Multitudes of organisms display metameric compartmentalization of their body plan. Segmentation of these compartments happens sequentially in diverse phyla. In several sequentially segmenting species, periodically active molecular clocks and signaling gradients have been found. The clocks are proposed to control the timing of segmentation, while the gradients are proposed to instruct the positions of segment boundaries. However, the identity of the clock and gradient molecules differs across species. Furthermore, sequential segmentation of a basal chordate, Amphioxus, continues at late stages when the small tail bud cell population cannot establish long-range signaling gradients. Thus, it remains to be explained how a conserved morphological trait (i.e., sequential segmentation) is achieved by using different molecules or molecules with different spatial profiles. Here, we first focus on sequential segmentation of somites in vertebrate embryos and then draw parallels with other species. Thereafter, we propose a candidate design principle that has the potential to answer this puzzling question.

Lung cancer cells often show elevated levels of reactive oxygen species (ROS) and nicotinamide adenine dinucleotide phosphate (NADPH). However, the connections between deregulated redox homeostasis in different subtypes of lung cancer and acquired drug resistance in lung cancer have not yet been fully established. Herein, we analyzed different subtypes of lung cancer data reported in the Cancer Cell Line Encyclopedia (CCLE) database, the Cancer Genome Atlas program (TCGA), and the sequencing data obtained from a gefitinib-resistant non-small-cell lung cancer (NSCLC) cell line (H1975GR). Using flux balance analysis (FBA) model integrated with multiomics data and gene expression profiles, we identified cytosolic malic enzyme 1 (ME1) and glucose-6-phosphate dehydrogenase as the major contributors to the significantly upregulated NADPH flux in NSCLC tissues as compared with normal lung tissues, and gefitinib-resistant NSCLC cell line as compared with the parental cell line. Silencing the gene expression of either of these two enzymes in two osimertinib-resistant NSCLC cell lines (H1975OR and HCC827OR) exhibited strong antiproliferative effects. Our findings not only underscored the pivotal roles of cytosolic ME1 and glucose-6-phosphate dehydrogenase in regulating redox states in NSCLC cells but also provided novel insights into their potential roles in drug-resistant NSCLC cells with disturbed redox states.

The thioredoxin pathway is an antioxidant system present in most organisms. Electrons flow from a thioredoxin reductase to thioredoxin at the expense of a specific electron donor. Most known thioredoxin reductases rely on NADPH as a reducing cofactor. Yet, in 2016, a new type of thioredoxin reductase was discovered in Archaea which utilize instead a reduced deazaflavin cofactor (F420 H2 ). For this reason, the respective enzyme was named deazaflavin-dependent flavin-containing thioredoxin reductase (DFTR). To have a broader understanding of the biochemistry of DFTRs, we identified and characterized two other archaeal representatives. A detailed kinetic study, which included pre-steady state kinetic analyses, revealed that these two DFTRs are highly specific for F420 H2 while displaying marginal activity with NADPH. Nevertheless, they share mechanistic features with the canonical thioredoxin reductases that are dependent on NADPH (NTRs). A detailed structural analysis led to the identification of two key residues that tune cofactor specificity of DFTRs. This allowed us to propose a DFTR-specific sequence motif that enabled for the first time the identification and experimental characterization of a bacterial DFTR.

Transthyretin (TTR) is a carrier protein for thyroid hormone thyroxine (T4 ) in plasma, placental cytosol, and cerebrospinal fluid. While the potential toxicity of small molecules that compete with T4 for binding to TTR should be carefully studied, these small molecules can also serve as anti-ATTR amyloidosis drugs by stabilizing the TTR structure. Here, we demonstrated that rafoxanide, an EU-approved anthelmintic drug for domesticated animals, binds to the T4 -binding site of TTR. An intrinsic fluorescence quenching assay showed that rafoxanide also binds to the thyroid hormone-related proteins, including serum albumin and thyroid hormone receptor β. Rafoxanide strongly inhibited TTR amyloidogenesis in fibrillization assay, but the binding of rafoxanide to TTR was interfered with in human plasma, probably due to interactions with thyroid hormone-related proteins. Protein crystallography provided clues for the optimization of binding affinity and selectivity. Our findings emphasize the importance of considering rafoxanide as both a possible thyroid-disrupting chemical and a lead compound for the development of new ATTR amyloidosis inhibitors.

MicroRNAs (miRNAs) are small noncoding RNAs that control essential cellular processes. For several decades, the molecular mechanisms underlying the functions and biogenesis of miRNAs have been clarified, whereas the molecular dynamics of miRNAs are poorly understood. We recently found that muscle-enriched miRNAs were reduced by only 20 ~ 50% in the skeletal muscles even 4 weeks after the suppression of miRNA processing through an inducible depletion of Dicer1 gene. These data suggest that miRNAs are stably expressed in skeletal muscle. In this study, we investigated the half-lives of those miRNAs in adult skeletal muscle with an in vivo metabolic labeling strategy and a genetic mouse model. In contrast to the hypothesis, in vivo metabolic labeling revealed that the half-lives of skeletal-muscle-enriched miRNAs were approximately 11-20 h. Furthermore, the levels of mature miR-23a decreased rapidly in the skeletal muscle of mice lacking miR-23 clusters in a tamoxifen-inducible manner. These data suggest that skeletal-muscle-enriched miRNAs are not highly stable in vivo. We also observed that the transfer of miR-150 into Dicer1-deficient muscle increased the miR-150 level to the same as that in control muscle. Taken together, our data demonstrate that miRNAs are degraded within a few days in adult skeletal muscle and that a Dicer-independent biogenetic pathway may produce mature miRNAs.

The dimethyl sulfone monooxygenase system is a two-component flavoprotein, catalyzing the monooxygenation of dimethyl sulfone (DMSO2 ) by oxidative cleavage producing methanesulfinate and formaldehyde. The reductase component (DMSR) is a flavoprotein with FMN as a cofactor, catalyzing flavin reduction using NADH. The monooxygenase (DMSMO) uses reduced flavin from the reductase and oxygen for substrate monooxygenation. DMSMO can bind to FMN and FMNH- with a Kd of 17.4 ± 0.9 μm and 4.08 ± 0.8 μm, respectively. The binding of FMN to DMSMO is required prior to binding DMSO2 . This also applies to the fast binding of reduced FMN to DMSMO followed by DMSO2 . Substituting reduced DMSR with FMNH- demonstrated the same oxidation kinetics, indicating that FMNH- from DMSR was transferred to DMSMO. The oxidation of FMNH- :DMSMO, with and without DMSO2 did not generate any flavin adducts for monooxygenation. Therefore, H2 O2 is likely to be the reactive agent to attack the substrate. The H2 O2 assay results demonstrated production of H2 O2 from the oxidation of FMNH- :DMSMO, whereas H2 O2 was not detected in the presence of DMSO2 , confirming H2 O2 utilization. The rate constant for methanesulfinate formation determined from rapid quenched flow and the rate constant for flavin oxidation were similar, indicating that H2 O2 rapidly reacts with DMSO2 , with flavin oxidation as the rate-limiting step. This is the first report of the kinetic mechanisms of both components using rapid kinetics and of a method for methanesulfinate detection using LC-MS.

The vertebrate nervous system is composed of a wide range of neurons and complex synaptic connections, raising the intriguing question of how neuronal diversity is generated. The spinal cord provides an excellent model for exploring the mechanisms governing neuronal diversity due to its simple neural network and the conserved molecular processes involved in neuron formation and specification during evolution. This review specifically examines two distinct progenitor domains present in the zebrafish ventral spinal cord: the lateral floor plate (LFP) and the p2 progenitor domain. The LFP is responsible for the production of GABAergic Kolmer-Agduhr neurons (KA″), glutamatergic V3 neurons, and intraspinal serotonergic neurons, while the p2 domain generates V2 precursors that subsequently differentiate into three unique subpopulations of V2 neurons, namely glutamatergic V2a, GABAergic V2b, and glycinergic V2s. Based on recent findings, we will examine the fundamental signaling pathways and transcription factors that play a key role in the specification of these diverse neurons and neuronal subtypes derived from the LFP and p2 progenitor domains.

Transcription factor RUNX1 plays important roles in hematopoiesis and leukemogenesis. RUNX1 function is tightly controlled through posttranslational modifications, including ubiquitination and acetylation. However, its regulation via ubiquitination, especially proteasome-independent ubiquitination, is poorly understood. We previously identified DTX2 as a RUNX1-interacting E3 ligase using a cell-free AlphaScreen assay. In this study, we examined whether DTX2 is involved in the regulation of RUNX1 using in vitro and ex vivo analyses. DTX2 bound to RUNX1 and other RUNX family members RUNX2 and RUNX3 through their C-terminal region. DTX2-induced RUNX1 ubiquitination did not result in RUNX1 protein degradation. Instead, we found that the acetylation of RUNX1, which is known to enhance the transcriptional activity of RUNX1, was inhibited in the presence of DTX2. Concomitantly, DTX2 reduced the RUNX1-induced activation of an MCSFR luciferase reporter. We also found that DTX2 induced RUNX1 cytoplasmic mislocalization. Moreover, DTX2 overexpression showed a substantial growth-inhibitory effect in RUNX1-dependent leukemia cell lines. Thus, our findings indicate a novel aspect of the ubiquitination and acetylation of RUNX1 that is modulated by DTX2 in a proteosome-independent manner.

The human Nod-like receptor protein NOD1 is a well-described pattern-recognition receptor (PRR) with diverse functions. NOD1 associates with F-actin and its protein levels are upregulated in metastatic cancer cells. A hallmark of cancer cells is their ability to migrate, which involves actin remodelling. Using chemotaxis and wound healing assays, we show that NOD1 expression correlated with the migration rate and chemotactic index in the cervical carcinoma cell line HeLa. The effect of NOD1 in cell migration was independent of the downstream kinase RIPK2 and NF-ĸB activity. Additionally, NOD1 negatively regulated the phosphorylation status of cofilin, which inhibits actin turnover. Co-immunoprecipitation assays identified HCLS1-associated protein X-1 (HAX-1) as a previously unknown interaction partner of NOD1. Silencing of HAX-1 expression reduced the migration behaviour to similar levels as NOD1 knockdown, and simultaneous knockdown of NOD1 and HAX-1 showed no additive effect, suggesting that both proteins act in the same pathway. In conclusion, our data revealed an important role of the PRR NOD1 in regulating cell migration as well as chemotaxis in human cervical cancer cells and identified HAX-1 as a protein that interacts with NOD1 and is involved in this signalling pathway.

Cyanobacteriochrome (CBCR) photoreceptors are distantly related to the canonical red/far-red reversible phytochrome photoreceptors. In the case of the CBCRs, only the GAF domain is required for chromophore incorporation and photoconversion. The GAF domains of CBCR are highly diversified into many lineages to sense various colors of light. These CBCR GAF domains are divided into two types: those possessing only the canonical Cys residue and those with both canonical and second Cys residues. The canonical Cys residue stably ligates to the chromophore in both cases. The second Cys residue mostly shows reversible adduct formation with the chromophore during photoconversion for spectral tuning. In this study, we focused on the CBCR GAF domain AnPixJg2_BV4, which possesses only the canonical Cys residue. AnPixJg2_BV4 covalently ligates to the biliverdin (BV) chromophore and shows far-red/orange reversible photoconversion. Because BV is a mammalian intrinsic chromophore, BV-binding molecules are advantageous for in vivo optogenetic and bioimaging tool development. To obtain a better developmental platform molecule, we performed site-saturation random mutagenesis and serendipitously obtained a unique variant molecule that showed far-red/blue reversible photoconversion, in which the Cys residue was introduced near the chromophore. This introduced Cys residue functioned as the second Cys residue that reversibly ligated with the chromophore. Because the position of the introduced Cys residue is distinct from the known second Cys residues, the variant molecule obtained in this study would expand our knowledge about the spectral tuning mechanism of CBCRs and contribute to tool development.

Generating active, pure, and monodisperse protein remains a major bottleneck for structural studies using X-ray crystallography and cryo-electron microscopy (cryo-EM). The current methodology heavily relies on overexpressing the recombinant protein fused with a histidine tag in conventional expression systems and evaluating the quality and stability of purified protein using size exclusion chromatography (SEC). This requires a large amount of protein and can be highly laborious and time consuming. Therefore, this approach is not suitable for high-throughput screening and low-expressing macromolecules, particularly eukaryotic membrane proteins. Using fluorescent proteins fused to the target protein (applicable to both soluble and membrane proteins) enables rapid and efficient screening of expression level and monodispersity of tens of unpurified constructs using fluorescence-based size exclusion chromatography (FSEC). Moreover, FSEC proves valuable for screening multiple detergents to identify the most stabilizing agent in the case of membrane proteins. Additionally, FSEC can facilitate nanodisc reconstitution by determining the optimal ratio of membrane scaffold protein (MSP), lipids, and target protein. The distinct advantages offered by FSEC indicate that fluorescent proteins can serve as a viable alternative to commonly used affinity tags for both characterization and purification purposes. In this review, I will summarize the advantages of this technique using examples from my own work. It should be noted that this article is not intended to provide an exhaustive review of all available literature, but rather to offer representative examples of FSEC applications.

Interleukin-6 (IL-6)-type cytokines not only have key immunomodulatory functions that affect the pathogenesis of diseases such as autoimmune diseases, chronic inflammatory conditions, and cancer, but also fulfill important homeostatic tasks. Even though the pro-inflammatory arm has hindered the development of therapeutics based on natural-like IL-6-type cytokines to date, current synthetic trends might pave the way to overcome these limitations and eventually lead to immune-inert designer cytokines to aid type 2 diabetes and brain injuries. Those synthetic biology approaches include mutations, fusion proteins, and inter-cytokine swapping, and resulted in IL-6-type cytokines with altered receptor affinities, extended target cell profiles, and targeting of non-natural cytokine receptor complexes. Here, we survey synthetic cytokine developments within the IL-6-type cytokine family and discuss potential clinical applications.

Poly(ADP-ribosyl)ation is predominantly catalyzed by Poly(ADP-ribose) polymerase 1 (PARP1) in response to DNA damage, mediating the DNA repair process to maintain genomic integrity. Single-strand (SSB) and double-strand (DSB) DNA breaks are bona fide stimulators of PARP1 activity. However, PAR-mediated PARP1 regulation remains unexplored. Here, we report ZnF3, BRCT, and WGR, hitherto uncharacterized, as PAR reader domains of PARP1. Surprisingly, these domains recognize PARylated protein with a higher affinity compared with PAR but bind with weak or no affinity to DNA breaks as standalone domains. Conversely, ZnF1 and ZnF2 of PARP1 recognize DNA breaks but bind weakly to PAR. In addition, PAR reader domains, together, exhibit a synergy to recognize PAR or PARylated protein. Further competition-binding studies suggest that PAR binding releases DNA from PARP1, and the WGR domain facilitates DNA release. Unexpectedly, PAR showed catalytic stimulation of PARP1 but hampered the DNA-dependent stimulation. Altogether, our work discovers dedicated high-affinity PAR reader domains of PARP1 and uncovers a novel mechanism of allosteric regulation of DNA-dependent and DNA-independent activities of PARP1 by its catalytic product PAR.

Neonatal brachial plexus injury (NBPI), a leading cause of pediatric upper limb paralysis, results in disabling and incurable muscle contractures that are driven by impaired longitudinal growth of denervated muscles. A rare form of NBPI, which maintains both afferent and sympathetic muscle innervation despite motor denervation, protects against contractures. We have previously ruled out a role for NRG/ErbB signaling, the predominant pathway governing antegrade afferent neuromuscular transmission, in modulating the formation of contractures. Our current study therefore investigated the contributions of sympathetic innervation of skeletal muscle in modulating NBPI-induced contractures. Through chemical sympathectomy and pharmacologic modification with a β2 -adrenergic agonist, we discovered that sympathetic innervation alone is neither required nor sufficient to modulate contracture formation in neonatal mice. Despite this, sympathetic innervation plays an intriguing sex-specific role in mediating neonatal muscle growth, as the cross-sectional area (CSA) and volume of normally innervated male muscles were diminished by ablation of sympathetic neurons and increased by β-adrenergic stimulation. Intriguingly, the robust alterations in CSA occurred with minimal changes to normal longitudinal muscle growth as determined by sarcomere length. Instead, β-adrenergic stimulation exacerbated sarcomere overstretch in denervated male muscles, indicating potentially discrete regulation of muscle width and length. Future investigations into the mechanistic underpinnings of these distinct aspects of muscle growth are thus essential for improving clinical outcomes in patients affected by muscle disorders in which both length and width are affected.

The P450 monooxygenase CYP109A2 from Bacillus megaterium DSM319 was previously found to convert vitamin D3 (VD3) to 25-hydroxyvitamin D3. Here, we show that this enzyme is also able to convert testosterone in a highly regio- and stereoselective manner to 16β-hydroxytestosterone. To reveal the structural determinants governing the regio- and stereoselective steroid hydroxylation reactions catalyzed by CYP109A2, two crystal structures of CYP109A2 were solved in similar closed conformations, one revealing a bound testosterone in the active site pocket, albeit at a nonproductive site away from the heme-iron. To examine whether the closed crystal structures nevertheless correspond to a reactive conformation of CYP109A2, docking and molecular dynamics (MD) simulations were performed with testosterone and vitamin D3 (VD3) present in the active site. These MD simulations were analyzed for catalytically productive conformations, the relative occurrences of which were in agreement with the experimentally determined stereoselectivities if the predicted stability of each carbon-hydrogen bond was taken into account. Overall, the first-time determination and analysis of the catalytically relevant 3D conformation of CYP109A2 will allow for future small molecule ligand screening in silico, as well as enabling site-directed mutagenesis toward improved enzymatic properties of this enzyme.

Zinc Finger DNA-binding domain-containing proteins are the most populous family among eukaryotic transcription factors. Among these, members of the BTB domain-containing ZBTB sub-family are mostly known for their transcriptional repressive functions. In this Viewpoint article, we explore molecular mechanisms that potentially diversify the function of ZBTB proteins based on their homo and heterodimerization, alternative splicing and post-translational modifications. We describe how the BTB domain is as much a scaffold for the assembly of co-repressors, as a domain that regulates protein stability. We highlight another mechanism that regulates ZBTB protein stability: phosphorylation in the zinc finger domain. We explore the non-transcriptional, structural roles of ZBTB proteins and highlight novel findings that describe the ability of ZBTB proteins to associate with poly adenosine ribose in the nucleus during the DNA damage response. Herein, we discuss the contribution of BTB domain scaffolds to the formation of transcriptional repressive complexes, to chromosome compartmentalization and their non-transcriptional, purely structural functions in the nucleus.

Glycoside hydrolase family 31 (GH31) contains α-glycoside hydrolases with different substrate specificities involved in various physiological functions. This family has recently been classified into 20 subfamilies using sequence similarity networks. An α-galactosidase from the gut bacterium Bacteroides salyersiae (BsGH31_19, which belongs to GH31 subfamily 19) was reported to have hydrolytic activity against the synthetic substrate p- nitrophenyl α-galactopyranoside, but its natural substrate remained unknown. BsGH31_19 shares low sequence identity (around 20%) with other reported GH31 α-galactosidases, PsGal31A from Pseudopedobacter saltans and human myogenesis-regulating glycosidase (MYORG), and was expected to have distinct specificity. Here, we characterized BsGH31_19 and its ortholog from a soil Bacteroidota bacterium, Flavihumibacter petaseus (FpGH31_19), and demonstrated that they showed high substrate specificity against α-(1→4)-linkages in α-(1→4)-galactobiose and globotriose [α-Gal-(1→4)-β-Gal-(1→4)-Glc], unlike PsGal31A and MYORG. The crystallographic analyses of BsGH31_19 and FpGH31_19 showed that their overall structures resemble those of MYORG and form a dimer with an interface different from that of PsGal31A and MYORG dimers. The structures of FpGH31_19 complexed with d-galactose and α-(1→4)-galactobiose revealed that amino acid residues that recognize a galactose residue at subsite +1 are not conserved between FpGH31_19 and BsGH31_19. The tryptophan (Trp153) that recognizes galactose at subsite -1 is homologous to the tryptophan residues in MYORG and α-galactosidases belonging to GH27, GH36, and GH97, but not in the bacterial GH31 member PsGal31A. Our results provide structural insights into molecular diversity and evolutionary relationships in the GH31 α-galactosidase subfamilies and the other α-galactosidase families.

Fluorine forms the strongest single bond to carbon with the highest bond dissociation energy among natural products. However, fluoroacetate dehalogenases (FADs) have been shown to hydrolyze this bond in fluoroacetate under mild reaction conditions. Furthermore, two recent studies demonstrated that the FAD RPA1163 from Rhodopseudomonas palustris can also accept bulkier substrates. In this study, we explored the substrate promiscuity of microbial FADs and their ability to defluorinate polyfluorinated organic acids. Enzymatic screening of eight purified dehalogenases with reported fluoroacetate defluorination activity revealed significant hydrolytic activity against difluoroacetate in three proteins. Product analysis using liquid chromatography-mass spectrometry identified glyoxylic acid as the final product of enzymatic DFA defluorination. The crystal structures of DAR3835 from Dechloromonas aromatica and NOS0089 from Nostoc sp. were determined in the apo-state along with the DAR3835 H274N glycolyl intermediate. Structure-based site-directed mutagenesis of DAR3835 demonstrated a key role for the catalytic triad and other active site residues in the defluorination of both fluoroacetate and difluoroacetate. Computational analysis of the dimer structures of DAR3835, NOS0089, and RPA1163 indicated the presence of one substrate access tunnel in each protomer. Moreover, protein-ligand docking simulations suggested similar catalytic mechanisms for the defluorination of both fluoroacetate and difluoroacetate, with difluoroacetate being defluorinated via two consecutive defluorination reactions producing glyoxylate as the final product. Thus, our findings provide molecular insights into substrate promiscuity and catalytic mechanism of FADs, which are promising biocatalysts for applications in synthetic chemistry and bioremediation of fluorochemicals.

Damage-associated molecular patterns (DAMPs) are typically derived from the endogenous elements of necrosis cells and can trigger inflammatory responses by activating DAMPs-sensing receptors on immune cells. Failure to clear DAMPs may lead to persistent inflammation, thereby contributing to the pathogenesis of immunological diseases. This review focuses on a newly recognized class of DAMPs derived from lipid, glucose, nucleotide, and amino acid metabolic pathways, which are then termed as metabolite-derived DAMPs. This review summarizes the reported molecular mechanisms of these metabolite-derived DAMPs in exacerbating inflammation responses, which may attribute to the pathology of certain types of immunological diseases. Additionally, this review also highlights both direct and indirect clinical interventions that have been explored to mitigate the pathological effects of these DAMPs. By summarizing our current understanding of metabolite-derived DAMPs, this review aims to inspire future thoughts and endeavors on targeted medicinal interventions and the development of therapies for immunological diseases.

Lysosomal pH setpoint and H+ homeostasis is key to the lysosome's functions. The Parkinson's disease-risk protein TMEM175, originally identified as lysosomal K+ channel, works as a H+ -activated H+ channel and discharges the lysosomal H+ store when it is hyper-acidified. Yang et al. indicate that TMEM175 is permeable for both K+ and H+ in the same pore and charges the lysosome with H+ under certain conditions. The charge and discharge functions are under regulation of the lysosomal matrix and glycocalyx layer. Their presented work indicates that TMEM175 performs as a multi-functional channel regulating lysosomal pH in response to physiological conditions.

Neuronal differentiation and maturation are extended developmental processes. To determine whether neurons at different developmental stages have divergent chemosensitivities, we screened differentiating and maturing neuronal populations using a small compound library comprising FDA-approved and investigational drugs. Using a neurotoxicity assay format, both respective neuronal population-based screening campaigns performed robustly (Z-factors = 0.7-0.8), although the hit rate for the differentiating neurons (2.8%) was slightly higher than for maturing neurons (1.9%). While the majority of hits were toxic to both neuronal populations, these hits predominantly represented promiscuous drugs. Other drugs were selectively neurotoxic, with receptor tyrosine kinase inhibitors disproportionally represented after confirmation. Ponatinib and amuvatinib were neuroinhibitory for differentiating and maturing neurons, respectively. Chemoinformatic analyses confirmed differences in potential drug targets that may be differentially expressed during neuronal development. Subsequent studies demonstrated neuronal expression of AXL, an amuvatinib target, in both neuronal populations. However, functional AXL activity was confirmed only in the maturing neuronal population as determined by AXL phosphorylation in response to GAS6, the cognate ligand of AXL, and concurrent STAT3Y705 phosphorylation. Differentiating neurons were unresponsive to the effects of GAS6 suggesting that the AXL-STAT3 signaling axis was nonfunctional. Amuvatinib treatment of maturing neuronal cultures significantly reduced pAXL levels. These studies indicate that neuronal developmental states may exhibit unique chemosensitivities and that drugs may have different neuro-inhibitory effects depending upon the developmental stage of the neuronal population.

The prostate is an androgen-responsive organ, but the complex cellular and molecular interactions that mediate these responses remain incompletely defined. Here, I synthesise the existing literature to derive a simple conceptual framework describing the androgen-dependent regulation of prostate epithelial dynamics. In this framework, epithelial androgen receptor (AR) cell-autonomously controls luminal cell height, whereas stromal AR regulates the synthesis of growth factors that promote luminal cell survival and proliferation. With the additional aid of a reanalysis of single-cell RNA-seq data, I also propose that insulin-like growth factor 1 (IGF1) functions as a key androgen-dependent growth factor coordinating stromal-to-epithelial paracrine communication. A novel mathematical model based on this framework was able to quantitatively fit experimental data describing prostate regression and regeneration. Model analysis demonstrates how the luminal cell population can maintain a stable equilibrium size via competition for and degradation of stroma-derived IGF1 and how this population size can be controlled by androgen levels, without a requirement for distinct luminal cell subsets. Moreover, model simulations were able to qualitatively recapitulate experimental observations in inflammatory and cancerous states, thereby providing insights into potential disease mechanisms. This simple model could therefore serve as a foundation for more comprehensive modelling of both the healthy and diseased prostate.