Regulation of ABCG4 transporter expression by sterols and LXR ligands

https://doi.org/10.1016/j.bbagen.2020.129769Get rights and content

Highlights

  • ABCG4 is a sterol transporter expressed in cell types in the brain.

  • The regulation of this transporter is currently poorly characterized.

  • ABCG4 mRNA was only upregulated by LXR ligands in astrocytes.

  • ABCG4 protein expression was stabilized by cholesterol and cholesterol synthesis intermediates.

  • ABCG4 protein is potentially regulated by its substrates post-translationally.

Abstract

Background

Oxysterols, which are derivatives of cholesterol produced by enzymic or non-enzymic pathways, are potent regulators of cellular lipid homeostasis. Sterol homeostasis in the brain is an important area of interest with regards to neurodegenerative conditions like Alzheimer's disease (AD). Brain cells including neurons and astrocytes express sterol transporters belonging to the ABC transporter family of proteins, including ABCA1, ABCG1 and ABCG4, and these transporters are considered of interest as therapeutic targets. Although regulation of ABCA1 and ABCG1 is well established, regulation of ABCG4 is still controversial, in particular whether the transporter is an Liver X receptor (LXR) target. ABCG4 is thought to transport cholesterol, oxysterols and cholesterol synthesis intermediates, and was recently found on the blood brain barrier (BBB), implicated in amyloid-beta export. In this study, we investigate the regulation of ABCG4 by oxysterols, cholesterol-synthesis intermediates and cholesterol itself.

Methods

ABC transporter expression was measured in neuroblastoma and gliablastoma cell lines and cells overexpressing ABCG4 in response to synthetic LXR ligands, oxysterols and cholesterol-synthesis intermediates.

Results

In contrast to previous reports, ABCG4 expression was induced by a synthetic LXR ligand in U87-MG astrocytes but not in neuroblastoma and BBB endothelial cell lines. In addition, ABCG4 protein was stabilized by cholesterol as was previously shown for ABCG1. ABCG4 protein was furthermore stabilized by cholesterol-synthesis intermediates, desmosterol, lathosterol and lanosterol.

Conclusions

These results identify new aspects of the post-translational control of ABCG4 that warrant further exploration into the role of this transporter in the maintenance of sterol homeostasis in the brain.

Introduction

Oxysterols are derivatives of cholesterol that are produced via a number of enzymatic or non-enzymic steps [1]. Many oxysterols are potent regulators of metabolic pathways and have been investigated extensively in the context of atherosclerosis and macrophage lipid homeostasis [2]. More recently, oxysterols have become associated with the pathology of Alzheimer's disease, with studies indicating changes in oxysterol levels in brains of patients compared to controls [[3], [4], [5]]. Sterol homeostasis in the brain is distinct from that in circulation, with the absence of circulating lipoproteins contributing to brain sterol homeostasis due to their inability to cross the blood brain barrier (BBB). However, similar to peripheral cells, export of excess cholesterol and oxysterols can be carried out by a number of transporters belonging to the family of ATP-binding cassette (ABC) transporters [6]. The cholesterol and oxysterol transporters ABCA1, ABCG1 and ABCG4 have been shown to be expressed in various cell types in the brain [6], including neurons and astrocytes. More recently, ABCG4 was found to be expressed on BBB endothelial cells [7].

The precise sterol substrates of these ABC lipid transporters expressed in the brain are only partially characterized, based primarily on in vitro cell culture studies and the phenotype of knockout mouse models. ABCA1 is thought to primarily export cholesterol and phospholipids, including phosphatidylcholine and sphingomyelin [8,9], utilizing apolipoprotein A-I as extracellular acceptor in the circulation and ApoE in the brain. More recently, lanosterol, a cholesterol-synthesis intermediate, was added to the list of potential ABCA1 substrates [10]. Proposed ABCG1 and ABCG4 substrates include cholesterol, sphingolipids, but also a number of oxysterols and cholesterol synthesis intermediates [11,12], based on analysis of knockout mice. Abcg1 knockout mice accumulate a number of oxysterols, including 25-hydroxycholesterol (25-HC) and 3β,5α,6β-cholestanetriol in lung and brain tissues [13]. Two separate research groups characterized brain tissue from Abcg1/Abcg4 double knockout mice and reported significant accumulation of 24(S)-hydroxycholesterol (24S-HC) and 27-hydroxycholesterol (27-HC) in addition to the cholesterol synthesis intermediates, desmosterol, lanosterol and lathosterol [11,14]. Considering abnormal levels of 24S-HC and 27-HC have been observed in brain samples from AD patients [5], these transporters may have an important function in the context of AD development that is so far poorly understood. Single nucleotide polymorphisms in the ABCG4 gene have been associated with AD development [15], however it is also known that these transporters are highly regulated at the post-translational level which may reduce potential effect sizes seen in genetic association studies. Moreover, Do and colleagues [16] showed that overexpression of ABCG4 in HEK293 cells increased the export of radiolabeled amyloid-β (1–40) peptide, which accumulates in AD plaques. It is unclear whether this export function was due to direct transport of the peptides or changes in the lipid environment due to the activity of the transporter. Nonetheless, it suggests that ABCG4 may play an important role in the removal of amyloid-β peptides from brain due to its expression on BBB endothelium.

Unlike ABCA1 and ABCG1, the regulation of ABCG4 expression has been controversial and poorly understood. ABCA1 and ABCG1 are highly regulated by the Liver X Receptor (LXR), which forms a dimer with Retinoid X receptor (RXR) to upregulate expression of a number of cholesterol-related genes [17]. Early characterization of the ABCG4 gene indicated the location of LXR response elements and suggested the transporter gene expression could be potentially regulated by endogenous LXR ligand such as oxysterols [18]. Endogenous LXR ligands can be synthesized intracellularly from cholesterol, including the potent oxysterols 25-HC, 27-HC and the brain-specific cholesterol metabolite, 24S-HC [2]. However, reports on whether ABCG4 is indeed an LXR target gene are varied and with conflicting results [12,19].

In addition to transcriptional regulation, our group and colleagues have shown that ABCA1 and ABCG1 are highly regulated at the post-translational level, by mechanisms that include phosphorylation and protein ubiquitination [[20], [21], [22], [23]]. We furthermore showed that deprivation of cellular cholesterol via serum starvation lead to degradation of the transporter proteins in overexpressing cells, while loading with cholesterol, a primary substrate for the transporters, could prevent this degradation [20]. This suggests that cholesterol status in the cell affects transporter protein levels via a post-translational mechanism.

Considering the lack of clarity around the regulation of ABCG4, the aim of this study was to investigate in more detail the regulation of ABCG4 by LXR as well as post-translational regulation of the transporter by sterol substrates in a number of commonly used neuroblastoma and glioblastoma cell lines as well as cells overexpressing the transporter. These ABC lipid transporters have been implicated as potential therapeutic targets, and hence unravelling their regulation may open up new avenues for drug development in future.

Section snippets

Materials

Cell lines were purchased from Cell Bank Australia, Sigma-Aldrich or Merck-Millipore. Cell culture media, fetal bovine serum (FBS), supplements and phosphate-buffered saline (PBS) were all purchased from ThermoFisher Scientific, except for media used for growing hCMEC/D3 cells (EGM2MV Endothelial Medium Bullet Kit), which was purchased from Lonza Australia. Oxysterols and cholesterol synthesis intermediates were all purchased from Avanti Polar Lipids. Compactin (also known as mevastatin),

Cell culture

CHO-K1 cells stably overexpressing either human ABCG1 (expressing the C-terminal myc-tagged ABCG1(−12) isoform; [25]) or ABCG4 [21] have been described previously and were cultured in Ham's F12 medium with standard additions (10% FBS, supplemented with l-glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 μg/ml), plus zeocin (200 μg/ml).

The neuroblastoma cell line Be(2)C (also referred to as SK-N-Be(2)) was maintained in Ham's F12/MEM (50:50 v/v) while SH-SY5Y cells were cultured in MEM

Cholesterol/oxysterol/cyclodextrin complexes

Sterols were complexed to methyl-β-cyclodextrin (CD) as described in Luu et al for cholesterol/CD [24], and adapted for oxysterols as follows. Methyl-β-CD was dissolved in nano-pure water at a concentration of 5% (w/v). Sterols were dissolved in ethanol at a concentration up to 15 mg/ml, depending upon solubility. 500 μl of methyl-β-CD stock solution was heated in a 1.5 ml Eppendorf or Cryo tube to 80 °C while stirring with a tiny magnetic bar. Sterol stock solutions were added in 10 μl

Cell lysis, SDS-PAGE and Western blotting

Cells were washed twice in PBS and harvested in 1% (v/v) IGEPAL-based cell lysis buffer as described in Aleidi et al [21]. Cell protein levels were determined using the BCA assay and equal amounts of cell protein loaded onto 10% (w/v) SDS-PAGE gels as described in Aleidi et al [21]. Proteins were transferred onto nitrocellulose and detected via Western blotting using antibodies for myc (for detection of myc-tagged ABCG1; 1:5000) or ABCG4 (1:5000), followed by HRP-conjugated secondary antibody

RNA extraction, reverse-transciption and qRT-PCR

Cells were washed three times and total RNA was extracted using TRIzol Reagent, followed by generation of cDNA using a High Capacity cDNA Reverse Transcription Kit (ThermoFisher Scientific). The mRNA levels of various candidate genes were determined by qRT-PCR using a Rotor-Gene Q (Qiagen Pty Ltd., Australia). Primer sequences are listed in Table 2. mRNA expression levels were normalised to that of the housekeeping gene (hydroxymethybilanesynthase; HMBS) and expressed relative to the control

Cholesterol efflux assay

Determination of cholesterol efflux is described in detail in Yang and Gelissen [26]. Briefly, cells were labelled by incubation overnight with 1 μCi/ml of [1α, 2α(n)-3H]-cholesterol in serum-containing media. The medium was then removed, followed by twice washing of cells with PBS and subsequent equilibration for 30 min in medium containing 0.1% (w/v) fatty-acid free BSA without serum. Following the equilibration, the labelled cells were washed once with PBS and then incubated for the

Data analysis

Data are expressed as means ± Standard Error of the Mean (SEM). Significance was determined where appropriate using the two tailed Student's t-tests using Prism software version 7 (GraphPad Software, USA), with p < 0.05 considered as significant.

Results

The first aim of our study was to investigate the transcriptional regulation of ABCG4 in cell types that are present in brain. For this, we utilized two commonly used neuroblastoma cell lines (Be(2)C and SH-SY5Y) as well as BBB endothelial cells (hCMEC-D3) and investigated whether ABCG4 mRNA expression could be induced by LXR ligands. In Fig. 1A, cells were incubated with the synthetic LXR ligand, TO901317, and ABCG4 expression compared to ABCA1 (a known LXR target gene). ABCA1 mRNA expression

Discussion

In this study, we aimed to investigate the regulation of ABCG4, a brain lipid transporter, in a number of commonly used cell lines including neuroblastoma, astrocytes and BBB endothelial cells, which are commonly used cell lines to study processes in the equivalent cell types in brain. Unlike its close family member ABCG1, ABCG4 is almost exclusively expressed in brain. The transporter is of importance in the context of Alzheimer's disease as it was recently found to be expressed in BBB

Authorship statement

Alryel Yang: Conceptualization; Data curation; Formal analysis; Writing – original draft; Writing – review and editing.

Amjad Z. Alrosan: Conceptualization; Data curation; Formal analysis; Writing – original draft.

Laura J. Sharpe: Data curation; Methodology; Writing – original draft; Writing – review and editing.

Andrew J. Brown: Conceptualization; Methodology; Writing – original draft.

Richard Callaghan: Conceptualization; Supervision; Writing – original draft.

Ingrid C. Gelissen:

Declaration of Competing Interest

We have no conflicts to declare.

Acknowledgements

We thank Proffesor Wendy Jessup from the Anzac Research Institute, The University of Sydney, for her generous donation of HDL2 for our experiments. AY was supported by a scholarship from the Australian Government. AZA was supported by a scholarship from the Hashemite University in Jordan.

References (38)

1

Combined first authors.

View full text