Specific detection of CB1 receptors; cannabinoid CB1 receptor antibodies are not all created equal!

https://doi.org/10.1016/j.jneumeth.2008.02.014Get rights and content

Abstract

The study of endogenous cannabinoid CB1 receptor proteins in neuronal tissues and cells relies on the availability of highly specific antibodies. We have tested the ability of a series of CB1 antibodies to detect endogenous receptors in brain as well as hemagglutinin (HA)-tagged receptors transfected into HEK-293 cells using a combination of immunological methods. An initial comparison of several CB1 antibodies in mouse brain revealed substantial differences in staining pattern to ligand binding by autoradiography. Antibodies were then tested immunocytochemically against HEK cells expressing HA-tagged rat and human CB1 receptors. None of the commercial antibodies tested were able to detect the receptor in this context. All antibodies were then screened by Western blotting using lysates from the HEK cells and rodent brain homogenates. Again, none of the commercially available antibodies detected proteins of the correct molecular weight in transfected cell lines or brain homogenates, although all recognized multiple proteins in brain tissues. We conclude that the commercially available antibodies we tested failed to detect CB1 receptors abundantly expressed in HEK cells or native receptors in brain slices or homogenates. As such, comprehensive validation of the specificity of these CB1 antibodies for a particular application is essential before use.

Introduction

Much cannabinoid research worldwide is focused on understanding the molecular events that govern the behavior of endogenous cannabinoid receptors in the brain. This includes aspects of expression, signaling, endocytosis, trafficking, and degradation, all of which ultimately relate to the function of endogenous CB1 receptors within the brain. This area of research is highly relevant clinically, especially in light of the multiple cannabinoid receptor-targeted drugs currently in the market (Marinol, Cesamet, Sativex, and Rimonabant-Acomplia), with more in various stages of development (Rimonabant for smoking cessation, IP751 see www.indevus.com).

An important requirement in order to study endogenous CB1 receptor proteins in brain or neuronal cultures is the availability of reliable and highly specific antibodies. Over the last decade several CB1 antibodies targeting different regions of the receptor have been developed by research groups with others becoming available through commercial sources. The rationale for requiring region-specific CB1 antibodies is driven partly by experimental design and partly by the consideration that a single antibody is unlikely to detect all receptor species (i.e. post-translationally modified, splice-variants, receptor oligomers, and degradation products).

The first antibodies to recognize CB1 in brain tissue by histology and also recognize denatured CB1 proteins on immunoblots were raised against the extreme amino-terminus (Pettit et al., 1998, Tsou et al., 1998, Tsou et al., 1999). Antibodies raised to the carboxy-terminus of CB1 have also been used to successfully detect receptors in rodent brain tissue, encompassing the entire carboxy-terminal region (Hajos et al., 2000, Katona et al., 2001, Bodor et al., 2005, Matyas et al., 2006) and also the last 13–15 amino acids (Egertova and Elphick, 2000, Egertova et al., 2003, Nyiri et al., 2005, Eggan and Lewis, 2006). The last 13–15 amino acids (461–472) have recently been shown to interact with GASP-1 to mediate receptor degradation (Martini et al., 2007), while other regions within the C-terminus are involved in G-protein coupling and signaling (Mukhopadhyay and Howlett, 2001, Nie and Lewis, 2001). Antibodies directed to the N-terminus of CB1 have the valuable advantage of detecting cell surface receptors in living cells; thus, enabling visualization of receptor trafficking in real-time. However, the N-terminus of CB1 is glycosylated and very little is known regarding the conformational folding of the N-terminus in vivo. Studies of other G-protein-coupled receptors strongly suggest that the receptors likely exist as part of a much larger complex structure termed a ‘receptosome’ (McDonald et al., 2000, Bockaert et al., 2004). The CB1 receptosome complex may include other receptors, like the dopamine D2 receptor which has been shown to dimerise with CB1 under specific conditions (Kearn et al., 2005), in addition to G-proteins (Mukhopadhyay and Howlett, 2001) and as yet unidentified proteins involved in receptor signaling and trafficking. These receptor modifications and interactions with other proteins may be stable or transient and should be considered carefully as they have the potential to mask an epitope (G-proteins and GASP-1) and prevent antibody binding or even allow epitope availability under certain receptor conformations, such as receptor activation or oligomerization.

Considering each of the above points it is clear that a single antibody is unlikely to detect all receptor species. Therefore, specific antibodies that detect both N-terminal (NT) and C-terminal domains of the CB1 receptor are vital in order to investigate the various conformers and complexes in which CB1 receptors may be found. However, in the course of validating CB1-specific antibodies obtained from reputed commercial sources our preliminary histological studies revealed surprising discrepancies in the specificity of the CB1 antibodies. We, therefore, investigated their specificity using a combination of immunocytochemistry and Western blotting of hemagglutinin (HA)-tagged CB1 receptors in HEK cells and endogenous CB1 proteins from mouse and rat brain. We report here that the commercially available anti-CB1 antibodies tested in this study show poor specificity for the receptor, and of particular concern, detect non-specific proteins of similar molecular weight.

Section snippets

Tissue preparation

All tissues and sections used in this study were obtained under the strict guidance and approval of the University of Auckland Animal Ethics Committee.

Mice were sacrificed by cervical dislocation and the brains removed and snap frozen for −80 °C storage. Brains (n = 4) were cryostat cut and sections of 16 μm thickness were thaw mounted onto chrome-alum-gelatine coated slides.

For preparation of western homogenates, blocks of mouse and rat brain encompassing the hippocampus and striatum were prepared

Immunohistochemical distribution of CB1 in the mouse brain

CB1 distribution has been well characterized in the mammalian brain by receptor autoradiography and in situ hybridization (Glass et al., 1997). Immunohistochemistry on tissue sections enables identification of regional as well as cellular distribution. In the current study we have used the well characterized L15 CB1 antibody (Nyiri et al., 2005, Eggan and Lewis, 2006, Chen et al., 2007, Haring et al., 2007) and [3H]-CP55, 940 ligand binding to locate CB1 receptors in mouse brain for comparison

Discussion

The expression of the CB1 receptor has been extensively mapped by histology in the rodent brain (Pettit et al., 1998, Tsou et al., 1998, Tsou et al., 1999, Egertova and Elphick, 2000) using several epitope-specific antibodies. The C-terminal L15 antibody which was developed by Mackie and colleagues (Nyiri et al., 2005) recognizes the last 15 amino acids of CB1. A similar epitope (last 13 residues) is detected by the antibody developed by Elphick and colleagues (Egertova and Elphick, 2000).

Acknowledgements

We would like to thank Prof. Ken Mackie for generously providing the L15 and L73 antibodies. This research was funded by the Neurological Foundation of New Zealand and the Royal Society of New Zealand Marsden Fund.

References (29)

  • K. Tsou et al.

    Cannabinoid CB1 receptors are localized primarily on cholecystokinin-containing GABAergic interneurons in the rat hippocampal formation

    Neuroscience

    (1999)
  • M. Uhlen

    Mapping the human proteome using antibodies

    Mol Cell Proteomics

    (2007)
  • J. Wager-Miller et al.

    Dimerization of G-protein-coupled receptors: CB1 cannabinoid receptors as an example

    Chem Phys Lipids

    (2002)
  • J. Bockaert et al.

    GPCR-interacting proteins (GIPs): nature and functions

    Biochem Soc Trans

    (2004)
  • Cited by (0)

    View full text