There have been significant advances in the discovery and existence of the endocannabinoid system since the isolation and elucidation of the structure of Δ9-tetrahydrocannabinol (Δ9-THC), the main active component of Cannabis sativa [1]. Despite a considerable amount of research occurring since this discovery, the mode of action of the cannabinoids still remained a mystery. The first cannabinoid receptor CB1 was identified by Devane et al. in 1988 [2] and cloned in 1990 [3]. This led to the discovery of the first endocannabinoid, termed anandamide [4]. A second cannabinoid receptor subtype, CB2, was found in spleen but absent in brain [5], and this was followed by the discovery of another key ligand, now known as 2-arachidonoylglycerol (2-AG) [6]. Over the last decade, work on the endocannabinoids has expanded greatly, transforming endocannabinoid research into mainstream science. Anandamide and 2-AG have become key molecular players in this, with many implications in health and disease.
The endocannabinoid signalling system
There a number of molecules involved in endocannabinoid signalling, including the endocannabinoids, the cannabinoid receptors they bind to, and the enzymes involved in their synthesis and degradation [7]. The endocannabinoids are a group of lipid-derived second messengers that bind to and activate the cannabinoid receptors. They are thought to be involved in the regulation of a number of brain functions, including memory, learning, pain, mood and appetite [8]. Unlike classical neurotransmitters, endocannabinoids are rapidly synthesised on demand in neurons from membranous lipid precursors. This is thought to be triggered by a number of mechanisms. Postsynaptic depolarisation and subsequent Ca2+ influx activates various enzymes involved in the synthesis of the endocannabinoids [9]. Activation of postsynaptic Gαq/11-coupled group I metabotropic glutamate (mGluRα1) or acetylcholine receptors and the phospholipase C (PLC) pathway also leads to the production of 1,2-diacylglycerol (DAG) [10]. Both pathways appear to be independent of each other, allowing cooperative stimulation of endocannabinoid synthesis [11].
Recently, anandamide and 2-AG have been shown to be involved in the retrograde regulation of synaptic transmission, most commonly at inhibitory presynaptic cannabinoid receptors [12]. The release of a number of neurotransmitters is inhibited as a result. Endocannabinoid-mediated retrograde signalling was originally located in the cerebellum [10] and hippocampus [13], and later on in other brain regions [14]. One common short-term retrograde mechanism of action is described as depolarisation-induced suppression of inhibition (DSI) [13], whereby postsynaptic depolarisation causes Ca2+-dependent endocannabinoid synthesis. This causes activation of presynaptic CB1 cannabinoid receptors on inhibitory interneuron terminals, inhibiting presynaptic voltage-gated Ca2+ channels. This reduces GABA neurotransmitter release, particularly in hippocampal pyramidal neurons [15], resulting in an overall increase in neuron excitability. Recordings from hippocampal neurons with inhibitory synaptic connections show transient suppression of inhibitory synaptic currents in presynaptic terminals, mainly due to the reduction of GABA release [16]. Similarly, depolarisation-induced suppression of excitation (DSE) causes short-term inhibition of glutamatergic transmission, decreasing excitability, particularly in cerebellar Purkinje cells [12]. Endocannabinoids are also involved in the induction of longer lasting forms of plasticity, namely long-term depression (LTD) and long-term potentiation (LTP) [12]. A summary of the endocannabinoid signalling mechanism is illustrated in Figure 1 [17].
Occurrence, biosynthesis and metabolism of the endocannabinoids
Until the end of the 20th century only two endocannabinoids, N-arachidonoylethanolamine (anandamide, AEA) and 2-AG had been discovered [4,6]. However, three more endocannabinoids have recently been proposed: 2-arachidonoylglyceryl ether (noladine ether), O-arachidonoylethanolamine (virodhamine) and N-arachidonoyldopamine (NADA) [18]. Whilst anandamide was discovered prior to 2-AG, there is reason to believe that 2-AG is the true natural ligand for the cannabinoid receptors. Being the most abundant endocannabinoid in the brain [19], 2-AG is the only ligand which is a full agonist at both CB1 and CB2 receptors, with similar affinity for both [20]. It is also more potent than anandamide in stimulating CB1 receptor-dependent G-protein activity [21]. Anandamide has similar potency to Δ9-THC at the CB1 receptor, but acts as a partial agonist at the CB2 receptor [19].
Anandamide and 2-AG are regulated by distinct biosynthetic and inactivation pathways, involving numerous enzymes and precursors. This reflects the diverse range of functions that the endocannabinoids are involved in. Clarification of these pathways thus plays an important step in understanding the complexity of the endocannabinoid signaling system. Figure 2 indicates the pathways involved in the biosynthesis, action and inactivation of the two key endocannabinoids, anandamide and 2-AG [22].
As a common intermediate in different metabolic pathways, 2-AG is highly abundant in many cells, with brain levels two to three times higher than those of anandamide [20]. In neurons, 2-AG is synthesised on demand from DAG by sn-1-specific DAG lipases (DAGL; DAGLα and DAGLβ) [20]. These have recently been cloned and characterised by Bisogno et al. [23]. They have a molecular mass of 120 kDa and 70 kDa respectively and have been shown to contain lipase-3 and serine lipase motifs. It is believed that they have four transmembrane domains but differ in the C terminal tail. They are mostly found in the plasma membrane and are stimulated by Ca2+. The expression of DAGLα and DAGLβ has been shown to correlate well with 2-AG and CB1 levels in the rodent brain, with the expression pattern matching its proposed physiological roles during development and in the adult (discussed later) [23,24].
As previously highlighted, endocannabinoid biosynthesis appears to be mediated by a number of distinct enzymatic pathways: Ca2+-dependent endocannabinoid release, activation of the Gαq/11-coupled metabotropic glutamate receptor/PLC pathway, and a combination of both methods [9,10,11]. The biosynthetic and inactivation pathways for anandamide differ to 2-AG. These are illustrated in Figure 2. There are two primary routes of 2-AG biosynthesis in neurons. One route involves the hydrolysis of phosphatidylinositol (PI) by phospholipase A1 (PLA1) and the subsequent generation of a lysophospholipid (lyso-PI), which is hydrolysed by a lyso-PI specific PLC to form 2-AG [25]. The more widely described route involves the PLC-mediated hydrolysis of membrane phospholipids in order to produce DAG. PLC synthesises inositol triphosphate (IP3) and DAG from PI, where IP3 releases Ca2+ from internal stores. DAG is metabolised by DAG kinase (DAGK) or converted to 2-AG by subsequent sn-1-specific DAGL activity [25]. DAG and 2-AG are key second messenger signalling molecules, highlighting the importance of DAGLα/β regulation. The use of various metabolic inhibitors of PLC and DAGL in preventing 2-AG formation in cultured cortical neurons indicates that the PLC/DAGL pathway plays a primary role in 2-AG biosynthesis [20]. The importance of this pathway in depolarisation-induced generation of 2-AG in rat brain synaptosomes has also been observed [26].
Following its synthesis, 2-AG is released from postsynaptic neurons to act on presynaptic CB1 receptors. As endocannabinoids are lipophilic molecules, it would be expected that their release following synthesis and their reuptake prior to inactivation occurs by passive diffusion. However, experimental evidence indicates that this postsynaptic release is mediated by an endocannabinoid membrane transporter [27]. It is yet to be cloned and characterised, and as such the existence of this transporter remains controversial. Termination of endocannabinoid signalling is achieved by neuronal reuptake through endocannabinoid transporters, followed by a number of enzymatic reactions. Both 2-AG and anandamide are inactivated intracellularly by enzymatic hydrolysis, yielding glycerol and arachidonic acid, which is itself a potent second messenger. 2-AG acts as a substrate for several enzymes including monoacylglycerol lipase (MAGL), fatty acid amide hydrolase (FAAH), cylooxygenase-2 (COX-2) and lipoxygenase (LOX). FAAH, which demonstrates a broad specificity, is the principal enzyme involved in the inactivation of anandamide and has also been shown to hydrolyse 2-AG [26]. Whilst levels of anandamide rise greatly in FAAH knockout (KO) mice, 2-AG levels are unaffected, suggesting the involvement of a number of alternative metabolic pathways for 2-AG [28]. Hence hydrolysis of 2-AG by MAGL is the most common degradation pathway [26]. This has been supported by a number of findings which indicate 2-AG degradation on transfer of the MAGL gene [29]. Its expression pattern correlates well with the DAGL enzymes, indicating a close proximity between the various endocannabinoid signalling components (see later) [24]. The involvement of 2-AG and anandamide in non-cannabinoid pathways reflects the diverse range of functions that these messengers are involved in. For example the synthesis of a prostaglandin PGH2-G, which is derived from the action of COX-2 on 2-AG, is inhibited by the DAGL inhibitor RHC80267 [30].
The cannabinoid receptors
It is now widely established that the psychoactive effects of Δ9-THC in Cannabis sativa are primarily mediated through CB1 cannabinoid receptors, whilst its therapeutic properties are mediated mainly through CB2 receptors [7]. The transduction of signals in response to the endocannabinoids also occurs via the CB1 receptor. The CB1 and CB2 cannabinoid receptors are seven-transmembrane spanning G-protein coupled receptors (GPCRs) which belong to the class A rhodopsin GPCR family [31]. Both receptors are coupled to Gαi/G0, and treatment with pertussis toxin results in the abolition of various biological responses mediated through them [26]. CB1 is mainly expressed in the nervous system and in various peripheral tissues [3], whilst CB2 is believed to be predominantly expressed in the immune system, including the spleen, where it was first localised [5].
The CB1 receptor is one of the most abundant receptors in the brain and it has been shown to be expressed at high levels in the molecular layer of the cerebellum, the hippocampus, cerebral cortex and basal ganglia [13,31]. Increasing evidence suggests that CB1 is the predominant presynaptic cannabinoid receptor in the brain at both inhibitory and excitatory synapses. This includes immunohistochemical, biochemical and electrophysiological studies using CB1 knockout mice [32]. Whilst CB1 and CB2 are well established as the two main cannabinoid receptor subtypes, evidence for additional cannabinoid receptors is also emerging. This includes the orphan receptor GPR55, which has been shown to bind a synthetic cannabinoid ligand, as well as anandamide and virhodamine [33]. Other groups have not found such evidence, however, making this a disputed matter. Such examples highlight the complexity of the endocannabinoid signalling system.
The cannabinoid receptors have the potential to couple to a wide array of intracellular responses. These include the regulation of ion channel activity, the release of Ca2+ from internal stores and regulation of mitogen-activated protein kinase (MAPK) signalling. Following its postsynaptic synthesis and release, 2-AG binds to and activates presynaptic CB1 receptors. These couple through Gαi/G0 to regulate Ca2+ and K+ channels and reduce neurotransmitter release. CB1 receptor activation leads to the inhibition of adenylate cyclase (AC), reducing cyclic AMP (cAMP) and protein kinase A (PKA) activity [17]. This results in the inhibition of N- and P/Q-type Ca2+ channels and the activation of hyperpolarising K+ channels [34]. CB1 receptor signaling has been shown to positively couple to N- and L-type Ca2+ channels on neuronal growth cones [35]. Hence attenuation of Ca2+ influx results in the suppression of neurotransmitter release. This can inhibit or disinhibit neuronal signalling depending on whether the CB1 receptor is expressed on glutamatergic or GABAergic terminals [17].
Physiological roles and function of the endocannabinoid system
Whilst Δ9-THC has been shown to be involved in many physiological processes, including memory, learning and appetite [8], less is known about the roles of the endocannabinoids. Although the endocannabinoid system (via the CB1 receptor) is involved in the regulation of a number of brain functions including those just mentioend, associations with the immune, endocrine, cardiovascular and reproductive systems have also been found in animals and humans [31]. These roles include regulation of cell development and growth, immune function, vasodilation and reproduction [31]. This is shown in CB1 KO mice which exhibit motor learning deficits, hypoalgesia, anxiety reduction and impaired memory extinction compared to wild type (wt) mice [36]. The diversity of symptoms reflects the extensive CB1 distribution in the brain.
Much recent evidence points to the endocannabinoid system in playing a key role in the development of the nervous system. The expression of a number of key components including the CB1 receptor and DAGLs has been shown in neurons during development [23,24]. Essential roles include the stimulation of axonal growth and guidance during development, proliferation and differentiation of neural progenitor cells, and chemoattractive and chemorepulsive effects on cortical interneurons [24]. Activation of the FGF receptor by FGF-2 or cell adhesion molecules has shown a DAGL-dependent stimulation of neurite (neuronal process) growth. This suggests that DAGLα/β couples growth factor signalling to the synthesis of 2-AG, which activates CB1 receptors during development and thus promotes axonal growth. Importantly, CB1 agonists stimulate this growth, whilst CB1 antagonists prevent this [37]. CB1 antagonists have also been shown to cause defects in axonal growth and guidance, indicating that endocannabinoid signalling is required in early development for normal axonal growth and fasciculation [24].The expression of DAGLs in highly proliferating cells in the subventricular zone (SVZ) of the adult mouse indicates a role for these enzymes in adult neurogenesis. DAGL and CB2 antagonists have thus been shown to inhibit cell proliferation in the SVZ, which is a major site of neurogenesis in the adult [38].
In the adult brain, the main role of the endocannabinoids is in the modulation of synaptic transmission and plasticity. Ca2+-dependent endocannabinoid release is believed to play a role in several forms of synaptic plasticity, with short- and long-term depression and potentiation being important [12,13]. These illustrate the ability of endocannabinoid signalling to modulate neurotransmission at both inhibitory and excitatory synapses. For example, Stella et al. [20] have reported the suppression of LTP by 2-AG in rat hippocampal slices. 2-AG plays an important role in controlling the rate of firing of presynaptic neurons by activating the CB1 receptor and reducing neurotransmitter release. This negative feedback regulation allows neuron stimulation to be attenuated after excitation. This may be of great physiological significance in alleviating pain and protecting neurons, as sustained neuronal activation is known to cause cell exhaustion and may lead to neuronal cell death, which is implicated in a number of neurodegenerative diseases [23,25]. In support of this, administration of 2-AG resulted in a better recovery and reduced neuronal death and infarct size following head injury in mice [39]. In contrast, CB1 KO mice showed a higher mortality rate [40]. It therefore appears that this neuroprotective effect is mediated by an inhibition of excitatory neurotransmitter release.
DAGL expression in the developing and adult nervous system
The cloning and characterisation of the first sn-1-specific DAGLs by Bisogno et al. [23] resulted in the elucidation of two closely related genes. These were identified as DAGLα and DAGLβ isoforms, with their expression in cells correlating with 2-AG biosynthesis and release. In the developing mouse brain, DAGL expression is seen in presynaptic axonal tracts where it is required in the synthesis of 2-AG for axonal growth and guidance. This activates CB1 receptors in the same presynaptic axonal growth cone [38]. All developing axonal tracts appear to coexpress both DAGL isoforms, and expression was seen in the spinal cord and optic nerve [23]. In the adult mouse brain, however, both enzymes are absent from axonal tracts, which contrasts with strong CB1 receptor expression in these areas. Instead, DAGL activity is restricted to postsynaptic dendritic fields where it is involved in retrograde synaptic signalling via the production of 2-AG and subsequent activation of presynaptic CB1 receptors. Overall, these findings are in accord with the changing role of DAGL during development and in the adult. This change in expression of DAGL from axonal tracts to dendritic fields reflects the role of DAGL during brain development as a mediator of neurite growth, and in the adult as a retrograde synaptic messenger [18]. This indicates a high degree of spatial and temporal regulation of DAGLα/β expression and 2-AG synthesis in the brain [23].
Expression of DAGLα appears to be highest in the hippocampus and cerebellar cortex, with moderate levels in the cerebral cortex, olfactory bulb, basal ganglia and thalamus. In the cerebellum, DAGLα and DAGLβ expression appears to be most prominent in the dendritic field and in deep cerebellar nuclei, with DAGLα specifically expressed in tubular-like structures in Purkinje cell dendrites. The more limited expression of DAGLβ suggests a downregulation of this enzyme during development [23]. More detailed studies of DAGL expression have revealed the specific targeting of DAGLα to postsynaptic dendritic spines, with expression also seen in somatodendritic regions [32]. In the cerebellum, DAGLα is predominantly expressed in Purkinje cells, with considerable enrichment at the base of spine neck. Expression is absent from the main body of spine neck and head. In hippocampal pyramidal cells, DAGLα is distributed in the spine head and neck. This indicates a fine specificity in DAGLα distribution depending on the neuron type, and may suggest differential control of the distance between the postsynaptic 2-AG site of synthesis and the CB1 receptor [32].
Many studies have now shown that expression of the DAGLs and the metabolic enzymes for 2-AG correlates with the expression of the CB1 receptor at central synapses. DAGLα is seen to be expressed in close apposition to CB1 receptors, PLC and mGluRα1, indicating a close proximity between key endocannabinoid signalling molecules and allowing for functional signalling [32]. In a study looking at CB1 expression, Watson et al. [24] showed that CB1 is expressed widely in the developing nervous system and that its expression follows neuronal differentiation in the embryo from the earliest stages. DAGLα and DAGLβ are expressed in an overlapping manner in the early embryo, whilst MAGL is expressed at later stages, indicating that neurons in the early stages of development have the ability to both synthesise and respond to 2-AG. MAGL has been shown to occupy presynaptic sites, with its expression correlating with that of DAGL and CB1. This confirms the role of these enzymes in 2-AG signalling. In contrast, FAAH is often expressed postsynaptically on intracellular membranes and does not show a strong correlation with CB1 expression [41].
Aims and objectives
The main aim of this report is to identify the localisation and expression of key endocannabinoid signalling molecules in the nervous system with the use of immunolocalisation and Western blot techniques. 2-AG has been shown to play a major role in the endocannabinoid system and is involved in mediating short and long-term retrograde signalling [12]. Thus the expression of the DAGL enzymes is of particular interest as this will enable the site of 2-AG synthesis to be determined and will clarify the critical role of these enzymes in endocannabinoid signalling. The expression of CB1 receptors, FAAH and MAGL will also be of interest as they have crucial roles in signalling. Tests will be carried out on adult mouse cerebellum, as this region has been shown to have high endocannabinoid pathway component expression [23,32]. Tissue from DAGLα KO and CB1 KO mice will also available, enabling questions regarding the expression of key components of the endocannabinoid system to be addressed.
Preliminary tests on cerebellum tissue and on cell lines will be carried out using a Western blot technique to determine the expression of DAGLα, DAGLβ, FAAH, MAGL, green fluorescent protein (GFP) and actin in wt, DAGLα KO and DAGLβ KO mice and in cell lines. Detailed immunolocalisation studies will then be carried out in wt, DAGLα KO and CB1 KO mice. The use of immunofluorescence and immunohistochemical staining techniques will enable the specific localisation and expression profiles of these key components to be identified and confirmed with previous findings, and any major anatomical differences between the wt and KO brain samples can be viewed. The use of numerous antibodies against the same antigen will allow the comparison and validation of each antibody. Staining the wt and KO tissue with antibodies against the various components of the endocannabinoid pathway will indicate whether loss of the enzymes that make 2-AG impact on the expression of the other pathway components. This will highlight the relationship between the various components in functional signalling and will enable one to determine if there is a feedback loop present where the components regulate each other's expression. Previous findings have noted a correlation between DAGL and CB1 expression [32], and thus removal of one of these genes may have an affect on the expression of the other protein. Identification of the receptors, ligands and enzymes will provide a full elucidation of endocannabinoid signalling in the brain and its physiological function. This may then be able to highlight the therapeutic potential that manipulation of this crucial system may offer.
Materials and Methods
Generation of DAGL KO mice
The DAGL KO mice were generated in a similar way to that of a recent study by Gao et al. [42]. Mice lacking DAGLα were generated through gene targeting and those lacking DAGLβ through gene trapping. A targeting vector was used to replace a portion of exon 1 of the mouse DAGLα gene with a selection cassette. A GFP gene was inserted in the promoter region where DAGLα should be expressed, disrupting its expression. This GFP tag is inert; when GFP is expressed this proves the promoter is working effectively. Mouse lines were derived from targeted clones using standard procedures and maintained on a C57Bl/6 background. DAGLβ KO mice were generated from a Lexicon OmniBank embryonic stem cell clone OST195261 (Lexicon Pharmaceuticals) which contains a gene trap cassette insertion in the first exon of the mouse DAGLβ gene.
Cell line information and cell culture
COS-7 and Tango cell lines were chosen as model systems to investigate the expression of endocannabinoid signalling pathway components alongside wt, DAGLα KO and DAGLβ KO mouse cerebellum tissue. These were used to confirm the presence of DAGLα and DAGLβ in the cells, acting as an additional control for the antibodies. Where the DAGL enzymes should not be expressed in their respective KO tissues, their expression should be abundant in the respective transfected cell lines (cells overexpressing recombinant human DAGL), reflecting effective KO and transfection processes. If results obtained are not what expected this can therefore be attributed to methodological procedures and not the tissue or cell lines. The COS cell line was derived from transformation of a CV-1 African green monkey kidney fibroblast cell line with a version of the SV40 Simian virus genome [43]. The altered plasmids can replicate in high copy numbers in cells infected with SV40 helper viruses, and the cells are easy to maintain in culture, making this cell line favourable for studying gene expression. COS cells only have a low level functioning endocannabinoid system, naturally expressing low levels of DAGLα and DAGLβ. This makes their use in studying endocannabinoid signalling interesting. Tango™ GPR92-bla U2OS cells contain the human GPCR 92 (GPR92) linked to a TEV protease site and a Gal4-VP16 transcription factor stably integrated into the parental cell line. This expresses a beta-arrestin/TEV protease fusion protein and the beta-lactamase reporter gene [44].
The cell lines were transfected with either plasmids containing V5 tagged human DAGLα or FLAG tagged DAGLβ, enabling tracking of the enzymes using antibodies. This was done using lipofectamine plus (Invitrogen). The DAGLα plasmid used was human DAGLα in pcDNA3.1D/V5-His-TOPO (Invitrogen), whilst a mouse DAGLβ plasmid in pCMV-Tag4A (Stratagene) was used [23]. This results in separate COS and Tango cell lines which show normal expression, or overexpress DAGLα (COSα and Tangoα) or DAGLβ (COSβ and Tangoβ). COS cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 2 mM L-glutamine (Invitrogen), 1mM sodium pyruvate, 10% foetal bovine serum (FBS) and 100 µg/ml neomycin in a humified incubator containing 8% CO2. For passaging, COS cells are trypsinised for three to five minutes with 5 ml trypsin at 37°C, washed off and collected in expansion media, spun down at 240g for five minutes, before finally being re-suspended in fresh expansion media. Cells were plated out onto new gelatine-coated tissue culture plastic and incubated at 37°C in an 8% CO2 incubator. Stable transfected clones were selected using G418. Cell lines were maintained in media containing 300 μg/ml G418 and screened for successful insertion of the vector constructs by western blotting and immunolocalisation as described below. The Tango cell lines were cultured in a similar manner.
Cell lysis, gel electrophoresis and Western blotting
Cell lysis/protein extraction
Cells were cultured in 6-well plates or 10 cm dishes until reaching 80-90% confluence. Cells were washed with ice-cold PBS (2ml/1ml) and lysed in 50-500 μl lysis buffer (137 mM NaCl, 20 mM Tris-HCl (pH 7.4), 10% glycerol, 2mM EDTA, 1% Triton X). 1 mM phenylmethylsulfonyl fluoride (PMSF), complete protease inhibitors (1:50, Roche) and 0.1 mM sodium orthovanadate were also added. The lysate was collected in 1.5 ml eppendorf tubes and left to rotate for half an hour at 4°C, followed by centrifugation at 11,000g for 10 minutes at 4°C to separate out cell debris. The supernatant was collected and stored at -20°C. Protein extraction from mouse brain tissue was followed in a similar manner, instead using a standard solubilisation buffer of 250mM sucrose, 10 mM Tris-HCl (pH 7.4), protease inhibitors and 1 mM PMSF. Mouse cerebellum was homogenised in the buffer on ice at 100 μl/brain part and 0.1% Triton was added. Protein concentration was measured with a Pierce BCA Protein Assay Kit (no. 23223) following the manufacturer's protocol. The reactions were carried out in a clear 96-well plate.
Gel electrophoresis
The cerebellum, COS and Tango samples were diluted to obtain equal protein concentrations, and 15-60 µl of each sample were loaded into wells ready to run through the gel. The samples were separated on SDS-polyacrylamide gels (Table 1) at 120 V for approximately two hours in running buffer (Table 2), and then transferred to nitrocellulose hybond membranes (Amersham) at 100 V for 60 minutes in transfer buffer (Table 2).
Table 1: Components of SDS-acrylamide gel
Two different gels were made, consisting of the 12.5% or 7.5% gel with the 5% stacking gel. The 12.5% gel was used for components that had low molecular weights as these were easier to detect on the Western blots obtained from these.
Chemical
12.5% gel
7.5% gel
5% stacking gel
30% acryl amide/Bis (VWR International)
2.5 ml
1.67 ml
1.5M Tris, pH 8.8 (VWR International)
3.7 ml
1.25 ml
10% SDS (VWR International)
100 μl
100 μl
H2O
3.63 ml
6.91 ml
TEMED (National Diagnostics)
6.6 μl
20 μl
10% Ammonium persulphate, APS
(VWR International)
66 μl
50 μl
Table 2: Components of running and transfer buffers
The buffer solutions were made up to 1l with distilled water and thoroughly mixed, before being poured in to cover the gels.
Chemical
10x running buffer
10x transfer buffer
Tris Base
30.3 g
30.3 g
Glycine
144.2 g
144.2 g
SDS
10 g
-
Methanol
-
200 ml
Western blotting
The membrane was blocked in 1% Tris Buffered Saline (TBS) and 5% milk at room temperature for one hour, in order to block unspecific membrane binding of the primary antibody. Primary antibody in 0.1% TBS-T 5% milk was applied to the membrane at room temperature and left to rotate for an hour (Table 3). This was followed by washing the membrane four times in TBS-T for 10 minutes each to remove loosely attached antibody. The membrane was then incubated with the secondary antibody solution in 0.1% TBS-T 5% milk for one hour at room temperature. The secondary antibody was chosen against the animal source from which the primary antibody was derived (Table 4). Four more washings in TBS-T followed for 10 minutes each, and finally two 10 minute washings in TBS. The blot was developed using an ECL (or ECL Plus) Western Blotting Detection System kit (GE Healthcare) and exposed to X-ray film according to conditions recommended by the manufacturer. Following blot development, membranes can be reused by stripping with Re-Blot Plus Strong (Millipore), and blocking and antibody application can be repeated with another antibody.
Table 3: Primary antibodies for Western blotting
Rabbit DAGLα affinity-purified antibodies were raised against mouse DAGLα in the method outlined by Yoshida et al. [32]. DAGLβ rabbit antibodies were raised and affinity-purified against the SSDSPLDSPTKYPTL epitope in DAGLβ [23]. The remaining four primary antibodies used for Western blotting were obtained from commercial sources.
Antibody
Species
Dilution
Source/ company
Antibody detail
DAGLα
Rabbit
1:1000
Prof Masahiko Watanabe
Works best with Western blots
DAGLβ
Rabbit
1:250
Eurogentec
As used by Bisogno et al. [23]
586
FAAH
Mouse
1:1250
Abcam
Ab54615
MAGL
Goat
1:1000
GFP
GFP should be expressed in DAGLα KO mice
Actin
Mouse
1:20000
Invitrogen
Actin acts as a loading control
Table 4: Secondary antibodies for Western blotting
Antibody
Species
Dilution
Source/ company
Anti-mouse IgG
1:3000
Anti-rabbit IgG
1:3000
Anti-goat IgG
1:3000
Immunostaining techniques
Immunoperoxidase staining
A protocol similar to that followed in the study by Bisogno et al. [23] was adopted. Saggital sections of formalin-fixed, paraffin wax embedded mouse cerebellum tissue were obtained from adult wt, DAGLα and CB1 KO mice. They were mounted on Superfrost Plus slides and left overnight at room temperature. Sections were dewaxed in xylene and 100% alcohol, and incubated in 3% hydrogen peroxide for 10 minutes, before being thoroughly rinsed in water. Heat-mediated antigen retrieval was carried out in citric acid (pH 6.5) until boiling to disclose antigenic sites. Slides were washed for 10 minutes with water and cooled, before being briefly drained and covered with blocking solution (1% BSA in 1x TBS and sodium azide, pH 7.6) for 15 minutes. Overnight incubation (16 hours) in the primary antibody solutions followed at room temperature (Table 5). After washing in TBS, the sections were incubated with the corresponding biotinylated secondary antibody diluted in blocking buffer, instead for 60 minutes at room temperature (Table 6). This was followed by washing and detection with a StreptABComplex/HRP kit (VectorLabs). Slides were developed for 10 minutes in DAB solution (Sigma D5637) to stain antigen sites brown, and counterstained using haematoxylin to stain nuclei blue. Differentiate using 0.5% HCl in 70% alcohol if required and wash under running water. Finally, the samples were dehydrated by rinsing in 100% alcohol followed by xylene, and the slides were mounted using a solvent-based plastic mountant (DPX), leaving to set. Slides were viewed on an Axiovert 135 microscope (Carl Zeiss) and images captured using an AxioCamMR3 camera (Zeiss) using AxioVision software. Images were processed using Adobe Photoshop, with brightness and contrast modifications made.
Table 5: Primary antibodies for immunoperoxidase staining
Commercial and lab-derived DAGLα affinity-purified antibodies were used to enable comparison of the antibodies, as with the CB1 antibodies. The anti-CB1 antibody supplied by Professor MR Elphick was raised and affinity-purified against the KYTMSVSTDTSAEAL epitope of the C-terminal 13 amino acids of the rat CB1 receptor [45].
Antibody
Species
Dilution
Source/ company
Antibody detail
DAGLα
Goat
1:400
Everest
Commercial antibody
DAGLα
Rabbit
1:50
Wolfson CARD
Lab antibody (651)
CB1
Rabbit
1:1000
Prof MR Elphick
Typical CB1 antibody used
CB1
Mouse
1:10
Sigma
CB1
Rabbit
1:2.5
Affinity BioReagents
Antibody PA1-745
CB1
Rabbit
1:50
CAMAN
CB1
Goat
1:100
Everest
GFP
Chick
1:200
GFP
1:150
Invitrogen
Table 6: Secondary antibodies for immunoperoxidase staining
Antibody
Species
Dilution
Source/ company
Anti-mouse IgG
Goat
1:200/300
Dako E0433/ Vectorlab BA-9200
Anti-rabbit IgG
Goat
1:200/300
Dako E0432/ Vectorlab BA-1000
Anti-goat IgG
Anti-chick IgG
Immunofluorescence staining
The procedures undertaken were similar to that of the immunoperoxidase staining, except saggital sections of formalin-fixed, paraffin wax embedded mouse cerebellum tissue were only obtained from adult wt and DAGLα KO mice. Sections were placed in 50mM ammonium chloride for 20 minutes before being blocked. They were subject to double labelling and incubated with two primary antibodies (Table 7), followed by incubation with a fluorochrome-conjugated secondary antibody (Table 8). Mounted slides were stored at 4°C and protected from light. Slides were viewed on a Zeiss Axioplan 2 ApoTome imaging microscope with a motorised stage for taking z stacks, using AxioVision software. Similarly, Adobe Photoshop was used to process images.
Table 7: Primary antibodies for immunofluorescence staining
Antibody
Species
Dilution
Source/ company
Antibody detail
DAGLα
Rabbit
1:20
Wolfson CARD
Lab antibody (651)
GFP
Chick
1:30
GFAP
Mouse
1:100
Sigma
Astroglial cell marker G3893
Table 8: Secondary antibodies for immunofluorescence staining
Hoechst dye 33342 (1:10000, Invitrogen) was added to the secondary antibody solution to fluorescently highlight nuclei blue.
Antibody
Species
Dilution
Source/ company
Antibody detail
Anti-mouse IgG Alexa Fluor 594
Goat
1:100
Molecular Probes
Red GFAP fluorescence
Anti-rabbit IgG Alexa Fluor 594
Goat
1:100
Molecular Probes
Red DAGLα fluorescence
Anti-mouse IgG Alexa Fluor 488
Goat
1:100
Molecular Probes
Green astrocyte fluorescence
Anti-chick IgG
1:100
Green GFP
