Genome-Wide Exploration of Thiamin Pyrophosphate Riboswitches in Medically Relevant Fungi Reveals Diverse Distribution and Implications for Antimicrobial Drug Targeting Authors: Valdemir Vargas-Junior, Ana Carolina Ramos Guimarães, Ernesto Raul Caffarena, Deborah Antunes Published: 2024 ABSTRACT The rising incidence of fungal infections coupled with limited treatment options underscores the urgent need for novel antifungal therapies. Riboswitches, particularly thiamin pyrophosphate (TPP) class, have emerged as promising antimicrobial targets. This study presents a comprehensive genome-wide analysis of TPP riboswitches in 156 medically relevant fungi utilizing advanced covariance models (CMs) tailored for fungal sequences. Our investigation identified 378 conserved TPP riboswitch sequences distributed across 140 distinct species, revealing a broader prevalence than that previously recognized. Notably, we provide evidence for a novel putative group of TPP riboswitches, designated TPPswSUGAR, associated with sugar transporters in Mucoromycota and Basidiomycota. This group exhibits distinctive structural features while maintaining key TPP-binding motifs, potentially expanding our understanding of the riboswitch diversity in fungi. Our analysis highlights the impact of P3 stem variability on riboswitch detection and characterization, demonstrating the superiority of fungal-specific CMs over generic models. We observed multiple TPP riboswitches in over 50% of the examined species, including clinically significant pathogens involved in aspergillosis and mucormycosis. Remarkably, Aspergillus latus, a species associated with COVID-19 coinfections, harbors six distinct TPP riboswitch sequences, whereas the extremophilic black fungus Hortaea werneckii possesses nine. These findings not only elucidate the diverse distribution of TPP riboswitches in pathogenic fungi but also emphasize their potential as multifaceted targets for antifungal drug development. By addressing key limitations of previous detection methods and providing insights into riboswitch structural diversity, this study lays a foundation for future investigations into riboswitch-mediated regulation in fungi and the development of novel antifungal strategies. Introduction The incidence of fungal infections in humans has grown alarmingly in recent years.1 These infections affect over 300 million individuals globally, with an estimated annual mortality of 1.6 million.2,3 Notably, certain fungal infections have garnered significant international attention during the Covid-19 pandemic, as they have been identified as coinfections in patients with this viral disease.4−18 The list of fungal coinfections related to Covid-19 encompasses aspergillosis, candidiasis, cryptococcosis, mucormycosis, pneumocystosis, and several endemic mycoses.4−7,10,13−19 Among these, aspergillosis and mucormycosis have emerged as substantial diseases with alarming mortality rates.4,14,16,20 Even before the pandemic, chronic pulmonary aspergillosis affected more than 3 million people each year.2 This infection is caused by various species within the genus Aspergillus, with Aspergillus fumigatus being the most frequently associated pathogen.8,19,21−27 Unfortunately, this challenging scenario is compounded by limited availability of effective antifungal therapies and the emergence of drug-resistant fungal strains.25,28,29 As aggravating factors, most currently available antifungal drugs rely on only a few primary mechanisms. This means that when a fungus develops resistance to one of them, it significantly reduces the treatment options.30−38 In fact, some species, such as Lomentospora prolificans, have already exhibited resistance to most currently available antifungal drugs.38−42 Additionally, current antifungal therapies often lead to significant side effects due to the similarity between human and fungal cells. Therefore, the need of developing novel antifungal therapies becomes evident.30,31,43−49 To address the global burden of fungal infections, the World Health Organization (WHO) released a priority list of fungal pathogens, highlighting the urgent need for new therapies.50 This initiative echoes a comparable undertaking in 2017 when the WHO released a priority list for bacterial pathogens.51 Riboswitches are molecular structures present in the introns and untranslated regions (UTRs) of the mRNA (mRNA) of several species of bacteria, archaea, fungi, and plants.52,53 These structures play a crucial role in the recognition and post-transcriptional regulation of specific metabolites including vitamins, ions, and nitrogenous bases.54,55 The riboswitch consists of an evolutionarily conserved aptamer domain associated with the expression platform.56−58 The fundamental principle of the riboswitch function relies on two structurally stable conformations that can interchange depending on the presence or absence of their respective ligands in their binding site.54,59,60 Among the over 40 classes of known riboswitches, the thiamin pyrophosphate (TPP) riboswitch appears to be the only class of riboswitches found in fungi.57,61−65 The TPP riboswitch, a type of riboswitch that recognizes and binds to TPP, is the most prevalent riboswitch class found in three domains of life: Archaea, Bacteria, and Eukarya.56,66,67 TPP is an essential cofactor for several organisms, participating in central metabolic pathways, such as the Krebs cycle and pentose-phosphate pathway.68−71 Consequently, TPP plays a crucial role in the growth and survival of many organisms,71−73 making the TPP riboswitch an intriguing target for the development of novel antimicrobial drugs. Nonetheless, the antimicrobial potential of TPP analogs in bacteria and fungi has only been demonstrated in a limited number of studies,66,74,75 whereas others have provided confirmatory evidence.63 Thus, far, five groups of TPP riboswitches have been reported in fungi.61,62,65 In the first two, the TPP riboswitch regulates thi4 (TPPswTHI4) and nmt1 (TPPswNMT1) genes associated with the biosynthesis of the two moieties constituting TPP. In the three others, the TPP riboswitch is believed to control the gene expression of distinct families of transporter proteins: urea (TPPswUREA), nucleoside (TPPswNUC), and amino acid (TPPswAA) transporters. Previous studies also found other target genes associated with riboswitch sequences, suggesting the possibility of additional groups of TPP riboswitches in fungi that could not be classified at that time.61,65 Certain fungal species have multiple TPP riboswitches within their genomes.61,65 Moreover, the TPP riboswitch exhibits remarkable conservation across species, particularly regarding its secondary and tertiary structures. In our recent study, which focused on three distinct TPP riboswitches discovered in Aspergillus oryzae, we revealed similarities in TPP dynamics and binding free energy.63 These findings highlight the importance of gaining comprehensive understanding of their functional characteristics. Given the potential use of TPP riboswitches as drug targets and their involvement in critical post-transcriptional processes, it is imperative to explore their distribution and characteristics in medically relevant fungal species. In this study, we conducted a comprehensive search for TPP riboswitches in the genomes of 156 fungal pathogens, encompassing all fungal groups in the WHO’s priority list, as well as other fungal species associated with relevant infections in humans.2,4−6,18,28,29,50,76−84 In addition, we conducted a comparative analysis of riboswitches from these species using multiple covariance models (CMs) to augment the precision of TPP riboswitch detection. By elucidating the presence and diversity of TPP riboswitches in medically important fungi, we aim to contribute to the development of novel antifungal therapies with improved efficacy against these infections. Results TPP Riboswitch Distribution in Fungal Pathogens Our comprehensive analysis of 156 medically relevant fungal species revealed a widespread distribution of TPP riboswitches, with 140 species (89%) harboring at least one regulatory element. A total of 378 TPP riboswitch sequences were identified, of which 322 were assigned to previously known riboswitch groups. Notably, 116 species (74%) possessed two or more TPP riboswitches, highlighting the prevalence of multiple regulatory elements within the individual genomes. The distribution of TPP riboswitches across different groups was as follows: 107 sequences of TPPswTHI4, 91 sequences of TPPswNMT1, 81 sequences of TPPswUREA, 26 sequences of TPPswNUC, and 18 sequences of TPPswAA. Additionally, we identified 43 sequences associated with sugar transporters, providing evidence for a potentially new group of fungal TPP riboswitches, which we designated as TPPswSUGAR. The remaining 12 identified sequences were associated with other target genes that could not be included in any of the established TPP riboswitch groups. Interestingly, the TPPswSUGAR group exhibited a remarkable prevalence in Mucoromycota species, emerging as the most abundant group within this taxon (Figure 1). This distribution pattern suggests a specialized role for the TPP-based regulation of sugar transport in these fungi. Notably, other transporter families were largely absent in most Mucoromycotina species, except for the TPPswAA group found in Lichtheimia corymbifera. Distribution and diversity of TPP riboswitches across Mucoromycota species. The phylogenetic tree shows the relationships among the Mucoromycota species, with Mucoromycotina highlighted in green. Colored circles represent different TPP riboswitch groups identified in each species as defined in the legend. Circle quantity indicates the number of detected riboswitches. This visualization revealed the prevalence of TPPswSUGAR (red), TPPswTHI4 (blue), and TPPswNMT1 (cyan) in Mucoromycota, and demonstrated the diversity of TPP riboswitch types across the phylum. Our analysis also revealed the co-occurrence of two TPP riboswitches that regulate TPP biosynthesis in multiple Mucormycota species, indicating a potentially more complex regulatory network in these fungi. In Mucor irregularis and Mucor lusitanicus, we identified proteins annotated as ATP-binding domains related to ABC transporters. However, further examination using deep TMHMM revealed the absence of transmembrane domains in these proteins, suggesting that they are unlikely to be involved in membrane transport processes. This diverse distribution of TPP riboswitches across different functional groups underscores their importance in regulating various aspects of fungal metabolism, from thiamine biosynthesis to transport of essential molecules. The identification of the putative TPPswSUGAR group, particularly prevalent in Mucoromycota species, suggests a previously unrecognized role of TPP-based regulation in fungal sugar transport. Distribution of TPP Riboswitches in Basidiomycota and Ascomycota According to our analysis, the most prevalent TPP riboswitch group In Basidiomycota, was TPPswAA (Figure 2), which regulates amino acid transporters. TPP riboswitch distribution in Basidiomycota. The tree depicts the phylogenetic relationships among Basidiomycota species, with classes indicated by colors in the legend. Colored circles represent the different TPP riboswitch groups found in each species, as defined in the legend. The quantity of circles corresponds to the number of detected riboswitches. This figure highlights the predominance of TPPswAA (pink) and TPPswSUGAR (red) in Basidiomycota, and illustrates the varying distribution of TPP riboswitch types across the phylum. These transporters are typically responsible for the uptake of various amino acids, including branched-chain and aromatic amino acids. Additionally, we identified a TPP riboswitch associated with a protein annotated as an amino butyric acid transporter in a few species. TPPswUREA was exclusively found in Malassezia furfur. Notably, TPPswNUC and TPPswNMT1 were absent in Basidiomycota, except for the presence of TPPswTHI4 in Rhodotorula mucilaginosa. The prevalence of TPPswTHI4 and TPPswUREA was similar across the examined species. These findings suggest that TPP riboswitches have limited regulatory roles in governing TPP biosynthesis within Basidiomycota, with greater emphasis on regulating amino acid transport. In Ascomycota, all TPP riboswitch groups were present, except for TPPswSUGAR (Figure 3), whereas TPPswUREA was abundant across most species. However, in some Aspergillus and certain Dothideomycete species, TPPswNUC was observed instead of TPPswUREA. Additionally, TPPswTHI4 and TPPswNMT1 were prevalent in Ascomycota, except for their absence in Candida spp. Prevalence and diversity of TPP riboswitches in Ascomycota. The phylogenetic tree shows relationships among Ascomycota species, with classes indicated by colors defined in the legend. Colored circles represent the different TPP riboswitch groups identified for each species. Circle quantity indicates the number of detected riboswitches. This visualization shows the wide distribution of TPPswTHI4 (blue), TPPswNMT1 (cyan), and TPPswUREA (yellow) across Ascomycota, while also highlighting the absence of TPPswSUGAR in this phylum. Our bioinformatic analysis identified some Ascomycota species with a notably high number of TPP riboswitches. Among these, Hortaea werneckii was found to have the highest number, with a total of nine TPP riboswitches. These consisted of four pairs (two each of TPPswTHI4, TPPswNMT1, TPPswNUC, and TPPswUREA) and a single TPPswAA. The presence of these paired riboswitches in H. werneckii can be directly attributed to complete genome duplication,85−87 which resulted in two copies of each riboswitch-regulated gene. Among the Aspergillus species, we typically detected three riboswitches: TPPswNMT1, TPPswTHI4, and TPPswNUC. However, Aspergillus latus stood out with six TPP riboswitches, specifically, three pairs corresponding to TPPswNMT1, TPPswTHI4, and TPPswNUC. This doubling of TPP riboswitches in A. latus is consistent with its known allodiploid genome structure resulting from an interspecies hybridization event. As an allodiploid species, A. latus contains a full set of chromosomes from each parent species, effectively doubling its genomic content, including TPP riboswitch-regulated genes. Characteristics of TPPswSUGAR Our comprehensive analysis uncovered a group of 43 sequences associated with sugar transporters that exhibit the key characteristics of TPP riboswitches while displaying distinct features. We named this group the TPPswSUGAR. These sequences maintain the hallmark elements of TPP riboswitches, including highly conserved nucleotides in the J2–3 junction region, which are typically involved in binding to the pyrimidine ring of TPP (Figure 4). Furthermore, the pyrophosphate binding pocket, characterized by the conserved GCG sequence, is present and retains the residues responsible for interacting with the pyrophosphate moiety of TPP. Consensus secondary structure of the putative TPPswSUGAR riboswitch. The structure is color-coded to indicate the nucleotide conservation levels, as defined in the legend. Golden asterisks mark nucleotides predicted to interact with TPP based on their structural similarity to known TPP riboswitches. The key structural elements (P1–P5, L2-L5, J2–4) are labeled. This model highlights both the conserved features shared with other TPP riboswitches and the unique structural characteristics of the TPPswSUGAR group. The overall secondary structure of the TPPswSUGAR sequences conforms to the characteristic three-way junction architecture of the TPP riboswitches. A distinguishing feature of the TPPswSUGAR group is its P3 stem, which shows notable variability compared with other fungal TPP riboswitch groups. While the average length of the P3 stem in TPPswSUGAR was shorter (115 nucleotides) than that in the other groups, it exhibited considerable sequence diversity. This variability in the P3 region, combined with differences observed in the P2 and P4 stems and the J2–4 region, contributes to the unique structural profile of the TPPswSUGAR group. Genomic context analysis revealed that TPPswSUGAR sequences were consistently located in the 5′-UTR of genes annotated as sugar transporters across multiple fungal species. This consistent position suggests a conserved regulatory function related to sugar transport. Phylogenetically, TPPswSUGAR sequences were primarily identified in Mucoromycota, with some occurrences in Basidiomycota. This distribution pattern indicates potential lineage-specific adaptations of these regulatory elements. Although these structural and genomic features provide evidence for a distinct group of TPP riboswitches associated with sugar transporters, it is important to note that experimental validation is crucial. Future studies, including in vitro binding assays and in vivo functional analyses, are necessary to confirm the regulatory activity of these sequences and their specificity for TPP. Such experimental work is essential to fully establish the role of TPPswSUGAR as a new functional group of TPP riboswitches in fungi. Structural Conservation and Impact of P3 Size on TPP Riboswitch Analysis TPP riboswitches exhibit distinct sequence and structural conservation patterns across diverse fungal species. Our analysis of the consensus secondary structures revealed a high degree of conservation in most motifs, with nucleotides crucial for TPP binding being consistently preserved (Figure 5). The P2, P4, and P5 regions demonstrated the highest conservation levels and played pivotal roles in maintaining the overall structure and function of the riboswitch. Consensus secondary structures of established fungal TPP riboswitch groups. Each structure represents a distinct TPP riboswitch group as labeled. Nucleotide conservation is indicated by color, as defined in the legend. Golden asterisks mark nucleotides predicted to interact with TPP. The experimentally validated structure of the E. coli ThiA riboswitch is provided as a reference. This comparison highlights both conserved elements across TPP riboswitches and group-specific structural variations. Notably, the P3 motif exhibited significant variation in sequence identity, length, and base pairing. This variability in P3 contrasts with the more conserved nature of other structural elements. The interaction between the proximal segment of the P3 and P5 nucleotides is crucial for stabilizing the “on-state” of the TPP riboswitch, highlighting the functional importance of P3 despite its variability. Among the established TPP riboswitch groups, TPPswAA showed lower conservation, specifically in the L5 region, and lacked the conserved sequence 5′-RUAAUR-3, ′ which is typically found in this area in other fungal TPP riboswitch groups. This distinctive feature of TPPswAA may have implications for its function and regulation, and warrants further investigation. Our analysis revealed that the putative TPPswSUGAR group displayed the highest overall variability compared with the well-established TPP riboswitch groups (Figures 4 and 5). This group showed notable variations across multiple structural elements, particularly in the P2, P3, and P4 stems, as well as in the J2–4 region. Despite this global variability, some structural elements and nucleotides corresponding to the TPP binding sites in known riboswitches appear to be conserved. It is important to note that the apparent higher variability in the putative TPPswSUGAR group may be influenced by the relatively small sample size of the sequences currently available. The variability in P3 length significantly affects the detection and analysis of TPP riboswitches in fungi. We observed that the P3 stem in fungal TPP riboswitches exceeded the combined length of all the other aptamer regions. For instance, in Scedosporium species, we identified TPP riboswitches with P3 lengths reaching 172 nucleotides, whereas the total aptamer size encompassed 234 nucleotides. This is in stark contrast to bacterial TPP riboswitches, such as the Escherichia coli ThiA riboswitch (PDB ID: 2GDI), which has a total length of 80 nucleotides and only 18 nucleotides in the P3 region. In our data set, the average P3 length among fungal sequences was approximately 150 nucleotides (Figure 6). Distribution of P3 stem lengths in fungal TPP riboswitches. (A) Histogram showing the frequency distribution of P3 stem lengths across all the identified TPP riboswitches. The y-axis represents the relative frequency of the sequences, and the vertical red dashed line indicates the mean P3 stem length. (B) Average P3 stem length for each TPP riboswitch. Bars represent the mean P3 stem length for sequences in each group, with the exact values shown above each bar. This analysis revealed substantial variability in P3 stem length among fungal TPP riboswitches, particularly when compared with their bacterial counterparts. This extensive variability in P3 length poses challenges for TPP riboswitch detection using standard CMs. We found that RF00059 CM from Rfam, which is built with riboswitch sequences from various domains of life, often fails to accurately identify the full sequence of fungal TPP riboswitches, particularly those with longer P3 stems. When the P3 helix exceeds a certain length threshold (typically >60 nucleotides), the output often reports an incomplete sequence, with the detected region ending at some point within the P3 stem. To address this issue, we constructed and calibrated fungal-specific CMs using the TPPswNMT1, TPPswTHI4, and TPPswUREA sequences. These custom CMs demonstrated improved performance in detecting TPP riboswitches in fungal genomes, including those with extended P3 regions that were missed by RF00059 CM. The impact of P3 variability also extends to sequence alignment and clustering. Initial clustering based on the entire riboswitch length resulted in random groupings that did not correlate with the genes regulated (Figure 7A). However, when we excluded P3 stem nucleotides from each sequence and performed reclustering, we observed distinct groupings that correlated with the regulated genes (Figure 7B). Impact of P3 stem on TPP riboswitch clustering. Trees represent hierarchical clustering of TPP riboswitch sequences based on their structural similarity. Colors indicate different TPP riboswitch groups as defined in the legend. (A) Clustering of full-length TPP riboswitch sequences. (B) Clustering of TPP riboswitch sequences with P3 stem nucleotides excluded. This comparison demonstrates that excluding highly variable P3 regions results in improved grouping of sequences according to their associated genes, highlighting the impact of P3 variability on sequence analysis and classification. These findings underscore the importance of considering the unique structural features of fungal TPP riboswitches, particularly the variable P3 region, in detection and strategy. The development of fungal-specific models and careful consideration of P3 variability in sequence comparisons are crucial for the accurate identification and classification of TPP riboswitches in fungal genomes. Further investigations are required to elucidate the underlying factors contributing to P3 variability and confirm the potential regulatory function of the putative TPPswSUGAR sequences. Understanding these structural nuances is essential to unravel the diverse roles of TPP riboswitches in fungal biology and their potential as targets for antifungal therapies. Genomic Context and Regulation Modes of TPP Riboswitches The precise genomic position of the TPP riboswitch dictates its mode of regulation, consistent with previous findings reported in other studies.61,65 In this regard, our results corroborate previous observations from other studies.61,65 Within our data set, type I regulation (Figure S1) emerged as the predominant mode, with the majority of the TPP riboswitch sequences located in the 5′ UTR of the respective genes. Type III regulation (Figure S3) was the second most frequently observed mode, while Type II regulation (Figure S2) was only identified in a subset of the TPPswUREA sequences. Type IV regulation is relatively rare and found in a limited number of species, including Malassezia furfur and Sporothrix schenckii. We observed that TPPswSUGAR primarily employed type I regulation as its main mechanism, although type III regulation was also present. Consequently, most TPPswSUGAR sequences were positioned in the 5′-UTR of the corresponding genes. Discussion Bioinformatics has a significant potential to drive breakthroughs in riboswitch research, particularly in identifying novel riboswitch groups. This approach allows for a comprehensive study of large genome data sets and the correlation of various aspects of riboswitch secondary structure conservation and genomic context, enhancing search robustness88. Our analysis revealed limitations of previous search methods for TPP riboswitches in fungal organisms. For instance, we observed that several fungal TPP riboswitch sequences reported by Mukherjee et al.65 exhibited shorter lengths in the P3 region compared to our findings. This reduced length was often associated with apparently missing or poorly conserved 5′ portions of P1 and P2, as well as a portion of P3. These shorter sequences are potentially problematic as they may lack crucial structural elements of the complete riboswitch. These observations led to incomplete identification of crucial structural elements in some riboswitch sequences. Specifically, Moldovan et al. previously argued that Hortea werneckii lacks TPP riboswitches that regulate thiamine biosynthesis genes.61 However, we identified both TPPswNMT1 and TPPswTHI4 duplicates in the H. werneckii genome. Similarly, Mukherjee et al. did not detect certain TPPswAA sequences in Cryptococcus species.65 The size of the P3 stem plays a crucial role in detecting TPP riboswitches in fungal sequences, often constituting >60% of the total aptamer length. This extension significantly exceeded that observed in bacterial TPP riboswitches.57,67 Our results demonstrate that the P3 stem variability affects sequence alignment and clustering, highlighting the importance of accounting for P3 size when designing CMs for accurate detection of fungal TPP riboswitches. The potential of TPP riboswitches as drug targets has been highlighted in previous studies.92−94 In vitro investigations using TPP analogs as antifungal agents have shown promise in controlling fungal growth, although there are limitations in their suitability as commercial drug candidates.66,95 Our comprehensive search identified TPP riboswitches in over 89% of the fungal species examined, with most genomes harboring two or more instances of these regulatory elements. This prevalence presents an exciting opportunity for antifungal drug development, potentially allowing the design of more potent therapeutic agents capable of simultaneously disrupting TPP production and availability in multiple instances.66,95−99 The discovery of a putative new group of TPP riboswitches associated with sugar transporters (TPPswSUGAR) raises intriguing questions regarding the diversity of riboswitch functions in fungi. While this group shares key structural similarities with known TPP riboswitches, such as conserved nucleotides in the pyrimidine and pyrophosphate binding regions, it also presents distinctive characteristics.100−107 Notably, the TPPswSUGAR sequences exhibited variations in the P2, P3, and P4 stems as well as in the J2–4 region. While the P3 stem in TPPswSUGAR sequences is shorter on average compared to other fungal TPP riboswitch groups, it shows considerable sequence variability. This structural diversity, particularly in the P3 region, distinguishes TPPswSUGAR from other fungal TPP riboswitch groups, and may reflect adaptations to its specific regulatory role in sugar transport. Genomic analysis of these TPPswSUGAR sequences revealed a consistent association with sugar transport genes in the 5′ UTR, primarily in Mucoromycota species, with some occurrences in Basidiomycota. This pattern suggests a potential regulatory role and lineage-specific adaptation of the TPPswSUGAR. However, experimental validation is necessary to confirm their function as riboswitches and their specific roles in regulating sugar transporter genes. Further investigation of these structural variations could provide insight into the evolution and functional diversity of TPP riboswitches in fungi. The prevalence of multiple TPP riboswitches within individual fungal genomes, particularly in species associated with aspergillosis or mucormycosis, may indicate that these pathogens could be more susceptible to therapeutic interventions targeting TPP riboswitches. The absence of TPP riboswitches in the human genome57,67 might offer an additional advantage, potentially resulting in fewer side effects and reduced harm to patients. Our findings also highlight the need for further research into the diversity of fungal riboswitches. The differences in P3 stem length between bacteria and fungi, which hindered the detection of certain fungal organisms, suggest the possibility of undiscovered riboswitch classes in fungi that may possess substantial structural differences compared with bacterial riboswitches. In summary, this study expands our understanding of TPP riboswitch diversity in fungi and emphasizes its potential as a promising target for antifungal therapies. The unique characteristics and prevalence of these riboswitches across diverse fungal species underscore their significance in fungal biology and in the development of novel therapeutic interventions. Future research should focus on experimental validation of the putative TPPswSUGAR group and exploration of the functional roles of multiple TPP riboswitches in individual fungal genomes. Conclusions The discovery of a previously unknown group of putative TPP riboswitches in fungi that are associated with a family of sugar transporters presents exciting opportunities for drug targeting in fungal organisms. Our comprehensive analysis revealed that TPP riboswitches have the potential to interfere with various stages of the TPP transport and biosynthesis pathways. Remarkably, our investigation of a wide range of fungal species demonstrated that 89% of them possessed at least one TPP riboswitch, with the majority harboring two or more TPP riboswitches within their genomes. The TPPswTHI4 group exhibited the highest abundance in Ascomycota, whereas the TPPswAA group predominated in Basidiomycota, and TPPswSUGAR was primarily found in Mucoromycota. Furthermore, our study highlights the impact of P3 size on the search for and alignment of fungal TPP riboswitches. To address the limitations of existing models, we developed novel covariance models tailored specifically for fungi, enabling the accurate identification of TPP riboswitch sequences. These findings not only expand our understanding of TPP riboswitch diversity in fungi but also emphasize their potential as promising targets for antifungal therapies. Although the identification of the putative TPPswSUGAR group is intriguing, we acknowledge the need for experimental validation to confirm its regulatory function and specificity for TPP. The unique characteristics and prevalence of these riboswitches across diverse fungal species underscore their significance in fungal biology and set the stage for future research on novel therapeutic interventions. Methods Data Collection We retrieved 60 fungal TPP riboswitch sequences from the NCBI Nucleotide database based on the genome coordinates reported by Mukherjee et al.65 Sequences were equally distributed among the three fungal TPP riboswitch groups: NMT1,108 THI4,109 and urea transporters.62 We categorized sequences based on their associated target genes and used RNAfold to predict secondary structures. These were compared to the consensus structures reported by Mukherjee et al.65 using Varna110 for visualization. For sequences with shorter lengths in the P3 region, often associated with apparently missing or poorly conserved 5′ portions of P1 and P2 as well as part of P3, we extended the sequence by incorporating additional 5′ nucleotides until matching the consensus structure. This approach ensured that we captured the full length of the riboswitch, including all the crucial structural elements. We compiled a list of 131 medically relevant fungal species (Table S1) and obtained their complete genomes from the RefSeq database.111 CM Construction We downloaded the RF00059 covariance model (CM) from Rfam.89−91 Using the 60 refined sequences (20 each from NMT1, THI4, and urea transporter groups), we constructed three new CMs using Infernal’s cmbuild and cmcalibrate functions.112−114 These CMs served as comparison parameters to investigate the presence of riboswitches in medically relevant fungal genomes. TPP Riboswitch Search We employed Infernal’s cmsearch to compare RF00059 CM and our three custom CMs against 131 fungal genomes. As the CM scores were not directly comparable, we selected the broadest genomic coordinates and manually refined them using the approach described for the initial 60 sequences. We validated secondary structures using RNAfold115 energy minimization and compared them to reported consensus structures, visualizing them with Varna.110 Sequence length was adjusted to match the consensus structures. To identify open reading frames (ORFs) and verify the genomic context for each sequence, we used the NCBI ORFfinder and Artemis databases, respectively.116,117 To infer the gene product of the sequences and confirm annotation data, we conducted BLASTx searches against the UniProtKB database and the Transporter Classification Database (TCDB).116,120,121 Additionally, we detected conserved structural domains of putative transporters using DeepTMHMM.122 Clustering of TPP Riboswitch Sequences and Tree Generation The sequences were subjected to clustering using RNAclust, a Perl script that integrates tools from the Vienna RNA package, and LocARNA.115,118 RNAclust utilizes the base pair probability matrix of each sequence’s secondary structure distribution, calculated using RNAfold, to identify groups of RNA sequences sharing similar secondary structure motifs. The LocARNA tool generates a sequence-structure alignment for each pair of base-pair probability matrices. Hierarchical clustering (weighted pair group method with arithmetic mean) was then applied by RNAclust to construct a hierarchical cluster tree based on pairwise alignment distances. The optimal number of clusters is determined using the resulting tree. Finally, we used iTOL version 6 to visualize the cluster tree.119 Alignment and Visualization of Secondary Structures Alignments of TPP riboswitch sequences against the RF00059 CM were generated using the Infernal Cmalign tool.113,114 The resulting alignments were visualized using the Aline software, providing a comprehensive depiction of sequence conservation and variation.123 To further analyze the structural characteristics, consensus secondary structure models were generated using R2R software.124 These structural models provide an overview of the conserved folding patterns within TPP riboswitches, highlighting key elements crucial for their function and regulation.