Genomics of acute lung injury

Acute lung injury (ALI) is a complex syndrome involving the interplay of both environmental (such as the addition of mechanical ventilation) and genetic factors. Clinical models have identified risk factors for development and poor outcome but these
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Genomics of Acute Lung Injury  Carlos Flores, Ph.D., 1 Shwu-Fan Ma, Ph.D., 1 Karen Maresso, M.P.H., 1 Omer Ahmed, M.D., 2 and Joe G.N. Garcia, M.D. 1 ABSTRACT Acute lung injury (ALI) is a complex syndrome involving the interplay of bothenvironmental (such as the addition of mechanical ventilation) and genetic factors. Clinicalmodels have identified risk factors for development and poor outcome but these strategiesremain imprecise. To better understand the mechanisms of pathogenesis associated withmechanisms of ALI, candidate genes identified by global expression profiling or relatedliterature searches are being explored for relevant polymorphisms (single base pairsubstitutions) that can affect both ALI susceptibility and outcome. This article summarizesseveral specific genetic association studies that have been conducted in ALI and reviewssupporting data from in vitro and in vivo models of the disease and clinical observations.Although valuable information has been reported to date, intense analyses are needed inthis developing discipline to assure significant clinical utility. The detailing of specificassociated polymorphisms will continue to provide new insights in the understanding of disease pathogenesis, and promise to reveal novel molecular targets and personalizedtreatments to prevent the disease. KEYWORDS:  Case-control association study, SNP, conserved expression patterns A cute lung injury (ALI) is a syndrome consist-ing of acute hypoxemia with respiratory failure as a resultof bilateral pulmonary edema. The incidence in theUnited States is   86 cases per 100,000 person-years. 1 Despite monitoring technological innovations and re-search-driven improvementsof the supportive care of thecritically ill patients, 2 the mortality rate remains high(around 30 to 50%). Even though the pathologicalmanifestations of ALI are caused by acute lung inflam-matory processes, including sepsis (generalized infec-tion) with a suspected pulmonary source, 1 the geneticpredisposition has been suggested as playing a major rolein the onset of ALI. This notion is supported by the factthat mortality rate in ALI is higher in African Ameri-cans than in other ethnic groups in the United States, 3 and that the outcome in patients with similar clinicalcharacteristics may vary from death to complete reso-lution. Several well-known risk factors, including age,gender, alcohol abuse, and diabetes have been reported;however, the pathological mechanisms of ALI remainunclear. It is therefore a challenging task to determine which environmental and genetic factors are relevant andhow these factors interact with each other.Both linkage and association studies have beenused for the identification of genetic susceptibility loci inrespiratory diseases. 4 However, the sporadic nature of ALI precludes a conventional linkage mapping strategy  where large families with both affected and unaffectedindividuals are examined for loci linked to the trait of interest. Because of these limitations, candidate gene-based association studies become the preferred type of analysis, where the frequencies between unrelated casesand controls of common variants, usually single nucleo-tide polymorphisms (SNPs), in genes of relevance for thedisease, are compared. This article reviews the associa-tion studies in ALI conducted to date, some of which 1 Department of Medicine, Section of Pulmonary and Critical CareMedicine, University of Chicago, Chicago, Illinois;  2 Louis A. WeissMemorial Hospital, Chicago, Illinois.Address for correspondence and reprint requests: Joe G.N. Garcia,M.D., Department of Medicine, Section of Pulmonary and CriticalCare Medicine, University of Chicago, 5841 S. Maryland Ave., W604,Chicago, IL 60637. E-mail: Lung Injury and Acute Respiratory Distress Syndrome; GuestEditors, John A. Belperio, M.D., Michael A. Matthay, M.D.,F.C.C.P.SeminRespir Crit Care Med 2006;27:389–395.Copyright # 2006by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York,NY 10001, USA. Tel: +1(212) 584-4662.DOI 10.1055/s-2006-948292. ISSN 1069-3424. 389    D  o  w  n   l  o  a   d  e   d   b  y  :   U  n   i  v  e  r  s   i   t  y  o   f   C   h   i  c  a  g  o .   C  o  p  y  r   i  g   h   t  e   d  m  a   t  e  r   i  a   l .  may have potential therapeutic implications. However, itis important to note that some of the association studiesare difficult to replicate unless well-designed case-con-trol studies including environmental covariables areconducted, with special emphasis of the power attained.In addition, heterogeneity between different studies andrisk factors also vary across populations. Because ALI is aclinical syndrome that lacks unique, easily measuredmarkers, but, rather, represents a spectrum of overlap-ping phenotypes, bias still may be present in the studiesbecause of the heterogeneous nature of ALI patients,reducing the power to detect a significant association. PRIORITIZING THE CANDIDATE-GENE LISTFOR ASSOCIATION WITH ALI  The crucial step in candidate-gene association studiesis the selection of the gene(s). Because both the settingand the progression of ALI involve a variety of local andsystemic interactions between cells, pathways, and fac-tors in space and time, the dissection of the geneticfactors underlying the pathogenesis of the disease un-doubtedly constitutes a multifaceted challenge. Clinicalobservations and pathway connections to biomarkers of the disease have been used with great success as a sourcefor ALI candidate genes. Comprehensive analyses of gene expression profiles have also unraveled new targetsof unsuspected contribution to ALI. Recently, we gen-erated a list of candidate genes phylogenetically involvedin the pathogenesis of ALI (Fig. 1) by analyzing multipleanimal and in vitro ALI models. 5,6 Apart from inflam-mation, a component known to be the key process inALI development, four other prominent biological proc-esses (ontologies), including chemotaxis, immune re-sponse, regulation of cell proliferation, and bloodcoagulation were revealed. 7  This approach successfully identified genes previously linked to ALI, as well asnovel ones that offer new insights into disease patho-genesis (Table 1). Further identification and analysis of functionally relevant genetic variants of the candidategenes may offer the opportunity to understand individ-ual variability in disease predisposition and severity andmay provide new therapeutic targets and the ability totailor the management of critically ill patients. The nextsection focuses on the genetic variants of candidategenes, along with in vitro, in vivo, and clinical evidencesfor their contribution in ALI development, and theirassociation with ALI susceptibility and outcome. ASSOCIATION STUDIES AND GENETICSUSCEPTIBILITY TO ALI Angiotensin Converting Enzyme  The rennin-angiotensin system has an important role inmaintaining blood pressure homeostasis and regulatescirculatory salt and water. The angiotensin convertingenzyme (ACE) and ACE2, an ACE homologue withnegative regulator activity, play a central role in thissystem by balancing the levels of angiotensin I and II. Anintronic  Alu   insertion/deletion (I/D) polymorphism of   ACE   has been shown to be associated with at least 28 to Figure1  Schematic representation of the global expression analysis to identify acute lung injury (ALI) associated candidate genes. Inbrief,totallungribonucleicacids(RNAs)frommouse,rat,andcanineALImodelsaswellasRNAsfromhumanendothelialcellsexposedto mechanical stretch were hybridized to microarrays. Gene expression profiles were then analyzed to generate the orthologous geneprofiles. Significantly up- or downregulated genes were extracted using the Significance Analysis of Microarray (SAM; followed by gene ontology classification. 390  SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 27, NUMBER 4 2006    D  o  w  n   l  o  a   d  e   d   b  y  :   U  n   i  v  e  r  s   i   t  y  o   f   C   h   i  c  a  g  o .   C  o  p  y  r   i  g   h   t  e   d  m  a   t  e  r   i  a   l .  47% of variance in plasma ACE activity where the Dallele possesses higher enzyme activity. 8,9 Congruently,the D allele has been shown to produce larger amountsof   ACE   messenger ribonucleic acid (mRNA) in whiteblood cells. 10  Whether genetic variants of   ACE   areassociated with the susceptibility and/or outcome of acute respiratory distress syndrome (ARDS), an ALIsubtype characterized by a more severe hypoxemia, is not yet clear. Marshall et al 11 first established that thehomozygous  ACE   deletion genotype (DD) may beassociated with susceptibility and outcome of ARDS inthe European population. However, a recent replicationstudy in a Chinese population showed it to be associatedonly with outcome but not with susceptibility. 12  ACE  and  ACE2  have also been implicated in the susceptibility to severe acute respiratory syndrome (SARS)-coronavi-rus infection and disease outcomes. 13 However, recentstudies by Chiu et al 14 and Chan et al 15 concluded thatneither the  ACE   I/D polymorphism nor the  ACE2  genepolymorphism are directly associated with increasedsusceptibility or poor outcomes after SARS-coronavirusinfection in the Chinese population. Interleukin-6 Interleukin-6 (IL-6) is a pleiotropic proinflammatory cytokine playing a key role in the acute-phase responseand the activation of B and T cells. Inflammatory cyto-kines,includingIL-6,areessentialfortheimmunesystemhomeostasis, but their exaggerated productions have beenshown to have devastating effects. Adverse outcome inALI is related to the increased levels of IL-6 both inhumans 16 and in animal models. 17 IL-6 is an efficientbiomarkeroftheoutcomeinALI/ARDSandthusagoodcandidate to search for genetic variants that can beassociated with the syndrome. A G/C SNP located atposition -174 of the  IL6  gene promoter has been studiedfor association with a large list of conditions (see Holle-gaard and Bidwell 18 for an updated database). Although-174 G/C SNP is located in a functional site of the  IL6 promoter, 19 results from both association studies withother diseases 18 and correlations between alleles andIL-6 levels in plasma have been conflicting. 19–21 In rela-tion to ALI, Marshall et al 16 compared ARDS cases withtwo high-risk ARDS groups, intensive care unit (ICU)non-ARDS patient samples and coronary artery bypassgrafting samples of European descent. They reported noassociation of this SNP with susceptibility to ARDS,albeit the CC genotype was found less frequently inICU nonsurvivors. More recently, taking advantage of SeattleSNPs, a National Heart Lung and Blood Institute(NHLBI)-funded effort to resequence candidate genesthat underlie inflammatory responses in humans (, Sutherland et al 22 implicatedthe  IL6  gene in ALI development by studying the-174 G/C SNP along with two other deep-rootingSNPs of the genealogy of the  IL6  gene in a cohort of septic patients of European descent. Although nosingle SNP was found to be associated with any of the outcomes tested, genotypes carrying two copies of specific three-SNP-haplotypes showed association witha higher mortality, and other secondary clinical out-comes, including shorter survival and free of ALI. Thefact that both alleles of the -174 SNP were present inthe group of risk haplotypes further supports that theposition -174 may not be the causal locus for theassociation, or at least not the only one. Recent studiesin  IL6  are refocusing on other polymorphisms 23 or evenon several of them in the  IL6  region, which has led tothe hypothesis that haplotypes of the promoter may bemore functionally relevant to the disease. 24,25 Interleukin-10 Interleukin-10 (IL-10) is an important cytokine withpleiotropic effects in immunomodulation and inflamma-tion. IL-10 acts as an anti-inflammatory cytokine, in-hibiting the production of several other cytokines,including interferon gamma (IFN- g ), IL-2, IL-1 b ,IL-6, IL-8, IL-12, and tumor necrosis factor-alpha Table 1 Candidate Genes in Acute Lung Injury Candidate GenesPubMatrix TermsAcute Lung Injury Mechanical Ventilation Lung Inflammation Angiotensin converting enzyme 91 58 77Angiotensin converting enzyme 2 3 0 0Interleukin-6 255 154 378Interleukin-10 97 56 210Myosin light chain kinase 9 23 1Pre-B cell colony enhancing factor 4 1 0Tumor necrosis factor- a  596 178 714Tumor necrosis factor- b  5 4 14Vascular endothelial growth factor 34 95 13Surfactant proteins 348 285 268 Shown are the number of references identified by PubMatrix ( citing the genes and relevant terms. GENOMICS OF ACUTE LUNG INJURY / FLORES ET AL  391    D  o  w  n   l  o  a   d  e   d   b  y  :   U  n   i  v  e  r  s   i   t  y  o   f   C   h   i  c  a  g  o .   C  o  p  y  r   i  g   h   t  e   d  m  a   t  e  r   i  a   l .  (TNF- a ). During sepsis, IL-10 release is increased, 26 and the increased levelsare correlated with the severity of the injury  27 and mortality. 28 At least 50% of interindi- vidual differences in IL-10 production are attributable togenetic factors. 29,30  These differences have been ascribedto genetic variation in the 5 0 end, including the pro-moter, of the gene. 30 Among several known polymor-phisms in the 5 0 end of the gene, 31,32 a G/A SNP atposition   1082 have been recognized as the mostimportant variation regulating the constitutive mRNAlevel of   IL10  . 33 Although the   1082 GG homozygousindividuals have been suggested to produce the highestlevels of IL-10, 34,35 the haplotypes of this regulatory region of the gene, and not isolated single variants, alsoaffect the production of   IL10   mRNA and proteindepending on the stimulus. 36–38 Recently, Gong et al 39 tested the association of   1082 G/A SNP with develop-ment and outcome of ARDS in a well characterized andlarge ICU European American cohort. Despite somemethodological limitations (focusing on a single poly-morphism, power of the tested interactions, and devia-tions from Hardy-Weinberg equilibrium), the datasuggested that the   1082 GG genotype may be a risk factor associated with the development of ARDS, de-pending on age. Strikingly, the same genotype may confer protection among the ARDS patients in termsof mortality and organ failure. This is, however, notsurprising because the role of this cytokine in systemicinflammations is so complex that, when used as a treat-ment in a model of sepsis, it may display opposite effectsdepending on the timing of intervention. 40 Myosin Light Chain Kinase Myosin light chain kinase gene (  MYLK  ), encompassingmore than 200 kb, encodes three proteins: nonmuscleendothelial cell myosin light chain kinase (EC MLCK),smooth muscle MLCK (sm MLCK), and telokin. ECMLCK is centrally involved in cytoskeleton rearrange-ment participating in apoptosis, inflammation, cell traf-ficking, cell division, and angiogenesis (Dudek andGarcia 41 and references therein). Both in vivo and in vitro studies have shown that lung endothelial cell barrierdysfunction plays a key role in pathophysiology of ALI,andthatlung edemaformation involves MLCK-depend-ent cytoskeletal rearrangement. 41 EC MLCK knockoutmice (retaining thesmMLCK isoform) have been shownto be less susceptible to LPS-induced ALI and thereforelonger survival rate during subsequent mechanical ven-tilation. 42 Addition of MLCK inhibitor prior to lip-opolysaccharide (LPS) treatment in wild-type mouseattenuated lung barrier permeability and inflamma-tion. 42,43  We assessed the genetic association of MLCK  with susceptibility to sepsis and ALI with an in-depthanalysis. 44 Briefly, direct sequencing of the 32 exons(including exon-intron boundaries) and 2 kb of 5 0 UTR of gene in healthy, sepsis-alone, and sepsis-associatedALI individuals of European and African Americandescent ( n ¼ 6 per group) led to over 50 SNPs identified.An SNP map of the gene with an average of one SNPevery    8 kb was constructed for linkage disequilibriumstudies. Subsequently, association analysis of both singleSNPs and haplotypes demonstrated very strong associa-tions, sometimes shared between sepsis and ALI, of haplotypes in 5 0 of the gene in both population groups.Although the results are encouraging due to its replica-tion in two different population groups, this associationneeds a further exploration to unravel the functionally relevant variant(s). Pre-B Cell Colony Enhancing Factor Pre-B cell colony enhancing factor (PBEF), a relatively unknown cytokine, has been shown to be upregulated by acute mechanical distension 45 and multiple proinflam-matory molecules. 46  We recently identified this cytokineas a novel biomarker in ALI by extensive analysis of expression profiling in animal and cell culture models andprotein levels in bronchoalveolar lavage (BAL) fluidsamples from ALI patients. 47 In vitro studies demon-strated that PBEF increases the pulmonary permeability by regulating Ca 2 þ -dependent cytoskeletal rearrange-ments in endothelial cells. 48 PBEF therefore acted as adual transducer involved in both mechanical stress andinflammation. To perform the association study, Yeet al 47 further resequenced part of the gene of healthy,sepsis alone, and sepsis-associated ALI subjects of Euro-pean and African American descent ( n ¼ 6 per group)and genotyped two of the SNPs (positions C-1543T and T-1001G) close to mapped transcription factor bindingsites. In addition, the strongest association was found with the   1543C/  1001G haplotype conferring risk for susceptibility to sepsis and ALI, which was furthersupported by functional assays. The recent finding thatPBEF expression is increased in neutrophils during sepsisand delays apoptosis allows new pathogenic schemes tobe developed. Surfactant Proteins Surfactant proteins (SPs) are important for host defenseand lung function. They contribute in reducing the sur-face tension in the lung alveoli allowing for the normalexpansion of the lungs. Injury in the alveolar epithelia is ahallmark of ALI/ARDS, and experimental models of ALI indicate that serum levels of SP increase progres-sively, reflecting lung injury and increased permeability, 49  with the concomitant reduction of SP lung levels. 50 Plasma levels in ARDS patients have also been shownto be correlated with a worse clinical outcome. 51 Severalstudies have been done to examine if SP polymorphismscould be associated with susceptibility and outcome of  392  SEMINARS IN RESPIRATORY AND CRITICAL CARE MEDICINE/VOLUME 27, NUMBER 4 2006    D  o  w  n   l  o  a   d  e   d   b  y  :   U  n   i  v  e  r  s   i   t  y  o   f   C   h   i  c  a  g  o .   C  o  p  y  r   i  g   h   t  e   d  m  a   t  e  r   i  a   l .  ALI/ARDS. Although the polymorphisms analyzed were different and the number of ARDS cases studied was limited, all studies point to the association of   SP-B  ( SFTPB  ) variants with susceptibility to ARDS. 52–54 Tumor Necrosis Factor  TNF- a  and lymphotoxin- a  (LT- a , also known as TNF- b ) are two potent proinflammatory cytokines. TNF- a  is commonly used as a marker of inflammation, with a key role as an early mediator in the developmentof ALI. 2  TNF- a is known to increase endothelial hyper-permeability leading to subsequent vascular leakage. 55 Although its soluble receptors have been associated withmorbidity and mortality in ALI patients, 56  TNF- a levelsmeasured from different fluids have not been consis-tently correlated with severity in patients with ALI. 57 Both  TNF   and  LTA   genes lie in close proximity withinthe major histocompatibility complex (MHC), and sev-eral polymorphisms have been described in the region. AG/A SNP located at the position   308 of the  TNF  promoter and a G/A SNP at position þ 90 in intron 1(also known as TNFB  Nco I) of the  LTA   gene are, by far,the most commonly examined polymorphisms of theregion in association studies. 58 Both SNPs have beenimplicated in the modulation of expression levels but whether these polymorphisms have functional signifi-cance is conflicting. In great part, due to the extent of linkage disequilibrium in the region, 35,58 it is still amatter of debate whether these are regulatory SNPs orif the protein level is modulated by a third locus or ahaplotype. 59 Recently, Gong et al, 60 using a case-controlsample nested in a large and well characterized cohort of European descent, studied the association of the  TNF   308 G/A SNP and the TNFB  Nco I with ARDS. By stratifying the sample of cases in direct or indirectpulmonary injury, they showed both SNPs to be asso-ciated as protective factors with susceptibility to ARDS( TNF    308 G/A SNP only in the direct pulmonary injury group, and the TNFB  Nco I only in the indirectpulmonary injury group). In terms of mortality, only the TNF    308 G/A SNP showed a significant associationbut, remarkably, as a risk factor, which bears a resem-blance, albeit inverted, to the situation described by thesame group for the association with  IL10  . Haplotypes of the two SNPs resembled the associations of the  TNF   308 G/A SNP by itself. Vascular Endothelial Growth Factor Vascular endothelial growth factor (VEGF) is a key cytokine in regulation of angiogenesis. VEGF stimulatesendothelial cells to proliferate and migrate but alsopromotes cell permeability, allowing for protein accu-mulation in the extravascular space. 61  Thus the action of this cytokine may be directly related to pulmonary edema, one of the hallmarks in ALI. Congruent withthis idea, it has been shown that plasma VEGF issignificantly elevated in ARDS patients, whereas intra-pulmonary levels are decreased compared with non-ARDS patients. 62 Several polymorphisms have beendescribed in the  VEGF   gene, and most of them havebeen studied in association with cancer susceptibility andseverity. However, a recent work studied the associationof a  VEGF   SNP with ARDS susceptibility and severity.Medford et al 63 analyzed the C/T SNP at þ 936 of the3 0 UTR of the gene, which was previously associated withthe VEGF plasma levels in healthy subjects, 64 in a case-control sample of European descent and found associa-tion with both ARDS susceptibility and severity, asmeasured by the Acute Physiology and Chronic HealthEvaluation (APACHE) III score. CONCLUDING REMARKS  The study of variation in genes relevant for ALI patho-genesis is a developing field. In several genes studied sofar, the functional consequences of the associated var-iants are known, providing biological plausibility as risk or protective factors for ALI susceptibility and outcome.For most genes,however, it remains a challenge to definethis functional association. For that purpose, well de-signed studies, with SNP discovery on candidate genesin samples of different ancestries, including analysis of linkage disequilibrium and functional assays, will beneeded to extend our current understanding of thepathophysiology of ALI. In this respect, studying ge-netic variants with well-supported associations in otherinflammatory conditions may be useful to disentangleindividual susceptibility to ALI. Sepsis, widely recog-nized as one of the main risk factors leading to thedevelopment of ALI, has been shown to be associated with genes involved in multiple processes of the bacteria-induced cellular response. 65  Thus genes of relevance forsusceptibility to sepsis constitute potential candidates of relevance for susceptibility to ALI. In fact, some of thesepsis-associated genes, such as  ACE  ,  IL6 ,  TNF,  and LTA  , have already been associated with ALI suscepti-bility. Other sepsis-associated genes, such as  PAI-1 , PLAT,  and  CXCL2 , although not yet studied for ALIassociation, have been directly selected as candidategenes in ALI cross-species expression patterns.Currently, several procedures and drugs are beingevaluated for the treatment of ALI/ARDS patients inclinical trials, though the most effective treatment todate involves ventilator adjustments known to reducelocal and systemic inflammation. Thus a better classi-fication of patients at risk for ALI/ARDS is importantto develop new strategies to prevent the development of the syndrome. Ethnic differences in the incidence andmortality from ALI may provide with clues to reveal key processes for the disease. The identification of genetic GENOMICS OF ACUTE LUNG INJURY / FLORES ET AL  393    D  o  w  n   l  o  a   d  e   d   b  y  :   U  n   i  v  e  r  s   i   t  y  o   f   C   h   i  c  a  g  o .   C  o  p  y  r   i  g   h   t  e   d  m  a   t  e  r   i  a   l .
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks