The cannabinoid receptor 1 (CB1), a member of the G protein-coupled receptor (GPCR) family, is an important member of the endocannabinoid system and is known to be expressed in the Central Nervous System (CNS)¹ and multiple other tissues and cell types². In the CNS, the CB1 plays a key role in multiple processes such as pain perception, brain development and body temperature regulation^1,2. It has been widely studied and is considered a potential pharmacological target for multiple diseases, including neuropsychological and neurodegenerative disorders².

This receptor is encoded by the CNR1 gene, which in humans is located at the chromosome 6 (6q14-q15), this gene is comprised by 4 exons; although the entire coding region is contained within exon 4³. CB1 expression levels are known to variate during embryo neurodevelopment⁴ and other development stages³. Altered expression of this receptor is directly linked to some pathological conditions such as Huntington’s⁵, Alzheimer’s⁶, Parkinson’s diseases⁷, obesity and diabetes⁸. Multiple mechanisms have been reported to be responsible for the expression of CB1, causing a broad number of dysregulation effects depending on the affected tissue⁹. The regulation of the expression of this receptor represents one of the major therapeutic target for the development of new drugs that could restore its normal physiological levels^3,10. Modulation of CB1 expression has also been observed upon prolonged endocannabinoids exposure, such as the case of Neuropathic pain, where the CB1 expression is downregulated¹¹, as well as under pharmacological modulation¹².

Variants of CNB1 have been found to result in changes in the amount of the expressed CB1, such as the rs1406977 SNP, carriers of this allele (G) have reduced CNR1 prefrontal mRNA expression amount compared with A/A subjects¹³. Other polymorphisms might not alter receptor expression but its affinity and response to agonists, that is the case of the rs2023239 SNP, a potential biomarker for susceptibility to cannabis use, which causes an alternative splicing of the CNR1 increasing CB1 affinity for its agonists¹⁴. To personalize treatments, there is need to analyze changes of expression caused by different polymorphisms or mutations. An example of the change in CB1 localization is point mutation F237L that reduces the membrane expression of this receptor¹⁵. Mutation T210I is characterized not only by reduced membrane expression but also by several times greater affinity for ligands, making it essentially constitutively active¹⁶, while mutation T210A results in a completely inactive CB1 despite membrane localization, although it seems to lack internalization capabilities¹⁷.

Many techniques have been developed and employed for the quantification of GPCRs cell surface expression; the most common include flow cytometry using fluorescent probes^18,19, ELISA²⁰, surface biotinylation²¹, among others, which require long incubations and multiple washing steps that increase quantitative determination variability per assay, or involve complex imaging equipment. In order to solve these problems, here we report an inexpensive and easy bioassay for the relative quantification of the CB1 receptor expression at the cell surface; it is based on the genetic fusion of the CB1 and Gaussia luciferase (GLuc), a widely used enzymatic reporter in eukaryotes. As a proof of principle, we analyzed three previously reported point mutations (CB1^F237L, CB1^T210I and CB1^T210A) and compared them to the wild-type (WT) counterpart²².

We took advantage of the natural properties of GLuc which does not require cofactors and it is normally secreted—it contains a secretion signal peptide²⁶. This allows us to quantify the CB1 relative expression on the surface of HEK293T cells. The relative quantification is easily done by measuring the luminescence generated by the GLuc enzyme linked to the CB1 protein, since this luminescence depends on the amount of receptor, changes on its expression will be quantifiably noted. A schematic representation of this method is shown in Fig. 1C.

As described in the Construct Description section, the signal peptide (SP in Fig. 1) used to localize the receptor at the membrane is encoded in the GLuc gene, in agreement with previous research works that have employed artificial signal sequences at the N-terminus to allow membrane localization²⁷ since the majority of the GPCR receptors including CB1 do not possess a signal peptide to be trafficked to the cell surface²⁸. Foreign sequences at the N-terminal of GPCRs, are known to affect the addressing of the receptor to the cell surface due to a low efficiency of translocation through the endoplasmic reticulum (ER). For this reason, the addition of a signal sequence (signal peptide, SP) at the N-terminus of a fusion construct is necessary^27,29. In this case, GLuc, which already contains a SP, was inserted in front of CB1 separated by a short linker to avoid interfering with the CB1 folding and topology; in some other cases the fusion of such chimeric proteins has been done between residues 25 and 26 of the extracellular N terminus following the same principle²⁷.

At first, the initial architecture containing the GLuc linked to the CB1 by a flexible domain consisting of 4 amino acids (GLEG), resulted in that most of GLuc was found in the medium and not attached to the cells (Fig. 2A), what seems to be caused by the cleavage of the GLuc-CB1 by a protease. To address this, we incubated cells expressing this GLuc-CB1 with several protease inhibitors and found the Batimastat, a pan-metalloprotease inhibitor, diminished the amount of free GLuc (Fig. 2B). We then analyzed a short sequence that comprises the GLEG flexible domain and the proteins GLuc and CB1 (specifically DKIKGAGGDGLEGKSILD) for metalloprotease sites using a prediction software “SitePrediction” (https://www.dmbr.ugent.be/prx/bioit2-public/SitePrediction/) and found a potential cleavage site that could disrupt the generated chimeric protein causing GLuc enzyme to be released (Supplementary Fig. S1). This sequence was partially removed by inserting a FLAG Tag between the GLuc and the CB1, which plays the role of a linker and a spacer for proper folding Fig. 1. Subsequent experiments showed that the GLuc activity, after cells were transfected with the GLuc-FLAG-CNR1 plasmid, was virtually only from cells (Cells) and not in the culture medium (Medium) (Fig. 2C). Mock transfected cells and their medium showed baseline luminescence values, as shown in Fig. 2C. Based on the results, the generated luminescence specifically comes from the transfected cells, clearly demonstrating that the enzyme is not being secreted and that it is only readable in cells expressing the GLuc-FLAG-CB1. Full sequences of both architectures (GLuc-FLAG-CNR1 and GLuc-GLEG-CNR1) are provided in the Supplementary material, both as nucleic acids and amino acid sequences, as well, the plasmid maps are shown in Supplementary Figs. S2 and S3.Figure 2

The construct GLuc-GLEG-CB1 contains a cleavage site that causes GLuc to be released into the medium. (A) Cells expressing the GLuc-GLEG-CB1 were tested 48 h after transfection by adding coelenterazine, it can be observed how the luminescence is originating from the medium and not from the cells, suggesting that the generated GLuc is being cleaved and released into the medium. (B) The RLU level in culture medium after protease inhibitors treatment of transfected cells. The incubation of cells expressing GLuc-GLEG-CB1 with different protease inhibitors at fixed concentrations (Batimastat 1 µM, EDTA 2 mM, PMSF 0.2 mM) and time points (0 h, 2 h, 4 h and 6 h) suggests that the linker sequence contains a metalloprotease cleavage site (metalloprotease predicted site is presented in Supplementary Fig. S1), as it can be observed from the readings coming from the medium of transfected cells. (C) In contrast, 48 h after transfection luminescence from cells expressing the construct GLuc-FLAG-CB1 comes entirely from the cells, while virtually no signal was obtained from their medium. The RLU level for mock transfected cells (mock cells) and their medium (mock medium) is shown in the Fig. 2C. The results represent the mean ± SD analyzed with one-way ANOVA test and Tukey’s multiple comparison post-hoc test vs. the mock cells and medium control (A,C).

The above results show that adding a FLAG tag between the GLuc and the CB1 let the chimeric protein be expressed at the cell surface and being properly folded. This also suggests that any intracellular GLuc signal is minimal e.g. during biosynthesis and localization, as proteolytically processed GLuc would show activity within the cells while transporting, which is virtually unperceived.

The one-way ANOVA showed that there is a statistically significant difference among the means of the four factors (cells and medium from transfected GLuc-FLAG-CB1 and mock transfected cells) (p ≤ 0.001); the Tukey’s test revealed that the transfected cells (Cells) mean is significantly different from those of the other three factors, confirming the source of luminescence.

Having established that the receptor is at the cell membrane and can be quantified through this bioassay, we next generated a series of CNR1 mutants (CB1^F237L, CB1^T210I, CB1^T210A), in order to demonstrate changes in the cell surface expression caused by these point mutations are in line with previous observations. Indeed, the mutants behaved as expected, where the CB1^T210A exhibited similar expression to CB1^WT, while CB1^F237L and CB1^T210I showed considerable reduced expression on membrane (Fig. 3).Figure 3

WT and mutant CB1 relative expression on the surface of HEK293T cells. Cells expressing GLuc-FLAG-CB1^WT (WT), GLuc-FLAG-CB1^F237L (F237L), GLuc-FLAG-CB1^T210I (T210A) and GLuc-FLAG-CB1^T210A (T210I) are shown, in order to generate a read-out, coelenterazine was added to the cells after medium was washed away. Similar expression levels can be observed for CB1^T210A and the CB1^WT, showing that this mutation does not affect the expression of the receptor, however, mutations CB1^F237L and CB1^T210I reduced the expression at the surface of HEK293T cells, as compared to the CB1^WT.

The GLuc-CB1 reporter showed that there is a statistically significant difference among the means of the cells transfected by GLuc-FLAG-CB1^WT and GLuc-FLAG-CB1^F237L, or GLuc-FLAG-CB1^T210I (Fig. 3), however, no difference was observed between those transfected by GLuc-FLAG-CB1^WT and GLuc-FLAG-CB1^T210A, indicating similar cell surface receptor expression. Figures 2C and 3 also show the importance of running all controls [positive (WT) and negative] within the same run, as changes in cell number, cells state, reagents batch and other environmental factors might significantly affect the readout, as can be observed by a reduction in the RLU units obtained in the two different experiments (4 × 10⁶ in Fig. 2C vs 3 × 10⁴ in Fig. 3). These differences should not be considered as a 100-fold reduction in the expression but are in fact inter assay variations. Comparisons can only be done between samples of the same run, as we have done for WT vs mutants in Fig. 3.

Through this bioassay we corroborated previous data in that the CB1^F237L and CB1^T210I mutants cause a reduced CB1 receptor expression level on the cell surface^15,16. Our data also supports that CB1^T210A mutation results in unaffected expression¹⁷.

Next, we tested whether luciferase readouts agreed with membrane-anchored GLuc-FLAG-CB1 expression using an antibody against the FLAG tag by flow cytometry. The analyses are shown in Fig. 4. The CB1^WT and CB1^T210A mutant showed similar GLuc-CB1 expression both in permeabilized as well as in not permeabilized cells, while CB1^F237L and CB1^T210I have significantly lower (p ≤ 0.001) membrane expression (Fig. 4C) as expected from the GLuc readouts. GLuc-CB1 expression in permeabilized (total expression) and not permeabilized (membrane-anchored only) cells is shown in Fig. 4A,B, respectively. See Supplementary Table S1 for a statistical analysis of the collected data.Figure 4

The total and surface-only level of GLuc-CB1 in HEK293T cells. The representative SSC-H/FL2-H 2D-dot plots for (A) permeabilized and (B) non-permeabilized wild type (WT), GLuc-FLAG-CB1^F237L (F237L), GLuc-FLAG-CB1^T210I (T210I) and GLuc-FLAG-CB1^T210A (T210A) cells are shown. The cells treated with secondary antibody only (NC) and not transfected by plasmids (UT) were used as negative controls for gating. (C) The median expression [the mean ± SD (n = 3)] of HEK293T cells classified as GLuc-CB1-positive (gate R4) in permeabilized (P) and non-permeabilized (NP) HEK293T cells normalized vs 1 are shown.

According to the flow-cytometry results, the total expression for all the mutants was apparently the same as the wild type, however the results demonstrate how the point mutations CB1^F237L and CB1^T210I directly diminish the membrane-anchored receptor quantity. The membrane expression of the wild-type and CB1^T210A clearly indicate that the addition of the signal peptide aided the membrane localization of the receptor, and that the lower levels of expression seen for the mutants CB1^F237L and CB1^T210I are caused by different pathways, as alterations in biosynthesis, internalization, a phenomenon previously observed for the mutated receptor F237L¹⁵, or by a constitutive endocytosis. Either way, the bioassay is able to detect such changes which is essential for screening of hitherto mutations.

The results obtained by cell cytometry and GLuc-generated relative luminescence units were both in accordance with each other and those in literature, clearly indicating that the readouts generated by both experiments quantify correctly the receptor expressed at the cell membrane for all the analyzed cases. This is also supported by the GLuc-GLEG-CB1 test where virtually no intracellular readout was detected (Fig. 2A) and instead almost all luminescence is found in the culture media.

There are some limitations that should be considered; in this study we have used the CB1 to demonstrate that the luminescence comes only from those transfected cells; however, it has been reported that the expression of this receptor occurs at other intracellular structures such as the mitochondria³⁰. In addition, coelenterazine has a lipophilic nature and its able cross the cell membrane; a non-permeable substrate is available³¹, although this is easily detected by flow cytometry or confocal microscopy if the mutant is suspected to be mostly intracellularly located. Another limitation is that our assay cannot assess CB1 activity, for which the system could be coupled to some of the others signaling assays such as TANGO³², GloSensor³³, or other assays that measure cAMP, inositol phosphate or calcium accumulation methods³⁴ and two measurements could be accounted for, expression vs activity.

CB1 plays key homeostatic role in a variety of physiological processes, new tools have been constantly being designed to provide a better understanding of its mechanisms and variability. With our quantification assay, determining how mutations affect the receptor expression in a much easier way than commonly used methods will ease research regarding the endocannabinoid system and its alterations upon previously reported pathological conditions. Moreover, GLuc has been used for BRET and therefore it can be used to detect ligand–receptor or receptor-receptor interactions on the plasma membrane of living cells.

The main novelty of our work is determined by the fact that it could be possible to detect the effect of various alterations in the chemical environment (such as the constant binding of ligands) or genetical factors (such as the presence of point mutations or polymorphisms) on the GPCR receptors’ expression¹⁴.

The detection of membrane localization of particular GPCRs is fundamental to understand the roles of mutations and protein interactions. In fact, the altered expression of the CB1 receptor is in clear association with various mental-related issues such as major depression, schizophrenia, bipolar disorder, among others^35–37. Our research could provide the foundation to better understanding how dysregulation of the membrane expression of CB1 plays a role in the molecular basis for the afore mentioned diseases.

Multiple mutations, polymorphisms, and alternative splice variants of the CB1 receptor have been identified and recognized as important pharmacological targets in multiple diseases, these genetic variants of the CB1 have been deeply described elsewhere³⁸. Thus, screening how these genetic variants affect the level of expression of this and other GPCRs can speed up the diagnosis and treatment, with e.g. molecular chaperones³⁹, of multiple genetic-based diseases.

As previously described, other quantifying methods including flow-cytometry or ELISA can be employed to determine surface expression of the CB1 and other GPCRs, however, through a simplified bioassay as the one described here, we offer a new simple methodology with results that can be compared with those obtained by flow-cytometry, this, we believe, will potentially aid research on the GPCR field and many other membrane expressed receptors.

Altogether, these results demonstrate the reliability of this novel bioassay in quantifying the CB1 receptor expression on the cell surface.