Albacar et al. Microbial Cell Factories (2023) 22:63 Microbial Cell Factories https://doi.org/10.1186/s12934-023-02074-6 RESEARCH Open Access Transcriptomic profiling of the yeast Komagataella phaffii in response to environmental alkalinization Marcel Albacar1†, Abdelghani Zekhnini1†, Jorge Pérez-Valle1, José L. Martínez2, Antonio Casamayor1 and Joaquín Ariño1* Abstract Background Adaptation to alkalinization of the medium in fungi involves an extensive remodeling of gene expression. Komagataella phaffii is an ascomycetous yeast that has become an organism widely used for heterologous protein expression. We explore here the transcriptional impact of moderate alkalinization in this yeast, in search of suitable novel promoters able to drive transcription in response to the pH signal. Results In spite of a minor effect on growth, shifting the cultures from pH 5.5 to 8.0 or 8.2 provokes significant changes in the mRNA levels of over 700 genes. Functional categories such as arginine and methionine biosynthesis, non-reductive iron uptake and phosphate metabolism are enriched in induced genes, whereas many genes encoding iron-sulfur proteins or members of the respirasome were repressed. We also show that alkalinization is accompanied by oxidative stress and we propose this circumstance as a common trigger of a subset of the observed changes. PHO89, encoding a Na+/Pi cotransporter, appears among the most potently induced genes by high pH. We demonstrate that this response is mainly based on two calcineurin-dependent response elements located in its promoter, thus indicating that alkalinization triggers a calcium-mediated signal in K. phaffii. Conclusions This work defines in K. phaffii a subset of genes and diverse cellular pathways that are altered in response to moderate alkalinization of the medium, thus setting the basis for developing novel pH-controlled systems for heterologous protein expression in this fungus. Keywords Stress response, Oxidative stress, Promoter regulation, Protein production † Marcel Albacar and Abdelghani Zekhnini contributed equally to this work as co-first authors. *Correspondence: Joaquín Ariño joaquin.arino@uab.es 1 Institut de Biotecnologia i Biomedicina & Departament de Bioquímica i Biologia Molecular, Universitat Autònoma de Barcelona, Cerdanyola del Vallès 08193, Spain 2 Department of Biotechnology and Biomedicine, Section for Synthetic Biology, Technical University of Denmark, Kongens Lyngby, Denmark © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Albacar et al. Microbial Cell Factories (2023) 22:63 Page 2 of 17 Background densities, efficient mechanisms for secretion of produced Adaptation to changes in the pH of the environment is proteins to the medium, low level of secreted endoge- critical for survival of yeast cells. The mechanisms under- nous proteins (which facilitates purification of heterolo- lying such adaptation are largely mediated by remod- gous secreted products), and a more suitable pattern of eling of the transcriptional response and have been glycosylation of secreted proteins [17, 18]. Heterologous investigated with substantial detail in fungi such as Sac- protein expression in this organism is very often driven charomyces cerevisiae, Candida albicans, and Aspergil- from the powerful AOX1 promoter, required for metha- lus nidulans [1–4]. Because adaptation to niches with nol utilization [19]. This is a strongly controlled and very diverse pHs appears critical for fungal pathogenesis, pH potent promoter, but the toxic and flammable nature of signaling mechanisms have been proposed as possible methanol causes safety concerns and is considered a limi- antifungal targets [5]. Major signaling pathways responsi- tation in the process. This has led to the search of novel ble for this response include the conserved PacC/Rim101 promoters and induction systems that could be consid- pathway, the calcineurin phosphatase pathway, mediated ered an alternative [20, 21]. More recently, K. phaffii has by the transcription factors Crz1/CrzA, or the protein been proposed as an emerging model organism in biolog- kinase A (PKA) pathway. ical research, particularly suitable for the investigation of The PacC/Rim101 pathway is remarkably conserved methanol assimilation, mechanism of protein secretion, among fungi and is based in a signaling pathway (named peroxisome biogenesis and pexophagy, or the machinery the Pal pathway in filamentous fungi, and the Rim path- for maintenance of cell wall integrity [22]. way in yeast-like fungi) that is triggered by the activation Despite the relevance of this organism, the knowledge of a 7-transmembrane domain receptor that, ultimately, about the mechanisms activated in response to stress leads to the proteolytic processing and activation of the are, in comparison with other yeast such as S. cerevi- transcription factor PacC (Rim101 in S. cerevisiae). In S. siae or S. pombe, relatively scarce. This is a significant cerevisiae, most of the effects of Rim101 activation are limitation, since there are growing evidence that activa- due to the repression of the expression of Nrg1, which tion of stress responses positively influences the expres- in turn acts as a transcriptional repressor [6]. The calci- sion of recombinant proteins [23, 24]. In particular, the neurin pathway is activated by an increase in intracel- adaptive mechanisms allowing survival when pH in the lular calcium triggered by extracellular alkalinization [7, environment becomes moderately alkaline have not been 8] and results in activation of the protein phosphatase examined. In this work, we investigate the changes in calcineurin. Then, this enzyme dephosphorylates the the transcriptional landscape of K. phaffii occurring in transcription factor Crz1 and promotes its transit to the response to sudden alkalinization of the medium. We nucleus, where it activates a substantial number of genes present evidence that this situation triggers a widespread [7, 9]. The relevance of the PKA pathway in the adapta- remodeling of gene expression through mechanisms that tion to high pH has been also demonstrated, although its likely include the activation of calcineurin signaling. This role seems to be species dependent. Thus, in S. cerevi- knowledge constitutes a key step that will guide further siae, activation of PKA associates with decreased alkaline research to devise novel promoters able to respond to a tolerance [10], whereas in the human pathogen Crypto- very simple and inexpensive stimulus. coccus neoformans is involved in increasing tolerance to high pH [11]. Additional pathways have been identified Results in S. cerevisiae to be activated upon alkalinization, such Alkalinization alters mRNA levels for hundreds of genes in as the PHO regulon that responds to phosphate scarcity K. phaffii [12], the cell wall integrity Slt2/Rml1 pathway [13] or the K. phaffii X33 cells exponentially growing on glycerol Snf1 pathway [14]. It is worth noting that in S. cerevisiae were shifted from medium at pH ∼5.5 to pH 8.0 or 8.2, there are examples (such as ENA1 or PHO89) of promot- whereas similar cultures growing on glucose were shifted ers containing diverse structural elements that allow the to pH 8.0. Samples were taken just before the pH shift integration of signals mediated by various independent and after 15-, 30- and 60-min. RNA-seq analysis showed pathways to allow coordinated and timely transcription the expected differential profile, in the absence of stress, [15, 16]. between cells grown on glucose compared to those Komagataella phaffii (formerly called Pichia pastoris) grown on glycerol, namely increased expression of glyco- is an ascomycetous yeast characterized by an obligatory lytic genes, and repression of those relevant in gluconeo- aerobic metabolism and the ability to use methanol as genesis and glycerol utilization (Additional File 1, Supp. sole carbon and energy source. This yeast has become Figure 1). When the impact of alkalinization was exam- an organism extensively used for heterologous protein ined, we found 772 genes with a change in mRNA levels expression. This popularity derives from a number of of at least 2-fold (log2 > = 1 or <= -1, with a p-value < 0.01) advantages, such as the ability to grow at high cellular for at least one time or condition (Additional File 2). Of Albacar et al. Microbial Cell Factories (2023) 22:63 Page 3 of 17 these, 430 were induced and 378 repressed (36 genes Cys biosynthesis), while the latter contained diverse were induced under some circumstances and repressed genes related to arginine and thiamine biosynthesis. in others). As shown in Fig. 1A, the highest number of Conversely, the small cluster number 4, which includes changes was observed for cells grown on glycerol and genes particularly repressed in cells grown on glu- shifted to pH 8.2. The number of affected genes in cells cose, comprises the genes encoding the first and second shifted to pH 8.0 was slightly higher for cells grown on steps for Arg degradation (two putative Car1 arginases, glycerol than on glucose, mostly due to an excess of C4R298 and C4R8L9, and the Car2 (C4R4H3) ornithine induced genes (302 vs. 237). About 61% of genes induced aminotransferase). Cluster 5 contains genes generally at pH 8.0 in cells grown on glycerol were also induced in repressed after 30 and 60 min of stress, and it is highly cells grown on glucose, whereas for repressed genes the enriched in genes encoding proteins required for cel- overlap was of 68%. The transcriptional response was lular respiration and oxidative phosphorylation, as well very fast, with over 170 genes up-regulated and nearly as those containing iron-sulfur clusters. Cluster 7 con- 150 genes down-regulated 15 min after shifting cells to tains genes that code for enzymes with redox activity alkaline pH (Fig. 1B). In fact, in cells grown on glycerol and involved in transmembrane transport, and they are and shifted to pH 8.0, the value at 15 min was the peak constantly repressed from min 15 onwards. Genes tran- of the time-course for about 35% of both up-regulated siently induced after 15 and 30 min are found in cluster and down-regulated genes (Fig. 1C). In contrast, in cells 8, which is enriched in those encoding proteins involved shifted to pH 8.2, the peak was clearly delayed to 60 min, in oxidative stress response and, as in cluster 7, also with particularly for induced genes (56% of genes). redox activity. However, the overall temporal and func- The k-means algorithm was used to define 10 clusters tional enrichment profile of down-regulated (cluster 7) based in the expression profile of the genes (Fig. 2). Gene and up-regulated genes related to redox activity (clus- Ontology analysis of the different clusters revealed sev- ter 8) is very different (Fig. 3), since induction of redox- eral interesting associations. Cluster 2 and cluster 3 were related up-regulated genes is an early process (typically made of induced genes, and the former was enriched in at 15 and 30 min), whereas repressed genes are enriched genes involved in sulfur metabolism (mainly Met and after 60 min, particularly for cells shifted to pH 8.2. Fig. 1 Transcriptional impact of environmental alkalinization in K. phaffii. (A) Number of genes up-regulated or down-regulated (log2 > = 1 or <= -1 and p-value < = 0.01) for each of the conditions tested. (B) Time-course of emergence of up-regulated and down-regulated genes. (C) Number of genes show- ing the stronger expression change (peak) for each time and experimental condition Albacar et al. Microbial Cell Factories (2023) 22:63 Page 4 of 17 Fig. 2 Changes in expression for the 772 induced or repressed genes at least at one time-point and condition. Data was clustered with the Gene Cluster software v.3.0 [60] using k-means algorithm (10 clusters, 100 iterations, uncentered correlation) and visualized using Java TreeView [61]. The most relevant specific GO enrichment terms for the clusters generated are shown with the p-value in parentheses. The intensity of the expression change can be in- ferred by comparison with the enclosed scale (log2 values) Alkalinization of the medium triggers oxidative stress in K. becomes fluorescent in the presence of reactive oxygen phaffii species. As shown in Fig. 4B, exposure of the cells to The antioxidant response of K. phaffii to alkaline pH was moderate alkalinization (pH 7.4, by adding KOH up to further investigated. Figure 4A shows the expression 20 mM) already triggers a substantial increase in fluo- profile of diverse genes induced shortly after pH shifting rescence that is slightly augmented when pH is raised and encoding proteins widely recognized as involved in to 8.1 (30 mM KOH), as documented by flow cytometry the response to oxidative stress. These included C4R2S1 analysis. This effect can be also detected by fluorescence (Cta1, catalase A), C4QXN4 (Ahp1, a peroxiredoxin), microscopy Fig. 4C, revealing that cells become clearly C4R2H3 (Tsa1, Thioredoxin peroxidase), or C4R934 fluorescent after shifting the medium to pH 8.1. (Grx2, a Glutaredoxin). This profile suggested that expo- sure to alkaline pH triggers a situation of oxidative stress Impact of alkalinization on the metabolic transcriptome in K. phaffii. Such possibility was investigated by incu- As hinted in Fig. 2 (cluster 3) and 3, alkalinization causes bating cells with the dihydrorhodamine 123 dye, which the induction of diverse genes encoding proteins required Albacar et al. Microbial Cell Factories (2023) 22:63 Page 5 of 17 Fig. 3 Time course for up-regulated and down-regulated genes for diverse GO terms derived from the entire set of data. The enrichment for each time and condition is expressed as log10 of the p-value. The discontinuous line sets the p-value < 0.05 threshold for arginine biosynthesis. These included (Fig. 5A) both plasma-membrane arginine transporters (Can1/C4QYY6 subunits of carbamoyl-phosphate synthase (C4R6Y9/ and Gap1/ C4QYK8). Cpa1, and C4R5L7/Cpa2), C4QZV1 (Arg5,6), a bifunc- Genes involved in sulfate assimilation and Met and tional enzyme that catalyzes two steps in the cyclic trans- Cys biosynthesis are also induced upon alkalinization formation of glutamate into N2-acetyl-L-ornithine, as (Fig. 5B), including the ones encoding the sulfate trans- well as the consecutive steps catalyzed by C4R533 (Arg3, porter Sul1 (C4R4Q2), the zinc-finger DNA-binding ornithine carbamoyltransferase) and C4R3W1 (Arg1, transcription factor Met32 (C4QVF9), and a putative argininosuccinate synthase). Interestingly, in cells grown Mcy1 cysteine synthase (C4R122), able to catalyze the on glucose, the genes related to the steps involved in argi- conversion of O-acetyl-L-serine into cysteine. As it can nine degradation are repressed (Fig. 2, cluster 4), includ- be observed, the response was very fast and transient ing two genes encoding putative Car1 arginases (C4R298 for most genes (note that CYS3 was even repressed after and C4R8L9), the one that encodes Car2 (C4R4H3), and 45 min). The exception was MUP1, encoding a high affin- even Put2 (C4R1 X 5), albeit the latter does not reach ity methionine permease involved in both methionine the log2 < = 1 threshold. We also observe a moderate and cysteine uptake. decrease in the expression of genes encoding possible Albacar et al. Microbial Cell Factories (2023) 22:63 Page 6 of 17 Fig. 4 Alkalinization of the medium triggers an oxidative stress response in K. phaffii. (A) Changes in expression for diverse genes known to be responsive to oxidative stress in yeasts. Data corresponds to cells grown in glycerol and switched to pH 8.2. The discontinuous red line indicates, in this and subse- quent figures, the log2 > = 1 and <= -1 thresholds. (B) K. phaffii X33 cells were grown in YPD medium and incubated for 1 h with dihydrorhodamine 123. Aliquots were taken and treated with KCl (30 mM), or KOH (20 and 30 mM, raising the pH to 7.4 and 8,1, respectively). After 30 min of incubation, cells were processed for flow cytometry. The lower panel shows the fluorescence shift of the population treated with 30 mM KOH. (C) Cultures treated with 30 mM KOH) were observed under a Nikon Eclipse TE2000-E using the FIT-C filter. Scale bar denotes 5 μm Alkalinization triggers the induction of genes involved starvation, such as C4QZX3 (Pet18), C4QYB3 (Sno1-3), in the uptake and biosynthesis of thiamine, a response and C4QYB4 (Snz1-3) were also clearly induced by high very similar to that observed upon thiamine starvation. pH. The latter two genes encode activities required for As shown in Fig. 6, an increased expression of genes the synthesis of pyridoxal 5’-phosphate, the substrate for encoding C4QZN1 (a Pho11/12 phosphatase, able to Thi13. hydrolyze thiamin phosphates in the periplasmic space), As shown in Fig. 7A, the expression of genes involved two plasma membrane thiamin transporters (C4QWN5/ in copper and iron metabolism was potently altered. Thi73 and C4R561/Thi7), and several enzymes of the thi- Among them, at least three different Arn1-3 like sidero- amine biosynthetic pathway, such as Thi4, Thi20 and Thi6 phore transporters (C4R573, C4R468, and C4QZD7), two was observed. Curiously, Thi13, which catalyzes the first putative members of the ferric and cupric reductase sys- step in the pathway, was initially repressed and potently tem (Fre1 and Fre2), the low-affinity Fe(II) transporter induced later. The gene encoding C4R0 X 5, a putative Fet4, and the plasma membrane high-affinity copper mitochondrial thiamine pyrophosphate carrier, was transporter Ctr1 (C4R733) were induced. The gene strongly repressed. At least three genes encoding pro- encoding the high-affinity zinc transporter of the plasma teins known to be induced in S. cerevisiae upon thiamine membrane (C4R4S9, Ztr1) was among the most potently Albacar et al. Microbial Cell Factories (2023) 22:63 Page 7 of 17 Fig. 5 Changes in expression in genes related to the biosynthesis of arginine (A) and the metabolism of sulfur and the amino acid cysteine (B). Data for A and B correspond to cells grown on glucose or glycerol, respectively, and shifted to pH 8.0. Genes induced are denoted in red and those repressed in blue induced genes (not shown). Interestingly, two important protein C4R3N5 was among the most strongly repressed components of the iron uptake system, Fet3 (C4R1S1) genes. and Ftr1 (C4QX98), were clearly repressed. Fet3 is the Taken together, these results indicate that the alkalini- plasma membrane Ferro-O2-oxidoreductase that oxidizes zation of the medium triggers a transcriptional response Fe2+ to Fe3+ for subsequent uptake by the transmembrane that affects the metabolism of arginine and sulfur-con- permease Ftr1.To note, the gene encoding C4R794, a pro- taining amino acids, thiamine, as well as the homeostasis tein very closely related to S. cerevisiae Fre2 and Fre3, of copper, iron and phosphate. was also markedly repressed. Alkalinization causes a widespread transcriptional Phosphate metabolism appears also affected by alka- repression of respiratory metabolism. linization of the medium Fig. 7B. This included genes GO analysis of genes showing a down-regulation pat- encoding components of the vacuolar polyphosphate tern highlighted many genes related to oxidative phos- biosynthetic machinery (Vtc1, Vtc2/3, Vtc4), the puta- phorylation (Figs. 2 and 3). As shown in Fig. 8A, changes tive transcription factor Pho4 (C4QZW1) and the increased with time and were more intense at the latest Na+/Pi cotransporter Pho89 (C4R021), which was one of time tested (60 min). Down-regulated genes included the genes most potently induced. In contrast, we found widespread components of the electronic transport chain, two genes encoding putative plasma membrane phos- involved in generation of complex II (SDH1, SDH2, SDH3 phate transporters (C4R3N5 and CAR7K5), equivalent and SDH4), complex III (CYT1, COR1, RIP1, QCR6, to the S. cerevisiae Pho84 H+/Pi high affinity transporter, QCR7, QCR8, and QCR10), and complex IV (COX15), as that were repressed by alkaline pH. The one encoding well as several other related but less well characterized Albacar et al. Microbial Cell Factories (2023) 22:63 Page 8 of 17 Fig. 6 Transcriptional effects on the thiamine biosynthetic pathway. Data corresponds to cells gown on glycerol and shifted to pH 8.2. Induced genes are in red. The asterisk (*) in THI13 indicates that this gene is repressed at time 15 and 30 min, but induced at 60 min proteins. In addition, a decrease in the mRNA levels and PHO84 encode putative Na+/Pi and H+/Pi plasma of many genes encoding glycolytic and gluconeogenic membrane high affinity transporters, respectively, and enzymes was observed (Fig. 8B). For instance, all genes were selected because their unexpected opposite behav- encoding the subunits of phosphofructokinase-1 (PFK1), ior (see Discussion). GAS1 encodes a putative β-1,3- a key regulator of glycolysis, were repressed, including glucanosyltransferase, required for cell wall assembly, and the one coding for C4R7H4, a γ PFK1 subunit that is not GDH3 codes for a putative NADP+-dependent glutamate found in other fungi. We noted that ADH3, GPM1 and dehydrogenase. As shown in Fig. 9A, PHO89 and GAS1 PGK1, which suffered a weak repression in glycerol cul- reporters confirmed the strong activation of both pro- tures, were more potently repressed when grown on glu- moters by alkaline pH. We noted that they presented dif- cose, decreasing below the log2 <= -1 threshold. ferent pH sensitivity, since GAS1 was almost maximally induced already at the lowest pH tested (7.4), whereas Characterization of the response of selected genes by PHO89 displayed a marked dose-dependent response. translational promoter fusions The activation of PHO89 and GAS1 could be also easily To further test the transcriptional response to alkaline monitored by fluorescence microscopy (Fig. 9B). Analysis pH determined by RNA-Seq we selected two strongly of PHO84 and GDH3 confirmed the repression of these induced genes (PHO89 and GAS1) and two repressed promoters, although the changes were less potent than ones (C4R3N5, corresponding to one of the two likely that detected by RNA-Seq (compare with Additional file PHO84, and GDH3), for GFP-fused promoter reporter 1) and showed different kinetic profile. assays (Additional File 1). As mentioned above, PHO89 Albacar et al. Microbial Cell Factories (2023) 22:63 Page 9 of 17 Fig. 7 Transcriptional effects on the iron and copper (A) and phosphate uptake and utilization (B) pathways. In both cases, cells were grown on glycerol and shifted to pH 8.0. See main text for details Fig. 8 Repression of oxidative phosphorylation (A) and glycolysis/gluconeogenesis related (B) genes. Data corresponds to glycerol cultures shifted to pH 8.2. See main text for details Albacar et al. Microbial Cell Factories (2023) 22:63 Page 10 of 17 Fig. 9 Analysis of pH dependent promoter response of selected genes by translational reporters. (A) The indicated GFP-derived reporters were intro- duced in K. phaffii X-33 cells and cultures were shifted to the different pHs by addition of KOH. Samples were taken at the indicated times and processed for flow cytometry analysis. Data are mean ± SEM from 3 to 5 experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001 determined by paired Student’s t-test. (B) Cultures bearing PHO89-GFP and GAS1-GFP constructs were monitored by fluorescence microscopy after growing at normal pH (5.6) or alkaline pH (8.3) for 4 h In an earlier report it was shown that, in S. cerevisiae, was responsive only when rather high concentrations of the induction of PHO89 by alkaline pH is mainly medi- CaCl2 (≥ 0.6 M) were added to the medium and after a ated by the calcineurin-Crz1 pathway in response to the relatively long period of incubation (> 4 h). In any case, immediate influx of calcium into the cell [7]. We sought the induction caused by external calcium was partially to investigate if this would be the case in K. phaffii. Com- abolished by mutation of either CDRE 1 or CDRE 2, and putational analysis of the PHO89 promoter in K. phaffii fully eliminated when both sites were altered, thus sup- revealed diverse putative regulatory sites, including a porting the calcium-mediated nature of PHO89 induc- possible Pho4 site, but also two putative CDREs (Calci- tion in response of alkaline pH. neurin Dependent Response Element), that we named CDRE 1 and CDRE 2 (Fig. 10A). To test the possible func- Expression levels achieved by pH-regulatable alkaline tional role of these sites, we mutagenized both of them promoters are in the range of those of constitutive at the crucial GCC core either separately or in combina- promoters commonly used for expression of heterologous tion. As shown in Fig. 10B, mutation of either CDRE site proteins decreased the response of the reporter to alkalinization, To evaluate the usefulness of the newly discovered alka- although the effect of the mutation of CDRE 2 was more line pH-regulatable promoters for industrial application prominent than that of CDRE 1. Removal of both sites we sought to assess the amount of mRNA generated in further decreased the promoter response to less than comparison with that of genes whose expression is driven 20% of the observed for native version. These results indi- from well-known constitutive promoters widely used cate that both sites are functional (although not equally for heterologous protein production [25]. To this end relevant) and that in K. phaffii the PHO89 gene is likely we calculated the RPKM (reads per kilobase per million subjected to transcriptional regulation by the calcium- mapped reads) values for each condition, which allow calcineurin pathway. Further support for this notion was comparing gene expression values within a sample. As obtained by incubating the cells with CaCl2 and moni- shown in Fig. 11, expression of HSP12 and TSA1 nearly toring the activity of the native and mutated promot- matched that of GAP, and it was from 65 to 75% of that ers (Fig. 10C). We observed that the PHO89 reporter of TEF1, which was the second most potently transcribed Albacar et al. Microbial Cell Factories (2023) 22:63 Page 11 of 17 Fig. 10 Putative CDRE elements are crucial for pH-dependent induction of PHO89 expression. (A) Putative CDRE elements were identified in the PHO89 promoter (CDRE 1 and CDRE 2). (B) The constant GCC core was mutated individually or in combination (CDRE 1 + 2) and the response of cells bearing PHO89-GFP reporters carrying such mutations was compared by flow cytometry with that of the native version (denoted as PHO89). Data is represented as the mean ± SEM of the fold-change in comparison with unstressed cells from 4 experiments. (C) Cells bearing the reporters indicated above were incubated for 8 h with 0.6 M KCl or CaCl2 and samples taken at t = 0 and 8 h for flow cytometry analysis. Values were compared with cells incubated with KCl to avoid interference of possible osmotic effects. Data is the mean ± SEM of the fold-change from 3 or 4 experiments. *, p < 0.05; **, p < 0.01; ns, not significant (p > 0.05) gene in our dataset. PHO89 was the most intensely (8.2) the impact on growth rate was relatively mild (see induced gene by alkalinization (nearly 90-fold, see Addi- Additional File 1, Supp. Figure 4) in comparison with S. tional File 1, Supp. Figure 3 and Additional File 2), and cerevisiae, which hardly grows at pH 8.2 or above. We its expression level in cells grown on glycerol was higher also observe that the overall response is quite similar than that of PGK1 or TPI1 in cells growing on either irrespectively of the carbon source used (Fig. 1). On the carbon sources. Therefore, the mRNA levels attained other hand, increasing the intensity of the stress (from upon induction from alkaline pH-regulatable promot- pH 8.0 to 8.2) tends to delay the transcriptional response ers are high enough to suggest that they could be used to (Figs. 1B and C and 3). This is reminiscent of what has develop novel platforms for heterologous protein expres- been reported in response to osmotic stress in S. cerevi- sion in K. phaffii. siae [32–34], likely as a result of a more prolonged adap- tation phase until effective transcription occurs. Discussion We show that alkaline pH leads to the development of Work in the last twenty years has shown that environ- a situation of oxidative stress in K. phaffii. This was pre- mental alkalinization results in robust transcriptional viously observed for S. cerevisiae [7, 12, 35], although it remodeling that affects hundreds of genes not only in S. was not reported in the case of C. albicans [29]. We also cerevisiae [7, 12, 13, 26–28], but also in other fungi such detect a marked increase in a subset of genes required as C. albicans [29], Aspergillus nidulans [29, 30] or the for assimilation of iron and copper. The induction of this mycotrophic fungus Trichoderma virens [31]. Data pre- kind of genes upon alkalinization has been observed in sented in this report indicates that this is also the case S. cerevisiae and C. albicans [12, 28, 29] and the avail- for the methylotrophic yeast K. phaffii, for which we ability of these metals at alkaline pH has been shown detect over 700 genes up- or down-regulated. This was to be a limiting factor for growth in S. cerevisiae and S. somewhat surprising, since even at the higher pH tested pombe [35, 36]. However, the specific expression profile Albacar et al. Microbial Cell Factories (2023) 22:63 Page 12 of 17 Fig. 11 Relative expression of selected alkaline pH-inducible genes and constitutively expressed genes whose promoters are widely used for heterolo- gous protein expression. The RPKM (reads per kilobase per million mapped reads) were calculated for each gene and condition in our dataset and plotted for five selected alkaline-inducible genes and nine constitutive genes (the latter selected from Table 1 in reference [25]). The scale of the Y axes has been kept the same in all graphs to facilitate comparison of iron-related genes in K. phaffii is somewhat particu- non-reductive pathway, thus diminishing oxidative dam- lar. Indeed, many genes belonging to the non-reductive age derived from ferrous ions import through the Ftr1/ uptake system, encoding siderophore transporters as well Fet3 complexes [38]. In contrast, putative FTR1 and as the oxygen-independent low-affinity plasma mem- FET3 homologs are found induced upon environmental brane iron and copper transporter Fet4, were induced. alkalinization in C. albicans [29]. In contrast, the genes encoding the copper-dependent Related to iron metabolism, we also observe induc- high-affinity iron import complex Fet3/Ftr1, as well as a tion of the gene encoding C4R7J6 (named here CTH1), putative homolog of Fre3, one of the cell surface metal- which is the closest relative to S. cerevisiae Cth1 and loreductases that cooperates with Fet3/Ftr1 [see [37] Cth2 (Tis11) proteins. It is known that, in S. cerevisiae, and references therein] were clearly repressed. Whereas Cth2 acts as a mRNA-binding protein that, by posttran- induction of genes relevant for the non-reductive path- scriptional inhibition, restricts the expression of genes way could be mainly attributed to the decreased bio- coding for iron-containing proteins or that participate availability of iron, derived from the low solubility of this in iron-consuming processes [see [37] and references metal (particularly as Fe+ 3) at neutral and alkaline pH therein]. These include components of the mitochondrial [38], the decrease in mRNA of FET3 and FTR1 could be electron transport chain or the tricarboxylic acid (TCA) due to the concurrent condition of oxidative stress. In cycle. Indeed, we observe a marked decrease in the genes this regard, it has been shown that in S. cerevisiae under encoding proteins containing iron-sulfur clusters (Figs. 2 oxidative stress caused by ter-butyl hydroperoxide [39, and 3), such as SDH2, LEU1, LYS4, ACO1, ACO2, GLT1, 40] or arsenate [41] FTR1 and FET3 mRNA levels are and GRX3, among others. Repression of iron-sulfur maintained low by transcript destabilization through containing genes (SDH2, GLT1, COX15, among oth- the general 5’–3’ mRNA decay pathway. This response ers) can be also observed in C. albicans [29]. In addi- would allow redirecting iron assimilation through the tion, we detect an overall repression of genes involved in Albacar et al. Microbial Cell Factories (2023) 22:63 Page 13 of 17 oxidative phosphorylation, mostly encoding components strongly decrease tolerance to sulfur dioxide stress [46]. of the electron transport chain, as well as additional ele- Interestingly, in C. albicans, where arginine biosynthe- ments of the TCA cycle, such as isocitrate dehydroge- sis is repressed and methionine biosynthetic genes are nase (C4R142) or the Kgd2 subunit of α-ketoglutarate barely affected, exposure to alkaline pH does not gen- dehydrogenase, which possibly contribute to the smaller erate a strong transcriptional oxidative stress response growth rate of the culture in the first few hours (Addi- and, in fact, diverse genes induced in K. phaffii are even tional File 1, Supp. Figure 4). Therefore, it is likely that repressed (i.e., SOD2, CAT1, TTR1, AHP1 or HSP21). most of the changes described above would be the result The increased expression of thiamine biosynthetic of the combination of iron unavailability and oxidative genes observed here is intriguing because this is a very stress. It is worth noting that in the halotolerant yeast expensive process from the point of view of energy and Debaryomyces hansenii, which is mainly a respiratory it is mainly regulated at the transcriptional level [see [49] yeast, a down-regulation of genes involved in glycoly- and references therein]. Indeed, their expression seems sis and oxidative phosphorylation was also reported in largely unchanged in alkali-treated cultures of S. cerevi- response to alkalinization of the medium [42]. siae, C. albicans or A. nidulans. Still, the emergence of Diverse metabolic pathways are upregulated in K. these genes as induced in K. phaffii could be related to phaffii by high external pH, including the biosynthesis of oxidative stress. For instance, some of them (i.e. THI13, arginine and sulfur-containing amino acids, as well as the THI4, THI20, or THI6) were identified as activated in biosynthesis of thiamine. The behavior of arginine metab- response of heat shock [44] and, in addition, an increase olism in response to alkalinization appears to be species in both the transcript levels and the enzymatic activity dependent. Thus, in S. cerevisiae, genes involved in argi- was observed for Thi4 and Thi6 in yeast subjected to oxi- nine biosynthesis are consistently up-regulated [7, 10, 13, dative stress by hydrogen peroxide [50]. 26], although CAR1 and CAR2, involved in arginine uti- We have found that diverse genes involved in the lization, appear also up-regulated. Up-regulation of argi- response to phosphate scarcity are induced by alkalini- nine biosynthetic genes is also observed in A. nidulans, zation in K. phaffii. Among them, the Na+/Pi transporter accompanied by a strong down-regulation of AN8279, PHO89 was among the most potently induced genes. encoding a homolog of the plasma-membrane arginine Induction of PHO89 by low phosphate conditions in K. transporters CAN1 and GAP1 [30]. In sharp contrast, the phaffii was reported earlier [51] and these same authors expression pattern in C. albicans in response to alkaline highlighted the presence of two putative Pho4 binding pH is the opposite: biosynthetic genes are repressed, and sites (CACGTG/T) at positions − 583 and − 414 from degradative ones are induced [29]. Concerning methio- the initiating ATG (albeit these sites were not further nine metabolism, alkaline pH triggers the induction of characterized). We reported long ago [7, 12] that, in S. the biosynthetic pathway in S. cerevisiae, including genes cerevisiae, alkaline pH induction of PHO89 was largely (MUP1, MUP2) involved in the uptake of the amino acid dependent on the activation of the calcineurin pathway [7, 10, 26]. In contrast, the effect of alkalinization on this and the Crz1 transcription factor. Our data demon- pathway seems very limited in C. albicans and none in A. strates that in K. phaffii the PHO89 promoter contains nidulans [29, 30]. It is worth noting that, in addition to a tandem of two putative Crz1 binding sites, located at environmental alkalinization, a concomitant induction positions − 405/-461 that, working together, are almost of genes involved in biosynthesis of arginine and methio- fully responsible for the alkaline pH response of this nine has been observed by transcriptomic and/or metab- gene. These sites are also responsible for the induction of olomic analyses in S. cerevisiae cells subjected to different PHO89 in response to extracellular calcium. It must be forms of stress, such as heat stress [43, 44], arsenate [45] noted that K. phaffii seems rather refractory to the inter- and sulfur dioxide [46] treatments, or potassium starva- nalization of extracellular calcium under standard growth tion [47]. It is suggestive that all these stress conditions conditions, since very high amounts of the cation must are accompanied by a situation of oxidative stress, as we be added to elicit PHO89 induction. In any case, our data show here for K. phaffii under alkaline stress. Therefore, suggest that environmental alkalinization promotes the it is tempting to speculate that the concerted induction entry of extracellular calcium in K. phaffii which, through of arginine and methionine biosynthetic genes would activation of calcineurin and the Crz1 transcription fac- be a response to oxidative stress elicited by extracellular tor, would activate PHO89 expression. In addition to S. alkalinization. This scenario is supported by the previous cerevisiae [7, 12], the sensitivity of PHO89 to alkaline pH observation that the beneficial effect on ethanol tolerance has been also documented for Candida albicans [8, 29] in S. cerevisiae derived from arginine overproduction and Aspergillus nidulans [30]. Therefore, the regulation was, at least in part, due to decreased ROS production of this Na+/Pi antiporter gene by high pH through the [48]. In addition, deletion of diverse genes encoding calcium/calcineurin signaling seems a widespread feature methionine or arginine biosynthetic genes was shown to in fungi, pointing out to a conserved mechanism. Albacar et al. Microbial Cell Factories (2023) 22:63 Page 14 of 17 It is worth noting that, together with PHO89, previ- 20 g/L of glycerol (YPGly) or glucose (YPD) with con- ous work in S. cerevisiae [12, 28] and C. albicans [29] tinuous shaking (230 rpm/min). The medium contained has identified also the gene encoding the H+/Pi cotrans- 200 µg/ml of hygromycin B for selection of transfor- porter PHO84 as upregulated in response to high pH. K. mants. Growth data under different pH conditions were phaffii contains two genes encoding Pho84-like proteins obtained by monitoring the OD600 in a Bioscreen C appa- (C4R3N5 and C4R7K5) but, in this case, we observe a ratus (Thermo Fisher) as described in [52]. strong decrease in their mRNA levels, particularly in the case of C4R3N5. Indeed, analysis of the gene promoter by DNA techniques using a GFP reporter confirmed a moderate and relatively Escherichia coli (strain DH5α) was used as standard slow decrease in activity (Fig. 9). However, it is difficult to plasmid DNA host and was grown in LB medium at justify the very fast and potent decrease in PHO84 mRNA 37 °C supplemented with 100 µg/ml hygromycin B when levels solely based on promoter repression. One possible needed for plasmid selection. Transformation of E. coli, explanation would be a rapid destabilization of PHO84 and standard recombinant DNA techniques were per- mRNA. In fact, a fast mRNA destabilization upon envi- formed as described [53]. Yeast cells were transformed ronmental alkalinization was shown to be a general trait by electroporation following the procedure described in for genes showing a reduction in their mRNA levels in [54]. S. cerevisiae [27]. While the downregulation of putative Pho84-encoding genes in response to high pH observed Construction of gene reporters here might appear surprising, it is worth mentioning that Gene reporters were made by translational fusion the recent analysis of the transcriptome of A. nidulans to eGFP and subsequent integration in the K. phaffii in response to alkalinization [30] reveals a clear down- genome. To this end, we amplified by PCR from genomic regulation (∼4-fold) of ANIA_01612, which codes for the DNA obtained from strain X-33, using Q5 DNA poly- closest relative to S. cerevisiae Pho84. Therefore, it seems merase (New England Biolabs), the following genomic that whereas induction of PHO89 by high pH appears a regions (from the initiating ATG codon): -772/+11 from common feature in fungi, the specific response of Pho84- PHO89 (PAS_chr2-1_0235); -839/+6 from GAS1 (PAS_ encoding genes appears very much determined by the chr1-3_0227), -981/+21 from PHO84 (PAS_chr3_0141), fungal species. and − 752/+30 from GDH3 (PAS_chr1-1_0107). The first three fragments were digested with HindIII and KpnI, Conclusions whereas the GDH3 region was digested with BstBI and In conclusion, our work has uncovered a vast array of KpnI. All fragments were ligated into the corresponding genes that are sensitive to a moderate alkalinization in sites of plasmid pAHYB–GFP [55] and the mixtures used the yeast K. phaffii. Due to the interest of this organism to transform E. coli cells. Positive clones were selected for heterologous protein expression and that (1) modu- in LB plates containing hygromycin B. The inserts from lation of the pH of the medium is a very easy and eco- selected clones were sequenced to ensure the absence of nomical procedure, and (2) the expression level achieved unwanted mutations. in some cases is comparable to that of well-known strong Mutation of the putative CDRE consensus sequences constitutive promoters, some of these genes may provide in the PHO89 promoter was done as follows. Mutation of novel promoters useful for this purpose. Examples would CDRE-1 (nt -455/-446) was made by two-step PCR, using be HSP12, TSA1 or even PHO89, a gene potently induced the pAHYB-PHO89 reporter as template. In the first step, by alkaline pH and whose performance in response to two fragments were amplified using primer pairs Pp_ phosphate starvation was previously found to be compa- Pho89_Fw / PHO89-CDRE-1-Rev, and PHO89-CDRE- rable to that of well-known strong K. phaffii promoters, 1-Fw / Pp_Pho89_Rev (348 and 492 bp, respectively). such as of the genes encoding translation elongation fac- Both amplification fragments were diluted and re-ampli- tor 1α and glyceraldehyde-3-phosphate dehydrogenase fied with oligos Pp_Pho89_Fw and Pp_Pho89_Rev. For [51]. Understanding how K. phaffii signals and adapts to CDRE-2, the strategy was the same but using primers moderate alkalinization will provide information to fur- PHO89-CDRE-2-Fw and PHO89-CDRE-2-Rev in the ther refine the response of alkaline pH-regulatable pro- first step reactions, to yield fragments of 381 and 454 nt, moters to enhance their suitability for protein production respectively. Upon digestion with HindIII and KpnI and at the industrial level. cloning into pAHYB-GFP, the resulting constructs were denominated pPHO89-CDRE-1 and pPHO89-CDRE-2, Materials and methods respectively. The double mutation was constructed in Yeast strains and growth conditions parallel by mutating CDRE-2 using pPHO89-CDRE-1 as K. phaffii X-33 was grown at 28 °C in YP medium (10 g/L template, and mutating CDRE-1 using pPHO89-CDRE-2 yeast extract, 20 g/L peptone) supplemented with as amplification source, yielding in both cases plasmid Albacar et al. Microbial Cell Factories (2023) 22:63 Page 15 of 17 pPHO89-CDRE-1 + 2. In all cases, plasmids were linear- www.ncbi.nlm.nih.gov/assembly/GCA_000027005.1/), ized within the AOX1 promoter with PmeI, and 100 ng yielding 98.4–99.2% mapped reads. The SeqMonk soft- of DNA used to electroporate X-33 cells. Transformants ware v1.48.0 (https://www.bioinformatics.babraham. were recovered in YPD plates containing hygromycin ac.uk/projects/seqmonk/) was employed to analyze (200 µg/ml). The correct insertion of the plasmid into the the SAM files. Mapped reads were counted using the chromosomal AOX1 locus was confirmed by PCR. Oligo- RNA-Seq pipeline on mRNA features and differentially nucleotides used in this work and details on their use are expressed genes with respect to time zero were identi- documented in Additional file 3. Promoter analysis was fied using the built-in DESeq2 R package, selecting the performed with the RSAT platform [56]. multiple testing correction option [58]. Genes with a p-value < 0.01 (FDR < 0.05) and a log2 fold-change <-1 or RNA purification and RNA-Seq procedures > 1 were selected for further analysis. Functional annota- For the analysis of the transcriptional response of K. tions were obtained from NCBI’s “Feature Table” of the phaffii to alkalinization of the medium cells were grown GCA_000027005.1 assembly and enriched with data overnight on YPD or YPGly medium, and then cultures retrieved from Uniprot (Proteome ID UP000000314). were diluted in 200 ml of fresh medium at OD600 = 0.2 In addition, to gain for further information, a Blastp (∼ 107 cells/ml) and grown until an OD600 around 0.6. search against the proteomes of Saccharomyces cere- Then two 100 ml-aliquots were centrifuged (5 min at visiae, Schizosaccharomyces pombe, Candida albicans, 1000xg). One aliquot was resuspended in the same vol- Aspergillus nidulans and Cryptococcus neoformans pro- ume of either fresh YPD or YPGly medium plus 50 mM teomes was carried out, setting a p-value threshold of TAPS pH 5.5 (control cells) or in YPD or YPGly supple- 10− 30. Gene Ontology analyses for functional profiling mented with 50 mM TAPS adjusted to pH 8.0 or 8.2 were done with the g.Profiler server (https://biit.cs.ut.ee/ (stressed cells). Samples (15 ml) were taken immediately gprofiler/gost) using the K. phaffii GS115 data from the after resuspension in medium at pH 5.5 (time = 0) and, Ensemble Genomes fungi database. for each stressed culture, after 15, 30 and 60 min. In all cases, samples were placed on ice and immediately cen- Evaluation of reporter expression trifuged at 4 °C. Total RNA was extracted with the Ribo For investigation of the response of reporters to pH Pure™-Yeast kit (Ambion™) following the manufactur- changes, o/n cultures were diluted in YPD to OD600 = 0.3– er’s instructions. The quality was assessed by capillary 0.4 and grown until OD600 = 0.6. Cultures were split into electrophoresis in a Bioanalyzer equipment and RNA aliquots (8–12 ml), and the appropriate volume of 1 M amounts quantified by measuring the A260 in a Nanodrop KOH added to achieve a final concentration of 20–35 ND-1000 Spectrophotometer. mM (pH 7.4 to 8.4). Samples (0.875 ml) were taken at mRNA was purified from total RNA using poly-T different times and cells were fixed by mixing with 50 µl oligo-attached magnetic beads. After fragmentation, the of formaldehyde (37.5% solution) for 5 min at room first strand cDNA was synthesized using random hex- temperature, washed twice with the same volume of amer primers, followed by the second strand cDNA syn- cold PBS and finally resuspended in 200–300 µl of PBS. thesis using dTTP (non-directional library). The library Fluorescence of the samples was analyzed (20,000 cells/ was checked by Qubit™ assay (Thermo Fisher Scientific) sample) in a FACSCanto flow cytometer (Benton & Dick- and real-time PCR for quantification and size distribu- inson) with the FITC filter (excitation 488 nm / emission tion monitored with a Bioanalyzer 2100 equipment (Agi- 530 nm) and/or monitored by fluorescence microscopy lent). Quantified libraries were pooled and sequenced in a Nikon Eclipse TE2000-E using the FITC filter (excita- on an Illumina NovaSeq 6000 platform by Novagene Co. tion 482 nm / emission 536 nm) with 1000 ms of exposi- Ltd. An average of 22.8 million of single reads per sample tion. Evaluation of the response of the PHO89 promoter were used for subsequent analysis (98.7% were effective to external calcium was performed similarly but adding reads and 94.49% were Q30 or higher on average). Two calcium to 0.6 M from a 4 M stock solution. biological replicas were done for each condition, except for pH 8.0 at 15 min, for which data from four biological Monitoring of ROS generation replicas was obtained. DNA sequence data was deposited K. phaffii X-33 cells exponentially growing in YPD at NCBI’s Gene Expression Omnibus (GEO) and can be medium were incubated for 1 h with dihydrorhodamine retrieved under study GSE224427. 123 (D1054, Sigma) at a final concentration of 2.5 µg/ ml, as described in [7, 59]. Then, aliquots were taken and Analysis of RNA-Seq data treated with KCl or KOH (20 and 30 mM, raising the pH Mapping of fastq files to generate SAM files was carried to 7.4 and 8.1, respectively). After 30 min of incubation, out with the Bowtie2 software [57] in local / sensitive samples were taken, cells fixed with formaldehyde as mode against the GCA_000027005.1 assembly (https:// Albacar et al. Microbial Cell Factories (2023) 22:63 Page 16 of 17 5. Cornet M, Gaillardin C. pH signaling in human fungal pathogens: a new described above, and analyzed by fluorescence micros- target for antifungal strategies. Eukaryot Cell. 2014;13:342–52. copy and flow cytometry. 6. Lamb TM, Mitchell AP. The transcription factor Rim101p governs ion toler- ance and cell differentiation by direct repression of the regulatory genes NRG1 and SMP1 in Saccharomyces cerevisiae. Mol Cell Biol. 2003;23:677–86. Supplementary Information 7. Viladevall L, Serrano R, Ruiz A, Domenech G, Giraldo J, Barcelo A, Arino J. The online version contains supplementary material available at https://doi. Characterization of the calcium-mediated response to alkaline stress in Sac- org/10.1186/s12934-023-02074-6. charomyces cerevisiae. J Biol Chem. 2004;279:43614–24. 8. Wang H, Liang Y, Zhang B, Zheng W, Xing L, Li M. Alkaline stress triggers an Additional file 1. Supplemental Fig. 1 (ratio of expression of cells growing immediate calcium fluctuation in Candida albicans mediated by Rim101p on glucose or glycerol prior shifting cells to alkaline pH), 2 (transcriptomic and Crz1p transcription factors. FEMS Yeast Res. 2011;11:430–9. profiling of genes selected for construction of GFP reporters), 3 (Induction 9. Roque A, Petrezsélyová S, Serra-Cardona A, Ariño J. Genome-wide recruit- factor for selected genes based on RPKM values) and 4 (Growth of K. phaffii ment profiling of transcription factor Crz1 in response to high pH stress. BMC under moderate alkaline pH) Genomics. 2016;17:662. 10. Casado C, González A, Platara M, Ruiz A, Ariño J. The role of the protein kinase Additional file 2. Genes induced and repressed at least 2-fold A pathway in the response to alkaline pH stress in yeast. Biochem J. 2011. Additional file 3. Oligonucleotides used in this work https://doi.org/10.1042/BJ20110607. 11. O’Meara TR, Norton D, Price MS, Hay C, Clements MF, Nichols CB, Alspaugh JA. Interaction of Cryptococcus neoformans Rim101 and protein kinase a Acknowledgements regulates capsule. PLoSPathog. 2010;6:e1000776. The skillful technical assistance of Montserrat Robledo is acknowledged. 12. Serrano R, Ruiz A, Bernal D, Chambers JR, Arino J. The transcriptional response We appreciate the collaboration of Paula Iglesias and Natalia Brando in the to alkaline pH in Saccharomyces cerevisiae: evidence for calcium-mediated generation of the mutation of CDREs in the PHO89 promoter. signalling. MolMicrobiol. 2002;46:1319–33. 13. Serrano R, Martín H, Casamayor A, Ariño J. Signaling alkaline pH stress in the Author Contribution yeast Saccharomyces cerevisiae through the Wsc1 cell surface sensor and the MA prepared RNA from cultures and contributed to reporter studies. AZ and Slt2 MAPK pathway. J Biol Chem. 2006;281:39785–95. JP-V contributed to reporter studies. JLM supervised transcriptomic data. AC 14. Casamayor A, Serrano R, Platara M, Casado C, Ruiz A, Ariño J. The role of the prepared RNA from cultures and contributed to RNA-seq data analysis and Snf1 kinase in the adaptive response of Saccharomyces cerevisiae to alkaline reporter preparation. JA analyzed RNA-seq data, prepared and analyzed gene pH stress. Biochem J. 2012;444:39–49. reporters, and drafted the manuscript. All authors read and approved the final 15. Platara M, Ruiz A, Serrano R, Palomino A, Moreno F, Arino J. The Transcrip- manuscript. tional response of the yeast Na+-ATPase ENA1 gene to alkaline stress involves three Main Signaling Pathways. J Biol Chem. 2006;281:36632–42. Funding 16. Serra-Cardona A, Petrezsélyová S, Canadell D, Ramos J, Ariño J. Coregulated This research was funded by the Ministerio de Ciencia e Innovación, Spain, expression of the Na+/phosphate Pho89 transporter and Ena1 Na+- grant number PID2020-113319RB-I00, to JA and AC. AZ is a recipient of PIF ATPase allows their functional coupling under high-pH stress. Mol Cell Biol. fellowship from the Universitat Autònoma de Barcelona, Spain. The funding 2014;34:4420–35. bodies had no role in the design of the study and collection, the analysis, and 17. Karbalaei M, Rezaee SA, Farsiani H. Pichia pastoris: a highly successful expres- interpretation of data, or the writing of the manuscript. sion system for optimal synthesis of heterologous proteins. J Cell Physiol. 2020;235:5867–81. Data Availability 18. Pan Y, Yang J, Wu J, Yang L, Fang H. Current advances of Pichia pastoris as The datasets generated and analyzed during the current study concerning cell factories for production of recombinant proteins. Front Microbiol. 2022. RNA-seq are available at NCBI’s Gene Expression Omnibus (GEO) and can https://doi.org/10.3389/fmicb.2022.1059777. be retrieved under study GSE224427. Any other data are available from the 19. Hartner FS, Glieder A. Regulation of methanol utilisation pathway genes in corresponding author on reasonable request. yeasts. Microb Cell Fact. 2006. https://doi.org/10.1186/1475-2859-5-39. 20. Stadlmayr G, Mecklenbräuker A, Rothmüller M, Maurer M, Sauer M, Mat- tanovich D, Gasser B. Identification and characterisation of novel Pichia Declarations pastoris promoters for heterologous protein production. J Biotechnol. 2010;150:519–29. Ethics approval and consent to participate 21. Takagi S, Tsutsumi N, Terui Y, Kong XY, Yurimoto H, Sakai Y. Engineering the Not applicable. expression system for Komagataella phaffii (Pichia pastoris): an attempt to develop a methanol-free expression system. FEMS Yeast Res. 2019. https:// Consent for publication doi.org/10.1093/femsyr/foz059. Not applicable. 22. Bernauer L, Radkohl A, Lehmayer LGK, Emmerstorfer-Augustin A. Komaga- taella phaffii as emerging Model Organism in Fundamental Research. Front Competing Interest Microbiol. 2021. https://doi.org/10.3389/FMICB.2020.607028. The authors declare that they have no competing interests. 23. Zahrl RJ, Prielhofer R, Burgard J, Mattanovich D, Gasser B. Synthetic activation of yeast stress response improves secretion of recombinant proteins. N Received: 8 February 2023 / Accepted: 27 March 2023 Biotechnol. 2023;73:19–28. 24. Lin NX, He RZ, Xu Y, Yu XW. Oxidative stress tolerance contributes to heterolo- gous protein production in Pichia pastoris. Biotechnol Biofuels. 2021. https:// doi.org/10.1186/S13068-021-02013-W. 25. Vogl T, Glieder A. Regulation of Pichia pastoris promoters and its conse- quences for protein production. N Biotechnol. 2013;30:385–404. 26. Causton HC, Ren B, Koh SS, Harbison CT, Kanin E, Jennings EG, Lee TI, True HL, References Lander ES, Young RA. Remodeling of yeast genome expression in response to 1. Selvig K, Alspaugh JA. pH response pathways in Fungi: adapting to host- environmental changes. Mol Biol Cell. 2001;12:323–37. derived and environmental signals. Mycobiology. 2011;39:249–56. 27. Canadell D, García-Martínez J, Alepuz P, Pérez-Ortín JE, Ariño J. Impact of 2. Ariño J. Integrative responses to high pH stress in S. cerevisiae. OMICS. high pH stress on yeast gene expression: a comprehensive analysis of mRNA 2010;14:517–23. turnover during stress responses. Biochimica et Biophysica Acta (BBA) -. Gene 3. Maeda T. The signaling mechanism of ambient pH sensing and adaptation in Regul Mech. 2015;1849:653–64. yeast and fungi. FEBSJ. 2012;279:1407–13. 4. Serra-Cardona A, Canadell D, Ariño J. Coordinate responses to alkaline pH stress in budding yeast. Microb Cell. 2015;2:182–96. Albacar et al. Microbial Cell Factories (2023) 22:63 Page 17 of 17 28. Lamb TM, Xu W, Diamond A, Mitchell AP. Alkaline response genes of Sac- 46. Lage P, Sampaio-Marques B, Ludovico P, Mira NP, Mendes-Ferreira A. Tran- charomyces cerevisiae and their relationship to the RIM101 pathway. J Biol scriptomic and chemogenomic analyses unveil the essential role of Com2- Chem. 2001;276:1850–6. regulon in response and tolerance of Saccharomyces cerevisiae to stress 29. Bensen ES, Martin SJ, Li M, Berman J, Davis DA. Transcriptional profiling in induced by sulfur dioxide. Microb Cell. 2019;6:509–23. Candida albicans reveals new adaptive responses to extracellular pH and 47. Barreto L, Canadell D, Valverde-Saubí D, Casamayor A, Ariño J. The short-term functions for Rim101p. Mol Microbiol. 2004;54:1335–51. response of yeast to potassium starvation. Environ Microbiol. 2012. https:// 30. Picazo I, Etxebeste O, Requena E, Garzia A, Espeso EA. Defining the transcrip- doi.org/10.1111/j.1462-2920.2012.02887.x. tional responses of aspergillus nidulans to cation/alkaline pH stress and the 48. Cheng Y, Du Z, Zhu H, Guo X, He X. Protective Effects of Arginine on Sac- role of the transcription factor SltA. Microb Genom. 2020;6:1–18. charomyces cerevisiae against ethanol stress. Sci Rep. 2016. https://doi. 31. Trushina N, Levin M, Mukherjee PK, Horwitz BA. PacC and pH-dependent org/10.1038/SREP31311. transcriptome of the mycotrophic fungus Trichoderma virens. BMC Genom- 49. Perli T, Wronska AK, Ortiz-Merino RA, Pronk JT, Daran JM. Vitamin require- ics. 2013. https://doi.org/10.1186/1471-2164-14-138. ments and biosynthesis in Saccharomyces cerevisiae. Yeast. 2020;37:283–304. 32. Posas F, Chambers JR, Heyman JA, Hoeffler JP, de Nadal E, Arino J. The tran- 50. Kowalska E, Kujda M, Wolak N, Kozik A. Altered expression and activities scriptional response of yeast to saline stress. J Biol Chem. 2000;275:17249–55. of enzymes involved in thiamine diphosphate biosynthesis in Saccha- 33. Martínez-Montañés F, Pascual-Ahuir A, Proft M. Toward a genomic view of romyces cerevisiae under oxidative and osmotic stress. FEMS Yeast Res. the gene expression program regulated by osmostress in yeast. OMICS. 2012;12:534–46. 2010;14:619–27. 51. Ahn J, Hong J, Park M, Lee H, Lee E, Kim C, Lee J, Choi ES, Jung JK, Lee H. 34. Gutin J, Joseph-Strauss D, Sadeh A, Shalom E, Friedman N. Genetic screen of Phosphate-responsive promoter of a Pichia pastoris sodium phosphate the yeast environmental stress response dynamics uncovers distinct regula- symporter. Appl Environ Microbiol. 2009;75:3528–34. tory phases. Mol Syst Biol. 2019;15:8939. 52. Albacar M, Velázquez D, Casamayor A, Ariño J. The toxic effects of yeast Ppz1 35. Serrano R, Bernal D, Simón E, Ariño J. Copper and Iron are the limiting factors phosphatase are counteracted by subcellular relocalization mediated by its for growth of the yeast Saccharomyces cerevisiae in an Alkaline Environment. regulatory subunit Hal3. FEBS Lett. 2022;596:1556–66. J Biol Chem. 2004. https://doi.org/10.1074/jbc.M313746200. 53. Green MR, Sambrook J. Molecular cloning: a laboratory Manual. 4th ed. New 36. Higuchi Y, Mori H, Kubota T, Takegawa K. Analysis of ambient pH stress York: Cold Spring Harbor Laboratory Press, NY; 2012. response mediated by iron and copper intake in Schizosaccharomyces 54. Lin-Cereghino J, Wong WW, Xiong S, Giang W, Luong LT, Vu J, Johnson SD, pombe. J Biosci Bioeng. 2018;125:92–6. Lin-Cereghino GP. Condensed protocol for competent cell preparation and 37. Ramos-Alonso L, Romero AM, Martínez-Pastor MT, Puig S. Iron Regulatory transformation of the methylotrophic yeast Pichia pastoris. Biotechniques. Mechanisms in Saccharomyces cerevisiae. Front Microbiol. 2020. https://doi. 2005;38:44–8. org/10.3389/FMICB.2020.582830. 55. Yang J, Nie L, Chen B, Liu Y, Kong Y, Wang H, Diao L. Hygromycin-resistance 38. Kosman DJ. Molecular mechanisms of iron uptake in fungi. Mol Microbiol. vectors for gene expression in Pichia pastoris. Yeast. 2014;31:115–25. 2003;47:1185–97. 56. Thomas-Chollier M, Defrance M, Medina-Rivera A, Sand O, Herrmann C, van 39. Castells-Roca L, Mühlenhoff U, Lill R, Herrero E, Bellí G. The oxidative stress Thieffry D HJ. RSAT 2011: regulatory sequence analysis tools. Nucleic Acids response in yeast cells involves changes in the stability of Aft1 regulon Res. 2011;39:W86–W91. mRNAs. Mol Microbiol. 2011;81:232–48. 57. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat 40. Castells-Roca L, Pijuan J, Ferrezuelo F, Bellí G, Herrero E. Cth2 protein mediates Methods. 2012;9:357–9. early adaptation of yeast cells to oxidative stress conditions. PLoS ONE. 2016. 58. Love MI, Huber W, Anders S. Moderated estimation of fold change and https://doi.org/10.1371/JOURNAL.PONE.0148204. dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550. 41. Batista-Nascimento L, Toledano MB, Thiele DJ, Rodrigues-Pousada C. Yeast 59. Cabiscol E, Belli G, Tamarit J, Echave P, Herrero E, Ros J. Mitochondrial Hsp60, protective response to arsenate involves the repression of the high affinity resistance to oxidative stress, and the labile iron pool are closely connected iron uptake system. Biochim Biophys Acta. 2013;1833:997–1005. in Saccharomyces cerevisiae. J Biol Chem. 2002;277:44531–8. 42. Sánchez NS, Calahorra M, Ramírez J, Peña A. Salinity and high pH affect 60. de Hoon MJ, Imoto S, Nolan J, Miyano S. Open source clustering software. energy pathways and growth in Debaryomyces hansenii. Fungal Biol. Bioinformatics. 2004;20:1453–4. 2018;122:977–90. 61. Saldanha AJ. Java Treeview–extensible visualization of microarray data. 43. Pan D, Wiedemann N, Kammerer B. (2019) Heat Stress-Induced Meta- Bioinformatics. 2004;20:3246–8. bolic Remodeling in Saccharomyces cerevisiae. Metabolites. https://doi. org/10.3390/METABO9110266 44. Oromendia AB, Dodgson SE, Amon A. Aneuploidy causes proteotoxic stress Publisher’s Note in yeast. Genes Dev. 2012;26:2696–708. Springer Nature remains neutral with regard to jurisdictional claims in 45. Menezes RA, Amaral C, Batista-Nascimento L, Santos C, Ferreira RB, Devaux F, published maps and institutional affiliations. Eleutherio ECA, Rodrigues-Pousada C. Contribution of Yap1 towards Saccha- romyces cerevisiae adaptation to arsenic-mediated oxidative stress. Biochem J. 2008;414:301–11.