Molecular targets underlying benzo[a]phenoxazines antifungal activity: the role of ergosterol metabolism and endoplasmic reticulum stress

Ana Rita Santos Nunes Molecular targets underlying benzo[a]phenoxazines antifungal activity: the role of ergosterol metabolism and endoplasmic reticulum stress Abril de 2019 UMinho | 2019 Ana Rita Santos Nunes Molecular targets underlying benzo[a]phenoxazines antifungal activity: the role of ergosterol metabolism and endoplasmic reticulum stress Universidade do Minho Escola de Ciências Ana Rita Santos Nunes Molecular targets underlying benzo[a]phenoxazines antifungal activity: the role of ergosterol metabolism and endoplasmic reticulum stress Dissertação de Mestrado Mestrado em Genética Molecular Trabalho efetuado sob a orientação de Professora Doutora Maria João Marques Ferreira Sousa Moreira Professora Doutora Maria do Sameiro Torres Gonçalves Abril de 2019 Universidade do Minho Escola de Ciências ii DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ DE ACORDO COM A LEGISLAÇÃO EM VIGOR, NÃO É PERMITIDA A REPRODUÇÃO DE QUALQUER PARTE DESTA TESE/TRABALHO. iii Agradecimentos Começo com uma palavra de agradecimento muito grande às duas pessoas sem as quais a realização da minha tese não teria sido possível: à Professora Maria João, minha orientadora, agradeço não só pela orientação que me deu mas também pela simpatia, persistência e principalmente paciência que teve para comigo; à Professora Sameiro, que apesar de ter tido menos contacto comigo, mostrouse sempre muito prestável, simpática e atenta ao meu trabalho. Um sincero obrigado às duas. Continuo agradecendo a toda a minha "crew" da Micro I, por me terem aturado, mas principalmente pela ajuda que me deram e pela boa companhia que foram. Obrigada Cátia (Mini boss), Filipa, Marta, Flávia, António PhD (Pi-eij-di), Vitória (Vit), Carla e Joana. Aos meus amigos e fiéis irmãos de bancada e de orientadora, Belinha e João, obrigada pela vossa ajuda, por estarem sempre lá para me fazerem rir e rirem comigo. Sei que vai custar não me ter na bancada convosco, mas eu sei que vocês são capazes de aguentar essa dor. À minha "equipa de manutenção" pertencente ao LBA, que quase conseguiu tornar a hora do almoço a minha hora preferida do dia, devido à ótima companhia que são. Depois de uma manhã de trabalho difícil, nada melhor do que um almoço com esta malta. Sarinha, Maria, Ritinha, Mário e Pedro, um grande obrigado. Um agradecimento a todo o departamento de biologia e à equipa de técnicos e funcionários, sem os quais não seríamos capazes de trabalhar. Um especial e grande obrigado ao “vizinho da frente”, o Sr. Luís, que devia ganhar o prémio de melhor técnico de laboratório. Obrigada por existir no nosso departamento e que sorte a nossa o seu local de trabalho ser mesmo em frente à Micro I. Obrigada pela infinita ajuda, simpatia e por ser tão prestável. Já não sei o que é acabar o dia sem ter ouvido o Sr. Luís a chamar-me "Muchacha". Dirijo-me agora aos meus amigos do coração, sem os quais esta fase não seria concluída com sucesso e sem os quais a minha vida não seria a mesma. Tiago, Lagartas (Inês, Mariana, Raquel e Rita), Eloy, Miguel Palmares, Hugo Blanquet, Pedro, obrigada a vocês por existirem na minha vida e por sempre iv me ajudarem tanto. Um obrigado à Família da FEUP, que, apesar de ter entrado na na minha vida há pouco tempo, já contribuiu muito para o meu sucesso nesta etapa. De uma maneira ou de outra, todos vocês foram importantes para que eu conseguisse concluir mais esta etapa. Obrigada também aos amigos que não referi aqui pois não me quero alongar. Por fim, a minha maior gratidão é dirigida à minha família, em especial aos meus pais, que são aquilo que de mais importante tenho e terei sempre na vida. Pai e mãe, obrigada por estarem sempre ao meu lado. Acompanharam-me tanto, e tão bem, ao longo deste percurso que estão aptos a serem mestres, tal como eu vou ser. São os melhores pais que eu podia ter (e tenho) e isso, por si só, já diz muito. Obrigada por sempre me darem tanta força, por sempre me orientarem na direção certa nas várias vezes em que já me senti perdida, e, principalmente, por nunca me deixarem desistir. Aproveito ainda para agradecer ao Centro de Biologia Molecular e Ambiente (CBMA) e ao programa estratégico UID/BIA/04050/2013 (POCI-01-0145-FEDER-007569) financiado por fundos nacionais através da Fundação para a Ciência e a Tecnologia (FCT) e pelo Fundo Europeu de Desenvolvimento Regional (FEDER) através do COMPETE 2020 – Programa Operacional Competitividade e Intercionalização (POCI). v DECLARAÇÃO DE INTEGRIDADE Declaro ter atuado com integridade na elaboração do presente trabalho académico e confirmo que não recorri à prática de plágio nem a qualquer forma de utilização indevida ou falsificação de informações ou resultados em nenhuma das etapas conducente à sua elaboração. Mais declaro que conheço e que respeitei o Código de Conduta Ética da Universidade do Minho. vii Alvos moleculares subjacentes à ação antifúngica das benzo[ a ]fenoxazinas: o papel do metabolismo do ergosterol e do stress do retículo endoplasmático Resumo Com o passar dos anos, as infeções fúngicas têm-se tornado uma questão cada vez mais relevante, particularmente devido ao desenvolvimento de resistência contra as drogas antifúngicas em uso. Recentemente, o nosso grupo tem contribuído para a elucidação do mecanismo de ação de um grupo de compostos químicos com atividade anti-proliferativa, as fenoxazinas e os seus derivados, nomeadamente as benzo[ a ]fenoxazinas. Trabalhos anteriores do nosso laboratório têm-nos fornecido um vasto conhecimento acerca do mecanismo de ação da benzo[ a ]fenoxazina BaP1. O BaP1 é capaz de induzir um processo de morte celular regulada, mediado pela permeabilização do vacúolo. As potencialidades exibidas por este composto, levaram o nosso laboratório a iniciar os estudos com um composto semelhante, o C9, que apenas difere do BaP1 numa substituição do sistema aromático nuclear da benzo[ a ]fenoxazina. Esta pequena diferença estrutural, contudo, teve um grande efeito na atividade do composto, uma vez que o C9 é um agente anti-proliferativo mais potente do que o BaP1. As primeiras abordagens mostram que este pode atuar de uma forma diferente do BaP1. Trabalhos desenvolvidos anteriormente no nosso laboratório já tinham demonstrado que este composto acumula não só na membrana vacuolar, mas maioritariamente na membrana do retículo endoplasmático. Tal como o BaP1, o C9 é capaz de induzir um processo de morte regulada, apesar de, no caso do C9, haver também o envolvimento mitocondrial. Não obstante, o(s) alvo/alvos deste composto continua/continuam por descobrir. De forma a contribuir para a elucidação dos alvos do C9 e, tendo em conta que o seu principal local de acumulação na célula é a membrana do retículo endoplasmático, colocamos a hipótese de que o metabolismo do ergosterol, nomeadamente, a sua via de biossíntese, pode estar envolvido no modo de ação do C9. Contrariamente ao que foi inicialmente proposto, observou-se que uma suplementação externa de ergosterol no meio durante o tratamento com o C9 poderá ter um papel efetor nos efeitos nocivos do C9 no crescimento celular, não tendo, no entanto, qualquer efeito na viabilidade das mesmas. Finalmente, mostramos que o mecanismo de ação do C9 não é mediado pelo stresse do retículo endoplasmático nem pela depleção do ergosterol membranar. Palavras chave: antifúngico, benzo[ a ]fenoxazina; ergosterol; retículo endoplasmático; stress do retículo endoplasmático xiv UPC2 Sterol Uptake Control 2 WT Wild Type YEPD Yeast Extract Peptone Dextrose YEPDA Yeast Extract Peptone Dextrose Agar xv List of figures Figure 1.1.1-1: Schematic representation of the ergosterol biosynthetic pathway in S. cerevisiae……... .................................................................................................................................. 5 Figure 1.2.1-1: Schematic representation of the Ire1p pathway in yeast.. ................................. 10 Figure 1.3.1-1: Structure of BaP1 and C9 and respective MICs against a S.cerevisiae strain called PYCC 4072. ..................................................................................................................................... 13 Figure 4.1-1: Intracellular distribution of C9 in the ER. Representative fluorescence microscopy images of BY 4741 GFP-SEC66 cells treated with C9. Cells were visualized by fluorescence microscopy with a 100x objective with immersion oil. .......................................................................................... 29 Figure 4.2-1:Effect of C9 in yeast growth with and without the addition of external ergosterol. ... 31 Figure 4.3-1: Effect of C9 alone or in combination with ergosterol on cell viability. .................... 33 Figure 4.4-1: Schematic representation of the azole detection biosensor. ................................. 34 Figure 4.4-2: Effect of C9 on ERG11 expression. ..................................................................... 35 Figure 4.5-1: Effect of C9 in combination with Methyl-b-Cyclodextrin on cell viability. ................ 36 Figure 4.6-1: Distribution of membrane ergosterol in untreated cells and cells exposed to methylb-cyclodextrin. .................................................................................................................................. 37 Figure 4.6-2: Distribution of membrane ergosterol in cells exposed to C9 alone and cells exposed to C9 and methyl-b-cyclodextrin simultaneously. ............................................................................... 38 Figure 4.7-1: Schematic representation of UPRE-lacZ gene reporter. ........................................ 39 Figure 4.7-2: Assessment of unfolded protein response (UPR) machinery induction on endoplasmic reticulum (ER) in response to C9 treatment....................................................................................... 40 xvii List of tables Table 1: List of S. cerevisiae strains used in the present work. .................................................... 21 1. Introduction 3 1.1 Ergosterol, the major fungal sterol Sterols are a class of lipids widely distributed in nature being important constituents of eukaryotic cell membranes and having an essential role on both their organization and function (Dupont et al. , 2012; Hu et al ., 2017)). There are three predominant forms of sterols found in eukaryotes: mammalian sterols, phytosterols and fungal sterols. In the case of mammalian sterols, cholesterol is often the main one, while phytosterols essentially include stigmasterol, sitosterol and campesterol. On the other hand, ergosterol is the predominant sterol found in fungi (Hannich, Umebayashi, & Riezman, 2011; Hu et al ., 2017).This fungal sterol is an important component of the fungal cell membrane and presents a variety of cellular functions. In structural terms, ergosterol binds to membrane phospholipids and modulates the thickness, fluidity and permeability of the membrane itself (Dupont et al ., 2012; Rodriguez et al., 1985; Zhang & Rao, 2010; Zweytick et al . , 2000). Ergosterol is also able to promote the growth and proliferation of fungi, thus being considered a “fungal hormone” (Wangspa & Takemoto, 1998). Besides, ergosterol is also important in other fundamental processes such as cytoskeleton organization, mating, endocytosis, sporulation, among others (Kodedová & Sychrová, 2015; Zhang et al ., 2010). Moreover, it seems to be a key participant in yeast adaptation to a variety of environmental stresses. Ethanol, one of the main alcoholic fermentation products, can negatively impact many physiological mechanisms in yeast, namely their growth, viability, metabolism, transport processes, enzymatic activity, among several others (Aguilera et al ., 2006; Stanley et al ., 2010). Nevertheless, yeast have evolved to become more resistant to its toxicity. For instance, it has been shown that under an alcoholic environment, S. cerevisiae strains are able to increase the content of ergosterol in their cellular membranes, which minimizes membrane damage, maintains membrane permeability at a basal level and increases the fluidity of the membrane(Aguilera et al ., 2006; Vanegas et al ., 2012). Moreover, studies revealed that an increased content of ergosterol is involved in yeast resistance to freezing (Janković et al ., 2015; Villarreal et al ., 2018). Besides being important for many cellular routines and responses, ergosterol has also a great value from the pharmaceutical industry’s point of view since it is a precursor of vitamin D2 and steroid hormone drugs, such as cortisone and progesterone (Francavilla et al ., 2013; Huang, Cai, & Xu, 2016; Karpova et al ., 2016). 4 1.1.1 Biosynthesis of ergosterol, its regulation and importance The biosynthetic pathway that ultimately leads to the production of ergosterol is an energy-consuming process since it takes at least twenty-four ATP and sixteen NADPH molecules to produce this sterol. Interestingly, despite being a pathway that occurs in lower eukaryotes this process requires more energy than the biosynthetic process of mammalian cholesterol (Hu et al ., 2017; Kodedová & Sychrová, 2015). It is also an oxygen-requiring process. It is a complex pathway involving the participation of variety of enzymes (almost 30) known as Erg proteins, which, in turn, are encoded by ERG genes. Considering the intermediate products, this pathway can be roughly divided into three modules that involve numerous enzymatic steps and occur in the endoplasmic reticulum (ER) (Figure1.1) (Alvarez-Vasquez et al ., 2011; Kodedová & Sychrová, 2015). The first module is conserved among all eukaryotes and it comprises the production of mevalonate from three molecules of acetyl-CoA, which is achieved by the enzymatic action of acetoacetyl-CoA thiolase, encoded by ERG10 gene, hydroxymethylglutaryl-CoA synthase, encoded by the ERG13 gene and the HMG-CoA reductases, encoded by the HMG1 and HMG2 genes (Hayakawa et al ., 2017; Lv et al ., 2016; Yuan & Ching, 2014). During the second section of the pathway, mevalonate is used to synthesize farnesyl pyrophosphate, which is a significant intermediate metabolite that participates in various metabolic pathways and in the production of numerous substances in the cells. This section involves the action of enzymes such as mevalonate kinase (Erg12p), phosphomevalonate kinase (Erg8p), mevalonate pyrophosphate decarboxylase (Erg19p), IPP isomerase and geranyl/FPP synthase (Erg20p) (Hu et al ., 2017). Finally, the third and last module comprises the synthesis of ergosterol from farnesyl pyrophosphate. It is a more complex part of the pathway, requiring enzymes such as squalene synthase (Erg9p) and squalene epoxidase (Erg1p), the latter being an enzyme that catalyses the conversion of squalene to squalene epoxide, in an oxygen-dependent reaction. It is an important step, leading to the synthesis of the first sterol-like structure that, upon cyclization catalysed by lanosterol synthase/oxidosqualene-lanosterol cyclase (Erg7p), results in the formation of lanosterol, the first sterol to appear during the whole biosynthetic process. 5 Afterwards, lanosterol is converted in ergosterol through a series of reactions that involve enzymes such as cytochrome P450-dependent lanosterol 14a-demethylase (Erg11p), C-24 sterol methyltransferase (Erg6p), C24 (28) sterol reductase (Erg4p), among others (Klug & Daum, 2014). As referred above, a variety of enzymes (almost 30) take part in this complex pathway. These proteins are encoded by genes that can either be essential or non-essential genes, depending on whether their function is or not required for yeast survival. For instance, ERG9 and ERG7 are two examples of essential genes (Hu et al ., 2017; Klug & Daum, 2014). Nonetheless, even though the non-essential genes are not critical to yeast survival, their absence can impair cellular processes. Deletion strains for non-essential genes can display a disrupted ergosterol biosynthesis, which can impair many basic mechanisms. For instance, previous studies have already demonstrated that mutants in ERG6 , a non-essential gene, exhibit a slow growth as well as an impairment in membrane integrity, low tolerance to hyperosmotic stress, an Figure 1.1.1-1: Schematic representation of the ergosterol biosynthetic pathway in S. cerevisiae (Zeng et al. , 2017) 6 increased susceptibility to drugs, compromised tryptophan transport, among others (Kodedová & Sychrová, 2015; Parks & Casey, 2003). Another study reports abnormal mitochondrial structure and respiratory inability in strains harbouring deletions in the ergosterol pathway genes (Dimmer, 2002). On the other hand, deletions of ergosterol genes can also increase the resistance of the strains to therapeutically relevant antifungals (Bhattacharya, Esquivel, & White, 2018). As mentioned before, ergosterol plays a role in a variety of cell functions and for each function, a specific ergosterol concentration is required. Thus, ergosterol levels must be rigorously regulated so that these cellular functions can occur but also to avoid accumulation of free sterols, which presents a source of toxicity to the cell. This regulation occurs mainly through feedback mechanisms at transcriptional, translational and post-translational levels (Klug & Daum, 2014). For instance, in conditions of excess of sterols, the degradation of HMG-CoA reductase, an enzyme participating in the ergosterol biosynthetic pathway, is induced, decreasing mevalonate synthesis and consequently leading to the down-regulation of sterol synthesis. This degradation occurs via the proteasome degradation pathway, mediated by an endoplasmic reticulum-associated degradation (ERAD) pathway, that will be discussed further on (Hu et al. , 2017). On the other hand, oxygen levels have also a regulatory role since ergosterol synthesis requires oxygen in order to occur. Under conditions of low oxygen, they can either try to increase its synthesis or they can uptake it from the external environment. Thus, in some cases, a transcriptional up-regulation of ergosterol pathway genes is induced when oxygen levels are low or reduced. In S. cerevisiae , two transcriptional factors called Upc2 and Ecm22 bind to the sterol regulatory element (SRE), activating the expression of ergosterol synthesis enzymes (Davies & Rine, 2006; Espenshade & Hughes, 2007; Yang et al ., 2015). In other cases, when external sterol uptake is needed, Upc2 is also necessary, as it induces the expression of AUS1 and PDR11 genes, that code for ATP-binding transporters, important for the uptake of exogenous sterols (Yang et al ., 2015; Zavrel, Hoot, & White, 2013). Even with a diverse regulatory machinery, intracellular sterols can sometimes reach excessive levels under aerobic conditions and this poses a problem, since yeasts cannot degrade sterols. Thus, how can yeast cells maintain sterols at homeostatic levels, overcoming the toxic effects that arises from excessive intracellular sterols? They do it through the conversion of sterols to steryl esters, that can be stored in cytosolic lipid droplets. Moreover, they can also secrete the excess sterols as sterol acetates into the medium, which contributes not only for the detoxification but also to the removal of damaged lipids (Choudhary & Schneiter, 2012; Klug & Daum, 2014; Kohlwein, 2017). As already demonstrated, ergosterol plays a fundamental role in the overall growth and viability of yeast cells, as it is a major constituent of their cellular membrane. This turns ergosterol and its 13 oxobutyl)amino)-10-methyl-9H-benzo[ a ]phenoxazin-9-ylidene) ethanaminium chloride, or BaP1, one of the benzo[ a ]phenoxazines synthesized by our group, has had increasing attention due to its antifungal properties and its action mechanisms have been uncovered using yeast cell as a model. It was already suggested that this compound accumulates at the vacuolar membrane and in the endoplasmic reticulum of Saccharomyces cerevisiae cells. It leads to vacuolar membrane damage and vacuolar permeabilization, ultimately leading to cell death (Carvalho, Gonçalves, & Sousa, 2011; Ferreira, 2017; Lopes, 2015). Pep4, which is a vacuolar protease, has also an effector role in the cell death induced by BaP1, as it is released from the vacuole to the cytosol due to the vacuolar permeabilization caused by the benzo[ a ]phenoxazine treatment (Ferreira, 2017; Lopes, 2015). However, the specific cell death process by which this compound acts remains to be clarified, as it was already suggested that neither autophagy nor the apoptotic mitochondrial pathway are involved. A more recent study investigated another benzo[ a ]phenoxazine that was also synthesized by our group, N-(10-methyl-5-(propylamino)-9H-benzo[ a ]phenoxazin-9-ylidene)ethanaminium chloride, or as it is also named, C9 (Sousa, 2018). This compound is much more potent than BaP1, since it presents a MIC of 6.25 µM, whilst the BaP1 compound presents a MIC of 25 µM. This may be due to the only structural difference between the two benzo[ a ]phenoxazines, which is a different single substituent in the position 5 of the compound’s aromatic system. This substitution may potentiate the antifungal activity of C9, being a clear evidence of how structural differences influence the compound’s activity Figure 1.3.1.1. (Leitão, 2015). Compound R R1R2R3R4 MIC (μM) BaP1 CH2CH3H (CH 2)3CO2 Et CH3H25.0 C9 CH2CH3H(CH2)2CH3CH3H6.25 Figure 1.3.1-1: Structure of BaP1 and C9 and respective MICs against a S.cerevisiae strain called PYCC 4072. In this figure the small difference in the functional group R2 of the core benzo[ a ]phenoxazine aromatic system of both BaP1 and C9 is demonstrated. Also demonstrated are the MICs of each compound. 14 Thus, this study aimed to characterize the mechanisms of action of C9, once more making use of the yeast S. cerevisiae as a model eukaryotic cell. Just as BaP1 compound, C9 also accumulates in the ER and in the vacuolar membrane. Contrarily to what was observed during BaP1 treatment, in this case, Pep4 had a protective role during the C9-induced cell death pathway. On the other hand, it seems that there is a mitochondrial involvement in the cell death process triggered by C9 (Sousa, 2018). However, there are still many questions regarding the mode of action of this compound. Since it is a more powerful benzo[ a ]phenoxazine than BaP1 and its mode of action is less understood, this present work continued to investigate this benzo[ a ]phenoxazine through a different perspective and considering other aspects about the compound. 2. Aim of the work 17 2.1 Aims of this work Throughout the last few years, our group has been synthesizing a plethora of benzo[ a ]phenoxazines and exploring their potentialities (Leitão, 2015), aiming to discover promising compounds. In order to understand how these compounds exert their action, S. cerevisiae was extensively used as an eukaryotic cell model. As previously denoted, one of the first benzo[ a ]phenoxazines to be studied was BaP1, a compound that was later discovered to exhibit a high antifungal activity. Later on, C9, another benzo[ a ]phenoxazine, began to be explored. Interestingly, the only difference between these two compounds is a single substitution in the aromatic system, but this minor difference confers a much higher antifungal activity to C9 in comparison to BaP1. It was already found that both compounds accumulate in the vacuolar membrane and in the endoplasmic reticulum membrane. Moreover, they both seem to induce a regulated cell death. However, they present some distinctions. Although they both lead to the permeabilization of the vacuolar membrane, a mitochondrial involvement is only seen in the case of C9’s treatment. (Sousa, 2018). Lately, we have been focusing our attention on C9, since it is much more potent than BaP1 and is mechanism of action is less understood. Even though we have already answered many questions about this compound, others arise that also need an answer. Therefore, the main goal of the present work is to continue to clarify the mechanism of action of this compound and, more specifically, try to uncover its target or targets. The research done by our group has suggested that a possible target of this compound could be located at the ER membrane: an enzyme belonging to the biosynthetic pathway of ergosterol (Leitão, 2015). For this matter, we have studied ergosterol’s metabolism, since it could be affected by the compound and also the induction of ER stress. Yeast growth curves and viability assays were used to assess the effects of C9 in growth and viability of the cells, in the presence and in the absence of external ergosterol. An indirect UPR induction assay was employed to asses ER stress. Fluorescence microscopy was used to observe the intracellular localization of the compound but also to check its influence on the membrane lipid rafts, rich in ergosterol. To assess if the compound was having a direct influence on the expression on a specific ERG gene, a recently designed luminescence assay was employed (Carvalho, 2018). 3. Materials and Methods 21 3.1 Yeast strains and plasmids The Saccharomyces cerevisiae strains used in this work are listed in Table 1. BY 4741 and W303-1A were used as the wild type strains. Table 1: List of S. cerevisiae strains used in the present work. Yeast strain Genotype Source BY4741 MATa, his3 Δ 1, leu2 Δ 0, met15 Δ 0, ura3 Δ 0 EUROSCARF W303-1A MATa, ura3-52, trp1 Δ 2, leu2-3, 112 his3-11, ade2-1 EUROSCARF BF264-15D MATa, ade1,his3, leu2-3,trp1 (Costa, 2018) BF264-15D pJT30 BF264-15D; MATa, ade1,his3, leu23,trp1 ; pJT30 (URA3) (Costa, 2018) BY 4741 GFP-SEC66 MATa, his3 Δ 1, leu2 Δ 0, met15 Δ 0, ura3 Δ 0 Brian Zid BY4741 pAM10 BY4741 harboring pAM10 (URA3) (Carvalho, 2018) BY 4743 Δ ERG11 - yNluc MATa/α his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 LYS2/lys2Δ0 met15Δ0/MET15 ura3Δ0/ura3Δ0 Δ ERG11 -yNlucKanMX (Carvalho, 2018) W303-1A pDF01VBA1-YEGFP MATa, ura3-52, trp1 Δ 2, leu2-3, 112 his3-11, ade2-1 , pDF01-VBA1-YEGFP (URA3) Josef Nosek 3.2 Media and growth conditions Most S. cerevisiae cells of BY 4741 and W303-1A strains were grown in YEPDA (Yeast Extract Peptone Dextrose Agar) medium plates (1% yeast extract; 2% bactopeptone; 2% glucose and 2% agar) at 30ºC during 2 days. Afterwards, they were transferred to liquid SC (Synthetic Complete) medium (2% glucose; 0.5% (W/V) ammonium sulphate; 0.7% yeast nitrogen base without amino acids and ammonium sulphate; 0.2% dropout mix; 0.01% histidine, uracil; 0.02% leucine) and put in an orbital shaker at 200 22 rpm, at 30 ºC and with a flask volume/medium ratio of 5:1. Then, they were allowed to reach the exponential phase (OD640nm » 0.5-0.7). Some of the strains transformed with plasmids (BF264-15D pJT30, BY4741 pAM10, W303-1A pDF01-VBA1-YEGFP and BY 4741 GFP-SEC66) were grown in SC medium plates without the appropriate amino acids, depending on the strain, at 30 ºC during 2 days. After that, they were transferred to liquid SC (Synthetic Complete) medium (2% glucose; 0.5% (W/V) ammonium sulphate; 0.7% yeast nitrogen base without amino acids and ammonium sulphate; 0.2% dropout mix; appropriate amino acids) and put in an orbital shaker at 200 rpm, at 30 ºC and with a flask volume/medium ratio of 5:1. Then, they were allowed to reach the exponential phase (OD640nm » 0.5-0.7) 3.3 C9 stock solution preparation and treatment N-(10-methyl-5-(propylamino)-9H-benzo[ a ]phenoxazin-9-ylidene)ethanaminium chloride (C9) was dissolved in dimethyl sulfoxide (DMSO) to a final concentration of 4906 µM. As previously mentioned, yeast cells were grown in SC medium at 30 ºC with agitation at 200 rpm, until they reached the exponential phase (OD640 nm » 0.5 – 0.6). After that, cells were collected and centrifuged for 3 minutes at 5000 rpm. Then, the pellet was washed with deionized sterile water, centrifuged again and resuspended in fresh SC medium or in another medium, depending on the assay in question. C9 treatment was performed by adding C9 to the resuspended cells at different final concentrations: 2.5, 7.5, 12.5, 25 µM. The same volume of DMSO (DMSO concentration was never superior to 1%) was added to another tube that served as a negative control. Incubation took place at 30 ºC for a defined period of time, depending on the assay that was being performed. Also dependent on the assay were the time-points at which the cells were collected. In every assay the 0 minute sample was collected before adding the compound and DMSO. 3.4 Ergosterol stock solution preparation and treatment Ergosterol stock solution was prepared by dissolving ergosterol in absolute ethanol to a final concentration of 4 mg/mL, using an ultrasonic bath. Yeast cells were grown as previously described. 29 4.1 Confirming intracellular distribution of C9 Preliminary studies using S. cerevisiae as a model have revealed that benzo[ a ]phenoxazines accumulate at the vacuolar and endoplasmic reticulum membranes. More specifically, it was possible to establish a correlation between the potency of each compound and its site of accumulation; benzo[ a ]phenoxazines with lower minimum inhibitory concentrations (MIC) tend to accumulate more in the membrane of the endoplasmic reticulum than in the membrane of the vacuole, while benzo[ a ]phenoxazines with a higher MIC accumulate mostly in the vacuolar membrane (Leitão, 2015) In order to confirm that C9’s distribution two yeast mutants were used. One expressed a GFP-tagged ER membrane protein called Sec66, a protein involved in the post-translational translocation of proteins to the ER (Lee & Heitman, 2012). The other mutant expressed a GFP-tagged vacuolar membrane protein called Vba1, a transporter protein essential for the uptake of basic amino acids in the vacuolar membrane (Jakubkova et al ., 2016; Sekito et al ., 2008). Below, in Figure 4.1.1 it is possible to observe the results from the observation of the mutant cells expressing the GFP-tagged ER membrane protein Sec66. GFP-Sec66 C9 Overlay Figure 4.1-1: Intracellular distribution of C9 in the ER. Representative fluorescence microscopy images of BY 4741 GFP-SEC66 cells treated with C9. Cells were visualized by fluorescence microscopy with a 100x objective with immersion oil. 30 In terms of the ER, fluorescence microscopy images revealed that this compound co-localizes in the ER membrane, indeed. However, after observing the cells expressing the GFP-tagged vacuolar membrane protein, we were not able to conclude that the compound was in fact accumulated at the vacuolar membrane. This was probably due to some problems with the strain itself, since it was not even possible to observe any fluorescence emitted by the VBA1-GFP tagged cells. 4.2 External ergosterol seems to exacerbate the toxic effect of C9 on cell growth As previously referred, preliminary studies have revealed that less active benzo[ a ]phenoxazines had a tendency to accumulate more in the vacuole membrane, while the most active compounds would accumulate simultaneously in the membrane vacuole and in the ER, with the latter being the major site of accumulation (Ferreira, 2017; Leitão et al ., 2015; Leitão, 2015). The first results of this present work have shown and confirmed that C9 seems to accumulate more in the ER membrane. Thus, it seems that the antiproliferative activity of this benzo[ a ]phenoxazine can correlate with their ability to accumulate in the ER and that the main target of these compounds could be located in this organelle, rather than the vacuole membrane. The ER membrane contains several enzymes important for cell growth that could be potential targets of these compounds. One of those enzymes, oxidosqualene-lanosterol cyclase, is a key enzyme in the ergosterol biosynthesis pathway and has a substrate that shares with C9 the fact that it accommodates a 4 -ring system in his structure (Mo et al ., 2003). We hypothesized that if C9 would, somehow, be targeting this enzyme, cells would have their growth affected due to limiting ergosterol content. In line with this, we speculated that an external supplementation of ergosterol to the medium would render cells more resistant to C9’s deleterious effect. To test this, growth assays with BY4741 WT cells treated with different concentrations of ergosterol, C9 or the combination of both, were performed during 9 h. Finally, the specific growth rate was calculated and the results are represented below in Figure 4.2.1. 31 Considering the results, it is possible to observe that the inhibition of the cells’ growth is proportional to the concentration of the compound; the higher the compound’s concentration, the more inhibited the growth of the cells, and the lower the specific growth rate (Figure 4.2 1 A). On the other hand, looking at the results regarding the treatment of the cells with increasing concentrations of ergosterol, we can observe that the specific growth rate of cells treated with increasing concentrations of ergosterol are similar to the specific growth rate obtained for untreated cells, suggesting that cell growth was not affected by the addition of external ergosterol to the medium (Figure 4.2 1 B). However, in the case of cells treated with 7.5 μM of C9 and increasing concentrations of ergosterol simultaneously, it seems that, although not significant, there is a growth inhibition in comparison with cells treated only with 7.5 μM of C9. (Figure 4.2 1 C). Similar results were obtained for cells treated with 12.5 μM of C9 and increasing concentrations of ergosterol, simultaneously. However, in this case, the growth inhibition is significant (Figure 4.2 1 D). 0.0 0.1 0.2 0.3 0.4 Specific Growth Rate (h-1) Control EtOH C9 (7.5 µM) Ergost. (5 µg/mL) + C9 (7,5 µM) Ergost. (10 µg/mL) + C9 (7,5 µM) Ergost. (20 µg/mL) + C9 (7.5 µM) 0.0 0.1 0.2 0.3 0.4 Specific Growth Rate (h-1) Control EtOH C9 (12.5µM) Ergost. (5 µg/mL) + C9 (12,5 µM) Ergost. (10 µg/mL) + C9 (12,5 µM) Ergost. (20 µg/mL) + C9 (12.5 µM) * **** 0.0 0.1 0.2 0.3 0.4 Specific Growth Rate (h-1) Control EtOH C9 (7.5 µM) C9 (12.5µM) C9 (25 µM) **** **** 0.0 0.1 0.2 0.3 0.4 Specific Growth Rate (h-1) Control EtOH Ergost. (5 µg/mL) Ergost. (10 µg/mL) Ergost. (20 µg/mL) (D) (A) (C) (B) Figure 4.2-1:Effect of C9 in yeast growth with and without the addition of external ergosterol. Wild type BY4741 cells were incubated with C9 (2.5, 7.5 and 25 μM) and ergosterol (5,10 and 20 μg/mL), either alone or in combination for 9h. At different time points (90, 180, 270, 360, 450 and 540 minutes) OD640nm was measured and the specific growth rate was posteriorly calculated.; (A) cells treated with different C9 concentrations were compared with cells with no treatment (control); (B) cells treated with various concentrations of ergosterol compared with untreated cells (control); (C) cells treated with 7.5 μM of C9 compared with cells treated with both 7.5 μM of C9 and different concentrations of ergosterol; (D) cells treated with 12.5 μM of C9 compared with cells treated with both 12.5 μM of C9 and different concentrations of ergosterol. The reported values represent means ± SD of at least three independent experiments. Statistical analysis was performed by One-way ANOVA test.****P<0.0001,***P<0.001, **P<0.01,*P<0.5. 32 Since for the tested concentrations, ergosterol, per si did not display any toxicity, the results suggest that the external ergosterol addition potentiates C9’s toxicity. In fact, opposite to our hypothesis, the effect of C9 on cell growth does not seem to be due to a limiting ergosterol concentration, but rather due to the increase of ergosterol, that negatively affects cell growth in the presence of C9. 4.3 External supplementation of ergosterol has no protective effect on cell viability upon treatment with C9 Initially, we hypothesized that C9 could be targeting ergosterol biosynthetic pathway, more specifically, it could be targeting an enzyme that participates in this pathway and has a key role in the biosynthesis of ergosterol. Thus, if this was definitely C9’s mechanism, then cellular ergosterol would probably decrease and the growth of the cells would be compromised, since ergosterol is very important for the maintenance of the cell membrane, contributing to its fluidity, permeability among other functions (Kohlwein, 2017; Zhang & Rao, 2010). Nevertheless, our results suggest that after exposing cells to C9, an external supplementation of ergosterol has an effector role rather than a protective one, since cells that receive an external supplementation of ergosterol during treatment with C9 have a lower specific growth rate than the ones who are exposed to C9 alone, indicating that external ergosterol could be increasing the compound’s toxicity. To further understand the role of increasing ergosterol concentration in C9 effects, we aimed to elucidate if, besides influencing -growth inhibition, an external supplementation of ergosterol could also have an effect in C9-induced cell death. In line with this, viability of cells treated with different concentrations of C9, either alone or in combination with different concentrations of ergosterol, was evaluated during a 2 h assay. The resulting percentages of CFU over time are represented below in Figure 4.3.1. 33 It is possible to observe that after 120 minutes of treatment, as expected, there was a decrease in the viability of cells treated only with C9. This effect was enhanced as the concentration of C9 increased. The higher the concentration of C9, the bigger the decrease on the viability of the cells. When cells were exposed to the same concentration of C9 but also to ergosterol, there was a slight increase in cell’s viability. However, the differences in the viability of cells treated only with C9 and cells treated with both C9 and ergosterol were not statistically significant. 4.4 C9 is not leading to ergosterol depletion Very recently, and in a distinctive line of work, another group from our laboratory has developed a reporter yeast strain meant for detection of azole compounds, specifically, tebuconazole, in aqueous samples. Many research studies showed that, upon treatment with tebuconazole and other fungicides, transcripts of ERG11 gene were amongst the most up-regulated genes, which is in agreement with the fact that, as previously referred in the introduction section, azole fungicides target the biosynthetic pathway of ergosterol, namely, the conversion of lanosterol to ergosterol, for which the cytochrome P450dependent lanosterol 14a-demethylase enzyme, encoded by ERG11 gene, is strictly necessary (Prasad, Shah, & Rawal, 2016). Overall, this leads to ergosterol’s depletion in the cells. In line with this, ERG11 gene was selected to design the reporter biosensor for the detection of azole compounds. The S.cerevisiae strain used to design this biosensor was a diploid strain called BY 4743 with an ERG11 deletion. Moreover, this strain was transformed with a cassette that contained the gene of a reporter protein called 030 60 90 120 0 100 200 300 Time (minutes) %CFU DMSO + Ergosterol (10 µg/mL) C9 (25 µM) C9 (25 µM) + Ergosterol (10 µg/mL) C9 35 µM C9 (35 µM) + Ergosterol (10 µg/mL) C9 40 µM C9 (40 µM) + Ergosterol (10 µg/mL) Figure 4.3-1: Effect of C9 alone or in combination with ergosterol on cell viability. Analysis of the viability of BY 4741 WT cells treated with different concentrations of C9 (25, 35 and 40 M) and ergosterol (10 g/mL). DMSO and ergosterol (negative control) were added 30 minutes before C9 in combination with ergosterol was added. Cell viability was evaluated by CFU counting at different time points (0, 30, 60, 90 and 120 minutes). The reported values represent means ± SD of at least three independent experiments. Statistical analysis was performed by the two-way ANOVA test to compare cells treated with C9 alone with cells treated with C9 and ergosterol simultaneously, for each time of the treatment. 34 nanoluciferase gene optimized for yeast expression (yNluc) (Masser et al. , 2016) under the control of the ERG11 promoter (Carvalho, 2018). The correct functioning of this biosensor was successively achieved, as it was responsive to tebuconazole treatment. A simplified explanation of how this biosensor works is shown in Figure 4.4.1. In order to uncover if C9 could also be targeting ergosterol synthesis we employed this reporter strain. Firstly, control assays were performed to validate the functioning of the reporter strain. Untransformed BY 4741 and BY 4741 cells harbouring the pAM10 plasmid were used. This plasmid contained the yNluc gene that was under the control of heat-shock responsive promoter. For this assay we treated BY 4743% ERG11yNluc cells with C9 and tebuconazole during 4 hours. At 2h and 4h time points cells were mixed with luciferase substrate in white opaque 96-well microplates and after a 1h incubation, bioluminescence levels were recorded. The results of this assay are represented below in Figure 4.4.2. ERG11 promoter Luciferase (reporter gene) mRNA Tebuconazole Expression of luciferase (reporter protein) + External addition of luciferase substrate Bioluminescence Figure 4.4-1: Schematic representation of the azole detection biosensor. When cells are treated with tebuconazole, ergosterol production is affected since tebuconazole will target cytochrome P450-dependent lanosterol 14a-demethylase enzyme, which catalyses the conversion of lanosterol to ergosterol. Thus, ERG11 promoter is activated, which, in turn, will lead to the transcription and translation of the luciferase protein. Then, after an external addition of the enzyme’s substrate, there is the production of a bioluminescence signal. 35 As expected, treating the cells with 4 µg/mL of tebuconazole lead to a high luciferase signal, meaning that the fungicide is inducing the expression of ERG11 . In the case of cells treated with 25 μM of C9 it is possible to observe that there was a slight increase of luciferase signal in comparison with the negative control. Nonetheless, despite the levels of luciferase observed for C9-treated cells, it is possible to conclude that these bioluminescence levels are still very low when compared with the levels of luciferase signals observed in cells treated with tebuconazole. Additionally, there were no significant statistical differences between all of the conditions. Thus, the overall results suggest that C9 does not affect ERG11 expression, thus not leading to a decrease in ergosterol content. 4.5 Depletion of ergosterol from the plasma membrane increases cell viability after treatment with C9 We next raised the question whether membrane ergosterol could be a key participant in the action of C9. To answer this question, we evaluated if decreasing ergosterol at the plasma membrane could be affecting the toxicity of this compound. Control Control + Luc. Teb. 4 µg/mL Teb. 4 µg/mL + Luc. C9 25 µM C9 25 µM + Luc -2 -1 0 1 2 3 4 5 6 7 8 Bioluminescence (a.u.) T 2h T 4h Figure 4.4-2: Effect of C9 on ERG11 expression. BY 4743ERG11yNluc cells were incubated with 25 μM of C9 and 4 μg/mL of tebuconazole during 4 hours. Untreated cells and cells treated with tebuconazole were used as negative and positive controls, respectively. At 2 and 4h of treatment, cells were collected, centrifuged and resuspended in SC-Glu pH 8 medium and then mixed 1:10 with the luciferase substrate in white opaque 96-well microplates (OptiPlate-96). Finally, after a 1h incubation with the substrate, luciferase signal was measured. The reported values represent means ± SD of at least three independent experiments. Statistical analysis was performed by the two-way ANOVA test to compare cells treated with C9 with cells treated with tebuconazole, for each time point of the treatment. 36 For such we exposed cells to two different concentrations of C9 (25 μM and 40 μM) and, at the same time, to 50 mg/mL of a sterol-depleting agent, methyl-b-cyclodextrin (MbCD), that extracts sterols from the plasma membrane of the cells. Then, the cell’s viability was evaluated during a 2 h assay. The results are presented in Figure 4.5.1. Once more, as expected, cells treated with 25 µM and 40 µM of C9 presented a decrease in their viability during the 2 hours of treatment with the compound. When cells were exposed to the same concentrations of C9 but were also treated with MbCD, there was no decrease of viability, some growth being even observed during the course of the assay. Furthermore, treatment with MbCD seems to be even more effective on higher concentrations of C9, which is suggested by the results of the statistical analysis. Overall, these results strongly support that MbCD exerts a protective effect against C9 treatment. 4.6 Lipid rafts distribution upon C9’s treatment Disrupting membrane sterols is amongst the main antifungal mechanisms displayed by various compounds and drugs that have the ability to kill fungal cells. Specifically, these antifungal agents target ergosterol (fungal sterol) in different ways; depending on their type, they can either bind to key enzymes of the ergosterol biosynthetic pathway, ultimately leading to the inhibition of ergosterol production or they can also bind to ergosterol itself and form a pore-like complex that results in cell’s membrane disruption 030 60 90 120 0 100 200 300 Time (minutes) %CFU DMSO Met-β-cyc. (50 mg/mL) C9 25 µM C9 25 µM + Met-β-cyc. (50 mg/mL) C9 40 µM C9 40 µM + Met-β-cyc. (50 mg/mL) Figure 4.5-1: Effect of C9 in combination with Methylb -Cyclodextrin on cell viability Analysis of the viability of BY 4741 WT cells treated with two concentrations of C9 (25 40 μM) and MbCD (50 mg/mL). DMSO (negative control) was added 30 minutes before C9 in combination with MbCD was added. Cell viability was evaluated by CFU counting at different time points (0, 30, 60, 90 and 120 minutes). 30 min: (C9 40 µM vs C9 40 µM + MbCD)*; 60 min: (C9 25 µM vs C9 25 µM + MbCD)* , (C9 40 µM vs C9 40 µM + MbCD)*; 90 min: (C9 25 µM vs C9 25 µM + MbCD)**, (C9 40 µM vs C9 40 µM + MbCD)*; 120 min: (C9 25 µM vs C9 25 µM + MbCD)**, (C9 40 µM vs C9 40 µM + MbCD)****. The reported values represent means ± SD of at least three independent experiments. Statistical analysis was performed by the two-way ANOVA test to compare cells treated with C9 alone with cells treated with C9 and MbCD simultaneously, for each time of the treatment.****P<0.0001,***P<0.001, **P<0.01,*P<0.5. 37 and the outflow of cytosolic contents (Prasad et al ., 2016). Normally, ergosterol exists in cellular membranes as a part of lipid rafts, which are membrane microdomains enriched also in sphingolipids and transmembrane proteins and important for the correct function of the yeast cell. These domains are involved in signal transduction pathways and membrane protein trafficking (Curwin et al ., 2013; Mollinedo, 2012). In line with our hypothesis of ergosterol being a target of C9, we sought to monitor the sterol distribution, particularly in lipid rafts. For this assay, cells were stained with filipin, a polyene antibiotic that exhibits fluorescent properties and that bind to sterols, in this case, to ergosterol, one of the major constituents of lipid rafts (Van Leeuwen et al ., 2008). To validate our approach, we used methyl-b-- cyclodextrin (MbCD) as positive control, a reagent known to extract sterols from membranes. The main goal of this assay was to observe sterols distribution in cells either treated with C9 alone or C9 in combination with MbCD. Then, it was necessary to count at least 300 cells for each condition. Unfortunately, this was not possible due to the extreme instability of the filipin staining. During the course of the visualization of the cells under the microscope some of the conditions had no staining at all, which must have been due to some unwanted light that was entering the microscope room or even the excessive time that some of the slides took to be observed in the microscope. Therefore, the only analysis that can be made is a qualitative one, rather than a quantitative. Below in Figures 4.6.1 and 4.6.2. are representative images of each condition. Control (Untreated cells) MβCDA B Figure 4.6-1: Distribution of membrane ergosterol in untreated cells and cells exposed to methylb -cyclodextrin. Representative images of untreated W303-1A WT cells and cells treated with 50 mg/mL of methyl-b-cyclodextrin, for 1 h. Then, they were stained with filipin during 15 minutes in the dark and finally, they were visualized using fluorescence microscopy. (A) untreated cells; (B) cells treated with MbCD. 38 If we observe cells from the positive control (Figure 4.6.2 B) it is possible to conclude that it did not work as expected and, as such, is not suitable for the validation of this assay. Normally, after treatment with MbCD, cells should have all or, at least, the majority of its membranal ergosterol depleted and localized in small aggregations inside the cell. However, looking at the representative images of positive control cells we see the majority of the cells still have ergosterol localized in the cellular membrane, almost like a ring-like structure around the cell. Moreover, looking at cells treated with C9 (Figure 4.6.2 C) it is possible to see some of the cells with cluster-like structures inside the cell, which could mean that the treatment with C9 lead to the formation of these structures. Finally, in the case of cells exposed to both C9 and MbCD (Figure 4.6.2 D), it is possible to observe that a vast number of cells presents the same ring-like structures visualized for cells treated only with MbCD. Thus, it seems that C9, in the presence of MbCD is no longer able to induce the formation of the cluster-like structures, which represent accumulations of ergosterol. 4.7 C9 does not seem to induce ER stress As we have already shown, this compound accumulates mostly at the ER of the yeast cells. As such, we hypothesized that the compound could be impairing this organelle’s function contributing to its C9 (25 μM) + MβCD C9 (25 μM) C D Figure 4.6-2: Distribution of membrane ergosterol in cells exposed to C9 alone and cells exposed to C9 and methylb -cyclodextrin simultaneously. Representative images of W303-1A WT cells that were incubated with 25 µM of C9, 50 mg/mL of methyl-b-cyclodextrin, and the combination of the two compounds for 1 h. Then, they were stained with filipin during 15 minutes in the dark and finally, they were visualized using fluorescence microscopy. (C) cells treated with C9; (D) cells treated with C9 + MbCD. 45 lanosterol synthase/oxidosqualene-lanosterol cyclase enzyme and study the effect of C9 in the growth of this strain. However, since ERG7 is an essential gene, the strain for this study would have to be diploid and an effect of ERG7 deletion on growth may be unclear (Lees et al ., 1995). As external ergosterol seemed to exacerbate the toxic effect of C9 on cell’s growth, we next aimed to know what would be the role of ergosterol in C9-induced cell death. Thus, we assessed the viability of cells either treated witch C9 alone, ergosterol alone or the combination of both. The results showed that, again, as expected, C9 alone induces loss of cell viability. Ergosterol alone does not seem to affect the viability of the cells. However, when cells were treated with the combination of C9 and ergosterol their viability had a slight increase over the course of the assay, when compared with treatment with C9 alone, but this increase was not statistically significant. This indicates that ergosterol supplementation is not able to protect the cells from C9-induced cell death as well. In the future, it would also be important to check the viability of strains lacking ERG genes and see if their viability would increase upon treatment with C9 alone and in combination with ergosterol. Another group from our laboratory had recently constructed a reporter yeast strain for detection of azole compounds in aquatic samples. This reporter system is meant to detect the presence of tebuconazole and other azole fungicides whose target is already known: ERG11 gene, that codes for cytochrome P450-dependent lanosterol 14a-demethylase enzyme, an essential enzyme in the biosynthetic ergosterol pathway. Basically, when the reporter strain is in the presence of azole compounds, the ERG11 promoter, which is regulating the expression of a reporter protein called luciferase, is activated, leading to the transcription and translation of this protein, that, when in contact with its substrate, emits bioluminescence. (Carvalho, 2018; Masser et al ., 2016). Although this reporter strain was constructed for ecologic applications, it can also be used to check if a specific compound is affecting the expression of ERG11 , which could be suggestive of ergosterol depletion. In line with this, we have employed this reporter system to understand if C9 could be targeting ERG11 . If so, this could indicate that C9’s treatment was leading to ergosterol’s depletion. The overall results seem to show that C9 is not leading to the depletion of ergosterol, since there was not an induction of ERG11 gene expression. However, the results may not be as accurate as they could be because the luciferase substrate was very unstable and it is possible that, in some of the replicates, the substrate was not functioning correctly. Although the Nano-Glo substrate coupled with NanoLuc (the luciferase that is expressed by the reporter yeast strain) seems to be one of the best reporter assay systems nowadays, it 46 has to be employed very carefully (Brescia, Banks, & Landreman, 2016). Thus, in the future, and if this reporter assay system continues to be one of the most well-suited, it may be necessary to repeat this assay but with more restrictions. For example, for each independent assay, a fresh solution of luciferase substrate should be made in order to make sure that the substrate is unaltered. We have also questioned if C9 could be targeting cellular membrane ergosterol. To evaluate if extraction of ergosterol from the cellular membrane could, somehow, play a protector role against C9induced cell death, cells treated with two concentrations of C9 were exposed to a sterol-depleting agent, methyl-b-cyclodextrin (MbCD), that would extract the ergosterol from the plasma membrane. Then, the viability of such cells was evaluated. Interestingly, the results revealed that the extraction of ergosterol from the membrane renders the cells more resistant to C9’s treatment. In literature, there are many reports stating that cyclodextrins, specifically, methyl-b-cyclodextrin, have been gaining attention due to their possible applications as drug delivery systems, since the interior of these cyclic oligosaccharides is hydrophobic and can encapsulate molecules that have low affinity with aqueous environments, thus having problems entering the cells (Gómez et al ., 2018; Jóhannesson, Stefánsson, & Loftsson, 2016). C9 is a relatively hydrophilic compound, thus having no difficulties in entering the cells. Therefore, C9 will not have a high affinity with methyl-b-cyclodextrin’s lipophilic interior. In line with this, it is acceptable to think that treatment with MbCD would not exacerbate C9’s effects, at least from this perspective. Thus, it seems reasonable to suggest that a possible reason why MbCD would play a protective role against C9induced cell death is because, by extracting membrane ergosterols, it is, somehow, preventing C9 from targeting it. Thus, it would be interesting to further understand if it is possible to validate this hypothesis. For instance, this could be achieved through a fluorescence microscopy assay, in which the intracellular distribution of C9 should be observed, in order to understand if, after MbCD’s treatment, C9 is located in the ER and vacuolar membranes, or not. In line with the previous results, we went on to visualize the behaviour of cellular membrane lipid rafts after treating the cells with C9 alone and in combination with MbCD. In this fluorescence microscopy assay, cells were stained with filipin, a polyene antibiotic that exhibits fluorescent properties and that bind to sterols, in this case, to ergosterol, one of the major constituents of lipid rafts. This assay was supposed to be a quantitative assay, in order to understand which is the real phenotype displayed by the cells in each condition. However, filipin is an extremely unstable probe, suffering photobleaching very easily, even with the presence of Vectashield, a liquid medium that is applied to microscope slides in order to prevent the probes from suffering photobleaching caused by light exposure. We were not able to phenotypically 47 quantify all the conditions since some of them had already lost the filipin fluorescence at the time of the microscopic observation. Thus, in the future, it is mandatory to perform this assay during a lower period of time. Despite all of these problems, we still analysed the obtained images in order to, at least, understand what is the major phenotype displayed by the cells in each condition. The validation of the obtained results became impaired due to the fact that our positive control was not functioning correctly. It is important to have in mind that these results are preliminary. Thus, in the future, it is mandatory to repeat these assays. Probably, the positive control did not function correctly because it needed more time to have an effect on the cells, which is something to consider further on. A good example of how MbCD is supposed to act is reported in a very interesting assay published by Pacheco and co-workers (Pacheco et al ., 2013). The fact that C9 accumulates mostly at the ER membrane can indicate that this organelle’s function may be impaired after the cell’s exposure to C9. In terms of the organelles’ morphology, C9 does not seem to be disrupting its organization. Thus, C9 could be impaired cellular processes that take place in the ER and inducing ER stress. As no one had addressed this question before, we evaluated whether C9 could be causing ER stress. For this, we used an indirect assay that relates b-galactosidase activity with the induction of the UPR, a cellular process that is initiated upon ER stress. The obtained results seem to indicate that C9 is not an ER-stress inducer, since there is no induction of UPR in cells treated with the compound. To further confirm if C9 is not, indeed, an ER-stress inducer, viability of cells with deletions in HAC1 and IRE1 genes, which are very important for the UPR process, should be evaluated, upon treatment with C9. If C9 is not, in fact an inducer of ER stress, then these deficient strains should not be more resistant to the treatment. Another way to validate the hypothesis that C9 is not a stress inducer is to check if there is splicing of the transcription factor Hac1, using, for this matter, an RT-PCR assay. If there is no induction of the UPR, then there is no splicing of Hac1 (Chaillot et al ., 2015). Aside from ER stress, there are also other important cellular other processes occurring in the ER that could be affected by the action of C9. For instance, a crucial process that occurs in the ER is protein Nlinked protein glycosylation, a post-translational modification(Wang et al ., 2017). One of the assays that could be used to assess whether this important process is being affected or not by C9, is Western blotting, since glycosylation confers a higher molecular weight to the protein: the non-glycosylated proteins will migrate faster that glycosylated ones (Buentzel & Thoms, 2017). Tunicamycin would have to be used as a negative control, as it inhibits N-linked glycosylation (Samali et al. , 2010) 48 In essence, we demonstrate that, contrarily to what we first purposed, an external supplementation of ergosterol to the medium during C9’s treatment does not act as a protection against C9’s deleterious effects. Instead, it increases C9’s ability to provoke growth impairment. However, this external supplementation of ergosterol is neither deleterious to the viability of the cells, nor protective. Furthermore, cytochrome P450-dependent lanosterol 14a-demethylase does not seem to be a target of C9. Thus, it is possible that C9 is not targeting enzymes existing in the ER membrane and important for ergosterol synthesis, as we first proposed, although the induction of other enzymes of the pathway should be tested. On the other hand, it is possible that C9 is, instead, targeting cellular membrane ergosterol, since the extraction of ergosterol from the cell’s membrane rendered the cells more resistant to C9induced cell death. Although C9 accumulation occurs mainly at the ER membrane there is no disruption of the morphology of the organelle, as observed by the first images of C9 localization in the ER membrane. Moreover, C9’s mechanism of action is not mediated by ER stress, since we employed an assay to evaluate UPR induction after C9’s treatment and the results show that UPR is not induced. Hopefully, and although there are still many questions to be asked and answered about this compound, this work will lead us in new directions in order to fully understand how it is able to exert its antifungal activity. We hope that the different approaches employed in the present work will help us to achieve our ultimate goal, which is the use of benzo[ a ]phenoxazines as potent antifungal agents 6. References 51 Abe, A., Yamane, M., & Tomoda, A. (2001). Prevention of growth of human lung carcinoma cells and induction of apoptosis by a novel phenoxazinone, 2-amino-4,4α-dihydro-4α,7-dimethyl-3Hphenoxazine-3-one. Anti-Cancer Drugs , 377–382. doi.org/10.1097/00001813-20010400000011 Aebi, M. (2013). N-linked protein glycosylation in the ER. Biochimica et Biophysica Acta - Molecular Cell Research . doi.org/10.1016/j.bbamcr.2013.04.001 Aguilera, F., Peinado, R. A., Millán, C., Ortega, J. M., & Mauricio, J. C. (2006). Relationship between ethanol tolerance, H+-ATPase activity and the lipid composition of the plasma membrane in different wine yeast strains. International Journal of Food Microbiology , 110 (1), 34–42. doi.org/10.1016/J.IJFOODMICRO.2006.02.002 Aitman, T. J., Boone, C., Churchill, G. A., Hengartner, M. O., MacKay, T. F. C., & Stemple, D. L. (2011). The future of model organisms in human disease research. Nature Reviews Genetics . doi.org/10.1038/nrg3047 Alberti, A., Bolognese, A., Guerra, M., Lavecchia, A., Macciantelli, D., Marcaccio, M., … Paolucci, F. (2003). Antitumor Agents 4. Characterization of Free Radicals Produced during Reduction of the Antitumor Drug 5H-Pyridophenoxazin-5-one: An EPR Study. Biochemistry , 11924–11931. doi.org/10.1021/bi0346087 Alcazar-Fuoli, L., & Mellado, E. (2013). Ergosterol biosynthesis in Aspergillus fumigatus : its relevance as an antifungal target and role in antifungal drug resistance. Frontiers in Microbiology , 3 , 439. doi.org/10.3389/fmicb.2012.00439 Alvarez-Vasquez, F., Riezman, H., Hannun, Y. A., & Voit, E. O. (2011). Mathematical Modeling and Validation of the Ergosterol Pathway in Saccharomyces cerevisiae . PLoS ONE , 6 (12), e28344. doi.org/10.1371/journal.pone.0028344 Austriaco, N. (2012). Endoplasmic reticulum involvement in yeast cell death. Frontiers in Oncology , 2 , 87. doi.org/10.3389/fonc.2012.00087 Barros, S., Frade, V., Moura, J., & Gonçalves, M. S. (2006). Benzo[ a ]phenoxazine dyes as new fluorescent labels of L-valine. In Proceedings of The 10th International Electronic Conference on Synthetic Organic Chemistry (p. 1411). Basel, Switzerland: MDPI. doi.org/10.3390/ecsoc-10-01411 Beh, C. T., & Rine, J. (2004). A role for yeast oxysterol-binding protein homologs in endocytosis and in the maintenance of intracellular sterol-lipid distribution. Journal of Cell Science , 117 (14), 2983– 2996. doi.org/10.1242/jcs.01157 Berner, N., Reutter, K.-R., & Wolf, D. H. (2018). Protein Quality Control of the Endoplasmic Reticulum 52 and Ubiquitin–Proteasome-Triggered Degradation of Aberrant Proteins: Yeast Pioneers the Path. Annual Review of Biochemistry , 87 (1), 751–782. doi.org/10.1146/annurev-biochem-062917012749 Bhattacharya, S., Esquivel, B. D., & White, T. C. (2018). Overexpression or Deletion of Ergosterol Biosynthesis Genes Alters Doubling Time, Response to Stress Agents, and Drug Susceptibility in Saccharomyces cerevisiae . MBio , 9 (4), e01291–e018. doi.org/10.1128/mbio.01291-18 Brescia, P. J., Banks, P., & Landreman, A. (2016). Comparative Analysis of Dual-Luciferase ® Assay Technologies in a High Throughput Microplate Format . www.biotek.com Brodsky, J. L., & Morrow, M. W. (2010). Protein Import into the Yeast Endoplasmic Reticulum: Methods. In Encyclopedia of Life Sciences . Chichester, UK: John Wiley & Sons, Ltd. doi.org/10.1002/9780470015902.a0002619.pub3 Brodsky, J. L., & Skach, W. R. (2011). Protein folding and quality control in the endoplasmic reticulum: Recent lessons from yeast and mammalian cell systems. Current Opinion in Cell Biology . doi.org/10.1016/j.ceb.2011.05.004 Buentzel, J., & Thoms, S. (2017). The use of glycosylation tags as reporters for protein entry into the endoplasmic reticulum in yeast and mammalian cells. In Methods in Molecular Biology . doi.org/10.1007/978-1-4939-6937-1_21 Carvalho, C. (2018). Novel yeast biosensors for environmental monitoring . University of Minho. Carvalho, R. E. S., Gonçalves, M. S. T., & Sousa, M. J. (2011). Marking patterns and potential targets of action of a benzo[ a ]phenoxazinium chloride with antiproliferative activity. In MicroBiotec11 National Congress . http://repositorium.sdum.uminho.pt/handle/1822/17290 Chaillot, J., Tebbji, F., Remmal, A., Boone, C., Brown, G. W., Bellaoui, M., & Sellam, A. (2015). The Monoterpene Carvacrol Generates Endoplasmic Reticulum Stress in the Pathogenic Fungus Candida albicans . Antimicrobial Agents and Chemotherapy , 59 (8), 4584–4592. doi.org/10.1128/AAC.00551-15 Choudhary, V., & Schneiter, R. (2012). Pathogen-Related Yeast (PRY) proteins and members of the CAP superfamily are secreted sterol-binding proteins. Proceedings of the National Academy of Sciences , 109 (42), 16882–16887. doi.org/10.1073/pnas.1209086109 Cox, D. J., Strudwick, N., Ali, A. A., Paton, A. W., Paton, J. C., & Schröder, M. (2011). Measuring signaling by the unfolded protein response. Methods in Enzymology , 261–292. doi.org/10.1016/B978-0-12385928-0.00015-8 Curwin, A. J., LeBlanc, M. A., Fairn, G. D., & McMaster, C. R. (2013). Localization of Lipid Raft Proteins 53 to the Plasma Membrane Is a Major Function of the Phospholipid Transfer Protein Sec14. PLoS ONE , 8 (1), e55388. doi.org/10.1371/journal.pone.0055388 Davies, B. S. J., & Rine, J. (2006). A role for sterol levels in oxygen sensing in Saccharomyces cerevisiae . Genetics , 174 (1), 191–201. doi.org/10.1534/genetics.106.059964 De Magistris, P., & Antonin, W. (2018). The Dynamic Nature of the Nuclear Envelope. Current Biology , 28 (8), R487–R497. doi.org/10.1016/j.cub.2018.01.073 Demirci, E., Junne, T., Baday, S., Berneche, S., & Spiess, M. (2013). Functional asymmetry within the Sec61p translocon. Proceedings of the National Academy of Sciences , 18856–18861. doi.org/10.1073/pnas.1318432110 Denoth Lippuner, A., Julou, T., & Barral, Y. (2014). Budding yeast as a model organism to study the effects of age. FEMS Microbiology Reviews . doi.org/10.1111/1574-6976.12060 Dimmer, K. S. (2002). Genetic Basis of Mitochondrial Function and Morphology in Saccharomyces cerevisiae . Molecular Biology of the Cell , 847–853. doi.org/10.1091/mbc.01-12-0588 Dupont, S., Lemetais, G., Ferreira, T., Cayot, P., Gervais, P., & Beney, L. (2012). Ergosterol biosynthesis: A fungal pathway for life on land? Evolution , 2961–3008. doi.org/10.1111/j.15585646.2012.01667.x English, A. R., & Voeltz, G. K. (2013). Endoplasmic Reticulum Structure and Interconnections with Other Organelles. Cold Spring Harbor Perspectives in Biology , 5 (4), a013227–a013227. doi.org/10.1101/cshperspect.a013227 Espenshade, P. J., & Hughes, A. L. (2007). Regulation of Sterol Synthesis in Eukaryotes. Annual Review of Genetics , 401–427. doi.org/10.1146/annurev.genet.41.110306.130315 Farrugia, G., & Balzan, R. (2012). Oxidative Stress and Programmed Cell Death in Yeast. Frontiers in Oncology , 2 , 64. doi.org/10.3389/fonc.2012.00064 Ferreira, C., & Lucas, C. (2008). The yeast O-acyltransferase Gup1p interferes in lipid metabolism with direct consequences on the sphingolipid-sterol-ordered domains integrity/assembly. Biochimica et Biophysica Acta (BBA) - Biomembranes , 1778 (11), 2648–2653. doi.org/10.1016/j.bbamem.2008.08.011 Ferreira, J. C. C. (2017). Characterization of vacuole permeabilization and HMGB1 nuclear release in the yeast cell death induced by a benzo[a]phenoxazine derivative . University of Minho. https://repositorium.sdum.uminho.pt/handle/1822/45755 Feyder, S., De Craene, J. O., Bär, S., Bertazzi, D. L., & Friant, S. (2015). Membrane trafficking in the yeast Saccharomyces cerevisiae model. International Journal of Molecular Sciences . 54 doi.org/10.3390/ijms16011509 Finley, D. (2009). Recognition and Processing of Ubiquitin-Protein Conjugates by the Proteasome. Annual Review of Biochemistry , 477–513. doi.org/10.1146/annurev.biochem.78.081507.101607 Frade, V. H. J., Sousa, M. J., Moura, J. C. V. P., & Gonçalves, M. S. T. (2007). Synthesis, characterisation and antimicrobial activity of new benzo[ a ]phenoxazine based fluorophores. Tetrahedron Letters , 48 (47), 8347–8352. doi.org/10.1016/j.tetlet.2007.09.108 Frade, V. H. J., Sousa, M. J., Moura, J. C. V. P., & Gonçalves, M. S. T. (2008). Synthesis of naphtho[2,3a]phenoxazinium chlorides: Structure–activity relationships of these heterocycles and benzo[ a ]phenoxazinium chlorides as new antimicrobials. Bioorganic & Medicinal Chemistry , 16 (6), 3274–3282. doi.org/10.1016/j.bmc.2007.12.013 Francavilla, M., Franchi, M., Monteleone, M., & Caroppo, C. (2013). The red seaweed Gracilaria gracilis as a multi products source. Marine Drugs , 11 (10), 3754–76. doi.org/10.3390/md11103754 Gómez, E. C., Anguiano Igea, S., Gómez Amoza, J. L., & Otero Espinar, F. J. (2018). Evaluation of the promoting effect of soluble cyclodextrins in drug nail penetration. European Journal of Pharmaceutical Sciences , 270–278. doi.org/10.1016/j.ejps.2018.02.028 Hannich, J. T., Umebayashi, K., & Riezman, H. (2011). Distribution and functions of sterols and sphingolipids. Cold Spring Harbor Perspectives in Biology , a004762. doi.org/10.1101/cshperspect.a004762 Hayakawa, H., Sobue, F., Motoyama, K., Yoshimura, T., & Hemmi, H. (2017). Identification of enzymes involved in the mevalonate pathway of Flavobacterium johnsoniae . Biochemical and Biophysical Research Communications , 702–708. doi.org/10.1016/j.bbrc.2017.04.120 Hayashi, K., Hayashi, T., Miyazawa, K., & Tomoda, A. (2010). Phenoxazine Derivatives Suppress the Infections Caused by Herpes Simplex Virus Type-1 and Herpes Simplex Virus Type-2 Intravaginally Inoculated Into Mice. Journal of Pharmacological Sciences , 85–91. doi.org/10.1254/jphs.10027fp High, S., & Laird, V. (1997). Membrane protein biosynthesis - All sewn up? Trends in Cell Biology . doi.org/10.1016/S0962-8924(97)01035-0 Hoffman-Sommer, M., & Rytka, J. (2007). The yeast protein sorting pathway as an experimental model for lysosomal trafficking. Expert Review of Clinical Immunology , 3 (2), 225–239. doi.org/10.1586/1744666X.3.2.225 Hou, J., Tang, H., Liu, Z., Österlund, T., Nielsen, J., & Petranovic, D. (2014). Management of the endoplasmic reticulum stress by activation of the heat shock response in yeast. FEMS Yeast Research , 481–494. doi.org/10.1111/1567-1364.12125 61 Yuan, J., & Ching, C. B. (2014). Combinatorial engineering of mevalonate pathway for improved amorpha4,11-diene production in budding yeast. Biotechnology and Bioengineering , 608–617. doi.org/10.1002/bit.25123 Zattas, D., Berk, J. M., Kreft, S. G., & Hochstrasser, M. (2016). A Conserved C-terminal Element in the Yeast Doa10 and Human MARCH6 Ubiquitin Ligases Required for Selective Substrate Degradation. Journal of Biological Chemistry , 12105–12118. doi.org/10.1074/jbc.M116.726877 Zavrel, M., Hoot, S. J., & White, T. C. (2013). Comparison of Sterol Import under Aerobic and Anaerobic Conditions in Three Fungal Species, Candida albicans, Candida glabrata, and Saccharomyces cerevisiae . Eukaryotic Cell , 725–738. doi.org/10.1128/ec.00345-12 Zhang, H., & Hu, J. (2016). Shaping the Endoplasmic Reticulum into a Social Network. Trends in Cell Biology . doi.org/10.1016/j.tcb.2016.06.002 Zhang, Y. Q., Gamarra, S., Garcia-Effron, G., Park, S., Perlin, D. S., & Rao, R. (2010). Requirement for ergosterol in V-ATPase function underlies antifungal activity of azole drugs. PLoS Pathogens , 725– 738. doi.org/10.1371/journal.ppat.1000939 Zhang, Y. Q., & Rao, R. (2010). Beyond ergosterol: Linking ph to antifungal mechanisms. Virulence , 551– 554. doi.org/10.4161/viru.1.6.13802 Zweytick, D., Hrastnik, C., Kohlwein, S. D., & Daum, G. (2000). Biochemical characterization and subcellular localization of the sterol C-24(28) reductase, Erg4p, from the yeast Saccharomyces cerevisiae. FEBS Letters , 470 (1), 83–87. doi.org/10.1016/S0014-5793(00)01290-4 7. Supplementary data 65 Control (Untreated cells) MβCDA B Figure 5.1-1: Distribution of membrane ergosterol in untreated cells and cells exposed to and methylb -cyclodextrin. Representative images of W303-1A WT cells that were incubated with 25 µM of C9, 50 mg/mL of methyl-b -cyclodextrin, and the combination of the two compounds for 1 h. Then, they were stained with filipin during 15 minutes in the dark and finally, they were visualized using fluorescence microscopy. (A) untreated cells; (B) cells treated with M bCD. 66 C9 (25 μM) + MβCD C9 (25 μM) C D Figure 5.1-2: Distribution of membrane ergosterol in untreated cells and cells exposed to and methylb -cyclodextrin. Representative images of W303-1A WT cells that were incubated with 25 µM of C9, 50 mg/mL of methylb -cyclodextrin, and the combination of the two compounds for 1 h. Then, they were stained with filipin during 15 minutes in the dark and finally, they were visualized using fluorescence microscopy. (C) cells treated with 25; µM of C9 (D) cells treated with MbCD + 25 µM of C9