Machine learning models for predicting blood pressure phenotypes by combining multiple polygenic risk scores.

We construct non-linear machine learning (ML) prediction models for systolic and diastolic blood pressure (SBP, DBP) using demographic and clinical variables and polygenic risk scores (PRSs). We developed a two-model ensemble, consisting of a baseline model, where prediction is based on demographic and clinical variables only, and a genetic model, where we also include PRSs. We evaluate the use of a linear versus a non-linear model at both the baseline and the genetic model levels and assess the improvement in performance when incorporating multiple PRSs. We report the ensemble model's performance as percentage variance explained (PVE) on a held-out test dataset. A non-linear baseline model improved the PVEs from 28.1 to 30.1% (SBP) and 14.3% to 17.4% (DBP) compared with a linear baseline model. Including seven PRSs in the genetic model computed based on the largest available GWAS of SBP/DBP improved the genetic model PVE from 4.8 to 5.1% (SBP) and 4.7 to 5% (DBP) compared to using a single PRS. Adding additional 14 PRSs computed based on two independent GWASs further increased the genetic model PVE to 6.3% (SBP) and 5.7% (DBP). PVE differed across self-reported race/ethnicity groups, with primarily all non-White groups benefitting from the inclusion of additional PRSs. In summary, non-linear ML models improves BP prediction in models incorporating diverse populations.

A parametric bootstrap approach for computing confidence intervals for genetic correlations with application to genetically determined protein-protein networks.

Genetic correlation refers to the correlation between genetic determinants of a pair of traits. When using individual-level data, it is typically estimated based on a bivariate model specification where the correlation between the two variables is identifiable and can be estimated from a covariance model that incorporates the genetic relationship between individuals, e.g., using a pre-specified kinship matrix. Inference relying on asymptotic normality of the genetic correlation parameter estimates may be inaccurate when the sample size is low, when the genetic correlation is close to the boundary of the parameter space, and when the heritability of at least one of the traits is low. We address this problem by developing a parametric bootstrap procedure to construct confidence intervals for genetic correlation estimates. The procedure simulates paired traits under a range of heritability and genetic correlation parameters, and it uses the population structure encapsulated by the kinship matrix. Heritabilities and genetic correlations are estimated using the close-form, method of moment, Haseman-Elston regression estimators. The proposed parametric bootstrap procedure is especially useful when genetic correlations are computed on pairs of thousands of traits measured on the same exact set of individuals. We demonstrate the parametric bootstrap approach on a proteomics dataset from the Jackson Heart Study.

Organ transformation by environmental disruption of protein integrity and epigenetic memory in Drosophila.

Despite significant progress in understanding epigenetic reprogramming of cells, the mechanistic basis of "organ reprogramming" by (epi-)gene-environment interactions remained largely obscure. Here, we use the ether-induced haltere-to-wing transformations in Drosophila as a model for epigenetic "reprogramming" at the whole organism level. Our findings support a mechanistic chain of events explaining why and how brief embryonic exposure to ether leads to haltere-to-wing transformations manifested at the larval stage and on. We show that ether interferes with protein integrity in the egg, leading to altered deployment of Hsp90 and widespread repression of Trithorax-mediated establishment of active H3K4me3 chromatin marks throughout the genome. Despite this global reduction, Ubx targets and wing development genes preferentially retain higher levels of H3K4me3 that predispose these genes for later up-regulation in the larval haltere disc, hence the wing-like outcome. Consistent with compromised protein integrity during the exposure, the penetrance of bithorax transformations increases by genetic or chemical reduction of Hsp90 function. Moreover, joint reduction in Hsp90 and trx gene dosage can cause bithorax transformations without exposure to ether, supporting an underlying epistasis between Hsp90 and trx loss-of-functions. These findings implicate environmental disruption of protein integrity at the onset of histone methylation with altered epigenetic regulation of developmental patterning genes. The emerging picture provides a unique example wherein the alleviation of the Hsp90 "capacitor function" by the environment drives a morphogenetic shift towards an ancestral-like body plan. The morphogenetic impact of chaperone response during a major setup of epigenetic patterns may be a general scheme for organ transformation by environmental cues.

A parametric bootstrap approach for computing confidence intervals for genetic correlations with application to genetically-determined protein-protein networks.

Genetic correlation refers to the correlation between genetic determinants of a pair of traits. When using individual-level data, it is typically estimated based on a bivariate model specification where the correlation between the two variables is identifiable and can be estimated from a covariance model that incorporates the genetic relationship between individuals, e.g., using a pre-specified kinship matrix. Inference relying on asymptotic normality of the genetic correlation parameter estimates may be inaccurate when the sample size is low, when the genetic correlation is close to the boundary of the parameter space, and when the heritability of at least one of the traits is low. We address this problem by developing a parametric bootstrap procedure to construct confidence intervals for genetic correlation estimates. The procedure simulates paired traits under a range of heritability and genetic correlation parameters, and it uses the population structure encapsulated by the kinship matrix. Heritabilities and genetic correlations are estimated using the close-form, method of moment, Haseman-Elston regression estimators. The proposed parametric bootstrap procedure is especially useful when genetic correlations are computed on pairs of thousands of traits measured on the same exact set of individuals. We demonstrate the parametric bootstrap approach on a proteomics dataset from the Jackson Heart Study.

Anaerobic pathogens associated with OSA may contribute to pathophysiology via amino-acid depletion.

BACKGROUND: The human microbiome is linked to multiple metabolic disorders such as obesity and diabetes. Obstructive sleep apnoea (OSA) is a common sleep disorder with several metabolic risk factors. We investigated the associations between the gut microbiome composition and function, and measures of OSA severity in participants from a prospective community-based cohort study: the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). METHODS: Bacterial-Wide Association Analysis (BWAS) of gut microbiome measured via metagenomics with OSA measures was performed adjusting for clinical, lifestyle and co-morbidities. This was followed by functional analysis of the OSA-enriched bacteria. We utilized additional metabolomic and transcriptomic associations to suggest possible mechanisms explaining the microbiome effects on OSA. FINDINGS: Several uncommon anaerobic human pathogens were associated with OSA severity. These belong to the Lachnospira, Actinomyces, Kingella and Eubacterium genera. Functional analysis revealed enrichment in 49 processes including many anaerobic-related ones. Severe OSA was associated with the depletion of the amino acids glycine and glutamine in the blood, yet neither diet nor gene expression revealed any changes in the production or consumption of these amino acids. INTERPRETATION: We show anaerobic bacterial communities to be a novel component of OSA pathophysiology. These are established in the oxygen-poor environments characteristic of OSA. We hypothesize that these bacteria deplete certain amino acids required for normal human homeostasis and muscle tone, contributing to OSA phenotypes. Future work should test this hypothesis as well as consider diagnostics via anaerobic bacteria detection and possible interventions via antibiotics and amino-acid supplementation. FUNDING: Described in methods.

Genetic associations between sleep traits and cognitive ageing outcomes in the Hispanic Community Health Study/Study of Latinos.

BACKGROUND: Sleep phenotypes have been reported to be associated with cognitive ageing outcomes. However, there is limited research using genetic variants as proxies for sleep traits to study their associations. We estimated associations between Polygenic Risk Scores (PRSs) for sleep duration, insomnia, daytime sleepiness, and obstructive sleep apnoea (OSA) and measures of cogntive ageing in Hispanic/Latino adults. METHODS: We used summary statistics from published genome-wide association studies to construct PRSs representing the genetic basis of each sleep trait, then we studied the association of the PRSs of the sleep phenotypes with cognitive outcomes in the Hispanic Community Healthy Study/Study of Latinos. The primary model adjusted for age, sex, study centre, and measures of genetic ancestry. Associations are highlighted if their p-value <0.05. FINDINGS: Higher PRS for insomnia was associated with lower global cognitive function and higher risk of mild cognitive impairment (MCI) (OR = 1.20, 95% CI [1.06, 1.36]). Higher PRS for daytime sleepiness was also associated with increased MCI risk (OR = 1.14, 95% CI [1.02, 1.28]). Sleep duration PRS was associated with reduced MCI risk among short and normal sleepers, while among long sleepers it was associated with reduced global cognitive function and with increased MCI risk (OR = 1.40, 95% CI [1.10, 1.78]). Furthermore, adjustment of analyses for the measured sleep phenotypes and APOE-ε4 allele had minor effects on the PRS associations with the cognitive outcomes. INTERPRETATION: Genetic measures underlying insomnia, daytime sleepiness, and sleep duration are associated with MCI risk. Genetic and self-reported sleep duration interact in their effect on MCI. FUNDING: Described in Acknowledgments.

Machine learning models for blood pressure phenotypes combining multiple polygenic risk scores.

We construct non-linear machine learning (ML) prediction models for systolic and diastolic blood pressure (SBP, DBP) using demographic and clinical variables and polygenic risk scores (PRSs). We developed a two-model ensemble, consisting of a baseline model, where prediction is based on demographic and clinical variables only, and a genetic model, where we also include PRSs. We evaluate the use of a linear versus a non-linear model at both the baseline and the genetic model levels and assess the improvement in performance when incorporating multiple PRSs. We report the ensemble model's performance as percentage variance explained (PVE) on a held-out test dataset. A non-linear baseline model improved the PVEs from 28.1% to 30.1% (SBP) and 14.3% to 17.4% (DBP) compared with a linear baseline model. Including seven PRSs in the genetic model computed based on the largest available GWAS of SBP/DBP improved the genetic model PVE from 4.8% to 5.1% (SBP) and 4.7% to 5% (DBP) compared to using a single PRS. Adding additional 14 PRSs computed based on two independent GWASs further increased the genetic model PVE to 6.3% (SBP) and 5.7% (DBP). PVE differed across self-reported race/ethnicity groups, with primarily all non-White groups benefitting from the inclusion of additional PRSs.

Correlations between complex human phenotypes vary by genetic background, gender, and environment.

We develop a closed-form Haseman-Elston estimator for genetic and environmental correlation coefficients between complex phenotypes, which we term HEc, that is as precise as GCTA yet ∼20× faster. We estimate genetic and environmental correlations between over 7,000 phenotype pairs in subgroups from the Trans-Omics in Precision Medicine (TOPMed) program. We demonstrate substantial differences in both heritabilities and genetic correlations for multiple phenotypes and phenotype pairs between individuals of self-reported Black, Hispanic/Latino, and White backgrounds. We similarly observe differences in many of the genetic and environmental correlations between genders. To estimate the contribution of genetics to the observed phenotypic correlation, we introduce "fractional genetic correlation" as the fraction of phenotypic correlation explained by genetics. Finally, we quantify the enrichment of correlations between phenotypic domains, each of which is comprised of multiple phenotypes. Altogether, we demonstrate that the observed correlations between complex human phenotypes depend on the genetic background of the individuals, their gender, and their environment.

Genetic determinants of cardiometabolic and pulmonary phenotypes and obstructive sleep apnoea in HCHS/SOL.

BACKGROUND: Obstructive Sleep Apnoea (OSA) often co-occurs with cardiometabolic and pulmonary diseases. This study is to apply genetic analysis methods to explain the associations between OSA and related phenotypes. METHODS: In the Hispanic Community Healthy Study/Study of Latinos, we estimated genetic correlations ρg between the respiratory event index (REI) and 54 anthropometric, glycemic, cardiometabolic, and pulmonary phenotypes. We used summary statistics from published genome-wide association studies to construct Polygenic Risk Scores (PRSs) representing the genetic basis of each correlated phenotype (ρg>0.2 and p-value<0.05), and of OSA. We studied the association of the PRSs of the correlated phenotypes with both REI and OSA (REI≥5), and the association of OSA PRS with the correlated phenotypes. Causal relationships were tested using Mendelian Randomization (MR) analysis. FINDINGS: The dataset included 11,155 participants, 31.03% with OSA. 22 phenotypes were genetically correlated with REI. 10 PRSs covering obesity and fat distribution (BMI, WHR, WHRadjBMI), blood pressure (DBP, PP, MAP), glycaemic control (fasting insulin, HbA1c, HOMA-B) and insomnia were associated with REI and/or OSA. OSA PRS was associated with BMI, WHR, DBP and glycaemic traits (fasting insulin, HbA1c, HOMA-B and HOMA-IR). MR analysis identified robust causal effects of BMI and WHR on OSA, and probable causal effects of DBP, PP, and HbA1c on OSA/REI. INTERPRETATION: There are shared genetic underpinnings of anthropometric, blood pressure, and glycaemic phenotypes with OSA, with evidence for causal relationships between some phenotypes. FUNDING: Described in Acknowledgments.

Non-linear machine learning models incorporating SNPs and PRS improve polygenic prediction in diverse human populations.

Polygenic risk scores (PRS) are commonly used to quantify the inherited susceptibility for a trait, yet they fail to account for non-linear and interaction effects between single nucleotide polymorphisms (SNPs). We address this via a machine learning approach, validated in nine complex phenotypes in a multi-ancestry population. We use an ensemble method of SNP selection followed by gradient boosted trees (XGBoost) to allow for non-linearities and interaction effects. We compare our results to the standard, linear PRS model developed using PRSice, LDpred2, and lassosum2. Combining a PRS as a feature in an XGBoost model results in a relative increase in the percentage variance explained compared to the standard linear PRS model by 22% for height, 27% for HDL cholesterol, 43% for body mass index, 50% for sleep duration, 58% for systolic blood pressure, 64% for total cholesterol, 66% for triglycerides, 77% for LDL cholesterol, and 100% for diastolic blood pressure. Multi-ancestry trained models perform similarly to specific racial/ethnic group trained models and are consistently superior to the standard linear PRS models. This work demonstrates an effective method to account for non-linearities and interaction effects in genetics-based prediction models.

Machine and Deep Learning in Molecular and Genetic Aspects of Sleep Research.

Epidemiological sleep research strives to identify the interactions and causal mechanisms by which sleep affects human health, and to design intervention strategies for improving sleep throughout the lifespan. These goals can be advanced by further focusing on the environmental and genetic etiology of sleep disorders, and by development of risk stratification algorithms, to identify people who are at risk or are affected by, sleep disorders. These studies rely on comprehensive sleep-related data which often contains complex multi-dimensional physiological and molecular measurements across multiple timepoints. Thus, sleep research is well-suited for the application of computational approaches that can handle high-dimensional data. Here, we survey recent advances in machine and deep learning together with the availability of large human cohort studies with sleep data that can jointly drive the next breakthroughs in the sleep-research field. We describe sleep-related data types and datasets, and present some of the tasks in the field that can be targets for algorithmic approaches, as well as the challenges and opportunities in pursuing them.

Systemic Regulation of Host Energy and Oogenesis by Microbiome-Derived Mitochondrial Coenzymes.

Gut microbiota have been shown to promote oogenesis and fecundity, but the mechanistic basis of remote influence on oogenesis remained unknown. Here, we report a systemic mechanism of influence mediated by bacterial-derived supply of mitochondrial coenzymes. Removal of microbiota decreased mitochondrial activity and ATP levels in the whole-body and ovary, resulting in repressed oogenesis. Similar repression was caused by RNA-based knockdown of mitochondrial function in ovarian follicle cells. Reduced mitochondrial function in germ-free (GF) females was reversed by bacterial recolonization or supplementation of riboflavin, a precursor of FAD and FMN. Metabolomics analysis of GF females revealed a decrease in oxidative phosphorylation and FAD levels and an increase in metabolites that are degraded by FAD-dependent enzymes (e.g., amino and fatty acids). Riboflavin supplementation opposed this effect, elevating mitochondrial function, ATP, and oogenesis. These findings uncover a bacterial-mitochondrial axis of influence, linking gut bacteria with systemic regulation of host energy and reproduction.

Combining 16S rRNA gene variable regions enables high-resolution microbial community profiling.

BACKGROUND: Most of our knowledge about the remarkable microbial diversity on Earth comes from sequencing the 16S rRNA gene. The use of next-generation sequencing methods has increased sample number and sequencing depth, but the read length of the most widely used sequencing platforms today is quite short, requiring the researcher to choose a subset of the gene to sequence (typically 16-33% of the total length). Thus, many bacteria may share the same amplified region, and the resolution of profiling is inherently limited. Platforms that offer ultra-long read lengths, whole genome shotgun sequencing approaches, and computational frameworks formerly suggested by us and by others all allow different ways to circumvent this problem yet suffer various shortcomings. There is a need for a simple and low-cost 16S rRNA gene-based profiling approach that harnesses the short read length to provide a much larger coverage of the gene to allow for high resolution, even in harsh conditions of low bacterial biomass and fragmented DNA. RESULTS: This manuscript suggests Short MUltiple Regions Framework (SMURF), a method to combine sequencing results from different PCR-amplified regions to provide one coherent profiling. The de facto amplicon length is the total length of all amplified regions, thus providing much higher resolution compared to current techniques. Computationally, the method solves a convex optimization problem that allows extremely fast reconstruction and requires only moderate memory. We demonstrate the increase in resolution by in silico simulations and by profiling two mock mixtures and real-world biological samples. Reanalyzing a mock mixture from the Human Microbiome Project achieved about twofold improvement in resolution when combing two independent regions. Using a custom set of six primer pairs spanning about 1200 bp (80%) of the 16S rRNA gene, we were able to achieve ~ 100-fold improvement in resolution compared to a single region, over a mock mixture of common human gut bacterial isolates. Finally, the profiling of a Drosophila melanogaster microbiome using the set of six primer pairs provided a ~ 100-fold increase in resolution and thus enabling efficient downstream analysis. CONCLUSIONS: SMURF enables the identification of near full-length 16S rRNA gene sequences in microbial communities, having resolution superior compared to current techniques. It may be applied to standard sample preparation protocols with very little modifications. SMURF also paves the way to high-resolution profiling of low-biomass and fragmented DNA, e.g., in the case of formalin-fixed and paraffin-embedded samples, fossil-derived DNA, or DNA exposed to other degrading conditions. The approach is not restricted to combining amplicons of the 16S rRNA gene and may be applied to any set of amplicons, e.g., in multilocus sequence typing (MLST).

Microbiome-Germline Interactions and Their Transgenerational Implications.

It is becoming increasingly clear that most, if not all, animals and plants are associated with a diverse array of resident gut microbiota. This symbiosis is regulated by host-microbiome interactions which influence the development, homeostasis, adaptation and evolution of the host. Recent evidence indicated that these interactions can also affect the host germline and have a potential of supporting transgenerational effects, including inheritance of acquired characteristics. Taken together, the influence of gut bacteria on the host soma and germline could potentially give rise to emergent phenotypes, which may be partially inherited by three distinguishable modes of transgenerational influence of gut bacteria: 1) "soma-to-soma" 2) "soma-to-germline" and 3) "soma-germline-soma". Here, we discuss these possibilities in light of evidence supporting bacterial-mediated modes of transgenerational inheritance.

Impact of gut microbiota on the fly's germ line.

Unlike vertically transmitted endosymbionts, which have broad effects on their host's germ line, the extracellular gut microbiota is transmitted horizontally and is not known to influence the germ line. Here we provide evidence supporting the influence of these gut bacteria on the germ line of Drosophila melanogaster. Removal of the gut bacteria represses oogenesis, expedites maternal-to-zygotic-transition in the offspring and unmasks hidden phenotypic variation in mutants. We further show that the main impact on oogenesis is linked to the lack of gut Acetobacter species, and we identify the Drosophila Aldehyde dehydrogenase (Aldh) gene as an apparent mediator of repressed oogenesis in Acetobacter-depleted flies. The finding of interactions between the gut microbiota and the germ line has implications for reproduction, developmental robustness and adaptation.

A principle of organization which facilitates broad Lamarckian-like adaptations by improvisation.

BACKGROUND: During the lifetime of an organism, every individual encounters many combinations of diverse changes in the somatic genome, epigenome and microbiome. This gives rise to many novel combinations of internal failures which are unique to each individual. How any individual can tolerate this high load of new, individual-specific scenarios of failure is not clear. While stress-induced plasticity and hidden variation have been proposed as potential mechanisms of tolerance, the main conceptual problem remains unaddressed, namely: how largely non-beneficial random variation can be rapidly and safely organized into net benefits to every individual. PRESENTATION OF THE HYPOTHESIS: We propose an organizational principle which explains how every individual can alleviate a high load of novel stressful scenarios using many random variations in flexible and inherently less harmful traits. Random changes which happen to reduce stress, benefit the organism and decrease the drive for additional changes. This adaptation (termed 'Adaptive Improvisation') can be further enhanced, propagated, stabilized and memorized when beneficial changes reinforce themselves by auto-regulatory mechanisms. This principle implicates stress not only in driving diverse variations in cells tissues and organs, but also in organizing these variations into adaptive outcomes. Specific (but not exclusive) examples include stress reduction by rapid exchange of mobile genetic elements (or exosomes) in unicellular, and rapid changes in the symbiotic microorganisms of animals. In all cases, adaptive changes can be transmitted across generations, allowing rapid improvement and assimilation in a few generations. TESTING THE HYPOTHESIS: We provide testable predictions derived from the hypothesis. IMPLICATIONS OF THE HYPOTHESIS: The hypothesis raises a critical, but thus far overlooked adaptation problem and explains how random variation can self-organize to confer a wide range of individual-specific adaptations beyond the existing outcomes of natural selection. It portrays gene regulation as an inseparable synergy between natural selection and adaptation by improvisation. The latter provides a basis for Lamarckian adaptation that is not limited to a specific mechanism and readily accounts for the remarkable resistance of tumors to treatment.

Delayed development induced by toxicity to the host can be inherited by a bacterial-dependent, transgenerational effect.

Commensal gut bacteria in many species including flies are integral part of their host, and are known to influence its development and homeostasis within generation. Here we report an unexpected impact of host-microbe interactions, which mediates multi-generational, non-Mendelian inheritance of a stress-induced phenotype. We have previously shown that exposure of fly larvae to G418 antibiotic induces transgenerationally heritable phenotypes, including a delay in larval development, gene induction in the gut and morphological changes. We now show that G418 selectively depletes commensal Acetobacter species and that this depletion explains the heritable delay, but not the inheritance of the other phenotypes. Notably, the inheritance of the delay was mediated by a surprising trans-generational effect. Specifically, bacterial removal from F1 embryos did not induce significant delay in F1 larvae, but nonetheless led to a considerable delay in F2. This effect maintains a delay induced by bacterial-independent G418 toxicity to the host. In line with these findings, reintroduction of isolated Acetobacter species prevented the inheritance of the delay. We further show that this prevention is partly mediated by vitamin B2 (Riboflavin) produced by these bacteria; exogenous Riboflavin led to partial prevention and inhibition of Riboflavin synthesis compromised the ability of the bacteria to prevent the inheritance. These results identify host-microbe interactions as a hitherto unrecognized factor capable of mediating non-Mendelian inheritance of a stress-induced phenotype.

High-resolution microbial community reconstruction by integrating short reads from multiple 16S rRNA regions.

The emergence of massively parallel sequencing technology has revolutionized microbial profiling, allowing the unprecedented comparison of microbial diversity across time and space in a wide range of host-associated and environmental ecosystems. Although the high-throughput nature of such methods enables the detection of low-frequency bacteria, these advances come at the cost of sequencing read length, limiting the phylogenetic resolution possible by current methods. Here, we present a generic approach for integrating short reads from large genomic regions, thus enabling phylogenetic resolution far exceeding current methods. The approach is based on a mapping to a statistical model that is later solved as a constrained optimization problem. We demonstrate the utility of this method by analyzing human saliva and Drosophila samples, using Illumina single-end sequencing of a 750 bp amplicon of the 16S rRNA gene. Phylogenetic resolution is significantly extended while reducing the number of falsely detected bacteria, as compared with standard single-region Roche 454 Pyrosequencing. Our approach can be seamlessly applied to simultaneous sequencing of multiple genes providing a higher resolution view of the composition and activity of complex microbial communities.

Pax6 is required for normal cell-cycle exit and the differentiation kinetics of retinal progenitor cells.

The coupling between cell-cycle exit and onset of differentiation is a common feature throughout the developing nervous system, but the mechanisms that link these processes are mostly unknown. Although the transcription factor Pax6 has been implicated in both proliferation and differentiation of multiple regions within the central nervous system (CNS), its contribution to the transition between these successive states remains elusive. To gain insight into the role of Pax6 during the transition from proliferating progenitors to differentiating precursors, we investigated cell-cycle and transcriptomic changes occurring in Pax6 (-) retinal progenitor cells (RPCs). Our analyses revealed a unique cell-cycle phenotype of the Pax6-deficient RPCs, which included a reduced number of cells in the S phase, an increased number of cells exiting the cell cycle, and delayed differentiation kinetics of Pax6 (-) precursors. These alterations were accompanied by coexpression of factors that promote (Ccnd1, Ccnd2, Ccnd3) and inhibit (P27 (kip1) and P27 (kip2) ) the cell cycle. Further characterization of the changes in transcription profile of the Pax6-deficient RPCs revealed abrogated expression of multiple factors which are known to be involved in regulating proliferation of RPCs, including the transcription factors Vsx2, Nr2e1, Plagl1 and Hedgehog signaling. These findings provide novel insight into the molecular mechanism mediating the pleiotropic activity of Pax6 in RPCs. The results further suggest that rather than conveying a linear effect on RPCs, such as promoting their proliferation and inhibiting their differentiation, Pax6 regulates multiple transcriptional networks that function simultaneously, thereby conferring the capacity to proliferate, assume multiple cell fates and execute the differentiation program into retinal lineages.

Dual requirement for Pax6 in retinal progenitor cells.

Throughout the developing central nervous system, pre-patterning of the ventricular zone into discrete neural progenitor domains is one of the predominant strategies used to produce neuronal diversity in a spatially coordinated manner. In the retina, neurogenesis proceeds in an intricate chronological and spatial sequence, yet it remains unclear whether retinal progenitor cells (RPCs) display intrinsic heterogeneity at any given time point. Here, we performed a detailed study of RPC fate upon temporally and spatially confined inactivation of Pax6. Timed genetic removal of Pax6 appeared to unmask a cryptic divergence of RPCs into qualitatively divergent progenitor pools. In the more peripheral RPCs under normal circumstances, Pax6 seemed to prevent premature activation of a photoreceptor-differentiation pathway by suppressing expression of the transcription factor Crx. More centrally, Pax6 contributed to the execution of the comprehensive potential of RPCs: Pax6 ablation resulted in the exclusive generation of amacrine interneurons. Together, these data suggest an intricate dual role for Pax6 in retinal neurogenesis, while pointing to the cryptic divergence of RPCs into distinct progenitor pools.