In-Depth Proteomics Identifies a Role for Autophagy in Controlling

Subscriber access provided by Weizmann Institute of Science

Article

In-depth proteomics identifies a role for autophagy in controlling ROS-mediated endothelial permeability Francesca Patella, Lisa J Neilson, Dimitris Athineos, Zahra Erami, Kurt Anderson, Karen Blyth, Kevin M. Ryan, and Sara Zanivan J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00166 • Publication Date (Web): 01 Jun 2016 Downloaded from http://pubs.acs.org on June 8, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

In-depth proteomics identifies a role for autophagy in

controlling

ROS-mediated

endothelial

permeability Francesca Patella1, Lisa J Neilson1, Dimitris Athineos1, Zahra Erami1, Kurt Anderson1, Karen Blyth1, Kevin M Ryan1 and Sara Zanivan1,* 1

Cancer Research UK Beatson Institute, Glasgow G611BD, UK

*Correspondence: [email protected] , tel. +44(0)141 330 3971

Running Title: Autophagy controls endothelial permeability

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Abstract Endothelial cells (ECs) form the inner layer of blood vessels and physically separate the blood from the surrounding tissue. To support tissues with nutrients and oxygen, the endothelial monolayer is semipermeable. When EC permeability is altered, blood vessels are not functional and this is associated with disease. A comprehensive knowledge of the mechanisms regulating EC permeability is key in developing strategies to target this mechanism in pathologies. Here we have used an in vitro model of human umbilical vein endothelial cells mimicking the formation of a physiologically permeable vessel and performed time-resolved in-depth molecular profiling using SILAC MS-proteomics. Autophagy is induced when ECs are assembled into a physiologically permeable monolayer. By using siRNA and drug treatment to block autophagy in combination with functional assays and MS proteomics, we show that ECs require autophagy flux to maintain intracellular reactive oxygen species levels and this is required to maintain the physiological permeability of the cells.

Keywords Proteomics, SILAC, endothelial cell, contact inhibition, autophagy, permeability, ROS .


Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Endothelial cells (ECs) line the inner wall of blood vessels as a semipermeable monolayer that physically separates the blood from the surrounding tissues 1. The maintenance of endothelial cell permeability is critical to the physiological function of blood vessels in distributing oxygen and nutrients throughout the entire organism. In the mature vasculature, the endothelial monolayer is constituted of quiescent cells 2, which are tightly connected to each other through cell-cell adhesion protein complexes called tight junctions and adherens junctions 3. The disruption of these junctions results in a leaky vasculature, which is a hallmark

of

diseases,

such

as cancer

4,5

.

Several mechanisms

regulating

EC

hyperpermeability have already been well characterized. For example, upon binding to their receptors, the vascular endothelial growth factor (VEGF)

6,7

and the pro-inflammatory factors

thrombin and histamine8,9 activate Ca2+, Rho GTPase/ROCK, and myosin light chain kinase signaling pathways. Those pathways, in turn, induce the disruption of intercellular junctions between ECs. We have recently shown that EC metabolism is a key controller of vascular permeability, particularly that fatty acid oxidation is required to maintain the physiological permeability of the endothelial cells

10

. This highlights the complexity of the mechanisms

regulating vascular permeability, and suggests that other pathways may play a key role in this process. To address this question, we have applied unbiased time-resolved MS-based proteomics to an in vitro model where endothelial cells, cultured for one week with growth factor stimulation, assemble into a monolayer

11

. This model recapitulates aspects of the

formation of a physiologically permeable blood vessel. Autophagy is a catabolic process that leads to the degradation of cellular proteins and organelles, and it is crucial to maintain cellular homeostasis. Autophagy is often induced when cells are under stress conditions triggered by stimuli such as reactive oxygen species (ROS), amino acid starvation and hypoxia. For example, autophagy can generate free amino acids and fatty acids, which are used by the cells as building blocks to survive under nutrient limitation

12

. Autophagy is a multistep process whereby targeted cytoplasmic components 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

are engulfed by double membrane organelles called autophagosomes

13

. The fusion of an

autophagosome to a lysosome generates an autolysosome, which degrades its contents by virtue of its acidic pH and the presence of lytic enzymes

14,15

. Autophagy (ATG)-related

proteins tightly regulate the autophagy process. As an example, ATG5 controls the formation of autophagosomes. Notably, ATG proteins can be targeted to inhibit autophagy 12. Autophagy is active in ECs in vitro and in vivo, but its functional role in ECs is quite controversial. In human umbilical vein endothelial cells (HUVECs) autophagy is induced upon silencing of VEGF and this leads to cell death

16

. Conversely, autophagy promotes cell

survival in response to hypoxia in bovine aortic ECs (BAECs) cultured under glucosedeprived conditions

17

. Autophagy has also been implicated in regulating vessel growth. By

manipulating the expression of ATG5, it has been shown that autophagy enhances the ability of nutrient deprived BAEC to assemble in a capillary-like network when cultured on matrigel

18

. However, HUVECs silenced for ATG5 do not have any sprouting defects when 19

embedded into a three dimensional gel

. In contrast, inhibition of autophagy by silencing

ATG5 in HUVECs enhances their assembly into network-like structures when co-cultured with neuroblastoma cells

20

. In vivo, EC-specific deletion of Atg5 or Atg7 in mice did not

affect postnatal retinal angiogenesis

21 65

. However it decreased von Willebrand factor (vWF)

release upon epinephrine stimulation, with a consequent prolongation of bleeding time

21

.

Finally, EC-deletion of Atg5 in mice bearing a graft of B16-F10 melanoma cells produced tumors with smaller, more numerous and tortuous vessels

19

. Hence autophagy is necessary

for the regulation of endothelial functions. By in depth measurement of the proteomes of HUVECs cultured either sparsely or tightly confluent in the presence of nutrients and growth factors, we have discovered that autophagy levels are higher in a fully formed endothelial monolayer. We show that inhibiting autophagy in ECs, either pharmacologically or by silencing ATG5 with siRNA, impairs their permeability. Bioinformatic analysis of proteomic changes occurring upon inhibition of autophagy in ECs, predicted ROS as an upstream regulator of the measured changes. 4 ACS Paragon Plus Environment

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Finally, we show that HUVECs require autophagy to control ROS levels in order to maintain their permeability.

Experimental Section Cell culture and transfection. Human Umbilical Vein Endothelial Cells (HUVECs) isolated from 2-5 umbilical cords were pooled together and cultured in EGM-2 (Lonza, Basel, Switzerland) until passage 6 on 1% gelatin-coated dishes. Cells were nucleofected with 60 pmol of si-ATG5 (RNA-Stealth, pool three single siRNAs, Thermo Fisher Scientific) or si-NC (non-targeting siRNA, Thermo Fisher Scientific) per 106 cells using Amaxa nucleofector and nucleofector kit (Lonza) and experiments were performed 4 days after transfection. SILAC labeling. To generate the EC SILAC standard, HUVECs were cultured in custommade EGM-2 without arginine and lysine (Lonza) supplemented with and

13

13

C615N4 L-arginine,

C615N2 L-lysine (SILAC heavy) (Cambridge Isotope Laboratories) for 3 passages,

corresponding to more than 97% of heavy aminoacid incorporation. Cell lysis and sample preparation for proteomic analysis. Sample lysate in 2% SDS in 0.1 M Tris HCl buffer pH 7.4 was mixed with an equal protein amount (125 µg for the confluency experiment, 5 µg for the bafilomycin experiment, 65 µg for the si-ATG5 experiment) of SILAC internal standard. For the confluency and si-ATG5 experiment, three replicates (cells cultured separately and processed on the same day) were analyzed. Proteins were trypsin (sequencing grade modified, Promega) digested using the filter-aided sample preparation protocol

22

and

peptides fractionated into six fractions using pipette tip strong anion exchange separation microcolumn 23. For the bafilomycin experiment, three replicates (cells cultured separately and processed on the same day) were analyzed. Proteins were precipitated in methanol-chloroform, 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

resuspended in urea buffer (8M urea, 75mM NaCl, 50mM Tris), reduced with dithiothreitol, and alkylated with iodoacetamide. Endoproteinase Lys-C (Waco) followed by trypsin were used to digest the proteins (enzyme to protein ratio = 1:50). For the conditioned medium analysis, ECs were grown until confluence in EGM-2 medium. They were washed in PBS with Ca2+ and Mg2+ and treated or not with bafilomycin 100 nM in EGM-2 without serum. After 4 hours the supernatant was collected and spun at 4°C (300 g for 10 minutes, followed by 2,000 g for 10 minutes and 10,000 g for 30 minutes). The proteins in the supernatant were extracted as previously described

24

. Collected proteins

were precipitated with methanol and chloroform, resuspended in 8 M urea buffer, reduced, alkylated and Lys-C and trypsin digested. Peptides were desalted by using C18 StageTip

25

and eluted in 80% acetonitrile, 0.5% acetic

acid and stored at -20 °C until analyzed at the MS. Mass spectrometry analysis. Proteomic MS analysis was performed using a linear trap quadrupole (LTQ)-Orbitrap Elite (Thermo Fisher Scientific), operated in the high energy collision dissociation (HCD) fragmentation mode, coupled on-line with a nano-liquid chromatography (nLC) (EasynLC, Thermo Fisher Scientific) as described previously

10

. The

tryptic peptides were separated on a 20 cm reverse phase column packed with 1.9 µm C18 resin (Dr.Maisch, GmbH, Ammerbuch-Entringen, Germany) using a flow of 200 nl/min in 90 min gradient from 5% to 25% ACN in 0.5% acetic acid. The MS spectra were acquired in the Orbitrap analyzer at a resolution of 120,000 at 400 m/z, and a target value of 106 charges. HCD fragmentation of the 10 most intense ions was performed using a target value of 40,000 charges which were acquired in the Orbitrap at a resolution 15,000 at 400 m/z. Data were acquired with Xcalibur software. MS proteomic data analysis. MS data were processed using the MaxQuant computational platform

26

(version 1.3.6.0 for the cell confluency proteomes and 1.3.8.2 for the autophagy-

inhibited proteomes) and searched with Andromeda search engine

27

against the human 6

ACS Paragon Plus Environment

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


UniProt database (release-2012 01, 88,847 entries). The ‘match between runs’ 26 option was enabled. An initial maximal mass deviation of 7 ppm and 20 ppm was required to search for precursor and fragment ions, respectively. Trypsin with full enzyme specificity and peptides with a minimum length of seven amino acids were selected, and two missed cleavages were allowed. Oxidation (Met) and N-acetylation were set as variable modifications, whereas Carbamidomethylation (Cys) as fixed modification. False discovery rate (FDR) of 1% was used for peptides and proteins identification. Only proteins identified with a minimum of one unique peptide and quantified with a minimum of two ratio counts were used for the analysis. Only peptides uniquely identified were used for protein quantification. The relative quantification of the peptides against their SILAC-labeled counterparts was performed by MaxQuant. Label-free quantification was performed with the label-free algorithm integrated in MaxQuant 28. Common contaminants 26 were excluded from the analysis. Data analysis. MS data statistical analysis and 2D analysis were performed with Perseus software 29. Comparative pathway analysis was performed with Ingenuity® Pathway Analysis IPA®, QIAGEN Redwood City, www.qiagen.com/ingenuity . Cell proliferation and apoptosis. To assess cell proliferation, HUVECs were incubated with EdU for 1.5 h. After harvesting and staining with Click-iT EdU kit (Invitrogen) according to manufacturer’s recommendations, EdU incorporation was analyzed by FACS. To assess apoptotic cell death, Annexin V kit (Invitrogen) was used according to manufacturer’s recommendations and cells were analyzed by FACS. Immunoblot analysis. HUVECs were lysed in 2% SDS in 0.1 M Tris HCl pH 7.4 and mouse ECs in 50 mM TrisHCl, pH 7.5, 140 mM NaCl, 1% Igepal and complete protease inhibitor (Roche) buffer. Proteins were separated on 4-12% gradient NuPAGE Novex Bis-Tris gel (Life Technologies) and transferred to PVDF membranes (Millipore). The membranes were probed with the following primary antibodies: anti-ATG7 (H-300, sc-33211), anti-β tubulin (H235, sc-9104) from Santa Cruz Biotechnology; anti-vinculin (V9131) from Sigma; anti-ATG5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

(D1G9, #8540) and anti-LC3B (#2775) from Cell Signaling; anti-p62/SQSTM1 (#PM045) from MBL. Permeability assay (Trans Endothelial Electrical Resistance, TEER). HUVECs were plated on 0.4 µm pore size polyester membrane (Corning, New York, NY) pre-coated with 1% gelatin and grown 100% confluent. STX2 electrode connected to an EVOM2 voltohmmeter (World Precision Instruments) was used to measure TEER. The background resistance measured in transwells with no cells was subtracted from the transwells with cells. Permeability assay (FITC-dextran or FITC-albumin). HUVECs were plated and grown as for the TEER assay and treated as indicated. For the ROS scavenging experiment, the cells were pre-incubated with N-acetyl-L-cysteine (1mM, Sigma) and ascorbic acid (vitamin C, 500 µM, Sigma) for 15 minutes before treatment with 100 nM bafilomycin. FITC-dextran 40 KDa or FITC-albmin 65 KDa, 10 µM (Sigma-Aldrich), was then added in the top wells and after 30 minutes the transwells were disassembled and the fluorescence of the medium in the bottom chamber was measured by using a fluorescence plate reader. Mice and Miles assay The C57Bl/6 mice with endothelial-specific deletion of Atg7 were generated by crossbreeding Atg7flox/flox (kindly provided by Prof. Masaaki Komatsu) with Cdh5 (PAC)-CreERT2 mice (kindly provided by Prof. Ralf Adams) to produce Atg7flox/+Cdh5-Cre+/−. These mice were further crossed to generate Atg7flox/floxCdh5-Cre−/− (used as control) or Atg7flox/floxCdh5Cre+/− mice. Three mice per genotype were used to extract the lungs which were used to isolate endothelial cells. Mice were injected intraperitoneally with 2 mg of tamoxifen daily for three days and sacrificed two weeks later. Mouse lung endothelial cells (MLECs) were isolated from lungs of 8 weeks old mice with slight modification to the protocol previously described

30

. Mice were culled by cervical

dislocation and lungs placed in Opti-MEM (Life Technologies) with 20% fetal bovine serum 8 ACS Paragon Plus Environment

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and antibiotics. Minced tissue was left for 45 minutes at 37ºC in pre-warmed 2mg/ml type 1 collagenase (Sigma) in Dulbecco’s PBS (Sigma). The cell suspension was gently passed through a 19 gauge cannula twelve times then through a 70µm cell strainer prior to being centrifuged and the pellet re-suspended in cold PBS/0.1% bovine serum albumin. Cell suspension was incubated for 10 minutes at room temperature with Dynabeads (Dynal) precoated with anti-mouse CD102 (Icam-2) antibody (Pharmingen), 15 µl of beads/ml of cell suspension. After separation and washes using a magnetic separator, the cells were plated in M199 (Invitrogen) with antibiotics, 16% fetal bovine serum, 10 µg/ml heparin, 2 mM glutamine and endothelial cell growth supplement (Sigma). A second sort with the magnetic beads was performed after 3 days of culture. The Miles assay was performed as previously described with minor modifications

31

. Briefly,

30 min before Evans blue injection, mice were intra-peritoneally injected with 100 µl of pyrilamine maleate salt (4 mg/kg body weight in 0.9% saline, Sigma-Aldrich) to inhibit histamine release. Then 200 µl of 0.5% solution of Evans blue were injected tail vein and left in the circulation for 20 min. Mice were sacrificed by terminal anesthesia. Residual Evans Blue within the blood vessels was removed by heart perfusion with PBS followed by 0.1% PFA in PBS. Lungs were excised, weighed and processed to extract the Evans blue. After 48h incubation in formamide, the Evans blue was quantified with a spectrophotometer (wavelength = 610 nm) and normalized to the weight of the lungs. All mouse procedures were in accordance with ethical approval from University of Glasgow under the revised Animal (Scientific Procedures) Act 1986 and the EU Directive 2010/63/EU authorized through Home Office Approval (Project license number 60/4181). Reactive oxygen species (ROS) measurement. HUVECs were incubated for 30 minutes with 3 µM CM-H2DCFDA probe (Invitrogen), harvested and assayed by FACS. Alternatively, cells plated on gelatin-coated glass bottom dishes were incubated with the carboxy,



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

H2DCFDA probe (Image-iT ™ LIVE Green Reactive Oxygen Species Detection Kit, Invitrogen) and imaged live according to manufacturer’s recommendations. Experimental Design and Statistical Rational. All the proteomic analyses were performed using samples from three replicate experiments, where cells were cultured in separate dishes, and lysed and prepared for MS analysis on the same day. For the time resolved MS-proteomic analysis, in order to compare all cell conditions between them, an ANOVA test corrected for multiple testing analysis (maximum permutation-based FDR of 0.1%) was used. For the other MS-proteomic analyses, a cut-off of 2 standard deviation (SD) from the mean of the calculated SILAC ratios was used. This test was chosen because the data (SILAC ratios of ratios) followed a normal distribution, and because the distribution of the ratios was quite narrow. This allowed us to select proteins with moderate changes in levels, but consistent between replicates. For all other analyses, for each assay, a representative experiment of at least three reproducible independent experiments is shown. As controls we used a non-targeting siRNA for silencing experiments, and vehicle for drug-treatment experiments. In the plots, bars represent mean ± standard error of the mean (S.E.M) for n = 3 technical replicates. GraphPad Prism software was used for statistical analysis. A two-tailed unpaired t-test was used to calculate p-values: * = p

In-Depth Proteomics Identifies a Role for Autophagy in Controlling

Recommend Documents