A PROTEOMIC ANALYSIS OF GSD-1a IN MOUSE LIVERS

Subscriber access provided by ALBRIGHT COLLEGE

Article

A PROTEOMIC ANALYSIS OF GSD-1a IN MOUSE LIVERS: EVIDENCE FOR METABOLIC REPROGRAMMING, INFLAMMATION AND MACROPHAGE POLARIZATION Davide Cangelosi, Roberta Resaz, Andrea Petretto, Daniela Segalerba, Marzia Ognibene, Federica Raggi, Luca Mastracci, Federica Grillo, Maria Carla Bosco, Luigi Varesio, Antonio Sica, Irma Colombo, and Alessandra Eva J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.9b00309 • Publication Date (Web): 07 Jun 2019 Downloaded from http://pubs.acs.org on June 9, 2019

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 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 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.

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 48 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

A PROTEOMIC ANALYSIS OF GSD-1a IN MOUSE LIVERS: EVIDENCE FOR

METABOLIC

REPROGRAMMING,

INFLAMMATION

AND

MACROPHAGE POLARIZATION

Davide Cangelosi1^, Roberta Resaz1^, Andrea Petretto2, Daniela Segalerba1, Marzia Ognibene3, Federica Raggi1, Luca Mastracci4,5, Federica Grillo4,5, Maria Carla Bosco1, Luigi Varesio1†, Antonio Sica6,7, Irma Colombo8, Alessandra Eva1*.

1Laboratorio

di Biologia Molecolare, IRCCS Istituto Giannina Gaslini, Via G. Gaslini

n. 5, 16147 Genova, Italy 2Core

Facilities-Proteomics Laboratory, IRCCS Istituto Giannina Gaslini, Via G.

Gaslini n. 5, 16147 Genova, Italy 3Laboratorio

Cellule Staminali Post Natali e Terapie Cellulari, IRCCS Istituto

Giannina Gaslini, Via G. Gaslini n. 5, 16147 Genova, Italy. 4Department

of Surgical and Diagnostic Sciences (DISC), Anatomic Pathology Unit,

University of Genova, Viale Benedetto XV n. 6, 16132 Genova, Italy. 5IRCCS

Ospedale Policlinico San Martino, National Cancer Research Institute, Largo

Rosanna Benzi n. 10, 16132 Genova, Italy 6Department

of Pharmaceutical Sciences, Università del Piemonte Orientale "Amedeo

Avogadro", Largo Guido Donegani n. 2, 28100 Novara, Italy. 7Humanitas

Clinical and Research Center, Via Alessandro Manzoni n. 56, 20089

Rozzano, Italy. 8Dipartimento

di Scienze Farmacologiche e Biomolecolari, Università degli Studi di

Milano, Via Balzaretti n. 9, 20133 Milano, Italy.

ACS Paragon Plus Environment

1

Journal of Proteome Research 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

^Davide

Page 2 of 48

Cangelosi and Roberta Resaz contributed equally to this work

† Deceased *Corresponding author Dr. Alessandra Eva Address: Laboratorio di Biologia Molecolare, Istituto Giannina Gaslini, Largo Gaslini 5, 16147 Genova, Italy. Tel.: +39 010 5636633. Fax: +39 0103733346. E-mail address: [email protected].


2

Page 3 of 48 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


ABSTRACT Glycogen storage disease type 1a (GSD-1a) is a rare genetic disease caused by mutations in the catalytic subunit of the enzyme glucose-6-phosphatase-alpha (G6Pase-α). The majority of patients develop long-term complications including renal failure and hepatocellular adenoma/carcinoma. The purpose of this study was to ascertain the proteomic changes in the liver of LS-G6pc-/- mice, a murine model of GSD-1a, in comparison with wild type mice to identify potential biomarkers of the pathophysiology of the affected liver. We used liquid chromatography-tandem mass spectrometry (LC-MS/MS) to analyze liver lysates from a total of 20 LS-G6pc-/- and 18 wild type (WT) mice.

We compared the proteomic expression profile of LS-

G6pc-/- and WT mice. We identified 4138 significantly expressed proteins 1243 of which were differentially represented. Network and pathway analyses indicate that LS-G6pc-/- livers display an age-dependent modulation of the expression of proteins involved in specific biological processes associated with increased progression of liver disease. Moreover, we found upregulation of proteins involved in the process of tissue inflammation and macrophage polarization toward the M2 phenotype in LSG6pc-/- mice with adenomas.

Our results identify a metabolic reprogramming of

glucose-6-P and a pathologic environment in the liver compatible with tumor development and progression.

KEYWORDS: Glycogen storage disease type 1a, Macrophage polarization, Hepatocellular adenoma, Hypoxia, Animal model, Proteomics


3


Page 4 of 48

INTRODUCTION Glycogen storage disease type 1a (GSD-1a) is a rare autosomal recessive disorder caused by the deficiency in glucose-6-phosphatase-alpha (G6Pase-α). This enzyme, encoded by G6PC gene, is expressed primarily in the liver, kidney and intestine, where it catalyzes the hydrolysis of glucose-6-phosphate (G6P) to glucose in the terminal reaction of gluconeogenesis and glycogenolysis.1-3 The lack of production of glucose leads to the accumulation of glycogen and triglycerides in the liver and kidney, causing the progressive worsening of the clinical parameters and functions of liver and kidney affected by the chronic dysmetabolism. Nowadays, GSD-1a patients are treated by naso-gastric infusion of glucose and frequent meals of uncooked cornstarch to control symptomatic hypoglycemia.4;5 However, the primary disease remains uncorrected, and long-term complications, including renal failure, development of hepatocellular adenomas (HCA), and hepatocellular carcinoma (HCC) still occur in a high percentage of patients. In fact, 80% of the patients with GSD-1a above the age of 30 years developed HCA and 10% of these patients developed HCC.6 A mouse model for GSD-1a (G6PC null mice),2 has been generated and used to develop somatic gene therapies. It was shown that gene therapy with viral vectors carrying the G6PC transgene can improve the survival of these mice by correcting their liver metabolic abnormalities to various levels.7;8 However, these animals, unless treated, usually die within 3 weeks after birth and therefore could not be used to study long-term complications of the disease. We generated transgenic mice (LS-G6pc-/-) that carry a liver-specific deletion of G6Pase-α.9 LS-G6pc-/- mice exhibit a milder phenotype and therefore display normal life expectancy. We characterized the liver disease in these mice and demonstrated


4

Page 5 of 48 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


that they display hepatic abnormalities comparable to those of patients affected by GSD-1a, including late development of HCA and HCC.9 Because of these characteristics, these mice constitute a valuable animal model to study GSD1a longterm complications and to find biomarkers for new treatments. Gene expression profile allows the identification of transcriptional changes related to disease onset and progression but agreement between alteration in gene transcription and changes in protein levels has not always been observed.10;11 Therefore, the analysis of protein levels through a proteomic approach should allow a better characterisation of the pathophysiological alterations in a diseasd tissue such as liver.12 A proteomic analysis permits the concomitant characterization of hundreds of differentially expressed proteins allowing the identification of relevant changes in a tissue as a hallmark of distincive biological conditions. Thus far, many studies have focused on proteomics to characterize diseased livers.13;14 In this study, we determined the proteomic expression changes in the liver of LSG6pc-/- mice in comparison with wild type mice to identify potential biomarkers of the pathophysiology of the affected liver. We show that LS-G6pc-/- livers display an agedependent modulation of the expression of proteins belonging to specific biological processes associated with the progression of liver disease. Moreover, proteins that are hallmarks of hypoxia were persistently upregulated in the liver of these mice, while the acute inflammatory and immune response were among the significantly enriched biological processes in LS-G6pc-/- mice with adenomas, in comparison with mice without tumors. Finally, different proteins involved in the process of tissue inflammation and macrophage polarization toward the M2 phenotype were upregulated.


5


Page 6 of 48

Our results are compatible with a metabolic reprogramming of glucose-6-P leading to enhanced glycolysis, lactate production, increased lipogenesis and cholesterol synthesis, and dysfunction in oxidative phosphorylation. Moreover, our results identify in the liver a pathologic environment, represented by hypoxia, inflammation, and macrophage M2 polarization, which may lead to tumor development and progression.


6

Page 7 of 48 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


EXPERIMENTAL PROCEDURES Housing of mice The animal Use project n. 291, communicated to the Italian Ministry of Health and carried out at the animal facility of the National Institute for Cancer Research, Genova, Italy, was reviewed and approved by CSEA, the Italian Ethical Committee for Animal Experimentation.

Sample collection, histology, and phenotype analyses LS-G6pc-/- mice generation and characterization was previously described.9 Both females and males were included in the study. Mice were grouped according to their age into 6 groups, from 1 month to 18 months, three months apart. Animals were in fed state when sacrificed. Blood was drawn with a syringe from the left ventricle of anesthetized mice. Livers were sectioned, in part frozen for proteomic analysis and in part fixed for 24 hours in 10% buffered formalin and paraffin embedded. Four micrometer thick sections were stained with haematoxylin and eosin for histological analysis. All animals were phenotypically evaluated. The mean glucose concentration in the blood of LS-G6pc-/- mice in the fed state was 25% lower than in control mice, even though it always stayed within normal levels, as previously described.9 Mice were histologically evaluated for liver lesions. Mice showed different degree of glycogen and lipid accumulation, and six out of twenty LS-G6pc-/- mice developed hepatic adenoma (Table S-1). No HCA or areas close to HCA were included in the study. Livers of WT mice were normal.


7


Page 8 of 48

Proteomics setup Peptide separation, mass spectrometry analysis and protein identifications were performed as described by Santucci and collaborators15 with minor changes. Briefly, UHPLC Dionex Ultimate 3000 RSLC system, configured in preconcentration injection mode, using a trapping column (2 cm × 100 μm ID, Acclaim PepMap C18, 2 μm particles, 100 Å pore size; Thermo Scientific Cat. No. 164564) at flow rate of 5µl/min, was used to focalize the digested peptides. After 5 min, the trapping column is switched in line to a separation column, eluting the peptides in a 170 min multistep gradient on a EASY-Spray column (25 cm x 75 µm ID, PepMap C18, 2 µm particles, 100 Å pore size; Thermo Scientific Cat. No. ES803), mounted on the EASY-Spray Ion Source, thermostated at 40 °C with a voltage of 2.4 kV and a flow rate of 300nl/min. The mass spectrometer LTQ Orbitrap Velos Pro, acquiring in a top20 DDA mode, was used. The MS1 resolution was set to 60000 with an AGC of 1E10^6 and an IT of 250 ms. The lock mass, based on polydimethylcyclosiloxane background ion, was enabled. The triggering MS/MS has been set to 3000 ions using an isolation window of 2 Da. The MS/MS are collected in a linear trap with a maximal ion injection time of 100 ms and normalized collision energy of 35% The mass spectra were subsequently processed using the MaxQuant software version 1.5.3.3016 and a mouse database downloaded from UniProt (release 2016_02). For all processing, the default settings have been maintained. Lastly, quantification was performed using the built-in label-free quantification algorithm,17 enabling the ‘Match between runs’ option (time window 1 min) in order to compensate the effect of missing values. Proteomic data have been deposited to the ProteomeXchange Consortium via the PRIDE18 partner repository with the dataset identifier PXD011561.


8

Page 9 of 48 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


Experimental Design Comparative proteomics profiling was performed by Principal Component Analysis (PCA) implemented in ClustVis online resource.19 Time-course algorithm was used across the 6 experimental time points to identify differentially expressed proteins. Time-course analysis used the Significance Analysis of Microarrays method20 implemented in the Web-enabled and Cross-platform SAM via Shiny (https://github.com/MikeJSeo/SAM). For building the SAM interactive web app straight from R, we used the Shiny R package setting up a two classes, unpaired, time course analysis with slope function as summary measure for each time course. The slope analysis has been designed for identifying those proteins whose expression increases or decreases over time. Briefly, for a two class-unpaired problem, the slope of each time series is calculated and the average slope between the two groups of interest are compared to find proteins with a significant modulation as it happens in a normal differential expression analysis. Gene ontology (GO) and KEGG analyses were carried out on selected groups of proteins using the STRING-DB software.21 A FDR lower than 0.05 identified significantly enriched ontology terms and pathways. STRING-DB software was also used to evaluate protein-protein interaction (PPI) network from selected groups of proteins. PPI enrichment p-value was used to assess whether a group of proteins have more interactions among themselves than what would be expected from a random set of proteins of similar size, drawn from the genome. We considered significantly a PPI enrichment p value lower than 0.05. Statistical calculations were carried out using Prism for Macintosh V6 (GraphPad Software Inc., La Jolla, CA). The schematic representation of the bioinformatics analyses used in the present study is shown in Figure 1.


9


Page 10 of 48

Gene set enrichment analysis We used GSEA22 to evaluate the enrichment of known biological processes and pathways in LS-G6pc-/- mice. We carried out the analysis on all available protein symbols. An enrichment score (ES), a normalized enrichment score (NES), and a Nominal p value (Nom p value) is calculated by GSEA for each gene set of a collection.22 The significance of the enrichment of a gene set is estimated by a nonparametric method based on empirical permutation test.22 When multiple gene sets are tested in one single run, a False Discovery Rate q-value (FDR q-value)

is

associated with each gene set for accounting for the increased probability of generating false positive findings.22 The gene sets used in the analysis belong to the Hallmark

(H), KEGG

(C2.CP.KEGG) and Gene Ontology (C5.GO) collections of the Molecular Signature Database (MSigDB) v6.2 database.23;24 A gene set with nominal p value lower than 0.05 and FDR q-value lower than 0.25 is considered significant.

Western Blot Analysis. LS-G6pc-/- and WT mouse livers were lysed in StaphA buffer (10 mM sodium phosphate, pH 7.4; 0.1 M NaCl; 1% Triton X-100; 0.1% sodium dodecyl sulfate (SDS); 0.5% sodium deoxycholate (DOC); 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride (AEBSF), 10 mg/ml-1 aprotinin, 10 mg/ml-1 leupeptin), processed and transferred to PVDF membrane (Merck Millipore) as previously described.25 Membranes were incubated with specific antibodies (Santa Cruz Biotecnology) to reveal selected proteins. Blots were reprobed with anti-ezrin antibody (Santa Cruz Biotecnology) as loading control. The antibody used were: anti-MVD, sc-376975; anti-TPI, sc-166785; anti- STAT6, sc-374021; anti-Stat6, sc-


10

Page 11 of 48 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


1689; anti-Aldolase A, sc-390733; anti-GAPDH, sc-365062; anti-Gapdh, sc-166545; anti-PKLR, sc-133222; anti-CPT1, sc-393070; anti-NDUFB5, sc-514245; anti-Ezrin, sc-32759. All antibodies were diluted 1:200 in TTBS 1X with 0.5% of bovine serum albumin. Protein signals were visualized with ECL select Western Blotting Detection Reagent (GE-Healthcare) using a horseradish peroxidase (HRP)-conjugated goat antimouse antibody (Thermo Fisher Scientific), quantified using Image Lab 6.0 software (ChemiDoc, Bio-Rad) and normalized to loading control.


11


Page 12 of 48

RESULTS Expression profiling of hepatic proteins in LS-G6pc-/- and WT mice To identify potential biomarkers and key biological processes associated with liver degeneration in GSD-1a, we analyzed the proteomic profile of LS-G6pc-/- and wild type (WT) mice. The flowchart summarizing the main steps of these analyses is shown in Figure 1. Mice were divided into six groups according to their age (Table S-1). A total of 20 LS-G6pc-/- and 18 WT mice were analyzed. All LS-G6pc-/- mice utilized were characterized for pathological manifestations and found to display all the features already described for this murine model of GSD-1a.9 Six LS-G6pc-/- mice developed at least one HCA. None of the WT mice developed liver disease, as expected (Table S-1). Livers were profiled by a proteomic approach based on high-resolution mass spectrometry (HRMS) coupled with liquid chromatography techniques. We obtained 57893 peptides that mapped to 4763 unique mouse gene symbols. Of these, 4138 significantly expressed proteins were quantified and organized in a dataset.

GSEA reveals the enrichment of hypoxia-related proteins in the proteomic profile of LS-G6pc-/- mouse livers Gene set enrichment analysis (GSEA)22 is a computational technique for estimating whether a known set of biologically-related genes (gene set) shows a concordant up- or down-regulation between two conditions. Seven distinct collections of gene sets are publicly available in MSigDB database.23;24 We carried out GSEA between 20 LS-G6pc-/- and 18 WT protein expression profiles using the hallmark collection composed by 50 curated gene sets that represent well-defined biological states or processes, the C5.BP collection composed by 4436 gene sets derived from Gene


12

Page 13 of 48 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


Ontology database, and the C2.CP.KEGG collection composed by 186 gene sets derived from the KEGG pathway database. To use MSigDB collections, protein symbols were converted into the corresponding gene symbols. GSEA identified 71 differentially represented proteins in the proteomic profile of the LS-G6pc-/compared to WT groups of mice (Nom p-value

A PROTEOMIC ANALYSIS OF GSD-1a IN MOUSE LIVERS

Recommend Documents