Abstract Candida auris is a fungal pathogen with frequent development of multidrug-resistance or pan-drug resistance. Currently, the treatment options for Candida auris are limited. Therefore, there is an urgent need for alternative therapeutic strategies. Antimicrobial photodynamic therapy (aPDT), which generates reactive oxygen species (ROS) through light-activated photosensitizers, has shown promise against C. auris; however, its molecular mechanism remains unclear. To investigate COP1T-HA-mediated PDT-induced genomic alterations, we constructed a 3D genome map of Candida species, which uncovered the reorganization of chromatin architecture in response to PDT treatment. Our data showed that low-dose PDT causes subtle local adjustments in chromatin topology, whereas high-dose PDT leads to more pronounced changes in A/B compartmentalization, topologically associating domain (TAD) organization, and chromatin looping associated with key genes related to mitochondrial energy metabolism. Confocal imaging confirmed that high-dose COP1T-HA-mediated PDT induces localized ROS accumulation near the nucleus and a temporally ordered cellular stress response. Furthermore, functional validation through QCR10, NDUFA5, and MP knockouts confirmed the essential roles of these genes in mitochondrial integrity, ATP synthesis, ROS homeostasis, and biofilm formation. Mutants showed altered mitochondrial membrane potential, intracellular pH imbalance, and enhanced glycolytic compensation, highlighting the impact of electron transport disruption on energy metabolism. This study provides the first comprehensive insight into COP1T-HA-mediated PDT-induced chromatin reorganization in C. auris and establishes a direct connection between 3D genome remodeling and fungal energy metabolism, offering a foundation for chromatin-targeted antifungal strategies. Keywords: Photodynamic therapy (PDT), 3D genome architecture, Chromatin remodeling, Mitochondrial energy metabolism, Multidrug-resistant Candida auris Graphical abstract Photodynamic therapy (PDT)-induced chromatin remodeling and mitochondrial dysfunction in drug-resistant Candida auris. (a) The established antimicrobial mechanism of PDT, where light-activated photosensitizers generate reactive oxygen species (ROS), leading to an oxidative burst and fungal cell death. (b) The proposed mechanism from this study, showing PDT-induced chromatin remodeling in C. auris. Changes in A/B compartments, TADs, and chromatin loops affect gene regulation on energy metabolism, disrupting the mitochondrial respiratory chain, increasing ROS accumulation, reducing ATP production, and ultimately triggering cell apoptosis. Image 1 [41]Open in a new tab Highlights * • Multidrug- or pan-drug resistant Candida auris is highly resistant to current antifungals, underscoring the urgent need for new treatments; although photodynamic therapy (PDT) shows broad-spectrum activity, its specific mechanism remains unclear and limits clinical use. * • This study is the first to construct a three-dimensional genome map of C. auris (indeed, the first for any Candida species), revealing its chromatin reorganization under different PDT doses and offering insights into PDT's antimicrobial action. * • PDT induces dose-dependent 3D genome changes: low doses cause subtle local chromatin adjustments, while high doses trigger A/B compartment shifts, topologically associating domains (TAD) reorganization, and major alterations in chromatin loops linked to metabolic genes. * • Following PDT treatment, restructured genomic regions are enriched with key mitochondrial energy metabolism genes, and topology changes in QCR10, NDUFA5, and MP coincide with marked transcriptional downregulation. * • Gene knockout experiments confirm the importance of these PDT-affected genes; deleting QCR10, NDUFA5, or MP in C. auris leads to mitochondrial dysfunction, reduced ATP production, and growth defects, highlighting their critical role in responding to PDT-induced stress. * • Our work proposes a novel antifungal strategy targeting 3D genome reorganization, offering a new approach for treating drug-resistant C. auris and other pathogenic fungi, while also shedding light on PDT mechanisms in tumors and other superficial diseases. 1. Introduction Fungal infections represent a significant global health threat, leading to an estimated 3.75 million deaths annually [[42]1]. Among the various pathogens responsible for these infections, Candida species are the most prevalent, with Candida auris emerging as the most concerning one due to its resistance to multiple antifungal drugs [[43]2,[44]3]. First identified in 2009, C. auris has rapidly spread across more than 40 countries, causing outbreaks in healthcare settings [[45]4,[46]5]. With mortality rates reaching as high as 50 % in some cases [[47]6], C. auris presents a serious challenge to current therapeutic options, which are limited to azoles, echinocandins, and polyenes—drugs that target specific molecular pathways such as ergosterol synthesis and β-glucan biosynthesis. However, the overuse of these agents has led to drug resistance, especially in C. auris, which underscores the urgent need for novel antifungal therapeutics [[48]7,[49]8]. Antimicrobial photodynamic therapy (aPDT) is a promising alternative treatment for fungal infection. In aPDT, photosensitizers are activated by light to generate reactive oxygen species (ROS), which cause oxidative damage to multiple cellular components [[50]9,[51]10]. Unlike traditional antifungals, which typically target a single molecular pathway, aPDT induces broad-spectrum oxidative damage, significantly reducing the likelihood of development of resistance to antifungal drugs [[52]11]. In addition, aPDT has shown promising results in treating drug-resistant pathogens, such as Candida albicans, C. auris and even bacterial species like Pseudomonas aeruginosa and Staphylococcus aureus, demonstrating its potential to overcome drug resistance in a variety of microorganisms [[53][12], [54][13], [55][14]]. Despite these promising outcomes, the molecular mechanisms by which aPDT exerts its antifungal effects remain poorly understood, particularly its impact on genomic architecture and gene expression. COP1T-HA is a novel photosensitizer synthesized by our group [[56]11], structurally based on hypocrellin A (HA), modified via conjugation to a covalent organic polyhedron (COP1T) and polyethylene glycol (PEG). Previously, COP1T-HA exhibited significantly enhanced antifungal efficacy against multidrug-resistant Candida spp., including effective disruption of biofilms. Despite the demonstrated mitochondrial damage and apoptosis induction, detailed intracellular localization, sites of oxidative damage, and their temporal sequence have not been previously clarified. Recent research on chromatin organization has illuminated its critical role in regulating gene expression, especially under stress conditions such as drug treatments or oxidative damage [[57]15]. Eukaryotic genomes are not simply linear sequences of DNA; instead, they are organized into a dynamic 3D structure that facilitates the spatial organization of genes and regulatory elements [[58]16,[59]17]. This 3D architecture includes large-scale chromosomal compartments, topologically associating domains (TADs), and local chromatin loops [[60]18]. Chromosomal compartments are regions of genomes with specific transcriptional properties, with “active” compartments containing genes that are more likely to be transcribed. TADs are smaller units within compartments where chromatin segments interact more frequently with each other than with regions outside the TAD, promoting coordinated gene expression. Local chromatin loops further refine gene regulation by bringing distant enhancers and promoters into close proximity, facilitating the activation of target genes [[61]19,[62]20]. These structural features play a pivotal role in regulating gene expression by enabling or hindering the interactions between genes and their enhancers, transcription factors, and other regulatory elements. For example, in mammalian cells, exposure to oxidative stress or drug treatments can lead to significant reorganization of TADs and chromatin loops, resulting in the activation or silencing of genes involved in apoptosis, stress response, or cell cycle regulation [[63]21]. In Saccharomyces cerevisiae, chromatin rearrangements have been observed in response to heat shock and nutrient deprivation, suggesting that environmental stress can trigger large-scale chromatin reorganization to fine-tune gene expression in real time [[64]22,[65]23]. Furthermore, in Arabidopsis thaliana, chromatin loops and TADs have been shown to dynamically adjust in response to environmental stresses such as drought, influencing the expression of stress-responsive genes involved in water retention and metabolism [[66]24]. Additionally, in bacterial species such as Escherichia coli, studies have shown that oxidative stress can lead to rearrangements of chromatin-like structures, affecting gene expression related to DNA repair and stress response pathways [[67]25]. Despite these advances in chromatin research, the role of chromatin remodeling in response to aPDT-induced oxidative stress remains unexplored, especially in C. auris. Based on our previous findings that COP1T-HA mediated PDT disrupts mitochondrial structure and induces apoptosis in C. auris [[68]26], and according to our recent confocal microscopy data, COP1T-HA [COP1T-HA at high dose (HD-PDT, 3.125 μg/mL) and low dose (LD-PDT, 0.78 μg/mL) compared to the control group (equal volume of PBS), following previously reported effective concentrations [[69]26] was confirmed to enter C. auris cells in a dose-dependent manner. At low PDT doses, COP1T-HA fluorescence co-localized with mitochondria but did not overlap with DAPI-stained nuclei. At higher doses, COP1T-HA signals became partially co-localized with nuclear regions, though whether the compound enters the chromatin remains undetermined ([70]Fig. 1A and B). These observations provide cellular-level evidence that supports the rationale for investigating how photodynamically induced intracellular stress may influence higher-order chromatin organization. This study uses Hi-C and RNA-seq to investigate how COP1T-HA mediated PDT affects gene expression related to mitochondrial energy metabolism. We focus on changes in chromatin organization, particularly in compartments, TADs, and loops. Gene knockouts are performed to compare the energy metabolism phenotypes, gene expression, ultrastructural changes, pathogenicity, and mitochondrial function between wild-type and knockout strains. We present, to our knowledge, the first comprehensive 3D genomic map of C. auris and analyze how this architecture shifts under aPDT. This study aims to elucidate how PDT-induced chromatin remodeling influences the three-dimensional genome architecture of C. auris and its regulation on genes of energy metabolism. The results of this study will provide potential therapeutic targets for multidrug-resistant strains and other fungal pathogens. Fig. 1. [71]Fig. 1 [72]Open in a new tab COP1T-HA intracellular localization and the spatiotemporal sequence of PDT-induced oxidative and nuclear responses in Candida auris. (A) Confocal microscopy showing subcellular localization of COP1T-HA (red autofluorescence) in C. auris following low-dose (LD-PDT) or high-dose photodynamic treatment (HD-PDT). MitoTracker Green was used to label mitochondria and DAPI to stain nuclei. Scale bars represent 50 μm (B) Confocal co-staining of SOSG and Hoechst under different PDT conditions. Scale bars represent 50 μm (C) Fluorescence microscopy showing sequential changes in intracellular ROS production (SOSG), nuclear morphology (DAPI), and DNA fragmentation (TUNEL) after HD-PDT treatment. Scale bars represent 50 μm(SOSG), 25 μm(DAPI), 50 μm(TUNEL). 2. Results 2.1. Characterization of COP1T-HA intracellular localization and spatiotemporal cellular responses following PDT To clarify the initial sites of PDT-induced damage, confocal and fluorescence microscopy assessed the intracellular localization of COP1T-HA and temporal generation of ROS. Under LD-PDT, COP1T-HA fluorescence was strictly co-localized with mitochondrial markers (MitoTracker Green), without observable overlap with nuclear DAPI staining. At higher concentrations (HD-PDT), COP1T-HA showed additional partial co-localization with nuclear regions, although entry into chromatin could not be directly verified. To further evaluate the spatial relationship between ROS generation and nuclear regions, we performed confocal imaging using Singlet Oxygen Sensor Green (SOSG) together with Hoechst staining. In LD-PDT and HD-PDT groups, SOSG signals were clearly detected and appeared more intense in HD-PDT. Partial overlap between SOSG and Hoechst signals was observed, especially under HD-PDT, suggesting that ROS accumulated near or around the nucleus ([73]Fig. 1B). To investigate the temporal dynamics of these events, fluorescence microscopy was used to monitor SOSG, DAPI, and TUNEL signals over a 24-h period following PDT treatment. SOSG fluorescence became detectable as early as 1–3 h post-treatment, confirming the early onset of intracellular ROS generation. Nuclear changes, characterized by increased DAPI intensity and alterations in nuclear morphology, emerged at later time points (6–12 h). TUNEL-positive signals, indicating DNA fragmentation, were observed subsequently, peaking at 24 h. These results ([74]Fig. 1C) demonstrate a clearly ordered sequence of cellular responses—beginning with ROS production, followed by nuclear alteration, and culminating in DNA damage. 2.2. Changes in the three-dimensional genomic structure of C. auris with and without PDT To explore the changes in the spatial structure and interactions of chromosomes due to COP1T-HA mediated PDT, the 3D chromosome organization in muti-drug resistant C. auris CBS14918 with and without aPDT treatment were investigated through the high-throughput chromosome conformation capture (Hi-C) method. Because COP1T alone or dark conditions had no therapeutic effect, these control groups were not included in this study. The genome is divided into 5-kb bins, and interaction frequencies for each restriction fragment were assigned to corresponding bins. We systematically examined PDT-induced chromatin interaction changes, topological remodeling, and potential regulatory mechanisms. Three-dimensional genome modeling ([75]Fig. 2A, movies S1, S2, S3) showed that, under control conditions, chromosomes in C. auris exhibited a highly organized distribution. Centromeres cluster near the nuclear center, while telomeres appear closer to the nuclear periphery, forming a “Rabl-like” pattern similar to that in wild-type yeasts. After low-dose photodynamic therapy (LD-PDT), the overall chromatin folding remained largely unchanged, with only minor local rearrangements. In contrast, high-dose PDT (HD-PDT) led to pronounced structural reorganization across multiple chromosomes. Fig. 2. [76]Fig. 2 [77]Open in a new tab PDT-induced chromatin remodeling in C. auris. (A) 3D genome structures of C. auris under control, LD-PDT, and HD-PDT conditions. Chromosomes are color-coded, showing spatial organization and compaction differences in these conditions. (B) Hi-C contact maps at 10 kb resolution for control, LD-PDT, and HD-PDT conditions, illustrating chromatin interaction changes following PDT treatment. (C) Circos plot displaying A/B compartment shifts, TAD organization, and chromatin loop alterations across different conditions. (D) Differential chromatin interaction maps comparing control vs. LD-PDT (left) and control vs. HD-PDT (right), highlighting regions with significant changes in chromatin interactions. Warmer colors indicate increased interactions, while cooler colors indicate decreased interactions. Chromatin interaction matrices based on Hi-C data ([78]Fig. 2B) revealed chromosome territories, with intra-chromosomal interactions (on the diagonal) stronger than inter-chromosomal interactions (off-diagonal). After LD-PDT, the global interaction pattern stayed relatively stable, intra- and inter-chromosomal interactions both decreased slightly, suggesting minor chromatin decompaction with limited disruption to global genome architecture. After HD-PDT, interactions intensified in chromosomes 2, 4, and 6 but decreased in chromosome 3, indicating a region-specific effect of PDT on the 3D chromatin layout. Circos plot analysis ([79]Fig. 2C) further illustrated these topological shifts, capturing the distribution of A/B compartments, TADs, and loops. While the control group retained stable chromatin organization across multiple layers, LD-PDT caused minor readjustments in chromosomes 3, 4, 6 and 7. In contrast, HD-PDT generated substantial reorganization in chromosomes 2, 4, and 6, with altered loop signals, hinting deeper spatial rearrangements and potential gene regulatory effects. To clarify these differences, we computed chromosomal interaction changes ([80]Fig. 2D, [81]Fig. S1) by normalizing and comparing Z-score matrices between treated samples and controls. The LD-PDT vs. control comparison revealed localized changes in chromatin interactions, with certain regions showing increased interactions while others exhibited decreases, but overall, the impact remained limited. In contrast, the HD-PDT vs. control comparison showed more extensive modifications, with a broader range of regions displaying enhanced or reduced interaction frequencies, indicating a greater effect on chromatin organization. These outcomes suggest that PDT's impact on the 3D genome of C. auris is both dose-dependent and region-specific. 2.3. PDT induces region-specific A/B compartment transitions linked to metabolic regulation Chromatin A/B compartments are key architectural features of the 3D genome, with A compartments typically corresponding to open, transcriptionally active chromatin, and B compartments often harboring silenced genes. To determine whether PDT affects compartment organization in C. auris, we calculated compartment signals from Hi-C data, then examined associated genomic features and gene functions ([82]Fig. 3). Fig. 3. [83]Fig. 3 [84]Open in a new tab PDT-induced A/B compartment reorganization and functional implications in C. auris. (A) A/B compartment eigenvector heatmaps in chromosome 1 for control, LD-PDT, and HD-PDT. (B–D) Boxplots illustrating differences in eigenvector values (B), compartment A/B distribution (C), and compartment length distribution (D) across conditions, highlighting shifts in chromatin activity. (E) Pie charts depicting the number of genomic regions undergoing compartment transitions between control and PDT-treated samples, categorized as A2A (stable active), A2B (active-to-inactive), B2A (inactive-to-active), and B2B (stable inactive). (F) KEGG pathway enrichment analysis of genes located in regions undergoing compartment transitions. Hi-C–based mapping of A/B compartments ([85]Fig. 3A, [86]Fig. S2) revealed stable partitioning in the control group, reflecting a typical fungal genome arrangement. Under LD-PDT, the compartmental structure stayed largely intact, suggesting minimal impact on broad chromatin domains. However, HD-PDT led to more pronounced compartment changes compared to LD-PDT group, especially in chromosomes 2, 4, and 6, where certain regions shifted from B to A compartments, indicating a potential increase in local chromatin accessibility. Quantitative analyses of GC content ([87]Fig. 3B), gene density ([88]Fig. 3C), and compartment length ([89]Fig. 3D) demonstrated that A compartments had consistently higher GC content (∼0.46) and gene density than B compartments (∼0.44), with no significant global differences among treatment groups. Nonetheless, the localized B-to-A compartment switches seen under HD-PDT imply region-specific modifications rather than wholesale genomic rearrangements. To explore functional consequences, we tracked distinct types of compartment transitions ([90]Fig. 3E). In the Control vs. LD-PDT comparison, the most frequent events were A2A (499 bins) and B2B (93 bins), indicating that most compartments remained stable. Meanwhile, there was a small number of transitions (A2B and B2A) took place, suggesting limited remodeling at LD-PDT. The Control vs. HD-PDT group showed a high proportion of A2A and B2B events, but the observed A-to-B shifts (12 bins) indicate that HD-PDT may slightly reconfigure specific chromosomal regions to an inactive state. KEGG pathway analysis ([91]Fig. 3F–[92]Tables S1 and S2) revealed that genes undergoing these compartment transitions are significantly enriched in energy metabolism pathways, including the TCA cycle, oxidative phosphorylation, glycolysis, and pyruvate metabolism. Under LD-PDT, the impact primarily involved TCA cycle–related genes, whereas HD-PDT generated broader changes encompassing oxidative phosphorylation and RNA metabolic processes. Such findings suggest that PDT-driven compartment remodeling could influence mitochondrial function and metabolic adaptation in C. auris, fitting the notion that chromatin organization are associated with critical cellular pathways. 2.4. Dose-dependent effects of PDT on chromatin topology and TAD stability TADs are fundamental structural units of 3D chromatin organization, with intra-domain interactions and boundary stability playing a critical role in gene regulation. To examine whether PDT affects the TAD landscape in C. auris, we generated high-resolution chromatin interaction maps under control, LD-PDT, and HD-PDT conditions to identify changes in TAD architecture ([93]Fig. 4A, [94]Fig. S3). Fig. 4. [95]Fig. 4 [96]Open in a new tab PDT-induced TAD reorganization and its genomic features in C. auris. (A) Hi-C contact maps of chromosome 1 (Ch1) under control, LD-PDT, and HD-PDT conditions, with corresponding insulation scores indicating TAD boundary locations. (B–C) Boxplots comparing gene density (B) and GC content (C) between TAD boundaries and TAD interiors across different conditions, showing their distribution in relation to structural features. (D) Aggregate Hi-C maps showing TAD boundary enrichment across control, LD-PDT, and HD-PDT samples. (E) Pileup analysis of TAD domains under different conditions. (F) Boxplot comparing TAD boundary strength across conditions. Then, we evaluated gene density ([97]Fig. 4B) and GC content ([98]Fig. 4C) to examine whether PDT treatment altered basic genomic features. On average, TADs in C. auris contained about 4.8 genes, while TAD boundaries contained around 5.0 genes, without significant differences across treatments (NS). Similarly, GC content hovered at 0.46 inside TADs and 0.47 in boundary regions, again showing no statistical difference (NS) before or after PDT. These observations suggest that gene distribution and DNA sequence composition remain stable, implying that PDT primarily influences chromatin topology rather than large-scale genomic structure. Next, we investigated changes in TAD organization and intra-domain interactions ([99]Fig. 4A). In the control group, TADs exhibited defined structures, with strong interactions inside each TAD and weaker boundary interactions—a characteristic pattern for chromatin partitioning. Following LD-PDT, most TADs remained stable, although a localized interaction increase at 2.0–2.5 Mb, suggesting chromatin folding, while a reduction at 3.0–3.5 Mb indicated possible relaxation. By contrast, HD-PDT caused a more pronounced weakening of overall interactions, particularly at 1.5–2.0 Mb, which might signify chromatin decompaction, yet an increase in interactions near 3.5 Mb hinted localized condensation. Additionally, HD-PDT appeared to induce or strengthen certain TAD boundaries around 3.5–4.0 Mb, implying that HD-PDT can reorganize chromatin compartmentalization. To quantify these effects, we performed TAD boundary aggregation analysis ([100]Fig. 4D–[101]Tables S3 and S4). Under LD-PDT, boundary interaction strength averaged 1.09, comparable to the control, though a modest dip (0.92) was noted at −60 kb. In contrast, HD-PDT reduced the overall boundary signal to a minimum of 0.90 (−0.19 compared to control) yet caused localized gains near −60 kb and +60 kb (peaking at 1.15, +0.06 vs. control). This pattern suggests that while high-dose treatment weakens chromatin interactions globally, it may also reinforce selected TAD boundaries. Consistent with this, TAD domain insulation analysis ([102]Fig. 4E) showed that intra-TAD interactions under LD-PDT stayed largely unchanged (dropping slightly to 0.817 from 0.822). Under HD-PDT, however, insulation strength rose to 0.863 (from 0.822), reflecting a more extensive shift in chromatin compartmentalization. Lastly, TAD boundary strength ([103]Fig. 4F) revealed no significant difference (NS) among the control and treated groups, indicating that while local structures may undergo remodeling, the primary TAD framework remains generally intact. Collectively, these data demonstrate a dose-dependent influence of PDT on TAD structures. HD-PDT dampens intra-TAD interactions but can bolster certain TAD boundaries, potentially affecting chromatin stability at the local level. With gene density and GC content unchanged, PDT-driven remodeling likely hinges on topological rearrangements rather than large-scale genome sequence alterations. Such localized changes in TAD architecture may have downstream effects on gene regulation, shedding light on how PDT reshapes fungal chromatin organization. 2.5. PDT disrupts chromatin loops and mitochondrial metabolism in C. auris To further investigate the impact of PDT on the 3D genome structure of C. auris, we visualized the top 100 most significant chromatin loops ([104]Fig. 5A, [105]Fig. S4). In the control group, loops were evenly distributed, primarily in chr1 and chr3. However, after LD-PDT and HD-PDT treatment, loop numbers in these regions decreased significantly, suggesting disruption of long-range chromatin interactions. Conversely, new loops emerged in chr5 and chr6, indicating local chromatin reorganization. Loop interaction strength was analyzed across different conditions ([106]Fig. 5B). In the control group, loop interactions were strongest (maximum value 1.58), whereas LD-PDT treatment led to a noticeable reduction (maximum 1.41). After HD-PDT treatment, interaction strength partially recovered (maximum 1.62) but remained below control levels. These results suggest that LD-PDT weakens long-range chromatin interactions, while HD-PDT may induce localized chromatin remodeling. Fig. 5. [107]Fig. 5 [108]Open in a new tab PDT-induced chromatin looping changes and functional enrichment analysis in C. auris. (A) Top 100 chromatin interactions shown as Circos plots for control, LD-PDT, and HD-PDT conditions. Red lines represent long-range chromatin loops, while blue lines indicate interactions within the same chromosomal region. (B) Loop pileup heatmaps for control, LD-PDT, and HD-PDT conditions, displaying the distribution and intensity of chromatin loops across the genome. (C) Venn diagrams showing the overlap of differentially expressed chromatin loops between control vs. LD-PDT and control vs. HD-PDT conditions, with the number of significant loops in each category. (D) Gene Ontology (GO) enrichment analysis for genes associated with differentially regulated chromatin loops between control and LD-PDT (left) and control and HD-PDT (right). A comparison of loop numbers across groups ([109]Fig. 5C) revealed 6837 shared loops between the Control and LD-PDT groups, with 8055 loops unique to LD-PDT and 1439 unique to the Control. In the Control vs. HD-PDT group, 3750 loops were shared, 2961 were specific to HD-PDT, and 4526 were unique to the Control. This trend indicates that LD-PDT promotes the formation of new chromatin interactions, whereas HD-PDT leads to loop instability and loss. GO enrichment analysis of differential loops ([110]Fig. 5D–[111]Tables S5 and S6) showed that PDT significantly affects mitochondrial function and energy metabolism. LD-PDT specifically enriched pathways related to mitochondrial transmembrane transport, protein targeting to mitochondria, and mitochondrial protein transport, suggesting a role for chromatin reorganization in regulating mitochondrial function. HD-PDT impacted oxidative phosphorylation, glycolysis, and RNA metabolism, implying that HD-PDT disrupts energy metabolism and may impair fungal survival. Overall, PDT treatment profoundly alters chromatin loop structures and their functional implications in C. auris. Such loop-level remodeling likely serves as an antifungal mechanism, hampering the fungus's mitochondrial activity and capacity for survival. 2.6. PDT reshapes gene expression and disrupts mitochondrial energy metabolism in C. auris We performed transcriptome sequencing to compare gene expression changes between the control group and LD-PDT as well as the control and HD-PDT group, with three biological replicates per group to ensure data reliability and reproducibility. This analysis aimed to identify PDT-induced gene expression changes and explore the functional roles and pathways of differentially expressed genes (DEGs). To assess sample variability and intra-group consistency, we conducted principal component analysis (PCA) on gene expression data. PCA results showed clear separation between the Control and both LD-PDT-and HD-PDT-treated groups, indicating a significant impact of PDT on gene expression patterns in C. auris ([112]Fig. 6A and B). Differential expression analysis using DESeq2 identified 929 DEGs in the Control vs LD-PDT group (397 upregulated, 532 downregulated) ([113]Figs. 6C) and 2868 DEGs in the Control vs HD-PDT group (1412 upregulated, 1456 downregulated) ([114]Fig. 6D), suggesting a stronger transcriptional response to HD-PDT. Among these, 563 DEGs were shared between the two comparisons, while 366 were unique to the LD-PDT group and 2306 were exclusive to the HD-PDT group ([115]Fig. 6E). Fig. 6. [116]Fig. 6 [117]Open in a new tab Transcriptomic changes following PDT treatment in C. auris. (A) Principal Component Analysis (PCA) comparing control and LD-PDT conditions. (B) PCA for control and HD-PDT conditions. (C) Volcano plot comparing control vs. LD-PDT conditions, highlighting significantly upregulated (red) and downregulated (blue) genes, with the number of differentially expressed genes indicated. (D) Volcano plot for control vs. HD-PDT conditions, illustrating differential gene expression between control and HD-PDT, with genes categorized by fold-change and statistical significance. (E) Venn diagrams showing the overlap of differentially expressed genes between control vs. LD-PDT and control vs. HD-PDT, with the number of unique and shared genes across conditions. (F) GO enrichment analysis for genes differentially expressed in control vs. LD-PDT. (G) GO enrichment analysis for genes differentially expressed in control vs. HD-PDT. Functional enrichment analysis of DEGs revealed distinct biological processes and pathways affected by PDT treatment. In the Control vs LD-PDT group, DEGs were significantly enriched in mitochondrial energy metabolism-related processes, including mitochondrial respiratory chain complex assembly, cytochrome c oxidase assembly, and acetyl-CoA metabolism ([118]Fig. 6F). KEGG pathway analysis showed enrichment in ribosome, glycine, serine and threonine metabolism, beta-alanine metabolism, and methane metabolism pathways ([119]Table S7). In the Control vs HD-PDT group, DEGs were predominantly associated with rRNA processing and metabolism, as well as mitochondrial functions such as NAD biosynthesis, nicotinamide nucleotide biosynthesis, and ATP-dependent RNA processing ([120]Fig. 6G). KEGG pathway analysis highlighted metabolic pathways, including ribosome, N-glycan biosynthesis, ribosome biogenesis in eukaryotes, and thiamine metabolism ([121]Table S8). These transcriptomic data revealed significant gene expression differences between the control and PDT-treated groups, with a more pronounced effect at HD-PDT. GO and KEGG enrichment analyses indicated that PDT primarily affects mitochondrial energy metabolism and ribosomal function, indicating its impact on C. auris at the transcriptional level. 2.7. PDT disrupts chromatin architecture and impairs mitochondrial energy metabolism in C. auris Given that changes in chromatin compartments, loops, and RNA-seq data all pointed to the regulation of energy metabolism-related genes, we examined how PDT affects the energy metabolism network in C. auris to uncover its potential regulatory mechanisms. As shown in [122]Fig. 7A and B and [123]Fig. S5, PDT treatment significantly altered chromatin topology. [124]Fig. 7A indicates partial repositioning of TAD boundaries, changes in loop numbers, and reduced long-range chromatin interactions in certain regions, suggesting that PDT may disrupt coordinated gene regulation. [125]Fig. 7B further compares the control and HD-PDT groups, revealing extensive TAD rearrangement (as shown in arrows) and loop remodeling, with some loops lost and new ones formed, indicating a stronger spatial disruption under HD-PDT. Based on these findings, we focused on energy metabolism-related genes that were significantly downregulated in the HD-PDT group to analyze their expression patterns. Fig. 7. [126]Fig. 7 [127]Open in a new tab PDT-induced chromatin reorganization affects energy metabolism genes in C. auris. (A–B) TAD structures, insulation scores, and chromatin loops in control vs. LD-PDT (A) and control vs. HD-PDT (B). Red arcs represent chromatin loops, with specific enrichment near energy metabolism genes. (C–E) Hi-C contact maps, RNA-seq expression, and chromatin loop interactions for QCR10 (C), NDUFA5 (D), and MP (E), key genes involved in mitochondrial function and energy metabolism. [128]Table 1 showed that several mitochondrial function-related genes were significantly downregulated after HD-PDT treatment, with the most notable changes seen in QCR10 (FDK38_000278), NDUFA5 (FDK38_003365), and MP (FDK38_003723). Chromatin conformation and expression changes of these genes are shown in [129]Fig. 7C–E. [130]Fig. 7C demonstrates that QCR10, involved in complex III of the electron transport chain, exhibited loop structural changes accompanied by a marked reduction in RNA expression, suggesting that PDT may destabilize complex III and impair mitochondrial ATP synthesis. [131]Fig. 7D shows a significant downregulation of NDUFA5 (NADH dehydrogenase subunit 5) after PDT treatment, along with loop remodeling, potentially disrupting complex I function, impairing NADH oxidation, and reducing electron transport chain activity, thereby weakening mitochondrial respiration. [132]Fig. 7E illustrates the downregulation of MP, which encodes the mitochondrial pyruvate carrier responsible for transporting pyruvate from the cytoplasm into mitochondria. The decrease in mp expression, coupled with loop alterations, suggests that PDT may disrupt pyruvate import, impair aerobic metabolism, and reduce energy production efficiency. Together, these findings indicate that PDT-induced chromatin alterations directly impact the regulation of mitochondrial energy metabolism, potentially weakening the electron transport chain and overall cellular energy production. Table 1. List of mitochondrial-related genes affected by PDT-induced chromatin remodeling in C. auris. Gene ID Gene functional identification The Layers of Changes Fold change gene-FDK38_000176 Mitochondrion organization Loop −4.839764055 gene-FDK38_000278 Cytochrome b-c1 complex subunit 10 Complex III subunit 10 Complex III subunit XI Ubiquinol-cytochrome c reductase complex 8.5 kDa protein Loop −2.438523983 gene-FDK38_000455 Tropomyosin-1 Loop −4.592030827 gene-FDK38_000519 Regulation of Ty1 transposition protein 104 rDNA recombination mutation protein 3 Loop −2.636438718 gene-FDK38_000965 Mitochondrial import inner membrane translocase subunit TIM17 Mitochondrial inner membrane protein MIM17 Mitochondrial protein import protein 2 Loop −2.872018522 gene-FDK38_001087 Mitochondrial distribution and morphology protein 31 Precursor Loop −2.490872678 gene-FDK38_001460 Mitochondrial DnaJ homolog 2 Loop −2.562254756 gene-FDK38_001498 Protein ILM1 Increased loss of mitochondrial DNA protein 1 Loop −2.50336301 gene-FDK38_001525 MICOS subunit MIC26 Mitochondrial contact site complex 29 kDa subunit Mitochondrial inner membrane organization component of 27 kDa Mitochondrial organizing structure protein 2 (MitOS2) Loop −3.072255902 gene-FDK38_002018 Mitochondrial import inner membrane translocase subunit TIM21 Precursor Loop −2.83497193 gene-FDK38_002024 Protein ATP12, mitochondrial Precursor Loop −2.902694045 gene-FDK38_002076 Mitochondrial fission 1 protein Loop −2.389204526 gene-FDK38_002165 Actin-related protein 2/3 complex subunit 5 Arp2/3 complex 16 kDa subunit Loop −3.266147114 gene-FDK38_002428 Mitochondrial inner membrane magnesium transporter LPE10 Precursor Loop −2.814774638 gene-FDK38_002452 Mitochondrial intermediate peptidase (MIP) Loop −3.376061532 gene-FDK38_002622 Mitochondrial FAD-linked sulfhydryl oxidase ERV1 regulatory protein Loop −2.971037879 gene-FDK38_002854 Mitochondrial ATPase complex subunit ATP10 Loop −2.782877266 gene-FDK38_002908 Mitochondrial genome maintenance protein MGM101 Precursor Loop −4.09619778 gene-FDK38_002976 Solute carrier family 25 member 38 homolog Loop −2.389981567 gene-FDK38_003277 NADH dehydrogenase 1 alpha subcomplex assembly factor 2 homolog (N7BML) Loop −2.739369514 gene-FDK38_003365 NADH-ubiquinone oxidoreductase 29.9 kDa subunit, mitochondrial Complex I-29.9kD Loop −2.975377655 gene-FDK38_003425 Mitochondrion organization Loop −2.653452819 gene-FDK38_003723 Mitochondrial pyruvate carrier 3 Protein Compartment/Loop −2.374829474 gene-FDK38_004156 DNA mismatch repair protein MSH1, mitochondrial MutS protein homolog 1 Precursor Compartment/Loop −3.922852809 gene-FDK38_004165 Adenylyl cyclase-associated protein(CAP) Compartment/Loop −3.915031376 gene-FDK38_004241 Sorting assembly machinery 35 kDa subunit Mitochondrial 38 kDa outer membrane protein TOB complex 38 kDa subunit Compartment/Loop −3.161990351 gene-FDK38_004242 26S proteasome regulatory subunit RPN11 Compartment/Loop −2.930908775 gene-FDK38_004320 Probable mitochondrial transport protein FSF1 Fungal sideroflexin-1 Compartment/Loop −2.563890277 gene-FDK38_004508 Actin-related protein 2 Actin-like protein ARP2 (Actin-like protein 2) Loop −3.326044203 gene-FDK38_004679 Probable mitochondrial phosphate carrier protein Loop −3.086156433 [133]Open in a new tab Genes are categorized by function and regulatory changes, including chromatin loops and A/B compartment shifts. Fold change values indicate differential expression between PDT-treated and control conditions, with negative values representing downregulation. 2.8. Gene deletions of QCR10 and NDUFA5 impair growth and disrupt mitochondrial integrity in C. auris To functionally validate the link between the chromatin-mediated downregulation of energy metabolism genes and mitochondrial dysfunction, we performed gene knockout of QCR10, NDUFA5, and MP in C. auris. As shown in [134]Fig. 8A, on YPD agar plates, qcr10Δ and ndufa5Δ mutants exhibited significantly slower growth, with smaller and less dense colonies compared to the wild-type (WT) strain, indicating their essential roles in fungal proliferation. Moreover, the mpΔ mutant was nonviable (data not shown), confirming that MP is indispensable for C. auris survival. In liquid culture ([135]Fig. 8B), after 36 h, WT strain showed robust growth with highly turbid media, whereas qcr10Δ and ndufa5Δ cultures remained clear, reflecting impaired proliferation. OD[600] measurements ([136]Fig. 8C) further supported these findings, as WT cells entered the exponential growth phase after 24 h and reached OD[600] ≈ 1.2 at 36 h, while qcr10Δ and ndufa5Δ mutants displayed significantly reduced growth rates and OD[600] values. Fig. 8. [137]Fig. 8 [138]Open in a new tab Phenotypic and ultrastructural analysis of QCR10 and NDUFA5 mutants in C. auris. (A) Growth of wild-type (WT) and mutant strains (qcr10Δ and ndufa5Δ) on solid medium after 36, 48 and 60 h, showing colony morphology differences. (B) Liquid culture growth of WT and mutants after 36 h. (C) Growth curves of WT and mutant strains in liquid culture over 72 h. (D) Scanning electron microscopy (SEM) images of WT and mutant cells at 5000 × and 20000 × magnification. Scale bars represent 20 μm and 5 μm. (E) Transmission electron microscopy (TEM) images of WT and mutant cells at 12000 × and 40000 × magnification. bars represent 1 μm and 500 nm. Scanning electron microscopy (SEM) ([139]Fig. 8D) revealed structural abnormalities in the mutants, with WT cells exhibiting smooth, compact morphology, while qcr10Δ and ndufa5Δ cells appeared shriveled, with surface indentations and irregular shapes, suggesting potential defects in the cell wall or membrane. Transmission electron microscopy (TEM) ([140]Fig. 8E) showed mitochondrial abnormalities—WT cells had intact mitochondria with well-defined cristae, whereas most qcr10Δ and ndufa5Δ mutants exhibited varying degrees of morphological damage, including mitochondrial vacuolization, cristae reduction, and occasional membrane disruption. However, not all mitochondria were observable in each section due to variation in cellular orientation and sectioning depth. These observations indicate that losing QCR10 or NDUFA5 not only hampers growth but also damages mitochondrial structure, likely impairing the electron transport chain and energy metabolism, ultimately weakening fungal viability. 2.9. QCR10 and NDUFA5 are essential for mitochondrial function, biofilm stability, and stress adaptation in C. auris To further investigate the roles of QCR10 and NDUFA5 in C. auris, we analyzed their effects on biofilm formation, mitochondrial function, and energy metabolism. As shown in [141]Fig. 9A and B, qcr10Δ and ndufa5Δ mutants formed loose, irregularly arranged biofilms with wrinkled surfaces, and XTT assays confirmed a significant reduction in biofilm formation, indicating compromised biofilm stability. Mitochondrial dysfunction in the qcr10Δ and ndufa5Δ mutants was confirmed by multiple physiological indicators presented in [142]Fig. 9C. Both strains exhibited significantly elevated intracellular ROS levels as measured by DCFDA and SOSG staining, with the ndufa5Δ mutant showing a greater increase. JC-1 staining revealed a marked reduction in mitochondrial membrane potential (ΔΨm), and this was semi-quantitatively calibrated using FCCP and oligomycin controls ([143]Fig. S6). BCECF-AM staining further demonstrated a decrease in intracellular pH, consistent with dissipation of the proton gradient (ΔpH). While ATP levels were also reduced, the decrease was less pronounced compared to the loss of ΔΨm, indicating a partial compensation likely through enhanced glycolysis. These findings confirm that both QCR10 and NDUFA5 are critical for maintaining mitochondrial electrochemical homeostasis. Fig. 9. [144]Fig. 9 [145]Open in a new tab Impact of QCR10 and NDUFA5 deletions on biofilm formation, oxidative stress, ATP levels, and gene expression in C. auris. (A) Biofilm formation of wild-type (WT) and mutant (qcr10Δ and ndufa5Δ) strains observed under scanning electron microscopy (SEM) at 5000 × and 20000 × magnification. Scale bars represent 20 μm and 5 μm. (B) Quantification of biofilm formation using OD[492] measurements in qcr10Δ and ndufa5Δ mutants compared to WT. (C) Intracellular ROS levels detected by DCFH-DA and SOSG in qcr10Δ and ndufa5Δ mutants compared to WT. (D) Mitochondrial function assessed by ATP production, mitochondrial membrane potential (ΔΨm, JC-1 staining), and intracellular pH (ΔpH, BCECF-AM staining). FCCP (15 μM) and oligomycin (15 μM) were used to define lower and upper ΔΨm thresholds, respectively. (E) Relative mRNA expression of genes associated with biofilm formation (ALS4, EAP1, FKS1), stress response (CRZ1, HSP90), and iron metabolism (FET3) in WT and mutant strains by qRT-PCR. Error bars represent mean ± SD. Statistical significance: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. qPCR analysis ([146]Fig. 9D) showed that FET3 upregulation likely reflects a compensatory response to mitochondrial dysfunction affecting iron metabolism, while increased CRZ1 and HSP90 expression suggests activation of calcium signaling and stress response pathways. Additionally, ALS4 and EAP1, which encode fungal adhesins critical for biofilm formation, were downregulated in the mutants, consistent with the observed reduction in biofilm stability. The downregulation of FKS1 suggests alterations in cell wall integrity. Antifungal susceptibility testing ([147]Table 2) revealed that ndufa5Δ mutants were slightly more sensitive to caspofungin and voriconazole, while qcr10Δ mutants showed no significant changes, indicating that NDUFA5 plays a more critical role in stress adaptation. Table 2. Antifungal susceptibility testing of C. auris CBS14918. MIC range (μg/mL) FLC AMB VRC CAS POS ISA ndufa5Δ >128 0.5 2 8 0.5 0.5 qcr10Δ >128 0.5 4 16 0.25 1 WT >128 0.5 4 16 0.25 1 ATCC 6258 >128 0.5 0.25 0.5 0.25 0.125 [148]Open in a new tab MIC:minimum inhibitory concentration, using MIC[50] and MIC[100] to evaluate antifungal activity of C. auris (n = 3). FLC, fluconazole; VRC, voriconazole; POS, posaconazole; AMB, amphotericin B; CAS, caspofungin. ISA, isavuconozole. Overall, the loss of QCR10 and NDUFA5 disrupted mitochondrial function, impaired biofilm formation, induced oxidative stress, and activated compensatory pathways. Among them, NDUFA5 is particularly important for maintaining energy homeostasis and adapting to environmental stress. 3. Materials and methods 3.1. Strains and cultures C. auris strain (CBS 14918) [[149]27] and plasmid (pSFS2A) used in this study were gifted from professor Wanqing Liao, Shanghai Changzheng Hospital. The knockout plasmid pSFS2A containing the upstream and downstream fragments of the QCR10, NDUFA5 and MP open reading frames were transformed into E. coli strain (DH5α) cells and then grown overnight on Luria Broth solid medium containing 25 μg/mL Chloramphenicol. C. auris strain was grown overnight at 30 °C in yeast peptone dextrose (YPD) medium, which consists of 1 % yeast extract (Oxoid, Basingstoke, England), 2 % peptone (Solarbio, Beijing, China) and 2 % dextrose (Solarbio). The formation of biofilms of C. auris was on RPMI 1640 medium, which consists of 1.04 % RPMI 1640 powder (Gibco, USA), 3.45 % MOPS and 2 % dextrose. Candida parapsilosis ATCC22019 was used as a quality control strain in the broth microdilution assay. Lab conserved strains were stored at −80 °C. All the strains and plasmids used in this study were listed in [150]Table S9. 3.2. Reagents All chemical reagents were used as received without further purification. The ATP Luminescent Cell Viability Assay Kit and JC-1 Mitochondrial Membrane Potential Assay Kit were obtained from YEASEN (Shanghai, China). The ROS Assay Kit, MitoTracker Green (mitochondrial green fluorescent probe), TUNEL apoptosis detection kit, DAPI stain, Hoechst stain and Nourseothricin sulfate were purchased from MedChemExpress. Singlet Oxygen Sensor Green (SOSG) were obtained from Beyotime. Cell culture media and phosphate-buffered solution (PBS) were from BBI Life Sciences. 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H-t etrazolium hydroxide (XTT) and 3-(N-morpholine) propanesulfonic acid (MOPS) were purchased from Shanghai Yuanye Bio-Technology Co., Ltd. All antifungal agents (fluconazole, voriconazole, itraconazole, posaconazole, amphotericin B, terbinafine, caspofungin) were obtained as standard powders from Sigma-Aldrich, Germany. All antifungal agents were diluted in dimethyl sulfoxide (DMSO) at a concentration of 25,600 μg/mL as stock solutions and stored at −80 °C. Milli-Q grade water (18.2 MΩ) was used for solution preparation throughout the experiments. 3.3. aPDT on C. auris Our previous research demonstrated the significant antifungal efficacy of cop1T-HA mediated aPDT [[151]11]. Overnight C. auris cultures were diluted to approximately 10^6 CFU/mL in YPD medium, then treated with COP1T-HA at low-dose (LD-PDT, 0.78 μg/mL) or high-dose (HD-PDT, 3.125 μg/mL) for 30 min, with or without 470 nm laser irradiation (100 mW/cm^2). The cultures were subsequently incubated for 3 h at 30 °C in the dark. Because neither COP1T alone nor dark treatment exerted therapeutic effect, these control groups were not included in the present study. Confocal microscopy was used to assess the intracellular localization and co-localization of COP1T-HA based on its intrinsic red fluorescence. MitoTracker Green and DAPI were used to visualize mitochondria and nuclei, respectively, allowing for evaluation of COP1T-HA distribution in subcellular compartments. Additionally, SOSG was used in combination with Hoechst staining to assess ROS generation and nuclear co-localization. Fluorescent microscopy was then applied to monitor the sequential progression of intracellular events post-PDT. SOSG to detect intracellular ROS, DAPI was used to assess nuclear morphological changes, and TUNEL staining to evaluate DNA fragmentation. Staining was performed at multiple time points (0, 1, 3, 6, 12, and 24 h) to establish the chronological sequence of PDT-induced cellular responses. 3.4. In situ Hi-C library construction protocol for C. auris C. auris CBS14918 were grown in 5 mL YPDA [1 % (w/v) Tryptone (BBI), 1 % (w/v) Yeast Extract (BBI), 2 % (w/v) glucose (BBI),0.15 mM Adenine sulfate (BBI), and 2 % (w/v) Agar (BBI)] overnight, then diluted and grown in 50 ml YPDA for 5 h at 30 °C. Cells were resuspended by 1 × PBS to an OD[600] of 1.0. Cells were cross-linked with 3 % final concentration of fresh formaldehyde and the cross-linking was quenched with 0.15 M final concentration of glycine for 5 min. The cells were resuspended in 1 mL of 1 × NEBuffer 2.1 (NEB) and homogenized by grinding to a fine powder in liquid nitrogen. Then, the homogenized yeast material was washed by 25 mL 1 × NEBuffer 2.1, then suspended in 1 × NEBuffer 2.1 to an OD[600] of 10.0 (about 2.5 mL 1 × NEBuffer 2.1). Cells were split into aliquots (V = 456 μL) and the chromatin was solubilized in 0.1 % SDS for 10 min at 65 °C. Cross-linked DNA was digested with 200U MboI (NEB) per tube at 37 °C overnight. Restriction fragment ends were labeled with biotinylated cytosine nucleotides by biotin-14-dCTP (TriLINK). Blunt-end ligation was carried out at 16 °C overnight in the presence of 100 Weiss units of T4 DNA ligase (Thermo, 10.0 mL final volume per tube). After ligation, the cross-linking was reversed by 200 μg/mL proteinase K (Thermo) at 65 °C overnight. DNA purification was achieved through QIAamp DNA Mini Kit (Qiagen) according to manufacturers’ instructions. Purified DNA was sheared to a length of ∼400 bp. Point ligation junctions were pulled down by Dynabeads® MyOne™ Streptavidin C1 (Thermofisher) according to manufacturers’ instructions. The Hi-C library for Illumina sequencing was prepared by NEBNext® Ultra™ II DNA library Prep Kit for Illumina (NEB) according to manufacturers’ instructions. Fragments between 400 and 600 bp were paired-end sequenced on an Illumina HiSeq X10 platform (San Diego, CA, United States) with 150PEmode. Two replicates were generated for one group material [[152]28]. 3.5. Construction of the contact map Raw Hi-C reads were first quality filtered using fastp (v0.23.2), and subsequently processed with the Distiller pipeline ([153]https://github.com/mirnylab/distiller-nf) to map reads to the genome, filter dangling ends, and remove unusable data. Valid read pairs were used to assess reproducibility across two biological replicates via HiCRep. The data from replicates were pooled and then binned into non-overlapping genomic intervals at multiple resolutions (40 kb, 20 kb, 10 kb, 5 kb, 2 kb, 1 kb) to generate comprehensive contact maps [[154]29]. Raw contact maps were normalized iteratively to eliminate biases such as GC content, mappability, and uneven distribution of restriction enzyme sites. 3.6. Map resolution analysis The map resolution was defined as the smallest bin size at which 80 % of bins contained at least 1000 contacts, providing an estimate of the finest scale at which local chromatin features can be reliably detected. 3.7. Compartment A/B analysis Chromosomal compartments were identified by performing principal component analysis (PCA) on the Hi-C data binned at 100 kb [[155][30], [156][31]]. Genomic bins with positive values of the first principal component (PC1) were classified as A compartments, corresponding to gene-dense and transcriptionally active regions, while bins with negative PC1 values were assigned to B compartments, typically representing gene-poor, inactive regions. 3.8. TAD analysis TADs were delineated using Hi-C data binned at 40 kb [[157]32]. TAD boundaries were identified with the insulation score algorithm as implemented in cooltools (v0.6.1). This approach allowed us to determine the number and positions of TADs, and to quantify changes in intra-TAD interactions and boundary strengths across samples [[158]33]. 3.9. Calculation of chromosomal interactions Contact counts between 5 kb bins for chromosomal interactions were processed using Fit-Hi-C to calculate cumulative P-values and false discovery rate (FDR) q-values. Interactions with both a P-value and q-value less than 0.001, were considered significant. 3.10. Comparative analysis of Hi-C data To compare chromatin interactions between samples, raw contact matrices were transformed into Z-score matrices following the method described by Crane et al. [[159]33]. These normalized matrices facilitated the identification of differences in interaction frequencies between conditions. TAD boundaries were compared between samples using cooltools, classifying boundaries as stable, newly appeared, disappeared, weakened (at least a twofold decrease), or strengthened (at least a twofold increase) based on changes in boundary strength. 3.11. Library preparation and RNA sequencing Total RNA of the samples was isolated and purified by TRIzol (thermofisher, 15596018) according to the operation scheme provided by the manufacturer. Then, the quantity and purity of total RNA was determined by NanoDrop ND-1000 (NanoDrop, Wilmington, DE, USA) and the integrity of RNA was detected by Bioanalyzer 2100 (Agilent, CA, USA). The concentration was >50 ng/μL, RIN value > 7.0, and total RNA > 1 μg. Samples with concentration >50 ng/μL, RIN value > 7.0, and total RNA > 1 μg were used for library construction and sequencing, following the manufacturer's instructions. Sequencing was performed using the Illumina NovaSeq 6000 for all libraries with a paired-end sequencing length of 150 bp (PE150). 3.12. Transcriptome data analysis After using Cutadapt [[160]34] to filter out unqualified sequences (sequencing joints, low-quality sequences, etc.) from the original Data to obtain Clean Data, the reference genome is compared using Hisat2. According to the comparison results of Hisat2 [[161]35], The transcript was reconstructed using StringTie and the expression levels of all genes in each sample were calculated. Gene expression level analysis is mainly aimed at protein-coding genes (mRNA) annotated by the genome, and the expression level of genes is statistically measured to evaluate the correlation of gene expression characteristics and differentially expressed genes within and between groups. When measuring gene expression level, FPKM value (Fragments Per Kilobase Million, standardized based on the original reads count of genes) was used as a measure of gene expression level, and gene expression levels in different samples were counted. The FPKM value of a gene can be understood as the amount of gene expression. Then, we used DESeq2 v1.42.1 [[162]36] to identify differentially expressed genes. The GO and KEGG enrichment analysis were performed by the R package clusterProfiler v4.10.1 [[163]37]. Heat map of DEGs was drawn using R package heatmap v1.0.12 ([164]https://CRAN.R-project.org/package=pheatmap). Differential gene volcano map and box map were drawn using ggplot2 v3.5 [[165]38]. 3.13. Integration of Hi-C and RNA-seq data To correlate chromatin architecture changes with transcriptional alterations, we integrated the Hi-C contact maps with RNA-seq expression profiles. DEGs were identified using DESeq2 with a false discovery rate (FDR) cutoff of 0.05. Genes were subsequently mapped onto the Hi-C contact matrices to assess their spatial distribution relative to altered compartments and loops. This integration enabled us to determine whether regions exhibiting significant chromatin reorganization corresponded with transcriptional downregulation, particularly in areas associated with mitochondrial energy metabolism. 3.14. Constructing the vectors and gene knockout Upstream and downstream homologous fragments of the QCR10 (CJI96_0000277), NDUFA5 (CJI96_0003024), and MP (CJI96_0002903) genes were amplified from the cDNA library of C. auris strain. These PCR products were inserted into the pSFS2A vector between the KpnI/XhoI and SacII/SacI restriction sites using an In-Fusion cloning kit ([166]Fig. S7). All primers used in this study are listed in [167]Table S10. All constructed vectors were verified by Sanger sequencing. To prepare competent cells, wild-type colonies were first grown overnight in 5 mL YPD medium at 30 °C with 200 rpm shaking. The overnight culture was diluted 1:50 into 70 mL fresh YPD and incubated until the OD[600] reached ∼0.8. After harvesting 100 mL of cells and washing twice with sterile water, the pellet was resuspended in 1 mL 1 × TE/LiAc solution (100 μL 10 × TE, 100 μL 10 × LiAc, 800 μL deionized water) and incubated at room temperature for 30 min. Then, 100 μL of competent cells were mixed with 10 μL salmon sperm DNA and 40 μL purified, SacI-linearized fusion vectors. Next, 1 mL of “plate solution” (100 μL 10 × TE, 100 μL 10 × LiAc, 800 μL 50 % PEG3350) was added, and the mixture was shaken at 30 °C, 200 rpm for 4 h. Following a 100 μL DMSO addition, cells were heat-shocked at 44 °C for 25 min, placed on ice for 3 min, resuspended in fresh YPD, and incubated for 1 h at 30 °C, 200 rpm. Finally, cells were washed, plated on selective agar containing 100 mg/L nourseothricin, and incubated at 30 °C for 2 days to obtain transformants. PCR confirmed the SAT1 gene insertion and successful knockout of each target gene. Electrophoresis was performed on a 1 % agarose gel in 1 × TBE buffer at 145 V for 25 min ([168]Fig. S8), followed by staining with a nucleic acid dye and UV visualization. 3.15. Functional validation experiments To characterize the physiological impact of these gene deletions, colony growth was monitored on both solid and liquid media. On YPD agar, colony morphology and size were recorded at 30 °C at 36 h, 48 h, and 60 h post-inoculation; both wild-type and mutant strains were pre-cultured in liquid YPD medium to reach logarithmic phase (OD[600] ≈ 0.6), cell suspensions were then diluted to an identical OD[600] of 0.01 before being spotted or streaked onto YPD agar plates. In liquid culture, OD[600] was measured every 12 h for 60 h to monitor growth kinetics. All conditions, including inoculum density, medium composition, and incubation temperature, were strictly controlled to ensure consistency. Biofilm formation was induced by allowing C. auris cells to adhere to 96-well plates for 2–2.5 h, washing with PBS to remove non-adherent cells, and incubating for an additional 24 h (total of 48 h) for biofilm maturation. Metabolic activity was quantified by XTT reduction; after a 2-h incubation at 37 °C, absorbance at 492 nm was measured to assess biofilm biomass and cell viability. Mitochondrial function was assessed by measuring intracellular ATP (using a luciferase-based luminescence assay kit), ROS production using two probes, DCFDA for general intracellular ROS and SOSG for singlet oxygen. And mitochondrial membrane potential using JC-1 staining, where red-to-green fluorescence intensity indicates mitochondrial health. To semi-quantitatively assess ΔΨm, FCCP (15 μM) and oligomycin (15 μM) were used as uncoupler and inhibitor controls to define the lower and upper thresholds, respectively (The selection of these concentrations was based on optimization experiments detailed in the [169]Supplementary Information Fig. S6). Intracellular pH (ΔpH) was evaluated using BCECF-AM, a ratiometric fluorescent dye sensitive to pH changes, and fluorescence intensity was measured according to the manufacturer's protocol. These measurements allowed for integrated assessment of the components of the proton motive force (Δp), including ΔΨm and ΔpH. For quantitative PCR (qPCR) analyses, RNA was extracted via a TRIzol-based method, then reverse-transcribed with the Easy Script® One-Step gDNA Removal and cDNA Synthesis SuperMix Kit. Gene-specific primers were designed using BLAST, with ACT1 as the internal control. qPCR was carried out using the TransStart® Green qPCR SuperMix and a standard 40-cycle program (94 °C for 30 s, followed by 40 cycles of 94 °C for 5 s, and 60 °C for 20 s). All reactions were performed in triplicate, and relative expression levels were calculated using the 2^−ΔΔCt method. Minimum inhibitory concentrations (MICs) of common antifungal agents were determined following Clinical and Laboratory Standards Institute (CLSI) M27-A3 [[170]39] guidelines, with cultures incubated in 96-well plates at 30 °C for 48 h. MIC[100] was used for amphotericin B (AMB), while MIC[50] was applied for other antifungals. 3.16. SEM observation Both wild-type and mutant strains were inoculated at an initial OD[600] of 0.01 in liquid medium and cultured until reaching the logarithmic growth phase (OD[600] = 0.6). After centrifugation at 3000×g to collect the cells, the samples were fixed overnight using electron microscopy fixative. For SEM preparation, 100 μL of RPMI-1640 liquid medium was added to a sterile 96-well plate and incubated at 37 °C for 0.5 h, followed by the removal of the medium. The 100 μL of fungal suspension was added to the plate and incubated at 37 °C for 2–2.5 h. After discarding the suspension, the wells were washed three times with PBS. Following this, the samples were fixed overnight with electron microscopy fixative. The fixed samples were dehydrated through a graded ethanol series (30 %, 50 %, 70 %, 90 %, and 100 %) for 5 min at each concentration. To improve conductivity, the samples were sputter-coated with gold or platinum. Finally, the samples were observed under a scanning electron microscope (SEM) at magnifications ranging from × 5000 to × 20,000 to analyze their cellular morphological characteristics. 3.17. TEM observation Both wild-type and mutant strains were inoculated at an initial OD[600] of 0.01 in liquid medium and cultured until reaching the logarithmic growth phase (OD[600] = 0.6). The culture was centrifuged at 3000×g to collect the cells, which were then fixed overnight with electron microscopy fixative. After fixation, the samples were further fixed with 1 % osmium tetroxide or uranyl acetate for 1 h. The samples were dehydrated through a graded ethanol series (30 %, 50 %, 70 %, 90 %, and 100 %) and embedded in epoxy resin. Using an ultramicrotome, the samples were sectioned into 100 nm slices, transferred to carbon-coated copper grids, and stained. The samples were then observed under a TEM, with parameters adjusted to obtain high-resolution images for analysis of cellular and subcellular structures. 3.18. Statistical analysis The results were analyzed by GraphPad Prism software. Every experiment was independently performed at least three times, and the data were expressed as the mean ± SD. Statistical comparison of mean values was performed with the ANOVA test, and the results were considered statistically significant when the ∗P < 0.05, ∗∗P < 0.01 and ∗∗∗P < 0.001. 4. Discussion This study integrates Hi-C and RNA-seq data to generate, for the first time, a comprehensive 3D genome map of Candida species before and after PDT (it is important to note that the results obtained herein specifically pertain to COP1T-HA. The distinct chemical and photophysical characteristics of COP1T-HA, particularly its dose-dependent intracellular distribution and ROS generation profile, cannot be universally extrapolated to other photosensitizers without further empirical validation. Future studies using additional photosensitizers are necessary to generalize the findings related to chromatin remodeling and nuclear interactions observed here). Under HD-PDT, which induces oxidative stress, we observed significant remodeling of chromatin architecture at the levels of compartments, TADs, and loops [[171][40], [172][41], [173][42]]. And HD-PDT not only triggered genome-wide reorganization of TADs and switched compartment states, but also elicited marked local changes in compartments and dynamic reconfiguration of loops in regions harboring genes of energy metabolism (e.g., QCR10, NDUFA5, and MP). These local alterations are highly consistent with the observed downregulation of gene expression in these regions. Furthermore, gene knockout experiments confirmed the essential roles of these genes in maintaining mitochondrial function, cellular energy metabolism, and growth [[174][43], [175][44], [176][45]]. Previous studies have demonstrated that chromatin loops serve as critical bridges linking enhancers and promoters, with their formation or disruption directly impacting gene expression [[177]46,[178]47]. Our observation of local loop remodeling strongly correlating with the downregulation of genes of energy metabolism suggests that HD-PDT may attenuate transcription by disrupting these critical enhancer–promoter contacts [[179]46,[180]48,[181]49]. Although HD-PDT also led to extensive changes in TAD structure at the genome-wide level, we did not detect notable local adjustments of TAD boundaries in the regions of genes of energy metabolism. This implies that local regulatory mechanisms primarily depend on changes at the compartment and loop levels, while TADs largely reflect the global chromatin response to stress [[182][50], [183][51], [184][52]]. Consistent with our Hi-C data, RNA-seq analysis revealed significant downregulation of genes closely associated with mitochondrial function [[185]53] following HD-PDT [[186]11]. Moreover, our gene knockout experiments further confirmed the indispensable roles of QCR10, NDUFA5, and MP in maintaining cellular energy metabolism and mitochondrial integrity. Combining our Hi-C and RNA-seq data with previous literature, we infer that local chromatin remodeling plays a key role in regulating the expression of energy metabolism genes, consistent with previous findings [[187]54,[188]55]. Our findings are highly consistent with observations from other systems. For example, in mouse neurons, environmental stress triggers TAD and loop reorganization that is closely linked to gene expression reprogramming [[189][56], [190][57]]. In plants, external stress induces dynamic adjustments of chromatin structure that regulate the expression of stress response genes [[191]58,[192]59]. In budding yeast, oxidative stress-induced chromatin remodeling has been shown to correlate with changes in stress gene expression [[193]60]. Furthermore, in fungi, changes in chromatin architecture under stress are tightly coupled with gene regulation [[194]61], Selmecki et al. highlight the importance of genome structure in the development of drug resistance [[195]62]. Recent reviews detail the link between chromatin organization and transcription, and emphasize that dynamic chromatin remodeling is essential for cellular adaptation to external cues [[196]63,[197]64]. Our study not only reveals the overall remodeling of the 3D genome in C. auris following low & high dose PDT but also underscores a strong association between compartment and loop remodeling in energy metabolism gene regions and the downregulation of their expression. This finding offers a new perspective on the molecular mechanisms underlying the antifungal effects of PDT and lays a theoretical foundation for the development of novel antifungal strategies based on chromatin regulation. Our gene knockout experiments confirmed the functional necessity of these key energy metabolism genes. Based on our Hi-C and RNA-seq data, we propose that local chromatin remodeling plays a critical role in regulating the expression of energy metabolism genes. In support of the observed gene expression changes, our physiological data confirm that deletion of QCR10 and NDUFA5 results in significant mitochondrial dysfunction. The mitochondrial membrane potential (ΔΨm), measured by JC-1 staining and calibrated using FCCP and oligomycin, was markedly reduced in all mutant strains. In parallel, intracellular pH (ΔpH), assessed via BCECF-AM staining, also decreased significantly. These findings align with the well-established model of mitochondrial bioenergetics in which the proton motive force (Δp)—comprised of both ΔΨm and ΔpH—drives ATP synthesis. In eukaryotic cells, ΔΨm typically accounts for approximately 80–90 % of Δp. Therefore, substantial reductions in ΔΨm can trigger ΔpH dissipation due to proton re-entry into the mitochondrial matrix, ultimately compromising ATP synthase efficiency. Consistently, our data showed a marked drop in ATP production in both qcr10Δ and ndufa5Δ mutants. Both DCFDA and SOSG assays confirmed increased ROS levels in qcr10Δ and ndufa5Δ strains. Notably, DCFDA staining showed more variability among biological replicates, with ndufa5Δ-1 exhibiting a slightly higher signal, although this was not consistently observed in SOSG results. These findings support a general increase in oxidative stress in complex I-deficient mutants, while further studies are required to explain intra-strain variation. Additionally, Hi-C analysis revealed chromatin loop remodeling near the NDUFA5 locus after high-dose PDT. While a direct causal relationship remains to be established, this spatial reorganization may reflect broader transcriptional responses linked to mitochondrial dysfunction. Notably, however, the extent of ATP reduction was less severe than the loss of ΔΨm, suggesting that glycolytic compensation might have occurred. This metabolic adaptation is well documented in yeast species such as Candida albicans and Saccharomyces cerevisiae [[198][65], [199][66]], where cytosolic ATP production through fermentation can sustain energy demands under mitochondrial impairment. Together, these findings highlight that disruption of electron transport chain components not only affects ΔΨm but also disturbs mitochondrial pH homeostasis and downstream energy production in C. auris. In future work, we plan to further employ CRISPR-dCas9-based chromatin engineering, ATAC-seq, and ChIP-seq to directly manipulate and dissect the effects of local chromatin structural changes on gene expression, providing more robust experimental evidence for antifungal therapy. 5. Conclusion In conclusion, this study presents the first comprehensive 3D genomic map of C. auris with and without COP1T-HA mediated PDT, revealing that HD-PDT induces extensive chromatin reorganization at multiple levels. Our Hi-C analysis demonstrated a global reconfiguration of TADs and compartment state switches, while local remodeling—particularly in regions harboring key energy metabolism genes such as QCR10, NDUFA5, and MP—was characterized by dynamic changes in compartments and chromatin loops. These local structural alterations are strongly correlated with significant transcriptional downregulation of these genes, as evidenced by RNA-seq data. In addition, gene knockout experiments confirmed the critical role of these genes in maintaining mitochondrial function and cellular energy homeostasis. These findings establish a strong link between COP1T-HA mediated PDT-induced chromatin architecture remodeling and the regulation of energy metabolism in C. auris, providing new insights into the antifungal mechanisms operating in multidrug-resistant pathogens. CRediT authorship contribution statement Xinyao Liu: Conceptualization, Formal analysis, Funding acquisition, Investigation, Writing – original draft. Yiran Tao: Methodology. Linwan Zhang: Formal analysis, Methodology. Yuzhou Liu: Methodology. Dongmei Shi: Methodology. Jiao Wang: Methodology. Peng Xue: Methodology. Bin Xu: Conceptualization, Writing – review & editing. Wenjie Fang: Conceptualization, Funding acquisition, Writing – review & editing. Yuping Ran: Conceptualization, Funding acquisition, Writing – review & editing. Fundings The work was supported by the National Natural Science Foundation of China [82302546, 82202543]; Natural Science Foundation of Sichuan Province [2023NSFSC1550]; Project of the Ministry of Science and Technology [2022YFC2504803]; the National Key Research and Development Program of China [2022YFC2504800]; HX-Academician Project of West China Hospital, Sichuan University [HXYS19003]. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements