An explainable machine learning-driven proposal of pulmonary fibrosis biomarkers

Fanidis, Dionysiοs; Pezoulas, Vasileios C.; Fotiadis, Dimitrios I.; Aidinis, Vassilis

doi:10.1016/j.csbj.2023.03.043

Cited by 1 publication

(1 citation statement)

References 59 publications

(82 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this context, we have recently developed Fibromine, a database and data mining tool, hosting all publicly available IPF transcriptomic (and proteomic) datasets ( 5 ), thus allowing the further exploitation of legacy data. Comparative analysis selected several hundred genes as differentially expressed in IPF, while an explainable machine learning phenotype classification algorithm prioritized 76 genes that include previously identified IPF expression hallmarks (e.g., Col1a1), IPF biomarkers (e.g., MMP7), as well as many genes previously shown to be involved in the pathophysiology of IPF (e.g., SPP1) ( 6 ). Among the novel, commonly identified deregulated genes in IPF was Lipocalin-2 (LCN2), also known as neutrophil gelatinase B-associated lipocalin (NGAL), as it was initially identified in neutrophilic granules in association with matrix metalloproteinase 9 (MMP9; gelatinase B) ( 7 , 8 ).…”

Section: Introductionmentioning

confidence: 99%

Increased lipocalin-2 expression in pulmonary inflammation and fibrosis

Galaris,

Fanidis,

Tsitoura

et al. 2023

Front. Med.

Self Cite

View full text Add to dashboard Cite

IntroductionIdiopathic Pulmonary Fibrosis (IPF) is a chronic, progressive interstitial lung disease with dismal prognosis. The underlying pathogenic mechanisms are poorly understood, resulting in a lack of effective treatments. However, recurrent epithelial damage is considered critical for disease initiation and perpetuation, via the secretion of soluble factors that amplify inflammation and lead to fibroblast activation and exuberant deposition of ECM components. Lipocalin-2 (LCN2) is a neutrophil gelatinase-associated lipocalin (NGAL) that has been suggested as a biomarker of kidney damage. LCN2 has been reported to modulate innate immunity, including the recruitment of neutrophils, and to protect against bacterial infections by sequestering iron.MethodsIn silico analysis of publicly available transcriptomic datasets; ELISAs on human IPF patients' bronchoalveolar lavage fluids (BALFs); bleomycin (BLM)-induced pulmonary inflammation and fibrosis and LPS-induced acute lung injury (ALI) in mice: pulmonary function tests, histology, Q-RT-PCR, western blot, and FACS analysis.Results and discussionIncreased LCN2 mRNA expression was detected in the lung tissue of IPF patients negatively correlating with respiratory functions, as also shown for BALF LCN2 protein levels in a cohort of IPF patients. Increased Lcn2 expression was also detected upon BLM-induced pulmonary inflammation and fibrosis, especially at the acute phase correlating with neutrophilic infiltration, as well as upon LPS-induced ALI, an animal model characterized by neutrophilic infiltration. Surprisingly, and non withstanding the limitations of the study and the observed trends, Lcn2−/− mice were found to still develop BLM- or LPS-induced pulmonary inflammation and fibrosis, thus questioning a major pathogenic role for Lcn2 in mice. However, LCN2 qualifies as a surrogate biomarker of pulmonary inflammation and a possible indicator of compromised pulmonary functions, urging for larger studies.

show abstract