Acute-Phase α1-Antitrypsin−−A Novel Regulator of -like Protein 4 Transcription and Secretion

This information is current as Eileen Frenzel, Sabine Wrenger, Stephan Immenschuh, of September 28, 2021. Rembert Koczulla, Ravi Mahadeva, H. Joachim Deeg, Charles A. Dinarello, Tobias Welte, A. Mario Q. Marcondes and Sabina Janciauskiene J Immunol 2014; 192:5354-5362; Prepublished online 23

April 2014; Downloaded from doi: 10.4049/jimmunol.1400378 http://www.jimmunol.org/content/192/11/5354

Supplementary http://www.jimmunol.org/content/suppl/2014/04/23/jimmunol.140037 http://www.jimmunol.org/ Material 8.DCSupplemental References This article cites 56 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/192/11/5354.full#ref-list-1

Why The JI? Submit online. by guest on September 28, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Acute-Phase Protein a1-Antitrypsin—A Novel Regulator of Angiopoietin-like Protein 4 Transcription and Secretion

Eileen Frenzel,* Sabine Wrenger,* Stephan Immenschuh,† Rembert Koczulla,‡ Ravi Mahadeva,x H. Joachim Deeg,{,‖ Charles A. Dinarello,# Tobias Welte,* A. Mario Q. Marcondes,{,‖ and Sabina Janciauskiene*

The angiopoietin-like protein 4 (angptl4, also known as peroxisome proliferator–activated receptor [PPAR]g–induced angiopoietin- related protein) is a multifunctional protein associated with acute-phase response. The mechanisms accounting for the increase in angptl4 expression are largely unknown. This study shows that human a1-antitrypsin (A1AT) upregulates expression and release of angplt4 in human blood adherent mononuclear cells and in primary human lung microvascular endothelial cells in a concen- tration- and time-dependent manner. Mononuclear cells treated for 1 h with A1AT (from 0.1 to 4 mg/ml) increased mRNA of angptl4 from 2- to 174-fold, respectively, relative to controls. In endothelial cells, the maximal effect on angptl4 expression was Downloaded from achieved at 8 h with 2 mg/ml A1AT (11-fold induction versus controls). In 10 emphysema patients receiving A1AT therapy (Prolastin), plasma angptl4 levels were higher relative to patients without therapy (nanograms per milliliter, mean [95% confi- dence interval] 127.1 [99.5–154.6] versus 76.8 [54.8–98.8], respectively, p = 0.045) and correlated with A1AT levels. The effect of A1AT on angptl4 expression was significantly diminished in cells pretreated with a specific inhibitor of ERK1/2 activation (UO126), irreversible and selective PPARg antagonist (GW9662), or genistein, a ligand for PPARg. GW9662 did not alter the ability of A1AT to induce ERK1/2 phosphorylation, suggesting that PPARg is a critical mediator in the A1AT-driven angptl4 http://www.jimmunol.org/ expression. In contrast, the forced accumulation of HIF-1a, an upregulator of angptl4 expression, enhanced the effect of A1AT. Thus, acute-phase protein A1AT is a physiological regulator of angptl4, another acute-phase protein. The Journal of Immunology, 2014, 192: 5354–5362.

ngiopoietin-like protein 4 (angptl4, also known as hepatic -related disorders, such as cardiovascular diseases fibrinogen/angiopoietin-related protein, fasting-induced (cardiac hypertrophy and ) (8–12), ischemia (4, 13), adipose factor, or peroxisome proliferator–activated tumor growth (4, 14, 15), diabetes (16, 17), wound healing (18,

A by guest on September 28, 2021 receptor [PPAR]g-induced angiopoietin-related protein) is an 19), and inflammation (5). Angptl4 is also believed to play a role endogenous inhibitor of (1). Human angplt4 is in the regulation of airway remodeling in asthma and chronic expressed at high levels in the placenta, heart, muscle, lung, liver obstructive pulmonary disease (20, 21). Little is known about the (2), (3), endothelial cells (4), and also in monocytes/ regulation of angptl4 in vivo. The expression of angptl4 is induced macrophages (5). Adipose tissue is suspected to be an abundant by ischemic and hypoxic conditions and by treatment with glu- source of serum angptl4 (6). cocorticoids, PPAR agonists, and TGF-b, among others (4, 22– Angptl4 is implicated in the regulation of glucose homeostasis, 27). The overexpression of angptl4 associated with acute-phase sensitivity, and lipid metabolism (7) and is associated with reactions, is suggested to have an anti-inflammatory effect. Ex- perimental studies demonstrate that induction of angptl4 expres- sion reduces atherosclerosis development (28) and protects pulmonary *Department of Respiratory Medicine, Hannover Medical School, 30625 Hannover, Germany; †Institute for Transfusion Medicine, Hannover Medical School, 30625 microvascular endothelial cells against endotoxin-induced injury (21). Hannover, Germany; ‡Division of Pulmonary Diseases, Department of Internal Med- x Hence, angptl4 is a multifunctional protein, with potential significance icine, Philipps-Universita¨t Marburg, 35037 Marburg, Germany; Department of Re- spiratory Medicine, University of Cambridge, Cambridge CB2 0QQ, United for disease pathogenesis. Kingdom; {Department of Medicine, University of Washington, Seattle, WA In this study, we demonstrate that angptl4 expression is regulated ‖ 98195; Clinical Research Division, Fred Hutchinson Cancer Research Center, by another acute-phase protein, namely human a1-antitrypsin Seattle, WA 98109; and #Department of Medicine, University of Colorado Denver, Aurora, CO 80045 (A1AT). A1AT is one of the most abundant acute-phase in Received for publication February 10, 2014. Accepted for publication March 19, the circulation and one of the fastest acting inhibitors of neutrophil 2014. proteases. A1AT is a multifunctional protein involved in the reg- This work was supported by Baxter Healthcare, Hannover Medical School, Deutsches ulation of acute-phase responses (29–33). In recent years, previ- Zentrum fur€ Lungenforschung, and the Cambridge Biomedical Research Centre. ously unrecognized functions of A1AT have been identified (34), Address correspondence and reprint requests to Prof. Dr. Sabina Janciauskiene, Depart- including protective effects against atherosclerosis (35), inhibition ment of Respiratory Medicine, Hannover Medical School, Feodor-Lynen Strasse 23, of cell apoptosis (36, 37), and tumor growth (31). 30625 Hannover, Germany. E-mail address: [email protected] The clinical relevance of A1AT is highlighted in individuals The online version of this article contains supplemental material. with inherited ZZ (Glu342Lys) A1AT deficiency (plasma levels Abbreviations used in this article: A1AT, a1-antitrypsin; angptl4, angiopoietin-like protein 4, DCF, 29,79-dichlorofluorescein; DFOM, deferoxamine mesylate; DMOG, below 0.7 g/l, whereas normal values range between 1 and 2 g/l). dimethyloxalylglycine; HIF-1a, hypoxia-inducible factor-1a; HMVEC-L, primary These individuals are at high risk of developing early onset emphy- human lung microvascular endothelial cell; HSA, human serum albumin; PPAR, sema, liver and pancreatic diseases at any age, and in rare cases pan- peroxisome proliferator–activated receptor; ROS, reactive oxygen species. niculitis and vasculitis (38). Twenty-five years ago augmentation Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 therapy with A1AT isolated from pooled human plasma was de- www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400378 The Journal of Immunology 5355 veloped to treat patients with A1AT deficiency–related emphy- Patient plasma samples sema. Therapy with A1AT also modulates or prevents tissue injury Ten of 20 patients (4 PiSZ and 6 PiZZ: one female and nine males, age 45 in experimental animal models of human diseases, including graft- [15.3] y [mean [SD]) received weekly Prolastin (60 mg/kg), whereas 10 versus-host-disease, rheumatoid arthritis, autoimmune diabetes, patients (PiZZ: 5 females and 5 males at age 65 [8.34] y [mean [SD]) had no and renal ischemia–reperfusion injury, among others (39, 40). How- therapy (controls). All participants were nonsmokers. The mean forced ever, despite vast anti-inflammatory properties ascribed to A1AT expiratory volume in 1 s (FEV1%) was 44.4 and 64.8% in augmented and nonaugmented patients, respectively. All included patients had normal (41), the mechanisms of its anti-inflammatory effects remain in- levels of C-reactive protein. Plasma samples were obtained just before and completely understood. Our finding that A1AT regulates angptl4 2 h after therapy with Prolastin and at random times from controls and stored expression and release reveals previously unrecognized ac- at 280˚C until use. Participants had signed informed consent as required tivity of this protein. Moreover, in vitro results suggest adherent by the ethics committee of the University of Marburg (59/06). PBMCs and endothelial cells as important sources of circulating Experimental conditions angplt4. In some experiments, a part of human adherent PBMCs or HMVEC-L was pretreated for 30 min with specific MAPK/ERK kinase inhibitors UO126 Materials and Methods and PD98059 (10 mM; Sigma-Aldrich), with 10 mM GW9662, a selective AAT preparation and irreversible PPARg antagonist (Santa Cruz Biotechnology), or with PPARg activating compound genistein (1–25 mM; Roth), with 10 mM Pooled human plasma A1AT (Prolastin; Grifols) was used in all experi- CAY10585, a small molecule inhibitor of hypoxia-inducible factor-1a ments. A1AT preparations were dissolved in sterile water for injections (HIF-1a) accumulation and gene transcriptional activity (Cayman Chemical), provided by the manufacturer. Prior to the experiments, Prolastin was with 1 mM dimethyloxalylglycine (DMOG), a cell permeable inhibitor of repurified using centricon-30 (30 kDa cutoff; Sartorius), diluted in sterile HIF-1a degrading prolyl hydroxylase (Tocris), or 20 mM deferoxamine Downloaded from PBS, aliquoted, and stored at 220˚C. In some experiments, plasma-purified mesylate (DFOM), an HIF-1a stabilizing iron chelator (Sigma-Aldrich). A1AT preparations obtained from Baxter (Aralast), Calbiochem, or Sigma- Afterward, cells were cultured for various time points alone or in the pres- Aldrich were used. ence of A1AT (from 0.1 to 4 mg/ml), or LPS (Sigma Aldrich; 1 mg/ml LPS for HMVEC-L and 50 ng/ml LPS for adherent PBMCs), or combination of Isolation of PBMCs and neutrophils A1AT and LPS. In some experimental settings, cells were treated with human Human PBMCs were isolated from peripheral blood of healthy volunteers serum albumin (HSA, MP Biomedicals) as a control. At the end of incuba- using Lymphosep discontinuous gradient centrifugation (c.c. pro GmbH), tion, cell supernatants and lysates were collected for further analysis. http://www.jimmunol.org/ according to the manufacturer’s instructions. PBMCs were then resus- Angiogenesis array for HMVEC-L pended in RPMI 1640 medium with 2 mM N-acetyl-L-alanyl-L-glutamine (Life Technologies) supplemented with 1% nonessential amino acids, 2% Serum- and vascular endothelial –starved HMVEC-L were sodium pyruvate, and 20 mM HEPES and plated at a density of 5 3 106 treated with AAT (1 mg/ml) or PBS (as a control) for 8 h. Total RNA was cells/ml. Cells were incubated for 75 min at 37˚C in a 5% CO2 incubator to prepared using the RNeasy Mini (Qiagen) and quantified with Nano- allow monocytes to adhere to the cell culture plates. Afterward, nonadherent Drop 1000 spectrophotometer (Thermo Fisher Scientific). RNA purity was cells were washed away with PBS containing Mg2+ and Ca2+ (Life Tech- analyzed on 1% agarose gels. cDNA was synthesized from 1 mg RNA with nologies), and adherent mononuclear cells were used for experiments. RT2 First Strand kit (SABiosciences, Qiagen) and analyzed for the ex- Human neutrophils were isolated from freshly obtained peripheral blood pression of angiogenesis-related using Human Angiogenesis RT2 of healthy volunteers using Polymorphprep (Axis-Shield PoC) as described Profiler PCR Array (SABiosciences, Qiagen), according to the manu- previously (42). Purified neutrophils were washed in PBS and then facturer’s protocol. Quantitative PCR was conducted on ABI 7500 machine by guest on September 28, 2021 resuspended in RPMI 1640 medium. The neutrophil purity was typically (Applied Biosystems, Life Technologies). b2-Microglobulin, hypoxanthine 90% as determined by cytospin and cell viability was .95% according to phosphoribosyltransferase 1, ribosomal protein L13a, GAPDH, and b-actin 0.4% trypan blue staining. Cells were preincubated for 30–40 min before were measured in the same run and used as internal controls. the start of experiment. Relative (fold change) was determined using the ΔΔCT method on the SAB web portal (http://pcrdataanalysis. Primary human lung microvascular cell culture sabiosciences.com/pcr/arrayanalysis.php). Human microvascular endothelial cells from lung (HMVEC-L; Promocell) Specific gene expression analysis by RT-PCR were cultured in MV-2 endothelial cell growth medium (Promocell) at 37˚C with 5% CO2. HMVEC-L (passages 4–6, at 70–90% confluence) were Total cellular RNA was isolated using the RNeasy Mini Kit (Qiagen), starved from serum and vascular endothelial growth factor for 4 h before according to the manufacturer’s instructions. The amount of total RNA was start of experiment. According to recommendations of the provider, cells determined with NanoDrop 1000 Spectrophotometer (Thermo Fisher were kept in complete medium during the incubation times longer than 20 h. Scientific). cDNA was synthesized by reverse transcription using the high-

FIGURE 1. Time- and concentration-dependent effects of A1ATon angptl4 expression and release from human blood adherent PBMCs. Adherent human PBMCs were treated with constant concentration of A1AT (1 mg/ml) for different periods of time (A) or incubated with various concentrations of A1AT for 1h(B). Gene expression levels of angptl4 were analyzed by RT-PCR and normalized to GAPDH. Each point represents mean 6 SD of n = 13 independent experiments, each with two repeats. (C) PBMCs were treated with 1 and 2 mg/ml A1AT or 2 mg/ml HSA for 6 h, and the concentration of angptl4 in cell culture supernatants was determined by ELISA. Each bar represents mean 6 SD of n = 4 independent experiments, each with two repeats. 5356 A1AT IS AN INDUCER OF angptl4

FIGURE 2. Time- and concentration-dependent effects of A1AT on angptl4 expression and release from HMVEC-L. (A) HMVEC-L cells were treated with 1 mg/ml A1AT for the indicated periods of time. Angptl4 expression was determined by RT-PCR. Each point represents mean 6 SD of n =3orn = 4 independent experiments, each with three repeats. Cells were treated for 8 h with different concentrations of A1AT ranging from 0.03 to 2 mg/ml (B). The expression of angptl4 relative to GAPDH expression was assessed by RT-PCR. Each point represents mean 6 SD of n = 5 independent experiments, each with three repeats. (C) Angptl4 concentration in culture supernatants was determined by ELISA. Data are mean 6 SEM of n = 4 independent experiments, each with two repeats. Downloaded from capacity cDNA reverse transcription kit (Applied Biosystems, Life Tech- oxygen species (ROS) into fluorescent and cell-impermeable 29,79- nologies). Expression of angptl1, angptl4, HIF-1a, TNF-a, and PPARg dichlorofluorescin (DCF). Labeled cells were resuspended in HBSS and was analyzed by real-time PCR using the TaqMan Gene Expression Assay incubated with A1ATor PMA (Sigma-Aldrich), a well-described inducer of (Applied Biosystems, Life Technologies). The expression of the house- ROS, alone or with A1AT plus PMA. In some conditions, cells were pre- keeping gene GAPDH was used for normalization of cDNA levels in all incubated for 30 min with UO126 (10 mM) prior to addition of A1AT or experiments. All primers were purchased from Applied Biosystems. Rel- PMA. DCF fluorescence (excitation at 498 nm, emission at 522 nm) was

ative gene expression was calculated according to the ΔΔCT method. monitored for 1 h at 37˚C using Infinite M200 microplate reader (Tecan). http://www.jimmunol.org/ Electrophoresis (SDS-PAGE) and Western blot analysis Statistical analysis After incubation, cells were washed three times with ice-cold PBS and lysed The differences in the means of experimental results were analyzed using in radioimmunoprecipitation assay buffer (Santa Cruz Biotechnology). one-way ANOVA combined with a multiple-comparison procedure (Scheffe The protein concentration in the lysates was determined by the BC protein multiple range test), with an overall significance level of a = 0.05. An in- assay kit (Uptima and Interchim). Equal amounts of protein were loaded on dependent two-sample t test was also used where appropriate. Statistical 12.5, 10, or 7.5% polyacrylamide gels. After SDS-PAGE, proteins were Package (SPSS for Windows, release 19.0) was used for the statistical cal- transferred onto polyvinylidene fluoride membrane by semidry electro- culations. blotting. For specific detection, the following primary Abs were used: polyclonal anti-ERK1/2, monoclonal anti-phosphorylated ERK1/2, mono-

Results by guest on September 28, 2021 clonal anti–b-actin (AC-15; Sigma-Aldrich), and monoclonal anti–HIF-1a (BD Transduction Laboratories). The immunocomplexes were visualized Time- and concentration-dependent effects of A1AT on angptl4 with appropriate secondary HRP-conjugated Abs (DakoCytomation) and expression and release from human blood adherent PBMCs ECL Western blotting substrate (Thermo Fisher Scientific). The specific bands were quantified using ImageJ software. The effects of 1 mg/ml A1AT on angptl4 expression and release were measured after culture of adherent PBMCs from 1 to 24 h. As Quantitative analysis of angptl4 shown in Fig. 1A, exposure of cells to A1AT for 1 h resulted in Patient plasma samples and cell culture supernatants were analyzed for a 10-fold induction of angptl4 expression, whereas at 6 h, angptl4 angptl4 protein levels using DuoSet ELISA sets (R&D Systems). Detection expression was maximally induced by 28-fold (p , 0.001) relative limit was 1.25 ng/ml. to controls. At 24 h, angptl4 expression in A1AT-treated cells Reactive oxygen species assay in adherent PBMCs remained significantly higher (17.7-fold; p , 0.001) than in con- Cells were labeled with 29,79-dichlorofluorescein diacetate (Sigma-Aldrich), trols (Fig. 1A). Similar to A1AT (Prolastin), two other A1AT which is deacetylated by intracellular esterases and oxidized by reactive preparations (Aralast from Baxter and A1AT from Sigma-Aldrich)

FIGURE 3. Augmentation therapy with A1AT (Prolastin) increases plasma levels of angptl4. (A) Ten patients with A1AT deficiency (4 PiSZ and 6 PiZZ) not receiving A1AT therapy had significantly lower plasma levels of A1AT in comparison with matched 10 PiZZ patients receiving weekly A1AT aug- mentation therapy. Levels of A1AT were analyzed by nephelometry. (B) Within in the group of patients receiving A1AT therapy plasma angptl4 levels were higher as compared with patients never treated with A1AT therapy. In particular, plasma angptl4 levels increased directly after A1AT therapy. The levels of angptl4 were analyzed by ELISA. (C) A direct correlation was observed between plasma A1AT and angptl4 levels in all cohorts (Spearman’s correlations). The Journal of Immunology 5357

as compared with controls. Similarly, treatment of liver carcinoma cells (HepG2) with 1 or 4 mg/ml A1AT for 1, 4, or 24 h had no any effect on angptl 1 and 4 expression (data not shown).

Time- and concentration-dependent effects of A1AT on angptl4 expression and release from HMVEC-L To assess whether the effect of A1AT on angplt4 expression is not restricted to adherent PBMCs, we next used a HMVEC-L model. As shown in Fig. 2A–C, A1AT induced angpt14 expression and release in a time- and concentration-dependent manner with a maximal effect at 8 h (6-fold, p , 0.01). A significant induction of angplt4 expression occurred when cells were treated with A1AT of 1 mg and above for 8 h (Fig. 2B). Elevated angptl4 expres- sion paralleled with increased protein release in cell supernatants (Fig. 2C). There was no effect from HSA (1 mg/ml) relative to controls (mRNA angptl4 fold change, mean [SEM]: 0.8 [0.06] versus 1 [0.05], n = 3, and angplt4 release [nanograms per mil- liliter] mean [SEM]: 6713 [234] versus 6274 [154], n =4,re- spectively). Downloaded from Data from the angiogenesis array showed that endothelial cells treated with 1 mg/ml A1AT for 8 h, except for induction of angplt4 FIGURE 4. A1AT activates ERK1/2 in adherent PBMCs and HMVEC- mRNA (by 5-fold; p , 0.01) showed no changes in any other gene L. Western blots of cell lysates prepared from adherent PBMCs (A) and expression (Supplemental Table I). For comparison, 0.5 mg/ml HMVEC-L (B) illustrate phosphorylation of ERK1/2 at 30 min in response to the treatment with 1 mg/ml A1AT. Cell lysates were immediately an- A1AT purchased from Calbiochem also enhanced expression of

, http://www.jimmunol.org/ alyzed for phosphorylated and total ERK1/2. b-Actin was used as a load- angptl4 mRNA (by 8.5-fold; p 0.01) in HMVEC-L at 8 h ing control. Cell preincubation for 30 min with MEK inhibitors, 10 mM (Supplemental Fig. 2). UO126 or PD98059, completely abolished ERK1/2 activation in controls and A1AT-treated cells. Each blot is representative out of n = 5 indepen- Therapy with A1AT (Prolastin) upregulates plasma levels of dent experiments. angptl4 To determine the stimulatory effect of A1AT on angptl4 levels significantly induced mRNA of angplt4 at 1 h (Supplemental Fig. 1). in vivo, we assessed a small cohort of A1AT deficiency–related In contrast, A1AT had no effect on angplt 1 expression (data emphysema patients with and without A1AT therapy. Ten patients not shown). with severe A1AT deficiency (4 PiSZ and 6 PiZZ) not receiving We next examined A1AT concentration-dependent effect on A1AT therapy had plasma levels of A1AT (mean [SEM] 0.45 by guest on September 28, 2021 angptl4 expression (Fig. 1B). Adherent PBMCs treated for 1 h [0.09] g/l, n = 10). For comparison, another group of 10 clinically with A1AT (0.1–4 mg/ml) showed a rapid increase in levels of matched PiZZ patients receiving weekly A1AT therapy was tested angplt4 mRNA that peaked with 4 mg/ml A1AT (174-fold in- for plasma A1AT and angptl4 levels just before therapy and ∼2h crease as compared with controls; p , 0.001). At a concentration after therapy. As expected, before therapy plasma levels of A1AT of 0.1 mg/ml, A1AT also induced angplt4 expression, albeit at were low (mean [SEM] 0.78 [0.06] g/l, n = 10) but increased very modest levels (by 1.8-fold). In support of this, supernatants significantly directly after therapy (mean [SEM] 2.71 [0.11] g/l, from cells treated with 1 or 2 mg/ml A1AT for 6 h contained n = 10; p , 0.001) (Fig. 3A). significantly more angplt4 protein than nontreated controls. HSA Within the group of patients receiving A1AT infusions plasma had no effect on angptl4 expression (data not shown) and release angptl4 levels were higher than in patients never treated with A1AT (Fig. 1C). (Fig. 3B). In particular, plasma angptl4 levels increased directly We also tested whether A1AT induces expression of angptl 1 and after A1AT therapy (nanograms per milliliter, mean [95% confi- 4 in human blood neutrophils. Notably, basal expression levels dence interval] 127.1 [99.5–154.6] versus nontreated patients 76.8 of these proteins were very low or undetectable in neutrophils. [54.8–98.8]; p = 0.025). A trend was noted for a direct relationship Moreover, the incubation of neutrophils with 1 mg/ml A1AT for between plasma A1AT and angptl4 levels in a whole cohort (r = 2 or 4 h showed no effect on angplt1 and 4 expression and release 0.37; p = 0.044; n = 30) (Fig. 3C).

FIGURE 5. A1AT-induced angptl4 expression is suppressed by UO126 in adherent PBMCs (A) and HMVEC-L (B). Cells were pretreated for 30 min with 10 mM UO126 prior to addition of 1 mg/ml A1AT and incubated for additional 6 h (adherent PBMCs) and 3 h (HMVEC-L). Bars represent mean 6 SD of n = 9 in- dependent experiments, each with three repeats. 5358 A1AT IS AN INDUCER OF angptl4

PD98059+A1AT, mean [SEM]: 27.8 [6.1] versus 31.5 [9.9], ex- periment n = 13 and n = 6, respectively). A1AT-induced expression of angptl4 is related to HIF-1a protein levels in adherent PBMCs As illustrated in Fig. 6A, PBMCs incubated with 1 mg/ml A1AT increased HIF-1a expression in a time-dependent manner by ∼4- fold as compared with controls. Concomitantly, A1AT slightly increased HIF-1a protein levels at 6 h (p , 0.05) (Fig. 6B). Re- markably, 1 mg/ml A1AT had no effect on HIF-1a expression in HMVEC-L at 8 h (Supplemental Table I). No effect on mRNA of HIF-1a in HMVEC-L was found at other time points; therefore, the next set of experiments was performed only with adherent PBMCs. Novel studies suggest that angptl4 can be upregulated by the transcriptional enhancer, HIF-1a (15). When adherent PBMCs were preincubated with CAY10585 (inhibitor of HIF-1a tran- FIGURE 6. A1AT induces HIF-1a mRNA and protein levels in human scriptional activity and accumulation, Fig. 6B), no significant

PBMCs. (A) To determine the effect of A1AT on HIF-1a expression, ad- changes in A1AT-induced angptl4 expression were observed Downloaded from herent PBMCs were incubated with 1 mg/ml A1AT from 1 to 24 h. HIF-1a (Fig. 7A). Preincubation of PBMCs with HIF-1a stabilizer DFOM expression levels were normalized to GAPDH. Each point represents (Fig. 7B) not only induced angptl4 mRNA (by 26-fold; p , 0.001 mean 6 SD of n = 13 independent experiments. (B) To determine the relative to controls) but also enhanced the effects of A1AT on effect of A1AT on cellular HIF-1a protein levels, PBMCs were treated angptl4 expression (by 7.9-fold; p , 0.001). Similarly, there was with 1 mg/ml A1AT for 6 h, and total cell lysates were analyzed by a great increase in HIF-1a protein (Fig. 6B) and mRNA (Fig. 7C) Western blot. For the controls, cells were preincubated for 30 min with 10 a

levels in cells preincubated with DMOG, a preventer of HIF-1 http://www.jimmunol.org/ mM CAY10585, an inhibitor of HIF-1a accumulation and transcriptional deactivation. DMOG enhanced angptl4 mRNA (by 75-fold; p , activity, or 1 mM DMOG, an HIF-1a stabilizer, before adding A1AT. b-Actin was used as a loading control. Data are representative of n =4 0.001 relative to controls), whereas costimulation with A1AT led independent experiments. to the strong additive effect on angptl4 expression (Fig. 7C). GW9662 and genistein inhibit the effect of A1AT on angptl4 A1AT-induced angptl4 expression is dependent on ERK1/2 mRNA activation Angptl4 is a target gene of PPARg (3, 27). PPAR agonists are The activation of ERK1/2 has been identified as a potent modula- known to elevate the circulating levels of angptl4 in humans and

tor of angptl4 expression (20, 43). Previous finding that A1AT rodents (45). Treatment of adherent PBMCs with A1AT did not by guest on September 28, 2021 induces transient activation of ERK1/2 in human monocytes (44) and affect the mRNA of PPARg (Fig. 8A). However, the preincuba- neutrophils (42), we now confirmed in human adherent PBMCs tion of PBMCs or HMVEC-L with GW9662, a selective and irre- and HMVEC-L (Fig. 4A, 4B). Preincubation of cells with MEK/ versible PPARg antagonist, strongly inhibited stimulatory effect of ERK inhibitor UO126 for 30 min blocked ERK1/2 activation A1AT on angptl4 mRNA (Fig. 8B, 8C). Because it was proposed (Fig. 4) and dramatically inhibited the ability of A1AT to induce that GW9662 inhibits ERK1/2 activation (46), we investigated angplt4 expression (Fig. 5A, 5B). whether this pathway was responsible for the attenuation of Notably, preincubation of adherent PBMCs with MEK inhibitor A1AT-induced angplt4 mRNA. As shown in Fig. 8D, 10 mM PD98059 also blocked ERK1/2 activation at 30 min (Fig. 4A). GW9662 did not alter the property of A1AT to induce ERK1/2 Unfortunately, 10 mM PD98059 blocking effect on EKR1/2 acti- phosphorylation. Hence, the inhibition of A1AT-induced angptl4 vation vanished after 1 h, and because of toxicity for the primary expression by GW9662 is associated with a failure of PPARg cells, we were not able to increase the concentration of PD98059. activation. Thus, at the concentration used, PD98059 had no significant ef- Using adherent PBMCs, in the following set of experiments, we fect on A1AT-induced angplt4 expression at 6 h (A1AT versus checked whether genistein, a potent PPAR g ligand, has any effect

FIGURE 7. A1AT-induced expression of angptl4 is related to HIF-1a protein levels in adherent PBMCs. Cells were preincubated for 30 min with 10 mM CAY10585, an HIF-1a inhibitor (A) or with HIF-1a stabilizers, 20 mM DFOM (B) or 1 mM DMOG (C). Thereafter, cells were incubated with 1 mg/ml A1AT for 6 h. The expression of angptl4 was determined by RT-PCR. Expression levels were normalized to GAPDH. Each bar represents mean 6 SD of n = 6 independent experiments, each with three repeats. The Journal of Immunology 5359

FIGURE 8. A1AT-induced expression of angptl4 is related to PPARg activity. Cells were pretreated for 30 min with 10 mM GW9662, an irreversible PPARg antagonist, prior to treat- ment with 1 mg/ml A1AT for 6 h (for adherent PBMCs) or 3 h (for HMVEC-L). PBMC mRNA of PPARg and angptl4 (A and B) and HMVEC-L mRNA of angptl4 (C) were assessed by RT- PCR. Expression levels were normalized to GAPDH. Each bar represents mean 6 SD from six independent experiments for PBMC and three experiments for HMVEC-L. (D) Adherent PBMCs were preincubated alone or with 10 mM GW9662 for 30 min, and 1 mg/ml A1AT was added for another 30 min. At the end of incubation time, total cell lysates were directly analyzed for total and phosphorylated ERK1/2 by Western blot analysis. Downloaded from b-Actin was used as a loading control. Each blot is representative of n = 5 repeated experiments. http://www.jimmunol.org/ on A1AT-stimulated angptl4 expression. At 1 h, 0.5 mg/ml A1AT (Fig. 11A, 11B). The same result was also confirmed in adherent and 25 mM genistein induced angptl4 mRNA by .30- and 4.9- PBMCs (data not shown). fold, respectively (Fig. 9). When cells were preincubated with 5 or 25 mM genistein, A1AT effect on angptl4 expression was dimin- Discussion , ished by 58 and 70% (p 0.001), respectively. In this study, we demonstrate in human adherent PBMCs and A1AT decreases LPS-induced TNFa expression and HMVEC-L that A1AT induces angptl4 expression and protein PMA-induced free radical (ROS) generation independently secretion in a time- and concentration-dependent manner. Actually, of angptl4 upregulation mononuclear cells treated with 0.1–4 mg/ml A1AT for 1 h in- creased mRNA of angptl4 from 2- to 174-fold compared with the by guest on September 28, 2021 We evaluated the effects of A1AT on LPS-induced TNF-a ex- levels in controls. In endothelial cells, the maximal effect on pression and PMA-induced ROS production in adherent PBMCs angptl4 expression was achieved with 2 mg/ml A1AT at 8 h (11- pretreated with UO126, an inhibitor of A1AT-induced angptl4 fold induction versus control). Interestingly, in mononuclear cells, expression. As illustrated in Fig. 10A–D, A1AT inhibited LPS- A1AT-induced angplt4 expression occurred more quickly and to induced TNF-a and PMA-induced ROS production independently a greater extent than in the HMVEC-L. The observed differences from preincubation of cells with UO126. In addition, it is im- are likely related to the specific properties of these cells. It is also portant to point out that presence of LPS did not influence A1AT noteworthy to mention that A1AT had no effect on angplt4 mRNA stimulatory effect on angptl4 expression and release in HMVEC-L in neutrophils and liver carcinoma cells, suggesting cell-type specific effect of A1AT. In a small cohort of emphysema patients with inherited Pi*Z A1AT deficiency, we found variability in plasma angplt4 levels (ranging from 40 to 224 ng/ml). Data on plasma angptl4 con- centrations in humans are still scarce (47) and show high vari- ability. For example, a Finnish study revealed that angptl4 serum levels are ranging from 2 to158 ng/ml (48). Despite this vari- ability, we present data showing that augmentation therapy with A1AT increases plasma levels of angptl4. Moreover, indepen- dently of A1AT therapy, we observed a trend toward a direct re- lationship between angplt4 and A1AT. This latter observation is of particular interest because angptl4 is an inhibitor of lipoprotein lipase, a key regulatory enzyme in fatty acid metabolism (49, 50). Therefore, A1AT might indirectly contribute to the hyper- triglyceridemia that characteristically occurs during the acute-phase response. To answer this question is beyond the scope of our study. The above findings raise the question as to how A1AT regulates FIGURE 9. Genistein and GW9662 diminish the ability of A1AT to induce angptl4 expression. Adherent PBMCs were preincubated for 30 min angptl4 expression. Our earlier studies have shown that exogenous with various concentrations of genistein or 10 mM GW9662 prior to A1AT interacts with lipid rafts and rapidly activates ERK1/2 in treatment with 0.5 mg/ml A1AT for 1 h. mRNA of angptl4 was assessed by monocyte cultures (44). Similarly, when added to human blood RT-PCR. Each bar represents mean 6 SEM of n = 2 independent neutrophils, within minutes A1AT induces a transient polarization experiments, each with six repeats. and random migration of cells, events that are accompanied by 5360 A1AT IS AN INDUCER OF angptl4

FIGURE 10. A1AT decreases LPS-induced TNF-a expression and PMA-induced ROS generation inde- pendently of angptl4 up-regulation. Adherent PBMCs were preincubated alone (A) or with 10 mM UO126 (B) for 30 min prior to stimulation for 8 h with 50 ng/ml LPS alone or LPS plus 1 mg/ml A1AT. mRNA of TNF-a was assessed by RT-PCR. Each bar represents mean 6 SEM of n = 2 independent experiments, each with six repeats. Adherent PBMCs were labeled with 10 mM 29,79-dichlorofluorescein diacetate and treated with HBSS (d) and 1 mg/ml A1AT (s)(C) or 150 nM PMA (d) and A1AT+PMA (s)(D). In some conditions, prior to adding A1AT, PMA, or A1AT+PMA, cells were pretreated for 30 min with 10 mM UO126 (;) (C and D), an inhibitor of A1AT-induced angptl4

mRNA. ROS production was monitored by the intra- Downloaded from cellular accumulation of oxidized fluorescent DCF. Representative graphs are means of n = 4 independent experiments, each performed in two repeats. http://www.jimmunol.org/

ERK1/2 phosphorylation (42). In this study, we confirmed that Furthermore, genistein, a compound interacting directly with the A1AT induces a transient (15–60 min) ERK1/2 phosphorylation in PPARg ligand–binding domain, dose-dependently inhibited the ef- adherent PBMCs and HMVEC-L. The ERK pathway is involved fect of A1AT on angplt4 expression. Thus, A1AT-induced angptl4 in the regulation of angptl4 transcription (20, 43). Therefore, we expression occurs in a PPARg-dependent manner. speculated that A1AT-induced ERK1/2 phosphorylation might be The expression of angptl4 is also regulated by hypoxia and essential for the upregulation of angptl4. As predicted, in the chronic inflammatory responses (4, 6). Recent studies provide

presence of the MEK inhibitor (UO126), ERK1/2 activation was evidence that angptl4 can be induced by the transcription factor by guest on September 28, 2021 blocked, and consequently, the effect of A1AT on angptl4 ex- HIF-1a (53). HIF-1a functions by regulating many of the genes pression was markedly diminished or abolished. involved in angiogenesis, erythropoiesis, glycolysis, iron metab- In contrast, it is suggested that inhibition of ERK1/2 by UO126 olism, and cell survival (54). Although HIF-1a is regulated mainly decreases PPARg activity (51, 52). Angptl4 is one of the down- by oxygen tension, the transcriptional activity of HIF-1a also can stream target genes of PPARg (27), and it is well documented that be enhanced by the activation of ERK1/2 (55, 56). Because angptl4 PPARg agonists enhance expression of angptl4 (3). In this study, is a target gene of HIF-1a, it was important to determine the role of we show that GW9662, a selective and irreversible PPARg an- HIF-1a in A1AT-induced angptl4 mRNA. Preincubation of adherent tagonist, abolished the ability of A1AT to induce angptl4 expression, PBMCs with CAY10585 (to inhibit HIF-1a transcriptional activity however, did not prevent A1AT to induce ERK1/2 phosphorylation. and accumulation) had no effect on A1AT-induced angptl4 mRNA

FIGURE 11. LPS does not affect the ability of A1AT to induce angptl4 expression and release in HMVEC-L. (A) Cells were incubated with 1 mg/ml LPS, 1 mg/ml A1AT or A1AT+LPS for 8 h. The expression of angptl4 was analyzed by RT-PCR and calculated versus GAPDH levels. Each bar represents mean 6 SEM of n = 2 independent experiments, each with three repeats. (B) Angptl4 protein levels were determined in cell culture supernatants by ELISA. Bars represent mean 6 SEM of n = 3 independent experiments, each performed in two repeats. The Journal of Immunology 5361 levels, suggesting that A1AT effect is independent on HIF-1a tran- a proangiogenic factor produced during ischemia and in conventional renal cell carcinoma. Am. J. Pathol. 162: 1521–1528. scription. 5. Lichtenstein, L., F. Mattijssen, N. J. de Wit, A. Georgiadi, G. J. Hooiveld, R. van However, preincubation of cells with DFOM (which mimics der Meer, Y. He, L. Qi, A. Ko¨ster, J. T. Tamsma, et al. 2010. Angptl4 protects hypoxia and stabilizes HIF-1a via inhibition of prolyl hydrox- against severe proinflammatory effects of saturated by inhibiting fatty acid uptake into mesenteric lymph node macrophages. Cell Metab. 12: 580–592. ylases) induced angptl4 expression (by 26-fold) and enhanced the 6. Lu, B., A. Moser, J. K. Shigenaga, C. Grunfeld, and K. R. Feingold. 2010. The effect of A1AT. Similarly, DMOG, by preventing deactivation of acute phase response stimulates the expression of angiopoietin like protein 4. HIF-1a, also significantly enhanced angptl4 mRNA (by 75-fold), Biochem. Biophys. Res. Commun. 391: 1737–1741. 7. Grootaert, C., T. Van de Wiele, W. Verstraete, M. Bracke, and B. Vanhoecke. and again A1AT-induced upregulation of angptl4 expression was 2012. Angiopoietin-like protein 4: health effects, modulating agents and greater than in the absence of DMOG. Thus, upregulation of structure-function relationships. Expert Rev. Proteomics 9: 181–199. 8. Adachi, H., Y. Fujiwara, T. Kondo, T. Nishikawa, R. Ogawa, T. Matsumura, angptl4 mRNA by A1AT is enhanced by the accumulation of HIF- N. Ishii, R. Nagai, K. Miyata, M. Tabata, et al. 2009. Angptl 4 deficiency 1a protein but is independent on HIF-1a transcriptional activity. improves lipid metabolism, suppresses foam cell formation and protects against Altogether, our findings highlight the complex interrelationship atherosclerosis. Biochem. Biophys. Res. Commun. 379: 806–811. 9. Hauton, D., and G. M. Caldwell. 2012. Cardiac lipoprotein lipase activity in the between ERK1/2 phosphorylation, activation of PPARg, and ac- hypertrophied heart may be regulated by fatty acid flux. Biochim. Biophys. Acta cumulation of HIF-1a during the induction of angptl4 by A1AT. 1821: 627–636. Furthermore, these findings not only provide new insights into the 10. Miida, T., and S. Hirayama. 2010. Impacts of angiopoietin-like proteins on li- poprotein metabolism and cardiovascular events. Curr. Opin. Lipidol. 21: 70–75. regulation of angptl4 expression but also broaden our knowledge 11. Nettleton, J. A., K. A. Volcik, E. W. Demerath, E. Boerwinkle, and A. R. Folsom. on the biological functions of A1AT. 2008. Longitudinal changes in according to ANGPTL4[E40K] genotype and longitudinal body weight change in the atherosclerosis risk in It is well accepted that the rise in the plasma concentration of communities study. Ann. Epidemiol. 18: 842–846. A1AT during acute-phase response assists host defense. Hence, 12. Talmud, P. J., M. Smart, E. Presswood, J. A. Cooper, V. Nicaud, F. Drenos, Downloaded from A1AT is a potent anti-inflammatory protein (37, 57), and we J. Palmen, M. G. Marmot, S. M. Boekholdt, N. J. Wareham, et al. 2008. ANGPTL4 E40K and T266M: effects on plasma and HDL levels, thought that its activities might be directly linked to the induction postprandial responses, and CHD risk. Arterioscler. Thromb. Vasc. Biol. 28: of angptl4. However, UO126, an inhibitor of ERK1/2 activation 2319–2325. and angptl4 expression, did not abolish inhibitory effect of A1AT 13. Galaup, A., E. Gomez, R. Souktani, M. Durand, A. Cazes, C. Monnot, J. Teillon, S. Le Jan, C. Bouleti, G. Briois, et al. 2012. Protection against myocardial in- on 1) ROS production and 2) LPS-induced TNF-a expression. farction and no-reflow through preservation of vascular integrity by

Moreover, A1AT-driven angplt4 expression was independent on angiopoietin-like 4. Circulation 125: 140–149. http://www.jimmunol.org/ 14. Galaup, A., A. Cazes, S. Le Jan, J. Philippe, E. Connault, E. Le Coz, H. Mekid, LPS. On the basis of these observations, it is conceivable that L. M. Mir, P. Opolon, P. Corvol, et al. 2006. Angiopoietin-like 4 prevents me- A1AT-induced angplt4 expression contributes to anti-inflammatory tastasis through inhibition of vascular permeability and tumor cell motility and activities of A1AT but not necessarily direct them. In contrast, these invasiveness. Proc. Natl. Acad. Sci. USA 103: 18721–18726. 15. Li, H., C. Ge, F. Zhao, M. Yan, C. Hu, D. Jia, H. Tian, M. Zhu, T. Chen, G. Jiang, findings provide novel evidence that A1AT expresses biological et al. 2011. Hypoxia-inducible factor 1 a‑activated angiopoietin-like protein 4 functions not only under the influence of activating stimuli but also contributes to tumor metastasis via vascular cell adhesion molecule-1/ b1 under basal conditions. Under normal physiologic conditions, A1AT signaling in human hepatocellular carcinoma. Hepatology 54: 910–919. 16. Smart-Halajko, M. C., A. Kelley-Hedgepeth, M. C. Montefusco, J. A. Cooper, by regulating expression of angplt4 can perform endocrine func- A. Kopin, J. M. McCaffrey, A. Balasubramanyam, H. J. Pownall, D. M. Nathan, tions, which may include modulation of angiogenesis, wound I. Peter, et al. 2011. ANGPTL4 variants E40K and T266M are associated with

lower fasting triglyceride levels in Non-Hispanic White Americans from the by guest on September 28, 2021 healing, enhancing glucose homeostasis and insulin sensitivity, Look AHEAD Clinical Trial. BMC Med. Genet. 12: 89. and induction of lipogenesis in the adipose tissue. 17. Xu, A., M. C. Lam, K. W. Chan, Y. Wang, J. Zhang, R. L. Hoo, J. Y. Xu, The interest of healthcare providers in A1AT preparations recently B. Chen, W. S. Chow, A. W. Tso, and K. S. Lam. 2005. Angiopoietin-like protein 4 decreases blood glucose and improves glucose tolerance but induces hyperlipidemia has increased because of the beneficial effects of A1AT therapy in and hepatic steatosis in mice. Proc. Natl. Acad. Sci. USA 102: 6086–6091. single cases and in small cohorts with clinical conditions other than 18. Goh, Y. Y., M. Pal, H. C. Chong, P. Zhu, M. J. Tan, L. Punugu, C. R. Lam, lung emphysema (34). Hence, our results provide the framework for Y. H. Yau, C. K. Tan, R. L. Huang, et al. 2010. Angiopoietin-like 4 interacts with beta1 and beta5 to modulate keratinocyte migration. Am. J. Pathol. 177: further studies addressing how A1AT interacts with specific cells to 2791–2803. regulate angptl4 expression and how A1AT-induced angptl4 19. Goh, Y. Y., M. Pal, H. C. Chong, P. Zhu, M. J. Tan, L. Punugu, C. K. Tan, R. L. Huang, S. K. Sze, M. B. Tang, et al. 2010. Angiopoietin-like 4 interacts expression relates with inflammation and disease progression. with matrix proteins to modulate wound healing. J. Biol. Chem. 285: 32999– 33009. 20. Stapleton, C. M., J. H. Joo, Y. S. Kim, G. Liao, R. A. Panettieri, Jr., and Acknowledgments A. M. Jetten. 2010. Induction of ANGPTL4 expression in human airway smooth We thank laboratory technician Helena Lickei for excellent technical assis- muscle cells by PMA through activation of PKC and MAPK pathways. Exp. Cell Res. 316: 507–516. tance and Nupur Aggarwal for help performing assays with human neutro- 21. Wang, Y., H. Chen, H. Li, J. Zhang, and Y. Gao. 2013. Effect of angiopoietin-like phils and endothelial cells. We also thank colleagues at the Institute for protein 4 on rat pulmonary microvascular endothelial cells exposed to LPS. Int. Transfusion Medicine for providing laboratory facilities. J. Mol. Med. 32: 568–576. 22. Belanger, A. J., H. Lu, T. Date, L. X. Liu, K. A. Vincent, G. Y. Akita, S. H. Cheng, R. J. Gregory, and C. Jiang. 2002. Hypoxia up-regulates expres- Disclosures sion of peroxisome proliferator-activated receptor g angiopoietin-related gene The authors have no financial conflicts of interest. (PGAR) in cardiomyocytes: role of hypoxia inducible factor 1a. J. Mol. Cell. Cardiol. 34: 765–774. 23. Kaddatz, K., T. Adhikary, F. Finkernagel, W. Meissner, S. Muller-Br€ usselbach,€ and R. Muller.€ 2010. Transcriptional profiling identifies functional interactions References of TGF b and PPAR b/d signaling: synergistic induction of ANGPTL4 1. Zhu, P., Y. Y. Goh, H. F. Chin, S. Kersten, and N. S. Tan. 2012. Angiopoietin-like transcription. J. Biol. Chem. 285: 29469–29479. 4: a decade of research. Biosci. Rep. 32: 211–219. 24.Koliwad,S.K.,T.Kuo,L.E.Shipp,N.E.Gray,F.Backhed,A.Y.So,R.V.Farese, 2. Kim, I., H. G. Kim, H. Kim, H. H. Kim, S. K. Park, C. S. Uhm, Z. H. Lee, and Jr., and J. C. Wang. 2009. Angiopoietin-like 4 (ANGPTL4, fasting-induced adipose G. Y. Koh. 2000. Hepatic expression, synthesis and secretion of a novel factor) is a direct glucocorticoid receptor target and participates in glucocorticoid- fibrinogen/angiopoietin-related protein that prevents endothelial-cell apoptosis. regulated triglyceride metabolism. J. Biol. Chem. 284: 25593–25601. Biochem. J. 346: 603–610. 25. Stockert, J., T. Adhikary, K. Kaddatz, F. Finkernagel, W. Meissner, S. Muller-€ 3. Kersten, S., S. Mandard, N. S. Tan, P. Escher, D. Metzger, P. Chambon, Brusselbach,€ and R. Muller.€ 2011. Reverse crosstalk of TGFb and PPARb/d F. J. Gonzalez, B. Desvergne, and W. Wahli. 2000. Characterization of the signaling identified by transcriptional profiling. Nucleic Acids Res. 39: 119–131. fasting-induced adipose factor FIAF, a novel peroxisome proliferator-activated 26. Wiesner, G., R. E. Brown, G. S. Robertson, S. A. Imran, E. Ur, and receptor target gene. J. Biol. Chem. 275: 28488–28493. M. Wilkinson. 2006. Increased expression of the adipokine genes resistin and 4. Le Jan, S., C. Amy, A. Cazes, C. Monnot, N. Lamande´, J. Favier, J. Philippe, fasting-induced adipose factor in hypoxic/ischaemic mouse brain. Neuroreport M. Sibony, J. M. Gasc, P. Corvol, and S. Germain. 2003. Angiopoietin-like 4 is 17: 1195–1198. 5362 A1AT IS AN INDUCER OF angptl4

27. Yoon, J. C., T. W. Chickering, E. D. Rosen, B. Dussault, Y. Qin, A. Soukas, 43. Zhang, Q. L., B. R. Cui, H. Y. Li, P. Li, L. Hong, L. P. Liu, D. Z. Ding, and J. M. Friedman, W. E. Holmes, and B. M. Spiegelman. 2000. Peroxisome X. Cui. 2013. MAPK and PI3K pathways regulate hypoxia-induced atrial na- proliferator-activated receptor gamma target gene encoding a novel triuretic peptide secretion by controlling HIF-1a expression in beating rabbit angiopoietin-related protein associated with adipose differentiation. Mol. Cell. atria. Biochem. Biophys. Res. Commun. 438: 507–512. Biol. 20: 5343–5349. 44. Subramaniyam, D., H. Zhou, M. Liang, T. Welte, R. Mahadeva, and 28. Georgiadi, A., Y. Wang, R. Stienstra, N. Tjeerdema, A. Janssen, A. Stalenhoef, S. Janciauskiene. 2010. Cholesterol rich lipid raft microdomains are gateway for J. A. van der Vliet, A. de Roos, J. T. Tamsma, J. W. Smit, et al. 2013. Over- acute phase protein, SERPINA1. Int. J. Biochem. Cell Biol. 42: 1562–1570. expression of angiopoietin-like protein 4 protects against atherosclerosis devel- 45. Mandard, S., F. Zandbergen, N. S. Tan, P. Escher, D. Patsouris, W. Koenig, opment. Arterioscler. Thromb. Vasc. Biol. 33: 1529–1537. R. Kleemann, A. Bakker, F. Veenman, W. Wahli, et al. 2004. The direct per- 29. Aldonyte, R., T. E. Hutchinson, B. Jin, M. Brantly, E. Block, J. Patel, and oxisome proliferator-activated receptor target fasting-induced adipose factor J. Zhang. 2008. Endothelial a-1-antitrypsin attenuates cigarette smoke induced (FIAF/PGAR/ANGPTL4) is present in blood plasma as a truncated protein that apoptosis in vitro. COPD 5: 153–162. is increased by fenofibrate treatment. J. Biol. Chem. 279: 34411–34420. 30. Aldonyte, R., L. Jansson, and S. Janciauskiene. 2004. Concentration-dependent 46. Takahashi, H., K. Fujita, T. Fujisawa, K. Yonemitsu, A. Tomimoto, I. Ikeda, effects of native and polymerised a1-antitrypsin on primary human monocytes, M. Yoneda, T. Masuda, K. Schaefer, L. J. Saubermann, et al. 2006. Inhibition in vitro. BMC Cell Biol. 5: 11. of peroxisome proliferator-activated receptor g activity in esophageal carci- 31. Huang, H., S. C. Campbell, T. Nelius, D. F. Bedford, D. Veliceasa, N. P. Bouck, noma cells results in a drastic decrease of invasive properties. Cancer Sci. 97: and O. V. Volpert. 2004. a1-Antitrypsin inhibits angiogenesis and tumor growth. 854–860. Int. J. Cancer 112: 1042–1048. 47. Staiger, H., C. Haas, J. Machann, R. Werner, M. Weisser, F. Schick, F. Machicao, 32. Janciauskiene, S., S. Wrenger, and T. Welte. 2012. Conformational and functional N. Stefan, A. Fritsche, and H. U. Ha¨ring. 2009. Muscle-derived angiopoietin-like polymorphism of a1-antitrypsin: role in health and diseases. In Serine Proteases: protein 4 is induced by fatty acids via peroxisome proliferator-activated receptor Mechanism, Structure and Evolution. Isamu Chiba, and Takao Kamio, eds. Nova (PPAR)-delta and is of metabolic relevance in humans. Diabetes 58: 579–589. Publishers, Hauppauge, New York, p. 1–84 48. Robciuc, M. R., E. Tahvanainen, M. Jauhiainen, and C. Ehnholm. 2010. 33. Van Molle, W., C. Libert, W. Fiers, and P. Brouckaert. 1997. Alpha 1-acid Quantitation of serum angiopoietin-like proteins 3 and 4 in a Finnish population glycoprotein and a1-antitrypsin inhibit TNF-induced but not anti‑Fas-induced sample. J. Lipid Res. 51: 824–831. apoptosis of hepatocytes in mice. J. Immunol. 159: 3555–3564. 49. Sukonina, V., A. Lookene, T. Olivecrona, and G. Olivecrona. 2006. Angiopoietin- 34. Sabina, J., and W. Tobias. 2013. Augmentation therapy with a1-antitrypsin: like protein 4 converts lipoprotein lipase to inactive monomers and modulates li- Downloaded from novel perspectives. Cardiovasc. Hematol. Disord. Drug Targets 13: 90–98. pase activity in adipose tissue. Proc. Natl. Acad. Sci. USA 103: 17450–17455. 35. Talmud, P. J., S. Martin, G. Steiner, D. M. Flavell, D. B. Whitehouse, S. Nagl, 50. Yau, M. H., Y. Wang, K. S. Lam, J. Zhang, D. Wu, and A. Xu. 2009. A highly R. Jackson, M. R. Taskinen, M. H. Frick, M. S. Nieminen, et al. 2003. Pro- conserved motif within the NH2-terminal coiled-coil domain of angiopoietin-like gression of atherosclerosis is associated with variation in the a1-antitrypsin protein 4 confers its inhibitory effects on lipoprotein lipase by disrupting the gene. Arterioscler. Thromb. Vasc. Biol. 23: 644–649. enzyme dimerization. J. Biol. Chem. 284: 11942–11952. 36. Kalis, M., R. Kumar, S. Janciauskiene, A. Salehi, and C. M. Cilio. 2010. a 1- 51. Li, M., T. W. Lee, A. P. Yim, T. S. Mok, and G. G. Chen. 2006. Apoptosis in- Antitrypsin enhances insulin secretion and prevents cytokine-mediated apoptosis duced by troglitazone is both peroxisome proliferator-activated receptor-gamma-

in pancreatic b-cells. Islets 2: 185–189. and ERK-dependent in human non-small lung cancer cells. J. Cell. Physiol. 209: http://www.jimmunol.org/ 37. Petrache, I., I. Fijalkowska, T. R. Medler, J. Skirball, P. Cruz, L. Zhen, 428–438. H. I. Petrache, T. R. Flotte, and R. M. Tuder. 2006. alpha-1 antitrypsin inhibits caspase- 52. Prusty, D., B. H. Park, K. E. Davis, and S. R. Farmer. 2002. Activation of MEK/ 3 activity, preventing lung endothelial cell apoptosis. Am. J. Pathol. 169: 1155–1166. ERK signaling promotes adipogenesis by enhancing peroxisome proliferator- 38. Fregonese, L., and J. Stolk. 2008. Hereditary a-1-antitrypsin deficiency and its activated receptor gamma (PPARg) and C/EBPa gene expression during the clinical consequences. Orphanet J. Rare Dis. 3: 16. differentiation of 3T3-L1 preadipocytes. J. Biol. Chem. 277: 46226–46232. 39. Lewis, E. C. 2012. Expanding the clinical indications for a(1)-antitrypsin ther- 53. Knowles, H. J., A. M. Cleton-Jansen, E. Korsching, and N. A. Athanasou. 2010. apy. Mol. Med. 18: 957–970. Hypoxia-inducible factor regulates osteoclast-mediated bone resorption: role of 40. Marcondes, A. M., X. Li, L. Tabellini, M. Bartenstein, J. Kabacka, G. E. Sale, angiopoietin-like 4. FASEB J. 24: 4648–4659. J. A. Hansen, C. A. Dinarello, and H. J. Deeg. 2011. Inhibition of IL-32 acti- 54. Wenger, R. H. 2000. Mammalian oxygen sensing, signalling and gene regula- vation by a-1 antitrypsin suppresses alloreactivity and increases survival in an tion. J. Exp. Biol. 203: 1253–1263. allogeneic murine marrow transplantation model. Blood 118: 5031–5039. 55. Conrad, P. W., T. L. Freeman, D. Beitner-Johnson, and D. E. Millhorn. 1999. € 41. Jonigk, D., M. Al-Omari, L. Maegel, M. Muller, N. Izykowski, J. Hong, EPAS1 trans-activation during hypoxia requires p42/p44 MAPK. J. Biol. Chem. by guest on September 28, 2021 K. Hong, S. H. Kim, M. Dorsch, R. Mahadeva, et al. 2013. Anti-inflammatory 274: 33709–33713. and immunomodulatory properties of a1-antitrypsin without inhibition of elas- 56. Richard, D. E., E. Berra, E. Gothie´, D. Roux, and J. Pouysse´gur. 1999. p42/p44 tase. Proc. Natl. Acad. Sci. USA 110: 15007–15012. mitogen-activated protein kinases phosphorylate hypoxia-inducible factor 1a 42. Al-Omari, M., E. Korenbaum, M. Ballmaier, U. Lehmann, D. Jonigk, (HIF-1a) and enhance the transcriptional activity of HIF-1. J. Biol. Chem. 274: D. J. Manstein, T. Welte, R. Mahadeva, and S. Janciauskiene. 2011. Acute-phase 32631–32637. protein a1-antitrypsin inhibits neutrophil calpain I and induces random migra- 57. Tuder, R. M., S. M. Janciauskiene, and I. Petrache. 2010. Lung disease associ- tion. Mol. Med. 17: 865–874. ated with a1-antitrypsin deficiency. Proc. Am. Thorac. Soc. 7: 381–386. Supplemental Fig. 1

12 1mg/ml A1AT p<0.001

10

8 p=0.004

6 p=0.029

(fold change) 4

mRNA ofmRNA angptl4 at h 1

2

0 Control Prolastin Aralast A1AT Sigma

Supplemental Figure 1. Adherent PBMCs were incubated with 1 mg/ml A1AT from different sources (Prolastin, Grifols; Aralast, Baxter; or Sigma Aldrich) for 1h and mRNA levels of angplt4 were determined by RT-PCR. Bars represent the mean ± SD of four experiments.

Supplemental Fig. 2

12 8 h p=0.005 10

8

6

mRNA angplt4 mRNA change Fold 4

2

0 Control A1AT 0.5 mg/ml

Supplemental Figure 2. HMVEC-L were incubated with 0.5 mg/ml A1AT (Calbiochem) for

8 h and mRNA levels of angptl4 were determined by RT-PCR. Bars represent the mean ±

SEM of three experiments.

Supplemental Table I. Influence of A1AT (1 mg/ml, 8 h) on expression of angiogenesis genesa in HMVEC-L

Unigene Symbol Description Gname fold changeb Hs.525622 AKT1 V-akt murine thymoma viral AKT/MGC99656/PKB/PKB- 0,94 0,14 oncogene homolog 1 ALPHA/PRKBA/RAC/RAC- ALPHA Hs.369675 ANGPT1 AGP1/AGPT/ANG1 3,61 5,11 Hs.583870 ANGPT2 Angiopoietin 2 AGPT2/ANG2 0,98 0,01 Hs.209153 ANGPTL3 Angiopoietin-like 3 ANGPT5/FHBL2 1,10 0,59 Hs.9613 ANGPTL4 Angiopoietin-like 4 ANGPTL2/ARP4/FIAF/HFAR 5,79 1,66 P/NL2/PGAR/pp1158 Hs.1239 ANPEP Alanyl (membrane) APN/CD13/GP150/LAP1/P150 1,03 0,01 aminopeptidase /PEPN Hs.194654 BAI1 Brain-specific angiogenesis FLJ41988/GDAIF 0,89 0,26 inhibitor 1 Hs.54460 CCL11 Chemokine (C-C motif) ligand 11 MGC22554/SCYA11 1,04 0,15 Hs.303649 CCL2 Chemokine (C-C motif) ligand 2 GDCF- 1,09 0,14 2/HC11/HSMCR30/MCAF/M CP- 1/MCP1/MGC9434/SCYA2/S MC-CF Hs.76206 CDH5 Cadherin 5, type 2 (vascular 7B4/CD144/FLJ17376 0,96 0,14 endothelium) Hs.517356 COL18A1 Collagen, type XVIII, alpha 1 FLJ27325/FLJ34914/KNO/KN 0,99 0,09 O1/KS/MGC74745 Hs.570065 COL4A3 Collagen, type IV, alpha 3 - 0,62 0,51 (Goodpasture antigen) Hs.789 CXCL1 Chemokine (C-X-C motif) ligand 1 FSP/GRO1/GROa/MGSA/MG 0,93 0,09 (melanoma growth stimulating SA-a/NAP-3/SCYB1 activity, alpha) Hs.632586 CXCL10 Chemokine (C-X-C motif) ligand C7/IFI10/INP10/IP- 1,49 0,56 10 10/SCYB10/crg-2/gIP-10/mob- 1 Hs.89690 CXCL3 Chemokine (C-X-C motif) ligand 3 CINC-2b/GRO3/GROg/MIP- 0,66 0,23 2b/MIP2B/SCYB3 Hs.89714 CXCL5 Chemokine (C-X-C motif) ligand 5 ENA-78/SCYB5 1,54 0,36 Hs.164021 CXCL6 Chemokine (C-X-C motif) ligand 6 CKA-3/GCP-2/GCP2/SCYB6 0,90 0,13 (granulocyte chemotactic protein 2) Hs.77367 CXCL9 Chemokine (C-X-C motif) ligand 9 CMK/Humig/MIG/SCYB9/crg 0,45 0,33 -10 Hs.592212 TYMP Thymidine phosphorylase ECGF/ECGF1/MEDPS1/MNG 1,26 0,06 IE/MTDPS1/PDECGF/TP/hPD -ECGF Hs.154210 S1PR1 Sphingosine-1-phosphate receptor CHEDG1/D1S3362/ECGF1/E 1,02 0,25 1 DG-1/EDG1/FLJ58121/S1P1 Hs.516664 EFNA1 -A1 B61/ECKLG/EFL1/EPLG1/LE 0,80 0,11 RK1/TNFAIP4 Hs.516656 EFNA3 Ephrin-A3 EFL2/EPLG3/Ehk1-L/LERK3 0,87 0,23 Hs.149239 EFNB2 Ephrin-B2 EPLG5/HTKL/Htk- 0,79 0,17 L/LERK5/MGC126226/MGC1 26227/MGC126228 Hs.419815 EGF HOMG4/URG 1,01 0,36 Hs.76753 ENG Endoglin CD105/END/FLJ41744/HHT1/ 1,07 0,01 ORW/ORW1 Hs.437008 EPHB4 EPH receptor B4 HTK/MYK1/TYRO11 0,89 0,13 Hs.115263 EREG ER 1,50 0,10 Hs.483635 FGF1 1 (acidic) AFGF/ECGF/ECGF- 0,89 0,55 beta/ECGFA/ECGFB/FGF- alpha/FGFA/GLIO703/HBGF1 Hs.284244 FGF2 Fibroblast growth factor 2 (basic) BFGF/FGFB/HBGF-2 1,01 0,17 Hs.1420 FGFR3 Fibroblast 3 ACH/CD333/CEK2/HSFGFR3 1,36 0,29 EX/JTK4 Hs.11392 FIGF C-fos induced growth factor VEGF-D/VEGFD 1,37 0,29 (vascular endothelial growth factor D) Hs.654360 FLT1 Fms-related 1 FLT/VEGFR1 1,10 0,14 (vascular endothelial growth factor/vascular permeability factor receptor) Hs.388245 HAND2 Heart and neural crest derivatives DHAND2/FLJ16260/Hed/MG 3,13 3,75 expressed 2 C125303/MGC125304/Thing2/ bHLHa26/dHand Hs.396530 HGF DFNB39/F- 1,12 0,38 (hepapoietin A; scatter factor) TCF/HGFB/HPTA/SF Hs.597216 HIF1A Hypoxia inducible factor 1, alpha HIF-1alpha/HIF1/HIF1- 1,12 0,07 subunit (basic helix-loop-helix ALPHA/MOP1/PASD8/bHLH transcription factor) e78 Hs.44227 HPSE Heparanase HPA/HPA1/HPR1/HPSE1/HS 1,04 0,12 E1 Hs.504609 ID1 Inhibitor of DNA binding 1, ID/bHLHb24 0,71 0,18 dominant negative helix-loop-helix protein Hs.76884 ID3 Inhibitor of DNA binding 3, HEIR-1/bHLHb25 0,76 0,16 dominant negative helix-loop-helix protein Hs.37026 IFNA1 Interferon, alpha 1 IFL/IFN/IFN-ALPHA/IFN- 1,42 1,14 alphaD/IFNA13/IFNA@/MGC 138207/MGC138505/MGC138 507 Hs.93177 IFNB1 Interferon, beta 1, fibroblast IFB/IFF/IFNB/MGC96956 2,70 1,76 Hs.856 IFNG Interferon, gamma IFG/IFI 1,05 0,14 Hs.160562 IGF1 Insulin-like growth factor 1 IGF-I/IGF1A/IGFI 1,22 0,15 (somatomedin C) Hs.126256 IL1B 1, beta IL-1/IL1-BETA/IL1F2 1,52 0,41 Hs.654458 IL6 Interleukin 6 (interferon, beta 2) BSF2/HGF/HSF/IFNB2/IL-6 0,89 0,20 Hs.624 IL8 Interleukin 8 CXCL8/GCP- 1,10 0,21 1/GCP1/LECT/LUCT/LYNAP /MDNCF/MONAP/NAF/NAP- 1/NAP1 Hs.436873 ITGAV Integrin, alpha V (vitronectin CD51/DKFZp686A08142/MS 1,14 0,14 receptor, alpha polypeptide, K8/VNRA antigen CD51) Hs.218040 ITGB3 Integrin, beta 3 (platelet CD61/GP3A/GPIIIa 1,07 0,09 glycoprotein IIIa, antigen CD61) Hs.728907 JAG1 Jagged 1 AGS/AHD/AWS/CD339/HJ1/J 1,36 0,30 AGL1/MGC104644 Hs.479756 KDR Kinase insert domain receptor (a CD309/FLK1/VEGFR/VEGFR 0,96 0,12 type III ) 2 Hs.473256 LAMA5 Laminin, alpha 5 KIAA1907 1,11 0,19 Hs.421391 LECT1 Leukocyte cell derived chemotaxin BRICD3/CHM- 1,92 2,26 1 I/CHM1/MYETS1 Hs.194236 LEP Leptin FLJ94114/OB/OBS 1,05 0,14 Hs.82045 MDK (neurite growth- FLJ27379/MK/NEGF2 0,92 0,19 promoting factor 2) Hs.513617 MMP2 Matrix metallopeptidase 2 CLG4/CLG4A/MMP- 0,95 0,08 (gelatinase A, 72kDa gelatinase, II/MONA/TBE-1 72kDa type IV collagenase) Hs.297413 MMP9 Matrix metallopeptidase 9 CLG4B/GELB/MANDP2/MM 1,03 0,52 (gelatinase B, 92kDa gelatinase, P-9 92kDa type IV collagenase) Hs.436100 NOTCH4 Notch 4 FLJ16302/INT3/MGC74442/N 0,96 0,08 OTCH3 Hs.131704 NRP1 Neuropilin 1 BDCA4/CD304/DKFZp686A0 0,98 0,16 3134/DKFZp781F1414/NP1/N RP/VEGF165R Hs.471200 NRP2 Neuropilin 2 MGC126574/NP2/NPN2/PRO 1,08 0,18 2714/VEGF165R2 Hs.535898 PDGFA Platelet-derived growth factor PDGF-A/PDGF1 1,11 0,06 alpha polypeptide Hs.514412 PECAM1 Platelet/endothelial cell adhesion CD31/FLJ34100/FLJ58394/PE 1,10 0,29 molecule CAM-1 Hs.81564 PF4 CXCL4/MGC138298/SCYB4 0,72 0,30 Hs.252820 PGF D12S1900/PGFL/PLGF/PlGF- 1,06 0,21 2/SHGC-10760 Hs.77274 PLAU Plasminogen activator, urokinase ATF/UPA/URK/u-PA 1,12 0,13 Hs.143436 PLG Plasminogen DKFZp779M0222 0,76 0,35 Hs.125036 PLXDC1 Plexin domain containing 1 DKFZp686F0937/FLJ36270/F 0,49 0,13 LJ45632/TEM3/TEM7 Hs.528665 PROK2 Prokineticin 2 BV8/KAL4/MIT1/PK2 1,03 0,24 Hs.201978 PTGS1 Prostaglandin-endoperoxide COX1/COX3/PCOX1/PGG/H 0,73 0,12 synthase 1 (prostaglandin G/H S/PGHS- synthase and cyclooxygenase) 1/PGHS1/PHS1/PTGHS Hs.532768 SERPINF1 Serpin peptidase inhibitor, clade F EPC-1/PEDF 1,08 0,12 (alpha-2 antiplasmin, pigment epithelium derived factor), member 1 Hs.68061 SPHK1 Sphingosine kinase 1 SPHK 0,98 0,16 Hs.301989 STAB1 Stabilin 1 CLEVER-1/FEEL-1/FELE- 1,03 0,08 1/FEX1/KIAA0246/STAB-1 Hs.89640 TEK TEK tyrosine kinase, endothelial CD202B/TIE- 0,91 0,04 2/TIE2/VMCM/VMCM1 Hs.170009 TGFA Transforming growth factor, alpha TFGA 0,82 0,29 Hs.645227 TGFB1 Transforming growth factor, beta 1 CED/DPD1/LAP/TGFB/TGFb 0,96 0,02 eta Hs.133379 TGFB2 Transforming growth factor, beta 2 MGC116892/TGF-beta2 0,90 0,14 Hs.494622 TGFBR1 Transforming growth factor, beta AAT5/ACVRLK4/ALK- 1,28 0,03 receptor 1 5/ALK5/LDS1A/LDS2A/SKR 4/TGFR-1 Hs.164226 THBS1 1 THBS/THBS-1/TSP/TSP- 1,09 0,15 1/TSP1 Hs.371147 THBS2 Thrombospondin 2 TSP2 0,98 0,06 Hs.522632 TIMP1 TIMP metallopeptidase inhibitor 1 CLGI/EPA/EPO/FLJ90373/HC 1,01 0,14 I/TIMP Hs.633514 TIMP2 TIMP metallopeptidase inhibitor 2 CSC-21K 0,88 0,05 Hs.644633 TIMP3 TIMP metallopeptidase inhibitor 3 HSMRK222/K222/K222TA2/S 2,79 2,95 FD Hs.241570 TNF Tumor necrosis factor DIF/TNF- 1,06 0,10 alpha/TNFA/TNFSF2 Hs.525607 TNFAIP2 Tumor necrosis factor, alpha- B94/EXOC3L3 1,19 0,12 induced protein 2 Hs.73793 VEGFA Vascular endothelial growth factor MGC70609/MVCD1/VEGF/V 1,02 0,34 A PF Hs.435215 VEGFC Vascular endothelial growth factor Flt4-L/VRP 1,10 0,17 C Hs.534255 B2M Beta-2-microglobulin - 0,99 0,08 Hs.412707 HPRT1 Hypoxanthine HGPRT/HPRT 1,04 0,14 phosphoribosyltransferase 1 Hs.728776 RPL13A Ribosomal protein L13a L13A/TSTA1 0,96 0,03 Hs.592355 GAPDH Glyceraldehyde-3-phosphate G3PD/GAPD/MGC88685 0,93 0,05 dehydrogenase Hs.520640 ACTB Actin, beta PS1TP5BP1 1,10 0,12 aGene expression was analyzed using angiogenesis PCR arrays. bData represent mean ± SD (n=3).