Interrogating ARDS Risk and Severity Utilizing Genomic and Systems Biology Approaches
Item Type text; Electronic Dissertation
Authors Lynn, Heather D.
Citation Lynn, Heather D. (2020). Interrogating ARDS Risk and Severity Utilizing Genomic and Systems Biology Approaches (Doctoral dissertation, University of Arizona, Tucson, USA).
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 07/10/2021 14:43:39
Link to Item http://hdl.handle.net/10150/650808 INTERROGATING ARDS RISK AND SEVERITY UTILIZING
GENOMIC AND SYSTEMS BIOLOGY APPROACHES
by Heather Lynn
______Copyright © Heather Lynn 2020
A Dissertation Submitted to the Faculty of the GRADUATE INTERDISCIPLINARY PROGRAM IN PHYSIOLOGICAL SCIENCES
In Partial Fulfillment of the Requirements For the Degree of DOCTOR OF PHILOSOPHY
In the Graduate College THE UNIVERSITY OF ARIZONA
2020
Acknowledgements & Dedications
Many people’s efforts and time besides my own went into the creation and refinement of this work. I would first like to acknowledge my mentor, Dr. Garcia, without whose support none of this would be possible. Secondly, I want to acknowledge my committee members (Drs Coletta, Karnes, Ledford, and Sun) that have contributed to both the scientific aspects of this thesis and to the personal task of mentoring and teaching me their own areas of expertise they brought to this project. Third, I would like to thank the support of the many professors and staff of the Physiological Sciences GIDP for financial, logistical, and intellectual support during graduate school. Finally, I want to thank my family and friends (first among them being my husband, Charlie) for their patience while I was navigating the work of graduate school and the final task of completing this thesis.
3 | Lynn Table of Contents
List of Figures and Illustrations ...... Page 9
List of Tables ...... Page 11
Abstract ...... Page 12
Description of Dissertation Format ...... Page 15
Chapter 1: Introduction ...... Page 16
ARDS is a Critical Care Illness ...... Page 16
Endothelial Barrier Dysfunction in ARDS ...... Page 20
Biomarkers as a Clinically Useful Sub-Phenotyping Tool in ARDS ... Page 21
Genomics in ARDS ...... Page 24
Epigenetics in ARDS ...... Page 28
MYLK and NAMPT: Two Candidate Risk Genes in ARDS ...... Page 30
Summary and the Promise of System Biology...... Page 32
Hypothesis ...... Page 33
Chapter 2: Aim 1--Genomic and Genetic Approaches to Deciphering Acute
Respiratory Distress Syndrome Risk and Mortality ...... Page 35
Significance ...... Page 35
Recent Advances and ARDS in the Clinical and Genetic Literature .. Page 37
Potential ARDS Candidate Genes Identified by Dysregulated Gene Expression
...... Page 41
PBMC Gene Expression in ARDS as Predictors of Mortality ...... Page 43
Pathway Analysis ...... Page 46
4 | Lynn Reactive Oxygen Species (ROS) Pathways ...... Page 50
Immune-linked and Inflammation-linked Pathways ...... Page 53
Endothelial Vascular and Cellular Signaling Pathways ...... Page 57
Other Transcription Factor and Signaling Pathways ...... Page 61
Critical Issues ...... Page 61
ARDS Genetic Variants Identified by Genome-Wide
Association Studies ...... Page 62
GWAS Studies of ARDS Mortality ...... Page 65
ARDS Genetic Variants/Genes Identified by Candidate
Gene Studies ...... Page 67
Candidate Gene Studies of ARDS Mortality ...... Page 68
ARDS risk SNPs in African Americans ...... Page 71
Future Directions ...... Page 74
Chapter 3: Aim 2—NAMPT Haplotypes and Plasma eNAMPT levels Predict ARDS Mortality Risk ...... Page 76
Significance ...... Page 76
Methods ...... Page 78
Demographics of the ARDS Genotyping and Biomarker Cohort
...... Page 78
Patient Sample Collection ...... Page 79
Detection of Secreted NAMPT ...... Page 79
NAMPT Genotyping ...... Page 80
Endothelial Cell Culture ...... Page 80
5 | Lynn NAMPT Promoter Activity ...... Page 81
Statistical Analysis ...... Page 81
Results ...... Page 83
NAMPT Promoter Genotypes and eNAMPT Plasma Levels in
ARDS Cohorts ...... Page 83
The NAMPT Promoter SNP rs61330082 (-1535) is
Significantly Associated with ARDS Mortality Risk ...... Page 86
NAMPT Promoter SNP rs61330082 (-1535) Alters LPS-
and Mechanical Stress-induced NAMPT Promoter Activity .... Page 88
eNAMPT Levels are Associated with APACHE II Scores
and rs61330082 Genotypes ...... Page 89
Optimized ARDS Mortality Prediction Model: eNAMPT Plasma
Values and a rs61330082/rs59744560-based Haplotype Risk Score
...... Page 91
Discussion ...... Page 94
Future Directions ...... Page 98
NAMPT as a Biomarker in Other Pulmonary Diseases ...... Page 98
ARDS is an Endotype for COVID-19 ...... Page 102
Chapter 4: Aim 3—ARDS & Methylation...... Page 103
Significance ...... Page 103
Methods ...... Page 105
Patient Demographics and Sample Collections ...... Page 105
Genome-wide Methylation Profiling ChIP ...... Page 106
6 | Lynn mRNA Extraction and Genome-wide Transcription Profiling ... Page 107
Pathway Analysis ...... Page 108
ARDS Combined Lung Injury Preclinical Rat Model ...... Page 108
ARDS Combo Model...... Page 108
qPCR Assay ...... Page 109
Results ...... Page 109
A Unique Methylation Profile in PBMCs is Present for Severe ARDS
Outcomes ...... Page 109
c-MET Pathway is Associated with Methylation Enrichment in ARDS
...... Page 113
GSEA Analysis Reveals an Interconnected Network of Receptor
Tyrosine Kinase Signaling Genes and Transcription Factors . Page 114
Global Genome Methylation is Increased in Severe ARDS Outcomes
in a Validation ARDS Cohort ...... Page 116
Methylation in CpG Islands in Key Transcription Factors in ARDS
Show Altered Levels of Methylation in ARDS/VILI ...... Page 118
Discussion ...... Page 120
Future Directions ...... Page 125
Chapter 5: Discussion ...... Page 127
Conclusion ...... Page 127
Summary of Results ...... Page 128
Future Directions ...... Page 132
Appendix A: Supplemental Data ...... Page 134
7 | Lynn Appendix B: Related Publications ...... Page 149
References ...... Page 150
8 | Lynn List of Figures and Illustrations
Figure 1: Schema of the Overlapping Syndromes and Complications in the Critically Ill ...... Page 19
Figure 2: ARDS Causes ...... Page 19
Figure 3: A Timeline of ARDS Clinical and Genetic Contributions ...... Page 39
Figure 4: ARDS Gene Table and Top Pathways ...... Page 40
Figure 5: Heatmap of PBMC Gene Expression Predicting ARDS
Susceptibility and ARDS Mortality ...... Page 45
Figure 6: Inflammation Pathways ...... Page 55
Figure 7: Five interleukin-associated signaling specific pathways ...... Page 55
Figure 8: Genes represented in top enriched endothelial vascular pathways ......
...... Page 58
Figure 9: Platelet and Coagulation Pathways ...... Page 60
Figure 10: Risk Genes in African Descent and Hispanic Ethnicity ...... Page 73
Figure 11: Selection of NAMPT promoter SNPs for genotyping in an ARDS cohort ...... Page 85
Figure 12: NAMPT genotype-phenotype experimental design schematic ... Page 85
Figure 13: rs61330082 genotypes show a significant decrease in
APACHE II Scores ...... Page 87
Figure 14: NAMPT promoter activity is increased by the rs61330082 variant in human endothelial cells following LPS challenge ...... Page 88
Figure 15: Higher eNAMPT levels are significantly correlated with
APACHE II scores ...... Page 90
9 | Lynn Figure 16: ARDS subjects harboring the rs61330082/rs9770242 GG/CC diplotype exhibit significantly increased eNAMPT plasma levels and mortality risk in African Americans
...... Page 91
Figure 17: The optimized ARDS mortality risk prediction approach combines NAMPT rs61330082/rs59744560 haplotypes, eNAMPT levels, and ARDS covariates ...... Page
93
Figure 18: An increase in global methylation is associated with severe
ARDS outcomes (28-day mortality) ...... Page 101
Figure 19: 3 genes (ENTPD2, HPGD, CACNA1D) show an increase in mRNA transcription expression with severe ARDS outcomes (28-mortality) ...... Page 112
Figure 20 cMET is a significant pathway for ARDS severe outcomes ...... Page 113
Figure 21 cMET/HGF pathway schematic ...... Page 115
Figure 22 Overlapping genes from across 17 top GSEA pathways ...... Page 115
Figure 23 Global methylation in ARDS severe outcomes ...... Page 116
Figure 24 Transcription factor CpG islands are hypermethylated in combo rat LPS/VILI models ...... Page 119
Figure 25 Three genes had significantly upregulated transcripts in mRNA seq ...... Page 125
Figure 26 A systems biology approach to ARDS ...... Page 128
10 | Lynn List of Tables
Table 1: Consensus Pathway Data Base (CPDB) Analysis: 38 Enriched
ARDS Pathways ...... Page 48
Table 2: Top SNPs and Genes in ARDS Literature—GWAS Risk ...... Page 63
Table 3: Top SNPs and Genes in ARDS Literature—Candidate Gene Risk
...... Page 69
Table 4: Demographics of ARDS patients genotyped for 7 SNPs on the
NAMPT promoter ...... Page 83
Table 5: rs61330082 is associated with lowered mortality risk in ARDS (n=393)
...... Page 86
Table 6: eNAMPT plasma levels are associated with genotype rs61330082
...... Page 89
Table 7: rs61330082/rs9770242 G/C is a risk haplotype for increased eNAMPT levels in ARDS patients ...... Page 92
Table 8: Prediction models for mortality risk in ARDS ...... Page 93
Table 9: Demographics of ARDS (Illumina EPIC ChIP) ...... Page 110
Table 10: Significant CpG sites in 28-day outcome-based analysis ...... Page 110
Table 11: IGABS validation cohort demographics ...... Page 117
11 | Lynn Abstract
Introduction: Acute respiratory distress syndrome (ARDS) is a severe, highly
heterogeneous critical illness with staggering mortality that is influenced by environmental
factors (such as mechanical ventilation) and genetic factors. Significant unmet needs in
ARDS include addressing the paucity of validated predictive biomarkers for ARDS risk and susceptibility that hamper the conduct of successful clinical trials in ARDS and the complete absence of novel disease-modifying therapeutic strategies. The current ARDS definition relies on clinical characteristics that fail to capture the diversity of disease pathology, severity, and mortality risk.
Methods: A systems biology approach utilizing a variety of cellular and molecular biology
techniques, genetic and genomic data, and data analysis was performed to examine i)
novel candidate genes in ARDS (an agnostic approach) and ii) known candidate genes
associated with key areas of lung injury pathology. I undertook a comprehensive survey
of the available ARDS literature to identify genes and genetic variants (candidate gene
and limited GWAS approaches) implicated in susceptibility to developing ARDS in hopes
of uncovering novel biomarkers for ARDS risk and mortality and potentially novel therapeutic targets in ARDS. I further attempted to address the well-known health disparities that exist in susceptibility to and mortality from ARDS by utilizing a genetic risk
score incorporating both genetic risk SNPs and plasma protein biomarkers in NAMPT (a
known candidate gene previously associated with ARDS risk). I extended the exploration
of genomics in ARDS by performing the first genome-wide methylation study focused on
ARDS mortality. I validated the individual CpG sites, islands, and pathways from this
study.
12 | Lynn Results: Bioinformatic analyses identified 201 ARDS candidate genes with pathway
analysis indicating a strong predominance in key evolutionarily-conserved inflammatory
pathways including reactive oxygen species (ROS), innate immunity-related inflammation, and endothelial vascular signaling pathways. The rs61330082 was significantly associated with risk of ARDS mortality (p=0.036). The rs61330082 GA genotype was associated with lower mortality risk in an adjusted logistic regression model
(OR=0.46 [0.24, 0.88], p=0.019) and was linked to lower NAMPT promoter activity in response to either lipopolysaccharide (LPS) challenge (p<0.001) or to combined
LPS/18% cyclic stretch exposure (p<0.001). An additive rs61330082 genetic model was associated with lower APACHE II scores (p=0.012), and rs61330082 AA was associated with APAHCE II score in an adjusted linear model [β=-26.85[-48.81, -4.90], p=0.017).
Higher eNAMPT levels were also associated with higher APACHE II scores (p<0.001).
The NAMPT haplotype rs61330082/rs59744560 A/C was also associated with lowered plasma eNAMPT levels (β=-9.25 ng/mL [-17.76, -0.73], p=0.033). A mortality risk score utilizing NAMPT haplotype risk, eNAMPT plasma levels, and covariates (AUC: 0.645
[0.567, 0.722]) significantly predicted ARDS mortality across the cohort after adjustment for age, gender, and race (p=0.002). We identified 15 CpG sites in a mortality-based
ARDS outcome pilot study (p<0.05 after FDR multiple testing correction, n=22 survived, n=23 deceased) that were hyper-methylated in ARDS patients with more severe outcomes (28-day based mortality). Pathway-based expression analysis discovered the c-MET pathway as being significant in ARDS severity outcomes.
Conclusion: A systems biology approach allowed for the synthesis of novel and candidate
genes associated with ARDS to be incorporated into my studies of ARDS. I have
13 | Lynn established a novel genotype-phenotype relationship between NAMPT SNPs, eNAMPT levels, and mortality in ARDS. This mortality risk score may have utility in integrating genotype-phenotype data of high-risk subjects for clinical trial stratification.
14 | Lynn Description of Dissertation Format
Pertinent background information and my hypothesis is provided in Chapter 1.
Chapter 2 (Aim 1--Genomic and Genetic Approaches to Deciphering Acute Respiratory
Distress Syndrome Risk and Mortality) is a reworking of my previously published first author manuscript that is now copyrighted4 and has been adapted to this thesis to highlight my contributions. Chapter 3 (Aim 2—NAMPT haplotypes and plasma eNAMPT levels predict ARDS mortality risk) is based upon a current manuscript that is in submission. Chapter 4 (Aim 3—ARDS and Methylation) is an ongoing project with a manuscript current in preparation. Finally, Chapter 5 provides a summary of research conclusions that resulted from my research.
I performed the majority of the experiments, data analysis and interpretation of results, and I am the primary author for the studies presented in Chapters 3, 4, and 5.
The literature and pathway analysis presented in Chapter 3 was a strategy of my own design. In Chapter 4, all data analysis strategies applied were performed by me. Dr.
Xiaoguang Sun guided me while working with endothelial cells, and the University of
Genomics Core performed the high-throughput genotyping assays. In Chapter 5, the methylation ChIP arrays were run by the University of Colorado’s Genetics and Genomics
Core. Data processing, analysis, and follow-up validation of said genome-wide methylation results were performed in Garcia lab by me. While I am the primary author on publications resulting from these data, all publications acknowledge the appropriate lab members and collaborators that contributed to interpretation, data collection, and experimental design. Final decisions on all published manuscripts were approved by Dr.
Garcia.
15 | Lynn
Chapter 1: Introduction
ARDS is a Critical Care Illness
Critical illnesses, including infection, sepsis, trauma, pancreatitis, hemorrhage,
and acute respiratory failure/ARDS (Figure 1)4, account for 20% of US health care costs
($90 billion annually)5, 6. Estimates of the incidence of Acute Respiratory Distress
Syndrome (ARDS) vary between ~200,000 to 400,000 individuals annually in the United
States with mortality rates of ~30-40%3, 7-12. The pathologic hallmarks of ARDS are varied
(marked increases in high permeability pulmonary edema related to acute endothelial cell
activation/dysfunction resulting in paracellular gap formation13-15, alveolar flooding,
decreased lung compliance, severe hypoxemia and a requirement for mechanical
ventilation) but are driven by the breakdown in the alveolar barrier. From a pathobiological
standpoint, two categories of lung injury are recognized in ARDS: i) direct pulmonary
injury and ii) indirect pulmonary injury. Direct pulmonary injury from insults to the local
lung epithelium such as in pneumonia, aspiration, mechanical ventilation, inhalation
injury, and lung contusion. Indirect pulmonary injury occurs secondary to vascular
endothelial damage16 with sepsis the most common cause of indirect lung injury and the highest contributor to ARDS mortality (Figure 2)3. However, these broad categorizations
and classifications of ARDS fail to capture the nuances of the disease and the substantial
heterogeneity of phenotypes within ARDS that has made interrogation of basic
mechanisms and therapeutic development extremely challenging. This has contributed
to the abysmal track record of Phase II/III trials of novel therapies in ARDS5, 17, 18. Clinical
16 | Lynn factors alone have failed to predict which patients will develop ARDS or develop severe
ARDS19. While promising attempts exist to sub-phenotype ARDS cohorts via blood- derived biomarkers predictive of mortality in hope of identifying at risk individuals, successful validation has been elusive20-22. Scoring systems such as the acute physiology
and chronic health evaluation II (APACHE II) score5, 23 or the injury severity score24 in
critically ill patients predict patient outcomes but can only be applied to general intensive
care unit populations and do not provide consistent and accurate estimates of the risk of
death in specific ICU patient populations. For example, mean lung injury scores were not
significantly different between ARDS survivors and non-survivors25, 26 and attempts to characterize predictors of death in ARDS by developing a prognostic index20, 21 remain controversial and without replication or validation. There is a compelling unmet need to identify ARDS sub-phenotypes that stratify patients for both accurate prognostication and
clinical trial purposes. A risk score is a standardized metric for the likelihood that an
individual will experience a particular outcome—in this instance, higher risk for ARDS
mortality20, 21, 27. Risk stratification is the aggregation of multiple individual risk scores to
create a broader, more complex profile of risk.
In the recent year, a new vascular endotype and direct infectious cause of ARDS has afflicted the world—coronavirus 2 (SARS-CoV-2 or colloquially COVID-19)28.
Moderate to severe ARDS is a frequent complication of COVID-19 and requires ventilator
treatment29. A common feature of both ARDS induced by COVID-19 is increased levels of inflammation and higher gene expression of inflammatory mediators. The cytokine storm characteristic of severe ARDS, sepsis, and septic shock includes a wide range of cytokines, chemokines, interleukins, tumor necrosis factors, macrophage inflammatory
17 | Lynn signaling, and C-reactive proteins and CXC-ligand signaling29. SARS-Cov-2 utilizes angiotensin converting enzyme II (ACEII) to infiltrate cells in the host airway29. ACEI
(angiotensin converting enzyme I) was the first recognized risk gene associated with
ARDS and signaling from both ACEI and ACEII have long be recognized as propagators
of the cytokine storm30-35. Focusing on just one signaling gene misses the complexity of
the signaling pathways that propagate the inflammatory response in the injured airway.
Intracellular signaling of the damaged/infected cell to other airway cells occurs via tumor
necrosis factor signaling pathways, tyrosine receptor kinase growth pathways, and
interleukin 6 (IL-6) signaling. These signaling pathways activate NFκB and STAT3 in non- immune cells29, 36. At this juncture, key cell signaling pathways that mediate the cytokine
cascade and loss of endothelial/epithelial barrier integrity in the lower airway converge
between SARS-CoV-2 infection and other ARDS endotypes37, 38. While the risk factors
for ARDS and COVID-19 may differ or are unclear at best28, the downstream disease
pathways are similar enough that COVID-19/ARDS will often be treated together.
18 | Lynn
Figure 1: Schema of the Overlapping Syndromes and Complications in the Critically Ill. Depicted are the major causes of critical care illnesses and associated complications that contribute to the staggering morbidity, mortality, and health care costs in the critically ill.
Figure 2: ARDS Causes. An adaptation of the summary of Cochi et al. (2016) data from fifteen years of ARDS comorbidity data (1999-2013) that identified four major comorbidities that occur with ARDS3.
19 | Lynn Endothelial Barrier Dysfunction in ARDS
Disruption of the endothelial cell barrier occurs during inflammatory disease states
such as ARDS and acute lung injury (ALI)39. The endothelial cell monolayer is critical for
maintaining the blood/air barrier in the microvascular airway, and the disruption of
endothelial cell barrier integrity leads to fluid, macromolecule, and leukocyte infiltration
into the lung39-41. The endothelial cytoskeleton is composed of three primary elements
that maintain barrier integrity: i) actin microfilaments, ii) intermediate filaments, and iii)
microtubules39. Actin is connected at focal adhesion junctions to the endothelial cell membrane, but under excessive contractive forces and mechanical stress, actin pulls apart to form paracellular gaps. The actin/myosin binding relationship allows for
physiologically normal range of stretch but is disrupted by the abnormally high levels of
contraction present in ARDS. These contractive forces are generated by the
Ca2+/calmodulin-dependent myosin light chain kinase (MLCK)—specifically the non-
muscle MLCK isoform (nmMLCK) first identified by the Garcia laboratory in endothelial cells40, 42. nmMLCK is stimulated by thrombin receptors that signal through Rho-
GTPases, Rho-kinases, and Ca2+-dependent pathways. Kinase activity and thrombin signaling both lead to the phosphorylation and activation of nmMLCK, which phosphorylates myosin light chain and enhances myosin binding to actin and contributes
to cell contraction and permeability39, 42, 43.
Inflammatory signaling is key in perpetuating vascular barrier disruption. Activated
neutrophils release reactive oxygen species (ROS) in response to an inciting
inflammation event44. ROS signal to cellular and plasma membranes disrupting
membrane fluidity and facilitating neutrophil recruitment, adhesion, and signaling. ROS
20 | Lynn influence downstream, intercellular signaling pathways by altering the function of receptor
tyrosine kinase which are key in meditating ROS signaling. Many receptor tyrosine
kinases (human growth factor [HGF]45-48, sphingosine-1-phosphate receptor [S1PR1]49,
50, sphingosine-1-phosphate receptor 3 [S1PR3]49, 51, 52, vascular epithelial growth factor
[VEGF]53-56, epidermal growth factor receptor [EGFR]57, 58) participate in regulation of
inflammatory pathways. For example, the S1PR3 receptor is directly nitrated in ARDS50
and the Akt/PI3K/mTOR signaling pathways, which are necessary in promoting
endothelial barrier dysfunction46, 59, 60. Two other receptor tyrosine kinases (VEGF and
EGFR) are growth factor receptors that regulate angiogenesis, macrophage migration,
and tumor growth and metastasis54, 57.
Biomarkers as a Clinically Useful Sub-Phenotyping Tool in ARDS
One approach to address the subphenotyping challenges in ARDS is by utilizing
biologically and mechanistically-relevant plasma biomarkers22, 61-68. Biomarkers reflect underlying pathological mechanisms of disease and may aid in diagnosis, early detection, therapeutic stratification, risk stratification, and outcome prediction of disease. A meta- analysis of ARDS biomarker studies identified 54 studies that met the criteria of: i) a focus on ARDS risk prediction, ii) adult human subjects, iii) a plasma biomarker, iv) the study being available in English, and v) sufficient data for a meta anlaysis67. 20 unique plasma
biomarkers were identified from these 54 studies. Four biomarkers (KL-6 [Krebs von den
Lugen-6], LDH [lactate dehydrogenase], sRAGE [soluble receptor for advanced glycation
end products], and vWF [von Willebrand factor]) were associated with the development
of ARDS risk, and four biomarkers (IL-4 [interleukin-4], IL-2 [interleuking-2], ANG2
21 | Lynn [angiopoietin 2], and KL-6) were associated with mortality outcome in ARDS67. In this
meta-analysis, a high level of publication bias and heterogeneity between studies was reported67. The factors that the authors proposed contributed to these publication biases
(various etiologies of ARDS, different methodology for biomarker measurement, a wide
interval range between study inclusion compared to biomarker measurement, and the
long time window over which the studies were published) encapsulates the difficulties in
evaluating the relative effectiveness of biomarkers for ARDS risk prediction67. Of interest is that the biomarkers that were the strongest predictors of ARDS risk were (with the exception of KL-6) not additionally predictive of ARDS mortality outcome. The need to identify ARDS risk, predict ARDS mortality outcome, and stratify ARDS sub phenotypes
may require separate biomarkers to achieve successful clinical studies.
While single plasma biomarkers in ARDS have had limited success, a panel
approach to plasma biomarkers may better capture the complex pathophysiology of
ARDS69. In a study conducted in the Garcia lab63, eight pathological plasma biomarkers
were identified based upon prior evidence of association between ARDS risk, severity,
and outcome including two biomarkers uniquely discovered by the lab70-84. These
biomarkers represented groups of: i) inflammatory cytokine-chemokines (IL-6
[interleukin-6], IL-8 [interleukin-8], IL-1B [interleukin-1 Β], IL-1RA [interleukin-1 receptor
A]), ii) dual-functioning cytozymes that are enzymes with intracellular function as well as an extracellular cytokine function (MIF [macrophage migration inhibitory factor],
NAMPT/eNAMPT), and iii) vascular injury markers (S1PR3 [sphingosine-1-phosphate
receptor 3], ANG2) (Supplemental Table 1). Six of these eight biomarkers (ANG2, MIF,
IL-8, IL-1RA, IL-6, and eNAMPT) were found to predict mortality by utilizing classification
22 | Lynn and regression trees (CART) analysis and latent-class modeling (LCA)27, 63, 85-87. This
study population was primarily septic63 with a primary interest to predict ARDS severity
defined as 28-day mortality. Another lab studied 21 biomarker proteins represented in
prior studies as associated ARDS88 and chosen due to biological importance in
inflammation, lung epithelial and endothelial injury, fibrosis and coagulation, and
fibrinoloysis88. Two models that predicted ARDS risk compared to controls (utilizing
receiver operator characters (ROC) analysis of area under the curve (AUC)) were generated utilizing: i) the seven most significant biomarkers (RAGE [the receptor of advanced glycation end products], PCPIII [procollagen peptide III], BNP [brain natriuretic peptide], ANG2, TNF-α [tumor necrosis factor alpha], IL-10 [interleukin-10], IL-8), and ii)
a simplified model with the three (RAGE, PCPIII, and BNP) top biomarkers88. Both
prediction models had strong predictiveness (AUC>0.80) in differentiating trauma-
induced ARDS vs controls 88. While both panel-based biomarker studies had differing aims (ARDS outcome vs ARDS risk) and studied different ARDS populations (sepsis vs
trauma), several common themes emerged. In both multi-panel biomarker studies, a key aspect of the experimental design was to capture the diverse pathology that makes ARDS a challenging condition to treat. Both studies identified two common biomarkers (IL-8 and
ANG2) that had predictive powers for both ARDS risk and mortality. However, the weight each study placed upon the individual biomarkers and their respective predictive power in ARDS differed due to the aims of the research63, 88.
Biomarkers fill an unmet need in ARDS clinical studies and clinical trial
designs20, 63, 85. Diagnostic uncertainty in clinical trials and the complex, heterogenous
nature of ARDS have contributed to the failure of ARDS Phase II/III therapeutic clinical
23 | Lynn trials17. Severity scores can successfully link to ICU patient outcomes89, but severity
scores like APAHCE II/III fail to distinguish between mortality risk in specific disease
populations like ARDS90. Biomarker panel approaches offer an alternative endotype
stratification strategy to risk scores in ARDS clinical trials. This makes biomarkers an
attractive clinical tool in ARDS risk stratification.
Genomics in ARDS
Beyond plasma biomarkers, genetic variants (mutations/polymorphisms),
associated with ARDS risk, sub phenotype severity, or severity of outcome, are a potential
source of biomarker discovery in ARDS. The advent of the Human Genome Project and
the sequencing of the human genome in 2000 allowed for the genotyping of genetic
mutations/polymorphisms beyond traditional ‘Mendelian’ (single gene point mutation)
diseases. The most abundant type of mutation in the human genome are single
nucleotide polymorphisms (SNPs)91, and many of the early genome-wide discovery of
SNPs were done utilizing high-density nucleotide arrays92-94. By 2012, the 1000 Genome
Project had validated 38 million SNPs, 1.4 million short gene insertions, and 14,000 large
deletions95. The first ARDS study to utilize a case-control genome-wide association study
(GWAS) design was performed in 2012 and leveraged a multi-cohort collaboration to find
associations for ARDS in a trauma-focused patient population96. Even with a sizeable
ARDS cohort (800 patients) with a single cause for ARDS (trauma), this study struggled
to achieve adequate power for genome-wide significance96. A SNP in an expression
quantitative trait locus (eQTL) in the gene liprin alpha (PPFIA1) was found to be associated with ARDS risk. The finding of only a single significant SNP is in stark contrast
24 | Lynn to other GWAS studies performed in other fields including asthma GWAS’ performed
around the same time period of 2007-2012 which yielded hundreds of significant SNPs
across multiple loci97-101. This has allowed for large meta-analysis with over tens of
thousands of patients to be performed to validate prior assocaitions102. While multi-center
cohorts and larger collaborations have improved the power to detect significance in case-
control ARDS GWAS studies, the acute and time-sensitive nature of studying ARDS adds
an additional challenge in studying critical illness vs a chronic pulmonary disease like
asthma. Next generation sequencing technology coupled with larger ARDS cohorts have
yielded genome-wide significant SNPs in several genes (X-linked gene arylsulfatase D
[ASRD]103, XX Kell blood group complex member 3 gene [XXR3]103, selectin P ligand
gene [SELPLG]104, HEAT repeat containing 1 [HEATR1]104). The Garcia lab identified
SNPs associated with ARDS risk in SELPLG and HEATR1104, but like previous ARDS
GWASs96, 103, SELPLG and HEATR1 have a small number of SNPs with only a modest
effect sizes. The novel aspect of SELPLG and HEATR1 are that they are ARDS risk
SNPS specific to African Americans, and influence key vascular regulatory pathways i.e
P selectin and Akt signaling, however, the modest signal of these SNPs limits their
predictive power potential in ARDS104.
To optimize recovery of useful information from GWAS with modest effect sizes
requires new approaches to translate these gene and loci associations to clinical,
diagnostic, or therapeutic utility. An alternative approach to utilizing plasma biomarkers is
to generate a genetic risk score that utilizes significant SNP associations to stratify
patients into discrete risk groups22, 105-108. GWAS studies with a case-control experimental design have produced a plethora of SNP associations linked to risk. Outcome-focused
25 | Lynn GWASs within disease populations found mutations associated with severity and
outcomes in complex heterogenic diseases109-120. The majority of these associations have
limited odds ratios individually, however, cumulatively can represent a larger aggregate
risk than any SNP alone121, 122. A variety of studies have been utilized to combine and
stratify genetic risk scores to obtain a clinically-relevant metric122, 123. In prostate cancer,
combining family history and GWAS-associated SNPs have resulted in a variety of
methods for calculating genetic risk scores that estimate the relative risk of developing
prostate cancer121. In critical illnesses such as ARDS, this approach has not yet been
attempted due to the size of GWASs needed to generate 30+ SNPs used to calculate
these scores108, 109, 115, the lack of family history data107, 110, 113, 122, 123, and the inability to generate a stratified survival analysis which is the most common way of stratifying and achieving significance between relative risk scores in prior studies123.
In lieu of highly-powered GWAS studies, candidate gene studies in ARDS,
prioritizing a priori knowledge of the biological pathway and mechanisms of the specific
gene, have provided the majority of replicable genetic associations between ARDS
susceptibility and outcomes124. The first genetic mutations associated with ARDS
(surfactant protein B [SFTPB]125, angiotensin-converting enzyme [ACE]34, 35, IL6126) were
discovered utilizing a candidate gene approach with regard to the known mechanisms of these genes in pulmonary airway function (SFTPB)125, epithelial and vascular injury
(ACE)34, 35, and inflammatory pathways and the cytokine storm (IL6)126.
Discovery designs (focused primarily on mRNA microarrays) provided a major
expansion of known genes associated with ARDS risk and outcome severity79, 80, 82-84.
While mRNA arrays do not provide specific loci of mutations associated with disease,
26 | Lynn associated mRNA transcripts provided a crucial link between case-control GWASs and a
priori candidate gene studies. An early approach that identified many genes associated
with ARDS in humans was performed by analyzing orthologous mRNA expression in
animal models (mice, rats, dogs) and from human endothelial cell cultures84. Gene
expression of ontological genes were analyzed in silico across a multi-species
database84. The top GO pathways were immune response, inflammatory response, negative regulation of cell proliferation, cell cycle arrest, blood coagulation, antimicrobial humoral response, induction of apoptosis, cell-cell signaling, and chemotaxis83, 84. The two most commonly upregulated genes were IL-1β and IL-6, which are involved in a variety of biological systems that regulate inflammatory and immune responses83.
Cytokine/chemokine genes and their receptors were also heavily represented in this GO pathway analysis83. CXC chemokine ligand-2 (CXCL2)127, cell chemokine 2 (CCL2)128,
and CC chemokine receptor 5 (CCR5)129 encode genes involved in inflammatory
disorders and in amplifying lung inflammation. An independent study that summarized
these early orthologous studies that compared ARDS animal and human cell culture
ARDS models produced a list of genes 37 that were further mechanistically investigated
in ARDS models80, 83, 130-132 including IL-6, IL-1β, SERPINE1, and IL1R2 which were independently validated in multiple human mRNA array studies67, 76, 88, 133-139. Using these
early animal model and cell culture array studies, candidate gene studies from 2008-2014
investigated risk loci in human ARDS patients in genes of interest. A literature review of
56 independent candidate gene studies over this time period identified 41 candidate
genes in ARDS models that had risk loci associated with ARDS risk in human patients124.
Importantly, this study included loci derived from across multi-ethnic cohorts (6.6%
27 | Lynn African Americans, 66.7% European Americans, 15.0% Asian Americans, and 11.7% other multi-ethnic samples)124. While several of these genes (IL-6, SFTPB, ACE,
SERPINE1) had previously identified risk loci associated with ARDS, this new literature review identified several other genes (myosin light chain kinase [MYLK], interleukin-10
[IL-10], angiopoetin-2 [ANGPT2], fas cell surface death receptor [FAS], interleukin-19 [IL-
18], peptidase inhibitor 3 [PI3], tumor necrosis factor [TNF], nicotinamide phosphoribosyl
transferase [NAMPT]) that were validated by several groups that had risk or outcome
associated loci with either ARDS or ALI124.
Epigenetics in ARDS
Beyond transcriptomics and genetics studies, many areas of ‘omics’ have yet to
be applied to ARDS19. Even in transcriptomics, mRNA (protein coding exons) have been
the major focus of array and next generation sequencing studies. RNA is a diverse
molecule that encompasses transfer RNA (tRNA), ribosomal RNA (rRNA), microRNA
(miRNA), and non-coding or lincRNA. There have been over 10,000 papers on miRNAs
since their discovery140. Excluding mRNA, miRNA gene silencing of transcription is the
second most studied type of genetic regulation in ARDS141-146. miRNAs are small (~22
nucleotide base pairs) sequences that bind the 3’ untranslated region (3’ UTR) of targeted
mRNAs, and classically result in gene silencing and potentially promote mRNA
degradation144, 147. Extracellular vesicles facilitate miRNA transport to other cells, but free-
floating miRNAs have been found in the urine, saliva, and lymphatic fluid140, 148, 149.
Because of the relative ease of a blood draw compared to a biopsy, circulating miRNA is
an attractive potential target to profile as a biomarker of disease. In breast cancer,
28 | Lynn circulating miRNA profiles may detect early-stage breast cancer142. However, the
precedent established by miRNA profiling represents an attempt to transition biomarker profiling away from only proteins to targeting markers of genetic regulation.
Similar to miRNA profiling, there is an untapped potential for utilizing methylation and epigenetic profiling in ARDS. Epigenetics is the study of heritable changes to gene activity or function that is not dependent on altering the DNA sequence itself150. DNA
methylation is when methyltransferases (DNMTs) catalyze the transfer of a methyl group
150 (CH3) onto the fifth carbon of a cytosine residue . While genetics in ARDS has been
well studied, the literature on epigenetic changes in ARDS is sparse and focused mostly
on candidate gene studies151-155. A bioinformatics approach utilizing mRNA arrays and
methylation arrays from prior GEO candidate gene studies (coupled with GEO and DAVID
database annotation) found several genes (FYN proto-oncogene, Src family tyrosine
kinase [FYN], TBC1 domain family member 22B [TBC1D22B], C-X3-C motif chemokine
receptor 1 [CX3CR1]) with multiple CpG sites associated with ARDS156. Genome-wide methylation analysis could be useful for detecting susceptibility to ARDS risk or outcome157. A bias-free application of methylation has the potential to be coupled with
other ‘omics’ strategies with a focus towards extending precision medicine in ARDS158.
Another strategy to implement methylation data in ARDS clinical or therapeutic contexts
is to expand candidate gene studies with a focus on prior associated and well-validated
genetic targets19, 159. As with mRNA/transcriptomics and DNA/genetics, the heterogenous
nature of ARDS will necessitate a blend of discovery and candidate gene-based studies
to advance epigenetic regulation risks in ARDS.
29 | Lynn MYLK and NAMPT: Two Candidate Risk Genes in ARDS
Two of the major ARDS-associated genes studied by Garcia lab are myosin light
chain kinase [MYLK] and nicotinamide phosphoribosyl transferase [NAMPT]. Of these
two genes, MYLK was the earlier of these two candidate genes studied due to its prior
known mechanistic roles in regulating and maintaining the integrity of the endothelial cell
barrier160. These early genomic studies of MYLK proved advantageous, and MYLK
becoming one of the earliest transcripts to be an independently validated ARDS risk
gene67. An essential pathogenic element of ARDS is the loss of endothelial cell integrity
resulting in increased vascular permeability which is critical to development of alveolar
flooding and hypoxemia39. The actin-myosin contraction is one of the key regulatory
forces involved in maintaining the endothelial barrier, and an increase in contractile forces
is a driver of barrier dysfunction39. MYLK encodes two Ca2+/calmodulin (CaM)-dependent myosin light chain kinase [MLCK] isoforms associated with airway dysfunction and
endothelial barrier dysregulation: i) a 108 kD a smooth muscle myosin light chain kinase
(smMLCK)40, 161 and ii) a 210 kD non-muscle MLCK isoform (nmMLCK)33, 39, 42, 162-168.
that was initially cloned by the Garcia lab40, 42 The expression and function of nmMLCK
is critical to maintaining endothelial function in vitro and in vivo utilizing animal models of
lung injury40-42, 167, 169-171. Coding SNPs in both nmMLCK and smMLCK were found to be
associated with risk and severity of sepsis, sepsis- and trauma-induced ARDS, and asthma 97, 167, 170, 172-176. Studies of the MYLK promoter have shown SNPs associated with asthma and airway dysfunction alter MYLK activity and are associated with increased transcription of MYLK165, 167, 170, 173, 175, 176. While it has not been studied in ARDS, a
pseudo-gene (a partial duplication of a gene to another region of the genome which can
30 | Lynn be conditionally functional) of MYLK (pseudo gene MYLK [MYLKP1]) has been shown to
have promoter SNPs associated with a risk for developing prostate cancer177. Notably,
this region of the MYLKP1 overlaps with the region of MYLK that transcribes nmMLCK
and shares ~88% of the same genetic sequence177-180.
NAMPT, also known as pre-B cell colony-enhancing factor (PBEF) or visfatin, first
entered the critical illness literature when recognized as upregulated in neutrophils from
sepsis subjects in response to IL-1β stimulation77. The NAMPT structure was then
determined and NAMPT found to be a member of a dimeric type II family of
phosphoribosyltransferases181. The Garcia lab, utilizing genomic approaches in
preclinical ARDS models and human biologic samples, was first to identify NAMPT as an
ARDS biomarker68. Several studies determined NAMPT be an attractive ARDS candidate
gene and therapeutic target especially in ventilator-induced lung injury (VILI)68, 182 that is
essential to the development of VILI utilizing eNAMPT neutralizing antibody strategies as
well as heterozygous partial knock-out mouse model (NAMPT-/+)182 (double knock-out
NAMPT-/- embryonically lethal). Compared to wild type mice, NAMPT-/+ mice subjected to excessive mechanical ventilation to model VILI conditions in ARDS, there was attenuation of multiple inflammatory pathways including interleukin, neutrophil infiltration, toll-like receptor (TLR) and JAK/STAT signaling pathways182. Ultimately, this lab identified
toll-like receptor 4 (TLR4) as the eNAMPT-binding receptor triggering NFkB transcriptional upregulation and inflammatory cytokine expression183, again making eNAMPT an attractive ARDS therapeutic target.
31 | Lynn Prior studies in the Garcia lab designed to interrogate NAMPT promoter regulation,
discovered a sizeable region in the NAMPT promoter (-2,428 to -1,228 from the transcription start site) that is sensitive to mechanical stress154 termed the mechanical-
stress-inducible region (MSIR). Further deletion constructs of the NAMPT promoter and point-mutation transfection studies in endothelial cells proved the mechanistic significance of this mechanically stress sensitive region in the NAMPT promoter66, 151, 153,
154, 183-186. Several NAMPT promoter SNPs were found to be associated with ARDS risk,
sepsis risk, or mortality outcomes in ARDS patient populations68, 151, 153, 154, 185, 187, 188.
Further investigation of the NAMPT promoter revealed a ~1300 base pair CpG island that
spans the promoter and transcription start site of NAMPT151. Endothelial cells transfected
with a deletion construct of this CpG island and subjected to mechanical stress demonstrated several demethylated regions151. While these deletion constructs provide
crucial mechanistic data into the epigenetic regulation of NAMPT151, the regulation of
individual CpG sites in this large CpG island spanning the NAMPT promoter is unknown.
Epigenetic NAMPT changes in patients has yet to be explored, although endothelial cell
data provides strong mechanistic evidence that epigenetic regulation of NAMPT under
chemokine stress and mechanical stress could be crucial to NAMPT transcription in
ARDS and VILI151. NAMPT/eNAMPT is a particularly interesting novel target for both
ARDS diagnostics (NAMPT genomic associations with ARSD risk and outcome) and
therapeutics (a competitive eNAMPT antibody that binds to TLR4).
Summary and the Promise of System Biology
32 | Lynn Despite intensive research, definitive and disease-defining biomarkers for ARDS
remain in limbo. The probability of finding a single biomarker in a heterogenous disease
such as ARDS that covers risk but also sufficiently discriminates ARDS susceptibility from
other critical illnesses is unlikely. Systems biology is an integrative approach that is
focused on deciphering the relationships between genes, protein and cell elements of a
biological system189, 190. While systems biology is traditionally thought of as the ‘omics’
fields (genomics, proteomics, metabolomics, and transcriptomics), it represents much
more than simply interrogating big data. Systems biology attempts to understand how
biological systems are linked together in nonlethal and mechanistic combinations.
Systems biology allows new techniques and tools in the ‘omics’ fields to study how
genotype generates phenotype190. It is the combination of multiple ‘omics’ fields through
computational and bioinformatics to that allows for a broader study of disease
phenotypes189. The tools and methods offered by utilizing a systems biology approach and mindset present powerful opportunities for utilizing the plethora of genetic data in
ARDS to predict risk and develop therapeutics.
Hypothesis
The underlying hypothesis of my work is that the synthesis of genomic and
molecular/cell data in ARDS can facilitate development of novel strategies to predict
ARDS mortality and thus enhance clinical trial design. The aims of my thesis attempt to
use a systems biology approach to synthesize multiple models, studies, and techniques
in the field of ARDS. Aim 1 focuses on an ambitious literature review and database-
centered pathway analysis that aims to synthesize the complex literature of ARDS
33 | Lynn genetics, transcriptomics, and proteomics. Aim 2 focuses upon development of a novel genetic risk score to potentially serve as a novel biomarker approach integrating clinical
variables, genetic risk SNPs and plasma biomarkers. Aim 3 focuses upon epigenetic
methylation in ARDS and utilizes both genome-wide methylation profiling and pathway
analysis to investigate candidate genes that may discern ARDS subjects likely to survive
from those ARDS subjects who may not.
34 | Lynn Chapter 2: Aim 1: Genomic and Genetic Approaches to Deciphering Acute
Respiratory Distress Syndrome Risk and Mortality
Significance
The ability to leverage newly described systems biology and “omics” approaches and techniques (genetic and genomic, biomarker panels, and epigenetic profiling) hold promise for improving the capacity to characterize risk and prognosis for ARDS and in other ICU patients with critical illness and respiratory failure. ARDS genetic and genomic studies potentially provide the basis for identifying candidate genes, biomarker discovery, risk stratification, and novel ARDS therapeutic targets19, 83. While ARDS is not a known
inheritable condition, the pattern of host injury response-recovery has significant
heritability across populations19, 81, 191. Unlike rare and high penetrance monogenetic
(‘single’ mutation or gene) diseases, ARDS risk and severity are influenced by multiple
genes to a varying effect. The potential to utilize genomic approaches to identify ARDS
high inflammatory sub-phenotypes at higher risk for death is a currently untapped area of
therapeutic stratification in severe lung injury19, 27, 61, 67, 133. The diversity of potential genetic biomarkers in ARDS range from markers of epithelial injury (RAGE)15, 192,
endothelial activation/injury (ANGPT1 [Angiopoetin-1], ICAM-1 [intercellular adhesion
molecule 1], VEGF [vascular endothelial growth factor])15, 54, 193-195, pro-inflammatory (IL-
1Β, IL-18, IL-6, IL-8)70-72, 75, 76, 135, 136, 196, anti-inflammatory molecules (IL-10)72,
coagulation and fibrinolysis proteins (PA-1 [pediocin PA-1])139, 197-199, and macrophage
markers (HMGB1 [high mobility group box 1], MIF)65, 200-203. The true challenge in utilizing
these diverse biomarkers in ARDS diagnostics and therapeutics is synthesizing
35 | Lynn associations across different genomic platforms, diverse populations, performed under
differing definitions of ARDS, and over a thirty-year time period16, 20, 84, 158, 200, 204-206.
We1, 7, 39, 41, 52, 60, 68, 84, 153, 160, 163, 168, 174, 176, 182, 183, 207-210 and others19, 59, 79, 82, 96, 133,
151, 155, 211-221 have contributed to the notion that ARDS represents the ultimate in genetic stress, with a complex assortment of genes that individually contribute a limited over-all
effect size133 with limited value for risk prediction. However, in aggregate, these genes
significantly impact lung injury phenotypes (effect size) and ARDS pathology19. Unlike
other complex genetic diseases, the field of ARDS genetic research has not benefited
from family pedigree studies133. However, utilizing both candidate gene and GWAS
approaches, we have identified several novel genetic targets and biomarkers of ARDS
risk and outcome severity including NAMPT68, 154, 222, toll-like receptor 4 (TLR4)52, 188,
MYLK96, 167, 170, 173, 174, 176, DIO2 (iodothyronine deiodinase 2)174, 221, GADD45a (growth
arrest and DNA damage-inducible gene)59, 60, macrophage migration inhibitory factor
(MIF)65, and sphingosine 1-phosphate receptors 1 and 3 (S1P1, S1P3)52, 208. In Aim 1, I
have used pathway analysis tools to integrate three types of studies: i) studies utilizing
peripheral blood mononuclear cells (PBMCs) for identification of genetic signature in
ARDS, ii) meta-analysis of ARDS risk and mortality biomarker studies67, and iii)
discovery-based genomic studies. I speculate this strategy may expand the
understanding of ARDS pathobiology and potentially identify genes associated with
ARDS mortality that may serve as diagnostic makers and therapeutic targets.
I evaluated PubMed literature relevant to ARDS to identify a total of 201
dysregulated genes potentially associated with either ARDS risk or severity and then
performed pathway analysis on these genes. Pathway analysis included genes from both
36 | Lynn candidate and agnostic GWAS studies, genes from mRNA microarray and sequencing
studies, and proteomic evaluations that identified putative candidate genes but without
associated single nucleotide polymorphisms (SNPs) related to ARDS risk or ARDS
mortality. Blood biomarkers without genomic/genetic or mechanistic evidence were
excluded. I further summarized the current state of ARDS genetics in terms of
susceptibility risk SNPs and those that confer increased mortality risk based on adjusted
significance in their respective study. I have also chosen to evaluate ARDS studies that use mortality as an endpoint as this captures the most severe outcome for ARDS patients.
Our bioinformatically-derived results are consistent with the concept that evolutionarily-
conserved reactive oxygen species (ROS), innate immunity-related inflammatory
networks, and vascular signaling pathways are potent contributors to multiple organ
dysfunction-related ARDS mortality and pathobiology83, 84.
Recent Advances and ARDS in the Clinical and Genetic Literature
ARDS was first described in the citation classic report by Ashbaugh et al in
1967223. Figure 3 depicts a brief timeline of the initial important clinical studies for
common therapeutics in ARDS with clinical therapeutics the first natural avenue for ARDS
therapeutics206. Early clinical ARDS management methods to manage lung atelectasis
(alveolar) collapse utilizing positive end-expiratory pressure (PEEP) and the prone
position to aid in oxygenation224. The recognition of VILI or barotrauma-mediated lung
injury induced by the mechanical ventilator to be a major contributor to ARDS mortality
was first identified in the landmark ARDSNet clinical trial (2000) using lower tidal volumes
to reduce VILI and ARDS mortality represented another significant clinical therapeutic
37 | Lynn advance225,13. In contrast to this long history of clinical ARDS studies, the first candidate
gene study in ARDS involving ACE polymorphisms corresponding to higher ACE plasma
levels in ARDS was reported in 1992. Candidate gene studies in ARDS, rare prior to the
sequencing of the human genome35, have subsequently significantly populated the ARDS
literature18, 22, 30, 32, 34, 52, 55, 70, 72-76, 79, 90, 125, 137, 153, 154, 174, 203, 211, 219, 221, 226-242. The
chronology of the discovery of the major candidate genes in ARDS as well as the
transition into ARDS GWAS studies has been previously elegantly detailed133. In this review study, I have attempted to capture more recent genetic developments and
reporting in ARDS as well as to expand upon the previous literature by incorporating pathway analysis205, 206.
38 | Lynn
Figure 3: A Timeline of ARDS Clinical and Genetic Contributions: A selection of clinical contributions for ARDS (bottom) and genetic contributions to the ARDS literature (top). The relative recent history of genetic contributions to the ARDS literature should be noted compared to a much long history of clinical contributions to treatment. Clinical and genetic timelines adapted from Laffey et al. (2017) and Reilly (2017) 97,178.
I integrated the variety of genetic studies in the ARDS literature, and excluding the initial ACE polymorphism study35, all candidate gene studies were published after 2000
(Figure 3). GWAS studies enter the ARDS literature starting in 2012 but compared to
GWAS studies in other disorders and clinical arenas99, 102, 108, 109, 112, 115, 243, ARDS GWAS studies generally include smaller cohorts19. The literature of ARDS genetics is relatively small and recent compared to other academic fields, and the genes presented are a comprehensive list of all mapped genes in ARDS that were significant (respective of the individual study) for either: i) a specific polymorphism associated with ARDS; or ii) an overall gene expression level significant for ARDS risk. Below, the importance of gene expression studies and mortality risk genes are discussed as these genes potentially serve as novel therapeutic targets. The Pathway Analysis section below highlights our attempt to synthesize the recent and diverse genetic studies in the ARDS literature. A list
39 | Lynn of 201 mapped genes were identified via PMC/PubMed literature search of ARDS, Acute
Lung Injury (ALI), and ‘lung injury’ studies that identified genes that were differentially expressed or conferred risk for ARDS, severe sepsis, or mortality244 (Figure 4,
Supplemental Table 2).
Figure 4: ARDS Gene Table and Top Pathways: 201 genes (identified through differential gene expression (mRNA sequencing, mRNA CHIP array, RNA microarray, proteomics, candidate gene sequencing, or genome-wide association study (GWAS) that mapped onto either the Reactome or Wikipathways databases for pathway analysis (5,359 human pathways covered). The second, third and fourth columns reflect the number of citations for each gene158. Four broad categories of enriched pathways (immunological signaling, reactive oxygen species-related signaling, vascular signaling, and transcription factor-related pathways) and the number or genes represented in each (p<0.01, 5 members minimum per pathway).
40 | Lynn
Potential ARDS Candidate Genes Identified by Dysregulated Gene
Expression
The 201 ARDS genes were derived from studies with clinical populations, human-
derived cell lines, or genes validated across multiple animal models with genetically-
conserved regions80, 83, 84, 133, 205, 245. The Garcia lab has previously performed cross-
species analysis of ventilator-induced lung injury (VILI) models (rat, mouse, canine)
(orthologous approach) and human ARDS patients to yield a list of genes that are
conserved across species and of potential importance in the pathophysiology of ARDS
and VILI80. Eleven genes were ‘immune response’ genes that were highly significant with
Expression Analysis Systematic Explorer (EASE) scores, and two genes (IL-1Β, IL-6),
are noted to harbor SNPs independently associated with ARDS risk or ARDS mortality80,
134, 246. Six genes were involved in ‘inflammatory response’ and ‘innate immune response’
pathways. Taken together, these data indicate that multiple genes fall into the
evolutionary-conserved inflammatory and immunological-related pathways across species and are potentially important in ARDS and VILI pathology80. A limited mRNA
study of ARDS and healthy controls yielded 12 upregulated genes in ARDS73 with IL-1R2,
a decoy receptor that dampens IL-1 signaling73, identified as the top upregulated gene.
Three genes (ARG1, MHC-DRB1, CCR2) are macrophage-specific genes expressed by
activated macrophages73.
Another strategy to study pathways incorporating genes of interest is to utilize
genetically-engineered preclinical murine models involving exposure to ARDS and VILI
followed by genome-wide lung tissue gene expression and pathway analysis182. For
41 | Lynn example, NAMPT is an ARDS candidate gene67 that harbors several promoter SNPs (-
1001C/T, -2422A/G, -948G/T) associated with increased risk of ARDS and ARDS mortality52, 126, 153, 222, 247. Genomic comparisons of wild type mice and NAMPT
heterozygous mice (NAMPT-/+) exposed to eNAMPT, VILI, or LPS revealed significant
NAMPT-influenced pathways involved in ‘acute phase response signaling’, ‘IL-10
signaling’, ‘IL-6 signaling’, ‘NF-кB signaling’, ‘LXR/RXR activation’, ‘Leukocyte
Extravasation Signaling’, ‘PPAR signaling’, ‘Death Receptor Signaling’, ‘Apoptosis
Signaling’, and ‘TLR signaling’183, 210. This was the first study to link NAMPT and eNAMPT
to TLR signaling and pathways. Similar genomic-intensive studies independently identified Toll like receptor 1 (TLR1) and interleukin-1 receptor-associated kinase (IRAK1) as ARDS risk genes49, 248. We have further shown that eNAMPT directly interacts and
signals to TLR4183.
A complementary approach to identify pathways relevant to ARDS mortality is to
utilize proteomic analyses to identify ARDS biomarkers that identify ARDS sub-
phenotypes213, 214. Proteomic analysis of bronchoalveolar lavage fluid (BALF) in ARDS
survivors and non-survivors214 revealed differentially expressed proteins that fall within
‘acute phase signaling’ and ‘FXR/RXR Activation’ pathways (results remarkably similar to
results from preclinical models of ARDS)182, 214. The ‘oxidative ethanol degradation’ and
‘fatty acid α-oxidation’ pathways were significantly upregulated in BALF obtained from
ARDS non-survivors214. Utilization of quantitative electrophoresis-based proteomics
method (DIGE) identified 37 proteins differentially expressed between ARDS patients and
healthy controls249. The top network pathways (‘Wounding’ and ‘Inflammatory Response’)
included calgranulin A (S100A8), calgranulin B (S100A9), calgranulin C (S100A12),
42 | Lynn serum amyloid protein (SAA), complement C9 precursor (C9), hemopexin precursor
(HPX), peroxiredoxin 5 mitochondrial (PDX5), complement C3 precursor (C3), annexin
A1 (ANXA1), and alpha-1-antitrypsin (SERPINA1)249.
PBMC Gene Expression in ARDS as Predictors of Mortality
Peripheral blood mononuclear cells (PBMCs) are an easily obtainable blood cell
fraction that is broadly representative of innate immunity status84, 132, 202, 250, 251. A meta-
analysis of PBMC molecular biomarkers (54 distinct studies) attempted to validate ARDS
gene biomarkers67 and found two significant sets of blood plasma biomarkers67. One set consisted of ARDS risk genes and included Krebs von den Lugen-6 (KL-6), lactate dehydrogenase (LDH), soluble receptor for advanced glycation end products (sRAGE), and von Willebrand factor (vWF). A second set of genes associated with increased ARDS mortality67 included interleukin 4 (IL-4), interleukin 2 (IL-2), angiopoetin 2 (ANG2), and
KL-6. IL-4 is also an ARDS candidate gene with SNPs associated with ARDS risk, possibly via regulation of lung repair in cellular and animal models of ARDS75, 219, 230, 252.
To further determine the utility of a PBMC-derived gene signature in the ICU
setting, I interrogated differentially expressed PBMC genes in 55 ARDS survivors and
non-survivors (Affymetrix GeneChip Human Exon 2.0 ST microarray). Figure 5A depicts
a heatmap displaying results of bioinformatic analysis with 33 differentially-expressed
genes (DEGs) identified in a molecular signature in ARDS patients (n=23) vs controls
(n=80), 19 genes were downregulated, 14 upregulated (fold change >2 ± 4.41, p<7.26e-
23). Importantly, of the 21 genes predictive of survival207, the “Toll-like receptor signaling pathway” (23 up-regulated, 192 down-regulated) was the top enriched pathway. Figure
43 | Lynn 5B depicts DEGs in PBMCs from 23 ARDS patients that reflect survival with 16 genes
downregulated and 5 upregulated genes (Figure 5B) include previously reported
dysregulated genes in other ARDS studies (PLAUR, IL1B, VEGFA)54, 83, 246, indirectly
validating these data (Figure 5A, Figure 5B). IL1R2 harbors SNPs that confer increased
risk for ARDS and represents a biomarker potentially predictive of sepsis induced ARDS
mortality79. The gene with the greatest magnitude of upregulation was MMP8 (matrix
metallopeptidase 8) with >400-fold change. Although levels of plasma proteins reflecting
the expression of these genes were previously reported, this is the first report of MMP8
and TIMP-1 as genomic markers among non-survivors in ARDS233. This link between
MMP8 and IL1B as molecular biomarkers in blood and gene expression biomarkers predicting survival in ARDS, suggests this approach may yield clinically- and biologically
relevant ARDS biomarker candidates14, 75, 233, 252-254.
44 | Lynn Figure 5: Heatmap of PBMC Gene Expression Predicting ARDS Susceptibility and ARDS Mortality: Panel A. The 33 top genes identified in a molecular survival signature from PBMC mRNA in ARDS patients (n=23) vs controls (n=80, PMID:19222302). 19 genes are downregulated and 14 are upregulated (Fold change > 2 ± 4.41, p< 7.26e-23). Panel B. 22 genes in the ROS pathway were predictive of ARDS survival in ARDS patients. 5 of these genes are upregulated, 17 are downregulated25.
45 | Lynn Figure 5B results are also consistent with other reports that Toll-like receptor 4
(TLR4) signaling pathway is a top pathway in predicting survival in ARDS patients183, 217.
Extracellular NAMPT (eNAMPT), a validated ARDS blood biomarker whose NAMPT
promoter SNPs confer risk of ARDS and ARDS mortality52, 126, 153, 187, 222, is a novel TLR4
ligand183 and both TLR4 and NAMPT are differentially expressed in animal models of
ARDS80, 151, 219. There is increasing interest in potentially therapeutically targeting the
NAMPT pathway as a strategy to reduce ARDS mortality255. In addition, NAMPT
genotypes and plasma protein levels represent an opportunity to develop a panel of
biomarkers/genotypes that could be employed for clinical trial stratification based on
ARDS mortality outcome90, 200.
Pathway Analysis
High throughput screenings have provided a wealth of valuable genome-wide data,
and pathway analysis is the next logical step to integrating these results in order to
understand the biological phenomena that underpin these data and generate future hypotheses245, 256. While database analysis is currently coding protein focused, and
multiple database searches can be used to integrate GWAS, mRNA, and proteomic
studies256. I employed multiple genomic and biological pathway database searches to
compile the results of ARDS genomic studies to facilitate better understanding of the
pathways involved in a complex genetic and heterogenic diseases such as ARDS. While
pathway analysis is not a meta-analysis of ARDS patients, pathway analysis using
multiple databases allows the results of candidate gene, agnostic GWAS studies, and
unclassified studies to be combined for a better understanding of the cellular and
46 | Lynn molecular pathways that are involved in ARDS pathobiology. Pathway analysis has two
major benefits: i) it allows for thousands of genes to be reduced in complexity245, and, ii)
it allows for the development of active pathways that are significantly dysregulated in
ARDS potentially providing mechanistic insights that extend beyond creation of a simple
gene list245.
With a complex genetic disease like ARDS, many genes with varying effect sizes
will presumably be involved in ARDS pathobiology. Pathway analysis organizes the
results of GWAS studies, candidate gene studies, and meta-analysis to understand the biological pathways that contribute significantly to disease progression. I used pathway analysis for gene sets (GO ontology terms), protein-protein interactions, and gene interactions (Reactome and Wikipathways; 5,259 human pathways searched) to analyze our collected pool of 201 ARDS genes256. These studies were performed on the Max
Planck Institute for Molecular Genetics consensus pathway database (CPDB) database
across three pathway sources with the most relevance (Reactome and Wikipathways)
(Table 1)257-259. Enriched pathways were defined as containing more than five genes represented and a p-value of <0.01, which resulted in 72 total enriched pathways. In
addition, 38 relevant and/or highly enriched pathways were chosen based on clustering
of genes (Figure 4, Table 1). Of the enriched pathways, 17 resided in Reactome and 21
in Wikipathways (Table 1) and were further divided into reactive oxygen species (ROS)
pathways (n=6), immune and inflammatory pathways (n=9), cardiovascular signaling
pathways (n=11), transcription factor signaling pathways (n=6), and other pathways
(n=6).
47 | Lynn
48 | Lynn
49 | Lynn
Reactive Oxygen Species (ROS) Pathways
Reactive oxygen species (ROS) play an important role in sepsis and ARDS and contribute to the severe disruption of the endothelial barrier and the resulting inflammatory cascade in the lower airway260. Neutrophils migrate across the endothelial barrier in
response to endothelial secreted cytokines and chemoattractants. In addition to these
endothelial secreted pro-inflammatory signaling, neutrophils are further sources of
released pro-inflammatory cytokines, ROS, proteolytic enzymes, nitrogen species,
cationic proteins, and lipid mediators260. Each inflammatory cell type in the lung generates
and releases distinct profiles of ROS molecules215. Leukocytes express NADPH oxidase
50 | Lynn and nitric oxidase synthases (NOS), which together generate peroxynitrite and other ROS
species215. In ARDS, macrophages initiate prolific ROS activation215. Several isoforms of
NADPH oxidase (NOX1, NOX2, NOX4, NOX5) are expressed in the endothelium, and
these cells increased expression of NOX1, NOX2, and NOX4 drive endothelial and epithelial barrier dysfunction and generate substantial amounts of secondary ROS212, 215,
261, 262. The main role of NOX is to catalyze the reduction of molecular oxygen (O2) to
superoxide (O2-)212. Pulmonary endothelial cells express both NOX2 and NOX4 that
generate ROS under hypoxic conditions as well upon exposure to mechanical stress
caused by VILI261, 262.
ROS generation is linked to survival in sepsis patients52, 77, 81, 90, 167, 202, 207, 211, 216,
227, 263, 264, and we recently reported that a 21-gene ROS gene signature was significantly
linked to survival in sepsis207 (‘Oxidative phosphorylation’ was the top enriched pathway
result). Oxidative phosphorylation is the major pathway of ATP generation in eukaryotic
cells, including the vascular endothelium265. Endothelial mitochondria are also a major source of ROS under aerobic conditions, which include encode complex organelles like
multiple peroxisomes (the P450 complex, xanthine oxidases, and nicotinamide adenine
dinucleotide (NADPH) oxidase complexes) that are encoded by a large number of
genes265. Many mitochondrial genes involved in ATP production create ROS byproducts265, 266. The list of ROS genes (Figure 5B) include chaperon proteins (HSP40,
HSP70), which were not commented upon in the original paper207 but are included in the
KEGG oxidative phosphorylation pathway, and these chaperon proteins are of potential
mechanistic interest to understanding the role of oxidative phosphorylation in ARDS. The
21 ROS gene signature (CSDE1, DNAJC8, DNAJB9, PRDX5, GCLM, FTH1, DNAJA3,
51 | Lynn GSTP1, CCT7, NCF1, CCT8, DNAJB6, PRDX3, SOD2, DNAJC5, CYBA, PRDX2,
DNAJB11, HSPA1A, KEAP1, GSR) was used to create a sepsis risk score207 (Figure 5B)
and significantly outperformed a thousand randomly-picked genes in predicting survival
among ARDS patients207. My pathway analysis further solidifies the importance of ROS
species and this previously generated ROS signature as a potential genomic (mRNA) risk
signature.
In the wider ARDS literature, 27 other genes were identified in pathways related to
oxidative and cellular stress (Table 1). NOX4 represented a link across multiple pathways
(‘detoxification of ROS’, ‘Cellular stress’, ‘Cellular response to external stimuli’, and
‘oxidative stress’). Both the superoxide dismutase (SOD) family (SOD2, SOD3) and the peroxiredoxin gene family (PRDX2, PRDX3, PRDX5, PRDX6) were also represented across multiple pathways (‘Detoxification of ROS’, ‘Cellular stress’, ‘Cellular response to external stimuli’) (Table 1) and exhibit common variants associated with ARDS in case- control studies267. These genes, together with those that comprise the previously
identified 21 gene signature207, are relevant to a genetic ROS risk and survival signature
in ARDS.
‘Oxidative stress’ pathways join ‘inflammation’, and ‘apoptosis’ as pathways
implicated as being important to ARDS pathology182, 214, 215, 268, 269 with several ‘Oxidative
stress’ pathway genes containing inflammatory-related genes that are significantly linked
to ARDS (NAMPT, IL-6, IL4, IL-13) and to vascular signaling pathways215. These genes could serve as an important link between these disparate biological pathways and have potent potential as biomarkers. Important redox-sensitive pathways in ARDS are the mitogen-activated protein kinase (MAPK) and signal transducer and activator of
52 | Lynn transcription (STAT) pathways that regulate several ARDS candidate genes242, 270-273.
Another redox-sensitive pathway involved in signaling and fibrotic proliferation is the
PI3K/Akt pathway229. Individually, many genes (NOX1, NOX4, STAT4, STAT5) in the
MAPK/STAT and PI3K/Akt pathways exhibit mechanistic roles in ARDS pathology in animal models229.
One highly ROS-related pathway is the NRF2 (NFE2L2) transcription factor and
signaling pathway. NRF2 is a ubiquitous master transcription factor that regulates
antioxidant response elements (ARE) and mediates cytoprotective and antioxidant
protein expression (Figure 5A)274. In the healthy lung, NRF2 has a protective effect
against hyperoxia, mechanical stress, and VILI237, 274. Additionally, NRF2 regulates NOX4
in the mouse lung via ROS signaling in ARDS VILI and LPS models209, 237.
Mechanistically, NRF2 binds to KEAP1 [Kelch-like ECH associated protein 1], which was
a gene identified as differentially upregulated gene in the ROS ARDS survival gene
signature207, 237. The Keap1-Nrf2 complex translocates NRF2 to the matrix that binds to
AREs and transcribes heme oxygenase-1 (HO-1), NAD(P)H:quinone oxidoreductase 1
(NQPO1), catalase, and superoxide dismutase (SOD)275. Interestingly, we recently showed NRF2 to uniquely repress the expression of another ARDS candidate gene,
MYLK, via a novel mechanism involving the AREs33, 209, 276, 277, which again highlights the
crucial mechanistic involvement of NRF2 in ARDS.
Immune-Linked and Inflammation-Linked Pathways
Many innate immunity genes are implicated in severe lung injury that contribute to
neutrophil infiltration into the alveolar space and the cytokine storm182. In sepsis-induced
53 | Lynn ARDS, neutrophil-related genes (OLFM4, CD24, LCN2, BPI, RBP7, UTS2) are
significantly expressed compared in sepsis patients alone90. The most highly enriched
pathways in my analysis were related to immune signaling (Table 1) with the caveat that
many genes were shared between pathways. A strong 37-gene signature (ARG1, CAP1,
CAT, CCT8, CEACAM1, CEACAM8, CHIT1, COTL1, CRIPS3, CYBA, DEFA4, DNAJC5,
FABP5, FTH1, FTL, GPI, GSTP1, HBB, HSP90AB1, HSPA1A, HSPA8, LCN2, LGALS3,
MIF, MMP8, MMP9, OFLM4, PGAM1, PLAUR, PPIA, RNASE3, S100A8, S100A9) was
shared between ‘neutrophils’, ‘innate immune system’, and ‘immune system signaling’
pathways (Figure 6). The strong neutrophil/immune system signature highlights
dysregulation in neutrophil cellular biology is of primary importance in ARDS risk and
survival outcomes278. MMP8 and MMP9, two genes present in the ARDS survival
signature depicted in Figure 6 (unpublished data), are released by neutrophils as the site
of acute inflammation279, 280. Both MMP8 and MMP9 had increased protein levels in
models of lung injury233, 279, 280. Absence of MMP8 and MMP9 in MMP8-/- and MMP9-/-
mice exposed to VILI models showed decreased risk for acute lung injury279, 280. Both
MMP8 and MMP9 levels in BAL fluid correlate with increased lung injury278.
High-mobility group box 1 (HMGB1) is represented across all five of the top immune system pathways and in the IL-1 signaling pathway (Table 1, Figure 6, Figure
7). HMGB1 was identified as a cytokine in a murine model of endotoxin-mediated lethality281 and is upregulated in vitro under 18% cyclic stretch conditions of high
mechanical stress221. Like eNAMPT, HMGB1 binds TLR4 as well as RAGE, the primary
receptor of HMGB1195, 282. HMGB1 protein expression correlates with ARDS severity, 28-
day and 90-day mortality outcomes193, 263, 283-285.
54 | Lynn
Figure 6: Inflammation Pathways: Using the entirety of 201 genes identified by our eGWAS approach, we identified the top five enriched inflammation pathways: Neutrophils, Cytokine Signaling, Immune System, Innate Immune System, Leukocytes. Shown are the individual genes in each pathway with significant overlap1.
Figure 7: Five interleukin-associated signaling specific pathways (IL-10 Signaling, IL-4 and IL-13 signaling, Development of the ILC family, IL-1 Signaling, Leukocytes) and the genes that overlap1.
55 | Lynn Another important immunological cytokine in ARDS is macrophage migration
inhibitory factor (MIF), which is a product of activated T cells that inhibit macrophage
migration65, was initially described in the critical care literature via the use of ARDS animal models286. In humans, MIF levels in BAL were elevated in both ARDS and septic patients, and we identified two SNPs (rs755622 and rs2070767) in MIF to be associated with
African American ARDS patients65. In pathway analysis, MIF is present in immune
pathways (neutrophils, immune system signaling, innate immune system signaling),
hemostasis, and adipogenesis (Table 1).
IL-6 and IL-4 are well established biomarkers in ARDS67, 75, 76, 207, 215, 253, and
enrichment of interleukin signaling pathways in ARDS merits commentary (Figure 7).
Pathways for “IL-4 and IL-13 signaling’, ‘IL-1B signaling’, ‘IL-10 signaling’, development
of the ‘ILC’ family, and ‘leukocytes (general)’ were enriched in pathway analysis (Table
1). IL1B is involved in all pathways, and IL-1B is an early candidate biomarker whose
serum levels correlate with endothelial cellular injury75, 78. eNAMPT-driven gene pathways
following TLR4 ligation include ‘IL10- signaling’, ‘IL-6 signaling’, ‘leukocyte extravasation signaling’, and ‘toll-like receptor signaling’182. A total of 28 genes are involved in at least
one interleukin-related signaling pathway (AGER, AHR, ANXA1, AREG, CCL2, CDKN1A,
CEBPD, CSF2, CSF2RB, CXCL2, CXCL8, HMGB1, HSPA8, IL13, IL1B, IL1R2, IL1RN,
IL4, IL6, JUN, LCN2, MMP9, OSM, PTGS2, S100A12, SAA1, TNF, YWHAZ) (Table 1,
Figure 7). I have previously described the history of interleukins as early biomarkers and
candidate genes in ARDS, and this pathway analysis in Aim 1 links these crucial
interleukin biomarkers to other immune genes and pathways that have been associated
with ARDS.
56 | Lynn
Endothelial Vascular and Cellular Signaling Pathways
One of the most studied and diverse groups of genes responsible for the pathology
of ARDS are the highly conserved vascular signaling genes80. My ARDS pathway analysis identified eleven pathways involved with vascular biology relevant to ARDS, with the six most highly enriched Vascular and Cellular Signaling Pathways shown in Table 1 and Figure 8, and related genes from platelet-specific pathways shown in Figure 9.
Growth factors and coagulation factors are both up regulated in mRNA ontological studies across species (mouse, rat, canine, human)80. Of the 37 genes up-regulated in this cross-
species study, 5 were related to cell proliferation, 6 were related to wound healing, and 5
were related to extracellular spaces, and all were related to pro-fibrinolytic processes
associated with poor outcomes in ARDS80, 287. One of these genes, SERPINE1, encodes
PAI-1, a potential biomarker for ARDS80, 138. Signaling by VEGF (KEGG pathway map04370) is associated with ARDS in a large genomic ARDS study that yielded 44 significant genes of interest267 with 13 specifically involved in VEGF signaling, a critical
pathway for cellular proliferation in vascular signaling and acute lung injury267. Expression
of the receptor for advanced glycation end-products (RAGE) is correlated with severity in
ARDS patients192. RAGE is predominantly expressed in epithelial cells, and several
RAGE SNPs are potential ARDS risk SNPs288. I have previously described how NOS3,
IL1Β, NOX4, SERPINE1, and IL6 all have a history of associations with ARDS pathology,
57 | Lynn risk, and severity104, 246, 262. VEGF signaling pathway also triggers the downstream activation of many transcription factors like SP1, which regulates nmMLCK55.
Figure 8: Genes represented in top enriched endothelial vascular pathways: Shown are the top 6 cardiovascular signaling pathways and the number of genes in each pathway: Hemostasis, VEGFA-VEGFR2, MAPK Signaling, PI3K-Akt Signaling, AGE-RAGE.
58 | Lynn Thrombin (a pro-coagulation and increased mechanical stress ligand) signaling to
endothelial cells further exacerbates alveolar capillary barrier dysfunction in lung injury41,
42, 51, 160, 188. Because thrombin increases coagulation in addition to initiating barrier dysfunction in endothelial cells, I have considered coagulation in conjunction with genes that regulate vascular and endothelial pathways. Genes associated with platelet count and coagulation have been discovered to be indirect mediators of endothelial damage in
ARDS241. Five genes associated with platelet counts (BAD, LRRC16A, CD36, JMJD1C,
SLMO2) in a meta-analysis were studied in a larger population of ARDS and at-risk
controls240, 241, 289. Five pathways (‘hemostasis’, ‘platelet activation’, ‘signaling and
aggregation’, platelet degranulation, response to elevated platelet cytosolic Ca2+,
‘complement and coagulation cascades’, involved with platelet signaling or coagulation
(Table 1, Figure 9). In a canine model of lung injury, 7.4% of differentially regulated genes
were in blood coagulation pathways82. CPDB pathway analysis identified four separate
significant enrichment pathways involving platelets and coagulation (‘Hemostasis’,
‘Platelet activation, signaling and aggregation’, ‘Platelet degranulation’, ‘Response to
elevated platelet cytosolic Ca2+”) (Table 1, Figure 9). These pathways share 14 common
genes, which drive the platelet and coagulation pathway signal (Figure 9). Many of these
coagulation genes (ANXA1, APOA1, FGA, PPIA, SERPINA1) were identified in ARDS proteomic studies as well249. Genes involved in the sphingolipid generation and signaling
pathway, such as S1PR1 and S1PR3, are highly abundant in platelets and are also potentially novel biomarkers and risk SNPs49, 51, 52, 208.
59 | Lynn
Figure 9: Platelet and Coagulation Pathways: 14 genes are shared among top 4 enriched platelet and coagulation pathways.
60 | Lynn Other Transcription Factor and Signaling Pathways
Several important transcription factor and signaling pathways emerged from our
pathway analysis that did not fit neatly into either of the previous broad categories
discussed but still merit discussion (Table 1). Nuclear receptors directly interact with DNA as a ligand and the ‘Nuclear receptors meta-pathways’ represents a diverse group of genes that point to the overall importance of DNA regulation and transcription in ARDS.
Nuclear receptor meta-pathways are a nebulous category but present a large and highly significant pathway identified in this study (Table 1, p=6.42e-17). Of the 201 mapped
ARDS genes, 31 represent either nuclear receptors or key protein binding partners (Table
1) (ABCB1, AGER, AHR, APOA1, CCL2, CYP1A1, EGFR, FTH1, FTL, GCLC, GCLM,
GSR, GSTP1, HBEGF, HSP90AB1, HSPA1A, IL1B, JUN, KEAP1, NFE2L2, PDE4B,
FLK2, PRDX6, PTGS2, SERPINA1, SOD3, TGFB2, TNF, TNFAIP3, TXN, UGT2B7).
Expression of several genes (CCL2, EGFR, IL1B, JUN, SERPINA1, TNF) (mRNA) in
ARDS animal models was associated with lung injury risk81, 83, 130-132. Other transcription
factors (NFE2L2, KEAP1, PRDX6, SOD3)126, 211, 261, 274, 290, 291 are associated with ROS
pathways and overlap with ROS-focused gene signatures207. This group of transcription
factors should be viewed as important linkages between the other major categories of
ARDS biomarkers previously described (ROS, inflammatory, and endothelial), but should
also be considered in their own right as genes associated with ARDS associated risk,
endothelial barrier dysfunction, and initiators of the cytokine storm.
Critical Issues
61 | Lynn ARDS Genetic Variants Identified by Genome-Wide Association Studies. The
recent advances133 in identifying risk SNPs for ARDS present new therapeutic opportunities for ARDS clinical trials but also present challenges for validation and
replication across multiple cohorts in a heterogeneous genetic disease such as ARDS19,
67. Attempts to address the challenge of defining genetic risk factors involved in the
development of ARDS and the severity of the ARDS phenotype largely relies on two approaches: i) candidate gene studies and ii) genome-wide association studies133.
Candidate gene studies focus on specific gene(s) with probable biological and
mechanistic links to either vascular permeability, cytoskeletal protein dysregulation,
apoptosis pathways or pro-inflammatory cascades etc.242, 273. Genome-wide association
studies (GWASs) focus on genotyping the entire genome without requiring an a priori
hypothesis regarding specific genes or their biological significance292. GWASs benefit
from not requiring an understanding of mechanisms of gene involvement and allow for
the discovery of novel genes in ARDS133. Candidate gene studies have several limitations
that GWASs overcome, but candidate gene studies have the potential advantage of
facilitating defining SNP functionality. In complex diseases such as ARDS with a myriad
of environmental and genetic causes, GWASs are valuable as they may be performed without a complete mechanistic understanding of the many biological pathways involved292 with this agnostic approach allowing for discovery of unique genotype- phenotype relationships133. GWASs exploring ARDS risk are primarily divided between
European descent populations (80%) and African descent populations (20%)34, 79, 103, 104,
219, 293. Together, five GWASs have yielded eleven genes with fifteen independent SNPs associated with ARDS susceptibility in GWAS case-control studies (Table 2).
62 | Lynn Table 2: Top SNPs and Genes in ARDS Literature—GWAS Approach: ARDS Risk
Gene Predictive SNPs Population Study GWAS Risk SNPs
ABCC1 Rs3887893 (p=0.0001, meta) European descent (MGH, Boston 765 stage II ARDS trauma population,
MA) 838 stage II ARDS sepsis population;
direct vs. indirect ARDS association
and meta-analysis219
ARSD Rs78142040 (3.64e-47) ARDSNet 213 ARDS patients, 440 (379 EUR and
61 ASW) controls; Exome-seq103
case/control association235
FAAH Rs324420 (p=0.0131) European descent (MGH, Boston 765 stage II ARDS trauma population,
MA) 838 stage II ARDS sepsis population;
direct vs. indirect ARDS association
and meta-analysis219
HEATR1 Rs2115740(p=6.53x10-5, African American descent (Seattle 232 ARDS cases, 162 ICU controls104
unadjusted) WA and Chicago IL)
IL1RN Rs315952 (p=0.0023), rs380092 European descent (Philadelphia Association stage II (n=606) ARDS and
(p=0.026) PA) stage III (n=561) ARDS79
PDE4B Rs12080701 (p=0.0005, meta), European descent (MGH, Boston 765 stage II ARDS trauma population,
Rs17419964 (p=0.0002, meta) MA) 838 stage II ARDS sepsis population;
direct vs. indirect ARDS association
and meta-analysis219
POPDC3 rs1190286(p=0.0094) European descent (MGH, Boston 765 stage II ARDS trauma population,
MA) 838 stage II ARDS sepsis population;
direct vs. indirect ARDS association
and meta-analysis219
63 | Lynn PPFIA1 Rs471931(p=0.0021) European descent (Philadelphia Two stage GWAS; phase 1 compared
PA) 600 ARDS trauma-associated ALI,
2266 population based controls;
phase 2 compared 212 ALI cases and
238 at-risk controls96
SELPLG Rs109017898(p=1.5xe-04, African American descent (Seattle 232 ARDS cases, 162 ICU controls104
p=0.005, discovery, meta) WA and Chicago IL)
TACR2 Rs61732394(p=6.24x10-04) African American descent (Seattle 232 ARDS cases, 162 ICU controls104
WA and Chicago IL)
TNFRSF11A Rs9960450 (p=5.3x10-3,meta), European descent (MGH, Boston 765 stage II ARDS trauma population,
Rs17069902 (p=0.0001,meta) MA) 838 stage II ARDS sepsis population;
direct vs. indirect ARDS association
and meta-analysis219 XKR3 Rs9605146 (1.68x10-59) ARDSNet 213 ARDS patients, 440 (379 EUR and
61 ASW) controls; Exome-seq103
case/control association (161)
ZNF335 Rs3848719 (p=2.86e-04) ARDSNet 213 ARDS patients, 440 (379 EUR and
61 ASW) controls; Exome-seq103
case/control association (161)
GWAS Approach: Survival-Specific Risk SNPs
ADIPOQ Rs2082940 (p=0.0039) ICU (MGH and BIDMC, Boston) 2067 ICU patients; 567 ARDS patients;
prospective risk and mortality study294
FER Rs4957796 (p=0.0144) European descent (Gottingen, 441 total ARDS patients, 274 ARDS
Germany) patients with pneumonia; 90 day
survival; prospective case control232
Among European ARDS cohorts, nine ARDS genes, including XK-related 3
(XKR3), arylsulfatase D (ARSD), and Zinc-Finger/Leucine-Zipper Co-Transducer NIF1
(ZNF335), were identified (Table 2) by case-control whole exome sequencing of Asian
American and European American populations103. Another study in multiple ARDS
64 | Lynn populations (trauma- and sepsis-induced ARDS) of European descent, identified several
genes (POPDC3, FAAH, PDE4B, ABCC1, TNFRSF11A) to reach population-wide
significance in a meta-analysis219. Two additional independent ARDS studies in European
descent patients (Philadelphia, PA) had two genes (IL1RN, PPFIA1) reaching population-
wide significance in their respective case-control studies79, 96. IL1RN is linked to the development of both ARDS and sepsis72, 79; IL1RN levels were shown to be significantly
higher in ARDS patients compared to controls with IL1RN levels predicting mortality72.
Another immunity-related gene, TNFRSF11A, is a member of the tumor necrosis factor
receptor (TNFR) family, mechanistically important to developing ARDS295, 296. TNFR1
mediates cell death, which in turn leads to vascular leak and neutrophil infiltration of the alveolar space295. TNFRSF11A encodes the receptor activator of NF-kB (RANK) which is
the receptor for receptor activator of NF-kB ligand (RANKL), a key to altered NF-kB
signaling296, 297. While TNFRSF11A SNPs have been significantly associated with the
severity of Paget’s Disease, the reported ARDS risk SNPs are unique79, 296 and may
influence TNFRSF11A alternative splicing, a largely unexplored mechanism in ARDS220,
297, and NF-kB signaling296, 297.
GWAS Studies of ARDS Mortality. In the ARDS genetic literature, GWAS studies
are less common than candidate gene studies due to expense and the requirement for
larger patient populations to overcome the limitations of multiple association testing
(Bonferroni correction)133. Only two GWAS studies have evaluated ARDS with mortality as the severity outcome endpoint 227, 232, 293, 298, and both ARDS outcome studies were
conducted in European232 or European descent populations34, 227, 293. Each study had a
65 | Lynn single gene reach population significance for mortality association risk (ADIPOQ, FER,
ACE) with at least one SNP, although other SNPs in linkage disequilibrium were reported
(Table 2). ADIPOQ encodes adiponectin294 and has been linked in several meta-analyses
to type 2 diabetes or obesity299, 300 in Caucasians and South Eastern Asian populations299,
and with Type 2 diabetes in Chinese and Taiwanese populations294, 298, 300, 301. However, after adjusting for BMI (a measure of obesity) and for diabetes status, rs2082940 remained significantly associated with ARDS mortality294. These studies are the first
reported incidences of ADIPOQ being associated with ARDS risk, which highlights the importance of performing both case-control GWASs and outcome/severity based GWASs in ARDS.
Unlike ADIPOQ, ACE has a history of mechanistic studies that implicate angiotensin converting enzyme (ACE) as an important gene in ARDS pathology30, 32, 302.
ACE is the enzyme that degrades angiotensin I (Ang I) to angiotensin II (Ang II), which is
the peptide that is primarily responsible for maintaining blood pressure homeostasis and
fluid/salt balance in kidney filtration32. In animal models, ACE is strongly associated with
elevations in IL-6 and leukocyte counts30 and is elevated in human ARDS BALF30, 32. Ang
I has also been implicated as having an insertion/deletion (I/D) polymorphism that is
associated with mortality in an ICU ARDS cohort34. FER is a member of the FPS/FES
non-transmembrane receptor tyrosine kinase family and significantly associated with
ARDS outcome (Table 2) and survival in septic patients264. In a multi-cohort study, rs4957796 (a FER SNP) was significantly associated with survival in sepsis patients264
and with increased survival in ARDS232. Many ARDS patients have sepsis as a
comorbidity, and rs4957796 being both a SNP associated with mortality in ARDS and
66 | Lynn sepsis makes it a strong candidate gene to study in ARDS, and potentially useful in developing an ARDS risk SNP panel232, 264.
ARDS Genetic Variants/Genes Identified by Candidate Gene Studies.
Candidate gene studies require an a priori hypothesis of a gene’s function, which made candidate gene studies popular for studying genes with known pathological roles in
ARDS292. A limitation of candidate gene studies is a potential failure to account for genetic drift and population demographics on natural selection292. In ARDS, this is a particularly important limitation because of observed health disparities between African, Hispanic, and European populations10. The largest category of ARDS genetic studies are ARDS risk-association candidate gene studies31, 64, 104, 125, 187, 211, 238, 246, 303-305. Candidate gene studies are the original genetic approaches in the ARDS literature and are often associated with a gene or protein’s hypothesized role in lung injury133. The first reported candidate risk genes were SFTPB, ACE, and IL634, 125, 135, 293, 305. The advantage of candidate gene studies is that analyses can be performed in smaller populations with phenotypically-controlled case sizes to achieve significance and account for a fraction of
ARDS population heterogeneity when compared to GWAS approaches34, 125, 293. While candidate gene studies are numerous, replication of specific SNPs has proven difficult.
Larger study populations have produced more robust results for specific risk genes
(NFE2L2, NAMPT/PBEF, IL4, IL13, SP-B, AGER, PI3, MAP3K1, IL6, MYLK) and specific
SNPs of interest (Table 2). Among candidate gene risk studies, eighteen of twenty-three studies (~78%) have been performed exclusively in European populations or populations of European descent31, 64, 104, 125, 187, 227, 238, 303-305.
67 | Lynn The early research into candidate genes provided viable genotype-phenotype links
for biomarker studies, and one of the most successful has been IL-6293. A haplotype of
IL-6, -174G/C, has been identified and validated as a risk SNP for ARDS several case- control studies135, 136. Increased levels of interleukin-6 (IL-6) cause a rise in reactive oxygen species (ROS) via the TLR4-TRIF-TRAF6 pathway217. IL-6 is elevated in patients with ARDS and has been shown to have a significant role in the permeability of the lung
endothelium in multiple ARDS mouse models76. IL-6 has several risk SNPs and promoter
haplotypes associated with sepsis-related acute lung injury (ALI)76, 306. IL-6 levels are
determined by many genetic factors, and the SNPs associated with ALI in sepsis patients
were discovered in a Hispanic population306. Most importantly, two of the SNPs (-597G/-
174G) are associated with a risk haplotype136. A strong phenotype-genotype relationship
with IL-6, genotypes, clinical outcomes, and ARDS severity or mortality has been
documented, and the case for IL-6 genetics playing a role in ARDS severity risk and
mortality are strong136, 293, 306.
Candidate Gene Studies of ARDS Mortality. Similar to GWASs in ARDS, there
are more candidate gene studies that focus on ARDS risk over outcome severity, but
there is a higher proportion of candidate gene studies that have evaluated ARDS mortality and associated outcome risk gened187, 231, 294. When ARDS patients are stratified by the
Berlin definition (mild, moderate, and severe), the more severe ARDS patients are
significantly more likely to die from ARDS than patients with mild or moderate ARDS307.
Thus, patients that die from ARDS can be argued to have severe ARDS307. Of the four
candidate gene studies that reported on mortality, three genes (75%) were obtained in
68 | Lynn European descent populations, and one population is from a pediatric, Brazilian
population (Table 3). In the European populations, NAMPT/PBEF, IL-1β, and PHD2 were
identified as each having at least one SNP (NAMPT/PBEF has four) that are associated
with ARDS mortality231, 246, 294. In the Brazilian pediatric ARDS population, TNF had two
SNPS that were associated with death in septic and ARDS populations294.
Two candidate genes (TNF and IL-1β) were chosen for their long history as ARDS biomarkers75, 204, 227. IL-1β transcription is caused by stress and endotoxin triggers and is
secreted by macrophages, thrombocytes, and injured endothelial cells71, 204. The
promoter for IL-1β includes NFkB sites and activating protein-1 (AP-1) sites71. In an
attempt to link biomarkers to genotype and establish a genotype-phenotype relationship
in IL-1β, a significant SNP was found in the IL-1β promoter region (-511 upstream from the TATA box and transcription start site)71, 246. The site found to be related to ARDS and
sepsis mortality was previously reported to be an important site for the secretion of IL-
1β78.
Table 3: Top SNPs and Genes in ARDS Literature—Candidate Gene Approach:
ARDS Risk
Gene Predictive SNPs Population Study Candidate Gene Risk SNPs
ACE DD (p=0.00004, healthy European descent (University of 88 respiratory failure patients, 174 GABG
population, p=0.0008 CABG College London Hospitals) controls34
AGER Rs2070600ti t ) (A/A, Ser/Ser, European descent (Clermont- 59 ARDS, 405 controls; log rank test with
p<0.0001) Ferrand, France) case/control304
AGT Rs699 (0.028, dom) European descent (Moscow, 68 NP ARDS cases, 198 NP controls285
Russia)
69 | Lynn AhR Rs2066853 (0.0012, dom) European descent (Moscow, 68 NP ARDS cases, 198 NP controls285
Russia)
CYP1A1 Rs2606345 (0.0027, dom) European descent (Moscow, 68 NP ARDS cases, 198 NP controls285
Russia)
DIO2 Rs12885300 (p=0.039) African and European descent 327 European Americans: 139 sever
Rs225014 (p=0.009) (American-European sepsis, 78 severe sepsis+ARDS/ALI, 188
Consensus Criteria) controls; 261 African Americans: 78
severe sepsis, 41 severe
sepsis+ARDS/ALI, 187 controls308
EGF Rs4444903 (p=0.005, males), European descent (MGH, 416 ARDS cases (246 survivors, 170 died),
rs2298991 (p=0.019, males), Boston MA) 1052 ICU controls; 887 males and 581
Rs7692976 (p=0.005, males), females303
Rs6533485 (p=0.025, males)
GADD45a Rs581000 (p=0.009) African Americans Chicago African American Chicago cohort: 71
Study and Spanish Study severe sepsis, 40 sepsis+ALI, 182
(Chicago, IL) controls; Spanish cohort: 80 severe
sepsis, 66 sepsis+ALI, 95 controls59
IL-13 1 SNP (431 A>Gr, p=0.008) European descent (Moscow, 347 controls, 74 ARDS cases; logistic
Russia) regression adjusted case/control187
IL-4 1 SNP (-589 C>T, p=0.01) European descent (Moscow, 345 controls, 72 ARDS cases; logistic
Russia) regression adjusted case/control187
IL-6 -174G/C allele (p=0.03) European descent Twin study on lung function, 427 twins
(232 women, 195 men)76, 306
MAP3K1 Rs832582 (p=0.01, rec) FACTT and ARDSNet (Seattle 241 ARDS patients, 346 healthy, locally
WA) matched controls235
MIF 2 SNPs (rs755622, p=0.03; European descent, African 288 European: 113 severe sepsis, 90
Rs2070767, p=0.04) descent sepsis- associate ALI, 85 healthy controls;
218 African: 69 severe sepsis, 61 sepsis-
associated ALI, 88 healthy controls65
MYLK Rs820336 (p=0.002) European and African descent European: 92 ALI, 99 sepsis, 85 healthy
Rs936170 (p=0.009) (John Hopkins University and controls; African: 43 ALI, 51 sepsis, 61
Rs936170 (p=0.025) Medical College of Wisconsin) healthy control176
70 | Lynn NAMPT/PBEF 2 SNPs (-948, p=0.015 European descent 374 ARDS patients and 787 at-risk
-2422, p=0.03) controls; nested case control154
NFE2L2 7 SNPS (p-values: 0.0069- Spanish Network 321 severe sepsis and ARDS; 871
0.0089) population-based controls; case control126
PI3 Rs1983649 (p=0.034, add), European descent (MGH, 449 ARDS patients, 1031 at risk controls;
Rs2664581 (p=0.004, 0.023, Boston MA) Case control64, 238
add, dom)
S1PR3 2 SNPS (rs7022797, p= 0.017; European and African descent 71 European ARDS and 24 African ARDS;
Rs11137480, p= 0.042) (Chicago, IL) 186 European controls, 185 African
controls52
SP-B 1 SNP (606-bp variant allele, MGH (Boston, MA), European 72 ARDS cases, 117 controls; nested
1580 C/T) descent (German) case/control199
52 ARDS patients, 46 healthy controls125
TLR1 1 SNP (rs5743551, p=0.002) European descent (Seattle, WA) 138 severe sepsis ARDS, 107 ALI, 167
healthy controls81
Candidate Gene Approach: ARDS Survival-Specific Risk SNPs IL-1B 1 SNP (-511 G>A, p=0.019) European descent (Moscow, 321 controls, 91 mortality cases; adjusted
Russia) logistic regression for case/control
mortality246 NAMPT/PBEF -1001G (p=0.001) European descent 374 ARDS patients and 787 at risk controls;
-1543T (p=0.03) nested case control68
PHD2 1 SNP (rs516651, p=0.002) European descent (Duisburg- 264 ARDS (70 deceased); case/control231
Essen, Germany)
TNF TNF-308 (p=0.0006) Brazilian septic and ARDS 490 septic and ARDS patients; 610
TNF-863 (p=0.01) pediatric patients controls294
ARDS Risk SNPs in African Americans. Racial and ethnic disparities in ARDS mortality and disease susceptibility have been reported for nearly fifty years10, yet genetic studies in ARDS have focused mostly on larger, European cohorts. While this has provided a strong foundation for the understanding of ARDS genetics, population diversity among ARDS should not be discounted in the sub phenotyping, diagnosis, and treatment
71 | Lynn of the critically ill. Understanding genetic population diversity in ARDS is critical because
there is a significant difference in mortality rates between European and African descent
populations3, 10. Across multiple age populations (until the age of 65), African Americans
have significantly higher rates of both sepsis and ARDS than their matched European
American cohorts3, 7, 9, 11, 309, greater duration on mechanical ventilation than European
Americans10, and a higher risk of ARDS mortality7, 11, 309 when compared to age-matched
European American counterparts.
The health disparity borne by African Americans in ARDS and severe sepsis warrants additional research and attention to the role of genetics in identifying unique plasma and genetic biomarkers for African Americans at risk for ARDS. Several unique candidate genes (MYLK, HEATR1, MIF, GADD45a, DIO2, SELPLG, and S1PR3) are promising genetic markers for increased risk of ARDS and ARDS mortality susceptibility among African Americans (Figure 10)49, 51, 65, 104, 152, 167, 170, 176, 308. In the case of MYLK,
a risk haplotype was identified consisting of coding SNPs with one of these SNPs verified
in other inflammatory disorders including sickle cell disease and severe asthma in African
Americans175. The functionality of this MYLK SNP has been shown to cause a delay in restoration of the vascular barrier in inflammatory models as well as to cause secondary mRNA structure alterations which promote excessive expression of this major cytoskeletal regulatory protein166, 167, 170, 174, 175.
Our group recently conducted a GWAS study of African American ARDS patients
and ICU controls that was underpowered but after innovative pathway prioritization,
discovered three novel genes that achieved genome-wide significance104. Unlike in the
European American ARDS studies, the population size in this study was relatively small
72 | Lynn (n=232) (Table 2, Table 3, Figure 10)104. Two unique risk SNPs for ARDS (rs2115740,
HEATR1; rs109017898, SELPLG) were found in this African American GWAS study in two genes that had not previously been identified as ARDS risk genes104. Further, higher- powered GWASs in non-European populations may potentially provide more novel genes and SNPS that are risk factors for ARDS in other under-studied populations.
Figure 10: Risk Genes in African Descent and Hispanic Ethnicity. Depicted are the 7 top risk SNP-harboring Genes that are unique to ARDS risk in African descent individuals and to ARDS survival. The genes were identified via interrogation of GWAS results (HEATR1, SELPLG) or via candidate gene approaches and sequencing (S1PR3, MYLK, DIO2, GADD45a).
73 | Lynn
Future Directions
ARDS is a severe, high mortality, complex and heterogeneous critical illness
influenced by environmental and genetic factors. In Aim 1, I have collated the available preclinical and human ARDS literature and identified 201 pooled ARDS candidate genes
(Supplemental Table 2) in a multi-database approach. While I highlighted ARDS risk
SNPs from both candidate gene studies and GWASs, pathway analysis allowed genes without known SNPs but reported mRNA and notable protein fold changes to be included in our pathway analysis80, 207, 287. My pathway analysis strategies revealed results that were consistent with the concept that evolutionarily-conserved inflammatory and ROS
networks and vascular gene dysregulation are potent contributors to ARDS
pathobiology83, 84. A broader ‘omics’ approach to ARDS allows for the focus on biologically-relevant pathways and genotype-phenotype connections between established ARDS biomarkers and differentially expressed ARDS risk genes.
I have also chosen to evaluate ARDS studies that use mortality as an endpoint as this captures the most severe outcome for ARDS patients. I summarized the evidence from genetic studies in diverse populations that have the potential to uncover novel biomarkers for ARDS risk and mortality and potential therapeutic targets in ARDS. I highlighted information relevant to the role of genetic factors in ARDS susceptibility and mortality7 that address the well-known health disparities that exists in susceptibility to and
mortality from ARDS7, 310, 311. Improved strategies for sub-phenotyping of diverse ARDS patients via molecular signatures or SNP panels will facilitate the potential for successful
74 | Lynn clinical trials in ARDS and yield a better fundamental understanding of ARDS pathobiology90.
75 | Lynn Chapter 3: Aim 2: NAMPT Haplotypes and Plasma eNAMPT Levels Predict
ARDS Mortality Risk
Significance
Acute respiratory distress syndrome (ARDS) is a clinical condition triggered by
diverse direct and indirect injurious challenges to the lung resulting in diffuse inflammatory
injury to the alveolar lung barrier3. The rate of ARDS cases is ~400,000 persons annually
with mortality rates in excess of 30%3, 309. Detailed phenotyping of ARDS subjects has been challenging. Although ARDS severity is significantly predictive of survival using
Berlin criteria for ARDS severity (mild, moderate, severe ARDS)309, a sub-phenotyping tool with therapeutic utility has yet to emerge. A clinically-useful biomarker in a causal pathway, with the potential to identify specific ARDS sub-phenotypes and therapeutic targets, would be significantly beneficial in promoting effective clinical trial designs for novel therapeutics. With the advent of COVID-19, therapeutics for ARDS and VILI injury and improved clinical trial design becomes more crucial28, 37, 38, 312.
We have previously utilized genomic–intensive approaches to identify potentially
novel therapeutic targets in acute inflammatory lung disorders including ARDS59, 65, 66, 68,
83, 84, 104, 135, 153, 170, 182, 183. These studies identified the gene encoding nicotinamide
phosphoribosyl transferase (NAMPT) as a novel candidate gene in ARDS68, 83, 153, 154, 181.
NAMPT encodes a protein that exhibits dual functionality—as an intracellular enzyme
(iNAMPT) and, when secreted, as an extracellular inflammatory cytokine (eNAMPT)
resulting in NAMPT being characterized as a “cytozyme” with both cytokine (extracellular)
and enzymatic properties (intracellular) that regulate stress responses154, apoptosis186, 313
76 | Lynn nicotinamide adenine dinucleotide (NAD) synthesis313, 314, metabolism314, tissue
hypertrophy and remodeling145, 315, and multiple cancer responses316-325. We first
demonstrated that secreted eNAMPT is a damage-associated molecular pattern protein
(DAMP) and ligand for Toll-like receptor 4 (TLR4), potently stimulating dysregulated
inflammatory responses that contribute to lung inflammation182, 183. The magnitude of
acute lung injury was significantly decreased in a NAMPT heterozygous+/- mouse model
of ventilator-induced lung injury (VILI)74, 182 and eNAMPT levels directly contribute to the
development of LPS-induced lung injury and VILI182. The reported elevation in plasma
eNAMPT levels in human ARDS cohorts and association with ARDS severity63, 84, 326
provide promising evidence for eNAMPT as a potentially clinically-relevant biomarker in
ARDS.
Our prior mechanistic studies identified promoter regions and specific transcription
factors that regulate NAMPT promoter activity in response inflammatory stimuli. These
include key roles of excessive mechanical stress and hypoxia both in humans and
preclinical models with identification of a NAMPT promoter region sensitive to mechanical
stress (-3028 to +1 transcription start site)68, 153, 182, 183, 187. NAMPT expression also
involves genetic and epigenetic regulation and multiple CpG sites are demethylated in
response to 18% cyclic stretch (18% CS) or lipopolysaccharide (LPS) challenges that
increase NAMPT transcription63, 222, 326. Four previously reported NAMPT promoter SNPs
are sites for binding of several transcription binding factors including (STAT5 [signal transducer and activator of transcription 5), NFкB (nuclear factor kappa-light-chain- enhancer of activated B cells), HIF2α (hypoxia-inducible factor-2 alpha), SOX17 and
SOX18 (SRY-box transcription factor 17/18)151, 153, 183, 314. Importantly, a two-SNP NAMPT
77 | Lynn haplotype (-1001/-1535 G/C) markedly increased risk for ARDS and sepsis while
conferring increased risk for ARDS mortality in primarily European descent ARDS
cohorts68, 187, 327.
Given the participation of the highly targetable eNAMPT/TLR4 inflammatory
signaling pathway in ARDS, we posited that the establishment of a genotype-phenotype link between NAMPT promoter SNPs and eNAMPT biomarker levels to ARDS clinical outcomes may have clinical utility. In this study, we show that specific NAMPT genotypes that regulate NAMPT promoter activity and eNAMPT secretion into the circulation confer increased mortality risk in ARDS (with additional ARDS mortality risk in African
Americans). As a component of a mortality risk score that integrates genotype-phenotype relationships, plasma eNAMPT levels and covariate data, the ARDS mortality index may enhance eNAMPT’s role as an effective biomarker and potential therapeutic target in
ARDS and facilitate the selection of high-risk subjects for ARDS clinical trial stratification
thereby increasing therapeutic trial success in this vexing critical care illness.
Methods
Demographics of the ARDS Genotyping and Biomarker Cohort. A total of 474 ARDS patient samples were utilized in this study. These samples represent six unique cohorts in addition to five academic health centers within the Fluid and Catheter Treatment Trial
(FACTT)—a multi-centered randomized controlled trial performed by the NHLBI
ARDSNet61, 87, 225, 327, 328. We obtained 28-day mortality outcomes on 428 patients.
78 | Lynn Patient Sample Collection. DNA derived from peripheral blood mononuclear cells
(PBMCs) were collected in accordance with their respective IRBs and transferred with
appropriate material transfer agreements. Blood was collected within 36 hours of ARDS
onset defined as when all Berlin ARDS Criteria were met. Blood was collected in EDTA-
treated tubes, centrifuged for 1 hour (2000 x g for 20 min, RCF) and the platelet-depleted
plasma stored at -80°C.
Detection of Secreted NAMPT. A total of 50 uL of plasma was utilized for an enzyme-
linked immunosorbent assay (ELISA) to quantify plasma levels of eNAMPT as we have
previously described62, 63, 66. A Nunc MaxiSorp 96-well plate was coated with 100 uL/well
of proprietary polyclonal goat anti-NAMPT antibody (8 µg/ml) diluted in the coating buffer
(1.5 g Na2CO3, 2.93 g NaHCO3 /1L distilled water, pH 9.6). The plate was placed overnight at 4 °C. After 24 hrs, the plate was placed at 37 °C for 1 hr, followed by 3 times washing in 1x TBS-T (0.1%). The plate was incubated with 100 uL/well of 1% BSA-TBS
for 1 hr at 37 °C to block any non-specific binding. The plate was further washed 1 time
with 1x TBS-T (0.1%) followed by incubation with 100 uL/well of either i) proprietary
human rhNAMPT as standard (100 to 0 ng/mL, 8 concentrations utilized for standard
curve) or ii) with the human plasma samples diluted 1/10 in plasma diluent buffer
(Cygnus). The plate was further incubated overnight at 4 °C, and the next day it was
placed 1 hr at 37 °C before 3x wash in 1x TBS-T. Culture medium was incubated with
100 uL/well of the polyclonal rabbit anti-NAMPT antibody (Bethyl, dilution 1:10,000 in 1%
BSA-TBS) for 1 hr at 37 °C, followed by 3x wash in 1x TBS-T. The plate was incubated
with 100 uL/well of the secondary donkey anti-rabbit HRP-labelled antibody (1:10,000 in
79 | Lynn 1% BSA-TBS) at 37 °C for 1 hr (followed by 3x wash in 1x TBS-T). The plate was developed with 100 uL/well of the HRP substrate (SIGMAFAST OPD) for 5 min at room temperature. The reaction was stopped with 100 uL/well of 10% H2SO4. The plate was read at 490 nm on BioRad iMARK (Hercules, CA, USA) and the standardization of results for this assay have been previously described154.
NAMPT Genotyping. DNA was extracted from PBMCs stored in RNA later or Triazol
according to standard protocol (Qiagen DNA Mini kit: Hilden, Germany). Eight NAMPT
SNPs were targeted for genotyping in 474 patients using the Agilent MassArray platform
(Agena Bioscience, San Diego, CA, USA). These SNPs were selected on the basis of
CpG sites, transcription factor binding sites, minor allele frequency in populations with
European and African ancestry and known associations with ARDS risk 329-331
(Supplemental Table 2). A second round of genotyping in 173 ARDS patients with
available eNAMPT plasma was performed using the same protocol and platform. Seven
of eight SNPs passed genotyping quality control (QC metrics included genotyping success rate of >90% per SNP, >50% per individual, and Hardy-Weinberg equilibrium
(p>0.001)). There were 416/474 ARDS patients that were successfully genotyped and had available outcome data for 28-day mortality. Haplotypes for NAMPT SNPs (those that passed initial genotyping quality control) were calculated via a contingency table.
Endothelial Cell Culture. Human lung endothelial cells (Lonza Inc. Allendale, NJ) were cultured utilized as previously described66. Human pulmonary artery endothelial cells
(HPAECs) (Lonza Walkersville, MD, USA) were grown at 37°C in a 5% CO2 incubator to
80 | Lynn 70-75% confluence according to the manufacturer’s protocol. HPAECs were either
treated with i) pGL3 sham vector or ii) -1535 A vector. HPAECs were treated with nothing
or 100 ug/mL lipopolysaccharide (LPS) for 18 hrs. Sham and 1535 A cells (both untreated
and treated with LPS) were subjected to 4 hours of 18% cyclic stretch (0.5 Hz) on Bioflex
collagen I type cell culture plates (FlexCell International, Hillsborough, NC) on the FX-
5000 System to mimic high tidal volume as we have previously described154.
NAMPT Promoter Activity. NAMPT promoter activity in human lung endothelial cells
was studied using the promoter construct as we have previously described154. The
Promega Dual Luciferase assay (Promega, Madison, WI, USA) was followed according
to protocol to assess NAMPT promoter activity in sham vector control and NAMPT SNP-
transfected endothelial cells. Six (n=6) duplicate luciferase assays were performed per
condition and per the manufacturer’s instructions and as previously described154.
Luminescence (RLU) ratios were normalized to log2 untreated control (pGL3 with
rs61330082 G).
Statistical Analysis. Of the 474 genotyped ARDS subjects, 28-day mortality outcome data was available for 428 subjects (Table 4). Statistical was performed in STATA 15.1
(StataCorp. LLC: Release 15, College Station, TX) and in R (R Core Team, 2020). An
additive genetic model for rs61330082 (-1535) (Supplemental Table 3) was utilized for
a 28-day mortality logistic regression model. We utilized an additive genetic model of
rs61330082 to test association with APACHE II Scores. Linear regression was also
performed to determine the association between APACHE II Scores and eNAMPT
81 | Lynn baseline plasma levels. A two-way ANOVA was used to assess the interaction of self-
reported race and mortality on plasma eNAMPT levels. A Bonferroni adjusted pairwise
comparison of means between the four rs61330082/rs59744560 haplotypes (calculated
via contingency tables) was performed; an adjusted linear regression model with these four haplotypes and covariates was also performed. Genetic risk scores are statistical
methods developed to assess risk across multiple genotypes 107, 332, 333. A mortality risk
score using the rs61330082/rs59744560 haplotypes was developed based upon previous
methods utilizing genetic risk and hazard scores 107, 333. Logistic regression for 28-day
mortality was used to calculate odds ratios and area under curve (AUC) for all models.
Receiver operating characteristic (ROC) analysis was performed to compare the AUCs
of the relevant models.
82 | Lynn Results
Table 4: Demographics of ARDS patients genotyped for 7 SNPs on the NAMPT promoter ARDS cohort (n=428) P-value Survived (n=297) Deceased (n= 131) Sex (Male, Female) (171, 123) (76, 54) 0.95* Age, Median [Q1, Q3] 55 [44, 72] 52 [43, 62] 0.74‡
APACHE II, Median [Q1, Q3] (n) 67 [27, 98] (203) 40.5 [23, 121] (118) 0.58‡ African American APACHE II, 79.11 [69.76, 88.45] (76) 87.65 [72.52, 102.80] (46) 0.34 Median [Q1, Q3] (n) European American APACHE II, 59.40 [52.39, 66.42] (127) 58.31 [45.58, 71.03] (72) 0.88 Median [Q1, Q3] (n) APACHE III, Median [Q1, Q3] (n) 74 [53, 96] (90) 105 [96, 139] (12) <0.01‡ Self-reported race, N African American 166 59 0.04* European American 131 72 Sepsis, N 202 99 0.12* Plasma eNAMPT levels (ng/mL) 54.58 [29.14, 73.60] (n=148) 54.98 [32.03, 79.75] 0.74‡ Median [Q1, Q3] (n=74) *Pearson’s Χ2 test ‡Two-sample t-test with equal variance
NAMPT Promoter Genotypes and eNAMPT Plasma Levels in ARDS Cohorts. A schematic presentation of NAMPT promoter (TSS to -3020 bp upstream) in Figure 11
depicts the genomic location of the 7 NAMPT promoter SNPs (rs7789066, rs116647506,
rs61330082, rs151175330, rs9770242, rs59744560, rs1319501) that survived quality
control (QC) metrics. Minor allele frequencies (MAFs) for these 7 SNPs were calculated
for control populations from the 1000 genomes database for both European and African
populations and for ARDS subjects (Supplemental Table 4). Table 4 details the clinical
and demographic features of the 428 ARDS subjects from six unique sources and the
FACTT trial with 28-day mortality outcomes334. Sex, age, APACHE II score, septic status,
83 | Lynn and eNAMPT plasma values were not significantly different between survivors and
deceased ARDS patients in this cohort (Table 4). The proportion of deceased and
surviving European vs African Americans was significantly different as was APACHE III
score (Table 4). Similarly, plasma eNAMPT levels were not significantly different in ARDS patients that survived vs those who died (p=0.74) (Table 4). There was also a significant
difference between African American (n=72, mean=63.70 ng/mL [54.04, 73.36]) and
European American (n=150, mean=51.68 ng/mL [46.97, 56.39]) (p=0.01) eNAMPT
levels. Overall cohort mortality was 31% (n=131). The overall experimental design that
84 | Lynn utilizes outcome variables, eNAMPT values, and NAMPT SNPs are detailed in Figure
12.
Figure 11: Selection of NAMPT promoter SNPs for genotyping in a combined ARDS cohort. The NAMPT functional promoter length was determined in a prior study utilizing deletion construct analysis and resides between -3028 to +1 (TSS, transcription start site). SNPs in red have been previously associated with ARDS risk. SNPs in yellow are novel targets selected via: i) the capacity to alter transcription binding site/CpG sites, ii) a higher prevalence in African Americans with ARDS, or iii) location in a mechanistically-relevant gene region (Supplemental Table 3).
Figure 12: Experimental Design. 474 ARDS patients were genotyped and 28-day mortality outcome values available for 428 subjects (Table 4). A total of 222 ARDS subjects had 28- day mortality outcome values and plasma available for eNAMPT measurements. Quality control (detailed in methods) of SNP genotyping produced the totals used for haplotype analysis.
85 | Lynn
The NAMPT Promoter SNP rs61330082 (-1535) is Significantly Associated with
ARDS Mortality Risk. Of the 7 SNPs, rs61330082 (-1535) showed a cohort-wide trend
for mortality (p=0.036) in an additive genetic model (Table 5) which was determined to
be the most robust model (Supplemental Table 5) and utilized in an additive logistic
regression model (Table 5). For 28-day mortality, genotype rs61330082 GA (OR= 0.46
[0.24, 0.88], p=0.019) showed a significant protective survival effect in this ARDS cohort
adjusted for relevant covariates (Table 5).
Table 5: rs61330082 is associated with lowered mortality risk in ARDS p-value* rs61330082 additive genetic model# Genotypes Survived (n) Deceased (n) 0.036 OR [95% CI]* P-valuea,# GGb 193 82 -- -- GA 84 29 2.06 [0.66, 6.46] 0.214 AA 8 10 0.46 [0.24, 0.88] 0.019 a Likelihood ratio test b Reference genotype # n=406, R2=0.077, p=0.001; adjusted for age, race, sepsis status, sex, severity, and center *Pearson’s χ2
APACHE II (Acute Physiology And Chronic Health Evaluation II) score is one of a series
of severity-of-disease classifications utilized in the ICU21, 197, 335. At the time the majority
of these patients were admitted into their respective studies, APACHE II was the score
utilized to quantify risk (higher score equates to higher risk of mortality). An additive
genetic model utilizing linear regression of APACHE II scores showed a significant
decrease in APACHE II scores associated with rs61330082 A (minor allele) (p=0.012)
(Figure 13).
86 | Lynn
Figure 13: rs61330082 genotypes show a significant decrease in APACHE II Scores. An additive genetic model utilizing linear regression of APACHE II scores showed a significant decrease in APACHE II scores associated with rs61330082 genotypes. n=310, p=0.012, R2=0.018, y(APACHE II)= -11.34(x) + 73.43
87 | Lynn
NAMPT Promoter SNP rs61330082 (-1535) Functionally Alters LPS- and Mechanical
Stress-induced NAMPT Promoter Activity. The rs61330082 A allele showed a
decrease in NAMPT promoter activity compared to G after endothelial cell exposure to
LPS (n=6 per challenge group, p=0.008) (Figure 14A). After a 4-hour exposure to 18%
cyclic stress, rs61330082 A showed a decrease in NAMPT promoter activity with and
without LPS challenge compared to G (fold change=1.26, p<0.001) (Figure 14B). The
rs61330082 A also showed a decrease in NAMPT promoter activity in a combined
LPS+18% cyclic stretch exposure condition compared to rs61330082 G (fold
change=3.11, p<0.001) (Figure 14B).
Figure 14: NAMPT promoter activity is increased by the rs61330082 variant in human endothelial cells following LPS challenge. Human lung endothelium were transfected with the rs61330082 major allele (G) or minor allele (A) and grown to confluency prior to challenge with LPS 100 ng/mL (n=6 for all groups). RLU activity was normalized to log2 values of control (rs61330082 A, static conditions). A. Samples treated with LPS in static conditions showed decreased NAMPT promoter activity for rs61330082 A compared to controls (p=0.008). B. Endothelium subjected to either 18% cyclic stretch or to 18% cyclic stretch and LPS (for 22hrs) show decreased NAMPT promoter activity (p<0.001) in rs61330082 A-containing endothelium compared to control-transfected endothelium.
88 | Lynn eNAMPT Levels are Associated with APACHE II Scores and rs61330082 Genotypes.
While baseline eNAMPT levels are not associated with mortality, higher eNAMPT levels were found to be significantly correlated with higher APACHE II scores (Figure 15). (None of the patients with APACHE III scores had available blood plasma for eNAMPT measurements.) We investigated the rs61330082 genotype and its relationship to both eNAMPT plasma levels and APACHE II scores (Table 6). While no genotypes (GA and
AA) were significantly related to changes in eNAMPT values compared to the reference genotype (GG), mean APACHE II scores for the minor allele genotype (AA) were significantly decreased compared to the reference genotype (GG). In an adjusted linear regression model, rs61330082 AA continues to be associated with lower APACHE II scores (Table 6). The rs61330082/rs9770242 GG/CC diplotype is associated with significantly higher eNAMPT levels in African Americans compared to European
Americans (p=0.014) (Figure 16). Race and mortality interactions for the rs61330082/rs9770242 GG/CC diplotype suggests rs61330082/rs9770242 GG/CC is a significant risk genotype for higher eNAMPT levels in African Americans (p=0.045)
(Figure 16).
Table 6: rs61330082 (-1535AA) is associated with lowered APACHE II score APACHE II Score linear regression model# Genotype Mean eNAMPT p- Mean APACHE II p- β [95% CI]a# p- * ng/mL [95% CI] value† Score [95% CI] value value † rs61330082 58.11 [52.31, -- 72.12 [65.58, 78.67] ------b GG 63.90] (n=130) (n=193) rs61330082 50.84 [42.71, 0.392 67.63 [58.09, 77.16] 0.829 1.56 [-9.71, 12.83] 0.785 GA 58.97] (n=66) (n=91)
89 | Lynn rs61330082 50.34 [32.69, 0.796 35.94 [13.88, 58.00] 0.006 -26.85 [-48.81, - 0.017 AA 68.00] (n=14) (n=17) 4.90] a Regression coefficients # APACHE II score adjusted linear regression, n=298, p<0.001 R2=0.128, adjusted R2=0.116, adjusted for age, sex, race, and septic status *adjusted p-value †pairwise comparisons of means, Bonferroni adjusted
Figure 15: Higher eNAMPT levels are significantly correlated with APACHE II scores. A linear regression model was utilized, n=217, R2=0.111, adjusted R2=0.106, p<0.001, y(eNAMPT)=0.24+36.74
90 | Lynn
Figure 16: ARDS subjects harboring the rs61330082/rs9770242 GG/CC diplotype exhibit significantly increased eNAMPT plasma levels and mortality risk in African Americans. A) The rs61330082/rs9770242 GG/CC diplotype is associated with higher eNAMPT plasma levels in African American ARDS subjects (p=0.014). B) rs61330082/rs9770242 GG/CC diplotype is not significant for mortality (p=0.811), however, a two-way ANOVA to assess the interaction of race/ethnicity and mortality on plasma eNAMPT levels determined the interaction to be significant (F=3.15, p=0.045).
Optimized ARDS Mortality Prediction Model: eNAMPT Plasma Values and a rs61330082/rs59744560-based Haplotype Risk Score. While haplotypes and elevated
eNAMPT levels have been investigated in relation to ARDS risk154, 185, 222, our study is the
first to examine both in conjunction and to mechanistically assess the potential inter-
relationship. While there were no association was observed between rs59744560 and
eNAMPT levels, the rs61330082/rs59744560 A/C (1535 min/948 Maj) haplotype was
associated with decreased eNAMPT plasma levels (Table 7). Logistic regression for 28-
day mortality outcome was performed in four models: i) relevant covariates (detailed in
Table 8 and Figure 17), ii) rs61330082/rs59744560 haplotypes, iii) eNAMPT baseline
91 | Lynn plasma levels, and iv) a combined model. Receiver operator characteristic (ROC)
comparison was used to determine the best model for a predictive risk score for ARDS
mortality (Table 8, Figure 17). A mortality prediction model that included relevant
covariates, genetic risk, and eNAMPT plasma levels had higher predictive value
(AUC=0.645 [567, 0.722]) compared to any of the factors in their own respective models
(p=0.002) (Figure 17).
Table 7: rs61330082/rs59744560 A/C is a haplotype associated with decreased eNAMPT levels in ARDS patients Adjusted eNAMPT linear model* Haplotype Mean eNAMPT (ng/mL) P-value† β [95% CI]b P- (rs61330082/rs59744560) [95% CI] valuec
G/C (Maj/Maj)a 61.72 [54.23, 69.20] (n=94) ------
A/C (min/Maj) 47.80 [40.80, 54.80] (n=68) 0.009 -9.25 [-17.76, -0.73] 0.033
G/A (Maj/min) 49.58 [40.45, 58.72] (n=35) 0.065 -6.16 [-16.70, 4.38] 0.252
A/A (min/min) 67.48 [41.30, 93.65] (n=12) 0.570 11.55 [-4.02, 27.11] 0.146
a Reference haplotype b regression coefficient c adjusted p-value *Linear model with eNAMPT baseline values adjusted for age, sex, sepsis, center, and race, n=209, Wald Χ2(3)=8.80, p=0.032 †pairwise comparisons of means, Bonferroni adjusted
92 | Lynn
Figure 17: The optimized ARDS mortality risk prediction approach combines NAMPT rs16330082/rs59744560 haplotypes, eNAMPT levels, and ARDS covariates. A comparison of the four AUC models (testing for equal ROC areas) yielded a significant difference in the models (p=0.002). Models detailed in Table 5. Covariates: ROC=0.0.531 [0.440, 0.622]; the covariate model includes: race/ethnicity, age, sex, sepsis status, collection center. NAMPT rs61330082/rs59744560 haplotypes: ROC=0.575 [0.489, 0.661]. Plasma baseline eNAMPT: ROC=0.586 [0.499, 0.672]. Covariates, NAMPT haplotype rs61330082/rs59744560, and eNAMPT plasma: ROC=0.645 [0.567, 0.722]. The combined model includes all prior covariates, eNAMPT plasma levels, and the previously described haplotype risk score.
Model (205 observations each) AUROC # [95% CI] Model 1: Covariates (sepsis, gender, age, race, center) 0.531 [0.440, 0.623] Model 2: rs61330082/rs59744560 haplotypes 0.575 [0.489, 0.661] Model 3: eNAMPT plasma 0.586 [0.499, 0.672] Model 4: covariates + eNAMPT + rs61330082/rs59744560 0.645 [0.567, 0.722] #Null hypothesis: area(model1) = area(model2) = area(model3) = area(model4) *χ2(3) = 15.08; Prob> χ2 = 0.002
93 | Lynn
Discussion
ARDS is a challenging critical care entity due to the lack of FDA-approved
therapies and the sobering failure of therapeutically-driven ARDS clinical trials 334, 336, 337.
Contributing to the absence of successful Phase II/III trials is the inability to stratify ARDS
patients, either by mortality risk or likelihood of drug responsiveness, and the absence of
useful predictive biomarkers or genetic variants in the critically ill 338, 339. Although an inflammatory sub-phenotype among multiple ARDS cohorts was recently described as
significantly associated with sepsis outcomes and ARDS mortality 336, 340, a validated
prognostic biomarker that reliably identifies this ARDS inflammatory sub-phenotype has
not been identified or successfully utilized in clinical trials. ARDS subject stratification by
any modality would have significant utility in clinical management of ARDS patients and
in clinical trial design for novel therapeutics 61, 85, 341, 342.
Genomic and genetic approaches to deciphering ARDS risk and mortality have been reported 1, 219. Our prior system biology and translational approach to ARDS led us
and others to the identification of eNAMPT as a potentially useful ICU biomarker 153, 183,
343 and therapeutic target in ARDS and VILI 153, 183, 343. Our present report extends our
prior clinical and preclinical studies to demonstrate a genotype-phenotype link between
human ARDS plasma eNAMPT levels and NAMPT risk SNPs with ARDS mortality. We
conducted a covariate-adjusted genotype analysis of potentially functional NAMPT SNPs
selected on the basis of: i) SNPs residing in the functional NAMPT promoter region; ii)
SNPs that alter transcription factor binding or serve as CpG methylation sites; and iii) prior demonstration of NAMPT SNP linkage to ARDS risk or severity153, 183. Functional analysis
94 | Lynn of the NAMPT promoter via deletion constructs identified a region (-3,028 to -2,128 from
TSS) sensitive to 18% cyclic stretch in human lung endothelial cells151, 154.
The -948 (rs59744560) NAMPT promoter SNP has been previously shown via
deletion construct studies to be essential for NAMPT promoter activity154. STAT5, a
transcription factor that increases NAMPT mRNA transcripts, binds at the -948 NAMPT
promoter site, and a single-point mutation at -948 in a transfected NAMPT promoter
decreased STAT5 binding and NAMPT mRNA expression levels154. I have linked the -
948 SNP with eNAMPT levels (a potent cytokine response) and established a genotype-
phenotype link between the -1535 (rs61330082)/-1001 (rs59744560) A/C risk haplotype
and an ARDS outcome of interest (survival) with rs61330082 -1535AA associated with
decreased APACHE II scores. The current work is the first to show a linkage between
NAMPT haplotypes and increased eNAMPT plasma levels in ARDS patients (Table
7), and this result is consistent with the prior reported effects of these SNPs on increasing
NAMPT promoter activity and mRNA expression151, 188, 329. Another important aspect of
our study is the establishment of a link between the eNAMPT plasma biomarker and functional NAMPT SNPs associated with ARDS risk and worsened outcomes. We also showed that the rs61330082/rs59744560 A/C haplotype is associated with lower
eNAMPT levels (Table 7).
This rs61330082/rs59744560 haplotype was a key component of the risk score
developed as a pilot score to predict ARDS mortality (Table 8, Figure 17). The history of
genetic risk scores was initially implemented utilizing GWAS significant SNPs from large
prostate cancer cohorts332. This study established the concept the multiple risk SNPs can
outperform family history of disease and other relevant risk covariates in predicting
95 | Lynn disease risk332. The utility of genetic risk scores have been expanded to cardiovascular
diseases utilizing GWAS first SNP discovery106, 108, 344; stratification of risk score
percentages have been utilized in survival analysis to sub phenotype patients107. The goal
of this ARDS risk score is to further identify an ARDS endotype for clinical trial
stratification and expand the utility of the concept of genetic risk scores to an acute/critical
illness. Our strongest mortality prediction model included relevant covariates, haplotype
mortality risk, and eNAMPT plasma levels (Figure 17), which outperformed the individual
covariate, genotype, or plasma biomarker model. This combination of clinical variables
and biomarkers is a common strategy in developing genetic risk scores106, 345. The
integrated mortality risk score (NAMPT SNPs, eNAMPT plasma levels, covariates) produced an AUC of 0.645, which on first blush, may appear to reflect only a modest association with mortality. However, it is important to highlight that the AUC for our risk score is nearly identical to the atrial fibrillation risk scores that utilized 97 SNPs/loci in over 20,000 AF subjects106. In contrast, our analysis utilized only two NAMPT SNPs
(Figure 17).
Another important aspect of our study is the inclusion of a sizable African American
ARDS population, an ‘at risk’ population with disparities in incidence and severity 7, 10, 11
especially in the current landscape of the COVID-19 pandemic28, 37, 38, 312. Several genes
including MYLK [myosin light chain kinase] 170, 175-177, MIF [macrophage migration
inhibitory factor] 65, GADD45a [growth arrest DNA damage inducible alpha] 60, 346, S1PR3
[sphingosine-1-phosphate receptor-3] 51, and SELPLG [selectin P ligand gene] 347, have
previously been shown to harbor SNPs significantly associated with the development of
ARDS in African American populations. Our NAMPT genotype-phenotype link in African
96 | Lynn Americans offers the potential opportunity to utilize eNAMPT as a biomarker that offers
specific advantages to potential ARDS diagnosis/prognosis value in African American
patients who experience health care challenges in this critical illness.
This pilot project utilized plasma biomarkers and a genetic risk score approach to
bridge a translational gap between genetic risk and outcomes in ARDS. Limitations in this
current study include missing values for mortality and reduced availability for plasma
eNAMPT determinations across the cohort (Table 4); collectively, these limited the power
of our analysis of genotype-phenotype associations between NAMPT risk SNPs and
elevated eNAMPT plasma levels. Adjusted regression models were utilized respectively
to account for center, age, sex, and sepsis biases. As this is a candidate gene study, risk
SNPs were chosen based upon prior association with ARDS, ALI, sepsis, or critical
illness. When compared to the genetic risk scores in the atrial fibrillation studies cited
above106, 108, 344 that utilized SNPs chosen from multi-center GWAS collaborations106, 108,
344, ARDS GWAS studies remain challenging given the comparatively small cohort sizes
as well as the ARDS heterogeneity that limits the likelihood of identifying GWAS SNPs
that contribute to a significant and meaningful genetic risk score. As a result, we elected
to utilize a candidate gene approach based upon prior NAMPT SNP and gene
associations with ARDS risk and outcome severity. We utilized mixed effects regression
to account for biologically imprecise effects of clinically--relevant covariates and were
again limited by power and lower numbers of minor allele genotypes. To increase the
predictive power of our model, in the absence of a markedly increased sample size, we
added biologically-relevant protein and DNA biomarkers (eNAMPT levels and NAMPT
97 | Lynn genotypes) to refine our model in anticipation of a future validation study in a larger
replication cohort.
We and others previously have shown the significant association between
rs61330082 genotypes and 28-day mortality in ARDS68, 82, 187. We have now established
rs61330082 -1535AA (recessive genotype) as being significantly associated with lower
APACHE II Scores (Figure 13). While rs61330082 is not associated with eNAMPT trends,
higher eNAMPT levels were correlated with higher APACHE II scores (Figure 15) and
eNAMPT levels were also higher among African Americans with the
rs61330082/rs9770242 haplotype (Figure 16).
In summary, we propose an integrated ARDS mortality risk approach that utilizes: i) biologically-relevant phenotypes, ii) mechanistically-relevant and risk-associated ARDS
genotypes, and iii) an ARDS plasma biomarker, and iv) clinically-relevant covariates that capture additional (but biologically imprecise) risks (Table 8, Figure 17). The results reported here provide a potential roadmap for the elucidation of multi-omics approaches to clinical trial subject stratification with NAMPT as an informative test case supported by plasma eNAMPT levels and a NAMPT genotype which includes rs61330082’s strong association to mortality risk in ARDS, a potentially useful tool for future clinical trial design in ARDS.
Future Directions
NAMPT as a Biomarker in Other Pulmonary Diseases. Many of the genes found
associated with ARDS pathology have loci associated with other pulmonary diseases50,
98 | Lynn 97, 101, 128, 134, 174-176, 202, 262, 348-350. A recent multi-disease study identified a sizeable group of genes associated with both increased transcription across multiple chronic lung diseases (idiopathic pulmonary fibrosis, chronic obstructive pulmonary disorder, and severe asthma)348. This study design was underpinned by the rationale that severe
pulmonary disorders share several underlying pathological features: i) inflammation351, ii) reduced lung function352, 353, and iii) progressive deterioration of the alveolar barrier354.
While this study focused on chronic lung diseases, several genes associated with ARDS
risk (NAMPT and GADD45a) were found to be associated with idiopathic pulmonary
fibrosis (IPF) and COPD348. Ongoing studies in the Garcia lab in prostate cancer,
radiation-induced fibrosis and idiopathic pulmonary fibrosis lends support to the
hypothesis that NAMPT may serve as a potential therapeutic or diagnosis biomarker for
multiple pulmonary diseases.
Pulmonary arterial hypertension (PAH) is a progressive disease affecting the
pulmonary pre-capillary vasculature leading to severe pulmonary vascular remodeling, persistently increased vascular resistance, and heart failure355. NAMPT’s role in
promoting endothelial cell survival66 and increased cellular proliferation and cellular
redox356 state makes it a candidate gene of interest to study in PAH. A PAH study in
Garcia lab recently expanded the mechanistically understanding of NAMPT and eNAMPT
in lung injury185. eNAMPT expression was increased in PAH patients compared to
controls, and PAH-relevant stimuli increased NAMPT promoter activity185. NAMPT
promoter responsiveness to PAH stimuli (platelet-derived growth factor [PDGF], vascular
endothelial growth factor [VEGF], and transforming growth factor-1 [TGFB1]) leaves the
99 | Lynn tantalizing questions open for how the known functional SNPs studied in ARDS risk and
outcome severity are associated with PAH risk and survival.
In a large, well-phenotyped database of PAH type 1 patients, rs61330082
genotype AA is associated with increased survival in PAH (Figure 18, Supplemental
Data). A Cox regression in an adjusted model from the Cincinnati PAH Biobank cohort
found rs61330082 to be significant for decreased mortality risk in PAH (HR=0.24 [0.08,
0.78]). Because PAH is a chronic disease, patient follow-up visits allow for a valid survival
analysis to be performed (between right heart catheter and death or last follow-up).
Rs61330082 AA was significantly associated with survival compared to rs61330082 GG
(reference genotype). Of note, the heterozygote (rs61330082 GA) did not observe a significant decrease in elevated risk (HR=1.12 [0.76, 1.63]) unlike the homozygous minor allele genotype AA (Figure 18). In ARDS, a survival analysis with rs61330082 was not feasible due to the acute nature of the syndrome and the variation in enrollment times across the multiple cohorts used. The lack of a survival analysis was a limitation of the previous NAMPT study in ARDS, but in PAH, the extensive databank and follow-up period
(years) allows for a statistically and clinically valid survival analysis to be performed. A validation cohort from the University of Arizona (IGABS) has been genotyped to validate this survival analysis. These novel results with the rs61330082 AA genotype and its protective role against pulmonary diseases (decreased odds ratio for mortality in ARDS and a decreased hazard ratio for survival in PAH) are promising data and crucial for applying NAMPT disease risk-associated genotypes as biomarkers.
100 | Lynn
Figure 18: Cox regression adjusted model includes age at catheterization (diagnosis), gender, and PC1-3. Survival time is between data of RHC and death or last follow-up. Tested additive model for 4 SNPs, and dominant model for rsrs116647506. PTrend was calculated by treating genotype as an ordinal variable.
101 | Lynn ARDS is an Endotype for COVID-19. Coronavirus disease (COVID-19) is caused by an infection of SARS-CoV-2 and can result in sepsis and ARDS37, 38. While several risk
factors (obesity and age) for increased mortality in both COVID-19 and ARDS are the same, there is emerging evidence COVID-19 may have unique risk factors (diabetes and hypertension) that make it a distinct endotype category of ARDS37. Endothelial cell
signaling may play a key role in this COVID-ARDS endotype via signaling receptors ACE-
2 and CEACAM138. The endothelial barrier’s role in driving ARDS, ALI, and sepsis outcomes makes biomarkers associated with endothelial dysfunction and ARDS severity attractive options for COVID-19 biomarkers as well38, 357. NAMPT’s role as a biomarker in
ARDS and its use in a potential sub-phenotyping risk score model makes it an attractive
option for use in COVID-19 trials as well.
The barrier dysfunction genes associated with VILI are strong candidates for
COVID-19 biomarkers that could be associated with worsened COVID-19 outcomes.
Statistical genetic based risk models and pathway analysis studies could be tools used
to stratify and categorize biomarkers (SNP, mRNA gene expression, or plasma levels) that are more predictive of worse outcome in COVID-19 patients. As a distinct endotype of ARDS, COVID-19 will share many risk factors common with ARDS/VILI injury pathophysiology. As a sub-phenotype of ARDS, COVID-19 will have specific biomarkers that will be useful in stratifying patients at risk for worsened outcomes.
102 | Lynn Chapter 4: Aim 3: ARDS & Methylation
Significance
Acute Respiratory Distress Syndrome (ARDS) is a critical illness affecting
~200,000 to 400,000 individuals annually in the U.S.3, 7-9 with a 30-40% mortality rate.
African descent and Hispanics exhibit greater even mortality rates than non-Hispanic
Caucasians, even after adjustment for relevant confounding factors (age, sex, socio- economic quantiles)7, 104, 309. Both direct lung injury (trauma and pneumonia) and indirect lung injury (sepsis) are the leading comorbidities in ARDS patients1, 3. Unfortunately, the
heterogeneity of ARDS phenotypes and lack of validated biomarkers have impeded
advances in therapeutic developments in ARDS and to date, all drugs used for ICU care
of ARDS patients are generic. Every clinical trial designed to assess novel ARDS
therapeutics have failed18, 21. While sub-phenotyping ARDS patients into high and low inflammatory groups based upon biomarkers has shown promise in identifying ARDS patients at high risk of mortality22, 51, 67, 73, 104, 255, currently there are no validated
biomarkers, genotypes or molecular signatures that predict ARDS mortality67. Of
importance is to establish biomarkers consisting of genotype-phenotype links to sub-
phenotype high-mortality risk ARDS populations27. Over 200 genes have been associated
with ARDS1, 133, and all these genes have been discovered to be associated with ARDS
post sequencing of the human genome1, 133. The recent application of genetics and genomics to understand ARDS biology and pathology has provided insights into the complex sub-phenotypes present in ARDS. There is a need to move the understanding
103 | Lynn of ARDS as a complex genetic disease beyond candidate gene and SNP-focused
studies19.
Epigenetics is the study of heritable changes to gene activity or function that is not
dependent on altering the DNA sequence itself150. DNA methylation is when methyltransferases (DNMTs) catalyze the transfer of a methyl group (CH3) onto the fifth
carbon of a cytosine residue150. While genetics in ARDS has been well studied51, 55, 68, 81,
83, 84, 104, 133, 136, 167, 175, 176, 180, 183, 237, 308, 358, the literature on epigenetic changes in ARDS
is sparse and focused solely on candidate gene studies151-155. For example, the Garcia
lab has shown that ARDS candidate genes such as nicotinamide
phosphoribosyltransferase (NAMPT) and myosin light chain kinase (MYLK) showed
methylation changes in their promotor regions under conditions of mechanical stress that
mimic ARDS-associated ventilator-induced lung injury (VILI)153, 154. In MYLK, two
epigenetic variants (cg03892735 intron 4/5, cg23344121 exon 18) were associated with
ARDS risks after controlling for sex, age, and ethnicity152. For NAMPT, histone
deacetylase (HDAC) activity was shown to be critical for NAMPT gene transcription151.
Excessive mechanical stress (18% cyclic stretch) and lipopolysaccharide (LPS) both
elicited a large-scale demethylation of the NAMPT promoter151. Thus, epigenetic changes
across the genome reflects a key regulatory mechanism that influences ARDS mortality.
The application of genetics to ARDS34, 125 is recent (corresponding with the
sequencing of the human genome circa year 2000) in comparison to the long history of
clinical studies and therapies in ARDS (first described in 1967)206, 223. A limited number of
GWAS have focused on ARDS risk SNPs and their associated genes79, 80, 96, 104, 235. These
studies have produced a robust list of associated candidate genes that have advanced
104 | Lynn the understanding and potential treatment of ARDS1, 133, 206. Risk SNPs for mortality and
a focus on the most severe inflammatory subphenotypes of ARDS have also been
identified73, 75, 80, 187, 213, 214, 231, 232, 237, 288, 294. These studies provide the basis for
developing a mortality-based genotype signature with the potential for top genes to
predict mortality risk for high-inflammatory burden ARDS patients. Environmental
damage (direct or indirect) to the lungs is the initiator of ARDS, and a genome-wide methylation study, which has never previously been performed, could provide critical links between associated ARDS genes, their respective SNPs, and individual CpG sites or islands that could provide a powerful signature to predict ARDS mortality1. A genome-
wide methylation study could provide biological understanding of which genes and
pathways are the most robust for predicting ARDS mortality by providing a crucial
genotype-phenotype link.
This study on genome-wide methylome changes focuses on ARDS outcome (28-
day mortality). I have shown how methylation changes at individual CpG sites and across
biologically relevant promoter CpG islands are associated with ARDS mortality outcomes.
A GSEA (gene set enrichment analysis) bridges the divide between novel omics findings
(whole methylome analysis) and prior endothelial barrier dysfunction gene associations.
Methods
Patient Demographics and Sample Collection. Subjects were recruited from The
University of Arizona Health Sciences (IRB#1312168664R001), the University of Illinois
at Chicago (IRB# 20120192), and the University of Chicago (IRB#15194A) from 2007-
105 | Lynn 2018. Enrollment included subjects admitted to the ICU with confirmed ARDS or at risk
for development of ARDS (sepsis, septic shock, and trauma). Written consent was
obtained for all the participants. ARDS diagnosis was established according to the
diagnostic criteria per the American-European Consensus Conference (AECC)212, 359 or
the Berlin definition12.
A total of 48 ARDS patient samples were utilized in the genome-wide pilot study.
DNA derived from peripheral blood mononuclear cells (PBMCs) were collected in
accordance with their respective IRBs and transferred with appropriate material transfer
agreements. Blood was collected within 36 hours of ARDS onset defined as when all
Berlin ARDS Criteria were met. Blood was collected in EDTA-treated tubes, centrifuged
for 1 hour (2000 x g for 20 min, RCF) and the platelet-depleted plasma stored at -80°C.
DNA was extracted from PBMCs stored in buffy coat according to standard protocol
(Qiagen DNA Mini kit: Hilden, Germany). DNA quality and quantity were access with
NanoDrop. For the validation cohort, 50 samples were obtained from the IGABS BioBank and transferred with appropriate material transfer agreements. The same protocols to collect, process, and evaluate DNA samples were used as in the initial cohort.
Genome-wide Methylation Profiling ChIP. High quality DNA was derived from peripheral blood mononuclear cells (PBMCs) available from 48 subjects. Subjects were chosen on the basis of: i) to achieve outcome parity regarding 28-day mortality status, ii) to represent a multi-ethnic cohort, and iii) based upon the quality of existing DNA (260:280
> 1.80, total DNA > 1 ug). A detailed protocol can be found in the previously published literature151, 152 and via the manufacture’s website (bisulfite conversion Qiagen,
106 | Lynn hybridization and ChIP sequencing, Illumina). Bisulfite conversion and ChIP hybridization
were performed on all 48 DNA subjects prior to ChIP methylation on the MethylationEPIC
BeadChip platform (Illumina). Poor quality samples were excluded using a detection p- value cutoff greater than 0.05 using the R package minifi. After standard minfi and combat
quality control steps, 45 subjects were included in the analysis (3 subjects were removed
for incomplete or contradictory medical records). Methylation levels were represented as
beta values (methylated/unmethylated), which were used in our downstream analysis.
mRNA Extraction and Genome-wide Transcription Profiling. Peripheral blood
mononuclear cell (PBMC) isolation was performed using the Ficoll-Paque method. Total
RNA was isolated from PBMCs using RNAeasy MiniKit Qiagen™ following
manufacturer’s protocol. RNA concentration and quality (RIN>7) was assayed by
Nanodrop™ (Thermo Fisher) and 2100 Bioanalyzer RNA™ (Agilent). 48 unique samples
were sent for transcription profiling using RNA-sequencing. 1 μg total mRNA was used to
prepare the sequencing library. Total RNA was enriched by oligo (dT) magnetic beads
(rRNA removed) prior to RNA-seq library preparation (Illumina KAPA Stranded RNA-Seq
Library Prep Kit). The completed libraries were qualified with Agilent 2100 Bioanalyzer
and quantified by absolute quantification qPCR method. To sequence the libraries on the
Illumina HiSeq 4000 instrument, the barcoded libraries were captured on Illumina flow
cell, amplified in situ, and subsequently sequenced for 150 cycles for both ends on
Illumina HiSeq instrument. Genes with low expression (defined by low count <6) were
removed. A total of 18,652 genes were included in the final analysis.
107 | Lynn Pathway Analysis. Gene Set Enrichment Analysis (GSEA) was utilized for pathway
analysis360. The methods utilized for transcription gene set enrichment analysis were
modified to be applicable to a genome-wide methylation ChIP. The Molecular Signatures
Database (MSigDB) Biocarta curated gene database (298 canonical pathways) was utilized360.
ARDS Combined Lung Injury Preclinical Rat Model. 1 ug of DNA from the IGABS
validation cohort was used in an Msel enzyme digest overnight at 37 °C. Digested DNA
was hybridized with the LINE-1 probed prior to immobilization on a streptavidin-coated
plate. The plate was washed and blocked to reduce non-specific binding prior to 1 hr
incubation with 5-mC antibody. Another was performed, and anti-IgG HRP antibody was
conjugated and incubated for 1 hr. The plate was washed 3x prior to the addition of
stopping and developing solutions. A standard curve was generated with methylated and
unmethylated DNA controls provided via the manufacturer (methylation percentages
used: 0, 10, 20, 50, 75, 100). This assay was performed as per the manufacturer’s
instructions (Active Motif, Carlsbad, CA, USA) and has been shown to be sensitive and
repliciable361 to determine the methylation status between differentially methylated cell
lines362 and human blood samples363.
ARDS Combo Model. Intratracheal LPS (0.1 mg/kg) was injected into 150-200 g
Sprague Dawley male Rat from Charles River after being anesthetized with
ketamine/xylazine (100/5 mg/kg) and intratracheally intubated with a 16-G angiocath into
the trachea. After the injection, 3 ml of room air was injected twice into the catheter to
108 | Lynn clear the trachea from residual solution and distribute LPS inside both lungs. Rats were
anesthetized 20h later and intratracheally intubated and connected to VentElite small
animal ventilator from Harvard Apparatus for 4hrs with a tidal volume (Vt) 20ml/kg and
respiratory rate of 70 bpm. Rats were sacrificed 4 hrs later for subsequent analyses and
left lung tissue was snap frozen. 20-30 mg of left lung tissue was used (Qiagen kit) to
extract DNA for further experiments.
qPCR Assay. 1 ug of DNA was digested with either a mock enzyme (Mo), a methylation- sensitive enzyme (Ms), a methylation-dependent enzyme (Md), or a double enzyme digest (Msd) (Qiagen, EpiTect Methyl II DNA Restriction Kit). The digested DNA was run with SYBR Green qPCR master mix (Qiagen, RT2 SYBR Green qPCR Mastermix) and
run on the CFX96 (BioRad, Hercules CA USA) and analyzed with the CFX Maestro
software (BioRad, Hercules CA USA) and according to the manufacture’s analysis guides
(Qiagen EpiTect Methyl II Assay).
Results
A Unique Methylation Profile in PBMCs is Present for Severe ARDS Outcomes. An ethnically and racially diverse ARDS cohort was assembled for a pilot discovery-based experimental design with the focus of differentiating ARDS severe outcomes (28-day mortality) (Table 9). There were no significant differences in age or septic status between the 45 ARDS patients selected for this study (Table 9). 15 CpG sites associated with genes were found to be significant (p <0.05, Benjamini-Hochberg FDR corrected) in a
109 | Lynn Fisher’s X2 analysis (Table 10). A total of 22 CpG sites were found to be associated with severe 28-day mortality ARDS outcomes (Figure 19).
Table 9: Demographics of ARDS patients used for genome-wide methylation study
Total Mortality Gender European/African/Hispanic Sepsis % (n) Age (±sd)
ARDS (M:F) American
23 Live 10:13 7/9/7 78 (18) 53 (±15)
22 Deceased 16:6 7/6/9 81 (18) 53 (±12)
Table 10: Significant CpG sites in a 28-day outcome-based Fisher’s Χ2 analysis
Gene (CpG CpG location P-value/P-value Live Beta Avg Dead Beta Avg Delta Beta
site) adjusted
5'UTR-island -08 0.05 0.06 +0.01 C10orf18 1.91e /<0.01 (cg21302771)
Body-opensea -08 0.44 0.47 +0.03 WDR66 2.98e /<0.01 (cg02294570)
TSS1500-shore -08 0.43 0.45 +0.02 GPR137 4.97e /0.01 (cg22753417)
Body-opensea -07 0.91 0.92 +0.01 STAT4 1.69e /0.02 (cg18481642)
5’UTR-opensea -07 0.70 0.74 +0.04 KIAA0513 1.80e /0.02 (cg07450021)
Body-shelf -07 0.39 0.43 +0.04 MSL1 2.18e /0.02 (cg04309037 )
TSS1500- -07 0.50 0.53 +0.03 ZNF683 6.12e /0.02 (cg15118519) opensea
TSS1500-shore -07 0.39 0.42 +0.03 DPF1 2.33e /0.02 (cg23480284)
110 | Lynn 3’UTR-island -07 0.86 0.88 +0.02 NCLN 5.50e /0.03 (cg00922143)
Body-shelf -07 0.89 0.90 +0.01 SAE1 5.76e /0.03 (cg08239694)
Body-opensea -07 0.59 0.63 +0.04 INSR 6.18e /0.03 (cg09940832)
Body-shore -07 0.57 0.60 +0.03 MRAP 6.41e /0.03 (cg03216491)
Body-opensea -06 0.42 0.43 +0.01 PDCD6IP 5.67e /0.05 (cg02385556)
3’UTR-opensea -06 0.81 0.84 +0.03 SSH3 1.12e /0.05 (cg05295023)
Body-shelf -06 0.48 0.52 +0.04 TBC1D8 1.16e /0.05 (cg08862717)
111 | Lynn C10orf18 AD deceased WDR66 HD deceased GPR137 ED deceased STAT4 AD survived KIAA0513 MSL1 HD survived ZNF683 ED survived DPF1 NCLN SAE1 INSR MRAP PDCD6IP SSH3 TBC1D8
Figure 19: ARDS Mortality Signature by DNA Methylation. Heatmap of 22 significant CpG sites in live vs dead (n=23 survived, n=22 deceased) X2 Fisher’s exact model (q<0.05, FDR adjusted). Each row is a CpG methylation site and red denotes increased methylation (see color key histogram) and green denotes demethylation. 21 CpG sites are hypermethylated (increased methylation). The genomic location of these sites are further detailed in Supplemental Data Figure 2.
112 | Lynn
c-MET Pathway is Associated with Methylation Enrichment in ARDS. Gene set expression analysis (GSEA) ranks genes by their expression differences, computes the cumulative sum over ranked genes, and records the maximum deviation from zero to calculate an enrichment score (ES) per pathway360. Due to its well-curated pathway yet sizeable database, Biocarta database (298 total canonical pathways) had an ES generated. 17 gene sets were identified with an FDR < 25% (the GSEA standard cut-off), but a single pathway (c-MET/HGF pathway) passed a more stringent FDR cutoff
(FDR<0.05) (Figure 20).
Figure 20: cMET is a significant pathway for ARDS severe outcomes (28-day mortality). GSEA analysis discovered 17 gene sets with FDR < 25%, 5 Gene sets p-value < 5%, 17 gene sets p-value <1 %.
113 | Lynn GSEA Analysis Reveals anIinterconnected Network of Receptor Tyrosine Kinase
Signaling Genes and Transcription Factors. cMET/HGF is one of many receptor
tyrosine kinase signaling pathways that are crucial to regulating vascular cell layer
permeability and maintaining a functional blood/air barrier in the lower airway4. Other receptor tyrosine kinases associated with endothelial barrier dysfunction are VEGF,
EGFR, and FGF-250, 53, 54. Similar to EGFR, HGF activates RAS proteins, which further
signal to PIK3/AKT/mTOR45, 46. The upregulation of the mTOR1 pathway disrupts the
formation of focal adhesions in endothelial cells. HGF signaling via mTOR has been
shown to drive angiogenesis and is involved in vascular barrier regulation. cMET/HGF
also internal signal to upregulated Nf-kB, a key transcription factor that promotes a
hypoxia53, 364.
While the cMET/HGF pathway was the only significant pathway with an FDR < 5%,
16 other pathways passed a less stringent GSEA cutoff of 25% FDR and had an
enrichment score p-value < 0.05 (Figure 20). Across these pathways, several common
genes drove the pathway analysis, creating a network of potentially crucial genes in
methylation. MAP kinases (MAPK3, MAPK1, MAPK14) were all in the top 10 genes shared across all pathways and are part of the canonical cMET pathway in Biocarta
(Figure 21, Figure 22). Other major transcription factors (STAT1, STAT3, STAT4, and
Nf-κB) were in the top list of genes as well.
114 | Lynn
Figure 21: cMET/HGF pathway. A schematic reproducing a simplified version of the cMET/HGF pathway that highlights the role of MAPKinases and AKT/mTOR signaling2.
16
14
12
10
8
6
4
2
Number of Gene Sets 0 B κ IL7 IL15 FYN CD4 IRS1 PTK2 NF RELA RAC1 SOS1 STAT3 STAT1 STAT2 MAPK3 MAPK1 PIK3CG PPP3CB MAPK14 PRKACG PRKAR1A PRKAR2A Figure 22: Overlapping genes from across 17 top GSEA pathways. Top GSEA pathways were analyzed for genes that appeared most often.
115 | Lynn
Global Genome Methylation is Increased in Severe ARDS Outcomes in a Validation
ARDS Cohort. While high performance liquid chromatography (HPLC) has been the gold
standard to measure global methylation, alternative PCR and ELISA based methods that
require less DNA have been shown to be replicable and sensitive enough to detect
methylation changes between cell types and patient samples. Due to limited DNA from
precious DNA samples in the IGABS validation cohort, a PCR method that cleaves the
after major methylation genomic repeats was utilized to measure global methylation in an
independent ARDS cohort (Figure 23, Table 11). When site of infection is controlled for
(lung), a significant difference in global methylation percentage was observed in ARDS
patients that died within 28 days (Figure 23, Table 11).
Figure 23: Global methylation in ARDS severe outcomes (28-day mortality). ARDS patients with sepsis with a primary site of infection (lung) have significantly higher levels of global methylation (Lived: 34.12% [27.54, 40.70] (n=13); Died: 43.80% [37.95, 49.66] (14); p=0.025).
116 | Lynn Table 11: IGABS validation cohort demographics
Survived (n=23) Deceased (n=19) P-value
Age 44 [38, 50] 60 [52, 67] 0.002
Gender 13 (M), 10 (F) 13 (M), 6 (F) 0.429
Ethnicity 12 (NH), 11 (H) 12 (NH), 7 (H) 0.474
Sepsis 18 16 0.625
Source of Infection (Lung) 14 13 0.611
117 | Lynn
Methylation in CpG Islands in Key Transcription Factors in ARDS Show Altered
Levels of Methylation in ARDS/VILI. CpG promoter islands are key site of epigenetic
regulation in gene transcription150, 322, 365, 366. An unmethylated promoter leads to gene
expression and transcription, and higher (hyper) methylation leads to gene silencing in a
promoter150. While ROS and inflammatory genes are of key interest as ARDS therapeutic
targets, the transcription factors that regulate them are of equal importance due to their
ability to regulate DNA repair, the cell cycle, angiogenesis, and apoptosis.
In an ARDS rat model (combo LPS/VILI), two key transcription factors involved in
DNA damage/repair (Runx1 [runt-related transcription factor 1]367-370, Runx2 [runt-related
transcription factor 2]368, 369, 371) were found to have hypermethylated promoter CpG islands (Figure 24). Of key importance, both Runx1/Runx2 form complexes with p53, and
while they have not been studied in ARDS, they have been studied in cancer369.
RUNX1/acetylated p53 promoter DNA repair and arrest the cell cycle via GADD45A and
CDKN1A. RUNX2/phosphorylated p53 arrest the cell cycle and promote apoptosis. Both
Runx1 and Runx2 rodent models have shown pathogenic phenotypes. Knock-out models of Runx1 in mice have vascular defects and fail to establish hematopoiesis in embryogenesis372. Runx2 knockout mice were found to have skeletal structure bone and
tissue malformations373, 374.
118 | Lynn
* LPS/VILI Runx1
* LPS/VILI Runx2
PBS Runx1
PBS Runx2 Percentage methylation
Figure 24: Runx1 and Runx2 both show increased methylation levels across CpG promoter island. N=6 for each group (PBS control, combo LPS/VILI). * p<0.05
119 | Lynn
Discussion
DNA methylation evidence in ARDS has been limited to single genes and VILI
animal models151, 153, 154, 375, 376. A recent study utilized a multi-omics retrospective study approach to study methylation status in prior datasets156 with 10 genes related to
inflammation or immunity, endothelial function, epithelial function, and coagulation
showing differential changes in methylation. This pilot study (Illumina EPIC ChIP) in
ARDS is the first study to show a differentially regulated CpG sites in severe ARDS
outcomes (Figure 19, Table 10). While other methylation studies in ARDS have been
done, my pilot data is novel for its incorporation of ARDS human samples, expression
pathway analysis linking these whole-methylome findings to key pathways and
pathogenic gene drivers in ARDS (Figure 21, Figure 22), and in showing this hypermethylation pattern extends to larger regions of the genome (whole genome and
CpG promoter islands) (Figure 23, Figure 24).
GSEA analysis in this pilot dataset associated the canonical c-MET/HGF pathway with severe ARDS outcomes (Figure 20). The c-MET/HGF pathway was previously shown to increase proliferation in type II alveolar epithelial cells and be necessarily to induce lung injury in type II alveolar epithelial cells377. After LPS treatment in animal models of lung injury, HGF was elevated and promoted barrier dysfunction48. In
hepatocellular carcinoma, HGF signaling dysfunction is a critical pathway in maintaining
the tumor micro environment and has a role in angiogenesis, tumorigenesis, and
regeneration47. Treatment of hepatocellular carcinoma cell lines (HepG2) with 5’Aza-2’-
deoxycytidine (5’Aza) and subsequent analysis of mRNA levels showed increased mRNA
120 | Lynn levels in the key HGF pathway transcriptomes47. My GSEA analysis is the first study to
associate alternations in HGF pathway methylation with ARDS outcomes (28-day
mortality) (Figure 20). While a limitation of this study is the lack of RNA from these
samples to validate key mRNA transcription levels, this is a critical study that links a key
pathological pathway (c-Met/HGF) in maintaining the alveolar barrier with epigenetic
changes in severe ARDS outcomes.
Due to small sample sizes, a second well-phenotyped ARDS cohort was utilized to validate the methylation trends present in the initial study (Table 11). Global methylation changes have been proposed as a potential biomarker in cancer diagnosis and treatment monitoring due to wide-spread global hypomethylation of gene promotors by DNA methyltransferase (DNMT1)365, 366. In ARDS, global methylation (like genome
wide methylation) is a trend that has remained understudied, and this study is the first to
report an increase in global methylation in ARDS and septic patients (Figure 20). This
significant increase in global methylation in a separate ARDS cohort supports the
hypermethylation of multiple CpG sites in the pilot microarray methylation data (Figure
19, Table 10). Consistent methylation trends in PBMCs is a key step in profiling epigenetic
changes in ARDS patients.
Changes in methylation metabolism and DNMT1 regulation may play a role in the
methylation trends observed in ARDS. Methionine serves as the direction precursor to S-
adenosyl methionine, which is the substrate needed to methylate DNA378. In studies
involving methylation mechanisms in tumors, methionine acts in a feedback and
regulation system with mTOR complex 1 (mTORC1). Low levels of methionine inhibit the
activity of mTORC1, which regulates cell growth and metabolism379. The folic acid cycle
121 | Lynn is the metabolic pathway that converts folic acid into methionine380. Under normal dietary
conditions, folate and cofactors B6 and B12 main the 1-carbon flux, which ensures normal
demethylation/remethylation cycles378, 380. DNA-targeted hypermethylation is a
characteristic of tumors, and this pattern is non-random380. Folate depletion and other
changes to diet can result in methylation changes in tumor suppressor genes380. The combination of enzyme feedback regulation, cofactors (B6, B12), and folate levels can all contribute to aberrant DNA methylation patterns.
DMNT1 and the other two DNA methyltransferases found in mammals (DMNT2 and DMNT3) were first mapped to the human genome and described in 2000381. While
DMNT1 has no known SNPs associated with ARDS, ALI, or sepsis, DMNT1 SNPs have
been found to be associated with cardiac autonomic neuropathy382. A meta-analysis of
DMNT1 polymorphisms across multiple cancer types (cervical, gastric, esophageal,
prostate, and breast) found three separate SNPs (rs16999593, rs2228612, and
rs2228611) associated with cancer risk383. The genetic regulation of DNA methyltransferases could alter the DMNT1 mechanisms and biochemistry of the -CH3
methyl group transfer. Post-translational modifications to stabilizing transcription factors
(p300, HDAC1, Sirt1, Akt1) that interact with DMNT1 have the potential to be major
factors in how DMNT1 is regulated in ARDS/VILI384. Akt activates HIF1α via mTOR, and
HIF1α is crucial to maintaining the proteasomal degradation in maintaining normoxia385.
Transcription factors are the mediators between inflammatory-activated cell signaling and pro-inflammatory gene transcription. The Akt/mTOR pathway is a key
signaling pathway in endothelial cells that regulates inflammatory signaling and vascular
barrier integrity in ARDS. LPS signaling down-regulates autophagy via Akt/mTOR, and
122 | Lynn mTOR/STAT3 activation is mechanistically linked to the HGF pathway in mesenchymal
stem cells386, 387. The role of the STAT transcription factors in regulating gene expression
of candidate genes associated with ARDS risk and endothelial barrier dysfunction is well
documented242, 272, 388-394. Notably, Akt and p300 are mechanistically involved in both
mTOR and DMNT1 complexes. While the mechanistic regulation of these complexes is
beyond the scope of my project, epigenetic regulation to these stabilizing transcription
factors could alter the activity of major damage repair complexes in ARDS.
GADD45a is a VILI-related candidate gene and is related to a diverse array of cell
processes including cell cycle checkpoints395, apoptosis396, DNA methylation excision and repair397, TH1 differentiation398, and regulation of ubiquitination59. GADD45a is
regulated by p53 acetylation or phosphorylation activation, which determines the
availability of p53 as a binding partner to p300/RUNX1 or HDAC6/RUNX2369, 370. Long recognized as a master regulator of the cell cycle in cancer, p53 regulates cell cycle checkpoints, DNA repair, and apoptosis. A recent and interesting finding in LPS-
challenged p53 knock out mice showed increased edema and interleukin expression (IL-
1α, IL-1β, and TNFα), which established p53 and its regulation compounds may have a
key role in endothelial defenses against inflammatory stimuli399. p53 also attenuates the
proinflammatory transcription factor, Nf-κB400. The interactions of GADD45a, p53, and the
RUNX genes need to be further investigated in ARDS. GADD45a acts as a stress sensor
that regulates growth, proliferation, and apoptosis401. In ARDS, GADD45a knockout mice
were associated with an increase in mRNA expression of proinflammatory factors
(CXCL1, CXCL2, and IL-6 among others)59. While the role of GADD45a is well
established in ARDS, GADD45a regulation in ARDS remains vague. These epigenetic
123 | Lynn data provide evidence that the regulation of GADD45a by the RUNX genes and p53 may
play an important role in endothelial barrier dysfunction by signaling through receptor
tyrosine kinases and further activation of GADD45a.
The various elements of these methylation data support a novel trend of hyper-
methylation in ARDS. In humans PBMC samples, 22 CpG sites were hypermethylated at a genome-wide level, and pathway analysis revealed cMet/HGF to be a key enriched methylation pathway. This hypermethylation trend was support in an independent cohort were mortality outcome in an ARDS/septic population was studied. An ARDS combo rate model (LPS/VILI challenge) supported that this hypermethylation trend could extend to the CpG islands of promoters that interact with key genes involved in endothelial barrier integrity. Validation of individual CpG sites and further mechanistic studies to validate these data need to be done, but a strong trend in hypermethylation in ARDS has been presented.
124 | Lynn
Future Directions
Promoter methylation status affects the transcription levels of mRNA.
Hypomethylation (a decrease) is associated with unregulated gene transcription402, 403. In this pilot data of ARDS methylation, I have primarily observed patterns of hypermethylation (an increase) in severe ARDS outcomes. Therefore, a decrease in mRNA levels should be expected at a genome-wide level, but the complexity of ARDS
Figure 25: Three genes (ENTPD2, HPGD, CACNA1D) had significantly upregulated transcripts. A generalized linear model (adjusted for race/ethnicity, age, and gender) for ARDS outcome (28-day mortality) was used in edgeR.
125 | Lynn complicates a simple conclusion. As with other transcriptomic microarray approaches,
mRNA in a cohort of 50 ARDS patients (adjusted for race/ethnicity, gender, and age)
indicate no down-regulated mRNA transcripts (Figure 25). Three genes (ectonucleoside
triphosphate diphosphohydrolase 2 [ENTPD2], 14-hydroxyprostaglandin dehydrogenase
[HPGD], calcium voltage gated channel subunit alpha1 D [CACNA1D]) were upregulated
in deceased ARDS patients. Of the three genes, two of them (ENTPD2 and HPGD) have
log fold changes > 2 (ENTPD2=3.08, p=0.04, Bonferroni FDR adjusted; HPGD=2.20,
p=0.04, Bonferroni FDR adjusted) (Figure 25). ENTPD2 has been shown to upregulate
HIF-1α, a transcription factor that has been previously reported upon as a crucial mediator
of endothelial barrier dysfunction231, 364, 404. HPGD is mediated with ROS species, and I
and others have reported ROS genes and ROS-mediated pathways as one of the major
pathways in ARDS pathology44, 207, 229, 234, 260, 405, 406. While these genes are novel candidate genes associated with ARDS in prior known pathways, how these transcription factor changes relate to the methylation alterations in this cohort is uncertain. More investigation of the transcription changes that result from the hypermethylation signature will be crucial in understanding the nature of the global methylation changes observed in these data.
126 | Lynn Chapter 5: Discussion
Conclusion
ARDS was first described in a 1967 case report223. A seminal randomized set of
control trials (1998-2002)225, 407, 408 was the first to recognize low tidal volumes to manage
VILI, which had been previously reported in both animal models409 and human patients410
in the early ARDS literature. In the relative timespan of ARDS management and
therapeutic care, omics is a relatively new addition133, 206. In the past 20-25 years of
genetic and genomic studies in ARDS, a plethora of information has been generated67,
133, but synthesizing these data into a clinically applicable use continues to be a
challenge22, 63, 159. There have been several comprehensive studies and reviews on how
to use both plasma protein biomarkers and genetic/genomic associations to improve the
treatment and therapeutics in ARDS20, 21, 27, 63, 67, 85, 159, 213, 228.
While the challenges of utilizing precision medicine in critical care are many, I have
attempted to provide novel tools and data interpretation to advance the field in my thesis.
The heterogeneous and challenging nature of ARDS requires genetic methodologies
beyond GWAS associated SNPs and mRNA profiling to account for ARDS risks and
mortality outcome susceptibility. While other studies have utilized multiple plasma
biomarker panels and LCA to predict risk and outcome27, 63, 67, I have utilized a genetic
risk score based upon a single gene (NAMPT) with an associated ARDS haplotype and
prior evidence of elevated plasma protein (eNAMPT) associated with mortality62, 63, 222. A
genetic risk score has potential for combining multiple omics data from across biologically
relevant genes and pathways in ARDS.
127 | Lynn With the advent of COVID-19, there is an increased urgency to understand VILI and provide effective therapeutics for those at risk for ARDS. In my thesis, I have proposed several methods (mRNA profiling, genotype/phenotype risk models, and epigenetic profiles) to stratify ARDS severity endotypes/subphenotypes. The key concept underlying these data and approaches are that a systems biology approach (utilizing genomic and proteomic technologies) provides opportunities to characterize disease mechanisms in ARDS and VILI (Figure 26). Further expansion of ARDS biobanks will allow for further validation of promising pilot studies focusing on ARDS severity (mRNA seq and genome-wide methylation). Additional proteomics analysis will allow for structure-function analysis in endothelial cells.
Figure 26: A systems biology approach schematic to ARDS (candidate gene-based strategy). The future work in ARDS in Garcia lab relies upon utilizing multiple genomic/genetic and proteomic elements (DNA sequencing, RNA sequencing, DNA variant genotyping, and DNA methylation analysis) to further evaluate top candidate genes associated with endothelial cell barrier dysfunction for subphenotyping/therapeutic uses.
Summary of Results
128 | Lynn In Chapter 2, a database focused pathway analysis was the tool I utilized to
elucidate the many pathways and genes that have been associated with ARDS. While
there were over 201 genes with genetic, transcriptomic, and proteomic associations with
ARDS risk and outcome, 32 pathways were generated from this pathway database- focused literature-based design. Of these 32 pathways, they were further summarized into four broad categories with pathobiological significant in ARDS and based upon the over-lapping gene signatures: i) inflammatory/innate immune system based pathways ii)
ROS signaling iii) endothelial and vascular signaling pathways (including coagulation) and iv) transcription factors1. Independent validation of individual gene associations with
ARDS are crucial for establishing a biomarker (or biomarkers) for ARDS subphenotyping,
but these broad pathway trends should be kept in mind as to choose biomarkers that
reflect the broader, systemic changes of alveolar barrier damage.
A major difficulty in ARDS genetic research is the inability to compare studies due
to altering definitions of ARDS diagnosis, under-powered sample sizes, population
disparities, and a wide focus in study design and methodology. While this pathway
analysis does not capture the nuances of any particular ARDS omics field, it identifies the
broad biological pathways (inflammatory cytokines, ROS related, vascular remodeling,
and transcription factor relation) where the majority of ARDS associated genes across a
broad range of studies (genetic, transcriptome profiling, and proteomic) and study designs
(candidate gene risk, GWAS associated risk, candidate gene outcome association, and
GWAS outcome association) can be categorized. As a previous meta-analysis of ARDS
biomarkers showed, the study design (overall risk vs outcome severity) determines which
biomarker is the most effective at predicting the disease state in question67. A single
129 | Lynn biomarker of disease in ARDS is improbable due to the heterogeneity of the disease on-
set and progression12, 20, 21, 26, 27, 61, 85, 133, 157, 206, 327. We and others have also reported
significant racial and ethnic disparities in ARDS outcomes7, 10, 11, 310. A biomarker panel
approach has shown early promise at addressing the diversity present in treating and
diagnosing ARDS22, 27, 63, 67.
In Chapter 3, I focused on the associations between a single candidate gene
(NAMPT) and ARDS. NAMPT is one of several independently validated risk genes in
ARDS, which makes it an attractive target for an ARDS biomarker133. NAMPT was
reported as an early gene (as PBEF/visfatin) associated with ARDS in VILI and LPS
animal models77, 181, 210, 323, 326, 411. An extracellular form of NAMPT (later dubbed eNAMPT) was found to be a plasma protein associated with mortality62, 222. eNAMPT
binds to TLR4 and acts as a potent cytokine signal in endothelial cells183.
Beyond associations between NAMPT transcription, eNAMPT plasma proteins,
and ARDS injury and risk, several genotypes and SNPs have been associated with
ARDS, sepsis, and lung injury risk. A mechanically sensitive region (-2,428 to -1,228) in
the ARDS promoter (-3028 to 0 TSS) was reported to have several SNPs that altered
NAMPT promoter activity154. Within this promoter, several SNPs (rs7789066 at -2422, rs61330082 at -1535) were reported, and rs7789066 (-2422) increased NAMPT promoter activity. The rs61330082 (-1535) was found to have a minor allele was associated with decreased disease risk in ARDS, ALI, and sepsis68. In an additional study, the
rs61330082 G and rs9770242 C haplotype was shown to have a increased hazard ratio
for severe ARDS outcomes (both 28-day and 60-day mortality)187.
130 | Lynn NAMPT’s unique status with dual intracellular and extracellular roles in endothelial
cells presents an opportunity to use it as both a plasma and genetic biomarker.
Rs61330082 is associated with 28-day mortality risk in ARDS as well as increased
eNAMPT levels. Mechanistically, rs61330082 G also showed an increase in NAMPT promoter function in endothelial cells. The previously reported rs61330082 G/rs9770242
C haplotype (showing an increased hazard risk for ARDS 28-day mortality187) was found
to be associated with increased eNAMPT levels. Specifically, this haplotype was
associated with mortality risk in African American patients. A combination of the outcome
associated haplotype and plasma biomarker was utilized in a novel genetic risk score
model to predict mortality in ARDS.
Genetic risk scores have been utilized in prostate and colorectal cancer to stratify patients into distinct risk groups107, 109-112, 119, 121, 122. Over a decade of developing and
fine-tuning genetic risk scores have shown their predictive power in prostate cancer, and
a genetic risk score that utilized 5 distinct risk SNPs predicted prostate cancer risk better
than family history107, 119, 121. In critical care, small sample sizes and lack of high-powered
GWASs have hindered the utility of a genetic risk score. My study combines risk SNPs
and plasma biomarker levels with prior associations to severe ARDS outcomes (28-day
mortality)62, 68, 154, 187, 222, 412. A genetic risk score is a separate tool that could be used in conjunction with a biomarker panel to increase ARDS and mortality risk prediction clinically or in a clinical trial design.
In Chapter 4, I utilized a genome-wide approach to profile epigenetic changes
(methylation) in ARDS. In a diverse cohort (and after adjustment for gender and race/ethnicity), 15 significant CpG sites were found to be associated with ARDS mortality
131 | Lynn risk. Further, 3 genes in a genome-wide mRNA next generation exon sequencing were found to be associated with ARDS mortality risk (after adjustment for gender and race).
These data provide multi-omic evidence that a distinct genetic mortality signature for
ARDS could be developed, although this study is under-powered to do so. Genome-wide methylation trends were investigated in a well-phenotyped cohort (IGABS) curated in the last several years at the University of Arizona. Gene specific methylation changes and global methylation changes were reported in this independent cohort. Replicating
epigenetic changes in larger ARDS cohorts and combo VILI/LPS animal models will be
critical to making these data applicable to translational research in ARDS.
Future Directions
Clinical Phase II/III trials for ARDS therapeutics have been largely unsuccessful18,
20, 26, 27, 69. This is a critical problem because ~7% of ICU patients develop ARDS, and the
modern clinical mortality rate for ARDS typically fluctuates around 30%20, 21. To ensure a
generalizable treatment, large enrollment sizes in clinical trials are essential, but this
methodology may hinder optimal clinical trial designs in ARDS due to the heterogeneity
of causes of ARDS. The two major reasons to identify subphenotypes in ARDS clinical
trials are i) identifying patients that are at higher risk of dying based upon a variety of
clinical factors and ii) selecting a subset of patients that may be more responsive to
treatment20.
A powerful strategy to subphenotype ARDS patients is to utilize omics data. In a
complex genetic disease, each loci of associated risk typically have a relatively small odds ratio of associated risk, but the cumulative risk of these loci represents a larger and more
132 | Lynn complete fraction of risk. Genetic risk scores attempt to remedy this challenge by generating a risk score from top associated risk SNPs. However, due to the low numbers of GWAS associated SNPs, SNPs from candidate genes may have to be utilized. An additional strategy for stratifying ARDS outcome risk is by utilizing multiple plasma biomarkers63, 67. A combo strategy utilizing multiple omics data in an ARDS
subphenotyping risk model may provide the optimal strategy for subphenotyping higher-
risk ARDS cohorts for clinical trials.
Substantial epigenetic changes have been shown to be associated with increased
ARDS mortality151, 153. A genome-wide increase in methylation was found to be
associated with increased ARDS outcome severity risk. While these data are not yet
utilizable clinically, the success of circulating miRNA profiling to predict risk (via PBMCs) of epigenetic changes could be a relatively less invasive manner to utilize genomic biomarkers in critical care140, 142, 143, 145, 148. Changes in methylation, transcription, and
protein biomarkers have been reported utilizing PBMCs from ARDS patients.
Successfully incorporating a multi-omics approach to clinical trial disease and patients
subphenotyping provides an alternative strategy for identifying critically ill patients at risk
of developing severe ARDS27.
133 | Lynn Appendix A: Supplemental Data
Supplementary Table 1: Gene and Molecular Factor Names and Abbreviations
Abbreviation Common Gene Names
ABCB1 ATP binding cassette subfamily B member 1
ABCC1 ATP binding cassette subfamily C member 1
ACE Angiotensin-converting enzyme
ACTG1 Actin gamma 1
ADIPOQ Adioponectin C1Q and collagen domain containing
AGER Advanced glycosylation end-product specific receptor
AHR Aryl hydrocarbon receptor
AHSG Alpha 2-HS glycoprotein
AKT Serine/threonine kinase
ANG1 Angiogenin-1
ANG2 Angiogenin-2
ANXA1 Annexin A1
ANXA2 Annexin A2
AP-1 Activating protein-1
APOA1 Apolipoprotein A1
ARE Antioxidant response elements
AREG Amphiregulin
ARG1 Arginase 1
ARSD Arylsulfatase D
BAD BCL2 associated agonist of cell death
BPI Bactericidal permeability increasing protein
C3 Complement C3 precursor
134 | Lynn C9 Complement C9 precursor
CAP1 Cyclase associated actin cytoskeleton regulatory protein 1
CAT Catalase
CCL2 C-C motif chemokine ligand 2
CCR2 C-C motif chemokine receptor 2
CCT7 Chaperonin containing TCP1 subunit 7
CCT8 Chaperonin containing TCP1 subunit 8
CD24 CD24 surface antigen
CD36 CD36 surface antigen
CDKN1A Cyclin dependent kinase inhibitor 1A
CEACAM1 Carcinoembryonic antigen related cell adhesion molecule 1
CEACAM8 Carcinoembryonic antigen related cell adhesion molecule 8
CEBPD CCAAT enhancer binding protein delta
CHIT1 Chitinase 1
COTL1 Coactosin like F-acting binding protein 1
CSDE1 Cold shock domain containing E1
CSF2 Colony stimulating factor 2
CSF2RB Colony stimulating factor 2 receptor beta common subunit
CXCL2 C-X-C motif chemokine ligand 2
CXCL8 C-X-C motif chemokine ligand 8
CYBA P22phox
CYP1A1 Cytochrome P450 family 1 subfamily A member 1
DEFA4 Defensin alpha 4
DIO2 Iodothyronine deiodinase 2
DNAJA3 DNAJ heat shock protein family (HSP40) member A3
DNAJB6 DNAJ heat shock protein family (HSP40) member B6
DNAJB9 DNAJ heat shock protein family (HSP40) member B9
135 | Lynn DNAJB11 DNAJ heat shock protein family (HSP40) member B11
DNAJC5 DNAJ heat shock protein family (HSP40) member C5
DNAJC8 DNAJ heat shock protein family (HSP40) member C8
EGFR Epidermal growth factor receptor
eNAMPT Extracellular nicotinamide phosphoribosyl transferase
FAAH Fatty acid amide hydrolase
FABP5 Fatty acid binding protein 5
FER Malectin-like receptor kinase FERONIA
FGA Fibrinogen alpha chain
FLK2 FMS related tyrosine kinase 3
FTH1 Ferritin heavy chain 1
FTL Ferritin light chain
GADD45a growth arrest and DNA damage-inducible gene
GCLM Glutamate-cysteine ligase modifier subunit
GPI Glucose-6-phosphate isomerase
GSR Glutathione-disulfide reductase
GSTP1 Glutathione S-transferase pi 1
HBB Hemoglobin subunit beta
HBEGF Heparin binding EGF like growth factor
HEATR1 HEAT repeat containing 1
HMGB1 High mobility group box 1
HO-1 Heme oxygenase-1
HPX Hemopexin precursor
HSP90AB1 Heat shock protein 90 alpha family class B member 1
HSPA1A Heat shock protein family A (Hsp70) member 1A
HSPA8 Heat shock protein family A (Hsp70) member 8
ICAM-1 Intercellular adhesion molecule 1
136 | Lynn IL-1Β Interleukin-1 Β
IL-1R2 Interleukin-1 receptor type 2
IL-10 Interleukin-10
IL-13 Interleukin-13
IL-18 Interleukin-18
IL-6 Interleukin-6
IL-8 Interleukin-8
IRAK1 Interleukin-1 receptor-associated kinase
JMJD1C Jumonji domain containing 1C
JUN AP-1 family transcription factor
KEAP1 Kelch-like ECH associated protein 1
KL-6 Krebs von den Lugen-6
LCN2 Lipocalin 2
LDH Lactate dehydrogenase
LGALS3 S3 galectin 3
LRRC16A Leucine rich repeat containing 16A
MAPK Mitogen-activated protein kinase
MHC-DRB1 Major histocompatibility complex class II DR B1 domain
MIF Macrophage migration inhibitory factor
MMP8 Matrix metallopeptidase 8
MMP9 Matrix metallopeptidase 9
MYLK Encoding myosin light chain kinase
NAMPT Nicotinamide phosphoribosyl transferase
NCF1 Neutrophil cytosolic factor 1
NFE2L2 Nuclear factor erythroid 2 like 2
NF-кB Nuclear factor kappa B
NRF2 Nuclear factor erthroid 2 like 2
137 | Lynn nmMLCK Non-muscle myosin light chain kinase
NOS3 Nitric oxide synthase 3
NOX NADPH oxidase
NQPO1 NAD(P)H:quinone oxidoreductase 1
OLFM4 Olfactomedin 4
OSM Oncostatin M
PA-1 Pediocin PA-1
PBEF Nicotinate phosphribosyltransferase
PDE4B Phosphodiesterase 4B
PDX5 Peroxiredoxin 5 mitochondrial
PGAM1 Phosphoclycerate mutase 1
PHD2 Egl-9 family hypoxia inducible factor 1
PI3 Peptidase inhibitor 3
PI3K Peptidase inhibitor 3 kinase
PLAUR Plasminogen activator, urokinase receptor
POPDC3 Popeye domain containing 3
PPFIA1 PTPRF interacting protein alpha 1
PPIA Peptidylprolyl isomerase A
PRDX2 Peroxiredoxin 2
PRDX3 Peroxiredoxin 3
PRDX6 Peroxiredoxin 6
PTGS2 Prostaglandin-endoperoxide synthase 2
RAGE Receptor of advanced glycation end products
RANKL Nuclear factor kappa-B ligand
RBP7 Retinol binding protein 7
RNASE3 Ribonuclease A family member 3
S100A12 Calgranulin C
138 | Lynn S100A8 Calgranulin A
S100A9 Calgranulin B
S1P1 Sphingosine 1-phosphate 1
S1P3 Sphingosine 1-phosphate 3
S1P1 Sphingosine 1-phosphate receptor 1
S1P3 Sphingosine 1-phosphate receptor 3
SAA Serum amyloid protein
SELPLG Selectin P ligand
SERPINA1 Alpha-1-antitrypsin
SLMO2 PRELI domain containing 3B
SOD Superoxide dismutase
SOD2 Superoxide dismutase 2
SOD3 Superoxide dismutase 3
SP-B Surfactant-associated protein B
sRAGE Soluble receptor for advanced glycation end products
STAT Signal transducer and activator of transcription
STAT4 Signal transducer and activator of transcription 4
STAT5 Signal transducer and activator of transcription 5
TGFB2 Transforming growth factor beta 2
TIMP-1 Tissue inhibitor of metalloproteinases inhibitor 1
TLR1 Toll-like receptor 1
TLR4 Toll-like receptor 4
TNFAIP3 Tumor necrosis factor alpha induced protein 3
TNFR Tumor necrosis factor receptor
TNFRSF11A Tumor necrosis factor receptor superfamily member 11a
TRAF6 TNF receptor associated factor 6
TRIF TIR-domain containing adaptor-inducing interferon-B
139 | Lynn TXN Thioredoxin
VEGF Vascular endothelial growth factor
VEGFA Vascular endothelial growth factor A
vWF von Willebrand factor
UGT2B7 UDP glucuronosyltransferase family 2 member B7
UTS2 Urotensin 2
XKR3 XK-related 3
YWHAZ Tyrosine 3-monoxygenase/tryptophan 5-monooxygenase activation protein zeta
ZNF335 Zinc-Finger/Leucine-Zipper Co-Transducer NIF1
140 | Lynn Supplementary table 2: Genes used in CPDB pathway gene study
Proteomics ACTG1, AHSG, ANXA1, ANXA2, ANXA3, ANXA5, APOA1,
(BAL)90, 213, 249 ARHGDIB, B2M, C3, C4, C9, COL1A1, COL3A1, COL5A1,
COTL1, CP, CST2, CST4, ENO1, EZR, FABP4, FABP5, FGA,
FLNA, FTH1, FTL, GAPDHL6, GPI, GSTP1, HBB, HINT1, HPX,
HRG, HSPG2, MUC5AC, MYH9, PDIA3, PFN1, PGAM1, PGK1,
PPIA, PRDX2, PRDX5, RBP4, S100A8, S100A9, S100A12,
SAA1, SERPINA1, SFTP1, SOD3, STMN1, TMSB4X, TPI1,
TPM, TTR, TXN, VASP
Proteomic BPI, CD24, LCN2, MME, MMP8, OLFM4, RBP7,
(PAXgene)90 STMN1, UTS2
Gene Expression ADMR, ADPRH, AREG, ARG2, ARPC4, AQP1, ATF3, BHLHB2,
(Microarray)83, 130, 131 BTG1, BTG2, CCL2, CDKN1A, CEBPB, CEBPD, COX2, CXCR4,
CYCS, DNAJA1, EGFR, EIF2S1, EST, F3, FGA, GABRD,
GADD45A, GAPD, GBP2, GCLC, GJA1, HSPA8, IFIT1, IFRD1,
IFIT3, IL13, IL1B, IL1R2, IL6, ISG15, KCNJ6, LGALS3, MAT2A,
MX1, NR4A1, PLAUR, PRKAR2A, PTGS2, S100A9, S100A12,
SERPINE1, TCF21, TFF2, YWHAZ
141 | Lynn Gene Expression ARG, ANXA3, ATF3, BNIP3L, CEACAM1, CEACAM8,
(PBMC)‡70, 80, 159, 207 CCT7, CCT8, CHUK, CSDE1, CXCL2, CXCL8, CYBA,
CSDE1, DDIT3, DEF4A, DNAJA3, DNAJB6, DNAJB9,
DNAJB11, DNAJC5, DNAJC8, DUSP2, DUSP4, FGL2,
FTH1, GADD45a, GCLM, GPR177, GSR, GSTP1, HBEGF,
HSP90AB1, HSPA1A, IL1B, IL1R2, IL8, JUN, LCN2, LY75,
KEAP1, MGAM, MME, MMP8, MMP9, NCF1, NR4A2,
NR4A3, OLFM4, OSM, PI3, PIGF, PLAUR, PLK2, POLB,
PRDX2, PRDX3, PRDX5, PTGS2, RNASE3, SGK, SIK1,
SOD2, STAT1, TBCC, TCN1, TNFAIP3, VEGFA
Genetic Sequencing ABCB1, ACE, ADIPOQ, ADIPOR1, ADIPOR2, AGER, AGT,
(Candidate Genes)20, AGTR1, AHR, CAT, CSF2, CSF2RB, CXCL2, CYP1A1,
22, 30-32, 50-53, 55, 64, 65, 72, DIO2, GCLC, HMGB1, IL10, IL4, IL6, IREK3, LRRC16A,
73, 76, 81, 125-127, 135, 154, MAP3K1, MIF, MTHFR, MYLK, NAMPT/PBEF, NFE2L2,
174, 185, 188, 208, 209, 211, NOS3, NOX4, PHD2, PI3, S1P1, S1PR3, SOD2, SPB,
227, 230, 231, 235, 238, 240, SFTPB, TLR1, TNF, VEGF
262, 288, 294, 304, 308, 413,
414 Genome wide ABCC1, ADA, ADRBK2, ARSD, BCL11A, CBS, CHIT1,
association studies DYNC2H1, EPAS1, FAAH, FER, FLT1, FZD2, GPR98,
(GWASs)64, 72, 79, 96, GRM3, HEATR1, HSPG2, HTR2A, IL1RN, IL8RA, ISG15,
103, 104, 219, 232, 264, 415, ITGA1, KLK2, MAP3K6, PDE4B, POPDC3, PPFIA1,
416 PRKAG2, SELPLG, SFRS16, TACR2, TGFB2,
TNFRSF11A, UGT2B7, VLDLR, VWF, XKR3, ZNF335
‡ Unpublished data
142 | Lynn
Supplemental Table 3: Minor allele frequencies (MAFs) of 7 NAMPT promoter SNPs genotyped in ARDS European
and African American cohorts.
‡ rsID Function 1000 ARDS 1000 ARDS SNPs Genome European Genome African
European † African MAF* MAF MAF MAF
-2442 rs7789066 Alters Transcription factor 0.072 0.030 0.144 0.114
(A/G) binding
-2296 rs116647506 Alters Transcription factor, CpG 0.017 0.011 0.061 0.038
(A/G)
-1535 rs61330082 Alters Transcription factor 0.296 0.234 0.075 0.137
(G/A) binding
-1208 rs151175330 Alters Transcription factor 0.000 0.000 0.019 0.006
(G/A) binding
-1001 rs9770242 Alters Transcription factor 0.205 0.246 0.206 0.214
(A/C) binding
-948 rs59744560 Alters Transcription factor 0.116 0.156 0.006 0.041
(C/A) binding, CpG
-422 rs1319501 Alters Transcription factor 0.204 0.287 0.296 0.290
(T/C) binding
‡ SNP major/minor allele obtained from hg38 †N=204 *N=240
Supplementary Table 4: Mortality variation by DNA/Plasma cohort or center*
Fluid and University University University Consortium Genomic Total
Catheter Arizona Tennessee Washington Evaluate Association
Treatment Genomic Memphis (UW) Lung Studies
Trial associations (UTM) Edema (GAS)
(FACTT) and Genetics
(CELEG)
143 | Lynn
biomarkers
(GABS)
Survived 121 26 11 89 26 12 285
Deceased 59 8 11 12 23 18 131
Total 180 34 (23.5%) 22 (50.0%) 101 (11.9%) 49 (46.9%) 30 (60.0%) 416
(32.8%) (31.5%)
*na=58
144 | Lynn
Supplemental Table 5: Additive, recessive, and dominant genetic models for rs61330082.
Samples P-value Pearson χ2 P-value Pearson χ2 P-value Pearson χ2 SNP (n) Additive Additive Recessive Recessive Dominant Dominant
Total ARDS Cohort
rs61330082 393 0.025 7.405 0.036 6.631 0.201 3.207
European American ARDS cohort
rs61330082 172 0.110 7.547 0.109 4.426 0.829 0.375
African American Cohort
rs61330082 221 0.435 3.790 0.333 2.198 0.301 2.402
Supplementary Figure 1: Haplotypes for rs61330082/rs9770242 represented in 474
ARDS stratified into European and African American ARDS patients.
Supplemental Table 6
n (%) Unadjusted Adjusted
145 | Lynn
Survived Deceased HR (95% CI) PTrend HR (95% CI) PTrend
GG 230 (56.2) 89 (57.8) Reference 0.23 Reference 0.012
GA 142 (34.7) 60 (39.0) 1.10 (0.79-1.53) 1.12 (0.76-1.63)
AA 37 (9.0) 5 (3.2) 0.38 (0.15-0.93) 0.24 (0.08-0.78)
146 | Lynn
Supplemental Methods
PAH NAMPT analysis methods
Patient Samples. A total of 694 patients with IPAH and FPAH from the PAH Biobank were included. They were European American individuals who were age 18 or older at the time of enrollment.
Genotype data. Five SNPs (rs1319501, rs59744560, rs9770242, rs61330082, and
rs116647506) were selected based on previous studies showing associations with ARDS
risk and other pulmonary disease traits. Genotyping and imputation was performed as a
part of genome wide association study (GWAS) of PAH published previously417. Briefly,
Illumina HumanOmni5 array was used for genotyping. Illumina recommended protocol
was used for genotype calling of PAH patients using GenomeStudio. For each dataset,
SNPs with genotyping call rate <95%, minor allele frequency (MAF) <1%, and Hardy
Weinberg Equilibrium (HWE) P-value <0.001 were removed. SNPs with genotype call
rate <97%, MAF 3%, and HWE P-value <0.01 were removed after merging case and control datasets. We excluded potentially related individuals and individuals whose race/ethnicity was discordant between recorded clinical and genomic data. Ancestral background was assessed using 1000 Genomes Project date with principle component analysis. Michigan Imputation Server was used to impute ungenotyped variants418. Before
imputation, duplicated SNPs and SNPs with ambiguous alleles were removed, and
147 | Lynn discordant strand was corrected. Eagle version 2.3 was used for phasing of genotype data419. Minimac3 algorithm and the Haplotype Reference Consortium panel used for imputation420. After the imputation, SNPs with poor imputation quality (Rsq<0.3), MAF
0.1%, and HWE P-value <0.00001 were removed.
9
8
7
6
5
4
3
2 CpGcites (Count) 1
0 CpG location CpG Island 5' UTR Body/Exon TSS1500 3'UTR
Supplemental Figure 2
148 | Lynn
Appendix B: Related Publications
These data presented in this thesis has been previously utilized in publication,
presentations (oral and poster), and utilized in grants. Listed here are manuscripts and
their corresponding abstracts that have either been previously published or are in
submission and review where I contributed to substantially to the manuscript preparation
and have repurposed data from these works.
Heather Lynn1, Xiaoguang Sun1, Christian Bime1, Nancy Casanova1, Radu Oita1, Nikolas
Ramos1, Mark M. Wurfel2, Gianfranco U. Meduri3, David C. Christiani4, Dawn K. Coletta1,
Sara M. Camp1, Edward J. Bedrick5, Ken Batai1, Jason H. Karnes6,7, Yves A. Lussier1,
Nathan A. Ellis1, Joe G.N. Garcia1,6. NAMPT haplotypes and plasma eNAMPT levels predict ARDS mortality. Thorax (2020) In review.
Lynn H, Sun X, Casanova N, Gonzales-Garay M, Bime C, Garcia JGN. (2019). Genomic and
Genetic Approaches to Deciphering Acute Respiratory Distress Syndrome Risk and Mortality.
Antioxid Redox Signal, 31(14):1027-1052. Doi:10.1089/ars.2018.7701
149 | Lynn
References
1. Lynn H, Sun X, Casanova N, Gonzales Ganay M, Bime C, Garcia J. Genomic and Genetic Approaches to Deciphering Acute Respiratory Distress Syndrome Risk and Mortality. Redox and Anti-oxidant signaling.; 2019. 2. Adlung L, Kar S, Wagner M-C, et al. Protein abundance of AKT and ERK pathway components governs cell type-specific regulation of proliferation. Molecular Systems Biology. 01/24 2017;13doi:10.15252/msb.20167258 3. Cochi S, Kempker J, Annangi S, Kramer M, Martin G. Mortality Trends of Acute Respiratory Distress Syndrome in the United States from 1999 to 2013. Annals of the American Thoracic Society. 2016/10/01 2016;13(10):1742-1751. doi:10.1513/AnnalsATS.201512-841OC 4. Lynn H, Sun X, Casanova N, Gonzales-Garay M, Bime C, Garcia JGN. Genomic and Genetic Approaches to Deciphering Acute Respiratory Distress Syndrome Risk and Mortality. Antioxid Redox Signal. Nov 10 2019;31(14):1027-1052. doi:10.1089/ars.2018.7701 5. Ruhl AP, Huang M, Colantuoni E, et al. Healthcare utilization and costs in ARDS survivors: a 1-year longitudinal national US multicenter study. Intensive Care Medicine. 2017/07/01 2017;43(7):980-991. doi:10.1007/s00134-017-4827-8 6. Zimlichman E, Henderson D, Tamir O, et al. Health care–associated infections: A meta-analysis of costs and financial impact on the us health care system. JAMA Internal Medicine. 2013;173(22):2039- 2046. doi:10.1001/jamainternmed.2013.9763 7. Bime C, Poongkunran C, Borgstrom M, et al. Racial Differences in Mortality from Severe Acute Respiratory Failure in the United States, 2008–2012. Annals of the American Thoracic Society. 2016/12/01 2016;13(12):2184-2189. doi:10.1513/AnnalsATS.201605-359OC 8. Checkley W, Martin GS, Brown SM, et al. Structure, process, and annual ICU mortality across 69 centers: United States Critical Illness and Injury Trials Group Critical Illness Outcomes Study. Critical care medicine. 2014;42(2):344-356. doi:10.1097/CCM.0b013e3182a275d7 9. Cooke CR, Erickson SE, Eisner MD, Martin GS. Trends in the incidence of noncardiogenic acute respiratory failure: the role of race. Critical Care Medicine. 2012;40(5):1532-1538. doi:10.1097/CCM.0b013e31824518f2 10. Erickson SE, Shlipak MG, Martin GS, et al. Racial and Ethnic Disparities in Mortality from Acute Lung Injury. Critical care medicine. 2009;37(1):1-6. doi:10.1097/CCM.0b013e31819292ea 11. Moss M, Mannino D. Race and gender differences in acute respiratory distress syndrome deaths in the United States: An analysis of multiple-cause mortality data (1979–1996)*. Critical Care Medicine. 2002;30(8):1679-1685. 12. The ADTF. Acute respiratory distress syndrome: The berlin definition. JAMA. 2012;307(23):2526- 2533. doi:10.1001/jama.2012.5669 13. Dreyfuss D, Saumon G. Ventilator-induced Lung Injury. American Journal of Respiratory and Critical Care Medicine. 1998/01/01 1998;157(1):294-323. doi:10.1164/ajrccm.157.1.9604014 14. Du P, Zhang X, Huang C-C, et al. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics. 2010/11/30 2010;11(1):587. doi:10.1186/1471-2105-11-587 15. Eklund L, Saharinen P. Angiopoietin signaling in the vasculature. Experimental Cell Research. 2013/05/15/ 2013;319(9):1271-1280. doi:https://doi.org/10.1016/j.yexcr.2013.03.011 16. Shaver CM, Bastarache JA. Clinical and Biological Heterogeneity in ARDS: Direct versus Indirect Lung Injury. Clinics in chest medicine. 09/23 2014;35(4):639-653. doi:10.1016/j.ccm.2014.08.004 150 | Lynn
17. Frank AJ, Thompson BT. Pharmacological treatments for acute respiratory distress syndrome. Current Opinion in Critical Care. 2010;16(1):62-68. doi:10.1097/MCC.0b013e328334b151 18. Janz DR, Ware LB. Biomarkers of ALI/ARDS: Pathogenesis, Discovery, and Relevance to Clinical Trials. Semin Respir Crit Care Med. // 11.08.2013 2013;34(04):537-548. doi:10.1055/s-0033-1351124 19. Meyer N. Beyond SNPs—genetics, genomics and other ‘omic approaches to ARDS. Clinics in chest medicine. 09/23 2014;35(4):673-684. doi:10.1016/j.ccm.2014.08.006 20. Ware L. Prognostic determinants of acute respiratory distress syndrome in adults: Impact on clinical trial design. Critical Care Medicine. 2005;33(3):S217-S222. doi:10.1097/01.ccm.0000155788.39101.7e 21. Ware L, Koyama T, Billheimer D, et al. Prognostic and pathogenetic value of combining clinical and biochemical indices in patients with acute lung injury. Chest. 2010;137(2):288-296. doi:10.1378/chest.09- 1484 22. Ware L, Koyama T, Zhao Z, et al. Biomarkers of lung epithelial injury and inflammation distinguish severe sepsis patients with acute respiratory distress syndrome. Critical care (London, England). 2013;17(5):R253-R253. doi:10.1186/cc13080 23. Kallet RH, Zhuo H, Ho K, Lipnick MS, Gomez A, Matthay MA. Lung Injury Etiology and Other Factors Influencing the Relationship Between Dead-Space Fraction and Mortality in ARDS. 10.4187/respcare.05589. Respiratory Care. 2017;62(10):1241. 24. Murray JF, Matthay MA, Luce JM, Flick MR. An Expanded Definition of the Adult Respiratory Distress Syndrome. American Review of Respiratory Disease. 1988/09/01 1988;138(3):720-723. doi:10.1164/ajrccm/138.3.720 25. Chiu L-C, Tsai F-C, Hu H-C, et al. Survival Predictors in Acute Respiratory Distress Syndrome With Extracorporeal Membrane Oxygenation. The Annals of Thoracic Surgery. 2015/01/01/ 2015;99(1):243- 250. doi:https://doi.org/10.1016/j.athoracsur.2014.07.064 26. Kangelaris KN, Calfee CS, May AK, Zhuo H, Matthay MA, Ware LB. Is there still a role for the lung injury score in the era of the Berlin definition ARDS? Annals of intensive care. 2014;4(1):4-4. doi:10.1186/2110-5820-4-4 27. Calfee C, Delucchi K, Parsons P, et al. Subphenotypes in acute respiratory distress syndrome: latent class analysis of data from two randomised controlled trials. The Lancet Respiratory medicine. 2014;2(8):611-620. doi:10.1016/S2213-2600(14)70097-9 28. Tzotzos SJ, Fischer B, Fischer H, Zeitlinger M. Incidence of ARDS and outcomes in hospitalized patients with COVID-19: a global literature survey. Critical Care. 2020/08/21 2020;24(1):516. doi:10.1186/s13054-020-03240-7 29. Hojyo S, Uchida M, Tanaka K, et al. How COVID-19 induces cytokine storm with high mortality. Inflammation and Regeneration. 2020/10/01 2020;40(1):37. doi:10.1186/s41232-020-00146-3 30. Yilin Z, Yandong N, Faguang J. Role of angiotensin-converting enzyme (ACE) and ACE2 in a rat model of smoke inhalation induced acute respiratory distress syndrome. Burns. 2015/11/01/ 2015;41(7):1468-1477. doi:https://doi.org/10.1016/j.burns.2015.04.010 31. Salnikova LE, Smelaya TV, Vesnina IN, Golubev AM, Moroz VV. Genetic Susceptibility to Nosocomial Pneumonia, Acute Respiratory Distress Syndrome and Poor Outcome in Patients at Risk of Critical Illness. Inflammation. 2014/04/01 2014;37(2):295-305. doi:10.1007/s10753-013-9740-x 32. Wösten-van Asperen RM, Lutter R, Specht PA, et al. Acute respiratory distress syndrome leads to reduced ratio of ACE/ACE2 activities and is prevented by angiotensin-(1–7) or an angiotensin II receptor antagonist. The Journal of Pathology. 2011/12/01 2011;225(4):618-627. doi:10.1002/path.2987
151 | Lynn
33. Mirzapoiazova T, Moitra J, Moreno-Vinasco L, et al. Non–Muscle Myosin Light Chain Kinase Isoform Is a Viable Molecular Target in Acute Inflammatory Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 01/ 2009;44(1):40-52. 34. Marshall RP, Webb S, Bellingan GJ, et al. Angiotensin Converting Enzyme Insertion/Deletion Polymorphism Is Associated with Susceptibility and Outcome in Acute Respiratory Distress Syndrome. American Journal of Respiratory and Critical Care Medicine. 2002/09/01 2002;166(5):646-650. doi:10.1164/rccm.2108086 35. Tiret L, Rigat B, Visvikis S, et al. Evidence, from combined segregation and linkage analysis, that a variant of the angiotensin I-converting enzyme (ACE) gene controls plasma ACE levels. American Journal of Human Genetics. 1992;51(1):197-205. 36. Matsuyama T, Kubli SP, Yoshinaga SK, Pfeffer K, Mak TW. An aberrant STAT pathway is central to COVID-19. Cell Death & Differentiation. 2020/10/09 2020;doi:10.1038/s41418-020-00633-7 37. Mangalmurti NS, Reilly JP, Cines DB, Meyer NJ, Hunter CA, Vaughan AE. COVID-19–associated Acute Respiratory Distress Syndrome Clarified: A Vascular Endotype? American Journal of Respiratory and Critical Care Medicine. 2020/09/01 2020;202(5):750-753. doi:10.1164/rccm.202006-2598LE 38. Huertas A, Montani D, Savale L, et al. Endothelial cell dysfunction: a major player in SARS-CoV-2 infection (COVID-19)? The European respiratory journal. 2020;56(1):2001634. doi:10.1183/13993003.01634-2020 39. Dudek SM, Garcia JGN. Cytoskeletal regulation of pulmonary vascular permeability. Journal of Applied Physiology. 2001;91(4):1487-1500. 40. Garcia JG, Lazar V, Gilbert-McClain LI, Gallagher PJ, Verin AD. Myosin light chain kinase in endothelium: molecular cloning and regulation. American Journal of Respiratory Cell and Molecular Biology. 1997/05/01 1997;16(5):489-494. doi:10.1165/ajrcmb.16.5.9160829 41. Birukov KG, Jacobson JR, Flores AA, et al. Magnitude-dependent regulation of pulmonary endothelial cell barrier function by cyclic stretch. 10.1152/ajplung.00336.2002. American Journal of Physiology - Lung Cellular and Molecular Physiology. 2003;285(4):L785. 42. Alexander DV, Lydia IG-M, Carolyn EP, Joe GNG. Biochemical Regulation of the Nonmuscle Myosin Light Chain Kinase Isoform in Bovine Endothelium. American Journal of Respiratory Cell and Molecular Biology. 1998;19(5):767-776. doi:10.1165/ajrcmb.19.5.3126 43. Vouret-Craviari V, Boquet P, Pouysségur J, Van Obberghen-Schilling E. Regulation of the actin cytoskeleton by thrombin in human endothelial cells: role of Rho proteins in endothelial barrier function. Molecular biology of the cell. 1998;9(9):2639-2653. doi:10.1091/mbc.9.9.2639 44. Chow C-W, Abreu MTH, Suzuki T, Downey GP. Oxidative Stress and Acute Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2003;29(4):427-431. doi:10.1165/rcmb.F278 45. Wozney JL, Antonarakis ES. Growth factor and signaling pathways and their relevance to prostate cancer therapeutics. Cancer metastasis reviews. 2014;33(2-3):581-594. doi:10.1007/s10555-013-9475-z 46. Fu P, Usatyuk PV, Jacobson J, et al. Role played by paxillin and paxillin tyrosine phosphorylation in hepatocyte growth factor/sphingosine-1-phosphate-mediated reactive oxygen species generation, lamellipodia formation, and endothelial barrier function. Pulmonary Circulation. 05/18 2015;5(4):619- 630. doi:10.1086/683693 47. Xie C-R, Sun H, Wang F-Q, et al. Integrated analysis of gene expression and DNA methylation changes induced by hepatocyte growth factor in human hepatocytes. Molecular medicine reports. 2015;12(3):4250-4258. doi:10.3892/mmr.2015.3974 48. Hu S, Li J, Xu X, et al. The hepatocyte growth factor-expressing character is required for mesenchymal stem cells to protect the lung injured by lipopolysaccharide in vivo. Stem Cell Research & Therapy. 2016/04/29 2016;7(1):66. doi:10.1186/s13287-016-0320-5
152 | Lynn
49. Sammani S, Moreno-Vinasco L, Mirzapoiazova T, et al. Differential Effects of Sphingosine 1– Phosphate Receptors on Airway and Vascular Barrier Function in the Murine Lung. American Journal of Respiratory Cell and Molecular Biology. 2010/10/01 2010;43(4):394-402. doi:10.1165/rcmb.2009-0223OC 50. Sun X, Ma S-F, Wade MS, et al. Functional variants of sphingosine-1-phosphate receptor 1 gene associate with asthma susceptibility. The Journal of allergy and clinical immunology. 08/ 2010;126(2):241- 249.e3. 51. Sun X, Singleton PA, Letsiou E, et al. Sphingosine-1-phosphate receptor-3 is a novel biomarker in acute lung injury. American journal of respiratory cell and molecular biology. 2012;47(5):628-636. doi:10.1165/rcmb.2012-0048OC 52. Sun X, Ma S-F, Wade MS, et al. Functional promoter variants in sphingosine 1-phosphate receptor 3 associate with susceptibility to sepsis-associated acute respiratory distress syndrome. American Journal of Physiology - Lung Cellular and Molecular Physiology. 2013;305(7):L467-L477. doi:10.1152/ajplung.00010.2013 53. Tsao P-N, Wei S-C, Huang M-T, et al. Lipopolysaccharide-induced Notch signaling activation through JNK-dependent pathway regulates inflammatory response. Journal of biomedical science. 2011;18(1):56-56. doi:10.1186/1423-0127-18-56 54. Shibuya M. Vascular endothelial growth factor and its receptor system: physiological functions in angiogenesis and pathological roles in various diseases. Journal of Biochemistry. 2013;153(1):13-19. doi:10.1093/jb/mvs136 55. Shimizu Y, Camp SM, Sun X, Zhou T, Wang T, Garcia JGN. Sp1-mediated nonmuscle myosin light chain kinase expression and enhanced activity in vascular endothelial growth factor-induced vascular permeability. Pulmonary circulation. 2015;5(4):707-715. doi:10.1086/684124 56. Song Y-J, Cao J-Y, Jin Z, et al. Inhibition of microRNA-132 attenuates inflammatory response and detrusor fibrosis in rats with interstitial cystitis via the JAK-STAT signaling pathway. Journal of Cellular Biochemistry. 2019/06/01 2019;120(6):9147-9158. doi:10.1002/jcb.28190 57. Tian Y, Yao J, Liu S, et al. Identification and expression analysis of 26 oncogenes of the receptor tyrosine kinase family in channel catfish after bacterial infection and hypoxic stress. Comparative Biochemistry and Physiology Part D: Genomics and Proteomics. 2015/06/01/ 2015;14:16-25. doi:https://doi.org/10.1016/j.cbd.2015.02.001 58. Aschner Y, Zemans RL, Yamashita CM, Downey GP. Matrix Metalloproteinases and Protein Tyrosine Kinases: Potential Novel Targets in Acute Lung Injury and ARDS. Chest. 2014/10/01/ 2014;146(4):1081-1091. doi:https://doi.org/10.1378/chest.14-0397 59. Meyer N, Huang Y, Singleton P, et al. GADD45a is a novel candidate gene in inflammatory lung injury via influences on Akt signaling. The FASEB Journal. 2009;23(5):1325-1337. doi:10.1096/fj.08-119073 60. Mitra S, Wade MS, Sun X, et al. GADD45a Promoter Regulation by a Functional Genetic Variant Associated with Acute Lung Injury. PLOS ONE. 2014;9(6):e100169. doi:10.1371/journal.pone.0100169 61. Famous KR, Delucchi K, Ware LB, et al. Acute Respiratory Distress Syndrome Subphenotypes Respond Differently to Randomized Fluid Management Strategy. American Journal of Respiratory and Critical Care Medicine. 2017/02/01 2016;195(3):331-338. doi:10.1164/rccm.201603-0645OC 62. Oita RC, Bime C, Casanova N, Camp S, Wang T, Garcia GJ. Blood eNAMPT is an ARDS and VILI biomarker differentially expressed and secreted by lung epithelium and endothelium in response to mechanical stress. Am J Respir Cell Mol Biol. 2020; 63. Bime C, Casanova N, Oita R, et al. Development of a biomarker mortality risk model in acute respiratory distress syndrome. Critical care (London, England). 2019;23(1):410-410. doi:10.1186/s13054- 019-2697-x
153 | Lynn
64. Wang Z, Beach D, Su L, Zhai R, Christiani D. A Genome-Wide Expression Analysis in Blood Identifies Pre-Elafin as a Biomarker in ARDS. American Journal of Respiratory Cell and Molecular Biology. 2008;38(6):724-732. doi:10.1165/rcmb.2007-0354OC 65. Gao L, Flores C, Fan-Ma S, et al. Macrophage migration inhibitory factor in acute lung injury: expression, biomarker, and associations. Translational research : the journal of laboratory and clinical medicine. 2007;150(1):18-29. doi:10.1016/j.trsl.2007.02.007 66. Oita R, Camp S, Ma W, et al. Novel Mechanism for Nicotinamide Phosphoribosyltransferase Inhibition of TNF-α-mediated Apoptosis in Human Lung Endothelial Cells. American journal of respiratory cell and molecular biology. 2018;59(1):36-44. doi:10.1165/rcmb.2017-0155OC 67. Terpstra ML, Aman J, van Nieuw Amerongen GP, Groeneveld ABJ. Plasma Biomarkers for Acute Respiratory Distress Syndrome: A Systematic Review and Meta-Analysis*. Critical Care Medicine. 2014;42(3):691-700. doi:10.1097/01.ccm.0000435669.60811.24 68. Ye S, Simon B, Maloney J, et al. Pre–B-Cell Colony-enhancing Factor as a Potential Novel Biomarker in Acute Lung Injury. American Journal of Respiratory and Critical Care Medicine. 2005/02/15 2005;171(4):361-370. doi:10.1164/rccm.200404-563OC 69. Ware L, Calfee C. Biomarkers of ARDS: what’s new? Intensive Care Medicine. 2016/05/01 2016;42(5):797-799. doi:10.1007/s00134-015-3973-0 70. Wacharasint P, Nakada TA, Boyd John H, Russell James A, Walley Keith R. AA genotype of IL‐8 −251A/T is associated with low PaO2/FiO2 in critically ill patients and with increased IL‐8 expression. Respirology. 2012;17(8):1253-1260. doi:10.1111/j.1440-1843.2012.02244.x 71. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood. 1996;87(6):2095. 72. Parsons PE, Moss M, Vannice JL, Moore EE, Moore FA, Repine JE. Circulating IL-1ra and IL-10 levels are increased but do not predict the development of acute respiratory distress syndrome in at-risk patients. American Journal of Respiratory and Critical Care Medicine. 1997/04/01 1997;155(4):1469-1473. doi:10.1164/ajrccm.155.4.9105096 73. Kovach MA, Stringer KA, Bunting R, et al. Microarray analysis identifies IL-1 receptor type 2 as a novel candidate biomarker in patients with acute respiratory distress syndrome. Respiratory Research. 2015;16(1):29. doi:10.1186/s12931-015-0190-x 74. Moreno-Vinasco L, Quijada H, Sammani S, et al. Nicotinamide Phosphoribosyltransferase Inhibitor Is a Novel Therapeutic Candidate in Murine Models of Inflammatory Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2014/08/01 2014;51(2):223-228. doi:10.1165/rcmb.2012-0519OC 75. Meduri GU, Headley S, Kohler G, et al. Persistent Elevation of Inflammatory Cytokines Predicts a Poor Outcome in ARDS: Plasma IL-1β and IL-6 Levels Are Consistent and Efficient Predictors of Outcome Over Time. Chest. 1995/04/01/ 1995;107(4):1062-1073. doi:https://doi.org/10.1378/chest.107.4.1062 76. Goldman JL, Sammani S, Kempf C, et al. Pleiotropic effects of interleukin-6 in a “two-hit” murine model of acute respiratory distress syndrome. Pulmonary Circulation. 2014;4(2):280-288. doi:10.1086/675991 77. Jia SH, Li Y, Parodo J, et al. Pre–B cell colony–enhancing factor inhibits neutrophil apoptosis in experimental inflammation and clinical sepsis. Journal of Clinical Investigation. 2004;113(9):1318-1327. doi:10.1172/JCI200419930 78. Pociot F, MØLvig J, Wogensen L, Worsaae H, Nerup J. A Taql polymorphism in the human interleukin-1β (IL-1β) gene correlates with IL-1β secretion in vitro. European Journal of Clinical Investigation. 1992/06/01 1992;22(6):396-402. doi:10.1111/j.1365-2362.1992.tb01480.x
154 | Lynn
79. Meyer N, Feng R, Li M, et al. IL1RN Coding Variant Is Associated with Lower Risk of Acute Respiratory Distress Syndrome and Increased Plasma IL-1 Receptor Antagonist. American Journal of Respiratory and Critical Care Medicine. 2013;187(9):950-959. doi:10.1164/rccm.201208-1501OC 80. Wurfel MM. Microarray-based Analysis of Ventilator-induced Lung Injury. Proceedings of the American Thoracic Society. 2007;4(1):77-84. doi:10.1513/pats.200608-149JG 81. Wurfel MM, Gordon AC, Holden TD, et al. Toll-like Receptor 1 Polymorphisms Affect Innate Immune Responses and Outcomes in Sepsis. American Journal of Respiratory and Critical Care Medicine. 2008;178(7):710-720. doi:10.1164/rccm.200803-462OC 82. Simon BA, Easley RB, Grigoryev DN, et al. Microarray analysis of regional cellular responses to local mechanical stress in acute lung injury. 10.1152/ajplung.00463.2005. American Journal of Physiology - Lung Cellular and Molecular Physiology. 2006;291(5):L851. 83. Grigoryev D, Ma S-F, Irizarry R, Ye S, Quackenbush J, Garcia J. Orthologous gene-expression profiling in multi-species models: search for candidate genes. Genome Biology. 2004;5(5):R34-R34. 84. Grigoryev DN, Finigan JH, Hassoun P, Garcia JGN. Science review: Searching for gene candidates in acute lung injury. Critical Care. 06/30 2004;8(6):440-447. doi:10.1186/cc2901 85. Calfee CS, Delucchi KL, Sinha P, et al. Acute respiratory distress syndrome subphenotypes and differential response to simvastatin: secondary analysis of a randomised controlled trial. The Lancet Respiratory medicine. 2018;6(9):691-698. doi:10.1016/S2213-2600(18)30177-2 86. Matthay M, Zimmerman G, Esmon C, et al. Future Research Directions in Acute Lung Injury. American Journal of Respiratory and Critical Care Medicine. 2003/04/01 2003;167(7):1027-1035. doi:10.1164/rccm.200208-966WS 87. Matthay M, Brower R, Carson S, et al. Randomized, Placebo-controlled Clinical Trial of an Aerosolized β2-Agonist for Treatment of Acute Lung Injury. American Journal of Respiratory and Critical Care Medicine. 2011/09/01 2011;184(5):561-568. doi:10.1164/rccm.201012-2090OC 88. Fremont RD, Koyama T, Calfee CS, et al. Acute Lung Injury in Patients With Traumatic Injuries: Utility of a Panel of Biomarkers for Diagnosis and Pathogenesis. Journal of Trauma and Acute Care Surgery. 2010;68(5) 89. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: A severity of disease classification system. Critical Care Medicine. 1985;13(10) 90. Kangelaris KN, Prakash A, Liu KD, et al. Increased expression of neutrophil-related genes in patients with early sepsis-induced ARDS. American Journal of Physiology - Lung Cellular and Molecular Physiology. 2015;308(11):L1102-L1113. doi:10.1152/ajplung.00380.2014 91. Dong S, Wang E, Hsie L, Cao Y, Chen X, Gingeras TR. Flexible Use of High-Density Oligonucleotide Arrays for Single-Nucleotide Polymorphism Discovery and Validation. Genome Research. 2001;11(8):1418- 1424. 92. Bacolod MD, Schemmann GS, Giardina SF, Paty P, Notterman DA, Barany F. Emerging Paradigms in Cancer Genetics: Some Important Findings from High-Density Single Nucleotide Polymorphism Array Studies. Cancer Research. 2009;69(3):723. doi:10.1158/0008-5472.CAN-08-3543 93. Carr SM, Marshall HD. Intraspecific Phylogeographic Genomics From Multiple Complete mtDNA Genomes in Atlantic Cod (Gadus morhua): Origins of the “Codmother,” Transatlantic Vicariance and Midglacial Population Expansion. Genetics. 2008;180(1):381. doi:10.1534/genetics.108.089730 94. Wenzl P, Carling J, Kudrna D, et al. Diversity Arrays Technology (DArT) for whole-genome profiling of barley. Proceedings of the National Academy of Sciences of the United States of America. 2004;101(26):9915. doi:10.1073/pnas.0401076101 155 | Lynn
95. McVean GA, Altshuler DM, Durbin RM, et al. An integrated map of genetic variation from 1,092 human genomes. Nature. 2012/11/01 2012;491(7422):56-65. doi:10.1038/nature11632 96. Christie JD, Wurfel MM, Feng R, et al. Genome Wide Association Identifies PPFIA1 as a Candidate Gene for Acute Lung Injury Risk Following Major Trauma. PLOS ONE. 2012;7(1):e28268. doi:10.1371/journal.pone.0028268 97. Ma S-F, Flores C, Dudek SM, Nicolae DL, Ober C, Garcia JGN. A Common Cortactin Gene Variation Confers Differential Susceptibility to Severe Asthma. Genetic epidemiology. 2008;32(8):757-766. doi:10.1002/gepi.20343 98. Torgerson DG, Ampleford EJ, Chiu GY, et al. Meta-analysis of genome-wide association studies of asthma in ethnically diverse North American populations. Nature genetics. 2011;43(9):887-892. doi:10.1038/ng.888 99. Demenais F, Margaritte-Jeannin P, Barnes KC, et al. Multiancestry association study identifies new asthma risk loci that colocalize with immune-cell enhancer marks. Nature Genetics. 2018/01/01 2018;50(1):42-53. doi:10.1038/s41588-017-0014-7 100. Moffatt MF, Kabesch M, Liang L, et al. Genetic variants regulating ORMDL3 expression contribute to the risk of childhood asthma. Nature. 2007/07/01 2007;448(7152):470-473. doi:10.1038/nature06014 101. Vercelli D. Discovering susceptibility genes for asthma and allergy. Nature Reviews Immunology. 2008/03/01 2008;8(3):169-182. doi:10.1038/nri2257 102. Moffatt MF, Gut IG, Demenais F, et al. A large-scale, consortium-based genomewide association study of asthma. The New England journal of medicine. 2010;363(13):1211-1221. doi:10.1056/NEJMoa0906312 103. Shortt K, Chaudhary S, Grigoryev D, et al. Identification of Novel Single Nucleotide Polymorphisms Associated with Acute Respiratory Distress Syndrome by Exome-Seq. PLOS ONE. 2014;9(11):e111953. doi:10.1371/journal.pone.0111953 104. Bime C, Pouladi N, Sammani S, et al. Genome Wide Association Study in African Americans with Acute Respiratory Distress Syndrome Identifies the Selectin P Ligand Gene as a Risk Factor. American Journal of Respiratory and Critical Care Medicine. 2018;doi:10.1164/rccm.201705-0961OC 105. Desikan R, Fan C, Wang Y, et al. Genetic assessment of age-associated Alzheimer disease risk: Development and validation of a polygenic hazard score. PLOS Medicine. 2017;14(3):e1002258. doi:10.1371/journal.pmed.1002258 106. Kloosterman M, Santema BT, Roselli C, et al. Genetic risk and atrial fibrillation in patients with heart failure. European Journal of Heart Failure. 2020/01/09 2020;n/a(n/a)doi:10.1002/ejhf.1735 107. Seibert TM, Fan CC, Wang Y, et al. Polygenic hazard score to guide screening for aggressive prostate cancer: development and validation in large scale cohorts. BMJ (Clinical research ed). 2018;360:j5757-j5757. doi:10.1136/bmj.j5757 108. Gladding PA, Legget M, Fatkin D, Larsen P, Doughty R. Polygenic Risk Scores in Coronary Artery Disease and Atrial Fibrillation. Heart, Lung and Circulation. 2019/12/27/ 2019;doi:https://doi.org/10.1016/j.hlc.2019.12.004 109. Al-Tassan NA, Whiffin N, Hosking FJ, et al. A new GWAS and meta-analysis with 1000Genomes imputation identifies novel risk variants for colorectal cancer. Scientific reports. 2015;5:10442-10442. doi:10.1038/srep10442 110. Berndt SI, Potter JD, Hazra A, et al. Pooled analysis of genetic variation at chromosome 8q24 and colorectal neoplasia risk. Human molecular genetics. 2008;17(17):2665-2672. doi:10.1093/hmg/ddn166 111. Study C, Houlston RS, Webb E, et al. Meta-analysis of genome-wide association data identifies four new susceptibility loci for colorectal cancer. Nature genetics. 2008;40(12):1426-1435. doi:10.1038/ng.262
156 | Lynn
112. Houlston RS, Cheadle J, Dobbins SE, et al. Meta-analysis of three genome-wide association studies identifies susceptibility loci for colorectal cancer at 1q41, 3q26.2, 12q13.13 and 20q13.33. Nature genetics. 2010;42(11):973-977. doi:10.1038/ng.670 113. Tomlinson I, Webb E, Carvajal-Carmona L, et al. A genome-wide association scan of tag SNPs identifies a susceptibility variant for colorectal cancer at 8q24.21. Nature Genetics. 2007/08/01 2007;39(8):984-988. doi:10.1038/ng2085 114. Tomlinson IPM, Webb E, Carvajal-Carmona L, et al. A genome-wide association study identifies colorectal cancer susceptibility loci on chromosomes 10p14 and 8q23.3. Nature Genetics. 2008/05/01 2008;40(5):623-630. doi:10.1038/ng.111 115. Tomlinson IPM, Carvajal-Carmona LG, Dobbins SE, et al. Multiple Common Susceptibility Variants near BMP Pathway Loci GREM1, BMP4, and BMP2 Explain Part of the Missing Heritability of Colorectal Cancer. PLOS Genetics. 2011;7(6):e1002105. doi:10.1371/journal.pgen.1002105 116. Jaeger E, Webb E, Howarth K, et al. Common genetic variants at the CRAC1 (HMPS) locus on chromosome 15q13.3 influence colorectal cancer risk. Nature Genetics. 2008/01/01 2008;40(1):26-28. doi:10.1038/ng.2007.41 117. Tenesa A, Farrington SM, Prendergast JGD, et al. Genome-wide association scan identifies a colorectal cancer susceptibility locus on 11q23 and replicates risk loci at 8q24 and 18q21. Nature genetics. 2008;40(5):631-637. doi:10.1038/ng.133 118. Whiffin N, Hosking FJ, Farrington SM, et al. Identification of susceptibility loci for colorectal cancer in a genome-wide meta-analysis. Human molecular genetics. 2014;23(17):4729-4737. doi:10.1093/hmg/ddu177 119. Zanke BW, Greenwood CMT, Rangrej J, et al. Genome-wide association scan identifies a colorectal cancer susceptibility locus on chromosome 8q24. Nature Genetics. 2007/08/01 2007;39(8):989-994. doi:10.1038/ng2089 120. Peters U, Jiao S, Schumacher FR, et al. Identification of Genetic Susceptibility Loci for Colorectal Tumors in a Genome-Wide Meta-analysis. Gastroenterology. 2013;144(4):799-807.e24. doi:10.1053/j.gastro.2012.12.020 121. Zheng S, Sun J, Wiklund F, et al. Cumulative Association of Five Genetic Variants with Prostate Cancer. New England Journal of Medicine. 2008/02/28 2008;358(9):910-919. doi:10.1056/NEJMoa075819 122. Weigl K, Chang-Claude J, Knebel P, Hsu L, Hoffmeister M, Brenner H. Strongly enhanced colorectal cancer risk stratification by combining family history and genetic risk score. Clinical epidemiology. 2018;10:143-152. doi:10.2147/CLEP.S145636 123. Frampton MJE, Law P, Litchfield K, et al. Implications of polygenic risk for personalised colorectal cancer screening. Annals of Oncology. 2016;27(3):429-434. doi:10.1093/annonc/mdv540 124. Acosta-Herrera M, Pino-Yanes M, Perez-Mendez L, Villar J, Flores C. Assessing the quality of studies supporting genetic susceptibility and outcomes of ARDS. Frontiers in genetics. 2014;5:20-20. doi:10.3389/fgene.2014.00020 125. Lin Z, Pearson C, Chinchilli V, et al. Polymorphisms of human SP‐A, SP‐B, and SP‐D genes: association of SP‐B Thr131Ile with ARDS. Clinical Genetics. 2000/09/01 2001;58(3):181-191. doi:10.1034/j.1399-0004.2000.580305.x 126. O'Mahony DS, Glavan BJ, Holden TD, et al. Inflammation and Immune-Related Candidate Gene Associations with Acute Lung Injury Susceptibility and Severity: A Validation Study. PLOS ONE. 2012;7(12):e51104. doi:10.1371/journal.pone.0051104 127. Belperio JA, Keane MP, Burdick MD, et al. Critical role for CXCR2 and CXCR2 ligands during the pathogenesis of ventilator-induced lung injury. The Journal of clinical investigation. 2002;110(11):1703- 1716. doi:10.1172/JCI15849 128. Rose Jr CE, Sung S-SJ, Fu SM. Significant Involvement of CCL2 (MCP-1) in Inflammatory Disorders of the Lung. Microcirculation. 2003/07/01 2003;10(3‐4):273-288. doi:10.1038/sj.mn.7800193 157 | Lynn
129. Dixon AE, Mandac JB, Madtes DK, Martin PJ, Clark JG. Chemokine expression in Th1 cell-induced lung injury: prominence of IFN-γ-inducible chemokines. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2000/09/01 2000;279(3):L592-L599. doi:10.1152/ajplung.2000.279.3.L592 130. Altemeier WA, Matute-Bello G, Frevert CW, et al. Mechanical ventilation with moderate tidal volumes synergistically increases lung cytokine response to systemic endotoxin. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2004/09/01 2004;287(3):L533-L542. doi:10.1152/ajplung.00004.2004 131. Copland IB, Kavanagh BP, Engelberts D, McKerlie C, Belik J, Post M. Early Changes in Lung Gene Expression due to High Tidal Volume. American Journal of Respiratory and Critical Care Medicine. 2003/11/01 2003;168(9):1051-1059. doi:10.1164/rccm.200208-964OC 132. Dolinay T, Kaminski N, Felgendreher M, et al. Gene expression profiling of target genes in ventilator-induced lung injury. Physiological Genomics. 2006/06/01 2006;26(1):68-75. doi:10.1152/physiolgenomics.00110.2005 133. Reilly JP, Christie JD, Meyer NJ. Fifty Years of Research in ARDS.Genomic Contributions and Opportunities. American Journal of Respiratory and Critical Care Medicine. 2017/11/01 2017;196(9):1113- 1121. doi:10.1164/rccm.201702-0405CP 134. van den Borst B, Souren NY, Gielen M, et al. Association between the IL6-174G/C SNP and maximally attained lung function. 10.1136/thx.2009.128108. Thorax. 2011;66(2):179. 135. Flores C, Ma S-F, Maresso K, Wade MS, Villar J, Garcia JGN. IL6 gene-wide haplotype is associated with susceptibility to acute lung injury. Translational Research. 2008/07/01/ 2008;152(1):11-17. doi:https://doi.org/10.1016/j.trsl.2008.05.006 136. Martín-Loeches I, Solé-Violán J, Rodríguez de Castro F, et al. Variants at the promoter of the interleukin-6 gene are associated with severity and outcome of pneumococcal community-acquired pneumonia. Intensive Care Medicine. 2012/02/01 2012;38(2):256-262. doi:10.1007/s00134-011-2406-y 137. Qi D, Wang D, Zhang C, et al. Vaspin protects against LPS-induced ARDS by inhibiting inflammation, apoptosis and reactive oxygen species generation in pulmonary endothelial cells via the Akt/GSK-3β pathway. International Journal of Molecular Medicine. 2017;40(6):1803-1817. doi:10.3892/ijmm.2017.3176 138. Ozolina A, Sarkele M, Sabelnikovs O, et al. Activation of Coagulation and Fibrinolysis in Acute Respiratory Distress Syndrome: A Prospective Pilot Study. Frontiers in Medicine. 2016;3:64. doi:10.3389/fmed.2016.00064 139. Prabhakaran P, Ware LB, White KE, Cross MT, Matthay MA, Olman MA. Elevated levels of plasminogen activator inhibitor-1 in pulmonary edema fluid are associated with mortality in acute lung injury. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2003/07/01 2003;285(1):L20-L28. doi:10.1152/ajplung.00312.2002 140. Cortez MA, Bueso-Ramos C, Ferdin J, Lopez-Berestein G, Sood AK, Calin GA. MicroRNAs in body fluids--the mix of hormones and biomarkers. Nature reviews Clinical oncology. 2011;8(8):467-477. doi:10.1038/nrclinonc.2011.76 141. Lau NC, Lim LP, Weinstein EG, Bartel DP. An Abundant Class of Tiny RNAs with Probable Regulatory Roles in Caenorhabditis elegans. Science. 2001;294(5543):858. doi:10.1126/science.1065062 142. Schrauder MG, Strick R, Schulz-Wendtland R, et al. Circulating micro-RNAs as potential blood- based markers for early stage breast cancer detection. PloS one. 2012;7(1):e29770-e29770. doi:10.1371/journal.pone.0029770 143. Fortunato O, Borzi C, Milione M, et al. Circulating mir-320a promotes immunosuppressive macrophages M2 phenotype associated with lung cancer risk. International journal of cancer. 2019;144(11):2746-2761. doi:10.1002/ijc.31988
158 | Lynn
144. Lee RC, Ambros V. An Extensive Class of Small RNAs in Caenorhabditis elegans. Science. 2001;294(5543):862. doi:10.1126/science.1065329 145. He Y, Liu H, Wang S, Chen Y. In Silico Detection and Characterization of microRNAs and Their Target Genes in microRNA Microarray Datasets from Patients with Systemic Sclerosis–Interstitial Lung Disease. DNA and Cell Biology. 2019/09/01 2019;38(9):933-944. doi:10.1089/dna.2019.4780 146. Bosia C, Pagnani A, Zecchina R. Modelling Competing Endogenous RNA Networks. PLOS ONE. 06/26 2013;8(6):e66609. 147. Lee RC, Feinbaum RL, Ambros V. The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell. 1993/12/03/ 1993;75(5):843-854. doi:https://doi.org/10.1016/0092-8674(93)90529-Y 148. Ji C, Guo X. The clinical potential of circulating microRNAs in obesity. Nature Reviews Endocrinology. 2019/12/01 2019;15(12):731-743. doi:10.1038/s41574-019-0260-0 149. Treiber T, Treiber N, Meister G. Regulation of microRNA biogenesis and its crosstalk with other cellular pathways. Nature Reviews Molecular Cell Biology. 2019/01/01 2019;20(1):5-20. doi:10.1038/s41580-018-0059-1 150. Moore LD, Le T, Fan G. DNA methylation and its basic function. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2013;38(1):23-38. doi:10.1038/npp.2012.112 151. Elangovan V, Camp S, Kelly G, et al. Endotoxin- and mechanical stress–induced epigenetic changes in the regulation of the nicotinamide phosphoribosyltransferase promoter. Pulmonary Circulation. 2016;6(4):539-544. doi:10.1086/688761 152. Szilágyi KL, Liu C, Zhang X, et al. Epigenetic contribution of the myosin light chain kinase gene to the risk for acute respiratory distress syndrome. Translational Research. 180:12-21. doi:10.1016/j.trsl.2016.07.020 153. Adyshev DM, Elangovan VR, Moldobaeva N, Mapes B, Sun X, Garcia JGN. Mechanical Stress Induces Pre–B-cell Colony-Enhancing Factor/NAMPT Expression via Epigenetic Regulation by miR-374a and miR-568 in Human Lung Endothelium. American Journal of Respiratory Cell and Molecular Biology. 2014/02/01 2013;50(2):409-418. doi:10.1165/rcmb.2013-0292OC 154. Sun X, Elangovan V, Mapes B, et al. The NAMPT Promoter Is Regulated by Mechanical Stress, Signal Transducer and Activator of Transcription 5, and Acute Respiratory Distress Syndrome–Associated Genetic Variants. American Journal of Respiratory Cell and Molecular Biology. 2014/11/01 2014;51(5):660-667. doi:10.1165/rcmb.2014-0117OC 155. Barkan D, Kleinman H, Simmons JL, et al. Inhibition of Metastatic Outgrowth From Single Dormant Tumor Cells by Targeting the Cytoskeleton. Cancer research. 08/01 2008;68(15):6241-6250. 156. Zhang S, Wu Z, Xie J, Yang Y, Wang L, Qiu H. DNA methylation exploration for ARDS: a multi-omics and multi-microarray interrelated analysis. Journal of translational medicine. 2019;17(1):345-345. doi:10.1186/s12967-019-2090-1 157. Hernández-Beeftink T, Guillen-Guio B, Villar J, Flores C. Genomics and the Acute Respiratory Distress Syndrome: Current and Future Directions. International journal of molecular sciences. 2019;20(16):4004. doi:10.3390/ijms20164004 158. Rogers AJ, Meyer NJ. Precision Medicine in Critical Illness: Sepsis and Acute Respiratory Distress Syndrome. Precision in Pulmonary, Critical Care, and Sleep Medicine: A Clinical and Research Guide. 2019:267-288. doi:10.1007/978-3-030-31507-8_18 159. Meyer NJ, Calfee CS. Novel translational approaches to the search for precision therapies for acute respiratory distress syndrome. The Lancet Respiratory medicine. 2017;5(6):512-523. doi:10.1016/S2213- 2600(17)30187-X
159 | Lynn
160. Garcia JGN, Verin AD, Schaphorst KL. Regulation of Thrombin-Mediated Endothelial Cell Contraction and Permeability. Semin Thromb Hemost. 1996;22(04):309-315. 161. Gallagher PJ, Herring BP, Griffin SA, Stull JT. Molecular characterization of a mammalian smooth muscle myosin light chain kinase. The Journal of biological chemistry. 1991;266(35):23936-23944. 162. Acosta-Herrera M, Pino-Yanes M, Ma S-F, et al. Fine mapping of the myosin light chain kinase gene replicates the association with asthma in Spanish descent populations. The Journal of allergy and clinical immunology. 05/27 2015;136(4):1116-1118.e9. doi:10.1016/j.jaci.2015.04.025 163. Zhou T, Wang T, Garcia J. Genes Influenced by the Non-Muscle Isoform of Myosin Light Chain Kinase Impact Human Cancer Prognosis. PLoS ONE. April 8 2014;9(4):e94325. 164. Wang T, Zhou T, Saadat L, Garcia J. A MYLK variant regulates asthmatic inflammation via alterations in mRNA secondary structure. European Journal of Human Genetics. 2015;23(6):874-876. doi:10.1038/ejhg.2014.201 165. Wang T, Moreno-Vinasco L, Ma S-F, et al. Nonmuscle Myosin Light Chain Kinase Regulates Murine Asthmatic Inflammation. American Journal of Respiratory Cell and Molecular Biology. 2014/06/01 2014;50(6):1129-1135. doi:10.1165/rcmb.2013-0434OC 166. Zhou T, Wang T, Garcia J. A nonmuscle myosin light chain kinase–dependent gene signature in peripheral blood mononuclear cells is linked to human asthma severity and exacerbation status. Pulmonary Circulation. 2015;5(2):335-338. doi:10.1086/680357 167. Gao L, Grant AV, Rafaels N, et al. Polymorphisms in the myosin light chain kinase gene that confer risk of severe sepsis are associated with a lower risk of asthma. Journal of Allergy and Clinical Immunology. 2007/05/01/ 2007;119(5):1111-1118. doi:https://doi.org/10.1016/j.jaci.2007.03.019 168. Shen K, Ramirez B, Mapes B, et al. Structure–Function Analysis of the Non-Muscle Myosin Light Chain Kinase (nmMLCK) Isoform by NMR Spectroscopy and Molecular Modeling: Influence of MYLK Variants. PLOS ONE. 2015;10(6):e0130515. doi:10.1371/journal.pone.0130515 169. Brand-Arpon Vr, Rouquier S, Massa H, et al. A Genomic Region Encompassing a Cluster of Olfactory Receptor Genes and a Myosin Light Chain Kinase (MYLK) Gene Is Duplicated on Human Chromosome Regions 3q13–q21 and 3p13. Genomics. 1999;56(1):98-110. 170. Gao L, Grant A, Halder I, et al. Novel Polymorphisms in the Myosin Light Chain Kinase Gene Confer Risk for Acute Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2006/04/01 2006;34(4):487-495. doi:10.1165/rcmb.2005-0404OC 171. Lazar V, Garcia JGN. A Single Human Myosin Light Chain Kinase Gene (MLCK; MYLK)Transcribes Multiple Nonmuscle Isoforms. Genomics. 1999;57(2):256-267. 172. Han YJ, Ma S-F, Wade MS, Flores C, Garcia JGN. An intronic MYLK variant associated with inflammatory lung disease regulates promoter activity of the smooth muscle myosin light chain kinase isoform. Journal of Molecular Medicine. 2012/03/01 2012;90(3):299-308. doi:10.1007/s00109-011-0820- 9 173. Mascarenhas J, Tchourbanov A, Fan H, Danilov S, Wang T, Garcia J. Mechanical Stress and Single Nucleotide Variants Regulate Alternative Splicing of the MYLK Gene. American Journal of Respiratory Cell and Molecular Biology. 2017/01/01 2016;56(1):29-37. doi:10.1165/rcmb.2016-0053OC 174. Wang T, Brown M, Kelly G, et al. Myosin light chain kinase (MYLK) coding polymorphisms modulate human lung endothelial cell barrier responses via altered tyrosine phosphorylation, spatial localization, and lamellipodial protrusions. Pulmonary Circulation. 2018;8(2):2045894018764171. doi:10.1177/2045894018764171 175. Flores C, Ma S, Maresso K, Ober C, Garcia J. A variant of the myosin light chain kinase gene is associated with severe asthma in African Americans. Genet Epidemiol. May 2007;41:4.
160 | Lynn
176. Christie JD, Ma SF, Aplenc R, et al. Variation in the myosin light chain kinase gene is associated with development of acute lung injury after major trauma. Crit Care Med. Oct 2008;36(10):2794-800. 177. Lynn H, Sun X, Ayshiev D, et al. Single nucleotide polymorphisms in the MYLKP1 pseudogene are associated with increased colon cancer risk in African Americans. PLOS ONE. 2018;13(8):e0200916. doi:10.1371/journal.pone.0200916 178. Poliseno L, Salmena L, Zhang J, Carver B, Haveman WJ, Pandolfi PPP. A coding-independent function of gene and pseudogene mRNAs regulates tumour biology. Nature. June 24 2010;465:6. doi:doi:10.1038/nature09144 179. Pink RC, Wicks K, Caley DP, Punch EK, Jacobs L, Francisco Carter DR. Pseudogenes: Pseudo- functional or key regulators in health and disease? RNA. 05/ 2011;17(5):792-798. 180. Han Y, Ma S, Yourek G, Park Y, Garcia J. A transcribed pseudogene of MYLK promotes cell proliferation. FASEB J. Mar 25 2011;25:7. 181. Wang T, Zhang X, Bheda P, Revollo J, Imai S-I, Wolberger C. Structure of Nampt/PBEF/visfatin, a mammalian NAD+ biosynthetic enzyme. Nature Structural &Amp; Molecular Biology. 06/18/online 2006;13:661. doi:10.1038/nsmb1114 https://www.nature.com/articles/nsmb1114#supplementary- information 182. Hong S-B, Huang Y, Moreno-Vinasco L, et al. Essential Role of Pre–B-Cell Colony Enhancing Factor in Ventilator-induced Lung Injury. American Journal of Respiratory and Critical Care Medicine. 2008/09/15 2008;178(6):605-617. doi:10.1164/rccm.200712-1822OC 183. Camp SM, Ceco E, Evenoski CL, et al. Unique Toll-Like Receptor 4 Activation by NAMPT/PBEF Induces NFκB Signaling and Inflammatory Lung Injury. Scientific Reports. 2015;5:13135. doi:10.1038/srep13135 184. Chen Y, Liang Y, Hu T, et al. Endogenous Nampt upregulation is associated with diabetic nephropathy inflammatory-fibrosis through the NF-κB p65 and Sirt1 pathway; NMN alleviates diabetic nephropathy inflammatory-fibrosis by inhibiting endogenous Nampt. Experimental and Therapeutic Medicine. 2017;14(5):4181-4193. doi:10.3892/etm.2017.5098 185. Sun X, Sun B, Babicheva A, et al. Extracellular NAMPT involvement in development of pulmonary hypertension: Transcriptional regulation by HIF2 a and NAMPT SNPs. 2019 186. Imai S-I, Yoshino J. The importance of NAMPT/NAD/SIRT1 in the systemic regulation of metabolism and aging. Diabetes, obesity & metabolism. 2013;15(0 3):10.1111/dom.12171. doi:10.1111/dom.12171 187. Bajwa E, Yu C-L, Gong M, Thompson B, Christiani D. Pre-B-cell colony-enhancing factor gene polymorphisms and risk of acute respiratory distress syndrome*. Critical Care Medicine. 2007;35(5) 188. Ye SQ, Zhang LQ, Adyshev D, et al. Pre-B-cell-colony-enhancing factor is critically involved in thrombin-induced lung endothelial cell barrier dysregulation. Microvascular Research. 2005/11/01/ 2005;70(3):142-151. doi:https://doi.org/10.1016/j.mvr.2005.08.003 189. Richens JL, Urbanowicz RA, Lunt EAM, et al. Systems biology coupled with label-free high- throughput detection as a novel approach for diagnosis of chronic obstructive pulmonary disease. Respiratory research. 2009;10:29-29. doi:10.1186/1465-9921-10-29 190. Kirschner MW. The Meaning of Systems Biology. Cell. 2005/05/20/ 2005;121(4):503-504. doi:https://doi.org/10.1016/j.cell.2005.05.005 191. Mehta NN, Heffron SP, Patel PN, et al. A human model of inflammatory cardio-metabolic dysfunction; a double blind placebo-controlled crossover trial. Journal of Translational Medicine. 2012;10:124-124. doi:10.1186/1479-5876-10-124
161 | Lynn
192. Jabaudon M, Futier E, Roszyk L, Sapin V, Pereira B, Constantin J-M. Association between intraoperative ventilator settings and plasma levels of soluble receptor for advanced glycation end- products in patients without pre-existing lung injury. Respirology. 2015/10/01 2015;20(7):1131-1138. doi:10.1111/resp.12583 193. Calfee C, Ware L, Eisner M, et al. Plasma receptor for advanced glycation end products and clinical outcomes in acute lung injury. Thorax. 2008;63(12):1083-1089. doi:10.1136/thx.2008.095588 194. Dustin Ml Fau - Rothlein R, Rothlein R Fau - Bhan AK, Bhan Ak Fau - Dinarello CA, Dinarello Ca Fau - Springer TA, Springer TA. Induction by IL 1 and interferon-gamma: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). (0022-1767 (Print)) 195. Neeper M, Schmidt AM, Brett J, et al. Cloning and expression of a cell surface receptor for advanced glycosylation end products of proteins. Journal of Biological Chemistry. 1992;267(21):14998- 15004. 196. Soubrier F, Chung WK, Machado R, et al. Genetics and Genomics of Pulmonary Arterial Hypertension. Journal of the American College of Cardiology. 2013/12/24/ 2013;62(25, Supplement):D13- D21. doi:https://doi.org/10.1016/j.jacc.2013.10.035 197. Bouros D, Alexandrakis MG, Antoniou KM, et al. The clinical significance of serum and bronchoalveolar lavage inflammatory cytokines in patients at risk for Acute Respiratory Distress Syndrome. BMC Pulmonary Medicine. 2004;4:6-6. doi:10.1186/1471-2466-4-6 198. Cartin-Ceba R, Hubmayr RD, Qin R, et al. Predictive value of plasma biomarkers for mortality and organ failure development in patients with acute respiratory distress syndrome. Journal of Critical Care. 2015/02/01/ 2015;30(1):219.e1-219.e7. doi:https://doi.org/10.1016/j.jcrc.2014.09.001 199. Doyle IR, Hermans C, Bernard A, Nicholas TE, Bersten AD. Clearance of Clara Cell Secretory Protein 16 (CC16) and Surfactant Proteins A and B from Blood in Acute Respiratory Failure. American Journal of Respiratory and Critical Care Medicine. 1998/11/01 1998;158(5):1528-1535. doi:10.1164/ajrccm.158.5.9712097 200. Blondonnet R, Constantin J-M, Sapin V, Jabaudon M. A Pathophysiologic Approach to Biomarkers in Acute Respiratory Distress Syndrome. Disease Markers. 2016;2016:3501373. doi:10.1155/2016/3501373 201. Cohen MJ, Brohi K Fau - Calfee CS, Calfee Cs Fau - Rahn P, et al. Early release of high mobility group box nuclear protein 1 after severe trauma in humans: role of injury severity and tissue hypoperfusion. (1466-609X (Electronic)) 202. Gao L, Tsai YJ, Grigoryev DN, Barnes KC. Host defense genes in asthma and sepsis and the role of the environment. Curr Opin Allergy Clin Immunol. Dec 2007;7(6):459-67. doi:10.1097/ACI.0b013e3282f1fb9a 203. Nakamura T, Sato E, Fujiwara N, Kawagoe Y, Maeda S, Yamagishi S-i. Increased levels of soluble receptor for advanced glycation end products (sRAGE) and high mobility group box 1 (HMGB1) are associated with death in patients with acute respiratory distress syndrome. Clinical Biochemistry. 2011/06/01/ 2011;44(8):601-604. doi:https://doi.org/10.1016/j.clinbiochem.2010.12.014 204. Butt Y, Kurdowska A, Allen TC. Acute Lung Injury: A Clinical and Molecular Review. Archives of Pathology & Laboratory Medicine. 2016/04/01 2016;140(4):345-350. doi:10.5858/arpa.2015-0519-RA 205. Devaney J, Contreras M, Laffey JG. Clinical Review: Gene-based therapies for ALI/ARDS: where are we now? Critical Care. 06/20 2011;15(3):224-224. doi:10.1186/cc10216 206. Laffey J, Kavanagh B. Fifty Years of Research in ARDS.Insight into Acute Respiratory Distress Syndrome. From Models to Patients. American Journal of Respiratory and Critical Care Medicine. 2017/07/01 2017;196(1):18-28. doi:10.1164/rccm.201612-2415CI
162 | Lynn
207. Bime C, Zhou T, Wang T, Slepian M, Garcia J, Hecker L. Reactive oxygen species–associated molecular signature predicts survival in patients with sepsis. Pulmonary Circulation. 2016;6(2):196-201. doi:10.1086/685547 208. Natarajan V, Dudek SM, Jacobson JR, et al. Sphingosine-1–Phosphate, FTY720, and Sphingosine- 1–Phosphate Receptors in the Pathobiology of Acute Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2013/07/01 2013;49(1):6-17. doi:10.1165/rcmb.2012-0411TR 209. Pendyala S, Moitra J, Kalari S, et al. NRF2 Regulates Hyperoxia-induced NOX4 expression in Human Lung Endothelium: Identification of functional antioxidant response elements on NOX4 promoter. Free radical biology & medicine. 03/26 2011;50(12):1749-1759. doi:10.1016/j.freeradbiomed.2011.03.022 210. Zhang LQ, Adyshev DM, Singleton P, et al. Interactions between PBEF and oxidative stress proteins-a potential new mechanism underlying PBEF in the pathogenesis of acute lung injury. FEBS letters. 05/16 2008;582(13):1802-1808. doi:10.1016/j.febslet.2008.04.061 211. Acosta-Herrera M, Pino-Yanes M, Blanco J, et al. Common variants of NFE2L2 gene predisposes to acute respiratory distress syndrome in patients with severe sepsis. Critical Care. 2015;19(1):256. doi:10.1186/s13054-015-0981-y 212. Bernard K, Hecker L, Luckhardt TR, Cheng G, Thannickal VJ. NADPH Oxidases in Lung Health and Disease. Antioxidants & Redox Signaling. 2014;20(17):2838-2853. doi:10.1089/ars.2013.5608 213. Bhargava M, Becker T, Viken K, et al. Proteomic Profiles in Acute Respiratory Distress Syndrome Differentiates Survivors from Non-Survivors. PLOS ONE. 2014;9(10):e109713. doi:10.1371/journal.pone.0109713 214. Bhargava M, Viken K, Wang Q, et al. Bronchoalveolar Lavage Fluid Protein Expression in Acute Respiratory Distress Syndrome Provides Insights into Pathways Activated in Subjects with Different Outcomes. Scientific Reports. 2017/08/07 2017;7(1):7464. doi:10.1038/s41598-017-07791-8 215. Chow C-W, Herrera Abreu MT, Suzuki T, Downey GP. Oxidative Stress and Acute Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2003/10/01 2003;29(4):427-431. doi:10.1165/rcmb.F278 216. Flores C, Pérez-Méndez L, Maca-Meyer N, et al. A common haplotype of the LBP gene predisposes to severe sepsis*. Critical Care Medicine. 2009;37(10):2759-2766. doi:10.1097/CCM.0b013e3181a57b90 217. Imai Y, Kuba K, Neely G, et al. Identification of Oxidative Stress and Toll-like Receptor 4 Signaling as a Key Pathway of Acute Lung Injury. Cell. 2008/04/18/ 2008;133(2):235-249. doi:https://doi.org/10.1016/j.cell.2008.02.043 218. Ryu H, Lee J, Zaman K, et al. Sp1 and Sp3 Are Oxidative Stress-Inducible, Antideath Transcription Factors in Cortical Neurons. The Journal of Neuroscience. 2003;23(9):3597. 219. Tejera P, Meyer N, Chen F, et al. Distinct and replicable genetic risk factors for acute respiratory distress syndrome of pulmonary or extrapulmonary origin. Journal of medical genetics. 10/09 2012;49(11):671-680. doi:10.1136/jmedgenet-2012-100972 220. Varley I, Hughes DC, Greeves JP, et al. RANK/RANKL/OPG pathway: Genetic associations with stress fracture period prevalence in elite athletes. Bone. 2015/02/01/ 2015;71:131-136. doi:https://doi.org/10.1016/j.bone.2014.10.004 221. Wolfson R, Mapes B, Garcia J. Excessive Mechanical Stress Increases HMGB1 Expression in Human Lung Microvascular Endothelial Cells Via STAT3. Microvascular research. 12/24 2014;92:50-55. doi:10.1016/j.mvr.2013.12.005 222. Lee K, Huh J, Lim C-M, Koh Y, Hong S-B. Clinical role of serum pre-B cell colony-enhancing factor in ventilated patients with sepsis and acute respiratory distress syndrome. Scandinavian Journal of Infectious Diseases. 2013/10/01 2013;45(10):760-765. doi:10.3109/00365548.2013.797600
163 | Lynn
223. Ashbaugh D, Boyd Bigelow D, Petty T, Levine B. ACUTE RESPIRATORY DISTRESS IN ADULTS. The Lancet. 1967/08/12/ 1967;290(7511):319-323. doi:https://doi.org/10.1016/S0140-6736(67)90168-7 224. Suter PM, Fairley HB, Isenberg MD. Optimum End-Expiratory Airway Pressure in Patients with Acute Pulmonary Failure. The New England journal of medicine. 1975;292(6):284-289. doi:10.1056/NEJM197502062920604 225. Brower R, Matthay M, Morris A, Schoenfeld D, Thompson B, Wheeler A. Ventilation with Lower Tidal Volumes as Compared with Traditional Tidal Volumes for Acute Lung Injury and the Acute Respiratory Distress Syndrome. New England Journal of Medicine. 2000/05/04 2000;342(18):1301-1308. doi:10.1056/NEJM200005043421801 226. Armstrong L, Thickett D, Christie S, Kendall H, Millar A. Increased Expression of Functionally Active Membrane-Associated Tumor Necrosis Factor in Acute Respiratory Distress Syndrome. American Journal of Respiratory Cell and Molecular Biology. 2000/01/01 2000;22(1):68-74. doi:10.1165/ajrcmb.22.1.3728 227. Azevedo Z, Moore D, Lima F, et al. Tumor necrosis factor (TNF) and lymphotoxin-alpha (LTA) single nucleotide polymorphisms: Importance in ARDS in septic pediatric critically ill patients. Human Immunology. 2012/06/01/ 2012;73(6):661-667. doi:https://doi.org/10.1016/j.humimm.2012.03.007 228. Calfee C, Gallagher D, Abbott J, Thompson B, Matthay M, the NAN. Plasma angiopoietin-2 in clinical acute lung injury: Prognostic and pathogenetic significance*. Critical care medicine. 2012;40(6):1731-1737. doi:10.1097/CCM.0b013e3182451c87 229. Carnesecchi S, Dunand-Sauthier I, Zanetti F, et al. NOX1 is responsible for cell death through STAT3 activation in hyperoxia and is associated with the pathogenesis of Acute Respiratory Distress Syndrome. International Journal of Clinical and Experimental Pathology. 2014;7(2):537-551. 230. D'Alessio FR, Craig JM, Singer BD, et al. Enhanced resolution of experimental ARDS through IL-4- mediated lung macrophage reprogramming. American journal of physiology Lung cellular and molecular physiology. 2016;310(8):L733-L746. doi:10.1152/ajplung.00419.2015 231. Dötsch A, Eisele L, Rabeling M, et al. Hypoxia Inducible Factor-2 Alpha and Prolinhydroxylase 2 Polymorphisms in Patients with Acute Respiratory Distress Syndrome (ARDS). International Journal of Molecular Sciences. 2017;18(6):1266. doi:10.3390/ijms18061266 232. Hinz J, Büttner B, Kriesel F, et al. The FER rs4957796 TT genotype is associated with unfavorable 90-day survival in Caucasian patients with severe ARDS due to pneumonia. Scientific Reports. 2017;7:9887. doi:10.1038/s41598-017-08540-7 233. Hästbacka J, Linko R, Tervahartiala T, et al. Serum MMP-8 and TIMP-1 in Critically Ill Patients with Acute Respiratory Failure: TIMP-1 Is Associated with Increased 90-Day Mortality. Anesthesia & Analgesia. 2014;118(4):790-798. doi:10.1213/ane.0000000000000120 234. Liu H, Zhang D, Zhao B, Zhao J. Superoxide Anion, the Main Species of ROS in the Development of ARDS Induced by Oleic Acid. Free Radical Research. 2004/12/01 2004;38(12):1281-1287. doi:10.1080/10715760400006940 235. Morrell E, O’Mahony D, Glavan B, et al. Genetic Variation in MAP3K1 Associates with Ventilator- Free Days in Acute Respiratory Distress Syndrome. American Journal of Respiratory Cell and Molecular Biology. 2018/01/01 2017;58(1):117-125. doi:10.1165/rcmb.2017-0030OC 236. Nick JA, Caceres SM, Kret JE, et al. Extremes of Interferon-Stimulated Gene Expression Associate with Worse Outcomes in the Acute Respiratory Distress Syndrome. PLOS ONE. 2016;11(9):e0162490. doi:10.1371/journal.pone.0162490 237. Rojo de la Vega M, Dodson M, Gross C, et al. Role of Nrf2 and Autophagy in Acute Lung Injury. Current pharmacology reports. 02/06 2016;2(2):91-101. doi:10.1007/s40495-016-0053-2
164 | Lynn
238. Tejera P, Wang Z, Zhai R, et al. Genetic Polymorphisms of Peptidase Inhibitor 3 (Elafin) Are Associated with Acute Respiratory Distress Syndrome. American Journal of Respiratory Cell and Molecular Biology. 2009/12/01 2009;41(6):696-704. doi:10.1165/rcmb.2008-0410OC 239. Wang M, Yan J, He X, Zhong Q, Zhan C, Li S. Candidate genes and pathogenesis investigation for sepsis-related acute respiratory distress syndrome based on gene expression profile. Biological Research. 2016/04/18 2016;49(1):25. doi:10.1186/s40659-016-0085-4 240. Wei Y, Tejera P, Wang Z, et al. A Missense Genetic Variant in LRRC16A/CARMIL1 Improves Acute Respiratory Distress Syndrome Survival by Attenuating Platelet Count Decline. American Journal of Respiratory and Critical Care Medicine. 2017/05/15 2016;195(10):1353-1361. doi:10.1164/rccm.201605- 0946OC 241. Wei Y, Wang Z, Su L, et al. Platelet Count Mediates the Contribution of a Genetic Variant in LRRC 16A to ARDS Risk. Chest. 2015/03/01/ 2015;147(3):607-617. doi:https://doi.org/10.1378/chest.14-1246 242. Zhao J, Yu H, Liu Y, et al. Protective effect of suppressing STAT3 activity in LPS-induced acute lung injury. American Journal of Physiology - Lung Cellular and Molecular Physiology. 2016;311(5):L868-L880. doi:10.1152/ajplung.00281.2016 243. Rivas MA, Beaudoin M, Gardet A, et al. Deep resequencing of GWAS loci identifies independent rare variants associated with inflammatory bowel disease. Article. Nature Genetics. 10/09/online 2011;43:1066. doi:10.1038/ng.952 https://www.nature.com/articles/ng.952#supplementary- information 244. Becker KG, Hosack DA, Dennis G, Jr., et al. PubMatrix: a tool for multiplex literature mining. BMC bioinformatics. 2003;4:61-61. doi:10.1186/1471-2105-4-61 245. Braun R. Systems Analysis of High–Throughput Data. Advances in experimental medicine and biology. 2014;844:153-187. doi:10.1007/978-1-4939-2095-2_8 246. Belopolskaya O, Smelaya T, Moroz V, Golubev A, Salnikova L. Clinical associations of host genetic variations in the genes of cytokines in critically ill patients. Clinical and Experimental Immunology. 2015;180(3):531-541. doi:10.1111/cei.12592 247. Bajwa EK, Yu C-L, Gong MN, Thompson BT, Christiani DC. Pre-B-cell colony-enhancing factor gene polymorphisms and risk of acute respiratory distress syndrome*. Critical Care Medicine. 2007;35(5) 248. Pino-Yanes M, Ma S-F, Sun X, et al. Interleukin-1 Receptor–Associated Kinase 3 Gene Associates with Susceptibility to Acute Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2011/10/01 2011;45(4):740-745. doi:10.1165/rcmb.2010-0292OC 249. Chang DW, Hayashi S, Gharib SA, et al. Proteomic and computational analysis of bronchoalveolar proteins during the course of the acute respiratory distress syndrome. Am J Respir Crit Care Med. Oct 1 2008;178(7):701-9. doi:10.1164/rccm.200712-1895OC 250. Howrylak JA, Dolinay T, Lucht L, et al. Discovery of the gene signature for acute lung injury in patients with sepsis. Physiological genomics. 2009;37(2):133-139. doi:10.1152/physiolgenomics.90275.2008 251. Kamp R, Sun X, Garcia JGN. Making Genomics Functional. Proceedings of the American Thoracic Society. 2008/04/15 2008;5(3):348-353. doi:10.1513/pats.200709-152DR 252. Headley AS, Meduri GU, Tolley E. Infections and the Inflammatory Response in Acute Respiratory Distress Syndrome. Chest. 1997/05/01/ 1997;111(5):1306-1321. doi:https://doi.org/10.1378/chest.111.5.1306 253. Bauer TT, Montón C, Torres A, et al. Comparison of systemic cytokine levels in patients with acute respiratory distress syndrome, severe pneumonia, and controls. 10.1136/thorax.55.1.46. Thorax. 2000;55(1):46.
165 | Lynn
254. Lee Y-L, Chen W, Chen L-Y, et al. Systemic and bronchoalveolar cytokines as predictors of in- hospital mortality in severe community-acquired pneumonia. Journal of Critical Care. 2010/03/01/ 2010;25(1):176.e7-176.e13. doi:https://doi.org/10.1016/j.jcrc.2009.05.002 255. Oita RC, Bime C, Casanova N, Camp S, Wang T, Garcia GJ. Blood eNAMPT is an ARDS and VILI biomarker differentially expressed and secreted by lung epithelium and endothelium in response to mechanical stress. Am J Respir Cell Mol Biol. 2020; 256. Jin L, Zuo X-Y, Su W-Y, et al. Pathway-based Analysis Tools for Complex Diseases: A Review. Genomics, Proteomics & Bioinformatics. 2014/10/01/ 2014;12(5):210-220. doi:https://doi.org/10.1016/j.gpb.2014.10.002 257. Herwig R, Hardt C, Lienhard M, Kamburov A. Analyzing and interpreting genome data at the network level with ConsensusPathDB. Nature Protocols. 09/08/online 2016;11:1889. doi:10.1038/nprot.2016.117 https://www.nature.com/articles/nprot.2016.117#supplementary- information 258. Kamburov A, Pentchev K, Galicka H, Wierling C, Lehrach H, Herwig R. ConsensusPathDB: toward a more complete picture of cell biology. Nucleic Acids Research. 2011;39(suppl_1):D712-D717. doi:10.1093/nar/gkq1156 259. Kamburov A, Stelzl U, Lehrach H, Herwig R. The ConsensusPathDB interaction database: 2013 update. Nucleic Acids Research. 2013;41(D1):D793-D800. doi:10.1093/nar/gks1055 260. Di A, Mehta D, Malik AB. ROS-activated calcium signaling mechanisms regulating endothelial barrier function. Cell calcium. 02/17 2016;60(3):163-171. doi:10.1016/j.ceca.2016.02.002 261. Griffith B, Pendyala S, Hecker L, Lee PJ, Natarajan V, Thannickal VJ. NOX Enzymes and Pulmonary Disease. Antioxidants & Redox Signaling. 2009;11(10):2505-2516. doi:10.1089/ars.2009.2599 262. Hecker L, Vittal R, Jones T, et al. NADPH Oxidase-4 Mediates Myofibroblast Activation and Fibrogenic Responses to Lung Injury. Nature medicine. 08/23 2009;15(9):1077-1081. doi:10.1038/nm.2005 263. Jabaudon M, Futier E, Roszyk L, et al. Soluble form of the receptor for advanced glycation end products is a marker of acute lung injury but not of severe sepsis in critically ill patients*. Critical Care Medicine. 2011;39(3):480-488. doi:10.1097/CCM.0b013e318206b3ca 264. Rautanen A, Mills TC, Gordon AC, et al. Genome-wide association study of survival from sepsis due to pneumonia: an observational cohort study. The Lancet Respiratory Medicine. 01/ 2015;3(1):53-60. doi:10.1016/S2213-2600(14)70290-5 265. Park J-B, Nagar H, Choi S, et al. IDH2 deficiency impairs mitochondrial function in endothelial cells and endothelium-dependent vasomotor function. Free Radical Biology and Medicine. 2016/05/01/ 2016;94:36-46. doi:https://doi.org/10.1016/j.freeradbiomed.2016.02.017 266. Victor MV, Juan VE, Antonio Hernandez-Mijares and Milagros R. Oxidative Stress and Mitochondrial Dysfunction in Sepsis: A Potential Therapy with Mitochondria-Targeted Antioxidants. Infectious Disorders - Drug Targets. 2009;9(4):376-389. doi:http://dx.doi.org/10.2174/187152609788922519 267. Hernandez-Pacheco N, Guillen-Guio B, Acosta-Herrera M, et al. A vascular endothelial growth factor receptor gene variant is associated with susceptibility to acute respiratory distress syndrome. Intensive Care Med Exp. Jul 9 2018;6(1):16. doi:10.1186/s40635-018-0181-6 268. Li G, Malinchoc M, Cartin-Ceba R, et al. Eight-Year Trend of Acute Respiratory Distress Syndrome. American Journal of Respiratory and Critical Care Medicine. 2011;183(1):59-66. doi:10.1164/rccm.201003-0436OC 269. Rubenfeld GD, Caldwell E, Peabody E, et al. Incidence and Outcomes of Acute Lung Injury. New England Journal of Medicine. 2005/10/20 2005;353(16):1685-1693. doi:10.1056/NEJMoa050333 166 | Lynn
270. Benveniste EN, Liu Y, McFarland BC, Qin H. Involvement of the Janus Kinase/Signal Transducer and Activator of Transcription Signaling Pathway in Multiple Sclerosis and the Animal Model of Experimental Autoimmune Encephalomyelitis. Journal of Interferon & Cytokine Research. 2014;34(8):577- 588. doi:10.1089/jir.2014.0012 271. O’Shea JJ, Plenge R. JAKs and STATs in Immunoregulation and Immune-Mediated Disease. Immunity. 2012;36(4):542-550. doi:10.1016/j.immuni.2012.03.014 272. Villarino AV, Kanno Y, O’Shea JJ. Mechanisms of Jak/STAT signaling in immunity and disease. Journal of immunology (Baltimore, Md : 1950). 2015;194(1):21-27. doi:10.4049/jimmunol.1401867 273. Wang H-W, Yang W, Lu J-Y, et al. N-acetylcysteine Administration is Associated with Reduced Activation of NF-kB and Preserves Lung Dendritic Cells Function in a Zymosan-Induced Generalized Inflammation Model. Journal of Clinical Immunology. 2013/04/01 2013;33(3):649-660. doi:10.1007/s10875-012-9852-3 274. Cho H-Y, Kleeberger SR. Nrf2 protects against airway disorders. Toxicology and Applied Pharmacology. 2010/04/01/ 2010;244(1):43-56. doi:https://doi.org/10.1016/j.taap.2009.07.024 275. Chen J, Zhang Z, Cai L. Diabetic cardiomyopathy and its prevention by nrf2: current status. Diabetes & metabolism journal. 2014;38(5):337-345. doi:10.4093/dmj.2014.38.5.337 276. Liu P, Rojo de la Vega M, Sammani S, et al. MYLK transcriptional repression by NRF2-ARE and RPA1 enhances vascular integrity. Proceedings of the National Academy of Science (PNAS); 2018. 277. Tao S, Rojo de la Vega M, Quijada H, et al. Bixin protects mice against ventilation-induced lung injury in an NRF2-dependent manner. Article. Scientific Reports. 01/05/online 2016;6:18760. doi:10.1038/srep18760 https://www.nature.com/articles/srep18760#supplementary-information 278. Grommes J, Soehnlein O. Contribution of Neutrophils to Acute Lung Injury. Molecular Medicine. Mar-Apr 2011;17(3-4):293-307. doi:10.2119/molmed.2010.00138 279. Albaiceta GM, Gutiérrez-Fernández A, Parra D, et al. Lack of matrix metalloproteinase-9 worsens ventilator-induced lung injury. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2008/03/01 2008;294(3):L535-L543. doi:10.1152/ajplung.00334.2007 280. Albaiceta GM, Gutierrez-Fernández A, García-Prieto E, et al. Absence or Inhibition of Matrix Metalloproteinase–8 Decreases Ventilator-Induced Lung Injury. American Journal of Respiratory Cell and Molecular Biology. 2010/11/01 2010;43(5):555-563. doi:10.1165/rcmb.2009-0034OC 281. Wang H, Bloom O, Zhang M, et al. HMG-1 as a Late Mediator of Endotoxin Lethality in Mice. 10.1126/science.285.5425.248. Science. 1999;285(5425):248. 282. Shirasawa M, Fujiwara N, Hirabayashi S, et al. Receptor for advanced glycation end-products is a marker of type I lung alveolar cells. Genes to Cells. 2004/02/01 2004;9(2):165-174. doi:10.1111/j.1356- 9597.2004.00712.x 283. Jabaudon M, Blondonnet R, Roszyk L, et al. Soluble Receptor for Advanced Glycation End-Products Predicts Impaired Alveolar Fluid Clearance in Acute Respiratory Distress Syndrome. American Journal of Respiratory and Critical Care Medicine. 2015/07/15 2015;192(2):191-199. doi:10.1164/rccm.201501- 0020OC 284. Jabaudon M, Blondonnet R, Roszyk L, et al. Soluble Forms and Ligands of the Receptor for Advanced Glycation End-Products in Patients with Acute Respiratory Distress Syndrome: An Observational Prospective Study. PLOS ONE. 2015;10(8):e0135857. doi:10.1371/journal.pone.0135857 285. Uchida T, Shirasawa M, Ware LB, et al. Receptor for advanced glycation end-products is a marker of type I cell injury in acute lung injury. American journal of respiratory and critical care medicine. 2006;173(9):1008-1015. doi:10.1164/rccm.200509-1477OC
167 | Lynn
286. Bernhagen J, Calandra T, Mitchell RA, et al. MIF is a pituitary-derived cytokine that potentiates lethal endotoxaemia. Nature. 10/21/online 1993;365:756. doi:10.1038/365756a0 287. Steinberg K, Hudson L, Goodman R, et al. Efficacy and Safety of Corticosteroids for Persistent Acute Respiratory Distress Syndrome. New England Journal of Medicine. 2006/04/20 2006;354(16):1671- 1684. doi:10.1056/NEJMoa051693 288. Brown S, Grissom C, Rondina M, et al. Polymorphisms in key pulmonary inflammatory pathways and the development of acute respiratory distress syndrome. Experimental lung research. 12/16 2015;41(3):155-162. doi:10.3109/01902148.2014.983281 289. Qayyum R, Snively BM, Ziv E, et al. A Meta-Analysis and Genome-Wide Association Study of Platelet Count and Mean Platelet Volume in African Americans. PLOS Genetics. 2012;8(3):e1002491. doi:10.1371/journal.pgen.1002491 290. Kataoka K, Ekuni D, Tomofuji T, et al. Visualization of Oxidative Stress Induced by Experimental Periodontitis in Keap1-Dependent Oxidative Stress Detector-Luciferase Mice. International journal of molecular sciences. 2016;17(11):1907. doi:10.3390/ijms17111907 291. Padgett LE, Anderson B, Liu C, et al. Loss of NOX-Derived Superoxide Exacerbates Diabetogenic CD4 T-Cell Effector Responses in Type 1 Diabetes. Diabetes. 2015;64(12):4171. doi:10.2337/db15-0546 292. Akey JM. Constructing genomic maps of positive selection in humans: Where do we go from here? Genome Research. 2009;19(5):711-722. doi:10.1101/gr.086652.108 293. Marshall RP, Webb S, Hill MR, Humphries SE, Laurent GJ. Genetic Polymorphisms Associated With Susceptibility and Outcome in ARDS. Chest. 2002/03/01/ 2002;121(3, Supplement):68S-69S. doi:https://doi.org/10.1378/chest.121.3_suppl.68S 294. Ahasic A, Zhao Y, Su L, Sheu C-C, Thompson B, Christiani D. Adiponectin Gene Polymorphisms and Acute Respiratory Distress Syndrome Susceptibility and Mortality. PLOS ONE. 2014;9(2):e89170. doi:10.1371/journal.pone.0089170 295. Proudfoot A, Bayliffe A, O’Kane CM, et al. Novel anti-tumour necrosis factor receptor-1 (TNFR1) domain antibody prevents pulmonary inflammation in experimental acute lung injury. 10.1136/thoraxjnl- 2017-210305. Thorax. 2018;73(8):723. 296. Gianfrancesco F, Rendina D, Di Stefano M, et al. A nonsynonymous TNFRSF11A variation increases NFκB activity and the severity of Paget's disease. Journal of Bone and Mineral Research. 2012/02/01 2011;27(2):443-452. doi:10.1002/jbmr.542 297. Sirinian C, Papanastasiou AD, Zarkadis IK, Kalofonos HP. Alternative splicing generates a truncated isoform of human TNFRSF11A (RANK) with an altered capacity to activate NF-κB. Gene. 2013/08/01/ 2013;525(1):124-129. doi:https://doi.org/10.1016/j.gene.2013.04.075 298. Wu J, Liu Z, Meng K, Zhang L. Association of Adiponectin Gene (ADIPOQ) rs2241766 Polymorphism with Obesity in Adults: A Meta-Analysis. PLOS ONE. 2014;9(4):e95270. doi:10.1371/journal.pone.0095270 299. Lu J-f, Zhou Y, Huang G-h, Jiang H-x, Hu B-l, Qin S-y. Association of ADIPOQ polymorphisms with obesity risk: A meta-analysis. Human Immunology. 2014/10/01/ 2014;75(10):1062-1068. doi:https://doi.org/10.1016/j.humimm.2014.09.004 300. Tsai M-K, Wang H-MD, Shiang J-C, et al. Sequence Variants of ADIPOQ and Association with Type 2 Diabetes Mellitus in Taiwan Chinese Han Population. The Scientific World Journal. 2014;2014:650393. doi:10.1155/2014/650393 301. Chung H-F, Long KZ, Hsu C-C, et al. Adiponectin gene (ADIPOQ) polymorphisms correlate with the progression of nephropathy in Taiwanese male patients with type 2 diabetes. Diabetes Research and Clinical Practice. 2014/08/01/ 2014;105(2):261-270. doi:https://doi.org/10.1016/j.diabres.2014.04.015 302. Ghosh CC, David S, Zhang R, et al. Gene control of tyrosine kinase TIE2 and vascular manifestations of infections. 10.1073/pnas.1519467113. Proceedings of the National Academy of Sciences. 2016;113(9):2472.
168 | Lynn
303. Sheu CC, Zhai R, Su L, et al. Sex-specific association of Epidermal Growth Factor gene polymorphisms with ARDS. The European respiratory journal : official journal of the European Society for Clinical Respiratory Physiology. 11/14 2009;33(3):543-550. doi:10.1183/09031936.00091308 304. Jabaudon M, Berthelin P, Pranal T, et al. Receptor for advanced glycation end-products and ARDS prediction: a multicentre observational study. Scientific Reports. 2018;8:2603. doi:10.1038/s41598-018- 20994-x 305. Gong MN, Wei Z, Xu L-L, Miller DP, Thompson BT, Christiani DC. Polymorphism in the Surfactant Protein-B Gene, Gender, and the Risk of Direct Pulmonary Injury and ARDS. CHEST. 125(1):203-211. doi:10.1378/chest.125.1.203 306. Flores C, del Mar Pino-Yanes M, Villar J. A quality assessment of genetic association studies supporting susceptibility and outcome in acute lung injury. Critical Care. 2008;12(5):R130-R130. doi:10.1186/cc7098 307. Bellani G, Laffey JG, Pham T, et al. Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA. 2016;315(8):788-800. doi:10.1001/jama.2016.0291 308. Ma S-F, Xie L, Pino-Yanes M, et al. Type 2 deiodinase and host responses of sepsis and acute lung injury. American journal of respiratory cell and molecular biology. 2011;45(6):1203-1211. doi:10.1165/rcmb.2011-0179OC 309. Bellani G, Laffey J, Pham T, et al. Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA. 2016;315(8):788-800. doi:10.1001/jama.2016.0291 310. Celedón JC, Burchard EG, Schraufnagel D, et al. An American Thoracic Society/National Heart, Lung, and Blood Institute Workshop Report: Addressing Respiratory Health Equality in the United States. Annals of the American Thoracic Society. 2017;14(5):814-826. doi:10.1513/AnnalsATS.201702-167WS 311. Garcia JGN, Sznajder JI. Healthcare Disparities in Patients with Acute Respiratory Distress Syndrome. Toward Equity. American Journal of Respiratory and Critical Care Medicine. 2013/09/15 2013;188(6):631-632. doi:10.1164/rccm.201307-1394ED 312. Tobin MJ, Laghi F, Jubran A. Reply to: On Happy Hypoxia and on Sadly Ignored Acute Vascular Distress Syndrome in COVID-19 Patients. American Journal of Respiratory and Critical Care Medicine. 2020;doi:10.1164/rccm.202007-2940LE 313. Zhou CC, Yang X, Hua X, et al. Hepatic NAD(+) deficiency as a therapeutic target for non‐alcoholic fatty liver disease in ageing. British Journal of Pharmacology. 2016;173(15):2352-2368. doi:10.1111/bph.13513 314. Garten A, Schuster S, Penke M, Gorski T, de Giorgis T, Kiess W. Physiological and pathophysiological roles of NAMPT and NAD metabolism. Review Article. Nature Reviews Endocrinology. 07/28/online 2015;11:535. doi:10.1038/nrendo.2015.117 315. Pillai VB, Sundaresan NR, Kim G, et al. Nampt secreted from cardiomyocytes promotes development of cardiac hypertrophy and adverse ventricular remodeling. American Journal of Physiology - Heart and Circulatory Physiology. 2013;304(3):H415-H426. doi:10.1152/ajpheart.00468.2012 316. Shackelford RE, Mayhall K, Maxwell NM, Kandil E, Coppola D. Nicotinamide Phosphoribosyltransferase in Malignancy: A Review. Genes & Cancer. 2013/11/01 2013;4(11-12):447-456. doi:10.1177/1947601913507576 317. Hufton SE, Moerkerk PT, Brandwijk R, de Bruïne AP, Arends J-W, Hoogenboom HR. A profile of differentially expressed genes in primary colorectal cancer using suppression subtractive hybridization. FEBS Letters. 1999;463(1-2):77-82. doi:10.1016/S0014-5793(99)01578-1
169 | Lynn
318. van Beijnum JR, Moerkerk PTM, Gerbers AJ, et al. Target validation for genomics using peptide- specific phage antibodies: A study of five gene products overexpressed in colorectal cancer. International Journal of Cancer. 2002;101(2):118-127. doi:10.1002/ijc.10584 319. Wang B, Hasan MK, Alvarado E, Yuan H, Wu H, Chen WY. NAMPT overexpression in prostate cancer and its contribution to tumor cell survival and stress response. Oncogene. 2011/02/01 2011;30(8):907-921. doi:10.1038/onc.2010.468 320. Patel ST, Mistry T, Brown JEP, et al. A novel role for the adipokine visfatin/pre-B cell colony- enhancing factor 1 in prostate carcinogenesis. Peptides. 2010/01/01/ 2010;31(1):51-57. doi:https://doi.org/10.1016/j.peptides.2009.10.001 321. Tian W, Zhu Y, Wang Y, et al. Visfatin, a potential biomarker and prognostic factor for endometrial cancer. Gynecologic Oncology. 2013/06/01/ 2013;129(3):505-512. doi:https://doi.org/10.1016/j.ygyno.2013.02.022 322. Ehrich M, Nelson MR, Stanssens P, et al. Quantitative high-throughput analysis of DNA methylation patterns by base-specific cleavage and mass spectrometry. Proceedings of the National Academy of Sciences of the United States of America. 10/21 2005;102(44):15785-15790. doi:10.1073/pnas.0507816102 323. Venkateshaiah SU, Khan S, Ling W, et al. NAMPT/PBEF1 enzymatic activity is indispensable for myeloma cell growth and osteoclast activity. Experimental Hematology. 2013/06/01/ 2013;41(6):547- 557.e2. doi:https://doi.org/10.1016/j.exphem.2013.02.008 324. Reddy PS, Umesh S, Thota B, et al. PBEF1/NAmPRTase/Visfatin: A potential malignant astrocytoma/glioblastoma serum marker with prognostic value. Cancer Biology & Therapy. 2008/05/01 2008;7(5):663-668. doi:10.4161/cbt.7.5.5663 325. Shackelford R, Hirsh S, Henry K, Abdel-Mageed A, Kandil E, Coppola D. Nicotinamide Phosphoribosyltransferase and SIRT3 Expression Are Increased in Well-differentiated Thyroid Carcinomas. Anticancer Research. 2013;33(8):3047-3052. 326. Lee K, Gong M. Pre-B-Cell Colony-Enhancing Factor and Its Clinical Correlates With Acute Lung Injury and Sepsis. Chest. 2011/08/01/ 2011;140(2):382-390. doi:https://doi.org/10.1378/chest.10-3100 327. Harrington JS, Schenck EJ, Oromendia C, Choi AMK, Siempos II. Acute respiratory distress syndrome without identifiable risk factors: A secondary analysis of the ARDS network trials. Journal of Critical Care. 2018/10/01/ 2018;47:49-54. doi:https://doi.org/10.1016/j.jcrc.2018.06.002 328. Wiedemann H, Wheeler A, Bernard G, et al. Comparison of Two Fluid-Management Strategies in Acute Lung Injury. New England Journal of Medicine. 2006/06/15 2006;354(24):2564-2575. doi:10.1056/NEJMoa062200 329. Ye SQ, Simon BA, Maloney JP, et al. Pre–B-Cell Colony-enhancing Factor as a Potential Novel Biomarker in Acute Lung Injury. American Journal of Respiratory and Critical Care Medicine. 2005/02/15 2005;171(4):361-370. doi:10.1164/rccm.200404-563OC 330. Sun X, Elangovan VR, Mapes B, et al. The NAMPT Promoter Is Regulated by Mechanical Stress, Signal Transducer and Activator of Transcription 5, and Acute Respiratory Distress Syndrome–Associated Genetic Variants. American Journal of Respiratory Cell and Molecular Biology. 2014/11/01 2014;51(5):660-667. doi:10.1165/rcmb.2014-0117OC 331. Elangovan VR, Camp SM, Kelly GT, et al. Endotoxin- and mechanical stress–induced epigenetic changes in the regulation of the nicotinamide phosphoribosyltransferase promoter. Pulmonary Circulation. 2016;6(4):539-544. doi:10.1086/688761 332. Zheng SL, Sun J, Wiklund F, et al. Cumulative Association of Five Genetic Variants with Prostate Cancer. New England Journal of Medicine. 2008/02/28 2008;358(9):910-919. doi:10.1056/NEJMoa075819
170 | Lynn
333. Desikan RS, Fan CC, Wang Y, et al. Genetic assessment of age-associated Alzheimer disease risk: Development and validation of a polygenic hazard score. PLOS Medicine. 2017;14(3):e1002258. doi:10.1371/journal.pmed.1002258 334. Ware LB, Koyama T, Billheimer DD, et al. Prognostic and pathogenetic value of combining clinical and biochemical indices in patients with acute lung injury. Chest. 2010;137(2):288-296. doi:10.1378/chest.09-1484 335. Stewart RM, Park PK, Hunt JP, et al. Less Is More: Improved Outcomes in Surgical Patients with Conservative Fluid Administration and Central Venous Catheter Monitoring. Journal of the American College of Surgeons. 2009/05/01/ 2009;208(5):725-735. doi:https://doi.org/10.1016/j.jamcollsurg.2009.01.026 336. Calfee CS, Delucchi K, Parsons PE, et al. Subphenotypes in acute respiratory distress syndrome: latent class analysis of data from two randomised controlled trials. The Lancet Respiratory medicine. 2014;2(8):611-620. doi:10.1016/S2213-2600(14)70097-9 337. Ware LB. Prognostic determinants of acute respiratory distress syndrome in adults: Impact on clinical trial design. Critical Care Medicine. 2005;33(3):S217-S222. doi:10.1097/01.ccm.0000155788.39101.7e 338. Ware LB, Koyama T, Zhao Z, et al. Biomarkers of lung epithelial injury and inflammation distinguish severe sepsis patients with acute respiratory distress syndrome. Critical care (London, England). 2013;17(5):R253-R253. doi:10.1186/cc13080 339. Ware LB, Calfee CS. Biomarkers of ARDS: what’s new? Intensive Care Medicine. 2016/05/01 2016;42(5):797-799. doi:10.1007/s00134-015-3973-0 340. Calfee CS, Ware LB, Eisner MD, et al. Plasma receptor for advanced glycation end products and clinical outcomes in acute lung injury. Thorax. 2008;63(12):1083-1089. doi:10.1136/thx.2008.095588 341. Spadaro S, Park M, Turrini C, et al. Biomarkers for Acute Respiratory Distress syndrome and prospects for personalised medicine. Journal of Inflammation. 2019/01/15 2019;16(1):1. doi:10.1186/s12950-018-0202-y 342. Sarma A, Calfee C, Ware L. Biomarkers and Precision Medicine: State of the Art. Crit Care Clin. 2020;36(1):11. 343. Grigoryev DN, Ma S-F, Irizarry RA, Ye SQ, Quackenbush J, Garcia JGN. Orthologous gene- expression profiling in multi-species models: search for candidate genes. Genome Biology. 2004;5(5):R34- R34. 344. Alonso A, Krijthe BP, Aspelund T, et al. Simple risk model predicts incidence of atrial fibrillation in a racially and geographically diverse population: the CHARGE-AF consortium. Journal of the American Heart Association. 2013;2(2):e000102-e000102. doi:10.1161/JAHA.112.000102 345. Seibert T, Fan C, Wang Y, et al. Polygenic hazard score to guide screening for aggressive prostate cancer: development and validation in large scale cohorts. BMJ (Clinical research ed). 2018;360:j5757- j5757. doi:10.1136/bmj.j5757 346. Meyer NJ, Huang Y, Singleton PA, et al. GADD45a is a novel candidate gene in inflammatory lung injury via influences on Akt signaling. The FASEB Journal. 2009;23(5):1325-1337. doi:10.1096/fj.08-119073 347. Bime C, Pouladi N, Sammani S, et al. Genome Wide Association Study in African Americans with Acute Respiratory Distress Syndrome Identifies the Selectin P Ligand Gene as a Risk Factor. American Journal of Respiratory and Critical Care Medicine. 2018;doi:10.1164/rccm.201705-0961OC 348. Maghsoudloo M, Azimzadeh Jamalkandi S, Najafi A, Masoudi-Nejad A. Identification of biomarkers in common chronic lung diseases by co-expression networks and drug-target interactions analysis. Molecular Medicine. 2020/01/17 2020;26(1):9. doi:10.1186/s10020-019-0135-9
171 | Lynn
349. Kessler MD, Yerges-Armstrong L, Taub MA, et al. Challenges and disparities in the application of personalized genomic medicine to populations with African ancestry. Nature Communications. 2016;7:12521. 350. Prasse A, Binder H, Schupp JC, et al. BAL Cell Gene Expression Is Indicative of Outcome and Airway Basal Cell Involvement in Idiopathic Pulmonary Fibrosis. American Journal of Respiratory and Critical Care Medicine. 2019/03/01 2018;199(5):622-630. doi:10.1164/rccm.201712-2551OC 351. Duck A, Pigram L, Errhalt P, Ahmed D, Chaudhuri N. IPF Care: a support program for patients with idiopathic pulmonary fibrosis treated with pirfenidone in Europe. Advances in therapy. 2015;32(2):87-107. doi:10.1007/s12325-015-0183-7 352. García-Sancho C, Buendía-Roldán I, Fernández-Plata MR, et al. Familial pulmonary fibrosis is the strongest risk factor for idiopathic pulmonary fibrosis. Respiratory Medicine. 2011/12/01/ 2011;105(12):1902-1907. doi:https://doi.org/10.1016/j.rmed.2011.08.022 353. Raghu G, Collard HR, Egan JJ, et al. An Official ATS/ERS/JRS/ALAT Statement: Idiopathic Pulmonary Fibrosis: Evidence-based Guidelines for Diagnosis and Management. American Journal of Respiratory and Critical Care Medicine. 2011/03/15 2011;183(6):788-824. doi:10.1164/rccm.2009-040GL 354. Vestbo J, Hurd SS, Agustí AG, et al. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease. American Journal of Respiratory and Critical Care Medicine. 2013/02/15 2013;187(4):347-365. doi:10.1164/rccm.201204-0596PP 355. Lai Y-C, Potoka KC, Champion HC, Mora AL, Gladwin MT. Pulmonary Arterial Hypertension: The Clinical Syndrome. Circulation research. 2014;115(1):115-130. doi:10.1161/CIRCRESAHA.115.301146 356. Gallí M, Van Gool F, Rongvaux A, Andris F, Leo O. The Nicotinamide Phosphoribosyltransferase: A Molecular Link between Metabolism, Inflammation, and Cancer. Cancer Research. 2010;70(1):8. doi:10.1158/0008-5472.CAN-09-2465 357. Huertas A, Guignabert C, Barberà JA, et al. Pulmonary vascular endothelium: the orchestra conductor in respiratory diseases. European Respiratory Journal. 2018;51(4):1700745. doi:10.1183/13993003.00745-2017 358. Nonas SA, Vinasco LM, Ma SF, et al. Use of consomic rats for genomic insights into ventilator- associated lung injury. American journal of physiology Lung cellular and molecular physiology. 04/27 2007;293(2):L292-L302. doi:10.1152/ajplung.00481.2006 359. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. American Journal of Respiratory and Critical Care Medicine. 1994/03/01 1994;149(3):818-824. doi:10.1164/ajrccm.149.3.7509706 360. Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences. 2005;102(43):15545. doi:10.1073/pnas.0506580102 361. Irahara N, Nosho K, Baba Y, et al. Precision of pyrosequencing assay to measure LINE-1 methylation in colon cancer, normal colonic mucosa, and peripheral blood cells. The Journal of molecular diagnostics : JMD. 2010;12(2):177-183. doi:10.2353/jmoldx.2010.090106 362. Lisanti S, Omar WAW, Tomaszewski B, et al. Comparison of Methods for Quantification of Global DNA Methylation in Human Cells and Tissues. PLOS ONE. 2013;8(11):e79044. doi:10.1371/journal.pone.0079044 363. Baccarelli A, Wright R, Bollati V, et al. Ischemic heart disease and stroke in relation to blood DNA methylation. Epidemiology (Cambridge, Mass). 2010;21(6):819-828. doi:10.1097/EDE.0b013e3181f20457 364. Mohlin S, Hamidian A, von Stedingk K, et al. PI3K–mTORC2 but not PI3K–mTORC1 Regulates Transcription of HIF2A/EPAS1 and Vascularization in Neuroblastoma. Cancer Research. 2015;75(21):4617. doi:10.1158/0008-5472.CAN-15-0708
172 | Lynn
365. Stewart DJ, Issa J-P, Kurzrock R, et al. Decitabine Effect on Tumor Global DNA Methylation and Other Parameters in a Phase I Trial in Refractory Solid Tumors and Lymphomas. Clinical Cancer Research. 2009;15(11):3881. doi:10.1158/1078-0432.CCR-08-2196 366. Levenson VV. DNA methylation as a universal biomarker. Expert review of molecular diagnostics. 2010;10(4):481-488. doi:10.1586/erm.10.17 367. Osato M. Point mutations in the RUNX1/AML1 gene: another actor in RUNX leukemia. Oncogene. 2004/05/01 2004;23(24):4284-4296. doi:10.1038/sj.onc.1207779 368. Cohen Jr MM. Perspectives on RUNX genes: An update. American Journal of Medical Genetics Part A. 2009/12/01 2009;149A(12):2629-2646. doi:10.1002/ajmg.a.33021 369. Samarakkody AS, Shin N-Y, Cantor AB. Role of RUNX Family Transcription Factors in DNA Damage Response. Molecules and cells. 2020;43(2):99-106. doi:10.14348/molcells.2019.0304 370. Kamikubo Y. CROX (Cluster Regulation of RUNX) as a Potential Novel Therapeutic Approach. Molecules and cells. 2020;43(2):198-202. doi:10.14348/molcells.2019.0268 371. Martin JW, Zielenska M, Stein GS, van Wijnen AJ, Squire JA. The Role of RUNX2 in Osteosarcoma Oncogenesis. Sarcoma. 2011;2011:282745-282745. doi:10.1155/2011/282745 372. Wang Q, Stacy T, Binder M, Marin-Padilla M, Sharpe AH, Speck NA. Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis. Proceedings of the National Academy of Sciences. 1996;93(8):3444. doi:10.1073/pnas.93.8.3444 373. Komori T, Yagi H, Nomura S, et al. Targeted Disruption of Cbfa1 Results in a Complete Lack of Bone Formation owing to Maturational Arrest of Osteoblasts. Cell. 1997/05/30/ 1997;89(5):755-764. doi:https://doi.org/10.1016/S0092-8674(00)80258-5 374. Otto F, Thornell AP, Crompton T, et al. Cbfa1, a Candidate Gene for Cleidocranial Dysplasia Syndrome, Is Essential for Osteoblast Differentiation and Bone Development. Cell. 1997/05/30/ 1997;89(5):765-771. doi:https://doi.org/10.1016/S0092-8674(00)80259-7 375. Dunn J, Thabet S, Jo H. Flow-Dependent Epigenetic DNA Methylation in Endothelial Gene Expression and Atherosclerosis. Arteriosclerosis, Thrombosis, and Vascular Biology. 2015/07/01 2015;35(7):1562-1569. doi:10.1161/ATVBAHA.115.305042 376. Zhang Q, Cao X. Epigenetic regulation of the innate immune response to infection. Nature Reviews Immunology. 2019/03/27 2019;doi:10.1038/s41577-019-0151-6 377. Zeng L, Yang X-t, Li H-s, et al. The cellular kinetics of lung alveolar epithelial cells and its relationship with lung tissue repair after acute lung injury. Respiratory Research. 2016/12/07 2016;17(1):164. doi:10.1186/s12931-016-0480-y 378. Guéant J-L, Oussalah A, Zgheib R, Siblini Y, Hsu SB, Namour F. Genetic, epigenetic and genomic mechanisms of methionine dependency of cancer and tumor-initiating cells: What could we learn from folate and methionine cycles. Biochimie. 2020/06/01/ 2020;173:123-128. doi:https://doi.org/10.1016/j.biochi.2020.03.015 379. Gu X, Orozco JM, Saxton RA, et al. SAMTOR is an S-adenosylmethionine sensor for the mTORC1 pathway. Science. 2017;358(6364):813. doi:10.1126/science.aao3265 380. Crider KS, Yang TP, Berry RJ, Bailey LB. Folate and DNA methylation: a review of molecular mechanisms and the evidence for folate's role. Advances in nutrition (Bethesda, Md). 2012;3(1):21-38. doi:10.3945/an.111.000992 381. Bestor TH. The DNA methyltransferases of mammals. Human Molecular Genetics. 2000;9(16):2395-2402. doi:10.1093/hmg/9.16.2395 382. Santos-Bezerra DP, Admoni SN, Mori RC, et al. Genetic variants in DNMT1 and the risk of cardiac autonomic neuropathy in women with type 1 diabetes. Journal of Diabetes Investigation. 2019/07/01 2019;10(4):985-989. doi:10.1111/jdi.12988
173 | Lynn
383. Li H, Liu J-w, Sun L-p, Yuan Y. A Meta-Analysis of the Association between DNMT1 Polymorphisms and Cancer Risk. BioMed Research International. 2017/04/03 2017;2017:3971259. doi:10.1155/2017/3971259 384. Scott A, Song J, Ewing R, Wang Z. Regulation of protein stability of DNA methyltransferase 1 by post-translational modifications. Acta Biochimica et Biophysica Sinica. 2014;46(3):199-203. doi:10.1093/abbs/gmt146 385. Zhang Z, Yao L, Yang J, Wang Z, Du G. PI3K/Akt and HIF‑1 signaling pathway in hypoxia‑ischemia (Review). Molecular medicine reports. 2018;18(4):3547-3554. doi:10.3892/mmr.2018.9375 386. Meng S-S, Guo F-M, Zhang X-W, et al. mTOR/STAT-3 pathway mediates mesenchymal stem cell– secreted hepatocyte growth factor protective effects against lipopolysaccharide-induced vascular endothelial barrier dysfunction and apoptosis. Journal of Cellular Biochemistry. 2019/03/01 2019;120(3):3637-3650. doi:10.1002/jcb.27642 387. Xie T, Xu Q, Wan H, et al. Lipopolysaccharide promotes lung fibroblast proliferation through autophagy inhibition via activation of the PI3K-Akt-mTOR pathway. Laboratory Investigation. 2019/05/01 2019;99(5):625-633. doi:10.1038/s41374-018-0160-2 388. Jin H, Ciechanowicz A, Kaplan A, et al. Surfactant protein C dampens inflammation by decreasing JAK/STAT activation during lung repair. American journal of physiology Lung cellular and molecular physiology. 2018;314(5):L882-L892. doi:10.1152/ajplung.00418.2017 389. Robinson K, Vona-Davis L, Riggs D, Jackson B, McFadden D. Peptide YY Attenuates STAT1 and STAT3 Activation Induced by TNF-α in Acinar Cell Line AR42J. Journal of the American College of Surgeons. 2006/05/01/ 2006;202(5):788-796. doi:https://doi.org/10.1016/j.jamcollsurg.2006.01.007 390. Kamezaki K, Shimoda K, Numata A, Matsuda T, Nakayama K-I, Harada M. The role of Tyk2, Stat1 and Stat4 in LPS-induced endotoxin signals. International Immunology. 2004;16(8):1173-1179. doi:10.1093/intimm/dxh118 391. Si J, Collins SJ. IL-3–induced enhancement of retinoic acid receptor activity is mediated through Stat5, which physically associates with retinoic acid receptors in an IL-3–dependent manner. 10.1182/blood-2001-12-0374. Blood. 2002;100(13):4401. 392. Lentsch AB, Kato A, Davis B, Wang W, Chao C, Edwards MJ. STAT4 and STAT6 regulate systemic inflammation and protect against lethal endotoxemia. The Journal of clinical investigation. 2001;108(10):1475-1482. doi:10.1172/JCI13763 393. Matsukawa A, Kaplan MH, Hogaboam CM, Lukacs NW, Kunkel SL. Pivotal role of signal transducer and activator of transcription (Stat)4 and Stat6 in the innate immune response during sepsis. The Journal of experimental medicine. 2001;193(6):679-688. doi:10.1084/jem.193.6.679 394. Takeda K, Akira S. STAT family of transcription factors in cytokine-mediated biological responses. Cytokine & Growth Factor Reviews. 2000/09/01/ 2000;11(3):199-207. doi:https://doi.org/10.1016/S1359- 6101(00)00005-8 395. Hollander MC, Philburn RT, Patterson AD, Wyatt MA, Fornace JAJ. Genomic Instability in Gadd45a- /- Cells is Coupled with S-Phase Checkpoint Defects. Cell Cycle. 2005/05/02 2005;4(5):704-709. doi:10.4161/cc.4.5.1675 396. Zerbini LF, Wang Y, Czibere A, et al. NF-κB-mediated repression of growth arrest- and DNA- damage-inducible proteins 45α and γ is essential for cancer cell survival. Proceedings of the National Academy of Sciences of the United States of America. 2004;101(37):13618. doi:10.1073/pnas.0402069101 397. Barreto G, Schäfer A, Marhold J, et al. Gadd45a promotes epigenetic gene activation by repair- mediated DNA demethylation. Nature. 2007/02/01 2007;445(7128):671-675. doi:10.1038/nature05515 398. Salvador JM, Mittelstadt PR, Belova GI, Fornace AJ, Ashwell JD. The autoimmune suppressor Gadd45α inhibits the T cell alternative p38 activation pathway. Nature Immunology. 2005/04/01 2005;6(4):396-402. doi:10.1038/ni1176
174 | Lynn
399. Uddin MA, Akhter MS, Kubra KT, Barabutis N. P53 deficiency potentiates LPS-Induced acute lung injury in vivo. Curr Res Physiol. Dec 2020;3:30-33. doi:10.1016/j.crphys.2020.07.001 400. Barabutis N. Regulation of lung endothelial permeability by NEK kinases. IUBMB Life. 2020/04/01 2020;72(4):801-804. doi:10.1002/iub.2251 401. Liu L-Q, Tian F-J, Xiong Y, Zhao Y, Song J-B. Gadd45a gene silencing by RNAi promotes cell proliferation and inhibits apoptosis and senescence in skin squamous cell carcinoma through the p53 signaling pathway. Journal of Cellular Physiology. 2018/09/01 2018;233(9):7424-7434. doi:10.1002/jcp.26588 402. de Leon M, Cardenas H, Vieth E, et al. Transmembrane protein 88 (TMEM88) promoter hypomethylation is associated with platinum resistance in ovarian cancer. Gynecologic oncology. 2016;142(3):539-547. doi:10.1016/j.ygyno.2016.06.017 403. Schulz H, Ruppert A-K, Herms S, et al. Genome-wide mapping of genetic determinants influencing DNA methylation and gene expression in human hippocampus. Nature Communications. 2017/11/15 2017;8(1):1511. doi:10.1038/s41467-017-01818-4 404. Chiu DK-C, Tse AP-W, Xu IM-J, et al. Hypoxia inducible factor HIF-1 promotes myeloid-derived suppressor cells accumulation through ENTPD2/CD39L1 in hepatocellular carcinoma. Nature communications. 2017;8(1):517-517. doi:10.1038/s41467-017-00530-7 405. Wang W, Hu Y, Wang X, Wang Q, Deng H. ROS-Mediated 15-Hydroxyprostaglandin Dehydrogenase Degradation via Cysteine Oxidation Promotes NAD+-Mediated Epithelial-Mesenchymal Transition. Cell Chemical Biology. 2018/03/15/ 2018;25(3):255-261.e4. doi:https://doi.org/10.1016/j.chembiol.2017.12.008 406. Lee I, Yang C-M. Role of NADPH oxidase/ROS in pro-inflammatory mediators-induced airway and pulmonary diseases. Biochemical Pharmacology. 2012/09/01/ 2012;84(5):581-590. doi:https://doi.org/10.1016/j.bcp.2012.05.005 407. Brower R. Mechanical ventilation in acute lung injury and ARDS. Tidal volume reduction. CritCare Clin. 2002;18:1-13. 408. Amato MBP, Barbas CSV, Medeiros DM, et al. Effect of a Protective-Ventilation Strategy on Mortality in the Acute Respiratory Distress Syndrome. New England Journal of Medicine. 1998/02/05 1998;338(6):347-354. doi:10.1056/NEJM199802053380602 409. Webb HH, Tierney DF. Experimental Pulmonary Edema due to Intermittent Positive Pressure Ventilation with High Inflation Pressures. Protection by Positive End-Expiratory Pressure. American Review of Respiratory Disease. 1974/11/01 1974;110(5):556-565. doi:10.1164/arrd.1974.110.5.556 410. Dreyfuss D, Basset G, Soler P, Saumon G. Intermittent Positive-Pressure Hyperventilation with High Inflation Pressures Produces Pulmonary Microvascular Injury in Rats. American Review of Respiratory Disease. 1985/10/01 1985;132(4):880-884. doi:10.1164/arrd.1985.132.4.880 411. Sethi JK. Is PBEF/Visfatin/Nampt an Authentic Adipokine Relevant to the Metabolic Syndrome? Current hypertension reports. 2007;9(1):33-38. 412. Song HK, Lee MH, Kim BK, et al. Visfatin: a new player in mesangial cell physiology and diabetic nephropathy. 10.1152/ajprenal.90231.2008. American Journal of Physiology - Renal Physiology. 2008;295(5):F1485. 413. Villar J, Pérez-Méndez L, Flores C, et al. A CXCL2 polymorphism is associated with better outcomes in patients with severe sepsis*. Critical Care Medicine. 2007;35(10):2292-E1. doi:10.1097/01.ccm.0000284511.73556.59 414. Diebold I, Flugel D, Becht S, et al. The hypoxia-inducible factor-2alpha is stabilized by oxidative stress involving NOX4. Antioxid Redox Signal. 2010;13:425-436.
175 | Lynn
415. Hernandez-Pacheco N, Guillen-Guio B, Acosta-Herrera M, et al. A vascular endothelial growth factor receptor gene variant is associated with susceptibility to acute respiratory distress syndrome. Intensive care medicine experimental. 2018;6(1):16-16. doi:10.1186/s40635-018-0181-6 416. Wakai K, Matsuo K, Matsuda F, et al. Genome-wide association study of genetic factors related to confectionery intake: Potential roles of the ADIPOQ gene. Obesity. 2013/11/01 2013;21(11):2413-2419. doi:10.1002/oby.20316 417. Rhodes CJ, Batai K, Bleda M, et al. Genetic determinants of risk in pulmonary arterial hypertension: international genome-wide association studies and meta-analysis. The Lancet Respiratory medicine. 2019;7(3):227-238. doi:10.1016/S2213-2600(18)30409-0 418. Das S, Forer L, Schönherr S, et al. Next-generation genotype imputation service and methods. Nature Genetics. August 29 2016;48:1284. 419. Loh P-R, Danecek P, Palamara PF, et al. Reference-based phasing using the Haplotype Reference Consortium panel. Nature genetics. November 2016;48(11):1443-1448. 420. McCarthy S, Das S, Kretzschmar W, et al. A reference panel of 64,976 haplotypes for genotype imputation. Nature Genetics. 2016/10/01 2016;48(10):1279-1283. doi:10.1038/ng.3643
176 | Lynn