gadd45a is a novel candidate gene in inﬂammatory lung ...€¦ · novel vili-related candidate...

The FASEB Journal • Research Communication

GADD45a is a novel candidate gene in inflammatorylung injury via influences on Akt signaling

Nuala J. Meyer,* Yong Huang,† Patrick A. Singleton,* Saad Sammani,* Jaideep Moitra,*Carrie L. Evenoski,* Aliya N. Husain,‡ Sumegha Mitra,* Liliana Moreno-Vinasco,*Jeffrey R. Jacobson,* Yves A. Lussier,† and Joe G. N. Garcia*,1

*Section of Pulmonary and Critical Care Medicine and †Section of Genetic Medicine, Departmentof Medicine, and ‡Department of Pathology, Pritzker School of Medicine, University of Chicago,Chicago, Illinois, USA

ABSTRACT We explored the mechanistic involve-ment of the growth arrest and DNA damage-induciblegene GADD45a in lipopolysaccharide (LPS)- and venti-lator-induced inflammatory lung injury (VILI). Multiplebiochemical and genomic parameters of inflammatorylung injury indicated that GADD45a�/� mice are mod-estly susceptible to intratracheal LPS-induced lung in-jury and profoundly susceptible to high tidal volumeVILI, with increases in microvascular permeability andbronchoalveolar lavage levels of inflammatory cyto-kines. Expression profiling of lung tissues from VILI-challenged GADD45a�/� mice revealed strong dysregu-lation in the B-cell receptor signaling pathway comparedwith wild-type mice and suggested the involvement ofPI3 kinase/Akt signaling components. Western blotanalyses of lung homogenates confirmed �50% reduc-tion in Akt protein levels in GADD45a�/� mice accom-panied by marked increases in Akt ubiquitination. Elec-trical resistance measurements across human lungendothelial cell monolayers with either reducedGADD45a or Akt expression (siRNAs) revealed signifi-cant potentiation of LPS-induced human lung endothe-lial barrier dysfunction, which was attenuated by over-expression of a constitutively active Akt1 transgene.These studies validate GADD45a as a novel candidategene in inflammatory lung injury and a significantparticipant in vascular barrier regulation via effects onAkt-mediated endothelial signaling.—Meyer, N. J.,Huang, Y., Singleton, P. A., Sammani, S., Moitra, J.,Evenoski, C. L., Husain, A. N., Mitra, S., Moreno-Vinasco, L., Jacobson, J. R., Lussier, Y. A., Garcia,J. G. N. GADD45a is a novel candidate gene in inflam-matory lung injury via influences on Akt signaling.FASEB J. 23, 000–000 (2009)

Key words: mechanical ventilation � biomarker � vascular bar-rier regulation � ubiquitination

Acute lung injury (ALI), a common and highlymorbid inflammatory syndrome, is characterized byincreased pulmonary endothelial and epithelial perme-ability leading to alveolar flooding (1) and a mortality�35% (2). While infectious causes, such as sepsis and

pneumonia, are the most common ALI precipitants,supportive mechanical ventilation therapy may incite orworsen preexisting ALI, a syndrome known as ventila-tor-induced lung injury (VILI) (3, 4). Consistent withthe concept that ventilatory support, while vital, isnevertheless potentially injurious (4, 5), multiple ani-mal models and in vitro studies (6–10) support aninjurious role for excessive mechanical stress and ob-servational studies (4) suggest that �25% of critically illpatients without ALI at the initiation of mechanicalventilation will develop the syndrome during their first5 days on the ventilator. Despite identification of sev-eral risk factors for the development of ALI (includingsepsis, pneumonia, aspiration, and high tidal volumes;refs. 5, 11), only a minority of patients with these riskfactors develop the syndrome. This marked heteroge-neity, combined with the observed disparities in ALIbetween different ethnic and racial populations, lendssupport to the hypothesis that a genetic contributionmay underlie ALI susceptibility (12, 13).

We previously searched for novel ALI/VILI candi-date genes via orthologous gene expression profiling ofin vivo (murine, rat, and canine) and in vitro [humanpulmonary endothelial cells (ECs)] models of in-creased mechanical stress consistent with VILI (14).These studies generated a list of mechanosensitivecandidate genes unidirectionally and differentially ex-pressed across all species, including genes stronglyimplicated in ALI pathogenesis as well as novel candi-dates previously unassociated with lung injury, ventila-tion, or pulmonary pathophysiology (14, 15). Onenovel VILI-related candidate gene thus identified wasthe growth arrest and DNA damage-inducible geneGADD45a (14). GADD45a exhibits low constitutive ex-pression with transcriptional activation by cellulargenotoxic and nongenotoxic stressors, including ultra-violet and ionizing radiation, hyperoxia, and endotoxin[lipopolysaccharide (LPS); refs. 16–19]. GADD45a is

1 Correspondence: Department of Medicine, W604, Pritz-ker School of Medicine University of Chicago, 5841 S. Mary-land Ave., W604 Chicago, IL 60637, USA. E-mail: [email protected]

doi: 10.1096/fj.08-119073

10892-6638/09/0023-0001 © FASEB

The FASEB Journal article fj.08-119073. Published online January 5, 2009.

recognized as a participant in the regulation of the cellcycle, apoptosis, maintenance of genomic stability,DNA methylation excision and repair, and regulationof Th1 differentiation (20–26). The involvement ofGADD45a in inflammatory lung processes, however, isunknown. We utilized in vivo models of LPS-inducedlung injury and VILI and genetically engineered micewith targeted GADD45a deletion to examine the partic-ipation of GADD45a in inflammatory lung injury. Ourresults are consistent with a significant role forGADD45a in inflammatory lung injury with genomicand cellular analyses, suggesting GADD45a participa-tion in vascular barrier regulation via effects on Akt-mediated endothelial signaling.

MATERIALS AND METHODS

Cell culture and reagents

Standard reagents including LPS (batch O127B8) were ob-tained from Sigma-Aldrich (St. Louis, MO, USA) unlessotherwise specified. Human pulmonary arterial endothelialcells (HPAECs) were obtained from Cambrex (Walkersville,MD, USA) and cultured as described previously (27). ForSDS-PAGE, reagents were purchased from Bio-Rad (Rich-mond, CA, USA), Immobilon-P transfer membrane was fromMillipore (Bedford, MA, USA), and gold microelectrodeswere from Applied Biophysics (Troy, NY, USA). Primaryantibodies for GADD45a as well as short interfering RNAs(siRNAs) specific for GADD45a, Akt1, and control sequencewere purchased from Santa Cruz Biotechnology (Santa Cruz,CA, USA). Primary antibodies for Akt, phospho-Akt (Ser-473), and ubiquitin, as well as secondary antibodies, werepurchased from Cell Signaling Technology (Danvers, MA,USA). The Akt1/PKB� (active) cDNA expression vector waspurchased from Millipore.

GADD45a-engineered mice

GADD45a�/� mice (129/Ola background) generously pro-vided by Dr. Michael O’Reilly (University of Rochester, Roch-ester, NY, USA) and Dr. Albert Fornace (Brigham andWomen’s Hospital, Boston, MA, USA) were backcrossed ontothe C57BL/6 background for eight generations (28). Wild-type (WT) C57BL/6 mice were obtained from Jackson Labo-ratories (Bar Harbor, ME, USA).

Preclinical model of VILI

All experiments were approved by the Animal Care and UseCommittee at the University of Chicago. Mice were housedunder standard conditions with free access to food and water.Mechanical ventilation experiments were performed in age-matched male WT C57BL/6 and GADD45a�/� mice (8–12wk) after anesthesia with inhaled isofluorane followed byintraperitoneal ketamine/acetylpromazine (150/15 mg/kgrespectively). Mice were intubated (20-gauge catheter) andplaced on mechanical ventilation (Harvard Apparatus, Hol-liston, MA, USA) at room air, tidal volume of 30 ml/kg (VILIgroups, n�8/group), 65 breaths/min, and positive end expi-ratory pressure of 0 cm H2O for 4 h (29). Control mice wereallowed to breathe spontaneously for 4 h (n�6/group).Ventilated WT and GADD45a�/� mice were monitored withintermittent blood pressure and arterial blood gas monitor-

ing to ensure adequate perfusion and received 8 ml/kg 0.9%saline at the initiation of ventilation and at 2 h after the onsetof ventilation. Deep anesthesia was maintained with ket-amine/acetylpromazine throughout the experiment.

Preclinical model of LPS-induced lung injury

A second model was used to determine the involvement ofGADD45a in inflammatory lung injury. Age-matched maleC57BL/6 and GADD45a�/� mice (8–12 wk) were anesthe-tized and intubated as described above. LPS (1.25 mg/kg;n�15–16/group) or an equal volume of water (1.25 ml/kg;n�8/group) was administered intratracheally as describedpreviously (30, 31). After extubation, mice from the fourexperimental groups were allowed to breathe spontaneouslyfor 18 h with free access to food and water and thenunderwent harvesting as described for the mouse VILI model.

Bronchoalveolar lavage (BAL) and lung tissue expressionmeasurements

BAL was recovered as we previously described (30) fordetermination of cell counts and differentials, protein (Bio-Rad DC protein assay), albumin (ELISA; Bethyl Laboratories,Montgomery, TX, USA) (31) and levels of mouse cytokines[interleukin (IL)-1�, IL-6, keratinocyte-derived chemokine(KC), macrophage inflammatory protein (MIP)-2, and tumornecrosis factor (TNF-�)] using a Bio-Plex bead assay (Bio-RadLaboratories; ref. 32). Lung tissue was homogenized in ex-traction buffer and assayed for albumin by ELISA (BethylLaboratories).

Histology and immunohistochemistry

To characterize VILI- or LPS-mediated alterations in lungmorphology and to localize GADD45a expression, left lungs(2 animals/group) were excised from the mainstem bron-chus at death and placed immediately in formalin overnight,followed by embedding in paraffin for histological evaluationby hematoxylin and eosin staining as we previously described(30). A masked lung pathologist (A.N.H.) graded the severityof each slide for edema, inflammation, and the presence ofhyaline membranes utilizing the following criteria: 1) edema:absent, mild (�10% alveoli involved), moderate (involving10–50% alveoli), or severe (involving�50% alveoli); 2) in-flammation; absent, mild [�10 inflammatory cells/highpower field (hpf)], moderate (10–50 inflammatory cells/hpf), or severe (�50 inflammatory cells/hpf); and 3) hyalinemembranes: present or absent. Immunohistochemistry wasperformed on paraffin-embedded sections using rabbit anti-GADD45a (Santa Cruz Biotechnology) at a 1:40 dilution asdescribed previously (33). Each sample was graded (n�2/group) from 0 to 3� in the following locations: pulmonaryepithelium (including airway epithelium and type II pneumo-cytes), pulmonary endothelium, and inflammatory cells.

RNA isolation and microarray analysis

Total RNA was isolated from whole lungs for expressionprofiling as described previously (34) using Affymetrix Mouse430_2 arrays and protocols (Affymetrix, Santa Clara, CA,USA). Chip quality and “present” or “absent” expression callswere determined by the GeneChip Operating Software. In-tensities and normalization of probe sets were calculated byBioconductor software (GCRMA package; refs. 35, 36). Themicroarray data have been submitted to the National Centerfor Biotechnology Information’s Gene Expression Omnibus

2 Vol. 23 May 2009 MEYER ET AL.The FASEB Journal

repository (GSE11662). To identify differentially expressedgenes, two-group comparisons were conducted using signifi-cance analysis of microarrays (SAM) as described previously(37). Gene filtering parameters and results are summarizedin Supplemental Table S1. “Dysregulated genes” are thosethat were differentially expressed with �2-fold change vs.control. For redundant probe sets representing the sameEntrez Gene or UniGene ID, only the probe set with thelowest false discovery rate or highest fold change was includedin the gene list.

Dysregulated genes were uploaded into the IngenuityPathway Analysis (IPA) software (http://www.ingenuity.com), a web-delivered application that utilizes the IngenuityPathways Knowledge Base (IPKB) containing a large amountof individually modeled relationships between gene objects,e.g., genes, mRNAs, and proteins, to dynamically generatesignificant regulatory and signaling networks or pathways.The genes submitted for mapping to corresponding geneobjects in the IPKB are called “focus genes.” The significanceof a canonical pathway is controlled by P value, which iscalculated using the right-tailed (referring to the overrepre-sented pathway) Fisher’s exact test for 2 2 contingencytables. This is done by comparing the number of focus genesthat participate in a given pathway, relative to the totalnumber of occurrences of those genes in all pathways storedin the IPKB. The significance threshold of a canonicalpathway is set to 1.3, which is derived by �log10 [P value],with P � 0.05.

Real-time RT-PCR and analysis

Transcript levels of CXCL1, CXCL2, PIK3CD, and Akt1 inhomogenized mouse lungs from VILI-challenged and spon-taneously breathing mice (n�3 per condition) were mea-sured in 96-well microtiter plates with an ABI Prism 7700Sequence Detector System (Applied Biosystems, Foster City,CA, USA) as we have described (38). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internalcontrol for normalization. All primers and probes werepurchased from Applied Biosystems in a 20 mixture. Exper-imental protocols were based on the manufacturer’s recom-mendation using the TaqMan Gold RT-PCR Core ReagentsKit (Applied Biosystems). Experimental parameters were48°C for 30 min followed by 40 cycles of 95°C for 15 s and60°C for 1 min. A relative quantitative method was used toanalyze changes in gene expression in a given sample relativeto an untreated control sample, and specific mRNA transcriptlevels were expressed as fold difference, calculated by raising2 to the power of the difference in geometric means. Com-parisons of the fold change in each transcript were performedusing unpaired t tests.

Immunoprecipitation and immunoblotting

Left lung homogenates were sonicated in immunoprecipita-tion buffer (50 mM HEPES, pH 7.5; 150 mM NaCl; 20 mMMgCl2; 1% Nonidet P-40; 0.4 mM Na3VO4; 40 mM NaF; 50M okadaic acid; 0.2 mM phenylmethylsulfonyl fluoride; and1:250 dilution of protease and phosphatase inhibitors (Cal-biochem, San Diego, CA, USA). In companion experiments,HPAEC lysates were normalized for protein concentrationfollowed by SDS-PAGE in 4–15% gradient polyacrylamidegels, transferred onto Immobilon membranes, and incubatedwith specific primary and secondary antibodies. Immunore-active bands were visualized using enhanced chemilumines-cence (Amersham Biosciences, Piscataway, NJ, USA). Stan-dardized average gray values processed from ImageQuant

software (Amersham Biosciences) were obtained from immu-noreactive bands for quantification.

Transfection of siRNA and cDNA constructs

HPAECs were transfected with siRNA (Santa Cruz Biotech-nology) specific for GADD45a (75 nM), Akt1 (75 nM), or ascrambled sequence (50 nM) using siPORT Amine (Ambion,Austin, TX, USA) as the transfection reagent according to themanufacturer’s protocol. The Akt1/PKB� (active) cDNA con-struct (Millipore, Bedford, MA, USA) was diluted 1:1 inFuGENE HD (Roche, Basel, Switzerland) and transfectedconcurrently with siRNA (scramble or GADD45a). Transfec-tions were carried out 48 h before transendothelial electricalresistance (TER) measurements.

Measurement of TER

HPAECs were grown to confluence in polycarbonate wellscontaining evaporated gold microelectrodes, and TER mea-surements were performed using an electrical cell substrateimpedance sensing system (Applied Biophysics) as we havepreviously described in detail (27, 39, 40). TER values fromeach microelectrode were pooled at discrete time points andplotted as the mean � se.

Statistical methodology

Results are presented as means � se, with the exception ofthe array and quantitative PCR (qPCR) gene expression datathat are presented as mean fold change � sd. Physiologicalmeans were compared using Stata (Stata, College Station, TX,USA) by one-way ANOVA and corrected for multiple com-parisons using the Bonferroni method. Parameters demon-strating significantly unequal variances and non-normal dis-tributions (Akt abundance) were analyzed by nonparametrictesting (Mann-Whitney rank sum test; Stata). Gene expres-sion data (array and qPCR) were transformed from absoluteexpression values to a log 2 scale by determining the expres-sion ratio [�2ˆ(experimental expression value�control ex-pression value)]. When the expression ratio is positive, foldchange equals the ratio, whereas a negative ratio is convertedto fold change by the formula �1/expression ratio]. Foldchanges were compared using Student’s t test. Predeterminedsignificance level was P � 0.05.

RESULTS

Susceptibility of GADD45a�/� mice to VILI

To assess the role of GADD45a in inflammatory lunginjury, WT C57Bl/6 (WT) and GADD45a�/� micewere exposed to extensive levels of mechanical stressvia mechanical ventilation (4 h) at a high tidalvolume (30 ml/kg). WT mice exhibited modest lunginjury with increased BAL cellularity (Fig. 1) andenhanced microvascular permeability reflected byincreases in BAL protein and albumin (Fig. 2). BALlevels of proinflammatory cytokines (KC, MIP-2, IL-1�, IL-6, and TNF-�) were also increased in VILI-challenged WT mice, with IL-6 achieving statisticalsignificance (P�0.03; Fig. 3).

In contrast to WT mice, VILI-challenged GADD45a�/�

mice exhibited greater increases in BAL cellularity,

3GADD45A IN MOUSE ACUTE LUNG INJURY

especially in neutrophils [polymorphonuclear cells(PMNs); Fig. 1], as well as increased lung vascularpermeability (BAL protein and albumin; Fig. 2). Inaddition, VILI-exposed GADD45a�/� mice displayedstriking and significant BAL elevations in innate immu-nity cytokine levels (KC, MIP-2, IL-6, IL-1�, and TNF-�)compared with similarly challenged WT mice (Fig. 3).

Histological examination from spontaneously breath-ing WT and GADD45a�/� mice did not detect evidenceof lung edema formation whereas both WT andGADD45a�/� ventilated groups demonstrated “mild”lung edema formation (fewer than 10% alveoli in-volved) with minimal PMN infiltration and withouthyaline membrane formation (Supplemental Fig. S1).

Figure 1. GADD45a�/� mice exposed to high tidal volumemechanical ventilation exhibit a neutrophilic alveolitis. A) Cel-lular content of BAL fluid from spontaneously breathing andVILI-challenged WT and GADD45a�/� mice. VILI inducessignificantly higher cell counts in GADD45a�/� mice comparedwith VILI-challenged WT mice (*P�0.002) or spontaneouslybreathing GADD45a�/� mice (**P�0.02). B) Dramatic in-creases in BAL PMNs in VILI-challenged GADD45a�/� but notin WT mice (*P�0.001, **P�0.001 vs. GADD45a�/� controls).BAL PMNs comprise �70% of BAL cells in VILI-challengedGADD45a�/� mice (n�8/VILI group; n�6/spontaneouslybreathing mice). C) Representative cytospin of BAL cells from aVILI-exposed GADD45a�/� mouse. Inset: cytospin from a VILI-challenged WT mouse (40).

Figure 2. GADD45a�/� mice demonstrate increased ventilator-induced vascular permeability. A) BAL protein, an index ofalveolo-capillary permeability, is significantly elevated in VILI-challenged GADD45a�/� mice compared with VILI-exposed WTmice (*P�0.02) or spontaneously breathing GADD45a�/�controls (**P�0.001). BAL protein content in VILI-challenged WTmice was also significantly elevated relative to spontaneously breathing WT mice (***P�0.02). B) BAL albumin content wassignificantly elevated by VILI challenge in WT and GADD45a�/� mice (**P�0.001, GADD45a�/� mice; ***P�0.05, WT mice).


Susceptibility of GADD45a�/� mice to LPS-inducedvascular leak and injury

We next examined the responses of GADD45a�/� miceexposed to a model of LPS-induced lung injury as we havepreviously described (29–31). LPS-challenged WT andGADD45a�/� mice both sustained dramatic but comparableincreases in BAL cellularity and a prominent neutrophilicalveolitis (Supplemental Table S2). However, LPS-challenged GADD45a�/� mice elaborated significantlygreater BAL protein and lung tissue albumin levels com-pared with WT mice (Fig. 4), indicating that geneticallyengineered GADD45a�/� mice demonstrate increased sus-

ceptibility to increased alveolar capillary permeability in twodistinct models of inflammatory lung injury. Similarly, WTand GADD45a�/� mice both demonstrated markedly ele-vated levels of innate immunity inflammatory cytokinespost-LPS challenge, with significantly higher BAL cytokinelevels in LPS-challenged GADD45a�/� mice (Fig. 4). Histo-logical examination of WT mice exposed to LPS demon-strated mild edema formation and “moderate inflamma-tion” (10–50 inflammatory cells/hpf), where PMNs were thedominant inflammatory cell (Fig. 5). LPS-challengedGADD45a�/� mice exhibited similar mild edema formationbut were graded as “severe inflammation” (�50 inflamma-tory cells/hpf), again with prominent PMNs (Fig. 5).

Figure 3. Elevated BAL inflammatory cytokines in VILI-challenged GADD45a�/� mice. A) BAL IL-6 levels are shown for all 4experimental groups. IL-6 levels were negligible in spontaneously breathing animals of both genotypes. VILI-challenged WTmice elaborated significantly more BAL IL-6 than WT controls (***P�0.03), while VILI-exposed GADD45a�/� micedemonstrated BAL IL-6 levels significantly higher than ventilated WT mice (*P�0.004) or GADD45a�/� controls (**P�0.03).B) BAL cytokine levels in VILI-challenged GADD45a�/� mice are shown relative to VILI-challenged WT mice (fold change aboveWT-VILI response). Ventilated GADD45a�/� mice demonstrated a �2-fold increase over the WT VILI response of each cytokinetested; KC (*P�0.001), MIP-2 (*P�0.001), TNF-� (*P�0.001), IL-1� (*P�0.001), and IL-6 (*P�0.004; n�8/VILI group;n�6/spontaneously breathing group).

α β

Figure 4. LPS-challenged GADD45a�/� mice exhibit greater increases in alveolo-capillary permeability than WT mice.A) Depicted are several indices of lung microvascular permeability (BAL protein content, BAL albumin content, and lung tissuealbumin content) in LPS-challenged GADD45a�/�mice compared with LPS-challenged WT mice (fold increase over LPS-challenged WT mice). LPS evoked significant elevations in each index of permeability compared with water-treated controls.LPS-treated GADD45a�/� mice demonstrated greater levels of BAL protein (*P�0.012) and lung albumin (*P�0.008) relativeto LPS-treated WT mice. BAL albumin content did not differ significantly between LPS-treated GADD45a�/� and WT mice.B) BAL levels of innate immunity cytokines in LPS-treated GADD45a�/� mice were compared with LPS-treated WT mice (foldincrease above WT-LPS levels). GADD45a�/� mice exhibited significantly greater levels of BAL KC (*P�0.017), BAL MIP-2(*P�0.031), and BAL TNF-� (not statistically significant, †P�0.123; n�14–15/LPS group; n�8/water group).


Immunohistochemical localization of GADD45aexpression in murine inflammatory lung injury

Immunohistochemical evaluation of spontaneously breath-ing or vehicle-treated WT mouse lungs revealed GADD45aexpression in airway epithelium and type II pneumocytes(Fig. 5; inflammatory cells were not present in controllungs), which was unchanged by 4 h exposure to VILI(Supplemental Fig. S2). WT mice challenged by LPSdemonstrated increased GADD45a expression in pul-monary endothelium and alveolar epithelium com-pared with control animals but also showed increasedGADD45a expression in pulmonary endothelium andavid GADD45a staining in inflammatory cells present inLPS-challenged lung tissues (Fig. 5). Lungs of GADD45a�/�

mice were stained with GADD45a antibody to confirmspecificity of the antibody and demonstrated negligiblestaining.

GADD45a influences VILI-regulated gene expression

Given the observed susceptibility of GADD45a�/� mice toinflammatory lung injury (evoked by VILI or LPS), wenext identified molecular signatures that characterize therole of GADD45a in the response to VILI-mediated me-chanical stress. Expression profiling of WT lung tissuesidentified significant VILI-mediated gene dysregulation(571 genes) by two-group comparison (SAM softwaresummarized in Supplemental Table S1). In contrast, fewgenes were dysregulated by GADD45a�/� status alone (21genes), whereas VILI-challenged GADD45a�/� mice dem-onstrated the greatest number of dysregulated genes(658 genes; see full gene list: http://phenos.bsd.uchicago.edu/publication/GADD45). Consistent withthe increased inflammatory injury in VILI-challengedGADD45a�/� mice, gene expression analysis revealedsignificant up-regulation of VILI-associated marker genes,

Figure 5. GADD45a�/� mice exhibit increased inflammatory lung histology after LPS exposure. A, B) LPS-challenged WT(A) and GADD45a�/� (B) mice demonstrate diffuse interstitial neutrophilic infiltration, which is not evident in water-challengedlungs (insets: 100). In LPS-challenged WT mice (200), inflammation was graded as “moderate, 10–50 inflammatorycells/hpf,” whereas LPS-treated GADD45a�/� mice (200) exhibited “severe” inflammation with �50 inflammatory cells/hpf.PMNs were the predominant inflammatory cell visualized in both LPS groups, and inflammation was accompanied by mildedema without hyaline membrane formation (n�4/group). C) Immunohistochemical staining of GADD45a (brown; 200) isshown for a water-challenged mouse where GADD45a expression was graded as 2� in the epithelium (indicated by asterisk) and0 to 1� in pulmonary endothelium (indicated by arrows). Minimal inflammatory cells were observed. D) Section from anLPS-challenged WT mouse demonstrates 2� GADD45a expression in the epithelium (asterisk) and 2� in pulmonaryendothelium (thick arrow), with prominent GADD45a immunoreactivity in inflammatory cells (3�; thin arrow; n�2/group).


such as IL-6, IL-1�, CXCL1, and CXCL2, in GADD45a�/�

mice (Fig. 6; ref. 41). Fold changes of CXCL1 and CXCL2by array were validated by real-time qPCR, as were foldchanges of Akt1 and PIK3CD (Fig. 6). IPA revealed signif-icant overlap between signaling pathways dysregulated byVILI-exposed WT mice and VILI-exposed GADD45a�/�

mice (Fig. 7). However, the targeted deletion of GADD45aresulted in prominent increases in signaling pathwaysrelated to the B-cell receptor and axonal guidance signal-ing with reduced dysregulation in integrin signaling. Inaddition, despite significant overlap in the gene listsregulated by VILI-challenged WT and GADD45a�/� mice,direct comparison of the 2 groups demonstrated dramaticup-regulation of several immunoglobulin genes andmembers of the phosphoinositide-3-kinase (PI3K)/Aktsignaling family in GADD45a�/� mice (SupplementalTable S1, gene list 4).

Role of PI3K and Akt in GADD45a-mediatedsusceptibility to lung injury

The B-cell receptor signaling pathway, containing bothimmunoglobulin genes and PI3K pathway genes (PI3Kand Akt), emerged as a highly dysregulated canonicalsignaling pathway in VILI-challenged GADD45a�/�

mice, second only to specific B-cell components thatencode various chains of an autoreactive immunoglob-ulin of the J558 family, (implicated in autoimmunedisorders) (42–45). While the J558 immunoglobulin

chains were dramatically elevated (100- to 1000-foldincreased expression) in GADD45a�/� mice (http://phenos.bsd.uchicago.edu/publication/GADD45), these au-toreactive immunoglobulins were not viewed as likely keyVILI targets, as their expression was modified almostexclusively by GADD45a status and not by VILI exposure.In addition, an immune complex-driven inflammatorylung injury phenotype would be unlikely to manifestwithin the 4 h of VILI exposure. Consequently, we fo-cused on PI3K and Akt, members of the VILI-relevanthighly dysregulated B-cell receptor pathway, as targets inVILI in GADD45a�/� mice.

Initial studies investigated the contribution of thePI3K/Akt pathway to GADD45a�/� VILI susceptibilityby evaluating the abundance of total Akt in mouse lunghomogenates. Marked differences in Akt levels werenoted in lung homogenates from GADD45a�/� micecompared with WT mice, with GADD45a�/� lungsdemonstrating a significant reduction in total Akt pro-tein levels whether comparing lungs from spontane-ously breathing mice (55�21% of total WT Akt expres-sion; P � 0.02) or VILI-challenged mice (38�8% of WTtotal Akt expression; P � 0.04; Fig. 8A).

To investigate the extent of Akt activation in thesemouse models, p-Akt levels were detected by immuno-reactivity to phospho-specific antibody recognizing Ser-473, an Akt site phosphorylated by PI3K. Akt activityvaried by lung injury model in WT mice, with LPSchallenge provoking an �6-fold increase in the ratio of

0

2

4

6

8

10

12

14

CXCL1 CXCL2 AKT1 PIK3CD

Fold

cha

nge

vs W

T-C

ontro

l: qP

CR

-2

3

8

13

18

23

28

CXCL1 CXCL2 IL6 IL1B AKT1 PIK3CD

Fold

cha

nge

vs W

T-C

ontro

l: A

rray

WT-VILIGADD45a-/--VILI

WT-VILIGADD45a-/--VILI*

*

*

*

**

*

*

*

*

*

**

†

BA

Figure 6. Fold changes of ALI biomarker genes (CXCL1, CXCL2, IL6, and IL1ß) and PI3K family members relative to thespontaneously breathing WT controls. The significance of gene dysregulation is determined by two-group comparison using at test. A) Lung mRNA microarray. VILI-exposed GADD45a�/� mice demonstrated significantly increased expression of eachmarker gene [CXCL1 (*P�0.001), CXCL2 (*P�0.001), IL6 (*P�0.005), and IL1ß (*P�0.004)] compared with WT controls.VILI challenged WT mice also demonstrated increased expression of CXCL1 (*P�0.003), CXCL2 (*P�0.006), and IL1ß)(*P�0.009), with IL6 failing to achieve statistical significance (P�0.10). Fold changes for PI3K family members Akt1 and PIK3CDare also shown. Both VILI groups showed increased expression of PIK3CD (*P�0.05 WT, *P�0.03 GADD45a�/�), whereasneither VILI group demonstrated a change in Akt1 expression. B) Real-time qPCR validation of selected genes from lunghomogenates. VILI-challenged WT and GADD45a�/�mice demonstrated increased expression of CXCL1 (*P�0.02 WT;*P�0.002 GADD45a�/�) and CXCL2 (*P�0.03 WT; *P�0.03 GADD45a�/�), whereas neither VILI group demonstrated asignificant change in Akt1 expression. The change in PIK3CD expression in VILI-exposed GADD45a�/�mice did not achievestatistical significance (†P�0.21).


activated (phosphorylated) Akt to total Akt (p-Akt/Akt), whereas VILI evoked only an �2-fold increase(Fig. 8B). In GADD45a�/� lungs, despite the dimin-ished total Akt abundance, LPS challenge still pro-duced an approximate 6-fold increase in p-Akt/Akt

levels, whereas it was difficult to detect any increase inAkt activation in VILI-challenged GADD45a�/� lungs(Fig. 8B).

To address a mechanism for the observed reductionin Akt levels in GADD45a�/� mice, we explored alter-

Figure 7. Expression profiling reveals a signatureof increased B-cell receptor signaling dysregula-tion in VILI-challenged GADD45a�/� mice. IPAreveals canonical pathways enriched in dysregu-lated genes in VILI-challenged groups relative tospontaneously breathing WT control lungs.Dashed line represents threshold level for signifi-cance (1.3, negative log of P value 0.05 by Fisher’sexact test). The B-cell receptor signaling pathwayexhibited the greatest dysregulation in the VILI-exposed GADD45a�/� group, with componentsthat include members of the PI3K/Akt pathway.

Figure 8. GADD45a�/� mice demonstrate decreasedlung expression of Akt protein. A) Inset: representativeimmunoblot of spontaneously breathing and VILI-challenged lung homogenates probed for Akt withdramatic decreases in total Akt protein expression inGADD45a�/� lungs. Bar graph: pooled Akt densitomet-ric quantitation expressed as standardized average grayvalues, normalized for protein loading and standard-ized to WT control lungs. GADD45a�/� lungs exhibitreduced Akt expression (*P�0.02 for spontaneouslybreathing GADD45a�/� mice vs. WT mice, **P�0.05for VILI- or LPS-challenged GADD45a�/� lungs com-pared with WT mice). B) Relative Akt activation isdepicted as the ratio of phosphorylated Akt (Ser-473)/total Akt. Inset: lung homogenates of spontaneously

breathing, VILI-challenged, and LPS-challenged mice probed for Akt, p-Akt, and �-actin. In WT mice, VILI induced an�2-fold increase in Akt activation while LPS challenge provoked an �6-fold increase (**P�0.04). Despite the reductionof total and p-Akt in GADD45a�/� lungs, LPS-challenged mice demonstrated increased Akt phosphorylation similar toLPS-challenged WT mice (*P�0.04). VILI challenge, in contrast, did not significantly increase the p-Akt/Akt ratio inGADD45a�/� mice. C) Pooled results (n�3 mice/group) of relative ratio of immunoreactive ubiquitin and Akt (anti-Akt)obtained from Western blots of immunoprecipitates from WT and GADD45a�/� mouse lung homogenates (*P�0.01).Inset: Western blots for ubiquitin and Akt in lung homogenates from 3 WT and 3 GADD45a�/� mice.


ations in Akt transcriptional regulation but failed toobserve differential regulation of Akt gene expression(Affymetrix array, qPCR) between GADD45a�/� andWT lungs (Fig. 6). As these results suggested thatreduced Akt levels may reflect post-translational modi-fication and protein degradation, we next generatedAkt immunoprecipitates, from WT and GADD45a�/�

mouse lung homogenates followed by Western blotting,to detect ubiquitin immunoreactivity consistent withAkt protein ubiquitination. Figure 8C depicts the mark-edly increased levels of ubiquitinated Akt in lunghomogenates from GADD45a�/� but not WT mice,strongly suggesting an increase in Akt protein degrada-tion in GADD45a�/� mice.

GADD45a involvement in LPS-induced EC barrierregulation via an Akt-dependent pathway

Given the observed significant increases in LPS- andVILI-induced vascular permeability and decreased Aktprotein abundance in GADD45a�/� mice, we nextinvestigated the mechanistic contribution of Akt signal-ing to increased microvascular permeability duringlung inflammation. Initial experiments examined theeffects of GADD45a depletion (siRNA) on human lungEC total Akt levels. Similar to the reduced Akt levels inlung homogenates derived from GADD45a�/� mice,in vitro GADD45a depletion resulted in decreased levelsof total Akt (20�9%) compared with control ECs(scramble sequence siRNA; Fig. 9A, B). In addition,Akt activation increased after LPS stimulus in bothcontrol and GADD45a-silenced endothelium afterLPS, although the total Akt abundance was decreasedin GADD45a-silenced cells. We next utilized anin vitro model of vascular barrier regulation withhuman lung endothelium grown on gold microelec-trodes with continuous measurement of TER acrosscell monolayers. Despite GADD45a depletion, unstimu-lated EC monolayers demonstrated normal basal bar-rier integrity (Fig. 9C, D). Control endothelial mono-layers exposed to exogenous LPS (1 mcg/ml) exhibitedmodest declines in TER values peaking at 6 h; however,LPS challenge of endothelium with reduced GADD45aexpression exhibited greater declines in TER values aswell as slower and incomplete barrier recovery, indicat-ing augmented vascular barrier dysfunction. A fall inTER values of a similar magnitude was observed whenECs were depleted of Akt1 (siRNA) and subsequentlychallenged with LPS (Fig. 9B, C). The simultaneousaddition of siRNA oligonucleotides directed againstboth GADD45a and Akt did not result in further LPS-induced TER decrements compared with either siRNAalone, indicating the absence of additive or synergisticeffects. However, overexpression of constitutively activeAkt1/PKB� mitigated the LPS-induced declines in TER,even in the presence of GADD45a depletion (Fig.9B–D). Transfection of the active Akt1/PKB� transgenedid not alter TER in the control siRNA-treated cells.Together these results are consistent with a key role forAkt in GADD45a�/� susceptibility to inflammation-

induced lung vascular barrier dysfunction and supportthe hypothesis that the increased susceptibility to in-flammatory lung injury and barrier dysfunction inGADD45a�/� mice involves the critical loss of Akt andAkt-related signaling components.

DISCUSSION

Mechanical ventilation, a life-saving intervention for crit-ically ill patients with respiratory failure, exhibits thepotential to contribute to inflammatory lung injury ashighlighted by the decreased mortality in acute respira-tory distress syndrome (ARDS) patients treated with alow tidal volume ventilation strategy (46). Furthermore,patients without ALI at the initiation of mechanical ven-tilation are at risk to progress to fully develop the syn-drome, particularly if higher tidal volume ventilationstrategies are utilized (4, 5, 47). Thus, a deeper under-standing of the genetic factors involved in susceptibility toor severity of ALI/ARDS is key to our understanding ofALI pathogenesis and to the design of novel therapies(12, 13, 48–51). To our knowledge, our prior genomicstudies (14, 52) were the first to suggest GADD45a as anovel ALI/VILI candidate gene, a suggestion stronglysupported by our current studies indicating GADD45a as anovel therapeutic target in inflammatory lung injury. Weused similar genomic approaches to successfully identifypre-B-cell colony enhancing factor (PBEF) as a novelALI/VILI candidate gene and biomarker (33). Our re-cent studies validated PBEF as a potential VILI therapeu-tic target (38) and, interestingly, noted increasedGADD45a expression in VILI- and PBEF-challenged micewith GADD45a expression strongly correlated with VILIseverity (38). In the present study, GADD45a expressionwas also increased 1.98-fold in VILI-challenged WT mice(data not shown), a result consistent with prior VILImodels (14, 52). While our immunohistochemical stain-ing of VILI-challenged lungs did not demonstrate in-creased GADD45a protein expression (Fig. 5), we feel thismay reflect the time course of our VILI challenge; lungswere harvested immediately after 4 h ventilation, whichmay have been insufficient to see changes in GADD45atranslation. More recently, GADD45a expression wasnoted to be increased in the peripheral whole blood ofpatients during the acute stages of ARDS, declining insurviving patients during the recovery phase (53).

Although there is limited information as to the role orfunction(s) of GADD45a in lung homeostasis and disease,GADD45a is well recognized as a participant in cell cycleregulation, DNA repair, maintenance of genomic stability,as well as in regulation of p53-dependent and p53-inde-pendent apoptosis (18, 23, 24, 54). In addition, GADD45ainteracts with innate and adaptive immune systems toregulate T-cell differentiation and proliferation (26). Tolink GADD45a availability and function to inflammatorylung injury, we exposed mice with targeted deletion ofGADD45a to two models of lung inflammation and notedthat GADD45a�/� mice exhibited increased susceptibilityto inflammatory lung injury after either LPS or excessive


mechanical stress (VILI), the first characterization of alung injury phenotype in GADD45a�/� mice. Prior re-ports (55) describing exposure of GADD45a�/� mice tohyperoxic oxidative stress revealed a similar phenotype asWT mice, highlighting the distinctive mechanisms under-lying VILI-, LPS-, and hyperoxia-induced lung injury. Inaddition, whereas GADD45a expression in hyperoxia-chal-lenged mice was confined to the lung epithelium (55), weobserved GADD45a expression in epithelial cells and ECsafter LPS and VILI challenge, with further increases inpulmonary endothelial expression 18 h after LPS. In bothour LPS and VILI models of lung injury, GADD45a�/�

mice exhibited both increased neutrophilic inflammationand increased alveolo-capillary permeability. We feelthese results strongly implicate a role for GADD45a inalveolar and microvascular barrier regulation during LPS-induced inflammatory lung injury and VILI.

To better dissect the role of GADD45a in inflamma-tory processes such as permeability, apoptosis, andleukocyte infiltration, we used genomic approachesthat demonstrated significantly altered gene expressionresponses in GADD45a�/� mice after ventilation. Re-markably, despite only modest physiological aberra-

tions in VILI-challenged WT mice, 4 h of high tidalvolume ventilation triggered dramatic alterations ingene expression with extensive up-regulation in a num-ber of ALI candidate genes (CXCL1, CXCL2, IL-6, andIL-1�; refs. 41, 52). The molecular signature of venti-lated GADD45a�/� animals was qualitatively similar tothat of ventilated WT mice but quantitatively amplified,with increased expression of each putative ALI markergene (CXCL1, CXCL2, IL-6, and IL-1�) at levels thatwere at least two times greater than VILI-exposed WTmice (Fig. 6), results that were verified by qPCR and byBAL protein levels (Fig. 3B). These results appear to beconsistent with the potential utility of these chemokinesas early biomarkers of ALI or VILI (41).

We used IPA to identify biological pathways relevantto the mechanistic consequences of GADD45a deletionin the context of inflammatory lung injury. Severalsignaling pathways (ERK/MAPK, acute-phase response,IL-6, and axonal guidance; Fig. 7) were exclusivelydysregulated in VILI-challenged GADD45a�/� miceand may yield further insights into the role ofGADD45a’s in resisting lung injury, which we hope toinvestigate further. It was also interesting to note that

Figure 9. Expression of a constitutively active Akt transgene reversesthe effects of GADD45a depletion in LPS-induced EC barrier dysfunc-tion. A) HPAECs were treated with scramble siRNA, GADD45a siRNA,or AKT1 siRNA and immunoblotted for GADD45a or AKT, resultingin effective knockdown of GADD45a and AKT expression. Bottombands demonstrate effective transfection of a Myc-His tagged consti-tutively active AKT1/PKB� transgene in HPAECs. B) Lung endothe-lium with GADD45a depletion (siRNA) or control siRNA was treatedwith LPS or vehicle and total Akt levels determined at 4 h. Depletionof GADD45a reduced Akt expression, though did not abolish Aktphosphorylation. C, D) Human ECs were transfected with siRNAs todeplete either GADD45a or AKT1 or both, or with an Akt/PKB cDNAto increase expression of constitutively active Akt. Treated cells werethen assessed for TER after vehicle or LPS (arrow). Data are means �se for 3–5 experiments/condition. C) TER tracings in which GADD45adepletion exaggerates LPS-induced declines in TER (gray circles).

However, ECs depleted of GADD45a (siRNA) but with Akt overexpression (black solid line) failed to demonstrate theGADD45a siRNA-mediated exacerbation of LPS-induced declines in TER values, but instead mirrored the resistance ofscramble siRNA-LPS treated cells, as shown in D. D) Normalized TER values at 6 h following LPS in endotheliumtransfected with either scrambled siRNA or GADD45a siRNA and 1 of 4 treatments: vehicle, LPS, AKT siRNA and LPS, orAKT overexpression and LPS. GADD45a depletion increased LPS-induced permeability at 6 h (*P�0.006 vs. siScr/LPS;**P�0.001 vs. siG45/vehicle). When Akt1 was depleted or cells were transfected with both Akt1 and GADD45a siRNAsconcurrently, there were no differences noted between depletion of GADD45a alone, Akt1 alone, or the two incombination. Overexpression of c/a Akt1 abolished the effect of GADD45a depletion on LPS-induced barrier function(***P�0.043 vs. siG45/LPS), while it did not alter the resistance of scramble siRNA LPS-treated cells. Dashed lineemphasizes the LPS effect on scramble siRNA-treated cells.


VILI-challenged GADD45a�/� mice displayed only amodest increase in dysregulation of the NF-�B pathway,despite the marked increase in TNF-dependent inflam-matory cytokines compared with ventilated WT ani-mals. Examination of the pathway members annotatedin the NF-�B pathway by the IPA software revealed thatonly IL-1� and IL1R2 are included in this pathway,whereas CXCL1 and CXCL2 are not.

We were struck by the dramatic dysregulation occur-ring in the B-cell receptor signaling pathway of venti-lated GADD45a�/� mice, with marked up-regulation ofgenes involved in the PI3K/Akt pathway. Included inthis pathway are upregulated genes encoding the cata-lytic domain (p110 ) of PI3K (PIK3CD), the PI3Kregulatory region, and a PI3K adaptor protein (PIK3APor B-cell adaptor protein), which exhibited �20-foldincreases in gene expression (http://phenos.bsd.uchica-go.edu/publication/GADD45). PIK3AP is the tyrosinekinase substrate that bridges B-cell antigen receptor-asso-ciated kinases to PI3K activation (56). Activity of the PI3Kp110 catalytic domain (PIK3CD) is required for efficientneutrophil trafficking to cytokine-stimulated endothe-lium (57) and is the dominant PI3K isoform dictatingcytokine secretion by natural killer cells (58), a possiblelink between innate and adaptive immunity (59–61).

A key target for the enzymatic activity of PI3K is theprosurvival signaling protein known as Akt, with bothPI3 and Akt implicated in regulating responses to VILI.Recently, mice deficient in PI3K� were found to bephysiologically and morphologically protected fromex vivo VILI (62). Pharmacological inhibition of PI3Kwas also protective in an ex vivo murine VILI model,whereas lung injury was potentiated in animals pre-treated with an Akt inhibitor (63). We sought todetermine whether alterations in the PI3K pathwaywere evident in VILI-challenged GADD45a�/� mice andnoted striking reductions in total Akt levels inGADD45a�/� mice, an observation suggesting that al-terations in Akt might underlie the GADD45a�/�-in-duced susceptibility to VILI. Whereas several mecha-nisms exist that potentially couple the loss of GADD45ato dramatic reductions in Akt protein levels, bothmicroarray analysis and qPCR failed to identify differ-ences in Akt gene expression between groups, signify-ing that the GADD45a�/� effect is unlikely to reflectaltered Akt transcriptional regulation. Further investi-gation into the potential role of GADD45a in promotingAkt protein stability indicated that GADD45a depletionis strongly associated with increases in Akt ubiquitina-tion, indicating that GADD45a modulates protein deg-radation pathways that affect Akt stability. The potentialfor GADD45a to modify the availability of Akt chaper-one proteins (e.g., heat shock protein 90) or otherposttranscriptional modifiers such as microRNAs is alsocurrently being examined.

Our observations complement prior reports of poten-tiation of VILI in the presence of Akt inhibition as well asincreasing recognition of the role of PI3K and Akt signal-ing in lung vascular barrier regulation (64–66). Weexamined whether GADD45a depletion affected human

lung EC barrier function as suggested by in vivo results inLPS- and VILI-challenged GADD45a�/� mice. We deter-mined exaggerated declines in TER values after LPStreatment in human ECs treated with siRNA directedagainst GADD45a, suggesting a direct and significant rolefor GADD45a in EC barrier regulation after inflammatorystress. Interestingly, reductions in Akt1 expression (siRNA),followed by LPS challenges, produced TER values thatessentially mirrored EC barrier responses after GADD45adepletion. Furthermore, reductions in the expression ofboth GADD45a and Akt failed to produce an additive effect,suggesting a common mechanism. Importantly, increasedexpression of Akt1 attenuated LPS-induced declines in en-dothelial monolayer integrity associated with GADD45a de-pletion without altering the TER of control cell monolayers.These data indicate that GADD45a regulates Akt availabilityvia control of ubiquitination and protein degradation andthat Akt is a key molecular target in mechanical stress- andinflammation-induced lung injury. As we have previouslyimplicated Akt in the promotion of EC barrier functionvia transactivation of the sphingosine-1-phosphate recep-tor (S1P1) resulting in Rac-dependent increases in corticalactin (64), we speculate that one mechanism by whichGADD45a may contribute to vascular barrier functionmay be via stabilization of Akt interaction with S1P1.Further investigation to determine the precise mecha-nisms by which GADD45a interacts with Akt may yield anenhanced understanding of the significance in Akt in thepathogenesis of VILI. In summary, this work stronglyvalidates the candidate gene approach in identifyingnovel candidate genes such as GADD45a and elucidatinggene involvement in inflammatory lung injury. Further-more, this work supports a direct role for GADD45a in theregulation of lung microvascular barrier function. Futureinvestigations should consider strategies designed to in-crease GADD45a and Akt lung expression as a platform forfuture targeted genotype-based therapy in critically illpatients at risk for ALI and VILI.

We are extremely grateful to Dr. Michael O’Reilly (Univer-sity of Rochester, Rochester, NY, USA) and Dr. Albert For-nace (Brigham and Women’s Hospital, Boston, MA, USA) forgenerously providing the GADD45a�/� mice. We acknowl-edge the invaluable contributions of Dr. Shwu; Fan Ma, forher genomic expertise and other important insights. We alsoacknowledge Lakshmi Natarajan, Darrel Sparkman, AnnetteWesterberg, and Nicholas Shank, for their outstanding exper-tise and invaluable assistance. This work was supported byU.S. National Institutes of Health grants PPG HL-58064, F32HL-088858-01, R01 HL-088144, K08 HL-077134-01, and K22LMV-008308.

REFERENCES

1. Ware, L. B., and Matthay, M. A. (2000) The acute respiratorydistress syndrome. N. Engl. J. Med. 342, 1334–1349

2. Rubenfeld, G. D., Caldwell, E., Peabody, E., Weaver, J., Martin,D. P., Neff, M., Stern, E. J., and Hudson, L. D. (2005) Incidenceand outcomes of acute lung injury. N. Engl. J. Med. 353,1685–1693

3. Hudson, L. D., Milberg, J. A., Anardi, D., and Maunder, R. J.(1995) Clinical risks for development of the acute respiratorydistress syndrome. Am. J. Respir. Crit. Care Med. 151, 293–301


4. Gajic, O., Dara, S. I., Mendez, J. L., Adesanya, A. O., Festic, E.,Caples, S. M., Rana, R., St. Sauver, J. L., Lymp, J. F., Afessa, B.,and Hubmayr, R. D. (2004) Ventilator-associated lung injury inpatients without acute lung injury at the onset of mechanicalventilation. Crit. Care Med. 32, 1817–1824

5. Gajic, O., Frutos-Vivar, F., Esteban, A., Hubmayr, R. D., andAnzueto, A. (2005) Ventilator settings as a risk factor for acuterespiratory distress syndrome in mechanically ventilated pa-tients. Intensive Care Med. 31, 922–926

6. Webb, H. H., and Tierney, D. F. (1974) Experimental pulmo-nary edema due to intermittent positive pressure ventilationwith high inflation pressures. Protection by positive end-expira-tory pressure. Am. Rev. Respir. Dis. 110, 556–565

7. Dreyfuss, D., Basset, G., Soler, P., and Saumon, G. (1985)Intermittent positive-pressure hyperventilation with high infla-tion pressures produces pulmonary microvascular injury in rats.Am. Rev. Respir. Dis. 132, 880–884

8. Dreyfuss, D., and Saumon, G. (1993) Role of tidal volume, FRC,and end-inspiratory volume in the development of pulmonaryedema following mechanical ventilation. Am. Rev. Respir. Dis.148, 1194–1203

9. Corbridge, T. C., Wood, L. D., Crawford, G. P., Chudoba, M. J.,Yanos, J., and Sznajder, J. I. (1990) Adverse effects of large tidalvolume and low PEEP in canine acid aspiration. Am. Rev. Respir.Dis. 142, 311–315

10. Birukov, K. G., Jacobson, J. R., Flores, A. A., Ye, S. Q., Birukova,A. A., Verin, A. D., and Garcia, J. G. (2003) Magnitude-dependent regulation of pulmonary endothelial cell barrierfunction by cyclic stretch. Am. J. Physiol. Lung Cell. Mol. Physiol.285, L785–L797

11. Pepe, P. E., Potkin, R. T., Reus, D. H., Hudson, L. D., andCarrico, C. J. (1982) Clinical predictors of the adult respiratorydistress syndrome. Am. J. Surg. 144, 124–130

12. Meyer, N. J., and Garcia, J. G. (2007) Wading into the genomicpool to unravel acute lung injury genetics. Proc. Am. Thorac. Soc.4, 69–76

13. Moss, M., and Mannino, D. M. (2002) Race and gender differ-ences in acute respiratory distress syndrome deaths in theUnited States: an analysis of multiple-cause mortality data(1979–1996). Crit. Care Med. 30, 1679–1685

14. Grigoryev, D. N., Ma, S. F., Irizarry, R. A., Ye, S. Q., Quacken-bush, J., and Garcia, J. G. (2004) Orthologous gene-expressionprofiling in multi-species models: search for candidate genes.Genome Biol. 5, R34

15. Nonas, S. A., Finigan, J. H., Gao, L., and Garcia, J. G. (2005)Functional genomic insights into acute lung injury: role ofventilators and mechanical stress. Proc. Am. Thorac. Soc. 2,188–194

16. O’Reilly, M. A., Staversky, R. J., Watkins, R. H., Maniscalco,W. M., and Keng, P. C. (2000) p53-independent induction ofGADD45 and GADD153 in mouse lungs exposed to hyperoxia.Am. J. Physiol. Lung Cell. Mol. Physiol. 278, L552–L559

17. Altemeier, W. A., Matute-Bello, G., Gharib, S. A., Glenny, R. W.,Martin, T. R., and Liles, W. C. (2005) Modulation of lipopo-lysaccharide-induced gene transcription and promotion of lunginjury by mechanical ventilation. J. Immunol. 175, 3369–3376

18. Hollander, M. C., Alamo, I., Jackman, J., Wang, M. G., McBride,O. W., and Fornace, A. J., Jr. (1993) Analysis of the mammaliangadd45 gene and its response to DNA damage. J. Biol. Chem. 268,24385–24393

19. Fornace, A. J., Jr., Alamo, I., Jr., and Hollander, M. C. (1988)DNA damage-inducible transcripts in mammalian cells. Proc.Natl. Acad. Sci. U. S. A. 85, 8800–8804

20. Hollander, M. C., Philburn, R. T., Patterson, A. D., Wyatt, M. A.,and Fornace, A. J., Jr. (2005) Genomic instability inGadd45a�/� cells is coupled with S-phase checkpoint defects.Cell Cycle 4, 704–709

21. Zerbini, L. F., Wang, Y., Czibere, A., Correa, R. G., Cho, J. Y.,Ijiri, K., Wei, W., Joseph, M., Gu, X., Grall, F., Goldring, M. B.,Zhou, J. R., and Libermann, T. A. (2004) NF-kappa B-mediatedrepression of growth arrest- and DNA-damage-inducible pro-teins 45alpha and gamma is essential for cancer cell survival.Proc. Natl. Acad. Sci. U. S. A. 101, 13618–13623

22. Befort, K., Karchewski, L., Lanoue, C., and Woolf, C. J. (2003)Selective up-regulation of the growth arrest DNA damage-inducible gene Gadd45 alpha in sensory and motor neuronsafter peripheral nerve injury. Eur. J. Neurosci. 18, 911–922

23. Sheikh, M. S., Hollander, M. C., and Fornance, A. J., Jr. (2000)Role of Gadd45 in apoptosis. Biochem. Pharmacol. 59, 43–45

24. Barreto, G., Schafer, A., Marhold, J., Stach, D., Swaminathan,S. K., Handa, V., Doderlein, G., Maltry, N., Wu, W., Lyko, F.,and Niehrs, C. (2007) Gadd45a promotes epigenetic geneactivation by repair-mediated DNA demethylation. Nature445, 671– 675

25. Salvador, J. M., Mittelstadt, P. R., Guszczynski, T., Copeland,T. D., Yamaguchi, H., Appella, E., Fornace, A. J., Jr., andAshwell, J. D. (2005) Alternative p38 activation pathway medi-ated by T cell receptor-proximal tyrosine kinases. Nat. Immunol.6, 390–395

26. Salvador, J. M., Mittelstadt, P. R., Belova, G. I., Fornace, A. J., Jr.,and Ashwell, J. D. (2005) The autoimmune suppressorGadd45alpha inhibits the T cell alternative p38 activationpathway. Nat. Immunol. 6, 396–402

27. Garcia, J. G., Liu, F., Verin, A. D., Birukova, A., Dechert, M. A.,Gerthoffer, W. T., Bamburg, J. R., and English, D. (2001)Sphingosine 1-phosphate promotes endothelial cell barrier in-tegrity by Edg-dependent cytoskeletal rearrangement. J. Clin.Invest. 108, 689–701

28. Hollander, M. C., Sheikh, M. S., Bulavin, D. V., Lundgren, K.,Augeri-Henmueller, L., Shehee, R., Molinaro, T. A., Kim, K. E.,Tolosa, E., Ashwell, J. D., Rosenberg, M. P., Zhan, Q., Fernan-dez-Salguero, P. M., Morgan, W. F., Deng, C. X., and Fornace,A. J., Jr. (1999) Genomic instability in Gadd45a-deficient mice.Nat. Genet. 23, 176–184

29. McVerry, B. J., Peng, X., Hassoun, P. M., Sammani, S., Simon,B. A., and Garcia, J. G. (2004) Sphingosine 1-phosphate reducesvascular leak in murine and canine models of acute lung injury.Am. J. Respir. Crit. Care Med. 170, 987–993

30. Peng, X., Hassoun, P. M., Sammani, S., McVerry, B. J., Burne,M. J., Rabb, H., Pearse, D., Tuder, R. M., and Garcia, J. G. (2004)Protective effects of sphingosine 1-phosphate in murine endo-toxin-induced inflammatory lung injury. Am. J. Respir. Crit. CareMed. 169, 1245–1251

31. Jacobson, J. R., Barnard, J. W., Grigoryev, D. N., Ma, S. F.,Tuder, R. M., and Garcia, J. G. (2005) Simvastatin attenuatesvascular leak and inflammation in murine inflammatory lunginjury. Am. J. Physiol. Lung Cell. Mol. Physiol. 288, L1026 –L1032

32. Young, S. H., Antonini, J. M., Roberts, J. R., Erdely, A. D., andZeidler-Erdely, P. C. (2008) Performance evaluation of cytomet-ric bead assays for the measurement of lung cytokines in tworodent models. J. Immunol. Methods 331, 59–68

33. Ye, S. Q., Simon, B. A., Maloney, J. P., Zambelli-Weiner, A., Gao,L., Grant, A., Easley, R. B., McVerry, B. J., Tuder, R. M.,Standiford, T., Brower, R. G., Barnes, K. C., and Garcia, J. G.(2005) Pre-B-cell colony-enhancing factor as a potential novelbiomarker in acute lung injury. Am. J. Respir. Crit. Care Med. 171,361–370

34. Nonas, S. A., Moreno-Vinasco, L., Ma, S. F., Jacobson, J. R.,Desai, A. A., Dudek, S. M., Flores, C., Hassoun, P. M., Sam, L.,Ye, S. Q., Moitra, J., Barnard, J., Grigoryev, D. N., Lussier, Y. A.,and Garcia, J. G. (2007) Use of consomic rats for genomicinsights into ventilator-associated lung injury. Am. J. Physiol.Lung Cell. Mol. Physiol. 293, L292–L302

35. Team, R. D. C. (2006) R: A language and environment forstatistical computing. R Foundation for Statistical Computing,Vienna, Austria

36. Wu, Z., Irizarry, R. A., Gentleman, R., Martinez-Murillo, F.,and Spencer, F. (2004) A model-based background adjust-ment for oligonucleotide expression arrays. J. Am. Stat. Assoc.99, 909 –917

37. Kittleson, M. M., Minhas, K. M., Irizarry, R. A., Ye, S. Q.,Edness, G., Breton, E., Conte, J. V., Tomaselli, G., Garcia,J. G., and Hare, J. M. (2005) Gene expression analysis ofischemic and nonischemic cardiomyopathy: shared and dis-tinct genes in the development of heart failure. Physiol.Genomics 21, 299 –307

38. Hong, S. B., Huang, Y., Moreno-Vinasco, L., Sammani, S.,Moitra, J., Barnard, J. W., Ma, S. F., Mirzapoiazova, T., Evensoki,C., Reeves, R. R., Chiang, E. T., Lang, G. D., Husain, A. N.,Dudek, S. M., Jacobson, J. R., Ye, S. Q., Lussier, Y. A., and Garcia,J. G. (2008) Essential role of pre B cell colony enhancing factor(PBEF) in ventilator-induced lung injury. Am. J. Respir. Crit. CareMed. 178, 607–617


39. Shi, S., Verin, A. D., Schaphorst, K. L., Gilbert-McClain, L. I.,Patterson, C. E., Irwin, R. P., Natarajan, V., and Garcia, J. G.(1998) Role of tyrosine phosphorylation in thrombin-inducedendothelial cell contraction and barrier function. Endothelium 6,153–171

40. Garcia, J. G., Schaphorst, K. L., Shi, S., Verin, A. D., Hart, C. M.,Callahan, K. S., and Patterson, C. E. (1997) Mechanisms ofionomycin-induced endothelial cell barrier dysfunction. Am. J.Physiol. Lung Cell. Mol. Physiol. 273, L172–L184

41. Frank, J. A., Parsons, P. E., and Matthay, M. A. (2006) Pathoge-netic significance of biological markers of ventilator-associatedlung injury in experimental and clinical studies. Chest 130,1906–1914

42. Monestier, M., Bonin, B., Migliorini, P., Dang, H., Datta, S.,Kuppers, R., Rose, N., Maurer, P., Talal, N., and Bona, C. (1987)Autoantibodies of various specificities encoded by genes fromthe VH J558 family bind to foreign antigens and share idiotopesof antibodies specific for self and foreign antigens. J. Exp. Med.166, 1109–1124

43. Monestier, M., Manheimer-Lory, A., Bellon, B., Painter, C.,Dangt, H., Talal, N., Zanetti, M., Schwartz, R., Pisetsky, D.,Kuppers, R., Rose, N., Brochier, J., Klareskog, L., Holmdahl, R.,Erlanger, B., Alt, F., and Bona, C. (1986) Shared idiotypes andrestricted immunoglobulin variable region heavy chain genescharacterize murine autoantibodies of various specificities.J. Clin. Invest. 78, 753–759

44. Eilat, D., Hochberg, M., Tron, F., Jacob, L., and Bach, J. F.(1989) The VH gene sequences of anti-DNA antibodies in twodifferent strains of lupus-prone mice are highly related. Eur.J. Immunol. 19, 1241–1246

45. Aguado, M. T., Balderas, R. S., Rubin, R. L., Duchosal, M. A.,Kofler, R., Birshtein, B. K., Secher, D. S., Dixon, F. J., andTheofilopoulos, A. N. (1987) Specificity and molecular charac-teristics of monoclonal IgM rheumatoid factors from arthriticand non-arthritic mice. J. Immunol. 139, 1080–1087

46. Anonymous. (2000) Ventilation with lower tidal volumes ascompared with traditional tidal volumes for acute lung injuryand the acute respiratory distress syndrome. The Acute Respi-ratory Distress Syndrome Network. N. Engl. J. Med. 342, 1301–1308

47. Mascia, L., Zavala, E., Bosma, K., Pasero, D., Decaroli, D.,Andrews, P., Isnardi, D., Davi, A., Arguis, M. J., Berardino, M.,and Ducati, A. (2007) High tidal volume is associated with thedevelopment of acute lung injury after severe brain injury: aninternational observational study. Crit. Care Med. 35, 1815–1820

48. Gong, M. N., Zhou, W., Williams, P. L., Thompson, B. T.,Pothier, L., Boyce, P., and Christiani, D. C. (2005) �308GA andTNFB polymorphisms in acute respiratory distress syndrome.Eur. Respir. J. 26, 382–389

49. Bajwa, E. K., Yu, C. J., Gong, M. N., Thompson, B. T., andChristiani, D. C. (2006) PBEF gene polymorphisms influencethe risk of developing ARDS. Proc. Am. Thorac. Soc. 3, A272

50. Gong, M. N., Wei, Z., Xu, L. L., Miller, D. P., Thompson, B. T.,and Christiani, D. C. (2004) Polymorphism in the surfactantprotein-B gene, gender, and the risk of direct pulmonary injuryand ARDS. Chest 125, 203–211

51. Marshall, R. P., Webb, S., Bellingan, G. J., Montgomery, H. E.,Chaudhari, B., McAnulty, R. J., Humphries, S. E., Hill, M. R.,and Laurent, G. J. (2002) Angiotensin converting enzymeinsertion/deletion polymorphism is associated with susceptibil-ity and outcome in acute respiratory distress syndrome. Am. J.Respir. Crit. Care Med. 166, 646–650

52. Ma, S. F., Grigoryev, D. N., Taylor, A. D., Nonas, S., Sammani,S., Ye, S. Q., and Garcia, J. G. (2005) Bioinformatic identifi-cation of novel early stress response genes in rodent models

of lung injury. Am. J. Physiol. Lung Cell. Mol. Physiol. 289,L468 –L477

53. Wang, Z., Beach, D., Su, L., Zhai, R., and Christiani, D. C.(2008) A genome-wide expression analysis in blood identifiespre-elafin as a biomarker in ARDS. Am. J. Respir. Cell Mol. Biol.38, 724–732

54. Jung, H. J., Kim, E. H., Mun, J. Y., Park, S., Smith, M. L., Han,S. S., and Seo, Y. R. (2007) Base excision DNA repair defect inGadd45a-deficient cells. Oncogene 26, 7517–7525

55. Roper, J. M., Gehen, S. C., Staversky, R. J., Hollander, M. C.,Fornace, A. J., Jr., and O’Reilly, M. A. (2005) Loss of Gadd45adoes not modify the pulmonary response to oxidative stress.Am. J. Physiol. Lung Cell Mol. Physiol. 288, L663–L671

56. Okada, T., Maeda, A., Iwamatsu, A., Gotoh, K., and Kurosaki, T.(2000) BCAP: the tyrosine kinase substrate that connects B cellreceptor to phosphoinositide 3-kinase activation. Immunity 13,817–827

57. Puri, K. D., Doggett, T. A., Huang, C. Y., Douangpanya, J.,Hayflick, J. S., Turner, M., Penninger, J., and Diacovo, T. G.(2005) The role of endothelial PI3Kgamma activity in neutro-phil trafficking. Blood 106, 150–157

58. Kim, N., Saudemont, A., Webb, L., Camps, M., Ruckle, T.,Hirsch, E., Turner, M., and Colucci, F. (2007) The p110deltacatalytic isoform of PI3K is a key player in NK-cell developmentand cytokine secretion. Blood 110, 3202–3208

59. Hebeis, B. J., Vigorito, E., and Turner, M. (2004) The p110deltasubunit of phosphoinositide 3-kinase is required for the lipo-polysaccharide response of mouse B cells. Biochem. Soc. Trans.32, 789–791

60. Okkenhaug, K., Ali, K., and Vanhaesebroeck, B. (2007) Antigenreceptor signalling: a distinctive role for the p110delta isoformof PI3K. Trends Immunol. 28, 80–87

61. Okkenhaug, K., Patton, D. T., Bilancio, A., Garcon, F., Rowan,W. C., and Vanhaesebroeck, B. (2006) The p110delta isoform ofphosphoinositide 3-kinase controls clonal expansion and differ-entiation of Th cells. J. Immunol. 177, 5122–5128

62. Lionetti, V., Lisi, A., Patrucco, E., De Giuli, P., Milazzo, M. G.,Ceci, S., Wymann, M., Lena, A., Gremigni, V., Fanelli, V.,Hirsch, E., and Ranieri, V. M. (2006) Lack of phosphoinositide3-kinase-gamma attenuates ventilator-induced lung injury. Crit.Care Med. 34, 134–141

63. Miyahara, T., Hamanaka, K., Weber, D. S., Drake, D. A.,Anghelescu, M., and Parker, J. C. (2007) Phosphoinositide3-kinase, Src, and Akt modulate acute ventilation-induced vas-cular permeability increases in mouse lungs. Am. J. Physiol. LungCell. Mol. Physiol. 293, L11–L21

64. Singleton, P. A., Dudek, S. M., Ma, S. F., and Garcia, J. G.(2006) Transactivation of sphingosine 1-phosphate receptorsis essential for vascular barrier regulation. Novel role forhyaluronan and CD44 receptor family. J. Biol. Chem. 281,34381–34393

65. Liu, F., Schaphorst, K. L., Verin, A. D., Jacobs, K., Birukova, A.,Day, R. M., Bogatcheva, N., Bottaro, D. P., and Garcia, J. G.(2002) Hepatocyte growth factor enhances endothelial cellbarrier function and cortical cytoskeletal rearrangement: poten-tial role of glycogen synthase kinase-3beta. FASEB J. 16, 950–962

66. Schreibelt, G., Kooij, G., Reijerkerk, A., van Doorn, R., Gring-huis, S. I., van der Pol, S., Weksler, B. B., Romero, I. A.,Couraud, P. O., Piontek, J., Blasig, I. E., Dijkstra, C. D., Ronken,E., and de Vries, H. E. (2007) Reactive oxygen species alterbrain endothelial tight junction dynamics via RhoA, PI3 kinase,and PKB signaling. FASEB J. 21, 3666–3676

Received for publication August 20, 2008.Accepted for publication November 26, 2008.


gadd45a is a novel candidate gene in inﬂammatory lung ...€¦ · novel vili-related candidate...

Documents