• Users Online: 29
  • Print this page
  • Email this page


 
 
Table of Contents
ORIGINAL ARTICLE
Year : 2018  |  Volume : 7  |  Issue : 1  |  Page : 35-42

Network Mining Indicated the Triglycerides as the Most Related Clinical Relevance to Age-related Transcriptional Changes in the Aorta


Department of Plant Breeding and Biotechnology, Sari Agricultural Sciences and Natural Resources University, Mazandaran, Iran

Date of Web Publication26-Feb-2018

Correspondence Address:
Dr. Fereshteh Izadi
Sari Agricultural Sciences and Natural Resources University, Mazandaran
Iran
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/rcm.rcm_14_17

Get Permissions

  Abstract 


Background: Aging is believed to be one of the main causes of cardiovascular diseases. The incidence of cardiovascular dysfunctions has increased substantially over the past few years. However, our understanding of molecular mechanisms of age-related vascular disorders remains somehow unclear, and an effective treatment has not been developed. A biological network is a collection of interactions between molecular regulators and their targets in cells governing gene expression level that is usually built by employing omics data, facilitating the inference of molecular basis of complex diseases. Materials and Methods: GSE50833 series containing aorta samples of 6-month-old mice (n = 6) and 20-month-old mice (n = 6) obtained from Janvier labs (Saint Berthevin, France) were downloaded from Gene Expression Omnibus database and the verified Agilent probe IDs were subjected to build a weighted gene coexpression network by a bioinformatics tool known as Weighted Gene Coexpression Network Analysis. We then conducted a network-driven integrative analysis to find significant modules and underlying pathways. Results: The unique genes extracted from normalized gene expression values were parsed into six modules. Among the incorporated clinical traits, the most significant module was associated with triglycerides enriched in biological terms, including proteolysis, blood circulation, and circulatory system process. Moreover, Enpp5, Fez1, Kif1a, F3, H2-Q7, and Pa × 8 were taken as putative hallmark molecules by further screening. Conclusion: the main goal of this analysis was the prioritization of genes that likely play a role in the pathogenesis of vascular diseases. We attempted to provide a system understanding of the potential connections among these genes.

Keywords: Age-related vascular diseases, gene coexpression network, gene modules


How to cite this article:
Izadi F. Network Mining Indicated the Triglycerides as the Most Related Clinical Relevance to Age-related Transcriptional Changes in the Aorta. Res Cardiovasc Med 2018;7:35-42

How to cite this URL:
Izadi F. Network Mining Indicated the Triglycerides as the Most Related Clinical Relevance to Age-related Transcriptional Changes in the Aorta. Res Cardiovasc Med [serial online] 2018 [cited 2021 Mar 9];7:35-42. Available from: https://www.rcvmonline.com/text.asp?2018/7/1/35/226158




  Introduction Top


Over the course of a lifespan, cells due to losing their functions would be vulnerable against stresses therefore become more susceptible to diseases. Although vascular aging studies in rodent models have been led to promising results toward the determination of age-associated vascular dysfunction,[1] there is still lack of knowledge in the details of physiological arterial aging at the molecular and cellular levels. Fundamental changes of the vascular wall, loss of vascular integrity, and vessel homeostasis are characteristics by which vascular aging can be differentiated from atherosclerotic pathology.[2] Generally, by aging through the alterations in collagen and elastin contents as well as reorganization of extracellular matrix, diameter of arterial increased and walls be remodeled.[2] Physiological aging phenomena mostly thought to be controlled by signaling system affected by several of endogenous and exogenous factors.[3],[4],[5] The detrimental changes in aortic characteristics mainly occur due to a number of processes in cellular level including DNA, protein and lipid oxidation, decreased protein synthesis, increased angiotensin II levels, and mitochondrial dysfunction.[6],[7],[8] The alterations of cellular processes in gross dimensions will clearly followed by severe inflammation and immune responses in aged aorta.[9]

The elucidation of the putative genes and pathways underlying cardiovascular changes by increasing age is then critical for understanding the molecular mechanisms of vascular diseases. Previous researches have discussed the role of insulin-like growth factor-1,[10] mitochondria,[11] telomere attrition,[12] and FOXOs/sirtuins angiogenesis [13] in vascular aging. However, in spite of conducting enormous researches on understanding heart diseases including vascular dysfunction, these challenging diseases remain a leading cause and 40% of all deaths worldwide.[14] On the other hand, aging is an unavoidable cardiovascular risk factor known to be associated with another parameters such as lipid level. For example, aging-induced aorta stiffening demonstrated to be a natural consequence of increasing age that might be due to the several impairment including alteration in aortic wall.[15]

Coexpression network analysis enables us to cluster tremendous of genes by assigning them to known functions in which they are involved. Among the coexpression network inference algorithms, Weighted Gene Coexpression Network Analysis (WGCNA) is a relatively new statistical consistent software not only to infer correlation patterns between two genes but also to cover neighborhoods across expression data.[16] While the constructed gene networks generally could be divided into modules by tools such as ModuLand,[17] topologically analyzed by NetworkAnalyzer [18] or Cytohubba,[19] and gene clusters functionally annotated by BiNGO,[20] WGCNA contains a comprehensive set of functions such as visualization, clustering, topology, and enrichment analysis. Thereon, a comparative assay of different network analysis methods indicated that WGCNA is able to network construction, subnetwork extraction, and hub genes identification from  Escherichia More Details coli data.[21] A module can be described as a stable unit underlying biological networks when its function can remain the same, while individual gene expression can be changed or replaced by other genes with similar redundant functions.[17] The genomic differences and the genes potentially responsible for vascular aging have not been identified in C57BL/6 mice sufficiently and those mice are, although the most widely used “background” of laboratory animals used for vascular aging research. However, expression profiling of aged aorta in C57BL/6 mice displayed arterial dysfunction by increased stiffness and associated systolic hypertension.[22]

In this study, we hypothesized a cross-talk between transcriptional changes in aorta by increasing age and fat components. We then compared the transcriptional profile of aorta in young and old mice by conducting coexpression network analysis to explore putative modules and prognostic genes involved in a triangle of aging, vascular dysfunctions, and lipid level.


  Materials and Methods Top


Data acquisition and preprocessing

The Agilent TXT files of GSE50833 series was downloaded from the National Center for Biotechnology Information Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/geo/). These data consisted of a total of 12 samples based on the GPL10787 platform correspond to the aortas of 6-month-old mice and 20-month–old-matched mice. Next, raw files were preprocessed with quantile method (https://www. bioconductor.org/). After quantile normalization, Linear Models for Microarray Data (limma) R package (https://www.bioconductor.org/) was applied to extract the variable genes between 6- and 20-month-old mice with the cutoff of false discovery rate correction (FDR) <0.05. Ultimately, the extracted Agilent probe IDs were transformed into Agilent MIT IDs. More detailed information can be found in the Supplementary Notes.

Network construction

The most variable genes (FDR <0.05) from the normalized expression data on aorta between adult and old mice were used to construct a weighted network using the WGCNA package (https://labs.genetics.ucla.edu/horvath/CoexpressionNetwork/Rpackages/WGCNA/) in R software v3.2.5 (https://www.bioconductor.org/). To do so, first, calculated absolute values of the Pearson's correlation coefficient for all pair-wise comparisons of gene expression values were transformed into a similarity matrix. Next, the similarity matrix was applied to step-by-step network construction and a module detection method.[16] To this end, a weighted matrix was built with a scaling factor beta (β = 6) and using the assumption that biological networks are scale-free.[18] The modules were computed by assigning a minimum of 30 genes per module and keeping the default value of SplitDepth in 2 for a medium sensitivity of cluster splitting. These parameters would optimize scale-free topology and robust node connectivity criteria when any two genes were connected and the edge weight was determined by the topology overlap measure (TOM) provided in WGCNA. Furthermore, genes were clustered into modules by utilizing average linkage hierarchical clustering using “1-TOM” as the distance measure and modules determined by the Dynamic Hybrid Tree Cut algorithm.[16] Finally, in order to trimming genes whose correlation with module eigengene is less than the defined threshold (0.25), similar modules with highly eigengenes (the first principal component of modules) were merged. WGCNA determines highly interconnected nodes as modules designated in different colors. Moreover, this algorithm is able to relate the identified modules to external provided information including clinical outcomes of quantitative traits, annotating these modules to biological terms, and finally network visualization and exporting of networks to external software such as VisANT [23] that visualizes biological interactions.

Gene ontology analysis and visualization

Gene ontology (GO) analysis reveals functions of gene products of significant modules from three aspects: biological process, cellular component (CC), and molecular function (MF). WGCNA also uses the online software Database for Annotation, Visualization, and Integrated Discovery (DAVID)[24] to determine whether the particular modules are significantly enriched regarding known GOs. DAVID facilitates high-throughput gene functional analysis not only for a list of genes but also for biological information for an individual gene. Here, to study the potential functions of significant modules, module annotation enrichment function embedded in WGCNA using DAVID tool was exploited with default parameters.

Exploring transcription factors: miRNA–mRNA interactions in particular modules

By the default setting, WGCNA identifies the most 30 high-rank intramedullary hub genes for each module to be visualized by visualization functions. These genes are being calculated by intramodular connectivity and tend to be highly connected within modules as well as correlated with the first principle competent measures.[16] The most 90 interconnected genes within the significant modules were graphically depicted by Cytoscape 3.4.0.[25] Further study on the interplay between a network of hub genes in the significant modules and microRNAs was done by utilizing predicted miRNA-target networks for Mus Musculus from TargetScan (http://www.targetscan.org/vert_71/) and MicroCosm Mouse (Release 5) (http://www.ebi.ac.uk/enright-srv/microcosm/htdocs/targets/v5/). For expanding the network, CyTargetLinker Cytoscape plugin was employed.[26] CyTargetLinker enhances biological networks using the provided information in the frame of Regulatory Interaction Networks (RegIN). RegIN is a network in XGMML format containing regulatory interactions. Finally, we extracted putative miRNAs generated by the intersection of the miRNA targets from TargetScan and microcosm. Transcription factors control the expression of gene within modules; thus, iRegulon Cytoscape plugin [27] was used to identify the possible regulons and cofactors associated with a collection of intramedullary hub genes extracted from the prognosticate modules separately.

Module preservation

To determine the reliability of results and comparing the modular structure in inferred coexpression network against a reference network, we downloaded a list of 1163 significantly differentially expressed probe sets (P< 0.05) from a study done by Gräbner et al.,[28] from which we reconstructed a test network and contrasted the preservation of coexpression across testing and reference datasets to detect the conservation of gene pairs between two networks. This analysis aided to detect if any of the modules were dysregulated or perturbed in testing samples relative to reference samples. In sum, the network module preservation statistics determines whether modules that were identified in one network (reference) remain connected in the other network (test) by calculating a Zsummary statistic computed by basic R functions as a composite measure of statistics related to network density and connectivity. A module shows no evidence of preservation among the datasets if its Zsummary statistic is smaller than two, whereas a statistics between two and ten corresponds to moderately preserved (reproducible) modules and above ten strongly preserved modules. Of note, the genes within preserved modules are differentially connected between two networks however might be or not differentially expressed. Zsummary statistic is being calculated by this equation;




  Results Top


Network construction and module detection

The normalized Agilent Probe IDs were more filtered for removing ambiguous probes and duplicated genes. Ultimately, the expression values of unique genes were selected for subsequent analysis. In graph theory, certain nodes with higher connections are more likely to be pivotal connectors, and likewise, critical points controlling important dynamic components in biological networks [29] that removal of them causes biological systems fail to save their coherence. To the identification of such an interconnected genes from aging-associated coexpression network in mice aorta, abstractly, adjacency matrix was obtained from the Pearson's correlation matrix to a power β = 6 based on the scale-free topology criterion. Next, the represented matrix was transformed to similarity matrix in clustering process when genes grouped in a cluster likely to be biologically relevant to the same pathway. The power β = 6 can emphasize robust correlations and removing unreliable correlations on an exponential level. [Figure 1]a shows the determination of β parameter. In brief, given Agilent Series GSE50833 WGCNA was used to establish modules in coexpression network where we obtained six modules [Figure 1]b. As illustrated in [Figure 1]b, modules in gene dendrogram are displayed in different colors, and based on dynamic branch cutting algorithm, underneath row color assigns the modules membership after setting a minimum module size of 30 and DeepSplit in 2.
Figure 1: (a) Parameter analysis of inferred coexpression network. The β value of 6 was selected which resulted in signed networks that were approximately scale-free and yielded a significantly high scale-free topology. (b) Hierarchical clustering of genes in significant modules. The colors are assigned to each module by the Dynamic Tree Cut

Click here to view


Modules were associated with aging and lipid level

The main goal in this study was gaining more insights into the role a given gene module may play in vascular dysfunction in aged mice. We therefore associated module eigengenes with phenotypic traits, lipid level in this study. The first principal component of the gene expression matrix for each module was retained as the representative module eigengenes (MEs) which represents the average normalized gene expression for a module. Two blue and turquoise modules were established to be positively correlated with triglycerides, leptin, and free fatty acid (FFA) with cutoff of absolute correlation >0.6 and P < 0.05 [Figure 2]a. [Figure 2]b illustrates the correlation between turquoise module size and triglycerides. Furthermore, we examined whether significant modules (turquoise, blue, and red) enriched for GO terms relevant to aging vascular disorders. GO enrichment of turquoise module has been provided in [Table 1] and the GO analysis of the other modules can be found in Supplementary Notes.
Figure 2: (a) Scatterplot shows a highly significant correlation between gene significant versus module membership in the turquoise module with triglyceride. (b) The heatmap showing correlation between assayed traits and module eigengene values. Green and red colors represent the negative and positive correlation, respectively

Click here to view
Table 1: Gene ontology enrichment analysis of genes assigned to the turquoise module

Click here to view


miRNA–mRNA-transcription factor interaction network in significant modules

Our logic for applying TargetScan and MicroCosm databases was enriching our data mining with biological regulators such as miRNAs and their predicted targets. Interestingly, miRNA-target analysis showed that in the turquoise module, gene F3 was putatively regulated by 23 and 38 predicted miRNAs from TargetScan and MicroCosm, respectively [Figure 3]a. F3 is related to angiogenesis and coagulation cascades and has been suggested to be involved in pathways that are relevant with apoptosis in aged mice.[30] In addition, in the blue module, H2-Q7 was illustrated to be regulated by 78 miRNAs in MicroCosm [Figure 3]b. H2-Q7 (Qa-2) is a member of murine nonclassical MHC class I molecules involved in the modulation of immune responses; its expression displayed gradual increasing according to age.[31] H2-Q7 differential expression may contribute to the distinct patterns of mouse susceptibility/resistance to infectious and noninfectious disorders. However, the analysis showed no gene found to be regulated by miRNAs in the red module. Among the miRNAs, mmu-miR-449a, mmu-miR-449c, mmu-miR-34c, mmu-miR-34b-5p, mmu-miR-15a, and mmu-let-7 were listed as common between networks built by turquoise and blue modules. In a study by Melo-Lima et al.,[32] mir-34, mir-15, and mir-449 miRNA families exhibited significant distinct expression patterns according to age in mice. miR let-7 plays a role in tissue homeostasis, repair, and stem cell aging.[33] Due to the inconvenience of considering a highly dimensional set of transcription factors, we only investigated transcription factors in a network of the 90 connected hub nodes depicted for mRNA-miRNA interaction analysis; hereby, Pax8 and Hsf1 were ranked as the most significant based on normalized enrichment score [Table 2].
Figure 3: The network of the most top 30 interconnected genes in the turquoise module and predicted miRNAs from TargetScan and microcosm. (b) The network of the most top 30 interconnected genes in blue module and predicted miRNAs from TargetScan and Microcosm. The hub genes and miRNAs are shown in red circles and light yellow squares. mRNA-miRNA interactions from TargetScan and Microcosm are shown in green and blue lines, respectively. Dark yellow circles show the genes that are predictably regulated by miRNAs

Click here to view
Table 2: Promoter analysis of the 90 top hub genes obtained by significant modules, sorted by NES score

Click here to view


Module preservation test

To the identification of age-dependent modules preserved between constructed network and a testing network, a reference network was built from extracted gene symbols of 1063 significantly differentially expressed probe sets in cutoff of P < 0.05 from GSE10000 series. Among the expression profiles of unique genes obtained by GSE50833 datasets, only 126 gene symbols were matched with the modules defined by Toledano,[34] by which we had already built the testing network. Median rank and Zsummary values were calculated for module preservation test; however, the both of the used statistics revealed a low preservation of green and blue modules between two networks (Supplementary Notes). The blue module was among the significant modules and enriched for terms such as immune response, leukocyte differentiation, and hemopoiesis.


  Discussion Top


The purpose of this present study was the identification of potential mechanisms related to aortal transcriptional changes toward aging. In the context of network analysis approach, we attempted to explore modules that could be associated with the impacts of aging on vascular dysfunction. Given the most variable transcripts between aorta of 6- and 20-month-old mice, a coexpression network by WGCNA was built. This method summarizes genes with similar expression profiles into the same modules; therefore, cogrouped genes thought to be involved in the same pathways.[35] By Dynamic Tree Cut method, six modules ranging from 160 genes in the turquoise module to 50 genes in the red and green modules were detected. After merging modules with highly correlated eigengenes, the turquoise, blue, and red modules were chosen for functional and subsequent analysis. Hereby, to reveal the transcriptional changes trade-off with aging in vascular, we further detected trait-associated gene modules and hub genes by incorporating lipid level-related traits including triglycerides, FFA, and leptin into the constructed coexpression network. The modules are closely associated with the clinical traits likely to be an important biological clue to unveil unclear gaps between aging and the incidence of vascular disorders. Hence, this step potentially can draw interactions between the included clinical traits and aging-associated gene expression changes in the aorta. The red module did not display any correlation with the assayed traits while the turquoise module found to be the most significant and correlated module with incorporated traits [Figure 2]. Functional enrichment analysis of turquoise module by DAVID tool resulted in several terms apparently related with aging-associated vascular dysfunctions including extracellular matrix, blood circulation, circulatory system process, and proteolysis. The cross-talk between extracellular matrix and vascular disorders has been already reviewed in detail.[36]

Surprisingly, the clinical implication of significant modules indicated the triglycerides and leptin and triglycerides and FFA correlated with the turquoise and blue modules, respectively; however, the correlation between triglycerides with turquoise module was higher than the correlation between triglycerides content and the blue module. Accordingly, leptin has been suggested as a key modulator in pathways involved in vascular proliferation.[37] Furthermore, leptin sensitivity pointed to be declined during aging in rodents.[38] In addition, the hypertriglyceridemia was demonstrated as a prevalent risk factor for cardiovascular diseases [39] as well as FFA level reportedly increased with aging.[40] Module preservation via permutation testing was weakly validated by the blue and green modules in aging-related microarray dataset-derived reference network although the blue module was somehow preserved between two networks and enriched for genes involved in processes with an apparent connection to different mechanisms implying association with aging including immune response, leukocyte differentiation, and hemopoiesis. In concordance,[41] Talayero and Sacks demonstrated age-associated detrimental alterations in leukocytes that thereof affects the innate immune system. In addition, correlation between leukocyte counts with hypertriglyceridemia has been previously displayed.[42] Moreover, FFA activates leukocytes following by endothelial dysfunction through enhanced angiotensin II production.[43] The involvement of angiotensin system has been reported with vascular aging.[22]

The WGCNA R package also computes two values as gene significance (GS) and module membership (MM) when the smaller corresponding P value denotes higher Pearson's correlation coefficient between gene expression profile, incorporated clinical traits and modules, respectively.[16] Candidate mRNAs with the highest GS and MM could be considered as the genes significantly associated with phenotypic traits and module eigengene of a given module. Hereby, we focused on the most correlated turquoise module with triglycerides. The top genes regarding MM and GS were then selected as the most biologically relevant genes with triglycerides and turquoise module [Table 3]. Consequently, Enpp5, Fez1, and Kif1a were more interesting genes as common among the top 30 connected genes visualized by Cytoscape [Figure 2]b and highly ranked genes based on GS and MM scores. Fez1 encodes an elongation protein that is being abnormally aggregated in aged mice.[44] Kif1a was exhibited as a crucial regulator of synaptic aging.[45] As a result, Enpp5, Fez1, and Kif1a as highly connected intermodular genes can be considered hub genes and likely to play pivotal roles in maintaining the network functions. Centrality analysis of gene within turquoise and blue modules with CytoNCA [46] revealed the Enpp5 among the first 30 genes with the highest betweenness centrality. Enpp5 known to have catalytic activity as well as its expression was changed in Agt-shRNA-transfected adipocytes reflecting to be involved in blood pressure and lipid accumulation.[47] The role of Enpp5 in regulating blood pressure might be biologically relevant with a study by Rammos et al.[22] that confirmed age-dependent aortic dysfunction by increased systolic hypertension because of stiffness in vascular. It was also shown that Enpp5 acts as extracellular signaling molecules in a broad variety of tissues.[48] Seemingly, lipid components such as triglycerides and FFA interact with blood cells indeed stimulating immune system followed by vascular endothelial disruptions. Regarding highlighted GO terms, modules and clinical traits, genes such as F3, H2-Q7, and Enpp5 are therefore considered to be potential key elements in aortic changes by aging. Moreover, motif discovery in the upstream of hub genes revealed Pax8 and Hsf1 as the most potential regulators. Pax8 was shown to play a pivotal role in the regulation of cardiomyocyte growth and senescence [49] as well as a crucial transcription factor for developing epithelial organs.[50]
Table 3: The most correlated genes with triglycerides and the turquoise module

Click here to view


Aging is believed to be one of the most important nonmodifiable cardiovascular risk factors that lead unequivocally to a number of detrimental changes in the cardiovascular system, increasing cardiovascular morbidity and mortality.[8] During the past decades, network-based data mining approaches have successfully deciphered causal genes and relate them with human disorders.[51] Keeping with this, compared to the methods merely based on single genes derived by differential expression analysis, biological modules may represent more credible information. To provide a comparison of vascular dysfunction in two age stages in mice, therefore, in the frame of network construction and modularity analysis, a small number of genes, miRNAs, and transcription factors were discovered whose interplays could lead to the cardiovascular dysfunction toward aging. An expected outcome of such a work would possibly shed light the bridges between inevitable age-dependent changes in transcriptome and external sample traits, in addition to presenting evidences in clinical diagnosis and treatment in vascular disorders. However, we should be cautious about several notes in our study. First, small sample size was a limitation of the present study, while by utilizing more adequate number of samples, the results would be more reliable. In addition, network analysis at transcriptome level could be more intensified through merging studies with protein networks to draw more precise conclusions regarding predicted master regulators. Finally, we inferred an undirected network, whereas connectivity between nodes does not mean the causal relationships.


  Conclusion Top


Our study aimed in bioinformatics analysis of age-related alterations in gene expression in the nonatherosclerotic vascular system. Using a collection of functions for systematic analysis of biological networks implemented in the WGCNA R package, we observed several interesting results that are thought to have important influences in aging-associated cardiovascular system. Triglyceride, FFA, and leptin were significantly and positively correlated with transcriptional changes in aorta by increasing age. Moreover, genes such as Enpp5, Fez1, Kif1a, F3, H2-Q7, Pax8, and Hfs1 as well as interacting miRNAs exhibited the most connectivity with external traits and significant modules which could serve as potential candidate genes involved in vascular dysfunction toward aging. In sum, triglycerides were found as the most possible correlated trait with aortal transcriptional changes affected by aging.

We hope our study could aid in the understanding of age-related vascular diseases and could provide more information for vascular disease biomarker discovery for the future therapeutic manipulations.

Acknowledgment

We thank Dr. Nooshin Omranian, scientific staff in Systems Biology and Mathematical Modelling Group, Max Planck Institute for Molecular Plant Physiology, Potsdam, Germany, for her precious guidance.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Sindler AL, Fleenor BS, Calvert JW, Marshall KD, Zigler ML, Lefer DJ, et al. Nitrite supplementation reverses vascular endothelial dysfunction and large elastic artery stiffness with aging. Aging Cell 2011;10:429-37.  Back to cited text no. 1
[PUBMED]    
2.
Lakatta EG, Levy D. Arterial and cardiac aging: Major shareholders in cardiovascular disease enterprises: Part I: Aging arteries: A “set up” for vascular disease. Circulation 2003;107:139-46.  Back to cited text no. 2
[PUBMED]    
3.
Barzilai N, Huffman DM, Muzumdar RH, Bartke A. The critical role of metabolic pathways in aging. Diabetes 2012;61:1315-22.  Back to cited text no. 3
[PUBMED]    
4.
Greer EL, Brunet A. FOXO transcription factors in ageing and cancer. Acta Physiol (Oxf) 2008;192:19-28.  Back to cited text no. 4
[PUBMED]    
5.
Kenyon CJ. The genetics of ageing. Nature 2010;464:504-12.  Back to cited text no. 5
[PUBMED]    
6.
Kovacic JC, Moreno P, Nabel EG, Hachinski V, Fuster V. Cellular senescence, vascular disease, and aging: Part 2 of a 2-part review: Clinical vascular disease in the elderly. Circulation 2011;123:1900-10.  Back to cited text no. 6
[PUBMED]    
7.
Sawabe M. Vascular aging: From molecular mechanism to clinical significance. Geriatr Gerontol Int 2010;10 Suppl 1:S213-20.  Back to cited text no. 7
[PUBMED]    
8.
Ungvari Z, Sonntag WE, Csiszar A. Mitochondria and aging in the vascular system. J Mol Med (Berl) 2010;88:1021-7.  Back to cited text no. 8
[PUBMED]    
9.
Busse PJ, Mathur SK. Age-related changes in immune function: Effect on airway inflammation. J Allergy Clin Immunol 2010;126:690-9.  Back to cited text no. 9
[PUBMED]    
10.
Higashi Y, Quevedo HC, Tiwari S, Sukhanov S, Shai SY, Anwar A, et al. Interaction between insulin-like growth factor-1 and atherosclerosis and vascular aging. Front Horm Res 2014;43:107-24.  Back to cited text no. 10
[PUBMED]    
11.
Dai DF, Rabinovitch PS, Ungvari Z. Mitochondria and cardiovascular aging. Circ Res 2012;110:1109-24.  Back to cited text no. 11
[PUBMED]    
12.
Fyhrquist F, Saijonmaa O, Strandberg T. The roles of senescence and telomere shortening in cardiovascular disease. Nat Rev Cardiol 2013;10:274-83.  Back to cited text no. 12
[PUBMED]    
13.
Oellerich MF, Potente M. FOXOs and sirtuins in vascular growth, maintenance, and aging. Circ Res 2012;110:1238-51.  Back to cited text no. 13
[PUBMED]    
14.
Ohira T, Iso H. Cardiovascular disease epidemiology in Asia: An overview. Circ J 2013;77:1646-52.  Back to cited text no. 14
[PUBMED]    
15.
Fleenor BS, Sindler AL, Eng JS, Nair DP, Dodson RB, Seals DR, et al. Sodium nitrite de-stiffening of large elastic arteries with aging: Role of normalization of advanced glycation end-products. Exp Gerontol 2012;47:588-94.  Back to cited text no. 15
    
16.
Langfelder P, Horvath S. WGCNA: An R package for weighted correlation network analysis. BMC Bioinformatics 2008;9:559.  Back to cited text no. 16
[PUBMED]    
17.
Kovács IA, Palotai R, Szalay MS, Csermely P. Community landscapes: An integrative approach to determine overlapping network module hierarchy, identify key nodes and predict network dynamics. PLoS One 2010;5. pii: e12528.  Back to cited text no. 17
    
18.
Assenov Y, Ramírez F, Schelhorn SE, Lengauer T, Albrecht M. Computing topological parameters of biological networks. Bioinformatics 2008;24:282-4.  Back to cited text no. 18
    
19.
Chin CH, Chen SH, Wu HH, Ho CW, Ko MT, Lin CY, et al. CytoHubba: Identifying hub objects and sub-networks from complex interactome. BMC Syst Biol 2014;8 Suppl 4:S11.  Back to cited text no. 19
    
20.
Maere S, Heymans K, Kuiper M. BiNGO: A cytoscape plugin to assess overrepresentation of gene ontology categories in biological networks. Bioinformatics 2005;21:3448-9.  Back to cited text no. 20
[PUBMED]    
21.
Allen JD, Xie Y, Chen M, Girard L, Xiao G. Comparing statistical methods for constructing large scale gene networks. PLoS One 2012;7:e29348.  Back to cited text no. 21
[PUBMED]    
22.
Rammos C, Hendgen-Cotta UB, Deenen R, Pohl J, Stock P, Hinzmann C, et al. Age-related vascular gene expression profiling in mice. Mech Ageing Dev 2014;135:15-23.  Back to cited text no. 22
[PUBMED]    
23.
Hu Z, Chang YC, Wang Y, Huang CL, Liu Y, Tian F, et al. VisANT 4.0: Integrative network platform to connect genes, drugs, diseases and therapies. Nucleic Acids Res 2013;41:W225-31.  Back to cited text no. 23
[PUBMED]    
24.
Huang DW, Sherman BT, Tan Q, Kir J, Liu D, Bryant D, et al. DAVID bioinformatics resources: Expanded annotation database and novel algorithms to better extract biology from large gene lists. Nucleic Acids Res 2007;35:W169-75.  Back to cited text no. 24
[PUBMED]    
25.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res 2003;13:2498-504.  Back to cited text no. 25
[PUBMED]    
26.
Kutmon M, Kelder T, Mandaviya P, Evelo CT, Coort SL. CyTargetLinker: A cytoscape app to integrate regulatory interactions in network analysis. PLoS One 2013;8:e82160.  Back to cited text no. 26
[PUBMED]    
27.
Janky R, Verfaillie A, Imrichová H, Van de Sande B, Standaert L, Christiaens V, et al. iRegulon: From a gene list to a gene regulatory network using large motif and track collections. PLoS Comput Biol 2014;10:e1003731.  Back to cited text no. 27
    
28.
Gräbner R, Lötzer K, Döpping S, Hildner M, Radke D, Beer M, et al. Lymphotoxin beta receptor signaling promotes tertiary lymphoid organogenesis in the aorta adventitia of aged apoE-/-mice. J Exp Med 2009;206:233-48.  Back to cited text no. 28
    
29.
Faith JJ, Hayete B, Thaden JT, Mogno I, Wierzbowski J, Cottarel G, et al. Large-scale mapping and validation of Escherichia coli transcriptional regulation from a compendium of expression profiles. PLoS Biol 2007;5:e8.  Back to cited text no. 29
[PUBMED]    
30.
Puzzo D, Bizzoca A, Loreto C, Guida CA, Gulisano W, Frasca G, et al. Role of F3/contactin expression profile in synaptic plasticity and memory in aged mice. Neurobiol Aging 2015;36:1702-15.  Back to cited text no. 30
[PUBMED]    
31.
Melo-Lima BL, Evangelista AF, de Magalhães DA, Passos GA, Moreau P, Donadi EA, et al. Differential transcript profiles of MHC class ib (Qa-1, qa-2, and qa-10) and aire genes during the ontogeny of thymus and other tissues. J Immunol Res 2014;2014:159247.  Back to cited text no. 31
    
32.
Victoria B, Dhahbi JM, Nunez Lopez YO, Spinel L, Atamna H, Spindler SR, et al. Circulating microRNA signature of genotype-by-age interactions in the long-lived Ames dwarf mouse. Aging Cell 2015;14:1055-66.  Back to cited text no. 32
    
33.
Toledano H. The role of the heterochronic microRNA let-7 in the progression of aging. Exp Gerontol 2013;48:667-70.  Back to cited text no. 33
[PUBMED]    
34.
Ghazalpour A, Doss S, Zhang B, Wang S, Plaisier C, Castellanos R, et al. Integrating genetic and network analysis to characterize genes related to mouse weight. PLoS Genet 2006;2:e130.  Back to cited text no. 34
[PUBMED]    
35.
van Nas A, Guhathakurta D, Wang SS, Yehya N, Horvath S, Zhang B, et al. Elucidating the role of gonadal hormones in sexually dimorphic gene coexpression networks. Endocrinology 2009;150:1235-49.  Back to cited text no. 35
[PUBMED]    
36.
Ponticos M, Smith BD. Extracellular matrix synthesis in vascular disease: Hypertension, and atherosclerosis. J Biomed Res 2014;28:25-39.  Back to cited text no. 36
[PUBMED]    
37.
Noblet JN, Goodwill AG, Sassoon DJ, Kiel AM, Tune JD. Role of leptin in coronary vascular dysfunction and proliferation. FASEB J 2016;30 1 Suppl:733-1.  Back to cited text no. 37
    
38.
Filippi BM, Lam TK. Leptin and aging. Aging (Albany NY) 2014;6:82-3.  Back to cited text no. 38
[PUBMED]    
39.
Kaltsas GA, Korbonits M, Isidori AM, Webb JA, Trainer PJ, Monson JP, et al. How common are polycystic ovaries and the polycystic ovarian syndrome in women with Cushing's syndrome? Clin Endocrinol (Oxf) 2000;53:493-500.  Back to cited text no. 39
[PUBMED]    
40.
Talayero BG, Sacks FM. The role of triglycerides in atherosclerosis. Curr Cardiol Rep 2011;13:544-52.  Back to cited text no. 40
[PUBMED]    
41.
Stervbo U, Meier S, Mälzer JN, Baron U, Bozzetti C, Jürchott K, et al. Effects of aging on human leukocytes (part I): Immunophenotyping of innate immune cells. Age (Dordr) 2015;37:92.  Back to cited text no. 41
    
42.
Babio N, Ibarrola-Jurado N, Bulló M, Martínez-González MÁ, Wärnberg J, Salaverría I, et al. White blood cell counts as risk markers of developing metabolic syndrome and its components in the PREDIMED study. PLoS One 2013;8:e58354.  Back to cited text no. 42
    
43.
Azekoshi Y, Yasu T, Watanabe S, Tagawa T, Abe S, Yamakawa K, et al. Free fatty acid causes leukocyte activation and resultant endothelial dysfunction through enhanced angiotensin II production in mononuclear and polymorphonuclear cells. Hypertension 2010;56:136-42.  Back to cited text no. 43
[PUBMED]    
44.
Butkevich E, Härtig W, Nikolov M, Erck C, Grosche J, Urlaub H, et al. Phosphorylation of FEZ1 by microtubule affinity regulating kinases regulates its function in presynaptic protein trafficking. Sci Rep 2016;6:26965.  Back to cited text no. 44
    
45.
Li LB, Lei H, Arey RN, Li P, Liu J, Murphy CT, et al. The neuronal Kinesin UNC-104/KIF1A is a key regulator of synaptic aging and insulin signaling-regulated memory. Curr Biol 2016;26:605-15.  Back to cited text no. 45
[PUBMED]    
46.
Tang Y, Li M, Wang J, Pan Y, Wu FX. CytoNCA: A cytoscape plugin for centrality analysis and evaluation of protein interaction networks. Biosystems 2015;127:67-72.  Back to cited text no. 46
[PUBMED]    
47.
Carroll WX, Kalupahana NS, Booker SL, Siriwardhana N, Lemieux M, Saxton AM, et al. Angiotensinogen gene silencing reduces markers of lipid accumulation and inflammation in cultured adipocytes. Front Endocrinol (Lausanne) 2013;4:10.  Back to cited text no. 47
[PUBMED]    
48.
Vollmayer P, Clair T, Goding JW, Sano K, Servos J, Zimmermann H, et al. Hydrolysis of diadenosine polyphosphates by nucleotide pyrophosphatases/phosphodiesterases. Eur J Biochem 2003;270:2971-8.  Back to cited text no. 48
    
49.
Wu Y, Zhou X, Huang X, Xia Q, Chen Z, Zhang X, et al. Pax8 plays a pivotal role in regulation of cardiomyocyte growth and senescence. J Cell Mol Med 2016;20:644-54.  Back to cited text no. 49
[PUBMED]    
50.
Bartram MP, Amendola E, Benzing T, Schermer B, de Vita G, Müller RU, et al. Mice lacking microRNAs in paX8-expressing cells develop hypothyroidism and end-stage renal failure. BMC Mol Biol 2016;17:11.  Back to cited text no. 50
    
51.
Altshuler D, Daly MJ, Lander ES. Genetic mapping in human disease. Science 2008;322:881-8.  Back to cited text no. 51
[PUBMED]    


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3]



 

Top
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed1793    
    Printed123    
    Emailed0    
    PDF Downloaded170    
    Comments [Add]    

Recommend this journal