The Cannabinoid Receptor Interacting Proteins 1 of zebrafish are not required for morphological development, viability or fertility
Randall Division of Cell and Molecular Biophysics, New Hunt’s House, Guy’s Campus, King’s College London, London, SE1 1UL UK
Abstract
The Cannabinoid Receptor Interacting Protein 1 (Cnrip1) was discovered as an interactor with the intracellular region of Cannabinoid Receptor 1 (CB1R, also known as Cnr1 or CB1). Functional assays in mouse show cannabinoid sensitivity changes and Cnrip1 has recently been suggested to control eye development in Xenopus laevis. Two Cnrip1 genes are described in zebrafish, cnrip1a and cnrip1b. In situ mRNA hybridisation revealed accumulation of mRNA encoding each gene primarily in brain and spinal cord, but also elsewhere. For example, cnrip1b is expressed in forming skeletal muscle. CRISPR/Cas9 genome editing generated predicted null mutations in cnrip1a and cnrip1b. Each mutation triggered nonsense-mediated decay of the respective mRNA transcript. No morphological or behavioural phenotype was observed in either mutant. Moreover, fish lacking both Cnrip1a and Cnrip1b both maternally and zygotically are viable and fertile and no phenotype has so far been detected despite strong evolutionary conservation over at least 400 Myr.
Article notes
Untitled section
Received 2017 Feb 27; Accepted 2017 May 23; Collection date 2017.
Introduction
The discovery of (−)-trans-Δ9-tetrahydrocannabinol (THC) as the major psychoactive component of marijuana (Cannabis sativa)1 and subsequent cloning of the THC receptor2 helped to identify the endogenous cannabinoid system (eCS). The eCS is a signalling network responding to endogenous lipid transmitters (endocannabinoids) that bind to the G-protein coupled cannabinoid receptors CB1R and CB2R3. These receptors act via Gαi/o binding to their second and third intracellular loops and C-terminal 8th helix to initiate signalling events typical of this class of transducing proteins, such as inhibition of adenylyl cyclase and activation of MAPK and FAK4, 5. CB1R has also been suggested to affect PI3K and PKB signalling and a variety of ion channels3, 6. Endocannabinoids are frequently produced post-synaptically in response to neurotransmitter signalling and act retrogradely on pre-synaptic terminals to attenuate pre-synaptic depolarisation and diminish further neurotransmitter release, forming a negative feedback loop3, 7. The eCS is involved in numerous pathological conditions, ranging from neurological disorders, such as Parkinson’s, Huntington’s and Alzheimer’s diseases8, to gastrointestinal, cardiovascular and reproductive disorders9 and is also potentially involved in relief of pain, chemotherapy-induced nausea and vomiting, and anorexia10. It thus has potential for therapeutic drug development11.
To understand the biochemical mechanism of endocannabinoid action and find possible drug targets, screens have been performed for proteins interacting with CB1R12–14. Two alternative splicing variant proteins, named CRIP1a and CRIP1b, deriving from the human Cannabinoid Receptor Interacting Protein 1 gene (CNRIP1)12 have been reported to bind to the conserved C-terminal intracellular portion of CB1R, a region thought to be critical for receptor activation, sensitivity and recycling15, 16. Human CRIP1a and CRIP1b share identical N-termini, but diverge after residue 110 (Fig. 1a). The longer CRIP1a isoform (164 amino acids in human) is conserved across vertebrates and invertebrates, whereas CRIP1b (128 amino acids in human) is restricted to certain primates17. Functionally, CRIP1a, but not CRIP1b, has been reported to modulate CB1R-mediated tonic inhibition of voltage-gated Ca2 + channels12 and recent evidence has indicated that CRIP1a can compete with β-arrestins for the C-terminal region of CB1R, suggesting a role of internalization of the receptor18.
Neuronal activity, endocannabinoid and D2 dopamine receptor (D2R) signalling can all regulate CNRIP1 expression19–21. Over-expression of CNRIP1 in rat dorsal striatum led to reduced ERK phosphorylation, decrease in transcript levels of certain opioid peptide precursors and the up-regulation of the opioid receptor DOR1 at both mRNA and protein level, mimicking effects observed when either CB1R or D2R were knocked-down21. Further over-expression studies have implicated CNRIP1 in CB1R signalling and turnover18, 22–24. Knockdown of CNRIP1 protein in a neuronal cell line can increase plasma membrane CB1R, alter G-protein subtype preference, augment basal phosphor-ERK level and enhance ERK phosphorylation in response to the CB1R agonist CP55940, but not to certain other CB1R agonists25. These data suggest the hypothesis that a role of CNRIP1 is modulation of endocannabinoid signalling through CB1R. An important observation, however, is that whereas CB1R is primarily expressed in neurons of the central and peripheral nervous systems26–30, CNRIP1 is expressed more broadly (http://www.ebi.ac.uk/gxa). Western blot analyses confirm that mouse CNRIP1 proteins are present in brain but also widely in heart, lung, kidney, intestine, testis, spleen, liver and muscle12. Association of CNRIP1 promoter hypermethylation with colorectal and liver cancer and lymphoma has suggested the gene is a promising biomarker and may function in non-neural tissues31–34. Antisense morpholino knockdown of Cnrip1 in Xenopus has been reported to cause brain and also eye defects early in development35, even though the eye lacks abundant Cnrip1 and CB1R in adult mouse36. Moreover, a CNRIP1 gene is found in species that lack CB1R, such as the diploblast sea anemone Nematostella vectensis 17. Thus, CNRIP1 may have other functions independent of CB1R.
Here, we report the analysis of CNRIP1 expression and function in the zebrafish. Zebrafish have duplicate cnrip1a and cnrip1b genes expressed widely during early development. CRISPR/Cas9-mediated mutation of each gene leads to viable and fertile fish with no obvious phenotype. Double mutant animals are also viable and fertile, showing that CNRIP1 proteins are not essential for life.
Results
Zebrafish Cnrip1 genes
Two genes encoding Cnrip1 homologues were discovered in the zebrafish genome (Zv10). Cnrip1a, located on Chromosome 1, has three coding exons predicting a 162 amino acid protein with, respectively, 60% and 59% identity with the 164 amino acids of human CRIP1a and mouse CNRIP1. The region of greatest divergence was in the central region of the predicted proteins. The second gene, cnrip1b, is located on Chromosome 13, also has three exons and 162 amino acids with 69% identity with Cnrip1a and 59% identity to mammalian CNRIP1 (Fig. 1a). Identity between the two zebrafish paralogue cDNAs was 72% in the coding region, yielding 93% similarity at the amino acid level, with differences primarily in their N terminal 28 amino acids. Analysis in GeneTree (ensemble.org) shows that each gene clusters next to a cavefish orthologue as the sister groups to euteleost cnrip1a and cnrip1b clusters, consistent with the known teleost-specific genome duplication37. These groups cluster together and branch basally from the single amniote/amphibian CNRIP1 gene cluster, following known evolutionary relationships.
Synteny analysis confirms the homology of both zebrafish genes with CNRIP1 of human (Fig. 1b). Human and mouse genes have PPP3R1, PNO1, WDR92, CD1 and ETAA1 genes downstream and PLEK, FBXO48 and APLF genes upstream. Zebrafish cnrip1a has ppp3r1b and etaa1 downstream and two fbxo48 homologues and aplf upstream. Cnrip1b has ppp3r1a, pno1, wdr92 downstream and plek upstream. Thus, almost all flanking genes are conserved across ≥ 400 Myr of evolution, suggesting conserved regulation.
Although human CNRIP1 RNA can be alternatively spliced, yielding proteins named CRIP1a and CRIP1b12 (Fig. 1a), no evidence of alternative splicing of either zebrafish gene was apparent in ENSEMBL. BLAST searches with the human alternate third exon of CRIP1b failed to identify sequences on zebrafish Chromosome 1 or 13. Indeed, neither zebrafish gene had significant homology with the alternate third exon of CRIP1b. This suggests that sub-functionalisation of an ancestral alternatively-spliced gene into two genes each of which encodes one splice variant does not explain the evolutionary retention of two Cnrip1 genes in teleosts. Cnrip1a and cnrip1b have 24% and 28% identity to the last three exons of a C. elegans orthologue, F29A7.4, which has two additional 5′ exons encoding an N-terminal extension. Taken together, these data indicate strong conservation of structure, regulation and probably function of CNRIP1 across the vertebrates and possibly beyond.
Expression of cnrip1a and cnrip1b transcripts
To analyse the expression pattern of cnrip1a and cnrip1b genes in zebrafish, embryos at several developmental stages were subjected to whole mount in situ hybridisation (Fig. 2; quantitative evidence of reproducibility for all experiments is presented in Table S1). Cnrip1a mRNA was detected throughout the embryo at around 10 hpf (Fig. 2a; Table S1). Hybridisation with a cnrip1a sense probe, in contrast, showed no significant staining. Consistent with a previous report38, at 24 and 48 hpf, cnrip1a was most highly expressed in the brain and spinal cord (Fig. 2a; Table S1). At 24 hpf, cnrip1a shows strongest expression in telencephalon, anterior and post-optic commissure, pineal gland, retinal ganglion cells, other head ganglia and neuron clusters in the hindbrain and in spinal cord (Fig. 2a; Table S1). At 48 hpf, the brain is widely stained. Olfactory receptors appear to be stained flanking the telencephalon. In hindbrain, cnrip1a mRNA is most abundant in radial glia on the sides of the rhombomeres. Signal in radial neuronal cells in the retina around the lens was suggestive of retinal ganglion cells. The spinal cord appeared less strongly stained at 48 hpf compared to 24 hpf (Fig. 2a; Table S1), but when 48 hpf fish tails were stained overnight separately from heads, a pattern similar to the one observed at 24 hpf appeared (data not shown). In general, cnrip1a mRNA was detected in many regions where post-mitotic neurons were abundant.
Differential gene expression could provide initial information on the possible additional functions of Cnrip1b in other body regions. Cnrip1b mRNA appears at around 10 hpf and, at this developmental stage, the pattern is similar to the one observed for cnrip1a mRNA (Fig. 2b; Table S1). At 24 and 48 hpf, the expression of cnrip1b diverges from that of cnrip1a. Cnrip1b is expressed in the brain, the eyes and the spinal cord, but the staining remains diffuse, whereas cnrip1a mRNA showed a clear enrichment in regions of post-mitotic neurons concentration (Fig. 2a; Table S1). In addition, unlike cnrip1a, cnrip1b mRNA was detected in developing somites and pectoral fins (Fig. 2b; Table S1). Thus, cnrip1b is expressed somewhat differently from cnrip1a, suggesting it could have a distinct function.
Mutagenesis of Cnrip1 genes
Zebrafish mutants in cnrip1a and cnrip1b were generated using CRISPR/Cas9 genome editing (Fig. 3). Embryos from genotyped AB strain parents chosen to lack polymorphism in the target region were injected at 1-cell stage with gRNA and Cas9 RNA and these F0 fish outcrossed to similarly non-polymorphic wild-type AB to identify F1 fish heterozygous for mutations in each gene. Among 12 mutations obtained in one gRNA to cnrip1a, an allele cnrip1a kg98 with a 4 bp deletion leading to a frameshift at amino acid 33 followed by in-frame stop codon after six further amino acids was selected (Fig. 3a). The novel stop codon in this allele is 20 amino acids ( = 60 bp) 5′ to the next in-frame methionine, making ribosomal re-initiation highly unlikely because the ribosome footprint on mRNA is only ~28 bp. A second allele, cnrip1a kg96, was selected from three alleles produced with a distinct gRNA to cnrip1a; in this case a 53 bp insertion caused a frameshift at amino acid 47 followed by in-frame stop codon after six further amino acids (Fig. 3a). Beyond this stop codon at 7 and 19 bp downstream are two additional out of frame AUG codons which are themselves shortly followed by two additional stop codons, making in-frame re-initiation of the ribosome highly unlikely. For cnrip1b. among nine distinct mutations obtained with the single gRNA, a 10 bp deletion allele cnrip1b kg101 was selected, which led to a frameshift at amino acid 31 followed by in-frame stop codon after six further amino acids (Fig. 3a). The nearest in-frame AUG codon is 40 bp downstream of the deletion, giving little possibility of ribosomal re-initiation.
For both genes, only mutations that inserted a stop codon as close as possible to the CRISPR target site were selected. In this way, only short truncated polypeptides are produced, which are typically unfolded and rapidly degraded. Moreover, as the mutant mRNAs lack translation beyond the first coding exon, they were degraded by the nonsense-mediated decay pathway39, as revealed by the greatly reduced in situ mRNA hybridisation signal observed in mutants (Fig. 3b,c; Table S1). Compared to wild type siblings, significantly less mRNA was also discerned in heterozygotes (Fig. 3b,c; Table S1).
Embryos homozygous mutant in each of the three alleles developed normally and no morphological abnormalities were detected (Fig. 3b,c; Table S1 and data not shown). In particular, eye morphology, pigmentation, lens and retina formation appeared normal (Fig. 3b,c; Table S1). From heterozygote in-crosses, an average of 20 embryos were genotyped at 48 hpf from three independent lays and χ2 tests showed no significant difference from expected Mendelian values (Table S1). Sibling larvae were again monitored at 5 dpf and weekly during growth in the nursery, but no consistent morphological or behavioural defects were observed in the predicted 25% of mutants (Table S1). Larvae were grown to adulthood and mutants (detected from finclip DNA) were observed at the expected 25% frequency and were indistinguishable from wild type and heterozygous siblings (Table S1). Cnrip1a kg98 homozygous mutants survived as well as siblings until at least 19 months of age; cnrip1b kg101 homozygous mutants survived until at least 16 months (Table S1). Both cnrip1a kg98 and for cnrip1b kg101 homozygous mutants were normally fertile and maternal/zygotic mutants (produced by breeding from a homozygous mutant mother) also showed no phenotype (Table S1), eliminating the possibility that Cnrip1 protein placed in the egg by a heterozygous mother supressed a more severe phenotype.
Functional analysis of double mutants
Cnrip1a kg98 and cnrip1b kg101 carriers were crossed to produce dual cnrip1a kg98/+;cnrip1b kg101/+ heterozygotes, which were in-crossed. Double mutants were observed at the expected frequency and again showed no morphological or behavioural defects (Fig. 4a; Table S1). Double mutants survived as well as their siblings until at least 15 months of age and, when in-crossed produced normal sized lays of healthy and viable larvae that grew to adulthood (Fig. 4b–d; Table S1). No defects in eye, brain, spinal cord, body shape and size, muscle, notochord, fins, pigmentation, swim bladder, pectoral fins, jaw, gut peristalsis, eating behaviour, eye movement, swimming, shoaling and mating behaviour or fertility were observed. Thus, insertion of early stop codons into both Cnrip1 genes of zebrafish yielded healthy and fertile adult fish.
Discussion
The present work demonstrates that homozygous predicted loss of function mutations of all CNRIP1 genes in a vertebrate leads to viable and fertile organisms with no discernible characteristics different from wild type. Cnrip1 expression is observed in the outer plexiform layer (OPL) of the mouse retina36 and morpholino loss of function studies have suggested that Xenopus laevis cnrip1 regulates eye development35. However, we detected no eye defects in our maternal/zygotic cnrip1a;cnrip1b double mutants. Similarly, although Cnrip1 is expressed in mouse heart12, careful analysis of the developing heart revealed no defects and adult mutants did not show increased sudden cardiac death. Lastly, CNRIP1 is highly expressed in the nervous system in species from mammals to fish, yet no morphological neural deficits or behavioural changes were observed in our mutants. These findings indicate that the function of the well-conserved CNRIP1 protein(s) confers a subtle increase in fitness that is not readily detected in the unchallenged aquarium environment.
There remains a formal possibility that our mutations, despite creating early stop codons in the first exon, are not null, either because the short truncated polypeptides can play some role or because of some kind of ribosomal re-initiation or stop codon read-through permits synthesis of a long protein. In the absence of functional antibodies recognizing the zebrafish proteins, we cannot be certain that no functional folded Cnrip1 protein is synthesised. However, we think this unlikely for several reasons. Firstly, nonsense mediated mRNA decay for each gene suggests that exon splice junction complexes are not removed downstream of the in-frame stop codons, indicating that ribosomes do not progress normally to the end of the three exon transcripts from each gene. Consistent with this view, there are no in-frame methionine codons nearby downstream of the novel stop codons that might encourage re-initiation by a stalled ribosome before it dissociates from the mRNA. Secondly, the reduced mRNA levels will mean that fewer ribosomal initiation events occur on each transcript and thus the absolute number of truncated polypeptides synthesised per unit time is likely lower than that of the full length proteins in wild types. Thirdly, the truncated polypeptides are under 40 amino acids in length (compared with 162 amino acids for each full length protein) and thus are unlikely to create a stable folded domain and will therefore be more rapidly degraded than the full length proteins. For these reasons, we suspect our mutants are functionally null.
Genomic analysis reveals that the genomes of zebrafish and most other fish contain duplicated copies of a clear homologue of the human CNRIP1 gene. This strongly suggests that (a) CNRIP1 is a conserved gene that has had an important function in vertebrates for at least 400 Myr, (b) that after the extra teleost genome duplication the duplicates developed essential functions sufficiently rapidly for both genes to be retained and well conserved. An invertebrate homologue of CNRIP1, F29A7.4, is present in C. elegans, has an ~80 amino acid N-terminal extension compared to vertebrates but is expressed in the ventral nerve cord and various neurons40. Akin to our findings in zebrafish, F29A7.4 has not been observed to have a loss of function phenotype in two independent RNAi screens (http://www.wormbase.org/species/c_elegans/gene/WBGene00017914).
A consistent characteristic of CNRIP1 family genes, also observed in zebrafish, is strong expression in neural tissue (recently reviewed41). Both cnrip1a and cnrip1b are expressed in the zebrafish nervous system, with cnrip1a mRNA sometimes concentrated in particular regions of neuronal terminal differentiation. Cells in specific regions of the brain and in placode-derived cranial ganglia express highly. This conserved location of expression is consistent with the conserved synteny of the chromosomal region around CNRIP1 coding sequence across vertebrates. Expression in neurons is consistent with the suggested ability of CNRIP1 to bind to the intracellular region of CB1R, but does not exclude other functions in neurons. Cnrip1b is expressed more broadly than cnrip1a, appearing in muscle and fin tissue in the late embryo. As CB1R is expressed in many tissues, this finding could indicate functions beyond neuronal action of endocannabinoids. In future, it will be essential to assess the response of cnrip1a and cnrip1b mutant fish to gain and loss of function of endogenous and exogenous cannabinoids, as well as explore alternative functional avenues.
Methods
Zebrafish lines and maintenance
Genetically-altered Danio rerio on a primarily AB background were created and reared at King’s College London on a 14/10 hr light/dark cycle at 28.5 °C42. All experiments were performed in accordance with licences held under the UK Animals (Scientific Procedures) Act 1986 and later modifications and conforming to all relevant guidelines and regulations.
In situ mRNA hybridisation
In situ mRNA hybridisation (ISH) was performed as described previously43, 44. IMAGE clones for both cnrip1a (clone ID: 7149263) and cnrip1b (clone ID: 8760081) genes (Thermo Scientific) were grown as in Table S2 and confirmed by sequencing. A PCR reaction was then performed to add T3 and T7 RNA polymerase sites to create antisense and sense probes. Generic pDNR-LIB primers were used for cnrip1a, while specific primers for cnrip1b were designed. Primers (Integrated DNA Technologies; Table S3) were used to amplify template DNA (Table S4), purified using the QIAquick PCR Purification Kit (QIAGEN) and digoxigenin-labelled RNA probes synthesised, treated with DNase and purified with the Illustra Microspin G-50 kit (GE Healthcare).
CRISPR/Cas9 genome editing
Genome editing was adapted from ref. 45. CRISPR target sites in the first coding exon were selected with ZiFiT46 and potential off-targets minimized with the specific ZiFiT tool. Optimised flanking primers creating a ~120 bp PCR product for High Resolution Melt Analysis (HRM; 20 bp each, Tm 60 °C) and 200–400 bp product for DNA sequencing were selected for each gRNA with Primer 3 software47 and oligos bought from MWG Eurofins. CRISPR oligos were annealed, ligated into BsaI-digested pDR274 (Addgene), plasmid DNA purified, sequenced, digested with DraI and the 284 bp fragment gel-purified and used to synthesise gRNA with T7 RiboMAX large scale RNA production kit (Promega) which was phenol/ethanol purified, quantified by gel and Qubit (Invitrogen), aliquoted in 5 µl samples at 200 ng/µl and stored at −80 °C. NotI-HF-linearised pCS2-Cas9 was transcribed using mMessage mMachine SP6 kit (Ambion) and product purified as for gRNA.
HRMA-selected AB wild-type fish were DNA sequenced over the target loci to avoid polymorphisms, crossed and the resulting embryos injected with 1 nl containing 60 ng/µl gRNA, 200 ng/µl Cas9 mRNA and phenol red and GFP mRNA to select injected embryos. Twenty morphologically-normal 48 hpf larvae were analysed by HRMA to verify mutagenesis, their F0 siblings grown to adulthood and backcross F1 progeny analysed for transmission by HRMA. Mutant loci of F1s were sequenced to identify F0s transmitting mutations of interest, F1 siblings grown to adulthood and F1 heterozygotes identified by HRMA and sequencing of fin-clip DNA. Subsequent generations were bred by outcross to non-polymorphic wild-type AB. To generate double mutants, cnrip1a and cnrip1b F2 heterozygotes were crossed to obtain F3 progeny in which 25% of fish were dual heterozygotes (cnrip1a +/− ;cnrip1b +/−). These fish were in-crossed to generate a F4 progeny of which 6.25% were predicted to be homozygous for both genes (cnrip1a −/− ;cnrip1b −/−).
High resolution melt analysis (HRMA)
Total genomic DNA was extracted from single embryos at 24 or 48 hpf or fin-clip tissue using 100 µl of PeqLab lysis solution (31-401-E) with 20 µg proteinase K incubated at 55 °C overnight and inactivated at 85 °C for 45 minutes. For the HRMA, 1 µl of each DNA lysate was mixed with 2× MeltDoctor HRM mix (Life Technologies), forward and reverse HRM primers (0.14 µM) to 32 µl and half loaded in duplicate wells on a MicroAmp Optical 384-well plate (Life Technologies), sealed with optical adhesive film, centrifuged at maximum speed for 30 seconds on a bench plate centrifuge to eliminate bubbles and HRMA performed on a ViiA7 real-time PCR system (Applied Biosystems).
Screening for phenotype
Multiple lays of in-crosses between fish carrying identical or distinct alleles were screened for phenotypes. In each lay, about a hundred F2–F4 embryos at 4–10 hpf were monitored for alterations in gastrulation, epiboly and formation of the embryonic shield. From 10–36 hpf, defects in somites, myogenesis, brain, spinal cord, notochord, eye and heart were screened. Subsequently, alteration in brain, spinal cord, eye, heart, cardiovascular system and fin morphology, pigmentation hatching, touch response and swimming patterns were analysed. At 96 hpf, larvae were screened again for touch response and motility, defects in pigmentation, pectoral fins, jaw, branchial arches, blood, ear, liver and gut. At 5 dpf, larvae were transferred to the nursery, where they were monitored on a weekly basis for later developmental defects or behaviour abnormalities.
Statistics
χ2 tests were performed in Excel taking p < 0.05 as significant.
Electronic supplementary material
Untitled section
Acknowledgements
We thank Vikki Williams for genotyping, Pat Doherty, Anthony Graham, Yaniv Hinits and members of the Hughes lab for advice and Bruno Correia da Silva and his staff for care of the fish. LF had an EPSRC PhD studentship in the laboratory of RAS. GB was funded by the Biotechnology and Biological Sciences Research Council (BB/K0101151) to SMH. SMH is a Medical Research Council (MRC) Scientist with Programme Grant G1001029 and MR/N021231/1 support.
Untitled section
Competing Interests
The authors declare that they have no competing interests.
Footnotes
Footnote Group
Contributor Information
Roberto A. Steiner, Email: roberto.steiner@kcl.ac.uk
Simon M. Hughes, Email: simon.hughes@kcl.ac.uk
References
Untitled section
References
- 1.Gaoni Y, Mechoulam R. Isolation, structure, and partial synthesis of an active constituent of hashish. J Amer Chem Soc. 1964;86:1646–1647. doi: 10.1021/ja01062a046.
- 2.Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI. Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature. 1990;346:561–564. doi: 10.1038/346561a0.
- 3.Di Marzo V, Bifulco M, De Petrocellis L. The endocannabinoid system and its therapeutic exploitation. Nature reviews. Drug discovery. 2004;3:771–784. doi: 10.1038/nrd1495.
- 4.Piomelli D. The molecular logic of endocannabinoid signalling. Nature reviews. Neuroscience. 2003;4:873–884. doi: 10.1038/nrn1247.
- 5.Anavi-Goffer S, et al. Helix 8 Leu in the CB1 cannabinoid receptor contributes to selective signal transduction mechanisms. J Biol Chem. 2007;282:25100–25113. doi: 10.1074/jbc.M703388200.
- 6.De Petrocellis L, Cascio MG, Di Marzo V. The endocannabinoid system: a general view and latest additions. Br J Pharmacol. 2004;141:765–774. doi: 10.1038/sj.bjp.0705666.
- 7.Flores, Á., Maldonado, R. & Berrendero, F. Cannabinoid-hypocretin cross-talk in the central nervous system: what we know so far. Frontiers Neurosci7, doi:10.3389/fnins.2013.00256 (2013).
- 8.Fernández-Ruiz, J., Romero, J. & Ramos, J. In Endocannabinoids Vol. 231 Handbook of Experimental Pharmacology (ed Roger G. Pertwee) Ch. 8, 233–259 (Springer International Publishing, 2015).
- 9.Maccarrone M, et al. Endocannabinoid signaling at the periphery: 50 years after THC. Trends in Pharmacol Sci. 2015;36:277–296. doi: 10.1016/j.tips.2015.02.008.
- 10.Maida V, Daeninck PJ. A user’s guide to cannabinoid therapies in oncology. Curr Oncol. 2016;23:398–406. doi: 10.3747/co.23.3487.
- 11.Sharma C, et al. Small molecules from nature targeting G-protein coupled cannabinoid receptors: potential leads for drug discovery and development. Evidence-based Complementary and Alternative Medicine: eCAM. 2015;2015:238482. doi: 10.1155/2015/238482.
- 12.Niehaus JL, et al. CB1 cannabinoid receptor activity is modulated by the cannabinoid receptor interacting protein CRIP 1a. Molecular pharmacology. 2007;72:1557–1566. doi: 10.1124/mol.107.039263.
- 13.Sanchez C, et al. The CB(1) cannabinoid receptor of astrocytes is coupled to sphingomyelin hydrolysis through the adaptor protein fan. Molecular pharmacology. 2001;59:955–959. doi: 10.1124/mol.59.5.955.
- 14.Martini L, et al. Ligand-induced down-regulation of the cannabinoid 1 receptor is mediated by the G-protein-coupled receptor-associated sorting protein GASP1. Faseb j. 2007;21:802–811. doi: 10.1096/fj.06-7132com.
- 15.Nie J, Lewis DL. Structural domains of the CB1 cannabinoid receptor that contribute to constitutive activity and G-protein sequestration. J Neuroscience. 2001;21:8758–8764. doi: 10.1523/JNEUROSCI.21-22-08758.2001.
- 16.Tiburu EK, et al. Human cannabinoid 1 GPCR C-terminal domain interacts with bilayer phospholipids to modulate the structure of its membrane environment. The AAPS journal. 2011;13:92–98. doi: 10.1208/s12248-010-9244-7.
- 17.Elphick MR. The evolution and comparative neurobiology of endocannabinoid signalling. Philosophical transactions Royal Society London. Series B. 2012;367:3201–3215. doi: 10.1098/rstb.2011.0394.
- 18.Blume LC, et al. Cannabinoid receptor interacting protein 1a competition with beta-Arrestin for CB1 receptor binding sites. Molecular pharmacology. 2017;91:75–86. doi: 10.1124/mol.116.104638.
- 19.Ludanyi A, et al. Downregulation of the CB1 cannabinoid receptor and related molecular elements of the endocannabinoid system in epileptic human hippocampus. J Neuroscience: 2008;28:2976–2990. doi: 10.1523/JNEUROSCI.4465-07.2008.
- 20.Bojnik E, et al. Changes in the cannabinoid (CB1) receptor expression level and G-protein activation in kainic acid induced seizures. Epilepsy research. 2012;99:64–68. doi: 10.1016/j.eplepsyres.2011.10.020.
- 21.Blume LC, et al. Striatal CB1 and D2 receptors regulate expression of each other, CRIP1A and delta opioid systems. J neurochemistry. 2013;124:808–820. doi: 10.1111/jnc.12139.
- 22.Guggenhuber S, et al. Cannabinoid receptor-interacting protein Crip1a modulates CB1 receptor signaling in mouse hippocampus. Brain structure & function. 2015 doi: 10.1007/s00429-015-1027-6.
- 23.Smith TH, et al. Cannabinoid receptor-interacting protein 1a modulates CB1 receptor signaling and regulation. Molecular pharmacology. 2015;87:747–765. doi: 10.1124/mol.114.096495.
- 24.Blume LC, et al. Cannabinoid receptor interacting protein suppresses agonist-driven CB1 receptor internalization and regulates receptor replenishment in an agonist-biased manner. J neurochemistry. 2016;139:396–407. doi: 10.1111/jnc.13767.
- 25.Blume LC, Eldeeb K, Bass CE, Selley DE, Howlett AC. Cannabinoid receptor interacting protein (CRIP1a) attenuates CB1R signaling in neuronal cells. Cellular signalling. 2015;27:716–726. doi: 10.1016/j.cellsig.2014.11.006.
- 26.Mailleux P, Parmentier M, Vanderhaeghen JJ. Distribution of cannabinoid receptor messenger RNA in the human brain: an in situ hybridization histochemistry with oligonucleotides. Neuroscience letters. 1992;143:200–204. doi: 10.1016/0304-3940(92)90265-9.
- 27.Matsuda LA, Bonner TI, Lolait SJ. Localization of cannabinoid receptor mRNA in rat brain. J Comp Neurol. 1993;327:535–550. doi: 10.1002/cne.903270406.
- 28.Tsou K, Brown S, Sanudo-Pena MC, Mackie K, Walker JM. Immunohistochemical distribution of cannabinoid CB1 receptors in the rat central nervous system. Neuroscience. 1998;83:393–411. doi: 10.1016/S0306-4522(97)00436-3.
- 29.Egertova M, Elphick MR. Localisation of cannabinoid receptors in the rat brain using antibodies to the intracellular C-terminal tail of CB. J Comp Neurol. 2000;422:159–171. doi: 10.1002/(SICI)1096-9861(20000626)422:2<159::AID-CNE1>3.0.CO;2-1.
- 30.Pertwee RG, Stevenson LA, Elrick DB, Mechoulam R, Corbett AD. Inhibitory effects of certain enantiomeric cannabinoids in the mouse vas deferens and the myenteric plexus preparation of guinea-pig small intestine. Br J Pharmacol. 1992;105:980–984. doi: 10.1111/j.1476-5381.1992.tb09088.x.
- 31.Andresen K, et al. Four DNA methylation biomarkers in biliary brush samples accurately identify the presence of cholangiocarcinoma. Hepatology. 2015;61:1651–1659. doi: 10.1002/hep.27707.
- 32.Bethge N, et al. Colorectal cancer DNA methylation marker panel validated with high performance in Non-Hodgkin lymphoma. Epigenetics. 2014;9:428–436. doi: 10.4161/epi.27554.
- 33.Oster B, et al. Identification and validation of highly frequent CpG island hypermethylation in colorectal adenomas and carcinomas. Int J Cancer. 2011;129:2855–2866. doi: 10.1002/ijc.25951.
- 34.Lind GE, et al. Identification of an epigenetic biomarker panel with high sensitivity and specificity for colorectal cancer and adenomas. Mol Cancer. 2011;10:85. doi: 10.1186/1476-4598-10-85.
- 35.Zheng X, Suzuki T, Takahashi C, Nishida E, Kusakabe M. cnrip1 is a regulator of eye and neural development in Xenopus laevis. Genes to cells. 2015;20:324–339. doi: 10.1111/gtc.12225.
- 36.Hu SS, et al. Architecture of cannabinoid signaling in mouse retina. J Comp Neurol. 2010;518:3848–3866. doi: 10.1002/cne.22429.
- 37.Pasquier J, et al. Gene evolution and gene expression after whole genome duplication in fish: the PhyloFish database. BMC Genomics. 2016;17:368. doi: 10.1186/s12864-016-2709-z.
- 38.Thisse, B. & Thisse, C. Fast Release Clones: A High Throughput Expression Analysis (2004).
- 39.Boehm V, Gehring NH. Exon Junction Complexes: Supervising the Gene Expression Assembly Line. Trends Genet. 2016;32:724–735. doi: 10.1016/j.tig.2016.09.003.
- 40.Shaye DD, Greenwald I. OrthoList: a compendium of C. elegans genes with human orthologs. PLoS One. 2011;6:e20085. doi: 10.1371/journal.pone.0020085.
- 41.Oltrabella, F. et al. Role of the endocannabinoid system in vertebrates: Emphasis on the zebrafish model. Dev Growth Differentiation, doi:10.1111/dgd.12351 [Epub ahead of print] (2017).
- 42.Westerfield, M. The Zebrafish Book - A guide for the laboratory use of zebrafish (Danio rerio) (University of Oregon Press, 2000).
- 43.Thisse, C. & Thisse, B. High-resolution in situ hybridization to whole-mount zebrafish embryos. Nat Protoc3, 59–69, doi:nprot.2007.514 [pii] 10.1038/nprot.2007.514 (2008).
- 44.Jin L, Lloyd RV. In situ hybridization: methods and applications. J clinical laboratory analysis. 1997;11:2–9. doi: 10.1002/(SICI)1098-2825(1997)11:1<2::AID-JCLA2>3.0.CO;2-F.
- 45.Hwang WY, et al. Efficient genome editing in zebrafish using a CRISPR-Cas system. Nature biotechnology. 2013;31:227–229. doi: 10.1038/nbt.2501.
- 46.Sander JD, et al. ZiFiT (Zinc Finger Targeter): an updated zinc finger engineering tool. Nucleic Acids Res. 2010;38:W462–468. doi: 10.1093/nar/gkq319.
- 47.Koressaar T, Remm M. Enhancements and modifications of primer design program Primer3. Bioinformatics (Oxford, England) 2007;23:1289–1291. doi: 10.1093/bioinformatics/btm091.