Agonist treatment did not affect association of mu opioid receptors with lipid rafts and cholesterol reduction had opposite effects on the receptor-mediated signaling in rat brain and CHO cells

Abstract

Lipid rafts are small cholesterol- and glycosphingolipid-enriched membrane subdomains. Here we compared the mu opioid receptor (MOR)–lipid rafts relationship in the rat brain, where neurons have non-caveolae rafts, and in CHO cells stably transfected with HA-rat MOR (CHO-HA-rMOR), which are enriched in caveolae. Membranes of rat caudate putamen (CPu) and thalamus or CHO-HA-rMOR cells were homogenized, sonicated in a detergent-free 0.5 M Na₂CO₃ buffer and fractionated through sucrose density gradients. Western blot and [³H]diprenorphine binding showed that ∼ 70% of MOR in CHO-HA-rMOR was present in low-density (5–20% sucrose) fractions enriched in cholesterol and/or ganglioside M1 (GM1) (lipid rafts) in plasma membranes, whereas about 70% and 45% of MOR in CPu and thalamus, respectively, were associated with lipid rafts. Incubation with a saturating concentration of etorphine or morphine at 37 °C for 30 min failed to change the MOR location in rafts in CHO-HA-rMOR, indicating that the internalized MOR does not move out of rafts, in contrast to the delta opioid receptor. In vivo, rafts association of MOR in CPu and thalamus was not affected significantly in rats implanted with two 75-mg morphine pellets for 72 h. In addition, cholesterol reduction by methyl-β-cyclodextrin (MCD) disrupted rafts and shifted MOR to higher density fractions in both CHO-HA-rMOR and CPu membranes. However, MCD treatment had opposite impacts on MOR signaling in the two tissues: it attenuated MOR-mediated [³⁵S]GTPγS binding in CPu but enhanced it in CHO-HA-rMOR.

Introduction

The actions of opioid drugs are mediated by at least three types of opioid receptors: mu, delta and kappa (MOR, DOR and KOR, respectively) (Pasternak, 1988). These receptors belong to rhodopsin subfamily of 7-transmembrane receptors (7-TMRs)/G-protein-coupled receptors (GPCRs) (Bockaert and Pin, 1999). They are coupled via pertussis toxin-sensitive G_i/G_o proteins to many downstream effectors such as adenylate cyclase, potassium channels, calcium channels, and a mitogen-activated protein kinase pathway [for a review, see Law et al., 2000]. Although opioid drugs have long been used as analgesics, their clinical use is limited by the fear of addiction and side effects such as respiratory depression, decreased gastrointestinal motility, sedation, mood changes, and tolerance (Kieffer, 1999). Morphine acts primarily on the MOR (Matthes et al., 1996) and is among the most widely used and abused drugs.

Lipid rafts, including non-caveolae rafts and caveolae, are small cell membrane domains (10–200 nm), which are thought to serve as platforms for many cellular processes (Pike, 2006). They are enriched in cholesterol and glycosphingolipids (e.g., GM1). Recently, lipid rafts were suggested to be renamed as “membrane rafts,” because it has become clear that proteins play a major role in their formation and contribute to their functions (Pike, 2006). While caveolae are relatively stable because the protein caveolins support the structure, non-caveolae rafts are considered to be unstable, heterogeneous and dynamically produced and degraded (Pike, 2006). Electron micrographs show that caveolae are flask-shaped membrane invaginations at plasma membranes in most differentiated cells (Cohen et al., 2004). One the contrary, morphological identification of non-caveolae rafts has been elusive (Shaw, 2006). The existence of non-caveolae rafts is largely based on their biochemical nature of insolubility in nonionic detergents at low temperature and high buoyancy in density gradients (Brown and Rose, 1992), which does not differentiate non-caveolae rafts from caveolae.

Many 7-TMRs/GPCRs have been shown to locate in and regulated by lipid rafts (Pike, 2003, Chini and Parenti, 2004, Ostrom and Insel, 2004), including opioid receptors (Head et al., 2005, Xu et al., 2006, Zhao et al., 2006, Patel et al., 2006, Huang et al., 2007). For example, MOR expressed in HEK293 cells was found by Zhao et al. (2006) to be mainly localized in lipid rafts domains of plasma membranes and this localization was required for MOR-mediated adenylyl cyclase superactivation. Our recent findings demonstrated that the kappa and delta opioid receptor (KOR and DOR, respectively) expressed in CHO cells were mostly associated with and regulated by lipid rafts (Xu et al., 2006, Huang et al., 2007). We have also demonstrated that the acute treatment with a full agonist move some DORs out of lipid rafts in both CHO cells and NG108-15 cells (Huang et al., 2007). However, most of those studies were performed in caveolae-rich non-neuronal cells, while numerous GPCRs, including opioid receptors, are present in neurons in the brain, which were demonstrated to be devoid of caveolae and deficient in its structural protein(s), the caveolins (Cameron et al., 1997).

By the use of an anti-MOR antibody (anti-μC), we have recently found that the MOR associated with lipid rafts at different levels in rat caudate putamen (CPu) and thalamus, with cholesterol and ganglioside M1 as markers for lipid rafts (Huang et al., in press). In this study, we determined to what extent MOR in CHO-HA-rMOR was localized in lipid rafts. In addition, we investigated rafts association of MOR in cells following acute treatment of CHO-HA-rMOR with agonists and in CPu and thalamus after chronic morphine treatment of rats in vivo. Moreover, effects of disrupting rafts by cholesterol reduction on MOR–G protein coupling in the two systems were examined.