Integrated transcriptomic and DNA methylomic analysis of Yarrowia lipolytica in response to La(III)/Ce(III) stress

Abstract

Rare earth elements (REEs), as strategic resources, cause severe pollution and ecological degradation through chemical leaching processes. Bioremediation technology offers an efficient green alternative for REE recovery from wastewater. This study investigated Yarrowia lipolytica to elucidate its efficient adsorption mechanisms and stress adaptation towards La(III)/Ce(III). Phenotypic analysis revealed that the strain enhances adsorption capacity for La(III)/ Ce(III) by increasing specific surface area through dimorphic transition. Under optimal conditions, adsorption rates reached 84.33% for La(III) and 87.21% for Ce(III). Adsorption kinetics followed a pseudo-second-order model (indicating chemisorption dominance), and isotherms conformed to the Langmuir model (suggesting monolayer adsorption). FTIR and XPS analyses identified cell surface -OH groups as key active sites, directly capturing REE ions via complexation. Integrated transcriptomic and DNA methylomic analyses uncovered interaction mechanisms and stress responses: La(III) exposure inhibited glycolysis/TCA cycle genes while activating peroxisome pathways (antioxidant defense) and ABC transporters (ion efflux). Ce(III) exposure specifically suppressed amino acid metabolism (e.g., glutamate pathway). Whole-genome methylation levels decreased significantly with preferential methylation in CHH contexts. 10 (La(III)) and 4 (Ce(III)) were identified differentially expressed genes accompanied by altered methylation levels, demonstrating DNA methylation-mediated regulation of La(III)/Ce(III) resistance genes. This study lays a theoretical foundation for bioremediation of REE pollution.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12934-026-03041-7.

Introduction

Rare earth elements (REEs), as indispensable strategic resources in modern industry, are widely used in fields such as new energy, electronic devices, catalysts, and defense technology [9]. China possesses the world's largest reserves of rare earths, with over 250 discovered rare earth minerals, of which approximately 50 to 60 have industrial value [13, 40]). The most important rare earth minerals include bastnaesite, monazite, xenotime, and ion-adsorbed rare earth minerals [34, 62]). Rare earth minerals can be broadly categorized into the following three types [39]: Mineral-type rare earth ores, such as monazite and bastnaesite, In these minerals, rare earth elements exist as ionic compounds within the mineral lattice; Minerals containing rare earth elements, such as apatite and fluorite, In these minerals, rare earth elements enter the crystal structure of other metal minerals as impurities through isomorphic substitution.; Ion-adsorbed rare earth minerals, such as monazite placer deposits, mica minerals, and clay minerals. In these, rare earth elements are adsorbed onto mineral surfaces as ions. It is noteworthy that most rare earth minerals primarily contain lanthanum and cerium [40]. Currently, the predominant extraction processes for rare earth elements from minerals—both domestically and internationally—rely on chemical leaching [41]. This process requires substantial amounts of acid and inevitably generates large quantities of environmentally polluting chemical waste and acidic wastewater [17]. Long-term mining has led to severe ecological degradation, while traditional physicochemical remediation methods often cause secondary pollution [23]. To address these environmental challenges, bioremediation technology demonstrates significant potential for recovering and concentrating rare earth elements from wastewater and waste residues. Wang et al. [59] isolated Penicillium sp. ZD28 from soil in a rare earth mining area. This strain exhibited high tolerance to yttrium (Y), achieving an adsorption rate of up to 99% for Y(Ⅲ) under environmental conditions of 600 μM Y(Ⅲ). Song et al. [49] demonstrated that carboxyl and phosphate groups on microalgal surfaces can adsorb rare earth ions through coordination complexation, thereby enabling the remediation of rare earth-containing wastewater.

Yarrowia lipolytica (Y. lipolytica) is an unconventional yeast that is considered a promising microorganism for application in various biotechnologies due to its biochemical characteristics [66]. In numerous environmental water bodies contaminated with metals, multiple strains of Y. lipolytica exhibiting tolerance to metal ions and adsorption capabilities have been isolated. Xiaoman Xie et al. [63] enhanced the adsorption capacity of Y. lipolytica for rare earth ions by surface-displaying the rare earth-binding protein lanmodulin (LanM), achieving adsorption capacities of 48.72–50.38 mg/g for multiple rare earth elements. According to a report by Bankar et al. [4], two marine strains—Y. lipolytica NCIM 3589 and Y. lipolytica NCIM 3590, both originating from metal-contaminated seawater—demonstrated the ability to adsorb Cr(VI). Under optimal conditions (35 °C, pH 1.0, 130 rpm), the specific uptake capacities reached 63.73 ± 1.3 mg/g for NCIM 3589 at an initial Cr(VI) concentration of 950 ppm and 46.09 ± 0.23 mg/g for NCIM 3590 at 955 ppm, respectively. Similarly, in mercury (Hg)-contaminated river sediments, Y. lipolytica Idd1 and Y. lipolytica Idd2 were screened out, and their Hg(II) adsorption capacities were found to be 32 mg/g and 59 mg/g, respectively [42]. Further studies on these two strains revealed that during growth, the yeast strains achieved excellent Hg removal (> 97%) from the medium containing ≤ 16 μg/mL Hg²⁺ via bioaccumulation, volatilization and micro-precipitation [43].

Transcriptomics has been widely applied to unravel the molecular mechanisms underlying microbial responses to metal stress [22, 32]). In this study, the marine yeast Y. lipolytica was used as the model organism to investigate microbial responses to rare earth ion stress. In the marine yeast Y. lipolytica, Kolhe et al. performed transcriptomic analysis following exposure to 50 µM uranium, revealing that oxidative stress response mechanisms were the major reaction to uranium toxicity, with 56 differentially expressed genes identified, including those involved in transport, DNA damage repair, and oxidative stress [31]. Beyond uranium, the metal interactions of Y. lipolytica have been comprehensively reviewed, highlighting its intrinsic capacity to tolerate and respond to various metal pollutants [30]. Furthermore, Shen et al. investigated the biosorption behavior and mechanism of Y. lipolytica for five rare earth ions (La³⁺, Nd³⁺, Er³⁺, Y³⁺, and Sm³⁺), demonstrating that the yeast employs a monolayer chemical biosorption process involving amino, carboxyl, and hydroxyl groups to biosorb rare earth ions, with a maximum biosorption capacity of 76.80 mg/g under optimal conditions [47]. Collectively, these studies demonstrate that transcriptomics plays a pivotal role in elucidating microbial response mechanisms to metal stress. Consequently, transcriptomics emerges as an indispensable analytical tool in research on microbial resistance mechanisms under rare earth ion stress. DNA methylation, a primary epigenetic mechanism, regulates pre-transcriptional gene expression in eukaryotes without altering the DNA sequence [10]. Current research on DNA methylation dynamics under metal stress is primarily focused on plant systems, while studies in non-plant eukaryotes remain limited. Accumulating evidence suggests that when plants are exposed to heavy metal-contaminated environments, dynamic changes in DNA methylation are associated with the activation and suppression of specific gene expression [26, 36, 51]). This epigenetic mechanism enables plants to respond to metal stress by modulating the expression of relevant genes. Beyond investigating DNA methylation alterations under various metal ion stresses, studies have also reported enhanced metal ion tolerance in eukaryotes through deliberate manipulation of DNA methylation levels. For instance, treatment with the DNA methylation inhibitor 5-azacytidine reduced global DNA methylation levels in grape seedlings, thereby alleviating aluminum toxicity and enhancing plant tolerance [65]. Similarly, the overexpression of a methyltransferase gene (CIMT1) was sufficient to enhance cadmium tolerance in Arabidopsis [24]. Collectively, these findings underscore the pivotal role of DNA methylation in eukaryotic responses to metal stress. However, current research on fungal DNA methylation dynamics under metal stress remains limited. This study aims to address this gap to some extent.

In this study, phenotypic analysis was first conducted on Y. lipolytica under La(III)/Ce(III) stress. Subsequently, RNA-seq and whole-genome bisulfite sequencing (WGBS) were employed to characterize transcriptomic profiles and genome-wide DNA methylation dynamics in Y. lipolytica exposed to La(III)/Ce(III) stress. Differentially expressed genes (DEGs) and differentially methylated genes (DMGs) were functionally annotated through enrichment analysis to elucidate the molecular mechanisms underlying the yeast's response to rare earth ion stress. Finally, integrative analysis of transcriptomics and methylomics was performed to investigate the association between DNA methylation and the expression regulation of tolerance-related genes in Y. lipolytica under La(III)/Ce(III) stress, thereby revealing the mechanistic basis of Y. lipolytica -rare earth ion interactions.

Materials and methods

Strains, culture media, and chemicals

The Y. lipolytica W29 (ATCC 20460) strain used in this experiment was stored in our laboratory.

YPD medium was prepared containing 20 g/L glucose, 20 g/L tryptone, and 10 g/L yeast extract, and was sterilized at 115 °C for 30 min. To prepare solid medium, 15 g/L agar was added.

The 0.1% Tween 80 solution was prepared by adding 20 μL Tween 80 to 20 mL deionized water, followed by filter sterilization through a 0.22 μm membrane.

YNB medium was prepared with 6.7 g/L Yeast Nitrogen Base and 20 g/L glucose, and was sterilized by autoclaving at 115 °C for 30 min. As the minimal medium for Y. lipolytica, filter-sterilized LaCl₃ or CeCl₃ stock solutions (0.22 μm filtered) were supplemented to target concentrations for tolerance assays of Y. lipolytica W29.

The additional reagents used in this experiment included lanthanum chloride hexahydrate (LaCl₃·6H₂O), cerium chloride hexahydrate (CeCl₃·6H₂O), rare earth ion standard solutions, sodium hydroxide (NaOH), and hydrochloric acid (HCl). All reagents were purchased from Sinopharm Chemical Reagent Co., Ltd. (China) and were of analytical grade purity.

Determination of La(III)/Ce(III) tolerance capacity

After revival of Y. lipolytica W29, 1 mL of activated yeast culture in the logarithmic growth phase was inoculated into YNB medium without added external rare earth ions. Samples were taken at 6, 12, 18, 24, 30, and 48 h to measure OD600 and assess the growth capacity of the strain. Similarly, 1 mL of revived yeast suspension in the logarithmic growth phase was inoculated into YNB medium with initial La(Ⅲ)/Ce(Ⅲ) concentrations ranging from 50 to 500 mg/L. After 24 h of incubation, samples were taken at various time points (6, 12, 18, 24, 30, and 48 h) to measure OD600 and assess the lipid-degrading yeast's tolerance to La(Ⅲ)/Ce(Ⅲ).

SEM–EDS characterization of cell morphology

Collect Y. lipolytica W29 cells cultured in control medium without rare earth ions and in medium containing the highest tolerable concentration of La(III)/Ce(III) for 24 h, freeze at -80 °C overnight, dry using a vacuum freeze dryer, and observe morphological changes in the strain under La(III)/Ce(III) stress using SEM. Sample preparation followed the standard operating procedure provided by the SEM testing facility. SEM images were acquired at a magnification of × 10,000 (scale bar = 10 µm).

Batch adsorption experiments

This study aims to investigate the adsorption capacity of Y. lipolytica W29 under different pH conditions (3.0–7.0), initial La(III) concentrations (30–70 mg/L), and bioadsorbent addition levels (300–2000 mg/L), in order to determine the optimal conditions for rare earth ion adsorption. All adsorption experiments were conducted in an oscillating incubator at 30 °C and 150 rpm. Before adsorption (i.e., the initial metal solution without Y. lipolytica W29 cells) and after adsorption equilibrium, the samples were filtered using a 0.22 μm membrane filter, and the La(III)/Ce(III) concentrations were determined by ICP-OES. The adsorption rate (A_e) was calculated using the following formula:

1

C₀(mg/L) and C_t(mg/L) represent the initial concentration of La(III)/Ce(III) before adsorption and the concentration of La(III)/Ce(III) at adsorption equilibrium, respectively.

Kinetic and Isotherm Models

The equations for the pseudo-first-order and pseudo-second-order kinetic models are expressed as follows:

2

3

(mg/g) represents the adsorption capacity of the Y. lipolytica strain for La(III) at time “t”, k₁(min⁻¹) and k₂(mg·g⁻¹·min⁻¹) denote the rate constants for the pseudo-first-order and pseudo-second-order models, respectively;

The adsorption data were fitted using the two most widely applied adsorption isotherm models: Freundlich model (2–4) and Langmuir model (2–5)

4

5

C_e (mg/L): La(III) concentration at adsorption equilibrium; (mg/g): Maximum adsorption capacity of the strain; K_F (mg/g): Characteristic constant of the Freundlich isotherm; K_L (L/mg): Adsorption equilibrium constant of the Langmuir model; n: Dimensionless parameter related to adsorption intensity.

FTIR and XPS analysis

The yeast cells before adsorption (i.e., cells cultured in control medium without La(III)/Ce(III)) and after adsorption (i.e., cells harvested from the adsorption medium containing La(III)/Ce(III)) were freeze-dried and ground into powder. For FTIR analysis, the powder was mixed thoroughly with dried potassium bromide, pressed into pellets, and then scanned using a Fourier transform infrared spectrometer (NEXUS 670, Thermo Nicolet, USA), with the scanning range from 4000 cm⁻¹ to 500 cm⁻¹. For XPS analysis, an X-ray photoelectron spectrometer (ESCA-LAB 250Xi, Thermo Fisher Scientific, USA) was directly applied to the freeze-dried and ground powder to investigate changes in the chemical element composition and state on the surface of the microbial cells before and after adsorption.

RNA-seq analysis

After revival, Y. lipolytica W29 was inoculated into YPD liquid medium containing 400 mg/L La(III), 400 mg/L Ce(III), and no rare earth ions, respectively. All cultures were incubated at 28 °C and 200 rpm for 3 days. The cells were then centrifuged at 4 °C and 8000 × g for 5 min to collect the cell pellets, which were subsequently stored at -80 °C in a ultra-low temperature freezer. Each group of samples included three biological replicates. The RNA sequencing libraries were prepared by Beijing Novogene Co., Ltd. (Beijing, China).

Raw reads were quality-controlled and trimmed to remove adapters and low-quality sequences. The clean reads were aligned to the Y. lipolytica reference genome (CLIB122, GenBank assembly accession: GCF_000002525.2). Gene expression levels were quantified, and differentially expressed genes (DEGs) were identified using DESeq2 with criteria of |log₂(fold change)|≥ 1 and adjusted P-value < 0.05. Functional enrichment analysis of DEGs was performed using Gene Ontology (GO) and KEGG pathway databases.

RT-qPCR validation

To validate the RNA-seq results, DEGs were selected for RT-qPCR analysis from each treatment group. Specifically, for the Ce(III) treatment group, five upregulated and five downregulated genes were selected; for the La(III) treatment group, another five upregulated and five downregulated genes were selected. The names of all selected genes and their primer sequences are listed in Supplementary Table 1. YlACT1(YALI1_D10600g, Gene ID: 2911067) was used as the endogenous control. Primers were designed using SnapGene (v6.1.1). Each RT-qPCR reaction (20 µL total volume) contained 0.4 µL of each primer, 2 µL of cDNA, and 10 µL of SYBR Green mix. The thermal cycling protocol was as follows: 95 °C for 3 min, followed by 45 cycles of 95 °C for 5 s and 60 °C for 30 s.

WGBS and Data Analysis

The same three experimental groups as in Sect. "RNA-seq analysis" (La(III) treatment group, Ce(III) treatment group, and control group) were subjected to whole-genome bisulfite sequencing (WGBS). DNA samples were sequenced on the Illumina NovaSeq 6000 platform (Illumina, USA). Low-quality data in the sequencing data were filtered using TrimGalore software (v 0.4.4). The filtered data were validated using FastQC software (v 0.11.2). Valid data were aligned with the reference genome using BSMAP software (v 2.9.0). Differentially methylated regions (DMRs) were identified using the sliding window method.

Statistical analysis

All experiments were performed in triplicate, and the results are expressed as mean ± standard deviation (SD). Error bars in all figures represent SD from three independent replicates.

Results and discussion

Phenotypic analysis of Y. lipolytica W29 under La(III)/Ce(III) stress

Tolerance of Y. lipolytica W29 to La(Ⅲ)/Ce(Ⅲ)

The tolerance of Y. lipolytica W29 to La(Ⅲ)/Ce(Ⅲ) was investigated by measuring the OD₆₀₀ of yeast suspensions. As shown in Fig. 1A, the growth curve of Y. lipolytica W29 indicates that the strain entered the logarithmic phase at 12 h and reached its maximum OD₆₀₀ after growing for 24 h. Further assessment of its tolerance to La(Ⅲ)/Ce(Ⅲ) was conducted by measuring the OD₆₀₀ after 24 h of growth in La(Ⅲ)/Ce(Ⅲ) concentrations ranging from 50 to 500 mg/L. The 24 h time point was chosen because the strain reaches the late logarithmic/early stationary phase at this time, where the inhibitory effect on final cell density can be more clearly observed. The growth of strain was only slightly inhibited at La(Ⅲ)/Ce(Ⅲ) concentrations of 50–300 mg/L. However, at 500 mg/L of La(Ⅲ) and Ce(Ⅲ), the OD₆₀₀ dropped sharply, indicating maximal growth inhibition. Thus, the maximum tolerance level of strain to both La(Ⅲ) and Ce(Ⅲ) was determined to be 400 mg/L. (Fig. 1B).

Fig. 1.

Growth curves and stress tolerance of Y. lipolytica W29 (A, B)

La(Ⅲ)/Ce(Ⅲ)-induced cellular morphological alterations in Y. lipolytica W29

SEM–EDS was employed to examine the morphological changes of Y. lipolytica W29 cells under La(Ⅲ)/Ce(Ⅲ) stress at the maximum tolerable concentration (400 mg/L), as determined in the tolerance assay (Sect. "Tolerance of Y. lipolytica W29 to La(Ⅲ)/Ce(Ⅲ)", Fig. 1B). Y. lipolytica exhibits dimorphic transition capability, whereby it switches between yeast and hyphal morphologies in response to environmental changes. This adaptive mechanism is considered to be associated with the yeast's ability to maintain cellular homeostasis under adverse environmental stress [55]. Multiple metal ions have been reported to induce dimorphic transition in Y. lipolytica. Studies demonstrate that upon exposure to Cr(VI) or Pb(II) containing environments, morphological changes occur in Y. lipolytica NCIM3589, with a subset of cells observed to transform into hyphal morphology [3]. As shown in Fig. 2, control-cultured Y. lipolytica cells exhibited oval-shaped yeast morphology with bipolar budding, consistent with literature reports [30]. In contrast, under La(III) and Ce(III) stress (400 mg/L), cells exhibited both hyphal and yeast morphologies (Fig. 2B, C), indicating that both La(III) and Ce(III) induce dimorphic transition from yeast to hyphal forms in Y. lipolytica. SEM images were captured at magnifications of × 10,000 to clearly visualize the morphological transition. Studies have demonstrated that Cu(II) specifically triggers this dimorphic transition in Y. lipolytica, revealing it as a coping strategy against copper stress [46]. Furthermore, research indicates that compared to oval yeast forms, hyphal morphology reduces the surface area-to-volume ratio of cells, which facilitates enhanced biomass accumulation in yeasts [67]. Therefore, based on the aforementioned results, it can be inferred that the La(III)/Ce(III)-induced dimorphic transition represents a specific response of Y. lipolytica to rare earth ion stress. This morphological shift confers a survival advantage in La(III)- and Ce(III)-containing environments. Subsequent transcriptomic analysis of Y. lipolytica W29 under La(III) and Ce(III) stress revealed differential expression in a subset of genes associated with dimorphic transition, further corroborating this postulate.

Fig. 2.

SEM–EDS analysis of Y. lipolytica W29 cultured under different conditions: A SEM image of control culture, B SEM image of culture with La(III), C SEM image of culture with Ce(III), D EDS spectrum of control culture, E EDS spectrum of culture with La(III), F EDS spectrum of culture with Ce(III)

Biosorption capacity of Y. lipolytica W29 for La(III)/Ce(III)

Determination of optimal biosorption conditions

To determine the optimal biosorption conditions, experiments were conducted to evaluate the effects of pH, initial ion concentration, and biosorbent dosage using La(III) as a representative rare earth ion. Figure 3 illustrates the effects of various factors on biosorption of La(III) by Y. lipolytica W29.

Fig. 3.

Effects of various factors on biosorption of La(III) by Y. lipolytica W29. A pH dependence; B initial ion concentration dependence; C biosorbent dosage dependence

pH critically influences biosorption by modulating the activity of cell wall functional groups, the chemical behavior of metal ions, and inter-ion competition[7]. The biosorption efficiency and equilibrium capacity of Y. lipolytica W29 for La(III)/Ce(III) increased concomitantly with rising pH (Fig. 3A). This pH-dependent adsorption profile is governed by competitive ion interactions at binding sites. At low pH, high H₃O⁺ concentrations compete with rare earth ions for adsorption sites. Concurrent protonation of functional groups induces electrostatic repulsion toward cationic La(III)/ Ce(III), significantly reducing biosorption efficiency [6, 29]). As pH elevates, increased OH⁻ availability exposes negatively charged groups (phosphate, carboxyl, and amino groups), enhancing electrostatic attraction to metal cations [25]. This maximizes biosorption capacity near the isoelectric point of cell walls. Above pH > 7.0 [5], La(III) and Ce(III) form insoluble hydroxides (La(III), Ce(Ⅲ)), which confounds accurate assessment of biosorption capacity [22].

The effect of initial ion concentration on the adsorption of Y. lipolytica W29 was further explored. According to Fig. 3B, the adsorption efficiency of Y. lipolytica W29 all decreased gradually with the increase of the initial rare earth ion concentration, and the adsorption capacity reached the highest value of 63.62 mg/g at the initial rare earth ion concentration of 50 mg/L. The probability of the collision of the rare earth ions with the effective adsorption sites of the strain increased with the increase of the initial rare earth ion concentration, but the surface of the strain for the adsorption of rare earth ions is limited, so the adsorption of rare earth ions by the strain will gradually reach saturation, which may be the reason why the adsorption capacity at equilibrium increased with the increase of the initial rare earth ion concentration and there was a maximum value [21].

The addition amount of biosorbent determines the number of metal ion binding sites in the adsorption system, which is an important factor affecting its adsorption efficiency on metal ions [27]. As shown in Fig. 3C, the adsorption rate of Y. lipolytica W29 was directly proportional to the addition amount of biosorbent, and the maximum adsorption rate was reached at the addition amount of 1500 mg/L. In addition, the adsorption capacity of Y. lipolytica W29 at equilibrium were all inversely proportional to the adsorbent addition. With the increase of biosorbent addition, the binding sites for the adsorption of rare earth ions gradually increased, and thus the adsorption efficiency was enhanced, but too high a concentration of biosorbent would lead to the shielding effect of the peripheral yeast on the internal biosorbent, which would result in the reduction of binding sites available for the adsorption of metal ions to the extent that the adsorption efficiency would be reduced. Furthermore, it has been reported that at a fixed adsorbate concentration, an increase in the biosorbent dosage results in an excess of binding sites, which may also account for the observed decline in adsorption capacity[54].

Based on the integrated experimental findings, the optimal biosorption conditions for Y. lipolytica W29 were identified as pH 7.0, initial ion concentration of 50 mg/L, and biosorbent dosage of 1.5 g/L. Under these optimized parameters, the adsorption rate of Y. lipolytica W29 for La(III) reached 84.33%. Owing to the similar chemical properties and adsorption behavior of La(III) and Ce(III) [56, 57]), the optimal conditions determined for La(III) are applicable to Ce(III) biosorption. Subsequently, Ce(III) was tested under the same optimal conditions to evaluate the adsorption performance of Y. lipolytica W29 for this rare earth element, yielding an adsorption rate of 87.21%, as summarized in Table 1.

Table 1.

Biosorption capacity of Y. lipolytica W29 for La(III)/Ce(III) under optimal conditions

Strain	Adsorbed ion type	Adsorption rate (%)	Adsorption capacity (mg/g)
Y. lipolytica W29	La(Ⅲ)	84.33 ± 1.21	28.11 ± 0.41
	Ce(Ⅲ)	87.21 ± 0.95	29.07 ± 0.32

Biosorption kinetics and isotherms

The biosorption process of Y. lipolytica W29 was modeled using pseudo-first-order and pseudo-second-order kinetics. The fitting results and associated parameters are presented in Fig. 4A and Table S2. The pseudo-second-order model yielded a higher R² value compared to the pseudo-first-order model, indicating that the biosorption process better conforms to pseudo-second-order kinetics. This model estimated an equilibrium biosorption capacity of 64.9827 mg/g, consistent with experimental results. These findings demonstrate that chemisorption dominates the biosorption mechanism, serving as the rate-limiting step. This behavior is attributed to valence forces arising from electron exchange/sharing between functional groups on the cellular surface and metal ions [11].

Fig. 4.

Kinetic and Isotherm Models for Biosorption by Y. lipolytica W29, A Kinetic model fitting; B Adsorption isotherm fitting

The experimental biosorption data of Y. lipolytica W29 were fitted using Langmuir and Freundlich isotherm models, with the fitting results and corresponding parameters presented in Fig. 4B and Table S3. The Langmuir model exhibited a higher R² value than the Freundlich model in fitting the biosorption data of Y. lipolytica W29, indicating its superior suitability for describing the adsorption process. This alignment confirms monolayer adsorption as the dominant mechanism. Prior studies report analogous findings for Ni(II) and Zn(II) biosorption by Y. lipolytica W29, consistent with the Langmuir dominance observed herein [61].

Mechanism of La(III)/Ce(III) biosorption by Y. lipolytica W29

FTIR was employed to characterize Y. lipolytica W29 cells before and after La(III)/Ce(III) biosorption, aiming to identify key chemical groups involved in metal ion binding (Fig. 5). Post-adsorption spectral analysis revealed peak shifts in hydroxyl (-OH), amino (-NH₂), aliphatic (-CH), carboxyl (-COOH), carbonyl (C = O), and phosphate (PO₄³⁻) groups, confirming their direct participation in rare earth ion coordination [48, 63]). Notably, a new absorption peak emerged at 1541 cm⁻¹, corresponding to the amide II band (C-N stretching and N–H bending vibrations of peptide bonds). According to studies by Xie et al. on ytterbium (Yb(III)) biosorption by Y. lipolytica, carboxyl and phosphate groups play a dominant role in the adsorption process, forming stable complexes with Yb(III) through ligand exchange and coordination reactions [61]. Our experiments similarly demonstrate the involvement of amino, hydroxyl, carbonyl, carboxyl and phosphate groups on the surface of Y. lipolytica W29 in the adsorption of La(III) and Ce(III). These functional groups likely facilitate adsorption through analogous ligand exchange and coordination mechanisms.

Fig. 5.

FTIR of Y. lipolytica W29 before and after biosorption

To further validate the FTIR characterization results, XPS was employed to analyze the elemental composition and chemical valence states of Y. lipolytica W29 before and after biosorption. The XPS survey spectra of La/Ce-loaded biomass (Fig. 6A) clearly showed characteristic peaks of La 3d and Ce 3d, confirming successful biosorption of La(III) and Ce(III) by the strain. In the La 3d spectrum (Fig. 6B), four distinct peaks were observed at binding energies of 834.87 eV (La 3d_5/2), 837.93 eV (La 3d_5/2), 851.77 eV (La 3d_3/2), and 854.87 eV (La 3d_3/2), consistent with La(III) oxidation state. Similarly, the Ce 3d spectrum (Fig. 6C) exhibited two spin–orbit splitting peaks at 884.48 eV (La 3d_5/2) and 903.39 eV (La 3d_3/2)[14], confirming exclusive presence of trivalent cerium (Ce(III)). As shown in Fig. 6 E, two peaks, N–C = O/-NH₂ and -NH₃⁺[33], were observed before and after the adsorption of Y. lipolytica W29, and both absorption peaks were shifted after adsorption, and the peak area of -NH₃⁺ was greatly reduced, while the peak area of N–C = O/-NH₂ was significantly elevated, which indicated that both N–C = O/-NH₂ and -NH₃⁺ participated in the yeast adsorption of La(III) and Ce(III) adsorption process. The O1s profile of Y. lipolytica W29 (Fig. 6 F) could be decomposed into two peaks, H–O-H (530.76 eV) and -OH (532.11 eV).After the adsorption, the characteristic peak of -OH was shifted to 530.38 eV, and the characteristic peak at 531.84 eV corresponded to the formation of a -OH and metal ion ligand bond which may indicate the complexation of -OH with La(III) and Ce(III) after adsorption[60].

Fig. 6.

XPS spectra of Y. lipolytica W29 before and after La(III)/Ce(III) biosorption (G-L): A Survey spectrum, B La 3d, (C) Ce 3d, D C 1 s, E N 1 s, and F O 1 s

Transcriptomic analysis of Y. lipolytica W29 under La(III)/Ce(III) stress

DEGs under La(III)/Ce(III) stress

To investigate the molecular response mechanisms of Y. lipolytica W29 to La(III) and Ce (III) stress, cells cultured in medium containing 400 mg/L La(III), 400 mg/L Ce(III), and control medium (three groups: La(III)-treated, Ce(III)-treated, and control) were harvested during mid-log phase for RNA-seq analysis. The screening results of differentially expressed genes (DEGs) are shown in Fig. 7. The expression of some genes was changed in Y. lipolytica W29 under either La(Ⅲ) or Ce(Ⅲ) stress, and the number of up-regulated DEGs was greater than that of down-regulated DEGs under both rare earth ion stresses (Fig. 7A, B). With |log₂ (FoldChange)|≥ 1 and p adjust < 0.01 as the threshold, 2034 and 243 DEGs were screened under La(Ⅲ) and Ce(Ⅲ) stress conditions, respectively. 1405 DEGs were up-regulated and 629 DEGs were down-regulated under La(Ⅲ) stress, whereas 133 DEGs were up-regulated and 110 DEGs were down-regulated under Ce(Ⅲ) stress. In addition, 182 DEGs were identical among those screened under both La(III) and Ce(III) stress conditions (Fig. 7D). The log₂(fold change) and adjusted p-value for all DEGs are visually presented in Supplementary Fig. S1 and Supplementary Fig. S2, respectively. The above data demonstrate that exposure to environments containing La(III) and Ce(III) leads to changes in the gene expression levels of Y. lipolytica W29. Moreover, the differentially expressed genes (DEGs) identified under the two rare earth ion treatments are not entirely identical, suggesting that cells employ distinct detoxification mechanisms in response to different rare earth ion stresses.

Fig. 7.

Transcriptomic analysis and RT-qPCR validation of DEGs in Y. lipolytica W29 under La(III) and Ce(III) stress: A Volcano plot of DEGs under Ce(III) stress, B volcano plot under La(III) stress, C statistics of up- and down-regulated DEGs, D Venn diagram showing overlapping and specific DEGs between La(III) and Ce(III) treatments, E RT-qPCR validation of selected DEGs under La(III) stress, F RT-qPCR validation under Ce(III) stress

To verify the reliability of the RNA-seq data under La(III)/Ce(III) stress, ten differentially expressed genes (DEGs) were randomly selected from each of the La(III) and Ce(III) treatment groups for RT-qPCR validation, including five up-regulated and five down-regulated DEGs per group. The validation results are shown in Fig. 7E, F. The results demonstrated a high correlation between the RT-qPCR data and the RNA-seq data for the validated genes under both La(III) and Ce(III) stress, confirming the reliability of the RNA-seq data.

Functional enrichment analysis of DEGs: GO and KEGG

Functional enrichment analysis of the DEGs under La(III)/Ce(III) stress was performed using GO and KEGG analyses. Under La(III) stress, 21 GO terms were significantly enriched, including 5 molecular function terms, 10 biological process terms and 5 cellular component terms, The most highly enriched terms were transport activity, organic acid metabolic process, and membrane components (Fig. 8A). Under Ce (III) stress, 19 GO terms showed significant enrichment, distributed as: 4 molecular function terms, 9 biological process terms and 6 cellular component terms. Prominently enriched terms included transport activity, cell cycle processes, and small molecule catabolic processes (Fig. 9A).

Fig. 8.

GO and KEGG pathway enrichment analysis of DEGs under La(III) stress. A GO enrichment analysis of DEGs, B KEGG pathway enrichment of upregulated genes under La(III) stress, (C) KEGG pathway enrichment of downregulated genes under La(III) stress

Fig. 9.

GO and KEGG pathway enrichment analysis of DEGs under Ce(III) stress. A GO enrichment analysis of DEGs, B KEGG pathway enrichment of upregulated genes under Ce(III) stress, C KEGG pathway enrichment of downregulated genes under Ce(III) stress

KEGG pathway analysis of upregulated genes under La(III) stress revealed significant enrichment in 12 pathways, with the most prominent being: SNARE interactions in vesicular transport, Peroxisome, MAPK signaling pathway and Protein processing in endoplasmic reticulum(Fig. 8B). Conversely, downregulated genes were enriched in 14 pathways, most notably: Glycolysis/Gluconeogenesis, Citrate cycle (TCA cycle), Fatty acid biosynthesis, Arginine and proline metabolism, Tryptophan metabolism (Fig. 8C). Under Ce(III) stress, the KEGG pathways enriched by DEGs showed distinct patterns compared to the La(III) treatment group. Upregulated genes were significantly enriched in 6 pathways, with the most prominent being: Cell cycle, and MAPK signaling pathway (Fig. 9B). Conversely, Downregulated genes demonstrated enrichment in 14 pathways, primarily involving: Alanine, aspartate and glutamate metabolism, Tyrosine metabolism and Glycolysis (Fig. 9C).

Response mechanisms of Y. lipolytica W29 to La(III)/Ce(III) stress

Based on the KEGG enrichment results described above, this study further analyzes the major enriched metabolic pathways and functions of DEGs to elucidate the response mechanisms of Y. lipolytica W29 to La(III)/Ce(III) stress.

La(III)/Ce(III)-induced disruption of energy metabolism

The glycolytic pathway (EMP) and tricarboxylic acid cycle (TCA cycle) serve as the primary metabolic pathways in microorganisms, whose coordinated operation provides essential materials and energy for cellular activities(Ryan and O'Neill, 2020). KEGG enrichment analysis revealed that both La(III) and Ce(III) stress induced downregulation of the EMP pathway, with relevant DEGs listed in Table S4. Under La(III) stress, multiple genes encoding key glycolytic enzymes were downregulated, including: 6-Phosphofructokinase 1 (PFK1), Fructose-1,6-bisphosphate aldolase (FBA), Phosphoglycerate kinase (PGK) and Phosphoglycerate mutase (PGM). Previous studies indicate that toxic ions at high concentrations disrupt glycolysis. For instance, Rhodotorula mucilaginosa exposed to high-concentration Pb(II) stress showed downregulation of key glycolytic enzymes, leading to intracellular ATP deficiency via impaired glycolysis[12].

The TCA cycle (tricarboxylic acid cycle) is a fundamental metabolic pathway present in all living organisms, playing a pivotal role in nutrient oxidation and energy production. Moreover, it serves as the core hub for organic acid metabolism in Y. lipolytica [38]. La(III) exposure induced downregulation of the TCA cycle pathway, with corresponding DEGs documented in Table S4. Specifically, the following key enzyme-encoding genes showed significant downregulation: Citrate synthase (CS, YALI1_D32268g), Isocitrate dehydrogenase (IDH, YALI1_F06197g), Fumarate hydratase (FH, YALI1_C09033g), Malate dehydrogenase (MDH, YALI1_E17214g). These enzymes play essential roles in maintaining proper TCA cycle function. Previous research has demonstrated that under metal stress conditions, downregulation of CS, FH and MDH genes leads to suppressed production of critical TCA cycle intermediates including citrate, malate and oxaloacetate [37]. Studies have demonstrated that under high-concentration Cu(II) stress, the expression levels of multiple TCA cycle key proteins in Penicillium janthinellum were significantly reduced. This suppression led to insufficient intracellular ATP for sustaining normal growth, consequently resulting in enhanced Cu(II) sensitivity in the fungal strain [18]. The observed downregulation of multiple TCA cycle key enzyme genes indicates that La(III) similarly disrupts the TCA cycle in Y. lipolytica. This disturbance leads to impaired cellular ATP homeostasis, ultimately resulting in inhibited fungal growth.

In summary, both La(III) and Ce(III) stress suppress the expression of key enzyme genes in Y. lipolytica's primary energy metabolic pathways (including the glycolytic pathway and TCA cycle), thereby inducing cellular energy metabolism dysregulation. The resulting energy imbalance leads to insufficient growth efficiency, representing one of the fundamental causes for the observed biomass reduction under La(III)/Ce (III) stress.

Oxidative stress induced by La(III)

The peroxisome pathway plays a pivotal role in diverse cellular metabolic processes through catalytic oxidation reactions within peroxisomes, serving as a crucial metabolic conversion route in cells. Studies have demonstrated that multiple key enzymes in this pathway participate in counteracting metal ion-induced oxidative stress and toxicity effects [58]. Multiple rare earth ions can induce oxidative stress. For instance, exposure to ytterbium (Yb) triggers upregulation of 18 oxidative stress-responsive genes in Saccharomyces cerevisiae20. In this study, under La(III) stress, the peroxisome pathway was significantly upregulated, with relevant DEGs listed in Table S5. Key upregulated genes included: Mitochondrial inner membrane protein MPV17 (YALI1_C32823g), Catalase (CAT, YALI1_F38629g), Superoxide dismutase (SOD, YALI1_E14988g). Studies have reported that exposure to high concentrations of rare earth ions elevates intracellular ROS levels [35]. Upregulation of the Mpv17 gene enhances ROS scavenging capacity. Furthermore, both SOD (superoxide dismutase) and CAT (catalase) have been demonstrated to play critical roles in counteracting ROS-mediated toxicity induced by environmental stress. Under oxidative stress conditions, the levels of these two enzymes are typically upregulated. According to Kolhe et al. [30], Y. lipolytica under uranium (U) stress exhibits elevated levels of both SOD and CAT, which are proposed to enhance cellular defense against ROS overproduction during oxidative stress.

In summary, La(III) stress induces intracellular ROS generation, leading to oxidative stress. However, upregulation of genes encoding MPV17, SOD, and CAT provides antioxidant defense mechanisms to counteract this stress.

Impacts of La(III) and Ce(III) on amino acid metabolism

Disruption of amino acid metabolism can destabilize cellular metabolic homeostasis, impair normal growth and development, and even induce cell death [2]. KEGG enrichment analysis revealed that both La(III) and Ce(III) stress downregulated distinct amino acid metabolic pathways, with relevant DEGs listed in Table S6. Under La(III) stress, Arginine/proline metabolism and Tryptophan metabolism pathways were down-regulated. Under Ce(III) stress, Alanine, aspartate and glutamate metabolism and Tyrosine metabolism pathways were down-regulated. Prior studies have documented similar metal-induced disturbances in yeast amino acid metabolism. For example, the Rhodotorula mucilaginosa exhibited disrupted arginine/proline metabolism, aspartate/glutamate metabolism, and arginine biosynthesis under Cd stress [64].

Under La(III) stress conditions, significant downregulation was observed in several key metabolic genes: YALI1_B12993g encoding 1-pyrroline-5-carboxylate dehydrogenase (P5CDH) involved in glutamate synthesis, YALI1_B12966g encoding proline dehydrogenase (PRODH), and YALI1_C05908g encoding ornithine aminotransferase (OAT). The decreased expression of these enzymes, particularly PRODH, leads to proline accumulation which serves crucial protective functions. Proline acts as an effective hydroxyl radical scavenger, reducing oxidative damage through mitigation of lipid peroxidation and DNA lesions, thereby enhancing cellular metal tolerance. This mechanism is supported by previous findings in Saccharomyces cerevisiae, where PRODH knockout strains exhibited significantly higher survival rates and proline accumulation compared to wild-type when exposed to cadmium and arsenic stress. Similarly, in Y. lipolytica under La(III) exposure, the downregulation of PRODH gene expression results in elevated intracellular proline levels that contribute to improved antioxidant defense capacity against metal-induced oxidative stress.

In this study, exposure to Ce(III) stress resulted in the downregulation of gene YALI1_F33681g encoding succinic semialdehyde dehydrogenase (SSADH). In Saccharomyces cerevisiae, certain nitrogen sources are intracellularly converted by transaminases into succinic semialdehyde (SSA) and glutamate. However, SSA exhibits cellular toxicity, while SSADH rapidly catalyzes its conversion to succinate [64]. The observed downregulation of SSADH-encoding genes provides further evidence of nitrogen metabolism disruption. Additionally, aspartate serves as a precursor for threonine, methionine, lysine, pyrimidines, pantothenate, and nicotinamide adenine dinucleotide (NAD), supplying essential substrates for protein and nucleic acid synthesis. Alanine not only participates in the biosynthesis of various nitrogen-containing compounds but also contributes to cell wall formation. Notably, tyrosine metabolism has been reported to regulate fungal hyphal growth. Under high-concentration Cd stress, inhibited mycelial growth in Stropharia rugosoannulata was associated with reduced activity of key proteins in tyrosine metabolism [16]. The downregulation of genes related to alanine, aspartate, and glutamate metabolism pathways, along with tyrosine metabolism under Ce(III) stress, collectively leads to: (1) nitrogen metabolic disorders, (2) cellular membrane damage, and (3) suppression of hyphal growth.

ABC transporters and vesicle-mediated La(III)/Ce(III) efflux mechanisms

Under both La(III) and Ce(III) stress conditions, the gene encoding ATP-binding cassette transporter ABCB1 (ABC) was significantly upregulated (Table S7). ABC transporters play a crucial role in fungal resistance to metal toxicity by mediating the efflux of metal ions, thereby enhancing cellular metal tolerance[8]. Previous studies have demonstrated that ABCB1 in Arabidopsis thaliana participates in the efflux of Cd(II) and Pb(II), contributing to improved tolerance against these metal ions[53]. These findings collectively suggest that under La(III)/Ce(III) stress, cells can regulate the expression of ABC transporter genes to facilitate the efflux of intracellular rare earth ions, consequently alleviating their cytotoxic effects.

KEGG enrichment analysis revealed that under La(III) stress, pathways associated with SNARE interactions in vesicular transport were significantly upregulated, with relevant DEGs listed in Table S7. SNARE proteins (Soluble NSF Attachment Protein Receptor) serve as crucial mediators in vesicle trafficking and membrane fusion, facilitating specific vesicle-target membrane binding and fusion processes essential for intracellular secretion, material transport, and endo/exocytosis[1]. Notably, several genes encoding key vesicle fusion-related proteins exhibited upregulation, including: YALI1_E21203g (syntaxin 8, STX8), YALI1_B14441g and YALI1_D34232g (syntaxin 1, STX1), along with YALI1_A03487g and YALI1_E00963g (vesicle-associated membrane protein 4, VAMP4) which participates in synaptic vesicle docking to presynaptic membranes. Vesicle-mediated transport represents a primary cellular response to lanthanide toxicity in yeast. As reported by Roger et al.[20], Saccharomyces cerevisiae knockout strains with impaired genes involved in vesicle-Golgi fusion, intra-Golgi trafficking, and cytoplasm-to-vacuole targeting demonstrated enhanced sensitivity to high-concentration La(III) compared to wild-type strains. Both Golgi apparatus and vacuoles function as biologically significant cation reservoirs and transient containers prior to metal exocytosis. These observations support a proposed detoxification mechanism whereby intracellular La(III) is transported via vesicles to the Golgi for storage/processing, followed by vesicle-mediated exocytosis to extrude La(III) from the cell, thereby reducing intracellular La(III) toxicity.

In summary, to counteract La(III) toxicity, Y. lipolytica enhances vesicle-mediated La(III) transport, consequently increasing the demand for key transport proteins. This likely explains the upregulation of multiple genes associated with vesicle-mediated trafficking. ABC transporter-mediated efflux and vesicle-facilitated excretion may constitute crucial detoxification mechanisms against La(III) in this yeast.

Regulation of dimorphic transition in Y. lipolytica by MAPK signaling pathway in response to La(III)/Ce(III) stress

Under exposure to either La(III) or Ce(III) stress, the mitogen-activated protein kinase (MAPK) signaling pathway was significantly upregulated, with relevant genes listed in Table S8. As a crucial eukaryotic pathway for extracellular signal perception, the MAPK cascade enables cellular response to environmental stimuli through activation of specific transcription factors that subsequently regulate genes involved in stress adaptation and stimulus sensing [28]. Current studies have demonstrated the pivotal role of MAPK signaling in controlling dimorphic transition (yeast-to-hypha switching) in Y. lipolytica [46].

Under La(III) stress, the MAPK signaling pathway exhibited upregulation of gene YALI1_E27697g encoding Mitogen-activated protein kinase Fus3. Previous studies have demonstrated that Fus3 promotes filamentous growth in Saccharomyces cerevisiae through phosphorylation-mediated activation of transcription factors that induce hypha-specific gene expression [46]. Conversely, Ce(III) exposure triggered upregulation of YALI1_F22241g encoding the actin cytoskeletal protein Spa2, which serves as the primary scaffold of the polarisome multiprotein complex. This complex coordinates actin polymerization by recruiting other actin-binding proteins upon receiving dimorphic transition signals, thereby facilitating hyphal development in yeasts. As established earlier, both La(III) and Ce(III) induce dimorphic transition in Y. lipolytica. The observed upregulation of filamentation-related genes associated with the MAPK pathway under both rare earth stresses suggests that these ions function as signaling molecules, activating MAPK-mediated cascades to trigger the yeast-to-hypha morphological shift.

In summary, the response mechanisms of Y. lipolytica to La(III)/Ce(III) stress are systematically summarized (Fig. 10). Under La(III) exposure: (1) suppressed expression of key enzyme genes in glycolysis and TCA cycle disrupts organic acid synthesis and causes energy metabolism dysfunction; (2) upregulation of peroxisomal pathway genes (e.g., SOD, CAT) and proline metabolism-related genes enhances oxidative stress resistance by mitigating ROS toxicity; and (3) ABC transporters and vesicle-mediated efflux mechanisms reduce intracellular La(III) accumulation, thereby improving tolerance. Conversely, Ce(III) stress similarly inhibits glycolytic enzymes to impair energy metabolism, while downregulation of genes in alanine/aspartate/glutamate metabolism and tyrosine metabolism pathways leads to nitrogen metabolic disorders, cell membrane damage, and delayed hyphal growth. Notably, both La(III) and Ce(III) function as signaling molecules that induce dimorphic transition in Y. lipolytica through MAPK pathway activation.

Fig. 10.

Mechanisms of Y. lipolytica in response to La(III)/Ce(III) stresses

DNA methylome analysis of Y. lipolytica W29 under La(Ⅲ)/Ce(Ⅲ) stress

Whole-genome DNA methylation analysis of Y. lipolytica W29 under La(Ⅲ) /Ce(Ⅲ) stress

To further analyze changes in DNA methylation under La(III) or Ce(III) stress, whole-genome bisulfite sequencing (WGBS) was performed on the same three experimental groups as the transcriptome analysis: La(III)-treated group, Ce(III)-treated group, and control group. The whole-genome DNA methylation analysis revealed average DNA methylation levels of 0.13%, 0.09%, and 0.11% in the control, La(III)-treated, and Ce(III)-treated groups, respectively, consistent with the reported range of 0.014%-0.34% for yeast whole-genome DNA methylation levels[52]. Compared with the control, DNA methylation levels decreased following both La(III) and Ce (III) treatments. Furthermore, the DNA methylation levels in CG, CHG, and CHH sequence contexts were globally lower in both treatment groups than in the control (Fig. 11A). Under La(III) stress, methylation decreased by 36.26%, 30.76%, and 30.76% in the CG, CHG, and CHH contexts, respectively, while under Ce(III) stress, the reductions were 18.18%, 15.38%, and 15.38% (Fig. 11B). All three groups exhibited similar distribution patterns across CG, CHG, and CHH contexts, indicating that La(III)/Ce(III) stress did not alter the distribution patterns of methylation sequence contexts. Methylation was consistently biased toward the CHH context in all groups. Further analysis of methylation levels in mCG, mCHG, and mCHH contexts (Fig. 11C) revealed predominantly low methylation levels (< 10%) across all sequence contexts. However, the proportion of CHH sites with higher methylation levels (10%-20%) increased under both La(III) and Ce(III) stress. Combined with the observed reduction in the number of low-methylated CHH sites under stress conditions, these results collectively demonstrate that La(III) and Ce(III) stress reduces the overall cytosine methylation levels in Y. lipolytica W29, with demethylation mainly occurring in the CHH sequence context.

Fig. 11.

Effects of La(III)/Ce(III) Stress on Whole-Genome DNA Methylation Levels, A Methylation levels in CG, CHG, and CHH sequence contexts, B Number of methylated cytosine sites in CG, CHG, and CHH sequences, C Methylation level distribution of cytosines in CG, CHG, and CHH contexts, D Methylation levels across gene bodies and flanking 2-kb upstream/downstream regions

To further investigate the distribution patterns of DNA methylation across distinct genomic regions, this study partitioned coding gene sequences (including gene bodies and their flanking 2-kb upstream/downstream regions) into four distinct transcriptional element regions. Average cytosine methylation levels were analyzed across these regions to elucidate the functional roles of DNA methylation modifications at the whole-genome level. As shown in Fig. 11D, DNA methylation levels across all three sequence contexts (CG, CHG, CHH) declined sharply downstream of the transcription start site (TSS), a pattern unaffected by La(III) or Ce(III) stress. Furthermore, only in the CG context was methylation significantly lower within gene bodies compared to flanking regions. In contrast, CHG and CHH contexts exhibited comparable methylation levels between gene bodies and adjacent regions.

Differential DNA Methylation in Y. lipolytica W29 under La(III)/Ce(III) Stress

Differentially Methylated Regions (DMRs) refer to genomic DNA segments exhibiting distinct methylation patterns between sample groups, which are widely recognized as functional elements regulating transcriptional activity [19]. Using Metilene with an adjusted p-value threshold of < 0.05, we identified DMRs under La(III) or Ce(III) stress, categorizing them as hypo-methylated (indicating decreased methylation) or hyper-methylated (indicating increased methylation). As shown in Fig. 12, La(III) stress induced 18 DMRs, comprising 13 hypo-methylated (72.22%) and 5 hyper-methylated DMRs (27.78%). Similarly, Ce(III) stress yielded 18 DMRs with 10 hypo-methylated (55.56%) and 8 hyper-methylated (44.44%) regions. These results demonstrate a predominance of hypo-methylated DMRs under both stress conditions, further confirming that La(III) and Ce(III) stress reduce global DNA methylation levels in Y. lipolytica W29. Notably, DMRs overlapping with gene bodies or 2-kb flanking regions were designated as DMR-associated genes (DMGs). We identified 41 and 46 DMGs under La(III) and Ce(III) stress, respectively.

Fig. 12.

Number of DMRs under La(III)/Ce(III) stress

Functional annotation analysis of DMGs under La(III)/Ce(III) stress via GO and KEGG

GO enrichment analysis (adjusted p-value < 0.05) of DMGs screened under La(III) /Ce(III) stress revealed significant functional categories (Fig. 13). Under La(III) stress, 11 GO terms were significantly enriched, including 6 biological processes (BP), 4 molecular functions (MF), and 1 cellular component (CC). The most prominently enriched terms included cellular secretion, oxidoreductase activity, and cytoplasm. For Ce(III) stress, 15 GO terms were enriched exclusively in BP (11 terms) and CC (4 terms), with organelle organization and cellular component organization exhibiting the highest enrichment.

Fig. 13.

GO enrichment analysis of DMGs under La(III) (A) and Ce(III) (B) stress

To further characterize DMG functions, KEGG pathway enrichment analysis was performed (adjusted p-value < 0.05). As summarized in Table S9, La(III) stress significantly enriched 5 pathways, most notably the MAPK signaling pathway, cell cycle regulation, and autophagy. Under Ce(III) stress, 6 pathways were enriched, with TCA cycle, cysteine and methionine metabolism, pyruvate metabolism, and endocytosis showing the strongest enrichment signals.

Integrated analysis of transcriptomics and DNA Methylomics in Y. lipolytica W29 under La(III)/Ce(III) stress

Leveraging transcriptomic data from previous sections, we performed correlation analysis between DEGs and DMGs) screened under La(III) and Ce(III) stress. As shown in Table S10, 10 and 4 genes exhibiting significant correlations between DNA methylation and gene expression were identified under La(III) and Ce(III) stress, respectively. In these correlated genes, decreased methylation levels predominantly occurred in upstream gene regions. This prevalent hypomethylation suggests that demethylation—particularly in promoter regions—is a primary epigenetic response to La(III)/Ce(III) stress. Similar metal-stress-induced demethylation mechanisms have been reported in Triticum aestivum and Triticosecale[20]. Furthermore, methylation changes in upstream regions exhibited an inverse correlation with gene expression (i.e., hypomethylation associated with upregulated expression). Conversely, gene-body methylation changes showed a positive correlation with expression levels. This regulatory pattern aligns with established models: hypermethylation in promoter/transcription start site (TSS) regions typically suppresses transcription, while hypomethylation enhances it; conversely, gene-body hypermethylation promotes expression[15]. Collectively, these findings indicate that Y. lipolytica W29 likely activates stress-responsive gene expression primarily through promoter demethylation to counteract La(III)/Ce(III) toxicity.

Integrated functional analysis revealed that under La(III) stress, gene YALI1_B05295g (encoding methionine sulfoxide reductase, Msr) and YALI1_F11481g (encoding a mitochondrial inner membrane protease) exhibited decreased methylation with concomitant upregulation of expression. Msr catalyzes the reduction of methionine sulfoxide to methionine, repairing oxidatively damaged proteins and playing a critical role in countering oxidative stress induced by environmental stressors. Notably, Saccharomyces cerevisiae Msr knockout mutants show enhanced sensitivity to chromium (Cr), establishing Msr as a key enzyme for eukaryotic Cr-stress tolerance [50]. The mitochondrial inner membrane protease serves as a key regulator of mitochondrial function, maintaining intramembrane proteostasis and mitochondrial dynamics. Plant studies demonstrate its activation confers aluminum (Al) resistance [44]. Under Ce(III) stress, gene YALI1_C09033g (encoding fumarate hydratase, FH) similarly showed hypomethylation and transcriptional upregulation. FH, a pivotal TCA cycle enzyme, causes oxidative stress when inhibited due to disrupted metabolic flux. Y. lipolytica upregulates FH activity under low-pH stress to mitigate oxidative damage through metabolic reprogrammin [45]. Intriguingly, FH also participates in DNA repair: nuclear translocation occurs upon double-strand breaks, where it catalyzes malate-to-fumarate conversion, though its precise DNA repair mechanism remains elusive.

Collectively, YALI1_B05295g (Msr), YALI1_F11481g (protease), and YALI1_C09033g (FH) represent key La(III)/Ce(III) resistance genes in Y. lipolytica W29. Their stress-responsive methylation changes indicate DNA methylation modulates rare-earth tolerance by regulating these resistance genes. Further investigation is warranted to elucidate the detailed epigenetic regulatory mechanisms.

Conclusion

Y. lipolytica demonstrates significant potential as a sustainable biosorbent for REE bioremediation, exhibiting high tolerance and efficient adsorption capacity toward La(III) and Ce(III) ions. Under REE stress, the strain undergoes a dimorphic transition that increases its specific surface area and enhances adaptive resilience—a critical trait for real-world remediation scenarios. Through systematic process optimization, we identified ideal adsorption conditions (pH 7.0, 100 mg/L initial ion concentration, 2000 mg/L biomass dosage), achieving removal efficiencies of 84.33% for La(III) and 87.21% for Ce(III). Mechanistic studies reveal that adsorption is driven by chemisorption (pseudo-second-order kinetics,) and monolayer surface coverage (Langmuir model). Critically, FTIR and XPS analyses confirm that surface hydroxyl groups act as core active sites, facilitating REE binding via ligand complexation. This synergistic combination of robust stress adaptation, high removal efficiency, and well-defined adsorption mechanisms positions Y. lipolytica as a promising biocatalytic platform for REE recovery.

Multi-omics analysis revealed the mechanism by which Y. lipolytica maintains its function under rare earth element stress. Transcriptomics reveals the molecular mechanisms underlying the stress response. Under La(III) stress, the downregulation of genes encoding key enzymes in the glycolytic pathway and the TCA cycle interferes with the synthesis of various organic acids and triggers cellular energy metabolism disorders. Concurrently, the upregulation of genes associated with the peroxisome pathway and proline metabolism is linked to the cell's defense against La(III)-induced oxidative stress. Furthermore, ABC transporters and vesicles are capable of mediating the efflux of excess La(III) from the cell to maintain homeostasis. Under Ce(III) stress, the downregulation of related genes in the glycolytic pathway and the metabolism of multiple amino acids, such as glutamate, leads to disturbances in cellular energy and nitrogen metabolism, cell membrane damage, and slow hyphal growth. Additionally, under both stress conditions, the upregulation of filamentation genes within the MAPK signaling pathway indicates that La(III) or Ce(III) can act as signaling molecules to induce the dimorphic transition in Y. lipolytica.

Genome-wide methylation analysis showed that DNA methylation levels decreased by 30.76% and 15.38% under La(III) and Ce(III) stress, respectively. And the DNA methylation were biased to occur in CHH sequence. 18 DMRs were screened under both La(III) and Ce(III) stress, which contained 41 and 46 DMGs, respectively. Functional enrichment analysis showed that more DMGs under La(III) stress were associated with pathways such as the MAPK signaling pathway, whereas more DMGs under Ce(III) stress were involved in pathways such as the TCA cycle.

Joint analysis of DNA methylomics and transcriptomics showed that 10 and 4 genes with significant differences in gene expression levels accompanied by altered methylation levels were screened under La(III) and Ce(III) stress, respectively. The methylation levels of these genes in the upstream region showed a negative correlation with the expression changes, whereas a positive correlation was observed in the gene body. In addition, among the relevant genes screened, the gene functions of YALI1_B05295g, YALI1_F11481g found under La(III) stress, and YALI1_C09033g found under Ce(III) stress were all related to responding to oxidative stress and were key resistance genes, so it was hypothesized that DNA methylation was involved in regulating the expression of La(III)/Ce(III) resistance-related gene expression, which in turn improves the tolerance of Y. lipolytica to La(III)/Ce(III).

Y. lipolytica initially adsorbs La(III)/Ce(III) through surface functional group complexation. Subsequently, ABC transporters and vesicles efflux excess intracellular La(III)/Ce(III) to prevent cellular metabolic disruption. Finally, the strain optimizes long-term stress adaptation by regulating antioxidant genes—such as upregulating the expression of the hypomethylated gene YALI1_B05295g.

Author contributions

QHF and LJQ contributed equally to this study and should be considered co-first authors. QHF and LJQ wrote the main manuscript text and completed the experiments. TDH drew Figs. 1, 2, and 3 based on the data. LYT and WHY searched for some materials. YY and ZHB provided guidance on the paper.

Funding

This work was supported by the Hunan Provincial Department of Education Key Project [24A0009]; and National Natural Science Foundation of China [project No. 52222406].

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Competing interests

The authors declare no competing interests.