|Home | About | Journals | Submit | Contact Us | Français|
Over the past decade, an increasing number of studies have shown that G-protein-coupled receptors including opioid and cannabinoid receptors associate to form heteromers. Moreover, G-protein-coupled receptor heteromerization leads to the modulation of the binding, signaling, and trafficking properties of individual receptors. Although very little information is available about the physiological role of receptor heteromers, some studies have shown that the levels of some heteromers are upregulated in disease states such as preeclamptic pregnancy, schizophrenia, Parkinson’s, ethanol-induced liver fibrosis, and development of tolerance to morphine. The recent generation of antibodies that selectively recognize distinct heteromers and, of peptides that selectively disrupt them, have started to elucidate the contribution of heteromers to the disease state. Here, we describe the methods for the generation of heteromer-selective antibodies and elucidation of their levels and localization under normal and pathological conditions.
Drugs of abuse such as opioids and cannabinoids act through G-protein-coupled receptors (GPCRs), the opioid and cannabinoid receptors. Three opioid (μ, δ, and κ) and cannabinoid (CB1R, CB2R, and GPR55) receptor subtypes have been identified (Balenga, Henstridge, Kargl, & Waldhoer, 2011; Dietis, Rowbotham, & Lambert, 2011; Di Marzo, Piscitelli, & Mechoulam, 2011). Both receptors signal via Gαi/o proteins to activate similar signal transduction cascades leading to decreases in intracellular cyclic AMP levels, inhibition of neurotransmitter release, and to increases in mitogen-activated protein kinase phosphorylation (Bushlin, Rozenfeld, & Devi, 2010; Cichewicz, 2004; Howlett et al., 2002; Vigano, Rubino, & Parolaro, 2005). Moreover, activation of either receptor induces similar physiological responses such as antinociception, sedation, reward, and emotional responses (Maldonado, Valverde, & Berrendero, 2006; Manzanares et al., 1999). This similarity in mechanisms of action and physiological responses suggests the possibility of interactions between the opioid and cannabinoid systems.
Opioid receptor subtypes can associate to form higher-order structures, a process known as heteromerization. For example, μ (μOR) and δ (δOR) opioid receptors heteromerize and these modulate binding, signaling, and morphine-mediated analgesia (Gomes et al., 2004, 2000; Gomes, Ijzerman, Ye, Maillet, & Devi, 2011; Kabli et al., 2010; Levac, O’Dowd, & George, 2002; Rozenfeld & Devi, 2007). Heteromerization between δOR and κ opioid receptors (κOR) leads to novel pharmacology and alteration of individual receptor-trafficking properties (Berg et al., 2012; Bhushan, Sharma, Xie, Daniels, & Portoghese, 2004; Jordan & Devi, 1999). Furthermore, opioid receptors can heteromerize with other family A GPCRs such as α2A adrenergic (Jordan, Gomes, Rios, Filipovska, & Devi, 2003; Rios, Gomes, & Devi, 2004), β2 adrenergic (Jordan, Trapaidze, Gomes, Nivarthi, & Devi, 2001), chemokine (Chen et al., 2004; Hereld & Jin, 2008; Pello et al., 2008), substance P (Pfeiffer et al., 2003), or somatostatin receptors (Pfeiffer et al., 2002). Interestingly, heteromerization between CB1R and μOR, δOR, or angiotensin AT1 receptors (AT1Rs) leads to alterations in signaling and localization of CB1R (Rios, Gomes, & Devi, 2006; Rozenfeld et al., 2012, 2011). However, little information is available about the physiological role of GPCR heteromers due to a lack of appropriate tools to study them in endogenous tissues and to distinguish from receptor homomers. Studies using mainly coimmuno-precipitation techniques suggest the involvement of some GPCR heteromers in disease. Heteromers between dopamine D1–D2 receptors have been implicated in major depression (Pei et al., 2010), between AT1R and adrenergic α1D or AT1R and bradykinin B2 receptors with preeclamptic pregnancy (AbdAlla, Abdel-Baset, Lother, el Massiery, & Quitterer, 2005; Gonzalez-Hernandez Mde, Godinez-Hernandez, Bobadilla-Lugo, & Lopez-Sanchez, 2010) and between dopamine receptor subtypes as well as dopamine D2 and adenosine 2A receptors in schizophrenia (Dziedzicka-Wasylewska, Faron-Gorecka, Gorecki, & Kusemider, 2008; Faron-Gorecka, Gorecki, Kusmider, Wasylewski, & Dziedzicka-Wasylewska, 2008; Fuxe et al., 2005; Maggio & Millan, 2010; Perreault, O’Dowd, & George, 2011). However, direct demonstration of heteromers in vivo has not been possible due to a lack of appropriate reagents.
We recently generated monoclonal antibodies (mAbs) that selectively recognize heteromers over individual receptor homomers using a subtractive immunization strategy. This enabled studies to directly explore the physiological role of GPCR heteromers. For example, these antibodies can be used to detect the presence of a heteromer in a specific tissue/region. A case in point is the detection of δOR–κOR heteromers in peripheral sensory neurons using δOR–κOR selective antibodies (Berg et al., 2012). Alternatively, the antibodies could implicate the heteromer in a disease state. μOR–δOR heteromer-selective antibodies detect increased heteromer levels in brain regions involved in pain processing following chronic morphine administration under conditions leading to the development of tolerance (Gupta et al., 2010), suggesting that they may play a role in tolerance. This is supported by studies showing that μOR–δOR heteromer disruption leads to enhanced morphine analgesia with a concomitant decrease in tolerance (He et al., 2011). CB1R–AT1R heteromer-selective antibodies detect a significant heteromer upregulation in hepatic stellate cells of rats chronically treated with ethanol (Rozenfeld et al., 2011), suggesting its involvement in ethanol-induced liver fibrosis. Here, we describe the generation of heteromer-selective antibodies and their use with enzyme-linked immunosorbent assays (ELISAs), immunofluorescence, immunoprecipitation, and Western blotting to detect levels and localization of heteromers in native tissues under normal and pathological conditions.
The advantages of mAbs to probe for heteromer levels in normal and disease states are that they recognize a single epitope, are highly specific, and can be produced in large quantities. The challenge in the generation of heteromer-selective mAbs is that the “heteromer-specific” epitope may be present in a cell or tissue at relatively low levels thus limiting the chances of being detected by antibody-producing cells. This limitation can be overcome through the use of a subtractive immunization strategy (Salata et al., 1992; Sleister & Rao, 2001, 2002). The first step in subtractive immunization involves tolerizing mice with membrane preparations that either do not express the receptor of interest (usually HEK-293 membranes) or express one of the two receptors (e.g., Neuro2A neuroblastoma cell membranes used to generate CB1R–AT1R heteromer-selective mAb (Rozenfeld et al., 2011) endogenously express CB1R). Tolerization is achieved by killing activated antibody-producing cells through the use of a chemical agent such as cyclophosphamide. Once a consistently low titer, as ascertained by ELISA, is obtained with the membranes used for the tolerization step, the mice are immunized with membranes from cells coexpressing the receptors forming the targeted heteromer. The immunizations are repeated until a high titer is obtained by ELISA at which time the mice are killed and spleens are removed for the generation of mAbs. Once the mAbs are generated, their heteromer selectivity is characterized by ELISA, immunofluorescence, or Western blot analysis using cells that express the individual or a combination of receptors or tissues from wild type and animals lacking one of the receptors. These steps are described below using the μOR–δOR heteromer as a model system.
For the generation of μOR–δOR heteromer-selective mAbs (μOR–δOR mAb), we used membranes from human embryonic kidney 293 (HEK-293) cells (American Type Culture Collection, Manassas, VA) for the tolerization step and from cells coexpressing Flag-μOR and myc-δOR (HEK-μδ) for the second immunogenic step.
Subtractive immunization is carried out in Balb/c mice (6-week-old female). Use at least three mice for immunization. Prior to immunization, collect blood from the tail vein to prepare preimmune serum (described in Section 2.6).
To generate hybridoma-secreting clones, splenocytes from immunized mice are fused with mouse myeloma cells (Gupta et al., 2010; Kohler & Milstein, 1975). We use SP2/0-Ag-14 myeloma cells (American Type Culture Collection, Manassas, VA) that have the machinery necessary for antibody secretion although on their own they do not secrete antibodies (Shulman et al., 1978). SP2/0-Ag-14 cells have a mutation in one of the enzymes of the salvage pathway for purine nucleotide biosynthesis rendering the latter nonfunctional (Shulman et al., 1978). The fusion process is relatively inefficient, given that only 1 in 105 of starting cells form viable hybrids (Shulman et al., 1978). The unfused splenocytes do not grow in culture and eventually die. To kill unfused myeloma cells, compounds that block de novo purine biosynthesis such as aminopterin, methotrexate, or azaserine are used (Shulman et al., 1978). Fused myeloma cells do not die under these conditions since they have a functional salvage pathway derived from the splenocyte fusion partner (Shulman et al., 1978).
We routinely purify mAbs using a protocol described by Ey, Prowse, and Jenkin (1978) that takes advantage of protein A-Sepharose CL-4B beads.
Serum is prepared from blood collected from the tail vein of mice before (preimmune serum) and after (immune serum) immunization.
ELISA, a technique commonly used to screen for the presence on an antigenic epitope on membrane preparations, in whole cells or tissue sections, can also detect GPCRs by using antibodies that recognize endogenous receptors or epitope tags (Flag, myc, or HA) present in the N-terminal region of the receptors. We use ELISA to screen for and determine the selectivity of heteromer-selective antibodies (Gupta et al., 2010; Rozenfeld et al., 2011). In the case of μ-δ mAb, we used membranes from (i) HEK cells, (ii) HEK cells expressing individual receptors, and (iii) HEK-μδ cells to detect hybridoma clones that gave a high titer with only HEK-μδ (Gupta et al., 2010). We also examined the selectivity of μ-δ mAb clones using membranes from the brain of wild type and mice lacking μOR, δOR, or both. We found that the μ-δ mAb recognized an epitope present only in membranes from wild type but not from knockout animals (Gupta et al., 2010). We examined whether the μ-δ mAb exhibited cross-reactivity with other heteromers involving μOR or δO using membrane preparations from cells coexpressing μOR–α2A adrenergic receptor (α2AR), μOR–CB1R, δOR–α2AR, or δOR–CB1R (Gupta et al., 2010). We found that the μ-δ mAb did not exhibit a significant degree of cross-reactivity with other heteromers involving μOR or δOR (Gupta et al., 2010). In addition, we used ELISA to determine the effect of a pathological condition such as development of tolerance to morphine (μOR–δOR heteromers) or chronic ethanol treatment (CB1R–AT1R heteromers) on heteromer levels (Gupta et al., 2010; Rozenfeld et al., 2011). Thus, ELISA is a versatile technique that can be used to determine heteromer levels under pathological techniques. Below, we describe a routinely used ELISA protocol.
This ELISA protocol can be adapted to monitor surface heteromer levels following chronic treatment with drugs. For example, cells (~2.5×105 cells/well) coexpressing μOR and δOR are seeded in growth media in a 24-well plate. After cells have attached, they are treated with morphine or any lipophilic opiate, preferably an antagonist (1 μM final concentration in growth media) for different time periods (0–48 h). Controls are treated with same amount of growth media. At the end of the incubation period, media are removed, plates are placed on ice, and wells are rinsed with 200 μl of cold PBS. Cells are fixed with ice-cold methanol for 5 min followed by two washes with 500 μl PBS. Cells are then subjected to ELISA using primary and secondary antibodies as described above.
ELISA can also be carried out in tissue sections, to compare regional differences in antibody recognition in control and pathological conditions and to examine the effect of drugs/ligands on receptor levels. For this, 10-μm tissue sections are placed on Fisher Brand Superfrost Plus slides (Fisher Scientific, PA) and circled immediately with ImmEdge PAP pen (Vector Laboratories, Inc., CA) to form a water-proof barrier; the resulting wells hold approximately 200 μl solution. ELISA is then carried out as described above.
The most common problems encountered during the ELISAs are elevated background, poor, or absence of a signal. Elevated background signal can be due to inadequate washing and draining of the wells, contamination of the substrate solution with metal ions or oxidizing reagents. The latter can be avoided by using only distilled/deionized water in the preparation of the different solutions and by using clean plastic-ware. Prior substrate exposure to light leads to poor signal. The substrate should be prepared about 10 min before use and should be kept in the dark till use. Sometimes, the signal develops very rapidly (i.e., in less than a minute), giving very high titers that are outside the range of the plate reader. For this reason, pilot experiments using different dilutions of primary and secondary antibodies should be carried out to establish conditions where optimum signal is obtained after 10 min of incubation with substrate.
Immunofluorescence uses fluorescently labeled secondary antibodies to visualize proteins in cells and tissues and can provide information about the tissue distribution of a given protein as well as its subcellular distribution. We used immunofluorescence to show that coexpression of μOR with δOR leads to intracellular retention of μOR–δOR heteromers in the Golgi apparatus, and this is rescued by the expression of RTP4 (Decaillot, Rozenfeld, Gupta, & Devi, 2008). We used this technique to visualize μOR–δOR heteromers in the brains of wild-type mice (but not μOR or δDOR knockout mice) treated chronically with morphine under a paradigm that leads to the development of tolerance. The increases in μOR–δOR heteromer levels were robust in the medial nucleus of the trapezoid body and in the rostral ventral medulla, brain regions involved in the processing of painful stimuli (Gupta et al., 2010). In the case of CB1R–AT1R heteromers, immunofluorescence studies showed that heteromerization leads to surface expression of CB1R and colocalization of CB1R and AT1R at the cell surface of hepatic stellate cells that were chronically treated with ethanol (Rozenfeld et al., 2011). The protocol that we use for immunofluorescence is as follows:
Common problems encountered with immunofluorescence are high background and sensitivity particularly of heteromer-selective antibodies to fixatives. High background can be reduced by adding 0.1% Tween-20 to washes or increasing the concentration of normal serum (5–10%) used for blocking. If antibodies are sensitive to fixatives, then changing the fixation conditions (glutaraldehyde, methanol, or 10% formalin) followed by extensive washing to remove all traces of the fixative is recommended.
Immunoprecipitation is a useful technique to demonstrate heteromerization between differentially epitope-tagged GPCRs. After cell lysis, one of the receptors is immunoprecipitated using antibodies recognizing the epitope tag present on the receptor (myc-tagged receptors are immunoprecipitated using anti-myc antisera). The immunoprecipitates are then subjected to SDS-PAGE under nonreducing conditions, and Western blots are probed with antibodies to the epitope tag present to the other receptor. In order to avoid cross-reactivity, antibodies from two different species are used. A signal is detected in the blots only if there is an association between the two epitope-tagged receptors. We used this strategy to demonstrate heteromerization between μOR and δOR (Gomes et al., 2004, 2000), δOR and κOR (Jordan & Devi, 1999), μOR and α2AR (Jordan et al., 2003), δOR and α2AR (Rios et al., 2004), δOR and CB1R (Rozenfeld et al., 2012), as well as CB1R and AT1R (Rozenfeld et al., 2011). When examining heteromerization in endogenous tissue, antibodies that selectively recognize each receptor or the heteromer are needed (Gomes et al., 2004; Gupta et al., 2010; Rozenfeld et al., 2012, 2011). In the case of the heteromer, the selective antibody could be used to immunoprecipitate the heteromer, and antibodies to individual receptors to probe the Western blots (Gupta et al., 2010). Coimmunoprecipitation and Western blot analysis can also be used to examine interactions between GPCR heteromers and signaling molecules such as β-arrestins, or chaperones such as RTP-4, AP-3, or AP-2. Such studies showed that μOR–δOR heteromers are constitutively associated with β-arrestin (Rozenfeld & Devi, 2007), and that RTP-4 interacts with μOR–δOR heteromers (Decaillot et al., 2008). Thus, immunoprecipitation and Western blotting could be used to (i) probe heteromer levels under healthy and pathological conditions, (ii) identify differences in the levels of heteromer-associated proteins, or (iii) help identify novel proteins associated with the heteromer under pathological conditions. The following sections describe the different steps involved in coimmunoprecipitation and Western blotting.
Different buffers can be used to lyse transfected cells or tissues. Two important considerations in the choice of lysis buffer are the efficient solubilization of the desired receptor without affecting receptor associations and whether the lysis buffer interferes with receptor recognition by the antibody used for immunoprecipitation. Variables that can drastically affect the solubilization of proteins are salt concentration, pH, and type of detergent used. Lysis buffers commonly used to examine receptor interactions include
Artifactual receptor aggregation during solubilization/immunoprecipitation due to the inherent hydrophobic nature of GPCRs is a major concern when examining receptor heteromerization. To rule this out, a variety of solubilization conditions including different combinations of detergents have to be used. In addition, controls where cells expressing individual receptors are mixed prior to solubilization and immunoprecipitation have to be used. If the heteromers are observed only in cells coexpressing both receptors and not in the mixed cells, this would suggest that they are not the result of artifactual aggregation.
Harsh solubilization procedures can disrupt receptor associations. Cross-linking reagents have been used to address this concern (Cvejic & Devi, 1997). The presence of receptor heteromers in the presence of cross-linking reagents, irrespective of the latter’s functional properties, would then suggest that they stabilize the interactions and do not induce receptor heteromerization.
To prevent the masking of the antigenic epitope due to the presence of nonspecific proteins, it is advisable to preclear the cell lysate with normal serum (e.g., rabbit serum if using rabbit polyclonal antibody for immunoprecipitation) followed by binding to protein A beads. This removes proteins that bind nonspecifically to the antibody or to the beads.
The presence of a high background or nonspecific bands in Western blots can be minimized through the use of a different blocking buffer, reducing the time of incubation with primary/secondary antibody, using harsher conditions for washing membranes after antibody incubation (e.g., 50 mM Tris–Cl, pH 7.5 containing 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, and 0.1% sodium dodecyl sulfate), or adding detergent to the primary/secondary antibody preparation up to a concentration of 1%.
Recent evidence indicates that some GPCR heteromers are upregulated in disease states. However, the physiological role of these heteromers in pathology is not clearly understood due to the lack of tools to distinguish between heteromer- and homomer-mediated effects. The development of heteromer-selective antibodies and of specific heteromer disrupting TAT peptides could help elucidate their role in pathological conditions. This is clearly shown in the case of μOR–δOR heteromers where selective antibodies detect increased heteromer levels in brain regions involved in pain perception following chronic treatment with morphine. Studies with TAT peptides that selectively disrupt μOR–δOR heteromers suggest that the latter may keep morphine-mediated signaling via μOR in the desensitized state. Thus, GPCR heteromers could be novel targets for the development of therapeutics to treat diseases. This, however, would require an understanding of the functional role of GPCR heteromers during pathology. The different protocols described in this review used in combination with the heteromer-selective antibodies or with peptides that selectively disrupt the heteromers could help in elucidating these roles.
L. A. D. is supported by NIH grants DA008863 and DA019521.