bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1 Toxicological impacts and likely protein targets of bisphenol A in Paramecium caudatum

2 Marcus V. X. Senra† & Ana Lúcia Fonseca

3 Instituto de Recursos Naturais, Universidade Federal de Itajubá, 37500-903, Itajubá, Minas Gerais – Brazil 4 †To whom correspondence should be addressed – [email protected]; Orcid - 0000-0002-3866-8837

5 Abstract 6 Bisphenol A (BPA) is a chemical agent widely used in plastic production and a well-known ubiquitous endocrine 7 disruptor, frequently associated with a series of reproductive, developmental, and transgenerational impacts over 8 wildlife, livestocks, and humans. Although widely studied, toxicological data on the effects of BPA are mostly 9 restricted to mammalian models, remaining largely underexplored for other groups of organisms such as protists, 10 which represents a considerable proportion of eukaryotic diversity. Here, we used acute end-point toxicological 11 assay to evaluate the impacts of BPA over the survival of the cosmopolitan Paramecium caudatum; and a proteome- 12 wide inverted virtual-screening (IVS) to predict the most likely P. caudatum proteins and pathways affected by 13 BPA. This xenobiotic exerts a time-dependent effect over P. caudatum survival, which may be a consequence of 14 impairments to multiple core cellular functions. We discuss the potential use of this ciliate as a biosensor for 15 environmental BPA and as a new model organism to study the general impacts of this plasticizer agent over 16 Eukaryotes. Finally, our data stress the relevance of bioinformatic methods to leverage the current knowledge on the 17 molecular impacts of environmental contaminants over a diversity of biological systems.

18 Keywords 19 Contaminants of emerging concern, xenobiotics, reverse virtual screening, Ciliophora.

20 Declarations 21 Funding 22 This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit 23 sectors. 24 Conflicts of interest/Competing interest 25 The authors declare no conflicts of interest nor competing interests. 26 Availability of data and material 27 Data and material will be available at request.

1 1 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

28 Code availability 29 Code will be available at request. 30 Authors’ contributions 31 Marcus Senra: Conceptualization, methodology, investigation, and writing-original draft preparation: Ana Fonseca: 32 Supervision, writing-reviewing, and editing.

33 Introduction 34 The widespread contamination of aquatic ecosystems with a myriad of synthetic toxic compounds represents a 35 significant threat to biodiversity (Karaouzas et al. 2018) and human safety (Wang et al. 2018), and can also intensify 36 water shortage processes, through contamination of water supplies, all around the globe (Ma et al. 2020). Among 37 these pollutants, bisphenol A (BPA), an organic synthetic compound largely employed in polycarbonate plastics and 38 epoxy resins production has been attracting considerable attention due to its environmental ubiquity, consequence of 39 its massive global demand, which is estimated in 3 million tons per year (Allard and Colaiácovo 2011), and because 40 of its endocrine disruptor activity, frequently associated with a series of reproductive and developmental 41 impairments over a variety of invertebrates and vertebrates (Flint et al. 2012), including humans (Abraham and 42 Chakraborty 2019). 43 The impacts of BPA over exposed cells are broad (Aghajanpour-Mir et al. 2016; Can et al. 2005; 44 Dumitrascu et al. 2020; Kundakovic and Champagne 2011) and are associated with impairments to a wide range of 45 cellular pathways, among which the best studied protein targets are the nuclear hormone receptors, i.e. androgen, 46 estrogen, and thyroid receptors (Usman and Ahmad 2016). In fact, according to Li and collaborators (Li et al. 2015), 47 BPA can activate human estrogen receptor α (hERα), human estrogen-related receptor γ (hERRγ), and human 48 peroxisome proliferator activated receptor γ (hPPARγ), by mimicking the structure of their natural agonist and 49 binding to their corresponding site. Other cellular influences involves induction of oxidative stress (Yoon et al. 50 2014), apoptosis (Guo et al. 2017), and transgenerational epigenetic alterations through DNA methylation and 51 histone acetylation pattern changes (Yoon et al. 2014) among many others. 52 Although extensive toxicological data are currently available, most of what is known about molecular 53 impacts of BPA is restricted to metazoans and particularly, to mammalian cells (Flint et al. 2012). Therefore, the 54 proteins targeted by BPA within unicellular microeukaryotes, organisms that are greatly diverse and crucial for the 55 energy/carbon flow along trophic chains (Weisse 2017), remain largely underexplored. 56 Protein-ligand interactions can be determined through different in vitro approaches, such as crystallography 57 (Delfosse et al. 2012), nuclear magnetic resonance (Yang et al. 2017), and competitive assays (Yang et al. 2016). 58 However, these methods are extremely time-consuming, demand significant resources and their use in high- 59 throughput analyses are greatly limited. Thus, in recent years, in silico approaches, such as molecular docking-based

2 2 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

60 simulations, which could be used in both small (Li et al. 2015) and large scales (Xu et al. 2013) have been regarded 61 as reliable alternative methods to overcome such limitations. 62 Among these docking-based approaches, the most widely used for proteome-wide target identification 63 analysis is the inverted virtual screening (IVS), sometimes referred to as reverse virtual screening (Chen and Zhi 64 2001). In brief, this method starts with a searching algorithm to generate several possible binding modes between a 65 given ligand and structures within a defined protein database, and then, an energy function is called to rank these 66 predicted poses according to their free energy (Xu et al. 2018). Finally, proteins with the strongest binding energies 67 are typically considered as the most likely targets for the desired ligand. In these terms, IVS could greatly boost our 68 current understanding on the molecular mechanism of actions of a variety of environmental contaminants over 69 model and non-model organisms with sequenced genomes. 70 Here, we used an acute end-point toxicological assay to investigate the detrimental impacts of BPA over 71 the survival of an unicellular microeukaryote, the brackishwater and cosmopolitan ciliate (Ciliophora), Paramecium 72 caudatum. Next, after predicting 3D protein structures based on the proteome of this species, we used IVS to 73 identify likely protein targets of BPA, shedding some more light into the molecular bases underlying the effects of 74 this xenobiotics over biological systems.

75 Material and Methods 76 Paramecium in vitro culture and acute toxicological assay 77 In vitro cultures of Paramecium caudatum strain JP1 (Boas et al. 2020) were maintained in petri dishes filled with 78 10 mL mineral water supplemented with macerated rice grains (Foissner et al. 2002). Stocks were kept at 23.0 °C 79 and refreshed every week. Acute end-point toxicological assays were performed in 48-well polystyrene plates. 80 Where, to each well 10 log-phase specimens were incubated for 2, 4, and 8h in the presence of 0.0001, 0,001, 0,01, 81 0,1, 1, 10, 100, or 1000 µM of BPA in mineral water at 23.0 ºC. BPA stock solutions (Sigma-Aldrich, purity of 82 ≥99%) were prepared to a concentration of 1 M in 50% DMSO (dimethyl sulfoxide) and then, serially diluted in 83 mineral water to reach desirable concentrations. For this assay, a total of 5 experiments were performed, each one 84 consisting in 5 replicates of each condition per time point. Mineral water and serial dilutions of 50% DMSO were 85 used as negative controls. Dead ciliates were counted using a stereo microscope and mean lethal concentrations 86 (LC50) were estimated for each time point using Probit (Bliss 1935; Bliss 1934a; Bliss 1934b).

87 Genome-wide spatial protein structure predictions 88 Since the genome sequence of P. caudatum strain JP1 is currently unavailable, we based our analyses on the 89 reference genome of this species, P. caudatum strain 43c3d. Accordingly, its proteome (v.2), consisting of 18,673 90 predicted proteins, were retrieved from ParameciumDB (Arnaiz et al. 2020) and protein 3D structures were 91 calculated using two homology-based modeling programs: Modeller v9.19 (Webb and Sali 2016) and Protein

3 3 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

92 Homology/analogY Recognition Engine v.2 (Phyre2) (Kelley et al. 2015). For Modeller, best templates were 93 selected using Blastp (Camacho et al. 2009) against proteins with defined 3D structures available from Protein Data 94 Bank (PDB) (Berman et al. 2000). To ensure the quality of the models only proteins with sequence identities of 95 >30% and coverage of >75% to the selected best template were included in the analysis. Then, 30 models for each 96 P. caudatum protein were generated and the one with the lowest DOPE (Discrete Optimized Protein Structure) 97 score, was selected for further analyses. Although Phyre2 also applied a knowledge-based approach to generate 98 models from the desired protein, it uses Hidden Markov Monte-Carlo (HMM) profile searches against PDB for best 99 template selection, which can be more efficient to detect and to align distant homologs (Söding 2005), theoretically 100 enhancing the quality of the final models. Again, the same sequence identity and coverage cutoff values used for 101 Modeller were applied here to limit the universe of modeled proteins to the highest confidence possible. Next, model 102 quality was accessed using MolProbity (Chen et al. 2010) and only the ones with ≥90% of amino acid residues 103 within favorable regions of the Ramachandran plot were selected. In this last filtering step, when there were 104 duplicate protein structures (modeled by both Modeller and Phyre2), the one with the highest number of amino acid 105 residues within favorable regions of the Ramachandran plot were selected to feed the IVS assay.

106 Inverse virtual screening (IVS) 107 Candidate protein targets for BPA within the P. caudatum modelled proteome were investigated using an IVS 108 approach. The rationale here was based on the assumption that strong binding energies should indicate the most 109 likely true associations. To start, all proteins were minimized in vacuum using Gromacs v5 (Abraham et al. 2015), 110 applying OPLS-AA force field (Jorgensen et al. 1996) with 50.000 steps of steepest descent to resolve clashes and 111 torsions. The search space, within each model, was defined by a grid enclosing the entire protein (blind docking) 112 using Obabel (O’Boyle et al. 2011). BPA 3D structure was directly retrieved from PubChem 113 (https://pubchem.ncbi.nlm.nih.gov/), and we used Obabel (O’Boyle et al. 2011) to desalt, add polar hydrogens, and 114 minimize (steepest descent) to resolve clashes and torsions. Then, an in-house python script was used to run 115 AutoDock Vina (Trott and Olson 2010) and to compile the results. Afterwards, proteins with the lowest (strongest) 116 binding energies to BPA, from the first percentile of frequency distribution, were selected as the most likely P. 117 caudatum targets for this compound. Structure visualizations and image manipulations were done using Chimera 118 (Pettersen et al. 2004), and 2D diagrams of polar contacts between targeted proteins and BPA were done using 119 PoseView, as implemented in Protein Plus (Fährrolfes et al. 2017).

120 Functional characterization of potential protein targets for BPA 121 Putative protein targets for BPA were characterized in terms of closest homologs, using blastp against SwissProt 122 database (Bateman 2019); conserved domains searches, using InterProScan (Jones et al. 2014), Prosite (de Castro

4 4 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

123 et al. 2006), and CD-search (Marchler-Bauer and Bryant 2004); subcellular location, using Deeploc (Almagro 124 Armenteros et al. 2017); pathways prediction, using KEGG automatic annotation (Moriya et al. 2007); and cofactor 125 and binding site predictions, using Cofactor (Zhang et al. 2017).

126 Functional enrichment analysis 127 To reconstruct the protein-protein interaction (PPI) network of the 34 putative protein targets for BPA, a functional 128 enrichment analysis was performed querying the amino acid sequences of these proteins against the STRING 129 database (Szklarczyk et al. 2019). All settings were in default mode with exception of the minimum required 130 interaction score, which was changed to the highest confidence level (0.90) and max number of interactors, set to 131 100. PPI network was analysed and manipulated using Cytoscape (Shannon et al. 2003).

132 Results 133 To evaluate the impacts of BPA over the freshwater unicellular microeukaryote Paramecium caudatum strain JP1, 134 acute endpoint toxicological assays were performed for 2, 4, and 8h and data are summarized in Table 1. As

135 expected, BPA produced a time-dependent effect on ciliate’s survival, with the lowest LC50 value (7.42 µM or 15.28 136 mg/L) observed past 8h of exposure, indicating this widespread plasticizer agent, may also represent a potential 137 environmental threat to this protist. 138 We next asked what would be the molecular bases underlying this impact to P. caudatum survival rate and 139 to address this issue, we used an inverted virtual screening (IVS) approach to identify the most likely P. caudatum 140 protein targets for this plasticizer agent. The first step for this analysis was to model 3D structures of P. caudatum 141 proteome based on data from the reference genome of this species (P. caudatum strain 43c3d), using two homology- 142 based algorithms (Modeller and Phyre2). After successive rounds of filtering steps, 852 high-quality protein models 143 were generated (4.5% of the predicted proteome size) (Table 2 and in Table S1), including members of 64 distinct 144 cellular pathways, such as metabolic functions, ribosome-related enzymes, exosome membrane trafficking systems, 145 and chromosome-associated proteins (Table S2). 146 Binding energies between these high-quality proteins and BPA were calculated using IVS and data are 147 presented in Table S1. Such values varied considerably, ranging from strong values, such as -9.1 kcal/mol estimated 148 for BPA and P00070322:Alkyl dihydroxyacetone phosphate synthase, to weak values, such as -3.8 kcal/mol 149 between BPA and P00300095:RIO kinase. To improve the success rate of our predictions - correctly discriminate 150 unlikely/spurious interactions from true associations - only proteins with the strongest binding energies to BPA, 151 within the first percentile of the frequency distribution (<-7.9 kcal/mol), were considered for further investigations 152 (Figure 1). 153 Thirty four proteins were identified within this group and were functionally annotated (Table 3 and Table 154 S3). These proteins were predicted to different subcellular compartments, where they play crucial biological roles,

5 5 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

155 such as DNA replication and cell cycle control, transcription and RNA splicing, protein transport and degradation, 156 cell signaling, and lipid production among others. Many of these enzymes were assigned to one or more (KEGG) 157 pathways (Table S3), and 10/34 can also interact with enzymes from distinct pathways, such as ribosome 158 biogenesis, porphyrin metabolism, folate biosynthesis, and mismatch and nucleotide excision repair systems (Figure 159 2), as revealed by reconstructing the protein-protein interaction network of these 34 candidate protein targets for 160 BPA, using data available from STRING database, suggesting that BPA may have a broad influence over P. 161 caudatum. 162 Among this group of proteins with strong binding energies to BPA (Table 3), 13/34 may interact with BPA 163 within functionally important sites for cofactors (i.e. GDP/GTP), coenzymes (i.e. FAD) and inhibitors (i.e. 164 staurosporine) (Table 4 and Figure S1), stressing that the predicted interactions between BPA and these enzymes 165 may really produce biological effects.

166 Discussion 167 The freshwater ciliate Paramecium caudatum was firstly described by Ehrenberg in 1838 and, since then, mostly 168 because of its worldwide distribution, simple taxonomy and in vitro maintenance, small body sizes, fast generation 169 rates, and availability of genetic manipulation tools, P. caudatum have been serving as models in a vast array of 170 disciplines, including cell biology (Sabaneyeva et al. 2009), climate change (Krenek et al. 2012), ecology (Violle et 171 al. 2010), evolution (Johri et al. 2017), genetics and genomics (McGrath et al. 2014), and also toxicology (Boas et 172 al. 2020). 173 Here, we used a Paramecium caudatum strain JP1, isolated from an eutrophic urban stream in Brazil (Boas

174 et al. 2020), to evaluate how acute exposures to bisphenol A impacts its survival. The obtained LC 50 measurements 175 (Table 1) are relatively lower than values previously reported using a different P. caudatum strain sampled from an 176 unrelated geographic region (Miyoshi et al. 2003). This should be a consequence of differences in their genetic 177 background or even within the experimental conditions. Nevertheless, both estimates are within the same order of 178 magnitude typically observed for invertebrates and vertebrates (Mathieu-Denoncourt et al. 2016), indicating this 179 ciliate species should be regarded as a good biosensor for the environmental impacts of BPA. 180 Although BPA concentration in superficial waters (0-56 휇g/L) (Corrales et al. 2015) may not induce acute 181 toxicological effects over P. caudatum nor to other aquatic organisms, it is now a consensus that long-term 182 exposures, even at these environmental concentrations, are sufficient to produce diverse reproductive and 183 developmental effects over wildlife (Flint et al. 2012) and epigenetic modifications in mice (Bansal et al. 2019; 184 Wolstenholme et al. 2013) and in zebrafish (Santangeli et al. 2019). Considering the above mentioned benefits of 185 using Paramecium as model organisms, we would like to highlight the great potential of this ciliate to improve our 186 comprehension on the chronic effects of BPA over Eukaryotes.

6 6 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

187 Broad impacts are expected in P. caudatum during exposures to BPA. According to our data BPA can 188 target 34 proteins from distinct subcellular compartments and different cellular functions, such as basic metabolism, 189 DNA replication, transcription, translation, protein transport, signaling systems, and cell cycle (as schematized in 190 Figure 3); many of which (10/34) able to interact with proteins of unrelated pathways (Figure 2), indicating that the 191 observed effects on the survival of this ciliate may result from direct and indirect interferences over multiple core 192 cellular functions. 193 Different lines of evidence stress that these predicted interactions may really occur in nature. For instance, 194 binding energies measured for all 34 candidate targets are considerably lower than the typical accepted threshold 195 value for stable protein-ligand associations (-6.0 kcal/mol) (Shityakov and Förster 2014) (Table 3); best binding 196 poses calculated for many of these enzymes (14/34) are within conserved cofactor or coenzyme binding sites (Figure 197 S1 and Table 4); and also, several likely affected pathways or biological functions (Table 3) have already been 198 reported for other biological systems. On the other hand, whether such associations will cause enzyme activation, as 199 described for small GTPase Ras proteins, after the binding of BPA to their GTP site (Schöpel et al. 2018; Schöpel et 200 al. 2013) or inactivation, such for thyroid receptors, which occurs through the displacement of T3 hormone from the 201 receptor (Moriyama et al. 2002), cannot be clearly accessed through bioinformatics and would require further 202 investigations through biochemical approaches. 203 Different mechanisms can be used by BPA to cross plasmatic membranes and reach the cytosol of exposed 204 cells. It can occur passively, after binding and inducing pore formation within plasmatic membranes (Chen et al. 205 2016), but also actively, through ATP-Binding Cassette (ABC) transporters (Mazur et al. 2012). Here, strong 206 binding energy was measured between BPA and an ABCB-11 transporter (P00070270) within the plasmatic 207 membrane of P. caudatum (Table 3). In fact, many ABCB transporters are involved in influx/efflux of a variety of 208 metabolites in plants (Hwang et al. 2016); and interestingly, the best binding pose predicted in our analysis indicate 209 that BPA may interact within amino acid residues facing the lumen side of the transmembrane channel (Figure S2), 210 suggesting a plausible route of entrance toward P. caudatum cytoplasm. 211 Once in P. caudatum cytosol, BPA may transit to golgi apparatus, lysosome, mitochondria, nucleus, and 212 endoplasmic reticulum, while potentially interfering with a wide range of cellular functions (Figure 3). Many of 213 these interferences have already been described within other model organisms, such as extensive impacts over basic 214 metabolic processes, including changes to the metabolism of galactose (Li et al. 2016), isoprenoid (Fic et al. 2015), 215 lipid (Guan et al. 2019), ketone body (Meng et al. 2019), and purine (Li et al. 2018) (Table 3). In this sense, our data 216 provide a possible molecular explanation for these observed phenotypic alterations. For instance, interferences to β- 217 oxidation of fatty-acids within peroxisomes and purine metabolism within mitochondria may be a consequence of 218 interaction of BPA within the binding sites flavin-adenine dinucleotide (FAD) within FAD-dependent acyl-CoA 219 oxidase 1 (P00100039) and with the ATP site of guanylate kinase I (P00930049), the same used by staurosporine to 220 inhibit the activity of a variety of kinases (Hirozane et al. 2019), respectively (Figure S1, and Table 4).

7 7 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

221 Besides the drastic transcriptomic (Yin et al. 2014), proteomic, and metabolomic (Huang et al. 2020) 222 profile changes induced in organisms exposed to BPA, which may also occur in P. caudatum (although not 223 evaluated here), our data suggest that general transcription and translation machineries are targets for BPA as well 224 (Table 3). Concerning transcription, we predict impacts to rDNA transcription and RNA splicing, through 225 interactions with RNA polymerase I (P00300132) (Table 3) and with dual specificity protein kinase CLK4 226 (P00040322), an enzyme that participates in the spliceosomal regulation through phosphorylating serine and 227 arginine residues of members of this complex (Rosenthal et al. 2010), which may occur within its ATP binding site, 228 the same region occupied by the alkaloid staurosporine (STU) to inhibit a variety of kinases (Zhou et al. 2006) 229 (Figure S1, and Table 4). 230 Impacts over P. caudatum translation may be even more complex, involving impairments to tRNA 231 biosynthesis (P00210034 and P00620128); ribosome export (P00450170), and protein degradation processes 232 (P00300125, P00430133, and P00220032) (Table 3). Likewise, protein chain elongation and regular turnover may 233 also be affected, since the high affinities found between BPA and diphthine methyl ester synthases (P00210141), 234 within its coenzyme (S-adenosyl-L-homocysteine) site, an enzyme that participates in the unique post-translational 235 modification of eukaryotic elongation factors (EF-2), resulting in the conversion of a histidine to diphthine residue, 236 which is essential for its normal function (Zhu et al. 2010); and with leucine aminopeptidase 2 (Waditee-Sirisattha et 237 al. 2011), within the same site used by bestatin (BES) to inhibit this enzyme activity (Stamper et al. 2004) (Figure 238 S1, and Table S4). 239 Furthermore, serious effects to transport and signaling systems are also predicted. Three key regulators of 240 intracellular membrane trafficking system (Gray et al. 2020), small GTPases Ras-related Rab proteins (P00680119, 241 P00260129, and P00700035) are likely to interact with BPA through their GTP/GDP binding sites, which would 242 result in impairments to vesicle and protein transport within this ciliate. With concerns to P. caudatum’ signaling 243 systems, a likely target for BPA is the ubiquitous calcium signaling pathway (Ca2+/CaM/CaMKII) that regulates 244 diverse calcium-dependent intracellular processes, including transcription, activation of regulatory enzymes, 245 proliferation, and development, among others (Berridge et al. 2000). One of the major components of this pathway, 246 calmodulin (CaM) (P00420197) (Table 3) is a likely target for BPA and; in fact, previous reports indicate that its 247 interaction with BPA results in allosterically changes that reduce the ability to bind Ca2 + (Murayama et al. 2015), 248 leading to male germ cell injuries in metazoans (Qian et al. 2015). In paramecium, motor functions (KINK et al. 249 1991) and trichocyst extrusion (Rauh and Nelson 1981) are controlled by calcium signaling pathway, suggesting that 250 movement and cell defence mechanisms should also be impaired during exposures to BPA. 251 Impacts to P. caudatum cell cycle control is also predicted, in part a consequence of interactions of BPA 252 with cytosolic extracellular signal-regulated kinase 1 (ERK) (P00150104 and P0004016) (Table 3, Figure S1, and 253 Table 4). These enzymes are members of the conserved ERK pathway, which regulates a diversity of cellular 254 functions, including cell entrance in mitosis and meiosis (Liu et al. 2004). Therefore, impairments to these enzymes'

8 8 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

255 normal functions would lead to G1/S cell arrestments, as previously reported in other organisms (Bilancio et al. 256 2017; Chu et al. 2018). Moreover, BPA would likely interfere within G2/M checkpoint of P. caudatum, as proposed 257 to occur in mice (Liu et al. 2013), given the strong binding energies observed with cyclin-dependent kinase 14 258 (CDK14) (P00150260), a cell cycle regulator involved in G2/M transition (Wang et al. 2016); with DNA replication 259 licensing factor mcm7 (P00370082), a participant of the MCM2-7 complex, which is essential for the initiation of 260 DNA replication and elongation throughout Eukaryotes (Liang et al. 2017); and with protein kinase dsk1 261 (P00300149), a protein involved in G2/M progression in Yeast (Takeuchi and Yanagida 1993). Still, we believe that 262 BPA may also influentiate normal cytokinesis and cause mitotic exit of P. caudatum by interacting with Dual 263 specificity protein phosphatase CDC14A (P00040164), an key enzyme responsible for the down-regulation of 264 mitotic Cdks (Gray et al. 2003) (Table 3). 265 Finally, BPA can also interact with two Phenylalanine-4-hydroxylases (EC:1.14.16.1) (P00090254 and 266 P00680028), which are oxidoreductases involved in the conversion of L-phenylalanine to L-tyrosine. The

3+ 267 configuration assumed by BPA, with one of its aromatic rings oriented toward the Fe2 center (FigureS3), resembles 268 the binding mode of L-phenylalanine within the active site of these enzymes (Andreas Andersen et al. 2002). 269 Therefore, we hypothesise that BPA could act either as an competitive inhibitor or as a substrate for these two 270 oxidoreductases, which could have a role in detoxification processes of intracellular BPA.

271 Conclusion 272 As a consequence of the enormous world-wide demand for plastic production and its recalcitrant properties, BPA 273 can be considered a ubiquitous and persistent environmental contaminant, with great potential to cause acute and 274 chronic damages to a variety of microeukaryotes, invertebrates, and vertebrates. Therefore, it should be closely 275 monitored and more effective legislation and risk management programs should be established to mitigate its 276 impacts. In this context, we showed that the brackishwater cosmopolitan ciliate Paramecium caudatum could be a 277 good biosensor for environmental BPA and also an efficient model to better evaluate both acute and chronic 278 toxicological impacts of this plasticizer agent. 279 The cellular impacts of BPA over P. caudatum are broad and may involve interferences to multiple 280 conserved core functions. Many new molecular targets for BPA were uncovered here and since they are conserved 281 throughout Eukaryotes, our predictions should be further evaluated in other unicellular and multicellular organisms. 282 Moreover, we believe that the framework applied here, using molecular docking approaches, should be used more 283 often to provide a better comprehension on how contaminants of emerging concern affect a wide variety of 284 biological systems.

285 Acknowledgements 286 The authors acknowledge the “Coordenação de Aperfeiçoamento de Pessoal de Nível Superior” (CAPES) for the 287 post-doctoral fellowship (PNPD) (88882.317976/2019-01) conferred to MS.

9 9 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

288 References 289 290 Abraham A, Chakraborty P. A review on sources and health impacts of bisphenol A [Internet]. Rev. Environ. 291 Health. De Gruyter; 2019 [cited 2021 May 25]. Available from: https://pubmed.ncbi.nlm.nih.gov/31743105/

292 Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, et al. Gromacs: High performance molecular 293 simulations through multi-level parallelism from laptops to supercomputers. SoftwareX [Internet]. Elsevier; 2015 294 Sep 1 [cited 2019 Nov 12];1–2:19–25. Available from: 295 https://www.sciencedirect.com/science/article/pii/S2352711015000059

296 Aghajanpour-Mir SM, Zabihi E, Akhavan-Niaki H, Keyhani E, Bagherizadeh I, Biglari S, et al. The genotoxic and 297 cytotoxic effects of Bisphenol-A (BPA) in MCF-7 cell line and amniocytes. Int. J. Mol. Cell. Med. [Internet]. Babol 298 University of Medical Sciences; 2016 [cited 2021 May 11];5(1):19–29. Available from: /pmc/articles/PMC4916780/

299 Allard P, Colaiácovo MP. Bisphenol A. Reprod. Dev. Toxicol. Elsevier Inc.; 2011. p. 673–86.

300 Almagro Armenteros JJ, Sønderby CK, Sønderby SK, Nielsen H, Winther O. DeepLoc: prediction of protein 301 subcellular localization using deep learning. Hancock J, editor. Bioinformatics [Internet]. Oxford Academic; 2017 302 Nov 1 [cited 2021 Jan 5];33(21):3387–95. Available from: 303 https://academic.oup.com/bioinformatics/article/33/21/3387/3931857

304 Andreas Andersen O, Flatmark T, Hough E. Crystal Structure of the Ternary Complex of the Catalytic Domain of 305 Human Phenylalanine Hydroxylase with Tetrahydrobiopterin and 3-(2-Thienyl)-l-alanine, and its Implications for 306 the Mechanism of Catalysis and Substrate Activation. J. Mol. Biol. [Internet]. Academic Press; 2002 Jul [cited 2021 307 Apr 11];320(5):1095–108. Available from: https://pubmed.ncbi.nlm.nih.gov/12126628/

308 Arnaiz O, Meyer E, Sperling L. ParameciumDB 2019: Integrating genomic data across the genus for functional and 309 evolutionary biology. Nucleic Acids Res. Oxford University Press; 2020;48(D1):D599–605.

310 Bansal A, Li C, Xin F, Duemler A, Li W, Rashid C, et al. Transgenerational effects of maternal bisphenol: a 311 exposure on offspring metabolic health. J. Dev. Orig. Health Dis. [Internet]. Cambridge University Press; 2019 Apr 312 1 [cited 2020 Aug 19];10(2):164–75. Available from: https://www.cambridge.org/core/journals/journal-of- 313 developmental-origins-of-health-and-disease/article/abs/transgenerational-effects-of-maternal-bisphenol-a-exposure- 314 on-offspring-metabolic-health/A578D5273CC8124C922C081F4FFC80B9

315 Bateman A. UniProt: A worldwide hub of protein knowledge. Nucleic Acids Res. [Internet]. Oxford University 316 Press; 2019 Jan 8 [cited 2021 Jan 5];47(D1):D506–15. Available from: https://www.ebi.ac.uk/proteins/api/doc/

317 Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, et al. The Protein Data Bank. Nucleic Acids 318 Res. [Internet]. Narnia; 2000 Jan 1 [cited 2019 Nov 12];28(1):235–42. Available from: 319 https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/28.1.235

320 Berridge MJ, Lipp P, Bootman MD. The versatility and universality of calcium signalling [Internet]. Nat. Rev. Mol. 321 Cell Biol. European Association for Cardio-Thoracic Surgery; 2000 [cited 2021 Apr 16]. p. 11–21. Available from: 322 https://www.nature.com/articles/35036035

10 10 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

323 Bilancio A, Bontempo P, Di Donato M, Conte M, Giovannelli P, Altucci L, et al. Bisphenol a induces cell cycle 324 arrest in primary and prostate cancer cells through EGFR/ERK/p53 signaling pathway activation. Oncotarget 325 [Internet]. Impact Journals LLC; 2017 [cited 2020 Dec 22];8(70):115620–31. Available from: 326 /pmc/articles/PMC5777798/?report=abstract

327 Bliss CI. THE METHOD OF PROBITS - A correction. Science [Internet]. American Association for the 328 Advancement of Science; 1934a Jan 12 [cited 2019 Sep 24];79(2035):409–10. Available from: 329 http://www.ncbi.nlm.nih.gov/pubmed/17813446

330 Bliss CI. THE METHOD OF PROBITS. Science (80-. ). [Internet]. American Association for the Advancement of 331 Science; 1934b Jan 12 [cited 2019 Aug 13];79(2037):38–9. Available from: 332 http://www.ncbi.nlm.nih.gov/pubmed/17813446

333 Bliss CI. THE CALCULATION OF THE DOSAGE-MORTALITY CURVE. Ann. Appl. Biol. [Internet]. John 334 Wiley & Sons, Ltd (10.1111); 1935 Feb 1 [cited 2019 Aug 13];22(1):134–67. Available from: 335 http://doi.wiley.com/10.1111/j.1744-7348.1935.tb07713.x

336 Boas L do AV, Senra MVX, Fernandes K, Gomes AM da A, Pedroso Dias RJ, Pinto E, et al. In vitro toxicity of 337 isolated strains and cyanobacterial bloom biomasses over Paramecium caudatum (ciliophora): Lessons from a non- 338 metazoan model organism. Ecotoxicol. Environ. Saf. Academic Press; 2020 Oct 1;202:110937.

339 Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and 340 applications. BMC Bioinformatics [Internet]. BioMed Central; 2009 Dec 15 [cited 2019 Oct 16];10(1):421. 341 Available from: http://www.biomedcentral.com/1471-2105/10/421

342 Can A, Semiz O, Cinar O. Bisphenol-A induces cell cycle delay and alters centrosome and spindle microtubular 343 organization in oocytes during meiosis. Oxford Academic; 2005 Jun 1 [cited 2020 Dec 21];11(6):389–96. Available 344 from: http://academic.oup.com/molehr/article/11/6/389/974447/BisphenolA-induces-cell-cycle-delay-and-alters

345 de Castro E, Sigrist CJA, Gattiker A, Bulliard V, Langendijk-Genevaux PS, Gasteiger E, et al. ScanProsite: 346 Detection of PROSITE signature matches and ProRule-associated functional and structural residues in proteins. 347 Nucleic Acids Res. [Internet]. Nucleic Acids Res; 2006 Jul [cited 2021 Jan 5];34(WEB. SERV. ISS.). Available 348 from: https://pubmed.ncbi.nlm.nih.gov/16845026/

349 Chen L, Chen J, Zhou G, Wang Y, Xu C, Wang X. Molecular Dynamics Simulations of the Permeation of 350 Bisphenol A and Pore Formation in a Lipid Membrane. Sci. Rep. [Internet]. Nature Publishing Group; 2016 Sep 15 351 [cited 2021 Mar 8];6(1):1–7. Available from: www.nature.com/scientificreports

352 Chen VBVBVB, Arendall WBB, Headd JJJ, Keedy DADADA, Immormino RMRM, Kapral GJGJ, et al. 353 MolProbity: All-atom structure validation for macromolecular crystallography. Acta Crystallogr. Sect. D Biol. 354 Crystallogr. [Internet]. International Union of Crystallography; 2010 Dec 21 [cited 2019 Nov 12];66(1):12–21. 355 Available from: http://scripts.iucr.org/cgi-bin/paper?S0907444909042073

356 Chen YZ, Zhi DG. Ligand - Protein inverse docking and its potential use in the computer search of protein targets of 357 a small molecule. Proteins Struct. Funct. Genet. 2001;43(2):217–26.

358 Chu P-W, Yang Z-J, Huang H-H, Chang A-A, Cheng Y-C, Wu G-J, et al. Low-dose bisphenol A activates the ERK 359 signaling pathway and attenuates steroidogenic gene expression in human placental cells†. Biol. Reprod. [Internet].

11 11 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

360 Oxford University Press; 2018 Feb 1 [cited 2021 Mar 17];98(2):250–8. Available from: 361 https://academic.oup.com/biolreprod/article/98/2/250/4689091

362 Corrales J, Kristofco LA, Baylor Steele W, Yates BS, Breed CS, Spencer Williams E, et al. Global assessment of 363 bisphenol a in the environment: Review and analysis of its occurrence and bioaccumulation. Dose-Response 364 [Internet]. University of Massachusetts; 2015 Jul 1 [cited 2020 Dec 18];13(3). Available from: 365 /pmc/articles/PMC4674187/?report=abstract

366 Delfosse V, Grimaldi M, Pons JL, Boulahtouf A, Le Maire A, Cavailles V, et al. Structural and mechanistic insights 367 into bisphenols action provide guidelines for risk assessment and discovery of bisphenol A substitutes. Proc. Natl. 368 Acad. Sci. U. S. A. [Internet]. National Academy of Sciences; 2012 Sep 11 [cited 2021 Jan 28];109(37):14930–5. 369 Available from: www.pnas.org/cgi/doi/10.1073/pnas.1203574109

370 Dumitrascu MC, Mares C, Petca RC, Sandru F, Popescu RI, Mehedintu C, et al. Carcinogenic effects of bisphenol A 371 in breast and ovarian cancers (Review) [Internet]. Oncol. Lett. Spandidos Publications; 2020 [cited 2021 May 11]. p. 372 1–1. Available from: http://www.spandidos-publications.com/10.3892/ol.2020.12145/abstract

373 Fährrolfes R, Bietz S, Flachsenberg F, Meyder A, Nittinger E, Otto T, et al. Proteins Plus: A web portal for structure 374 analysis of macromolecules. Nucleic Acids Res. 2017;45(W1):W337–43.

375 Fic A, Jurković Mlakar S, Juvan P, Mlakar V, Marc J, Sollner Dolenc M, et al. Genome-wide gene expression 376 profiling of low-dose, long-term exposure of human osteosarcoma cells to bisphenol A and its analogs bisphenols 377 AF and S. Toxicol. Vitr. [Internet]. Elsevier Ltd; 2015 Aug 1 [cited 2021 Apr 11];29(5):1060–9. Available from: 378 https://pubmed.ncbi.nlm.nih.gov/25912373/

379 Flint S, Markle T, Thompson S, Wallace E. Bisphenol A exposure, effects, and policy: A wildlife perspective 380 [Internet]. J. Environ. Manage. J Environ Manage; 2012 [cited 2020 Dec 18]. p. 19–34. Available from: 381 https://pubmed.ncbi.nlm.nih.gov/22481365/

382 Foissner W, Agatha S, Berger H. Soil Ciliates (Protozoa , Ciliophora) from Namibia (Southwest Africa), with 383 Emphasis on Two Contrasting Environments , the Etosha Region and the Namib Desert . Part I . Text and Line 384 Drawings . Part II . Photographs . Text. 2002;II(June):81–2.

385 Gray CH, Good VM, Tonks NK, Barford D. The structure of the cell cycle protein Cdc14 reveals a proline-directed 386 protein phosphatase. EMBO J. [Internet]. EMBO J; 2003 Jul 15 [cited 2021 Apr 9];22(14):3524–35. Available from: 387 https://pubmed.ncbi.nlm.nih.gov/12853468/

388 Gray JL, von Delft F, Brennan PE. Targeting the Small GTPase Superfamily through Their Regulatory Proteins 389 [Internet]. Angew. Chemie - Int. Ed. Wiley-VCH Verlag; 2020 [cited 2021 May 16]. p. 6342–66. Available from: 390 https://doi.org/10.1002/anie.201900585.

391 Guan Y, Zhang T, He J, Jia J, Zhu L, Wang Z. Bisphenol A disturbed the lipid metabolism mediated by sterol 392 regulatory element binding protein 1 in rare minnow Gobiocypris rarus. Aquat. Toxicol. [Internet]. Elsevier B.V.; 393 2019 Feb 1 [cited 2021 Mar 30];207:179–86. Available from: https://pubmed.ncbi.nlm.nih.gov/30579156/

394 Guo J, Zhao MH, Shin KT, Niu YJ, Ahn YD, Kim NH, et al. The possible molecular mechanisms of bisphenol A 395 action on porcine early embryonic development. Sci. Rep. [Internet]. Nature Publishing Group; 2017 Dec 1 [cited 396 2020 Aug 14];7(1):1–9. Available from: https://pubmed.ncbi.nlm.nih.gov/28819136/

12 12 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

397 Hirozane Y, Toyofuku M, Yogo T, Tanaka Y, Sameshima T, Miyahisa I, et al. Structure-based rational design of 398 staurosporine-based fluorescent probe with broad-ranging kinase affinity for kinase panel application. Bioorganic 399 Med. Chem. Lett. Elsevier Ltd; 2019 Nov 1;29(21):126641.

400 Huang W, Zhu L, Zhao C, Chen X, Cai Z. Integration of proteomics and metabolomics reveals promotion of 401 proliferation by exposure of bisphenol S in human breast epithelial MCF-10A cells. Sci. Total Environ. [Internet]. 402 Elsevier B.V.; 2020 Apr 10 [cited 2021 May 12];712. Available from: https://pubmed.ncbi.nlm.nih.gov/31945527/

403 Hwang JU, Song WY, Hong D, Ko D, Yamaoka Y, Jang S, et al. Plant ABC Transporters Enable Many Unique 404 Aspects of a Terrestrial Plant’s Lifestyle [Internet]. Mol. Plant. Cell Press; 2016 [cited 2021 May 25]. p. 338–55. 405 Available from: https://pubmed.ncbi.nlm.nih.gov/26902186/

406 Johri P, Krenek S, Marinov GK, Doak TG, Berendonk TU, Lynch M. Population Genomics of Paramecium Species. 407 Mol. Biol. Evol. [Internet]. Oxford University Press; 2017 May 1 [cited 2020 Dec 18];34(5):1194–216. Available 408 from: https://academic.oup.com/mbe/article/34/5/1194/2992914

409 Jones P, Binns D, Chang HYH-Y, Fraser M, Li W, McAnulla C, et al. InterProScan 5: Genome-scale protein 410 function classification. Bioinformatics [Internet]. Narnia; 2014 May 1 [cited 2019 Oct 16];30(9):1236–40. Available 411 from: https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btu031

412 Jorgensen WL, Maxwell DS, Tirado-Rives J. Development and testing of the OPLS all-atom force field on 413 conformational energetics and properties of organic liquids. J. Am. Chem. Soc. [Internet]. American Chemical 414 Society; 1996 [cited 2019 Nov 12];118(45):11225–36. Available from: 415 https://pubs.acs.org/doi/abs/10.1021/ja9621760

416 Karaouzas I, Theodoropoulos C, Vardakas L, Kalogianni E, Th. Skoulikidis N. A review of the effects of pollution 417 and water scarcity on the stream biota of an intermittent Mediterranean basin. River Res. Appl. [Internet]. John 418 Wiley and Sons Ltd; 2018 May 1 [cited 2020 Dec 3];34(4):291–9. Available from: 419 http://doi.wiley.com/10.1002/rra.3254

420 Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJE. The Phyre2 web portal for protein modeling, 421 prediction and analysis. Nat. Protoc. [Internet]. 2015;10(6):845–58. Available from: 422 https://doi.org/10.1038/nprot.2015.053

423 KINK JA, MALEY ME, LING K ‐ Y, KANABROCKI JA, KUNG C. Efficient Expression of the Paramecium 424 Calmodulin Gene in Escherichia coli after Four TAA‐‐ to CAA Changes through a Series of Polymerase Chain 425 Reactions. J. Protozool. [Internet]. J Protozool; 1991 [cited 2021 Apr 16];38(5):441–7. Available from: 426 https://pubmed.ncbi.nlm.nih.gov/1920142/

427 Krenek S, Petzoldt T, Berendonk TU. Coping with Temperature at the Warm Edge – Patterns of Thermal 428 Adaptation in the Microbial Eukaryote Paramecium caudatum. Petchey O, editor. PLoS One [Internet]. 2012 Mar 429 9;7(3):e30598. Available from: https://dx.plos.org/10.1371/journal.pone.0030598

430 Kundakovic M, Champagne FA. Epigenetic perspective on the developmental effects of bisphenol A [Internet]. 431 Brain. Behav. Immun. NIH Public Access; 2011 [cited 2021 May 11]. p. 1084–93. Available from: 432 /pmc/articles/PMC3703316/

13 13 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

433 Li C, Lu Q, Ye J, Qin H, Long Y, Wang L, et al. Metabolic and proteomic mechanism of bisphenol A degradation 434 by Bacillus thuringiensis. Sci. Total Environ. Elsevier B.V.; 2018 Nov 1;640–641:714–25.

435 Li L, Wang Q, Zhang Y, Niu Y, Yao X, Liu H. The Molecular Mechanism of Bisphenol A (BPA) as an Endocrine 436 Disruptor by Interacting with Nuclear Receptors: Insights from Molecular Dynamics (MD) Simulations. Shioda T, 437 editor. PLoS One [Internet]. Public Library of Science; 2015 Mar 23 [cited 2020 Dec 26];10(3):e0120330. Available 438 from: https://dx.plos.org/10.1371/journal.pone.0120330

439 Li S, Jin Y, Wang J, Tang Z, Xu S, Wang T, et al. Urinary profiling of cis-diol-containing metabolites in rats with 440 bisphenol A exposure by liquid chromatography-mass spectrometry and isotope labeling. Analyst [Internet]. Royal 441 Society of Chemistry; 2016 Feb 7 [cited 2021 Apr 11];141(3):1144–53. Available from: 442 https://pubs.rsc.org/en/content/articlehtml/2016/an/c5an02195b

443 Liang Z, Li W, Liu J, Li J, He F, Jiang Y, et al. Simvastatin suppresses the DNA replication licensing factor MCM7 444 and inhibits the growth of tamoxifen-resistant breast cancer cells. Sci. Rep. [Internet]. Nature Publishing Group; 445 2017 Feb 2 [cited 2020 Dec 21];7(1):1–11. Available from: https://www.nature.com/articles/srep41776

446 Liu R, Xing L, Kong D, Jiang J, Shang L, Hao W. Bisphenol A inhibits proliferation and induces apoptosis in 447 micromass cultures of rat embryonic midbrain cells through the JNK, CREB and p53 signaling pathways. Food 448 Chem. Toxicol. Pergamon; 2013 Feb 1;52:76–82.

449 Liu X, Yan S, Zhou T, Terada Y, Erikson RL. The MAP kinase pathway is required for entry into mitosis and cell 450 survival. Oncogene [Internet]. Nature Publishing Group; 2004 Jan 22 [cited 2021 Apr 23];23(3):763–76. Available 451 from: www.nature.com/onc

452 Ma T, Sun S, Fu G, Hall JW, Ni Y, He L, et al. Pollution exacerbates China’s water scarcity and its regional 453 inequality. Nat. Commun. [Internet]. Springer US; 2020;11(1):1–9. Available from: 454 http://dx.doi.org/10.1038/s41467-020-14532-5

455 Marchler-Bauer A, Bryant SH. CD-Search: Protein domain annotations on the fly. Nucleic Acids Res. [Internet]. 456 Nucleic Acids Res; 2004 Jul 1 [cited 2021 Jan 5];32(WEB SERVER ISS.). Available from: 457 https://pubmed.ncbi.nlm.nih.gov/15215404/

458 Mathieu-Denoncourt J, Wallace SJ, de Solla SR, Langlois VS. Influence of Lipophilicity on the Toxicity of 459 Bisphenol A and Phthalates to Aquatic Organisms [Internet]. Bull. Environ. Contam. Toxicol. Springer New York 460 LLC; 2016 [cited 2020 Nov 5]. p. 4–10. Available from: https://pubmed.ncbi.nlm.nih.gov/27169527/

461 Mazur CS, Marchitti SA, Dimova M, Kenneke JF, Lumen A, Fisher J. Human and rat ABC transporter efflux of 462 bisphenol A and bisphenol A glucuronide: Interspecies comparison and implications for pharmacokinetic 463 assessment. Toxicol. Sci. [Internet]. Toxicol Sci; 2012 Aug [cited 2020 Dec 21];128(2):317–25. Available from: 464 https://pubmed.ncbi.nlm.nih.gov/22552776/

465 McGrath CL, Gout JF, Doak TG, Yanagi A, Lynch M. Insights into three whole-genome duplications gleaned from 466 the Paramecium caudatum genome sequence. Genetics [Internet]. Genetics; 2014 [cited 2020 Dec 18];197(4):1417– 467 28. Available from: /pmc/articles/PMC4125410/

468 Meng Z, Zhu W, Wang D, Li R, Jia M, Yan S, et al. 1 H NMR-based serum metabolomics analysis of the age- 469 related metabolic effects of perinatal exposure to BPA, BPS, BPF, and BPAF in female mice offspring. Environ. Sci.

14 14 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

470 Pollut. Res. [Internet]. Springer Verlag; 2019 Feb 28 [cited 2021 Apr 11];26(6):5804–13. Available from: 471 https://link.springer.com/article/10.1007/s11356-018-4004-9

472 Miyoshi N, Kawano T, Tanaka M, Kadono T, Kosaka T, Kunimoto M, et al. Use of Paramecium species in 473 bioassays for environmental risk management: Determination of IC50 values for water pollutants. J. Heal. Sci. 474 [Internet]. 2003 [cited 2019 Sep 20];49(6):429–35. Available from: http://joi.jlc.jst.go.jp/JST.JSTAGE/jhs/49.429? 475 from=CrossRef

476 Moriya Y, Itoh M, Okuda S, Yoshizawa AC, Kanehisa M. KAAS: An automatic genome annotation and pathway 477 reconstruction server. Nucleic Acids Res. [Internet]. Nucleic Acids Res; 2007 Jul [cited 2020 Dec 29];35(SUPPL.2). 478 Available from: https://pubmed.ncbi.nlm.nih.gov/17526522/

479 Moriyama K, Tagami T, Akamizu T, Usui T, Saijo M, Kanamoto N, et al. Thyroid hormone action is disrupted by 480 bisphenol A as an antagonist. J. Clin. Endocrinol. Metab. [Internet]. J Clin Endocrinol Metab; 2002 Nov 1 [cited 481 2021 May 12];87(11):5185–90. Available from: https://pubmed.ncbi.nlm.nih.gov/12414890/

482 Murayama K, Sonoyama M, Matsuda S. Strong Interaction of Bovine Brain Calmodulin with Bisphenol A: Effects 483 on Secondary Structure, Conformation, Ca 2+ -Binding Affinity, Gibbs Energy, and Domain Cooperativity. Bull. 484 Chem. Soc. Jpn. [Internet]. Chemical Society of Japan; 2015 Jun 15 [cited 2020 Dec 22];88(6):880–7. Available 485 from: http://www.journal.csj.jp/doi/10.1246/bcsj.20150045

486 O’Boyle NM, Banck M, James CA, Morley C, Vandermeersch T, Hutchison GR. Open Babel: An open chemical 487 toolbox. J. Cheminform. [Internet]. BioMed Central; 2011 Dec 7 [cited 2019 Nov 13];3(1):33. Available from: 488 https://jcheminf.biomedcentral.com/articles/10.1186/1758-2946-3-33

489 Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, et al. UCSF Chimera. A visualization 490 system for exploratory research and analysis. J. Comput. Chem. [Internet]. John Wiley & Sons, Ltd; 2004 Oct 1 491 [cited 2019 Nov 12];25(13):1605–12. Available from: http://doi.wiley.com/10.1002/jcc.20084

492 Qian W, Wang Y, Zhu J, Mao C, Wang Q, Huan F, et al. The toxic effects of Bisphenol A on the mouse 493 spermatocyte GC-2 cell line: The role of the Ca2+-calmodulin-Ca2+/calmodulin-dependent protein kinase II axis. J. 494 Appl. Toxicol. [Internet]. John Wiley and Sons Ltd; 2015 Nov 1 [cited 2020 Dec 22];35(11):1271–7. Available 495 from: https://pubmed.ncbi.nlm.nih.gov/26096086/

496 Rauh JJ, Nelson DL. Calmodulin is a major component of extruded trichocysts from Paramecium tetraurelia. J. Cell 497 Biol. [Internet]. J Cell Biol; 1981 [cited 2021 Apr 16];91(3 I):860–5. Available from: 498 https://pubmed.ncbi.nlm.nih.gov/6276412/

499 Rosenthal AS, Tanega C, Shen M, Mott BT, Bougie JM, Nguyen D-T, et al. An inhibitor of the Cdc2-like kinase 4 500 (Clk4) [Internet]. Probe Reports from NIH Mol. Libr. Progr. National Center for Biotechnology Information (US); 501 2010 [cited 2021 Mar 30]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21735605

502 Sabaneyeva E V., Derkacheva ME, Benken KA, Fokin SI, Vainio S, Skovorodkin IN. Actin-Based Mechanism of 503 Holospora obtusa Trafficking in Paramecium caudatum. Protist [Internet]. 2009 May 23 [cited 2020 Dec 504 18];160(2):205–19. Available from: https://linkinghub.elsevier.com/retrieve/pii/S1434461009000030

15 15 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

505 Santangeli S, Consales C, Pacchierotti F, Habibi HR, Carnevali O. Transgenerational effects of BPA on female 506 reproduction. Sci. Total Environ. [Internet]. Elsevier B.V.; 2019 Oct 1 [cited 2020 Aug 19];685:1294–305. 507 Available from: https://doi.org/10.1016/j.scitotenv.2019.06.029

508 Schöpel M, Jockers KFG, Düppe PM, Autzen J, Potheraveedu VN, Ince S, et al. Bisphenol a binds to Ras proteins 509 and competes with guanine nucleotide exchange: Implications for GTPase-selective antagonists. J. Med. Chem. 510 [Internet]. American Chemical Society; 2013 Dec 12 [cited 2020 Oct 23];56(23):9664–72. Available from: 511 https://pubs.acs.org/doi/full/10.1021/jm401291q

512 Schöpel M, Shkura O, Seidel J, Kock K, Zhong X, Löffek S, et al. Allosteric activation of GDP-bound ras isoforms 513 by bisphenol derivative plasticisers. Int. J. Mol. Sci. [Internet]. MDPI AG; 2018 Apr 10 [cited 2020 Oct 23];19(4). 514 Available from: /pmc/articles/PMC5979466/?report=abstract

515 Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: A software Environment for 516 integrated models of biomolecular interaction networks. Genome Res. [Internet]. 2003 Nov 1 [cited 2021 Jan 517 5];13(11):2498–504. Available from: https://pubmed.ncbi.nlm.nih.gov/14597658/

518 Shityakov S, Förster C. In silico predictive model to determine vector-mediated transport properties for the blood- 519 brain barrier choline transporter. Adv. Appl. Bioinforma. Chem. [Internet]. Dove Medical Press Ltd; 2014 [cited 520 2021 Mar 16];7(1):23–36. Available from: /pmc/articles/PMC4159400/

521 Söding J. Protein homology detection by HMM-HMM comparison. Bioinformatics. 2005;21(7):951–60.

522 Stamper CC, Bienvenue DL, Bennett B, Ringe D, Petsko GA, Holz RC. Spectroscopic and X-ray crystallographic 523 characterization of bestatin bound to the aminopeptidase from Aeromonas (Vibrio) proteolytica. Biochemistry 524 [Internet]. Biochemistry; 2004 Aug 3 [cited 2021 May 25];43(30):9620–8. Available from: 525 https://pubmed.ncbi.nlm.nih.gov/15274616/

526 Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. STRING v11: Protein-protein 527 association networks with increased coverage, supporting functional discovery in genome-wide experimental 528 datasets. Nucleic Acids Res. [Internet]. Oxford University Press; 2019 Jan 8 [cited 2021 Jan 5];47(D1):D607–13. 529 Available from: https://pubmed.ncbi.nlm.nih.gov/30476243/

530 Takeuchi M, Yanagida M. A mitotic role for a novel fission yeast protein kinase dsk1 with cell cycle stage 531 dependent phosphorylation and localization. Mol. Biol. Cell [Internet]. American Society for Cell Biology; 1993 532 [cited 2020 Dec 22];4(3):247–60. Available from: /pmc/articles/PMC300923/?report=abstract

533 Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, 534 efficient optimization, and multithreading. J. Comput. Chem. [Internet]. NIH Public Access; 2010 Jan 30 [cited 2019 535 Nov 13];31(2):455–61. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19499576

536 Usman A, Ahmad M. From BPA to its analogues: Is it a safe journey? Chemosphere. Elsevier Ltd; 2016. p. 131–42.

537 Violle C, Pu Z, Jiang L. Experimental demonstration of the importance of competition under disturbance. Proc. Natl. 538 Acad. Sci. U. S. A. [Internet]. National Academy of Sciences; 2010 Jul 20 [cited 2020 Dec 18];107(29):12925–9. 539 Available from: www.pnas.org/cgi/doi/10.1073/pnas.1000699107

16 16 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

540 Waditee-Sirisattha R, Hattori A, Shibato J, Rakwal R, Sirisattha S, Takabe T, et al. Role of the arabidopsis leucine 541 aminopeptidase 2. Plant Signal. Behav. [Internet]. Taylor & Francis; 2011 Oct [cited 2021 Mar 30];6(10):1581–3. 542 Available from: /pmc/articles/PMC3256389/

543 Wang L, Zhang L, Lv J, Zhang Y, Ye B. Public Awareness of Drinking Water Safety and Contamination Accidents: 544 A Case Study in Hainan Province, China. Water [Internet]. MDPI AG; 2018 Apr 9 [cited 2020 Dec 26];10(4):446. 545 Available from: http://www.mdpi.com/2073-4441/10/4/446

546 Wang X, Jia Y, Fei C, Song X, Li L. Activation/proliferation-associated protein 2 (Caprin-2) positively regulates 547 CDK14/cyclin Y-mediated lipoprotein. J. Biol. Chem. [Internet]. American Society for Biochemistry and Molecular 548 Biology Inc.; 2016 Dec 16 [cited 2020 Dec 21];291(51):26427–34. Available from: /pmc/articles/PMC5159503/? 549 report=abstract

550 Webb B, Sali A. Comparative protein structure modeling using MODELLER. Curr. Protoc. Bioinforma. [Internet]. 551 Hoboken, NJ, USA: John Wiley & Sons, Inc.; 2016 Jun 20 [cited 2019 Nov 12];2016(June):5.6.1-5.6.37. Available 552 from: http://doi.wiley.com/10.1002/cpbi.3

553 Weisse T. Functional diversity of aquatic ciliates. Eur. J. Protistol. Elsevier GmbH; 2017. p. 331–58.

554 Wolstenholme JT, Goldsby JA, Rissman EF. Transgenerational effects of prenatal bisphenol A on social 555 recognition. Horm. Behav. [Internet]. 2013 Nov 1 [cited 2020 Aug 19];64(5):833–9. Available from: 556 /pmc/articles/PMC3955720/?report=abstract

557 Xu X, Huang M, Zou X. Docking-based inverse virtual screening: methods, applications, and challenges. Biophys. 558 Reports [Internet]. Springer Berlin Heidelberg; 2018 Feb 1 [cited 2021 Jan 29];4(1):1–16. Available from: 559 https://doi.org/10.1007/s41048-017-0045-8

560 Xu XJ, Su JG, Liu B, Li CH, Tan JJ, Zhang XY, et al. Reverse virtual screening on persistent organic pollutants 561 4,4′-DDE and CB-153. Wuli Huaxue Xuebao/ Acta Phys. - Chim. Sin. 2013 Sep 26;29(10):2276–85.

562 Yang F, Xu L, Zhu L, Zhang Y, Meng W, Liu R. Competitive immunoassay for analysis of bisphenol A in 563 children’s sera using a specific antibody. Environ. Sci. Pollut. Res. [Internet]. Springer Verlag; 2016 Jun 1 [cited 564 2021 May 25];23(11):10714–21. Available from: https://pubmed.ncbi.nlm.nih.gov/26888526/

565 Yang H, Huang Y, Liu J, Tang P, Sun Q, Xiong X, et al. Binding modes of environmental endocrine disruptors to 566 human serum albumin: Insights from STD-NMR, ITC, spectroscopic and molecular docking studies. Sci. Rep. 567 [Internet]. Nature Publishing Group; 2017 Dec 1 [cited 2021 Jan 28];7(1):1–11. Available from: 568 www.nature.com/scientificreports/

569 Yin R, Gu L, Li M, Jiang C, Cao T, Zhang X. Gene expression profiling analysis of bisphenol A-induced 570 perturbation in biological processes in ER-negative HEK293 cells. PLoS One [Internet]. Public Library of Science; 571 2014 Jun 5 [cited 2021 Apr 11];9(6):98635. Available from: /pmc/articles/PMC4047077/

572 Yoon K, Kwack SJ, Kim HS, Lee BM. Estrogenic endocrine-disrupting chemicals: Molecular mechanisms of 573 actions on putative human diseases. J. Toxicol. Environ. Heal. - Part B Crit. Rev. [Internet]. Taylor and Francis Inc.; 574 2014 Apr 3 [cited 2020 Dec 26];17(3):127–74. Available from: https://pubmed.ncbi.nlm.nih.gov/24749480/

17 17 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

575 Zhang C, Freddolino PL, Zhang Y. COFACTOR: Improved protein function prediction by combining structure, 576 sequence and protein-protein interaction information. Nucleic Acids Res. [Internet]. Oxford University Press; 2017 577 Jul 3 [cited 2021 Feb 21];45(W1):W291–9. Available from: https://pubmed.ncbi.nlm.nih.gov/28472402/

578 Zhou TJ, Sun LG, Gao Y, Goldsmith EJ. Crystal structure of the MAP3K TAO2 kinase domain bound by an 579 inhibitor staurosporine. Acta Biochim. Biophys. Sin. (Shanghai). [Internet]. Acta Biochim Biophys Sin (Shanghai); 580 2006 Jun [cited 2021 May 18];38(6):385–92. Available from: https://pubmed.ncbi.nlm.nih.gov/16761096/

581 Zhu X, Kim J, Su X, Lin H. Reconstitution of diphthine synthase activity in vitro. Biochemistry [Internet]. NIH 582 Public Access; 2010 Nov 9 [cited 2021 Mar 30];49(44):9649–57. Available from: /pmc/articles/PMC3049195/ 583

584 Figure captions

585 Fig 1 Frequency distribution of binding energies between BPA and 852 high-quality proteins modelled from the 586 predicted proteome of P. caudatum. Third-four proteins with the strongest binding energies, indicated in the first 587 percentile of the distributions were selected as likely targets for BPA

588 Fig 2 Protein-protein interaction (PPI) network of likely protein targets for BPA in P. caudatum based on data 589 available from STRING database. Only candidate protein targets that interact with proteins from other pathways are 590 presented. Circles in red represent the candidate protein targets, while in green are proteins retrieved during 591 functional enrichment analysis. When known, the pathways to which each protein belongs are presented as color 592 patterns displayed around the respective circle. P00040164 interacts with a kinase domain-containing protein 593 (UNIPROT:A2DMG0) from T. vaginalis not allocated to any pathway

594 Fig 3 Schematic eukaryotic cell illustrating the putative protein targets for BPA in P. caudatum and their subcellular 595 compartment and biological function. Protein codes are relative to Table 3. Proteins that may interact with BPA 596 within the same site used by natural cofactors/coenzymes are marked (*)

597 Supplementary Fig S1 Three-dimensional structures of 14 candidate protein targets for BPA in P. caudatum, in 598 which the predicted besting binding pose is located within regions used by the natural cofactor/coenzyme of the 599 enzyme according to analysis using COFACTOR. Each candidate protein target is represented twice: In the top part 600 is presented the best binding pose for BPA (in purple); and in the bottom part the binding site used by the 601 cofactor/coenzyme (in red)

18 18 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

602 Supplementary Fig S2 Three-dimensional structure of candidate protein targets P00070270, an ABC transporter B 603 family member 11 of P. caudatum, which may be involved in the translocation of BPA from extracellular region to 604 cytosol. The predicted best binding pose for BPA is represented in purple and occurs within the lumen-side of the 605 transmembrane channel

606 Supplementary Fig S3 Three-dimensional structures of 2 candidate protein targets for BPA in P. caudatum, which 607 could be involved in detoxification processes, through oxidation, to protect cells from impacts caused by BPA. (A) 608 The cytoplasmic P00090254 and (B) the mitochondrial P00680028 Phenylalanine-4-hydroxylases. Iron ions are 609 represented as a golden sphere and BPA are in purple. The small boxes depict hydrophobic (green spindles) and 610 polar (dotted lines) interactions between BPA and amino acid residues from the enzymes

19 19 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. TablebioRxiv 1. preprint Mean lethaldoi: https://doi.org/10.1101/2021.06.24.449746 concentration (LC ) of BPA over P.; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder,50 who has granted bioRxiv a license to display the preprint in perpetuity. It is made caudatum along exposures of 2, 4, andavailable 8h. under aCC-BY-NC-ND 4.0 International license.

Exposure LC50 (µM) Standard deviation (±) 2h 43,641 2,839 4h 16,727 1,586 8h 7,415 2,048 bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (whichTable was 2. Summarynot certified of by the peer P. review) caudatum is the proteome author/funder, modelling who has process. granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-NDTotal models 4.0† InternationalHigh-quality license. models‡ Total high- Proteome Reference genome Modeller PHYRE2 Modeller PHYRE2 quality size models§ Paramecium caudatum 43c3d 18,673 1,347 1,865 784 838 852 † Considering cutoff values of 75% for coverage and 30% for identity in template selection step. ‡ >90% of residues within favored positions in Ramachadran plot. § Considering high-quality protein models generated from both Modeller and PHYRE2. bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (whichTable was3. Functional not certified annotation, by peer binding review) modes, is the and author/funder, binding energies who of has 34 grantedputative proteinbioRxiv targets a license of BPA. to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 InternationalBinding license. Protein ID† Gene product Biological process energies Hydrogen bonds to BPA (mol/kcal) ABC transporter B family Transmembrane transport P00070270 -8,3 ARG 913/2HH2; ASN 71/OD1 member 11 (GO:0055085) GLU 196/OE2; GLU 196/N; P00300149 Protein kinase dsk1 Cell cycle (GO:0007049) -8,7 ASN 134/1HD2; GLU 134/HN P00090254 Phenylalanine-4-hydroxylase Redoxi process (GO:0055114) -8,6 TYR 365/OH Regulation of G2/M transition P00030265 Casein kinase I isoform epsilon -8,4 ASP 120/OD1 (GO:0010389) Extracellular signal-regulated Activation of MAPK activity P00150104 -8,4 LYS 79/NZ kinase 1 (GO:0000187) Probable glutamine--tRNA Glutaminyl-tRNA P00210034 -8,4 ARG 441/2HH1; GLU 436/OE2; LYS 209/O ligase aminoacylation (GO:0006425) Diphthine methyl ester Diphthamide biosynthesis P00210141 -8,3 CYS 91/O synthase (GO:0017183) Galactose metabolism P00710081 UDP-glucose 4-epimerase -8,3 ARG 233/2HH2; ARG 233/2HH1 (GO:0006012) dTDP biosynthesis P00100308 Thymidylate kinase -8,2 ASN 79/ND2 (GO:0006233) Calcium-mediated signaling P00420197 Calmodulin -8,2 LYS 110/NZ (GO:0019722) Extracellular signal-regulated Protein phosphorylation P00470113 -8,2 SER 192/HG; ASN 201/O; VAL 202/O kinase 1 (GO:0006468) Dual specificity protein P00040164 Cell division (GO:0051301) -8,1 GLY 182/HN; ASP 164/O phosphatase CDC14A Geranylgeranyl pyrophosphate Isoprenoid biosynthesis P00870013 -8,1 ASP 46/OD1 synthase penG (GO:0008299) ABC transporter E family Positive regulation of translation P00450170 -8 GLN 171/O; SER 245/HG member 2 (GO:0045727) 26S proteasome non-ATPase Proteasome assembly P00300125 -8,4 GLU 226/O; LYS 275/HZ2 regulatory subunit 1 homolog B (GO:0043248) Calcium ion transport P00340021 Calcium-transporting ATPase 2 -8,2 PRO 398/O (GO:0006816) Protein catabolism P00220032 Leucine aminopeptidase 2 -8 PRO 324/O; LYS 613/HN (GO:0030163) Calcium ion transport P00530165 Calcium-transporting ATPase -8 THR 236/HG1 (GO:1903515) P00430133 Cathepsin D Proteolysis (GO:0006508) -8,1 PHE 106/O P00680119 Ras-related protein Rab-8B Protein export (GO:0009306) -8,5 SER 157/HG; LYS 159/HN; ALA 158/HN Protein transport P00260129 Ras-related protein Rab6 -8 SER 155/HG; LYS 157/HN; ALA 156/HN (GO:0015031) Protein transport P00700035 Ras-related protein ORAB-1 -8 LYS 155/HN; ALA 154/HN (GO:0015031) P00680028 Phenylalanine-4-hydroxylase Redoxi process (GO:0055114) -8,9 GLU 273/OE1 Peptidyl-prolyl cis-trans P00530125 Protein folding (GO:0006457) -8,7 ASP 40/OD2; LYS 111/HN isomerase CYP19-2 Phenylalanyl-tRNA P00620128 Phenylalanine--tRNA ligase -8,2 PHE 197/O aminoacylation (GO:0006432) Probable succinyl-CoA:3- Cellular ketone body P00800061 ketoacid coenzyme A -8,2 ASN 303/2HD2 metabolism (GO:0046950) transferase Purine nucleotide metabolism P00930049 Guanylate kinase 1 -8 GLU 110/OE2 (GO:0006163) DNA replication licensing DNA replication initiation P00370082 -8,2 SER 275/HN; ASP 274/HN; LEU 142/O factor mcm7 (GO:0006270) Serine/threonine-protein Protein autophosphorylation P00950041 -8,6 LEU 127/HN kinase AFC3 (GO:0046777) Dual specificity protein Regulation of RNA splicing P00040322 -8,3 LEU 119/HN; SER 42/HG kinase CLK4 (GO:0043484) Regulation of mitosis THR 191/O; VAL 190/O; GLU 221/OE2; P00150260 Cyclin-dependent kinase 10 -8,1 (GO:0007346) LYS 155/HZ2 DNA-directed RNA polymerase P00300132 Transcription (GO:0006351) -8 TYR 369/HH I subunit rpa1 Peroxisomal acyl-coenzyme A Fatty acid metabolism P00100039 -8,2 TYR 128/O; GLN 130/HN; THR 131/HN oxidase 1 (GO:0006631) Lipid biosynthesis ARG358/HH2; ARG 505/1HH2; P00070322 Alkyl-DHAP synthase -9,1 (GO:0008610) ARG 505/HE Proteins in bold represent the ones in which BPA may occupy functionaly important sites (see Table 4). †According to original proteome annotation. bioRxiv preprint doi: https://doi.org/10.1101/2021.06.24.449746; this version posted June 24, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Table4. P. caudatum proteins that may be targeted by BPA within functionaly important sites, as predicted in analyses using Cofactor. COFACTOR Protein Predicted ligand (Bio-Lip) Binding site residues (C-score) P00930049 Guanosine-5'-monophosphate (5GP) 0,34 33, 37, 40, 49, 68, 72, 77, 78, 79, 98, 99, 102 P00950041 Adenosine diphosphate (ADP) 0,59 44, 45, 46, 47, 48, 49, 51, 63, 65, 95, 112, 114, 115, 117, 162, 163, 165, 178 40, 41, 42, 48, 61, 63, 99, 115, 116, 117, 118, 119, 120, 121, 166, 167, 169, P00040322 Staurosporine (STU) 0,99 188, 189 P00090254 Iron (fe) 0,61 251, 256, 296 26, 128, 129, 134, 135, 166, 168, 219, 227, 406, 409, 410, 411, 413, 415, 416, P00100039 Flavin-adenine dinucleotide (FAD) 0,7 419 20, 21, 22, 23, 28, 41, 43, 61, 74, 95, 96, 97, 98, 99, 100, 101, 144, 147, 157, P00150104 Staurosporine (STU) 0,99 158 P00220032 Bestatin (BES) 0,95 134, 262, 263, 264, 265, 266, 287, 290, 291, 294, 312, 313, 373, 378 10, 11, 12, 13, 14, 15, 16, 26, 27, 28, 30, 32, 33, 59, 114, 115, 117, 118, 144, P00260129 Phosphoaminophosphonic acid-guanylate ester (GNP) 0,99 145, 146 46, 47, 54, 67, 69, 101, 121, 122, 123, 124, 125, 126, 127, 173, 174, 176, 188, P00300149 Staurosporine (STU) 0,98 189 N-(ethoxycarbonyl)-l-leucyl-n-[(1r,2s,3s)-1-(cyclohexylmethyl)- P00430133 0,61 107, 109, 150, 151, 152, 192, 195, 290, 292, 294, 295, 296, 297, 392 2,3-dihydroxy-5-methylhexyl]-l-leucinamide (0ZL) 19, 20, 21, 22, 23, 24, 25, 35, 36, 37, 39, 41, 42, 68, 123, 124, 126, 127, 153, P00700035 Phosphoaminophosphonic acid-guanylate ester (GNP) 0,94 154, 155 P00680028 Iron (fe) 0,4 255, 264, 312 22, 23, 24, 25, 26, 27, 28, 38, 39, 40, 44, 45, 71, 127, 128, 130, 131, 157, 158, P00680119 Guanosine (GMP) 0,84 159 P00210141 S-adenosyl-l-homocysteine (SAH) 0,9 10, 36, 37, 87, 88, 91, 116, 117, 166, 167, 225, 249, 250, 251 C-score is the confidence score of predicted binding site. C-score values range in between [0-1]; where a higher score indicates a more reliable prediction.