Metabolomic Analysis of Response to Nitrogen-Limiting Conditions in Yarrowia Spp

H OH metabolites OH

Article Metabolomic Analysis of Response to Nitrogen-Limiting Conditions in Yarrowia spp.

Sivamoke Dissook, Sastia Prama Putri * and Eiichiro Fukusaki

Department of Biotechnology, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka 565-0871, Japan; [email protected] (S.D.); [email protected] (E.F.) * Correspondence: [email protected]; Tel.: +81-6-6879-7416

Abstract: Yarrowia is a yeast genus that has been used as a model oleaginous taxon for a wide array of studies. However, information regarding metabolite changes within Yarrowia spp. under different environmental conditions is still limited. Among various factors affecting Yarrowia metabolism, nitrogen-limiting conditions have a profound effect on the metabolic state of yeast. In this study, a time-course LC-MS/MS-based metabolome analysis of Y. lipolytica was performed to determine the optimal cultivation time and carbon-to-nitrogen ratio for studying the effects of nitrogen-limiting conditions on Yarrowia; we found that cultivation time of 36 h and carbon-to-nitrogen ratio of 4:1 and 5:0 was suitable for studying the effects of nitrogen-limiting conditions on Yarrowia and these conditions were applied to six strains of Yarrowia. These six strains of Yarrowia showed similar responses to nitrogen-limiting conditions; however, each strain had a unique metabolomic profile. Purine and pyrimidine metabolism were the most highly affected biological pathways in nitrogen- limiting conditions, indicating that these conditions affect energy availability within cells. This stress leads to a shift in cells to the utilization of a less ATP-dependent biological pathway. This information will be beneficial for the development of Yarrowia strains for further scientific and industrial applications.

Keywords: Yarrowia; LC-MS/MS; metabolome

Citation: Dissook, S.; Putri, S.P.; Fukusaki, E. Metabolomic Analysis of Response to Nitrogen-Limiting 1. Introduction Conditions in Yarrowia spp. Yarrowia is a fungal genus in the family Dipodascaceae. This genus was initially Metabolites 2021, 11, 16. https://doi. considered monotypic, containing only Yarrowia lipolytica, which has been used as a model org/10.3390/metabo11010016 oleaginous organism for a wide range of studies, including studies of lipid accumulation and degradation, peroxisome biogenesis, dimorphism, and protein secretion pathways [1]. Received: 10 November 2020 Wild-type Y. lipolytica can utilize glucose, fructose, glycerol, and hydrophobic substrates as Accepted: 24 December 2020 carbon sources [2,3]. Owing to this property, the yeast species has potential applications in Published: 29 December 2020 industrial microbiology, particularly for the biosynthesis of lipids [4]. Based on molecular phylogenetics studies, the genus now includes several additional species [5–7]. Publisher’s Note: MDPI stays neu- Altering the carbon-to-nitrogen (C:N) ratio in the culture medium could drastically tral with regard to jurisdictional claims in published maps and institutional change the metabolomic profile in yeasts, and the precise effects may vary among different affiliations. strains [8–11]. Substantial work over the past few years has focused on characterizing and engineering oleaginous yeast [12]. Nonetheless, the metabolic response of Yarrowia spp., especially the recently added species, under nitrogen-limiting conditions is not well char- acterized. The nitrogen-limiting condition is a metabolic condition. Therefore, a complete Copyright: © 2020 by the authors. Li- understanding of its effects cannot be achieved without studying intracellular metabo- censee MDPI, Basel, Switzerland. This lites [13]. article is an open access article distributed In this study we conducted the time-course sampling of one Y. lipolytica strain to probe under the terms and conditions of the the suitable time points for multiple strains study. Since the variable changes in this study Creative Commons Attribution (CC BY) highly affect the growth of Y. lipolytica, it might not be suitable to choose the sampling license (https://creativecommons.org/ point based on the growth phase alone. Therefore, we considered the metabolome profile licenses/by/4.0/).

Metabolites 2021, 11, 16. https://doi.org/10.3390/metabo11010016 https://www.mdpi.com/journal/metabolites Metabolites 2021, 11, x FOR PEER REVIEW 2 of 15

study highly affect the growth of Y. lipolytica, it might not be suitable to choose the sam- Metabolites 2021, 11, 16 pling point based on the growth phase alone. Therefore, we considered the metabolome2 of 14 profile to help understand the dynamic of metabolic changes occuring during the cultivation with several different C:N ratios and determined the crucial sampling time point and C:Nto help ratio understand for multiple the strains dynamic study. of We metabolic further studied changes the occuring metabolic during responses the cultivation of several withYarrowia several strains different to nitrogen C:N ratios-limiting and conditions determined, including the crucial one samplinglaboratory time strain point, Y. lipo- and lyticaC:N ratio PO1d for [14] multiple, three strainsformer study. Candida We lipolytica further studiedspecies thefrom metabolic different responses study groups of several (cur- Yarrowiarently identifiedstrains to as nitrogen-limiting Yarrowia spp.), C conditions,. lipolytica JCM including 2304 one[15], laboratory C. lipolytica strain, JCM Y.21924 lipolytica [16], andPO1d C. [lipolytica14], three JCM former 8061Candida, and two lipolytica recentlyspecies identified from species, different Y. studydeformans groups JCM (currently 1694 [17] andidentified Y. keelungensis as Yarrowia JCMspp.), 14894C. [1 lipolytica8]. This informationJCM 2304 [will15], beC. beneficial lipolytica JCM for the 21924 development [16], and C.and lipolytica manipulationJCM 8061, of Yarrowia and two sp recentlyp. for fu identifiedture applications. species, Y. deformans JCM 1694 [17] and Y. keelungensis JCM 14894 [18]. This information will be beneficial for the development and 2.manipulation Results and of DiscussionYarrowia spp. for future applications. 2.1. Growth of Y. Lipolytica under Different Carbon-to-Nitrogen Ratios 2. Results and Discussion To determine the crucial sampling time point and C:N ratio for analyses of the effect 2.1. Growth of Y. Lipolytica under Different Carbon-to-Nitrogen Ratios of nitrogen-limiting conditions, time course sampling under various C:N compositions was Toperformed determine using the crucialY. lipolytica sampling PO1d. time The point extracellular and C:N ratio glucose for analyses concentrations of the effect and growthof nitrogen-limiting of Y. lipolytica conditions, cultivated time with course different sampling C:N underratios variousare shown C:N in compositions Figure 1. There was performedwere three usingdifferentY. lipolytica stable physiologicalPO1d. The extracellular states: (1) C glucosearbon-restri concentrationscted conditions and growth (0:5), (2) of Y.intermediate lipolytica cultivated C:N ratios with (4;1, different 2:2, and C:N1:4), ratiosand (3) are nitrogen shown- inrestricted Figure1 .condition There weres (5:0) three; all statisticaldifferent stable test results physiological can be found states: in (1)Table Carbon-restricted S1. Under carbon conditions-restricted (0:5), conditions, (2) intermediate growth increasedC:N ratios steadily (4;1, 2:2, from and 0 1:4), h to and 12 h (3), remained nitrogen-restricted constant for conditions 24 h, and cells (5:0); reached all statistical the high- test results can be found in Table S1. Under carbon-restricted conditions, growth increased est OD600 at 36 h. For intermediate C:N ratios, growth increased rapidly from 0 h to 12 h, steadily from 0 h to 12 h, remained constant for 24 h, and cells reached the highest OD600 slowed from 12 h to 24 h, and the highest OD600 was reached at 36 h. Under nitrogen- at 36 h. For intermediate C:N ratios, growth increased rapidly from 0 h to 12 h, slowed restricted conditions, growth increased slowly from 0 h to 24 h and increased steadily from 12 h to 24 h, and the highest OD600 was reached at 36 h. Under nitrogen-restricted from 24 h until cells reached the highest OD600 at 36 h. In this study, all OD600 values de- conditions, growth increased slowly from 0 h to 24 h and increased steadily from 24 h creased after 36 h of cultivation. In both carbon- and nitrogen-restricted conditions, the until cells reached the highest OD at 36 h. In this study, all OD values decreased growth profiles suggested that cells600 took approximately 24 h to adjust600 to the new condi- after 36 h of cultivation. In both carbon- and nitrogen-restricted conditions, the growth tion before entering the exponential growth phase. For intermediate C:N ratios, especially profiles suggested that cells took approximately 24 h to adjust to the new condition before 4:1 and 2:2, three phases of cell growth were observed, 0–12 h, 12–24 h, and 24–36 h. These entering the exponential growth phase. For intermediate C:N ratios, especially 4:1 and data suggest that the conditions changed during cultivation, and cells had to adapt for 2:2, three phases of cell growth were observed, 0–12 h, 12–24 h, and 24–36 h. These data increasedsuggest that growth. the conditions Interestingly, changed this during phenomen cultivation,on was andnot observed cells had tofor adapt a C:N for ratio increased of 1:4. Thgrowth.ese results Interestingly, show that this Y. lipolytica phenomenon is very was sensitive not observed to changes for ain C:N the ratioC:N ratio. of 1:4. Surpris- These ingly,results Y. show lipolytica that growthY. lipolytica did notis very depend sensitive on glucose to changes availability in the, as C:N evidenced ratio. Surprisingly, by the con- Y.tinued lipolytica growthgrowth in glucose did not-depleted depend onconditions glucose availability,(C:N 0:5 and as 1:4 evidenced at 24 h, 2:2 by at the 36 continued h). Yeast ingrowth the C:N in glucose-depleted5:0 condition could conditions not utilize (C:N glucose 0:5 and effectively 1:4 at 24 h,, even 2:2 atwith 36 h). abundant Yeast in glucose the C:N. Moreover,5:0 condition Y. lipolytica could not entered utilize glucosethe death effectively, phase, regardless even with of abundantglucose availability glucose. Moreover, (C:N 4:1, Y.5:0). lipolytica entered the death phase, regardless of glucose availability (C:N 4:1, 5:0).

Figure 1. Growth curves of Y. lipolytica PO1d and glucose consumption for different carbon-to-nitrogen ratios. Y. lipolytica

cultures at a starting an OD600 of 0.5 for various carbon-to-nitrogen ratios. Error bars represent the standard deviation from three replicates; 5:0 (diamonds with a dashed line), 4:1 (squares with a dark line), 2:2 (triangle), 1:4 (double crosses with a dashed line), 0:5 (asterisks with a dark line). (A) Growth curve determined by optical density at 600 nm. (B) Glucose (g/L). Metabolites 2021, 11, x FOR PEER REVIEW 3 of 15

Figure 1. Growth curves of Y. lipolytica PO1d and glucose consumption for different carbon-to-nitrogen ratios. Y. lipolytica cultures at a starting an OD600 of 0.5 for various carbon-to-nitrogen ratios. Error bars represent the standard deviation from Metabolitesthree replicates;2021, 11, 16 5:0 (diamonds with a dashed line), 4:1 (squares with a dark line), 2:2 (triangle), 1:4 (double crosses with a 3 of 14 dashed line), 0:5 (asterisks with a dark line). (A) Growth curve determined by optical density at 600 nm. (B) Glucose (g/L).

2.2. Intracellular Metabolome Analysis of Y. Lipolytica for Different Carbon-to-Nitrogen Ratios 2.2. Intracellular Metabolome Analysis of Y. Lipolytica for Different Carbon-to-Nitrogen Ratios A total of 93 metabolites were annotated (Table S2). The time-course results for all A total of 93 metabolites were annotated (Table S2). The time-course results for all detected metabolites are summarized in Figure S1. Annotated compounds mainly con- detected metabolites are summarized in Figure S1. Annotated compounds mainly consisted sisted of organic acids, sugars, amino acids, phosphate compounds, and other metabo- of organic acids, sugars, amino acids, phosphate compounds, and other metabolites. Widely lites. Widely targeted LC-MS/MS data were analyzed by a principal component analysis targeted LC-MS/MS data were analyzed by a principal component analysis (PCA) to (PCA) to differentiate among cultivation conditions. The PCA score plot offers a visual differentiate among cultivation conditions. The PCA score plot offers a visual image of image of sample variation from a global view, and, as a non-parametric analysis, the cre- sample variation from a global view, and, as a non-parametric analysis, the created model atedis model independent is independent of the user; of therefore,the user; therefore, it is unsupervised it is unsupervised [19–21]. [19–21]. As Asshown shown in inFigure Figure 22, ,when when all all samples samples at allall timetime pointspoints were were evaluated evaluated by by PCA, PCA, two twomain main clusters clusters were were separated separated along along PC1. PC1. Cluster Cluster oneone includedincluded samplessamples withwith C:NC:N ratiosra- tiosof of 0:5, 0:5, 1:4, 1:4, 2:2, 2:2, and and 4:1 4:1 at at 12 12 h toh to 36 36 h ofh of cultivation. cultivation. Cluster Cluster two two separated separate samplesd samples into intotwo two sub-clusters; sub-clusters sub-cluster; sub-cluster one one included included samples samples with with a C:Na C:N ratio ratio of of 5:0 5:0 at at 12 12 h toh to 48 h, 48 hand, and sub-cluster sub-cluster two two included included samples samplesobtained obtained underunder a C:N C:N ratio ratio of of 4:1 4:1 at at 48 48 h. h.These These resultsresults showed showed that that the the most most important important factor factor separat separatinging the the samples samples was was the the nitrogen nitrogen-- limitinglimiting condition condition;; the the C:N C:N ratio ratio of of4:1 4:1 at at 48 48 h hclustered clustered with with the the C:N C:N ratio ratio of of 5:0 5:0,, suggest- suggesting ingthat that at at48 48 hh ofof cultivation,cultivation, the the nitrogen nitrogen source source with with a aC:N C:N ratio ratio of of 4:1 4:1 was was depleted. depleted. To To comprehensivelycomprehensively analyze analyze the the data, data, we we evaluated evaluated each each time pointpoint separately,separately, as as summarized summa- rizedin Figurein Figure3. 3.

Figure 2. Summary of principal component analysis (PCA) of Y. lipolytica cultivated under different Figure 2. Summary of principal component analysis (PCA) of Y. lipolytica cultivated under differ- carbon-to-nitrogen ratios. The PCA score plot indicates differences in metabolite proﬁles for different ent carbon-to-nitrogen ratios. The PCA score plot indicates differences in metabolite profiles for differentcarbon-to-nitrogen carbon-to-nitrogen ratios basedratios onbased 93 metabolites. on 93 metabolites. The grey The ellipse greyindicates ellipse indicates a 95% conﬁdence a 95% confi- border dencebased border on Hotelling’s based on Hotelling’s T2. Each pointT2. Each represents point represents one replicate one replicate for each C:Nfor each ratio; C:N 0:5 ratio; (grey 0:5 circles), (grey1:4 circ (greyles), squares), 1:4 (grey 2:2 squares), (grey triangles), 2:2 (grey 4:1 triangles), (grey stars), 4:1 (grey 5:0 (white stars), circles). 5:0 (white The circles). yellow ellipse The yellow indicates ellipsethe samplesindicates fromthe samples C:N 1:4 from−4:1 C at:N 12–36 1:4−4:1 h, Theat 12−36 cyan h, ellipse The cyan indicates ellipse the indicates samples the from samples C:N 5:0 at from12–48 C:N h, 5:0 The at purple12−48 h, ellipse The purple indicates ellipse the samplesindicates from the sa C:Nmples 4:1 atfrom 48 h.C: TheN 4:1 complete at 48 h. The numerical com- PCA pleteloading numerical scores PCA are loading presented scores in Table are presented S3. in Table S3.

Metabolites 2021, 11, 16 4 of 14 Metabolites 2021, 11, x FOR PEER REVIEW 4 of 15

Figure 3. PCA results for Y. lipolytica cultivated in different carbon-to-nitrogen ratios. PCA score plot indicating differences Figure 3. PCA results for Y. lipolytica cultivated in different carbon-to-nitrogen ratios. PCA score plot indicating differences in metabolite profiles for different carbon-to-nitrogen ratios based on 93 metabolites. The grey ellipse indicates a 95% in metabolite profiles for different carbon-to-nitrogen ratios based on 93 metabolites. The grey ellipse indicates a 95% confidenceconfidence border border based based on on Hotelling’s Hotelling’s T2.T2. Each point point represents represents one one replicate replicate in each in each condition; condition; 0:5 (grey 0:5 circle), (grey circle),1:4 (grey 1:4 (grey square),square), 2:2 (grey2:2 (grey triangle), triangle), 4:1 4:1 (grey (grey star), star), 5:0 5:0 (white(white circle). (Scale: (Scale: Auto Auto-scale;-scale; n =n 3);= ( 3);A) (profileA) profile at 12 at h, 12 (B h,) profile (B) profile at 24 at 24 h, (C) profileh, (C) profile at 36 at h, 36 and h, and (D) ( atD) profile at profile at at 48 48 h. h. The The cyancyan ellipse indicates indicates the the nitrogen nitrogen-limited-limited samples, samples, the purple the purpleellipse ellipse indicates the carbon-limited samples, the yellow and blue ellipse indicates samples in which neither carbon nor nitrogen indicates the carbon-limited samples, the yellow and blue ellipse indicates samples in which neither carbon nor nitrogen were limited. The full list of PCA loading scores is presented in Table S4. were limited. The full list of PCA loading scores is presented in Table S4. As shown in Figure 3A, there were three distinct clusters at 12 h of cultivation, cor- respondingAs shown to the in nitrogen Figure3-A,restricted therewere condition, three the distinct carbon clusters-restricted at condition 12 h of cultivation,, and the cor- respondingcondition where to the both nitrogen-restricted carbon and nitrogen condition, were available the carbon-restricted. PC1 separated the condition, nitrogen- and the conditionlimiting condition where both and other carbon conditions, and nitrogen while PC2 were separated available. the PC1carbon separated-limiting condi- the nitrogen- limitingtion and condition other conditions. and other Most conditions, of the metabolites while PC2 with separated high PCA thescores carbon-limiting (top 10 scores on condition andthe otherpositive conditions. and negative Most sides of of thethe metabolitesPC axis) at 12with h were high amino PCA acids scores involved (top in 10 the scores on theaminoacyl positive-tRNA and negativebiosynthesis sides pathway. of the Aminoacyl PC axis) at-tRNAs 12 h wereare substrates amino acids for translation involved in the aminoacyl-tRNAand determine the biosynthesisinterpretation pathway.of the genetic Aminoacyl-tRNAs code [22,23]. This might are substrates indicate that for cells translation andin both determine nitrogen the- and interpretation carbon-limiting of theconditions genetic adjust code [to22 the,23 ].new This environment might indicate at this that cells intime both point nitrogen-. At 24 andh of carbon-limitingcultivation (Figure conditions 3B), the clustering adjust to results the new were environment similar to those at this time at 12 h, except the C:N 1:4 profile was more similar to that under the carbon-restricted point. At 24 h of cultivation (Figure3B), the clustering results were similar to those at 12 h, condition in relation to glucose consumption (Figure 1B). This probably reflects glucose except the C:N 1:4 profile was more similar to that under the carbon-restricted condition depletion in the medium, leading to adjustments in cell metabolism to the carbon-re- instricted relation condition to glucose, similar consumption to 12 h. PC1 at (Figure 24 h separated1B). This samples probably in nitrogen reflects-limiting glucose con- depletion indition the medium,s and other leading conditions, to adjustments while PC2 separate in cell metabolismd the carbon to-limiting the carbon-restricted condition from condition,other similar conditions. to 12 However, h. PC1 atthe 24 metabolites h separated with samples a high PCA in nitrogen-limitingscore in PC1 are made conditions up of and other conditions, while PC2 separated the carbon-limiting condition from other conditions. However, the metabolites with a high PCA score in PC1 are made up of sugar-phosphate groups from purine and pyrimidine metabolism. These results suggested that purine and pyrimidine metabolism are essential for the response to nitrogen-limiting conditions. At 36 h of cultivation (Figure3C), the PCA score plot clearly shows that the C:N 1:4 profile merged with the C:N 0:5 profile, indicating that at this time point, the metabolomic profile Metabolites 2021, 11, 16 5 of 14

for the C:N 1:4 condition completely shifted to that of the carbon-restricted condition. The C:N 2:2 and 4:1 profiles clustered together; presumably, the metabolomic profile at this point represents the condition in which carbon and nitrogen are available. Considering the remaining glucose at this time, in the C:N 4:1 condition (Figure1B), approximately 10 g/L glucose still remains, while glucose in C:N 2:2 is depleted; this might explain why one of the samples in C:N 2:2 started to shift toward the profile of the carbon-restricted condition. Finally, at 48 h of cultivation (Figure3D), there were four distinct clusters: C:N 4:1, C:N 2:2, C:N 5:0, and C:N 1:4, 0:5 clusters. At this time point, the growth profile (Figure1A ) suggested that cells entered the death phase, and all glucose was depleted except in C:N 5:0 and 4:1. Interestingly, the C:N 2:2 group did not cluster with the carbon-restricted group, suggesting that metabolism during the death phase is different, even in the carbon- or nitrogen-restricted condition. Metabolites with a high PCA score at 36 h and 48 h were similar to those at 24 h, where PC1 consisted mostly of sugar-phosphate compounds, and PC2 consisted of metabolites from amino acid and central carbon metabolism. Purine and pyrimidine are involved in the synthesis of thiamine, riboflavin, folic acid pteridines, and histidine or are constituents of cytokinins, purine alkaloids, and other unusual nitrogen compounds related to nitrogen accumulation or nitrogen excretion [24,25]. As purine and pyrimidine are nitrogen sources stored in cells, the lack of nitrogen supply in the medium might directly or indirectly affect purine and pyrimidine biosynthesis in yeast. Moreover, purine and pyrimidine and their derivatives are linked to the synthesis, transfer, and utilization of stored energy in energy metabolism [26]. The principal component loading scores on PC1 and PC2 at each time point was converted to an absolute value to rank their contribution to the separation of the data on the PC scatter plot. Metabolites with a PC score in the upper 75th percentile for the same cultivation time and principal component were selected for Venn diagram construction, as shown in Figure4. We found ten core metabolites at all cultivation time points, i.e., ADP, arginine, F6P, lysine, trehalose, IMP, BPG, FAD, PEP, and NADP. These metabolites with a high PCA score throughout the cultivation period have a wide range of fluctuation with respect to C:N ratios in every stage of growth. Trehalose is a non-reducing disaccharide found ubiquitously in fungi and is widely found in bacteria and animals [27]. Trehalose was initially thought of as a reserve carbohydrate in yeast. However, its role as a stress protectant in yeast has been recognized [28]. The primary role of trehalose is to protect the cytosol against adverse conditions, such as desiccation, frost, and heat. In the budding yeast Saccharomyces cerevisiae, trehalose levels are low during rapid growth periods and increase during periods of slow or no growth [29]. These results agree with our results for Y. lipolytica (Figure S2), except in the condition where carbon or nitrogen was restricted. As discussed above, purine and pyrimidine metabolism were affected. Notably, IMP and ADP were among the metabolites with high PCA scores at every time point. It is possible that these metabolites were at the center of the biological reaction in purine metabolism. Therefore, the effect of changes in conditions could be observed in these metabolites. Consequently, FAD, which requires GTP from the purine biosynthesis pathway, was also affected. For arginine, carbamoyl phosphate, an upstream metabolite for arginine biosynthesis, was directly involved in pyrimidine metabolism and nitrogen metabolism. The high PCA score for arginine might reflect the effect of the upstream pathway. PEP, F6P, and BPG are essential intermediates in the TCA cycle, suggesting that glycolysis/gluconeogenesis or related metabolites in the pathway were utilized differently under different C:N ratios; consequently, lysine, which is linked to central metabolism, was also affected. Finally, NADP is a cofactor used in anabolic reactions, such as the Calvin cycle and lipid and nucleic acid syntheses, which require NADPH as a reducing agent. As the carbon and nitrogen sources were interrupted, the NADP level changed drastically. The time course of relative intensities is shown in Figure S2, all t-test p-value could be found in Table S5. The exact mechanisms of how the nitrogen-limiting condition affects the glucose uptake in Y. lipolytica was unclear but we hypothesized that the lower energy molecule might cause the ATP-dependent pathway in Y. lipolytica related to carbon metabolisms to slow down. Metabolites 2021, 11, x FOR PEER REVIEW 6 of 15

nitrogen sources were interrupted, the NADP level changed drastically. The time course of relative intensities is shown in Figure S2, all t-test p-value could be found in Table S5. The exact mechanisms of how the nitrogen-limiting condition affects the glucose uptake Metabolites 2021, 11, 16 in Y. lipolytica was unclear but we hypothesized that the lower energy molecule might 6 of 14 cause the ATP-dependent pathway in Y. lipolytica related to carbon metabolisms to slow down.

FigureFigure 4. Venn 4. diagramVenn diagram of selected of metabolites selected based metabolites on principal based component on principal loading scores. component The loading scores. The metabolites with high scores from PCA 1 and 2 at each cultivation time were included in the anal- metabolites with high scores from PCA 1 and 2 at each cultivation time were included in the analysis. ysis. A list of metabolites in each subset is provided in Table S6. A list of metabolites in each subset is provided in Table S6. 2.3. Intracellular Metabolome Analysis of Different Yarrowia spp. in Normal and Nitrogen-Lim- iting2.3. Conditions Intracellular Metabolome Analysis of Different Yarrowia spp. in Normal and Nitrogen-LimitingThe time-course analysis Conditions of the Y. lipolytica PO1d metabolomics profile provides in- formationThe regarding time-course metabolic dynamics analysis with of therespectY. to lipolytica different C:NPO1d ratios metabolomics and time to profile provides studyinformation the response regardings of Yarrowia metabolicstrains to nitrogen dynamics-limiting with condition respects. Based to different on growth, C:N ratios and time to glucose consumption, and metabolomic profile results, we concluded that the appropriate timestudy point for the sample responses collection of Yarrowiafor this purposestrains was to 36 nitrogen-limiting h, as the metabolomic conditions. profile of Based on growth, C:Nglucose 4:1 was superimposed consumption, with and that metabolomic for C:N 5:0 at 48 profile h (Figure results, S3), suggesting we concluded that the that the appropriate nitrogentime source point at for 48 h sample for C:N collection4:1 was depleted for. this In addition, purpose the wasgrowth 36 of h, C:N as 5:0 the at metabolomic 24 profile of h wasC:N limited 4:1.was To study superimposed the response to with nitrogen that-limiting for C:N conditions 5:0 at, a 48 standard h (Figure C:N ratio S3), suggesting that the of 4:1 was selected for comparison against the nitrogen-limiting condition of 5:0 as the nitrogen source at 48 h for C:N 4:1 was depleted. In addition, the growth of C:N 5:0 at 24 h C:N 5:0 gave the distinct metabolome profile throughout the cultivation. The growth pro- fileswas of different limited. strains To study confirmed the that response Yarrowia to sp nitrogen-limitingp. have similar standard conditions, growth pro- a standard C:N ratio of files4:1 (Figure was S4) selected. In this forstudy comparison, we investigated against the responses the nitrogen-limiting of six Yarrowia strains condition, Y. of 5:0 as the C:N lipolytica5:0 gave PO1d the, Y. deformans distinct JCM metabolome 1694, Y. keelunge profilensis throughoutJCM 14894, C. lipolytica the cultivation. JCM 2304, The growth profiles C. lipolyticaof different JCM 21924, strains and confirmedC. lipolytica JCM that 8061.Yarrowia The completespp. numerical have similar results standard of all growth profiles detected normalized peak areas of six Yarrowia spp. in normal and nitrogen-limiting condition(Figures are presented S4). In in this Table study, S7. we investigated the responses of six Yarrowia strains, Y. lipolytica PO1d,As shownY. deformansin Figure 5, PC1JCM separated 1694, Y. the keelungensis yeast cultivatedJCM under 14894, normalC. and lipolytica nitrogenJCM- 2304, C. lipolytica limitingJCM conditions 21924, and, whileC. PC2 lipolytica separatedJCM each strain 8061. of TheYarrowia. complete While all numerical of the yeast resultsin of all detected normalized peak areas of six Yarrowia spp. in normal and nitrogen-limiting conditions are

presented in Table S7. As shown in Figure5, PC1 separated the yeast cultivated under normal and nitrogen- limiting conditions, while PC2 separated each strain of Yarrowia. While all of the yeast in this study belonged to the genus Yarrowia, the results suggested that each Yarrowia strain has a unique metabolomic proﬁle, including variation at the subspecies level. The high PC score for PC1 agrees with the initial analysis of Y. lipolytica PO1d, in which most of the metabolites with high PC scores were from purine and pyrimidine metabolism, suggesting that Yarrowia spp. respond to nitrogen-limiting conditions in a similar way. We further analyzed differences among Yarrowia strains under nitrogen-limiting conditions. As shown in Figure6, PC1 separated Y. deformans JCM 1694 and the remaining taxa, while PC2 separated the following three groups: (1) Y. keelungensis, JCM 14894 (2) Y. lipolytica PO1d, C. lipolytica JCM 21924, and C. lipolytica JCM 8061, and (3) C. lipolytica JCM 2304. The high PC1 score suggested that Y. deformans JCM 1694 has a different response to nitrogen-limiting conditions than those of other Yarrowia spp. Interestingly, PC2 separated not only the different Yarrowia species but also the subspecies C. lipolytica JCM Metabolites 2021, 11, x FOR PEER REVIEW 7 of 15

Metabolites 2021, 11, 16 7 of 14 this study belonged to the genus Yarrowia, the results suggested that each Yarrowia strain has a unique metabolomic profile, including variation at the subspecies level. The high PC score for PC1 agrees with the initial analysis of Y. lipolytica PO1d, in which most of the metabolites2304; this with result high shows PC scores that Yarrowia were fromis purine a highly and diverse pyrimidine genus metabolism, even at the suggest- subspecies level. ingThe that results Yarrowia for sp allp.detected respond to metabolites nitrogen-limiting are summarized conditions in ina similar Figure way. S5.

FigureFigure 5. PCA 5. PCA results results for six for Yarrowia six Yarrowia spp. cultivatedspp. cultivated in normal inand normal nitrogen and-limiting nitrogen-limiting conditions. conditions. PCAPCA score score plot plot indicating indicating differences differences in metabolite in metabolite profiles profilesfor two carbon for two-to- carbon-to-nitrogennitrogen ratios ratios based basedon 100 on metabolites.100 metabolites. The The grey grey ellipse ellipse indicatesindicates a a 95% 95% confidence confidence border border based based on Hotelling on Hotelling’s’s T2. The T2. The purple ellipse indicates the nitrogen-limited samples. Each point represents one replicate; purple ellipse indicates the nitrogen-limited samples. Each point represents one replicate; Y. lipolytica Y. lipolytica PO1d in normal conditions (light grey triangles), Y. lipolytica PO1d in nitrogen-limiting conditionPO1d ins (light normal grey conditions four-point (lightstars), C. grey lipolytica triangles), JCM 8061Y. lipolytica in normalPO1d condition in nitrogen-limitings (white four- conditions point(light star greys), C. four-pointlipolytica JCM stars), 8061 inC. nitrogen lipolytica-limitingJCM 8061condition in normals (black circle conditionss), C. lipolytica (white JCM four-point stars), 2304C. lipolytica in normalJCM condition 8061sin (black nitrogen-limiting four-point stars conditions), C. lipolytica (black JCM 2304 circles), in nitrogenC. lipolytica-limitingJCM con- 2304 in normal ditions (dark grey circles), Y. lipolytica JCM 21924 in normal conditions (dark grey squares), Y. lipolyticaconditions JCM (black21924 in four-point nitrogen-limiting stars), conditionC. lipolyticas (darkJCM grey 2304 triangle in nitrogen-limitings), Y. deformans JCM conditions 1694 in (dark grey normalcircles), conditionY. lipolyticas (darkJCM grey 21924four-point in normal stars), Y. conditions deformans JCM (dark 1694 grey in nitrogen squares),-limitingY. lipolytica condi- JCM 21924 in tionnitrogen-limitings (yellow four-point conditions star), Y. keelungensis (dark grey JCM triangles), 14894 inY. normal deformans conditionJCMs 1694(green in five normal-point conditions (dark stargreys), Y. four-point keelungensis stars), JCM Y.14894 deformans in nitrogenJCM-limiting 1694 in condition nitrogen-limitings (blue circle conditionss). Right, top (yellow 25% of four-point star), metabolites based on PCA scores from each principal component. The full list of PCA loading scoresY. keelungensis is presentedJCM in Table 14894 S8. in normal conditions (green five-point stars), Y. keelungensis JCM 14894 in nitrogen-limiting conditions (blue circles). Right, top 25% of metabolites based on PCA scores from eachWe principal further analyzed component. differences The full among list of PCA Yarrowia loading strains scores under is presentednitrogen-limiting in Table con- S8. ditions. As shown in Figure 6, PC1 separated Y. deformans JCM 1694 and the remaining taxa, whileSurprisingly, PC2 separated the the metabolites following three from group PC2s include: (1) Y. keelungensis MEP and, JCM MEcPP; 14894 ( (Table2) Y. S8) these lipolyticametabolites PO1d, areC. lipolytica in the JCM MEP 21924, pathway, and C. whichlipolytica has JCM not 8061 been, andreported (3) C. lipolytica in this JCM yeast genus. 2304.However, The high Soliman PC1 score etal. suggested [30] detected that Y. adeformans MEP-like JCM pathway 1694 has in a fungi.different We response hypothesized that tothe nitrogen nitrogen-limiting-limiting conditions condition than those reduces of other purine Yarrowia and sp pyrimidinep. Interestingly, activity PC2 sepa- in yeast, which rated not only the different Yarrowia species but also the subspecies C. lipolytica JCM 2304; affects the energy availability in the cell. This stress leads to a cellular adjustment to the less ATP-dependent pathway, i.e., a switch from the MVA pathway to the MEP pathway, which requires less ATP. However, further biological investigations are required. Peak areas for all detected metabolites of purine and pyrimidine metabolism are summarized in Figures7 and8, respectively, all t-test p-value could be found in Table S9. These results clearly demonstrate that Yarrowia strains respond to the nitrogen-limiting condition in a similar way; notably, R5P, ADP, ATP, GDP, and GTP in purine metabolism were similar among strains. On the other hand, each species exhibited distinct metabolomic profiles for other metabolites in purine metabolism and all other detectable metabolites in pyrimidine metabolism. The laboratory Y. lipolytica PO1d strain had a distinct profile in Metabolites 2021, 11, 16 8 of 14

which the orotate, PRPP, IMP, CTP, CDP, and CMP levels were exceptionally high compared to those in other strains. In general, the levels of metabolites in purine and pyrimidine metabolism were lower under nitrogen-limiting conditions than under normal conditions; however, C. lipolytica JCM 8061 had an interesting pattern in which metabolite profiles in the nitrogen-limiting condition were reversed compared to those of other strains, including glutamine, adenine, and uridine. C. lipolytica JCM 2304 had a unique profile in which many metabolites in purine and pyrimidine metabolism were not affected by nitrogen-limiting conditions, including R5P, guanine, cAMP, xanthine, hypoxanthine, UTP, uridine, and PRPP. C. lipolytica JCM 21924 metabolites under nitrogen-limiting conditions, including adenine and uridine, showed the opposite patterns compared to those of other strains; however, unlike JCM 8061, glutamine was not affected. Y. deformans JCM 1694 had the highest glutamine content, and levels of all metabolites in purine and pyrimidine metabolism were lower in the nitrogen-limiting condition. Y. keelungensis JCM 14894 was the only strain in which metabolites from purine metabolism were mostly unaffected; however, metabolites Metabolites 2021, 11, x FOR PEER REVIEW 8 of 15 of pyrimidine metabolism were quite similar to those of other strains. The different profiles of Yarrowia spp. observed in this study can be utilized for metabolic manipulations for special applications in the future. Based on these results, to engineer microorganisms for a thisparticular result show application,s that Yarrowia it isis necessarya highly diverse to consider genus even the preciseat the subspecies metabolic level. traits The of the strain resultsfor optimization.for all detected Wemetabolites demonstrated are summarized that metabolomics in Figure S5. is a powerful tool for this purpose.

FigureFigure 6. PCA 6. PCA results results for six forYarrowia six Yarrowia spp. cultivatedspp. cultivated in normal inand normal nitrogen and-limiting nitrogen-limiting conditions; conditions; (left)(left) PCA PCA score score plot plotindicates indicates differences differences in metabolite in metabolite profiles proﬁles among strains among in strains the nitrogen in the- nitrogen-limiting limitingcondition condition based based on on 100 10 metabolites.0 metabolites. The grey grey ellipse ellipse indicates indicates a 95% a 95% confidence conﬁdence border border based on based on Hotelling’s T2. The cyan ellipse indicates samples from Y. keelungensis JCM 14894, the yellowHotelling’s ellipse indicates T2. The samples cyan ellipse from Y indicates. deformans samples JCM 1694, from theY. purple keelungensis ellipse indicatesJCM 14894, samples the yellow ellipse fromindicates C. lipolytica samples JCM 2304, from theY. bluedeformans ellipseJCM indicates 1694, samples the purple from ellipseC. lipolytica indicates JCM 21924, samples 8061 from C. lipolytica andJCM Y. lipolytica 2304, thePO1d. blue Each ellipse point indicatesrepresents samplesone replicate; from Y. C.lipolytica lipolytica PO1dJCM (light 21924, grey four 8061-point and Y. lipolytica starPO1d.s), C. lipolytica Each point JCM represents8061 (black onecircle replicate;s), C. lipolyticaY. lipolytica JCM 2304PO1d (dark (light grey greycircles four-point), C. lipolytica stars), C. lipolytica JCM 21924 (dark grey triangles), Y. deformans JCM 1694 (yellow four-point stars), and Y. keelungen- JCM 8061 (black circles), C. lipolytica JCM 2304 (dark grey circles), C. lipolytica JCM 21924 (dark grey sis JCM 14894 (blue circles); (right) top 25% of metabolites based on PCA scores from each princi- paltriangles), component.Y. The deformans full list JCMof PCA 1694 loading (yellow scores four-point is presented stars), in Table and Y. S8 keelungensis. JCM 14894 (blue circles); (right) top 25% of metabolites based on PCA scores from each principal component. The full list of PCASurprisingly, loading scores the metabolites is presented from in Table PC2 S8. include MEP and MEcPP; (Table S8) these metabolites are in the MEP pathway, which has not been reported in this yeast genus. However, Soliman et al. [30] detected a MEP-like pathway in fungi. We hypothesized that the nitrogen-limiting condition reduces purine and pyrimidine activity in yeast, which affects the energy availability in the cell. This stress leads to a cellular adjustment to the less ATP-dependent pathway, i.e., a switch from the MVA pathway to the MEP pathway, which requires less ATP. However, further biological investigations are required. Peak areas for all detected metabolites of purine and pyrimidine metabolism are summarized in Figures 7 and 8, respectively, all t-test p-value could be found in Table S9. These results clearly demonstrate that Yarrowia strains respond to the nitrogen-limiting condition in a similar way; notably, R5P, ADP, ATP, GDP, and GTP in purine metabolism were similar among strains. On the other hand, each species exhibited distinct metabolomic profiles for other metabolites in purine metabolism and all other detectable metabolites in pyrimidine metabolism. The laboratory Y. lipolytica PO1d strain had a distinct

MetabolitesMetabolites 20212021,, 1111,, x 16 FOR PEER REVIEW 109 of of 15 14

Figure 7. Normalized peak areas for all detected metabolites in purine metabolism. The pathway skeleton was obtained Figure 7. Normalized peak areas for all detected metabolites in purine metabolism. The pathway skeleton was obtained fromfrom the the KEGG KEGG pathway pathway database database [31] [31].. SC SC indicates indicates C:N C:N 4:1, 4:1, - -NN indicates indicates C:N C:N 5:0. 5:0. Error Error bars bars indicates indicates standard standard deviation, deviation, ** indicates indicates p--valuevalue < 0.05, 0.05, ** ** indicates indicates pp--valuevalue < < 0.001. 0.001.

MetabolitesMetabolites2021 2021, 11, 11, 16, x FOR PEER REVIEW 1011 of of 14 15

Figure 8. Normalized peak areas for all detected metabolites in pyrimidine metabolism. The pathway skeleton was obtained fromFigure the 8. KEGG Normalized pathway peak database. areas for SC all indicates detected C:N metabolites 4:1, -N indicates in pyrimidine C:N 5:0. metabolism. Error bars The indicates pathway standard skeleton deviation, was ob- * indicatestained fromp-value the KEGG < 0.05, pathway ** indicates database.p-value SC < 0.001.indicates C:N 4:1, -N indicates C:N 5:0. Error bars indicates standard deviation, * indicates p-value < 0.05, ** indicates p-value < 0.001.

Metabolites 2021, 11, 16 11 of 14

3. Materials and Methods 3.1. Yeast Strains, Cultivation, and Sample Collection The Y. lipolytica laboratory strain PO1d was obtained from the National University of Singapore, C. lipolytica JCM 2304, C. lipolytica JCM 21924, C. lipolytica JCM 8061, Y. deformans JCM 1694, and Y. keelungensis JCM 14894 were obtained from the Japan Collection of Mi- croorganisms (JCM). Yarrowia strains were pre-cultivated overnight in synthetic complete (SC) medium, all components were prepared according to the manufacturer’s instruction (20 g/L glucose, 1.7 g/L, yeast nitrogen-based without amino acids and ammonium sulfate purchased from BD Difco product number 233520, 5 g/L ammonium sulfate, drop-out medium supplements purchased from Sigma-Aldrich product number Y1501 consists of all standard amino acids at 76 mg/L except for leucine, which is present at 380 mg/L, adenine (18 mg/L), inositol (76 mg/L), p-aminobenzoic acid (8 mg/L) in 100-mL shake flasks at 30 ◦C and 200 rpm. The pre-cultured cells were inoculated into 20 mL of SC medium with different C:N ratios (e.g., C:N ratio 4:1 consists of 20 g/L glucose, 5 g/L ammonium sulfate, all standard amino acids at 76 mg/L except for leucine, which is present at 380 mg/L, 18 mg/L adenine, 76 mg/L inositol, 8 mg/L p-aminobenzoic acid, 1.7 g/L, yeast nitrogen-based without amino acids and ammonium sulfate. C:N 5:0 consists of 25 g/L glucose, all standard amino acids at 76 mg/L except for leucine, which is present at 380 mg/L, 18 mg/L adenine, 76 mg/L inositol, 8 mg/L p-aminobenzoic acid, 1.7 g/L, yeast nitrogen-based without amino acids and ammonium sulfate) to obtain an initial optical density at 600 nm of 0.5 in 100-mL shake flasks at 30 ◦C and 200 rpm. For sample collection, cells were collected at an OD600 of 5.0 at 12, 24, 36, and 48 h by fast filtration using a 0.45-µm pore size, 47-mm diameter nylon membrane (Millipore, Billerica, MA, USA). The cells were immediately immersed in liquid nitrogen for quenching and stored at −80 ◦C until extraction.

3.2. Fast Filtration and Extraction of Intracellular Metabolites Before LC-MS/MS, pre-analysis processes, such as quenching and storing the samples at −80 ◦C, were performed. Subsequently, data acquisition and analysis were performed to determine the relative concentration of each metabolite. The acquired signal intensities of metabolites were compared among samples after the normalization process. Here, the samples were normalized using champorsulfonic acid. For extraction, 1.8 mL of extraction solvent (methanol/water/chloroform = 5:2:2 v/v/v%, with 20 µg/L camphorsulfonic acid as an internal standard) was added to each 2-mL sampling tube with the ﬁltered sample, followed by incubation at −30 ◦C for 1 h. After incubation, 700 µL of the solution was transferred to a new tube containing 350 µL of water. The mixture was homogenized by vortexing and centrifuged at 9390× g. for 10 min at 4 ◦C to separate polar and non-polar phases. Then, 700 µL of the upper polar phase was transferred to a new tube via syringe ﬁltration (0.2-µm PTFE hydrophilic membrane; Millipore). The sample was centrifugally concentrated for 2 h and freeze-dried overnight. After reconstitution in 50 µL of ultrapure water, the sample was centrifuged at 16,000× g for 3 min at 4 ◦C and transferred to a glass vial for the LC/MS analysis.

3.3. Analysis of Intracellular Metabolites by LC-MS/MS The samples were analyzed based on the method adopted from Dempo et al. [32]. Brieﬂy, ion-pair reversed phase LC/MS/MS was carried out using a Nexera UHPLC system (Shimadzu, Kyoto, Japan) coupled with LCMS 8030 Plus (Shimadzu) for the time course analysis and coupled with LCMS 8050 (Shimadzu) for analyses of multiple strains. The column was a CERI (Chemicals Evaluation and Research Institute, Tokyo, Japan) L-column 2 metal-free ODS (150 mm 2.1 mm, particle size 3 µm). Mobile phase (A) was 10 mM tributylamine and 15 mM acetate in ultra-pure water, and mobile phase (B) was pure methanol. The ﬂow rate was set at 0.2 mL/min, and the column oven temperature was set at 45 ◦C. The concentration of mobile phase (B) was programmed to increase from 0% to 15%, 50%, and 100% from 1.0 to 1.5 min, 3.0 to 8.0 min, and 8.0 to 10.0 min, respectively, Metabolites 2021, 11, 16 12 of 14

held until 11.5 min, then decreased to 0% from 11.5 min and held at 0% for 20 min. The analysis mode was set to negative ion detection mode. The injection volume was 3 µL, desolvation line temperature was set at 250 ◦C, probe position was +1.5 mm, heat block temperature was set at 400 ◦C, nebulizer gas ﬂow was set at 2 L/min, and drying gas ﬂow was set at 15 L/min.

3.4. Multivariate Analysis Data analysis was carried out by converting the raw data from Shimadzu file format (.lcd) to an analysis base file (.abf) format using an open source file converter (Reifycs Inc., Tokyo, Japan). After data conversion, MRMPROBS 2.19 [33] was used for automatic peak selection and peak area integration. The detected peaks were manually confirmed using Lab Solution (Shimadzu Co.). A principal component analysis (PCA) was carried out using SIMCA-P+ version 13 (Umetrics, Umeå, Sweden). The metabolome data were normalized using an internal standard, mean-centered, and scaled to unit variance. A t-test to determine statistically significant differences was performed using MS Excel.

3.5. Glucose Measurement The remaining glucose in the culture broth was determined using the GL Science GL-7400 high-performance liquid chromatography system coupled with a column oven GL-7432 and refractive index detector GL-7454 (HPLC-RI). The column was Shim-pack SPR-Pb (250 mm length and ID 7.8 mm (Shimadzu). The supernatant of the culture broth was centrifuged at 10,000× g for 5 min at 4 ◦C; 300 µL of the sample was ﬁltered in a Whatman syringeless LC vial (Merck, Kenilworth, NJ, USA). The sample injection volume was 10 µL; the oven temperature was set to 80 ◦C. Ultrapure water was used as the mobile phase with a 0.6 mL/min ﬂow rate and 189 psi of back pressure.

4. Conclusions A metabolomics analysis revealed that Yarrowia is a very diverse genus, although the response to the nitrogen-limiting condition is similar among strains. However, despite the similarity, each strain had a unique metabolome profile. Nitrogen-limiting conditions cause lower purine and pyrimidine activity in Yarrowia spp., which affected the availability of energy molecules in the cell. This stress might lead to the adjustment to a less ATP- dependent pathway. This information will be beneficial for the development of Yarrowia strains for further scientific and industrial applications.

Supplementary Materials: The following are available online at https://www.mdpi.com/2218-1 989/11/1/16/s1, Figure S1: Heat map describes the time-course change of the normalized peak areas metabolome profile of 93 metabolites, Figure S2: Line chart of core metabolites extracted from top 25 percent of PC1 and PC2 from 12 to 48 h of cultivation error bars indicates standard deviation, Figure S3: All-time point score plot (A) all sample from 12–36 h of cultivation (B) all sample from 12–48 h of cultivation, Figure S4: Growth profile comparing Y. lipolytica spp. error bars indicates standard deviation, Figure S5: Heat map describes the average normalized peak areas of the metabolome profile of 100 metabolites from six Yarrowia spp. cultivated in two conditions. Table S1: p-value from t-test of growth and glucose consumption profile; yellow highlight indicates p-value < 0.05, Table S2: complete numerical results of normalized peak area in the time-course sampling of Y. lipolytica PO1d in different C:N ratio, Table S3: complete numerical results of PCA loading score from all samples of Y. lipolytica PO1d in different C:N ratio. Table S4: complete numerical results of PCA loading score from each time point of Y. lipolytica PO1d in different C:N ratio. Table S5: p-value from t-test of core metabolites line chart in Figure S2; yellow highlight indicates p-value < 0.05, Table S6: complete results of Venn diagram analysis from Figure4, Table S7: complete numerical results of all detected normalized peak area of six Yarrowia spp. in normal and nitrogen-limiting conditions, Table S8: complete numerical results of PCA loading score of six Yarrowia spp. in normal and nitrogen-limiting conditions. Table S9: p-value from t-test of Figures7 and8; yellow highlight indicates p-value < 0.05. Metabolites 2021, 11, 16 13 of 14

Author Contributions: Conceptualization, S.D. and S.P.P.; methodology, S.D.; validation, S.D., and S.P.P.; formal analysis, S.D. and S.P.P.; investigation, S.D.; resources, E.F.; data curation, S.D.; writing— original draft preparation, S.D.; writing—review and editing, S.P.P.; visualization, S.D.; supervision, E.F.; project administration, E.F.; funding acquisition, E.F. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available in the supplementary data and Supplementary Tables S1–S9. Acknowledgments: This work was supported by the Ministry of Education, Culture, Sports, Science, and Technology (MEXT). This study will be included in a dissertation submitted by Sivamoke Dissook to Osaka University in partial fulfillment of the requirement for his doctoral degree. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Papanikolaou, S.; Aggelis, G. Yarrowia lipolytica: A model microorganism used for the production of tailor-made lipids. Eur. J. Lipid Sci. Technol. 2010, 112, 639–654. [CrossRef] 2. Fickers, P.; Benetti, P.H.; Waché, Y.; Marty, A.; Mauersberger, S.; Smit, M.S.; Nicaud, J.M. Hydrophobic substrate utilisation by the yeast Yarrowia lipolytica, and its potential applications. FEMS Yeast Res. 2005, 5, 527–543. [CrossRef] 3. Ledesma-Amaro, R.; Nicaud, J.-M. Metabolic Engineering for Expanding the Substrate Range of Yarrowia lipolytica. Trends Biotechnol. 2016, 34, 798–809. [CrossRef] 4. Zhu, Q.; Jackson, E.N. Metabolic engineering of Yarrowia lipolytica for industrial applications. Curr. Opin. Biotechnol. 2015, 36, 65–72. [CrossRef] 5. Michely, S.; Gaillardin, C.; Nicaud, J.M.; Neuvéglise, C. Comparative Physiology of Oleaginous Species from the Yarrowia Clade. PLoS ONE 2013, 8, 1–10. [CrossRef] 6. Crous, P.W.; Wingfield, M.J.; Burgess, T.I.; Carnegie, A.J.; Hardy, G.E.S.J.; Smith, D.; Summerell, B.A.; Guarro, J.; Houbraken, J.; Lombard, L.; et al. Fungal Planet Description Sheets: 625–715. In Persoonia—Molecular Phylogeny and Evolution of Fungi; Naturalis Biodiversity Center: Leiden, The Netherlands, 2017; Volume 39, pp. 270–467. 7. Nagy, E.; Dlauchy, D.; Medeiros, A.O.; Péter, G.; Rosa, C.A. Yarrowia porcina sp. nov. and Yarrowia bubula f.a. sp. nov., two yeast species from meat and river sediment. Antonie Van Leeuwenhoek Int. J. Gen. Mol. Microbiol. 2014, 105, 697–707. [CrossRef] 8. Manikandan, K.; Viruthagiri, T. Optimization of C/N ratio of the medium and fermentation conditions of ethanol production from tapioca starch using co-Culture of Aspergillus niger and Sachormyces cerevisiae. Int. J. ChemTech Res. 2010, 2, 947–955. 9. Egli, T.; Quayle, J.R. Influence of the carbon: Nitrogen ratio of the growth medium on the cellular composition and the ability of the methylotrophic yeast Hansenula polymorpha to utilize mixed carbon sources. J. Gen. Microbiol. 1986, 132, 1779–1788. [CrossRef] 10. Gao, L.; Sun, M.H.; Liu, X.Z.; Che, Y.S. Effects of carbon concentration and carbon to nitrogen ratio on the growth and sporulation of several biocontrol fungi. Mycol. Res. 2007, 111, 87–92. [CrossRef][PubMed] 11. Jackson, M.A.; Bothast, R.J. Carbon concentration and carbon-to-nitrogen ratio influence submerged-culture conidiation by the potential bioherbicide Colletotrichum truncatum NRRL 13737. Appl. Environ. Microbiol. 1990, 56, 3435–3438. [CrossRef][PubMed] 12. Sheng, J.; Feng, X. Metabolic engineering of yeast to produce fatty acid-derived biofuels: Bottlenecks and solutions. Front. Microbiol. 2015, 6, 554. [CrossRef][PubMed] 13. Ryan, D.; Robards, K. Metabolomics: The greatest omics of them all? Anal. Chem. 2006, 78, 7954–7958. [CrossRef][PubMed] 14. Knutsen, A.K.; Robert, V.; Poot, G.A.; Epping, W.; Figge, M.; Holst-Jensen, A.; Skaar, I.; Smith, M.T. Polyphasic re-examination of Yarrowia lipolytica strains and the description of three novel Candida species: Candida oslonensis sp. nov., Candida alimentaria sp. nov. and Candida hollandica sp. nov. Int. J. Syst. Evol. Microbiol. 2007, 57, 2426–2435. [CrossRef] 15. Barth, G.; Gaillardin, C. Yarrowia lipolytica. In Nonconventional Yeasts in Biotechnology; Springer: Berlin/Heidelberg, Gemany, 1996; pp. 313–388. 16. Nakase, T.; Goto, S.; Koyama, Y.; Komagata, K.; Iizuka, H. Microbiological Studies on Cheese (II). Food Hyg. Saf. Sci. (Shokuhin Eiseigaku Zasshi) 1977, 18, 353–361. [CrossRef] 17. Groenewald, M.; Smith, M.T. The teleomorph state of Candida deformans Langeron & Guerra and description of Yarrowia yakushimensis comb. nov. Antonie Van Leeuwenhoek 2013, 103, 1023–1028. 18. Chang, C.-F.; Chen, C.-C.; Lee, C.-F.; Liu, S.-M. Identifying and characterizing Yarrowia keelungensis sp. nov., an oil-degrading yeast isolated from the sea surface microlayer. Antonie Van Leeuwenhoek 2013, 104, 1117–1123. [CrossRef] 19. Trygg, J.; Holmes, E.; Lundstedt, T. Chemometrics in metabonomics. J. Proteome Res. 2007, 6, 469–479. [CrossRef] 20. Vichi, M.; Saporta, G. Clustering and disjoint principal component analysis. Comput. Stat. Data Anal. 2009, 53, 3194–3208. [CrossRef] Metabolites 2021, 11, 16 14 of 14

21. Bylesjö, M.; Eriksson, D.; Sjödin, A.; Jansson, S.; Moritz, T.; Trygg, J. Orthogonal projections to latent structures as a strategy for microarray data normalization. BMC Bioinform. 2007, 8, 207. [CrossRef] 22. Soutourina, J.; Plateau, P.; Blanquet, S. Metabolism of d-Aminoacyl-tRNAs in Escherichia coli and Saccharomyces cerevisiae Cells. J. Biol. Chem. 2000, 275, 32535–32542. [CrossRef] 23. Ibba, M.; Söll, D. Aminoacyl-tRNAs: Setting the limits of the genetic code. Genes Dev. 2004, 18, 731–738. [CrossRef][PubMed] 24. O’donovan, G.A.; Neuhard, J. Pyrimidine Metabolism in Microorganisms. Bacteriol. Rev. 1970, 34, 278. [CrossRef][PubMed] 25. Vogels, G.D.; van der Drift, C. Degradation of Purines and Pyrimidines by Microorganisms. Bacteriol. Rev. 1976, 40, 403. [CrossRef] [PubMed] 26. Henderson, J.F.; Paterson, R.P. Catabolism of purine nucleotides. In Nucleotide Metabolism: An Introduction; Academic Press: New York, NY, USA, 1973; pp. 3–20. 27. De Virgilio, C.; Hottiger, T.; Dominguez, J.; Boller, T.; Wiemken, A. The role of trehalose synthesis for the acquisition of thermotolerance in yeast I. Genetic evidence that trehalose is a thermoprotectant. Eur. J. Biochem. 1994, 219, 179–186. [CrossRef] 28. Hottiger, T.; de Virgilio, C.; Hall, M.N.; Boller, T.; Wiemken, A. The role of trehalose synthesis for the acquisition of thermotolerance in yeast II. Physiological concentrations of trehalose increase the thermal stability of proteins in vitro. Eur. J. Biochem. 1994, 219, 187–193. [CrossRef] 29. Leslie, B.; Teter, S.A.; Crowe, L.M.; Crowe, J.H. Trehalose lowers membrane phase transitions in dry yeast cells. Biochim. Biophys. Acta BBA Biomembr. 1994, 1192, 7–13. [CrossRef] 30. Soliman, S.S.M.; Tsao, R.; Raizada, M.N. Chemical inhibitors suggest endophytic fungal paclitaxel is derived from both mevalonate and non-mevalonate-like pathways. J. Nat. Prod. 2011, 74, 2497–2504. [CrossRef] 31. Kanehisa, M.; Goto, S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000, 28, 27–30. [CrossRef] 32. Dempo, Y.; Ohta, E.; Nakayama, Y.; Bamba, T.; Fukusaki, E. Molar-based targeted metabolic proﬁling of cyanobacterial strains with potential for biological production. Metabolites 2014, 4, 499–516. [CrossRef] 33. Tsugawa, H.; Arita, M.; Kanazawa, M.; Ogiwara, A.; Bamba, T.; Fukusaki, E. MRMPROBS: A data assessment and metabolite identiﬁcation tool for large-scale multiple reaction monitoring based widely targeted metabolomics. Anal. Chem. 2013, 85, 5191–5199. [CrossRef]