Biology subjects

Zhu, W.

Publications and source records attributed to Zhu, W..

At least 73 records · Page 4Linked to original sources

bioRxiv2025

Coro1a promotes the multivesicular body and plasma membrane fusion by facilitating PKM2-mediated SNAP-23 phosphorylation

During the formation of endosomal pathway-dependent extracellular vesicles (EVs), intraluminal vesicles in multivesicular bodies (MVBs) fuse with the plasma membrane (PM), releasing intraluminal vesicles to produce exosomes. Soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex, membrane fusion effector, is essential in this process. However, which SNARE complex is involved in MVB and PM fusion and how its assembly is regulated remain elusive. We have demonstrated that neddylation of Coro1a inhibits EV secretion by inducing Rab7-mediated degradation of MVBs. Here, we find that Coro1a increases EV biogenesis by promoting the assembly of SNARE complex STX-12-SNAP-23-VAMP-7 in a neddylation-independent manner. Coro1a activates PKM2 to enhance SNAP-23 phosphorylation in the presence of neddylation inhibitor MLN4924, which drives SNAP-23 to recruit more STX-12 and VAMP7, leading to increased formation of the STX-12-SNAP-23-VAMP-7 complex. Correspondingly, Coro1a-induced EV biogenesis is eliminated by PKM2 inhibitor or SNAP-23 silencing independent of neddylation. Reduced EVs from tumors due to Coro1a knockout cannot be observed in MLN4924-treated tumor mice with PKM2 inhibitor treatment either. Furthermore, Coro1a increases in tumor tissues from lung tumor patients with tumor progression, and high Coro1a indicates unfavorable survival of tumor patients, suggesting that increased Coro1a-mediated enhanced EV production from tumors accelerates tumor progression. Altogether, our data demonstrate that Coro1a facilitates EV biogenesis by promoting the assembly of STX-12-SNAP-23-VAMP-7 complex during MVB and PM fusion independent of neddylation, thereby being involved in the progression of EV-related diseases.

cell biology↗

bioRxiv2025

Nonlinear Relationship Between the C-Reactive Protein to Albumin Ratio and Chronic Kidney Disease: A Cross-Sectional Analysis Using the NHANES Database

ObjectiveThe C-reactive protein to albumin ratio (CAR) has gained attention as an inflammatory biomarker with potential relevance in various diseases. However, its relationship with chronic kidney disease (CKD) remains inadequately understood. This study aims to investigate the association between CAR and CKD prevalence in a diverse, multi-ethnic U.S. cohort. MethodsThis cross-sectional analysis utilized data from the 1999-2010 National Health and Nutrition Examination Survey (NHANES). Multivariable regression and subgroup analyses were conducted to control for confounding factors and examine the link between CAR and CKD. A restrictive cubic spline model was employed to explore potential nonlinear associations. ResultsAmong 30,049 participants aged 18 years or older, 5,423 (18.05%) had CKD. Compared to the lowest quartile (Q1) of log-transformed CAR, CKD risk was higher in the second (Q2), third (Q3), and fourth (Q4) quartiles, with odds ratios of 1.28, 1.29, and 1.63, respectively. A dose-response relationship was observed, indicating that higher CAR levels were significantly associated with increased CKD risk. ConclusionElevated CAR is independently associated with a higher prevalence of CKD, suggesting its potential utility as a biomarker for CKD risk assessment in diverse populations. Further studies are needed to confirm these findings and explore underlying mechanisms.

bioinformatics↗

bioRxiv2025

SRS microscopy identifies inhibition of vitellogenesis as a mediator of lifespan extension by caloric restriction in C. elegans

The molecular mechanisms of aging are not fully understood. Here, we used label-free Stimulated Raman scattering (SRS) microscopy to investigate changes in proteins and lipids throughout the lifespan of C. elegans. We observed a dramatic buildup of proteins within the body cavity or pseudocoelom of aged adults that was blunted by interventions that extend lifespan: caloric restriction (CR) and the reduced insulin/insulin-like growth factor signaling (IIS) pathway. Using a combination of microscopy, proteomic analysis, and validation with mutant strains, we identified vitellogenins as the key molecular components of the protein buildup in the pseudocoelom. Vitellogenins shuttle nutrients from intestine to embryos and are homologous to human apolipoprotein B, the causal driver of cardiovascular disease. We then showed that CR and knockdown of vitellogenins both extend lifespan by >60%, but their combination has no additional effect on lifespan, suggesting that CR extends the lifespan of C. elegans in part by inhibiting vitellogenesis. The extensive dataset of more than 12,000 images stitched into over 350 whole-animal SRS images of C. elegans at different ages and subjected to different longevity intervention will be a valuable resource for researchers interested in aging.

cell biology↗

bioRxiv2025

The evolutionary dynamics of organellar pan-genomes in Arabidopsis thaliana

In plants, comparative analyses of organellar genomes often depend on draft assemblies, with large-scale investigations into the complex structural rearrangements of mitochondrial genomes remaining scarce. Here, we conducted a comprehensive analysis of the dominant conformations and dynamic heteroplasmic variants of organellar genomes in the model plant Arabidopsis thaliana, utilizing high-quality long-read assemblies validated at read-level resolution from 149 samples. We found that mitochondrial and plastid genomes share common types of structural and small-scale variants driven by similar DNA sequence features. However, rearrangements mediated by repetitive sequences in mitochondrial genomes evolved so rapidly that they often became decoupled from other types of variants. Rare complex events involving elongation and fusion were also observed, contributing to the unalignable regions commonly found at the interspecies level. Additionally, we demonstrated that disrupting and rescuing organellar DNA maintenance could drive the rapid evolution of dominant mitochondrial genome conformations. Our study provides an unprecedentedly detailed view of the dynamics of organellar genomes at pan-genome scale in A. thaliana, paving the way to unlock the full potential of organellar genetic resources. SignificancePlant mitochondrial genomes exhibit rapid rearrangements within species. Large-scale analyses using high-quality assemblies offer detailed insights into their evolutionary dynamics, paving the way for organellar genome engineering and functional research in the future.

evolutionary biology↗

bioRxiv2025

Characterising the periodontal granulation tissue using scRNAseq.

AimsOur research investigated the cellular composition and molecular mechanisms of periodontal granulation tissue using single-cell RNA sequencing (scRNA-seq), aiming to enhance the understanding of periodontal disease pathogenesis and identify potential targets for regenerative therapies. Materials and MethodsGranulation tissue samples were collected from patients undergoing periodontal surgery. Fresh tissues were processed into single-cell suspensions and subjected to scRNA-seq. The data were integrated with existing datasets from healthy gingiva and periodontal ligament. Computational analyses were performed, and validation through immunofluorescence staining. ResultsTen distinct cell clusters were identified across the samples. Granulation tissue exhibited a higher abundance of immune cells compared to healthy tissues. A novel endothelial cell subpopulation, exclusive to granulation tissue, was discovered, characterized by NOTCH3 expression and involvement in ossification pathways. Additionally, granulation tissue Fibroblast subpopulations demonstrated a progenitor-like state, characterized by extracellular matrix reorganization and low differentiation, similar to cancer-associated fibroblasts. Conclusion(s)This study advances the understanding of periodontal disease by characterizing key regulatory cell populations within granulation tissue. The identification of the novel endothelial subpopulation offers new insights into the diseases pathogenesis and presents potential targets for regenerative therapies. These findings suggest opportunities for developing biomaterials that modulate specific cellular pathways to improve periodontal disease treatment.

cell biology↗

bioRxiv2025

Sparse Sequencing permits accurate and efficient quantification of genome-wide cytosine modification levels

5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) regulate gene expression and exhibit dynamic levels during development and disease. While high-depth, base-resolution studies offer the most detailed view of epigenetic landscapes, many open questions are answered by surveying changes in 5mC/5hmC levels across larger cohorts. Nonetheless, current global quantification methods, including mass spectrometry, are typically limited in accessibility, accuracy, or throughput. Here, to evaluate the viability of low-coverage sequencing as an alternative, we first computationally downsampled deeply sequenced data to resolve the three-way relationship between sequencing coverage, modification levels, and measurement error. This relationship allowed us to develop a facile online tool for error calculation and to define experimental targets: <0.24% genome coverage can quantify 5mC and low-abundance 5hmC with minimal and predictable errors (<5%). Importantly, in direct comparisons, low-depth sequencing (Sparse-Seq) demonstrated high accuracy and less variability than mass spectrometry, while distinctively preserving genomic context. Applied serially to developing mouse brains, Sparse-Seq revealed an earlier emergence of 5hmCpG compared to 5mCpH and uncovered previously overlooked, genomic feature-specific epigenetic changes. This work establishes a rigorous foundation for employing Sparse-Seq as a highly accessible approach for 5mC/5hmC quantification, enabling economical first-pass analysis of epigenetic landscapes suited for large cohort studies and new hypothesis generation.

genomics↗

bioRxiv2025

Pathobiont-triggered induction of epithelial IDO1 drives regional susceptibility to Inflammatory Bowel Disease

The structure and function of the mammalian gut vary by region, yet why inflammatory diseases manifest in specific regions and not others remains unclear. We use a TNF-overexpressing Crohns disease (CD) model (Tnf{Delta}ARE/+), which typically presents in the terminal ileum (TI), to investigate how environmental factors interact with the hosts immune susceptibility to drive region-specific disease. We identified Chlamydia muridarum, an intracellular bacterium and murine counterpart to the human sexually transmitted C. trachomatis, as necessary and sufficient to trigger disease manifestation in the ascending colon (AC), another common site of human CD. Disease manifestation in the AC depends on indoleamine 2,3-dioxygenase (IDO1) activity induced by hypersensitive surface secretory cells in genetically susceptible hosts. Single-cell and microbial analyses of human specimens also implicates this pathobiont-epithelial IDO1 pathway in patients with a history of CD in the AC. Our findings demonstrate that genetic and microbial factors can independently drive region-specific disease and provide a unique model to study CD specific to the AC.

cell biology↗

bioRxiv2024

Protection against N. gonorrhoeae induced by OMV-based Meningococcal Vaccines are associated with cross-species directed humoral and cellular immune responses

Limited protective immunologic responses to natural N. gonorrhoeae infection and a lack of knowledge about mechanisms of protection have hampered development of an effective vaccine. Recent studies in humans and mice have found meningococcal outer membrane vesicle-containing vaccines (OMV) induce cross species immune responses against gonococci and are associated with protection. The exact mechanisms or how humoral and cellular immunity are related to protection, remain unclear. To study this, we immunized mice with two meningococcal OMV-containing vaccines known to accelerate clearance of N. gonorrhoeae, 4CMenB and OMV from an engineered N. meningitidis strain lacking major surface antigens PorA, PorB, and Rmp (MC58 {Delta}ABR). We assessed serologic and cellular immune signatures associated with these immunizations and assessed bacterial clearance in the mice using a vaginal/cervical gonococcal infection model. Mice immunized with 4CMenB or MC58 {Delta}ABR demonstrated shortened courses of recovery of vaginal N. gonorrhoeae compared to control mice immunized with alum alone. Vaccination with 4CMenB or MC58{Delta}ABR OMV elicited serum and vaginal cross-reactive anti-Ng-OMV antibody responses that were augmented after vaginal challenge with N. gonorrhoeae. Further, splenocytes in 4CMenB and MC58 {Delta}ABR immunized mice exhibited elevated cytokine production after restimulation with heterologous N. gonorrhoeae OMV when compared to splenocytes from Alum immunized mice. We further tested for correlations between bacterial burden and the measured anti-gonococcal immune responses within each vaccination group and found different immunologic parameters associated with reduced bacterial burden for each vaccine. Our findings suggest the cross-protection against gonococcal infection induced by different meningococcal OMV vaccines is likely multifactorial and mediated by different humoral and cellular immune responses induced by these two vaccines.

microbiology↗

bioRxiv2024

TFAM as a sensor of UVC-induced mitochondrial DNA damage

Mitochondria lack nucleotide excision repair; however, mitochondrial DNA (mtDNA) is resistant to mutation accumulation following DNA damage. These observations suggest additional damage sensing or protection mechanisms. Transcription Factor A, Mitochondrial (TFAM) compacts mtDNA into nucleoids and binds differentially to certain forms of DNA damage. As such, TFAM has emerged as a candidate for protecting mtDNA or sensing damage. To examine the possibilities that TFAM might protect DNA from damage or act as a damage sensing protein for irreparable forms of mtDNA damage, we used live-cell imaging and HeLa cell-based assays, atomic force microscopy (AFM), and high-throughput protein-DNA binding assays to characterize the binding properties of human TFAM to ultraviolet-C (UVC) irradiated DNA and the cellular consequences of UVC irradiation. Our cell data show increased TFAM mRNA after exposure and suggest an increase in mtDNA degradation without a loss in mitochondrial membrane potential that might trigger mitophagy. Our protein-DNA binding assays indicate a reduction in sequence specificity of TFAM following UVC irradiation and a redistribution of TFAM binding throughout the mitochondrial genome. Our AFM data show increased compaction of DNA by TFAM in the presence of damage. Despite the TFAM-mediated compaction of mtDNA in vitro, we do not observe any protective effect of increased TFAM protein on DNA damage formation in cells or in vitro. Increased TFAM protein did not alter levels of mtDNA damage over time after UVC exposure in vivo, but knockdown of TFAM did alter mtDNA damage levels in HeLa cells both at baseline and after UVC exposure. Taken together, these studies indicate that UVC-induced DNA damage alters TFAM binding and promotes compaction by TFAM in vitro. We hypothesize that that TFAM may act as a damage sensing protein in vivo, sequestering damaged genomes to prevent mutagenesis by facilitating removal or suppression of replication.

biochemistry↗

bioRxiv2024

DEMINERS enables clinical metagenomics and comparative transcriptomic analysis by increasing throughput and accuracy of nanopore direct RNA sequencing

Nanopore direct RNA sequencing (DRS) advances RNA biology but is limited by relatively low basecalling accuracy, low throughput, yet high RNA input and costs. Here we introduce a novel DRS toolkit, DEMINERS, which integrates an RNA multiplexing experimental workflow, a machine-learning barcode classifier based on Random Forest and a novel basecaller built on an optimized convolutional neural network providing an additional species-specific training module. With the increased accuracy in barcode classification and basecalling, DEMINERS can demultiplex up to 24 samples and the required RNA input and running time are both substantially reduced. We demonstrated the applications of DEMINERS in clinical metagenomics, cancer transcriptomics and parallel comparison of transcriptomic features in different biological conditions, revealing altered airway microbial diversity in COVID-19 and a potential role of m6A in increasing transcriptomic diversity in glioma and the mature blood-stage of malaria parasites. Overall, DEMINERS is a simple, robust, high-throughput DRS method for accurately estimating transcript levels, poly(A) lengths, and mutation and RNA modification heterogeneity at single-read level, with minimal sequencing biases.

bioinformatics↗

bioRxiv2024

Antipsychotic-like effects of the selective Rho-kinase 2 inhibitor KD025 in genetic and pharmacological mouse models of schizophrenia

1Copy number variations in the ARHGAP10 gene encoding Rho GTPase-activating protein 10 are significantly associated with schizophrenia. ARHGAP10 negatively regulates RhoA/Rho-kinase (ROCK) signaling. We previously demonstrated that fasudil, a non-selective ROCK inhibitor, exhibited antipsychotic-like effects in several mouse models of schizophrenia. ROCK has two subtypes, ROCK1 and ROCK2. ROCK1 is mainly expressed in the thymus and blood, while ROCK2 is predominantly expressed in the brain. Therefore, it is expected that like fasudil, selective ROCK2 inhibitors will exhibit antipsychotic-like effects, accompanied by a lower incidence of adverse effects due to ROCK1 inhibition. Here, we used genetic and pharmacological models of schizophrenia to investigate whether the selective ROCK2 inhibitor KD025 would show antipsychotic-like effects with a favorable adverse effect profile. Oral administration of KD025 suppressed the abnormal increase in the phosphorylation level of myosin phosphatase-targeting subunit 1, a substrate of ROCK, and ameliorated the decreased spine density of layer 2/3 pyramidal neurons in the medial prefrontal cortex of Arhgap10 S490P/NHEJ mice. Furthermore, KD025 mitigated the methamphetamine-induced impairment of visual discrimination (VD) in Arhgap10 S490P/NHEJ and wild-type mice. KD025 also reduced MK-801-induced impairments of VD, novel object recognition, and hyperlocomotion. Regarding side effects that are commonly seen with typical antipsychotics, KD025 did not affect systolic blood pressure and did not induce extrapyramidal symptoms, hyperprolactinemia, or hyperglycemia at the effective dosage in naive wild-type mice. Taken together, KD025 shows antipsychotic-like effects with a favorable adverse effect profile in genetic and pharmacological mouse models of schizophrenia.

pharmacology and toxicology↗

bioRxiv2024

Deep contrastive learning enables genome-wide virtual screening

Numerous protein-coding genes are associated with human diseases, yet approximately 90% of them lack targeted therapeutic intervention. While conventional computational methods, such as molecular docking, have facilitated the discovery of potential hit compounds, the development of genome-wide virtual screening against the expansive chemical space remains a formidable challenge. Here we introduce DrugCLIP, a novel framework that combines contrastive learning and dense retrieval to achieve rapid and accurate virtual screening. Compared to traditional docking methods, DrugCLIP improves the speed of virtual screening by up to seven orders of magnitude. In terms of performance, DrugCLIP not only surpasses docking and other deep learning-based methods across two standard benchmark datasets, but also demonstrates high efficacy in wet-lab experiments. Specifically, DrugCLIP successfully identified agonists with < 100 nM affinities for 5HT2AR, a key target in psychiatric diseases. For another target NET, whose structure is newly solved and not included in the training set, our method achieved a hit rate of 15%, with 12 diverse molecules exhibiting affinities better than bupropion. Additionally, two chemically novel inhibitors were validated by structure determination with Cryo-EM. Finally, a novel potential drug target TRIP12, with no experimental structures and inhibitors for reference, was used to challenge DrugCLIP. DrugCLIP achieved a hit rate of 17.5% by screening a pocket identified on an AlphaFold2-predicted structure, verified with multi-cycle SPR assays. Molecules with the highest affinities also showed a dose-dependent inhibition to the enzymatic function of TRIP12. Building on this foundation, we present the results of a pioneering trillion-scale genome-wide virtual screening, encompassing approximately 10,000 AlphaFold2 predicted proteins within the human genome and 500 million molecules from the ZINC and Enamine REAL database. This work provides an innovative perspective on drug discovery in the post-AlphaFold era, where comprehensive targeting of all disease-related proteins is within reach.

bioinformatics↗

bioRxiv2024

Field conditions greatly modify a major growth-defense tradeoff in Arabidopsis thaliana

When plants defend themselves from pathogens, this often comes with a trade-off: the same genes that protect a plant from disease can also reduce its growth and fecundity in the absence of pathogens. One protein implicated in a major growth-defense trade-off in the plant Arabidopsis thaliana is ACCELERATED CELL DEATH 6 (ACD6), an ion channel that modulates salicylic acid (SA) synthesis to potentiate a wide range of defenses. Wild plant populations maintain significant functional variation in the ACD6 gene, with some alleles making the protein hyperactive. In the greenhouse, plants with hyperactive ACD6 alleles are resistant to diverse pathogens, yet are of smaller stature, their leaves senesce earlier, and they set fewer seeds. We hypothesized that such hyperactive alleles would not only affect the growth of microbial pathogens, but also more generally leaf microbiome assembly in the wild. To test this, we grew plants with hyperactive, standard, and defective ACD6 alleles in the same field-collected soil, both in climate-controlled conditions and outdoors. We surveyed visual phenotypes, gene expression, hormone levels, seed production, and the microbiome in each environment. To our surprise, we discovered that mature field plants, in stark contrast to greenhouse plants, were unaffected by their ACD6 genotype, suggesting that additional abiotic and/or microbial signals present outdoors - but not in the greenhouse - greatly modulate ACD6 activity.

plant biology↗

bioRxiv2024

PepCleav: a classification method for evaluating the cleavability of peptide fragments presented by MHC class I

Neoantigens, due to their tumor specificity and lack of central tolerance, are promising targets for cancer immunotherapy. Evaluating the cleavability of peptide fragments generated by proteasomes and aminopeptidases is crucial for neoantigen identification, but tools for this purpose are scarce. To address this, we developed PepCleav, a method for evaluating peptide cleavability by classifying the amino acids at the proteasome and aminopeptidase recognition sites of the C- and N-termini. By integrating C- and N-terminal cleavability, PepCleav accurately predicted the overall cleavability of peptide fragments in test datasets, correctly identifying 84-89% of cleavable peptides and 63-92% of non-cleavable peptides. We also found that highly cleavable peptides have a higher likelihood of being effective neoantigens, highlighting PepCleavs potential to improve neoantigen identification. PepCleavs source code is publicly available at https://github.com/Dulab2020/PepCleav.

bioinformatics↗

bioRxiv2024

F-CPI: Prediction of activity changes induced by fluorine substitution using multimodal deep learning

There are a large number of fluorine (F)-containing compounds in approved drugs, and F substitution is a common method in drug discovery and development. However, F is difficult to form traditional hydrogen bonds and typical halogen bonds. As a result, accurate prediction of the activity after F substitution is still impossible using traditional drug design methods, whereas artificial intelligence driven activity prediction might offer a solution. Although more and more machine learning and deep learning models are being applied, there is currently no model specifically designed to study the effect of F on bioactivities. In this study, we developed a specialized deep learning model, F-CPI, to predict the effect of introducing F on drug activity, and tested its performance on a carefully constructed dataset. Comparison with traditional machine learning models and popular CPI task models demonstrated the superiority and necessity of F-CPI, achieving an accuracy of approximately 89% and a precision of approximately 67%. In the end, we utilized F-CPI for the structural optimization of hit compounds against SARS-CoV-2 3CLpro. Impressively, in one case, the introduction of only one F atom resulted in a more than 100-fold increase in activity (IC50: 22.99 nM vs. 28190 nM). Therefore, we believe that F-CPI is a helpful and effective tool in the context of drug discovery and design.

bioinformatics↗

bioRxiv2024

Droplet-AI-empowered high-throughput screening of cell-free gene expression

Cell-free gene expression (CFE) systems enable transcription and translation using crude cellular extracts, offering a versatile platform for synthetic biology by eliminating the need to maintain living cells. This allows direct manipulation of molecular components and the focused synthesis of specific products. However, the optimization of CFE systems is constrained by cumbersome composition, high costs, and limited yields due to numerous additional components required to maintain biocatalytic efficiency. While optimizing such complicated systems is daunting for existing high-throughput screening means, we introduce DropAI, a droplet-based, AI-driven screening strategy designed to optimize CFE systems with high throughput and economic efficiency. DropAI employs microfluidics to generate picoliter reactors and utilizes a fluorescent color-based coding-decoding system to address and screen a vast array of additive combinations. The in-droplet screening is complemented by in silico optimization, where experimental results train a machine-learning model to estimate the contribution of the components and predict high-yield combinations, which are then validated in vitro. Applying DropAI to an Escherichia coli-based CFE system, we simplified a set of 12 additives to only 3 essential components. Through further optimization, we achieved a 2.1-fold cost reduction and a 1.9-fold increase in yield for the expression of superfolder green fluorescent protein (sfGFP). This optimized formulation was further validated across 12 different proteins. Notably, the established E. coli model is successfully adapted to a Bacillus subtilis-based system through transfer learning, leading to doubled yield through prediction. DropAI thus offers a generalizable and scalable method for optimizing CFE systems, enhancing their potential for biochemical engineering and biomanufacturing applications.

bioengineering↗

bioRxiv2024

CRISPR screens reveal convergent targeting strategies against evolutionarily distinct chemoresistance in cancer

Resistance to chemotherapy has been a major hurdle that limits therapeutic benefits for many types of cancer. Here we systematically identify genetic drivers underlying chemoresistance by performing 30 genome-scale CRISPR knockout screens for seven chemotherapeutic agents in multiple cancer cells. Chemoresistance genes vary between conditions primarily due to distinct genetic background and mechanism of action of drugs, manifesting heterogeneous and multiplexed routes towards chemoresistance. By focusing on oxaliplatin and irinotecan resistance in colorectal cancer, we unravel that evolutionarily distinct chemoresistance can share consensus vulnerabilities identified by 26 second-round CRISPR screens with druggable gene library. We further pinpoint PLK4 as a therapeutic target to overcome oxaliplatin resistance in various models via genetic ablation or pharmacological inhibition, highlighting a single-agent strategy to antagonize evolutionarily distinct chemoresistance. Our study not only provides resources and insights into the molecular basis of chemoresistance, but also proposes potential biomarkers and therapeutic strategies against such resistance.

cancer biology↗

bioRxiv2024

Evolution of a Eukaryotic Transcription Factor's co-TF Dependence Involves Multiple Intrinsically Disordered Regions Affecting Activation and Autoinhibition

Combinatorial control by transcription factors (TFs) is central to eukaryotic gene regulation, yet its mechanism, evolution, and regulatory impact are not well understood. Here we use natural variation in the yeast phosphate starvation (PHO) response to examine the genetic basis and species variation in TF interdependence. In Saccharomyces cerevisiae, main TF Pho4 relies on the co-TF Pho2 to regulate [~]28 genes, whereas in the related pathogen Candida glabrata, Pho4 has reduced Pho2 dependence and regulates [~]70 genes. We found C. glabrata Pho4 (CgPho4) binds the same motif with 3-4 fold higher affinity. Machine learning and yeast one-hybrid assay identify two intrinsically disordered regions (IDRs) in CgPho4 that boost its activation domains activity. In ScPho4, an IDR next to the DNA binding domain both allows for enhanced activity with Pho2 and inhibits activity without Pho2. This study reveals how IDR divergence drives TF interdependence evolution by influencing activation potential and autoinhibition.

evolutionary biology↗