Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer_journal/nup155-insufficiency-recalibrates-a-pluripotent-transcriptome-with-AWgLNjxf63
- Publisher
- Springer Journals
- Copyright
- Copyright © 2018 by The Author(s).
- Subject
- Life Sciences; Bioinformatics; Systems Biology; Simulation and Modeling; Computational Biology/Bioinformatics; Physiological, Cellular and Medical Topics; Algorithms
- eISSN
- 1752-0509
- DOI
- 10.1186/s12918-018-0590-x
- Publisher site
- See Article on Publisher Site
Abstract
Background: Atrial fibrillation is a cardiac disease driven by numerous idiopathic etiologies. NUP155 is a nuclear pore complex protein that has been identified as a clinical driver of atrial fibrillation, yet the precise mechanism is unknown. The present study employs a systems biology algorithm to identify effects of NUP155 disruption on cardiogenicity in a model of stem cell-derived differentiation. Methods: Embryonic stem (ES) cell lines (n = 5) with truncated NUP155 were cultured in parallel with wild type (WT) ES cells (n = 5), and then harvested for RNAseq. Samples were run on an Illumina HiSeq 2000. Reads were analyzed using Strand NGS, Cytoscape, DAVID and Ingenuity Pathways Analysis to deconvolute the NUP155- disrupted transcriptome. Network topological analysis identified key features that controlled framework architecture and functional enrichment. Results: In NUP155 truncated ES cells, significant expression changes were detected in 326 genes compared to WT. These genes segregated into clusters that enriched for specific gene ontologies. Deconvolution of the collective framework into discrete sub-networks identified a module with the highest score that enriched for Cardiovascular System Development, and revealed NTRK1/TRKA and SRSF2/SC35 as critical hubs within this cardiogenic module. Conclusions: The strategy of pluripotent transcriptome deconvolution used in the current study identified a novel association of NUP155 with potential drivers of arrhythmogenic AF. Here, NUP155 regulates cardioplasticity of a sub-network embedded within a larger framework of genome integrity, and exemplifies how transcriptome cardiogenicity in an embryonic stem cell genome is recalibrated by nucleoporin dysfunction. Keywords: NUP155, Atrial fibrillation, RNAseq, Embryonic stem cells, Network bioinformatics Background increases with aging [3]. AF is defined as a sustained Electrical disorders in the heart are a hallmark feature of supraventricular tachyarrhythmia with disorganized a class of clinical cardiac pathologies called arrhythmias atrial activation and ineffective contraction that has dis- that are the underlying substrate for heart failure, stroke tinctive electrocardiogram characteristics including: fast and sudden cardiac death [1, 2]. The most common sus- atrial rate of ~ 300 beats/min; absence of P waves; and tained arrhythmia observed in a clinical setting is atrial irregular R-R intervals [1, 4]. This type of sustained fibrillation (AF), with a population prevalence that arrhythmia is accompanied by co-morbidities in the eld- erly, where the majority of this cohort presents with * Correspondence: Randolph.Faustino@SanfordHealth.org concomitant structural alterations of the heart [3]. Even Genetics and Genomics Group, Sanford Research, 2301 E. 60th Street N, though AF is more prevalent in octogenarians, a Sioux Falls, SD 57104, USA Department of Pediatrics, Sanford School of Medicine of the University of percentage of patients less than 60 years of age appear in South Dakota, 1400 W. 22nd Street, Sioux Falls, SD 57105, USA the clinic with a “healthy heart” history. These Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Preston et al. BMC Systems Biology (2018) 12:62 Page 2 of 13 individuals are diagnosed with idiopathic or lone AF, an maintaining pH homeostasis in cardiomyocytes [25]. unexplained arrhythmia where clinical studies report in- NUP155, being a critical scaffolding component of the conclusive or negative results [5]. Studies to address this NPC, in a homozygous mutant form has been shown to gap in knowledge have focused on ion channel gene var- impair atrial electrical signaling and give rise to clinical iants, however recent work has attributed significant atrial fibrillation [19]. This presents as ectopic initiation contributions of several non-ion channel substrates to of contraction, reentrant impulses, and futile cycling that AF [6–8]. Among these, nucleoporins (nups) have ultimately compromises cardiac function and leads to emerged as potential epigenomic regulatory proteins. sudden death in early childhood [19]. In contrast to the Nups comprise the nuclear pore complexes (NPCs), nup-associated cardiopathologies described above, the which are large toroidal structures with a main function precise cellular and molecular mechanisms by which of directing the selective transport of macromolecules NUP155 contributes to supraventricular arrhythmias between the cytoplasm and the nucleus. NPCs were first such as atrial fibrillation remains unknown. described in Xenopus oocytes using electron micros- Systems and network biology algorithms can identify copy, with modern understanding of its intricate cryptogenic drivers of AF through high throughput data- structure and morphology revealed by advanced tech- set cartography. This approach has been used to profile niques such as cryo-electron microscopy (cryo-EM) cardiogenic transcriptome changes and capture remod- and super-resolution microscopy [9, 10]. NPCs are eled intermolecular relationships to identify categories of composed of cytoplasmic, inner, and nucleoplasmic functional perturbation in a cardiopathological model of rings, each of which consists of multiple copies of differentiation [26, 27]. Here, a network-based bioinfor- nups stacked and linked together to form distinct matic strategy was applied to decipher complex systems NPC subcomplexes [10, 11]. Furthermore, interactions biology impacts of NUP155 disruption in an embryonic among discrete NPC subcomplexes create specialized stem cell line that models mammalian arrhythmogenesis. structural and functional domains within the pore. This work is the first to characterize pluripotent For example, the Y-subcomplex is a well characterized transcriptome remodeling regulated by NUP155, where component of the NPC that interacts with the inner we identify novel and high value NUP155-regulated ring subcomplex to form the NPC scaffold [12]. Re- candidates associated with AF etiology. Significantly, cent studies in eukaryotes have revealed that apart transcriptome networks that arise from NUP155 insuffi- from their canonical function as architectural compo- ciency revealed alterations in membrane function and nents of the pore and nucleocytoplasmic transport extracellular interactions, and specifically identified mediators, nups play a significant role in regulation NTRK1/TRKA and SRSF2/SC35 as hubs essential to the of transcriptional activity and chromatin structure/ integrity of a cardiogenic sub-network. organization that impacts phenotype [13–15]. Indeed, nup-driven differentiation is conserved among a var- Methods iety of metazoans, where nups play active roles in de- All media and reagents have been procured from Fisher velopment [16–18]. Scientific, unless specifically noted. NUP155-deficient Moreover, altered nup dynamics have been associated embryonic stem cell lines along with wild type paren- with normal and pathologic cardiogenesis, with a range tal cell lines were obtained from Bay Genomics of clinical phenotypes that range from morphological de- (Berkeley, CA). fects to contractile and electrical impairment of heart 2+ function [19, 20]. For example, some components of the Embryoid body formation and Ca imaging NPC inner ring subcomplex affect nuclear localization Embryoid body (EB) formation and imaging to measure 2+ and histone acetylation of the HOXA gene cluster that Ca transients in contractile EBs was performed as pre- underlies mesodermal development and proper cardiac viously described [26, 28]. Mouse embryonic stem (ES) morphology [21, 22]. Disrupted NUP188 results in con- cells were maintained in Glasgow’s Minimum Essential genital heart defects (CHD) associated with left-right Medium (GMEM, Gibco) supplemented with penicillin patterning disorders [23]. Idiopathic and dilated cardio- G/streptomycin (Pen/Strep), sodium pyruvate (Lonza myopathy, in which myocardial function is progressively BioWhittaker), non-essential amino acids (NEAA, impaired, is associated with NDC1, NUP160, NUP153, Corning), β-mercaptoethanol (β-ME, Sigma-Aldrich), NUP93, and NUP62 expression changes that together 7.5% fetal bovine serum (FBS, EMD Millipore) and disrupt nuclear transport [20, 24]. Intracellular acidifica- ESGRO leukemia inhibitory factor (LIF, EMD Millipore), tion associated with ischemic cardiac disease was and passaged three times to establish stable growth be- reported to be regulated in part by NUP35 through its fore embryoid body (EB) formation. ES cell lines were ability to bind the 5’ UTR of nhe1, an mRNA that en- differentiated into three-layered EBs using the codes a sodium-hydrogen exchanger essential for hanging-drop method. Briefly, cells were harvested and Preston et al. BMC Systems Biology (2018) 12:62 Page 3 of 13 resuspended in differentiation medium that contained via Genespring GX to prioritize cardiogenic nup candi- 20% FBS without LIF, to a concentration of 8 × 10 cells/ dates. To determine nup genes that demonstrated con- ml. To facilitate EB formation, hanging drops were cre- sistent and significant changes in expression during ated by depositing 25 μl of the cell suspension on the cardiogenesis, differential gene expression analysis and lids of 500-cm2 square culture plates and incubated for self-organizing map (SOM) classification were independ- 48 h. To induce spontaneous differentiation, EBs were ently performed between undifferentiated LIF+ ES cells flushed and transferred to floating suspension for an- (ES LIF+) and stem cell-derived cardiomyocytes (CM). other 48 h. Following differentiation, cells were cultured For increased resolution of WT and NUP155-deficient in differentiation media containing GMEM supple- ES transcriptomes, five independent biological replicates +/− mented with Pen/Strep, sodium pyruvate, NEAA, β-ME of WT and NUP155 lines (n = 10 total) were submit- and 20% FBS. Alternatively, Aggrewell (STEMCELL ted for RNAseq. RNA libraries were prepared according Technologies Inc., Cambridge, MA) plates were used to to the manufacturer’s instructions for the TruSeq RNA promote uniform EB size and organization and were Sample Prep Kit v2 (Illumina, San Diego, CA) from cultured in differentiation media as described above. EBs 100 ng of total RNA. Briefly, polyA mRNA was purified were grown for three days, with media changes as neces- from total RNA using oligo dT magnetic beads. Purified sary before transferring to gelatin coated dishes. Beating mRNA was fragmented at 95 °C for 8 min and eluted foci could be observed between 5 and 7 days after from the beads. Double stranded cDNA was prepared plating. using SuperScript III reverse transcriptase, random 2+ Ca imaging: Contractile EBs were incubated in primers (Invitrogen, Thermo Fisher Scientific, Waltham, 2+ Tyrode’s solution at 37 °C and then loaded with the Ca MA) and DNA polymerase I and RNase H. The cDNA indicator dye, Fluo-4-AM (5 μM) for 15 min. Stained ends were repaired and an “A” base added to the 3′ EBs were imaged with a Zeiss LSM Live 5 laser confocal ends. TruSeq paired end index DNA adaptors (Illumina, microscope (Zeiss, Oberkochen, Germany). Spontaneous San Diego, CA) with a single “T” base overhang at the 2+ Ca transients were recorded at 37 °C using ZEN 2.1 3′ end were ligated and resulting constructs were puri- software (Zeiss, Oberkochen, Germany), and plotted as a fied using AMPure SPRI beads (Agencourt Bioscience, function of time using Excel (Microsoft, Redmond, WA). Beverly, MA). The adapter-modified DNA fragments were enriched by 12 cycles of PCR using Illumina Cell culture and RNA extraction TruSeq PCR primers (Illumina, San Diego, CA). The Wild type (WT) and NUP155 exon truncated concentration and size distribution of the libraries were +/− E14TG2a.4 (NUP155 ) feeder independent mouse ES determined using an Agilent Bioanalyzer DNA 1000 chip cell lines were cultured on 0.1% gelatin coated 100 mm (Agilent Technologies, Santa Clara, CA) and Qubit dishes grown in 10 ml of GTES medium consisting of fluorometry (Invitrogen, Thermo Fisher Scientific, 85% Glasgow MEM (GMEM), 15% ES qualified Fetal Waltham, MA). Bovine Serum (FBS), sodium pyruvate, non-essential Libraries were sequenced at 5 samples per lane to gen- amino acids (NEAA), penicillin/streptomycin (PenStrep), erate 70–90 million reads per sample following Illumi- β-mercaptoethanol (β-ME) and ESGRO Leukemia na’s standard protocol using the Illumina cBot and cBot Inhibitory Factor (LIF). After initial plating (seeding Paired end cluster kit version 3. The flow cells were se- 6 6 density at 1.0 × 10 –1.5 × 10 cells), cells were main- quenced as 101 × 2 paired end reads on an Illumina tained in culture for 2–3 passages, changing GTES HiSeq 2000 using TruSeq SBS sequencing kit version 3 media as required. At approximately 80% confluency, and HCS v2.0.12 data collection software. Base-calling cells were passaged by treatment with 5 ml of 0.25% was performed using Illumina’s RTA v1.17.21.3. This trypsin for 4 min at 37 °C. Trypsin digestion was data was deposited into the NIH GEO database with ac- arrested by addition of an equal amount of GTES media. cession number GSE111596. This suspension was centrifuged at 1500 rpm for 4 min, and pellets of cells were either resuspended in GTES Determination of loss of function intolerance metrics and media to be re-plated or in PBS prior to RNA extrac- differential expression analysis tion. Total RNA was extracted with an RNeasy Mini The Exome Aggregation Consortium (ExAC) browser kit according to manufacturer’s protocol (Qiagen, was used to investigate pathological potential of our Germantown, MD) in preparation for sequencing on identified nups [29]. Data extracted from the ExAC HiSeq 2000 System (Illumina, San Diego, CA). browser included probability of loss of function (LoF) intolerance (pLI) metric, and z scores for missense Transcriptome deconvolution metrics [30]. Interrogation of previously published Gene Expression For RNAseq bioinformatics, raw reads (as *.bam files) Omnibus (GEO) dataset ID# GDS3729 was performed were imported into Strand NGS for expression analysis Preston et al. BMC Systems Biology (2018) 12:62 Page 4 of 13 (Agilent Technologies, Santa Clara, CA). Samples were Results aligned to the Mus musculus genome (Build mm10) and Discrete nucleoporin gene expression changes in annotated using RefSeq (Release 80). Unmatched cardiogenesis paired-end reads were filtered out for downstream qual- Cardiac fate is regulated by temporospatial gene ex- ity control. These datasets were further refined by re- pression [26], and nucleoporins are emerging as key moving reads that did not reach a mapping quality players in the determination of cardiac structure and above 20, as well as those that did not surpass function. Previous gene expression analysis in a model vendor-established quality control (QC) criteria in of stem cell-derived cardiogenesis revealed global Strand NGS (Agilent Technologies, Santa Clara, CA). down-regulation of nuclear transport genes with Final QC reads were normalized by DESeq with cardiac differentiation [36]. To gain novel insights median of all samples used as baseline. All reads rep- into nup expression dynamics in cardiogenesis, we resented a total of 36,172 gene entities that were performed a nup-focused SOM cluster analysis of our indexed according to fold change and statistical sig- original GEO dataset GDS3729 and identified a nificance, using the criteria of a 2.0-fold change (or unique nup gene set that contained Nup153, Nup155, greater) and possessing a p-value of 0.05 or less to Nup85, Rae1, and Tpr. (Fig. 1a). Analysis of these identify a quality filtered transcriptome, for a total of genes using the. 326 genes. Exome Aggregation Consortium (ExAC) browser re- vealed that only Nup155 and Nup153 possessed negative Gene ontology analysis, and network cartography missense constraint Z-scores (more variants than ex- This signature transcriptome was separated into up- pected) with probability of loss of function (LoF) intoler- regulated (176 entities) and downregulated (150 en- ance (pLI) metrics of > 0.9, which infers extreme LoF tities) groups for functional annotation, KEGG, and intolerant genes, while the remaining three NPC pro- Reactome pathway enrichment analysis using DAVID teins (Nup85, Rae1, and Tpr) had a positive Z-score (in- Bioinformatics Resources [31–34]. To determine over ferred as increased constraints with fewer variants) with representation or enrichment, the DAVID algorithm high pLI (Fig. 1b). Furthermore, Nup155 and Nup153 employs a modified Fisher’s exact test that is incorpo- demonstrated consistent significance confirmed by inde- rated into a score that reports relative priority [32]. pendent volcano plot and quality control thresholding Gene lists defined for each cluster were submitted to analyses. Of these, Nup155 emerges as the most signifi- DAVID using Entrez Gene identifiers for downstream cantly changed nucleoporin transcript (p = 0.000879), analyses. The highest classification stringency was downregulated by more than 3.5-fold in stem selected to maintain robust groups and scores were cell-derived cardiomyocytes (Fig. 1c). reported for KEGG and Reactome pathways when applicable. Dysrhythmia of NUP155 deficient contractile embryoid To map functional interactions among genes within bodies the quality filtered transcriptome, the total gene list Differentiation of ES cells into beating embryoid bodies was submitted to Ingenuity Pathway Analysis (IPA; (EBs) recapitulates cardiac phenotypes of automaticity Qiagen, Germantown, MD) toidentifysubnetworks and electromechanical coupling (Fig. 2). Fluorescent 2+ within the transcriptome, and construct an integrated quantitation of Ca transients in wild type control (WT de novo gene regulatory network (GRN) to determine Ctrl) EBs demonstrated constant and regular rhythm overall functional priorities and network topology. A (Fig. 2a, b). Treatment of WT Ctrl with the β-adrenergic total of 10 subnetworks were identified that were as- receptor agonist isoproterenol (Iso, 10 μM) increased 2+ sembled into one inclusive network using the “Merge the frequency of Ca cycling (Fig. 2c, d) depicted by a 2+ Networks” function within IPA. Edges within this significant decrease of time between Ca signal peaks in collective network indicate functional interactions WT treated EBs compared with WT Ctrl (Fig. 2i). In among genes, supported by published empirical obser- contrast, unstimulated control NUP155 deficient +/− vations curated within the IPA database. These rela- (NUP155 Ctrl) contractile EBs (Fig. 2e) exhibited tionship data were collated and exported in .xls drastically increased beating frequency, reminiscent of format using the “Export Data ➔ Export ➔ All myocardial fibrillation, with variable and diminished amp- 2+ Relationships” feature within IPA, and served as an litude of Ca waves compared to WT EBs (Fig. 2f, j). Iso- +/− input file for network analysis in Cytoscape [35]. proterenol stimulation in the NUP155 EBs (Fig. 2g) 2+ Graph theory metrics, to quantify network structure exacerbated the irregularity of the Ca waves, with and topology, were determined using the “Network responses that ranged from a blunted chronotropic effect Analyzer” tool in Cytoscape, and data was used to to loss of agonist response (Fig. 2h), however no difference +/− prioritize gene targets for further analysis. in interval times was observed compared with NUP155 Preston et al. BMC Systems Biology (2018) 12:62 Page 5 of 13 Transcriptome remodeling in NUP155-disrupted embryonic stem cells Dysfunctional contractility in beating EBs is supported by the clinical role of NUP155 in arrhythmogenesis [19], together with previous reports that have identified gene activation and repression associated with NUP155 in neonatal rat ventricular myocytes [37]. These data sug- gest a broad capacity for NUP155 to remodel global gene expression profiles in a cardiac setting. To investi- gate the effects of NUP155 in a cardiogenic context, mouse ES cells that harbor disrupted NUP155 were examined by RNAseq to understand transcriptome changes precipitated by NUP155 in a pluripotent back- ground. Principal component analysis (PCA) plots +/− distinguished WT from NUP155 transcriptomes (Fig. 3a). Hierarchical clustering of individual transcrip- tomes demonstrated clear segregation and reproducibil- ity of gene expression profiles for each biological sample category (Fig. 3b). Replicate analysis was performed to delimit transcripts to those changing by 2.0 fold or greater as well as meeting significance criteria of p < 0.05. A total of 176 and 150 up and downregulated genes were identified that met these criteria (Fig. 3c). Pathway enrichment analysis using DAVID [31] identi- fied several thematic clusters for up (14 clusters) and downregulated (11 clusters) gene lists (Additional file 1: Tables S1 and S2). The significant functional terms, depicted in Table 1 (p ≤ 0.05), included functions related to integrin alpha for both up and downregulated genes (Upregulated and Downregulated Clusters 1), protein phosphorylation (Upregulated Cluster 2), transmem- brane tyrosine protein kinase activity (Up Cluster 3), and cysteine switch/zinc binding (Upregulated Cluster 7, Table 1). Molecular cartography prioritizes genes within transcriptome subnetworks Up and downregulated genes were integrated using In- genuity Pathway Analysis to form a collective de novo gene regulatory network. Gene relationships were ex- tracted and exported for analysis in Cytoscape where they were visualized in a circular layout to emphasize Fig. 1 Nucleoporins in cardiac differentiation. a Nup153, Nup155, nodes with high edge density (Fig. 4a). Topology analysis Nup85, Rae1, and Tpr expression profiles during cardiac specification. confirmed scale-free and hierarchical properties innate Undifferentiated stem cells (ES LIF+); early differentiated stem cells (ES LIF-); to biological networks (Fig. 4b and inset). Neighborhood cardiac precursors (CP), cardiomyocytes (CM). b ExAC constraint metrics and pLI values for Nup153, Nup155, Nup85, Rae1, and Tpr. c Venn diagrams connectivity (Fig. 4c), betweenness centrality (Fig. 4d) intersect discrete gene groups independently parsed from the same high and closeness centrality (Fig. 4e) plots of the integrated throughput dataset to identify nucleoporins (nup) that emerge as robust NUP155 transcriptome identified preferential attach- candidates involved in cardiogenesis. Nup155 is the highest priority ment nature and bridging nodes key to network molecule (p < 0.001). SOM = self-organizing map; QC = quality control connectivity and information flow. The highest between- ness scores in this NUP155-remodeled transcriptome Ctrl (Fig. 2i). Taken together these results show that were, in order of priority: APP, HNF4A, TP53, NTRK1, NUP155-deficient EBs are prone to electrical instability and CTNNB1 (Fig. 4d), whereas the top closeness cen- which may serve as a substrate for atrial fibrillation. trality scores were associated with the same 5 molecules Preston et al. BMC Systems Biology (2018) 12:62 Page 6 of 13 2+ Fig. 2 Contractile NUP155 deficient embryoid bodies exhibit dysrhythmia. a Embryoid bodies (EBs) were loaded with Fluo4-AM to visualize Ca handling during systolic and diastolic phases of contractile cycling, as indicated. Color legend to left of image identifies fluorescence of regions 2+ 2+ that range from high Ca concentration in red, medium concentration in cyan, to low concentration in blue. b Ca handling for each region of interest (ROI), with timescale in seconds (s) and range of fluorescent intensity units (i.u.) indicated in lower right. c, d Visualization and measurement of contractile EBs following treatment with 10 mM isoproterenol (Iso). e, f Untreated beating areas in NUP155 deficient EBs demonstrates increased 2+ frequency of contractile cycling, with variable Ca handling. g, h Agonist treatment of NUP155 deficient EBs aggravates the irregular contractile cycling observed in unstimulated controls that ranged in severity from hypercontractility to loss of rhythmic contraction. I Measurement of intervals 2+ +/− between peaks highlight a significant decrease of time between Ca waves in NUP155 EBs compared to controls, independent of isoproterenol 2+ +/− treatment (n =3, *p <0.05 vs WT Ctrl, **p< 0.05 vs WT Iso). j Changes in mean amplitude of Ca waves. WT Iso treated, and NUP155 EBs with and without Iso treatment did not show significant differences, but were significantly decreased compared to WT Ctrl. (n= 3, *p< 0.05 vs WT Ctrl) +/− in a different order of priority, i.e. TP53, CTNNB1, revealed no significant differences between NUP155 NTRK1, APP, and HNF4A (Fig. 4e). and WT (Additional file 1: Figure S1). Ingenuity Pathways Analysis reports molecular functional enrichment of each subnetwork within the Discussion collective network, and when complemented with A growing body of evidence supports multiple roles for DAVID-based analyses, comprehensive functional cat- nups in cell fate acquisition, yet the contributions of egories can be identified that one approach alone may nups to normal differentiation are incompletely charac- not detect, as well as corroborate robust enrichment of terized. The present study employs a network strategy to consistent gene ontology categories [26–28]. This deconvolute multivariate systems biology impacts of approach revealed the sub-network with the highest sig- NUP155 in a pro-arrhythmogenic embryonic stem cell nificance score (Score = 67) that prioritized Cardiovascu- model of cardiogenesis and captures a capacity for lar System Development and Function (Fig. 5a). Here, NUP155 to remodel a pluripotent transcriptome. Key NTRK1/TRKA was integrated as the primary hub with molecules associated with cardiac innervation and the highest degree, betweenness and closeness centrality fibrosis were identified in a module enriched for Cardio- coefficients, followed by SRSF2/SC35. (Fig. 5b, c). Spe- vascular Development within a larger NUP155 remod- cific examination of NTRK1 and SRSF2 expression data eled network. This work identifies transcriptome +/− in NUP155 compared to WT control ES cell lines recalibration caused by NUP155 deficiency that under- confirmed significance and magnitude of expression lies arrhythmogenic elements, and supports a develop- change (Fig. 5d, e). Western blot analysis (Additional file mental function for nups beyond canonical roles in NPC 1: Supplemental Methods) showed that SRSF2 and architecture and nucleocytoplasmic transport. TRKA protein level changes followed the same trend as To gain insights into allele frequencies of nups that gene expression data, however densitometric analysis may predispose to cardiac disease, we used the ExAC Preston et al. BMC Systems Biology (2018) 12:62 Page 7 of 13 +/− Fig. 3 Molecular signature of NUP155 truncation in a pluripotent genome. Deep transcriptome profiling of WT and NUP155 embryonic stem cells was performed using RNAseq. a Principal component analysis (PCA) revealed distinct hallmark gene expression profiles with clear segregation of +/− +/− WT from NUP155 transcriptomes. Filled circles represent ES (dark grey) and NUP155 (light green) transcriptomes of distinct biological replicates, plotted in a three dimensional volumetric space. Axes:X – PC1 (24.08%), Y – PC2 (13.2%), Z – PC3 (10.38%). b Pairwise correlation of samples reveals +/− reproducible clustering of discrete up and downregulated gene expression patterns that define ES and NUP155 populations. Lower right: colorscale indicates normalized intensity, where red, yellow, and blue represent upregulated, no change, and downregulated trends, respectively. c Volcano plot +/− of gene expression changes to enumerate up and downregulated mRNA in the NUP155 transcriptome, according to criteria of absolute Fold Change (FC) > 2.0 and p < 0.05. Filled circles in red represent up-regulated genes that meet this criteria, while blue circles represent downregulated transcripts. Circles in grey indicate genes that fall below the indicated threshold values. A total of 326 genes were identified that met the filtering parameters, with 176 up and 150 downregulated, respectively. PC – Principal Component browser to investigate the anticipated number of vari- on global mechanisms of gene regulation. In support of ants for the nups we identified in our study, as well as this, previous work has reported a capacity for NUP155 their tolerance (or intolerance) to variation. All nups to bind, tether and regulate discrete regions of chromatin identified in the present work possess a high pLI score that control gene expression in yeast and Drosophila that is expected given the essential role of nucleoporins models [15, 40]. NUP155 may also act indirectly through in eukaryotic cell viability. However, of the 5 identified interactions with histone modifiers, as described above for in this study, Nup85, Rae1, and Tpr demonstrated a studies in rat models of cardiac hypertrophy [37], or in- positive constraint metric z-score for missense muta- ferred by electronic protein interaction datasets [41, 42]. tions, while Nup155 and Nup153 had negative missense In the present study, analysis of functional annotation constraint z-scores. This suggests that Nup155 and clusters for both up and downregulated revealed consist- Nup153 exist within the population with a high toler- ent enrichment of terms related to membrane interactions ance to variation that may present as developmental dis- and/or transmembrane biology. This was supported by ease rather than terminal nonviability [30, 38]. Indeed, a non-clustered significant functional terms (Additional file landmark clinical study by Zhang et al. identified a mu- 1: Tables S3 and S4). Select nucleoporins regulate cell tation in NUP155 that led to atrial fibrillation [19], while adhesion primarily through altered nucleocytoplasmic recent work by Nanni et al. identified an altered role for trafficking [43], and NUP155 may influence cell mem- NUP153 in cardiac chromatin regulation in patients with brane biology via this functional modality since NUP155 Duchenne muscular dystrophy [39]. In the former study, is critical for nuclear pore complex assembly, formation, NUP155 protein levels were normal, with experimental and nuclear transport [13, 44, 45]. models revealing disrupted nucleocytoplasmic transport Controlled temporospatial execution of molecular pro- that was concluded to be the main cause of arrhythmo- grams is required for normal differentiation, driven by genic compromise. The latter study identified a patho- the composition and architecture of underlying gene logical up-regulation of NUP153 combined with networks. Disruption of these networks recalibrates plur- increased NUP153 acetylation that led to dysregulated ipotency that leads to compromised phenotypes. In the expression of nexilin with calcium channel gain of func- present study, TP53 was prioritized in a molecular tion [39]. It is of note that NUP155 regulates cardiac framework driven by NUP155 truncation, supported by hypertrophy through HDAC4 [37], providing additional recent demonstration of TP53 as a master regulator that evidence for the potential role of nups as epigenomic controls hypertrophic responses of the myocardium [46]. regulators. Though investigation of our dataset revealed a Indeed, the diversity of genes whose expression is non-significant increase in TP53 expression in heterozy- dysregulated by NUP155 insufficiency suggests impacts gous ES cell lines (data not shown), its prioritization in a Preston et al. BMC Systems Biology (2018) 12:62 Page 8 of 13 Table 1 Pathway enrichment analysis of differentially expressed genes Database Term Description p-Value FE Upregulated Cluster 1 (1.84) Uniprot Seq Feature short sequence motif GFFKR motif 0.005 27.897 InterPro IPR018184 Integrin alpha chain, C-terminal cytoplasmic 0.007 22.714 region, conserved site IPR000413 Integrin alpha chain 0.008 21.452 IPR013649 Integrin alpha-2 0.008 21.452 IPR013517 FG-GAP repeat 0.008 21.452 IPR013519 Integrin alpha beta-propeller 0.009 20.323 SMART SM00191 Integrin alpha 0.012 17.511 EMBL-EBI GO:0008305 Integrin complex 0.016 15.321 Cluster 2 (1.50) InterPro IPR000719 Protein kinase, catalytic domain 0.019 2.499 IPR011009: Protein kinase-like domain 0.029 2.315 EMBL-EBI GO:0004672 Protein kinase activity 0.029 2.314 GO:0006468 protein phosphorylation 0.040 2.180 GO:0016310 phosphorylation 0.054 2.052 Cluster 3 (1.35) InterPro IPR020635 Tyrosine-protein kinase, catalytic domain 0.025 6.356 IPR008266 Tyrosine-protein kinase, active site 0.041 5.201 SMART SM00219 TyrKc, Tyrosine kinase, catalytic domain 0.036 5.477 EMBL-EBI GO:0007169 transmembrane receptor protein tyrosine 0.045 5.023 kinase signaling pathway UniProtKB Tyrosine-protein kinase 0.052 4.750 Cluster 7 (0.64) Uniprot Seq Feature metal ion-binding site Zinc binding site, in inhibited form 0.024 12.505 short sequence motif Cysteine switch 0.041 9.299 Downregulated Cluster 1 (1.84) Uniprot Seq Feature short sequence motif GFFKR motif 0.005 27.897 InterPro IPR018184 Integrin alpha chain, C-terminal 0.007 22.714 cytoplasmic region, conserved site IPR000413 Integrin alpha chain 0.008 21.452 +/− Pathway enrichment analysis of significantly up and downregulated genes in Nup155 ES cells performed with DAVID. Here are depicted the clusters with functional terms that reached significance of p ≤ 0.05. Each cluster shows their respective enrichment score in parentheses. Database represents the online functional database used to extract each term; Term represents the pathway identification; Description is the pathway symbol; and FE refers to the fold enrichment score pluripotent transcriptome network underlying cardio- as a hub that integrates discrete network neighborhoods pathological manifestation is reinforced by recent and determines informational flow among those regions. systems biology meta-analyses of disease-causing NTRK1/TRKA and SRSF2/SC35 were identified here protein-protein interaction (PPI) networks [47]. Pinero as the most upregulated and downregulated genes with et al. describe in their study a general multiscale meso- highest degrees, respectively, within the sub-network scopic molecular signature that underlies disease, where that prioritizes Cardiovascular System Development, tumor suppressors such as TP53 possess high centrality with mRNA and protein expression changes trending in and are essential hubs within the network structure that the same direction. Although protein expression changes are the most sensitive to genomic perturbation. The next did not reach significance, such transcript and protein prioritized network hubs are dominant disease genes, expression discrepancies are expected, given that followed by recessive disease genes in modules with low multiple post-transcriptional and post-translational centrality located at the network periphery [47]. Top- mechanisms may regulate final expression level [48]. ology analysis of the present NUP155-recalibrated tran- This does not preclude the potential for NUP155 to im- scriptome is in line with this biological network pact cardiac development through a NUP155-TRKA sig- property, as homozygous TP53 possesses the highest naling cascade, however. NTRK1/TRKA is a tyrosine degree and closeness centrality scores that identify TP53 receptor kinase that drives cholinergic differentiation Preston et al. BMC Systems Biology (2018) 12:62 Page 9 of 13 Fig. 4 Network cartography of a nucleoporin-disrupted transcriptome. (a) The collective transcriptome inclusive of up and downregulated transcripts were analyzed by Ingenuity Pathways Analysis to identify experimentally observed interactions among the 326 genes. The network generated from this data was visualized in a circular layout that positions nodes circumferentially with their connections (edges) plotted diametrically. Singletons are network nodes with only one connection to the larger network and are arrayed on the outside of the circle plot. This layout emphasizes nodes that have high edge density, seen in this network on the right. Right panel: Magnification of network arc with high edge density. Nodes were colored properties according to degree, or number of connections, where high degree is represented by dark red and low degree in white, shown here in the colorscale above panel. b Topological analysis revealed a clustering coefficient distribution associated with hierarchical network structure. Inset: Degree distribution demonstrates a power law relationship indicative of scale-free architecture (c) Neighborhood connectivity plot identifies dissortative nature of the network, where highly connected nodes tend to connect to nodes with a lower number of edges. d Nodes with high betweenness centrality are critical to maintaining network integrity as they connect other regions of the network to one another. APP, HNF4A, TP53, NTRK1/ TRKA, and CTNNB1 possessed distinct betweenness centrality scores that segregated them from other nodes in the network. Inset: Legend identifies genes with the topmost betweenness centrality scores, ranked in order from highest to lowest, and are colored to facilitate identification within the plot. e Closeness centrality scores are important for speed of informational transmission within a network. Here, nodes with the highest closeness centrality clustered together. The nodes prioritized for high betweenness centrality measures were identical to the molecules with critical closeness centrality scores. Inset: Legend depicts nodes rank ordered from high to low. Identity of nodes with discrete centrality metrics are preserved as identical, yet distinct reprioritization of those molecules is observed on comparison of closeness versus betweenness [49], and has a defined role in promoting cardiac innerv- upstream NGF stimulation of NTRK1/TRKA that drives ation and repair [50, 51]. NTRK1/TRKA directs target CORONIN-1 mediated calcium release, which negatively cardiac innervation through its effector, CORONIN-1 regulates axon growth and arborization within the myo- [52]. This developmental cascade is initiated upon cardium [52]. Increases in NTRK1/TRKA would Preston et al. BMC Systems Biology (2018) 12:62 Page 10 of 13 +/− Fig. 5 Deconvolution of modular functional enrichment within a sub-network of the NUP155 transcriptome. a Sub-networks that comprise the larger network possess characteristic identities at the mesoscopic level. Identification of the most significant sub-network revealed robust and consistent functional enrichment in Cardiovascular System Development, and incorporated a variety of up and down-regulated genes. Magnification of hubs NTRK1/TRKA and SRSF2/SC35 shown on the right. b, c Betweenness and closeness centrality plots of this small network confirm the same molecule, NTRK1/TRKA, as critical for integration and information transmission within the module, labeled in both plots and highlighted in red. SRSF2/ SC35 possessed the next highest betweenness and closeness centrality measures, labeled and highlighted in green. d, e RNAseq abundances for NTRK1/TRKA and SRSF2 confirmed significant expression changes for both transcripts. Normalized intensities shown on y-axis Preston et al. BMC Systems Biology (2018) 12:62 Page 11 of 13 facilitate CORONIN-1 suppression of cardiac innerv- from Wild Type (WT, n = 4) and Nup155+/− (n= 4) ES cell lines. β tubulin was ation that would manifest as electrophysiological deficits. used as loading control for both proteins. Comparison of normalized ratios from densitometry analyses of c SRSF2/SC35 and (D) NTRK1/TRKA protein This is corroborated by studies that identify increased levels in WT and Nup155+/− cell lines (p >0.05). (PDF 416 kb) NTRK1/TRKA levels associated with atrial fibrillation, and is further supported by data that revealed concomi- Abbreviations tant autocrine and paracrine regulation of NTRK1/ 5’ UTR: 5′ untranslated region; AF: Atrial fibrillation; APP: Amyloid beta TRKA expression by upstream NGF [53]. Further work 2+ precursor protein; Ca : Calcium; CM: Cardiomyocytes; CP: Cardiac precursors; will clarify this potential NUP155-TRKA axis in the CTNNB1: Beta catenin; DAVID: Database for annotation, visualization and integrated discovery; EBs: Embryoid bodies; ES: Embryonic stem; FBS: Fetal context of cardiac development. bovine serum; GEO: Gene expression omnibus; GTES: GMEM ES cell media; The most downregulated hub identified within this HNF4A: Hepatocyte nuclear factor 4 alpha; HOXA: Homeobox A cluster; Cardiovascular Development sub-network was SRSF2/ LIF: Leukemia inhibitory factor; NDC1: Nuclear division cycle homolog 1; NEAA: Non-essential amino acids; NGF: Nerve growth factor; NHE1: Sodium/ SC35, implicated by protein expression data where the hydrogen exchanger 1; NPC: Nuclear pore complex; NTRK1/TRKA: Tropomyosin trend matched the decrease in SRSF2 mRNA. SRSF2 is a +/− related kinase A; NUP: Nucleoporin; Nup155 : NUP155 exon trapped cell line; serine/arginine rich splicing factor essential for pluripo- PBS: Phosphate buffered saline; QC: Quality control; RAE1: Ribonucleic acid export 1; ROI: Region of interest; SC35/SRSF2: Serine/arginine rich splicing factor tent self-renewal, with decreased SRSF2 promoting stem 2; SOM: Self-organizing map; TBX5: T-box protein 5; TP53: Tumor protein 53; cell differentiation [54]. SRSF2 dynamics are highly Tpr: Translocated promoter region; WT: Wild type; β-ME: Beta-mercaptoethanol regulated in cardiac tissue, as cardiac-specific ablation of SRSF2 results in dilated cardiomyopathy and abnormal Acknowledgements 2+ Ca handling linked to down-regulation of the cardiac We are grateful for the help provided by Dr. Jin Jen and the Genome Analysis Core within the Medical Genome Facility at the Mayo Clinic, Rochester MN; and specific ryanodine receptor [55]. Balanced interactions discussions with Dr. Johan Martijn Bos on the genetics of arrhythmias and the between SRSF2 and TBX5 are necessary for proper ExAC browser. pre-mRNA splicing critical for cardiac development [56]. Indeed, disruptions to the TBX5/SRSF2 equilibrium Funding result in Holt-Oram Syndrome, which include cardiac All aspects of this work, including study design, data collection, analysis, interpretation, and writing of the manuscript was supported by Sanford conduction diseases such as AF [57, 58] that can occur Research and by the American Heart Association (Grant # 14SDG20380322). alone or in combination with atrial and ventricular septal defects [59]. Availability of data and materials All data generated or analyzed during this study are included in this Conclusions published article (and its Additional file). Gene networks are highly regulated, stratified structures that can be regulated at multiple levels [60]. Our results Authors’ contributions CP analyzed data, prepared figures for manuscript, as well as prepared and identify a mesoscopic cardiac sub-network impacted by edited the manuscript; SW assisted with cell culture and cardiogenic NUP155-deficient recalibration of a pluripotent tran- differentiation; SR performed calcium imaging and data analysis; BE assisted scriptome. Here, NUP155 insufficiency re-organizes a with preparation and processing of samples for next generation sequencing; ES performed western blot analysis and prepared supplemental material, as molecular network that prioritizes TRKA and SRSF2 as well as revised the manuscript; RF designed the study, acquired data, potential factors in the development of cardiac AF. The performed bioinformatic analyses, and wrote and revised manuscript. All idea that nups may epigenomically remodel the cardiac authors have read and approved the manuscript. program is supported by the role of various nups in repressing and/or activating discrete chromatin regions Ethics approval and consent to participate Not applicable. [15, 40, 61–63] and previous work that identified regu- lated nup expression in cardiogenesis [26]. In particular, Competing interests future work focused on elucidating the role of the spli- The authors declare that they have no competing interests. cing factor SRSF2 will provide deeper insights into the epigenomic function and cardiogenic role of NUP155 Publisher’sNote predicted by the present systems biology study. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional file Author details Genetics and Genomics Group, Sanford Research, 2301 E. 60th Street N, Additional File 1: Table S1. Pathway Enrichment Analysis of Upregulated Sioux Falls, SD 57104, USA. Department of Dermatology, Mayo Clinic, 200 Genes. Table S2. Pathway Enrichment Analysis of Downregulated Genes. 1st St SW, Rochester, MN 55905, USA. Department of Surgery, Wake Forest Table S3. Functional terms not clustered during pathway enrichment University Health Sciences, Medical Center Boulevard, Winston-Salem, NC analysis of Upregulated genes. Table S4. Functional terms not clustered 27157, USA. Medical Genome Facility, Mayo Clinic, 200 1st St SW, Rochester, during pathway enrichment analysis of Downregulated genes. Supplemental MN 55905, USA. Department of Pediatrics, Sanford School of Medicine of Methods:Western Blot Analysis. Figure S1. Protein Expression data for SRSF2/ the University of South Dakota, 1400 W. 22nd Street, Sioux Falls, SD 57105, SC35 and NTRK1/TRKA. Western blots of a SRSF2/SC35 and b NTRK1/TRKA USA. Preston et al. BMC Systems Biology (2018) 12:62 Page 12 of 13 Received: 10 January 2018 Accepted: 24 May 2018 21. Labade AS, Karmodiya K, Sengupta K. HOXA repression is mediated by nucleoporin Nup93 assisted by its interactors Nup188 and Nup205. Epigenetics Chromatin. 2016;9:e54. https://doi.org/10.1186/s13072-016-0106-0. 22. Di-Poi N, Koch U, Radtke F, Duboule D. Additive and global functions of HoxA cluster genes in mesoderm derivatives. Dev Biol. 2010;341(2):488–98. References https://doi.org/10.1016/j.ydbio.2010.03.006. 1. January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC Jr, et al. 23. Del Viso F, Huang F, Myers J, Chalfant M, Zhang Y, Reza N, et al. Congenital 2014 AHA/ACC/HRS guideline for the management of patients with atrial heart disease genetics uncovers context-dependent organization and fibrillation: executive summary: a report of the American College of function of nucleoporins at cilia. Dev Cell. 2016;38(5):478–92. https://doi.org/ Cardiology/American Heart Association task force on practice guidelines 10.1016/j.devcel.2016.08.002. and the Heart Rhythm Society. Circulation. 2014;130(23):2071–104. https:// 24. Cortes R, Rosello-Lleti E, Rivera M, Martinez-Dolz L, Salvador A, Azorin I, et al. doi.org/10.1161/CIR.0000000000000040. Influence of heart failure on nucleocytoplasmic transport in human 2. Reinier K, Marijon E, Uy-Evanado A, Teodorescu C, Narayanan K, Chugh H, et cardiomyocytes. Cardiovasc Res. 2010;85(3):464–72. https://doi.org/10.1093/ al. The association between atrial fibrillation and sudden cardiac death. cvr/cvp336. JACC Heart Fail. 2014;2(3):221–7. https://doi.org/10.1016/j.jchf.2013.12.006. 25. Xu L, Pan L, Li J, Huang B, Feng J, Li C, et al. Nucleoporin 35 regulates 3. Mirza M, Strunets A, Shen WK, Jahangir A. Mechanisms of arrhythmias and cardiomyocyte pH homeostasis by controlling Na+-H+ exchanger-1 conduction disorders in older adults. Clin Geriatr Med. 2012;28(4):555–73. expression. J Mol Cell Biol. 2015;7(5):476–85. https://doi.org/10.1093/jmcb/ https://doi.org/10.1016/j.cger.2012.08.005. mjv054. 4. Fye WB. Tracing atrial fibrillation–100 years. N Engl J Med. 2006;355(14): 26. Faustino RS, Behfar A, Perez-Terzic C, Terzic A. Genomic chart guiding 1412–4. https://doi.org/10.1056/NEJMp068059. embryonic stem cell cardiopoiesis. Genome Biol. 2008;9(1):R6. https://doi. 5. Blagova OV, Nedostup AV, Kogan EA, Sulimov VA, Abugov SA, Kupriyanova AG, org/10.1186/gb-2008-9-1-r6. et al. Myocardial Biopsy In “Idiopathic” Atrial Fibrillation And Other Arrhythmias: 27. Faustino RS, Wyles SP, Groenendyk J, Michalak M, Terzic A, Perez-Terzic C. Nosological Diagnosis, Clinical And Morphological Parallels, And Treatment. J Systems biology surveillance decrypts pathological transcriptome remodeling. Atr Fibrillation. 2016;9(1):1414. https://doi.org/10.4022/jafib.1414. BMC Syst Biol. 2015;9:36. https://doi.org/10.1186/s12918-015-0177-8. 6. Perez-Serra A, Campuzano O, Brugada R. Update about atrial fibrillation 28. Faustino RS, Chiriac A, Niederlander NJ, Nelson TJ, Behfar A, Mishra PK, et al. genetics. Curr Opin Cardiol. 2017;32(3):246–52. https://doi.org/10.1097/HCO. Decoded calreticulin-deficient embryonic stem cell transcriptome resolves latent cardiophenotype. Stem Cells. 2010;28(7):1281–91. https://doi.org/10. 7. Tucker NR, Ellinor PT. Emerging directions in the genetics of atrial 1002/stem.447. fibrillation. Circ Res. 2014;114(9):1469–82. https://doi.org/10.1161/ 29. The Exome Aggregation Consortium (ExAC) browser. http://exac. CIRCRESAHA.114.302225. broadinstitute.org/. Accessed 10 Nov 2017. 8. Olesen MS, Nielsen MW, Haunso S, Svendsen JH. Atrial fibrillation: the role 30. Lek M, Karczewski KJ, Minikel EV, Samocha KE, Banks E, Fennell T, et al. of common and rare genetic variants. Eur J Hum Genet. 2014;22(3):297–306. Analysis of protein-coding genetic variation in 60,706 humans. Nature. 2016; https://doi.org/10.1038/ejhg.2013.139. 536(7616):285–91. https://doi.org/10.1038/nature19057. 9. Callan HG, Tomlin SG. Experimental studies on amphibian oocyte nuclei. I. 31. The Database for Annotation, Visualization and Integrated Discovery Investigation of the structure of the nuclear membrane by means of the (DAVID) v6.8. https://david.ncifcrf.gov/. Accessed 5 May 2017. electron microscope. Proc R Soc Lond B Biol Sci. 1950;137(888):367–78. 32. Huang DW, Sherman BT, Tan Q, Collins JR, Alvord WG, Roayaei J, et al. The 10. von Appen A, Beck M. Structure determination of the nuclear pore complex DAVID gene functional classification tool: a novel biological module-centric with three-dimensional Cryo electron microscopy. J Mol Biol. 2016;428(10 Pt algorithm to functionally analyze large gene lists. Genome Biol. 2007;8(9): A):2001–10. https://doi.org/10.1016/j.jmb.2016.01.004. R183. https://doi.org/10.1186/gb-2007-8-9-r183. 11. Alber F, Dokudovskaya S, Veenhoff LM, Zhang W, Kipper J, Devos D, et al. 33. Huang DW, Sherman BT, Tan Q, Kir J, Liu D, Bryant D, et al. DAVID The molecular architecture of the nuclear pore complex. Nature. 2007; bioinformatics resources: expanded annotation database and novel 450(7170):695–701. https://doi.org/10.1038/nature06405. algorithms to better extract biology from large gene lists. Nucleic Acids Res. 12. Vollmer B, Lorenz M, Moreno-Andres D, Bodenhofer M, De Magistris P, 2007;35(Web Server issue):W169–75. https://doi.org/10.1093/nar/gkm415. Astrinidis SA, et al. Nup153 recruits the Nup107-160 complex to the inner 34. Huang d W, Sherman BT, Lempicki RA. Systematic and integrative analysis nuclear membrane for Interphasic nuclear pore complex assembly. Dev Cell. 2015;33(6):717–28. https://doi.org/10.1016/j.devcel.2015.04.027. of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009; 4(1):44–57. https://doi.org/10.1038/nprot.2008.211. 13. Schwartz M, Travesa A, Martell SW, Forbes DJ. Analysis of the initiation of 35. Cytoscape software platform. http://www.cytoscape.org/. Accessed 5 May nuclear pore assembly by ectopically targeting nucleoporins to chromatin. Nucleus. 2015;6(1):40–54. https://doi.org/10.1080/19491034.2015.1004260. 36. Perez-Terzic C, Faustino RS, Boorsma BJ, Arrell DK, Niederlander NJ, Behfar A, 14. Gomez-Cavazos JS, Hetzer MW. The nucleoporin gp210/Nup210 controls et al. Stem cells transform into a cardiac phenotype with remodeling of the muscle differentiation by regulating nuclear envelope/ER homeostasis. J Cell nuclear transport machinery. Nat Clin Pract Cardiovasc Med. 2007;4(Suppl 1): Biol. 2015;208(6):671–81. https://doi.org/10.1083/jcb.201410047. S68–76. https://doi.org/10.1038/ncpcardio0763. 15. Breuer M, Ohkura H. A negative loop within the nuclear pore complex 37. Kehat I, Accornero F, Aronow BJ, Molkentin JD. Modulation of chromatin controls global chromatin organization. Genes Dev. 2015;29(17):1789–94. position and gene expression by HDAC4 interaction with nucleoporins. https://doi.org/10.1101/gad.264341.115. J Cell Biol. 2011;193(1):21–9. 16. Capelson M, Hetzer MW. The role of nuclear pores in gene regulation, development and disease. EMBO Rep. 2009;10(7):697–705. https://doi.org/ 38. Petrovski S, Wang Q, Heinzen EL, Allen AS, Goldstein DB. Genic intolerance 10.1038/embor.2009.147. to functional variation and the interpretation of personal genomes. PLoS 17. Dickmanns A, Kehlenbach RH, Fahrenkrog B. Nuclear pore complexes and Genet. 2013;9(8):e1003709. https://doi.org/10.1371/journal.pgen.1003709. nucleocytoplasmic transport: From Structure to Function to Disease. Int Rev 39. Nanni S, Re A, Ripoli C, Gowran A, Nigro P, D'Amario D, et al. The nuclear Cell Mol Biol. 2015;320:171–233. https://doi.org/10.1016/bs.ircmb.2015.07.010. pore protein Nup153 associates with chromatin and regulates cardiac gene expression in dystrophic mdx hearts. Cardiovasc Res. 2016;112(2):555–67. 18. Beck M, Hurt E. The nuclear pore complex: understanding its function https://doi.org/10.1093/cvr/cvw204. through structural insight. Nat Rev Mol Cell Biol. 2017;18:73-89. https://doi. 40. Van de Vosse DW, Wan Y, Lapetina DL, Chen WM, Chiang JH, Aitchison JD, org/10.1038/nrm.2016.147. et al. A role for the nucleoporin Nup170p in chromatin structure and gene 19. Zhang X, Chen S, Yoo S, Chakrabarti S, Zhang T, Ke T, et al. Mutation in silencing. Cell. 2013;152(5):969–83. https://doi.org/10.1016/j.cell.2013.01.049. nuclear pore component NUP155 leads to atrial fibrillation and early 41. STRING v10.5. https://string-db.org/. Accessed 6 Apr 2018. sudden cardiac death. Cell. 2008;135(6):1017–27. https://doi.org/10.1016/j. cell.2008.10.022. 42. GeneCards: Human Gene Database http://www.genecards.org/. Accessed 6 20. Tarazon E, Rivera M, Rosello-Lleti E, Molina-Navarro MM, Sanchez-Lazaro IJ, Apr 2018. Espana F, et al. Heart failure induces significant changes in nuclear pore 43. Funasaka T, Balan V, Raz A, Wong RW. Nucleoporin Nup98 mediates complex of human cardiomyocytes. PLoS One. 2012;7(11):e48957. https:// galectin-3 nuclear-cytoplasmic trafficking. Biochem Biophys Res Commun. doi.org/10.1371/journal.pone.0048957. 2013;434(1):155–61. https://doi.org/10.1016/j.bbrc.2013.03.052. Preston et al. BMC Systems Biology (2018) 12:62 Page 13 of 13 44. De Magistris P, Tatarek-Nossol M, Dewor M, Antonin W. A self-inhibitory interaction within Nup155 and membrane binding are required for nuclear pore complex formation. J Cell Sci. 2018;131:1-9. jcs208538. https://doi.org/ 10.1242/jcs.208538. 45. Eisenhardt N, Redolfi J, Antonin W. Interaction of Nup53 with Ndc1 and Nup155 is required for nuclear pore complex assembly. J Cell Sci. 2014; 127(Pt 4):908–21. https://doi.org/10.1242/jcs.141739. 46. Mak TW, Hauck L, Grothe D, Billia F. p53 regulates the cardiac transcriptome. Proc Natl Acad Sci U S A. 2017;114(9):2331–6. https://doi.org/10.1073/pnas. 47. Pinero J, Berenstein A, Gonzalez-Perez A, Chernomoretz A, Furlong LI. Uncovering disease mechanisms through network biology in the era of next generation sequencing. Sci Rep. 2016;6:24570. https://doi.org/10.1038/srep24570. 48. Gan H, Cai T, Lin X, Wu Y, Wang X, Yang F, et al. Integrative proteomic and transcriptomic analyses reveal multiple post-transcriptional regulatory mechanisms of mouse spermatogenesis. Mol Cell Proteomics. 2013;12(5): 1144–57. https://doi.org/10.1074/mcp.M112.020123. 49. Wang L, He F, Zhong Z, Lv R, Xiao S, Liu Z. Overexpression of NTRK1 promotes differentiation of neural stem cells into cholinergic neurons. Biomed Res Int. 2015;2015:857202. https://doi.org/10.1155/2015/857202. 50. Meloni M, Caporali A, Graiani G, Lagrasta C, Katare R, Van Linthout S, et al. Nerve growth factor promotes cardiac repair following myocardial infarction. Circ Res. 2010;106(7):1275–84. https://doi.org/10.1161/ CIRCRESAHA.109.210088. 51. Lorentz CU, Alston EN, Belcik T, Lindner JR, Giraud GD, Habecker BA. Heterogeneous ventricular sympathetic innervation, altered beta-adrenergic receptor expression, and rhythm instability in mice lacking the p75 neurotrophin receptor. Am J Physiol Heart Circ Physiol. 2010;298(6):H1652– 60. https://doi.org/10.1152/ajpheart.01128.2009. 52. Suo D, Park J, Young S, Makita T, Deppmann CD. Coronin-1 and calcium signaling governs sympathetic final target innervation. J Neurosci. 2015; 35(9):3893–902. https://doi.org/10.1523/JNEUROSCI.4402-14.2015. 53. Saygili E, Schauerte P, Kuppers F, Heck L, Weis J, Weber C, et al. Electrical stimulation of sympathetic neurons induces autocrine/paracrine effects of NGF mediated by TrkA. J Mol Cell Cardiol. 2010;49(1):79–87. https://doi.org/ 10.1016/j.yjmcc.2010.01.019. 54. Lu Y, Loh YH, Li H, Cesana M, Ficarro SB, Parikh JR, et al. Alternative splicing of MBD2 supports self-renewal in human pluripotent stem cells. Cell Stem Cell. 2014;15(1):92–101. https://doi.org/10.1016/j.stem.2014.04.002. 55. Ding JH, Xu X, Yang D, Chu PH, Dalton ND, Ye Z, et al. Dilated cardiomyopathy caused by tissue-specific ablation of SC35 in the heart. EMBO J. 2004;23(4):885–96. https://doi.org/10.1038/sj.emboj.7600054. 56. Fan C, Chen Q, Wang QK. Functional role of transcriptional factor TBX5 in pre-mRNA splicing and Holt-Oram syndrome via association with SC35. J Biol Chem. 2009;284(38):25653–63. https://doi.org/10.1074/jbc.M109.041368. 57. Cerbai E, Sartiani L. Holt-oram syndrome and atrial fibrillation: opening the (T)-box. Circ Res. 2008;102(11):1304–6. https://doi.org/10.1161/CIRCRESAHA. 108.178079. 58. Baruteau AE, Probst V, Abriel H. Inherited progressive cardiac conduction disorders. Curr Opin Cardiol. 2015;30(1):33–9. https://doi.org/10.1097/HCO. 59. Jhang WK, Lee BH, Kim GH, Lee JO, Yoo HW. Clinical and molecular characterisation of Holt-Oram syndrome focusing on cardiac manifestations. Cardiol Young. 2015;25(6):1093–8. https://doi.org/10.1017/ S1047951114001656. 60. Kobayashi T, Masuda N. Fragmenting networks by targeting collective influencers at a mesoscopic level. Sci Rep. 2016;6:37778. https://doi.org/10. 1038/srep37778. 61. Seo HS, Blus BJ, Jankovic NZ, Blobel G. Structure and nucleic acid binding activity of the nucleoporin Nup157. Proc Natl Acad Sci U S A. 2013;110(41): 16450–5. https://doi.org/10.1073/pnas.1316607110. 62. Lapetina DL, Ptak C, Roesner UK, Wozniak RW. Yeast silencing factor Sir4 and a subset of nucleoporins form a complex distinct from nuclear pore complexes. J Cell Biol. 2017;216(10):3145–59. https://doi.org/10.1083/jcb.201609049. 63. Toda T, Hsu JY, Linker SB, Hu L, Schafer ST, Mertens J, et al. Nup153 interacts with Sox2 to enable bimodal gene regulation and maintenance of neural progenitor cells. Cell Stem Cell. 2017;21(5):618–34 e7. https://doi.org/10. 1016/j.stem.2017.08.012.
Journal
BMC Systems Biology – Springer Journals
Published: May 30, 2018