Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Aug 28;8(9):997.
doi: 10.3390/cells8090997.

Endoglin Trafficking/Exosomal Targeting in Liver Cells Depends on N-Glycosylation

Affiliations

Endoglin Trafficking/Exosomal Targeting in Liver Cells Depends on N-Glycosylation

Steffen Meurer et al. Cells. .

Abstract

Injury of the liver involves a wound healing partial reaction governed by hepatic stellate cells and portal fibroblasts. Individual members of the transforming growth factor-β (TGF-β) superfamily including TGF-β itself and bone morphogenetic proteins (BMP) exert diverse and partially opposing effects on pro-fibrogenic responses. Signaling by these ligands is mediated through binding to membrane integral receptors type I/type II. Binding and the outcome of signaling is critically modulated by Endoglin (Eng), a type III co-receptor. In order to learn more about trafficking of Eng in liver cells, we investigated the membranal subdomain localization of full-length (FL)-Eng. We could show that FL-Eng is enriched in Caveolin-1-containing sucrose gradient fractions. Since lipid rafts contribute to the pool of exosomes, we could consequently demonstrate for the first time that exosomes isolated from cultured primary hepatic stellate cells and its derivatives contain Eng. Moreover, via adenoviral overexpression, we demonstrate that all liver cells have the capacity to direct Eng to exosomes, irrespectively whether they express endogenous Eng or not. Finally, we demonstrate that block of N-glycosylation does not interfere with dimerization of the receptor, but abrogates the secretion of soluble Eng (sol-Eng) and prevents exosomal targeting of FL-Eng.

Keywords: BMP; Caveolin-1; TGF-β; endoglin; exosomes; fibrosis; hepatic stellate cells; hepatocytes; lipid raft; liver; portal myofibroblasts; shedding.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest. The German Research Foundation or the Interdisciplinary Centre for Clinical Research (IZKF) within the Faculty of Medicine at the RWTH Aachen University had no role in the design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Figures

Figure 1
Figure 1
Endoglin is localized in Caveolin-1 enriched lipid rafts. (A) CFSC and (B) primary rat hepatic stellate cells (HSC) were cultured to near confluence under basal conditions (10% FCS). Thereafter, the cells were harvested in carbonate buffer, sonicated, and the extracts separated on a sucrose gradient. Eleven consecutive fractions were collected starting from the top (low sucrose) to the bottom (high sucrose). Proteins of the individual fractions (equal volumes) were separated by SDS-PAGE and analyzed by Western blot analysis. Lipid rafts were identified by detection of Caveolin-1. The presence of Endoglin (Eng), Alix, and CD81 were analyzed by specific antibodies.
Figure 2
Figure 2
Structure, sequence analysis and functional expression of a virus coding for full-length Endoglin from rat. (A) The full-length endoglin (FL-Eng) form is composed of a signal peptide, an orphan domain, a zona pellucida (ZP) domain, a transmembrane region (TM), and a cytosolic domain (CD). Amino acid positions spanning individual regions are numbered. (B) Sequence analysis of FL-Eng expression plasmid. For clarity the sequence data of the 3′end was inversed and complemented for depicting the encoded amino acids. (CE) Human LX-2, primary rat HSC and pMF were infected with a control virus (Luc), FL-Eng or left untreated (NC). Thereafter, cells were treated or not with TGF-β1 (1 ng/mL) for 24 h (C,D), 48 h (E) or cultured without any treatment (Co) (D). Cells were harvested and proteins analyzed using Western blot for transgene expression (FL-Eng), connective tissue growth factor (CTGF) and α-smooth muscle actin (α-SMA) or activation of Smad2 (pSmad2). Equal loading of lanes was proven by re-hybridization of the membranes with an antibody specific for β-actin.
Figure 3
Figure 3
Structure, sequence analysis and functional expression of soluble Endoglin from rat. (A) Scheme of the FL-Endoglin (FL-Eng) domain structure including the amino acid environment surrounding the cleavage site (scissors and red arrow) of human, mouse and rat. (B) Nucleotide sequence of FL-Eng and the corresponding primer composition (SM091) to amplify sol-Eng and introduce a stop codon (TGA) and a NotI restriction site (upper panel). Sequence analysis of the sol-Eng expression plasmid (lower panel). For clarity the sequence data of the 3′end was inversed and complemented for depicting the encoded amino acids. (C) COS-7 cells (left) and CHO cells (right) were transiently transfected with empty vector (pcDNA) or sol-Eng expression vector (sol-Eng). sol-Eng expression was displayed in Western blot in the presence (+) or absence (−) of DTT. (DE) HepG2 cells were transiently transfected using empty vector (pcDNA) or sol-Eng expression vector (sol-Eng). Thereafter, cells were either treated or not with TGF-β1 (1 ng/mL) for the indicated time points and sol-Eng expression and CTGF expression was monitored in Western blot (D) or the supernatant was supplemented with TGF-β1. Binding was accomplished for 2 h rotating and sol-Eng complexes were precipitated overnight using a rat Eng-specific antibody (PPabE2) and analysed in Western blot for the presence of bound TGF-β1 and precipitated sol-Eng. (F) HSC Col-GFP cells were transiently transfected using empty vector (pcDNA) or sol-Eng expression vector (sol-Eng). After starvation for 16 h, the cells were stimulated or not for 48 h with TGF-β1 (0.1, 1.0 ng/mL). Cellular proteins were analysed in Western blot for transgene expression (sol-Eng), CTGF and α-SMA. Equal loading of proteins was proven by re-hybridization of the membranes with a β-actin antibody.
Figure 4
Figure 4
Exosomal marker expression in liver cells. Cells were plated on 10 cm dishes and either cultured in growth medium without further treatment (NC), or subjected to viral infection with a control virus (Luc). Thereafter, exosomes were prepared from the supernatants and marker expression was analyzed by Western blot using antibodies directed against Alg2-interacting protein (Alix), CD81 and Caveolin-1 (Cav-1).
Figure 5
Figure 5
Identification of full-length Endoglin and soluble Endoglin in liver cell lines. (A,B) Cell protein extracts, supernatant (pre), wash solution (post), and final exosome preparations from (A) LX-2 or (B) HepG2 infected with adenoviral constructs expressing luciferase (Luc), full-length endoglin (FL-Eng), or soluble endoglin (sol-Eng) were analysed by Western blot for expression of indicated proteins. (C) Direct comparison of proteins included in exosomes purified from cultured LX-2 and HepG2 cells which were infected with adenoviral expression vectors driving expression of Luc, FL-Eng, and sol-Eng. (D) Cell lysates and corresponding exosomes fractions from HSC Col-GFP were analyzed in parallel in Western blot analysis. Exosomal markers and TGF-β-receptors were detected with specific antibodies. To monitor equal protein loading, the membrane was re-probed with a β-actin specific antibody. Please note, that only Endoglin and no other TGF-β-receptors analyzed were detected in significant amounts in exosomes. Abbreviations used are: NC, normal control meaning untreated cells cultured in the presence of normal growth medium containing 10% FCS; Co/TGF-β1, cells that were starved overnight (~16 h) in medium containing 0.5% FCS before stimulation with TGF-β1 (1.0 ng/mL) (TGF-β1) or not (Co) for 24 h.
Figure 6
Figure 6
Identification of full-length Endoglin in primary mesenchymal liver cells. (A) Cell protein extracts, supernatant (pre), wash solution (post), and final exosome preparations from primary rat HSC untreated (NC), infected with adenoviral constructs expressing luciferase (Luc) or full-length endoglin (FL-Eng) were analysed by Western blot for expression of indicated proteins. (B) Primary rat HSC were stimulated or not with TGF-β1 (1 ng/mL) for 24 h and exosomes were prepared. The corresponding proteins were analysed by Western blot for the expression of exosomal marker proteins and FL-Eng. (C) Cell lysates and corresponding exosomal proteins of three independent experiments of primary rat pMF were analysed by Western blot for exosomal markers and TGF-β-receptors. For displaying equal protein loading, the membrane was reprobed with a β-actin specific antibody. Note that in contrast to other liver cell lines and primary cells, rat pMF express significant amounts of Tβ-RIII and Tβ-RI in exosomes.
Figure 7
Figure 7
Blocking of N-glycosylation by tunicamycin leads to aberrant trafficking of Endoglin. (AD) CHO or COS-7 cells were transfected with plasmids directing expression of rat FL-Endoglin (rFL-Eng), betaglycan (TβRIII), and rat soluble Endoglin (sol-Eng). A transfection with empty vector (pcDNA) served as control. Thereafter, cells were treated with tunicamycin (0.5 µg/mL, 24 h) or left untreated. (A) Cell extracts were prepared and analysed for expression of rFL-Eng and betaglycan. To monitor equal protein loading the membrane was reprobed with a β-actin specific antibody. (B) Cell lysates of COS-7 cells taken from (A) were incubated with ConA beads for precipation of glycosylated proteins. The ConA-bound fraction as well as the unbound fraction of the precipitates were then analysed by Western blot for rFL-Eng and betaglycan. (C) Supernatants of COS-7 cells transfected with vectors expressing rFL-Eng or sol-Eng were treated or not with tunicamycin (0.5 µg/mL, 24 h). Protein extracts were analysed by Western blot in the presence or absence of DTT (50 mM) for expression of sol-Eng. (D) Supernatants of COS-7 cells taken from (C) were incubated with ConA beads for precipitation of glycosylated proteins. The ConA-bound fraction as well as the unbound fraction of the precipitates were then analysed by Western blot for sol-Eng. (E) HSC Col-GFP cells were treated or not with tunicamycin (0.5 µg/mL, 24 h) and stimulated or not with TGF-β1 (1 ng/mL; 30 min). Thereafter, proteins were extracted and analysed by Western blot in the presence or absence of DTT (50 mM) for expression or activation of indicated proteins. Equal protein loading was demonstrated with a β-actin specific antibody. (F) Protein lysates of HSC Col-GFP (only TGF-β1 stimulated samples) taken from (E) were incubated with ConA beads for precipitation of glycosylated proteins. The ConA-bound fraction as well as the unbound fraction of the precipitates were then analysed by Western blot for the indicated receptor proteins. (G) HSC Col-GFP cells were transfected with a plasmid directing expression of rat soluble-Endoglin (sol-Eng) or empty vector control (pBK). Thereafter, cells were treated with tunicamycin (0.5 µg/mL, 24 h) or left untreated. Cell protein extracts were prepared and analysed in Western blot in the presence or absence of DTT (50 mM) for expression of sol-Eng.
Figure 8
Figure 8
Isolation of exosomes from HSC treated with tunicamycin. (A) Cell extracts or exosome fractions prepared from LX-2 which were infected with FL-Eng or a control virus expressing luciferase (Luc). Subsequently, the cells were treated with TGF-β1 or an inhibitor of N-glycosylation (tunicamycin) and analysed by Western blot for expression of indicated proteins. Please note, that only properly glycosylated Eng is directed to the exosomal compartment. (B) Cell protein extracts, supernatants, and exosomes were prepared from primary mouse HSC and analysed for the presence of indicated proteins demonstrating that endogenous Eng is only directed to exosomes when properly glycosylated.

Similar articles

Cited by

References

    1. Nickel J., Ten Dijke P., Mueller T.D. TGF-β family co-receptor function and signaling. Acta Biochim. Biophys. Sin. 2018;50:12–36. doi: 10.1093/abbs/gmx126. - DOI - PubMed
    1. Weiskirchen R., Tacke F. Liver fibrosis: From pathogenesis to novel therapies. Dig. Dis. 2016;34:410–422. doi: 10.1159/000444556. - DOI - PubMed
    1. Weiskirchen R., Meurer S.K., Gressner O.A., Herrmann J., Borkham-Kamphorst E., Gressner A.M. BMP-7 as antagonist of organ fibrosis. Front. Biosci. 2009;14:4992–5012. doi: 10.2741/3583. - DOI - PubMed
    1. Shi M., Zhu J., Wang R., Chen X., Mi L., Walz T., Springer T.A. Latent TGF-β structure and activation. Nature. 2011;474:343–349. doi: 10.1038/nature10152. - DOI - PMC - PubMed
    1. Chen L., Chen R., Velazquez V.M., Brigstock D.R. Fibrogenic signaling is suppressed in hepatic stellate cells through targeting of connective tissue growth factor (CCN2) by Cellular or Exosomal MicroRNA-199a-5p. Am. J. Pathol. 2016;186:2921–2933. doi: 10.1016/j.ajpath.2016.07.011. - DOI - PMC - PubMed

Publication types

LinkOut - more resources