Pietrovito L; Nguyen NT; Jin YF; Modesti
PA; Lindsey ML; Modesti A

Research Article

Austin J Proteomics Bioinform & Genomics. 2015;2(1): 1007.

Interacting Network Analysis and Functional Profiling to Look Inside Adverse Ventricular Remodeling Post- Myocardial Infarction

Pietrovito L¹*, Nguyen NT^2,3, Jin YF^2,3, Modesti PA⁴, Lindsey ML^2,5,6 and Modesti A¹

¹Dipartimento di Scienze Biomediche, Sperimentali e Cliniche, Università degli Studi di Firenze, Italy

²San Antonio Cardiovascular Proteomics Center, University of Texas Health Science Center at San Antonio, USA

³Department of Electrical and Computer Engineering, University of Texas at San Antonio (UTSA), San Antonio, USA

⁴Dipartimento di Medicina Sperimentale e Clinica, Università degli Studi di Firenze, Italy

⁵Mississippi Center for Heart Research, Department of Physiology and Biophysics, University of Mississippi Medical Center,USA

⁶Research Service, G.V. (Sonny) Montgomery Veterans Affairs Medical Center, USA

*Corresponding author: Pietrovito L, Dipartimento di Scienze Biomediche, Sperimentali e Cliniche,Viale G. Morgagni 50 50134, Università degli Studi di Firenze, Italy

Received: October 01, 2014; Accepted: February 21, 2015; Published: February 23, 2015

Abstract

Dysfunction of the left ventricle occurs to a varying degree in the most of surviving patients’ post-myocardial infarction. In these patients, adverse remodeling frequently culminates in heart failure. Being able to predict patients who will progress to congestive heart failure would greatly advance in clinical prognostic capabilities.

The aim of this study was two-fold: to improve the knowledge on functional pathways of early and late left ventricleremodeling processes, and to generate new hypotheses to identify putative prognostic indicators for heart failure. To this purpose, we carried out a systems biology study using protein lists previously identified by proteomic studies.

Twenty-seven journal articles were included in our analysis. We generated two protein lists:a list of circulating proteins associated with adverse left ventricleremodeling (early changes); anda list of proteins found differentially expressed in ventricle tissue of patients with heart failure (later phase). We separately analyzed the protein sets by a combination of pioneering bioinformatics portals available on web.

We obtained significant enrichments of blood proteins involved in extracellular matrix remodeling, collagen catabolism, response to stress, and the inflammatory response, while the alterations in the left ventricle reflected remarkable activation of the respiratory chain coupled with ATP production and oxidative metabolism. We provided new insights into the pathogenesis of adverse ventricularremodeling and heart failure and we brought to light some intermediate proteins likely involved in the disease mechanisms not previously associated with the failing status, supplying a new rationale for drug development and further discovery of biomarkers of these heart pathologies.

Keywords: Proteomics; Ventricular Remodeling; Heart Failure; Systems Biology; Matrix Metalloproteinase; Respiratory chain

Abbreviations

LV: Left Ventricle; HF: Heart Failure; MI: Myocardial Infarction; LVSD: Left Ventricular Systolic Dysfunction; ICM: Ischemic Cardiomyopathy; ORA: Over-Representation Analysis; BP: Biological Process; MF: Molecular Function; CC: Cellular Component; ECM: Extracellular Matrix; BPLVR: Blood Proteins List from Patients with Adverse Remodeling Post-MI; LVTP: LV Tissue Proteins List From Patients with HF

Introduction

Adverse remodelingof the left ventricle (LV) defines a pathological process during which molecular, biochemical, and cellular changes lead to alterations in shape, dimensions, and function of the LV. This cascade of eventsincludes dilatation, hypertrophy, and the formation of a discrete collagen scar anditcan occur in response to different conditions including myocardial infarction (MI), pressure overload (hypertension), volume overload (valvular heart disease) or cardiomyopathy [1]. Following MI, the remodeling response begins within hours after the ischemic insult and can continue for months, even years, involving both the infarcted and non-infarcted regions of the LV. The remodelingprocess persists until a balance has been reached between the distention forces generated following the dilatation of the ventricle chamberand the traction forces exerted by the collagen scar [2]. If the adaptive response fails, progressive dilatation and fibrosis will provoke extensive worsening of LV function until develop into HF [3].

Although persistent cardiac remodeling is widely accepted to be associated with high risk of HF and death, the early diagnosis of remodeling has only recently been appreciated as a pivotal time for intervention.Deriving new insights into the temporal evolution and biochemical pathways that drive LV remodeling representsa current challenge in efforts toreduce the mortality, morbidity, and the costs of HF.Unfortunately,adverse LVremodeling is a compensatory mechanism that often progresses without obvious changes in the collected clinical variables[4]. Wang and colleagues recently reported that the incidence of asymptomatic left ventricular systolic dysfunction (LVSD) in the community ranges from 3% to 6%, though as common as systolic HF but often occurring without known cardiovascular diseases[5].Currently, first screening of patients with known or suspected myocardial dysfunction is achieved by electrocardiography, cardiac imaging, and blood chemistry. However, each measure screened for cardiac remodelingpresents some limitations and it is indicative of a different aspect of the disease,without providing a comprehensive picture of the pathological state.

Systems biology, through the identification of the biological networks connecting the different molecular elements, may supply new powerful insights into the pathogenesis of complex diseases [6]. In particular, by combining the systems biology approach to a bioinformatics analysis, it is possible to identifythe protein-protein interactionsnetworks and their functional enrichments, thus vastly improving the knowledge on the central biological mechanisms of the disease.

In the current study, we performed an extensive systems biology analysis of human adverse LV remodeling by functional profiling and network analysis to improve the understanding on functional pathways of the early and late LV remodeling process and to provide novel hypotheses on likely prognostic indicators, with the final goal to improve the treatment of adverse cardiac remodeling and HF.Ourstudy was focused on two different systems: blood and LV tissue. We generated two different protein lists based on the type of the examined sample and we started the analysis with the list of blood proteins previously associated with adverse left ventricular remodeling (BPLVR)[7-32]. Subsequently, we uploaded the list of LV proteins found differentially expressed in patients with congestive HF triggered by ischemic cardiomyopathy (ICM) in comparison with healthy tissues[33]. The collected data were analysed using a combination of over-representation analysis (ORA) tools available as Cytoscape apps or as web-based portals (www.bioprofiling.de and bioinfo.vanderbilt.edu/webgestalt/) [34-36].

Methods

Data collection

We performed a Medline search by using individually or in combination the following criteria: “proteomics”, “adverse remodeling”, “heart failure”, “serological biomarkers”, and “human ventricular tissue”.The search was time-restricted to November, 2013.

Concerning the association between circulating proteins and adverse remodeling, we focused on changes that occurred during the first month post-MI (early alterations). We considered only cohort studies or clinical trials carried out with at least 30 patients admitted with acute MI (AMI). The blood proteins considered in our study, their Swiss-Prot Protein database accession numbers, the main features of the twenty-six publications and the relative references [7- 32] considered are displayed in table 1.

Table 1: List of BPLVR and main clinical features of the studies selected from literature.




  
    Protein Name 
    AN^a 
    Time of Sampling ^b 
    Change in Circulating Level Post-MI 
    
    AMI cohort^d 
    Ref.^e 
  
  
    Cnt^c
    Mean age
    Sex
      (male %)
    No. of Patients
  
  
    Adiponectin
    Q15848
    1 d
    "-
    †
    55
    NS
    75
    [7]
  
  
    Apelin 
    Q9ULZ1
    12h-2 d; 6 mo
    "- / -
    #
    59
    77
    100
    [8]
  
  
    Atrial natriuretic peptide (ANP)
    P01160
    12 h, 1, 3 mo
    "+ / + /+
    #
    59
    76
    33
    [9]
  
  
    
     
    2 d; 1, 2 wk; 1 mo
    "+ / + / + / +
    #
    61
    73
    30
    [10]
  
  
    Brain natriuretic peptide (BNP)
    P16860
    12h, 1, 3 mo
    "+ / + /+
    #
    59
    76
    33
    [9]
  
  
    
     
    2 d; 1, 2 wk; 1 mo
    "+ / + / + / +
    #
    61
    73
    30
    [10]
  
  
    
     
    1 mo
    "+
    §
    63
    72
    246
    [11]
  
  
    
     
    1 d
    "+
    †
    63
    80
    97
    [12]
  
  
    
     
    46 h; 3, 6 mo
    "+ / + / =
    †
    59
    77
    100
    [13]
  
  
    C-reactive protein (CRP)
    P02741
    1 d
    "-
    †
    55
    NS
    75
    [7]
  
  
    
     
    1-3 d
    "+
    †
    58
    50
    35
    [14]
  
  
    
     
    2 d; 1 wk; 2 mo; 1 y
    "- / - / = / =
    †
    59
    81
    42
    [15]
  
  
    
     
    3 d; 1 wk; 6 mo
    "+ / + / +
    #
    57 
    63
    75 
    [16] 
  
  
    
     
    12 h; 5 d
    "+ / +
    §
    59
    NS
    30
    [17]
  
  
    Fas ligand 
    Q53ZZ1 
    1 mo
    "+
    §
    63
    72
    246 
    [11] 
  
  
    Ferritin
    P02792
    1-3 d
    "-
    †
    64
    71
    258
    [18]
  
  
    Growth differentiation factor 15 (GDF15)
    Q99988
    1 d
    "+
    †
    63
    80
    97
    [12]
  
  
    
     
    2 d; 1 wk; 2 mo; 1 y
    "- / - / - / -
    †
    59
    81
    42
    [15]
  
  
    Interleukin-1 (IL-1) receptor-like 1, soluble    isoform (ST2)
    Q01638
    46 h; 3, 6 mo
    "+ / + / +
    §
    59
    77
    100
    [19]
  
  
    
     
    1 d; 2 wk; 3 mo
    “+ / = / =
    §
    61
    78
    69
    [20]
  
  
    
     
    1 d-1 mo
    "+
    †
    58
    80
    1239
    [21]
  
  
    Interleukin-10 (IL-10) 
    P22301
    3 d; 1 wk; 6 mo
    "+ / + / +
    #
    57 
    63
    75 
    [16]
  
  
    Interleukin-18 (IL-18) 
    Q14116
     
    "+
    #
    55
    74
    399
    [22]
  
  
    Interleukin-3 (IL-3)
    P08700
    12 h; 5 d
    "+ / +
    §
    59
    NS
    30
    [17]
  
  
    Interleukin-6 (IL-6)
    P05231
    3 d; 1 wk; 6 mo
    "+ / + / +
    #
    57
    63
    75
    [16]
  
  
    Macrophage colony-stimulating factor 1 (CSF-1)
    P09603 
    12 h; 5 d
    "+ / +
    §
    59
    NS
    30
    [17]
  
  
    Macrophage inflammatory protein-1 alpha
      (MIP-1a)
    P10147
    1-7 d; 1 mo
    "+ / +
    †
    58
    50
    35
    [14]
  
  
    Matrix metalloproteinase-1 (MMP-1)
    P03956
    3-7 d; 1, 3 mo; 1 y
    "= / = / = / =
    §
    57
    81
    239
    [23]
  
  
    Matrix metalloproteinase-2 (MMP-2)
    P08253
    46 h; 3, 6 mo
    "= / = / =
    †
    59
    77
    100
    [13]
  
  
    
     
    3-7 d; 1, 3 mo; 1 y
    “-/ +/ +/ +
    §
    57
    81
    239
    [23]
  
  
    
     
    1, 3 d; 1 wk; 1 mo
    "- / + / + / +
    §
    63
    83
    108
    [24]
  
  
    
     
    2, 5 d; 1, 3, 6 mo
    "- / - / - / - / -
    #
    58
    75
    32
    [25]
  
  
    
     
    12 h; 1, 2, 3, 4 d; 6 mo
    "+ / + / + / + / + / +
    †
    63
    73
    91
    [26]
  
  
    
     
    1, 2, 3, 4, 5 d
    "+ / + / + / = / =
    §
    63
    75
    60
    [27]
  
  
    Matrix metalloproteinase-3 (MMP-3)
    P08254
    46 h; 3, 6 mo
    "+ / = / =
    †
    59
    77
    100
    [13]
  
  
    
     
    3-7 d; 1, 3 mo; 1 y
    "- / + / + / +
    §
    57
    81
    239
    [23]



Table 1:  List of BPLVR and main clinical features of the studies selected from literature.

Table 1 of 2: List of BPLVR and main clinical features of the studies selected from literature.




  
    
     
    12 h; 1, 2, 3, 4 d
    "+ / + / + / + / +
    †
    64
    74
    382
    [28]
  
  
    Matrix metalloproteinase-7 (MMP-7)
    P09237
    2, 5 d; 1, 3, 6 mo
    "+ / + / = / = / =
    #
    58
    75
    32
    [25]
  
  
    Matrix metalloproteinase-8 (MMP-8)
    P22894
    3-7 d; 1, 3 mo; 1 y
    "+ / - / - / -
    §
    57
    81
    239
    [23]
  
  
    
     
    2, 5 d; 1, 3, 6 mo
    "+ / = / = / = / =
    #
    58
    75
    32
    [25]
  
  
    Matrix metalloproteinase-9 (MMP-9)
    P14780
    46 h; 3, 6 mo
    "- / - / -
    †
    59
    77
    100
    [13]
  
  
    
     
    3-7 d; 1, 3 mo; 1 y
    "+ / - / + / -
    §
    57
    81
    239
    [23]
  
  
    
     
    1, 3 d; 1 wk; 1 mo
    "+ / - / - / -
    §
    63
    83
    108
    [24]
  
  
    
     
    2, 5 d; 1, 3, 6 mo
    "+ / + / + / + / +
    #
    58
    75
    32
    [25]
  
  
    
     
    12 h, 1, 2, 3, 4 d; 6 mo
    + / = / = / = / = / =
    †
    63
    73
    91
    [26]
  
  
    
     
    1, 2, 3, 4, 5 d
    "+ / = / = / + / =
    §
    63
    75
    60
    [27]
  
  
    Monocyte chemoattractant protein-1 (MCP-1)
    P13500
    2 d; 3, 6 mo
    "+ / + / +
    †
    59
    77
    100
    [13]
  
  
    
     
    1-7 d; 1 mo
    "+ / +
    †
    58
    50
    35
    [14]
  
  
    Myocardial connective tissue growth factor (CTGF)
    P29279
    2 d; 1 wk; 2 mo; 1 y
    "+ / + / + / +
    †
    59
    81
    42
    [15]
  
  
    Pterin-4-alpha-carbinolamine dehydratase
    P61457
    1 d
    "+
    †
    64
    85
    108
    [29]
  
  
    T-cell-specific protein RANTES
    P13501
    1-7 d; 1 mo
    "+ / +
    †
    58
    50
    35
    [14]
  
  
    TGF-beta receptor type-1    (TGFBR1) 
    P36897
    12 h
    "+
    †
    56 
    85
    115 
    [30]
  
  
    Tissue inhibitor of metalloproteinase-1 (TIMP-1)
    P01033
    3-7 d; 1, 3 mo; 1 y
    "+ / - / - / -
    §
    57
    81
    239
    [23]
  
  
    
     
    1, 3 d; 1 wk; 1 mo
    "+ / + / - / -
    §
    63
    83
    108
    [24]
  
  
    
     
    2, 5 d; 1, 3, 6 mo
    "+ / + / + / = / =
    #
    58
    75
    32
    [25]
  
  
    
     
    1, 2, 3, 4, 5 d
    "= / = / = / + / =
    §
    63
    75
    60
    [27]
  
  
    Tissue inhibitor of metalloproteinase-2 (TIMP-2)
    P16035
    3-7 d; 1, 3 mo; 1 y
    "- / + / + / +
    §
    57
    81
    239
    [23]
  
  
    
     
    2, 5 d; 1, 3, 6 mo
    "= / + / + / + / +
    #
    58
    75
    32
    [25]
  
  
    Tissue inhibitor of metalloproteinase-4 (TIMP-4)
    Q99727
    3-7 d; 1, 3 mo; 1 y
    "- / + / + / +
    §
    57
    81
    239
    [23]
  
  
    
     
    2, 5 d; 1, 3, 6 mo
    "= / + / + / + / +
    #
    58
    75
    32
    [25]
  
  
    Tissue plasminogen activator    (t-PA)
    P00750
    46 h; 3, 6 mo
    "+ / = / =
    †
    59
    77
    100
    [13]
  
  
    Transforming growth factor    beta-1 (TGF-ß1) 
    P01137
    12 h
    "+
    †
    56 
    85
    115 
    [30] 
  
  
    Tumor necrosis factor (TNF- a )
    P01375
     
    "+
    #
    55
    74
    399
    [22] 
  
  
    Vasopressin-neurophysin    2-copeptine 
    P01185
    1, 3, 5 d
    "+ / = / =
    §
    66
    80
    980
    [31]
  
  
    
     
    3, 5 d
    "+ / +
    §
    59
    77
    274
    [32]
  
  
    von Willebrand factor (vWf)
    P04275
    46 h; 3, 6 mo
    = / = / =
    †
    59
    77
    100
    [13]
  
  
    ^a AN= Accession Number in Swiss-Prot    Protein database
      ^bTime of blood collection post-MI
      ^c Control subjects age- and sex-matched    recruited into the study: # Healthy subjects, † Patients with preserved LV function post-MI, § Comparison into    the same group of patients post-MI (considering others correlation factors    e.g. anterior AMI vs inferior AMI)
      ^dBaseline characteristics of acute    myocardial infarction patients
      ^e References



Table 1 of 2:  List of BPLVR and main clinical features of the studies selected from literature.

The complete list of the LV tissue proteins (LVTP) is reported in tableS1. To our knowledge, the proteomic profile of LV tissue from patients with HF triggered by MI was reported only in one recently published paper by Roselló-Lletí[33].We selected the LV proteins found differentially expressed in patients with congestive HF triggered by ischemic cardiomyopathy (ICM) in comparison with healthy tissues and we excluded the proteins whose modifications in expression levels were caused by dilated cardiomyopathy (DCM).

Bioinformatics and statistical analysis

TheORA of the two input lists (Tables 1, S1)was carried out using the WebGestalt online tools (https://bioinfo.vanderbilt.edu/ webgestalt/) [34].WebGestalt is a bioinformatics platform designed for functional genomics, proteomics and large-scale genetic studies from which large number or genes/proteins are continuously generated.The program manages supports and uses data from multiple different databases, such as GO, IntAct, Reactome, KEGG, Pathway Commons and WikiPathways. In WebGestalt, following the upload of the input list, the user can choose the type of analysis to perform. The current version of WebGestalt covers eight organisms including human, mouse, yeast and zebrafish, and it is able to analyze the uploaded list for functional enrichments in various biological contexts.

To obtain functional advanced enrichments and to build interacting networks of the submitted protein lists we used the web portal BioProfiling de(www.bioprofiling.de) [35].Bioprofiling. de is an easy tool of analysis for the interpretation of lists of genes/ proteins. The program accepts a directory of Accesion Number (AN) as input and provides a list of genes as output.For this study, we used three different BioProfilingapplications: (i) ProfCom_GO, (ii) PPI spider and (iii) R spider [43]. ProfComGO wasused for functional profiling. The output of ProfCom_GOis a list of enriched “complex classes”. Such classes, in general characterized by a more specific biological function compared to the single GO terms, are constructed by combination of three Boolean operations: intersection, union and difference (“OR”, “AND”, and “NOT”) of available functional terms within GO and FunCat databases and enriched in input protein list.

PPI and R spider are two tools of BioProfiling pipeline able to provideenriched protein maps. of the submitted list. PPI spider is based on protein-protein interactions networks derived from IntAct, while R spider uses a combination of signaling and metabolic pathways from Reactome and KEGG databases. Both programs profile the protein networks according to three different models. In model D1, the networks were built only considering the direct interactions between the input proteins, while in the models D2 and D3 one or two intermediate nodes were added. The significant enrichments provided by BioProfiling were determined with the default parameter settings. The p-value presented, computed by Monte Carlo simulation (https:// www.bioprofiling.de/statistical_frameworks.html), referred to the probability of obtaining a model of the same quality for a random gene list of the same size.

Visualization with cytoscape

The significant enriched networks (p-value < 0.01) were exported as an ‘xgmml’ file and visualized by Cytoscape 3.1.0 (https://www. cytoscape.org/), aopen-source software for the large-scale integration of interaction network data [36].

Results and Discussion

Bioinformatics analysis of the blood proteins list from patients with adverse remodeling post-MI (BPLVR)

At first we analysed theinputlist of circulating proteins (Table1) by using WebGestalt resources, in order to obtain the ORA of proteomics data. Moreover, toincrease the specificity of the functional enrichments and the number of putative proteins associated with the disease, we subsequently uploaded the input liston BioProfiling.de.

GO over-representation analysis

Figure 1 reports the GO Treesrelated to Biological Process (BP, Panel A), Molecular Function (MF, Panel B) and Cellular Component (CC, Panel C)classes obtained after performing the ORA of GO categories by the Webgestalt portal. In BP sub-root are present40 enriched GO categories. As expected,during the first month post- MI there was a significant increase in circulating levels of proteins involved in ‘response to stress’ (GO:0006950 – 25 proteins), ‘response to wounding’ (GO:0009611 - 20 proteins), ‘inflammatory response’ (GO:0006954 – 13 proteins), ‘response to decreased oxygen levels’ (GO:0036293 – 13 proteins) and ‘response to hypoxia’ (GO:0001666 – 12 proteins). The bioinformatics analysis also revealed enrichments of proteins implicated in ‘regulation of cell proliferation’ (GO:0042127 - 19 proteins) and ‘cell migration’ (GO:0016477 – 16 proteins). However, the key enrichments werethose ofproteins engaged in ‘extracellular matrix disassembly’ (GO:0022617 – 8 proteins) and ‘collagen catabolic process’ (GO:0030574 – 6 proteins). In fact, despite the number of proteins belonging to these categories has been relatively small compared to the entire submitted list, the values of the enrichment ratios were extremely high (full reports of enriched GO categories of circulating biomarkers provided by WebGestalt are available in table S2). Concerning the MF sub-root, it included five ontologies involved in protein binding: ‘receptor binding’(GO:0005102 - 21 proteins), ‘cytokine activity’ (GO:0005125 – 12 proteins), ‘cytokine receptor binding’ (GO:0005126 - 11 proteins), ‘growth factor activity’ (GO:0008083 - 7 proteins),and ‘hormone activity’ (GO:0005179 - 5 proteins). Moreover, in accordance with the BP categories, the MF enrichmentswere related to ‘metalloendopeptidase activity’ (GO:0004222 –7 proteins), and ‘endopeptidase activity’ (GO:0004175 – 7 proteins). Finally, the CCpanel revealed that 31 proteins of the input list (more than 90%) localize in ‘extracellular region’ (GO:0005576). More in detail, 13 proteins pinpointed in ‘ECM’ (GO:0031012), 8 proteins in ‘cell surface’ (GO:0009986) and 4 proteins in ‘cytoplasmic membrane-bounded vescicle lumen’ (GO:0060205). The set of blood proteinswas then analyzed with respect to theKEGG, PathwayCommons, and Wikipathways databases (Table 2). The entries provided by KEGG pathways disclosed significant enrichments of proteins common to other pathologies, including rheumatoid arthritis (10 proteins of the input list) and cancer (7 proteins), in addition to a remarkable involvement of proteins related to cytokine-cytokine receptor interaction (12 proteins). The outcomes obtained withPathwayCommonsdatabase reinforced the previous results, underlining a strong activation of intracellular signaling cascadestriggered by cytokines and growth factors, as well as the opening of pathways involved in cell migration (Arf6, focal adhesion kinase and nestin). By using the Wikipathway platform,we obtained a significant enrichment of‘matrix metalloproteinases’ (10 proteins of the input list),‘cytokines and inflammatory response’ (5 proteins) and, interestingly, of‘TWEAK signaling pathway’. TNF related weak inducer of apoptosis (TWEAK) is a small pleiotropic cytokine of the TNF super family. The multiple biological activities of TWEAK include stimulation of cell growth and angiogenesis, induction of inflammatory cytokines, and stimulation of apoptosis.

Table 2: Over-representation of enriched categories in BPLVR list provided by using KEGG(A),PathwayCommons(B)andWikipathways (C)databases. pvalue < 0.0001 after Benjamini-Hochberg correction.




  
    Pathway Name 
    #^a 
    AN^b 
    Statistics^c 
  
  
    (A) KEGG pathways 
  
  
    Rheumatoid arthritis
    10
    P01137 P01375 P08254 P13500    P05231 P09603 P03956 Q14116 P10147 P13501
    C=91; O=10; E=0.07; R=139.39; rawP=1.32e-19; adjP=1.58e-18
  
  
    Cytokine-cytokine receptor interaction
    12
    P01137 P01375 P08700P13500P05231    P09603 Q53ZZ1Q14116P36897P10147P22301P13501
    C=265; O=12; E=0.21; R=57.44; rawP=1.10e-18; adjP=6.60e-18
  
  
    Chagas disease (American    trypanosomiasis)
    9
    P01137 P01375 P13500    P05231 Q53ZZ1P36897P10147P22301P13501
    C=104; O=9; E=0.08; R=109.77; rawP=9.65e-17; adjP=3.86e-16
  
  
    Malaria
    6
    P01137 P01375 Q14116P22301P13500P05231
    C=51; O=6; E=0.04; R=149.23; rawP=2.64e-12; adjP=7.92e-12
  
  
    African trypanosomiasis
    5
    Q53ZZ1P01375 Q14116P22301P05231
    C=35; O=5; E=0.03; R=181.21; rawP=7.15e-11; adjP=1.72e-10
  
  
    NOD-like receptor signaling pathway
    5
    P01375 Q14116P13501P13500P05231
    C=58; O=5; E=0.05; R=109.35; rawP=9.96e-10; adjP=1.99e-09
  
  
    Pathways in cancer
    7
    Q53ZZ1P03956 P14780P01137P36897P05231P08253
    C=326; O=7; E=0.26; R=27.24; rawP=5.97e-09; adjP=1.02e-08
  
  
    Hematopoietic cell lineage
    4
    P01375 P08700P05231P09603
    C=88; O=4; E=0.07; R=57.66; rawP=7.16e-07; adjP=1.07e-06
  
  
    Toll-like receptor signaling pathway
    4
    P01375 P10147P13501P05231
    C=102; O=4; E=0.08; R=49.74; rawP=1.29e-06; adjP=1.72e-06
  
  
    Amoebiasis
    4
    P01137P01375 P22301P05231
    C=106; O=4; E=0.08; R=47.87; rawP=1.51e-06; adjP=1.81e-06
  
  
    Osteoclast differentiation
    4
    P01137P01375 P36897P09603
    C=128; O=4; E=0.10; R=39.64; rawP=3.20e-06; adjP=3.49e-06
  
  
    MAPK signaling pathway
    4
    Q53ZZ1P01137P01375 P36897
    C=268; O=4; E=0.21; R=18.93; rawP=5.84e-05; adjP=5.84e-05
  
  
    (B) PathwaysCommons 
  
  
    ErbB1 downstream signaling
    20
    P08700P01375P01137P14780P13500P01160P03956P36897Q99988P22301P13501P09237P08254P01033P29279P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    PAR1-mediated thrombin signaling    events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1299;    O=20; E=1.02; R=19.53; rawP=3.05e-22; adjP=6.90e-22
  
  
    VEGF and VEGFR signaling network
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1304;    O=20; E=1.03; R=19.45; rawP=3.29e-22; adjP=6.90e-22
  
  
    S1P1 pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    PDGF receptor signaling network
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1293;    O=20; E=1.02; R=19.62; rawP=2.79e-22; adjP=6.90e-22
  
  
    Nectin adhesion pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1295;    O=20; E=1.02; R=19.59; rawP=2.87e-22; adjP=6.90e-22
  
  
    Urokinase-type plasminogen activator    (uPA) and uPAR-mediated signaling
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22



Table 2:  Over-representation of enriched categories in BPLVR list provided by using KEGG(A),PathwayCommons(B)andWikipathways (C)databases. pvalue < 0.0001
after Benjamini-Hochberg correction.

Table 2 of 2:




  
    Signaling events mediated by VEGFR1    and VEGFR2
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1296; O=20;    E=1.02; R=19.57; rawP=2.92e-22; adjP=6.90e-22
  
  
    LKB1 signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1308;    O=20; E=1.03; R=19.40; rawP=3.50e-22; adjP=6.90e-22
  
  
    Alpha9 beta1 integrin signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1305;    O=20; E=1.03; R=19.44; rawP=3.34e-22; adjP=6.90e-22
  
  
    Syndecan-1-mediated signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P0    9237P29279P01033P08 254P09603P05231P 08253Q53ZZ1P02741
    C=1300;    O=20; E=1.02; R=19.51; rawP=3.10e-22; adjP=6.90e-22
  
  
    Class I PI3K signaling events
    20
    P08700P 01375P01137P1 4780P13500P01160P03    956Q99988P36897 P22301P13501P09237 P29279P01033P0825 4P09603P05231P08253Q 53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    Endothelins
    20
    P08700P01375P01137P14 780P13500P0    1160P03956Q99988P36897 P22301P13501P09237    P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1307;    O=20; E=1.03; R=19.41; rawP=3.45e-22; adjP=6.90e-22
  
  
    Internalization of ErbB1
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    IL5-mediated signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1292;    O=20; E=1.02; R=19.64; rawP=2.74e-22; adjP=6.90e-22
  
  
    Glypican 1 network
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1299;    O=20; E=1.02; R=19.53; rawP=3.05e-22; adjP=6.90e-22
  
  
    Thrombin/protease-activated receptor    (PAR) pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1300;    O=20; E=1.02; R=19.51; rawP=3.10e-22; adjP=6.90e-22
  
  
    GMCSF-mediated signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1292;    O=20; E=1.02; R=19.64; rawP=2.74e-22; adjP=6.90e-22
  
  
    Plasma membrane estrogen receptor    signaling
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1301;    O=20; E=1.03; R=19.50; rawP=3.15e-22; adjP=6.90e-22
  
  
    Arf6 signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    Arf6 trafficking events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    PDGFR-beta signaling pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    Insulin Pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    EGFR-dependent Endothelin signaling    events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1289;    O=20; E=1.02; R=19.68; rawP=2.62e-22; adjP=6.90e-22
  
  
    Arf6 downstream pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    IL3-mediated signaling events
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1295;    O=20; E=1.02; R=19.59; rawP=2.87e-22; adjP=6.90e-22
  
  
    Signaling events mediated by    Hepatocyte Growth Factor Receptor (c-Met)
    20
    P08700 P01375 P01137 P14780 P13500 P01160 P03956 Q99988 P36897 P22301P13501 P09237 P29279 P01033 P08254 P09603 P05231 P08253 Q53ZZ1 P02741
    C=1293;    O=20; E=1.02; R=19.62; rawP=2.79e-22; adjP=6.90e-22
  
  
    IFN-gamma pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1296;    O=20; E=1.02; R=19.57; rawP=2.92e-22; adjP=6.90e-22
  
  
    Signaling events mediated by focal    adhesion kinase
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    EGF receptor (ErbB1) signaling pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    IGF1 pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1291;    O=20; E=1.02; R=19.65; rawP=2.70e-22; adjP=6.90e-22
  
  
    Class I PI3K signaling events mediated    by Akt
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    mTOR signaling pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1288;    O=20; E=1.02; R=19.70; rawP=2.58e-22; adjP=6.90e-22
  
  
    Integrin family cell surface    interactions
    21
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P00750P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1378;    O=21; E=1.09; R=19.33; rawP=2.12e-23; adjP=6.90e-22
  
  
    ErbB receptor signaling network
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1312;    O=20; E=1.03; R=19.34; rawP=3.72e-22; adjP=6.92e-22
  
  
    Sphingosine 1-phosphate (S1P) pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1311;    O=20; E=1.03; R=19.35; rawP=3.66e-22; adjP=6.92e-22
  
  
    TRAIL signaling pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1328;    O=20; E=1.05; R=19.10; rawP=4.72e-22; adjP=8.55e-22
  
  
    Glypican pathway
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1338;    O=20; E=1.05; R=18.96; rawP=5.47e-22; adjP=9.64e-22
  
  
    Proteoglycan syndecan-mediated    signaling events
    20
    P08700 P01375 P0    1137 P14780 P13 500 P01160 P03956 Q9998 8 P36897 P22    301P13501 P09237  P29279 P01033 P0 8254 P09603 P05231 P0    8253 Q53ZZ1  P02741
    C=1345;    O=20; E=1.06; R=18.86; rawP=6.06e-22; adjP=1.04e-21
  
  
    Beta1 integrin cell surface    interactions
    20
    P08700P01375P01137P14780P13500P01160P03956Q99988P36897P22301P13501P09237P29279P01033P08254P09603P05231P08253Q53ZZ1P02741
    C=1351;    O=20; E=1.07; R=18.78; rawP=6.62e-22; adjP=1.11e-21
  
  
    AP-1 transcription factor network
    15
    P08700P01375P01137P14780P13500P01160P03956P36897P22301P01033P29279P05231P08253Q53ZZ1P02741
    C=623;    O=15; E=0.49; R=30.54; rawP=3.03e-19; adjP=4.95e-19
  
  
    Integrin-linked kinase signaling
    15
    P08700P01375P01137P14780P13500P01160P03956P36897P22301P01033P29279P05231P08253Q53ZZ1P02741
    C=656;    O=15; E=0.52; R=29.00; rawP=6.55e-19; adjP=1.04e-18



Table 2 of 2:

Table 2 of 3:




  
    CDC42 signaling events
    15
    P08700P01375P01137P14780P13500P01160P03956P36897P22301P01033P29279P05231P08253Q53ZZ1P02741
    C=757; O=15;    E=0.60; R=25.13; rawP=5.50e-18; adjP=8.57e-18
  
  
    Regulation of CDC42 activity
    15
    P08700P01375P01137P14780P13500P01160P03956P36897P22301P01033P29279P05231P08253Q53ZZ1P02741
    C=770;    O=15; E=0.61; R=24.71; rawP=7.07e-18; adjP=1.08e-17
  
  
    Validated transcriptional targets of    AP1 family members Fra1 and Fra2
    8
    P08700P01375P14780P13500P03956P05231P08253Q53ZZ1
    C=136;    O=8; E=0.11; R=74.61; rawP=1.34e-13; adjP=2.00e-13
  
  
    amb2 Integrin signaling
    6
    P01375P14780P00750P29279P05231P08253
    C=41;    O=6; E=0.03; R=185.63; rawP=6.64e-13; adjP=9.67e-13
  
  
    IL12-mediated signaling events
    6
    P01375P01137Q14116P05231Q53ZZ1P10147
    C=113;    O=6; E=0.09; R=67.35; rawP=3.58e-10; adjP=5.10e-10
  
  
    IL27-mediated signaling events
    4
    P01375P01137Q14116P05231
    C=26;    O=4; E=0.02; R=195.14; rawP=4.75e-09; adjP=6.63e-09
  
  
    ALK1 signaling events
    7
    P08700P01375P01137P36897P22301P29279Q53ZZ1
    C=321;    O=7; E=0.25; R=27.66; rawP=5.37e-09; adjP=7.34e-09
  
  
    ALK1 pathway
    7
    P08700P01375P01137P36897P22301P29279Q53ZZ1
    C=324;    O=7; E=0.26; R=27.40; rawP=5.72e-09; adjP=7.66e-09
  
  
    TGF-beta receptor signaling
    6
    P08700P01375P36897P22301P29279Q53ZZ1
    C=305;    O=6; E=0.24; R=24.95; rawP=1.36e-07; adjP=1.72e-07
  
  
    Regulation of nuclear SMAD2/3    signaling
    6
    P08700P01375P36897P22301P29279Q53ZZ1
    C=305;    O=6; E=0.24; R=24.95; rawP=1.36e-07; adjP=1.72e-07
  
  
    Regulation of cytoplasmic and nuclear    SMAD2/3 signaling
    6
    P08700P01375P36897P22301P29279Q53ZZ1
    C=305;    O=6; E=0.24; R=24.95; rawP=1.36e-07; adjP=1.72e-07
  
  
    Peptide ligand-binding receptors
    5
    Q9ULZ1P13500P13501P10147P01185
    C=167;    O=5; E=0.13; R=37.98; rawP=2.08e-07; adjP=2.58e-07
  
  
    IL23-mediated signaling events
    4
    P01375P13500Q14116P05231
    C=66;    O=4; E=0.05; R=76.88; rawP=2.24e-07; adjP=2.73e-07
  
  
    LPA receptor mediated events
    4
    P01375P14780P05231P08253
    C=100;    O=4; E=0.08; R=50.74; rawP=1.20e-06; adjP=1.44e-06
  
  
    Class A/1 (Rhodopsin-like receptors)
    5
    Q9ULZ1P13500P13501P10147P01185
    C=276;    O=5; E=0.22; R=22.98; rawP=2.47e-06; adjP=2.90e-06
  
  
    Signal Transduction
    8
    P01137P14780Q9ULZ1P13500P36897P13501P10147P01185
    C=1231;    O=8; E=0.97; R=8.24; rawP=4.04e-06; adjP=4.67e-06
  
  
    GPCR ligand binding
    5
    Q9ULZ1 P13500 P13501 P10147 P01185
    C=338;    O=5; E=0.27; R=18.76; rawP=6.63e-06; adjP=7.53e-06
  
  
    CXCR4-mediated signaling events
    4
    P08700P01375P14780Q53ZZ1
    C=192;    O=4; E=0.15; R=26.43; rawP=1.59e-05; adjP=1.78e-05
  
  
    Posttranslational regulation of    adherens junction stability and dissassembly
    4
    P14780P09237P08254P08253
    C=231;    O=4; E=0.18; R=21.96; rawP=3.28e-05; adjP=3.60e-05
  
  
    N-cadherin signaling events
    4
    P14780P09237P08254P08253
    C=251;    O=4; E=0.20; R=20.21; rawP=4.53e-05; adjP=4.90e-05
  
  
    E-cadherin signaling in the nascent    adherens junction
    4
    P14780P09237P08254P08253
    C=275;    O=4; E=0.22; R=18.45; rawP=6.45e-05; adjP=6.75e-05
  
  
    Stabilization and expansion of the    E-cadherin adherens junction
    4
    P14780P09237P08254P08253
    C=275;    O=4; E=0.22; R=18.45; rawP=6.45e-05; adjP=6.75e-05
  
  
    E-cadherin signaling events
    4
    P14780P09237P08254P08253
    C=280;    O=4; E=0.22; R=18.12; rawP=6.92e-05; adjP=7.13e-05
  
  
    (C) WikiPathways 
  
  
    Matrix Metalloproteinases
    10
    P01375P14780P16035P09237P08254P01033P22894P08253Q99727P03956
    C=31;    O=10; E=0.02; R=409.17; rawP=9.39e-25; adjP=1.31e-23
  
  
    TWEAK Signaling Pathway
    5
    P01375P14780P13500P05231P13501
    C=56;    O=5; E=0.04; R=113.25; rawP=8.31e-10; adjP=5.82e-09
  
  
    Cytokines and Inflammatory Response
    5
    P01375P01137P09603P05231P22301
    C=67;    O=5; E=0.05; R=94.66; rawP=2.09e-09; adjP=9.75e-09
  
  
    Integrated Pancreatic Cancer Pathway
    6
    P08700P01375P01137P02729Q53ZZ1P36897
    C=181;    O=6; E=0.14; R=42.05; rawP=6.13e-09; adjP=2.15e-08
  
  
    Selenium Pathway
    5
    P01375P13500P05231P00750P02741
    C=108;    O=5; E=0.09; R=58.72; rawP=2.36e-08; adjP=6.61e-08
  
  
    Senescence and Autophagy
    5
    P08700P01137P05231P10147P00750
    C=120;    O=5; E=0.09; R=52.85; rawP=4.00e-08; adjP=9.33e-08
  
  
    Angiogenesis overview
    4
    P14780P16035P08253Q99727
    C=79;    O=4; E=0.06; R=64.22; rawP=4.64e-07; adjP=9.28e-07
  
  
    Oncostatin M Signaling Pathway
    4
    P08254P01033P13500P03956
    C=85;    O=4; E=0.07; R=59.69; rawP=6.23e-07; adjP=1.09e-06
  
  
    MicroRNAs in cardiomyocyte hypertrophy
    4
    P01375P01137P01160P16860
    C=103;    O=4; E=0.08; R=49.26; rawP=1.35e-06; adjP=2.10e-06
  
  
    Toll-like receptor signaling pathway
    4
    P01375P05231P10147P13501
    C=116;    O=4; E=0.09; R=43.74; rawP=2.16e-06; adjP=3.02e-06
  
  
    Adipogenesis
    4
    P01375P01137P05231Q15848
    C=130;    O=4; E=0.10; R=39.03; rawP=3.41e-06; adjP=4.34e-06
  
  
    Regulation of toll-like receptor    signaling pathway
    4
    P01375P05231P10147P13501
    C=154;    O=4; E=0.12; R=32.95; rawP=6.67e-06; adjP=7.78e-06
  
  
    MAPK signaling pathway
    4
    P01375P01137Q53ZZ1P36897
    C=165;    O=4; E=0.13; R=30.75; rawP=8.76e-06; adjP=9.43e-06
  
  
    SIDS Susceptibility Pathways
    4
    P01375P05231P22301P01185
    C=214;    O=4; E=0.17; R=23.71; rawP=2.43e-05; adjP=2.43e-05
  
  
    ^aNumber of submitted    proteins involved in the pathway
      ^bAccession number in Swiss-Prot Protein database
      ^cStatistics column    lists: C=the number of reference genes in the category; O=the number of    proteins in the gene set and also in the category; E= the expected number in    the category; R=ratio of enrichment; rawP= p value from the Fisher exact test; adjP=p value adjusted for multiple testing.



Table 2 of 3:

Figure 1: Over-representation analysis of GO categories in BPLVR. The analysis with Web Gestalt disclosed functional GO enrichments grouped according to Biological Process (A), Molecular Function (B), and Cellular Component (C). GO terms in red are the enriched GO categories, while the black ones are their nonenriched parents. Each node shows the name of the GO category, the number of genes in the category and the adjusted p-value indicating the significance of enrichment.

    
    
    Figure 1:  Over-representation analysis of GO categories in BPLVR. The analysis with Web Gestalt disclosed functional GO enrichments grouped according to
Biological Process (A), Molecular Function (B), and Cellular Component (C). GO terms in red are the enriched GO categories, while the black ones are their nonenriched
parents. Each node shows the name of the GO category, the number of genes in the category and the adjusted p-value indicating the significance of
enrichment.

GO functional enrichments

The submission of the blood protein list to the BioProfiling platform revealed statistically significant functional enrichments by using ProfCom_GO. Table 3 provides a short summary of ProfCom_GO output Table. Categories of degree 0 correspond to single GO term, while categories of degree 1, 2 and 3 are obtained by combining pair, triplet, and quadruplet of GO terms, respectively. The main enrichments concerned proteins involved in ‘extracellular region’, ‘response to hypoxia’, ‘inflammatory response’and ‘protein binding’(see Table S3 for full report).Among the enriched categories of degree 0 reported in Table 3, the GO terms ‘extracellular matrix disassembly’and ‘extracellular region’ identified 8 and 26 proteins of input list (IA), respectively. In the human whole genome (IB), the proteins classified according these GO termswere29 for ‘extracellular matrix disassembly’and 1487 for ‘extracellular region’. By combining two functional categories (degree 1), the number of identified proteins in submitted list remained the same but in the whole genome only 23 proteins belonged to this class. The complex function ‘extracellular matrix disassembly AND extracellular region EXCLUDING extracellular matrix EXCLUDING digestion’supplied by ProfCom (degree 3) was even more specific, since only 15 genes in the human whole genome were classified by this complex function. The 8 proteins of input list related by this function complex were:MMP-1, MMP-2, MMP-3, MMP-7, MMP-8, MMP-9, TIMP-1, and TIMP-2.

Table 3: Functional enriched categories inBPLVR list supplied by ProfCom_GO (short summary).




  
    GO term 
    Description 
    l^A 
    l^B 
    p-value 
  
  
    Enriched categories of degree 0 
  
  
    GO:0022617
    EXTRACELLULAR MATRIX DISASSEMBLY 
    8 (34) 
    29 (17987) 
    3.60e-13
  
  
    GO:0005576
    EXTRACELLULAR REGION 
    27 (34) 
    1487 (17987) 
    1.81e-20
  
  
    GO:0005615 
    EXTRACELLULAR SPACE 
    25 (34) 
    811 (17987) 
    7.12e-24 
  
  
    GO:0001666 
    RESPONSE TO HYPOXIA 
    9 (34) 
    163 (17987) 
    1.84e-08 
  
  
    Enriched categories of degree 1 
  
  
    (GO:0022617    and GO:0005576) 
    (EXTRACELLULAR MATRIX DISASSEMBLY andEXTRACELLULAR    REGION)
    8 (34) 
    23 (17987) 
    5.36e-11 
  
  
    (GO:0005615    and GO:0001666) 
    (EXTRACELLULAR SPACE and RESPONSE TO HYPOXIA) 
    9 (34) 
    42 (17987) 
    6.88e-11 
  
  
    Enriched categories of degree 2 
  
  
    (GO:0022617 and GO:0005576 not GO:0031012)
    ((EXTRACELLULAR MATRIX DISASSEMBLY and EXTRACELLULAR    REGION) not EXTRACELLULAR MATRIX)
    8 (34)
    18 (17987)
    2.07-09
  
  
    (GO:0005615 and GO:0001666 not GO:0030154) 
    ((EXTRACELLULAR SPACE and RESPONSE TO HYPOXIA) not CELL    DIFFERENTIATION)) 
    9 
      (34) 
    37 
      (17987) 
    8.32e-09 
  
  
    Enriched categories of degree 3 
  
  
    (GO:0022617 and GO:0005576    not GO:0031012 not GO:0007586) 
    (((EXTRACELLULAR MATRIX DISASSEMBLY and EXTRACELLULAR    REGION) notEXTRACELLULAR MATRIX) not DIGESTION)
    8 (34)
    15 (17987)
    9.90e-08
  
  
    (GO:0022617 notGO:0031012    and GO:0005576 not GO:0007586) 
    (((EXTRACELLULAR MATRIX DISASSEMBLY not EXTRACELLULAR    MATRIX) and EXTRACELLULAR REGION) not DIGESTION)
    8 (34)
    15 (17987)
    9.90e-08
  
  
    (GO:0022617 notGO:0031012    not GO:0005886 and GO:0030198) 
    (((EXTRACELLULAR MATRIX DISASSEMBLYnot EXTRACELLULAR    MATRIX) not PLASMA MEMBRANE) andEXTRACELLULAR MATRIX ORGANIZATION)
    8 (34)
    16 (17987)
    1.98e-07
  
  
    l^A=The    number of proteins from the input list classified by the GO term (in brackets    the size of the input list is reported);
      l^B=The    total number of genes in the whole genome classified by the GO term (in    brackets the number of genes in the whole genome is given).



Table 3:  Functional enriched categories inBPLVR list supplied by ProfCom_GO (short summary).

Similarly, by combining the GO annotations‘extracellular space AND response to hypoxia’(degree 1), we related9 proteins from input list(CRP, IL-18, MMP-9, BNP, t-PA, MCP-1, TGF-β1, TNF-a, and adiponectin) and 42in the whole genome. In whole human genome, by considering the single GO annotations, the program identifies 811proteins with the term ‘extracellular space’and 163proteins with the annotation ‘response to hypoxia’.

Bioinformatics analysis of the LV tissue proteins list from patients with HF (LVTP)

The proteomic study in cardiac tissue from patients with HFevaluated by Roselló-Lletí[36]was used in this work and the LV protein list found differentially regulated in ICM patients (Table S1) was uploaded into the WebGestalt portal to perform ORA. Then, by using BioProfiling devices, we enriched the input list in terms of a network-based statistical framework. Finally, we again investigated the enriched protein-protein interactions networks provided by R and PPI spiders by WebGestal.

GO over-representation analysis

Figure 2 shows the significant GO enrichments obtained in BPand CC area(Panels A and B, respectively) following the submission of LV protein list into theWebGestalt platform. No statistically significant enrichmentwas providedwith GO MF terms. TheBP enriched classes revealed that the majority of the submitted proteins were involved in ‘cellular respiration’ (GO:0045333 - 11 proteins), electron transport chain and ATP production (‘mitochondrial ATP synthesis coupled electron transport’ - GO:0042775 - 5 proteins), and metabolism (‘acetylCoA metabolic process’ – GO:0006084 – 4 proteins). Theenrichment analysis in CC area confirmed that more than 50% of the submitted proteins localized in the ‘mitochondrion’ (GO: 0005739 –17 proteins). 6 of these were proteins of the ‘mitochondrial matrix’ (GO:0005759), while 11 were associated to the ‘mitochondrial inner membrane’ (GO:0005743). More in details, 4proteins were part of the ‘mitochondrial respiratory chain’ (GO:0005746). The full report of enriched GO categories provided by WebGestaltis available in supplementary information (Table S4).

Figure 2: Over-representation analysis of GO categories in LVTP. The analysis with WebGestalt disclosed functional GO enrichments grouped according to Biological Process (A), and Cellular Component (B). GO terms in red are the enriched GO categories while the black ones are their non-enriched parents. Each node shows the name of the GO category, the number of genes in the category and the adjusted p-value indicating the significance of enrichment.

    
    
    Figure 2:  Over-representation analysis of GO categories in LVTP. The analysis with WebGestalt disclosed functional GO enrichments grouped according to
Biological Process (A), and Cellular Component (B). GO terms in red are the enriched GO categories while the black ones are their non-enriched parents. Each
node shows the name of the GO category, the number of genes in the category and the adjusted p-value indicating the significance of enrichment.

The LV protein set was then explored by KEGG, PathwayCommons and Wikipathways databases. The outcomes supplied by these bioinformatics devices further supported a remarkable involvement of pathways associated to electron transport chain, oxidative phosphorilation, and metabolism, with strong activation of the Krebs cycle (data not shown).

GO functional enrichments

In Bioprofiling, we obtained significant enrichments with ProfCom_GO (Table S5), PPI and R spider tools. The analysis of the networks provided by both spider applications may be deeply helpful to identify new proteins not previously associated with the disease and, accordingly, to develop newhypotheses regarding putative prognostic indicators which may subsequently be validated.

In Figure 3 are illustrated the outcomes by PPI spider (Panel A) and R spider (Panel B). The graphical illustrations of the models were obtained with Cytoscape 3.1. Concerning the analysis with PPI spider, we acquired only model D2 (one intermediate allowed), that involved 11proteins of submitted list, 8 new intermediate proteins, and 36 interacting protein pairs, with a p-value < 0.005. Among theintermediate nodes, the protein that showed the highest number of interactionswas the NADH dehydrogenase (ubiquinone) 1 alpha sub-complex 2 (NDUFA2). The PPI network D2 reinforced the enrichment of proteins associated with the inflammatory response showing as intermediates some proteins involved in immune response[receptor-interacting serine-threonine kinase 2 ([RIPK2], granzyme A [GZMA] and IkappaB kinase complex-associated protein[IKBAP).

Figure 3: Results obtained by BioProfiling analysis following the submission of LVTP. (A), Graphic illustration of PPI spider’s model D2, in which is addedone intermediate node to the protein network. (B), Graphic illustrations of R spider’s model D3, in which are added two intermediate nodes to the protein network. In both network the nodes arising from the input list are shown as white squares, whereas the intermediate nodes are represented by red triangles. The edges represent the interactions among the protein nodes. The yellow circles in panel B indicate the metabolites involved in metabolic and signaling pathways.

    
    
    Figure 3:  Results obtained by BioProfiling analysis following the submission of LVTP. (A), Graphic illustration of PPI spider’s model D2, in which is addedone
intermediate node to the protein network. (B), Graphic illustrations of R spider’s model D3, in which are added two intermediate nodes to the protein network. In both
network the nodes arising from the input list are shown as white squares, whereas the intermediate nodes are represented by red triangles. The edges represent
the interactions among the protein nodes. The yellow circles in panel B indicate the metabolites involved in metabolic and signaling pathways.

Figure 3, Panel Bshows the protein network obtained with respect tothe reference knowledge from Reactome and KEGG databases. Among the models supplied by R spider, we just considered the D3 network, covering 13 proteins from the input list and including 88 interacting protein pairs (p-value < 0.005). The inferred model contained also sixintracellular metabolites, represented in the network by yellow circles, and 21 intermediate nodes. Among the intermediate proteins, we found once againmitochondrial proteins involved in the respiratory chain, in ATP synthesis,and metabolism. Among the proteins provided by the enriched PPI network and missing in the input list, we disclosed specific proteins of the respiratory chain. In addition, the R spider’s network highlighted a considerable participation of members of ATP/ADP mitochondrial carrier subfamily, like the ADP/ATP translocase 1, 2 and 3. The complete list of intermediate nodes engaged in both protein networks (PPI spider and R spider) is reportedin table S6.

GO over-representation analysis of the enriched networks

Following the generation of enriched networks by Bioprofiling, we created two novel protein lists by combining the LV proteinspreviously identifiedin failing hearts with the new intermediates proteinsrecognized by PPI and R spider tools. Accordingly, we obtained one set of 41 protein IDs adding the intermediate nodes identified by the model D2 of PPI spider, and one list of 54 LV proteins inserting the intermediates disclosed by the model D3 of R spider. The new input lists were separately submitted to WebGestalt and the bioinformatics analysis was limited to enriched GO categories. The results obtained were elaborated and summarized in table 4. The full reports of GO categories in enriched PPI and R spider networks are available in supplementary material (Table S7, S8). Analyzing the enriched networks, it was possible to better describe the molecular mechanisms which underlie the failure status. The predominant MFcategory was catalytic activity (27/34 proteins in PPI network and 34/54 proteins in R network). In particular, 11/34 proteins in the PPI network and 21/54 proteins in the R networkwere associated to oxidoreductase activity that involve NADH or NADPH as electron carrier. Moreover, the analysis of protein networks with two intermediate nodes showed asignificant enrichment of the multi-protein complex Iof the respiratory chain proteins.Finally, we demonstrated that the considerable increase of energy production mainly arose from carboxylic acid catabolic processes, and from the citric acid cycle.

Table 4: Over-representation analysis of GO terms in enriched PPI spider interacting network (A) and R spider interacting network (B) supplied by WebGestalt.




  
    GO category ^a
    ID ^b
    # ^c
    adjP ^d
  
  
    
      Enriched Model D2    PPI spider
    
  
  
    Biological Process
  
  
    Cellular respiration
    GO:0045333
    13
    6.08e-15
  
  
    Respiratory electron transport chain
    GO:0022904
    10
    7.47e-12
  
  
    Energy derivation by oxidation of organic compounds
    GO:0015980
    13
    2.61e-11
  
  
    Oxidation-reduction process
    GO:0055114
    15
    4.26e-11
  
  
    Generation of precursor metabolites and energy
    GO:0006091
    14
    4.26e-11
  
  
    Electron transport chain
    GO:0022900
    10
    5.10e-11
  
  
    ATP synthesis coupled electron transport
    GO:0042773
    6
    1.20e-07
  
  
    Mitochondrial ATP synthesis coupled electron    transport
    GO:0042775
    6
    1.20e-07
  
  
    Oxidative phosphorylation
    GO:0006119
    6
    3.11e-07
  
  
    Mitochondrial electron transport, NADH to ubiquinone
    GO:0006120
    5
    1.11e-06
  
  
    Aerobic respiration
    GO:0009060
    5
    1.13e-06
  
  
    Molecular Function
  
  
    Oxidoreductase activity
    GO:0016491
    11
    2.65e-05
  
  
    NADH dehydrogenase activity
    GO:0003954
    4
    2.82e-05
  
  
    Catalytic activity
    GO:0003824
    27
    2.82e-05
  
  
    Oxidoreductase activity, acting on NADH or NADPH
    GO:0016651
    5
    2.82e-05
  
  
    NADH dehydrogenase (ubiquinone) activity
    GO:0008137
    4
    2.82e-05
  
  
    NADH dehydrogenase (quinone) activity
    GO:0050136
    4
    2.82e-05
  
  
    Oxidoreductase activity, acting on NADH or NADPH,    quinine or similar compound as acceptor
    GO:0016655
    4
    7.36e-05
  
  
    Cellular Component
  
  
    Mitochondrial part
    GO:0044429
    17
    1.24e-12
  
  
    Mitochondrion
    GO:0005739
    20
    4.27e-11
  
  
    Mitochondrial inner membrane
    GO:0005743
    12
    7.10e-11
  
  
    Organelle inner membrane
    GO:0019866
    12
    1.37e-10
  
  
    Mitochondrial membrane
    GO:0031966
    12
    2.80e-09
  
  
    Mitochondrial envelope
    GO:0005740
    12
    3.79e-09
  
  
    Mitochondrial membrane part
    GO:0044455
    8
    4.44e-09
  
  
    Organelle envelope
    GO:0031967
    13
    3.16e-08
  
  
    Cytoplasm
    GO:0005737
    35
    3.16e-08
  
  
    Envelope
    GO:0031975
    13
    3.43e-08
  
  
    Respiratory chain
    GO:0070469
    6
    5.09e-08
  
  
    Cytoplasmic part
    GO:0044444
    30
    4.01e-07
  
  
    Mitochondrial respiratory chain
    GO:0005746
    5
    1.23e-06
  
  
    Mitochondrial matrix
    GO:0005759
    7
    5.64e-06
  
  
    Mitochondrial respiratory chain complex I
    GO:0005747
    4
    7.53e-06
  
  
    NADH dehydrogenase complex
    GO:0030964
    4
    7.53e-06
  
  
    Respiratory chain complex I
    GO:0045271
    4
    7.53e-06
  
  
    Intracellular part
    GO:0044424
    36
    1.77e-05
  
  
    Intracellular
    GO:0005622
    36
    4.34e-05
  
  
    
      Enriched Model D3 R spider
    
  
  
    Biological Process
  
  
    Cellular respiration
    GO:0045333
    21
    2.78e-27
  
  
    Oxidation-reduction process
    GO:0055114
    29
    2.78e-27
  
  
    Energy derivation by oxidation of organic compounds
    GO:0015980
    24
    2.39e-27
  
  
    Generation of precursor metabolites and energy
    GO:0006091
    25
    7.54e-24
  
  
    Respiratory electron transport chain
    GO:0022904
    16
    2.70e-21
  
  
    Electron transport chain
    GO:0022900
    16
    3.25e-19
  
  
    Small molecule metabolic process
    GO:0044281
    36
    5.92e-17
  
  
    Aerobic respiration
    GO:0009060
    9
    3.50e-13
  
  
    Tricarboxylic acid cycle
    GO:0006099
    7
    3.93e-11
  
  
    Acetyl-CoA catabolic process
    GO:0046356
    7
    4.75e-11
  
  
    Acetyl-CoA metabolic process
    GO:0006084
    8
    9.69e-11
  
  
    Coenzyme catabolic process
    GO:0009109
    7
    1.88e-10
  
  
    ATP synthesis coupled electron transport
    GO:0042773
    8
    1.91e-10
  
  
    Mitochondrial ATP synthesis coupled electron    transport
    GO:0042775
    8
    1.91e-10
  
  
    Thioester metabolic process
    GO:0035383
    9
    2.12e-10
  
  
    Acyl-CoA metabolic process
    GO:0006637
    9
    2.12e-10
  
  
    Cofactor catabolic process
    GO:0051187
    7
    4.70e-10
  
  
    Oxidative phosphorylation
    GO:0006119
    8
    6.31e-10
  
  
    Mitochondrial electron transport, NADH to ubiquinone
    GO:0006120
    6
    5.73e-08
  
  
    Coenzyme metabolic process
    GO:0006732
    9
    2.99e-07
  
  
    Carboxylic acid metabolic process
    GO:0019752
    15
    4.69e-07
  
  
    Small molecule catabolic process
    GO:0044282
    9
    1.23e-06
  
  
    Single-organism catabolic process
    GO:0044712
    9
    1.23e-06
  
  
    Cofactor metabolic process
    GO:0051186
    9
    1.23e-06
  
  
    Carboxylic acid catabolic process
    GO:0046395
    8
    1.86e-06
  
  
    Organic acid catabolic process
    GO:0016054
    8
    1.86e-06
  
  
    Oxoacid metabolic process
    GO:0043436
    15
    1.86e-06
  
  
    Dicarboxylic acid metabolic process
    GO:0043648
    6
    2.03e-06
  
  
    Organic acid metabolic process
    GO:0006082
    15
    2.05e-06
  
  
    Cellular metabolic process
    GO:0044237
    44
    3.32e-06
  
  
    Fatty acid beta-oxidation
    GO:0006635
    5
    7.91e-06
  
  
    Metabolic process
    GO:0008152
    45
    1.93e-05
  
  
    Fatty acid catabolic process
    GO:0009062
    5
    2.93e-05
  
  
    Single-organism metabolic process
    GO:0044710
    44
    4.17e-05
  
  
    Lipid modification
    GO:0030258
    6
    4.61e-05
  
  
    Monocarboxylic acid catabolic process
    GO:0072329
    5
    6.10e-05
  
  
    Fatty acid oxidation
    GO:0019395
    5
    6.10e-05
  
  
    Cellular catabolic process
    GO:0044248
    17
    6.74e-05
  
  
    Lipid oxidation
    GO:0034440
    5
    6.79e-05
  
  
    Phosphorus metabolic process
    GO:0006793
    21
    7.59e-05
  
  
    Molecular Function
  
  
    Oxidoreductase activity
    GO:0016491
    21
    4.84e-14
  
  
    Coenzyme binding
    GO:0050662
    11
    3.40e-10
  
  
    Cofactor binding
    GO:0048037
    11
    8.98e-09
  
  
    Flavin adenine dinucleotide binding
    GO:0050660
    6
    1.42e-06
  
  
    NADH dehydrogenase (ubiquinone) activity
    GO:0008137
    5
    2.09e-06
  
  
    NADH dehydrogenase (quinone) activity
    GO:0050136
    5
    2.09e-06
  
  
    NADH dehydrogenase activity
    GO:0003954
    5
    2.09e-06
  
  
    NAD binding
    GO:0051287
    5
    3.12e-06
  
  
    Oxidoreductase activity, acting on NADH or NADPH
    GO:0016651
    6
    3.12e-06
  
  
    Catalytic activity
    GO:0003824
    34
    5.45e-06
  
  
    Oxidoreductase activity, acting on NADH or NADPH,    quinine or similar compound as acceptor
    GO:0016655
    5
    5.45e-06
  
  
    Electron carrier activity
    GO:0009055
    6
    4.68e-05
  
  
    Hydrogen ion transmembrane transporter
    GO:0015078
    5
    7.19e-05
  
  
    Cellular Component
  
  
    Mitochondrial part
    GO:0044429
    33
    5.82e-32
  
  
    Mitochondrial inner membrane
    GO:0005743
    23
    6.00e-25
  
  
    Mitochondrion
    GO:0005739
    35
    6.00e-25
  
  
    Organelle inner membrane
    GO:0019866
    23
    2.95e-24
  
  
    Mitochondrial membrane
    GO:0031966
    24
    4.60e-23
  
  
    Mitochondrial envelope
    GO:0005740
    24
    1.05e-22
  
  
    Organelle envelope
    GO:0031967
    24
    2.67e-18
  
  
    Envelope
    GO:0031975
    24
    3.35e-18
  
  
    Mitochondrial matrix
    GO:0005759
    16
    1.93e-16
  
  
    Mitochondrial membrane part
    GO:0044455
    12
    3.89e-14
  
  
    Cytoplasmic part
    GO:0044444
    45
    1.16e-13
  
  
    Cytoplasm
    GO:0005737
    49
    2.07e-13
  
  
    Respiratory chain
    GO:0070469
    9
    3.03e-12
  
  
    Mitochondrial respiratory chain
    GO:0005746
    8
    6.95e-11
  
  
    Organelle membrane
    GO:0031090
    26
    3.67e-10
  
  
    Intracellular organelle part
    GO:0044446
    40
    5.52e-09
  
  
    Organelle part
    GO:0044422
    40
    8.11e-09
  
  
    Intracellular part
    GO:0044424
    49
    2.44e-07
  
  
    Mitochondrial respiratory chain complex I
    GO:0005747
    5
    4.93e-07
  
  
    NADH dehydrogenase complex
    GO:0030964
    5
    4.93e-07
  
  
    Respiratory chain complex I
    GO:0045271
    5
    4.93e-07
  
  
    Intracellular
    GO:0005622
    49
    7.28e-07
  
  
    Membrane-bounded organelle
    GO:0043227
    43
    4.57e-06
  
  
    Intracellular membrane-bounded
    GO:0043231
    43
    4.57e-06
  
  
    Intracellular organelle
    GO:0043229
    44
    3.39e-05
  
  
    Organelle
    GO:0043226
    44
    3.42e-05
  
  
    ^a Name of the enriched GO category
      ^b GO category ID
      ^c Number of input protein in the category
      ^d Adjusted p value of enrichment



Table 4:  Over-representation analysis of GO terms in enriched PPI spider interacting network (A) and R spider interacting network (B) supplied by WebGestalt.

Conclusions

The main purpose of this study was to use a systems biology approach for proposing novel hypothesis on mechanisms of LV remodeling and HF. The most significant findings of this study were: (i) blood analysis of early remodeling disclosed enrichments for ECM, stress response, response to hypoxia and inflammatory response,(ii) LV samples from end stage HF showed enrichments for respiratory chain, ATP production, and lipid metabolism, with deeply activation of the citric acid cycle.

As many authors reported, within a few hours following MI the release of pro-inflammatory cytokines stimulates the chemotaxis of the neutrophils into the injured zone.Subsequently the neutrophils draw the macrophages,which phagocyte the necrotic cardiomyocytes[37]. Moreover, during the inflammatory response to MI, a strong increase of the synthesis and the secretion of both matrix metalloproteinases and their tissue inhibitorsoccur. The net balance between these two systems coordinates the wound healing by helping the disposal of necrotic myocytes and the activation of myofibroblasts to initiate new ECM deposition to form the infarct scar [38].

The progression of LV remodeling into HF is the result of the loss of myocytes and maladaptive changes in the surviving myocytes and ECM. In our systems biology analysis, we highlighted an enrichment of ECM, cytokines and the inflammatory response. These results are in good accordance with the post-MI knowledge map recently published by Nguyen et al. [39]. Pro-inflammatory cytokines are expressed in all myocardial cells and have structural and functional consequences on the post-MI myocardium. Inflammatory mediators alone can induce several components in the HF pathway, including LV dysfunction, pulmonary edema, cardiomyopathy, endothelial dysfunction, cachexia/anorexia, adenylatecyclase receptor uncoupling, activation of the fetal gene program, and cardiac myocyte apoptosis [40]. In addition, alterations in the cardiac metabolism are key factors in the transition to HF. By analyzing the ORA of the enriched networks from PPI and R spider, we found an elevated involvement of protein interactions and metabolic pathways linked to an alteration of the mitochondrial function.Gupteand coworkerssuggested a global reduction in substrate oxidation,by finding reduced Krebs cycle intermediates[41]. These results were associated with decreased transcript levels for enzymes that catalyze fatty acid oxidation and pyruvate metabolism. Moreover,Karamanlidisand colleagues found out that mitochondrial respiratory dysfunction is linked to the pathogenesis of HF, predisposing the heart to injury by redox-sensitive mechanisms[42].

Systems biology is an emerging field of biomedical research that aims to understand the properties of a biological system by investigating the interrelationships and the interactions among its molecular components, such as genes, proteins and metabolites [44].The central point of this new research approach is identify the maps of such interactions using systematic and standardized approaches and assays that are unbiased as possible [45]. To date, the basic ideas of systems and network biology are already experimentally tested and applied to relevant biological problems, remarkably to multi-factorial diseases with complex etiologies. In the field of cardiovascular disease,Diez and colleagues have recently published a combined gene association and correlation network of atherosclerosis by performing systems biology analysis on data collected from47microarray [46].Isserlinand coworkers, by proteomic profiling and enriched maps, granted a more comprehensiveunderstanding of the progressive alterations associated with functional decline in dilatedcardiomyopathy in mouse model[47]. Finally, an unbiased systems approach was published to define energy metabolic events that occur in mouse model during the pathological cardiac remodeling and early stage of HF [48].

To date, adverse LV remodeling following MI still remains difficult to distinguish from the normal wound healing repair, and HF is often discovered late during disease progression at a time when it is difficult to intervene. Based on this evaluation, future studies should evaluate whether early changes in ECM can predict later changes in cardiac metabolism. One limitation of this study was the inability to compare the early changes in the blood to late changes in the LV, due to differences in sampling types. Future studies will be needed to show and compare the two compartments alterations, in order to identify blood biomarkers that reflect tissue status along the time continuum of disease.

Acknowledgement

This work was partially supported by grants from the Ministerodell’Universita’ e dellaRicercaScientifica e Tecnologica [Grant No. 2003063257-006 to P. A. Modesti], from the University of Florence [Grant No. 239, 2001 to P.A. Modesti], from the San Antonio Cardiovascular Proteomics Center funded from NHLBI [HHSN 268201000036C (N01-HV-00244), and NIH R01 HL075360] to M.L. Lindsey.

References

Download PDF

Citation: Pietrovito L, Nguyen NT, Jin YF, Modesti PA, Lindsey ML and Modesti A. Interacting Network Analysis and Functional Profiling to Look Inside Adverse Ventricular Remodeling Post-Myocardial Infarction. Austin J Proteomics Bioinform & Genomics. 2015;2(1): 1007. ISSN : 2471-0423

Instruction for Authors

Submit Your Article

Protein Name	AN^a	Time of Sampling ^b	Change in Circulating Level Post-MI		AMI cohort^d			Ref.^e
Protein Name	AN^a	Time of Sampling ^b	Change in Circulating Level Post-MI	Cnt^c	Mean age	Sex (male %)	No. of Patients	Ref.^e
Adiponectin	Q15848	1 d	"-	†	55	NS	75	[7]
Apelin	Q9ULZ1	12h-2 d; 6 mo	"- / -	#	59	77	100	[8]
Atrial natriuretic peptide (ANP)	P01160	12 h, 1, 3 mo	"+ / + /+	#	59	76	33	[9]
		2 d; 1, 2 wk; 1 mo	"+ / + / + / +	#	61	73	30	[10]
Brain natriuretic peptide (BNP)	P16860	12h, 1, 3 mo	"+ / + /+	#	59	76	33	[9]
		2 d; 1, 2 wk; 1 mo	"+ / + / + / +	#	61	73	30	[10]
		1 mo	"+	§	63	72	246	[11]
		1 d	"+	†	63	80	97	[12]
		46 h; 3, 6 mo	"+ / + / =	†	59	77	100	[13]
C-reactive protein (CRP)	P02741	1 d	"-	†	55	NS	75	[7]
		1-3 d	"+	†	58	50	35	[14]
		2 d; 1 wk; 2 mo; 1 y	"- / - / = / =	†	59	81	42	[15]
		3 d; 1 wk; 6 mo	"+ / + / +	#	57	63	75	[16]
		12 h; 5 d	"+ / +	§	59	NS	30	[17]
Fas ligand	Q53ZZ1	1 mo	"+	§	63	72	246	[11]
Ferritin	P02792	1-3 d	"-	†	64	71	258	[18]
Growth differentiation factor 15 (GDF15)	Q99988	1 d	"+	†	63	80	97	[12]
		2 d; 1 wk; 2 mo; 1 y	"- / - / - / -	†	59	81	42	[15]
Interleukin-1 (IL-1) receptor-like 1, soluble isoform (ST2)	Q01638	46 h; 3, 6 mo	"+ / + / +	§	59	77	100	[19]
		1 d; 2 wk; 3 mo	“+ / = / =	§	61	78	69	[20]
		1 d-1 mo	"+	†	58	80	1239	[21]
Interleukin-10 (IL-10)	P22301	3 d; 1 wk; 6 mo	"+ / + / +	#	57	63	75	[16]
Interleukin-18 (IL-18)	Q14116		"+	#	55	74	399	[22]
Interleukin-3 (IL-3)	P08700	12 h; 5 d	"+ / +	§	59	NS	30	[17]
Interleukin-6 (IL-6)	P05231	3 d; 1 wk; 6 mo	"+ / + / +	#	57	63	75	[16]
Macrophage colony-stimulating factor 1 (CSF-1)	P09603	12 h; 5 d	"+ / +	§	59	NS	30	[17]
Macrophage inflammatory protein-1 alpha (MIP-1a)	P10147	1-7 d; 1 mo	"+ / +	†	58	50	35	[14]
Matrix metalloproteinase-1 (MMP-1)	P03956	3-7 d; 1, 3 mo; 1 y	"= / = / = / =	§	57	81	239	[23]
Matrix metalloproteinase-2 (MMP-2)	P08253	46 h; 3, 6 mo	"= / = / =	†	59	77	100	[13]
		3-7 d; 1, 3 mo; 1 y	“-/ +/ +/ +	§	57	81	239	[23]
		1, 3 d; 1 wk; 1 mo	"- / + / + / +	§	63	83	108	[24]
		2, 5 d; 1, 3, 6 mo	"- / - / - / - / -	#	58	75	32	[25]
		12 h; 1, 2, 3, 4 d; 6 mo	"+ / + / + / + / + / +	†	63	73	91	[26]
		1, 2, 3, 4, 5 d	"+ / + / + / = / =	§	63	75	60	[27]
Matrix metalloproteinase-3 (MMP-3)	P08254	46 h; 3, 6 mo	"+ / = / =	†	59	77	100	[13]
		3-7 d; 1, 3 mo; 1 y	"- / + / + / +	§	57	81	239	[23]