Steroid Hormone Effects on the Mitochondrial Morphology and Physiology of Primary Astrocytes Implications for Neuroprotection

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Publikationsserver der RWTH Aachen University

Steroid hormone effects on the mitochondrial morphology and physiology of primary astrocytes

Implications for neuroprotection

Von der Fakultat¨ fur¨ Mathematik, Informatik und Naturwissenschaften der RWTH Aachen University zur Erlangung des akademischen Grades einer Doktorin der Naturwissenschaften genehmigte Dissertation

vorgelegt von

Doctorandus Licenciada em Biologia Aplicada

GILDA CATARINA DE CASTRO LOPO WRIGHTDE ARAUJO

aus Braga, Portugal

Berichter: Universitatsprofessor¨ Dr. Cordian Beyer Universitatsprofessor¨ Dr. Hermann Wagner

Tag der mundlichen¨ Prufung:¨ 15. Marz¨ 2012

Diese Dissertation ist auf den Internetseiten der Hochchulbibliothek online verfugbar.¨

To my parents ii Contents

1 General introduction 1 1.1 Neuroprotection by steroid hormones in aging and disease ...... 3 1.2 Steroid hormones in brain development ...... 4 1.3 Astrocyte mediation of steroid hormone neuroprotection ...... 6 1.4 Mitochondria ...... 7 1.4.1 Mitochondria in aging and disease ...... 7 1.4.2 Interaction of steroid hormones with mitochondria ...... 9 1.4.3 Mitochondrial structure ...... 10 1.4.4 Cellular respiration ...... 11 1.4.5 Mitochondrial dynamics - fusion and ﬁssion ...... 12 1.4.6 Mitochondrial apoptosis regulation ...... 16 1.5 Aim of this work ...... 17

2 Materials and methods 19 2.1 Reagents ...... 21 2.2 Animals ...... 21 2.3 Cell culture ...... 21 2.4 Cell treatments ...... 22 2.5 Cell viability ...... 22 2.6 Reverse Transcription ...... 23 2.7 Semi-quantitative reverse transcriptase - polymerase chain reaction (sqRT- PCR) analysis ...... 24 2.8 Quantitative real time-PCR (qRT-PCR) analysis ...... 28 2.9 Measurement of mitochondrial and nuclear DNA ...... 29 2.10 Isolation of mitochondria from primary astrocytes ...... 29 2.11 Polarographic measurements ...... 30

iii iv CONTENTS

2.12 Protein content determination ...... 30 2.13 Statistical analysis ...... 31

3 Steroid hormone effects on mitochondrial respiration 33 3.1 Introduction ...... 35 3.2 Materials and Methods ...... 37 3.3 Results ...... 37 3.3.1 Regulation of the expression of catalytic respiratory chain subunits by estrogen ...... 37 3.3.2 Effect of Estrogen on the ratio of mitochondrial versus nuclear DNA levels ...... 40 3.3.3 Long-term estrogen effects on mitochondrial activity ...... 40 3.3.4 Short-term estrogen effects on mitochondrial respiratory chain activity 42 3.4 Discussion ...... 46

4 Steroid hormone effects on mitochondrial fusion and fission 53 4.1 Introduction ...... 55 4.2 Materials and Methods ...... 57 4.3 Results ...... 57 4.3.1 Gender-specific effects of steroid hormones on cell viability and apoptosis ...... 57 4.3.2 Gender-specific effects of steroid hormones on fusion and fission gene transcription ...... 60 4.4 Discussion ...... 63

5 Mitochondrial fusion and ﬁssion processes in the mouse cortex development 69 5.1 Introduction ...... 71 5.2 Materials and Methods ...... 74 5.3 Results ...... 75 5.3.1 Ontogenic regulation of mitochondrial fusion and ﬁssion gene transcription ...... 75 5.4 Discussion ...... 76

6 General discussion 81

7 Summary / Zusammenfassung 89 CONTENTS v

Abbreviations 95

References 99

Contributions 113

Acknowledgements 117

Curriculum vitae 121 vi CONTENTS Chapter 1

General introduction

1.1. Neuroprotection by steroid hormones in aging and disease 3

With the considerable increase in life expectancy and a consequently older population, there are increased demands in successful aging strategies that lower the probability of disease and allow an active life with high cognitive and physical capacity. Dementing diseases will soon become main causes of disability, without proper neuroprotective treatments (Schumacher et al., 2003). The fact that levels of some steroid hormones markedly decrease with age and the evident sex-speciﬁc differences in incidence and course of many neurodegenerative disorders (Schulman, 2007) proposed the involvement of sex hormones in neurodegeneration. The risk of Alzheimer’s and Parkinson’s Diseases in post-menopausal women was reported to have a direct correlation with the estrogen decline (Schumacher et al., 2003). Several studies have shown that estrogen replacement therapy has successfully reduced the incidence of these diseases in post-menopausal women as well as improved cognitive and memory tests (Schulman, 2007; Peri and Serio, 2008).

1.1 Neuroprotection by steroid hormones in aging and dis-

ease

Steroid hormones have become favorites for neuroprotection treatments on the one side because of their vital importance on the functioning of the central nervous system (Behl, 2003), the regulation of neurotransmitter systems, the promotion of neuron viability and myelinization and a positive influence in cognitive processes of learning and memory. On the other side, unconjugated steroids easily cross the blood-brain-barrier and accumulate throughout the brain. Additionally, neurosteroids are produced locally in the brain (Schu- macher et al., 2003; Prange-Kiel et al., 2006; Von Schassen et al., 2006), and the stimulation of this process is another therapeutic possibility. The increasing incidence of neurodegenerative diseases demands for further research in the field of neuroprotection and the clarification of the cell global process. The cellular mechanisms of the steroid-mediated 4 Chapter 1. General introduction neuroprotection have been shown to involve direct interaction with neural cells and indirect effects through other cells, from which astrocytes are the most probable intermediaries (see chapter 1.3).

1.2 Steroid hormones in brain development

Steroid hormones are known to be important regulatory signals in biological functions and for brain development (Hutchison et al., 1997; Beyer, 1999; Mendez et al., 2005; Daniel, 2006; McCarthy, 2008). The developing and adult brain cells express receptors for progesterone (PR) and estrogen (ER) (Zhang et al., 2002; Mitra et al., 2003; Quadros et al., 2007), being therefore unmistakable targets of these hormones, even though during development, male have a stable PR presence, while female has little. The PR is transcribed by activated ER, showing that the effects caused by these hormones are intimately connected. Estrogen plays an essential role in late embryonic and early postnatal brain development, controlling cell growth, neurons and astrocytes morphometry and tissue differentiation, regulating cell apoptotic degeneration, synaptogenesis, organization of brain circuits and also the sex-specific differentiation of the CNS (Mesano and Jaffe, 1997; Beyer, 1999; Woo et al., 2001; McCarthy, 2008). Interestingly, some of the most obvious sex differences are in neurological and psychiatric disorders, women being more likely to suffer from depression, multiple sclerosis and Alzheimer’s disease, and male youngsters showing more often signs of attention-deficit hyperactivity, autism and dyslexia, while older men have more often Parkinson’s disease (Schumacher et al., 2003; Schulman, 2007). As most brain sexual differences are derived from differences in early steroid hormone exposure, it is possible that sex differences in some disorders are partly influenced by abnormal steroid exposure or action in brain development. Moreover, the tissue- or brain region-specificity of some mutations could be connected to a time-point in development. Estrogen does not only originate from the foetuses gonads and the feto-placenta- 1.2. Steroid hormones in brain development 5 maternal unit (Tuchinsky et al., 1972; Trotter et al., 1999), but is also synthesized locally in the nervous tissues. Aromatase P450, the enzyme responsible for the transformation of testosterone in estrogen, is expressed in neurons of the developing and adult brain (Schumacher et al., 2003; Prange-Kiel et al., 2006; Von Schassen et al., 2006). Estrogen can also be synthesized de novo from cholesterol and the necessary enzymes are also present in the female and male developing brain. Because of local synthesis, steroid hormones can reach much higher local concentrations than anticipated and the greater aromatase activity in males than in females might be responsible for the sex brain dimorphism (Hutchison et al., 1997; Beyer, 1999; Quadros et al., 2007). Estrogen, estrogen’s receptors and P450 aromatase are at their highest levels in the brain either prenatally or in the first post-natal days, later abating to the adult base levels. This time interval is believed to be the estrogen’s sensitivity-window which allows for major permanent brain modifications, including brain sex differences connected to adult sexual behaviour (McCarthy, 2008). Because of these substantial and permanent effects in brain morphology and function potentially due to estrogen action, the developing brain has unique characteristics to guide or restrict the effects of this hormone, ensuring proper establishment of functional features, the early embryonic cell differentiation and modulation of motor and cognitive processes (Mendez et al., 2005; Daniel, 2006; McCarthy 2008). The aromatase sub cellular localization next to synaptic terminals shows a possible E action like a neurotransmitter. Estrogen receptors were also found in the axon and dendrites terminals, seemingly too far to return to the nucleus for transcription activation, suggesting a non-transcriptional effect of ER in the nervous system, or most probably, a local interaction with neighbouring mitochondria which locate next to synapses (McCarthy, 2008). To understand these diverse sex and tissue-specific phenotypes, it is critical to elucidate the processes of steroid hormone action and the possible interactions with mitochondria, in development and in connection to abnormal cell conditions. 6 Chapter 1. General introduction

1.3 Astrocyte mediation of steroid hormone neuroprotec-

tion

In the past years, the neuroprotective aspects of estrogen (E) and progesterone (P) in the central nervous system (CNS) have been analysed using in vitro and in vivo approaches. Besides direct interactions of E and P with neurons (Kipp et al., 2006; Stein et al., 2008), there are indications of indirect protective effects. Estrogen’s neuroprotective characteristics have repeatedly been demonstrated in vivo and in organotypical cultures (Dhapandani and Brann, 2002; Simpkins et al., 2005), suggesting the involvement of a non-neuronal brain cell type in the protective function. Of the non-neuronal brain cell types, astrocytes have the greatest potential for a critical role in the mediation of steroid-related neuroprotective effects. Astrocytes are the most abundant type of glial cells in the brain and are located juxtaposition to neurons, outnumbering them by a 10:1 ratio in the cortex. Additionally, ablation of astrocytes in vivo results in a signiﬁcant decrease of neuronal survival (Cui et al., 2001). Furthermore, astrocytes are well-described target cells for E and P, expressing the related steroid receptors (Jung-Testas et al., 1992; Santagati et al., 1994; Beyer, 1999; Quadros et al., 2007). As to the mechanisms involved in the indirect steroid-mediated neuroprotection through astrocytes, recent research has uncovered several vital functions which show that astrocytes, as well as other glial cells, are not mere structural support. Astrocytes play vital roles on the CNS, such as energy metabolism and protection of neurons, regulation of cerebral blood ﬂow, coupling neuronal activity and regulation of synapses. Astrocytes extend processes that contact the brain vasculature and envelop synapses. They transport nutrients to neurons, supporting neuronal metabolism and neurotransmitter synthesis. Astrocytes further maintain the extracellular environment, regulating ion concentrations and acid-base balance and importing extracellular glutamate (Azcoitia et al., 2001; Maragakis and Rothstein, 2006, Pellerin et al., 2007; Schousboe et al., 2007). They are involved in the regulation of growth, cell proliferation, and neuroprotection 1.4. Mitochondria 7 in the brain (Beyer, 1999; Beyer et al., 2003; Garcia-Segura et al., 2003; Kajta and Beyer, 2003; Kipp et al., 2006; McCarthy, 2008). Estrogen-induced changes of physiological parameters occur either in a direct non-genomic fashion (Beyer and Raab, 1998), or indirectly via interactions with nuclear and non-nuclear estrogen receptors (ER) that appear to be essential for estrogen-dependent gene expression and activation of various intracellular signalling pathways (Garcia-Segura, 2001; Kajta and Beyer, 2003; Beyer and Raab, 1998; Collins and Webb, 1999; Behl, 2003; Beyer et al., 2003; Arnold and Beyer, 2009). Besides the above described supportive and protective function, astrocytes role in the energy homeostasis of the brain, point at astrocyte mitochondria as an interesting compartment involved in neuroprotection.

1.4 Mitochondria

Mitochondria are particular cell organelles with specializations that point to their main function, the production of energy required for cellular processes. They are unique in having their own genetic material, called mitochondrial DNA (mtDNA) in multiple copies. Mitochondria are essential for a wide range of vital cell functions as metabolism, cell signaling, growth, cell proliferation and also apoptosis, thus playing an important role in development, aging and many diseases.

1.4.1 Mitochondria in aging and disease

Mitochondria play a fundamental role in many essential cell functions, including energy metabolism, thermogenesis, intracellular calcium homeostasis, cell growth, division, signalling and also regulation of apoptosis (Bettini and Maggi, 1992; Behl, 2002; Chen et al., 2004). Mitochondrial dysfunction affects not only the biochemical pathways located in mitochondria, but also the pathways located in the cytoplasm and can therefore cause important imbalance due to diminished cell energy, production of reactive oxygen species, 8 Chapter 1. General introduction disturbance of ionic balance and cytochrome C release, lately leading to apoptosis (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Chen et al., 2005). Mitochondria have their own genetic encoding molecules, the mitochondrial DNA (mtDNA) which is a 16.56 kbp circular double-stranded molecule, encoding 13 polypeptides, 12S and 16S rRNA and 22 transfer-RNAs, all essential for the proper functioning of mitochondria, and interestingly, there are no non-coding sections of genetic material in mtDNA. The 13 polypeptides encoded in mtDNA are all subunits of the respiratory chain complexes I, III, IV and V, only complex II is exclusively encoded by the nuclear genome. The compact structure of mtDNA, the lack of histone protection as well as the insufficient reparation mechanisms, and the exposition to the ROS by-products of the respiration tak- ing place in the organelle, make the mtDNA ten to twenty times more prone to mutations than the nuclear DNA (ncDNA) (Zeviani and Donato, 2004). But due to heteroplasmy, which is the simultaneous presence of mtDNA in multiple copies, possibly wild type and mutated, and mitochondrial dynamics (see Chapter 1.4.5), mutations accumulate without major symptoms until a threshold is reached (Nakada et al., 2001). Mitochondrial essential functions and characteristics explain the direct connection of their dysfunction to aging and various diseases including cancer, obesity, diabetes, cardiovascular disorders and several neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS), autosomal dominant optic atrophy, stroke, Charcot-Marie-Tooth, Parkinson’s, Huntington’s and Alzheimer’s diseases and others, stressing the importance of understanding the mechanisms underlying mitochondrial functions (Alexander et al., 2000; Delettre et al., 2000 Bach et al., 2003; Zuchner et al., 2004; Zeviani and Donato, 2004; Chen et al., 2004; Kijima et al., 2005; Chen and Chan, 2005; Singh et al., 2006; Thomas and Beal 2007; Poole et al., 2008; Schapira, 2008; Tatsuta and Langer, 2008). The new field of ’mitochondrial diseases’ includes a heterogeneous group of clinical phenotypes derived from a dysfunction of mitochondria and/or from abnormalities in mitochondrial DNA. Defects of the respiratory chain and oxidative phosphorilation are the main cause of 1.4. Mitochondria 9 mitochondrial disorders (Schapira, 2008) but these can also derive from abnormalities in metabolic pathways partially located in mitochondria. Furthermore, symptoms may be manifested at any age and the disorder may affect any tissue isolated or in combination with others, the tissues having higher energy demands, like muscles, brain, kidney and endocrine organs being more prone to being affected (Weissig et al., 2004; Armstrong, 2007). Mitochondrial diseases have unique characteristics not only because mitochondrial functions have vital importance for the cell function, but also due to the way they are often inherited. The mtDNA is normally derived only from the mother’s oocyte, the father’s sperm mtDNA usually not contributing to the fetus mtDNA. Depending whether the genetic defect is located on the mtDNA of the mother or on the ncDNA of one of the parents, so there is the respective inheritance pattern.The recent research on mitochondrial diseases, led to the emergence of a whole new field of biomedical research, the ’mitochondrial medicine’ focusing on mitochondrial-target therapeutics (Weissig et al., 2004; Armstrong, 2007) and stressing the importance of understanding mitochondrial mechanisms and their regulation.

1.4.2 Interaction of steroid hormones with mitochondria

The cell functions managed by mitochondria are interestingly also modulated by steroid hormones (Nilson et al., 2002; Kipp et al., 2006; Arnold and Beyer 2009). Estrogens have repeatedly been reported to play an important role in the regulation of mitochondrial function, such as energy metabolism, cell signalling, calcium homeostasis, and apoptosis (Bettini and Maggi, 1992; Behl, 2002; Chen et al., 2004; Kipp et al., 2006). Because mitochondria function as integrators of diverse cellular stress and toxic signals (Greenamyre et al., 1999; Reynolds, 1999; Choo et al., 2003; Beal, 2005), they seem to be perfect targets for steroid hormone regulation. Although the mechanisms for these actions are still not fully understood, it was found that mitochondria interact with glucocorticoid, estrogen, androgen, and progesterone receptors (Chen et al., 2004; Yang et al., 2004; Gavrilova-Jordan, 10 Chapter 1. General introduction

2007). Recent studies have shown that E promotes the expression of mitochondria-encoded subunits of the respiratory chain, thereby influencing mitochondrial respiratory activity and enhancing functional efficiency (Araujo et al., 2008; Irwin et al., 2008). Estrogen was also shown to have direct interactions with the respiratory chain complexes (Zheng and Ramirez, 1999). The modes of action of E are diverse, as it was shown to influence mitochondria not only through its nuclear receptors, but also directly binding to mitochondria in vitro and in situ (Yang et al., 2004; Stirone et al., 2005; Yager and Chen, 2007; Milner et al., 2008). Some of these effects were not subject to inhibition by classical nuclear ER antagonists, suggesting a non-genomic pathway of action (Beyer and Raab, 1998; Ivanova et al., 2001 (MAPK); Ivanova et al., 2002 (PI3K)). However, there is only limited information available on interactions between sex steroids and mitochondria, and the underlying intracellular mechanisms of cell protection by estrogen require further clarification.

1.4.3 Mitochondrial structure

Flexible in shape and size, mitochondria can assume a somewhat spherical to a tubular form, ranging from 1 to 10 µM in size and from a single to a thousand units in number inside each cell. Mitochondria are usually organized into a connected and dynamic network. They are especially abundant in high energy demanding cells, like muscle cells. Their distribution differs also inside a single cell, accumulating in areas where there is higher energy demand, such as near the basal membrane, near synapses or at the base of the ﬂagellum of ﬂagellated cells. This and the extremely ordered structure of mitochondria in high-energy demanding cells, like in muscle, suggest their form is intimately connected to their function (Bakeeva et al., 1981; Chen and Chan, 2005). The mitochondrial double membrane separates the organelle into very different com- partments in structure, biochemistry and function. The outer membrane is a simple phospholipid bilayer similar to the cell membrane, containing protein pores that make it permeable to small molecules like ions, simple nutrients, ATP and ADP. Therefore, 1.4. Mitochondria 11 the intermembrane space, between the two mitochondrial membranes, is similar to the cytosol, except for higher weight proteins. The inner membrane on the other side, is highly complex and protein rich. It is folded in cristae which greatly increase the total surface area, where the complexes for electron transport, the ATP synthase and transport proteins are organized. The inner room or matrix is a lightly alkaline space with a pH of about 7.7-8.0, housing the enzymes and molecules of the citric acid cycle, the energy carriers and substrates for the oxidative phosphorilation, mitochondrial ribosomes, tRNA and several copies of the mitochondrial DNA.

1.4.4 Cellular respiration

To produce energy for the cell, mitochondria extract energy from simple nutrient molecules like single fatty acid molecules, simple sugars and amino acids. These molecules derive from organic molecules like fats, carbohydrates and proteins from food, converted into simpler forms through cytoplasmic and mitochondrial enzymes, in very well described biochemical pathways (β-oxidation, citric acid or Krebs’ cycle). The set of reactions for this conversion is called catabolism. Mitochondria are responsible for several essential functions in metabolism such as fatty acid degradation (β-oxidation), degradation of metabolites (e.g. urea cycle), biosynthesis of heme-groups, nucleotides and aminoacids. The transformation of a wide range of molecules to a limited amount of biochemical forms allows for an optimization of the process. During metabolism, enzymes release the free energy of nutrients through oxidation, and transfer it to other molecules that become reduced, in so-called oxidation-reduction or redox reactions. The simple nutrient molecules generated in the intermediate metabolism, like the Acetyl-Coenzyme A (Acetyl- CoA), reduced nicotinamide-adenine-dinucleotide (NADH) and reduced ﬂavin-adenine- dinucleotide (FADH2), are subsequently conveyed to the respiratory chain, which is the last step of energy production, when the free energy is enough to generate ATP. This last part of the process consumes oxygen, and the whole is therefore called an aerobic metabolism. 12 Chapter 1. General introduction

The elements that process the successive steps of aerobic metabolism constitute the so called respiratory chain or oxidative phosphorylation system (Fig. 1.1), and are five protein complexes (I, II ,III, IV and V). Complexes I, III and IV are the H+ translocation machines that carry H+ from the matrix to the intermembrane space, leading to an electrochemical gradient, which is essential for the ATP production. The complex V or ATP synthase allows the H+ to pass through it in the direction of the gradient to produce ATP molecules from ADP and inorganic phosphate, storing energy as ATP. In the process, oxygen is needed as the final electron acceptor and therefore, the consumption of oxygen during the aerobic metabolism is called cellular respiration (Zheng and Ramirez, 1999; Araujo et al., 2008; Irwin et al., 2008). Interestingly, several studies suggest that mitochondria are biochemically and functionally diverse in different tissues (Benard et al., 2006) and even inside the same cell (Kuznetsov et al., 2006) suggesting an adaptation to different demands of energy at different sites. This is especially interesting in brain energy management, where different cell types intervene, demanding further research in specific cell types and their respiratory function.

1.4.5 Mitochondrial dynamics - fusion and ﬁssion

Mitochondria usually assume a tubular network morphology, but are present in cells in different shapes and quantity and also have distinct distributions inside the same cell. Mitochondria are dynamic organelles, its network being in constant, yet not fully understood, rearrangements. The mitochondrial morphology seems to be controlled by two opposing and balanced processes: fusion and ﬁssion (Figs. 1.2 and 1.3). These processes of organelle shaping occur far more times than the necessary to maintain mitochondrial morphology, and are therefore thought to be essential for the proper functioning of mitochondria. In fact, ﬁssion is necessary for the even distribution of mitochondria to the daughter cells during cell division, and is essential in apoptosis (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Stojanovski et al., 2004). Fusion, on the other side, serves the 1.4. Mitochondria 13

Figure 1.1: The respiratory chain, also called the oxidative phosphorylation system, is constituted by ﬁve protein complexes (I, II ,III, IV and V). Complexes I, III and IV are the H+ translocation machines that move H+ from the matrix to the intermembrane space, leading to an electrochemical gradient essential for the ATP production. The complex V or ATP synthase allows the H+ to pass through it in the direction of the gradient, that is, to the matrix, to produce ATP molecules from ADP and inorganic phosphate, storing energy.

exchange of carriers and nutrients, and allows for mtDNA molecules exchange, a cleaver gene complementation strategy, that prevents the detrimental effects of accumulating mtDNA mutations and cellular dysfunction, until a certain threshold is met (Nakada et al., 2001). Recent research has revealed several proteins involved in the processes regulating mitochondrial number and morphology, fusion and fission, each regulated by a specific set of proteins. Mitochondrial fusion is dependent on mitochondrial GTPases mitofusin 1 and 2 (Mfn1, Mfn2), located in the mitochondrial outer wall and being responsible for the membrane tethering between two mitochondria during fusion, and dominant optic atrophy-associated protein (OPA1), an intermembrane protein thought to participate in the mitochondrial inner membrane fusion and cristae modeling (Fig. 1.2). Mitochondrial fission on the other hand, is regulated by mitochondrial fission 1 protein (Fis1) an outer mitochondrial membrane-bound protein, and dynamin-related protein 1 (Drp1, also called dynamin-like protein 1, Dnm1l), a cytoplasmic protein, recruited to the mitochondrial 14 Chapter 1. General introduction surface during (Fig. 1.3). Fis1 and Drp1 are also responsible for the peroxisome fission, while their fusion intermediaries are still unclear. Other proteins were found to be involved in the processes of mitochondrial remodeling, like Ugo1, Rab32 and Mdv1 and there is currently active investigation to disclose these mechanisms (Santel et al., 2001; Rojo et al., 2002; Chen et al., 2003; Karbowski et al., 2003; Olichon et al., 2003; Cipolat et al., 2004;

Griparic et al., 2004; Chen and Chan,MITOCHONDRIAL 2005). FUSION

Mfn1/Mfn2

mitochondria

OMM fusion

Opa1

IMM fusion

Figure 1.2: Mitochondrial fusion is dependent on the mitochondrial GTPases mitofusin 1 and 2 (Mfn1, Mfn2) and the dominant optic atrophy-associated protein (OPA1). The mitofusins 1 and 2 are located in the mitochondrial outer wall and are responsible for the outer mitochondrial membrane (OMM) tethering between two mitochondria during fusion. OPA1 is an intermembrane protein located at the inner mitochondrial membrane (IMM) and is thought to participate in the IMM fusion and cristae remodelling, after the OMM fusion. There are other probable intermediates in this process.

The mitochondrial morphology maintenance is dependent on the equilibrium of these two opposing and constant processes. An imbalanced expression of fusion or ﬁssion proteins disrupts this equilibrium and causes dramatic changes in mitochondrial morphology leading to dysfunction (Yaffe, 1999; Karbowski and Youle, 2003). In humans, mutations in Mfn2 cause Charcot-Marie-Tooth neuropathy type 2A (Zuchner et al., 2004; Kijima et al., 2005) and mutations in OPA1 lead to autosomal dominant optic atrophy 1.4. Mitochondria 15 MITOCHONDRIAL FISSION

Drp1

Fis1

mitochondria

Drp1

fission

Figure 1.3: Mitochondrial fission is regulated by mitochondrial fission 1 protein (Fis1) and dynamin-related protein 1 (Drp1). Fis1, an outer mitochondrial membrane-bound protein, is thought to accumulate in greater amount at the fission sites. Drp1, a cytoplasmic protein, is recruited to the mitochondrial surface probably through the action of cofactors. Drp1 molecules are thought to connect to each other in form of a belt, which tightening provokes the fission of the mitochondria. Other intermediate proteins are probably involved in the process. After fission, Drp1 is again liberated into the cytoplasm.

(Alexander et al., 2000; Delettre et al., 2000). A down-regulation of fusion proteins or an over-expression of fission proteins causes fragmentation of the tubular mitochondrial network. Cells lacking Mfn1 and Mfn2, or OPA1, lacked mitochondrial fusion and showed severe dysfunctions, such as widespread heterogeneity of mitochondrial membrane potential, decreased cellular respiration, poor cellular growth, increased susceptibility to cell death, and apoptosis (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Chen et al., 2005). Mfn1 or Mfn2 Knock-out mice dye during embryogenesis, demonstrating the essential role of mitochondrial fusion in embryonic development and cell viability (Chen et al., 2003). Small interfering RNA (siRNA) knock-down of OPA1 induced fragmented mitochondria with dissipation of membrane potential and cristae disassembly, which render highly sensitized cells to exogenous apoptotic stimuli (Olichon et al., 2003; Griparic et al., 2004; Lee et al., 2004). In Caenorhabditis elegans, over-expression of fission protein Drp1 increased 16 Chapter 1. General introduction the number of mitochondrial divisions (Labrousse et al., 1999). Fis1 over-expression caused mitochondrial fragmentation and aggregation (Stojanovski et al., 2004). It is important to retain that mitochondrial fragmentation does not always result in apoptosis (Lee et al., 2004). On the other side, over-expression of fusion proteins or down-regulation of fission proteins stimulates mitochondrial fusion and the development of an elongated mitochondrial network, inhibiting the apoptosis-related mitochondrial fragmentation and therefore providing protection against different apoptotic stimuli (Smirnova et al., 2001; Santel et al., 2003; Sugioka et al., 2004).

1.4.6 Mitochondrial apoptosis regulation

Mitochondria play an essential role in coordinating programmed cell death or apoptosis. After a death signal, mitochondria divide profusely and release mediators of the apoptotic signalling. On the onset of apoptosis, the pro-apoptotic Bax translocates from the cytoplasm to the outer mitochondrial membrane, co-localizing with one of the fission related proteins, Drp1, on prospective fission sites. Drp1-mediated mitochondrial fission was found to be the initiating step of mitochondrial outer membrane permeabilization which leads to the release of cytochrome c, Smac/Diablo and other mediators from the intermembrane space into the cytoplasm, leading to caspase family proteases activation and the ignition of the apoptotic cascade (Karbowski and Youle, 2003). Interestingly, Fis1 knock-down by small interference RNA blocked the recruitment and activation of Bax to the surface of mitochondria after a death stimulus, indicating an essential role for Fis1 in apoptosis (Lee et al., 2004). Similarly, a loss of Drp1 blocked mitochondrial fission and cytochrome c release, although it did not significantly interfere with Bax activation (Breck- enridge et al., 2003; Lee et al., 2004). In the inner side of mitochondria, small interfering RNA (siRNA) knock-down of OPA1 induced fragmented mitochondria with dissipation of membrane potential and cristae disassembly, which render highly sensitized cells to exogenous apoptotic stimuli (Olichon et al., 2003; Griparic et al., 2004; Lee et al., 2004). 1.5. Aim of this work 17

The connection between mitochondrial fusion and apoptosis inhibition through cell signalling first came from the identification of Mfn2 (previously called hyperplasia suppressor gene, HSG) as an important anti-proliferating protein, interfering with the Ras pathway and blocking the signalling from growth factor receptors at the plasma membrane (Chen et al., 2004). Furthermore, the pro-apoptotic protein Bax was found to co-localize not only with Drp1, but also with Mfn2 on mitochondria during apoptosis (Karbowski and Youle, 2003) and an activated form of Mfn2 blocked Bax activation and cytochrome c release, protecting cells from apoptosis (Neuspiel et al., 2005). The over-expression of fusion proteins or down-regulation of fission proteins stimulates mitochondrial fusion and the development of an elongated mitochondrial network, inhibiting the apoptosis- related mitochondrial fragmentation and therefore providing protection against different apoptotic stimuli (Smirnova et al., 2001; Santel et al., 2003; Sugioka et al., 2004). These discoveries strongly point towards the importance of the morphological state of mitochondria in relation to their activity and cellular functions, through the involvement of cellular signalling cascades.

1.5 Aim of this work

Due to the great importance of mitochondrial functions in the cell, and the consequences of their dysfunction, it is of vital importance to understand the mechanisms underlying mitochondrial actions and their regulation, and with that simultaneously identify possible ways to keep proper function. Because mitochondria manage diverse cellular stress and toxic signals, an improvement of the respiratory energy production and a reduction of oxidative stress may represent a valid mechanism to stabilize energy homeostasis and to defend neural cells from death. Generally, an imbalance of the mitochondrial fusion and fission processes causes mitochondrial and cellular dysfunction indicating that fusion and fission processes regulating mitochondrial morphology and function are essentially linked 18 Chapter 1. General introduction to cell function and protection or apoptosis. Steroid hormones regulate many of the cellular processes depending on mitochondria. Understanding what are the connecting points between them, might clarify the regulation of those mechanisms and suggest ways to overcome dysfunction and find therapies. Fi- nally, having an insight of what effects steroid hormones have in mitochondria during development might enlighten processes of later degeneration and suggest ways to prevent it or postpone it, possibly improving our quality of life. In an endeavour to better understand the protective action of E and P on mitochondrial related processes, focus was put on respiration and mitochondrial fusion and fission gene transcription in astrocytes and an attempt was made to correlate their regulation with cell viability and apoptosis. The chapters 3 and 4 contain work previously published which was included in this thesis in accordance with the further authors. Chapter 2

Materials and methods

2.1. Reagents 21

2.1 Reagents

Chemicals were purchased from Roth (Karlsruhe, Germany), while reagents for molecular biological techniques and cell culture were acquired from Invitrogen (Karlsruhe, Germany), unless stated otherwise.

2.2 Animals

BALB/c mice were acquired from Harlan Winkelmann GmbH (Borchen, Germany). All procedures were in strict accordance with the published welfare rules for the care and use of laboratory animals at the University Clinics Aachen and the government of the State of Nordrhein-Westfalen, Germany.

2.3 Cell culture

Astrocytes were prepared from postnatal day 1 BALB/c mice (Pawlak et al., 2005). For separate female or male cell cultures, the gender of the donor was previously determined by visual inspection of the ano-genital distance which is larger in males, otherwise, mixed sex cultures were prepared. Brains were isolated and transferred into preparation buffer consisting of 10 mM HEPES, 154 mM NaCl, 10 mM glucose, 2 mM KCl, and 15 µM bovine serum albumine (BSA). The desired brain regions, cortex and/or mesencephalon, were separated and the meninges removed. Afterwards, each brain region was incubated in PBS containing 0.1% trypsin and 0.02% ethylenediamine-tetraacetic acid (EDTA) for 15 min, minced with a Pasteur pipette and ﬁltered through a 50 µm nylon mesh. The cell suspension was centrifuged at 300 g (Eppendorf, Hamburg, Germany) for 4 min. The cell pellet was re-suspended and the cell suspension plated onto poly-L-ornithine-coated (Sigma-Aldrich, Munich, Germany) culture dishes and cultured in Dulbecco’s modiﬁed Eagle’s medium (DMEM; PAA, Coelbe, Germany) supplemented with 20% (v/v) fetal calf 22 Chapter 2. Materials and methods serum (PAA, Coelbe, Germany), 50 U/ml penicillin, 50 µg/ml streptomycin, 0.25 µg/ml amphotericin B (Fungizoner), 2 mM glutamine (Glutamaxr). Cultures were maintained

◦ at 37 C in a humidified atmosphere of 95% air/5% CO2. Before reaching confluence, astrocytes were trypsinized and plated at lower density. After the third cell passage, sub-confluent cells were incubated in neurobasal medium (NBM) supplemented with B27, penicillin, streptomycin, Fungizoner, and L-glutamine for 48 hours, and then used for experiments. In some studies the cells were incubated in Neurobasal medium after the first passage, as indicated in the material and methods section of the applicable chapter. Astroglial cell cultures prepared from postnatal mouse brains revealed a high yield (> 95%) of GFAP-positive cells as determined by immunocytochemistry against the glial fibrillary acidic protein (GFAP), a known astrocyte-specific protein marker. Neurons, oligodendrocytes, and microglia were virtually absent from astrocyte cell cultures (Pawlak et al., 2005).

2.4 Cell treatments

Astroglial cell cultures placed on culture dishes at 37◦C in a humidiﬁed atmosphere in the presence of NBM were treated with 10−6 or 10−7 M 17β-estradiol (E) and/or progesterone (P) for different time intervals, as indicated in each of the following chapters. To inhibit nuclear estrogen receptors or progesterone receptors, cells were pre-treated with 10−6 M ICI 182,780 (Tocris, Bristol, UK) or RTI 3021-022 (a generous gift from Dr. Trotter, Ulm) respectively, for 30 min prior and in parallel to E or P treatment.

2.5 Cell viability

To access cell viability under treatments and control conditions, cells were stained with 1 µg/ml Hoechst 33342 (Invitrogen, Karlsruhe, Germany), a fluorescent marker for the viability of cells, for 5 min under culturing conditions. Subsequently, cells were washed 2.6. Reverse Transcription 23 twice with PBS, fixed with 4% (w/v) paraformaldehyde (Merck, Darmstadt, Germany) for 15 min in the dark, at room temperature, and washed twice with PBS. Pictures of the Hoechst-stained cells were taken from five distinct areas of the same size from each of the samples of 3-5 independent experiments, with a Hitachi HV-C20A Camera (Tokyo, Japan) connected to an inverted fluorescence microscope (Leica DMIL, Leica Microsystem, Wetzlar, Germany) at an excitation wavelength of 365 nm. The labeled nuclei of Hoechst- stained normal, apoptotic, and necrotic cells were detected accordingly to their specific morphological characteristics (Horvat et al., 2006). Cells were counted analysing the pictures taken, by using the software Adobe Photoshop 7.0 (Adobe Systems Inc., USA). Cell numbers are expressed as a percentage of the total cell number in control conditions, referred to as 100%. Additionally, cells were stained with propidium iodide to visualize necrotic astrocytes. Briefly, cells were fixed in 100% (v/v) ice-cold methanol at -20 ◦ C for 10 min, washed, incubated in a buffer containing 0.3 M NaCl and 0.03 M sodium citrate, pH 7.0, for 20 min and then stained with 500 nM propidium iodide for 10 min. Fluorescent necrotic cells were visualized at an excitation wavelength of 535 nm and using an emission filter of a wavelength > 605 nm and were quantified as described above for the Hoechst-stained samples.

2.6 Reverse Transcription

Total RNA was isolated from treated and non-treated control astrocytes cultures using PeqGold RNA pure (PeqLab, Erlangen, Germany), according to the manufacturer’s in- structions. RNA concentration was measured photometrically using BioPhotometer (Ep- pendorf). RNA integrity was tested randomly by 1% (w/v) agarose denaturing gel electrophoresis and ethidium bromide staining and visualized under UV-illumination. First strand complementary DNA (cDNA) was synthesized from 0.5 µg total RNA. In brief, total RNA dissolved in 7 µl diethyl pyrocarbonate- (DEPC-) H2O was pre-incubated at 24 Chapter 2. Materials and methods

85◦C for 5 min and placed immediately on ice. Subsequently, the reaction buffer consisting of 8 U/ µl M-MLV Reverse Transcriptase, 8 mM dithiotreitol, 40 mM Tris-HCl, 60 mM KCl,

r and 2.4 mM MgCl2, and 0.4 mM each dNTP (Roti-mix PCR3, Roth, Karlsruhe, Germany), was added to the RNA giving a ﬁnal volume of 12.5 µl. After one hour incubation at 37◦C, reverse transcription was stopped by heat-inactivating the enzyme at 70◦C for 15 min. Addition of water instead of RNA served as negative control. Transcripts of 18S rRNA and the housekeeping gene hypoxanthine guanine phosphoribosyl transferase (HPRT), spanning over intron-exon borders, served as control for RNA purity and normalization of sample quantity.

2.7 Semi-quantitative reverse transcriptase - polymerase chain

reaction (sqRT-PCR) analysis

To estimate the mRNA transcript levels of catalytic subunits of the respiratory chain, sqRT- PCR was performed. 18S rRNA (18S ribosomal RNA) and HPRT (hypoxanthine guanine phosphoribosyl transferase) served as internal controls and were amplified to normalize the relative levels of cDNA in different samples. Forward and reverse primers (Tables 2.1 and 2.2) were designed by the author of this thesis for the specific amplification of the selected genes of interest: mitochondria-encoded subunits of the respiratory chain ND1 (complex I, NADH ubiquinone oxidoreductase, subunit 1), cytB (complex III, ubiquinol cytochrome c oxidoreductase), cox2 (complex IV, cytochrome c oxidase, subunit 2), ATP6 (complex V, ATP synthase, subunit 6), the mitochondrial fusion and fission intermediaries mitofusin 1 (Mfn1), mitofusin 2 (Mfn2), fission 1 homolog (Fis1), dynamin-related protein 1 (Drp1, also called dynamin 1-like protein, Dnm1l), a proliferation marker, proliferating cell nuclear antigen (PCNA), an anti-apoptotic marker, B-cell lymphoma protein 2 (Bcl2), and a pro-apoptotic marker, Bcl2-associated X protein (Bax). For each set of primers, a basic local alignment search tool (BLAST, NCBI) search revealed that sequence homology 2.7. Semi-quantitative reverse transcriptase - polymerase chain reaction (sqRT-PCR) analysis 25 was obtained only for the target gene.

Table 2.1: Primer information for the mitochondrial-encoded respiratory chain subunits and the mitochondrial fusion and ﬁssion intermediaries.

Gene Genbank accession Primer sequence 5’-3’ Mitochondrial-encoded respiratory chain subunits ND1 NC 005089.1 TTTACGAGCCGTAGCCCAAA GGCCGGCTGCGTATTCTAC CytB NC 005089.1 ACGAAAAACACACCCATTATTT CCGTTTGCGTGTATATATCGGATT Cox2 NC 005089.1 AACCGAGTCGTTCTGCCAAT AACCCTGGTCGGTTTGATGTT ATP6 NC 005089.1 AGCAGTCCGGCTTACAGCTAA GGAGGGTGAATACGTAGGCTTG Mitochondrial fusion and ﬁssion intermediaries Mfn1 NM 024200 CAG AGA AGA GGG TTT ATT CA ACT CAT CAA CCA AAA CAG AT Mfn2 NM 133201 TGA ATG TTG TGT TCT TTC TG AAG TGC TCT CTG CTA AAT GT Fis1 NM 025562 AGA TGG ACT GGT AGG CAT GG CTA CAG GGG TGC AGG AGA AA Drp1 NM 152816 TTT GCT CGT GTG AAG ACT GG TCC TGG AGC TTC CTT TCT GA

To establish optimal PCR conditions, a linear regression analysis was performed for all the genes of interest as well as for the housekeeping genes, 18S rRNA and HPRT, at a range of 12-38 cycles under standard conditions. Analysis of the sqRT-PCR products revealed single bands for all cDNAs of catalytic subunits at the correct molecular weight (see table 2.3). The cycle numbers applied in the sqRT-PCR analysis met the requirements of linear regression for each gene. Optimal cycle numbers between 20 and 32 for all genes studied reflected approximately 50-70% of maximal OD values obtained at 38 cycles, when gene amplification became saturating. This guaranteed that all mRNAs of interest were analysed within the linear PCR amplification range and suitable for further semi- and quantitative gene expression studies. As primers for HPRT were designed as intron over-spanning gene sequences, the observation of a single band for HPRT indicated that 26 Chapter 2. Materials and methods the isolated RNA and subsequently the synthesized cDNA were free of genomic DNA impurities.

Table 2.2: Primer information for the markers of apoptosis and proliferation and the internal housekeeping controls.

Gene Genbank accession Primer sequence 5’-3’ Markers for apoptosis and proliferation Bcl2 BC089016 CAT CCC AGC TTC ACA TAA CC GCA ATC CGA CTC ACC AAT AC Bax NM 007527 GGC AGA CAG TGA CCA TCT TT AGG GGA CCT GAG GTT TAT TG PCNA X53068 TTT TTC ACA AAA GCC ACT CC TCA GGA GCA ATC TTC AAA GG Internal controls HPRT NC 000086.6 GCTGGTGAAAAGGACCTCT CACAGGACTAGAACACCTGC 18S rRNA X00686 CGG CTA CCA CAT CCA AGG AA GCT GGA ATT ACC GCG GCT

The PCR reaction was performed in a reaction buffer containing 0.032 U/µl Taq DNA polymerase, 1.5 mM MgCl2, 150 mM Tris-HCl, 60 mM KCl, 0.2 mM dNTPs, and 0.2 µM primers (Sigma Genosys, Steinheim, Germany) for 0.75 µl of cDNA of each gene of interest in a total volume of 12.5 µl. The negative control contained water instead of cDNA. The PCR conditions (Table 2.3) for the respiratory chain subunits of interest started with an initial denaturation at 95◦C for 5 min followed by 20 (ATP6) or 26 (ND1, cytB, cox2) amplification cycles composed of 95◦C for 45 s, 53◦C for 45 s, and 72◦C for 60 s, and ended with a final elongation of 5 min at 72◦C. The PCR conditions for the mitochondrial fusion and fission genes of interest was similar, only the amplification cycles varied: 26 (Bcl2) or 30 (Mfn1, Mfn2, Fis1 and Drp1) cycles composed of 95◦C for 45 s, 53◦C (Mfn1, Mfn2) or 58◦C (Fis1, Drp1) and 72◦C for 60 s. The PCR for the internal control genes 18S rRNA and HPRT, and for the proliferation marker (PCNA), anti-apoptotic marker (Bcl2), and pro-apoptotic marker (Bax) was again similarly constructed, with specific amplification cycles (Table 2.3): 14 (18S rRNA), 26 (Bcl2), 28 (HPRT) or 32 (PCNA, Bax) amplification cycles composed 2.7. Semi-quantitative reverse transcriptase - polymerase chain reaction (sqRT-PCR) analysis 27 of 95◦C for 45 s, 59◦C (Bcl2, Bax, PCNA) 60◦C (18S rRNA) or 61◦C (HPRT) for 45 s, and 72◦C for 60 s. The PCR products were then separated together with a semi-quantitative DNA marker ladder, E-Gelr low range quantitative DNA ladder by a 1.5% (w/v) agarose gel electrophoresis containing ethidium bromide, and visualized under UV-illumination (AlphaImagerr, Alpha Innotech, Biozym Scientific GmbH, Oldendorf, Germany). Semi- quantitative analysis of the PCR product bands normalized to the quantitative gel marker bands was performed using AlphaEaseFCTM Software (Alpha Innotech Corporation 2003). Results of at least three independent experiments normalized against the internal controls are presented as percentage of control.

Table 2.3: Polymerase Chain Reaction (PCR) details: annealing temperature, cycle number and product size.

Gene Annealing temperature Cycle number Product size (bp) Mitochondrial-encoded respiratory chain subunits ND1 53◦C 26 260 CytB 53◦C 26 246 Cox2 53◦C 26 155 ATP6 53◦C 20 182 Mitochondrial fusion and ﬁssion intermediaries Mfn1 53◦C 30 191 Mfn2 53◦C 30 164 Fis1 57◦C 30 211 Drp1 58◦C 30 241 Markers for apoptosis and proliferation Bcl2 59◦C 26 175 Bax 59◦C 32 228 PCNA 59◦C 32 194 Internal controls HPRT 61◦C 28 250 18S rRNA 60◦C 14 186 28 Chapter 2. Materials and methods

2.8 Quantitative real time-PCR (qRT-PCR) analysis

Quantitative real-time PCR (qRT-PCR) analysis of the genes of interest was performed using SYBRr Green technology and carried out on the iQ5 detection system (Bio-Rad, Munich, Germany). Standard and sample cDNA obtained after reverse transcription (see above) were diluted 1:10 and added to a solution containing 5 µM primers and IQ SYBR Green Supermix (Bio-Rad, Munich, Germany) consisting of 25 U/ml iTaq polymerase,

50 mM KCl, 20 mM Tris-HCl, 0.2 mM each dNTP, 3 mM MgCl2, SYBR Green I and stabilizers. Using a standard curve of cDNA dilutions the amount of sample cDNA was quantified. The qRT-PCR protocol was composed of an initial denaturation step for 3 min at 95◦C followed by 40 cycles consisting of 10 s at 95◦C, 30 s at the appropriate annealing temperature for the target gene (Table 2.3), 30 s at 72◦C, and 10 s at 78◦C. To obtain melting curves for the resulting PCR products, a final step was added to the qRT-PCR consisting of 81 cycles for 10 sec each stepwise increasing the cycle temperature by 0.5◦C, beginning with the cycle 1 at 55◦C and ending with cycle 81 at 95◦C. The threshold cycle (Ct) for real-time quantitative PCRs is reported as the cycle at which a significant increase in fluorescence takes place. Ct values change linearly with increasing target amount. The PCR products were quantified using the relative ∆Ct method. Relative quantification relates the PCR signal of the target transcript to that of HPRT in treated with respect to untreated cells. A test for an approximately equal efficiency of target amplification was performed by looking at how the ∆Ct value varies with template dilutions. 18S rRNA and HPRT served as endogenous control in the validation experiments. The absolute value of the slope of log input amounts versus ∆Ct should be less than -3.3 and the efficiency approximately 100%. The validation experiments passed this test. The results are expressed as the average of at least three samples of three independent experiments for control and treated cells. 2.9. Measurement of mitochondrial and nuclear DNA 29

2.9 Measurement of mitochondrial and nuclear DNA

To quantify the amount of mitochondria in a deﬁned number of E/P treated versus control astroglial cells, the content of mitochondrial DNA was taken as an indirect measure for the mitochondrial mass. Total DNA was extracted using DirectPCR lysis reagent (Viagen Biotech, Los Angeles, USA) by incubating cells for 5 h at 55◦C and subsequent inactivation of the enzyme reaction at 85◦C for 45 min. The suspension was then used for PCR of the mitochondrial marker gene ATP6 (17 cycles at 53◦C) and HPRT as a nuclear control gene (28 cycles at 61◦C). The results are expressed as the ratio of ATP6 to HPRT expression in steroid-treated cells, as a percentage of control (untreated cells).

2.10 Isolation of mitochondria from primary astrocytes

Mitochondria were isolated from primary astrocytes in the previously described culture conditions (see section 2.3). Treated and control cells from culture dishes (about 1x108 astrocytes) were collected by scraping the cells from the dish surface and precipitating them by centrifugation at 300 g (Eppendorf 5415R) for 4 min. The cell pellet was re-suspended in 1 ml ice-cold isotonic mitochondrial isolation buffer consisting of 0.25 M sucrose, 5 mM HEPES, and 1 mM EDTA. Suspended cells were homogenized in a pre-cooled plastic vessel with pestle (Eppendorf). To precipitate nuclei and cell fragments the disrupted cells were centrifuged at 500 g for 10 min at 4◦C. The resulting supernatant was transferred into a pre-cooled tube, while the pellet was re-suspended in 500 µl mitochondrial isolation buffer for a second turn of homogenization to optimize mitochondria isolation. Both supernatants were collected and centrifuged at 9400 g for 15 min at 4◦C. The pellet containing mitochondria was re-suspended in 22 µl mitochondrial buffer and kept on ice for the duration of the experiments. 30 Chapter 2. Materials and methods

2.11 Polarographic measurements

Mitochondria were isolated as described above (section 2.10), from primary cortical and mesencephalic astrocytes after E treatment and under control conditions. Mitochondria were kept on ice during polarographic measurements. Respiration of isolated mitochondria was measured polarographically at 25◦C in a volume of 500 µl of a buffer consisting of 70 mM sucrose, 220 mM mannitol, 2 mM HEPES, 5 mM MgCl2, 5 mM KH2PO4, 1 mM EDTA. The measurements were conducted after subsequent additions of 20 µM rotenone, 10 mM succinate, 30 µM ADP (state 3 mitochondrial respiration), and 3 µM carbonyl cyanide m- chlorophenylhydrazone (CCCP; uncoupled mitochondrial respiration), using an Oxygraph system (Helmut Saur Laborbedarf, Reutlingen, Germany). The activity of mitochondria is given as Turnover Number (TN) which is the consumed O2 (µM) / [(time (min) x total amount of protein (mg)]. TN was determined at state 3 (in the presence of ADP) and state 4 (after ADP has been phosphorylated reﬂecting ATP-dependent mitochondrial respiration) respirations as well as at the uncoupled state of mitochondrial respiration. The Respiratory Control Ratio (RCR), an index of the electron transport chain coupling, was calculated as the ratio of state 3 to state 4 respiration rates.

2.12 Protein content determination

The protein content of isolated mitochondria was measured applying the Bio-Rad Dc Protein Assay (Munich, Germany) microtiterplate procedure according to the manufacturer’s protocol. Samples of frozen mitochondria were solubilised in mitochondrial buffer resulting in a 1:5 dilution of original mitochondrial samples, sonicated for 30 sec using Bransonic 52 ultrasonic bath (Branson, Soest, Netherlands) and then used for the protein assay. Samples that were centrifuged prior to the protein assay gave the same results as those that were not. BSA served as standard. 2.13. Statistical analysis 31

2.13 Statistical analysis

Data are presented as the mean ± SEM of at least three to five independent experiments. For statistical analysis, data was analyzed by applying ANOVA followed by a posthoc Student’s unpaired t-test and regarded as statistically significant at P≤0,05. The statistical significance was further divided into sections P<0,05, P<0,01 and P<0,001 according to their fit. 32 Chapter 2. Materials and methods Chapter 3

Steroid hormone effects on mitochondrial respiration

Estrogen regulates mitochondrial gene expression and respiratory chain activity in cortical and mesencephalic astrocytes1

1Adapted from: Araujo, Beyer and Arnold 2008. Journal of Neuroendocrinology 20, 930-941.

3.1. Introduction 35

Abstract

The regulation of mitochondrial energy metabolism plays an essential role in the CNS. Abnormalities of the mitochondrial respiratory chain often accompany neurodegenerative diseases. This makes mitochondria a perfect target for strategies of cellular protection against toxic compounds and pathological conditions. Steroid hormones, such as estrogen, are well-known to fulﬁl a protective role in the brain during ischemic and degenerative processes. Since astrocytes function as the major energy supplier in the CNS, the estrogen effects on the mitochondrial respiratory chain of this cell type were here analysed. In these studies, semi- and quantitative PCR analysis of gene expression and polarographic measurements of the respiratory chain activity of mitochondria were applied. It was observed that structural and functional properties were regulated dependent on the estrogen exposure time and the brain region but independently of the nuclear estrogen receptors were observed. These studies demonstrate that long-term estrogen exposure increases the subunit gene expression of respiratory chain complexes and the mitochondrial DNA content, thereby indicating an up-regulation of the amount of mitochondria per cell together with an increase of mitochondrial energy production. This could represent an important indirect mechanism by which long-term estrogen exposure protects neurons from cell death under neurotoxic conditions. On the other hand, short-term effects of estrogen were observed on the activity of mitochondrial, proton-pumping respiratory chain complexes. In astrocytes from cortex, respiratory chain activity was decreased, whereas it was increased in astrocytes from mesencephalon. An increased production of reactive oxygen species would be the consequence of an increased respiratory chain activity in mesencephalic astrocytes. This could serve as an explanation for different efﬁciencies of estrogen-mediated short-term protection in distinct brain regions, but indicates also the limitations for a therapeutic short-term application of estrogen.

3.1 Introduction

Estrogen, a steroid sex hormone, affects numerous physiological processes in the central nervous system (CNS), such as the early embryonic cell differentiation and modulation of motor and cognitive processes (Mendez et al., 2005; Daniel, 2006). Beside these, estrogen exhibits neuroprotective properties in the embryonic and adult brain, preventing nerve cells from oxidative stress, toxic factors, and chemically induced apoptosis (McEwen, 1992; Beyer, 1999; Harms et al., 2001; Behl, 2002; Garcia-Segura, 2001; Kajta and Beyer, 2003; Kipp et al., 2006). In the CNS, astrocytes play an important role in mediating the effects of estrogen. They represent not only the direct target cells for estrogen, but are also of major importance for mediating neural cell protection (Behl, 2002; Kipp et al., 2006; Dhandapani and Brann, 2007). Estrogen-induced changes 36 Chapter 3. Steroid hormone effects on mitochondrial respiration

of physiological parameters occur via interactions with nuclear and non-nuclear estrogen receptors (ER) that appear to be essential for estrogen-dependent gene expression and activation of various intracellular signalling pathways (Garcia-Segura, 2001; Kajta and Beyer, 2003; Beyer and Raab, 1998; Collins and Webb, 1999; Behl, 2003; Beyer et al., 2003). Previously, estrogen was also found to interact with mitochondria (Chen et al., 2004; Yang et al., 2004). Because mitochondria function as integrators of diverse cellular stress and toxic signals (Greenamyre et al., 1999; Reynolds, 1999; Choo et al., 2003; Beal, 2005), an improvement of energy production and a reduction of oxidative stress may represent a valid mechanism to stabilize energy homeostasis and to defend neural cells from death. In this context, estrogens have repeatedly been reported to play an important role in the regulation of mitochondrial function, such as energy metabolism, cell signalling, calcium homeostasis, and apoptosis (Behl, 2002; Kipp et al., 2006; Chen et al., 2004; Bettini and Maggi, 1992). However, the underlying intracellular mechanisms of neuroprotection by estrogen are not fully understood. This raises the question of how estrogen may regulate mitochondrial function. One possibility is to control expression of mitochondrial proteins encoded by the nuclear genome involving the activation of classical nuclear estrogen receptors. Another possibility is to regulate mitochondrial genes or to directly modify enzyme activity of respiratory chain complexes. Supporting both pathways, the first evidence came from studies by Bettini and Maggi (1992) and Law et al.(1994), which have shown that oestradiol treatment is associated with increased transcript levels of mitochondrial (mt) DNA- and nuclear (nc) DNA-encoded genes for mitochondrial respiratory chain proteins. The screening of a cDNA library from the hippocampus of estrogen-stimulated ovariectomized rats identified subunit III of complex IV of the mitochondrial respiratory chain as a target gene of estrogen (Bettini and Maggi, 1992). Furthermore, the non-classical signalling of estrogen is a wide-spread phenomenon. It was shown that mitochondrial cytochrome c oxidase subunit III expression was induced in the midbrain through non-classical steroid signalling mechanisms (Bjorklund and Lindvall, 1984). These direct and rapid interactions of estrogen with mitochondria may play an important role in the control of cellular energy homeostasis. A signalling pathway regulating mitochondrial function and involving a direct interaction of estrogen with mitochondria is indicated by mitochondrial location of ERs in hippocampal neurons and cerebral blood vessels (Chen et al., 2004; Yang et al., 2004; Stirone et al., 2005). However, the exact mechanism of the regulation of astrocytic mitochondria remains unknown. In the present study, data is presented on the regulation of catalytic subunit expression of the proton translocating respiratory chain complexes by estrogen. It is shown that estrogen induced an up-regulation of mRNA transcripts of catalytic respiratory chain subunits affecting respiratory chain activity. These results suggest that estrogen influences cellular energy homeostasis via a long-term action of estrogen by up-regulating the expression of catalytic subunits and also via a short-term action by directly affecting mitochondrial ATP production, both through non-classical estrogen signalling mechanisms. 3.2. Materials and Methods 37

3.2 Materials and Methods

For information on the materials and methods used in this study, please refer to Chapter 2. For information on the author’s contributions please refer to Contributions. In this investigation, cells were treated with 10−6 M 17β-estradiol (E2, generally referred to as estrogen, E) for 4, 8, 12 or 24 hours, as described in the ﬁgure legends.

3.3 Results

3.3.1 Regulation of the expression of catalytic respiratory chain sub-

units by estrogen

The expression of catalytic subunits of the mitochondrial respiratory chain complexes was assessed in primary neonatal mouse mesencephalic and cortical astroglial cell cultures. Immunocytochemical analysis for glial fibrillary acidic protein (GFAP) revealed a high yield (> 95%) of GFAP-positive astrocytes, whereas neurones, oligodendrocytes, and microglia were virtually absent from the astroglial cell cultures. This is in accordance with previous studies (for further details please see Chapter 2). To determine whether estrogen regulates the gene expression of catalytic respiratory chain subunits, one representative mitochondrial gene was selected for each of the mitochondrial proton translocating respiratory chain complexes: ND1 (complex I, NADH ubiquinone oxidoreductase), cytB (complex III, ubiquinol cytochrome c oxidoreductase), cox2 (complex IV, cytochrome c oxidase, COX), ATP6 (complex V, ATP synthase). Firstly, the expression of these genes was analysed, by applying a qualitative, semi- quantitative RT-PCR technique (sqRT-PCR; for information on the methods and materials used, please refer to Chapter 2). To establish optimal PCR conditions, linear regression analysis was performed for ND1, cytB, cox2, ATP6, as well as for the housekeeping genes, 18S rRNA and HPRT, at a range of 12-32 cycles under standard conditions. Analysis of the sqRT-PCR products revealed single bands for all cDNAs of catalytic subunits at the correct molecular weights. The cycle numbers applied in our sqRT-PCR analysis met the requirements of linear regression for each gene. Optimal cycle numbers between 20 and 28 for all genes studied reflected approximately 50-70% of maximal optical density values obtained at 38 cycles, when gene amplification became saturating. This guaranteed that all mRNAs of interest were analysed within the linear PCR amplification range and suitable for further semi- and quantitative gene expression studies. As primers for HPRT were designed as intron over-spanning gene sequences, the observation of a single band for HPRT indicated that the isolated RNA, and subsequently the synthesized cDNA, were free of genomic 38 Chapter 3. Steroid hormone effects on mitochondrial respiration

DNA impurities (for further information, please refer to Chapter 2). The preliminary qualitative studies on cultured astrocytes from cortex and mesencephalon revealed that all mitochondrial genes of interest are expressed in considerable amounts. This is in accordance to the observation of mitochondria organelles and mitochondrial activity in these cells (see below). To study the effect of estrogen on the expression of catalytic subunits of the mitochondrial respiratory chain in astroglial cell cultures, the optimal concentration and time range of estrogen treatment had to be established. Primary astrocytes were treated with different concentrations of 17β-estradiol (E2) between 10−10 and 10−6 molar and tested for significant changes of the catalytic subunit expression. A significant up-regulation of mRNA was observed for all catalytic subunits (ND1, cytB, cox2, ATP6) at all E2 concentrations tested (data not shown). For further studies, micromolar E concentrations were chosen as to cover all potential pathways of E action (i.e. classical and non-classical ER signalling as well as direct interactions of E with proteins). Thus, primary astrocytes from cortex and mesencephalon were treated with 10−6 M E2 for different time intervals. Subsequently, the mRNA expression of catalytic subunits of respiratory chain complexes was assessed by sqRT-PCR and compared with the expression profile of the housekeeping genes, 18S rRNA and HPRT, which served as control set 100% (Fig. 3.1). The results show a maximal observed increase (20-50%) for the expression of ND1 (complex I), cytB (complex III), cox2 (complex IV), and ATP6 (complex V) after 8 and 12 hours of E treatment of cortical (Figs. 3.1 A) and mesencephalic astrocytes (Figs. 3.1 B), respectively. ND1 and cox2 were up-regulated by ca. 20 and 40%, respectively, in astrocytes from both brain regions. Brain region-specific differences were observed for E-induced up- regulation of the catalytic subunits cytB and ATP6. In cortical astrocytes, cytB and ATP6 were increased by 16±5% and 29±2%, respectively (Fig. 3.1 A), whereas mesencephalic astrocytes showed an elevation of cytB by 51±16% and ATP6 by 56±15% (Fig. 3.1 B). This indicates that the E-induced up-regulation of cytB and ATP6 is more pronounced in mesencephalic astrocytes than in cortical astrocytes. To substantiate the semi-quantitative gene expression results, a quantitative RT-PCR (qRT-PCR) analysis followed. HPRT expression values served as control and were set to 100%. The qRT-PCR data analysis revealed an up-regulation of the catalytic respiratory chain subunits cytB by 13±8%, cox2 by 28±14%, and ATP6 by 86±27%, except for ND1 expression which showed a down-regulation by 37±1% in cortical astrocytes (Fig. 3.2 A, darker bars). In mesencephalic astrocytes, gene expression of all catalytic subunits was increased (Fig. 3.2 B, darker bars). ND1 transcript levels were elevated by 41±29%, cytB by 43±11%, cox2 by 42±12%, ATP6 by 101±48%. Comparing astrocytes from two different brain regions, qRT-PCR and sqRT-PCR both revealed increase of catalytic subunit expression that was more pronounced in mesencephalic than in cortical astrocytes. An E exposure time for longer than 8 h and 12 h for cortical and mesencephalic 3.3. Results 39

(A) Cortex 180 160 ** 140 ** * * ** * * ** 120

100 ND1 80 CytB Cox2 60 ATP6 (% of control)

mRNA expression 40 20 0 C 4 8 12 24 Time of E treatment (h)

(B) Mesencephalon 180 * * 160 * * 140 * * * * ** 120 * **

100 ND1 80 CytB Cox2 ATP6

(% of control) 60

mRNA expression 40 20 0 C 4 8 12 24 Time of E treatment (h)

Figure 3.1: Time course of estrogen (E) effects on the expression of catalytic subunits of the mitochondrial respiratory chain. Cortical astrocytes (A) and mesencephalic astrocytes (B) in culture were incubated in the presence of 10−6 M 17β-estradiol for 4, 8, 12, 24 h. Total RNA was isolated and ﬁrst strand cDNA was synthesized for ND1, cytB, cox2 and ATP6 in a sqRT-PCR. Data were normalized to housekeeping genes 18S rRNA and HPRT and represent means ± SEM of three independent experiments of three different cell preparations. Statistically signiﬁcant differences (* - P<0.05, ** - P<0.01, *** - P<0.001) between samples were determined using Student’s unpaired t-test.

astrocytes, respectively, did not show any further increase in gene expression. Therefore, these time ranges were applied for further investigations of the E effects on mitochondrial function. 40 Chapter 3. Steroid hormone effects on mitochondrial respiration

To determine whether the nuclear ERs would be involved in regulating mitochondrial gene expression, the ER antagonist ICI 182.780 was applied to astroglial cell cultures 30 min prior to and throughout the E treatment. The simultaneous application of an ER antagonist did not interfere with the observed E effect on catalytic respiratory chain gene expression in astrocytes from both brain regions (Fig. 3.2 C, D). This observation is in line with our expectations indicating that E stimulates mitochondrial gene expression independently of the nuclear estrogen receptor pathway.

3.3.2 Effect of Estrogen on the ratio of mitochondrial versus nuclear

DNA levels

An up-regulation of mRNA transcripts of catalytic respiratory chain subunits could be paralleled, on one hand, by regulation of mitochondrial activity, and, on the other hand, by elevation of the amount of mitochondria per cell. The latter could be reflected by an overall increase of mitochondrial protein and/or mitochondrial DNA content. To assess the effect of E on the amount of mitochondria in astrocytes, the ratio of mitochondrial DNA to nuclear DNA was measured in a defined number of astroglial cell cultures from cortex and mesencephalon before and after E2 treatment at a concentration of 10−6 M for 8 and 12 h, respectively. In Fig. 3.3, a significant increase of the mitochondrial versus nuclear DNA content by 22±13% becomes evident in cortical astrocytes, whereas mesencephalic astrocytes show a greater elevation by 54±15%.

3.3.3 Long-term estrogen effects on mitochondrial activity

On the basis of these gene expression studies, it is proposed here that there is a concomitant change of mitochondrial activity. To determine whether there would be a correlation between an increased expression of mitochondrial catalytic respiratory chain subunits and mitochondrial respiratory activity by E, the subsequent measurement of the succinate-stimulated respiration of isolated intact mitochondria from E- treated in comparison to control astrocytes from cortex and mesencephalon followed. After E treatment, cortical and mesencephalic astrocytes were incubated in NBM for an additional 4h to allow translation of mRNA transcripts and protein targeting into mitochondria (Strausberg and Butow, 1977; Barrientos et al., 2004). Subsequently, intact mitochondria were isolated from E-treated and -untreated cortical and mesencephalic astroglial cell cultures. Respiratory chain activity was measured immediately thereafter by assessing mitochondrial oxygen consumption stimulated by addition of succinate as substrate. Unexpectedly, signiﬁcant changes were observed neither for mitochondrial respiration of mitochondria from E-treated 3.3. Results 41

(A) (B) 250 Cortex 250 Mesencephalon * * * * 200 * 200 * * * * * 150 * sqPCR 150 * * sqPCR qPCR qPCR 100 100 (% of control) (% of control)

mRNA expression 50 mRNA expression 50

0 0 C ND1 CytB Cox2 ATP6 C ND1 CytB Cox2 ATP6

200 200 Cox2 ATP6 Cox2 ATP6 * * * * * * * 150 * * 150 *

100 100 (% of control) 50 (% of control) 50 mRNA expression mRNA expression

0 0 CEE+ICICEE+ICI CEE+ICICEE+ICI

Figure 3.2: Effect of estrogen (E) on the expression of catalytic subunit genes of the mitochondrial respiratory chain complexes. Primary astrocytes from cortex (A, C) and mesencephalon (B, D) in culture were incubated in the presence of 10−6 M 17β-estradiol for 8 and 12 h, respectively. Total RNA was isolated and ﬁrst strand cDNA was synthesized. Figures A and B show the results of a semi-quantitative sqRT-PCR (left bars in light grey) and a quantitative qRT-PCR (right bars in dark grey) using a Bio-Rad iQ5 detection system (Munich, Germany). The expression of the catalytic subunits ND1 (complex I), cytB (complex III), cox2 (complex IV), ATP6 (complex V) was analysed in E-treated astrocytes from cortex (A) and mesencephalon (B), normalized to HPRT and 18S rRNA as housekeeping genes and related to the expression in untreated control astrocytes set 100% (C). The qRT-PCR data were analyzed using the relative ∆Ct method. Figures C and D represent a sqRT-PCR analysis of the expression levels of Cox2 (lighter bars) and ATP6 (darker bars) of cortical (C) and mesencephalic (D) astrocytes treated with 10−6 M 17β-estradiol (estrogen, E) and 10−6 M 17β-estradiol plus 1 µM ICI 182.780 (E+ICI) normalized to housekeeping genes and related to that in untreated control astrocytes set 100% (C). Data represent means ± SEM of three independent experiments of three different cell preparations. Statistically signiﬁcant differences (* - P<0.05, ** - P<0.01, *** - P<0.001) between samples were determined using Student’s unpaired t-test.

versus untreated cortical and mesencephalic astrocytes stimulated by succinate, ADP, or CCCP (Figs. 3.4 A, C, E and B, D, F, respectively), nor for the efﬁciency of mitochondrial energy production determined as RCR (Figs. 3.4 G and H, respectively). With respect to the low RCR values of approximately 1.4-1.8, which suggest that mitochondria were uncoupled, the ADP/O ratio was additionaly calculated, as another parameter of 42 Chapter 3. Steroid hormone effects on mitochondrial respiration

(A) Cortex (B) Mesencephalon 200 200 * 150 * 150

100 100 (% of control) (% of control) mtDNA/ncDNA 50 mtDNA/ncDNA 50

0 0 CE CE

Figure 3.3: Effect of estrogen (E) on the ratio of mitochondrial to nuclear DNA content in cortical (A) and mesencephalic (B) astrocytes. Mitochondrial and nuclear DNA contents were determined as an indirect measure for the amount of mitochondria in a deﬁned number of 10−6 M 17β-estradiol-treated versus control astroglial cells. Total DNA was extracted using DirectPCR lysis reagent (Viagen Biotech, Los Angeles, USA). The suspension has directly been used for PCR of the mitochondrial gene ATP6 and HPRT as a nuclear control gene. The results are expressed as the ratio of ATP6 versus HPRT expression in E-treated cells normalized to control cells set 100%. Statistically signiﬁcant differences (* - P<0.05) between samples were determined using Student’s unpaired t-test.

mitochondrial integrity. ADP/O ratios were in the range 1.1-1.4, which is indicative of coupled mitochondria respiration stimulated with succinate. Similar results were obtained in preliminary experiments applying glutamate/malate as mitochondrial energy substrates (data not shown). However, comparing the succinate-, ADP-, and CCCP-stimulated respiration of mitochondria from cortical with those from mesencephalic astrocytes, an up to two-fold higher activity of mitochondria isolated from cortical astrocytes can be observed.

3.3.4 Short-term estrogen effects on mitochondrial respiratory chain ac-

tivity

To distinguish E-induced long-term effects of several hours of E exposure on mitochondrial protein expression from the short-term effects of several minutes of exposure which could affect mitochondrial activity directly, isolated intact mitochondria were incubated in the presence of 10−6 M E2 for 5 and 30 min prior to the measurement of mitochondrial activity. This revealed a signiﬁcant decrease by 28±9% of succinate- stimulated respiration of cortical astrocyte mitochondria, whereas mesencephalic astrocyte mitochondria showed an increase of respiration by 57±30% after E treatment of isolated mitochondria for 5 min (Figs. 3.5 3.3. Results 43

A, B). The same tendency was observed after 30 min of E2 exposure for succinate-stimulated respiration of astrocyte mitochondria from cortex (decrease by 15±9%) and mesencephalon (increase by 33±16%). ADP-stimulated (state 3) and uncoupled (CCCP-stimulated) mitochondrial respiration did not show any signiﬁcant changes for cortical astrocytes (Figs. 3.5 C, E), whereas for mesencephalic astrocyte mitochondria, a signiﬁcant increase of state 3 mitochondrial activity by 52±8% and 40±14% (Fig. 3.5 D) and an elevation of uncoupled mitochondrial activity by 30±12% and 28±16% were measured after 5 and 30 min of E2 treatment, respectively (Fig. 3.5 F). The RCR values remained unchanged for cortical and mesencephalic astrocyte mitochondria after micromolar E treatment of intact mitochondria (Figs. 3.5 G, H). 44 Chapter 3. Steroid hormone effects on mitochondrial respiration

(A) Succinate (B) Succinate 80 80

60 60

40 40 M/min mg) M/min mg) 20 20 TN ( TN (

0 0 C1224 C1624

Time of E treatment (h) Time of E treatment (h)

60 60

40 40 M/min mg) M/min mg) 20 20 TN ( TN (

0 0 C1224 C1624

Time of E treatment (h) Time of E treatment (h)

(E) CCCP (F) CCCP 120 120

100 100

80 80

60 60 M/min mg) M/min mg) 40 40

TN ( 20 TN ( 20

0 0 C1224 C1624 Time of E treatment (h) Time of E treatment (h)

(G) RCR (H) RCR 3 3

2 2 RCR RCR 1 1

0 0 C1224 C1624

Time of E treatment (h) Time of E treatment (h)

Figure 3.4: Long-term effects of estrogen (E) on the mitochondrial respiration. Intact mitochondria were isolated from primary cortical (graphics on the left side, A, C, E, G) and mesencephalic astrocytes (graphics on the right side, B, D, F, H) after 8 and 12 h of 10−6 M 17β-estradiol treatment and further 4 h incubation on NBM. (A) Succinate respiration of intact mitochondria was measured polarographically at 25◦C in a buffer containing 70 mM sucrose, 220 mM mannitol, 2 mM HEPES, 5 mM MgCl2, 5 mM KH2PO4, 1 mM EDTA. The measurements were conducted after subsequent additions of 20 µM rotenone, 10 mM succinate (A, B), 30 µM ADP (C, D), and 3 µM CCCP (E, F) using an Oxygraph system (Helmut Saur Laborbedarf, Reutlingen, Germany). The activity of mitochondrial respiration is given as Turnover Number (TN, consumed µMO2 / min x mg of total protein). The Respiratory Control Ratio (RCR) value (G, H) was calculated as the ratio of respiration state 4 to state 3. Data represent means ± SEM of three independent experiments of three different cell preparations. Statistically signiﬁcant differences between samples determined using Student’s unpaired t-test were not observed. 3.3. Results 45

(A) Succinate (B) Succinate 80 80

60 60 * 40 40 M/min mg) M/min mg) 20 20 TN ( TN (

0 0 C530 C530

Time of E treatment (h) Time of E treatment (h)

60 60

40 40 M/min mg) M/min mg) 20 20 TN ( TN (

0 0 C530 C530

Time of E treatment (h) Time of E treatment (h)

(E) CCCP (F) CCCP

140 140 120 120

100 100

80 80

M/min mg) 60 M/min mg) 60 40 40 TN ( TN ( 20 20

0 0 C530 C530

Time of E treatment (h) Time of E treatment (h)

(G) RCR (H) RCR 3 3

2 2 RCR RCR 1 1

0 0 C530 C530

Time of E treatment (h) Time of E treatment (h)

Figure 3.5: Short-term effects of estrogen (E) on the mitochondrial respiration. Intact mitochondria were isolated from primary cortical (graphics on the left side, A, C, E, G) and mesencephalic (graphics on the right side, B, D, F, H) astrocytes. Immediately after isolation, mitochondria were incubated in the absence (control) and presence of 10−6 M 17β-estradiol for 5 and 30 min prior to the measurement of mitochondrial activity at 25◦C. Subsequently, the succinate respiration of intact mitochondria was assessed applying the Oxygraph (for details, see Fig. 3.4). The measurements were conducted after subsequent additions of 20 µM rotenone, 10 mM succinate (A, B), 30 µM ADP (C, D), and 3 µM CCCP (E, F). The activity of mitochondrial respiration is given as Turnover Number (TN, consumed µMO2 / min x mg of total protein). The Respiratory Control Ratio (RCR) value (G, H) was calculated as the ratio of respiration state 4 to state 3. Data represent means ± SEM of three independent experiments of three different cell preparations. Statistically signiﬁcant differences (* - P<0.05) between samples were determined using Student’s unpaired t-test

. 46 Chapter 3. Steroid hormone effects on mitochondrial respiration

3.4 Discussion

In the present study, two different (long- and short-term) effects of E can be observed on mitochondria in astroglial cells cultured from mouse brain cortex and mesencephalon. Numerous reports allocate that neurones and non-neuronal cells are target cells for E (Behl, 2002; Kipp et al., 2006; Dhandapani and Brann, 2007; Beyer et al., 2003). Astrocytes express estrogen receptors (ER) α and β in the cortex as well as in the mesencephalon (Beyer et al., 2003). In addition to the predominant nuclear staining, various degrees of cytoplasmic staining for both ERα and ERβ are evident in a variety of tissues and cells including various brain cell types (Li et al., 1997; Simonian and Herbison, 1997; Price and Handa, 2000; Zhang et al., 2002; Mitra et al., 2003; Pawlak et al., 2005). Supporting the direct effects of E on mitochondria, earlier reports have suggested an association of E-binding capacity and E-binding proteins with mitochondria (Grossman et al., 1989; Monje and Boland, 1999; Moats and Ramirez, 2000; Monje et al., 2001; Mitra et al., 2003). Monje and Boland, (1999) and Monje et al., (2001) provided more direct evidence for the presence of ERα and ERβ in mitochondria from rabbit uterus and ovary tissue. More recently, Chen et al. (2004) demonstrated the localisation of ERα and ERβ to mitochondria in MCF7 cells and related the presence of ERs in mitochondria to the mediation of functional effects caused by E treatment. However, there are no reports to date demonstrating a functional role of an E interaction with astrocyte mitochondria. Therefore, the respiratory activity of mitochondria is here investigated, along with the regulation of mitochondrial gene expression after E treatment of astrocytes. These studies revealed that E affects mitochondrial function in a time- and concentration-dependent manner reaching maximal effects at a E concentration of 10−6 µM and after 8 and 12 h for cortical and mesencephalic astrocyte mitochondria, respectively. The long-term effects were reﬂected by elevated mtDNA levels in astrocytes from cortex and mesencephalon accompanied by increased transcript levels of mtDNA-encoded catalytic subunits of the respiratory chain. Both effects indicate an elevation of mitochondrial mass per cell, especially because, under these conditions, astrocyte proliferation remained unaffected by E (data not shown). Furthermore, increases in mtDNA have previously been proposed to serve as a feedback mechanism that compensates for a decline in respiratory chain function (Kim et al., 2004). Besides the long-term effects on mitochondrial gene transcription, here are shown short-term (5, 30 min) effects of micromolar E concentrations on the respiratory chain activity in a brain region-speciﬁc manner. With regard to E-mediated signalling mechanisms, it has been reported (Dhandapani and Brann, 2002; Beyer et al., 2002; Beyer et al., 2003; Kipp et al., 2006) that E is capable of regulating gene expression in cultured neurones and astrocytes by two different mechanisms, i.e., classically through nuclear receptors and non-classically through the stimulation of membrane-E receptors which are coupled to distinct intracellular signal transduction cascades. This is substantiated by the presence of nuclear as well as mitochondrial 3.4. Discussion 47

ERs in astrocytes from both brain regions studied. The observations presented here were well in line with reports about increased levels of mtDNA- and ncDNA-encoded subunits of complex IV in rat brain cells, hepatocytes, human Hep G2 cells etc. (Chen et al., 2004; Bettini and Maggi, 1992; Demonacos et al., 1996; Chen et al., 2003; Hsieh et al., 2006). However, by contrast to these reports, there was no detection of any inhibitory effect of ICI182,780, a pure ER antagonist, on the elevation of mitochondrial gene transcription (Chen et al., 2004; Stirone et al., 2005) in this study. Therefore, the classical nuclear ER pathway is excluded as a regulatory mechanism for mitochondrial genes, proposing the stimulation of a non-classical signalling pathway for cortical and mesencephalic astrocytes. However, a potential involvement of mitochondrial ERs in the regulation of mitochondrial genes needs to be further elucidated along with the exact mechanism of E-stimulated up-regulation of respiratory chain subunit expression. The E concentration used in this study (10−6 M) might at first glance appear to be above physiological relevant E levels. Nevertheless, several previous in vitro studies also show a requirement of higher E concentrations to generate physiological effects (Pawlak et al., 2005; Banerjee et al., 1997; Luconi et al., 1999; Ivanova et al., 2002; Rune et al., 2006; Von Schassen et al., 2006). Intriguingly, the brain itself produces E by means of the aromatase activity (Mukai et al., 2006; Prange-Kiel et al., 2006). This possibly allows higher tissue steroid concentrations in the brain than in the blood plasma (Schumacher et al., 2003). Preliminary data using mass spectrometry to reach highest sensitivity revealed that juvenile mouse brain tissue contains several-fold higher estrogen levels compared to the corresponding plasma values. These findings suggest that the concentration chosen for the in vitro experiments described here may reflect the in vivo situation rather than being non-physiologically high. In addition, transient expression of the E-forming enzyme aromatase at varying concentrations may contribute to region-specific differences in E concentrations affecting differently the expression of E-regulated genes (Prange-Kiel et al., 2006). Besides, some reports indicate an antioxidant role for E, especially when applied at micromolar pharmacological concentrations (Culmsee et al., 1999; Moosmann and Behl, 1999; Sugishita et al., 2003; Nilsen and Brinton, 2004). With regard to the aromatic nature of the E2 molecule, which promotes penetration of lipid membranes and scavenging free radicals, it should be taken into account that protective short-term E effects on mitochondrial respiratory chain function could be paralleled or initiated by this antioxidant pathway. Interestingly, the up-regulation of the catalytic subunit transcription of mitochondrial respiratory chain complexes by long-term exposure of cortical and mesencephalic astrocytes to E were not paralleled by any obvious significant changes of mitochondrial activity or efficiency of energy production (Fig. 3.4). This is in contradiction to numerous observations demonstrating an activation of mitochondrial respiratory activity by E. For example, E treatment of ovariectomized female rats caused an increase of the transcript levels of mtDNA- and ncDNA-encoded subunits of complex IV accompanied by increased enzyme and mitochondrial 48 Chapter 3. Steroid hormone effects on mitochondrial respiration

respiratory chain activity in cerebral blood vessel mitochondria (Stirone et al., 2005). In vitro studies showed similar results (Chen et al., 2003). Nilsen and Brinton (2004) demonstrated an E-mediated increase of the oxidative phosphorylation that would preserve ATP levels in neurones and increase mitochondrial respiration efficiency. These effects are likely to be enhanced by antioxidant effects of estrogen. The possible reasons for the apparent discrepancy include: (i) very low RCR values impeding the detection of changes in the efficiency of mitochondrial respiration; (ii) an up-regulation of mitochondrial gene expression not being paralleled by an elevation of nuclear gene expression of mitochondrial proteins in astrocytes; (iii) the increase in mRNA expression not necessarily being translated into an increase of protein synthesis in astrocytes; and (iv) a different isoform expression pattern of regulatory subunits of the respiratory chain complexes causing a decrease instead of the expected increase of mitochondrial activity. With respect to point (i), the RCR values measured here of approximately 1.4-1.8 were very low. Despite the fact that other studies (Almeida and Medina, 1997; Benard et al., 2006) also reported relatively low RCR values of 1.7 and 2.6 for astrocytes and brain mitochondria stimulated by succinate and pyruvate/malate, respectively, which support the results presented here, RCR values of 9-10 for astrocyte mitochondria have been recently reported (Kristin et al., 2006). Low RCR values are an indicator for at least partially uncoupled mitochondria. There could be several reasons for the discrepancy in RCR values obtained by different groups. Compared to Kristin et al., (2006), the methods for the isolation of mitochondria and the measurement of state 4 respiration, which is directly implicated in the RCR calculation, differed. Furthermore, the accumulation of residual traces of CCCP, although the electrode chamber was washed thoroughly after applications of the uncoupler CCCP, cannot be excluded. As an additional parameter of the integrity of isolated mitochondria, the ADP/O ratio was determined. Calculated ADP/O values for mitochondria isolated from estrogen-treated and untreated astrocytes were in the range 1.1-1.4. Similar values were reported in several other studies for a coupled mitochondrial respiration stimulated by succinate (Hafner et al., 1998; Gnaiger et al., 2000). Interestingly there weren’t any significant differences for the ADP/O ratios of mitochondria from treated and untreated cells (data not shown). To further substantiate the present data, intact cell respiration studies were performed. The preliminary data showed a stimulating effect of long-term estrogen treatment of astrocytes on mitochondrial activity, which will be subject to further investigations. To rule out reason (ii), the analysis of the nuclear-encoded COX subunit IV gene expression reflected an up-regulation of the transcript (data not shown, incorporated into another submitted manuscript). Furthermore, as the measured mitochondrial activity was normalized to the whole mitochondrial protein content, an increase of mitochondrial activity due to an increased expression of specific genes might be camouflaged by an increased overall mitochondrial protein content and would not be evident from the normalized respiration rates. Supporting this assumption, an increase in mtDNA was detected, which could be an indication for the 3.4. Discussion 49 mitochondrial protein content being increased along with the amount of mitochondria per cell. That would suggest a higher mitochondrial activity due to a higher number of mitochondria and a larger mitochondrial protein amount in E-treated versus untreated cells. This increase in respiratory activity would lead to a higher ATP production that could meet high energy consuming processes in astrocytes, especially when cells undergo stress conditions. This might also explain why cells treated with E prior to a toxic stimulus (e.g. MPP+) show a higher survival rate than cells not being treated with E (Bains et al., 2007). In the same context, it is shown here an up to two-fold higher succinate-, ADP-, and CCCP-stimulated mitochondrial respiration for cortical astrocytes compared with mesencephalic astrocytes that occurred independently of the E treatment. This suggests that cortical astrocytes are more protected from energy deprivation than astroglia from mesencephalon by sustaining higher levels of energy production. Nevertheless, mitochondrial respiration coupling is dependent on the membrane conditions of the isolated mitochondria, and another attempt was made to prove the intactness of the isolated mitochondria, using scanning electron microscopy, SEM (many thanks to D. Alves) and the pictures produced show well kept membranes, inclusively very clear inner membrane invaginations (Fig. 3.6), which show that a fair degree of intactness of the mitochondria is kept. This is supported by the proposed 3D model of mitochondrial imaging of Manella (Manella, 2000). Even though isolated mitochondria do not compare to mitochondria in physiological conditions, their form resembles one of the mitochondria forms depicted by others, described as smaller, elliptical mitochondria with randomly disposed cristae appearing to be floating free in the inner compartment (Ambu et al., 2000). Isolated mitochondria have been described since decades as small and spherical to elliptical. During the extraction, which obliges to the breakage of cells and necessarily, breakage of the mitochondrial net, the method cannot preserve mitochondria in their original physiological structure.

Figure 3.6: Isolated mitochondria scanning electron microscopy, SEM. Mitochondria were isolated according to the description in Chapter 2, section 2.10. Cryofraction SEM was performed by D. Alves, many thanks. 50 Chapter 3. Steroid hormone effects on mitochondrial respiration

To study the short-term pathways of E signalling, 1 µM E2 was applied for 5 and 30 min on intact mitochondria isolated from mouse brain cortical and mesencephalic astroglial cells. This short-term application of E exerted a brain region-specific increase of succinate-stimulated oxygen consumption of mitochondria from mesencephalic, but not from cortical astrocytes which showed a decrease of mitochondrial respiration (Fig. 3.5). The ADP- and CCCP-stimulated respiration in the presence of succinate as substrate also demonstrated an increase in astrocyte mitochondria from mesencephalon, but no changes in astrocyte mitochondria activity from the cortex. Interestingly, the RCR values remained unchanged for mitochondria from astrocytes of both brain regions. This indicates that short-term E treatment of astrocytes does not influence the efficiency of mitochondrial energy production, but affects the activity of proton-translocating complexes of the respiratory chain. As the short-term E treatment of intact mitochondria leads neither to an up-regulation of the expression of ATP synthase subunits (data not shown), nor does E stimulate ATP synthase activity in astrocytes (Kipp and Ramirez 2001), no compensation for a higher proton-translocating activity of complexes I, III, and IV by a faster consumption of the increased proton gradient can occur. Zheng and Ramirez (1999) support this aspect by showing that E at micromolar concentrations inhibits F0F1-ATPase activity by binding to the oligomycin-sensitivity conferring protein (OSCP), a subunit of the ATP synthase that is required for coupling the proton gradient to ATP synthesis. Thus, the increase of the succinate respiration in mitochondria from mesencephalic astrocytes is an indicator of a hyperpolarisation of the mitochondrial membrane potential, which, in turn, causes a higher production of reactive oxygen species (ROS) by the complexes III and I of the respiratory chain (Lee et al., 2001; Kadenbach et al., 2004). As indicated by the ADP-stimulated respiratory chain activity in the presence of succinate, this effect is even more pronounced under conditions when cellular ATP levels are low and energy production becomes crucial for cell survival. Therefore, the observed elevation of respiratory chain activity might be concomitant with increased deleterious ROS production, specifically in mesencephalic astrocytes. This would indicate that a short-term E treatment is less effective for improvement of mitochondrial function and survival of astrocytes in the mesencephalon, whereas E protects cortical astrocytes by decreasing mitochondrial respiratory chain activity to a level where ROS production does not occur. It might be that a decrease of succinate-stimulated respiration paralleled by a decrease of oxygen consumption is a mechanism to prevent ROS production. This may, at least in part, explain why astrocytes from mesencephalon are more prone to be subject to cell death induced by toxic and stress factors than astrocytes from cortex. However, the extent to which each of these mechanisms contributes to the overall protective defence state initiated by E remains to be elucidated further. In summary, this study shows that E influences mitochondrial physiology independently of the nuclear ERs in a long- and short-term manner. There weren’t any significant long- or short-term effects of E on the 3.4. Discussion 51 efficiency of mitochondrial energy production. However, after long-term E treatment of astrocytes from both brain regions, an elevated mitochondrial gene expression and mtDNA content was observed, which can be considered as a measure of the amount of mitochondria per cell. This is presumably accompanied by an increase in the activity of proton-translocating and proton-consuming respiratory chain complexes leading to increased ATP levels but lower ROS production in astrocytes. In turn, this would support the survival of astrocytes undergoing stress and degenerative processes by providing them with higher energy levels. The results described here emphasize the neuroprotective role of E in long-term treatment of astrocytes in different brain regions and could represent an important indirect mechanism by that estrogen protects neurons from cell death under neurotoxic conditions. Interestingly, cortical astrocytes, independent of estrogen treatment, showed higher mitochondrial respiration rates than mesencephalic astrocytes. This indicates a higher level of protection of astrocytes from brain cortex compared with mesencephalon. Short- term E treatment showed different effects on the respiratory chain activity in astrocytes from cortex and mesencephalon. This could, at least in part, explain why cortical astrocytes are better preserved from stress conditions and degenerative processes than astrocytes in the mesencephalon during short-term E exposure. However, to better understand the in vivo role of E, the time and concentration range of E would need to be measured under physiological and pathological conditions. Nevertheless, these studies indicate that a short-term application of micromolar E concentrations appears to be protective for cortical astrocytes. 52 Chapter 3. Steroid hormone effects on mitochondrial respiration Chapter 4

Steroid hormone effects on mitochondrial fusion and ﬁssion

Gender-speciﬁc regulation of mitochondrial fusion and ﬁssion gene transcription and viability of cortical astrocytes by steroid hormones2

2Adapted from: Arnold, Wright Araujo and Beyer 2008. Journal of Molecular Endocrinology 41, 289-300.

4.1. Introduction 55

Abstract

Astroglia and steroid hormones such as estrogen and progesterone regulate cell growth, function, and protection in the central nervous system (CNS). It appears that astrocytes and steroids act in concert to promote cell survival under pathological conditions. With respect to the role of mitochondrial fusion and fission in energy metabolism, apoptosis, and proliferation, astrocyte mitochondria resemble a perfect intracellular target for steroids to modulate these processes, thereby promoting cell vitality after damage. Here focus was put on the effects of estrogen and progesterone on cell viability in comparison with mitochondrial fusion and fission gene transcription in primary cortical astrocytes from female and male mouse brains. Estrogen- and progesterone-treated female astrocytes demonstrated an increase in cell number and proliferation marker accompanied by an upregulation of fusion and fission gene transcription, which were apparently balancing pro- and anti-apoptotic processes. On the other hand, male astrocytes exhibited no change in cell number after estrogen treatment, but a decrease after progesterone administration. This could be the consequence of stimulated apoptosis in male astrocytes by both steroids, which was counterbalanced by an increased proliferation in the presence of estrogen, whereas it was strengthened in the presence of progesterone. Supportively, estrogen promoted and progesterone decreased the transcription of fusion and fission genes. It seems that estrogen and progesterone affect mitochondrial fusion and fission gene transcription in cortical astrocytes in a gender-specific way, thereby influencing mitochondrial function differently in both genders. Thus, interaction of sex steroids with mitochondria may represent one possible cause for gender differences in cellular pathology in the CNS.

4.1 Introduction

Estrogen (E) and progesterone (P) exert their well-known neuroprotective effects in the CNS by initiating a multitude of cellular mechanisms which are often mediated by astrocytes (Azcoitia et al., 2002; Arnold, 2005; Pawlak et al., 2005; Garcia-Segura and Melcangi, 2006; Leonelli et al., 2007). Astrocyte mitochondria represent an interesting intracellular target organelle for steroid hormones, since they provide ample of services to neural cells including energy production, calcium buffering, and regulation of apoptosis. These processes are also known to be modulated by gonadal steroid hormones (Nilsen and Brinton, 2004; Kipp et al., 2006). However, mitochondria-mediated mechanisms in steroid-dependent protection are not fully understood. In very recent studies, it has been shown that E promotes the expression of mitochondria-encoded subunits of the respiratory chain, thereby influencing mitochondrial respiratory activity and enhancing functional efficiency (Araujo et al., 2008; Irwin et al., 2008). Mitochondrial functions are not only affected by regulation 56 Chapter 4. Steroid hormone effects on mitochondrial fusion and fission

of gene expression and activity of the respiratory chain (Zheng and Ramirez, 1999; Araujo et al., 2008; Irwin et al., 2008), but also by the morphology of these organelles. Two sets of proteins and their relative activities are involved in shaping the morphology of mitochondria. While mitochondrial GTPases, mitofusin 1 and 2 (Mfn1, Mfn2) and optic atrophy 1 homolog (Opa1), are essential for fusion processes, fission 1 (mitochondrial outer membrane) homolog (Fis1) and dynamin-related protein 1 (Drp1) are essential fission proteins (Santel and Fuller, 2001; Rojo et al., 2002; Chen et al., 2003; Olichon et al., 2003; Griparic et al., 2004; Cipolat et al., 2004; Chen and Chan, 2005). Given the highly ordered arrangement of mitochondrial networks in some cell types (Bakeeva et al., 1981), it is likely that mitochondrial shape indeed has important consequences for mitochondrial function. Under physiological conditions, mitochondria are characterized by a dynamic equilibrium of fusion and fission, whereas an imbalanced expression of fusion versus fission proteins leads to a disturbance of this equilibrium and causes dramatic changes in mitochondrial morphology and function (Yaffe, 1999; Karbowski and Youle, 2003). In humans, mutations in Mfn2 cause Charcot-Marie-Tooth neuropathy type 2A (Zuchner et al., 2004; Kijima et al., 2005). Mutations in Opa1 lead to autosomal dominant optic atrophy (Alexander et al., 2000; Delettre et al., 2000). A down-regulation of fusion proteins together with an impairment of mitochondrial fusion processes or an over-expression of fission proteins causes fragmentation of the tubular mitochondrial network (James et al., 2003; Olichon et al., 2003; Yoon et al., 2003; Lee et al., 2004; Stojanovski et al., 2004; Chen et al., 2005; Detmer and Chan, 2007). As a consequence, an increased susceptibility to cell death and apoptosis occurred. Knock-out mice of either Mfn1 or Mfn2 are both embryonic lethal demonstrating the essential role of mitochondrial fusion for embryonal development and cell viability (Chen et al., 2003). In contrast, over-expression of the two mitofusin proteins together provides some protection against different apoptotic stimuli (Sugioka et al., 2004). On the other hand, mitochondrial fission proteins (Fis1 and Drp1) and processes are essential for cell division as well as for programmed cell death and decreased cell survivability (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Stojanovski et al., 2004). In Caenorhabditis elegans, an over-expression of fission protein Drp1 increased the number of mitochondrial divisions (Labrousse et al., 1999). Fis1 over-expression also caused mitochondrial fragmentation and aggregation (Stojanovski et al., 2004). On the contrary, a disruption of fission by over- expressing a mutant form of Drp1 or over-expression of fusion proteins provoked an elongation and formation of interconnected mitochondrial network due to enhanced mitochondria fusion (Smirnova et al., 2001; Santel et al., 2003). Considering the severe consequences of imbalanced mitochondrial fusion and fission for neurodegeneration, the effect of E and P on the transcription level of mitochondrial fusion and fission genes in cortical astrocytes in vitro was studied. Astrocytes were elected, for being an outstanding cell type in the CNS supporting neuronal viability and function. It is apparent that neurodegenerative processes and their steroid responsiveness often occur in a gender-specific way (Miller et al., 1998; Roof and Hall, 2000; 4.2. Materials and Methods 57

Dluzen and McDermott, 2004; Kenchappa et al., 2004; Marchetti et al., 2005). Therefore, female and male astroglia were analysed separately.

4.2 Materials and Methods

For information on the materials and methods used in this study, please refer to Chapter 2. For information on the author’s contributions please refer to Contributions. In this investigation, cells were treated with 10−7 M 17β-estradiol (E2, generally referred as E) or progesterone (P) for 24 h.

4.3 Results

4.3.1 Gender-speciﬁc effects of steroid hormones on cell viability and

apoptosis

In a first attempt, the gender-specific effects of E- and P-treatment on proliferation and apoptotic death of cortical astrocytes were analysed. Female and male astrocyte cultures were treated with 10−7M E or P in the absence or presence of the appropriate receptor inhibitor, i.e. ICI and RTI as inhibitors for ER and PR, respectively. Viable, apoptotic, and necrotic astrocytes were distinguished by fluorescent labelling of cell nuclei with Hoechst 33342. Before and after treatments, no significant changes in the number of necrotic cells were detected (0-2 cells per total number of approx. 100 cells; data not shown). This finding was supported by experimental data obtained from propidium iodide staining which labels necrotic cells (data not shown). Due to the fact of a constant number of necrotic astrocytes, focus was put on the determination of the number of viable and apoptotic astrocytes before and after hormone application. Treatment of cortical astrocytes from female brains with E or P for 24 h induced a significant increase in the number of Hoechst-stained viable cells by approximately 50% (Fig. 4.1 A). Co-application of P and PR inhibitor suppressed this effect, whereas the ER inhibitor did not affect the E-mediated effect (Fig. 4.1 A). Independently of the steroid hormone applied, the number of apoptotic cells was not significantly changed in female astrocyte cultures (Fig. 4.1 C). In contrast, the application of E did not exert any effect on the number of viable male cells, whereas P caused a decrease (Fig. 4.1 B). The latter effect was counteracted by RTI. Compared to female astrocytes, opposite effects were observed concerning apoptosis of male astrocytes. Both hormones provoked a significant 3- to 4-fold induction which was partially suppressed by ICI and completely antagonized by RTI (Fig. 4.1 D). 58 Chapter 4. Steroid hormone effects on mitochondrial fusion and fission

(A) Female Viability (B) Male Viability 200 * 200 * * * *

150 # 150 # 100 100 * Cell number Cell number (% of control) (% of control) 50 50

0 0 C E E+ICI P P+RTI C E E+ICI P P+RTI

(C) 500 Female Apoptosis (D) 500 Male Apoptosis * 400 * 400 * * 300 300 # 200 200 # (% of control) 100 (% of control) # Apoptotic cell death

Apoptotic cell death 100

0 C E E+ICI P P+RTI 0 C E E+ICI P P+RTI

Figure 4.1: Effect of estrogen (E) and progesterone (P) on the viability (A, B) and apoptotic cell death (C, D) of female and male cortical astrocytes. Primary astrocyte cultures from female (A, C) and male (B, D) Balb/c mice were treated with 10−7 M E or P or co-treated with 10−6 M of the appropriate inhibitors of steroid receptors ICI 182.780 (ICI for ER) and RTI 3021-022 (RTI for PR) for 24 h. The number of viable astrocytes (A, B) and apoptotic cells (C, D) were scored after staining astrocyte nuclei with Hoechst 33342. Data represent the percentage of the ratio of sample cell numbers to untreated control cell numbers referred to as 100% as means ± SEM of triplicates of ﬁve independent experiments. *p<0.05, **p<0.01, ***p<0.001, E-/P-treated samples versus untreated controls; #p<0.05, ##p<0.01, inhibitor plus E-/P-treated samples versus E-/P-treated samples.

In a next step, the transcription levels of pro-, anti-apoptotic, and proliferation markers (Bax, Bcl2, and PCNA, respectively) where measured, to assess if changes in the viable and apoptotic cell numbers resemble changes in the marker gene expression. Primers used for RT-PCR ampliﬁcation (Chapter 2, section 2.7) were designed as intron over-spanning sequences except for 18S rRNA. Only single bands at correct molecular weight (Chapter 2, Table 2.3) were observed for the PCR products of the housekeeping gene HPRT and the genes of interest in an agarose gel. This indicates that the isolated RNA was free of genomic DNA impurities. In female cortical astrocytes, E exerted no signiﬁcant effect on the transcription of anti- and pro-apoptotic marker genes, whereas P induced an approximately 6-fold up-regulation of Bcl2 and Bax transcription 4.3. Results 59

(A) Female Apoptosis (B) Male Apoptosis 1000 1000

800 * * 800 * * * * 600 Bcl2 600 * Bcl2 Bax * Bax # # * 400 400 (% of control) (% of control) # # mRNA expression

mRNA expression # # 200 200 # # * # * * * * 0 0 C E E+ICI P P+RTI C E E+ICI P P+RTI

600 # 600 * # PCNA PCNA * 400 400 # (% of control) * (% of control) # mRNA expression 200 mRNA expression 200 # * * 0 0 C E E+ICI P P+RTI C E E+ICI P P+RTI

Figure 4.2: Effect of estrogen (E) and progesterone (P) on the transcription of apoptosis and proliferation markers in cortical astrocytes. Cultured astrocytes from female (A,C) and male (B, D) brains were treated with 10−7 M E or P in the absence or presence of 10−6 M ICI or RTI for 24 h. Specific quantification of Bcl-2 (A, B; lighter columns) and Bax transcripts (A, B; darker columns) as well as PCNA transcripts (C, D) was performed by RT-PCR. Data were normalized to 18S rRNA and HPRT. Untreated astrocytes from the same cell culture preparation served as negative control and were set 100%. Data represent the percentage of control expressed as means ± SEM of triplicates of five independent experiments. *p<0.05, **p<0.01, ***p<0.001, E-/P-treated samples versus untreated controls; #p<0.05, ##p<0.01, ###p<0.001, inhibitor plus E-P-treated samples versus E-P-treated samples.

(Fig. 4.2 A), which was partially down-regulated by RTI. Unlike females, male astrocytes responded to E application with an increased transcription of the apoptotic marker Bax and a decrease of anti-apoptotic Bcl2 (Fig. 4.2 B). Treatment of male astrocytes with P caused a decreased transcription of both marker genes. The P effect was completely inhibited by RTI, whereas ICI reversed only the E effect on Bcl2 transcription. In female astrocytes, PCNA transcription was 5- to 8-fold up-regulated, irrespective of the hormone (Fig. 4.2 C). Once more, ICI did not prevent the E effect, but RTI partially abolished the P effect. PCNA transcription levels in male astrocytes were stimulated by E, but signiﬁcantly down-regulated by P (Fig. 4.2 D). ICI did not interfere with E’s effect while RTI antagonized P’s effect. 60 Chapter 4. Steroid hormone effects on mitochondrial fusion and ﬁssion

4.3.2 Gender-speciﬁc effects of steroid hormones on fusion and ﬁssion

gene transcription

After treatment of female and male cortical astrocytes with E or P, the transcription levels of Mfn1, Mfn2, Fis1, and Drp1 were determined by RT-PCR (Fig. 4.3). In female astrocytes, E stimulated Mfn2 but not Mfn1 transcription, whereas P exposure caused an increase of both fusion transcripts (Fig. 4.3 A). In male astrocytes, E decreased Mfn1 but increased Mfn2 transcription, whereas P down-regulated both fusion gene transcript levels (Fig. 4.3 B). Regarding fission gene transcription, in female astrocytes, E and P stimulated both Drp1 and Fis1 (Fig. 4.3 C). In male astrocytes, E also increased Drp1 and Fis1 transcription, but P decreased the transcription of both fission genes (Fig. 4.3 D). Looking at the receptor inhibitors effects, in female astrocytes, ICI did not antagonize the E effects whereas RTI partially blocked the P effect (Figs. 4.3 A, C). In male astrocytes, ICI blocked only the E effect on Mfn1 transcription, and RTI reversed the P effect on the transcription of all fusion and fission genes (Figs. 4.3 B, D). Correlation analyses of Mfn1 with Bcl2 transcripts and Drp1 with PCNA for female and male astrocytes are given in Figures 4.4 A-D. It becomes evident that Mfn1 correlates well with Bcl2 and Drp1 with PCNA for both steroid hormones and both genders (Figs. 4.4 A, B and C, D, respectively) by showing similar transcription levels. Several observations concerning gender-specific differences of the hormonal regulation deserve attention: (i) In females, P increased Mfn1, Bcl2, Drp1 and PCNA transcript levels (Figs. 4.4 A, C); in males, gene transcription was decreased (Figs. 4.4 B, D). (ii) Treatment of astrocytes with E caused higher transcript levels of Mfn1, Bcl2, Drp1 and PCNA in females than in males (Figs. 4.4 A, C versus 4.4 B, D). In females, Mfn1 and Bcl2 are not changed by E, but decreased in males. Drp1 and PCNA are increased by E in females to a greater extent than in males. A possible correlation of the ratios of Drp1/Fis1 transcription and PCNA/Bax transcription with viable/apoptotic cell number is shown in figures 4.5 A and B for female and male astrocytes, respectively. When comparing all three ratios, the following gender-specific differences become clear: (i) The PCNA/Bax transcript level and viable/apoptotic cell ratios are > 1 for female astrocytes treated with E or P, whereas these levels are < 1 for the same treatments in male astrocytes; (ii) The Drp1/Fis1 ratios are decreased in P- treated versus E-treated male astrocytes, but slightly increased in females. Generally, for both genders: (i) the ratios of PCNA/Bax are decreased in P-treated compared to E-treated cells; (ii) the ratios of viable/apoptotic cell numbers do not differ between the two steroid treatments. 4.3. Results 61

(A) Female Fusion (B) Male Fusion 1200 1200

1000 1000

800 800 Mfn1 Mfn1 * * Mfn2 Mfn2 600 * * * 600 * # 400 # 400

# (% of control) (% of control)

mRNA expression * mRNA expression # # 200 200 # # # * * * * * 0 0 C E E+ICI P P+RTI C E E+ICI P P+RTI

(C) (D) 1200 Female Fission 1200 Male Fission * 1000 * * 1000 * * 800 800 Drp1 Drp1 * * Fis1 Fis1 600 * # # 600 * # * 400 400 (% of control)

(% of control) * mRNA expression

mRNA expression # # 200 200 # # * * * 0 0 C E E+ICI P P+RTI C E E+ICI P P+RTI

Figure 4.3: Effect of estrogen (E) and progesterone (P) on the transcription of mitochondrial fusion and fission genes in cortical astrocytes from female and male brains. Primary cortical astrocytes from female (A, C) and male (B, D) mouse brains were incubated in the presence of 10−7 M E or P for 24 h. Antagonists of estrogen and progesterone receptors (ICI and RTI, respectively) were added simultaneously to the respective steroid hormone treatment. Specific quantification of fusion genes, Mfn1 (A, B; lighter columns) and Mfn2 (A, B; darker columns), and fission genes, Drp1 (C, D; lighter columns) and Fis1 (C, D; darker columns) was performed by RT-PCR. Data were normalized to 18S rRNA and HPRT and related to the transcription levels in untreated control astrocytes set 100%. Data represent the percentage of control as means ± SEM of triplicates of five independent experiments. *p<0.05, **p<0.01, ***p<0.001, E-/P-treated samples versus untreated controls; #p<0.05, ##p<0.01, inhibitor plus E-/P-treated samples versus E-/P-treated samples. 62 Chapter 4. Steroid hormone effects on mitochondrial fusion and fission

(A) Female (B) 1000 1000 Male

800 800

600 Mfn1 600 Mfn1 Bcl2 Bcl2

400 400 (% of control) (% of control) mRNA expression mRNA expression 200 200

0 0 CEP CEP

800 800

600 Drp1 600 Drp1 Pcna Pcna

400 400 (% of control) (% of control) mRNA expression 200 mRNA expression 200

0 0 CEP CEP

Figure 4.4: Correlation analysis between Mfn1 and Bcl2 gene transcription (A, B) as well as Drp1 and PCNA gene transcription (C, D) in female (A, C) and male (B, D) astrocytes. Data used for this graph derived from Figures 4.2-4.3. Gene transcription levels for Mfn1 (A, B; lighter columns) and Bcl2 (A, B; darker columns) as well as for Drp1 (C, D; lighter columns) and PCNA (C, D; darker columns) after E/P-treatment in the absence or presence of ICI/RTI for 24 h are presented as % of control set 100%. Note that P exerted opposite effects in female and male astrocytes. Treatment of astrocytes with E caused higher transcription levels of all four genes in females (A, C) compared to males (B, D). 4.4. Discussion 63

(A) (B) 2 Female 2 Male Drp1/Fis1 Pcna/Bax Drp1/Fis1 Viab/Apop Pcna/Bax Viab/Apop

1 1 Ratios (a.u.) Ratios (a.u.)

0 0 CEP CEP

Figure 4.5: Correlation analysis between the ratios of Drp1/Fis1 transcription levels (Drp1/Fis1), PCNA/Bax transcription levels (PCNA/Bax), and viable/apoptotic cell numbers (Viab/Apop) for female (A) and male (B) astrocytes. Gender-speciﬁc differences become obvious when comparing the ratios of Drp1 to Fis1 (lighter grey columns) with those of PCNA to Bax (darker grey columns) or with the ratios of viable to apoptotic cell number (darker columns). Note that the PCNA/Bax transcripts and viable/apoptotic cell ratios are > 1 for female astrocytes treated with E or P (A), whereas these levels are < 1 for E-/P-treated male astrocytes (B). For both, female (A) and male (B) astrocytes, the ratios of PCNA/Bax are decreased in P-treated compared with E-treated cells, whereas the Drp1/Fis1 ratios are also decreased in P-treated versus E-treated male astrocytes, but marginally increased in females. The ratios of viable/apoptotic cell numbers do not differ between the two steroid treatments neither in females (A) nor in males (B). a.u., arbitrary units.

4.4 Discussion

Astrocytes are essential for the support of neurons in the CNS (Pellerin et al., 2007; Schousboe et al., 2007). They are implicated in the regulation of growth, cell proliferation, and neuroprotection in the brain (Beyer, 1999; Beyer et al., 2003; Garcia-Segura et al., 2003; Kajta and Beyer, 2003; Kipp et al., 2006; McCarthy, 2008). In the past years, the neuroprotective aspects of E and P effects in the CNS have been studied using in vitro and in vivo approaches. Besides direct interactions of E and P with neurons (Kipp et al., 2006; Stein et al., 2008), there are indications that neurons may be indirectly protected through other cells. However, estrogen has consistently been demonstrated to be protective in vivo and in organotypical cultures (Dhapandani and Brann, 2002; Simpkins et al., 2005). This suggests the involvement of another non-neuronal brain cell type exerting a protective function. Of these, astrocytes have the greatest potential for a possible involvement in neuroprotection. Astrocytes are the most abundant type of glial cells in the brain and are located juxtaposition to neurons, outnumbering them by a 10:1 ratio in the cortex. This cell type might most probably have a 64 Chapter 4. Steroid hormone effects on mitochondrial fusion and ﬁssion

critical role in steroid-mediated protection of neurons in the brain, since ablation of astrocytes in vivo results in a significant decrease of neuronal survival (Cui et al., 2001). Astrocytes are also well-described target cells for E and P, expressing the corresponding nuclear steroid receptors (Jung-Testas et al., 1992; Santagati et al., 1994; Beyer, 1999; Quadros et al., 2007). Confirming previously published data (Pawlak et al., 2005), transcripts for ERα, ERβ, and PR were observed in cortical astrocytes from both genders. Besides the above described supportive function, astrocytes play a major role for the energy homeostasis of the brain, pointing at astrocyte mitochondria as an interesting compartment involved in neuroprotection. However, there is only limited information available on interactions between sex steroids and mitochondria. Recent studies have shown that E promotes the expression of mitochondria-encoded subunits of the respiratory chain, thereby influencing mitochondrial respiratory activity and enhancing functional efficiency (Araujo et al., 2008; Irwin et al., 2008). Mitochondrial functions are not only affected by regulation of nuclear and mitochondrial gene expression and activity of the respiratory chain (Zheng and Ramirez, 1999; Araujo et al., 2008; Irwin et al., 2008), but also by the morphology of these organelles. It is noteworthy that E influences mitochondria not only through its nuclear receptors, but acts also directly by binding of E to mitochondria in vitro and in situ (Yang et al., 2004; Stirone et al., 2005; Yager and Chen, 2007; Milner et al., 2008). Interestingly, some effects, in particular those exerted by E but not by P, were not subject to inhibition by classical nuclear ER antagonists. Proliferation and fission gene transcription were obviously mediated by E via non-classical mechanisms in male and female astrocytes. The context of mitochondrial fission and cell division additionally supports the role of one for the other. Such a proliferative E action that is intracellularly transmitted through, for instance, the MAP kinase system has already been reported for non-neural cells (Bouskine et al., 2008). This is in line with a previous observation of a rapid activation of ERK in astroglia (Ivanova et al., 2001). Intriguingly, a recent report suggests a mitochondrial ER-mediated pathway in stimulation of cell division (Chen et al., 2008). Besides proliferation, the ER antagonist ICI did not abolish any of the E effects on female astrocytes studied, whereas in male astrocytes, ICI prevented the E effects by decreasing apoptotic cell death and increasing Bcl2 and Mfn1 transcription. This indicates gender-specific E actions in the context of anti- apoptotic processes and an involvement of Mfn1 therein. It is known that a wide range of neurodegenerative disorders show distinct sex-specific differences in course and incidence. Thus, a female prevalence is known for Alzheimer’s disease and multiple sclerosis, whereas a male prevalence was observed for Parkinson’s disease and stroke. In particular, intact adult female rodents sustain lower mortality and less neuronal damage as compared to age-matched male rodents following middle cerebral artery occlusion (Alkayed et al., 1998). Parkinson’s disease reveals a clear gender disparity (Shulman, 2007). Interestingly, estrogen appears to exert opposite effects in females and males in an experimental animal model for Parkinson’s disease being neuroprotective only in females (Gillies et al., 2004). Estrogen replacement therapy leads to 4.4. Discussion 65 an improvement of Parkinsonian symptoms of women in the menopause (Shulman, 2007). Both disorders, stroke and Parkinson’s disease, are characterized by a disruption of mitochondrial function (Singh et al., 2006; Thomas and Beal, 2007; Poole et al., 2008; Schapira, 2008). This raised the question whether steroid-mediated regulation of mitochondrial morphology and function could be involved in the gender- specificity of neurodegenerative and neuroprotective processes. Besides functional characteristics, such as energy production, the morphology of mitochondria including the expression of fusion and fission genes are indicators of mitochondrial vitality. Increased fusion processes cause an elongation of mitochondria, whereas a raise in fission leads to mitochondrial fragmentation (Chan, 2007). Generally, an imbalance of mitochondrial fusion and fission causes mitochondrial and cellular dysfunction indicating that fusion and fission processes are essentially linked to cell function and protection/apoptosis. In an attempt to better understand the protective action of E and P, focus was put on fusion and fission processes in astrocytes and an attempt was made to correlate their regulation with cell viability and apoptosis. Cultured cortical astrocytes were treated with E concentrations known to be effective in the brain (Luconi et al., 1999; Ivanova et al., 2002; Pawlak et al., 2005; Rune et al., 2006; von Schassen et al., 2006; Araujo et al., 2008). Under basal (physiological) conditions, cultured cortical astrocytes from males and females did not differ in their characteristics for mitochondria morphology and cell viability. E appeared to stimulate proliferation in females, whereas P facilitates cell death in males. The age-related drop of E levels in females may, thus, play a role for the vulnerability of cortical neurons in late-onset neurodegenerative disorders, such as Alzheimer’s disease, due to reduced astrocyte stimulation and survival. We might speculate that the prevalence for Alzheimer’s disease observed for females is causally linked to decreased E levels and astrocyte activation. Notwithstanding the importance of our observations, we did not provide a pathological model with our experimental set-up. For E-treated cortical astrocytes of both genders, a correlation between Mfn1 and anti-apoptotic marker Bcl2 transcription as well as Drp1 and proliferation marker PCNA transcription was made. This indicates a role of Mfn1 in anti-apoptotic processes and Drp1 in cell proliferation and is in agreement with the generally accepted role of fusion genes in anti-apoptotic processes and fission genes in proliferation and apoptosis (Yaffe, 1999; James et al., 2003; Karbowski and Youle, 2003; Olichon et al., 2003; Yoon et al., 2003; Lee et al., 2004; Stojanovski et al., 2004; Chen and Chan, 2005; Chen et al., 2005; Chan, 2007). These studies also revealed gender differences in the transcription of Mfn1 and Drp1 showing higher levels in females which were accompanied by lower apoptotic cell death and elevated number of viable cells. The role of Drp1 in mitochondrial fragmentation was previously supported by the discoveries that Drp1 mutations blocked and Drp1 over-expression increased mitochondrial proliferation (Bleazard et al., 1999; Labrousse et al., 1999). Furthermore, Drp1 was co-localized with constrictions in mitochondria that can coincide with actual division events (Bleazard et al., 1999; Labrousse et al., 1999). During mitosis, Drp1 is involved in fission 66 Chapter 4. Steroid hormone effects on mitochondrial fusion and fission

of mitochondria, a phenomenon that might facilitate the partitioning of fragmented mitochondria to daughter cells during cytokinesis (Taguchi et al., 2007). This makes Drp1 a factor of controlling mitochondrial division and a potential candidate for steroid-mediated regulation of cell proliferation, although the involvement of Drp1 in apoptosis cannot be completely ruled out (Frank et al., 2001). The Drp1/Fis1 ratio was deliberately chosen to describe proliferative and apoptotic events in astrocyte cultures. This enables the discrimination of both opposite mechanisms in cultures of equal numbers of viable cells. A constant number of cells does not allow to conclude whether there is an effect or no effect on both processes by hormones. Comparing the anti- and pro-apoptotic marker gene transcription with apoptotic cell death, it becomes evident that an increase or the absence of differences in Bax compared to Bcl2 is related to increased or unchanged apoptosis level, respectively. The apparent discrepancy between the effect of P on male astrocytes showing a reduced Bax transcription and increased apoptotic cell death can be clarified by determination of the Bax/Bcl2 ratio, which is elevated. This points at an increase of apoptotic cell death, although the individual values of Bax and Bcl2 in males are below the basal levels found in controls. A clear gender-specific difference in the apoptosis rate of steroid-treated astrocytes was observed and could be correlated with fusion and fission gene transcription. Female astrocytes showed a mainly balanced increase of both, fusion and fission gene transcription. This can be considered as a reason for a balanced transcription of anti- and pro-apoptotic marker genes. On the contrary, fusion and fission gene transcription in males was imbalanced. Consequently, this imbalance caused a disturbance of the equilibrium between anti- and pro-apoptotic marker genes, thus promoting apoptotic processes specifically in male astrocytes treated with gonadal steroids. A higher level of fusion gene transcription supports the increased astrocyte viability in females (Zuchner et al., 2004; Kijima et al., 2005). This could be partially due to the role of mitochondrial fusion processes in mixing mitochondrial DNA copies, thereby keeping high loads of mutated mitochondrial DNA below the threshold of developing diseases (Nakada et al., 2001; Chen and Chan, 2005; Chan, 2007). The increased apoptosis in male astrocyte cultures could only be compensated by an elevation of proliferation resulting in an unchanged level of viable cells after steroid hormone treatment. Indeed, E- treated male astrocytes exhibited a higher PCNA/Bax ratio than P-treated cells. A ratio of PCNA/Bax > 1 together with an increased proliferation marker transcription at constant levels of apoptosis in female astrocytes after steroid treatment caused an elevated number of viable cells. In contrast, male astrocytes demonstrated an increased proliferation after E treatment which compensated for the increased apoptosis, thereby demonstrating a similar number of viable cells as in untreated controls. However, a decreased proliferation after P treatment of male astrocytes was not able to counterbalance the increased apoptotic cell death leading to a net-decrease of viable cells. Other authors have similarly demonstrated E to increase 4.4. Discussion 67 glial cell proliferation (Jung Testas et al., 1992; Dhandapani and Brann, 2002), whereas no effect of P on cell proliferation was described. This could be due to the fact that gender-specific differences observed for P-treated astrocytes are counterbalanced in cell cultures of mixed genders. In conclusion, these results show for the first time that sex steroids are capable of influencing fusion and fission processes in astroglia mitochondria. Importantly, this effect occurred in a gender-specific way and revealed distinct differences in the effectiveness between E and P. Under physiological conditions, sex steroids may be important for the balance of mitochondrial morphology and function in the brain. After toxic events and under neuropathological conditions, sex steroids may differently contribute to the homeostasis and integrity of cell function, thus providing a basis for the well-known gender differences in neuronal vulnerability and cell death. These findings suggest a pronounced sensitivity of male astroglia for apoptosis, which may serve as an explanation for the higher incidences of neurodegenerative diseases in males (Miller et al., 1998; Roof and Hall, 2000; Dluzen and McDermott, 2004; Kenchappa et al., 2004; Marchetti et al., 2005). 68 Chapter 4. Steroid hormone effects on mitochondrial fusion and fission Chapter 5

Mitochondrial fusion and ﬁssion processes in the mouse cortex development

5.1. Introduction 71

Abstract

Mitochondria are responsible for major cellular functions including metabolism, signalling and apoptosis. Deciding between survival and death, mitochondria play a central role in tissue development and maintenance as well as in senescence and disease. Recent research has revealed the intimate connection of mitochondrial physiology to their morphology. The complex and dynamic processes of mitochondrial fusion and fission regulate the maintenance of the mitochondrial network. When imbalanced, they cause deep changes in mitochondrial architecture and consequently on their function. Mitochondrial diseases include many neurodegenerative disorders and mitochondria are perfect targets for neuroprotection. Steroid hormones, as progesterone and estrogen, influence brain development and neuroprotection directly and indirectly through their action on the neuroprotective astrocytes. These studies aim at understanding the effects steroid hormones have on the fission and fusion events in mitochondria and what contribution this might play in brain development and the involved tissue specializations, influencing neuroprotection. To investigate how mitochondrial fusion and fission processes might be involved in developmental, tissue-specialization mechanisms, the fusion Mfn1 and Mfn2, and the fission Drp1 and Fis1 mRNA levels were measured during mouse brain cortex development. The results correlate very well with the previous published work by others on the stages of cell growth and specializations during mice cortex development, and have shown several sex-specific differences.

5.1 Introduction

Mitochondria play a fundamental role in many important cell functions including energy metabolism, thermo genesis, calcium homeostasis, cell growth, division, signalling and regulation of apoptosis. Their dysfunction can therefore cause important cellular disequilibrium due to diminished cell energy, production of reactive oxygen species, disturbance of ionic balance, cytochrome C release and lately apoptosis (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Chen et al., 2005). Mitochondrial functions and characteristics explain their direct connection to aging and various diseases including cancer, obesity, diabetes, cardiovascular disorders and several neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS), autosomal dominant optic atrophy, Charcot-Marie-Tooth, Parkinson, Huntington’s and Alzheimer’s diseases and others, stressing the importance of understanding the mechanisms underlying mitochondrial functions (Chapter 1.4.1) (Alexander et al., 2000; Delettre et al., 2000; Bach et al., 2003; Chen et al., 2004; Zeviani and Donato, 2004; Zuchner et al., 2004; Chen and Chan, 2005; Kijima et al., 2005; Tatsuta and Langer, 2008) Mitochondria usually assume a tubular network morphology, but are present in cells in different shapes 72 Chapter 5. Mitochondrial fusion and ﬁssion processes in the mouse cortex development

and quantity and also have distinct distributions inside the cell, accumulating in high energy demanding cellular regions. Several studies suggest that mitochondria do not just differ in their morphology and distribution inside the cell, but that they are also biochemically and functionally diverse in different tissues (Benard et al., 2006), and even inside the same cell (Kuznetsov et al., 2006). The mitochondrial morphology is controlled by two opposing and balanced processes, fusion and fission. These processes of organelle shaping occur far more times than the necessary to maintain mitochondrial morphology, and are therefore thought to be essential for the proper functioning of mitochondria. In fact, fission is necessary for the even distribution of mitochondria to the daughter cells during cell division, and is essential in apoptosis (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Stojanovski et al., 2004). Fusion, on the other side, serves the exchange of carriers and nutrients, and allows for mtDNA molecules exchange and the gene complementation that prevents the detrimental effects of accumulating mtDNA mutations and cellular dysfunction (Nakada et al., 2001; Chen and Chan, 2005; Chan, 2007). The fusion and fission processes controlling mitochondrial morphology, are each regulated by a specific set of proteins. Fusion is dependent on mitochondrial GTPases mitofusin 1 and 2 (Mfn1, Mfn2), located in the mitochondrial outer wall and being responsible for the membrane tethering between two mitochondria during fusion, and dominant optic atrophy-associated protein (OPA1), an intermembrane protein thought to participate in the mitochondrial inner membrane fusion and cristae modelling. Mitochondrial fission on the other hand, is regulated by mitochondrial fission 1 protein (Fis1) an outer mitochondrial membrane-bound protein, and dynamin- related protein 1 (Drp1), a cytoplasmic protein, recruited to the mitochondrial surface during fission. Other proteins were found to be involved in the processes of mitochondrial remodelling, like Ugo1 and Mdv1 and there is currently active investigation to disclose these mechanisms (Santel et al., 2001; Rojo et al., 2002; Chen et al., 2003; Olichon et al., 2003; Cipolat et al., 2004; Griparic et al., 2004; Chen and Chan, 2005). The mitochondrial morphology maintenance is dependent on the equilibrium of these two opposing and constant processes. An imbalanced expression of fusion or fission proteins disrupts this equilibrium and causes dramatic changes in mitochondrial morphology leading to dysfunction (Yaffe, 1999; Karbowski and Youle, 2003). In humans, mutations in Mfn2 cause Charcot-Marie-Tooth neuropathy type 2A (Zuchner et al., 2004; Kijima et al., 2005) and mutations in OPA1 lead to autosomal dominant optic atrophy (Alexander et al., 2000; Delettre et al., 2000). The tissue-specificity of some mutations could be connected to a time- point in development. To understand these diverse tissue-specific phenotypes, it is critical to elucidate the relative importance of mitochondrial fusion and fission processes in development and in connection to abnormal cell conditions. A down-regulation of fusion proteins or an over-expression of fission proteins causes fragmentation of the tubular mitochondrial network. Cells lacking Mfn1 and Mfn2, or OPA1, lacked mitochondrial fusion and showed severe dysfunctions, such as widespread heterogeneity of mitochondrial 5.1. Introduction 73 membrane potential, decreased cellular respiration, poor cell growth, increased susceptibility to cell death, and apoptosis (James et al., 2003; Yoon et al., 2003; Lee et al., 2004; Chen et al., 2005). Mfn1 or Mfn2 Knock-out mice dye during embryogenesis, demonstrating the essential role of mitochondrial fusion in embryonic development and cell viability (Chen et al., 2003). Small interfering RNA (siRNA) knock-down of OPA1 induced fragmented mitochondria with dissipation of membrane potential and cristae disassembly, which render highly sensitized cells to exogenous apoptotic stimuli (Olichon et al., 2003; Griparic et al., 2004; Lee et al., 2004). In C. elegans, over-expression of fission protein Drp1 increased the number of mitochondrial divisions (Labrousse et al., 1999). Fis1 over-expression caused mitochondrial fragmentation and aggregation (Stojanovski et al., 2004). It is important to retain that mitochondrial fragmentation does not always result in apoptosis (Lee et al., 2004). On the other side, over-expression of fusion proteins or down-regulation of fission proteins stimulates mitochondrial fusion and the development of an elongated mitochondrial network, inhibiting the apoptosis-related mitochondrial fragmentation and therefore providing protection against different apoptotic stimuli (Smirnova et al., 2001; Santel et al., 2003; Sugioka et al., 2004). Fis1 knock-down by small interference RNA blocked the recruitment and activation of Bax to the surface of mitochondria after a death stimulus indicating an essential role for Fis1 in apoptosis (Lee et al., 2004). Similarly, a loss of Drp1 blocked mitochondrial fission and cytochrome c release, although it did not significantly interfere with Bax activation (Breckenridge et al., 2003; Lee et al., 2004). The connection between mitochondrial fusion and apoptosis inhibition through cell signalling first came from the identification of Mfn2 (called previously hyperplasia suppressor gene HSG) as an important anti-proliferating protein, which interferes with the Ras pathway and blocks signalling from growth factor receptors at the plasma membrane (Chen et al., 2004). Furthermore, the pro-apoptotic protein Bax was found to co-localize with Mfn2 and Drp1 sites on mitochondria during apoptosis (Karbowski and Youle, 2003) and an activated form of Mfn2 blocked Bax activation and cytochrome c release, protecting cells from apoptosis (Neuspiel et al., 2005). These discoveries strongly point towards the importance of the morphological state of mitochondria in relation to their activity and cellular functions, through the involvement of cellular signalling cascades. Interestingly, the cell functions managed by mitochondria are also modulated by steroid hormones (Nilson et al., 2002; Kipp et al., 2006). Although the mechanisms for these are still not fully understood, mitochondria were found to have interactions with glucocorticoid, estrogen, androgen, and progesterone receptors (Gavrilova-Jordan, 2007). In previous studies, it was observed that estrogen stimulates the expression of mitochondria-encoded subunits of the respiratory chain, and influences mitochondrial respiratory activity (Araujo et al., 2008), while others have shown the direct interaction of estrogen with respiratory chain complexes (Zheng and Ramirez, 1999). Steroid hormones are known to be important regulatory signals for brain development and in biological 74 Chapter 5. Mitochondrial fusion and fission processes in the mouse cortex development

functions (Hutchison et al., 1997; Beyer, 1999; Mendez et al., 2005; Daniel, 2006; McCarthy, 2008). The developing and adult brain cells express receptors for progesterone and estrogen (Zhang et al., 2002; Mitra et al., 2003; Quadros et al., 2007), being therefore unmistakable targets of these hormones, even though during development, male have a stable PR presence, while female have little. The PR is transcribed by activated ER, showing the effects caused by these hormones are intimately connected. Estrogen plays an essential role in late embryonic and early postnatal brain development, controlling cell growth, neurons and astrocytes morphometry and tissue differentiation, regulating cell apoptotic degeneration, synaptogenesis, organization of brain circuits and sex-specific differentiation of the CNS (Mesano and Jaffe, 1997; Beyer, 1999; Woo et al., 2001; McCarthy, 2008). Estrogen does not only originate from the foetuses gonads and the feto-placenta-maternal unit (Tuchinsky et al., 1972; Trotter et al., 1999), but is also synthesized locally in the nervous tissues. The aromatase subcellular localization next to synaptic terminals shows a possible E action like a neurotransmitter. Estrogen receptors were also found in the axon and dendrites terminals, seemingly too far to return to the nucleus for transcription activation, suggesting a nontranscriptional effect of ER in the nervous system or, most probably, a local interaction with neighbouring mitochondria which locate next to synapses (McCarthy, 2008). The neuroprotective effects of estrogen and progesterone have been described by several research groups. Astrocytes play an essential role in the support and protection of neurons, being therefore able to influence neuroprotective processes in the brain (Beyer, 1999; Beyer et al., 2003; Garcia-Segura et al., 2003; Kajta and Beyer 2003; Kipp et al., 2006). Astroglial cells express estrogen and progesterone receptors, being therefore known targets of these steroid hormones (Jung-Testas et al., 1992). There are some studies on the effects of estrogen and progesterone and their interaction with mitochondria, but the connection of these steroid hormones to the mitochondrial fusion and fission processes is still not elucidated. Knowing mitochondria play an important part in tissue differentiation, this small project aimed at clarifying the relative proportion of mitochondrial fission and fusion, apoptosis and proliferation mechanisms during the development of the mouse cortex. These studies aim at the understanding of the effects estrogen and progesterone might have on the fission and fusion events in mitochondria, thus influencing their morphology and function, and what contribution this might have during mouse brain cortex development and astrocyte survival, influencing later neuroprotection.

5.2 Materials and Methods

For information on the materials and methods used in this study, please refer to Chapter 2. 5.3. Results 75

5.3 Results

5.3.1 Ontogenic regulation of mitochondrial fusion and ﬁssion gene

transcription

To verify the connection of mitochondrial fusion and fission processes to the development of cortical brain cells, the transcription of fusion (Mfn1, Mfn2) and fission genes (Drp1, Fis1) was analysed along with markers of proliferation (PCNA) and apoptosis (Bax and Bcl2) in mouse brain cortices on late embryonic development days E15 and E17 and early postnatal days P0 (birth), P7, and P15. Therefore, cortex tissue was isolated from mice in the several developmental stages, total RNA was extracted and used to produce cDNA for PCR analysis. Semi-quantitative RT-PCR (sqRT-PCR) was performed for the genes of interest (for details please refer to Chapter 2, section 2.7) and PCR results were normalized against the respective housekeeping genes, 18S rRNA and HPRT, results. The significance of differences between the groups of different ages was determined according to a T-test by comparing samples of different ages with the youngest group E15. During development, Mfn1 transcription was mostly leveled in the first 3 stages of development, then lowering considerably in P7 when female and male cortex showed a significant decrease of the transcription of Mfn1 from E15 to P7 by approximately two-fold (Fig. 5.1 A, D). In females, the lowest level of Mfn1 transcription on P7 then greatly and significantly increased on P15, almost reaching the transcription levels in E15 (Fig. 5.1 A). In males, Mfn1 expression did not significantly change from P7 to P15 (Fig. 5.1 D). The Mfn2 transcription levels were quite different to the Mfn1 levels, showing no significant changes in females during the development stages studied, while in males, Mfn2 expression significantly increased from P7 to P15 by approximately a third (Fig. 5.1 D). To these changes in the transcription profile of each of the fusion genes studied, should be added the significant differences in the levels of Mfn1 versus Mfn2 transcription at the same development stage. Thus, in females, the level of Mfn1 in P7 is less than half of the level of Mfn2, the same holding true for males in both P7 and P15. Mfn2 levels being about the double transcription level to Mfn1, showing a significant mitofusin 2 transcription preference (Fig. 5.1 D). There was also a sex-related significant difference in Mfn1 versus Mfn2 levels in P15 in males in contrast to females (Fig. 5.1 A, D). The transcription of fission genes did not show any significant changes until P15, when Drp1 was significantly augmented in female cortical brain tissue, but not in male brain cortex, denoting another sex-specific effect (Fig. 5.1 A, D; Fig. 5.2 B). Fis1 transcriptions were not significantly altered during brain cortical development over the time period studied (each stage in comparison to E15), even though Fis1 shows an increasing tendency with the developmental age in female brain cortex but not in male (Fig. 5.1 B, E). Transcription of the proliferation marker, PCNA, along with the pro- and anti-apoptotic marker, Bax and Bcl2, were gradually 76 Chapter 5. Mitochondrial fusion and fission processes in the mouse cortex development

and significantly down-regulated in female cortical tissue from E15 to P15 (Fig. 5.1 C). In male cortical tissue, PCNA also reduced gradually and significantly during development, while Bcl2 transcription was not significantly changed and there was only a significant reduction of Bax in P15 (Fig. 5.1 F). At embryonic ages, proliferation, anti-, and pro-apoptotic marker levels were higher in female than in male cortices, but aligned in the post-natal days. These results show very obvious sex-differences in the transcription levels of the genes studied. The levels of Fis1 and Drp1 in female and male show a significant difference in P15 (Fig. 5.2 A, B). Further significant sex-differences are shown on Bcl2 and Bax gene transcription (Fig 5.2 C, D), on E15 and E17, further also on P0 for Bcl2 only (Fig 5.2 C). At those developmental stages, female have higher transcription values of Fis1, Drp1, Bcl2 and Bax.

5.4 Discussion

In mouse development, cortex forms at embryonic day 14 (E14), when migrating cells from telencephalon compose the superficial cortical area of the primary cortex, which is then easily recognized at E15, enlarging considerably in the following days. The brain continues developing and differentiation of accessory struc- tures further takes place until two weeks after birth (Theiler, 1989). Estrogen blood concentrations during mouse development include a gradual increase from day E13 until a maximum level is reached at E17, dimin- ishing to the lowest estrogen levels on P0 and increasing during the first week after birth (Beyer, 1999). On the other side, Aromatase P450, the enzyme responsible for the transformation of testosterone in estrogen, is expressed in neurons of the developing and adult brain (Schumacher et al., 2003; Prange-Kiel et al., 2006; Von Schassen et al., 2006). Estrogen can also be synthesized de novo from cholesterol and the necessary enzymes are also present in the female and male developing brain. Because of local synthesis, steroid hormones can reach much higher local concentrations than anticipated and the greater aromatase activity in males than in females might be responsible for the sex brain dimorphism (Hutchison et al., 1997; Beyer, 1999; Quadros et al., 2007). Thus, the local brain synthesis of estrogen seems to compensate for the lower estrogen concentration from the blood stream during the late embryonic and early post-natal days. This time interval is believed to be the estrogen’s sensitivity-window which allows for major permanent brain modifications, including brain sex differences connected to adult sexual behavior (McCarthy, 2008). Although it is well known that estrogen increases cell proliferation (Jung-Testas et al., 1998), how it would influence mitochondrial fusion and fission remained unanswered. The results presented here confirm the high pre-natal cellular proliferation rates through a high transcription of the proliferation marker PCNA and mitochondrial fusion and fission proteins, needed for the proper distribution of mitochondria to daughter cells and morphology maintenance. The pro- and anti-apoptotic markers Bax and Bcl2 are also highly expressed, probably for the simultaneous 5.4. Discussion 77

(A) (B) 1,0 Female Fusion 1,0 Male Fusion # # # # Mfn1 # # Mfn2 Mfn1 0,5 0,5 Mfn2

* * mRNA expression (a.u.) mRNA expression (a.u.)

0,0 0,0 E15 E17 P0 P7 P15 E15 E17 P0 P7 P15 Age Age

1,5 * 1,5

Fis1 Fis1 Drp1 1,0 Drp1 1,0

0,5 0,5 mRNA expression (a.u.) mRNA expression (a.u.)

0,0 0,0 E15 E17 P0 P7 P15 E15 E17 P0 P7 P15

(E) Female Proliferation (F) 2,0 2,0 Male Proliferation # # # # # # # # # # # # # 1,5 # 1,5 # # * * Bcl2 # # Bcl2 * * Bax # Bax * PCNA PCNA 1,0 * * * 1,0 * * * * * * * * * 0,5 0,5 * mRNA expression (a.u.) mRNA expression (a.u.)

0,0 0,0 E15 E17 P0 P7 P15 E15 E17 P0 P7 P15 Age Age

Figure 5.1: Transcription of mitochondrial fusion and fission proteins, apoptosis and proliferation markers in the developing cortex of female and male mice. Brain cortices from female (A, B, C) and male Balb/c mice (D, E, F) of embryonic day 15 and 17 (E15 and E17, respectively), and postnatal days 0, 7, and 15 (P0, P7, and P15, respectively) were isolated. Total RNA was extracted and first strand cDNA was synthesized for amplification of Mfn1, Mfn2, Fis1, Drp1, Bcl2, Bax, and PCNA by sqRT-PCR. Data were normalized to housekeeping genes 18S rRNA and HPRT and represent means ± SEM of three to five independent experiments. Statistically significant differences between groups as compared to the E15 (* - P<0.05, ** - P<0.01, *** - P<0.001) and between other groups as connected by a line (# - P<0.05, ## - P<0.01, ### - P<0.001) were determined using Student’s unpaired t-test. 78 Chapter 5. Mitochondrial fusion and fission processes in the mouse cortex development

(A) Fis1 (B) 2,0 2,0 Drp1 # # # Female # Female Male # 1,5 1,5 Male

1,0 1,0

0,5 0,5 mRNA expression (a.u) mRNA expression (a.u.)

0,0 0,0 P15 P15 Age Age

# # Female Female # Male Male 1,5 # # 1,5 # # # # 1,0 1,0

0,5 0,5 mRNA expression (a.u.) mRNA expression (a.u.)

0,0 0,0 E15 E17 P0 E15 E17 Age Age

Figure 5.2: Sex differences on the transcription of mitochondrial fission proteins (Fis1 and Drp1), apoptosis (Bax) and anti-apoptotic (Bcl2) markers in the developing brain cortex of female (lighter columns) and male mice (darker columns) at the time in development when the differences became significant. Brain cortices from female and male Balb/c mice of embryonic day 15 and 17 (E15 and E17, respectively), and postnatal days 0, 7, and 15 (P0, P7, and P15, respectively) were isolated. Total RNA was extracted and first strand cDNA was synthesized for amplification of the genes of interest by sqRT-PCR. Data were normalized to housekeeping genes 18S rRNA and HPRT and represent means ± SEM of three to five independent experiments. Statistically significant differences (#- P<0.05, ## - P<0.01, ### - P<0.001) between groups were determined using Student’s unpaired t-test.

apoptotic and survival mechanisms connected to tissue differentiation. In the two weeks after birth, brain cell proliferation is lower, cell differentiation is intensified and accompanied by a rise in the number of mitochondria in cell somata (Gregson & Williams, 1969; Samson et al., 1960). Accordingly in the results described here, after birth there is a significant decrease of transcription of the proliferation marker PCNA and a much higher transcription of the anti-apoptotic marker Bcl2 compared to the pro-apoptotic marker Bax, resulting in cell survival. Simultaneously, Mfn1 transcription was decreased and Drp1 transcription increased, although in female only, on P7 and P15, respectively (Fig. 5.1 A, B, D), indicating enhanced mitochondrial fission in a sex-dependent fashion. These observations support reports on a rapid increase of mitochondrial division and volume between P7 and P15, which then rise gradually until adulthood (Pysh, 1970; Erecinska et al., 2004). An increased proliferation is mostly physiologically advantageous for a cell 5.4. Discussion 79 population during development or undergoing cell death. Under maintenance conditions however, the expression of fusion proteins and pro-apoptotic proteins is induced to counteract mitochondrial fission and cell proliferation, thereby keeping the cell number on a constant level and preventing cells from deterioration into cancer cells. The latter is an unwanted side-effect of estrogen therapy for postmenopausal women developing breast cancer (for review, see Yager and Davidson, 2006). Higher mitochondrial fission in female cortical astrocytes could result on either apoptosis or elevated cell growth in female cortex (Lee et al., 2004; Breckenridge et al., 2003). It is well known that brain sex differentiation involves apoptosis. The more extensive mitochondrial division in female could possibly lead to apoptosis in some areas of the brain, maybe even regarding tissue specializations leading to sexual differences in the brain cortical tissue. In the preoptic area, the sexually dimorphic nucleus of male mice, responsible for male-sex-behaviour, is five to seven times larger than in the females, due to estrogen directed apoptosis in female, between P3 and P7. Through the same estrogen regulated apoptosis mechanism, but in reversed sex, so is the anteroventral periventricular nucleus bigger in females than in males (McCarthy, 2008). Several studies have shown that sex-related differences are also evident in the human cortex tissue, as for example in the Broca’s and Wernicke’s areas related to language-associated thoughts, which were significantly larger in women. Using magnetic resonance imaging, the scientists measured the volumes of gray matter in several cortical regions and reached the conclusion that women had 23% (in Broca’s area, in the dorsolateral prefrontal cortex) and 13% (in Wernicke’s area, in the superior temporal cortex) more volume than men (Schlaepfer et al., 1995). Another research presented morphological evidence that men have more neurons in the cerebral cortex, while women have a more developed neuropil, the space between cell bodies, which contains synapses, dendrites and axons, and is responsible for communication between neurons. According to the scientists, this may be the reason why women are more prone to dementia (for example Alzheimer’s disease) than men, as although both could possibly lose the same amount of neurons due to the disease, the functional reserve may be greater in males, as a larger number of nerve cells are present, and this could prevent some functional losses (Rabinowicz et al., 1999). How these mechanisms are triggered and occur in a sex- and brain region -specific way is still to be clarified. In physiological conditions, cellular effects are induced by a net-effect of several substances and pathways, in the interconnected systems belonging to an organism. The tissue-specificity of some mutations could be connected to a time-point in development. To understand these diverse tissue-specific phenotypes, it is critical to elucidate the relative importance of mitochondrial fusion and fission processes in development and in connection to abnormal cell conditions. Further research on the single phenomena occurring in the physiological conditions needs to be done as to continue unraveling these processes. 80 Chapter 5. Mitochondrial fusion and fission processes in the mouse cortex development Chapter 6

General discussion

Due to the increasing world’s older population, dementing diseases could soon become main causes of disability, what demands for active research in neurodegeneration protection, generally named neuroprotection. Astrocytes play an important role in neuroprotection, for they support neuronal metabolism, neurotransmitter synthesis and energy homeostasis of the brain (Behl, 2002; Dhandapani and Brann, 2007; Pellerin et al., 2007; Schousboe et al., 2007), pointing at mitochondria in astrocytes as a most probable compartment involved in neuroprotection. Mitochondria are key players in many essential cell functions, including energy metabolism, thermogenesis, intracellular calcium homeostasis, cell growth, division, signalling and also apoptosis (Bettini and Maggi, 1992; Behl, 2002; Chen et al., 2004). Because mitochondria not only are key players in essential cell functions, but also function as integrators of diverse cellular stress and toxic signals (Greenamyre et al., 1999; Reynolds, 1999; Choo et al., 2003; Beal, 2005), their dysfunction may therefore cause major cellular imbalance. Interestingly, the cell functions managed by mitochondria are also modulated by steroid hormones (Nilson et al., 2002; Kipp et al., 2006). Herewith, it is known that for a wide range of neurodegenerative diseases, there is a sex-specific incidence and course (Gillies et al., 2004; Poole et al., 2008; Schapira, 2008; Schulman, 2007), most probably connected to the sex-specific brain differences described (Hutchison et al., 1997; Beyer, 1999; Quadros et al., 2007, McCarthy, 2008). This adds to the brain-region specificity of several diseases and clearly indicates that further research comparing brain regions and sex differences as well as brain development and aging, is essential for the therapy research. With this in mind, some studies presented here were performed in isolated female and male astrocyte cultures, as well as on the developing cortex tissue of female and male mice. The processes by which steroid hormones are related to, and possibly regulate, mitochondria, need to be clarified further and are subject of current research. In this study, chapter 3 is dedicated to the steroid hormone effects on mitochondrial respiration and it is shown there that E affects mitochondria in a time and concentration dependent manner. E promotes the expression of selected representative mitochondrial-encoded subunits of the respiratory chain in different proportions and up to 100% more than in control conditions, influencing mitochondrial respiratory activity and enhancing functional efficiency (Chapter 3). Studies by others support these results (Zheng and Ramirez, 1999; Chen et al., 2004; Irwin et al., 2008). These effects were also brain-region-specific, being much higher in the mesencephalic cells than on the cortical cells, as is the case for cytB and ATP6 expression. Why some genes were transcribed more than others and what consequences this might have, needs to be clarified further, but it is here hypothesized that this might be related to the sensitivity of the different respiratory complexes to malfunction, the most sensitive ones being the most transcribed genes as to substitute the malfunctioning proteins in a protective mechanism.According to the results described in chapter 3, the most sensitive respiratory complexes, being the ones which genes are transcribed most, would be Complex IV and 84 Chapter 6. General discussion

V (Cox2 and ATP6) in brain cortex and Complex III and V (CytB and ATP6) in mesencephalon. Why these proteins would be more sensitive as hypothesized here, remains to be elucidated. It is also target of further research what other effects in the cell E provokes and their relation to this subunit transcription augmentation. The co-treatment of cells with estrogen and the nuclear estrogen’s receptor antagonist ICI, did not influence the results obtained with estrogen’s treatment alone, showing that the stimulation of mitochondrial gene expression is independent of the nuclear ER, the classical pathway, and is rather due to direct E effects on mitochondria. Others have demonstrated the localisation of ERα and ERβ to mitochondria and related their presence to the mediation of functional effects caused by E treatment (Monje et al., 2001; Mitra et al., 2003; Chen et al., 2004; Yager and Chen, 2007; Milner et al., 2008). Furthermore, the ratio of mitochondrial to nuclear DNA, was increased in both cortical and mesencephalic astrocyte cultures, but much higher in mesencephalic cultures as on the cortical, here also, agreeing with the previous results of higher respiratory chain gene expression (Araujo et al., 2008). The elevated mtDNA levels and the increased transcript levels of mtDNA-encoded subunits of the respiratory chain, indicate an elevation of the mitochondrial mass in the cells, as under the study conditions, there were no changes in the astrocyte proliferation levels under E treatment as compared to controls. To understand what effect the estrogen’s stimulated higher expression of respiratory chain genes, and the amount of mitochondrial to nuclear DNA, might have on the actual respiratory activity of the cells, mitochondria were isolated and polarographic measurements of their oxygen consumption stimulated by the addition of the substrate succinate were made. Interestingly, cortical astrocytes had up to two-fold higher activity of isolated mitochondria than mesencephalic. Unexpectedly, after a long-term treatment (8 and 12h) of E there were no significant differences in the respiration of E-treated versus untreated cortical and mesencephalic astrocytes stimulated by succinate, ADP or CCCP, nor on the efficiency of energy production, RCR. This discrepancy was thoroughly discussed in chapter 3, and the values presented were supported by the work of others according to the conditions and substrate used. Nevertheless, mitochondrial respiration coupling is dependent on the membrane conditions of the isolated mitochondria, and another attempt was made to evaluate the intactness of the isolated mitochondria, using scanning electron microscopy, SEM (many thanks to D. Alves). The pictures show well kept membranes, inclusively very clear inner membrane invaginations (Fig. 3.6), which show that a fair degree of intactness of the mitochondria is kept and which are supported by the proposed 3D model of mitochondrial imaging of Manella (Manella, 2000). Isolated mitochondria have been described since decades as small and spherical to elliptical. During the extraction, which obliges to the breakage of cells and necessarily, breakage of the mitochondrial net, the method cannot preserve mitochondria in their original physiological structure. Even though isolated mitochondria do not compare to mitochondria in physiological conditions, the form showed here is similar to one of the isolated mitochondria forms depicted by others, as being smaller, elliptical 85 mitochondria with randomly disposed cristae appearing to be floating free in the inner compartment (Ambu et al., 2000). Besides the studies on the respiration of mitochondria after a long-term treatment of E, a short time treatment was also tested, as to see what immediate effects E might have on mitochondrial respiration. A short time treatment with E (5 min) led to a brain-region specific decrease of the succinate-stimulated oxygen consumption of cortical mitochondria, and a tendency to increase that of the mesencephalic respiration, although not significantly. Interestingly, RCR values remained unchanged for mitochondria from astrocytes of both brain regions, indicating that short-term E treatment of astrocytes does not influence the efficiency of mitochondrial energy production, and only affects the activity of proton-translocating complexes of the respiratory chain. Without compensation in form of up-regulation of ATP synthase units or ATP synthase activity, this effect could lead to the hyperpolarisation of the mitochondrial membrane and the higher production of deleterious reactive oxygen species (ROS). This indicates that short-term E treatment is less effective for improvement of mitochondrial function and survival of astrocytes in the mesencephalon, while E protects cortical astrocytes by decreasing respiration to a level where there is lower ROS production. This might in fact be a mechanism of cell protection. Furthermore, the differences in the expression of the respiratory chain genes and respiratory activities in the different brain regions might be related to, and even explain, the different sensitivity of brain regions to ischemic processes and degeneration, and also to therapies. Another point of interest in studying mitochondrial function is the relation of physiology to morphology, or vice-versa. The mitochondrial net architecture seems to dictate the optimal functioning and cellular distribution of mitochondria according to the cells energetic and physiological demands. Several studies suggest that mitochondria differ not only in their morphology and cellular location but also that they are biochemically and functionally diverse in different tissues (Benard et al., 2006) and also inside the same cell (Kuznetsov et al., 2006). The mitochondrial net architecture is controlled by fusion and fission processes and the mitochondrial form re-adaptations occur far more often than the necessary to maintain mitochondrial morphology, pointing at their physiological processes as the reason for such mechanisms. Additionally, a higher fusion to fission ratio results in some protection against apoptosis, while a higher fission could lead to apoptosis, but is also necessary during cell division for the equal distribution of mitochondria to the daughter cells (Taguchi et al., 2007; Stojanovski et al., 2004; Sugioka et al., 2004). In chapter 4 of this work, the effects of steroid hormones on the mitochondrial fusion and fission gene transcription in astrocytes are addressed. It is shown there that steroid hormones influence the mitochondrial fusion and fission gene transcription of cultured astrocytes from cortex and mesencehalon, of female and male mice. For cortical astrocytes of both genders, a correlation between Mfn1 (fusion) and anti-apoptotic marker Bcl2 transcription as well as 86 Chapter 6. General discussion

Drp1 (fission) and proliferation marker PCNA transcription was made, as supported by the results from others, showing the role of fusion genes in anti-apoptotic processes and fission genes in proliferation and apoptosis (Yaffe, 1999; James et al., 2003; Karbowski and Youle, 2003; Olichon et al., 2003; Yoon et al., 2003; Lee et al., 2004; Stojanovski et al., 2004; Chen and Chan, 2005; Chen et al., 2005; Chan, 2007). The studies presented here also revealed gender differences in the transcription of Mfn1 and Drp1 showing higher levels in females which were accompanied by lower apoptotic cell death and elevated number of viable cells. Treatment of female cortical astrocytes with E or P 10−7 M for 24 h induced a pronounced increase by approximately 50% in the number of viable cells, while in male cells, the application of E did not shown any effect, whereas P decreased the number of viable cells. A clear gender-specific difference in the apoptosis rate of steroid-treated astrocytes was observed and could be correlated with fusion and fission gene transcription. While female astrocytes showed a mainly balanced increase of both fusion and fission gene transcription and probably therefore also a balanced transcription of anti- and pro-apoptotic marker genes, in males the fusion and fission gene transcription was imbalanced, which seems to have caused a disturbance of the balance between anti- and pro-apoptotic marker genes, thus promoting apoptotic processes specifically in male astrocytes treated with gonadal steroids. A higher level of fusion gene transcription supports the increased astrocyte viability in females (Zuchner et al., 2004; Kijima et al., 2005), while it is very possible that the increased apoptosis in male astrocyte cultures could be compensated by an elevation of proliferation, which would result in an unchanged level of viable cells after steroid hormone treatment, as compared to untreated controls. In fact, male astrocytes treated with E exhibited a higher PCNA/Bax ratio than P-treated cells. Female astrocytes, on the other hand, showed a ratio of PCNA/Bax > 1 and the resulting proliferation increase at constant levels of apoptosis elevated the number of viable cells. In male astrocytes, however, a decreased proliferation after P treatment was not able to counterbalance the increased apoptotic cell death leading to a decrease of viable cells. Other authors have similarly demonstrated E to increase glial cell proliferation (Jung Testas et al., 1992; Dhandapani and Brann, 2002), whereas no effect of P on cell proliferation was described. This could be due to the fact that gender-specific differences observed for P-treated astrocytes are counterbalanced in cell cultures of mixed genders. The transcription levels of pro-, anti-apoptotic and proliferation markers (Bax, Bcl2, and PCNA, respectively), correlated very well to the viability results. Interestingly, the PR inhibitor usually blocked P’s effects partially to totally, while ER inhibitor did not affect E’s effects. It is noteworthy that E influences mitochondria not only through its nuclear receptors, but acts also directly by binding of E to mitochondria in vitro and in situ (Yang et al., 2004; Stirone et al., 2005; Yager and Chen, 2007; Milner et al., 2008). Thus, some effects, in particular those exerted by E but not by P, were not subject to inhibition by classical nuclear ER antagonists. The brain development is another very interesting field to research upon, aiming at understanding the 87 characteristics of mitochondria during development, the essential tissue specializations and the possible relation to steroid hormones. The results presented in Chapter 5 of this thesis are supported by the described stages of brain development in the time window researched. It is shown here that there is a high pre-natal cellular proliferation rate through a high transcription of the proliferation marker PCNA and mitochondrial fusion and fission proteins. These would be needed for the proper distribution of mitochondria to daughter cells and morphology maintenance. The pro- and anti-apoptotic markers Bax and Bcl2 are also highly expressed, probably for the simultaneous apoptotic and survival mechanisms connected to tissue differentiation. After birth there is a significant decrease of transcription of the proliferation marker PCNA and a much higher transcription of the anti-apoptotic marker Bcl2 compared to the pro-apoptotic marker Bax, which might be resulting in cell survival. This is in accordance with studies by others, in which, in the two weeks after birth, brain cell proliferation is reduced, cell differentiation is intensified and accompanied by a rise in the number of mitochondria in cell somata (Gregson & Williams, 1969; Samson et al., 1960). Simultaneously, Mfn1 transcription was decreased and Fis1 and Drp1 transcription increased, although in female only, on P7 and P15, respectively indicating enhanced mitochondrial fission in a sex-dependent fashion for Drp1. These observations are supported by reports on a rapid increase of mitochondrial division and volume between P7 and P15, which then rise gradually until adulthood (Pysh, 1970; Erecinska et al., 2004). In physiological conditions, cellular effects are induced by a net-effect of several substances and pathways. E is known to induce P receptor transcription (McCarthy, 2008), therefore sensitizing cells to progesterone action. There are most probably other processes involved in the physiological conditions, and further research is being done currently. Generally, this work presents a small contribution to the clarification of the effects of steroid hormones on mitochondrial properties of astrocytes, more specifically in the respiratory physiology and in the morphologic characteristics of astrocytic mitochondria. Future research will clarify these mechanisms further, in order to possibly explain region and sex-specific sensitivity differences in the brain, connected to the incidence and course of neurodegenerative diseases and help shed light into possible mechanisms of neuroprotection. To understand the mechanisms by which steroid hormones act on the mitochondrial morphology and physiology, and how the respiration and the fusion and fission mitochondrial balance is affected, further research on apoptosis challenged cell cultures could be pursued. As techniques improve, imaging would also surely provide exciting pictures of the mitochondrial changes during different cell status and under steroid hormone treatments. Further knowledge on the mitochondrial characteristics is surely going to play a meaningful role in future disease therapies. 88 Chapter 6. General discussion Chapter 7

Summary / Zusammenfassung

Summary of the Thesis

Several points of interest in the neuroprotection field are approached in this work: steroid hormones, astrocytes and mitochondria. Steroid hormones, such as estrogen and progesterone, are well-known to regulate cell growth, function and protection in the central nervous system (CNS). Astrocytes play a major role in the brain’s energy metabolism, growth, regulation of cerebral blood flow, coupling neuronal activity and regulation of synapses, as well as neuroprotection. Furthermore, the essential cellular energy metabolism, regulation of cell growth, division, signaling and also of apoptosis, is dependent on the mitochondrial functions. This explains why mitochondrial dysfunction can cause important cellular imbalance and is connected to aging and disease, including many neurodegenerative infirmities. Interestingly, the cell functions managed by mitochondria are also modulated by steroid hormones, so that it seems mitochondria are perfect targets for strategies of cellular protection. In the first part of this thesis, focus is put on the energy metabolism regulation in the CNS. Steroid hormones are well-known to fulfill a protective role in the brain during ischemic and degenerative processes. Mitochondrial dysfunctions due to abnormalities of the mitochondrial respiratory chain often accompany neurodegenerative diseases, in which, degeneration often shows brain region specificity. Since astrocytes function as the major energy supplier in the CNS, the estrogen effects on the mitochondrial respiratory chain were here analyzed in primary astrocytes from the brain regions cortex and mesencephalon. It was observed that estrogen influences subunit gene expression of respiratory chain complexes, the mitochondrial DNA content and the functional respiration of isolated mitochondria differently, according to the estrogen exposure time as well as depending on the brain region. This could be related to the different sensitivity of brain regions to degeneration and also point at an important indirect mechanism by which estrogen protects neurons from cell death under neurotoxic conditions. On a subsequent chapter, the processes of mitochondrial fusion and fission that regulate energy metabolism, apoptosis, and cell proliferation, were approached. In this respect, astrocyte mitochondria seem to be a perfect intracellular target for steroids to modulate these processes. Therefore, the effects of estrogen and progesterone on cell viability in comparison with mitochondrial fusion and fission gene transcription in primary cortical astrocytes from female and male mouse brains are studied here. Estrogen- and progesterone- treated astrocytes demonstrated differences in cell number and proliferation marker accompanied by a corresponding difference of fusion and fission gene transcription, in a gender-dependent way. It seems estrogen and progesterone affect mitochondrial function differently in both genders. Therefore, interaction of sex steroids with mitochondria may represent one possible cause for gender differences in cellular pathology in the CNS. The last part of this thesis focuses on mitochondrial fusion and fission processes during mouse 92 Chapter 7. Summary / Zusammenfassung

brain development. Mitochondria are responsible for major cellular functions, therefore playing a central role in tissue development and maintenance as well as in senescence and disease. The complex and dynamic processes of mitochondrial fusion and fission regulate the maintenance of the mitochondrial network. When imbalanced, they cause deep changes in mitochondrial architecture and consequently on their function. Steroid hormones, as progesterone and estrogen, influence brain development and neuroprotection directly and indirectly through their action on the neuroprotective astrocytes. These studies aim at understanding the effects steroid hormones have on the fission and fusion events in mitochondria and what contribution this might play in brain development, tissue-specialization mechanisms and astrocyte survival, influencing neuroprotection. Therefore , the mRNA levels of the target fusion and fission genes were measured during mouse brain cortex development. The results correlate very well with the previous published work by others, on the stages of cell growth and specializations during mice cortex development, and have shown some sex-specific differences.

Zusammenfassung der Dissertation

Mehrere Schwerpunkte aus dem Bereich Neuroprotektion werden in dieser Arbeit behandelt: Steroid- hormone, Astrozyten und Mitochondrien. Steroidhormone wie Ostrogen¨ und Progesteron regulieren das Zellwachstum, die Funktion und den Schutz im zentralen Nervensystem (ZNS). Astrozyten spielen eine wichtige Rolle bei solchen Prozessen und auch beim Energiestoffwechsel des Gehirns. Daruber¨ hinaus sind der wesentliche zellulare¨ Energiestoffwechsel, die Regulation von Zellwachstum, Zellteilung, Signaling und auch die Apoptose von der Mitochondrienfunktion abhangig.¨ Dies erklart,¨ warum mitochondriale Dysfunktion ein nennenswertes zellulares¨ Ungleichgewicht verursachen kann und mit Alterung sowie vielen neurodegenerativen Erkrankungen in Verbindung steht. Interessanterweise werden die Zellfunktionen, die von Mitochondrien gewahrleistet¨ werden, auch durch Steroidhormone moduliert, so dass die Mitochondrien als perfekte Ziele fur¨ Strategien zum Zellschutz erscheinen. Im ersten Teil dieser Arbeit liegt der Schwerpunkt auf der Regulation des Energiestoffwechsels im ZNS. Steroidhormone spielen im Gehirn eine schutzende¨ Rolle bei ischamischen¨ und degenerativen Prozessen. Mitochondriale Funktionsstorungen¨ aufgrund von Veranderungen¨ der mitochondrialen Atmungskette be- gleiten oftmals neurodegenerativen Erkrankungen, bei denen die Degeneration oft Gehirnregion-spezifisch 93 ist. Da Astrozyten als der wichtigste Energieversorger im ZNS fungieren, wurde in dieser Arbeit die Ostrogen-Wirkung¨ auf die mitochondriale Atmungskette in primaren¨ Astrozyten von Kortex und Mesen- cephalon untersucht. Es zeigte sich, dass Ostrogen¨ die Genexpression der Atmungskettenkomplexe, den mitochondrialen DNA-Gehalt,die Menge an Mitochondrien pro Zelle, und dadurch, die funktionelle At- mung in isolierten Mitochondrien, unterschiedlich in Abhngigkeit von der Ostrogen-Expositionszeit¨ und der Gehirnregion geandert¨ hat. Dies konnte¨ mit der unterschiedlichen Sensitivitt verschiedener Hirnregionen gegenber Degenerationsprozessen verbunden sein, und auch auf einen wichtigen indirekten Mechanismus hindeuten, uber¨ den eine Ostrogen-Exposition¨ Neuronen vor dem Zelltod unter neurotoxischen Bedingun- gen schutzt.¨ In einem spateren¨ Kapitel wird die mitochondrialen Fusions- und Spaltungsprozesse, die Energiestof- fwechsel, sowie Apoptose und Zellteilung regeln, untersucht. Diesbezglich stellen Astrozyten-Mitochondrien einen perfekten intrazellularen¨ Angriffspunkt fur¨ Steroide zur Modulierung dieser Prozesse dar. Deshalb werden hier die Einflsse von Ostrogen¨ und Progesteron auf die Lebensfahigkeit¨ der Zellen im Vergleich mit der Gentranskription der mitochondrialen Fusion und Spaltung in primaren¨ kortikalen Astrozyten weiblicher und mannlicher¨ Mause-Gehirne¨ untersucht. Ostrogen-¨ und Progesteron-behandelte Astrozyten zeigten geschlechtsabhngig Anderungen¨ in ihrer Anzahl sowie hinsichtlich der Proliferationsmarker, be- gleitet von einer mit Fusion und Spaltung verbundenen Gen-Transkription. Es scheint, dass Ostrogen¨ und Progesteron die Mitochondrien-Funktion bei beiden Geschlechtern unterschiedlich beeinflussen. Somit konnen¨ die Wirkungen von Sexualhormonen auf Mitochondrien eine mogliche¨ Ursache fur¨ geschlechtsspez- ifische Unterschiede in der Zellpathologie des ZNS darstellen. Der letzte Teil dieser Arbeit konzentriert sich auf die mitochondrialen Fusions- und Spaltungs-Prozesse wahrend¨ der Gehirnentwicklung der Maus. Mitochondrien sind fur¨ wichtige Zellfunktionen verantwortlich und spielen daher eine zentrale Rolle bei der Gewebentwicklung und dessen Erhalt sowie beim Altern und bei Erkrankung. Die komplexen und dynamischen Ablaufe¨ bei der Mitochondrien-Fusion und Spaltung regeln die Aufrechterhaltung des mitochondrialen Netzwerks. Wenn sie unausgewogen sind, provozieren sie tiefgreifende Veranderungen¨ in der mitochondrialen Architektur und damit in ihrer Funktion. Steroid- hormone wie Progesteron und Ostrogen¨ beeinflussen die Entwicklung des Gehirns und die Neuroprotektion direkt und indirekt ber ihre Wirkung auf die neuroprotektiven Astrozyten. Die vorliegenden Untersuchun- gen zielen darauf ab, die Auswirkungen, die Steroidhormone auf die Spaltungs- und Fusions-Vorgange¨ der Mitochondrien haben, zu verstehen und zu erklaren,¨ welchen Beitrag dies bei der Gewebeentwicklung und -differenzierung und beim Uberleben¨ der Astrozyten und damit bei der Beeinflussung der Neuro- protektion liefern konnte.¨ Dazu ist der mRNA-Spiegel von Fusions- und Spaltungs Genen wahrend¨ der Hirnrindenentwicklung der Maus gemessen worden. Die Ergebnisse korrelieren in Bezug auf die Phasen von 94 Chapter 7. Summary / Zusammenfassung

Zellwachstum und -differenzierung wahrend¨ der Entwicklung der Mause-Hirnrinde¨ sehr gut mit den bereits veroffentlichten¨ Arbeiten anderer Gruppen und haben auerdem einige geschlechtsspeziﬁsche Unterschiede gezeigt. Abbreviations

18S rRNA 18S ribosomal RNA; ADP adenosine diphosphate; ATP adenosine triphosphate; ATP6 ATP synthase (complex V), subunit 6; a.u. arbitrary units; Bax Bcl2-associated X protein; Bcl2 cell lymphoma protein 2; BSA bovine serum albumine CCCP carbonyl cyanide m-chlorophenylhydrazone; cDNA first strand complementary DNA; CNS central nervous system; COX cytochrome c oxidase; Cox2 cytochrome c oxidase (respiratory complex IV), subunit 2; CytB ubiquinol cytochrome c oxidoreductase (complex III), subunit cytochrome b; Ct threshold cycle; DEPC diethyl pyrocarbonate; DMEM Dulbecco’s modified Eagle medium; Dnm1l dynamin 1 like protein, synonym of Drp1; Drp1 dynamin-related protein 1; E estrogen (general term); E2 17β-estradiol; E15/17 embryonic day 15 or 17; EDTA ethylenediamine-tetraacetic acid ER estrogen receptor; FCS fetal calf serum; Fis1 fission 1 homolog (mitochondrial outer membrane); GFAP glial fibrillary acidic protein;

H, H2O2 hydrogen peroxide; HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; HPRT hypoxanthine guanine phosphoribosyl transferase; ICI ICI182,780, estrogen nuclear receptor inhibitor; IMM inner mitochondrial membrane; Mfn1 mitofusin 1; Mfn2 mitofusin 2; mtDNA mitochondrial DNA; NBM neurobasal medium; ncDNA nuclear DNA; 98 Abbreviations

NADH nicotinamide adenine dinucleotide, reduced form ND1 NADH ubiquinone oxidoreductase (complex I), subunit 1; OD optical density; OMM outer mitochondrial membrane; Opa1 optic atrophy associated protein 1; OSCP oligomycin-sensitivity conferring protein; P progesterone; P0/7/15 post-natal day 0 (birth), 7 or 15; PCNA proliferating cell nuclear antigen; PR progesterone receptor; qRT-PCR quantitative real time-PCR; RCR Respiratory Control Ratio; ROS reactive oxygen species; RTI, RTI3021-022 progesterone receptor inhibitor; S staurosporine; siRNA small interfering RNA; sqRT-PCR semi-quantitative, reverse transcriptase-PCR; TN Turnover Number. References

101

Alexander C, Votruba M, Pesch UE, Thiselton DL, Mayer S, Moore A, Rodriguez M, Kellner U, Leo-Kottler B, Auburger G, Bhattacharya SS and Wissinger B (2000). OPA1, encoding a dynamin-related GTPase, is mutated in autosomal dominant optic atrophy linked to chromosome 3q28. Nature Genetics 26 211-215.

Alkayed NJ, Harukuni I, Kimes AS, London ED, Traystman RJ and Hurn PD (1998). Gender-Linked Brain Injury in Experimental Stroke. Stroke 29 159-166.

Almeida A, Medina JM (1997). Isolation and characterization of tightly coupled mitochondria from neurons and astrocytes in primary culture. Brain Research 764 167-172.

Ambu R, Riva A, Lai ML, Loffredo F, Testa Riva F and B Tandler (2000). Scanning Electron Microscopy of the Interior of Cells in Huerthle Cell Tumors. Ultrastructural Pathology 24 211-219.

Araujo GW, Beyer C and Arnold S (2008). Estrogen inﬂuences on mitochondrial gene expression and respiratory chain activity in cortical and mesencephalic astrocytes. Journal of Neuroendocrinology 30 930-941.

Armstrong JS (2007). Mitochondrial medicine: pharmacological targeting of mitochondria in disease. British Journal of Pharmacology 151 1154-65.

Arnold S, Wright Araujo G, Beyer C (2008). Gender-speciﬁc regulation of mitochondrial fusion and ﬁssion gene transcription and viability of cortical astrocytes by steroid hormones. Journal of Molecular Endocrinology 41 289-300.

Arnold S, Beyer C (2009). Neuroprotection by estrogen in the brain: the mitochondrial compartment as presumed therapeutic target. Journal of Neurochemistry 110 1-11.

Arnold S (2005). Estrogen suppresses the impact of glucose deprivation on astrocytic calcium levels and signaling independently of the nuclear estrogen receptor. Neurobiology of Disease 20 82-92.

Arnold S, Kadenbach B (1997). Cell respiration is controlled by ATP, an allosteric inhibitor of cytochrome-c oxidase. European Journal of Biochemistry 249 350-354.

Azcoitia I, Garcia-Ovejero D, Chowen JA and Garcia-Segura LM (2001). Astroglia play a key role in the neuroprotective actions of estrogen. Progress in Brain Research 132 469-478.

Bains M, Cousins JC, Roberts JL (2007). Neuroprotection by estrogen against MPP+-induced dopamine neuron death is mediated by ERalpha in primary cultures of mouse mesencephalon. Experimental Neurology 204 767-776.

Bakeeva LE, Chentsov YS and Skulachev VP (1981). Ontogenesis of mitochondrial reticulum in rat diaphragm muscle. European Journal of Cell Biology 25 175-181.

Banerjee SK, Campbell DR, Weston AP, Banerjee DK (1997). Biphasic estrogen response on bovine adrenal medulla capillary endothelial cell adhesion, proliferation and tube formation. Molecular and Cellular Biochem- istry 177 97-105. 102 References

Barrientos A, Zambrano A, Tzagoloff A (2004). Mss51p and Cox14p jointly regulate mitochondrial Cox1p expression in Saccharomyces cerevisiae. EMBO Journal 23 3472-3482.

Beal MF (2005). Mitochondria take center stage in aging and neurodegeneration. Annals of Neurology 58 495-505.

Behl C (2003). Estrogen can protect neurons: modes of action. The Journal of Steroid Biochemistry and Molecular Biology 83 195-197.

Behl C (2002). Estrogen as a neuroprotective hormone. Nature Reviews Neuroscience 3 433-442.

Benard G, Faustin B, Passerieux E, Galinier A, Rocher C, Bellance N, Delage J, Casteilla L, Letellier T, Rossignol R (2006). Physiological diversity of mitochondrial oxidative phosphorylation. American Journal of Cell Physiology 291 C1172-C1182.

Beyer C (1999). Estrogen and the developing mammalian brain. Anatomy of Embryology 199 379-390.

Beyer C, Ivanova T, Karolczak M, Kuppers E (2002). Cell type-speciﬁcity of nonclassical estrogen signaling in the developing midbrain. The Journal of Steroid Biochemistry and Molecular Biology 81 319-325.

Beyer C, Pawlak J, Brito V, Karolczak M, Ivanova T, Kppers E (2003). Regulation of gene expression in the developing midbrain by estrogen. Annals of the New York Academy of Sciences 1007 17-28.

Beyer C, Pilgrim C and Reisert I (1991). Dopamine content and metabolism in mesencephalic and dien- cephalic cell cultures: sex differences and effects of sex steroids. Journal of Neuroscience 11 1325-1333.

Beyer C, Raab H (1998). Nongenomic effects of estrogen: embryonic mouse midbrain neurones respond with a rapid release of calcium from intracellular stores. European Journal of Neurosciences 10 255-262.

Benard G, Faustin B, Passerieux E, Galinier A, Rocher C, Bellance N, Delage JP, Casteilla L, Letellier T, Rossignol R (2006). Physiological diversity of mitochondrial oxidative phosphorylation. American Journal of Physiology - Cell Physiology 291 C1172-C1182.

Bettini E, Maggi A (1992). Estrogen induction of cytochrome c oxidase subunit III in rat hippocampus. Journal of Neurochemistry 58 1923-1929.

Bjorklund A, Lindvall O (1984). Dopamine-containing systems in the CNS. In: Bjorklund A, Hokfelt T, eds. Classical Transmitters in the CNS Amsterdam Elsevier 2 55-111.

Bleazard W, McCaffery JM, King EJ, Bale S, Mozdy A, Tieu Q, Nunnari J and Shaw JM (1999). The dynamin- related GTPase Dnm1 regulates mitochondrial ﬁssion in yeast. Nature Cell Biology 1 298-304.

Breckenridge DG, Stojanovic M, Marcellus RC and Shore GC (2003). Caspase cleavage product of BAP31 induces mitochondrial ﬁssion through endoplasmic reticulum calcium signals, enhancing cytochrome c release to the cytosol. Journal of Cell Biology 160 1115-1127. 103

Carroll RS, Brown M, Zhang J, DiRenzo J, Font De Mora J, Black PM (2000). Expression of a subset of steroid receptor cofactors is associated with progesterone receptor expression in meningiomas. Clinical Cancer Research 6 3570-3575.

Chan DC (2007). Mitochondrial dynamics in disease. New England Journal of Medicine 356 1707-1709.

Chen H and Chan DC (2005). Emerging functions of mammalian mitochondrial fusion and ﬁssion. Human Molecular Genetics 14 R283-R289.

Chen H, Chomyn A and Chan DC (2005). Disruption of fusion results in mitochondrial heterogeneity and dysfunction. Journal of Biological Chemistry 280 26185-26192.

Chen H, Detmer SA, Ewald AJ, Grifﬁn EE, Fraser SE and Chan DC (2003). Mitofusins Mfn1 and Mfn2 coordinately regulate mitochondrial fusion and are essential for embryonic development. Journal of Cell Biology 160 189-200.

Chen H, McCaffery JM and Chan DC (2007). Mitochondrial fusion protects against neurodegeneration in the cerebellum. Cell 130 548-562.

Chen J, Delannoy M, Odwin S, He P, Trush MA, Yager JD (2003). Enhanced mitochondrial gene transcript, ATP, Bcl2 protein levels, and altered glutathione distribution in ethinyl estradiol-treated cultured female rat hepatocytes. Toxicological Sciences 75 271-278.

Chen JQ, Delannoy M, Cooke C, Yager JD (2004). Mitochondrial localization of ERalpha and ERbeta in human MCF7 cells. American Journal of Physiology - Endocrinology and Metabolism 286 E1011-E1022.

Chen KH, Guo X, Ma D, Guo Y, Li Q, Yang D, Li P, Qiu X, Wen S, Xiao RP & Tang J (2004). Dysregulation of HSG triggers vascular proliferative disorders. Nature Cell Biology 6 872-883.

Cipolat S, Martins de Brito O, Dal Zilio B and Scorrano L (2004). OPA1 requires mitofusin 1 to promote mitochondrial fusion. Proceedings of the National Academy of Science U. S. A. 101 15927-15932.

Collins P, Webb C (1999). Estrogen hits the surface. Nature Medicine 5 1130-1131.

Cui W, Allen ND, Skynner M, Gusterson B and Clark AJ (2001). Inducible ablation of astrocytes shows that these cells are required for neuronal survival in the adult brain. Glia 34 272-282.

Culmsee C, Vedder H, Ravati A, Junker V, Otto D, Ahlemeyer B, Krieg JC, Krieglstein J (1999). Neuropro- tection by estrogens in a mouse model of focal cerebral ischemia and in cultured neurons: evidence for a receptor-independent antioxidative mechanism. Journal of Cerebral Blood Flow and Metabolism 19 1263-1269.

Daniel JM (2006). Effects of estrogen on cognition: what have we learned from basic research? Journal of Neuroendocrinology 18 787-795.

Delettre C, Lenaers G, Griffoin JM, Gigarel N, Lorenzo C, Belenguer P, Pelloquin L, Grosgeorge J, Turc-Carel 104 References

C, Perret E, Astarie-Dequeker C, Lasquellec L, Arnaud B, Ducommun B, Kaplan J, Hamel CP (2000). Nuclear gene OPA1, encoding a mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nature Genetics 26 207-210.

Demonacos CV, Karayanni N, Hatzoglou E, Tsiriyiotis C, Spandidos DA, Sekeris CE (1996). Mitochondrial genes as sites of primary action of steroid hormones. Steroids 61 226-232.

Detmer SA and Chan DC (2007). Complementation between mouse Mfn1 and Mfn2 protects mitochondrial fusion defects caused by CMT2A disease mutations. Journal of Cell Biology 176 405-414.

Detmer SA and Chan DC (2007). Functions and dysfunctions of mitochondrial dynamics. Nature Reviews of Molecular Cell Biology 8 870-879.

Dhandapani KM, Brann DW (2002). Protective effects of estrogen and selective estrogen receptor modulators in the brain. Biology of Reproduction 67 1379-1385.

Dhandapani KM, Brann DW (2007). Role of astrocytes in estrogen-mediated neuroprotection. Experimental Gerontology 42 70-75.

Dluzen DE and McDermott JL (2004). Developmental and genetic inﬂuences upon gender differences in methamphetamine-induced nigrostriatal dopaminergic neurotoxicity. Annals of the New York Academy of Sciences 1025 205-220.

Erecinska M, Cherian S and Silver IA (2004). Energy metabolism in mammalian brain during development. Progress in Neurobiology 73 397-445.

Frank S, Gaume B, Bergmann-Leitner ES, Leitner WW, Robert EG, Catez F, Smith CL and Youle RJ (2001). The role of dynamin-related protein 1, a mediator of mitochondrial ﬁssion, in apoptosis. Developmental Cell 1 515-525.

Garcia-Segura LM, Azcoitia I, DonCarlos LL. (2001). Neuroprotection by estrogen. Progress in Neurobiology 63 29-60.

Garcia-Segura LM and Melcangi RC (2006). Steroids and glial cell function. Glia 54 485-498.

Garcia-Segura LM, Veiga S, Sierra A, Melcangi RC and Azcoitia I (2003). Aromatase: a neuroprotective enzyme. Progress in Neurobiology 71 31-41.

Gavrilova-Jordan LP and Price TM (2007). Actions of steroids in mitochondria. Seminars in Reproductive Medicine 25 154-64.

Gillies GE, Murray HE, Dexter D and McArthur S (2004). Sex dimorphisms in the neuroprotective effects of estrogen in an animal model of Parkinsons disease. Pharmacology Biochemistry and Behaviour 78 513-522.

Gnaiger E, Mendez G, Hand SC (2000). High phosphorylation efﬁciency and depression of uncoupled 105 respiration in mitochondria under hypoxia. Proceedings of the National Academy of Sciences USA 97 11080-11085.

Greenamyre JT, MacKenzie G, Peng TI, Stephans SE (1999). Mitochondrial dysfunction in Parkinson’s disease. Biochemical Society Symposia 66 85-97.

Gregson NA and Williams PL (1969). A comparative study of brain and liver mitochondria from new-born and adult rats. Journal of Neurochemistry 16 617-626.

Griparic L, van der Wel NN, Orozco IJ, Peters PJ and van der Bliek AM (2004). Loss of the intermembrane space protein Mgm1/OPA1 induces swelling and localized constrictions along the lengths of mitochondria. Journal of Biological Chemistry 279 18792-18798.

Hafner RP, Brand MD (1988). Hypothyroidism in rats does not lower mitochondrial ADP/O and H+/O ratios. Biochemistry Journal 250 477-484.

Harms C, Lautenschlager M, Bergk A, Katchanov J, Freyer D, Kapinya K, Herwig U, Megow D, Dirnagl U, Weber JR, Hortnagl H (2001). Differential mechanisms of neuroprotection by 17 beta-estradiol in apoptotic versus necrotic neurodegeneration. Journal of Neuroscience 21 2600-2609.

Horvat S, Beyer C and Arnold S (2006). Effect of hypoxia on the transcription pattern of subunit isoforms and the kinetics of cytochrome c oxidase in cortical astrocytes and cerebellar neurons. Journal of Neurochemistry 99 937-951.

Hsieh YC, Yu HP, Suzuki T, Choudhry MA, Schwacha MG, Bland KI, Chaudry IH (2006). Upregulation of mitochondrial respiratory complex IV by estrogen receptor-beta is critical for inhibiting mitochondrial apoptotic signaling and restoring cardiac functions following trauma-hemorrhage. Journal of Molecular and Cellular Cardiology 41 511-521.

Hutchison JB, Byer C, Hutchison RE, Wozniak A (1997). Sex differences in the regulation of embryonic brain aromatase. The Journal of Steroid Biochemistry and Molecular Biology 61 315-22.

Irwin RW, Yao J, Hamilton RT, Cadenas E, Brinton RD, Nilsen J (2008). Progesterone and estrogen regulate oxidative metabolism in brain mitochondria. Endocrinology 149 3167-3175.

Ivanova T, Mendez P, Garcia-Segura LM and Beyer C (2002). Rapid stimulation of the PI3-kinase/Akt signalling pathway in developing midbrain neurones by estrogen. Journal of Neuroendocrinology 14 73-79.

James DI, Parone PA, Mattenberger Y and Martinou JC (2003). hFis1, a novel component of the mammalian mitochondrial ﬁssion machinery. Journal of Biological Chemistry 278 36373-36379.

Jung-Testas I, Baulieu EE (1998). Steroid hormone receptors and steroid action in rat glial cells of the central and peripheral nervous system. The Journal of Steroid Biochemistry and Molecular Biology 65 243-251.

Jung Testas I, Renoir M, Bugnard H, Greene Gl and Baulien EE (1992). Demonstration of steroid hormone receptors and steroid action in primary cultures of rat glial cells. Journal of Steroid Biochemistry and Molecular 106 References

Biology 41 621-631.

Kadenbach B, Arnold S, Lee I, Huttemann M (2004). The possible role of cytochrome c oxidase in stress- induced apoptosis and degenerative diseases. Biochimica Biophysica Acta 1655 400-408.

Kajta M and Beyer C (2003). Cellular strategies of estrogen-mediated neuroprotection during brain development. Endocrine 21 3-9.

Karbowski M and Youle RJ (2003). Dynamics of mitochondrial morphology in healthy cells and during apoptosis. Cell Death Differentiation 10 870-880.

Kenchappa RS, Diwakar L, Annepu J, Ravindranath V (2004). Estrogen and neuroprotection: higher constitu- tive expression of glutaredoxin in female mice offers protection against MPTP-mediated neurodegeneration. FASEB Journal 18 1102-1104.

Kijima K, Numakura C, Izumino H, Umetsu K, Nezu A, Shiiki T, Ogawa M, Ishizaki Y, Kitamura T, Shozawa Y and Hayasaka K (2005). Mitochondrial GTPase mitofusin 2 mutation in Charcot-Marie-Tooth neuropathy type 2A. Human Genetics 116 23-27.

Kipp M, Karakaya S, Pawlak J, Araujo-Wright G, Arnold S and Beyer C (2006). Estrogen and the development and protection of nigrostriatal dopaminergic neurons: concerted action of a multitude of signals, protective molecules, and growth factors. Frontiers of Neuroendocrinology 27 376-390.

Kipp JL, Ramirez VD (2001). Effect of estradiol, diethylstilbestrol, and resveratrol on F0F1-ATPase activity from mitochondrial preparations of rat heart, liver, and brain. Endocrine 15 165-175.

Kristin T, Hopkins IB, McKenna MC, Fiskum G (2006). Isolation of mitochondria with high respiratory control from primary cultures of neurons and astrocytes using nitrogen cavitation. Journal of Neuroscience Methods 152 136-143.

Kuznetsov AV, Troppmair J, Sucher R, Hermann M, Saks V and Margreiter R (2006). Mitochondrial subpopu- lations and heterogeneity revealed by confocal imaging: possible physiological role? Biochimica et Biophysica Acta 1757 686-691.

Labrousse AM, Zapaterra M, Rube DA and van der Bliek AM (1999). C. elegans dynamin-related protein drp-1 controls severing of the mitochondrial outer membrane. Molecular Cell 4 815-826.

Law SW, Apostolakis EM, Samora PJ, O’Malley BW, Clark JH (1994). Hormonal regulation of hypothalamic gene expression: identiﬁcation of multiple novel estrogen induced genes. The Journal of Steroid Biochemistry and Molecular Biology 51 131-136.

Lee YJ, Jeong SY, Karbowski M, Smith CL and Youle RJ (2004). Roles of the mammalian mitochondrial ﬁssion and fusion mediators Fis1, Drp1, and Opa1 in apoptosis. Molecular Biology of the Cell 15 5001-5011.

Leonelli E, Bianchi R, Cavaletti G, Caruso D, Crippa D, Garcia-Segura LM, Lauria G, Magnaghi V, Roglio 107

I and Melcangi RC (2007). Progesterone and its derivatives are neuroprotective agents in experimental diabetic neuropathy: a multimodal analysis. Neuroscience 144 1293-1304.

Luconi M, Muratori M, Forti G and Baldi E (1999). Identiﬁcation and characterization of a novel functional estrogen receptor on human sperm membrane that interferes with progesterone effects. The Journal of Clinical Endocrinology and Metabolism 84 1670-1678.

Mannella CA (2000). Introduction: our changing views of mitochondria. Journal of Bioenergetics and Biomem- branes 32 1-4.

Maragakis NJ, Rothstein JD (2006). Mechanisms of disease: astrocytes in neurodegenerative disease. Nature Clinical Practice Neurology Vol. 2 12 679-689.

Marchetti B, Serra PA, L’Episcopo F, Tirolo C, Caniglia S, Testa N, Cioni S, Gennuso F, Rocchitta G, Desole MS, Mazzarino MC, Miele E and Morale MC (2005). Hormones are key actors in gene x environment interactions programming the vulnerability to Parkinson’s disease: glia as a common ﬁnal pathway. Annals of the New York Academy of Sciences 1057 296-318.

McCarthy MM (2008). Estradiol and the developing brain. Physiological Reviews 88 91-124.

McEwen BS (1992). Steroid hormones: effect on brain development and function. Hormone Research 37 1-10.

Mendez P, Azcoitia I, Garcia-Segura LM (2005). Interdependence of estrogen and insulin-like growth factor-I in the brain: potential for analysing neuroprotective mechanisms. Journal of Endocrinology 185 11-17.

Mesano S and Jaffe RB (1997). Developmental and functional biology of the primate fetal adrenal cortex. Endocrine Reviews 18 378-403.

Miller DB, Ali SF, O’Callaghan JP, Laws SC (1998). The impact of gender and estrogen on striatal dopaminergic neurotoxicity. Annals of the New York Academy of Sciences 844 153-165.

Milner TA, Lubbers LS, Alves SE and McEwen BS (2008). Nuclear and extranuclear estrogen binding sites in the rat forebrain and autonomic medullary areas. Endocrinology 149(7) 3306-3312.

Mitra SW, Hoskin E, Yudkovitz J, Pear L, Wilkinson HA, Hayashi S, Pfaff DW, Ogawa S, Rohrer SP, Schaeffer JM, McEwen BS, and Alves SE (2003). Immunolocalization of estrogen receptor beta in the mouse brain: comparison with estrogen receptor alpha. Endocrinology 144 2055-2067.

Monje P and Boland R (2001). Subcellular distribution of native estrogen receptor alpha and beta isoforms in rabbit uterus and ovary. Journal of Cellular Biochemistry 82 467-479.

Moosmann B, Behl C (1999). The antioxidant neuroprotective effects of estrogens and phenolic compounds are independent from their estrogenic properties. Proceedings of the National Academy of Sciences USA 96 8867-8872. 108 References

Nakada K, Inoue K, Ono T, Isobe K, Ogura A, Goto YI, Nonaka I and Hayashi JI (2001). Inter-mitochondrial complementation: mitochondria-speciﬁc system preventing mice from expression of disease phenotypes by mutant mtDNA. Nature Medicine 7 934-940.

Neuspiel M, Zunino R, Gangaraju S, Rippstein, P and McBride H (2005). Activated Mitofusin 2 signals mitochondrial fusion, interferes with Bax activation, and reduces susceptibility to radical induced depolarization. The Journal of Biological Chemistry 280 25060-25070.

Nilsen J and Brinton RD (2004). Mitochondria as therapeutic targets of estrogen action in the central nervous system. Current Drug Targets CNS Neurological Disorders 3 297-313.

Nilson J and Brinton RD (2003). Mechanism of estrogen-mediated neuroprotection: Regulation of mitochondrial calcium and Bcl2 expression. PNAS 100 2842-2647.

Olichon A, Baricault L, Gas N, Guillou E, Valette A, Belenguer P and Lenaers G (2003). Loss of OPA1 perturbates the mitochondrial inner membrane structure and integrity, leading to cytochrome c release and apoptosis. Journal of Biological Chemistry 278 7743-7746.

Omulecka A, Papierz W, Nawrocka-Kunecka A and Lewy-Trenda I (2006). Immunohistochemical expression of progesterone and estrogen receptors in meningiomas. Folia Neuropathologica 44 111-115.

Pawlak J, Karolczak M, Krust A, Chambon P and Beyer C (2005). Estrogen receptor-alpha is associated with the plasma membrane of astrocytes and coupled to the MAP/Src-kinase pathway. Glia 50 270-275.

Pellerin L, Bouzier-Sore AK, Aubert A, Serres S, Merle M, Costalat R and Magistretti PJ (2007). Activity- dependent regulation of energy metabolism by astrocytes: an update. Glia 55 1251-1262.

Peri A Serio M (2008). Estrogen receptor-mediated neuroprotection: The role of the Alzheimer’s disease- related gene seladin-1. Neuropsychiatric Disease and Treatment 4 817-824.

Poole AC, Thomas RE, Andrews LA, McBride HM, Whitworth AJ and Pallanck LJ (2008). The Pink1/Parkin pathway regulates mitochondrial morphology. Proceedings of the National Academy of Science USA 105 1638-1643.

Prange-Kiel J, Fester L, Zhou L, Lauke H, Carrtero J, Rune GM (2006). Inhibition of hippocampal estrogen synthesis causes region-speciﬁc downregulation of synaptic protein expression in hippocampal neurons. Hippocampus 16 464-471.

Pysh JJ (1970). Mitochondrial changes in rat inferior colliculus during postnatal development: an electron microscopic study. Brain Research 18 325-342.

Quadros PS, Pfau JL and Wagner CK (2007). Distribution of progesterone receptor immunoreactivity in the fetal and neonatal rat forebrain. Journal of Comparative Neurology 504 42-56.

Rabinowicz T, Dean DE, Petetot JM, de Courten-Myers GM (1999). Gender differences in the human cerebral 109 cortex: more neurons in males; more processes in females. Journal of Child Neurology 14(2) 98-107.

Reynolds IJ (1999). Mitochondrial membrane potential and the permeability transition in excitotoxicity. Annals of the New York Academy of Sciences 893 33-41.

Rojo M, Legros F, Chateau D and Lombes A (2002). Membrane topology and mitochondrial targeting of mitofusins, ubiquitous mammalian homologs of the transmembrane GTPase Fzo. Journal of Cell Science 115 1663-1674.

Roof RL and Hall ED (2000). Gender differences in acute CNS trauma and stroke: neuroprotective effects of estrogen and progesterone.Journal of Neurotrauma 17 367-388.

Rune GM, Lohse C, Prange-Kiel J, Fester L and Frotscher M (2006). Synaptic plasticity in the hippocampus: effects of estrogen from the gonads or hippocampus? Neurochemical Research 31 145-155.

Samson Jr FE, Balfour WM and Jacobs RJ (1960). Mitochondrial changes in developing rat brain. American Journal of Physiology 199 693-696.

Santagati S, Melcangi RC, Celotti F, Martini L and Maggi A (1994). Estrogen receptor is expressed in different types of glial cells in culture. Journal of Neurochemistry 63 2058-2064.

Santel A and Fuller MT (2001). Control of mitochondrial morphology by a human mitofusin. Journal of Cell Science 114 867-874.

Santel A, Frank S, Gaume B, Herrler M, Youle RJ, Fuller MT (2003). Mitofusin-1 protein is a generally expressed mediator of mitochondrial fusion in mammalian cells. Journal of Cell Science 116 2763-2774.

Schapira AH (2008). Mitochondria in the aetiology and pathogenesis of Parkinson’s disease. The Lancet Neurology 7 97-109.

Schlaepfer TE, Harris GJ, Tien AY, Peng L, Lee S, Pearlson GD (1995). Structural differences in the cerebral cortex of healthy female and male subjects: a magnetic resonance imaging study. Psychiatry Research 61(3) 129-35.

Schousboe A, Bak LK, Sickmann HM, Sonnewald U and Waagepetersen HS (2007). Energy substrates to support glutamatergic and GABAergic synaptic function: role of glycogen, glucose and lactate. Neurotoxicity Research 12 263-268.

Schulman LM (2007). Gender differences in Parkinsons disease. Gender Medicine 4 8-18.

Schumacher M, Weill-Engerer S, Liere P, Robert F, Franklin RJM, Garcia-Segura LM, Lambert JJ, Mayo W, Melcangi RC, Parducz A, Suter U, Carelli C, Baulieu EE, Akwa Y (2003). Steroid hormones and neurosteroids in normal and pathological aging of the nervous system. Progress in Neurobiology 71 3-29.

Simpkins JW, Yang SH, Wen Y and Singh M (2005). Estrogens, progestins, menopause and neurodegeneration: 110 References

basic and clinical studies. Cellular and Molecular Life Sciences 62 271-280.

Singh M, Dykens JA and Simpkins JW (2006). Novel mechanisms for estrogen-induced neuroprotection. Experimental Biology and Medicine (Maywood) 231 514-521.

Smirnova E, Griparic L, Shurland DL and van der Bliek AM (2001). Dynamin-related protein Drp1 is required for mitochondrial division in mammalian cells. Molecular Biology of the Cell 12 2245-2256.

Stein DG, Wright DW and Kellermann AL (2008). Does progesterone have neuroprotective properties? Annals of Emergency Medicine 51 164-172.

Stirone C, Duckles SP, Krause DN, Procaccio V (2005). Estrogen increases mitochondrial efﬁciency and reduces oxidative stress in cerebral blood vessels. Molecular Pharmacology 68 959-965.

Stojanovski D, Koutsopoulos OS, Okamoto K and Ryan MT (2004). Levels of human Fis1 at the mitochondrial outer membrane regulate mitochondrial morphology. Journal of Cell Science 117 1201-1210.

Strausberg RL, Butow RA (1977). Expression of petite mitochondrial DNA in vivo: Zygotic gene rescue. Proceedings of the National Academy of Sciences USA 74 2715-2719.

Sugioka R, Shimizu S and Tsujimoto Y (2004). Fzo1, a protein involved in mitochondrial fusion, inhibits apoptosis. Journal of Biological Chemistry 279 52726-52734.

Sugishita K, Li F, Su Z, Barry WH (2003). Anti-oxidant effects of estrogen reduce [Ca2+]i during metabolic inhibition. Journal of Molecular and Cellular Cardiology 35 331-336.

Taguchi N, Ishihara N, Jofuku A, Oka T and Mihara K (2007). Mitotic phosphorylation of dynamin-related GTPase Drp1 participates in mitochondrial ﬁssion. Journal of Biological Chemistry 282 11521-11529.

Tatsuta T and Langer T (2008). Focus Quality Control. Quality control of mitochondria: protection against neurodegeneration and ageing. The EMBO Journal 27 306-314.

Thomas B and Beal MF (2007). Parkinson’s disease. Human Molecular Genetics 16 R183-R194.

Trotter A, Maier L, Grill HJ, Kohn T, Heckmann M, Pohlandt F (1999). Effects of postnatal estradiol and progesterone replacement in extremely preterm infants. The Journal of Clinical Endocrinology and Metabolism 84 4531-4535.

Von Schassen C, Fester L, Prange-Kiel J, Lohse C, Huber C, Bottner M and Rune GM (2006). Estrogen synthesis in the hippocampus: role in axon outgrowth. Journal of Neuroendocrinology 18 847-856.

Weissig V, Cheng SM, D’Souza GG (2004). Mitochondrial pharmaceutics. Mitochondrion 3 229-44.

Yaffe MP (1999). The machinery of mitochondrial inheritance and behaviour. Science 283 1493-1497. 111

Yager JD and Davidson NE (2006). Estrogen carcinogenesis in breast cancer. New England Journal of Medicine 354 270-282.

Yager JD and Chen JQ (2007). Mitochondrial estrogen receptors–new insights into speciﬁc functions. Trends in Endocrinology and Metabolism 18 89-91.

Yang SH, Liu R, Perez EJ, Wen Y, Stevens SM Jr, Valencia T, Brun-Zinkernagel AM, Prokai L, Will Y, Dykens J, Koulen P and Simpkins JW (2004). Mitochondrial localization of estrogen receptor beta. Proceedings of the National Academy of Science USA 101 4130-4135.

Yoon Y, Krueger EW, Oswald BJ and McNiven MA (2003). The mitochondrial protein hFis1 regulates mitochondrial ﬁssion in mammalian cells through an interaction with the dynamin-like protein DLP1. Molecular Cell Biology 23 5409-5420.

Zeviani M and Donato S D (2004). Mitochondrial disorders. Brain 127 2153-2172.

Zhang JQ, Cai WQ, Zhou de S, and Su BY (2002). Distribution and differences of estrogen receptor beta immunoreactivity in the brain of adult male and female rats. Brain Research 935 73-80.

Zheng J and Ramirez VD (1999). Rapid inhibition of rat brain mitochondrial proton F0F1-ATPase activity by estrogens: comparison with Na+, K+ -ATPase of porcine cortex. European Journal of Pharmacology 368 95-102.

Zuchner S, Mersiyanova IV, Muglia M, Bissar-Tadmouri N, Rochelle J, Dadali EL, Zappia M, Nelis E, Patitucci A, Senderek J, Parman Y, Evgrafov O, Jonghe PD, Takahashi Y, Tsuji S, Pericak-Vance MA, Quattrone A, Battologlu E, Polyakov AV, Timmerman V, Schroder JM and Vance JM (2004). Mutations in the mitochondrial GTPase mitofusin 2 cause Charcot-Marie-Tooth neuropathy type 2A. Nature Genetics 36 449-451. 112 References Contributions

113

115

Article: Araujo GW, Beyer C and Arnold S (2008). Estrogen inﬂuences on mitochondrial gene expression and respiratory chain activity in cortical and mesencephalic astrocytes. Journal of Neuroendocrinology 30, 930-941. Contributions: Araujo GW: project conception, gene selection, primer design, all experiments, preparation of the manuscript, references, review discussion. Beyer C: cell culturing, conception of the study, preparation of the manuscript, grant holder Arnold S: conception of study, preparation of manuscript, grant holder

Article: Arnold S, Wright Araujo G and Beyer C (2008) Gender-speciﬁc regulation of mitochondrial fusion and ﬁssion gene transcription and viability of cortical astrocytes by steroid hormones. Journal of Molecular Endocrinology 41, 289-300. Contributions: Arnold S: quantitative PCR experiments, cell culturing, project conception, manuscript preparation, grant holder Wright Araujo G: project conception, gene selection, primer design, semi-quantitative PCR and cell survival experiments, manuscript preparation, references. Beyer C: cell culturing, project conception, manuscript preparation, grant holder

This work was supported by the START-Programm and IZKF BIOMAT of the Faculty of Medicine, RWTH Aachen, and the Emmy-Noether-Programm of the Deutsche Forschungsgemeinschaft. 116 Contributions Acknowledgements

117

119

Academics

I would ﬁrstly like to thank Prof. Dr. C. Beyer, for the opportunity to make my PhD in his Neuroanatomy group, for the liberty given to choose and conduct my projects and the correction of the publications including this work. Special thanks to Prof. Dr. H. Wagner for agreeing to be my second evaluator, to Prof. Dr. J. Bohrmann, for being my third evaluator, and Prof. Dr. W. Baumgartner, for presiding over the PhD Defence. Many thanks to my dear colleagues S. Singh and J. Pawlak for the constructive scientiﬁc discussions and the collegial lab work. Thank you to U. Zahn and P. Ibold for the technical lab support. Thank you very much D. Alves for the electron microscopy analysis and J. Mauler for correcting the translation of the summary. Special thanks to Prof. Dr. H. Herbert of the Max-Planck Graduate School of Neural and Behavioural Science in Tubingen,¨ for his support. Thanks to the colleague PhD and graduate ”neuro-students” for the interesting seminars and fun get-togethers. Special thanks to my previous research team in Tubingen,¨ the collaborating groups of Dr. M. Simons and Dr. R. Weissert, especially to my colleagues S. Barth and M. Herrman, for the laboratory techniques teachings so helpful for my PhD, and also to the whole teams for the great collegial atmosphere. I’d also like to especially thank Dr. T. Bangsow from ESPLORA GmbH for igniting my interest for neuroscience and for the valuable teachings of molecular biology techniques, very helpful for my PhD.

Family and Friends

Very special thanks to my parents, whom I also dedicate this PhD, for the love and support and specially for the trust they have always given me. Very special thanks also to my sisters and brother, and my whole family, for their support and care. Very special thanks to S. Karunakaran for the essential support in the last, more demanding PhD years and the motivation to complete the task! I would like to heartily thank all my friends and all the people that in one way or the other motivated and helped me during the PhD years. I’d like to specially mention A. Kiran, N. Santos, K. Molina-Luna and D. Alves, and also S. Hofer¨ for the last motivating push. Thank you! 120 Acknowledgements Curriculum vitae

121

123

Personal data

Name Gilda Catarina de Castro Lopo Wright de Araujo Address Aeussere Kanalstr. 81, 50827 Cologne, Germany Phone +49 176 24163424 Email [email protected] Date, place of birth 02 Jan 1979, Chaves, Portugal

Work Experience

01/09 - present Clinical Research Associate, Aptiv Solutions GmbH, Cologne, Germany 10/05 - 12/07 Research Scientist and Teaching Assistant in Microscopic Anatomy\Histology, Neuroanatomy Institute, University Clinics, Aachen, Germany. 10/04 - 09/05 Research Scientist, Anatomy Institute, University Clinics, Tubingen,¨ Ger- many. 01/03 - 04/04 Research Scientist, University Clinics, Tubingen,¨ Germany. 09/02 - 11/02 Research Scientist, ESPLORA GmbH, Darmstadt, Germany. 04/02 - 05/02 Research Technician, University Clinics, Ulm, Germany

Education

06/08 -11/08 Qualiﬁcation course ”Clinical Research Associate/ Data Manager”, mibeg- Institut Medizin, Cologne, Germany 10/05-05/08 PhD Student, ”RWTH” University of Aachen, Neuroanatomy Institute of the University Clinics, Aachen, Germany 10/06 DAAD Interchange Program for PhD Students, Pharmacology Institute, Academy of sciences, University of Cracow, Poland 01/03 - 09/05 PhD Student, Graduate School of Neural and Behavioural Sciences, Max- Planck Institute and Eberhard Karls University of Tubingen,¨ Tubingen,¨ Germany 09/97 - 10/01 Graduation: Applied Biology (Biologia Aplicada) Universidade do Minho, Braga, Portugal. Masters Project, University of Luton, Luton, England 09/91-07/97 Graduation: Sciences; High School ”Escola Secundaria de Carlos Ama- rante”, Braga, Portugal 124 Curriculum vitae

Publications

2008 ”Oestrogen regulates mitochondrial gene expression and respiratory chain activity in cortical and mesencephalic astrocytes” Gilda Wright Araujo, Cordian Beyer, Susanne Arnold. Journal of Neuroendocrinology 2008 Jul;20(7):930-41. Epub 2008 Apr 28. 2008 ”Gender-speciﬁc regulation of mitochondrial fusion and ﬁssion gene transcription and viability of cortical astrocytes by steroid hormones” Susanne Arnold, Gilda Wright Araujo, Cordian Beyer. J. Mol. Endocrinology. 2008 Nov;41(5):289-300. Epub 2008 Aug 27. 2006 ”Estrogen and the development and protection of nigrostriatal dopaminergic neurons: concerted action of a multitude of signals, protective molecules, and growth factors” Markus Kipp*, Serkan Karakaya*, Justyna Pawlak, Gilda Araujo-Wright, Susanne Arnold, Cordian Beyer. Frontiers in Neu- roendocrinology 2006, 27:376-390, Epub 2006 Sep 1 *contributed equally 2004 ”Hyperphosphorylation and aggregation of tau in experimental autoim- mune encephalomyelitis” Anja Schneider*, Gilda Wright Araujo*, Katarina Trajkovic, Martin M. Herrmann, Doron Merkler, Eva-Maria Mandelkow, Robert Weissert, Mikael Simons*. J Biol Chem. 2004 Dec 31;279(53):55833-9. Epub 2004 Oct 1 *contributed equally