Metal homeostasis as critical determinant for cellular fitness

Metal homeostasis as critical determinant for cellular fitness

Jutta Diessl    Metal homeostasis as critical determinant for cellular fitness

Department of Molecular Biosciences, The Wenner-Gren Institute

ISBN 978-91-7911-512-8

The moment that turned this thesis on its head.

Jutta Diessl preparing a genome- wide screen to identify targets of manganese toxicity in the yeast Saccharomyces cerevisae.

Metal homeostasis as critical determinant for cellular fitness

Jutta Diessl

Academic dissertation for the Degree of Doctor of Philosophy in Molecular Bioscience at Stockholm University to be publicly defended on Friday 11 June 2021 at 14.00 in Vivi Täckholmsalen (Q-salen), NPQ-huset, Svante Arrhenius väg 20, online via Zoom, public link https://stockholmuniversity.zoom.us/j/68896819500.

Abstract

Metals play a crucial role in cellular biology. Bulk and trace metals such as calcium and manganese regulate a plethora of cellular processes ranging from signaling and oxidative stress to proteostasis and energy metabolism. Fine-tuning metal levels and distribution safeguards all forms of life from compromised cellular fitness and cell death elicited by metal deficiency or overload. However, the underlying molecular mechanisms eventually leading to cellular demise remain elusive. In this thesis, we studied the fundamental impact of disrupted metal homeostasis on cellular survival focusing on mitochondrial and lysosomal processes in Saccharomyces cerevisiae and Drosophila melanogaster. In Paper I, we establish Coenzyme Q (CoQ) biosynthesis in mitochondria as the prime target of cellular manganese overload and propose a molecular mechanism underlying manganese toxicity. Combining proteomics, genome-wide screening and comprehensive metal analyses, we identify mismetallation of the di-iron hydroxylase Coq7, an enzyme of CoQ biosynthesis, as cause for the CoQ deficiency upon manganese overload. Overexpression of Coq7 not only restored CoQ-mediated electron transport through the respiratory chain but also prevented age-associated death. Expanding from trace to bulk metals, we further assessed the impact of disrupted calciumand manganese homeostasis on cellular survival. In Paper II, we created a fluorescence-based reporter system for the Ca²⁺/calmodulin-dependent phosphatase calcineurin, a nexus for cell stress-induced signaling. Combining our reporters with a live/dead staining allows for quantification of acute and chronic changes in calcium signaling in living, unperturbed cells. In Paper III, we elucidate the impact of nutritional regimes known to improve cellular survival on cells compromised in the handling of calcium and manganese due to the absence of Pmr1, a Ca²⁺/Mn²⁺ ATPase of the secretory pathway. We demonstrate that caloric restriction prevents manganese- induced disruption of mitochondrial energy metabolism and improves survival independent of calcineurin activity and CoQ biosynthesis. In Papers IV and V, we studied the interplay of metal levels and calcium signaling in the context of neurodegeneration and report that calcineurin stimulates lysosomal proteolysis, thereby preventing proteotoxicity in yeast and Drosophila models for Parkinson’s disease. Collectively, our results provide new insights into the consequences of disrupted metal homeostasis for cellular fitness and unravel a novel link between manganese overload, mitochondrial bioenergetics and CoQ biosynthesis conserved across species.

Keywords: metal homeostasis, manganese toxicity, coenzyme Q synthesis, mitochondrial respiration, calcineurin signaling, Pmr1, SPCA1, calcium, survival, caloric restriction, Parkinson’s disease models, proteotoxicity, Pep4, cathepsin D.

Stockholm 2021

http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-191876

ISBN 978-91-7911-512-8 ISBN 978-91-7911-513-5

Department of Molecular Biosciences, The Wenner- Gren Institute

Stockholm University, 106 91 Stockholm

METAL HOMEOSTASIS AS CRITICAL DETERMINANT FOR CELLULAR FITNESS

 Jutta Diessl

Metal homeostasis as critical determinant for cellular fitness

Jutta Diessl

©Jutta Diessl, Stockholm University 2021

 ISBN print 978-91-7911-512-8 ISBN PDF 978-91-7911-513-5

 Printed in Sweden by Universitetsservice US-AB, Stockholm 2021

"Life is a mystery."

Madonna

List of papers

I J. Diessl, J. Berndtsson, F. Broeskamp, L. Habernig, V. Kohler, C. Vazquez-Calvo, A.

Nandy, C. Peselj, S.P. Drobysheva, L. Pelosi, F.N. Vögtle, F. Pierrel, M. Ott, S. Büttner, Manganese overload disrupts mitochondrial energy metabolism via inhibition of Coen- zyme Q biosynthesis, (manuscript).

II J. Diessl, A. Nandy, C. Schug, L. Habernig, S. Büttner, Stable and destabilized GFP report- ers to monitor calcineurin activity in Saccharomyces cerevisiae, Microb Cell. 7 (2020) 106–114. https://doi.org/10.15698/mic2020.04.713.

III J. Diessl, C. Prado Morales, L. Habernig, F. Pierrel, S. Büttner, Caloric restriction prevents manganese-induced disruption of mitochondrial bioenergetics, (manuscript).

IV A. Aufschnaiter, L. Habernig, V. Kohler, J. Diessl, D. Carmona-Gutierrez, T. Eisenberg, W.

Keller, S. Büttner, The Coordinated Action of Calcineurin and Cathepsin D Protects Against α-Synuclein Toxicity, Front Mol Neurosci. 10 (2017).

https://doi.org/10.3389/fnmol.2017.00207.

V L. Habernig, Broeskamp, Filomena, A. Aufschnaiter, J. Diessl, C. Peselj, E. Urbauer, T.

Eisenberg, A. de Ory, S. Büttner, Ca²⁺ administration prevents α-Synuclein proteotoxicity by stimulating calcineurin-dependent lysosomal proteolysis, (manuscript).

Related publications (not included in this thesis)

I J. Berndtsson, A. Aufschnaiter, S. Rathore, L. Marin-Buera, H. Dawitz, J. Diessl, V. Kohler, A. Barrientos, S. Büttner, F. Fontanesi, M. Ott, Respiratory supercomplexes enhance electron transport by decreasing cytochrome c diffusion distance, EMBO Reports. n/a (2020) e51015. https://doi.org/10.15252/embr.202051015.

II A. Aufschnaiter, V. Kohler, J. Diessl, C. Peselj, D. Carmona-Gutierrez, W. Keller, S. Büttner, Mitochondrial lipids in neurodegeneration, Cell Tissue Res. (2016).

https://doi.org/10.1007/s00441-016-2463-1.

III P. Rockenfeller, M. Smolnig, J. Diessl, M. Bashir, V. Schmiedhofer, O. Knittelfelder, J. Ring, J. Franz, I. Foessl, M.J. Khan, R. Rost, W.F. Graier, G. Kroemer, A. Zimmermann, D. Car- mona-Gutierrez, T. Eisenberg, S. Büttner, S.J. Sigrist, R.P. Kühnlein, S.D. Kohlwein, C.W.

Gourlay, F. Madeo, Diacylglycerol triggers Rim101 pathway–dependent necrosis in yeast: a model for lipotoxicity, Cell Death & Differentiation. 25 (2018) 765–781.

https://doi.org/10.1038/s41418-017-0014-2.

Aims of thesis

Main question: How do calcium and manganese homeostasis impact cellular fitness involv- ing mitochondrial and lysosomal functions?

To answer this question, the present thesis was guided by the following aims:

• To investigate mitochondrial function upon cellular metal overload (Paper I)

• To develop a fluorescence-based reporter system for Ca²⁺ signaling throughout growth and aging using calcineurin activity as a readout (Paper II)

• To characterize the effect of dietary interventions on cellular fitness upon cellular metal overload (Paper III)

• To study the interplay of Ca²⁺ homeostasis, lysosomal proteolysis and α-synuclein toxicity (Paper IV+V)

Popular summary

Metals play a crucial role in the processes that happen within a cell. Bulk and trace metals, which occur in high or low amounts, for example, calcium and manganese, regulate pro- cesses such as communication, coping with stress and energy household. Carefully calibrat- ing the metal levels protects all life forms from cell death, which can occur when dangerously high or low levels of metals are registered in a cell. How too much or too little of a certain metal causes cell death is not very well known. In this thesis, we studied how an imbalanced metal household affects the survival of yeast cells and flies. We focused on how imbalances in the metal household affect mitochondria, the powerhouses of the cell, and lysosomes, the recycling stations. In our first study, we find that mitochondria are especially sensitive to extremely high levels of manganese. A protein in the mitochondria, called Coq7, usually works with iron to produce its product coenzyme Q. However, when there is too much man- ganese in the mitochondria, the iron in Coq7 is replaced by manganese, causing the protein to produce less coenzyme Q. This product is an important part of the mitochondrial machin- ery that converts sugar into energy. So too much manganese impairs the function of mito- chondria and results in the cell suffering and eventually dying. In our other studies, we fo- cused on the role that calcium plays in the survival of cells. We first made a reporter that would allow us to know when the cell uses calcium to communicate in a specific way. The reporter gives a light signal when the cell uses calcium to activate the protein calcineurin.

Calcineurin is an important protein in all cells and can cause many changes, for example, it can help the cell to adapt to stress. In our third study, we used this reporter to find out how important calcium and calcineurin were for an aging yeast cell. We found that calcium is more important when cells are growing fast and that reducing growth by limiting the sugar source, like putting them on a diet, can actually be beneficial for aging when cells cannot regulate their calcium levels. In the last studies, we looked at how calcium and calcineurin are involved in a yeast model of Parkinson’s disease. Even though yeast does not have a brain, the region that is most affected by cell death in Parkinson’s disease, we can still use one of the proteins involved in Parkinson’s disease, α-synuclein, and see how yeast manages to survive with it. In our fourth study, we found that calcineurin is very important to help the cells’ recycling station, the lysosome, deal with damaged or old proteins when there is α- synuclein in the cell. In our fifth study, we use yeast cells and flies that have too much of the broken α-synuclein protein causing them to die. By simply giving them more calcium we can increase their activity of calcineurin which in turn helps them to recycle more proteins in their lysosomes. Combining all of these results we could demonstrate important aspects of how carefully calibrating metal levels is important for the survival of yeast and flies. As the cellular machinery affected by metals in yeast and flies is very similar to that of humans, our results are also important to understand how an imbalanced metal household can affect human cells and their survival.

Populärvetenskaplig sammanfattning

Metaller spelar en central roll i cellbiologiska processer. Förekomsten av metaller i den intra- cellulära miljon kan vara relativt hög, t.ex. kalcium, eller låg, t.ex. mangan. Båda kalcium och mangan är dock nödvändiga för många kemiska processer som sker inom cellen och styr allt från signalering till metabolism. Över evolutionens gång har nivån och distributionen av metaller inom celler finjusterats. Vid kroniska förändringar i metall balansen riskerar cellen att genomgå celldöd till följd av toxiskt höga eller låga metallnivåer. Dock är många av de underliggande mekanismerna okända. I den här avhandlingen har vi studerat sambandet mellan icke-fysiologiska nivåer av metaller och celldöd med ett fokus på cellens energifabrik, mitokondrien, och cellens återvinningscentral, lysosomer. För att studera de ovannämnda fundamentala cellbiologiska processerna använde vi jästceller och bananflugor. I vår första studie (Paper I) fann vi hur en ökad nivå av mangan leder till celldöd. Ett protein i mitokondrien, Coq7, som i normala fall förlitar sig på järn för sin aktivitet var inaktivt till följd av att mangan bytt ut järn. Detta lede till en försämrad förmåga att utvinna energi från socker via Koenzym Q (CoQ). En av svårigheterna med att studera metallers roll i biokemiska processer på cellulär nivå är avsaknaden av verktyg som kan följa intracellulära koncentrat- ionsförändringar av metaller. För att vidare studera mekanismerna som leder till celldöd till följd av toxiska nivåer av metaller utvecklade vi ett system som mäter aktiviteten hos ett protein vars aktivitet korrelerar med nivån av kalcium, calcineurin (Paper II). I kombination med en markör för celldöd kunde vi därmed studera calciumsignalering i levande celler. Me- tallnivåen inuti olika delar av cellen, cellens organeller, regleras med hjälp av membranpro- teiner som transporterat specifika metaller in eller ut ur organeller. Pmr1 är ett sådant pro- tein som transporterar kalcium och mangan. Tidigare studier har visat att celler som saknar Pmr1 dör i större utsträckning än vildtypsceller under normala förhållanden men inte när kaloriintag är begränsad. I vår tredje studie (Paper III) undersökte vi sambandet mellan ka- loribegränsning, celldöd och Pmr1. Vi kunde visa att avsaknaden av Pmr1 hämmar mitokondriens energimetabolism till följd av en förhöjd intracellulär koncentration av mangan, vilket motverkas av kaloribegränsning. Vidare mättes inga skillnader i calcineurin aktivitet eller biosyntesen av CoQ vid kaloribegränsning. Hos människor spelar kalcium en viktig roll i utvecklingen av neurodegenerativa sjukdomar som Parkinson’s sjukdom. I nästa studie (Paper VI and V) undersökte vi kopplingen mellan metallnivåer och kalciumsignalering i kontexten av Parkinson’s sjukdom. Vi kunde se att ökad calcineurin aktivitet stimulerade nedbrytning av proteinaggregat, en av orsakerna till neurodegeneration i Parkinsons sjuk- dom. Sammantaget visar våra resultat hur nivån av metaller i celler påverkar cellers levnads- förmåga. Vidare visar våra resultat hur toxiska nivåer av mangan hämmar mitokondriens funktion och energimetabolism.

Kurzzusammenfassung

Metalle spielen eine entscheidende Rolle in vielen Prozessen, die innerhalb einer Zelle ab- laufen. Sie kommen in großen (z.B. Kalzium) oder geringen Mengen (z.B. Mangan) vor und beeinflussen beispielsweise Kommunikation, Umgang mit Stress und Energiehaushalt. Eine sorgfältige Kalibrierung der Metallkonzentrationen schützt alle Lebensformen vor Zelltod, der auftreten kann, wenn gefährlich hohe oder niedrige Metallkonzentrationen in einer Zelle registriert werden. Wie zu viel oder zu wenig eines bestimmten Metalls zum Zelltod führt, ist nach wie vor nicht genau bekannt und Gegenstand unzähliger wissenschaftlicher Arbeiten.

In unserer Arbeit haben wir untersucht wie ein unausgeglichener Metallhaushalt das Über- leben von Hefezellen und Fliegen beeinflusst. Wir konzentrierten uns auf die Auswirkungen eines unausgeglichenen Kalzium- und Manganhaushalts auf Mitochondrien, die Kraftwerke der Zelle, und Lysosomen, die Recyclingstationen. In unserer ersten von fünf Studien stellen wir fest, dass Mitochondrien besonders empfindlich gegenüber zu viel Mangan sind. Ein Pro- tein in den Mitochondrien, Coq7 genannt, arbeitet normalerweise mit Eisen, ist jedoch zu viel Mangan in den Mitochondrien, wird das Eisen in Coq7 durch Mangan ersetzt. Das führt dazu, dass Coq7 nicht richtig funktioniert und weniger von seinem Produkt produziert. Die- ses Produkt, Coenzym Q, ist ein wichtiger Bestandteil der mitochondrialen Maschinerie, die Zucker in Energie umwandelt. Zu viel Mangan führt also dazu, dass die Funktion von Mito- chondrien beeinträchtigt wird, die Zellen leiden und schließlich sterben. In unseren weiteren Studien haben wir uns darauf konzentriert, welche Rolle Kalzium für das Überleben von Zel- len spielt. Die zweite Studie erläutert, wie wir einen Reporter hergestellt haben, der uns zeigt, wann die Zelle Kalzium verwendet um zu kommunizieren. Dieser Reporter gibt ein Lichtsig- nal, wenn die Zelle Kalzium verwendet, um das Protein Calcineurin zu aktivieren. Calcineurin ist ein wichtiges Protein in allen Zellen und kann viele Veränderungen verursachen. Bei- spielsweise kann es der Zelle helfen, sich an Stress anzupassen. In unserer dritten Studie haben wir diesen Reporter verwendet, um herauszufinden, wie wichtig Kalzium und Calcine- urin für alternde Hefezellen sind. Wir fanden heraus, dass Kalzium wichtiger ist, wenn Hefe- zellen schnell wachsen, und dass eine Verringerung des Wachstums durch Begrenzung der Zuckerquelle für das Altern von Vorteil sein kann, wenn Zellen ihren Kalziumhaushalt nicht regulieren können. In den letzten beiden Studien haben wir untersucht wie Kalzium und Calcineurin an einem Hefemodell der Parkinson-Krankheit beteiligt sind. Bei der Parkinson- Krankheit sind Teile des Gehirns besonders empfindlich gegenüber Ablagerungen des Pro- teins α-Synuclein, was dazu führt, dass die Zellen in dieser Gehirnregion sterben. Hefezellen haben zwar kein Gehirn, sie sind dennoch empfindlich gegenüber α-Synuclein-Ablagerungen und wir können sie verwenden um zu analysieren wie α-Synuclein zum Zelltod führt. In un- serer vierten Studie fanden wir heraus, dass Calcineurin eine große Rolle für das Überleben der Zellen mit α-Synuclein spielt: Calcineurin hilft dem Lysosom, und dem darin befindlichen Protein Cathepsin D, mit α-Synuclein und kaputten Proteinen umzugehen und damit das Überleben der Zellen zu verbessern. In unserer fünften Studie verwendeten wir Hefe- und

Fliegenmodelle, die aufgrund von α-Synuclein-Ablagerungen sterben. Durch Zugabe von mehr Kalzium in deren Nahrungsquellen erhöhen wir in beiden Fällen deren Calcineurin- Aktivität und helfen ihnen so, mehr Proteine in ihren Lysosomen zu recyceln und besser zu überleben. Wenn wir all diese Ergebnisse kombinieren, konnten wir zeigen, wie wichtig die sorgfältige Kalibrierung der Metallkonzentrationen für das Überleben von Hefezellen und Fliegen ist. Da die Zellmaschinerie von Hefezellen und Fliegen jener von Menschen sehr ähn- lich ist, sind unsere Ergebnisse eine wichtig Grundlage um zu verstehen, wie ein unausgegli- chener Metallhaushalt die menschlichen Zellen und ihr Überleben beeinflussen kann.

Contents

List of papers ... i

Aims of thesis ... iii

Popular summary ... iv

Populärvetenskaplig sammanfattning ... v

Kurzzusammenfassung ... vi

(1) Introduction: Metals in biology ... 1

(2) Manganese homeostasis ... 4

Manganese toxicity in higher eukaryotes ...4

Manganese homeostasis in budding yeast ...4

Pmr1 Act I – Ca²⁺ and/or Mn²⁺ homeostasis...5

Pmr1 Act II – Mn²⁺ homeostasis ...6

Identifying the molecular targets of cellular Mn overload ...6

Mitochondria as the target organelle of Mn toxicity ...7

CoQ nomenclature ...7

Regulation of Coenzyme Q biosynthesis ...8

The diiron hydroxylase Coq7 ... 10

Protein metallation and mismetallation ... 10

The diiron center of Coq7 is critical for structure and function ... 10

(3) Calcium homeostasis ... 12

Calcium homeostasis at a glimpse ... 12

The Ca²⁺/CaM -dependent phosphatase Calcineurin ... 13

Pmr1 Act III – Ca²⁺ homeostasis ... 14

Aging and caloric restriction (CR) ... 15

Calcineurin is required to promote lysosomal proteolysis to counteract α-synuclein toxicity ... 16

Outlook ... 17

Acknowledgements ... 18

Abbreviations ... 20

References... 21

(1) Introduction: Metals in biology

The occurrence of metals and the development of their roles on the stage of organic life has been mainly based on their availability in the environment throughout evolution (1). Metals can act in biology either incorporated as co-factors into proteins (metalloenzymes and other metalloproteins) or by themselves (e.g. Ca²⁺ as a ubiquitous second messenger). Additionally, they can give structure to RNA and DNA (2, 3) and are part of smaller biomolecules, such as copper in cobalamin (vitamin B12) and iron in hemes. These metal complexes contribute to various processes including signaling, respiration, photosynthesis, neurotransmission, ferti- lization and apoptosis (4) and are essential for all organic life forms (4–7).

Metals occur in biology across a wide range of concentrations. The essential metals in hu- mans can be divided into bulk metals (Na⁺, K⁺, Ca²⁺ and Mg²⁺) in the kg to g range and trace metals (Mn, Fe, Co, Cu, Zn, Mb) in the mg to g range (4). Similar to other nutrients, metal requirements can vary across species. For example, molybdenum is considered essential in humans (4), but not in yeast (8). For all life forms metals fulfill structural or catalytic functions.

Sodium and potassium are the most mobile metals and, in their ionic forms (Na⁺ an K⁺), are necessary to establish electrochemical gradients across membranes (4). Magnesium is esti- mated to be the most abundant metal used as a co-factor, occurring almost exclusively in its divalent form (Mg²⁺) (9). Calcium is best known for its role as second messenger and also predominantly occurs in its divalent form (Ca²⁺) in biology. It can induce major structural changes (e.g. in the Ca²⁺-modulated protein calmodulin) based on its unique coordination chemistry, which also enables it to bind to proteins even when other cations are in excess (10, 11). Among the transition metals, manganese, iron, cobalt, copper and molybdenum (Mn, Fe, Co, Cu and Mb) participate in redox reactions as they have more than one oxidation state, whereas zinc (Zn) only occurs as Zn²⁺ (4). Transition metals are mostly found in their divalent forms, as those have a higher ligand exchange rate than their trivalent forms, enabling faster chemical reactions with their environment (1).

Over time cells have developed mechanisms to import, utilize, sequester and export metals as part of the metabolic machinery that drives life. If one of these systems in metal homeo- stasis fails, or a cell is exposed to non-essential metals, the consequences can be fatal. Cer- tain aspects of metal homeostasis, such as the identity of transporters and channels as well as their transcriptional regulation in vivo are the subject of intense research (12). Similarly, metalloproteins and their metal requirements in vitro are actively studied in biochemistry (13). However, little is known about their synergy in vivo, about how metal levels regulate metal-dependent processes in the cell and how this contributes to sustaining cellular sur- vival.

As a unicellular organism, the yeast Saccharomyces cerevisiae is less complex and has fewer redundancies compared to mammalian cells. Many important components contributing to calcium and manganese homeostasis are evolutionarily conserved from yeast to human (14, 15). Similarly, S. cerevisiae has served as a valuable model to study aging and the impact of organellar biology on lifespan (16, 17). In this thesis, we set out to study the fundamental impact of disrupted calcium and manganese homeostasis on cellular fitness focusing on mi- tochondrial and lysosomal (yeast vacuolar) biology. We find that manganese toxicity primar- ily affects coenzyme Q homeostasis in mitochondria (Paper I and III), which leads to impaired development in Drosophila melanogaster and premature cell death in S. cerevisiae. The nega- tive impact of disrupted calcium homeostasis is greater in proliferating yeast cells during fermentation compared to cells on calorie restriction (Paper III) and calcium supplementa- tion is able to prevent the toxicity of α-synuclein and concomitant cell death by stimulating lysosomal proteolysis in both yeast and Drosophila models for Parkinson’s disease (Paper IV and V).

(2) Manganese homeostasis

Manganese toxicity in higher eukaryotes

In all eukaryotic organisms, overexposure to Mn results in a failure of cellular energy metab- olism, though the underlying molecular mechanisms remain unclear. Consequences of cel- lular Mn overaccumulation include defective photosynthesis and chlorosis in plants and compromised mitochondrial function in animals (18–21). A large body of knowledge about Mn utilization has come from plants, where Mn is an important factor in superoxide dis- mutases (together with Fe), enzymes of the tricarboxylic acid cycle as well as the photosys- tem. Sufficient metal uptake from the soil not only fosters plant growth but serves as an underlying principle in biofortification, a strategy to increase the nutritional value of crops (5). However, there are numerous examples of metal toxicity in plants, which occurs espe- cially in acidic soils, where Mn²⁺ becomes highly soluble and overrepresented in the plant tissue. Even though many phenotypes have been associated with Mn toxicity, the primary molecular targets remain unknown (22).

For decades, evidence suggesting an involvement of Mn-requiring processes in neurodegen- erative diseases, such as Parkinson’s disease, Alzheimer’s disease and Huntington’s disease, has been accumulating (23). Especially Parkinson’s disease resembles the symptoms of man- ganism, a neurotoxic condition resulting from Mn accumulation in the brain. Neuronal de- mise driven by excess Mn has been associated with several cellular processes, including dysregulation of glutamate transport, compromised dopaminergic function and perturba- tions of cellular Ca²⁺ handling as well as a failure of cellular energy metabolism, often asso- ciated with mitochondrial dysfunction and oxidative stress (20, 24–27).

Manganese homeostasis in budding yeast

Yeast can grow in and adapt to a wide range of intra- and extracellular Mn concentrations (28). Diverse processes such as nucleotide synthesis, protein glycosylation and redox reac- tions require Mn-dependent enzymes (28). A screen of non-essential genes required for growth under low Mn availability identified cell cycle and protein synthesis as significant pathways (29). Further, non-protein Mn complexes, such as Mn-phosphate, Mn-carbonate, and Mn-lactate have been reported to act as scavengers of reactive oxygen species (ROS) (30, 31). Cellular Mn homeostasis is not known to be regulated by transcription, but rather through post-transcriptional modifications and protein turnover, which is directed by changes in Mn levels (28).

While Mn transporters into the cell (Smf1, Pho84), into luminal vesicles (Smf1, Smf2), the Golgi (Ca²⁺+Mn²⁺/H⁺ exchanger Gdt1) and into the vacuole (Ccc1, Ypk9) have been identified, how Mn crosses mitochondrial membranes is not as well-characterized (28, 31). One study reports that Mtm1, a member of the mitochondrial solute carrier family, is required for metallation of the Mn-requiring superoxide dismutase Sod2 within mitochondria. However, this is most likely not an effect of Mtm1 acting as Mn transporter, as loss of Mtm1 results in iron accumulation in the mitochondria and mismetallation of Sod2 with iron (32, 33). To date, Mn is thought to cross the outer mitochondrial membrane unspecifically through pores, whereas uptake into the matrix requires dedicated, yet unidentified transport mechanisms (34). In the endoplasmic reticulum (ER) membrane, the still putative Ca²⁺/Mn²⁺-ATPase Spf1 is thought to contribute to Mn trafficking. Its loss or overexpression manifests as decreased or increased levels of Mn, respectively, in isolated microsomes (ER-derived vesicles) as well as phenotypes related to disrupted Mn homeostasis (35). However, no direct evidence of Mn-specific translocation via Spf1 exists. One of the central transporters regulating not only Mn, but also Ca²⁺ homeostasis is the Ca²⁺/Mn²⁺ ATPase Pmr1 (mammalian SPCA1). Heterol- ogous expression of human SPCA1 in S. cerevisiae can complement phenotypes associated with loss of Pmr1 (36), arguing for functional conservation.

Pmr1 Act I – Ca²⁺ and/or Mn²⁺ homeostasis

In mammals and yeast, Pmr1 localizes mainly to the Golgi apparatus, but in yeast small frac- tions have been found in the ER as well (37, 38). In flies, three Pmr1 isoforms exist, which localize to the Golgi, the ER and peroxisomes, respectively (39). The Rao lab identified the domains and residues relevant for Ca²⁺ and Mn²⁺ transport in Pmr1. First, Ca²⁺ transport was mapped to an N-terminal EF-hand-like motif and the aspartate residues 53 and 55 (40). Cor- responding single and double point mutants (Pmr1^D53A, Pmr1^D55A and Pmr1^{D53A, D55A}) showed a greatly decreased ability to transport Ca²⁺. Later, glutamine 783 was identified as an im- portant residue for Mn²⁺ transport (41). In another study, the analysis of 35 different point mutations in Pmr1 revealed that at least 13 amino acid residues are important for the coor- dination and binding of Ca²⁺ and Mn²⁺ (42). These point mutants have been used in various studies to dissect the impact of Pmr1-mediated Ca²⁺ or Mn²⁺ transport on diverse cellular processes (43–45). However, many studies have been carried out with the complete loss of Pmr1/SPCA1 function, which complicates the attribution of phenotypes to one of the two metals. Figure 1 schematically depicts wild type Pmr1 and the mutant proteins Pmr1^D53A and Pmr1^Q783A. Cells lacking the Pmr1 protein completely lack the PMR1 gene (∆pmr1 cells).

Figure 1: Schematic of wild type and mu- tant Pmr1 proteins with impaired Ca²⁺ or Mn²⁺ transport in yeast. The PMR1 dele- tion strain lacks the gene for Pmr1.

Further complicating the distinction between Ca²⁺-and Mn²⁺-related processes, a yeast study outlines a mechanism of how Ca²⁺ supplementation, widely used to discern between the impact of Ca²⁺ or Mn²⁺-related processes in cells lacking Pmr1 (∆pmr1 cells), alleviates known Mn²⁺-specific phenotypes (46). The restoration of Golgi Mn²⁺ homeostasis in ∆pmr1 cells upon CaCl2 treatment is enabled through organelle-to-organelle Mn²⁺ transport. Specifically, this requires Smf2, a trans-Golgi/late endosome Mn²⁺ transporter and functional intracellular vesicle trafficking to supply the cis-Golgi with Mn²⁺ loaded vesicles from the trans-Golgi.

With a profound impact on both Ca²⁺ and Mn²⁺ homeostasis, it is not surprising that com- promised Pmr1/SPCA1 function has been shown to cause pleiotropic phenotypes. Homozy- gous loss of murine SPCA1 causes defects in neuronal development (47) and homozygous mutations in nematode Pmr1 lead to defective embryogenesis (48). In both studies, it re- mained undetermined which aspect of metal homeostasis was affected. However, it is known that SPCA1 promotes Mn²⁺ detoxification in rat liver cells (49). Mutations in the human SPCA1 result in Hailey-Hailey Disease, a blistering skin disease (50). While the etiology of the disease has been connected mostly to disrupted Ca²⁺ homeostasis (51, 52), investigation of 177 known disease-causing mutations did not reveal any apparent clustering. A yeast-based heterologous expression screen of 14 out of 180 known disease-causing mutations of SPCA1 identified Mn²⁺ transport as being more affected than Ca²⁺ transport (53). A clear contribu- tion of impaired Ca²⁺ or Mn²⁺ homeostasis remains to be determined.

Pmr1 Act II – Mn²⁺ homeostasis

Loss of Pmr1 in yeast results in defective protein glycosylation, which is attributed to com- promised function of Mn²⁺-dependent sugartransferases in the secretory pathway (54). This defect can also be restored with Mn²⁺ supplementation (54). Further, mitotic misregulation has been linked to increased Mn²⁺ levels upon loss of yeast Pmr1 (55). Studies in nematodes and yeast demonstrated an involvement of Pmr1 in oxidative stress (56–58). Compromised Pmr1 function and subsequent accumulation of intracellular Mn in the form of non-protein complexes are thought to confer cytoprotection against oxidative stress (30). However, stud- ies in the yeast Kluyveromyces lactis have found that oxidative stress and altered mitochon- drial function compromise survival in cells lacking Pmr1, which can be alleviated by overex- pression of the mitochondrial chaperonin Hsp60 and the Glutathione S-transferase ϴ-subu- nit, both involved in oxidative stress defense (59, 60).

Identifying the molecular targets of cellular Mn overload

Given the controversial role of Mn in oxidative stress and the still elusive cellular determi- nants of Mn toxicity, we set out to identify the molecular targets primarily affected by cellular Mn overload (Paper I). After an initial screen of yeast strains with single deletions in genes connected to regulating intracellular Mn and other divalent metal levels, we identified the PMR1 deletion mutant as the only candidate with up to a 10-fold increase in total Mn levels.

Subsequent whole cell proteomics revealed that the gene ontology terms respiratory elec- tron transport chain and tricarboxylic acid cycle, both associated with mitochondrial func- tion, were significantly deregulated. We could further ascribe disrupted mitochondrial func- tion in the form of impaired oxygen consumption and respiratory growth defects to loss of Pmr1-mediated Mn²⁺ transport employing the Pmr1^Q783A mutant. Similarly, oxygen consump- tion was impaired in wild type cells exposed to high levels of Mn in a dose-dependent man- ner.

Mitochondria as the target organelle of Mn toxicity

Mitochondria are the powerhouses of the cell and as such have evolved as critical determi- nants of cellular growth and survival. Harboring such vital processes as the tricarboxylic acid cycle and the respiratory electron transport chain for energy metabolism, they offer several points for attack by cytotoxic insults. Several in vitro and in vivo studies indicate that Mn overload disrupts several crucial mitochondrial functions, resulting in a failure of cellular energy metabolism. This includes the induction of oxidative stress and mitochondrial DNA damage, a dysregulation of mitochondrial fission and fusion events as well as perturbations in oxidative phosphorylation and the tricarboxylic acid (TCA) cycle (61–65). Although details in respect to the molecular determinants and mechanisms directly targeted by excess Mn remain unclear, toxicity might involve an inhibition of several enzymes critical for mitochon- drial energy metabolism, such as the iron-sulfur cluster enzyme aconitase, as well as indirect damage by Mn-induced oxidative stress (65). In a whole genome screen, we identify COQ7 as the only gene able to suppress the respiratory growth defect of ∆pmr1 cells. Subsequent in- depth analysis established Coq7 as the primary target of mitochondrial Mn overload (Paper I). Coq7 overexpression was sufficient to fully restore respiration and to prevent subsequent cell death despite Mn overaccumulation. Coq7 is part of the CoQ synthome, a multiprotein complex peripherally associated with the inner mitochondrial membrane facing the matrix and producing the lipophilic electron carrier coenzyme Q (CoQ) (66). Even though respiratory chain enzymes have been postulated as targets of mitochondrial Mn toxicity, complex III and complex IV were fully functional (Paper I). NADH and succinate dehydrogenase activities, defective upon Mn overload, could be restored with exogenous CoQ (Paper I), a hallmark feature of mutants lacking components of CoQ synthesis (67). Our findings establish a new connection between Mn²⁺ and CoQ homeostasis.

CoQ nomenclature

COQ1-11 all capitals and italic, refers to a gene or mRNA thereof.

coq1-11 all lower case and italic, refers to a mutated gene.

Coq1-11 first letter capitalized, refers to a protein.

CoQ first and last letter capitalized, refers to the lipid coenzyme Q, which is also known as ubiquinone or Q. A number in subscript after CoQ, e.g. CoQ6, refers to the num- ber of isoprene units in the lipid tail.

Regulation of Coenzyme Q biosynthesis

The redox-active lipid CoQ is an important electron-carrier in the electron transport chain, where the quinone form is reduced by the actions of either NADH dehydrogenases (Nde1/2 and Ndi1) or succinate dehydrogenase (Complex II, CII). CoQH2 (ubiquinol) transfers its elec- trons to cytochrome c-reductase (Complex III, CIII), where it is oxidized and recycled in the Q-cycle. CIII in turn reduces cytochrome c in the intermembrane space (IMS), where it freely diffuses to cytochrome c oxidase (Complex IV, CIV). CIV transfers electrons from cytochrome c to molecular oxygen. Figure 2 depicts a schematic representation of the electron flow in the yeast electron transport chain.

CoQ is a redox agent not only found in mitochondrial membranes but in most cellular mem- branes (68). It has a lipid tail, retaining it in lipid bilayers, and a redox-active head group, which enables electron shuttling (66). The precursors for the head group come from either 4-hydroxybenzoate (and/or para-aminobenzoic acid in yeast (69)) and can be a rate-limiting step for its synthesis (66). Once in the inner mitochondrial membrane, Coq2 attaches the head group to the hydrophobic tail synthesized by Coq1. The length of the lipid tail, i.e. the number of isoprene units is species-specific (e.g. CoQ6 in S. cerevisiae; mostly CoQ9 in Dro- sophila, and CoQ10 in humans) (68). The assembled molecule goes through a series of head group modifications in the so-called CoQ synthome, consisting of Coq3-9 (70, 71). The reac- tion catalyzed by Coq7 is the hydroxylation of demethoxy-ubiquinone (DMQ) on the C5-po- sition of the head group to produce demethyl-ubiquinone (Figure 3A). Coq10-11 are thought to contribute to regulation and distribution of CoQ in yeast (66).

The regulation of CoQ biosynthesis is multifaceted and not fully elucidated. All COQ genes are encoded in the nucleus and gene expression is highly dependent on the available carbon source (72). Post-transcriptional regulation of COQ5 mRNA levels has been shown to be me- diated by the mRNA-binding protein Puf3 (73). Synthesized polypeptides are transported into the mitochondrial matrix in a transmembrane potential-dependent manner, even though strains with minimal transmembrane potential maintain the ability to synthesize CoQ (74).

Several Coq proteins are also regulated by phosphorylation, and Coq7 dephosphorylation can greatly stimulate CoQ synthesis (75, 76).

The assembly of the CoQ synthome occurs in several steps and is also thought to contribute to the regulation of CoQ synthesis. One model proposed the preassembly of a 700 kDa pre- complex consisting of Coq3, Coq4, Coq5, Coq6 and Coq9, which synthesizes the substrate

Figure 2: The electron transport chain in S. cerevisiae with a focus on the role of the re- dox-active lipid coenzyme Q (CoQ). For details see text. red. = reduced. ox. = oxidized.

for Coq7, DMQ. The association with Coq7 completes the full CoQ synthome and enables the last two reactions in the synthesis of CoQ (74). The proteins involved in CoQ synthesis are to a large degree functionally conserved across species, and the lack of yeast Coq pro- teins can be complemented by human homologs. Primary and secondary CoQ deficiencies in humans result in mitochondrial dysfunction and disease (66).

To understand the mechanism of impaired CoQ synthesis upon Mn overload we tested sev- eral of these aspects in our study (Paper I). However, neither transcriptional regulation nor mitochondrial import or localization of Coq proteins involved in CoQ synthesis seemed to be severely altered upon Mn overload. Moreover, only expression of Coq7 and supplemen- tation with 2,4-diHB, shown to specifically bypass Coq7-defective CoQ synthesis (77), could restore respiratory growth. Figure 3B shows a schematic of the bypass of the Coq7 reaction.

As the CoQ head group precursor already possesses a hydroxyl in the C5 position (indicated in red), the enzymatic activity of Coq7 is not required to produce CoQ.

In all scenarios tested to increase Coq7 protein levels, we never observed a marked accumu- lation of Coq7 protein even though mRNA levels were comparable to wild type upon over- expression (Paper I). In combination with our results showing mismetallation of Coq7 by Mn, replacing iron, we conclude that excess Mn during folding of the nascent polypeptide in the mitochondrial matrix replaces iron in the catalytic di-metal center and renders the enzyme dysfunctional.

Figure 3: The role of Coq7 in the synthesis of coenzyme Q (CoQ). (A) Coq7 catalyzes the hydroxylation of DMQ to DMeQ in the CoQ synthome. (B) The CoQ head group precursor 2,4-dihydroxybenzoic acid (2,4-diHB) bypasses the requirement for the Coq7 reaction.

For detail see text. DMQ = demethoxy-ubiquinone, DMeQ = demethly-ubiquinone, IMS = Intermembrane space, IMM = inner mitochondrial membrane.

The diiron hydroxylase Coq7

Coq7 is a diiron hydroxylase performing the penultimate step in the CoQ biosynthesis path- way (78, 79). Its structure is a four-helix bundle with a metal-binding site, where imidazole and carboxylate residues (in the case of Coq7 from 2 histidine and 4 glutamate side chains) coordinate two iron atoms (79). This binding motif is shared by other enzymes in the carbox- ylate-bridged diiron protein family, e.g. monooxygenases, mitochondrial alternative oxidase and ribonucleotide reductase (79). Although no experimental evidence exists, the iron in the diiron center is most likely bound during folding in the mitochondrial matrix. The reaction of DMQ hydroxylation has been elucidated with the human homolog of Coq7 (COQ7/CLK-1) (80). It involves electrons supplied from NADH and oxygen from dioxygen, which are trans- ferred onto DMQ to form the product demethyl-ubiquinone via reduction and oxidation of the diiron center.

Protein metallation and mismetallation

The abundance of metals in the intra- and extracellular space and the requirement of a met- alloprotein for a certain metal are often not aligned (81). In in vitro studies, metalloenyzmes are frequently shown to exert their function while metallated with different metals. How- ever, in vivo, the majority of metalloproteins seem to require one specific metal for function, and binding to the wrong metal will in most cases lead to an inactive enzyme. Thus, a cell has to constantly monitor intracellular metal levels to ensure proper metallation and thus function of metalloenzymes. Given this complex balance, it is also important to consider more than one metal when studying metal homeostasis in vivo, as a deficiency of one metal can simultaneously result in the overload of one or more other metals (82). Iron and Mn are highly likely to compete for metal-binding sites due to their similarity in physicochemical properties, such as ionic radius (11) and thus tend to outcompete each other for similar binding sites on metalloproteins (81). One prominent example of protein mismetallation im- pacting function in vivo is the case of mitochondrial manganese superoxide dismutase 2 (Sod2) (33). Upon mitochondrial iron overload, Sod2 is mismetallated with iron and rendered inactive.

The diiron center of Coq7 is critical for structure and function

The importance of the diiron center for enzyme stability and function has been observed with alternative oxidase of Arabidopsis thaliana (83). Four residues constituting the diiron center contribute to protein activity and stability. Individual mutations of two glutamate and two histidine residues coordinating the diiron center rendered the enzyme inactive, and the two mutations of the histidine residues destabilized the protein. Similar observations were made with the alternative oxidase of the parasitic protozoan trypanosome (84).

The importance of the diiron center for Coq7 structure and function has been well-estab- lished using the Coq7-E194K mutant, which is predicted to have a disrupted diiron center (85). The mutant fails to grow with glycerol as a carbon source and has no measurable CoQ, but twice the amount of DMQ compared to cells with wild type Coq7. While NADH- and suc- cinate-dehydrogenase activities are about half-maximal, activities requiring CoQ (NADH- /succinate-CIII) are practically abolished. The Coq7-E194K mutant is also more sensitive to oxidative stress induced by hydrogen peroxide and linolenic acid compared to wild type cells. In the performed clonogenicity assay to determine viability, Coq7-E194K mutants were especially sensitive to high doses of linolenic acid once cells had reached stationary phase, indicating that DMQ lacks anti-oxidant activity (85). In line with these results, we observe higher oxidative stress levels in ∆pmr1 cells (Paper I, Paper III). Analyses of steady-state pro- tein levels in the Coq7-E194K mutant (86) further corroborate our finding that Mn poisoning destabilizes Coq7. Protein levels of Coq7-E194K as well as Coq3, Coq4, Coq6, but not the respiratory complexes, were diminished compared to wild type cells. This study further shows that respiratory growth, absent in Coq7-E194K mutants, is restored by xenotopic ex- pression of UbiF, a bacterial equivalent of Coq7. The E. coli flavin adenine dinucleotide-de- pendent monooxygenase (UbiF) catalyzes the hydroxylation step analogous to Coq7 and does not require a diiron center for activity. It was found that UbiF, while restoring respira- tory growth of Coq7-E194K mutants, only produced small amounts of CoQ on glycerol (86).

We speculate that in our setup, where yeast Coq7 activity and stability are influenced by mitochondrial Mn overload, UbiF overexpression could increase respiration on glucose.

In conclusion, we establish the diiron hydroxylase Coq7 as the primary target of cellular Mn overload. Excess Mn causes mismetallation, loss of Coq7 and a subsequent disruption of CoQ biosynthesis. Absence of CoQ as a crucial electron carrier impairs oxidative phosphor- ylation, leading to defective respiratory growth and cell death. We propose a model in which the unique sensitivity of a diiron enzyme towards mismetallation underlies the disruption of mitochondrial bioenergetics upon Mn overaccumulation.

(3) Calcium homeostasis

Calcium homeostasis at a glimpse

Maintaining proper Ca²⁺ homeostasis is essential for cellular function and proliferation in all eukaryotic cells, in particular long-lived cells such as neurons (87, 88). Cellular Ca²⁺ homeo- stasis is achieved by a vast number of molecules that (i) affect Ca²⁺ signaling, transport and buffering and/or (ii) are affected by general changes in Ca²⁺ concentrations or subtler, spa- tially and temporally confined Ca²⁺ transients in a highly complex and interconnected man- ner. The unique role of Ca²⁺ in biology is accredited to its physical properties, which enable certain flexibility in the way it binds to ligands, i.e. biomolecules. This is exemplified by its highly specific binding to the EF-hand motif (10). Even though cytosolic Ca²⁺ concentrations are lower than K⁺, Na⁺ and Mg²⁺, only Ca²⁺ binding to the EF-hand induces structural changes in proteins like calmodulin, which then relay signals upon changes in cytosolic Ca²⁺ levels (11).

Many important components of the Ca²⁺ network are evolutionarily conserved from yeast to human (Ton and Rao, 2004). In yeast, Ca²⁺ enters the cell through the high-affinity Ca²⁺ up- take system (HACS), composed of Mid1/Cch1/Ecm7 and yet unidentified channels or trans- porters in the plasma membrane (89). Ca²⁺ signaling throughout the cell requires the possi- bility to rapidly transition from low to high concentrations and reestablish low cytosolic Ca²⁺

levels. Thus, the cytosolic levels are kept at a low concentration in the nM range (∼100 nM) (90). Most Ca²⁺ entering the yeast cell from the extracellular space is transported into the vacuole, the major storage organelle for metal ions and other nutrients. Ca²⁺ enters the vac- uole through the Ca²⁺ ATPase Pmc1 and the Ca²⁺/H⁺ exchanger Vcx1 and exits it through the stretch-activated channel Yvc1. The ER and Golgi membranes harbor the putative Ca²⁺/Mn²⁺

ATPase Spf1, the Ca²⁺/Mn²⁺ ATPase Pmr1 and the Ca²⁺+Mn²⁺/H⁺ exchanger Gdt1 (89, 91). Spf1 has been implicated in Ca²⁺-related processes in the ER (92), however, evidence for Ca²⁺

transport is missing. On the contrary, purified Spf1 did not show any Ca²⁺-activated or EGTA- sensitive ATPase activity in vitro and ATP hydrolysis was attributed to the phosphatase Pho8 that was purified as a contaminant (93). To date, no mitochondrial Ca²⁺ transporters have been identified in yeast. Similar to what is suggested for Mn²⁺, mitochondrial uptake of Ca²⁺

could be unspecific through pores in the outer mitochondrial membrane, followed by active transfer by a yet unidentified transporter or channel in the inner mitochondrial membrane.

To cope with increased cytosolic levels, the cell sequesters excess Ca²⁺ in the vacuole com- plexed to polyphosphate (90). Figure 4 schematically depicts the Ca²⁺ transporters in S. cere- visiae. This occurs through calcineurin-dependent activation of the vacuolar Ca²⁺ pump Pmc1 in conditions of high external Ca²⁺ or upon loss of Ca²⁺ pumping in the ER/Golgi, requiring increased Ca²⁺ uptake to refill the luminal stores. Calcineurin is highly activated in cells lack- ing Pmr1 (94), causing, for instance, transcriptional upregulation of Pmc1 to sequester excess

cytosolic Ca²⁺ (90, 95). While calcineurin function seems dispensable in unstressed cells, it is assumed to be crucial for survival upon disruption of Ca²⁺ homeostasis.

The Ca²⁺/CaM -dependent phosphatase Calcineurin

The evolutionarily conserved Ca²⁺/calmodulin (CaM)-dependent serine/threonine phospha- tase calcineurin is an important Ca²⁺ signaling molecule. Its name stems from initial studies observing its Ca²⁺ binding properties and its prominent presence in neuronal tissues. Cal- cineurin participates in neuritogenesis, regulation of ion channels, endocytosis and apopto- sis amongst others (96). Given this broad spectrum of function, it is not surprising that de- regulated calcineurin activity has been implicated in neurotoxicity (97–99). Specifically, abol- ishing or hyperactivating calcineurin has both been shown to increase cytotoxicity of α-synu- clein in a yeast model for Parkinson’s disease (100), demonstrating the need for tight spatiotemporal control of this phosphatase to exert proper function.

Calcineurin is a heterodimeric metalloenzyme requiring three different metals for activity.

The regulatory subunit (CnB) binds Ca²⁺ through 4 EF-hand motifs with different affinities, which enables a range of structural and functional properties (101, 102). The catalytic subunit (CnA) has binding domains for CnB and calmodulin, which binds Ca²⁺ itself and is required

Figure 4: Schematic representation of Ca²⁺ transporters in S. cerevisiae. Pmr1 is a Ca²⁺ and Mn²⁺

transporter. Gdt1 is a Ca²⁺+Mn²⁺/H⁺ exchanger and Spf1 is a putative Ca²⁺/Mn²⁺ transporter. For further details see text.

metal site, which has been modeled with an Fe³⁺-Zn²⁺ cluster. Mn²⁺, as well as Ni²⁺, can in- crease activity of calcineurin in vitro (96).

In fungi, calcineurin is one of the major stress-induced signaling molecules and exerts nu- merous functions including growth control, ion homeostasis and adaptation in response to a fluctuating environment (103–106). In pathogenic fungi such as Aspergillus, Cryptococcus and Candida, calcineurin contributes to virulence and is thus a target for combating infections (107, 108). Across phyla, the phosphatase can be inactivated pharmacologically with the se- lective inhibitors FK506 and cyclosporine A (109).

There are two lines of action in which calcineurin regulates cellular changes. First, calcineurin directly dephosphorylates target proteins involved in processes such as protein trafficking, Ca²⁺ homeostasis and lipid metabolism (110). The other line is indirect, through the zinc-fin- ger transcription factor Crz1. Upon dephosphorylation by calcineurin, Crz1 translocates to the nucleus and activates transcription of genes with calcineurin-dependent response ele- ments (CDREs) (106, 111).

This second line has been exploited to study calcineurin activity in vivo. Established reporters are based on Crz1 binding to CDRE and activating the transcription of a reporter gene (51, 110, 112, 113). Most of these assays require cell lysis in order to determine the activity of the commonly used beta-galactosidase reporter. As calcineurin plays such a prominent role in stress signaling, we wanted to be able to follow its activity in unperturbed, living cells and elucidate its contribution to survival (Paper II). To enable high-throughput in vivo measure- ments we generated a fluorescence-based reporter. This way, the readout can be combined with a simple live/dead staining with fluorescent dyes, which can both be recorded via flow cytometry or microscopy. To allow for a higher temporal resolution we also generated a variant with destabilized GFP via fusion to a degron. This is especially useful when quantify- ing rapid or transient changes in calcineurin activity upon extracellular stimulation of cells.

Pmr1 Act III – Ca²⁺ homeostasis

As mentioned above, loss of Pmr1 in the yeast S. cerevisiae results in a multitude of pheno- types. Contrary to the rather detrimental effects upon compromised Pmr1 function dis- cussed this far, loss of Pmr1 Ca²⁺ transport confers cytoprotection in a yeast model for Par- kinson’s disease (43). Nevertheless, deletion of PMR1 causes severe dysregulation of Ca²⁺ ho- meostasis, resulting in an increase in total, free cytosolic and vacuolar Ca²⁺ as well as a de- crease in ER Ca²⁺ (90, 114, 115). Cells lacking Pmr1 exhibit growth defects in low Ca²⁺ or the presence of divalent metal chelators (e.g. EGTA), protein sorting defects and vacuolar frag- mentation (54, 116–118). These phenotypes are alleviated by administration of modest doses of Ca²⁺ in the media or replenishing luminal stores through overexpression of mammalian Ca²⁺-specificpumps. Interestingly, the overexpression of Vps10/Pep1 (a sorting receptor for vacuolar proteases) can also counteract the growth defect in the presence of EGTA (54). Ad- ditionally, lack of yeast PMR1 has been shown to reduce replicative lifespan in a high- throughput screen and to compromise survival during chronological aging in the presence