence and death, and this can be caused for two main altruistic reasons (Longo et al., 2005). Firstly, it benefits a small group of related organisms that acquire mutations to extend their lifespan; and, secondly, it must be necessary for the species as a group, as it could provide long-term benefits such as genetic diversity or acceleration of the pace of adaptation. In this context, the term “phenoptosis” (Skulachev, 1999) defines the death of an organism as a whole, alike to apoptosis (programmed cell death) which is a mechanism that has evolved to eliminate a portion of a multicellular organism. Similar phenomena are also already described at subcellular level, for example, the suicide of mitochondria is termed mitoptosis. Apoptosis is involved in many physiological processes such as cancer defence mechanism or immune response. 4 Introduction There are other sub-categories of programmed theories that view aging as a genetically regulated timetable. The endocrine theory postulates that the evolutionarily conserved insulin/IGF-1 signalling (IIS) pathway plays a key role in controlling the pace of aging (van Heemst, 2010). Also, the immunological theory states that the immune system is programmed to decline over time and that antibodies lose their effectiveness leading to an increased vulnerability to infectious disease and impaired immune function (Jin, 2010). Almost 50 years ago, the evolutionary theorists proposed several arguments against these programmed theories, and a new series of evolutionary theories aroused, trying to explain how the aging process is a result of the evolution, as explained by Darwin in “The origin of species”. The first argument postulated that the strength of natural selection declines with age, and, as a consequence, deleterious genetic mutations would accumulate and finally manifest at old age. This argument is known as the mutation accumulation theory and was proposed by Medawar in 1952 (Medawar, 1952). Few years later, in 1957, George Williams proposed another possibility: genetic trade-off alleles could cause harmful effects late in life but are selected because they are beneficial early in life. This proposal is known as the antagonistic pleiotropy theory. It suggests that aging evolves and does not act in any altruistic way for the species, but rather aging is adjusted by antagonistic pleiotropic genes. Another argument against programmed theories is introduced in the disposable soma theory, focused on metabolic trade-offs (Kirkwood, 1977). This theory stated that organisms have limited energy resources that have to be diverted between fertility and maintenance functions. The inability to allocate all resources makes repair functions less efficient over time and eventually it leads to damage accumulation. Among the sub-category of damage and error theories, one of the most accepted is the free radical theory developed by Harman (Harman, 1956). This theory suggests that the ROS generated in cell metabolism randomly damage cellular components such as lipids, proteins, sugars and nucleic acids. Although several researchers have tried to prove this theory, the results have been mostly negative (Bokov et al., 2004). Reduction of ROS or increase of antioxidants is effective against oxidative stress but fail to increase longevity (Pérez et al., 2009). Furthermore, ROS species are also signalling molecules that can control gene 5 Introduction expression and other physiological processes, so total inhibition is detrimental (Sohal and Orr, 2012). 2. The muscle tissue The muscle tissue is classified into three types according to its structure and function: skeletal, cardiac, and smooth muscle. Structurally, the skeletal muscle is attached to the bones and its function is to control voluntary movements. The cardiac muscle composes the contractile heart-wall and, thus, its function is to pump blood through the body. Finally, the smooth muscle recovers the internal organs and is responsible for their involuntary movement. The smooth muscle constitutes the contractile component of the digestive, urinary, and reproductive systems as well as the pulmonary airways and the arteries. 2.1. Skeletal muscle structure Skeletal muscles are composed of bundles, termed fasciculus, of muscle cells or myofibers 1 . The muscle cell is tubular, multinucleated and no longer capable of division. Myofibers contain numerous tubular myofibrils, composed of repeating units of sarcomeres (Fig. 2A). Sarcomeres are composed of two types of contractile proteins, myosin and actin, that form thick and thin filaments, respectively. These filaments slide past each other when a muscle contracts or relaxes and, thus, their interaction is carefully regulated by the regulatory proteins called tropomyosin and troponin. Tropomyosin is a long, double-stranded, helical protein that wraps up the actin and serves to block its active site. Troponin is a small, globular protein complex composed of three subunits: troponin-C (TN-C) binds Ca 2+ ions, troponin-T (TN-T) binds troponin to tropomyosin, and troponin-I (TN-I) inhibits the binding of actin to myosin in the resting state. The sarcomere is the basic contractile unit of skeletal muscles which is repeated between two Z-lines (Fig. 2B). The Z-line acts as an anchor of the actin filaments. Surrounding the Z1 Both terms, muscle cell and myofiber, will be used as synonyms along the document. 6 Introduction line, it is located the region of the I-band, that is composed only by actin filaments. Following the I-band, it exists the A-band which contains both thick and thin filaments. Within the A-band is the M-line (in the middle of the sarcomere) formed of cross-connecting elements of the cytoskeleton (Henderson et al., 2017). Figure 2. The skeletal muscle structure. A) Skeletal muscle tissue consists of bundles of myofibers, each one contains millions of myofibrils that are composed of repeating units of sarcomeres. The sarcomere is the minimum contractile unit, which consists of two types of contractile proteins, myosin and actin, that form thick and thin filaments, respectively. Figure modified from (Ojima, 2019). B) The picture represents the structure of the sarcomere. The sarcoplasmic reticulum is reproduced in yellow, while mitochondria (on both sides of the Z–lines) are illustrated in green. The white elongated structures display the plasma membrane T-tubules and the blue spots are the ryanodine receptors. Orange bubbles represent other sarcomere associated proteins. Figure modified from (Guarnieri et al., 2013). SARCOMERE I-band I-band A-band M-line Z-lineZ-line B A Sarcomere Myofibril Myofiber Skeletal muscle Thick filament Thin filament 7 Introduction The sarcoplasmic reticulum (SR) is a membrane-delimited intracellular organelle that spans the sarcomere and wraps up the contractile myofilaments in the skeletal muscle of almost all species and, thus, it is functionally involved in muscle contraction (Sorrentino, 2004). The SR is a specialized version of the endoplasmic reticulum (ER). The ER is a multifunctional organelle actively involved in Ca 2+ signalling and other functions such as protein synthesis and correct folding, lipid metabolism and drug detoxification (Prins and Michalak, 2009). Morphologically, it appears as a continuous network with three distinct parts specialized in different functions (Pozzan et al., 1994; Friedman and Voeltz, 2011): the rough ER, the smooth ER and the nuclear envelope. The rough ER appears as flattened sacs and contains ribosomes for protein synthesis. Surrounding the nucleus is an extension of the ER called the nuclear envelope, also dotted with ribosomes, that acts as a discrete Ca 2+ store in the perinuclear region, and supplies the nucleus with Ca 2+ for gene transcription (Reddish et al., 2017). The smooth ER is an elongated, cylindrical network with three key functions in Ca 2+ storage: Ca 2+ uptake, mediated by pumps and exchangers; storage, enhanced by luminal Ca 2+ binding proteins (CaBPs) that act as buffers; and, Ca 2+ mobilization mediated by specific ion channels (Lam and Galione, 2013). Morphologically, the SR consists on large tubes denominated terminal cisternae (TC) connected through elongated portions termed longitudinal SR. The Ryanodine receptor type 1 (RyR1) is located in TC whereas the sarco/endoplasmic reticulum Ca 2+ -ATPase (SERCA) pump is found in the longitudinal SR, suggesting that the TC domain is in charge of Ca 2+ release and the longitudinal SR of Ca 2+ uptake (Rossi et al., 2008). In muscle cells, the TC domain is in close contact with the plasma membrane or sarcolemma, particularly with the tubule invaginations called transversal tubules or T-tubules. The dihydropyridine receptors (DHPRs), which are L-type voltage-dependent Ca 2+ channels that activate upon muscle cell depolarization, are located in the T-tubules. In mammals DHPRs are physically in contact with the RyR type 1 of the TC to activate Ca 2+ release when membrane depolarization occurs (Sorrentino, 2011). Skeletal muscle cells contain ER and SR regions, and the borders among them are not as obvious; the ER-SR compartment can be visualized as a patchwork of specialized domains (Rossi et al., 2008). 8 Introduction 2.2. Skeletal muscle function The muscle cells are functionally organized as motor units that consists on a single nerve (motoneuron) and all the muscle cells it innervates. In mammals, each skeletal myofiber is innervated by a single motoneuron; although the same motoneuron may also innervate other myofibers and, thus, the force of contraction of muscles is regulated by how many motor units are activated. Muscle contraction is initiated by the transmission of an impulse (action potential) from the motoneuron to the neuromuscular junction (NMJ), which is a chemical synapse between the motoneuron (presynaptic cell) and the muscle cell (postsynaptic cell). The motoneuron releases the neurotransmitter acetylcholine (ACh) to the synaptic cleft, which is the space between the preand postsynaptic cells, where it binds to its receptor in the muscle cell; which, in turn, leads to an action potential in the muscle. Skeletal muscle contraction (Fig. 3) is then activated by depolarization of cell membrane or sarcolemma. This signal spreads through the T-tubule system and activates DHPRs. The depolarization signal produces a conformational change in DHPRs that activates RyR1 channels. The opening of the RyR1 leads to a massive Ca 2+ release from the SR towards the cytosol. Cytosolic Ca 2+ binds to troponin-C (TN-C) which induces a conformational change in the regulatory complex such that troponin-I (TN-I) exposes a site on the actin molecule that is able to bind to the myosin ATPase located on the myosin head. This binding results in the adenosine triphosphate (ATP) hydrolysis that produces a conformational change in the actinmyosin complex. As a result, the actin and myosin filaments slide past each other, thereby shortening the sarcomere length (Allen et al., 2008; Zalk et al., 2007; Efremov et al., 2015). Ca 2+ is sequestered to the SR by SERCA pump, thus, restoring the low-basal cytosolic Ca 2+ concentration ([Ca 2+ ] C ) and removing Ca 2+ from the TN-C. Unbinding of Ca 2+ from TN-C induces a conformational change in the troponin complex leading, once again, to TN-I inhibition of the actin binding site. At the end of the cycle, a new ATP binds to the myosin head, displacing the adenosine diphosphate (ADP), and the initial sarcomere length is restored. 9 Introduction Figure 3. Skeletal muscle contraction. 1) An action potential arrives at the neuromuscular junction, and triggers acetylcholine release (ACh) that diffuses across the synaptic cleft and binds to its receptors on the plasma membrane. 2) The action potential depolarizes the sarcolemma, and the signal spreads through the T-tubule system. 3) The depolarization signal produces a conformational change in the dihydropyridine receptors (DHPR) that activates the Ryanodine receptor type 1 (RyR1). 4) The opening of the RyR1 leads to an increase in cytosolic Ca 2+ concentration. 5) Cytosolic Ca 2+ binds to the troponin complex, resulting in a conformational change such that the actin molecule exposes the active binding site and, binds to the myosin head. This shortens the sarcomeres length. 6) The sarco/endoplasmic reticulum Ca 2+ -ATPase (SERCA) pumps Ca 2+ inside the sarcoplasmic reticulum (SR) and, therefore, refilling the SR store and restoring the lowbasal cytosolic Ca 2+ concentration. 7) Unbinding of Ca 2+ from troponin induces a conformational change in the troponin complex leading to inhibition of the actin binding site. The actin molecule is now unable to bind to the myosin head and, thus, the initial sarcomere length is restored. Modified from (Reece, 2011). 2.3. Skeletal muscle aging: sarcopenia One of the tissues that mostly suffer in aging is skeletal muscle. The importance of age-related changes in muscle function is increasing in the last few years and there are multiple reasons to understand their underlying mechanism and to design specific treatments. First, due to an increase in lifespan, together with a decrease in fertility, the human population is aged. The prediction for 2050 is that the Action potential (2) Motoneuron ACh (1) SR RyR1 (3) SERCA (6) Cytosol Ca 2+ (4) Sliding Myofilaments (5) DHPR T-Tubule Restore lenght (7) Actin Troponin complex Tropomyosin Sarcolemma Synaptic cleft Muscle cell Neuromuscular junction Myosin 16 Introduction unchanged in skeletal muscle from aged rats (Baehr et al., 2016) and in human aged muscle (Ogborn et al., 2014). e) Defects in the autophagy-lysosome system. Mammalian aged muscles show accumulation of lipofuscin or age pigment (Fig. 5), which is a recognized hallmark of aging (Hütter et al., 2007). Lipofuscin is a polymeric protein located in the lumen of the lysosome that has been demonstrated to accumulate under oxidative stress (Brunk and Terman, 2002). Lipofuscin accumulation may also interfere with the cell autophagocytotic capacity by acting as a sink for newly produced lysosomal enzymes and, therefore, result in improper turnover of cellular components and organelles. In Drosophila, defects in the autophagy-lysosome system are exemplified by an accumulation of poly-ubiquitinated protein aggregates during aging. This suggests that there is a progressive protein damage that, together with a decrease in the turnover of muscle proteins, may result in the age-related decline of muscle strength. FOXO (Forkhead box; class O) transcription factors are key players in muscle protein homeostasis, given their ability to activate multiple systems of protein disposal, among them the autophagylysosome proteolytic system. Overexpression of wild-type (WT) FOXO in muscles of adult flies is protective because it prevents the age-related decline in protein homeostasis and, therefore, it preserves muscle function. Conversely, during aging, FOXO null flies accumulate more protein aggregates in skeletal muscles than WT flies (Demontis and Perrimon, 2010). f) Alterations in the nuclear and plasma membrane integrity. Below the plasma membrane there is a protein network known as the dystrophin glycoprotein complex (DGC) that helps to maintain membrane integrity and protects muscle cells from the damage produced by mechanical muscle contraction (Doherty and McNally, 2003). Mutations in dystrophin gene cause Duchenne muscular dystrophy (DMD) in mammals and mutations in the DGC homologs in Drosophila cause mobility defects and shortening of lifespan (Shcherbata et al., 2007). Nuclear membranes also show age-related deterioration. Both mice and Drosophila display nuclei with aberrant shape (Fig. 5), condensed chromatin and spatial disorganization with age (Brandt et al., 2008; Cristea et al., 2010). The nuclear lamina, as the DGC complex, maintains nuclear shape and stability. It is composed 17 Introduction by fibrous proteins from the nuclear intermediate filaments family called lamins. In addition to their structural roles, they are implicated in basic nuclear functions such as chromatin organization, DNA replication, DNA transcription, DNA repair, and cell cycle progression. Mutations in lamins produce diseases named laminopathies, among them the Hutchinson–Gilford progeria syndrome (HGPS). The most frequent mutation in HGPS patients is a de novo mutation in the lamin A gene which produces a truncated protein with a 50 amino acid deletion termed progerin and its phenotype is characterized by accelerated aging and muscle degeneration (Prochniewicz et al., 2007). Flies carrying lamin mutations also show locomotor defects (Muñoz-Alarcón et al., 2007). Nuclei Lysosomes Mitochondria Sarcoplasmic reticulum Sarcomeres Figure 5. Intrinsic factors involved in sarcopenia. During aging, muscle cells progressively accumulate damaged proteins. Sarcomeres (red), exhibit disorganization and reduced length (myofibrils are shown in pink). Lysosomes (yellow) accumulate lipofuscin deposits and have decreased capability for degradation. Mitochondria (green) show functional and morphological abnormalities. Aggregation of membrane proteins of the sarcoplasmic reticulum (light blue) can result in tubular aggregates and improper Ca 2+ handling in old age. Nuclei (grey) present altered shape, spatial disorganization (black star) and loss of nuclear membrane integrity (shown as the discontinuous line). Taken from (Demontis et al., 2013). 18 Introduction 3. Ca 2+ signalling mechanisms Ca 2+ signalling relies on the capacity of the cells to maintain the gradient between intracellular and extracellular Ca 2+ concentrations (Clapham, 2007). Under resting conditions, [Ca 2+ ] C is around 100 nM whereas extracellular [Ca 2+ ] is 1-2 mM. The ER is the major Ca 2+ storage organelle of the cell and the [Ca 2+ ] ER is similar to the extracellular, between 0.2 and 1 mM (Alonso et al., 1998; de la Fuente et al., 2013). Other organelles such as the Golgi apparatus (GO), the lysosome, the endosome or the peroxisome can also serve as Ca 2+ reservoirs (Prins and Michalak, 2009; Lam and Galione, 2013). When cells are activated, Ca 2+ entry and/or Ca 2+ release from the stores produces a rise in the [Ca 2+ ] C that controls numerous cellular functions, such as cell proliferation, neurotransmitter release, muscular contraction or gene expression (Berridge et al., 2003; Berridge, 2016). Upon the arrival of a stimulus, Ca 2+ can enter from the extracellular medium through a variety of channels located in the plasma membrane and named according to their activation or deactivation properties. Ca 2+ can also be released from the intracellular stores through two receptor families, the Inositol 1,4,5triphosphate receptors (IP 3 Rs) and the Ryanodine receptors (RyRs), both located in the ER and the GO membranes. These channels are closed under resting conditions and they open upon stimulation, generating a rise in [Ca 2+ ] C , and, then they rapidly close after stimulation. The transient increase in [Ca 2+ ] C triggers a cellular response that can vary in a window time from milliseconds in muscular contraction, to days in cell proliferation. This Ca 2+ transient is shaped by calciumbinding proteins (CaBPs) and by intracellular organelles such as the mitochondria that can uptake Ca 2+ through the Mitochondrial Calcium Uniporter (MCU). Both of them act as Ca 2+ buffering systems. Given that a sustained increase in [Ca 2+ ] C is toxic for the cells, restoring of Ca 2+ to basal levels is necessary. The cytosolic Ca 2+ clearance mechanisms include the Ca 2+ -ATPases, which pump Ca 2+ against the concentration gradient outside the cells or inside the stores, and the exchangers that collaborate with the pumps to restore the basal Ca 2+ levels. This allows the cell to return to a resting state where it is able to respond to another stimuli (Berridge, 1997). 19 Introduction 3.1. The Ca 2+ signalling toolkit In order to generate Ca 2+ signals, cells contain a number of Ca 2+ channels in the plasma membrane and in the intracellular organelles. The channels, pumps, exchangers and the buffering systems that participate in Ca 2+ mobilization form the Ca 2+ signalling toolkit (Fig. 6). Figure 6. The Ca 2+ signalling toolkit. Increases in cytosolic Ca 2+ concentration trigger many physiological processes. They are due to Ca 2+ entry through the numerous Ca 2+ channels in the plasma membrane and to Ca 2+ release from the intracellular stores. The plasma membrane Ca 2+ channels are classified according to the stimulus that produce their opening in: Receptor-Operated Calcium Channels (ROCCs), Second-MessengerOperated Calcium Channels (SMOCCs), Voltage-Operated Calcium Channels (VOCCs), Transient Receptor Potential channels (TRPCs), and Store-Operated Calcium Channels (SOCCs). Ca 2+ release from the endoplasmic reticulum is mediated by two main families: the Inositol 1,4,5-triphosphate receptors (IP 3 Rs) and the Ryanodine receptors (RyRs); whereas the sarco/endoplasmic reticulum Ca 2+ -ATPase (SERCA) pump is in charge of refilling after stimulation. Moreover, upon cell stimulation, mitochondria can sequester Ca 2+ rapidly through the Mitochondrial Calcium Uniporter (MCU) and then release it slowly through the Na + /Ca 2+ exchanger (mNCX). Finally, the plasma membrane contains two main mechanisms that regulate Ca 2+ extrusion: pumps (Plasma Membrane Ca 2+ -ATPase; PMCA) and Ca 2+ exchangers (NCX). Modified from (Shigetomi et al., 2016). Extracellular space1-2 mM Ca 2+ ROC SMOC VOC TRP SOC NCX PMCA Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Ca 2+ Na + 2+ Ca 2+ Ca 2+ Gq PLC Na + ADP ATP Cytosol 100 nM Ca 2+ 200 nM 0.2 - 1 mM Ca 2+ ADP ATP IP 3 SERCA Mitochondria mNCX MCU IP3R RyR Ca Endoplasmic Reticulum Ca 2+ Ca 2+ Ca 2+ 20 Introduction 3.1.1. Ca 2+ entry from the extracellular medium Ca 2+ signalling capacity is driven by the large electrochemical gradient that exists across the plasma membrane. The membrane potential is -60 mV (negative inside) and [Ca 2+ ] is four orders of magnitude higher in the extracellular medium than in the cytosol. Cells are remarkably impermeable to Ca 2+ under resting conditions; however, when a signal arrives to the cell, a variety of Ca 2+ channels with different properties are activated, allowing Ca 2+ entry. Ca 2+ channels are classified according to their gating properties in: ReceptorOperated Calcium Channels (ROCCs), Second-Messenger-Operated Calcium Channels (SMOCCs), Voltage-Operated Calcium Channels (VOCCs), Transient Receptor Potential Channels (TRPCs), and Store-Operated Calcium Channels (SOCCs) (Fig. 6). The ROCCs are plasma membrane channels opened as a result of the binding of an agonist to its receptor, being the agonist a neurotransmitter, a hormone or even a nucleotide (Barritt, 1999). ROCCs are non-selective cation channels that are permeable to Ca 2+ , but can also admit Na + and K + . ROCCs can be classified in two main types, ionotropic or metabotropic. Ionotropic receptors, also called ligandgated ion channels, combine transmitter-binding and channel functions into a single molecular entity and, thus, the receptor is itself a channel. Examples of this type of channels include: the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor, widely distributed in the central nervous system (CNS); the glutamate N-methyl-D-aspartate (NMDA) found at the excitatory synapses; the gamma-aminobutyric acid (GABA) receptor, located in the limbic area of the brain that deals with memories and emotions; and the nicotinic acetylcholine (nACh) receptor, abundant in the neuromuscular junctions. The metabotropic receptors do not have ion channels as part of their structure; they are proteins physically separated from the channel and, so, the mechanism of channel opening implies one or more metabolic steps. Metabotropic receptors are either G protein-coupled receptors (GPCRs) that activate second messengers such as cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) indirectly gating the ion channel, or receptor tyrosine kinases that open the channels directly or indirectly by phosphorylation. Examples of metabotropic receptors include the 21 Introduction metabotropic glutamate receptors (mGluRs) that exist primarily in the CNS; or the muscarinic acetylcholine receptors (mAChRs), of which there are five subtypes (M 1 -M 5 ) that play important roles, such as the smooth muscle contraction that recovers the blood vessels and the pulmonary airways. The SMOCCs are activated by second messengers such as cyclic nucleotides (cAMP or cGMP), that activate cyclic nucleotide-gated (CNG) channels, whose function has been established in retinal photoreceptors and olfactory sensory neurons (Kaupp and Seifert, 2002); or arachidonic acid that induces activation of arachidonate-regulated Ca 2+ channels (ARCs), implicated in modulating the frequency of oscillatory Ca 2+ signals in various cell types (Shuttleworth, 2009). The VOCCs are activated by membrane depolarization and, thus, are key transducers of the electrical signal of the action potential into intracellular Ca 2+ transients that initiate many physiological events (Catterall, 2011). These channels close when the plasma membrane repolarizes and they can be activated during in vitro experiments with high extracellular K + concentration. There are ten members of the VOCC family in mammals, classified in three subfamilies (Ca V 1, Ca V 2 and Ca V 3) according to the different central pore forming α1 subunits that define the types of Ca 2+ currents. Ca 2+ currents are classified according to their activation properties in Low or High Voltage for Activation (LVA or HVA); and, according to their kinetics of inactivation in three subtypes: long-lasting (L-type), no long-lasting (no-L) and transient (T-type). The LVA currents are T-type, and the HVA are either L-type currents or no-L that include N, P/Q and R currents. L-type Ca 2+ currents are mediated by the Ca V 1 type of α1 subunit. Ca V 1 channels can trigger an array of cellular responses: excitation-contraction (EC) coupling in skeletal, cardiac, and smooth muscle; excitation-secretion coupling in endocrine cells and at ribbon synapses in specialized sensory cells; and excitation-transcription coupling or gene expression in nerve and muscle. Ca V 1 channels are sensitive to dihydropyridine (DHP) molecules, that act as antagonist blocking the channel and, thus, these channels are also called dihydropyridine receptors (DHPRs). The Ca V 2 subfamily conducts N, P/Q and R currents and is primarily responsible for the initiation of synaptic transmission at fast synapses. These channels are blocked by spider, snake or scorpion toxins. The Ca V 3 subfamily is important for repetitive firing of action potentials (T-type currents) in rhythmically firing cells such as 22 Introduction cardiac myocytes and thalamic neurons. They are insensible to DHP or the toxins that block the Ca v 2 channels. The classification of the VOCC family is summarized in Table 1. α1 subunit Ca 2+ Current Specific Blocker Physiological function Ca v 1 HVA. L-type DHP Excitation-contraction, Excitation-secretion, and Excitation-transcription coupling Ca v 2 HVA. N, P/Q and R type Toxins Synaptic transmission at fast synapses Ca v 3 LVA. T-type Insensible Rhythmically firing cells Table 1. The voltage operated Ca 2+ channels. The TRPCs constitute a large superfamily of cation channels forming proteins which exhibit a large variety of functional properties and play diverse cellular and physiological roles (Gees et al., 2012). In mammals, 28 different genes have been identified, subdivided into multiple subfamilies; and, all related to the gene product of the trp locus, originally discovered in Drosophila (Montell and Rubin, 1989). In terms of gating mechanisms, there is a huge variation within the TRP superfamily, being many of them polimodal and some of them constitutively active channels. They can be activated by a variety of factors including depolarization, temperature, ligands, osmolarity or mechanical stimuli. Originally these channels were found in the plasma membrane, although an increasing number of them are being described in intracellular organelles, such as the ER, the Golgi apparatus or the endo-lysosome (Zhang et al., 2018). The SOCCs are plasma membrane channels that mediate Ca 2+ influx from the extracellular medium upon emptying of the intracellular stores (Prakriya and Lewis, 2015). Before the genes codifying for these channels were identified, they were referred to, in general, as Calcium Release-Activated Channels (CRACs) but nowadays the channels are known as Orai1-3 proteins (Feske et al., 2006). Ca 2+ release from the ER lowers [Ca 2+ ] ER , this reduces the Ca 2+ affinity from the EF- 23 Introduction hands (Ca 2+ -binding domain) of STIM (stromal interaction molecule) proteins that act as ER Ca 2+ sensors (Liou et al., 2005). In turn, activated STIM monomers oligomerize and translocate to the plasma membrane to form punctate structures at the existent ER/PM junctions, where the endoplasmic reticulum (ER) and the plasma membrane (PM) are kept in close appositions. Orai channels are also redistributed from diffusely distributed clusters along the plasma membrane into the punctate structures forming complexes with STIM. Oligomerized STIM directly activates Orai channels, which, in turn, lead to selective Ca 2+ influx and refilling of the ER-stores through the sarco/endoplasmic reticulum Ca 2+ -ATPase (SERCA) pump. This phenomenon has been termed as capacitive entry or Store-Operated Ca 2+ Entry (SOCE) (Putney, 1986; Ambudkar et al., 2017). Regardless of its nature and gating properties, the opening of the plasma membrane channels generate a high [Ca 2+ ] at its mouth, on the inner side of the plasma membrane, termed subplasma membrane Ca 2+ microdomains (Rizzuto and Pozzan, 2006). These Ca 2+ microdomains control a number of cellular functions, such as secretion of neurotransmitters at the presynaptic membrane or modulation of mitochondria Ca 2+ uptake capacity. Mitochondria have been shown to sense Ca 2+ gradients near Orai channels and, therefore, modulate their Ca 2+ - dependent inactivation (Gilabert and Parekh, 2000). 3.1.2. Cytosolic Ca 2+ exit to the extracellular medium The plasma membrane contains two main mechanisms that regulate Ca 2+ extrusion, the Ca 2+ -pumps or Ca 2+ -ATPases and the ionic exchangers (Fig. 6), which operate to pump Ca 2+ out of the cytoplasm against the electrochemical gradient (Brini and Carafoli, 2011). Ca 2+ -ATPases are primary active transport pumps that need to be phosphorylated to transport Ca 2+ , whereas the exchangers accomplish Ca 2+ extrusion by using the electrochemical gradient of Na + , so there is no direct coupling of ATP. This type of transport is known as secondary active transport or co-transport. The Plasma Membrane Ca 2+ -ATPase (PMCA) has a low transport capacity (turnover frequency of 150 s -1 ) and a high Ca 2+ affinity (K D of 10-30 µM in resting 24 Introduction state), so it can operate at low [Ca 2+ ] C and contribute to maintain the resting [Ca 2+ ] C . The PMCA was originally discovered in erythrocytes (Schatzmann, 1966) and, then, it was described and characterized in other numerous cell types. There are four basic PMCA isoforms (the PMCA is the product of four separate genes) that differ in tissue distribution. PMCA1 and 4 are ubiquitous whereas PMCA2 and PMCA3 expression is restricted to some tissues: PMCA2 is expressed in the nervous system and in the mammary gland and PMCA3 in the nervous system (Brini and Carafoli, 2009). The PMCA pump belongs to the family of P-type ATPases, it operates transporting Ca 2+ upon ATP hydrolysis with a 1:1 stoichiometry (Hao et al., 1994) and, hence, it is electrogenic (1 Ca 2+ in exchange for 1H + ). The activity of the pump is highly regulated: it is activated by acidic phospholipids and calmodulin (Carafoli et al., 1996); and also inhibited by many agents such as vanadate, lanthanum or caloxin. Structurally, it consists of ten transmembrane domains, two large intracellular loops, and the aminoand the carboxy-terminal cytoplasmic tails. In contrast to Ca 2+ -ATPases, Ca 2+ exchangers have a low Ca 2+ affinity (K D of 0.10.3 µM) and a high transport capacity (turnover frequency up to 5000 s -1 ) and, thus, are able to remove high amounts of Ca 2+ from the cytosol in a short time. The Na + /Ca 2+ exchanger (NCX) and the Na + /Ca 2+ /K + exchanger (NCKX) are the two main classes of Na + /Ca 2+ antiporters found in animals that play a critical role in Ca 2+ exit from the cytosol (Sharma and O’Halloran, 2014). The NCX includes three proteins (NCX1, 2 and 3) encoded by three separate genes (DiPolo and Beaugé, 2006): NCX1 is practically ubiquitous although it is highly expressed in the heart, in the brain and in the kidney; NCX2 is selectively expressed in the brain; and, NCX3 expression is restricted to the brain and the skeletal muscle. NCX accomplishes the exit of 1Ca 2+ ion in exchange for 3Na + ions that enter the cell. The NCKX comprises five members: NCKX1 is expressed in retinal rod photoreceptors; NCKX2 is expressed in cone photoreceptors; NCKX3 and NCKX4 are abundant in the brain, but also in other tissues; and, NCKX5 is expressed in skin, retinal epithelium and brain (Schnetkamp, 2004; Lytton, 2007). NCKX transports 1K + together with 1Ca 2+ in exchange for 4Na + ions. The operation of the exchangers is fully reversible, and the direction of the movement of the transported ions depends on the electrochemical gradients of Na + and Ca 2+ . 25 Introduction 3.1.3. Ca 2+ buffering systems The level of intracellular Ca 2+ is determined by a balance between the ‘on’ reactions that introduce Ca 2+ into the cytoplasm and the ‘off’ reactions through which Ca 2+ is removed by the combined action of pumps, exchangers and buffers. During the ‘on’ reactions, most of the Ca 2+ gets bound to the buffers that shape the amplitude, the duration, and the extent of the cytosolic Ca 2+ transient, whereas a small proportion of the Ca 2+ binds to the sensors or effectors that are responsible for stimulating numerous Ca 2+ -dependent processes. Both of them, buffers and sensors are CaBPs. During the ‘off’ reactions, Ca 2+ unbinds the sensors and buffers (Berridge et al., 2003). There are many protein families among the CaBPs, some of them are extracellular, like the Ca 2+ binding epidermal growth factor like proteins (cbEGF), but most of them are intracellular. Within the intracellular class, all of them contain negatively charged groups arranged for chemical Ca 2+ coordination, although they differ in the evolutionary conserved motifs or domains capable of binding Ca 2+ ions. Among them, the annexins; the C2-domain proteins; or the EF-hand proteins. The majority of the EF-hand protein family belongs to the group of Ca 2+ sensors; these proteins, upon binding of Ca 2+ ions, undergo a conformational change which allows them to act as regulators of specific targets (Schwaller, 2010). The prototypical example of Ca 2+ sensor is the calmodulin protein (Chin and Means, 2000). However, some of these EF-hand proteins act as Ca 2+ buffers, including parvalbumin, calbindin or calretinin. The “EF-hand” domain (Fig. 7) was first described by Kretsinger in 1973 as a Ca 2+ -binding variant of a helix-loop-helix motif in the structure of parvalbumin (Kretsinger and Nockolds, 1973). The name EF-hand was termed given the resemblance of the spatial distribution of the α-helixes E and F of the parvalbumin, to the letter “L” that forms the spread thumb and forefinger of a human hand. The loop of the EF-hands is composed by 12 amino acid residues. Amino acids in positions 1, 3, 5, 7, 9 and 12 (denoted by X, Y, Z, -Y, -X and –Z) are involved in Ca 2+ coordination and, thus, are acidic and highly conserved. The glutamate residue in the last position (12) of the loop serves as a bidentate ligand and, thus, contributes two oxygen atoms of its γ-carboxyl group, and the central residue (7) binds Ca 2+ with the main-chain carbonyl oxygen atom. Next to this central residue, 32 Introduction intracellular stores through Ca 2+ release channels as well as predicted countermovement of ions across the SR membrane to balance the transient negative potential generated by Ca 2+ release. The transmembrane protein named Trimeric Intracellular Cation (TRIC) channels (TRICA and B), formerly known as mitsugumin which is the Japanese for junction, is A SR-K + channel thought to provide counterion currents that facilitate the active release of Ca 2+ from intracellular stores (Yazawa et al., 2007). However, it has been shown that RyR and likely IP 3 R channels might mediate its own countercurrent given its ionic non-selectivity and high permeability to monovalent ions (Gillespie and Fill, 2008). Figure 9. Structure of the ryanodine receptor type 1. Schematic of the Ryanodine receptor type 1 (RyR1) and its modulators. RyR1 (in green) is localized in the tubular cisternae region of the sarcoplasmic reticulum where it physically interacts with the dihydropyridine receptors (DHPR; in orange) to activate Ca 2+ release when membrane is depolarized. RyRs are activated by cytosolic Ca 2+ or ATP; and inhibited by Mg 2+ . A number of binding partners that regulate the channel has been described: e.g. calmodulin (CaM; purple), S100A proteins (yellow), or FK506 binding proteins (FKBP; blue), from the cytosolic side; and junctin (pink), triadin (yellow) or calsequestrin (CSQ; blue), from the luminal side. Modified from (Lanner et al., 2010). Skeletal muscle SR T-tubule RyR1 DHPR Ca 2+ FKBP CaM Ca 2+ Ca 2+ Mg 2+ CSQ CSQ Junctin ATP Cell membrane S100A Triadin 33 Introduction 3.1.4.2. The sarco/endoplasmic reticulum Ca 2+ -ATPase The Ca 2+ -ATPase found on the membrane of the SR/ER is the SERCA pump that contributes, along with PMCA and the Ca 2+ exchangers in the plasma membrane, to restore [Ca 2+ ] C to 100 nM after cell stimulation by actively pumping the Ca 2+ ion into the SR/ER lumen and, thus, refilling the intracellular stores (Brini and Carafoli, 2009; Berridge, 2012). The Golgi apparatus also contributes in this task with two pumps: the SERCA located in the cis-Golgi region (cGO) and the Secretory Pathway Ca 2+ -ATPase (SPCA) in the trans-Golgi apparatus (tGO) (Vandecaetsbeek et al., 2011; Aulestia et al., 2015). Like PMCA, SERCA pump is a P-type ATPase pump, powered by ATP hydrolysis to move Ca 2+ ions against the concentration gradient. The SERCA pump, as PMCA, has low transport capacity (turnover frequency of 200 s -1 ) and high Ca 2+ affinity, with a K D for Ca 2+ of 0.2 μM under resting conditions and 1 μM under conditions of maximal activation (Strehler and Treiman, 2004). However, SERCA is more efficient than PMCA, as it transports two Ca 2+ ions into the ER lumen for each hydrolyzed ATP molecule instead of one (Toyoshima, 2009), and it is electrogenic (it exchanges 2Ca 2+ for 2H + ). There are three genes, each of them encode one protein isoform (SERCA13) that display tissue-specific expression: SERCA1 is primarily distributed in fasttwitch (type II) skeletal muscle; SERCA2 is found in fast and slow-twitch (type I) muscle, in cardiac and smooth muscle, and also in non-muscle cells; and, SERCA3 is expressed in non-muscle cells (Periasamy and Kalyanasundaram, 2007). Structurally (Fig. 10), SERCA is a monomer of 110 kDa molecular weight, consisting on a large cytosolic domain that includes the phosphorylation site (domain P), the nucleotide-binding site (domain N), and the anchor (domain A); and, a transmembrane domain of 10 α-helixes that contain two Ca 2+ binding sites (Toyoshima et al., 2003). SERCA activity is highly regulated by phosphorylation and protein interactions. For example, increases in [Ca 2+ ] C activate calmodulin that, in turn, activates Ca 2+ /calmodulin kinase II (CaMKII), which phosphorylates and activates SERCA pump (Vangheluwe et al., 2005). Furthermore, it is well established that the transmembrane proteins phospholamban and sarcolipin regulate the activity of the pump (MacLennan et al., 2003). Moreover, SERCA is inhibited by three specific inhibitors: the irreversible inhibitor thapsigargin, isolated from Thapsia garganica (Thastrup et al., 1989); and, the reversible inhibitors 34 Introduction cyclopiazonic acid (CPA) isolated from fungi such as Aspergillus and Penicillum (Demaurex et al., 1992), and 2,5-di(tert-butyl)-1,4-benzohydroquinone (TBH) (Moore et al., 1987). Figure 10. Structure of the sarco/endoplasmic reticulum Ca 2+ -ATPase. The Sarco/Endoplasmic Reticulum Ca 2+ -ATPase (SERCA) crystal structure. SERCA protein contains a large cytosolic domain and a transmembrane domain. The cytosolic part of the protein consists on the phosphorylation site (domain P, shown in dark blue), the nucleotide-binding site (domain N, shown in green), and the anchor (domain A, in red). The transmembrane domain (in cyan) is composed of 10 α-helices that contain two Ca 2+ binding sites (Ca 2+ ion is shown in yellow). Taken from (Sun et al., 2019). 3.1.4.3. Ca 2+ leak channels In addition to regulated Ca 2+ release pathways, there are also leak pathways that can be revealed by Ca 2+ loss from ER Ca 2+ stores when SERCA pump is inhibited (Camello et al., 2002). Therefore, the resting free Ca 2+ concentration maintained inside the ER is the result of a pump/leak steady state between the SERCA pump which mediates Ca 2+ entry and the ER Ca 2+ leak, whose molecular nature is still 35 Introduction controversial. Two types of Ca 2+ release pathways have been proposed: novel, still non-identified proteins; or known Ca 2+ release channels (such as RyRs and IP 3 Rs) that are deregulated. A number of channel proteins have been implicated in basal Ca 2+ leak pathways and, therefore, proposed to be the leak channel. These include: the translocon or Sec61 complex, the major entry site of newly synthesized proteins in the ER (Lang et al., 2011); the presenilins, whose mutations are linked to Alzheimer’s disease, have been associated with decreased Ca 2+ leak from the ER (Tu et al., 2006); the pannexins or protein pores, associated with ribosomes in rough ER (Abeele et al., 2006); the anti-apoptotic protein B-cell lymphoma-extralarge (Bcl-x L ), which interacts with IP 3 Rs to enhance the opening of the channel (Li et al., 2007); Bax inhibitor-1 (BI-1), that protects against apoptosis and ER stress (Bultynck et al., 2012); FKBP proteins, whose dissociation from RyR destabilizes the channel and render it leaky (Brillantes et al., 1994); or proteins involved in RyR channel complex remodelling, including PKA phosphorylation and S-nitrosylation (Bellinger et al., 2009, 2008). FKBP12 and FKBP12.6 also known as calstabin1-2 (calcium channel stabilizing binding protein) are peptidyl-propyl-cistransisomerases that associate with RyR1-2, respectively, via amphiphilic β-sheet structures. In order to regulate the channel gating through protein-protein interactions (Brillantes et al., 1994) and prevent pathological intracellular Ca 2+ leak (Santulli and Marks, 2015), one calstabin molecule binds to each RyR monomer (Timerman et al., 1993; Xin et al., 1995). This association helps to stabilize the closed state of the channel (Efremov et al., 2015); and, is crucial for EC coupling in both skeletal and cardiac muscles (Lanner et al., 2010); as well as in the brain where the RyR-mediated leakiness hypothesis was also recently proposed as an underlying mechanism in Alzheimer’s disease (Lacampagne et al., 2017). FKBPs are a highly conserved protein family, existing eight FKBP proteins in Drosophila which share its homology with the human FKBP12 protein (Ghartey-Kwansah et al., 2018). Finally, a new ER transmembrane membrane protein named Transmembrane and coiled-coil domains 1 (TMCO1) was recently discovered (Wang et al., 2016). TMCO1 has been proposed to have a role in preventing Ca 2+ stores from overfilling; in such a way that, upon Ca 2+ overloading, the protein forms homotetramers that would open and, thus, trigger Ca 2+ release from the ER. Then, the complex undergoes disassembly upon Ca 2+ depletion. 36 Introduction 3.1.4.4. Endoplasmic reticulum junctions with other organelles In the last decades, the contact sites between the ER membrane and other organelles have gained increased attention (Phillips and Voeltz, 2016). Plasma membrane-ER contact sites, such as the punctate structures involved in SOCE (Smyth et al., 2010), or ER-lysosome contact sites (Venkatachalam et al., 2015), have been linked to the regulation of Ca 2+ homeostasis. Besides their role in Ca 2+ homeostasis, MAMs are also involved in mitochondrial functions such as the transport of phospholipids or direct Ca 2+ transmission to mitochondria that activates the enzymes implicated in the tricarboxylic acid cycle; and, in other physiological functions such as: inflammation, autophagy, and apoptosis. Moreover, the ER and the mitochondria are tightly associated with very dynamic structures termed mitochondria-associated membranes (MAMs) that are approximately 10 nm distance at the smooth ER and 25 nm at the rough ER (Csordás et al., 2006). MAMs provide a platform that allow rapid exchange of biological molecules and, thus, is fundamental for several cellular functions, such as: Ca 2+ homeostasis, autophagy, lipid metabolism, and apoptosis. To the date, numerous MAM-specific proteins have been identified (Hayashi et al., 2009). Most of these proteins are ER proteins (ion channels, metabolic enzymes or molecular chaperones), with only a few belonging to mitochondria (electron transport chain proteins or mitochondrial fusion proteins). In fact, alterations in the composition of MAMs or in the crosstalk between the two organelles lead to different pathological conditions such as cancer or neurodegenerative diseases (Bravo et al., 2011). 3.2. The “Ca 2+ hypothesis” of aging The “Ca 2+ hypothesis” of aging was proposed by Khachaturian in 1994 (Khachaturian, 1994) to explain the neurophysiological mechanisms involving Ca 2+ signalling related to aging and neurodegeneration. Several changes in the components of the Ca 2+ signalling (channels, pumps, exchangers or mitochondria) have been described to be altered in aging (Toescu and Vreugdenhil, 2010). For instance it has been reported: an increased in Ca 2+ influx through VOCCs 37 Introduction (Campbell et al., 1996); an increased ER Ca 2+ release mediated mostly by RyR, proposing this channel as a biomarker of aging in neurons (Gant et al., 2006; Thibault et al., 2007); an altered mitochondrial buffering capacity (Xiong et al., 2002); and more recently, a decrease in SOCE (Calvo-Rodríguez et al., 2016). However, different models trying to verify the “Ca 2+ hypothesis” of neuronal aging yielded a set of rather inconsistent results. Indeed, the increased plasma membrane Ca 2+ entry reported in hippocampal neurons of brain slices (Thibault and Landfield, 1996) was not observed in other neuronal types like dorsal-root ganglia neurons (Kostyuk et al., 1993) or cerebellar neurons (Kirischuk and Verkhratsky, 1996), indicating that aging might have affect differentially the various types of neurons in the brain or even tissues in the body. At present, the reason for these differential effects in each neuronal type is not known (Toescu and Verkhratsky, 2003). Considering the association between Ca 2+ dyshomeostasis and brain aging or neurodegenerative diseases, the development of novel drugs against target molecules of the Ca 2+ signalling toolkit, such as ER leak channels or the Mitochondrial Calcium Uniporter (MCU), could be promising (Chandran et al., 2019). 3.3. The Ca 2+ sensors Ca 2+ ion is the most ubiquitous intracellular second messenger, involved in almost every physiological process, from gene expression to muscle contraction or neurotransmitter release (Berridge et al., 2003). Unlike other molecules that serve as transduction elements, Ca 2+ is neither synthesized nor metabolized. Instead, the ion storage and circulation is controlled by the Ca 2+ signalling toolkit. Under normal conditions, only transient increases of [Ca 2+ ] C occur, and Ca 2+ overload compromises cell viability and is a signal for apoptosis. Alterations in Ca 2+ homeostasis have been proposed as possible sources of many pathological conditions such as cancer or neurodegenerative diseases (Berridge, 2016). Therefore, it is important to accurately measure the [Ca 2+ ] to fully understand physiological and pathological processes. For this reason, a number of Ca 2+ 38 Introduction sensors or indicators 2 with different nature and properties have been developed that can be divided in two main classes: synthetic Ca 2+ indicators (also known as chemical) and genetically encoded Ca 2+ indicators (GECI). 3.3.1. Synthetic Ca 2+ indicators Synthetic Ca 2+ indicators are small molecules that can chelate Ca 2+ ions due to the presence of carboxylic acid groups in their structure. These molecules are based on the BAPTA (1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid) molecule. Synthetic Ca 2+ indicators are unable to cross lipid membranes due to their hydrophilic nature, making necessary the use of physical or chemical methods to load them inside the cell. The protection of the carboxylic groups as acetoxymethyl (AM) esters makes the dye 3 hydrophobic, so it can cross the cell membrane by simple diffusion (Tsien, 1981). Once inside the cell, the esterases cleave the AM groups, and the hydrophobic indicator becomes hydrophilic and, thus, trapped inside the cell. Indicators can be classified into either ratiometric or single wavelength dyes. Ratiometric indicators shift the peak wavelength of either their excitation or emission spectra upon binding of Ca 2+ ions. Based on this fact, this group can be subdivided in dual excitation or dual emission dyes. Fura-2 is a dual excitation dye, widely considered the standard for quantitative intracellular Ca 2+ measurements (Grynkiewicz et al., 1985). The excitation peak shifts from 340 nm in the Ca 2+ - bound state to 380 nm in the Ca 2+ -free state; with fluorescence emission at 500 nm for either excitation wavelength (Fig. 11A). Indo-1 is a dual emission dye, with a single excitation peak at 350 nm and a dual peak emission that occurs at 405 nm and 485 nm in the Ca 2+ -bound and Ca 2+ -free states, respectively (Fig. 11B). This enables ratiometric measurements that provide a more accurate quantification of the [Ca 2+ ] and are less prone to artefacts such as uneven dye loading, dye leakage, photo bleaching or changes in cell volume. On the contrary, single wavelength 2 Both terms, Ca 2+ sensor and Ca 2+ indicator, will be used along the document as synonyms. 3 A dye is a coloured substance that chemically binds to the substrate to which it is being applied and, thus, changes its colour. The term will be used to refer to different types of dyes: synthetic Ca 2+ indicators (mainly Fluo-4); Hoescht that stains nuclei in blue; and, Brilliant Blue G dye used to determine protein concentration in the Bradford assay. 39 Introduction indicators exhibit significant Ca 2+ dependent changes in fluorescence intensity without shifting their excitation or emission wavelengths and, thus, they are also called intensiometric. This sensors normally yield higher dynamic ranges, which is the ratio of maximum and minimum fluorescence intensity (F max /F min ). Within this group, the indicators based on the rhodamine (e.g. Rhod-2) or the fluorescein chromophores (e.g. Fluo-3; Fig. 11C and D) (Minta et al., 1989). Figure 11. Fluorescence spectra and structure of the main synthetic Ca 2+ indicators. The figures represent fluorescent spectra with Ca 2+ concentration ranging from 0 to 39.8 µM Ca 2+ . A) Excitation spectra of Fura-2. B) Emission spectra of Indo-1. C) Emission spectra of Fluo-3. D) Chemical structure of Fluo-3. (Modified from “The Molecular Probes Handbook, a guide to fluorescent probes and labeling technologies”; chapter 19, “Indicators for Ca 2+ , Mg 2+ , Zn 2+ and other metal ions”). BA C D Fura-2 Indo-1 Fluo-3 40 Introduction In this thesis we used a fluorescent Ca 2+ -dye based on the fluorescein chromophore, Fluo-4 AM, to measure cytosolic Ca 2+ in vitro. Fluo-4 is a single wavelength Ca 2+ indicator with an excitation wavelength of 494 nm and emission wavelength of 516 nm. Ca 2+ binding to the dye produces increases of 100 times in fluorescence. The high affinity of the synthetic dye to Ca 2+ (K D = 335 nM) makes it appropriate for cytosolic Ca 2+ measurements. A major advantage of synthetic indicators is the broad range of Ca 2+ affinities that are commercially available for the user (from <50 nM to >50 μM). Higher affinity indicators can be used to quantify Ca 2+ levels in the cytosol while lower affinity indicators can be optimized for measuring Ca 2+ in subcellular compartments with higher Ca 2+ concentrations. Also, they do not have to be transfected or expressed in cells, as cell loading protocols have been very well established (Takahashi et al., 1999). A major disadvantage is that the cellular localization of Ca 2+ indicators cannot be specifically targeted to a particular organelle. In addition, synthetic indicators tend to compartmentalize and are eventually extruded from the cell during long recording experiments. 3.3.2. Genetically encoded Ca 2+ indicators (GECIs) Genetically encoded Ca 2+ indicators (GECIs) are proteins composed by at least one light emitting protein and one Ca 2+ -responsive element, so that Ca 2+ binding changes its optical properties. The main advantages of GECIs over synthetic Ca 2+ indicators is their precise targeting to organelles by the fusion with signal peptide sequences; and, the generation of transgenic animals, which makes them inheritable. GECIs can be classified in three main groups: bioluminescent, based on the aequorin photoprotein; fluorescent, composed by a single fluorescent protein and a Ca 2+ -responsive element, e.g. camgaroos, pericams, GCaMPs and GECOs; and fluorescence resonance energy transfer (FRET) based, composed by two fluorescent proteins between a Ca 2+ -responsive element, e.g. camaleons and troponin-based Ca 2+ sensors. However, in our laboratory, we have generated the GAP (GFP-Aequorin Protein) family of GECIs (Rodriguez-Garcia et al., 2014). GAP is a unique dual-mode Ca 2+ indicator, able to function either as a fluorescent 41 Introduction or a luminescent sensor, depending on whether the photoprotein aequorin is in its apo-state or reconstituted with its cofactor coelenterazine (Rodríguez-Prados et al., 2015). In this thesis we used GAP in the fluorescent mode and other fluorescent (non-FRET based) GECIs to measure [Ca 2+ ]. 3.3.2.1. GAP (GFP-Aequorin Protein) To measure endoplasmic reticulum Ca 2+ we used the GAP indicator, composed by the fusion of the two proteins from the jellyfish Aequorea victoria: Green Fluorescent Protein (GFP) and aequorin. Figure 12. Bioluminescence reaction catalysed by aequorin. Apo-aequorin is reconstituted with the cofactor coelenterazine, and in the presence of O 2 (represented in green) it binds three Ca 2+ ions (blue). Aequorin undergoes a conformational change, oxides coelenterazine into coelenteramide and emits blue light (470 nm), being also the product of this reaction CO 2 (yellow). Aequorin is a Ca 2+ sensitive photoprotein that emits blue light when binding to Ca 2+ ions and in the presence of O 2 and its prosthetic group, the cofactor coelenterazine (Shimomura et al., 1962; Shimomura and Johnson, 1972). In the bioluminescence reaction (Fig. 12), aequorin undergoes a conformational change and oxides coelenterazine into coelenteramide, being also the products of this reaction CO 2 and blue light (470 nm). At a structural level, aequorin is a globular protein containing a hydrophobic core cavity that accommodates the ligand coelenterazine. With a molecular weight of 21.5 kDa, aequorin contains four helixloop-helix 'EF-hand' domains (I, II, III and IV) or Ca 2+ binding sites arranged in pairs, of which the second domain is not functional. Mutations in these 'EF-hand' domains can reduce the aequorin affinity for Ca 2+ . For example, the mutation D119A, located Apo - aequorin Coelenterazine Ca 2+ O 2 C O 2 h Ʋ 470 nm Coelenteramide Coelenterazine O 2 48 Introduction Figure 16. Drosophila life cycle. The whole life cycle of Drosophila takes around 10–12 days at 25 °C. The Drosophila development is divided into various stages: embryo (1 day), larva (first, second and third instars; 5 days), pupa (3-4 days) and adult. Taken from (Ong et al., 2015). 4.2. Genetics Another advantage of the fly as an animal model is that they have a compact genome, mostly because intergenic and intronic sequences are reduced in size compared to other vertebrate genomes, which makes it easier to manipulate genetically. Also, they have many orthologous genes associated with human diseases. The genome of Drosophila contains four pairs of chromosomes, an X/Y pair and three autosomes, labelled 2, 3 and 4. The first pair or the sex chromosomes are composed by an acrocentric X chromosome and a submetacentric Y chromosome, that is almost entirely composed of heterochromatin (highly compacted, transcriptionally silent DNA and dense in Adult Embryo Pupa First instar larva Second instar larva Third instar larva FemaleMale 1 day 1 day 1 day 3 days 3-4 days 49 Introduction repeats). The remaining three pairs are autosomes. Chromosomes 2 and 3 are large metacentric autosomes, and chromosome 4 (the dot chromosome) is very small and often ignored, apart from the genetic markers yellow and white, affecting body and eye colour, respectively. Other classic examples of these genetic markers are: Curly (Cy), on the second chromosome (affecting wing shape); and Stubble (Sb), on the third chromosome (affecting bristle length) (Greenspan, 1997). Genetic markers are commonly used in Drosophila to identify a phenotype and, thus, accurately track genetic crosses by selection of offspring that inherited one version or the other of a chromosome. For example, a geneticist may collect nonCy winged flies to obtain those that inherited the parent’s other copy of the second chromosome with a mutation of interest. However, recombination of the mutation of interest onto the Cy chromosome could occur during meiosis in the parent. To prevent this, the chromosomes with genetic markers also contain multiple inversions to prevent viable recombinant offspring; thus, the only viable offspring are those with one or the other of the parent’s two intact chromosomes (either with the mutation of interest or with the genetic marker). These engineered chromosomes are named “balancer chromosomes.” Besides genetic markers and multiple inversions, balancer chromosomes have a third feature, recessive lethal mutations, which prevent the mutations of interest from being selected out of an inbred population. A turning point of genetic technology that made the fruit fly such a powerful animal model system was the identification and the development of the P-element as a germ-line transformation vector. The P-element is a classic transposable DNA that contains the gene encoding the transposase enzyme, flanked by inverted repeats, which allows movement within the genome. First, Rubin and Spradling replaced the transposase gene with a gene of interest and co-injected the construct into Drosophila embryos with an intact P-element that supplies the transposase enzyme, finding an excellent system for inserting DNA into the fly (Rubin and Spradling, 1982). Then, Brand and Perrimon took the P-element transformation vector and used it to create a gene expression system that would eventually allow for the expression of any gene of interest in any particular tissue within the fly (Brand and Perrimon, 1993). They cloned the yeast transcription factor GAL4 (galactose-induced gene 4) into a P-element vector and showed that a defined 50 Introduction promoter could be cloned upstream of GAL4. To drive the gene expression, they created a corresponding P-element vector, pUAST, containing the upstream activating sequences (UAS), to which GAL4 protein can bind. These UAS sequences were connected to a general promoter and a cloning site to allow for the insertion of any gene of interest. However, the system had some pitfalls such as lack of control over the final resting place of the transposable element, the risk of disruption of an endogenous gene, or the possibility of gene insertion in a locus that may reduce or switch-off the expression. For these reasons, the GAL4-UAS system has been modified and improved over the years (Duffy, 2002). Nowadays, a new transformation system is routinely used derived from the ɸC31 bacteriophage: the pUAST vector containing the UAS sequence and the gene of interest is co-injected in Drosophila embryos along with the integrase ɸC31 mRNA. The integrase ɸC31 is a serine recombinase protein from the bacteriophage that recombines sequences using the attB and attP sequences integration sites in a site specific manner (Groth et al., 2004). For recombination to occur, one of this integration sites is in the pUAST vector and the other is located in the fly genome. 4.3. Drosophila anatomy and physiology The body of the fly is divided into three segments: head (A); thorax (B); and abdomen (C) represented in Figure 17. Firstly, in the head region (A) it can be distinguished: the antennae, which are sensory elements used to detect sound (Matsuo and Kamikouchi, 2013); the compound eye, which is composed by thousands of cells with distinct functions, from photoreception to non-neuronal cells (Cagan, 2009); and the ocellar triangle, housing the ocelli, which are another type of photoreceptor. The organ termed proboscis is a tubular part of the head with various functions: feeding; detection of toxic compounds and of non-volatile pheromones; and, egg laying. The thoracic region (B) is subdivided in three segments, each of them with a pair of legs located on the ventral side. The wings are on the dorsal side of the second thoracic segment; and, a pair of rudimentary halteres, are placed on the next posterior segment (Fig. 17). The halteres are modified hind wings used for balancing during flight (Mureli and Fox, 2015). Specifically, the halteres are mechanosensory organs that provide rapid feedback 51 Introduction to the wing-steering muscles, as well as to the muscles responsible for stabilizing the head. The bristles are also sensory organs placed across the whole surface of the adult fly body. Finally, the abdomen is subdivided in 8 abdominal segments, of which the last four are compressed and modified for reproductive functions. Figure 17. Drosophila external anatomy. The body of the fly follows a segmented pattern and is subdivided into three parts: A, head; B, thorax; and C, abdomen. The head contains the antennae (1, first; and 2, second antennal segment), the compound eye (3) and the ocelli (4). The thoracic region (7) is where the main muscle groups are located and also where the three pairs of legs (5, 8, and 10) and the wings (11) are inserted. The sensory bristles (6) are located through all the body. The abdomen (9) is also segmented and modified for reproductive functions. (Retrieved from https://www.cherrybiotech.com/scientific-note/drosophila-life-cycle-and-fly-anatomy). 4.3.1. Drosophila musculature The three vertebrate muscle types (skeletal, cardiac and smooth) have their anatomical counterparts in Drosophila (skeletal, cardiac and visceral). The skeletal muscle system is the largest organ in motile animals, constituting around 40% of the human body mass and up to 75% of the body mass in Drosophila, of which 65% is comprised by the flight musculature. The skeletal muscle is in charge of the 52 Introduction precision of motor commands and, thus, is highly relevant in many common behaviours, including stabilization of body posture, heading control and directed escape responses. The skeletal muscles of humans and flies are very similar at the molecular and the structural level (Taylor, 2013). Both are composed of tandem arrays of sarcomeres containing the thin and thick filaments, which, in a typical muscle twitch, slide past each other in response to Ca 2+ release from the SR and, eventually, result in force generation. Furthermore, muscle metabolism in humans is also similar to those in the fruit fly with myofibers divided in type I (slow-twitch) and type II (fast-twitch). In both cases, mammals and Drosophila, it has been shown that the muscle is an adaptive tissue that responds to exercise, nutrient supply, and endocrine factors with alterations in myofiber composition and size (Taylor, 2013; Piccirillo et al., 2014). However, motor units are organized slightly differently in invertebrates; all myofibers within a single muscle are simultaneously driven by the same motoneuron (Chakraborty et al., 2015). In vertebrates, each myofiber is innervated by a single motoneuron and, thus, a single muscle is innervated by numerous motoneurons whose synergistic activity controls muscular precision (Fuglevand, 2011). Figure 18. The flight musculature of Drosophila. The flight muscles consist of two groups: the asynchronous indirect flight muscles (A-IFMs) and the synchronous direct flight muscles (DFMs). A) Side view of the A-IFMs located inside the fly thorax. A-IFMs are subdivided into dorsal longitudinal muscles (DLMs; represented in grey) and dorsal ventral muscles (DVMs; dark blue). Structurally, they are arranged in an antagonistic manner. B) Side view of the DFMs located at the fly’s wing hinge and classified according to their insertion site into basalars (b1-3; in grey) and axillaries (I1-2 and III2–4; in dark blue). C) Top view of the location of A-IFMs in the thorax. The numbers (45-48) indicate the number of the shown myofiber. The tergotrochanter muscles (TTM, tergotrochanter) are located in the base of the wing; and functionally they are involved in the jumping and the walking locomotion. Modified from (Lehmann and Bartussek, 2017). c DVM47ac DLM45ac DLM45df TTM DVM46,48 a b DVM 46-48 DLM45 b1 b3 b2 I1 III1 III2-4 I2 A B C 53 Introduction The flight musculature of flies consists of two anatomically, physiologically, and functionally distinct groups: the asynchronous indirect flight muscles (A-IFMs) and the synchronous direct flight muscles (DFMs). The main muscle group is the A-IFMs, located in the thoracic region and subdivided into dorsal longitudinal muscles (DLMs) and dorsal ventral muscles (DVMs) (Fig. 18A). The A-IFMs of insects are characterized by the asynchrony between muscle electrical and mechanical activity, so that their contractions are not triggered by motoneuron spikes in the conventional one-for-one mode (i.e. one impulse-one contraction) like synchronous muscles (Josephson et al., 2000). When these muscles are stimulated by action potentials, they contract in an oscillatory manner and, thus, individual contractions are, instead, activated mechanically by stretch. In this way, A-IFMs are not activated and deactivated in concert with neurogenically controlled cycling of [Ca 2+ ] C but rather are stretching and shortening at a constant level of activating [Ca 2+ ] C (Gordon and Dickinson, 2006; Lehmann et al., 2013). This oscillatory mechanism is possible due to the antagonistic arrangement of DLMs and DVMs that operates as a semi-autonomous oscillator. The elevated power required for flight is, thus, delivered by the A-IFMs that generate wing flapping by the mechanical linkage between the muscles and the wings (Dickinson and Lighton, 1995; Lehmann and Dickinson, 1997). The second minor group is the DFMs, responsible for the subtle changes in wing motion required for rapid manoeuvres and flight stability. These muscles are a set of tiny, highly specialized muscles that insert directly on the wing hinge (Lindsay et al., 2017). The DFMs are classified according to the insertion site into (Fig. 18B): the basalars (b1, b2, and b3) and the axillaries (I1-2 and III2–4). In addition, there are three other sets of thoracic muscles that do not insert directly on the wing hinge: tergopleurals, pleurosternals, and tergotrochanter. The first two sets are situated on the base of the wing and their function is poorly understood. However, they are thought to influence wing motion by altering the mechanics of the thorax. The third set, the tergotrochanter muscles (Fig. 18C; TTM, tergotrochanter), are located just behind the base of the wing and, they are involved in the jumping and the walking locomotion. Finally, there are also other muscles in the legs and others attached to the body wall. 54 Introduction 4.3.2. Drosophila nervous system Drosophila central nervous system (CNS) is composed of the brain and the ventral nerve cord (equivalent to the spinal cord in vertebrates) which is a fusion of thoracic and abdominal ganglia (Fig. 19A). Ganglia are often composed of multiple, fused units termed neuromeres. The adult Drosophila CNS contains over 100,000 neurons (Lovick et al., 2013) and 15,000 glial cells of various types (Kremer et al., 2017), these cells provide neurotransmitter and ionic homeostasis to neurons and serve as immune cells. Although flies and humans show the same subdivision of the CNS, the structure of the fly neurons is different, being unipolar while most mammalian CNS neurons are multipolar (Martin and Krantz, 2014). However, in terms of electrophysiological properties they are similar to mammalian neurons. They fire proper Na + /K + -based action potentials and, they use neurotransmitters for synaptic vesicle release. Also, although Drosophila brain is comprised by a lower number of neurons, compared to humans, their activity regulates similar behavioural functions such as: motor activity, reward and aversion, memory formation, feeding or sexual appetite. The neurons involved in negative geotaxis behaviour are dopaminergic neurons located in the protocerebral anterior medial region (Kasture et al., 2018). The CNS can be subdivided into two histological regions: the cortex regions, where neuronal cell bodies are located; and, the neuropiles areas to which axons and dendrites project and form synaptic connections (Pyza and Meinertzhagen, 1993). The dense synaptic neuropil areas are covered by astrocytes and glia. The peripheral nervous system includes all of the nerves that branch out from the CNS and connect with the sensory organs and the fly musculature (Fig. 19B). Given the semitransparent nature of Drosophila wing, one of the most studied peripheral nerves are the wing sensory neurons that can be used as models of axon injury (Soares et al., 2014). Each wing neuron extends a long axonal projection that merge to form a thick nerve track that connects to the CNS by projecting into the thoracic ganglion located in the ventral nerve cord (Fang and Bonini, 2015) . Structurally, the wing is a thin and simple structure. The most easily visible features of the wing are the veins. These are not veins in the vertebrate sense but ectodermal tubes that serve as structural supports for the wing and as 55 Introduction vessels for haemolymph (analogue to blood in vertebrates), and the peripheral nerves. In the nomenclature most commonly used in Drosophila studies (Fig. 19C), there are five main longitudinal veins (L1–5) that run proximodistally; two smaller abbreviated veins (L0 and L6); and three cross veins, the anterior and posterior crossveins, which bridge L3–L4 and L4–L5, respectively, and the humeral cross vein, which runs between the anterior wing margin and L0. The neurons are mainly located in the anterior wing margin (along L1 and the coastal vein) and, in less number, in the L3 vein (Fig. 19C). Figure 19. Drosophila nervous system. A) The fly central nervous system (CNS; shown in yellow) includes the central brain, located in the head, and a ventral nerve cord (VNC), located in the thoracic cavity. Figure modified from (Namiki et al., 2018). B) Histological regions of the CNS. The brain is flanked by the optic lobes. The cortex regions (grey areas) contain the neuronal and most of the glial cell bodies; the neuropils (light blue) harbor synaptic connections; and the small and large axon tracts (dark blue) connect the different neuropiles. Figure modified from (Batelli et al., 2017). C) An illustration of the peripheral nervous system in the Drosophila wing. There are five main longitudinal veins (L1–5) and two smaller abbreviated veins (L0 and L6); and three crossveins: the anterior, the posterior and, the humeral. The neurons (cell bodies are represented in red; and their projections in green) are located in the anterior wing margin (along L1 and the coastal vein) and, in less number, in the L3 vein. Modified from (Fang and Bonini, 2015). BA optic lobe central brain optic lobe Central brain VNC Tract region Neuropil region Cortex region C Anterior wing margin Posterior wing margin L0 L6 Anterior Posterior Coastal Humeral L1 L2 L3 L4 L5 56 Introduction 4.4. Ca 2+ signalling toolkit in Drosophila Drosophila genome sequencing was completed in 2000 (Adams et al., 2000). This enabled the identification of genes encoding the Ca 2+ signalling toolkit, represented by vertebrate homologs, and most of them single genes. Thus, studying Ca 2+ signalling in Drosophila is considerably simplified (Chorna and Hasan, 2012). The research approach for this study was the generation of Drosophila mutants in Ca 2+ signalling genes. However, given the importance of Ca 2+ signalling in embryogenesis, mutations in genes affecting Ca 2+ signalling proteins were frequently homozygous lethal. In this regard, the use of genetic methods for spatialtemporal expression of mutant genes, such as the GAL4-UAS system, has been of vital importance (Brand and Perrimon, 1993). Ca 2+ entry from the extracellular medium is also mediated by numerous channels located in the plasma membrane and named according to their gating properties. Among Receptor-Operated Calcium Channels (ROCCs), glutamategated ionotropic receptors (iGluRs) are represented in Drosophila by 15 genes encoding different subunits. As in other animal species, Drosophila uses glutamate as a fast neurotransmitter in NMJs, and highly Ca 2+ permeable iGluRs are clustered in active zones in postsynaptic motoneuron terminals (DiAntonio, 2006). As mentioned before, vertebrate Voltage-Operated Calcium Channels (VOCCs) are classified in three families (Ca V 1, Ca V 2 and Ca V 3) according to the pore forming subunit α1 (Catterall, 2011). The Drosophila genome also encodes three α1 subunits (Dmca1D, Dmca1A, and Ca-α1T), which can be classified as Ca V 1, Ca V 2 and Ca V 3 channels, respectively. Channels formed by Dmca1D are dihydropyridine (DHP) sensitive, similar to L-type channels of vertebrates, and are expressed in the adult muscles (Eberl et al., 1998). The Dmca1A channel is encoded by the cacophony (cac) gene, it is insensitive to DHP and widely expressed in the embryonic nervous system (Smith et al., 1996). The transient receptor potential (TRP) superfamily includes 28 mammalian members, which are subdivided into multiple subfamilies. Each of these subfamilies is represented by at least one of the 13 members found in Drosophila, suggesting common evolutionary relationships. Importantly, about the Store-Operated Ca 2+ Entry (SOCE) players, Drosophila genome contains single STIM (dSTIM) and Orai 57 Introduction (dOrai) genes, whereas mammals have two and three copies, respectively (Venkiteswaran and Hasan, 2009). Cytosolic Ca 2+ exit to the extracellular medium is mediated by pumps and exchangers. In contrast to humans and other mammals, where there are four PMCA isoforms (PMCA1, 2, 3 and 4) encoded by distinct genes, the Drosophila genome contains a single gene (CG2165) encoding a plasma membrane Ca 2+ - ATPase, which is expressed in all tissues, including the muscle (Bai et al., 2008). The Drosophila Na + /Ca 2+ exchanger, CALX, shares 55% identity with the three mammalian isoforms (NCX1, NCX2 and NCX3) and is mainly expressed in the fly adult brain and in the eye (Ruknudin et al., 1997). Drosophila also has a homolog for the vertebrate Na + /Ca 2+ /K + exchanger, NCKX30C, which is expressed in the adult brain, in larval imaginal discs and during embryonic development (HaugCollet et al., 1999). In regard to the Ca 2+ buffering systems, a Mitochondrial Calcium Uniporter (MCU) homologue has been identified and characterized in Drosophila, the gene CG18769 (Walkinshaw et al., 2015; Drago and Davis, 2016). A homolog of the mammalian chaperone BiP has also been identified in Drosophila, known as heat shock cognate 70 (HSC70-3) (Rubin et al., 1993). Importantly, there is no protein similar to calsequestrin (CSQ), abundant in the muscle and with Ca 2+ binding properties, in Drosophila. However, the CaBP characteristic of the ER of nonmuscle cells, calreticulin (CR), is encoded in the fly by the Calr gene (Smith, 1992). Finally, regarding the intracellular Ca 2+ transport systems, a single gene dubbed Ca-P60A, that encodes SERCA protein was identified (Magyar and Váradi, 1990). It displays a higher homology with mammalian SERCA1 and SERCA2 (71–73%) than with SERCA3 (67–69%). It is expressed at high levels in the central nervous system and in the muscles (Magyar et al., 1995). A dominant temperature-sensitive loss-of-function allele of the Ca-P60A gene (Ca-P60AKum170) was isolated in a screen for temperature sensitive paralytic mutants and has been very valuable for understanding the physiological role of SERCA in Drosophila (Sanyal et al., 2005, 2006). The homologous gene of mammalian Secretory Pathway Ca 2+ -ATPase (SPCA) in Drosophila is known as SPoCk or CG32451, that produces three isoforms (SPoCk-A-C) each localized in different cellular compartments: SPoCk-A 64 Methods 67 Methods 1. Gene construction The fusion gene encoding GAP3 targeted to the ER (erGAP3) was cloned into a pUASTattB vector to generate pUASTattB-erGAP3. erGAP3 protein contains the calreticulin (CR) signal peptide sequence fused to the amino-terminus of the GAP protein and the ER retention signal KDEL, the tetrapeptide sequence, in the carboxy-terminus of the protein (Navas-Navarro et al., 2016). The fusion gene erGAP3 was expressed under the control of the upstream activating sequence (UAS) promoter in the pUASTattB plasmid. This vector is designed to integrate a sequence of interest into the fly genome and contains the following genetic elements: the white selectable marker that increases pigment levels in the eye, conferring red colour 4 ; the loxP recombination site; the UAS-erGAP3-SV40 expression cassette; and an attB integration site (Fig. 20). Basic molecular biology techniques were used in the cloning: isolation of plasmid DNA by miniprep (PureYield™ Plasmid Miniprep; Promega, A1222) and maxiprep (Quiagen, 12163) protocols; bacterial transformation by heat shock method (DH5α™ Competent Cells; ThermoFisher, 18258012); DNA agarose gel electrophoresis; DNA gel purification (Wizard® SV Gel and PCR Clean-Up System; Promega, A9282); the Polymerase Chain Reaction (PCR); restriction enzyme digests; DNA ligation reaction; primer design; and, DNA quantification by NanoDrop™ spectrophotometer. The resulting DNA plasmid was sequenced by the Sanger method (Secugen S.L) to verify the construct integrity. The SnapGene® software version 5.0 was used to plan, visualize, and document molecular biology procedures. 2. Generation of transgenic flies To generate transgenic flies, the pUASTattB-erGAP3 vector above described was microinjected into Drosophila embryos ‘‘y1 M{vas-int.Dm}ZH-2A w*; M{3xP3RFP.attP} ZH86Fb’’ (BL24749 strain; BestGene® Inc.), containing the locus 86F in the third chromosome which includes an attP integration sequence and it lacks the 4 Mutations on the white gene produce viable flies with white eyes, and hence, its name. 68 Methods white marker that affects eye colour (Fig. 20). This locus is also composed by the artificial 3xP3 promoter followed by a Red Fluorescent Protein (RFP) marker flanked by loxP recombination sequences. The pUASTattB-erGAP3 vector was microinjected along with the integrase ɸC31 mRNA, which is required to integrate the UAS-erGAP3-SV40 expression cassette into the fly genome (Bischof et al., 2007). After the integration has occurred, the Cre recombinase mRNA was microinjected to eliminate the RFP and the white markers (Chen et al., 2013). Figure 20. Generation of UAS-erGAP3 transgenic flies. The integrase ɸC31 catalyses the integration of pUASTattB-erGAP3 vector in locus 86F located in the third chromosome (Chr.3) of the fly genome by attB and attP recombination and, thus, generating the attR and attL sites. Next, Cre recombinase deletes the Red Fluorescent Protein (RFP) and the white markers flanked by the two more distant loxP sites. In order to express erGAP3 in muscle or neuronal tissues, we used the GAL4-UAS system (Brand and Perrimon, 1993; Hampel et al., 2011). This system consists of the GAL4 (galactose-induced gene 4) transcription factor that specifically binds the UAS sequence (Fig. 21). The UAS-erGAP3 flies (Navas-Navarro et al., 2016), generated by the ɸC31 system, were crossed with flies expressing the transcription factor GAL4 under the control of the muscle specific myosin heavy chain (Mhc) promoter (Mitra et al., 1996) or the nervous system specific embryonic lethal abnormal visual system (elav) promoter (Yao et al., 1993). The elav gene is expressed in most or all types of neurons and, thus, is referred to as pan-neuronal. In addition, the cytosolic Ca 2+ indicator GCaMP3 was expressed in the muscle, by White loxP attB ɸ C31 attP loxP loxP pUASTattBerGAP3 Locus 86F ( Chr. 3) UAS-erGAP3-SV40 3 xP3-RFP loxP loxP White loxP attR Recombinase Cre 3 xP3-RFP UAS-erGAP3-SV40 attL 69 Methods crossing the UAS-GCaMP3 flies (Tian et al., 2009) with the Mhc-GAL4 flies. We assigned each line a specific code 5 , summarized in the table below (Table 2). The wild type flies, used as control, were from the w 1118 strain; as the transgenic flies were constructed in this white genetic background. Fly line Genotype GECI Targeted tissue Targeted organelle DM006 Mhc-GAL4 // erGAP3 erGAP3 muscle SR DN005 Elav-GAL4 // erGAP3 erGAP3 nervous system ER DM010 Mhc-GAL4 // GCaMP3 GCaMP3 muscle Cytosol Wild type w 1118 - - - Table 2. Fly lines used in this thesis. 3. Lifespan assay Fly stocks were maintained in an incubator (Ibercex, M-17004) at 29 °C, constant 12 h 6 light/dark cycles, and 60% humidity. Flies hatched in the same day were collected and sorted by gender. So that flies can be sexed and scored, they were anaesthetised with CO 2 for 1-10 min using a square CO 2 station. The station is composed by a holed base covered with white felt, which is a porous material designed to permit CO 2 to flow through, while preventing Drosophila from falling through the holes beneath. The station is fed by a CO 2 tank and the gas flow is regulated by a valve system. After sorting, groups of 20 flies were placed into food vials. The food was prepared with: 1 packet (177.67g) of Nutri-Fly® Molasses Formulation (Genesee Scientific, 66-112); 0.9% (v/v) propionic acid (SigmaAldrich, 81910); and, 0.6% (w/v) Agar type I (iNtron Biotechnology, 25999); diluted in 1 L of bottled water. Changes onto new vials with fresh food were made every two or three days in order to avoid larvae disrupting food. In each change of food, 5 DM= Drosophila muscles; DN= Drosophila neurons. 6 Please note that, along the document, the most common units of: time, concentration, mass, distance, voltage, revolutions, frequency, and others; are written in symbols, according to the metric system. 70 Methods we counted the dead flies in the old vial and, also, the dead flies carried onto the new vial in order not to double-count the same dead flies. Flies that escaped or were dead by accident, immediately left the experiment. The assay ended when the last survivor was dead. The data were represented as the percentages of survival at the day indicated. Figure 21. Generation of erGAP3 transgenic muscle and neuronal lines. The GAL4UAS system allowed tissue-specific expression of erGAP3 sensor. Flies expressing the GAL4 transcription factor (pink) downstream the tissue specific promoter myosin heavy chain (Mhc; represented in blue) were crossed with UAS-erGAP3 flies. As a result, in the first generation, GAL4 transcription factor is only expressed in the skeletal muscle tissue where it binds to the UAS sequence (yellow) allowing the expression of erGAP3 protein. The same procedure was followed to generate the neuronal transgenic line with the pan neuronal tissue specific promoter embryonic lethal abnormal visual system (elav). Mhc - GAL4 // UAS - erGAP3 Elav-GAL4// UAS-erGAP3 erGAP3 X UAS - erGAP3 Mhc - GAL4 Elav-GAL4 erGAP3 UAS GAL4 Mhc 71 Methods 4. Climbing assay The climbing assay is a well standardized procedure based on Drosophila negative geotaxis (their tendency to go up) that can be used as an indicator of motor function across age in flies. The aim is to measure a known distance climbed in a certain time, in this case we used 5 cm in 18 s. The day prior to the experiment, we changed the flies of a known age onto vials with fresh food. The day of the experiment, groups of ten flies were transferred onto vials without food and previously labelled with the known height mark of 5 cm. The experiment was always performed at the same time of the day, early in the morning, to avoid bias due to changes in the circadian rhythms of the flies. We let them acclimate to the new environment for a few min and, then, the experiment was started. The vials were tapped vigorously onto the table surface making sure that all the flies were placed onto the bottom of the tube. After three taps we counted simultaneously the number of flies able to climb 5 cm in 18 s, and the time in which 50% of the flies surpassed the 5 cm mark. We let the flies rest for 15 min and, then, the assay was repeated another 3 times with the same group of flies. The climbing assays were performed on days 7 (i.e., when the flies were 7 days old), 14, 21, 28 and 35. 5. Protein induction and extraction First, the gene encoding GAP3 was cloned into the bacterial expression vector pET28a and transformed in E. coli BL21 strain (Stratagene). Second, bacteria were grown in 3 ml of Luria-Bertani Broth medium (LB Broth; Sigma-Aldrich, L3022) containing 40 µg/ml of kanamycin (Sigma-Aldrich, K4000), at 37°C and 250 rpm for 12-15 h. The next day, the bacterial cultures were diluted 50 times in the same medium containing LB and kanamycin, and grown until the optical density of the sample, measured at a wavelength of 600 nm, was between 0.6 and 1 (OD>0.6); which was usually 2 h. The optical density was measured in the UV-visible spectrophotometer (Thermo TM electro corporation uv1). Then, protein expression was induced by the addition of 0.01 mM isopropyl β-D-thiogalactoside (IPTG; Sigma-Aldrich, I5502) for 6 h at 30°C and 250 rpm. The cells were, then, pelleted by centrifugation at 6.000 g for 10 min; and, resuspended in a buffer containing: 72 Methods 150 mM NaCl (Merck; 106406); 0.5 mM DTT (Dithiothreitol; Sigma-Aldrich, D9760); 20 mM Tris-HCl (Trizma® base; Sigma-Aldrich, T1503); pH 8.8; diluted in Milli-Q® ultrapure water; and, a cocktail of protease inhibitors with EDTA (Roche, 11836153001) diluted 10 times in the buffer. Next, the cells were lysed by sonication (Vibra cell 75115, Biolock Scientific). The sonication protocol was performed on ice and consisted on 5 cycles, 15 s ON/ 1 min OFF, at 30% vibration amplitude. Finally, the bacterial lysate was centrifuged at 30.000 g for 10 min, and the supernatant was used as a source of GAP protein. The supernatant was aliquoted, frozen in liquid nitrogen, and stored at -80°C. 6. Protein quantification Total protein extracts were quantified by the Bradford assay. The procedure is based on the formation of a complex between the dye, Brilliant Blue G, and the proteins in the solution. The protein-dye complex causes a shift in the absorption maximum of the dye from 465 to 595 nm, measured in the UV-visible spectrophotometer. The Bradford Reagent concentrate (Bio-Rad, 500-006) is prepared by diluting 1 part of reagent in 4 parts of water. Then, we mix 1 part of the protein sample with 30 parts of the Bradford Reagent. Protein concentrations were determined by comparison to a standard curve. The standard curve was performed with bovine serum albumin (BSA; Sigma-Aldrich, A2153) diluted in 50 mM Tris-HCl, pH 8.8; in a concentration range between 0 and 2 mg/ml. 7. Fluorescence spectra Fluorescence spectra for GAP3 protein were performed in fluorescence spectrophotometer (HITACHI 650-150) equipped with a xenon lamp. The excitation spectra were recorded between 380 and 500 nm wavelengths with the emission set at 520 nm. First, the spectrophotometer cuvette was filled with 1 ml of extracellular medium (without Ca 2+ ) composed of: 145 mM NaCl; 5 mM KCl (Merck; 104938); 1 mM MgCl 2 (Merck; 102367); 10 mM glucose (Merck; 108342); and, 10 mM Na-HEPES (Sigma-Aldrich, H3375); pH 7.4; diluted in Milli-Q® 73 Methods ultrapure water; and 8.5 µg of the protein extract containing GAP3 7 . Second, the same recording was performed with the addition of 1 mM CaCl 2 (Merck; 102382) to the extracellular medium. Then, a new sample was prepared with another 8.5 µg of the same protein extract in extracellular medium; and, immediately heated for 10 min in an Eppendorf tube® at 50°C in the thermal bath. After heating, another spectrum was recorded under the same conditions, i.e. in the absence and in the presence of Ca 2+ . 8. Immunofluorescence Larvae were dissected under the Stereo Microscope (ZEISS Stemi DV4) in HL3 buffer containing: 70 mM NaCl; 5 mM KCl; 10 mM NaHCO 3 (Merck, 106329); 5 mM Na-HEPES; 5 mM trehalose (Merck, 108216); 115 mM sucrose (Merck; 107651); 1.5 mM CaCl 2 ; and, 20 mM MgCl 2 ; pH 7.2 in Milli-Q® ultrapure water. First, larvae were dorsally immobilized in the dissection plate with two pins, one placed in the anterior part of the animal and the other in the posterior part; second, larvae were opened through the mid-dorsal line with micro-scissors; third, the cuticle was opened and fixed very straight with four pins (two anterior and two posterior), making visible all the internal organs; finally, the internal organs were removed with the tweezers, leaving intact the CNS and the muscle wall. This larvae preparations were fixed for 10 min in 4% (v/v) paraformaldehyde (PFA; Sigma-Aldrich, 76240) diluted in phosphate-buffered saline 1X (PBS; ThermoFisher, Gibco™, 10010023) pH 7.4. PFA was removed by washing with PBS 1X for three times, then, tissues were permeabilized for 1 h with the permeabilization solution: 0.15% (v/v) Triton X100 (TX-100; Merck, 108603); 5% (v/v) goat serum (GS; ThermoFisher, Gibco™, 16210064) and 2% (w/v) BSA; diluted in PBS 1X. After three washes with PBS 1X, larvae were incubated overnight with the SERCA primary antibody (1:500; Sanyal et al., 2005) diluted in the permeabilization solution without TX-100 detergent. The next day, after removing the primary antibody, the samples were washed with PBS 1X and revealed with the corresponding Alexa Fluor 568-conjugated secondary antibody (1:100; ThermoFisher, A11031), by incubation for 1 h at room 7 Obtained as described in section 5.Protein induction and extraction. 80 Methods with the “Limit to Threshold” checked in the “analyse” menu, so that only thresholded pixels were included in the measurement calculations. Figure 22. Sarcoplasmic reticulum Ca 2+ imaging in vivo in Drosophila. Flies were immobilized in a small drop of agar and two fluorescence images at each GAP3 wavelength (405 and 470 nm) were acquired in the muscles of the fly thorax under resting conditions (basal levels). Then, to determine R min (sarcoplasmic reticulum depletion), each fly was carefully heated in the thermal bath for 10 min at 50 °C and, then, imaged using the same settings. Fluorescence images were acquired with a fluorescence stereo microscope (M205FA, Leica) using either 470/40 nm or 405/40 excitation filters and a 525/50 nm emission filter, 16 bits, 4X4 binning and 6.5X zoom. Images were analysed with the Image J software and the measurements were performed in the region of interest (ROI, represented by a white square in the fly image) in the thorax of each fly. 12.2. Measuring cytosolic Ca 2+ in Drosophila Flies of the DM010 line (Table 2) were anesthetized and immobilized in agarose as previously described. Recordings of the dorsal longitudinal muscles in the fly thorax were obtained by stimulation with CS-20 stimulator (Cibertec) of the giant fiber system (GFS) through two tungsten electrodes impaled into the fly eyes (Allen and Godenschwege, 2010). The Drosophila GFS is a well-characterized neuronal Rmin R470/405 50 °C 10 min Mhc:erGAP3 Basal levels SR depletion F470 (GAP) F405 (GAP) 81 Methods circuit that mediates the escaping behaviour in the fly (Fig. 23). It is named after the two largest interneurons in the fly, the giant fibers (GFs), which pass along the signal from the brain to the mesothoracic ganglion, located in the ventral nerve cord. In the mesothoracic ganglion each giant fiber makes two synapses: one to a large motoneuron (TTMn) that drives the tergotrochanter “jump” muscle (TTM); and another to the peripherally synapsing interneuron (PSI), which, in turn, synapses the dorsal longitudinal motoneurons (DLMns) of the dorsal longitudinal (flight) muscles (DLMs). These muscles are, along with the dorsoventral muscles (DVMs), the main muscle group located in the thoracic region, the A-IFMs 11 . By stimulating the giant fiber neurons directly in the brain of the adult fly we can obtain recordings from the flight muscles. Figure 23. Electrical stimulation protocol in Drosophila. Flies were immobilized in agarose, and two tungsten electrodes (stimulating electrodes) were impaled into the fly eyes in order to stimulate the two giant fibers (GFs; in red). The GFs pass along the signal from the brain to the mesothoracic ganglion, where each giant fiber makes two synapses: one to a large motoneuron (TTMn; in purple), that drives the tergotrochanter “jump” muscle (TTM; in blue); and another to the peripherally synapsing interneuron (PSI; in green), which, in turn, synapses the dorsal longitudinal motoneurons (DLMns; in yellow) that drive the flight muscles (DLMs; in red). Taken from (Allen and Godenschwege, 2010). 11 See Figure 12 of the introduction, in section 6.3.2. Drosophila Muscles. 82 Methods The stimulation protocol consisted of repetitive stimulation with 5 V/50 ms stimuli at different frequencies (0.5 or 10 Hz), and was performed during 5 s. Each experiment started with a visual check of twitches in the wing after stimulation to assure that the electrodes were correctly positioned. For these experiments, fluorescence images were acquired every 100 ms only at the 470 nm wavelength 12 , using the 470/40 nm excitation filter and the 525/50 nm emission filter. Images were acquired and analysed 13 as mentioned previously. To normalize the results obtained in different experiments fluorescence (F) was divided by the average of the F values obtained during the first 5–10 frames (F 0 ), resulting in F/F 0 . 12.3. Ca 2+ measurements in cells Cell imaging was performed in a Zeiss Axioplan 2 upright microscope equipped with a 20X water-immersion objective (W-Achroplan, Zeiss; NA = 0.5) and a Zeiss AxioCam camera MRm (12 bit) connected through a software interface (Axiovision Rel 4.6.3, Zeiss) to a 75 W Xenon fluorescent excitation source (Osram, XBO75) and a filter wheel. GAP3 was excited sequentially at 405 and 470 nm using the dichroic mirror 505DRPL (XF2031) and the 405IL25 (Comar, T-GQD) and 470DF35 (Omega, XF1013) excitation filters, coupled in the filter wheel. Fluorescence emission was registered between 517-552 nm with 535DF35 (Omega, XF3007) emission filter. Images were acquired with 5x5 binning sequentially at each wavelength every 10 s. All experiments were recorded at 25 °C. Neutral density filters (DN 1.0 and DN 0.5, Omega Optical 142588 and 164526, respectively) were also used to prevent bleaching. Briefly, 12 mm-diameter round coverslips were placed in the aluminium chamber and carefully adjusted in the center with a silicone piece that contains two grooves, one for the perfusion and the other for the vacuum. The chamber is placed in a specific platform that is then set in the microscope and connected to the valves12 The fluorescent spectra of GCaMP proteins display a single excitation peak at 489 nm and an emission peak at 509 nm. 13 GCaMP3 is intensiometric, the protein increases its fluorescence intensity upon binding to Ca 2+ ions. Therefore, the analysis is performed only in the F 470 wavelength. With the acquisition of only one wavelength, the calculation of the ratio is not possible; and, thus, these measurements are not calibrated into [Ca 2+ ]. 83 Methods perfusion system. This is a gravity, bath perfusion system composed by eight channels, each of them can contain a different solution. These channels are controlled by the valve controller which allows quick switching of the solutions flowing into the perfusion chamber. First, cells were perfused at 5 ml/min with extracellular medium (145 mM NaCl; 5 mM KCl; 1 mM CaCl 2 ; 1 mM MgCl 2 ; 10 mM glucose; and 10 mM Na-HEPES; pH 7.4) for 4-6 min. Then, cells were stimulated with either ATP (100 µM; Sigma-Aldrich, A3377), an agonist of ionotropic (P2X) and metabotropic (P2Y) purinergic receptors; or carbachol (CCh, 100 µM; SigmaAldrich, C4382) as an agonist of the acetylcholine receptors (AChR; both muscarinic and nicotinic); for 30 s. All the agonists were diluted in extracellular medium. Finally, at the end of each recording, the ER was depleted by stimulating with Ca 2+ -free extracellular medium (145 mM NaCl; 5 mM KCl; 1 mM MgCl 2 ; 10 mM glucose; 0.5 mM EGTA (Ethylene Glycol Tetraacetic Acid; Sigma-Aldrich, E4378); and 10 mM Na-HEPES; pH 7.4) containing ATP (100 µM), and the SERCA reversible inhibitor TBH (10 µM; Sigma-Aldrich, 112976), for 6-8 min. For the heating experiments, immediately after the recording, the coverslips were placed in extracellular medium at 50 °C for 10 min. Then, the same protocol was applied and fluorescence images were captured using the same settings. The output images were analysed using the ImageJ software: first background was subtracted; second, a threshold was set for each image stack, which correspond to the individual recording for the two GAP3 wavelengths; finally, pixel-by-pixel ratios were calculated by dividing the two individual wavelengths with the Image Calculator plugin. The regions of interest (ROIs) were carefully drawn for each individual cell and, the measurements were performed with the “Limit to Threshold” checked in the “analyse” menu, so that only thresholded pixels were included in the measurement calculations. For GAP3, the ratio (R=F 470 /F 405 ) was used as an index of [Ca 2+ ] ER and in order to normalize the results obtained in different experiments this ratio was divided by the average of the R values obtained during the first 5–10 frames (R 0 ), resulting in R/R 0 . To measure cytosolic Ca 2+ in vitro we used the fluorescent Ca 2+ -dye Fluo-4 AM (ThermoFisher, F14201). Cells were loaded with 2 µM of the dye diluted in extracellular medium. The incubation time was of 45 min, during which cells were maintained at RT and in dark conditions. Cell imaging was performed in the same 84 Methods microscope, and the acquisition conditions were the same as for the GAP3 470 nm wavelength. Briefly, Fluo-4 loaded cells were excited at 470 nm using the 470DF35 (Omega, XF1013) excitation filter, and fluorescence emission was acquired every 5 s between 517-552 nm with 535DF35. The protocol was as follows: first, cells were challenged with ATP (100 uM) diluted in extracellular medium for 30 s. Then, the same cells were heated as explained previously and the ATP response was checked with another 30 s pulse. Finally, cells were permeabilized with digitonin (50 µM; Sigma-Aldrich, D141) for 5 min diluted in Ca 2+ -free extracellular medium. For cytosolic measurements with Fluo-4 14 , the image analysis was performed only in the F 470 wavelength image stack. The results were also normalized by the average of the F values obtained during the first 5-10 frames (F 0 ) of each experiment, resulting, in this case, in F/F 0. 12.4. Calibration of Ca 2+ measurements The transformation of GAP3 fluorescence data into [Ca 2+ ] ER is based on the following equation (1):  =       (Equation 1) Being “Y” the normalized fluorescence with values between 0 and 1; “R” the ratio (F 470 /F 405 ) at a given time point of the experiment; “R min ” is the “R” when GAP sensor is not bound to Ca 2+ , which is obtained at the end of each experiment by depleting the ER with EGTA, IP 3 -producing agonists and SERCA inhibitors (in vitro) or by heating 15 (in vivo); and “R max ” is a calculated by multiplying “R min ” by the dynamic range (DR) of the sensor, in this case 3. After calculating “Y”, [Ca 2+ ] ER can be calculated with the Hill equation (2):  =   ·[  ]     ·[  ]  (Equation 2) 14 Fluo-4 is intensiometric, as the indicator GCaMP3 used for cytosolic Ca 2+ measurements in flies. Therefore, the analysis is performed only in the F 470 wavelength, and, these measurements are not calibrated into [Ca 2+ ]. 15 This calibration method was developed in this thesis for in vivo measurements, but it also proved to be valuable in vitro. See results section 2 . 85 Methods Equalizing the equations number (1) and (2), and, assuming V max is 1, we can solve for [Ca 2+ ] ER (3): [  ] =   (   )   (Equation 3) It has been determined, both in vitro and in situ (Navas-Navarro et al., 2016), that the coefficient Hill value (n) for the binding of GAP3 to Ca 2+ is 1. Substituting this value (n=1) in equation number (3) will lead us to the same equation (4) that has been previously described to calibrate Fura-2 measurements (Grynkiewicz et al., 1985): [  ] =   ·(  )    (Equation 4) Knowing the DR of GAP3 (R max /R min ) is 3.0, and the K D (Ca 2+ -GAP3 dissociation constant) is 489 µM (Alonso et al., 2017b) an approximate calibration of the [Ca 2+ ] signals can be performed using R/R min values according to the equation (5):    =  + ( − )( [  ]    [  ] ) (Equation 5) 13. Statistical analysis The software OriginLab® (version 7 or Pro 8) was used to graph, analyse and interpret data. Results are expressed as mean ± SD (standard deviation) or mean ± SEM (standard error of the mean), as indicated. The statistical significance was evaluated using Student’s t-test or One-way ANOVA with GraphPad InStat® software version 3.0. The significance level is shown as * for p<0.05; ** for p<0.01; and, *** for p<0.005. The vector graphics software Adobe Illustrator® version CS5 was used to create the final figures. 86 Methods 14. Reagents and resources Reagent Source Identifier Antibodies SERCA Drosophila (Sanyal et al., 2005) Dr.Ramaswami Ryanodine receptor Drosophila (Gao et al., 2013) Robert Scott GRP78/BiP Sigma-Aldrich G9043 Tubulin Sigma-Aldrich T6074 Goat anti Mouse IgG (H/L):HRP Bio-Rad 170-6516 Goat anti Rabbit IgG (H/L):HRP Bio-Rad 170-6515 Goat anti Guinea Pig IgG (H/L):HRP Bio-Rad AHP863P Alexa Fluor 568 goat anti-rabbit ThermoFisher A-11031 Bacterial strains MAX Efficiency™ DH5α™ Competent Cells ThermoFisher 18258012 BL21(DE3) Epicurian Coli Competent Cells Stratagene N/A Chemicals Tissue-Tek® Sakura Finetek 25608-930 Paraformaldehyde (PFA) ThermoFisher 76240 Phosphate saline buffer (PBS) Sigma-Aldrich P5119 TritonX-100 Merck 1086031000 Goat serum (GS) ThermoFisher 16210064 Bovine serum albumin (BSA) Sigma-Aldrich A2153 Hoescht 33322 cell permeant ThermoFisher H3570 Vectashield® Vector laboratories H-1000 Propionic acid Sigma-Aldrich 81910 Nutri-Fly® Genesee Scientific 66-112 Agar type I iNtron Biotechnology 25999 Agarose low gelling temperature Sigma-Aldrich A6560 Lipofectamine® 2000 ThermoFisher 1668-019 Geneticin® (G-418) ThermoFisher 11811-023 Dulbecco’s modified Eagle’s medium ThermoFisher 10566-016 Penicillin Streptomycin Lonza DE17-602-E Foetal bovine serum (FBS) Lonza DE-14-801F Trypsin-EDTA ThermoFisher 25200-056 Poly-L-lysine Sigma-Aldrich P1274 OPTI-MEM™-I ThermoFisher 11524456 Hank’s balanced salt solution ThermoFisher 14175-053 Papain Worthington Biochemical Corporation LS003119 DNase Roche 10104159001 Neurobasal medium ThermoFisher 17504-044 glutaMAX™-I ThermoFisher 35050-038 87 Methods B27 ThermoFisher 177504-044 Horse serum (HS) ThermoFisher 16050122 Isopropyl b-D-thiogalactoside (IPTG) Sigma-Aldrich I5502 DTT (Dithiothreitol) Sigma-Aldrich D9760 Digitonin Sigma-Aldrich D141 Fluo-4, AM, cell permeant ThermoFisher F14201 ATP Sigma-Aldrich A3377 Carbachol (CCh) Sigma-Aldrich C4382 EGTA (Ethylene Glycol Tetraacetic Acid) ThermoFisher E1219 Tert-Butylhydroquinone (TBH) Sigma-Aldrich 112976 Sodium dodecyl sulphate (SDS) Sigma-Aldrich L3771 Sodium deoxycholate Sigma-Aldrich D6750 Trizma® base Sigma-Aldrich T1503 Glycerol Sigma-Aldrich G5516 Bromophenol blue Sigma-Aldrich B0126 Glycine Sigma-Aldrich G8898 Methanol Merck 106009 Protein Assay Dye Reagent Concentrate Bio-Rad 500-006 Acrylamide Bio-Rad 161-0158 Precision Plus Protein All Blue Standards Bio-Rad 161-0373 TWEEN® 20 Sigma-Aldrich P1379 Defatted dry milk Nestle Sveltesse Nestlé España, S.A p-Coumaric acid Sigma-Aldrich C-90008 Luminol Sigma-Aldrich A-8511 Hydrogen peroxide (H 2 O 2 ) Sigma-Aldrich H-1009 Sodium chloride (NaCl) Merck 1064060500 Potassium chloride (KCl) Merck 1049380500 Calcium chloride (CaCl 2 ) Merck 10035-04-8 Magnesium chloride (MgCl 2 ) Merck 7791-18-6 Glucose Sigma-Aldrich G7021 Sucrose Merck 107651 Trehalose Sigma-Aldrich T0167 Sodium bicarbonate (NaHCO 3 ) Merck 106329 Na-HEPES Sigma-Aldrich 75277-39-3 LBroth Sigma-Aldrich L3152 LB-Agar Sigma-Aldrich L3027 Ampicillin ThermoFisher 11593027 Kanamycin Sigma-Aldrich K4000 Critical Commercial Assays SuperSignal™ West Pico PLUS Chemiluminescent Substrate ThermoFisher 34577 Plasmid Maxi Kit QIAGEN 12163 Wizard® SV Gel and PCR Clean-Up System Promega A9282 88 Methods PureYield™ Plasmid Miniprep Promega A1222 Experimental Models: Organisms/strains w[1118]; [+]; [+] Lola Ganfornina (IBGM) N/A y[1] w[*]; M{UAS-erGAP3}ZH-86Fb Bloomington Drosophila Stock Center 80904 w[*]; P{w[+mC]=Mhc-GAL4.K}2/TM3, Sb[1] Bloomington Drosophila Stock Center 55133 y[1] w[*];P{w[+mC]= Mhc - GAL4.K}2/TM3, Sb[1] M{UASerGAP3}ZH-86Fb (Navas-Navarro et al., 2016) N/A P{w[+mW.hs]=GawB}elav-GAL4; w[*];Gal80 TS / TM3 Ser A. Ferrús (Instituto Cajal, Madrid, Spain) N/A P{w[+mW.hs]=GawB} elav - GAL4 ; w[*]; M{UAS-erGAP3}ZH-86Fb This thesis N/A erGAP3 transgenic mice C57BL/6JCrl (Charles Rivers) (Navas-Navarro et al., 2016) N/A Experimental Models: Cell Lines Human: HeLa cells ATTC CCL-2 HeLa stably-expressing erGAP3 (Navas-Navarro et al., 2016) N/A HEK293T ATTC CRL-3216 C2C12 ATTC CRL-1772 Recombinant DNA Plasmid: pUASTattB-erGAP3 (Navas-Navarro et al., 2016) N/A Plasmid: pcDNA3-erGAP3 Addgene 78118 Plasmid: pCMV-G-CEPIA1er Addgene 58215 Software and Algorithms Axiovision Rel Version 4.6.3 Zeiss Image J Version 1.52a NIH, USA Graphpad InStat® Version 3.0 Graphpad Leica Application suite X Version LasX Leica SnapGene® Version 5.0 SnapGene OriginLab® 7 or Pro 8 OriginLab Table 5. Reagents and resources used in this thesis. Results 96 Results WT flies had a mean lifespan (50% survival) of 25 days (Fig. 26A), with the survival of the females being longer than that of the males: for example, at day 33, the survival is 0% for male flies whereas it is still 45% for female flies. Regarding the muscle line (Fig. 26B), the mean lifespan is 26 days, very similar to that of the WT line, whereas for the neuronal line (Fig. 26C), the mean lifespan is slightly higher, 30 days. In both transgenic lines, female flies also lived longer than male flies. For example, in the muscle line, at 19 days the survival of the males is 50% whereas that of the females is of 97%. In the neuronal line, at the same time point, the survival is 88% for both genders, however at day 28 it is 79% for females and 54% for males. The maximal lifespan reached values of 33-37 for male flies and 40-44 in female flies. In the three lines, survival slowly declines until day 12-25 and, then, between 25-32 days the survival is around 50%. At that time point, flies started to die abruptly until there are no individuals alive at day 33-40. From this point onwards, we divided the cohort into five age groups: 7, 14, 21, 28 (males and females) and 35 (only for females) days. The mean and the maximal longevity values are summarized in Table 6. To conclude, the expression of the erGAP3 indicator did not alter the fly survival, as no significant changes in these parameters are observed when transgenic lines were compared with WT flies from the w 1118 strain (Table 6). Wild type DM006 line DN005 line Mean lifespan (days) 25 26 30 Maximal lifespan (days) 40 44 40 Table 6. Mean and maximal longevity values obtained from the survival curves shown in Fig. 26. 1.4. Monitoring muscle function over age with the climbing assay Being the progressive loss of muscle function, known as sarcopenia, a hallmark of aging, we decided to characterize the loss of function of the skeletal muscle in DM006 transgenic flies by the well-characterized climbing assay (Fig. 27). 97 Results Figure 27. Climbing ability over age in the erGAP3 transgenic muscle fly line. The figure shows the climbing assay, that measures the muscle function, performed in DM006 flies (erGAP3 transgenic muscle line) over age (7, 14, 21, 28 and 35 days). The data are represented in mean ± SEM of groups of 20 flies, being 10 flies used for each individual experiment. A) The plot represents the number of flies (in percentage) able to climb 5 cm height in 18 s over age (in days). This number significantly declines over age. The last four points fit a linear correlation, with a coefficient of 0.997 (p<0.005). B) The plot represents the time (in seconds) in which 50% of the flies performing a given experiment surpass the 5 cm mark over age (in days). The differences between the means of consecutive age groups were statistically significant for all the three pairs (p<0.05 to p<0.001, One-way ANOVA, Bonferroni multiple comparisons tests). B A 7 14 21 28 35 0 20 40 60 80 100 Climbing (% flies) 0 10 20 30 40 50 60 70 80 90 Half time (s) 7 14 2821 Age (days) Age (days) 35 98 Results This test consists on counting the number of flies able to climb 5 cm height in 18 s (Fig. 27A), and was performed in males and females separately (results not shown). The data are represented in mean ± SEM of groups (7, 14, 21, 28 and 35 days) of 20 flies, being 10 flies used for each individual experiment. We found a slight non-significant decrease between the young flies of 7 and 14 days-old, where 89 and 84% of the flies, respectively, performed the task; but from that age onwards, flies progressively lost their ability to climb, till the oldest age group measured, 35 days where only 3% of the flies were able to climb within the limits set in the assay. The data of the last four points fits a straight line (linear correlation coefficient, r=0.997, p<0.005). In a different climbing test, we also monitored the time in which 50% of the flies performing the test 16 reached the 5 cm height mark (Fig. 27B), and observed a gradual increment over age, from (mean ± SEM) 5 ± 0.4 s for young flies (7-day old), up to 68 ± 12 s for old flies (28-day old). The differences between the means of consecutive age groups were statistically significant for all the three pairs (p<0.05 to p<0.001, One-way ANOVA, Bonferroni multiple comparisons tests). 1.5. Sarcoplasmic reticulum Ca 2+ measurements in the thorax muscles of aging flies One of the most convincing theories of skeletal muscle aging, postulates there is a disruption in Ca 2+ uptake and release with age that leads to an impairment of the excitation-contraction (EC) coupling. Whereas the cytosolic Ca 2+ has been widely studied, little is known about the role of Ca 2+ dynamics within the SR in muscle excitation. In order to study the effects of aging on the [Ca 2+ ] SR, DM006 transgenic flies were maintained at 29 °C and divided into five age groups (7, 14, 21, 28 and 35 days). This allows to measure [Ca 2+ ] SR along the whole fly life, according to the survival curve of the line, previously shown in Fig. 26B. Each age group contained 20 flies, 10 of each sex. For the experiment, flies were immobilized in agar and fluorescent images 17 were acquired at the two excitation wavelengths of erGAP3 16 Usually 10 flies performed an individual experiment and, thus, we counted the time in which 5 of them surpass the height mark. 17 Five fluorescence images at each GAP3 wavelength (405 and 470 nm) were collected at 5 s intervals. 99 Results sensor in the thorax muscles. The erGAP3 fluorescence ratio (R=F 470 /F 405 ) was, then, calculated by dividing, pixel-by-pixel, the individual wavelengths acquisitions, so that erGAP3 ratio is proportional to [Ca 2+ ] SR/ER (Alonso et al., 2017b; NavasNavarro et al., 2016). Therefore, this ratiometric measurement in the thoracic muscle mass of the DM006 fly should give a quantification of the [Ca 2+ ] SR . The erGAP3 fluorescence ratio was measured for all five age groups. As this value can vary if the experimental settings change, the same settings were maintained for all the measurements (e.g., gain or integration period). Figure 28. Sarcoplasmic reticulum Ca 2+ measurements in vivo in the erGAP3 transgenic muscle fly line. DM006 flies (erGAP3 transgenic muscle line) from different ages (7, 14, 21, 28 and 35 days) were imaged at the two individual wavelengths of erGAP3 sensor (405 and 470 nm) in the thorax muscles. erGAP3 ratios (R= F 470 /F 405 ) were calculated pixel-by-pixel from the individual wavelengths. Recordings for each fly were performed before (black circles; baseline values expressed as R) and, after 10 min-heating at 50 °C (red triangles; empty store, R min ). Each value is the mean ± DS of 20 flies, one half females and one half males, except for flies aged 35 days groups, which contained only 10 female flies. The results show that erGAP3 ratio (R) decreased progressively and significantly over age (p<0.001, One-way ANOVA; Bonferroni test). Note that R min values were constant during the whole fly life. 7 14 21 28 35 1.0 1.5 2.0 2.5 3.0 Age (days) R a t i o F 470 / F 405 R Rmin 100 Results The results are shown in Fig. 28, which represents the erGAP3 fluorescence ratios in the DM006 transgenic line over age. In the thorax muscles of the DM006 flies in vivo, erGAP3 ratio (Fig. 28; black circles) decreased progressively over age from 2.68 ± 0.28 (mean ± SD; n=20) in the young flies (7-day old) down to 1.48 ± 0.10 (mean ± SD; n=10) in the old flies (35-day old). There is a two-fold difference in the erGAP3 ratio between young and old flies, considered statistically significant (p<0.001, One-way ANOVA; Bonferroni multiple comparisons test). 2. Calibration of erGAP3 fluorescent signal into [Ca 2+ ] ER or [Ca 2+ ] SR The data shown in Fig. 28 demonstrates an age-dependent decrease in the SR Ca 2+ content, although the calibration of the fluorescent signal into [Ca 2+ ] was not possible. Knowing that the dynamic range (DR) of GAP3 (R max /R min ) is about 3, and the K D is 489 µM, it is possible to calibrate the fluorescent signal into [Ca 2+ ] SR/ER with the following equation:    =  + ( − )( [  ]    [  ] ) (Equation 5) Where “R” is a given GAP3 fluorescence ratio (R= F 470 /F 405 ), “R min ” is the minimal fluorescence ratio obtained when the ER store is completely empty, “DR” is the dynamic range of GAP3, “K D ” is the Ca 2+ -GAP3 dissociation constant (489 µM) and “[Ca 2+ ]” is the Ca 2+ concentration. The resulting values of R/R min should range between 1 at [Ca 2+ ]= 0 and 3 at [Ca 2+ ]= infinite. The calibration curve for DR = 3 and K D = 489 µM allows the transformation of the fluorescent signal into Ca 2+ concentration (Fig. 29); however a method to obtain the R min value is necessary for [Ca 2+ ] computation. In the in vitro experiments, perfusion allows the addition of agonists and inhibitors that act to deplete the ER. However, this manipulations are not feasible during in vivo measurements due to the inaccessibility of the Ca 2+ indicator. 101 Results [Ca 2+ ] (M) Figure 29. Ca 2+ calibration curve of GAP3. GAP3 protein isolated from E.coli was titrated for Ca 2+ . The curve was drawn using the following parameters: dynamic range (DR) = 3; Ca 2+ -GAP3 dissociation constant (K D ) = 489 µM; Hill number (n) = 1. On the left axis, R/R min that ranges between 1.0 at [Ca 2+ ] = 0 and 3.0 at [Ca 2+ ] = infinite. On the abscissa, [Ca 2+ ] in log scale. The red line corresponds to the following equation: R/R min = 1 + (DR - 1)*( [Ca 2+ ] / (K D + [Ca 2+ ]) ) (Equation 5). 2.1. Calibration procedure for the in vitro experiments Calibration of the fluorescent signal into Ca 2+ concentration during in vitro experiments in HeLa cells stably-expressing erGAP3 is performed by completely emptying the ER at the end of each experiment. In this way, the minimal value of fluorescence (F min ) is obtained and computation of [Ca 2+ ] from the ratio (R=F 470 /F 405 ) is possible. A representative experiment of ER Ca 2+ measurements in Hela cells is shown in Figure 30. Panel A shows the two individual excitation fluorescent wavelengths of the erGAP3 sensor (F 470 and F 405 ). They display, the typical reciprocal responses of a ratiometric Ca 2+ -sensitive sensor; upon Ca 2+ dissociation, one wavelength (470 nm, in green) decreases in parallel with an increase of the other wavelength (405 nm, in blue). Panel B shows the ratio (R) of the two individual fluorescences. 1.0 1.5 2.0 2.5 3.0 1 10 100 1000 10000 102 Results Figure 30. Calibration of erGAP3 fluorescence signal into endoplasmic reticulum Ca 2+ concentration in vitro. A) Endoplasmic reticulum (ER) Ca 2+ measurements in HeLa cells stably-expressing erGAP3 sensor. Traces show partial (first challenge) and complete (second challenge) ER Ca 2+ emptying by perfusion with inositol 1,4,5-triphosphate (IP 3 )- producing agonists (ATP+CCh (carbachol); 100 µM each) or the ER “depletion cocktail” (ATP+CCh in Ca0+TBH) that contains the same stimulus in Ca 2+ -free solution and with the sarco/endoplasmic reticulum Ca 2+ -ATPase (SERCA) reversible inhibitor 2,5-di(tertbutyl)-1,4-benzohydroquinone (TBH; 10 µM), respectively. Both stimulus are highlighted in grey shading. The traces show one representative experiment, average of 100 cells from the same microscopic field. The individual GAP fluorescence, F 405 in blue (scale at the left) and F 470 in green (scale at the right), in the control cells is shown. B) The same cells were heated at 50 °C for 10 min and immediately subjected to the same protocol. The panel shows erGAP3 ratio (R= F 470 /F 405 ) with the traces before (Control, black trace) and after heating (Heated, red trace). Note that erGAP3 ratio after heating remains stable during the whole experiment, with no significant changes upon neither of the challenges. 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2 min ATP+CCh Heated Control ATP+CCh in Ca0+TBH 0.8 B 180 200 220 220 260 300 340 380 420 F405 F470 ATP+CCh ATP+CCh in Ca0+TBH A 103 Results The protocol is as illustrated in Figure 30. In order to check that cells were physiologically responsive, they were stimulated during a pulse of 30 s with a mixture of ATP+CCh, at maximal stimulation concentrations, and this stimulus produces a large but partial depletion of the ER (Fig. 30B, “control” black trace). Next, the ER stores were refilled by perfusion with extracellular medium containing 1 mM Ca 2+ for 5 min and, then, the cells were challenged again with an ER “depletion cocktail” containing the same stimulus, in Ca 2+ -free solution and with the SERCA reversible inhibitor TBH (10 µM). This challenge produces a complete ER discharge and, thus, the F min value can be obtained at the end of the experiment (Fig. 30B, “control” black trace). After washing the ER “depletion cocktail” by perfusion with extracellular medium containing 1 mM Ca 2+ , the ER trace recovered its basal value, a sign of ER Ca 2+ refilling. Finally, the same cells were heated at 50 °C for 10 min in 1 mM Ca 2+ extracellular medium, and immediately subjected to the same protocol (Fig. 30B, “heated”, red trace). Note that erGAP3 ratio remains stable during the whole experiment, with minimal changes during the challenge with ATP+CCh or the ER “depletion cocktail” (ATP+CCh in Ca 2+ free solution with TBH). The application of agonists or inhibitors is not possible during the in vivo experiments performed with flies, given the impermeable nature of the chitin compound of the fly cuticle that prevents medium exchange. However, the fluorescence value obtained by heating at 50 °C for 10 min HeLa cells stablyexpressing erGAP3, was the same as that produced by the complete ER “depletion cocktail” as shown in Figure 30B (compare traces, “control” and “heated”). Results are highly reproducible, as shown in Figure 31A, where mean values ± SEM of 6 different experiments are represented. The ratios (R) were normalized to the initial fluorescence values (R 0 ) in order to homogenize the baseline among different cell batches. The R/R 0 value obtained by heating is not significantly different from the one obtained by depletion with Ca 2+ -free solution containing 10 µM TBH and IP 3 - producing agonists (Fig. 31A). The conclusion is that the heating treatment could be used for calibration, even during in vivo experiments with Drosophila. Importantly, the morphology of the cells (observed as green GFP expression) was not modified by the heating treatment, and there were no signs of GAP3 aggregation (compare panels, “control” and “heated”, in Fig. 31B). Moreover, the 104 Results changes in fluorescence were similar in the different cells and also among different ER regions within the same cell, as demonstrated by the homogeneous intensity of all the pixels in the ratio image (right panel in Fig. 31B). Figure 31. Calibration of erGAP3 fluorescence signal into endoplasmic reticulum Ca 2+ concentration in vitro. The Ca 2+ levels in control and heated cells. A) The figure shows the mean ± SEM values of 6 independent experiments as the one presented in Figure 30. erGAP3 ratios (R= F 470 /F 405 ) were normalized to the initial fluorescence values (R 0 ) in order to homogenize the baseline among different cell batches. The histogram shows the endoplasmic reticulum (ER) Ca 2+ levels, expressed as R/R 0 , at: baseline (Basal); after stimulation with the agonists alone (+ATP); after stimulation with the ER “depletion cocktail” (F min ); or after the heating treatment (Heated); according to the protocol illustrated in Figure 30. All the values were significantly smaller than the baseline value (p<0.001; One-way ANOVA, Bonferroni test), whereas the “F min ” and the “Heated” bars did not differ significantly. B) erGAP3 fluorescence images recorded at 470 nm excitation before (Control) and after heating (Heated), both in the baseline condition. The right image shows the pixel-by-pixel ratio of both images (F/F min ) coded in pseudocolour (scale on the right) where warm colours (such as red or yellow) represent high ER Ca 2+ concentration ([Ca 2+ ] ER ) and the cold colours (blue or cyan) represent low high [Ca 2+ ] ER . Calibration bar, 10 µm. A B Control (Basal) Heated 2 0 4 F/F min Basal 0.0 0.2 R/R 0 0.4 0.6 0.8 1.0 Heated F min +ATP 105 Results The same set of experiments was also performed in other cell types, such as transiently-expressing erGAP3 HEK293T cells or primary hippocampal cortical astrocytes obtained from erGAP3 transgenic mice. In both cases the results obtained were similar to those obtained in HeLa cells. A summary of these results of 4-6 individual experiments, with 100-150 individual cells per experiment, are shown in Table 7, where the values of F min obtained by perfusion with the ER “depletion cocktail” and by heating the cells were very similar. The conclusion is that the heating treatment can be used as a general method for calibrating fluorescent signals into [Ca 2+ ] SR/ER in different cell types. R/R 0 (mean ± SEM) HeLa HEK 293T Astrocytes Basal 1.00 1.00 1.00 ATP 0.62 ± 0.02 0.91 ± 0.02 0.79 ± 0.01 F min 0.49 ± 0.05 0.62 ± 0.01 0.65 ±0.01 Heated 0.43 ± 0.04 0.45 ± 0.01 0.54 ± 0.01 Table 7. Calibration of erGAP3 fluorescent signal into [Ca 2+ ] SR/ER in different cell types. Table comparing the ER Ca 2+ levels, expressed as R/R 0 , at baseline (Basal), after stimulation with IP 3 -producing agonists alone (ATP), together with TBH (F min ) and after the heating treatment (“heated”) in three different cell types: stably-expressing erGAP3 HeLa cells, transiently-expressing erGAP3 HEK293T and cortical astrocytes obtained from erGAP3 transgenic mice. The mean ± SEM values of 4-6 independent experiments, with 100-150 individual cells per experiment, are shown. Note that, for each cell type, “F min ” and “Heated” values were very similar. As we have generated the DM006 transgenic fly line expressing erGAP3 in muscle cells, and the calibration heating protocol was intended to be used in the DM006 flies in vivo; we next investigated whether erGAP3 was appropriate for measuring [Ca 2+ ] SR in this muscle cell type in mammals. For this purpose, mouse myoblast cells (C2C12) were transiently transfected with the erGAP3 gene cloned in the mammalian expression vector pcDNA3. The protocol is represented in Fig. 32 as the ratio of the two individual erGAP3 fluorescence wavelengths (R=F 470 /F 405 ). First, the ER stores are kept filled by perfusion with extracellular medium containing 1 mM Ca 2+ to obtain the resting [Ca 2+ ] ER levels. Then, the dynamic range of erGAP3 in C2C12 myoblasts was tested by complete depletion of the ER achieved by perfusion with the “depletion cocktail” composed by ATP (100 µM) and the SERCA 112 Results 2.4. GAP3 remains functional after heating In order to use the heating method for calibration of [Ca 2+ ] SR/ER , functionality of GAP3 after heating must be demonstrated. Therefore, we checked whether GAP3 sensor remained Ca 2+ -sensitive and maintained the structure after heating. For this purpose, we isolated the recombinant GAP3 protein produced in bacteria. Firstly, we checked if GAP3 remained Ca 2+ -sensitive with the heating treatment: the protein was heated at 50 °C and a fluorescence excitation spectrum, in a Ca 2+ -free or in a Ca 2+ -containing medium, was obtained in the fluorescence spectrophotometer. The excitation spectrum of the “heated” GAP3 protein continued to be Ca 2+ -sensitive and, thus, changed upon Ca 2+ binding to the sensor, so that the fluorescence excited at 470 nm increases with a parallel decrease of the fluorescence excited at 405 nm (Fig. 36). Figure 36. GAP3 remains functional after the heating treatment. Excitation spectra of 50 °C heated GAP3 protein produced in bacteria. The spectrophotometer cuvette was filled with extracellular-like medium containing 8.5 µg of the heated protein and excitation spectra were performed (between 380 and 500 nm with the emission set at 520 nm) in the absence (0.1 mM EGTA; blue line) or in the presence (1 mM CaCl 2 ; red line) of Ca 2+ . Characteristic Ca 2+ -dependent peaks at 405 and 470 nm (black dashed line) are preserved and, thus, the protein is functional after heating. Fluorescence (AU) C a 2+ EGTA 100 80 60 40 20 0 Excitation wavelenght (nm) 380 400 420 440 460 480 500 113 Results Secondly, to characterize protein stability, we performed a denaturation assay that quantifies the thermal stability of a protein under different denaturation conditions such as basic pH of 8.8, the reducing agent DTT, or the Ca 2+ chelating agent EGTA. The outcome is the transition unfolding temperature (T m ) where the concentrations of folded and unfolded protein are equal. According to the GAP3 thermal denaturation assay, the concentration of folded protein at 50 °C temperature is almost 1 and the T m is of 79 °C (Fig. 37), almost 20 °C above the temperature used for the heating treatment. The conclusion is that, after 50 °C heating, GAP3 protein remains functional and, also, the structure is maintained. Figure 37. Thermal stability of GAP3. Denaturation assay of GAP3 protein under different denaturation conditions such as basic pH (pH= 8.8; pink trace), the reducing agent DTT (Dithiothreitol; orange trace), or the Ca 2+ chelating agent EGTA (Ethylene Glycol Tetraacetic Acid; green trace). GAP3 remained folded at 50 °C and the transition unfolding temperature (T m ; black dashed line) is of 79 °C. Performed at the EMBL Sample Preparation and Characterization Facility, Hamburg, Germany. Temperature ( o C) 10 20 30 40 50 60 70 80 90 0.0 0.5 1.0 Unfolded Fraction Tm = 79 o C pH 8.8 +DTT +EGTA 114 Results 2.5. Calibration procedure for other low affinity Ca 2+ indicators We next asked whether the erGAP3 calibration protocol was also valid for other Ca 2+ indicators. We chose one of the CEPIA family members with low Ca 2+ affinity, the G-CEPIA1er. This a green Ca 2+ indicator, with a K D of 672 µM and a DR of 4.7, also adequate to measure [Ca 2+ ] in the lumen of the ER. Figure 38. Calibration of the fluorescence signal into endoplasmic reticulum Ca 2+ concentration in vitro for other low affinity Ca 2+ indicators. A) Endoplasmic reticulum (ER) Ca 2+ measurements in G-CEPIA1er transiently transfected HeLa cells recorded at 470 nm. The fluorescence values were normalized by the initial fluorescence (F 0 ) and expressed as F 470 /F 0 . Traces show one representative experiment, average of 100 cells from the same microscopic field, before (Control, black trace) and after heating (Heated, red trace). The protocol was identical to the one described for erGAP3 (Figure 30). Traces show partial (first challenge, ATP+CCh) and complete ER Ca 2+ emptying (second challenge, ATP+CCh in Ca0+TBH). Both stimulus are highlighted in grey shading. B) Histogram representing the normalized values by the initial fluorescence (F/F 0 ), expressed: at baseline (Basal); after stimulation with IP 3 -producing agonists alone (+ATP, first challenge); the agonists together with the SERCA inhibitor TBH (F min, second challenge); or after heating the cells at 50 °C for 10 min (Heated). The mean ± SEM values of 4 independent experiments are shown. B A 0 0.4 0.8 1.2 F 470 /F 0 1 min Heated Contro l ATP+CCh ATP+CCh in Ca0+TBH 0 , 0 0,4 0,8 1,2 F/F 0 Basal Heated F min +ATP 115 Results To search whether the calibration method could be used for CEPIA, we tested the protocol described above for erGAP3 in G-CEPIA1er transiently transfected HeLa cells and imaged the cells 24 h after transfection (Fig. 38A). G-CEPIA1er is an intensiometric Ca 2+ indicator and, therefore, we measured fluorescence at 470 nm and normalized the data by dividing by the initial fluorescence (F 0 ). We compared the F min value (mean ± SEM of 4 individual experiments) obtained by perfusion with the ER “depletion cocktail” (0.25 ± 0.03) with the F min value obtained after heating (0.28 ± 0.07), and conclude that they did not differ significantly (Fig. 38B). These results indicate that the heating treatment may be a general method that can be applied to different low affinity Ca 2+ sensors. 3. Calibrated Ca 2+ measurements in vivo 3.1. Sarcoplasmic reticulum Ca 2+ concentration decreases progressively with age Given that the calibration method had proved to be valuable for in vitro and in vivo experiments, we next applied the calibration procedure to the erGAP3 measurements previously obtained in the SR of the DM006 transgenic line in vivo (Fig. 28). Thus, the obtained erGAP3 ratio values (Fig. 28, black circles) for each fly were normalized by dividing by its corresponding R min value (Fig. 28, red triangles), measured in the same fly after heating at the end of the experiment (R/R min ). This treatment of the data homogenized the outcomes from different experiments. For example, the coefficients of variation estimated for five sets of flies of different ages were 11.70 ± 3.17 (%, mean ± SD; n=5) for the erGAP3 ratio (R) values and 8.06 ± 1.54 (%, mean ± SD; n=5) for R/R min ; the difference was significant (p<0.05; paired t-test). Note that complete Ca 2+ emptying of the SR would be reflected by a decrease of R/R min to 1, whereas saturation with Ca 2+ inside the SR would produce, according to the estimated dynamic range, a R/R min value of 3. The normalized data are shown in Fig. 39. Each value is the mean ± SEM of 10-20 flies, half females and half males, except for the group of flies aged 35 days, which contained only 10 female flies. The SR Ca 2+ content (R/R min ) in the skeletal muscle decreased monotonically with age from a value of 2.22 ± 0.02 (mean ± 116 Results SEM; n=20) in the young (7-day old) flies to 1.03 ± 0.03 (mean ± SEM; n=10) in the old flies (35-day old), which makes a two-fold difference (Fig. 39, left axis). All the values of the following groups were significantly smaller than the value at 7 days (p<0.001; One-way ANOVA; Bonferroni multiple comparisons test). Finally, the calibration procedure can be applied to calculate the real [Ca 2+ ] SR for each age group (Fig. 39, right axis). The approximate value of [Ca 2+ ] SR was near 600 µM in young flies (7-day old) and dropped dramatically, to one tenth of this concentration, 50 µM, in the old flies (35-day old). As a conclusion of these results, there was a gradual and significant decrease of R/R min that reflects the progressive emptying of the SR Ca 2+ store with age. Age (days) Figure 39. Decrease of the resting sarcoplasmic reticulum Ca 2+ concentration in aging flies. The previously obtained erGAP3 ratios (R= F 470 /F 405 ) in the DM006 flies (erGAP3 transgenic muscle line) were normalized by dividing by the erGAP3 ratio obtained in the same fly after heating (R min ), that correspond to a completely empty sarcoplasmic reticulum (SR) store. The normalized data (R/R min ; left axis) represent a quantification of the SR Ca 2+ content that range between 1 at [Ca 2+ ]= 0 and 3 at [Ca 2+ ]= infinite. Then, the calibration into sarcoplasmic reticulum Ca 2+ concentration ([Ca 2+ ] SR ) was calculated with the normalized values (R/R min ), the Ca 2+ -GAP3 dissociation constant (K D = 489 µM) and the dynamic range (DR= 3). The approximate [Ca 2+ ] SR is given on the right axis. All the values were significantly smaller than the value at 7 days (p<0.001; One-way ANOVA; Bonferroni test). The values correspond to the females flies (pink triangles, n= 10), the male flies (blue inverted triangles, n= 10) or the mean of both genders (solid black circles, males and females n= 20). 117 Results 3.2. The muscle function correlates with [Ca 2+ ] SR We next studied the relationship between the decrease in [Ca 2+ ] SR and the loss of muscle function with age in the DM006 line. The results are shown in Fig. 40. Figure 40. Correlation between the resting sarcoplasmic reticulum Ca 2+ concentration and fly climbing ability with age. The sarcoplasmic reticulum (SR) Ca 2+ content is measured as R/R min on the lower axis and ranges between 1 at [Ca 2+ ]= 0 and 3 at [Ca 2+ ]= infinite; calibration of erGAP3 signal into sarcoplasmic reticulum Ca 2+ concentration ([Ca 2+ ] SR ) is represented on the upper abscissa axis and ranges between 50 µM and 600 µM; and, the climbing ability is represented as percentage of the number of flies that surpassed 5 cm height in 18 s over the total of flies in each experiment on the left ordinate axis. There was close correlation between the motor function and the [Ca 2+ ] SR . Values represent mean ± SEM of groups of 10-20 flies. According to our calibration, the [Ca 2+ ] SR was near 600 µM in young flies (7-day old) and 89% of these flies were able to perform the climbing assay. Then, we observed that there is a threshold around 400 µM that corresponds to the fly age of 14 days, above this value the motor function was maintained, but below this 0 20 40 60 80 100 [Ca 2+ ] SR ( M) 1.0 2.22.01.81.61.41.2 20 50 100 200 300 400 600 800 Climbing (% flies) SR Ca 2+ content (R/R m i n ) 118 Results value both [Ca 2+ ] SR and climbing ability collapsed with very similar time courses. In the middle age group (21 day-old) 50% of the flies can perform the climbing test, and the [Ca 2+ ] SR was about 200 µM and, thus, has decreased three times compared to the youngest flies (7-day old). The oldest flies (35-day old) showed a [Ca 2+ ] SR of 50 µM, about one tenth of the [Ca 2+ ] SR value found in the youngest flies. This dramatic decrease of [Ca 2+ ] SR was associated to an almost absolute inability to climb, as only 3% of the flies were able to perform the climbing test. As a conclusion, we found an excellent correlation between SR Ca 2+ concentration and muscle function, measured in the climbing assay. 3.3. Cytosolic Ca 2+ dynamics is altered with age We next checked whether the variations in resting [Ca 2+ ] SR observed with age were associated to the dynamics of cytosolic Ca 2+ . For this purpose, we studied the [Ca 2+ ] C responses to muscle stimulation through motoneurons in DM010 transgenic flies that express the cytosolic Ca 2+ sensor GCaMP3. This indicator does not have any basal fluorescence in the muscle and is not ratiometric, so quantitative measurements of resting cytosolic Ca 2+ were not feasible. Instead we measured the cytosolic Ca 2+ response in the dorsal longitudinal (flight) muscles located in the fly thorax by stimulation of the giant fiber system (GFS). Flies were immobilized in agarose and the giant fibers were stimulated through two tungsten electrodes impaled into the fly eyes. After checking the correct positioning of the electrodes by observing twitches in the wing after stimulation, a protocol of repetitive stimulation with 5 V/50 ms stimuli at a frequency of 0.5 or 10 Hz was performed during 5 s. Fluorescent measurements were recorded as green GFP fluorescence, acquired every 100 ms, and normalized to the values obtained during the first 5-10 frames (F/F 0 ). The protocol is illustrated in Fig. 41A, where the mean ± SEM of 6 -11 individual recordings of young (7-day old) and old (21-day old) flies is shown. In both age groups, we can observe that at 0.5 Hz, there are small repetitive increases in [Ca 2+ ] C . However, at 10 Hz stimulation, there is one individual event fused and peaked resulting in a much larger cytosolic Ca 2+ increase. It seems clear that, at both frequencies, the increase of [Ca 2+ ] C is larger 119 Results in the younger (7-day old) flies. This output is consistent with the higher resting [Ca 2+ ] SR found in the younger flies. Figure 41. Effect of age on cytosolic Ca 2+ dynamics upon muscle stimulation in Drosophila. A) Recordings of the cytosolic Ca 2+ response in the flight muscles of young (7 days) and old (21 days) flies from the DM010 line, expressing the GCaMP3 cytosolic Ca 2+ indicator in the thorax muscles. Flies were immobilized in agarose and impaled through two tungsten electrodes in the eyes. Recordings were obtained by stimulation of the giant fibers which synapse the motoneurons that drive the flight muscles. The stimulation protocol consisted of repetitive stimuli of 5 V/50 ms and 5 s duration at low (0.5 Hz) and high (10 Hz) frequencies. Fluorescent measurements were acquired at 470 nm excitation every 100 ms, normalized to the values obtained during the first 5-10 frames (F 0 ) and represented as F 470 /F 0. The black trace represents the mean ± SEM of 6 young flies and the red trace the mean ± SEM of 11 old flies. B) Histogram comparing the area under the curve of the cytosolic Ca 2+ transients shown in (A) at low (0.5 Hz) and high (10 Hz) frequencies and in young (7 days old) and old flies (21 days old). The reduction of the cytosolic Ca 2+ transients in the old flies was statistically significant (red *; p<0.05; unpaired t-test). A 120 Results The Ca 2+ transients were quantified by measuring the area under the curve, and the results expressed as the mean ± SEM of 6-11 flies are shown in Fig. 41B. Comparing age groups, at low frequencies, the area showed a reduction of 50% in the old flies. At high frequencies the reduction was of 65%. Both of these values were statistically significant (p<0.05; unpaired t-test). This outcome validated the results obtained with erGAP3 in the DM006 muscle line and confirmed that the old flies had a lower SR Ca 2+ levels and, thus, release less Ca 2+ to the cytosol, in comparison to young flies. 3.4. Endoplasmic reticulum Ca 2+ concentration does not change with age in brain neurons In order to investigate whether the reduction of SR Ca 2+ content in the aged flies was a general phenomenon occurring in all tissues, we studied the changes of [Ca 2+ ] ER with age in the brain neurons, using the DN005 transgenic fly line. We measured [Ca 2+ ] ER in the whole brain (Fig. 24C and D). The experimental procedure was the same as in the muscle. First, flies were divided by age and fluorescent images were acquired in the brain before and after heating, to calculate R and R min , respectively. Then, values were normalized to obtain ER Ca 2+ content (R/R min ) and finally [Ca 2+ ] ER was calculated for each age group. The ER Ca 2+ content in the brain (Fig. 42, left scale) had a value of 1.75 ± 0.05 (mean ± SEM; n=20) in the young flies (7-day old) and did not change over age, yielding a value of 1.70 ± 0.05 (mean ± SEM; n=12) in the old flies (35-day old). Finally, we calibrated the measurements into [Ca 2+ ] ER (Fig. 42, right ordinate scale) and, unlike the aged muscle, the [Ca 2+ ] ER of the brain neurons remained stable at about 300 µM along the whole fly life. Results show that the variation among the means of the different age groups (12-22 flies) was not significantly greater than those expected by chance (p=0.995; One-way ANOVA; Bonferroni multiple comparisons test). 121 Results Figure 42. The resting endoplasmic reticulum Ca 2+ concentration does not change with age in brain neurons. erGAP3 fluorescence (F 405 and F 470 nm) was measured in the whole brain of DN005 flies (erGAP3 transgenic neuronal line) from different ages. erGAP3 ratios were calculated before (R= F 470 /F 405 ) and after (R min ) heating for each fly. Normalized data (R/R min ), which are proportional to endoplasmic reticulum (ER) Ca 2+ content, are shown on the left axis. An approximate calibration of erGAP3 fluorescence into endoplasmic reticulum ER Ca 2+ concentration ([Ca 2+ ] ER ) is given on the right axis. Each value is the mean ± SEM of 12-22 flies, one half females and one half males, except for flies aged 35 days groups, which contained only 12 female flies. Results show that the variation among the means of the different age groups is not significantly greater than expected by chance (One-way ANOVA; Bonferroni Multiple Comparisons Test). 3.5. Endoplasmic reticulum Ca 2+ concentration decreases with age in wing sensory neurons In the DN005 fly line, erGAP3 expression is controlled by the pan-neuronal promoter elav and, thus, the indicator is expressed in all neurons (Fig. 24C and D). As we have observed that [Ca 2+ ] ER does not change with age in brain neurons, we decided to test whether the behaviour was similar in other neuronal types such as the peripheral sensory wing neurons. We measured [Ca 2+ ] ER in the wing margin of the DN005 fly line following the same procedure used with the whole brain. 7 14 21 28 35 1.0 1.2 1.4 1.6 1.8 2.0 2.2 Age (days) [Ca 2+ ] ER ( M) 20 50 100 200 300 400 600 ER Ca 2+ content (R/R min )