Page 137 - Read Online

Broadwin et al. Vessel Plus 2023;7:25 https://dx.doi.org/10.20517/2574-1209.2023.103 Page 5 of 14

Pathway analysis was performed using GeneGo Metacore (Clarivate, PA, USA). Metacore is a large curated
database of known factor-to-factor interactions and allows designing networks based on these known
“interactions”. The “direct interactions” algorithm thus builds a graphical network of factors from a list
(without adding intermediaries) connected by interactive lines between them, each depicting a published
interaction between the two factors.

Pathway mapping in Metacore relies on pre-designed graphical maps of specific pathways which form the
pathway database. The algorithm determines what genes (factors) in the source list also occur in the
graphically pre-designed existing pathway maps and selects the top pathways enriched.

RESULTS
After initial analysis, 441 loci had a nominal P value < 0.05 when examining the effect of ischemia vs. normal
heart tissue on a normal diet in the absence of treatment (i.e., saline control perfused segment vs. saline
control ischemic segment). When enrichment of these loci was performed using Metacore, the alterations in
both molecular pathways and molecular functions were elucidated [Figure 1A and Figure 2A]. Specifically,
alterations involved the molecular functions with terms “protein kinase binding”, “kinase binding”, “sterol
response element binding”, and “proline dehydrogenase activity”. GoProcess annotation revealed that most
of these effects were concentrated on processes related to “regulation of lipid metabolism”, “insulin-
dependent stimulation of SREBP-1 in type 2 diabetes in liver”, and “role of ER stress in obesity and type 2
diabetes”. Top pathway names and related P values are shown in Figure 2A. After Bonferroni correction, no
P values were < 0.05.

426 loci at P value threshold < 0.05 were identified when comparing the ischemic vs. normal tissue from
high-fat diet animals. After undergoing enrichment analysis, the most significant change among molecular
functions was associated with terms “binding”, “protein binding”, “enzyme activator activity” and “C-8
sterol isomerase activity” [Figure 1B]. After GoProcess annotation, the pathways identified with alteration
in methylation were regulation of lipid metabolism, development role of IL-8 in angiogenesis, inhibition of
ephrin receptors, and transcription sirtuin6 regulation and functions [Figure 2B]. Similar to the effect of
ischemia in normal diet, after Bonferroni correction, no P values were < 0.05.

While the numbers of DMLs in the high-fat background and normal background were approximately the
same, network analysis in Metacore [Figure 3] demonstrates a substantially more interacting network in the
high-fat than in the normal diet.

When examining the effect of EV treatment on ischemic tissue under normal diet conditions, there were
574 loci with nominal P values < 0.05, with two loci FN3KRP (P < 0.001) and SNTG1 (P = 0.03) significant
after Bonferroni correction. Enrichment analysis showed the molecular functions of “binding”, “protein
binding”, “kinase binding”, and “ion binding” involved [Figure 1C]. Enrichment of molecular pathways
revealed deregulation of “PSD-95-dependent signaling in Huntington’s disease”, “signal transduction
ERK1/2 signaling pathway”, and “LKB1 signaling pathway in lung cancer cells” [Figure 2C].

When examining the effect of EV treatment in ischemic tissue in HFD, there were 511 loci with nominal
P values < 0.05. After Bonferroni correction, two loci had P values < 0.05, BTC (P = 0.008) and PCSK7
(P = 0.01). After enrichment, most effects on molecular function were related to type “2 angiotensin
receptor binding”, “protein binding”, “type 1 angiotensin receptor binding”, and “alpha1-adrenergic
receptor activity” [Figure 1D]. After pathway enrichment, changes appeared to be most concentrated on
pathways involving “chemotaxis”, “lysophosphatidic acid signaling via GPCRs”, “protein folding and