Individual transcription factors modulate both the micromovement of chromatin and its long-range structure.

The control of eukaryotic gene expression is intimately connected to highly dynamic chromatin structures. Gene regulation relies on activator and repressor transcription factors (TFs) that induce local chromatin opening and closing. However, it is unclear how nucleus-wide chromatin organization responds dynamically to the activity of specific TFs. Here, we examined how two TFs with opposite effects on local chromatin accessibility modulate chromatin dynamics nucleus-wide. We combine high-resolution diffusion mapping and dense flow reconstruction and correlation in living cells to obtain an imaging-based, nanometer-scale analysis of local diffusion processes and long-range coordinated movements of both chromatin and TFs. We show that the expression of either an individual transcriptional activator (CDX2) or repressor (SIX6) with large numbers of binding sites increases chromatin mobility nucleus-wide, yet they induce opposite coherent chromatin motions at the micron scale. Hi-C analysis of higher-order chromatin structures shows that induction of the pioneer factor CDX2 leads both to changes in local chromatin interactions and the distribution of A and B compartments, thus relating the micromovement of chromatin with changes in compartmental structures. Given that inhibition of transcription initiation and elongation by RNA Pol II has a partial impact on the global chromatin dynamics induced by CDX2, we suggest that CDX2 overexpression alters chromatin structure dynamics both dependently and independently of transcription. Our biophysical analysis shows that sequence-specific TFs can influence chromatin structure on multiple architectural levels, arguing that local chromatin changes brought by TFs alter long-range chromatin mobility and its organization.

Single Live Cell Monitoring of Protein Turnover Reveals Intercellular Variability and Cell-Cycle Dependence of Degradation Rates.

Cells need to reliably control their proteome composition to maintain homeostasis and regulate growth. How protein synthesis and degradation interplay to control protein expression levels remains unclear. Here, we combined a tandem fluorescent timer and pulse-chase protein labeling to disentangle how protein synthesis and degradation control protein homeostasis in single live mouse embryonic stem cells. We discovered substantial cell-cycle dependence in protein synthesis rates and stabilization of a large number of proteins around cytokinesis. Protein degradation rates were highly variable between cells, co-varied within individual cells for different proteins, and were positively correlated with synthesis rates. This suggests variability in proteasome activity as an important source of global extrinsic noise in gene expression. Our approach paves the way toward understanding the complex interplay of synthesis and degradation processes in determining protein levels of individual mammalian cells.

An out-of-equilibrium definition of protein turnover.

Protein turnover (PT) has been formally defined only in equilibrium conditions, which is ill-suited to quantify PT during dynamic processes that occur during embryogenesis or (extra) cellular signaling. In this Hypothesis, we propose a definition of PT in an out-of-equilibrium regime that allows the quantification of PT in virtually any biological context. We propose a simple mathematical and conceptual framework applicable to a broad range of available data, such as RNA sequencing coupled with pulsed-SILAC datasets. We apply our framework to a published dataset and show that stimulation of mouse dendritic cells with LPS leads to a proteome-wide change in PT. This is the first quantification of PT out-of-equilibrium, paving the way for the analysis of biological systems in other contexts.

A role for mitotic bookmarking of SOX2 in pluripotency and differentiation.

Mitotic bookmarking transcription factors remain bound to chromosomes during mitosis and were proposed to regulate phenotypic maintenance of stem and progenitor cells at the mitosis-to-G1 (M-G1) transition. However, mitotic bookmarking remains largely unexplored in most stem cell types, and its functional relevance for cell fate decisions remains unclear. Here we screened for mitotic chromosome binding within the pluripotency network of embryonic stem (ES) cells and show that SOX2 and OCT4 remain bound to mitotic chromatin through their respective DNA-binding domains. Dynamic characterization using photobleaching-based methods and single-molecule imaging revealed quantitatively similar specific DNA interactions, but different nonspecific DNA interactions, of SOX2 and OCT4 with mitotic chromatin. Using ChIP-seq (chromatin immunoprecipitation [ChIP] combined with high-throughput sequencing) to assess the genome-wide distribution of SOX2 on mitotic chromatin, we demonstrate the bookmarking activity of SOX2 on a small set of genes. Finally, we investigated the function of SOX2 mitotic bookmarking in cell fate decisions and show that its absence at the M-G1 transition impairs pluripotency maintenance and abrogates its ability to induce neuroectodermal differentiation but does not affect reprogramming efficiency toward induced pluripotent stem cells. Our study demonstrates the mitotic bookmarking property of SOX2 and reveals its functional importance in pluripotency maintenance and ES cell differentiation.

Single-molecule dynamics and genome-wide transcriptomics reveal that NF-kB (p65)-DNA binding times can be decoupled from transcriptional activation.

Transcription factors (TFs) regulate gene expression in both prokaryotes and eukaryotes by recognizing and binding to specific DNA promoter sequences. In higher eukaryotes, it remains unclear how the duration of TF binding to DNA relates to downstream transcriptional output. Here, we address this question for the transcriptional activator NF-κB (p65), by live-cell single molecule imaging of TF-DNA binding kinetics and genome-wide quantification of p65-mediated transcription. We used mutants of p65, perturbing either the DNA binding domain (DBD) or the protein-protein transactivation domain (TAD). We found that p65-DNA binding time was predominantly determined by its DBD and directly correlated with its transcriptional output as long as the TAD is intact. Surprisingly, mutation or deletion of the TAD did not modify p65-DNA binding stability, suggesting that the p65 TAD generally contributes neither to the assembly of an "enhanceosome," nor to the active removal of p65 from putative specific binding sites. However, TAD removal did reduce p65-mediated transcriptional activation, indicating that protein-protein interactions act to translate the long-lived p65-DNA binding into productive transcription.

Modulation of transcriptional burst frequency by histone acetylation.

Many mammalian genes are transcribed during short bursts of variable frequencies and sizes that substantially contribute to cell-to-cell variability. However, which molecular mechanisms determine bursting properties remains unclear. To probe putative mechanisms, we combined temporal analysis of transcription along the circadian cycle with multiple genomic reporter integrations, using both short-lived luciferase live microscopy and single-molecule RNA-FISH. Using the Bmal1 circadian promoter as our model, we observed that rhythmic transcription resulted predominantly from variations in burst frequency, while the genomic position changed the burst size. Thus, burst frequency and size independently modulated Bmal1 transcription. We then found that promoter histone-acetylation level covaried with burst frequency, being greatest at peak expression and lowest at trough expression, while remaining unaffected by the genomic location. In addition, specific deletions of ROR-responsive elements led to constitutively elevated histone acetylation and burst frequency. We then investigated the suggested link between histone acetylation and burst frequency by dCas9p300-targeted modulation of histone acetylation, revealing that acetylation levels influence burst frequency more than burst size. The correlation between acetylation levels at the promoter and burst frequency was also observed in endogenous circadian genes and in embryonic stem cell fate genes. Thus, our data suggest that histone acetylation-mediated control of transcription burst frequency is a common mechanism to control mammalian gene expression.

Endogenous fluctuations of OCT4 and SOX2 bias pluripotent cell fate decisions.

SOX2 and OCT4 are pioneer transcription factors playing a key role in embryonic stem (ES) cell self-renewal and differentiation. How temporal fluctuations in their expression levels bias lineage commitment is unknown. Here, we generated knock-in reporter fusion ES cell lines allowing to monitor endogenous SOX2 and OCT4 protein fluctuations in living cells and to determine their impact on mesendodermal and neuroectodermal commitment. We found that small differences in SOX2 and OCT4 levels impact cell fate commitment in G1 but not in S phase. Elevated SOX2 levels modestly increased neuroectodermal commitment and decreased mesendodermal commitment upon directed differentiation. In contrast, elevated OCT4 levels strongly biased ES cells towards both neuroectodermal and mesendodermal fates in undirected differentiation. Using ATAC-seq on ES cells gated for different endogenous SOX2 and OCT4 levels, we found that high OCT4 levels increased chromatin accessibility at differentiation-associated enhancers. This suggests that small endogenous fluctuations of pioneer transcription factors can bias cell fate decisions by concentration-dependent priming of differentiation-associated enhancers.

Circadian Dbp transcription relies on highly dynamic BMAL1-CLOCK interaction with E boxes and requires the proteasome.

The transcription factors BMAL1 and CLOCK drive the circadian transcription of clock and clock-controlled genes, such as Dbp. To investigate the kinetics of BMAL1 binding to target genes in real time, we generated a cell line harboring tandem arrays of Dbp repeats and monitored the binding of a fluorescent BMAL1 fusion protein to these arrays by time-lapse microscopy. BMAL1 occupancy at the Dbp locus was highly circadian and strictly dependent on CLOCK. Moreover, BMAL1-CLOCK associations with Dbp were extremely unstable and displayed stochastic, proteasome-dependent fluctuations. Proteasome inhibition prolonged the residence time of BMAL1-CLOCK but resulted in an immediate attenuation of Dbp transcription. In cells harboring a single Dbp-luciferase reporter gene copy, this silencing was shown to be caused by a decrease in both the frequencies and sizes of transcriptional bursts. Thus, BMAL1 and CLOCK may act as Kamikaze activators, in that they are rapidly degraded once bound to Dbp chromatin.

Spatial organization of RNA polymerase II inside a mammalian cell nucleus revealed by reflected light-sheet superresolution microscopy.

Superresolution microscopy based on single-molecule centroid determination has been widely applied to cellular imaging in recent years. However, quantitative imaging of the mammalian nucleus has been challenging due to the lack of 3D optical sectioning methods for normal-sized cells, as well as the inability to accurately count the absolute copy numbers of biomolecules in highly dense structures. Here we report a reflected light-sheet superresolution microscopy method capable of imaging inside the mammalian nucleus with superior signal-to-background ratio as well as molecular counting with single-copy accuracy. Using reflected light-sheet superresolution microscopy, we probed the spatial organization of transcription by RNA polymerase II (RNAP II) molecules and quantified their global extent of clustering inside the mammalian nucleus. Spatiotemporal clustering analysis that leverages on the blinking photophysics of specific organic dyes showed that the majority (>70%) of the transcription foci originate from single RNAP II molecules, and no significant clustering between RNAP II molecules was detected within the length scale of the reported diameter of "transcription factories." Colocalization measurements of RNAP II molecules equally labeled by two spectrally distinct dyes confirmed the primarily unclustered distribution, arguing against a prevalent existence of transcription factories in the mammalian nucleus as previously proposed. The methods developed in our study pave the way for quantitative mapping and stoichiometric characterization of key biomolecular species deep inside mammalian cells.

Single-molecule imaging of transcription factor binding to DNA in live mammalian cells.

Imaging single fluorescent proteins in living mammalian cells is challenged by out-of-focus fluorescence excitation. To reduce out-of-focus fluorescence we developed reflected light-sheet microscopy (RLSM), a fluorescence microscopy method allowing selective plane illumination throughout the nuclei of living mammalian cells. A thin light sheet parallel to the imaging plane and close to the sample surface is generated by reflecting an elliptical laser beam incident from the top by 90° with a small mirror. The thin light sheet allows for an increased signal-to-background ratio superior to that in previous illumination schemes and enables imaging of single fluorescent proteins with up to 100-Hz time resolution. We demonstrated the single-molecule sensitivity of RLSM by measuring the DNA-bound fraction of glucocorticoid receptor (GR) and determining the residence times on DNA of various oligomerization states and mutants of GR and estrogen receptor-α (ER), which permitted us to resolve different modes of DNA binding of GR. We demonstrated two-color single-molecule imaging by observing the spatiotemporal colocalization of two different protein pairs. Our single-molecule measurements and statistical analysis revealed dynamic properties of transcription factors.

Stimulus-induced modulation of transcriptional bursting in a single mammalian gene.

Mammalian genes are often transcribed discontinuously as short bursts of RNA synthesis followed by longer silent periods. However, how these "on" and "off" transitions, together with the burst sizes, are modulated in single cells to increase gene expression upon stimulation is poorly characterized. By combining single-cell time-lapse luminescence imaging with stochastic modeling of the time traces, we quantified the transcriptional responses of the endogenous connective tissue growth factor gene to different physiological stimuli: serum and TGF-ß1. Both stimuli caused a rapid and acute increase in burst sizes. Whereas TGF-ß1 showed prolonged transcriptional activation mediated by an increase of transcription rate, serum stimulation resulted in a large and temporally tight first transcriptional burst, followed by a refractory period in the range of hours. Our study thus reveals how different physiological stimuli can trigger kinetically distinct transcriptional responses of the same gene.

Simultaneous automated ascertainment of prevalent vertebral fracture and abdominal aortic calcification in clinical practice: role in fracture risk assessment.

Whether simultaneous automated ascertainments of prevalent vertebral fracture (auto-PVFx) and abdominal aortic calcification (auto-AAC) on vertebral fracture assessment (VFA) lateral spine bone density (BMD) images jointly predict incident fractures in routine clinical practice is unclear. We estimated the independent associations of auto-PVFx and auto-AAC primarily with incident major osteoporotic and secondarily with incident hip and any clinical fractures in 11 013 individuals (mean [SD] age 75.8 [6.8] years, 93.3% female) who had a BMD test combined with VFA between March 2010 and December 2017. Auto-PVFx and auto-AAC were ascertained using convolutional neural networks (CNNs). Proportional hazards models were used to estimate the associations of auto-PVFx and auto-AAC with incident fractures over a mean (SD) follow-up of 3.7 (2.2) years, adjusted for each other and other risk factors. At baseline, 17% (n = 1881) had auto-PVFx and 27% (n = 2974) had a high level of auto-AAC (≥ 6 on scale of 0 to 24). Multivariable-adjusted hazard ratios (HR) for incident major osteoporotic fracture (95% C.I.) were 1.85 (1.59, 2.15) for those with compared to those without auto-PVFx, and 1.36 (1.14, 1.62) for those with high compared to low auto-AAC. The multivariable-adjusted HRs for incident hip fracture were 1.62 (95% C.I. 1.26 to 2.07) for those with compared to those without auto-PVFx, and 1.55 (95% C.I. 1.15 to 2.09) for those high auto-AAC compared to low auto-AAC. The 5-year cumulative incidence of major osteoporotic fracture was 7.1% in those with no auto-PVFx and low auto-AAC, 10.1% in those with no auto-PVFx and high auto-AAC, 13.4% in those with auto-PVFx and low auto-AAC, and 18.0% in those with auto-PVFx and high auto-AAC. While physician manual review of images in clinical practice will still be needed to confirm image quality and provide clinical context for interpretation, simultaneous automated ascertainment of auto-PVFx and auto-AAC can aid fracture risk assessment.

Individuals with calcification of their abdominal aorta (AAC) and vertebral fractures seen on lateral spine bone density images (easily obtained as part of a bone density test) are much more likely to have subsequent fractures. Prior studies have not shown if both AAC and prior vertebral fracture both contribute to fracture prediction in routine clinical practice. Additionally, a barrier to using these images to aid fracture risk assessment at the time of bone density testing has been the need for expert readers to be able to accurately detect both AAC and vertebral fractures. We have developed automated computer methods (using artificial intelligence) to accurately detect vertebral fracture (auto-PVFx) and auto-AAC on lateral spine bone density images for 11 013 older individuals having a bone density test in routine clinical practice. Over a 5-year follow-up period, 7.1% of those with no auto-PVFx and low auto-AAC, 10.1% of those with no auto-PVFx and high auto-AAC, 13.4% of those with auto-PVFx and low auto-AAC, and 18.0% of those with auto-PVFx and high auto-AAC had a major osteoporotic fracture. Auto-PVFx and auto-AAC, ascertained simultaneously on lateral spine bone density images, both contribute to the risk of subsequent major osteoporotic fractures in routine clinical practice settings.

Gene expression flux analysis reveals specific regulatory modalities of gene expression.

The level of a given protein is determined by the synthesis and degradation rates of its mRNA and protein. While several studies have quantified the contribution of different gene expression steps in regulating protein levels, these are limited by using equilibrium approximations in out-of-equilibrium biological systems. Here, we introduce gene expression flux analysis to quantitatively dissect the dynamics of the expression level for specific proteins and use it to analyze published transcriptomics and proteomics datasets. Our analysis reveals distinct regulatory modalities shared by sets of genes with clear functional signatures. We also find that protein degradation plays a stronger role than expected in the adaptation of protein levels. These findings suggest that shared regulatory strategies can lead to versatile responses at the protein level and highlight the importance of going beyond equilibrium approximations to dissect the quantitative contribution of different steps of gene expression to protein dynamics.

Unsupervised Learning for Maximum Consensus Robust Fitting: A Reinforcement Learning Approach.

Robust model fitting is a core algorithm in several computer vision applications. Despite being studied for decades, solving this problem efficiently for datasets that are heavily contaminated by outliers is still challenging: due to the underlying computational complexity. A recent focus has been on learning-based algorithms. However, most of these approaches are supervised (which require a large amount of labelled training data). In this paper, we introduce a novel unsupervised learning framework: that learns to directly (without labelled data) solve robust model fitting. Moreover, unlike other learning-based methods, our work is agnostic to the underlying input features, and can be easily generalized to a wide variety of LP-type problems with quasi-convex residuals. We empirically show that our method outperforms existing (un)supervised learning approaches, and also achieves competitive results compared to traditional (non-learning-based) methods. Our approach is designed to try to maximise consensus (MaxCon), similar to the popular RANSAC. The basis of our approach, is to adopt a Reinforcement Learning framework. This requires designing appropriate reward functions, and state encodings. We provide a family of reward functions, tunable by choice of a parameter. We also investigate the application of different basic and enhanced Q-learning components.

An Information-Theoretic Method to Automatic Shortcut Avoidance and Domain Generalization for Dense Prediction Tasks.

Deep convolutional neural networks for dense prediction tasks are commonly optimized using synthetic data, as generating pixel-wise annotations for real-world data is laborious. However, the synthetically trained models do not generalize well to real-world environments. This poor "synthetic to real" (S2R) generalization we address through the lens of shortcut learning. We demonstrate that the learning of feature representations in deep convolutional networks is heavily influenced by synthetic data artifacts (shortcut attributes). To mitigate this issue, we propose an Information-Theoretic Shortcut Avoidance (ITSA) approach to automatically restrict shortcut-related information from being encoded into the feature representations. Specifically, our proposed method minimizes the sensitivity of latent features to input variations: to regularize the learning of robust and shortcut-invariant features in synthetically trained models. To avoid the prohibitive computational cost of direct input sensitivity optimization, we propose a practical yet feasible algorithm to achieve robustness. Our results show that the proposed method can effectively improve S2R generalization in multiple distinct dense prediction tasks, such as stereo matching, optical flow, and semantic segmentation. Importantly, the proposed method enhances the robustness of the synthetically trained networks and outperforms their fine-tuned counterparts (on real data) for challenging out-of-domain applications.

Machine-Learning Assessed Abdominal Aortic Calcification is Associated with Long-Term Fall and Fracture Risk in Community-Dwelling Older Australian Women.

Abdominal aortic calcification (AAC), a recognized measure of advanced vascular disease, is associated with higher cardiovascular risk and poorer long-term prognosis. AAC can be assessed on dual-energy X-ray absorptiometry (DXA)-derived lateral spine images used for vertebral fracture assessment at the time of bone density screening using a validated 24-point scoring method (AAC-24). Previous studies have identified robust associations between AAC-24 score, incident falls, and fractures. However, a major limitation of manual AAC assessment is that it requires a trained expert. Hence, we have developed an automated machine-learning algorithm for assessing AAC-24 scores (ML-AAC24). In this prospective study, we evaluated the association between ML-AAC24 and long-term incident falls and fractures in 1023 community-dwelling older women (mean age, 75 ± 3 years) from the Perth Longitudinal Study of Ageing Women. Over 10 years of follow-up, 253 (24.7%) women experienced a clinical fracture identified via self-report every 4-6 months and verified by X-ray, and 169 (16.5%) women had a fracture hospitalization identified from linked hospital discharge data. Over 14.5 years, 393 (38.4%) women experienced an injurious fall requiring hospitalization identified from linked hospital discharge data. After adjusting for baseline fracture risk, women with moderate to extensive AAC (ML-AAC24 ≥ 2) had a greater risk of clinical fractures (hazard ratio [HR] 1.42; 95% confidence interval [CI], 1.10-1.85) and fall-related hospitalization (HR 1.35; 95% CI, 1.09-1.66), compared to those with low AAC (ML-AAC24 ≤ 1). Similar to manually assessed AAC-24, ML-AAC24 was not associated with fracture hospitalizations. The relative hazard estimates obtained using machine learning were similar to those using manually assessed AAC-24 scores. In conclusion, this novel automated method for assessing AAC, that can be easily and seamlessly captured at the time of bone density testing, has robust associations with long-term incident clinical fractures and injurious falls. However, the performance of the ML-AAC24 algorithm needs to be verified in independent cohorts. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Machine learning for abdominal aortic calcification assessment from bone density machine-derived lateral spine images.

BACKGROUND: Lateral spine images for vertebral fracture assessment can be easily obtained on modern bone density machines. Abdominal aortic calcification (AAC) can be scored on these images by trained imaging specialists to assess cardiovascular disease risk. However, this process is laborious and requires careful training. METHODS: Training and testing of model performance of the convolutional neural network (CNN) algorithm for automated AAC-24 scoring utilised 5012 lateral spine images (2 manufacturers, 4 models of bone density machines), with trained imaging specialist AAC scores. Validation occurred in a registry-based cohort study of 8565 older men and women with images captured as part of routine clinical practice for fracture risk assessment. Cox proportional hazards models were used to estimate the association between machine-learning AAC (ML-AAC-24) scores with future incident Major Adverse Cardiovascular Events (MACE) that including death, hospitalised acute myocardial infarction or ischemic cerebrovascular disease ascertained from linked healthcare data. FINDINGS: The average intraclass correlation coefficient between imaging specialist and ML-AAC-24 scores for 5012 images was 0.84 (95% CI 0.83, 0.84) with classification accuracy of 80% for established AAC groups. During a mean follow-up 4 years in the registry-based cohort, MACE outcomes were reported in 1177 people (13.7%). With increasing ML-AAC-24 scores there was an increasing proportion of people with MACE (low 7.9%, moderate 14.5%, high 21.2%), as well as individual MACE components (all p-trend <0.001). After multivariable adjustment, moderate and high ML-AAC-24 groups remained significantly associated with MACE (HR 1.54, 95% CI 1.31-1.80 & HR 2.06, 95% CI 1.75-2.42, respectively), compared to those with low ML-AAC-24. INTERPRETATION: The ML-AAC-24 scores had substantial levels of agreement with trained imaging specialists, and was associated with a substantial gradient of risk for cardiovascular events in a real-world setting. This approach could be readily implemented into these clinical settings to improve identification of people at high CVD risk. FUNDING: The study was supported by a National Health and Medical Research Council of Australia Ideas grant and the Rady Innovation Fund, Rady Faculty of Health Sciences, University of Manitoba.

Down-regulation of phosphatase and tensin homolog by hepatitis C virus core 3a in hepatocytes triggers the formation of large lipid droplets.

UNLABELLED: Hepatitis C virus (HCV) perturbs the host's lipid metabolism and often results in hepatic steatosis. In nonalcoholic fatty liver disease, the intrahepatic down-regulation of phosphatase and tensin homolog deleted on chromosome 10 (PTEN) is a critical mechanism leading to steatosis and its progression toward fibrosis and hepatocellular carcinoma. However, whether an HCV infection triggers the formation of large lipid droplets through PTEN-dependent mechanisms is unknown. We assessed PTEN expression in the livers of patients infected with HCV genotype 1 or 3 with or without steatosis. The role of PTEN in the HCV-induced biogenesis of lipid droplets was further investigated in vitro with hepatoma cells transduced with the HCV core protein of genotype 1b or 3a. Our data indicate that PTEN expression was down-regulated at the posttranscriptional level in steatotic patients infected with genotype 3a. Similarly, the in vitro expression of the HCV genotype 3a core protein (but not 1b), typically leading to the appearance of large lipid droplets, down-regulated PTEN expression by a mechanism involving a microRNA-dependent blockade of PTEN messenger RNA translation. PTEN down-regulation promoted in turn a reduction of insulin receptor substrate 1 (IRS1) expression. Interestingly, either PTEN or IRS1 overexpression prevented the development of large lipid droplets, and this indicates that the down-regulation of both PTEN and IRS1 is required to affect the biogenesis of lipid droplets. However, IRS1 knockdown per se did not alter the morphology of lipid droplets, and this suggests that other PTEN-dependent mechanisms are involved in this process. CONCLUSION: The down-regulation of PTEN and IRS1 is a critical event leading to the HCV genotype 3a-induced formation of large lipid droplets in hepatocytes.

Isoform- and dose-sensitive feedback interactions between paired box 6 gene and delta-catenin in cell differentiation and death.

Pax6, a mammalian homolog of the Drosophila paired box gene family member expressed in stem and progenitor cells, resides at the top of the genetic hierarchy in controlling cell fates and morphogenesis. While Pax6 activation can lead to mitotic arrest, premature neurogenesis, and apoptosis, the underlying molecular mechanisms have not been resolved. Here we report that either Pax6(+5a) or Pax6(-5a) was sufficient to promote, whereas their knockdown reduced the expression of delta-catenin (CTNND2), a neural specific member of the armadillo/beta-catenin superfamily. Pax6(+5a) elicited stronger effects on delta-catenin than Pax6(-5a). Inducible Pax6(+5a) expression demonstrated a biphasic and dose-dependent regulation of delta-catenin expression and cell fates. A moderate upregulation of Pax6(+5a) promoted delta-catenin expression and induced neurite-like cellular protrusions, but increasing expression of Pax6(+5a) reversed these processes. Furthermore, sustained high expression of Pax6(+5a) triggered apoptosis as determined by the reduction of phospho-Bad, Bcl-2, survivin and procaspases, as well as the increases in Bax and cleaved poly(ADP-ribose) polymerase. Importantly, re-introducing delta-catenin by ectopic expression elicited a feedback suppression on Pax6(+5a) expression and reduced Pax6(+5a) induced apoptosis. Therefore, delta-catenin expression is not only controlled by Pax6, but it also provides a feedback suppression mechanism for their functional interactions with important implications in cellular morphogenesis, apoptosis, and cancer.